LIBRARY

OF THE

UNIVERSITY OF CALIFORNIA.

GIFT OF

Class

UNIVERSITY OF CALIFORNIA PUBLICATIONS

IN

ZOOLOGY

Vol. 6, No. 15, pp. 353-468, plates 33-48

June 17, 1911

THE GENUS GYROCOTYLE, AND ITS

SIGNIFICANCE FOR PROBLEMS

OF CESTODE STRUCTURE

AND PHYLOGENY

BY

EDNA EARL WATSON

BERKELEY

THE UNIVERSITY PRESS

UNIVERSITY OF CALIFORNIA PUBLICATIONS

Note. — The University of California Publications are offered in exchange for the publi-
cations of learned societies and institutions, universities and libraries. Complete lists of
all the publications of the University will be sent upon request. For sample copies, list!
of publications or other information, address the Manager of the University Press, Berkeley,
California, U. 8. A. All matter sent in exchange should be addressed to The Exchange
Department, University Library, Berkeley, California, U. S. A.

OTTO HAKEASSOWITZ, R. FKIEDLAENDER & SOHN,

LEIPZIG. BERLIN.

Agent for the series in American Arch- Agent for the series in American Arch-
aeology and Ethnology, Classical Philology, aeology and Ethnology, Botany, Geology,
Education, Modern Philology, Philosophy, Mathematics, Pathology, Physiology, Zool-
Psychology. ogy, and Memoirs.

ZOOLOGY. — W. E. Bitter and 0. A. Kofoid, Editors. Price per volume $3.50. Commenc-
ing with Volume II, this series contains Contributions from the Labora-
tory of the Marine Biological Association of San Diego.

Cited as Univ. Calif. Publ. Zool.
Vol. 1. 1. The Hydroida of the Pacific Coast of North America, by Harry Beal

Torrey. Pp. 1-104; plates 1-11. November, 1902 $1.00

2. A Case of Physiological Polarization in the Ascidlan Heart, by Frank

W. Bancroft and O. O. Esterly. Pp. 105-114. April, 1903 10

3. Embryology and Embryonic Fission in the Genus Crisia, by Alice

Bobertson. Pp. 115-156, plates 12-15. June, 1903 .60

4. Correlated Protective Devices in some California Salamanders, by

Marion E. Hubbard. Pp. 157-170, plate 16. November, 1903

6. Studies on the Ecology, Morphology and Speciology of the Young of
some Enteropneusta of Western North America, by William E. Bitter
and B. M. Davis. Pp. 171-210, plates 17-19. February, 1904 .50

6. Regeneration and Non-Sexual Reproduction in Sagartia davisi, by

Harry Beal Torrey and Janet Buth Mery. Pp. 211-226, 7 text-
figures. May, 1904 15

7. The Structure and Begeneration of the Poison Glands of Plethodon, by

C. O. Esterly. Pp. 227-268, plates 20-23. June, 1904 1.00

8. The Distribution of the Sense-organs in Microscolex elegans, by John

F. Bovard. Pp. 268-286, plates 24-25. December, 1904 .50

9. Some New Tintinnidae from the Plankton of the San Diego Region,

by Charles A. Kofoid. Pp. 287-306, plates 26-28. July, 1905 25

Index, pp. 307-317.

Vol. 2. (Contributions from the Laboratory of the Marine Biological Association

of San Diego.)

Introduction. A General Statement of the Idea? and the Present Aims
and Status of the Marine Biological Association of San Diego, by
William E. Bitter, Director of the Station. Pp. i-xvii. 2 Maps .25

1. The Hydroids of the San Diego Region, by Harry Beal Torrey. Pp.

1-43. 22 text figures. December, 1904.

2. The Ctenophores of the San Diego Region, by Harry Beal Torrey.

Pp. 45-51, plate 1. December, 1904.

Nos. 1 and 2 in one cover 60

3. The Pelagic Tunicata of the San Diego Region, excepting the Larvacea,

by William E, Bitter. Pp. 51-112, plates 2-3. 31 text-figures. Jan-
uary, 1905 .65

4. The Pelagic Copepoda of the San Diego Region, by 0. O. Esterly. Pp.

113-233, 62 text figures. September, 1905 1.25

5. The Non-encrusting Chilostomatous Bryozoa of the West Coast of North

America, by Alice Robertson. Pp. 235-322, plates 4-16. December,
1905 '..- 1.00

6. Differentiation in Hydroid Colonies and the Problem of Senescence,

by Harry Beal Torrey. Pp. 232-332, 4 text-figures. December, 1905.

7. The Behavior of Corymorpha, by Harry Beal Torrey. Pp. 333-340,

5 text-figures.

Nos. 6 and 7 in one cover - 25

8. Dinoflagellata of the San Diego Region. I. On Heterodinium, a New

Genus of the Peridinidae, by Charles Atwood Kofoid. Pp. 341-368,

plates 17-19. January, 1906 _ 26

Index, pp. 369-382.

* Supply limited; sold only with complete volumes.

UNIVERSITY OF CALIFORNIA PUBLICATIONS

IN

ZOOLOGY

Vol. 6, No. 15, pp. 353-468, plates 33-48 June 17, 1911

/" 'Of THE A

I UNIVERSITY )

v or y

X^UFORJj^

^^^SBHESOMBB^^^^

THE GENUS GYROCOTYLE, AND ITS

SIGNIFICANCE FOR PROBLEMS

OF CESTODE STRUCTURE

AND PHYLOGENY*

BY

EDNA EARL WATSON

CONTENTS.

PAGE

A. Introduction 354

B. History of the Genus 356

C. Gyrocotyle of the Coast of California 363

I. Occurrence and Host 363

II. Gross Anatomy 366

III. Behavior of Living Animal 371

IV. Characteristics for Separation of Species 376

V. Key to Species of Gyrocotyle 382

D. Morphology and Histology of Gyrocotyle 383

I. Material and Methods 383

II. Body-covering and Spines 384

III. Musculature 389

IV. Reproductive Organs 395

V. Formation of Compound Egg and Cleavage , 406

VI. Excretory System 407

VII. Nervous system 409

VIII. Sense Organs 417

E. General Discussion 418

I. Cuticula 419

II. Orientation 421

F. Summary 431

* A dissertation presented to the Faculty of the College of Natural
Sciences, in the University of California, in partial satisfaction of the
requirements for the degree of Doctor of Philosophy. April, 1910.

218265

354 University of California Publications in Zoology. CV°L- 6

A. INTRODUCTION.

The Cestodaria (Monticelli, 1892) or monozoic cestodes
(Lang, 1891), including Amphilina, (Rud.), Archigetes (Ratz)
Caryophyllaeus, (Rud.) Gyrocotyle, (Diesing) and Wageneria
(Wag.) are of peculiar interest to students of the platyhelminths.
They occupy a position intermediate between the merozoic ces-
todes and the trematodes, being allied to the latter by the general
arrangement of the reproductive organs and by their funda-
mentally unitary character, and distinguished from them by the
absence of a digestive tract. This last characteristic allies them
with the merozoic cestodes, from which they are distinguished
on the other hand by their unsegmented condition and by the
accompanying absence of segmental repetition of the genital
organs, as stated by Monticelli (1892). The group is difficult
to consider as a whole because of the remarkable diversity shown
in its four well-established genera, both as to structure, life-his-
tory and host. The characters mentioned above practically
exhaust those common to the group, and each genus possesses
characters of such rank as fittingly to receive ordinal distinction
in classification. Two of the genera, Archigetes and Wageneria,
have but one species each. Two each have been assigned to
Caryophyllaeus and Amphilina, and at least four are now recog-
nized in the genus Gyrocotyle. The grounds on which specific
distinctions have been made are far from satisfactory. These
considerations indicate that the group is not only phylogenetic-
ally intermediate but composed of very old, long-established
forms, on which the forces that make for speciation have long
since ceased to act. The Cestodaria are either remnants of a
once large and differentiated group, or they are members of a
class never differentiated to any considerable extent. They are
presumably of greater age than the cestodes, their hosts being
invertebrates or primitive vertebrates. Cestodes are phylogen-
etically the youngest of the platyhelminths, since the host in
the sexual period is always a vertebrate. They have been greatly
and variously modified by their parasitic habit, and they present
problems of structure and development of extreme difficulty. It
seems probable that much help towards the solution of these can

1911] Watson: The Genus Gyrocotyle. 355

be obtained through an exhaustive study of the Cestodaria, the
nearest surviving relatives of the merozoic cestodes; and in this
possibility lies the chief interest pertaining to the former group
and to this investigation.

Lonnberg (1897), in a study of the phylogeny of parasitic
platyhelminths, discusses the possible phylogenetic significance
of the Cestodaria and comes to the conclusion that they cannot
be regarded as ancestral forms of the merozoic cestodes :

"Die jetzigen polyzoischen Bandwurmstrobilen sind natiirlich aus
monozoischen Formen hervorgegangen. Die noch jetzt existierenden
monozoischen Cestoden konnen aber kaum als Stammformen der poly-
zoischen angesehen werden. . . . Diese [Gyrocotyle and Amphilina]
sind wahrscheinlieh beide urspriinglieh, weichen aber in mehreren Hin-
sichten von den echten Cestoden recht sehr ab. Gyrocotyle ist durch das
Erweben von Haftorganen an beiden Korperenden, das Trichterorgan
am Vorderende und das Acetabulum am Hinterende, unfahig geworden,
sich weiter zu entwickeln. ' '

While the writer is convinced of the essential correctness of
Lonnberg ?s general contention, that the ancestors of cestodes
must be sought, not among Cestodaria or trematodes, but among
Turbellaria, it is felt that he undervalues the evolutionary sig-
nificance of the Cestodaria. The merozoic cestode is probably
not the descendant of any monozoic cestode now in existence ;
but because of the fundamental similarity between the two
groups we may ^.ope to find, in the simpler forms meeting the
same problems in essentially the same way, some clue to the
manner in which occurred the complicated processes by which
the hypothetical turbellarian-like ancestor was transformed into
the merozoic cestode. Naturally, those cestodes which are second-
arily monozoic, Arckigetes and Caryophyllaeus (Lonnberg, 1897)
are of little value in this sense. Amphilina (Acipenser) and
Gyrocotyle (Chimaera) are primarily monozoic, inhabit hosts
of great phylogenetic age, and show fundamental correspondence
to the typical cestode structure in all important respects. Gyro-
cotyle, the genus with which this paper is concerned, Lonnberg
regarded as off the line of cestode evolution, side-tracked by the
development of organs of attachment at both ends. The writer
hopes to show that only one functional organ of attachment is
present, that this is homologous to the cestode scolex, although

356 University of Calif or n ia Publications in Zoology. tv°L- 6

much simpler, and that its structure and position afford very
significant indications of the origin and primitive location of the
cestode scolex. The evidence of this homology and its theoretical
significance will be discussed later; the point of chief interest
now is that the very structural peculiarity which Lonnberg
regarded as removing Gyrocotyle from any great phylogenetic
interest is the source of its most significant and interesting
contribution to the solution of the problem of cestode origins.

This investigation has been carried on in the Zoological Labo-
ratory of the University of California and at the Station of the
Marine Biological Association at La Jolla. For material I am
indebted to Professor C. A. Kofoid, and through him to the
United States Bureau of Fisheries and the Marine Biological
Association at La Jolla; also to Professor Jacques Loeb, for
privileges at the Herzstein Research Laboratory at Monterey.
For substantial assistance in the preparation of drawings for
publication I am indebted to Miss E. J. Eigden. For the point-
ing out of this line of investigation, for help in obtaining the
obscure and not easily accessible literature of the subject, and
for counsel and encouragement, I wish to express my debt to
Professor Kofoid, under whose direction this work has been
brought to its present state of completion.

B. HISTORY OF THE GENUS.

The history of the genus Gyrocotyle is complicated by con-
flicting reports as to its systematic position, its habitat, its orien-
tation and its morphology. The genus was established by
Diesing (1850), for a parasite obtained by Gueinzius in 1842
from Antilope pygarga, a South African ungulate. This report
was an error (Diesing, 1858, p. 492), but was accepted as correct
for several years, contributing in no small degree to the confusion
concerning the habitat and systematic position of the genus. In
1844 Krb'yer showed Diesing a parasite similar to that collected
by Gueinzius, taken from inside the shell of an edible mollusk
(Diesing, 1855). These two specimens were grouped in one
species by Diesing (1850) under the new genus Gyrocotyle, and
described in the following terms:

1911] Watson: The Genus Gyrocotyle. 357

"Corpus sub-ellipticum depressum; os subterminale corpore con-
tinuum; os subterminale anticum exiguum; acetabulum unum in extremi-
tate caudali terminale, sessile, orbiculare, disce in gyros plicata; penis
ventralis superus lateralis; apertura feminea infra penem centralis;
porus excretorius dorsalis supra acetabulum." [Quoted from Braun
(1889)].

The genus as thus denned contained only a single species,
Gyrocotyle rugosa. Diesing referred the genus to the trematodes,
placing it near the genus Amphistomum because of the posterior
position of its acetabulum. In 1855, in his description of the
collection of endoparasitic worms made by Natterer in Brazil,
Diesing published a second description of the genus identical
with the first, with five figures and a discussion of the habitat
and systematic affinities of the genus. He regards it as improb-
able that both Mactra edulis and Antilope pygarga are true hosts
of Gyrocotyle, and believes that the report of its occurrence in
Mactra is an error. In case Mactra should prove to be the true
host, he concludes that Gyrocotyle as an ectoparasite would have
to be referred to the Bdellidea. As an endoparasite of Antilope,
he places it under the trematodes. Diesing made no attempt at
a study of the internal structure of the parasite and apparently
only a cursory examination of its external appearance. From
his figures it seems probable that his specimens were in a state
of decomposition when preserved.

In 1852 Wagener described and figured a parasite found by
himself and Grube in the spiral valve of Chimaera monstrosa,
at Nice. For this he established the genus Amphiptyches, con-
taining a single species, urna, and referred this genus to the
cestodes because of its morphological resemblance to that group.
Wagener made a careful study of the internal structure of the
worm, both in the living animal and preserved specimens made
partially transparent by clearing. Except for confusion, as to
the ducts of the genital organs and in a few minor details, his
figures are correct. Diesing 's (1855) description and figures
convinced Wagener that Amphiptyches was identical with Gyro-
cotyle. In a letter to Diesing (1857) he calls the latter 's atten-
tion to Amphiptyches urna, and in his ' ' Enthelminthica No. V. ' '
(Wagener, 1858), he discards Amphiptyches as a generic name
and retains it as a specific name for the form discovered by him :

358 University of California Publications in Zoology. ITOL-

Gyrocotyle amphiptyches. Diesing ("Revision der Myzhelmin-
then," 1858) recognizes Amphiptyches as a distinct genus, plac-
ing it near Amphistomum under trematodes, in the place occupied
in his earlier paper (1855) by Gyrocotyle, then considered an
ectoparasite of Antilope pygarga. He supposes Amphiptyches
to have a mouth, digestive tract and anus, in the face of Wag-
ener's (1852, p. 547) explicit statement to the contrary. He
gives as its habitat the gills and intestinal tract of Chimaera
monstrosa. His own genus, Gyrocotyle, he puts under the
Bdellidea near Malacobdella (Diesing, 1858, p. 492). He makes
certain changes in the generic description, the most important of
these being in the interpretation of the ventral canal opening
(proboscis of Spencer) ; in his description of 1850 he regards
it as an excretory pore, in 1858 as the anus. The change was
made in order to place the genus under the Bdellidea, where
Diesing was convinced it belonged if Mactra was its true host.
That the report of its occurrence in Antilope was an error
Diesing had discovered in the interval between 1855 and 1858r
and it was probably due to this discovery that the change in the
position of the genus was made (Diesing, 1858, p. 492).

Diesing 's disposition of Amphiptyches was apparently not
satisfactory to Wagener, for in his paper "Ueber Amphilina
foliacea, Gyrocotyle Diesing und Amphiptyches (Grube and
Wagener)" (1858), Wagener definitely withdraws Amphip-
tyches, recognizing his form to be a species of Diesing 's older
genus. In Diesing 's ' * Nachtrage und Verbesserungen zur Revision
der Myzhelminthen " (1859), he agrees with Wagener in placing
the two species under one genus, which he refers to the "Bdel-
lidea monocotylea. "

The two species are distinguished one from the other only
by the absence of spines and lateral frills and the smaller size
of the "tail rosette" in Gyrocotyle rugosa.

Diesing (1859, p. 448) regards the genus as typically ecto-
parasitic on marine mollusks, assuming that the species of Grube
and Wagener has its normal habitat, not in Chimaera monstrosa,
but rather on some of those mollusks whose fragments occur in
the intestine of the fish. "Da Gyrocotyle rugosa aus einem

1911] Watson: The Genus Gyrocotyle. 359

Mollusken, namlich der Mactra edulis stammt, so 1st es sehr wahr-
scheinlich, dass auch die von Grube und Wagener beschriebene
Art ihren eigentlichen Wohnort nicht in der Chimaera mon-
strosa, sondern vielmehr in einer jener Mollusken, deren Frag-
mente sich im Darme des Fisches verfanden, habe."

Van Beneden and Hesse (1864), maintain the two genera and
refer both to the Hirudinea as the " Malacobdellaria dioici."
These authors agreed with Diesing in regarding a mollusk as the
true host of Amphiptyches, stating explicitly that it is not to be
considered a parasite of Chimaera, but as having been taken in
by chance with the mollusk on which the fish feeds. Later Van
Beneden (1869) withdrew from this position and classed Amphi-
ptyches as a true parasite of Chimaera.

No further work was done on either of these forms until
Monticelli's "Saggio di una morfologia dei trematodi" (1888) in
which he excludes Gyrocotyle and Amphiptyches from the
trematodes, referring Gyrocotyle to the bdellodes, near Malacob-
della, following Hesse and Van Beneden. He affiliates Amphi-
ptyches with the cestodes, suggesting that Amphilina and Amphi-
ptyches are closely related forms, the latter being the simpler of
the two, and together should be regarded as forms transitional
between trematodes and cestodes. Shortly after this Monticelli
(1889b) compared the specimens of Gyrocotyle and Amphi-
ptyches available in the museums of Leipzig, Berlin and Vienna.
As a result of the comparison he says :

' ' Io ho potuto stabilire con certezza la loro posizione sistematica

desumendola dai loro caratteri anatomic! ed embryologici. Dalle mie
recerche sono perVenuto alle seguenti conclusion!:

"1. Che 1' Amphiptyches ed il Gyrococtyle devono considerarsi, come gia
pensavano il Wagener ed il Diesing, due specie distinte del gen. Gyro-
cotyle. Le due specie si chiameranno: G. rugosa Diesing e G. urna Grube
e Wagener.

"2. Che il gen. Gyrocotyle deve allogarsi fra i Cestodi e riguardarsi
affine all ' Amphilina. ' '

He goes on to say that the first conclusion rests on remarkable
uniformity of structure; the sole external difference being the
lack or slight development of the lateral folds in Gyrocotyle.
The second conclusion rests on the mode of development and
internal structure of both species. The similarity of cestodarian

360 University of California Publications in Zoology. [VoL- 6

organization to that of the cestodes is then pointed out in the
musculature, the lack of a digestive tract, the arrangement of
the excretory system, the general form of the nervous system,
the arrangement of the genital organs, and the presence of a
six-hooked embryo, in Gyrocotyle rugosa.

In the same year (1889c), in examining the Entozoa of the
British Museum, Monticelli found an undoubted example of
G. rugosa, taken from Callorhynchus antarcticus at Dunedin, New
Zealand. As he points out (1890), G. rugosa had up to this time
been reported only from Mactra edulis, G. urna only from
Chimaera monstrosa. He summarizes his conclusion as follows :

"1. Che il genere Gyrocotyle e parassita proprio della famiglia delle
Chimaeridae.

"2. Che i due generi finora conosciuti della famiglia (Chimaera e
Callorhynchus) albergano ciascuno una specie del genere Gyrocotyle.

"3. Che il genere Gyrocotyle perviene nelle Chimaeridae per mezzo
di molluschi bivalvi. "

Monticelli discusses the question whether Mactra edulis is to
be regarded as the true intermediate host of G. rugosa, and con-
cludes that, since the example found in the mollusk contained
hooked embryos, a third host must be found to convey G. rugosa
to Mactra. He suggests that possibly some of the mollusks on
which Chimaera monstrosa feeds (probably some member of the
Cyprinidae), may be the intermediate hosts of G. urna.

This work of Monti celli's definitely settles the question as to
the generic identity of Gyrocotyle and Amphiptyches, and estab-
lishes two species of the genus Gyrocotyle, separated by three
distinct characters. Further, it indicates that G. rugosa is a true
parasite of Callorhynchus, though this point could hardly be
considered as established by a single occurrence of the parasite
in the fish.

Further evidence on this question was supplied by Spencer's
(1889) report on the morphology of three specimens of Gyro-
cotyle found by him in the mouth of a specimen of Callorhyn-
chus antarcticus. These Spencer referred to Wagener's Amphi-
ptyches urna, but from the form of the lateral frills and the
presence of hooked embryos, it seems quite clear that his species
was not G. urna, but G. rugosa. (See my pi. 38, fig. 36). This

1911] Watson: The Genus Gyrocotyle. 361

view is substantiated by Monticelli's identification of a specimen
from C. antarcticus as G. rugosa, Diesing 's original species,
though the size, number and character of the frills of the
terminal rosette as figured by Spencer do not agree with Diesing 's
original figure of those structures.

Braun (1889) published a careful review of the literature
of the genus up to and including Monticelli's (1890) paper on
the finding of Gyrocotyle rugosa in Callorhynchus antarcticus,
but not including Spencer's paper. He reduces Amphiptyches
to a synonym of Gyrocotyle.

Lonnberg (1890a) mentions the finding of Gyrocotyle in
July and August, near Bergen. He later (1890b) published a
short paper, "Ueber Amphiptyches Wag. oder Gyrocotyle urna
(Grube et Wagener) Diesing." In a subsequent paper (1891)
he includes a detailed study of Amphiptyches urna, covering the
questions of synonymy, systematic position, occurrence, habitat,
behavior, orientation, and morphology, concluding with a com-
parison of Amphiptyches with the other genera of the Cesto-
daria. Lonnberg agrees with Wagener in regarding Gyrocotyle
as a true cestode ; he considers it the most primitive of the mono-
zoic cestodes and the most closely related to the trematodes. He
does not accept the combination of Gyrocotyle and Amphi-
ptyches in one genus, and regards the form investigated by him,
from Chimaera monstrosa, as Amphiptyches urna. Aside from
his careful study of this species, which will be referred to later,
Lonnberg 's most important contribution to the knowledge of
the genus was his description of the animal in the living condi-
tion and the various forms assumed by it. Its behavior and
appearance he describes as being in the majority of cases similar
to that described by Wagener, namely, pointed at one end, with
marginal frills and a terminal rosette. But this typical body-
form he saw transformed into a totally different one in which
the worms "ganz platt und am beiden Enden gleich sind" (p.
17). (See my pi. 33, figs. 1-4). The ordinary frilled form he
regards as a contracted condition of this ' ' Ligulaa hnlicher "
worm; the intermediate stage between the two he describes as
one in which "die Seitenrander in grossen Wellen gebogen sind
und nur der ausserste Teil des Trichters in einen kleinen krausen

362 University of California Publications in Zoology. tv°L- 6

Kopf zusammengezogen. ' ' He describes also a third contraction
stage, differing from the ordinary frilled stage even more widely
than the * * Ligulaahnlicher. ' ' This is the most expanded state
of the worm, described by Lonnberg as follows:

"Wenn der Amphiptyches sich lebhafter bewegen will, verlangert er
sein Korper: die Querrunzeln an der Mitte des Korpers glatten sich aus,
die dichten seitlichen Krausen wandeln sich dann zuerst in weniger
zahlreicher Fallen urn und diese gehen darauf in einige wenige grosse
Wellen iiber, die bald auch verschwinden und die Seitenrander glatt
erscheinen lassen. Der Korper ist nun im Ligulastadium, also platt und
lanzettenformig. Aber gleichzeitig hiermit vollziehen sich am Trichter
und dem Halsteil grosse Veranderungen. Dieser verliert seine seitlichen
Fallen, wird langer und schmaler und erhalt eine cylindrische Form.
Jener verlangert sich unverhaltnissmassig mehr und zwar in der Weise,
dass er von hinten und vorwarts sich in ein cylindrisches Eohr umwan-
delt, der krause 'Kopf' wird immer kleiner und gleichwie vorwarts
geschoben; dies geht natiirlicher Weise so zu, dass nach un nach mehr
von den proximalen Teilen des Trichters zur Bildung des cylindrischen
Eohres angewandt werden; bald sind nur die aussersten Lippen ein wenig
kraus und schliesslich werden auch diese glatt und man hat eine solche
eigentiimliche Form, wie fig. 36 [my pi. 33, fig. 2] zeigt, vor sich.

Bei hochster Ausdehnung hat dieser Cylinder, wie oben

geschrieben ist, beinahe dieselbe Lange wie der ganze iibrige Korper,
in welchen er allmalig ohne Absatz iibergeht. Es ist mir mit schnell
totenden Eeagentien gelungen einige Tiere in diesem Stadium zu
fixieren, und wenn ich sie mit den gewohnlichen, krausen Formen ver-
gleiche, muss ich gestehen: hatte ich nicht selbst mit eigenen Augen die
Umwandlungen gesehen und die Zwischenstadien studieren konnen, so
wiirde es mir kaum klar werden konnen nur zwei verschiedene Kontrak-
tionszustande vor mir zu haben. " (Lonnberg, 1891, p. 17.)

