HARVARD UNIVERSITY

LIBRARY

OF THE

Museum of Comparative Zoology

BULLETIN

MUSEUM OF COMPARATIVE ZOOLOGY

HARVARD COLLEGE, IN CAMBRIDGE.

VOL. III.
Nos. 1-16.

CAMBRIDGE, MASS., U. S. A.
1871-1876.

Reprinted with the permission of the original publisher

KRAUS REPRINT CORPORATION

New York

1967

~>

Printed in U.S.A.

CONTENTS.

Page
No. 1. — Report on the Brachiopoda obtained by the United States Coast
Survey Expedition in charge of L. F. de Poortales. With a Revision
of the Craniidaa and Discinidae. By W. H. Dall. (2 Plates) . . 1

No. 2. — Application of Photography to Illustrations of Natural History.
With two Figures printed by the Albert and Woodbury Processes. By
Alexander Agassiz .47

No. 3. — A Letter concerning Deep-Sea Dredging, addressed to Professor
Benjamin Peirce, Superintendent United States Coast Survey. By
Louis Agassiz 49

No. 4. — Preliminary Notice of a few Echini. By Alexander Agassiz . 55

No. 5. — Fossil Cephalopods of the Museum of Comparative Zoology. —

Embryology by Alpheus Hyatt. (4 Plates) 59

Embryology, 66 ; Umbilicus, 76 ; Whorls, 76 ; Septa, 81 ; The Shell, 103.

No. 6. — Notes of an Ornithological Reconnaissance of Portions of Kansas,
Colorado, Wyoming, and Utah. By J. A. Allen . . . .113

I. List of Birds observed at Leavenworth, Kansas, from May 2 to

May 11, and at Topeka, Kansas, from May 11 to May 24, 1871.
With Annotations 122

II. List of Birds observed in the Vicinity of Fort Hays, Kansas, from

May 26 to July 3, 1871. With Annotations .... 131

III. List of Birds observed in Northwestern Kansas, December 25,

1871, to January 12, 1872. With Annotations . . . .142

IV. List of Birds observed at Cheyenne, Wyoming Territory, from

August 16 to August 28, 1871. With Annotations . . .144
V. List of Birds observed at the Eastern Base of the Rocky Mountains
in Colorado Territory, between Colorado City and Denver, in
July and August, 1871. With Annotations . . . .146
VI. List of Birds observed in South Park, Park County, Colorado Ter-
ritory, in July, 1871. With Annotations 153

VII. List of Birds observed in the Vicinity of Mount Lincoln, Park
County, Colorado, from July 19 to July 26, 1871. With Anno-
tations 159

iv CONTENTS.

VIII. List of Birds collected in the Vicinity of Ogden, Utah Territory,

"from September 1 to October 8, 1871. With Annotations . . 164
IX. Summary List of Birds observed in Kansas, Colorado, Wyoming,

and Utah, in 1871 173

No. 7. — Interim Report of the Hydroids collected by L. F. de Pourtales,
during the^Gulf Stream Exploration of the United States Survey. By
George J. Allman 185

No. 8. — The Echini collected in the Hassler Expedition. By Alexander
Agassiz . . . 187

No. 9. — Catalogue of the Terrestrial Air- breathing Mollusks of North Amer-
ica, with Notes on their Geographical Range. By W. G. Binney. (1 Plate) 191

No. 10. — Ophiuridae and Astrophytidse, new and old. By Theodore

Lyman. (7 Plates) 221

On the Species of Ophiothrix from the Waters of Western Europe and

of the Mediterranean .... 240

Catalogue of the Ophiuridae and Astrophytidse collected by Prof. C.
Semper, and now belonging to the Museum of Comparative Zoology 252

Homologies of Chewing Apparatus in Ophiuridte 254

Explanation of Plates 260

No. 11. — Exploration of Lake Titicaca. By Alexander Agassiz and
S. W. Garman. (1 Plate) 274

I. Fishes and Reptiles. By S. W. Garman 274

Fishes, 274 ; Batrachians, 276 ; Reptiles, 278.

No. 12. —Exploration of Lake Titicaca. By Alexander Agassiz and
S. W. Garman 279

II. Notice of the Palaeozoic Fossils. By Orville A. Derby, M. S.,

with Notes by Alexander Agassiz 279

No. 13. — Recent Corals from Tilibiche, Peru. By Alexander Agassiz
and L. F. de Pourtales. (1 Plate) 287

No. 14. — The Development of Salpa. By Wm. K. Brooks, Ph. D. (34

cuts) 291

Embryology of Solitary Salpa (Female) 306

Development of the Salpa Chain 326

Summary and General Conclusions ....... 337

No. 15. — Exploration of Lake Titicaca. By Alexander Agassiz and
S. W. Garman 350

III. List of Mammals and Birds. By J. A. Allen, with Field Notes

by Mr. Garman. Mammals, 350 ; Birds 353

No. 16. — Exploration of Lake Titicaca. By Alexander Agassiz and
S. W. Garman 361

IV. Crustacea. By Walter Faxon. (With 37 cuts) ... 361

No. 1. — Report on the Brachiopoda obtained by the United
States Coast Survey Expedition, in Charge of L. F. de Pour-
tales, with a Revision of the Craniidye and Discinid.e by
W. H. Dall.

(Communicated by Professor Benj. Peikce, Superintendent U. S. Coast

Survey.)

In the preparation of this paper I have been indebted to the Smith-
sonian Institution, under the direction of Professor Joseph Henry, for
the use of their library and collection of recent brachiopods ; to J. Gwyn
Jeffreys, Esq., F. R. S., for kindly lending specimens of the brachiopods
obtained by the English Deep-Sea Dredging Expedition, for comparison ;
and to Thomas Davidson, Esq., F. G. S., for many favors.

The animals which compose this class are of peculiar interest to the
naturalist and geologist, as being represented in rocks of very early
ages, and continuously through the various formations up to the present
period. Their position in the natural system of classification being stil1
a matter of discussion, all facts bearing on their anatomy and embryology
are of the highest interest. I have endeavored, therefore, instead of
compiling a mere list of species and descriptions, to enter as thoroughly
into the details of anatomy as the means at hand would allow, and have
considered the present a fit opportunity for rectifying the synonymy
of some groups which, from the confusion in which they have been
involved, have long been avoided by naturalists as stumbling-blocks.

It is hardly necessary to add, that I am indebted for the opportunity
of doing this work to the kindness of Professor Agassiz, who placed
the materials in my hands for examination, with the kind concurrence
of M. de Pourtales and Dr. William Stimpson

Class BRACHIOPODA Cuvier.

Animals provided with two shelly valves, each of which, normally, is
bilaterally symmetrical. Valves united by three or more pairs of muscles,
which, with all the other soft parts (except occasionally the intestine) are
arranged in bilateral symmetry with relation to the longitudinal axis of
the valves, respectively. Organs consisting essentially of a mantle com-
VOL. III. 1

BULLETIN OF THE

posed of two lobes, which have their anterior edges always disconnected,
and which correspond to the valves of the shell ; a disk of membrane,
variously modified, with its edges fringed with a series of tubular brachia ;
a mouth situated within the posterior edge of this disk ; a stomach with a
more or less differentiated and anteriorly recurved intestine ; a circulatory
system more or less contained within a series of vessels and an atrial
system of sinuses or lacunes; with a unilocular heart and usually one or
two pairs of accessory pulsatile vesicles ; with the genitalia usually sus-
pended in the vascular sinuses and expelling their products through one or
two pairs of oviducts opening externally ; nervous ganglia in a ring sur-
rounding the oesophagus ; respiring oxygen by absorption through contact
of the sea water with the surface of the tissues of the mantle and brachia;
dioecious, and exclusively marine.

This diagnosis comprises all the characters which, after careful consid-
eration, I find to be common to all the members of the class. There are
other characters which are more or less characteristic of the more familiar
forms of recent brachiopods, but which are not characteristic of the group
as a whole. Thus, many of the recent forms are attached by a pedicel,
while others in the same family are attached by the substance of the
valves, and others of nearly allied groups are without an attachment of
any kind. The shells of many brachiopods are perforated by minute tubuli
lined by ca^cal prolongations of the outer laminae of the mantle lobes, while
others in the same family, and perhaps, in some cases, in the same genus,
are without these perforations. The mantle edge of many genera is jiro-
vided with a more or less closely set border of seta?, while others in the
same family are entirely without seta;, and even the same individual, in
the earlier stages of its growth (but after the other organs are nearly
complete) may be devoid of them. In some brachiopods the setae are
stated to be movable, while in others no muscles exist by which they
might be moved. In some the blood is colored and in others colorless.
The chemical composition of the shell differs in different genera, though
in the gi-eat majority it is principally composed of carbonate of lime. The
embryonic forms differ widely among themselves, some being segmented
and possessing eye-spots like the fry of Pneumodermon and Dentalium,
while others are unsegmented. In one genus the pedicel is developed out
of the middle of the dorsal area of the embryo, so that the valves both
bear a dorsal relation to the animal, while in others it would appear to
extend from one extremity of the embryo, when the valves would bear a
dorsal and ventral relation respectively.

It is evident that characters such as these, which are few of family, and
none of ordinal value, can have no important bearing upon the classifica-
tion of the group and its systematic position as a whole.

MUSEUM OF COMPARATIVE ZOOLOGY. 3

Order ARTHROPOMATA Owen.

** Syn. = Arthropomata Owen, Enc. Brit. Ed., VIII, XV, Art. Mollusca, p. 336,
1858.
= Brachiopodes, valves articule'es Deshayes, An. s. Vert. Ed , II, VII, p.

309, 1836.
= PaUiobranchiata, testa cardine inslructa Van der Hceven, Handb. der

Zool., p. 092, 1850 *
= Apygia Bronx, Klass. Ordn. Thierr., Ill, 1 Abth. p. 301, 1862.
= Articulata Huxley, Lect. Class. 1864. Intr. Class. Anim., p. 116, 1869.
>- -< Ancylobrachia ■+■ Cryptobrachia + Sclerobrachia + Sarcicobrachia Gray,

Ann. Mag. Nat. Hist., II, pp. 435-438, 1848.
=~ -= Brachiopodes brachide's + B. cirride's D'Orbigny, Cours Ele'm. Pal., II,
p. 82, 1849.
>- Pedunculata Latreille, Fam. Nat. Reg. An., p. 196, 1825.

Characters. Intestine ending in a closed sac. Lobes of the mantle
united posteriorly. Valves articulated by teeth and sockets.

Family TEREBRATULIDiE.

Tcrebratulidce Dall, Am. Journ. Conch., VI, p. 101, 1870.
( Thecid'ddoz exclus. )

Subfamily TEREBRATULINiE Dall.
Terebratulinai Dall, 1. c. p. 101, 1870.

Genus TEREBRATULA Auct. ex Llhwyd.

Terebratula Lliiwyd, Litli. Britt. Ichn., 1699. Lam. Prodrome, 1799. Dall

Am. Jour. Conch., VI, p. 101, 1870.
Type T. perovalis Sowekby, Lamarck. T. maxillata, Sowerby.

Terebratula cubensis Pourtales.

Terebratula cubensis Pourtales, Bull. Mus. Comp. Zool., I, No. 7, p. 109, 1867.

Dall, Am. Jour. Conch., Vol. VI, pp. 105, 166, 1870.
Terebratula vitrea, var. Davidson, Mon. Ital. Tert. Brach., I, p. 9, 1870 ; also

Jeffreys, in litt. (not of Born).
Florida reefs, May, 1868, in 100 - 200 fathoms, rarer toward the east end of the

reefs. Coast of Cuba near Havana, in 270 fathoms. Pourtales, U. S.

Coast Survey.
In upholding the specific distinctness of this species, which is withou;
doubt closely allied to T. vitrea, I regret that I am obliged to differ from

* I have adopted in the synonymy the very excellent system of notation proposed by
Bronn and Strickland. This, or some similar system, is absolutely necessary for the
comprehension of intricate synonymy. It is a matter of surprise that it has not been
more generally adopted and made use of. The single asterisks denote references which
I have not been able to verify by actual examination in person.

