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
V
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
VI
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
1.
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.
AF
[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
3
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
4
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
5
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.
6
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
7
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
8
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
9
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
10
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
11
•O'*-! 93 93
2 °'I5 %
IS u
“ CQ Qh
3 S bOC
-K £ c c
w S.3 o
g
3 ft'O
”3 “ 3 0*
>
•3 £t-
•5 £*3 «
— « J?
aS O
LO
•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
OX
.. 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
93
S,eS^
2-"
> w
0)
P4
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
12
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
13
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
14
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
15
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-
16
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
17
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,
18
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
19
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).
20
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
21
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
of
Start.
No. of
Compan-
ions.
Meters
per
Hour.
Condition.
Remarks.
Fish. ■
1
Aug. 21
1:45
0
72.96
Open tank with mirror.
) One
A
2
Aug. 21
2:05
0
76.39
Open tank with mirror.
> continuous
A
3
Aug. 21
2:25
0
87.20
Open tank with mirror.
) reading.
A
4
Aug. 22
11:20
0
101.45
Open tank with mirror.
A
S
Aug. 22
1:55
0
93.60
Open tank with mirror.
A
6
Aug. 22
4:30
0
84.20
Open tank with mirror.
A
7
Aug. 27
9:30
3
27.98
Open tank with mirror.
A
8
Aug. 27
1:05
3
38.45
Open tank with mirror.
A
9
Aug. 27
4:00
3
26.45
Open tank with mirror.
A
10
Aug. 28
9:35
3
42.37
Open tank with mirror.
A
11
Sept. 5
3:10
3
17.17
Open tank with mirror.
A
12
Aug. 27
2:55
3
50.70
Open tank with mirror.
B
13
Aug. 28
10:00
3
34.38
Open tank with mirror.
B
14
Sept. 5
2:00
3
14.91
Open tank with mirror.
B
( 15 min. after
B
IS
Sept. 5
4:30
0
49.38
Open tank with mirror.
•j companions
( were removed.
16
Sept. 6
9:25
0
70.12
Open tank with mirror.
B
17
Sept. 6
3:30
0
78.85
Open tank with mirror.
B
18
Sept. 7
9:05
0
58.45
Open tank with mirror.
B
19
Sept. 7
1:35
0
68.35
Open tank with mirror.
B
20
Sept. 8
11:15
0
62.50
Open tank with mirror.
B
21
Sept. 10
11:25
0
81.59
Open tank with mirror.
B
22
Sept. 11
11:00
0
89.48
Open tank with mirror.
B
23
Apr. 13
10:30
0
162.15
In box. 100 Watts.
C
24
Apr. 15
11:00
3
166.98
In box. 100 Watts.
C
25
Apr. 18
2:00
3
61.70
Out of box. Daylight
C
only.
26
Apr. 19
11:30
3
33.39
Out of box. 40 Watts.
C
27
Apr. 19
12:00
3
63.35
Out of box. 40 Watts.
C
(Rotor).
28
Apr. 19
4:30
3
91.95
Out of box. 40 Watts.
C
(No rotor).
29
Apr. 22
2:30
3
79.48
Out of box. Daylight
C
only.
30
Apr. 18
3:00
3
34.49
Out of box. Daylight
D
only.
31
June 4
0
46.49
In box. 10 Watts.
D
32
June 5
0
92.85
In box. 40 Watts.
D
33
June 6
0
86.95
In box. 100 Watts.
D
34
June 7
0
87.48
In box. 200 Watts.
D
35
June 10
0
95.10
In box. (No light).
D
36
Apr. 22
3:10
1
Out of box. Daylight
Simultaneous
C
only.
with 37.
37
Apr. 22
3:10
1
Out of box. Daylight
Simultaneous
D
only.
with 36.
38
Apr. 22
4:10
0
78.45
Out of box. 40 Watts.
E
39
Apr. 23
0
110.2
In box. 100 Watts.
E
40
Apr. 26
3
52.49
In box. 100 Watts.
E
41
Apr. 26
3:10
3
103.1
In box. 40 Watts.
E
22
[XXIII :1
Zoologica: New York Zoological Society
TABLE II.
V. vertical index from top down. L. longitudinal index from left to right. T, trans-
verse index from front to back. Numbers under Position indicate each 2-inch
space of projection. Figures in body of Table equal percentage of period
occupied.
Experiment.
1
2
3
4
5
6
Position.
V
L
T
V
L
T
V
L
T
V
L
T
V
L
T
V
L
T
1
IS
07
59
20
08
88
16
04
93
22
05
62
31
04
09
09
09
25
2
09
05
11
18
11
08
26
06
05
19
05
06
09
11
09
02
04
09
3
13
04
08
13
02
04
11
07
02
11
05
01
13
01
05
08
03
06
4
11
03
04
07
02
00
14
02
00
08
03
05
13
05
05
04
04
09
S
11
03
04
08
04
00
13
01
00
13
02
05
08
05
03
08
02
09
6
14
01
04
11
01
00
06
04
00
09
01
05
13
01
08
09
04
06
7
27
01
05
23
08
00
14
04
00
18
06
06
13
03
13
60
13
10
8
03
05
04
00
06
00
10
10
06
48
18
26
9
13
15
15
13
13
15
10
60
45
51
50
51
28
Experiment.
7
8
9
10
11
12
1
07
04
20
06
05
09
12
10
08
10
05
23
01
00
03
00
00
03
2
19
01
10
13
02
21
10
17
28
27
07
24
14
02
02
05
00
07
3
25
00
20
14
21
37
14
07
16
19
21
14
36
01
13
12
00
05
4
23
09
20
19
18
05
23
12
14
10
13
19
18
13
19
17
00
09
S
10
25
12
18
22
16
07
17
10
05
09
12
20
20
25
24
00
05
6
03
19
10
07
13
05
06
21
09
05
07
02
02
16
38
18
01
07
7
13
06
07
23
05
06
28
10
13
24
12
05
09
36
00
24
03
03
8
16
01
02
01
03
02
15
01
09
00
01
61
9
10
06
02
09
02
10
10
10
06
01
02
01
85
Experiment.
13
14
IS
16
17
18
1
01
00
01
14
01
01
01
81
02
00
05
13
00
23
03
00
01
05
2
47
00
08
12
01
04
02
08
08
00
00
12
00
07
01
01
01
05
3
17
00
05
30
13
28
14
00
00
01
01
06
03
08
01
12
01
05
4
07
02
03
30
13
37
28
01
02
13
01
06
17
01
01
22
01
02
5
03
09
10
02
25
17
18
01
05
21
00
02
16
07
01
23
01
08
6
OS
24
20
02
17
10
21
00
02
29
02
03
39
06
01
17
00
05
7
20
10
31
10
05
02
16
00
04
36
01
10
25
03
01
25
01
05
8
10
22
12
01
01
77
01
48
08
91
00
65
9
12
12
01
07
14
06
10
33
01
07
82
23
88
Experiment.
19
20
21
22
23
24
1
00
13
05
01
00
24
01
00
32
01
05
36
10
32
32
00
35
31
2
00
02
07
05
01
14
02
01
15
03
05
15
03
12
08
01
06
11
3
OS
01
06
13
01
03
17
00
09
07
02
01
06
05
05
03
08
06
4
12
04
02
21
01
05
13
00
05
15
02
05
13
04
05
07
04
06
S
16
04
02
23
00
05
15
01
07
13
02
05
11
05
04
24
04
08
6
16
01
02
13
01
05
26
01
10
24
02
02
15
03
06
21
03
06
7
51
06
07
24
01
21
26
01
06
37
02
05
42
06
12
44
03
08
8
04
69
01
23
01
16
02
31
06
28
06
24
9
06
06
02
10
05
06
10
59
88
93
68
22
25
1938]
Breder & Nigrelli: Mass Physiology of Fishes
23
TABLE II. — Continued
Experiment.
25
26
27
28
29
30
Position.
V
L
T
V
L
T
V
L
T
V
L
T
V
L
T
V
L
T
1
01
63
23
00
32
10
00
28
17
01
08
04
01
38
28
00
04
08
2
06
08
28
00
05
14
01
07
20
05
06
10
03
11
19
01
04
20
3
11
04
08
00
05
04
01
04
12
23
09
09
05
07
08
01
12
08
4
06
01
13
00
20
03
04
04
06
18
11
09
19
02
07
03
05
09
S
12
01
06
03
08.
04
05
03
06
07
06
10
12
04
09
05
06
08
6
18
04
05
05
00
22
16
03
09
16
05
31
18
01
10
13
10
06
7
46
01
08
92
00
42
73
04
13
30
06
16
42
03
13
77
09
13
8
03
09
05
01
06
17
12
11
03
06
14
28
9
06
05
12
18
11
17
10
09
20
29
19
20
19
Experiment.
31
32
33
34
35
36
1
01
97
05
04
40
46
14
32
23
17
13
13
13
15
83
00
16
2
09
01
21
04
04
13
05
02
12
07
04
08
06
08
10
01
03
3
17
00
15
04
04
07
06
05
08
06
04
03
21
07
01
01
04
4
23
00
13
09
08
03
07
04
08
10
07
04
20
09
01
09
14
T3
5
26
01
13
11
04
03
07
06
08
11
05
04
16
08
01
20
06
<v
6
17
00
12
24
04
04
14
07
07
20
07
05
14
05
01
24
07
7
07
00
17
44
06
07
47
01
10
29
09
16
10
06
02
45
04
£
8
00
04
07
17
09
24
07
47
12
01
07
9
00
06
10
10
19
07
10
01
17
24
34
11
32
Experiment.
37
38
39
40
41
1
04
04
01
44
14
05
19
52
00
10
09
01
18
29
2
00
01
06
11
12
11
08
04
01
10
13
00
06
05
3
01
06
12
05
08
10
04
04
00
12
15
01
04
06
4
09
08
•ri
17
06
06
16
08
04
00
09
18
09
06
11
5
09
08
u,
14
06
05
19
08
04
00
13
18
22
04
12
6
32
07
■w
16
04
06
12
10
04
00
10
12
23
06
09
7
45
06
£
34
04
08
27
05
06
99
09
09
44
04
14
8
03
09
41
06
22
09
06
03
14
9
09
06
11
09
06
10
48
05
21
09
41
24
Zoologica: New York Zoological Society
[XXIII :1
TABLE III.
Data of others.
Schuett
( 1934) 1
Escobar,
Minahan and
Shaw (1936)2
Compilation of published data for comparative purposes.
Volume of Aquaria.
7,500 cc.
15,000 cc.
Two aquaria of
similar shape
Companions.
Meters per Hour.
were used but
0
200—
150—
the dimensions
3
50—
80—
were not given.
7
90+
40+
15
110+
90+
Volume of Aquaria.
Fish.
A
A
B
B
C
C
Companions.
0
3
0
3
0
3
1,625 cc.
3,250 cc.
4,875 cc.
6,500 cc.
Meters per Hour.
83.40
83.44
51.33
118.18
17.26
55.47
32.54
58.43
103.40
69.43
61.95
100.09
39.10
39.65
13.08
50.04
81.06
59.58
17.88
108.69
24.15
51.84
51.14
17.58
Comparative travel of Carassius from all data.
Tank
Size.
No. of
Alone.
With 3
Companions.
Number of
Observation
Periods.
Difference.
Values Alone
Reduced to Unity.
Then Value with
+3 =
Escobar
i3
92.84
27.58
47
65.26
.30—
Escobar
23
73.95
48.20
47
25.75
.64—
Escobar
33
50.87
27.03
47
23.84
.53+
Escobar
43
109.07
48.77
47
60.30
.45—
Schuett
5
2 00.00 ±
50.00±
?
1S0.00±
.25
Schuett
6
150.00±
80.00±
?
70.00±
.53 +
B.&N.
7
76.77
31.55
22
45.22
.41 +
1 Approximate values picked from the published graph.
2 Heterotypic associations omitted from this Table. Fish A and B : each value represents the mean
of 20 separate 20-minute periods. Fish C : each value represents the mean of 7 separate 20-minute
periods. The aquarium used measured 24 x 12 x 24 cm. deep and only the volume of water was varied.
3 These values are weighted A & B — 20, C — 7. (Range 75 to 34)
1938]
Breder & Nigrelli: Mass Physiology of Fishes
25
TABLE IV.
Comparative travels of Carassius in certain experiments.
Rate of Travel M/H.
Exp. No.
Fish.
Minimum.
Mean.
Maximum.
No. of Periods.
Alone.
1-6
A
72.96
86.00
101.45
6
15-22
B
49.38
69.84
89.48
8
With 3 Companions.
7-11
A
17.17
30.48 +
42.37
5
12-14
B
14.91
36.66+
50.70
3
Averages.
Alone
72.96
76.77—
101.45
14
3 Companions
14.91
31.55 +
50.70
8
TABLE V.
Mean distribution values for Fish A and B, expressed in percentage.
A (1-6)
B (15-22)
Average
14
V
L
T
V
L
T
V
L
T
Alone.
19
6
56
01
16
15
8
12
33
14
7
8
02
3
10
7
5
8
12
4
4
9
2
4
10
2
4
9
3
4
17
1
3
14
2
4
10
3
3
18
2
4
15
2
4
10
2
4
23
2
4
18
2
4
26
6
6
30
2
7
28
4
7
8
15
3
53
5
36
14
7
10
47
62
56
(7-11)
(12-14)
8
With 3
7
5
12
5
01
2
6
3
8
Companions.
16
6
17
21
01
6
18
4
13
21
10
20
20
4
13
21
8
18
18
13
16
18
5
16
18
10
16
14
18
15
10
11
11
12
16
13
4
15
13
8
14
12
6
15
13
20
14
06
18
6
12
19
11
8
9
01
8
28
8
11
6
11
8
4
39
17
Effect of light on position and speed.
Exp. No.
Watts.
% at Surf.
% at Bott.
% Above Mid.
% Below Mid.
Speed.
35
0
13
10
40
60
95.10
31
10
01
07
39
61
46.49
32
40
04
44
17
82
92.85
33
100
14
47
29
71
86.95
34
200
17
29
35
65
87.48
[XXIII :1
26
Zoologica: New York Zoological Society
TABLE VI.
Movements of animals in a shallow circular pan 6 feet in diameter, 2 inches deep.
Fishes.
No. on Text-
fig. 2.
Species.
Clockwise
Loops.
Counter-clock-
wise Loops.
Direction at End of Path.
i
Carassius
auratus No. 1
i
i
Counter-clockwise.
2
Carassius
auratus No. 2
0
0
Counter-clockwise.
3
Lebistes
reticulatus $
0
i
Counter-clockwise.
4
Brachydanio
rerio
1
i
Counter-clockwise.
S
Villarius
catus
0
0
Counter-clockwise.
Total
2
3
Snails (First Trial).
Length of Observation.
Text-fig. 3
Physa No. 1
0
3
10.0 min.
6
Physa No. 2
3
0
13. S
6
Physa No. 3
0
1
16.5
4
Planorbis No. 1
0
0
12.5
6
Vivipara No. 1
1
0
30.0
6
Vivipara No. 2
1
0
30.0
Snails (Second Trial).
Text-fig. 3
Physa No. 1
0
0
2.5
6
Physa No. 2
0
1
8.5
6
Physa No. 3
0
2
26.0
4
Planorbis No. 1
0
1
22.5
Snails (Third Trial).
Text-fig. 6
Physa No. 3
1
4
24.0
4
Planorbis No. 1
0
1
Snail (Fourth Trial).
Text-fig. 4
Planorbis No. 1
0
0
TOTAL
6
13
1938]
Breder & Nigrelli: Mass Physiology of Fishes
27
Bibliography.
Allee, W. C.
1931. Animal Aggregations. Chicago. University of Chicago Press.
Allee, W. C. and Bowen, E. S.
1932. Studies in Animal Aggregation: Mass Protection against Colloidal
Silver among Goldfishes. Jour. Exp. Zool. 61: 1-32.
Beebe, W.
1935. Resume of the 1935 Expedition of the Department of Tropical Re-
search. Bull. N. Y. Zool. Soc. 38 (6) : 191-196.
Bowen, E. S.
1931. The Role of the Sense Organs in Aggregation of Ameiurus melas.
Ecological Monographs. 1 (1): 1-35.
1932. Further Studies of the Aggregating Behavior of Ameiurus melas.
Biol. Bull. 63: 252-270.
Breder, C. M., Jr.
1929. Certain Effects in the Habits of Schooling Fishes, as Based on the
Observation of Jenkinsia. Amer. Mus. Novitates. (382) : 1-5.
1934. An Experimental Study of the Life-History and Breeding Habits of
the Cichlid Fish, Aequidens latifrons. Zoologica. 18 (1) : 1-42.
1936. The Reproductive Habits of the North American Sunfishes, Centrar-
chidae. Zoologica. 21 (1) : 1-48.
Breder, C. M., Jr. and Coates, C. W.
1934. Sex Recognition in the Viviparous Poecilid, Lebistes reticulatus. Anat.
Rec. 60 (4) : 45 (Abst. No. 45).
1935. Sex Recognition in the Guppy, Lebistes reticulatus Peters. Zoologica.
19 (5) : 187-207.
Breder, C. M., Jr. and Harris, J. E.
1935. Effect of Light on Orientation and Stability of Young Plectognath
Fish. Papers Tortugas Lab. 29 (3) : 23-36. Carnegie Inst. Wash.
Pub. (452).
Breder, C. M., Jr. and Nigrelli, R. F.
1934. The Influence of Temperature and other Factors on the Winter Ag-
gregations of the Sunfish, Lepomis auritus, Anatomical Record. 60 (4) :
97 (Abstract No. 160).
1935a. The Influence of Temperature and other Factors on the Winter Ag-
gregations of the Sunfish, Lepomis auritus, with Critical Remarks on
the Social Behavior of Fishes. Ecology. 16 (1) : 33-47.
1935b. A Technique for the Quantitative Measure of Tropisms and Social
Behavior in Active Animals, with an Illustration. Anatomical Record.
64 (1) : 119 (Abstract No. 208).
Brownlee, J.
1911. The Mathematical Theory of Random Migration and Epidemic Dis-
tribution. Pro. Roy. Soc. Edin. 31: 262-289.
Cole, W. H.
1922. Note on the Relation Between the Photic Stimulus and the Rate of
Locomotion in Drosophila. Science. 55: 678-679.
Eddy, M. W.
1925. Metabolic Rate and Bunching in Catfish. Anatomical Record. 31: 1332
(Abstract No. 79).
28
Zoologica: New York Zoological Society
[XXIII :1
Edgeworth, F. Y.
1920. Entomological Statistics. Metron. 1: 75-80.
Einstein, A.
1905. Zur Elektrodynamik bewegter Korper. Ann. d. Physik. 17: 549.
Escobar, R. A., Minahan, R. P. and Shaw, R. J.
1936. Motility Factor in Mass Physiology: Locomotor Activity of Fishes
under Conditions of Isolation, Homotypic Grouping. Physiol. Zool.
9 (1): 66-78.
Gordan, A.
1936. Paths of Flight. Science. 84 (2167) : 40-41.
Hardy, W. B.
1922. The Development Commission. 12th Report of the Development Com-
mission. Nature (110) : 865-866.
Hollister, G.
1934. Salt and Fresh-Water Viability of Fish. Bull. N. Y. Zool. Soc. 37 (6) :
183-187.
Langlois, T. H.
1935. A Study of the Small-Mouth Bass, Micropterus dolomieu (Lacepede)
in Rearing Ponds in Ohio. Bull. Ohio Biol. Survey. 4 (4) : 191-225.
Lissmann, H. W.
1932. Die umwelt des Kampfisches ( Betta splendens Regan). Zeitschr.
V eigleichende Physiologie. 18 (1) : 65-111.
Lotka, A. J.
1925. Elements of Physical Biology. 460 pp. Williams and Wilkens. Balti-
more.
Newman, E.
1876. Mr. Saville Kent’s Lectures at the Society of Arts on the “Aquarium
Construction and Management.” Zoologist. 4856.
Noble, G. K.
1934. Sex Recognition in the Sunfish, Eupomotis gibbosus (Linne). Copeia.
(4) : 151-155.
Noble, G. K. and Curtis, B.
1935. Sexual Selection in Fishes. Anatomical Record. 64 (1) : 84 (Abstract
No. 126).
Parr, A. E.
1927. A Contribution to the Theoretical Analysis of the Schooling Behavior
of Fishes. Occ. Pap. Bingham Oceanographic Coll. ( 1 ) : 1-32.
1931. Sex Dimorphism and Schooling Behavior among Fishes. Amer. Nat.
65: 173-180.
Pearson, K. and Blakeson, J.
1906. Mathematical Contributions to the Theory of Evolution. XV. A Mathe-
matical Theory of Random Migration. Draper’s Co. Research Memoirs.
Ser. 3 (15) Biometric Ser. : 1-54.
Przibram, K.
1913. liber die ungeordnete Bewegung niederer tiere. Phliiger Archiv. 153:
401-405.
Reighard, J. E.
1915. An Ecological Reconnaissance of the Fishes of Douglas Lake, Cheboy-
gan County, Michigan, in Mid-Winter. Bull. U. S. Bur. Fisheries. 33:
215-249.
1938]
Breder & Nigrelli: Mass Physiology of Fishes
29
Ross, R.
1923. Memoirs, with a Full Account of the Great Malaria Problem and Its
Solution. London. 547 pp.
Schaeffer, A. A.
1928. Spiral Movements in Man. Jour. Morph, and Physiol. 45: 293-398.
1931. On Molecular Organization in Ameban Protoplasm. Science. 74
(1906) : 47-51.
SCHUETT, F.
1934. Studies in Mass Physiology: the Activity of Goldfish Under Different
Conditions of Aggregation. Ecology. 15: 258-262.
Spooner, G. M.
1931. Some Observations on Schooling in Fish. Journ. Marine Biol. Assoc.
United Kingdom. 17 : 421-448.
Thompson, D. H.
1933. The Migration of Illinois Fish. Biol. Notes (1) : 1-25. Nat. Hist.
Survey 111. (Mimeographed).
Townsend, C. H.
1916. Hibernating Fishes. Bull. N. Y. Zool. Soc. 19 (2) : 1345.
Welsh, J. H.
1933. Light Intensity and the Extent of Activity of Locomotor Muscles as
Opposed to Cilia. Biol. Bull. 65 (2) : 168-174.
Welty, J. C.
1934. Experimental Exploration into the Group Behavior of Fishes : A Study
of the Influences of the Group on Individual Behavior. Physiol. Zool.
7: 85-128.
Arnold: Anoplocephaline Cestodes of N. Amer. Rabbits
31
2.
A Study of the Anoplocephaline Cestodes of
North American Rabbits.1
John G. Arnold, Jr.
Department of Biology , University College,
New York University.
(Text-figures 1-23).
1. Introduction.
Previous work on the anoplocephaline cestodes of North American rab-
bits has resulted in much confusion. Observations on internal anatomy, time
of appearance of genital primordia, number of testes and other morpholog-
ical details are so lacking in agreement that any interpretation is difficult.
Baer (1927) and Sprehn (1932) have incorrectly cast all American leporine
species of the genus Cittotaenia into synonymy with the European species
C. pectinata (Goeze, 1782). The purpose of the present investigation was
to secure pi’ecise information on the North American representatives of
this genus, correct the errors, and eliminate the confusion that has existed
in the knowledge of the rabbit cestodes of the family Anoplocephalidae.
2. Acknowledgements.
The present study was suggested by Professor Horace W. Stunkard
and carried out under his direction. The author wishes to thank Professor
Stunkard for helpful criticisms, for the loan of the European material used
in comparison, and for the use of his private library. Grateful acknowledge-
ment is also extended to Dr. Maurice C. Hall, U. S. Bureau of Animal
Industry, U. S. Department of Agriculture, for the privilege of studying
original maternal deposited in his care. The kindness of Dr. R. V. Boughton
of the University of Manitoba, in lending specimens of Cittotaenia pectinata
americana, is recognized. Lastly, the author gratefully acknowledges the
assistance of Marion E. Arnold for criticisms and aid in the preparation
of this paper.
3. Materials and Methods.
The material used in this study consisted of both North American and
European species. Specimens of all the American species and four of the
five European ones provided material for a comprehensive treatment of the
subject. Andrya rhopalocephala from European hares is rare and no mate-
rial of that species was available. Since doubt has been expressed con-
1 Contribution from the Biological Laboratory, New York University, University Heights, New
York.
32
Zoologica: New York Zoological Society
[XXIII :2
cerning the specific distinctness of the North American species, a detailed
comparison has been made between North American and European forms.
Specimens of Cittotaenia denticulata (Rudolphi, 1804), C. ctenoides
(Railliet, 1890), C. pectinata (Goeze, 1782), and Andrya cunicidi (Blan-
chard, 1891) were loaned to the writer by Professor Stunkard, who col-
lected them during 1931 and 1932 from the vicinity of Hamburg, Germany.
The American specimens of the family Anoplocephalidae were obtained
from various sources. The writer examined the intestines of 9 cottontail
rabbits, Sylvilagus floridanus mallurus, killed near New York City; 7 of
the same species from Boylesville, Pennsylvania; 7 from cottontails killed
in northern New York State; and 156 from cottontails, Sylvilagus floridanus
alacer, killed near Wichita, Kansas. The material in the last two groups
was obtained from slaughter houses in New York City. All the parasites
were killed and fixed in a saturated solution of corrosive sublimate and
were washed, stained, dehydrated, cleared, mounted and studied either as
sections or in toto. Alcoholic and mounted specimens of the genera Citto-
taenia and Schizotaenia were secured through the kindness of Dr. M. C.
Hall. Similar specimens of C. pectinata americana were loaned by Dr. R. V.
Boughton.
4. Historical Review.
The subfamily Anoplocephalinae was erected by Blanchard (1891),
and the family Anoplocephalidae by Kholodkovsky (1902). Fuhrmann
(1907) characterized the family Anoplocephalidae as follows: “Scolex meist
kugelig, seltener gestreckt, unbewaffnet; Saugnapfe verhaltnismassig gross;
Hals fehlt; Gleider kurz und breit; Genitalien einfach oder doppelt. Geni-
talpori randstandig; Eier ,oft mit einem ‘birnformigen Apparat.’ In
Saugetieren und Vogeln.” In his recent monograph, Fuhrmann (1931) gave
a more extended diagnosis of the family. He stated: “Scolex immer ohne
Rostellum. Glieder meist breiter als lang. 1 oder 2 Genitalapparate in
einer Proglottis. Genitaloffnung beidseitig, einseitig, regelmassig oder un-
regelmassig alterierend. Weibliche Genitaloffnung oft verschwindend, selten
fehlend. Hoden zahlreich. Weibliche Geschlechtsdriisen haufig poral ver-
schoben. Uterus sackformig, retikular oder sich in Ei-Kapseln auflosend
oder mit 1 bis zahlreichen Paruterinorganen versehen. Eier mit 3 Hiillen,
von welchen die innerste oft einen birnformigen Apparat aufweist. Ent-
wickelungsgeschichte vollstandig unbekannt. In Reptilien (2 Genera), in
Vogeln (7 Genera), in Saugern (21 Genera).”
Five species of anoplocephaline cestodes occur in European hares and
rabbits. Three of them belong to the genus Cittotaenia, the others to the
genus Andrya. The American leporine cestodes belong to the genera Citto-
taenia and Schizotaenia. Although the genus Andrya is represented in North
America, species have not yet been found in rabbits and hares. To facili-
tate description and avoid repetition, the diagnostic features of the three
genera are listed.
Andrya Railliet, 1893, Anoplocephalinae: Segments much broader than
long except in most distal parts of strobila. Single set of reproductive
organs in each segment; genital pores irregularly alternate. Testes mostly
on aporal side of segment; female genitalia on poral side. Uterus typically
reticular to saccular. Pyriform apparatus present. Adults in mammals.
Type species'. A. rhopalocephala (Riehm, 1881).
Cittotaenia Riehm, 1881. Anoplocephalinae: Segments broader than
long. Double set of reproductive organs in each proglottid; genital pores
in each lateral wall. Uterus saccular. Pyriform apparatus present. Adults
in mammals and birds.
Type species'. C. denticulata (Rudolphi, 1804) Stiles, 1896.
Schizotaenia, Janicki, 1906. Anoplocephalinae: Segments much broader
1938 J Arnold: Anoplocephaline Cestodes of N. Amer. Rabbits
33
than long. Single set of reproductive organs to each segment; genital pores
typically alternate. Testes in median field or segment; female genitalia
slightly to aporal side. Uterus reticular. No pyriform apparatus present.
Adults in mammals.
Type species: S. decrescens (Diesing, 1856).
A brief historical account of each species considered in the present
paper is given later with the specific description. For more detailed his-
torical data, the reader is referred to the original sources cited in the
bibliography.
5. ANOPLOCEPH ALINE CESTODES FROM NORTH AMERICAN HARES AND RABBITS.
Anoplocephaline cestodes were first reported in North American rab-
bits by Curtice (1888), who described supposedly the early larval stages
of Taenia pectinata in Lepus sylvaticus.
At present, four species of anoplocephaline tapeworms are recognized
in North American hares and rabbits: (1) Schizotaenia americana (Stiles,
1895) (= Andrya americana — Andrya americana leporis — Bertia ameri-
cana — Bertia americana leporis — Schizotaenia americana leporis) ; (2)
Cittotaenia perplexa (Stiles, 1895) (= Ctenotaenia perplexa = Cittotaenia
mosaica) ; (3) Cittotaenia pectinata americana Douthitt, 1915 (= Cittotae-
nia pectinata of Lyman, 1902) ; (4) Cittotaenia variabilis (Stiles, 1895)
(= Ctenotaenia variabilis = Cittotaenia variabilis variabilis — Cittotaenia
variabilis angusta — Cittotaenia variabilis imbricata) .
Schizotaenia americana (Stiles, 1895) Janicki, 1906.
(Text-figs. 1, 10).
This cestode was first described as Andrya americana leporis by Stiles
(1895), and was transferred by him (1896) to the genus Bertia on the
basis of its resemblance to Bertia americana (Stiles, 1895) Stiles, 1896,
from the porcupine. Stiles postulated that B. americana leporis might
eventually be elevated to specific rank, but that his five poorly preserved
specimens did not warrant such a step. The material was collected by C.
Curtice from Lepus sp. and the locality in the United States from which it
was taken is unknown. Stiles and Hassall (1902) proposed the name
Bertiella for Bertia, since the latter was preoccupied.
Cohn (1906) stated that Bertia americana was synonymous with Taenia
laticephala Leidy. He retained the genus Bertia (he had apparently not seen
Stiles’ and Hassall’s new allocation of the species), but placed the specific
name, americana, in synonymy.
In his description of the cestodes of mammals, Janicki (1906) erected
the new genus Schizotaenia with S. macrorhyncha (Rudolphi, 1810) as type
and in it he included Bertia americana (Stiles, 1895) and Bertia americana
leporis (Stiles, 1895).
Douthitt (1915) declared that the variety Schizotaenia americana
leporis (Stiles, 1895) should be dropped in favor of Schizotaenia americana.
He stated that the variety was too little known and too incompletely described
to permit any distinction from S. americana. Douthitt gave sound and logi-
cal arguments to disprove Cohn’s suggestion that Bertia americana and
Taenia laticephala Leidy were identical.
Meggitt (1924) reestablished Schizotaenia americana leporis as a valid
species. He reduced the species S. americana, which is from the porcupine,
to synonymy with S. laticephala. In this he followed Cohn’s contention and
disregarded Douthitt’s argument.
34
Zoologica: New York Zoological Society
[XXIII :2
Baer (1927) removed Schizotaenia americana from synonymy with S.
laticephala. Since he did not mention S. americana leporis, he apparently
followed Douthitt and considered the leporine variety identical with the
porcupine species.
Sprehn (1932), in his Lehrbuch der Helminthologie, did not even men-
Text-figures 1-9.
Abbreviations : c — cirrus sac. e — excretory duct, n — nerve fiber, o — ovary, p — pars prostatica.
r — seminal receptacle, s — shell gland, t — testis, u — uterus, v — vitelline gland, va — vagina, vd —
vas deferens, vs — seminal vesicle. All measurements are to a scale of 0.1 mm., with the exception
of Text-figure 8 in which the scale is 0.01 mm. All drawings are camera lucida.
1. Scolex, Schizotaenia americana. 2. Scolex, Cittotaenia variabilis. 3. Scolex,
Cittotaenia perplexa. 4. Scolex, Cittotaenia denticulata. 5. Scolex, Andrya
cuniculi. 6. Scolex, Cittotaenia pectinata americana. 7. Scolex, Cittotaenia
pectinata. 8. Ovum, Cittotaenia denticulata. 9. Scolex, Cittotaenia ctenoides.
35
1938J Arnold: Anoplocephaline Cestodes of N. Amer. Rabbits
tion the genus Schizotaenia, or give Janicki as a reference. He also failed to
mention either Bertiella americana or B. americana leporis, although he rec-
ognized the generic name, Bertiella, proposed by Stiles and Hassall (1902).
The writer was fortunate to obtain a cotype of Stiles’ original material
of Schizotaenia americana leporis (U. S. Nat. Mus. 1170).
Since the specimen was incomplete and in poor condition, the total
length and number of proglottids could not be determined. The fragment,
consisting of 75 proglottids, measured 33 mm. in length and 6.5 mm, in
maximal width. A neck, if present, was too contracted to be recognizable.
The scolex (Text-fig. 1) was 0.63 mm. wide (Stiles found it to be 0.64 mm.)
and the suckers measured 0.25-0.26 mm. in length and 0.21-0.25 mm. in width.
This is a wider variation than found previously in either Schizotaenia ameri-
cana or its variety, S. americana leporis.
The male genital primordia with evidences of the male ducts appeared
first in the sixth proglottid. The cirrus and cirrus sac were well developed
by the 20th segment, and the cirrus was seen extruded in the 23rd. Douthitt
stated that the genital primordia were present in the first segment. Accord-
ing to Douthitt, sexual maturity was attained in the 80th proglottid (Text-
fig. 10). The cirrus sac, containing the seminal vesicle and a long thin
cirrus, was very muscular, especially at its proximal end. It measured as
much as 0.69 mm. in length. It was Douthitt’s opinion that part of the
seminal vesicle was outside the cirrus sac. In one of the best preserved
proglottids, 65-70 testes (Stiles reported 50) were observed in an irregular
row in the distal half of the proglottid. The convoluted vas deferens ex-
tended from the region of the ovary to the cirrus sac and, just before it
entered the latter, it was surrounded by glandular cells, probably the pars
prostatica. The cirrus sac opened to the exterior at approximately the middle
of the lateral margin. The vagina emptied just below the male opening.
There were indications of a genital papilla.
The primordium of the seminal receptacle first appeared definitely in the
eighth segment, although indications were seen in the seventh proglottid.
The structure present was rather spherical and was found alternately to the
right and left of the median line. The details of the female reproductive
system could not be made out. Only two ovaries were measured and their
probable widths were 0.53 and 0.63 mm.; Douthitt’s measurement of the
ovary width was 1.3 mm. The uterus arose from the female genital complex
as a fan-shaped structure, and was filled with eggs by the 58th segment.
The ova measured 30-40 p in diameter, with an average of 39 p. According
to Douthitt, the outer embryonic membrane ranges from 55-61 p in diameter.
Comparison of the specific diagnoses (Stiles, 1896) of Bertia americana
and B. americana leporis disclosed only minor differences which may be
regarded as individual variations rather than specific characters. After
examination of specimens of S. americana leporis, the writer agrees with
Douthitt and Baer that the variety name should be suppressed.
From the above description and from a historical review, the following
specific diagnosis may be derived.
Diagnosis : Schizotaenia americana (Stiles, 1895) Janicki, 1906.
Strobilae from 23-47 mm. in length, and from 5-6.5 mm. in maximal breadth.
Number of proglottids may exceed 95. Scolex varies from 0.6-0. 7 mm. in
breadth. Neck absent, stroblization beginning immediately. Genital organs
single; pores lateral and regularly alternate. Male genital primordia appear
first at the 6th segment; the female, between the 7th (?)-14th proglottids.
Testes, 50-70 in number, found in an irregular row in distal half of proglot-
tid. Muscular cirrus sac contains the seminal vesicle and a long thin cirrus.
The vagina opens just below the cirrus sac, and the two tend to form a genital
papilla. Ovary usually indistinct, median (?). Douthitt reported it was 1.3
mm. in width. Uterus arises from female genital complex in fan-like fash-
36
Zoologica: New York Zoological Society
| XXIII :2
Text-figures 10-17.
Abbreviations: c — cirrus sac. e — excretory duct, n — nerve fiber, o — ovary, p — pars prostatica.
r — seminal receptacle, s — shell gland, t — testis, u — uterus, v — vitelline gland, va — vagina, vd —
vas deferens, vs — seminal vesicle. All measurements are to a scale of 0.1 mm. All drawings are
camera lucida.
1 0. Mature proglottid, Schizotaenia americana. 1 1 , Mature proglottid, Cittotaenia
perplexa. 12. Mature proglottid, Cittotaenia variabilis. 13. Mature proglottid,
Cittotaenia pectinata americana. 14. Mature proglottid, Cittotaenia denticul-
ata. 15. Mature proglottid, Cittotaenia ctenoides. 16. Mature proglottid,
Cittotaenia pectinata. 17. Mature proglottid, Andrya cuniculi.
1938] Arnold: Anoplocephaline Cestodes of N. Amer. Rabbits
37
ion to fill median field. Ova range from 30-40 p in material studied, but vary
from 55-61 p, according to Douthitt.
Hosts : Yellow-haired porcupine, Erethizon epixanthes; Canadian por-
cupine, E. dorsatus; and Lepus sp.
Habitat : Wyoming and New York.
Cittotaenia perplexa (Stiles, 1895).
Stiles and Hassall, 1896.
(Text-figs. 3, 11).
This species was described by Stiles (1895) as Ctenotaenia perplexa,
but was later transferred (Stiles and Hassall, 1896) to the genus Citto-
taenia for reasons of priority. Later in 1896, Stiles published an inclusive
diagnosis of the species.
In 1908, Hall described this same form as Cittotaenia mosaica, and
later (1912) listed it from the intestine of Sylvilagus nuttalli pinetis, a
cottontail rabbit of Colorado. The synonymy was pointed out by Douthitt
(1915) who restudied Stiles’ original specimens of C. perplexa (U. S. Nat.
Mus., No. 1110). He found that Stiles’ account was in error concerning the
length of the cirrus sac and distribution of testes, and that the features
which had been used to distinguish C. mosaica from C. perplexa were not
significant differences.
Meggitt (1924) recognized Cittotaenia perplexa as a valid species with
C. mosaica as a synonym. Baer (1927) reduced C. perplexa to synonymy
with Cittotaenia pectinata (Goeze, 1782), which he considered to be ex-
ceedingly variable and widely distributed. Sprehn (1932) failed to mention
either C. perplexa or C. mosaica.
In the present study, 26 incomplete specimens of Cittotaenia perplexa
(Stiles, 1895) and Cittotaenia mosaica Hall, 1908, were loaned to the writer
by Dr. M. C. Hall, U. S. Bureau of Animal Industry, U. S. Department of
Agriculture. The specimens of C. perplexa bore the following numbers,
U. S. Nat. Mus. Nos. 17226, 17246, and 17449. The specimens of C. mosaica
were catalogued as U. S. Nat. Mus. Nos. 28429 and 24845. These included
cotype material.
Of the fragments studied, the longest contained 150 proglottids and
measured 70 mm. in length with a maximal width of 11 mm., the largest
breadth yet reported. The greatest length reported previously (Hall, 1908)
for C. mosaica was 100 mm. The scolex (Text-fig. 3) ranged from 0.32-0.45
mm. in width. The suckers were 0.11 mm. in diameter, which agrees with
the measurement of Stiles. The scolex was not differentiated from a short
and broad unsegmented neck, which measured from 0.30-0.35 mm. in length.
The genital organs were double and the pores opened anterolaterally.
The female genital primordia appeared in the 10th proglottid and their
ducts were first evident in the 16th. The male genital primordia first ap-
peared about the 30th and their ducts in the 35th segment. The follicular
ovary was well developed in the 70th segment, and disappeared in the 95th.
This was correlated with the appearance of eggs in the uterus. This con-
dition agreed with the observations of Hall and Douthitt. The ovary (Text-
fig. 11) measured from 0.50-0.61 mm. in width. Douthitt reported a width
of 0.85 mm. According to Hall, the yolk gland measured 0.20-0.23 in width,
and the shell gland 0.074-0.092 mm. The seminal receptacle was rather large
and situated within the longitudinal canals. The vagina was long and thin,
opening below the cirrus sac. The ova ranged from 57-69 p in diameter.
Hall stated that they reached a maximum of 105 p.
The testes, 120-125 in number, were in the distal half of the segment.
In the younger proglottids, the testes were continuous from lateral canal to
38
Zoologica: Neiv York Zoological Society
[XXIII :2
lateral canal, while in more mature segments there was a definite median
break in continuity of the chain. This was apparent both in Stiles’ cotype
material and in Hall’s specimens of C. mosaica. The testes varied from
0.049-0.065 mm. in diameter. The maximal size recorded by Hall was 0.09
mm. The convoluted vas deferens proceeded from the region of the female
glands to the long, narrow cirrus sac, which measured from 0.43-0.50 mm.
in length. The maximal sizfe recorded by Hall was 0.64 mm. In some of
Stiles’ material, Douthitt observed cirrus sacs as long as 0.55 mm.
The following diagnosis may be offered.
Diagnosis : Cittotaenia perplexa (Stiles, 1895) Stiles and Hassall, 1896.
Mature specimens measure 3.8-10 cm. in length, and 11 mm. in maximal
width. The scolex, 0.27-0.45 mm. in breadth, is not differentiated from a
short, unsegmented neck region. Proglottids number over 150. Female gen-
ital primordia appear first in the 10th and their ducts in the 16th segment.
Ovary well developed by the 70th segment; disappears at the 95th. Ovary,
0.50-0.85 mm. in width ; ova 50 to 105 g in diameter. Male genital primordia
appear first in the 30th and their ducts in the 35th proglottid. Testes con-
tinuous from longitudinal canal to longitudinal canal in young proglottids,
but divided into two groups in more mature ones. Testes 60-125 in number
and 0.049-0.09 mm. in diameter. Cirrus sac 0.43-0.64 mm. long.
Hosts : Sylvilagus nuttalli pinetis, Sylvilagus floridanus mallurus, and
Sylvilagus floridanus alacer.
Habitat: Colorado, Maryland and Oklahoma.
Cittotaenia pectinata amerieana Douthitt, 1915.
(Text-figs. 6, 13, 19).
In 1896, Stiles declared that Cittotaenia variabilis (Stiles, 1895) is an
American variant of Cittotaenia pectinata (Goeze, 1782) of Europe. Later
discoveries have shown that C. pectinata amerieana of North America is
much more closely related to C. pectinata. This American cestode was
described by Lyman (1902) from Lepus melanotis, the common jackrabbit.
Hall (1908) noted: “A hasty comparison of specimens of the European and
American C. pectinata shows certain differences that should be determined
as accidental or shown to be of specific or subspecific importance.”
Later, Douthitt (1915) concluded that the American form “should be
designated as a distinct variety to avoid confusion,” and he proposed the
name C. pectinata amerieana. This terminology was followed by Meggitt
(1924), John (1926), Boughton (1932), and Rees (1933a).
Baer (1927) and Sprehn (1932) failed to recognize any distinctly
American variety.
The author was fortunate in obtaining several paratype specimens of
C. pectinata amerieana from Dr. R. V. Boughton. The original specimens
of C. pectinata america^ia, which were described by Douthitt, were not
available.
Ten almost complete specimens were studied. The longest, consisting
of at least 360 proglottids, was 110 mm., and the broadest 7.5 mm. The
scolex (Text-fig. 6), varying from 0.23-0.30 mm. in width, was set off only
slightly from the neck, which measured 0.14-0.30 mm. in length. The
suckers, ranging from 0.11-0.17 mm. in diameter, were shallow, situated
at an angle, and very close together.
The female genital primordia were indicated first in the 5th-6th, and
their ducts in the 6th-7th proglottids. The ovary (Text-fig. 13), averaging
about 0.60 mm. in breadth, reached maturity between the 60th-90th seg-
ments, and disappeared abruptly between the 90th-145th proglottids (Text-
fig. 19). The cirrus sac and, just below it, the vagina, opened into a shallow
1938J Arnold: Anoplocephaline Cestodes of N. Amer. Rabbits
39
Text-figures 18-23.
Abbreviations : c — cirrus sac. e — excretory duct, n — nerve fiber, o — ovary, p — pars prostatica.
r — seminal receptacle, s — shell gland, t — testis, u — uterus, v — vitelline gland, va — vagina, vd —
vas deferens, vs — seminal vesicle. All measurements are to a scale of 0.1 mm. All drawings are
camera lucida.
18. Gravid proglottid, Cittotaenia variabilis. 19. Gravid proglottid, Cittotaenia
pectinata americana. 20. Gravid proglottid, Cittotaenia denticulata. 21.
Gravid proglottid, Cittotaenia ctenoides. 22. Gravid proglottid, Cittotaenia
pectinata. 23. Gravid proglottid, Andrya cuniculi.
40
Zoologica: New York Zoological Society
[XXIII :2
depression or atrium in the posterior half of the proglottid. The vagina
continued mediad and very soon enlarged to form the seminal receptacle.
The latter, thin-walled and irregular in outline, led to the ovarian complex.
Ova ranged from 70-130 g in diameter, with an average of 80 y. The uterus
was typically single.
The male gential primordia appeared first in the 37th-45th segments,
and their ducts were visible one or two proglottids behind the primai’y con-
densation. The testes, about 125 per segment, 0.062-0.11 mm. in diameter,
were situated in the posterior half of the proglottid, between the longitudinal
excretory canals, posterior to the ovaries. The short vasa efferentia emptied
into the vas deferens, which joined the convoluted seminal vesicle located
median to the longitudinal excretory canal. The cirrus sac, 1.0-1.76 mm.
in length, was well developed and extended mediad from the longitudinal
excretory canal.
The following brief diagnosis is proposed.
Diagnosis : Cittotaenia pectinata americana Douthitt, 1915. Mature
specimens may contain 360 proglottids, with a length of 220 mm. and a
width of 7.5 mm. Scolex 0.23-0.30 mm. broad; neck 0.14-0.30 mm. long.
Female genital primordia appear in the 5th or 6th and their ducts in the
6th or 7th segments. Ovary, about 0.6 mm. in breadth, reaches maturity
in the 60th-90th and disappears between the 90th-145th segments. Ova
70-130 g in diameter. The male genital primordia appear first between the
37th-45th proglottids, and their ducts in each instance one -or two segments
later. Testes, 112-125 per proglottid, measure from 0.062-0.11 mm. in
diameter. They extend from one longitudinal excretory canal to the other,
in the posterior half of the segment. Cirrus sac, 1-1.76 mm. long, extends
mediad from the longitudinal excretory canal.
Hosts : Lepus calif ornicus melanotis and Lepus americanus.
Habitat : Nebraska, Kansas and western Canada.
After a careful comparison of C. pectinata americana, of North Amer-
ica, and C. pectinata, of Europe, it was felt that the only justifiable alloca-
tion of these species is to retain the former as a variety of the latter.
The chief difference between the two is that C. pectinata attains a length of
400 mm. with 190 proglottids, whereas C. pectinata americana, with a length
of 220 mm., has more than 360 segments. This difference is distinctive and
worth variety rank, considering that the specimens were from different
hosts and from different continents. On the other hand, careful comparison
affords no justification in separating them by more than variety status.
The writer, therefore, agrees with Douthitt (1915) that the American rep-
resentative of C. pectinata should be designated as a variety, C. pectinata
americana.
Citfotaenia variabilis (Stiles, 1895).
Stiles and Hassall, 1896.
(Text-figs. 2, 12, 18).
This species was described by Stiles in 1895 as Ctenotaenia variabilis.
The next year, Stiles and Hassall transferred it to the present genus because
of priority. Stiles felt proper classification of C. variabilis necessitated
breaking up the species into three varieties : C. variabilis variabilis, C. vari-
abilis angusta, and C. variabilis imbricata. These varieties were recog-
nized by Lyman (1902) but not by Hall (1908), Douthitt (1915) or John
(1926). Meggitt (1924) considered the varieties as synonyms of C. vari-
abilis.
In the present study some of Stiles’ original material of C. variabilis
and 97 other specimens of this species were studied. Maximal length of
1938J Arnold: Anoplocephaline Cestodes of N. Amer. Rabbits
41
strobilia was 450 mm., with as many as 750 proglottids, and maximal width
was 10.5 mm. The scolex, measuring 0.44-0.61 mm. in breadth, was typically
set off from the neck, although in a few contracted specimens there was no
line of demarcation (Text-fig. 2). The suckers, spherical in outline, varied
from 0.16-0.28 mm. in diameter and averaged 0.21 mm. The neck measured
from 0.26-0.84 mm. in length, with an average of 0.50 mm.
The female genital primordia appeared first between the 40th-50th and
their ducts between the 95th-105th segments (Text-fig. 12). The follicular
ovary measured from 0.48-0.71 mm. in width. The seminal receptacle, first
differentiated at about the 160th segment, lay median to the longitudinal
excretory canal, with its proximal end near the ovarian complex. Near its
origin, the oviduct was joined by the duct from the seminal receptacle and
the combined duct received shortly those from the yolk and shell glands.
The tubular uterus then proceeded to the saccular uterus which was either
single or double. The opening of the vagina was immediately below that of
the cirrus sac. The ova ranged from 52-68 /x in diameter, with an average
of 64 /x (Text-fig. 18).
The first appearance of the male genital primordia was thought to be in
the vicinity of the 125th proglottid, although their ducts were not located
definitely until the 175th. The testes were confined between the two ovaries
in the distal half of the segment. They numbered between 60-135 and ranged
from 0.053-0.071 mm. in diameter, with an average of 0.062 mm. The short
vasa efferentia emptied into the vas deferens which proceeded to the highly
convoluted seminal vesicle, lying for the most part median to the longitudinal
excretory canal. Distally the vesicle joined the small cirrus sac (0.32-0.45
mm. long, averaging 0.38 mm.) lying lateral to the longitudinal excretory
canal. The cirrus sac, testes, ovary, and female ducts degenerate in ripe
proglottids.
From the above data the following diagnosis may be given.
Diagnosis : Cittotaenia variabilis (Stiles, 1895), Stiles and Hassall,
1896. Mature specimens with as many as 750 proglottids attain a maximal
length of 450 mm. and a maximal breadth of 10.5 mm. Scolex 0.44-0.61 mm.
in width; may or may not be set off from the neck. Suckers 0.16-0.28 mm.
in diameter, average 0.21 mm. Neck well defined; length 0.26-0.84 mm.,
average 0.50 mm. Female genital primordia appear between the 40th-50th
and their ducts between the 95th-105th proglottids. Ovary measures 0.43-
0.71 mm. in width, with an average of 0.58 mm. Ova 52-68 /x in diameter,
with an average of 64 /x. Male genital primordia appear about the 125th
proglottid and their ducts at the 175th. Testes (60-135) in the distal half
of the proglottid between the two ovaries ; diameter 0.053-0.071 mm. Cirrus
sac small, 0.32-0.45 mm. long, lateral to longitudinal excretory canal.
Hosts : Sylvilagus floridanus mallurus, S. floridanus alacer and S.
palustris.
Habitat : New York, Maryland, Kansas and Pennsylvania.
6. Correlation of Host-Parasite Relations.
In the present survey, 179 cottontail rabbits were examined. For
identification the systematic arrangement of Nelson (1909) was employed.
Sylvilagus floridanus mallurus was obtained from New York and Pennsyl-
vania, and S. floridanus alacer from Kansas. Cittotaenia variabilis (Stiles,
1895) was the only cestode recovered. The parasites were usually found
near the middle of the small intestine, although more anteriorly in a few
cases.
Tables I and II give the data on infection in the rabbits studied.
42 Zoologica: New York Zoological Society [XXIII :2
TABLE I. Sylvilagus floridanus mallurus.
Locality.
No.
Examined.
No.
Infected.
%
Infected.
Lake Mahopac, N. Y.
3
2
66%
Pinebush, N. Y.
3
1
33%
Carmel, N. Y.
1
1
100
New York, N. Y.
2
1
50
Northern New York.
7
6
85.9
Boylesville, Pa.
7
2
28.5
Total.
23
13
56.5
TABLE II. Sylvilagus floridanus alacer.
Locality.
No.
Examined.
No.
Infected.
%
Infected.
Wichita, Kan.
1932
28
19
67.8
Wichita, Kan.
1933
50
19
38
Wichita, Kan.
1934
78
32
41
Total.
156
70
44.8
Cittotaenia pectinata americana (Douthitt, 1915) is found only in hares,
namely, Lepus calif ornicus melanotis Mearns and Lepus americanus Erx-
leben. Cittotaenia perplexa (Stiles, 1895) and C. variabilis (Stiles, 1895)
are found only in cottontail rabbits. Both species are present in Sylvilagus
floridanus mallurus (Thomas) and Sylvilagiis floridanus alacer (Bangs),
while only Cittotaenia perplexa has been found in Sylvilagus nuttalli pinetis
(Allen) and only C. variabilis in Sylvilagus palustris (Bachman).
7. Anoplocephaline Cestodes from European Hares and Rabbits.
Cittotaenia denticulata (Rudolphi, 1804).
Stiles and Hassall, 1896.
(Text-figs. 4, 8, 14, 20).
C. denticulata, type species of Cittotaenia, was first described by
Rudolphi as Taenia denticulata. The original description was expanded in
later papers (Rudolphi, 1805, 1810). The same species was subsequently
described by Baird (1853) as Taenia goezei.
The next important work is that of Riehm (1881a), who described
this worm first as Cittotaenia latissima, gen. nov., sp., nov., and then as a
species of Dipylidium. Dipylidium latissima was transferred to the genus
Taenia by Neumann in 1888.
Taenia goezei Baird was reduced to synonymy with D. latissima Riehm
by Blanchard (1891) on reexamination of Baird’s original material.
Riehm’s D. latissima was placed in the genus Ctenotaenia by Railliet in
1893.
Stiles and Hassall (1896), upon study of the original specimens of
Rudolphi’s Taenia denticulata, Baird’s Taenia goezei, and Riehm’s Dinv-
1938] Arnold: Anoplocephaline Cestodes of N. Amer. Rabbits
43
lidium latissima, declared that they were all the same species which they
recognized as Cittotaenia denticulata. Later descriptions of C. denticulata
include those of Stiles (1896), John (1926), Baer (1927) and Sprehn (1932).
In the present study, 19 specimens of Cittotaenia denticulata were ex-
amined. They ranged from minute, immature forms to large, fully matured
cestodes with as many as 260 proglottids. They measured 0.149-260 mm.
in length and 8.5 mm. in maximal breadth (Text-fig. 4). The solices of
the immature forms (wonns 0.149-10.5 mm. in length) ranged from 0.10-
0.50 mm. in width; those of mature worms (21-260 mm. in length) varied
from 0.57-0.76 mm. in breadth. The scolex appeared to be rectangular in
cross section. There was a short broad neck, not previously reported,
which measured from 0.21-0.92 mm., depending on the degree of contrac-
tion. The suckers ranged from 0.23-0.30 mm. in diameter, averaging 0.27
mm. The sucker size was in close agreement with that found by other
workers.
The female genital primordia were first seen between the 10th-15th, and
their ducts between the 17th-40th proglottids. The ovarian follicles were
not well developed until the 80th-100th proglottid (Text-fig. 14). The
follicles continued to increase in size and then suddenly disappeared between
the 135th-175th segments. The ovary measured from 0.32-1.42 mm. in width
and averaged 0.97 mm. The disappearance of the ovary is correlated with
the presence of eggs in the uterus (Text-fig. 20). The ovaries were found
in the posterior two-thirds of the proglottid and just median of the longi-
tudinal excretory canals. The oviduct originated about the middle of the
ovarian mass, and after a short distance was joined by the short duct from
the bulbous seminal receptacle. The combined duct then passed to the shell
gland, where it was joined by the vitelline duct. The tubular uterus then
proceeded anteriad to the saccular uterus. The ova measured from 46-75 p.
in diameter and averaged 61 p (Text-fig. 8).
The male genital primordia were first seen between the 35th-60th, their
ducts, between the 36th-70th proglottids. The testes increased in number
very rapidly after the first appearance of their primordia. The follicles
extended between the longitudinal excretory canals and were scattered be-
tween the anterior and posterior limits of the proglottid. The testes totaled
between 225 and 250, a number very much greater than that previously
recorded. Baer and Sprehn reported that there were 100 testes. Stiles and
John merely stated that they were numerous. However, Stiles (1896)
gave a drawing of one-half of a mature segment of C. denticidata. The
writer counted the testes represented in that half segment and found that
there were 124. Since Stiles worked on the original material of Rudolphi,
Baird and Riehm, and his drawing was apparently taken from that mate-
rial, it seems safe to assume that the number of testes is nearer 225-250
than 100. The vasa efferentia emptied on either side of the proglottid into
a vas deferens which proceeded to the cirrus sac containing the seminal
vesicle and a large cirrus. The cirrus sac measured from 0.50-0.97 mm.
and averaged 0.77 mm. in length, which was within the range given by
other workers. The testes ranged from 0.041-0.12 mm. and averaged 0.073
mm. in diameter. This range is much smaller than that given by other
authors, but it is inclusive of some of the higher values found by others.
The testes tend to disappear in ripe proglottids.
From the above description the following diagnosis may be drawn.
Diagnosis : Cittotaenia denticulata (Rudolphi, 1804) Stiles and Hassall,
1896. Type of the genus. Mature specimens range 20-800 mm. in length,
with as many as 300 proglottids, and from 8-15 mm. in maximal breadth.
Scolex 0.57-1.18 mm. wide; 0.43-1 mm. long. Suckers 0.2-0. 3 mm. in
diameter. The genital primordia appear early. Those of the female system
are found between the 10th-15th proglottids, and their ducts between the
17th-40th proglottids. Ovary, 0.32-1.42 mm. in width, becomes well devel-
44
Zoologica: Nerv York Zoological Society
[XXIII :2
oped at about the 84th-100th and disappears about the 135th-170th seg-
ment. Male genital primordia appear first between the 35th-60th and their
ducts between the 36th-70th segments. Testes range from 225-250 in num-
ber in mature proglottids, and vary between 0.041-0.12 mm. in diameter.
Cirrus sac measures from 0.50-1.12 mm. in length and 0.26-0.3 mm. in
width. Ova measure 46-75 y in diameter.
Host: Oryctolagus cuniculus.
Habitat: Europe.
C iitotaenia ctenoides (Riehm, 1881).
Stiles and Hassall, 1896.
(Text-figs. 9, 15, 21).
Cittotaenia ctenoides was first described as Dipylidium leuckarti by
Riehm in 1881. Later it was transferred to the genus Taenia by Neumann
(1888) and in 1892 it was replaced in the genus Dipylidium by the same
author.
In 1890, Railliet changed the specific name from Taenia leuckarti to
Taenia ctenoides. R. Blanchard (1891) placed Dipylidium leuckarti Riehm
in the genus Moniezia. Railliet (1893) erected the genus Ctenotaenia and
placed the species Taenia ctenoides in it. The synonymy of this genus with
Cittotaenia was pointed out by Stiles and Hassall (1896). Later accounts of
this species were given by Stiles (1896) (who studied Riehm’s type speci-
mens and others from Blanchard’s collection), by Baer (1927), and by
Sprehn (1932).
The writer examined 100 specimens of C. ctenoides, ranging from 6-460
mm. in length and from 1-10.5 mm. in width. The maximal length found
was less than that previously reported, while the maximal width was some-
what greater. The scolex was small, although wider than the neck region
and proximal proglottids. It measured 0.32-0.48 mm. in width (Text-fig. 9).
This range is smaller than that reported by Stiles, Baer or Sprehn, although
the higher value was in agreement. The suckers measured 0.12-0.25 mm. in
diameter, with an average of 0.16 mm. There was a short broad neck
which varied from 0.25-0.42 mm. and averaged 0.32 mm. in length. The
neck was not measured by Stiles, and Baer and Sprehn did not mention it.
The proglottids reached a number greater than 560. Stiles reported a
maximum of 750 proglottids. The original, terminal 10-15 proglottids, when
present, were sterile.
The female genital primordia appeared in the first few proglottids, but
their ducts were not distinct until the 45th-50th segments. A well developed
follicular ovary was first seen in the 130th-160th segments (Text-fig. 15).
The ovary then continued to develop and later suddenly disappeared between
the 183rd-232nd proglottids. This disappearance of the ovary was corre-
lated with the presence of eggs in the uterus. The ovary measured from
0.40-0.88 mm. in width and averaged 0.61 mm. The oviduct arose posteriorly
from the ovary and was joined almost immediately by the duct from the
seminal receptacle. The convoluted oviduct proceeded posteriad to join the
rather long vitelline duct, in the shell gland mass. The tubular uterus then
proceeded anteriad and slightly laterad for a short distance, where it en-
larged into the saccular uterus (Text-fig. 21). The ova ranged from 62-69 y,
and averaged 66 y in diameter. This figure is slightly higher than those
given by Stiles, Baer, and Sprehn. The seminal receptacle was not clearly
delimited. The vagina opened to the exterior immediately below the
opening of the cirrus sac. It proceeded as a small tube mediad and slightly
posteriad for a short distance and then expanded to form the seminal re-
ceptacle. The seminal receptacle gradually narrowed to a duct which
proceeded anteriad and mediad to join the oviduct.
1938] Arnold: Anoplocephaline Cestodes of N. Amer. Rabbits
45
The male genital primordia appeared first between the 85th-100th seg-
ments, and the ducts were found a few proglottids farther back. The testes
were divided into two groups, one on either side of the proglottid, posterior
to the ovaries and median to the longitudinal excretory canal. The testes
numbered from 35-75 in each group, and the two groups contained approxi-
mately the same number of follicles. The number of testes found was
less than that given previously by other workers. The testes varied from
0.046-0.081 mm. and averaged 0.056 mm. in diameter. The short vasa
efferentia joined the vas deferens which proceeded anteriad and laterad to
the cirrus sac. The small lateral cirrus sac, containing the seminal vesicle
and a well developed cirrus, varied from 0.16-0.25 mm. in length, with an
average of 0.23 mm. These measurements are slightly higher than those
reported by Stiles, Baer and Sprehn.
The following brief diagnosis is suggested.
Diagnosis : Cittotaenia ctenoides (Railliet, 1890) Stiles and Hassall,
1896. Strobila, of 500-750 proglottids, may attain a length of 800 mm. and
a maximal breadth of 10.5 mm. Scolex, 0.32-0.50 mm. wide, is broader than
the neck and proximal proglottids. Suckers 0.12-0.25 mm. in diameter.
Female reproductive primordia appear very soon after segmentation begins;
their ducts at 45th-50th segments. Ovary, 0.40-0.88 mm. wide, develops
between 130th-160th, and disappears about 183rd-232nd proglottids. Ova
62-69 /x in diameter. Male reproductive primordia appear first at 85th-100th
segments and their ducts very soon thereafter. Testes, 0.046-0.081 mm. in
diameter, arranged in two groups (35-80 in each). Cirrus sac 0.16-0.25 mm.
long and 0.138-0.23 mm. wide.
Hosts: Oryctolagus cuniculus and Oryctolagus cuniculus domesticus.
Habitat: Europe.
Cittotaenia pectinate (Goeze, 1782).
Stiles and Hassall, 1896.
(Text-figs. 7, 16, 22).
Cittotaenia pectinata was first described by Goeze (1782) as Taenia
pectinata from hares and wild rabbits. His description was vague and in-
adequate, and pertained to material of more than one species.
Zeder (1800) transferred the species to the genus Alyselminthes. His
description was supposed to cover T. pectinata, but subsequent workers feel
he actually studied Andrya rhopalocephala. In 1803, Zeder placed T. pecti-
nata in the genus Halysis.
Rudolphi (1810), Bremser (1824), and Diesing (1850) studied T.
pectinata but their characterizations for this species were broad enough to
include several species.
Riehm (1881) was the first to study the species in detail. After recti-
fying previous errors, he placed T. pectinata in the genus Dipylidium. In
1891, Blanchard reported this species at Brianqon.
The species was included in the genus Ctenotaenia by Railliet in 1893,
but Stiles and Hassall (1896) transferred it to the genus Cittotaenia.
Stiles (1896) obtained some of Riehm’s original material and elabo-
rated Riehm’s diagnosis. Stiles summarized the works of previous authors
and gave a more complete description.
Douthitt (1915) stated that C. pectinata was found only in Europe.
This allocation was followed by Meggitt (1924), John (1926), and Rees
(1933a).
Baer (1927) claimed that C. pectinata was very extensively distributed.
He considered all American representatives of the genus Cittotaenia in
46 Zoologica : Neiv York Zoological Society [XXIII :2
rabbits to be Cittotaenia pectinata. Sprehn (1932) followed Baer in his
consideration of this species.
The author studied 29 specimens of Cittotaenia pectinata (Goeze, 1782),
varying from 0.97-84 mm. in length and from 0.17-8 mm. in maximal
breadth. Stiles reported a greater length than this, 400 mm. Blanchard
found a width of 10 mm. The small scolex was rarely wider than the
proglottids immediately following it. The scolex measured from 0.20-0.35
mm. in width, with an average of 0.28 mm. (Text-fig. 7). The suckers were
set at an angle and were very close together. They ranged from 0.071-0.15
mm. in diameter, with an average of 0.12 mm.. The neck region measured
from 0.10-0.35 mm. in length and averaged 0.20 mm. The proglottids were
always much broader than long and numbered as high as 190. The terminal
proglottids were found to be sterile, an anoplocephaline characteristic.
The female genital primordia appeared by the 8th proglottid, while
their canals were found between the 10th-15th segments. The ovaries,
measured only in mature proglottids, ranged from 0.21-0.67 mm. in width,
with an average of 0.48 mm. (Text-fig. 16). The ovary developed steadily
from the primordial stage to that found in fully mature proglottids, and
then disappeared rather abruptly between the 75th-150th proglottids. The
follicles of the ovary were not well developed until the 45th-60th proglottids.
The oviduct arose medially from within the mass of the ovary and then
proceeded anteriad. Within a short distance it joined the duct from the
seminal receptacle. The oviduct then proceeded posteriad and soon met the
ducts from the vitelline and shell glands. The convoluted tubular uterus
passed anteriad to meet the saccular uterus near its posterior limits. The
uterus was either double or the two uteri became confluent in the middle
of the segment to form a single continuous one. The former type was the
one commonly encountered (Text-fig. 22). The ova ranged from 62-67 g
in diameter and averaged 65 g.
The male and female reproductive systems opened to the exterior by a
common genital pore, or by two pores which were situated exceedingly close
together. The vagina opened to the exterior below the cirrus sac, or into
the cirrus sac just at its opening to the outside. The latter condition
seemed the more common. The vagina soon expanded into the seminal
receptacle, a large, thin-walled structure, which was closely applied to the
cirrus sac and seminal vesicle. Medially, the seminal receptacle became a
duct that led to the oviduct.
The male genital primordia first appeared between the 25th-45th
proglottids, or 1.4-5. 6 mm. from the anterior end, and the male ducts be-
tween the 27th-47th segments. The testes, in mature proglottids, ranged
from 80-160 in number, and averaged 130. They lay along the posterior
border of the proglottid, between the longitudinal excretory canals, below
the distal half of the ovary. They measured from 0.043-0.077 mm. in
diameter, with an average of 0.06 mm. The vasa efferentia emptied shortly
into the vas deferens, which proceeded to join the seminal vesicle near
the anterior end of the proglottid. The seminal vesicle was much con-
voluted, and medial to the longitudinal excretory canal. The glandular
cells of the pars prostatica surrounded the seminal vesicle. The cirrus sac
was thick-walled, muscular and contained a well developed cirrus. The
cirrus sac ranged from 0.36-1.03 mm. in length and averaged 0.69 mm. The
cirrus sac and the seminal vesicle lay at a slight angle, and the genital pores
(one on either side of the segment) were situated in the posterior half
of the proglottid.
From the literature and the study of both whole mounts and sections,
the specific diagnosis, given below, has been derived.
Diagnosis'. Cittotaenia pectinata (Goeze, 1782) Stiles and Hassall,
1896. Mature specimens 400 mm. long and 10 mm. wide. Scolex 0.20-0.35
mm. wide. Broad, short neck measures 0.10-0.15 mm. in diameter. Female
1938] Arnold: Anoplocephaline Cestodes of N. Amer. Rabbits
47
genital primordia appear by 8th segment, their ducts, by 10th-15th segment.
Ovary, 0.21-0.67 mm. wide, disappears between 75th-150th segments. Ova
54-90 /x in diameter. Male genital primordia appear first between the 25th-
45th, their ducts between 27th-47th segments. Testes, 80-160 in number,
0.043-0.077 mm. in diameter, lying in distal half of proglottid, extend from
one longitudinal excretory canal to other. Cirrus sac, 0.36-1.03 mm. in
length, extends mediad beyond the longitudinal excretory canal.
Hosts: Oryctolagus cuniculus, Lepus europaeus, Lepus timidus, Lepus
variabilis and Marmota marmota.
Habitat : Asia and Europe (Germany, France and England).
Andrya euniculi (R. Blanchard, 1891).
Railliet, 1893.
(Text-figs. 5, 17, 23).
Andrya euniculi, first described by Riehm (1881) as Taenia rhopalioce-
phala, was placed in the genus Anoplocephala in 1891 by Blanchard when he
changed the specific name to euniculi. In regard to this transfer, he stated :
“A. euniculi R. Blanchard, 1891 ( Taenia pectinata Goze, 1782, pro parte;
T. rhopaliocephala Riehm, 1881). — Chez le Lapin de garenne. Le nom pro-
pose par Riehm ne peut pas etre conserve, en raison de son identite avec le
nom de Taenia rhopalocephala; ce dernier seul est valable : il a la priorite,
puisque la ver auquel il s’applique est decrit par Riehm quelques pages avant
le Taenia rhopaliocephala (3), et d’ailleurs est le seul dont le nom soit
correctment forme.”
Railliet (1893) erected the genus Andrya to contain A. rhopalocephala
and A. euniculi. Stiles (1896) restudied Riehm’s original material and gave
his own characterization of this species. He stated that he has never recov-
ered this parasite from rabbits in North America. Douthitt (1915) and
Meggitt (1924) both recognized this species as valid, but gave no further
characterization of it. Baer (1927) and Sprehn (1932) studied this form,
but gave nothing new in their monographs.
The present writer studied 38 specimens of Andrya euniculi. They
ranged from 13-325 mm. in length. The maximal breadth obtained was
3.5 mm. These measurements were only about one-third the maximum length
and width previously reported. The scolex was large, varying from 0.40-
0.67 mm. in diameter. It was definitely set off from a well differentiated
neck (Text-fig. 5). The latter measured from 0.80-1.42 mm. in length. The
large suckers ranged from 0.16-0.30 mm. in diameter, with an average of
0.25 mm. The immature segments were much broader than long, while the
mature proglottids tended toward a quadrate condition. The segments
numbered as many as 500. Stiles (1896) reported them to be as many
as 800. He undoubtedly had more mature specimens than were studied in
this survey. The original 2-7 terminal proglottids were sterile.
The reproductive organs were single. Their genital openings were ir-
regularly alternate on right and left lateral margins (Text-fig. 17), but
opened more often on one side than on the other, as previously observed by
Stiles. The female genital primordia appeared first about the 5th-10th prog-
lottids, and their ducts between the 95th-110th segments. The ovarian fol-
licles were not well developed until the 225th-275th proglottids. The ovary
measured from 0.43-0.50 mm. in width, and was median or on the poral
side of the segment. The vagina was distal to the cirrus sac and its openings.
It extended mediad to the longitudinal excretory canal and enlarged into
the seminal receptacle. The latter extended to the ovarian mass, where it
joined the oviduct. The latter duct, upon uniting with the yolk and shell
glands, gave rise to the uterus (Text-fig. 23). Stiles (1896) stated that the
48
Zoologica: New York Zoological Society
[XXIII :2
latter was characteristically a network structure. The ova measured from
49-57 /x in diameter, which was within the range given by Stiles (1896).
The male genital primordia were first evident about the 45th segment,
and their ducts at the 110th segment. The testes extended between the
two longitudinal excretory canals, and the anterior and posterior limits of
the proglottid. They practically surrounded the ovary, although they were
more predominantly distributed on the side opposite the genital pore. They
numbered between 70-90 and varied from 0.049-0.065 mm. in diameter. The
vasa efferentia very soon joined the vas deferens. Just before the latter
emptied into the cirrus sac, it united with the pars prostatica. The cirrus
sac was club-shaped, extended mediad of the longitudinal excretory canal
and contained a definite seminal vesicle. The cirrus sac measured from
0.27-0.43 mm. in length. This range is lower than that given by Stiles, but
may be explained by the fact that the specimens used in this study were not
as mature (long) as those used by Stiles. There seemed to be a tendency
for the formation of a genital papilla, which contained both male and
female genital openings.
Summarizing the above surveys, the following diagnosis is derived.
Diagnosis : Andrya cuniculi (R. Blanchard, 1891) Railliet, 1893.
Mature specimens, with as many as 800 segments, 1,000 mm. long and 8 mm.
wide. Scolex 0.4-0.67 mm. wide. Well differentiated neck, 0.80-1.42 mm.
long. Suckers 0.16-0.30 mm. in diameter, with an average of 0.25 mm.
Female genital primordia appear first in 5th-10th and their ducts in 95th-
110th proglottids. Ovary 0.43-0.50 mm. wide. Ova 49-57 y in diameter.
Male genital primordia appear first in about 45th proglottid and their
ducts in 110th. Testes in median field, 70-90 per segment, varying from
0.049-0.065 mm. in diameter. Cirrus sac, 0.27-0.8 mm. long, extends mediad
of the longitudinal excretory canal.
Hosts: Lepus timidus and Oryctolagus cuniculus.
Habitat : Europe.
8. Comparison of European and American Anoplocephalidae of Rabbits.
The members of the Anoplocephalidae from North American rabbits
show constant differences from those of European rabbits.
Schizotaenia americana (Stiles, 1895) and Andrya cuniculi (R. Blan-
chard, 1891) are immediately distinguished from the several species of
rabbit cestodes of the genus Cittotaenia by the fact that they possess only
one set of genital organs to a proglottid. S. americana differs from A.
cuniculi in regularity of alternation of the genital pores, length and width
of body, lack of neck, number of proglottids, size of ovary, first appearance
of male genital primordia and ducts, and number of testes.
Cittotaenia denticulata (Rudolphi, 1804) differs from C. ctenoides
(Railliet, 1890) in size of scolex, number of proglottids, earliest appear-
ance of female genital ducts, position of ovarian maturity and disappear-
ance, first indication of male genital primordia and ducts, number of testes
and their distribution, and size of cirrus sac. It is unlike C. pectinata
(Goeze, 1782) in maximal length, number of proglottids, first appearance
of female genital primordia and ducts, position of ovarian maturity, number
and distribution of testes, and location of cirrus sac. C. perplexa (Stiles,
1895) is different from C. denticulata in maximal length, size of scolex,
diameter of sucker, number of proglottids, appearance of ovarian follicles,
number and distribution of testes, and location of cirrus sac. C. denti-
culata is dissimilar to C. pectinata americana Douthitt, 1915, in maximal
length, maximal width, size of scolex, size of sucker, number and distribu-
tion of testes, and position of cirrus sac. C. denticulata is unlike C. varia-
bilis (Stiles, 1895) in maximal length and width, size of neck, number of
1938] Arnold: Anoplocephaline Cestodes of N. Amer. Rabbits 49
proglottids, position of genital primordia and ducts of both male and
female reproductive systems, number and distribution of testes, and size
of cirrus sac.
C. ctenoides is unlike C. denticulata in scolex size, sucker diameter,
proglottid number, first indication of female reproductive ducts, site of
ovarian maturity and disappearance, earliest appearance of male genital
primordia and ducts, number and distribution of testes, and size of cirrus
sac. C. ctenoides differs from C. pectinata in maximal length, number of
proglottids, first indication of female ducts, ovarian maturity and disap-
pearance, location of male genital primordia and ducts, distribution of
testes, and length and position of cirrus sac. C. perplexa is distinct from
C. ctenoides in maximal length, size of suckers, number of proglottids, site
of female genital duct appearance, maturity and disappearance of ovary,
first appearance of male genital primordia and ducts, and size of cirrus sac.
C. ctenoides is dissimilar to C. pectinata americana in maximal length and
width, size of scolex, number of proglottids, first indications of female
genital primordia, first appearance of male genital primordia and ducts,
maturity and disappearance of ovary, distribution and size of testes, and
position of cirrus sac. C. variabilis is different from C. ctenoides in
maximal length, appearance of genital primordia and ducts of both male
and female reproductive systems, distribution of testes, and length and
position of cirrus sac.
C. pectinata differs from C. denticulata in maximal length, scolex width,
sucker diameter, number of proglottids, location of ovarian maturity, testes
number and distribution, and cirrus sac size. C. pectinata is unlike C.
ctenoides in maximal size, number of proglottids, time of appearance of
genital primordia and of ducts of both male and female reproductive sys-
tems, maturity and disintegration of ovary, distribution of testes and size
and location of cirrus sac. The comparison of C. pectinata with the next
three American species is of especial interest, since Baer (1927) and Sprehn
(1932) considered them identical with C. pectinata. The latter is distinct
from C. perplexa in maximal length, first appearance of female genital
primordia and their ducts, site of first ovarian follicular development, and
distribution of testes in mature proglottids. C. pectinata has a variant in
C. pectinata americana but differs from it in maximal length and breadth
and number of proglottids. C. pectinata is unlike C. variabilis in size of
scolex, length of neck, size of suckers, number of proglottids, appearance
of primordia and ducts of both male and female reproductive systems, dis-
tribution of testes, and length and location of cirrus sac.
C. perplexa differs from C. denticulata in maximal length, scolex width,
sucker diameter, proglottid count, site of earliest complete follicular develop-
ment and disintegration, testes number and distribution, and cirrus sac
location. C. perplexa is different from C. ctenoides in maximal length,
sucker size, number of proglottids, first appearance of female genital
canals, location of ovarian follicular maturity and disappearance, earliest
appearance of male genital primordia and ducts, and cirrus sac position.
C. perplexa is unlike C. pectinata in maximal length, earliest appearance of
female genital primordia and their ducts, location of first ovarian follicular
development, and distribution of testes in mature segments. C. perplexa is
dissimilar to C. pectinata americana in maximal length and breadth, maxi-
mal number of proglottids, earliest appearance of female ducts, first ap-
pearance of the male genital primordia and ducts, testes distribution, and
cirrus sac characteristics. C. perplexa is distinct from C. variabilis in
maximal length, scolex size, length of neck, sucker size, number of proglot-
tids, place of appearance of genital primordia and ducts of both male and
female reproductive systems, distribution of testes, and size and location
of cirrus sac.
C. pectinata americana is unlike C. denticulata in maximal length, size
of scolex and suckers, number of proglottids, number and distribution of
50
Zoologica: Neiv York Zoological Society
[XXIII :2
testes, and cirus sac characteristics. C. pectinata americana differs from
C. ctenoides in maximal length and breadth, number of proglottids, first
appearance of female genital ducts, first appearance of male genital primor-
dia and ducts, site of ovarian maturity and disappearance, testes distribu-
tion, and cirrus sac size and position. C. pectinata is different from C. pecti-
nata americana, its variety, in maximal length and breadth, and number of
proglottids. C. pectinata americana is unlike C. perplexa in maximal length
and breadth, number of proglottids, first appearance of genital primordia
and ducts of both male and female reproductive systems, testes distribu-
tion, and cirrus sac size. C. pectinata americana is distinct from C. varia-
bilis in maximal length and breadth, size of scolex, number of proglottids,
first appearance of genital primordia and of ducts of both male and female
reproductive systems, testes distribution, and cirrus sac size and position.
C. variabilis is divergent from C. denticulata in maximal length, length
of neck, number of proglottids, appearance of genital primordia and of ducts
of both male and female reproductive systems, number and distribution of
testes, and size of cirrus sac. C. ctenoides is distinct from C. variabilis in
maximal length, length of neck, earliest appearance of male and female
genital primordia and ducts, testes, distribution, and cirrus sac size. C.
variabilis is dissimilar to C. pectinata in scolex width, length of neck, sucker
size, segment count, appearance of male and female genital primordia and
ducts, testes distribution, and cirrus sac length. C. variabilis is unlike
C. perplexa in maximal length, size of scolex, length of neck, size of suckers,
site of appearance of genital primordia and ducts of both male and female
reproductive systems, distribution of testes, and length of position of
cirrus sac. C. variabilis is unlike C. pectinata americana in maximal length
and breadth, width of scolex, number of proglottids, earliest appearance of
genital primordia and ducts of both male and female reproductive systems,
distribution of testes, and size and position of cirrus sac.
9. Key to Species.
Contrary to the findings of Baer (1927) and Sprehn (1932), the ano-
plocephaline species of leporine cestodes in North America and Europe are
separate and distinct. Representatives of three genera of the Anoploce-
phalidae have been found in hares and rabbits. The genus Schizotaenia
Janicki, 1906, is represented by S. americana (Stiles, 1895) and this species
is found only in North America. The genus Andrya Railliet, 1893, is repre-
sented by A. cuniculi (R. Blanchard, 1891) and is found only in Europe.
Finally the genus Cittotaenia Riehm, 1881, has members in both North
America and Europe and contains six species. They include C. denticulata
(Rudolphi, 1804), C. ctenoides (Railliet, 1890), and C. pectinata (Goeze,
1782), all found in Europe, and C. perplexa (Stiles, 1895), C. variabilis
(Stiles, 1895), and C. pectinata americana Douthitt, 1915, all found in
North America. C. pectinata and C. pectinata americana are parasitic prin-
cipally in hares ( Lepus ), whereas the other species of Cittotaenia occur
chiefly in rabbits ( Oryctolagus and Sylvilagus) .
The Anoplocephalidae of rabbits may be distinguished by the following
key:
1. (2) Two sets of reproductive organs to a segment. (5)
2. (1) One set of reproductive organs to a segment. (3)
3. (4) Genital pores alternating regularly to right and left lateral margins.
Schizotaenia americana.
4. (3) Genital pores alternating irregularly to right and left lateral mar-
gins. Andrya cuniculi.
5. (6) Cirrus sac extending medial of longitudinal excretory canal. (10).
6. (5) Cirrus sac lying lateral of longitudinal excretory canal. (7).
1938J Arnold: Anoplocephaline Cestodes of N. Amer. Rabbits
51
7. (8) Maximal number of proglottids — 300; first appearance of female
genital ducts between 17th-40th proglottids; earliest indication of
male genital primordia and ducts from the 35th-60th and 36th-70th
proglottids, respectively. Cittotaenia denticulata.
8. (9) Maximal number of proglottids — 750; first appearance of female
genital ducts between 45th-50th segments; first indication of male
genital primordia and ducts between 75th-100th and 76th-105th
proglottids, respectively. Cittotaenia ctenoides.
9. (7) Maximal number of proglottids — 750; female ducts indicated first
between the 95th-105th segments; male genital primordia and ducts
originating at 125th-175th proglottids, respectively; well defined
neck present. Cittotaenia variabilis.
10. (11) Cirrus sac just barely crossing longitudinal excretory canal, and
with a maximal length of 0.64 mm.; maximal body length, 100 mm.
Cittotaenia perplexa.
11. (12) Cirrus sac extending considerably mediad of excretory canal, and
with a maximal length of 1.03 mm. ; proglottids up to 190 ; maximal
body length, 400 mm. Cittotaenia pectinata.
12. (10) Cirrus sac extending much mediad of longitudinal excretory canal;
cirrus sac with a maximal length of 1.8 mm.; proglottid number,
more than 360; maximal body length, 220 mm.
Cittotaenia pectinata americana.
10. Literature Cited.
Baer, J. G.
1927. Monographic des Cestodes de la Famille des Anoplocephalidae. Sup-
plements a u Bulletin Biologique de France et de Belgique, Suppl. X.,
4 pis.
Baird, W.
1853. Catalogue of the species of Entozoa or Intestinal Worms contained in
the Collection of the British Museum. London. 132 pp.
Blanchard, R.
1891. Notices Helminthologique. (Deuxieme serie.) Mem. de la Soc. Zool. de
France. IV: 420-489; figs. 1-38.
Boughton, R. V.
1932. The Influence of Helminth Parasitism on the Abundance of the Snow-
shoe Rabbit in Western Canada. Can. Jour. Res., 7 : 524-547.
Bremser, J. G.
1824. leones Helminthum Systema Rudolphii Entozoologicum illustrantes.
Viennae.
Cohn, L.
1906. Zur Anatomie zweier Cestoden. Centralbl. f. Bakt. 1, Abt., orig., 40:
362-367.
Curtice, C.
1888. Early stages in the life of Taenia pectinata. Science, March 23, 1888.
1892. Parasites. Jour. Comp. Med. & Vet. Arch., 13: New York; 223-236.
Diesing, K. M.
1850. Systema Helminthum. Vol. 1. Vindobonae. 679 pp.
1856. Neunzehn Arten von Trematoden. Kenkschriften d. k. Akad. d.
Wissensch. mathem. Naturw. CI. X. Bd.
52
Zoologica: Neiv York Zoological Society
[XXIII :2
Douthitt, H.
1915. Studies on the Cestode Family Anoplocephalidae. Illinois Biol. Monogr.,
1: 351-446; 6 pis.
Fuhrmann, O.
1907. Die Systematik der Ordnung der Cyclophyllidea. Zool. Anz., 32:
289-297.
1931. Kiikenthal-Krumbach-Handbuch der Zoologie, Cestoidea. Walter De
Gruyter Co., Berlin and Leipzig, 141-416.
Goeze, J. A. E.
1782. Versuch einer Naturgeschichte der Eingeweidewiirmer thierischer
Korper. Blankenburg, 471 pp., 35 pis.
Hall, M. C.
1908. A new rabbit cestode, Cittotaenia mosaica. Proc. U. S. Nat. Mus., 34:
691-699.
1912. The Parasite Fauna of Colorado. Col. Coll. Pub., Sci. Ser., 12, no. 10
Janicki, C. von.
1906. Studien an Saugetiercestoden. Zeitschr. f. wiss. Zool., 81 : 505-597.
John, D. D.
1926. On Cittotaenia denticulata (Rudolphi, 1804), with Notes as to the
Occurrence of other Helminthic Parasites of Rabbits Found in the
Aberystwyth Area. Parasit., 18: 436-454; 4 figs.
Kholodkovsky, N. A.
1902. Contribution a la Connaissance des Tenias des Ruminants. Arch.
Parasit., 6: 145-148.
Leidy, J.
1855. Notices on some Tapeworms. Proc. Acad. Nat. Sci. Phila., 7 : 433.
Lyman, R. A.
1902. Studies on the genus Cittotaenia. Trans. Amer. Micro. Soc. 23: 173-190.
Meggitt, F. J.
1924. The Cestodes of Mammals. 282 pp. Pub. Edward Goldston, London.
Nelson, E. W.
1909. The Rabbits of North America. U. S. Dept, of Agri., 314 pp., 13 pis.
Neumann, L. G.
1888. Traite des maladies parasitaires non microbiennes des animaux
domestiques. Paris. 673 pp., 306 figs.
1892. Traite des maladies parasitares, etc., Paris 2 ed. 767 pp., 364 figs.
Railliet, A.
1890. Les parasites de nos animaux domestiques. Rev. d. sc. nat. appliq. Paris.
Tome XXXVII: 745-756.
1892. Une nouvelle affection parasitaire du lievre et du lapin de garenne. Rev.
d. sc. nat. appliq., Paris. Tome XXXVII : 345-352.
1893. Traite d. zool. med. et agric. 1 : 283.
Rees, F. G.
1933a. Studies on Cittotaenia pectinata (Goeze, 1782) from the common rab-
bit, Oryctolagus cuniculus. Part I. Anatomy and histology. Proc. Zool.
Soc. London (1933) : 239-252 (Plates I-V)
1933b. Studies on Cittotaenia pectinata (Goeze, 1782) from the common rab-
bit, Oryctolagus cuniculus. Part II. Developmental changes in the egg,
and attempts at direct infestation. Idem. (1933) : 253-257. (Plates
MI).
1938J Arnold: Anoplocephaline Cestodes of N. Amer. Rabbits
53
Riehm, G.
1881. Utersuchungen an den Bandwiirmern der Hasen und Kaninchen.
Zeitschr. f. ges. Naturwiss. 3 ser., VI : 200.
1881a. Studien an Cestoden. Zeitschr. f. d. ges. Naturwiss. Bd. 6: 545-610.
Rudolphi, C. A.
1804. Bemerkungen aus dem Gebiet der Naturgeschichte, Medizin und
Thierarzneykunde, auf einer Reise durch einen Theil von Deutschland,
Holland und Frankreich. 1 Th. Berlin.
1805. Id. 2 Th. Berlin.
1810. Entozoorum sive vermium intestinalium historia naturalis. Vol. II, pt.
2, Amstelaedami.
Sprehn, C. E. W.
1932. Lehrbuch der Helminthologie. 996 pp. 374 figs. Pub. Borntraeger
Brudern.
Stiles, C. W.
1895. Notes on Parasites. Vet. Mag. II: 341-346.
1896. A Revision of the Adult Tapeworms of Hares and Rabbits. Proc. U. S.
Nat. Mus. 19: 145-235; 25 pis.
Stiles, C. W. and Hassall, A.
1896. Notes on Parasites. Vet. Mag. Ill : 6-9.
1896. Notes on Parasites. Vet. Mag. Ill: 407.
1902. Bertiella, a new name for the cestode genus Bertia Blanchard, 1891.
Science, n. s. 16: 434.
1902-1912. Index-Catalogue of Medical and Veterinary Zoology. Authors’
Index. Bur. An. Ind. Bull. 39.
1912. Index-Catalogue of Medical and Veterinary Zoology. Subjects: Cestode
Cestodaria. Hyg. Lab. Bull. 85.
Stunkard, H. W.
1934. Studies on the Life-History of Anoplocephaline Cestodes. Zeit. f.
Parasitenk., 6: 481-507.
Zeder, J. G. H.
1800. Erster Nachtrag zur Naturgeschichte der Eingeweidewurmer. Leipzig.
320 pp., 6 pis.
1803. Anleitung zur Naturgeschichte der Eingeweidewurmer. 432 pp*., 4
pis. Bamberg.
Burkenroad: Penaeidae
55
3.
The Templeton Crocker Expedition. XIII. Penaeidae from the
Region of Lower California and Clarion Island,
with Descriptions of Four New Species.1
Martin D. Burkenroad
Bingham Oceanographic Laboratory,
Yale University.
(Text-figures 1-34).
[Note: This is the thirteenth of a series of papers dealing with the specimens
collected on the Twenty-fourth or Templeton Crocker Expedition of the Depart-
ment of Tropical Research of the New York Zoological Society; William Beebe,
Director. For data on dredges, localities, dates, etc., concerning the capture of
specimens treated in this paper, refer to the Volume XXII of Zoologica, No. 2
pp. 33 to 46. Unless otherwise noted the catalogue and station numbers referred
to relate to the above-mentioned expedition.]
Contents.
Page
Introduction 56
1. Gennadas sordidus Kemp 57
Female described and placed in key to genus; from Pacific America,
new record.
2. Gennadas scutatus Bouvier 59
From Pacific America, new record.
3. Hymenopenaeus doris (Faxon) 60
Semi-nektonic habit.
4. Solenocera mutator, sp. nov 61
Described.
5. Solenocera florea, sp. nov 64
Described.
6. Penaeus calif orniensis Holmes 67
Resurrected and differentiated from P. brevirostris Kingsley.
7. Penaeopsis kishinouyei (Rathbun) 72
Described; from Revillagigedo Islands, new record.
8. Penaeopsis beebei, sp. nov 74
Described.
1 Contribution No. 550, Department of Tropical Research, New York Zoological Society.
56
Zoologica: New York Zoological Society
[XXIII :3
Page
9. Penaeopsis mineri Burkenroad 77
Seminal receptacles described.
10. Trachypeneus similis pacificus Burkenroad 78
11. Trachypeneus brevisuturae Burkenroad 79
Female described; from Lower California, new record.
12. Eusicyonia disparri Bux-kenroad 80
Male described.
13. Eusicyonia disedwardsi Burkenroad 82
Female described.
14. Eusicyonia penicillata (Lockington) 83
Juveniles described.
15. Eusicyonia aliaffinis Burkenroad 84
16. Eusicyonia species 84
Possible intergradation with E. aliaffinis and E. affinis Faxon.
17. Eusicyonia picta (Faxon) 87
Compared with E. stimpsoni (Bouvier).
18. Eusicyonia disdorsalis Burkenroad 87
From Lower California, new record.
19. Eusicyonia ingentis, sp. nov 88
Described.
Literature Cited 90
Introduction.
Of the seventeen or eighteen species of Penaeidae represented in the
present collection, more than half were previously known more or less in-
completely if at all. There are no less than five species in the collection
which have not heretofore been recorded from Pacific America, of which
three are new to science; in addition the material aids in the resurrection
of a named species not generally recognized as valid and in the completion
of descriptions of five species hitherto known only from juveniles or from
a single sex, and also assists in the recognition as a new species of one form
not included in the collection. That so large a part of the gathering is of
the ultimate in systematic value is a tribute not only to the surprising rich-
ness of the region but to the capacity of those by whom the material was
obtained. I am deeply indebted to Dr. William Beebe, the Director of the
Expedition, and to his associates, for so valuable an opportunity to continue
a study of the Pacific American peneids.
In a previous paper (Burkenroad, 1936, p. 6), it was stated that
whereas the littoral peneids of the Atlantic and Pacific coasts of America
are much more nearly related each to the fauna of the other coast than is
either to the fauna of any other region, the deep-water peneids of the two
regions seem to exhibit the inverse relationship. In explanation of this re-
versal of degree of relationship, it was suggested that the littoral peneids
of the two coasts have had contacts less intimate (in terms of epoch and
intensity) with the peneids of other regions than were their contacts with
each other supplied by shallow Pleistocene and preceding transcontinental
channels. It was suggested that by contrast, the oceanic faunas of the
two coasts have presumably had no direct communication since the elevation
of the Panama ridge before the late Cretaceous and have at the same
time by their oceanic habit perhaps been placed in communication with
1938 j
Burkenroad: Penaeidae
57
the faunas of other regions with relatively greater frequency than have the
littoral forms. The new records supplied by the present collection are in
agreement with the arrangement suggested above. Of the two newly added
deep-water species, Gennadas sordidus is otherwise known only from the
Indo-Pacific; while G. scutatus, although it occurs in the American Atlantic,
is also known from the intervening Indo-Pacific regions.
It seems of interest to observe that the proportion of Eusicyoninae to
Penaeinae in the region of Lower California is rather different from what
seems usual elsewhere, the Eusicyoninae forming a relatively very con-
spicuous part of the littoral population. Thus, for instance, DeMan, 1911,
records as the catch of the Siboga in East Indian waters 388 specimens of
Penaeinae of about 30 species to 19 specimens of Eusicyoninae of 9 species
(a ratio which I believe to be not unusual in unselected collections from
most parts of the world). The Zaca took 156 specimens of Eusicyoninae of
7 or 8 species to 182 specimens of Penaeinae of 5 species, and the Pawnee
in 1926 took 92 specimens of Eusicyoninae of 4 species to 43 of Penaeinae
of 3 species, in the Lower California region.
Gennadas Bate.
Gennadas, Burkenroad, 1936, p. 59.
Gennadas sordidus Kemp.
Gennadas sordidus, Kemp, 1910, p. 177.
Amalopenaeus sordidus, Balss, 1927, p. 262.
(Text-figure 1).
Range : Indian Ocean off India; Pacific off Lower California, Gulf of
California; in midwater.
Material : A total of 37 specimens (about one-third of them males)
was taken in the Gulf of California in the general neighborhood of Tortuga
(Station 139) and San Ildefonso (Station 148) Islands; in the mouth of the
Gulf of California (Stations 158 and 159) ; off the southern tip of Lower
California (Station 134), and north of the Revillagigedo Islands (Station
165), at depths above 300 to 500 fathoms both inside and outside the 1,000
fathom contour, as follows:
Station 130: T-l (1 3); Station 134: T-2 (2 5), T-3 (1 9); Station
139: T-2 (1 $), T-3 (1 $) , T-4 (1 9); Station 148: T-2 (1 9), T-4 (2 $,
5 9) ; Station 158: T-4 (2 $, 7 9) ; Station 159: T-2 (1 $, 2 9), T-3 (2 $, 4 9) ;
Station 165: T-3 (3 $, 1 9). Cat. Nos. 36,950, 36,951, 36,952, 36,953, 36,954,
36,955, 36,956, 36,957, 36,958, 36,959, 36,960, 36,961.
Dimensions and Sexual Condition : Material ranges from adult in-
dividuals of carapace length 11 mm., total about 32 mm., to an uncertainly
determinable early post-mysis probably of this species, of carapace 2 mm.
The petasmal endopods are united in males above 8 mm. carapace length ;
females may and usually do bear ripened ovaries above 9 mm. Of the
twelve fruitful hauls, the six made at 500 fathoms or more accounted for 28
of the 37 specimens; no clear indication of segregation according to size
is apparent.
Remarks: The present Pacific American catches of Gennadas sordidus
are unexpected, inasmuch as previous records have suggested a localization
of the species in the Indian Ocean.
The female of the species has not previously been described. The
thelycum bears a very great resemblance to that of G. parvus Bate, from
which it differs in the slighter development of the median longitudinal
carina of sternite XIV, the more narrowed anterior portion of the triangular
58
Zoologica: Netv York Zoological Society
[XXIII :3
shield of sternite XIII, and the smaller size of the protuberance at the
sides of sternite XIII behind the bases of the third legs. The median
carina of XIV is variable in length and definition, ranging in form from
a short, tooth-like crest at the posterior margin of the sternite to a very
low ridge fading out anteriorly before it reaches a third of the way to
the anterior margin (although a faint trace of median elevation sometimes
reappears on the anterior part of the sternite and is very occasionally con-
tinuous with the more perceptible posterior ridge). In G. parvus the
much stronger carina of XIV displays comparable variations, but seems al-
ways to reach beyond the anterior third of the sternite. The protuberance
behind the bases of the third legs varies in strength in G. sordidus from
complete absence to a low and inconspicuous weakly setose swelling (and is
even, in one specimen, a conspicuous projection) ; but it seems never to
acquire the strength of its homologue in G. parvus, which reaches as far
median as to the lips of the openings of the sperm receptacles.
Text-figure 1.
Gennadas sordidus Kemp. Thely-
cum; adult $, D. T. R. 36,960,
x 13.
The petasma of G. sordidus is adequately represented by Balss, 1927,
fig. 18. It differs from that of G. parvus most conspicuously in that the
distoventral lobe is not clearly subdivided, the cleft between the lobules in
G. parvus being represented in G. sordidus only by a slight crenellation of
the margin of the lobe and a short vertical crest on its posterior face (some-
what as in G. tinayrei ) ; and also in that the accessory and the distomedian
lobes are not subdivided, as they are in G. parvus.
In my discussion of Gennadas in 1936, p. 85, I have stated, on the basis
of the undivided form of the distolateral lobe, that “The only other species
of which the female is unknown, G. sordidus Kemp, undoubtedly belongs
to a different section [I], and may be predicted with some confidence to have
a thelycum with separated, independent spermathecal orifices;” a prediction
which now stands confirmed. In a key to the species (1936, p. 64), however,
G. sordidus has been placed in Group I A with the species (G. capensis and
G. kempi) in which the distoventral lobe of the petasma is undivided and
the orifices of the sperm receptacles very widely separated and not guarded
posteriorly by conspicuous prominences; rather than in Group I B of
forms (G. elegans, G. brevirostris, G. tinayrei, G. parvus) with divided
distoventral lobe and guarded and less widely separated spermathecal
orifices. The structure of the female of G. sordidus is clearly that of
Group I B ; and it is evident that I was mistaken in ignoring the trace of a
1938J
Burkenroad : Penaeidae
59
cleft in the distal border of the distoventral lobe of the petasma indicated
in Balss’ figure (but not, however, in that of Kemp, 1910, plate XIV, figs.
1, 2). It may be observed that both Kemp and Balss have placed the male
of G. sordidus in its correct relationship, as most closely allied to G. parvus.
The diagnostic key published by me in 1936 thus requires to be modified
by elimination of I A 1 and transfer of G. sordidus to I B 2 b (2) as follows:
IB 2b (2). MALE. Distolateral lobe much narrower than the distoventral,
and not reaching so far distally as does the latter; accessory lobe much
larger than the distolateral.
FEMALE. Elevation of XIV with a median longitudinal ridge.
I B 2 b (2) (a) MALE. Division of the distoventral lobe indicated only by
a slight crenellation and thickening of the margin; accessory and dis-
tomedian lobes entire.
FEMALE. Median longitudinal ridge of XIV very short, not well
defined, if present, on the anterior two-thirds of the sternite.
6. G. sordidus Kemp; Balss, 1927, $ [$, figure 1],
I B 2 b (2) (b). MALE. Distoventral lobe conspicuously bipartite; accessory
lobe tripartite; distomedian lobe bipartite.
FEMALE. Median longitudinal ridge of XIV well defined and
extending two-thirds or more the length of the sternite.
7. G. parvus Bate; Balss, 1927, $ and 2.
Gennadas scutatus Bouvier.
Gennadas scutatus, Bouvier, 1906, p. 9 ; Burkenroad, 1936, p. 83.
Gennadas clavicarpus, part, DeMan, 1911, p. 19.
? Gennadas clavicarpus, Boone, 1930, p. 129.
Range : Cosmopolitan, in midwater.
Material: Two female specimens were taken 145 miles north of Clarion
Island (Station 165 T-3), 20° 36' N. Lat., 115° 07' W. Long., at a depth
above 500 fathoms (Cat. No. 36,962).
Dimensions and Sexual Condition: Carapace length 9 mm., total about
31 mm. The specimens equal in length the largest heretofore recorded (Milne
Edwards and Bouvier, 1909, p. 194). Both have well-developed ovaries; in
one a small mass of material in one sperm-receptacle seems to mark an
interrupted mating.
Remarks: The record from Pacific America is new.2
The present material differs in genital sculpture from that known from
the Atlantic and the Indo-Pacific in that the free flap from the anterior
margin of the fourteenth sternite is very slender (about three times as long
as broad), tapering, and pointed, being widest at the base; the anterior
margin of the anterior lip of the sperm receptacles is heavily setose; there
are a pair of conspicuous setose prominences between the bases of the fourth
pair of legs; and the coxal projection of the fourth legs has an elongate
deeply concave distal margin. Comparison of the present specimens and
others (including both sexes) taken by the Arcturus in the American
Pacific, with Atlantic material of G. scutatus in the Bingham Oceanographic
Collection and with the literature, leads however to the belief that the Zaca
females are to be considered as an extreme variation falling within the
specific limits.
2 It is possible that the record by Boone, 1930, of “Gennadas clavicarpus DeMan” from off Cape
Mala, may refer to the present species. However, although Miss Boone’s description of the petasma
applies to Gennadas scutatus, her description of the thelycum applies to Gennadas propinquus Rath-
bun, just as in the account of “G. clavicarpus” by DeMan, and what her specimens may actually
correspond to therefore seems doubtful.
\
60
Zoologica: New York Zoological Society
[XXIII :3
In females of G. scutatus from the Gulf of Mexico and the Caribbean,
the free flap of XIV is very broad, not more than one and one-third times
as long as wide. In Plate II, figure 3k, DeMan, 1911, of the thelycum of an
Indo-Pacific female of G. scutatus (“G. clavicarpus” ) , the flap is shown
as about one and three-quarters times as long as wide; in Bouvier’s figure
of a North Atlantic female (1906, fig. 8), as two and one-third times as long
as wide. In an Arcturus female from the American Pacific the flap is two
and one-third times as long as wide. In available Atlantic females the tip
of the flap is sometimes truncate or even emarginate, sometimes pointed.
The Atlantic females bear a pair of very inconspicuous and weakly
setose prominences, placed just median to the bases of the fourth legs;
these prominences, which were overlooked in my previous account, seem
the equivalents of the much more conspicuous ones in the Zaca females ; and
indeed, are not very much less distinct in Atlantic specimens than in small
Arcturus females from the American Pacific.
The inner margin of the coxa of the fourth legs in Atlantic females
generally resembles that indicated in DeMan’s Plate II, figure 3k, of an
Indo-Pacific specimen; in some Atlantic females, however, the distal projec-
tion is farther below the coxobasisal articulation and is more pronounced,
thus approaching the form characteristic of the Pacific American specimens.
Finally, the petasma of an Arcturus male from the American Pacific,
as compared with the same organ in Atlantic specimens, seems to show no
differences lying outside the range of those fairly considerable variations in
relative length of the different lobules which are indicated in the literature.
Hymenopenaeus Smith.
Hymenopenaeus , Burkenroad, 1936, pp. 101, 102.
Hymenopenaeus doris (Faxon).
Haliporus doris, Faxon, 1893, p. 214; 1895, p. 191.
Range: Off the Pacific coasts of southern Mexico and Guatemala;
mouth of the Gulf of California. Midwater (300-500 fathoms) and bottom,
in 1,500-2,000 fathoms.
Material: One female was taken in the mouth of the Gulf of California
(Station 159 T-3, Cat. No. 36,963) at a depth above 500 fathoms.
Dimensions and Sexual Condition: The specimen is a juvenile of cara-
pace 9 mm., total length about 29 mm.
Remarks : The thelycum of the present small female differs consider-
ably from that of the much larger types of H. doris in that the median
vertical projection of the posterior margin of the thirteenth sternite has
a very bluntly acuminated rather than a truncated distal margin; and the
pair of projections laterally flanking the median projection are very much
lower than the median one instead of reaching as high as it does. The
posterior margin the twelfth sternite lacks the pair of posteriorly directed
projections present in the types.
In my opinion, the thelycum described above cannot be that of a juvenile
of H. nereus, particularly since a strong cusp like that of the adult H. doris
is present on the fourteenth sternite, instead of the low untoothed ridge of
H. nereus adult. In Penaeinae, a tooth or spine is usually present on the
fourteenth sternite in larval or juvenile stages, which may later disappear
in the adult. According, however, to the results of examination of mysis
and post-mysis larvae and of juveniles of Solenocerinae, a tooth develops on
the fourteenth sternite of this subfamily only in juveniles of species which
continue to bear it as adult.
As to whether the present specimen actually refers to H. doris rather
1938]
Burkenroad: Penaeidae
61
than to some undescribed foi-m it is impossible to be certain. However, the
thelycum in juveniles of the closely related species H. laevis differs from
that of adults in somewhat the same manner as, although to a lesser degree
than, the thelycum of the present juvenile differs from that of H. doris.
A comparison of the present specimen of H. doris and of another sim-
ilar but still smaller juvenile with two adult females of H. nereus (the three
last-mentioned specimens being undescribed Pacific American material taken
by the Arcturus) , reveals no striking differences between the two forms
other than those in genitalia. A less detailed comparison of the female types
of H. nereus with those of H. doris (the males being known in the former
species only), made during a visit to the Museum of Comparative Zoology,
also indicated that, about as stated by Faxon, the two species agree so
closely except in thelycum “that one description would serve for both.’’ Both
forms are extremely like H. laevis (Bate), and differ from the remainder
of the genus [save possibly H. villosus (Alcock and Anderson) ] in bearing
both pterygostomian and branchiostegal spines (cf. Burkenroad, 1936, pp.
103 seq.). The only non-genital differences from H. laevis seem to be that
the eyes are somewhat larger in the two Pacific American species and the
tip of the telson, beyond the lateral spines, is more slender and elongate
than in H. laevis.
It may be noted that in a preceding reference to H. nereus and H. doris
(1936, p. Ill), I have done an injustice to Faxon by the incorrect state-
ment that “only three” of the five teeth arming the median lobule of the
distolateral lobe of the petasma of H. nereus “are indicated in Faxon’s
figure.”
Faxon (1895, p. 192) calls attention to the possibility that those of his
specimens of H. doris which were obtained at Albatross Station 3414 might
possibly have been taken pelagically, inasmuch as Agassiz records the cap-
ture of some “transparent Penaeidae” in the tow-net at this station; it is
possible, however, that Agassiz refers to the “Sergestes edwardsii Kr.”
which Faxon records from Station 3414. On the other hand, there is no
possible doubt that the present juvenile of H. doris, and another taken by
the Arcturus off Cocos Island at 300 fms., were captured pelagically, since
both were caught in nets lowered to not more than 500 fathoms, in areas
well outside the 1,000-fathom contour. In the present Californian juvenile,
the statocyst contains a flattened agglomeration of transparent, sharp, irreg-
ularly shaped granules, the largest of which is about 0.15 mm. in greatest
dimension. These granules, which are not affected by dilute HC1, appear
siliceous; although they are rather readily fractured by pressure with the
dissecting needle. In the Cocos juvenile, the statocyst contains a flattened
plate of the delicate shells of minute pelagic foraminifera cemented together
with cuticular secretion. In two large specimens of Hymenopenaeus nereus,
taken by the Arcturus on bottom at Station 74 and examined for comparison,
the statolith is composed of pelagic foraminifera, heavy calcareous frag-
ments apparently of molluscan shells, and granules of siliceous appearance
similar in nature to those composing the statolith of the Californian juve-
nile of H. doris. I am inclined to think that as in the case of the pelagically
captured specimens of H. laevis and H. aphoticus discussed by Burkenroad,
1936, p. 106, the present juveniles of H. doris had obtained their statoliths
from bottom, and had later swum up to the levels where they were captured.
SOLENOCERA Lucas.
Sofenocera mutator, sp. nov.
(Text-figures 2-5).
Type: Type male and cotypes: Cat. No. 36,969, Department of Tropical
Research, New York Zoological Society; taken at Station 150, Dredge 5
62
Zoologica: New York Zoological Society
LXXIII :3
(5 $, 2 2) ; from Gorda Banks, tip of Lower California, 23° 01' 30" N. Lat.,
109° 30' W. Long., 4-foot dredge at 40 to 100 fathoms; April 21, 1936.
Range: Bay of Panama; Pacific coast of Mexico off Manzanillo and
Mazatlan; Lower California, both coasts; 20-117 fathoms.
Material : A total of 33 specimens (somewhat more than half of them
males) was taken, at depths of 20 to 100 fathoms, east of Cedros Island
(Stations 125 and 126), in Santa Inez Bay (Stations 141 and 143), on
Gorda Bank (Station 150) and off Mazatlan (Stations 154 and 155), as
follows :
Station 125: D-l (2 $) ; Station 126: D-3 (2 $, 1 5), D-4 (1 $, 1 9) ;
Station 141: D-4 (1 $, 1 9) ; Station 143: D-3 (5 $, 2 9) ; Station 150: D-5
(5 $, 2 2) ; Station 154: D-l (3 9 plus 1 fragment) ; Station 155: (3 S, 3 9).
Cat. Nos. 36,964, 36,965, 36,966, 36,967, 36,968, 36,969, 36,970, 36,971.
In addition to this material, I have been enabled through the kindness
of Dr. W. L. Schmitt of the U. S. National Museum to examine four males
ranging in carapace length from 12 to 13.5 mm. and five females ranging in
carapace length from 12.5 to 16 mm., taken off Manzanillo, Mexico, in 117
fathoms (U. S. N. M. 28489) ; as well as a large female 19 mm. in carapace
length from Panama Bay ( Albatross Station 2804) at a depth of 47 fathoms
(U. S. N. M. 28488).
Dimensions and Sexual Condition : Males ranging in size from cara-
pace length 16 mm., total about 62 mm., to carapace 9 mm.; females from
carapace 14 mm., total 56 mm., to carapace 7 mm. In the four males of
Solenocera mutator, n. sp. 2. Carapace (lateral view) ; juvenile Mazatlan $,
D. T. R. 36,970, x 6. 3. Petasma (left half, posterior view) ; type $,
T. R. 36,969, x 6. 4. Fourteenth sternite; adult $, D. T. R. 36,965, x 6.
Thelycum; subadult 9, D. T. R. 36,968, x 6.
1938J
Burkenroad : Penaeiclae
63
carapace 10 mm. or less, and in one of carapace 11 mm., the petasmal endo-
pods are unjoined; in the rest, of carapace 11 mm. or more, the petasma is
of adult form.
Diagnosis : Pterygostomian spine meeting the frontal margin at nearly
a right angle; cervical carina not notched above the level of the hepatic
tooth; postorbital spine well developed; orbital angle dentiform. Cervical
sulcus not continuous across the dorsum; postrostral carina of the carapace
obliterated behind the level of the cervical sulcus. The rostrum is short and
deep, with convex ventral margin; it bears from 5 to 7, usually 6, teeth in
addition to the epigastric, of which the posteriormost (and sometimes the
preceding also) is behind the orbit. The posteriormost tooth of the rostral
series is separated from the preceding by an interspace equal to one-third
to one-fifth, usually more than one-fourth, of the interval between the pos-
teriormost rostral tooth and the dorsal end of the cervical sulcus. The
epigastric tooth is placed at a point from three-fifths to more than two-
thirds, usually about two-thirds, of the way from the orbital margin to the
cervical sulcus. In specimens from other localities than Lower California
there is a small slender spine on the dorsum of the carapace near its pos-
terior end, and still other spines not present in Lower Californian material
may occur on carapace and pleon. It is to the inconstant presence of this
armature that the specific name refers.
The fourth, fifth and sixth pleonic somites are dorsally carinated; the
third rather weakly ridged. The telson is short, with a conspicuous pair of
fixed marginal teeth.
The antennular flagella vary from somewhat shorter than the carapace
minus the rostrum, to more than one and two-thirds this length. They are
usually quite slender, except in a few of the larger males where the distal
portion of the inferior flagellum is expanded to twice its basal breadth.
The emargination of the posterior end of the fourteenth sternite of the
male is deep but broad. The paired teeth of the anterior part of the four-
teenth sternite of the female are usually represented by broad, low, obtuse
swellings; which may in large specimens, however, be somewhat produced,
acuminated and turned medioanteriorly. The posterior margin of the thir-
teenth sternite of the female is, although narrowly incised by a median
groove, on the whole conspicuously convex; the V-shaped anterior edge of
the posterior part of the stei’nite is sharply defined; the anterior part of the
sternite bears a low ridge not rising into a setose protuberance at its hinder
end. The coxae of the fourth legs of the female are often considerably pro-
duced, so that they may overlap considerably in the midline.
The petasma is particularly characterized by the considerable size of
the triangular lamellate projection of the dorsal face of the distolateral lobe
(which is even visible in ventral view, projecting beyond the lateral margin
of its lobe) ; and by the elongated subtriangular shape of the acuminated
median lobule of the distolateral lobe. The distoventral lobe is crowned by
a free distoventral projection.
Remarks : In a preceding paper (1936, p. 123), I have described in
detail, under the heading “Solenocera species,” a small male with joined
petasmal endopods taken in the Gulf of Panama, which differs in several
respects from the much larger types of S. agassizii Faxon from the same
region. It was thought possible at that time that this specimen might rep-
resent a young form of S. agassizii; chiefly for the reason that a small
female of Solenocera from the Gulf of Mexico, described in the same paper
(1936, p. 122) under the heading “Solenocera vioscai Burkenroad” and re-
garded as a young adult of that species, was found to differ in bodily char-
acters from the much larger types of S. vioscai in somewhat the same man-
ner as the small Panamanian male differed from the types of S. agassizii.
A study of the very much more adequate material of Solenocera from
64
Zoologica: Neiv York Zoological Society
[XXIII :3
American /waters which has now become available from various sources,
demonstrates, however, that not only do both “Solenocera species” and the
supposed young specimen of S. vioscai represent species distinct from those
to which it was previously thought that they might refer, but there are
still other distinct forms, one from the Gulf of Mexico, and the present one
included in the Zaca collection. The American species of the genus therefore
number six, of which only two, S. agassizii Faxon and S. vioscai Burkenroad,
have been named until the present. The third form, S. mutator sp. nov., has
been named and diagnosed in preceding paragraphs. A fourth form is a
peculiar one from the Gulf of Mexico represented by the supposed “juvenile”
erroneously referred to S. vioscai in 1936; a fifth form, also from the Gulf
of Mexico, is closely related to S. mutator; these will be described in a
forthcoming paper. The sixth form, the “Solenocera species” from Panama,
which is also closely related to S. mutator, may be given a name in this
place, as follows :
Solenocera florea, sp. nov. Holotype, B. O. C. No. 132 (a male described
by Burkenroad, 1936, p. 123, figs. 68-69). Paratypes (Text-figs. 6-7 below)
U. S. N. M. No. 28486, five males and five females, all adult, taken in Panama
Bay at a depth of 33 fathoms. The paratype material has been made avail-
able to me by the very great kindness of Dr. W. L. Schmitt of the U. S.
National Museum. The males range in size from carapace 17.2 mm., total
length 68 mm., to carapace 15 mm.; the females from carapace 22.8 mm.,
total length 77 mm. to carapace 21.5 mm. The paratype material agrees very
well with the previous description of the holotype; it may however be noted
that the antennal scale usually overtops the antennular peduncle by a fif-
teenth or more of the length of the scale. The figure of the petasma of the
holotype (Burkenroad, 1936, fig. 69) is in error, in that the distal margin
of the median lobule of the distolateral lobe should be shown as folded over
anteriorly, not posteriorly ; the description is also in error in that the “pair
of rigid lamellar projections” on the posterior face of the distolateral lobe
actually represent the two ends of a single lamella, the midsection of which
is broken in the holotype. Diagnostic distinctions between S. florea and
related species of the genus are given in succeeding paragraphs.
The three Pacific American species of Solenocera are distinguished
from their three Atlantic American congeners by the fact that the blade of
the epipodite of the tenth through thirteenth somites is in all of the former
deeply bifurcate, the shorter ramus being half or more the length of the
undivided portion of the blade (not including peduncle) ; whereas in the
Solenocera florea, sp. nov.
Text-figures 6 & 7.
Solenocera florea, n. sp. 6. Fourteenth sternite; adult, U. S. N. M.
28,486, x 5. 7. Thelycum; adult $, U. S. N. M. 28,486, x 5.
1938J
Burkenroad: Penaeidae
65
Atlantic forms the epipodites are only shallowly furcated or even mitten-
like, the shorter ramus being at most less than a third as long as the un-
divided portion of the blade. S. mutator, S. ftorea and one of the undescribed
species from the Gulf of Mexico are all three distinguished from S. agassizii
and S. vioscai by their less numerous rostral teeth (minimum 6, maximum 8
in the former as compared with minimum 8, maximum 10 in the latter) ; the
right angle made by the juncture of their pterygostomian spine with the
frontal margin of the carapace dorsal to it, this juncture being a smooth
concavity in the two latter forms; and the presence of a scale-like projec-
tion on the anterior face of the distolateral lobe of their petasma, which is
not represented in S. vioscai and S. agassizii. The former three species are
also, so far as their maxima are known, smaller (55-80 mm.) than the latter
two (total lengths respectively exceeding 100 and 200 mm.). The second
unnamed species from the Gulf of Mexico, which was erroneously described
as the juvenile of S. vioscai in 1936, agrees with the S. mutator group as
regards number of rostral teeth, but with the S. agassizii group in pterygo-
stomian spine; its petasma is unknown.
In non-sexual characters, typical (Californian) material of S. mutator
is not very easily distinguished from S. ftorea although the epigastric tooth
of the latter is usually placed somewhat nearer to the orbit (only about one
and one-quarter times as far from the orbit as from the cervical sulcus) and
the posteriormost tooth of its rostral series is less widely separated from
the tooth preceding it (interval between the two teeth from a fourth to a
fifth of that beween the posterior tooth and the cervical sulcus) than in
S. mutator. The scale of the inner margin of the basal segment of the
antennular peduncle (prosartema) of S. ftorea, which reaches well beyond
the eye, seems to be longer than that of S. mutator.
Females of S. ftorea are readily distinguished from those of S. mutator
by the fact that in the former the posterior margin of the thirteenth ster-
nite is acutely concave rather than roundedly convex; and that there is no
sharp boundary between the anterior and posterior parts of the same ster-
nite except medially, where the median groove of the posterior part of the
sternite ends at a conspicuous, unpaired, setose conical projection set just
in advance of the fourth legs. The pair of teeth of the anterior part of the
fourteenth sternite also seems to be somewhat more sharp and slender in
females of S. ftorea, and the coxal nibs of the fourth legs are less extended
than is sometimes the case in S. mutator.
The emargination of the posterior edge of the fourteenth sternite is a
narrow notch with parallel sides in males of S. ftorea, rather than a sloping
concavity as in S. mutator. In petasma the two species are strikingly dif-
ferent; particularly as regards the subtriangular rather than subrectan-
gular shape of the median lobule of the distolateral lobe of S. mutator, and
the form of the distoventral lobe which is in S. mutator very nearly identical
with that in S. agassizii.
Certain peculiar variations in structure within the material which I
have here assigned to Solenocera mutator seem to require attention. In the
specimens from Mexico and Panama, the integument is thin and mem-
braneous, the tissues are rather flabby, and the pleon seems relatively quite
slender. In Californian material, on the contrary, the integument is thick
and rigid, the flesh solid, and the habitus is of the Penaeine type usual in
the American species of the genus. These differences are difficult to define
precisely, and there is considerable intergradation between the groups; but
nevertheless specimens from Lower California do give a general impression
different from that of the Mexican material. That this impression may not
be the result of differences in method of preservation is suggested by the
fact that the Lower Californian and one of the available lots of Mexican
material are derived from one collection, the other lot of Mexican and Pan-
amanian material from another.
66
Zoologica: New York Zoological Society
[XXIII :3
Of more precisely definable differences between the peninsular and the
mainland material of S. mutator, in all but one of the nineteen Mexican and
Panamanian specimens the dorsum of the carapace about midway between
the cervical sulcus and the posterior margin bears a small, slender spine
which is absent in all of the twenty-four Californian specimens. In the
mainland material there is also a well-developed vestige of the posterior
dorsal organ found in many peneid larvae, which is not perceptible in Cali-
fornian specimens. In small mainland specimens from Mazatlan (the
smallest a female of carapace length 9 mm.), the carapace bears four addi-
tional pairs of spines of which no traces appear in peninsular specimens,
two pail’s on the branchial region below the cardiaco-branchial carina, and
two pairs on the sides behind the cervical sulcus. The cardiaco-branchial
carina, which is in Californian specimens obsolescent, is in these small indi-
viduals strongly developed ; and there is also a short carina, not represented
in peninsular material, which runs between the posterior section of the
cardiaco-branchialis and the cervical carina. One of the pairs of lateral
spines is set on the middle of this carina, from the midpoint of which an-
other ridge not perceptible in Californian specimens slants in a sinuous
oblique toward the posteroventral margin of the carapace. There is also in
these small specimens a spine at the posterior end of the dorsal carina of
the fourth and fifth pleonic somites as well as of the sixth to which such
spines are limited in the peninsular material. Altogether, the small speci-
mens recall to mind in many ways various of the species of Hymenopenaeus ;
thus the carapacic carination and pleonic armature, as well as the habitus,
are to some extent comparable with those of H. laevis (Bate) ; the dorso-
lateral postcervical spines with the pair present in H. diomedeae (Faxon)
(or in Solenocera hextii Wood Mason) ; and the posterior dorsal spine of
the carapace with that of H. steindachneri (Balss). In somewhat larger
specimens from Mazatlan, the extra carapacic carinae are weaker, and the
extra lateral spines of the carapace and the posterior spine of the fourth
pleonic somite are absent, although the dorsal carapacic spine and that of the
fifth pleonic somite are present. In the largest specimen from Mazatlan,
a male of carapace length 13.5 mm., all of the extra spines of carapace and
pleon are absent and the animal nearly approaches the Californian types in
habitus. In all available specimens from Manzanillo, of whatever size, the
dorsal spine and the two dorsal of the extra lateral pairs are present. A
spine is present on the fifth pleonic somite in all but one small male from
Manzanillo; that of the fourth pleonic somite occurs in only two of the ten
specimens. In the large female from Panama, of carapace length 19 mm.,
the dorsal and two pairs of extra lateral carapacic spines are present, but
only the sixth pleonic somite is dorsally armed.
There are a number of other inconstant differences between mainland
and peninsular material; thus there are eight rostral teeth in '5 of 7
Mazatlan specimens and in 2 of 9 Manzanillo specimens as compared with a
maximum of seven in all of the 22 Californian individuals with unbroken
rostrum; and the antennular flagella are shorter, or at most no more than
slightly longer, than the carapace in Mazatlan and Manzanillo specimens.
However, despite the many differences pointed out above, the mainland
and the peninsular specimens are identical in petasma and thelycum, and
for this reason I am unable to consider them as systematically distinct. It
seems possible, especially because of the change with size among the
Mazatlan specimens, that the peculiar features of the mainland material
are of larval derivation. However, the mainland specimens are quite as well
developed sexually as are Californian specimens of similar size, and the
Panamanian female is by far the largest of the known individuals of S.
mutator; while at the same time juveniles from Lower California, smaller
and less developed sexually than are any from the mainland, display no
traces of the extra spines and carinae characteristic of the latter. The
1938 |
BurJcenroad : Penaeidae
67
question of the significance of the differences between the Californian and
the mainland specimens is a most interesting one, and it is to be regretted
that no more satisfactory an account than the above can be offered.
Penaeus Fabricius.
Penaeus, Burkenroad, 1934a p. 74.
division 2, Burkenroad, 1934a, p. 75.
Penaeus calif orniensis Holmes.
Penaeus calif orniensis, Holmes, 1900, p. 218; Pesta, 1915, p. 113.
Penaeus canaliculatus , Holmes, 1895, p. 581.
Penaeus brevirostris, part, Rathbun, 1902, p. 287; 1904, p. 146; Schmitt,
1921, p. 21; Burkenroad, 1934a, p. 91.
Penaeus brasiliensis, part, Schmitt, 1935, p. 16.
(Text-figures 10, 11, 14, 15).
Range: San Francisco, California; Santa Monica, California; Gulf
coast of Lower California; west coast of Mexico near Esquinapa. Beach to
45 fathoms; Juveniles in lagoons and estuaries.
Material: A total of 3 specimens (1 $, 2 9) was taken on Arena Bank,
45 fathoms (Station 136 D-2, Cat. No. 36,972).