These statements of Lonnberg 's are in part corroborated by
the observations of Professor Collett (quoted in Lonnberg, 1891,
p. 17, footnote) : "Dass zwei verscheidene Cestoden die Spiral-
klappe Chimaeras bewohnte und zwar ein platter Ligulaahn-
licher und der gewohnliche krause Amphiptyches." In addi-
tion Olsson (1896) reports that among thirty-three individuals
he found "Deux exemplaires, longs de 30mm, avaient la forme
curieuse qui a ete dessinee par Lonnberg (1891, fig. 36) et la
conservent encore en alcool. La longueur du cylindre creux, qui
est forme du cou et de 1'entonnoir, est de pres de 13mm, celle
du corps n'est que de 17 mm." (See Olsson 's fig. 9, p. 509).

Braun (1894) gives a review of the Cestodaria, including
Gyrocotyle, He recognizes two species of the genus, G. urna

1911] Watson: The Genus Gyrocotyle. 363

(Gr. et Wag.) (from Chimaera monstrosa) and G. rugosa Diesing
(from Callorhynchus antarcticus) , but adds (p. 1157) that they
"differiren so wenig von einander, dass es fraglich 1st, ob die
Unterscheidung zweier Species sich rechtfertigen lasst^ moglich-
erweise bestehen Unterscheide in der Lage der Genitalpori. "
According to Lonnberg and Wagener, the vaginal opening is the
most anterior and lateral to the penis ; but according to Spencer,
dealing with G. rugosa, the penis opening is anterior to the
vagina and marginal.

Benham (1891) recognized the single genus, Gyrocotyle, "in
the intestine of Chimaera and Callorhynchus."

Haswell (1902) reported a new species of Gyrocotyle, G.
nigrosetosa, from a new species of Chimaera, C. ogilbyi, trawled
off the Australian coast by the "Thetis." Haswell had one
specimen of G. rugosa, and two specimens of the new species,
"not in good condition for investigation." He did not study
the living form. He distinguishes G. rugosa from G. urna (fol-
lowing Lonnberg 's description of the latter) on the following
grounds :

(1) Relative positions of the apertures of the penis and
vagina.

(2) Presence of an eversible cirrus, adapted to self -impreg-
nation, in G. rugosa.

(3) Spinules lining ejaculatory duct, in G. urna.

(4) Size of eggs.

(5) Presence of hexacanth embryo in G. rugosa.

(6) Shape of spines, which are simpler in G. rugosa.

C. GYROCOTYLE OF THE COAST OF CALIFORNIA.

I. OCCURENCE AND HOST.

As this review of the literature shows, the genus Gyrocotyle
includes three species, G. rugosa, G. urna, and G. nigrosetosa^
all occuring as intestinal parasites in the family Chimaeridae.
The occurrence of a form of the genus in Chimaera colliei, found
off the coast of California, has not to my knowledge been previ-
ously reported. So far as can be determined, no work has ever
been done on the parasites of C. colliei. This is the only species

364 University of California Publications in Zoology. [VOL. 6

of Chimaera found on the Pacific coast of the United States
(Dean, 1906, pp. 6, 7) and was first described by Lay and Bennet,
1839, p. 71, pi. 23, in Zoology of Captain Beechey's Voyage.
Professor Bashford Dean (1906) bases his work largely on eggs
of C. colliei, and gives a description of the living fish, with notes
on its occurrence, habitat, food and breeding habits. It occurs
in depths of from 5 to 10 fathoms, being found in shallower
water in the Puget Sound region than along the California coast.
The specimens examined by the writer have come from the fishing
grounds oft3 Pifios buoy in Monterey Bay, mentioned by Dean
(1906, p. 15) ; from Cabral's banks off San Diego; and to the
north of San Diego, off La Jolla, in depths up to fifty fathoms.
Dean's observations (1906, p. 20) on the food habits of this
species, are as follows :

"In view of the special character of the dentition of Chimaera, one
would naturally expect its food supply to be definite in character. The
examination of the contents of its gut, however, showed (C. colliei) singu-
larly omnivorous habits. It is true that the broken shells of mollusks
are commonly found, -as well as fragments of good-sized crustaceans, as
indeed the scanty literature records. Thus, in the gut of C. monstrosa
Faber finds Crustacea and shell-fish fragments; Monticelli, quoting
Liitken, Cyprina islandica; and Olsson, broken shells (Leda and Venus)
and bits of large decapods. Olsson finds also (and his observations are
the most detailed hitherto published on the feeding of Chimaera) chaeto-
pods, amphipods, echinoids and polyps. In C. colliei observations on
about a score of individuals showed a singular mixture of foods. The
most numerous were vertebral columns of small isospondylous fishes, a
few mollusk shells, usually greatly crushed, a quantity of sand and fine
gravel, squid, nudibranchs and opisthobranchs, bits of cases, jaws and
setae of annelids, and occasionally a fragment of a crustacean. In one
instance the gut was filled with seaweed. One is not surprised, there-
fore, that this species is taken readily with various baits."

The observations on stomach contents made by Professor
Kofoid and myself yield results agreeing in general with the
above; with the addition that the stomachs contained a great
quantity of echinoderm spines, plates, etc. Small fish were com-
mon, as were fragments of lamellibranchs, nudibranchs and gas-
teropods. Cephalopod beaks were almost always found. Crus-
tacean fragments, especially of Hippa and Blepharada, were not
uncommon in the San Diego specimens. The fish is so nearly
omnivorous in food habits that no definite clue as to the life-
history of its parasites can be obtained from these data.

1911] Watson: The Genus Gyrocotyle. 365

The examination of C. colliei for parasites has yielded the
following results :

(1) The fish is almost invariably infested with a parasite
belonging to the genus Gyrocotyle. The parasite was- found in
34 of the 38 specimens examined.

(2) The parasite rarely occurs singly, but usually two indi-
viduals in one host. More than three in one fish I have never
found. In four cases a single individual was found.

(3) No other adult cestodes have been found in the alimen-
tary tract of the fish. Encysted cysticerci are common in the
walls of the tract, in the mesenteries, liver, etc., especially during
the summer months. Parasitic crustaceans were frequently
found on the gills, and an aspidocotylean, probably belonging to
the genus Macraspis, was found embedded in the muscles of the
rectum. A distome, bearing a strong superficial resemblance to
a small Gyrocotyle without frills, was found once in the posterior
region of the rectum, and once in the body-cavity.

(4) Two distinct species of Gyrocotyle are found in C. col-
liei; one, the less common of the two, is closely related to and
perhaps identical with, Gyrocotyle urna; the other differs in
several characteristics from any species heretofore described.
The incompleteness of the figures and descriptions of G. urna
make it impossible to determine accurately the identity of the
form first mentioned with G. urna. It will be referred to as G.
urna (var. ?) . The second form, which, in view of its well-defined
peculiarities and of the absence of any intermediate forms link-
ing it to any described form, I regard as a new species, will be
refered to as G. fimbriata.

Gyrocotyle fimbriata, sp. nov.

Diagnosis. — Color, creamy white ; length 30-55 mm, width
7-12 mm.; rosette posterior, folds complex; lateral frills 3-5 mm.
deep, much folded, passing into smooth lateral fin near anterior
extremity; spines numerous, distributed in definite pattern,
points directed anteriorly; eggs about 0.09 mm. in average long
diameter, no hooked embryos; penis opens mediad of vagina, at
about the same antero-posterior level. Occurs in spiral valve of
Chimaera colliei; can leave host and live for some days free.

366 University of California Publications in Zoology. [VOL- 6

These two species occur in fish from the same locality, even
taken on the same trawl. Both species have never been found in
the same individual. G. fimbriata is the more abundant of the
two.

II. GROSS ANATOMY OF GYROCOTYLE.

The characteristics on the basis of which species are separated,
as given in the literature of the genus, are of various degrees of
usefulness and trustworthiness. This is due in great measure
to the manifold changes in form and appearance of the living
animal and of specimens preserved by different methods in all
stages of contraction and deterioration. Before they can be
intelligently considered, a brief description of the gross struc-
ture of the animal and of the external form and behavior of the
living specimen will be necessary. First, the orientation adopted
in this paper must be defined. The pointed acetabular end is
the anterior, the rosette or canal end the posterior. The surface
on which lie the uterine pore and opening of the canal is the
creeping surface and the one to which the animal returns in rest.
It is therefore ventral. The following description refers espec-
ially to G. fimbriata.

The worm when alive and attached is almost translucent.
After detachment it becomes opaque, whitish-yellow in color,
decidedly deeper in tone in the marginal frills and the folds of
the terminal posterior rosette. The anterior end is not frilled, is
highly contractile, and in the living animal is in constant motion
(pi. 33, figs. 7-9). It consists of a very muscular acetabulum,
whose margin can be retracted somewhat, while the whole can be
drawn back into the body by the retraction of the longitudinal
muscles attached to the sucker. Posterior to the acetabulum, in
the median dorsal line, lies the opening of the uterus (ut. po.,
pi. 39, fig. 42). Anterior to this, one-third of the distance be-
tween the uterine pore and the posterior margin of the aceta-
bulum, lie the vaginal opening (vag. op.), on the ventral surface,
and the penis opening (p. op.), on the dorsal surface. Both lie
almost on the margin, but the penis opening is the more mediad
and anterior to the vaginal pore. In the median third of the
body of the adult, the much coiled and distended uterus occupies

1911] Watson: The Genus Gyrocotyle. 367

most of the space. In front and to either side of it, running
forward to the posterior margin of the sucker, are the follicles
of the testes. Peripheral to these, running out into the lateral
folds, are the follicles of the vitellaria, distributed from the
acetabular region to the base of the posterior rosette. Along
the dorsal surface and to the left of the uterus runs the vagina,
leading to the large receptaculum seminis (rec. sem.) located
just posterior to the uterus. On each side of the receptaculum
seminis lie the follicular ovaries (ovar.) ; immediately in front
of it are the vitelline ducts (vit. d.) and the first coils of the
uterus. The shell-glands (sh. gl.) are found around these ducts.
It is in this region that the ova are fertilized, unite with the yolk-
cells and receive their shell-coating. Behind the receptaculum
seminis and the ovaries lies the ventral opening of the rosette
canal. This structure may be regarded as consisting of two
parts. The first is a flaring " funnel," its opening terminal, its
margin frilled and folded till it resembles a carnation, or as the
Germans put it, a "Kohlkopf." This funnel passes directly
acetabulad ; it leads into a narrow canal, which turns at almost a
right angle to the course of the funnel, passing to the ventral
surface (pi. 34, fig. 15). This ventral opening I have called the
" canal opening" (can. op., pi. 46, fig. 76), in distinction from
the terminal funnel opening above described. Funnel and canal
together are called by German workers the ' ' Trichter ' ' ; Spencer
speaks of the funnel-margin as the " rosette" and calls the canal
opening the "proboscis."

Gyrocotyle fimbriata (pi. 48, figs. 80, 81), while bearing a
general resemblance to the described species of the genus, differs
markedly in certain external characteristics. The body in ex-
panded condition is about four times as long as wide ; the lateral
frills never totally disappear in any stage of expansion, but can
be distinguished from the median portion of the body under all
conditions. They are about one-fourth the total width of the
body in depth. In contracted specimens the folds of the two
sides are invariably drawn toward each other on the ventral
(canal opening) surface. The posterior terminal rosette (post,
ros., pi. 34, figs. 10, 12-15), is from one-half to three-quarters the
greatest width of the body in transverse diameter. Its depth is

368 University of California Publications in Zoology. [VOL- 6

about one-third its width, and its dorso-ventral diameter varies
from one to one-half the transverse diameter, or even, in excep-
tional cases, to twice this diameter. The folds of the rosette are
not simple, but are secondarily folded, standing in marked con-
trast to the lateral folds, which consist of a single series of undu-
lations. The appearance suggests that the outer wall of the funnel
is more contractile than the inner, the result being the production
of a multitude of small secondary folds on the inner surface of
the funnel. The gross appearance of this type of rosette suggests
a finely villous surface, such as an intestinal mucosa. The large
folds with long axis parallel to the long axis of the body, so
prominent in other species, are almost obliterated here, especially
in preserved specimens.

The anterior or acetabular extremity (pi. 33, figs. 7-9 ; pi. 36,
figs. 22, 23), is bluntly pointed and bears a lateral "fin" of tissue,
to a point about half-way to its tip. The acetabulum (acet.) is
contractile within itself to a high degree and the structure as a
whole can be drawn back into the body by the strong longitudinal
muscles. This occurs in the most strongly contracted stage (acet.,
pi. 34, fig. 10). The body is markedly asymmetrical, owing to the
formation of a deep "genital notch" in the left margin (gen.
notch). The vaginal opening lies at the angle of this notch on
the ventral (canal opening) surface, appearing as a narrow slit
bounded by a slightly elevated ring; the penis opens dorsally in
a pit at the base of a very low and inconspicuous mound, about
two-thirds of the distance from the apex of the notch to the
median line (p. op., pi. 36, fig. 23). The relation of penis to
vagina varies within narrow limits, but in general they may be
said to be at practically the same antero-posterior level.

The lateral frills extend anteriorly to about the level of the
genital notch, in front of which they are continued as the un-
f rilled "lateral fin" above mentioned. Posteriorly they extend
to the tip of the posterior rosette. The funnel, whose margin is
formed by the posterior rosette, narrows rapidly to a small canal,
which turns at an angle slightly greater than 90°, and opens by a
small aperture on the ventral (vaginal opening) surface (can. op.,
pi. 46, fig. 76). The margin of the aperture is raised into a ring,
and is crenate (can. op., pi. 34, figs. 14, 15).

Watson: The Genus Gyrocotyle. 369

Specimens taken from decaying fish have very generally
shown a marked diminution in both lateral frills and terminal
rosette. Comparison of these with behavior of fresh specimens
kept in culture media shows that the diminution is due to an
actual disintegration of tissue. The lateral frills are cut off along
a perfectly definite line, running in a slightly irregular antero-
posterior course. The line can be made out some twenty-four
hours before autotomy actually occurs. In every case where a
specimen appears to have no lateral frills or noticeably reduced
ones, examination shows a cut edge, along which the frills have
dropped off. It is possible that Diesing's original figure was
taken from a specimen in which this autotomy had taken place.
The frills of the terminal rosette do not appear to be cut off in
this regular fashion, but disintegrate on the margins only ; ragged
strings of tissue will be found attached to the body of the rosette.
Most of the reduction in the size of the rosette in decayed speci-
mens I believe to be due to intense contraction of that region,
though undoubtedly accompanied by a certain amount of actual
disintegration of tissue. When the parasite is left in the dead
fish over twelve hours, at ordinary temperature, this decay begins.

The spines characteristic of this genus are very prominent in
G. fimbriata. Their arrangement is in a constant pattern, char-
acteristic of this species. This pattern is described in detail
later, under the discussion of the cuticula, but its general outline
will be given here also for the sake of convenience. In the
anterior region, borne on the lateral "fins" above mentioned, are
from five to seven rows of very large black spines (pi. 36, figs. 22,
23.) They are borne on very low papillae. There are about
twenty-five of these spines ; they vary in length, the more anterior
and marginal ones being in general the larger. Their form and
structure are discussed below, in connection with the description
of the cuticula.

In the posterior region of the body, the spines are longer than
those above described. They are borne on large rounded papillae.
They form a "collar" (pi'. 34, figs. 10, 12) around the region of
the body between the funnel opening of the rosette and the canal
opening. On the dorsal surface, no spines are found in front of
this collar, which does not extend quite to the level of the canal

370 University of California Publications in Zoology. [y°L- 6

opening. On the ventral surface a line of scattered spines
streams off laterally and anteriorly from the tips of this semi-
circle to about the level of the receptaculum seminis. No spines
are found on the lateral frills. The arrangement of the spines on
the ventral surface (pi. 34, fig. 10) is a fairly constant one. It
is of interest when considered in connection with Lonnberg's
(1891) theory of the formation of the funnel and canal by means
of the partial fusion of folds of the ventral body wall.

Gyrocotyle urna (Wagener) var.?

In the material collected by Dr. Kof oid near Monterey, off the
California coast in 1904, were found three specimens differing
from the rest in certain characters which seem to constitute a
basis for specific distinction. Later, in 1907-1908, the writer
found several similar specimens. The characters peculiar to this
form are as follows:

1. The lateral frills are less voluminous.

2. The folds of the posterior rosette are simple (i.e., not
thrown into irregular secondary frills, as in G. fimbriata), vary-
ing in diameter from one-sixth to one-third of the width of the
body.

3. The lateral frills do not extend to the base of the rosette
but taper off gradually, disappearing posterior to the recepta-
culum seminis.

4. Spines are present over the whole of the ventral surface
as far forward as the middle of the length of the uterus, and
extend thence in two lateral wings forward to a point just back
of the level of the opening of the uterus. There is no distinct
pattern discernible as in G. fimbriata. There are only a few
scattered spines on the dorsal surface, except around the
"collar," immediately in front of the base of the frill of the
posterior rosette.

5. The relative position of the opening of the penis, uterus,
and vagina is markedly different from that seen in G. fimbriata.
(See measurements, p. 381.)

Except in two particulars, i.e., the distribution of the spines
and the fact that penis and vagina open at approximately the
same antero-posterior level, this form seems to be identical with

1911] Watson: The Genus Gyrocotyle. 371

Gyrocotyle urna as described by Wagener and Lonnberg. The
distribution of spines may vary with age, though it has been
found constant in specimens of all sizes of G. fimbriata. It is
also possible that the non-appearance of the spines may be due
to action of reagents on preserved material. The figure given by
Wagener (1852) shows a very close approach to the pattern given
above as characteristic for G. fimbriata. The whole question of
distribution of spines cannot be settled until more material is
available.

The question whether penis or vagina opens more anteriorly
is a difficult one. In specimens killed between glass plates, these
relations can be very easily distorted. On the other hand, in
specimens not killed thus, it is very difficult to compare the posi-
tions of the two openings, as they are on opposite surfaces of a
body too thick to be made transparent by clearing agents. The
specimens in my possession show the two openings at practically
the same level, with a small range of individual variation, depend-
ing upon the degree of extrusion of the penis-papilla.

Not having specimens of G. urna for comparison, it has been
impossible to decide whether this form is really a distinct one, or
belongs to Wagener 's species. Until comparison can be made, it
will be referred to as G. urna (var. ?).

III. BEHAVIOR OF THE LIVING ANIMAL.

The shape of the body varies greatly in the living animal, and
also in preserved specimens. Wagener 's figures (pi. 48, figs. 84,
85) are the best in the literature, so far as giving a faithful re-
production of the appearance of the living worm. His figure of
the preserved specimen, on the other hand, resembles very slightly
any specimen that has come under the writer's notice. The
acetabular extremity of the living animal is extended into a long,
slender, cylindrical proboscis, along the sides of which the spines
above referred to are distributed. This part of the worm is in
constant motion. Posterior to the genital openings, in the region
of the lateral frills, the body is flattened, slightly convex dorsally.
Posterior to the uterus the lateral frills become less deep, and the
body less flattened. This " neck ' ' region can be very considerably
extended. On the posterior extremity is borne the rosette, of

372 University of California Publications in Zoology. [V°L. 6

greater size in G. fimbriata than in G. urna, and its folds many
times as complex, but generally held in a position exactly like that
shown in Wagener's second figure, that is, the opening of the
funnel, the rosette surface, lying, not at right angles to the sur-
face of the body, but parallel to it.

The .stage shown in Wagener 's first figure represents the most
extreme stage of expansion noted by the writer. No stages corre-
sponding to those figured by Lonnberg (pi. 33, figs. 1-4) were
seen, although the specimens observed were in many cases per-
fectly fresh and very active, contracting and expanding with
great freedom, and moving about in the dish. But the shape of
the posterior rosette was never materially altered, nor did the
lateral folds completely disappear. Further, in the observations
by Wagener, by Dr. Kofoid, and by the writer on the worm, an
expanded condition always involves expansion of the acetabular
extremity as well as the neck of the rosette. None of Lonnberg 's
figures shows anything suggesting the very characteristic aceta-
bular extremity as first figured by Wagener. In observations of
twenty-three living worms the writer has never failed to find the
acetabular extremity assuming and maintaining this appearance
throughout the period of activity. Lonnberg 's observations, as
above noted, are corroborated by Olsson (1896), who adds the
astonishing statement that the worms preserve this form in
alcohol. As the parasite is exceedingly sensitive to chemical
stimuli, and invariably reacts to even a very dilute solution (5
per cent.) of alcohol by strong contraction, this seems to indicate
that Olsson at least was dealing with some form other than
Gyrocotyle. The writer found on two occasions, while searching
for Gyrocotyle in Chimaera colliei, a distome strongly suggesting
Lonnberg 's figure and maintaining its shape in killing fluids,
which might have been mistaken for Gyrocotyle. But since Lonn-
berg states explicitly that these transformations in form took
place under his eyes, this explanation seems scarcely possible.
We are forced to accept a discrepancy in results here, only to be
removed by further investigation.

The living worm, in the intestine of a fish that has just been
caught, is translucent, of a dull pink color, exactly similar to
the intestine on which it lies. The worm in perfectly fresh fish

1911] Watson: The Genus Gyrocotyle. 373

is firmly attached by the terminal rosette, which is spread over
an area of ten to twelve millimeters in diameter. The folds of the
rosette fit over the vi.lli of the intestine. The canal opening (on
the ventral surface) is closed as long as the rose tte_ remains
attached. A series of waves of contraction passes from the
margin of the rosette toward the canal opening, during the
attachment of the rosette. The lateral frills are present even in
the most expanded state of the worm, constituting one-half of
the total width of the body. The acetabulum is very active,
thrusting itself backward, under, over or to the side of the body.
The whole worm contracts and expands frequently without any
apparent stimulus. The parasite is not easily affected by
mechanical stimuli under these conditions, but will contract if
exposed to direct sunlight, and is very sensitive to chemical
stimuli. The acetabulum has never been found attached, or in
any sense functioning as a sucker. This statement is borne out
by all investigators.

Shortly after the death of the fish, the fluid intestinal con-
tents become thick and opaque, and the worm detaches itself and
contracts, losing its translucence and becoming creamy-white in
color. The region of the mucous membrane to which the rosette
has been attached is of a deep purplish red, and appears inflamed
after the rosette has detached itself. Many of the worms are
found in this opaque condition, and do not again become active.
From observations on the behavior of the attached specimens, and
from the fact that inactive specimens are found in all parts of
the intestinal tract, in the mouth and on the gills, it appears
fairly certain that the worm attempts to leave the dead host, and
very frequently succeeds in doing so. In one case, the worm was
found crawling up the first turn of the valve, when the intestine
was opened. The fact that specimens are so often found,, either
in the first or next to the last turn of the valve, also points to this
conclusion. Finally, the finding of a living specimen crawling
on the bottom of a jar of sea-water in which a Chimaera had
been placed, is fairly conclusive evidence that the worm occa-
sionally succeeds in escaping from its host. The parasite is never
found attached elsewhere than in the spiral valve. Spencer
(1889, p. 138) remarks in this connection: "The specimens

374 University of California Publications in Zoology. [VOL. 6

[of Gyrocotyle rugosa] obtained by myself — three in number —
were taken from the mouth of Callorhynchus antarcticus, and
their presence in such a position was doubtless associated with
the fact that the fish had been dead some twenty-four hours,
though the parasites were still living and evidently trying to find
their way out of the dead body." Wagener (1852, p. 543) says:
"Nur einmal fand es sich an den Kiemen, wobei jedoch bemerkt
werden muss, dass die Fisch schon 12 Stunden ausserhalb des
Wassers sich befand." Lonnberg (1891) mentions the finding of
G. urna living free, but supposes the worms to have finished
their cycle and to have been cast out with the excrement of the
host. He seems to imply in the following statement that the
worms leave the dead host (writing of the almost invariable
occurrence of the parasite in Chimaera) : ''In zwei anderen, die
ich in Upsala untersuchte, die aber von Trondhjem stammten,
habe ich ihn nicht gefunden, aber diese Fische waren schon seit
mehreren Tagen ausserhalb des Wassers." It may then be re-
garded as an established fact that the parasite is capable of a
very considerable amount of locomotion, and that it can, under
favorable conditions, leave its. host and live for some days outside
the intestinal tract. The most important condition seems to be
temperature; a low temperature (about 10°-12° C.) is most
favorable. In warm weather the worms remain active for only
a very short time, and have never succeeded in getting out of
the intestine in any case under my observation. This ability
to leave the host, taken in connection with the behavior of the
isolated parasite as described below, has an important bearing
both on the problem of orientation and on the question of the
habitat of the parasite, which will be discussed later.

The worms on being removed from the fish were placed in
culture fluids. The worm is very active, if removed soon after
the death of the fish and kept at a low temperature. The culture
solution quickly becomes contaminated with bacteria, and the
worm contracts and grows sluggish. When changed to fresh
solution the parasite expands and renews its movements. These
consist of the following :

1. Thrusting and exploring movements of the acetabular ex-
tremity, accompanied by contraction and expansion of the body.

1911] Watson: The Genus Gyrocotyle. 375

2. " Kighting-up " movements, in which the worm turns from
one surface to the other. These take place when the worm is
placed on its dorsal- surf ace and also when it is exposed to strong
light.

3. This turning over is accomplished in two typical ways,
either by turning completely over on one side on the frills as an
axis, or by a more elaborate process of tucking under the aceta-
bulum until it emerges posterior to the rosette, when the rosette
is elevated and the overturning accomplished after the manner of
a somersault.

4. A rotation from side to side of the terminal rosette, accom-
panied by a wave-motion in the margin of the frills, takes place
during the worm's most active period.