I BULLETIN OF THE

the distinguished naturalists whose names are quoted above. It must bo
admitted that honest differences of opinion may exist in regard to the
specific limits of almost any species of animal ; in this, as in other cases,
. I can only give utterance to my own personal opinion, based upon the
material at my command. I have, in another publication, stated that I
have found constant, though not extreme, differences between this species
and T. vilrea, and, as no transition from the one to the other has yet been
shown, I feel justified in considering these differences as of specific value.
They were partly pointed out by M. de Pourtales in his original descrip-
tion ; and after a careful examination of many hundred specimens, and a
critical comparison of them with a large series of T. vilrea, I have not been
able to detect any inconstancy in the form of the loop in each species.
While the other characters arc more variable, yet even those show no
more approach to each other than may usually be observed in two closely
allied species.

The following comparative diagnoses will serve to point out these differ-
ences : —

T. cubensis has the. margin of the valves laterally flexuous, varying to
some extent in degree of flexuosity, with age. This diagnosis, however,
refers to adult shells, in which a certain amount of flexuosity is always
present. The convexity being in a homial direction, the margin of the
hamuli valve is excavated on each side, giving the valve a subquadrangular
aspect. The valves are usually rather inflated, giving the shell a tumid
aspect. The hinge teeth are stout and thick, the deltidium moderate ;
foramen rather large. The shell is widest near the anterior margin of the
valves, the cardinal border is strongly arched ; the cardinal process is
stout, blunt, broad, rounded and much recurved. The cardinal plate is
divided ; no shelly matter extending between the apophysal ridges. The
shelly plates on each side of these ridges, extending to the dental ridges,
are deeply concave, with the anterior border somewhat produced ami
rounded. The crura are short and blunt. The anterior part of the loop
is characteristic and peculiar. It is strongly squarely convex in the mid-
dle, and a deep narrow gutter extends on each side of this convexity and
is produced at each side into a point. Between these points and the
median convexity on each side is a deep slit or fissure. The anterior end
of the convexity is much produced, blunt, and square. It terminates be-
hind in a slight sinus or indentation. The apex of the haemal valve is
much incurved.

T.vitrea has the lateral margins of the valves almost rectilinear, if there
be any flexuosity the direction of the convexity is neural. Hence the
outline of the hannal valve is rounded ovate. The valves are more or less
compressed, and there are frequently indications of a broad median ridge,

MUSEUM OF COMPARATIVE ZOOLOGY. 5

flattened and bounded by two obscure carina?, which is never present in
T. cubensis. The hinge teeth are slender and delicate, the deltidium much
narrower, and the foramen usually smaller than in T. cubensis. The shell
is widest behind the middle of the valves, giving a slight coffin-shaped
aspect to the valves ; the cardinal border is nearly straight, the cardinal
process is slender, produced, and square at the end. The cardinal plate is
divided, as in the last, the shelly plates on each side are nearly flat and
anteriorly emarginated. The crura are sharply pointed. The anterior
portion of the loop is but slightly convex in the middle, the gutters on
each side are broad and shallow, there are no lateral slits, and instead
of a median prolongation there is a deep, broad emargination or concavity
in the anterior edge of this part of this loop. Instead of a sinus behind,
there is a sharp point, which is, however, often broken off* in dry speci-
mens. The apex of the haemal valve is not incurved. In fact, the shell
of this species is flat where T. cubensis is concave, emarginated where
cubensis is produced, entire where cubensis is fissured, broadest where
cubensis is narrowest, narrowest where cubensis is widest, subcarinate
where cubensis is smoothly rounded, etc., etc. In both obscure fine radiat-
ing lines may be often seen. That these characters are constant through-
out hundreds of specimens of T. cubensis I can testify from actual exami-
nation. I have found those here recorded of T. vitrea constant in some
twenty specimens, and in all the figures of this species extant. Hence I
have not the slightest hesitation in considering them as distinct species.
The differences of the loops, solely, would distinguish them anywhere, and
no more satisfactory specific character could be advanced.

In its general anatomy this species presents some similarity to Wahllu imia
Jiorblana. I shall r<"=erve a more explicit account of the structure for that
species, and only mention here the anatomical points which appear more
striking, and which are more or less peculiar to this species. Most of the.
soft parts are of a translucent yellowish white color. The mantle is of
stouter consistency than in W. fioridana, and may often be removed from
the shell with but little injury if care be exercised. The muscles are.
similar in disposition to those of the other members of the Terebratulidee,
and present no new features. The peduncle is solid, cup-shaped at its
extremity, and has the edge produced in cylindrical horny rootlets, which
are attached to foreign bodies. The regular arrangement in layers of the
muscles and corium, as well as the axial tube of the peduncle, found in
Lingula, is less evident or absent in these forms. In this species the
peduncle is very short and stout, broadly cordiform at its inner extremity
when enveloped by its various tunics.

The brachia are arranged as in T. vitrea, as figured by Woodward; the
central coil makes about four turns. The cirrhi are very short behind the

G BULLETIN OF THE

mouth, in front of the supra-oesophageal body. A striking feature in its
anatomy, which I believe has not yet been noted in any publication on
Brachiopods, is the absence of that great series of sinuses in the ante-
rior part of mantle, 'which was termed by Hancock the "great pallial
sinuses." So extraordinary did this appear to me, that I could not be-
lieve, at first, that I was not deceived by the translucency of the mem-
branes, and it was only after an examination of many specimens that I be-
came convinced that they did not exist in this species. There is in the
free lobes of the mantle an extensive and extremely close and fine network
of minute channels ; or perhaps it might be said that the whole of the
mantle lobes form one great lacune, the upper and lower walls of which are
held apart by a profuse number of pillars of tissue, which appear like dark
spots under the microscope, and which are situated so close together that
the spaces about them are reduced to minute channels. This system occu-
pies the anterior lobes of the mantle, which in some species also contain large
branching sinuses, here absent. On the outer surface of the perivisceral
chamber, above and below, on each side of the attachments of the principal
muscles, a small system of sinuses exists, and here are situated the genitalia
which necessarily assume a reticulated aspect quite different from the loops
and branches seen in Waldheimia and Terebratella. In the inner lining of
the mantle are scattered, everywhere, delicate branching spiculae, looking
like briers more than like deer-horns, and, while more or less interlocked,
and here and there stout and thick, are still much more delicate and slen-
der than those of Tcrebratidina caput-serpeiitis and Megerlia truncata, and
do not often exhibit a stellar arrangement. They are much more numer-
ous in some individuals than in others, and when present in abundance are
found in almost every part of the epithelium, even to the brachial cirrhi,
where the spicules are slender and not branched. They are especially
numerous over the perivisceral chamber and in the supra-cesophageal tissue.
The oral aperture presents no special peculiarities. The oesophagus is
wide and funnel-shaped, narrowest at its junction with the stomach, which
it enters at an acute angle. The stomach is small and oval, tapering to-
ward the intestine, which is nearly twice as long as the oesophagus. In the
stomach was a dark mass of calcareous granules, fragments of Foraminifera,
etc., filling it quite full ; among the debris was, in one specimen, the re-
mains of a small red crustacean with a large carapax and ( ? six) legs,
somewhat resembling a young Limulus, but much smaller. Other unrecog-
nizable crustacean fragments were noticed in other cases. Notwith-
standing the crowded state of the stomach, the intestine was always
empty. Its caecal end was somewhat blunt and rounded, and several fold-
like thickenings of the mesentery recalled Hancock's figures of the termi-
nation of the intestine in W. cranium. The pointed lower ends of the

MUSEUM OF COMPARATIVE ZOOLOGY. 7

plicated openings of the oviducts were free from the body wall and attached
to the parietal band on the anterior edge of the intestine a little way above
the heart : first, by a tendinous process of considerable toughness ; and,
secondly, by the end of the plicated membrane itself. Between the two
attachments a small foramen appeared, which seemed to be normal, but
may possibly have been due to a lesion of the tissues. The heart in most
specimens was pyriform and of moderately large size ; in one or two it
appeared of an hourglass shape, probably due to contraction. The dis-
position of the vessels was similar to that described by Hancock in Tere-
bratulina caput-serpentis, as far as I was able to determine.

A very careful search was made for accessory pulsatile vesicles, but
none were discovered, though I do not feel positive that they may not ex-
ist. The vessels which supply the genitalia are much looped and reticu-
lated. The genitalia, as before mentioned, are situated in a reticulated
series of sinuses, on the surfaces of the sides of the perivisceral tissues ;
this series does not pass in front of the muscular attachments, as far as I
have been able to discover by repeated and careful examinations. It
would seem as if this portion represented the pallial sinuses of Hancock,
which exist in other genera, but which, in this species at least, seem to be
suppressed anteriorly. The lacunes of the anterior portion of the mantle-
lobe are homologous with the inner and outer pallial lacunes of Hancock.
The genitalia agree in general features in all the specimens examined.
They are of a yellowish color, and all appeared destitute of the reddish
granular substance noticed in other species. On the other hand, a similar
accumulation of reddish-yellow granules appears in the glandular funnels
of the oviducts, which open by an oval and rather large aperture on each
side, behind, and a short distance above the mouth.

Above and behind the mouth, and directly in front of the anterior
occlusor (retractor) muscles, the external tissues of the perivisceral mem-
brane are thickened, or a mass of cellular tissue is interposed between the
laminae of the membrane. This causes a protuberance almost exactly
resembling in shape and appearance a human nose. Below the inferior
and most prominent portion of this protuberance is a deep groove or
incised line, under which is another protuberance, short, wide, and trans-
verse, shaped like a roll of parchment. For want of a better name I pro-
pose to call the lower protuberance the supra-ccsophageal body; the
fissure, the inter-corporeal groove; and the upper protuberance, the nasi-
form body.

These organs do not exist in any of the species of Brachiopods (except
T. cubetisis) with which I am acquainted. Nothing of the kind is to be
seen in W. floridana, Terebratella coreanica, T. caurina, Waldheimia
Jlavescens, etc., etc. I am unable to say whether it occurs in T. vilrea,

BULLETIN OF THE

but it is very likely that it does. What may be the office of these forma-
tions I cannot imagine, unless it may be to protect the muscles. In
many specimens the nasiform body was crammed with spiculae in one
heterogeneous mass, forming an excellent shield for the muscles. The
brachial cirrhi before these prominences were very much shorter than the
others. I am not aware that these peculiar features have been noticed in
any publication on this group. The hepatic digitations enter the stomach
by two ducts on each side as in W. Jloridana, but are longer and more
slender than in that species. The seta? are longer and more closely set in
front than at the sides ; they rarely are double in the same follicle ; and in
no case were more than two so noticed. They seemed to be almost uni-
formly broken off just beyond the edge of the mantle, but in those
which remained unbroken no transverse markings were seen. A few dark
pigment granules were noticed around the bases of the follicles, and a line
of similar granules was seen between each two seta?, passing round the
bases- of the follicles and joining the next line, and so on continuously.
The eircumpallial muscle was narrow and slender.

No peculiarities of note were observed in the shell structure. The per-
forations appeared to be slightly further apart than in T. vitrea, but the
difference was not much greater than that which may be observed in the
shells of different individuals of the same species.

The external layer, mentioned by Hancock as occurring at the edges of
the shells of other species, was well marked in perfect examples, and
extended over a large part of the shells.

Attached to a piece of rock, dredged off the Samboes on the Florida
coast, was a minute polished hyaline shell 4-100 of an inch in length,
which, from its general appearance and the locality in which it was found,
I believe to be the young of T. cubensis. That species is abundant in that
locality, and the only other known species to which it might be referred
is TT. Jloridana. The latter, even in very small specimens, has quite
a different aspect.

The shell in question was ovate,*with the beak of the neural valve quite
prominent, and with a small but sharply defined area. There was no
deltidium, and the apex of the haemal valve Avas somewhat prominent,
recalling that of an embryonic (?) Brachiopod described by Mr. Jeffreys,
under the name of Terebratula capsula. The punctures were very small
and widely separated, arranged in quincunx order. The ends of the
prisms of which the shelly matter is composed, by impinging upon the
surface, gave it a beautifully reticulated or lacelike aspect. By gumming
the lower valve to a piece of card, and allowing the end of a thread moist-
ened with gum to dry fast upon the upper valve, I was fortunately able to
separate the two without breaking them or injuring the remains of the an-
imal within. These afforded some interesting notes.

MUSEUM OF COMPARATIVE ZOOLOGY. 9

Muscular System. — The muscles were of a dark reddish-brown, and
consisted of a pair of cardinal muscles attached on each side of the notch
in the haamal valve, two pair of adductors in the usual position, and the
pedicel muscle. No others could be distinguished.