Dimensions and Sexual Condition: Larger female impregnated, of cara-
pace 34 mm., total length about 140 mm. ; smaller female of carapace 30 mm.,
unimpregnate. Male fully adult, carapace length 30 mm.
Remarks: No more than two American species of Section 2 of Penaeus
have in the past been recognized. The names generally applied to these have
been, for Atlantic specimens, P. brasiliensis Latreille; for Pacific ones, P.
brevirostris Kingsley. In the two most recent considerations of the group,
Burkenroad (1934a, pp. 91 seq.) has advanced some reasons for believing
that Atlantic is specifically distinct from Pacific material, while Schmitt
(1935, pp. 16-17) refers all American material to a single name, P. brasi-
liensis.
It now appears, however, that there are no less than five species of
Division 2 in the American waters, three Atlantic and two Pacific ones. As
the account of the Atlantic forms will appear in a forthcoming paper, it is
proposed to describe and to differentiate only the two Pacific species in this
place.
Although there seems in the past to have been no suspicion of the ex-
istence of more than one species of the P. brasiliensis group in Pacific Amer-
ica, names are already available for both of the newly separated forms.
Some juvenile specimens from Nicaragua were described by Kingsley, 1878,
p. 98, as the new species Penaeus brevirostris .3 In 1895, Holmes incorrectly
referred specimens of the brasiliensis group from Southern California and
San Francisco Bay (which he thought likely to be identical with Kingsley’s
material) to the Indo-Pacific P. canaliculatus (Olivier). In 1900 Holmes re-
trieved this error by describing the Californian specimens as the new spe-
cies P. calif orniensis ; but as he failed at this time to refer to Kingsley’s
species, his name has subsequently been regarded by most workers as syn-
onymous with P. brevirostris. P. calif orniensis Holmes is entered in his
list of peneid species by DeMan, 1911, p. 10, but it is intentionally omitted
from the account of Penaeus by the same author, who according to p. 131
accepts its identity with P. brevirostris. Pesta, 1915, p. 113, identifies mate-
3 This name was later incorrectly referred to the synonymy of P. brasiliensis by Kingsley him-
self (1879, p. 427), thus antedating the similar action by Verrill in 1922 and by Schmitt, 1935, p. 17.
68
Zoologica: New York Zoological Society
[XXIII :3
rial from San Francisco Bay as P. calif or niensis Holmes, but considers it
likely that Kingsley’s specimens are juveniles of the same form; he remarks
“Es scheint mir jedoch zweckmassiger den P. californiensis Holmes, von
dem eine gut Diagnose mit abbildungen gegeben ist, bestehen zu lassen, und
vielleicht die Spezies Kingsley’s als fragliches Synonym hierher zu stellen.”
Since I am informed by Dr. F. M. MacFarland of the California Acad-
emy of Sciences that the types of P. californiensis Holmes were destroyed
in the earthquake and fire of 1906, a specimen from Santa Monica, Cali-
fornia, is designated below as neotype. The specimen from San Francisco
which is figured by Pesta seems to belong to the same species as the neo-
type. Cotypes of Penaeus brevirosiris Kingsley, although juvenile, are eas-
ily differentiated from the form which occurs in the type locality of P.
californiensis. Material of the two Pacific American species of the group
available to me in addition to that obtained by Dr. Beebe, is listed below.
For material credited to the respective institutions, I am deeply indebted to
the authorities of the American Museum of Natural History, the Museum
of Comparative Zoology, and the U. S. National Museum.
Penaeus californiensis Holmes.
NEOTYPE. U. S. N. M. 23942. Santa Monica, California. 1 8 adult,
carapace 33 mm.; total length 140 mm.
B. 0. C. 18. Concepcion Bay, Lower California; May 3, 1926. Pawnee
II. 1 $ adult, impregnated, carapace length 43 mm.; 8 $ adult, carapace 29
to 34 mm.; 3 $ juvenile, carapace 14.5 to 20 mm.; 2 $ juvenile, carapace
17.5 to 20 mm.
B. 0. C. 14. Angeles Bay, Lower California; May 13, 1926; 17-23 fms.
Pawnee II. 3 9 adult (2 impregnated), carapace lengths 31.5 to 38.5 mm.;
2 $ adult, carapace lengths 29.5 and 30 mm.
A. M. N. H. 4620 [A], Lagunaje de las Lomitas, Esquinapa, Sinaloa,
Mexico (just south of the Gulf of California) ; November 15, 1919. Sr. M.
Gallegos. 2 $ juvenile, carapace lengths 18 mm. and 20 mm.
Penaeus brevirostris Kingsley.
(Text-figures 8, 9, 12, 13).
COTYPES. M. C. Z. Estero at Realejo, Nicaragua; March, 1867. 1 9,
juvenile, carapace length 13.2 mm.; 1 $, juvenile, carapace length 11 mm.
B. O. C. 15. Pearl Islands, Bay of Panama (8° 29' 40" N, 78° 52' 30"
W) ; March 31, 1926. Pawnee II. 5 9, subadult to juvenile, carapace length
9 mm. to 22 mm.; 2 $, juvenile, carapace lengths 10.5 mm. and 11 mm.
B. O. C. 102. Panama City Market (reported locality, Chame River) ;
February 10, 1934. M. D. B. 1 9 adult, impregnated, carapace length 38 mm.
B. O. C. 103. Tide-pool, Panama City; February 12, 1934. M. D. B.
1 $ juvenile, carapace length 8 mm.
B. O. C. 16. West coast of southern Mexico (14° 48' 40" N, 92° 54' 40"
W ; April 9, 1926 ; 30 fms. Pawnee II. 3 9 adult, impregnated, carapace
lengths 46 mm. to 50 mm.
U. S. N. M. 28417. Panama Bay ( Albatross Station 2795) ; 33 fms.
1 3 adult, carapace 30.5 mm. ; 1 9 adult, impregnated, carapace 43.5 mm.
U. S. N. M. 28419. Panama Bay ( Albatross Station 2804) ; 47 fms.
2 8 adult, carapace 31 mm.
A. M. N. H. 4620 [B]. Lagunaje de las Lomitas, Esquinapa, Sinaloa,
Mexico; November 15, 1919. Sr. M. Gallegos. 2 9, juvenile, carapace
lengths 18 mm. and 20 mm.
1938]
Burkenroad : Penaeidae
69
A. M. N. H. 4621. Estero del Puyeque, Esquinapa, Sinaloa, Mexico;
November 29, 1919. Sr. M. Gallegos. 1 $ juvenile, carapace length 19 mm.;
1 $ juvenile, carapace length 17.5 mm.
A. M. N. H. 4622. Pesqueria Caliguey, Esquinapa, Sinaloa, Mexico;
November 22, 1919. Sr. M. Gallegos. 1 $ juvenile, carapace length 19 mm.
A. M. N. H. 4623. Estero del Puyeque, Esquinapa, Sinaloa, Mexico;
November 29, 1919. Sr. M. Gallegos. 1 $ juvenile, carapace length 15.5
mm.; 2 $ juvenile, carapace lengths 18 and 19 mm.
The sharpest non-sexual diagnostic between P. brevirostris and P.
calif orniensis is supplied by the carina forming the postero-dorsal margin of
the antennal sulcus. In P. calif orniensis this carina reaches to within near-
ly half its length at least of the orbital angle (usually, and in adults almost
always, to within one-third) ; whereas in P. brevirostris the carina does not
reach to within more than three-fifths its length of the orbital angle, often
not within its length.
The orbital angle is more produced and the postorbital carina more pro-
nounced in P. calif orniensis than in P. brevirostris; thus in the former the
sharply defined posterior part of the crest (best seen in dorsal view) is
longer, usually considerably longer, than the more obtuse anterior portion,
while in the latter species the anterior obtuse part is usually the longer.
The posterior part of the ridge is in P. brevirostris often no sharper than is
the anterior part in P. calif orniensis, and the anterior part is sometimes
hardly discernible.
Text-figures 8-11.
Peneus brevirostris Kingsley. 8. Carapace (anterior part, lateral view) ;
juvenile $ cotype, M. C. Z., x 4. 9. Thelycum; impregnated adult $,
B. O. C. 102, x 4.
Peneus calif orniensis Holmes. 10. Carapace (anterior part, lateral view) ;
juvenile $, B. 0. C. 18, x 4. 11. Thelycum; unimpregnated adult $,
D. T. R. 36,972, x 4.
70
Zoologica: Neiv York Zoological Society
[XXIII :3
The adrostral carinae of P. calif orniensis are somewhat longer and more
widely separated than are those of P. brevirostris, generally extending
toward the posterior margin of the carapace to within the breadth of the
interval separating their posterolateral ends; and are posteriorly better
defined than in P. brevirostris, being continued medially at the posterior
ends of the adrostral sulci. In P. brevirostris these carinae usually extend
to no more than about three-quarters of the interval between them toward
the posterior margin, and are usually not continued medially, so that the
adrostral sulci lack a defined posterior margin. These differences in form
of adrostral carina are very much less distinct in juveniles than in adults.
The modal number of dorsal carapacic teeth is different in the two spe-
cies, as far as the present material serves to indicate, being nine in P.
calif orniensis, ten or eleven in P. brevirostris. The exact counts are shown
in the table:
Number of Rostral Teeth.
Number of
P. calif orniensis.
Individuals.
P. brevirostris.
7+1 epigastric
1
0
8 + 1
14
2
9 + 1
7
11
10 + 1
1
11
11 + 1
0
1
Finally, the two species seem distinct in thelycum, petasma and appendix
masculina. In adult Panamanian females of P. brevirostris, the posterior
part of the median elevation of the thirteenth sternite forms, in ventral
view, a roughly oval or diamond-shaped cup with elevated rim and de-
pressed center, against which the anteromedian lips of the lateral plates of
the fourteenth sternite abut, and by which these lips are held apart so that
they do not meet in the midline anteriorly. The anterior part of the ven-
tral surface of the lateral plates is quite heavily pubescent. The character-
istic form of the median elevation of XIII and the pubescence of the ventral
surface of the lateral plates are recognizable in Panamanian juveniles down
to a carapace length of 14 mm., though the pubescence cannot be discovered
in smaller specimens. Since the lateral plates of XIV gape widely in juveniles
of Penaeus, the separation of the anteromedian lips of the lateral plates
loses its value as a diagnostic in juveniles. In the juvenile female cotype
from Nicaragua, of carapace 13.2 mm., the form of the median elevation
of XIII is characteristic, but the specimen seems below the size at which
the pubescence of the lateral plates appears. In adult females of P. breviros-
tris from southern Mexico, the structure of the thelycum is similar to that
in Panamanian specimens, but the anterior parts of the ventral plates are
in some individuals only very lightly pubescent. In juvenile females from
Esquinapa, of carapace lengths 15-20 mm., the ventral surfaces of the lateral
plates are not pubescent at all ; and while the posterior part of the median
elevation of XIII has the characteristic cup-like form, its posterior end is
more extensively produced in a posterior direction, to form an obliquely
vertical ridge, than in the Panamanian juveniles.
In adult females of P. calif orniensis from the Gulf of California, the
posterior part of the median elevation of XII shows at most no more than
a trace of the cup-like form found in P. brevirostris; instead, it bears a
median longitudinal carina representing an elongation of the posteromedian
rim of the cup in P. brevirostris. The lateral plates of XIV meet in the
1938J
Burkenroad: Penaeidae
71
midline for their entire length, their anterior ends thus overlying and con-
cealing the posterior part of the elevation of XIII, as is clearly indicated in
Pesta’s figure of a female from San Francisco. The lateral plates are not
pubescent on their ventral surface. Juveniles of P. calif orniensis maintain
the characteristic form of the thelycum (except as regards the meeting of
the lips of the lateral plates) down to a carapace length of 18 mm. at least.
Although the juveniles of P. calif orniensis are not too readily distinguish-
able in these features from the Sinaloa juveniles of P. brevirostris de-
scribed above, it is my belief that the Sinaloa juveniles would as adults re-
semble material from southern Mexico, and would thus, although differing
slightly from the Panamanian stock of P. brevirostris, fall easily within
the same specific limits and be clearly distinct in female genital characters
from P. calif orniensis.
The petasma of adult males of P. calif orniensis displays the following
critical characters: The medially curved distal ends (distoventral lobes) of
the heavily chitinized lateral ribs of the petasma terminate in a sharply
pointed recurved tip which is free from the membranes forming the median
parts of the petasma. The distolateral or free margin of the distal ends
of the ribs is unarmed ; the proximomedian margin which is attached to the
membranes forming the median parts of the petasma is most frequently
armed on its anterior or dorsal side with a single proximally directed tooth-
like spine (the range in armature being from no such teeth to five). Just
median to the tip of the lateral rib is a large, fleshy posteriorly projecting
flap the area of which is nearly comparable with that of the distomedian
lobe. This flap, which probably represents a part of the distolateral lobe, is
armed on its free edge with several series of spines like those arming the
anterior surface of the membranes median to the lateral rib and those on
the median edge of the lateral rib itself. The posteriorly curved disto-
median lobes are quite long, and overlap the distal ends of the lateral ribs
by a considerable amount. The characteristic features of the adult petasma
are well-developed in the two available juvenile males with separated petas-
mal endopods.
In adult males of P. brevirostris from Panama the petasma differs from
that of P. calif orniensis as follows: The medially directed distal ends of the
lateral ribs terminate in a blunt tip not projecting free of the median mem-
branes; this tip is armed on its distolateral or free edge with one or two
Peneus brevirostris Kingsley. 12.
Petasma, (distal part, lateral
view from right side) ; adult
8, U. S. N. M. 28,419, x 5. 13.
Appendix masculina (of right
second pleopod, anterior view) ;
as in 12, x 5.
Text-figures 12-15.
Peneus calif orniensis Holmes. 14.
Petasma (distal part, lateral
view from left side) ; adult $,
D. T. R. 36,972, x 5. 15. Ap-
pendix masculina (of right
second pleopod, anterior view) ;
as in 14, x 5.
72
Zoologica: New York Zoological Society
[XXIII :3
conspicuous axially-directed teeth which seem not to be homologous with
the projecting, recurved, acutely-pointed tip in P. calif orniensis. The proxi-
momedian margin of the anterior, or dorsal side of the distal parts of
the lateral ribs bears a row of 8 to 11 spines along its juncture with
the membraneous median parts of the petasma. The folded distal edge
of the median parts, at the point of juncture with the tip of the lateral
rib, does not project inward as a conspicuous fleshy flap, and is here
unarmed. The distomedian lobes of the petasma are relatively short,
and do not much overlap the distal ends of the lateral ribs. In the four
juvenile males of P. brevirostris from Sinaloa, the disunited petasmal en-
dopods are very much like the above, although there is in all but one tooth
visible on the free margin of the tip of the lateral ribs, and although there
may be as few as four teeth on the proximomedian margin. Juvenile males
from Nicaragua and Panama are too small to display recognizable petasmal
characters.
In adult and juvenile males of P. calif orniensis, the external margin of
the appendix masculina of the second pair of pleopods is approximately
straight. In adult males of P. brevirostris from Panama and in juveniles
from Sinaloa the external margin of the appendix is deeply concave. In
Panamanian and Nicaraguan juveniles, the carapace length of which is
less than 10.5 mm., the external margin of the appendix masculina is
straight, but this organ, like the petasma, is extremely small and undevel-
oped in these very small specimens and presumably does not yet display the
adult characters.
Penaeopsis A. Milne Edwards (Bate).
Penaeopsis, Burkenroad, 1934b, p. 7.
Subgenus metapen aeopsis Bouvier.
Metapenaeopsis, Burkenroad, 1934b, p. 8.
SECTION I, Burkenroad, 1934b, p. 8.
Penaeopsis I Metapenaeopsisi kishinouyei (Rathbun).
Parapenaeus kishinouyei, Rathbun, 1902, p. 288.
Penaeopsis kishinouyei, Schmitt, 1924, p. 161.
Not Penaeopsis kishinouyei, Boone, 1930, p. 108; 1931, p. 176.
(Text-figures 16, 17).
Range: Galapagos Islands; Revillagigedo Islands; 2-55 fathoms.
Material: Two specimens (1 $, 1 9) were taken 3 miles off Pyramid
Rock, Clarion Island, Revillagigedo Islands (Station 163 D-2, Cat. No.
36,973) in 55 fathoms.
Dimensions and Sexual Condition: Female an impregnated adult of
carapace 12 mm., total length about 56 mm. Male a subadult with joined
petasmal endopods of carapace 5 mm., total length about 27 mm.
Remurks : The present record of Penaeopsis kishinouyei, although from
a locality nearly 20° to northward of the Galapagos from which the species
is otherwise definitely known, and from deeper water, refers like the
original record to the neighborhood of an isolated group of oceanic islands.
Whether the species occurs on the mainland, from which it has twice been
recorded by Boone, is extremely doubtful; thus the Panamanian specimen
in the collection of the American Museum of Natural History, referred to
P. kishinouyei by Boone, 1931, p. 176, is entirely distinct from Rathbun’s
species and will be discussed in a further paragraph as P. beebei sp. nov.
(not P. mineri Burkenroad, to which closely related form I have erroneously
1938]
Burkenroad: Penaeidae
73
referred the Panamanian specimen in 1934b, p. 28). As to the Costa Rican
specimen referred to P. kishinouyei by Boone, 1930, p. 108, according to
her Plate 33 the fifth legs of this specimen are nearly as long as the third,
the telson lacks conspicuous lateral armature, the antennular flagella are
longer than their peduncle, and the exopodites of the walking-legs are
swollen and bladder-like ; I would therefore infer that it may perhaps belong
to another genus than Metapenaeopsis. As a possible analogy, mention may
may be made of figure 4B in the same paper, which although referred to
“Penaeopsis goodei,” obviously represents the petasma of Trachypeneus.
If the Costa Rican specimen is not a Metapenaeopsis, it is necessary to
assume that such portions of Boone’s description as that (p. 110) referring
to the sternal spines characteristic of Metapenaeopsis may, like the descrip-
Text-figures 16 & 17.
Penaeopsis kishinouyei (Rathbun). 16. Thelycum; adult
$, D. T. R. 36,973, x 10.
Penaeopsis kishinouyei (Rathbun). 17. Petasma (posterior
view, “dust-cap” bent down to expose distal parts) ;
subadult $, D. T. R. 36,973, x 25.
tions of thelycum and petasma in the same account, be drawn from Rath-
bun’s remarks on P. kishinouyei rather than from the animal figured on
Plate 33. Boone’s unrecognizable figure 3A, given as an original drawing
of the thelycum of P. kishinouyei (although the material which she has
examined is stated to have consisted of “one adult male”) bears a great
resemblance to the similarly unrecognizable figure 4A given by the same
author as that of “ Penaeopsis goodei.” It must be admitted that this figure
is rather suggestive of Metapejiaeopsis, by reason of the occurrence of a
spine at the anterior end of what seems intended to represent the median
plate of XIII ; however, it seems unlikely the drawing can refer to P. kish-
inouyei, since at a carapace length of 11 mm. the median plate seems to
have lost its armature in that species.
Accepting that P. kishinouyei has not yet been taken on the mainland,
its limitation to oceanic islands off the Pacific American coast would seem
a matter of considerable interest, inasmuch as the mainland would seem
as readily accessible to a littoral animal of the Galapagos as would the
Revillagigedo Islands.
For comparison with the present Revillagigedo specimens, one of the
two juvenile females taken off Indefatigable Island by the Williams Gala-
pagos Expedition, and reported by Schmitt, 1924, p. 161 ; as well as three
74
Zoologica: New York Zoological Society
[XXIII :3
juvenile females (the largest of carapace 5.2, total length about 25 mm.)
and a juvenile male (of carapace 4 mm.; petasmal endopods not joined)
taken at Arcturus Station 54 (Gardiner Bay, Hood Island, Galapagos;
April 25, 1925; 15 feet), from the collection of the Department of Tropical
Research, have been available.
P. kishinouyei differs from the related American species in that the
basis of the second leg is constantly unarmed. Its rostrum also seems to
bear a smaller number of teeth than in the other species, the average for
the six available specimens being 8+1.
The petasma of the juvenile male from the Galapagos displays in
recognizable form the peculiarities of the Revillagigedo subadult. The dis-
tinctive features of the petasma of P. kishinouyei are, that the lateral
shoulder of the distolateral lobe of the right endopod bears a short obtuse
conical projection; the main or ventral lobule of the distolateral lobe is a
blunt cone armed on the dorsal rim of its distal end with a single lunate
row of spinules; the dorsal lobule of the distolateral lobe is an undivided
projection much smaller than the ventral lobule; the projection or “dust-
cover” of the distoventral lobe is distally deeply cleft into a small right
and large left lobule, the right partially folded under the left.
The present Revillagigedo female, by far the largest specimen of
P. kishinouyei so far recorded, agrees with Rathbun’s figure 15, PI. XII,
1902, of the Galapagan type of 33 mm., total length, in that the anterior
margin of the median plate of the thelycum is unarmed. As regards cer-
tain other features of Rathbun’s drawing, particularly the enormous bilobed
form indicated to characterize the coxa of the fourth legs, I find no cor-
respondence with available material. Rathbun’s figure is incomplete, the
posterior part of the fourteenth sternite not being represented. In the
available juvenile females from the Galapagos, the thelycum differs from
that of the northern adult in that the posterior parts of the median plate
are shorter and broader, and the anterior margin armed with a median
spine; these differences seem however to be referable to youth. The thely-
cum of P. kishinouyei differs from that of the related Pacific forms (aside
from the disappearance of the anteromedian spine in adults, a unique fea-
ture) in that the semicircular median elevation of the thirteenth sternite is
not posteromedially depressed; and the posterior parts of the median plate
of XIII are moderately elongate, little elevated, and taper to their posterior
ends. There appears to be a pair of sac-like membraneous sperm-receptacles
invaginated beneath the lateral hoods of sternite XIV. The transverse plate
of XIV is about three or four times as broad as long; its posterior half
slopes sharply down to well below the level of the median part of the trans-
verse ridge of the posterior margin of the sternite. The anteromedian edge
of the transverse plate does not overhang the posterior parts of XIII or the
anteromedian prolongation of XIV.
Penaeopsis ( Metapenaeopsisl b eebei1 sp. nov.
(Text-figures 18, 19).
Type: Type and cotypes, Cat. No. 36,983 (4 $, 4 $), Department of
Tropical Research, New York Zoological Society. Taken at Station 136,
Dredge 30, Arena Bank, Gulf of California, 23° 27' N. Lat., 109° 24' W.
Long., 4-foot dredge, 35 fathoms. May 1, 1936.
Range : Gulf of Panama; Lower California, 3 to 50 fathoms.
Material: A total of 106 specimens (almost equally divided as to sex)
was taken at San Lucas Bay (Station 135) and at Arena Bank (Station
4 Named for Dr. William Beebe of the Department of Tropical Research, New York Zoological
Society.
1938 J
Burkenvoad : Penaeidae
75
136), at the southern end of Lower California, as follows: Station 135:
D-l (4 3, 1 9), D-9 (20 3, 21 9), D-ll (9 3, 16 9), D-20 (1 3), D-18,
D-19, D-21 to D-25 (4 3, 5 9), D-26 (1 3, 1 2) ; Station 136: D-6 (1 3,
1 9), D-21 (1 3), D-24 (2 3, 5 9), D-26 (1 3)", D-27 (2 3, 1 9), D-30
(4 3, 4 9). Cat. Nos. 36,974, 36,975, 36,1027, 36,977, 36,978, 36,979, 36,980,
36,981, 36,982, 36,983, 36,1028.
Dimensions and Sexual Condition : Females ranging in carapace length
from 13.5 mm. (impregnated adult) to 4.5 mm. (juvenile) ; total length at
carapace 10 mm. about 44 mm.; at 5 mm. about 21 mm. Males ranging in
carapace length from 11 mm. (adult) to 4 mm. (juvenile) ; total length at
carapace 9 mm. about 42 mm., at 5 mm. about 23 mm. Petasmal endopods
joined and of adult appearance down to a carapace length of 6 mm. Both
juveniles and adults are included in all of the more extensive samples, with-
out any indication of segregation as to either locality or depth.
Text-figures 18 & 19.
Penaeopsis heebei, n. sp. 18. Thelycum (position of right
sperm-receptacle indicated by cross-hatching) ; adult
9 type, D. T. R. 36,983, x 7. 19. Petasma (posterior
view, “dust-cap” bent down to expose distal parts) ;
adult 3 type, D. T. R. 36,983, x 15.
Diagnosis: Vestigial anterior arthrobranch present on the thirteenth
somite. Distoventral projection of the left side of the asymmetrical petasma
reduced to a vestige.
Rostrum modally with 9 teeth in addition to the epigastric. The sec-
ond pleonic tergite lacks any trace of median longitudinal carina; the third
is compressed but not sharply carinated. Posterior pair of mobile lateral
spines of the telson exceeding by little or nothing the fixed pair succeeding
them. Stylocerite less than half as long as the external margin of the basal
segment of the antennular peduncle. Basis of second leg usually spinose
but sometimes unarmed.
Median elevation of thirteenth sternite of the female with a subtri-
angular outline, its posteromedian portion much depressed below the an-
terior and lateral rim; the anteromedian margin armed with an anteriorly
directed spine of moderate size. Posterolateral parts of the median plate
of XIII tear-shaped in outline, anteromedially depressed, posterolaterally ele-
vated. A pair of membraneous sperm-receptacles invaginated to lie internal
to the lateral hoods of the fourteenth sternite. Transverse plate near the
posterior margin of XIV more than six times as broad as long; the antero-
median margin slightly convex but not produced to overhange the postero-
76 Zoologica: New York Zoological Society [XXIII :3
lateral parts of XIII; the posterior margin depressed below the level of the
posterior rim of the stern ite.
Lateral shoulder of the distolateral lobe of the right endopod of the
petasma a gently curved swelling; main or ventral lobule in ventral view
a saddle-shaped transverse elevation crowned with a row of spinules on
each of its two peaks, the lateral of the two rows of spinules much more
extensive than the median one. Dorsal lobule of the distolateral lobe a
subdivided projection nearly as bulky as the ventral lobule but almost en-
tirely concealed behind it in ventral view.
Remarks: Penaeopsis beebei very nearly resembles P. mineri Burken-
road; and indeed, was confused with the latter by Burkenread, 1934b, pp.
25-26, since the Panamanian male in the collection of the American Museum
of Natural History which was there considered to be a variant individual
of P. mineri can now be recognized as a representative of P. beebei.
P. beebei is distinguishable in non-sexual characters from P. gooclei
(Smith) [and from the African P. pubescens (Bouvier) ] by its relatively
shorter stylocerite and the weaker c-arination of its pleon; and from P. pube-
scens by the relatively much shorter mobile lateral spines of its telson. It
differs from all of the American members of its section in the inconsistency
of the armature of the basis of its second leg, which appendage appears to
by constantly unarmed in P. kishinouyei, constantly armed in P. mineri and
the two Atlantic American species. In the available material of P. beebei,
a well-developed spine is present on the basis of the second leg in 78 in-
dividuals, a rudimentary one in 8 individuals, and no spine at all in 19 in-
dividuals. Although the spine modally diminishes in relative size with
decrease in size of the individual, its absolute presence or absence seems to
be independent of the size or sex of the individual. Even at its maximum,
the spine is smaller in P. beebei than in P. mineri.
In rostral, as in pereiopodal armature, P. beebei falls between P. kishi-
nouyei and P. mineri, as tabularly indicated below.
Number of Rostral
Teeth.
Nun
P. kishinouyei.
nber of Individu
P. beebei.
als.
P. mineri.
13+1 epigastric
0
0
1
12 + 1
0
0
5
11 + 1
0
0
23
10 + 1
0
4
23
9 + 1
2
63
5
8 + 1
2
22
0
7 + 1
2
0
0
In the 4 individuals of P. beebei with ten rostral teeth, and in somewhat
more than half of the 63 with nine, the anteriormost tooth is rudimentary.
Variations in rostral ai'mature show no conspicuous correlation with size
of individual, but the number of teeth tends to be greater in the females.
The rostrum of P. beebei, as that of other species of the section, varies in
length with size of the individual, ranging in extent from the end of the
first to the middle of the second segment of the antennular peduncle. As
in P. kishinouyei and P. mineri, the rostrum of P. beebei is somewhat
deeper proximally than it is in the Atlantic American species.
P. beebei agrees with all of the American species except P. goodei
1938J
Burkenroad : Penaeidae
77
(Smith) in that the females possess a pair of sac -like, membraneous sperm-
receptacles invaginated from the lateral portions of the groove between the
twelfth and thirteenth sternites. The thelycum is very similar to that of
P. mineri from which it differs as regards the subtriangular rather than
semicircular outline of the median elevation of XIII which is armed with a
somewhat smaller anteromedian spine (this spine being relatively larger in
P. mineri than in any of the other American species). The posterior parts
of XIII are in P. beebei much less narrow, elongate and elevated than in
P. mineri; the transverse plate of XIV is shorter and its anterior margin
much less produced ; the thelycum thus appears much less cuplike than in
P. mineri. The anteromedian margin of the transverse plate of XIV is
convex rather than notched as in P. mineri.
To an even greater extent than as regards thelycum, the petasma of
P. beebei resembles that of P. mineri much more nearly than that of any
other species. The principal differences are that in P. mineri the lateral of
the two rows of spinules crowning the eminences of the ventral lobule of
the distolateral lobe is hardly more extensive than the median row, instead
of much more extensive as in P. beebei; and the lateral part of the dorsal
lobule of the distolateral lobe projects considerably beyond the ventral lobule
in ventral view instead of being hidden behind it as in the present species.
Penaeopsis I Metapenaeopsis) mineri Burkenroad.
P. (M.) mineri, Burkenroad, 1934b, p. 25, part.
Range : Lower California, both coasts, 7 to 30 fms.
Material: A total of 61 specimens (the males slightly outnumbering
the females) was taken at Magdalena Bay, Pacific coast of Lower Cali-
fornia (Station 131), San Lucas Bay on the southern coast (Station 135)
and Santa Inez Bay on the Gulf coast (Stations 141 and 142) at depths of
6 to 30 fathoms, as follows: Station 131 : D-l (6 $, 10 2) ; Station 135: D-ll,
D-12 (2 a); Station 141: D-l (3 $), D-4 (2 $, 5 2) ; Station 142: D-l
(23 $, 10 2). Cat. Nos. 36,984, 36,985, 36,986, 36,987, 36,988.
Dimensions and Sexual Condition : Females ranging in carapace length
from 11.5 mm. (impregnated adult) to 5 mm. (juvenile) ; males from 11
to 5 mm. The petasmal endopods are already joined distally in the smallest
specimen available, but are of fully adult condition only above 6 mm. No
particular indication of local segregation as to size or sex is discernible.
Remarks : Although P. mineri has so far been recorded only from Lower
California, the fact that there is but one specifically determinable record of
the genus from the Pacific American mainland south of 22° N. permits no
speculation as to the actual range of the species. It is perhaps interesting
that although their ranges overlap, P. mineri was taken in company with
P. beebei only once, at Station 135, and then in very small number.
The present extensive material requires modification of the original
description in certain particulars. The rostral armature varies from 9 + 1
to 13 + 1, the modal number being 10 or 11 teeth in addition to the epi-
gastric; the extremes are infrequently found. Recognition of the “variant”
Panamanian male erroneously referred to the present species in 1934b
(p. 26) as representing the closely related but distinct species P. beebei
disposes of the opinion previously held that the basisal spine of the second
leg in P. mineri is variable in occurrence. The distoventral projection of
pars externa of the right petasmal endopod is not, as described in 1934b
(p. 28), shallowly cleft into “a small right and a large left lobe,” but into
two subequal lobes the left of which partially conceals the right in ventral
view; the essential distinction between the projection in P. mineri and in
P. kishinouyei is therefore that in the latter the left lobule is actually as
78
Zoologica: New York Zoological Society
[XXIII :3
well as apparently considerably larger than the right. Rathbun’s statement
that the “left half” of the “petasma” of P. kishinouyei is longer than the
right evidently refers to the distoventral projection of pars externa of the
right endopod only; and not, as interpreted in 1934b (p. 28) to the actual
halves of the petasma, the two endopods.
In the comparison of the thelycum of P. mineri with that of the
Atlantic American species of Metapenaeopsis, given in 1934b, the statement
is made (p. 28) that the transverse groove between sternites XIII and XIV
is in P. mineri “anteromedially so shallow . . . that it is probably no more
than a vestige bereft of the sperm-storing function which this portion of
the groove possesses in P. goodei and P. smithi;” and I have also stated
(p. 10) that “in the American species of Metapenaeopsis, no portion of this
depression [the transverse groove] is especially expanded; the spermato-
phores enter each lateral half of the groove by median openings, the sperm
being extruded at the lateral, anterior ends of the groove.” These state-
ments are not entirely correct, and a more precise description of sperm-
storage in these forms is therefore offered here, as follows:
In P. goodei (as described in 1934b, p. 18), the transverse groove is
on either side quite deep in its entire length, and the spermatophore is in-
troduced into the elongate, U-shaped receptacle of either side at its antero-
median end, on the inner side of the posterior part of the median plate.
The median limb of the U is occupied by an elongate cylinder of sperm-
free material; while the sperm lie near the exit of the receptacle, at the
anterolateral end of the U, which is not particularly expanded for this func-
tion. In P. smithi, on the contrary, the part of the transverse groove
homologous with the median limb of the U in P. goodei is shallow, and is not
employed in spermatophore storage; and the entrances of the receptacles
seem to lie near the posterior ends of the lateral limbs of the transverse
groove, on the outer sides of the posterior parts of the median plate ; instead
of on the inner sides of the posterior parts at the anterior ends of the
median limbs of the ti'ansverse groove, as in P. goodei. Further, the anterior
ends of the lateral limbs of the transverse groove in P. smithi, near the
exits, are quite definitely expanded as membraneous sacs into which the
sperm-bearing portion of the spermatophores is introduced; the sperm-free
secretion of the spermatophore seems much reduced in quantity as com-
pared with P. goodei, commensurate with the reduction of that part of the
groove in which the secretion might be stored. The receptacular arrange-
ments in the three Pacific American species of the subgenus are very
similar to those of P. ( M .) smithi, although there is little resemblance
in superficial sculpture. It therefore appears that as regards the method of
sperm storage, P. goodei is unique among its American congeners; and it
seems possible that P. mineri, P. beebei, and P. kishinouyei are more closely
related to P. smithi than to P. goodei and P. pubescens (to which P. kishi-
nouyei displays some superficial resemblance in sculpture of the thelycum).
It is probable, however, that the three Pacific species are more closely re-
lated to one another than they are to any of the Atlantic forms.
Trachypeneus Alcock.
Subgenus trachysalambria Burkenroad, 1934b, p. 49.
section 1, Burkenroad, 1934a, p. 94.
Trachypeneus similis paciticus Burkenroad.
T. ( T .) s. pacificus, Burkenroad, 1934b, p. 50.
Range : Gulf of Panama; Lower California, both coasts; 12-24 fms.
Material : Five specimens (1 $, 4 5) were taken at Magdalena Bay,
1938J
Burkenroad : Penaeidae
79
Pacific coast of Lower California, in 12 fathoms, March 29, 1936. (Station
131 D-l, Cat. No. 361,026).
Dimensions and Sexual Condition : Females ranging in carapace length
from 8 to 12 mm.; male 7.5 mm. Juvenile.
Remarks’. In the adults of T. s. pad ficus the anterior margins of the
produced posterior lips of the transverse groove are conspicuously truncated,
as stated in the original description. In the juveniles, however, this trunca-
tion is hardly discoverable, the anterior margins of the lips being roundedly
convex somewhat as in the Atlantic American species T. constrictus (Stimp-
son). In the Atlantic American T. s. similis (Smith) the truncation develops
with age as in T. s. pacificus, but the lips are never so decidedly convex as
they are in the juveniles of the Pacific subspecies. Other diagnostic dif-
ferences between T. similis and T. constrictus, listed in 1934b, are little
affected by growth in subadult stages.
SECTION 2, Burkenroad, 1934a, p. 94.
Trachypenens (Trachysalambrial brevisuturae Burkenroad.
T. (T.) brevisuturae, Burkenroad, 1934b, p. 55.
(Text-figures 20, 21).
Range: El Salvador; Lower California; 3-35 fms.
Material : A total of seven specimens (of which two are males) was
taken at San Lucas Bay (Station 135) and at Arena Bank at the southern
end of Lower California (Station 136) in 3 to 35 fathoms, as follows:
Station 135: D-l (1 $, 3 5), D-6 (1 $) , B-18 (1 9) ; Station 136: D-30 (1 9).
Cat. Nos. 36,989, 36,990, 36,991, 36,992.
Dimensions and Sexual Condition : Females ranging from carapace
length 15 mm., total about 58 mm. (impregnated adult) to carapace length
5 mm. (juvenile). Both males of carapace length 7 mm. (subadults with
joined petasmal endopods).
Remarks : The present Californian material seems in close agreement
with the Salvadorean holotype of the species, a subadult male of carapace
length 6.5 mm. There is very little difference in non-genital features be-
tween adults and juveniles or males and females; except that in the larger
21
Text-figures 20 & 21.
Trachypeneus brevisuturae MDB. 20. Carapace (anterior part,
lateral view); adult 9, D. T. R. 36,991, x 5.5. 21. Thelycum;
as in 20, x 12.
80
Zoologica: Neiv York Zoological Society
[XXIII :3
specimens the rostrum is somewhat longer than in the holotype, reaching
as far as to the middle of the second segment of the antennular peduncle.
In 4 of the present specimens there are seven rostral teeth in addition to
the epigastric; in 3, eight. The branchiostegal carina is very shallow, but
quite perceptible; it is obsolescent, rather than “obsolete” as stated in the
original description.
Confirming the allocation of Trachysalambria brevisuturae to Section 2
of the subgenus, the thelycum resembles that of the other, Indo-Pacific,
species included in the Section, in that the posteromedian lip of the trans-
verse groove is not produced forward as a pair of flaps divided by a deep
incision in the manner diagnostic of Section 1. The posterior lip of the
transverse groove of the present species, although it is continuous across
the midline in the fashion diagnostic of the subgenus, overhangs a median
cavity much shallower even than that in the other species of the Section.
The thelycum is superficially distinguishable from that of T. curvirostris and
its allies in that the posterior lip of the transverse groove is in adults (but
not in juveniles) medially convex, rather than concave; and in that the
portion of sternite XIV lying behind the level of the bases of the fifth legs
is longer than the anterior part, rather than shorter. The heart-shaped
median plate, which is considerably depressed in the midline, especially
posteriorly, bears an anteromedian spine reminiscent of that in Meta-
penaeopsis in the three juvenile females of D. T. R. 36989.
Eusicyonia Stebbing.
division I, Burkenroad, 1934b, p. 71.
Eusicyonia disparri Burkenroad.
E. disparri, Burkenroad, 1934b, p. 83.
(Text-figures 22, 23).
Range : Southern and Gulf coasts of Lower California, beach5 to 35
fathoms.
Material : A total of 3 specimens (2 8, 1 2) was taken at Arena Bank
(Station 136) and at Santa Inez Bay (Station 142) in 30 to 35 fathoms, as
follows: Station 136: D-30 (1 3); Station 142: D-l (1 8, 1 2). Cat. Nos.
36,993, 36,994.
Dimensions and Sexual Condition : Female of carapace length 7 mm.,
total about 29 mm. Larger male of carapace length 6.9 mm.; smaller of
carapace 4 mm., total length 15.5 mm. The petasma of both males is of
adult form.
Remarks : The female of the present collection is in detailed agreement
with the types. The larger male, taken in the same catch as the female,
agrees well as to sculpture and armature of carapace and pleon, except that
the posteriormost rostral tooth is somewhat farther in advance of the
orbital margin than in material previously recorded. The small male (D. T.
R. 36993) differs quite considerably from other known specimens of E. dis-
parri, which are all of considerably larger size, in that the rostral tip bears
only four teeth, the third rostral tooth is not a great deal farther behind the
tip than it may be in E. laevigata (Stimpson), the first or posteriormost
rostral tooth is not a great deal closer to the orbit than it sometimes is in
E. laevigata (Stimpson), the anterior postrostral tooth is conspicuously
smaller than the two following ones (although it is relatively much larger
than even the extreme in E. laevigata), and finally, the anterior three
pleonic somites bear a trace of posterior pleural sulcus, much shorter and
5 “Beach” record based on specimens taken in the seine by the Pawnee, at depths presumably
not greater than 5 fathoms.
1938J
Burkenroad: Penaeidae
81
shallower than in E. parri Burkenroad, but quite definite; and the anterior
tergal of the fourth somite is continuous, as in E. carinata (Olivi), not in-
terrupted as in E. laevigata, E. parri, and the other available specimens of
E. disparri. In spite of these differences from the specific norm it seems
to me safest for the present to identify D. T. R. 36993 with E. disparri.
The petasma of E. disparri, which has not previously been described,
very nearly resembles that of E. laevigata; the distolateral projections are
however turned conspicuously inward, somewhat as in E. carinata, instead
of extending straight distally in ventral view as in E. laevigata. A male of
E. parri taken by Dr. Beebe in Bermuda (which agrees very well in non-
sexual characters with the female holotype from the Bahamas) resembles
the present males of E. disparri in form of sexual appendage, except that
the proximal border of the notch in the lateral margins of the petasma pro-
trudes beyond the margin proximal to it as a prominent shoulder.
Text-figures 22 & 23.
Eusicyonia disparri MDB. 22.
Petasma (posterior
view) ; adult $, D. T. R.
36,994, x 14. 23. Eye (of
left side, dorsal view) ;
adult 9, D. T. R. 36,994.
x 14.
Since the variant small male referred with some doubt to E. disparri
above softens several of the differences previously given as distinguishing
this species from related forms, especially E. parri, certain additional dis-
tinctions between the Atlantic and the Pacific forms may be considered.
In all five of the available specimens of E. disparri the middle part of
the posterior margin of the pleura of the fourth pleonic segment is decidedly
concave and is ventrally set off by a well-developed angle which in the larg-
est specimens is acute. In both specimens of E. parri (as in E. laevigata
and E. carinata), the midsection of the posterior margin of the fourth
pleui'a is straight and the posterior rounds into the ventral margin without
a defined angle. In the five specimens of E. disparri the ventral margins
of the rostrum, near its distal end, bear one or two small but conspicuous
mobile spinules on either side, which are absent in both specimens of
E. parri. In four of the five specimens of E. disparri the rostrum bears a
fourth tooth on its dorsal margin just behind the tip, which is not present
in either example of E. parri.
The first of the above characters may perhaps be subject to variation;
the second probably is (inasmuch as in a male and a female of E. laevigata
from the west coast of Florida the mobile rostral tooth [similar to those of
E. disparri ] present in all fourteen members of the species previously ex-
amined by me, is absent) ; and the third is certainly variable; these differ-
ences therefore appear not to be diagnostic. A diagnostic between E. parri
and E. disparri seems however to be provided by size and shape of eye. In
the Atlantic form the distal, cornea-bearing section of the ocular peduncle
is relatively longer and is much less expanded at its distal end; and the
cornea is smaller and is set more transversely than in the Pacific material.
The difference in peduncular proportions may be expressed by the ratio of
the length of the outer margin of the peduncle to that of the carapace and
Zoologica: New York Zoological Society
82
[XXIII :3
to the breadth of the peduncle across its dorsal surface at base of cornea,
as follows:
Species.
Eusieyonia parri.
Eusieyonia disparri.
Sex.
9
8
8
8
9
9
9
Carapace length
in mm.
6.3
8.0
4.0
6.9
7.0
9.2
9.3
Ratio of length of
outer margin of dis-
tal pai’t of ocular
peduncle to carapace
length.
.14
.13
.10
.11
.11
.09
.09
Ratio of length to
breadth of ocular
.83
.76
.58
.52
.53
.52
.53
peduncle.
division II, Burkenroad, 1934b, p. 73.
Eusieyonia disedwardsi Burkenroad.
E. disedwardsi, Burkenroad, 1934b, p. 86.
Range: Lower California, Gulf coast and southern tip; 6 to 50 fathoms.
Material : A total of 33 specimens (about half of them males) was
taken at San Lucas Bay (Station 135), Arena Bank (Station 136) and Santa
Inez Bay (Stations 141, 142, 143) in 6 to 50 fathoms, as follows: Station
135: D-l (3 8, 1 9), D-9, D-10 (1 9), D-26 (1 9); Station 136: D-l (1
9), D-6 (2 8), D-24 (2 8, 2 9), D-30 (4 8, 2 9); Station 141: D-l (1 9),
D-3 (1 8, 1 9), D-4 (3 8, 1 9) ; Station 142: D-l (2 8, 3 9), Station 143: D-l
(19). Cat. Nos. 36,995, 36,996, 36,997, 36,998, 36,999, 361,000, 361,001,
361,002, 361,003, 361,004, 361,005, 361,006.
Dimensions and Sexual Condition: Females ranging in carapace length
from 16.2 mm. to 7 mm.; males from 15.3 mm. to 4.6 mm. Total length
of the largest male, about 59 mm. The haul from Station 136, D-24, com-
posed of the four largest specimens, is the deepest one yielding material of
the species; but although the smaller specimens are often those from the
least depths, the correspondence is by no means complete. The size of the
males at maturity seems quite irregular; there are specimens with unjoined
petasmal endopods ranging in carapace length from 6.8 mm. to 8 mm., and
specimens with the endopods hooked together at the distal end only ranging
in carapace length from 5.5 mm. to 10 mm.
Remarks: Eusieyonia disedwardsi was described with some misgivings
from a single subadult male; and it is therefore with considerable satisfac-
tion that a confirmation of its right to specific distinction is now offered
on the basis of an ample material.
Petasma, male genital sternites, and thelycum are practically identical
with those of E. edwardsi, except that the tips of the distolateral projec-
tions of the petasma are often directed rather conspicuously away from the
midline instead of parallel to it. Differences from E. penicillata in petasma,
previously described, are completely constant in adults and are recognizable
even in very juvenile stages. Differences in male genital sternite and in
thelycum between E. disedwardsi and E. penicillata, identical with those be-
tween the latter and E. edwardsi, are constant in adults but are somewhat
obscure in very juvenile individuals. The divergence of the tips of the
stylets of the ocular somite, constant at all sizes, provides a clear though
minute distinction from E. penicillata.
1938]
Burkenroad: Penaeidae
83
Differences in pleonic sculpture from E. edwardsi are constant and
diagnostic; although the traces of posterior pleural sulcus are often stronger
than in the holotype, there is generally a faint trace of posterior tergal on the
first somite, and the dorsal end of the posteromedian pleural often reaches
the anterior margin as a groove which although shallower than in E. ed-
wardsi and E. penicillata is not completely obliterated. In large adults the
pleura of the fourth somite often bear a posterior ventral tooth or angle.
The rostrum is almost constantly longer than the maximum in E. peni-
cillata, and is usually more elevated than in the related Pacific form. As a
very obvious distinction from both E. edwardsi and E. penicillata, it may be
observed that the rostrum of every available individual of E. disedwardsi
bears two teeth behind the bifurcated tip (in addition to the “anterior
tooth of the carapace” which is occasionally placed in advance of the
orbital margin) ; whereas in the two related species not more than one
individual in ten bears more than one rostral tooth behind the tip.
It may be noted that in one specimen of E. disedwardsi the middle tooth
of the carapace, in contrast to its usual position, lies anterior to the level
of the hepatic spine.
Like E. penicillata and several other Pacific species less closely related,
E. disedwardsi is marked with a large and conspicuous ring or ocella of
pigment on the branchial region which seems, peculiarly enough, to be absent
in E. edivardsi and in the other Atlantic members of the genus.
Eusicyonia penicillata (Lockington) .
Sicyonia penicillata, Lockington, 1879, p. 164; Pesta, 1915, p. 118.
Eusicyonia penicillata, Burkenroad, 1934b, p. 88.
Range: Lower California, both coasts; beach to 40 fathoms.
Material: A total of 37 specimens (somewhat more than half of these
being males) was taken at Santa Inez Bay (Station 144), at San Lucas
Bay (Station 135), east of Cedros Island (Station 126), and at Magdalena
Bay, in 2 feet to 40 fathoms, as follows: Magdalena Bay, “In 2 to 3 feet of
water under stone” (17 $, 14 9) ; Station 126: D-3 (3 $, 1 9) ; Station 135:
D-ll, D-12, (1 $ ); Station 144: D-7 (1 9). Cat. Nos. 361,007, 361,008,
361,009, 361,010.
Dimensions and Sexual Condition: Material from Station 126 (a female
of carapace length 18 mm. and males of 12.4 to 16.5 mm.), falls within the
range of sizes previously recorded (1934b, p. 88) and is fully adult; while
the male from Station 135, of carapace length 9.6 mm., although of smaller
size, appears to be adult. The series from very shallow water in Magdalena
Bay is however composed exclusively of juveniles, the first which I have been
able to examine. Females of this series range from carapace length 12 mm.
down to carapace 4 mm., total length 17.5 mm.; and males from carapace
8 mm. to carapace 4.9 mm., total length 18.5 mm. The single female from
Station 144 is also a juvenile, of carapace length 6.9 mm.
Remarks : The present seem to be the first records of the species from
outside the Gulf of California.
In my previous account of E. penicillata, I have unfortunately over-
looked the description by Pesta, 1915. It may be noted that the otherwise
excellent figure 7 of the petasma, there given, is inaccurate in one detail,
in that the accessory filaments originate near the middle of the distolateral
projections, rather than, as shown, at their bases.
The present juveniles of E. penicillata display little change from adult
form in non-sexual characters, except that as in many related species the
sharp angle terminating the hinder end of the dorsal carina of the fifth
pleonic somite becomes more rounded with diminishing size until in the
84
Zoologica: New York Zoological Society
[XXIII :3
smallest individuals there is no trace of it. The notch in the posterior
margin of the thirteenth sternite loses its narrow subrectangular outline in
the juveniles and becomes a shallow concavity; and the ridge across the
posterior margin of the thirteenth sternite of the male loses its well-bowed
appearance, chiefly by the reduction in relative size and the increased separa-
tion of the lateral protuberances.
In males of carapace length 7.5 to 8 mm. the petasmata are hooked
together only at their anterior ends, and the filaments of the various pro-
jections are relatively shorter than in the adult. Below 7.5 mm. the endopods
are unattached, but the adult characteristics of the petasma are recognizably
indicated even in the smallest individuals, where the accessory filament
is represented by a well-defined shoulder about one-third from the tip
of the curtailed but very slender distolateral projection.
Eusieyonia aliaffinis Burkenroad.
Eusicyonia aliaffinis, Burkenroad, 1934b, p. 92.
(Text-figures 25, 27).
Range: Pacific coast of Mexico; southern tip of Lower California; 13
to 30 fathoms.
Material: One male was taken at San Lucas Bay in 13 fathoms (Sta-
ton 135 D-26, Cat. No. 361,011).
Dimensions and Sexual Condition: The specimen, of carapace length
8.4 mm., is smaller than the holotype, but its petasma is fully developed and
of adult form, rather than disunited and juvenile as in the type.
Remarks : The present male agrees precisely in critical non-sexual char-
acters with the Mexican holotype. The petasma shows no very striking
differences from that of E. affinis, except in the non-projection of a small
cusp near the median base of the distolateral projections which is con-
spicuous in an available large type male of Faxon’s species.
In addition to the large specimens of E. affinis contained in the col-
lection of the Museum of Comparative Zoology, a comparison of which with
E. aliaffinis has been presented in my previous discussion of that species
(1934b), I have recently been enabled by the kindness of Dr. W. L. Schmitt
of the U. S. National Museum to examine the juvenile male of E. affinis
(U. S. N. M. 21169) which Faxon, 1895, p. 179, records from Albatross
St. 3369, off Cocos Island, in 52 fathoms. This specimen (figure 24), of
carapace 8.8 mm., total length 35 mm., is of the same size as the holotype
of E. aliaffinis. The agreement of this specimen in all critical characters
with the much larger individuals of E. affinis previously examined, and its
sharp contrast with E. aliaffinis, seems to provide justification for my
dismissal in 1934b, p. 94, of the possibility that E. aliaffinis might “repre-
sent a developmental stage of E. affinis.” The possibility that E. aliaffinis
and E. affinis may represent extremes of variation connected by intergrad-
ing forms is discussed in the next paragraph.
Eusicyonia species.
(Text-figures 26, 28, 29, 30).
Range: Known only from Arena Bank at the southern tip of Lower
California; 45-50 fathoms.
Material: Two specimens (1 $, 1 2) were taken at Arena Bank at depths
of 45 to 50 fathoms, as follows: Station 136: D-13 (IS), D-24 (12). Cat.
Nos. 361,012, 361,013.
1938J
Burkenroad, : Penaeidae
85
Dimensions and Sexual Condition: The male from Station 136 has a
carapace length of 12 mm., total length about 60 mm. The carapace length
of the female is 12 mm. Both specimens appear to be fully adult.
Remarks : The present material, representing a species closely related
to if distinct from E. aliaffinis Burkenroad, agrees with the two known
specimens of the latter and differs markedly from E. affinis Faxon (figure
24 above) as regards the sharp definition of the angle terminating the
dorsal carina of the fifth pleonic somite; the angular outline of the ventral
margins of the pleonic pleura (especially those of the fourth somite, the
posterior ventral angle of which is in the male even spiniform; and of the
first somite, the anteroventral margin of which is in all decidedly concave
rather than conspicuously convex as in E. affinis ) ; and the coarse tubercula-
Eusicyonia affinis (Faxon). 24. Pleon (lateral view) ; juvenile $, U. S. N. M.
21,169, x 5.5.
Eusicyonia aliaffinis MDB. 25. Pleon (lateral view of anterior part, tuber-
cles indicated for first two somites only) ; subadult $, D. T. R. 361,011,
x 5.5.
Eusicyonia species. 26. Pleon (lateral view, tubercles indicated for first
two somites only) ; adult 9, D. T. R. 361,013, x 5.5
86
Zoologica: Neiv York Zoological Society
[XXIII :3
tion of the pleonic surface. On the other hand, however, the two specimens
from Station 136 differ considerably from E. aliaffiinis and approach
E. affinis in that the posteromedian pleural sulci of the second and third
pleonic somites extend dorsally only to within a third of the height of the
somite from the dorsal midline (instead of to within a fourth or nearer),
turn anteriorly at their dorsal ends, and are margined above by a conspicuous
longitudinal ridge.
Text-figures 27-30.
Eusicyonia aliaffinis MDB.
27. Rostrum (lateral
view) ; as in 25, x 6.
Eusicyonia species. 28. Ros-
trum (lateral view) ; as
in 26, x 6. 29. Rostrum
(lateral view) ; adult $,
D. T. R. 361,012, x 6. 30.
Petasma (distal part,
posterior view) ; as in
29, x 13.5.
The anteromedian pleural sulci of the first, second and third pleonic
somites of the two specimens from Station 136 seem to fall about midway
in length and definition between those of E. affinis and E. aliaffinis. The
rostrum seems longer than that of either E. affinis or E. aliaffinis, being
approximately half instead of two-fifths or less as long as the carapace and
extending far beyond the eyes; and it bears six teeth instead of five in
both specimens. While, however, the rostrum of the male from Station 136
is recurved nearly to the horizontal and the third tooth of the dorsal margin
lies far behind the trifurcate tip, that of the female is recurved to an
angle of elevation much greater even than in the two available specimens
of E. aliaffinis, the third dorsal tooth forms part of a trifurcate tip, and the
sixth tooth is placed on the ventral margin more than half as far from
the tip as the penultimate dorsal tooth; the rostrum of the female thus
superficially somewhat resembles that of E. carinata (Olivi). The trans-
verse ridge at the posterior margin of the fourteenth sternite of the male
from Station 136 resembles that of E. aliaffinis as regards its dimensions,
but is strongly arched. The petasma differs slightly both from that of
E. aliaffinis and that of the available large male E. affinis, in that the
distolateral projections are quite convergent; and in that the small cusp
near the median base of the distolateral projections is more conspicuous than
in E. aliaffinis, less so than in E. affinis.
Whether the present material should be regarded as a distinct species,
as a variant form of E. aliaffinis, or as an intergrade between E. aliaffinis
and E. affinis seems difficult of decision. The fact that the pleonic sculpture
is the same in E. affinis of both large and small size, which is in harmony
with what is known of other members of the genus, suggests that differences
between the various available groups of individuals of the E. affinis super-
species are at least not the correlates of differences in size. The provenance
of the present material seems to rule out local variation as an explanation
of its attributes. However, the total number of available individuals of
the superspecies is not yet sufficient to exclude the possibility that differences
between groups of these individuals are referable simply to individual
variation (and it may be noted that if the small specimen discussed on
1938J
Burkenroad: Penaeidae
87
p. 80 actually refers to E. disparri, its pleonic characters imply that con-
siderable departure from the specific norm may sometimes occur in species
of the genus). Although it seems to me possible that the present specimens
represent a third species distinct from both E. affinis and E. aliaffinis, their
systematic status may be left undefined until further material has become
available.
Eusicyonia picta (Faxon).
Sicyonia picta, Faxon, 1893, p. 210.
Eusicyonia picta, Burkenroad, 1934b, p. 95.
Range : Off Mariato Point and in the Gulf of Panama, 127 to 182
fathoms; southern and Gulf coasts, Lower California, beach11 to 60 fathoms.
Material: A total of six specimens (including 1 male) was taken at
Arena Bank (Station 136) and at Santa Inez Bay (Station 147) in depths
of 30 to 45 fathoms; as follows: Station 136: D-ll (1 2), D-12 (1 2), D-15
(1 2) ; Station 147: (IS, 2 2). Cat. Nos. 361,014, 361,029, 361,016.
Dimensions and Sexual Condition: Females ranging in length from
carapace 14 mm., total 56 mm., to carapace 8 mm. Male of carapace length
7.5 mm. The largest female is considerably smaller than Faxon’s type-speci-
men of 70 mm. The petasma of the male is of adult form.
Remarks : The rostral armature varies in the present material from
three dorsal and two terminal to four dorsal and three terminal teeth.
The spacing and position of the dorsal rostral teeth, and the length, eleva-
tion and curvature of the rostral blade are quite variable. Although Cali-
fornian material of E. picta has not been directly compared with the
Panamanian types, there seems little reason to doubt its identity with the
latter.
A direct comparison of E. picta with two females and a male, paratypes
of E. stimpsoni (Bouvier), from Blake Station 293, Barbados, 82 fms.,
which were very kindly loaned by Dr. F. A. Chace, Jr., of the Museum of
Comparative Zoology, confirms the suggestions previously made (Burken-
road, 1934b, p. 96) as to characters distinguishing the two species. In addi-
tion to the conspicuously greater breadth and depth of its carapace and
pleon, its longer rostrum with ventral tooth very distinctly separated from
the two dorsal elements of the tip, and the deeper sculpture of its pleon,
E. stimpsoni seems to differ from E. picta as follows : Its pleon is tuber-
culate, not punctate only as in E. picta. The lateral ridge of its rostrum
is usually concave dorsally, so that the ridge is closer and more nearly
parallel to the ventral margin of the rostrum than in E. picta where the
ridge is usually concave ventrally. Its eye and the peduncle of its antenna
seem somewhat smaller and shorter and its stylocerite longer. The dis-
tolateral projections of its petasma are curved much more to median than
is usual in E. picta. Finally, whereas in a male of E. picta of carapace
11 mm. (B. O. C. No. 81) the petasma is 2.9 mm. long by 1.3 mm. broad,
and in the present male of E. picta of carapace 7.5 mm. the petasma
measures 1.3 x 0.7 mm., in the available male of E. stimpsoni, of carapace
7.0 mm., the petasma measures 2.5 x 1.2 mm., a size relatively much greater
than in E. picta,.
Eusicyonia disdorsalis Burkenroad.
E. disdorsalis, Burkenroad, 1934b, p. 96.
Range: Gulf of Panama, southern tip of Lower California, 3 to 24
fathoms.
6 “Beach” record based on specimen taken in the seine by the Pawnee, at a depth presumably
not greater than 5 fathoms.
88
Zoologica: Neiv York Zoological Society
[XXIII :3
Material: A total of 10 specimens (of which about half are males)
was taken at San Lucas Bay (Station 135) in 3 to 20 fathoms, as follows:
Station 135: D-ll, D-12 (2 S, 4 2), D-18, D-19 (2 $), D-20 (2 2). Cat.
Nos. 361,017, 361,018, 361,019.
Dimensions and Sexual Condition : All of the specimens are of very
small size, the females ranging in carapace length from 6 mm. to 8 mm.,
the males from 5 mm. to 7 mm. The petasma is, however, of adult form
and condition in all of the males, which are presumably adult.
Remarks: No distinction between the present material and that pre-
viously known, from the Gulf of Panama, is apparent.
Eusieyonia ingentis, sp. nov.
(Text-figures 31-34).
Type: Type and Cotypes, Cat. No. 361,025, Department of Tropical
Research, New York Zoological Society (6 $, 4 2). Station 127, Dredge 1,
off east coast of Cedros Island, west coast of Lower California, 28° 05'
N. Lat., 115° 09' W. Long. 4-foot dredge, 38 fathoms, March 27, 1936.
Range: Known only from the present records, from east of Cedros
Island on the Pacific coast of Lower California, in 38 to 60 fathoms.
Material: A total of 63 specimens (of which number about three-fifths
are males) was taken east of Cedros Island (Stations 125, 126 and 127) in
depths of 38 to 60 fathoms, as follows: Station 125: D-l (7 $, 1 2);
Station 126: D-2 (4 $, 6 2), D-3 (21 $, 12 2), D-4 (1 5), D-10 (1 2) ; Station
127: D-l (6 $, 4 2). Cat. Nos. 361,020, 361,021, 361,022, 361,023, 361,024,
361,025.
Dimensions and Sexual Condition: Females ranging in length from
carapace 27 mm., total length 105 mm., to carapace 7 mm., total 30 mm.
Males ranging in length from carapace 22 mm., total 86 mm. to carapace
5 mm., total 20 mm. There is no clear evidence of local segregation as to
size or sex. The petasmal endopods are of adult form and are joined to-
gether in males down to 10.5 mm. in carapace length.
Diagnosis : Antennal angle with a buttressed tooth ; dorsal carina of the
second pleonic somite not incised; basis and ischium of the first chelipeds
unarmed. The carapace is armed with two middorsal teeth, one before and
one behind the level of the hepatic spine; the rostrum is armed above with
three teeth and has a bifurcate tip.
The rostrum is more than one-third as long as the carapace and is
distally quite slender; it is proximally decurved, but the tip is often turned
up somewhat at an angle to the proximal part. The lateral ridge of the
rostrum parallels the ventral margin of the blade throughout its length.
The ridge (cardiacobranchial) extending posteriorly on the sides of the
carapace from behind the hepatic spine, is conspicuous, especially in large
individuals.
The telson is long and slender, the width of its base being no more
than a quarter of its length; and is armed with a conspicuous pair of
fixed lateral spines. The middorsal carina of the fifth pleonic somite de-
scends very gently at its posterior end. The anteromedian pleural sulcus
of the first pleonic somite is faintly continued below its well-defined dorsal
section as a very shallow depression. The pleural margin of the first
pleonic somite, anterodorsal to the ventral angle, is straight.
The dorsal surface of the distal part of the ocular peduncle bears, near
its distolateral margin, a pencil of hairs which when unbroken reaches far
beyond the eye in large adults, although it is sometimes much shorter in
small specimens. The terminal segment of the antennal peduncle is very
1938 J
BurJcenroad: Penaeidae
89
long and slender, its diameter being no more than one-fifth its length. The
telson does not extend as far as do the uropods in adults, although it may
exceed them in juveniles.
The blunt-tipped distoventral projections of the petasma bear a shal-
low proximolateral prominence, distal to which the projections extend paral-
lel to and nearly or quite as far as do the distolateral lobes. The distolateral
lobes are divergent from the median plane.
The depressed median interspace between the raised posterolateral
parts of the buttress of the median spine of the thirteenth sternite in the
female is little if at all wider than are the raised parts. The spine between
the first pleopods of the female is broadly subtriangular in ventral outline,
with convex lateral margins.
Remarks : The length attained by E. ingentis is somewhat greater than
seems heretofore to have been recorded for the subfamily, although the
animal, being a relatively slender one, is outbulked by the corpulent species
next it in length, E. brevirostris (Stimpson).
Eusicyonia ingentis is closely related to E. dorsalis (Kingsley) and its
Pacific congener E. disdorsalis Burkenroad, from which it is most strikingly
distinguished by the form of its petasma. Aside from being a much larger
species, its habitus is relatively more elongate and slender than in the
related forms, a tendency particularly pronounced as regards rostrum, telson,
and antennal peduncle. The details in which E. ingentis differs most mark-
edly from E. disdorsalis are those listed in the paragraphs of the diagnosis
subsequent to the first ; thus in E. disdorsalis the lateral ridge of the rostrum
usually arches up from the ventral margin near the distal end; the broad
ridge posteriorly continuing the hepatic buttress is ill-defined; the lateral
teeth of the telson are very minute and inconspicuous, especially in large
Text-figures 31-34.
Eusicyonia ingentis, n. sp. 31. Carapace (lateral view) ; juvenile 2, D. T. R.
361,021, x 4.5. 32. Telson and right uropod (dorsal view) ; adult $ of
total length 64 mm., D. T. R. 361,021, x 4.5. 33. Fifth pleonic somite
(lateral view) ; as in 32, x 4.5. 34. Petasma (posterior view) ; as in
32, x 10.
90
Zoologica: New York Zoological Society
[XXIII :3
specimens ; the posterior end of the dorsal carina of the fifth pleonic somite
is in adults sharply angular, although in juveniles it may descend gently;
the anteromedian pleural sulcus of the first pleonic somite is not continued
below its well-cut dorsal section; the anteroventral pleural margin of the
first pleonic somite is concave in adults, although it is straight in juveniles;
the pencil of setae on the dorsal surface of the ocular peduncle extends no
more than half way across the cornea; the telson exceeds the uropods; the
distolateral lobes of the petasma curve toward the midline; the proximo-
lateral ramus of the distoventral projection of the petasma exceeds the acute,
laterally directed distal part in size; the depressed area of the base of the
spine of sternite XIII of the female is much wider than the raised parts
flanking it; and the tooth between the first pleopods in females is narrow
and usually with concave lateral edges.
At dimensions at which E. ingentis is juvenile, with minute undevel-
oped petasmal endopods, E. disdorsalis presents all the characters of
maturity.
Literature Cited.
Balss, H.:
1927. Macrura der Deutschen Tiefsee-Expedition. 2. Natantia, Teil A. ILiss.
Ergeb. D. T.-E., XXIII, 6.
Boone, P. L.:
1930. Crustacea: Anomura, Macrura . . . Bull. Vanderbilt Marine Mus.,
III. Huntington, N. Y.
1931. Anomuran, Macruran Crustacea from Panama and Canal Zone. Bull.
American Mus. Nat. Hist., LXIII, 2.
Bouvier. E. L. :
1906. Sur les Gennadas ou Peneides bathypelagique. Bull. Mus. Oc. Monaco,
LXXX.
Burkenroad, M. D.:
1934a. The Penaeidea of Louisiana. Bull. American Mus. Nat. Hist., LXVIII,
2.
1934b. Littoral Penaeidea chiefly from the Bingham Oceanographic Collection.
Bull. B. O. C. IV, 7.
1936. The Aristaeinae, Solenocerinae and Pelagic Penaeinae of the Bingham
Oceanographic Collection. Bull. B. O. C., V, 2.
Faxon, W.:
1893. Reports on the Dredging Operations ... by the U. S. F. C. S. “Alba-
tross.” VI. Preliminary Descriptions of New Species of Crustacea. Bull.
Mus. Comp. Zool. Harvard, XXIV, 7.
1895. Reports on an Exploration off the west coasts of Mexico, Central and
South America, and off the Galapagos Islands . . . XV. The Stalk-
Eyed Crustacea. Mem. Mus. Comp. Zool. Harvard, XVIII.
Holmes, S. J.:
1895. Notes on West American Crustacea. Proc. Cal. Acad. Sci., (2) IV.
1900. Synopses of California stalk-eyed Crustacea. Occ. Pap. Cal. Acad.
Sci., VII.
Kemp, S. W. :
1910. Notes on Decapoda in the Indian Museum. I. The Species of Gennadas.
Rec. Indian Mus., V, 17.
Kingsley, J. S. :
1878. Notes on North American Caridea in the Museum of the Peabody Aca-
demy of Science. Proc. Acad. Nat. Sci. Philadelphia, XXX.
1879. On a collection of Crustacea from Virginia, North Carolina and
Florida. Proc. Acad. Nat. Sci. Philadelphia, XXXI.
1938 J
Burkenroad: Penaeidae
91
Lockington, W. N.:
1879. Notes on Pacific Coast Crustacea. Bull. Essex Inst., X (1878).
Man, J. G. De:
1911. Decapoda of the Siboga Expedition. I. Penaeidae. Siboga-Expeditie,
XXXIXa [and plates, 1913, suppl. XXXIXa].
Milne Edwards, A. & Bouvier, E. L.:
1909. Reports on the Results of Dredging ... by the U. S. C. S. S. “Blake”
. . . XLIV. Peneides et Stenopides. Mem. Mus. Comp. Zool. Harvard,
XXVII, 3.
Pesta, 0.:
1915. Die Penaeidea des Wiener Naturhistorischen Hofmuseums. Arch. f.
Naturg., LXXXI, A, 1.
Rathbun, M. J.:
1902. Papers from the Hopkins Stanford Galapagos Expedition . . . VIII.
Brachyura and Macrura. Proc. Washington Acad. Sci., IV.
1904. Decapod Crustaceans of the Northwest Coast of North America. Har-
riman Alaska Exped., X.
Schmitt, W. L. :
1921. Marine decapod Crustacea of California. Univ. California Publ. Zool.,
XXIII.
1924. The Macrura and Anomura collected by the Williams Galapagos Ex-
pedition, 1923. Zoologica (Sci. Contr. New York Zool. Soc.), V, 15.
1935. The West American Species of Shrimps of the Genus Penaeus. Proc.
Biol. Soc. Washington, XLVIII.
Smith & Coates: Fibro- epithelial Growths in Turtles
93
4.
Fibro-epithelial Growths of the Skin in Large Marine Turtles,
Chelonia my das (Linnaeus).
G. M. Smith
Department of Anatomy, Yale School of Medicine,
and Neiv York Aquarium
&
C. W. Coates
New York Aquarium
(Plates I-XVIII).
Diseases of turtles have received but relatively little study up to the
present time and not much is known about the benign or malignant tumors
of these reptiles. In a report on the causes of deaths in the London Zoo-
logical Gardens, Plimmer (1912, 1913) reported two cases of such growths.
Scott and Beattie (1927), referring to Plimmer’s observation, state that
one tumor was a glandular cancer of the stomach in an elephantine tortoise
and that the second was a multi-nucleated tumor involving heart muscle
of Sternothaerus niger, a fresh water turtle from the Cameroon area of
West Africa. These authors point out that of six thousand reptiles of
various kinds autopsied at the London Zoological Gardens, only two showed
tumors as the cause of death. Pick and Poll (1903) described an adenoma-
tous growth of the thyroid of the Brazilian fresh water turtle, Platemys
geoffroyana ( Hydraspis geoffroyana) . This was a large mass measuring
6 x 4 x 2.5 cm., consisting of narrow cylindrical cells in small acinar ar-
rangement with lumina containing granular material.
In December, 1936, it came to our attention that one of the large marine
turtles, Chelonia mydas (Linnaeus), of the New York Aquarium presented
multiple warty growths of the skin. This turtle, shipped from Key West,
Florida, two years previously, occupied a large harbor water tank with two
others of the same species and with two large loggerhead turtles, Thalas-
sochelys caretta (Linnaeus). None of the other four turtles exhibited
papillomatous lesions, although an exposure to possible infection had existed
for more than a year. The water supply of the tank is a continuous flow of
harbor water, which is brackish and changes density with the change of tide.
The turtle showing multiple warts weighed approximately one hundred
and fifty pounds. Papillomata were distributed in the region over the
dorsal part of the neck in both non-pigmented and pigmented areas and
in the light gray or whitish areas of both axillae and both groins (Figs. 1
and 2). There were in addition several small warty excrescences in the
outer regions of the upper eyelids involving the conjunctivae. Tissues were
94
Zoologica: New York Zoological Society
[XXIII :4
removed from the neck for microscopic study and a month later more
tissues were excised from several sites for transplantation experiments.
Recently (December, 1937) it was determined that cutaneous fibro-
epithelial growths occur in Chelonia mydas (Linnaeus) under normal con-
ditions of wild life. Three instances of this disease were found in these
large marine turtles (weighing from sixty to two hundred pounds) in the
course of an examination of some two hundred turtles of this species
caught in the waters south of Key West. The distribution of the growths
in the wild turtles likewise occupied the dorsal aspect of the neck, the
axillae, groins and eyelids. The histological details of the lesions in the
wild turtle resembled in general those of the original turtle studied in
the New York Aquarium and it was thought probable that the disease in
wild life and that seen in the Aquarium represented the same condition.
For purposes of pathological description, the cutaneous growths of
the large marine turtle now under consideration may be referred to as
fibro-epithelial in character. There appeared to be considerable variety
in the size, form and arrangement of the masses of tissue (Figs. 3 to 8).
Outstanding were the papillary, coral or cauliflower-like growths (Figs. 5
to 8). Figure 7 represents a growth arising from non-pigmented skin, while
the tumor in Fig. 8 is derived from a pigmented area. Less frequently
smooth oval or round fibromata projected above the level of the skin (Figs.
12 and 13). Both papillomata and fibromata, caused perhaps by the same
agent, appeared as single or confluent masses in the same turtle. Papil-
lomata and fibromata at times grew in immediate apposition. All growths
had a firm gritty consistency and areas of calcification at the surface of
tumors were not uncommon. In the preparation of histological sections, it
was found advisable to subject all tissues to decalcification. Papillomata
varied in size from 3 mm. to 3 cm. while solitary fibromata measured
as much as 2.5 cm. in diameter.
The internal structure of papillary growths became clear on sectioning
the gross specimens and inspecting their cut surfaces (Figs. 9 to 11). At-
tached to the skin by a sessile or pedunculated base, the stroma of the
various growths was composed of dense intermingling fibrous tissue which
spread out toward the surface in many lesser ramifications to form rounded,
bulbous or pointed terminal tips covered with thickened surface epithelium.
The microscopic structure of normal skin of Chelonia mydas is shown
in Fig. 17. Compared with this the epithelium covering papillomata appeared
to be somewhat thickened and covered with an irregular mass of keratin
(Fig. 21). The size and shapes of epithelial cells varied greatly and inter-
cellular bridges of prickle cells were strongly developed (Fig. 23). Epithelial
columns and cords often dipped down deeply into the supporting stroma
(Fig. 22). So-called epithelial “pearls” were frequently encountered, not
unlike those found in mammalian epithelial tumors (Fig. 24). The vascular
supply of papillomata is a network of small blood vessels. It was not unusual
to have collections of lymphocytes grouped around small blood vessels, indi-
cating possibly a low grade of inflammation (Fig. 27). Cell inclusions were
not demonstrated in the sections examined. Dendritic pigment cells fre-
quently accompanied blood vessels (Fig. 25), or they existed in diffuse dis-
tribution in the stroma below the surface epithelium (Fig. 26). In highly
pigmented areas, numerous melanophores were found at the surface between
epithelial cells.
The fibromata of the turtle’s skin (Figs. 12 and 13) were relatively
simple in microscopic architecture, being composed of intermingling bands
of fibrous tissue varying in density (Figs. 28 and 29), covered on the
surface with but slightly thickened epithelium (Fig. 28). Although the
skin covering fibromata could appear non-pigmented so that tumors looked
white on external examination, the interiors were usually gray or a streaky
black due to the presence of pigment cells which followed the course of blood
1938]
Smith & Coates: Fibro-epithelial Growths in Turtles
95
vessels (Fig. 28). Small nerve trunks were found in one of the sections
of a fibroma (Fig. 30).
It was noted that papilloma and fibroma existed at times in a combined
form of tumor. This was illustrated very well in a large growth measuring
8x6x5 cm., taken from the right side of the neck of one of the Key West
turtles (Fig. 15). The large bosses (F) of this growth were fibromata,
while at point P a distinct papilloma existed, fused with the rest of the
mass. Figure 16 shows the interior of this tumor to be composed of five
smaller fibromata, white or partly pigmented. Nearly all fibromata showed
microscopic evidences of peri-vascular melanophores (Fig. 28).
These fibro-epithelial tumors of the turtle show no malignant changes.
Mitotic figures are few in number.
Comment.
The stimulus to the formation of cutaneous fibro-epithelial growths in
Chelonia mydas remains for the present undetermined. It is interesting to
note that fishermen of the white population at Key West, exposed for many
years to the sun, frequently develop keratoses and malignant changes of the
skin, whereas it is said that with fishermen of the colored population
such changes are rare occurrences. It is conceivable that these great sea
turtles, even though in an aquatic environment, may respond to exposure to
the sun prolonged over many years by papillomatous changes affecting the
skin. A different explanation for the etiology of the disease other than
the above mentioned is more likely, such as infection by parasites or by an
unknown virus. No parasites were noted so far in the tumor tissues of
the four turtles examined. It is known that certain infectious warts of man,
dog and cattle are of virus origin. Infectious papillomatosis of rabbits has
been shown by Shope (1933) and Rous and Beard (1934) to have its origin
in a virus. Such an etiology for the cutaneous turtle fibro-papillomata is not
unlikely. The study of the transmissibility of the turtle papilloma has been
begun and it is hoped to report on this at a later date.
Summary.
Cutaneous fibro-epithelial growths in four large marine sea turtles,
Chelonia mydas (Linnaeus), are described, with a discussion of their micro-
scopic structure.
Bibliography.
Pick, L. and Poll, H.
1903. Ueber einige bemerkenswerthe Tumorbildungen aus der Thier-
pathologie, insbesondere fiber gutartige und krebsige Neubildungen
bei Kaltblfitern. Berlin Klin. Woch. 24:547.
Plimmer, H. G.
1912. Report on the deaths which occurred in the Zoological Gardens dur-
ing 1911. Proceedings of the Zoological Society of London. 1:235.
1913. Report on the deaths which occurred in the Zoological Gardens dur-
ing 1912, together with the blood-parasites found during the year.
Proceedings of the Zoological Society of London. 1:141.
Rous, P. and Beard, J. W.
1934. A virus induced mammalian growth with characters of a tumor (the
Shope rabbit papilloma). Jonrn. of Exper. Med. 60:701.
96
Zoologica: Neiv York Zoological Society
[XXIII :4
Scott, H. H. and Beattie, J.
1927. Neoplasm in a Porose Crocodile. Journal of Pathology and Bac-
teriology. 30:61.
Shope, R. E.
1933. Infectious papillomatosis of rabbits. Journ. of Exper. Med. 58:607.
1938]
Figs. 1 i
Fig. 3.
Fig. 4.
Figs. 5 t
Fig. 7.
Fig. 8.
Fig. 9.
Fig. 10.
Fig. 11.
Fig. 12.
Fig. 13.
Fig. 14.
Fig. 15.
Fig. 16.
Fig. 17.
Fig. 18.
Smith & Coates: Fibro-epithelial Growths in Turtles 97
EXPLANATION OF THE PLATES.
Plate I.
& 2. Marine turtle, Chelonia mydas, weighing about 150 pounds, show-
ing distribution of papillomata on neck, axillae and eyelid.
Plate II.
Small papillomata. A pigmented growth at x.
Papillomata of various shapes. E is tissue removed from upper eyelid
with papilloma growing on the margin of eyelid.
Plate III.
& 6. Larger forms of papillomata.,
Plate IV.
One of the larger forms of papillomata. Shows papilloma arising from
non-pigmented skin.
Plate V.
One of the larger forms of papillomata. Shows papilloma derived from
pigmented skin.
Plate VI.
Cut surfaces of gross specimens of papillomata.
Plate VII.
Cut surface of gross specimen of papilloma.
Cut surface of gross specimen of papilloma, showing a deeply pigmented
stroma.
Plate VIII.
Solitary fibroma of the skin. F marks the fibroma; C marks the site
of a calcified plaque of the skin.
Solitary fibromata of the skin.
Plate IX.
Inner surfaces of fibromata with pigmentation.
Plate X.
Large fibromatous masses, FF, with associated papilloma, P.
Plate XI.
Interior of tumor shown in Fig. 15, exhibiting smaller fibromata; 1,
2, 3, 4 and 5.
Plate XII.
Normal skin of Chelonia mydas. x 80.
Papilloma with broad base of fibrous tissue supporting irregularly-
shaped papillary processes tipped with thickened epithelium, x 40.
98
Zoologica: New York Zoological Society
Plate XIII.
Figs. 19 & 20. Low power magnification of papillomatous tissue, x 20.
Plate XIV.
Fig. 21. Bulbous tips of a papilloma. Stroma slightly hyalin in character, x 100.
Fig. 22. Surface epithelium of papilloma illustrating down-growth of epithelium,
x 60.
Plate XV.
Fig. 23. Surface epithelium of papilloma showing prickle cells united by thick
intra-cellular bridges. Masson stain, x 650.
Fig. 24. Deep-lying epithelial cord with epithelial “pearl.” x 250.
Plate XVI.
Fig. 25. Section from a papilloma with melanophores arranged around small
blood vessels, x 125.
Fig. 26. Melanophores in diffuse arrangement lying in stroma close to epi-
thelium. x 200.
Plate XVII.
Fig. 27. A collection of lymphoid cells around blood vessels in a papilloma,
x 350.
Fig. 28. Section taken from solitary pigmented fibroma showing compact fibrous
tissue below skin. Numerous pigment cells accompany blood vessels,
x 50.
Plate XVIII.
Fig. 29. Dense, wavy, hyalin, fibrous tissue taken from the center of a non-
pigmented fibroma.
Fig. 30. In center of photograph a small nerve trunk traversing loosely ar-
ranged fibers of a fibroma, x 275.
SMITH a COATES.
PLATE I.
FIG. 1.
FIG. 2.
FI BRO-EPITH ELI AL GROWTHS OF THE SKIN IN LARGE
MARINE TURTLES, CHELONIA MYDAS (LINNAEUS).
SMITH a COATES.
PLATE II.
ini|im
llll
llll
llli|llll
llll|!lll
llll|ll!l
lll!|llll
llll|IHI
llll|llll
lllllll
4
5
6
7
8
9
1
0 8
8
FIG. 3.
1
llll
llll
llll
llll
llll
|
II
llll
III
III
llll
3
4
5
6
7
8
FIG. 4.
FI BRO- EPITHELIAL GROWTHS OF THE SKIN IN LARGE
MARINE TURTLES, CHELONIA MYDAS (LINNAEUS).
SMITH a COATES.
PLATE III.
FIG. 5.
FI BRO-EPITH ELI AL GROWTHS OF THE SKIN IN LARGE
MARINE TURTLES, CHELONIA MYDAS (LINNAEUS).
SMITH a COATES.
PLATE IV.
FI BRO- EPITHELIAL GROWTHS OF THE SKIN IN LARGE
MARINE TURTLES, CHELONIA MYDAS (LINNAEUS).
SMITH a COATES.
PLATE V.
1 1 1
Ml!
| I 1 1 1
1 1 1 1 1 1 1 1 1
4
1
5
6
FIG. 8.
FI BRO- EPITHELIAL GROWTHS OF THE SKIN IN LARGE
MARINE TURTLES, CHELONIA MYDAS (LINNAEUS).
SMITH & COATES.
PLATE VI.
111!
llll
IHIII!!!
III!
llll
III!
llll
llll
mi
II
METRIC 1
2
3
4
5
FIG. 9.
FI BRO- EPITHELIAL GROWTHS OF THE SKIN IN LARGE
MARINE TURTLES, CHELONIA MYDAS (LINNAEUS).
SMITH a COATES.
PLATE VI!.
FIG. 10.
FIG. 11.
FI BRO-EPITH ELI AL GROWTHS OF THE SKIN IN LARGE
MARINE TURTLES, CHELONIA MYDAS (LINNAEUS).
SMITH & COATES.
PLATE VIII.
TUT
m
m
TTTT
llll
TUT
impr
llll||lll
mpr
TUTJUTT
TUT
llll
2
3
4
5
6
7
3
9
1
FIG. 13.
FI BRO- EPITHELIAL GROWTHS OF THE SKIN IN LARGE
MARINE TURTLES, CHELONIA MYDAS (LINNAEUS).
SMITH & COATES
PLATE IX.
MARINE TURTLES, CHELONIA MYDAS (LINNAEUS).
SMITH a COATES.
PLATE X.
FI BRO-EPITH ELI AL GROWTHS OF THE SKIN IN LARGE
MARINE TURTLES, CHELONIA MYDAS (LINNAEUS).
SMITH & COATES.
PLATE XI.
METRIC
II
6
FIG. 16.
FI BRO- EPITHELIAL GROWTHS OF THE SKIN IN LARGE
MARINE TURTLES, CHELONIA MYDAS (LINNAEUS).
SMITH & COATES.
PLATE XII.
FIG. 17.
FIG. 18.
FI BRO- EPITHELIAL GROWTHS OF THE SKIN IN LARGE
MARINE TURTLES, CHELONIA MYDAS (LINNAEUS).
SMITH a COATES.
PLATE XIII.
FIG. 19.
FIG. 20.
FI BRO- EPITHELIAL GROWTHS OF THE SKIN IN LARGE
MARINE TURTLES, CHELONIA MYDAS (LINNAEUS).
SMITH & COATES.
PLATE XIV.
FIG. 21.
FIG. 22.
FI BRO-EPITH ELI AL GROWTHS OF THE SKIN IN LARGE
MARINE TURTLES, CHELONIA MYDAS (LINNAEUS).
tk
FIG. 24.
SMITH & COATES.
PLATE XV.
’■* *■
.1 ^
\ ♦-V-1
FIG. 23.
FI BRO-EPITH ELI AL GROWTHS OF THE SKIN IN LARGE
MARINE TURTLES, CHELONIA MYDAS (LINNAEUS).
SMITH a COATES.
PLATE XVI.
FIG. 26.
FI BRO- EPITHELIAL GROWTHS OF THE SKIN IN LARGE
MARINE TURTLES, CHELONIA MYDAS (LINNAEUS).
V/ ■>
SMITH a COATES.
PLATE XVII.
FIG. 27.
FI BRO- EPITHELIAL GROWTHS OF THE SKIN IN LARGE
MARINE TURTLES, CHELONIA MYDAS (LINNAEUS).
SMITH & COATES.
PLATE XVIII.
FIG. 29.
FIG. 30.
FI BRO- EPITHELIAL GROWTHS OF THE SKIN IN LARGE
MARINE TURTLES, CHELONIA MYDAS (LINNAEUS).
Jleto ^orfe 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 }3arb
W. Reid Blair, Director
Raymond L. Ditmars, Curator of Mammals and Reptiles
Claude 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
3®tpartmont of tCropical &eseardj
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
Lee S. Crandall, Curator of Birds
Charles R. Schroeder, Veterinarian
W. Reid Blair
William Beebe
Charles M. Breder, Jr.
George Bird Grinnell
William Bridges
■ ■ •
V,- • :
ZOOLOGICA
J9J93Q
SCIENTIFIC CONTRIBUTIONS
OF THE
NEW YORK ZOOLOGICAL SOCIETY
VOLUME XXIII
Part 2
Numbers 5-9
PUBLISHED BY THE SOCIETY
THE ZOOLOGICAL PARK, NEW YORK
July 14, 1938
CONTENTS
PAGE
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 lintoni Gurley. By Ross F.
Nigrelli & G. M. Smith. (Plates I-VII) 195
8. Electrical Characteristics of the Electric Tissue of the Elec-
tric Eel, Electro phorus electricus (Linnaeus). By R. T.
Cox & C. W. Coates. (Plate I; Text-figures 1 8c 2) 203
9. Colonial Rattlesnake Lore, 1714. By James R. Masterson 213
Notes
217
Bigelow: Medusae Taken During 1929 and 1930
99
5.
Plankton of the Bermuda Oceanographic Expeditions. VIII.
Medusae Taken During the Years 1929 and 19301.
Henry B. Bigelow
Museum of Comparative Zoology, Harvard College.
(Text-figs. 1-23).
[This is one of a number of papers dealing with the planktonic contents
of nets drawn at various levels off the coast of Bermuda on the Bermuda
Oceanographic Expeditions of the New York Zoological Society under the
direction of Dr. William Beebe. Full details of the nets, locality, etc., will
be found in Zoologica, Volume XIII, Numbers 1, 2 and 3, and Volume XXI,
Numbers 3 and 4.]
Contents. Page
Introduction 101
Anthomedusae 102
Zancleopsis 102
Zancleopsis dichotoma Mayer 102
Bougainvillia .104
Bougainvillia niobe Mayer 104
Pandea 106
Pandea conica Quoy and Gaimard 106
Pandea rubra Bigelow 107
Heterotiara 108
Heterotiara anonyma Maas 108
Calycopsis 108
Calycopsis sp ? 108
Leptomedusae 109
Chromatonema 109
Chromatonema rubrum Fewkes 109
Aequorea 110
Aequorea floridana L. Agassiz Ill
Trachomedusae 112
Olindias 112
Olindias phosphorica tenuis Fewkes 113
1 Contribution No. 554, Department of Tropical Research. New York Zoological Society.
Contribution No. 147 of the Woods Hole Oceanographic Institution.
Contribution from the Bermuda Biological Station for Research, Inc.
A
100 Zoologica: New York Zoological Society [XXIII :5
Page
Rhopalonema 113
Rhopalonema velatum Gegenbaur 113
Pantachogon 114
Pantachogon haeckeli Maas 115
Colobonema 116
Colobonema typicum Maas 117
Crossota 118
? Crossota brunnea Vanhoffen 119
Trachynemidae genus ? 120
Halicreas 120
Halicreas minimum Fewkes 122
? Halicreas glabrum Vanhoffen 124
Halitrephes 125
Halitrephes valdiviae Vanhoffen 125
Liriope 126
Liriope tetraphylla Chamisso and Eysenhardt 127
Geryonia 127
Geryonia proboscidalis Forskal 127
Narcomedusae 128
Cunina ,, 128
Cunina sp? 128
Solmissus 129
? Solmissus incisa Fewkes 129
Aegina 130
Aegina citrea Eschscholtz 131
Aeginura 132
Aeginura giimaldii Maas 132
Pegantha 134
? Pegantha clara R. P. Bigelow 134
SCYPHOMEDUSAE 135
Carybdeida 135
Carybdea 136
Carybdea marsupialis xaymacana Conant 139
Carybdea alata Reynaud 144
Tamoya 150
Tamoya haplonema Muller 151
Periphylla 155
Periphylla hyacinthina Steenstrup 155
Nausithoe 159
Nausithoe punctata Kolliker 159
Atolla 160
Atolla wyvillei Haeckel 160
Linuche 163
Linuche unguiculata Schwartz 163
1938]
Bigeloiv: Medusae Taken During 1929 and 1930
101
Page
Discophora 164
Pelagia 164
Pelagia noctiluca Forskal 164
Poralia 165
? Poralia rufescens Vanhoffen 165
Aurellia 167
Aurellia aurita Linne 168
Faunistic Considerations 169
Vertical Distribution of Medusae as a Whole 170
Relative Abundance of Different Species 172
Absolute Abundance 172
Quantitative Comparison with Other Regions 172
Bibliography 174
Appendix 180
Introduction.
The medusae recorded in the following pages were taken by Dr. William
Beebe within a roughly circular area, some eight miles in diameter, cen-
tering about 9 miles about south-southeast from the eastern end of the Ber-
muda archipelago. Within the area of study, the depth ranges from 1,357 to
somewhat less than 1,500 fathoms.
Horizontal towing with open nets was carried out in each year at fre-
quent intervals from April through September, at various levels from the
surface down to 2,195-0 meters, the total number of tows being 528 for 1929,
488 for 1930. An account of methods of collection, with list of tows, has
been given by Beebe (1931). In this list, the depths stated are those at
which the horizontal parts of the tows were made, determined — at least in
some cases — by a recording pressure-gauge or “Bathygraph” (Beebe, 1931,
p. 13, Fig. 6). But one must remember that as the nets used were of the
ordinary open type, which fish nearly as effectively while being lowered as
while being towed horizontally, and as effectively while being hauled up
again, there is no way of determining from what depth any particular speci-
men may have come. The depths of the hauls, given under the several
species in the following pages, are therefore stated accordingly. However, so
many hauls were made on each towing day, and at so many different levels,
that differential results show fairly well the upper limits of occurrence; also
the chief levels of abundance for the more common species. But none of the
species was taken in numbers large enough to allow calculation of the prob-
able degree to which the catches were contaminated by specimens picked up
during the vertical parts of the tows (for further discussion of this subject
see Bigelow and Sears, 1937, p. 69).
The collection contains no new species. But it adds to morphological
knowledge of several. And it is especially interesting from the distribu-
tional standpoint, being the only intensive and long-continued campaign of
deep towing that has yet been made at one place in the open ocean. With
so many hauls, made at such close intervals, it is unlikely that any medusa
was missed that was at all common, or that occurred other than sporadically.
It is therefore safe to assume that the catches listed below give a repre-
sentative picture of all the species of medusae that occurred with any regu-
larity between the surface and the bottom of the sea, at the chosen station,
during the summers of 1929 and 1930, except Gyanea, which as Dr. Beebe
informs me, was occasionally seen at the surface.
102
[XXIII :5
Zoologica: New York Zoological Society
ANTHOMEDUSAE.
Cladonemidae.
Zancleopsis.
This genus was proposed by Hartlaub (1907, p. 116) for a bitentaculate
medusa from the Tortugas, described by Mayer (1900) as Gemmaria dicho-
toma, but which differs from typical Gemmaria (the latter is probably a
synonym of Zanclea) in the fact that its tentacles are branched, with the
branches as well as the main trunk terminating in swollen bulbs; structures
very different from the stalked nematocyst knobs with which the tentacles
of Gemmaria (or Zanclea) are armed. Zancleopsis also possesses ocelli,
which are lacking in Gemmaria. And the new genus was accepted by Mayer
(1910) in his subsequent monograph.
Zancleopsis was not reported again until 1928 when Kramp (1928,
p. 40) described a second species — tentaculata — from a single specimen from
the Pacific. The Bermuda series now yields a third record of the genus.
Mayer’s specimens from the Tortugas had only two developed tentacles.
But it appears that they were juveniles, for while he stated that the gonads
are developed on the inter-radial sides of the manubrium, his illustrations
give no indication that the sex tissue had actually appeared, which accords
with the small size of the specimens (3 mm. high), contrasted with the
Bermudian example (20 mm. high). Therefore, it is entirely possible that
the bulbs alternating with the formed tentacles of Mayer’s specimens would
later have developed filaments, thus raising the tentacle number to four.
And this is made the more likely by the fact that Kramp’s Pacific specimen
(4.5 mm. high) showed just this state, i. e., one pair of large branched
tentacles and one pair of much smaller simple tentacles. It also showed a
more advanced stage in the sexual development, with the gonads forming
“a number of vertical folds, four on each of the inter radial sides of the
stomach” (Kramp, 1928, p. 41). The Bermuda example illustrates a stage
still more advanced, the sex folds being more numerous (p. 104), the second-
ary tentacles about as long as the primary, though still continuing simple,
and the size much greater. In short, it seems altogether probable that we
have to do here with the final stage in growth of Mayer’s juvenile Z. dicho-
toma, a conclusion favored by the fact that there is no faunistic discon-
tinuity, so far as medusae are concerned, between Bermuda and southern
Florida.
But it appears that the Pacific Z. tentaculata is distinct, for not only
does it have large gonads at a much smaller size, but the tentacular spurs,
bearing the ocelli, are not so highly developed, and the edges of its radial
canals are pictured by Kramp as smooth, whereas they are jagged in the
Bermuda specimen.
Zancleopsis dichotoma Mayer.
Gemmaria dichotoma, Mayer, 1900, p. 35, PI. 17, Fig. 40.
Zancleopsis dichotoma, Hartlaub, 1907, p. 115; Kramp, 1928, p. 42.
Material : Net No. 922, 1,097-0 meters, September 20, 1930, 1 specimen,
20 mm. high by 14 mm. broad, in good condition.
As the adult has not been seen previously, an account of the significant
features follows :
The bell is high domed, moderately thick walled, especially apically
(Text-fig. 1), resembling Kramp’s (1928, Fig. 14) picture of the Pacific
tentaculata in general form, except that the apex is rounded instead of
pointed. And this minor difference may be merely the result of the preserva-
tive. The exumbrella is smooth as Hartlaub has also noted, without any
1938]
Bigelow: Medusae Taken During 1929 and 1930
103
Text-fig. 1.
Zancleopsis dichotoma. Bermudian specimen,
20 mm. high.
Text-fig. 2.
Zancleopsis dichotoma. Tentacular bulb. A., Abaxial
view; B., Lateral view, x about 12.
104
Zoologica: Neiv York Zoological Society
[XXIII :5
trace of the nematocyst-ribs so conspicuous in Zanclea. The subumbrella
occupies about % the heighth of the bell, with the bell-cavity projecting,
pouch-like, apically, in the interradii, in form easier pictured (Text-fig. 1)
than described ; again much as Kramp figured it for the Pacific species. The
four radial canals are of moderate breadth, their margins distinctly jagged
(smooth in tentaculata) . The margins of the circular canal are smooth.
The four subspherical tentacular bulbs, about equal in size, are prom-
inent axially. Just distal to its bulb, the tentacle base is also expanded,
abaxially, as a spur (Text-fig. 2) lying in a furrow of the exumbrella, and
with a curious cap-like mass of large, clear cells (apparently ectoderm) on
its outer (abaxial) side. On its outer angle the spur bears a prominent,
long oval ocellus (Text-fig. 2A). The four tentacular filaments are all of
about the same length — in the preserved state about % as long as the bell
is high. One pair are simple, the other pair branched. Of the latter, one
bears a branch out near the tip, two sessile knobs about mid-way the
length of the filament (distal to which the latter narrows abruptly), be-
sides a swelling apparently the forerunner of an additional branch. The
other tentacle of this pair bears two small branches, but is so strongly con-
tracted that it is impossible to tell how extensible the latter may have been.
The two branched primary tentacles terminate in large knobs. The two
secondary tentacles do not, but their tips are in poor condition.
The area of connection between the base of the manubrium and the
subumbrella is cruciform, much as described by Kramp for tentaculata; its
upper % is thrown into a very characteristic series of vertical folds. These
are no doubt the sites of the sexual tissue, as interpreted by Kramp, but eggs
not being visible (i.e., the specimen apparently being a male), this point
could not be determined positively without microscopic sectioning. In the
Pacific specimen of tentaculata, there were 3-4 such folds in each interradial
sector of the manubrium (Kramp, 1928, p. 41, Fig. 15). In the present
specimen the number is greater, 4-6 per interradius, precise enumeration
depending on what does, or does not, deserve the name “fold.” Below the
level of attachment of the radial canals the folds apparently form a continu-
ous series surrounding the manubrium, but as Kramp points out, sectioning
would be required to show to what extent the masses of sex tissue are con-
fluent. In the preserved state, the distal tube-like portion of the manubrium
hangs to about the mid-level of the bell cavity. But as it is much wrinkled,
it is probably contracted, hence may have been considerably longer in life.
After preservation with formalin, manubrium and tentacular bulbs are
of an opaque, dull brownish hue; the pigment ring of each ocellus, dull red.
A comparison of Text-fig. 1 with Kramp’s (1928, Figs. 14-18) illustra-
tions of Z. tentaculata will sufficiently emphasize the close resemblance be-
tween the two, especially as regards the structure of the tentacles. The
chief differences — probably sufficient to separate two species — are summar-
ized above (p. 102). Minor differences in the outline of the bell and the
shape of the manubrium are probably to be credited to the preservative.
Bougainvilliidae.
Bougainvillia.
Bougainvillia niobe Mayer.
Bougainvillia niobe, Mayer, 1894, p. 236, PI. 1, Fig. 2; 1900, p. 42;
1904, p. 11, PI. 2, Figs. 14-15c; 1910, p. 166, PI. 18, Figs. 1-3;
Bigelow, 1918, p. 368.
Material : Surface, July 11, 1929, 1 specimen, 9 mm. diameter, with
medusa buds.
1938] Bigelow: Medusae Taken During 1929 and 1930 105
Net No. 562, 914-0 meters, May 10, 1930, 1 specimen, 9 mm. diameter,
with medusa buds.
Net No. 738, 1,463-0 meters, June 28, 1930, 1 specimen, 9 mm. diameter,
with gonads.
Net No. 812, 1,463-0 meters, August 28, 1930, 1 specimen, 10 mm. diam-
eter, with gonads.
Net No. 862, 1,463-0 meters, September 8, 1930, 1 specimen, 10 mm.
diameter, with gonads.
Net No. 915, 549-0 meters, September 19, 1930, 1 specimen, 6 mm. diam-
eter, with medusa buds.
Net No. 935, 1,463-0 meters, September 23, 1930, 1 specimen, 10 mm.
diameter, with gonads.
These specimens confirm conclusions earlier reached from the study
of the more extensive Bache series (Bigelow, 1918, p. 368) as to the rela-
tionship of B. niobe to other species of the genus, for they fall well within
the recorded range of niobe in number of marginal tentacles and in branch-
ing of the oral tentacles. As in previous collections, the smaller specimens,
up to 9 mm. in diameter, bear the medusa buds well described by Mayer
(1910), whereas the gonads have so far been found only in specimens of
8 mm. in diameter or larger. In the specimens listed above, as in those of
the Bache collection, the sex-masses are adradial, in which niobe resembles
fulva and britannica. It was not possible to determine on the Bache speci-
mens whether or not the masses of sexual tissue were discontinous in the
interradii, at the base of the manubrium. This, however, seems definitely
to be the case in the specimen from Net No. 862; apparently so in the
specimen from Net No. 812 (see list above). Hence, while the other two
specimens with gonads are too much contracted to show this feature, it
can now be added to the specific diagnosis.
The medusa buds in the smaller specimens add nothing to Mayer’s
(1910) account, except for confirmation.
I may note also that one specimen (Net No. 562) is infested with the
parasitic larval stages of a Narcomedusa, probably Cunoctantha.
The large specimens are so much contracted that exact counting of
the marginal tentacles is difficult. They appear to range from 13-14 up
to 16; in one bundle possibly 18, the latter a somewhat greater number
than reported previously. For this same reason, the number of branchings
of the oral tentacles can be stated only tentatively; the maximum number
of branchings appears to be not less than 8 or 9.
On the basis of cumulative evidence the following diagnosis may now
be given for this species: Marginal tentacles, usual maximum 12-16 per
bundle at sexual maturity, perhaps up to 18-19; gonads adradial, discon-
tinuous in the interradii and entirely confined to the manubrium proper;
manubrium sessile, i.e., there is no gelatinous peduncle; ocelli are present;
oral tentacles branch a maximum of 7-9 times. Medium sized individuals
may produce medusa-buds. It is not known whether the same individuals
produce first buds, then gonads in succession, or whether some individuals
exhibit the asexual reproduction only, others the sexual.
The closest ally of niobe is the wide-ranging Pacific species, fulva; so
close in fact, that no constant difference has been found, to separate them,
other than the common occurrence of budding in the former, its absence, so
far as known, in the latter.
General Distribution: Records for B. niobe are so far confined to the
waters between southern Florida, Cuba, the Bahamas, and Bermuda. The
common occurrence of this species, close to the surface, proves it a shoal
water form, hence it is probable that the specimens taken in deep hauls (all
with open nets) were picked up not far below the surface.
106
[XXIII :5
Zoologica: New York Zoological Society
Pandeidae.
Pandea.
Pandea conica Quoy and Gaimard.
Dianaea conica, Quoy and Gaimard, 1827, p. 182, PI. 6A, Figs. 3-4.
For synonymy, see, Mayer, 1910, p. 118; Hartlaub, 1914, p. 338; Kramp,
1924, p. 8; Uchida, 1927, p. 214; Ranson, 1936, p. 84.
Material : Net No. 649, 1,646-0 meters, May 29, 1930, 1 specimen, about
17 mm. high.
The single example is much contracted, the margin being strongly re-
curved; gonads, radial canals, and tentacular bulbs are, however, in good
condition, and it is chiefly on the basis of the former that the specimen is
identified as P. conica.
The several recent accounts of this species agree that, at maturity, its
gonads form an irregular net-work completely occupying the interradial
areas of the gastric wall (see especially, Mayer, 1910, p. 117, Fig. 63B;
Hartlaub, 1914, Figs. 283, 286; Uchida, 1927, Fig. 38; Ranson, 1936, PI. 1,
Fig. 2). Conditions in the present specimen suggest a somewhat more ad-
vanced stage in development, for while the upper portion of the manubrium
shows much the type of network — the most distinctive generic feature —
usually figui’ed, this gives place in the distal zone next the lip to an irreg-
ular series of folds and knobs, a formation not shown in any of the earlier
figures, though perhaps indicated by Uchida (1927).
There are 34 or 35 tentacles — a number falling well within the recorded
range, the maximum so far recorded being 44 in a specimen 30 mm. high
(Uchida, 1927). The tentacles also show the characteristic laterally com-
pressed basal bulbs, each with a well marked, dark red, abaxial ocellus,
repeatedly described for P. conica.
Earlier accounts have varied as to whether the walls of the radial
canals are smooth (Hartlaub, 1914, p. 339), or more or less jagged (Mayer,
1910, p. 117, Fig. 63; Uchida, 1927, p. 214). In the present example, they
are of the latter type, much as pictured by Uchida (1927) and reported by
Kramp (1928) for a large specimen from Japan. The only feature specifi-
cally characteristic of P. conica that is not clearly shown is the ribbing of
the exumbrella with nematocyst ridges. The bell does show a definite ridge
in the radius of each tentacle. But the exumbral surface has been so badly
rubbed, that one cannot be certain whether these ridges actually represent
the locations of nematocyst ribs, or are merely contraction phenomena. But
the specimen agrees so well in all other respects with the accounts of conica,
and with good specimens of the latter from Naples, with which I have been
able to compare it, that its identity, as such, seems assured.
General Distribution-. P. conica has already been found at a station not
far from Bermuda (Bigelow, 1918). 2 Other captures show a wide distribu-
tion, including many localities in the Mediterranean (Kramp, 1924), off
Scotland, the South Atlantic, Japan, and the Philippines.
Vertical Range-. Recorded depths of capture indicated that P. conica oc-
cupies a considerable depth zone, for while all stages have been found com-
monly at the surface (Kramp, 1924; Ranson, 1936), a fair proportion of
the records have been from nets working down to considerable depths.3 But
the deep hauls yielding it have all been with open nets, nor are data exten-
sive enough to allow any conclusion as to the lower limit to its normal oc-
currence.
2 Identification of this specimen was provisional, but subsequent studies of P. conica show that
it fell well within the varietal range of this species.
3 See especially the table of Thor captures, given by Kramp (1924, p. 9).
1938]
107
Bigelow: Medusae Taken During 1929 and 1930
Pandea rubra Bigelow.
Pandea rubra, Bigelow, 1913, p. 14, PI. 2, Figs. 1-7; Hartlaub, 1914, p.
340; Kramp, 1920, p. 4; 1926, p. 96, PL 2, Fig. 15; Ranson, 1936,
p. 81.
Material : Net No. 562, 914-0 meters, May 10, 1930, 1 specimen, about
25 mm. high.
Net No. 607, 914-0 meters, May 20, 1930, 1 specimen, about 40 mm. high.
Net No. 613, 1,097-0 meters, May 21, 1930, 1 specimen, about 40 mm.
high.
Net No. 800 1,463-0 meters, July 15, 1930, 1 specimen, about 40 mm.
high.
The specimens from Net Nos. 562, 613, and 800 are in good enough
condition to show that they agree closely, even as to details, with the orig-
inal series from the North Pacific, and with those subsequently collected in
the North Atlantic by the Armauer Hansen (Kramp, 1920; 1926.) The most
diagnostic morphological features of this species, as contrasted with its
relative P. conica, are its lack of exumbral nettle ribs, the thin bell wall, the
very complex and fine meshed network formed on the walls of the manu-
brium by the development of the gonads, the complexly crenulated lips, the
small number of tentacles, and the large size attained at maturity. Another
difference is that most accounts of conica credit it with ocelli, whereas such
organs have not been detected in any of the specimens of rubra so far seen.
The largest example yet recorded (from the Pacific) about 75 mm. high,
had about 20 tentacles (Bigelow, 1913) ; Pacific examples of 35-47 mm. had
14-18. Present indications are that the number of tentacles may average
somewhat smaller in medium-sized examples from the Atlantic than from
the Pacific, for in 2 Bermudian specimens about 38-40 mm. high, and in
one of Kramp’s (1920; 1926), of about the same size, the number ranged
from 12-14. But larger series might well show that there is actually no
regional difference in this respect. Specimens 25-31 mm. high (1 Pacific,
2 Atlantic) may have 8 (Kramp, 1926, p. 96), 9 (Bermudian specimen, Net
No. 562), or 12 or 13 tentacles (Bigelow, 1913, PI. 2, Fig. 6).
The gonads of P. rubra are extremely characteristic, the sexual con-
volutions of the manubrium taking the form of an extremely fine and nu-
merous-meshed network; comparison of a photograph of a Pacific specimen
(Bigelow, 1913, PI. 2, Fig. 2) — which would equally well represent the Ber-
mudian specimen from Net No. 613 — with illustrations of P. conica by
Mayer (1910, Fig. 63) and by Ranson (1936, PI. 1, Fig. 2) will illustrate
how widely the two species differ in this respect; how widely, in fact, rubra
differs from any other pandeid yet known. Seen from the inner side, the
sexual thickenings appear as a great number of knobs.
The most arresting feature of P. rubra, among its relatives, is that it
shows extreme development of what might now be termed “bathypelagic”
pigmentation. The Bermuda specimens about 40 mm. high show the same
reddish or chocolate brown coloration as the Pacific specimens of the same
size or larger, the entire manubrium, tentacles, and subumbrella surface
being densely pigmented, with the more palely colored radial canals show-
ing as pale bands. In the Bermudian example of 25 mm., however, in
which the manubrium and tentacles are equally densely colored, the sub-
umbrella is pigmented only in the immediate vicinity of the base of the
manubrium, so that the pigmented canals — radial and circular — show as
dark bands in the transparent bell, as was also the case in the smaller of
the Pacific specimens. As Pacific examples of intermediate size (35-40
mm.) showed an intermediate stage (Bigelow, 1913, p. 16), it appears that
it is not until the medusa is well-grown that pigment spreads out over the
subumbrella — perhaps accompanying a descent into deeper water.
108
Zoologica: Neiv York Zoological Society
[XXIII :5
Ranson (1936, p. 82) does not consider the coloration diagnostic in this
genus. The pigmentation of P. rubra is, however so peculiar in its density
and opacity, and in its eventual extension over the subumbrella (quite apart
from the precise shade of color), and P. rubra differs so widely in these
respects from P. conica, that it must be regarded as a specific characteristic
even though it may not be a safe criterion for identification except for
large specimens.
General Distribution: Previous records include high latitudes in the
northwest Pacific and Bering Sea (Bigelow, 1913) ; and, more recently, two
oceanic stations to the west of Ireland (Kramp, 1920). The captures at
Bermuda now make it likely that its range will finally prove to be as exten-
sive as that of other bathypelagic medusae.
Vertical Range : The shoalest of the 6 hauls that have so far yielded it
was from 411-0 meters ( Albatross Sta. 4800) ; others have been from 549-0
meters or deeper.
Bythotiaridae.
Heterotiara.
Heterotiara anonyma Maas.
Heterotiara anonyma, Maas, 1905, p. 19, PI. 3, Figs. 19-21; Bigelow,
1909, p. 216, PI. 41, Figs. 12-13; 1913, p. 25; 1918, p. 382; Mayer,
1910, p. 107, 489; Vanhoffen, 1911, p. 211, PI. 22, Figs. 3, 4; Hart-
laub, 1914, p. 350.
Material : Net No. 79, 1,280-0 meters, May 8, 1929, 1 very fragmentary
specimen, about 20 mm. high by 17 mm. in diameter.
Net No. 977, Surface, May 12, 1931, 1 specimen, 13 mm. high by 9 mm.
in diameter, with 10 tentacles.
The specimen from Net No. 977 is in excellent condition. As regards
size, in relation to 4he number of tentacles, it falls well within the range
already recorded (Bigelow, 1913, p. 26; 1918, p. 382; 1919, p. 287.
The larger specimen is in such poor condition that identification is only
tentative. Its general appearance seems, however, to warrant reference to
Heterotiara, and its large size to H. anonyma rather than to H. minor.
General Distribution : Previous Atlantic records are between Florida,
the Bahamas, and Cuba (Bigelow, 1918). Elsewhere, it is known from
Malayan waters, the Indian Ocean and several localities in the Pacific,
South and North, including Bering Sea (for list of Pacific localities, see
Hartlaub, 1914, p. 351).
Vertical Range: The recorded vertical range includes various depths
from the surface downward. The comparative frequency with which the
Bache took it shoaler than 100 meters (Bigelow, 1918, p. 382), makes it
likely that specimens from deep hauls with open nets were actually taken
in the upper water layers.
Calycopsis.
Calyeopsis sp.?
Material: Net No. 959, 914-0 meters, September 29, 1930, 1 specimen
about 22 mm. high ; 1 specimen, much crumpled, about 9 mm. high.
The larger specimen is fragmentary, having lost all but the cruciform
base of the manubrium. And it apparently represents an abnormality of
the same type as pictured by Kramp (1924, p. 15, Fig. 11B, 11D) for
Bythotiara murrayi, for the canals are connected in an irregular network,
1938]
Bigeloiv: Medusae Taken During 1929 and 1930
109
the arrangement differing on the halves of the bell. To attempt to give it
a specific name would only add to the difficulties with which the student
of the genus is faced.
LEPTOMEDUSAE.
Laodiceidae.
Chromatonema.
The systematic location of this genus, originally described by Fewkes
(1882), is still a moot question: it is placed by Kramp (1919; 1933) among
the Laodiceidae, which by common consent, are the leptomedusan group
most closely related to the higher Anthomedusae, whereas Ranson (1936)
refers it to the latter, as a member of the Williidae. Discussion of this
point necessarily revolves chiefly around the relationship of the gonads to
the gastrovascular system, i.e., whether the perradial diverticula from the
corners of the stomach, along which the sexual tissue is developed, are
actually part of the manubrium, whether they are merely the dilated
proximal sectors of the radial canals, or whether (as I believe) no sharp
distinction can be drawn between gastric and canalar portions. This last
view has already been proposed by Kramp (1933, p. 552) who says, for
the Laodiceidae as a whole, that the separation of the gonads from the
manubrium is not “durchgeflihrt.” The Bermudian series are not in good
enough condition to throw light on this point. But successive growth
stages in the closely allied C. erythrogonon, from the Pacific (Bigelow,
1909), show, on the one hand, that the development of the gonads does
take place along the radial canals (cf. Bigelow, 1909, PI. 39, Fig. 6 with
Figs. 2-3), but, on the other hand, that as the sex masses increase in size
and complexity, the canal-sectors that they occupy expand in breadth and
so become morphologically incorporated into the cruciform angles of the
manubrium proper. Thus, it is only distal to the extremities of the gonads
that the canals continue to retain their narrow tubular nature. It also
seems probable from the corresponding expansion of entodermal pigment
that Kramp (1919) was correct in crediting a digestive function to the
expanded proximal portions.
If this view be correct, the primary question is no longer whether we
are to regard Chromatonema as a Leptomedusa or as an Anthomedusa,
for it bridges the gap, but to which of the borderline groups it belongs, or
is most closely related. Basing the decision chiefly on the marginal sensory
clubs, it seems wisest to follow Kramp (1919; 1933) in referring it to the
Laodiceidae — at least, provisionally. But future study may well show that
Chromatonema deserves a separate family, for as Kramp (1919) has em-
phasized, its gonads are fundamentally similar to those of Calycopsis,
rather than to those of Laodicea or Ptychogena. I cannot follow Ranson
(1936) in referring it to the Williidae (if his own definition of the latter is
to be accepted), because he confines this family to forms lacking a circular
canal, which is present in Chromatonema as Kramp (1919, p. 9) has shown
and as Ranson (1936, p. 103) himself remarks, and because the typical
members of the Williidae (Willia and Proboscidactyla ) lack marginal organs
other than tentacles.
Chromatonema rubrum Fewkes.
Chromatonema rubrum, Fewkes, 1882, p. 305, PL 1, Fig. 41.
For subsequent references, see Kramp, 1919, p. 7; 1933, p. 552; Ranson,
1936, p. 102.
Material : Net No. 368, 1,280-0 meters, August 14, 1929 and Net No.
110 Zoologica: New York Zoological Society [XXIII :5
959, 914-0 meters, September 29, 1930, 2 very fragmentary specimens, re-
spectively about 17 and 15 mm. in diameter.
One of these specimens still bears fragments of a few tentacles,
though no cordyli ; the other has lost the marginal zone altogether. Identity
seems, however, reasonably assured by what remains of the gonads, which
are of very characteristic appearance in this genus.
The Bermuda capture extends the known range of C. rubrum south-
ward, previous records being west of the Azores (Ranson, 1936) ; off the
American continental slope in Latitude about 40° N. (Fewkes, 1882) ;
and at numerous localities farther north, including Davis Strait (Kramp,
1919, Chart I, p. 13).
The depths of capture of the Bermudian specimens are in line with
previous experience that C. rubrum inhabits chiefly the mid-depths, the
great majority of captures (all with open nets) having been in hauls reach-
ing down to at least 700 meters, and as Kramp (1933, p. 553) remarks
“nur ein paarmal in hoheren Wasserlagen.”
Aequoridae.
Aequorea.
The few Aequoreas in the collection represent the subdivision of the
genus in which the stomach is only 1/4-1/5 as broad as the diameter of
the disc. In contrast to the multiplicity of names under which large-
stomached Aequoreas have appeared in literature, the small-stomached
group includes only two named forms, tenuis and floridana. The first of
these is now generally accepted as a good species (Bigelow, 1913; Kramp,
1933; Ranson, 1936), size of stomach having proved one of the most depend-
able characters in this widely variable genus. But the status of the
second has remained doubtful, awaiting critical examination of the sup-
posed differences — all minor — between it and tenuis. The Bermuda speci-
mens agree so closely with A. Agassiz’s (1865) account of floridana as to
leave no doubt of their identity with the latter. And the need of giving
them a name makes pertinent a decision as to the relationship of floridana
to tenuis; names which appeared on the same page in the original publica-
tion. According to the original accounts, and to Mayer’s (1910) more re-
cent comparison, the chief differences between the two are that floridana
has more tentacles at a smaller size, about four times as many otocysts as
tentacles instead of an equal number, a somewhat smaller stomach (about
1/7 instead of 1/5 as wide as the diameter of the bell), and that its radial
canals are very narrow, its tentacles minute.4 Hargitt (1905), however, had
already shown that in tenuis from Woods Hole (i.e., from the type region)
numerical characters are all highly variable. Thus, a specimen of only 36
mm. may have even more tentacles than one of 47 mm. Canals not only vary
in number, but show much irregular branching and anastomosis. The
number of otocysts (described by A. Agassiz as alternating regularly with
the tentacles) may be one, is often two, and sometimes even more between
every two tentacles and series of typical tenuis from Woods Hole in the col-
lection of the Museum of Comparative Zoology, similarly show from 1-4;
most frequently 2 or 3; the number depending, it seems, on the distance
apart of the large tentacles, and on how many rudimentary tentacles there
are in the interspaces. I may note in passing, that while A. Agassiz (1865,
p. 96, Fig. 138) stated that there is always a single otocyst between 2 ten-
tacles in tenuis, the segment of margin he illustrated was one in which two
large tentacles stood close together with no small ones intervening.
The contracted condition of the specimens in the present collection
4 Mayer’s (1910, PI. 43, Fig. 6) illustration of Ae. floridana is obviously labelled incorrectly,
for it shows a very wide stomach.
1938]
111
Bigelow: Medusae Taken During 1929 and 1930
makes it difficult to make sure of the total number of otocysts. But in other
specimens from Bermuda, in the collection of the Museum of Comparative
Zoology, there are from one to three between every two large tentacles, most
frequently one or two in the sectors on the margin on which the otocysts
are still intact, which agrees with the Woods Hole specimens of tenuis. And
the number of small tentacles, between 2 adjacent large ones, varies from
none to an observed maximum of six, usually 1-3.
Evidently, then, one can no longer invoke numerical precision in the
number of otocysts relative to tentacles as an alternative character between
tenuis and floridana. Neither can the number of large tentacles finally de-
veloped, for while, by published accounts, this is somewhat larger for flori-
dana (up to 100) than for tenuis (up to 90), there is no discontinuity in
this respect. Comparison of the Woods Hole series, just mentioned, with
the Bermudian specimens shows, however, that at diameters of 24-35 mm.
( tenuis is only half grown at this size), tenuis consistently has significantly
fewer large tentacles than floridana, the number in the one case ranging
from 43-58, in the other from 63-103. And a still sharper distinction ap-
pears in the number of canals, for while in the tenuis within this size range
that I have examined, the number of complete canals arising from the stom-
ach, varies from 22-31 (plus 1-4 blind centrifugals in 5 of the seven speci-
mens), which corroborates previous reports of 24-32, the number in the
Bermuda specimens is 16 in every specimen (large or small) where count-
able, except for two (respectively of 25 and 28 mm.) in which it is 15. And
no blind centrifugals are to be seen in any. A. Agassiz’s (1865, Fig. 139)
illustration of floridana also shows 16 complete canals, though with one
short centrifugal also, besides indications of an early stage in formation
of several more, while Mayer (1910, p. 330) reports 16-24 complete canals.
Thus it appears that while occasional specimens of floridana may have
as many canals as an occasional — perhaps stunted — tenuis, the number is
usually only % to % as great in the former as in the latter, with strong in-
dications of a normal determinate number of 16 in the one case, of 32 in
the other. And this difference seems sufficient to warrant recognition of
both species. A minor difference is that floridana (so far as known) is full
grown and has its maximum number of large tentacles at a maximum re-
ported diameter of only about 50 mm.; whereas tenuis, at Woods Hole, is
known to reach a diameter of 100 mm. (Mayer, 1910, p. 332).
Aequorea floridana L. Agassiz.
Rhegmatodes floridanus, L. Agassiz, 1862, p. 361 ; A. Agassiz, 1865, p.
97, Fig. 139.
Rhegmatodes floridana, Haeckel, 1879, p. 223.
Aequorea floridana, Mayer, 1910, p. 330; Bigelow, 1913, p. 37; 1919,
p. 309.
Material : Net No. 840, Surface, September 3, 1930, 1 specimen about
19-20 mm. in diameter, stomach diameter, 4 mm., 16 canals, 72 tentacles;
1 specimen, 23-24 mm. in diameter, stomach diameter about 5 mm., 16 canals
and about 50 large tentacles. As the margin in each case is damaged, exact
measurement of diameter is not possible.
Also, 16 other Bermudian specimens, 12-37 mm. in diameter (besides
fragments) collected in the summers of 1903, 1904, 1911 and 1916, now in
the Museum of Comparative Zoology.
The following tabulation for the Bermuda specimens is pertinent to the
foregoing discussion of the relationship of Ae. floridana to Ae. tenuis.
112 Zoologica: New York Zoological Society [XXIII :5
Date of
Collection.
Approximate
Diameter.
Approximate
Diameter of
Stomach.
Approximate
Number of
Large Tentacles.
Canals.
1904
12
3
37
16
1904
15
3
45
16
1930
20
4
72
16
1930
23
5
50
16
1904
23
6
63
16
1904
25
5
67
16
1911
25
6
63
15
1911
28
7
46
15
1911
28
5
81
16
1911
33
7
74
16
1916
33
7
87
16
1916
33
7
85
?
1904
35
7-8
83
16
1916
37
8
103
16
Relative counts of large tentacles and of small cannot be precise, there
being no sharp demarcation between the latter and rudimentary knobs of
various sizes, the variety in size of which is evidence that while some would
have developed into large, others would have remained permanently rudi-
mentary.
The specimens of 12-15 mm. have no gonads; in those of 20-23 mm.,
gonads have appeared on the outer V3-V2 of the radial canals; in those of
25-28 mm., they occupy the outer %-% and the outer %-% in the larger
specimens of 33-37 mm.
In Ae. tenuis of corresponding sizes, the gonads occupy about the same
sectors of the canals, but with continued growth they finally come to occupy
nearly the wdiole length of canals, though (as Mayer, 1910, p. 332, remarks)
“not quite reaching the circular canal.”
According to Mayer (1910, p. 330), the stomach is relatively smaller in
floridana (only 1/7 as wide as the disc) than in tenuis (about 1/5 the width
of the disc). However, A. Agassiz (1865, Fig. 139) pictures its greatest
diameter as about 1/5. And as the preceding tabulation shows, it ranges
about 1/4-1/6 the diameter of the disc in the Bermuda specimens. But since
the stomach is rarely — if ever — circular, its breadth depends on what par-
ticular radius be chosen for measurement. Total diameter similarly depends
on the degree of contraction (cupping) or relaxation of the disc.
General Distribution: So far as I am aware, previous records of small-
stomached Aequoreas referable to floridana have been confined to the waters
around the southern extremity of Florida. Bermuda is thus a considerable
extension of the known range.
TRACHOMEDUSAE.
Olindiidae.
Olindias.
Recent studies of considerable series from different parts of the world
(Bigelow, 1909; 1919; Mayer, 1910; Neppi and Stiasny, 1913) have made
it reasonably certain that the two named forms from the western Atlantic
( sambaquiensis F. Muller and tenuis Fewkes) represent but a single species,
for the supposed differences between them in number of tentacles, of mar-
ginal knobs, and of blind centripetal canals, and in the diameter of the disc
indicate, at most, that tenuis is a dwarf race of sambaquiensis, or that the de-
1938]
Bigelow: Medusae Taken During 1929 and 1930
113
scriptions of the former have been based on specimens that had not attained
their maximum growth and development. It has in fact been repeatedly sug-
gested that both of these are simply varieties — if as much— of the well-
known 0. phosphorica of the Mediterranean. And the necessity of naming
the Bermudian example listed below makes discussion of this question per-
tinent.