5. Locomotion in a definite direction is accomplished by means
of a series of expansions and contractions, with some help from
the mobile lateral frills. The acetabular extremity is in every
case directed anteriorly; the rosette is held in the typical posi-
tion, raised, with the canal opening ventral and the funnel open-
ing dorsal. The acetabulum is thrust out, and the whole body
elongates, the rosette remaining stationary. This expansion
leaves the rosette end where it was before, the acetabular end
some two centimeters further toward the goal than at start.
Then a wave of contraction sets in, bringing the rosette end
nearer the acetabular end, which also moves backward a short
distance, approaching the rosette. The net result is an advance
in the acetabular direction. The acetabulum remains contracted,
while the rest of the body expands. In this expansion the rosette
must either retreat, or the whole contracted anterior extremity
be shoved forward. The latter is what actually occurs, as shown
by careful marking of the relative positions of the two ends
before and after the wave of expansion. This method of pro-
cedure is very effective, resulting in an average gain of about
two centimeters for each wave of contraction. A distance of
fourteen centimeters was traveled in ten minutes by one speci-
men ; but in this case more than the average gain was made, the
whole distance being accomplished in five contractions.

376 University of Calif ornia Publications in Zoology. [VOL. 6

IV. CHARACTERISTICS FOR SEPARATION OF SPECIES.

The characteristic peculiarities of each species have been set
forth by its discoverer, but there has been no systematic attempt
to assess the worth of these characteristics for the genus as a
whole since that of Braun (1894). He was not acquainted with
at least one very important set of species characteristics, that
proposed by Haswell (1902).

The first species characteristics were proposed by Diesing
(1859). He distinguishes G. rugosa from G. urna by the pres-
ence of lateral frills and spines and the greater size of the tail
rosette in that species.

Monticelli also recognizes these two species, separating them
by the following points :

1. The absence in G. rugosa of the frilled lateral margins.

2. The greater length of the uterus in G. rugosa, possibly cor-
related with the following character.

3. The presence of hooks in the embryo within the uterus of
G. rugosa.

Braun (1894) recognized G. urna and G. rugosa, but only as
doubtfully distinct. The basis for separation he finds in the
position of the opening of the vagina and penis; the former
being the more anterior in G. urna, the latter in G. rugosa.

Haswell (1902) recognized G. urna and G. rugosa and tenta-
tively proposed a new species, G. nigrosetosa. He distinguished
G. rugosa from G. urna on the following grounds :

1. Relative position of the apertures of penis and vagina.

2. Presence of an eversible cirrus adapted to self-impregna-
tion in G. rugosa.

3. Spinules lining the ejaculatory duct in G. urna.

4. Presence of hexacanth embryo in the uterine egg of G.
rugosa.

5. Shape of spines.

G. nigrosetosa he distinguishes on the following grounds :

1. Size of eggs, 0.08 mm., as distinct from eggs of G. rugosa,
0.1 mm.

2. Presence of operculum, distinct from G. rugosa and G.
urna.

1911 J Watson: The Genus Gyrocotylc. 377

3. Absence of hooked embryos.

After careful comparison of the available figures and descrip-
tions and a study of about forty specimens, ranging in state of
preservation from those perfectly fresh, attached to the_ intestine
of the host, down to badly disintegrated specimens from various
regions of the alimentary tract of decomposing hosts, the writer
has found certain of these proposed criteria to be useful and
valid. Certain others seem untrustworthy. These specific char-
acters may be listed and criticized as follows :

1. Presence of lateral frills. — It seems fairly certain that
lateral frills are present in all species of the genus, when the
specimen is in a good state of preservation. Presence of the frills
is therefore not a basis for specific distinctions. The amplitude
of the frills, and their relative width (in terms of total body
width) for any given state of contraction or expansion, is how-
ever a characteristic of real value in the recognition of species.
G. rugosa (pi. 38, fig. 36) has comparatively scanty and narrow
frills ; in G. nigrosetosa the frills are deeper and more volu-
minous; in G. urna (pi. 47, figs. 84, 85) they are twice the depth
of those in G. nigrosetosa and many times as voluminous. In
G. fimbriata (pi. 34, fig. 10) they are still more ample and about
the same depth as in G. urna. While these differences can be
easily seen, the extreme contractility of the animal makes it im-
practicable to attempt to use this characteristic as a basis for
separation of species. It is a concomitant, rather than a critical
characteristic, not available for exact description.

Diesing's specimen without frills was taken, it will be recalled,
from Mactra. According to the best of our knowledge the only
host of the adult parasite is some member of the family Chimaer-
idae; this specimen, having- been set free from its host in some
fashion, probably entered the shell of the mollusk by accident.
Certainly not adapted to life as an ectoparasite, processes of
disintegration would undoubtedly speedily set in, and my own
experiments with Gyrocotyle in culture media show that dete-
rioration invariably begins by cutting off of the lateral frills in
increasingly deeper layers.

2. Presence of spines. — This is critical only when applied to
living specimens and to those which have been preserved in such

378 University of California Publication* in Zoology. tv°L- 6

a manner as not to injure the spines. In addition to the effect of
reagents, disintegration results in the early loosening and dis-
appearance of the spines. In no case where the living worm has
been examined have the spines been absent.

The distribution of the spines affords a useful basis of dis-
tinction. In G. rugosa (Spencer, 1889), "spines are scattered
over the whole surface of the body"; in G. nigrosetosa (Haswell,
1902) "they are mainly confined to the dorsal surface, except at
the anterior end. Over the rest of the ventral surface are
scattered a very few, much smaller than those on the dorsal sur-
face." In G. urna (Lonnberg, 1891), the distribution is confined
to the extremities and the lateral margins of the body, leaving
the middle of the body, from just in front of the canal opening
to the level of the acetabulum, free from spines. In G. fimbriata
the condition is as described by Lonnberg for G. urna, except
that spines are absent from the greater part of the lateral margin
of the body, and do not occur on the lateral frills. Possible
variation with age and the ease with which spines may be lost
by slight disintegration or action of reagents, makes any dis-
tinction resting on them somewhat impracticable. Here again
we have differences coordinated with, but scarcely useful as,
specific distinctions.

3. Size of tail rosette. — This character appears to depend less
on the state of contraction than do the width and amplitude of
the lateral folds, and if reduced to a definite statement in terms
of some other dimension of the animal, would probably be found
to vary about a mean characteristic for each species. The
astonishing illustrations given by Lonnberg (pi. 33, figs. 1-4)
would, if corroborated, not necessarily invalidate the worth of
this characteristic, provided comparisons were made between
specimens in a similar state of contraction, G. rugosa (pi. 33,
figs. 5, 6) in a contracted state has a rosette about one-fourth its
greatest body-width in diameter. In G. nigrosetosa, the rosette
is about five-sixths the body- width; in G. urna about one-half;
while in G. fimbriata it varies from two-thirds to three-fourths
or even more.

4. Length of uterus. — This is a character which it is exceed-
ingly difficult to apply. The length of the much-contorted

1911] Watson: The Genus Gyrocotyle. 379

uterus is hard to estimate, the number of turns may vary with
the state of contraction, and certainly varies with age. I have
been unable to reduce this to definite measurements. From
Spencer's diagram it seems probable that the uterus occupies
more space in G. rugosa than it does in G. fimbriata, or, accord-
ing to Haswell, in G. nigrosetosa.

5. Presence of hooked embryos in uterine eggs. — This is a
character easily applied and apparently thoroughly critical. It
serves to set off G. rugosa from the rest of the genus.

6. The relative positions of the openings of the penis and
vagina. — This character is perfectly definite and, applied to ma-
terial not flattened under pressure when killed, seems trustworthy.
There is one possible source of confusion, in the fact that the penis
opening is on the summit of a papilla, very mobile and capable
of considerable extension. The vagina is fixed in position. In
case this papilla should be fully extended, the penis opening
might lie in front of the vagina ; while if withdrawn it would lie
well behind it. The same criticism might hold for medio-lateral
relations. The fact that G. rugosa, the only form in which the
penis papilla has been observed greatly extruded, is also the only
one in which the opening is definitely in front of the vaginal
opening, suggested this possible source of error. Spencer's figure
seems to point in this direction.

7. Presence of an eversible cirrus, adapted to self-impregna-
tion.— This arrangement has been reported only by Haswell for
G. nigrosetosa. It does not seem probable that only one species
of the genus possesses a type of cirrus adapted to self-impregna-
tion, especially when it is reflected that single individuals of all
reported species have been found repeatedly in their hosts. It
is strange that no eversion of the cirrus has ever been noted in
any except the one specimen of G. rugosa studied by Haswell,
and possibly in the one figured by Spencer.

8. — Spinules lining ejaculatory duct. — This character is of
service in setting off G. rugosa from the rest of the genus;
spinules (interpreted by Lonnberg as cilia) have been observed
in the ejaculatory duct of G. nigrosetosa by Haswell, of G. urna
by Lonnberg, and in G. fimbriata by the writer.

9. Shape of spines. — The data on this point need revision.

380 University of California Publications in Zoology. [VOL. 6

The shape of the spines may vary somewhat in different regions
of the same animal. Furthermore, all the spines thus far de-
scribed correspond exactly to those figured by Lonnberg, with
the exception of Spencer's figure of the spines of G. rugosa.
Possibly a study of more individuals of the latter species may
show closer similarity to the type of the genus.

10. Size of eggs. — This character must be regarded with some
distrust, in view of the fact that one individual may yield ripe
eggs varying in size from 0.075 mm. to 0.115 mm.

11. Presence of an operculum in the egg-shell. — Haswell
(1902) figures an egg of G. nigrosetosa, having a thick shell,
with a thin plate at one pole, in diameter less than one-third the
greatest transverse diameter of the egg (see my plate 47, figure
81.) "A circular area of the shell at one pole is much thinner
than the rest, and is probably differentiated as an operculum."
The same observer studied eggs of G. rugosa, and states that in
them the shell does not appear to be provided with an operculum.
The eggs of G. fimbriata show when discharged no such thin plate ;
but eggs taken from poorly preserved specimens show an oper-
culum, its diameter a little more than half the greatest transverse
diameter of the egg, in which the shell is slightly thinner than
elsewhere. Its margin is serrated. This cap differs in size,
thickness, and character of margin from the operculum of Has-
well, but seems to be an homologous structure. The significant
point is that there is no trace of it in the freshly discharged ova,
and that it does appear very clearly later. Probably the ova
of G. urna and G. rugosa would also show a similar structure,
at a similar stage. The problem is whether Haswell's oper-
culum is present in a freshly discharged or intra-uterine egg, or
whether it is merely an early stage of the formation of this cap.
Only in the former case does the operculum become a good
specific character. Haswell explicitly observes that "the speci-
mens were not in good condition for minute investigation." His
figure (see my plate 47, figure 80) shows the neck of the rosette
as swollen, recalling an appearance characteristic of worms kept
too long in culture media. But in the absence of any definite
statement as to the condition of the two specimens of G. nigro-

19111 Watson: The Genus Gyrocotyle. 381

setosa, the presence of an operculum in uterine eggs must be
adopted as a specific characteristic of G. nigrosetosa.

In addition to these characters which have been proposed and
found trustworthy, the writer suggests the following:

1. The character of the folds of the terminal rosette, whether
simple or complex. This is correlated with the greater relative
size of the rosette ; but the fact that the size varies greatly and is
difficult to express in a ratio that could be applied to any speci-
men to be determined, makes the character of the folds them-
selves seem a more easily applicable test. The writer is convinced
that this character is constant in all stages of contraction, at all
ages, and after treatment with all the ordinary reagents. It
becomes unrecognizable only after decay has proceeded very far
indeed.

2. The ratio between the distance from the opening of the
uterus to the tip of the acetabulum, and the distance from the
opening of the uterus to the level of the opening of the penis.
While these absolute distances vary enormously with the state of
contraction, a series of measurements shows that the ratio is
fairly constant about a mean characteristic of the species.

TABLE I.
Eatios of Measurements of Position of Genital Pores in Gyrocotyle.

G. Jimbriata G. urna G. rugosa G. urna (Wag.) G. nigrosetosa

.4615 .266 .515 .333 .490

.4500 .222 .582

.4444 .200 .592

.4444 .200

.4375
.4375
.4295
.4295
.4295
.4290
.4000
.4000
.3809
.3333
.3333
.3333

382 University of California Publications in Zoology. [VOL. 6

V. KEY TO SPECIES OF GYBOCOTYLE.

On the basis of the above discussion, the following classifica-
tion is proposed :

A. Frills of posterior rosette simple.

I. Uterine eggs containing hooked embryos. Opening of penis

laterad of, and anterior to, vaginal opening.

G. rugosa.

Host: Mactra edulis. Diesing (1850, p. 408) Brazil. Mon-
ticelli (1889b, p. 228).

Callorhyncus antarcticus. Monticelli (1889c, p. 327),
Dunedin, N. Z. Spencer (1889, p. 138), Australia.

II. Uterine eggs not containing hooked embryos. Opening of penis
mediad to vaginal opening.

(1) Opening of penis posterior to vaginal opening.

(a) Eggs without operculum.

G. urna.

Host: Chimaera monstrosa. Wagener (1852, p. 545). Nice.
Kroyer, I)anmark's Fiske, v. 3, p. 813. Van Beneden
and Hesse (1864, p. 54). Olsson (1868, p. 58; 1896, p.
508). Lonnberg (1891, p. 14). Bergen, Christiania.
Monticelli (1889b, p. 229; 1889b, p. 327). Mediterranean.
Hansson. Free-living off the west coast of Sweden.
(Lonnberg, 1891, p. 14).

(b) Eggs with operculum.

G. nigrosetosa.

Host: Chimaera ogilbyi, Haswell (1902, p. 48). Australia.

(2) Opening of penis at the same level as vaginal opening.
G. urna (var?)

Host: Chimaera colliei, from the California coast. Has
been found attached by terminal rosette to mucosa of
spiral valve.

B. Frills of posterior rosette complex. Opening of penis mediad to
vaginal opening, at same level.

G. fimbriata.

Host: Chimaera colliei: off the California coast. Has been
found attached as above, free in intestine, and in mouth
and on gills of host.

1911] Watson: The Genus Gyrocotyle. 383

D. GENERAL MORPHOLOGY AND HISTOLOGY OF
GYROCOTYLE.

I. MATERIAL AND METHODS.

The structure of the worm has been studied in living speci-
mens, in whole mounts stained and cleared, and in serial sections.
The following killing fluids were used: Gilson's fluid, aceto-
sublimate, Zenker's fluid and Vom Rath's mixture. The greatest
difficulty is found in preventing extreme contraction of the speci-
mens. Stupefaction with chloretone resulted in contraction as
intense as that produced by the killing fluid : the use of hot
reagents was of some small advantage. Flattening between plates
of glass gave the least contraction, but had the disadvantage of
producing some distortion. The best results were obtained from
specimens killed in Zenker's fluid; the poorest from those killed
in Gilson's fluid. It is difficult to obtain good infiltration with
paraffine, especially in the normal (not flattened) specimens.
The modification of Apathy's method given by Lee was very
successful, leaving the tissues in good condition. High tempera-
tures (over 50° C.) are rapidly fatal to the integrity of the tis-
sues. The shortest possible time in paraffine of the lowest melting
point that can be used gives the most satisfactory result.

The worm is very difficult to section because of the great
mass of eggs it contains. Various macerating agents were tried,
but none was of any great service. Sections were cut as thin as
4/x; most of the series ranged from 8/* to 16/*, the latter in col-
lodion.

The stains used were : Heidenhain 's iron haematoxylin,
Benda's iron haematoxylin, Mayer's acid haemalum, Mallory's
connective tissue stain, Delafield's haematoxylin, Ehrlich's
haematoxylin, Lonnberg's borax carmine-Lyons blue, toluidin
blue. Iron haematoxylin preparations are excellent for the study
of cuticula, musculature and sex-cells; they are worthless for
the study of the nervous system and unreliable in many details.
No statements are based on results given by this stain alone.

Mallory's connective tissue stain colored nervous tissue, with
the exception of the nuclei, bright blue. Its action was too

384 I' n i versify of California Publications in Zoology. [VOL. 6

uneven to be of special value. Mayer's acid haemalum, Dela-
field's haematoxylin and Ehrlich's haematoxylin gave satisfac-
tory results for general morphology. Lonnberg's borax carmine-
Lyon's blue method gave the best result for the study of the
nervous system. This in general is the method used by Lonn-
berg in his work on Gyrocotyle. A brief description of the
process is given by him (Lonnberg, 1891). The material is
stained in toto in Grenadier's alcoholic borax carmine, 60%
alcohol, decolorized in acid alcohol (5 drops cone. HC1 to lOO
cc.) ; after sectioning and mounting the sections are run very
quickly through absolute and 90% alcohol to a saturated solution
of Lyon's blue in 60% alcohol, plus three drops n/10 HC1 to
35 cc. of stain, where they remain fifteen minutes. The sections
are decolorized in ammoniacal 85% alcohol made up of one part
85% just basic to litmus to 4 parts neutral 85%. Decolorization
requires from two to three minutes and must be stopped when
the sections appear violet in color. On washing in neutral 90%
the differentiation takes place. The sections must not be left
in 90% or absolute longer than one or two seconds, else -the stain
is blurred. This method gives beautiful preparations in so far
as differentiation of tissues is concerned, but is rather poor for
cytological differentiation, especially in cell-boundaries.

The nervous system stains uniformly a light blue, easily dis-
tinguished from all other tissues except condensed connective
tissue. The blue is clearer in the nerves than in the connective
tissue; the nervous tissue can easily be recognized with a fair
degree of certainty by this fact, coupled with its distinctive
histological structure.

The statement generally made that Lyon's blue should be
used in very dilute solution for corrosive sublimate material is
not applicable in this case. Weak solutions give no result, for
all the color drops out at once, leaving no differentiation.

II. BODY-COVERING AND SPINES.

The body is covered by a thin membrane, composed of finely
felted fibres imbedded in a homogeneous matrix (cut., pi. 42, figs.
55, 58). These fibres are exceedingly delicate, tangled in an
irregular fashion, but in general parallel to the surface of the

1911] Watson: The Genus Gyrocotyle. 385

body. The outer surface of the membrane is bounded by a fine,
deeply stained layer of ragged fibres (ext. cut. I.). This is seen
only in perfectly preserved specimens. Vesicles occur frequently
in the membrane. There are also breaks in the margin,_resem-
bling sections of the "pore-canals" figured by Blochmann (1896),
The thickness of this membrane varies in different regions, being
slightly less on the lateral folds and on the surface of the aceta-
bulum. The average thickness on the surface of the body proper
is about 3.8/x ; on the lateral folds, etc., about 2/x,. This membrane
will be referred to as a " cuticula, ' ' following the common usage
among writers on the subject; but the word is not used in its
strict sense, as denoting a body-covering derived from an under-
lying epidermis. The cuticula of Gyrocotyle may or may not be
derived from the * * subcuticular ' ' layer referred to below ; struc-
tural relations indicate that it is, but in the opinion of the writer
the question can only be settled by embryological data which are
not available.

Lonnberg (1891) describes a two-layered cuticula, agreeing
in general with the appearance of that of G. fimbriata; but he
describes the fibres seen in the " Hauptschichte " or principal
layer, as running perpendicular to the surface. He finds this
only in sections in which the cuticula is torn or affected by
reagents; from appearances presented in similar sections, the
writer thinks he was probably dealing, not with the cuticular
layer, but with a layer of perpendicular fibres lying beneath the
cuticular muscles, to be described later.

Immediately beneath the cuticula (cut. trans., pi. 42, fig.
55) is a layer of cuticular muscle-fibres running round the body
at right angles to its long axis (cut. trans., pi. 42, fig. 58). Close
beneath this layer lies a set of longitudinally arranged muscle-
fibres (cut. long.) ; these are heavier than the transverse fibres.
These layers of muscles are distinguished as the cuticular muscu-
lature. Some distance below these muscle fibres is an irregular
layer of large nuclei, about twice the size of the ordinary paren-
chyma nucleus, resting on and often imbedded in the outer
transverse layer of body musculature (sub. cut.). This layer
may be one or two nuclei deep, or it may be from five to eight
deep. In the neighborhood of a spine the layer becomes three

386 University of California Publications in Zoology. tv°L- 6

to five times its ordinary depth. These nuclei, the "Matrix-
zellen" of Lonnberg, are richly supplied with chromatin.
Definite cell-boundaries can sometimes be seen; the cytoplasm
stains deeply and shows a granular structure. The writer Avas
unable to establish any processes running from these cells out
to the cuticula through the cuticular muscle-fibres, or to make
out any structural connection existing between cuticula and
subcuticula. The space between the subcuticular layer and the
longitudinal cuticular muscles is filled with very delicate muscle-
fibres (sag. fib.) and thread-like processes of the subcuticular cells,
taking in general a direction perpendicular to the surface. There
are also many fibres running parallel to the surface, as shown in
transverse section by their cut ends, but these are not so conspicu-
ous as the perpendicular threads. It was probably this layer
which Lonnberg mistook for the principal layer of the cuticula
in his sections. The writer has found similar appearances, due
to the fact that the cuticula is very easily affected by reagents.
In a specimen preserved in early stages of decomposition or
imperfectly fixed by the killing fluid, the cuticula is almost
invariably found to be sloughing off in thin strips, broken,
notched and ragged. It presents both horizontal and perpendi-
cular planes of cleavage, but shows a marked tendency to split
in layers parallel to the body-surface. This is the only indica-
tion of lamination noticed; it is more justly interpreted as a
splitting in the direction taken by the greater portion of the
fibres of which it is composed.

The cuticula in the region of the rosette shows marked
changes. That lining the funnel is noticeably thinner, the
cuticular muscular layers are much reduced, and the subcuticular
cells are very few. As the much-folded margin of the funnel is
approached, the subcuticular cells of the outer wall also become
much reduced, sink deeper and cease to be a distinct layer; lying
within them large gland cells can be seen. In the writer's
opinion these are not related to the subcuticular cells but belong
to the central part of the body just as do the shell-glands and
the prostate glands, arising in the neighborhood of the inner
longitudinal muscular layer.

The most prominent feature of the body-covering is the

1911] Watson: The Genus Gyrocotyle. 387

spines, frequently referred to in the preceeding discussion of
the literature of the subject. In G. fimbriata, the spines are
arranged in a definite pattern, from which there is no great
amount of individual variation, though the number of spines
varies considerably. At the anterior extremity, on the lateral fin
on each side of the acetabulum are two groups of large spines
(pi. 36, figs. 22, 23). They are arranged in five rows, two each
on the dorsal and ventral surfaces and one row on the margin.
There are from five to seven spines in a row, from 20 to 30 in
each group. Here, as elsewhere, the spines are borne on rounded
papillae, which are in this region of greater size than elsewhere
on the body, except around the "neck" of the rosette. The
papillae extend back along the lateral fin to the beginning of
the lateral folds, but bear spines only at the region shown in
the figure, at the level of the posterior half of the acetabulum.
In addition to these large spines, of the shape shown in fig. 28,
there are a number of very much smaller, less definitely shaped
spines (pi. 37, fig. 32), in among the larger ones.

There are no spines on either surface, throughout the length
of the body, back to the posterior border of the receptaculum
seminis. Here on the ventral surface two groups of spines appear,
near the base of the lateral folds, extending backward in con-
verging lines toward the canal opening. They flank this opening,
increasing in number and size, and spread out posteriorly on the
surface of the neck of the rosette in its median half. At the
posterior border of the neck they extend laterally, passing around
to the dorsal surface. On the extreme lateral margins of the
ventral surface appear two groups of spines, which pass over on
the dorsal surface, forming with the ring above described a belt
of spines on the dorsal surface of the neck. The number and
direction of these spines is shown in the figures (pi. 34, figs.
10, 12).

In G. urna (var. ?) the distribution of spines (pi. 36, fig. 24;
pi. 34, figs. 11, 13) is markedly different. The ventral surface,
to the level of the uterus opening, is thickly set with large
spines ; there is no special pattern in the arrangement about the
neck, there being six or seven irregular rows of spines, encircling
the base of the rosette. On the dorsal surface the spines are

388 University of California Publications in Zoology. [VOL- 6

fewer and scattered irregularly over the whole surface. It
should be noted that the two marginal clumps of spines at the
level of the acetabulum occur as they do in G. fimbriata.

The distribution of spines was worked out in living as well
as preserved material, that no error due to loss of spines by
deterioration or action of reagents might enter. The distribu-
tion found resembles that given by Lonnberg for G. urna, except
that he describes spines as present along the whole length of the
lateral margins of the body, and along the lateral folds. The
distribution as given by Spencer (1889, p. 140) for G. rugosa is
substantially that found in G. urna (var.?) :

"They are distributed generally over the body surface, but are most
numerous along the side folds, and more especially at the anterior endr
both on the dorsal and ventral surfaces, and again at the posterior end,
beyond the region of the side folds. They are sparsely distributed over
the central part of the body, both dorsally and ventrally. ' '

Haswell's figures (see my plate 47, figure 80) show spines of
great size distributed over the whole of the surface of the animal.
Wagener's (1852) figures show the acetabular clumps, and a
distribution of the spines in the posterior region similar to that
described by Lonnberg and given above for G. fimbriata.

The small spines, irregularly distributed, vary in size from 40/*,
to 60/n by 15/x to 25/x, and in shape from simple rounded spinules
to a blunt-tipped, swollen-based form. Broad, triangular,
sharply pointed forms also occur, but less frequently (pi. 37r
fig. 32). These spinules are not borne on papillae; they rarely
contain more than three concentric layers.

The spines proper, from the extremities of the body, are of
fairly constant shape and size. In the acetabular region (pi. 37,
figs. 28, 29) they are bluntly rounded at the tip, swollen in the
middle third of their length to nearly three times the diameter
of the tip, narrowing again at the bluntly rounded base. They
are about 130/x by 36/x in size. In the rosette region, the spines
are of the same general shape (pi. 37, figs. 30, 35), but are of
considerably greater size, ranging from 185/t by 45/x to 220/w.
by 55/ut.