Brachia. — There were no apophyses, and the brachia were supported
by membranes of a horseshoe shape attaching them to the adductor mus-
cles. They consisted of a single row of distant cirrhi attached to the
edge of a horseshoe-shaped membrane, which passed behind the mouth and
was broadest on each side of the mouth, and prolonged anteriorly about
half the length of the valves, diminishing in breadth until it terminated in
a point on each side. There was no loop in the literal sense of the word.
On the outer edge of the membrane were long, slender, distant cirrhi,
about ten on each side. On the inner side of the membrane were a few
very short cirrhi, and the series was discontinued before passing below the
mouth. The external cirrhi were continuous above or behind the mouth.
They were tubular, hyaline, and presented transverse markings at short
intervals, somewhat as if they were annulated. They were about .018 of
an inch in length. The mantle was exceedingly thin, hyaline, with a
plain edge, not furnished with seta?.

Organs of Digestion. — The mouth was transverse and small, the upper
" lip " somewhat produced in the median line ; the stomach was straight,
short, bag-shaped, and a little constricted behind the mouth. Its termina-
tion was csecal. Around the stomach a few yellowish hepatic digitations
were observed. There were no other organs ; intestine, ovaries, etc., being
absent. The peduncle was short and slender.

The apex of the larger valve presented a curious appearance. It was
not pointed, but kidney-shaped, and its consistency appeared to be some-
what granulated, differing from the rest of the shell in texture. The in-
dentation lay in the median line just above the foramen. In the middle
of the nucleus two well-marked pores or hyaline points, apparently perfo-
rations, were clearly visible. Around the edges of the nucleus the growth
of the shell appeared to have been rather toward a bag-shape, and this
gave an appearance of constriction around the edges of the nucleus. The
upper margin of the arch of the foramen was a short distance below this.
The upper part of the arch was closed by a very thin, transparent septum
of shelly matter. The edge of the apex of the smaller valve appeared to
be of the same granular texture as the nucleus, but this merged impercepti-
bly into the rest. There was no cardinal process, and a shallow emargina-
tion or notch of rounded shape completed the opening of the foramen when
the valves were together. The cardinal muscles were attached on each
side of this notch to the interior of the shell. The teeth of the hinge were
already well marked.

10 BULLETIN OF THE

Terebratulina D'Orb.

Terebratulina D'Orbigny, Comptes Rendus, XXV, p. 268, 1847.
Type T. caput-serpentis Lin. sp. Syst. Nat. Ed., XII, 1153, 1767.

Terebratulina Cailleti Crosse.
Terebratulina Cailleti Crosse, Journ. de Conchyl., XIII. (3d Series, V), p. 27,
pi. i, figs. 1 -3, 1865. Pourtai.es, Bull. Mus. Comp. Zool., I, No. 6, p.
109, 1867. Dall, Am. Journ. Conch., VI, p. 106, 1870.

This very distinct species was obtained by M. de Pourtales off Chorrera,
Cuba, in 270 fathoms; near Cojima, in 450 fathoms; off Double-headed-
Shot Key, in 471 fathoms; and near Tennessee Reef, in 115 fathoms.
Two specimens, from which the diagnosis of M. Crosse was drawn up,
were obtained at the island of Guadaloupe in two hundred fathoms, by an
Italian party who were searching for beds of coral. Although obtained in
several localities, it does not appear to be an abundant species, as the num-
ber of specimens obtained by the United States Coast Survey Expedition
was quite limited. It is well distinguished from other species by its granu-
lated ribs, but varies so much in form and other characters, that I doubted
whether all the specimens could be referred to Crosse's species, upon my
first examination. They all differ from his figures in a remarkable auricu-
lation of the valves and in the straightness of the hinge-line. These char-
acters, though present, vary so much in the different specimens that I have
come to the conclusion that the species is identical with his, and that his
specimens were merely an extreme variety. The normal specimens,
though varying in amount of inflation, almost exactly resemble Terebratu-
lina Miclielottina Davidson, described by that eminent palaeontologist, in
his monograph of the Italian Tertiary Brachiopoda (Geol. Mag., VII, No.
9), p. 14, September, 1870, pi. xix, figs. 22, a, b, c, from the Eocene (stage
E) at Mossano, Italy. Were the two found living in the same seas, no
one, I think, would hesitate to consider them identical. The median fiex-
uosity is very variable, and often entirely absent. The nodulation of the
ribs is more evident in young shells. They also vary from quite broad and
flat to elongated and much inflated.

The smallest specimens of this species which I was able to find among
those sent by M. de Pourtales were nearly .1 of an inch in length. The
characteristic sculpture was developed upon them to the very apex of the
shell. The nucleus was already gone, being probably deciduous or soon
lost by attrition upon the rocks to which the young shells attach them-
selves. The various muscles were already well developed. The mouth
was as described in the young of the T. cubensis. The intestine was short,
cylindrical, and straight. The lower portion was embraced by a few
hepatic digitations. These lobes wei*e very dark brown, the muscles of a

MUSEUM OF COMPARATIVE ZOOLOGY. 11

deep reddish-brown, and the brachia of a flesh-color. The latter were in
the shape of a horseshoe, with no trace of a median lobe. They were
close set and marked with transverse lines, as in T. cubensis. The mem-
brane which covered the viscera was covered internally with irregular
hyaline spots with well-marked boundaries, which no doubt are the limits
of the lacunar channels of circulation. The mantle was quite transparent,
with a brownish edge, and in each of the internal channels, corresponding
to the ribs of the outside of the shell, was a single bristle, composed of
longitudinal fibres of chitine, without any of the transverse markings
which are seen in the seta? of the adult. The extreme tip of the bristle
alone protruded from the mantle, and its inner extremity was slightly bul-
bous. It was of a glistening yellow color throughout. In those adults
which I examined there were only five or six of these setae in each mantle
lobe. These specimens were obtained off Havana, in two hundred and
seventy fathoms water.

The very extraordinary manner in which all the soft parts were crowded
and crammed with masses of calcareous spicula? defied my best efforts to
obtain any very satisfactory results from the two or three alcoholic speci-
mens at my command. A flocculent mass of white matter resisted the
action of acid, and filled all the interstices of the membranes, so as to ren-
der them quite opaque. The genitalia were in such a condition that they
were quite invisible, and the animals appeared to be out of season. The
intestine was cylindrical, and ended much as it does in T. ccqiut-scrpcntis.
The mouth was surrounded by a dark-brown line. There were no struc-
tures above and behind the mouth, such as are described as existing in T.
cubensis. The attached extremities of the muscles were of a very bright
red-brown. Most of the specimens were overgrown with a tough, spongy
organism, like velvet. The peduncle is white, slender, and exceedingly
long, the exposed portion sometimes equalling in length one third of the
shell. A brownish tinge pervaded all the tissues of the adult. Transverse
markings were noticed on the brachia, as described in other species by
Hancock.

One specimen growing on a rock which had become covered with
sponge afforded an interesting observation. The peduncle was exceed-
ingly long, and, on cleaning off the sponge, it was seen that the creature, on
the growth of the sponge toward it, had apparently lengthened its peduncle
to get out of the way ; and while the original attachment still remained
(and the glossy opalescent color of that part of the peduncle testified to
its healthy condition), somewhat farther on, nearer the shell, a second at-
tachment of the peduncle had taken place by the outgrowth, from the
underside, of a bunch of cylindrical rootlets, exactly resembling the attach-
ment of an ivy to a stone. The under side of the peduncle and the root-

12 BULLETIN OF THE

lets were brown ; the rest, opalescent white. It is true that there is no
absolute proof that the peduncle had been lengthened, but 1 know not
how else to explain the extraordinary length and second attachment, and
I see nothing intrinsically improbable in the supposition.

Genus WALDHEIMIA King.

Waldheimia Kixg, Permian Fossils, p. 81, 1850. + Eudesia + Macandrevia

Kixg.
Type Waldheimia flavescens Lam. sp. Hist., VII, p. 3,30, 1836.

Waldheimia floridana Pourtales.

Waldheimia floridana Pourtales, Bull. Mas. Comp. Zoul., I, No. 7, p. 127,
1868. — Dale, Am. Journ. Conch., VI, p. 112, 1870.

Tcrebratula septata Jeffreys, Proc. Royal Soc, 121, p. 446, f 79, 1870.

Terebratulei septigera Jeffreys, 1. c.

Not T. (TerebrateUa) septata Philitpi, Moll. Sicil., II, p. 68, t. 18, f. 7, 1844.

Not T. ( Waldheimia) septigera Loven, Ind. Moll. Scand., p. 29, 1846.

T. ( Waldheimia) peloritana, var. Jeffreys, 1. c, not Seguenza Sicil. Brach.,
pi. vi, figs. 1-10, 18C5.

Florida reefs, between 110 and 200 fathoms, rocky bottom, Pourtales.

This species belongs to a peculiar group of the subgenus Waldheimia
(sensu stricto), containing several recent and some fossil species. Tcrebra-
tula septata Philippi, to which species peloritana, septigera, and floridana
have been referred, proves to belong to a different genus (see Davidson,
Mon. Ital. Tert. Brach.) from any of them. T. peloritana is referred by
.Mr. Davidson to T. septigera, in which Signor Seguenza concurs. We
have, then, three allied but sufficiently distinct forms, as follows : Wald-
heimia floridana Pourt., W. septigera Loven, and W. Iiaphaelis Dall.* The
first is from the Florida coast, the second from the seas of Northern Europe,
and the third from Japan.

The following table of measurements of large adult specimens will give
an approximate idea of their respective forms : —

Length, inch. Width. Diameter.

W. floridana 0.90 O.90 0.70

W. septigera 1.20 1.10 0.80

W. Raphaelis 1.75 1.30 1.00

Thus it is seen that the smallest species is by far the widest and most in-
flated, proportionately; the second species is the flattest, in proportion to
its length ; and the third is the most elongated. I have taken the largest
adult specimens of each species for comparison ; that of the septigera being
far larger than the ordinary form of that species, as it is one collected by

* Am. Journ. Conch., VI, p. Ill, pi. vii, figs, a, b, c, d, 1870.

MUSEUM OF COMPARATIVE ZOOLOGY. 13

Mr. Jeffreys, F. Ii. S., on the British Deep-Sea Dredging Expedition,
which was presented by him to M. de Pourtales.* The W.Jlorblana j)re-
sents very little variation among specimens of similar ages. A compar-
ative diagnosis is here given.

W.floridana. — Color grayish or brownish white. Form nearly an equi-
lateral triangle, widest near the anterior edge ; much inflated. Anterior
margin very strongly flexuous, the concavity being in the haemal (dorsal)
valve. The anterior corners of the haemal valve sharply pointed. Area
very narrow and short, deltidia just completing the foramen. Sides almost
flat, neural (ventral) valve broadly channelled in the middle. Apex
but slightly produced, short, rather acute. Cardinal process minute,
pointed, not recurved. Hinge plate wider than long ; anterior point over
the septum, behind the crura of the apophyses. Hinge teeth very short
and slender. Anterior ends of the lateral loops of the apophyses broadly
flaring, the shelly portion of these loops broadest near their ends ; hamial
arms of the apophyses close together and parallel for half the length of the
shell. Narrowest part of the recurved loop near its posterior end. Visceral
area small, muscular impressions within the posterior third of the dorsal
valve. Stomach spherical with a long cylindrical intestine.

W. septigera. — Color as in the last. Form roundly ovate, somewhat
truncated and wider in front. Anterior margin more or less flexuous.
Anterior corners of the haemal valve obtusely rounded. Area much wider
and longer, solid, with no median line indicating a separation or division
into deltidial plates. Sides rounded, inflated. • Neural valve not chan-
nelled, slightly concave near the exuosity flat the anterior margin. Apex
somewhat produced, blunt. Cardinal process broad, blunt, short, hardly
differentiated from the hinge plate. Hinge plate longer than wide, ante-
rior point passing forward between the crura. Hinge teeth very stout and
strong. Anterior ends of the lateral loops incurved, their laminae widest
near their posterior portion. Haemal arms of the apophyses diverging in
a wide curve from the hinge plate. Recurved part of the loop short, sides
nearly parallel. Visceral area very small, muscular attachments even
more posterior than in the last. Stomach posteriorly produced into a
point, without differentiation of the intestine, and very much shorter than
in the last.