The only ostensible differences between phosphorica and sambaquiensis
are that the original account of the latter (Muller, 1861) credits it with a
larger number of secondary tentacles (200-300) and of centripetal canals
(21-27 per quadrant), than have ever been recorded for phosphorica, for
which the maxima are 100-120 tentacles and 19 centripetal canals in one
quadrant (for tabulation of numerical characters, see Bigelow, 1909, p. 109;
Mayer, 1910, p. 353). That is to say, phosphorica bears the same sort of
relationship to sambaquiensis as does tenuis to phosphorica. By present in-
dications it is not possible to make any definite specific separation, within
this series, because the three forms intergrade in numerical characters. On
the other hand, to unite the three definitely under one name, would obscure
the very interesting fact that all the records from the West Indian-Ber-
mudian region are for the tenuis (or to some extent paedogenetic) form,
which is not known to occur at all in the Mediterranean. Neither, on the
other hand, has Olindias from the latter sea, ever been found to attain the
extreme development described for the Brazilian form. The case, analogous
to that of the representatives of Charybdea from these same regions (p. 137) ,
seems best handled taxonomically, by the recognition of subspecies. For
further discussion of the reasons for this procedure, in the case of geo-
graphic races of medusae, see page 137.
Olindias phosphorica tenuis Fewkes.
Halicalyx tenuis, Fewkes, 1882, p. 277, PI. 7, Fig. 15.
To the references given by Mayer, 1910, p. 354 (" Olindias tenuis”),
add:
Olindias phosphorica [Partim], Bigelow, 1919, p. 317.
Olindias tenuis, Boone, 1933, p. 32.
Material: Net No. 840, Surface, September 20, 1929, 3 specimens, ap-
proximately 10, 10, and 25 mm. in diameter.
The larger specimen has large gonads, between 45 and 50 primary
tentacles, and about 40 centripetal canals, the contracted and recurved state
of the margin preventing exact enumeration. The chief interest of the
series is as cumulative evidence that the Bermudian-West Indian representa-
tives of O. phosphorica are characterized by fewer primary tentacles and
canals at maturity than are their Mediterranean relatives; consequently
that the recognition in nomenclature of tenuis, as a local subspecies, is jus-
tifiable. Owing to the condition of the large specimen, I was not able to
count the secondary tentacles.
General Distribution: O. tenuis has already been reported from Ber-
muda (Bigelow, 1909, p. 109) ; other records are from the Bahamas and
the coast of southern Florida (Fewkes, 1882; Mayer, 1910; Boone, 1933).
Trachynemidae.
Rhopalonema.
Rhopalonema velatum Gegenbaur.
Rhopalonema velatum, Gegenbaur, 1856, p. 251, PI. 9, Figs. 1-5.
For synonymy, see Bigelow, 1909, p. 129; Mayer, 1910, p. 378; Thiel,
114
Zoologica: New York Zoological Society
[XXIII :5
Material-. Net Nos. 32, 35, 39, 43, 44, 67, 69, 97, 98, 100, 102, 362, 581,
620, 632, 634, 682, 720, 730, 793, 915, 916, 945, 959, 967; hauls at 549-0 to
1,829-0 meters; 39 specimens, 6-12 mm. in diameter; 1-5 specimens per haul.
These specimens, all of which are more or less fragmentary, call for
no special comment, this being one of the best known of Trachomedusae. I
need merely note that as all of them are large enough to show the situation
of the gonads, there is no possibility that the list includes any examples of
Sminthea eurygaster, a species somewhat similar in appearance, and which
Thiel (1936, Fig. 2) has recently shown to be much more generally dis-
tributed on the high seas than formerly supposed.
General Distribution: R. velatum was to be expected near Bermuda,
there being many previous records for the eastern side of the Sargasso Sea
region, as well as elsewhere over the warmer belts of all three oceans, in-
cluding also the Mediterranean and Red Sea. For a recent chart of dis-
tribution, see Thiel, 1936, Fig. 1.
It is interesting that 108 hauls shoaler than 549-0 meters did not yield
it, for it often occurs close to the surface, with its center of abundance
shoaler, in general, than 100 meters (Thiel, 1935, Fig. 21).
Pantachogon.
The definitions for this genus adopted by Mayer (1910), also recently
by Broch (1929) and by Thiel (1932), would likewise include the medusa
reported here (p. 117) as Colobonema. But for reasons stated on page 116, it
seems to me preferable to retain the latter genus as distinct, because of the
peculiar order in which its tentacles are developed, and because of the struc-
ture of its radial canals. The structure of its tentacles may perhaps be of
generic significance, likewise. Opinions have also differed as to whether the
generic limits of Pantachogon should be interpreted broadly enough to in-
clude the various species usually reported as Homoeonema, in which the
gonads are localized along restricted sectors of the radial canals (Bigelow,
1913, p. 44), or should be restricted to forms in which these organs are de-
veloped more diffusely along the whole length of the canals (Broch, 1929;
Thiel, 1932, 1936; Ranson, 1936). As the present collection contains repre-
sentatives of the latter sort, only, it does not aid in the solution of this
question.
The group with diffuse gonads includes three named forms, haeckeli
(Maas, 1893), rubrum (Vanhoffen, 1902), and scotti (Browne, 1910). I
have already expressed doubt as to whether the first two could be separated,
the color being the only difference. As Ranson (1936) again emphasizes,
successive studies of considerable series have revealed no anatomical differ-
ences whatever between them, hence he has now united them provisionally,
Thiel (1936) definitely, and I believe with good reason. I may add that the
capture near Bermuda of several specimens, lacking any trace of pigment
(i.e., of the haeckeli type), now answers the last argument in favor of con-
sidering the two color-forms as distinct, i. e., that the colorless haeckeli was
restricted to high latitudes. Nomenclatural rules require the choice of
haeckeli Maas as the name for the joint species, because Maas’ account,
with clearly recognizable illustration showing the gonads and tentacles cor-
rectly, was published many years prior to the description of rubrum (Van-
hoffen, 1902).
P. scotti, an Antarctic species, is separated from haeckeli by much more
numerous tentacles (104-128, contrasted with a maximum of 64 in haeckeli).
And so many specimens of the latter have now been examined that this
difference seems established.
1938]
Bigeloiv: Medusae Taken During 1929 and 1930
115
Pantachogon haeckeli Maas.
Pantachogon haeckeli, Maas, 1893, p. 17, PI. 1, Fig. 2; 1904, p. 29;
Mayer, 1910, p. 389; Bigelow, 1913, p. 44, PI. 3, Figs. 4-8; Broch,
1929, p. 503; Thiel, 1932a, p. 152; 1936, p. 26.
Pantachogon rubrum, Vanhoffen, 1902a, p. 63, PI. 9, Fig. 9, PI. 10,
Figs. 19-20, PI. 11, Fig. 25; Maas, 1905, p. 55, PI. 10, Fig. 66;
Mayer, 1910, p. 389; Kramp, 1913, p. 274; 1914, p. 433; 1920, p. 5;
1924, p. 22; Bigelow, 1913, p. 44; Browne, 1916, p. 195; Broch,
1929, p. 504; Thiel, 1931, p. 330; 1932a, p. 153; Ranson, 1936, p.
147.
Pantachogon rubrum [Partim], Thiel, 1936, p. 24.
Material : Net Nos. 36, 42, 52, 64, 234, 296, 539, 552, 554, 575, 582, 598,
628, 648, 657, 658, 659, 666, 684, 685, 691, 695, 698, 709, 717, 718, 736, 739,
742, 757, 780, 789, 805, 809, 854, 862, 864, 866, 881, 897, 916, 964; 549-0 to
1,829-0 meters; 65 specimens; 1-5 per haul.
Also, 12 very fragmentary specimens, from Net Nos. 556, 640, 698, 709,
721, 748, 862, 929; 1,280-0 to 1,646-0 meters, ai'e provisionally referred to
this species.
The specimens are all more or less fragmentary — most of them very
much damaged — but those included in the first of the two preceeding lists
have enough of the margin and of the subumbrella intact to show that the
tentacles are so spaced as to accord with the number characteristic of this
species (maximum, 64), and that the radial canals are of even bore through-
out their lengths. I have also been able to compare them with the series
collected by the Albatross in the northwestern Pacific (Bigelow, 1913).
Tentacles : Vanhoffen (1902a) counted 7 tentacles per octant, between
every two perradials, in all his specimens. Maas (1905), however, found
smaller numbers in some octants, i.e., intermediate stages in development.
And the Bermuda series again illustrates this, for while the sub-radials are
seven in number in most of the octants in which complete counts can be
made, only 6 are to be seen in some octants. And one specimen seems to
have not more than 6 in any octant, i.e., a total of only 56. On the other
hand, a greater number than 7 sub-radials has never been found in any
octant, nor have additional tentacular bulbs been observed. That the num-
ber is determinant in Pantachogon (as it is in Colobonema ) thus seems
sufficiently established. The order of development of the different series of
tentacles has not been traced; nor is this point clarified by the present
series, preservation not being good enough for dependence to be placed on
differences in size between different tentacles. It is, however, suggestive
that a tentacle does not always stand exactly opposite a canal, i.e., it appears
that in some cases none can be described as strictly “perradial ;” neither
are those most nearly opposite the canals appreciably larger than the others.
An interesting question, still awaiting answer, is whether the tentacles
of Pantachogon, like those of Colobonema, are stump-like, as suggested by
the fact that they have been very short, and of sub-equal length, in all the
specimens so far examined, or whether Maas (1905) was correct in his
suggestion that they are of ordinary filamentous type, but so easily broken
off that all except the bases have been lost. None of the Bermudian and
northwest Pacific specimens are in good enough condition to settle this
point, though the general appearance of the tentacles — especially the fact
that in many cases, they are square tipped — favors the second alternative.
Gonads: In one small specimen, about 8 mm. high, in which the gonads
are intact, they represent about the same early stage in development illus-
trated for a North Pacific specimen of corresponding size (Bigelow, 1913,
PI. 3, Fig. 5). In the larger specimens, the gonads have either been lost
116 Zoologica: New York Zoological Society [XXIII :5
entirely, or are represented by fragments only, here and there along the
canals.
Musculature : The circular musculature of the subumbrella is strongly
developed (Bigelow, 1913; Ranson, 1936); a development as striking, in
fact, as it is in Colobonema. In Pantachogon, as in Colobonema, however, the
musculature is not only entirely discontinuous along the lines of the radial
canals, but terminates, apically, at the level where the general contour of
the subumbrella changes from sub-cylindrical to domed, leaving a consider-
able clear space surrounding the base of the manubrium. And Ranson
(1936) has recently pointed out that the outlines of the apical ends of the
interradial muscular fields provide at least a specific character, being trans-
versely truncate in P. haeckeli, as appears clearly in the few Bermuda speci-
mens in which the musculature is still intact, whereas in Colobonema they
are ovoid, corresponding to the contours of the interspaces between the
apically expanded radial canals (p. 118). The latter, as Ranson (1936) has
emphasized, are narrow and of uniform caliber, throughout their lengths,
from bell margin to union with the manubrium, contrasting strongly with
their state in Colobonema.
Color : Occasional specimens show a pinkish tinge. But the few in which
the subumbral musculature is more or less intact are colorless: interesting
evidence, as noted above (p. 114), that the unpigmented form of this species,
as well as the rosy-colored, occurs in the warm belt of the North Atlantic.
Thiel (1936, p. 24), in fact, has already shown that the two may exist side
by side, the Meteor having taken one of each at a station in the South
Atlantic. No recent student of Pantachogon has considered this color differ-
ence a sufficient ground per se for specific separation. But it is an interest-
ing question, how the existence of the two color phases is to be interpreted.
As the present series throws no light on this point, I need only note that the
pigmentation of the rosy form ( P . rubrum, auct.) is not of the peculiarly
dense and opaque sort characteristic of many bathypelagic medusae.
General Distribution: Locality records for P. haeckeli, as here defined,
are widespread in the Atlantic, including Spitzbergen, Irminger Sea and
Davis Strait in the north, sundry localities off Ireland, in the Bay of Biscay,
and in the region between the Canaries, Azores, and Gibralter; the Mediter-
ranean ; the eastern side of the tropical and South Atlantic south to Lat.
64° S.; the Antarctic (Weddell Sea) ; South and North Indian Oceans and
Red Sea; Malaysia; and the extreme northern Pacific including Bering Sea
and the Sea of Okhotsk. With the Bermuda records extending the known
range to the western Sargasso Sea, — the nearest previous record was in mid-
Atlantic — P. haeckeli is evidently world wide in the ocean basins, right up
to the subpolar zones.
Vertical Range: The great majority of captures have been in open nets
working as deep as 500-0 meters, the only exception being a single young
specimen taken by the Valdivia in a haul from 60 meters (Vanhoffen, 1902).
Colobonema.
There has been much discussion recently as to whether Colobonema is
generically separable from Pantachogon. Thiel (1936) at the one extreme,
has even included its unique representative in the synonymy of P. rubrum
(= P. haeckeli) . But this seems definitely precluded not only by the constant
difference in the number of tentacles, but also by the structure of the
proximal sectors of the radial canals. Other students who have recently
written of Colobonema (Bigelow, 1913; 1919; Kramp, 1924; Broch, 1929;
Ranson, 1936) have therefore retained it as a separate genus, either defi-
nitely or provisionally. And that course is followed here.
The question what specific name should be applied to the one known
representative of Colobonema is one that will not down, because a final an-
1938] Bigelow: Medusae Taken During 1929 and 1930 117
swer depends on decision as to the accuracy, or the reverse, of the published
description of a specimen no longer extant. Accepting Maas’ (1905) subse-
quent statement that his earlier (1897) account of his Homoeonema typicum,
which credited the latter with more than 32 tentacles, was an error resulting
from the poor condition of the specimen, Mayer (1910), Uchida (1928), and
I (Bigelow, 1913; 1919) have used this specific name, for in all other re-
spects, Maas’ typicum appears to resemble the form later described by
Vanhoffen (1902a) as Colobonema sericeum. Kramp (1920; 1924), Broch
(1929), and Ranson (1936), however, take the opposite view, that Maas’
(1897) original typicum cannot be identified with sericeum, hence use the
latter name.5
The present series adds nothing toward solution of this nomenclatural
puzzle, beyond the opportunity for pointing out that choice of typicum as the
specific designation has the advantage of providing a resting place for a
name which is likely otherwise to remain permanently without connection
with any actually existant medusa.
Colobonema typicum Maas.
Homoeonema typicum, Maas, 1897, p. 22, PI. 3, Figs. 1-3.
For early synonymy, see Mayer, 1910, p. 385.
It has been referred to, subsequently, under the following names:
Colobonema sericeum, Kramp, 1920, p. 5; 1924, p. 28; Ranson, 1936,
p. 152, PI. 2, Figs. 14-15.
Homoeonema ( Colobonema ) sericeum. , Broch, 1929, p. 500.
Colobonema typicum, Bigelow, 1913, p. 46; 1919, p. 322.
Homoeonema typicum, Uchida, 1928, p. 81.
Material: Net Nos. 26, 39, 94, 95, 99, 105, 240, 292, 293, 295, 296, 400,
434, 565, 597, 605, 606, 607, 610, 625, 626, 637, 641, 653, 680, 732, 760, 777,
802, 805, 813, 853, 867, 874, 937, 939, 964; 732-0 to 1,829-0 meters; 47
specimens; 1-3 per haul, mostly in poor condition.
The specimens range from about 8 to about 33 mm. in heighth. Fea-
tures especially to be examined on any considerable series of this species
that may now come to hand, are: (a), number and order of appearance of
the tentacles, whether as constant as earlier records have indicated; (b),
number of otocysts; and (c), contours of the proximal parts of the radial
canals, which Ranson (1936, PI. 2, Fig. 14) has found to differ significantly
from those of Pantachogon.
Tentacles : Successive examinations have yielded cumulative evidence,
not only that the final number of tentacles in this species is 32, but that
variations from this number are so unusual that none has been found among
123 specimens collected by the Valdivia (Vanhoffen, 1902a) Siboga (Maas,
1905), Research (Browne, 1906), Albatross (Bigelow, 1909; 1913; 1919),
Thor (Kramp, 1924), and by the Prince of Monaco (Ranson, 1936). The
peculiar order of development — perradials, adradials, and finally interradials
— first reported by Maas (1905), has proved equally unvarying, judging
from the relative sizes of the different orders of tentacles on small and
medium-sized specimens. Both these findings are further corroborated by
the present series, for in every octant on which they can be counted, there
is the usual interradial, and pair of adradials in the larger specimens. In
one of the smaller specimens about 20 mm. high (Net No. 26), the inter-
radial has not yet appeared in one of the octants, while in 6 of the other
octants6 it is represented by a rudimentary knob only, illustrating a stage
between that in which the per- and adradials alone have appeared (Maas,
5 The species subsequently described by Maas (1905) from the Siboga collection, as typicum,
undoubtedly was identical with Vanhoffen’s sericeum.
6 The margin of the eighth octant has been torn off.
[XXIII :5
118 Zoologica: Neiv York Zoological Society
1905, Fig. 64), and the final stage in which the interradials are also well
developed.
In all specimens so far examined (including the Bermuda series) the
tentacles have been short, stump-like. And it seems now established that
this is normal, and not the result of mutilation, for in the better preserved
specimens the tips are rounded and apparently uninjured, i.e., with their
ectodermal covering complete. And the fact that in large specimens the
members of the per- and adradial series (sometimes the interradials also)
are all about equal in length, suggests that they are but little extensible.
Otocysts : The otocysts of this species must be very easily detached, for
only in one case has a single otocyst been detected, among the earlier collec-
tions (Bigelow, 1919, p. 323). One of the Bermuda specimens (Net No. 26)
also bears one otocyst — agreeing in structure with the earlier account — be-
tween a per- and an adradial tentacle. But I have not been able to detect
even a trace of any others, in this, or on any of the other specimens. To
determine positively whether otocysts normally alternate with tentacles, as
earlier observations suggest (Bigelow, 1913, p. 321), requires examination
of specimens with undamaged margin.
Radial Canals: Ranson (1936, p. 154, PI. 2, Fig. 14) has recently pointed
out (what seems to have been overlooked previously) that the radial canals
of Colobonema are expanded proximally so that their walls are in close
contact (though discontinuous) in the interradii across the domed apex of
the subumbrella, narrowing again toward their union with the narrow base
of the manubrium, which gives characteristic ovoid contours to the apical
ends of the 8 muscular zones. The better preserved of the Bermuda series
corroborate this interpretation, so far as can be determined from surface
views, though in all cases the base of the stomach is damaged.
General Distribution: Colobonema, as distinguished from Pantachon
haeckeli7 had been taken already at many stations in regions as widely scat-
tered, as the central Sargasso Sea; the eastern North Atlantic; west of Ire-
land ; between the Azores, Canaries, and the coast of Spain ; Straits of
Gibraltar; Mediterranean; equatorial and South Atlantic down to Latitude
42°S. ; the Indian Ocean from Lat. 9°N. to Lat 34°S. ; Malaysian region;
Philippines; Japanese waters; and the eastern tropical Pacific. It is, in
short, cosmopolitan in the warm and temperate belts, at appropriate depths,
but has not yet been reported from Arctic or from Antarctic latitudes.
Vertical Range: So far as I have been able to learn, the shoalest haul
that has so far yielded Colobonema was with an open net, from about 241-0
meters, most of the captures (all but one with open nets) being from hauls
deeper than 500-0 meters; and there is one record from a closing net from
900-1,500 meters. Its bathypelagic habit is thus well established.
Crossota.
Crossota, originally classed with the Ptychogastridae (Vanhoffen, 1902a;
Bigelow, 1909), was soon shown to be a trachynemid (Bigelow, 1913). But
it is not yet possible to give any precise generic definition for it that can be
depended upon to stand the test of time. During the first years of its his-
tory it was, indeed, regarded as a very sharply defined genus, for while the
type species (C. brunnea Vanhoffen) agrees with Aglaura and Aglantha in
the structure of the gonads, it differs markedly from them not only in lack-
ing any trace of a gelatinous peduncle, but also — and from all other Trachy-
nemidae as well — in having more than one row of tentacles. Shortly, how-
ever, other species were described, one of which ( rufobrunnea , Kvamp,
1913) apparently had the tentacles in a single row,8 while in another (C.
7 Thiel’s (1936) chart of distribution for Pantachogon also includes the Colobonema records,
for he classes the latter as a synonym of the former.
8 Broch (1929, p. 506) says “Ersichtlich in einer Reihe.”
1938] Bigelow: Medusae Taken During 1929 and 1930 119
pedunculata Bigelow, 1913), there is a short gelatinous peduncle. The for-
mer partially bridges the gap between Crossota and Pantachogon, the latter
between Crossota and Aglantha, i.e., between the subfamilies Rhopalone-
minae and Aglaurinae, as defined by Broch (1929) and by Thiel (1936, p. 9).
Indeed, pedunculata is tentatively referred to Aglantha by Thiel (1936, p.
22). But this course would involve expanding that genus to include species
with tentacles in more than one series, for in C. pedunculata these are in
three or four irregular rows, much as in Crossota brunnea. Until Kramp
and Damas’ (1925, p. 317) characterization of the tentacles of norvegica
as being in a single row is confirmed, it seems simplest to confine Crossota
to Trachynemidae with sausage-shaped gonads hanging free from the sub-
umbrella and with many closely crowded tentacles, in more than one row,
irrespective of whether or not there is a short peduncle. Data as to the
number of otocysts is much to be desired. In the only species in which any
estimate of this has been possible, they have proved much more numerous
than they are in Aglantha.
? Crossota brunnea Vanhoffen.
Crossota brunnea, Vanhoffen, 1902a, p. 73, PI. 9, Figs. 11-13, PI. 12, Figs.
34-38, 43-47; Bigelow, 1909, p. 135, PI. 2, Fig. 7, PI. 45, Fig. 9;
Mayer, 1910, p. 396.
Crossota brunnea [Partim], Thiel, 1936, p. 20°.
Crossota brunnea, var. norvegica, Bigelow, 1913, p. 48.
Crossota norvegica, Broch, 1929, p. 5Q7.
? Crossota norvegica, Kramp, 1920, p. 5.
Non Crossota norvegica, Vanhoffen, 1902a, p. 75; Kramp and Damas,
1925, p. 317.
See Thiel (1936, p. 21) for a recent summary of the checkered history
of C. norvegica Vanhoffen and for the reasons why norvegica can-
not be used as the name for a variety of C. brunnea as I formerly
did (Bigelow, 1913).
Material : Net Nos. 241, 642, 665, 733, 787, 794, 854, 869; 1,097-0 to
1,829-0 meters; 11 very fragmentary specimens, all approximately 14 mm.
in diameter.
These specimens are all in such poor condition that identification is
only tentative; in every case all trace of the manubrium and gonads has
been lost and only faint indications of the radial canals are to be seen. At
first sight, it seemed that they might belong to C. rufobrunnea Kramp, for
in most cases the tentacular bases — in their present state — have the appear-
ance of being in a single row. But in one specimen in which the margin is
in somewhat better condition than in the others, the larger (older) tentacle-
bases arise at a slightly higher level than do the smaller (younger) ones, for
which reason it seems more probable that the whole series belongs to
C. brunnea. In some parts of the margin, 3 successive sizes — or rows — of
tentacles can be recognized, in others only two, as was the case in the
Meteor specimen recently described by Thiel (1936, p. 20). In the speci-
mens of brunnea that I have seen, in good enough condition to show the
tentacles (Bigelow, 1909; 1913), the arrangement in different rows has been
less regular than Vanhoffen (1902a, PI. 12, Fig. 47) pictured it, the tentacles
being more crowded and the older only a very short distance above the
younger, instead of well above the latter: this distribution is more cor-
rectly represented in Vanhoffen’s 1902a, PI. 12, Fig. 46) section of the
margin.
,J Thiel (1936) includes C. alba Bigelow in the synonymy of C. brunnea.
120
Zoologica: Neiv York Zoological Society
[XXIII :5
It was not possible to make a reliable count of the tentacles, even in
any one octant, owing to the uncertainty as to the precise location of the
radial canals. In the specimen just mentioned, the spacings of the ten-
tacular-bases suggests a total of at least 200. No otocysts were to be found.
In most of the specimens enough of the subumbral pigmentation is
still intact to show that it was of the characteristic opacity and chocolate or
reddish brown color.
General Distribution: Previous records probably referable to C. brun-
nea, as mapped by Thiel (1936, p. 23), are widely scattered in the eastern
side of the Atlantic from Latitude about 50° N. down to the Antarctic —
extended, now in all probability to the western half of the Sargasso Sea — ;
in both sides of the Indian Ocean ; and in the parts of the eastern tropical
and far northern Pacific traversed by the Albatross on the cruises of 1904-
1905 and 1906. It has not been taken, however, in the Mediterranean. And
while reaching high latitudes in the south, it apparently does not inhabit the
North Polar regions.
Vertical Range: All captures have been in deep hauls.
? Crossota.
The following fragments (now mere shells) are tentatively referred to
Crossota because such indications of the locations of the tentacles as are
still to be seen suggest that these were numerous and close crowded, and
because most of the specimens still show some trace of subumbral pigment.
But specific identification is out of the question.
Net Nos. 41, 66, 74, 75, 101, 114, 137, 138, 220, 295, 640, 644, 647, 666,
724, 748, 770, 779, 807, 812, 815, 844, 849, 867, 885, 891, 917, 935, 959, 967,
970; 914-0 to 1,830-0 meters; fragments of 37 specimens.
Trachynemidae gen. ?
Medusae, apparently belonging to the Trachynemidae, but in such poor
condition as to preclude even provisional identification, were taken in Net
Nos. 110, 145, 269, 295, 317, 322, 327, 341, 449, 574, 577, 598, 613, 644, 680,
715, 718, 726, 738, 767, 785, 795, 799, 807, 824, 827, 867, 882, 885, 896, 926,
968; a total of 51 specimens, 1-4 per haul.
Halicreidae.
Halicreas.
Opinions have differed whether such of the Halicreidae as have only
8 canals, and tentacles in a continuous row (i.e., not grouped as they are in
Botrynema') should all be referred to the one genus, Halicreas, or to two
genera; and if the latter, whether generic separation should be based on
number of tentacles alone, as by Vanhoffen (1902a), on this character com-
bined with the form of the umbrella, as by Ranson (1936) ; or on the pres-
ence or absence of exumbral papillae, as formerly by me (1909).
Even granting eventual agreement as to the relative taxonomic value
of different characters, we still face a nomenclatural difficulty (if two gen-
era are to be recognized), in the fact that the decision as to the correct name
of the second genus would depend on the true identity of the specimen
originally described by Maas (1893) as Homoeonema platygonon, and later
declared by him (Maas, 1906, p. 3) to be a halicreasid, not a trachynemid as
he originally supposed.
Examination of the type specimen of H. platygonon, if this be still in
existence, would settle the point. Meanwhile it seems simplest to follow
Broch (1929) and Thiel (1936), in uniting, in Halicreas, all halicreasids
1938]
Bigelow: Medusae Taken During 1929 and 1930
121
with 8 canals and continuous series of tentacles, whether the exumbrella be
papillate or smooth, and whether the tentacles be few or numerous, all of
one size, or of two sizes. For further discussion, see Ranson (1936) and
Thiel (1936).
Among the named forms referable to Halicreas as thus expanded, two
( H . minimum Fewkes and H. papillosum Vanhoffen) are set apart by the
presence of exumbral papillae. Thiel (1936), it is true, bases the specific
distinction within Halicreas solely on whether the tentacles are of one size
or of two sizes. I see no reason, however, for abandoning the presence or
absence of exumbral papillae as of specific value here, even if it be not of
generic worth as I formerly suggested ; not only are these structures highly
distinctive features, but no variation has ever been found in their number.
And Ranson (1936) also considers them as specific.
H. papillosum is the best-known member of the genus, and large series
of it have been taken on most of the recent expeditions that have done deep
towing. It has, however, long been generally recognized that it might
eventually prove to be identical with the H. minimum of Fewkes (type
species of the genus) which was described 20 years the earlier, consequently
minimum has repeatedly been classed as a doubtful synonym of papillosum
in the lists of references to the latter. Fortunately, the collection of the
Museum of Comparative Zoology contains one of Fewkes’ original two speci-
mens (previously overlooked) ; evidently the one described by him (1882, p.
306) as “expanded” and “preserved in spirit.” Although this specimen has
subsequently been dessicated, then restored to alcohol, the typical exumbral
papillae (each with several conical projections) are still to be seen, while
the locations of the 8 radial canals are marked as broad, flat ridges on the
subumbrella. And fragments of the circular lip are still recognizable. What
Fewkes took for the velum was in reality the marginal zone of the bell,
distal to the exumbral papillae. Thus, while the tentacles have all been lost
— as indeed Fewkes pointed out — there is no reason to doubt that it is the
same species that Vanhoffen (1902a) later described as papillosum, while
Fewkes’ specimens were from a locality where many papillosum have since
been taken, on the cruises of the Woods Hole Oceanographic Institution.
Consequently, there need no longer be any hesitancy in superseding papillo-
sum by minimum, thus making one more step toward stability in the
nomenclature of the medusae.
It has long been obvious that the five named representatives of the
smooth-belled subdivision of the genus ( conicum , album, glabrum and ro-
tundatum Vanhoffen, 1902a; recovitzae Maas, 1906) stand in need of drastic
reduction, successive studies (Bigelow, 1909; Ranson, 1936) having proved
that the features on which Vanhoffen (1902a) laid chief stress (degree of
doming of the exumbrella, and precise location of the gonads on the radial
canals), are too variable to be of taxonomic significance. Thiel’s (1936)
recent grouping, based on whether there are two sizes of tentacles or only
one, is a decided advance. It is certain that the tentacles are all alike in
racovitzae, for Maas (1906) definitely states as much; likewise in conicum
(Vanhoffen, 1902a, PI. 11, Fig. 33). On the other hand, it seems equally
certain that the original specimens of glabrum had tentacles of two sizes,
because Vanhoffen (1902a, p. 70) describes them as “ebenso wie bei H.
papillosum angeordnet.” Hence — as glabrum has page priority in Vanhoffen’s
original paper — this name may be accepted for smooth-belled Halicreas with
2 sizes of tentacles.
Unfortunately, Vanhoffen gave no definite information on this point,
either for his Halicreas album or his Halicreas rotundatum. The first of
these is placed by Thiel (1936) in the group in which the tentacles are all
of one size (i.e., grouped with racovitzae and conicum ). And the original
accounts certainly would allow this interpretation. But Vanhoffen’s (1902a,
p. 68) statement that “so wohl Halicreas wie Haliscera” has tentacles of
122
Zoologica: New York Zoological Society
[XXIII :5
two sizes, makes it seem more likely that this was true of his specimens of
album and rotundatum. Hence it seems justifiable to class these names
provisionally as synonyms of glabrum, recognizing, however, that the wide
variation in numbers of tentacles between the several forms may finally
warrant a further subdivision on this basis.
Halicreas minimum Fewkes.
Halicreas minimum, Fewkes, 1882, p. 306.
Halicreas papillosum, Vanhoffen, 1902a, p. 68, PI. 9, Figs. 7-8; PI. 11,
Fig. 30.
For the more important subsequent references, see Halicreas papil-
losum, Mayer, 1910, p. 391; Ranson, 1936, p. 164; Halicreas papil-
losum [Pai’tim], Thiel, 1936, p. 34.
Material : Net Nos. 25, 26, 41, 43, 44, 56, 60, 62, 65, 66, 67, 68, 69, 75,
76, 81, 83, 108, 110, 115, 124, 138, 144, 157, 203, 241, 247, 269, 270, 271, 293,
295, 297, 357, 360, 383, 401, 405, 406, 442, 466, 468, 481, 488, 501, 566, 569,
570, 576, 590, 599, 640, 648, 654, 682, 683, 684, 701, 703, 717, 718, 722, 723,
724, 727, 728, 730, 736, 741, 746, 750, 751, 756, 765, 767, 774, 781, 787, 788,
795, 799, 815, 852, 857, 858, 863, 864, 869, 875, 887, 893, 898, 919, 931, 932,
933, 935, 940, 942, 947, 963, 969, 970; at depths ranging from 0 to 2,012-0
meters; a total of 186 specimens; 1-6 specimens per haul, except for one
catch of 17 specimens in Net No. 271.
The specimens, ranging from about 13 mm. to about 36 mm. in diam-
eter, are all fragmentary, having lost every trace of marginal organs and of
the gastrovascular system. Hence, they add nothing to previous accounts of
the morphology of this interesting bathypelagic medusa. This species is,
however, made so easily recognizable by the marginal papillae — structures
so resistant that they persist even after the roughest handling — that iden-
tity seems assured. Most of them also show more or less trace of the
gelatinous apical projection, though the discs of others are apically rounded,
in their present condition.
General Distribution: Earlier records of captui’e are widespread over
the eastern side of the tropical Pacific, also the tropical and subantarctic
belts of the Indian Ocean, the East Indies and Philippines, Japan, Sea of
Okhotsk, Bering Sea, the South Atlantic down to the Antarctic, north boreal
and east temperate belts of the North Atlantic (Thiel, 1935, Fig 38; 1936,
p. 36, Fig. 7; Ranson, 1936). It had been taken on the continental slope of
America in the offing of Woods Hole. Curiously, however, there was no
previous record of it in the Sargasso Sea region. The Bermuda records,
proving it abundant there, are therefore interesting as cumulative evidence
that it is cosmopolitan in the ocean basins at suitable depths, at low and
mid-latitudes.
However, it has not been taken in the Mediterranean, in spite of the
great number and wide distribution of deep hauls that have been made there
by the Thor and by the Prince of Monaco (Kramp, 1924; Ranson, 1936).
And its absence, or at least rarity, there is puzzling, for the comparative
frequency of records from small depths at low latitudes in the Atlantic (see
below), makes it unlikely that the shoalness of the sill depth of the Straits
of Gibraltar can offer a permanent barrier to its entrance into the Mediter-
ranean, as it apparently does in the case of certain more strictly bathy-
pelagic animals (Bigelow and Sears, 1937, p. 128). For further discussion
of its distribution and dispersal in the South Atlantic, see Thiel, 1935, p.
72.
Seasonal Distribution : The seasonal distribution of the hauls yielding
H. minima, for the two years combined, appear from the following tabula-
tion :
1938] Bigelow: Medusae Taken During 1929 and 1930 123
Month.
Hauls Yielding
H. minimum.
Total Hauls,
1,280-0 Meters
or Deeper.
% Yielding
H. minimum.
April.
6
20
30
May.
24
77
31
June.
22
135
16
July.
17
109
15
August.
5
40
12
September.
29
158
19
This shows a somewhat greater abundance in spring, for the two collecting
seasons combined, than during the summer months.
Vertical Range : The early records were all from hauls with open nets,
leaving in doubt the precise depths from which the specimens came, but
the great majority of the catches were from hauls that sampled down to
500 meters or deeper. Thus, the shoalest of the 40 Valdivia hauls that
yielded specimens positively identified as H. minimum was from 600-0
meters, 37 of them at least 1,000-0 meters10 or deeper (Vanhoffen, 1902).
All of the 30 Albatross catches were, similarly, in hauls from 550-0 meters
(Bigelow, 1909; 1913; 1919), while the records of the Prince of Monaco
were all from 1,000-0 meters or deeper (Ranson, 1936).
The Meteor records have shown, however, that H. minimum is not ex-
clusively restricted to great depths, but may populate the whole column
from, say, 75 meters down to 800-900 meters, as appears from the distribu-
tion of catches in closing nets, tabulated below, from data presented by
Thiel (1935, Tables 10, 21, Fig. 27; 1936, p. 34) :
Depth in
Meters.
Number
of Cases.
Number of
Specimens.
50- 70
1
1
50-100
7
8
100-200
2
2
200-400
5
7
400-600
4
5
600-800
2
1
And still more recent captures by Atlantis of the Woods Hole Oceanographic
Institution make it unlikely that there is any definite lower boundary to its
occurrence, any more than for various siphonophores (Bigelow and Sears,
1937, p. 137), for horizontal closing net hauls of uniform duration at 3
stations as described by Leavitt (1935) in the summer of 1935, yielded the
following catches:
Depth in
Meters.
Number
of Cases.
Number of
Specimens.
600
1
6
1,200
1
4
1,800
1
3
2,000
1
2
2,200
2
2
10 One juvenile, perhaps of this identity, was taken at 200-0 meters.
124
Zoologica: New York Zoological Society
[XXIII :5
The Bermuda collection amplifies the foregoing, proving that H. mini-
mum may occasionally come right up to the surface (Net No. 654, 1 speci-
men). But in this part of the Sargasso Sea, where the thickness of the
superficial stratum of warm water is great, the chief center of abundance
lies deeper than in most parts of the oceans, or at least as deep as 1,400-
1,800 meters, as appears from the following tabulation for the two seasons
combined :
Depth in
Meters.
No. of Hauls
with
minimum.
Total
Number
Hauls.
% with
minimum.
732-0
2
40
5
913-0
3
132
2
1,097-0
2
134
2
1,280-0
8
132
6
1,463-0
21
132
15
1,646-0
34
135
25
1,829-0
31
135
22
2,012-0
3
4
45
H. minimum (seemingly also H. glabrum and Halitrephes ) thus is a marked
exception to the general rule that medusae having their center of abundance
so deep, are densely pigmented.
Thiel (1935, p. 47) has shown, from the catches made in closing nets by
the Meteor, that the individuals of the genus Halicreas, as a whole, average
larger, the greater the depth. This suggests that the young stages are
passed chiefly in the shoaler strata and that the medusae tend to sink as
they grow larger. But we still await positive evidence whether the eggs, as
a whole, rise to the surface, as he suggests (Thiel, 1935, p. 72).
? Halicreas glabrum Vanhoffen.
Halicreas glabrum, Vanhoffen, 1902a, p. 70, PI. 9, Fig. 3; Mayer, 1910,
p. 392; Ranson, 1936, p. 167.
Homoeonema glabrum, Bigelow and Leslie, 1930, p. 564.
Haliscera alba, Vanhoffen, 1902a, p. 71, PI. 9, Fig. 5.
Homoeonema alba, Bigelow, 1909, p. 142, PI. 3, Fig. 1-2, PI. 33, Figs. 6,
11, PI. 34, Fig. 9.
Halicreas album [Partim], Thiel, 1935, p. 48; 1936, p. 37.
Halicreas rotundatum, Vanhoffen, 1902a, p. 71, PI. 9, Fig. 4; Maas, 1905,
p. 57 ; Mayer, 1910, p. 392.
Halicreas papillosum [Partim], Thiel, 1936, p. 34. [Non H. papillosum,
Vanhoffen, 1902.]
Material : Net Nos. 25, 43, 56, 62, 68, 75, 83, 89, 106, 114, 144, 145, 157,
218, 242, 247, 269, 270, 271, 295, 297, 315, 339, 345, 346, 347, 354, 359, 449,
466, 488, 501, 506, 539, 710, 742, 890, 942, 962; hauls from 914-0 meters
to 2,012-0 meters; 66 specimens, 1-5 per haul.
These specimens — like Ranson’s — agree very closely with Vanhoffen’s
(1902, p. 9, Fig. 3) illustration of H. album in the form of the bell, with
high conical apex, and thin marginal zone, without any sign of the exumbral
protuberances. But as already suggested (Bigelow and Leslie, 1930, p. 564),
and as indicated above (p. 122), it is probable that album is specifically
identical with Vanhoffen’s H. glabrum and the latter name has page priority.
1938]
Bigeloiv: Medusae Taken During 1929 and 1930
125
In every case, all trace of all the tentacles has been lost, preventing posi-
tive identification; in fact the great majority are mere shells. But in a few,
parts of the subumbrella still remain. And one from Net No. 890, another
from Net No. 942, are especially interesting, being in good enough condi-
tion to show the oval gonads situated proximally, leaving the distal parts of
the radial canals free, as was the case in the Albatross specimens (Bigelow,
1909). And this corroborates Ranson’s (1936, p. 168) observation that in
his specimens (which had lost the gonads) the traces of the radial canals
suggested proximal enlargements.
General Distribution: Previous records positively referable to glabrum
+ album are from the vicinity of the Azores (Ranson, 1936) ; and scattered
a'ong the route of the Valdivia in the eastern side of the Atlantic, from the
offing of Morocco to Lat. 42° S., besides one locality record for the southern
Indian Ocean (Vanhoffen, 1902a) ; several in the eastern tropical Pacific
(Bigelow, 1909)11; and off California (Bigelow and Leslie, 1930). With the
Bermuda records extending its known range to the western side of the
North Atlantic, there seems no reason to doubt that the distribution of H.
glabrum will eventually prove to be as wide as that of its better known
relative, H. minimum (p. 122).
Vertical Range: It is not yet possible to make a final statement as to the
vertical distribution of H. glabrum, as here defined, because it is combined
with H. racovitzae and H. conicum in Thiel’s (1935, 1936) discussions of
the Meteor closing-net data. However, the fact that the Valdivia, Albatross,
and Bermudian specimens all came from (open net) hauls from 550-0
meters, or deeper, suggests that it is chiefly bathypelagic in habit, though
perhaps not exclusively so.
Halitrephes.
This genus, closely allied to Halicreas by the structure of the tentacles
and of the sense organs, as well as by the simple circular mouth, without
separate lips, but separated from it by the large number of radial canals,
was first described from the collection made by the Albatross in the eastern
tropical Pacific as H. maasi (Bigelow, 1909, p. 146, PI. 33, Figs. 1-5, 7, 10,
PI. 45, Fig. 13). In the type specimen of the latter, 55 mm. in diameter, 28
canals were given off from the stomach, 5 of them branching dichotomously
so that the number at the bell margin was 33; and there were about 70
tentacles of different sizes. Three years later, Vanhoffen (1912) described
a second species, as H. valdiviae, from the collection of the German South
Polar Expedition, with as many tentacles or even more, but separated from
maasi by having only 16 canals. As the specimens examined by him
ranged in diameter from 18 to 70 mm. — i.e., included full grown individuals
as well as juveniles — the specific distinction seems well founded.
As the genus has not been seen since, the captures listed below from
Bermuda are of interest, for although the specimens are extremely frag-
mentary, their identity seems reasonably certain.
Halitrephes valdiviae Vanhoffen.
Halitrephes valdiviae, Vanhoffen, 1912, p. 384.
Material: Net Nos. 135, 154, 247, 357, 451, 562, 580, 614, 635, 709,
714, 745, 767, 778, 818, 830, 843, 967; hauls from 914-0 to 1,829-0 meters;
19 specimens about 30 to 80 mm. in diameter.
All the specimens are in poor condition. Even when much battered,
11 Thiel’s (1936, p. 36) chart of distribution also includes locality records for racovitzae and
conicum.
126 ' Zoologica: New York Zoological Society [XXIII :5
however, the flat, lenticular form, but with central thickening, combined
with numerous endodermal tentacular roots, (which are so resistant that
they are recognizable even in the most fragmentary specimens), give a
general aspect so characteristic as to make Halitrephes easily recognizable,
even when the canals have been lost. And the two known species differ so
widely in the number of tentacles that in most cases a specimen is referable
either to the one or to the other, with reasonable certainty. All of the
Bermudian specimens show the endodermal roots of some of the tentacles;
and in one example, several tentacles are still sufficiently intact to show the
contrast between stiff distal, and soft proximal, portions, characteristic of
the Halicreidae. It was not possible to make certain of the total number
of tentacles, in any specimen. But the minimum number (specimen of about
55 mm.) was certainly as great as 102. In another of 45 mm., there were at
least 105, while in others of 60-80 mm., 145-170 tentacular roots were
counted, total numbers being evidently somewhat greater. This corresponds
well to the distribution with size recorded by Vanhoffen (1912) of 32
tentacles in a specimen of 12 mm.; 32-64 tentacles at a diameter of 18-20
mm.; 64-96 at a diameter of 30-40 mm.; and 200 at 70 mm. As pointed out
for H. maasi (Bigelow, 1909, p. 146), the tentacular bases are of various
sizes, indicating their successive development. But as no trace of canals is
to be seen in the marginal zone of any of the Bermudian specimens, nothing
can be said as to the order of development of tentacles relative to canals.
In one specimen, a sector of the margin still bears several rhopalia,
alternating with the tentacular roots, as is the case in H. maasi, and agree-
ing closely with the rhopalia of the latter in structure and club-like appear-
ance (Bigelow, 1909, PI. 33, Fig. 10). The only example in which there is
any trace of the gastrovascular system still to be seen, shows the circular
lip characteristic of the genus and its allies, with the bases of about 8, at,
broad canals in about one-half the circumference, suggesting a total of
about sixteen; or the same number described by Vanhoffen (1912) for the
original series of H. valdiviae.
General Distribution : The original collections of H. valdiviae were from
the tropical Atlantic between Lat. 6° N. and Lat. 20° S. ; and fi'om the
Indian Ocean between Lat. 2° N. and Lat. 34° S. The Bermuda captures
are thus the most northerly yet recorded, and the farthest extra-tropical
for the Atlantic. Present indications are that this is a tropical and sub-
tropical species, else it would probably have been picked up in some of the
many deep towings that have been made farther north and east in the
Atlantic.
Vertical Range: All the records for H. valdiviae have so far been from
hauls that worked down to at least 500 meters; in most cases to at least
1,000 meters. Its bathypelagic status thus seems established.
More precise information as to the depth of its chief center of abun-
dance would be of interest, because, while colorless, most of the captures,
like those of Halicreas minimum, have been in nets working so deep as to
indicate that the specimens were from levels inhabited also by the deeply
pigmented groups of medusae, and by the black bathypelagic fishes.
Geryonidae.
Liriope.
Although this sharply defined genus is perhaps the most often encoun-
tered of Trachomedusae, there has been no agreement as to the specific
relationships of its many described representatives, owing to the difficulty
(I believe we may say “impossibility”) of finding any differences, the gaps
between which cannot be bridged by examination of sufficiently large series.
Final decision that all known representatives of Liriope represent but a
1938]
Bigelow: Medusae Taken During 1929 and 1930
127
single variable species was already foreshadowed in Browne’s (1927) inter-
esting exposition of the situation that faces every student of the genus.
Thiel (1936, p. 45) has now definitely made this union of all the named
forms under the oldest designation, L. tetraphylla Chamisso and Eysen-
hardt. And the reader is referred to his discussion for a clear exposition
of the reasons justifying this procedure. Granting, however, that the vari-
ous forms of Liriope cannot be considered “species,” in the sense in which
the term is generally employed in zoological nomenclature, it still remains
an interesting question how they are to be interpreted, and how recorded
in scientific literature. While it is now established that intergrades can be
found with regard to every character which has been suggested as specific,
we have still to recognize that the representatives of different swarms, or
of different populations (seasonal or regional), often do differ so widely in
many respects, as Browne (1927) has vividly described, that they would be
referred without hesitation to different species had not the existence of
intermediates been proved.
Liriope tetraphylla Chamisso and Eysenhardt.
Geryonia tetraphylla, Chamisso and Eysenhardt, 1821, p. 357, PI. 27,
Fig. 2.
For lists of references to the named forms of Liriope, all of which, fol-
lowing Thiel (1936) are now referred to L. tetraphylla, see Mayer,
1910, p. 413-424; also, Thiel, 1936, p. 45.
Material: Net No. 51, Surface, April 29, 1929, 1 specimen about 5 mm.
in diameter.
The gonads in this young specimen are of about the same form as in a
7 mm. specimen, already pictured from the eastern tropical Pacific (Bigelow,
1909, PI. 4, Fig. 3), namely, squarish with rounded angles. This stage has
already been proved preliminary to the triangular or rhomboid outlines
often, but not always, attained by adult L. tetraphylla.
The most interesting thing about this record is that 974 hauls in vari-
ous months from May to October, in two different years, yielded only the
one specimen of a genus so widespread and frequently abundant in warm
seas. Nor can the paucity be credited to the depths at which the hauls were
made, for 108 of them were in the upper 200 meters ; 86 right at the surface.
This, added to the corresponding fact that the Bache hauls in the winter of
1914, yielded only 4 Liriope at 2 stations on lines run between Cape Hat-
teras, Bermuda, the Bahamas, and the Straits of Florida, can only mean
that it was decidedly rare in this side of the Sargasso Sea area during the
periods in question, which included both the warm and the cold months. This
is the more astonishing, when one remembers that the Plankton Expedition
(Maas, 1893) took Liriope regularly from Latitude about 40° N., southward
to Bermuda; and thence eastward all along the track across the tropical
Atlantic.
Geryonia.
Geryonia proboscidalis Forskal.
Medusa proboscidalis, Forskal, 1775, p. 108.
For synonymy, see Bigelow, 1909, p. 116, and Mayer, 1910, p. 425.
Material : Net No. 613, 1,097-0 meters, 1 fragmentary specimen, diam-
eter about 30 mm.
Net No. 621, 1,097-0 meters, 1 fragmentary specimen, diameter about
16 mm.
128 Zoologica: New York Zoological Society [XXIII :5
Net No. 850, 1,463-0 meters, 1 fragmentary specimen, diameter about,
25 mm.
These very fragmentary specimens (identified by the number of
canals) add nothing to previous knowledge of the morphology of this well
known species.
Geryonia had already been reported to the north and to the east of
Bermuda (Maas, 1893) as well as at many other localities in the warm belt
of the Atlantic.
Narcomedusae.
In the classification adopted by Broch (1929), by Ranson (1936), by
Uchida (1928), and by me in earlier papers (Bigelow, 1909; 1913; 1918;
1919), the Narcomedusae that have gastric pockets are divided into two
families, Cuninidae and Aeginidae, depending on whether the pockets are
perradial, equal in number to the tentacles, and not extending out beyond
the points of origin of the latter (Cuninidae), or whether the definitive
pockets are interradial in location, and at least twice as numerous as the
tentacles, with the tentacles arising in the notches between the pockets
(Aeginidae). This scheme is followed here. Thiel (1936, p. 55), on the
contrary, follows Mayer (1910), in referring all Narcomedusae with
peripheral gastric pockets to one family, Aeginidae.
Such of the typical Narcomedusae as lack gastric pockets are now gen-
erally grouped in the family Solmaridae. If the curious sand-dwelling form,
Halammohydra, actually is an aberrant Narcomedusa, as Remane (1927),
Uchida (1928), and Leloup (1935) believe, it calls for a separate family
(Halammohydriidae, Remane).
Cunina.
Cunina sp.?
Material : Net No. 242, 1,646-0 meters, 1 specimen, about 30 mm. in
diameter.
Although .this specimen is very much damaged, enough remains to show
that it had perradial gastric pockets, and a well developed peronial canal
system, a combination of characters locating it in the genus Cunina, as con-
trasted with Solmissus, which lacks canals. Unfortunately, the exumbrella
is so much rubbed or macerated, that it is impossible to tell whether, or not,
otoporpae were present. And all the otocysts have been destroyed. There
are 23 large tentacles and one very small one. Identification of any speci-
men of Cunina with as many as 20 tentacles depends, in the first place,
on decision whether the Medusa mucilaginosa of Chamisso and Eysenhardt
(1821), with 21-24 tentacles, belonged to Cunina, to which Blainville (1834)
referred it, or whether it actually was a Solmissus according to the defini-
tions for these two genera now generally adopted. The name mucilaginosa
was used by Vanhoffen (1908a, as “Solmaris mucilaginosa”) for a series of
Cuninidae of large size (25-60 mm. in diameter), with 21-24 tentacles. But
since his specimens lacked peronial canals (he gave no information as to the
presence or absence of otoporpae), they are more likely to have belonged to
Solmissus than to Cunina.12 And I have argued (Bigelow, 1909, p. 55; 1918,
p. 392) that it was wisest to discard mucilaginosa as a nomen nudem, be-
cause it will never be possible to tell from Chamisso and Eysenhardt’s ac-
count, to which of these two genera their original specimens did actually
belong, lacking information on the critical points — canals and otoporpae.
Ranson (1936, p. 200), however, has recently revived this name, for a
Cunina of 55 mm. with 21 tentacles.
12 See Bigelow, 1919, p. 327, for discussion of Vanhoffen’s treatment of the genus Solmissus.
1938] Bigelow: Medusae Taken During 1929 and 1930 129
Without attempting to settle this question, I may point out that it
mucilaginosa be finally accepted as a Cunina, it was the earliest named mem-
ber of this genus. Hence, if it lies within the varietal range of the species
commonly called rubiginosa, a possibility suggested by Thiel (1936), mu-
cilaginosa must take precedence over rubiginosa, and the latter becomes a
synonym of it.
Solmissus.
? Solmissus incisa Fewkes.
Solmissus incisa, Fewkes, 1886, p. 954, PI. 9; Bigelow, 1909, p. 67, PI.
21, Figs. 1-3, 5; 1913, p. 57; Mayer, 1910, p. 483; Broch, 1929, p.
527; Uchida, 1928, p. 90; Bigelow and Leslie, 1930, p. 560; Ranson,
1936, p. 206.
Solmissus faberi, Haeckel, 1879, p. 350.
Solmissus bleekii, Haeckel, 1879, p. 351.
? Solmaris rliodoloma, Vanhoffen, 1908a, p. 60, PI. 1, Fig. 5. [non
Aequorea rhodoloma Brandt, 1838.]
Material : Net Nos. 599, 628, 632, 680, 700, 713, 735, 753, 765, 799, 848
885, 891; 914-0 to 1,829-0 meters; 13 specimens; those of which measure-
ment is possible, range from about 50 mm. to about 75 mm. in diameter;
also other fragments.
These specimens — all extremely fragmentary — are identifiable as
Solmissus because they show indications of the presence of gastric pockets
in the radii of the tentacles, but no signs of a canal system. In all cases, the
free gastric wall has been destroyed, but persistant bands of subumbral
tissue, around the outer margins of the pockets, and in the septal spaces
between them, show the pockets to have been oval in outline and somewhat
longer than broad. This fact, combined with the large number of tentacles
(see below), warrant tentative reference to S. incisa. And this identifica-
tion is rendered the more probable by the fact that even in their present
state, they agree in thinness of the disc, in softness of the gelatinous sub-
stance, and generally in appearance, with the incisa collected by the Al-
batross in the Pacific, with which I have been able to compare them.
In no case was an exact count of the tentacles possible, but the presence
of at least 23 in one specimen of about 55 mm., and of 25 to 30, in three
others of, roughly, 60 and 75 mm., corresponds with earlier counts, which
have shown a range of from 23-40 in specimens 50 mm. in diameter or
larger. This contrasts with a maximum of 16 tentacles in S. albescens and in
S. marshalli.
All specimens of S. incisa previously seen have lost all trace of the
gonads, nor is the present series any better in this respect. They do, how-
ever, confirm Fewkes’ original illustration of the marginal outline as
straight or nearly so, from peronium to peronium. All the otocysts have
been lost, and even the pads on which these organs stand.
General Distribution : S. incisa was to be expected in the vicinity of
Bermuda, for it has been taken at several stations along the American con-
tinental slope, between the latitudes of Cape Hatteras and of Sable Island,
Nova Scotia, on the one side, and mid-way between Bermuda and the
Azores on the other. Other Atlantic records are off the coasts of France and
of Spain. It has also been taken at a number of localities in the eastern
tropical Pacific; off Monterey, California; between San Francisco and the
Aleutians; in the Bering Sea region; off Kamchatka; and in Japanese
waters.
130
Zoologica : New York Zoological Society
[XXIII :5
Vertical Range : Most of the captures of this species have been from
hauls made with open nets at considerable depths, but the list includes three
captures from the surface, one from the northwestern Atlantic (Fewkes,
1886), the others from the eastern tropical Pacific (Bigelow, 1909).
Aeginidae.
Aegina.
Up to 1910, fourteen supposed “species” referable to Aegina had been
described. Mayer’s (1910) summary made it clear, however, that this list
required drastic reduction. And as the result of subsequent studies, by
various authors, evidence has gradually accumulated to the effect that all
Aeginas yet described represent at most two species, the Ae. rosea and Ae.
citrea of Eschscholtz (Bigelow, 1913; 1919; Broch, 1929; Ranson, 1936);
or perhaps only one as Thiel (1936) has most recently argued. Decision
here rests on whether the difference between the “citrea” type, in which the
margin of each of the eight adradial gastric pockets is more or less indented
in the mid-line, and the “rosea” type in which it is entire, is genetic, or is
simply the result of sex or of age differences, of differing degrees of disten-
tion of the pockets, or of contraction, as Thiel (1936) maintains. The situ-
ation may be summarized as follows (see Thiel, 1936, p. 74, for details) :
Juveniles, up to perhaps 9 mm. in diameter, have undivided pockets, i. e.,
are, as a rule, of the “rosea” type; all very large specimens (35 mm. in
diameter or larger), so far recorded, (of which adequate descriptions were
given) have likewise been of the “rosea” type;13 for example, Haeckel’s
(1879) Ae. rhodina, and the rosea reported by me (1913) from the north-
west Pacific. Among middle-sized individuals, however, of, say, 9-25 mm.
in diameter, both types have been recorded. Otherwise stated, all typical
citrea have fallen within this size range, whereas the much more numerous
records of the rosea type have covered the entire size range so far recorded
for the genus. Furthermore, several specimens of 7-16 mm. in the present
series show intermediate conditions, with some of the adradial pockets
notched, others not, or showing a slightly sinous outline. One point not
previously stressed is that in specimens of the citrea type the gonads usually
(but not always) form two swellings on the oral surface of each primarily
adradial pocket, whereas in the rosea type, the swelling involves the oral
surface of each of the 8 pockets as a whole. In oral view, that is to
say, there are up to 16 very conspicuous swollen areas (sometimes only 14
or 15) in the citrea type, but only 8 in the rosea type. And this difference
is much more conspicuous than is the difference in outline of the pockets,
for even at its maximum, the latter is slight.
The question as to the relationship of the two types cannot be settled
until a microscopic study is made to determine whether we may not be
dealing here with a sexual dimorphism — nor would this be unique among
Narcomedusae, for the gonads of the two sexes of Aeginura differ widely,
in appearance. But the present series does not offer hopeful material for
histological study. It is also possible that the very large representatives of
the rosea type are spent individuals. Nor can the role played in this matter
by contraction be determined without a study of living material.
Meantime it seems wisest to follow Thiel, at least provisionally, and
to class citrea and rosea as forms (of undetermined significance) of one
species. This, however, must be named citrea, not rosea as Thiel has
called it, because the former has page priority in Eschscholtz’s (1829)
original accounts of the two.
13 Ae. alternans (Bigelow, 1909) was probably a large damaged rosea (Bigelow, 1913, p. 59).
1938]
Bigelow: Medusae Taken During 1929 and 1930
131
Aegina citrea Eschscholtz.
Aegina citrea, Eschscholtz, 1829, p. 113, PI. 10, Fig. 3.
It is probable, as explained above, that all the citations to supposed
species of Aegina, listed by Mayer, 1910, p. 451-454, refer to one
variety or another of this species. For a recent list of references,
see Thiel, 1936, p. 73 (“Ae. rosea”).
Material: Net Nos. 263, 292, 294, 296, 488, 564, 566, 576, 627, 649, 657,
666, 686, 692, 693, 708, 745, 762, 779, 800, 809, 857, 874, 882, 885, 971; hauls
from 0 to 1,829-0 meters; 31 specimens, 5-20 mm. in diameter, in various
stages of preservation.
Among the specimens that are well enough preserved to show the
gastric pockets, 9 (ranging in diameter from 5-20 mm.) are definitely of the
rosea type, i.e., the adradial pockets show no signs of additional subdivision.
Two, of 15-16 mm., are as definitely of the citrea type. The most interest-
ing specimens are nine others (7-16 mm. in diameter) which show various
intermediate states as follows :
A. Each adradial pocket shows but one swelling {“rosea” type) ; margins
of some pockets entire {“rosea” type), of others notched {“citrea” type).
B. Some pockets show “citrea” swellings, some “rosea”] margins of all
pockets entire {“rosea” type).
C. Swellings as above; margins of some pockets notched {“citrea” type),
of others entire {“rosea” type).
D. Each pocket with 2 swellings {“citrea”) ; but with margin entire
{“rosea”).
Obviously, these corroborate Thiel’s (1936) view that it is not pos-
sible to segregate two groups (however named) on this basis. However,
we are still faced with the fact that while a great majority of specimens
of the “citrea” type show the normal metamerism, a large proportion of
those of the “rosea” type that have been seen, have had 5 or more tentacles.
This was, indeed, the case with Eschscholtz’s original series of rosea (4
specimens, 3 with 5 tentacles, one with 6 tentacles), and has repeatedly
been reported since; see especially Vanhoffen (1908a) and Uchida (1928).
The present series again yields 3 pentamerous specimens among the 9 of
rosea type just mentioned, the others having 4 tentacles each. A possible
explanation is that variants of this sort are more frequent in one sex than
in the other.
Reduction of the tentacles also occurs, though more rarely; the series
includes one specimen of 9 mm. with only three tentacles. But it is too
much contracted to allow classification, as to type.
While most authors who have written of Aegina have described it as
having only a small number of otocysts, Vanhoffen (1908a) found about 20
per marginal sector in one of the “rosea” type, 16 in one of the “citrea” type.
And while I have not been able to make a complete count of otocysts, even
for a single sector, of any of the Bermuda specimens, the spacing of such
of these organs as are still recognizable, or of their basal pads, is con-
sonant with Vanhoffen’s counts.
General Distribution: Aegina citrea is widespread in the warm belts
of all oceans — this was already known to apply both to the citrea and rosea
forms — also in the Mediterranean. Southward, it had been taken in the
sub- Antarctic ; and northward to the Bering Sea region in the Pacific; to
Lat. 58° N. in the northeastern Atlantic. The present series is, however,
interesting geographically because there was no previous record of Aegina
in the western half of the north Atlantic except close to the coast (see
Thiel, 1936, p. 76, Fig. 15, for a recent chart of this distribution).
132
Zoologica: New York Zoological Society
[XXIII :5
Vertical Range : Ae. citrea has repeatedly been taken at the surface (the
Bermuda series includes 2 surface hauls), but also in open nets lowered to
various levels, and — more instructive — in closing nets from 1,000-700 meters
(Vanhoffen, 1908a, as “Ae. lactea ”) ; about 800 meters (Bigelow, 1909) ;
200-100, 600-400, 700-500, 800-600, and 1,030-830 meters (Thiel, 1936, “Ae.
rosea”). Its bathic range is thus very wide.
Thiel (1935, p. 81) has also pointed out that the specimens taken deep-
est averaged the largest, those taken shoalest, the smallest; evidence that
Aegina, like various other medusae, reproduces chiefly near the surface,
tending to sink, with increasing age. This is corroborated by the Bermudian
collection, the distribution of catches with depth being:
Depth in
Meters.
Number of
Specimens.
0
2
914-0
2
1,097-0
3
1,280-0
4
1,463-0
11
1,646-0
5
1,827-0
4
Aeginura.
It is now generally agreed that all representatives of Aeginura that
have been examined recently, are referable to the wide-ranging, bathypelagic
species, named Ae. grimaldii by Maas (1904) from the collections of the
Prince of Monaco. Thiel (1936), in his historical survey of the genus, goes
still further, in definitely uniting grimaldii, and its more recent synonyms,
with the Aeginura myosura, Cunoctona nausithoe, and C. lanzerotae of
Haeckel (1879) ; hence he employs the latter name (it has page priority over
nausithoe and myosura). This point is discussed in an earlier publication
(Bigelow, 1913, p. 60). I need only repeat here, that (quite apart from
the presence or absence of a peronial canal system), if we are to conclude
that the Aeginura of modern students is identical with these older Haeckel-
ian species we must assume (a) that the otoporpae so clearly shown in
Haeckel’s (1879, PI. 20, Figs. 1-2) illustrations of his two species of “Cunoc-
tona” were imaginary; and (b) that the absence of secondary tentacles and
the small number of otocysts (16 indicated) in his Ae. myosura were the
result of mutilation. As his descriptions were based on preserved (no doubt
alcoholic) material, the second assumption is not unreasonable. But there
is no way to establish the correctness of either of these assumptions; and
it is possible that an Aeginura lacking secondary tentacles and with very
few otocysts, may some day be found. It, therefore, seems to me wiser to
treat Haeckel’s old names as doubtful synonyms of grimaldii, rather than
to select one of them as the name for the species in question.
I need only note farther that if Haeckel’s accounts were to be taken at
face value, grimaldii could not be referred to his genus Aeginura at all,
because he described the latter as having a well-developed canal system. But
as Maas (1905, p. 78) pointed out, what Haeckel (1881, PI. 13, Fig. 7)
pictured as canals were in reality parts of the gastric pouches.
Aeginura grimaldii Maas.
Aeginura grimaldii, Maas, 1904, p. 38, PI. 3, Figs. 19-28; Bigelow, 1909,
p. 80, PI. 9, Fig. 4; 1913, p. 61; Mayer, 1910, p. 470; Kramp, 1913,
1938]
Bigelow: Medusae Taken During 1929 and 1930
133
p. 276; 1914, p. 436; 1920, p. 6; 1924, p. 37; Broch, 1929, p. 533,
Fig. 37; Thiel, 1932a, p. 156; Ranson, 1936, p. 211.
Aeginura weberi, Maas, 1905, p. 77, PI. 11, Fig. 73, PI. 12, Fig. 76, PI.
14, Figs. 90-99.
Cunoctona grimaldi var munda, Vanhoffen, 1908a, p. 53, PI. 2, Fig. 6.
Cunoctona guinensis, Vanhoffen, 1908a, p. 53, PI. 3, Fig. 29.
Cunoctona obscura, Vanhoffen, 1908a, p. 52, PI. 2, Fig. 7, PI. 3, Figs.
25-28, 30.
Aeginura lanzerotae, Thiel, 1936, p. 86.
? Aeginura myosura, Haeckel, 1879, p. 343, PI. 19, Figs. 8-9; 1881, p.
41, PI. 13-14; Mayer, 1910, p. 468.
? Cunoctona lanzerotae, Haeckel, 1879, p.. 318, PI. 20, Figs. 1-6.
? Cunoctona nausithoe, Haeckel, 1879, p. 318.
? Aeginura lanzerotae, Mayer, 1910, p. 469.
Material : Net Nos. 18, 22, 24, 36, 41, 48, 49, 52, 53, 64, 66, 67, 69, 73,
78, 111, 121, 122, 137, 270, 293, 295, 362, 400, 427, 442, 455, 468, 481, 539
562, 582, 605, 606, 612, 620, 621, 627, 628, 631, 632, 635, 638, 639, 645, 646,
652, 654, 665, 680, 681, 685, 698, 700, 702, 708, 712, 715, 719, 720, 721, 724,
726, 732, 740, 742, 745, 746, 753, 759, 760, 761, 778, 779, 781, 793, 794, 804,
805, 810, 811, 812, 842, 849, 852, 861, 885, 894, 959; hauls from 540-0 to
1,829-0 meters; 184 specimens, 1-6 per haul. Also, Net Nos. 26, 42, 65, 76,
110, 143, 295, 360, 447, 488, 598, 716, 929, 949; 19 specimens, probably -
belonging here, but so fragmentary as to preclude positive identification.
In spite of its extent, the collection adds nothing to previous knowledge
of the morphology of this species, for the specimens are all in such poor
condition that it is only the very characteristic appearance of Aeginura —
especially the large tentacle roots, the outlines of gastric pockets, and the
dense pigmentation — that makes them recognizable at all. It was disap-
pointing that the margins are in every case so battered that neither the
secondary tentacles peculiar to the genus, nor the otocysts, are to be seen.
The very large white eggs are, however, so resistant that they are still
visible in specimens from Net Nos. Ill, 122, 468, 606, 726, 761, 794, 810,
and 959. These catches were made in the months of May, June, July,
August, and September, evidence that in this region the breeding season of
Aeginura covers spring, summer, and early autumn— very likely the entire
year.
General Distribution : The wide distribution of previous records (Thiel,
1936, Fig. 18) had already established the cosmopolitan nature of this
medusa in the ocean basins. The present captures are, however, interesting
as the first for the eastern side of the North Atlantic; interesting also as
showing that this is one of the most regularly occurring medusae at suit-
able depths in the Bermuda region, which may be taken as representative
of the Sargasso Sea as a whole.
Vertical Range : Up to the present, there is no positive evidence that
any Aeginura larger than 8-10 mm. has been taken from a depth less than
400-500 meters, for while the great majority of captures have been in open
nets, those taking adults have in every case sampled the water down to at
least this depth — in most cases much deeper. And there is record of closing
net captures of adults from about 780 meters (Maas, 1904), as well as from
1,000-700 meters and deeper (Thiel, 1935, Table 16). Thus, when adult,
Aeginura belongs to the same bathypelagic community as Atolla and Peri-
phylla, as indeed, its dense pigmentation would indicate. The depth distri-
bution of the Bermuda hauls yielding Aeginura, relative to the total number
made at each depth, tabulated below, suggests a rather definite concentra-
tion between, say, 1,000 and 1,600 meters.
134
Zoologica: New York Zoological Society
[XXIII :5
Depth in
Meters.
Total Number
Hauls.
Number Hauls
with Aeginura.
% with
Aeginura.
549-0
23
1
4
732-0
40
2
5
914-0
132
9
7
1,097-0
134
19
13
1,280-0
132
27
19
1,463-0
132
19
14
1,646-0
135
17
13
1,829-0
135
8
6
2,012-0
4
0
0
Thiel (1935), however, has recently made the interesting discovery
that young stages of 1-5 mm. occur in the upper water layers, the Meteor
having taken them on four occasions shoaler than 100 meters, once shoaler
than 50 meters (Thiel, 1936, p. 86). From this, he suggests that the eggs
rise to the surface, and that, as growth proceeds, the young medusae sink
deeper and deeper, as seems commonly to be the case among bathyplanktonic
animals.
SOLMARIDAE.
Peganfha.
? Peganfha elara R. P. Bigelow.
Pegantha clara, R. P. Bigelow, 1909, p. 80; Mayer, 1910, p. 445; H. B.
Bigelow, 1918, p. 397.
Polycolpa forskali, Vanhoffen, 1908a, p. 56; 1912, p. 391; 1912a, p. 32.
[Non P. forskalii Haeckel, 1879; 1881.]
Pegantha smaragdina, H. B. Bigelow, 1909, p. 90, PI. 14, Fig. 1, 2, PI.
19, Figs. 1-9, PI. 22-26.
Material: Net Nos. 50, 157, 271, 720, 812, 882, 902, 959; 914-0 to 1,463-0
meters; 8 fragmentary specimens, about 10-about 20 mm. in diameter.
Although damaged, having lost all trace of exumbral sculpture, and
with the stomach much battered, the specimens all show the peronial canals,
and the outlines of the stomach without any trace of gastric pockets, that
are characteristic of Pegantha. Owing to the obvious contraction of some
specimens and the flattening of others, it is not possible to state the normal
relationship of heighth to breadth: in the pi’esent state they suggest that
at diameters of 10-20 mm., the bell of this species is from *4 to % as high
as broad; perhaps still higher relatively in some cases.
The tentacles range in number from about 20 to about 24; the largest
number being in a specimen of about 12 mm., while the largest specimen
has only 22 tentacles. The small specimens (6 in number, 10-12 mm. in
diameter) without gonads, are referred to this species chiefly because this
is the only Pegantha, yet known, in which upwards of 18 tentacles have
been found before the development of the gonads. This identification is
strengthened, in the case of the 20 mm. specimen, by the fact that traces
are to be seen of simple interradial swellings — forerunners of the gonads —
at the margin of the stomach.
Most of the otocysts have been lost; but the spacing of those still
visible suggests the presence of 3-5 per marginal lappet; or about the num-
ber to be expected in juveniles of P. clara, in adults of which there may be
up to 8 per lappet. In only one specimen are traces of the otoporpae still
to be seen.
1938]
Bigelow: Medusae Taken During 1929 and T930
135
General Distribution: The original record of P. clara was from the
continental slope in the offing of Woods Hole, and young specimens were
taken near the Bahamas by the Baclie (Bigelow, H. B., 1918), so that it
was to be expected at Bermuda. Other records referable to it are from the
vicinity of the Cape Verde Islands (Vanhoffen, 1912, “Polycolpa forskali”) ,
off Chile (Vanhoffen, 1912a, “Polycolpa forskali”) , off Peru, (Bigelow,
1909, “Pegantha smaragdina”) , and from the tropical Pacific between
Hawaiian and Caroline Islands (Vanhoffen, 1912a, “Polycolpa forskali”) .
Vertical Range : Previous records include both the surface, and open net-
hauls from considerable depths.
SCYPHOMEDUSAE.
Carybdeida.
Mayer (1910) grouped all known Carybdeida in a single family — Caryb-
deidae. For convenience, however, — even if for no other reason — it seems
preferable to follow Krumbach (1925) and Uchida (1929) in limiting that
family to forms which lack gastric pouches projecting from the manubrium
into the subumbrella cavity, and in which there are only 8 (adradial)
marginal pockets. The more complexly organized forms with free gastric
pouches and more numerous (16) marginal pockets are divided by Krum-
bach (1925) into 2 families, but (I believe more logically) grouped by
Uchida in one, Chirodropidae. The latter need not be discussed here, as
it is not represented in the Bermuda collection.
Carybdeidae.
Few pelagic coelenterates came to scientific attention earlier than did
the quadritentaculate members of this family, for the well known Mediter-
ranean Carybdea was described by Linnaeus, (1758). Since that time, the
number of named forms multiplied, until Mayer (1910) gave a list of
twenty-one. It was already obvious, however, that the number of actual
species represented was much smaller. And although insufficiency of most
of the early descriptions precluded revision, Mayer (1910) tentatively
classed all of them as varieties of four species at most. The carybdeids have
subsequently been the subject of repeated discussion, notably by Stiasny
(1919, 1926, 1929, 1930, 1934, 1935, 1937), by Uchida (1929), by Thiel
(1928) and by Menon (1930). But views held, as to generic and specific
characters, have been so divergent that a resume is requisite, before the
Bermudian specimens can be discussed even though these belong to forms
known for many years.
Much of the confusion has resulted from uncertainty as to how much
weight should be given (in classification) to the position of the groups of
gastric filaments (phacellae) ; i. e., whether in clusters or horizontal bands
at the 4 interradial corners of the stomach ( Carybdea type), or extending
as vertical bands along the interradial sides of the latter, as described by
Muller (1858), for Tamoya haplonema. Unfortunately, Muller’s illustration
did not show the location of the filaments. And while his account of their
location has repeatedly been quoted, or paraphrased, no first-hand descrip-
tion of this type of phacella appeared subsequently until Stiasny (1919,
p. 40) found the phacellae forming “interradiale vertikale Fadenreihen langs
der Magenseiten,” in Sumatran specimens otherwise agreeing so closely
with Muller’s Tamoya that generic identity was assured. And in 1934
Stiasny (1934) was at last in the position to present a good illustration of
a West African carybdeid, evidently referable to Muller’s T. haplonema, ■
clearly showing this same vertical arrangement, with phacellae extending
along % of the length of the stomach. It is thus proved that the type species
136
Zoologica: New York Zoological Society
[XXIII :5
of Tarnoya differs from the type of Carybdea ( C . marsupialis, Linne) not
only in the larger stomach and in the presence of mesenteries (often used
as the chief generic separation), but equally sharply in the arrangement
of the phacellae.
Uchida (1929, p. 175), it is true, described the phacellae as “numerous
short gastral filaments in the interradial crescentic area of the stomach,”
for a Japanese carybdeid agreeing otherwise with Tarnoya (i.e., in the struc-
ture of the stomach, mesenteries, and canalization), hence, no doubt cor-
rectly, referred by him to that genus. But he did not illustrate the phacellae.
And Stiasny (1930, p. 10) included their arrangement as one of the alterna-
tive characters for the genera, Tarnoya and Carybdea, a course followed
here, as well.
On the other hand, it proves necessary to abandon size and the promi-
nence of the exumbrella sculpture, as generic characters — also included by
Stiasny in his recent synopsis — because the Bermudian collection (p. 144)
verifies Agassiz and Mayer’s (1902) account of the existence of at least one
species with the Carybdea type of stomach (i.e., flat and lacking mesenteries)
and phacellae, but with weak sculpture and growing to a larger size than any
other carybdeid yet seen. Thus, if the generic separation is to be based pri-
marily on structural features — and that this is the more logical procedure
need hardly be defended — we arrive at the following:
1. Stomach flat, without mesenteries: gastric cirri forming brush-like
bundles at the interradial cornel’s of the stomach, or (if expanded)
crescentic areas extending horizontally Carybdea.
2. Stomach deep, connected with the subumbrella by well developed per-
radial mesenteries : gastric cirri in bands, extending vertically along
the walls of the stomach, in the interradii Tarnoya.
This, the reader will observe, is a return to the scheme long ago pro-
posed by Haeckel (1880) ; an instance where the latter’s classic system has
proved more prophetic of knowledge to come than it was justified by facts at
hand at the time.
The family, Carybdeidae (as here defined), also includes one well known
genus, Tripedalia, which shows a multiplication of tentacles, but otherwise
resembles Carybdea and Tarnoya in simplicity of organization.14
Carybdea.
Among the Carybdeas that have yet come to light, one, recently de-
scribed by Stiasny (1930), but unnamed, is set apart by the fact that its
tentacles bear lateral branches. As the tentacles in all other members of the
family — whether single or grouped — are simple, this form certainly deserves
specific recognition. The new species, Carybdea stiasnyi, is therefore pro-
posed for it. Stiasny (1930, p. 5) suggests that “auf Grund dieses einen
Merkmals konnte ein neues Genus aufgestellt werden.” But in this regard,
we may well await the action of its discoverer.
Successive studies by different authors have brought cumulative evi-
dence to support Mayer’s (1910) contention that all other named forms of
Carybdea in reality represent nothing more than varieties or races of a very
small number of good species. But views have differed as to how many such
deserve recognition, largely because of the uncertainty as to which, if any,
of the supposed “Carybdeas,” should actually be referred to its companion
genus, Tarnoya. A survey of recent literature, added to the specimens I have
myself seen, corroborates Mayer’s (1910) view that the great majority of
Carybdeas, the generic identity of which is established by adequate descrip-
14 Haeckel’s (1879-80) problematical genera Procharagma (lacking velum) and Procharybdis
(with velarium, but lacking velar canals) would also fall among the Carybdeidae, should it ever
prove that any existing carybdeids show these primitive states when adult, i. e., that they were
not founded either on multilated specimens, or on juvenile stages.
1938]
Bigelow: Medusae Taken During 1929 and 1930
137
tion or by illustration of the critical characters, fall in one or other of the
following well differentiated groups:
Group A. Of small or moderate size (for the most part less than 50 mm.
high) ; with strong exumbral sculpture; with the gastric cirri in each cluster
arising either from a single basal stalk or from stalks closely clustered at
the four interradial corners of the base of the manubrium. The earliest
named Carybdea, C. marsupialis Linne, of the Mediterranean, was of this
type; so also two other well known forms, C. xaymacana Conant from the
West Indian region, and C. rastonii Haacke of the Indo-Pacific — likewise
C. murrayana Haeckel, from West Africa.
Mayer (1910, p. 508) in his tabular view classes xaymacana as a variety
of marsupialis from which it differs only in somewhat smaller size, in having
fewer and less complexly branched velar canals (4 per quadrant instead of
up to 6 or 8), and phacellae arising from a single primary trunk, instead of
from 8-10 primary branches.
The unity between xaymacana and marsupialis with respect to the pha-
cellae is in fact, even closer than Mayer supposed. On the one hand, Claus
(1878) long ago pointed out that in marsupialis each group of filaments
arises from a single main trunk, while on the other, my own examination
of Bermudian and Bahaman specimens of xaymacana described below (p.
139) shows that the basal trunk divides into several primary branches, each
of which bears a cluster of filaments either simple, or showing further den-
dritic branching (Text-figs. 3, 4, 5), much as described for marsupialis.
Nor are there any differences in the outline of gonads and of pedalia, in the
shape of the rhopalar nitches, or in exumbral sculpture, that can not be
explained as due to contraction, or — at most — to individual variation.
Comparison does, however, substantiate the supposed difference in
canalization, for whereas in the xaymacana that I have seen, as well as in
those described by Conant (1898), by Mayer (1910), and by Stiasny (1919),
there have been at most four (sometimes three, according to Stiasny), un-
branched or simply forked canals per quadrant, there may be 6 or even 8
in marsupialis , some of them complexly subdivided. In canalization, then,
xaymacana corresponds to half-grown marsupialis; as it does in the fact
that the largest so far seen (with large gonads) have been only % to %
as high as mature marsupialis.
Present indications, therefore, are that xaymacana is a dwarf — perhaps
paedogenetic — form of marsupialis ; one, too, with a different geographic dis-
tribution, for the typical ( marsupialis ) form has never been seen in the west-
ern side of the Atlantic, nor the dwarf ( xaymacana ) form in the eastern.
Furthermore, C. murrayana bears much the same relationship to marsupialis
that the latter does to xaymacana, for while it agrees with marsupialis in
sculpture and in the structure of its phacellae, its velar canals are not only
more complexly branched than in typical marsupialis, but more numerous,
i.e., 10-12 per quadrant. It is, indeed, doubtful whether a half-grown mur-
rayana would be distinguishable from a full-grown marsupialis — except per-
haps, by the state of development of the gonads — or a half-grown marsupialis
from a fully-grown xaymacana, though there would be no danger of con-
fusing full-grown specimens.
Forms bearing a relationship of this sort, one to the other, would un-
hesitatingly be named “subspecies” by students of molluscs, of Crustacea,
or of the higher vertebrates. And as they certainly require recognition in
nomenclature, more precise than the vague designation “variety,” it seems
justifiable to introduce trinomials into the classification of pelagic coelen-
terates, also, to cover such cases.
The discussion must also touch on the status of C. rastonii, because if
the latter be, in fact, indistinguishable from xaymacana, as Stiasny (1919)
has suggested, the combined subspecies must be named rastonii Haacke
138
Zoologica: New York Zoological Society
[XXIII :5
(1886), this being the older of the two names; or perhaps even prototypus
or cuboides, if it finally prove that Haeckel’s (1880) problematical Proch-
aragma prototypus and Procharybdis cuboides are young rastonii as Mayer
(1910, p. 509) has classed them.
Apart from geographic distribution, the only distinctions between
xaymacana and rastonii, listed in Mayer’s (1910) tabular view, are slight
differences in shape and relative length of pedalia, and in length of ten-
tacles, which certainly would not justify separation. Furthermore, the canali-
zation of rastonii is intermediate, in type, between that of typical marsupi-
alis, and of the subspecies, xaymacana.
On the other hand, rastonii does differ from xaymacana in the structure
of its phacellae, for, whereas in xaymacana all, or most of the primary
branches arise from a single basal trunk (p. 140), in rastonii they arise inde-
pendently from the gastric wall, in a row, flanked on either side by several
simple cirri, as long ago described by Haacke (1887, PI. 35, Fig. 4), a state
derived (in young medusae) from a row of simple cirri.15 And this differ-
ence, by which specimens of rastonii can be recognized at any stage after
the appearance of the gastric cirri, seems sufficient to warrant its continued
recognition as a distinct species, to emphasize the fact that the Indo-Pacific
representative of this group differs more widely from the Atlantic repre-
sentatives, than do the latter, one from another.
Group B. This group includes much larger forms up to 230 mm. high;
relatively narrower in outline than the marsupialis group; with weak exum-
bral sculpture, if any; with the gastric cirri grouped along crescent-shaped
rows centering at the interradial corners of the base of the gastric cavity.
Recent collections have repeatedly yielded large Carybdea of this type, most
frequently reported as C. alata Reynaud, as C. grandis Agassiz and Mayer,
as C. moseri Mayer, or as one or other of the last two considered as a variety
of the first. According to various accounts the group differs also from the
marsupialis group, in the form of the pedalia, in the relative heighth of the
rhopalia above the bell margin; in the outline of the rhopalar nitches, and
in the more numerous velar canals, as described below (p. 146, 147).
The earliest named form that can be positively referred to this group
(because of the distinctive characters of bell and of phacellae) was C. gran-
dis Agassiz and Mayer (1902) from the tropical Pacific, half-grown indi-
viduals (or a dwarf race) of which were redescribed shortly afterward by
Mayer (1906) as C. moseri. But it seems probable that C. alata Reynaud
(1830) was based on a young Carybdea of this same type, for the original
illustration shows a rounded, non-sculptured bell, nor is there anything in
Reynaud’s illustration or account to argue to the contrary. Consequently,
Vanhoffen (1908) revived the name alata, for Carybdeas from the Indian
Ocean and Straits of Malacca, that agreed with grandis in canalization and
in structure of phacellae. Mayer in his tabular view (1910, p. 508), followed
Vanhoffen in classifying both grandis and moseri as varieties of alata Rey-
naud, as have subsequent authors, generally.16 And while one must admit
that it will never be possible to determine, with absolute certainty, the iden-
tity of the specimen pictured by Reynaud, because no description was given
of the morphological features on which identification primarily depends, to
follow Vanhoffen will tend toward stability of nomenclature. The situation
is, however, confused by the fact that Uchida (1929), while agreeing that
grandis and moseri belong in the synonymy of alata Reynaud, transferred the
latter name to a species of Tamoya, on the supposition that Agassiz and May-
er’s (1902) original figures of grandis referred to a member of that genus,
not to a Carybdea. And this course has been followed by Stiasny (1929;
15 1 have been able to confirm Haacke’s (1887) account of stages in development for small
and middle-sized specimens from the tropical Pacific (Bigelow, 1909), and for a large series from
Japan, in the collection of the Museum of Comparative Zoology.
16 Thiel (1928) has revived the name grandis.
1938]
Bigelow: Medusae Taken During 1929 and 1930
139
1930, 1935, 1937), likewise by Rao (1931). Successive examinations of
specimens of the alata-grandis type have, however, proved that actually they
have the shallow stomach and the arrangement of phacellae characteristic of
Carybdea.
So far as can be judged, from published accounts, and by the specimens
(Atlantic and Pacific) that I have myself examined, the members of the
alata-grandis group, so far named, represent only one species, including C.
madrasputana recently described by Menon (1930) which seems to repre-
sent an immature stage in development. It is even doubtful whether any
of the named forms deserve the term variety.
Group C. Mayer (1900; 1910) described (as C. aurifera ) a juvenile
Carybdea from the Tortugas, differing from C. xaymacana in the facts that
the velar canals next the tentacular radii were broadly palmate, and that the
exumbrella was sprinkled with nematocyst warts, as well as in strong color-
ation. Recently, Stiasny (1926) has described a second species, C. sivickisi,
from the Philippines, in which all the canals are of this palmate form, and
with similarly warty exumbrella. Its rhopalar nitches also differ from those
either of the marsupialis group, or of the alata group, in lacking a well-
developed covering scale. But it is separated from aurifera by the fact that
its exumbrella is strongly sculptured; that it has large gonads though only
slightly larger (10-12 mm. high, as against 7 mm.) ; and in the tentacles
and sensory nitches. Examination of a larger series alone can show how
far these differences can be credited to different stages in growth ; especially
since Mayer’s figures of aurifera are evidently somewhat diagrammatic and
conflicting as regards the phacellae; meantime the two species are best
retained.
Summation of the foregoing yields the following synopsis of Carybdea,
as here recognized:
1. Tentacles simple.
A. Sensory nitches overhung by well developed covering scales;
exumbrella not warty.
a. With prominent exumbral sculpture; sensory nitches with
covering scale above, but entry open below.
(1) . Phacellae dendritic, the filaments primarily arising
from a single trunk
marsupialis (including xaymacana and murrayana) .
(2) . Phacellae more diffuse, of several trunks arranged in a
row rastonii.
b. Exumbrella without prominent sculpture; sensory nitches en-
closed by a pair of scales below, as well as by one above alata.
B. Sensory nitches without well developed covering scale, exumbrella
warty sivickisi {aurifera?) .
2. Tentacles bearing lateral branches stiasnyi.
Carybdea marsupialis xaymacana Conant.
Charybdea xaymacana, Conant, 1897, p. 8, Fig. 8; 1898, p. 4, Figs. 1-16,
31-34, 36-43, 57, 67, 69, 70, 72; Berger, 1898, p. 223; 1900, p. 1, 3
pis.; Mayer, 1904, PI. 7, Fig. 60; 1910, p. 509, PI. 56, Figs. 5-7, PI.
57, Fig. 1; Stiasny, 1919, p. 34.
Tamoya punctata, Fewkes, 1883, p. 84, PI. 1, Figs. 4-6.
Material : Net No. 745, Surface, June 29, 1930, 4 specimens, about 12-23
mm. high.
Net No. 840, Surface, September 3, 1930, 1 specimen, about 11 mm. high.
Dip net, Surface, October 16, 1930, 1 specimen about 15 mm. high.
140
Zoologica: New York Zoological Society
[XXIII :5
Net. No. 935, 1,463-0 meters, September 23, 1930, 1 specimen, about 19
mm. high.
Net No. 977, Surface, May 12, 1931, 14 specimens, about 10-20 mm. high.
Net No. 978, Surface, May 15, 1931, 18 specimens, about 10-18 mm. high.
Net No. 979, Surface, May 16, 1931, 66 specimens, about 10-19 mm. high.
The general features of xaymacana, including exumbral sculpture, need
no comment here, having been well described by Conant (1898) and by
Mayer (1910) ; the phacellae, and rhopalar nitches in more detail by Stiasny
(1919). But notes are pertinent on the range of variation in these respects,
and on the branching of the velar canals, questions bearing on the relation-
ship of this form to typical marsupialis on the one hand, and to rastonii on
the other.
Phacellae : Conant (1898) and Stiasny (1919) both describe the gastric
cirri of each group as arising from a single trunk, as is likewise the case
macana. One of the
groups of phacellae from
a specimen 16 mm. high
(Net No. 745), showing
the typical state in which
the filaments all arise
from a single primary
trunk, x 30.
Text-fig. 4.
Carybde a marsupialis xaymacana. One
of the groups of phacellae from
a specimen 23 mm. high (Net No.
745). x 30.
Text-fig. 5.
Carybdea marsupialis xaymacana.
One of the primary branches,
with terminal filaments, from
the group of phacellae shown
in Text-fig. 3.
1938] Bigelow: Medusae Taken During 1929 and 1930 141
in most of the Bermudian specimens (Text-fig. 3). And while in one of the
clusters in the specimen from Net No. 745 (Text-fig. 4), there is a secondary
cluster of filaments close beside the primary group (the other 3 phacellae are
of the more usual type), the fact that the primary trunk invariably bears
most, if not all of the cirri, is a ready distinction between xaymacana and
juvenile alata of the same size (p. 147).
According to Stiasny (1919, p. 36), the filaments are not branched17;
and they present this same appearance of simplicity when viewed through
the bell wall. But when the groups are dissected out and the filaments
teased apart (Text-figs. 3-5), it is seen that actually, the basal trunk gives
rise to several primary branches, each of which in turn divides, close to the
base, into a group of filaments, much as is the case in the typical marsupi-
alis (p. 137). In one group, 5 such primary branches were detected, with a
cluster of juvenile filaments at one side, apparently forecasting the develop-
ment of a sixth; the total number of filaments being about 40 (Text-fig.
3).
Canalization : Up to the present time, Stiasny (1919) alone has reported
any variation (3) from the number of velar canals per quadrant (4) orig-
inally reported by Conant; even when most deeply forked, the 4 primary
trunks are identifiable as such in all specimens I have seen. However, the
present series amplifies earlier accounts, in showing considerable variation
in the degree and type of branching. Conant’s (1898, Figs. 3-5) original
figures pictured all of the canals, in adult medusae, as forked, either bifid
or trifid. Mayer’s (1910, PI. 56, Fig. 7) illustration, however, shows each of
the four canals in one quadrant as four-branched. And the present series
proves that there may be as many as 5 terminal subdivisions, with indica-
tions even of a sixth (Text-fig. 6). On the other hand, Stiasny (1919)
found most of the canals in Haitian specimens of large size (22-25 mm.
high) to be unbranched, except for a tendency to dichotomous division at
the tips. And the Bahaman specimens listed above are of this same type,
the canals being simple (Text-fig. 7), though the specimens — evidently full-
grown, or nearly so, judging from the large gonads (p. 144.) — are otherwise
indistinguishable from the Bermudian series, and from the specimens orig-
inally described by Conant (1898) from Jamaica. Since both the simple
types of canalization and the more complex have been found in the West
Indian region (and exclusively one or the other at any one locality on a
given occasion), it is likely that we have here to do with a case of what has
been termed “swarm” variation.
The Bermudian series, added to earlier accounts, shows that when the
canals are branched, there is much variation in the degree and type of sub-
division, even from canal to canal in a given individual, though in most (if
not all) cases, it is primarily a dendritic forking, very different from the
lateral lobation such as characterizes Tamoya (p. 154, Text-fig. 22). And
only a very rough correlation — if any — exists between the complexity of
branching and the size of the medusa. Thus, the canals may already be bifid
and trifid, in specimens not more than 10-14 mm. high as exemplified by the
specimen from Net No. 745. One of 19 mm., shows much the same state,
but one of 23 mm. (Text-fig. 6) shows a somewhat more complex branching.
In most cases, the canals next the tentacular radii are somewhat more com-
plexly branched than those next the rhopalar radii (Text-fig. 6) ; occa-
sionally, however, one of the latter may have as many as four well developed
branches.
Pedalia: The only variations, worth mention, from the shape illustrated
by Conant (1898) and by Mayer (1910) are slight individual differences in
the precise outline of the fin-like basal expansion of the inner (oral) side
17 Conant (1898, p. 14) was not explicit on this point.
142
Zoologica: New York Zoological Society
[XXIII :5
T
Carybdea marsupialis xaymacana. Velar canals
of one quadrant of Bahaman specimen
about 23 mm. high, x about 5.
(Text-fig. 8). And these are so slight as to be perhaps explicable as con-
traction phenomena (Text-fig. 8). The pedalia of all I have seen agree very
closely with these structures in an excellent specimen of typical marsupialis
from Naples with which I have been able to compare them; also with rastonii
(Bigelow, 1909, PI. 10, Fig. 1; Uchida, 1929, Fig. 60A). The lengths of the
pedalia, in terms of bell height, in a series of 10 specimens ranging in height
from 10 mm. to 23 mm., are as follows :
Net No.
Bell
Height.
Length
of Pedalia.18
Pedalia, % of
Bell Height.
977
10
6
60
745
12
5
42
977
15
7
47
977
16
8
50
977
17
8
47
977
17
8
47
977
17
8
47
745
19
8
42
935
20
9
45
745
23
10
43
The foregoing shows some slight but irregular decrease in the relative
length of the pedalia with the growth of the medusa; the average of 43%
confirms Mayer’s (1910) report of between % and % the bell heighth.
18 Measured from the base of the fin-like expansion to the origin of the ringed portion of the
tentacle.
1938]
Bigelow: Medusae Taken During 1929 and 1930
143
Text-fig. 8.
Carybdea marsupialis xaymacana.
Pedalia to show variations in
outline. A.t From specimen 14
mm. high from Jamaica; B.,
From Bahaman specimen 23
mm. high.
A B
Text-fig. 9.
Carybdea marsupialis xaymacana. Outlines of two
specimens about 20 mm. high to show variations
in breadth and outlines of gonads. A., From Net
No. 979; B„ From Net No. 935.
Gonads : The feature which shows most obvious variation, among pre-
served specimens of a given size, is the relative length and breadth of the
gonads. In their final state (Conant, 1898, PI. 1, Fig. 1), these extend dis-
tally, nearly to the margin of the bell. None of the Bermudian or Bahaman
specimens are, however, as far advanced as this. In the largest of the for-
mer, the gonads reach only to about the level of the distal ends of the rho-
palia. In specimens 17-18 mm. high, i.e., % grown, the gonads terminate
about as high above the rhopalia as are the latter above the bell margin.
Among the Bermudian series, the gonads in specimens 18-23 mm. high vary
from narrow-linear, to so wide that their free margins are close together
(Text-fig.. 9). Furthermore, the lamellae from which the sex organs arise,
may be either so narrow that the gonads of each pair seem almost in union
(Text-fig. 9A), or so wide that they are separated by a considerable space
144
Zoologica: New York Zoological Society
[XXIII :5
(Text-fig. 9B). This last may mirror, simply, a flaccid condition. But the
variation in length of the gonads, in specimens of equal size, is evidence
that the growth of these organs does not necessarily parallel the growth
rate of the medusa as a whole.
So far as can be determined in surface view, the Bermudian specimens
are all males. Both sexes are, however, represented among the Bahaman
series, the females being recognizable by the large eggs with which the leaf-
like gonads are crowded, and which are easily visible with a hand lens.
Rhopalar Nitches: The present series confirms Stiasny’s (1919) obser-
vation that the outlines of the pits in which the rhopalia are situated, and
of the scales that partially cover the pits, offer specific characters in this
genus. As Conant (1898, PI. 1, Fig. 1) long ago showed (and Stiasny, 1919,
Fig. 2 more recently), the covering scale of xaymacana (Text-fig. 10) has
a simple undulatory outline, and the nitch is wide open toward the bell mar-
gin, i.e., there are no subsidiary scales such as characterize C. alata (p. 146).
This is also true of typical marsupialis, as pictured by Claus (1878, PI. 1,
Figs. 1, 2), and as I have been able to verify on the specimen mentioned
above; also in C. rastonii (Bigelow, 1909, PI. 10, Fig. 1, 4; Uchida, 1929,
Fig. 61).
Text-fig. 10.
Carybdea marsupialis xay-
macana. Sensory pit, with
covering scale, of a Ba-
haman specimen, x about
10.
General Distribution : Previous records of xaymacana, under this name,
are all from the West Indian region; Jamaica (Conant, 1898); Bahamas
(Mayer, 1904; 1910) ; Haiti (Stiasny, 1919). But it is no surprise to find
it near Bermuda, for it is probable that the T. punctata recorded thence by
Fewkes (1883) was actually a young xaymacana.
Vertical Range : The list of captures (see above) confirms previous ex-
perience that the species occurs chiefly close to the surface.
Carybdea alata Reynaud.
Carybdea alata, Reynaud, 1830, p. 95, PI. 33, Fig. 1 ; Haeckel, 1880,
p. 441; Vanhoffen, 1908, p. 34; Mayer, 1910, p. 510; Bigelow, 1918,
p. 400.
Marsupialis alata, Lesson, 1843, p. 276.
Tamoya alata, L. Agassiz, 1862, p. 174; Rao, 1931, p. 27 [partim],
- - -, Semper, 1863, p. 561, PL 39, Figs. 8, 9.
Charybdea pyramis, Haeckel, 1879, PI. 25, Fig. 5-8; 1880, p. 440.
Charybdella alata, Haeckel, 1880, p. 441.
Charybdea grandis, Agassiz and Mayer, 1902, p. 153, PL 6, Figs. 26-31 ;
Bigelow, 1909, p. 19; Thiel, 1928, p. 4.
Charybdea moseri, Mayer, 1906, p. 1135, PL 1, Figs. 2-26.
1938] Bigelow: Medusae Taken During 1929 and 1930 145
Carybdea alata var. pyramis, Mayer, 1910, p. 511.
Carybdea alata var. grandis, Mayer, 1910, p. 511; 1915, p. 171; 1917,
p. 189; Light, 1921, p. 26.
Carybdea alata var. moseri, Mayer, 1910, p. 512; Light, 1914, p. 196;
Stiasny, 1919, p. 37.
Charybdea madraspatana, Menon, 1930, p. 3, PI. 1, Fig. 3.
? , Semper, 1863, p. 561, PI. 39, Fig. 8.
? Charybdella philippina, Haeckel, 1877, No. 404.
? Charybdea philippina, Haeckel, 1880, p. 440.
? Charybusa obeliscus, Haeckel, 1877, No. 406.
? Charybdea obeliscus, Haeckel, 1880, p. 441.
? Charybdea sp? Browne, 1905, p. 157.
Non Tamoya alata, Uchida, 1929, p. 172; Stiasny, 1929, p. 196; 1930,
p. 5; 1935, p. 10; 1937, p. 210.
Material : Net No. 448, 732-0 meters, September 9, 1929, 1 specimen,
about 75 mm. high.
Net No. 459, 732-0 meters, September 11, 1929, 1 specimen, about 20
mm. high.
Net No. 663, 732-0 meters, June 4, 1930, 1 specimen, about 80 mm. high.
Net No. 979, Surface, May 16, 1931, 1 specimen, about 90 mm. high.
Also, 4 specimens, 55-63 mm. high, from Jamaica, in the collection of
the Museum of Comparative Zoology.
For the reasons for the choice of alata Reynaud (1830) as the name for
this species, see p. 138. Apparently, the next record of it was one of Semper’s
(1863, PI. 39, Fig. 9) unnamed carybdeids from the Pacific; while a sec-
ond of his unnamed forms (Semper, 1863, PI. 39, Fig. 8), may have been
a young stage of it, judging from the number (6) of canals in one quad-
rant.11* It is for this same reason that C. obeliscus Haeckel is provisionally
included here in the synonymy of alata. As Mayer (1910) has already de-
cided, this seems certainly to apply to Haeckel’s C. pyramis, for the latter
had not only the number of canals now known to be characteristic of
alata, but also the lunate groups of phacellae, while the relatively great
length of its pedalia is explicable as associated with the small size of Haeck-
el’s specimen (20 mm. high).
So far as I am aware, alata was not seen again until 1902, when Agassiz
and Mayer gave the first adequate description of it (large adult), from the
tropical Pacific, as C. grandis, followed by Mayer’s (1906) account of half-
grown (or dwarf) examples, from the Hawaiian Islands, as C. moseri.
Browne (1905) had meantime reported, from Ceylon, a small Carybdea that
may have been of this parentage, though its velar canals were less numer-
ous. Soon thereafter it was reported from the South Seas (Bigelow, 1909,
p. 20) and from Philippine waters (Mayer, 1915; 1917, “C. alata var.
grandis”) .
Since that time, specimens definitely referable to alata have been re-
ported from Sumatra (Stiasny, 1919) and from the west tropical Pacific
(Mariana Islands, Thiel, 1928) in the one hemisphere, from the Florida-
Bahamas region (Bigelow, 1918) and from unnamed Atlantic localities
(Thiel, 1928) in the other. The figures by Menon (1930) of his C. madras-
patana from India, also agree closely with C. alata, so far as general form
and shape of pedalia are concerned, while the depression of the apex, thought
by him to be distinctive, seems more likely to have been the result of con-
traction, or of distortion. And as the large number (up to 10) of canals
iy This later formed the basis for Haeckel’s (1880) Charybdea 'philippina.
146
Zoologica: New York Zoological Society
[XXIII :5
per quadrant, reported (but not figured) by Menon may likely be derivable
from the primary 6, characteristic of alata, it seems probable that madras-
patana belongs in the synonymy of the latter. Some of the specimens re-
corded by Rao (1931) as Tamoya alata seem also to belong here, because of
the number (6 per quadrant) and simplicity of their canals, whereas others
(because of complexity of the latter) seem to have been true Tamoya.
As no account — other than Reynaud’s (1830) original — of the Atlantic
representatives of this species has ever appeared, details are added here,
as warrant for definitely uniting it with its Indo-Pacific relative. The latter
has been redescribed by Stiasny (1919).
General Form: The Atlantic series agrees with previous accounts in be-
ing not only of large size, and in the absence of any pronounced exumbral
sculpture, but in the fact that the bell is relatively narrower than in any
member of the marsupialis group; measurements are as follows:
Net No.
Approximate
Height in mm.
Approximate
Breadth in mm.
Pedalia, % of
Bell Heighth.
459
20
14
70
663
80
43
56
979
85-90
42
50 (approx.)
Also,
Jamaica
63
35
65
ii
60
35
60
U
60
35
60
u
55
35
64
62%, average
In the Indo-Pacific examples, measured by Agassiz and Mayer (1902),
by Mayer (1906; 1915; 1917), and by Stiasny (1919), the width also aver-
aged 50-60% of the height, contrasting with an average of about 80% in
xaymacana, and in the marsupialis group generally. However, there is no
specific discontinuity in this respect — not, at least, in preserved specimens
— for Mayer (1915; 1917) has reported large alata (as var. grandis ) from
the Philippines, in which the heighth was 73% and 81% of the breadth. It
should be noted, however, that preserved specimens are usually so much
flattened that breadth is exaggerated. And the better preserved of the alata
I have seen all taper toward the apex, which is not normally the case in mar-
supialis and its immediate allies.
In the specimens of alata I have seen, exumbral sculpture has been con-
fined to a faint trace of a circumferential furrow, at the level of the base
of the stomach, and to the prominences in which the rhopalar nitches are
situated; nor has any trace of the longitudinal ridges and furrows, so
prominent in the marsupialis group, been reported by any observer for
alata; neither have nettle warts.
Rhopalia : In the largest two of the Bermudian specimens, the distance
of the rhopalia above the bell margin equals about 12% of the bell heighth;
about 15% in the somewhat smaller Jamaican examples, about 15-20% in
the smallest Bermudian, while in Indo-Pacific specimens, so far measured,
it has ranged from 13-19% (Agassiz and Mayer, 1902; Mayer, 1906; 1915;
1917; Stiasny, 1919).
Rhopalar Nitches : The series confirms Stiasny’s observation that these
differ from the nitches of the marsupialis group, in being partially covered,
on the side toward the bell margin, by scale-like projections of the umbrella
on either hand, in addition to the roof-like scale that overhangs the nitch
1938]
Bigelow: Medusae Taken During 1929 and 1930
147
from above; a difference illustrated by Text-fig. 10, compared with Text-fig.
11 (cf. also Text-fig. 20 of Tamoya liaplonema) .
Text-fig. 11.
Carybdea alata. Sensory pit,
with covering scales, of
specimen from Net No.
979. x about 6.
Gonads : No gonads are visible on the 20 mm. specimen, but as this is
in poor condition, they may have been lost. In the 80 mm. specimen, they
reach about to the level of the rhopalia; to a point about halfway between
the latter and the bell margin in the 90 mm. specimen. Perhaps as the result
of contraction of the bell, they overlap somewhat near their upper ends in
the 80 mm. specimen (Text-fig. 12).
Stomach : As the name alata has also (but incorrectly) been applied to
a Tamoya from the Indo-Pacific, it should be emphasized, as Stiasny (1919,
p. 37) has pointed out, that in Carybdea alata, the stomach is flat and with
very short lips, much as it is in C. marsupialis, and with no trace of mesen-
teries; indeed it is so shown in Agassiz and Mayer’s (1902) and Mayer’s
(1906; 1910; 1915) illustrations.
Phacellae: Previous students have already shown that the numerous
gastric cirri of each group, occupy crescentic areas extending horizontally,
at the corners of the stomach (Text-fig. 12), an arrangement differing
sharply from the dendritic phacellae of the marsupialis group (p. 140) ; but
no detailed account of these structures in alata has yet appeared. In the best
pi’eserved of the Bermudian series, the 4 crescents occupy about % of the
periphery of the stomach (Text-fig. 12) confirming Agassiz and Mayer’s
and Mayer’s illustrations. Stiasny (1919, p. 37) has described the filaments
as “einfach, unverzweigt.” And in surface view, the Bermudian specimens
presented this same appearance. But when one cluster (Text-fig. 13) was
dissected, and its component filaments spread apart it proved that the latter
are the terminal branches of numerous primary trunks, standing in a single
row. In most cases, a primary trunk bears 2-8 such branches (Text-fig. 14).
It is the fact that the end filaments are much longer than the (very short)
primary trunks that gives the deceptive appearance of simplicity. In one
group of phacellae examined, there were approximately 20 primary trunks,
and more than 75 filaments. As Uchida (1929) has expressed doubt, whether
Atlantic specimens earlier recorded by me (Bigelow, 1918, p. 400) as alata
really belonged to this species, I may note that one of these in the collec-
tion of the Museum of Comparative Zoology, has phacellae of this same
type, besides agreeing with the Bermudian series in other respects.
The fact that the small specimen, 20 mm. high, has only 6-8 primary
trunks, in each crescentic cluster, contrasted with the much larger number
in large specimens, gives a measure of the increase in number that takes
place with growth.
Canals: All recent accounts credit this species with 6 velar canals per
quadrant, as is also true of the Bermudian specimens (Text-fig. 15) ; evi-
148
Zoologica: New York Zoological Society
[XXIII :5
Text-fig. 12.
Carybdea alata. Somewhat sche-
matic apical view of specimen
75 mm. high, from Net No.
448, to show the location of
the lunate groups of gastric
filaments.
Text-fig. 13.
Carybdea alata. One cluster of gastric filaments, specimen from Net
No. 448. x about 12.
dently, variants from this state are rare, except that a canal next one of the
tentacular radii may be so deeply cleft that unless it be traced back to its
proximal end, it might appear that the primary number was greater (Text-
fig. 15B). Evidence thus accumulates to the effect that the number of canals
is a dependable specific difference between alata and the marsupialis group
(p. 141). Accounts have, however, differed widely as to the extent to which
the canals branch. On the other hand, Mayer (1906) and Stiasny (1919)
have described them as simple, or at most as showing the commencement of
division. But Agassiz and Mayer (1902), copied by Mayer (1910), picture
them as digitate, with short lateral lobes, while in Vanhoffen’s (1908) two
specimens, they were variously forked — those of the tentacular radii usually
the most deeply- And the two large Bermudian specimens not only show
forking, with similar contrasts between the perradial and the interradial
pairs, and with complexity increasing with growth (cf. Text-fig. 15A with
Text-fig. 15B), but also with wide variation from octant to octant even of
1938]
Bigelow: Medusae Taken During 1929 and 1930
149
Text-fig. 14.
Carybdea (data. One primary branch
with terminal filaments, from the
group of phacellae in Text-fig.
13.
T
T
Text-fig. 15.
Carybdea alata. To show velar canals. A., One quadrant
of velarium of specimen about 75 mm. high from
Net No. 448; B.. One quadrant of specimen about
90 mm. high, from Net No. 979. T., Tentacular
radii; R., Rhopalar radii, x about 3.
150 Zoologica: New York Zoological Society [XXIII :5
a given specimen, in the precise degree of subdivision, and in the lateral
lobing of the canal margins.
Pedalia: By earlier accounts, and in the Bermudian and Jamaican series,
the length of the pedalia (measured along the inner edge) equals about 25-
30% of the bell heighth, in medium-sized specimens of 60-90 mm. In larger
specimens they are relatively shorter: 17% according to Agassiz and Mayer
(1902) at a bell heighth of 230 mm. In young specimens they are relatively
longer: 33% in the Bermudian example of 20 mm.; and shown as about
50% in a young medusa by Agassiz and Mayer (1902).
The fin-like expansions of the pedalia are relatively narrower in the
Bermudian and Jamaican specimens (Text-fig. 16) than they are pictured
by Agassiz and Mayer (1902, PI. 6, Fig. 37) or by Mayer (1906, PI. 1, Fig.
2; 1917, Fig. 3); their outlines less pronouncedly semi-lunar. But this
may well be the result of different states of contraction, between preserved
and fresh material.
Text-fig. 16.
Car>~ybdea alata. Basal part of
tentacle o f specimen
from Net No. 448.
General Distribution: The wide distribution, Atlantic and Pacific, of
locality records for this species (listed above), suggests that it is cosmo-
politan, in the warm belts of the oceans. But it probably does not occur in
the Mediterranean (perhaps barred thence by low water temperatures), for
it could hardly have been overlooked there.
Tamo ya.
If the Bursarius cytherae of Lesson (1830, p. 108, PI. 14, Fig. 1) was,
in fact, congeneric with the form later named Tamoya by F. Muller (1858),
as Haeckel (1880) suggested, it would be necessary to substitute Bursarius
as the generic name. However, as Haeckel pointed out, Lesson’s account and
illustration differ from all known carybdeids in showing only 2 tentacles
(though 4 pedalia) ; nor do they give any indication as to the structural
characters which determine generic position in this group. It is thus wisest
to abandon the name Bursarius for good and all, on the ground that the type
species of the genus is and must always remain unrecognizable. For this
same reason the specific name bursaria, substituted by Haeckel (1880) for
Lesson’s B. cytherae, is likewise best regarded as a nomen nudem.
1938]
Bigeloio: Medusae Taken During 1929 and 1930
151
During the years that have elapsed since the type species of the genus
Tamoya ( T . haplonema ) was described by Muller (1858), carybdeids that
can be referred to this genus as here defined20 have been described from the
two sides of the Atlantic (Haeckel, 1880, as “T. prismatica” ; Mayer, 1910;
Stiasny, 1934) ; from Samoa (Haeckel, 1880) ; from Japan (Uchida, 1929) ;
from the Malay Archipelago (Stiasny, 1919; 1930) ; and from India (Menon,
1930; Rao, 1931) ; showing a world wide distribution within the warm belt.
All Atlantic records undoubtedly belong to T. haplonema. And so far
as can be judged from available information, all Indo-Pacific Tamoyas so
far reported also represent a single species. But it is still an open question
whether or not the Atlantic and Indo-Pacific forms are distinct. If so, the
correct name for the Indo-Pacific species is gargantua, for the earliest
named Indo-Pacific carybdeid that is definitely proved, by the description,
to have been a Tamoya, is the T. gargantua of Haeckel (1880), for which
he revived a name under which Lesson (1830) had long before reported a
large carybdeid from New Guinea. And while neither Lesson’s account, nor
his illustration, gave any indication of the details needed to show whether
his specimen was actually a Tamoya, or a large Caryhdea, stability of nomen-
clature will be served by following Haeckel in this regard. Should the Atlan-
tic and Pacific Tamoyas be finally united, haplonema would become a syno-
nym of gargantua.
Reasons are given above for abandoning altogether the specific name
bursaria which Haeckel (1880) substituted for Lesson’s Bursarius cytherae,
on the assumption that it also was a Tamoya.
Tamoya haplonema Muller.
Tamoya haplonema, Muller, 1858, p. 1, PL 1, 2; L. Agassiz, 1862, p. 174;
Haeckel, 1880, p. 443; Brooks, 1882, p. 138; Von Lendenfeld, 1884,
p. 245; Mayer, 1904, p. 28, PI. 7, Figs. 60-64; 1910, p. 513, PL 57,
Fig. 2; Stiasny, 1934, p. 339.
Tamoya prismatica, Haeckel, 1880, p. 443.
Chary bdea ( Tamoya ) haplonema, Fewkes, 1889, p. 526.
Non Tamoya haplonema, Boone, 1933, p. 39, PL 6.
Material: Surface, June 25, 1929, 1 specimen, heighth about 70 mm.,
breadth about 45-50 mm. (depending on the degree of flattening), length of
pedalia along inner edge, about 22 mm., of fin-like expansion of latter, 15
mm., heighth of rhopalia above bell margin, 13 mm.
Although the Atlantic species of Tamoya has been mentioned repeatedly,
by name, in the literature of medusae, few students have been fortunate
enough to have seen specimens of it. First described by Muller (1858), it
was next reported by Haeckel (1880) whose description (as “T. prismatica ”)
of a West Indian specimen agreed so closely with the original account, both
in arrangement of phacellae, in size of stomach, and in complexity of velar
canals, as to make it certain that he was dealing with haplonema. As the
next reports were by name only (Brooks, 1882; Fewkes, 1889), it is possible
that they may actually have referred to some Cary bdea. And the “T. haplo-
nema” recorded from the Bahamas by Boone (1933, p. 39, PL 6) seem cer-
tainly to have been Carybdea (probably C. xaymacana) , for her illustration
shows a very flat stomach, with brush-like groups of phacellae at its corners.
Mayer (1904; 1910), however, had typical examples of T. haplonema, as
had Stiasny (1934) more recently.
The single Bermudian specimen (considerably crumpled) agrees so
20 Omitting references by name only.
152 Zoologica: New York Zoological Society [XXIII :5
closely in general form with Mayer’s (1910, PI. 57, Fig. 2) picture that illus-
tration is not required.
In the preserved condition, the gelatinous substance is stiffer than in
any other carybdeid I have seen. And the nematocyst warts described by
previous students (see particularly, Stiasny, 1934) are still to be seen thickly
strewn here and there on the exumbrella down to the bell margin, likewise
on the outer faces of the pedalia, though the surface has been so badly
rubbed that nothing can be said as to the regularity of their distribution.
As the size of the stomach is a distinctive feature of the genus, I may note
that this organ (Text-fig. 17) is about as long as shown by Muller (1858,
PI. 1, Fig. 1) and by Mayer (1910, PI. 57, Fig. 2), i.e., with lips hanging
to about the mid-level of the bell; this is somewhat shorter than in Stiasny’s
(1934) West African specimen — a difference perhaps due to the degree of
contraction. The well differentiated, pointed lips are also much as pictured
by Muller and by Stiasny. They are not shown clearly by Mayer (1910, PI.
57, Fig. 2).
Text-fig. 17.
Tamoya haylonerna. Dissection of
upper part of bell of Bermu-
dian specimen, to show out-
lines of stomach, lips, mesen-
teries, and area occupied by
the phacellae. Somewhat sche-
matic.
Mesenteries : Previous statements regarding mesenteries in Tamoya have
been conflicting. According to Muller (1858) and to Haeckel (1880), the
pendulous stomach is bound to the subumbrella by well developed mesenteries
in the radii of the rhopalia, as Uchida (1929) also found for the Pacific
Tamoya. Stiasny (1919, Fig. 6; 1934, Fig. 1), however, pictures the
mesenteries as in the radii of the tentacles; while Mayer (1910, p. 512)
states that the “so-called mesenteries . . . are merely the flattened perradial
sides of the cruciform stomach.”
It is therefore worth mention that the Bermudian specimen confirms
early accounts, in the presence of mesenteries (Text-fig. 17), and that the
latter are in the radii of the rhopalia (perradial). Structurally, they are
thin, transparent plates (apparently ectodermic), their free margin cres-
centic, and by connecting the 4 sides of the stomach with the exumbrella,
they subdivide the upper part of the bell cavity into as many blind pockets.
Phacellae : The gastric cirri agree with previous accounts, in being
grouped in 4 bands extending vertically along the sides of the stomach, alter-
nating with the lips, i.e., in the radii of the tentacles (thus interradial, as
is also the case in Carybdea) . In Stiasny’s (1934) illustration, these bands
are shown extending as far as the distal end of of the dilated portion of
the manubrium. In the present example, however, they reach only about
1938]
Bigelow: Medusae Taken During 1929 and 1930
153
Tamoya haplonema. One of the groups
of gastric filaments of the Ber-
mudian specimen. P., Proximal end.
x about 10.
Text-fig. 19.
Tamoya haplonema. One of the
primary trunks with terminal
filaments, from the group illus-
trated in Text-fig. 18. x about
18.
to the mid-level of the latter (Text-fig. 17). As no detailed account has
previously appeared, I should add that each group consists of a single row
of about 40 primary trunks, arranged (but somewhat irregularly) in a
single row, and each dividing, close to its base into several filaments (Text-
figs. 18, 19).
Rhopalar N itches : Earlier accounts (Muller, 1858, PI. 1, Fig. 1; Stiasny,
1919, p. 39, Fig. 7; 1934, p. 341, Fig. IB; Uchida, 1929, Fig. 81B) had al-
ready shown that the shape of the rhopalar nitches is one of the minor char-
acters separating Tamoya from Carybdea. In T. haplonema (and in its Pa-
cific relative also), the nitches, situated on well marked, ovoid gelatinous
prominences of the exumbrella (Text-figs, 20, 21), are continuously bounded,
on the lower side as well as on the upper, by a rounded covering-scale, which
gives the nitch a pit-like confh'mation, very different in appearance, from
the more open nitches of Carybdea (cf. Text-fig. 20, with Text-figs. 10, 11).
The edges of the marginal pouches, where they enclose the nitches, and
for some distance beyond, are strongly lobate (Text-fig. 20) ; in Carybdea
(Text-fig. 10), they are smooth.
154
Zoologica: New York Zoological Society
[XXIII :5
Text-fig. 20.
Tamoya haplonema. One of
the sensory nitches of
the Bermudian speci-
men, to show covering
scales and lobate outlines
of the marginal pouches,
x about 7.
Text-fig. 21.
Tamoya haplonema. Lateral
view of sensory nitch. x
about 6.
T
lit5 ***
Text-fig. 22.
Tamoya haplonema. One quadrant of velarium of Bermudian
specimen to show velar canals. T., Tentacular radius; R.,
Rhopalar radius.
Canals : The readiest field-mark to distinguish Tamoya from any Caryb-
dea yet described is the lobing of the margins of its complexly forked velar
canals (Text-fig. 22) ; this type of branching was long ago pictured by
Muller (1858) ; more recently by Stiasny (1919, Figs. 8, 9; 1934, Fig. 1, C),
and by Uchida (1929, Fig. 85) .21 Previous authors (Mayer, 1910; Stiasny,
1934) record 10 canals per quadrant for T. haplonema. In the Bermudian
specimen, the number is less regular, there being only 3 primary canals in
one octant (between tentacular and rhopalial radii) ; but so complexly sub-
divided that there are 33 terminal branches, whereas in the adjacent quad-
21 Mayer’s (1910, PI. 57, Fig. 2) picture suggests a somewhat different type of subdivision, but
it is evidently diagrammatic.
1938]
Bigelow: Medusae Taken During 1929 and 1930
155
rant, the number of major trunks is 7 (some of which may have been
derived by subdivision close to the base) with 30 terminal branches.
Pedalia : These (Text-fig. 23) are intermediate in outline between
those figured by Mayer (1910, PI. 57, Fig. 2) and by Stiasny (1934, Fig. la).
According to Mayer’s figure, the tentacular canals are widely dilated at the
distal ends of the pedalia, whereas Stiasny shows them as of uniform bore.
Here, again, the present specimen is intermediate, the canals showing only
slight dilation distally, but with an irregularity of the margins suggesting
that when the pedalia are relaxed, the dilation might be greater.
Text-fig. 23.
Tamoya haplonema. Basal
part of tentacle, x about
2.
Gonads : These organs are somewhat farther advanced in development
than those pictured by Mayer, not only reaching about to the level of the
points of origin of the tentacular canals, but so broad that their free
edges overlap slightly. In their present state, they are folded, but so
irregularly as to suggest the effects of preservation.
General Distribution : Previous records of T. haplonema include Brazil
(the type locality) ; West Indies; east coast of the United States, north to
Latitude about 41° N., and to the offing of the French Congo.
Periphylla.
Periphylla hyacinthina Steenstrup.
For early synonymy, see Mayer, 1910, p. 544, 546 (“P. hyacinthina” and
“P. hyacinthina forma regina ”).
For subsequent discussions, see especially Broch, 1913, p. 4; Stiasny,
1934, p. 342.
Material : Net Nos. 13, 16, 17, 23, 26, 33, 34, 39, 43, 44, 45, 53, 83, 88,
89, 100, 103, 114, 116, 119, 121, 135, 141, 142, 144, 145, 154, 157, 178,
156
Zoologica: New York Zoological Society
[XXIII :5
291, 292, 293, 295, 296, 405, 434, 449, 543, 545, 546, 547, 548, 551, 552,
555, 562, 577, 581, 590, 595, 618, 619, 620, 621, 624, 625, 626, 631, 638,
644, 645, 646, 654, 674, 686, 688, 692, 699, 707, 709, 710, 712, 716, 721,
722, 723, 724, 726, 728, 730, 731, 732, 734, 736, 746, 750, 752, 756, 758,
759, 762, 766, 767, 772, 773, 774, 775, 778, 780, 785, 791, 795, 796,
797, 798, 799, 801, 804, 809, 818, 822, 825, 827, 832, 835, 836, 837, 838,
842, 846, 848, 850, 851, 852, 854, 855, 858, 860, 880, 882, 883, 885,
886, 895, 896, 898, 905, 916, 917, 918, 921, 925, 928, 930, 932, 933, 935,
937, 939, 940, 951, 963, 970; 317 specimens; 1-14 (in most cases, 1-3) per
haul.
Successive studies (see especially Stiasny, 1934) have now definitely
established that the two representatives of this genus most often reported
— dodecabostrycha and hyacinthina — are nothing more than- growth stages,
contraction phases, or color variants of one species; and that this probably
is equally true of the third, regina. Stiasny (1934, 1937), on the basis of the
Discovery collection, has concluded that the three represent successive stages
in growth, the dodecabostrycha form (small, with low rounded stomach and
transparent peripheral zone) being the youngest, the hyacinthina form
(with peripheral entodermal system also pigmented, and usually with high,
pointed stomach), an intermediate stage — usually 75-80 mm. in diameter —
while the regina form represents the largest, usually more than 80 mm. in
diameter, with extensive and dense peripheral pigmentation, but again
usually low-domed, as are the juveniles.
The Bermudian series (many of which are in excellent condition) cor-
roborates this general growth relationship, so far as the dodecabostrycha
and hyacinthina types of pigmentation are concerned, for among 38 small
specimens taken at random, 5-12 mm. in diameter at the level of the coronal
furrow, 37 have the peripheral zone so hyaline that the gonads are entirely
visible, only one being so densely pigmented that they are concealed. Among
37 specimens of 15-20 mm. the gonads are completely visible in only 21;
their upper portions are obscured by the spreading pigment in 7 ; and com-
pletely so obscured ( hyacinthina form) in 9. And among 27 specimens of
25-40 mm., the gonads are entirely obscured in 18; partially so in 8; and
completely visible only in 1. However, still larger specimens occasionally
retain the restricted (“ dodecabostrycha ”) type of pigmentation; this being
true of one Bermudian example of 45 mm.; and of another of 55 mm. The
series does not include any specimens larger than 60 mm., hence, throws
no additional light on the regina form.
It seems that the very smallest specimens (smaller, say, than 8-9
mm. in diameter at the coronal furrow) usually, or always, have the gastric
floor low-domed. It is already more or less definitely conical, however, in
the majority of the Bermudian specimens of 10-12 mm., in several cases
pronouncedly so (extreme hyacinthina type). It has already been suf-
ficiently emphasized, by various students, that medium-sized specimens show
wide variation in this respect, with complete integradation from the lowest
and rounded, to the highest and most acutely conical — depending, I believe,
entirely on phases of contraction.
Broch (1913, p. 7) and Stiasny (1934, p. 356) have both observed that
the apical projection of the aboral gastric wall into the mesoglea, so often
observed in Periphylla and frequently referred to as a “stiel canal,” does
not occur in very small specimens. This the Bermudian series corroborates,
for while the gastric apex is acutely conical in some, even of the smaller
(down to 11 mm. in diameter at the level of the coronal furrow), the smallest
in which the apex shows anything that could be named a canalar elongation
is about 17 mm. in diameter. We can, then, finally conclude that this struc-
1938] Bigelow: Medusae Taken During 1929 and 1930 157
ture has nothing whatever to do with a true stiel canal — i.e., does not indi-
cate derivation from a Scyphistoma stage, but merely represents one ex-
treme in the phases of contraction that govern the contour of the floor of
the stomach in middle-sized individuals.