The spines are composed of concentric layers of uniform
thickness, and are hollow. In macerated specimens the layers.

1911] Watson: The Genus Gyrocotyle. 389

are seen to be composed of a finely felted mass of fibres, circular
in general direction, embedded in a homogeneous interstitial sub-
stance which is acted on by the macerating agent (which also
removes the cuticula from the body) (pi. 37, fig. 33). Each spine
is set in a pit lined with a very thin cuticula of felted fibres (par.
felt, pi. 37, fig. 34) which passes indistinguishibly into the fibres
of the surrounding parenchyma. The spine is surrounded by
muscle-bundles, which serve to protrude and retract the spine.
The retractors are muscle-bundles from the outer transverse
layer of the body musculature; they insert at the base of the
spine (pi. 37, fig. 30). The protractors are bundles of fibres
inserted at the base and for some distance on the sides of the
spine ; they originate or are attached to the cuticular musculature,
principally to its transverse layer.

The direction of the spines is shown in the figures (pi. 34,
figs. 10-13). In general they may be said to be directed anter-
iorly and laterally. This is the most advantageous arrangement
possible for maintaining the attachment of the parasite to its
host ; the position of the greatest number of spines on the ventral
surface and around the rosette also further this end. For loco-
motion, the spines seem to be arranged to hinder rather than
help; but they are probably so completely retracted that their
rounded ends offer no special resistance to the forward move-
ments of the worm. At all events, however theoretically difficult
their arrangement and direction render locomotion, the worm
does move, and with considerable rapidity, in spite of them.
Naturally, ease of locomotion is a secondary and firmness of
attachment a primary consideration in the economy of this

creature.

III. MUSCULATURE.

The musculature of Gyrocotyle is exceedingly well developed,
and composed of powerful, neatly balanced sets of muscles. It
constitutes the great bulk of the body. This is not surprising,
when the extreme contractility and general activity of the worm
is considered.

In general, muscles are grouped in pairs running at right
angles to each other. There is throughout the body, but espe-
cially prominent in the region between the acetabulum and the

390 University of California Publications in Zoology. [y°L- 6

canal opening, a set of sagittal fibres, very delicate, not organized
into bundles, running from the outer cuticular muscle layer of
the dorsal surface to that of the ventral surface. The fibres
insert directly in the cuticula, as do those of the protractors of
the spines.

In addition to these sagittal fibres, six distinct coats of muscle
fibres can be distinguished, in the body proper. These may be
grouped according to their action, which gives us three sets of
opposing muscles, each set a longitudinal opposed to a transverse ;
or they may be grouped according to- their position with respect
to the other tissues of the body. Such a division gives us two
groups, one of which may be again subdivided into two.

1. Peripheral, including all muscles lying without the vitel-
laria. May be divided into :

(a) Cuticular, including all muscles lying without the sub-

cuticula or "Matrix-zellen."
(6) Intermediate, including all muscles lying between the

subcuticula and the vitellaria.

2. Central, muscles lying within the vitellaria.

The cuticular muscles have been described in connection with
the cuticula (p. 385). Immediately beneath the subcuticular cells,
frequently invading their territory, is a layer of muscle-bundles
directed around the body, at an angle of about 60° to its main
axis. Thus in a cross-section we get only broken segments of this
layer, not a continuous sheet, as in the deeper-lying transverse
muscle layer. These fibres run in two interlacing sets whose
directions are at an angle of 60° to each other. These bundles all
break through the subcuticula to insert by fine threads on the
cuticula (pi. 42, fig. 55).

Immediately beneath the outer transverse layer above
described are found the bundles of fibres running parallel to the
longitudinal axis of the body. These are heavier and grouped
into more definite bundles than any previously described. The
layer is one or two bundles thick and occurs over the whole body,
but becomes very thin in the region of the lateral folds. These
fibres insert on the cuticular musculature, and with the outer
transverse just described produce the transverse ridges so char-
acteristic of the worm in its contracted state. Within this laver

1911] Watson: The Genus Gyrocotyle. 391

lie the inner transverse muscles, a very heavy sheet of fibres
running at right angles to the longitudinal axis, around the body.
These fibres also are continued for some distance into the region
of the lateral folds, but are considerably reduced. These muscle
layers constitute the peripheral musculature and arF found
throughout the whole body, though less and less strongly devel-
oped the further laterad they pass.

Within the inner transverse muscles lie the vitellaria, the
large excretory canals, the central nervous system, the glands
of the reproductive system, and the strongest muscles of the
body, the inner longitudinal. In the region of the body anterior
to the testes and posterior to the ovaries, these muscle-bundles
occupy the whole of the space inside the vitellaria. In the inter-
mediate region, especially in the region of the uterus, the layer is
considerably reduced, and pushed to one side. It is quite thick
ventrally, but there are very few fibres dorsal to the uterus.
These longitudinal fibres insert on the acetabulum, and around
the neck of the rosette and on the walls of the canal. It is by
the contraction of these powerful muscles that the worm moves
about, thrusts forth and around and withdraws the proboscis-
like acetabulum, and firmly attaches the posterior rosette.
Furthermore, as Lonnberg pointed out, this muscle layer being
absent in the lateral frills, and the outer longitudinal fibres being
very weakly developed there, on contraction of the longitudinal
muscles the lateral folds are increased in amplitude. Lonnberg
regards the frills as due wholly to contraction of the body mus-
culature ; the writer, in agreement with Wagener, regards them as
independent structures, existing in all states of expansion of the
animal.

The musculature of the acetabular region (pi. 40, fig. 43)
shows plainly the origin of the acetabulum as an invagination of
the anterior extremity of the body. There is a doubling of the
layers, giving twelve instead of six. By inverting the anterior
extremity we get a mass of predominantly longitudinal muscles,
corresponding to the inner longitudinal body-layer, on the out-
side of the acetabulum. These fibres are meridional with refer-
ence to the acetabulum as a whole and constitute more than half
its bulk. Inserting along the surface of this mass are the inner

392 University of California Publications in Zoology. [VOL- 6

longitudinal muscles of the body, most numerous toward the
anterior extremity and in large masses around the margin of
invagination ; on contraction of the inner longitudinal muscles,
the whole mass of the acetabulum is drawn back (pi. 40, fig. 43).
This is the state invariably found in killed specimens. In the
living animal in the expanded state the mouth of the acetabulum
is at the anterior extremity (pi. 33, figs. 7, 9) ; further evagina-
tion is impossible. The great lengthening of this region observed
in active specimens must be due to expansion of the acetabulum
itself.

The fibres of the outermost layer (out. mer.), constituting
the bulk of the acetabulum, are derived from the inner longi-
tudinal layer of the body and are meridional in direction, pre-
senting in section a margin of cut edges surrounding the aceta-
bulum. This layer is densely supplied with finer radial fibres
(rad. fib.), at right angles to the long axis of the acetabulum,
passing from dorsal to ventral surface. These are the homo-
logues of the sagittal fibres of the body musculature, greatly
increased in size.

Immediately within this outer heavy layer lies a set of fibres
predominantly transverse (out. arc.), passing from left to right
of the acetabulum, the homologue of the inner transverse layer
of the body. This layer is even more abundantly supplied with
radial fibres than is the outermost one. These pass from dorsal
to ventral, at right angles to the radial fibres of the outer layer.
Within this layer is a second set of meridional fibres (in. mer.) •
the homologues of the outer longitudinal muscles of the body,
followed by a thin layer of fibres passing circularly around the
opening of the acetabulum, corresponding to the outer transverse
muscles of the body. Lying next to the cuticula are thin longi-
tudinal and transverse layers, the homologues of the cuticula r
musculature elsewhere. At the mouth of the acetabulum a
special sphincter muscle has been developed, composed of fibres
from the inner transverse, the inner longitudinal and the circular
muscle layers of the acetabulum (acet. sphinc.). Most of the
fibres are circular in direction ; they are grouped in heavy bundles
and form a large ring-muscle about the acetabular opening.

The musculature of the funnel-region has been worked out

Watson: The Genus Gyrocotyle. 393

by Lonnberg (1891). There is here a doubling of layers, due to
folding on the ventral surface, and an increase in the inner trans-
verse layer to produce two "wing-muscles." Lonnberg 's tabula-
tion of the funnel muscles and their homologues in Gyrocotyle
urna the writer finds substantially correct for G. fimbridJarHere
as in the acetabulum the development has been mainly in the
inner transverse layer, and in the sagittal fibres. By means of
these a very effective sphincter is produced at the base of the
rosette. Posterior to the collar or "neck" (the region occupied
by this sphincter) and anterior to it near the canal opening, the
development of the transverse muscle layer is much reduced,
though still greater than in the body in general.

Lonnberg concludes from the above described arrangement of
muscle layers that the rosette and canal were formed by a fold-
ing from the posterior extremity forward in the ventral surface.
This was first a furrow or trough, which functioned as a sucker, a
condition common among the lower cestodes; later the ventral
walls fused to form a tube, and finally only the most posterior
part of the tube functioned as an organ of attachment, its
anterior extremity remaining open as the present ventral open-
ing of the canal. Lonnberg hazards no opinion as to the signifi-
cance of this anterior canal opening. Observations on the
attached worm show that the mouth of the opening is always
closed wrhile the rosette is attached to the mucosa. When the
canal mouth opens, flaring out as in Spencer's figure 2, the hold
of the rosette loosens and the worm drops from the mucosa.
During attachment there is a series of waves of contraction, run-
ning from the margin of the rosette forward to the canal open-
ing. The whole posterior extremity thus forms a suction-cup of
high efficiency. The musculature of the ventral canal-opening,
developed from the peripheral muscles, is shown in plate 40,
figure 44.

The finer structure of the muscles in Gyrocotyle is very
simple. The sagittal fibres (radial fibres of acetabulum and
rosette collar) are not aggregated into bundles but are simple
strands running through the body, distinguished from parenchy-
ma! fibres by their size and also by the position of the nucleus
(pi. 42, fig. 57). This lies to one side of the fibre, forming a

394 University of California Publications in Zoology. [VOL. 6

bulging prominence. These nuclei are larger and more easily
seen than those of the other muscle fibres; this is in harmony
with the generally primitive and undifferentiated character of
this set of muscle fibres. In the outer transverse muscle layer
the fibres are longer, very slightly swollen in the middle of their
length, where the nucleus lies very closely applied to one side
of the fibre. The nucleus also is elongated, rather than round as
in the dorso- ventral fibre (pi. 42, fig. 57). In the two inner
layers, the inner circular and the inner longitudinal, the nucleus
is very hard to distinguish (pi. 42, fig. 56). The fibres are heavy,
elongated, tapering gradually toward their extremities. The
nucleus is exceedingly slender, flattened against the surface at
one side of the fibre, but still within the fibre. In cross-sections
of fibres the nucleus is difficult to distinguish, appearing merely
as a thickening of the cell- wall; in longitudinal sections it can
be made out more easily. Had not the nucleus appeared so un-
mistakably in the outer and sagittal muscles, it probably would
have been overlooked in these inner fibres.

In the cuticular muscle fibres there is no trace of a nucleus.
The fibres are slender, long, of even diameter. Those of the
transverse layer are exceedingly fine, while the longitudinal ones
are of ordinary diameter. The great number of processes run-
ning from the subcuticular cells, apparently to insert in the cuti-
cular musculature, suggests most strongly that some of these
cells, at least, are the myoblasts of the cuticular musculature.
The fact that elsewhere in the body the nucleus lies in the fibre
makes this seem improbable, yet the writer is at a loss to explain
the processes and their connection with the musculature in any
other way.

All the fibre bundles are penetrated throughout and en-
wrapped by parenchymal threads, with their accompanying
nuclei. The latter can always be distinguished from muscle-
nuclei by their shape and size, as well as by position.

It is of interest to note that Salensky (1874) found in Amphi-
lina foliacea smooth muscle fibres with laterally attached myo-
blast, this being one of the earliest cases in which a nucleus for
a muscle fibre was discovered. Furthermore, the description
above given of laterally attached nucleus in the dorso- ventral,

1911] Watson: The Genus Gyrocotyle. 395

transverse and longitudinal muscles and of nucleus some distance
removed from the fibre in the subcuticular muscles, corresponds
to the classification given by Braun (1901, p. 1351). The only
difference is that the laterally attached nucleus in Gyrocotyle is
inside the fibre, not in an adjacent myoblast. The occurrence of
nuclei within muscle fibres is reported, according to Braun (1894,
p. 1351) in the scolex of the Tetrarhyncha, and in the longitudinal
muscles of the proglottides of Taenia dendritica. That a form
as primitive as Gyrocotyle should exhibit a muscle fibre relatively
simple and undifferentiated as compared with that of the mero-
zoic cestodes, is exactly what other facts in its structure would
lead one to expect.

The staining reactions of the muscle-fibres are of some
interest. With iron hematoxylin and toluidin blue the whole
fibre stains very intensely. Preparations by these methods are
excellent morphologically, mapping out the muscles most sharply ;
but are quite worthless histologically. With Mallory 's connective
tissue stain, muscle fibres stain a bright red. As nuclei also stain
red, this method is not useful for a study of muscle-nuclei. The
same holds true for borax carmine-Lyon 's blue preparations.
The best results are given by Delafield's hematoxylin, hematoxy-
lin-eosin, or Mayer's acid haemalum, and counter-stains. With
these the muscle-fibre stains a dull blue-gray, while the nucleus
comes out sharply in blue-black.

IV. REPRODUCTIVE ORGANS.

The organs of Gyrocotyle include the following :

1. Female. Ovaries, receptaculum ovorum, vitellaria, shell-
glands, uterus, vagina, and receptaculum seminis.

2. Male. Testes, vasa efferentia, vesicula seminalis, vas
deferens, penis, and prostate glands.

All of these organs, with the exception of the vitellaria, lie
within the inner longitudinal muscles. The female organs are in
the second and third quarters of the length of the body ; the male
organs in the first quarter.

The ovaries (ovar., pi. 39, fig. 42) lie laterad of the uterus,
at its posterior border. They are roughly triangular in shape,
the lateral lobes united in the median line just posterior to the

396 University of California Publications in Zoology. [VOL. 6

receptaculum seminis by a small median portion. They are com-
posed of numerous rounded follicles, from which run collecting
tubules to empty into five main oviducts, one posterior, two
lateral and two anterior ducts. These lead into the receptaculum
ovorum, which is embedded in a mass of loose tissue in a depres-
sion on the anterior dorsal surface of the receptaculum seminis.
The receptaculum ovorum (rec. ov., pi. 45, figs. 73, 75), is
about 3 mm. in its antero-posterior and its dorso-ventral diam-
eter and about 6 mm. in transdiameter. These measurements are
the average for several sexually mature worms.

The vitellaria are composed of loosely-grouped follicles, each
containing ten or fifteen cells, lying just within the inner trans-
verse musculature and outside the inner longitudinal. The fol-
licles are present throughout the body with the following excep-
tions : (1) dorsal and ventral to the uterus, that is, in the median
third of the body in the two middle quarters or more of its
length; (2) anterior to the posterior border of the acetabulum;
(3) posterior to the level of the ventral canal opening.

The lateral folds are densely supplied with vitellarian fol-
licles, these forming the greater part of the folds. The ducts
of these yolk glands unite into four main lateral ducts, which
empty into a yolk reservoir, or "Endstiick, " which lies in the
dorsal depression of the receptaculum seminis above mentioned,
just posterior to the receptaculum ovorum. It gives off a single
efferent vitellary duct, which enters the afferent oviduct, as
described below, a short distance back of the entrance of the duct
connecting the efferent oviduct with the receptaculum seminis.

The efferent yolk-ducts (pi. 39, fig. 42, vit. d.), appear in
stained and cleared preparations as a dark brown anastomosing
network of delicate threads, spread over the receptaculum
seminis and the first three or four coils of the uterus, and con-
verging to the yolk-reservoir in the concavity of the recep-
taculum.

From the ventral surface of the receptaculum ovorum is
given off in the median line a single efferent duct (ef. ovd.)
which receives first a short thick-walled duct from the recep-
taculum seminis (duct, sem.) ; and then an efferent duct from
the yolk reservoir (ef. vit.d., pi. 45, figs. 71-74). This duct

1911] Watson: The Genus Gyrocotyle. 397

(oot.) then becomes the uterus, passing to the left of the recep-
taculum seminis and around it to the ventral surface, across the
ventral surface and forward, at once increasing in size. The first
two or three coils of the uterus, as above described,_may be
regarded as an ootype. They are surrounded by shell-glands, and
within them the compound eggs are formed. In the dorsal
(proximal) part of the first coil, and to the left of the receptac-
ulum seminis, the yolk-cells are seen to be aggregating about
a single ovum and the uterus is full of droplets of a yellow homo-
geneous material, which form a coating around the combined
ovum and yolk-cells. The uterus contains fully formed eggs in
the convolutions anterior to the receptaculum seminis. It winds
back and forth across the median third of the body through
from fifteen to twenty convolutions, the duct increasing steadily
in size, ending finally in a greatly dilated pouch which opens
to the exterior by a large aperture on the dorsal surface in the
median line posterior to the opening of penis and vagina.

The uterus is lined at its inception and throughout its course
by a thin fibrillated layer, increasing in thickness from the pos-
terior coils forward. This layer is covered with fine cilia, in
the first five or six coils of the uterus. These cilia are connected
with scattered nuclei beneath the cuticula, by means of delicate
fibrils strongly suggesting those figured by Lonnberg (1891)
for the cilia lining the excretory canals. Their direction is in
general at right angles to the course of the uterus. These fibres
lie in a nucleated meshwork of parenchymal fibres, which pass
indistinguishably into the cuticular lining of the duct. Further
forward, the nucleated circular muscle fibres disappear, and a
typical subcuticular layer of cells appears coincidently with the
appearance of a typical cuticular musculature, suggesting very
strongly a structural connection between subcuticular and cuti-
cular musculature.

The shell-glands (sh. gl., pi. 39, fig. 42) are an aggregation of
large cells, with characteristic nuclei, lying close to the recepta-
culum seminis on its dorsal, ventral, and posterior surfaces.
This relation to the receptaculum seminis is of course due to the
fact that the earliest convolutions of the uterus lie on these sur-
faces of the receptaculum. The greatest mass of cells lies to the

398 University of California Publications in Zoology. [VOL. 6

right and left; the posterior median mass is very thin. The
lateral masses extend dorsally and ventrally well into the main
longitudinal muscle mass, and in general the dorsal mass is
greater than the ventral. The efferent oviducts' and yolk-ducts
pass through this mass of gland cells, but no shell-material is
seen except in the first coils of the uterus. The cells of the gland
are large, swollen at the base, with slender "necks," leading into
intracellular ducts (pi. 44, fig. 68). These ducts appear to open
independently in the anterior wall. The cytoplasm of the gland
cells is dense, granular, and stains very intensely. The cell wall
is sharply defined. The nucleus is large, cloudy, with deeply
staining border and faint nucleolus. The nucleus lies in the
dilated base of the cell. The ducts have a sharply defined lumen,
and are of considerable size.

The receptaculum seminis, originally described by Wagener
(1852) as the testis, is a large chamber, convex posteriorly,
slightly concave dorso-anteriorly, filled with a dense mass of
semen (rec. sem., pi. 39, fig. 42). It might be regarded as the
dilated blind end of the vagina, which enters its dorsal surface
near the anterior border. The receptaculum seminis measures
about 1.5 mm. by 1 mm. by 1 mm. It is surrounded, as above
noted, by the first coils of the uterus and their accompanying
shell-glands; on its dorsal surface lie the yolk reservoirs, the
receptaculum ovorum, and their efferent and afferent ducts.
From the dorsal surface of the receptaculum is given off a short,
thick-walled duct, the ductus seminalis, to the efferent oviduct.

The wall of the receptaculum seminis (pi. 42, fig. 59) is com-
posed of a mass of finely felted fibres, elongated cells with large
nuclei and granular, non-fibrillated cytoplasm. Outside this
layer are found scattered cells of the shell-glands. This differs
markedly from Lonnberg's description (1891, p. 41) of the his-
tology of the wall of the receptaculum. "Das Receptaculum ist
von einen diinnen Pflasterepithel ausgekleidet, aber es besitzt eine
dicke fibrose Wand, die reich an eingeschalteten Kernen ist."
The writer cannot distinguish cell walls in the lining of the
receptaculum, but finds scattered nuclei, embedded in a mass of
delicate interlacing fibres.

The receptaculum seminis' contains, embedded in the mass of

1911 J Watson: The Genus Gyrocotyle. 399

spermatozoa, occasional deeply staining cells, large, and with
very large nuclei. These were noted and figured by Spencer
(1889, p. 146) : he suggests that they "may be simply the sperm-
blastophores from which the ripe sperm have separated^ but
there is no proof of this. ' ' The writer is convinced that they are
ova, which have entered the receptaculum seminis through the
duct which leads from this structure to the oviduct, the ductus
seminalis. The finding of these cells at the entrance of this duct
into the receptaculum and their resemblance to the ova, renders
this certain.

The passage to the oviduct, the ductus seminalis, has a very
thick muscular wall lined with cilia and an exceedingly small
lumen, often completely obliterated in sections. For this reason,
Spencer was unable to demonstrate it to his own satisfaction,
though convinced that it must exist.

The vagina passes forward, in the early part of its course
much convoluted and lying close to the receptaculum seminis.
Further forward it lies in the parenchyma close to the dorsal
wall of the uterus ; it turns ventrally near the middle of its course
and passes gradually to the ventral surface, where it finally
opens, laterally and posteriorly to the penis, near the right
margin of the body. The lumen of the vagina is lined with a
layer similar to that described for the receptaculum seminis, not
ciliated. It is not epithelial, as described by Lonnberg. Its wall
is composed of a thin layer of longitudinal muscle fibres ; it is
embedded in parenchyma and lies within the main mass of longi-
tudinal muscles, close to the wall of the uterus. As the vagina
nears the anterior margin of the uterus its lumen increases in
diameter. During its course past the convolutions of the vas
deferens this increase continues, and its wall is also increased in
thickness. The lining of the tube becomes much convoluted
in its course toward the left margin, and increases suddenly in
diameter in the latter half of its transverse course, opening
finally by a large aperture with much-folded margins and heavy
muscular wall. This region is plainly adapted to copulation.

The testes are arranged in two groups not connected and not
symmetrical (pi. 39, fig. 42). The left testis extends from the
posterior border of the anterior third of the body, forward to

400" University of California Publications in Zoology. tv°L- 6

the opening of the penis. The right testis is composed of two
lobes, the right and larger of which extends from the posterior
level of the left testis forward to the anterior extremity of the
body. The left lobe, attached by a narrow bridge posterior to
the base of the acetabulum, lies to the left of the acetabulum
and in front of the vaginal opening. The testes lie within the
longitudinal muscle mass, in a position identical with that occu-
pied by the ovaries in the posterior portion of the body. They
are considerably larger than the ovaries, as is shown in plate 39,
figure 42, and are composed of loosely-aggregated follicles. Each
follicle is covered with a fibrous nucleated layer, the "tunica
propria" of Lb'nnberg, continuous with the walls of the tubules
into which the follicle opens. These unite into two main vasa
efferentia and empty into a large median vesicula seminalis
situated immediately in front of the uterus, nearest the ventral
surface. This structure is perhaps more accurately described
as a vas deferens, more or less uniformly dilated by the masses
of spermatozoa. It coils from left to right and from dorsal to
ventral, in three large convolutions (pi. 41, fig. 46), dilating
near its anterior extremity to form a bulb, and opening through
a muscular papilla into a thick-walled tube, the cirrus-pouch or
ejaculatory duct. The wall of this pouch is thick and well sup-
plied with muscles. Its inner lining is greatly folded and covered
with delicate spinules. This duct passes almost straight dorsad,
turning a little to the left, opening on a rounded papilla on the
dorsal surface, a little to the left of the median line and anterior
to the mouth of the vagina on the ventral surface. The wall of
the duct grows much thinner toward its distal end and its lumen
increases somewhat. Heavy bundles of muscle fibres running at
right angles to the course of the ejaculatory duct are attached
to its wall throughout the whole of its course (rad. muse. /., pi.
43, fig. 65). Outside these lies a large mass of gland-cells extend-
ing the whole length of the ejaculatory duct. They are very large
granular cells, with definite cell-walls forming intracellular
canals (prost. gL, pi. 43, fig. 64). These empty into numerous
delicate ducts opening on the surface of the ejaculatory duct
and probably constitute a prostate gland. Outside this gland lie
heavy longitudinal muscle bundles, a part of the body muse u In-

1911] Watson: The Genus Gyrocotyle. 401

ture. The cells of the gland are embedded in the innermost of
these muscle-bundles. The lumen of the ejaculatory duct is
lined with a thin cuticula, beneath which are found cuticular
muscle fibres and subcuticular cells. Outside this layer is a
thick mass of circular muscle fibres, and outside these a heavy
longitudinal layer of fibres. To this coat are attached the
radially arranged muscle fibres above described", inserting in
general at the angle shown in figure 65, plate 43.

Spermato genesis. — The follicles of the testes are lined by a
syncytium in which are embedded rather small rounded nuclei
of the spermatogonia. The middle of the follicle is filled with
dividing cells and developing spermatozoa. No attempt has been
?nade to work out the details of the process of spermatogenesis.
It may be mentioned that in none of the preparations observed
by the writer were there any indications of amitosis, but several
mitotic figures were observed. The mature spermatozoon is a
slender thread tapering at the posterior end, with a well-marked
head, several times the diameter of the body and staining in-
tensely (pi. 41, fig. 50).