W. liaphaelis. — Color deep brown. Form rather elongate, squarish,
widest near the middle. Anterior margin sharply, shortly flexed. The

* I have since had the opportunity, through the kindness of Mr. Jeffreys, of examin-
ing other and more normal forms of this and several other species obtained by him.
These included one specimen of the septigera, which he regarded as a transition toward
W. fioridana, which has been figured for the Zoological Society, and which he very
politely lent for comparison. It is elsewhere referred to.

14 BULLETIN OF THE

convex flexuosities of the haemal valve pointed, not at the outer corners, but
nearer the median line. Area moderate, without any median line. Sides
not inflated. Neural valve channelled for two thirds of its length, with two
prominent rounded carina? corresponding to the flexuosities of the margin.
Apex very short and blunt. Cardinal process quadrate, long, abruptly re-
curved, like the blade of a hoe. Hinge plate longer than wide, anterior
point passing between and almost beyond the crura. The latter are longer
and more slender than in the previous species. Anterior ends of the lateral
loops nearly parallel, the widest part of the shelly lamina; being near their
posterior terminations, but the width of this part of the apophyses is nearly
uniform from one end to the other. Hamial arms of the apophyses diverging,
in nearly straight lines, from the hinge plate. Recurved part of the loop
proportionately much longer than in the two previous species ; neural
portion forming a regular ovoid. Visceral area very large, muscular im-
pressions reaching the middle of the shell. Soft parts mostly unknown.

I have been thus explicit, perhaps more so than the subject requires, be-
cause the first two of our species have been united by Mr. Jeffreys, whose
opinion is justly entitled to weight, though I am forced to disagree with
him upon the present occasion. I consider septigera and Jioridana as two
well-marked and distinct species, in which opinion I have reason to believe
Mr. Davidson concurs. W. septigera and Raphaelis are more nearly allied
but the points of difference already noted are quite sufficient to distinguish
them, aside from the habitat and the fact that the adult Raphaelis is twice
the size of the largest septigera hitherto collected.*

The greater portion of the mantle of W. jioridana is of the most ex-
treme tenuity and perfect transparency. It is furthermore so closely
attached to the shell as to render its removal intact — even with the aid of
acid — a matter of great difficulty. With this exception, the examination

* In the specimen already alluded to, and regarded by Mr. Jeffreys as a transitional
form between septigera and jioridana, all the characters of septigera as above given are well
marked. It differs from the ordinary forms of septigera in being proportionately wider
than many of them, and in the sharper angles of the marginal flexuosities. But it is
noticeable that these last are not at the anterior corners, as is always the case in jiori-
dana, but are strictly within the anterior margin at some little distance from the anterior
corners, as is always the case in septigera. Hence I cannot admit that there is any tran-
sition exhibited in this specimen, but merely an exaggeration of the usual characters of
the species. The apophyses are missing. One of the other specimens of septigera, in the
same lot. fortunately preserved the ovaria, and I am glad to be able to state that they
differ entirely in form and extent from the same organs in W. Raphaelis. This is a good
character, though it varies somewhat, within certain limits. I must again thank Mr.
Jeffreys for the kindness which he has shown in forwarding specimens for examination,
— an example worthy of imitation by other naturalists, and well calculated to assist in
dispelling false impressions, and in adding to the accuracy of scientific work.

MUSEUM OF COMPARATIVE ZOOLOGY. 15

of its anatomy is easy. The following notes are the result of a careful
dissection of several specimens. The soft parts are mostly of a translucent
whitish color. The number and disposition of the muscles are similar to
those of W. australis, already described by various authors. The muscles
themselves are of a glistening tendinous appearance, except at their points
of attachment, where they are of a more or less dark yellowish-brown.
The peduncle is moderately long, and the portion which is external or
contained in the foramen is covered with a dark, horny reddish-brown
membrane or skin, and the attached extremity is trumpet-shaped. Upon
opening the shell in its normal position, the median spires of the brachia
are seen to be somewhat widely separated, and between them is stretched
a fine translucent membrane extending forward from the under lip of the
mouth and following the downward curve of the median lobes. In this
great extension.of this membrane this species differs from T. caput-serpenlis
and W. australis, in which species the cirrhi of the median lobes touch at
their extremities, and are separated by only a very narrow strip of mem-
brane between their bases, so that the appearance is almost as if there was
but a single broad band of cirrhi in the median line. This intervening
membrane in ordinary specimens of W. floridana is about .24 of an inch
in width at its narrowest visible portion. The upper and lower bands of
cirrhi in the lateral loops are also much more widely separated by a similar
membrane, than in W. australis. The reason of this appears in the
fact that the brachial band follows the outer edge of the apophyses in both
species until it curves downward in the middle, and the shelly portions of
the apophyses in W. floridana are very much wider than in W. australis ;
hence the greater separation. The longest of the brachial cirrhi, in front,
measure about .14 inch in length ; those of that part of the band which
passes behind the mouth are about the same length. Th6y are, as in other
species, disposed in a double row, the cirrhi of one row being opposite the
spaces of the other. The spiral portion in the middle lobe makes about
two complete turns. With regard to their disposition and the manner
in which the cirrhi are set upon the brachial band, I can add nothing
to the observations of Mr. Hancock, with which my own agree in every
particular. A series of transverse lines at regular intervals was observed
on the individual cirrhi, somewhat resembling in appearance the trans-
verse markings on the setae. The mouth is, as usual, just in front of the
posterior junction of the brachial bands, and is in a rather long flexuous
groove, the edges of which are of a dark brown color, and somewhat
thickened. The upper or posterior lip, if such it may be called, has a for-
ward prolongation or convexity in the median line, to which a slight con-
cavity or indentation in the lower lip corresponds. The oesophagus is
about half as long as the intestine, and has a slight curve, of which the

10 BULLETIN OF THE

convexity is anterior ; it is transversely flattened close to the mouth, and
is a little compressed laterally, behind that portion. It is of a nearly uni-
form calibre throughout. It has quite a thin lining membrane, which be-
comes thicker, though still smooth, in the stomach, and quite thick and
longitudinally plicated in the intestine. The stomach is well differentiated
from the alimentary canal and intestine, and is of an oval shape. It is
embraced by the hepatic digitations, which are of a greenish-yellow
color, and empty into the stomach by four ducts. The orifices of these
ducts are of a compressed oval shape, obliquely inclined, and the anterior
pair, which correspond to the right and left anterior congeries of hepatic
digitations, are twice as large as the posterior pair, which similarly corre-
spond to the anterior lobes or bunches of digitations. The individual
digitations appear to be longer, larger, and less numerous than those of
W. australis, etc., as described by Hancock. They are traversed by
numerous ducts and bloodvessels, and the hepatic matter, when separated,
appeared to be of a granular consistency. Among the yellowish granules
in the hepatic matter, both before breaking it down and afterward, were
noticed certain darker granules, similar in general appearance to those
found in the ovary. The digitations are distinctly arranged in four groups,
of which the anterior pair are the larger. The upper and posterior sur-
face of the stomach is bare, and the arrangement of the mesentery and
the gastro- and ileo-parietal bands essentially agrees with the description
of the , same parts in other species of this group, as given by Hancock.
The intestine is twice as long as the oesophagus, of uniform calibre, and
perfectly straight. It leaves the stomach abruptly without any dilatation
of the portion adjacent to the latter organ, and reaches about half-way to
the dorsal valve. The heart is situated behind the junction of the stom-
ach and intestine. The termination of the intestine is abruptly rounded
off and not at all pointed. It is entirely closed, and is upheld by the
mesentery. It is also of a much darker color than the rest of the alimen-
tary canal, beiDg of a deep chestnut-brown hue.

The great pallial sinuses and their ramifications in W. floridana are of
much less extent and disposed in quite a different manner from that which
obtains in W. australis. The haemal pallial system consists essentially of
four branches which are remarkable for their straight course and the pau-
city of their ramifications. The neural pallial system is very similar, with
a greater number of small sinuses about the perivisceral cavity, but in
both lobes the narrowness and small extent of the sinuses, as compared
with those of other species, is very marked, and the same is true with
regard to the ovaries. But a very few exceedingly delicate spiculce were
observed in the floor of the greater sinuses. The heart consists of a very
minute pyriform vesicle situated behind the intestine at its junction with

MUSEUM OF COMPARATIVE ZOOLOGY. 17

the stomach, and sending one vessel in the haemal direction along the
median line of the stomach, and another on each side laterally. It is
attached by its lesser extremity, and, contracted by the alcohol, appears
exceedingly minute. A very careful search failed to reveal any accessory
pulsatile vesicles, yet it is possible that, from their extremely small size,
they may have been overlooked. The ovaries are very limited in extent
and principally confined to that portion of the sinuses which surrounds
the visceral cavity, only their ultimate extremities entering the larger
branches of the great sinuses. Those in the haemal valve are vermiform,
slightly hooked at their posterior extremity. Those in the neural valve
form open loops, with the " bight " posterior, and the two anterior extrem-
ities just entering the two outer sinuses. Their manner of suspension is
the same as in the other species of the genus. The ova were visible in all
stages of growth. Those floating free in the lacunes were nearly spherical,
and of a flesh color ; their substance seemed of a granular consistency, due
perhaps to the action of the spirit in which they were preserved. The
immature ova were pyriform, attached to the ovary by their pointed ends.
With the yellowish matter of the ovary were interspersed specks of a
brownish granular matter, which appeared dark yellowish under a high
power and intermixed with what seemed to be fat-globules; Somewhat
similar specks were observed in the hepatic matter. This was more
abundant toward the middle line of the ovary, but was irregularly distrib-
uted. No spermatophorae or spermatozoa were observed in any of the
specimens examined. The oviducts were situated as in W. australis. The
lining membrane of their trumpet-shaped portion was .drawn into thin
plicae. Their apices were teat-shaped, with very small orifices.

The mantle margin is folded as in other species ; the fold is deeper in
the sides than'in front, and not wide anywhere. The setae are very slender
and fine, irregularly marked with transverse lines, but smoother toward
their outer ends. They protrude from their follicles, hardly more than one
third of their length. In no instance was more than one seen to issue
from a single follicle. The circumpallial muscular band is very slender
and narrow. No coloring matter was observed in or about the follicles.
The mantle edge was brownish, and seemed to have a slightly villous epi-
thelium. No setellae, such as I have elsewhere described as existing on the
setae of Discina and Lingula, were to be found.

I did not observe any noticeable peculiarity in the perforations of the
shell-structure. The " suture " or breaking point, described by Messrs.
Jeffreys and Carpenter in W. cranium, exists in all the species with a
reflected loop, and is due to the deposition of the shelly matter of the loop
in laminae parallel with the longer axis of the shell, which makes the loop
weaker at the point of reflection than elsewhere.

VOL. II. 2

18 BULLETIN OF THE

Subfamily PLATIDIIN^E.
Platidiinm Dall, Am. Journ. Conch., VI, p. 142, 1870,

Genus PLATIDIA Costa.

Platidia 0. G. Costa, Faun, del Reg. Napoli, p. 47, January, 1852. — Dall,
Am. Journ. Conch., VI, p. 142, 1870.

Morrisia Davidson, An. Mag. Nat. Hist., p. 371, May, 1852, and the gener-
ality of authors.

Platidia anomioides Scacchi sp.
Terebratula anomioides Scacchi ; Phil. Moll. Sicil., II, p. 69, pi. xviii, fig.

9, 1844.
Platidia anomioides Costa, 1. c. ; Dall, Am. Journ. Conch., VI, p. 143, figs.

20, 21, 1870.

This species was dredged off the Samboes on the Florida coast in two
hundred and thirty-seven fathoms. This is the second genus (Crania being
the other) of Brachiopoda which has been added to our fauna by the
researches of the United States Coast Survey Expedition. It has since
been obtained by Mr. J. Gwyn Jeffreys, F. R. S., of the English Deep-Sea
Dredging Expedition on the Porcupine, in 1869, in the Shetland Channel,
at a depth of three' hundred and forty-five fathoms. Previously this species
had only been known as an inhabitant of the Mediterranean in deep water.
The specimen from the Florida coast presents no differences in size or
general appearance from the Mediterranean form. The calcareous prisms
of which the shell is composed, and the perforations, are remarkably large
and conspicuous.

Subfamily MEGATHYRINyE Dall.

Megathyrinaz Dall, Am. Journ. Conch., VI, p. 143, 1870.
Argiopidcv King, Perm. Foss., p. 142, 1850.