It is interesting that, in Periphylla (as in Atolla, p. 161), the gonads
appear while the medusa is still not more than 5-8 mm. in diameter at the
level of the coronal furrow22. By the time a diameter of 15-20 mm. is
reached, the sexual organs may be proportionately about as large as they
are in large specimens, the females with large eggs. But we have still to
learn through how extensive a growth-phase any individual specimen may
actually set free its sexual produce.
General Distribution : Enough records have now accumulated to prove
that Periphylla is truly cosmopolitan in seas of appropriate depth and tem-
perature23, Atlantic, Pacific, and Indian, right up to the polar fronts. Having
been already taken in the Straits of Florida (Bigelow, 1918), and to the
southeast of Bermuda (Bigelow, 1928), it was a foregone conclusion that
any campaign of deep towing would yield it in the immediate vicinity of
the latter.
Seasonal and Annual Distribution : The following tabulation of the
percentage of the hauls, at 732-0 meters or deeper, that took Periphylla (the
number taken shoaler was negligible) illustrates the relative frequency of
occurrence, in different months, and as between the two years 1929 and 1930 :
1929
1930
Months.
Total
Hauls
%
Months.
Total
Hauls.
%
Hauls.
with P.
with P.
Hauls
with P.
with P.
April
47
13
29
April
May
75
15
20
May
87
26
30
June
93
9
9
June
82
23
28
July
107
14
13
July
52
23
44
Aug.
51
0
0
Aug.
11
1
9
Sept.
106
3
3
Sept.
136
43
31
Thus it appears that Periphylla — like Atolla (p. 163) — occurred with
much greater frequency in 1930 than in 1929. But the annual difference
in actual abundance was not as great for Periphylla (180 specimens in 343
deep hauls, or 0.5 per haul in 1930; 127 specimens in 462 deep hauls, or 0.3
per haul in 1929) as for Atolla (p. 162). And the data for the two years
agree in showing the greatest frequency in spring and early summer, with
a decided decrease in late summer, followed by some recovery in early
autumn, especially in 1930.
The 95 hauls that yielded juveniles of 15 mm. or smaller, were well
distributed through the season, most numerous and in roughly comparable
numbers (17-23) for May, July and September, evidence that Periphylla,
near Bermuda, reproduces throughout the spring and summer. Probably
it does so throughout the year there, as Kramp (1924) found to be the case
in the Mediterranean.
Vertical Range : No Periphylla were caught in hauls from 366-0 meters
or shoaler. The distribution of the catches in both years combined, in the
deeper hauls, was as follows :
22 'The smallest individuals are in poor condition, but gonads are visible in several not more
than 5 mm. in diameter, though still very small.
23 It is not likely that it will ever be found in the Red Sea, for example.
158
Zoologica: Neiv York Zoological Society
[XXIII :5
Depth
in Meters.
No. of Hauls
Made.
No. of
Specimens.
Av. No.
Specimens
per Haul.
366-0
12
0
0.0
549-0
23
2
0.08
732-0
40
4
0.1
914-0
132
53
0.4
1,097-0
134
85
0.6
1,280-0
132
29
0.2
1,463-0
135
28
0.2
1,646-0
135
60
0.4
1,829-0
135
48
0.4
2,012-0
4
2
0.5
From the foregoing, it appears that in the region in question, Periphylla
regularly lives only deeper than about 600 meters and has its chief abun-
dance considerably deeper (from about 900 meters downward) than in the
Mediterranean, or off the coast of Spain and Portugal where the Thor
(Kramp, 1924) found it most numerous in hauls from about 150-200 to
600-700 meters24, while one Mediterranean catch was made close to the sur-
face. Farther north in the Atlantic, also, the upper limit to its most fre-
quent occuri’ence was found to be between about 250 and 500 meters
(occasional specimens as shoal as 50 meters) along the route of the Michael
Sars Expedition (Broch, 1913) ; in the Sognefjord, Broch (1913) reports
it as numerous as shoal as 150 meters — again with its upper limit a little
above 50 meters. And in west Greenland waters the Ingolf found it about
equally frequently at one level as at another, right up to the surface (Kramp,
1913). In high southern latitudes, also, Periphylla has been taken, not only
in deep hauls, but in the superficial stratum, and at the surface as well, on
several occasions (Browne, 1910; Stiasny, 1934, Table, p. 364, 365). The
combined picture is thus of a species strictly bathypelagic throughout most
of its range, but with the upper limit to its normal occurrence lying at a
progressively higher level, toward high latitudes, until, in subpolar seas,
it comes right up to the surface. Stiasny’s (1934) observations show, too,
that in upwelling regions it may frequently be brought up to the super-
ficial strata, as well as casually elsewhere, as noted above.
So far as is yet known, it is in the Sargasso Sea region (typified by
the vicinity of Bermuda) that the upper boundary to its regular occurrence
lies deepest, at least for the Atlantic. On the basis of these data, it seems
likely that the upper boundary for Periphylla is set by light combined with
temperature. As to the latter, one can at least say that water of 12°-13°
is not unfavorably warm for it, witness its regular occurrence in the deeps
of the Mediterranean (Kramp, 1924). But, so far as I have been able to
learn, it has never been found, more than casually, in temperatures much
higher than that.
Recorded captures do not suggest any definite lower limit; like sundry
siphonophores (Bigelow and Sears, 1937), however — and perhaps the
majority of bathypelagic animals — its numbers decrease as truly abyssal
depths are reached.
Segregation by sizes shows that the upper zone of concentration, near
the 1,000-1,200 meter level, represents the zone of greatest abundance of
juveniles, the depth distribution of specimens of 15 mm., or smaller, being
as follows:
-■i Kramp (1924) estimates the depths of the Thor hauls as equalling about one-half the length
of wire out. But it seems likely, from Jespersen’s (1915) experiments, that the towing depths were
actually somewhat greater, with 2/3 the wire length as perhaps maximal (Bigelow and Sears
1938]
Bigelow: Medusae Taken During 1929 and 1930
159
Depth
in Meters.
Number
Juveniles.
Depth
in Meters.
Number
Juveniles.
366-0
0
1,463-0
6
549-0
4
1,646-0
18
914-0
41
1,829-0
8
1,097-0
44
2,012-0
1
1,280-0
9
2,195-0
1
The largest individuals were even more definitely concentrated in the
deepest strata — responsible in part for the apparent concentration at the
1,400-1,800 meter level — for out of the 16 specimens, of 35 mm. or larger,
7 were from 1,646-0 meters25, 8 from 1,829-0 meters, one only from a
shoaler haul (549-0 meters).
It is especially interesting that in West Greenland waters, Kramp
(1913) found juveniles most abundant at about 600-700 meters26, i.e., but
little shoaler than at Bermuda. This perhaps indicates that Periphylla is not
permanently endemic in the Arctic water of the Greenland-Labrador sea
region, but that the supply there is maintained by the deep indraft of
Atlantic water, a hydrologic phenomenon now well recognized.
Nausithoidae.
Nausifhoe.
Nausithoe punctata Kolliker.
Nausithoe punctata, Kolliker, 1853, p. 323.
For synonymy, see Bigelow, 1909, p. 35; 1928, p. 498; Mayer, 1910, p.
554; Thiel, 1928, p. 25.
Material : Net No. 627, 1,097-0 meters, May 23, 1930, 1 specimen.
Net No. 719, 1,280-0 meters, June 25, 1930, 2 specimens.
Net No. 773, 1,829-0 meters, July 4, 1930, 1 specimen.
Net No. 793, 1,280-0 meters, July 9, 1930, 1 specimen.
Net No. 967, 914-0 meters, September 30, 1930, 1 specimen.
The specimens, ranging from 5 to 9 mm. in diameter, are specifically
recognizable, though fragmentary; one alone (Net No. 793) still shows the
characteristic dark-pigmented ocellus on one of the rhopalia.
Earlier accounts have credited N. punctata with spherical gonads, as
has been true of the specimens of it from the eastern tropical Pacific, from
Japanese waters, from the Mediterranean, and from Florida and the West
Indies that I have previously examined; likewise of the specimen from Net
No. 793 (a ? with eggs of various sizes). A specimen from Net No. 719
and one from Net No. 967 seem at first sight to have horse-shoe shaped
gonads, with concavities directed centripetally. But it appears from more
careful examination that the pockets (globular in life) within which the
sex products are developed, had been torn open on the centripetal sides by
the rough treatment the specimens had received, correspondingly disrupting
the sexual tissue still adhering to their inner walls. That is to say, their
present resemblance to the horse-shoe shaped gonads pictured by Haeckel
(1879, PI. 27), for his genera, Ephyra and Palephyra27, is purely accidental.
—’This tends to corroborate Stiasny’s (1934, p. 363) conclusion that, during its development,
it tends to move slowly into the deeper layers.
26 Recorded at 1,000 meters of wire out.
27 Vanhoffen s (1902, PI. 3, Fig. 10) illustration of Palephyra indica , they are kidney-shaped.
160
Zoologica: New York Zoological Society
[XXIII :5
This being a neritic species, it is likely that the specimens here listed
were the product of the Bermudian Bank. To find N. punctata a few miles
out from the land is no surprise, for it has long been known that the free-
living stage of N. punctata endures long enough for it to drift much greater
distances. It had not been recorded previously from Bermudian waters, but
was to be expected there, for it is common in the West Indies (Mayer,
1910), as well as widespread in warm seas generally. See Mayer (1910,
p. 555), for a summary of the earlier accounts of its developmental stages.
As the attached stage is known to be commensal within certain sponges
of the littoral zone, and as adults have commonly been taken close to the
surface elsew’here, it is somewhat astonishing that the Bermudian speci-
mens were all from deep hauls. But the numbers concerned are so small
that no estimation is possible of the actual depths of capture.
Atollidae.
Atolla.
Atolla wyvillei Haeckel.
Atolla ivyvillei, Haeckel, 1880, p. 488; 1881, p. 113, PI. 29, Figs. 1-9.
For early synonymy, see Mayer, 1910, p. 563, 565, 567 (“A. bairdii”
+ “A. valdiviae” + “A. gigantea” + “A. wyvillei”).
For subsequent discussions of relationships, see especially Broch (1913,
p. 13), Browne (1916, p. 203), Bigelow (1928, p. 505), and Stiasny
(1934, p. 365.).
Material : Net Nos. 23, 26, 29, 36, 39, 41, 53, 62, 66, 74, 78, 100, 102,
105, 111, 118, 122, 142, 154, 196, 221, 240, 247, 253, 292, 293, 295, 296, 368,
434, 539, 544, 545, 546, 552, 553, 556, 561, 562, 563, 565, 568, 574,
575, 582, 588, 596, 597, 598, 599, 601, 605, 606, 611, 618, 619, 625, 626,
631, 632, 633, 637, 639, 647, 648, 652, 657, 659, 666, 674, 676, 681, 682,
685, 687, 688, 691, 698, 701, 707, 719, 720, 721, 727, 739, 741, 745, 748,
767, 770, 774, 777, 778, 781, 787, 789, 798, 799, 800, 801, 804, 806, 808,
810, 812, 824, 826, 829, 831, 838, 846, 850, 860, 861, 863, 865, 875, 881,
882, 885, 892, 917, 926, 928, 937, 938, 943, 964; 549-0 to 1,829-0 meters;
232 specimens, 1-6 per haul, except for Net Nos. 565, 707, and 767, which
captured 10, 8, and 9 specimens respectively. The specimens range from 7 to
100 mm. in diameter.
It now seems so thoroughly established that all known Atollas, that
have smooth marginal lappets, belong to the one species wyvillei, that this
viewpoint needs no further defence (see Kramp, 1924; Bigelow, 1928; and
Stiasny, 1934, for recent pronouncements on this subject). Stiasny (1934,
p. 366) has thought it possible from his study of the extensive Discovery
series to divide the comprehensive species wyvillei into three forms, accord-
ing as the margin of the central disc is indented with broad radial notches
( wyvillei form), or with narrow notches ( verrillii form), or is entirely
smooth ( bairdii form), Conditions, however, in the Bermuda series and
in the others I have studied suggest that these differences (so far studied
only in preserved material), like the relative prominence of the annular
zone, chiefly reflect the state of contraction, or relaxation of the animal
when preserved. But individual variation may also enter into the case, to
a degree that it is not yet possible to estimate. Certainly, we do not have
to do with geographic variation, for the Arcturus and Discovery collections
have already shown that it is not unusual for narrow-notched, broad-notched,
and smooth disced individuals to be taken side by side in the same haul.
In a given individual some of the notches may, furthermore, be broad,
others narrow (Bigelow, 1909, p. 40) ; and it frequently happens that a
1938]
161
Bigelow: Medusae Taken During 1929 and 1930
given individual may be equally well characterized as “broad-notched,
or as “narrow,” there being a complete inter-graduation between the two.
The Bermuda collection now contributes at least one specimen (Net No.
221) about 65 mm. in diameter, in which the margin of the disc is entirely
smooth on one side, but shows clear traces of broad notches on the other.
Recent collections have yielded the smooth and notched or furrowed
forms in varying proportions, the relative counts being 112 notched (ver-
rillii, wyvillei) , and 92 smooth ( bairdii ), for the Discovery series (Stiasny,
1934, p. 370), 12 notched to 3 smooth, for the Thor series (Kramp, 1924),
and 103 smooth (or nearly so) to 57 furrowed, for the Michael Sars collec-
tion (Broch, 1913). Among such of the Bermuda series as are in good
enough condition to be significant in this respect, there are 25 smooth to
38 with the margin of the disc variously notched.
Previous collections had shown that in specimens upwards of, say, 30
mm. in total diameter, i.e., presumably with the final number of tentacles,
there may be as many as 29 or as few as 14 of these; the majority of large
specimens having from 22 to 28 (see Vanhoffen, 1902, Bigelow, 1909, and
Kramp, 1924, for tabulations of tentacle numbers at different sizes).
Eighteen to twenty-two tentacles have already been recorded in small speci-
mens 8-10 mm. in diameter (Vanhoffen, 1902; Kramp, 1924). Among the
smallest of the Bermudian specimens, of 7-12 mm., the tentacle number
ranges from 19 to 24, most frequently 20, the smallest specimen (7 mm.)
having the latter number.
The most interesting contribution by the Bermudian series to knowledge
of Atolla is its confirmation of earlier observations (Maas, 1897; 1904,
Vanhoffen, 1902) that the gonads regularly appear while the medusa is
still very small, for these organs are already of moderate size in 11 out of
14 specimens of 7-9 mm., while all but two out of the 54 specimens of 15 mm.
or smaller, that are in good enough condition for study, have gonads.
And eggs of various sizes are clearly visible in one specimen of only 10
mm., in two of 11 mm., in one of 13 mm. and in one of 14 mm. But it
seems that in occasional specimens these organs may not develop until much
later, for in one of 30 mm., the gonads are still very small. At the other
extreme, Maas (1904) has reported a 72 mm. Atolla as spent. Even the
largest, however, of the Bermudian specimens, from 60-100 mm. in diameter,
still have full, or partly full gonads, both sexes being represented.
In short, it appears that Atolla — like Periphylla — is sexually active
throughout most of the growth period. But nothing is yet known as to
how long, in point of time, this latter may continue.
Broch (1913) has argued, from the depth distribution of the extensive
series collected by the Michael Sars in the North Atlantic, that the wide
variations that occur in the characteristic pigmentation of Atolla are corre-
lated with the depths at which given specimens live, those in which only
the stomach is densely pigmented (Broch’s group I) having been taken most
numerously in hauls from 500 meters, those with more extensive pigmenta-
tion in hauls from 750 and 1,000 meters. Stiasny (1934) likewise found
the Discovery captures at least compatible with this, except in the Antarctic
and in the upwelling waters off West Africa. Unfortunately, the Bermudian
series throws little light on this point, partly because the great majority of
the hauls sampled the water to a depth of 900 meters or more, and partly,
because the great majority of the specimens were so much rubbed in the
nets, that it is impossible to tell how extensive their pigmentation was in
life. The most that can be said is that 14 of the 16 specimens in which the
ring muscle is strongly pigmented (or apparently was so in life), were from
hauls from 1,097-0 meters, or deeper, whereas the genus was of most
frequent occurrence in the 914-0 meter hauls, as noted below. But it is not
safe to estimate the vertical distribution of the contrasting specimens in
162
Zoologica: New York Zoological Society
[XXIII :5
which the pigment is now confined to the stomach (Broch’s group I), because
many of these still show some slight trace of peripheral pigment, as well.
General Distribution: A. wyvillei has already been reported close to
Bermuda (Bigelow, 1928, p. 509), as well as at many other localities wide-
spread in the North Atlantic from side to side; it is in fact now known to
be one of the most cosmopolitan of bathypelagic medusae, to be expected
anywhere in the basins of the open oceans. But like some other bathypelagic
forms, it seems to be barred from access to the Mediterranean (Kramp,
1924; Bigelow and Sears, 1937).
To the northward, it has been taken to Lat. 64° N. in the western side
of the Atlantic, north of the Faroe-Shetland ridge in the eastern (Kramp,
1914; Broch, 1913), to Bering Sea in the Pacific (Bigelow, 1913). To the
southward, its recorded range extends right down to the Antarctic, namely to
Lat. 63° S. in the Indian Ocean (Vanhoffen, 1902) : to 68° S. in the Atlantic
(Stiasny, 1934), to 70° 30' S. in the Pacific (Browne, 1910).
Seasonal Distribution: In 1930, Atolla was taken rather regularly right
through the collecting season, from early May until late September. In
1929, however, when it occurred regularly from April through July, only
one was taken in August, only 6 (in one net haul) in September, although
169 hauls to suitable depths (500-0 meters) were made during those two
months. Similarly, the total number taken was only about % as great in
1929 (54 specimens) as in 1930 (171 specimens) although many more hauls
were made to 900-0 meters or deeper, in the former year (462 hauls) than
in the latter (343 hauls). In our present ignorance of the ecological rela-
tionships of this species, any attempt to explain this annual difference in its
abundance would be pure guesswork. As elsewhere emphasized (Bigelow,
1918, p. 509) winter towing, in the Cape Hatteras-Bermuda-Bahamas tri-
angle, did not yield it at all. But the interpretation of this seasonal perio-
dicity must await a knowledge of whether it passes through a fixed stage
in its development, or whether it is holoplanktonic as seems the more likely
from its bathymetric range.
In 1929, so few very small specimens (of 12 mm. or less) were taken
(3 in April, 1 in May, 1 in July, 1 in August) that no weight can be given
to their seasonal distribution. In 1930, however, there was an unmistakable
concentration of young stages in spring and autumn (24 were taken in May,
6 in June, 11 in September) contrasting with the comparative scarcity in
July (5) and August (1). A combination of the data for the two years
yields the picture of a species breeding at least from mid-spring to early
August, but with a definite slackening of production in the late summer.
Vertical Range: The distribution of the specimens according to the
depth of haul is as follows:
Depth
in Meters.
No. of
Specimens.
No. of
Hauls.
Specimens,
per Haul.
366-0
0
12
0.0
549-0
2
23
0.09
732-0
3
40
0.08
914-0
56
133
0.42
1,097-0
41
134
0.3
1,280-0
38
132
0.3
1,463-0
42
132
0.3
1,646-0
32
135
0.2
1,829-0
14
135
0.1
1,829-0
0
5
0.0
1938]
Bigelow: Medusae Taken During 1929 and 1930
163
This indication o,f an upper limit not far from 500 meters, and of a pre-
ponderance between 1,000 and — perhaps — 1,500 meters, is in line with the
catches of the Michael Sars farther north in the Atlantic (Broch, 1913).
The catches of very small specimens (of 13 mm. or smaller) show even
more definite concentration near the 1,000 meter level, with no indication
whatever of any tendency on the part of the juveniles to live at any higher
level in the water than do the adults ; data are as follows :
Depth in
Meters.
No. of
Specimens.
No. of
Hauls.
Specimens
per Haul.
366-0
0
12
0.0
549-0
0
23
0.0
732-0
1
40
0.02
914-0
27
133
0.20
1,097-0
8
134
0.06
1,280-0
8
132
0.06
1,463-0
5
132
0.04
1,646-0
6
135
0.05
1,829-0
2
135
0.02
There is, however, some indication that as Atollas attain their full size
they tend to sink, for no specimens of 50 mm. or larger were taken as shoal
as 732-0 meters, only one from 914-0 meters, 3 at 1,097-0 meters, 2 at
1,280-0 meters, 7 at 1,463-0 meters, 6 at 1,646-0 meters (a preponderance
at, say, 1,400 meters or deeper), and one at 1,829-0 meters.
Linuchidae.
Linuche.
Linuehe unguiculata Schwartz.
Medusa unguiculata, Schwartz, 1788, p. 195, PI. 6, Fig. 1.
For synonymy, see Mayer, 1910, p. 558 (“L. unguiculata”) , p. 560 (“L.
aquila”) ; Vanhoffen, 1913, p. 429; Bigelow, 1928, p. 510; Stiasny,
1931, p. 30, 1935, p. 13; Darby, 1933, p. 268.
Material: Dip-net, surface, June 8, 1929, hundreds of specimens, about
1.5 mm. in diameter.
Dip-net, surface, July 23, 1929, 1 specimen, 5 mm. in diameter.
Net No. 293, 1,280-0 meters, July 12, 1930, 1 specimen, 6.5 mm. in
diameter.
Net No. 690, 2,012-0 meters, June 9, 1930, 2 specimens, 4.5-11 mm.
in diameter.
Net No. 694, 1,646-0 meters, June 12, 1930, 1 specimen, 5.5 mm. in
diameter.
Reasons are given in an earlier paper (Bigelow, 1928) for including,
under the old name unguiculata, the form aquila, first described by Haeckel
(1880), which has at various times been regarded as a distinct Pacific
species, or as a variety sufficiently well marked to deserve recognition in
nomenclature. And this union has subsequently been accepted bv Stiasnv
(1931). 3
Thiel (1928) has argued that L. draco Haeckel is a distinct species, sepa-
rated from unguiculata by its large size (up to 20 mm. in diameter), by
the arrangement of the subumbral pockets in two series, by the number
164
Zoologica: Neiv York Zoological Society
[XXIII :5
(48) of horse-shoe shaped gonads, and by green color (due to the infesting
zooxanthellae) . It is doubtful, however, whether these distinctions will
stand the test of time. Size, in particular, is hardly likely to prove of
specific significance. At most, a large variety might be indicated — whether
regional, or seasonal — or representing the growth of the members of a
particular brood under specially favorable surroundings. So far as the
number of pockets and their arrangement is concerned, I may point out that
while it is usual in typical unguiculata for these to be in 3 rows, 32 in the
outer, 16 in the median, 4-8 in the inner, an occasional large specimen (12
mm.) may lack the inner row entirely. And the number in the intermediate
row may even be as small as 8, reducing the total number to 40 (Bigelow,
1928, p. 512, Fig. 184). It has also been found in typical unguiculata that,
while the gonads originate in pairs, the members of each pair may, with
growth, either “come together in the perradii, resulting in four horse-shoe
shaped figures” (Bigelow, 1928, p. 513, Fig. 184) ; i.e., attain the “draco”
condition, or may still continue separate until the medusa is well grown
(Bigelow, 1928, p. 513, Fig. 183).
Nor does color seem any more likely to prove an alternative character,
Darby (1933) having recently found that L. unguiculata, at the Tortugas,
contains 2 pigments28, which — depending on the degree of oxidation — give
the subumbrella a yellowish brown color by day (the usual unguiculata
color), but a spotted dark green pattern by night, i.e., the “draco” hue. In
short, it now seems probable that specimens having the “draco” charac-
teristics, represent variants of unguiculata, though we are still entirely in
•the dark, as what conditions — seasonal, environmental, or genetic — favor
their appearance.
General Distribution : Previous records for the general region of Ber-
muda (Fewkes, 1883; Bigelow, 1928) had already made it likely that the
Bermuda Bank is a center of production for Linuche, the farthest extra-
tropical such center in the Atlantic. And this is corroborated by the catch
of juveniles, listed above.
Vertical Range : Linuche commonly swarms at the surface. The few
specimens listed above from deep hauls were probably picked up by the nets
on their way down or up.
DISCOPHORA.
Pelagidae.
Pelagia.
Pelagia noctiluca Forskal.
Medusa noctiluca, Forskal, 1775, p. 109.
For synonymy, see Mayer, 1910, p. 572-576 (“P. noctiluca,” “P. noc-
tiluca var. neglecta,” “P. cyanella,” “P. panopyra,” “P. panopyra
var. placenta,” “P. flaveola,” “P. perla,” “P. phosphora,” “P.
crassa ,”) ; Stiasny, 1914, p. 529, p. 531 (“P. purpuroviolacea” and
“P. rosea”) ; 1924, p. 83 (“P, curagaoensis”) .
Material : Net No. 863, 1,646-0 meters, September 8, 1930, 1 specimen,
about 26 mm. in diameter.
Surface, September 12, 1930, 1 specimen, about 45 mm. in diameter.
The several authors who have recently studied this wide-ranging and
widely varying genus, have successively reduced the number of so-called
species. On the basis of the Arcturus collection, I concluded that one species
of Pelagia (P . noctiluca) , at any rate, “occurs widespread, over the warmer
28 Probably these are actually in the contained zooxanthellae.
1938]
Bigelow: Medusae Taken During 1929 and 1930
165
parts of the Indian and Pacific and Atlantic oceans” (Bigelow, 1928, p. 519)
and doubted whether any of the other named forms could be separated from
it, specifically. Stiasny (1934, p. 388), in his most recent discussion of the
genus, likewise makes equally drastic reduction, writing “there is most prob-
ably only a single species of Pelagia with a world wide distribution over all
tropical and subtropical parts of the ocean.” This species must bear the
name that was earliest given to a medusa recognizable as Pelagia, the P.
noctiluca of Forskal. Menon (1930) has likewise applied this name, without
discussion, to Pelagia from the Indian Ocean.
Subsequently, however, Stiasny (1935) has maintained that flaveola is
also a good species.
Whether any of the other named forms deserve recognition at all, as
geographic or environmental races, or whether they merely represent indi-
vidual or swarm variations, different stages in growth, or contraction
phases, is still an open question, the solution of which is not aided by the
present small series. I can only point out that, in the larger specimen, linear
exumbral warts greatly predominate in the marginal zone, but are inter-
mingled with round-oval warts in nearly equal numbers in the central zone.
That is to say, the so-called perla and pelagia types are combined with the
noctiluca type (narrow linear warts) on different parts of the bell. Menon
(1930) also reports finding both long warts and rounded. This is additional
evidence that specific distinctions cannot be based on the shape of these
structures.
General Distribution: Pelagia was to be expected in the collection, hav-
ing already been found near Bermuda (Fewkes, 1883; Bigelow, 1928). It is,
however, interesting that so large a number of hauls, shoal as well as deep,
made in so many different months, should have yielded so few specimens of a
species which has now been encountered — often in swarms — at many sta-
tions, in the warm belts of all oceans, as well as in the Mediterranean.
Poralia.
? Poralia rufescens Vanhoffen.
Poralia rufescens, Vanhoffen, 1902, p. 41, PI. 4, Fig. 15, 16; Bigelow,
1909, p. 45, PI. 13, Figs. 1-5.
Material : Net Nos. 50, 247, and 726; hauls from 1,646-0, 1,829-0, and
1,463-0 meters ; fragments of the central parts of 3 specimens ; diameters of
base of stomach, respectively about 30, about 35, and about 18 mm.
This species is one of the most interesting finds in the collection, for it
is only the third record for the genus. Poralia was first described by Van-
hoffen (1902) for a fragmentary specimen (marginal zone torn off) with
very simple gonads consisting of outpocketings projecting direct from the
gastric wall into the subumbral cavity, in seven groups of 3 or 4 pockets
each. This type of symmetry, as Vanhoffen pointed out, was previously
known (for Scyphomedusae) only as an abnormality of Aurelia. There were
21 or 22 narrow radial canals.
Two specimens in the Albatross collection (Bigelow, 1909), though like-
wise very fragmentary, allowed some amplification of Vanhoffen’s original
account. In the smaller, there were 8 groups of sexual folds, and 21 canals
in about V2 the circumference, suggesting octoradial arrangement and about
twice as many canals as Vanhfiffen’s example. In the larger and somewhat
better preserved Albatross specimen, about 250 mm. in diameter, the sexual
folds formed a practically continuous ring, surrounding the base of the
stomach, but separated by gelatinous pillars, into 18 or 19 groups of 6 or 7
pockets each. And there were 41 canals. As already remarked (Bigelow,
1909, p. 46), a growth series is here indicated, with multiplication of canals,
166 Zoologica: New York Zoological Society [XXIII :5
and of sex folds, accompanied by a corresponding formation of additional
pillars.
In the largest Bermudian specimen (so far as can be seen in the present
fragmentary state), the sex folds are in five groups, separated by as many
broad gelatinous pillars, corresponding to as many angles at the base of the
manubrium. The number of folds per group is 5, 5, 6, 5(?), 5, while in
most of the groups the larger central folds are flanked in addition by indi-
cations of very small folds, which (if they developed) would raise the num-
ber per group to 6-8. Several of the folds contain large eggs. But this
cannot be taken as proof that the specimen had reached its full growth, for
it is equally possible that the development of sex products may extend over
a considerable period, allowing opportunity for subdivision of the primary
groups of six folds by the development of additional pillars. The number
of pillars is also apparently five in the smallest specimen (stomach, 18 mm.
in diameter) ; the condition of the third specimen is too bad for counting.
The canals are very narrow, as in Vanhoffen’s example of corresponding
size, approximately 38 in number in the smallest specimen, about 40 in the
example 30 mm. in diameter of stomach, but only about 30 in number in
the largest.
These conditions of metamerism in the Bermudian specimens make it
likely, not that Poralia is primarily octoradial as I formerly suggested, but
that it is primarily quadriradial, but that with the growth of the medusa,
the original state is obscured by the formation of additional canals, and by
subdivision of the original groups of sex folds through the development of
additional gelatinous pillars. Different specimens would then show different
numbers of groups of sex folds — as is actually the case — if these develop-
ments take place at different rates, in different radial sectors, as may very
well happen. Decision whether geographical races of the species exist in
this respect, or whether it has been a matter of pure chance that different
numerical, conditions have been illustrated by the few specimns so far seen
from the Indian, Pacific, and Atlantic oceans, must await a study of larger
series.
In the Bermudian, as in the Albatross examples (Bigelow, 1909, PI. 13,
Figs. 1-3), the gastric cirri are arranged in a single fringe-like row, in as
many groups as there are groups of sex folds, there being none on the
gelatinous pillars.
Nothing can be said as to the form of the manubrium, all but the base
having been destroyed.
The marginal zones have been torn off, as was also the case in Van-
hfiffen’s and in the smaller of the Albatross specimens. And while the
larger of the latter showed the distal terminations of the canals, and allowed
a description of the rhopalia, the tentacles of this genus have yet to be
seen.
The Bermudian specimens, like those previously seen, show the bathy-
pelagic type of pigmentation, for they are given a reddish chocolate hue by
pigment granules so large as to be individually visible under a magnification
of only 3 diameters. Even the gelatinous substance is sparsely pigmented
throughout; such of the subumbrella surface as is intact more densely so,
likewise the surface of the sexual folds, though in places where the latter
are torn, allowing the eggs to show through (specimen from Net No. 247),
these are opaque white, as is also the case in Aeginura.
The depths of capture, listed above, added to preexisting records of
1,100-0 meters (Vanhoffen, 1902) and 549-0 meters (Bigelow, 1909), com-
bined with the fact that a form so easily recognizable, and so striking in
appearance has been taken on only 3 occasions, is sufficient proof that its
habitat is exclusively bathypelagic.
1938] Bigelow: Medusae Taken During 1929 and 1930 167
Aurelliidae.
Aurellia.29
The written history of Aurellia extends back nearly two centuries,
during which period many descriptions of its members have appeared, and
discussion of their interrelationships. It has been the object of much inves-
tigation, embryologic, physiologic, and varietal; and it is commonly used
as material for biological teaching. None the less, we still await critical
revision of the seventeen “species” of Aurellia that have been named.
It is, however, generally accepted that most of these represent nothing
more than contraction-phases, growth stages, individual variants, or at most
environmental races of a much smaller number. Thus, Mayer (1910, p.
620, 622), in his tabular view of the genus, grouped all the named forms
as either “aurita and its varieties” or “labiata and its varieties.” Com-
parison, a few years later, between specimens of the aurita group from
various Atlantic and Pacific localities, lead me to the conclusion that they
are not separable into varieties, by any of the characters previously sug-
gested (Bigelow, 1913, p. 98). And Stiasny (1919a; 1922, p. 524, 525),
after a preliminary survey of the genus, concluded that all known Aurellias
fall in one or another of four “wohlunterscheidbare gruppen, nicht Varie-
taten” of the “grossen Species aurita” characterized by differences in the
number of canals arising from each genital sinus.
The difficulty in defining the forms of Aurellia (whatever be the taxo-
nomic rank of these) lies in the fact that the character chiefly involved, in
recent discussions, is one in which differences are gradual, not discontinuous,
namely, the complexity of branching of the canals, and the extent to which
the branches anastomose. The extremes in this respect are far apart. The
one is represented by the simple state illustrated by Mayer (1910, p. 624,
Fig. 397) for a Mediterranean example, in which the main perradial canals
show no branching between the subdivision that takes place close to their
points of origin, and the immediate vicinity of the bell margin, and in which
anastomosis is also confined to the latter region. The other extreme is repre-
sented by the form first reported by Brandt (1838) from Kamchatka, and
subsequently by Uchida (1934) and by me (1913; 1920) from northern
Japan, from the Sea of Okhotsk, and from the Bering Sea-Alaska region, in
which the perradials give off numerous branches throughout the outer % of
their lengths, and in which there is any extensive anastomosis between all
the primary systems, perradial, interradial, and adradial. I have already
argued (Bigelow, 1913; 1920) that this form differs so sharply from typical
aurita in this respect, (also in the dense pigmentation of the margin) as to
deserve specific recognition as A. limbata Brandt. However, it proves that
the further difference reported by Uchida (1934), namely wrinkled exumbral
sense pits in limbata, smooth in aurita, is not a dependable specific char-
acter (unless perhaps for full-grown specimens), for while the pits in a large
example from the Kurile Islands agreed with Uchida’s description in their
wrinkling (Bigelow, 1913, p. 100, PI. 5, Fig. 2), they are smooth in the
smaller specimens with limbata canalization that I have seen. Neither have
I been able to demonstrate, on any of the Aurellias of this type, the partial
closure of the mouth by flaps from the oral arms, reported by Uchida (1934)
and thought by him to suggest rhizostome affinities. This perhaps represents
a contraction phase.
Aurellia limbata, in short, is very closely allied to aurita though differ-
ing enough from the latter (at least in its typical form) for specific recog-
nition. It also has a much more circumscribed range, for all records of
Aurellias, the canals of which have shown the extreme complexity and exten-
29 This name has usually been spelled Aurelia. But Peron and Lesueur’s (1809) original spelling
was Aurellia , as Mayer (1910, p. 619) points out.
168
Zoologica: Neiv York Zoological Society
[XXIII :5
sive anastomosis characteristic of limbata have (so far as I am aware)
been from the Arctic or Subarctic; namely, from northern Japan, Kam-
chatka, Sea of Okhotsk, the Bering Sea region (Brandt, 1838; Uchida,
1934; Bigelow, 1913; 1920), and Greenland (Kramp, 1913, as “A. flavi-
dula”) . Thus it is circumpolar, as I have previously pointed out (Bigelow,
1920, p. 14).
The status of A. maldivensis Bigelow, and of A. solida Browne — the
former recently redescribed by Stiasny (1935), the latter by Menon (1930)
and by Rao (1931) — still remains uncertain.
Aurellia aurita Linne.
Medusa aurita, Linne, 1758, p. 660.
For early synonymy, see “A. aurita,” Mayer, 1910, p. 623, 627, and
“A. labiata,” p. 628.
The more important subsequent discussions are listed above (p. 167).
Material: Net No. 296, 1,829-0 meters, July 12, 1929, 1 specimen, about
85 mm. in diameter.
Dip Net, Surface, October 16, 1930, 10 specimens, about 30-80 mm. in
diameter.
Net No. 880, 914-0 meters, September 12, 1930, 1 specimen, frag-
mentary, about 85 mm. in diameter.
Also, 9 other Bermuda specimens, 25-90 mm. in diameter in the col-
lection of the Museum of Comparative Zoology, collected in the summers
of 1902, 1903, 1916, 1917.
The number of canal roots, arising from each genital sinus, shows the
following range, in specimens of different sizes:
Diameter
in mm.
Number of
Specimens.
Number of
Canals per
Sinus.
25
2
3
30
1
3
33
1
3-4
35
1
3
37
1
3
40
2
3
45
1
3
50
1
3-4
52
1
3
55
1
1-3
75
2
3-5
80
2
3-5
90
1
4-5
Thus, most of the smaller examples would fall in the “aurita” group, ac-
cording to Stiasny’s (1922) subdivision, while the larger specimens bridge
the gap between the latter and his “colpota” group. The first branching of
the primary interradial canals is trichotomous, as Stiasny (1922) has em-
phasized, with further branching leading to the formation of what Mayer
(1910, p. 624) has aptly named a “pitchfork-shaped system.” In small
Aurellias, the first branching of the interradial canals usually lies some dis-
tance distal to the boundary of each genital sinus. It is, in fact, probable
that all Aurellias pass through this “aurita” stage in this respect, when
1938]
Bigelow: Medusae Taken During 1929 and 1930
169
they are small. The outer boundaries of the sinuses spread centrifugally
with growth. And it appears that the number of separate trunks that arise
from a given sinus, in larger individuals, depends chiefly in how far out-
ward the canal-system has become incorporated into the sinus, by the
growth of the latter. With Aurellia proverbially variable and subject to
abnormality, it is not astonishing that the number of canals often differs
for the different sinuses of a given individual.
The Bermudian series gives evidence, in the foregoing tabulation, of the
tendency toward an irregular increase in the number of canals, with growth,
that is to be expected, on this basis. Failure of any of the specimens30 to
show a larger number of canal-roots than 5 per sinus is no doubt due to
the fact that none of them was of large size. For example, the numbers per
sinus in a large Cuban specimen, 250 mm. in diameter, are 6, 6, 8, 7.
I may note, in passing, that anastomosis of canals in the Bermudian
specimens, is much as it has frequently been described for A. aurita.
General Distribution: It has long been known that A. aurita is one of
the most abundant of medusae among the West Indies, and northward from
Florida along the east coast of North America. It is also a familiar object
at Bermuda, although no printed record of its presence seems to have ap-
peared. Being strictly a neritic form, the specimens in the present collec-
tion were no doubt the product of the Bermuda Bank.
Faunistic Considerations.
Neritic and Holoplanktonic Communities.
It is safe to conclude — from the geographic location of the station
where the towing was done — that all the medusae taken, of species that
pass through a fixed stage in their development, were the products of Ber-
muda. Hence, a comparison of this neritic element, in the catches, with the
holoplanktonic species gives a good indication of the extent to which a
small isolated Bank may be expected to contribute to the medusan fauna
of the open ocean nearby.
In the case of the Hydromedusae, out of a total of 717 specimens, the
number belonging to species that are either known to pass through an at-
tached hydroid stage, or may reasonably be assumed to do so,31 does not
exceed 17. This, I may point out, does not include the small series of
Bougainvillia niobe, which has a budding phase, hence is placed in the holo-
planktonic category, so far as its dispersal is concerned. That is to say,
the contribution made by the neighboring slopes and shallows of Bermuda,
to a point not ten miles distant, was less than 3% of the total catch. And
the numerical paucity of the neritic element is the more striking, when one
recalls that it represents at least 9 species, the holoplanktonic element
(Tracho- and Narcomedusae) only 14. It is probable that if the station
had been located but a few miles further offshore, the neritic element in the
medusa fauna would have been negligible, as Thiel (1935) has shown it to
be for the Central and South Atlantic as a whole.
In the case of the Scyphomedusae, the situation was the reverse, if
judged from the standpoint of total numbers only, for the number of neritic
specimens was raised considerably above that of the holoplanktonic by one
catch of several hundreds of juveniles of Linuclie, a genus which almost
certainly passes through a scyphistoma stage, though not yet actually proven
to do so. And a second large catch (66) was also made of Carxjbdea
xaymacana, which also probably has a polyp stage in coastal waters. How-
so Except for two abnormal tripartite specimens, each about 80 mm. in diameter, in which
the number of canals per sinus were 9, 7, 9, and 9, 6, 7.
31 Zancleopsis, Pandea, Heterotiara, Calycopsis, Chromatonema, Aequorea, Olindias.
170
Zoologica: New York Zoological Society
[XXIII :5
ever, if the neritic and holoplanktonic groups of Scyphomedusae be judged,
not by numbers of specimens, but by the frequency of occurrence, the latter
group ranks far in advance, for 252 hauls yielded representatives of genera
certainly or probably holoplanktonic ( Atolla , Periphylla, Pelagia), while
only 16 hauls and surface collections yielded genera which may be safely
called neritic ( Carybdea , Tamoya, Nausithoe, Linuche, Aurellia) . And
apart from the two species just mentioned, neritic (or probably neritic)
Scyphomedusae numbered only 23 specimens, holoplanktonic Scyphomedusae
549.
Apart, then, from sporadic contributions of 2 species, produced on the
Bermuda Bank, the regular population was as characteristically holoplank-
tonic for the Scyphomedusae as for the Hydromedusae. Why it is that the
young of Linuche and Carybdea drift offshore more frequently, or in greater
numbers, than do other medusae produced on the Bermuda Bank, is a
problem for the future.
Vertical Distribution of Medusae as a Whole.
If we omit the few specimens, for which depth data are not available;
fragments which may not have represented individual medusae; and also
the swarm of juvenile Linuche, mentioned above, the distribution of all the
medusae combined with respect to depth of haul, was as follows :
Depth in
Meters.
Number of
Hauls Made.
Number of
Medusae Taken.
Specimens
per Haul.
0
86
126
1.5
183-0
10
0
0
366-0
12
0
0
549-0
23
10
0.43
732-0
40
18
0.45
914-0
132
189
1.44
1,097-0
134
246
1.83
1,280-0
132
245
1.85
1,463-0
132
235
1.79
1,646-0
135
234
1.73
1,829-0
135
220
1.63
2,012-0
4
21
5.25
The fact that the yields of 85 hauls at depths of 183-0 to 732-0 meters
was so trifling, is good evidence that catches made in the deeper hauls can
have owed very little, if anything, to what the nets picked up while being
lowered and hauled up again. Consequently, the tabulation indicates that,
numerically, the immediate surface, on the one hand, and a stratum from
about the 900 meter level downward, were about equally populated, but that
the intervening zone, some 700 meters thick, was practically barren. And
omission of the neritic specimens from the surface catches, shows that this
barren zone extended right up to the surface, for only 4 specimens of the
holoplanktonic group were taken at the surface, namely, 1 Liriope, 1 Hal-
icreas, and 2 Aegina.
Consideration of the comparative regularity of occurrence at different
depths leads to a similar conclusion, as appears from the following tabu-
lation :
1938]
Bigelow: Medusae Taken During 1929 and 1930
171
Depth in
Meters.
Number of
Hauls Made.
Number with
Medusae.
% with
Medusae.
0
86
10 (3)
11 (-3)
183-0
10
0
0
366-0
12
0
0
546-0
23
4
17
732-0
40
12
30
914-0
132
55
41
1,097-0
134
67
50
1,280-0
132
66
50
1,463-0
132
64
49
1,646-0
135
71
52
1,829-0
135
61
45
2,012-0
4
4
100
By this criterion, for all medusae combined, and even including the
swarms of juvenile Linuche, the medusae occurred less regularly at the sur-
face than from about 600-0 meters downward, not at all in the intervening
strata. And omission of the neritic forms lowers the surface frequency to
a very small fraction of that of the deep layers, for only 3 out of the 86
surface hauls yielded holoplanktonic medusae (about 3%), or 1 for 28
hauls.
If the foregoing tabulations were taken at face value, they would sug-
gest that medusae were about equally abundant and equally frequent from
about 800 meters, or thereabouts, downward, with a decided maximum as
deep as 2,000 meters. But consideration of the method by which the hauls
were made shows that such a conclusion would almost certainly be incor-
rect, because the catches from the deepest hauls were no doubt augmented,
both in frequency and in numerical yield, by what the nets picked up as
they were lowered through the comparatively rich 500-800 meter stratum,
and then hauled up through it again. The data are not of a sort to allow
numerical calculation of the probable contamination resulting from this.
But it may be safely assumed that with depth, below, say, 800 meters, this
increased significantly and rapidly. And it is not unlikely that all the speci-
mens yielded by the 4 hauls at 2,012-0 meters actually came from much
smaller depths. In short, the data yields no positive evidence that there
were any medusae as deep as this.
Reasoning then from the collections actually made at Dr. Beebe’s sta-
tion, it seems safe to conclude for the part of the Sargasso Sea represented
by the Bermuda region:
1. That the neritic element among the Hydromedusae is numerically
negligible; and that this also applies to the Scyphomedusae, except on occa-
sions when swarms of one species or another may drift out for a few miles
from the Bermuda Bank.
2. That the whole upper 500 meters of water is practically barren of
medusae from spring, through summer, to autumn, except when such events
take place. On these occasions, the immediate surface may be the most
thickly populated zone. But it appears that these periods are of brief dura-
tion; nor is it likely that they aifect the situation on a broad scale more
than a few miles out from Bermuda.
3. That medusae as a whole in this part of the ocean basin are usually
most abundant and occur most regularly at about 500-800 meters, below
which they decrease, in both these respects, with increasing depth.
172
[XXIII :5
Zoologica: New York Zoological Society
Relative Abundance of Different Species.
The relative abundance of the several species is shown in the following
table :
Species.
Number.
Species.
Number.
Zancleopsis dichotoma
1
Liriope tetraphylla
1
Bougainvillia niobe
7
Geryonia proboscidalis
3
Pandea conica
1
Cunina sp?
1
Pandea rubra
4
Solmissus incisa
13
Heterotiara anonyma
2
Aegina citrea
31
Calycopsis sp?
2
Aeginura grimaldii
201
Chromatonema rubrurn
2
? Pegantha clara
7
Aequorea floridana
2
Carybdea xaymacana
105
Olindias tenuis
3
Carybdea alata
3
Rhopalonema velatum
39
Tamoya haplonema
1
Pantachogon haeckeli
65
Periphylla hyacinthina
317
Colobonema typicum
47
Nausithoe punctata
6
? Crossota brunnea
50
Atolla wyvillei
232
Trachynemidae?
51
Linuche unguiculata
many
Halicreas minimum
186
Pelagia noctiluca
2
Halier eas glabrum
66
Poralia rufescens
3
Halitrephes valdivii
19
Aurellia aurita
12
Absolute Abundance.
Some estimate of the absolute abundance has proved interesting. Since
the neritic forms may be regarded as sporadic invaders of the offshore
waters, they may be left out of account in this connection. The number
of holoplanktonic medusae of all kinds taken in the hauls from 549-0
meters and deeper, i.e., in the most productive zone, was about 1,406;
the number of hauls 869; the number of hours of towing was 3,276; or
an average yield of only 1.6 specimens per tow, or 0.42 specimen per
hour’s towing. Divided into groups, the catch was approximately 0.23
Trachomedusae and Narcomedusae combined, and 0.16 of the bathypelagic
Scyphomedusae ( Atolla and Periphylla ) per towing hour.
Quantitative Comparison with Other Regions.
A quantitative comparison of the medusa population of the Bermuda
region with that of the Mediterranean and of the Central and South Atlantic
is interesting, as an indication of how the former — and its part of the Sar-
gasso Sea as a whole — ranks as to productivity for this group, at different
levels in the water, compared with other seas. Calculation of the Mediter-
ranean catches made by the Thor as listed by Kramp (1924), but omitting
one large catch of “many thousands” of Liriope eurybia in the Sea of Mar-
mora ( Thor Sta. 175, Kramp, 1924, p. 32), shows an average catch per
hour’s towing of about 41.5 medusae of all sorts, of about 36-37 Tracho-
medusae and Narcomedusae combined, and about 2.5 Scyphomedusae. Since
the Thor collections were made with a net with a mouth area of 3 square
meters, towing at a rate of about 2 sea miles an hour, the Bermudian collec-
tion with one of only about 0.78 square meters, towed at a rate of 2-2%
knots, a factor of about 3.8 must be introduced to make the catches com-
parable. So adjusted, it is obvious that the Mediterranean was many times
the more productive region, both for the holoplanktonic Hydromedusae, and
for the medusae as a whole. In the case of the bathypelagic Schyphomed-
usae, however — represented at Bermuda chiefly by Atolla and Periphylla,
173
1938] Bigelow: Medusae Taken During 1929 and 1930
but in the Mediterranean by the latter only — the two regions were about
equally rich.
The Bermudian region is also poor in the holoplanktonic Hydromedusae,
in general, as compared to the Central and South Atlantic as a whole, where
the Meteor tows, with nets only half as large in diameter, yielded an aver-
age of about 5.4 specimens per 200 meters of towing at all depths and sta-
tions combined — a number that must be multiplied by a factor of perhaps
1.8 to correct for the size of the net alone, to be made comparable with
the Bermuda average catch of only 0.23 per hour’s towing. And the Atlantic
figure would be much increased if the vertical Meteor tows were to be cal-
culated on an hourly basis.
Reference to the foregoing discussion of vertical distribution (p. 170)
makes it clear that this relative poverty of the Bermuda region results
chiefly from the barrenness there of the upper 500 meters of water, in holo-
planktonic medusae of any sort. In the Mediterranean, by contrast, most of
the hauls from the upper 100 meters of water yielded considerable numbers
of medusae, of one species or another (see Kramp’s, 1924 lists), while most
of the larger catches of the species most plentiful there, namely Rhopalon-
ema velatum and Liriope, were from depths no greater than 300 meters. In
the Central and South Atlantic also, the Meteor found Tracho- and Narco-
medusae most abundant (10-11 individuals per 200 meters tow) between
the surface and 200 meters, much less so in the deep underlying waters
(Thiel, 1935, p. 43, Fig. 20, Table 3). And evidence in the same direction,
in the Pacific, results from the rich catches of medusae of this category
made in tWe Humboldt Current off northwestern South America, by the
Albatross (Bigelow, 1909).
From the foregoing, it appears that the very warm superficial stratum
of the Sargasso Sea is notably barren of medusae of the holoplanktonic
category, but so far as the evidence goes, it suggests that the productivity
of the deep waters in the more strictly bathypelagic types, represented by
Periphylla and Atolla, is much more nearly of the same general order of
magnitude there, as it is over the ocean basins as a whole32, or in the Medi-
terranean.
It is interesting as illustrating the relative numerical abundance of
two allied groups of pelagic coelenterates that the average Mediterranean
catch of about 41 medusae per hour’s towing mentioned above (p. 172), con-
trasts with a corresponding average of between 200 and 300 calycophorid
siphonophores, both in the Mediterranean and in the neighboring parts of
the Atlantic (Bigelow and Sears, 1937, p. 137). Evidently, the latter group
is much the more numerous in that particular region. But it is doubtful
how far regionally this generalization would apply, for the average Meteor
catches per 200 meters of towing, of siphonophores and of Tracho- plus
Narcomedusae, for the Central and South Atlantic did not differ greatly
one from the other (cf. Leloup and Hentschel, 1935, p. 24, Fig. 17 with
Thiel, 1935, p. 43, Fig. 20, Table 3).
In the Bermuda region, we have no direct comparison between the two
groups, as there are no published records of siphonophores for the years
1929-1930, and as the numbers captured are not recorded for the one year
on record (1931, Totton, 1936). We may, however, point out that they
were only taken in 55 out of 374 hauls (usually only one species in a haul),
or 14% of the tows in 1931, whereas the medusae were taken at 41% of
all hauls in the summers of 1929-1930. Hence assuming that the several
summers were comparable, it appears that medusae are actually more fre-
quently encountered in the Bermuda region than siphonophores. And this
area may thus prove to be as barren of the latter as the former.
.. ]:C,om?a,re’ for example, the catches of Atolla and Periphylla made in the North Atlantic bv
the Michael Sars (Broch, 1913).
174
Zoologica: New York Zoological Society
[XXIII :5
Bibliography.
Agassiz, Alexander.
1865 North American Acalephae. Mem. Mus. Comp. Zool., Vol. 1, No. 2, pp.
1-234, 360 figs.
Agassiz, A., and Mayer, A. G.
1902 Reports on the Scientific Results of the Expedition to the Tropical
Pacific, 1899-1900. III. Medusae. Mem. Mus. Comp. Zool., Vol. 26,
pp. 139-176, 14 pis.
Agassiz, Louis
1860 Contributions to the Natural History of the United States. Vol. 3,
11+301+26 pp., 19 pis. Boston.
1862 Contributions to the Natural History of the United States. Vol. 4,
8+380+12 pp., Pis. 20-35. Boston.
Beebe, Wlliam
1931 Bermuda Oceanographic Expeditions 1929-1930. Zoologica, New York
Zool. Soc., Vol. 13, Nos. 1-2, pp. 1-36.
Berger, E. W.
1898 The Histological Structure of the Eyes of Cubomedusae. J. Comp.
Neur., Vol. 8, pp. 223-230, 5 text-figs.
1900 Physiology and Histology of the Cubomedusae. Mem. Biol. Lab., Johns
Hopkins Univ., Vol. 4, No. 4, 84 pp., 3 pis.
Bigelow, H. B.
1909 Reports on the Scientific Results of the Expedition to the Eastern
Tropical Pacific, 1904-1905. XVI. The Medusae. Mem. Mus. Comp.
Zool., Vol. 37, 243 pp., 48 pis.
1913 Medusae and Siphonophorae collected by the U. S. Fisheries Steamer
“Albatross” in the Northwestern Pacific, 1906. Proc. V. S. Nat. Mus.,
Vol. 44, pp. 1-119, Pis. 1-6.
1918 Some Medusae and Siphonophorae from the Western Atlantic. Bull.
Mus. Comp. Zool., Vol. 62, pp. 365-442, 8 pis.
1919 Hydromedusae, Siphonophores and Ctenophores of the “Albatross”
Philippine Expedition. U. S. Nat. Mus. Bull. 100, Vol. 1, Pt. 5, pp. 279-
362, Pis. 39-43.
1920 Medusae and Ctenophora. Rept. Canadian Arctic Exped., 1913-1918.
Vol. 8, Pt. H, 22 pp., 2 pis.
1928 Scyphomedusae from the Arcturus Oceanographic Expedition. Zoo-
logica, New York Zool. Soc., Vol. 8, pp. 495-524.
Bigelow, H. B., and Leslie, Maurine.
1930 Reconnaissance of the waters and plankton of Monterey Bay, July
1928. Bull. Mus. Comp. Zool., Vol. 70, p. 429-581.
Bigelow, H. B., and Sears, Mary.
1937 Siphonophorae. Rept. Danish Oceangr. Exped., 1908-1910, to the
Mediterranean and adjacent seas, Vol. 2, Biol., Pt. 2, H. 2, 144 pp., 83
figs.
Bigelow, R. P.
1909 A New Narcomedusa from the North Atlantic. Biol. Bull., Vol. 16, pp.
80-82.
Blainville, H. M. D. de.
1834 Manuel d’Actinologie ou de Zoophytologie. Paris, 8+688+12 pp., Atlas
of 100 pis.
Boone, Lee.
1933 Coelenterata. Sci. Results of Cruises of the Yachts “Eagle” and “Ara,”
1921-1928 . . . Bull. Vanderbilt Marine Mus., Huntington, N. Y., Vol.
4, pp. 27-68, Pis. 1-24.
Brandt, J. F.
1838 Ausfiihrliche Beschreibung der von C. H. Mertens auf seiner Weltum-
segelung beobachteten Schirmquallen . . . Mem. Acad. Imp. Sci. St.
Petersburg. Ser. 6, Vol. 4, pp. 237-411, Pis. 1-31.
1938]
Bigelow: Medusae Taken During 1929 and 1930
175
Broch, Hj.
1913 Scyphomedusae from the “Michael Sars” North Atlantic Expedition
1910. Rept. Sci. Res. of the “ Michael Sars” North Atlantic Deep Sea
Expedition 1910 . . . Vol. 3, Pt. 1, Zool., 23 pp. 1 pi., 12 text-figs.
1924 Hydroida, in Kiikenthal und Krumbach, Handbuch der Zool., Vol. 1,
pp. 422-458.
1924 Trachylina, Ibid, pp. 459-484.
1929 Craspedote Medusen. Pt. 2. Trachylinen (Traehymedusen und Nar-
comedusen). Nordisches Plankton, Zool. Vol. 6: Coelenterata, pp. 481-
540, Figs. 1-40.
Brooks, W. K.
1882 List of Medusae found at Beaufort, N. C., during the summers of
1880 and 1881. Studies Biol. Lab., Johns Hopkins Univ., Vol. 2, pp.
135-146.
1895 The Sensory Clubs or Cordyli of Laodice. Journ. Morph., Vol. 10, pp.
287-304, PI. 17.
Browne, E. E.
1905 Report on the Medusae (Hydromedusae, Scyphomedusae, and Cteno-
phora) collected by Professor Herdman at Ceylon in 1902. Suppl. Rep.
XXVII, Ceylon Pearl Oyster Fisheries, publ. Roy. Soc. London, Pt.
IV. pp. 131-166, 4 pis.
1906 Biscayan Plankton. Part IX. The Medusae. Trans. Linn. Soc., London,
Vol. 10, pp. 163-187, Pl. 13.
1910 Coelenterata. V. Medusae. Nat. Antarctic Exped., Nat. Hist., Vol. 5,
62 pp. 7 pis.
1916 The Percy Sladen Trust Expedition to the Indian Ocean 1905 . . .
IV. Medusae from the Indian Ocean. Trans. Linn. Soc., London, Vol. 17,
pp. 169-209, PI. 39.
1927 Report on the Medusae. Cambridge Exped. to the Suez Canal, 1924.
Trans. Zool. Soc., London, Vol. 22, pp. 105-115.
Chamisso, A. de, and Eysenhardt, C. G.
1821 De animalibus quibusdam e Classe Vermium Linneana . . . Nova
Acta, Fasc. 2. Acad. Caes. Leop., Vol. 10, pp. 343-374, Pis. 24-33.
Claus, C.
1878 Ueber Charybdea marsupialis. Arb. Zool. Inst. Wien, Vol. 1, pp. 1-56
or (221) - (276) , 5 pis.
CONANT, F. S.
1897 Notes on the Cubomedusae. Johns Hopkins Univ. Circ., Vol. 17, No. 132,
p. 8-10, Figs. 8-9.
1898 The Cubomedusae. Mem. Biol. Lab., Johns Hopkins Univ., Vol. 4, No.
I, 61 pp., 8 pis.
Darby, H. H.
1933 Studies on Pigments of Linuche unguiculata (Eschscholtz) . Carnegie
Inst. Washington, Year Book, No. 32, p. 268.
Delle Chiaje, S.
1841 Descrizione e notomia degli animali invertebrati della Sicilia citeriore
. . . Vol. 4, 142 pp., Atlas (to Vols. 1-8), 181 pis., Naples.
Eschscholtz, Fr.
1829 System der Acalephen . . . 6+190 pp., 16 pis., Berlin.
Fewkes, J. W.
1882 On the Acalephae of the East Coast of New England. Bull. Mus. Comp.
Zool., Vol. 9, pp. 291-310, 1 pl.
1883 On a few Medusae from the Bermudas. Bull. Mus. Comp. Zool., Vol.
II, pp. 79-90, 1 pl.
1886 Report on the Medusae collected by the U. S. F. C. Steamer Albatross,
in the region of the Gulf Stream, in 1883-’84. Rept. U. S. Fish. Comm.
for 1884, pp. 927-977, 10 pis.
1889 Report on the Medusae collected by the U. S. Fish Commission Steam-
er Albatross in the region of the Gulf Stream, in 1885-’86. Rept.
U. S. Commissioner of Fish and Fisheries for 1886, pp. 513-534, 1 pl.
176
Zoologica: New York Zoological Society
[XXIII :5
Foerster, R. E.
1923 The Hydromedusae of the West Coast of North America, with Special
Reference to the Vancouver Island Region. Contr. Canadian Biol., N. S.,
Vol. 1, No. 12, pp. 219-277, 5 pis.
Forskal, Petrus.
1775 Descriptiones animalium . . . quae in itinere Orientali observavit . . .
Post mortem auctoris edidit Carsten Niebuhr. 19+1 + 34+164 pp.
Hauniae.
Gegenbaur, C.
1856 Versuch eines Systemes der Medusen . . . Zeit Wiss. Zool., Vol. 8, pp.
202-273, Pis. 7-10.
Haacke, Wilhelm.
1886 liber die Ontogenie der Cubomedusen. Zool. Anz., Vol. 9, pp. 554-555.
1887 Die Scyphomedusen des St. Vincent Golfes. Jena Zeitschr. Naturwiss.
Vol. 20, pp. 588-638, Pis. 35-37.
Haeckel, Ernst.
1877 Prodrom. System. Medus. (Cited from Haeckel, 1880).
1879 Das System der Medusen. Theil 1. System der Craspedoten. Jena
Denkschr. Med.-Naturwiss. Gesell., Vol. 1, 25+360 pp., Atlas of 40 pis.
1880 System der Acraspeden. Zweite halfte des Systems der Medusen. Jena
Denkschr. med.-naturwiss. Gesell., Vol. 1, Pt. 2, pp. 361-672.
1881 Report on the deep sea Medusae . . . Rept. Sci. Res. H. M. S. Chal-
lenger. Zool. Vol. 4, 105+154 pp., 32 pis.
Hargitt, C. W.
1905 Notes on the Variations in Rhegmatodes. Biol. Bull., Vol. 9, pp. 368-
377.
Hartlaub, C.
1907 Craspedote Medusen. 1 Teil, 1 Leif.: Familie I, II. Codoniden und
Cladonemiden Nordisches Plankton, Zool., Vol. 6, Coelenterata, pp.
1-135, 126 text-figs.
1914 Craspedote Medusen. 1 Teil, 3 Leif.: Familie IV. Tiaridae. Nordisches
Plankton, Zool., Vol. 6 Coelenterata, pp. 237-364.
Jespersen, P.
1915 Sternoptychidae ( Argryopelecus and Sternoptyx). Rep. Danish Ocean.
Exp. 1908-1910 to the Mediterranean and Adjacent Seas, No. 3, Vol. 2,
{Biology) , A. 2, pp. 1-33, 6 text-figs.
Kolliker, Albert.
1853 liber Scheibenquallen. Zeit. Wiss. Zool., Vol. 4, pp. 320-329.
Kramp, P. L.
1913 Medusae collected by the “Tjalfe” Expedition. Vidensk. Meddl. Dansk
naturh, foren. Kjobenhavn, Vol. 65, pp. 257-286, 4 text-figs.
1914 Meduser og Siphonophorer Medd. om Gronland, Pt. 23, pp. 381-456.
1919 Medusae. Part 1. Leptomedusae. Danish Ingolf-Exped., Vol. 5, No. 8,
111 pp., 5 pis.
1920 List of Medusae collected by the M. S. “Armauer Hansen” in the
North Atlantic in 1913. Bergens Mus. Aarbog, 1917-1918, No. 8, 7 pp.
1920a Anthomedusae and Leptomedusae. Rept. Sci. Res. “Michael Sars”
North Atlantic Deep Sea Exped., Vol. 3, Pt. 2, 13 pp., 1 pi., 6 text-figs.
1924 Medusae. Rep. Dan. Ocean. Exp. 1908-1910 to the Mediterranean and
Adjacent Seas, No. 8, Vol. 2, Biol. H. 1, 67 pp., 40 text-figs., 12 charts.
1926 Medusae. II. Anthomedusae. Danish Ingolf-Exped., Vol. 5, Pt. 10, 102
pp., 2 pis,
1928 Papers from Dr. Th. Mortensen’s Pacific Exped. 1914-1916. XLIII.
Hydromedusae. I. Anthomedusae. Vidensk. Meddl. Dansk Naturhist.
Foren. i Kjobenhavn, Vol. 85, pp. 27-64.
1933 Craspedote Medusen. Teil III. Leptomedusen. Nordisches Plankton.
Zool., Vol. 6: Coelenterata, pp. 541-602, Figs. 1-68.
1938]
Bigelow: Medusae Taken During 1929 and 1930
177
Kramp, P. L. and Damas, D. .
1925 Les Meduses de la Norvege. Introduction et Partie speciale I. Vidensk.
Meddl. Dansk Naturhist. Foren. i Kjobenhavn, Vol. 80, pp. 217-323,
PI. 35.
Krumbach, T.
1925 Scyphozoa, in Kiikenthal und Krumbach, Handbuch der Zool., Vol. 1,
Pt. 4, pp. 522-608.
Leavitt, B. B.
1935 A Quantitative Study of the Vertical Distribution of the Larger
Zooplankton in Deep Water. Biol. Bull., Vol. 68, pp. 115-130.
Leloup, E.
1935 Halammohydra et les Siphonophores Chondrophorides. Bull. Mus. roy.
hist. nat. Belg., Vol. 11, No. 23, 4 pp.
Leloup, E., and Hentschel, E.
1935 Die Verbreitung der Calycophoren Siphonophoren im Sudatlantischen
ozean. RY'ss. Ergeb. Deutsclien Atlantischen Exped. auf dem . . .
“Meteor,” 1925-1927, Vol. 12, Pt. 2, pp. 1-31.
Lesson, R. P.
1830 Voyage autour du monde sur la Corvette de sa majeste, La Coquille
. . . 1822, 1823, 1824, et 1825 . . . Zool., Vol. 2, Pt. 2, Div. 2, Zoophytes,
35 pp. and Atlas, Zoophytes, 16 pis.
1843 Histoire naturelle des Zoophytes. Acalephes. Paris, 8+596 pp. Atlas of
12 pis.
Light, S. F.
1914 Some Philippine Scyphomedusae . . . Philippine Jour. Sci., Vol. 9, Sect.
D, pp. 195-231.
1921 Further notes on Philippine Scyphyomedusan Jellyfishes. Philippine
Jour. Sci., Vol. 18, pp. 25-47.
Linne, Carl von
1758 Systema Naturae . . . Ed. 10, Vol. 1, 823 pp. Holmiae.
Maas, Otto.
1893 Die craspedoten Medusen. Ergebn. Plankton Exped. . . . Vol. 2, K. c.,
107 pp., 6 pis., 2 charts.
1897 Reports on an Exploration off the West Coasts of Mexico, etc. XXL Die
Medusen. Mem. Mus. Comp. Zool., Vol. 23, 92 pp., 15 pis.
1904 Meduses provenant des campagnes des Yachts Hirondelle et Princesse
Alice (1886-1903). Res. camp. sci. . . . par Albert ler, Prince souverain
de Monaco. Vol. 28, 71 pp., 6 pis.
1905 Die craspedoten Medusen der Siboga Expedition. Siboga Exped.
Monogr. 10, 84 pp., 14 pis.
1906 Medusen. Res. Voyage S. Y. Belgica en 1897-1898-1899 . . ., Zoologie,
pp. 1-30, 3 pis.
Mayer, A. G.
1894 Cruise of the Steam Yacht “Wild Duck” in the Bahamas . . . III. An
Account of some Medusae obtained in the Bahamas. Bull. Mus. Comp.
Zool., Vol. 25, pp. 235-242, 3 pis.
1900 Some Medusae from the Tortugas, Florida. Bull. Mus. Comp. Zool.,
Vol. 37, pp. 11-82, 44 pis.
1904 Medusae of the Bahamas. Mem. Nat. Sci., Brooklyn Inst. Arts and
Sci., Vol. 1, 33 pp., 7 pis.
1906 Medusae of the Hawaiian Islands collected by the Steamer Albatross
in 1902. Bull. U. S. Fish Comm., Vol. 23, Pt. 3, pp. 1131-1143, 3 pis.
1910 Medusae of the world. 3 vols. Carnegie Inst., Washington, Publ. No.
109.
1915 Medusae of the Philippines and of Torres Straits. Papers from Dept.
Mar. Biol., Vol. 8, Publ. 212, Carnegie Inst., Washington, pp. 157-202,
3 pis., 7 text-figs.
1917 Report upon the Scyphomedusae collected by the U. S. Bureau of
Fisheries Steamer “Albatross” in the Philippine Islands and Malay
Archipelago. Bull. U. S. Nat. Mus., 100, Vol. 1, Pt. 3, pp. 175-233.
178
Zoologica: New York Zoological Society
[XXIII :5
Menon, M. G. K.
1930 The Scyphomedusae of Madras and the neighboring coast. Bull. Madras
Govt. Mas., N. S. Nat. Hist, sect., Vol. 3, No. 1, 28 pp. 3 pis.
Muller, F.
1858 Zwei neue Quallen von Santa Catharina (Brasilien). Abhandl. natf.
Ges. Halle, Vol. 5, pp. 1-12, 3 pis.
1861 Polypen und Quallen von Santa Catharina. Olindias sambaquiensis n.
sp. Archiv. fur Naturges. Jahrg. 27, pp. 312-319, Fig. 9.
Neppi, V. and Stiasny, G.
1913 Die Hydromedusen des Golfes von Trieste. Arb. Zool. Inst. Wien, Vol.
20, pp. 1-70 or (23) -(92), 4 pis.
PeRON, F. and Lesueur, C. A.
1809 Histoire generate et particuliere de tous les animaux qui composent la
famille des Meduses. Ann. Mas. Hist. Nat. Paris, Vol. 14, pp. 218-228.
Quoy, J. R. C. and Gaimard, J. P.
1827 Observations Zoologiques faites a bord de V Astrolabe en mai 1826 dans
le detroit de Gibraltar. Ann. Sci. Nat., Vol. 10, pp. 172-193, Pis. 4-6.
Ranson, G.
1936 Meduses provenant des Campagnes du Prince Albert ler de Monaco.
Res. camp. sci. Monaco . . . Vol. 92, 245 pp., 2 pis.
Rao, H. S.
1931 Notes on Scyphomedusae in the Indian Museum. Rec. Indian Mus. Cal-
cutta, Vol. 33, Pt. 1, pp. 25-62, pis. 3-4.
Remane, A.
1927 Halammohydra, ein eigenartiges Hydrozoon der Nord- und Ostsee.
Zeit. f. Wiss. Biol., Abt. A, Zeit f. Morphol. u. Ukol. der Tiere, Vol.
7, pp. 643-677, 21 text-figs.
Reynaud, M.
1830 In Lesson, R. P. Centurie zoologique . . . Paris 10, 244 pp., 80 pis.
Semper, Carl.
1863 Riesebericht. Zeit. f. Wiss. Zool., Vol. 13, pp. 558-570, Pis. 38-39.
Stiasny, G.
1914 Zwei neue Pelagien aus der Adria. Zool. Anz., Vol. 44, pp. 529-533.
1919 Die Scyphomedusen-Sammlung des naturhistorischen Reichsmuseums
in Leiden. I. Die Carybdeiden (Cubomedusen) . Zool. Mededeel. Rijks.
Mus. Nat. Hist. Leiden, Vol. 5, pp. 33-58, 14 text-figs.
1919a Die Scyphomedusen-Sammlung der Naturhistorischen Reichsmuseums
in Leiden. II. Stauromedusae, Coronatae, Semaeostomeae. Zool.
Mededeel., Rijks. Mus. Nat. Hist., Leiden, Vol. 5, pp. 66-99, 3 text-figs.
1922 Papers from Dr. Th. Mortensen’s Pacific Expedition 1914-1916. XIII.
Die Scyphomedusen-Sammlung von Dr. Th. Mortensen . . . Vidensk.
Meddel Dansk Naturh. Foren., Kjobenhavn, Vol. 73, pp. 513-558.
1924 Ueber einige von Dr. C. J. Van der Horst bei Curasao gesammelte
Medusen. Bijdagen tot der Dierkunde. Amsterdam, Vol. 23, pp. 83-91.
1926 Ueber einige Scyphomedusen von Puerto Galera, Mindoro (Philip-
pinen). Zool. Mededeel., Rijks Mus. Nat. Hist., Leiden, Vol. 9, pp. 239-
248.
1929 Ueber einige Scyphomedusen aus dem Zoologischen Museum in Am-
sterdam. Zool. Mededeel., Rijks Mus. Nat. Hist., Leiden, Vol. 12, pp.
195-216.
1930 Scyphomedusen. Res. sci. Voyage aux Indes Orientales Neerlandaises
. . . Mem. Mus. Roy. Hist. Nat. de Belgique. Hors serie, Vol. 2, Pt. 4,
11 pp.
1931 Ueber einige Coelenterata von Australien. Zool. Meded., Rijks Mus.
Nat. Hist., Leiden, Vol. 14, pp. 27-40.
1934 Scyphomedusae. Discovery Repts., Vol. 8, pp. 329-396, Pis. 14-15.
1935 Die Scyphomedusen der Snellius expedition. Verh. konink. acad. Wet.
Amsterdam afd. natuurk., 2nd Section, Vol. 34, No. 6, pp. 1-44, 9
text-figs., 1 pi.
1938]
Bigelow: Medusae Taken During 1929 and 1930
179
1937 Scyphomedusae. Sci. Rep. John Murray Exped., 1933-34, Brit. Mus.
(N. H.), Vol. 4, No. 7, pp. 203-242, 1 pi., 14 text-figs., 2 maps.
Schwartz, C.
1788 Neue Abhandl. Schwed Acad. Deutsche liebers., 1789, p. 195, PI. 6, Fig.
1, (not accessible).
Thiel, M. E.
1928 Die Scyphomedusen des Zoologischen Staatsinstituts und Zoologischen
Museums in Hamburg. 1. Cubomedusae, Stauromedusae und Coronatae.
Mitt. Zool. Staats. Inst. und. Zool. Mus. Hamburg, Vol. 43, pp. 1-34.
1931 Beitrage zur Planktonbevolkerung der Weddellsee nach den Ergebnis-
sen der Deutschen Antarktischen Expedition, 1911-1912. Internat. Rev.
Hydrobriol u. Hydrograph., Vol. 25, pp. 305-332, 1 text-fig.
1932 Die Hydromedusenfauna des Nordlichen Eismeeres . . . Arch. f. Natur-
gesch., n. s., Vol. 1, pp. 435-514.
1932a Ubersicht fiber die Hydromedusen der Arktis. Fauna Arctica, Vol. 6,
pp. 121-158.
1935 Die Besiedlung des Sfidatlantischen Ozeans mit Hydromedusen. TViss.
Ergeb. Deutschen Atlantischen Exped. auf dem . . . “Meteor", 1925-
1927, Vol. 12, Pt. 2, pp. 32-100.
1936 Systematische Studien zu den Trachylinae der Meteorexpedition. Zool.
Jahrb. Abt. Syst. Vol. 69, pp. 1-92.
Totton, A. K.
1936 Plankton of the Bermuda Oceanographic Expeditions. VII. Siphon-
ophora taken during the year 1931. Zoologica. New York Zool. Soc.,
Vol. 21, Pt. 4, pp. 231-240.
UCHIDA, TOHRU.
1927 Studies on Japanese Hydromedusae. 1. Anthomedusae. Jour. Fac. Sci.
Imp. Univ., Tokyo, Sect. 4, Zool., Vol. 1, Pt. 3, pp. 145-241, Pis. 10-11.
1928 Studies on Japanese Hydromedusae. 2. Trachomedusae and Narcome-
dusae. Japanese Jour. Zool., Vol. 2, pp. 73-97.
1929 Studies on the Stauromedusae and Cubomedusae, with Special Refer-
ence to their Metamorphosis. Japanese Jour. Zool., Vol. 2, pp. 103-193,
Pis. 3-5.
1934 A Saemostome Medusa with Some Characters of Rhizostomae. Proc.
Imp. Acad., Tokyo, Vol. 10, No. 10, pp. 698-700.
Vanhoffen, Ernst.
1902 Die acraspeden Medusen der deutschen Tiefsee-Expedition 1898-1899.
JFiss. Ergeb. deutschen Tiefsee- Exped. auf dem Dampfer “Valdivia”,
1898-1899, Vol. 3, pp. 1-52, Pis. 1-8.
1902a Die craspedoten Medusen der deutschen Tiefsee-Expedition 1898-1899.
1. Trachymedusen. Wiss. Ergeb. deutschen Tiefsee-Exped. auf dem
Dampfer “ Valdivia ”, 1898-1899, Vol. 3, pp. 53-86, Pis. 9-12.
1908 Die Lucernariden und Scyphomedusen. Deutsche Siidpolar Exped. 1901-
1903 . . . Vol. 10, Zool. Vol. 2, pp. 25-49, 11 text-figs.
1908a Die Narcomedusen. Wiss. Ergeb. Deutschen Tiefsee-Exped. auf dem
Dampfer “Valdivia” , 1898-1899, Vol. 19, Pt. 2, pp. 43-74, Pis. 7-9, 5
text-figs.
1911 Die Anthomedusen und Leptomedusen der Deutschen Tiefsee-Exped.
1898-1899. TPiss. Ergeb. Deutschen Tiefsee-Exped. auf dem Dampfer
“Valdivia”, 1898-1899, Vol. 19, Pt. 5, pp. 193-233, PI. 22, 21 text-figs.
1912 Die Craspedoten Medusen der Deutschen Sfidpolar-Expedition.
Deutsche Siidpolar-Expedition 1901-1903. Vol. 13, Zool. Vol. 5, pp. 351-
396, PI. 24-25, 25 text-figs.
1912a Die craspedoten Medusen des “Vettor Pisani.” Zoologica, Stuttgart,
Vol. 26, Pt. 67, pp. 1-34, 2 pis., 6 text-figs.
1913 Uber westindische Medusen. Zool. Jahrb. Supp. Vol. 11, pp. 413-432.
Von Lendenfeld, R.
1884 The Scyphomedusae of the Southern Hemisphere. Pt. 2, Proc. Linn.
Soc., New South Wales, Vol. 9, pp. 242-249.
APPENDIX1
180
Zoologica: New York Zoological Society
[XXIII :5
aiqBgpnaptuft
»-t 03 — I 03 03 03 03 »-< *~t 03 — < 03
EliJive Bigamy
snaosajnj 'cqujoj i
-
■eoiqi^oou BiStejaj
^Bjnoinaun aqonurj
lajjiAjtn B|[<ny
03 03 i-4 *— < CO *— < 03 <-H ->t< O* «-•
■e^ound aoq^isnejq
■euiq^umAq
’ ,enAqdiJ9(j
»— < CO 03 -H lO 03 CO r-i i-l «-4 f-4 i-l
i3tnaao[dBq afonrex
'BapqAlBQ
eueouoiAux
sqEidns.n?ni eapqAjB^
■ejup uqjueSaj i
-
np|eaiiJ3 BJimiSay ^
- -
npiuaiij8 ujnuiaay
« — < — ' «-H 03 03 03 H03 CO CO 03 -r»<
B9J1I0 uui8ay
bsioui snesiuqog i
<ds salting
gqupiOBOqcud
' uiuoAia*)
c|[XqdEj^3} adouiq
-
reuipjcA gaqdaj^qujj
niiuqs[8 esajoi|Sjj
03 i-< ' 03
nmaiiuira gnajoqyg
*-h rH «-H -H CO — ' — ' — • 03 — < — < 03 —• 03
^sna98
aepiraauAqouJX
B'JOSSOJQ l
- «
cauuiuq ujoggojQ i
amoicLty BraaaoqojOQ
03 i-<
qaqoasq
uoStoqoBjmja
03 — « ■** — '
nm^|9A
^ai9no|Bdoq^j
O «-i 03 ^ 1-H «-H 03
sinaa;
uouoqdgoqd seipnijQ
Buupijoq Eajonbay
ranjqru
^raaaoi^raojqQ
^ds sigdooAjBQ
■BOIAUOU^ BJBI^OJ9;9g
ujqru BapuBj
Boinoo eapusj
aqoia uiqiAuieSnog
uraojoqoip
8isdo9ioa«2
Net
Number
MCN00»HC3M^'«5«0C5'HN«,1'«0O0>*-i03W^»0000>O^e3WO00ON,<'»0ON00ai«’rn5O00
^-,^^^03C303Cso3C303cococoeoeocoeo'^"tf"<f^’^^^r»o»o»o»o»o»o«o«o«c>o«oeococDr^r^t^t>»t»
1938] Bigeloiv: Medusae Taken During 1929 and 1930 181
1 This tabic does not include a number (less than 10) of specimens for which no Net Number was available at the time of compiling table.
APPENDIX'
Net
Number
l!
j|
f
8
1
i £
E
1
|
E
1
1
|l
J2
1
a
8
S
•a
.§1
la
|4
lx
13
i£
I
E
3
1
5
|
1
ll
e s.
E-*
1
S
■a
a
1
|
4
i
1
4
g
1
%
|l
S1
i
1
s
1
1
i
•g>
<
=5
1
1
'I
1
1
1
|
!i
it
t3
.a
s
3
I
1
|
1
Pcriphylla
hyacinthina
|
1
I
|
|
1
8
|
J
1
1
3
|
1
13
IS
17
18
21
22
23
24
28
28
29
31
32
33
34
38
3fl
39
41
42
43
44
49
60
61
52
53
58
62
65
66
67
69
73
74
75
76
7B
5
1
2
4
1
1
2
2
1
4
1
2
1
2
1
1
l
1
3
1
2
1
2
1
2
2
1
1
2
1
l
2
2
2
2
3
3
2
1
1
3
1
3
2
5
2
3
2
2
1
3
1
1
4
2
1
•
1
2
1
2
2
1
1
1
1
1
2
1
2
1
Thi. table doet not include a number (lean than 10) ot .pecimeus Tor which no Not Number
i available at the time of compiling tablo.
Zoologica: Ne,o York Zoological Society [XXIII :5 | ig38] Bigelow: Medusae Taken During 1929 and 1930
182
Zoologica: New York Zoological Society
[XXIII :5
9iqegi)U9piufi
■ejijne «^|9jny
-
suaosajru mjBJOj i
Boiqpoou
'Bj'qnoinSun aqonurj
-
lajjiAitM
CO *— l T-H *-H O
abound 9oq)isrve^
euiq^uioeAq
’ 'B[|^qdiJ9<j
1-t T-t M-HiM-HTfNO 05 -*f »-H *-«
Bai3U0|dBq ByfomEj,
'B9pq^JBQ
~
BU'BDBUlAeX
SljBldnSJBOI T39pqXjBQ
bjbjo CTfnreSsj j
-
iipimau3 BjnuiSsy <
<M 1-< »-H
upjurau3 ejnui39y
y-> »— • »— • *-< *-H CO r-< CO — <
1291)19 Bui39y
CS) H OJ
bsioui snssiaqog i
,;ds buiuuq
sqepiosoqojd
’ biuo^j9q
T?jlAqdx?j)9) adourj
9BlAip[BA S9qd9J)jIBJJ
*
raiuqB[3 sbsjoi[Bjj
HinO CM CO *-t y-> r-i Hf)
ranaiiuiai SB9J0i[eg
(N'J'N O CO 03 1-H •— 1 T-t r-l CO *-H CN CO CO hh
^snu93
9Bpira9UA'qOBJJi
b- < CM i-H
B)OSSOJQ l
-
B9aanjq bjossojq i
ainoidA) Bai9uoqo[03
CO CO *-H ’-H l C<J
q9qo9Bq
uoioqoejuej
ran)B[9A
Bra9uo[Bdoq^j
"
sinus;
Bouoqdsoqd seipuqQ
BUBpuop B9ionb9y
ranjqru
Bai9no)Braoiq3
^ds sisdoo^BQ
BUI^UOUB BJBl^019)9g
Bjqm B9PUBJ
BOIUOO BSpUBJ
sqoiu Bi|[iAuiB3nog
Brao^oqoip
SISd09[0UB2
Net
Number
CON0005O*HC0'}|O5O'-'NC0't>C«0NC0NaiON00'-IC0O'-ll0cON^NN00O5i-lu:O)t000'H'»J'00-H
CDOtD-ONNNNN00 03 05 O5O5 05 05 05'}'i0iC^>5O®N»OOOO(NC0'^,^'J,'^»0i0i0C0!0C0M00O
<M<M(M(M(M0qC^C^<M<M(M(MC^C^CM<MC^C0e0COC0C0C0C0C0>0<-^'«^Tr^J'TtiTrrf'Tti-^T}»Tf’<f'^''^^'^''Tj’»O
1938]
Bigelow: Medusae Taken During 1929 and 1930
183
Net
Number
ij
1 3
|
•p
8
J
( 2
e
2
1
|
|
g
o
Chroraatonema
rubrum
1
i
|
8
j
|
.11
li
« "
U
Pi
Coloboncma typicuml
8
|
5
I
3
6
1
'L
| 3
1 E
Halicreas minimum
Halicreas glabrum
J
J
1
1
Liriopo tctrapbylla
1 Gcryonin
proboscidalis
i
1
8
j
|
8
f
i
■i
e
|
i
■&
e
I
M
J
1
1
if
IS
5
8
3
CJ
j Tamoyn haploncma |
ouqiuumXq I
’ unfcpluaj
3
2
1
1
1
|
3
I
§
| Pelagia noctiluca ||
E
1
g
3
1
3
1
'1
263
267
268
269
270
271
273
274
279
280
291
292
293
294
295
290
297
343
357
359
360
362
368
371
383
400
401
405
406
427
434
442
447
448
449
451
455
459
468
481
484
488
1
*
3
3
1
2
■
1
2
17
3
2
1
1
1
3
2
2
3
3
1
1
5
0
2
3
1
1
1
2
1
2
1
1
i
i
1
2
1
I
3
1
3
2
1
1
i
1
1
1
2
1
2
*7
0
4
1
11
1
3
1
2
1
1
6
'
1
1
2
1
1
Zoologica: New York Zoological Society [XXIII :5 ( 19381 Bigelow: Medusae Taken During 1929 and 1930
184 Zoologica: New York Zoological Society [XXIII :5
a[qBgi^uapniQ
- - « - - c,~
■KjUTve Biqajny
SU80S3JIU 'BqBJOJ l
'Bonji^oou BiSujaj
B^'ejnainSun aqonuiq
<N
raijiAAn ^||oiy
NCOH Tf< O* CM ^ t-H f-H 1-H »-H — 1 CM 1-i
BjTqound aoqjisnB^
~
BuiqjuiO'BXq
'C|[Aqdija<j
~ ~ ~ N
Enrauo]di3q
BapqXjBQ
-
BUeDBOL/LBX
sijBidnsjBui -eapq^JBQ
lutqo Bq^ireSaa <
iipix}oiij8 BmuiSay i
iip|tjaiij2 BjnuiSay
co -*f >— < co »~t hn <m <m coi cm o coco oa
Bai^io Baiaay
- « -
■bsioui snssinqog i
- -
,:ds BuiunQ
sqBpiosoqojd
uiuoXjaQ
'B||AqdTJj^a^ adoiiiq
aBiAippjA saqdaj^qeji
-
ranjqB[3 sBajoipry
ramniuira sBajoqBjj
' *-H «-< CO -I •— 1
^snua8
aBpiraauXqoejj,
■*f 1-H
B^OSSOJQ i
T-< *-t »-H CO
Bauuiuq bjossojq i
CM .-I
innoicLty BinauoqoiOQ
-
qa^oa-eq
uoSoqaBjUBj
»-• T-< CM >-< CM CM 1-tiM y-t
um;'B|aA
'eraauoi'Bdoq^j
— CM <M f-H
smua;
Bouoqdsoqd SBipuqo
BQBpuop uajonbay
miuqiu
Baiauo^BraojqQ
;ds sisdoOiCjaj
boiAuoub BiBi^oja;aH
Bjqnj BapuBj
boiuoo BapuBj
-
aqoiu BiuiAureSnog
Braojoqoip
sisdoa[ouB2
Net
Number
h-Q0'H(NC0'l(>flN0005O'HiNC0'J,‘0C0N00O}NC0'^N00®C0i0eO^C0OiHiNC0'C>0cDN00O»-iNC0
(N(NCOCOCOCOCOCOCOCO'fl,'^Tt'T}i'<j<'tl<^^Tji'^»OU3U5»C»CiCcDcOcOr^NOOOOOOOOOOOOCOOOOOC»OSa)®
OCOCD©«OcDCOO®OCDcOeDcOcOcOOCOcDcOcDcDCDCDcDCOCOcDcO®CDOcOcDCDcOtOCOcDCO«OOcOCO
1938]
Bigelow: Medusae Taken During 1929 and 1930
185
Net
Number
||
¥
'£
i
8
1
|
2
J
|
J
a
1
11
2
|
a
1
8
1
a-|
■s |
o
ii
■jg
N
1!
§
j
1
|
8
|
f
6
5
i
|
f]
£
|
a
'a
1
a
s
•i
1
S
J
.a
|
ii
d®1
1
i
8
1
§
|
1
■E
s
1
I
ll
!l
5
3
3
o
Tamoya haploncma
Pcripbylla
hyacinthina
Nausithoc punctata
1
3
f
8
|
p-i
3
1
1
2
J
3
3
1
3
i
C27
G28
631
632
633
634
635
637
638
639
640
642
643
645
646
647
648
649
652
653
654
657
658
659
663
665
G60
674
670
680
681
682
683
684
685
086
687
688
690
691
692
693
■
2
2
1
2
2
2
1
2
1
1
1
1
l
i
i
i
3
•
l
3
1
1
l
i
i
i
1
l
3
1
3
1
2
2
2
3
2
6
6
3
2
1
1
1
1
3
1
1
1
1
2
1
2
3
1
2
2
1
1
1
1
1
1
1
1
1
1
2
1
2
1
3
1
2
Zoologica: New York Zoological Society [XXIII :5 I 1938] Bigelow: Medusae Taken During 1929 and 1930
186 ' Zoologica: New York Zoological Society [XXIII :5
9jqtjgiiu9pmjq
- - ~
bjurb Eijiajny
snaosajiu ci]cjoj i
caiqpoon
BjTqnoinSnn gqomnx
t:||0}y
Oi CM CM CM CM *-> lO *-» CM Tt* i-H r— ( T— < i-H CM CM i— < CM
tf^onnd goqiisne^
T?aiq^niotiAq
‘ uqXqdugj
CM — i CM CM i-H CM T-t i— l CO i—i -h CO i— 1 CM i— I r-uC CO CM CO
uai9uojdBq 'B^oraux
BSpqXJBQ
BaBOBOI^BX
sqBidnsj'Bni BapqXiBQ
bjbjo <
-
iip[Btnij8 ttjnniSay <
np|BraiJ3 BjnuiSay
CO CM T-I Tji i-H i-l i-H CO 1-1 CO
TOJJIO EUI83V
CM CM i-l
bsiooi snssiuqog i
-
^ds buiuiiq
sqepiosoqojd
BIU0i£J9£)
BqXqdBj^a^ adourq
9BiArpjBA saqdaj^qBg
CM i-H y-t
nnuqBja SB3JoqEji
umraiuiai segjoqBjj
~ ~ -
^snuaS
8BpiraauAqoBJX
CM CM r-< CM CM i-l
B^OSSOJQ i
i-l CM i-l CM i—t
B9UUIUq B^OSSOJQ £
CM i-H
ranoicLfy BraguoqojOQ
CM CM i-H
i[9q09Bq
noSoqoB^nBx
- -
ran;B|9A
Bra9nojBdoqH
-
smu9^
uouoqdsoqd sBipaqo
Bu^pijop B9ionb9y
unuqiu
BD39no^Braojq3
<ds SISdOOAj'BQ
UtILftlOUB BaBI}Oa3}3JJ
Bjqnj B9pnBx
-
BOIQOO B9piTBX
9qom BiqiAureSnog
-
uaicqoqoip
sisdo9p’uB2
Net
Number
iCONOHCMCOifiCNOOOO'HiONOOOiiHCOiJ'iOtDNOOaiO'HitiflCDNOOCJO'-iCMcO^iCOOOeM^
Ot0®r«NNNNNSNN00CC00 00WIX)0S0)Cl©05a 05 05OOOOOOOOHHHHHHMHN(N
l^t^t^.t>.t>-t^t^t>-t^t>.t^t>.t^t>.t^.t-t^t^t-t^t^t^.t^t^.f>it^OOOOOOOOOOOOCOOOOOOOOOCOOOC»ooOOOOOO
1938]
Bigelow: Medusae Taken During 1929 and 1930
187
C"$
(M
r- 1
(M
•H 1-H rHrt'-f MC^
iO <M 1— ( <M (M
<M
<M
<M CO *—* ION
<M
CO
i0'-0N05OHNif350N00ONM’fU5t00005O'-'0)»'},l0'0N00O'-'C')M^,»0ON000JTjiU3O*iMC0>CONO'-iNC0'},i0<0N
MNNNMP5COmMMCO'^,tTl'rf'a'^'f'fl'iCiniOiOlO»OlC>CiO«D^COcOCiCDtDCDy5cDNNOOOOOOOOOOCl0000503aO»0505C305
CO GC 00 00 CO 00 00 CO GO QO OO CO OO QO CO 0O CO CO GO CO CO 00 QO GO OO 00 OO 00 OO OO OO 00 00 OO GO OO 00 00 GO OO GO 00 OO 00 OO GO OO OO OO 00 CO OO OO OO OO
Net
Number
.= 1
as
gjj
s-M
s
|
•a
|
8
J
1
5
1
Heterotiara anonyma
1
Chromatonema
rubrum
1
i
1
Olimlias phosphorica
I
II
fj
Colobonema typicum
8
|
2
i
1 Trachyncmidae
genus?
|
i
$
1
a
5
jj
1
=3
€
1
1
i
■i i
!
a
1
1
£
1
■<
|
'§
§
J
|
1
j
1
!!
II
S3
3
S
3
|
I
I
a
=s|
g-S1
Oh
|
1
1
J
3
1
|
•s>
.a
3
%
2
|
|
1
705
766
767
770
771
772
773
774
775
777
778
779
780
781
785
787
788
789
701
793
794
795
796
797
798
799
800
801
804
805
806
807
808
809
810
811
812
813
814
815
816
818
822
824
1
■
1
2
1
2
1
1
2
1
2
1
2
2
1
2
2
i
4
1
1
1
2
■
2
1
3
2
1
1
1
1
3
1
3
1
2
1
2
2
1
3
1
1
2
1
5
3
3
1
9
2
2
2
5
1
2
1
1
2
2
2
1
1
1
1
2
1
2
186 * Zoological New York Zoological Society [XXIII :5 j 1938] Bigelow: Medusae Taken During 1929 and 1930 187
188 Zoologica: New York Zoological Society [XXIII :5
3R^S!?naP!an
r-l — < CO „ t 1— « ^ <
eqpjny
8U99S9JIU l
uorqi^oou ciSttpj
B^inomSun gqonuiq
reqiA&ii Bn°»y
1-H <M 1-H -Hoi r- —
B^ound goq^isnu^
-
■euiq^uiOBAq
■eqAqdugj
1-H CO 1-H 1-H (M cqj t-h 1-h .-h HHHN Nh Ofl i-h i-h
,eai9aojdBq BAoraBj,
•e^vye 'egpqX.reQ
naucmuLtax
sq'Bidnsj'eai ugpqA.reQ
^h ifooo
1-H 1-H CO
'BJiqO Bq^U1339<J ^
npi^niuS ^jnniasy i
1-H 03
iippeuiu3 ,Bjnni39y
CO
*B9J^I0 ^Ul39y
“
■bsioui snssiaqog i
jds BuiunQ
sipipiosoqojd
‘ 'BlU0AJ9{)
B[[Aqdrai3i adouiq;
geiAipjEA S9qd9j^iit?u
-
ratuqEjS src3JOi[i3]j
CO
umraiurai s^joqujj
T-H 1-H HNNH CO 1-h 1-h 1-h CO ZO
,;snu93
9Bpira9uAqoBJX
CO T-H
'B^OSSOJQ i
- - -
B9uuruq «;ossojq i
canDicU} ^raauoqo[03
- -
q9qo9Bq
uoSoqoejuu^
1-H 1-H CO
um^^A
uamuopJdoq'y;
<M CS 1-H 1-1 1-H
smu9^
uouoqdsoqd eiqpuqo
■euEpiJog B9ionb9y
uiruqru
■eni9uo)Baiojq3
-
,;ds sisdoo^Q
03
UU1<£U0UU ‘BJBi;OJ9^9H
-
'Bjqiu lB9pu,B<j
Boinoo BgpuBjj
aqotu BiiiiAuiGanoa
-
Brao^oqoip
sisdo9[0UB2
-
Net
Number
CCO3lNC0>CiC«0NC005iH(N»0C000O5O'H(NC0iCN0005OMC0>ONOJiH^i0iNC0H}iNC005O’HNC0ai
Q0)OOO'HrHiHrHrHC'JWM05(N«C0MC0C0C0««C0il<iJ,'ti}''fiJ'iCi0‘0cDC0C0fflC0C0NNNNN
OOOO 0303C50i05050505030jroO‘-0>05010'-OjC3 0Gi05 0'-C»a3 0i050505 0<®®0505ffi0305 0i050>©003
1938]
Bigelow: Medusae Taken During 1929 and 1930
189
C<l <M ^ CO
os cn o> g os
l> <U 0) CO" OJ on- O tO
o »— i ooj o •— i
2 ^
3 3 3 3 3 *3 =3 « =3 "
2>-5i72'-s(y2'-5COl/3cQO
3 3
02 *-3
Net
Number
.= 1
l\
1
1
S
a
1
2
2
f
£
1
EE
I
1
II
i 2
1
i
§
<
Olindiaa phosphorica
tenuis
l
!i
sj
IP
£
1
1
|
&
2
|
|
J
a
Is
I1
a
.1
1
a
5
!
3
3
4
0
8
£
a
t
2
§.
3
ll
I*
I
.1
8
1
J
2
1
i
|
I
1
fee
I
■<
|
§
p.
Is
“ 2
ll
o
3
■3
2
I
|
I
I
a
ll
Oh
|
|
1
£
|
3
1
i
3
2
1
1
1
1
s
1
3
I
|
s
898
899
902
903
905
915
916
918
919
921
922
925
928
929
930
931
932
933
937
938
939
940
942
943
945
947
949
951
954
959
962
963
964
967
968
969
970
971
977
978
1
1
1
1
2
1
2
2
1
1
1
I
3
1
1
1
3
1
1
2
2
3
1
1
1
3
6
5
3
l
I
3
1
2
l
18
66
1
3
1
2
2
1
1
1
1
1
2
1
2
1
1
1
2
1
1
1
1
1
1
1
3
1
2
2
1
1
Surfaco
June 25. 1929
Surface
July II, 1929 1
Surface
July 23, 1929
Surfaco
Sept. 8, 1930
Surface
Oct. 16. 1930
Surface
June 8, 1929
Zoologica: New York Zoological Society [XXIII :5 f 1938] Bigelow: Medusae Taken During 1929 and 1930
Nichols: Classification of Carp-like Fishes
191
6.
Classification of Carp-Like Fishes.
J. T. Nichols.
The American Museum of Natural History.
In attempting to classify the carp-like fishes (Eventognathi or Cypri-
noidea) it is generally assumed that the suckers (Catostomidae) are the
most primitive group. This is thoroughly in accord with the writer’s views
and he surmises that it is among ancestral bottom-feeding suckers that these
fishes lost oral dentition, the lack of which is an outstanding feature with
them all. Suckers are closely related to the true carps (Cyprinidae) , but
their recognition as a full family is at least convenient in this case. The
comparatively few genera of suckers need no further subdivision. It may
be noted that the northern genus Catostomus represents generalized or
standardized forms, and that the single peculiar genus Myxocyprinus in
China resembles one of the specialized genera in the lower Mississippi
Valley. These peculiar suckers also have characters which are perhaps
ancestral; and it may be questioned whether resemblance between Chinese
and American genera is relationship or parallelism, whether these are
specializations of more standard northern suckers or of some different
ancestral forms.
More different in the main from the true carps are the loaches, a large
varied group with characters generally spoken of as degenerate. It is the
writer’s view that the loaches are a recent specialization from the carps,
perhaps more or less polyphylatic, but this may not be generally accepted.
Classification of the loaches is very difficult. Two families are usually
recognized, true loaches (Cobitidae), and flattened bottom forms (Homa-
lopteridae). In the latter a series from Crossostoma to Gastromyzon is
rather clear, its relationship to the others of which Lepturichthys seems to
be a terminal specialization, obscure. It is probably impossible to divide
the loaches on the basis of structural resemblances without violating rela-
tionships. To place the emphasis on apparent relationships the writer
would recognize a single family for them (Cobitidae), consider the obvious
groups or series central for four subfamilies, and tentatively place aberrant
forms wherever they seem to fit least badly.
Cobitis and related genera (Cobitinae) are more or less elongate with
erectile spine under the eye, a peculiar un-carp-like specialization. As their
ancestors must have been without it we have some justification in consider-
ing Misgurnus, a widely distributed, abundant and somewhat aberrant eel-
like genus which lacks the spine and otherwise resembles Cobitis in various
ways, a primitive member of this series. Botia and related genera might
be secondarily free-swimming standardized forms derived from Cobitis.
Nemaclieilus, Barbatula and related genera lacking the erectile spine
may be grouped in a subfamily (Nemacheilinae) very abundant in High
Asia, also with a claim, which we do not recognize, to the genus Misgurnus.
192
Zoologica: Neto York Zoological Society
[XXIII :6
The aberrant carp-like genus Gobiobotia, for lack of a better place, may
go as a primitive Homalopterin ; and the Crossostoma-Gastromyzon series
stand as a subfamily (Gastromyzoninae) .
The true carps are the most abundant family in this group, and repre-
sented by many genera. Their relationships to one another are sufficiently
close and sufficiently confused to make their subdivision hardly worthwhile,
but it is certainly a convenience. In so doing one may also give tangible
expression to one’s views on relationships. The carps are most differen-
tiated in Asia; all the main groups are represented and most divergent there.
Weber and de Beaufort (Vol. Ill, 1916) recognize 3 subfamilies for
Indo-Australian forms, Abramidinae, Rasborinae and Cyprininae, of which
the last is the most composite. Turning to the list of species from China
on the Asiatic continent we find various genera which fit into these sub-
families without much question, and others with northern affinities which
do not. The keeled breams, Megalobrama, Hemiculter, Erythroculter and
so forth obviously go into the Abramidinae, Opsariichthys which is I’elated
to Rasbora in the Rasborinae, Cyprinus and Barbus and their allies in the
Cyprininae. One hesitates to place Leuciscus and Phoxinus, which with
their allies are widely distributed northward, in the Rasborinae and may
recognize a superficially somewhat parallel subfamily (Leuciscinae) for
these. Though probably derived from this group Schizothorax and allied
genera, abundant in High Asia, are sufficiently numerous and distinct to
recognize as another subfamily (Schizothoracinae) . The peculiar transverse
mouth of the abundant Chinese genus Xenocypris is presumedly derived
from a more or less sucking mouth of bottom feeding forms, and this is
perhaps the secondarily actively free swimming terminal member of a
series (Chondrostomatinae) that runs from Labeo, through Varicorhinus
to it, with various aberrant side specializations. The little fishes belonging
to Rhodeus and allied genera form a uniform well differentiated group
(Rhodeinae). The whole series of genera allied to Gobio, showing grada-
tion from standard free-swimming to specialized bottom forms, may con-
veniently be considered a subfamily (Gobioninae) , perhaps the most recently
evolved, now replacing in the Chinese center of specialization earlier Chon-
drostomatinae, just as these or some other bottom Cyprinids presumably
replaced the Catostomidae.
The standardized Leuciscinae may be the oldest group. Native American
carps belong to it, with the exception of Notemigonus, an Abramidin. The
Abramidinae and Schizothoracinae seem to be specializations of the Leucis-
cinae. Whence the Rhodeinae came is uncertain but one may suspect from
the Abramidinae. The Cyprininae are now flowering in southern Asia,
and Cyprinus itself is possibly a relict from an earlier period of differentia-
tion in this group. The Gobioninae may be derived from the Cyprininae.
There are plenty of Cyprinid genera, aberrant or otherwise, to be
assigned to this or that of these eight divisions according to one’s idea of
their relationships, but these eight as outlined can be considered to cover
the group in all parts of the world, and afford a basis for discussion of
the relationships of any genus.
Tentative Subdivision of the Cyprinoidea.
Suborder Cyprinoidea. Carp-like fishes.
Family Catostomidae. Suckers.
Family Cyprinidae. True Carps.
(1) Subfamily Leuciscinae — Leuciscus, Phoxinus, Richardsonius,
Hybopsis, etc.
1938]
Nichols: Classification of Carp-like Fishes
193
(2) Subfamily Rasborinae — Rasbora, Danio, Opsariichthys,
Barilius, etc.
(3) Subfamily Abramidinae. Abramis, Erythroculter, Hemiculter,
Notemigonus, etc.
(4) Subfamily Schizothoracinae. Schizothorax, Schizopygopsis,
Diptychus, etc.
(5) Subfamily Rhodeinae. Rhodens, Pseudoperilampus, Acantho-
rhodeus, etc.
(6) Subfamily Cyprininae. Cyprinus, Osteochilus, Barbus, Labeo,
etc.
(7) Subfamily Chondrostomatinae. (Labeo), Varicorhinus, Chon-
dr ostoma, Xenocypris, etc.
(8) Subfamily Gobioninae. Gnathopogon, Gobio, Pseudogobio,
Saurogobio, etc.
Family Cobitidae. Loaches.
(1) Subfamily Cobitinae. Misgurnus, Cobitis, Botia, etc.
(2) Subfamily Nemacheilinae. (Misgurnus) , Nemacheilus, Bar-
batula, etc.
(3) Subfamily Homalopterinae. (Gobiobotia) , Homaloptera, Lep-
turiclithys, etc.
(4) Subfamily Gastromyzoninae. Crossostoma, Hemimyzon, Gas-
tromyzon, etc.
Dr. Y. T. Chu has recently (1935, Biol. Bull. St. John’s Univ., Shanghai,
No. 2, p. ix) listed Chinese genera of Cyprinidae in eight subfamilies. He
does not differentiate Rasborinae from Leuciscinae, and recognizes a sub-
family for the aberrant genus Hypophthalmichthys of uncertain relation-
ships. There are presumably other aberrant Cyprinid genera which might
be so treated. He confines the Chondrostomatinae to genera close to
Xenocypris, leaving less terminal ones in the Cyprininae, a more conserva-
tive procedure especially as the phylaticy of the series from Labeo to
Xenocypris is assumed rather than proved. Also the name Acheilognathinae
is used by him in place of Rhodeinae.
Nigrelli & Smith: Tissue Response to Parasite
195
7.
Tissue Responses of Cyprinodon variegatus to the Myxosporidian
Parasite, Myxobolus lintoni Gurley.
Ross F. Nigrelli
New York Aquarium
&
G. M. Smith
Department of Anatomy, Yale School of Medicine,
and New York Aquarium.
(Plates I-VII).
Introduction.
Little is known about the histopathology of myxosporidian infections.
The majority of these protozoan parasites are found in the gall-bladder
and urinary bladder of fresh water and marine fishes but cause little or no
damage to the host tissues. However, when the infections become localized
in the epithelium of the intestine, in the liver, or in other organs, a con-
siderable number of pathological changes may occur. If the fins, skin or mus-
culature are invaded, the resulting lesions are very conspicuous, producing in
many instances tumor-like growths.
One of the best known myxosporidian tumors is found on the European
Barbel, Barbus barbus, and is called the “boil disease” or “Beulenkrankheit.”
This condition is caused by Myxobolus pfeifferi Thelohan and has been
reported for other European fresh water fishes such as Barbus fluviatilis and
B. plebejus. Keysellitz (1908) has found tumors on these fish varying in size
from that of a millet grain to one as big as a hen’s egg, and as many as
twenty-three small ones on a single fish. According to Doflein (1928), the
causative agent may be found in all organs of the body. When the muscula-
ture becomes infected tumor-like growths are formed. Microscopically, such
structures show a hypertrophy of the “interfibrillares” connective tissues,
Myxosporidia, degenerated muscle, and a secondary infection with bacteria.
Fiebiger and Kahls (1929) have also described the “boil disease” from
certain Austrian fishes. In the Barbel, the typical nodules were found, show-
ing the characteristic pathological picture. However, a similar infection
in the Giant Perch ( Lucioperca sandra ) resulted in no pathological changes,
except for some vascular congestion.
Plehn (1910) reported that Myxobolus piriformis Thelohan produced in
a European minnow, Leuciscus, growths which resemble a papillary adeno-
cystoma, but which were not regarded as true tumors. These are usually
found in the tail and histologically show folds of the epithelium of the skin
with the parasites within the folds.
Hahn (1913) described tumors from the common killifish, Fundulus
196
Zoologica: New York Zoological Society
[XXIII :7
heteroclitus , resulting from an infection of Myxobolus musculi = M. funduli
Kudo, 1919. The lesions, according to Hahn, are caused primarily by worm
parasites or mechanical injury and only secondarily do they become infected
with Myxosporidia. Microscopically, the tumors show an infiltration with
small lymphocytes and a preponderance of vascular tissue and erythrocytes.
Numerous non-staining granules of unknown nature and origin were found
in infected epidermis, in connective tissue of the dermis and in atrophied
muscle fibers. The latter were completely degenerated, showing but few
fibers with normal fibrillae and cross striations. The entire tumor mass
eventually becomes infected with bacteria. According to Hahn, the vegeta-
tive stages of the parasite are intracellular and found within the muscle
fibers.
Kudo (1929, 1931) has noted tumor formation due to Myxobolus and
related Myxosporidia in several species of fresh water fishes. One form,
Myxobolus notatus Mavor, is found in the sub-dermal connective tissues of
the tail muscles of the blunt-nosed minnow, Pimephales notatus ( Hyborhyn -
chus notatus) and Leuciscus rutilis. The tumor enclosing a trophozoite
reaches a diameter of 7 mm. Histologically, the growths are composed of
large cysts which are surrounded by the sub-dermal cells of the hosts. The
tissue around the parasite becomes so highly changed that it appears as an
“epithelium.” The tissues not in contact with the cysts show no cytological
changes. However, such areas have an abundance of blood capillaries. The
nuclei of the cells in direct contact with the cysts become hypertrophied;
the cells stain deeper than the normal cells, and lack the distinct membrane
characteristic of epithelial cells. They vary considerably in size and shape,
being columnar, club-shaped or irregularly rounded, and do not form a
continuous layer. Kudo believed that these “epithelial cells” are modified
connective tissue cells. Similar pathological pictures were described by
Mavor (1916) and Debaisieux (1925).
There are other genera of Myxosporidia that produce definite responses
of host tissues. In some cases such infections result in a complete degenera-
tion of the invaded areas, often causing the death of the fish; other infec-
tions result in a simple hypertrophy of the connective tissue cells, enclosing
the spores or vegetative stages of the parasite (Plehn, 1905; Kudo, 1919,
1926; Davis, 1923; Dunkerly, 1925).
Recently, the writers have encountered several specimens of Cyprinodon
variegatus Lacepede with large myxosporidial tumors of the body. This
has afforded the opportunity of making a study of various histological
responses of the host resulting from such an infection. It is recalled that
this disease was first noted by Linton (1889) in sheepshead minnows caught
at Woods Hole, Massachusetts. Gurley (1891) definitely allocated the para-
site causing the tumors and named it Myxobolus lintoni. In 1894, he gave
a more detailed description of the species. Similar parasites and tumors
were observed by Hahn (1913).
Material and Methods.
The infected Cyprinodon were caught at Sandy Hook Bay, New Jersey,
and the Connecticut River. The tissues were fixed in 10% neutral formalin,
and embedded in paraffin after decalcification. These were then sectioned
at 3-10 y thick and stained with Giemsa’s, Wright’s, hematoxylin-eosin and
Masson’s special connective tissue stain. An examination of the spores in
these preparations showed them to be Myxobolus lintoni Gurley.
Description of the Parasites.
Vegetative Stages. Histological examination of the tumors revealed
many stages in the development of the parasite. Although an occasional
1938]
Nigrelli & Smith: Tissue Response to Parasite
197
spore was found free in the lumen of the intestine, there was no evidence of
sporoplasm or multiplicative stages in the gut-epithelium. The tumors them-
selves are composed of enormous numbers of spores and vegetative parasitic
masses surrounded by host tissues (Figs. 4-7). These masses vary as to
number. Thus in certain flat tumors (Fig. 1), there are only a few present
and more or less widely scattered. In more rounded pendant types of tumors
(Figs. 2 & 3), the vegetative groups are numerous but are situated mostly
at the base or the periphery near the body wall. The parasites towards the
surface of the growths are usually fully developed spores. The size and
shape of the circumscribed vegetative masses also vary a great deal. A few
are spherical, measuring from 30 g to 66 p in diameter. The majority of
them, however, are oval and range in size from 18 x 29 to 62 x 128.7 p.
Each vegetative mass demonstrates clearly various stages in sporogene-
sis. In the larger groups may be found sporonts with two, four and six
nuclei and minute basophilic staining granules in the cytoplasm, or young
spores with a nucleus at the base of each polar capsule and two or four
nuclei in the sporoplasm, depending upon whether or not the shell is developed.
In these young spores, the polar filaments are in the form of basophilic
granules aligned spirally. Scattered throughout the vegetative masses may
be found nuclei of different forms together with degenerating host tissue.
Other large circumscribed masses contain only matured spores (Fig. 7),
many with but a single nucleus in the sporoplasm. In smaller masses (10-15 p
in diameter) trophozoites with many nuclei are present. According to Kudo,
and other investigators, the larger of the nuclei are vegetative, while the
smaller become generative nuclei. The latter are “budded” off with a small
bit of cytoplasm and become the sporonts. The nucleus of the sporont divides
several times until, in this species at least, six nuclei are produced. Such
cells have basophilic granules in the cytoplasm. These may be discarded
nucleoli; a phenomenon previously reported for other Myxosporidia (see
Kudo, 1926). In so far as could be determined, the sporont at this stage
becomes transformed into the definitive spore. It is assumed that in the
final stage of sporogenesis two nuclei are used in the formation of the polar
capsule, two for the shell and two remain as the nuclei of the sporoplasm.
Therefore, unlike the majority of forms belonging to the genus Myxobolus
the pansporoblast of M. lintoni gives rise to a single spore, instead of two or
more.
Spore. The present observations, on the fully developed spores agree
with those of Linton and Gurley, except in size. The measurements given
by Gurley and also used by Kudo (1919) are as follows: length 13.9 p ;
breadth 11 p ; thickness 8 p. Certain of our measurements agree with those
given by Gurley, but the average size, for over 200 spores taken at random,
is slightly smaller than those given above, measuring 10.96 x 7.47 x 6.46 p.
The extreme measurements are as follows: length 9.13 to 14.9; width 4.98 to
10.40; thickness 4.15 to 7.47 /x. However, the smaller size recorded here for
Myxobolus lintoni is no doubt partly due to shrinkage following fixation.
The spores (Fig. 7) are oval in shape. The shell is more or less thick
and with a fairly marked sutural ridge. There are two polar capsules, having
an average measurement of 4 p in length and 2 p in width. The sporoplasm
usually contains two nuclei and a large “iodinophilous” vacuole. The more
mature spores have but a single nucleus which, according to Kudo (1931),
is the result of a fusion of the two nuclei of the younger spore.
With Masson’s stain the shell, capsule and vacuole of the sporoplasm
are colored green ; the polar filaments and sporoplasm are colored red ; while
the nuclei take on a deeper red color.
Although the organisms are usually localized in the tumor mass, an
occasional spore was found in the lumen of the intestine, on the gills, in
the liver, kidney and external to the meninges of the cord. None of these
spores resemble the species reported by Davis (1917) as Myxobolus
198
Zoologica: New York Zoological Society
[XXIII :7
capsulatus which was found in the visceral connective tissues of Cyprinodon
from Beaufort, North Carolina. According to this investigator, M.
capsulatus is present in a state of diffuse infiltration, and differs from M.
lintoni mainly in having longer polar capsules, almost two-thirds the length
of the spore.
Description of the Tumors.
As was previously mentioned, two types of tumors were encountered.
Fig. 1 shows the flattened type with the overlying skin slightly ulcerating
and pigmented. Figs. 2 and 3 demonstrate the large pendant types of
tumors. The fish in Fig. 2 measures about 4 cm. in length. The tumor is
situated on the dorsal surface of the body and measures 8 x 10 mm., the
upper border extending above the surface of the body. The fish in Fig. 3 is
4.24 cm. in length and the tumor, irregular in shape, measures approximately
1 x 1.2 cm. This is situated behind the left pectoral fin and extends dorsally
for about one-half the body width. The ventral border of the tumor reaches
about 4 mm. below the surface of the body. In the last two fish, the skin
appears to be sloughed off, while the tumor of the fish in Fig. 3, is beginning
to show signs of ulceration.
In most of the microscopic preparations the loss of the epithelium cov-
ering the surface of the growths is apparent. The external boundary of
the tumors consists of a single layer of elongated flattened melanophores
(Fig. 6, p). These pigment cells are of the usual corial type found in
this fish, normally lying under the transparent epithelium. It is immedi-
ately below this narrow outer layer of pigment cells that characteristic
changes of host tissue occur (Figs. 4-15). The diffusely scattered spores
and vegetative masses of varying sizes are embedded in a meshwork of
delicate fibroblasts of the host. As was previously mentioned, the diffuse
arrangement of the spores is more frequently encountered near the free
surface of the tumor, whereas the circumscribed collections of spores or
vegetative stages are more numerous near the base or the mesial aspect
of the tumor. It is in the deeper parts of the growth that the fibroblastic
response of the host is most marked. Many delicate spindle-shaped, ir-
regular or often stellate forms of fibroblasts are loosely arranged between
the parasitic masses forming a stroma of the tumor which separates and
supports the circumscribed collections of the organisms (Fig. 6, fi). It is
only rarely that the connective tissue stroma takes on a denser appearance
(Fig. 8). Lying in the stroma are occasional polynuclear cells, lymphocytes,
eosinophiles and mast cells. This leucocytic inflammatory reaction is a very
mild one, and there is practically no evidence of necrosis. The connective
tissue stroma contains also a few melanophores which perhaps have migrated
into the mass of Myxosporidia (Figs. 8 and 10, p). Bacteria, bacilli and a
few cocci, may be seen near the surface of some of the myxosporidial
tumors. These doubtless represent a secondary infection.
In most microscopic fields there co-exists with the fibrous stroma, a
serous or albuminous substance staining faintly pinkish with eosin in
which Myxosporidia lie scattered (Fig. 9).
The vascular supply of myxosporidial growths consists of a very fine
network of capillary blood vessels, very often only wide enough to permit
the passage of a single file of red blood cells. In some fields a few
erythrocytes lie in the tissue spaces outside the lumen of the capillaries.
These cells are normal in appearance, and indicate a mild form of extra-
vasation of blood. No hemorrhages of importance were noted. A few
small nerve trunks were found in areas involved in myxosporidial infection,
but these did not seem to be affected by the diseased conditions.
The tumors of Cyprinodon are not encapsulated, as an irregular infiltra-
tion by the organisms into adjacent structures occurs along the margins of
1938]
Nigrelli & Smith: Tissue Response to Parasite
199
these growths. Myxosporidia gain access to the deeper lying muscular
structure by penetrating through the relatively dense external limiting
fascia which normally separates muscles from the layers of the skin (Fig.
10, f). The fibers of this fascial structure become fragmented and split in
the longitudinal direction into finer bundles between which Myxosporidia in
small groups are found. Organisms which collect below the level of the
external limiting membrane spread out along the fibrous septa which sepa-
rates muscle bundles. This streaming infiltration along intermuscular septa
is shown in Fig. 11. In some places the vegetative masses were found well
organized in the fibrous tissue separating groups of muscle fibers (Fig. 12).
Where the diseased process is advanced, atrophy, loss of striation, hyalin
degeneration and necrosis of muscle fibers may occur (Fig. 12, n). In the
sections examined we have found no evidence of spores or vegetative stages
lying directly in the individual muscle cells as reported by Hahn (1913).
In microscopic sections prepared in the region of the body cavity of
one fish, there were found certain areas where a penetration of organisms
through the pigmented peritoneal lining had occurred (Fig. 13). This was
usually only a slight involvement of the peritoneum, and interestingly
enough, was not associated with an inflammatory exudate. The normally
densely massed pigmented cells of the peritoneum were disarranged so that
groups of melanophores were seen separated by collections of Myxosporidia.
Although this penetration of myxosporidial tumor had occurred imme-
diately in the region of the liver, an infiltration of liver substance had not
taken place to any extent, as only a very few spores could be identified in
the liver tissue.
Partial destruction of scales in the diseased regions was not uncommon.
Bony tissue and cartilage, when lying in a myxosporidial growth, were
found surrounded and infiltrated by spores (Fig. 14). This was particu-
larly true in the case of spongy bone, where the meshes between the bony
spicules were filled with varying numbers of spores (Fig. 15). Here and
there the bony spicules appeared degenerated and necrotic.
Discussion.
Microscopic tissue changes caused by myxosporidial infections have
been described by previous investigators, and have been referred to in the
introduction. The present studies are restricted to material obtained from
Sandy Hook Bay, New Jersey, and the mouth of the Connecticut River.
These fish were caught during August and the early part of September,
1937, in the two locations about 100 miles apart. The lesions in the fish
from both localities were similarly advanced and much the same in char-
acter and distribution. It became apparent, after microscopic studies, that
there was a mild general infestation in each fish examined, as indicated
by the occurrence of spores in the liver, kidney, on the gills and in the
lumen of the intestine. In these organs no histological lesions were evoked.
Contrasting with this, the skin and subcutaneous lesions were most ex-
tensive, as there resulted in these regions relatively large tumor masses.
Here the Myxosporidia were seen as diffuse spores or as circumscribed
vegetative masses in various phases of development and embedded in a
delicate meshwork of fibroblastic tissue varying somewhat in density. This
meshwork of fibroblasts, with its rich capillary blood supply, represents one
of the major responses of the host to the myxosporidial infection. Asso-
ciated with it there occurs a wide-spread exudate which stains a light
pinkish color with eosin and thus resembles a serous or albuminous mate-
rial. In certain areas, the exudate exists as a homogenous substance in
which spores and vegetative masses lie free. In other places, it is granular
or appears to contain fine threads not unlike fibrin. The exudate accom-
panies fibrous tissue organization. It is least conspicuous where the fibro-
200
Zoologica: New York Zoological Society
[XXIII :7
blastic network is most highly developed and condensed. The fact that the
exudate is present at sites without fibroblastic development suggests the
possibility that it precedes the fibrous organization of the tumors.
The spread of the myxosporidial parasites from the subcutaneous
region to the muscles of the body wall is accomplished by their passing
between the loosened fibers of the external limiting fascia. Intermuscular
fibrous septa form pathways for the parasites to reach the deep seated
musculature and the peritoneum. In one specimen a limited involvement
of the peritoneum had occurred.
Myxosporidia infecting intermuscular fascia may result in the de-
generation of muscle tissue itself. This is evidenced by the loss of striation,
hyalinization and atrophy of the fibers.
Bony and cartilaginous structures, interposed in the spread of the
infection, are surrounded and infiltrated by spores and may become necrotic.
A secondary invasion of bacteria, which frequently occurs, may con-
tribute to bring about destruction of various tissues infected by Myxos-
poridia.
Summary.
1. Several specimens of Cyprinodon variegatus were found showing
tumors caused by the Myxosporidian, Myxobolus lintoni Gurley.
2. Certain stages in sporogenesis are reported and the morphology of
the spore redescribed.
3. In so far as could be determined, the sporont of M. lintoni gives rise
to a single spore.
4. This sporont has six nuclei and several discarded nucleoli in the
cytoplasm. As in many Myxosporidia, two of the nuclei probably give
rise to the polar capsules, two to the shell and two remain as the nuclei
of the sporoplasm.
5. With Masson’s special connective tissue stain the shell, capsule and
vacuole of the sporoplasm are colored green ; the polar filaments and sporo-
plasm are colored red; while the nuclei take on a deeper red color.
6. Various histological changes in the host tissue due to the infection
with Myxobolus lintoni are described.
References.
Davis, H. S.
1917. The Myxosporidia of the Beaufort Region. Bull. Bur. Fish. 35: 201-243.
1923. A new Myxosporidian Parasite, the Cause of “Wormy” Halibut. Re-
port of U. S. Comm. Fish. Appendix VIII, pp. 1-5.
Debaisieux, P.
1925. Etudes sur les myxosporidies, III. Myxobolus notatus Mavor. Arch.
Zool. Exper. et Gen. Tome 64.
Doflein, F.
1928. Lehrbuch der Protozoenkunde. Part II. pp. 440-1262. Verlag von
Gustav Fischer. Jena.
Dunkerly, J. S.
1925. The Development and Relationships of the Myxosporidia. Quart. J.
Micr. Sci. Vol. 69.
Fiebiger, J. and Kahls, 0.
1929. Ueber Masseninfektion mit Myxosporidien bei Fischen. Zentralbl
Bakt. I Abt. Orig. 113: 8-19.
1938]
Nigrelli & Smith: Tissue Response to Parasite
201
Gurley, R. R.
1891. On the Classification of the Myxosporidia, a Group of Protozoan Para-
sites Infecting Fish. Bull. U. S. Fish Comm. 11: 407-420.
1894. The Myxosporidia, or Psorosperms of Fishes, and the Epidemics Pro-
duced by Them. U. S. Comm, of Fish and Fisheries. Report for 1892.
pp. 65-304.
Hahn, C. W.
1913. Sporozoon Parasites of Certain Fishes in the Vicinity of Woods Hole,
Massachusetts. Bull. Bur. Fish. 33: 193-214.
Keysselitz, G.
1908. Die Entwicklung von Myxobolus pfeifferi. Arch. f. Protist. 11: 252-308.
Kudo, R.
1919. Studies on Myxosporidia. III. Biol. Mongr. 5: 1-265.
1926. On Myxosoma catostomi Kudo 1923, a Myxosporodian Parasite of the
Sucker, Catostomus commersonii. Arch. f. Protist. 56: 90-115.
1929. Histozoic Myxosporidia Found in Fresh-water Fishes of Illinois,
U. S. A. Arch. f. Protist. 65 : 364-378.
1931. Handbook of Protozoology. 451 pp. *C. C. Thomas, Publ. Baltimore, Md.
Linton, E.
1889. On Certain Wart-like Excrescences Occurring on the Short Minnow,
Cyprinodon variegatus, Due to Psorosperms. Bull. U. S. Fish Comm.
9: 99-102.
Mavor, J. W.
1916. Studies on the Protozoan Parasites of Georgian Bay. Trans. Roy. Soc.
Canada. Ser. 4, 10: 63-74.