The Ovum and Oogenesis. — Unfortunately the writer has seen
no young specimens of Gyrocotyle. While there has been great
variation in size, in all individuals the uterus has been full of
developing embryos which mark the specimens as sexually
mature. Lonnberg, the only investigator who has had the good
fortune to work with the immature form, described the ovaries
in the young individuals as follows :

"Das Ovarium hat bei jiingeren Individuen eine nicht gewohnliche
Gestalt, indem es viel mehr traubig gelappt als bei anderen Cestoden 1st.
Es besteht also in diesem Stadien aus kleinen rundlichen Follikeln, die
durch weite Ausfuhrungsgange zu Trauben vereinigt werden, und diese
Trauben erster Ordnung werden durch ihre Ausfiihrungsgange zu Trauben
zweiter Ordnung verbunden. Alle Trauben vereinigen sich zu je einer
Sammlung aufschmelzen aber allmalig die Trauben zu unregelmassigen
Lappen zusammen und es seheint daher nicht treffend, wenn Spencer auf
seiner schematischen Figur ein so distinkt traubenfb'rmiges Ovarium bei
einem Tiere mit von Eiern prall gefiillten Uterus zeichnet. "

Spencer (1889, p. 144) dealt with sexually mature forms, but
found two stages of reproductive activity :

"In the first-mentioned the ova were evidently passing down into the
uterus, in which they were but very slightly developed. The ovaries

402 University of California Publications in Zoology. [VOL- 6

consequently were full of fully formed ova, having the nature of distinct
cells with clearly defined nuclei. In the second the uterus was full of
much more highly developed embryos, and no ova, apparently, were pass-
ing into it. In this case the ovaries were evidently in the act of develop-
ing a fresh supply of ova. Each consisted of a mass of protoplasm con-
taining nuclei, evidently dividing rapidly, whilst the outlines of the cells
could only here and there be seen with anything approaching to clearness.
Each little ovary has thus, when the ova are not fully formed, the struc-
ture of a polynuclear mass of protoplasm, which only subsequently
becomes divided up into a number of distinct cells. ' '1

In all the preparations studied by the writer, the ovaries con-
tained at the same time both fully formed and developing ova;
the uterus contained young embryos in its most posterior coils,
and older ones towards its anterior opening. It seems that in
G. fimbriata the process of egg formation goes on uninterruptedly
the year round, rather than in a rhythmical fashion, as indicated
for G. rugosa by Spencer 's observations. This may be correlated
with the fact that apparently the period of intra-uterine life is
longer in G. rugosa than in any other species, this being the only
form in which the uterus contains hooked embryos.

Each follicle of the ovary is surrounded by a fibrous layer,
called by Lonnberg the "tunica propria," continuous with the
walls of the oviducts into which the ova are discharged. Plate
41, figure 52 represents a typical follicle with ova in early stages
of development. There is a syncytium containing small, round
nuclei with recticular chromatin and either a very small nu-
cleolus or none whatever. The other half of the same follicle is
cut up into cells with definite walls, denser cytoplasm, clear
nuclei nearly double the size of those in the syncytium, with
dense marginal chromatin reticulum and very large excentric
nucleolus. Plate 41, figure 53 represents stages intermediate
between these two, showing the growth of the nuclei and par-
ticularly the appearance of a large extra-nuclear body, staining

i Dr. M. Hungerbiihler's "Studien an Gyrocotyle und Cestoden "
(1910) was received too late for its results to be incorporated in the body
of this paper. He makes the suggestion, based on the facts above quoted
mentioned, that Spencer was dealing with two different species of Gyro-
cotyle, that is, with G. urna and G. rugosa. While this explains numerous
discrepancies in Spencer 's account, and while my own results show the
presence of two species of Gyrocotyle in one species of Chimaera, yet the
forms described by Spencer differ widely in several essentials from G.
urna, and Hungerbiihler 's grounds for concluding that one of Spencer's
forms was G. urna do not seem adequately to account for these differences.

Watson: The Genus Gyrocotyle. 403

by most methods as deeply as does the nucleus but easily dis-
tinguishable from it in strongly decolorized haematoxylin
preparations, and also in borax carmine-Lyon 's blue prepara-
tions. Plate 41, figure 54 shows the origin of this body as an
extruded nucleolus, distinguishable even when within the nuclear
membrane from the chromatin nucleolus proper. Probably only
one such body, or * ' yolk-nucleolus, " is formed during the
development of the ovum. In fully formed ova this yolk-
nucleolus has greatly decreased in staining intensity, appearing
as a mere shadowy ring with a dark center (pi. 41, fig. 51).

The ripe ovum passes from the follicle into the oviduct where
it takes on an irregular elongated form, strongly suggesting
amoeboid movements. This form has also been noted by Lonn-
berg (1891, p. 40). These ova are the largest cells in the body,
measuring about 26^, by 15/A. Their cytoplasm is very dense, full
of shapeless masses of material. The nuclei are about 12/A in
diameter, clear, with heavy deeply staining chromatin reticulum
and large round nucleus, from 5/x to 6/x in diameter. The nu-
cleolus is larger than any other nucleus in the body, except those
of the ganglion cells of the first order and possibly the nuclei of
yolk cells. The large, clear bright nucleus with its dense deep-
staining nucleolus makes the ovum easily recognizable even under
low magnification.

The processes of maturation, which have not been observed,
probably occur after the ova reach the receptaculum ovorum.
Division of oogonia in the follicles of the ovary appears to take
place by true mitotic division. Equatorial plates and anaphases
have been observed, but no attempt has been made to work out
the phases of mitosis. There are no indications of amitosis, either
here or in the follicles of the testes. It seems probable from the
evidence in Gyrocotyle, that Child's (1907) amitotic figures are
the result of confusing the "yolk-nucleolus" with the nucleus
proper. Such figures as are shown in plate 41, figures 52, 53, 54,
could easily be taken for unequal mitotic divisions of the nucleus,
were it not for the differential staining.

The vitellaria are follicular, each follicle being surrounded
by a fibrous tunica propria as in ovary and testes. The early
stages of yolk-cells resemble those of the young ova. A single

404 University of California Publications in Zoology. tv°L- 6

follicle frequently contains a great variety of stages. Near one
margin are found scattered nuclei in a syncytium; the rest of
the follicle is filled with cells with well-defined walls and full of
yolk-spheres (pi. 41, fig. 49). The cells are large, about 20/A,
with round clear nuclei, dense marginal chromatin and a round
central nucleolus. The cytoplasm is reduced to a thin marginal
layer in which the nucleus is embedded. The body of the cell is
packed with yolk-platelets, from %/* to 2/u in diameter, granular
in composition, and staining rather faintly except with Lyon's
blue and toluidin blue, with both of which the platelets stain
very intensely. The whole cell breaks out of the follicle and
enters the system of yolk-ducts, through which it makes its way
to the yolk-reservoir and thence to the uterus. The yolk-plates
have been observed by me only within a nucleated cell. The cell
as a unit becomes one of the components of the compound egg
formed in the uterus.

This statement is not in agreement with Lonnberg's descrip-
tion of conditions in G. urna. He says: "Die Dotterzellen zer-
f all en schliesslich, so dass nur die Kornchen durch die Grange zu
den Eizellen gelangen. ' ' That is, only the yolk-platelets, not the
yolk-cells themselves, enter into the composition of the uterine
egg. Aside from the fact that yolk-ducts, reservoir, and the
beginning of the uterus are all full of typical nucleated yolk-
cells, the most cursory examination of the early coils of the uterus
shows that the eggs, even before the shell is completely formed,
are multicellular. Lonnberg saw this, but interpreted it as evi-
dence of a very early cleavage of the ovum. "Die Eifurchung
tritt sehr friihzeitig ein, so dass man Eier findet, deren Schalen
noch nicht fertig gebildet sind, aber wo das Embryo schon
gebildet ist. ' ' But a careful examination of such an early uterine
egg shows that it contains cells of two kinds, one having the
characteristic size, staining reactions and nuclear structure of
the ovarian ovum (ov.), and the others typical vitellarian cells
(yk. c., pi. 9, fig. 47). Cleavage of the ovum does not begin for
some time after the shell is completely formed. Spencer's (1889,
p. 145) observations concerning the yolk-gland and his figures
of yolk-follicles are very puzzling. His statements are as follows :

"A difference of structure has been noted above in the case of the

Watson: The Genus Gyrocotyle. 405

ovaries of the two examples examined, containing embryos at different
stages of development in the uterus, and a curious difference obtains also
in the yolk-glands of the two forms. In the one containing highly
developed embryos the yolk-glands, like the ovaries, are evidently provid-
ing a fresh supply of material in prospect of the next period of repro-
ductive activity. Each consists of a mass of cells, the outlines~of~ which
are somewhat more clearly marked than in the case of ovaries, with large
nuclei evidently undergoing division. The cells are remarkably similar
to ova, but the relative size of the yolk masses and their definite super-
ficial position renders them distinct from the ovaries. In the case of the
form containing the ova passing down into the uterus, the yolk-glands
are in a much more advanced stage. Each is filled with a mass composed
partly of distinct yellow globular bodies, and partly of nucleated cells.
So far as can be seen there are no definite "shell-glands'7 present; all
the other structures connected with the reproductive organs could be
distinctly made out by means of sections, and presumably shell-glands
would have been able to be recognized if they were present as distinct
and separate structures. In plate 13, figure 2, is represented a portion of
the first part of the uterus, in which evidently the shells are being
formed around the ova. In addition to nucleated cells, the uterus con-
tains very numerous little drop-like yellow structures, which resemble
exactly those which have been previously described as present in the
yolk-glands. It appears as if these, as it were, "ran together," and
formed a case enclosing certain of the nucleated cells, some of which are
ova, and some probably cells from the yolk-glands which will serve as
food for the developing ova. This appears to be the only construction
which can be placed upon the appearances."

Failing to find any shell-glands in G. rugosa, Spencer's
hypothesis (though not definitely stated to be such) seems to be
that both shell-material and yolk-platelets are formed in the yolk-
glands, as indicated in his plate 13. Nothing resembling this
type of follicle, or indicating such a function of the vitellaria,
has been seen in G. fimbriata. It is hard to believe that Spencer
could have observed so delicate and obscure a structure as the
network of yolk- ducts in the region of the receptaculum seminis,
and overlooked the large and unmistakable* masses of the shell-
glands, had they been present in G. rugosa. Lonnberg (1891),
who observed the shell-glands in G. urna, suggests that Spencer
mistook the central mass of the shell-glands for the central part
of the ovary. This suggestion is rendered less plausible by the
fact that the main mass of the shell-glands lies, not in the median
line, but laterad of the receptaculum seminis, and is simply
ma'rked off from the mass of the ovaries. A revision of Spencer 's
material seems to be the only way in which the discrepancy in
observations can be explained.

406 University of California Publications in Zoology. ITOL- 6

V. FORMATION OP THE COMPOUND EGG AND CLEAVAGE
OP THE OVUM.

The formation of the compound egg of Gyrocotyle in the first
coils of the uterus has been described above. The completed
egg presents the appearance shown in plate 41, figure 47, taken
from the fifth coil of the uterus. Cleavage begins very shortly ;
at the same time the walls of the yolk cells become less definite
and their nuclei fainter. There is no indication of cleavage of
the yolk-cells. The mass of cells resulting from cleavage of the
ovum appears to be a syncytium (pi. 41, fig. 48) ; this mass
increases in size with the diminution of the yolk cells. It is
impossible to make out cell walls or to discover any orderly
arrangement of the nuclei, in the preparations available. The
eggs cannot be made sufficiently transparent for study in toto;
and the shells are so resistant that no successful infiltration with
paraffme was obtained. The best preparations, and the ones on
which these statements and figures are based, were celloidin sec-
tions stained with Delafield's hematoxylin. The egg-shells show
no opercula, and no trace of hooks on the embryo could be dis-
cerned. G. rugosa seems to be the only member of the genus
which possesses hooked embryos in uterine eggs.

The eggs when extruded are surrounded by a gelatinous sub-
stance which forms a jelly on contact with sea-water. They are
discharged with considerable force. The discharge has been
observed repeatedly when the intestine of the host was first slit
open, and also when the animal was changed from one solution
to another. Eggs are always found in the intestinal contents
of Chimaera infected with Gyrocotyle. The eggs when first dis-
charged are white and glistening, resembling finely cut sand-
grains. They are about .095 mm. by .065 mm. and ellipsoidal in
shape. The size varies widely in eggs from the same individual,
discharged at the same time, ranging from .075 mm. to .112 mm.,
in longest dimension.

The newly discharged eggs do not possess opercula (pi. 38,
fig. 41). There is a faint differentiation occasionally seen at one
pole of the egg, but nothing which could be definitely identified
as an operculum has ever been found in fresh eggs. In speci-

1911] Watson: The Genus Gyrocotyle. 407

mens which had begun to decay in the intestines, and in prepara-
tions made by Von Rath's method, a perfectly definite and
clearly marked operculum was found (operc., pi. 38, figs. 38, 39,
40). This does not appear to be comparable to the operculum
figured by Haswell (see my plate 47, figure 81)*. Its margins are
finely toothed, and the shell is somewhat thinner in the opercular
cap than elsewhere.

VI. EXCRETORY SYSTEM.

The network of excretory canals, so richly developed in
Gyrocotyle, is one of the first features to strike the eye in obser-
vation of the living animal. Wagener (1851) saw and described
it, and noted particularly its wonderfully elaborate development.
The excretory system consists of the following parts : ( 1 ) " flame-
cells, " (2) capillaries, (3) excretory canals distributed outside
the inner longitudinal muscle layer, (4) excretory canals lying
within and among the fibres of the inner longitudinal muscles.

The "flame-cells" (pi. 43, figs. 62, 63) are large, with swollen
base in which lies an oval nucleus with rich, deeply-staining
chromatin reticulum. The surrounding cytoplasm is granular,
occasionally vacuolated, and stains as heavily as does that of the
gland-cells. Running apparently from the apex of the nucleus is
a fine thread, extending for some distance through the hollowed-
out body of the cell. Each flame-cell thus forms an intracellular
canal, which leads into a capillary. These flame-cells are found
only in the outer layers of the body in what has been described
as the "intermediate" region, within the subcuticular layers and
outside the inner longitudinal muscle mass. Neither Spencer
(1889) nor Wagener (1852) found any such structures; Lonn-
berg (1891) found them, but tells us nothing about them, except
that they lie in the " Rinden-schicht. "

The capillaries are the fine tubules into which the intra-
cellular ducts of the flame-cells lead. They have very thin walls,
and differ from the excretory canals in not having a cuticular
lining or a muscle layer. They are found throughout the body
wherever flame-cells appear.

The excretory canals lying outside the inner longitudinal
muscle mass differ from those lying within this layer chiefly in

408 University of California Publications in Zoology. [VOL. 6

size. They are lined with a cuticula resembling that covering
the body, and are surrounded by a well-developed band of cir-
cular muscle fibres (circ. muse. I., pi. 11, fig. 60) resembling in
size and shape the cuticular longitudinal fibres of the body. Out-
side of this circular muscle layer is a mass of parenchyma fibres
passing in the same direction. Nuclei are scattered about irregu-
larly in the neighborhood of this layer, but often at some little
distance from it.

In the larger of these canals, a large tuft of cilia projects
into the lumen, through a break in the cuticular lining, running
the whole length of the canal. These ciliated canals receive the
smaller non-ciliated ones. The non-ciliated canals are plainly to
be seen in the living animal in the lateral frills, in close relation
to the follicles of the vitellaria, emptying on the one side
(lateral) into a small non-ciliated "sinus terminalis" (Wag-
ener) and on the other side (medial) into a ciliated canal of
about twice their own diameter (pi. 35, fig. 20). Still further
mediad, near the base of the lateral folds, is a second ciliated
canal, of the same size and appearance as the first. In it, how-
ever, the waves of motion traverse the cilia in the opposite direc-
tion from that taken in the more lateral canal. Thus, if in the
outer one waves pass from anterior to posterior, in the inner one
they run from posterior to anterior. The motion is almost in-
credibly swift and very regular.

The largest excretory canals lie within the central core of the
body, among the inner longitudinal muscles, the ovaries and
testes. Running the length of the body, on either side of the
uterus, is a very large ciliated canal, the largest in the body. The
structure of these canals is like that of the smaller peripheral
canal described above, except that the lining is thicker, the muscu-
lar layer better developed and nuclei more numerous in the neigh-
borhood of the wall of the canal. In no case has it been possible
to recognize any connection between the cilia of the canals and
the neighboring nuclei, such as was figured by Lonnberg (1891,
Taf. 3, figs. 39,40).

Around the anterior margin of the acetabulum there is a
fairly well-defined ring-canal, receiving many small anastomos-
ing longitudinal branches. Its diameter and relations shown

Watson: The Genus Gyrocotyle. 409

by sections do not indicate that it is connected with the central
canal-system, but rather with the peripheral non-ciliated system
of the lateral folds. The same is true for the dense network
of non-ciliated canals found in the posterior rosette. This rosette
region is riddled through and through with small carialsV much
like capillaries in size and structure of wall. No flame-cells
appear in this region. A ring-canal appears in the "neck," at
the level of the canal opening. Like the acetabular ring, this
appears to be immediately connected with the peripherally
situated canals, and through them with the deeper-lying larger
vessels (ant. ex. r., pi. 36, fig. 26.)

In several specimens a dilation of one of the large longitudinal
canals has been found in the region of the vaginal opening, usu-
ally posterior to it. These dilations take the form of a thin-
walled sphere, into which the large canal empties, containing
droplets of a structureless yellow material. It is certain that
this "bladder" is not a constantly occurring structure. No
external openings have ever been found, except by Spencer
(1889) on G. rugosa : " Wagener was unable to find any external
opening of the excretory system, but, after long searching, I have
been able to find two unmistakable openings on the ventral sur-
face, one on either side of the body, slightly in front of the open-
ing of the uterus to the external surface." Lonnberg (1891)
was unable to find these openings ; the writer has never seen them.
But it seems highly probable that these temporary "bladders"
may burst through the wall of the body to form a temporary
external opening, closing up after the collapse of the * ' bladder, ' '
due to the discharge of its contents.

VII. NERVOUS SYSTEM.

The nervous system of Gyrocotyle is of great interest with
reference to the problem of orientation, both in the genus itself
and in merozoic cestodes. It may be divided into central and
peripheral parts, according to the muscle layers with which the
nerve stems are related. The sense in which the phrase "central
nervous system" is used is of course quite distinct from the
meaning usually attached to it when applied elsewhere, as for
example to vertebrates. Neither should this use of the term be

410 University of California Publications in Zoology. [VoL- <>

confused with that advocated by Cohn (1898) for merozoic ces-
todes. He believes that all of the longitudinal nerves, together
with their transverse connections in scolex and proglottides, con-
stitute the central nervous system. The branches from these to
the various organs and to the surface of the animal he regards
as the peripheral system. While all the evidence indicates that
such a division would be justifiable in Gyrocotyle, it is not in this
sense that the terms are here used, but purely with reference to
position, not at all with reference to structure or function.

The peripheral nervous system consists of eight longitudinal
stems, lying in the intermediate muscle layers, just outside the
outer longitudinal set of fibres, and communicating with the cen-
tral system by means of the anterior nerve ring around the mar-
gin of the acetabulum. There are no ganglion cells in these
strands; they are very small and exceedingly difficult to trace.
With borax carmin and Lyon's blue they stain a very clear light
blue, and can be recognized with some ease in the neighborhood of
the acetabulum, especially near their junction with the anterior
ring. These strands have not been previously described. They
innervate the intermediate muscle layers (outer transverse and
outer longitudinal), wherever these occur in the body proper,
in the inner layers of the acetabulum, and in the inner layers of
the funnel.

In dealing with the central system it is important to remem-
ber and recognize the existence of these extra-central nerves, for
it is only when the central system is clearly distinguished from
the others that the relations of its parts become intelligible. It is
perhaps because of their failure to take account of this division
that the results of investigators of the nervous system of Gyro-
cotyle have shown so little agreement in details.

It is hardly necessary to say that the study of the nervous
system by means of serial sections, already made difficult by the
great contractility of the body, is rendered a much more serious
problem by the necessity of dealing not with two stems, their
branches and connectives, but with six or perhaps ten such stems.
To determine whether a complete ring is present in any part of
the body becomes a task of serious difficulty, and indeed one im-
possible without the assistance of the relations of the different

1911] Watson: The Genus Gyrocotyle. 411

sets of nerves to the muscle layers, which can always be distin-
guished from one another.

The central nervous system lies within (mediad to) and
among the inner longitudinal muscle-fibres, and within (mediad
to) the large longitudinal excretory canals. It consists of the
following parts :

1. Two lateral longitudinal stems.

2. An anterior bridge commissure and an anterior ring com-
missure, in the acetabular region.

3. A posterior bridge and two ring commissures, joined by
eight longitudinal connectives.

The nervous system was first recognized by Wagener (1852),
who saw only the anterior commissure. Monticelli (1889a) and
Spencer (1889) recognized the posterior commissure. According
to Spencer, this commissure is continuous around the canal, form-
ing a complete ring surrounding the canal ; the dorsal half of the
commissure extending farther posterior than the ventral half.
According to Monticelli this is not the case. He found that on
the ventral surface the two parts did not unite, but merely ran
alongside each other, then separating passed posteriorly, each
ending independently in the margin of the "Trichter. " Both
these investigators agree in placing the anterior, heavy and in-
dubitable part of the commissure on the same surface as the
canal opening, i.e., the ventral face (dorsal of Spencer and Monti-
celli). This puts the anterior (acetabular) commissure on the
opposite surface from the posterior (scolex) commissure.

Lonnberg (1891) described both commissures as lying on the
ventral surface. He found the posterior commissure to be a
bridge, not a complete ring, posterior to which the longitudinal
stems are continuous to the margin of the funnel, where each
breaks up into many branches, which probably form by their
anastomoses a ring about the margin of the funnel opening.
Thus his account differs from Monticelli 's in (1) the presence
of a marginal ring at the posterior extremity, and ( 2 ) the absence
of any statement of the near approximation of the posterior
nerve stems to each other posterior to the commissure. It differs
from Spencer in the first point, and also in the absence of any
indication of the completion of the posterior commissure to form

412 University of California Publications in Zoology. [V°L- 6

a ring surrounding the canal. Lonnberg also differs from Spen-
cer and Monticelli in placing the posterior commissure "ven-
trally. " The apparent contradiction in these results is due to
incomplete rather than erroneous observation.

The longitudinal nerve stems or lateral connectives (long. n.
st., pi. 39, fig. 42) lying in the dorsal half of the central region
of the body, run from the anterior to the posterior commissure.
They are separated from each other by one-third of the width
of the body in the anterior quarter of the body, but spread
further apart at the level of the birth pore, lying near the lateral
margins of the uterus. In the region of the ovaries they ap-
proach the median line again. Each stem gives off two sets of
branches, one in the sagittal and the other in the horizontal plane.
These branches are heaviest and most profuse in the regions of
the ovaries and the testes.

The acetabular nervous system consists, as elsewhere in the
body, of a central and peripheral portion. The peripheral system
comprises eight longitudinal strands with many anastomosing
branches. These strands innervate the peripheral muscle layers
of the body, and the homologous muscle layers of the acetabulum.
They come into relation with the central nervous system by
means of the anterior ring in the margin of the acetabular open-
ing (acet., pi. 39, fig. 42.)

The central system includes an anterior bridge commissure,
lying just in front of the posterior margin of the acetabulum on
its dorsal surface ; a pair of anterior lateral stems with branches ;
and an anterior nerve ring (pi. 39, fig. 42). The commissure
forms a bridge between the two longitudinal nerve stems. There
is a ganglionic enlargement of the lateral stem at a point where
it is joined by the bridge commissure. The commissure is in the
shape of an arch, enlarged at the ends and smaller in the middle
of its course. Its mass with respect to the rest of the body varies
widely, perhaps with the state of contraction of the animal ; it is
fairly constant with respect to the nervous system as a whole.
No branches are given off from the commissure itself. The mar-
ginal ganglionic knots are about double the diameter of the
lateral stem (ant. br. comm., pi. 39, fig. 42). They are enlarge-
ments of the lateral stems, beginning at the point where the

1911] Watson: The Genus Gyrocotyle. 413

anterior bridge commissure is given off and reaching their great-
est size in that region. Anteriorly they pass into the anterior
longitudinal stems, extending fully a third of the total distance
from the commissure to the anterior ring. From eaeh_ganglion
the following branches are given off (pi. 36, fig. 27).

1. An anterior lateral nerve stem, a prolongation of the
anterior nerve stem in front of the anterior bridge commissure,
runs forward along the dorso-lateral margins of the acetabulum.
From these stems there are given off dorsal, ventral and lateral
branches to the body musculature, and median branches to the
outer coat of the acetabular muscles. Certain branches from the
dorsal surface run forward, to the dorsal sensory ridges later
described, and from the ventral surface of each stem near its
anterior extremity a large nerve runs forward ventro-laterally to
the sensory pits on the ventral serface. At the anterior extremity
of the worm near the acetabular opening, each lateral nerve stem
divides into two branches which pass, one ventrad, the other
dorsad, at right angles to the longitudinal stem. These nerves
break up in the median line to form the anterior nerve ring.

2. Four nerves are given off from the ventral surface of the
knot, two of which are directed posteriorly and two anteriorly.
The posteriorly directed pair of branches are distributed to the
follicles of the testes (test, n., pi. 36, fig. 27). The anteriorly
directed pair pass around the posterior margin of the acetabulum
on its ventral surface and send out branches on this surface.

3. Four nerves arise from the dorsal surface, running for-
ward and laterally into the inner longitudinal muscles.

4. Two nerves arise from the lateral surface of the ganglion,
one running forward and ventrally, the other towards the dorsal
surface, into the central musculature of the body and the vitel-
laria.