Shell with a straight wide hinge line ; apophyses consisting of a sub-
marginal loop, attached to the hinge margin, provided with crura and
intersected by one or more submarginal elevations or septa.

Brachia in a single series, following the loop, surrounding a smooth disk
or membrane, in the posterior median portion of which the mouth is
situated. Setae absent from the mantle edge. Pedunculated.

Genus MEGATHYRIS D'Orb.

Megathyris D'Orbigny, Pal. Fran. Ter. Cret., p. 147, 1847. — Dall, Am.

Journ. Conch., VI, p. 144, 1870.
Argiope Deslongchamps, Mem. Soc. Lin. Norm., VII, p. 9, 1842.
Not Argiope Savigny, Desc. de l'F^gypte, XXII, p. 334, 1827. — Thorell,

Ann. Mag. Nat. Hist., p. 190, 1868; a genus of spiders.

MUSEUM OF COMPARATIVE ZOOLOGY. 19

It has been stated by some authors that Savigny's name was j^rgyope
or Argyopes, and hence not synonymous with Argiope, Desl. ; but this is an
error, which a reference to Thorell's paper, or to the original work of
Savigny, will enable any one to correct.

Subgenus CISTELLA Gray.

Cisiella Gray, B. M. Cat., p. 114, January, 1853. — Dall, 1. c. p. 145.

Zellania Moore, Proc. Som. Arch. Nat. Hist. Soc, 1854.

Shell with a single submarginal septum and bilobed loop. Surface
smooth or radiately ribbed. Brachia deeply emarginated by the septum.
Cardinal process absent or inconspicuous.

Type Cistella cuneuta Eisso sp., 1826.
Habitat : living in the Mediterranean.

Cistella {1 Schrammi var.) rubrotincta.

?? Argiope Schrammi (var.) Crosse and Fischer, Journ. de Conchyl., XIV
(3me Ser. VI), p. 269, pi. viii, fig. 6, 1866.= Cistella Schrammi Dall,
Am. Journ. Conch., VI, p. 146, 1870.

West of Tortugas, 30 to 43 fathoms, January 14, 1869, Pourtales.
Guadaloupe, W. I., 200 to 250 fathoms, Crosse and Fischer.

Shell small, semicircular, with the area at right angles to the plane of
the ha?mal (dorsal) valve. Hamial valve rather flat, with about ten pale
yellow rather strong ribs with brilliant scarlet interspaces ; a slight depres-
sion externally may be noticed on the surface of the valve, and occasionally
an attempt at a median rib, near the margin. Interior whitish, marked
by the punctations which are clearly visible to the naked eye in a good
light. Margin smooth, except for the fimbriated appearance caused by the
incomplete marginal perforations which are visible as grooves under a
lens. Hinge line straight, without area, hinge plate, or distinct cardinal
process. Septum triangular, extending from the hinge margin to the ante-
rior border of the shell. Most elevated point, forming the apex of the
triangle in the middle of the valve, rather bulbous and of a red color.
Anterior slope of the septum to the border of the shell, straight without
nodules ; this part of the septum is thin and even. Posterior slope of the
septum irregularly concave, thick, and nodulous, tapering to a point at the
hinge margin. On either side of the septum below its apex, is a transverse
wing or plate at right angles to the septum, of a thick nodulous form, the
two wings, taken together, presenting a heart-shaped plate with the broad
end downward. These extensions, however, are not confluent with the
valve, except close in by the base of the septum. Apophyses attached to
the hinge margin, provided with rather broad crura pointing toward each
other horizontally ; the lower edges of the laeunae of the apophyses con-

20 BULLETIN OF THE

fluent -with the shell throughout its entire length, and attached to the sep-
tum a short distance in front of the transverse plate, and running up on
the sides of the thin part of the septum for a short distance. Area behind
the lamina; much thickened for the muscular impressions, excavated be-
neath the lower edges of the transverse septal plate. Hinge margin as
wide as the shell, deeply grooved for the reception of the teeth of the
neural valve.

Neural valve convex, with a straight hinge margin and broad area.
Foramen usually much eroded ; deltidia rudimentary, widely separated ;
hinge teeth strong. Interior of the valve with a low well-marked septum,
rounded, broadest near the middle of the valve, where its upper edge is
somewhat excavated ; extending from the edge of the foramen to the an-
terior border of the shell, where there is a slight indentation. Length of
the shell .15 inch, width .18 inch.

This shell, of which a moderately large series was obtained, has a gen-
eral resemblance in form to C. Schrammi, but has a greater number of
ribs, wants the smooth mesial area, and is of a totally different coloration,
being scarlet with pale yellow ribs, while Schrammi is figured of a rufous
brown, paler on the beaks. There are some discrepancies between the
figure and the description of C. Schrammi, and, in spite of the apparent differ-
ences, I do not feel confident that the shells before me, and those described
by Crosse and Fischer, are more than varieties of one species. The form
of the septum and the transverse lamina at once separate the Coast Sur-
vey shells from any other species, but of C. Schrammi, unfortunately, the
apophyses are not figured. I have, therefore, indicated the present form
under a provisional varietal name, which will serve to distinguish it until
more exact knowledge is attained. I should add that the scarlet color is
not distributed in solid rays, but, under a lens, appears in concentric lines
transverse to the ribs and broadest in the interspaces.

The examination of the soft parts of this species added nothing new.
It appeared to resemble the next species in" every particular. The ovaria
were three-branched in the neural valve. The size and extent of the
transverse plate of the septum varied in different specimens.

Cistella (l Barrettiana var.) lutea.
11 Arc/iope Barrettiana Dav., P. Z. S., February, 1866, p. 103, pi. xii, fig. 3.

= Cistella Barrettiana Dall, Am. Journ. Conch., VI, p. 146, 1870.
? 1 Arm'ope Antillarum Crosse and Fiscuer, Journ. de Conchyl., XIV (3mo Se'r.

VI), March, 1866, p. 270, pi. viii, fig. 7. = Cistella Antillarum Dall,

Am. Journ. Conch., VI, p. 146, 1870.
Tortugas, 30-43 fathoms, Pourtales. Northeast coast of Jamaica, 150

fathoms, Barrett, Dav. Guadaloupe, W. I., 200 - 250 fathoms, Crosse

and Fischer.

MUSEUM OF COMPARATIVE ZOOLOGY. 21

Shell uniform light brownish-white; with twelve principal radiating ribs
on each A'alve, and secondary riblets between them, toward the margin.
Neural valve with a more or less marked depression extending from the
beak to the anterior margin, where it forms a slight convexity. Corre-
sponding to this internally is a slight rounded ridge. Hinge line straight,
sides and anterior margin slightly rounded. Area flat, smooth, as wide as
the shell. Pscudo-deltidia large, triangular, widely separated. Foramen
very large and usually much eroded. Hinge teeth moderately large and
strong. Muscular impressions very posterior, hidden beneath the area
when viewed from above. Margin slightly crenulated. A few faint stria?
discernible upon the surface of the ribs. Haemal valve smaller, flatter,
with a straight hinge line slightly emarginated in the middle, no area or
cardinal process ; teeth and sockets large and strong. Septum large and
stout, composed of three or four radiating ribs, with thin shelly matter be-
tween them, forming nodules and notches on the upper edge ; the whole
of a subtriangular form somewhat resembling a half-opened fan. Pos-
terior edge slightly concave, reaching a little behind the middle of the
shell ; anterior edge reaching the anterior margin of the shell, which
is here slightly concave or emarginated, giving the valve a somewhat
bilobed appearance. Muscular impressions much thickened, forming
two rather concave disks. Apophyses consisting of two haemal bands
attached to the hinge margin, first with two broad crura pointing
toward the median line, the arms of the apophyses extending in a
rounded curve within the middle third of the shell, and attached by
their lower edges to the thick disk-like muscular scars, and, lastly, to
the septum on each side about its middle, close to the shell. Cardinal
plate, or hinge plate, absent. Area behind the muscular disks somewhat
excavated.

The anterior portion of the apophyses is more posterior than in C.
Neapolitana, and the margin is not granulated as in that species. It would
seem from Mr. Davidson's figures that the loop of Cistclla Barretliana Dav.,
is more anterior than in this species ; the latter being also unprovided
with the posterior extension of the septum seen in the figures of the for-
mer, and being, moreover, entirely destitute of the red markings between
the ribs. It agrees with C. Antillarum Crosse and Fischer, as figured in gen-
eral appearance, but wants the red markings attributed, in the description,
to that species, and the ribs are also carried over the apex, while that por-
tion of C. Antillarum is described as smooth. No comparisons can be drawn
in regard to the apophyses, as Crosse and Fischer did not figure those of
their species. It is possible that the present species, C. Antillarum and C.
Barretliana are forms of one species, in which case the last name has pri-
ority, or it may be that the two latter are distinct from the present species,

2 BULLETIN OF THE

though synonymous with each other ; but they should not, as M. Cross©
observes in a letter on the subject, be united until clear proofs are shown
of identity, and therefore I have proposed for the present form a pro-
visional varietal name,"which may serve to distinguish it until the question
is settled by the comparison of specimens. My largest specimen meas-
ured .26 inches wide by .18 long.

Small as is this species, it is considerably larger than those of the
Mediterranean, and it was with much interest that I submitted it to an
anatomical examination.

I have not met with much success among these small species in the use
of acid in dissolving away the shell from the animal, and have been prin-
cipally obliged to work with specimens forcibly removed from their shells,
— a process which is not calculated to present the parts in the best con-
dition. Nevertheless, I have been able to determine some points of interest
in a satisfactory manner.

The brachia in this and the other species of the genus are arranged
around the edge of a broad membrane, which covers the concavity of the
shell, like a drumhead. The hoop of the drum is represented by the
apophyses. The brachia differ from the same organs in the Terebratulince
in being arranged in a single series instead of a double one. Of this there
can be no doubt, it is very evident upon a casual inspection, and is entirely
confirmed by careful dissections. In this species the drumhead membrane
is divided into two lobes by the septum. The edges of these lobes are
fringed with the brachia. The latter, in the alcoholic specimens, show
distinct transverse markings. They are usually curled up in front and on
each side, while those which are situated behind the mouth are longer than
the others, and usually lie smoothly over them, extending forward without
any marked curve, pointing toward the anterior margin of the shell, and
extending clear over the central membrane, even beyond the posterior
edge of the septum. The brachia are covered with an epithelium fur-
nished with cilia, are tubular, and communicate with a series of brachial
channels, which did not appear to differ from those of Waldheimia as de-
scribed by Hancock, as far as I was able to discover. The great brachial
canal was rendered conspicuous by a band of cartilaginous substance which
seemed to form its external covering, or rather beneath Avhich it was
situated, and which was longitudinally striated. The external edge of the
membranes, between which the apophyses were formed, was directly at-
tached to the pallial lobes at the points where the apophyses are attached
to the muscular disks of attachment already described. On either side of
this attachment, however, was a kind of pocket, opening externally, where
the brachial and pallial membranes did not coalesce ; and, there being one
on each side of the point of union, there were consequently four in all,

MUSEUM OF COMPARATIVE ZOOLOGY. 23

two on each side of the septum.* The drumhead membrane, covering the
space inside of the brachia, was translucent white or opalescent, and quite
thick and tough toward the middle of each lobe.

I am inclined to think that an error has been perpetuated in regard to
the position of the mouth of Megathyris decollata. It has been figured
and described by Woodward as being of a circular form, and situated in
the midst of the drumhead membrane. It is certainly not so situated in
Cistella ; and I do not believe that it is in Megatliyris, though I have only
seen dry specimens. In the present species it is placed, as in all the Tere-
bratulkhe, at the back of this membrane, just in front of the posterior
junction of the brachia, and at the bottom of a deep transverse groove
which is of a stout membranous consistency, and the two sides of which, for
convenience' sake, I have called the lips {labia). In the present species
the oral groove is situated far back and close to the brachia, which are ex-
ceptionally long behind it, as already described. It is, in fact, entirely
hidden by them until they are laid back. The groove is very long and
quite deep, the entrance to the oesophagus being trumpet-shaped and flat-
tened transversely. Were the brachia disjwsed as in Woodward's figure,
the oral groove would be hidden. I am disposed to think that this was
really the case in the specimen figured, and that the extraordinary circular
mouth there figured was an accidental lesion of the dry tissues, which might
easily be taken for a mouth in so small an animal. The labia, in all the
Brachiopods I have examined so far, have invariably exhibited a tinge of
darker color than the surrounding tissues. The present case forms no ex-
ception. The posterior lip presents a small prominence in the median
line, and the anterior lip a small emargination or concavity below this
prominence. This structure is also common to all the Brachiojwds I have
examined.