Plehn, M.
1905. Ueber die Drehkrankheit der Salmoniden (Lentospora cerebralis)
(Hofer). Arch. f. Protist. 5: 145-166.
1910. Die Pathogene Bedeutung der Myxoboliden fur Fische. Sitz. Ges.
Morph, u. Physiol. Miinchen, 26: 20-27.
202
Zoologica: New York Zoological Society
EXPLANATION OF THE PLATES.
Plate I.
Fig. 1. Live Cyprinodon taken at Sandy Hook, N. J. This fish shows flat myxo-
sporidian tumors.
Figs. 2 & 3. Cyprinodon with large pendant types of myxosporidian tumors.
Killed and fixed specimens taken from Sandy Hook.
Plate II.
Fig. 4. Low power photomicrograph through the myxosporidian tumor of the
skin, x 40. Stained with Giemsa.
Fig. 5. A slightly higher magnification of the tumor shown in Fig. 4. Note the
vegetative parasitic masses in the deeper parts of the tumor, x 75.
Plate III.
•
Fig. 6. Tumor showing circumscribed vegetative parasites and spores. Note the
fine network of fibroblasts (fi) between the parasitic masses. Outer sur-
face lined by pigment cells (p) ; the epidermis has disappeared. About
x 250.
Fig. 7. Fully developed spores of Myxobolus lintoni. x 1500.
Plate IV.
Fig. 8. Fibroblastic respose of the host supporting vegetative, parasitic masses.
Note that in this section the fibroblasts (fi) are denser than those shown
in Fig. 6. A large single melanophore (p) with its dendrites may be seen
in the center. No albuminous exudate present, x 375.
Fig. 9. Section showing a single vegetative mass with developing spores. A few
solitary spores are present in the adjacent albuminous or serous exudate.
- About x 500.
Plate V.
Fig. 10. Low power photomicrograph showing extension of the myxosporidian
parasites through the loosened external limiting fascia (f). (m), muscle
layer, (p), pigment cells, x 90.
Fig. 11. Section showing spores scattered along the intermuscular fibrous tissue,
(m) muscle, x 375.
Plate VI.
Fig. 12. Section showing infiltration of intermuscular connective tissue by the
myxosporidian parasites, (n) necrotic muscle fibers, x 90.
Fig. 13. Fragmentation of the pigmented peritoneum by Myxosporidia. About
x 375.
Plate VII.
Fig. 14. Myxosporidia surrounding bony spicules (b). x 100.
Fig. 15. Infiltration of spongy bone by the spores of Myxobolus lintoni. x 750.
NIGRELL! & SMITH.
PLATE I.
FIG. 1.
FIG. 3.
TISSUE RESPONSES OF CYPRINODON VARIEGATUS TO THE MYXOSPORIDI AN
PARASITE, MYXOBOLUS LINTONI GURLEY.
NIGRELLI & SMITH.
PLATE II.
iMdlWSfifl
•0.1 "?•
A V-'.
» Serf
FIG. 5.
TISSUE RESPONSES OF CYPRINODON VARI EGATUS TO THE MYXOSPORI Dl AN
PARASITE, MYXOBOLUS LINTONI GURLEY.
N1GRELLI & SMITH.
PLATE III.
FIG. 7.
TISSUE RESPONSES OF CYPRINODON VARI EGATUS TO THE M YXOSPORI Dl AN
PARASITE, MYXOBOLUS LINTON1 GURLEY.
NIGRELLI & SMITH.
PLATE IV.
FIG. 8.
TISSUE RESPONSES OF CYPRINODON VARI EGATUS TO THE MYXOSPORIDIAN
PARASITE, MYXOBOLUS LINTON1 GURLEY.
NIGRELLI a SMITH.
PLATE V.
FIG. 10.
«>
FIG. 11.
TISSUE RESPONSES OF CYPRINODON VARI EGATUS TO THE MYXOSPORIDIAN
PARASITE, MYXOBOLUS LINTONI GURLEY.
NIGRELLI & SMITH.
PLATE VI.
FIG. 12.
TISSUE RESPONSES OF CYPRINODON VARIEGATUS TO THE MYXOSPORIDIAN
PARASITE. MYXOBOLUS LINTONI GURLEY.
NIGRELLI a SMITH.
PLATE VII.
ms
'Vi AT-lS
iit'y
$&§$$
|>*fr
3 • - »
Hi
FIG. 15.
TISSUE RESPONSES OF CYPRINODON VARI EGATUS TO THE MYX0SP0R1DIAN
PARASITE, MYXOBOLUS LINTONI GURLEY.
Cox & Coates: Electric Eel Tissue
203
8.
Electrical Characteristics of the Electric Tissue of the Electric Eel,
Electrophorus electric us (Linnaeus).
R. T. Cox
Department of Physics, New York University,
University Heights
&
C. W. Coates
New York Aquarium
(Plate I; Text-figures 1 & 2).
A previous paper1 by the present authors jointly with L. P. Granath
described observations made with a cathode-ray oscillograph of the dis-
charge of the electric eel. In that work two types of discharge, the “major”
and “minor,” characterized by repeatable peak voltages, were identified as
the discharges of the large organs and the organs of Sachs respectively.
Other discharges, of indefinite peak voltage but identified by their occur-
rence only between a minor and an ensuing train of major discharges, were
assigned to a third type under the name “intermediate,” and were tentatively
associated with the organs of Hunter. Observations made with external
electrodes on the skin of the eel out of water showed that the major dis-
charge runs along the large organ as a pulse of potential gradient at a
speed of the order of 500 to 1,000 meters per second.
On the completion of this part of the work it was considered advisable
to carry the research for a time to a place where the electric eel was
plentiful and could be observed in the wild state or at least under nearly
natural conditions. Accordingly an expedition was organized by New York
University and the New York Aquarium. Its expenses were met by a
generous grant from the Doctor Simon Baruch Foundation, for which we
wish to express our gratitude. To the Goeldi Museum of Para, Brazil, we
are indebted for a laboratory and facilities for the work, and to Dr. Hag-
mann and Snhr. Pira of the staff of the Museum we are particularly
obliged for their constant kind assistance. Specimens for observation were
supplied by Mr. Karl Griem. He and his associates, Messrs. Hory, and
Weber, were most helpful. We should like to give our thanks also to the
Brazilian Embassy in Washington, to Mr. George E. Seltzer and Mr.
Reginald S. Kazanjian of the U. S. Consulate at Para, and to Dr. Eladio
da Cruz Lima and a number of other residents of Para for facilitating
arrangements for the expedition. To Mr. Robert S. Mathews and Miss
Shelby Shackelford, members of the expedition, we are indebted for a
great deal of varied assistance.
1 Coates, C. W., R. T. Cox & L. P. Granath. The Electric Discharge of the Electric Eel,
Electrophorus electricus (Linnaeus). Zoologica, Vol. XXII (Part 1), No. 1, April 5, 1937.
204
Zoologica: New York Zoological Society
[XXIII :8
The observations made at the Goeldi Museum confirmed in general
and extended those made earlier in New York. A report of them has been
published in abstract.2 In the present paper some of these observations
are combined with others made later at the New York Aquarium. Whereas
our first observations at the Aquarium were almost wholly concerned with
the variation in the external voltage along the electric organs at different
instants during the discharge, the present work undertakes to determine
some of the electrical characteristics of the electric tissue and to make
some inferences concerning the process of electric discharge.
It has long been surmised that the electric organ of electric fish is in
some sort a battery of cells. Thus Faraday remarked of the torpedo:
“In concluding this summary of the powers of torpedinal electricity, I cannot
refrain from pointing out the enormous absolute quantity of electricity which the
animal must put in circulation at each effort. . . . These circumstances indicate
that the torpedo has power (in the way probably that Cavendish describes) to
continue the evolution for a sensible time, so that its successive discharges rather
resemble those of a voltaic arrangement, intermitting in its action, than those of a
Leyden apparatus, charged and discharged many times in succession.”3
There is a variety of other evidence pointing to the same conclusion.
The possibility of generation of electromotive force by the conversion of
mechanical energy, as in an electrostatic or an electromagnetic generator,
is precluded by the absence of any structure in the organ suitable to this
conversion and by the observation that no muscular action is required
for the discharge. The possibility that the discharge is generated by the
thermoelectric effect is similarly ruled out by the lack of suitable structure
and of the requisite differences in temperature.
There remain to be considered only electrochemical processes, and their
consideration here is rendered reasonable by the fact that the minute
electrical effects observed in nerve and muscle tissue have been convincingly
ascribed by recent research to such processes. The voltages observed in
nerves and muscles have been identified as concentration potentials. The
theory of their production is that the cell membrane is permeable to one
ion of an electrolyte and impermeable to the other ion. If then the electro-
lyte is present in different concentrations on the two sides of the membrane,
there will be a migration through the cell membrane of the ions to which
it is permeable. The region on the side of lower concentration will thus
acquire an electric charge of the sign of these ions and an equal charge
of the opposite sign will be left on the other side. The ions which pass through
the membrane will hinder the migration of other like ions by the electro-
static repulsion they exert on charges of their own sign, and in time an
equilibrium will be reached characterized by a definite voltage across the
membrane. This voltage is determined by the chemical valence of the
diffusing ion, by the relative concentrations of the electrolyte on the two
sides of the membrane, and by the temperature. These conditions being
known, the voltage is calculable. In the case of the boundary of a nerve
fiber, the relative concentration of potassium on the inside and outside of
the fiber, as found by Fenn, Cobb, Hegnauer and Marsh,4 would give rise
at room temperature to a concentration voltage of 118 millivolts. This is
much greater than the voltages actually observed in experiments with
nerve, but the means of observation are not such as to reveal the maximum
voltage.
The voltages observed in the discharge of the electric eel, on the other
hand, are very much greater than the concentration voltage found across
any single membrane. We have observed voltages as high as 500 between
the anterior and posterior ends of the large electric organ, and voltages
2 Cox, R. T. & R. S. Mathews. Journal of Applied Physics , Feb., 1938. (Abstract).
3 Experimental Researches in Electricity. Everyman’s Library Edition, London, 1922. Page 26.
4 Cited by Erlanger & Gasser. Electrical Signs of Nervous Activity, Philadelphia, 1937.
Page 134.
1938]
Cox & Coates: Electric Eel Tissue
205
above 300 are common. Such voltages as these must be the sums of many
concentration voltages. Each unit or electroplax of the electric tissue is
divided by a vertical partition, and we suppose this pai'tition is the boundary
across which the concentration voltage is developed, so that each electro-
plax is a single concentration cell. E. du Bois-Reymond5 gives the dimension
of one electroplax of the large organ in the direction in which the electric
polarity is developed as 0.1 mm., so that there would be about 100 cells to
the centimeter. If the cells in line along the organ are all joined in series,
the electromotive force (which is the voltage measured when no appreciable
current is flowing) of the whole series is the sum of the electromotive forces
of the separate cells. With 100 cells to the centimeter and 100 millivolts to
the cell, an electromotive force of 10 volts would be developed in one
centimeter. This is approximately what we measure near the anterior
end of the large organs.
It seems then altogether probable that the electromotive force of the
electric organs is that of a series of cells, in each of which there is de-
veloped a concentration voltage by the selective diffusion of ions through
a boundary. The cells in any cross-section of the electric organ we suppose
act in parallel. The electromotive force is not thereby increased, since a
number of similar cells in parallel have only the electromotive force of each
one of them. But when a parallel array of cells is joined to an external
conductor, so that a circuit is made and a current flows, the total current in
the external conductor is the sum of the currents in the cells. Thus we suppose
that the parallel action of the cells of the electric organ serves to produce
the large currents (of the order of one ampere at maximum) which we
have observed.
We have now to account for the release of energy by the electric organ
in brief transient discharges. Evidently the electromotive forces of the cells
and their connection in series cannot both be maintained all the time, for if
it were so, there would not be transient discharges, but at all times an elec-
tric current would flow in the water around the fish. Hence either the elec-
tromotive forces of the cells, or their connection in series, or possibly both,
must be transient, existing only for the duration of the discharge.
The structure of the electric tissue seems adapted to a transient series
connection rather than to a transient generation of electromotive force.
The discharge at any part of the organ is evidently governed by the nerve
branching to that part, for transsection of the nerve cord stops the progress
of the discharge down the organ at the point of transsection. On the other
hand, transsection of the electric organ without injury to the nerve cord
makes no appreciable difference in either the open circuit voltage or the
rate of propagation of the discharge. There is an ending of a nerve fiber
on the posterior face of each cell of the electric tissue. A fall of the
resistance of the sheath of a nerve fiber during the nervous impulse is well
established. It seems reasonable to suppose that, except during the dis-
charge of the organ, cells adjacent along its axis are effectively insulated
from each other by the high resistance of the nerve sheath, but when a
nervous impulse reaches the end of a fiber the drop in this resistance makes
an electrical connection between the posterior (electrically negative) face
of each cell and the anterior (electrically positive) face of the cell next
behind, so that there follows a progressive connection in series of the
cells all along the organ.
Several observations on the discharge give qualitative evidence that
the discharge is produced by a transient connection in series of cells with
constant electromotive force rather than by the transient generation of
electromotive force in cells permanently connected in series. If two points
on the large organ are connected to the oscillograph with the circuit open,
5 du Bois-Reymond. Emil. Dr. Carl Sachs Untersuehungen am Zitteraal Gymnotus electricus,
Leipsig, 1881. Page 49.
206
Zoologica: New York Zoological Society
[XXIII :8
so that no current flows except that which makes a circuit within the
tissue of the eel, one of the most striking features of the oscillographic
trace is the uniformity of the peak voltage throughout a long series of
discharges. In this case, since the current flowing in the electric tissue
must be rather small, the observed peak voltage must be near the full
electromotive force. Thus it appears that the electromotive force, at least
at the peak, is nearly the same in successive discharges. But when a circuit
is made through the organ and an external conductor of low resistance, so
that a large current flows, the peak voltages are observed to be less uniform.
In this case, the peak voltage is the electromotive force less the drop in
voltage caused by the dissipation of energy in maintaining a current through
the internal resistance of the tissue. It appears then that, while the elec-
tromotive force is nearly uniform at the peaks of successive discharges, the
internal resistance is somewhat variable. This would seem to lend some
support to the hypothesis that the electromotive force is uniform not only
at the peaks of discharges but also during and between them, the discharge
being produced by a variation in the internal resistance such as would
occur in the transient series connection of the cells.
Another evidence, more convincing than this, is obtained when the
oscillograph is connected to two points near the posterior end of the large
organ and the two ends of the whole organ are short-circuited by a low
resistance. The voltage recorded by the oscillograph is the electromotive
force of the segment included between the points of contact less the drop
in voltage caused by the current which the electromotive force of the entire
organ sends through this segment. The electromotive force of this segment
alone would make the anterior end of the segment positive with respect to
the posterior. The current flowing through the resistance of the segment
would cause by itself a voltage in the opposite sense. Under the action
of these two opposing voltages, the net voltage of the anterior end of the
segment wflth respect to the posterior end will be positive or negative as
the electromotive force in the segment is greater or less than the voltage
drop caused there by the current. Moreover, since the discharge runs
along the organ from anterior to posterior, the relation between these two
opposing voltages may be different at different instants during the dis-
charge, and the voltage recorded by the oscillograph may change sign, i.e.,
it may be diphasic. If so, the order in which the two phases occur will
be just opposite according to whether it is the electromotive force or the
resistance of the electric tissue which varies to produce the discharge. If
the electromotive force varies progressively from anterior to posterior
while the resistance remains constant, then, before the pulse of electro-
motive force has reached this segment near the posterior end, the current
caused by the electromotive force of the anterior parts will have started
in the entire organ and the negative phase of the discharge will be observed
in the segment. Only later, when the electromotive force has been gen-
erated in the segment and has declined in the anterior parts, will the posi-
tive phase occur. On the other hand, if it is the resistance which varies
progressively from anterior to posterior, the electromotive force remaining
constant, no large current will flow until the resistance of the whole organ,
including that of the entire posterior segment, has fallen to a low value. In
this case, the voltage across the posterior segment will first become posi-
tive, as its cells are thrown into series by the falling resistance, and after-
ward will become negative, as the impulse reaches the posterior end of the
organ, dropping the resistance there and so completing the circuit to allow
the passage of a large current. When the observation is made, it is found
that the voltage across the segment is first positive, rising sharply for a
very brief time, then sharply falling and reversing. Consequently the
hypothesis that the discharge is produced by a drop in the internal re-
sistance is favored.
1938]
Cox & Coates: Electric Eel Tissue
207
(a)
R E
Text-fig. 1.
(a) Electrical connections used in making oscillo-
graphic traces, (b) The same with a schematic
representation of the electrical characteristics
of the eel.
Another evidence of the variation of the internal resistance of the
organ during the discharge appears in the oscillographic traces reproduced
in Plate I, Figs. 1, 2 and 3. These traces were obtained with electrical
connections as shown in Text-fig. 1 (a). The eel was 123 cm. long. The
electrodes were in contact with the skin, one 23 cm. from the snout at
the anterior end of the large organs, the other 51 cm. from the snout, so
that there was included between the electrodes a convenient length of the
large organs and necessarily about the same length of Hunter’s organs, but
none of the organs of Sachs. The electrodes were made of aluminum plates
on rubber belts, which were fastened around the eel. They were designed
to have an area large enough to avoid a very high current density through
the skin. When the eel discharges in the water, the flow of current through
the skin is distributed over a large surface. With electrodes of small area
in a circuit of low resistance, the high current densities produced in the
discharge temporarily bleach and, if continued, seriously injure the skin.
208
Zoologica: New York Zoological Society
[XXIII :8
A circuit for the discharge was completed through a resistor and a battery,
which could be connected either to reinforce or oppose the discharge of the
eel through the resistor. It was necessary in making observations to con-
nect the resistor and battery only briefly in the circuit. The repeated dis-
charge of the eel through a low resistance not only produces the injury to
the skin already mentioned but also, of course, exhausts the electric organs.
The exhaustion shows itself, at least in eels of large size, not in an appre-
ciable reduction of the peak voltage on open circuit but in a reluctance of
the eel to discharge. The effects of connection to a battery are more
striking. On being connected through a low resistance to a battery of
voltage even considerably less than its own and with either polarity of
connection, the eel becomes restive. If the connection is maintained, the
peak voltage on open circuit declines and the form of the oscillographic trace
is markedly altered, the peak becoming much sharper. The alteration per-
sists for some time after the eel is disconnected from the battery. To
avoid it in these observations, our procedure was as follows: The eel was
made to discharge by gentle prodding. When it was discharging with
some regularity, the resistor was thrown into the circuit by a button
switch and the camera shutter was opened to photograph the screen of the
oscillograph. As soon as traces of the discharge were obtained, the shutter
was closed and the button switch released. Since the shutter was always
open for the interval of several discharges, the traces of these discharges
overlap on the photographs.
In Plate I, Figs. 1, 2 and 3, the resistance was 1,000 ohms. In Fig. 1
the battery voltage was 90 reinforcing the discharge of the eel. In Fig. 2
there was no battery voltage. In Fig. 3 the battery voltage was 90
opposing the discharge of the eel. A constant voltage between the terminals
of the oscillograph does not show on the oscillographic trace. If the re-
sistance of the electric tissue were constant during the discharge, the
battery would add only a steady voltage to that of the electric tissue. If
then the electromotive force of the electric tissue varied during the dis-
charge in the same way for each of the three connections, the same trace
would be obtained each time. On the other hand, if the internal resistance
of the electric tissue varies during the discharge, the current produced in it
by the battery will vary and also the contribution of the battery to the
recorded voltage. Thus different traces will be obtained with the three
connections. It will be seen that the traces in Plate I are different, the
voltage being raised when the battery reinforces the electric organ and
lowered when the battery opposes the electric organ. Hence we infer that
the resistance varies in the discharge.
In Plate I, Figs. 4 and 5, as in Fig. 2, there was no battery voltage.
Plate I, Fig. 4, is the trace of the discharge on open circuit, so that the
external resistance was effectively infinite. In Fig. 2 the resistance was
1,000 ohms and in Fig. 5 it was 300 ohms. The current flowing in the
organ is increased as the external resistance is lowered, and the drop
of the observed peak voltage below the electromotive force is consequently
also increased.
Using various battery voltages and external resistances, we have ob-
tained traces of a large number of discharges, and from measurements on
these we have undertaken to calculate for different instants during the
discharge the values of the internal resistance of the segment of the electric
organ employed in the observations, assuming the resistance changing and
the electromotive force constant. For this calculation it was necessary to
make some assumption concerning the leaking of current through the tissue
of the eel adjacent to the electric organ. The paths followed by this
current could doubtless be accurately represented in diagram only by a
complicated network, but for purposes of rough reckoning we have employed
the simple scheme shown in Text fig. 1 (b). In this figure, E denotes the elec-
1938]
Cox & Coates: Electric Eel Tissue
209
tromotive force of the electric tissue and R its resistance, which we assume
changing throughout the discharge. The leakage path is represented by a
single resistance r, shunting the electric organ. The electromotive force of
the battery is denoted by E' and the external resistance by R'.
At the junction of the three resistances, the current, I, in the electric
tissue unites with the current, I', in the external part of the circuit to pro-
duce the current, i, in the leakage path, and we have:
i = I + I'
In the circuit comprising E', R', and r, there is a drop in voltage I'R'
through the resistance R' and a drop ir through the resistance r. These
must together be equal to the electromotive force E' of the battery, for
around the whole circuit the net drop in voltage must be zero. Hence we
have:
I'R' + ir = E'
Similarly for the circuit comprising E, R, and r, we have:
IR + ir = E
With these three equations we can eliminate the two currents, I and I',
and obtain for the third current, i, an expression in terms of the electro-
motive forces and resistances alone. This expression is:
ER' + E'R
1 RR' + rR + rR'
Except during the discharge, we suppose that R is very large. If this
is so, the only appreciable current flowing when the organ is not discharging
is that produced by the battery in a circuit closed through the external re-
sistance R' and the leakage resistance r. If we denote by io the current flow-
ing in r when the organ is not discharging, we have:
E’
10 — r + R'
The voltage between the terminals of the oscillograph is ir-E' during
the discharge and ioii — E' between discharges. (The resistance of the bat-
tery is low enough that the voltage between its terminals may be regarded
as constant and equal to E'.) Hence the variable part of this voltage is
ir — ior. This is the voltage as recorded on the oscillographic trace. If we
denote this voltage by V, we obtain from the two preceding equations:
V:
i -r ^ ^ ^
(RR' -)- rR + rR' R' +r
Finally, if we solve this last equation for R, we obtain:
R =
rR' JE — V
r + R'\ V
E' r \
V r+R'J
In this expression, E' and R' are known, being chosen at will by the
connection of different batteries and resistors in the circuit, and V is the
voltage measured on the oscillographic trace for any instant during the
discharge. E, the electromotive force of the segment of the electric organ,
and r, the leakage resistance of the adjacent tissue, are not directly given
and values must be sought which will bring into the best possible agreement
the values of R calculated for the same instant from different traces. In
the calculations E was taken as 260 volts and r as 3,600 ohms.
The values of R calculated from a number of traces obtained with
widely varying electrical connections are shown in Text-fig. 2. Each plotted
point denotes an average of measurements made on the traces of from four
to thirteen discharges. The traces of discharges obtained with the same
connections are sometimes rather widely variant. Moreover the measure-
ments could not be made with accuracy, by reason of the lack of high preci-
RESISTANCE, OHMS.
210
Zoologica: New York Zoological Society [XXIII :8
Text-fig. 2.
Kesistance of a segment of the large electric organ at different instants during
the discharge, calculated from voltages observed when the organ discharges
through various circuits.
Open circles . .
Crosses
Open triangles
Open squares
Solid circles. . .
Solid triangles
Solid squares .
open circuit
external resistance, 1,000 ohms] v.Q+tOT.„
300 ohms [■ voltaee^
battery voltage, 90 volts reinforcing discharge] External
157.5 volts opposing discharge j- resistance,
202.5 volts opposing discharge] 1,000 ohms.
sion of the apparatus and also of the impossibility of finding at all closely
on the trace the point corresponding to the instant at which the discharge
1938]
Cox & Coates: Electric Eel Tissue
211
began. It will be observed that the extreme values of the resistance calcu-
lated for any instant during the discharge differ in most cases by a factor
somewhat greater than two. On the other hand, in the observations from
which the calculations were made, the net electromotive force acting was
varied by a factor of about six and the effective resistance through which
it acted was varied by a factor of about forty. Considering these wide
variations in the experimental conditions, the inaccuracy of the observa-
tions, and the extreme crudity of the schematic representation in assuming
the leakage current as flowing through a single resistance, we think the
discrepancies in the results are not significant and we consider that the
results confirm the general hypothesis from which the calculations were
made. Our conclusion is then that the electromotive force of the electric
tissue is almost certainly that of a series of concentration cells, that the
discharge is very probably produced by a transient drop in the internal
resistance of the electric tissue, and that this drop is probably an action of
the nervous impulse at the ends of the nerve libers which changes a virtual
insulation between adjacent cells into an effective connection.
Two other observations seem worth noting. One of these is that the
electric tissue is rectifying, that is, during the discharge, when its resis-
tance to an electromotive force in the same direction as its own falls
markedly, its resistance to an electromotive force in the opposite direction
remains high. As the opposing electromotive force of the battery shown in
Text-fig. 1 is increased, the discharge diminishes in voltage and at a certain
value of the opposing electromotive force disappears altogether. A further
increase in the opposing electromotive force, up to some 90 volts at least,
produces no additional effect; discharge is not reversed.
The other observation concerns the intermediate discharge. Since the
major and minor discharges can be definitely attributed to the large organs
and the organs of Sachs, and since the only other electric organs are those
of Hunter, it seems natural to associate the third type of discharge with
these. However, our observations on the discharge through various resis-
tances indicate that this discharge has a power altogether out of proportion
to the bulk of Hunter’s organs. These organs, while nearly as long as the
large organs, have a very much smaller cross-section. Consequently their
internal resistance should be much higher and the voltage they could develop
externally across a low resistance should be only a very small fraction of
that developed in the major discharge. Plate I, Fig. 3, shows one and prob-
ably two intermediate discharges, those of the lowest and next lowest peak
voltages. (They are distinguished from the major discharge by their more
gradual fall as well as by their lower peak voltage.) The peak voltages of
these two discharges, while definitely lower than those of the major dis-
charges, are still relatively about as large as on open circuit. Consequently
it would seem that only the large organs would have enough power to pro-
duce the intermediate discharge. It may be that the intermediate discharge
is a discharge of the large organs modified, in some way of which as yet we
have only a very vague idea, by the simultaneous discharge of Hunter’s or-
gans. On open circuit, the intermediate discharge has sometimes a spur
before the main peak. This spur may show the discharge of Hunter’s
organs.
212
Zoologica: New York Zoological Society
EXPLANATION OF THE PLATE.
Plate I.
Oscillographic traces of the major discharge of a segment of the large organ
with various electrical connections.
Fig. 1. Battery voltage, 90 volts reinforcing discharge] External
Fig. 2. No battery voltage j- resistance,
Fig. 3. Battery voltage, 90 volts opposing discharge j 1,000 ohms.
Fig. 4. Open circuit.
Fig. 5. External resistance 300 ohms, no battery voltage.
COX & COATES.
PLATE 1.
CO
CD
ELECTRICAL CHARACTERISTICS OF THE ELECTRIC
TISSUE OF THE ELECTRIC EEL.
ELECTROPHORUS ELECTRICUS (LINNAEUS).
i/loaScLruj -th s of asecon d .
Masterson: Colonial Rattlesnake Lore, 1714
213
9.
Colonial Rattlesnake Lore, 1714.
James R. Masterson
Hillsdale College
The subjoined document,1 here printed for the first time, is in all
likelihood the fullest account before the Nineteenth Century of the charac-
teristics and habits of rattlesnakes. The author was a Captain Walduck
(probably Thomas Walduck2), who at an earlier time had sent to his friend
James Petiver/1 apothecary and Fellow of the Royal Society of London, eight
letters concerning the trade, government, society, and natural curiosities of
Barbados.4 While at Barbados he had made wistful efforts to bring himself
to the attention of the Royal Society; and in his “Account of the Rattle-
snake,” read before the Society January 7, 1714, he achieved his ambition.
From a reference in the manuscript it appears likely that Walduck sent it to
Petiver from New England.
The whole paper, presented with due solemnity before the most dis-
tinguished scientific society of the time, is an extraordinary collection of
folklore pertaining to a creature almost as fictitious as the basilisk, the
hydra, or the unicorn. One can imagine the honest Captain, full of inquiry
and thirsty for scientific glory, noting down for the Royal Society of London
the mendacities of Indians and fur traders. He displays no less assiduity in
this task than his distinguished contemporary, the Reverend Cotton Mather,
D.D., who in 1712 had contributed snake lore to the archives of the Royal
Society. Why Captain Walduck’s letter was not similarly honored by
inclusion in the Society’s Philosophical Transactions is not clear; perhaps
the editors doubted the authenticity of the Captain’s information.
Capt Walducks Acc* of ye Rattle Snake Read before
ye Royall Society [word illegible] Jan: 7th. 171J
The most pernicious Creature in ye English Empire upon ye Main of
America is the Rattle Snake, as well for their Number as Effect, whose bite
is as mortall as Fate, unless ye part be imediately cut out : beleiving it may
not be unpleasant to you, I will describe as many of it’s Qualities as I can,
1 The MS. forms Number 21 (ff. 113b-116b) in Sloane MS. 3339 (British Museum), which is
Volume VIII of the collections of James Petiver from 1691 to 1717, entitled Adversaria. See Samuel
Ayscough, A Catalogue of the Manuscripts Preserved in the British Museum (London, 1782), II,
653; E. J. L. Scott, Index to the Sloane Manuscripts in the British Museum (London, 1994), p. 556.
2 Tho. Waldock was appointed Register of the Court of Admiralty for the Bahamas, February
25, 1697 ( Calendar of State Papers, Colonial Series : America and West Indies , 15 May, 1696-31 Oct.,
1697, ed. J. W. Fortescue {London, 1904}, p. 383). The Minutes of the Council of Barbados, March
16, 1703, refer to a lawsuit of Thomas Walduck ( Calendar of State Papers, Colonial Series: America
and West Indies, Dec. 1, 1702-1703, ed. Cecil Headlam {London, 1913}, pp. 265,771).
3 Petiver’s name appears frequently in Philosophical Transactions (see particularly XXII, 795).
Regarding his collections see Sir Hans Sloane, A Voyage to . . . Jamaica (London), II (1725), iv-v.
4 Of these the first appears to have been lost ; the remaining seven are preserved in Sloane MS.
2302. The fourth is dated October 29, 1710, and the last September 17, 1712. In a letter from
Barbados dated November 12, 1710, Walduck speaks of himself as having passed fourteen years in
the West Indies, “and I support my self with the hopes of Seeing England once More ” (Sloane
MS. 2302, f. 16).
214 Zoologica: New York Zoological Society [XXIII :9
leaving ye Descriptions of its Parts to Surgeons & Anatomists. He is bred
from an Egg an Inch & half long, & about ye bigness of a Mans little
finger; ye Shell is of ye Colour of Parchment, & when first laid near as Tuff,
& what surprized me, I found some of these Eggs once, & they hapned to
be near Maturity to hatch, & I brake them with a Stick, & out ran a
Snake all perfect, att least 7. inches long, & as big as a Goose-Quill, for they
are quoill’d oblong in ye Shell, their head in ye middle & their Tayl on ye
outside, & when they hatch ye Shell groweth dry, & they break it with their
heads & come out: In Carolina & Virginia they engender in ye Month of
Aprill (when ye Sun is near ye Tropick of Cancer, & warms ye Earth, &
vivifies those dead Animalls buried in ye Winter) for a fortnight or 3 weeks
together: In New-England & Pensylvania not untill May, or June; when
they [ folio 114] engender some Thousands of them gott together (I heard
a man say he saw a heap as big as a hhd) some flying about others hissing
& makeing a horrible noise, in a strange permiscuous manner — And this
to me obviates a vulgar Opinion that there is a king among ye Snakes, in
whose head is a Carbuncle that shines in ye night with wonderfull Lustre,
& this is only ye Agrigation of their Eyes, by such a number getting
together, their Spirits being high & volatile & full of venom, that by ye
Reflection of water, or some such Accident gives such a light in ye Night as
Travellers love to report) a little time after they lay their Eggs at ye Roots
of dry Trees, amongst leaves & dry moss, where ye heat of ye Sun & Earth
vivifyes & hatches them, & in August young Snakes have been seen ; their
Eggs are laid like a chain 8. 10. 12. together if they lay but once or
oftner is uncertain : at a year old they have a Rattle in their Tayls, &
about 2 foot long & as bigg as a Mans thumb, & then they come to be deadly
poysonous, so they shall grow to be 8. 10.«12 foot long & as bigg as a Mans
thigh, & every year increases a Rattle (some belive they grow like ye
Alligator as long as they live) there have been some kill’d with 24 Rattles
in them, & it is beleived they do not exceed that age, they live upon
Squirrells & all sorts of Birds that they can catch, & by a wonderfull
Facination they shall charm them, both Squirrells & Birds into their
mouths, from ye Top of a Tree 50 foot high, for they are a Slothfull
heavy Creature (to other Snakes) otherways they could not gett their prey,
they swallow all whole, for their bite is as deadly to themselves as to those
they bite, as soon as they bite they apply themselves to a certain Root
w:ch they eat & is an Antidote to their own poyson, by ye want of which
they shall dye in an houres time by their own poyson [ye Indians [folio
1145] & some of ye English that trade amongst ye Indians know this Root &
keep it always about them, & when they are bitt, they chew this Root &
apply it to ye wound, by w:ch means ye Poyson will be expelled: ye Indians
likewise have another Notion, or rather a peice of superstition as a Remedy
agJ ye bite of ye R: Snake that by painting a R: Snake upon their Bodyes,
they beleive ye Rattle Snake will not bite them, & several both Indians, &
White men have Snakes drawn with black strokes most curiously upon their
bodies: We may Quest:11 whither this might not be taken from ye brazen
serpent in ye Wilderness, & handed down to them by Tradition, or whither
there might not be some Magick in it, for all Diabolical Magick came from
Natural Magick (as all Idolatry is a Corruption of ye true Religion), as
soon as they bite they run away, & they never bite but once, ye Experiment
of their killing themselves by their own bite hath been tryed thus : An
Indian hath been hired to bring in a Rattle snake (for there be some of
them, & white men too, that will take up a Rattle snake as one would
take up an Eel, & it hath been put into a Cask standing up with one head
out, & ye Snake hath been provok’d by a stick to bite ye Stick, which it
hath done, but could not be provoked to bite ye 2d time, & in an hour after
hath died for want of this Root. Another Experiment of ye virulence of
its Poyson hath been this a Snake hath been confined as above & by green
switches their Rind being pealed off, ye Snake hath been provoked to bite ye
1938]
Masterson : Colonial Rattlesnake Lore, 1714
215
end of ye switch, & ye poyson hath been observed to run up ye switch, &
turned ye green of ye switch \folio 115] to a blackness all ye way up to ye
Top, this hath been tryed some score of times by several hands, & it is ye
Nature of ye Poyson to ascend ; for if a man be bitt by ye Snake in ye Arm,
or Leg, left him hold that part up higher than his body he shall live some
days, but if it be below he shall not live 2 houres, but their body shall
break full of large Chapps, & Crakks as if their Skinn had been roasted
by ye Fire. The R. Snake cannot bite running, he must be quoiled round
with his head in ye middle & then he will rise as high as his navel, & no
higher & so chop to bite, Dogs that are used to them shall make them kill
themselves by running round & baying at them, that by chopping att
ye Dogg they have bitt themselves. Their Poyson lieth in som little bladder
att ye Root of their two lower Tusks, which when they bite breaks & is
squeaz’d out, & their is two socketts in their upper jaw to receive them,
they never make use of them in their ordinary feeding but swallow all
whole, their Stomach’s are of a large Capacity young Fauns being taken
out whole. The Indians never are in fear of them, nor will they be
perswaded to kill any of them. The Indians believe that ye R. Snakes are
ye Executors of ye Divine Vengeance, & that he that is bitt by them hath
been a Murderrer, w :ch likewise was the Opinion of ye Inhabitants of Maltha
when S :l Paul was shipwrek’d there according to ye sacred Text. Indeed
they never do bite but when provoked, & a Man may as securely sleep in ye
wood for them as in a feather bed. [folio 115b] They never bite but first
make a Rattling with their tayls, & they may be heard 20 yards of. Those
Men y‘ use ye Woods say they never come near a R. Snake but they are
aprised of it before they either hear or see them, & they are in a fright as
tho’ a Spectre was near them, & that their breath inflames ye Air & before
they either hear or see them they are seized with sorrow. So that ye
Fiction of Phyton is verify’d in ye Rattle Snake, or they are all one Crea-
ture, they are locked up all ye Winter in ye Earth, & when they first come
forth in ye Spring they are weak & have lost all their Stock of poyson, &
their bite is of no danger, untill by feeding & warm weather they are
supply’d with fresh venom, & in a Month’s time their bite shall become
Mortall, & their venom is deadly lett it be lodg’d any where so long as ye
Snake lives but not longer Experienced by ye Following History: A Man in
Virginia being in ye Woods having a pair of boots on was bitt by a RSnake
thro’ his boot came home to his wife & dyed, his Boots were hung up in
ye house, & his widdow marryed a second husband, who putt on those Boots,
& by rideing a Small Journey complain’d with a pain in his Leg & likewise
dyed, the Boots were hung up again, & ye Woman married ye third
Husband, who made use of ye same Boots, & ye first time of wearing
them complained as ye former of a small tumor in his legg & likewise died;
Upon this ye Surgeon cutts ye boot in peices & found ye Tooth of ye Rattle
Snake that bitt ye first Husband, & did all ye Execution since as small as
a hair, they took it out with a [folio 116] pair of Forceps, & prick’d a dog
with it that within few hours dyed, ye Surgeon took ye tooth some time
after tryed it upon another Dog & it did him no hurt, & it was supposed
then ye Snake was dead. Wherever ye English come & settle ye Snakes
leave ye place in a little time, which keeps up ye superstition of ye Indians
that there is some extraordinary virtue in ye English, when I can conceive
no other Reason in it but by planting & weeding ye Earth, plowing &
breaking up ye glebe, they eradicate ye Root out of ye Ground that is ye
Antidote & self preservation to ye Snake against his own poyson, & there-
fore they are forced to leave ye place & go where it is to be found. They
are not all over America, not to ye Southward of ye Tropick & to ye
Norward of 55°, whither ye heat destroys them in one & ye Cold in ye
other I am not Judge, there is none in ye Kingdom of Mexico amongst ye
Spaniards, but they say they were once all over there & drove away by ye
prayers of one of their Fryers; this is one of their Legends. We may say ye
216
Zoologica: New York Zoological Society
[XXIII :9
same thing 100 years hence in N. England, a Man may live there 20 Years
now & not see one, & in some Ages a RSnake will be as great a Rarity in
N. England as a Wolf is now in Old England, tho’ they were once very
populous in each Country. There be Water R. Snakes as well as Land &
as deadly in some Rivers in Virginia they shall creep up ye Calles of Shipps,
& assault a boat when near ye Shore, there’s aboundance of Strange Reptiles
in ye Woods in Virginia & Carolina Viper about 2 foot long more deadly than ye
R. S. [ folio 116b] but not so Common, & for whose bite ye Indians them-
selves know no remedy, & of which they stand very much in fear, there be
likewise in ye Woods of Carolina Snakes made like a Coach whip as long &
as small, that will twist their head round a horse’s Leg, & with their Tayl
Lash a horse with great Violence untill ye blood comes, there is like wise
a Glass Snake about 2 foot long, & as green & as brittle as Glass, that if they
happen to fall off a bough they will break as short as glass, & with a little
switch it is more easye to break one of them than a Tobacco pipe, & what
adds to ye Wonder they never bleed.
This is what I have learned (having particularly enquired) of ye Indians,
& white Men conversant & Tradeing amongst ye Indians, & beleive most
of this is Matter of Fact.
1938]
Notes
217
NOTES
External Parasites of the Electric Eel, Electrophones electricus (Linnaeus).
Two external parasites have been found so far on the electric eel, Electrophones
electricus. On one eel, measuring about five feet in length, numerous “anchor”
parasites (copepods) were found embedded in the thick skin along the dorsal sur-
face of the body. The copepods were identified as Lernaecocera cyprivmcea L.
and have been previously reported by many investigators from several species of
cyprinid fishes. On September 30, 1937, five of the smaller eels (ranging in size
from 18" to 2'6") were found to be heavily infected with leeches. More than 2,375
parasites were removed from one eel measuring 24% inches in length. These
leeches are the common Placobdella parasitica (Say) which are usually found
on the soft parts of turtles, but which also have been reported on other fishes.
When bearing eggs or young, the leeches often leave the host for a time and lead
a free life underneath stones, and feed on worms and larvae.
The interesting thing about these infections is that the parasites were able
to withstand the electric discharges of the eels. These discharges, according to
Coates, Cox and Granath (1937 11 may reach a voltage of 300 in eels exceeding
50 cm. in length. One 11%" eel gave off about 200 volts (Coates and Cox, 1936) 2.
It is definitely known that the eels were discharging at the time the parasites
were present. — Ross F. Nigrelli, New York Aquarium.
1 C. W. Coates, R. T. Cox, and L. P. Granath (1937). Zoologica 22: 1-32.
2 C. W. Coates and R. T. Cox (1936). Zoologica 21 : 125-128.
jHeto fforfe Zoological Society
General Office: 90 Broad Street, New York City
©fficenJ
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 $arb
W. Reid Blair, Director
Raymond L. Ditmars, Curator of Mammals and Reptiles
Claude 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
Squarium
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 tropical fte&arci)
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
Lee S. Crandall, Curator of Birds
Charles R. Schroeder, Veterinarian
W. Reid Blair
William Beebe
Charles M. Breder, Jr.
William Bridges
ZOOLOGICA
SCIENTIFIC CONTRIBUTIONS
OF THE
NEW YORK ZOOLOGICAL SOCIETY
VOLUME XXIII
Part 3
Numbers 10-16
PUBLISHED BY THE SOCIETY
THE ZOOLOGICAL PARK, NEW YORK
September 28, 1938.
CONTENTS
PAGE
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-
tro phorus 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
Bell: Hypophysis of the Goldfish
219
10,
Morphology of the Hypophysis of the Common Goldfish
(Car as sius auratus L.).1
W. Randal Bell
Washington Square College, New York University
(Plates I & II; Text-figures 1 & 2).
There is much speculation concerning the possibility of pituitary-
gonadal relationships in fish similar to those found in mammals. Implants
and extracts of fish pituitaries have been found to cause a premature ovula-
tion in fish, (Houssay, 1930, von Ihring, 1935) and an increase in gonad size
(Cardoso, 1934). Noble, Kumpf and Billings (1936) have induced brooding
behavior in the Jewel fish ( Hemichromis bimaeulatus ) with injections not
only of fish pituitary extracts but of sheep extracts and prolactin as well.
Wunder (1931), Fleischman and Kann (1932), and Ranter (1934), have
found the color and ovipositor length of the bitterling ( Rhodeus amarus )
to be affected by hormones from the fish pituitary but not by gonad stimu-
lating hormones from the pituitary of mammals. Smith (1931), Zondek
and Krohn (1932), Matthews (1933), Parker (1934), Zondek (1935), Hog-
ben (1936), and numerous other workers have studied the pigmentation
effects caused by changes in the amounts of intermedin and other “chroma-
tophorotropic” hormones. The effects of the physiological removal of the
pituitary have been observed by Lundstrom and Bard (1932), Matthews
(1933), Parker (1935), and Smith, Burr and Ferguson (1935). Recently
Abramowitz (1937) has reported the successful removal of the pituitary of
the catfish and the killifish, using an opercular approach. Examination of
the goldfish gives reason to believe that such an approach would be effica-
cious in its hypophysectomy. However, an improved technique for this opera-
tion, especially one which would make possible a partial ablation, would be
highly desirable.
It is the purpose of this investigation to demonstrate the anatomy and
histology of the pituitary and its surrounding structures in a form of fish2
well adapted to experimention. Carassius auratus (the goldfish) was selected
because it is highly adaptable to laboratory conditions and has an excep-
tional resistance to surgical shock. Also, its compact pituitary is suspended
by a stalk which passes through a connective tissue membrane, covering
a foramen in the brain-case floor. These protective structures should make
it quite feasible to remove the gland without injury to the brain. It is
hoped that the observations presented here will provide an adequate founda-
tion for future investigations.
At this time, I wish to acknowledge the thoughtful guidance of Dr.
Harry A. Charipper in the accomplishment of this work. I would also like
1 Accepted in partial fulfillment of the requirements for the degree of Master of Science, New
York University.
2 A resume and key references on the pituitary of various teleosts may be found in the
symposium report of H. A. Charipper (1937).
220
Zoologica: New York Zoological Society
[XXIII :10
to thank Mr. C. M. Breder, Jr., of the New York Aquarium for providing
the laboratory facilities and for his helpful advice on the ichthyological
aspects of the problem.
Materials and Methods.
The specimens of Carassius auvatus were obtained from commercial
fisheries in Saddle River, New Jersey. Their over-all length varied from
seven to eight inches, while their body length from the tip of the snout to
the proximal border of the tail fin averaged five inches. Material for histo-
logical preparations was fixed between the end of December and the latter
part of March. Most of this was from female animals.
For convenience, the fish were removed from the large main tank into
small individual tanks before being used. All tanks were supplied with ac-
tively flowing tap water, which maintained a temperature of about 8° Centi-
grade. This water was cold enough to prevent any active maturation of the
eggs in the ovaries.
The fish were taken from the water and rapidly anaesthetized by flood-
ing the gills with chloroform. The roof of the cranium was removed first by
cutting transversely through the dermal bone between the eyes, and con-
tinuing the incision through the dorso-lateral angles of the neurocranium.
The entire roof was then lifted and broken away at its posterior margin.
The neurocranium was then freed posteriorly by a transverse incision in the
region of the medulla, and anteriorly by another incision midway between
the olfactory lobes and the olfactory bulbs. Finally the entire neurocranium
was separated from the rest of the animal by cutting through the opercular
attachments and separating the gill bars from their cartilaginous attach-
ments on the neurocranium. The optic nerves and muscles were carefully
cut to avoid altering their relation to the pituitary. The neurocranium with
the brain and pituitary were removed to the fixative. The whole procedure
was accomplished in about three minutes.
Helly’s, Bouin’s and Zenker’s fluids were employed for fixation. At first
the crania from anesthetized fish were immersed in these for about seven-
teen hours. Later it was found more desirable to kill the animals directly
by injecting the fixing fluid through the thin medial wall of the olfactory pit
into the cavity above the brain. The neurocranium was then rapidly re-
moved as before and more fluid injected into the myodom near the gland
before immersion. The fixation time was reduced to six hours by this pro-
cedure.
When the pituitary was to be sectioned alone, it was carefully removed
from the cranium after fixation. All attempts to remove the brain and
the pituitary together were unsuccessful, due to the fragile nature of the
pituitary stalk and to the intimate relation of the endocranial membranes
and the bone floor to the pituitary body. Specimens designed for sectioning
of the neurocranium were decalcified for four hours in 10% solution of
nitric acid in 70% alcohol. Sections of the pituitary were cut from 5-7
micra in thickness and those of the neurocranium at 10-15 micra. Serial
sections, longitudinal, horizontal and transverse, were made of the pituitary
in situ in the neurocranium and also of the gland alone. The Masson stains
were used for cytoplasmic differentiation and Delafield’s haematoxylin and
eosin as a routine procedure. The latter technique was used for the ana-
tomical preparations.
Relation of the Pituitary Gland to Surrounding Structures.
The pituitary body of Carassius auratus lies deeply embedded in the
neurocranium. It does not occupy a sella turcica in the basisphenoid bone
of the cranial floor as is the case in higher vertebrates. Instead, it lies in
a tunnel-like cavity, the myodom, below the floor of the brain case (Plate I,
1938]
Bell: Hypophysis of the Goldfish
221
Fig. 1). This space is enclosed in the apex of the V-shaped neurocranial
floor, just above the prominent parasphenoid or “keel bone” (Plate I, Fig. 4).
The myodom extends posteriorly from the optic foramina to the region of
the medulla, the pituitary lying near its anterior end.
The lateral walls of the teleost neurocranium in the region of the
pituitary are formed by the prootic and alisphenoid bones. The alisphenoid
bones extend forward and medially to form the walls and floor of the brain
case proper in the region of the orbit. Posterior to the pituitary, the roof
of the myodom forms the floor of the brain case. This is made up of two
horizontal plate-like processes, extending medially from the prootic bones
and fusing in the midline, forming the “prootische Briicke” of Holmgron
and Stenzio (1936).
When viewed from above, a large oval optic-pituitary foramen may be
observed in the floor of the brain case. The posterior rim of this is formed
by the prootic bridge and the anterior, more dorsal rim by the alisphenoid
bones (Plate I, Fig. 1). Its lateral edges are formed by delicate bony pro-
cesses projecting from the lateral walls. A thin, tough extension of the
endocranial membrane completely covers the foramen except where it sur-
rounds the optic nerves and pituitary stalk as they enter the brain case
(Plate I, Fig. 3). A sac-like fold of this membrane is reflected over the
pituitary body, which lies immediately below the posterior half of the
foramen. The optic nerves enter the myodom through the optic foramina,
and then pass through the anterior half of the optic-pituitary foramen to
reach the brain.
The pituitary body is placed dorsally in the myodom (Plate I, Fig. 4)
above the rectus externus muscles of the eye, which enter this bony chamber
ventral to the optic nerves. In transverse section, the muscles appear as two
oval masses ventro-lateral to the pituitary. A medial bony septum rises from
the parasphenoid a short distance between them.
The pituitary lies with its dorsally flattened surface pressed against
the membrane covering the optic-pituitary foramen. Its posterolateral edge
is enclosed by a cupping of the rim of the prootic bridge (Plate I, Fig. 3).
Likewise the lateral edges of the foramen project past the sides of the
gland. (Plate I, Fig. 2). The above three structures effectively separate
the pituitary from the cavity of the brain case.
As a result of this separation, the pituitary is suspended from the brain
by a short and heavy stalk. This solid process is about twice as wide
as it is long in a transverse section (Plate I, Fig. 2). It rises in the mid-
line from the floor of the hypothalamis at the posterior end of the lobus
medialis. The latter is a small eminence of the lamina post-optica lying
between the anterior ends of two large, oval lateral lobes, the lobi inferior
(or mammillary lobes). These latter bodies are over-shadowed dorsally by
the larger optic lobes (Plate I, Fig. 4). In a transverse section, the narrow,
medial, slit-like cavity of the third ventricle may be seen to penetrate the
ventral infundibular region for a short distance but it does not extend into
the pituitary stalk.
No large blood vessels were observed entering the pituitary body, al-
though a large vein occupies the angle formed by the posterior indentation
of the pars intermedia (Plate I, Fig. 3). The membranes surrounding the
gland contain a profusion of small vessels, which were frequently observed
to penetrate the pituitary, especially in the region of the pars anterior. The
gland, as well as the brain itself, lies in a semi-fluid fatty mass.
Anatomy of the Gland.
The pituitary in Carassius is an irregular, solid, rounded mass about
1.3 millimeters in length. Its form can best be appreciated by consulting
222
Zoologica: New York Zoological Society
[XXIII :10
Text-fig. 1 showing sketches of its various aspects. From above (Text-fig.
1, B) it is almost circular with a flattened posterior side. A slight crevice
connects the stalk with the posterior extremity. Ventrally and laterally
(Text-fig. 1, A and C), it shows a rounded cone-shaped body projecting
from the main mass of the gland. The apex of this cone-like portion of the
intermedia points posteriorly and somewhat ventrally. The separation of
this portion from the remainder of the pars intermedia is not always so
pronounced.
The posterior aspect (Text-fig. 1, D) is almost triangular with the apex
being formed by the cone just described. The dorsal, more massive, rec-
tangular portion represents the ubergangsteil and the remainder of the in-
termedia. This latter is faintly outlined as a central triangular portion.
The pars intermedia is the most ventral portion and occupies somewhat
less than one-half the volume of the gland. A crevice may be seen to sepa-
rate this lobe from the ubergangsteil everywhere except at the very core of
the gland. In addition to the cone-shaped, antero-ventral portion the inter-
media is made up of a more dorsal plate (Text-fig. 2, and Plate I, Fig. 3),
lying intimately in contact with nearly the whole ventral surface of the
ubergangsteil. Anteriorly this plate has a flat dorsal surface, but posterior
to the center it rises medially to form a triangular cross section. In some
cases the plate disappears entirely from transverse sections taken near the
posterior extremity.
Drawings of various aspects of the hypophysis. (A) Ventral aspect, (B)
Dorsal aspect, (C) Lateral view and !D) Caudal view. In A, B, and C
the nasal end is to the left. Approximately x 12.
1938]
Bell: Hypophysis of the Goldfish
223
The ubergangsteil is formed by a deep plate of cells, indented antero-
dorsally by the pars anterior and postero-ventrally by the ridge of the
intermedia. The pars nervosa forms a solid core-like mass in this lobe
posterior to the midpoint (Text-fig. 2, and Plate I, Fig. 3). Branching
root-like processes extended from this into all the other lobes but especially
into the pars anterior and the intermedia. A heavy process curves ventrally
toward this latter portion in the median plane and on reaching it, broadens
out into a mass which sends processes radiating throughout the lobe.
A diagrammatic reconstruction of a median sagittal section of the pituitary
body showing the anatomical and area relationships of the various
portions. P.A. — Pars anterior; P.N. — Pars nervosa; P.l. — Pars inter-
media; Ubg. — Ubergangsteil. Approximately x 15.
The pars anterior (Text-fig. 2, and Plate I, Fig. 3) may be con-
sidered to consist of two portions; first, a flat plate-like region anterior to
the pituitary stalk and dorsal to the ubergangsteil — the pars anterior proper;
and second, a layer of peculiar cells enveloping the dorsal nervosa mass and
the roots of the processes which radiate from it (Plate I, Fig. 5).
Histology op the Pituitary Gland.
Pars Nervosa. This tissue appears as a loose fibrous mass with scat-
tered neuroglia nuclei and occasional glandular cells (Plate I, Figs. 2, 3, 5).
The coarse basophilic fibers run lengthwise in the cords, as do the nuclei of
the neuroglia cells. These latter are fusiform or irregularly elongate in shape
and are characterized by scattered coarse granules. They are stained red
with the Masson technique. Occasionally an area appears around the nuclei
which stains as would cytoplasm, but no cell borders are visible. Numerous
large basophilic glandular cells, resembling those of the ubergangsteil, are
found in the nervosa masses but not in the cords. These are always de-
generate and often appear as irregular blue masses. Eosinophilic colloid
material was observed in rare cases only. Numerous blood vessels penetrate
the cords of the nervosa and occasional large vessels appear in the dorsal
mass.
224
Zoologica: New York Zoological Society
[XXIII :10
Pars Intermedia. A compact irregular cord-like arrangement is char-
acteristic of the pars intermedia (Plate I, Fig. 6). The cords are separated
from each other and from the processes of the nervosa by thin connective
tissue membranes. The lobe appears as a typical epithelial gland, and the
roots of the nervosa occupy only a small portion of its volume.
Two cell types were observed — a large, rounded one, flattened by adjoin-
ing cells, and a smaller, more polygonal type (Plate II, Fig. 7). The rounded
cells are predominant and are responsible for the faintly basophilic staining
reaction of this lobe. They occupy the center of the cords and much of the
periphery, while the polygonal cells are always located adjacent to the con-
nective tissue membrane.
The vesicular nuclei of the basophils are large and oval and occa-
sionally indented. They are made prominent with the Masson stains, by
a deep red-staining rim, scattered large red granules, and often a round,
central nucleolus. The cytoplasm is marked by numerous faint blue granules.
The polygonal cells are smaller than the basophils, although their
nuclei are identical in size as well as staining reaction. The cytoplasm of
the former is dense and finely granular. It takes both acid and basic stains
better than that of the basophilic cells and stains a reddish purple with the
Masson technique. These may be called amphiphils.
The goldfish intermedia is moderately well supplied with blood vessels.
The largest vessel observed in the pituitary appears in the region adjacent
to the lower nervosa mass (Plate I, Fig. 3). Numerous other smaller vessels
are associated with the connective tissue elements throughout the intermedia
and between this lobe and the iibergangsteil.
iibergangsteil. The iibergangsteil is the largest portion of the pituitary
body, occupying about one-half of its volume (Plate I, Fig. 3). Although
connective membranes separate it from the pars intermedia and the pars
nervosa, no similar structure lies between its cells and those of the pars
anterior. The iibergangsteil is quite homogeneous in structure; there are
no connective tissue elements within it and nervosa processes extend into it
for only a short distance. Strongly contrasting areas appear, however, as a
result of differential staining. Small irregular masses of deep-staining
acidophils are characteristically distributed throughout a deeply basophilic
background (Plate I, Fig. 3, and Plate II, Fig. 8). These masses increase
somewhat in number, as one proceeds posteriorly. The transition from
acidophilic to basophilic areas is strikingly sharp and is visible even in un-
stained sections.
The acidophilic areas (Plate II, Fig. 8) are made up of large, polygonal
cells, having a finely granular cytoplasm, which takes an intense red color
with the Masson stains. The cell borders are exceptionally sharp and dis-
tinct while the nuclei are very prominent and about the size of those of the
pars intermedia. The round or oval vesicular nuclei possess a few large
granules and frequently a large central nucleolus. Like the nuclei of
the intermedia, these take a deep red stain and have a dense rim. Scattered
among the acidophils are occasional polygonal cells with the typical red
nuclei but whose cytoplasm is vesicular, containing large faintly violet
granules. These are chromophobic cells, presumably degranulated eosino-
phils.
The basophilic background (Plate II, Fig. 8) is composed of large
polygonal cells, having indistinct borders and spherical or picnotic nuclei,
that are more granular than those of the acidophils. The nuclei of most of
the cells appear degenerate and many are missing entirely. With the Mas-
son stains, several differentiated areas may be distinguished. The cells of
all regions possess large red cytoplasmic globules which are usually located
near the cell borders. The cells of the central region have relatively few
globules in a pure blue cytoplasm in which individual granules cannot be
1938]
Bell: Hypophysis of the Goldfish
225
distinguished. Those of the posterior region are similar but have more
numerous globules. The most anterior region and also an area of cells
lying below the pars anterior present a different picture. Here the number
of globules is greatly increased so that they may occupy a large portion of
the volume of the cells. The cytoplasm is not deep blue but shows a coarse
violet or violet and blue granulation. These cells appear to be largely de-
granulated although they have not lost their red globules.
The vascular supply of the iibergangsteil is poor. Only small capil-
laries have been observed, and rarely do more than several of these appear in
any one field.
Pars Anterior. The characteristic cellular arrangement in the pars an-
terior is due to the presence of a large number of cords of the nervosa
which course through the lobe in an anterior-posterior direction (Plate I,
Fig. 5, and Plate II, Figs. 10, 11). In tranverse section, the lobe has a
perforated appearance, while in a sagittal section, the cells seem to be divided
into horizontal cords several cells in depth.
The cells are small as compared with those of the iibergangsteil and
the intermedia (Plate II, Figs. 9, 10). They are rounded or polygonal and
have distinct borders and large typically ovoid nuclei. Many indented or
folded nuclei are also present. The majority of the nuclei are vesicular,
having large, deep red granules and a smooth prominent red rim outlined
against a yellowish nucleoplasm. On the basis of cytoplasmic staining re-
actions, three cell types may be distinguished ; e.g., purplish-red amphiphilic
cells, basophils, and chromophobes. The lobe is composed predominantly of
the amphiphils, which have considerably less affinity for acid fuchsin than
the amphiphils of the pars intermedia. The chromophobes (Plate II, Fig.
10) are relatively scarce and are distributed uniformly throughout the lobe.
They have normal nuclei but appear pale violet in color like the degranulated
cells of the iibergangsteil. The basophils (Plate II, Fig. 10) are smaller
than the other cells but have nuclei which are identical in size and staining
reaction. Their cytoplasm is filled with fine deep blue granules.
A transition from a predominantly basophilic condition next to the
iibergangsteil to an acidophilic one at the dorsal free border, occurs in this
lobe. Most of this is due to the depth of staining of the amphiphilic cells.
At the ventral border, the majority of these assume a deep purple color and
have a dense cytoplasm without visible granulation, while those at the dorsal
border are more red than purple and have varying numbers of coarse
granules. Some of the transition is also due to the increase in the number
of basophils toward the ventral border. At most, however, these cells
account for only a small fraction of the change.
A few of the amphiphilic cells possess nuclei unlike the vesicular
acidophilic nuclei of the remainder of the cells of the pars anterior. They
are uniformly filled with fine basophilic granules, are without nucleoli, and
have a deep blue rim (Plate II, Fig. 9).
The processes of the nervosa that penetrate the pars anterior are highly
fibrous and especially loose in consistency (Plate II, Figs. 10, 11). Asso-
ciated with most of them are blood vessels of various sizes; these always
lie within the connective tissue membranes separating the processes from
the surrounding cells (Plate II, Fig. 11). A few vessels lie in the paren-
chyma itself, and occasionally one of these near the free border may be seen
to connect with a vessel in the connective tissue capsule of the gland. The
pars anterior is the most vascular lobe of the goldfish pituitary; neverthe-
less, its vascular supply is relatively poor as compared with that lobe in the
mammal and most higher vertebrates.
The layer of cells, enveloping the dorsal mass of the pars nervosa
(Plate I, Fig. 5) is considered part of the pars anterior, because of their
resemblance to the cells found in that part of the gland and should be
226
Zoologica: New York Zoological Society
[XXIII :10
confirmed further by morphogenic studies. The layer is thickest next to the
mass and thins down to a few cells as it follows the processes a short way.
Some nesting of the cells is evident. A few cells follow the main stalk all
the way to the pars intermedia. The cells are small and polygonal with large
vesicular nuclei and a scanty rim of cytoplasm. When stained with Dela-
field’s hematoxylin and eosin, the nuclei are deep blue and the cytoplasm
is a faint blue. Masson staining produces a pale violet coloration of the
fine cytoplasmic granules. The nuclei are pale yellowish-red with deep red
rims and a few scattered large red granules.
Discussion.
The anatomical relationships of the pituitary in Carassius auratus to
the neurocranium and brain resemble those found in higher vertebrates more
than in most other teleosts. Matthews (1936) found the pituitary of Fundu-
lus heteroclitus to be closely applied to the floor of the brain and merely cov-
ered ventrally by the parasphenoid bone. This condition is characteristic
of most of the teleost pituitaries that have been described, e.g., Anguilla vul-
garis and the Mormyridae (Stendell, 1914), Anguilla anguilla (Florentin
and Weiss, 1931, and v. Hallerstein, 1934), Gasterosteus aculeatus (Bock,
1928), Gadus morrhua (Herring, 1908 and 1913) and the haddock (de Beer,
1926).
In Carassius, on the other hand, the gland is separated from the brain
by the shelf-like cranial floor and the extension of the endocranial membrane
over the optic-pituitary foramen. As a result of this change, the pituitary is
attached to the lobus medialis by a relatively long stalk, which penetrates
the foramen. Two teleost pituitaries, those of Cyprinus carpio, the carp,
and of Esox lucius, the pike, as described by Stendell (1914), are very simi-
lar in structure to the gland of the goldfish. Whether they also have similar
anatomic relationships is unknown. In fact, previous accounts of teleost
pituitaries have included only brief comments on this subject.. The only
available literature dealing with it is that provided in anatomical discus-
sions by Jordan (1905), Gregory (1933), von Hallerstein (1934) and Holm-
gren and Stenzio (1936).
Four main divisions were differentiated in the goldfish pituitary: pars
nervosa, pars intermedia, iibergansteil, and pars anterior. These terms have
been used by de Beer (1926) and other recent workers in describing the
teleost pituitary. The configuration and structure of these divisions show a
marked similarity to those in Esox and especially in Cyprinus.
Certain changes in the orientation of the various lobes have taken place
as a result of the development of the pituitary stalk. The glands have as-
sumed a compact ovoid form with a somewhat variable orientation of the
lobes. The pars anterior of Carassius and Esox are antero-dorsal in position
while the pars intermedia is ventral and ventro-caudal respectively. In the
acorn-shaped gland in Cyprinus, the pars anterior is dorsal and the inter-
media is directed antero-ventrally. The change in the position of the gland
in Cyprinus has been associated by Stendell (1914) with its greater anterior
rotation as a result of a more extensive stalk formation. These glands rep-
resent a considerable advance over the condition in other teleosts, in which
the lobes are serially extended in an anterior-posterior direction, i.e., pars
anterior, iibergangsteil, and pars intermedia.
The goldfish has a typical teleost pars nervosa, which is composed of a
mass of tissue in the postero-dorsal region, from which root-like processes
radiate in all directions into the glandular tissue. Like Cyprinus, however,
it has an additional long internal process leading to the distant, compact,
lobe-like pars intermedia. Here the process ends in an additional mass of
nervosa tissue from which processes radiate throughout the intermedia. The
fine structure of the nervosa is very similar to that described by other work-
1938]
Bell: Hypophysis of the Goldfish
227
ers. Numerous acidophilic colloid droplets have been noted in the nervosa
adjacent to the pars intermedia by Herring (1908), Stendell (1914), Bock
(1928), and Matthews (1936). Some such strongly acidophilic material was
observed in the pars intermedia and occasionally in the adjacent nervosa in
Carassius. This material was missing, however, in the dorsal mass of the
nervosa. Stendell (1914) and Bock (1928) noted glandular cells distributed
throughout the nervosa and stated that these had migrated there from the
pars intermedia and the pars anterior. These cells are present in Carassius
but they definitely resemble the basophils of the iibergangsteil.
The pars intermedia of the goldfish forms a distinct lobe and, like that
of the carp, it is separated from the iibergangsteil by a partial cleft and a
connective tissue membrane. In most other teleosts it consists of a region,
occupying the posterior or postero-ventral extremity and enveloping the roots
of the nervosa. This is the case in Anguilla, Mormyrus, Mugil, Salmo, Chry-
sops and Esox (Stendell, 1914) and in Gasterosteus (Bock, 1928). In Fun-
dulus it takes the form of a layer, several cells in thickness, following the
contours of the nervosa (Matthews, 1936).
Stendell (1914) found no pattern of cell arrangement and no blood ves-
sels in the intermedia of the teleosts he studied. The only vascular supply
in this lobe was that carried into it by the roots of the nervosa. The condi-
tion is strikingly different in Carassius. Here the parenchyma is broken up
into irregular cords by a connective tissue network which extends through-
out the lobe. Moreover, blood vessels are more numerous in this lobe than
are free in the parenchyma of all the pituitary regions. The cord-like ar-
rangement resembles that seen in the intermedia of some of the mammals.
As in all vertebrates, the intermedia of the goldfish is predominantly
basophilic in staining reaction. The cells in the center of the cords are baso-
philic while many of those next to the nervosa and connective membranes
are amphiphilic in reaction. The latter take both acid and basic stains bet-
ter than the basophilic cells. Matthews (1936) noted large intensely stain-
ing acidophiles adjacent to the nervosa cords in Fundulus and Stendell found
a similar condition in Esox. It seems likely that these cells are identical with
the amphiphils in Carassius. As in Matthews’ description in Fundulus, many
of the nuclei in this lobe of the goldfish pituitary are folded or indented.
A marked similarity also exists between the iibergangsteil of Carassius
and that of Cyprinus (Stendell, 1914). In both it is the largest portion of
the complex; that of Carassius is central in position while it is central and
posterior in Cyprinus. The position of this lobe in other teleosts is highly
variable. The simple pituitaries, lacking hypophyseal stalks, have a mass of
differentiated cells occupying a central position between the pars anterior
and the pars intermedia. Bock (1928) found it to be located centrally in
Gasterosteus and to be covered ventrally by the pars anterior and the pars
intermedia. In Esox, it is a compact mass in the anterior region while in
Fundulus (Matthews, 1936), it is composed of an anterior mass and a layer
which covers the periphery of the gland everywhere except at the posterior
extremity.
Many irregular patches of large polygonal acidophils lying in a back-
ground of similar basophilic cells have been described in the iibergangsteil
of the goldfish. Stendell (1914) found acinar groups of acidophils lying in a
chromophobic background in the closely related Cyprinus and in a basophilic
area in Esox. The basophils of the goldfish appear degenerate, at least in the
late Winter and Spring, but nevertheless they take a much stronger baso-
philic stain than the basophils of the intermedia.
The staining reactions of the iibergangsteil of the various other teleost
pituitaries are variable and widely divergent from these. Tilney (1911) de-
scribed columns of acidophils, separated by blood vessels in Anguilla vulgaris,
while Stendell (1914) found the cells to be predominantly basophilic in this
228
Zoologica: Neiv York Zoological Society
[XXIII :10
form and in Mormyrus. This lobe is predominantly acidophilic in Fundulus
(Matthews, 1936) and Gadus (Herring, 1908) and equally acidophilic and
basophilic in Esox (Stendell, 1914). Usually the cells of the other teleost
iibergangsteils are smaller and more rounded than those of Cyprinus, Esox
and Carassius.
Stendell (1914) held that there is a progressive increase in the number
of basophils and a decrease in acidophils from the region adjacent to the
pars anterior to that next to the pars intermedia in the teleost iibergangsteil.
Bock (1928) could not confirm this in Gasterosteus nor does it occur in
Carassius. In the latter, an increase in the number of acidophilic masses
occurs toward the caudal extremity.
Few nervosa cords penetrate the teleost iibergangsteil. Connective tis-
sue has not been reported in it, and there is no sharp boundary separating
it from the pars anterior. In Gasterosteus there is a partial separation
(Bock, 1928).
A pars anterior, very similar to that described by Stendell (1914) in
Cyprinus and Esox, is present in Carassius. Matthews (1936) was unable
to distinguish such a differentiated portion in Fundulus. However, all other
descriptions in teleosts have included such a portion in an anterior or antero-
dorsal region of the pituitary.
A characteristic cellular arrangement in the form of cords, separated
by processes of the nervosa was noted in Carassius. Stendell (1914) noted
a similar cord-like arrangement of the cells in Cyprinus and Esox, but found
them to be separated by blood sinusoids. Many large blood vessels were
found enclosed within the nervosa cords in the goldfish and it seems pos-
sible that Stendell mistook such cords for blood sinuses.
Most of the cells of the goldfish pars anterior stain purple or a purplish-
red with the Masson stains and have been described as amphiphilic. Some
basophils and chromophobes are also present. The cells of Cyprinus and
Esox (Stendell, 1914) and Anguilla vulgaris (Tilney, 1911) are reported to
be basophilic. Stendell (1914) also described large acidophils next to the
blood sinusoids in Esox. He and Bock (1928) have found the pars anterior
to be acidophilic in the primitive Mormyrus and Gasterosteus, respectively.
It is interesting to note that an antero-dorsal chromophobic zone was found
in the pars anterior of Gasterosteus (Bock, 1928) and in Gadus morrhua
(Herring, 1908). In the former, the nuclei are closely packed and sur-
rounded by a small ring of basophilic granules.
A small specialized region has been included in the discussion of the
pars anterior of Carassius although its significance is uncertain. This is the
area of small polygonal cells surrounding the pars nervosa and fusing with
the posterior end of the pars anterior proper. The cells resemble those of
the pars anterior in size but have less cytoplasm and are nearly chromo-
phobic. On the other hand, its position and form, as a layer surrounding the
nervosa, immediately recalls the descriptions of the pars intermedia in Fun-
dulus (Matthews, 1936), in the haddock and the cod (Butcher, 1936) and in
other teleosts. It seemed advisable to classify this region with one of the
other lobes of the gland until more is known about it rather than further
confuse the pituitary terminology.
No large blood vessels were found entering the pituitary body in Caras-
sius. Occasionally capillaries were observed, however, leaving the rich net-
work in the pituitary capsule and entering the pars anterior. The blood
supply of the gland is poor as compared with those of most higher verte-
brates; this is true in general of all teleosts. The pars anterior is most
vascular of the teleost lobes with the exception of that in Gasterosteus (Bock,
1928). In Carassius, the vessels are located in the nervosa cords while in
Anguilla (Tilney, 1911) and Esox and Cyprinus (Stendell, 1914) numerous
blood sinusoids are reported. The pars nervosa carries numerous vessels
1938]
Bell: Hypophysis of the Goldfish
229
throughout the gland. Few vessels are present in the iibergangsteil in Caras-
sius, Anguilla and Cyprinus (Stendell, 1914) while Mormyrus, and Esox
(Stendell, 1914) and Fundulus (Matthews, 1936) are somewhat better sup-
plied. Blood vessels are reportedly absent in the parenchyma of the pars
intermedia, nevertheless they are present in considerable number in Caras-
sius.
In the accounts of the pituitary of the teleosts, most investigators have
included a thin-walled, folded pouch, the saccus vasculosus. This lies pos-
terior to the gland and is a separate outgrowth of the posterior hypothal-
amis. It is not present in all Teleostei; von Hallerstein (1934) reported the
presence of one in Cyprinus but there is none in Carassius. The saccus vas-
culosus is related in no way to the pituitary body. According to von Haller-
stein (1934), its thin epithelial wall contains numerous characteristic sen-
sory cells, as well as nerves, which is evidence for its being a special sense
organ. For these reasons, it is suggested that this organ should be omitted
from discussions of the teleost pituitary.
If granulation and regranulation, as evidenced by the presence of the
similar cells with varying numbers of granules can be taken as evidence of
secretory activity, as is usually done, and if certain changes in the form and
staining reaction of the nuclei may also be taken as such evidence (as Sever-
inghaus (1938) believes valid for the rat at least) then the cells of the
pituitary of the goldfish may be considered to be in a state of active secre-
tion in the late Winter and early Spring. Such variations in granulation
occur in Carassius both in the pars anterior and in the iibergangsteil. Scat-
tered about the pars anterior, cells have been observed the nuclei of which
stain uniformly with basic stains instead of the usual vesicular manner with
acid ones. Otherwise these cells cannot be differentiated from the normal
amphiphilic cells.
In this discussion of the pituitary, the terminology developed by Sten-
dell (1914), Herring (1908), Tilney (1911) and de Beer (1926) and used
as such by de Beer, has been adopted. This is based on Stendell’s homology
of the various portions of the teleost pituitary with those of this gland in
higher vertebrates. The suitability of this terminology has been questioned.
Indeed, Stendell himself stated that the iibergangsteil of the Teleostei sug-
gests the pars anterior more than it does the pars intermedia. He thought
it not impossible that the anterior lobe of higher vertebrates arose from the
iibergangsteil of fish.
Bock (1928) carried this idea to a positive conclusion as a result of his
study in Gasterosteus. He said that the anterior lobe of Teleostei is thus to
be regarded as a structure typical of this group, to which there is no homol-
ogy in amniotes and amphibia.
In the absence of detailed embryological evidence Stendell, (1914), pre-
ferred to retain a homology based largely on relative position. It seems to
the author, however, that a modification of this type of “homology” to in-
clude histological and cytological similarities, such as Bock (1928) sug-
gested with respect to the iibergangsteil, would be of definite value. Such a
homology would undoubtedly approach closer to a fundamental physiological
analogy. Such an analogy and the simplification it might bring should be the
goal of comparative physiological experimentation. Very little has been ac-
complished in this direction as yet, with the exception of the work on the
pars intermedia. The functions of the other portions of the pituitary in ver-
tebrates below the mammals are almost completely unknown.
Both the evidence of secretory activity and the cytological structure of
the cells of the pars anterior and the iibergangsteil in the goldfish pituitary,
suggest that these two portions ought to be considered as a glandular com-
plex, which is equivalent to the pars anterior of higher vertebrates. Al-
though no attempt has been made here to study the embryological origin of
230
Zoologica: New York Zoological Society
[XXIII :10
the various portions and thus arrive at a true homology, if such a thing is
possible, the lack of separation of the iibergangsteil from the pars anterior
suggests a common embryological origin. In contrast to this, there is a very
definite separation of the iibergangsteil from the pars intermedia.
The pars anterior (Hauptlappen) in teleosts was considered by Stendell
(1914) to be homologous to the lobe of that name in higher forms, while Bock
(1928) thought it was distinctive of this group. Certain facts suggest an-
other possibility, which would afford a definite simplification in terminology.
This lobe is strikingly similar to the pars tuberalis in reptiles, birds and
mammals in various morphological respects. The lobe is typically composed
of small polygonal basophils with a few acidophils in some forms. Only the
primitive pituitaries of Mormyrus and Gasterosteus are acidophilic. The
acinar, vesicular, cord-like or columnar arrangement of the cells found in
the tuberalis of various higher vertebrates can all be duplicated in the teleost
pars anterior. Stendell (1914) has reported the presence of intercellular
colloid in Esox lucius.
The position and relationships of the pars anterior may also be equated
to the uberalis. It is dorsal or antero-dorsal in the more complex teleost
pituitaries and anterior in the lower forms. The position of the pars
tuberalis in amphibians is antero-ventral, in reptiles it is antero-dorsal,
and in birds and mammals it is dorsal. The tuberalis usually fuses with
the pars anterior without a line of separation much as the teleost pars
anterior does with the iibergangsteil. In the stalked pituitaries, the pars
anterior lies in contact with the infundibular stalk much as does the tuber-
alis in the birds and mammals.
It seems desirable to simplify the pituitary terminology in the teleosts
and to correlate it with that applied to higher vertebrates. The morpho-
logical similarity of the teleost iibergangsteil and pars anterior with the
pars anterior and pars tuberalis respectively of higher forms suggests that
the latter terms might well be substituted for the former in referring to
the teleost.
No thorough cytological investigation has been attempted in this paper.
An elaboration of this material with information of that type, secured
throughout the year and including seasonal variations would allow a more
critical examination of the above question. At best, however, this can only
elaborate on the morphological similarities. By combining this method with
physiological experiments, an analogy with the functions of the lobes of the
pituitary of higher vertebrates might be established.
Summary.
1. The anatomical relationship and position of the pituitary gland in the
teleost fish, Carassius auratus, are distinctive not only in the vertebrate
group as a whole but also in the teleost group itself. It is well below
the floor of the brain in the region of the diencephalon, above the promi-
nent parasphenoid and surrounded by the prootic and alisphenoid bones.
Its dorsal surface is separated from the floor of the brain by a tough
membrane.
2. The pituitary body is attached to the brain in the infundibular region of
the hypothalamis by a cylindrical stalk, which penetrates this membrane.
3. Only a few small blood vessels were observed entering the gland from
the vascular network in its connective tissue sac. As in most teleosts
the gland is poorly vascularized as compared with those of higher verte-
brates.
4. The pituitary is made up of the following four main divisions.
(A) A pars nervosa made up of two irregular masses connected by a
1938]
Bell: Hypophysis of the Goldfish
231
heavy process. Root-like processes ramify throughout the pituitary from
these regions.
(B) A pars intermedia which is a large compact ventral lobe almost
completely separated from the ubergangsteil by a narrow cleft.
(C) An ubergangsteil in the form of a large thick plate-like region
dorsal to the pars intermedia. Its structure resembles most closely the
pars anterior of higher vertebrates.
(D) A small plate-like pars anterior in the antero-dorsal region. In
structure it is very similar to the pars tuberalis of many of the higher
vertebrates.
5. Evidence is offered to confirm Bock (1928) in the suggestion that the
ubergangsteil in Teleostei is really the pars anterior in the sense that
this term has been applied to higher vertebrates.
6. Similarly evidence is presented and the suggestion is made that the
“pars anterior” in Teleostei is in fact the pars tuberalis of higher forms.
Cell type, arrangement, and the position and relationships of the lobe
as a whole were used as criteria.
Bibliography.
Abramowitz, A. A.
1937 The opercular approach to the pituitary. Science, v. 85, p. 609.
Bock, Friedrich
1928 Die Hypophyse des Sticklings (Gasterosteus aculeatus L.) unter
besonderer Berichtigung der jahrescyklischen Veranderungen. Zeitschr.
f. wissens. Zool., Bd. 131 ,S. 645-710.
Butcher, Earl 0.
1936 Histology of the pituitaries of several fish. Bull. Mt. Desert Island
Biol. Lab., pp. 18-20.
Cardoso, D.
1924 Relations entre l’hypophyse et les Organes Sexuels chez les Poissons.
Compt. rend. Soc. Biol., T. 115, pp. 1347-1349.
Charipper, Harry A.
1937 The morphology of the hypophysis in lower vertebrates particularly
fish and amphibia, with some notes on the cytology of the pituitary
of Carassius auratus (the Goldfish) and Necturus maculosus (the
Mudpuppy). Symposia on Quantitative Biology. V, pp. 151-164.
de Beer, G. R.
1926 The Comparative Anatomy, Histology and Development of the
Pituitary Body. Oliver & Boyd, London.
Fleischmann, W. and Suzanne Kann
1932 Uber eine Funktion des weiblichen Sexual hormones bei Fischen
(Wachstum der Legerrohre des Bitterlings) . Pfluger’s Arch. f. d.
pies. Physiol., Bd. 230, S. 662-667.
Florentin, P. and M. Weiss
1931 Etude histologique de l’hypophyse de l’anguille (Anguilla anguilla
L.). Compt. rend. Soc. Biol., T. 107, pp. 718-720.
Gregory, William K.
1933 Fish skulls: A study of the evolution of natural mechanisms. Trans.
Amer. Philosoph. Soc., New Series, vol. 23, part II.
232
Zoologica: New York Zoological Society
[XXIII :10
Herring, P. T.
1908 A contribution to the comparative physiology of the pituitary body.
Quart. Jour. Exp. Physiol., v. 1, pp. 261-280.
1913 Further observations upon the comparative anatomy and physiology
of the pituitary body. Quart. Jour. Exp. Physiol., v. 6, pp. 73-108.
Hogben, L.
1936 The pigmentary effector system. VII. The chromatic function in
Elasmobranch fishes. Proc. Roy. Soc., Series B, v. 120, pp. 142-158.
Holmgren, Nils and Erik Stenzio
1936 Kranium and Visceralskelett der Akranier, Cyclostomen und Fische.
Bolk, Goppert. Kallius und Lubosch’s Handbuch der vergleichenden
Anatomie der Wirbeltiere. Bd. 4, S. 233-500. Urban & Schwarzenberg,
Berlin.
Houssay, B. A.
1930 Accion Sexual de la Hipofisis en los peces y Reptiles. Rev. Soc.
Argentina Biol., T. 6, pp. 686-688.
Jordan, David Starr
1905 A guide to the Study of Fishes. Henry Holt & Co., New York.
Kanter, A. E.
1934 A new biologic test for hormones in pregnancy urine. J.A.M.A., v.
103, pp. 2026-2027.
Lundstrom, Helen M. and Philip Bard
1932 Hypophysial control of cutaneous pigmentation in an Elasmobranch
fish. Biol. Bull., v. 62, pp. 1-9.
Masson, P.
1928 Carcinoids (argentafRne-cell tumors) and nerve hyperplasia of the
appendicular mucosa. Am. J. Path., v. 4, pp. 181-211.
Matthews, Samuel A.
1933 Color changes in Fundulus after hypophysectomy. Biol. Bull. v. 64,
pp. 315-320.
1936 The pituitary gland of Fundulus. Anat. Rec., v. 65, pp. 357-69.
Noble, G. K., K. F. Kumpf and V. N. Billings
1936 The induction of brooding behavior in the Jewel fish. Anat. Rec.,
v. 67, Proc. Amer. Soc. Zool., p. 50.
Parker, G. H.
1934 Color changes in the catfish Ameiurus in relation to neurohumors.
J. Exp. Zool., v. 69, pp. 199-224.
Severinghaus, Aura E.
1938 The cytology of the pituitary gland. The pituitary gland, pp. 69-117.
The Williams & Wilkins Co., Baltimore.
Smith, D. C.
1931 The influence of humoral factors upon the melanophores of fish, espe-
cially Phoxinus. Zeitschr. f. vergl. Physiol., Bd. 15, S. 613-636.
Smith, G. M., H. S. Burr and R. S. Ferguson
1935 A study of the effects of intermedia and injury of the hypophysis on
traumatic corial melanophores in goldfishes. Endocrinology, v. 19, pp.
409-412.
Stendell, Walter
1914 Die Hypophysis Cerebri. Oppel’s Lehrbuch der vergl. mikros. Anat.,
Teil 8, S. 1-162.
1938]
Bell: Hypophysis of the Goldfish
233
Tilney, Frederick
1911 Contribution to the study of the hypophysis cerebri with especial ref-
erence to its comparative histology. Memoirs of the Wistar Institute
of Anatomy and Biology, No. 2.
von Hallerstein, Viktor Haller
1934 Aussere Gleiderung des Zentralnervensystems. Bolk, Goppert, Kallius,
and Lubosch’s Handbuch der vergleichenden Anatomie der Wirbeltiere.
Bd. 2, 1. Halfte, S. 319.
von Ihering, R.
1935 Die Wirkung von Hypophyseninjektion auf den Laichakt von Fischen.
Zool. Anzeiger, Bd. Ill, S. 273-279.
WUNDER, W.
1931 Experimented Erzeugung des Hochzeitskleides beim Bitterling (Rho-
deus amarus) durch Einspritzung von Hormonen. Zeitschr. vergl.
Physiol., Bd. 13, S. 696-708.
Zondek, Bernard
1935 Chromatophorotropic principle of the pars intermedia of the pituitary.
Glandular Physiology and Therapy, pp. 133-138. Amer. Med. Assoc.,
Chicago.
Zondek, B. and H. Krohn
1932 Hormon des Zwischenlappens der Hypophyse. 1. Die Rotfarbund der
Elritze als Testobjekt. Klin. Woch., Bd. 11, S. 405-408.
234
Zoologica: Neiv York Zoological Society
EXPLANATION OF THE PLATES.
Plate I.
Fig. 1. Median sagittal section of a portion of the neurocranium with the brain
and pituitary in situ. Nasal end at the right. The section shows the
pituitary (P) lying in the myodom (MY) below the floor of the brain
case. The cerebrum (CB), cerebellum (CBL), lobus viscero-sensibilis
(L.V.S), parasphenoid (PS) and optic nerve (O.V.) are also indicated.
Zenker’s; Delafield’s haematoxylin-eosin. x 4.
Fig. 2. Transverse section of the pituitary near the posterior end and of adja-
cent structures, further magnified from Fig. 4. Note the cavity of the
third ventricle above the pituitary stalk and the thin lateral bone shelves
between the gland and the brain. P.N., pars nervosa; UBG, Ubergang-
steil; P.I., pars intermedia, x 13.
Fig. 3. An area from Fig. 1 under higher magnification, showing the attachment
of the pituitary to the brain, the cupped anterior rim of the prootic
bridge, the connective tissue capsule of the gland, and the large blood
vessel lying in the posterior indentation of the intermedia. Nasal end at
the right. P.A., pars anterior, x 13.
Fig. 4. Transverse section of a portion of the neurocranium with the brain and
pituitary in situ. The section shows the attachment of the pituitary
to the floor of the hypothalamis and the bony shelves that separate them
laterally. B.S., bony shelf; L.ln., lobus interior; L.M., lobus medialis;
O.L., optic lobe; P.. pituitary; R.M., x-ectus muscle. Bouin’s; Delafield’s
haematoxylin-eosin. x 4.
Fig. 5. Dorsal mass of the pars nervosa and the adjacent specialized region of
the pars anterior (the gi'oup of cells above the arrows). The dark cells
in the nervosa are glandular cells similar to the basophils of the iibei--
gangsteil. Zenker’s; Masson stains, x 160.
Fig. 6. Transverse section of epithelial portions of the pituitary showing his-
tological structui-e of the (P.l.) pars intei’media; (UBG) iibergangsteil
and (P.A.) pai's antei’ior. The dai’k ai-eas in the iibergangsteil represent
acidophils and the light areas in the pars anterior are nei’vosa cords.
Zenker’s; Masson stains, x 200.
Plate II.
Fig. 7. Pai’s intei’media showing amphiphilic cells adjacent to the nei'vosa cord
at left. Bemaining cells ai-e weakly basophilic. Zenker’s; Masson stains,
x 1080.
Fig. 8. tibergangsteil area. Acidophils appear dark while the basophils appear
chromophobic with haematoxylin. Several chromophobes may be noted
among the acidophils. Bouin’s; Delafield’s haematoxylin and Masson A.
x 1800.
Fig. 9. Pars anterior at the dorsal border, showing typical amphiphilic cells,
three with dark basophilic nuclei, also occasional chi’omrphobes. Zenk-
er’s; Masson stains, x 1800.
Fig. 10. Pars anterior near the ventral border with deeper-staining amphiphils
than in the more doi’sal region. C., chromophobe; B., basophil; A., am-
phiphil. Zenker’s; Masson stains, x 1800.
Fig. 11. Section of an ai’ea fi’om the pars anterior showing cords of the nervosa
with blood vessels typically enclosed within them. Zenker’s; Masson
stains, x 1800.
BELL.
PLATE I.
FIG. 5. FIG. 6.
MORPHOLOGY OF THE HYPOPHYSIS OF THE
COMMON GOLDFISH (CARASSIUS AURATUS L. ).
BELL.
PLATE II.
FIG. 9.
FIG. 7.
FIG. 10.
FIG. 11.
MORPHOLOGY OF THE HYPOPHYSIS OF THE
COMMON GOLDFISH (CARASSIUS AURATUS L.).
Hartley: Pathology of Dirofilaria Infestation
235
11.
Pathology of Diroflaria Infestation.1
Report of a case with chronic pulmonary arteritis.
Joel Hartley, M.D.
Mt. Sinai Hospital, New York City.
(Plates I-V).
The prominent increase in the recognition of Dirofilaria immitis infes-
tation among dogs in the United States during recent years has made this
a subject of some importance. Anatomic reports, although plentiful, have
been generally incomplete and inconclusive. A case of extensive Dirofilaria
immitis infestation in a gray wolf has recently come under our observation,
the study of which forms the basis of this report. In view of the gaining
importance of the disease it is of value to review briefly the literature on
the subject and to discuss its various features.
The occurrence of this type of parasitism in the gray wolf, as far as we
can determine, has not been reported in detail although there are frequent
records of its appearance in dogs. Textbooks, however, refer to the occur-
rence of Dirofilaria in foxes, wolves, muskrats, raccoons, and occasionally
other animals. Infestation with Dirofilaria immitis does not occur in man
and conversely Filaria bancrofti is not observed in animals except perhaps
in primates.
The presence of worms in the heart of the dog was first reported by
Peysson in 1806 and by Von Gruby and Delafond in 1843. Leidy named the
parasite Filaria immitis. Raillet and Henry in 1911 established the genus
Dirofilaria, the most common species of which is Dirofilaria immitis (dirus,
meaning cruel), a designation which has remained to the present. The
members of the group besides D. immitis are : D. magalhaes and D. repens.
The genus Dirofilaria is characterized by the absence of oral labia and by
the possession of very inconspicuous cephalic papillae.
Shattock in 1881 observed worms in the cardiac chambers of a dog in
Japan and found many parasites partly embedded in a post-mortem blood
clot, which extended into the pulmonary artery and its larger branches.
In 1886, J. R. Figueira de Saboia described the presence of male and
adult filaria in the chamber of the left ventricle of the heart of a child and
referred them to De Magalhaes for study. A statement of the illness of
the child was not given ; the blood was not known to have been examined
for larvae. The role of the parasite in the child’s illness is therefore un-
known. These were designated D. magalhaes. This occurrence in the child
plus one case of D. repens found in man by Skrjbin, Althausen and Schul-
man (1930), represent the only recorded instances of infestation in man
by a related species of Dirofilaria.
1 From the laboratories of the Mt. Sinai Hospital and the New York Zoological Park.
)
236
Zoologica: New York Zoological Society
[XXIII :11
Woods, in 1885, observed that Dirofilaria infestation was a common
disease throughout the coastal cities of China. In one of his cases a con-
tinuous coil of filaria was found extending from the right hepatic vein,
through to the inferior vena cava, into the right auricle, through the tri-
cuspid valve orifice into the chamber of the right ventricle. The left
auricular chamber was packed with the nematodes connected with a mass
fourteen inches long extending into both lungs through the pulmonary-
arteries. He was of the opinion that the ova were introduced into the animal
by drinking from stagnant pools. F. Fullerborn and others, however,
definitely established the fact that the disease is transmitted by means of
the mosquito which functions as the intermediate host.
De Magalhaes (1892) emphasized the difference between Dirofilaria
immitis and Filaria bancrofti. This investigator felt that whereas Dirofilaria
immitis was generally found in the chamber of the right heart and in its
great vessels, Filaria bancrofti was constantly encountered within the
lymphatics. The recent studies of Augustine and his co-workers demon-
strated, on the other hand, that the microfilaria of D. immitis could occa-
sionally also be found within the lymphatic channels.
The first comprehensive articles on Dirofilaria immitis appeared in 1889
with the reports by Janson of post-mortem findings in 41 dogs. The para-
sites were most frequently found in the right cardiac chambers and its
great vessels but only rarely within the cavity of the left ventricle, aorta or
femoral artery. Pulmonary emboli were found in the branches of the
pulmonary artery with infarction, necrosis and even abscess formation of
the pulmonary parenchyma. The adult parasites wei'e found in the inferior
vena cava, extending into the hepatic vein, producing thereby congestion of
the liver with secondary congestion of the portal vein and subsequent ascites.
In one heart a worm was seen passing through a patent foramen ovale with
one half of its body located in each auricle. This offers a ready explanation,
at least in some instances, of a mechanism for the entrance of the parasite
into the general systemic circulation. Janson further described 26 cases of
hypertrophy of the right ventricle and in some specimens marked dilata-
tion, 5 of obstruction of the pulmonary artery, 2 examples of aneurism of
the pulmonary artery without thrombosis, 9 of aneurism of the pulmonary
artery with thrombosis, 13 of “interstitial hepatitis,” 19 instances of hydro-
thorax and ascites, 1 of edema of the subcutaneous tissues, 5 examples of
worms entangled in the chordae tendinae. He referred to the presence of
chronic endocarditis which he felt had proceeded the filariasis, and was
aggravated by the presence of the nematodes. In the lungs of acute cases
he found “anemia and atelectasis with and without edema.” A “nut-meg
liver” was not infrequently observed. The kidney exhibited “parenchy-
matous, catarrhal and interstitial changes.” On the basis of these ana-
tomical findings Janson believed that the embryos of Dirofilaria immitis
escaped from the body mostly in the urine. This concept has been subse-
quently shown to be erroneous, for although microfilaria are known to pass
through the kidneys, the renal lesions, described by Janson, correspond to
the so-called spontaneous interstitial nephritis, so commonly observed in
animals (Horn, 1937). However the possibility of any exaggeration of this
process by the toxins elaborated by the parasite cannot be negated.
Van Meter (1892) reported the gross finding of a case which included
chronic endocarditis of the tricuspid valve, attributed to the presence of the
parasite. The inter-auricular septum was perforated (foramen ovale?) and
the nematodes were also found in the pulmonary vein. French (1899) re-
ported the gross findings of a case in a bitch whose litter of pups three
months previously died with diarrhea. He found some small worms in the
intestines of the offsprings and raised the question of the possibility of
the placental transmission of the disease. Riesman (1903) described an
animal with dyspnea, who fell to the ground while trotting and on examina-
1938]
Hartley: Pathology of Dirofilaria Infestation
237
tion revealed markedly accentuated heart sounds, a murmur (type not de-
scribed), cyanosis and ascites. Necropsy disclosed a “white thrombus” in
the right heart enmeshing a parasite and slightly hypertrophied right
cardiac chambers. The left branch of the pulmonary artery was completely
occluded with an ante-mortem thrombus. Larvae were not demonstrated.
He stated that 50% of the canine population of China and Japan harbored
the worm and that the embryos were found more easily during the night
than the day. The attention of Hopkins three years afterwards was drawn
to this disease when three of his dogs being used for experimental pur-
poses died under anesthesia. These and three additional animals showed
gross findings typical of immitis infestation. In one case, the wall of the
pulmonary artery two centimeters above the valve was studded with fine
granules which on section proved to be subendothelial fibrous nodules.
In 1919 Lynch anesthetized an infected dog in whose blood he had
demonstrated many larvae and proceeded to observe the vessels of the
omentum drawn over the stage of a microscope. He observed the micro-
filaria moving about in the capillaries but found no evidence of plugging
of the vessels. In the same year Ryan reported the case of a dog which
had contracted the disease in the northern temperate zone of the United
States. Kowakami, in 1919, studied the distribution of the microfilaria in
the body and found them to be most commonly within the lungs, heart and
respiratory muscles. Yano (1927) gave the order of frequency of the in-
volved organs as heart, lung, kidney and asserted that 20% of the dogs
about Tokio were infected. Histological studies of the kidneys of his animals
disclosed the typical lesion found in the so-called interstitial nephritis of
dogs. In one instance, however, the author found multiple hemorrhages
and in another, multiple zones of infarction. Microfilaria were found in
scattered capillaries of the cortex. In advanced cases there was renal
atrophy. It was Yano’s opinion that the microfilaria did not form glom-
erular emboli but suggested that the renal changes might be of toxic origin.
A few of the vitally stained microfilaria injected into healthy dogs were re-
covered in the urine and were at times associated with renal hemorrhage.
Blood eosinophilia was not observed. His attitude concerning the toxic
origin of renal changes was strengthened by the fact that when a centrifuged
extract of female Dirofilaria immitis was injected into mice, distinct
changes, resembling those found in dogs, were elicited in the kidneys.
Yamanouchi (1928) attempted to study the distribution of experiment-
ally introduced larvae in the kidneys of mice. When he injected living larvae
into mice both intravenously and into the renal artery, he found the em-
bryos in the capillaries of the cortex located, for the most part, in the capil-
laries of the glomeruli. Heat-killed larvae were distributed mostly in the
afferent vessels of the glomeruli but none in the excretory tubular system,
the kidney, ureters or bladder. Blackberg and Ashman (1930) emphasized
the pronounced effect of filarial infestation upon cardiac function. Electro-
cardiographic tracings taken on dogs during exercise disclosed an inversion
of the T wave in lead I and almost complete disappearance of the T wave
in lead II. In view of the similiarity of the tracings to those seen during
attacks of angina pectoris in man the changes were attributed to an in-
adequate blood supply resulting in a deficient oxygen supply to the cardiac
muscle.
Hayes (1933) reported that 80% of the dogs in Florida that he ex-
amined had Dirofilaria immitis infestation. Hayes asserted that in 1931-
1933 this disease caused more loss than canine distemper. He referred to
the increase of the disorder in recent years throughout the southern tiers
of the United States. It was panzootic in 1932, throughout Georgia, Florida,
Alabama, Mississippi and Louisiana. The most common symptoms were
poor general condition, a rough coat, lack of endurance, a lazy attitude de-
veloping into extreme weakness, progressive anemia, often chronic in-
238
Zoologica: Neiv York Zoological Society
[XXIII :11
digestion, accompanied by ascites and generalized dropsical swellings. Mild
infestations produced less marked symptomotology and with good care the
animals could go on indefinitely without any apparent discomfort.
In recent years Augustine and his co-workers studied the question of
the filtration of microfilaria by lymph nodes. They found that the larvae
were not phagocytosed, as were the erythrocytes, and that they could pass
through the lymph nodes with slight hindrance. In another experiment,
after exsanguinating a normal dog and substituting the blood of a heavily
infested dog, they were able to collect living embryos in the lymphatic
channels and in the spinal fluid. Thus, they demonstrated that the larvae
may escape from the blood vessels, traverse tissues and enter the lymphatic
system. The histological findings were the same in both donor and re-
cipient. Microfilaria were disseminated in every part of the body. Great
numbers were present in the lungs, liver and kidney. None was found in
the stomach or intestines. In the lungs many were discovered in the large
arteries, veins, capillaries and occasionally in the alveoli. Many lay in the
veins and sinusoids of the liver. The kidneys manifested the microfilaria
most frequently in the glomeruli, some edema was present but no patho-
logical changes. They usually followed the course of the capillaries. Very
few were recovered in 40 cc. of urine of a heavily infected dog. Evidence
that they passed through the glomeruli to the efferent vessels was offered
by the presence of the organisms in the renal veins. Few existed in the
capsule of the spleen but many were in the pulp. The lymph nodes likewise
showed few and those were mostly in tne intermediary sinuses in the depths
of the node. Great numbers were embedded in the myocardium, particu-
larly the left ventricle. They were mostly in the capillaries but in some in-
stances lay outside the capillaries parallel to the fibers. A tremendous num-
ber rested in the bood vessels of the pericardium. No cellular reaction to
the microfilaria was evident anywhere.
To settle the question of the behavior of the larvae in the capillaries
and the reason for the absence of embolic phenomena these workers injected
the microfilaria into bats and observed their behavior in the capillaries of
the bat’s wing. This was a superior method to that of Lynch, who ex-
amined the exposed mesentery of a dog, because it did not disturb the normal
physiological conditions of the host. “The anterior end of the microfilaria
were constantly active in searching movements. As the blood vessels branch
and become smaller the organisms may eventually enter capillaries which
are obviously somewhat contracted and proceed into them until the lumen
gets too small, when they never were observed to escape or make permanent
plugs. They simply backed out. That is the reason why microfilaria do not
form emboli. Whenever these active organisms became stuck in a capillary
they merely moved backwards against the current until they were safely
on their way elsewhere.”
Joyeux and Cabassu (1935) found microfilaria in the blood of 19 out
of 26 dogs examined in the area known as Camargue. They identified the
adult Dirofilaria in one of two foxes they studied. Yamamoto (1936) de-
scribed microfilaria in the uterus and detected their escape in the menstrual
flow. However, he did not regard this fact as definite proof of passage
through the placenta. Ohashi, in the same year, established the fact that
the larvae could be found in the skin of the eyelids, bulbar conjunctivae and
membrana nictitans. A few embryos were present in the lacrimal secretion.
He occasionally found that they worked their way into the vitreous body or
produced hemorrhage about the optic nerve.
Hinman and Baker (1936) made a helminthic survey of 1,305 dogs
from New Orleans and found 321, or 24.4%, infected with Dirofilaria im-
mitis. They recovered the embryos in immature dogs and concluded that
they may pass through the placenta. Diagnostic serological and intra-
dermal tests in filaria were introduced independently by Taliaferro and
1938]
Hartley: Pathology of Dirofilaria Infestation
239
Hoffman, and also by N. H. Fariley (cited by Manson-Bahr 1935). The
antigen was obtained from Dirofilaria immitis.
Nematodes have been found in the hearts of the harbor seal. Accord-
ing to Ross Nigrelli of the New York Aquarium this species belongs to the
genus Halocercus, the members of which are common parasites of the lungs
of other species of seals, dolphins and porpoises. This parasite not in-
frequently is mistaken for Dirofilaria immitis.
Report of a Case.
On December 12, 1937, a male wolf ( Canis nubilis ) died in the New
York Zoological Park. He was observed to have had a distended abdomen
during the previous four years.
Post-mortem Examination : The body was well developed and in fairly
good condition. The dentition was good. The abdomen was moderately dis-
tended. The wall of the right heart was strikingly hypertrophied and the
chambers tremendously dilated so that it appeared to be about twice the
size of the left ventricle (Plate I, Fig. 1). The pulmonary ring was dilated
and measured 6.5 cm. in circumference whereas the aortic ring measured
5.5 cm. A thrombus filled the right ventricle in which many adult round
worms were enmeshed. These were identified by Lucker and McIntosh in
E. W. Price’s laboratory of the Bureau of Animal Industry, Washington,
D. C., as Dirofilaria immitis. A reddish-brown, rough, irregular, endocardial
nodule, 3 mm. in diameter, was found in the outflow tract of the right ven-
tricle. A thrombus, with entrapped Dirofilaria, extended into the lumen of
the pulmonary artery and of many of its branches. Many of the branches
of the right pulmonary artery and some of the left showed aneurysmal
dilatations and some were thrombosed. In the branches the thrombi were,
for the most part, firmly adherent. Some were reddish-brown and others
grayish-white, fibrotic and evidently organized (Plate I, Fig. 2). Some
of the nematodes were still motile. The intima of the main stem of the
pulmonary artery was roughened, due to shallow, fine, pit-like, closely
grouped depressions. Along the intimal surfaces of the branches of the pul-
monary arteries there was longitudinal and transverse bridging. Tranvers-
ing the lumina of some branches, in secant-like fashion was a network of
slender, fibrous bands, strings and thin membranes, originating in the in-
tima at one point and inserting at another (Plate I, Fig. 2), (Safir, 1932).
The lungs weighed 1,370 grams together and presented many infarcted
areas with associated areas of pneumonitis (Plate II, Fig. 1). The pul-
monary veins exhibited no pathological changes. The larger radicles of the
tracheo-bronchial tree were clear throughout. The hilar lymph nodes were
large and succulent.
The peritoneal cavity was filled with 2,000 cc. of clear amber fluid. The
liver was enlarged, weighed 1,900 grams and displayed evidence of marked
acute and chronic passive congestion. Some areas manifested pseudo-adeno-
matous formation. There was one large cyst 4 cm. in diameter in the right
lobe. A large thrombus containing Dirofilaria filled the right main branch
of the hepatic vein and its smaller branches (Plate II, Fig. 2). The gall
bladder and biliary system were normal. The spleen did not appear en-
larged. A firm hemangioma, about 1 cm. in diameter, arose from the mid-
portion. The gastro-intestinal tract showed evidence of severe passive con-
gestion. Multiple minute mucosal hemorrhages were present. In addition
many small, yellow, hard nodules 2 mm. in diameter were disseminated over
the serosa of the small intestine. The kidneys were markedly congested.
The spinal fluid was cloudy and on smear disclosed many gram positive cocci.
The meninges had a dull fibrinous appearance. Pure cultures of staphy-
lococcus were grown from the heart’s blood and the spinal fluid. The or-
240
Zoologica: New York Zoological Society
[XXIII :11
ganisms produced yellowish to light orange pigment but could not be defi-
nitely classified by fermentation tests.
Microscopic Findings : The myocardium revealed only degenerative
changes. The intima of the main stem of the pulmonary artery was markedly
thickened and showed pit formation. The adventitia was thickened and
showed numerous perivascular infiltrations consisting chiefly of lymphocytes
but also plasma cells and polymorphonuclear cells. The pericardium in this
region revealed a membrane composed of fibrin with entrapped necrotic
cells lying on the surface. The elastica Van Giesen stain disclosed marked
reduplication and fragmentation of the elastic fibres. The elastic lamellae
were of irregular thickness and often their continuity was interrupted.
The walls of the intra-pulmonic branches of the pulmonary artery
were thickened and exhibited frequently a striking destruction with replace-
ment of the elastic and muscle fibres of the media by connective tissue. The
elastica was markedly fragmented (Plate III, Fig. 1). There was excessive
thickening of the adventitia. The intima manifested a conspicuous focal
thickening. In one place there was necrosis of the wall with inflammatory
cell infiltration (Plate IV). The lumina of these vessels displayed thrombi
in all stages of organization. The adjacent alveolar septa were very strik-
ingly thickened and infiltrated with inflammatory cells. This interstitial
reaction extended into the finer ramifications. The lung tissue in the im-
mediate vicinity of the larger branches of the pulmonary artery showed
collapsed alveoli, frequently containing polymorphonuclear cells and macro-
phages. Many of the smaller branches of the pulmonary artery were sur-
rounded by a mantle of inflammatory cells and display a marked thick-
ening of the wall and narrowing of the lumen (Plate III, Fig. 2). Some ves-
sels disclosed entire replacement with fibrous tissue. There was also evi-
dence of purulent bronchitis, obliterating bronchiolitis, pneumonia and
edema. Organizing pneumonia was in evidence. One wide patch of lung
tissue was completely fibrotic, suggesting a completely organized infarct.
The ghost-like remains of its vessel of supply was present in its center
(Plate III, Fig. 1). The lumen of this artery was completely filled with scar
tissue indicating an organization of the original thrombus. A gram stain
showed a small number of gram positive cocci in the lung tissue while an
iron stain revealed large amounts of iron pigment in the walls of the larger
arteries and in the macrophages (heart failure cells) which occupied the
alveolar spaces. In the lumen of one bronchus an adult Dirofilaria was seen
(Plate V, Figs. 1 and 2). The hilar lymph nodes revealed evidence of hyper-)
plasia. A gram stain disclosed only an occasional gram positive coccus.
The capsule of the liver was slightly thickened. The hepatic architec-
ture was severely distorted. There was marked central congestion and
marked secondary atrophy of liver cells. In some places entire patches were
devoid of liver cells and were replaced by cavernous sinusoids. Disse’s
spaces were widened and filled with granular material. The Kupfer cells
were markedly swollen and some showed erythrophagocytosis. The biliary
ducts were not increased. A large branch of the hepatic vein contained an
organizing thrombus. An adult worm lay within the lumen next to the
thrombus. An iron stain disclosed very little iron and the gram stain only
occasional gram positive cocci.
The spleen was markedly congested. The pancreas displayed vascular
congestion. The adrenals presented slight hemorrhage in the zone reticu-
laris. The kidneys were the seat of marked congestion, edema and de-
generation of the tubular epithelium. There was slight interstitial focal in-
filtrations of plasma cells.
Unfortunately the brain and spinal cord were lost and could not be
examined microscopically. No larvae were detected in any of the tissues.
All tissues were fixed in formalin.
1938]
Hartley: Pathology of Dirofilaria Infestation
241
The final diagnosis was : Dirofilaria immitis infestation of the right
heart with thrombosis in the right heart, pulmonary artery and its branches,
and the right branch of the hepatic vein ; chronic arteritis of the pulmonary
artery and its branches with aneurism formation of many of the branches
and pulmonary arteriosclerosis; mai'ked hypertrophy of the right ventricle
and auricle with tremendous dilatation ; chronic endocarditis of the out-
flow tract of the right ventricle; multiple pulmonary infarcts; chronic
purulent bronchiolitis; chronic pneumonitis of all lobes; staphylococcus
sepsis ; severe acute and chronic passive congestion of the liver with atrophy
and pseudoadenomatous formation; acute purulent meningitis; large cyst
of the liver; and slight interstitial nephritis.
Comment.
A review of the reported cases, including the instance herein de-
scribed, indicates that the usual cause of death in dirofilariasis of the
pulmonary arterial tree is congestive right heart failure. Anatomically, it
was evidenced in this case by right cardiac hypertrophy and excessive dila-
tation, an extreme degree of chronic passive congestion of the liver and
ascites. From the character of the arterial lesions observed there is indi-
cation that one of the earliest effects on the pulmonary arteries may be a
toxic degeneration of the pulmonary arterial wall caused by the humoral
products of the adult parasites. This arteritis in turn may have caused
the thrombosis in the pulmonary vessels, followed by the development of
marked chronic changes in the vessel walls and organization of the thrombi.
The hypertension of the pulmonary circuit thus engendered by the presence
of such an occluding mechanism, caused dilation of the right ventricle and
subsequent hypertrophy. The pulmonary hypertension may in turn be a
contributing factor in the causation of the thickening of the intima and the
endocardium. The endocarditis may well be a change de novo or possibly
an exaggeration of a preceeding endocarditis which is so commonly found in
animals of the older age group. Thrombus formation may also be provoked
by this endocardial lesion. As the right heart fails, the slowing of the cir-
culation produced in the great veins leading to the right heart, plus the
presence of the adult parasites, would tend to foster further thrombosis in
these places. We have then, four factors that may have a causative relation-
ship to thrombus formation, namely: toxic arteritis, chronic endocarditis,
slowing of the venous circulation, and the adult Dirofilariae acting as foreign
bodies. The pulmonary arteriosclerosis in the case herein reported may be
considered as secondary to the pulmonary hypertension.
The presence of this obstructing mechanism, together with the asso-
ciated infarcts, whatever the pathogenesis of the latter may be, causes gen-
eralized anoxemia, as a result of the interference with the pulmonary cir-
culation. This readily accounts for part of the symptom-complex that the
diseased animals present, such as, dyspnea, cyanosis and weakness.
It is interesting in connection with the arterial lesions encountered to
briefly mention the report of Clark and Graef of a case of Schistosomiasis
mansoni in a twenty-one year old Puerto Rican girl who died of congestive
failure. Post-mortem examination revealed a markedly dilated right heart.
Evidence of arteritis and dilatation of the branches of the pulmonary artery
with associated severe arteriosclerosis was noted. The authors were able
to demonstrate the ova in the media and concluded that the arteritis was
a specific response to the parasite, while the arteriosclerotic thickening was
assumed to be secondary to the increased tension within the pulmonary
arterial tree. We do not feel that the arteritis observed by us was caused
by the presence of adult or embryonic parasites in the wall of the vessels.
We could not demonstrate such a lesion as did Clark and Graef. On the
242
Zoologica: New York Zoological Society
[XXIII :11
contrary, we are inclined to the view that the toxins elaborated by the adult
Dirofilaria and their larvae may have caused the arteritis. The work of Yano,
referred to above, in which he produced renal changes after injection of a
centrifuged extract of female Dirofilaria, gives some support to this concept.
In the case reported here, the additional factor of severe infection must
be held accountable as the immediate cause of the animal’s death.
We failed to detect any microfilaria in the tissues we examined. This
is not surprising when one considers the life cycle of the Dirofilaria. For it
is known that the larvae produced by the adult filaria are not capable of
maturing in the host and tend to disappear. They must pass through an
intermediate host (mosquito) and be innoculated into a dog or other sus-
ceptible animal before they are capable of reaching an adult, sexually mature
state. In our case, where the adult nematodes were apparently dying as
they were entrapped in thrombi, fewer and fewer embryos were produced
and thus none was seen in the tissues. The blood of the wolf was not
examined before death.
It is apparent from what has been said that in infested animals with
cardiac embarrassment, the additional toxic effect of an anesthetic on the
cardiac muscle readily explains the cases of death ensuing during anesthesia
reported in the literature.
Summary.
1. A review of the literature on the morbid anatomy of Dirofilaria
immitis infestation in mammals is presented.
2. An instance of such infestation in a gray wolf dying of congestive
heart failure and superimposed sepsis is described.
3. An unusual feature of arteritis with associated aneurysmal dilata-
tion of the branches of the pulmonary artery is reported.
The author is indebted to Dr. Charles R. Schroeder for his kind co-
operation and permission to publish the case; and to Drs. Paul Klemperer
and Sadao Otani for their valuable aid in the preparation of this report.
Bibliography.
Andrews, Mary N.
1937. Hyperinfection of dogs with microfilaria. J. Helminth., 15, 167-168.
Augustine, D. L., Drinker, & Leigh:
1935. On the filtration of microfilaria by lymph nodes. Tr. Royal Soc. Troy.
Med. and Hygiene, 29, 51.
Augustine, D. L., & Drinker
1935. Migration of microfilaria (dirofilaria immitis) from blood vessels to
lymphatics. Tr. Royal Soc. Troy. Med. and Hygiene, 29, 303-306.
Augustine, D. L., Field, & Drinker
1936. Observations on living microfilaria immitis in the capillary circulation
of bats. Tr. Royal Soc. Troy. Med. and Hygiene, 30, 231.
Blackberg, S. N., & Ashman, R.
1930. Electrocardiographic studies of dogs infested with dirofilaria immitis.
J. A. Vet. M. A., 77. 204-211.
Clark, E., & Graef, I.
1935. Chronic pulmonary arteritis in schistosomiasis mansoni associated with
right ventricular hypertrophy. Report of a case. Amer. J. Path., 11,
693-707.
1938]
Hartley: Pathology of Dirofilaria Infestation
243
De Rosario, F.
1936. Dirofilaria immitis Leidy and its culicine intermediate host in Manila.
Philippine J. Sci., 60, 45-55.
French, W. F.
1899. Worms in a dog’s heart. Amer. Vet. Rev., 23, 432-434.
Fullerborn, F.
1929. In the “Handbuch der pathogenen Mikroorganismen” by Kolle and
Wasserman.
Gaillard, H.
1937. L’Evolution de dirofilaria immitis Leidy chez Aedes (Stegomya)
aegypti et A. olbopictus au Turkin. Comp. Rend. Soc. de Biol., 125, 130-
132.
Gruby & Delafond
Recuil de Medecine V eterinaire, Bd. 20.
Hall, M. C„ Price, E. W., & Wright, W. H.
1935. Parasites and parasitic diseases of dogs. U. S. Dept. Agriculture.
Dept. Circ. No. 338. May.
Hayes, I. M.
1933. Filariasis or heartworm in dogs. Vet. Med., 28, 140-143.
Hill, V. C.
1934. Some data on dirofilaria immitis. Vet. Bull. U. S. Army, 28, 252-253.
Hopkins, J. G.
1906. Six cases of infection with filaria immitis. Johns Hopkins Hosp. Bull.,
17, 377-379.
Hinman, E. H.
1935. Studies on dog heartworm. Amer. J. Trop. Med., 15. 371-382.
Hinman, E. H.
1936. Attempted reversal of periodicity in dirofilaria immitis. Proc. Soc.
Exp. Biol, and Med., 33, 524-527.
Hinman, E. H., & Baker, D. D.
1936. Helminthologic survey of 1,305 dogs from New Orleans with special
reference to age resistance. J. Trop. Med., 39, 101-104.
Hinman, E. H., Faust, E. C., & De Bakey, M. E.
1936. Filarial Periodicity in dog heartworm dirofilaria immitis after blood
transfusion. Proc. Exp. Bio. and Med., 31, 1043.
Horn, H.
1937. The Experimental Nephropathies. Archiv. Path., 23, 71-121.
Howard, F. H.
1903-1904. Case of filaria immitis in heart of a dog. Proc. Path. Soc. (Phila.),
7. 91-96.
Hu, Stephen M. K.
1931. Host-parasite relationships of dirofilaria immitis. Am. J. Hyg., 14, 614-
629.
Hutyra & Marak
1926. “Pathology and Therapeutics of the Diseases of Domestic Animals,”
Vol. 3. Budapest.
Janson, J. L. & Tokishige, H.
1899-1892. Filaria immitis. Mitteilungen Deutche Gesell f. Volkerk. Ostas-
iens, 5, 349.
244
Zoologica: Neiv York Zoological Society
[XXIII :11
Joyeux, Ch., & Cabassu, J.
1935. Etude sur le filariose des chiens de Camargue. Frequence de dirofilaria
immitis (Leidy). Bull, de le Soc. Path. Exot., 18, 187.
Kawakami, A.
1919. Ueber die Verteilung der Microfilarien in den verschiedenen Organen
von Hunden, die von filaria immitis infiziert worden sind. Verhandl. d.
Jay. Path. Gesellsch., Tokio, 9, 95.
Leidy
1856. Proc. Acad. Nat. Sci. Phila., 5. 118.
Lent, H., & Teixeira de Freitas, J.
1911. Dirofilaria Raillet and Henry. Mem. Inst. Oswaldo Cruz., 1937, 32.
37-54.
Lynch, K. M.
1919. Filarial periodicity. J. Am. Med. A., 73, 760.
Magalhaes, De P. S.
1892. Die filaria bancrofti; Cobbald und die filaria immitis. Centrbl. F. Bac-
teriol. u. Parasitenk {Jena), 12, 511-514.
Magalhaes, de, P. S.
1886. Rev. des. cursos. theor. e. prat. d. fac. de med. d. Rio di Janeiro, Anno
3, No. 3.
Manson-Baiir Philip, H
1935. “Lectures on Tropical Diseases.” W. T. Keene’s and Co., Chicago.
Markus, H.
1909-1910. Filaria immitis Bij. den hond in Nederland. Tijdschr. V veeart-
senijk Moandbl., Utrect, 37, 449-456.
McGinnis, C. L.
1933. Filariasis in a dog. Vet. Med., 28, 140-143.
Mazza, S.
1937. Filarideo del corozon de coati. (Filariasis in the heart of a raccoon).
Prensa med. Argent., 13, 1045-1048.
Mazza, S., Rosenbusch, F., & Antequeda, E.
1926~. Secunda nota preliminar sobre la microfilaria sp. de los. penos. del
Norte. (Microfilaria in dogs). Bol. Inst, de Clin. Quir., 2, 267-272.
Mullany, J.
1934. Heartworm in Hong Kong. J. Royal Army Vet. Corps, 6, 34-35.
Nigrelli, Ross
Personal communication.
Pavoloff, P.
1935. Dirofilaria immitis in Bulgaria. Vet. Sbir., 39, 133-135.
Perard, Ch.
1936. Au sujet de la filariose du chien. (Dirofilaria immitis). Bui. De la Soc.
Path. Exot., 19, 1036.
Peysson
1806. Observations de pathologie comparee sur des vers trouves dans le coeur
d’un chien. Annals Soc. de med. prat, de Montpel., 15, 49-53.
Popesco, F.
1935. Canine Filariasis in Roumania. Arch. Vet. Bucharest, 25, 145-148.
1938]
Hartley: Pathology of Dirofilaria Infestation
245
Prieto, R. R.
1934. Filariasis as a cause of sudden death. Rev. Hig. y San. pec., Madrid,
24. 53-54.
Roulard, E., Colas-Balcour, J., Touranoff, C., & Treillard, M.
1936. Transmission of dirofilaria immitis Leidy. Bull. Soc. Path. Exot., 29.
111-112.
Raillet and Henry
1911. Bui. de la Soc. de Path. Exot., Vol. 4, p. 285.
Ryan, J. F.
1919. Filaria immitis in a dog’s heart. J. Am. Vet. Assn., 55, 199.
Saphir, Otto
1932. Bands and Ridges in the Pulmonary Artery. Archiv. Path., 14. 10-20.
Shattock, S. G.
1881-1882. Filaria immitis in a dog’s heart. TV. Path. Soc., London, 33, 434.
Siiipley, Arthur E.
1894. Notes on a dog’s heart infected with filaria immitis. Proc. of Cam-
bridge Philosophical Soc., 8, 211-214.
Simonelli, A.
1936. Canine Filariasis in Umbria. Nuovo Ercol., 41, 169-178.
Skrjbin, K. J., Althausen, A. Y., & Schulman, E. S.
1930. First case of dirofilaria repens in man. Trop. Med. i. vet., 8. 9-11.
Smith, Allen J.
1915-1916. Filiariasis in the Americas. Proc. of the Second Pan Amer. Scient.
Cong., sect. 8, part 1, vol. 9, pp. 49-76.
Woods, G. W.
1885-1886. (U.S.N.) Report on filaria in dogs. U. S. Navy Dept. Bur. Med.
and Surg. Gen. Doc. U. S. N., pp. 100-103.
Underwood, P. C.
1933. A case of heartworm (dirofilaria immitis) in a Virginia dog. J.
Parisitol., 20, 77.
Vaills, L.
1930. Filaria and microfilaria infection in dogs. Rev. vet. Toulouse, 88, 133-
144.
Van Meter, S. D.
1892. The filaria immitis. TV. Colorado M. Soc., pp. 288-292.
Welch, F. H.
1873. On a species of filaria found in the interior of the vascular system of
a dog. Lancet, 1. 336-338
Yamanouch, M.
1928. Distribution of experimentally introduced larvae of dirofilaria immitis
in the kidney. Tr. Jap. Path. Soc. 1 8, 487-489.
Yamamoto, K.
1936. Histopathological findings of the uterus, placenta, and fetus of dogs
with dirofilaria immitis infections. Tr. Soc. Path. Jap., 26, 690-694.
Yano, A.
1927. Findings in the kidneys of a dog with dirofilaria immitis. Jap. Med.
World, 7, pp. 292 and also 324.
246
Zoologica: New York Zoological Society
EXPLANATION OF THE PLATES.
Plate I.
Fig. 1. Markedly dilated and hypertrophied right cardiac chamber.
Fig. 2. Right lung with pulmonary arteries opened to show old and fresh thrombi
(A) and adult Dirofilaria (B). Note organizing network over intima (C).
Plate II.
Fig. 1. Right lower lobe (larger) and right upper lobe (smaller) showing areas
of infarction and pneumonitis. Note aneurysmal dilation of pulmonary
artery (A).
Fig. 2. Liver with hepatic vein containing thrombus and adult Dirofilaria.
Plate III.
Fig. 1. Cross section of a branch of the pulmonary artery showing complete or-
ganization of a thrombus. Elastica exhibits marked fragmentation and de-
struction. (Section taken from an infarcted area. Elastica-carmine stain).
Fig. 2. Smaller branch of pulmonary artery showing markedly thickened intima
and narrowed lumen. (Elastica-carmine stain).
Plate IV.
Fig. 1. Wall of large branch of pulmonary artery showing purulent inflammation
of intima (A) and thrombus in lumen (B).
Plate V.
Fig. 1. Low power view of lung showing longitudinal section of an adult Diro-
filaria (A) lying within the bronchus, thickened wall of main branch of
pulmonary artery revealing fibrosis (B) and a thrombus in lumen (C).
Fig. 2. Enlarged photograph of right lower corner of Fig. 1 showing portion of
Dirofilaria lying within bronchus, and purulent bronchitis.
HARTLEY.
PLATE I.
FIG. 1.
FIG. 2.
PATHOLOGY OF DIROFILARIA INFESTATION.
HARTLEY.
PLATE II.
FIG. 2.
PATHOLOGY OF DIROFILARIA INFESTATION.
HARTLEY
PLATE III
FIG. 2.
PATHOLOGY OF DIROFILARIA INFESTATION
HARTLEY.
PLATE IV.
PATHOLOGY OF DIROFILAR1A INFESTATION.
HARTLEY.
PLATE V.
FIG. 2.
PATHOLOGY OF DIROFILARIA INFESTATION.
Coates, Cox & Smith: Papilloma in Electric Eel
247
12.
Papilloma of the Skin Occurring in an
Electric Eel, Electro phorus electric us (Linnaeus).
C. W. Coates
New York Aquarium
R. T. Cox
Department of Physics, New York University,
University Heights
&
G. M. Smith
Department of Anatomy, Yale School of Medicine,
and New York Aquarium.
(Plates I-VI).
With the increase in the reported instances of abnormal growths occur-
ring in fishes, it has become apparent that cutaneous tumors of teleosts
are more frequently encountered than are the visceral neoplasms affecting
deeper-lying organs. Among the skin tumors of fishes those arising in the
connective tissue of the derma or corium seem to be more common than
those growths composed of the squamous and mucus cells of the epidermis.