5. Several nerves are given off from the anterior face of the
ganglion and run forward along the surface of the acetabulum
to innervate the outer meridional acetabular muscle coat (the
homologue of the inner longitudinal coat of the body muscula-
ture).

The Anterior Ring Commissure. — The anterior end of the
worm is pierced by a round opening into the cavity of the

414 University of California Publications in Zoology. [VOL. 6

acetabulum. The circular margin of this opening is rounded and
thick, due to the presence of a heavy sphincter. Surrounding
this opening, within the sphincter of the margin, is the anterior
excretory ring and the anterior nerve ring. This is formed as
described by dorsal and ventral terminal branches of the anterior
lateral stems, united in the median line by a dorsal and ventral
anastomosis. This ring is thus composed laterally of a single
large nerve, but in the median line of many small anastomosing
threads (pi. 39, fig. 42). In its lateral portions at the points of
junction with the lateral nerve stem, there is a slight enlarge-
ment in which typical ganglion cells are present. From the ring
are given off nerves ramifying in the margin of the acetabular
opening, and in the outer layer of the acetabular muscle coat.
The ring also receives the peripheral stems mentioned above.

The Central Nervous System of the Rosette and Canal. — The
posterior rosette may be considered as composed of two parts, —
a funnel and a canal. The wide-mouthed funnel with glandular,
much folded walls is circular in cross-section, and roughly V-
shaped in sagittal section, the point of the V being directed
anteriorly. That is to say, the course of the funnel tube is almost
straight forward through the center of the body (pi. 46, fig. 76).
Its posterior margin is bordered by the frills which make up the
posterior rosette. At the base of these frills there is a thickening
of the walls of the funnel commonly referred to as the neck of
the rosette, due to the formation of a sphincter by the inner
transverse muscle. This funnel leads into the canal, a narrow,
non-glandular region, passing a little anteriorly and almost
directly ventrad to open on the ventral surface of the valve-like
canal opening. The region where the canal joins the funnel,
the apex of the V, will be called the tope, from the corresponding
region of an ordinary funnel. The walls of the canal itself show
no central innervation. There are apparent fine fibres of the
intermediate and cuticular nerves but no branches of the central
system. From the tope of the funnel posteriorly the walls of the
passage are weakly innervated by a complicated set of commis-
sures, stems, and branches, all belonging to the central system.
consisting of the following parts :

1. A posterior bridge commissure connecting laterally situ-
ated ganglionic knots.

1911] Watson: Tlie Genus Gyrocotyle. 415

2. A proximal ring commissure directly connected with the
posterior margin of the ganglionic knots.

3. A distal ring commissure, connected with the proximal
ring by eight connectives.

4. Anastomoses between these connectives in the lateral walls
of the funnel.

5. Eight branches running posteriorly from the distal ring
commissure into the folds of the terminal rosette.

The main longitudinal stems lie as above noted in the dorsal
half of the central region. At a point just posterior to the tope,
close to the wall of the funnel, each stem enlarges to form a
ganglionic knot (pi. 35, fig. 17). These knots are connected
dorsally by a transverse bridge which is fairly large laterally,
but narrows to the merest thread in the median region, not dis-
tinguishable with a magnification of less than 300 diameters.
The ganglionic knots at the ends of this bridge commissure pass
diagonally across the lateral walls of the cavity. This forms the
"bow-commissure" of Spencer. Each ganglionic knot divides
into two main branches, a dorsal and a ventral, which pass
toward the median line to join similar branches from the opposite
side, thus forming the proximal ring commissure, lying about
half-way between the neck or sphincter region and the tope of
the funnel. It was probably the ventral half of this ring which
Spencer regarded as forming the "dorsal" region of his bow-
commissure, embracing the funnel. Lonnberg's "ventral" com-
missure was probably the dorsal half of this same ring, plus the
lateral ganglionic knots of the posterior commissure, the half-
ring and knots being connected by one of the primary lateral
connectives to be described later. The middle third of the dorsal
posterior bridge commissure is so delicate that it is easily over-
looked. However, the continuity and independence of both the
dorsal commissure and the dorsal half of the ring can be easily
demonstrated in serial sections with a magnification of about 300
diameters.

In the neck or sphincter region there is formed a second ring
which completely encircles the posterior margin of the funnel at
the base of the frills. This posterior ring is connected with the
proximal ring by eight connectives (lat. long, conn., pi. 35, fig.

416 University of California Publications in Zoology. [V°L- 6

17). Of these, the two lateral pairs are derived from the primary
longitudinal stems, each divided into two and continued pos-
teriorly from the ganglionic knots. The two lateral connectives
of each side are interconnected by anastomosing branches, lateral
and irregular, nearly parallel to the proximal ring, and fre-
quently quite as heavy. The dorsal and ventral median pairs,
lying in the median third of the funnel wall, extend only from the
proximal ring to the distal ring. From the distal ring branches
run out into the frills at the points where the longitudinal con-
nectives enter the ring.

Distribution of Ganglion Cells. — No attempt has been made
to deal with the histology of the nervous system, further than
is necessary for understanding the significance of the structure of
the ganglionic knots at the lateral margins of the anterior and
posterior commissures. The nerve stems, both longitudinal and
peripheral, are composed of exceedingly delicate fibrils, woven
together in a dense meshwork (pi. 34, fig. 16). Around the stem
is a covering of nucleated parenchyma fibres, or sheath cells.
There are no nuclei within the meshwork. At intervals along the
two main lateral stems, where branches are given off, there appear
just inside these ''sheath cells" a cluster of cells having homo-
geneous cytoplasm and oval, clear nuclei, with a few chromatin
nucleoli (gang. 1st., pi. 44, fig. 67). These correspond to the
"first type" of ganglion cells described by Pintner (1880) and
by Niemiec (1886). They occur only at the margins of the main
lateral nerve stems and in the nerve rings : the anterior ring, the
proximal and distal ring commissures. In the ganglion knots
there occur very large cells, with coarsely granular, intensely
staining cytoplasm, and a large, clear nucleus containing one
large and several small chromatin nucleoli (pi. 44, fig. 66) . With
Lyon 's blue these cells stain a dark violet blue quite distinct from
the clear "Himmelblau" as Lonnberg calls it, of the the fibrous
tissue. These occur only in the ganglion knots of the anterior
and posterior bridge commissures, not at all in the median sec-
tions of the commissures, in the rings, or lateral stems. They are
not numerous, their total number not exceeding seventy or eighty.
They correspond to the second type of cells described by Pintner
and by Niemiec, commonly referred to as giant cells.

1911] Watson: The Genus Gyrocotyle. 417

VIII. SENSE ORGANS.

The presence of sensory end-organs in cestodes has been
several times suggested, but, never established to the satisfaction
of investigators in general. Lang (1891) says "The cestodes no
longer possess any specific sensory organs." Braun (1894) in
Bronn's Thierreich (p. 1300) quotes Blumberg (1877) as observ-
ing nerve endings in the limiting membrane or cuticle of the
Taenia of horses. These endings are in the form of delicate
threads terminating in a swollen knob. Braun thinks that, con-
sidering Zernecke 's observations, it seems probable that Blumberg
saw actual nervous end-organs. Linton (1891) briefly describes
an organ of hearing in Otobothrium crenacolle, as a small struc-
ture covered with hairs, situated on the bothridia. Beyond these,
and Schiefferdecker 's (1874) interpretation of flame-cells as
nerve endings; there are no references to sense-organs in the
literature of cestodes, in so far as that is known to the writer.

In Gyrocotyle the whole acetabulum functions much as does
the proboscis of the rhabdocoelan Proboscidea, as a highly
efficient organ of exploration, or one might say of touch. While
the whole surface of the acetabulum is richly innervated, there
are on the margin of the opening of the acetabulum two ridges
with a peculiar and significantly rich nerve supply, and two shal-
low pits in which lie flat plates of nervous tissue, end-organs of a
pair of heavy branches from the anterior lateral stem above
referred to.

1. The Sensory Ridges or Papillae. — These lie one on each
side of the latero-dorsal margin of the acetabular opening. A
nerve from the anterior lateral stem spreads out within each
"papilla," its branches running to the base of the very thin
limiting membrane.

2. The Sensory Pits (sens, pits, pi. 36, figs. 23, 25) lie
farther laterad than the papillae, and on the opposite or ventral
surface of the acetabular margin. They consist of a definite
depression, covered with a differentiated membrane, immediately
beneath which lies a plate of nervous tissue, formed by a very
heavy branch from the anterior longitudinal nerve. This does
not break up but ends abruptly as a plate of nervous tissue

418 I' Diversity of California Publications in Zoology. tv°L- 6

immediately beneath the limiting membrane of the pit. These
pits stain an intense blue when the living animal is treated with
methylen blue. They take up the stain quickly and hold it for
several hours after removal from the staining medium. The
function of these pits is totally unknown, but the presence of
central nervous tissue immediately beneath their surface indicates
their possession of some sensory function.

E. GENERAL DISCUSSION.

This investigation was undertaken and carried out in the hope
of obtaining evidence which would definitely settle the question
of antero-posterior orientation. This evidence was sought along
three lines: (1) morphological relationships of organs, in them-
selves and compared with other platyhelminths ; (2) behavior of
the living animal; (3) embryological history. The writer has
unfortunately failed to find any extra-uterine embryological
material, and has no evidence from this source to offer. The fact
that Chimaera colliei can not be kept in aquaria, even large ones,
with any measure of success (Dean, 1906, p. 16) makes the life-
history a hard problem to attack. I am convinced that we have
thus far no hint whatever as to the intermediate host of Gyro-
cotyle. The occurrence of decaying, sexually mature forms in
Mactra edulis has, it seems, no bearing on this question. The
youngest forms reported, the only immature ones in fact, are the
young individuals found by Lonnberg in the spiral valve of C.
monstrosa. This fact indicates that the worm enters the host in
a sexually immature condition. The fact that no hatched em-
bryos have been found in the intestinal contents of Chimaera
indicates that Gyrocotyle, like other cestodes, has at least two
hosts. Further than this we know nothing of its life-history.2

Conclusive morphological and functional evidence bearing on
the question of orientation has been found in abundance. In the
course of this work, evidence bearing on certain other questions

2 The discovery by Hungerbiihler (1910) of cysticercoids in the paren-
chyma of Gyrocotyle rugosa is a recent addition to our knowledge of the
embryology of the genus. The embryo resembles the ten-hooked embryo
already figured by Spencer as an extra-uterine embryo. Its position (near
the uterine pore) suggests that the wall of the uterus may have given way
in that region in the preparation of the specimen.

1911] Watson: The Genus Gyrocotyle. 419

of interest to students of the phylum has been found; this will
be briefly discussed before proceeding to consider the main prob-
lems of orientation.

I. CUTICULA.

The question of ectodermal origin of the "cuticula" or limit-
ing membrane of the body, and the significance of the "sub-
cuticular cells ' ' in the trematodes and cestodes has been recently
reviewed by Professor Pratt (1909). Blochmann's theory, re-
garding the subcuticular cells as a sunken epithelial layer, and
the limiting membrane as a true cuticula, such as is found in
arthropods and annelids, Pratt considers completely discredited
by various pieces of embryological and morphological evidence.
He considers the "cuticula" to be nothing but a closely matted
layer of parenchyma fibres, from which the nuclei have dis-
appeared. The observations made on Gyrocotyle seem to bear
out this view of the limiting membrane. Of particular interest
are the conditions of the vagina, receptaculum seminis and
uterus. Here the lining of these ducts is in direct continuity
with the cuticula at their openings ; there is a gradual transition
from this cuticular lining to one of comparatively loose-matted
fibres and indefinite boundaries, containing unmistakable paren-
chyma nuclei and passing on its inner surface indistinguishably
into a typical parenchymatous net in which lie muscle fibres.
This same transition can be seen on the inner surface of the folds
of the posterior rosette. Furthermore, nowhere in Gyrocotyle
have I found a definite layer of epithelial cells. Such tissue,
described by Lonnberg for the ducts of the reproductive system,
resolves itself in favorable preparations and under high magni-
fication into the dense parenchymatous layer, fibrillated and
without cell-walls, above described.

One statement made by Pratt in support of this view of the
cuticula as not associated with the subcuticular cells is not borne
out by conditions in Gyrocotyle. He says: "If now the cuticula
is the product of the underlying subcuticular cells, we should
expect to find some special development of them beneath the
hooks and spines, especially where these are very large, just as
in the integument of insects a cuticular hair or scale is invariably

420 University of California Publications in Zoology. [VOL- 6

situated over the enlarged hypodermal cell which produces it.
Nothing- of the sort exists in trematodes and cestodes. The sub-
cuticular cells beneath the hooks and spines do not differ in size,
number or arrangement, from the adjacent cells and in the
monogenetic trematodes, which are often provided with gigantic
hooks, no subcuticular cells at all are present."

In Gyrocotyle the subcuticular layer of cells is much increased
in thickness in the neighborhood of a spine and is closely related
to it.

A question of far greater difficulty is that of the function of
the subcuticular cells. If not related to the formation of the
cuticula, what is their function? Pratt makes two suggestions;
first that they are secretory in function, forming an antibody
for the -protection of the worm from the chemical action of the
medium in which it lives. This is supported by the fact that
these cells are altogether lacking in monogenetic trematodes. A
second suggestion (see Looss, 1894), is that these cells constitute
an undifferentiated embryonic layer, from which new cells of
various tissues are formed during the lifetime of the animal.
The only evidence in support of this theory is the statement that
in certain individuals known to be of advanced age, the sub-
cuticular layer was greatly reduced.

The glandular theory finds no definite support in the con-
ditions in Gyrocotyle. The unmistakable gland cells here present
are found as above noted in the central core of the body, not
in relation to the peripheral layer. The statement made by
previous investigators that the subcuticular layer gives rise on
the inner surface of the rosette folds to gland cells has not been
verified in the writer's preparations. However, this does not
militate against the possible glandular nature of these cells. The
intense staining reaction of the cytoplasm of at least some of
these cells recalls the appearance presented elsewhere by un-
mistakable gland cells.

Looss 's suggestion seems hardly susceptible of proof or dis-
proof. It is difficult to believe that the layer of cells so closely
related in position to the cuticula, varying in thickness with its
thickness and increasing in the region of special cuticular struc-
tures such as spines, should be totally unrelated to the body-

1911] Watson: The Genus Gyrocotyle. 421

covering. The conditions of Gyrocotyle indicate very clearly
that the subcuticular layer is related to the cuticular muscula-
ture, and that some at least of its cells are to be considered the
myoblasts of the cuticular muscles. There is much in the litera-
ture of the subject in harmony with this suggestion ; Blochmann
(1896) shows in his diagrammatic cross-section of Ligula myo-
blasts lying near the subcuticular layer. He was able to distin-
guish these cells from the rest of the subcuticular layer. This the
writer is unable to do in Gyrocotyle. The fact that the cuticular
musculature always increases with the thickness of the cuticula,
disappearing as the cuticula thins out and passing into the
fibrous nucleated layer above described, lends support to the
suggestion. In the lining of the receptaculum seminis and the
posterior end of the vagina, no subcuticular cells can be seen.
These ducts are surrounded by simple nucleated muscle fibres.
In the early coils of the uterus, where this same nucleated
fibrous lining is found but where cilia are also present, the cilia
pass through the fibrous layer and are in connection with scat-
tered nuclei lying just beneath the fibrous lining. Further
along the course of the uterus, where a definite non-nucleated
fibrous lining has appeared, beneath which lies a cuticular mus-
culature, the subcuticular layer appears, just as it does beneath
the cuticular musculature of the body-covering. All the facts
available indicate that at least a large part of the subcuticular
cells are related, not to the cuticula, but to the cuticular muscu-
lature as myoblasts. The subcuticular cells in the neighborhood
of the spines are probably related to the protractor musculature
of these structures.

II. ORIENTATION.

The question of antero-posterior orientation of cestodes is
one of peculiar difficulty. Their endoparasitic and attached
mode of life makes it impossible, in general, to settle the matter
by the test ordinarily applied, that of the direction of locomo-
tion. The usual custom, reflecting the influence of Leuckart, has
been to regard the scolex end as anterior, the free end as pos-
terior. Many early workers, among them Perrier, Grassi, and
Blanchard, reversed this orientation, looking on the scolex as

422 rnircrsity of California Publications in Zoology. [VOL- 6

the posterior end. In the Cestodaria, conflicting views as to the
orientation of the various genera have long existed, and in the
genus Gyrocotyle the question never has been conclusively
settled. It is of peculiar importance for the problem of cestode
orientation in general that these relations should be well estab-
lished in Gyrocotyle, for there is no functional antero-posterior
orientation in the adult merozoic cestode and the problem there
is one of comparative morphology and phylogenetic develop-
ment. Since Gyrocotyle is in every respect a primitive, rela-
tively simple form, parasitic in one of the most ancient of
vertebrates, it seems reasonable to assume that this cestode may
give some hint as to the extremity at which the ancestral cestode
most probably developed its organ of firm attachment. Observa-
tions of the living animal have shown that in Gyrocotyle there
is still a definite functional antero-posterior orientation, due to
the fact that it is not a permanently attached form but is still
capable of locomotion.

Diesing (1855) regarded the acetabular end of Gyrocotyle
as anterior, but his grounds for this decision are not clear.
Working with a few poorly preserved specimens, he had little
on which to base his conclusions. Wagener (1852), who did
careful work on the living animal and on sections, had proposed
the same orientation on the basis of the active exploring move-
ments of the acetabulum in the living animal, and of the loca-
tion of a bridge commissure of the central nervous system at the
base of the acetabulum. This orientation was followed by suc-
ceeding investigators up to Spencer (1889), who reversed it on
the strength of the discovery of a similar and much heavier bow-
commissure at the rosette extremity. He did not observe living
material. Lonnberg, working on a large quantity of living and
preserved material, followed Spencer in regarding the rosette as
anterior, basing this decision on the behavior of the living animal
(the stretching out of the rosette and funnel into a long canal
which performed exploring movements and is directed forward
in locomotion, according to his observation), on the great de-
velopment of its nervous system in the funnel region, and the
greater abundance of ganglionic cells in that as compared with
the acetabular commissure, on the direction of the spinules,

1911] Watson: The Genus Gyrocotyle. 423

which point toward the acetabulum; and also on the fact that
the worm is always attached to the rosette extremity, since
"cestodes always attach by the head end." Haswell (1902)
rejects Lonnberg's and Spencer's view on orientation on the
basis of homologies in position between the reproductive organs
of Gyrocotyle and of merozoic cestodes, to which in his opinion
it is very closely related. "The end which bears the sucker is
seen as the result of such a comparison, to correspond to the
scolex end in the segmented cestode." This homology I regard
as unjustifiable, as will be pointed out later. Benham (1891),
comparing the reproductive organs of Gyrocotyle with those of
Amphilina and the heterocotylean trematodes, concludes that the
acetabulum of Gyrocotyle corresponds to the anterior sucker of
the trematodes, while the rosette organ and its peculiar proboscis
possibly represents the posterior caudal disc of the latter class.
An examination of the literature of the genus thus shows
that the orientation of Gyrocotyle has been made on the follow-
ing grounds :

1. Behavior of living animal.

2. Cephalization of the nervous system.

3. Homologies of the reproductive organs with similar struc-
ture in the merozoic cestodes and in the trematodes.

4. Direction of spines.

Conclusions based on a consistent and constant functional
orientation of the living animal in locomotion and general move-
ments are unquestionably well grounded. Conclusions resting
on the cephalization of the nervous system assume that the
nervous system will be centralized and most richly developed in
the head region. This is true for worms in general and for all
platyhelminths which retain in any marked degree the power of
moving from place to place. In the trematodes there, takes
place, however, a remarkable development of ring-commissures
in connection with the development of powerful organs of attach-
ment. This is especially noticeable in the large posterior termi-
nal sucker of the heterocotylean trematode (pi. 47, fig. 79). It
seems very probable that in a permanently attached form, like
the cestode, in which the most powerful and highly specialized
musculature of the body is centered in the organ of attachment,

424 University of California Publications in Zoology. [VOL- 6

there will be found in that region the greatest and most highly
developed mass of the nervous system, quite independent of
whether the attached end is the homologue of the ancestral
anterior or cephalic extremity or not. Therefore, I am inclined
to question conclusions as to orientation based on "cephaliza-
tion" of the nervous system, unless confirmed by other unques-
tioned evidence.

The third basis on which the question has been decided, the
homologizing of the reproductive organs and openings with those
of trematodes or merozoic cestodes, can obviously only be applied
in the direction of trematodes as long as the orientation of
merozoic cestodes themselves is in question. Furthermore,
antero-posterior relations among these structures are fixed and
constant for trematodes ; while among cestodes the most astonish-
ing variations present themselves.

The direction of the spines is not at all a decisive piece of
evidence inasmuch as spines may be quite as useful to the animal
if directed anteriorly and serving as a means of attachment, as
if directed posteriorly and serving as aids to locomotion.

Wagener's orientation of Gyrocotyle, regarding the aceta-
bulum as anterior, the rosette as posterior, is justified in my
opinion on the following grounds:

1. This is the functional orientation of the living worm. The
rosette end is relatively quiescent while the acetabular end is
exceedingly active in exploring movements, is directed anteriorly
in well-defined progressive locomotion, and leads in all righting-
up movements. The rosette end never leads in locomotion except
when shoved backward by the doubling under of the active
acetabular end, and performs no movements other than a slight
rolling from side to side. This mode of behavior agrees with
that described by Wagener (1852) in his original account of
Gyrocotyle urna; but is totally at variance with Lonnberg's
(1891) observations on the same form. This discrepancy is
discussed above.

2. The position of the reproductive openings in Gyrocotyle
as compared with that in the heterocotylean trematodes homo-
logizes the rosette with the posterior sucker. The birth-pore of
Gyrocotyle is probably, according to Goto's (1891) view of the

1911] Watson: The Genus Gyrocotyle. 425

uterus in cestodes, the homologue of the female copulatory duct
in the Heterocotylea ; and the vagina is the homologue of the
heterocotylean uterus. This gives exactly similar anterior and
posterior relations to the openings of the ducts, the penis-open-
ings being very slightly anterior to the vaginal opening~and the
birth-pore lying most posteriorly and some distance to one side
of the other two.

3. Further morphological evidence in support of this orienta-
tion may be adduced as follows : In the first place there are two
pairs of abundantly innervated antero-lateral sensory areas, com-
parable in structure and location to similar areas in planarians
and certain heterocotylean trematodes. In the second place the
structure of the central nervous system, when compared with
that of the heterocotylean trematodes, affords morphological sup-
port to this orientation. The anterior commissure, giving off
sensory branches, the main and secondary longitudinal nerve
strands, the eight posterior branches and a posterior ring com-
missure are all common and similarly placed in Gyrocotyle and
the heterocotylean, as for example in Tristomum molae, whose
nervous system was described by Lang (1882), (pi. 47, fig. 79).
There is added in Gyrocotyle the delicate median portion of the
bridge commissure and the second ring commissure, which may
well have arisen in Gyrocotyle in correlation with the increased
mass and complexity of the musculature of the posterior organ
of attachment, the rosette. Thus a comparison of the nervous
system and the position and innervation of the organ of attach-
ment of Gyrocotyle with the heterocotylean compels us to homolo-
gize the rosette with the posterior sucker of the trematode.

The development of two bridge commissures at the two ex-
tremities of the body, approximately equal in abundance of
ganglion cells but the anterior supplied with a well-developed
median part which is very faint in the posterior one, indicates
the manner in which the evolution of the nervous system of the
merozoic cestode has taken place. This is, briefly, by the de-
generation of the anterior commissure associated with the
reduction in the locomotor and sensory functions of the animal,
and the great development of the posterior commissure and its

426 University of California Publications in Zoology. [V°L- 6

stems and rings, associated with the development of the powerful
musculature of the organ of attachment, the scolex.

That this is in harmony with the course of development
elsewhere in the phylum is shown by consideration of the
probable construction of the nervous system of the primitive
turbellarian-like ancestor and the changes it has undergone.
Throughout the phylum there is remarkable uniformity in the
ground-plan of the nervous system. The primitive structure in
the free-living Turbellaria is a sub-dermal plexus of fibres and
ganglion cells with a marked concentration of these at the
anterior end, and an increase in their number and size on the
ventral or creeping surface. This differentiates in two direc-
tions ; first in the segregation from the plexus of from six to eight
longitudinal strands, with irregular transverse connecting fibres ;
and second, in the increase in size of the main bridge-commissure
or brain, and in the development of secondary commissures in
the region of the brain and in the neighborhood of specially
developed musculature, notably in the pharynx and in organs of
attachment. This is to be seen in the remarkable development
of the posterior commissure in the heterocotylean in connection
with the development of the posterior sucker as the principal
organ of attachment. Further, two of the longitudinal nerve-
strands, the ventral, become more highly developed than the
rest. In the trematodes there are two longitudinal strands, con-
nected near the anterior extremity b}^ a bridge-commissure rich
in ganglion cells from the region of which arises a pair of sensory
nerves. There are in addition numerous peripheral longitudinal
strands and an indefinite number of cross-nerves, anastomosing
among themselves to form irregular ring-commissures about the
body. In the posterior region, in close relation to the large
sucker, is developed a complicated system of commissures and
rings. From this type the nervous system of Gyrocotyle has
been derived. The longitudinal strands are the same in both;
there are however two bridge-commissures in Gyrocotyle. But
the sensory function of the acetabulum and the forward direc-
tion of that extremity in locomotion, together with the well-
developed nature of that commissure throughout as compared

1911] Watson: The Genus Gijrocotyle. 427

with the delicate median thread of the rosette bridge-commis-
sure, all point to its unmistakable homology with the anterior
commissure of trematodes, and so with the typical "brain" of
the Turbellaria.