The oesophagus is wide, transversely flattened, with thin walls, and of
an orange color. It enters the stomach nearly at a right angle, without
much dilation. The stomach is oval with thicker and firmer walls ; the
inner lining appearing slightly villous and rugose. The intestine is not
differentiated from the stomach on the lower side, but on the upper side a
deep groove occurs at the juncture. The canal is stout and thick at its
lower extremity, tapers slightly, and terminates in a somewhat bulbous,
but pointed cascal extremity, attached to the perivisceral membrane. The
various membranous bands which support the alimentary system present
no differences from the homologous structures in other species of the Tere-
bratuliike. The stomach was filled in each case with a yellowish flocculent
matter. The hepatic lobules resembled those of other species, entering

* There is but one on each side, close to the hinge margin, in the last species.

24 BULLETIN OF THE

the stomach by two ducts on each side, of which the anterior were the
larger. They did not extend over or cover the sides of the intestine.

The heart is extremely small and difficult to find. It is situated lower
down than in most species and between and hidden by the hepatic lobules.
It is nearly spherical. No accessory pulsatile vesicles were found after
close scrutiny.

The ovaries differ in appearance from those of Waldhehnia and Tercbra-
tulina. They hang like a frill or puckered ribbon-like lamina from the pallial
membranes, and form a simple loop on each side of each valve. Those of
the haemal (dorsal) valve were most developed. The ends of the loops
extended into the great pallial sinuses. The rounded granules which
studded the frills were of two kinds. Those at or near the extreme edge
were of a pellucid deep brown hue, while those closer to the pallial mem-
branes were mostly of a pale yellowish color and quite opaque. The ovi-
ducts are very inconspicuous and not easily found. They are situated in
the usual position, but exhibit only a very few short folds, and the exter-
nal opening directly in the midst of them, instead of being at the end of a
rather long duct, as in other forms. There are only two of them. They
do not appear to be attached to the intestine or mesenteries, but lie flatly
upon the parietes.

The pallial sinuses are comparatively insignificant in this species, being
very narrow, almost linear, channels with few branches. A few spicule
were observed in some of them. The margin of the mantle is perfectly
plain, without seta?, and adhering closely to the shell. Yet the circumpal-
lial muscular band is much broader than usual and strongly marked.
W lien torn from the shell, the ca?cal prolongations of the mantle were
beautifully shown. They were often bifurcate and occasionally had three
or even four branches.

The punctate structure of the shell was very coarse. Even the crura
and laminae of the apophyses were punctate.

The nervous system was not traced out, but the oesophageal ganglia
presented no special peculiarities.

The border of the mantle appeared to be ciliated. The peduncle, so
wide and short as to resemble a mere muscular disk, was strongly at-
tached to the shell by the peduncular muscle, beside which a broad tendi-
nous band appeared to pass entirely across, in front of the dorsal adjusters
(posterior retractors of Owen), giving an additional solidity and firmness
to the attachments of the peduncle. The extremities of all the muscles
were very much enlarged and thickened, while their median portions were
slender and 'tendinous. No striated fibres were observed.

MUSEUM OF COMPARATIVE ZOOLOGY. 25

Order LYOPOMATA Owen.

Syn. = Lyopomata Owen, Enc. Brit., Ed. VIII, XV, Article Mollusca, p. 339,

1858.
= Drachiopodes, valves libres Deshayes, Lamk., An. s. Vert., 2'1" Ed., VII,

p. 309, 1836.
= PaUiobranchiata, testa acardis Van der Hceven, Handb. der Zool., p. 692,

1850.
= Pleuropygia Bronn, Klass. Ordn. Thicrr., Ill, 1 Abth., p. 301, 1862.
= Inarticulata Huxley, Lcct. Class., 1864. Intr. Class. Anim., p. 116,

1869.
x Pedimculata + Sessilia Latreille, Fam. Nat. Reg. Anim., p. 204,

1825.
-=: Brachiopodes brachide's D'Orbigny, Cours. Ele'm. Pal.,. II, p. 82, 1849.
-= Sarcicobrachia Gray, Ann. Mag. Nat. Hist. 2d Ser. II, p. 438, 1848.

Characters. Arms free, unsupported by shelly apophyses. Intestine
opening by a lateral anus. Lobes of the mantle disunited all around their
borders. Brachia without a distinct median lobe. Shell, in most cases,
without hinge teeth, articulation, or cardinal process.

Family CRANILD.E.

Syn. = Craniidce H. and A. Adams, Gen. Rec. Moll., II, p. 583, 1858. — Jef-
freys, Brit. Conch., II, p. 24, 1863.

= Cranidce Owen, Anat. Inv. Index, p. 683, 1855. — D'Orbigny, Cours
Ele'm. Pal., II, p. 90, 1849.

= Cranidees D'Orbigny, Pal. Fran. Ter. Cre't., IV, 1844* Comptes
Rendus, XXV, p. 269, 1847. An. Sci. Nat. c. xiii, p. 350, 1850*
(fide Gray, B. M. Cat).

= Craniadce Gray, Syn. Brit. Mus., p. 155, 1840* — Ibid.,1. c, p. 88, 1842.
P. Z. S., p. 202, 1847. Ann. Nat. Hist., 2d Ser., II, p. 438, 1848. —
Davidson, Int. Class. Brach., p. 51, 1851. — Woodward, Man. Rec.
and Foss. Shells, p. 235, 1854. — Owen, Anat. Inv., p. 503,1855.—
Clark, Brit. Test. Moll., p. 37, 1855. — Gosse, Mar. Zool., II, p. 80,
1856. — Davidson, Mem. Lin. Soc. Norm., X, p. 84, 1856. — Suess,
Wohns. der Brach., p. 38 (220), 1859. — Mrs. Gray's Moll., IV, p.
202, 1859. — Carpenter, Lect. Moll. Smithsonian Rep., p. 276, 1860.
— Chenu, Man. de Conchyl., II, p. 230, 1862. —Bronn, Klass. Ordn.
Thierr., Ill, 1 Abth., p. 301, 1862.

= Craniade'es Davidson, Mem. Soc. Lin. Norm., X, p. 226, 1856.

~=cLes Cranies Ferussac, Tabl. Syst, folio 38, 1819. — Rang, Man. Moll.,
p. 262, 1829. — Deshayes, Enc. Me'th., II, Table, 1830. Hist. An.
s. Vert., 2de Ed., VII, p. 309, 1836.

-=: Cranio Hkrrmannsen, Ind. Gen. Mai., I, p. 315, 1846 (as of Fer. Rang,
and Desii).

26 BULLETIN OF THE

-c Craniacece Menke, Syn., p. 56, 1828,* olim.

-c Craniacea Menke, Syn., Ed. II, p. 96, 1830. — ANTOTT Verz., p. xii,

1839. — Agassiz, Nomencl. Fasc. IX, p. 31, 1846. — Mcerch, Cat.

Yoldi, p. 64, 1852.
^zCraniidce Forbes, Mai. Mon., p. 38, 1838* — King, Ann. Nat. Hist.,

XVIII, p. 28, 1846. Perm. Fos., p. 78, 1850.
-=: Craniadai Forbes and Hanley, Brit. Moll., II, p. 364, 1853.
^cLes Orbicules Cuvier, Lecons d'Anat. Comp. An. VII, I, t. 5, 1798*

(fide Gray). Regne An., II, p. 504, 1817 ; Tabl. El. Hist. Nat., p.

435, 1799.
^cAthjridw McCoy, Carb. foss. Irel., p. 104, 1844.
-=cOrbieuke Herrmannsen, Index Gen. Mai., II, p. 156, 1847 (as of

Desha yes).
-c Les Oslrace'es Lam., Phil. Zool., 1809* (Ed. 1830, p. 31 7.)
~^Plaeunea Rafinesque, Anal. Nat., p. 148, 1815.
■^ Fixivalvia Latreille, Fam. Nat. Regn. An., p. 205, 1825. (Ed. Berth,

p. 196.)
-= Palliobranches a coquilles non symmetriques Blainville, Man. Mai., p.

515, 1825.
-=: Terebratulidea G. B. Sowerby, Trans. Lin. Soc., XIII, p. 469, 1822.

Shell calcareous ; hingeless ; without perforation for a pedicel ; attached
by the umbones of or the entire lower valve, or rarely free. Upper valve
suborbicular, with a subcentral apex. Lower valve subcircular or pyri-
form. Four principal muscular impressions in each valve. Shell struc-
ture punctate. Animal with free spiral arms, the direction of the apex of
the spires toward the concavity of the upper or ha?mal valve. Mantle
extending to the edge of the valves, closely adhering, without seta? upon
its external edge. Animal holding the same relation to the attached
valve which obtains in the Terebratulidcc, but actually reversed in relation
to surrounding objects, on account of the attachment being by the surface,
instead of by the recurved apex, of the neural valve.

Synopsis of the Family.

Genus Crania Retz: Shell attached ; upper valve with the muscular impres-
sions usually excavated, but occasionally convex, without apophyses of
any kind, inner surface vaulted, without septa; impressions of the pallial
sinuses flabclliform, separated in front. Margin of the valves tubercu-
lose or papillose. Type Crania craniolaris, Lin. sp. Syst. Nat., Ed. XII,
p. 1150, 1767.

?1 Subgenus Pseudocrania McCoy. Shell free, with the impressions of the
pallial sinuses fimbriated and confluent in front. Margins smooth. Ante-
rior muscular impressions larger than the posterior ones. Type Crania

MUSEUM OF COMPARATIVE ZOOLOGY. 27

antiquissima Eichwald, sp. McCoy, Annals Nat. Hist., VIII, p. 388, 1851.
The value of this section is doubtful.

Subgenus Cranopsis Dall. Attached; upper valve with two. pointed slender
apophyses divaricating from the internal apex, where the muscular im-
pressions of the anterior pair are situated in the typical Cranice. Type
Crania Parisiensis, Defrance, Davidson, Me'm. Soc. Lin. Norm., X, pi.
xiii, fig. 23 a, b, 1856.

Genus Craniscus Dall. Fixed valve divided by a transverse and a longitudi-
nal septum into three cells, the posterior of which contains the muscular
impressions and the rostrellum. Type Crania tripartita Munster, sp.
Davidson, 1. c fig. 21.

It is extraordinary that the two sections here indicated have not been
separated previously, and indicates that this group has received little
attention from modern authors. The differences between the genera
Crania and Craniscus are hilly as great as any existing between the
acknowledged genera of the Terebralulidce ; and the characters of Cra-
nopsis, as separated from Crania, are well marked. The genus Spondylo-
bolus of McCoy appears to have rather more affinity with the Lingulidoz
than with this group, as I have elsewhere observed, and it is not in-
cluded here for that reason.

The genus Pholidops Hall appears congeneric with PseudocraniaM.cCoy.
It is known principally from casts, however, and further researches may
establish its validity.

Genus CRANIA Retz.

Non binomial syn. Ostracitrs minimus . . . numulus Braltensburgcnsis diclus Sto-
b^eus, Act. Lit. Sci. Svec, pp. 14 and 21, 1731 * (fide Retz.), and Opuscl.
I, p. 31, Tab. l.figs. 1-3, 1752.

Helmintholithus craniolaris Linne, Syst. Nat., XII, III, p. 162, 1768.

Anomilcs craniolaris Drattensburgenscs et Ignaburgensis Wahlenberg, Act.
Upsala, 1821 * (fide Bronn).

Nummuli Drattenburgencs Waller, Syst. Min., II, p. 500, 1775.

Criopus Jimbriatus + Criopoderma (turbinatum) [taken collectively^ pars, Poli,
Test, utriusq. Sicil., I, p. 34, 1791 ; II, pp. 189, 255, 261, 1795 (fide G.
W. Tryon, Jr.