There are, finally, a group of papillary tumors of the skin in which the
hyperplastic epithelium is heavily supported by a stroma of connective
tissue carrying blood and lymphatic vessels and occasionally nerves to the
tumor. These growths represent both an epithelial and mesodermal re-
sponse with both types of cells about equally distributed. Such tumors may
be classified as fibro-epithelial growths of the skin. Unlike neoplastic
growths in mammals, teleost tumors which are composed principally of
epidermal cells appear to be exceedingly rare. Even when they occur in a
malignant form (epithelioma), their rate of growth is a relatively slow
one, and the tumor, although involving adjacent tissues, usually remains
localized in the region where it has its origin, rarely giving rise to the sec-
ondary growths in distant tissues or organs so frequently found in mam-
malian and avian tumors.
Of the few benign epithelial tumors in fishes which have been reported
in the literature, the following instance described by Takahashi (1929) may
be mentioned as an example. The growth was 3 x 2 x 0.5 cm., grayish-white
in color, and occurred in a Crucian carp, Carassius carassius, occupying
the region of the left operculum. It was sharply circumscribed and showed
on its surface dilated blood vessels. The tumor was composed histologically
of epidermal cells which were flattened near the surface and exhibited some
cornification. Along the deep surface of the tumor there was an active
proliferation of epithelial cells into the derma. Interstitial connective tissue
248
Zoologica: Neiv York Zoological Society
[XXIII :12
was scanty but fairly well supplied with blood vessels. There was no ne-
crosis, nor were there metastases. Attempts to transplant small pieces of
tissue into other fishes of the same species were unsuccessful.
There are no tumors reported up to the present time occurring in the
electric eel, Electrophorus electricus (Linnaeus). The small cutaneous
growth here described developed in the right mid-dorsal region of a large
adult electric eel, caught in the Amazon river basin and maintained in good
health in a fresh water tank at the New York Aquarium for a period of
more than four years. This particular eel, measuring 6 feet in length and
18 inches in largest circumference, had been tested frequently for electrical
discharges estimated by measurement to be approximately 350 volts and
40 watts (Coates, Cox & Granath, 1937). Such electrical discharges were
repeated at intervals of about 2/1,000 of a second when the eel was dis-
turbed. Other eels subjected to the treatment in measuring electrical dis-
charges for approximately the same period showed no similar growths.
The growth began as a small elevation of the skin, pinkish-gray in color,
and a few millimeters in diameter. Its maximum growth of 2 x 2.5 cm.
was reached in about ten weeks, at which time it was excised for purposes
of histological study. No recurrence took place during the remaining year
and a half of life.
The growth was a grape-like pedunculated mass (Plate I, Figs. 1 and 2)
and was attached to the skin by a narrow short stalk above the lateral line.
It did not appear to have its origin in the skin overlying the usual distribu-
tion of the electric organs. It was of considerable interest, therefore, when
the microscopic study disclosed evidences of nerve trunks and nerve ter-
minations in the epithelium of the growth itself (Plates, IV & V, Figs. 7
and 10).
The epithelial tumor, regarded as a benign papilloma, was composed
of solid masses of squamous epithelium commingled with very numerous
mucous cells, in an arrangement far more irregular than that exhibited by
the normal skin (Plate II, Figs. 3 and 4). The epithelium composing the
various lobes of the grape-like mass was supported by a central narrow
fibrous core of connective tissue containing many small blood vessels along
the course of which existed dendritic melanophores. The linear distribution
of pigmented cells external to the thin walls of blood vessels gave to the
interior of the tumor the peculiar branching or tree-like effect seen by
transillumination in the gross specimen cleared in cedar oil (Plate I, Fig.
2). It will be noted that pigmented lines identifying the various ramifica-
tions of the connective tissue stroma extended outward nearly to the free
surface of the growth.
It is not the purpose of the present paper to refer in detail to the
rather complicated structure of the skin of the electric eel. It may be stated
briefly, that the normal epidermis (Plate VI, Figs. 11 and 12) approximately
20-30 cells in depth, rests upon a well developed basement membrane which
in its turn separates epithelium (E) from the corium (C). The latter is
composed of dense connective tissue fibers supporting the blood capillaries,
lymphatic vessels, small nerve trunks and branches and numerous melano-
phores. The pigment cells are spread out in horizontal arrangement, at
times in several layers, immediately below the hyaline basement membrane.
In the normal skin, extending outward at right angles to the surface of the
body, are found at short regular intervals numerous prolongations of the
fibrous tissue of the corium, passing as supporting septa (T) outward into
the epithelium (Plate VI, Fig. 11). These supporting septa of connective
tissue are somewhat conical in shape with the broader base continuous
with the corium while externally they taper off to form finely pointed or
club-shaped tips of delicate connective tissue which in this way carry the
terminal branches of blood capillaries with closely lying melanophores to
points near the extreme outer limits of the epithelial surface. Such sup-
1938]
Coates, Cox & Smith: Papilloma in Electric Eel
249
porting septal strands, single or branching, form a region of attachment
for narrow elongated closely packed palisade-like epithelial cells (T) (Plate
VI, Fig. 12). In contrast with these the epidermal cells springing from the
basement membrane between the supporting septa form a system of low
cuboidal and small round epithelial cells several layers in density which in
all probability represent matrix cells that give rise to predominant squamous
and mucous cells of the many-layered epidermis. Relatively few elongated
or fusiform cells are found resting on the basement membrane between
septa.
Since both the above-mentioned prolongations of connective tissue from
the basement membrane and the tuft-like arrangement of narrow epithelial
cells covering their tips are very conspicuous features in the skin of the
electric eel, it seems rather reasonable to suppose that they are somehow
involved in the conduction of the electric discharge from the tissue of the
fish into the surrounding water. If these structures differ in electric con-
ductivity from the surrounding epithelial tissue, their presence must modify
the distribution of electric current through the epithelium. If they are
markedly more conducting than the surrounding tissue, their effect will be
to concentrate the current in their own structure through the layers of
cells near the basement membrane. The branching of the prolongations near
the outer boundary of the epithelium would diffuse the current again where
it passes through the outer layers. If such a distribution of current in the
epithelium exists, its advantage may be to by-pass the discharge around the
matrix cells and so to confine any injurious effect it may have on the epi-
thelial tissue to the outer layers of cells which are constantly replenished.
In our preparations, the extreme outer layer of cells of the normal
epidermis in contact with aquatic environment is composed of mucous cells.
In the papillomatous growth all forms and shapes of epithelial cells
already mentioned were present, but did not give the appearance of having
the precise and regular arrangement which is seen so well in the normal
skin. Besides mucous cells there was a great irregular commingling of
squamous cells with variations in shape and size, also with irregularities
in the size of the nucleus. In many places epithelial cells were trans-
versely arranged running parallel with the surface. For the most part
the cells forming the outer boundary of the various lobules composing the
tumor were of the mucous type. (Plate III, Fig. 5).
The entire tumor after excision was fixed in 10% formalin, embedded
in paraffin, sectioned serially and stained with both hematoxylin and eosin
and Masson’s Light Green stain. An opportunity was afforded therefore
to examine in sequence the stalk-like attachment of the growth to the skin
in order to determine the presence of nerve tissues.
Several small nerve trunks were found entering the growth via the
narrow stalk. One of these was composed of six separate bundles of
myelinated nerves, while another consisted of two nerve bundles. The en-
veloping connective tissue of each nerve trunk was well developed. Of par-
ticular interest was the appearance of the perineurium immediately sur-
rounding the smaller separate groups of nerve fibers. The perineurium
was conspicuous by the lamellated appearance of its cells arranged circularly
in several layers (Plate IV, Figs. 7 and 8). The two primary nerve trunks
which could be traced in detail entered the same lobule of the tumor at dif-
ferent levels by penetration of the basement membrane and were unaccom-
panied by blood or lymphatic vessels. On reaching the epithelium the
perineural cells were still further increased in number and rested in contact
with flattened adjacent epithelial cells, from which they were clearly
distinguished, particularly in sections stained by the Masson technique.
The lamellated arrangement of cells about the several groups of nerve
fibers persisted as the nerve bundles coursed through the hyperplastic
250
Zoologica: New York Zoological Society
[XXIII :12
epithelium (Plate V, Fig. 9) to reach positions near the outer surface of
the tumor. Here individual nerve fibers terminated with lamellated cells
still visible (Plate V, Fig. 10). Terminal modifications of nerve fibers were
thus formed resembling in many respects Pacinian bodies.
In numerous areas of the growth considerable oedema was noted. This
affected principally the region near the surface representing the terminal
branches of the small capillaries coursing along the fibrous septa. The
tips of connective tissue septa here formed a delicate reticulum in whose
meshes an oedematous or albuminous material existed.
Neither necrosis nor signs of inflammation were found in the papil-
loma. There was no evidence that parasites were present in the abnormal
tissue. No intracellular inclusion bodies could be demonstrated. Collections
of monocytes with ingested particles of pigment detritus were encountered
at intervals (Plate III, Fig. 6).
Summary.
The histological features of a papilloma arising in the skin of Electro-
phorus electricus (Linnaeus), containing certain terminal modifications of
nerve fibers, are described, and the electrical possibilities of the regions con-
taining palisade-like epithelial cells in the normal skin are briefly discussed.
Bibliography.
Coates, C. W., Cox, R. T. & Granath, L. P.
1937. The electric discharge of the electric eel, Electrophorus electricus
(Linnaeus). Zoologica, 22: 1.
Takahashi, K.
1929. Studie fiber die Fischgeschwfilste. Zeitschrift fur Krebsf orschung ,
29: 1.
1938]
Coates, Cox & Smith: Papilloma in Electric Eel
251
EXPLANATION OF THE PLATES.
Plate I.
Fig. 1. Gross specimen of lobulated papilloma excised from the skin of an elec-
tric eel, Electrophorus electricus (Linnaeus). Fixed in formalin.
Fig. 2. Gross specimen of papilloma, cleared in cedar oil, photographed by
transillumination to show the pigmentation of the supporting stroma
caused by melanophores which accompany small blood vessels. X 4.
Plate II.
Figs. 3 & 4. Photomicrographs of papilloma, composed of hyperplastic epithelium
formed by commingling of squamous and mucous cells. The outer lining
cells are columnar-shaped mucous cells. The supporting stroma of the
papilloma is fibrous tissue containing pigment cells along the course of
small blood vessels. X 65, 85.
Plate III.
Fig. 5. Mucous cells, columnar in form, covering the outer surface of the
epithelial growth. X 275.
Fig. 6. Collections of monocytic cells containing ingested pigment detritus.
X 325.
Plate IV.
Figs. 7 & 8. Bundles of nerve fibers with thickened lamellated perineurium lying
in fibrous stroma of the growth. X 150, 250.
Plate V.
Figs. 9 & 10. Terminal modifications of nerve fibers resembling Pacinian bodies
lying in the hyperplastic epithelium of the growth. X 250.
Plate VI.
Fig. 11. Normal skin of Electrophorus electricus (Linnaeus). Many layers of
squamous epithelial cells with numerous distended mucous cells, rest
upon a dense fibrous corium (C) containing pigment cells. Supporting
the epithelium are connective tissue septa (T) arising from the corium
and extending into the epithelium ( E ) . The outer lining of the
epithelium is formed by columnar shaped mucous cells. X 95.
Fig. 12. Normal epithelium of the skin of electric eel showing septa attached to
which are narrow elongated epithelial cells giving a tuft-like appearance
(T). (C) = corium, (M) = matrix cells, (S) — squamous cells, (X) —
mucous cells. X 250.
COATES, COX a SMITH.
PLATE I.
FIG. 1.
FIG. 2.
PAPILLOMA OF THE SKIN OCCURRING IN AN
ELECTRIC EEL, ELECTROPHORUS ELECTR1CUS (LINNAEUS)
COATES, COX a SMITH.
PLATE II
FIG. 3.
FIG. 4.
PAPILLOMA OF THE SKIN OCCURRING IN AN
ELECTRIC EEL, ELECTROPHORUS ELECTRICUS (LINNAEUS).
COATES, COX a SMITH.
PLATE III.
FIG. 5.
FIG. 6.
PAPILLOMA OF THE SKIN OCCURRING IN AN
ELECTRIC EEL, ELECTROPHORUS ELECTRICUS (LINNAEUS).
COATES, COX & SMITH.
PLATE IV.
FIG. 7.
FIG. 8.
PAPILLOMA OF THE SKIN OCCURRING IN AN
ELECTRIC EEL, ELECTROPHORUS ELECTRICUS (LINNAEUS).
COATES, COX a SMITH.
PLATE V.
FIG. 9.
FIG. 10.
PAPILLOMA OF THE SKIN OCCURRING IN AN
ELECTRIC EEL, ELECTROPHORUS ELECTRICUS (LINNAEUS).
COATES, COX & SMITH.
PLATE VI.
FIG. 12.
PAPILLOMA OF THE SKIN OCCURRING IN AN
ELECTRIC EEL, ELECTROPHORUS ELECTRICUS (LINNAEUS).
Ginsburg: Arithmetical Definition of Species
253
13.
Arithmetical Definition of the Species, Subspecies and Race
Concept, with a Proposal for a Modified Nomenclature.
Containing a simple method for the comparison of related populations.1
Isaac Ginsburg.
(Text-figures 1-4).
Introduction.
Ever since naturalists began to doubt the special creation of species,
in consequence of the accumulating incontrovertible evidence proving the
descent of species from preexisting forms of life, the question of what
constitutes a species became bothersome. The acuteness of the problem
was intensified when biologists ceased to be satisfied with describing and
cataloging species of plants and animals by a study of one or but a few
specimens, and began to study in detail the individuals comprising a species
en masse. The great individual variability of the characters employed for
specific distinctions and the consequent difficulty of drawing sharp lines of
demarcation between closely related species wras thus revealed.
This fundamental problem in biology engaged the attention of students
and a considerable literature has grown up dealing with the question of
what constitutes a species. A good deal of what has been written has
reference to its speculative aspects. It is not the aim of this article to
add anything to the purely abstract discussions of the problem. Instead,
it represents an attempt to solve this problem and determine definitely just
what is a species by reference to a series of actual data. It is my intention
to consider in detail a number of concrete cases showing the facts of
speciation as they occur in nature and an attempt is made to correlate the
facts and draw the proper conclusions therefrom.
The data employed to support the propositions advanced in the present
paper are entirely taken from populations of fishes. The data were not
specially collected for this paper, but have been gathered in connection
with my revisional studies of American fishes. Some of the data have been
published in reports on these studies; others are here published for the
first time, or previously published data are amplified. As far as the included
evidence is concerned this paper may be said to be a sort of a by-product
of my revisional studies of the genera of American fishes, and the support-
ing data presented below have been only casually selected to prove the
propositions advanced. While the examples cited have been taken entirely
from the class of fishes, the same method no doubt will be found applicable
to other groups of animals, and plants as well.
All the examples included below are based on quantitative characters.
The method developed here is most strikingly applicable to such characters
1 Published by permission of the U. S. Commissioner of Fisheries.
254
Zoologica: New York Zoological Society
[XXIII :13
which thus serve best for the purpose of illustration. For some specific
characters this method will be applied with greater difficulty and a lesser
degree of precision, and cases may be encountered in which it is inappli-
cable, especially in cases in which differences are based on qualitative char-
acters (see below). However, this is a problem in practical usage. Such
cases must be considered by themselves, and special means of expressing
particular characters in the form of frequency distributions, may be devised.
The fact that this method may be inapplicable practically in some cases does
not detract from the pertinence of the general principles evolved as a result
of its application in the great number of cases in which it may be used with
ease and precision.
The question of what constitutes a species cannot be considered by
itself. To solve this problem consistently it must be broadened to include
the subdivisions of the species, since, as will be shown definitely hereafter,
the different categories which may be established imperceptibly grade into
one another. This paper, therefore, goes into the whole problem of the
species and its subdivisions. In connection with this study the question of
the nomenclature of taxonomic categories below specific rank is examined
and a modification of the method now in general use, is proposed. A simple
method for the comparison of closely related populations is employed in
this paper, which may prove to be a useful tool in the taxonomic study of
populations of specific or lower rank.
Failure of Previous Attempts to Establish Absolute Limits
to the Species Concept.
A number of criteria have been used by previous writers for the pur-
pose of defining and establishing absolute boundaries to limit the species
concept. The literature on the subject is quite voluminous, and it is not my
present purpose to give a complete review of such previous attempts. This
has been done by a number of writers in greater or lesser detail, and the
reader may be referred to one of these writers, such as Robson (1928) who
also gives an extensive bibliography of the subject. In general, it may be
stated that all criteria which have been proposed for establishing absolute
boundaries by which we may definitely determine just what constitutes a
species, have been found wanting. One of these criteria, the morphological,
may be considered here briefly because the definitions proposed herewith
are based solely on that criterion. It is the only criterion which is most
generally useful in the practice of taxonomy.
A population of variable individuals was considered to be fully distinct
specifically from another population differing in at least one structural or
color character in such manner that no intergrading individuals occur, but
that every single individual may be referred either to one or the other of
the two populations. Such populations have been regarded as fully distinct
species. Even at the present time some systematists hold to this as a
criterion for distinguishing species, or as the most important criterion.
However, absolute lack of intergradation in nature is far from being the
usual condition. On the contrary, intergradation of related populations is
so general that it would not be far fetched to make the statement that it
is the rule rather than the exception. Certainly most closely related species
of fishes, in my experience, have been found to intergrade more or less.
Specific characters are roughly divisible into two classes: (1) quan-
titative characters, such as the number of fin rays, scales or vertebrae,
proportional measurements, etc.; (2) qualitative, such as differences in
color, or the presence or absence of certain structures or color marks, or
differences in their form. Among fishes the former is predominant. Qual-
itative structural characters which are absolute, that is, they practically
do not show any intergradation, are in many cases considered to be of
1938]
Ginsburg: Arithmetical Definition of Species
255
generic or subgeneric, rather than specific, importance. Also, two groups
of related species between which a comparatively wide gap exists with
respect to a given quantitative character are often placed in separate genera
or subgenera. Characters which are considered to be of specific importance
only, by general consent, more likely than not, will be found to intergrade
between two closely related populations when a sufficiently large number
of individuals are studied in detail. This is true not only of quantitative
characters, but frequently an imperceptibly gradual intergradation occurs
also in the case of qualitative characters, although in the latter case it may
be difficult to measure and express in terms of precise figures the degree
of intergradation.
The general existence of intergradation is not duly reflected at present
in taxonomic works. The reason taxonomists have been able to describe
species generally in such manner as to make it appear that no intergrading
individuals are present, is that these descriptions are usually based on but
a few individuals. Consequently, by the law of chance, such few individuals
were apt to fall, in most cases, near the center, and away from the extremes,
of a regular frequency distribution. Occasional bothersome specimens
may have been explained away as being abnormalities, spoi’ts or hybrids.
However, such border line specimens will be found in most closely related
species if a sufficient number of individuals are studied in detail. Except
for possible occasional hybrids, or atypical individuals for various reasons,
such specimens are normal individuals, but they fall in at the extreme of
the frequency distribution, and as far as any given single character is
concerned they may as well belong to one species as to another closely
related one.
The True Morphological Criterion Is the Degree of
Intergradation, or Divergence.
The intergradation between natural populations varies in degree. When
a sufficient number of pairs of closely related populations are compared and
the several intergradations, or divergences, are arranged in order of mag-
nitude, we obtain a series that is graduated by virtually infinitesimal steps.
It follows, therefore, that the determination of whether a given pair of pop-
ulations constitute two species or belong to a category of lower rank depends
on the degree of intergradation; or, to view it from another angle, the
degree of divergence. As a further consequence, it follows that species as
well as subspecies and races are not absolute entities. The lines drawn to
limit these classificatory units must be arbitrary. That this is not merely
a theory but the actual condition which exists in nature is attempted to
be proved in the paragraphs which follow. (I discussed this proposition in
another paper (1937a). Data to prove it are presented here.) After adopt-
ing a measure by which the degree of intergradation is expressed in terms
of a definite figure, a series of data obtained by the study of pairs of
closely related populations of fishes is presented and the figure expressing
the degree of intergradation for each pair is calculated by the method
employed. The figures thus obtained form a gradual series which may be
arranged in ascending or descending order and there are no breaks in the
gradual continuity of the series where sharp lines may be drawn to limit
absolutely our concept of species, subspecies or race.
Measure of Intergradation, or Divergence.
Our next step in the solution of the problem is to adopt a definite and
uniform measure by which the degree of intergradation, or divergence,
between any two closely related populations may be expressed in terms of a
single figure. Several methods of measuring intergradation or divergence
may be employed. For instance, Davenport (1898) proposes what are essen-
256
Zoologica: New York Zoological Society
[XXIII :13
tially two methods of measuring precisely the difference between two
closely related populations. Davenport represents the two related popula-
tions, in every instance cited by him, in the form of a single bimodal curve.
One measure which he designates as the “index of divergence” represents
“the ratio of the distance between the modes to the half-range ... of the
broader curve.” The other measure, designated by the author as the “index
of isolation,” represents “the ratio of the depression [between the two parts
of the bimodal curve] to the length of the shorter mode.” (The “depres-
sion” is the vertical distance between the apex of the smaller curve and
the lowest point of the depression). Both measures are expressed as per-
centages. A measure which is often employed to indicate the difference or
divergence between two populations may be represented by the formula
Mi — M;
V El2 + E?
in which Mj and M2 represent the means of the two respective populations,
and E! and E2 represent the probable errors of the two means, respectively.
This expression represents the ratio of the difference between the means to
the probable error of this difference (see Pearl, 1930, pp. 282-287). In
treatises dealing with the application of statistical methods to biological
problems the statement is often made that when the numerical value of this
ratio is 3 or 4, it is “significant.” Other methods may be suggested. For
our present purpose a simple method of expressing the degree of intergra-
dation, or divergence, between two related populations is proposed to be
used as indicated below.
For the present I am not going to discuss in detail the advantages and
disadvantages of the various methods which have been proposed or which
may be suggested with the method employed herein. This is a problem by
itself, a full discussion of which would lead us astray from our main thesis.
If any consistent method be adopted and a number of closely related pairs
of populations be compared by it, the results quite likely will form a gradual
series going to prove the continuity of intergradations, inter se, in nature;
but the relative position of the pairs of populations compared, in the series,
will no doubt change somewhat according to the method used, and some
methods will more nearly represent the facts of nature than others. A brief
comparison is made below (p. 279) between the method employed herein and
the standard method, and it is shown that the latter is not well adapted for
our purpose. Besides the fitness of the measure employed to represent the
facts more nearly in their true light, it has two salient advantages which
may be mentioned briefly. (1) It may be determined easily and quickly, a
very desirable consideration, especially from the point of view of the busy
taxonomist. (2) Because of its simple character its pertinence in explain-
ing the facts of nature is strikingly evident and the relationship of variable
and closely related populations may be appreciated readily when this measure
is used.
Our simple measure may be illustrated by the following hypothetical
examples. Let us assume two species of fishes, A and B, the chief differenti-
ating character of which is represented by the number of scales in the lateral
line, a character which is frequently employed in distinguishing closely re-
lated species of fishes. Let us suppose further that the scales of a hundred
specimens of each species have been counted, and the figures obtained were
as follows:
Number of scales (class) 24 25 26
Species A (frequencies) 92 8
Species B (frequencies) 97 3
It may be said then that species A intergrades with species B to the extent
of 8% and this figure may be suggested as our measure of intergradation.
1938]
Ginsburg: Arithmetical Definition of Species
257
While this figure obviously suggests itself, it is not the figure finally adopted.
The measure of intergradation for the above hypothetical case, by the method
employed, is 4%, for reasons which will become clear presently.
In nature examples similar to the above simple hypothetical case may
be encountered ; but the variability and relationship of closely related natural
populations is much more often not as simple. Let us, therefore, take the
next step and assume a hypothetical case where the dispersion of the fre-
quency distributions and the overlap are a little more pronounced; while at
the same time the frequency distributions are perfectly regular, as follows :
Number of scales (class) 24 25 26 27
Species A (frequencies) 5 90 5
Species B (frequencies) 5 90 5
In this perfectly regular hypothetical case each species overlaps the other by
5% and this figure may be taken as our measure of intergradation.
Again it may be stated that perfectly regular frequency distributions
such as the foregoing hypothetical case are seldom encountered in practice.
Frequency distributions are usually irregular or, to use the technical ex-
pression, skewed. Part of the irregularity encountered in practice is no
doubt due to incomplete sampling of the populations; but it is evident that
skewness in the frequency distributions of populations is the more usual
and normal condition in nature. Let us then assume a hypothetical case
where the frequency distribution is irregular while the overlap is more pro-
nounced than in the simple hypothetical case cited first, as follows:
Number of scales (class) 24 25 26 27
Species A (frequencies) 3 89 8
Species B (frequencies) 6 85 9
This case is a little more complicated and is nearer the majority of examples
encountered in actual practice. How are we to measure intergradation in
this case?
Frequency polygons of the number of scales of two hypothetical species;
see discussion in text. Solid line represents species A, broken line rep-
resents species B.
If two frequency polygons be constructed to represent graphically the
foregoing hypothetical data (Text-fig. 1), the polygons will intersect at a
point over the abscissal axis between the points representing 25 and 26
258
Zoologica: Neiv York Zoological Society
[XXIII :13
scales. If a vertical line be drawn from the point of intersection of the two
polygons, it will be found that 8 specimens of species A cross over to the
right of the dividing line, and 6 specimens of species B cross over to the
left. The specimens thus crossing over may be said to intergrade. Species
A therefore intergi’ades with respect to species B to the extent of 8% ; while
species B intergrades with species A in 6% of the individuals. The inter-
gradation is thus not the same in both species. This is a result of the skew-
ness of the frequency distributions and is the usual condition encountered
in practice, as stated. However, we want a single figure which will I'epre-
sent the intergradation between two species. This is obtained by taking the
simple average of the two figures. The intergradation of species A and B
in this hypothetical case is therefore 7%.
We will now take up an actual case and see how this measure works in
practice. Let us take the case of Sciaenops ocellatus, the northern Atlantic
and Gulf coast populations of which diverge sufficiently to place them well
up in the following gradated series of examples. The two populations diverge
chiefly in the frequency distribution of the number of rays in the second
dorsal fin for which the data are given in Table VII, p. 267. These data are
Text-figure 2.
Frequency polygons of the number of articulate rays in the second dorsal
fin of two populations of Sciaenops ocellatus, based on data given in
Table VII, the actual number of specimens being represented. The
solid line represents the Chesapeake Bay population; the broken line
represents the Texas population. The dotted vertical line represents
the dividing line between the two polygons as used throughout the
present discussion for the purpose of determining and measuring the
degree of intergradation.
represented graphically by Text-fig. 2 which illustrates the frequency poly-
gons of the two populations and the dividing line that forms the basis of
determining the measure of intergradation as used throughout the present
discussion. Of the northern population 15 specimens cross over to the right
of the dividing line, or 14.42% of the composite sample studied comprising
104 specimens; and 50 specimens of the Gulf coast population cross over to
the left of the line, or 44.64% of the sample which comprises 112 fish. The
simple average of these two percentages, in round figures, is 30, which rep-
1938]
Ginsburg : Arithmetical Definition of Species
259
resents the measure of intergradation of those two populations. In practice,
it is usually not necessary actually to construct the polygons; but the point
where the dividing line is to be drawn may be determined by inspection after
arranging the data in a frequency distribution table, preferably in the form
of percentages (see Table I).
This measure of intergradation, which is uniformly employed in this
paper, has the following statistical basis. If the histograms representing
the two populations compared be constructed on a percentage basis, the area
enclosed by the two overlapping histograms, expressed as a percentage of
the sum of their separate areas, equals the measure of intergradation deter-
mined as indicated above. In other words, the measure of intergradation as
employed for our present purpose represents the measure of the area en-
closed by the two overlapping histograms expressed as a percentage. This
may be illustrated graphically by the pair of intergrading populations of
Sciaenops ocellatus.
TABLE I.
Frequency distribution of the number of articulate fin rays in the sec-
ond dorsal of two races of Sciaenops ocellatus, expressed as percentages of
the total number of specimens counted of each race, respectively.
Locality
Number of dorsal rays
(percentages) :
Totals
22
23
24
25
26
Chesapeake Bay
3.85
31.73
50.00
13.46
.96
100
Texas coast
6.25
38.39
42.86
12.50
100
The smaller of the over-
lapping percentages
6.25
38.39
13.46
.96
59.06
Table I gives in percentage form the data represented in Table VII, p.
267. This is necessary for the purpose of constructing the graphs because
the number of specimens in the two samples is not the same, the usual con-
dition in practice; whereas, in order to show the normal amount of overlap
of the two populations it is necessary to have two samples containing the
same number of individuals. Text-figs. 3 and 4 illustrate graphically the
data presented in Table I. Text-fig. 3 shows the overlapping histograms of
the two populations, the part of each histogram which overlaps the other
being distinctively shaded. In Text-fig. 4, drawn to the same scale, the two
histograms are shown side by side with the shaded areas the same as in
Text-fig. 3. Text-fig. 4 shows at a glance the approximate relation of the
sum of the two shaded areas, or the area enclosed by the overlapping histo-
grams, to the sum of the areas of the two histograms. Precisely, this area
equals 30%, in round figures, of the sum of the areas of the two histograms.
The measure of intergradation is therefore 30% in this particular case.
This result may be obtained by adding the smaller of the overlapping fre-
quencies, as shown in Table I, and dividing the sum by 2 in order to get the
percentage of the area overlapped to the sum of the two histograms; or by
the simple arithmetical calculation as indicated above.
The measure of divergence may be indicated as a percentage also, by
subtracting the measure of intergradation as determined above from 100.
The calculation, and hence the number denoting intergradation, or di-
vergence, is always based on the character showing the greatest degree of
divergence. There may be some question as to the adequacy of the use of a
single character, since we know that populations usually differ in several
characters, some of which such as color differences are not readily expressible
in definite figures. Some attempts have been made to combine several char-
acters and express the hypothetical sum of their differences by a single fig-
260
Zoologica: New York Zoological Society
[XXIII :13
ure. However, I am not at all satisfied that such attempts have succeeded in
producing a measure which will more satisfactorily express for our purpose
the essential biological facts. Besides, in studying and combining the data
for several characters the question will always come up as to just where to
draw a line since practically it would be almost impossible to study them
all. While not entirely adequate the present measure should prove sufficient
for practical purposes, and conclusions arrived at as a result of studies by
the standard statistical formulae have been based on a single character. Cer-
tainly in the determination of the differences between species or subspecies
it is the character which shows the greatest divergence that is the important
one to consider. In the method here employed provision is made for taking
into consideration other differentiating characters besides the one showing
the greatest divergence (see p. 276) ; although such characters are considered
in a general way and not expressed in terms of definite figures. This is
probably the best that may be done for the present.
Arithmetical Definition of Species, Subspecies and Race.
Using the above measure, it is proposed tentatively to limit the desig-
nation of species and its subdivisions as follows: Other things being equal,
a given population is to be considered a race with respect to another closely
related population when the average intergradation of the character show-
ing the greatest divergence is between 30% and 40%; a subspecies consti-
tutes a population intergrading between 15% and 25% ; it is to be considered
a full species when the degree of intergradation is not more than 10%. Con-
Text-figure 3.
Overlapping histograms of the number of articulate rays in the second
dorsal fin of two populations of Sciaenops ocellatus; based on the same
data as Text-figure 2, but the number of specimens in each class ex-
pressed as a percentage of the entire sample studied. The solid line
represents the Chesapeake Bav population; the broken line represents
the Texas population; the hatched space represents the area by which
the latter histogram overlaps the former, and the stippled space rep-
resents the area by which the former histogram overlaps the latter.
1938]
Ginsburg: Arithmetical Definition of Species
261
Text-figure 4.
The same two histograms represented in Text-figure 3, separated and placed
side by side to give a better picture of the relation of the shaded areas
to the sum of the areas of the two histograms. In this particular case
the sum of the two shaded areas is 30%, in round figures, of the sum
of the areas of the two histograms, and this number represents the
measure of intergradation in this particular case. The series of three
Text-figures are graphic illustrations showing, by reference to a par-
ticular example, how the measure of intergradation as employed for
our present purpose, is derived.
comitantly, the divergence between races is 60% to 70% ; between subspecies
75% to 85% ; and full species diverge to an extent of 90% or more.
For the benefit of those who are used to thinking in terms of graphs the
above definition may be paraphrased as follows : When the area enclosed by
the two overlapping histograms, constructed on a percentage basis, equals
30 to 40% of the sum of their separate areas, the two populations are to be
considered as races; they are considered subspecies when the overlap is 15
to 25% ; they are full species when the overlap is 10% or less.
The above proposed boundary lines are discussed on page 275.
The Variety and Further Possible Subdivision of the Species.
In the above definition of the two principal subdivisions of a species,
namely, the subspecies and the race, the maximum intergradation allowed
for any population to be included under the latter category is 40%. All
pairs of populations which intergrade to a greater extent than 40% are pro-
posed to be grouped under the general category of “variety” and are not
further analyzed in our present preliminary study.
In addition to the subdivisions of a species proposed herewith it is
evident that the number of such subdivisions may be readily increased by
simply narrowing the boundary lines, which are arbitrary anyway, in this
completely gradated series. More extensive studies may indicate the de-
262
Zoologica: New York Zoological Society
[XXIII :13
stability of increasing the subdivisions, such as inserting a category between
the subspecies and the race, and further subdividing the inclusive category
here designated as the variety and consisting of populations which inter-
grade by more than 40%. The number of categories which may be found
desirable to recognize may differ with particular genera or groups. Any
such further subdivision should be based not on the rule of the thumb, but
on the study and correlation of a sufficient number of actual cases. Most
likely in the majority of cases the subdivisions of a species proposed here-
with will be found sufficient.
Sampling.
The present study would be incomplete without a consideration of the
question of sampling. Although I have tried to avoid the use of complex
statistical formulae, it will be noted that the present study is largely statisti-
cal in its nature. Indeed, if taxonomy is ever to be placed on a high scientific
plane it will perforce have to become to a large extent statistical in its
methods; although, quite likely, simple statistical deductions will be found
sufficient in most cases. One sometimes comes across the statement in
taxonomic works that a certain species is a “statistical species,” sometimes
stated in a rather disparaging manner, as though such a species is not of
much account. However, probably the majority of closely related species of
fishes are “statistical species.” The reason taxonomists were able to make
this distinction between “statistical species” and those supposedly not
statistical, is because their studies were largely based on but a few speci-
mens, as noted above (p. 255). A “statistical species” is simply one which
diverges from a closely related species to a comparatively low degree so
that even the study of a few specimens shows the close approach or even
intergradation of the frequency distributions of the differentiating
characters.
The importance of proper sampling in statistical studies in general is
universally appreciated, and it is not necessary to consider the subject here
at any length. It will be sufficient here to state that, except in a few cases
such as the complete enumeration of the population of a country during a
census, the description of certain attributes of a population is in reality a
description of the attributes of the sample which has been studied, and the
value of the description is dependent on how nearly the sample is represen-
tative of the population as a whole.
The same is true of the description of species, subspecies, races or
varieties. If a species is described from a single specimen or from a few
specimens, the account in reality represents a description of those specimens.
While in many cases such a description is sufficient for practical purposes to
identify and distinguish the species, in many other cases such an account
will be found entirely inadequate. Related species may be so close that it
takes the detailed study of many specimens to establish their divergence. In
such cases proper sampling becomes of importance; although in the case of
populations which reach a sufficiently high degree of divergence to be
regarded as full species, sampling is not of as transcendent importance as
in populations showing a relatively higher degree of intergradation.
In my own studies during which the data presented below were obtained,
I was impressed time and again with the importance of proper sampling.
It was noted frequently that specimens in the same lot bearing the same
data, evidently having been obtained in one or but a few drags of the net
in the same place at the same time, and consequently, most likely having a
common, immediate genotypic origin, would tend to group themselves, in a
predominant manner, within a narrowly circumscribed space, sometimes even
near either end of the frequency distribution of their species or race as a
whole. This was noted especially in cases where the specimens in the lot were
1938]
Ginsburg: Arithmetical Definition of Species
263
of nearly the same size. Consequently, it may be readily appreciated that if
the sample studied be obtained in one, or but a few drags of the net made
at the same time in the same place, it is quite likely that it would not give
a true picture of the population in many cases.
In obtaining the data presented below due attention was paid to the
question of sampling. As stated above, the data were obtained incidentally
in the course of taxonomic studies of the fishes. They were taken from
lots of specimens obtained at different times by various collectors. In no case
was a definite plan of sampling the particular population devised and carried
out. Under the circumstances, the method of sampling which I adopted was
as follows.
For the sake of brevity and clarity it is proposed to designate all the
specimens from which the final data in a Table are drawn, as the composite
sample and the smaller samples which go to make up the composite sample,
as constituent samples. It has been stated that if the composite sample is
obtained in one drag of the net, that is, it has but one constituent sample, it
will quite likely not give a true picture of the population. The greater the
number of constituent samples the more nearly will the data approach the
true distribution of the population as a whole. As a consequence of these
premises, it was my aim to include as many constituent samples as it was
possible to obtain from the available material. If, for instance, I had 25
containers of specimens representing as many lots of fish taken on different
dates, in different places, and 20 containers had but 1, 2, or 3 specimens each,
while the other 5 containers had much larger numbers, the 20 smaller lots
were included in the study, and only part of the specimens of each of the
larger lots. Of course, my sampling was limited by the material available,
but in every case I tried to come as near to my aim as was possible. The
number of constituent samples will be stated hereafter under each example
cited, so that the reader may judge as to the adequacy of the sampling.
(For the meaning of the notation adopted see footnote on p. 264). I am
confident that in most cases at least the given distributions represented by
the composite samples are more or less fairly representative of their
populations for practical purposes.
In some cases the number of specimens studied were too few to con-
stitute a satisfactory composite sample, such as in case of the two races of
Hippocampus zoster ae, in H. punctulatusfi and in Gobiosoma bosci and G.
robustum. Since it is my hope that the method adopted herein will serve
as a useful guide for taxonomists who frequently and of necessity have to
work with quite small samples, these examples are included in our series.
Further remarks regarding sampling in the above three cases are given
under their accounts.
Examples of Varieties.
Cynoscion regalis. This species is discussed below in another connec-
nection and the frequency distributions of the number of dorsal rays are
given in Table IX, p. 269. Turning to that Table and comparing the popula-
tion of Chesapeake Bay with that from the east coast of Florida, it will be
noted that the dividing line to be drawn according to the simple method
outlined above, is between the columns representing 27 and 28 rays. Of the
Chesapeake Bay population 21 specimens of a total of 40, or 52.50% of
the composite sample, cross over to the left of the dividing line. The
percentage of intergradation of the Chesapeake Bay population as compared
with that of Florida is therefore 52.50. Likewise, of the Florida population
2 The notation for subspecies employed in this paper is in accordance with the suggestion
made on p. 284.
264
Zoologica: New York Zoological Society
[XXIII :13
43 specimens out of a composite sample of 110, cross over to the right of the
dividing line, representing an intergradation of 39.09%. The simple average
of these two intergradations, 46 in round figures, represents our measure
of intergradation. The measure of divergence in this case is 100 minus 46,
or 54%. Constitution of composite samples: Chesapeake Bay 1 (4) 3, 2 (4),
7 (1), 8 (1), 13 (1); Florida 5 (1), 28 (1), 77 (1). The two larger con-
stituent samples from Florida do not bear any more definite data than the
locality and month and each one possibly contains more than one constituent.
Cynoscion nebulosus. The well known spotted sea trout of the southern
states is a common food and game fish from Chesapeake Bay to Texas, rang-
ing northward, in diminishing numbers, to New York. For an account of
the species the reader may be referred to any general work dealing with
the fishes of the region where it occurs, such as the “Fishes of Chesapeake
Bay” by Hildebrand and Schroeder (1928). No extensive study has been
made as yet of its racial differentiation; but there is found a statistically
measurable difference in the number of dorsal fin rays in fish from the Gulf
coast as compared with those from Chesapeake Bay on the Atlantic Coast.
It is possible that a more extensive study will reveal other characters show-
ing a greater degree of divergence, but this is quite doubtful. At any rate,
the dorsal fin ray count evidently shows sufficient divergence to be useful as
an illustration in the present study.
TABLE II.
Frequency distribution of the number of articulate rays in the second
dorsal of Cynoscion nebulosus.
Number of rays in second dorsal
23
24
25
26
27
Chesapeake Bay
1
17
53
31
6
Texas coast
1
12
38
48
5
The dividing line in this case is drawn between the columns repre-
senting 25 and 26 rays. The Texas population intergrades by 49.04%, while
the Chesapeake Bay population intergrades 34.26%. The average inter-
gradation in this case is 42% and the divergence 58%. Constitution of
composite samples: Chesapeake Bay 1 (15), 2 (3), 3 (2), 5 (2), 6 (2), 16
(1), 43 (1) ; Texas 1 (2), 2 (4), 3 (2), 4 (2), 5 (3), 6 (1), 7 (2), 10 (1),
17 (1), 18 (1).
Although this example does not form a happy choice for the purpose of
illustration, it is included in our series because similar cases no doubt will be
encountered in practice. It will be noted that if the polygons representing
the above data be constructed they will overlap at three points. Consequently,
our method of drawing the dividing line and calculating the intergradation
3 For the sake of brevity the sampling of every case cited in this paper is indicated by
figures which have the following meaning. As suggested above (p. 263), the entire number of
specimens of a given population, on which the final distribution in any one Table is based, is
designated as the composite sample, while each lot of specimens bearing the same data is known
as a constituent sample, a variable number of constituents going to make up the composite
sample, depending on the particular population used as an example. In the notation adopted a
figure outside a parenthesis indicates the number of specimens in one or more constituent samples,
while a figure within a parenthesis gives the number of constituent samples each one of which
contained the number of specimens indicated by the preceding figure. Thus, the composite sample
of C. regalis from Chesapeake Bay was made up of 11 constituent samples, four of which had 1
specimen each, four constituents had 2 specimens each, while the remaining three had 7, 8 and 13
specimens, respectively.
1938]
Ginsburg: Arithmetical Definition of Species
265
does not apply strictly to the present case. However, it will be noted also
that if one specimen from the Texas population be dropped at one extreme
end and two specimens from the Chesapeake Bay population be dropped at
the other end, the two polygons will overlap at one point as usual. This small
irregularity caused by one and two specimens out of a little over a hundred
in either composite sample, very likely is caused by the imperfection of
sampling, and any such irregularity at either extreme where the number of
specimens are few, is not likely to be compensated readily. At any rate,
whatever the cause, for the purpose of computing the degree of intergrada-
tion by our present method, this small irregularity is disregarded. It is to
be noted that computing by the standard formula (see Table XVII) the
degree of difference between these two distributions almost reaches a
“significant” figure.
Examples of Races.
Bairdiella chrysura. This is a very common fish on the east coast of the
United States. A recent and quite extensive account of the life history of
the species has been published by Hildebrand and Cable (1930). Like the
preceding species its races have not been studied extensively as yet, but
there is a difference in the frequency distributions of the dorsal fin rays
between southern and northern fish, as follows.
TABLE III.
Frequency distribution of the number of articulate rays in the second
dorsal of Bairdiella chrysura.
Number of rays in second dorsal
19
20
21
22
23
Chesapeake Bay
2
11
26
28
3
Texas coast
3
19
37
12
If a line be drawn between the columns representing 21 and 22 rays, it
will be found that the Chesapeake Bay population intergrades with that of
the Texas coast to the extent of 55.71%, while the Texas population inter-
grades 16.9%, giving an average intergradation of 36% and a divergence of
64%. Sampling: Chesapeake Bay 1 (14), 2 (8), 3 (6), 4 (3), 5 (2) ; Texas
1 (6), 2 (1), 4 (2), 5 (1), 6 (1), 19 (1), 25 (1).
Fundulus confluentus. This is a rather common, small, cyprinodont fish
originally described from Lake Monroe, Florida, by Goode and Bean (in
Goode 1879, p. 118). The original description is in error in some important
details, and the species has been confused with related species by all later
authors which I consulted. However, it is a well marked species which may
be distinguished without undue difficulty from its congeners occurring with
it through a greater part of its range. An account of the species will be
included in a revision of the genus which is under preparation. The Chesa-
peake Bay population of this species differs racially from that of Florida.
In addition to a rather slight and variable difference in the color pattern, the
structural character showing the greatest divergence is found in the number
of rays in the anal fin as follows.
266
Zoologica: New York Zoological Society
[XXIII :13
TABLE IV.
Frequency distribution of the number of anal rays in Fundulus
confluentus.
Locality
Anal rays
9
10
11
Norfolk, Va.
23
35
Beaufort, N. C.
19
1
Florida
2
21
2
It will be noted that the Chesapeake Bay population tends to have a
markedly lower anal fin ray count. The population from North Carolina
nearly agrees with that of Florida in the number of fin rays, but the color
pattern is more nearly like that of the Chesapeake Bay population. Drawing
the dividing line between the columns representing 9 and 10 rays and com-
paring the Chesapeake Bay population with that of Florida, it will be found
that the former population intergrades by 60.34% and the latter by 8%;
or an average intergradation of 34 % . Sampling : N orf oik 3 ( 1 ) , 9 ( 1 ) , 14 (1),
32 (1); Florida 1 (6), 2 (1), 5 (1), 12 (1). This is the only example, of
all those cited here, in which the well marked modes of both populations
fall on the same side of the dividing line.
Hippocampus zosterae. In reviewing the species of its genus found in
American waters I (1937) studied material of this species from Pensacola,
Captiva Pass and Key West, and the data presented herewith are taken from
my paper, where the question of racial differentiation is taken up in greater
detail. The populations from the three localities differ, on the average, in
the number of trunk segments and the number of rays in the pectoral fin,
the former character showing the greatest divergence as follows.
TABLE V.
Frequency distribution of the number of trunk segments of Hippo -
campus zosterae.
Locality
Trunk segments
9
10
11
Key West4
4
16
1
Captiva Pass
5
12
1
Pensacola5
7
6
It is to be noted that the greatest divergence exists between the
Pensacola and Key West populations, while the Captiva Pass population is
somewhat intermediate but nearer to that of Key West. This gradual dif-
ferentiation with latitude is a frequently recurring phenomenon which is
well known to students of fishes. In a case of this kind and in the absence
of a more elaborate study of the species, we may compare the extremes. If
a line be drawn between the columns representing 9 and 10 segments, it will
be found that the Key West population intergrades with that from Pensacola
to the extent of 19.05%, while the latter intergrades with the former by
46.15%, giving an average intergradation of 33%. Sampling: Pensacola 1
(2), 11 (1) ; Key West 1 (6), 2 (1), 3 (1), 4 (1), 6 (1).
4 Including 4 specimens from Newfound Harbor.
6 Including 1 specimen from Apalachicola.
1938]
Ginsburg: Arithmetical Definition of Species
267
The available material of this species is not sufficient to constitute a
satisfactory sample ; but the difference between the populations is significant,
especially when considered in connection with the small spread of the
frequency distribution, and approximately this difference very likely will be
found to exist after a more satisfactory sampling (see also remarks on
p. 263).
Leiostomus xanthurus. This species is the well known spot, a common
market fish on the east coast of the United States. The most comprehensive
account of its life history published so far is that by Hildebrand and Cable
(1930). Almost nothing is known now regarding the racial differentiation
of the species; but I found a significant difference in the number of rays in
the second dorsal on comparing fish from Chesapeake Bay with those from
the coast of Texas, as follows.
TABLE VI.
Frequency distribution of the number of articulate rays in the second
dorsal of Leiostomus xanthurus.
Locality
Number of rays
in second
dorsal
28
29
30
31
32
33
Chesapeake Bay
5
25
33
18
Texas coast
1
25
50
20
6
1
The dividing line is drawn between the columns representing 30 and
31 rays. The Chesapeake Bay population intergrades 37.04% and that of
the Texas coast 26.21% giving an average intergradation of 32%. Sampling:
Chesapeake Bay 1 (3), 2 (2), 3 (5), 4 (1), 5 (1), 10 (1), 20 (2) ; Texas 1
(5), 2 (2), 3 (1), 4 (1), 5 (1), 6 (1), 17 (1), 19 (1), 20 (2).
Sciaenops ocellatus. This species is the well known redfish in the mar-
kets of the Gulf coast, the celebrated channel bass of sportsmen. For an ac-
count of the species the reader may again be referred to Hildebrand and
Schroeder (1928). The racial differentiation of this species likewise has
not been studied to any extent, but there is a significant difference in the
number of rays in the second dorsal when the Chesapeake Bay population
is compared with that of the Gulf coast as follows.
TABLE VII.
Frequency distribution of the number of articulate rays in the second
dorsal of Sciaenops ocellatus.
Number of rays in second dorsal
22
23
24
25
26
Chesapeake Bay
4
33
52
14
1
Texas coast
7
43
48
14
The dividing line in this case is drawn between the columns represent-
ing 24 and 25 rays. The intergradation of the Texas coast population is
44.64%, that of the Chesapeake Bay population 14.42%, or an average inter-
gradation of 30%. Sampling: Chesapeake Bay 1 (1), 2 (1), 3 (1), 5 (1),
268
Zoologica: New York Zoological Society
[XXIII :13
6 (1), 22 (1), 28 (1), 37 (1); Texas 1 (5), 2 (1), 4 (2), 5 (1), 8 (1),
10 (1), 12 (2), 21 (1), 29 (1).
Examples of Subspecies.
Hippocampus punctulatus2 and H. hudsonius2 . These seahorses from
the east coast of the United States and the coast of Cuba have been gener-
ally recognized as independent species by American ichthyologists. How-
ever, they have been badly confused, and their structural characters have
been stated incorrectly in most current accounts. A discussion of their mor-
phological and geographical limits is given in my review of the species of
Hippocampus (1937) where it is determined that the character showing the
greatest divergence between them is found in the number of caudal seg-
ments, as follows.
TABLE VIII.
Frequency distributions of the numbers of caudal segments of Hippo-
campus hudsoniuso and H. punctulatus 2.
Locality and subspecies
Number of caudal segments
33
34
35
36
37
38
39
H. hudsonius-2
Chesapeake Bay to
Maine
3
7
15
7
4
North and South
Carolina
1
7
8
1
Mississippi to Texas
1
5
10
2
H. punctulatus*
Florida and Cuba
1
8
10
7
As in H. zosterae (see p. 266), the greatest divergence is found between
the northern population, Chesapeake Bay to Maine, and the southern, Florida
and Cuba. The population from the Carolinas and that from Mississippi to
Texas are somewhat intermediate but nearer to the northern population.
Other characters as well gradually differ with latitude. Although there is
a gradual change in morphology with latitude, it nevertheless seems desir-
able to recognize two subspecies as discussed on page 277.
In this case the dividing line is drawn between the columns represent-
ing 36 and 37 caudal segments. Comparing the extreme northern population
of hudsonius2 with the extreme southern population, punctulatus 2, the for-
mer intergrades the latter by 27.78%, and the latter intergrades the former
by 26.92%, or an average intergradation of 27%. Also, in order to show
the normal morphological range of hudsonius2 as a whole, the data from
North and South Carolina and Mississippi to Texas are combined with those
from Chesapeake Bay and northward. Combining the data as indicated,
the northern H. hudsonius2, intergrades with the southern H. punctulatus a,
to the extent of 33.80%, while the latter intergrades by 26.92%, giving an
average intergradation of 30%. The composite sample of H. hudsonius2 con-
sists of 54 constituent samples as follows: 1 (43), 2 (7), 3 (3), 5 (1). "Some
of the constituents having more than one fish do not have the data any more
specific than the name of the state on the coast of which they were captured,
and they quite likely represent more than one constituent. At any rate, the
number of constituents in the composite sample is not less than 54. The
composite sample of H. punctulatus2 consists of 23 constituent samples, as
follows: 1 (21), 2 (1), 3 (1). The specimens in the composite sample of
1938]
Ginsburg: Arithmetical Definition of Species
269
punctulatus 2 are very few, but they represent the limit of my available ma-
terial. While lacking in numbers it consists of many constituents and pos-
sibly fairly represents its population (see also remarks on p. 263).
Cynoscion arenarius2 and C. regalis2 • The differences between these
two common subspecies from the east coast of the United States have been
discussed by me (1929) and the reader is referred to that paper for details.
In grown specimens the character showing the greatest divergence seems
to be found in the number of gill rakers on the outer gill arch; while the
number of dorsal rays shows the next greatest divergence. However, the
number of gill rakers is not susceptible of precise determination for reasons
stated in the paper cited. Should it be found possible to discount the diffi-
culties inherent in a precise determination of that character, it is doubtful
whether it would prove to show greater divergence than the number of dorsal
rays. On the other hand, the latter character is susceptible of absolutely
precise determination at all stages of growth, soon after the fin rays have
developed in the young fry. For the purpose of the present discussion it
may be assumed that that character shows the greatest divergence.
TABLE IX.
Frequency distribution of the number of articulate dorsal rays in Cyno-
scion regalis2 and C. arenarius2.
Locality and subspecies
Number of
dorsal rays
24
25
26
27
28
29
C. regaliSi
Chesapeake Bay
North and South
1
5
15
16
3
Carolina
Cape Canaveral and
1
6
21
12
Fernandina, Fla.
3
15
49
35
8
C. arenariusi
Louisiana and Texas
2
35
59
22
1
The dividing line in this case falls between 26 and 27 rays; C. arenarius2
intergrading 19.33% and C. regalis2 16.32%, resulting in an average inter-
gradation of 18%. The sampling of the populations of regalis2 from Chesa-
peake Bay and from the coast of Florida has been indicated above (p. 264) ;
while that from the Carolinas is as follows: 3 (2), 4 (1), 7 (1), 8 (1), 15
(1). Of arenarius2 69 specimens do not have any specific data by which the
number of constituent samples could be determined; the rest consists of 14
constituents as follows: 1 (9), 2 (1), 5 (1), 6 (2), 22 (1).
Examples of Species.
Hippocampus regulus and H. zosterae. The former species is closely
related to the latter, the races of which have been discussed above. More
extensive accounts of the two species and a discussion of their relationship
are given in my (1937) review. It will be sufficient to state here that be-
sides its somewhat smaller size, and a slightly greater average number of
trunk segments, H. regulus differs chiefly from H. zosterae in having a
smaller number of caudal segments and fewer dorsal rays, the greatest di-
vergence occurring in the latter character, as follows:
270 Zoologica: New York Zoological Society [XXIII :13
TABLE X.
Frequency distribution of the number of dorsal rays in Hippocampus
regulus and H. zosterae.
Locality and species
Number of dorsal rays
10
11
12
13
14
H. regulus
Mississippi and Texas
2
15
1
Campeche, Mexico
5
H. zosterae
Key West, Fla.0
4
13
5
Captiva Pass, Fla.
16
2
Pensacola, Fla.7
2
9
1
An inspection of the foregoing Table shows that this character tends to
be fairly constant within the limits of each species, there being no pro-
nounced racial divergence as was found above for the number of trunk seg-
ments of H. zosterae. The Key West population shows a tendency to have a
slightly smaller average — thus surprisingly more nearly approaching regulus
from Mississippi and Texas than the Pensacola race of zosterae approaches
regulus- — but the difference is slight. The divergence of the two species in
this case is measured by drawing a dividing line between the columns repre-
senting 11 and 12 rays. H. zosterae is thus found to intergrade to the extent
of 11.54% ; H. regulus intergrades 4.35% ; giving an average intergradation
of 8%. Sampling: H. regulus 1 (3), 2 (5), 5 (2) ; H. zosterae 1 (7), 2 (2),
3 (2), 6 (1), 11 (1), 18 (1).
Gobionellus boleosoma and G. shufeldti. Accounts of these two species
of gobies and a discussion of their relationship will be found in my (1932)
revision of the genus. Briefly, the two species differ in the maximum size
to which they attain, in their color pattern, in the relative length of the
ventral fin, in the extent of squammation in front of the dorsal, and in the
number of dorsal and anal rays. However, none of these differences is abso-
lutely decisive when each one is considered by itself, and individual fish
often can not be distinguished and identified with assurance by any one
single character, although there is usually no trouble in referring individual
fish to their proper species when all the differentiating characters are taken
in consideration. The greatest divergence between the two species is shown
by the number of anal rays, as follows.
TABLE XI.
Frequency distribution of the number of anal rays in Gobionellus boleo-
soma and G. shufeldti.
Species
Number of anal rays
11
12
13
14
G. boleosoma
3
74
4
G. shufeldti
3
33
1
6 Including' 3 specimens from Newfound Harbor and 2 from Biscayne Bay.
7 Including 1 specimen from Apalachicola.
1938]
Ginsburg: Arithmetical Definition of Species
271
The dividing line in this case is between 12 and 13; the intergradation
calculated for G. shufeldti is 8.11%, and for G. boleosoma 4.94%, resulting
in an average intergradation of 7%. The counts of the fin rays in these two
species are fairly constant throughout their ranges, there being no pro-
nounced racial differences in this respect. Sampling: G. boleosoma 1 (13),
2 (8), 3 (1), 4 (1), 6 (1), 39 (1) ; G. shufeldti 1 (2), 8 (1), 13 (1), 14 (1).
Paralichthys lethostigma and P. albigutta. Some recent investigators
have expressed doubt in regard to the distinctness of these two common
species of flounders. This doubt is apparently caused by the fact that the
chief differentiating characters are of a meristic nature, that is, quantita-
tive, and vary within rather wide limits. Consequently, when specimens
near the beginning or near the end of the two frequency distributions of
the two species, respectively, are encountered, they may be referred to either
species when any single character is considered. However, after preparing
tables for the several characters showing the normal frequency distribu-
tions of each species, it becomes an easy matter, with the aid of such tables,
to refer individual fish to their proper species. An extensive discussion of
the differences between these two species, among others, is included in a
revision of the genus which I now have completed in manuscript form. In
connection with that study, I examined over 500 specimens representing
both species, and out of this large number I encountered only one specimen
the status of which was in doubt. Besides the structural differences the two
species differ also in their color pattern. Without going here into details,
it may be stated that the least intergradation is found in the number of anal
rays, as follows.
TABLE XII.
Frequency distribution of the number of rays in the anal fin of Para-
lichthys albigutta and P. lethostigma.
NUMBER OF RAYS IN ANAL FIN
SPECIES
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
P. albigutta
1
I
4
10
15
16
26
17
11
5
5
P. lethostigma
2
5
10
15
25
27
31
13
14
9
1
1
Drawing the dividing line between 63 and 64 rays, it is found that P.
lethostigma intergrades 1.31%; but no specimens of P. albigutta cross over
to the left of the line, the intergradation of the latter thus being 0. The
average intergradation is then 1% in round numbers. The composite sample
of P. albigutta includes 70 specimens from the vicinity of Beaufort, N. C.,
and 16 from the vicinity of Corpus Christi, Texas, for which no detailed
data are available, probably containing a number of constituents. The rest
are from the Gulf of Mexico, except two specimens from the Atlantic coast,
and altogether consist of 16 constituents as follows: 1 (9), 2 (6), 4 (1). The
composite sample of P. lethostigma contains 13 specimens from Beaufort,
N. C., without any further data, but probably obtained on different dates;
7 other specimens from the Atlantic coast and 133 from the Gulf coast. The
sampling of all except the 13 is as follows : 1 (11) , 2 (4) , 3 (2) , 4 (3) , 5(1),
6 (1), 7 (1), 8 (1), 9 (1), 10 (1), 12 (2), 13 (1), 21 (1).
272
Zoologica: New Yoi'Jc Zoological Society
[XXIII :13
Gobiosoma bosci and G. robustum. An extensive discussion of the re-
lationship between these two species of gobies was published in my (1933)
revision of the genus. Besides a difference in the color pattern which, how-
ever, is not always distinctive, the chief structural characters differentiat-
ing the two species are: the number of dorsal and anal rays and the length
of the ventral. The numbers of fin rays intergrade in the two species; but
the length of the ventral expressed as the number of times it enters into the
distance between the ventral and anal origins, does not show any intergrada-
tion in the specimens measured, as follows.
TABLE XIII.
Frequency distribution of the length of the ventral in Gobiosoma bosci
and G. robustum, expressed as the numerical value of the ratio of the dis-
tance from base of ventral to origin of anal, to the length of the ventral.
Length of ventral in the distance from its base to origin of anal
1.2
1.3
1.4
1.5
1.6
1.7
1.8
1.9
2.0
G. bosci
2
4
8
2
1
G. robustum
3
7
2
4
In this case then intergradation is zero and divergence 100%. The
heading numbers for the classes in the above table represent the mid-values,
this example being an illustration of a character the measurements of which
form a continuous series. Sampling: G. bosci 1 (9), 2 (4); G. robustum
1 (5), 2 ( 1 ) , 4 (1),5 (1). The composite samples are entirely inadequate
but there is no question in regard to the specific distinctness of the popula-
tions compared (see also remarks on p. 263).
It should be stated also that the chief differentiating character which
distinguishes these two species is not susceptible of determination with a
high degree of accuracy. Although proportional measurements are widely
employed for separating closely related populations of fishes, of specific or
lower rank, such measurements can be determined only by a rather rough
approximation to their true value. The values of the measurements, more
likely than not, are apt to vary with the state and method of preservation
of the specimens and also with the observer making the measurements. Even
the same observer measuring the same specimens with the same instruments
is apt to obtain somewhat different results by successive trials; although
when sufficient care is exercised and a vernier caliper employed for taking
the measurements, the results usually are sufficiently accurate for practical
purposes even in the case of quite small specimens. Anyway, since the data
represented in the above Table are not entirely satisfactory we may cite the
following case which occupies the same position in our series, and which is
based on a character that is susceptible of being determined with almost
absolute accuracy.
Lepiclogobius y-cauda and L. guaymasiae. These two gobies were de-
scribed originally by Jenkins and Evermann (1888) from the Gulf of
California. The two species are very closely related, and all later authors
considered the above two names as synonymous. Indeed, the original de-
scriptions are not sufficient to distinguish the two species, are erroneous
in some important details, and it is very doubtful whether even the original
describers separated all of their material properly. Nevertheless, the two
1938]
Ginsburg: Arithmetical Definition of Species
273
species are quite distinct. The differences distinguishing these two species
will be taken up in greater detail in a revision of their genus which is now
in process of preparation. It will suffice for our present purpose to state
that the greatest divergence is shown by the number of pectoral rays as
follows.
TABLE XIV.
Frequency distribution of the number of pectoral rays in Lepidogobius
y-cauda and L. guaymasiae.
Species
Number of pectoral rays
20
21
22
23
24
25
26
27
L. y-cauda
L. guaymasiae
6
23
11
7
51
42
5
1
In this case also intergradation is zero and divergence 100%, as far
as the samples studied are concerned. Sampling: G. y-cauda 1 (1), 2 (2),
5 (1), 6 (1), 24 (1) ; G. guaymasiae 2 (2), 14 (1), 34 (1), 54 (1).
Paralichthys dentatus and P. lethostigma. The latter species was com-
pared above with P. albigutta, with which it showed a slight amount of in-
tergradation. When compared with another species of its genus, with
dentatus, it does not show any intergradation. P. dentatus has a more
northern distribution. It is the well-known summer flounder or fluke and
is an important commercial species. Its range extends from Cape Cod to
northern Florida. On the coast of North Carolina and southward its geo-
graphical range overlaps with its two common congeners which were dis-
cussed above. An extensive account of this species is included in my manu-
script referred to above. For a published account of the species the reader
is referred to “Fishes of Chesapeake Bay” by Hildebrand and Schroeder
(1928), this being the only species of Paralichthys occurring in that body
of water. For the purpose of the present discussion it may be stated that
P. dentatus has nearly the same frequency distribution of the number of
fin rays as P. lethostigma, but it differs decidedly in the number of gill
rakers, as follows.
TABLE XV.
Frequency distribution of the number of gill rakers on the lower limb
of the first gill arch of Paralichthys dentatus and P. lethostigma.
Number of gill rakers on lower limb
8
9
10
11
12
13
14
15
16
17
18
P. lethostigma
P. dentatus
7
67
66
6
4
3
18
44
35
11
The foregoing is an illustration of a case where a gap exists between
two species with reference to a given character. The composite sample of
lethostigma consists of 24 specimens from localities ranging from Albemarle
Sound, N. C., to St. John’s River, Florida, on the Atlantic coast, and 122
from the coasts of Texas and Louisiana (including 1 specimen from Apala-
chicola, Fla.). The Atlantic coast specimens include 13 from Beaufort,
N. C., without any more detailed data, probably a mixed lot, and 7 other
constituents as follows: 1 (3), 2 (4). The composite sample from the
TABLE XVI.
Summary of the preceding data showing the gradual continuity of intergradation and divergence arranged in
order of decreasing intergradation, or increasing divergence.
274
Zoologica: New York Zoological Society
[XXIII :13
e
£ V2
g o «
o s
ft
(SCO
rO
8
©4
Ci
CO
(M
00
00
CO
O t-H O 00 <0 rH
8
be
<D i £
g&g
«.£.§
0) -
s
03
° U)
CO
cvl
o
CO
00 rH o o
.*_> c3 ^ V>
C ^ vn c3
0, tjo3
^ c a
0)^0 °
Ph .5 H ft
O o o
hn cj to C
21^-B
c {* «w c3
0i fcfi o "2
ft H C ft
OJ -g O o
Ph .£ Q3 C,
53
"53
*
8
©
•S
©
to
8
85
s
o
83
o
8
O
8
*8
8
"8
8
"8
8
*8
8
"8
8
"8
8
"8
8
£
8
83 ©
"W
&5
a
©
a
8
ft
•<s>
©
8
ft
.8
©
s
a
'o 1
©
Co
CO
CO
co
to
C3 Bh
I?
8
O
coO
•<s>
§>3
>»
8
m
CO
S g
C0
o c3
8* 0)
8 rv
©
m
0)
.r8
ou
S JO
s> 8
Is
h w
•a 03
° X
. a>
CQEh
s
s
52
s
'e.ts
e> t
g -£ s
55 05
2^ §
kfe iuw ujn
8
pq
K a>
SPi
ti ai
53 a;
O 0)
„ -JZ
too
c$
-Q
M 3
Wr ■)
HO
53 T3
~ C
■S e«
o
SiS
Pi.:-.
, o
ado
§>■-£
© H
c- 8
*2 -2
53 o
e -=
0<C m Ci (L c5
CO
co ^ co
Ig 8 ©
eW
o a>
x a
•<s> 03
© Ifl
CO CP
O r8
o
^ o
8 °
•2 M
© 03
CO X
O <D
o
gs»
S t3
§,«
O
^ ft
3 “
'S S
•cO O
8W
cq
">H 0j
pq
^ CD
^ c3
<D
co rv
8 E?
r8 ^
r
8
5-
C3
CO
©
ti
CO
S S
e es
O CO
g «
a <u
• B-Ph
s
J-
S v
c3
sPP
S o>
Co
| a
8 o3
© ui
-K> <D
•S
s
® cS
CO rrt
'eiS
53
S
535
55
sg
co u
|3
r- x
CV3 v*
to fe
s
■S^
8
8
©
8 m
. o3
8 o
.8 °
S«H
CO > — i
© 8
SO
ft -8
ft -o
© © lX: ^ ©
^ h5 to a; o Uti
53 o
t, ej
.g >>
ho or-
® 5
o -2
•'S 8
55 53
>-4 Ch
1938]
Ginsburg : Arithmetical Definition of Species
275
Gulf coast is constituted as follows: 1 (6), 2 (4), 3 (4), 5 (1), 6 (1),
8 (1), 9 (1), 10 (1), 12 (2). 13 (1), 21 (1). The composite sample of P.
dentatus consists of 71 specimens from Chesapeake Bay constituted as fol-
lows: 1 (28), 2 (8), 3 (5), 5 (1), 7 (1) ; 37 specimens from Beaufort, N. C.,
for which no further data are available, probably a mixed lot; 7 specimens
comprising 6 constituent samples from North Carolina to Georgia.
Recapitulation.
The data presented above of a series of actual cases in widely dissimilar
groups of fishes, arranged to show the degree of divergence in a gradually
ascending series, are recapitulated in tabular form on page 274, in order to
give a birdseye view of the gradual continuity of the series. Table XVI
shows the gradualness of divergence or intergradation and that any line
drawn between the species concept and its subdivisions must be arbitrary.
The continuity of the series is not as gradual as may be desired. A more
evenly gradual continuity will no doubt be obtained by studying and citing
a greater number of examples. However, those cited seem sufficient for this
preliminary study to establish in a definite manner the gradualness of
divergence in nature. Beside their number, the kinds of examples cited
are not as well chosen as I would have liked them to be. I have now under
study other populations which show promise of furnishing better proof to
clinch the present thesis. However, the differentiation of those populations
has not been satisfactorily established as yet, and the accumulated data are
not of sufficient extent to enable me to use those data with assurance. While
the examples cited are sufficient to prove the proposition in this preliminary
paper, the subject evidently can not be expected to be exhausted in a single
article.
The Arbitrary Boundaries Between Species, Subspecies and Race.
The propriety and expediency of the arbitrary boundaries between the
three main taxonomic units proposed in this paper may now be considered
after having determined just how intergradation occurs in nature.
It may be suggested by a study of Table XVI that a more logical boundary
to draw between the species and subspecies would be between two other
pairs of population, namely, Paralichthys lethostigma and P. albigutta,
and Gobiosoma bosci and G. robustum. That is, two populations are to be
regarded as fully distinct species only when they differ in any single char-
acter in such a manner that there is no intergradation between them. How-
ever, this suggestion can be defended neither on theoretical nor on practical
grounds.
First of all it may be stated that there is nothing of an inherently
fundamental nature in such a boundary line. In a gradually continuous
series such as we are dealing with, it is just as arbitrary as any other
dividing line which may be proposed. Even when a gap exists between
two species it may be possible to devise such a measure as will express
the degree of the extent of the gap and the series denoting divergence con-
tinued in a gradual manner, although for our present purpose it is not
deemed necessary to devise such a measure.
FOOTNOTES TO TABLE XVI.
8 This column is added for the purpose of comparing the method here employed with the
standard method, as discussed on page 279. However, this does not form an essential part of our
method, and the column as well as the discussion may be left out of consideration by those
readers who are not interested in making this comparison.
9 The sampling of the pairs of populations of S. ocellatus and H. hudsonius is not altogether
comparable as discussed on page 277. When the extreme populations of the latter species are
compared the intergradation is 27%.
276
Zoologica: New York Zoological Society
[XXIII :13
Second, it should be remembered that two diverging populations, espe-
cially when they reach such a high degree of divergence as to constitute
species or subspecies, usually differ in several characters. Sometimes a high
degree of divergence may be shown by two or more characters, although no
single character may show a divergence of 100%. In such cases individual
specimens may be referred with comparative ease to their proper species
by at least one of these characters falling decidedly near the mode of its
species. (This point is taken up more fully in my manuscript study of the
genus Paralichthys) . It is evident, therefore, that the boundary line under
discussion can not be defended on theoretical grounds.
On practical grounds also such a boundary line will not prove satisfac-
tory. If this boundary is adopted, possibly the majority or at least a large
percentage of closely related species of fishes now universally l’ecognized
will have to be reduced to subspecific rank because of the general or at
least frequent existence of intergrades. A case in point is the difference
between Gobionellus boleosoma and G. shufeldti which forms one of the pre-
ceding illustrations. Probably no taxonomist who would compare and study
authentic specimens of these two species would do otherwise but come to
the conclusion that they represent fully distinct and independent species.
In fact the differences between them are more numerous and saliently
marked than in many other species which are now generally recognized.
In spite of all this, their chief differentiating character intergrades in 8%
of the individuals of G. shufeldti, or 7% as an average between the two.
It is evident, therefore, that a boundary line drawn so that two popula-
tions in which the average intergradation of the most divergent character
is 7% or less are considered fully distinct, is not far fetched. The figure pro-
posed herein, 10%, is not based on sufficient definite data, but is chosen
chiefly for the simple reason that it represents a conveniently round number
in our decimal system. Since such a boundary line is arbitrary anyway, the
one proposed here tentatively is sufficient for the purpose of discussion. It
is evident that the boundary line will be drawn the more advantageously in
accordance with the facts of nature, the greater the number of cases which
are studied in detail and correlated, and quite likely will change with in-
creased knowledge. It may also be found advantageous to fix different boun-
daries in special cases or in certain groups in order to give in such cases an
interpretation of the facts which will more closely approach nature.
The proposed boundary between the subspecies and the race also seems
the most expedient as judged by the evidence considered.
In the arithmetical definition of species, subspecies and race given on
page 260, it is to be noted that gaps are left in the continuity of the per-
centages at the boundary lines between the species and subspecies on the
one hand and the subspecies and the race on the other. This is done on pur-
pose. Also, the phrase “other things being equal” is used advisedly. As a
matter of fact, other things usually are not equal. Populations, especially
those that reach a sufficiently high degree of divergence to be regarded
properly as species or subspecies, usually differ in many characters, the na-
ture, number and variability of which vary widely with the particular
populations, and all the characters have to be taken in consideration. Since
any single character does not absolutely determine the taxonomic status of
a population, it is evident that in appraising the sum total of values of the
other characters in which two populations differ, there will be room for
difference of opinion in cases where the difference between the two popula-
tions falls near any arbitrary lines which may be fixed. This is as it should
be. The very nature of the subject matter excludes the possibility, at least
for the present, of universal agreement in all cases. Consequently, the taxo-
nomic status of border line populations will be determined to a large extent
by a consideration of the other differentiating characters in addition to the
one showing the greatest divergence. Leaving gaps between the proposed
1938]
Ginsburg: Arithmetical Definition of Species
277
limits of our taxonomic units thus allows leeway for the exercise of judg-
ment in border line cases.
The necessity of exercising judgment in regard to populations falling
near an arbitrarily fixed border line is shown by two pairs of populations
cited above as illustrative examples, namely, the populations of Sciaenops
ocellatus, and those of Hippocampus hudsonius (including all populations of
the typical subspecies). Both pairs intergrade approximately by 30%. The
intergradation of the latter pair is even a fraction greater in the samples
examined. Nevertheless, the latter two populations are recognized as sub-
species, and the former as races only for the following reasons: (1) Specia-
tion in Hippocampus is rather unlike the usual. The species generally ap-
proach closely or even overlap (see Ginsburg 1937, p. 558 and passim ). Hip-
pocampus constitutes one of those groups in which different arbitrary lines
between the categories will possibly have to be drawn in order to adequately
represent the essential facts, as suggested on page 276. (2) The sampling of
the two l'espective pairs of populations is not strictly comparable. Of Sciae-
nops ocellatus two extreme populations are compared, while of Hippocampus
hudsonius intermediate populations are also included. Of the latter species,
when the extremes are compared (see p. 268), the intergradation, 27%, falls
within the arbitrarily fixed gap between the subspecies and the race as de-
fined on page 260, and is less then in the extreme populations of Sciaenops
ocellatus. (3) The populations of Hippocampus hudsonius diverge to a con-
siderable extent in a number of other characters also, such as the number
of dorsal rays, the number of pectoral rays, the relative length of the snout
and the relative depth. (For a more detailed discussion of their differences
see Ginsburg 1937, p. 557.) While the degree of intergradation in every
character is considerable, an appraisal of the sum total of all the differences
between the populations makes it evident that it is desirable to recognize
them as subspecies rather than races. But for the populations of Sciaenops
ocellatus no other well marked diverging characters are known at present.
The foregoing paragraph discusses some of the many factors involved
which need to be considered in forming a final conclusion in regard to the
status of a given pair of populations under study. A decision becomes es-
pecially difficult in cases where the samples available are manifestly inade-
quate, a frequent contingency in taxonomic practice. In such cases it would
take an experienced, able and careful taxonomist, one who has a highly de-
veloped intuitive capacity, to form a decision which may stand the test of
time and further, more adequate research. Even under the most favorable
circumstances opinions regarding the rank of a given pair of populations
may change as other, more divergent, characters are discovered. For in-
stance, it is possible that some other characters may be discovered by which
the populations of Sciaenops ocellatus may be shown to diverge to such a
degree as to constitute subspecies.
The arbitrary lines suggested in this first attempt are admittedly tenta-
tive; but I believe that they approach closely to those which may be fixed
after more extensive studies. Furthermore, any arbitrary line, no matter
on how many concrete examples it may be based, will be from its very nature
only generally indicative since all characters have to be considered in deter-
mining the taxonomic status of a pair of populations, as stated. Also, their
proper use will always depend on the experience and ability of the taxono-
mist. Nevertheless, although the arbitrary lines here proposed are tentative,
they are manifestly preferable to the rule of the thumb by which taxonomists
now seem to decide regarding the category of particular pairs of popula-
tions. At least, they form a basis for immediate practice, as a general guide
for taxonomists to determine whether any given pair of diverging popula-
tions are to be assigned specific, subspecific or racial rank. They also form
a basis for discussion, improvement and refinement.
Irrespective of the question of the adequacy of the arbitrary lines here
TABLE XVII.
Some commonly used statistics of the populations discussed above,10 based on the data presented in Tables II-XV.
278
Zoologica: New York Zoological Society
[XXIII :13
H
s
Q
05
tH
05
tH
TF
©3
o
00
rH
O
rH
O
00
O
rH
1
W
©
©i
LO*
co’
00
©3
co
©i
LO
©3
CO
tH
O
LO
1
rH
©i
t-H
©3
LO
rH
00
©3
Cl
rH
©1
05
CO
rH
LO
tH
00
©3
o
CO
©3
IH
O
^f
CO
05
LO
tH
rH
CO
©1
CO
©3
05
CO
00
H
rH
IH
05
CO
rH
00
tH
co
IH
tH
©3
00
IH
rH
O
o
O
rH
o
o
rH
o
o
O
rH
O
o
o
5
CO
05
LO
CO
co
LO
05
o
CO
LO
tH
00
co
CM
o
CO
LO
©1
CO
00
CO
CO
©3
tH
CO
O
o
LO
05
05
rH
LO
05
00
CO
GO
CO
©3
05
rH
©3
^F
CO
CO
tH
00
05
CO
05
CO
co
§
j
rH
rH
05
co
CO
^F 00
^F LO
tH LO
O CO
IH CO
©1 rH
00 V-
CO LO
H ^
©3 Ttf
o o
©3 tH
^ 05
tH ©3
tH LO
^ CO
©1 ©1
O O
^F CO
00 CO
^ 05
05 O
LO LO
©1 CO
CO O
CO ©3
rH rH
CO LO
tH 00
05 CO
C5 LO
LO LO
tH C©
LO Tj«
CO 05
CO to
rF LO
tH ©1
LO LO
CO ©3
©3 CO
rH rH
"cF CO
CO CO
W
o o
o o
o o
o o
O O
o o
O O
O rH
o o
o o
o o
rH rH
o o
o o
o o
o o
rF 00
O tH
LO 00
rH CO
rH CO
©1 ^
tH CO
O C5
©3 CO
tH rH
CO rH
tH CO
CO o
LO CO
2
^F H
o o
tH tH
©1 LO
©3 rH
00 O
rH O
IH ©1
05 rH
LO ©3
CO 00
LO rH
rH rH
O HF
rH
T— 1
rH rH
rH
tH rH
rH rH
t-H t — 1
rH
rH rH
co co
05 05
o tF
O ©1
CO LO
CO 05
O 00
©1
LO O
05 LO
CO IH
LO LO
©3 05
CO 00
05 rH
LO CO
©1 00
tH CO
O 05
CO 00
CO 00
©1 LO
^ 00
05 tH
00 00
^ CO
CO CO
O LO
00 05
tH TF
t
oo oo
05 O
tH LO
O 00
CO 05
LO 00
oo co
LO LO
CO LO
©3 05
©3 LO
O O
©3 CO
O LO
00 00
IH 00
00 tH
00 00
tH IH
Oj Op
00 tH
LO CO
CO ©3
©3 O
rH rH
tH CO
rH CO
©3 ©3
rH
LO rf
Ph
LO LO
LO LO
LO Tf
CO ©1
CO ©1
CO
tF LO
LO LO
LO LO
4
3
CO CO
LO rH
rH rH
oo
LO CO
CO ^F
O tH
rH ©1
05
o ^F
rH LO
rH V-
rH CO
©1
05 05
tH LO
05 CO
©3 tH
LO 00
00 o
CO CO
LO C l
CO ©1
rH CO
O CO
tH tH
O tH
CO 05
LO CO
IH CO
tH CO
LO ©3
©3 00
CO CO
©3 LO
©3 CO
IH tH
©3 CM
tH rH
o o
LO CO
05
rH LO
LO tH
LO LO
^ rH
©3 CO
IH
LO ©3
00 00
§
CO CM
xF ©1
©1 00
O CO
00
tH O
CO tH
00 00
©3 00
O 05
05 O
©3 tH
IH CO
^F rH
rH rF
tH ft
LO LO
rH ©
© 05
Cl Cl
© ©
ft CO
CO LO
tH LO
©3 O
©3 ©3
00 00
rH r — 1
^F rH
CO 05*
©3 Cl
©1 ©1
©1 ©1
rH
co co
©1 ©1
CO CO
©3 ©3
rH rH
rH rH
CO LO
©3 ©3
rH
>i
o3 cS
d
O
*-4-1
d
>»
cS
ffl
o
03 ©
0> .
fttf
Oj ^3
S3
M
4J <D
52 PC
o
s£
M C3
Cfi M
X <u
<u,c
►>>
d
m
0) 'S
cs
o3 o
0) o
O,
£7* w
» s
n> rS
Florida coast
Norfolk, Va.
Key West, FI
Pensacola, FI
Chesapeake E
Texas coast
a
M
m $
g*
o c;
ft
2 c$
d m
X <D
ei CO
CO ^
g
©
e
a
3
ft
o
(ft
£ o
co
•<s>
ft-
©
«*•
HO
CO
© „
'S'-
r©
©
£
qH
s'
§
CO 3
5=5
o
co
£
© -
©
©
oT
© ^
CO
©
CO
ft
CO
3
V-
£
e
&
co
Q) H
HO
co
S$
©
©
©
•s«
s s
©
co ©
§s
ft ft
CO
s
ft
co
CO £
^ ©
ft ?>
g e
pus zosterae
regulus
,s shufeldti
boleosoma
Si e
•<s> HO
so so
co S
ft -H
co
ft
. bosci
robustum
ius guaymasi
y-cauda
§
CC ©5
^*^
^ CO
-S °
^ ©
CO
ft
s
s
e
s
co
s
§
ft
©
r©
.©
©
CO v.
© -
©
•<s>
©
co -
© ~
©
ft3 ~
"3 _
■a-
§-
©
©
ft
§
© -
so '•
co
©
© -
§ -
©
©
ft
#©
©
co -
©
S3 -
©
©
ft
*©
©
© ^
-
o
©
CC -
© "
ft
© C
2
.© ^
s
5=
*<S>
£
ft
•2
a
s
ft
-©
V.
r©
ft
ft
ft
e
•-S-
©
•c*i
ft
•<s>
©
e
©
©
©
o
o
cq
fcn
£
H
CO
S3
o
Hi
CL
H
Oh
1938]
Ginsburg : Arithmetical Definition of Species
279
proposed, the fundamental thing to remember is that no matter where
drawn, the lines must be arbitrary. This is proved by the gradual series
here presented. It is either one or the other. Either the different categories
have no existence in fact, and in that case our use of the terms “species,”
“subspecies,” “race,” etc., should be abandoned. Or, if the categories do
exist in nature, they can be based only on relative degrees of divergence.
The sum total of experience obtained as a result of careful population studies
made up to now unmistakably shows that the latter interpretation represents
the true state of affairs in nature.
Comparison Between the Simple Method Employed in
This Paper and That in General Usage.14
Attention has been called previously (p. 256) to the use of the standard
method for comparing closely related populations as represented by the for-
mula ■ .. ==§ . This formula is not well adapted for our purpose,
V iiii- -|- Jcj2
as may be shown by a consideration of the several factors which go to make
up that expression.
First of all it may be stated that theoretically this expression in reality
is not a measure of divergence. Rather, it is meant to be a test indicating,
from a strictly mathematical point of view, the probable reliability of a
difference between two sets of data, and consequently the probable value
of any conclusions which may be drawn from such data. From a biological
point of view, the usefulness of this test is limited in such a problem as we
are confronted with. Since differences between populations exhibit all de-
grees of extensiveness, it follows as a consequence which can hardly be
doubted that some average differences between natural populations exist,
which are of such low magnitude as to be below the usually accepted mathe-
matical “significance” obtained by the use of the above formula with sam-
ples which are ordinarily limited in practice. Such small differences must
have a biological significance, although mathematically their significance
appears doubtful. This test merely serves as a warning to proceed with
caution in drawing conclusions in cases where the values are low, but it does
not prove that such low values do not have any biological significance. How-
ever, while the above expression is strictly speaking not a measure of di-
vergence, it may be used for that purpose. It is evident that, in general, the
greater the divergence between two populations the greater is the numerical
14 See also footnote to Table XVI, p. 275.
FOOTNOTES TO TABLE XVII.
10 The symbols at the head of the columns stand for the following : M = arithmetical mean ;
R = actual range of the distribution on the X — axis ; a = standard deviation ; N = number of speci-
mens in composite sample ; Em == probable error of the mean ; Mi — M2 = difference between
the means of the two populations compared ; Ed = probable error of Mi — M2.
11 Biological statisticians are now generally employing the standard error instead of the probable
error. . To reduce the figures given in this table to the standard error and the derivatives based on
it, divide the probable error of the mean, and the probable error of the difference between the means
by the modulus .6745, and multiply the figures in the last column, showing “significance,” by the same
number. The relative magnitudes of the comparative figures given in the last column will thus
remain the same.
12 The sampling of the populations of H. hudsonius is not altogether comparable with those of
5. ocellatus, as discussed on page 277. The statistics for the extreme northern population of the
former (see Table VIII, p. 268) are: M = 37.0556; cr — 1.0787; Em = 0.1213; and as compared
Mi — M2
with the southern population, punctulatus2, Mi — M2 = 1.2094, Ed = .1745 and —
Ed
= 6.9, or nearly the same as when the total sample of hudsoniusz is compared with punctulatus 2
(see discussion on p. 280).
13 Omitting the one specimen at the extreme right of the frequency distribution (see Table
XII, p. 271), the statistics for lethostigma are : M = 68.1645; R = 11 ; a — 2.1134; Em — 0.1156;
Mi — M2
and as compared with albigutta Mi — - M2 = 9.4258 ; Ed = 0.1752, and — — — 53.8.
280
Zoologica: New York Zoological Society
[XXIII :13
value of this expression. Consequently, by the use of the above formula
comparative figures are obtained which may indicate the relative degree of
divergence of a pair of populations in a given series of pairs investigated.
The question remains how useful is the standard method for our present
practical purpose.
It will be noted that the above expression depends on: (1) the differ-
ence between the means which may be determined directly from the data
by a simple arithmetical process of averaging, and (2) on the probable error
which depends, in its turn, on a number of factors. The statistical formulae
for the determination of the probable error are as follows:
The probable error then depends on the number of specimens in the com-
posite sample and on the value of sigma. Now, the magnitude of sigma is
determined primarily, (1) by the extent of the spread of the frequency
distribution, the greater the spread the greater its value; and (2) by the
form of the distribution; one having a sharply pronounced mode with the
number of specimens diminishing rapidly at either end will show a com-
paratively smaller value for sigma, while a distribution the curve of which
has a flattened form, or is skewed, will show a relatively greater value. (I am
not considering very irregular distributions which may be due to faulty sam-
pling, heterogeneity of the material or other causes.) It follows therefore
that the value of the probable error depends chiefly on three factors, namely,
(a) the number of specimens in the composite sample, (b) the spread of
the distribution, and (c) the form of the distribution. Some of the factors
involved are shown in Table XVII. The influence of the form of the dis-
tribution on the value of sigma is well shown by the two populations of
Fundulus confluentus, for instance.
With these brief remarks we may now compare the method employed
herein with the standard method. The final values obtained by these two
methods are shown conveniently in parallel columns in Table XVI, page 274.
Turning back to that Table it will be noted that of the 15 pairs of popula-
tions compared 10 occupy the same relative position by both methods, while 5
would occupy different positions if they were to be arranged in regular order
by the standard method. A consideration of the latter cases throws con-
siderable light on the relative merits of the two methods.
The first value out of place is that of Hippocampus zosterae, that ob-
tained by the standard method being too low. The chief reason for the low
value is shown in Table XVII. The composite samples are quite small, which
results in a comparatively greater value for the probable errors. Although
the spread of the distribution, and hence sigma, is small, its effect is not
sufficient to counteract the influence of the small sample. On account of the
relatively large probable error combined with the comparatively small dif-
ference between the means, the final expression obtained is relatively low.
The effect of the high probable error as a result of a small sample is
also shown by the relatively low value obtained in comparing H. punctulatus ,
with H. hudsonius.,. Although in this case the sample of only one of the two
populations compared is quite small and the difference between the means
is higher, the effect of the small sample of the one population is sufficient to
result in a low value.
The influence of the size of the sample on the final figure used for com-
parison is shown further by two different comparisons of the populations of
H. hudsonius. The northernmost population of hudsonius0 diverges to a
greater extent than the populations from intermediate localities- from the
probable error of the mean = .6745
<T
V N
while, ° (standard deviation)
1938]
Ginsburg: Arithmetical Definition of Species
281
southernmost populations, punctulatuso (see Table VIII. p. 268). The two com-
parisons of these populations here made are as follows: (1) The intermediate
populations are grouped with the northern one into the subspecies hudsonius ,
and the composite sample is compared with the southernmost populations,
designated as the subspecies punctulatus2 (see Table XVII, p. 278). (2) The
northernmost population only of hudsonius2 is compared with the southern-
most ones, punctulatus2 (see footnote to Table XVII). The “significant”
figures obtained in these two comparisons are about the same, 6.8 and 6.9,
respectively. The greater divergence of the populations from near the ex-
tremes of the geographic range of the species as a whole, is not indicated
by these figures. The same two comparisons made by the method here em-
ployed are 30% and 27%, respectively, which does show the lesser inter-
gradation, or greater divergence, of the populations from the extremes of
the geographic range. The figures obtained by the use of the standard for-
mula fail to show this greater divergence, evidently because the number of
specimens in one of the composite samples used in the second comparison is
considerably less than that in the first comparison.
On the other hand, the value for the comparison between Cynoscion
regalis2 and C. arenarius2 is relatively high and for the same reason. In
this case the samples are comparatively large, resulting in a low probable
error, and this combined with the rather large difference between the means
gives a relatively high figure as the final result. The magnitude of the figure
obtained by the standard method in comparing a pair of related populations
is thus greatly influenced by the number of specimens in the samples; but
by far not to such an extent by the method employed.
The influence of the figure representing the probable error may also be
shown by the following hypothetical examples. Let us suppose that of popu-
lation A we study a sample to, which is sufficiently representative for all
practical biological purposes. Let us now take another sample n2 of the
same population containing say ten times the number of specimens in the
first sample. Biologically, as far as we may draw any conclusions from their
study, the two samples are very nearly alike; but mathematically the prob-
able error of the second sample will be smaller. Of course, this is perfectly
logical. It simply shows that from a mathematical point of view the reliabil-
ity of the sample increases as the number of specimens is increased. How-
ever, from a practical biological point of view let us see what will happen
when population A is compared with another population. Let us suppose
that we take a sufficiently representative sample m of a population B. If
now we compare sample to, with m we will get one figure ; while by comparing
to, with m we will get a different and, quite likely, a widely divergent figure.
Two separate comparisons of population A with B will thus yield two results.
Mi — M2
In general, as n increases the numerical value of ~ =====
V Ei2 + E2
will increase. Consequently, the standard formula in the form in which it
is generally employed is not suitable for the practical purpose of obtaining
a series of comparative figures. On the other hand, by the method employed
here the results will differ but little with the size of the samples, especially
when such composite samples give fairly approximate representations of
their populations.
Of even greater importance in comparing the two methods is the factor
M, — M2. It is obvious that pairs of populations the chief differentiating
characters of which are not the same or in which the values of M, — M„ differ
widely can not be fairly compared by any method employing this factor. A
striking effect of the influence the factor M, — M„ exerts on the final value
obtained by the standard method, is shown by the comparison between P.
lethostigma and P. albigutta. The difference between the means in this case
is comparatively high, resulting in a very high value for the final figure,
although the probable error is also high because of the wide spread of the
282
Zoologica: New York Zoological Society
[XXIII :13
distribution. In spite of the fact that this pair of populations shows a slight
amount of intergradation, the index of their divergence obtained by the
standard method is much greater than that of other pairs which do not in-
tergrade at all. This example shows in a striking manner that the standard
method is not well adapted for our purpose which requires the comparison
of pairs of populations that differ by widely unlike characters.
The influence of the factor M, — M0 is further shown by the comparison
of the fifth pair of populations falling out of line, namely, that of Gobiosoma
bosci and G. robustum. In this case, although the samples are small, never-
theless the probable error is relatively small, evidently because of the small
range, according to the unit adopted. In spite of the small probable error,
the final figure obtained is yet relatively small, evidently due to the small
difference between the means.
It is evident that the standard formula is not well adapted for our pur-
pose because the several factors involved do not result in fair comparative
figures. It is preferable from the point of view of theoretical mathematics
in determining probabilities; but for our purpose it is all too often not
practically workable. One great drawback is that in practice we usually do
not have samples of sufficient size that may be adequate according to mathe-
matical usage. But irrespective of the size of the samples, the formula is
not properly applicable because of the other factors involved, as discussed
above. An application of the standard formula evidently is often bound to
lead us astray. Although the method employed lacks the mathematical ele-
gance of the standard method, in reality it gives a truer picture and inter-
pretation of the facts of nature in connection with the problem under con-
sideration. The relative degree of divergence of the pairs of populations, or
their position in the series, with which this paper deals, is shown more
nearly in keeping with the facts of nature or more accurately by our present
method.
The method here employed also has the advantage of greater clarity.
In using the standard formula biometricians seem to be chiefly concerned
whether the resulting figure is “significant” or not; but once “significant”
figures are obtained no particular stress seems to be laid generally on the
relative value of figures of differing magnitude. For instance, let us say
that we compare three pairs of populations and obtain the three “significant”
figures of 5, 10 and 15, respectively. Now, definitely, just what is the mean-
ing of these relative figures outside their implication of differing degrees
of probability? Certainly, this is not clear to the average busy biologist. Of
course, 5 is greater than 10, and 10 is greater than 15; but the differences
between these figures do not convey any special meaning or idea in connec-
tion with our problem, since the biologist is not in the habit of thinking in
such terms. Indeed, the relative magnitude of such figures hardly seem to
have any definite meaning within rather wide limits even to the statisticians.
Furthermore, such figures are not susceptible of conveying a very precise
meaning with respect to divergence; because their value changes materially
as the size of the sample is changed, and with other factors. Even their
relative value in the series is likely to change with a change in these fac-
tors. On the other hand, by the method employed, if we compare three pairs
of populations and find that the intergrades are 5, 10, and 15 per cent, re-
spectively, of the total number of individuals in the composite samples, on
the average, such figures immediately convey a definite and clear meaning
to every biologist.
Nomenclature of Taxonomic Units Below Full Specific Rank.
It has been shown that the boundary lines which may be drawn between
the species and its subdivisions are arbitrary, and tentative limits have been
proposed where to draw such lines. The nomenclature of the species and
1938]
Ginsburg: Arithmetical Definition of Species
283
its subdivisions may now be considered with profit, and a simpler method
than the one now in use is here proposed.
According to present usage which is sanctioned by the international
code of zoological nomenclature, subspecies are designated in the form of
trinomials. The international code goes into the subject only as far as
subspecies are concerned, and no units of lower rank are considered; but on
account of the arbitrary nature of these categories there is no fundamental
reason for stopping there, and some authors now are using quadrinomials,
generally inserting the term “variety” between the third and fourth words
of the full name. However, there is likewise no reason for stopping even
there. One, conceivably, may even propose to use quinquinomials, sexi-
nomials, etc., since in a gradual series such as we are dealing with where the
boundary lines necessarily must be arbitrary, such boundaries may be
increased.
It is evident that carried to its logical conclusion, the nomenclature of
taxonomic units below specific rank would become clumsy and unwieldy, and
assume a form very similar to the pre-Linnaean polynomial nomenclature;
although it is true that the fundamental nature of the binomial system would
be retained. It is evident, therefore, that there is room for improvement in
our present method of naming taxonomic units below specific rank.
Another practical difficulty inherent in the present system is found in
the relation of zoological considerations to nomenclature. It should be re-
membered that the Linnaean or binomial system of nomenclature serves two
virtually independent purposes: (1) It supplies names to species and groups
of higher or lower rank, such names being for the practical use of handles
by which we may discuss these entities. (2) It also attempts to show rela-
tionship. In practice these two purposes cannot be made to work in entire
accord. For the first purpose stability is a primary consideration, and it
could be served most effectively by a set of arbitrary rules, such as the inter-
national code; but since the second purpose must also be considered, abso-
lute stability is impossible, and part of the full names of organisms must
change with inci'eased knowledge, with our changing ideas of relationship,
or with differences in the interpretation of relationship by individual biolo-
gists. Changes in nomenclature caused by zoological considerations are, or
may be, chiefly of two kinds.
First, changes in the first word or the generic part of the name of a
species are caused by the everlasting shifting about of species from one
genus to another, or the frequent changing of the boundaries of genera by
individual zoologists. Species are thus constantly shuffled with respect to
their generic affiliations. Nomenclatorial instability of this kind is, of course,
inherent in the system and is frequently unavoidable; although it would be
well for zoologists to exercise restraint in their treatment of many cases of
this kind by refraining from changing the constitution of genera on the
slightest provocation.
Second, in genera in which the populations are now in an early and
active state of diversification and ramification, questions may arise: (a) as
to whether a given population is to be properly assigned specific or subspe-
cific rank, and (b) if the latter, as to the proper species into which it is to be
grouped. In such cases, any interpretation given of the relationship between
closely related populations, on the basis of the available zoological evidence,
is not the only possible one. The chief difficulty in the way of a single, con-
sistent and acceptable interpretation in cases of this kind often lies in the
uncertainty as to which one of the several populations is the more primitive
one, and the consequent uncertainty as to the starting point of the argument.
This is primarily due to our present very imperfect knowledge of the me-
chanism and methods of descent. While the available evidence may show to
a high degree of probability that the several populations under consideration
284
Zoologica: New York Zoological Society
[XXIII :13
should be assigned to taxonomic units of varying degree, yet the interpreta-
tion of their relationship on which such assignment must depend is of a
speculative nature to a large extent. Examples of such genera are Hippo-
campus (see, for instance, page 549 of my review, 1937), and Bathygobius
(a study of which I have now prepared in manuscript form). In such cases
scrambling and unscrambling of the trinomial names of subspecies may well
be expected to result from different interpretation of the data by individual
zoologists.
In view of the possible resulting complexity of nomenclature as outlined
above, the impossibility of being reasonably certain of relationship, in some
cases, at the present time, and the necessarily arbitrary nature of limiting
taxonomic units, the following method of naming species and their sub-
divisions is here proposed. All names are to be binomial. To indicate that
a taxonomic unit is to be regarded as a full species the numeral 1 is inserted
after the specific name of the binomial ; a taxonomic unit of the next lower
rank is to have the numeral 2 inserted in the same place; the next lower,
the figure 3. Related populations of organisms of the rank of full species or
lower are thus divided by arbitrary lines into categories and the categories
designated by number, namely, a category of the first order or a full species,
a category of the second order or a subspecies, a category of the third order
or a race. In the case of a full species which is not subdivided the numeral
may be left out in writing its binomial designation. By way of illustration,
a taxonomic unit of the second category would be designated as :
A — us a — us2;
while a unit of the third category would bear the designation :
A — us b — usR.
The subspecies treated of in this paper are designated above according
to the proposed method.
The proposed method is in consonance with the international code,
except for the required use of trinomials for subspecies, which is not uni-
versally followed anyway. If anything it will tend to reenforce the system
of binomial nomenclature which will thus become binomial in fact as well
as in theory, and the necessity for using the clumsy trinomial and quadri-
nomial designations will be obviated. Another advantage is that this method
does not imply a commitment as to the relationship of the populations named.
It has already been pointed out that given a certain set of evidence rela-
tionship may be interpreted sometimes in more than one way. In such
cases, if the nomenclature be used in the form here proposed it will be in
accordance with the known facts and no more ; the probable taxonomic rank
of the particular unit may thus be indicated without the necessity of making
a commitment in regard to relationship about which one may be uncertain.
An apparent advantage of this method of nomenclature is its flexibil-
ity. However, its very flexibility may turn out to be a disadvantage if the
method is not used with caution. It may tend to put a premium on care-
less work. Without the necessity of using trinomials, quadrinomials, etc., it
may result in the excessive multiplication of new names based on unim-
portant or irrelevant differences exhibited by a few specimens after a hap-
hazard examination of such specimens. Nevertheless, the advantages of
this method are so evident that they outweigh this possible disadvantage,
and, in any case, there is no absolute insurance against careless or half-
baked work. The undesirable contingency indicated may be obviated by a
general agreement among systematists to refrain from formally naming
categories of a rank lower than a subspecies, or at least the next lower or
third category which would be equivalent to a quadrinomial as now used;
otherwise the resulting great increase in new names will tend to make
zoological nomenclature too burdensome.
1938]
Ginsburg: Arithmetical Definition of Species
285
Summary.
1. No absolute criterion exists by which to determine just what is a
species. All proposed criteria hold only in part or in special cases. Morph-
ology is the only practical and most generally applicable criterion, al-
though it has only a relative value. The present study is based entirely on
that criterion.
2. Intergradation in nature is universal. It is the rule rather than the
exception. This is true of species as well as of subspecies or taxonomic cate-
gories of lower rank. (To a certain extent this is also true of categories
higher than species, but this paper deals only with the species and its sub-
divisions.) Consequently the limitation of the species concept or that of
any of its subdivisions must depend on the degree of intergradation.
3. The intergradation between, or the divergence of, natural popula-
tions is such that if some method be adopted for measuring intergradation,
,or divergence, and a sufficient number of pairs of closely related popula-
tions be compared, the figures expressing the measures and indicating the
degree of intergradation, or divergence, will form a continuous series with-
out any breaks. Consequently, it follows that any limitation given to our
concept of species, subspecies, race, etc., necessarily must be arbitrary.
4. A definite measure which is determined by a simple arithmetical
calculation is employed in the present study for the purpose of expressing
intergradation, or divergence. This measure is based on the character show-
ing the greatest divergence between the two populations.
5. The measure employed has the advantage of simplicity, clarity and
ease of determination, and it gives a correct though approximate interpre-
tation of the data.
6. Using this measure as a basis, definitions of the terms species, sub-
species, and race are presented.
7. The necessity of proper sampling of the material from which the
data are obtained, that form the basis of the measure, is indicated and dis-
cussed. The method of sampling the material used in the present study is
stated.
8. A series of examples of pairs of related populations of fishes is
presented, showing the application of the measure, proving the gradualness
of intergradations, inter se, and showing that the necessarily arbitrary
boundaries drawn in the definition of the species and its subdivisions, are
more or less in accordance with the facts of speciation as they occur in
nature.
9. A comparison is made between our method of measuring divergence
and the standard method.
10. A method for the nomenclature of populations below specific rank
is proposed.
Bibliography.15
Davenport, Charles B. and J. W. Blankinship.
1898. A precise criterion of species. Science (n. s.) vol. 7, pp. 684-695.
Ginsburg, Isaac.
1929. Review of the weakfishes (Cynoscion) of the Atlantic and Gulf coasts
of the United States, with a description of a new species. Bull. U. S.
Bur. Fish., vol. 45, pp. 71-85.
1932. A revision of the genus Gobionellus (Family Gobiidae). Bull. Bing-
ham Oceanog. Coll., vol. 4, art. 2.
1933. A revision of the genus Gobiosoma (Family Gobiidae) with an ac-
count of the genus Garmannia, idem, art. 5.
286
Zoologica: New York Zoological Society
1937. Review of the seahorses (Hippocampus) found on the coasts of the
American continents and of Europe. Proc. TJ. S. Nat. Mus., vol. 83,
pp. 497-594.
1937a. The species and its subdivisions. Copeia, pp. 184-188.
Goode, G. Brown.
1879. A preliminary catalogue of the fishes of the St. John’s River and
the east coast of Florida, with descriptions of a new genus and three
new species. Proc. U. S. Nat. Mus., vol. 2, pp. 108-121.
Hildebrand, Samuel F. and Louella E. Cable.
1930. Development and life history of fourteen teleostean fishes at Beaufort,
N. C. Bull. U. S. Bur. Fish., vol. 46, pp. 383-488.
Hildebrand, Samuel F. and William C. Schroeder.
1928. Fishes of Chesapeake Bay, idem, vol. 43, pt. 1, 366 pp.
Jenkins, Oliver P. and Barton W. Evermann.
1888. Description of eighteen new species of fishes from the Gulf of Cali-
fornia. Proc. U. S. Nat. Mus., vol. 11, pp. 137-158.
Pearl, Raymond.
1930. Introduction to medical biometry and statistics. Philadelphia and
London. W. B. Saunders Company.
Robson, G. C.
1928. The species problem. London. Oliver and Boyd.
15 Includes only those publications to which reference is made in the text.
Beebe: Eastern Pacific Expeditions — Introduction
287
14.
Eastern Pacific Expeditions
of the New York Zoological Society, XIV.
Introduction, Itinerary, List of Stations, Nets and Dredges of the
Eastern Pacific Zaca Expedition, 1937- 1938.1
William Beebe
Department of Tropical Research
New York Zoological Society
(Text-figures 1 & 2).
[This is the fourteenth of a series of papers dealing with the collec-
tions made in the eastern Pacific by the expeditions of the New York
Zoological Society. The first thirteen of the series appeared under the
heading of “The Templeton Crocker Expedition.” As it is necessary to sim-
plify the titles of future reports, all publications dealing with the materials
gathered in this region will appear from now on under the title of the
“Eastern Pacific Expeditions of the New York Zoological Society.” Under
this name will appear the results of work of the Noma (1923), Arcturus
(1925), Antares (1933), Templeton Crocker (1936) and the Eastern Pacific
Zaca (1937-1938) Expeditions.]
Introduction.
The Twenty-sixth Expedition of the Department of Tropical Research
was made possible by Mr. Templeton Crocker’s generous offer of the use of
his yacht Zaca for the second time. Mr. Crocker accompanied the expedition
together with his friend Mr. Maurice Willows. The scientific personnel was
the same as on the first trip and consisted of four members of the Tropical
Research Department, Dr. William Beebe, Director; Mr. John Tee-Van,
General Associate; Miss Jocelyn Crane, Technical Associate; and Mr.
George Swanson, Artist. Toshio Asaeda was, as before, photographer and
preparateur, and together with Captain Alfred Pedersen and the entire
crew were most efficient and helpful in every way.
The Zaca is a Diesel schooner, 118 feet over all, with a gross tonnage
of 84. She is supplied with all of the usual apparatus for capturing fish and
animals, such as seines, nets, submerged lights, etc.; in addition Mr. Crocker
provided for these expeditions, a gasoline-engined winch and a 7,500-foot
length of 34 -inch-diameter steel cable. With this apparatus we were able
to trawl down to a maximum depth of 500 fathoms and to dredge on the
bottom with deep-sea dredges. Two valuable features of the vessel were
the bow pulpit and the boom-walk copied after those which I first used on
the Arcturus in 1925.
1 Contribution Number 560, Department of Tropical Research, New York Zoological Society.
For corresponding List of Stations of the First Zaca Trip, see Zoologica, Vol. XXII, No. 2.
288
[XXIII :14
Zoologica: Neiv York Zoological Society
Itinerary.
The route of the expedition is shown on the two accompanying maps.
We left San Diego on November 6, 1937, and disbanded at Balboa on April
5, 1938. The route is indicated by the successive numbers of the Stations.
The dates and time spent at the various localities are as follows: San
Diego, November 5; MEXICO: Ensenada, November 6; Guadelupe Island,
November 8; San Benito Islands, November 9 and 10; Cedros Island, Novem-
ber 10; Cape San Lucas, November 13; Penas, Banderas Bay, November 15
and 16; Chamela Bay, November 17 to 20; Tenacatita Bay, November 20;
Manzanillo, November 21 and 22; Sihuatanejo Bay, November 24; Acapulco,
November 25 to 29 ; Port Angeles, December 1 ; Port Guatulco, December
2 to 7 ; Santa Cruz Bay, December 7 and 8 ; Tangola-Tangola, December
8 to 13; EL SALVADOR: La Libertad, December 16 to 19; La Union, De-
cember 20 and 21; Meanguera Island, December 21 and 22; Farrallone
Islands, December 22; Conchaguita Island, December 22; La Union, Decem-
ber 22 and 23, and 26 and 27; NICARAGUA: Potosi River, December 23 to
25; Monypenny Point, December 25 and 26; Corinto, December 28 to January
7; San Juan del Sur, January 9 to 12; COSTA RICA: Port Parker, January
12 to 23;; Murcielago Bay, January 23; Potrero Grande Bay, January 23
and 24; Port Culebra, January 24 to 31; Cocos Bay, January 26, 29 and 31;
Braxilito Bay, January 31 to February 1; Piedra Blanca Bay, February 1
to 6; Puntarenas, February 11 and 12; Cedro Island, February 12 and 13;
Puntarenas, February 13 to 21; Cedro Island, February 21 and 22; Negritas
Island, February 22; Jasper Island, February 22 to 25; Ballenas Bay, Feb-
ruary 25 and 26 ; Puntarenas, February 28 to March 1 ; Uvita Bay, March 2
to 4; Golfito, Gulf of Dulce, March 5 to 9; Pavon Bay, March 9 and 10;
PANAMA: Puerto Armuelles, March 11; Parida Island, March 11 to 13;
Bahia Honda, March 13 to 19; Hermosa Bay, Coiba Island, March 19 and 20;
Hannibal Bank, March 20; Balboa, C.Z., March 22 to 24; COLOMBIA:
Gorgona Island, March 27 to April 2; PANAMA: Balboa, April 5.
List of Stations, Nets and Dredges.
The data presented in the following tables are self-explanatory. Ab-
bi'eviations are as follows: The Station numbers refer to the continuously
numbered series of oceanographic stations of the Department of Tropical
Research. In the second column the letter “D” refers to dredges, “T” to
tow-nets and “L” to submarine night lights. Hence, 182, D-l means the
first dredge drawn at Station 182.
In the third column the dredges are separated into two- and four-foot
ones, the measurement being across the mouth. Both sizes were of the Blake
type, the bags being of two-inch and one-inch mesh respectively. The small
dredges were drawn from launch or dory, the large ones being handled from
the deck of the yacht, using the quarter-inch steel cable. The tow-nets of
one-half and one metre diameter were of standard Michael Sars’ type, with
2XX silk bolting cloth posteriorly and OXX bolting cloth anteriorly, the
mouth of the net having a collar of shrimp netting. The lights indicated
refer to night stations, where submerged electric lights were employed,
and the captures made by the use of long-handled dip-nets.
1938]
Beebe: Eastern Pacific Expeditions — Introduction
289
Sta-
Net
Type of
Depth
Position
General
Date
Bottom
Duration
No.
No.
Net
Fath-
oms
Metres
N. Lat.
W. Long.
Locality
1937
Hr s.
Mins.
Nov.
126
D-13
4' dredge
45
82
28° 12'
115° 07'
E. of Cedros Is., Mex.
10
Mud, algae.
—
10
D-14
4' dredge
45
82
28° 10'
115° 08'
“
10
—
10
D-15
4' dredge
42
76
28° 09'
115° 08'
“
10
Mud.
—
10
D-16
4' dredge
42
76
28° 07'
115° 08'
10
—
20
D-17
4' dredge
40
73
28° 05'
115° 09'
10
u
—
10
D-18
4' dredge
20
36
28° 01'
115° 10'
SE. of Cedros Is., Mex.
10
Rocks, algae.
—
10
D-19
4' dredge
25
45
2 7° 59'
115° 09'
“
10
“
—
10
135
D-27
2' dredge
2-6
3.6-11
22° 53'
109° 54'
San Lucas Bay, Mex.
13
Sand.
—
20
150
D-27
4' dredge
60
108
23° 01'
109° 28'
Gorda Banks, Mex.
13
—
20
176
L-l
Light
—
—
31° 25'
116° 58'
23 m. S.xW. of Banda
Point, Mex.
7
—
i
10
177
L-l
Light
—
—
28° 42'
117° 50'
30 m. E.xS. of Guad-
alupe Is., Mex.
8
—
—
30
178
L-l
Light
—
—
28° 18'
115° 34'
San Benito Is., Mex.
9
—
2
00
179
T-l
Yl Metre
0
0
24° 02'
111° 38'
18 m. SSE. of Pt.
Tosca, Mex.
12
—
—
10
180
T-l
y Metre
0
0
23° 31'
110° 54'
65 m. N W . of Cape
Falso, Mex.
12
—
10
181
L-l
Light
—
—
20° 36' 30"
105° 16'
Banderas Bay, Mex.
15
—
2
0
182
D-l
4' dredge
8
14.5
19° 34' 30"
105° 08' 15"
Chamela Bay, Mex.
17
Sand, algae.
—
10
D-2
4' dredge
12
22
19° 34'
105° 07' 30"
“
17
—
10
D-3
4' dredge
15
27
19° 33' 45"
105° 07' 45"
“
17
“
—
10
D-4
4' dredge
16
29
19° 33' 30"
105° 08' 15"
17
Sand, algae,
cr. shell.
—
10
L-l
Light
—
—
19° 34' 45"
105° 08' 30"
17
—
2
0
L-2
Light
—
—
19° 34' 45"
105° 08' 30"
18
—
2
0
183
D-l
4' dredge
15
27
19° 17'
104° 51'
TenacatitaBay, Mex.
21
Sand.
—
10
D-2
4' dredge
30
54
19° 15' 30"
104° 51'
21
Muddy sand.
—
15
D-3
4' dredge
40
73
19° 14' 30"
104° 51' 30"
a
21
Sandy mud.
—
15
D-4
4' dredge
40
73
19° 14' 30"
104° 52' 30"
“
21
Mud.
—
15
L-l
Light
—
—
19° 18'
104° 51'
“
20
—
2
30
184
D-l
4' dredge
25
45
19° 03' 45"
104° 20' 45"
Manzanillo, Mex.
22
Sand.
—
10
D-2
4' dredge
30
55
19° 04'
104° 22'
“
22
Grav’ly sand.
—
20
L-l
Light
—
—
19° 03' 30"
104° 19' 45"
21
—
3
0
185
T-l
Metre
300
545
17° 45'
103° 05'
43 m. SE. of Pt.
Telmo, Mex.
23
—
2
0
T-2
Metre
400
730
17° 45'
103° 05'
23
—
2
0
T-3
Metre
500
910
17° 45'
103° 05'
23
—
2
0
T-4
J4 Metre
0
0
17° 45'
103° 05'
23
—
0
30
186
L-l
Light
—
—
17° 38'
102° 00'
20 m. W. of Sihua-
tanejo, Mex.
23
—
2
30
187
L-l
Light
—
—
17° 17'
101° 16'
72 m. WNW. of Aca-
pulco, Mex.
24
—
0
30
188
L-l
Light
—
—
16° 15' 30"
99° 55' 30"
Acapulco, Mex.
25
—
2
0
L-2
Light
—
—
16° 15' 30"
99° 55' 30"
26
—
2
0
L-3
Light
—
—
16° 15' 30"
99° 55' 30"
27
—
2
0
L-4
Light
—
—
16° 15' 30"
99° 55' 30"
“
28
—
2
0
189
D-l
4' dredge
20
36
16° 40'
99° 43' 30"
17 m. SE.xE. of
Acapulco, Mex.
29
Sandy mud,
algae.
—
10
D-2
4' dredge
20
36
16° 39' 30"
99° 42'
*
29
—
20
D-3
4' dredge
13
23.5
16° 39'
99° 41'
“
29
Mud.
—
10
D-4
4' dredge
28
51
16° 38' 30"
99° 40'
a
29
—
14
190
L-l
Light
—
—
16° 30'
99° 13'
45 m. ESE. of
Acapulco, Mex.
29
—
1
0
191
D-l
2' dredge
8
14.5
16° 29'
98° 45'
Mouth of Dulce Riv.,
Mex.
30
—
0
10
192
D-l
4' dredge
26
47
16° 16' 30"
98° 37'
4 m. SSW. of Maldan-
ado Pt., Mex.
30
Mud.
—
10
290
Zoologica: New York Zoological Society
[XXIII :14
GUADELUPE IS.^>
177®
SAN BENITO ISH78@
CEDROS IS."l26^b
~o.
SAN DIEGO
MAGDALENA BAY
179©
180 ©
SAN LUCAS B.-ISS^^baKk
GORDA BANKS-150
.BANDERAS B.-I8I^
, ,o (tbes mar/etas /s.)
CHAMELA B.- 182
TENACATITA B." 183
MAN2ANIL LO-184
S I H U AT A NEJO
185®
186
B. -
187
ACAPULCO- 188
189
190®
DULCE RIVER-I9I"
192
193
PORT ANGELES - 194-
P O R T GUATULCOl
SANTA CRUZ B. / 9
TANGOLA-TANGOLA B. 196
ZACA EXPEDITION
1937-1938
NEW YORK
ZOOLOGICAL SOCIETY
1938]
Beebe: Eastern Pacific Expeditions — Introduction
291
LA LIBERTAD -198
GULF OF FONSECA-199
(CONCHAGUITA IS., LA UNION,
UCANGUERA IS, NONFPENNF PT..
POTOSI P., FARALLONE IS.)
CORI NTO-200
©
201
PORT PARKER
-203
MURCIELAGO BAY-204
POTRERO GRANDE B/205
PORT C U LE BRA— 206
(COCOS BJ
BR AX I LITO BAY-207
PIEDRA BLANCA B.-208
209©
21
210©
GULF OF NICOYA-213
(ALCATRAZ IS., BALLENAS B.,
CEDRO IS., JASPER IS.,
NEGRITOS IS., PUNTARENAs))
UVITA BAY
2170
GULF OF DULCE-218
(GOLr/rO, PAVON b)
PARI DA
CO I BA IS. “223
HANNIBAL BANK-224
225©
GORGONA IS.
232 "
292
Zoologica: New York Zoological Society
[XXIII :14
Sta-
tion
Net
Type of
Depth
Position
General
Date
Bottom
Duration
No.
No.
Net
Fath-
oms
Metres
N. Lat.
W. Long.
Locality
1937
Hrs.
Mins.
192
D-2
4' dredge
33
60
16° 16'
98° 36'
4 m. SSW. of Maldan-
Nov.
30
Sand.
10
D-3
4' dredge
38
69
16° 15'
98° 35' 30"
ado Pt., Mex.
a
30
Mud.
10
193
L-l
Light
—
—
16° 05'
98° 17'
21 m. SE. of Maldan-
30
—
1
10
194
L-l
Light
15° 39' 15"
96° 30' 40"
ado Pt., Mex.
Port Angeles, Mex.
Dec.
1
3
0
195
D-l
2' dredge
2.5
4.5
15° 44' 56"
96° 08' 03"
Port Guatulco, Mex.
4
Sand, algae.
—
3
D-2
2' dredge
3
5.5
15° 44' 51"
96° 07' 51"
4
Sand.
—
4
D-3
2' dredge
3.5
6.3
15° 44' 45"
96° 07' 53"
*
4
Sand, cr. shell
—
3
D-4
2' dredge
4.5
8.2
15° 44' 40"
96° 07' 53"
a
4
Sand, algae,
—
3
D-5
2' dredge
2
3.6
15° 44' 50"
96° 08' 09"
a
5
cr. shell.
Sand, algae.
3
D-6
2' dredge
3
5.4
15° 44' 45"
96° 08' 05"
U
5
Sand, algae,
—
3
D-7
2' dredge
4.5
8.2
15° 44' 35"
96° 08'
«
5
cr. shell.
Rocks.
_
3
D-8
2' dredge
6
11
15° 44' 30"
96° 07' 56"
“
5
Sand, algae,
—
3
D-9
2' dredge
7
12.6
15° 44' 28"
96° 07' 51"
a
5
cr. shell.
Gr. sand, cr.
_
5
D-10
2' dredge
4
7.3
15° 44' 53"
96° 08' 03"
«
6
shell.
Gr. sand, cr.
_
3
D-ll
2' dredge
5
9.1
15° 44' 49"
96° 07' 59"
a
6
shell, dead
coral.
Gr. sand, cr.
4
D-12
2' dredge
6
11
15° 44' 42"
96° 07' 55"
a
6
shell.
Sand.
_
3
D-13
2' dredge
8
14.6
15° 44' 33"
96° 07' 49"
a
6
«
—
3
D-14
2' dredge
4
7.3
15° 44' 27"
96° 07' 57"
a
6
CoraL
—
5
D-15
Diving
1.5
2.7
15° 44' 54"
96° 07' 57"
w
6
1
0
D-16
2' dredge
10
18.2
15° 44' 35"
96° 07' 45"
a
7
Sand.
—
3
D-17
2' dredge
6
11
15° 44' 39"
96° 07' 42"
“
7
—
5
D-18
2' dredge
6
11
15° 44' 39"
96° 07' 49"
a
7
—
3
D-19
4' dredge
17
31
15° 44' 22"
96° 07' 27"
a
7
Gr. mud, cr.
—
5
D-20
4' dredge
23
42
15° 44' 22"
96° 07' 08"
a
7
shell.
Mud.
_
10
D-21
4' dredge
18
33
15° 44' 45"
96° 06' 55"
Santa CruzBay, Mex.
7
“
—
10
L-l
Light
—
15° 44' 45"
96° 08'
Port Guatulco, Mex.
3
—
1
30
L-2
Light
—
—
15° 44' 45"
96° 08'
4
—
2
0
L-3
Light
—
—
15° 44' 45"
96° 08'
5
—
2
30
196
D-l
2' dredge
5
9
15° 45' 58"
96° 06'
Tangola-Tangola
9
Gr. sand.
0
3
D-2
2' dredge
5.5
10
15° 45' 53"
96° 05' 57"
Bay, Mex.
9
Sand.
0
3
D-3
2' dredge
7
12.8
15° 45' 48"
96° 05' 54"
“
9
“
0
3
D-4
2' dredge
8
14.5
15° 45' 44"
96° 05' 52"
a
9
0
3
D-5
2' dredge
9
16.3
15° 45' 40"
96° 05' 50"
“
9
“
0
5
D-6
2' dredge
7
12.8
15° 45' 34"
96° 06' 02"
9
Sand, cr.
0
5
D-7
2' dredge
6
11
15° 45' 34"
96° 06' 03"
a
9
shell.
Sand.
0
4
D-8
2' dredge
9
16.3
15° 45' 37"
96° 05' 54"
a
9
0
5
D-9
2' dredge
7.5
13.6
15° 45' 45"
96° 05' 59"
a
12
a
0
3
D-10
2' dredge
5
9.1
15° 45' 51"
96° 06' 01"
12
“
0
3
D-ll
2' dredge
4.5
8.2
15° 45' 57"
96° 06' 02"
12
u
0
3
D-12
2' dredge
4
7.3
15° 46' 02"
96° 05' 58"
a
12
a
0
5
D-13
2' dredge
10
18
15° 45' 32"
96° 05' 52"
“
12
Gr. sand, cr.
0
3
D-14
2' dredge
5
9.1
15° 45' 34"
96° 06' 03"
a
12
shell.
Cr. shell.
0
5
D-15
2' dredge
5
9.1
15° 45' 34"
96° 06' 03"
12
Cr. shell.
0
5
1938]
Beebe: Eastern Pacific Expeditions — Introduction
293
Sta-
tion
Net
Type of
Depth
Position
General
Date
Bottom
Duration
No.
No.
Net
Fath-
oms
Metres
N. Lat.
W. Long.
Locality
1937
Hrs.
Mina.
Dec.
196
D-16
4' dredge
16
29
15° 45' 22"
96° 05' 51"
Tangola-Tangola
Bay, Mex.
13
Mud.
0
5
D-17
4' dredge
23
42
15° 45'
96° 05' 34"
“
13
0
3
D-18
4' dredge
30
55
15° 44' 58"
96° 05' 13"
U
13
U
0
3
D-19
4' dredge
30
55
15° 44'
96° 05'
13
“
0
3
D-20
4' dredge
50
91
15° 43' 15"
95° 04' 15"
13
0
4
L-l
Light
—
—
15° 45' 40"
96° 06' 05"
“
8
—
1
0
L-2
Light
—
—
15° 45' 40"
96° 06' 05"
“
9
—
1
0
L-3
Light
—
—
15° 45' 40"
96° 06' 05"
10
—
3
0
L-4
Light
—
—
15° 45' 40"
96° 06' 05"
11
2
30
L-5
Light
—
—
15° 45' 40"
96° 06' 05"
12
—
3
0
197
D-l
4' dredge
14
25
14° 16'
92° 03'
7 m. W . of Champer-
ico, Guatemala.
15
Mud.
0
5
D-2
4' dredge
14
25
14° 13'
92° 02'
44
15
0
5
198
D-l
4' dredge
13
24
13° 27' 20"
89° 19' 20"
La Libertad, El
Salvador.
16
0
5
D-2
4' dredge
14
25
13° 25' 50"
89° 19' 20"
«
16
a
0
5
L-l
Light
—
—
13° 28' 11"
89° 19' 14"
°
16
—
1
30
L-2
Light
—
—
13° 28' 11"
89° 19' 14"
a
18
—
2
00
199
D-l
4' dredge
16
29
13° 08'
87° 43'
Meanguera Is., Gulf
of Fonseca, El Sal.
23
Sand, mud,
cr. shell.
0
3
D-2
4' dredge
5
9.1
13° 02' 30"
87° 29' 30"
Monypenny Pt., Gulf
of Fonseca, Nicar.
24
Mud.
0
1
D-3
4' dredge
6
11
13° 03'
87° 30'
24
0
2
D-4
4' dredge
7
12.8
13° 03' 30"
87° 30' 20"
“
24
0
1
D-6
4' dredge
7
12.8
13° 03' 30"
87° 30' 20"
“
24
0
1
D-6
4' dredge
4
7.2
13° 02' 30"
87° 29' 30"
“
24
0
1
D-7
4' dredge
6
11
13° 20' 07"
87° 49'
La Union, Gulf of
Fonseca, El Salv.
27
Mud, man-
grove leaves
0
1
D-8
4' dredge
6
11
13° 19' 53"
87° 48' 43"
“
27
a
0
2
D-9
4' dredge
5
9.1
13° 19' 45"
87° 48' 32"
a
27
a
0
2
D-10
4' dredge
5
9.1
13° 19' 35"
87° 48' 19"
“
27
Mud.