The posterior commissure with its accompaniments'" of com-
plicated rings and connectives has plainly been developed in con-
nection with the musculature of the funnel-shaped rosette-scolex.
This complex development of nervous structure in connection
with a highly developed musculature is strikingly shown
throughout the phylum. The ventral nerve stems, in connection
with the ventral creeping muscles, become heavier, more pro-
fusely branched, than their homologues near the dorsal surface.
The great complexity of the nervous system connected with the
posterior sucker of the heterocotylean has already been referred
to. The development of a posterior commissure, in itself rather
weak, but surrounded by a complex system of rings and connec-
tives such as is found in the rosette of Gyrocotyle, is exactly
what would be expected in connection with the development of a
complicated and powerful organ of attachment.

The comparison of the nervous system of Gyrocotyle with
that of the merozoic cestodes shows two main longitudinal stems
in both. These are more or less sharply differentiated but always
sufficiently clearly marked to be distinguished from the weaker
longitudinal stems, of which there are four or eight, correspond-
ing to the eight peripheral nerves of Gyrocotyle. Near the free
margin of the proglottid, which must be regarded as anterior,
these longitudinal stems are connected by a transverse commis-
sure, in close connection with the transverse canal of the
excretory system. In the scolex there is a heavy ganglionic
bridge-commissure, joining the longitudinal stems, lying midway
between the dorsal and ventral surfaces of the body. Peripheral
to this and in contact with it only at the points where the longi-
tudinal stems enter the commissures, is a more or less complete
ring-commissure. This ring reaches its most perfect develop-
ment in the Taeniadae, but is present in an incomplete form in
Ligula, in the Tetrarhynchidae, and in the Tetraphyllidea.
Distal to this commissure and ring is found a more or less clearly
developed ring joined to the former by numerous connectives,

428 University of California Publications in Zoology. [VOL- 6

usually six or eight in number. This ring appears in a rudi-
mentary form in the Tetraphyllidea and in the Tetrarhynchidae ;
it is well established and clearly marked in the Taeniadae and
the Dibothridiata.

It is at once evident that the nervous system of the rosette
in Gyrocotyle is much more easily homologized with the nervous
system of the merozoic cestode scolex than is the nervous system
of the acetabular region. There are two serious objections to at-
tempting to derive the nervous system of the cestode scolex from
the acetabular bridge-commissure and its anterior rings. In the
first place, the rosette and funnel constitute an efficient organ of
attachment, so strikingly like the scolices of many Tetraphyllidea
in mode of adhesion and probable developmental history — being
formed by partial fusion of the walls of a trough, and later dif-
ferentiation of the ends of the tube thus formed — that it seems
irrational to suppose that two structures of such fundamental
similiarity could have been developed independently in two
groups of organisms as closely allied as are the merozoic and
the monozoic cestodes. The acetabulum, on the contrary, never
functions as a sucker or organ of attachment; there seems to be
no possible relationship between this structure and any of the
familiar types of cestode scolex. There is every reason, on the
basis of function, derivation and structure for regarding the
rosette as a scolex of the phyllidian type. In the second place,
there is no "starting-point" for the formation of a ring about
the acetabular commissure, no matrix out of which to differentiate
the complex rings and connectives of the nervous system of the
scolex. Such a matrix is, however, afforded by the numerous
anastomosing branches and the two irregular rings and their con-
nectives, seen in the rosette extremity.

These facts, with other considerations previously given,
justify the homologizing of the scolex of the merozoic cestode
with the rosette of Gyrocotyle, a posteriorly situated organ of
attachment. This conclusion implies a functional reversal of the
nervous system of the ancestral flatworm in the course of its
development into a merozoic cestode. The greatest mass of
nervous tissue, cephalized in the primitive flatworm, comes to lie
in the posterior region of attachment of the cestode. The anterior

1911] Watson: The Genus Oyrocotyle. 429

commissure disappears and the ring grows weak with the assump-
tion of the sessile habit and the disappearance of sense-organs;
while the posterior commissure develops with the increase in
efficiency and complexity of the organ of attachment.

The orientation of cestodes here suggested has been advanced
by several investigators, on more or less substantial grounds,
from Perrier to the present day. This contention has been based
for the most part on embryological evidence, especially with
reference to the hexacanth onchosphaeres so characteristic of
cestodes. The well-established fact that the embryonic hooks
are at the extremity of the cysticercus opposite to the one on
which the organ of attachment is developed, and the further fact
that the hook-bearing part of the onchosphaere is directed for-
wards in the movement of the embryo, affords good ground for
seriously questioning, if not altogether- denying, the generally
accepted identification of the scolex as "head." Barrois (1889)
maintains that the anterior part of the scolex is that extremity
which bears the embryonic hooks; that this part of the scolex
gives rise to the first proglottis, which is therefore to be regarded
as the ' * Kopf theil ' ' of the primitive animal. Furthermore, the
establishment of a zone of growth in the "neck" of the strobila
suggests very strongly the penultimate "zone of growth" in
annelids, with which the "neck" of the cestode is homologized
if the scolex is recognized as posterior. A full presentation of the
evidence in favor of this orientation, derived from embryological
and comparative anatomical considerations such as the above,
was given by Cohn (1907). He remarks that the present orien-
tation of cestodes has been regarded as self-evident, incapable of
proof; and proceeds to show that, aside from the habitus of the
worm, there is no evidence in favor of this view. In his own
words: "Meine These ist, dass dem Geschlechtstiere der Ces-
toden ein Kopf iiberhaupt fehlt, und sein Hinterende zu einem
Haftorgane — dem Scolex — umgebildet ist." He regards the
hook-bearing tail-like appendage of the cysticercoids as the homo-
logue of the ancestral anterior extremity of the worm; this is
discarded, leaving the posterior organ of attachment and the
intermediate growing region of the body to constitute the adult
cestode. "Wir haben in den Proliferationsfahigen Scolices also

430 University of California Publications in Zoology. [VOL. 6

Tiere, die ohne ein wahres Vorderende, d. h. einen Kopf zu besit-
zen, mit dem aussersten Hinterende sich an der Darmwand fix-
ieren und mit ihren relativ vordersten Korperende frei in den
Darm hineinhangen. " He then proceeds to show that, first, the
presence of a differentiated intermediate portion between the
anterior and posterior segments of the body, secondly, the detach-
ment of the posterior segment and its transformation into the
sexually mature animal, and lastly, the location of the growing-
zone in the penultimate region of the body, are conditions whose
analogues can be readily found in other worms and also in
echinoderms, bryozoans, etc. On the last and probably most im-
portant point, the location of the growth-zone, he sums up the
evidence very briefly as follows :

"Ob wir also die normalen Wachstumserscheinungen, ob wir Kegenera-
tion oder die der autotomischen Teilung vorausgehenden Processe der Seg-
mentvermehrung betrachten: uberall finden wir dass sich die Wachstumzone
beiden genannten Tieren an aussersten Hinterende des Korpers befindet.
Bei der von mir vorgeschlagenen Orientierungen der Cestoden schaffen wir
also in bezug auf die Wachstumsverhaltnisse keinen Ausnahmefall sondern
erhalten im Gegenteil erst so die Moglichkeit, das Wachstum der Cestoden
durch Proliferation am Collum mit demjenigen anderer Vermes konform
auf zuf assen. ' '

These considerations, arising from facts of comparative
embryology and morphology of the invertebrates, taken together
with those arising from a study of the morphology of the primi-
tive genus Gyrocotyle, afford a warrant for serious question of
the validity of the generally accepted orientation of cestodes.
Furthermore, they constitute a more or less successful effort to
take this question out of the realm of "self-evident" hypotheses
incapable of either proof or disproof where, as Cohn pointed
out, it has too long existed. Further embryological research,
especially on such forms as Ampliiliiia and Gyrocotyle, is greatly
to be desired ; from this field the final word on the question must
be obtained. All the facts now at hand, however, seem to show
that this decision will be in direct opposition to the generally
accepted belief, and will place the organ of attachment in ces-
todes at the posterior extremity of the strobila.

1911] Watson: The Genus Gyrocotyle. 431

F. SUMMARY.

1. The genus Gyrocotyle is composed of the following species :
G. rugosa, G. urna, G. nigrosetosa, G. fimbriata. These are dis-
tinguished on the basis of the following specific characteristics :

(1) Character of folds of terminal rosette.

(2) The ratio between the distance from the opening of the
uterus to the tip of the acetabulum, and the distance from
the opening of the uterus to the level of the opening of the
penis.

(3) Character of lateral frills.

(4) Presence and distribution of spines.

(5) Size of tail-rosette.

(6) Presence of hooked embryo in uterine eggs.

(7) Presence of an eversible cirrus, adapted to self -impreg-
nation.

(8) Spinules lining ejaculatory duct.

(9) Operculated uterine eggs.

2. The normal habitat of the sexually mature individual is
in the spiral valve of the intestine of some species of the family
Chimaeridae. Reported occurrences of G. rugosa in bivalve
molluscs are probably accidental. Nothing is known of inter-
mediate host or life-cycle of the parasite.

3. The functional orientation of Gyrocotyle fimbriata directs
the acetabulum anteriorly, the rosette posteriorly. This is in
agreement with Wagener's observations on the living G. urna,
but in exact opposition to Lonnberg's observations on the same
form. The worm is capable of definitely directed locomotion and
is very active under favorable conditions. The exploring func-
tion of the acetabulum is strongly in evidence. The posterior
rosette functions strictly as an organ of attachment. The aceta-
bulum never functions as an organ of attachment.

4. This functional orientation is borne out by evidence from
the structure of the central nervous system and by the presence
on the margin of invagination of the acetabulum of a pair each
of sensory pits and sensory papillae, abundantly innervated by
heavy branches from the central nervous system.

432 University of California Publications in Zoology. tv°L- G

5. The acetabular portion of the nervous system is developed
in connection with the acetabular sense organs and with the
power of locomotion in a definite direction. It corresponds to the
" brain" of Turbellaria and to the anterior ganglionic commis-
sure in Trematoda.

6. The rosette portion of the nervous system is developed in
connection with the development of a powerful posterior organ
of attachment, and is comparable to the posterior ring-commis-
sure in the posterior sucker of a heterocotylean trematode.

7. The rosette of Gyrocotyle is in structure and function a
true scolex, and corresponds to that organ in merozoic cestodes.
This correspondence is strikingly shown in a comparison of the
nervous system of the rosette of Gyrocotyle and that of the scolex
of the merozoic cestodes.

8. On the basis of this evidence from comparative morphology
and of other evidence previously adduced from the embryology
of merozoic cestodes, it is proposed to regard the cestode scolex
as a posteriorly situated organ of attachment, the "neck" or
growing region as the antepenult region corresponding to the
antepenult segment in annelids, and the proglottis as the inter-
mediate region of the body. The anterior extremity has com-
pletely disappeared, according to this view.

9. The limiting-membrane in Gyrocotyle consists of a sur-
face layer, composed of delicate fibres in a homogeneous matrix,
and immediately beneath this a layer of transverse and a layer of
longitudinal muscle fibres, non-nucleated. These are connected
by fine processes with a layer of large cells lying in the paren-
chyma of the body, the subcuticular cells. Some at least of these
cells are to be regarded as myoblasts of the cuticular musculature.
There is no ground for regarding them as sunken epidermal cells.
There is no trace in any of the tissues of the body of an epithelial
layer of cells. The lining of the genital ducts is a meshwork of
fibres in a homogeneous matrix, with nuclei scattered through it.
This passes by gradual transition into the non-nucleated condition
described for the limiting membrane of the body.

10. The muscle fibres of Gyrocotyle are all nucleated except
those of the cuticular musculature. The latter are attached by

1911] Watson: The Genus Gyrocotyle. 433

delicate processes to deeper-lying myoblasts (in the subcuticular
layer).

11. The processes of cell-division in the maturation of the
ovum are mitotic. A large "nucleolus" is formed within the
nucleus and extruded into the cytoplasm. This process can
easily be mistaken for an amitotic figure.

Zoological Laboratory, University of California.
Transmitted April 25, 1910.

Of TH£

UNIVERSITY

OF
ILIF(

434 University of California Publications in Zoology. [VOL- 6

LITERATURE CITED.

Barrels, J.

1889. Une nouvelle conception de I'organisme Cestode. Rev. Biol.
Nord France, 2, 18-23, 199.

Benham, W. B.

1891. The Platyhelmia, Mesozoa and Nemertini. In Lankester's
Treatise on Zoology, part IV, 1-204, with figs.

Blochmann, F.

1896. Die Epithelfrage bei Cestoden und Trematoden. Vortrag ge-
halten auf der VI. Jahresversammlung der deutschen
zoologischen Gesellschaft zu Bonn. 16 pp., 2 pis. Hamburg.

Blumberg, C.

1877. Ein Beitrag zur Anatomic der Taenia plicata, T. perfoliata, und
T. mamillata. Arch. f. wiss. u. prakt. Thierhlkde, 3, 33-44.
pi. 1, figs. 1, 2.

Braun, M.

1889. Gyrocotyle, Amphiptyches und Verwandte. Zusammenfassender
Bericht. Centrbl. Bakter., 6, 436-441.

1894-1900. Cestodes. In Bronn's Klassen und Ordnungen des Thier-
reichs, 4, I b, pp. 927-1731, pis. 35-59.

Child, C. M.

1907. Studies on the relation between amitosis and mitosis. IV and
V. Biol. Bull., 13, 165-184, 19 figs.

Conn, L.

1898. Untersuchungen iiber das centrale Nervensystem des Cestoden.
Zool. Jahrb. Anat., 12, 89-160, pis. 6-9, text figs. A-J.

1907. Die Orientierung der Cestoden. Zool. Anz., 32, 51-56.

Dean, B.

1906. Chimaeroid Fishes and their development. 194 pp., 11 pis.,

144 text figs. Carnegie Institution of Washington.
Diesing, C. M.

1850. Systema Helminthum. I; xiii, 679 pp. Vindobonae.

1855. Sechzehn Gattungen von Binnenwurmern und ihre Arten.
Denkschr., d. k. Akad. d. Wiss. Wien, Math.-naturw. Cl., 9,
171-185, pis. 1-6.

Watson: The Genus Gyrocotyle. 435

1858. Eevision der Myzhelminthen. Abtheilung Trematoden. Sitzb.

k. Akad. d. Wiss. Wien, Math.-naturw. Cl., 32, 207-390, 2 pis.

1859. Eevision der Myzhelminthen. Abtheilung Bdellideen. Ibid. 33,

473-513.

1859. Nachtrage und Verbesserungen zur Eevision der Myzhelmin-
then. Ibid, 35, 421-454.

Goto, S.

1891. Diplozoon nipponicum n.sp. Journ. Coll. Sci. Imp. Univ. Japan,
4, 151-192, pis. 21-23.

Grube, E.

1855. Bemerkungen iiber einige Helminthen und Meerwiirmer. Arch.
f. Naturg., 21, I, 137-158, pis. 6, 7.

Haswell, W. A.

1902. On a Gyrocotyle from Chimaera ogilbyi, and on Gyrocotyle in
general. Proc. Linn. Soc., N. S. Wales, 1902, I, 48-54, pi. 7.

Hungerbiihler, M.

1910. Studien an Gyrocotyle und Cestoden. Ergebnisse einer von L.
Schultze ausgefiihrten zoologischen Forschungsreise in
Siidafrika. Inaug. Diss., Basel. 26 pp., 2 pis. Also in Jenais.
Denkschr., 16, 3 Lief.

Kofoid, C. A., and Watson, Edna E.

1910. On the Orientation of Gyrocotyle and of the Cestode strobila.
Advance print from Proc. Seventh Intern. Zool. Congr., 1907,
5 pp., 3 figs.

Lang, A.

1881. Untersuchungen zur vergleichenden Anatomie und Histologie
des Nervensystems der Plathelminthen. IV. Nervensystem
der Tricladen. Mt. Stat. Neapel, 3, 1-53, pis. 5, 6.

1891. Text-book of Comparative Anatomy, trans, by H. M. and M.
Bernard, 1, pp. 133-176; figs. 102-120.

Lay, G. T., and Bennet, E. T.

1839. "Fishes" in Zoology of Captain Beechey's voyage. Bohn,
London, 1839.

Lint on, E.

1891. Notes on Entozoa of marine fishes of New England with de-
scriptions of several new species. Part II. Eep. U. S. Fish
Comm., 1887, 719-899, pis. 1-15.

Lonnberg, E.

1890a. Helminthologische Beobachtungen von der Westkiiste Nor-
wegens. Cestoden. Bih. Svenska Ak., 16, IV, 5, 1-47.

436 University of California Publications in Zoology. tv°L- 6

1890b. Ueber Amphiptyches Wag. oder Gyrocotyle urna (Grube et
Wagener) Diesing Verb. Biol. Ver. Stockholm, 3, 55-61.

1891. Anatomische Studien tiber Skandinavische Cestoden. Svenska

Ak. Handl., N. F. 24, VI, 1-109, pis. 1-3, 47 figs.

1897. Beitrage zur Phylogenie der parasitischen Plathelminthen.
Centrbl. Bakter. 22, I, 674-684, 725-731, figs. 1-4.

Looss, A.

1894. Die Distomen unserer Fische und Frosche. Neue Untersuch-
ungen iiber Bau und Entwicklung des Distomen Korpers.
Zoologica, 6, Heft. 16, 1-296, 9 pis.

Monticelli, F. S.

1888. Saggio di una morfologia dei trematodi. Tese per ottonere la
privata docenza in Zoologia nella K. Universita di Napoli.
viii, 130 pp., ind. 1 p.

1889a. Sul sistema nervosa dell' Amphiptyches urna Grube et Wagener.
Zool. Anz., 12, 142-144.

1889b. Gyrocotyle Diesing, Amphiptyches Grube et Wagener. Atti
Ace. Lincei Kendic. cl. sci. fis. mat. e nat. (4), 5, 228-230.

1889c. Notes on some entozoa in the collection of the British Museum.
P. Zool. Soc. London, 1889, 321-325, pi. 33.

1890. Alcuni considerazioni biologiche sul genere Gyrocotyle. Atti
Soc. Ital. di Sci. Nat., 32, 327-329.

1892. Appunti sui Cestodaria. Atti E. Accad. Sci. fis. mat. Napoli

(2), 5, 67-78, 4 figs.

Niemiec, J.

1886. Untersuchungen iiber das Nervensystem der Cestoden. Arb.
Zool. Inst. Wien, 7, 1-60, pis. 1, 2.

Olsson, P.

1896. Sur Chimaera monstrosa et ses parasites. Mem. Soc. Zool.
France, 9, 499-512, figs. 1-9.

Pintner, Th.

1880. Untersuchungen iiber den Bau des Bandwurmkorpers mit be-
sonderer Beriicksichtigung der Tetrabothrien und Tetra-
rhynchen. Arb. Zool. Inst. Wien, 3, 163-242, 5 pis.

Pratt, H. S.

1909. The cuticula and subcuticula of trematodes and cestodes. Amer.
Nat., 43, 705-729.

Watson: The Genus Gyrocotyle. 437

Salensky, W.

1874. Ueber den Bau und die Entwickelungsgeschichte der Amphi-
lina G. Wagener (Monostomum foliaceum Kud.). Z. wiss.
Zool., 24, 291-342, pis. 28-32.

Schiefferdecker, P.

1874. Beitrage zu Kenntniss des feineren Baues der Taenien. Jena.
Zeitschr. f. Naturw. 8, 459-487, pi. 16.

Spencer, W. B.

1889. Anatomy of Amphiptyches urna Grube et Wagener. Tr. R. Soc.
Victoria, 1, 138-151, pis. 11-13.

Tower, W. L.

1900. Nervous system of the cestode Moniezia expansa. Zool. Jahrb.
Anat., 13, 359-384, pis. 21-26.

van Beneden, P. J., et Hesse, C. E.

1864. Recherches sur les bdellodes ou hirudinees et les trematodes
marins. Mem. Ac. Belgique, 38, Mem. I, 1-142, pis. 1-13.

Wagener, G.

1852. Ueber einen neuen in der Chimaera monstrosa gefundenen
Eingeweidewurm. Arch. f. Anat. und Physiol., 1852, 543-
554, pis. 14, 15.

1858. Enthelminthica No. V. Ueber Amphilina foliacea (Monostomum
foliaceum Rud.) Gyrocotyle Diesing und Amphiptyches Gr. W.
Arch. f. Natiirg., 24, I, 244-249, 1 pi.

Will, H.

1893. Anatomic- von Caryophyllaeus mutabilis Rud., ein Beitrag zur
Kenntniss der Cestoden. Z. wiss. Zool., 56, 1-39, pis. 1, 2,
2 figs, in text.

DESCRIPTION OF PLATES.

All figures are drawn with a Zeiss camera lucida, unless otherwise
stated.

PLATE 33.

Figs. 1-4. Changes in form of living Gyrocotyle, as seen by Lonnberg
(1891, Taf. Ill, figs. 34, 35, 36, 37). Fig. 1, the Ligula-like form; fig. 2,
rosette extended into long tube; figs. 3 and 4, form ordinarily assumed
by animal.

Figs. 5 and 6, Diesing's figures of Gyrocotyle rugosa. (Diesing 1855,
Taf. 1, figs. 17, 20.)

Figs. 7, 8, 9. Sketches of the acetabulum of a living specimen. Figs.
7, 9, extended; fig. 8, contracted.

acet. — acetabulum. p. op. — penis opening.

[438]

UNIV. CALIF. PUBL. ZOOL. VOL. 6

[WATSON] PLATE 33

7.

P. op

PLATE 34.

Fig. 10. Gyrocotyle fimbriata, ventral. Flattened, stained in borax-
carmine, cleared in cedar-oil. Showing arrangement of spines at pos-
terior extremity. Spines at anterior extremity not shown. X 4.

Fig. 11. G. urna, dorsal. Stained flattened specimen. Showing size
and character of folds of posterior rosette, and collar of spines about the
neck of the rosette. X 4.

Fig. 12. G. fimbriata, dorsal. Stained flattened specimen. X 4.

Fig. 13. G. urna (var.), ventral. Showing distribution of spines over
whole surface. X 7.

Figs. 14, 15. G. fimbriata. Showing canal-opening, with proboscis
inverted, fig. 14, and everted, fig. 15. Sketch without camera.

Fig. 16. Sagittal section of longitudinal nerve-stem, showing sheath-
cells, branch-nerves, and ganglion-cells of the first order. Iron haem-
atoxylin-erythrosin. X 100.

acet. — acetabulum.

can. op. — canal opening.

gang. 1st. — ganglion cell of the first order.

gen. notch — genital notch.

n. b. — nerve branch.

n. fib. — nerve fibre.

p. op. — penis opening.

par. — parenchyma.

post. ros. — posterior rosette.

prob . — proboscis.

rec. sem. — receptaculum seminis.

sh. c. — sheath cell.

ut. — uterus.

[440]

UNIV. CALIF. PUBL ZOOL. VOL. 6

[WATSON] PLATE 34

acet.

gen. notch
v. op.

ui.

rec.sem.

post. ros.

10.

post. ros.

can. op-

15.

post. ros.

11.

16.

PLATE 35.

Fig. 17. Gyrocotyle fimbriata. Diagram of the nervous system of the
posterior extremity. From Kofoid and Watson (1910, fig. 3).

Fig. 18. Diagram of the arrangement of muscle-layers in the saggital
section.

Fig. 19. Diagram showing track made by living specimen across
dish. Distance traversed, about 14cm.; four contractions.

Fig. 20. Sketch of network of excretory canals in the lateral fold
as seen in a living specimen. Zeiss-Greenough binocular.

Fig. 21. Diagram, reproductive system, typical heterocotylean trema-
tode. (After Benham, 1891, p. 51.)

cut. — cuticle.

cut. trans. — cuticular transverse muscle.

cut. long. — cuticular longitudinal muscles.

dist. r. comm. — distal ring commissure.

ex. can. — excretory canal.

in. long. — inner longitudinal muscle.

in. trans. — inner transverse muscle.

lat. A— lateral fold.

lat. long. conn. — lateral longitudinal connective.

marg. ex. sin. — marginal excretory sinus.

med. long. conn. — median longitudinal connective.

out. long. — outer longitudinal muscle.

out. trans. — outer transverse muscle.

post. ~br. comm. — posterior bridge commissure.

subcutic. 1. — subcuticular layer.

vit. — vitellaria.

v. op. — vaginal opening.

[442]

UNIV. CALIF, PUBL. ZOOL VOL. 6

[WATSON] PLATE 35

post. br. comm.

-^^7i med. long. conn.
lat. long. conn.

- — dist.r. comm.

17.

cut. trans.

out. trans
out. long. —

in.

ut.

— cut. long,
subcutic. I.
/
?$&$z-in. trans:

PP-** /

ex. can.

/

19.

w^m

Jj

I It f

,\: ««• ;/-i

SOF/

i/i. /on^.
ex. can.

21.

PLATE 36.

Figs. 22, 23. Gyrocotyle fimbriata, anterior extremity, ventral and
dorsal. Showing sensory pits and arrangement of spines. From pre-
served specimen. X 13.

Fig. 24. G. urna (var.). Sketch from life. Lateral view, expanded.
Showing spines, lateral fold and posterior rosette.

Fig. 25. G. fimbriata. Dorsal half of sagittal section of tip of aceta-
bulum, showing sensory pit. Borax-carmine, Lyon 's blue. X 430.

Fig. 26. G. fimbriata. Sketch, life, acetabular extremity. Showing
anterior excretory ring and deeper-lying ciliated canals.