Ostracites ?7iinimus, &c. Beuth, Jul. et Mont, subt, p. 130, 1776.
Actual syn. = Crania Retzius, Schrift., Berl. Ges. Naturf. Freunde, Bd. II, p.
72, 1781. — Philippson (?), Diss. Hist. Nat., p. 11, § v, No. 1, 1788.—
Schroter, Lith. Lex., IV, p. 265, 1785. — Bruguiere, Enc. Me'th. Vers.,
I, tabl. p. xiii, 1789. — Defrance, Diet. Sci. Nat., XI, p. 312, 1818.— G. B.
Sowerby, Gen. Shells, fasc. XII, n. d., 1821 (?). Trans. Lin. Soc, XIII,
p. 431, 1822. — Hceninciiaus, Isis, p. 108, 1822* (fide Engelman).—
Nilsson, Kong. Vet. Ak. Handl., p. 378, 1824. Act. Holm., p. 326, 1825*

28 BULLETIN OF HIE

Petref. Suec, p. 37, 1827. —Gray, Ann. Phil., XXVI (N. Ser. X), p. 244,
1825. — Menke, Syn. Ed. II, p. 96, 1830. — Philippi, Moll. Sicil.,p. 100,
1836.* — Grateloup, Cat. Zool., p. 55, 1838.*— Morris, Cat. Brit, foss.,
p. 121, 1843. — Loven, Index Moll. Scand., p. 29, 1846. — King, Ann. Nat.
Hist., XVIII, p. 28, 1846. Ibid., Perm. Foss., p. 84, 1850.— Sowerby,
Thes. I, p. 366, 1 847. — Jay, Cat. Shells, Ed. IV, p. 94, 1 850. — Davidson,
Int. Class, Brach., p. 122, 1853. — Forbes and Hanley, Brit. Moll., II, p.
365, 1853. — Owen, Anat. Invert., p. 503, 1855. — Clark, Brit. Test.
Moll., p. 27, 1855. — Davidson, Mem. Soc. Lin. Norm., X, pp. 24, 226,
1856. — Gosse, Mar. Zool., II, p. 30, 1856. — H. and A. Adams, Gen. Rec.
Moll., II, p. 583, 1858. — Bronn, Klass. Ordn. Thierr., Ill, 1st Abth. p.
302, 1862. — Chenu, Man. de Conchyl., p. 230, 1862.— Davidson, Mon.
Sil. Brach., p. 78, 1866. —Jeffreys, Brit. Conch., II, p. 24, 1863 ; V, p.
165, 1869 (Retz. non Piiilippson, as asserted by Jeffreys). — Hall, Pal.
New York, IV, p. 26, 1870.
= Crank Lamarck, Phil. Zool., Ed. 1830, p. 317 ; Ed. I, 1809*
>- Crania Lamarck, Prodr., p. 83, 1799.* — Megerle v. Muhlfeldt, Entw.
Syst. Schaalh., 1811* (fide Schum.). — Lamarck, Hist. An. s. Vert., lm
Ed. VI, p. 237, 1819. —Ferussac, Tabl. Syst., p. xxxviii, 1821. — Gray,
Lond. Med. Repos., 1821 * (fide Herrm.). — Blainville, Man. Mai., p.
515, 1825. —Rang, Man. Moll., p. 262, 1829. — Deshayes, Enc". Me'th.
Vers, II, C, p. 15, p. 553, Tab. ace'ph., 1830. — Lamarck, Hist. An. s.
Vert., 2de Ed. VII, p. 297, 1836. — Thomas Brown, Conch. Textb., Ed.
V, p. 108, 1839. — Macgillivray, Ibid., Ed. IX, p. 123, n. d. — Thorpe,
Brit. Mar. Conch., p. 125, 1844. — Cuvier, Regne An. Moll. Ed. Deshayes,
p. 251, 1845. — Quexstedt, Handb. Petref., p. 494, 1852. — Woodward,
Man., p. 236, 1854.
>■ -< Crania Schumacher, Essai, p. 37, 1817.

:=- Crania a, Schumacher, Essai, p. 101, 1817.
Not Crania /3, Schumacher, Essai, p. 102, 1817. = Discina.
Not Crania Gould, Moll. P. S. Expl. Expcd., p. 465, 1852. = Discina, sp.

= Craniolites Schlotheim, Petref., p. 247, 1820.

= Cranicella Rafinesque, Analys. Nat., p. 148, 1815.

= Orbicula Cuvier, Tabl. Ele'm. Hist. Nat. p. 435, 1798 ; and Regne An., l"«
Ed., p. 504, 1817. Regno, An. Moll., Ed. Deshayes, p. 250, 1S45. — La-
marck, Hist. An. s. Vert. I, VI, p. 242,1819.
>- -< Orbicula Lamarck, Hist. An. s. Vert., Ed. II, VII, p. 313, 1836.

>- Orbicula Lamarck, Prodr., 1799.* — Bosc, Hist. Nat. Coq., II, p. 243, 1801.

— Lamarck, Syst. An. s. Vert., p. 140, 1801. — Cuvier, An. Mus., I;
Me'm. sur laLingule, p. 9, 1802. — Schumacher, Essai, pp. 55, 176, 1817.

— Deshayes, Enc. Me'th. Vers., Ill, p. 668, 1832 (not tabl. 2, p. 553, 1830).
= Orbicula sp. Eichwald, Sil. Schicht. Syst., p. 169, 1840; Urwelt, Russl.,

I, p. 98, 1840; II, p. 75, 1842.
— Orbiculoidea sp. Ryckuolt, Melanges Pal., pi. iv, fig. 3* (fide Davidson,
Intr., p. 128).

MUSEUM OF COMPARATIVE ZOOLOGY. 29

>~ Orbicularius Dumeril, Zool. Analyt., p. 1G8, 1806.

= Distinct sp. Turton, Dith. Brit., p. 238, 1822 (Gen. diag. exclus.).

= Anomia sp. Linne, Syst. Nat., Ed. X, I, p. 700, 1760 ; Ed. XII, p. 1150, No.
216, 1767. — Gmelin, Syst. Nat., p. 3340, 1792. — Chemnitz, Conchyl.
Cab., VIII, p. 72, 1785. — SchrGter, Einl., Ill, p. 381, 1778.* — Poli,
Test, utriusq. Sicil., II, p. 189, 1795. — Dillwyn, Cat. Shells, I, pp. 285,
286, 1817.

= Anomites sp. Davidson, Me'm. Soc. Lin. Norm., X, p. 226, 1856, in syn.
(not of Linne, Syst. Nat., 1768).

= Patella sp. O. F. Muller, Prodr. Zool. Danica, p. 237, No. 2870, 1776 ;
Zool. Danica, I, p. 4, 1788. — Gmelin, Syst. Nat., p. 3721, 1792. — Mon-
tague, Trans. Lin. Soc., XI, p. 195, 1808. — Humphrey (ubi ? fide Sby.
and Rve). — Koch and Dunker, Beitr. Ool.-Geb., p. 51, 1837. —
Puemer, Verst. Ool.-Geb., p. 135, 1840.

= Terebratula (part) Schweigger, Naturgesch., p. 690, 1820* (fide Gray, An.
Phil., 1-825).

= Producta 1 sp. Klipstein, Beitr. Ost. Alp., VIII, pp. 60, 239, 1843.

= Siphonaria sp. Quenstedt, Handb. Petref., I, p. 442, 1852!

= Numulus Agassiz, Nomencl. fasc, IX, p. 60, 1846.

= Siphonotreta sp. Eichwald, Zool., I, p. 274, 1829 * (fide Bronn 1. c.)

= Criopododerma Agassiz, Nomencl. Ind., p. 301, 1848 (corr. Poli).

= Criopus Gray, Lond. Med. Repos., 1821 * (fide Herrm., Agassiz) —Flem-
ing, Phil. Zool., II, p. 499, 1822, and Brit. An., pp. 367, 377, 1828. — King,
Perm. Foss., p. 84, 1849.— Leach (Gray), Moll. Gt. Brit., p. 358, Dec.
1852.

= Cryopus Deshayes, Hist. An. s.Vert., VII, p. 314, 1836.

Shell with or without a more or less produced beak and false area in
the lower valve. The two posterior muscular impressions are near the
cardinal border, and usually larger and more widely separated than the
two anterior scars. The latter are near the centre of the valves, and in
the lower valve are confluent in front, or barely separated by a small nose-
like prominence, or rostellum, which is usually excavated at its most ele-
vated extremity for the attachment of a muscle, which at its other end is
attached near the cardinal border of the upper valve between the post-
adductor scars ; the shelly matter being slightly produced on each side of
this attachment in some species, forming two slight tooth-like prominences.
Margin of the shell more or less tuberculous or papillose. Lower valve
differing in position and extent of attachment to extraneous objects.
Upper valve conical, with the apex subcentral ; internally vaulted, simple,
without apophyses or septa. Lower valve without septa or apophyses,
unless the rostellum be so considered. Exterior foliated, concentrically
or radiately striate, or smooth. Great pallial sinuses leaving more or less
flabeliiform or paucidigitated impressions on the shelly matter, which im-
pressions are not confluent anteriorly.

30 BULLETIN OF THE

Soft parts with two spiral arms in the horizontal plane, with the apices
of the spires directed toward the concavity of the lower valve. Intestine
terminating between the lobes of the mantle on the ( ? right) side.

The genus as described bv Retzius, was founded on several species
which he confounded together under the name of Crania Brattensburgensis.
Under this name he included the " Ostracites minimus . . . Numulus
Brattensburgensis dictus " of Stobaeus, the Anomia craniolaris of Linne,
and a recent species said to be from the Philippines, but probably the same
previously described by Miiller under the name of Patella anomala, from
the Scandinavian seas.

The question now arises as to which of these shall be taken as the type
of the genus, and shall therefore retain the specific name given by Retzius.
With regard to this authors have differed, and the result has been a con-
fusion only equalled in the generic synonymy of this unfortunate group.
Most of them have transferred the C. Brattensburgensis of Retzius to the
synonymy of the recent species (C. personata Lam.), overlooking the fact
that Lamarck's name has not priority, and ignoring Midler's name en-
tirely, though it preceded that of Retzius. On the other hand, they have,
placed the JVumulus Brattensburgensis of Stobaeus in the synonymy of C.
nummulus Lam., with the Anomia craniolaris of Linne, which in its turn
is long pi'ior to that of Lamarck. This disregard of priority by the earlier
authors has always been a fruitful cause of confusion and annoyance to
subsequent students. As Retzius evidently had the species described by
Stobaeus in his mind as the species of which he supposed he was describ-
ing the recent form, I think that the only course left for us is to accept
Stobseus' species as the type. Schumacher, in his Essai (p. 102),"says that
Retzius had sent him specimens of the two species which he had
described, and that the C. Brattensburgensis Retz. was a fossil. Now
most, if not all, authors agree that Stobaeus' species was identical with
Anomia craniolaris of Linne, which is identified by Hanley and others
with the Crania nummulus of Lam., which of course becomes a synonyme.
Stobaeus was not a binomial author, and Linne's name being the first
binomial appellation, his specific name must stand. Lamarck, also, in
adopting the genus Crania (Prodrome, p. 83, 1799), took Anomia cranio-
laris as the type.* Schrbter, Gmelin, and Dillwyn, as well as Chemnitz,
continued to confound the recent and fossil species under the name of
craniolaris. Miiller, in 1776, was the first author to describe the European
form, under the name of Patella anomala, with a correct habitat, and it
afterwards received from Poli the specific name ofiurbinata, though not in
a binomial sense.

* Woodward also adopts it as the type, and Davidson, under the specific name of
Brattensburgensis

MUSEUM OF COMPAPwATIVE ZOOLOGY. 31

The synonymy of the type, according to these views, will stand as fol-
lows : —

Crania craniolaris Lin. sp.

Non-binomial synonymy. Ostracites 7ninimi(S parasiticus calvarium hominis utcun-

que referens, numuhis Brattensburgensis dictus K. Stobjeus, Diss, epist. Act.

Litt. et Sci. Svec. pp. 14-21, figs. 1, 2, 1731* (Retz., 1. c. p. 74, 1781).

Opuscl., p. 31, t. 1, figs. 1, 2, 1752.
Anomites craniolaris Brattensburgensis Wahlb., Act. Ups., VIII, p. 60, 1821*

(fide Bronx, Ind. Pal.).
Concha testa planiore orbiculata cranium humanum referente Linne, Fauna

Svecica, p. 384, No. 1347, t. 2, fig. 1347, 1746 (a later edition is probably

referred to in the Syst. Nat. 1. c.)
Ilelmintholithus (anomiie) craniolaris Linne, Syst. Nat., Ed. XII, III, p. 164,

1768.
Ostracites minimus sive ostracites nwnismalis Becth, Jul. et Mont, subt., p.

130, t. 7, No. 46, 1776.
Actual synonymy. Anomia craniolaris Linne, Fauna Svecica, 2150, fig. 2150,

Ed. II, 1761. Syst. Nat., Ed. X, t. 1, p. 700, No. 183, 1760. Syst. Nat.,

Ed. XII, t. l,pt.II,p. 1150, No. 216, 1767. — Gmelin, Syst. Nat., t. 1, pt.