0
3
D-ll
4' dredge
5
9.1
13° 19' 24"
87° 48' 00"
u
27
0
IK
D-12
4' dredge
5
9.1
13° 19' 08"
87° 47' 30"
“
27
0
3
D-13
4' dredge
6
11
13° 18' 50"
87° 47' 07"
27
U
0
IK
D-14
4' dredge
5
9.1
13° 19' 40"
87° 48' 23"
u
27
“
0
IK
D-15
4' dredge
6
11
13° 19' 52"
87° 48' 43"
“
27
U
0
IK
D-16
4' dredge
6
11
13° 20' 03"
87° 48' 57"
“
27
*
0
ik
D-17
2' dredge
4
7.2
13° 19' 54"
87° 49' 07"
44
27
Mud, man-
grove leaves.
0
IK
D-18
2' dredge
4
7.2
13° 19' 50"
87° 49'
*
27
a
0
IK
D-19
2' dredge
3
5.4
13° 20' 03"
87° 49' 26"
“
27
*
0
1A
D-20
2' dredge
3
5.4
13° 20' 03"
87° 49' 26"
27
“
0
IK
D-21
2' dredge
3
5.4
13° 20' 03"
87° 49' 26"
“
27
“
0
IK
D-22
2' dredge
3
5.4
13° 20' 03"
87° 49' 26"
“
27
0
IK
D-23
2' dredge
3
5.4
13° 20' 03"
87° 49' 26"
27
0
IK
D-24
2' dredge
6
11
13° 20' 03"
87° 49' 26"
*
27
0
IK
D-25
2' dredge
6
11
13° 20' 03"
87° 49' 26"
27
“
0
2
L-l
Light
—
—
13° 19' 59"
87° 49' 15"
20
—
3
0
L-2
Light
—
—
13° 11'
87° 41' 15"
Meanguera Is., Gulf
of Fonseca, El Salv.
21
—
2
0
L-3
Light
—
—
13° 19' 59"
87° 49' 15"
La Union, Gulf of
Fonseca, El Salv.
22
—
2
30
L-4
Light
—
—
CO
o
©
05
O
CO
Monypenny Pt., Gulf
of Fonseca, Nicar.
23
—
2
0
L-5
Light
—
—
13° 02'
87° 29' 30"
*
24
—
2
40
200
D-l
2' dredge
6.5
11.8
121" 28' 25"
87° 10' 59"
Corinto, Nicaragua
29
Mangrove
leaves.
0
IK
294
Zoologica: New York Zoological Society
[XXIII :14
Sta-
tion
No.
Net
No.
Type of
Net
Depth
Position
General
Locality
Date
1937
Bottom
Duration
Fath-
oms
Metres
N. Lat.
W. Long.
Hrs.
Mins.
Dec.
200
D-2
2' dredge
5.3
9.7
12° 28' 27"
87° 10' 55"
Corinto, Nicaragua
29
Mangrove
0
VA
leaves.
D-3
2' dredge
2
3.6
12° 28' 28"
87° 10' 51"
a
29
“
0
3
D-4
2' dredge
Vi
.9
12° 28' 32"
87° 10' 42"
u
29
0
VA
D-5
2' dredge
2
3.6
12° 28' 36"
87° 10' 48"
u
29
“
0
1H
D-6
2' dredge
2.5
4.6
12° 28' 41"
87° 10' 42"
a
29
“
0
2
D-7
2' dredge
2
3.6
12° 28' 45"
87° 10' 34"
“
29
“
0
2
D-8
2' dredge
6.6
12
12° 28' 24"
87° 10' 58"
M
29
“
0
2
D-9
2' dredge
6
11
12° 28' 27"
87° 11' 04"
u
29
“
0
2
1938
Jan.
D-10
2' dredge
7
12.8
12° 27' 46"
87° 11' 32"
“
5
0
VA
D-ll
2' dredge
8
14 6
12° 27' 46"
87° 11' 32"
M
5
Sand.
0
m
D-12
2' dredge
3
5.4
12° 27' 20"
87° 11' 38"
a
5
Mangrove
0
VA
leaves.
D-13
2' dredge
3
5 4
12° 27' 14"
87° 11' 45"
w
5
a
0
3
D-14
2' dredge
3
5.4
12° 27' 30"
87° 12' 06"
“
5
a
0
3
D-15
2' dredge
1
1.8
12° 27' 32"
87° 12' 09"
“
5
“
0
5
D-16
2' dredge
4-7
7-13
12° 27' 41"
87° 12' 08"
w
5
“
0
5
D-17
2' dredge
7-10
13-18
12° 27' 46"
87° 12' 17"
a
5
Sand.
0
7
D-18
2' dredge
5
9.1
12° 27' 38"
87° 12' 41"
“
5
Mangrove
0
5
leaves.
D-19
2' dredge
12-13
22-24
12° 28' 03"
87° 12' 39"
5
0
4
D-20
2' dredge
1.5
2.7
12° 27' 19"
87° 11' 39"
“
7
0
3
D-21
2' dredge
2
3.6
12° 27' 14"
87° 11' 38"
M
7
“
0
3
D-22
2' dredge
1.5
2.7
12° 27' 07"
87° 11' 37"
a
7
0
5
D-23
2' dredge
3
5.4
12° 27' 20"
87° 11' 35"
a
7
“
0
5
D-24
2' dredge
6.5
11.8
12° 27' 15"
87° 11' 55"
a
7
0
5
D-25
2' dredge
6.5-2.5
11. 8-1.5
12° 27' 10"
87° 11' 57"
“
7
“
0
10
D-26
2' dredge
2.5
4.5
12° 27' 24"
87° 11' 15"
a
7
“
0
3
D-27
2' dredge
3
5.4
12° 28' 38"
87° 10' 42"
“
7
0
5
D-28
2' dredge
3
5.4
12° 28' 31"
87° 10' 45"
a
7
“
0
5
D-29
2' dredge
3
5.4
12° 28' 16"
87° 10' 49"
M
7
“
0
5
D-30
2' dredge
3
5.4
12° 28' 27"
87° 10' 47"
a
7
0
6
1937
Dec.
L-l
Light
—
—
12° 28' 22"
87° 11' 06"
“
28
—
2
15
L-2
Light
—
—
12° 28' 22"
87° 11' 05"
“
29
—
1
30
L-3
Light
—
—
12° 28' 22"
87° 11' 05"
*
30
—
1
0
1938
Jan.
201
L-l
Light
—
—
12° 09'
87° 24'
22 m. SW. of Corinto,
7
—
1
30
Nicaragua.
202
L-l
Light
—
—
11° 15' 12"
85° 52' 58"
S. Juan del Sur, Nicar.
9
—
1
30
203
D-l
4' dredge
15
27
10° 56' 05"
85° 49' 25"
Port Parker, Costa
20
Sandy mud,
0
4
Rica.
cr. shell.
D-2
4' dredge
10
18.2
10° 55' 38"
85° 49' 21"
20
Shelly sand,
0
5
algae.
D-3
4' dredge
12
22
10° 55' 45"
85° 49' 05"
“
20
Shelly mud.
0
2
D-4
2' dredge
7
12.8
10° 55' 06"
85° 48' 53"
22
Gravel, algae.
0
3
D-5
2' dredge
3
5.4
10° 55' 10"
85° 49' 10"
W
22
Shells, dead
0
2
coral.
D-6
2' dredge
1
1.8
10° 55' 20"
85° 49' 18"
22
Rocks, gravel.
0
3
D-7
2' dredge
9-5
16. 4-9.1
10° 55' 43"
85° 49' 37"
22
Shells, algae.
0
5
D-8
2' dredge
9
16.4
10° 55' 43"
85° 49' 46"
“
22
“
0
5
D-9
2' dredge
1.5-4
2. 7-7.2
10° 55' 51"
85° 49' 52"
22
Coral.
0
5
D-10
2' dredge
6-2.5
10.9-4.5
10° 56' 15"
85° 48' 47"
22
Rocks.
0
3
1938]
Beebe: Eastern Pacific Expeditions — Introduction
295
Sta-
tion
Net
Type of
Depth
Position
General
Date
Bottom
Duration
No.
No.
Net
Fath-
oms
Metres
N. Lat.
W. Long.
Locality
1938
Hrs.
Mins,
203
D-ll
2' dredge
2-4
3. 6-7.2
10° 56' 07"
85° 48' 52"
Port Parker, Costa
Jan.
22
Rocks.
0
3
D-12
2' dredge
2
3.6
10° 55' 43"
85° 48' 44"
Rica.
22
Gravel, algae,
0
2
D-13
2' dredge
7-9
12.8-16.4
10° 55' 33"
85° 48' 30"
22
cr. shell.
Shells, algae.
0
3
D-14
2' dredge
4
7.3
10° 55' 29"
85° 48' 24"
u
22
“
0
5
D-15
2' dredge
9-2
16.4^3.6
10° 55' 03"
85° 48' 25"
“
22
Mangrove
0
3
L-l
Light
10° 55' 22"
85° 48' 28"
„
12
leaves, cr.
shell, algae.
3
0
L-2
Li^ght
—
—
10° 55' 22"
85° 48' 28"
“
14
—
1
30
L-3
Light
—
—
10° 55' 51"
85° 48' 58"
16
—
1
30
L-4
Light
—
—
10° 55' 51"
85° 48' 58"
“
17
—
1
45
204
D-l
2' dredge
3
5.4
10° 52' 08"
85° 52' 30"
Murcielago Bay,
23
Sand.
0
3
D-2
2' dredge
4
7.3
10° 52' 05"
85° 52' 27"
Costa Rica.
23
«
0
3
D-3
2' dredge
2.5
4.5
10° 52' 01"
85° 52' 23"
23
0
5
D-4
2' dredge
2
3.6
10° 51' 55"
85° 52' 19"
“
23
0
5
205
L-l
Light
—
—
10° 50' 45"
85° 45' 30"
Potrero Grande Bay,
23
—
2
30
206
D-l
4' dredge
14
25.5
10° 37' 03"
85° 41' 12"
Costa Rica.
Port Culebra, Costa
30
Sandy mud.
0
3
D-2
4' dredge
14
25.5
10° 36' 47"
85° 41' 10"
Rica.
30
u
0
4
D-3
4' dredge
14
25.5
10° 36' 22"
85° 41' 08"
“
30
“
0
5
L-l
Light
—
—
10° 37' 10"
85° 41' 18"
25
—
2
0
L-2
Light
—
—
10° 37' 50"
85° 39'
a
28
—
1
45
L-3
Light
—
—
10° 35' 45"
85° 40' 15"
a
29
—
1
45
L-4
Light
—
—
10° 35' 45"
85° 40' 15"
30
—
2
15
207
L-l
Light
—
—
10° 25' 20"
85° 48' 45"
Braxilito Bay, Costa
31
—
2
40
208
D-l
2' dredge
6
11
9° 51' 57"
85° 29' 55"
Rica.
Piedra Blanca Bay,
Feb.
5
Rocks, sand,
0
3
D-2
2' dredge
5
9.1
9° 51' 54"
85° 29' 53"
Costa Rica.
5
algae.
0
3
D-3
2' dredge
4
7.3
9° 51' 52"
85° 29' 46"
5
*
0
3
D-4
2' dredge
6
10.9
9° 51' 49"
85° 29' 41"
u
5
0
3
D-5
2' dredge
5
9.1
9° 51' 47"
85° 29' 34"
“
5
0
3
D-6
2' dredge
3
5.4
9° 51' 48"
85° 29' 32"
5
«
0
5
D-7
2' dredge
3
5.4
9° 52' 06"
85° 29' 49"
«
5
0
3
D-8
2' dredge
3
5.4
9° 52' 06"
85° 30'
5
«
0
5
D-9
2' dredge
3
5.4
9° 52' 05"
85° 30' 09"
5
“
0
5
D-10
2' dredge
2
3.6
9° 51' 51"
85° 30' 20"
«
5
“
0
4
L-l
Light.
—
—
9° 51' 47"
85° 29' 56"
“
1
—
2
40
L-2
Light
—
—
9° 51' 47"
85° 29' 56"
“
2
—
2
0
L-3
Light
—
—
9° 51' 47"
85° 29' 56"
“
3
—
2
0
209
L-l
Light
—
—
9° 45'
85° 26'
22 m. WNW. of Cape
6
—
0
45
T-l
Yi Metre
0
0
9° 41'
85° 24'
Blanco, Costa Rica
6
0
15
210
L-l
Light
—
—
9° 15'
85° 09'
20 m. S. of Gape
7
—
2
0
L-2
Light
9° 14'
85° 07'
Blanco, Costa Rica.
8
2
30
L-3
Light
—
—
9° 13'
85° 09'
u
26
—
3
0
T-l
Metre
300
540
9° 12'
85° 05'
7
—
4
0
T-2
Metre
400
730
9° 12'
85° 05'
«
7
—
4
0
T-3
Metre
500
910
9° 12'
85° 05'
7
—
4
0
T-4
Yi Metre
2
3.6
9° 12'
85° 05'
7
—
0
30
T-5
Yi Metre
2
3.6
9° 12'
85° 05'
7
—
0
30
296
Zoologica: New York Zoological Society
[XXIII :14
Sta-
tion
Net
Type of
Depth
Position
General
Date
Bottom
Duration
No.
No.
Net
Fath-
oms
Metres
N. Lat.
W. Long.
Locality
1938
Hrs.
Mins.
Feb.
210
T-6
Metre
(3 nets)
500
910
9° 12'
85° 05'
20 m. S. of Cape
Blanco, Costa Rica.
7
—
2
53
T-7
Metre
(3 nets)
500
910
9° 09'
85° 04'
“
8
—
4
25
T-8
Metre
(3 nets)
500
910
9° 12'
85° ic
27
—
3
26
T-9
Metre
200
360
9° 11'
85° 08' 30"
“
27
—
3
32
T-10
Metre
(2 nets)
500
910
9° 11'
85° 08' 30"
27
—
3
32
T-ll
Yi Metre
3
5.4
9° 11'
85° 08' 30"
“
27
—
1
0
211
L-l
Light
—
—
9° 29'
85° 08'
6 m. SW. of Cape
Blanco, Costa Rica.
10
—
0
20
212
L-l
Light
—
—
9° 58' 15"
84° 59' 45"
Puntarenas, Costa
Rica
11
—
8
0
213
D-l
2' dredge
8
14.6
9° 50' 52"
84° 53' 07"
Cedro Island, Gulf of
Nicoya, Costa Rica
13
Mud.
0
3
D-2
2' dredge
4
7.2
9° 50' 46"
84° 53' 05"
“
13
“
0
3
D-3
2' dredge
4
7.2
9° 50' 41"
84° 53' 03"
“
13
a
0
1
D-4
2' dredge
5
9.1
9° 50' 39"
84° 53' 07"
“
13
a
0
2
D-5
2' dredge
4
7.2
9° 50' 37"
84° 53' 12"
“
13
0
2
D-6
2' dredge
4
7.2
9° 50' 38"
84° 53' 14"
13
Mud, sand,
cr. shell
0
2
D-7
2' dredge
4
7.2'
9° 50' 42"
84° 53' 17"
“
13
“
0
2
D-8
2' dredge
4
7.2
9° 50' 45"
84° 53' 19"
13
“
0
3
D-9
2' dredge
6
11
9° 50' 52"
84° 53' 23"
“
13
“
0
4
D-10
2' dredge
10
18.2
9° 51' 01"
84° 53' 22"
*
13
Mud.
0
3
D-ll
4' dredge
35
63.7
9° 44' 52"
84° 51' 25"
Off BallenasBay, Gulf
of Nicoya, Costa Rica
25
0
3
D-12
4' dredge
35
63.7
9° 44' 24"
84° 51' 03"
“
25
“
0
5
D-13
4' dredge
35
63.7
9° 43' 45"
84° 51' 08"
25
“
0
5
D-14
4' dredge
35
63.7
9° 43'
84° 51' 17"
“
25
“
0
5
D-15
4' dredge
40
73
9° 42' 10"
84° 51' 25"
“
25
*
0
5
D-16
4' dredge
45
82
9° 41' 10"
84° 51' 45"
25
“
0
5
D-17
4' dredge
35
63.7
9° 42'
84° 56'
25
U
0
5
D-18
4' dredge
80
145
9° 29' 30"
85° 06' 40"
3 m. off Cape Blanco,
Costa Rica.
26
0
10
L-l
Light
—
—
9° 51' 30"
84° 52' 30"
Cedro Island, Gulf of
Nicoya, Costa Rica
21
—
2
30
L-2
Light
—
—
9° 47'
84° 53' 30"
Alcatraz Is., Gulf of
Nicoya, Costa Rica
22
—
2
0
L-3
Light
—
—
9° 46' 20"
84° 53' 05"
Jasper Is., Gulf of
Nicoya, Costa Rica
23
—
2
30
L-4
Light
9° 44' 25"
84° 59' 35"
BallenasBay, Gulf of
Nicoya, Costa Rica
25
Mar.
3
00
214
D-l
4' dredge
42
76.5
9° 19' 32"
84° 29' 30"
14 m. S.xE. of Judas
Pt., Costa Rica
i
Mud, shell.
0
5
D-2
4' dredge
43
78
9° 18' 45"
84° 29' 15"
i
Mud.
0
10
D-3
4' dredge
50
91
9° 18' 05"
84° 28' 25"
a
i
“
0
10
D-4
4' dredge
61
112
9° 17' 40"
84° 27' 30"
a
i
Mud, rocks
0
10
215
L-l
Light
—
—
9° 03'
84° 06'
23 m. W. of Uvita
Bay, Costa Rica.
i
—
1
30
216
L-l
Light
—
—
9° 08' 45"
83° 45' 25"
Uvita Bay, Costa
Rica
2
—
4
0
L-2
Light
—
—
9° 08' 45"
83° 45' 25"
3
—
2
0
217
L-l
Light
8° 19'
83° 36'
20 m. W. of Gulf of
Dulce, Costa Rica
4
0
30
1938]
Beebe: Eastern Pacific Expeditions — Introduction
297
Sta-
tion
Net
Type of
Depth
Position
General
Date
Bottom
Duration
No.
No.
Net
Fath-
oms
Metres
N. Lat.
W. Long.
Locality
1938
Hrs.
Mins.
218
D-l
4' dredge
100
183
8° 35' 30"
83° 15' 30"
Gulf of Dulce, Costa
Mar.
9
Mangrove
0
6
218
D-2
4' dredge
105
192
8° 35' 50"
83° 16' 40"
Rica.
Gulf of Dulce, Costa
9
leaves.
Mangrove
0
15
D-3
4' dredge
80-60
145-109
8° 38' 10"
83° 16' 45"
Rica.
9
leaves.
0
10
D-4
2' dredge
6
11
8° 38' 15"
83° 10' 25"
Golfito, Gulf of
9
M angrove
0
3
D-5
2' dredge
6
11
8° 38' 15"
83° 10' 25"
Dulce, Costa Rica
9
leaves, mud,
shells.
0
3
D-6
2' dredge
4
7.3
8° 38' 15"
83° 10' 25"
9
a
0
3
D-7
2' dredge
6
11
8° 38' 15"
83° 10' 25"
*
9
“
0
3
D-8
2' dredge
6
11
8° 38' 15"
83° 10' 25"
“
9
0
3
L-l
Light
—
—
8° 38' 12"
83° 10' 45"
“
6
—
2
0
L-2
Light
—
—
8° 38' 12"
83° 10' 45"
u
7
—
2
0
L-3
Light
—
—
8° 26'
83° 08' 40"
Pavon Bay, Gulf of
9
—
2
0
219
T-l
Metre
300
540
8° 08'
83° 17'
Dulce, Costa Rica
25 m. W.xN. of Pt.
10
4
3
T-2
(3 nets)
Metre
500
910
8° 03'
83° 12'
Burica, Costa Rica.
10
2
0
220
L-l
(3 nets)
Light
_
_
8° 07'
82° 22' 12"
Isla Parida, Gulf of
11
4
0
L-2
Light
_
_
8° 07' 52"
82° 18' 45"
Chiriqui, Panama.
12
2
30
221
D-l
4' dredge
35
64
7° 54' 45"
82° 04' 32"
Gulf of Chiriqui,
13
Sandy mud.
0
5
D-2
4' dredge
35
64
7° 54' 15"
82° 04' 25"
Panama.
13
«
0
5
D-3
4' dredge
35
65
7° 53' 12"
82° 02' 45"
13
0
5
D-4
4' dredge
38
69
7° 52' 45"
82° 02'
“
13
“
0
8
D-5
4' dredge
40
73
7° 52' 30"
82° 01'
13
0
10
222
D-l
2' dredge
3
5.4
7° 45' 51"
81° 32' 21"
Bahia Honda, Pan.
18
Rocks, dead
0
2
D-2
2' dredge
4-8
7.2-14.5
7° 45' 46"
81° 32' 23"
«
18
coral.
0
3
D-3
2' dredge
8
14.5
7° 45' 42"
81° 32' 24"
18
Dead coral.
0
3
D-4
2' dredge
11
20
7° 45' 39"
81° 32' 21"
“
18
Dead coral,
0
3
D-5
2' dredge
11
20
7° 45' 35"
81° 32' 18"
u
18
shells, gr.
mud.
Mud, shells,
0
10
L-l
Light
_
_
7° 45' 15"
81° 32' 10"
u
13
leaves.
2
15
223
L-l
Light
—
—
7° 32'
81° 52' 30"
Bahia Hermosa,
19
—
3
30
224
D-l
4' dredge
40
73
7° 23' 30"
82° 03'
Coiba Is., Panama.
Hannibal Bank, Pan.
20
Rocks, dead
0
4
D-2
4' dredge
35
64
7° 23' 30"
82° 03'
«
20
coral.
Rocks, mud,
0
10
D-3
4' dredge
35
64
7° 23' 30"
82° 03'
«
20
dead coral.
Sand, shells,
0
15
225
T-l
Metre
500
910
7° 08'
81° 57'
11 m. SW.xW. of Jic-
20
algae.
2
31
226
L-l
(3 nets)
Light
_
7° 05'
81° 27'
aron Is., Panama.
22 m. ESE. of Jicaron
20
0
30
227
T-l
Metre
500
910
7° 00'
80° 40'
Is., Panama.
20 m. SW. of Morro
21
3
41
228
T-l
(3 nets)
Metre
500
910
7° 00'
79° 16'
de Puercos, Panama.
52 m. SE.xE. of Cape
25
4
30
229
L-l
(3 nets)
Light
_
_
6° 27'
79° 00'
Mala, Panama.
83 m. SE. of Cape
25
0
35
Mala, Panama.
298
Zoologica: New York Zoological Society
Sta-
tion
Net
Type of
Depth
Position
General
Date
Bottom
Duration
No.
No.
Net
Fath-
oms
Metres
N. Lat.
W. Long.
Locality
1938
Hrs.
Mins.
230
T-l
Metre
500
910
5° 10'
78° 42'
71 m. W.xS. of Cape
Mar.
26
4
2
231
L-l
(3 nets)
Light
©
eo
o
78° 33'
Corrientes, Columbia.
85 m. SW. of Cape
26
0
35
232
D-l
2' dredge
2-8
3.6-14.6
2° 57' 30"
78° 11'
Corrientes, Columbia.
Gorgona, Island,
31
Sand.
0
30
L-l
Light
_
2° 59'
78° 11'
Columbia.
27
_
2
30
L-2
Light
—
—
2° 59'
78° 11'
28
—
4
0
L-3
Light
—
—
2° 59'
78° 11'
“
29
—
3
30
L-4
Light
—
—
2° 59'
78° 11'
30
—
4
0
L-5
Light
—
—
2° 59'
78° 11'
“
31
—
4
0
233
T-l
Metre
500
910
4° 45'
78° 02'
55 m. SSW. of Cape
Apr.
3
_
4
33
234
T-l
(3 nets)
Metre
500
910
7* 24'
78° 35'
Corrientes, Columbia.
24 m. W.xS. of Pinas
4
_
4
27
(3 nets)
Pt., Panama.
Beebe & Tee-Van: 7 New Marine Fishes
299
15.
Eastern Pacific Expeditions
of the New York Zoological Society, XV.
Seven New Marine Fishes from Lower California.1
William Beebe
&
John Tee-Van
Department of Tropical Research,
New York Zoological Society
(Plates I-III; Text-figures 1-5).
| This is the fifteenth of a series of papers dealing with the collections
of the Eastern Pacific Expeditions of the New York Zoological Society
made under the direction of William Beebe. For data on dredges, localities,
dates, etc., concerning the capture of specimens treated in this paper, refer
to Zoologica, XXII, No. 2, pp. 33-46.]
Contents.
Page
Mobula lucasana, sp. nov 299
Arenichthys apterus, gen. nov., sp. nov 301
Citharichthys gordae, sp. nov 302
Scorpaenodes cortezi, sp. nov 304
Ammodytes lucasanus, sp. nov 306
Emblemaria micropes, sp. nov 308
Acanthemblemaria crockeri, sp. nov 310
Mobulidae.
Mobula lucasana, sp. nov.
(Plate I, Figs. 1 & 2; Plate II, Figs. 3 & 4; Plate III, Fig. 5).
Type: Holotype 2; No. 24,793, Department of Tropical Research, New
York Zoological Society; San Lucas Bay, Lower California, Mexico; March
30, 1936; width across disk 1,035 mm. (40% inches) ; length from anterior
part of disk to posterior margin of the pelvic fins, 584 mm. (23 inches).
Type in the collections of the Department of Tropical Research, the head,
posterior body and tail preserved, the latter damaged, only a length of 7
inches projecting beyond the outer margin of the pelvic fins.
1 Contribution No. 561, Department of Tropical Research, New York Zoological Society.
300
Zoologica: Neiv York Zoological Society
[XXIII :15
Description and Dimensions of Type: Skin smooth, no trace of rugosi-
ties or spines anywhere. Width across disk 1,035 mm. ; length from anterior
margin of disk to posterior margin of the pelvic fin 584 mm. (anterior
margin of the disk considered as being at the mid-line of the body, thus not
including the cephalic projections) ; cephalic fins extending forward 80 mm.
beyond the anterior margin of the disk, the distance between their tips
approximately 150 mm., distance from upper margin of the cephalic fin
to lower margin of the fin when the fin is unfurled, 58 mm. Eye lateral,
17 mm. in diameter, its anterior margin 88 mm. from the tip of the cephalic
fin. Spiracle oblique, 31 mm. posterior to the eye. Mouth inferior, 128 mm.
wide, very slightly concave when viewed from below, the margin of the
upper lip 40 mm. from the anterior margin of the disk.
Teeth in each jaw in a narrow band, each band extending 70 percent,
of the width of the mouth. Upper jaw with 80 transverse rows of teeth and
5 rows fi’om back to front. Lower jaw with 97 teeth in a transverse series
and with 4 or 5 rows from back to front. Teeth small, flattened, their
surfaces slightly roughened, the posterior border with 2 to 5 dull, irregular
points. A typical tooth in the center of the lower jaw measures 1.2 mm.
in width and .5 to .6 mm. in depth.
Mouth to transverse line of first gill-openings 70 mm.; transverse dis-
tance between first gill-openings 133 mm. ; transverse distance between last
gill-openings 60 mm.; length of gill-slits of first four pairs of gill-openings
52 mm. ; length of last gill-slit 37 mm.
Pelvic fins rather elongate, their tips extending 31 mm. beyond the
posterior tip of the pectoral, the inner margin extending backward farther
than the outer, the fin nearly uniform in width, averaging 38 mm.
Dorsal fin with its posterior one-fourth situated above the free portion
of the tail, the base of the fin 57 mm., the height 46 mm.
Tail immediately beyond the dorsal fin flattened, the skin of its upper
surface reticulated, the reticulations being in the form of minute, low,
raised ridges of skin, the upper margins of which are roughened and occa-
sionally slightly ciliate. Although the area mentioned above has this peculiar
specialized skin there is no trace of a spine or of the beginnings of a spine.
Tail beyond the expanded portion becoming attenuated near the tips of the
pelvic fins, its diameter 50 mm. posterior to the tips of the pelvic fins being
3.5 to 4 mm.
Color : In life upper surfaces, including the area about the eye and the
upper part of the cephalic fins, blackish-gray; lower surfaces whitish be-
coming bluish-gray toward the tips of the wings, this color darkest on the
anterior portion of the fins. Leading edge of undersurfaces of pectoral fin
dusky, the posterior edges similarly colored but the band of color narrower.
An oval, dusky spot near the posterior margin of the lower side of the
pectoral fin, slightly nearer the pelvic fins than the tip of the pectoral. Tip
of the cephalic fin black.
The preserved type has the underside of the body, anterior to the
mouth, and a narrow band along the lower jaw as well as the entire inner
surface of the cephalic fin, brownish-black.
Remarks: Two species of Mobula are recorded from the eastern Pacific,
Mobula tarapacana (Philippi) from the coast of Chile and M. japanica,
recorded from Hawaii and known principally from the western Pacific.
Mobula tarapacana was described from a drawing of a dead specimen
and there is considerable question as to the accuracy of the description and
figure. The present species, M. lucasana, however, does not possess the
peculiar long neck of the drawing of tarapacana.
In all of the available descriptions of M. japanica the bands of teeth
are described as extending to, or nearly to, the angles of the jaws. The
teeth of the present species by no means extend that far. In addition
1938]
Beebe & Tee-Van: 7 New Marine Fishes
301
japanica possesses a caudal spine and, even though the caudal of lucasana
is damaged, there is sufficient to show that the spine is lacking. Jordan and
Fowler’s figure ( Proc . U. S. National Museum, XXVI, p. 666, fig. 10) of a
fetus of japanica shows quite clearly a spine on the tail close to the pelvic
fins, a condition that is not true of lucasana. However, in the present species
it is of interest that the skin of the upper portion of the base of the tail is
modified, a condition that may or may not be significant in view of the
absence of a spine. This modification is mentioned in the description of the
species.
During the 1936 Zaca Expedition these fish were common at Cape San
Lucas and San Lucas Bay and a number of individuals were seen somer-
saulting in the air and striking the water in descent with a sound audible
for considerable distances. In January and February, 1938, numerous indi-
viduals, presumably of this species, were seen off Elena Point and Cape
Velas, Costa Rica. Probably some of the west coast of Central America
sight records of Manta can be assigned to the present new species.
The type of lucasana was captured on a baited hook.
Myridae.
Arenichthys, gen. nov.
Body elongate; dorsal fin beginning close behind gill-opening; pectoral
fins absent; anterior nostril in a tube; posterior nostril in a flap-like tube
which is open on its inner posterior aspect, the tube as large and con-
spicuous externally as the anterior nostril; teeth as in the description of
the species.
Genotype: Arenichthys apterus Beebe and Tee-Van.
Arenichthys apterus, sp. nov.
(Text-fig. la, lb, lc).
Type: Holotype, No. 25,361, Department of Tropical Research, New
York Zoological Society; taken in dredge 136-D-16 in 45 fathoms, Arena
Bank, Lower California (23°29'30" N., 109°25'30" W.), April 20, 1936;
standard length 135 mm. Type in the collections of the Department of
Tropical Research.
Measurements of the type: Length, standard, 135 mm.; depth 4.1 mm.
(33) ; head 16.6 mm. (8.1) ; eye 1.6 mm. (10.2) ; snout 3.6 mm. (4.6) ; snout
to gape 5.8 mm. (2.85) ; snout to dorsal fin 19 mm. (7.1) ; snout to anus
46 mm. (2.95) ; snout to anal fin 47.5 (2.84).
Description: Body elongate, worm-like, very slightly compressed, the
depth 33 times in the length; head and trunk 2.95 in the length. Head 8.1
in the length, 1.76 in the trunk; gill opening oblique, at the level of the
mid-sides. Gape extending backward slightly beyond the vertical of the
posterior border of the orbit.
Maxillary teeth small, conical, in three rows, the teeth of the inner row
considerably larger than the others; all are slightly depressible. Vomer
anteriorly with a series of teeth about its border; progressing backward
from these border teeth are three teeth set transversely on the median line
followed by two teeth set on the median line; posterior to these are two
lines of teeth on the shaft of the vomer, the two rows rather widely sep-
arated from each other, the teeth of the central portion of each row slightly
heavier than those of the anterior and posterior portions. Lower jaw with
a double row of small conical teeth becoming an irregular three rows an-
teriorly, the teeth of the inner row largest.
302
Zoologica: New York Zoological Society
[XXIII :15
Text-figure 1.
A. Arenichthys apterus. B. Arenichthys apterus. Diagram of teeth of the upper
jaw. C. Arenichthys apterus. View from inside of mouth of right side of
upper lip, showing (left) the tubular anterior nostril, and (right) the pos-
terior nostril.
Anterior nostril in a tube; posterior nostril in a flap-like tube which is
open on its inner posterior aspect. Eye small, 10.2 in the head.
Dorsal fin beginning slightly less than an eye’s diameter posterior to
the vertical of the gill opening, confluent with the anal fin about the tip of
the caudal. Pectoral fins absent.
Color: Brown above, white below, the boundary sharply demarcated,
especially on the head. Here the brown of the interorbital space extends
forward to the snout as a band the width of the interorbital space, leaving
the upper lip and the side of the head white with the exception of a rather
long, triangular, brown spot extending forward from the eye toward the
snout. Lower jaw with a narrow brown band along the lip, most intense
interiorly, and hidden by the upper jaw when the mouth is closed.
Bothidae.
Citharichthys gordae, sp. nov.
(Text-fig. 2).
Type: Holotype, No. 25,785, Department of Tropical Research, New
York Zoological Society; taken at Station 150, Dredge Haul 14; depth 60
fathoms, center of outer Gorda Bank, Lower California, April 23, 1936;
standard length 97 mm. Type in the collections of the Department of Trop-
ical Research.
Dimensions of Type : Total length 120 mm. ; standard length 97 mm. ;
depth 40 mm. (2.4 in length) ; head 36 mm. (2.7 in length) ; eye 10.5 mm.
(3.7 in head) ; interorbital space 1.3 mm. (8 in eye) ; maxillary 12 mm.
(3 in head) ; pectoral length 22.5 mm. (1.6 in head).
Counts of Type: Dorsal fin 76; anal 57; pectoral 12; gill-rakers 19;
lateral line scales 50 ; vertebrae 33.
1938]
Beebe & Tee-Van: 7 Neiv Marine Fishes
303
Description : A medium-sized flounder (standard length up to 111 mm.),
moderately deep (2.4 in length) ; eyes separated by a narrow, concave, scaled
space which rises to a sharp ridge above lower eye; interorbital width sim-
ilar in both sexes; dorsal beginning on blind side opposite front of eye;
upper pelvic fin on ventral profile; head moderately compressed with the
upper profile very slightly concave; anterior margin of the eyes are level
or the lower slightly in advance; upper eye touching profile of head; max-
illary not reaching middle of eye; teeth slightly enlarged anteriorly but not
forming real canines; indications of an imperfect second row in the upper
jaw; gill-rakers long and slender; scales relatively strongly ctenoid.
Color : Immaculate below; brown above with various dark mottlings,
sometimes reduced to several small spots, or increasing to larger blotches,
the extreme is where the dark color has run together covering three-fourths
of the upper surface. The fins are dusky, sometimes indistinctly mottled or
spotted; opercular region and lateral line always dark.
In specimens of 50 mm. standard length or under, there is often a faint
but distinct regular pattern. The body is pale with 6 to 8 broken cross
bands, very distinct as mid-ray spots on the profile fins, and about as strong
but of greater extent on the baseosts ; these bands are very irregular on the
body proper but in general follow the oblique muscle bands. The pre-
peduncular band is almost complete and the dark markings on the peduncle
are reduced and concentrated to a pair of jet black dots in the profile.
Variation : Considering that 216 specimens of this new form were taken
on Gorda Bank, measuring from 31 to 111 mm. in length, the extremes
of variation shown were small. Depth in length 2.3 to 2.4 (average 2.3) ;
head in length 2.68 to 2.7 (2.7) ; eye in head 3.4 to 4.15 (3.75) ; snout in
head 5 to 6 (5.5) ; maxillary in head 2.78 to 3 (2.9) ; dorsal count 73 to 79
(76) ; anal count 57 to 62 (58.7) ; scales 48 to 50 (48.6) ; gill rakers 17 to
19 (18).
General Relationships: In general appearance and the more obvious
external characters C. gordae approaches C. fragilis, a species which has
been taken only in the upper half of the Gulf of California, from Guaymas
northward. On the other hand the low vertebral count of gordae (34) re-
moves it, according to Norman, from the subgenus Orthopsetta “species of
the North Pacific,” (with 37 to 40 vertebrae, including sordidus, fragilis,
Text-figure 2.
Citharichthys gordae. Drawing by George Swanson.
304
Zoologica: New York Zoological Society
[XXIII :15
xanthostigma and stigmaeus) , and places it in the subgenus Citharichthys
“species of the Atlantic and Tropical Pacific,” (with 33 to 36 vertebrae),
including the Pacific species platophrys and gilberti. In scale and gill-raker
count and interorbital width it is indistinguishable from fragilis, while in
dorsal and anal ray count it departs widely from this species and is within
the range of both species of subgenus Citharichthys. In depth and pectoral
length it is closest to stigmaeus. C. gordae differs from the two species
platophrys and gilberti, in possessing smaller scales, a greater number of
gill-rakers, a larger head, shorter maxillary and a much narrower inter-
orbital. The distinctions between the two subgenera Citharichthys and
Ortliopsetta are very slight and may disappear on more intensive study.
Local Distribution : Two hundred and sixteen specimens of this new
species were taken by us, 200 of which came up in ten di’edge hauls at
Station 150, on Gorda Banks. The single catch of these fish outside of the
banks but in the same general locality was from Station 151 D-l, when 16
came up from a depth of 60 fathoms on April 24, half a mile south of Cape
San Lucas and 20 miles south-west of Gorda Banks.
The ten catches on Gorda were made on April 21, 22 and 23, 1936. The
extremes in vertical limits were 45 to 80 fathoms, in size 31 to 111 mm. and
in abundance in hauls were 2 in D-6 and 50 in D-4. The ten hauls show a
remarkable relation to sandy bottoms, none being on muddy or rocky areas.
The 16 flounders taken at Station 151 measure from 31 to 111 mm.
in length, an average far and away smaller than those from Gorda, and this
may have significance in the proximity of the locality to San Lucas Bay
which proved, in the case of many other fish, to be a nursery for young
stages.
SCORPAENIDAE.
Szorpaenodes e© rfezi, sp. nov.
(Text-fig. 3).
Type: Holotype, No. 24,889-A, Department of Tropical Research, New
York Zoological Society. Taken at surface in weed, Station 138, 24° 55' N.
Lat., 110° 20' W. Long., 10 miles east of San Jose Island, Gulf of Cali-
fornia, April 8, 1936. Standard length 45.5 mm. Type in the collections of
the Department of Tropical Research.
Measurements and Counts of Type: Length, total, 54 mm.; length,
standard, 45.5 mm. ; depth 16 mm. ; width of body 8.8 mm. ; head 15.8 mm. ;
eye 4.2 mm. ; snout 5.3 mm. ; maxillary 8.2 mm. ; interorbital space, 3.2 mm. ;
snout to dorsal fin 16.1 mm.; snout to anal fin 31 mm.; dorsal fin XIII, 13;
anal fin, III, 6; pectoral fin 18; pelvic fin I, 5; scales, approximately 63 rows;
lateral line pores 41 ; gill-rakers 10 plus 20.
Description: Body considerably compressed, its width 5.2 in the length,
depth 2.84; the dorsal outline evenly rounded.
Body, with the exception of snout, chin, maxillary, branchiostegal mem-
branes and isthmus, covered with small, weakly ctenoid scales, approximately
63 rows from shoulder to caudal fin.
Lateral line continuous, prominent, following the dorsal outline and
composed of 41 pores.
Head 2.86 in the length, about 1% times as deep as wide. Spines of
the head low, the following present: nasal, preocular, supraocular, post-
ocular and parietal. Opercle posteriorly with two flat spines, the upper
slightly longer than the lower, the lower heavier at its base. Preopercle
with five rather prominent spines, the second and third slightly stronger
than the others. Suborbital stay weak but evident without dissection.
1938]
Beebe & Tee-Van: 7 New Marine Fishes
305
Text-figure 3.
Scorpaenodes cortezi. Drawing by George Swanson.
Snout 3 in head. Eye small, 3.75 in the head. Nostrils small, slightly
nearer eye than snout, the anterior with a slight raised rim.
Mouth terminal, the jaws equal. Maxillary 1.9 in the head, its pos-
terior tip extending to the vertical of slightly beyond the posterior margin
of the pupil. Upper and lower jaw with a band of small conical teeth, the
band widening anteriorly. Vomer with small teeth on its head, none on the
shaft. Palatine teeth apparently absent in the type specimen, but present
as a band of minute teeth in a 39 mm. specimen that has been stained with
alizarin and cleared in caustic potash.
Gill-rakers 10 plus 20 on the first arch; rakers slim, the longest about
half an eye’s diameter. No slit behind the fourth gill-arch.
Spines of the dorsal fin regularly increasing in length from the first to
the fifth, the length of the latter being two in the head, the spines then
decreasing in length, the last spine about half the height of the first ray.
Soft dorsal fin anteriorly about as high as the highest spine, slightly lower
posteriorly. First anal spine less than half the height of the second ; second
anal spine considerably stronger and heavier than the other two; tip of the
second extending slightly beyond that of the third. Anal rays slightly
longer than the second anal spine. Caudal fin truncate. Pectoral fin rounded
posteriorly, all its rays simple, the tips of the rays expanded and flattened,
the tips with heavy skin. Tips of the lower three to four rays free of mem-
brane. Pelvic fins inserted posterior to the vertical of the pectoral fin base.
Color : A color sketch made from the living fish is as follows: Body
dark reddish-brown with a yellowish tinge, paler below, the entire body
covered with black punctulations and irregular greenish-yellow blotches,
the latter especially marked posteriorly. Dorsal fin dark brown, the mem-
branes of the spinous dorsal black, the soft dorsal irregularly blotched with
lighter. Caudal fin yellowish-green, without pattern. Pectoral fin brown-
ish, yellow toward the tip and dusky at the base. Pelvic fins black at base,
yellow at tips and with scarlet along the anterior edge. Anal fin black at
base, yellow on outer half and with a small scarlet patch at base of the
first spine.
In preservative the body is dark brown with irregular, rather large,
lighter blotches.
306
[XXIII :15
Zoologica: New York Zoological Society
Ammodytidae.
Ammodytes lucasanus, sp. nov.
Type: Holotype, No. 25,249-A, Department of Tropical Research, New
York Zoological Society; from stomach of Euthynnus alletterata (Rafin-
esque), Cape San Lucas, Lower California, April 25, 1936; standard length
100 mm. Paratype: No. 25,541-A, from stomach of Seriola colburni Ever-
mann and Marsh, Cape San Lucas, Lower California, May 6, 1936 ; standard
length 103 mm. Types in the collections of the Department of Tropical
Research.
Thirty-one other specimens (Nos. 25,249, 25,541 and 25,555), 50 to 106
mm. standard length, are also in the collections. They were taken from the
stomachs of the following fish: Euthynnus alletterata (Rafinesque), Seriola
dorsalis (Gill) and Seriola colburni Evermann and Marsh, and from a cor-
morant, Phalacrocorax penicillata (Brandt) ; all of these from Cape San
Lucas, April 25, 1936 and May 6, 1936.
Dimensions of Type : Total length 114 mm.; standard length 100 mm.;
depth 11.3 mm. (8.85 in length) ; width of body 8.6 mm. (2.72 in head) ;
head 23.5 mm. (4.25 in length) ; eye 3.8 mm. (6.2 in head) ; interorbital
space 2.6 mm. (9 in head) ; snout 6.6 mm. (3.55 in head) ; maxillary 7.7
mm. (3.04 in head) ; predorsal length 25 mm. (4 in length) ; preanal length
66 mm. (1.6 in head) ; pectoral fin length 10.2 mm. (2.3 in length) ; caudal
peduncle depth 5.1 mm. (4.6 in length).
Counts of Type: Dorsal fin 47; anal fin 22; pectoral fin 15; 97 lateral
folds ; branchiostegal rays 7 ; gill-rakers 6 plus 23 ; vertebrae 56.
Description: Elongate, spindle-shaped, tapering toward head and caudal
fin, the head quite sharp and pointed. Upper and lower profiles similar.
Body, with the exception of the head, covered with scales, the scales
being cycloid but with traces of small points along the posterior edge, the
whole suggesting a weakly ctenoid scale.2 A series of 97 oblique folds along
the sides, the upper margin of each fold being anterior to the lower. A fold
of skin along the lower side of the belly ending near the beginning of the
anal fin, this fold not noticeable in the preserved type but remarked upon
in the field notes on fresh specimens.
Lateral line single, paralleling the back ; anteriorly the lateral line partly
surrounds the eye, having an angled branch descending to under the pos-
terior lower edge of the eye, and an upper branch which runs obliquely
forward ending posterior to the vertical of the eye; on the posterior head
the lateral line follows closely the upper border of the opercular margin,
then sharply ascends to the position of the line on the body; a short trans-
verse branch of the lateral line on the nape, not meeting its fellow of the
opposite side.
Head conical, 4.25 in the length, the lower jaw projecting considerably
beyond the tip of the upper and with a heavy fleshy tab on its tip. Eye 6.2
in head. Nostrils subcircular, the anterior slightly larger, situated midway
between the eye and the tip of the premaxillary, placed one before the
other and widely spaced from each other. Mouth widely protractile, the
jaws, vomer and palate toothless. Maxillary extending slightly beyond the
vertical of the anterior border of the eye.
Opercular and preopercular borders smooth, the former ending pos-
teriorly in a wide obtuse flap. Branchiostegal rays 7, the membranes free
from each other, leaving a wide opening interiorly.
Dorsal fin low, continuous, formed of rays only, originating slightly
back of the pectoral origin. Caudal fin deeply forked. Anal fin low, con-
2 This is also true of specimens of Ammodytes personatus from the San Juan Islands, Wash-
ington (Amer. Mus. Nat. Hist., 2712).
1938]
Beebe & Tee-Van: 7 Neiv Marine Fishes
307
tinuous, formed of rays only, originating under the 29th dorsal ray. Pec-
toral fins small, pointed, situated below the median line of the body. Pelvic
fins absent.
Color : Fresh specimens greenish above, silvery on sides with traces of
bright golden yellow on the sides of the head and upper body. A series of
rather widely placed black spots (8 in the holotype, 4 in the paratype)
along the upper margin of the dorsal fin.
Variation : In a series of ten specimens examined, including the para-
type, there is the following variation in proportions and counts: depth 7.8
to 9.45; head 4 to 4.25; eye 6.8 to 8; snout 3.5 to 3.8; dorsal fin count 45
to 46; anal fin count 20 to 23; lateral folds 97 to 100; gill-rakers 6 plus
22 to 23.
Remarks : During the last few years there has been a tendency to con-
sider many of the northern forms of Ammodytes as being a single circum-
polar species which sends representatives down the coasts of the continents
to a greater or lesser degree, the distance depending upon the temperature
of the water and other conditions.
As far as the Pacific is concerned, ichthyologists seem in agreement
that all of the previously reported Ammodytes in that region are of one
species. Thus the nominal form alascanus was merged during the latter part
of the last century with personatus ; Jordan in 1906 considered the Japanese
form as the same species and in 1930 Soldatov and Lindberg utilized the
name personatus for the northwestern Pacific fish. Finally Schultz in 1936
has related the Pacific form to the circumpolar species by using the name
Ammodytes tobianus personatus (Girard) for the Pacific fish.
With this taxonomic history in mind the counts and measurements of
two series of eastern Pacific Ammodytes, one series from Karluk Island,
Alaska, and the second from Straits of Juan de Fuca, Washington, together
with the records in the literature of California specimens, have been com-
pared with those of our San Lucas material. The most pertinent counts are
included in the following diagram; unfortunately we have not been able to
procure California specimens, so that we have no vertebrae counts for this
locality.
AMMO DYTES
LUCASANUS
PE R S ONATUS
' SAN LUCAS
CALIFORNIA
WASHINGTON
ALASKA
DO R S A L
F 1 N
RAYS
60 ;
so ;
40 ;
B
■
1
_ _ fi
ANAL
F 1 N
RAYS
35 :
30 ;
25 ;
20 '■
I
- « -
I""""
9 __
LATERAL
FOLDS
ISO -
140
too
8
1
V E RTEBRAE
75 h
70
65
60
55
•
1
8
308
Zoologica: Neiv York Zoological Society
[XXIII :15
Examination of the table above demonstrates the apparent close rela-
tionship of lucasanus to the northern Ammodytes. In each of the characters
shown there is a numerical diminution as one progresses southward, a con-
dition similar to that found in other groups of fishes. In the case of the
lateral folds, however, the fishes from California to Alaska are markedly
closer to each other than they are to lucasanus; this similarity is shown
to a lesser degree in the dorsal fin rays.
While it might be advisable to consider lucasanus as a southern repre-
sentative of personatus, we think it better, for the present, to establish it
as a separate form, principally for the following reasons :
1. A gap of 1,100 miles intervenes between the southernmost recorded
locality of personatus (Monterey Bay, California) and the type locality of
lucasanus. Specimens from this gap would help to establish the relationship
between the two forms.
2. The San Lucas fish come from a sub-tropical faunal region, as op-
posed to the cold water habitat of personatus. San Lucas is about 420 miles
south of Cedros Island where northern and southern faunas meet and cross.
3. With this faunal difference in mind it is possible that the lessened
numerical characters may be constantly associated with the warm water
habitation of lucasanus and that this species may have broken off completely
from its northern parent form, leaving no intermediates.
Emblem ariidae.
Emblemaria micropes, sp. nov.
(Text-fig. 4).
Type: Holotype, $, No. 24,895, Department of Tropical Research, New
York Zoological Society. Taken in loose weed floating along beach, Inez
Point, Santa Inez Bay, Gulf of California, April 9, 1936. Standard length
33 mm. Type in the collections of the Department of Tropical Research.
Dimensions and Counts of Type : Length, total, 37.6 mm. ; length, stand-
ard, 33 mm. ; depth 5.4 mm. ; head 7.8 mm. ; eye 1.4 mm. ; snout 1.9 mm. ;
maxillary 3 mm. ; interorbital space 1 mm. ; snout to dorsal fin 5 mm. ; snout
to anal fin 15 mm.; height of dorsal fin 7 mm.; pectoral fin length 6.8 mm.;
pelvic fin length 3.2 mm. ; height of supraocular tentacle 2 mm. ; dorsal fin,
XIX, 12; anal fin, II, 21; pectoral fin 13; pelvic fin 3.
Description: Body elongate, considerably compressed posteriorly, the
greatest width of body (across the swollen opercles) 6 in the length; depth
6.1.
Head 4.2 in the length ; preopercular area and head considerably
swollen. Upper profile of head a simple curve from nape to snout. A large
flattened supraorbital tentacle, slightly longer than eye diameter. Pre-
opercular margin not visible, hidden by the swollen sides of the head.
Opercular margin smooth, the branchiostegal rays and membranes extend-
ing considerably beyond its margin, the tip of the outermost branchiostegal
ray forming a slight forwardly-curved hook above; branchiostegal mem-
branes broadly joined to each other but free from the isthmus.
Snout 4.1 in the head. Eye small, 5.6 in head, not entering the dorsal
profile. Nostrils two on each side, the anterior with a small fleshy tentacle.
Mouth small, the lips rather full; maxillary extending to posterior border
of the eye. Teeth in the upper jaw strong, blunt and conical, the anterior
ones strongest, in a single row, with a trace of asperities along the inner
base of the anterior teeth resembling small teeth. Lower jaw with a
similar series of teeth, the lateral anterior ones exceptionally strong and
recurved. Vomerine teeth in a slightly curved row of four teeth. Palatine
teeth similar in size and not quite continuous with the vomerine teeth.
1938]
Beebe & Tee-Van: 7 New Marine Fishes
309
Dorsal fin beginning on the nape, of 19 spines and 12 rays, the last
ray connected by membrane with the caudal peduncle ; a slight flap of mem-
brane projecting forward from the edge of the 1st dorsal spine with the
exception of the basal portion of the spine. Caudal fin rounded. Anal fin
of 2 spines and 21 rays, similar to the dorsal but lower. Pectoral fins with
lowermost five rays slightly thickened and with their tips free of mem-
brane. Pelvic fins of three rays, short and thick, originating under the
pectoral base.
Text-figure 4.
Emblemaria micropes. Drawing by George Swanson.
Color : The following notes were made upon the living fish. Head,
supraocular tentacle, body and pectoral fins green, becoming more intense
green on the caudal fin. Anterior three-quarters of the dorsal fin with
alternate bands of green and bright red, each band including about two to
three rays; posterior quarter of dorsal fin with faint pinkish patches
distally in the webs. Anterior three-quarters of the anal fin green on its
basal third, similar to that of the body, black on its median third and with
the outer margin white; posterior quarter almost all green with only a
hint of dusky and whitish.
Remarks : The five species of Emblemaria known from the eastern
Pacific seem to be well differentiated from each other, not only by their
coloration but by various combinations of other characters. In the following
table the dorsal and anal fin counts and the head and pelvic fin proportions
of the various species are gathered together.
Species.
Dorsal fin
count.
Anal fin
count.
Pelvic fin
in length.
Head in
length.
nigra, Panama
XXVII, 11
26-27
7. 1-7. 5
4. 1-4. 2
nivipes, Panama
XXIII-XXIV,
12-14
25
3.8
3.4-3.75
hudsoni, Peru
XXII-XXIII,
15-16
27
3.9
3.9
oculocirrus, Gulf of
California
XXII, 13
25-26
4.9-5.75
3.75-4.1
micropes, Gulf of California
XIX, 12
23
10.3
4.2
This new species differs from the somewhat similar Emblemaria oculo-
cirrus from the same region, in fin ray counts, smaller and differently
shaped pelvic fins, low dorsal fin and in the supraocular tentacle. At first
it was though that the former was the opposite sex of the high-dorsalled
oculocirrus, but careful dissection disclosed that all of the available speci-
mens examined, four of oculocirrus and the type of micropes, were males.
While perhaps of no importance, it is of interest that the specimen of
310
Zoological New York Zoological Society
[XXIII :15
microbes was taken in weed along shore, while all three specimens of
oculocirrus were washed out of empty augur shells taken on sandy shallows.
For comparison with our materials of the genus we have examined the
single specimen of oculocirrus taken by Dr. C. H. Townsend and recorded
by Osburn and Nichols (Bull. Amer. Mus. Nat. Hist., XXXV, (1916),
p. 178).
Acanthemblemaria crockeri, sp. nov.
(Text-fig. 5).
Type: Holotype, No. 24,824, Department of Tropical Research, New
York Zoological Society; taken with fulminating cap, 3V2 fathoms deep,
San Lucas Bay, Lower California, April 2, 1936; standard length 41 mm.;
type in the collections of the Department of Tropical Research.
Dimensions of Type : Total length 46.2 mm. ; standard length 41 mm. ;
depth 6.2 mm. (6.6 in length) ; head 10.2 mm. (4 in length) ; eye 2.1 mm.
(4.85 in head) ; interorbital space 1.5 mm. (6.8 in head) ; snout 2.3 mm.
(4.4 in head) ; maxillary 4.8 mm. (2.1 in head) ; predorsal length 6.5 mm.
(6.3 in length) ; preanal' length 18.2 mm. (2.2 in length) ; pectoral fin length
6.2 mm. (1.65 in head) ; pelvic fin length 3.4 mm. (3 in head) ; caudal
peduncle depth 2.8 mm. (3.64 in head).
Counts of Type: Dorsal fin XXVI, 13; anal fin II, 27; pectoral fin 12;
pelvic fin 3; caudal fin 19; branchiostegal rays 6.
Text-figure 5.
Acanthemblemaria crockeri. Drawing by Miss Eloise Southern.
Description: Body moderately elongate, the greatest depth at the base
of the pelvic fins; the trunk compressed, especially posteriorly. The depth
of the caudal peduncle 3.64 in the head length. Body scaleless. Anterior
profile a simple curve from origin of dorsal fin to snout.
Head with short blunt spines anteriorly, those on the snout and at
the anterior edge of the eye largest, the spines extending backward on
the upper surface of the head slightly beyond the vertical of the posterior
border of the orbit, the posterior margin of the spinous area forming a
blunt angle at the mid-line of the body. Spines extending in a single
palisade-like series about the orbit, except posteriorly where there is a slight
elevated rim free of spines, and superiorly where there is a slight gap
for the supraorbital tentacle; anterior orbital spines strongest. Inter-
orbital space concave, its central portion free of spines. Supraorbital
tentacle multifid, tree-like, with a number of major trunks and branches
from which extend smaller ones, the height of one of the larger from its
tip to the base of the tentacle, slightly greater than an eye diameter.
Nasal tentacle of right side with a single-based bifid tentacle, that of the
left side single-based but quadrifid distally, the tentacles about equal to
an eye diameter in length. A group of pores extending slightly below and
along posterior border of the eye; a series of 8 pores along the posterior
border of the pi’eopercle and on to the underside of the posterior portion of
1938]
Beebe & Tee-Van: 7 New Marine Fishes
311
the mandible; 3 pores, forming a triangle, at the nape immediately before
the dorsal fin, and 4 pores on each side of the upper body above the pre-
opercle and opercle. Branchiostegal membranes joined to each other but
free from the isthmus, the rays extending upward beyond the opercular tip
to form a small hook.
Dorsal fin beginning above the center of the preopercle, the tips of both
rays and spines thickened; the fin connected by membrane to the caudal
peduncle but not to the caudal fin. Anal fin originating under the 13th dorsal
ray, similar to dorsal but its rays shorter and slightly heavier, the fin
similarly joined to the caudal peduncle. Pectoral fin with tips of lower
rays thickened. Pelvic fins originating before the pectoral, composed of 3
simple rays, the central ray longest, the inner ray shortest. Caudal fin
rounded.
Teeth : Upper jaw with a single row of strong, conical, inwardly curved
teeth, the teeth strongest anteriorly and much smaller along side of jaw;
inside of this row anteriorly is a band of small conical teeth. Palatines
with 2 rows of strong teeth. Vomer with about 5 strong teeth on each side.
Lower jaw expanded and flattened anteriorly, then constricted and ex-
panded again posteriorly, the jaw, when viewed from above, shaped like
an old-fashioned keyhole, or as suggested by Myers and Reid in their de-
scription of Acanthemblemaria hancocki, like the jaw of Hippopotamus.
Lower jaw with a single series of flattened strong, conical teeth, paralleling
which exteriorly, on the posterior half of the jaw, is a second row of some-
what similar teeth. Anteriorly behind the single strong series is a band of
very small conical teeth.
Color : In life, general color of body brownish-tan, darker above; sides
with small subcircular and oval spots of scarlet-red, these becoming dull
brown superiorly. Head yellowish laterally and below, dull reddish-orange
above and with a large oval patch of vinaceous-buff on the preopercle, this
patch surrounded anteriorly, superiorly and posteriorly by a narrow blackish
margin. Opercle with a small blackish patch near its posterior margin.
Pectoral and pelvic fins with greenish-yellow rays. Vertical fins brownish-
pink, the rays and spines pinkish. Caudal fin, rays of the dorsal fin and
posterior rays of the anal with pinkish-red spots along the rays. Anterior
portion of dorsal with a dull brown median band surmounted by a mar-
ginal orange band. Iris brownish-gold.
In preservative the scarlet and orange pigment has disappeared en-
tirely and the specimen may be described as follows: General color pale
buff, the entire body covered with minute brown chromatophores, the
chromatophores on the sides and lower sides forming the borders of sub-
circular clear patches (These are the scarlet spots of the life coloration).
Above, the chromatophores are more concentrated, forming irregularly-
shaped patches which extend onto the base of the dorsal fin. Dorsal fin
with brownish blotches, especially anteriorly where a spot of deep brown
and black extends down the center of the fin. Anal fin with dark spots
toward the outer margin, these especially marked anteriorly. Preopercle
with a large ovoid brownish spot, heavily bordered anteriorly, superiorly
and posteriorly with darker brown. Opercle mottled with brown.
Remarks'. The genus Acanthemblemaria was previously represented in
the eastern Pacific by the recently described Acanthemblemaria hancocki
Myers and Reid 19363, taken at Secas Island, Pacific coast of western Pan-
ama. The U. S. National Museum has kindly allowed us to have one of
the paratypes of hancocki for examination. Comparison of the single spec-
imen of crockeri with the original description of hancocki and the paratype
of the latter has enabled us to draw up the following key to show the differ-
ences between the two species.
3 Myers, G. S. and Reid, E. D., University of Southern California Publications , The Hancock
Pacific Expeditions, Vol. 2, No. 2, pp. 7-9.
312
Zoologica: New York Zoological Society
Key to Eastern Pacific Species of Acanthemblemaria.
A. Dorsal fin XXII to XXIII, 13 to 14; anal fin II, 24 to 25; spines of
head relatively small and fine; supraorbital tentacle simple; head with
a dark area like a large, hour-glass shaped saddle set transversely
across the occiput and extending down on opercles to opposite lower
rim of the eye; no oval, black-margined brown patch on the preoper-
cle; body, in preservative, pale with conspicuous brown spots hancocki.
AA. Dorsal fin XXVI, 13; anal fin II, 27; spines of head coarse and heavy;
supraorbital tentacle complex and multifid; no hour-glass shaped dark
figure on nape; an oval conspicuous, black-margined brownish spot on
preopercle; body, in preservative, with pale spots surrounded with
darker crockeri.
EXPLANATION OF THE PLATES.
Plate I.
Fig. 1. Mobula lucasana. Dorsal view of freshly caught type specimen. Photo-
graph by Toshio Asaeda.
Fig. 2. Mobula lucasana. Ventral view of freshly caught type specimen. Photo-
graph by Toshio Asaeda.
Plate II.
Fig. 3. Mobula lucasana. Under anterior surface of body showing folded cephalic
fins, mouth and teeth. The upper jaw teeth show as a dark band just
below the margin of the upper jaw. Photograph by Toshio Asaeda.
Fig. 4. Mobula lucasana. View of posterior end of animal from dorsal side,
showing the dorsal and pelvic fins and the expanded base of the tail.
Photograph by E. Osterndorff.
Plate III.
Fig. 5. Mobula lucasana. Photograph of teeth of the lower jaw. Enlarged 33
times. Photograph by Mrs. Ruth Needham Nauss.
BEEBE & TEE-VAN.
PLATE I.
FIG. I.
FIG. 2.
SEVEN NEW MARINE FISHES FROM LOWER CALIFORNIA.
BEEBE & TEE-VAN.
PLATE II
FIG. 3.
FIG. 4.
SEVEN NEW MARINE FISHES FROM LOWER CALIFORNIA,
BEEBE & TEE-VAN.
PLATE III.
FIG. 5.
SEVEN NEW MARINE FISHES FROM LOWER CALIFORNIA.
Taning: Family Anguillidae
313
16.
Deep-sea Fishes of the Bermuda Oceanographic Expeditions.
Family Anguillidae.1
A. Vedel Taning
Copenhagen, Denmark.
(Text-figure 1).
Dr. William Beebe has kindly entrusted to me the collection of lepto-
cephalus stages of the European and American eels made off Bermuda
during the years 1929-1931. As rather few eel larvae have been collected
in the deep water very close to Bermuda, it was of interest to me to see this
material procured during 1,350 hauls off Nonsuch Island.2 The specimens
are preserved in alcohol and are accordingly probably shrunk about 2 mm.
as compared with specimens preserved in formalin.
The collection is small, comprising only 46 specimens, of which 29
belong to the European eel and 17 to the American eel. The specimens were
collected during the months of May through September, none being taken
in April and none in October or November when relatively few hauls were
made.
Moreover, rather few hauls were carried out in the water layers in
which we should expect most of these larvae, that is, in subsurface water
in depths between 25 and 200 fathoms. Only about 60 hauls were made in
these upper layers from May through September, as against about 1,100
in deeper water layers.
The eel larvae are, according to Johannes Schmidt’s experiences, most
frequently taken during night hauls and of these comparatively very few
were made during the Nonsuch investigations. In the collection only three
specimens of Anguilla anguilla were caught during night hauls.
The facts mentioned explain why so few eel larvae have been caught,
and we may moreover emphasize that the small net commonly used (1-metre
net) is not the best implement for the capture of eel larvae of a length
greater than about 40 mm., as these are rather rapid stages which easily
escape the net. This is especially the case with larval stages of the Amer-
ican eel, of which some in the collection are approaching the metamorphosis-
stage; during the latter period they become very strong and rapid in their
movements.
In regard to literature on the subject, I refer the reader to Schmidt’s
various publications mentioned in the Bibliography.
In the accompanying Tables I give various particulars concerning the
material, such as myomere counts. A few show the total number of 112
and 113 myomeres, and as an overlapping in numbers may occur in the two
1 Contribution No. 562, Department of Tropical Research, New York Zoological Society.
Contribution, Bermuda Biological Station for Research, Inc.
2 For /detailed data in regard to nets, locality, dates, etc., refer to Zoologica, Vol. XIII,
Nos. 1, 2 and 3.
314
Zoologica: New York Zoological Society
[XXIII :16
species, it could be assumed that the identification is not quite certain.
Since, however, the pre- and postanal number of myomeres and the blood
vessels furnish some help in the identification and since the size of the
larvae and the season of capture give further help, I am rather certain that
the few specimens showing the number of myomeres mentioned are referred
to the proper species.
Tables I and II show the sizes during the different months.
TABLE I. TABLE II.
The larval stages belonging to the European eel ( Anguilla anguilla )
are easily referred to the year’s brood (O-group) and the previous year’s
brood (I-group) ; probably a few of a length of about 36-37 mm. in June
and July and a single one of 41 mm. in September should actually be re-
ferred to respectively the I-group and the O-group, but this is of minor
importance.
It is well known that during the movement of the larval stages a few
may be behind the main stock moving toward Europe and remain as the
1938]
Tailing: Family Anguillidae
315
I-group for some time in the western Atlantic; the 13 specimens measuring
between about 41 and 50 mm. taken from May to September belong to this
slow-moving (and late spawned) part of the I-group.
On the other hand, the 16 smaller specimens, between 18 and 37 mm.
long, taken from June to September, belong most probably to the faster-
moving (and early spawned) part of the O-group which have already
reached the vicinity of the Bermudas in June.
In considering now Table II, which shows the lengths of larval stages
of the American eel ( Anguilla rostrata ) taken during the months men-
tioned, it is immediately obvious that these arrive in the Bermuda area
later, in July and August, and are of larger size than those of the O-group
of the European eel; this is quite in agreement with the fact that the breed-
ing place of the American eel is situated more to the southwest than that
of the other species, and that the spawning takes place earlier in the year.3
According to Schmidt, metamorphosis takes place at a length of 60 to 65
mm. in the American eel. As is well known, the Bermuda fresh water eel is
the American, not the European species. In regard to this, see William
Beebe and John Tee-Van, “American Fresh-water Eels in Bermuda,” Bull.
N. Y. Zool. Soc., Vol. XXXVII, 1934, pp. 181-183, and Johs. Schmidt, 1909,
pp. 17-18.
The following table (Table III) gives a review of the actual number
of larval stages taken in the different months and the number per 100 hauls.
It is obvious that the European species is most common from June to
August, and the American only during August and September. Actually,
however, we know but very little concerning conditions during the months
from October to April, and it would certainly be most interesting to get
investigations also carried out during these months.
TABLE III.
April
May
June
July
Aug.
Sept.
Oct.
Nov.
Total
Total No. of hauls
1929-3 i
63
214
260
236
192
362
15
8
1,350
No. of
Anguilla anguilla ....
0
3
10
7
6
3
0
0
29
No. per 100 hauls
1.4
3.8
2.9
3.1
0.8
2.2
No. of
Anguilla rostrata
0
0
0
2
6
9
0
0
17
No. per 100 hauls. .....
0.8
3.1
2.5
1.2
As in other areas of the Sargasso Sea the number of larval stages of
Anguilla anguilla surpasses that of A. rostrata.
As to the relative frequencies in the different months and in different
depths, the following table (Table IV) gives a clear review, showing that
the larval stages of the European eel are by far the most common in the
3 It is unlikely that the American eel spawns so close to Bermuda that eggs and young larvae
could be met with there; at any rate post-larval stages smaller than 15 mm. have not yet been
captured near Bermuda. Johannes Schmidt has previously (The Danish “Dana” Expeditions
1920-22, No. 1, p. 16, footnote, 1929) pointed out that the eggs without oil globule described by
M. P. Fish ( Zoologica , Vol. VIII, No. 5, p. 289 ff.) cannot be eggs of Anguilla rostrata. I have
myself hatched eggs of the same or a very similar type and I am most inclined to refer the eggs in
question to the Muraenoid species of which the larval stage is known as Leptocephalus similis.
The information obtained by the 1,350 hauls made by Dr. Beebe off Bermuda supports the view that
Anguilla rostrata propagates at a considerable distance from these islands.
316
Zoologica: New York Zoological Society
[XXIII :16
subsurface waters off Bermuda from May through July, very few occurring
in the deeper strata, where, however, the larval stages of the American eel
are most commonly taken during August and September, owing probably
also to the fact that the fishing during daytime in the lighter subsurface
layers of the water is not able to catch the rapid stages of these larger
sizes of American eel larvae.
Text-figure 1.
Distribution in the western Atlantic of the different larval stages of the European
and American eel. Unbroken line = European eel. Broken line = American
eel. Curves show limits of occurrence: e.g., larvae less than 10 mm. have only
been found inside the 10 mm. curve, etc. ul: curve showing limit of occur-
rence of unmetamorphosed larvae. Striated area: only place where the newly
hatched larvae (5-7 mm. long) of the European eel have been recorded.
American eel larvae less than 15 mm. have never been taken outside the
dotted area; the actual spawning place inside the dotted area is, however,
not yet known. (Redrawn from Schmidt, 1935).
1938] Tailing: Family Anguillidae 317
TABLE IV.
Depth
Total No.
of hauls
A. anguilla
+ . rostrata
May- July
Aug. Sept.
May-Julv
Aug.-
Sept.
Fath-
oms
Metres
May-
July
Aug.-
Sept.
No.
No.
per
100
hauls
No.
No.
per
100
hauls
No.
No.
per
100
hauls
No.
No.
per
100
hauls
25-200
c. 45-370
43
20
11
25.6
1
5.0
0
0.0
0
0.0
>300
> c.550
606
506
9
1.5
8
1.6
2
0.3
15
3.0
Though the material to hand is very small it gives some valuable
information as to the occurrence of the larval stages of the two species of
fresh-water eels off Bermuda, and I wish to emphasize that the area is
obviously very interesting for studies on the rate of growth and movement
of the eel larvae. Further studies would certainly give us much information
which otherwise will be difficult to obtain. A glance at the small map (Text-
fig. 1) will show immediately how excellent the position of Bermuda is for
these studies.
I take this opportunity of thanking Miss Esther Hansen for valuable
assistance.
Study Material.
The following list gives the catalogue number, net, depth in fathoms,
date, length and myomere count of each Anguilla leptocephalus taken by the
Bermuda Oceanographic Expeditions. All were caught in the cylinder of
water off the Bermuda coast described in Zoologica, Vol. XVI, No. 1, p. 5.
Anguilla anguilla.
No. 9,858; Net 85; 200 fathoms; May 5, 1929; 50 mm.; 72+46 (118) myom.
No. 9,895; Net 90; 200 fathoms; May 11, 1929; 43 mm.; 70+45 (115) myom.
No. 9,895a; Net 90; 200 fathoms; May 11, 1929; 43 mm.; 68+48 (116) myom.
No. 11,046; Net 227; 800 fathoms; June 27, 1929; 32 mm.; 73+44 (117) myom.
No. 11,075; Net 229; 1,000 fathoms; June 27, 1929; 43.5 mm.; 71+42 (113) myom.
No. 11,184; Net 245; 800 fathoms; July 1, 1929; 30.5 mm.; 73+40 (113) myom.
No. 11,707; Net 313; 800 fathoms; July 22, 1929; 47.5 mm.; 73+43 (116) myom.
No. 11,772; Net 320; 1,000 fathoms; July 23, 1929; 37 mm.; 70+44 (114) myom.
No. 11,819; Net 326; 1,000 fathoms; July 24, 1929; 36.5 mm.; 71+45 (116) myom.
No. 15,625; Net 657; 700 fathoms; June 2, 1930; 27 mm.; 74+41 (115) myom.
No. 15,746; Net 683; 900 fathoms; June 7, 1930; 37.5 mm.; 71+44 (115) myom.
No. 17,789; Net 839; 700 fathoms; Sept. 3, 1930; 41 mm.; 70+44 (114) myom.
No. 18,351; Net 870; 100 fathoms; Sept. 11, 1930; 34.5 mm.; 68+44 (112) myom.
No. 18,386; Net 874; 500 fathoms; Sept. 11, 1930; 44 mm.; 69+46 (115) myom.
No. 20,985; Net 1035; 100 fathoms; June 25, 1931; 27.5 mm.; 70+43 (113) myom.
No. 20,985a; Net 1035; 100 fathoms; June 25, 1931; 33 mm.; 69+44 (113) myom.
No. 21,020; Net 1042; 100 fathoms; June 26, 1931; 45 mm.; 70+46 (116) myom.
No. 21,020a; Net 1042; 100 fathoms; June 26, 1931; 41 mm.; 70+45 (115) myom.
No. 21,043; Net 1047; 100 fathoms; June 27, 1931; 48 mm.; 71+45 (116) myom.
No. 21,043a; Net 1047; 100 fathoms; June 27, 1931; 46 mm.; 71+46 (117) myom.
No. 21,152; Net 1058; 300 fathoms; July 7, 1931; ca. 18 mm.; 66+47+ (113+)
myom.
No. 21,342; Net 1080; 100 fathoms; July 14, 1931; 44.5 mm.; 71+46 (117) myom.
318 Zoologica: New York Zoological Society
No. 21,342a; Net 1080; 100 fathoms; July 14, 1931; 28 mm.; 73+ca. 43 (ca. 116)
myom.
No. 21,858; Net 1129; 1,000 fathoms; Aug. 4, 1931; 34 mm.; 72+43 (115) myom.
No. 22,233; Net 1166; 900 fathoms; Aug. 11, 1931; 45.5 mm.; 72+46 (118) myom.
No. 22,409; Net 1184; 800 fathoms; Aug. 15, 1931; 35 mm.; 73+44 (117) myom.
No. 22,760; Net 1192; 1,000 fathoms; Aug. 17, 1931; 36 mm.; 72+44 (116) myom.
No. 22,934; Net 1244; 800 fathoms; Aug. 31, 1931; 35.5 mm.; 70+49 (119) myom.
No. 22,905; Net 1245; 1,000 fathoms; Aug. 31, 1931; 35 mm.; 73+43 (116) myom.
Anguilla rostrata.
No. 11,650; Net 305; 600 fathoms; July 16, 1929; 43.5 mm.; 68+39 (107) myom.
No. 11,722; Net 314; 900 fathoms; July 22, 1929; 39.5 mm.; 67+39 (106) myom.
No. 12,877; Net 406; 800 fathoms; Sept. 2, 1929; 51 mm.; 69+39 (108) myom.
No. 13,125; Net 427; 900 fathoms; Sept. 5, 1929; 48 mm.; 68+40 (108) myom.
No. 13,193; Net 435; 1,000 fathoms; Sept. 6, 1929; 51.5 mm.; 71+38 (109) myom.
No. 13,343; Net 449; 1,000 fathoms; Sept. 9, 1929; 53 mm.; 71+39 (110) myom.
No. 17,759; Net 835; 500 fathoms; Sept. 3, 1930; 49.5 mm.; 70+38 (108) myom.
No. 18,029; Net 856; 800 fathoms; Sept. 6, 1930; 47.5 mm.; 69+39 (108) myom.
No. 18,098; Net 862; 800 fathoms; Sept. 8, 1930; 50 mm.; 69+40 (109) myom.
No. 18,549; Net 881; 600 fathoms; Sept. 12, 1930; 45 mm.; 69+38 (107) myom.
No. 18,496; Net 883; 700 fathoms; Sept. 13, 1930; 46.5 mm.; 68+38 (106) myom.
No. 22,305; Net 1169; 700 fathoms; Aug. 12, 1931; 51 mm.; 67+39 (106) myom.
No. 22,306; Net 1172; 1,000 fathoms; Aug. 12, 1931; 53 mm.; 70+39 (109) myom.
No. 22,667; Net 1209; 1,000 fathoms; Aug. 20, 1931; 41 mm.; 69+38 (107) myom.
No. 23,039; Net 1217; 600 fathoms; Aug. 24, 1931; 48 mm.; 70+39 (109) myom.
No. 23,040; Net 1240; 1,000 fathoms; Aug. 29, 1931; 45 mm.; 70+38 (108) myom.
No. 23,707; Net 1325; 500 fathoms; Sept. 19, 1931; 52 mm.; 70+39 (109) myom.
Bibliography.
Johs. Schmidt
1909. Remarks on the metamorphosis and distribution of the larvae of the
Eel. (Anguilla vulgaris Turt.). ( Medd . Komm. Havunders0gelser,
Serie: Fiskeri, Bd. Ill, Nr. 3, pp. 1-17. Kpbenhavn).
1909. On the distribution of the fresh-water Eels (Anguilla) throughout
the world. I. Atlantic Ocean and adjacent Regions. {Medd. Komm.
Havunders0gelser, Serie: Fiskeri, Bd. Ill, Nr. 7, pp. 1-45. Kpbenhavn.
1912. Danish researches in the Atlantic and Mediterranean on the life-
history of the Fresh-water Eel (Anguilla vulgaris Turt.). ( Inter-
nationale Revue der gesamten Hydrobiologie und Hydrographie, V.
pp. 317-342, Leipzig).
1913. On the Identification of Muraenoid Larvae in their early (“Prelepto-
cephaline”) stages. {Medd. Komm. Havunders0gelser, Serie: Fiskeri,
Bd. IV, pp. 1-13. Kpbenhavn).
1913. First Report on Eel investigations 1913. {Rapports et Proces-V erbaux
du Conseil Intern, p. VExplorat. de la Mer, Vol. 18, pp. 1-30, Kpben-
havn) .
1915. Second report on Eel investigation 1915. {Rapports et Proces-V erbaux
du Conseil Intern, p. VExplorat. de la Mer, 23, 1. pp. 1-24, Kpbenhavn) .
1916. On the early larval stages of the Fresh-water Eel (Anguilla) and some
other North Atlantic Muraenoids. {Medd. Komm. Havunders0gelser,
Serie: Fiskeri, Bd. V, Nr. 4. pp. 1-20, Kpbenhavn).
1922. The breeding places of the Eel. {Phil. Trans. Royal. Soc. London,
B. Vol. 211, 385, pp. 179-208, London).
1924. The transatlantic migration of the Eel-larvae. Nature. Jan. 5, 1924.
1925. The breeding places of the Eel. {Smithsonian Report for 1924, pp.
279-316, Washington).
1931. Eels and Conger Eels of the North Atlantic. {Nature, Vol. 128, pp.
602-604. London).
1935. Danish Eel Investigations during 25 Years, 1905-1930. Published by
the Carlsberg Foundation, Copenhagen.
Jleto Horfe 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 |3arb
W. Reid Blair, Director
Raymond L. Ditmars, Curator of Mammals and Reptiles
Lee S. Crandall, Curator of Birds
Charles R. Schroeder, Veterinarian
Claude 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
Squarium
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 Croptcal &esearcf)
William Beebe, Director and Honorary Curator of Birds
John Tee-Van, General Associate
Gloria Hollister, Research Associate
Jocelyn Crane, Technical Associate
Cbttorial Committee
Fairfield Osborn, Chairman
Charles M. Breder, Jr.
William Bridges
W. Reid Blair
William Beebe
ZOOLOGICA
SCIENTIFIC CONTRIBUTIONS
OF THE
NEW YORK ZOOLOGICAL SOCIETY
1/
VOLUME XXIII
Part 4
Numbers 17 - 21
PUBLISHED BY THE SOCIETY
THE ZOOLOGICAL PARK, NEW YORK
December 31, 1938.
CONTENTS
PAGE
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
Gregory & Conrad: Phylogeny of the Characin Fishes
319
17.
The Phylogeny of the Characin Fishes.
William K. Gregory & G. Miles Conrad
The American Museum of Natural History
(Text-figures 1-37).
Table of Contents.
Page
Introduction 319
Phylogenetic Review of Principal Groups 325
Cheirodontinae 325
Serrasalmoninae 325
Characinae 333
Sarcodacinae 338
Erythrininae 343
Hemiodontinae 344
Anostomatinae 347
Citharininae 349
Distichodontinae 352
Gymnotidae 353
The Characins and the Supposed South American- African Bridge 356
Bibliography 356
Abbreviations Used in the Illustrations 359
Introduction.
In the systematic history of the family Characinidae, among the greatest
contributors were Muller and Troschel (1844, +), Sagemehl (1884, + ),
Boulenger (1887, +), Eigenmann (1899, +), Rowntree (1903, 1906), and
Regan (1911, +). In spite of all of these labors the broad evolutionary
trends were so easily obscured by the details that it was only in 1917, with
the publication by Eigenmann of the first part of his monograph on “The
American Characidae,” that the first attempt to outline the phylogenetic
relationships of all of the South American subfamilies was made (Text-fig.
5). Unfortunately this outline from which we have made a phylogenetic
diagram was never completely developed, for Dr. Eigenmann died before
the completion of his memoir on the Tetragonopterinae and allied sub-
families.
Rowntree (1903) examined the visceral anatomy of the characins and
made several contributions of phylogenetic interest, including evidence
against the idea of an amioid ancestry of the Erythrininae which had been
suggested by Sagemehl. Much of the data concerning the visceral anatomy
merely shows the differences in adaptation to an herbivorous diet on the one
320
Zoologica: New York Zoological Society
[XXIII :17
hand and a carnivorous diet on the other. Thus the differences and sim-
ilarities noted in the stomach and intestine have little phylogenetic signifi-
cance of subfamily rank, inasmuch as the herbivorous condition has appar-
ently been developed independently several times within the family.
In 1911 Regan revised the classification of the Characinidae and in so
doing synthesized the groups of previous authors to a more workable system.
However, in many respects the classification of Boulenger in the Cambridge
Natural History is still useful.
Cockerell (1912, 1913) studied the scales of both the African and South
American characins. His studies were very detailed and in some cases his
results are confusing when compared with other data, but on the whole they
served to clarify the phylogenetic relationships within certain groups.
Text-figure 1.
The phylogenetic relationships of the subfamilies of the Characidae.
The fossil records of this group are very meagre. From the Tertiary of
Brazil and Peru three more or less doubtful genera are known : Lignobrycon 1
Eigenmann and Myers 1929, Eobrycon2 Jordan 1907, Characilepis Cockerell
1920. These are apparently related to the Characinae. The teeth known as
Onchosaurus Gervais 1852^ which are found in the Upper Cretaceous of
North America, Europe and Egypt, have been shown by Eastman (1917)
to resemble the teeth of Hydrocyon and Hoplias. However, neither in gen-
eral body form nor in tooth structure are any of these genera primitive and
we must, therefore, with Eigenmann, fall back upon an analysis of the gen-
eralized conditions among living characins as the principal basis for a tenta-
tive reconstruction of phylogenetic history.
As a basis for our chart (Text-fig. 1) of the phylogenetic relationships
of the family Characinidae we have used first the monumental work, “The
American Characidae,” by C. H. Eigenmann, as well as many lesser papers
by the same author. Nor have we neglected the important systematic and
distributional studies of Regan, Boulenger and others. Our own material
includes a series of skeletal and preserved specimens representing the vari-
1 Based on Tetragonopterus ligniticus Woodward, Catalogue of Fossil Fishes in the British
Museum (Natural History), 1901, pt. 4, p. 298, pi. 17, figs. 2, 3.
2 Based on Tetragonopterus avus Woodward, ibid., p. 298, pi. 17, fig. 1.
PICTORIAL CLASSIFICATION OF
THE CHARACINS
CHEIRODONTINAE
SEKRA5ALM0NINAE
CHARACINAE
SARCODACINAE
AN 0ST0MAT1NAE
DI5TICH0D0NT1NAE
2WEKIC1W
AM.ER.1C2W
AMERICAN"
AMERICAN"
AFRICAN"
Cheirodon
Serrasalmo
Brqcoa
Chilodus
Distichodus
AFRICAN"
amr, T7
jjjh
" • -
Sarcodaces
Aaostoraus
Paraphiago
MoerShausia
Mijlefes
«sAc_
AMERICAN"
Procmlodus
Ichthqoborus
Tetragonopterus
Stethaprioa
Corijnopoma
Acestrorhqachus
Luc ocharcix
Curimatus
Pha^o
^r'!'^Chalciaus
Cqnodon
CITHARININAE
JA
AFRICAN"
iSP "
Gasteropelecus
ERYTHRININAE
Starks iaa
AFRICAN
AMERICAN"
Citharinus
jg§] „
HEMIODONTINAE
Alestes
AMERICAN"
.
Metqnnis
Hqdrocqoa
Lebiasiaa
Text-figure 2.
A pictorial classification of the characin fishes.
PICTORIAL AC I NS
CHEIRODONTTNAE
SERRASAISTICHODONTINAE
Tetraq,onopterus
ahekica.it
Cheirodon.
AFRICAN
1938] Gregory & Conrad: Phylogeny of the Characin Fishes
321
ous subfamilies; especially those brought back by Messrs. Lang and Chapin
of the American Museum Congo Expedition of 1909-1915, and the Brazilian
collections made by B. A. Krukoff in 1934 and 1935.
As an aid to the construction of a tentative phylogenetic chart we have
arranged a “Pictorial Classification of the Characins” (Text-fig. 2). This is
based on a phylogenetic interpretation of the characters used in defining
families, subfamilies, etc., by various authors. In view of the remarkable
uniformity in basic morphology of all characins and of the existence of more
or less annectant genera between the so-called families of earlier authors
we have treated the entire series as a single family, Characinidae, coordinate
in rank with the Gymnotidae, the latter being an extremely specialized and
presumably early side branch.
To supplement the chart showing the inferred phylogenetic relationships
of the general body forms we have added several other charts laid out on the
same plan. First of these is a series of scales as figured by Cockerell (1912,
1913). Although Cockerell figured the circuli and radii of the scales, we
have contented ourselves with the outlines (Text-fig. 3) which, by them-
selves, seem to present the several broad divergent trends within the family.
A series of diagrams of characin livers has been compiled from Rown-
tree’s data. These drawings (Text-fig. 4) are quite schematic and are merely
diagrams from verbal descriptions, but they afford an interesting contrast
to the more or less stable characters relied on in the above charts. Whether
the viscera, always in a more or less plastic state, have any really helpful
significance in these studies is open to question except in certain cases, but
it is noteworthy that in several places interrelationships are evident.
For convenience we insert at this point a summary of the subfamilies of
the characins, as used in the present paper.
1. Cheirodontinae (Cheirodontinae and Tetragonopterinae of Eigenmann).
Primitive, generalized characins; both maxilla and premaxilla bear-
ing teeth; marginal teeth varying from pluricuspid to conical;
mouth moderate to small. American: Grundulus, Mixobrycon,
Moenkhausia (Text-figs. 2, 8), Tetragonopterus (Text-figs. 2, 7, 8),
Cheirodon (Text-figs. 2, 6, 8), etc.
2. Serrasalmoninae (Stethaprioninae, Serrasalminae and Mylinae of Eigen-
mann). Extremely deep bodied; short, heavy jawed offshoots of
Tetragonopterus. Nicely graded series of constantly increasing
ratios of depth to length. Teeth grade from carnassial to molari-
form. American: Stethaprion (Text-fig. 2), Mylesinus, Serrasal-
mo (Text-fig. 2), Mylosoma (Text-figs. 7B, 8), Metynnis (Text-
fig. 2), etc.
3. Characinae (Characinae (in part), Hydrocyoninae and Gasteropelecidae
of Regan). The large central group of African and South Ameri-
can forms arising from the Cheirodontinae. Varying in body-form,
but central type as in Brycon; hypocoracoids usually forming
prominent vertical laminae, but grading into forms with short
median laminae and divergent coracoid fossae ( Cynopotamus ) ; lat-
eral line decurved ; no teeth on palatines. American : Charax, Bry-
con (Text-figs. 2, 9, 11, 12, 14, 16, 30), Iguanodectes, Diapoma
(Text-fig. 2), Corynopoma (Text-fig. 2), Chalcinus (Text-figs. 2, 13,
14), Gasteropelecus (Text-figs. 2, 13, 14). African: Alestes (Text-
figs. 2, 10, 15), Hydrocyon (Text-figs. 2, 11, 12, 17, 18, 19, 20,
21, 29).
4. Sarcodacinae (Sarcodacinae, Acestrorhamphinae, Cynodontinae and
Xiphostomatidae of Regan). Predatory “pikes” derived from the
Characinae. Jaws more or less elongate with caniniform teeth; a
large prefrontal plus supraorbital bone; supraoccipital small, not
elevated above flattened skull top; no interfrontal fontanelle. Ameri-
322
Zoologica: New York Zoological Society
[XXIII :17
Text-figure 3.
Chief variants in the form of characin scales. Based on data from Cockerell
(1912, 1913).
1938] Gregory & Conrad: Phylogeny of the Characin Fishes
323
Text-figure 4.
Chief variants in the form of the liver. Based on Rowntree’s (1903) data.
324 Zoologica: New York Zoological Society [XXIII :17
can: Luciocharax (Text-figs. 2, 23, 24), Acestrorhynchus (Text-fig.
2). African: Sarcodaces (Text-figs. 2, 23, 24).
5. Erythrininae (Erythrininae and Lebiasininae of Regan). Amm-like
relatives of Sarcodaces with more or less broad, rounded heads.
Opposite hypocoracoid fossae diverging sharply, median laminae
short ( Lebiasina ) to absent; lateral line straight (lacking in Lebia-
sina). Teeth on palatines. American : Erythrinus (Text-figs. 2, 25,
26), Hoplias (Text-fig. 24), Lebiasina (Text-fig. 2).
6. Hemiodontinae (Hemiodontidae of Regan). Small fusiform to slender
offshoots of the Characinae with short anal fin, adipose typically
reduced or absent; very small to small subterminal to terminal
mouth; teeth uniserial, minute; “pterygoid movably articulated
with quadrate, narrowed posteriorly ending in a small condyle”
(Regan). American: Hemiodus, Poecilobrycon (Text-fig. 2), Nan-
nostomus, Parodon.
7. Anostomatinae (Anostomidae of Regan). Slender to deep-bodied herbi-
vorous American characins. In scale and jaw characters paralleling
some of the Citharininae and Hemiodontinae. “Pterygoid rather
broad posteriorly, overlapping the quadrate” (Regan). Anostomus
(Text-fig. 2), Prochilodus (Text-fig. 2), Chilodus (Text-fig. 2).
Curimatus (Text-figs. 2, 28), Leporinus (Text-figs. 28, 29, 30,
31, 32).
8. Citharininae (Citharininae, Hemistichodontinae and Xenocharacinae of
Regan). An African group grading in scale characters from cyc-
loid to ctenoid. Parallel in many characters to South American
Anostomatinae. Citharinus (Text-fig. 2), Xenocharax.
9. Distichodontinae (Distichodontinae and Ichthyoborinae of Regan). This
African group seems to be related to the Citharininae, but is
sharply separated in that the dentaries are movably articulated
with the articulars. Distichodus (Text-figs. 2, 33), Phago (Text-
figs. 2, 35), Ichthyoborus (Text-fig. 2), Mesoborus (Text-fig. 34),
Paraphago (Text-fig. 2).
-Curimatinae
- Elopomorphinae
- Hemiodontinae
- Prochilodinae
-dxiLod.Ln.ae
~ Anastomatinae
- Hydrocyrunae
~ Cy rwdorttinae
- Acestrorhamphinae
- Characinae
- Salm.ininae
- Lebiasininae
~ Piabuscininae
- Pyrrhulininae
- lyu anodectinae
~ Aryconmae
-Gasteropelicinae
-Chalciainae
Aiylinae Serrasalmoninae
I Gymnocharacinae Stethaprioninae
1 CAtoenkhausia)
Tetragon optennae
GLandulocaudinae
_ Diapominae
Text-figure 5.
The phylogeny of the South American characins, according to Eigenmann (1917,.
pp. 38, 39).
1938] Gregory & Conrad: Phylogeny of the Characin Fishes
325
Phylogenetic Review of Principal Groups.
Cheirodontinae.
Several authors, including one of us (Gregory, 1933), have considered
the Erythrininae to be the most primitive existing subfamily of the Char-
acinidae, but further study indicates that this subfamily is relatively high
in the phylogenetic series. After close study of Eigenmann (1916), we are
indeed forced to the conclusion that the Erythrininae are less primitive than
the Cheirodontinae.
Text-figure 6 represents the skeleton of Cheirodon. It closely resembles
the generalized cheirodont which Eigenmann has pictured for us in words.
It is a relatively deep-bodied type, with a long anal fin having its origin
under the last dorsal ray and reaching almost to the origin of the caudal.
The caudal is deeply forked. The mouth is terminal and relatively small.
The teeth of the generalized type are in a single series, rather few in num-
ber and with lateral notches, occurring along the edge of the premaxilla, at
the upper angle of the maxillary, and along the front and sides of the
lower jaw.
The Cheirodontinae, or some of them, also retain several other out-
standing generalized characters from which those of almost any of the
remaining subfamilies may have sprung. One of these is the extremely
undifferentiated cycloid scales, almost elliptical in shape, from which the
others have evidently developed. Another is the greatly variable tooth
form, which, as noted by Eigenmann, ranges from the simple conical type
of Grundulus, through the bicuspid teeth of Macropsobrycon, the tricuspids
of Aphyocharax, Megalamphodus, Parecbasis to the octacuspids of Cheir-
odon. Thus, as far as teeth are concerned, this group is well prepared to
account for all of the bizarre tooth-types encountered in the family, such
as the caniniform teeth of Hydrocyon, the molariform teeth of the Mylinae,
the incisor-like teeth of Leporinus and many others.
Eigenmann’s generalized cheirodont exhibits the well developed chain
of suborbital bones which are constantly cropping out in various genera of
the Characinidae and which apparently first led Sagemehl to the belief that
Erythrinus had amioid affinities (since wholly disproved by several authors).
An adipose fin is present in the generalized type of the Cheirodontinae as
well as in most other Characinidae. However, the cheirodont genera Grun-
dulus and Spintherobolus lack one, this indicating that the possession of an
adipose is in an unstable condition in even the more primitive characins.
Some of the Cheirodontinae also have the median fronto-parietal fontanelle
which is so frequent in other subfamilies.
Arising from the Cheirodontinae ( sensu strictu), then, are the several
diverging subfamilies. The Tetragonopterinae of Eigenmann are very near
to the primitive stock of the characins and may have been derived from the
Cheirodontinae or from a common stem. In either event Eigenmann points
out the significant similarity of the heavy teeth and cheek armor in Mixo-
brycon (Cheirodontinae) and that of Hyphessobrycon (Tetragonopterinae).
A glance at Moenkhausia (Text-fig. 2), a tetragonopterine, will show the
general similarity of its body form to that of Cheirodon pulcher. Tetra-
gonopterus proper (Text-fig. 2) is a greatly deepened form which through
some species such as T. argenteus has given rise ultimately to the extremely
deep-bodied and strangely modified Serrasalmoninae.
Serrasalmoninae.
We have seen above that T etrag onopterus argenteus, which we refer to
the Cheirodontinae, approaches very closely to the body form of the Steth-
aprioninae and differs from the latter chiefly in the lack of a “pre-dorsal
spine” insofar as external gross characters are concerned. Both Cockerell
326
Zoologica: New York Zoological Society
[XXIII :17
(1913) and Eigenmann and Myers (1929) noted the relationship between
Stethaprion and another cheirodont, Moenkhausia, but Eigenmann (1907)
states, “the membei's of the Stethaprioninae mark the direct road from the
genus Tetragonopterus in its narrowest sense to the Mylinae and Serrasal-
moninae. In the deep T. argenteus the post-ventral region is trenchant, the
pre-ventral region flat. In Stichonodon both pre-ventral and post-ventral
regions are trenchant. In Stethaprion, Brachychalcinus , and Fowlerina the
post-ventral region is incipiently serrate and there is a pre-dorsal spine.
1938] Gregory & Conrad: Phylogeny of the Characin Fishes
327
\~p/m
<scalebone
cupel t
supra orouc
postorbital canal a
orbital canal
pmr
occipital canal
scalebone
... fossa forlei/ator arcus palatini
v. fossa for dilatator ope r cut i
fossa for lerntor ope r cut i
-supclt
-preop ercular canal
ang
mandibular canal \
\ r
\ I
suborbifal canal
Tetra^oriopterus
Text-figure 7.
A, skull of Tetragonopterus, showing lateral line canals and principal muscle
fossae. B, Mylosoma, detail of occipital region, showing canals.
328
Zoological New York Zoological Society
[XXIII :17
In Mylesinus of the Mylinae the abdomen is serrate behind the ventrals and
in the rest of the Mylinae and Serrasalmoninae the ventral edge is serrate
both in front and behind the ventrals.” This series is supported again by
Eigenmann and Myers in 1929. On the entire chart of body forms (Text-
fig. 2) there is no series so convincing as this one, especially in the constantly
increasing ratio of depth to length until finally, in Metynnis, the depth
almost equals the length. There is a transition from the sharp, shearing,
carnassial-like teeth of Serrasalmo to the more blunt, molariform teeth of
Myletes, while the jaws in both remain short and blocky. In liver form
(Text-fig. 4) Serrasalmo appears to be very generalized.
The skulls of the Cheirodontinae and the Serrasalmoninae are often
quite small and paper thin so that in order to facilitate the identification of
the bones a detailed study of the sensory canals was made. Text-figure 7,
an outline picture of the skull of the typical genus, Tetragonopterus, shows
the position and names of the canals and the diagnostic muscle fossae.
The lateral line passes from the body into the posttemporal bone and
from there it continues into the “scalebone.” In the scalebone it forks into
two branches, one leading toward the postorbital canal, the other, the occipi-
tal canal, passing dorsally through the parietal bone. Just above the scale-
bone the occipital canal is met by a branch from the supraorbital canal.
At the most ventral point of the surface of the pterotic the preopercular
canal arises. This preopercular canal runs directly into the mandible.
Starting at the origin of the preopercular canal and bordering the fossa
of the dilatator operculi is the postorbital canal which continues forward to
the junction of the suborbital canal with the main trunk. The suborbital
canal joins the main canal at the “peninsula” which marks the separation
of the fossa for the levator arcus palatini from that of the dilatator operculi.
From this junction forward the sensory canal is known as the supraorbital
canal and gives off several branches as it passes through the frontal bone.
This canal passes through the nasal bone and, in Tetragonopterus at least,
seems to continue into the premaxilla.
The nicely graded series of body forms which has been noted in the
cheirodont-serrasalmonine branch is reflected in the skull structures. Cheir-
odon (Text-fig. 8A) has the skull of a more or less fusiform habitus. The
dorsal curvature of the skull is slight and convex and there are sizable
parietals, which might imply a lack of the forward curling of the occiput
that is seen as we proceed up the scale. The dermosphenotic is present in
this primitive form. The jaws are relatively compact and strong and are
not elongate.
Moenkhausia (Text-fig. 8B) reflects the slightly deeper body and the
supraoccipital crest is a bit more steep than in Cheirodon. Indeed the entire
skull shows a dorso-ventral emphasis. The dermosphenotic disappears in
Moenkhausia and is lost in those latter specimens of this series which we
have examined. The parietal and pterotics have been reduced.
The skull of Tetragonopterus (Text-fig. 8C) resembles very closely that
of Moenkhausia. As might be inferred from the progressively increasing
body-depth, the supraoccipital bone is so upturned that the frontal is quite
concave in side view. The suture between the frontals and the parietals and
pterotics is in the form of two acute angled, overlapping bevels. Thus the
surface extent of the parietals is a great deal less than the cranial. This
suggests a mechanical shoving forward of the temporal region. The post-
orbital portion of the circumorbital chain has been dissected away so as to
show the various muscle fossae as well as the postorbital process of the
sphenotic bone.
As a representative of the Serrasalmoninae, Mylosoma (Text-fig. 8D)
serves admirably. Here is the ultimate in the deep-bodied forms in some
of which the depth almost equals the length. It is no wonder, therefore,
that the fronto-parieto-supraoccipital contour in side view is so concavo-
1938] Gregory & Conrad: Phylogeny of the Characin Fishes 329
Text-figure 9.
Skeleton of Brycon.
332
Zoological New York Zoological Society
[XXIII :17
convex as to form a feeble S. In Tetragonopterus a perpendicular line pass-
ing through the posterior tip of the supraoccipital would also pass close to
the rear margin of the operculum, while in Mylosoma such a line would
pass through the center of the postorbital portion of the circumorbital
chain. For the first time in this series a new bone appears, the supra-
orbital. The sensory canal system retains the basic plan of Text-fig. 7A,
but in response to the necessity of the openings passing through the thick
layer of fatty tissue (which cushions the concave part of the skull) there
have been developed several bony “craters” which carry the canals to the
surface. The word, crater, is really descriptive of them for they look much
like a volcanic crater — an eruption of the bone. As another innovation the
occipital canal passes behind the crest, which is usually placed at the pos-
terior edge of the parietal. However, in Mylosoma this crest has moved
forward on the parietal, while the canal presumably maintains its primitive
position (see inset, Text-fig. 7B).
The most salient feature of this phylogenetic branch is the constant
anterior movement of the occipital region while the rest of the skull remains
stationary. The loss of the dermosphenotic and the appearance of the
supraorbital are observed. The fontanelles are left intact throughout the
group.
carotid for
■Autogenous
%parapophysis
pt+poreth
trigemiho-faciatis Tio/c/i carotid for
Brycon
i pupra occipital buttress
\ - ? neural arch or
■ ’wcrsp/natbone
, neural archVZ
pto p ' neural arch \ Iff
’tyj transv. pr 142
' tfansv prV3
' transvprVb
fir/en Z'i/ra de£
Hydrocyon
boc\ j tripus-autogparapoph V3
ogot \inbcrcalarium=l autog.parapopb
scaphium^neurai arch VZ
r VI
Text-figure 11.
Neurocrania of the African Hydrocyon and the South American Brycon. The
first four vertebrae with their attendant Weberian apparatus are included
with the skull of Hydrocyon. Although the Weberian ossicles and anterior
vertebrae have not been added to the skull of Brycon, a close similarity in
basic plan of the neurocrania is noted in these two members of the Chara-
cinae.
1938] Gregory & Conrad: Phytogeny of the Characin Fishes
333
Characinae.
This group is the most extensive of the family, abounding as it does
in both South American and African forms which range from the primitive
African genus, Alestes, to the South American fresh-water flying fish, Gas-
teropelecus. Although the interrelationships of the group are obscure and
are in the sense used by Eigenmann (1917) probably polyphyletic, there is
little doubt that they are all traceable to the completely primitive Cheir-
odontinae. The Characinae include on our chart (Text-fig. 2) as representa-
tives of the group, the following genera:
(A) American
Brycon
Diapoma
Corynopoma
Chalcinus
Gasteropelecus
SdC
inter-fronta!
- trans. groove can ■
sm.impan
He /e-ris'Z /s/ca.
Hydrocyon
prodtic / [
/ en trance to cran. recess
trans. groove ran.s/n . forsaccutus
/. rnpar .
Text-figure 12.
Neurocrania of Hydrocyon and Brycon, longitudinal sections. These sections show
even more strikingly the basic similarity of the African and South American
genera of Characinae. The heavy black lines in Hydrocyon indicate the posi-
tion of the semicircular canals.
334
Zoologica: New York Zoological Society
[XXIII :17
Text-figure 13 (upper).
Cleithrum and primary pectoral arch (except radials) of A. Gasteropelecus pec-
torosus and B. Chalcinus trachypomus. After Regan, 1911.
Text-figure 14 (lower) .
Skeletons of Gasteropelecus, Chalcinus and Brycon.
1938] Gregory & Conrad: Phytogeny of the Characin Fishes
335
(B) African
Alestes
Hydrocyon
This group may be split into several series, all of which seem to be
ultimately derivable from cheirodont beginnings. The African members are
closely related to their South American allies in the fundamental pattern
of skull structure (Text-figs. 11, 12) as well as in many curious details of
body form, scalation, etc.
According to Eigenmann the Glandulocaudinae, here represented by
Diapoma and Corynopoma (Text-fig. 2), are linked quite definitely to the
Cheirodontinae by Paragoniates, with its general shape and backward posi-
tion of the dorsal fin, and by Compsura and Odontostilbe, with the peculiar
caudal scalation of the males.
Brycon (Text-figs. 2, 9) seems to be the central type of the group and
from it, according to Eigenmann (1917), arose one line leading to Iguano-
dectes and Pyrrhulina. Eigenmann and Myers (1929) describe the sub-
family Iguanodectinae as “slender, elongate, moderately compressed, smelt-
like fishes of small size . . .” Cockerell (1913) remarks that the iguano-
dectine scales are not far from the condition noted in Cheirodon. Pyrrhulina
is apparently near to the point at which the hemiodontines branched from
the cheirodont stem.
Chalcinus (Text-fig. 2), in its deepened coracoids and large pectoral
fins, seems to afford a favorable point of departure for Gasteropelecus
(Text-fig. 2). Regan (1911, p. 20) rejected this connection and stressed
the structural relationships with Tetragonopterus. But the latter differs
widely from Chalcinus (Text-fig. 14B) in the low position of the pectorals,
general form of body and mouth, and undoubtedly points rather to Serra-
salmo. After comparing the skeletons of Chalcinus and Tetragonopterus
with the skeleton of Gasteropelecus (Text-fig. 14A), we incline to the opin-
ion that Chalcinus is much the nearer to the structural ancestor of Gas-
teropelecus.
Alestes (Text-figs. 2, 10), a relatively primitive predator, is typical of
several African members of the Characinae, such as Bryconaethiops, Micra-
lestes, and Petersius, which vary from fusiform to fairly deep-bodied shapes.
The teeth vary from very small pluricuspid to large compressed shearing
types; some even have molar-like teeth on the inner row of the upper jaw.
That the African genus Alestes is surely close to the American Brycon
is undoubted. The general shape of the skulls (Text-figs. 15, 16) is very
much alike. They both possess the supraorbital, but whereas Brycon retains
the dermosphenotic, it is noteworthy that in Alestes the frontal is excluded
from the orbit by a prolongation of the suborbitals. However, in Alestes
the dermosphenotic may be so completely fused with the suborbitals as to
be indistinguishable from them. At any event, the postorbital portion of
the circumorbital chain serves the purpose of protecting the muscle fossae
noted in Text-fig. 8, as does the dermosphenotic when present. In both
genera the lacrymal is present (not shown in Text-fig. 14) and well sep-
arated from the parethmoid, as in primitive characins.
Hydrocyon (Text-figs. 17, 18) is an extreme predaceous, pike-like off-
shoot of this African stock. Regan (1911) separates it as a distinct sub-
family from the rest of his Characidae largely on the basis of the movable
premaxillae, but this character is developed independently in other groups
of characins (e.g., Anostomatinae, Hemiodontinae, Citharininae and Dis-
tichodontinae) and should not outweigh the many marks of close kinship
with other African Characinae. If we are to grade taxonomic rank accord-
ing to intensity of specialization then the complex symphysial hinge-joint of
Hydrocyon (Text-figs. 19, 20, 21) should entitle it to the grade perhaps of
a superfamily. However, other characins (Text-fig. 22) and even its own
336
Zoologica: New York Zoological Society
[XXIII : 17
Text-figure 15.
Skull of Alestes.
young clearly point the way toward the complex hinge-joint of Hydrocyon.
The complex symphysial hinge-joint of Hydrocyon and other characins
is described in more detail by Gregory and Conrad (1936) as follows:
“Thus the story of the development of the complex hinge-joint in Hydro-
cyon lineatus starts at a stage when the ‘hinge’ is nothing more than an
interdigitating articulation with but three knuckles to each dentary [Text-
fig. 21, I.]. The dorsal border of the dentary then produces another knuckle,
C, on the right and b on the left. The ventral border of the left dentary
gives rise to / [Text-fig. 21, II.]. The next stage [Text-fig. 21, III.] sees
the first advance toward the adult interlocked hinge for, as pointed out
above, the knuckles begin to curl from the posterior to the anterior borders
of the dentary, leaving an opening or fenestra at the pivotal point. By
inserting here an hypothetical phase in which this forward curling is wholly
finished and the knuckles are interdigitated, it can be seen that by the coales-
cence of the knuckles arising from the same dentary the condition in IV.
[Text-fig. 21] would occur. Now all the elements are present [Text-fig. 20,
V. ; Text-fig. 21, IV.] and the left hinge-stop is completely formed, while the
right one is in process of coalescence. Finally, in VI. [Text-fig. 20] the hinge-
joint reaches its completely formed complex condition, which will with
later growth be pulled out and distorted both vertically and horizontally,
while at the same time retaining its mechanical efficiency.
“A glance at [Text-fig. 22] shows that the most common condition of
the hinge in Characins is that seen in Moenkhausia ( Tetragonopterinae
Eigenmann), Piaractus ( Mylinae Eigenmann), and Alestes ( Characinae
Regan). In all of these the knuckles are arranged in planes radiating pos-
teriorly from the anterior border of the dentary. Our detailed studies on
the family suggest that Hydrocyon has evidently arisen from an Alestes- like
ancestor, and a comparison between the Alestes hinge in [Text-fig. 22] and
that of Hydrocyon in [Text-fig. 21] (II.) shows that the latter passes
1938J Gregory & Conrad: Phylogeny of the Characin Fishes
337
Brycon dentex
Text-figure 16.
Skull of Brycon. After Gregory, 1933, fig. 70.
through an Alestes stage in development. Returning to [Text-fig. 22] we
notice that the primitive hinge condition is upset in Erythrinus and Hoplias
which have moved the ‘area of radiation’ posteriorly, so that the knuckles
radiate both anteriorly and posteriorly. The condition in Erythrinus may
possibly be that of Hydrocyon in [Text-fig. 21] (IV.) just before any of
the knuckles have coalesced.
“In conclusion, the complex hinge-joint in the symphysis mandibulae of
Hydrocyon lineatus Bleeker develops its most complex features, the so-called
hinge-stops described above, after the post-larval and infantile stages have
been completed.
“The hinge is formed from interdigitating processes analogous to the
knuckles of a mechanical hinge. These knuckles represent subdivisions of
the subalveolar ridge of the dentary bone. They are typical polyisomeres in
the sense defined by one of us (Gregory, 1934). They grow inward toward
the mid-plane like fingers, decussate across the mid-plane, and curl around
and join themselves into the hinge-stops on either side of the mid-plane.
Here they behave like typical anisomeres, which arise by the differential
growth and fusion of polyisomeres.
338
Zoologica: New York Zoological Society
[XXIII :17
Text-figure 17.
Skull of Hydrocyon. After Gregory, 1933, fig. 69.
“The earlier structural stages of the condition in Hydrocyon are found
in Alestes and other primitive Characins and in an arrested stage in the
subfamily Erythrininae.”
Sarcodacinae.
Acestrorhyncus, with its comparatively long jaws and with its pre-
maxillaries fixed and firmly attached to the mesethmoid, has apparently
given rise to the elongate, pike-like Xiphostomatidae of Regan (represented
on our chart (Text-fig. 2) by Acestrorhyncus and Luciocharax) .
The similarity of the skulls of Sarcodaces (Text-figs. 23, 24), an Afri-
can form, and Luciocharax, a South American, is amazing. Text-figure 23
shows the dorsal views of these skulls. It is seen that the dorsal fontanelle
is completely closed over in both and that a triangle is formed by the
dermosphenotics, frontals and pterotics. This wedge-like arrangement of
the cranial elements is highly suggestive of relationship, coupled as it is
with a well developed prefrontal plus supraorbital which is so rare among
the other characins examined. The scale bone is not figured in Luciocharax.
There is no supraopercular bone in Luciocharax, but this is a derm bone
which might easily be lost. As seen in side view (Text-fig. 24) the two
skulls are likewise comparable, with their plate-like lacrymals which are so
different from those of the more primitive characins. The prefrontals lie
dorsal to the lacrymals instead of behind them as in more generalized
forms. The suborbital bones of both are very similar in cut, the bones
marked so2 being strikingly alike. There is a persistence of the dermosphen-
1938] Gregory & Conrad: Phylogeny of the Characin Fishes
339
Hydrocyon
Text-figure 18.
Skull of Hydrocyon, dorsal view.
otics of the primitive characin. The cut of the operculum is quite similar
in both. The premaxilla and dentary of Luciocharax have been elongated
and the maxilla is somewhat reduced. That the increased number of teeth
in Luciocharax is secondary is highly probable.
The similarity of these two is so great that it seems warranted to refer
Sarcodaces to the Sarcodacinae (which includes Luciocharax and others)
as here understood (page 321). The connection of Sarcodaces with Alestes
and allied genera seems much more remote than with Luciocharax. In
Sarcodaces the lacrymal and prefrontal overlap, as they do in Luciocharax,
whereas in Alestes which has the typical characin lacrymal placement, they
are well separated from the parethmoid-prefrontal complex. A supraorbital
is present in Sarcodaces and Luciocharax and is also present in Alestes.
Alestes lacks the dermosphenotic that is present in the others. The general
heavily armored character and the telescoped occipital region common to
Sarcodaces and Luciocharax is lacking in Alestes and the Characinae gen-
erally. The large, well developed circumorbitals of Sarcodaces are noted in
Hydrocyon, but the lacrymal and parethmoid are well separated in the latter.
Hydrocyon has a well developed supraorbital but at the same time retains
the dermosphenotics of the primitive Cheirodon. The supraopercular found
in some few characins, such as Sarcodaces, is lacking in Hydrocyon. The
340
Zoologica: New York Zoological Society
[XXIII :17
Lower jaw of Hydrocyon lineatus, showing the dentaries spread to greatest extent
in A and closed as tightly as possible in B. After Gregory and Conrad,
1936, fig. 1.
supraoccipital is produced sufficiently far back to include Hydrocyon among
the more central Characinae.
Luciocharax, while it retains most of the osteological characters of the
Characidae of Regan, has developed a somewhat movable upper jaw and the
maxillaries have become firmly united with the premaxillaries.
The scales of the Acestrorhamphinae (our Sarcodacinae) according to
Cockerell (1913) are like those of the Serrasalmoninae. However, this prob-
1938] Gregory & Conrad: Phylogeny of the Characin Fishes
341
Text-figure 20.
A, distal portion of left half of mandible of adult Hydrocyon, oblique mesial view
of S turned slightly. VI, median sagittal section of same specimen. V, section
of a much smaller specimen, measuring 23 cm. (Stippled portion represents
right dentary; unstippled, left.) After Gregory and Conrad, 1936, fig. 6.
ably indicates only a community of origin for the two groups have developed
into as many diverging types. While the Serrasalmoninae have become in-
creasingly deep-bodied as noted above, the sarcodacines have been pulled out
antero-posteriorly to form, in Luciocharax, an astonishingly pike-like fish.
Thus the Sarcodacinae like the Serrasalmoninae may be traced back to the
Cheirodontinae.
342
Zoologica: New York Zoological Society
[XXIII :17
II
Text-figure 21 (left).
Median sagittal sections of symphysis in Hydrocyon lineatus. IV, a specimen
about 21 cm. long; III, a fish about 5.5 cm. in length; II, a specimen 5 cm.
long; and I, about 2.5 cm. long. After Gregory and Conrad, 1936, fig. 7.
Text-figure 22 (right).
Diagrams of disarticulated symphysial hinges in various characins; mesial views.
The hatched portion represents the cavities and the light, the knuckles. The
dotted lines are merely an aid in noting the directions of the knuckles.
L, left dentary; R, right dentary; ant., anterior border of the dentary. After
Gregory and Conrad, 1936, fig. 5.
1938] Gregory & Conrad: Phylogeny of the Characin Fishes
343
Erythrininae.
As noted above the primitive appearance of the Erythrininae is ap-
parently only secondary. In a study of the visceral anatomy of the chara-
cins Rowntree (1903) notices the striking resemblance between the Eryth-
rininae and Sarcodaces, “not only in cranial characters, as shown by Sage-
mehl, but also in certain visceral characters, notably in the opening of the
ductus pneumaticus far to the left on the alimentary canal, in the character
of the ovaries, and in the features of the air-bladder.” This asymmetric
position of the ductus pneumaticus is, according to Rowntree, a specialized
condition inasmuch as a symmetric or mid-dorsal position of the duct is the
primitive one. That Rowntree and Sagemehl noted several points common
to the Erythrininae and Sarcodaces is an interesting observation for, of all
the skulls studied, the only ones forming possible links between the Eryth-
rininae and the remainder of the Characinidae are those of Sarcodaces and
related genera.
Rowntree further notes that Macrodon ( Hoplias ), alone among the
Characinidae, possesses but two of the usual three liver lobes. Insofar as
scales are concerned Cockerell (1913) shows that the Erythrininae combine
characters of the characins and cyprinids.
Studies on the symphysial hinge-joints in characins seemed to indicate
at first that the condition noted in Erythrinus (Text-fig. 22) was a precursor
of that found in Hydrocyon. Upon further consideration (see above, page
337), however, it is more probable that the Erythrinus condition of the
symphysial hinge-joint is merely that of Hydrocyon in an arrested stage.
In any event it becomes evident that the Erythrininae are well along
in the evolutionary series, but in spite of their amazing similarity to the
cyprinids in the scales it is doubtful that they are near to the stem leading
from the primitive Ostariophysi to that group. The well developed upper
and lower jaws are armed with conical, canine-like teeth much as in Sar-
codaces (Text-fig. 24). The condition of their jaws and teeth and the lack
of an adipose fin are early noted in the cheirodont Grundulus, and the Eryth-
rinus body-form has been developed time and again within the family
Characinidae.
The heavily armored, tightly built skull of the Erythrininae (Text-figs.
24B, 25, 26) closely resembles those of the Sarcodacinae. Hoplias (Erythrin-
inae) (Text-fig. 24B), with its elongate dermosphenotic and separate supra-
operculum, seems to form a connecting link between Sarcodaces and Eryth-
rinus. The lacrymal and the pre-fronto-parethmoid articulate in this sub-
family and are not separated from each other as they are in more primitive
characins. However, the Erythrininae resemble the primitive subfamilies
in not having the frontal excluded from the orbit, whereas in Sarcodaces and
Luciocharax the enlarged prefrontal plus supraorbital do exclude the frontal
from the orbital margin. Both Erythrinus and Sarcodaces have large nasals.
The interfrontal sutures and the sutures between the parietals are very
similar in the two, but Erythrinus lacks the characteristic triangular
dorsal view in Sarcodaces. The fontanelles are entirely covered over in both
genera. The dentition of Hoplias is quite like that of Sarcodaces. The cut
of the lower border of the operculum in the Erythrininae and Sarcodaces
is peculiar, but this is approached also in Curimatus (Anostomatinae) .
Lebiasina and Piabucina (Text-fig. 26), referred by Regan to his Char-
acidae (sensu strictu), show a close approximation to Erythrinus not only
in the body-form but even in the skull. The fronto-parietal fontanelle is
completely closed over; the operculum is very similar; the lacrymal-pareth-
moid relationship is as in the Erythrininae; and there is a sizable dermo-
sphenotic. The only specimen available does not seem to have a supra-
opercular. The dorsal aspect affords a favorable comparison with Eryth-
rinus. The coracoids of Lebiasina, according to Regan, have the median
344
Zoologica: New York Zoological Society
[XXIII :17
Text-figure 23.
Skulls of Luciocharax and Sarcodaces, dorsal views.
ventral laminae present but short whereas in the Erythrininae generally
these are absent.
Thus the Erythrininae may easily have been derived either from a
cheirodont near to the Alestes-Hydrocyon stem or from the immediate an-
cestors of the Sarcodacinae.
Hemiodontinae.
Regan (1911) points out that with the exception of certain diagnostic
skull characters, the osteological characters of the Hemiodontinae are “essen-
tially similar to the Characidae.” As noted above (p. 321) his “Characidae”
has been redefined in part as the Characinae. Cockerell (1913) notes that
1938] Gregory & Conrad: Phylogeny of the Characin Fishes
345
Text-figure 24.
Skulls of Sarcodaces, Hoplias and Luciocharax, lateral views.
the scales of the “Pyrrhulinae,” which we refer to the Characinae, are very
similar to those of Nannostomus of the subfamily Hemiodontinae. On the
other hand Cockerell states that the scales of Hemiodus, Anisistsia (Hemio-
dontinae), and Anostomus (Anostomatinae) are of the curimatine type
(< Curimatus being referred by us to the Anostomatinae), while the scales
of Poecilobrycon (Hemiodontinae) and Nannostomus (Hemiodontinae) are
like those of Leporinus (Anostomatinae), indicating for the hemiodontines
a position close to the anostomatine series. However, because of the
slight affinities of Pyrrhulina and other members of the Characinae to
the Hemiodontinae we consider the subfamily Hemiodontinae as a side
346
Zoologica: New York Zoological Society
[XXIII :17
soe _P+rn ,.supcl
scale borte
pf & pereth
deth
qu' an X>'
Erythrinus unitaenlatus
Text-figure 25.
Skull of Erythrinus, lateral view. After Gregory, 1933, fig. 67.
Text-figure 26.
Skulls of Erythrinus (after Gregory, 1933, fig. 68) and Piabucina, dorsal views.
1938] Gregory & Conrad: Phytogeny of the Cliaracin Fishes 347
shoot of the Cheirodon-Curimatus line. Parodon (Hemiodontinae) with its
rather short body seems to be a primitive forerunner of the elongate, fusi-
form Poecilobrycon (Text-fig. 2). The Hemiodontinae and Prochilodinae of
Regan both have a single headed hyomandibular in contrast to the double
headed condition seen in the remainder of the family.
The subfamily resemblances between the scales of various genera as
observed by Cockerell (1913) may, according to our classification, be tab-
ulated as in Text-fig. 27.
CHEIRODONTINAE SERRASALMONINAE CHARAC1NAE HEMIODONTINAE ANOSTO MATI NAE
>*Cheirodon. <
^■Pheaacocjaster «
l Deuterodon.
K' Astqanax
Hemi^rammus
\ HijpKessobrijCori;
'^Pnstella
MoeakKausia
Aphuocharax
kjuanodectes
Ctenobrqcon.
CBPLjcoaam.ericiis
CreatocKaaes
Gasteropelecus
Carnegie lla
Charax
Petersius <
Chalciaus <
Chalceus
Pqrrhulina
Alestes
CHemiodus
Aaisistsia
Curimatus
• Hqdrolqcus
- Hqdrocqnus
- Accstrorhqachus
Nannosfomus
Poecilobrqcoa
CKaracidium.
ERYTHRININAE
Lebiasiaa
Erqthriaus
Hoplias
Aaostomus^.
Leporiaus ^
Text-figure 27.
Resemblances between the scales of various genera of characins as noted by
Cockerell, 1913.
Anostomatinae.
The Anostomatinae, a rather compact group, are represented in the
“Pictorial Classification” (Text-fig. 2) by Anostomus, Prochilodus, Chilodus
and Curimatus. The series starts with Anostomus, which possesses a small,
non-protractile mouth and has the rami of the lower jaw short and stout.
Leporinus is very close to Anostomus ; it is said by Cockerell (1913) to have
curimatoid shaped scales with a very definite alestoid relationship. Chilodus
also has scales of curimatoid shape with evidence of a part of the alestiform
pattern. Cockerell goes so far as to state the belief that this group supplies
the link between Distichodus and Alestes because their scales undoubtedly
illustrate the beginning of the development which culminates in the special-
ized ctenoid scales of the African Xenocharax. In Prochilodus we have at
iast a real development of the ctenoid scale, combined with an alestoid radial
pattern. It is apparent from scale studies that Prochilodus leads us toward
a separate offshoot of the curimatoids, possibly to the African distichodonts,
but certainly near and parallel to them.
The curimatoids proper are represented in our “Pictorial Classification”
by Curimatus. This toothless form constitutes the “end-genus” of this line
of divergence from the Cheirodontinae.
As a whole the subfamily Anostomatinae is highly specialized and far
from its original ancestor, a generalized cheirodont. These conclusions con-
348
Zoologica: New York Zoological Society
[XXIII :17
pfo
sPh°\
Sup- or 6
pf+parpfhjt
cleth
pcalebone
- pirn
■mx '
Curimalus
cerning the relationships of this group, based largely upon scale structure,
are in accord with Eigenmann’s systematic studies (1917).
The skulls of this group that we have been able to study are Curimatus
and Leporinus (Text-fig. 28). A supraorbital bone is present in Curimatus
and Prochilodus but is absent in Leporinus. The fontanelle in both Curi-
matus and Leporinus completely separates the frontals from the occiput
1938] Gregory & Conrad: Phytogeny of the Characin Fishes
349
Hydrocyon
to the dermethmoid; the lacrymal and parethmoid are well separated; the
supraoccipital is produced posteriorly; the maxilla is greatly reduced, the
small nibbling jaws retain strong teeth in Leporinus but are edentulous in
Curimatus and some others. The lower border of the operculum in Curi-
matus is truncate postero-inferiorly as in Sarcodaces. The opercular region
of the skull extends postero-ventrally in Curimatus but to a much lesser
degree in Leporinus.
Text-figure 29 shows the great similarity between the occipital regions
of the skulls of the African characine, Hydrocyon, and the South American
anostomatine, Leporinus.
The implied relationship pn the chart (Text-fig. 1) between the Anos-
tomatinae and the Citharininae is strengthened by the skulls. It may be,
however, that these resemblances are merely due to parallelism for both
subfamilies reflect their cheirodont ancestry.
The ribs of Leporinus and other genera, along with those of the
remainder of the Characinidae, are articulated to the centrum by auto-
genous parapophyses. Text-figures 30, 31 give front and side views of
vertebrae in the abdominal and caudal regions. Text-figure 32 shows the
arrangement of the hypurals in Leporinus.
Citharininae.
Regan (1911) refers the African Citharinus (Text-figs. 2, 3) and its
ally Citharidium (Text-fig. 3), along with many other African genera, to
350
Zoologica : New York Zoological Society
[XXIII :17
Text-figure 30.
A, C, side and front views of abdominal vertebra of Leporinus, showing rib and
autogenous parapophysis. B, autogenous parapophysis and rib of Brycon;
D, with autogenous parapophysis attached to centrum in Brycon.
Text-figure 31.