Fig. 27. G. fimbriata. Sagittal section just laterad of the acetabulum.
Showing anterior longitudinal nerve stem, with branches and anterior
ganglionic knot. No histological detail; position of giant-cells within
ganglion knot marked by small circles.

acet. op. — acetabular opening.

ant. ex. r. — anterior excretory ring.

ant. gang. Jen. — anterior ganglion knot.

ant. lat. n. st. — anterior lateral nerve stem.

cil. — cilia.

cut. — cuticula.

ex. can. — excretory canal.

gen. notch — genital notch.

lat. /.—lateral fold.

par. nuc. — parenchyma nucleus.

p. op. — penis opening.

post. ros. — posterior rosette.

sens, pit — sensory pit.

sp. — spine.

test. — testis.

ut. po. — uterine pore.

test. n. — testicular nerve.

[444]

UNIV. CALIF. PUBL. ZOOL. VOL. 6

[WATSON] PLATE 36

ns. pit
acet. op.

27.

OF THE

{ UNIVERSITY J

PLATE 37.

Figs. 28, 29. Gyrocotyle fimbriata. Spines, from acetabular group.
Teased out. X 1000.

Fig. 30. G. fimbriata. Spine, from neck of rosette. From specimen
stained in borax-carmine and cleared in cedar-oil. Showing direction of
spine and attachment of muscles. X 430.

Fig. 31. G. fimbriata. Spine from neck of rosette. From same speci-
men as fig. 30. Drawn in situ. X 430.

Fig. 32. Spines from margin of anterior end, in front of genital notch.
Teased out. X 1000.

Fig. 33. G. fimbriata, transverse section. Cuticula absent. Showing
muscles of spine. X 430. Iron haematoxylin.

Fig. 34. Same as fig. 33. Showing structure of socket of spine.
X 1000.

Fig. 35. Spine from neck of rosette. X 1000.

muse. fib. — muscle fibre.
par. felt — parenchyma felt.
par. nuc. — parenchyma nucleus.
protr. m. — protractor muscle.
retr. m. — retractor muscle.
sp. — spine.
sp. base — spine base.
sp. sock. — spine socket.
sp. tip — tip of spine.

[446]

UNIV CALIF. PUBL. ZOOL. VOL. 6

[WATSON] PLATE 37

sp. tip

28

sp. tip

29

32.

rotr. m.

•par. nuc.

30.

35.

sp. base

PLATE 38.

Fig. 36. Gyrocotyle rugosa, from Spencer (1889, pi. 1, fig. 1). To show
size and character of lateral folds and posterior rosette and position of
genital pores. X 1.5, based on Spencer's reported magnification of 3
which is possibly an error.

Fig. 37-41. Eggs of G. fimbriata. Figs. 37-40, from decayed or macer-
ated specimens; fig. 41, from living worm. X 500.

acet. — acetabulum.

ant. br. comm. — anterior bridge commissure.

long. n. st. — longitudinal nerve stem.

operc. — operculum.

p. op. — penis opening.

post. br. comm. — posterior bridge commissure.

post. ros. — posterior rosette.

prob. — proboscis.

rec. ov. — receptaculum ovorum.

rec. sem. — receptaculum seminis.

ut. po. — uterine pore.

vag. — vagina.

vag. op. — vaginal opening.

vas def. — vas deferens.

[448]

UNIV. CALIF. PUBL. ZOOL. VOL. 6

[WATSON] PLATE 38

~acet.

ant. br. comm.

post. br. comm

prob.
post. ros.

40.

PLATE 39.

Fig. 42. Gyrocotyle fimbriata, dorsal. Specimen flattened, stained in
borax-carmine; cleared in cedar-oil. X 13.

acet. — acetabulum.

acet. op. — acetabular opening.

ant. br. comm. — anterior bridge commissure.

ant. lat. n. st. — anterior lateral nerve stem.

gen. notch — genital notch.

lat. f.— lateral fold.

long. n. st. — longitudinal nerve stem.

ovar. — ovary.

p. op. — penis opening.

pen. pap. — penis papilla.

post. br. comm. — posterior bridge commissure.

post. gang. Ten. — posterior ganglion knot.

post. ros. — posterior rosette.

prox. r. comm. — proximal ring commissure.

rec. sem. — receptaculum seminis.

ros. n. — "neck" of rosette.

sh. gl. — shell gland.

ut. po. — uterine pore.

vag. — vagina.

vas def. — vas deferens.

vit. d. — vitelline duct.

[4501

UNIV, CALIF. PUBL ZOOL. VOL 6

[WATSON] PLATE 39

acet. op.

gen. notch
vag.

pen. pap

long. n. St.

recjsem.
sh. gl.

.post. br. comni.
'post. gang. kn.
—vox r. comm.

ros. n.
t. ros.

42.

Of THE

UNIVERSITY

OF

PLATE 40.

Fig. 43. G. fimbriata, sagittal section. To show structure of aceta-
bulum and attachment of inner longitudinal muscles. Borax-carmine,
Lyon's blue. X 100.

Fig. 44. Diagram, showing musculature of canal opening.

Fig. 45. Same as fig. 43, showing detail of acetabular sphincter.
X 430.

acet. lum. — acetabular lumen.

acet. op. — acetabular opening.

acet. sphinc. — acetabular sphincter.

ant. br. comm. — anterior bridge commissure.

ant. lat. n. st. — anterior lateral nerve stem.

can. op. — canal opening.

in. long. — inner longitudinal muscle.

in. mer. — inner meridional muscle.

in. trans. — inner transverse muscle.

out. circ. — outer circular muscle.

out. long. — outer longitudinal muscle.

out. mer. — outer meridional muscle.

out. trans. — outer transverse muscle.

rad. fib. — radial fibre.

ros. sphinc. — rosette sphincter.

sphinc. — sphincter.

test. — testis.

[452]

UNIV, CALIF. PUBL. ZOOL, VOL 6

[WATSON] PLATE 40

test.

acet. op.
acpt.sphinc.

out. circ.
rad. fib.
acet. lum.

in. mer.

out. mer
in. trans,
out. long,
out. trans.

ant. br. comm.
in. long.

43.

acet. lum.

sphinc.

ant. lat. n. st.

45.

PLATE 41.

Fig. 46. Gyrocotyle fimbriata. Same specimen as fig. 42. Ventral.
Showing vagina and vaginal opening, vesicula seminalis, penis-papilla
and ejaculatory duct. X 100.

Fig. 47. Egg from fifth coil of uterus. Showing single ovum sur-
rounded by numerous yolk-cells. Delafield's haematoxylin. X 1000.

Fig. 48. Same, tenth coil of uterus. Showing division of ovum and
disintegration of yolk cells. Note disappearance of nuclei of yolk cells.
X 1000.

Fig. 49. Yolk cells in vitellarian follicle. Note reduction of cyto-
plasm and formation of yolk-platelets. X 1850.

Fig. 50. Spermatozoa from receptaculum seminis, showing head, tail
and acrosome. X 1850.

Fig. 51. Ovum in afferent oviduct; note shadow of yolk nucleolus.
X 1500.

Figs. 52, 53, 54. Follicle of ovary. Showing syncytial ova and various
stages in the formation and extrusion of the yolk-nucleolus. X 1500.

cyt. r. — cytoplasmic rim.

ejac. duct — ejaculatory duct.

in. long. — inner longitudinal muscle.

ov. — ovum.

pen. pap. — penis papilla.

sptz. — spermatozoa.

ut. — uterus.

vag. — vagina.

vag. op. — vaginal opening.

vas def. — vas deferens.

yk. c. — yolk cell.

yk. gr. — yolk platelets.

yk. nucleo. — yolk nucleolus.

[454]

UNIV. CALIF. PUBL. ZOOL. VOL. 6

[WATSON] PLATE 41

ov.

ov.

-yk. c.

ov.

yk. nucleo.

52.

k. nucleo.

53.

PLATE 42.

Fig. 55. Gyrocotyle fimbriata. Transverse section, showing cuticula
and cuticular musculature. Iron haematoxylin. X 1500.

Fig. 56. Showing structure of muscle fibres of the inner circular and
inner longitudinal layers. X 1850.

Fig. 57. Showing sagittal (upper) and outer transverse (lower)
muscle cells. X 1850.

Fig. 58. Frontal section of cuticula and cuticular musculature. Cut
through a fold of body-covering, in the plane of the fibres of the cuticu-
lar musculature. X 1850.

Fig. 59. Showing wall of receptaculum seminis, and undifferentiated
nucleated muscle fibres lying in the parenchyma of which it is composed.
X 1850.

cut. — cuticula.

cut. long. — cuticular longitudinal muscle.

cut. trans. — cuticular transverse muscle.

ext. cut. 1. — external cuticular layer.

in. long. fib. — inner longitudinal fibre.

in. trans. — inner transverse muscle.

in. trans, fib. — inner transverse fibre.

nuc. — nucleus.

out. long. — outer longitudinal muscle.

out. trans. — outer transverse muscle.

out. trans, fib. — outer transverse fibre.

par. felt — parenchyma felt.

par. nu. — parenchyma nucleus.

sag. fib. — sagittal fibre.

sub. cut. — subcuticula.

undif. muse. fib. — undifferentiated muscle fibre.

vit. — vitellaria.

[456]

UNIV. CALIF, PUBL. ZOOL, VOL. 6

[WATSON] PLATE 42

text. cut. l

:_ cut.

cut. trans,
cut. long.

sub. cut.
ut. trans,
out. long.

par. nu.

1 in. long. fib.
nuc.

56

55

57

out. trans, fib.

:ut

cut. trans.

cut. long,
sub. cut.

58

par. felt
par. nu.

par. nu

undif. muse,

Of THE

( UNIVERSITY j

IUF<

PLATE 43.

Fig. 60. Gyrocotyle fimbriata. Transverse section through ciliated
excretory canal. Showing circular muscle coat. Iron haematoxylin. X
1000.

Fig. 61. Longitudinal section through ciliated excretory canal (taken
at one side of attachment of cilia). Showing circular muscle coat. Iron
haematoxylin. X 1000.

Figs. 62, 63. Flame cells. X 1500.

Fig. 64. Cells of prostate gland. X 1500.

Fig. 65. Sagittal section through penis papilla and ejaculatory duct.
X 100.

can. — canal.

cil. — cilia.

circ. muse. 1. — circular muscle layer.

ejac. duct — ejaculatory duct.

fl. — flame.

long. muse. 1. — longitudinal muscle layer.

in. long. — inner longitudinal muscle.

pa r. — pa renchyma .

par. nu. — parenchyma nucleus.

pen. pap. — penis papilla.

prost. d. — prostate duct.

prost. gl. — prostate* gland.

rad. muse. f. — radial muscle fibre.

spnl. — spinule.

[458]

UNIV. CALIF. PUBL. ZOOL. VOL. 6

[WATSON] PLATE 43

ejac. duct
circ. muse, .
. m«sc. ^

65

64

PLATE «,

Kg. CS. gyrerefrte fm+n*t»L Ftan pesterair ga»gli««e
S**giio« ««Us of first writer sand * £»jnf£*wi wM «f tlw

Cc

trrai

whi ck & brmiwA to tbe OT^TV n «av«i «C. X

of

^. Di Ji^

: :. -- l

. t*t. <L — efferent riteUix

>•; p»n^li< . :•: . of 1 :•; finl

,,;

UNIV. CALIF. PUBL. ZOOL. VOL. 8

[WATSON] PLATE 44

vit. d.

yk. res.

PLATE 45.

Figs. 70, 75. Gyrocotyle fimbriata. Sagittal section, nearly median.
Diagrammatic. X 13.

Figs. 71, 72, 73. Transverse sections, showing (71) emergence of
ductus seminalis from receptaculum seminis; (73) fusion of the efferent
vitelline duct with the efferent oviduct, which thus becomes the uterus.
X 100.

Fig. 74. Sagittal section through dorsal half of neck of funnel. Show-
ing contrast between cuticula of funnel cavity and of the surface of the
body. X 1000.

acet. — acetabulum.

cut. — cuticula.

duct. sem. — ductus seminalis.

ef. ovd. — efferent oviduct.

ef. vit. d. — efferent vitelline duct.

ejac. duct — ejaeulatory duct.

ex. can. — excretory canal.

in. long. — inner longitudinal.

lat. long. conn. — lateral longitudinal connective.

long. n. st. — longitudinal nerve stem.

mus. bd. — muscle bundle.

oot. — ootype.

ovar.; ov. — ovary.

post. ros. — posterior rosette.

pros. gl. — prostate gland.

rec. ovor. — receptaculum ovorum.

rec. sem. — receptaculum seminis.

subcutic. 1. — subcuticular layer.

sh. gl. — shell gland.

test. — testis.

ut. — uterus.

ut. c. — uterine coil.

vag. — vagina.

vit. — vitellaria.

vit. d. — vitelline duct.

[462]

UNIV, CALIF. PUBL ZOOL. VOL. 8

[WATSON] PLATE 45

post. ros.

rec. sem.

miduct. sem.

ef. vit. d.

cut.

subcutic. I
ex. can.

in. long.

.bd.

mas

rec. sem.
— oot.
ef. vit. a.

mzz^^it.

Lei. vit, d.
sh. gl.
mus. bd.
in. long.

74.

75.

'

PLATE 46.

Fig. 76. Gyrocotyle fimbriata. Median sagittal section, posterior ex-
tremity. Showing funnel, canal, funnel-opening, canal-opening and sec-
tions of the ring commissures of the nervous system of the posterior
extremity. X 100.

Fig. 77. Same, to left of median. Reconstruction of five sections,
showing lateral view of nervous system. Faint dotted lines indicate
course of nerves as shown in the neighboring sections; heavy dotted lines
indicate position of funnel with reference to the central nervous system.
X 100.

can. op. — canal opening.

ex. can. — excretory canal.

lat. anast. — lateral anastomosis.

lat. long. conn. — lateral longitudinal connective.

post. br. comm. — posterior bridge commissure.

post. gang. Jen. — posterior ganglion knot.

prox. r. comm. — proximal ring commissure.

ros. n. st. — rosette nerve stem.

ros. r. comm. — rosette ring commissure.

ros. sphinc. — rosette sphincter.

</'• — tope.

[464]

UNIV. CALIF. PUBL. ZOOL. VOL. 8

[WATSON] PLATE 46

ros. sphinc.
ros. n. st.

r. comm.

can. op.
t. br.comm.

ex. can.

ros. n. st.
ros. r. comm.

lat. lone, conn

\\\

\A\> lat. anast.

lat. long, conn

>,'V \v^ prox. r. comm

post. gang. kn.

PLATE 47.

Fig. 78. Nervous system of Moniezia expansa, after Tower (1900).
Fig. 79. Nervous system of Triclad; after Lang (1881).
Fig. 80. G. nigrosetosa, Haswell (1902, pi. VII, fig. 1).

Fig. 81. Same, egg, showing operculum. Haswell (1902, pi. VII,
fig. 7).

ant. br. comm. — anterior bridge commissure.

ant. gang. kn. — anterior ganglion knot.

ant. r. comm. — anterior ring commissure.

dist. r. comm. — distal ring commissure.

lat. long. conn. — lateral longitudinal connective.

long. n. st. — longitudinal nerve stem.

med. long. conn. — median longitudinal connective.

post. br. comm. — posterior bridge commissure.

post. gang. kn. — posterior ganglion knot.

prox. r. comm. — proximal ring commissure.

[460]

UNIV. CALIF. PUBL. ZOOL. VOL, 8

[WATSON] PLATE 47

Ions, n, st.

ant. sans. kn.
- ant r. comm.

post. br. comm.
post. sans. kn.

prox. r. comm.

med. Ions. conn.

at. Ions. conn.

'dist. r. comm.

ant. br. comm.

81.

Ions. n. st.

prox. r. comm.

79.

Of THE

UNIVERSITY

OF

PLATE 48.

Figs. 82, 83. Photographs of living specimens of G. fimbriata and
G. urna (var.) All in state of contraction. About natural size.

Figs. 84, '85. G. urna, from Wagener (1852, pi. 14, figs. 1, 2).

[408]

UNIV. CALIF. PUBL. ZOOL. VOL.

[WATSON] PLATE 48

82

83

85

84

UNIVERSITY OF CALIFORNIA PUBLICATIONS— (Continued)

Vol. 8. 1. Some Observations on the Nervous System of Copepoda, by O. O.

Esterly. Pp. 1-12, plates 1-2. January, 1906 .25

2. (IX)* Ostracoda of the San Diego Region. 1. Halocypridae, by

Chancey Juday. Pp. 13-38, plates 3-7. April, X906 .30

3. (X) The California Shore Anemone, Bunodactis xanthogrammica, by

Harry Beal Torrey. Pp. 41-46, plate 8, April, 1906.

4. (XI) Sexual Dimorphism in Aglaophenia, by Harry Beal Torrey and

Ann Martin. Pp. 47-52, 9 text-figures. April, 1906.

Nos. 3 and 4 in one cover. .15

6. (XII) New Copepod Fauna from the San Diego Region, by Calvin Olin

Esterly. Pp. 53-92, plates 9-14. December, 1906 .36

6. (XII) Dinoflagellata of the San Diego Region, EL On Triposolenia, a

New Genus of the Dinophysidae, by Charles Atwood Kofoid. Pp.
93-116, plates 15-17.

7. A Discussion of the Species Characters in Triposolenia. I. The Nature

of Species Characters. II. The Adaptive Significance of Species
Characters. III. The Coincident Distribution of Related Species.
By Charles Atwood Kofoid. Pp. 117-126.

8. On the Significance of the Asymmetry in Triposolenia, by Charles

Atwood Kofoid. Pp. 127-133.

Nos. 6, 7, and 8 in one cover. December, 1906 .36

9. (XIV) Ostracoda of the San Diego Region. II. Littoral Forms, by

Chancey Juday. Pp. 135-156, plates 18-20.

10. (XV) Cladocera of the San Diego Region, by Chancey Juday. Pp.

157-158, 1 text figure.

Nos. 9 and 10 in one cover. January, 1907.. . .26

11. (XVI) The Marine Fishes of Southern California, by Edwin Chapin

Starks and Earl Leonard Morris. Pp. 159-251, plate 21. March, 1907. .75

12. Biological Studies on Corymorpha. II. The Development of C. palma

from the Egg. By Harry Beal Torrey. Pp. 253-298, 33 text figures.
June, 1907 ~ .50

13. (XVII) Dinoflagellata of the San Diego Region, in. Descriptions of

New Species. By Charles Atwood Kofoid. Pp. 299-340, plates 22-23.
April, 1907 - ._ .50

14. The Structure and Movements of Condylostoma patens, by John F.

Bovard. Pp. 343-368, 21 text figures. September, 1907 36

Index, pp. 369-383.

Vol. 4. 1. The Ascidians Collected by the United States Fisheries Bureau steamer
Albatross on the Coast of California during the Summer of 1904, by
William Emerson Ritter. Pp. 1-52, plates 1-3. October, 1907 .50

2. (XVIII) Behavior of the Starfish Asterias forreri de Lcrriol, by H. S.

Jennings. Pp. 53-185, 19 text figures. November, 1907 1.00

3. (XIX) The Early LifeJEEistory of Dolichoglossus pusillus Ritter, by B.

M. Davis. Pp. 187-226, plates 4-8. March, 1908 .50

4. Notes on two Amphipods of the Genus Corophium from the Pacific

Coast, by J. Chester Bradley. Pp. 227-252, plates 9-13. April, 1908. .30

5. (XX) The Incrusting Chilostomatous Bryozoa of the Western Coast of

North America, by Alice Robertson. Pp. 253-344, plates 14-24, May,
1908 . .. 1.00

6. (XXI) On Exuviation, Autotomy, and Regeneration in Cerattum, by

Charles Atwood Kofoid. Pp. 345-386, with text figures.

7. (XXII) Notes on some Obscure Species of Ceratium, by Charles Atwood

Kofoid. Pp. 387-393.

Nos. 6 and 7 in one cover. April, 1908 .50

Index, pp. 395-400.
Vol. 5. 1. The Biota of the San Bernardino Mountains, by Joseph Grinnell. Pp.

1-170, plates 1-24. December, 1908 _..... 2.00

2. Birds and Mammals of the 1907 Alexander Expedition to Southeastern

Alaska. Pp. 171-264, pis. 25-26, figs. 1-4. February, 1909 .75

3. Three New Song Sparrows from California, by Joseph Grinnell. Pp.

265-269. April 9, 1909 05

4. A New Harvest Mouse from Petaluma, California, by Joseph Dixon.

Pp. 271-273. August 14, 1909 05

5. A New Cowbird of the Genus Molothrus, with a note on the Probable

Genetic Relationships of the North American Forms, by Joseph
Grinnell. Pp. 275-281, 1 text figure. December, 1909 .05

6. Two New Rodents from Nevada, by Walter P. Taylor. Pp. 283-302,

plates 27-29.

7. A Northern Coast Form of the Calif ornia Gray Fox, by Joseph Dixon.

Pp. 303-305.

Nos. 6 and 7 in one cover. February, 1910 „•. ,20

* Roman numbers indicate sequence of the Contributions from the Laboratory of the
Marine Biological Association of San Diego.

UNIVERSITY OF CALIFORNIA PUBLICATIONS-(CONTINUED)

8. Two Heretofore Unnamed Wrens of the Genus Thryomanes, by Joseph

Grinnell. Pp. 307-309.

9. The Savannah Sparrow of the Great Basin, by Joseph Grinnell. Pp.

311-316.

10. A Second Record of the Spotted Bat (Euderma maculatum) for Cali-

fornia, by Joseph Grinnell. Pp. 317-320, plate 30.

Nos. 8, 9, and 10 in one cover. February, 1910 15

11. Mammals of the 1908 Alexander Alaska Expedition, with Descriptions

of the Localities Visited and Notes on the Flora of the Prince Wil-
liam Sound Region, by Edmund Heller. Pp. 321-360, plates 31-32.

12. Birds of the 1908 Alexander Alaska Expedition, with a Note on the

Avifauna! Relationships of the Prince William Sound District, by
Joseph Grinnell. Pp. 361-428, plates 33-34, 9 text-figures.

Nos. 11 and 12 in one cover. March, 1910 $1.00

Index, pp. 429-440.

1. (XXm) On the Weight of Developing Eggs. Part I, The Possible

Significance of Such Investigations, by William E. Bitter; Part II,
Practicability of the Determinations, by Samuel E. Bailey. Pp. 1-10.
October, 1908 10

2. (XXIV) The Leptomedusae of the San Diego Region, by Harry Beal

Torrey. Pp. 11-31, with text figures. February, '1909 , ... .20

3. (XXV) The Ophiurans of the San Diego Region, by J. F. McClen-

don. Pp. 33-64, plates 1-6. July, 1909 30

4. (XXVI) Halocynthia johnsoni n. sp.: A comprehensive inquiry as to

the extent of law and order that prevails in a single animal species,

by Wm. E. Ritter. Pp. 65-114, plates 7-14. November, 1909 50

5. (XXVII) Three Species of Cerianthus from Southern California, by

H. B. Torrey and F. L. Kleeberger. Pp. 115-125, 4 text-figures.
December, 1909 10

6. The Life History of Trypanosoma dimorplion, Dutton & Todd, by

Edward Hindle. Pp. 127-144, plates 15-17, 1 text-figure. December,
1909 50

7. (XXVIII) A Quantitative Study of the Development of the Salpa

Chain in Salpa fmiformis-runcinata, by Myrtle Elizabeth Johnson.

Pp. 145-176. March, 1910 35

8. A Revision of the Genus Ceratocorys, Based on Skeletal Morphology,

by Charles Atwood Kofoid. Pp. 177-187. May, 1910 10

9. (XXIX) Preliminary Report on the Hydrographic Work Carried on by

the Marine Biological Station of San Diego, by George F. McEwen.

Pp. 189-204; text-figure and map. May, 1910 15

10. (XXX) Biological Studies on Corymorpha. III. Regeneration of Hy-

dranth and Holdfast, by Harry Beal Torrey. Pp. 205-221; 16 text-
figures.

11. (XXXI) Note on Geotropism in Corymorpha, by Harry Beal Torrey.

Pp. 223-224; 1 text-figure.

Nos. 10 and 11 in one cover. August, 1910 20

12. The Cyclostomatous Bryozoa of the West Coast of North America, by

Alice Robertson. Pp. 225-284; plates 18-25. December, 1910 60

13. Significance of White Markings in Birds of the Order Passeriformes,

by Henry Chester Tracy. Pp. 285-312. December, 1910 25

14. (XXXIII) Third Report on the Copepoda of the San Diego Region, by

Calvin Olin Esterly. Pp. 313-352; plates 26-32. February, 1911 40

15. The Genus Gyrnroiyle, and Its Significance for Problems of Cestode

Structure and Phylogeny, by Edna Earl Watson. Pp. 353-468; plates
33-48. June, 1911 1.00

1. Two New Owls from Arizona, with Description of the Juvenal Plum-

age of Strix occidcntaUs occidcntaUs (Xantus), by Harry S. Swarth.

Pp. 1-8. May, 1910 10

2. Birds and Mammals of the 1909 Alexander Alaska Expedition, by

Harry S. Swarth. Pp. 9-172; plates 1-6; 3 text-figures. January, 1911. 1.50

3. An Apparent Hybrid in the Genus Dendroica, by Walter P. Taylor.

Pp. 173-177. February, 1911 - 05

4. The Linnet of the Hawaiian Islands: a Problem in Speciation, by

Joseph Grinnell. Pp. 179-195. February, 1911 _. .15

5. The Modesto Song Sparrow, by Joseph Grinnell. Pp. 197-199. Feb-

ruary, 1911 05

6. Two New Species of Marmots from Northwestern America, by H. S.

Swarth. Pp. 201-204. February, 1911 _ -. 05

1. The Vertical Distribution of Eucalanus elongatus in the San Diego

Region during 1909, by Calvin O. Esterly. Pp. 1-7. May, 1911 10

NON-CIRCULATING BOOK

CD

218265

10m-l,"

8RARY