A, lateral view and B, front
view of caudal vertebra in
Leporinus.
the family Citharinidae. We, however, prefer to segregate Citharinus and
Citharidium and their near allies, N annaethiops , Neolebias, Xenocharax
(Text-fig. 3), Hemistichodus, in a more primitive subfamily, Citharininae,
1938] Gregory & Conrad: Phylogeny of the Characin Fishes
351
Text-figure 32.
Arrangement of hypurals in Leporinus.
characterized by the absence of the high specializations of the mouth parts
seen in the Distichodontinae, as here understood.
Boulenger in the Cambridge Natural History, places both Prochilodus
(Text-figs. 2, 3), a definitely ctenoid South American anostomatine, and
Curimatus (Text-figs. 2, 3) in the Citharininae; nor was this classification
wholly unwarranted, for his conclusions are supported by other anatomical
features, i.e., both have movable upper jaws, premaxillaries articulating on
the mesethmoid, maxillaries articulated with or adherent to the premaxil-
laries, palate toothless, lateral line straight, etc.
Cockerell (1912) separates the scales of the African subfamily Cithar-
ininae into the Citharidium type and the Citharinus type, the former with
ctenoid and the latter with cycloid scales. It is evident that Citharinus
with the more primitive cycloid scale is the forerunner of the ctenoid
Citharidium.
The skull of Citharinus compares rather well with that of the Anos-
tomatinae, especially Prochilodus or Curimatus (Text-fig. 28). The simi-
larity and extent of the fontanelle and the presence of supraorbital are note-
worthy. The jaws, too, are quite similar in both subfamilies.
In further support of an early connection between the Anostomatinae
and the Citharininae, Rowntree (1903) writes of an accessory branchial
organ arising as a blind sac from the upper margin of the last gill cleft;
this character seems to be peculiar to the herbivorous forms, Hemiodus
(Hemiodontinae), Prochilodus, Caenotropus, Curimatus (Anostomatinae),
and Citharinus, Xenocharax (Citharininae). Sagemehl identified this struc-
ture as an organ arising from the rudimentary fifth branchial arch. “If this
conclusion,” says Rowntree, “be correct the presence of the organ in the
herbivorous Characinids, and in these only, becomes of great interest, in
view of the fact that traces of a fifth gill have not been found in any living
ganoid, but only in fishes of yet lower organization — certain Selachians and
Dipnoids.”
Apparently, however, both Sagemehl and Rowntree failed to realize
that selachians and dipnoans have no connection with teleosts and that
the so-called fifth branchial arch has no definite claim to be homologized
with those of selachians and dipnoans, but is more probably a secondary
response to the presence of an accessory branchial organ.
352
Zoologica: New York Zoological Society
[XXIII :17
Rowntree then procedes to draw these conclusions, “(1) that the herbi-
vorous Characinids which possess it (the accessory branchial organ) form
a natural group; and (2) that this division of the family cannot be derived
from either the Erythrinoids or the other carnivorous Characinids, but is
at least as ancient as either of these groups.” His first conclusion, at least,
seems safe and serves as more positive evidence in showing the inter-
relationship of the anostomine-citharinine group.
Distichodontinae.
Regan’s subfamilies, Distichodontinae and Ichthyoborinae, are closely
linked by the possession of more or less massive dentaries, firmly united at
the symphysis and movably articulated with the articulars.
Text-figure 33.
Skull of Distichodus. After Gregory, 1933, fig. 71.
That these African forms were derived from the ctenoid African Cith-
arininae seems more likely than that they came from the ctenoid South
American anostomatines. Apparently drift toward the herbivorous dis-
tichodonts took place in or near the stem of the Anostomatinae and Cith-
arininae (Text-fig. 1).
Distichodus (Text-fig. 33) is apparently more primitive than Mesoborus
(Text-fig. 34), Ichthyoborus, or Phago (Text-fig. 35). The lengthening of
1938] Gregory & Conrad: Phylogeny of the Characin Fishes
353
body and jaws in Ichthyoborus and the development of unicuspid teeth are
probably secondary conditions as contrasted with the primitive short jawed,
deep bodied Distichodus. With regard to the scales Cockerell (1912) notes
that there is nothing to distinguish the scales of Ichthyoborus from those
of Distichodus.
Text-figure 34.
Skull of Mesoborus.
Gymnotidae.
To judge from the cranial osteology the gymnotids have probably been
derived from some primitive member of the Characinae. The chief habitus
specializations are:
354 Zoologica: New York Zoological Society [XXIII :17
(1) Marked forward inclination of the entire suspensorium, primi-
tively correlated with a small projecting mouth;
(2) Varied reduction or emphasis of premaxilla and maxilla;
(3) Reduction and loss of true pterygoid correlated with enlargement
of entopterygoid (mesopterygoid) and symplectic;
(4) Loss of parethmoid (prefrontal) and varied reduction of meseth-
moid ;
(5) Loss of suborbital series including lacrymal;
(6) Reduction or loss of postorbital process (dermosphenotic) ;
(7) Varied emphasis or reduction and loss of interfrontal fontanelle.
Regan has noted that the pectoral girdle of Rhamphichthys is much
less specialized than that of the typical “Sternarchids.” This genus has a
greatly elongated, decui’ved snout with a very small mouth and thus parallels
some of the long-snouted mormyrids.
A possible explanation of the peculiar specializations of the gymnotid
skull as above noted may be as follows:
All primitive gymnotids being lost, the genus Rhamphichthys, even
with its “mental vent,” may be near the structural starting point for the
remaining genera. We have only to suppose that with the extreme elonga-
tion of the snout the maxilla lost its normal connection with the reduced
doth
A Eigenmannia macrops B Electrophorus electrlcus
Text-figure 36.
Skulls of Eigenmannia and Electrophorus, dorsal views. After Gregory, 1933,
fig. 72.
1938] Gregory & Conrad: Phytogeny of the Characin Fishes
355
palatine, retaining only its contact with the premaxilla, the latter resting
only upon the prolonged mesethmoid. As the palatine disappeared the
true pterygoid became reduced and the entopterygoid enlarged. Meanwhile
the entire suspensorium was swinging far forward as the mouth became
Text-figure 37.
Skulls of Eigenmannia and Electrophorus, lateral views. After Gregory, 1933,
fig. 73.
356
Zoologica: New York Zoological Society
[XXIII :17
smaller and smaller at the end of the lengthening edentulous snout. In
the line leading to Sternarchus and Eigenmannia (Text-figs. 36, 37) a
reverse movement then set in involving the secondary shortening of the
snout and the secondary widening of the mouth which even acquired
secondary teeth. The end-stage is Electrophorus (Text-figs. 36, 37) in
which the mesethmoid is secondarily enlarged and strengthened and the
median frontal fontanelle is completely closed to support the upward thrusts
of the stout upper and lower jaws. Meanwhile the maxilla, having long
since lost its primitive contact with the palatine, remains as a vestigial
tab on the distal end of the secondarily enlarged premaxilla.
To those who hold to current interpretations of “Irreversibility of Evo-
lution” such an explanation can only appear fantastic. But if we once grasp
the idea of secular rise and recession of specializations the above steps will
be seen to be supported by strong morphological evidence as well as by
closely analogous series among the long-snouted mormyrids Gymnarchus
and their secondarily short-snouted relatives, such as Petrocephalus (Greg-
ory, 1933, Figs. 62B, 64).
The Characins and the Supposed South American-African Bridge.
The unique distribution of the Characidae in Africa and South America
alone has long engaged the attention of ichthyologists. Africa and South
America each serve as subsidiary centers for adaptive radiation of the
group, the Cheirodontinae, Serrasalmoninae, Erythrininae, Anostomatinae
and Hemiodontinae being characteristic of South America while the Cith-
arininae and Distichodontinae are confined to Africa. The Characinae and
Sarcodacinae, however, as here understood, have representatives in both
continents and the resemblances between certain African and South Ameri-
can forms are so striking (Text-figs. 9, 10, 11, 12, 22, 23, 24, 29) as to
leave a strong impression of close relationship.
Accordingly, the preceeding study leads us to the following tentative
conclusions:
(1) that the African and South American characins are closely
related ;
(2) that so far as known characins are wholly absent from ancient
freshwater deposits of North America;
(3) that there is much evidence analyzed by C. W. Andrews, Schuchert
and others for the reality of a narrow isthmian land or archi-
pelago connecting Brazil with West Africa even in possibly late
Tertiary times.
Dr. Bequaert, however, in the light of his wide knowledge of the
faunae and florae of South America and Africa, tells us that neither the
botanists nor the entomologists would favor the assumption of an extreme
or prolonged contact between the two continents in Tertiary times on
account of the large number of endemic families on either side of the
Atlantic.
Bibliography.
Andrews, C. W.
1906. A descriptive catalogue of the Tertiary Vertebrata of the Fayum,
Egypt. Printed by order of the Trustees of the British Museum.
London.
Borodin, N. A.
1929. Notes on some species and subspecies of the genus Leporinus Spix.
Mem. Mus. Comp. Zool., vol. 50. no. 3.
1938] Gregory & Conrad: Phylogeny of the Characin Fishes
357
1931. On the genus Anostomus (Family Characinidae). Bull. Mus. Comp.
Zool., vol. 72, no. 2, pp. 37-52.
Boulenger, G. A.
1909. Catalogue of the fresh-water fishes of Africa in the British Museum.
Vol. I. London.
1910. Fishes (systematic account of Teleostei). In the Cambridge Natural
History. London.
Bridge, T. W. and Haddon, A. C.
1893. Contributions to the anatomy of fishes. II. The air-bladder and
Weberian ossicles in the siluroid fishes. Philo. Trans. Royal Soc.
London, vol. 184 (1893), B, pp. 65-333.
Bruhl, C. B.
1847. Anfangsgrunde der vergleichenden Anatomie aller Thierklassen.
Erster Abschnitt, Die Skeletlehre der Fische, Lief. 1-3. Wien. IV,
254 pp., 19 pis. 4to and obi. folio.
de Castelnau, Francis
1855. Expedition dans les parties centrales l’Amerique du Sud, de Rio de
Janeiro a Lima, et de Lima au Para: executee par ordre du gouverne-
ment frangaise pendant les annees 1843 a 1847 sous la direction du
comte Francis de Castelnau. Zoologie. Poissons. Paris.
Cockerell, T. D. A.
1912. Scales of African characinid fishes. Smith. Misc. Coll., vol. 56, pp.
1-10, 2 pis.
1913. The scales of the South American characinid fishes. Ann. Carnegie
Mus., vol. 9, 1913-15, pp. 92-113.
Cope, E. D.
1878. Synopsis of the fishes of the Peruvian Amazon obtained by Professor
Orton during his expeditions of 1873 and 1877. Proc. Amer. Philo.
Soc., vol. 17, pp. 673-701.
Eastman, C. R.
1904. On the dentition of Rhynchodus and other fossil fishes. Amer. Natural-
ist, vol. 38, no. 448, pp. 295-299.
1917. Dentition of Hydrocyon and its supposed fossil allies. Bull. Amer.
Mus. Nat. Hist., vol. 37, art. 26, pp. 757-760.
Eigenmann, C. H.
1905. Divergence and convergence in fishes. Bio. Bull. Woods Hole, vol. 8,
no. 2, pp. 59-66.
1906. The freshwater fishes of South and Middle America. Popular Science
Monthly, June.
1907. Fowler’s “Heterognathous Fishes” with a note on the Stethaprioninae.
Amer. Naturalist, vol. 41, no. 492, pp. 767-772.
1912. The freshwater fishes of British Guiana, including a study of the
ecological grouping of species, and the relation of the fauna of the
plateau to that of the lowlands. Mem. Carnegie Mus., vol. 5, pp. 1-554.
1914. Some results from the studies of South American fishes. Indiana Univ.
Bull., vol. 12, no. 1, pp. 20-48.
1915. The Serrasalminae and Mylinae. Ann. Carnegie Mus., vol. 9, pp. 226-
272.
1916. The Cheirodontinae, a sub-family of minute characid fishes of South
America. Mem. Carnegie Mus., vol. 7, no. 1, pp. 1-100.
1917-27. The American Characidae. Mem. Mus. Comp. Zool., vol. 43, pt. 1,
1917; pt. 2, 1918; pt. 3, 1921; pt. 4, 1927.
1921. Nature and origin of the fishes of the Pacific slope of Ecuador, Peru
and Chile. Proc. Amer. Philo. Soc., vol. 60, pp. 503-523.
1927. Fresh-water fishes of Chile. Mem. Nat. Acad. Sci., vol. 22, 2nd.
memoir.
358
Zoologica: New York Zoological Society
[XXIII :17
Eigenmann, C. H. and Eigenmann, R. S.
1889. A revision of the edentulous genera of Curimatinae. Ann. N. Y. Acad.
Sci., vol. 4, 1887-89, pp. 409-440.
1891. A catalogue of the fresh-water fishes of South America. Proc. U. S.
Nat. Mus., vol. 14, pp. 1-81.
Eigenmann, C. H. and Myers, G. S.
1929. The American Characidae. Mem. Mus. Comp. Zool., vol. 43, pt. 5.
Fowler, H. W.
1906. Further knowledge of some heterognathous fishes. Parts I and II.
Proc. Acad. Nat. Sci., vol. 58, pp. 293-351; 431-483.
1930. The fresh-water fishes obtained by the Gray African Expedition-1929.
With notes on other species in the Academy collection. Proc. Acad.
Nat. Sci., vol. 82, pp. 27-83.
Gill, T.
1896. 1. The differential characters of characinoid and erythrinoid fishes.
Proc. U. S. Nat. Mus., vol. 18, pp. 205-209.
1896. 2. Note on the fishes of the genus Characinus. Proc. U. S. National
Mus., vol. 18, pp. 213-215.
1896. 3. Notes on characinoid fishes with ctenoid scales, with a description
of a new Psectrogaster. Proc. U. S. Nat. Mus., vol. 18, pp. 199-203.
1896. 4. The nomenclature of the fishes of the characinoid genus Tetrag-
onopterus. Proc. U. S. Nat. Mus., vol. 18, pp. 225-227.
Gregory, William K.
1933. Fish skulls: a study of the evolution of natural mechanisms. Trans.
Amer. Philo. Soc., (N. S.) vol. 23, no. 2, pp. 75-481.
Gregory, W. K. and Conrad, G. M.
1936. The structure and development of the complex symphysial hinge-joint
in the mandible of Hydrocyon lineatus Bleeker, a characin fish. Proc.
Zool. Soc. London, 1936, pt. 4, pp. 975-984.
Gunther, A.
1864. Catalogue of the fishes of the British Museum. Vol. V. Physostomi.
London.
Haseman, J. D.
1911. Some new species of fishes from the Rio Iguassu. Ann. Carnegie Mus.,
vol. 7, nos. 3, 4, pp. 374-387.
1912. Some factors of geographical distribution in South America. Ann.
N. Y. Acad. Sci., vol. 22, pp. 9-112.
Jordan, D. S.
1923. A classification of fishes, including families and genera as far as
known. Stanford Univ. Pub., Biological Sci., vol. 3, no. 2.
Matthew, W. D.
1915. Climate and evolution. Ann. N. Y. Acad. Sci., vol. 24, pp. 171-318.
Muller, J. and Troschel, F. H.
1845. Horae ichthyologicae. Beschreibung und Abbildung neuer Fische; die
Familie Characinen. 3 pts. in 2 vols. Berlin.
Myers, G. S.
1929. 1. Cranial differences in the African characin fishes of the genera
Alestes and Brycinus, with notes on the arrangement of related genera.
Amer. Mus. Novitates, no. 342.
1929. 2. On curimatid characin fishes having an incomplete lateral line, with
a note on the peculiar sexual dimorphism of Curimatopsis macrolepis.
Ann. Mag. Nat. Hist., ser. 10, vol. 3, p. 618.
1938]
Gregory & Conrad: Pliylogeny of the Characin Fishes
359
Nichols, J. T.
1930. Speculation on the history of the Ostariophysi. Copeia, no. 4, pp.
148-151.
Nichols, J. T. and Griscom, L.
1917. Fresh-water fishes of the Congo Basin obtained by the American
Museum Congo Expedition, 1909-1915. Bull. Amer. Mus. Nat. Hist.,
vol. 37, art. 25, pp. 653-756.
Norman, J. R.
1928. The South American characid fishes of the subfamily Serrasalmoninae,
with a revision of the genus Serrasalmus Lacepede. Proc. Zool. Soc.
London, pt. 4, 1928.
Pearson, N. E.
1924. Fishes of the eastern slope of the Andes. 1. The fishes of the Rio
Beni Basin, Bolivia collected by the Mulford Expedition. Indiana Univ.
Studies, vol. 11, no. 64.
Regan, C. T.
1911. The classification of the teleostean fishes of the order Ostariophysi.
1. Cyprinoidea. Ann. Mag. Nat. Hist., ser. 8, vol. 8, pp. 13-32.
Rowntree, W. S.
1903. On some points in the visceral anatomy of the Characinidae, with an
enquiry into the relations of the ductus pneumaticus in the Physostomi
generally. Trans. Linn. Soc. London, 2 ser., vol. 9, pp. 47-81.
1906. On the dentition of the characinoid genus Piabuca; a new generic
character. Ann. Mag. Nat. Hist., ser. 7, vol. 17, pp. 240-243.
Sagemehl, M.
1885. Beitrage zur vergleichenden Anatomie der Fische. III. Das cranium der
Characiniden nebst allgemeinen Bemerkungen uber die mit einem
Weber’schen Apparat versehenen Physostomenfamilien. Morphol.
Jahrb., vol. 10, pp. 1-119.
Schuchert, C.
1932. Gondwana Land bridges. Bull. Geol. Soc. Ayner., vol. 43, pp. 875-915.
Spix, J. B. von and Agassiz, L.
1829-31. Selecta genera et species piscium quos in itinere per Brasiliam annis
1817-20 . . . peracto 'collegit et pingendos curavit Dr. J. B. de Spix
. . . digressit, descripsit et observationibus anatomicis illustravit Dr. L.
Agassiz, praefatus est et edidit itineris socius Dr. C. F. Ph. de
Martius (Memoriae J. B. de Spix). Monachii. xvi, 138 pp., and atlas
of 101 pis.
Abbreviations Used in the Illustrations.
“adn” — adnasal
“alsp” — alisphenoid
ang — angular
art — articular
boc — basioccipital
brstg — branchiostegal
cerhy — ceratohyal
cl, clt — cleithrum
cor — hypocoracoid
cran — cranial
Deth — dermethmoid
dn — dentary
dsph, dsphot — dermosphenotic
entp — entopterygoid
epiot — epiotic
exo — exoccipital
f, for — foramen
fr — frontal
hyo — hyomandibular
iop — interopercular
ju — jugal
la — lacrymal
m — mesocoracoid
mtp — metapterygoid
mx — maxilla
na — nasal
op, operc — opercular
opisthot, opot — opisthotic
orbsp — orbitosphenoid
pa — parietal
360
Zoologica: New York Zoological Society
pareth — parethmoid
pas — parasphenoid
pelt, pci — postcleithrum
pf — prefrontal
pi — palatine
pmx — premaxilla
pop, preoperc — preopercular
ptm — posttemporal
pto — pterotic
ptr — pterygoid
qu — quadrate
sc, scap — hypercoracoid (scapula)
so — suborbital
soc — supraoccipital
sop — subopercular
sphot — sphenotic
supcl, supclt — supracleithrum
sup. op. — supraopercular
sup. orb — supraorbital
sym — symplectic
tab — tabular
v — vertebra
vo — vomer
Deichmann: Holothurians
361
18«
Eastern Pacific Expeditions of the
New York Zoological Society. XVE
Holothurians from the Western Coasts of Lower California and
Central America, and from the Galapagos Islands.1
Elisabeth Deichmann
Assistant Curator of Invertebrates, Museum of Comparative
Zoology, Cambridge, Massachusetts.
(Text-figures 1-15).
[This is the Sixteenth of a series of papers dealing with the collections
of the Eastern Pacific Expeditions of the New York Zoological Society made
under the direction of William Beebe. The present paper is concerned with
specimens taken on the Eastern Pacific Zaca (1937-1938) and the Arcturus
Oceanographic (1925) Expeditions. For data on localities, dates, dredges,
etc., of these expeditions, refer to Zoologica, Vol. VIII, No. 1, pp. 1-45
( Arcturus ) and Zoologica, Vol. XXIII, No. 14, pp. 287-298 (Eastern Pacific
Zaca ) .]
Contents.
Page
Introduction 362
Order Aspidochirota
Family Stichopodidae
Genus Parastichopus H. L. Clark
Parastichopus calif ornicus (Stimpson) 362
Parastichopus parvimensis (H. L. Clark) 362
Genus Stichopus
Stichopus fuscus Ludwig 363
Family Holothuriidae
Genus Labidodemas Selenka
Labidodemas americanum sp. nov 363
Genus Holothuria Linnaeus
Holothuria arenicola Semper 364
Holothuria impatiens (Forskal) 365
Holothuria inhabilis Selenka 365
Holothuria inornata Semper 365
Holothuria languens Selenka 367
Holothuria lubrica Selenka 368
Holothuria marenzelleri Ludwig var. theeli var. nov 369
Holothuria gyrifer (Selenka) 371
Holothuria pardalis Selenka 372
Order Dendrochirota
Family Cucumariidae
Genus Cucumaria Blainville
Cucumaria calif ornica Semper 372
Genus Pentamera Ayres
Pentamera beebei sp. nov 374
Pentamera chierchia (Ludwig) 374
Pentamera zacae sp. nov 375
l Contribution No. 564, Department of Tropical Research, New York Zoological Society.
362
Zoologica: New York Zoological Society
[XXIII :18
Genus Thyone Oken
Tliyone benti Deichmann var. zacae var. nov 376
Thyone gibber (Selenka) 376
Family Phyllophoridae
Genus Euthyonidium gen. nov.
Euthyonidium ovulum (Selenka) 380
Genus Phyllophorus Grube
Phyllophorus zacae sp. nov 381
Family Psolidae
Genus Thyonepsolus H. L. Clark
Thyonepsolus beebei Deichmann 382
Order Molpadonia
Family Caudinidae
Genus Paracaudina
Paracaudina chilensis (J. Muller) 383
Bibliography 384
Introduction.
The holothurians brought back by the Eastern Pacific Zaca Expedition
of 1937-1938 are of even greater interest than those secured in 1936.2
Twenty different species were collected and only five are identical with those
brought back in 1936 (two of these were then described as new). Of the
remaining fifteen species four are new, while three are reported for the
first time since the types were described. Only eight may be designated as
common and even here the material in hand has in two cases given very
valuable additional information and helped to straighten out some moot
questions.
A few shore species from the Galapagos Islands (secured by the Arc-
turus Oceanographic Expedition in 1925) have been included, thereby bring-
ing the total number of species discussed up to twenty-three. Some of the
species in the Arcturus collection were also taken by the Zaca.
I beg Dr. William Beebe to accept my sincere thanks for the oppor-
tunity of studying this interesting collection.
Order Aspidochirota.
Family Stichopodidae.
Genus Parastichopus H. L. Clark, 1922.
Parastichopus Californians (Stimpson).
For diagnosis, synonyms, etc., see Deichmann, 1937, p. 163.
Local Range: Two large specimens from Cedros Island, 40 fathoms,
(Station 126 D-17).
Parastichopus parvimensis (H. L. Clark).
Stichopus parvimensis H. L. Clark, 1913, p. 234; 1922, p. 70, pi. 1, figs.
13-14; G. H. Parker, 1921, p. 205.
Parastichopus parvimensis, Deichmann, 1937, p. 163 (passim).
Diagnosis: Resembles P. calif ornicus but is normally dark brown with
black tips to the papillae. Spicules minute tables (disk up to 0.05 mm.) and
small buttons (length 0.08 mm.). Shallow water form.
Type: U.S.N.M.
Type Locality: Cedros Island, west coast of Lower California, 3.5 feet.
General Distribution: From San Pedro to Cedros Island, in shallow
water.
2 See Deichmann, 1937.
1938]
Deichmann: Holothurians
363
Local Range : One much contracted specimen from Cedros Island, 20
fathoms, (Station 126 D-18) .
Remarks : The specimen agrees in every respect with various individu-
als from Corona del Mar, California.
Genus Stichopus Brandt, 1835.
Stichopus fsfscMS Ludwig.
For diagnosis, synonyms, etc., see Deichmann, 1937, p. 163.
Local Range : Two large, well expanded specimens from Hood Island,
Galapagos, 15 feet depth ( Arcturus Oceanographic Expedition).
Remarks : The first record of this characteristic form from the Gala-
pagos. Previously it had been reported from Ecuador (Ludwig).
Family Holothuriidae.
Genus Labidodemas Selenka, 1867.
Labidodemas Selenka, 1867, p. 309; Sluiter, 1901, p. 21; Fisher, 1907,
p. 674.
Diagnosis: Medium sized forms with 20 small tentacles, terminally
placed; anus subterminal. Feet few, arranged in three double rows ventrally
and in two single to double rows on the dorsal side ; either all the_ feet are
cylindrical or the dorsal ones are more papilliform. Skin thin, soft. Inner
anatomy as in a typical Holothuria. Spicules form a more or less scattered
layer of tables with spire of various degrees of development; besides also a
few plates or buttons with large holes; C-shaped spicules said to be present
in some cases. In one species the spicules seem to be completely reduced.
Shallow water.
Type Species : Labidodemas semperianum Selenka.
Remarks: Three or four species were hitherto described, ranging from
the East Indies to Hawaii; [the latter locality represents supposedly the
type locality for L. semperianum, but the label in the M.C.Z. says Society
Islands.] According to Sluiter (1901, p. 21) three of the species are identi-
cal (see Fisher, 1907, p. 675), while the fourth species lacks spicules.
The species described below differs so markedly from the other species
known that it cannot be considered a variety of any of these.
Labidodemas amerieanum sp. nov.
Diagnosis: Ventral feet cylindrical, in double rows; dorsal feet papilli-
form, in two scattered rows. Spicules delicate tables with four central holes
and a circle of marginal holes. Spire low, mostly incomplete or reduced to
four knobs. Ventral feet with large end plate and plates with four to five
holes, possibly reduced tables. Dorsal papillae with no end plate but the
same kind of plates as the ventral feet. Skin soft, golden brown in color to
greenish, appendages mostly dark brown.
Type: M.C.Z.
Type Locality: Jasper Island, Costa Rica.
General Distribution: Probably widespread in the Panamic region.
Local Range: Two contracted specimens from Jasper Island, Gulf of
Nicoya, Costa Rica.
Remarks: The material has been compared with the type material of
L. semperianum from Hawaii, (or Society Islands) in the M.C.Z. The latter
364 Zoologica: New York Zoological Society [XXIII :18
Text-figure 1.
Labidodemas americanum sp. nov. 1-3, Tables with reduced spire from skin. 4-5,
Buttons in wall of feet, probably derived from tables. Scale 1/100 mm.
species has much stouter spicules with long spines on the top of the spire, so
it is out of question that the present species is identical with Selenka’s
species as one would be inclined to expect.
The specimens are small (5-6 cm. long) and not well suited for anatomi-
cal studies. The radials are large squarish while the interradials are low;
both are thin and delicate in structure. The Polian vesicle is small and
ventrally placed; the single stone canal is embedded in the dorsal mesentery,
its flattened head projecting free to the right. The intestine is partly torn
and filled with coarse coral debris. No traces of gonads are present in the
types.
Genus Holothuria Linnaeus.
Holothuria arenicola Semper.
Holothuria arenicola Semper, 1868, p. 81, pi. 20, pi. 30, fig. 13, pi. 35,
fig. 4; Theel, 1886a, p. 7; Fisher, 1907, p. 662; Deichmann, 1930,
p. 66, pi. 4, figs. 1-9; 1937, p. 165 (passim); Panning, 1935, IV,
p. 88, text-figure 73.
Holothuria maculata Ludwig, 1887, p. 2.
Nec Holothuria maculata Ludwig, 1894, p. 1 (= H. inhabilis Selenka;
see Deichmann, 1937, p. 165).
Diagnosis: Slender tapering form (up to 20 cm. long) ; mouth almost
terminal with 20 small tentacles; anus terminal. Feet cylindical, in scat-
tered bands almost equally numerous ventrally and dorsally and only very
slightly smaller on the dorsum. Calcareous ring low; ring canal unusually
far behind the calcareous ring (1-2 cm.) ; normally one small stone canal
attached to the mesentery; one or two Polian vesicles. Cuvieran organs
present but seem normally to be small; gonads numerous tubes in a tuft
behind the vascular ring.
Spicule tables with 4-8 marginal holes and a low four-pillared spire
with 8-12 spines on top. Buttons smooth, regular, usually with 6 holes often
decreasing in size with advancing age; few buttons present in young indi-
viduals. Feet with end plate and straight supporting rods with perforated
ends and often holes along the middle sometimes developed as regular large
buttons with numerous holes.
Color mottled gray, with either two rows of dark spots on the dorsum
or irregularly spotted.
Type : Possibly in Germany.
Type Locality: Bohol, Philippines.
General Distribution: Almost circumtropical. Common in the West
Indies and Hawaii, in shallow water often buried in sand.
1938]
Deichmann: Holothurians
365
Local Range : Two small specimens from Tangola-Tangola Bay, Mexico,
in coral (Station 196 D-15) ; one small specimen from Ballena Bay, Costa
Rica ( Zaca 1938) ; one from Gardner Bay, Hood Island, Galapagos ( Arctu -
rus, 1925).
Remarks : This widespread species was recorded from the Galapagos by
Theel, 1886a, and from Panama by Ludwig in 18873. Later Ludwig mistook
H. inhabilis from off Cocos Island, 66 fathoms, for this common species.
It has been reported by most collectors working in the Panamic waters.
Holothuria impatiens (Forskal).
Fistularia impatiens Forskal, 1775, p. 121, pi. 39, fig. B.
Holothuria impatiens Theel, 1886a, p. 7; Fisher, 1907, p. 667; Pan-
ning 1935, IV, p. 86, text-fig. 72 (complete list of references) ;
Deichmann, 1930, p. 64, pi. 3, figs. 17-18.
Diagnosis : Medium sized form (10-15 cm. long), slender, often dis-
tinctly bottle-shaped with long narrow “neck.” Tentacles 20, small, mouth
and anus both terminal. Feet cylindrical, slightly more papilliform on the
dorsal side, scattered over the entire body without any apparent order,
often placed distinctly on warts. Calcareous ring low; stone canal free;
Polian vesicles usually 2-4; Cuvieran organs unusually thick tubes; gonads
placed in anterior part of body.
Spicules regular tables with 8 large marginal holes; spire with one or
two cross beams and numerous short spines on the top. Buttons smooth
oval with 6 large holes. Feet with end plate, often reduced in the dorsal
appendages' and numerous slightly curved supporting rods, heavy, with few
holes in the ends and on each side of the middle, occasionally developed as
complete buttons. Color mottled, gray, skin decidedly sandy to the touch.
Type : Lost.
Type Locality: Red Sea.
General Distribution: Almost circumtropical. Common in the West
Indies and in Hawaii; shallow water to few fathoms depth.
Local Range: Five small individuals from Situatanejo, Mexico; 6 from
Station 195 D-15; two fragments from Galapagos ( Arcturus 1925).
Remarks: This common species was reported from the Galapagos in
1886, by Theel; from Lower California in 1913 by H. L. Clark.
Holothuria inhabilis Selenka.
For diagnosis, etc., see Deichmann, 1937, p. 164.
Local Distribution: Three large specimens from Corinto, Nicaragua,
2 and 1 fathoms (Station 200 D-5 and D-15).
Remarks: The specimens agree with those previously reported from
Lower California, Cocos Island and Clarion Island. The depth is excep-
tionally low but very likely the larger individuals migrate shorewards as is
known to be the case in other species.
Holothuria inornata Semper.
Holothuria inornata Semper, 1868, p. 252, pi. 40, fig. 1 ; Panning, 1934,
II, p. 33, text-fig. 28 (reproduction of Semper’s figure).
Diagnosis : Large form (20 cm. or more) with thick skin ; 20 large ten-
tacles; mouth ventrally directed; anus terminal; ventral feet numerous,
3 Deichmann (1937) erroneously writes “1875” instead of “1887.”
366
Zoologica: New York Zoological Society
[XXIII :18
cylindrical, forming a crowded sole; dorsal appendages mostly papillae,
many arranged on more or less distinct warts. Inner anatomy not remark-
able except for a cluster of 5-6 free stone canals; Cuvieran organs present.
Spicules a thinly scattered layer of tables; dorsally they are large with
mostly complete disk consisting of four central holes and four marginal
ones and with about 12 marginal spines; ventrally most of the tables are
smaller with disk completely reduced; spire with four pillars, usually one
cross beam and ending in four upright and 8 vertical spines. Feet with
end plate and no supporting rods or plates except a number of perforated
plates close to the end plate; dorsal appendages with a vestigial end plate
and numerous supporting rods, straight or curved, smooth with perforated
ends. Color almost black with a reddish tinge.
Type : Hamburg.
Type Locality : Mazatlan, Mexico.
General Range : Probably the entire Panamic region. Shallow water.
Local Range: Two small specimens from Ballena Bay, Costa Rica, shore;
one small from Corinto, Nicaragua; two large from Galapagos ( Arcturus ,
1925).
Text-figure 2.
Holothuria inornata Semper. 1-5, Tables of various development from skin and
dorsal appendages. 6-7, Typically reduced tables from ventral appendages.
8, Supporting plate from near end plate in ventral apendages. 9, Straight
supporting rod from dorsal appendage. Scale 1/100 mm.
Remarks : The species has not been reported since the original type
specimens were secured. The general exterior of the animal is slightly sug-
gestive of H. grisea Selenka from the West Indies, and the coasts of Brazil
and West Africa, but it is much larger and more darkly colored and has
numerous stone canals.
Although it is not impossible that H. atra occurs in the Galapagos
1938]
Deichmann: Holothurians
367
Islands and Clipperton Island (see Panning, chart, p. 29) it is very likely
that the two records actually refer to this species.
Holothuria Ianguens Selenka.
Holothuria Ianguens Selenka, 1867, p. 335, pi. 19, figs. 80-81; H. L.
Clark, 1920, p. 149; Panning, 1934, II, p. 45, text-fig. 37.
Holothuria imitans Theel, 1886a, p. 7.
Diagnosis : Slender form (rarely more than 10 cm. long); tentacles
small, mouth and anus terminal; feet ventrally cylindrical, dorsally papilli-
form, in indistinct longitudinal rows, scattered. Inner anatomy not remark-
able. Cuvieran organs apparently lacking.
Spicules a crowded layer of tables with reduced disk, except in very
young individuals (3 cm. long) where it is circular with smooth edge; spire
tall with normally one cross beam and ending in a Maltese cross of 8 flat
spines; in the juvenile tables the spire is more tapering, often with two
cross beams and insignificant teeth. Feet with end plate and curved sup-
porting rods, either smooth with perforated ends or with lateral projections
which ultimately may become united so they form a series of holes along
the sides. Dorsal papillae lack end plate but are filled with curved support-
ing rods with spinous or perforated ends.
Color mottled reddish-brown, lighter below; tentacles yellow. Skin
slightly sandy to the touch.
Type: In Germany; paratype in M.C.Z.
Type Locality : Panama.
General Distribution: Probably the entire Panamic region, including
Galapagos. Shallow water.
Local Range: One specimen from Santa Cruz Bay, Mexico, 4 and 15
fathoms (Station 195 D-14 and D-15) ; one from Jasper Island, Gulf of
Nicoya, Costa Rica (Station 213 L-3) ; two from Galapagos ( Arcturus ,
1925).
Text-figure 3.
Holothuria Ianguens Selenka. 1-2, Juvenile table, disk and lateral view, from
small specimen from the Galapagos Islands. 3-5, Normal reduced tables,
lateral view from adult specimens. 6, Straight supporting rod from
dorsal appendage. Scale 1/100 mm.
368
Zoologica: New York Zoological Society
[XXIII :18
Remarks: The species differs from the common West Indian form,
H. surinamensis Ludwig, in its smaller size, its total lack of large flat rods
scattered in the skin, and in the flat undivided spines on the top of the
spire.4 It differs from the Indo-Pacific form, H. imitans Ludwig from
the Navigator Islands, in the more delicate tables with smooth margin to
the disk, which is preserved only in very young individuals, while Ii. imitans
apparently retains the disk on most of its tables.
H. languens was described from Panama, and as far as known it is
restricted to that region. Semper referred (1868, p. 87) an incomplete
specimen from British Guiana to H. languens, as it seemed to resemble
Selenka’s species. Undoubtedly he was dealing with Ludwig’s H. surina-
mensis, described a few years later from Dutch Guiana (Surinam) — a
species which never has been reported from outside of the West Indies.
Sluiter’s record (1910, p. 331) of H. languens, from St. Thomas, W. I., also
undoubtedly refers to H. surinamensis .5
Ludwig withdrew, but as I consider incorrectly, H. surinamensis, mak-
ing it a synonym of H. imitans which he, in the same paper (1887), reported
from Panama and Galapagos. He did not mention Selenka’s species at all,
being probably misled by Selenka’s very untypical figure. It is not clear
whether he refers to his original material from the Navigator Islands or
to his Panamic-Galapagos material when he speaks of the presence of tables
with spinous margin. As far as the present material shows, the disk is
smooth in H. languens. H. L. Clark (1920, p. 148) describes an H. imitans
from somewhere in the Pacific Ocean — Albatross cruise, locality label lack-
ing— and emphasizes the spinous disk and the tapering spire with 12-20
teeth, different from H. languens which he records from the Gulf of Panama.
Holothuria lubrica Selenka.
Holothuria lubrica Selenka, 1867, p. 329, pi. 18, fig. 59; Panning, 1934,
p. 45, text-fig. 38 (copied from Lampert, 1896, fig. 2).
Holothuria kapiolaniae Bell, 1887, p. 533; Fisher, 1907, p. 653.
For diagnosis, etc., see Deichmann, 1937, p. 165.
Local Range: Four specimens from Ballena Bay, Costa Rica, shore; 1
from Corinto, Nicaragua, shore.
Remarks: The typical form seems to be restricted to Panamic waters
and possibly Hawaii ( H . kapiolaniae Bell). The records from the Malay
Archipelago, given by Sluiter (1901, p. 8), need re-investigation. Sluiter
gives no measurements of the spicules and it is possible that he was dealing
with the other species with similar but shorter spicules before him, viz.,
H. parva Lampert, type locality east coast of Africa, or possibly H. moebii
Ludwig.
Theel’s record of H. lubrica from St. Bartholomew, W. I., has unfor-
tunately been repeated uncritically by various writers. Actually Theel says
about his West Indian material: “either lubrica or the foregoing species,”
( =H . glab err ima) , and the description is decidedly of H. glaberrima.
Ludwig maintains (1898, p. 434) that he has found spicules of the
glaberrima type in specimens from Mazatlan, Mexico, but he has not
4 Deichmann (1926, pi. 1, fig. 1) selected the most reduced table she could find in H. suri-
namensis, as well as the most complete type. Unfortunately this figure was selected by Panning
(1934, II, text-fig. 34) instead of the more typical tables figured in 1930, pi. 3, figs. 12-15, 19.
Normally H. surinamensis has heavy double spines on the top of the spire.
5 The records of H. imitans from Bermuda are due to Panning who refers Clark’s material of
H. surinamensis (1898, p. 412, and 1899, p. 118) to Ludwig’s Indo-Pacific species. His reasons
are that Clark states that fiat bars apparently are lacking in the Bermuda specimens (but may
have been overlooked) and that he sometimes has found spines on the edge of the tables in speci-
mens from Bermuda as well as from Jamaica, adding that these tables may have more spines
(18-20) on the top of the spire. Presumably this is merely an untypical variation, and H. imitans
must be removed from the list of West Indian species until a careful comparison has been made.
1938] Deichmann: Holothurians 369
Text-figure 4.
Holothuria lubrica Selenka. 1-5, Typical spinous rods from integument.
Holothuria glaberrima Selenka. 6-8, Typical rods from integument. Scale
1/100 mm.
compared them with spicules of the true glaberrima. Although the spicules
vary somewhat in development and are of approximately the same size, I
have always found that they cannot be confused. Moreover, the two species
are rather different in exterior. H. glaberrima is dark brown and robust,
while H. lubrica is smaller and grayish in color, often with yellow feet and
two rows of dark spots on the dorsum.
Holothuria marenzelleri Ludwig var. theeli var. nov.
Holothuria marenzelleri var. ? Theel, 1886a, p. 7.
Holothuria marenzelleri Ludwig, 1887, p. 2, pi. 2, fig. 12; Panning,
1934, II, p. 47, fig. 41 (copy of Ludwig’s figure).
Diagnosis: Large robust form (up to 20 cm. long) with 20 large bushy
tentacles; ventral side covered by numerous cylindrical feet; dorsal side
with numerous small papillae. Single free stone canal.
Spicules in young individuals short spectacle-shaped rods, often devel-
oped as asymmetrical or symmetrical plates; surface smooth; margin scal-
loped. In larger individuals the spicules change into larger rods with more
or less rough surface and numerous holes in the ends and along the margins.
Color dark brown.
Type: M. C. Z. cat. no. 665. Ludwig’s types of the typical form from
the Nicobars are probably in Germany.
Type Locality: Galapagos Islands.
General Distribution: According to Panning (chart, p. 44) taken only
in the Galapagos Islands.
Local Range: Nine specimens ranging from small to large, from Tower
Island and Hood Island, Galapagos (Arcturus) .
Remarks: Ludwig gives no figures of the spicules in his species from
the Nicobars and his description may equally well refer to H. erinaceus
Semper — a widespread form in the East Indies. Theel was the first to
express his doubt whether the Galapagos specimens were identical with
the typical H. marenzelleri.
Only examination of the types of H. marenzelleri, or of specimens from
Nankauri, Nicobars, can decide whether two species are concealed under one
name or not. Apparently there are no records of H. marenzelleri occurring
between the Galapagos Islands and the Malay Archipelago.
370 Zoologica: Neiv York Zoological Society [XXIII :18
Text-figure 5.
Holothuria marenzelleri Ludwig var. theeli var. nov. 1-8, Tyical smooth plates
and rods from young individuals from Galapagos. Scale 1/100 mm.
Text-figure 6.
Holothuria marenzelleri Ludwig var. theeli var. nov. 1-3, Typical rods with
numerous holes and rough surface from large individuals from same
locality as the small individuals figured above. Scale 1/100 mm.
It is with great hesitation that the large individuals with rather differ-
ent spicules are referred to H. 'marenzelleri var. theeli. But it is quite pos-
sible that the small smooth spicules are completely reduced and the spicules
in the older individuals become large and rough.
1938]
Deichmann: Holothurians
371
Holothuria gyrifer (Selenka).
Stichopus gyrifer Selenka, 1867, p. 319.
Holothuria monacaria, Fisher, 1907, p. 659; H. L. Clark, 1923, p. 163;
1938, pi. 16, fig. 7 (colored figure) ; Panning, 1934, III, p. 69, text-
fig. (complete list of references).
Holothuria ? patagonica R. Perrier, 1904, p. 13.
Stichopus ( Holothuria ?) patagonicus R. Perrier, 1905, pp. 11-17,
pi. 1, figs. 1-3.
Nee P solus monacarius Lesson, 1830, p. 225, pi. 76.
Diagnosis'. Slender form, of medium length (10-5 cm.), slightly bottle-
shaped, with pronounced “neck” when normally expanded. Tentacles 20,
cylindrical feet small; mouth and anus terminal. Feet few, in indistinct
rows; ventrally cylindidcal feet, dorsally papillae. Inner anatomy rather
similar to that of H. impatiens.
Spicules a crowded layer of tables and buttons. Tables with round disk
with a complete circle of marginal holes, edge smooth, spire moderately high
with four pillars and mostly one cross beam and few teeth on the top. But-
tons smooth with six large holes. Ventral feet with large end plate but no
special supporting rods, except for a few reticulated plates next to the end
plate or the typical buttons may be slightly larger. Dorsal appendages with
no end plate and besides the typical buttons a number of long curved rods
with short transverse projections which often unite so they form a row of
lateral holes on each side.
Color bright brown, ventrally paler; appendages white, base surrounded
by a circular white area.
Types : M. C. Z. and Gottingen, Germany.
Type Locality : Hawaii.
General Distribution : Indo-West Pacific to Hawaii, and Gulf of Cali-
fornia (Clark, 1923). A specimen without spicules, tentatively described as
a new species, reported from Chile by R. Perrier, 1904.
Local Range: Two large specimens from Jasper Island, Gulf of Nicoya,
Costa Rica, shore.
Remarks : It has been deemed advisable to change the name monacaria
Lesson to gyrifer Selenka, since it is absolutely certain that the species
usually called monacaria has nothing to do with the species which Lesson
described and figured from Tahiti, while the type of gyrifer has been exam-
ined. It is a mystery how the present species with few appendages has
been confused with Lesson’s species which has numerous conical appendages
on the dorsal side while the ventrum is covered by numerous cylindrical
feet. What Lesson’s species actually is cannot be said offhand — possibly it
is a well known form and the name monacaria may be placed among the
synonyms.
H. gyrifer is a most strikingly colored species (see Clark’s colored
figure, 1938) and it can hardly be confused with any other form. It resem-
bles H. impatiens in general shape but it is more brightly colored and the
skin is less rough to the touch. The spicules resemble those found in H.
arenicola, but differ in the presence of a complete circle of holes in the
margin of the disk, and the holes are often rectangular, while in H. arenicola
the holes are more circular and frequently the marginal holes are reduced
to four. Also the buttons in gyrifer are large while in arenicola they tend
to become very small.
Stichopus patagonicus R. Perrier, from Sta. Cruz, Patagonia, is unques-
tionably this species. The single individual measured 5.3 cm. (strongly con-
tracted) and the spicules were dissolved and the calcareous ring slightly
corroded. Possibly it was wrongly labelled, as it is most unlikely that this
tropical form should occur so far south.
372 Zoologica: New York Zoological Society [XXIII :18
Holothuria pardalis Selenka.
Holothuria pardalis Selenka, 1867, p. 336, pi. 19; fig. 85; Fisher, 1907,
p. 664, pi. 69, figs. 1, la-g; Panning, 1935, V, p. 3, text-fig. 106.
Diagnosis: Small to medium-sized form (10-12 cm.) slender, with 20
small tentacles; mouth and anus terminal; feet cylindrical, in indistinct
bands, not crowded, rarely retracted into warts.
Spicules tables with reduced or low spire and small disks often com-
posed of four holes, margin spinous. Buttons with 6-8 holes, often incom-
plete and frequently twisted. Feet with end plate, slightly smaller in the
dorsal appendages and supported by curved supporting rods with per-
forated ends.
Color mottled gray, extremely variable, reminiscent of H. arenicola
Semper.
Type: M. C. Z. and Gottingen, Germany.
Type Locality: Hawaii.
General Distribution: Indo-West Pacific and Galapagos Islands.
Local Range: Three specimens from Darwin Bay, Galapagos Islands
( Arcturus ) .
Remarks: The specimens were kept in formalin but the spicules are
well preserved and agree with those figured by Fisher.
Order Dendrochirota.
Family Cucumariidae.
Genus Cucumaria Blainville, 1834.
For definition and diagnostic characters, see Deichmann, 1938, p. 103.
Cucumaria californica Semper.
Cucumaria californica Semper, 1868, p. 235, pi. 39, fig. 16, pi. 40, fig.
10; Theel, 1886a, p. 9.
Nec Cucumaria californica, Edwards, 1910, p. 601. (= Cucumaria fallax
Ludwig) .
Diagnosis: Medium sized forms (about 10 cm.) with ten bushy ten-
tacles of equal size. Skin soft, smooth; feet large, soft, completely retrac-
tile, arranged in five bands, not scattered in the interambulacra. Calcareous
ring simple, single stone canal and one or two Polian vesicles.
Spicules four-holed buttons or longer oblong plates with holes mostly
in two rows ; surface knobbed or smooth. Feet with no end plate or a vestige
and a few three-armed supporting rods. Tentacles with few plates or rods
or no spicules at all. In older individuals the spicules may be almost com-
pletely lacking.
Color varying from almost black to slate-colored or almost white with
dark tentacles and anterior end which always seems to be blackish.
Type: Possibly in Germany.
Type Locality: Mazatlan, Mexico.
General Distribution: Probably widespread in the Panamic region.
Local Range: Eighteen specimens from Situatanejo Bay, Mexico,
shore.
Remarks : A well marked species which can hardly be confused with any
other species described from the Panamic region.
The name C. californica Semper has by a curious error been applied to
1938]
Deichmann: Holothurians
373
a large frondosa-l ike form, C. fallax Ludwig, which is common in the Bering
Sea. A large individual of the latter species, from the Albatross cruises,
received in one way or another the label Galapagos. Edwards, who identified
the material, assumed that the label was correct and that this species was
identical with Semper’s C. calif ornica. He referred several of the specimens
from the Bering Sea to Semper’s species while other, mostly smaller, indi-
viduals from the same area, were correctly named C. fallax Ludwig.
Genus Pentamera Ayres, 1852.
Pentamera Ayres, 1852, p. 207 ; Deichmann, 1938, p. 105.
Diagnosis : Small to medium sized forms (rarely more than 10 cm.);
ventral tentacles small; feet long, non retractile, arranged in five bands but
never scattered in the interambulacra. Calcareous ring with long posterior
prolongations on the radials. Spicules two-pillared tables, or reduced deriva-
tives of these or developed as acorn-like bodies. Feet with large end plate
and numerous supporting tables usually with well developed spire, in some
cases completely reduced. Tentacles with rods or plates, in some forms no
spicules, at least in older individuals. Spicules usually numerous, but in
some species scarce and reduced with advancing age.
Type Species: Pentamera pulcherrima Ayres.
Remarks: The diagnosis is here modified to include also Pentamera
chier chia (Ludwig) and P. zacae sp. nov., the former with few and mostly
reduced spicules, the latter with peculiar acorn-shaped bodies, as it seems
unwise for the present to segregate these two forms which otherwise con-
form well with the typical members of the genus.
The type species seems to represent the only form in the tropical west-
ern Atlantic, while P. calcigera (Stimpson) is a typical form in the northern
waters, of both Atlantic and Pacific, and a series of seven related species
are known from the west coast of North America (see Deichmann, 1938).
From the Panamic region (and Chile) four species are known.
Key to the Species of Pentamera Known from the Panamic Region
and Chile.
1. Spicules in skin acorn-shaped bodies, possibly derived from tables.
Feet with large end plate and numerous curved supporting tables with
well developed spire composed of two rods with flattened top with blunt
lobes or teeth. Color white Pentamera zacae sp. nov.
1. Spicules in skin two pillared tables or derivates of these with spire
reduced 2.
2. Spicules scarce; disk with normally four holes and spinous edge; spire
low with few teeth or reduced to two knobs. Feet with large end plate
and a varying number of spectacle-shaped rods with or without two
knobs representing the spire. Color dark brown or black
Pentamera chierchia (Ludwig).
2. Spicules numerous with well developed spire, disk with smooth edge.
Feet with large end plate and well developed supporting tables with
curved disk and tall or even excessively tall spire 3.
3. Supporting tables in feet often excessively tall. Tables small with tuft
of slender spines on top Pentamera beebei sp. nov.
3. Supporting tables in feet not excessively tall. Tables large with fairly
stout short spines on top Pentamera chiloensis Ludwig, 1887.
( Syn . Cucumaria tabulata Perrier, 1904). Type Locality: Chiloe Island,
Patagonia.
374
[XXIII :18
Zoologica: New York Zoological Society
Pentamera heebei sp. nov.
Diagnosis : Small form (few cm. long), externally and internally typical
of the genus. Spicules minute tables with mostly four larger holes and four
marginal ones; spire two pillared, ending in a tuft of slender spines. Feet
with large end plate and curved supporting tables with from 2-7 cross
beams in the spire which ends in a few flattened lobes or teeth. Color white.
Type: M. C. Z.
Type Locality: Ballena Bay, Gulf of Nicoya, Costa Rica, 40 fathoms.
General Distribution: Known from the type locality. Costa Rica.
Local Range: One small specimen from Ballena Bay, Gulf of Nicoya,
Costa Rica, 40 fathoms (Station 213 D-15).
Text-figure 7.
Pentamera heebei sp. nov. 1-6, Supporting tables of various degree of devel-
opment. 7-9, Tables from integument. Scale 1/100 mm.
Remarks: Superficially the species resembles others of the smaller pen-
tamerids, but the peculiar spicules set it apart from any other species
known. The very tall spires may possibly disappear completely but the
excessively small tables will be sufficient to distinguish it from other related
forms.
Pentamera chierchia (Ludwig).
Cucumaria chierchia Ludwig, 1887, p. 13, pi. 1, fig. 5.
Diagnosis: Small form (3-6 cm.), soft skinned with numerous cylindri-
cal feet in five bands; ventral tentacles small. Calcareous ring with long-
posterior prolongations. Spicules present in Varying number but not
crowded. Tables with mostly four central holes and dentate margin ; spire
low, two-pillared, often completely lacking; feet with large end plate and few
short spectacle-shaped rods, rarely with a trace of a spire. Tentacles with
curved perforated plates and rods. Color dark brown, almost black.
1938]
Deichmann: Holothurians
375
Text-figure 8.
Pentamera chierchia (Ludwig). 1-3, Tables with reduced disk from integument.
4, Supporting rod from appendage. Scale 1/100 mm.
Type: Possibly in Germany.
Type Locality : Coast of an island in the Gulf of Panama (shore; one
single specimen secured).
General Distribution: Probably widespread in the Panamic region, in
shallow water.
Local Range: Five from Situatanejo Bay, Mexico, shore; one from
Jasper Island, Gulf of Nicoya, Costa Rica, shore; one from Port Parker,
Costa Rica, 1.5-4 fathoms (Station 203 D-9).
Pentamera zacae sp. nov.
Diagnosis: Small form (few cm. long) with body strongly curved,
tapering toward both ends. Feet cylindrical, nonretractile, in five bands,
Pentamera zacae sp. nov. 1 & 5, Supporting tables from appendages. 2-4 & 6-8,
Acorn-shaped spicules from integument. Scale 1/100 mm.
376
Zoologica: New York Zoological Society
[XXIII :18
most numerous on the ventrum, more sparingly on the dorsum and toward
the oral and anal ends. Spicules form a crowded layer of peculiar small
bodies resembling acorns, with a tapering spire, mostly composed of two
rods and basal cup-shaped part. Feet with large end plate and numerous
supporting tables with curved disk with four holes and a small hole in
each end ; spire with two pillars ending in few flattened lobes or blunt teeth.
Color dirty white.
Type: M. C. Z.
Type Locality: Tangola-Tangola Bay, Mexico, 23 fathoms.
General Distribution: Known only from the type locality.
Local Range: Tangola-Tangola Bay, Mexico, 23 fathoms (Station 196
D-17) .
Remarks: The tentacles and calcareous ring are lacking, but neverthe-
less it can hardly be doubted that the species belongs in the genus Penta-
mera.
Genus Thyone Oken, 1815.
Remarks : The genus is extremely heterogenous and needs a revision,
but the time is hardly yet ripe for undertaking the task. The type species
is Thyone fusus (0. F. Muller), a well known form from the northeastern
Atlantic with a closely related representative in the West Indies. The two
species listed below undoubtedly deserve separate genera.
Thyone benfi Deichmann var. zacae var. nov.
Thyone benti Deichmann, 1937, p. 170, text-figs. 2, 1-11 (partim) .
Diagnosis : Like the typical form, from Puget Sound, but the disk of
the tables seems to become more quickly reduced and large heavy rods are
totally lacking in the tentacles whereas rosettes are numerous.
Type: M. C. Z.
Type Locality: East of Cedros Island, Lower California; 40 fathoms.
General Distribution: The variety is apparently restricted to the waters
of southern California and as far south as Cedros Island. The typical form
is known from Puget Sound and may be expected as far south as Point
Conception, California.
Local Range: East of Cedros Island, Lower California; 40 fathoms
(Station 126 D-17).
Remarks: The specimen in hand measures about 3 cm. and has its ten-
tacle crown and calcareous ring. It resembles otherwise the larger individual
which had lost these organs, from the same locality ( Zaca 1936), except
that the feet have just barely begun to spread out into the interambulacra.
No gonads were developed in this small individual.
Thyone gibber (Selenka).
Stolus gibber Selenka, 1867, p. 356.
Thyone gibber Deichmann, 1921, p. 205, text-fig. 4; 1936, p. 64 (passim).
Thyone similis Ludwig, 1887, p. 23, pi. 2, fig. 7; Deichmann, 1936,
p. 64 (passim).
Diagnosis: Medium sized form (5-10 cm.), robust, with thick skin
packed with spicules and numerous feet covering the entire surface of the
body; tentacles large, bushy, the two ventral smaller. Calcareous ring with
long posterior prolongations on the radials. Spicules regular four-holed
knobbed buttons, often with the central knobs united into a handle which
1938]
Deichmann: Holothurians
377
Text-figure 10.
Thyone benti Deichmann var. zacae var. nov. 1-3, Tables with spire completely
reduced. 4, Supporting rod with no trace of spire. 5, Table from introvert.
6, Rosette from tentacles. Scale 1/100 mm.
Thyone gibber (Selenka). 1-2, Knobbed buttons from integument. 3, Knobbed
button with spinous handle, from external layer of buttons. 4, Supporting
tale from appendages. Scale 1/100 mm.
378
Zoologica: New York Zoological Society
[XXIII :18
on the external side of the surface spicules may carry spines. Feet with
large end plate and large supporting tables with a low spire composed of
several twisted rods. Introvert with two-pillared tables and disk perforated
by numerous holes, almost lace-like, forming transition to rosettes. Tentacles
with heavy rods or oblong plates and numerous rosettes. Color black or
purplish-brown or white with tentacles and anterior part more or less black.
Type: M. C. Z.
Type Locality : Panama.
General Distribution: Widespread in the Panamic region, in shallow
water.
Local Range: Four from Situatanejo Bay, Mexico, shore, (Station 186
L-l) ; one from Jasper Island, Gulf of Nicoya, Costa Rica.
Remarks : From the Panamic region only one other species is known
with similar spicules, viz., T. panamensis Ludwig (1887, p. 22). The latter
seems to lack the spines on the handles of the buttons and has no heavy rods
in the tentacles. From Californian waters a smaller species, Thyone rubra
H. L. Clark, is known. It has knobbed buttons but the entire surface of the
external side of the superficial buttons is covered by a reticulum and the
supporting tables have mostly large reticulated spires.
Family Phyllophoridae.
Diagnosis: Dendrochirotes with 12-30 tentacles, either definitely ar-
ranged in an external circle with 10 large tentacles and an inner with 5 or 10
much smaller ones, or the two circles more or less confluent and the size and
number of the tentacles variable. Feet either restricted to the ambulacra or
scattered over the entire surface. They are mostly cylindrical; in some
forms the dorsal appendages are conical, more papilliform. Calcareous ring
simple or with posterior prolongations. Third mesentery attached in left
interambulacrum.
Remarks: The family is at present in a state of upheaval, as are the
other Dendrochirotes and it is therefore necessary briefly to summarize the
history to understand the viewpoint adopted here. In this connection there
are no grounds for discussing the genera in which the feet are restricted
to the ambulacra as these forms are not known from the Panamic region.
(For these species, see Ohshima, 1912, and Engel, 1933). Moreover they
seem to form well defined genera which have given little cause for trouble.
Two genera, viz., Thyonidium Diiben & Koren, 1844, and Phyllophorus
Grube, 1840, have for a long time constantly been confused and most authors
have in despair chosen to unite all the species in the oldest genus Phyllo-
phorus. The reason for this state of affairs has been that the tentacles in
the common northern species, T. pellucidum Diiben & Koren, have either
been counted as 15 or 20 — never any number between these (except occa-
sionally 16 which has been considered an abnormality) . The species showed
therefore apparently almost as wide a variation as in Phyllophorus in which
the number ranged from 12 to 20, so there seemed no valid grounds for
separating the two genera on account of the tentacles. The differences in
the type of calcareous ring seems to have been completely ignored.
Recent studies of Heding (1936, pp. 19-26) have revealed that the five
inner pairs of small tentacles in T. pellucidum are merely five tentacles
which usually are so deeply cleft that they simulate five pairs, and were
described and figured as such by Diiben & Koren (1844, p. 217, pi. 11, fig.
57), and also that the full number is reached very quickly in this species
(Heding, p. 22). The same has been found by the writer to be true of T.
commune Forbes where specimens 1 cm. long exhibit the full number of
tentacles (in this species five small inner pairs of tentacles). In Phyllo-
phorus, on the other hand, the tentacle number remains low for a long
1938]
Deichmann: Holothurians
379
period. Sars describes (1857, p. 137) how specimens of P. urna which
measure 3-4 cm. in length have only 17 tentacles, (12 external of unequal
size and 5 small inner ones), and gradually more are added. The same
irregularity in the appearance of the last tentacles is also noticed by Engel,
1933, so it seems to be a characteristic feature of the genus Phyllophorus.
In the following an attempt is made to draw the line between the
genera belonging to the Phyllophoridae with feet scattered in the interam-
bulacra. Three new genera are proposed, one to take in Phyllophorus granu-
latus Grube, one to accommodate the Tliyonidium- like forms known from the
tropical areas, and one to include the forms which have 20 tentacles in two
well defined circles, as Tliyonidium, but a calcareous ring with long posterior
prolongations, as Phyllophorus. Furthermore Selenka’s old genus Pattalus
is re-established.
Key to the Genera of Phyllophoridae with Interambulacral Feet.
1. Tentacles arranged in an external circle of five pairs of large tentacles
and an inner circle of much smaller tentacles, five deeply divided or five
pairs 2.
1. Tentacles arranged in two indistinct circles, of unequal size although
those in the inner circle are small, as are also some of those in the outer
circle. Number of tentacles varying from 12-20, frequently 16. 4.
2. Calcareous ring tall, tubular with long deeply divided posterior prolonga-
tions and long narrow interradials Neothyonidium gen. nov.
Type Species : N. liawaiense (Fisher), 1907.
2. Calcareous ring low, with long anterior teeth but no distinct posterior
tails, at most low protuberances. 3.
3. Spicules four-pillared tables which rapidly disappear with age; tenta-
cles with perforated rods or plates, but no rosettes. Arctic and boreal
forms Tliyonidium Duben & Koren, 1844.
Type Species: T. pellucidum Duben & Koren, 1844.
3. Spicules four pillared tables with spire often reduced to knobs or spines.
Spicules sometimes reduced with advancing age. Tentacles with rosettes.
Tropical forms. Euthyonidium gen. nov.
Type Species: E. seguroensis (Deichmann), 1930.
4. Calcareous ring low, simple. Spicules apparently plates, which are quickly
reduced: Large forms, 20 cm. Pattalus Selenka, 1867.
Type Species: P. mollis Selenka, 1867.
4. Calcareous ring tall, with posterior prolongations 5.
5. Feet tubular, spicules tables with two to four pillars in spire and appar-
ently not reduced with advancing age.6 Phyllophorus Grube, 1840.
Type Species : P. urna Grube, 1840.
5. Feet partly conical, papilliform on the dorsal side.
Euphyllophorus gen. nov.
Type Species: E. granulatus Grube, 1840.
6 Heding’s proposal (1936, p. 23) to transfer T. commune (Forbes) to Phyllophorus cannot
possibly be accepted. Aside from the difference in the number of inner tentacles (which may
be a variable character in T. pellucidum ), the two species differ very slightly from each other
in general features, early appearance of the full number of tentacles, etc. It is a regrettable
slip when Heding also states (p. 23) that the genus Thyonidium "usually” has many spicules,
in contrast to “Phyllophorus” in which they disappear. Both T. pellucidum and T. commune
lose their spicules very rapidly — hence the difficulty in distinguishing between the two forms (see
among others Mortensen, 1927, pp. 411 and 413). But in no case is it known that the spicules
disappear completely in any member of Phyllophorus. Sars says that the tables often are scarce
in the adult individuals of P. urna (1857, p. 138) ; Theel, 1886, p. 150, mentions the presence
of end plate and supporting rods in the largest individuals while tables are practically lacking.
Usually careful examination of fragments of the integument reveals the presence of some tables.
More important, however, is the entirely different type of calcareous ring which is characteristic
of the members of the genus Phyllophorus.
380
Zoologica: New York Zoological Society
[XXIII :18
Euthyonidium gen. nov.
Diagnosis : Medium sized forms (10 cm. long) with fairly robust feet
scattered over the entire body, rarely with trace of being arranged in
bands. Tentacles five large pairs in an external circle and five small pairs
in an inner circle, often contracted so they are difficult to observe. Cal-
careous ring simple or with insignificant posterior protuberances. Spicules
tables or derivatives of tables with spire reduced, often scarce in older indi-
viduals. Feet with large end plate and a few supporting rods. Tentacles
with delicate rods and rosettes.
Type Species : Euthyonidium seguroensis (Deichmann).
Remarks : The genus represents the tropical counterpart of the northern
genus, Thyonidium Diiben & Koren, 1844. Two species are known from the
West Indies, viz., the type species and E. occidentalis (Ludwig). Re-exami-
nation of Selenka’s Thyone ovulum from Acapulco, Mexico, showed a complete
circle of five pairs of small inner tentacles, while a microscopical examination
of the skin revealed the presence of a few tables with reduced spire — of
exactly the same shape as those found in an Euthyonidium brought home by
the Zaca.
Euthyonidium ovulum (Selenka).
Stolus ovulum Selenka, 1867, p. 365, pi. 20, fig. 117.
Thyone ovulum, Deichmann, 1936, p. 64.
Diagnosis: Agrees in exterior with diagnosis given for the genus.
Calcareous ring with radials posteriorly deeply incised, and with long an-
terior tooth; interradials heart-shaped with posterior margin slightly in-
cised. Numerous free stone canals and Polian vesicles. Spicules a scattered
layer of tables with cross-shaped disk with spinous or branching edge and
spire in most cases reduced to 1-4 spines. Feet with large end plate and
apparently few supporting rods; tentacles with few rosettes and delicate
rods. Spicules more or less completely reduced with advancing age and
except for the large end plate, easily overlooked. Color reddish-brown.
Type: M.C.Z.
Type Locality: Acapulco, Mexico.
General Distribution: Probably widespread in the Panamic region.
Local Range: One specimen from Port Parker, Costa Rica, 1.5-4 fathoms
(Station 203 D-9).
Remarks: The Zaca specimen, which decidedly is a polychirote form,
showed so remarkable a similarity with Thyone ovulum that a re-examina-
Text-figure 12.
Euthyonidium ovulum (Selenka). 1-3, Plates with reduced spire from integu-
ment. 4-5, Supporting rods from appendages. (From Selenka’s type,
M. C. Z.) Scale 1/100 mm.
1938]
Deichmann: Holothurians
381
tion of Selenka’s material was undertaken and the presence of five pairs of
small tentacles demonstrated as well as the presence of a few tables with
reduced disks of the same type as those found in the Zaca specimen. “Thyone
ovulum” must therefore be transferred to Euthyonidium. The Zaca material
represents the first record since the types were described.
Genus Phyllophorus Grube, 1840.
Diagnosis : Medium sized forms (rarely more than 10 cm. long) with
robust cylindrical feet distributed over the entire surface and with 12-20
tentacles of different size in two indistinct circles; the last tentacles appear
at intervals after the animal has reached a considerable size. Calcareous
ring with distinct long posterior prolongations on the radials; interradials
shorter or longer and often overlapping the radials. Spicules delicate tables ;
feet with end plate and sometimes a few oblong, curved supporting tables;
tentacles with few delicate rods and rosettes. Spicules apparently preserved
throughout the animal’s entire life.
Type Species : Phyllophorus urna Grube.
Remarks: Defined this way the genus comprises the Mediterranean
P. urna Grube and five of the seven species known from the tropical western
Atlantic, some species from the Indo-Pacific (see Engel, 1933) and two
species from the Panamic region, including the species described below.
From the west coast of North America no Phyllophoridae are known (except
Thyonidium commune Forbes; unpublished record, material in U.S.N.M.)
nor has any been reported so far from the coast of Chile (except P attains).
Two Phyllophoridae were described in 1907 from Hawaii by Fisher.
Not included is P. granulatus Grube from the Mediterranean which
seems to differ in so many respects (papillae on the dorsal side and reticu-
lated plates, etc. [see Koehler, 1927, p. 199]), that it unquestionably de-
serves its own genus for which the name Euphyllophorus is proposed.
Key to the Species of Phyllophorus Known from the Panamic Region.
1. Spicules tables with oval disk with smooth edge and 8-12 small marginal
holes; spire with four pillars and ending in wreath of 8-12 spines. Feet
with end plate and supporting tables with an enormous four-pillared
spire ending in a long conical tooth which perforates the skin.
Phyllophorus aculeatus Ludwig.
(Panama, shallow water. Only the type specimen known, U.S.N.M.;
Ludwig, 1894, p. 128, pi. 13, figs. 6-11).
1. Spicules tables with oval disk with four large and four smaller marginal
holes, the latter more or less wedge shaped. Spire with two pillars and
ending in four tufts of few spines. Feet with large end plate and few
supporting tables with oval curved disk and low two-pillared spire ending
mostly in two diverging spines Phyllophorus zacae sp. nov
Phyllophorus zacae sp. nov.
Diagnosis: Medium sized form with 15 (probably up to 20) tentacles of
unequal size, in confluent circles. Feet stout, distributed over the entire
body, not crowded. Calcareous ring with well developed posterior prolonga-
tions on the radials; anteriorly the radials have a long rectangular tooth
with a constriction at its base; interradials with broad base and short
anterior tooth, and overlapping the radials. Spicules delicate tables with
oval disk with four large and four smaller marginal holes, often wedge
shaped, spire with two pillars, ending in four tufts of few spines. Feet with
large end plate and with few oblong supporting tables with mostly oval disk
382
Zoologica: New York Zoological Society
[XXIII :18
with numerous holes and a two-pillared spire ending mostly in two diverging
horns or teeth. Tentacles with few delicate rods and rosettes. Color reddish.
Type: M.C.Z.
Type Locality : Tangola-Tangola Bay, Mexico, 10 fathoms.
General Distribution : Known only from the type locality.
Local Range : One contorted specimen from Tangola-Tangola Bay,
Mexico, 10 fathoms (Station 196 D-13).
Text-figure 13.
Phyllophorus zacae sp. nov. 1-3, Tables from integument. 4, Supporting table
from appendages. 5, Rosette from tentacles. Scale 1/100 mm.
Remarks: The species is closely related to Phyllophorus destichadus
Deichmann (1930, p. 146, pi. 18, fig. 3) from the West Indies. It differs in
the presence of supporting tables in the feet and in minor details of the
calcareous ring and the spicules.
From Hawaii Fisher (1907, p. 712, pi. 79, figs. 2, 2a-c) has described
a species with similar spicules but with the tentacles arranged in two dis-
tinct circles and of pronouncedly different size and with a very different
calcareous ring — with long narrow interradials. For that species the genus
Neothyonidium has been proposed.
For diagnosis, etc., see Deichmann, 1937, p. 172, text-fig. 3.
Local Range: Two specimens from Situatanejo Bay, Mexico, shore;
one from Jasper Island, Gulf of Nicoya, Costa Rica, shore.
Remarks: The specimens are larger and more robust than the single
type specimen secured in 1936 from Arena Bank, Gulf of California. The
present records indicate that the species occurs widespread in the Panamic
region, as was to be expected.
L—J L—l L-l L._! — L 1 1
Family Psolidae.
Genus Thyonepsolus H. L. Clark, 1901.
Thyonepsolus beebei Deichmann.
Order Molpadonia.
Family Caudinidae.
Genus Paraeaudina Heding, 1931.
Caudina Auctores.
Pseudocaudina Heding, 1931, p. 283.
Paraeaudina Heding, 1931, p. 455; Clark, 1935, p. 267.
1938]
Deichmann: Holothurians
383
Diagnosis: Caudinids of medium size or large (up to 20 cm.) with 15
tentacles with two pairs of digits but no unpaired terminal digit. Calcareous
ring with short undivided posterior prolongation on the radials. Third loop
of intestine supported by powerful pseudomesenteries attached to the lateral
body walls while the normal ventral mesentery is reduced or lacking. Spicules
small cross-cups or four-holed plates with spinous to lobate margin, with age
degenerating into irregular deformed plates. No phosphatic bodies present
but older individuals may have certain spicules surrounded by a reddish
substance. Color white. Shallow water forms.
Type Species: Paracaudina chilensis (J. Muller).
I — I l LJ ! I I L_J 1
Text-figure 14.
Paracaudina chilensis (J. Muller). 1-4, Cross-cups from Zaca specimen, from 35
fathoms depth.
Remarks: The number of species belonging to this genus has been the
subject of much discussion. Clark (1935) holds that there are two Aus-
tralian species with fairly simple spicules while the type species can be
divided into a number of varieties which differ merely in the development
of the spicules. One argument against uniting the widespread forms under
one name has been the discontinued distribution of the species. It was first
in 1938 that some specimens were taken between Chile, the type locality, and
California, viz., off Guatemala and Mexico, (Deichmann, 1938, p. 23, text-
figure) and the zoogeographic gap thereby closed. The Zaca expedition has
furthermore secured two specimens from off Costa Rica, an indication that
P. chilensis is by no means uncommon in the Panamic region.
Paracaudina chilensis ( J. Muller) .
Molpadia chilensis, J. Muller, 1850, p. 139; 1854, pi. 4, fig. 14, pi. 9,
fig. 1.
Caudina chilensis, H. L. Clark, 1908, p. 175.
Paracaudina chilensis, H. L. Clark, 1935, pp. 267-284. (discussion of the
species problem). Deichmann, 1938, p. 23, text-fig.
Diagnosis: As for the genus. Spicules numerous cross-cups with
marginal projections, shorter or longer, often as rounded lobes; reduced to
simple plates in older individuals.
Type: Museum of Berlin.
Type Locality: Coast of Chile.
General Distribution: From the Strait of Magellan, along the west
coast of South, Central and North America to Japan, Coasts of China and
North Australia — varieties chiefly with long marginal projections on the
cross-cups; from New Zealand and West Indies — varieties chiefly with
384
Zoologica: New York Zoological Society
[XXIII :18
Paracaudina chilensis (J. Muller). Spicules from specimens from off Guate-
mala, 10 fathoms (upper row) and from off Mexico, 7 fathoms (lower
row). Reproduced from Deichmann, 1938.
reduced marginal projections (this type may also occur in numbers in speci-
mens from the west coast of Central America). Shallow water to about 40
fathoms.
Local Range: Two specimens, 4 and 5 cm., from Ballena Bay, Gulf of
Nicoya, Costa Rica, 35 and 40 fathoms (Station 213 D-12 and D-15).
Remarks : The spicules selected at random from the Zaca material shows
the extreme variability of the calcareous spicules in this species when con-
trasted with the figures of spicules from the specimens from Guatemala and
Mexico, from respectively 10 and 7 fathoms depth.7 Possibly the greater
depth from which the Zaca material came may account for the variation of
the spicules.
Bibliography.
Ayres, W. 0.
1852. Description of a new species of Holothuria. Proc. Boston Soc. Nat.
Hist., vol. 4, pp. 207-208 (June, 1852).
Bell, F. J.
1887. Studies in the Holothuroidea, VI. Descriptions of new species. Proc.
Zool. Soc., pp. 531-534, pi. 45.
Clark, H. L.
1898. Notes on the Echinoderms of Bermuda. Ann. N. Y. Acad. Sci., 11, no.
19, pp. 407-413.
1899. Further Notes on the Echinoderms of Bermuda. Ibid, 12, no. 7, pp.
117-138, pi. 4.
1902. Papers from the Hopkins Stanford Galapagos Expedition, 1898-99, 12,
Echinodermata. Proc. Wash. Acad. Sci., pp. 521-531. (Washington,
D. C.)
1908 The Apodous Holothurians. Smithsonian Contributions to Knowledge,
35, no. 1723, pp. 1-231, pis. 1-13.
7 Record of depth omitted in the original report (Deichmann, 1938).
1938]
Deichmann: Holothurians
385
1913. Echinoderms from Lower California with Descriptions of New Species,
Albatross cruise 1911. Bull. Amer. Mus. Nat. Hist., 32, pp. 185-236,
pis. 44-46.
1920. Holothurioidea. Mem. Mus. Com-p. Zool. 39, pp. 121-154, pis. 1-4.
1922. The Holothurians of the genus Stichopus. Bull. Mus. Comp. Zool., 65,
no. 3, pp. 39-74, pis. 1-2.
1923. Echinoderms from Lower California with Descriptions of new species,
Supplementary Report . . . Albatross cruise 1911. Bull. Amer. Mus.
Nat. Hist., 48, pp. 147-163.
1935. The Holothurian Genus Caudina. Ann. Mag. Nat. Hist., ser. 10, vol.
15, pp. 267-284.
1938. Echinoderms from Australia. Mem. Mus. Comp. Zool., 55, pp. 1-596,
pis. 1-28.
Deichmann, E.
1922. On some cases of Multiplication by Fission, etc. Papers from Dr. Th.
Mortensen’s Pacific Expedition, 1914-1916. Vid. Med. Nat. For., 73,
pp. 199-215, text-figures (Copenhagen).
1930. The Holothurians of the Western Part of the Atlantic Ocean. Bull.
Mus. Comp. Zool., 71, no. 3, pp. 43-226, pis. 1-24.
1936. A new species of Thy one from the West Coast of Mexico. Proc. New
England Zool. Club, 15, pp. 63-66, text-figure.
1937. The Templeton Crocker Expedition. IX. Holothurians from the Gulf
of California, the West Coast of Lower California and Clarion Island.
Zoologica, New York Zoological Society, 22, pt. 2, pp. 161-176, text-
figures 1-3.
1938. New Holothurians from the Western Coast of North America and
some Remarks on the Genus Caudina. Proc. New England Zool. Club,
16, pp. 103-115, 4 text-figures.
1938a. New Records of Paracaudina chilensis (J. Muller) from the West
Coast of Central America and Mexico. Proc. New England Zool. Club,
17, pp. 23-25, text-figure.
Duben, M. W. & Koren, J.
1844. Ofversigt Skandinaviens Echinodermer. Kgl. Vid. Akad. Handl.,
1844 (1846), pp. 229-328, pis. 6-11.
Edwards, C. E.
1910. Four Species of Pacific Ocean Holothurians allied to Cucumaria
frondosa (Gunnerus). Zool. Jahrb. Abt. Syst., 29, pp. 597-612, pi. 19.
Engel, H.
1933. Holothuries; Resultats Scientifiques du Voyage aux Indes Orienta-
les Neerlandaises LL. AA. RR. le Prince et la Princesse Leopold de
Belgique. Verhandelingen van het Koniklijk Natuurhistorisch Museum
van Belgie (Buiten Reeks) vol. Ill, Fascicule 13, pp. 1-42, pi. 1, text
figures 1-25, 4 charts.
Fisher, W. K.
1907. The Holothurians of the Hawaiian Islands. Proc. U. S. Nat. Mus., 32,
pp. 637-744, pis. 66-82.
Forskal, P.
1775. Descriptiones animalium . . . etc. Post mortem auctoris edidit Carsten
Niebuhr, pp. 1-164, 1 map. (Hauniae).
Grube, A. E.
1840. Actinien, Echinodermen und Wiirmer des Adriatischen und Mittel-
meers nach eigenen Sammlungen beschrieben, pp. 1-92, pi. 1.
(Konigsberg) .
386
Zoologica: New York Zoological Society
[XXIII :18
Heding, S. G.
1931. On the Classification of the Molpadia. Vid. Med. Nat. Hist. For. 92,
pp. 275-284. Correction of Pseudocaudina to Paracaudina, ibid, pp.
455-456.
1933. The Caudina of Asamushi. Sci. Rep. Tohoku Imp. Univ. (4), Biol. 8,
no. 2, pp. 127-243, pis 5-8. (Sendai, Japan).
1936. Echinoderms, 6 & 7 Thule Expedition. Medd. Gronland, 108, no. 1,
pp. 1-34, text-figs. 1-6.
Koehler, R.
1927. Les Echinodermes des Mers d’Europe, II, Encylopedie Scientifique,
pp. 1-339, pis. 10-18. (Paris).
Lampert, K.
1885. Die Seewalzen, Holothuroidea, Eine Systematische Monographic, pp.
1-310, pi. 1. (Wiesbaden).
1896. Die von Dr. Stuhlmann in den Jahren 1888 und 1889 an der Ostkfiste
Afrikas gesammelten Holothurien. Mit. Mus. Hamburg, 13, pp. 49-71,
text-figures.
Lesson, R. P.
1830. Centurie zoologique, pp. I-X, 1-244, pis. 1-80. (Paris).
Ludwig, H.
1874. Beitrage zur Kentniss der Holothurien mit Nachtrag. Arbeiten aus
d. Zool.-Zoot. Institut in Wurzburg, 3, Heft. 2, pp. 77-120.
1887. Die von G. Chierchia auf der Fart der Kgl. Ital. Corvette “Vettor
Pisani” gesammelten Holothurien. Zool. Jahrb., 2, pp. 1-36, pis. 1-2.
1894. Holothurioidea; Report on an Explor., etc., Albatross. Mem. Mus.
Comp. Zool., 17, pp. 1-183, pis. 1-19.
1898. Holothurien. Hamburger-Magelhaensische Sammelreise, pp. 1-98,
pis. 1-3.
Mortensen, Th.
1927. Handbook of the Echinoderms of the British Isles, pp. 1-471, text-
figures 1-269. (Oxford University Press).
Ohshima, H.
1912. On the system of Phyllophorinae with descriptions of the species found
in Japan. Annot. Zool. Jap. Tokyo, 8, 1912, pp. 53-96, pi. 1, text-
figures 1-7.
Muller, J.
1850. Anatomische Studien fiber die Echinodermen. Archiv fur Anatomie
und Physiologie, pp. 115-155.
1854. Uber den Bau der Echinodermen, pi. 9, figs. 1, 1+.
Panning, A.
1929. Die Gattung Holothuria. I. Mitt. Zool. Staatsinstitut u. Zool. Mus.
Hamburg, 44, pp. 91-138, text-figures 1-21.
1934. Die Gattung Holothuria, II & III. Ibid, 45, pp. 24-50, 65-84, text-
figures 22-43, 44-70.
1935. Die Gattung Holothuria, IV & V. Ibid. 45 & 46, pp. 85-107, 1-18, text-
figures 71-102, 103-121.
Parker, G. H.
1921. The locomotion of the Holothurian Stichopus parvimensis H. L.
Clark. Journ. Exp. Zool., 33, pp. 205-208. (Philadelphia).
Perrier, R.
1904. Holothuries du Cap Horn. Bull. Mus. d’Hist. nat., 10, pp. 13-16.
1905. Holothuries antarctiques. Ann. Sci. Nat., 9 ser., 1, pp. 1-146, pis. 1-5,
text-figures A-M. (Paris).
1938]
Deichmann: Holothurians
387
Sars, M.
1857. Bidrag til Kundskaben om Middelhavets Littoral-Fauna. Nyt Mag.
Naturvidenskab, pp. 57-155, pis. 1-2.
Selenka, E.
1867. Beitrage zur Anatomie und Systematik der Holothurien. Zeitsch.
TFiss. Zool., 38 pp. 291-374, pis. 17-20.
Semper, C.
1868. Reisen im Archipel der Philippinene, Pt. 2, vol. 1, Holothurien, pp.
1-288, pis. 1-40.
Sluiter, C. P.
1901. Die Holothurien der Siboga Expeditie, Mon. 44. Uitkomst H. M.
Siboga. . . . uitgeven van Max Weber, pp. 1-142, pis. 1-10. (Leiden).
1910. Westindische Holothurien. Zool. Jahrb., Suppl. 11, pp. 331-342, text-
figures A-F.
Stimpson, W.
1857. On the Crustacea and Echinodermata of the Pacific Shores of North
America. Journ. Boston Nat. Hist., 6, pp. 444-532, pis. 18-23.
Theel, H.
1886. Report on the Holothurioidea. Report on the Scientific Results of the
Voyage of H.M.S. “ Challenger ” during the years 1873-76, pt. 39,
Zoology, 14, pp. 1-290, pis. 1-16.
1886a. Report on the Holothurioidea, Report on the Results of Dredgings by
U. S. Coast Survey Steamer “Blake” Bull. Mus. Comp. Zool., 13, pp.
1-21, pi. 1.
Conant: Note on Leioheterodon madagascariensis
389
19.
A Note on Eggs and Young of Leioheterodon madagascariensis
(Dumeril & Bibron).
Roger Conant
Zoological Society of Philadelphia
(Plate I).
The life histories of the great majority of reptiles are so poorly
known that even the most fragmentary observations may be of value. Cer-
tainly herpetology, which has made extraordinary gains in the past few
decades, is still far behind other fields, especially ornithology, when the
lack of details on the sizes of eggs and young, etc., is considered. While
I have no particular interest in the fauna of Madagascar, and scarcely
know its literature, I publish this brief note in the belief that it may help
to bridge the gap in our knowledge of an interesting snake. Monsieur F.
Angel, of the Paris Museum, the principal worker on the Malagasy herpeto-
fauna, in response to an inquiry, writes under date of November 13, 1937,
“Je ne connais pas de travaux particulars traitant de reproduction des
Lioheterodon.”
A large female Leioheterodon madagascariensis arrived at the Phila-
delphia Zoological Garden on May 11, 1937, in an exchange shipment from
the London Zoological Gardens. On the morning of July 18, 1937, it was
discovered coiled about 13 eggs which it had laid during the night. It did
not move while photographs were being taken, and only slightly shifted
its position when it and the eggs, which were laid on the cage floor, were
covered with damp peat moss. However, it left them and lay in the rear
of the cage during the afternoon, but returned to its former position in the
evening. The next morning it moved away from the eggs again and,
possibly as a result of being accidently disturbed, did not approach them for
many hours. They were removed, therefore, to an aquarium and buried
in a mixture of slightly damp, rotten wood and peat. In this mixture they
were kept, at ordinary room temperatures, until they hatched.
The eggs were adherent to one another and grouped in a single cluster.
The shells were white, parchment-like and yielded slightly to the pressure
of one’s fingers. Small, hard nodules, slightly raised above the surrounding
surfaces of the eggs, were scattered over them.
Owing to their positions in the cluster it was impossible to measure
all of the eggs. The lengths, in five of them, varied from 46.1 to 50.0 and
averaged 47.8 mm. ; the widths, in seven, varied from 29.9 to 34.1 and
averaged 32.3 mm. Their weight, as a unit, was 376.1 grams, or an average
of 28.9 grams for each egg. The female was 1,483 mm. in length and
weighed 926.6 grams. (All weights and measurements recorded July 20,
1937).
The first indications of hatching were observed early during the after-
noon of October 16, 1937, when it was noticed that the shell of one of the
390
Zoologica: New York Zoological Society
[XXIII :19
eggs had a small slit in it. Protruding from this opening was the snout
of a young snake, surrounded by a mass of bubbles resembling spittle.
Two hours later snouts were found protruding from two other eggs; five
heads were out early in the morning of October 17 and a total of seven
were visible the evening of October 18. At this time, however, none of the
young snakes had yet left their shells. They seemed very wary, and even
when their heads were entirely out they withdrew them inside their shells
when an observer approached. Even passing one’s hand over the aquarium
or turning on a photo-flood lamp caused them all to retreat back into the
eggs. Masses of bubbles, such as the one described above, were seen
around or near each head. Some of the eggs had two or more slits in them.
The cluster of eggs and the hatching medium were examined in detail
the afternoon of October 19. Two young had escaped from the shells and
buried themselves in the peat. Another was out by early evening and by
the next afternoon a total of six had emerged. All were very lively, sought
to burrow in the peat when they were handled and one expanded its neck
in a manner very similar to that employed by Heterodon contortrix when
alarmed. All of them strongly resembled the female in coloration and pat-
tern except that their lighter markings were more vivid.
The last of the 13 little snakes left its shell shortly after noon, October
22, and the entire brood was weighed and measured soon afterward. The
figures obtained are as follows:
Snake
Weight
Length
1.
18.2 grams
324 millimeters
2.
19.1
<<
335
3.
18.3
U
336
4.
20.3
u
343
5.
19.0
it
. 337
6.
18.7
u
339
7.
16.2
“
314
8.
18.2
u
343
9.
19.4
«
348
10.
18.1
u
337
11.
19.5
<<
338
12.
19.8
342
13.
18.0
u
334
Average
18.68
«
336
With a single exception the eyes of all of the little snakes were over-
cast, indicative of an approaching moult. In the snake with clear eyes,
and one of the others, egg teeth were still in place; all the rest had lost
them. This was the only time during the entire hatching period that any
egg teeth were observed, although the snouts protruding from all the top-
most eggs were examined several times with a lens. The two young in
which egg teeth were seen must have come from eggs lower in the cluster
and must have been among the last to hatch. One specimen shed its skin
on October 27, two on October 28, seven on October 29, two on October 30
and one on October 31.
Attempts to feed the small snakes were almost 100% unsuccessful.
They consistently refused to eat the small frogs, baby mice, earthworms,
insects, etc., which were put in their cage, except as indicated in the fol-
lowing notes: One snake seized a frog by the head but let go and lost
1938]
Conant: Note on Leioheterodon madagascariensis
391
all interest when the frog kicked its legs. Specimens refused the legs of
freshly killed frogs. Two started to swallow them when they were
forced into their mouths but rejected them almost immediately there-
after. One snake chased a young mouse and caught it by the tail. It
attempted to constrict it, but instead coiled around its own body. The
mouse was removed, killed, returned to the cage and wiggled by the
fingers. The snake constricted it and started to swallow it tail first but
gave up the job before the mid-point of the body was reached. One little
snake pursued a small, green frog, Rana clamitans, seized it, but let it go
at once. Another snake caught it and swallowed it.
In contrast to the lack of interest in food displayed by the young
snakes, the adult female has thrived in captivity and has accepted several
kinds of animals as food. She has eaten Rana clamitans, Rana pipiens,
Rana catesbeiana, white mice, rats and baby chicks. Small prey she seizes
and swallows as it struggles; larger animals are constricted.
The several specimens of Leioheterodon were noticeably similar to
Heterodon in several respects. The keeled and turned up rostral, the robust
body and the habit of flattening the neck (noticed in the female and several
young of the group) are all suggestive of the hog-nosed snakes. In no
case, however, did the specimens of Leioheterodon keep their necks in the
flattened position for any length of time, as is such a common habit with
Heterodon. They returned to their normal positions almost at once. This
may have been due to captivity, however, for in Heterodon captive speci-
mens usually fail to perform after one or two demonstrations.
I wish to express my indebtedness to Mr. Arthur Loveridge, Dr. Howard
K. Gloyd and Dr. E. R. Dunn for helpful suggestions during the preparation
of this manuscript.
Summary.
1. Thirteen eggs, averaging 47.8 X 32.3 mm. in length and width,
and 28.9 grams in weight, were laid by a specimen of Leioheterodon mada-
gascariensis in the Philadelphia Zoological Garden, July 18, 1937.
2. The eggs started hatching 90 days later and the last of the brood
left its shell 96 days after the eggs were laid. The young snakes averaged
336 millimeters in length and 18.68 grams in weight.
3. The young snakes showed little interest in food although one ate
a frog and one started to eat a mouse. The female ate frogs, rats, mice
and chicks.
392
Zoologica: New York Zoological Society
EXPLANATION OF THE PLATE.
Plate I.
Fig. 1. Female Leioheterodon madagascariensis coiled around her clutch of
thirteen eggs a few hours after they were laid.
Fig. 2. Eggs of L. madagascariensis hatching. The snouts of two of the baby
snakes may be seen protruding from the shells. A third is visible be-
neath the slit in the shell of the right-hand egg. Photographs by Mark
Mooney, Jr.
CONANT.
PLATE I.
FIG. 1.
FIG. 2.
A NOTE ON EGGS AND YOUNG OF LEIOH ETERODON MADAGASCARIENSIS
(DUMERIL & BIBRON )
Breder: An Unusual Aberrantly Colored Pleuronectid
393
20.
An Unusual Aberrantly Colored Pleuronectid.
C. M. Breder, Jr.
New York Aquarium.
(Plate I).
Abnormal coloration in the Heterosomata has been under study by
Gudger and Firth for some time and in their records they have included
an analysis of the earlier literature. See Gudger (1934 and 1935) and
Gudger and Firth (1935, 1936a, b, c, and 1937). The case here recorded
(A.M.N.H. 14142) differs from any described by them in certain interesting
respects.
This specimen, a nearly ripe female, Pseudopleuronectes americanus
(Walbaum), measures 195 mm. in standard length and 244 mm. in total
length. It was caught by Mr. William Stratton near Bayside, Long Island,
on November 8, 1938, and sent to the New York Aquarium through the
courtesy of Mr. Fred Fletcher.
The abnormality in coloration amounts to a complete absence of
pigmentation on the upper side of the body. The vertical fins, however,
are normally pigmented, although in the state of preservation received
(iced) seemed to represent a pale phase, a feature not uncommon to iced
fish in a fairly soft condition. The pectoral fin of the upper side is pig-
mented, as are both the pelvics, while the pectoral of the lower side is with-
out pigment as is normal in this species. All the pigmented fins are en-
croached upon with varying pigmentless areas at their bases.
The head is also without pigment on the upper side, except for the pro-
truding parts of the eyes and a small triangle between and ahead of them,
which, however, does not reach the lips. The eyes themselves are normally
pigmented. This fish, then, represents a reversal of the more usually
reported abnormal color differential in which the dark pigment of the upper
side appears on the normally white lower surface. There is here a pretty
problem in the dynamics of coloration involving the inverse of those dark
on the normally white lower side.
There is no evidence of a hooked dorsal or other morphological anomaly
sometimes associated with atypical coloration in these fishes. There is,
however, a very evident traumatic peculiarity. As though it were not enough
alone for this specimen to grow to maturity, it clearly survived and healed
perfectly what appears to have been a bite out of the middle of its back.
Such healed injuries are common enough in fishes generally, there usually
being one or more such specimens present in the collections of the New
York Aquarium. These arrive along with other fishes and are gener-
ally not placed on exhibition for obvious reasons. This particular case is,
however, of unusual interest since it raises the question of the protective
value of flounder coloration. Specimens of this species, along with flounders
generally, are extremely inconspicuous as they lie on the sea floor. Not
394
Zoologica: New York Zoological Society
[XXII :20
only do they blend well with the bottom colors, but actually go through
extensive color and pattern changes in which the tone and texture of their
surroundings are very closely approximated. See, for example, Sumner
(1911) and Mast (1916). In addition to, or in spite of this, they further
hide themselves, when the nature of the bottom permits, by causing a
flurry in the sand or mud in which they then sink so that only their eyes
protrude. Such being the case, it might be inferred that since the present
individual was white on the top side, it became an easy mark. To offset
this conspicuousness it may be imagined that it buried itself deeper than
others so that most of the time only its pigmented eyes protruded. How-
ever, as a matter of fact, the greatest hazard to these flat fish occurs when
they are swimming, for even the normally colored ones can then be easily
seen, and, indeed, even when buried they are frequently caused to swim
by larger fishes rooting in the bottom, presumably to dislodge them.
This sketch of flounder life and hazards is given merely as a back-
ground against which to point out that in spite of the normal hazards of
all flatfish, this one without the alleged protection of changeable coloration
and the fact that something did bite a chunk out of it, nevertheless grew
to adulthood. It will be noted that this is all that is necessary for survival.
This fish may have spawned last winter, or at least could have this winter if
it had not suffered the fate of so many Long Island flounders — one not in
the least associated with any of its individual peculiarities.
Lest it be thought that perhaps its injury was in some way connected
with a decoloration of the upper surface, it may be pointed out that the
writer, as already indicated, is more than usually familiar with injuries
of this sort because of circumstances, and that in no case known to him is
there any instance where abnormal pigmentation followed even vastly great-
er injuries. See, for example, figures of other similar injuries in Breder
(1925 and 1934). It may be noted in Plate 1, along with the other items
discussed, that the pigmentation of the fin adjacent to the wound is normal
and in the pigmentless areas scales have covered the area of the wound.
References.
Breder, C. M., Jr.
1925. Tailless Pearl Roach. Bull. N.Y.Z.S. 28 (3) : 72-74.
1934. The Ultimate in Tailless Fish. Bull. N.Y.Z.S. 37 (5) : 141-145.
Gudger, E. W.
1934. Ambicoloration in the Winter Flounder, Pseudopleuronectes amer-
icanus. Amer. Mus. Novitates (717) : 1-8.
1935. Abnormalities in Flatfishes, (Heterosomata). Journ. Morph. 58 (1) :
1-39.
Gudger, E. W. & Firth, F. E.
1935. An almost totally ambicolorate halibut, Hippoglossus hippoglossus,
with partially rotated eye and hooked dorsal fin — the only recorded
specimen. Amer. Musi Novitates (811): 1-7.
1936a. Three partially ambicolorate four-spotted flounders, Paralichthys ob-
longus, two each with a hooked dorsal fin and a partially rotated eye.
Amer. Mus . Novitates (885) : 1-9.
1936b. A reversed, almost wholly ambicolorate summer flounder, Paralichthys
dentatus. Amer. Mus. Novitates (896) : 1-5.
1936c. Ambicoloration, partial and complete in the southern flounder, Para-
lichthys lethostigma. Amer. Mus. Novitates (897) : 1-7.
1937. Two reversed, partially ambicolorate halibuts, Hippoglossus hippoglos-
sus. Amer. Mus. Novitates (925) : 1-10.
1938]
Breder: An Unusual Aberrantly Colored Pleuronectid
395
Mast, S. O.
1916. Changes in shade, color and pattern in fishes and their bearing on
certain problems of behavior and adaptation. Bull. U. S. Bureau
Fisheries 34: 173-238 (1914).
Sumner, F. B.
1911. The adjustment of flat-fishes to various backgrounds. A study of adap-
tive color change. Journ. Exper. Zool. 10 (4) : 409-505.
396
Zoologica: New York Zoological Society
EXPLANATION OF THE PLATE.
Abnormal Pseudopleuronectes americanus.
Fig. 1. Eyed or upper side.
Fig. 2. Blind or under side.
BREDER.
PLATE I.
FIG. 2.
AN UNUSUAL ABERRANTLY COLORED PLEURONECTID.
Schroeder : Test for Tuberculosis in Primates
397
21.
A Diagnostic Test for the Recognition of Tuberculosis
in Primates; a Preliminary Report.
Charles R. Schroeder
Veterinarian, New York Zoological Park.
(Plates I & II).
A practical and simple test which might be applied to captive Primates
for the early recognition of tuberculous infection is urgently needed (Schroe-
der, 1938). Tuberculosis was recognized as a disease to be coped with in the
late 1890’s in many zoological gardens. Recently the wholesale use of the
Rhesus Macaque ( Macaca mulatta ) by laboratories has further shown the
immediate need for a test applicable to large groups of newly imported
monkeys. Penrose, White, Brown and Pearson, of the Penrose Laboratory of
the Philadelphia Zoological Society, adapted to monkeys the test used in
dairy cattle by the Bureau of Animal Industry (Fox, 1923). The test was
used successfully at the Philadelphia Zoological Garden by these early work-
ers. Their successful results were continued in the hands of White and Fox
(White & Fox, 1909). Alterations and improvements in the test were made
which depended upon a thermal response to an injection of K.O.T. The inter-
esting discovery by early workers, indicating that Primates have a wider
diurnal temperature variation than most other animals, made it imperative
to record pre-injection temperatures so that an increase subsequent to the
injection of tuberculin could be properly evaluated against the pre-injection
graph (Simpson & Galbraith, 1906).
The Bureau of Animal Industry early recognized the necessity of alter-
ing its test, which was dependant on a regular systemic temperature rise
following the subcutaneous administration of tuberculin. It was necessary to
simplify it and make unnecessary the detailed taking and recording of tem-
peratures on which depended the accuracy of the test. They adopted the
intradermal test, at first in conjunction with the ophthalmic installation of
tuberculin, later with a dual intradermal test (caudal fold and vaginal labia),
and finally with a single caudal fold intradermal injection which has given
best results because of its simplicity. The subcutaneous test is still admittedly
accurate.
European investigators have done some work but nothing on a suffi-
ciently large scale to be conclusive. The complete bibliography of Primate
tuberculosis prepared by Dr. Margaret A. Kennard of Yale University School
of Medicine gives no clew to a satisfactory test, in Europe or elsewhere, for
laboratory animals.
A modification of the Bureau of Animal Industry’s intradermal test in
cattle has been evolved in the New York Zoological Park for use in monkeys.1
l The author wishes to acknowledge the assistance given by Dr. N. D. C. Lewis and Dr. S. E.
Barrera of the New York State Psychiatric Institute for permitting the routine testing of incoming
Rhesus monkeys ; Dr. Herbert Fox of the Penrose Laboratory of the Philadelphia Zoological Society,
Dr. Jacob Traum of the University of California, Dr. A. R. Dochez and Dr. F. S. Cheever of the
Presbyterian Hospital, College of Physicians and Surgeons, Dr. Margaret A. Kennard of the Yale
University School of Medicine and Dr. Clarence Carpenter of Columbia University for their timely
aid and cooperation ; and Dr. Florence B. Seibert, who prepared and donated the tuberculin used.
398
Zoologica: New York Zoological Society
[XXIII :21
Site of Injection.
A simple site for injection, easily available in all Primates and readily
seen, was to be selected. For intradermal testing, the skin on the back proved
most satisfactory, but required shaving and would necessitate handling the
monkey to read the test (monkeys usually face an observer and the shaved
area could not easily be seen without handling). The belly and thighs, by
repeated trials, would not permit easy intradermal administration. The
tuberculin usually broke through to the subcutaneous tissues and rapidly
spread. The eyelid was a location easily seen, which could be readily
approached, and has remained our choice of site for the injection.
The Tuberculin.
It seemed important to select a tuberculin which could be prepared in
great concentration, so that when the injection was made the greatest dose
of sensitizing agent per cell could be given (Sabin & Joyner, 1938). It is a
well-known fact that monkeys do not react allergically to proteins as well as
humans or domestic animals (Kopeloff & Kopeloff, 1938). In addition, a
tuberculin was required which would be standard wherever used, and one
which would not lose potency. We selected Purified Protein Derivative as
prepared by Dr. Florence Seibert of the Henry Phipps Institute of the
University of Pennsylvania because it best met these requirements2 3 (Seibert,
1934).* It was necessary to use a tuberculin for the first test which would
have sufficient concentration to bring down all reactors, regardless of severity
of reaction. The final tuberculin used, which proved to be quite harmless to
all non-tuberculous Primates, was a 1% solution of P.P.D. (0.1 cc. = 1 mgm.
P.P.D.).4 (Human type of organism is usually demonstrated).
The Test.
The animal to be tested is restrained by placing its arms behind it, legs
extended. It is placed on its back, head extending over the edge of a bench or
table, arms pulled down over the edge of the table at a right angle to the long
axis of the body. The operator approaches from behind and grasps the head
in the left hand. A 1 cc. tuberculin syringe and %-inch 27-gauge needle is
used. The syringe is held in the right hand, the needle inserted (bevel up)
just under the skin in the upper lid of the right eye close to the margin.
One-tenth cc. of the tuberculin is injected. This quantity will leave a lentil-
size raised area. The operation is simple and not hazardous for attendant,
operator or monkey. The specimen is placed in a cage and observed, without
further restraint, after 16 hours. A reaction will usually begin to make its
appearance at this time in a tuberculous monkey, and persist for 72 hours,
depending upon the severity of the reaction. Primates too large for usual
manual restraint can be anesthetized with Nembutal (29 mgm. per kilo,
intraperitoneally) .
Interpretation of the Test.
A positive reaction consists of edema of the lid with redness and, on
rare occasions, a small area of necrosis. The extent of the reaction is imma-
terial. All reacting monkeys should either be destroyed or used for short-
2 In a personal communication from Dr. Florence B. Seibert the following data on the stability
of P.P.D. were offered: “We have found that a 1% P.P.D. solution kept in the incubator at 37.5°
for one month loses no potency. Furthermore, a 1% P.P.D. solution kept in the icebox for nearly 5
years has lost very little potency.”
3 Some laboratories are using O.T. (Human) 0.1 cc. = 1.0 mgm. tuberculin (= 0.2% P.P.D.).
4 No systemic reaction was demonstrated in non-tuberculous Orang-utans, Chimpanzees,
Gorillas, Gibbons, Baboons, the Macaques, Mangabeys, the more delicate Douracoulis and the related
New World monkeys.
1938]
Schroeder: Test for Tuberculosis in Primates
399
time experiments. Those monkeys having active, advanced clinical disease
will often succumb to this test dose of 1 mgm. P.P.D. In early or terminal
infections the monkey may not react. In terminal infection, the tuberculin
will either kill the specimen or bring about a profound systemic reaction,
in which event it should be destroyed. Retesting after two months is impor-
tant to detect infection in those specimens which did not have sufficiently
advanced disease at the time of the first test to produce a reaction.
Discussion.
The test as presented leaves much room for further experimentation.
It has proved to be quite satisfactory, however, and is being used routinely
in some laboratories and zoological gardens. Recently the author, with one
assistant, tested 15 Rhesus monkeys in a laboratory in 35 minutes. This
period of time included catching the monkeys, injecting the tuberculin, and
returning the animals to their cages. The following day the readings were
made simply by observing the monkeys in their cages. The value of such
a test can be determined only by extensive trial to accumulate experience,
statistics on a large scale, so that correlation of tuberculin tests and
autopsies can be made.
Probably 90% of Primate tuberculosis is acquired between time of
capture and delivery to the final user of the specimens (Schroeder, 1937).
Testing at origin, before shipment, will eliminate the ultimate planting of
tuberculosis in all monkeys shipped in the same crate with a tuberculous
specimen. When laboratories and zoological gardens demand tuberculosis-
free monkeys from commercial distributors, tuberculosis in Primates will
no longer be a problem.
Conclusions.
1. Monkeys will react locally to concentrated tuberculins.
2. The site of injection is the subcutaneous tissue at the margin of
the upper eyelid.
3. The allergin used is 1/10 cc. of a 1% solution of Purified Protein
Derivative tuberculin (0.1 cc. = 1 mgm. P.P.D.) .
4. Reactors show swelling and redness of the lid, visually evident with-
out removing the specimens from their cages, in 16 hours, and persisting
for 72 hours.
Bibliography.
Fox, H.
1923. Diseases in Captive Wild Mammals and Birds. J. B. Lippincott,
Philadelphia, p. 519.
Kopeloff, L. & Kopeloff, N.
1938. A Shock Reaction in the Monkey. Journ. Soc. Amer. Bad. Jan., 1938
(Abstract number).
Sabin, F. R. & Joyner, A. L.
1938. Tubercular Allergy Without Infection. Journ. Exp. Med. Vol. 68,
p. 659.
Schroeder, C. R.
1938. Acquired Tuberculosis in the Primate in Laboratories and Zoological
Collections. Journ. Amer. Pub. Health Assn. Vol. 28, p. 469.
Seibert, F. B., Arson, J. D., Reichel, John, Clark, L. T. & Long, E. R .
1934. Purified Protein Derivative, a standardized tuberculin for uniformity
in diagnosis and epidemiology. Amer. Rev. Tuberc. Suppl. Vol. 30,
No. 6, p. 713.
400
Zoologica: New York Zoological Society
Simpson, S. & Galbraith, J. J.
1906. Observations on the Normal Temperature of the Monkey and Its
Diurnal Variation and on the Effect of Changes in the Daily Routine
on This Variation. Trans. Royal Soc. Edinb. Vol. 45, pp. 65-104.
White, C. Y. & Fox, H.
1909. Results of Tuberculin Tests in Monkeys at the Philadelphia Zoological
Garden. Arch. Int. Med. Vol. 4, p. 517.
EXPLANATION OF THE PLATES.
Plate I.
Fig. 1. Manner of restraining monkeys weighing up to 10 kilos, for injection
of tuberculin. Larger specimens should be anesthetized. When the
operator touches the eyelid with the needle the monkey usually closes
his eyes, which facilitates the proper administration of the tuberculin.
Fig. 2. The 27-gauge needle, attached to a 1 cc. Luer end tuberculin syringe, is
inserted bevel up in the lid near its lower margin, and 0.1 cc. is admin-
istered.
Plate II.
Fig. 1. An average reaction consists of edema with or without apparent redness.
The edema prevents the specimen from completely uncovering the pupil.
Such a reaction can be recognized without restraining the specimen for
close observation. Some reactions are more intense, with pronounced
redness involving adjacent structures and edema which may close the eye.
Photographs by Hans Gehrung, N. Y. State Psychiatric Institute.
SCHROEDER.
PLATE I.
FIG. I.
FIG. 2.
A DIAGNOSTIC TEST FOR THE RECOGNITION OF TUBERCULOSIS IN PRIMATES;
A PRELIMINARY REPORT.
SCHROEDER.
PLATE II.
A DIAGNOSTIC TEST FOR THE RECOGNITION OF TUBERCULOSIS IN PRIMATES;
A PRELIMINARY REPORT.
Index to Parts 1-4
401
INDEX.
Names in bold lace indicate new genera, species or varieties; numbers in
bold lace indicate illustrations.
A
Abramis, 193
Acanthemblemaria hancocki, 311
crockeri, 310
Acanlhorhodeus, 193
Acestrorhynchus, Texl-fig. 2, 322, 324
Aegina, 130
allernans, 130
cifrea, 131
rhodina, 130
rosea, 130
Aegirtura, 132
grimaldii, 132
lanzerolae, 133
myosura, 132
weberi, 133
Aequorea, 110
floridana, 111
rhodoloma, 129
lenuis, 110
Aglaura, 118
Aglantha, 118
Alesles, Texl-fig. 2, 321, 322, 331, 336
kolschyi, 323
macrolepidolus, 323
nurse, 323
Amalopenaeus sordidus, 57
Ameiurus melas, 3
Ammodyles lucasanus, 306
personalus, 306
lobianus personalus, 307
Andrya americana, 33
americana leporis, 33
cuniculi, 34, 36, 39, 47, 50
rhopalocephala, 31, 47
Anisistsia, 345
Anodus, 322
Anoslomus, Text-fig. 2, 323, 324
Anguilla anguilla, 226, 313
roslrala, 314
vulgaris, 226
Aphyocharax, 322, 324
Arenichthys, 301
aplerus, 301, 302
Astyanax, 322
Atolla, 133
bairdii, 160
gigantea, 160
vivaldiae, 160
wyvillei, 160
Aurellia, 167
aurila, 168
flavidula, 168
labiata, 167, 168
limbala, 167
maldivensis, 168
solida, 168
B
Barbalula, 191
Barbus, 192
barbus, 195
fluviatilis, 195
plebejus, 195
Barilius, 193
Bairdiella chrysura, 265
Bertia americana, 33
americana leporis, 33
Bertiella americana, 35
Bella splendens, 4
Bolia, 191
Bougainvillia brilannica, 105
fulva, 105
niobe, 104
Brachychalcinus, 326
Brachydanio rerio, 7
Brycon, Texl-fig. 2, 321, 322, 330, 332-334, 337,
350
Bryconaelhiops, 322, 323, 335
Bursarius cylherae, 150
Bylholiara murrayi, 108
C
Caenolropus, 351
Calycopsis sp., 108
Canis nubilis, 239
Carassius auralus, 3, 219, 222, 223, PI. I & II
carassius, 247
Carybdea, 136
alaia, 144, 147-150
var. grandis, 145
var. moseri, 145
var. pyramis, 145
marsupialis, 137
xaymacana, 139, 140, 142-144
murrayana, 137
raslonii, 137
sliasnyi, 136
xaymacana, 137
Carybdeida, 135
Calostomus, 191
Caudina chilensis, 383
Chalceus, 322
Chalcinus, Text-fig. 2, 321, 322, 334
Irachypomus, 334
Charax, 321, 322
Charybdea grandis, 144
(Tamoya) haplonema, 151
madraspatana, 145
moseri, 144
402
Zoologica: New York Zoological Society
[XXIII
obeliscus, 145
pyramis, 144
philippina, 145
xaymacana, 139
Charybdella alata, 144
philippina, 145
Charybusa obeliscus, 145
Cheirodon, Text-fig. 2, 321, 326, 329
pulcher, 322, 325
Chelonia mydas, 93-98, PI. I-XVIII
Chilodus, Texl-fig. 2, 322, 324
Chondrosloma, 193
Chromalonema, 109
erylhrogonon, 109
rubrum, 109
Ciiharichihys gordae, 302, 303
Citharidium, 322, 349
Cilharinus, Texl-fig. 2, 322, 323, 324
Cillolaenia ctenoides, 34, 36, 39, 44, 51
denficulala, 34, 36, 39, 42, 50
lalissima, 42
mosaica, 33
peclinata, 34, 36, 39, 45, 51
americana, 34, 36, 38, 39, 51
perplexa, 34, 36, 37, 51
variabilis, 34, 36, 39, 40, 51
angusla, 33
imbricata, 33
variabilis, 33
Clupea harengus, 3
Cobilis, 191
Colobonema, 114, 116
sericeum, 117
lypicum, 117
Compsura, 335
Corynopoma, Texl-fig. 2, 321
Crealochanes, 322
Crossola, 118
alba, 119
brunnea, 119
var. norvegica, 119
norvegica, 119
pedunculata, 119
rufobrunnea, 118
Crossosloma, 191
Clenoiaenia perplexa, 33
variabilis, 33
Cucumaria californica, 372
chierchia, 374
fallax, 372
Cunina sp., 128
Cunoclona grimaldi var. munda, 133
guinensis, 133
lanzerolae, 132
nausilhoe, 132
obscura, 133
Curimalus, Texl-fig. 2, 322, 323, 324, 348
Cynodon, Text-fig. 2, 320
Cynopolamus, 321
Cynoscion arenarius, 269
nebulosus, 264
regalis, 263
Cyprinodon variegalus, 195, PI. I-VII
Cyprinus, 192
carpio, 226
D
Danio, 193
Diapoma, Text-fig. 2, 321
Diplychus, 193
Dipylidium latissima, 42
leuckarti, 44
Dirofilaria immifis, 235, PI. I-V
megalhaes, 235
repens, 235
Dislichodus, Text-fig. 2, 322, 323, 324, 352
E
Eigenmannia, 354, 355
Electrophorus, 354, 355
eleciricus, 203, 217, 247, PI. I-VI
Emblemaria hudsoni, 309
micropes, 308, 309
nigra, 309
nivipes, 309
oculocirrus, 309
Erethizon dorsalus, 37
epixanthes, 37
Erythrinus, Text-fig. 2, 320, 324, 346
Erylhroculter, 192
Esox lucius, 226
Eugnalhichlhys, 322, 323
Euphyllophorus, 379
granulatus, 379
Eupomotis gibbosus, 4
Eusicyonia affinis, 84, 85
aliaffinis, 84, 85, 86
carinala, 81
disdorsalis, 87, 89
disedwardsi, 82
disparri, 80, 81
dorsalis, 89
edwardsi, 82
ingenlis, 88, 89
laevigata, 80
parri, 81
penicillata, 82, 83
picta, 87
sp., 84, 85, 86
slimpsoni, 87
Euthynnus allelterata, 306
Euthyonidium, 379, 380
ovulum, 380
seguroensis, 379
F
Filaria bancrofti, 235
immitis, 235
Fislularia impatiens, 365
Fowlerina, 326
Fundulus confluentus, 265
heleroclilus, 195, 229
G
Gadus morrhua, 226
Gasteropelecus, Texl-fig. 2, 320, 321, 334
pectorosus, 334
Gaslerosieus aculeatus, 226
Gastromyzon, 191
Gemmaria dicholoma, 102
Gennadas brevirostris, 58
capensis, 58
clavicarpus, 59
1938]
Index to Parts 1-4
403
elegans, 58
kempi, 58
parvus, 57
propinquus, 59
scutafus, 59
sordidus, 57, 58
iinayrei, 58
Geryonia proboscidalis, 127
ietraphylla, 127
Gnalhopogon, 193
Gobio, 192
Gobiobofia, 192
Gobionellus boleosoma, 270
Gobiosoma bosci, 272
robuslum, 272
shufeldli, 270
Grundulus, 321
Gymnarchus, 356
H
Halicalyx tenuis, 113
Halicreas, 120
album, 121, 124
conicum, 121
glabrum, 121, 124
minimum, 121, 122
papillosum, 121, 124
racovilzae, 121
rolundafum, 121, 124
Haleporus doris, 60
Haliscera alba, 124
Halilrephes, 125
maasi, 125
valdiviae, 125
Hemichromis bimaculatus, 4, 219
Hemiculter, 192
Hemigrammus, 322
Hemimyzon, 193
Hemiodus, 324
Hemistichodus, 350
Helerodon contortrix, 394
Heterofiara anonyma, 108
minor, 108
Hippocampus hudsonius, 268
punctulatus, 268
regulus, 269
zosterae, 266
Holothuria arenicola, 364, 372
atra, 366
erinaceus, 369
glaberrima, 368, 369
grisea, 366
gyrifer, 371
imifans, 367
impaliens, 365
inhabilis, 365
inornata, 365, 366
kapiolaniae, 368
languens, 367
lubrica, 368, 369
maculata, 364
marenzelleri, 369
var. Iheeli, 369, 370
moebii, 368
monacaria, 371
pardalis, 372
parva, 368
? pafagonica, 371
surinamensis, 368
Homaloplera, 193
Homoeonema, 114
alba, 124
(Colobonema) sericeum, 117
glabrum, 124
platygonon, 120
lypicum, 117
Hoplerylhrinus, 322
Hoplias, 322, 323, 324, 345
Hybopsis, 192
Hyborhynchus notatus, 196
Hydraspis geoflroyana, 93
Hydrocyon, Text-tig. 2, 320, 321, 322, 323, 332,
333, 338-342, 349
lineaius 337
Hymenopenaeus apholicus, 61
diomedeae, 66
doris, 60
laevis, 61
nereus, 60
sleindachneri, 66
villosus, 61
Hyphessobrycon, 325
Hypophthalmichthys, 193
I
Ichthyoborus, Text-fig. 2, 323, 324
Iguanodectes, 321
J
Jenkinsia slolifera, 3
L
Labeo, 192
Labidodemas, 363
americanum, 363, 364
semperianum, 363
Laodicea, 109
Lebiasina, Text-fig. 2, 323, 324
Lebistes reticulatus, 4
Leioheierodon madagascariensis, 393, PI. I
Leiostomus xanthurus, 267
Lepidogobius guaymasiae, 272
y-cauda, 272
Lepomis auritus, 3
Leporinus, 322, 323, 324, 348-351
Lepturichlhys, 191
Lepus americanus, 40, 42
californicus melanolis, 40, 42
europaeus, 47
melanolis, 38
timidus, 47, 48
variabilis, 47
Lernaecocera cyprinacea, 217
Leuciscus, 192, 195
rutilis, 196
Linuche aquila, 163
draco, 163
unguiculata, 163
Liriope, 126
Ietraphylla, 127
Luciocharax, Text-fig. 2, 324, 344, 345
Lucioperca sandra, 195
404
Zoologica: New York Zoological Society
[XXIII
M
Macaca mulatta, 397
Macrodon, 343
Macropsobrycon, 325
Marmota marmota, 47
Marsupialis alata, 144
Medusa aurila, 168
mucilaginosa, 128
noctiluca, 164
proboscidalis, 127
unguiculala, 163
Megalamphodus, 324
Megalobrama, 192
Mesoborus, 324, 353
Metapenaeopsis, 72
Metynnis, Texl-fig. 2, 320, 321
Micralesles, 322, 335
Misgurnus, 191
Mixobrycon, 321
Mobula japanioa, 300
lucasana, 299, PI. I-III
larapacana, 300
Moenkhausia, Texl-fig. 2, 321, 322, 329
Molpadia chilensis, 383
Morone labrax, 3
Mylesinus, 321
Myleles, Text-fig. 2, 328
Mylosoma, 321, 327, 329
Myxobolus capsulalus, 198
funduli, 196
lintoni, 195
musculi, 196
pfeifferi, 195
piriformis, 195
Myxocyprinus, 191
N
Nannaelhiops, 350
Nannoslomus, 324
Nausilhoe puncfala, 159
Nemacheilus, 191
Neolebias, 350
Neolhyonidium, 379
hawaiense, 379
Nolemigonus, 192
O
Odonlostilbe, 335
Olindias, 112
phosphorica, 113
tenuis, 113
sambaquiensis, 112
tenuis, 113
Opsariichthys, 192
Oryclolagus cuniculus, 44, 45, 47, 48
cuniculus domesticus, 45
Osleochilus, 193
P
Pandea conica, 106
rubra, 107
Panlachogon, 114
haeckeli, 114, 115
rubrum, 114, 115
scotli, 114
Paracaudina chilensis, 383, 384
Paragoniates, 335
Paralichthys albigutla, 271
dentalus, 273
lethosligma, 271
Parapenaeus kishinouyei, 72
Paraphago, Texl-fig. 2, 324
Parastichopus californicus, 362
parvimensis, 362
Parecbasis, 325
Parodon, 324
Pattalus, 379
mollis, 379
Pegantha clara, 134
smaragdina, 134
Pelagia crassa, 164
curacaoensis, 164
cyanella, 164
flaveola, 164
noctiluca, 164
var. neglecta, 164
panopyra, 164
var. placenta, 164
perla, 164
phosphora, 164
purpuroviolacea, 164
rosea, 164
Penaeopsis beebei, 72, 74, 75
goodei, 73, 76, 78
kishinouyei, 72, 73, 76
(Metapenaeopsis) kishinouyei, 72
(Melapenaeopsis) beebei, 74
(Metapenaeopsis) mineri, 77
mineri, 72, 76, 77
pubescens, 76
smithi, 78
Penaeus californiensis, 67, 69, 71
canaliculaius, 67
brasiliensis, 67
brevirostris, 67, 69, 71
Pentamera, 373
beebei, 374
calcigera, 373
chierchia, 374, 375
pulcherrima, 373
zacae, 375
Periphylla, 133
dodecabostrycha, 156
hyacinthina, 155
regina, 156
Petersius, 322, 335
Petrocephalus, 356
Phago, Texl-fig. 2, 323, 324, 353
Phalacrocorax penicillala, 306
Phoxinus, 192
Phyllophorus, 379, 381
aculeatus, 381
destichodus, 382
urna, 379
zacae, 381, 382
Physa, 8
Piabucina, 343, 346
„ Piaracius, 336
Pimephales notatus, 196
Planorbis, 9
Placobdella parasitica, 217
Platemys geoffroyana, 93
Pneumalophorus grex, 3
1938]
Index to Parts 1-4
405
Poecilobrycon, Texi-iig. 2, 322, 324
Polycolpa forskali, 134
Poralia rufescens, 165
Proboscidactyla, 109
Prochilodus, Text-iig. 2, 322, 323, 324
Psectrogasler, 322
Pseudogobio, 193
Pseudoperilampus, 193
Pseudopleuronectes americanus, 393, PI. I
Psolus monacarius, 371
Ptychogena, 109
Pyrrhulina, 335
R
Rana clamitans, 395
Rasbora, 192
Rhamphichthys, 354
Rhegmalodes floridana, 111
floridanus, 111
Richardsonius, 192
Rhodeus, 192
amarus, 219
Rhopalonema velalum, 113
S
Sarcodaces, Text-fig. 2, 322, 323, 324, 344, 345
Saurogobio, 193
Sciaenops ocellalus, 258, 267
Schizopygopsis, 193
Schizolaenia amerioana, 33, 34, 36, 50
americana leporis, 33
decrescens, 33
lalicephala, 33
macrorhyncha, 33
Schizolhorax, 192
Scorpaenodes coriezi, 304, 305
Seriola colburni, 306
dorsalis, 306
Sergesles edwardsii, 61
Serrasalmo, Texl-fig. 2, 321, 322, 323
Sicyonia penicillala, 83
picta, 87
Sminlhea eurygasler, 114
Solenocera agassizii, 63
florea, 64"
hexlii, 66
mutator, 61, 62
vioscai, 63
Solmaris mucilaginosa, 128
rhodoloma, 129
Solmissus bleekii, 129
faberi, 129
inoisa, 129
Spintherobolus, 325
Starksina, Text-fig. 2 (opp. 321)
Sternarchus, 356
Slernothaerus niger, 93
Stethaprion, Texl-fig. 2, 321
Sliohopus fusous, 363
gyrifer, 371
parvimensis, 362
(Hololhuria?) palagonicus, 371
Stiohonodon, 326
Slolus gibber, 376
ovulum, 380
Sylvilagus floridanus alacer, 32, 38, 41, 42
floridanus mallurus, 32, 38, 41, 42
nuttalli pinetis 37, 38, 42
palustris, 41, 42
T
Taenia clenoides, 44
denticulala, 42
goezei, 42
lalicephala, 33
leuckarti, 44
pectinata, 45
rhopaliocephala, 47
Tamoya, 136, 150
alala, 144
garganlua, 151
haplonema, 135, 151, 152-155
prismalica, 151
punctata, 139
Tetrogonopterus, Texl-fig. 2, 321, 322, 327, 329
argentatus, 323
argenleus, 325
fascialus, 323
Thalassochelys caretta, 93
Thyone, 376
benti, 376
var. zacae, 376, 377
gibber, 376, 377
ovulum, 380
panamensis, 378
rubra, 378
similis, 376
Thyonepsolus beebei, 382
Thyonidium, 379
pellucidum, 379
Trachypeneus, 73
constrictus, 79
similis pacificus, 78
(Trachysalambria) brevisulurae, 79
Trachysalambria curvirostris, 80
V
Varicorhinus, 192
Villarius calus, 7
Vivipara, 11
W
Willia, 109
X
Xenocharax, 322, 323, 324
Xenocypris, 192
Z
Zancleopsis, 102
dichotoma, 102, 103
tentaculata, 102
jgeto |9ork Hoologtcal Stcietp
General Office: 90 Broad Street, New York City
©fitters
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 -Parti
W. Reid Blair, Director
Raymond L. Ditmars, Curator of Mammals and Reptiles
Lee S. Crandall, Curator of Birds
Charles R. Schroeder, Veterinarian
Claude 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
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 UEropical i^esearctj
William Beebe, Director and Honorary Curator of Birds
John Tee-Van, General Associate
Gloria Hollister, Research Associate
Jocelyn Crane, Technical Associate
Cbttortal Committee
Fairfield Osborn, Chairman
gquarium
W. Reid Blair
William Beebe
Charles M. Breder, Jr.
William Bridges
on 5
>» DjlpiO
n 5mIM1
agglEl
►3) IlSSBPO
ITIgM
»aj535».
s*3> jemm
®)3)W
!»» OTpl