VI, p. 3340, No. 1, 1792, partly (+ C. anomala + turbinata part).—

Chemnitz, Conchyl. Cab., VIII, p. 72, t. 76, pars, 1785. — Dillwyn,

Cat., I, p. 285, No. 1, 1817. — Hanley's Conchyl. Lin., p. 119, 1855.
Crania Brattensburgensis (pars) Retzius, Schrift. Berl. Ges. Naturf. Freunde,

Band II, p. 73, 1781 (fig. excl?) (-f C. anomala part?). — Schumacher,

Essai, p. 101, 1817.
Crania nummulus Lamarck, Hist. An. s. Vert., Ed. I, t. 6, p. 238, No. 2,

1819. — Lamarck, Hist. An. s. Vert, Ed. II, t. 7, p. 299, No. 2, 1836.

— ILeninghaus, Mon. Crania, p. 5, No. 5, figs. 5 a, b, c, 1828. —

Deshayes, Enc. Meth. vers, II, C, p. 17, 1830, &c, &c.
Crania personata part, Lamarck, Hist. An. s. Vert., Ed. I, VI, p. 238, 1819.

Not C. personata of Lamarck, Ed. II, VII, p. 299, 1836.

This species is found fossil in Sweden, where Stobaeus and Linne
obtained their specimens, and the lower valves, furnishing a rude imita-
tion of a face stamped on a coin, were sufficiently common to obtain the
popular designation of Brattensburg money, or pennies. According to
Deshayes and Sowerby, it presents the peculiarity of being attached by
only a small portion of its lower valve. This, however, is a character of
slight importance. I have omitted all the other ostensible synonymes of
craniolaris Lin. and nummulus Lam., because I have not had the opportu-
nity of certainly identifying them, and therefore have preferred to retain
only those of which there was no doubt whatever. In order to render the
matter more clear and throw as much light as possible upon the subject,
I subjoin the synonymy of the second species described by Retzius.

32 BULLETIN OF THE

Crania Egnabergensis Retz.

Non-binomial syn. Numtdus minor rarissimus ocali et r.aso prominent ibus e lapi-
cidina Egnabergensi in Gothungia K. Stob.eus, 1. c. figs. 3, 4* (fide
Retz., 1. c). Opusc, p. 31, t. 1, figs. 3, 4.

Anomites cranioluris Ignabergensis Wahlb., Act. LTps., VIII, p. 60, 1821 * (fide
Bronn, Ind. Pal.).

Actual syn. Crania Egnabergensis Retz., 1. c. p. 75, t. 1, figs. 4-7, 1781.
Crania striata Defrance, Diet. Sci. Nat., Vol. XI, p. 315, 1818. — La-
marck, Hist. An. s. Vert., VI, p. 239, 1819 ; Ed. II, VII, p. 301, No. 5,
1836. — Desiiayes, Eneyc. Me'th., II, C, p. 19, No. 9, 1830 (not C.
striata Schum.).

Found fossil at Balsberg and Charlottenlund in Scania. It is well dis-
tinguished from the preceding species by its radiating ribs.

The recent species, which may be referred to the genus Crania, are as
follows : —

Crania Suessii Rve., Mon. Crania, pi. i, fig. 2, 1862. Sydney, Australia.

Crania rostrata ILeningiiaus, Mon. Crania, p. 3, No. 3, fig. 3, 1828. — Rve.,
Coneh. Icon., pi. i, fig. 3, 1862. — Desiiayes, An. s. Vert, Ed. II, Vol.
VII, p. 302, 1836. (Syn. exclus.) Mediterranean, W. Africa?

I have received a Mediterranean specimen of this species from Mr.
Davidson ; it may also extend along the northwest coast of Africa, as
stated by Reeve. Deshayes' synonymy is very erroneous, and includes
both turbinata and anomala.

Crania (?? Cranopsis) japonica A. Adams, Ann. Mag. Nat. Hist., 3d Series,
XI, p. 100, 1863. Gotto Island, Japan, 71 fms.

And the following : ** —

Crania anomala Mull. sp.
A. Typical. Syn. Patella anomala Mull., Prodr. Zool. Dan., p. 237, 2S70,

1776. Zool. Dan., I, p. 4, t. 5, figs. 1-8, 1783. — Gmelin, Syst. Nat.,

p. 3721, No. 151, 1792.
Patella dislorta Mont., Trans. Lin. Soc, XI, p. 195, pi. xiii, fig. 5, 1808.

— Flem., Edin. Encyc, VII, 65, t. 204, fig. 4.
Patella Kermes Humphrey* (ubi?) fide Sby., Tr. Lin. Soc. — Rve., Couch.

Icon. Mon. Crania, sp. 4, 1 862.
Anomia cranioluris (pars) Chemn., VIII, p. 72, t. 76. — Gmelin, Syst. Nat., p.

3340, No. 1, 1792. — Dillwyn, Cat, I, p.285, 1817 (not of Linne).
Anomia turbinata Dillwyn, Cat., I, p. 286, 1817. Polii syn. exclus.
Criopus anomalies Fleming, Phil. Zool., II, p. 499, IS22, and Brit. An. p. 377,

1828.

** I have separated the synonymy of the var. ( ? ) turbinata, in order that those who
consider it a good species may make use of the synonymy, but I myself consider it as
a strict synonyme of anomala.

MUSEUM OF COMPARATIVE ZOOLOGY. 33

Criopus orcadensis Leach (Gray), Moll. Gt. Brit., p. 358, pi. xiii, figs. 6-

8, December,- 1852.
Orbiculu norvegica Lamarck, Syst, p- 140, 1801 (not Sby., Lin. Tr. and Gen.

Sh., Rang, Man., nor Blainville, Man., p. 515). — Lamarck, Hist.

An. s. Vert., Ed. I, Vol. VI, p. 242, 1819. Ibid., Ed. II, Vol. VII, p.

316, No. 1, 1836. — Deshayes, Enc. Meth., Ill, p. 668, 1832, partly

( + turbinata, part). — Schumacher, Essai, p. 176, pi. xxi, fig. 2, 1817.

— Thomas Brown, Conchol. Textb., Ed. V, p. 107, pi. xiv, fig. 32 (no
such figure there), 1839. — Macgillivray, Ibid., Ed. IX, p. 123, pi. xiv,
fig. 32 (same remark applies), n. d. — De Blainville, Diet. Sci. Nat.,
XXXVI, p. 292, 1825, partly (-(- Discina ostreoides, part).

Orbicula anomala Cuvier, Tabl. Ele'm. de l'Hist. Nat., p. 435, 1799, and

Regno An., II, p. 504, 1817; Ed. Desh. Moll., p. 251, 1845, partly

(+ turbinata, part).
Discina ostreoides Turton, Dith. Brit., p. 238, 1822 (not of Lam.).
? Crania Brattensburgensis, part, Retz., Sclirift. Berl., Ges. Naturf. Fr.,

Band II, p. 73, 1781 (+ craniolaris, part).
Crania turbinata Wood's Ind. Test., Ed. Hanley, pi. xi, fig. 2, 1856 (not

turbinata Poli).
Crania personata, part, Defrance, Diet. Sci. Nat., XI, p. 312 (+ turbinata,

part), 1818.
Crania personata Lamarck, Hist. An. s. Vert., Ed. I, VI, p. 238 (syn. exclus.),

1819. Ibid., Ed. II, VII, p. 298 (syn. excl.), 1836 (not personata

Blainv., Diet. Sci. Nat. Cah. V, figs. 2-9, fide Deshayes, 1. c. pr.).

— Sowerby, Trans. Lin. Soc, XIII, p. 431, 1822. Gen. Sh. Fasc, XII,
figs. 1 and 2, n. d. (1821 ?). — Blainville, Diet. Sci. Nat, XXXII, p.
304, 1824. Ibid., Man. Mai., p. 515, 1825.** — Nilsson, Kong. Vet. Ak.
Handl., Part II, p. 324, partly, 1825* (fide Fe'r. Bull. Sci. Nat. + crani-
olaris part). — Thomas Brown, Conch. Textb., Ed. V, p. 108, pi. xiv,
fig. 5, 1839. —Macgillivray, Ibid., Ed. IX, p. 123, pi. xiv, fig. 5, n. d.

— Suess, Wohns., I, p. 41 (223), 1859.

Crania norvegica Sowerby, Thes. Con., I, p. 368, pi. 73, figs. 15 and 17,
1847. — Forbes and Hanley, Brit. Moll., I, pi. U, fig. 2, 1853.

Crania rostrata Thorpe, Brit. Mar. Conch., p. 125, 1844 (not of Hiening-
haus).

Crania anomala Sowerby, Conch. Man., Ed. II, p. 125, fig. 197 a, 1842.

— Loven, Index Moll. Scand., p. 29, 1846. — Forbes and Hanley, Brit.
Moll., II, p. 366, pi. lvi, figs. 7 and 8, 1853. — Davidson, Int. Class.
Brach, p. 123, figs. 44-46, pi. ix, figs. 237, 238. — Woodward, Man.
Moll., p. 235, figs. 157-159, 1854. — Clark, Brit. Test. Moll., p. 37,

** I do not know whether Deshayes refers in his cited remarks to these two refer-
ences also, but there is nothing in the context to indicate the Mediterranean form
(turbinata), and the reference of Blainville is to C perso?iata Sowerby, which is un-
doubtedly anomala.

VOL. HI. 3

34 BULLETIN OF THE

1855. — Gosse, Marine Zool., II, p. 80, fig. 120, 1856. — Davidson,
Mem. Lin. Soc. Norm., X, p. 229, pi. xiii, figs. 14 - 1G, 24, 32, 33, 35, 3G,

1856. — H. and A. Adams, Gen. Rec. Moll., II, p. 583; III, pi. exxxii,
figs. 3, 3 a, 3 b, 1858. — Suess, Wohns., I, p. 39 (221), 1859. — Chenu,
Man. Conchyl., II, p. 230, figs. 1178, 1862. — Reeve, Conch. Icon. Mon.
Crania, pi. i, fig. 4, 1862. — Jeffreys, Brit. Conch., II, p. 24, pi. i, fig.
3, 1863 ; V, p. 165, pi. xix, fig. 6, 1869

Hah. North European seas.

B. var. turbinuta Poll

Anomia turbinuta Poli, Test. Utrius. Siciliae, II, p. 189, 261, t. 30, fig. 15,

1795, in synonymy. — Dillwyn, Cat., I, p. 286, No. 2, 1817, in part

(+ anomala, part).
Anomia craniolar is (part) Gmelin, Syst. Nat., p. 3340, 1792. — Dillwyn, Cat ,

I, p. 285, No. 1, 1817 (-f- craniolaris part).
Criopus fimbriatus (part) Poli, Test. Utrius. Sic., II, p. 189, 1795. "Hab. in

Anomia turbinuta" Poli, 1. c. (animal).
Criopodermu turbinatum Poli, II, p. 261, No. 1, 1795 (shell).
Patella anomala Dillwyn, Cat., I, p. 286, in syn.

Orbicula turbinuta Lamarck, Hist. An. s. Vert., Ed. II, VII, p. 317, 1836.
Crania persona ta Blainville, Diet. Set. Nat., XI, p. 312 ; XXXII, p. 304, pi.

lxxxiv, fig. 2, Cah. xv, 1818 (not of Lam., Hist., Ed. II, VII, p. 299, note,

fide Deshayes (?). Cf. previous note, p. 33).
Crania personata Deshayes, Encyc. Method., II, C, p. 16, 1830; partly;

(-4- anomala part, per citation of Retz.).
Crania rinrjens Hceninghaus, Mon. Crania, p. 3, No. 2, fig. 2. — Deshayes,

Encyc. Me'th., II, p. 16,No. 3, 1830. — Lamarck, Hist. An. s. Vert., Ed. II,

VII, p. 302, 1836.— Sby., Thes., I, p. 367, pi. lxxiii, figs. 10, 11,1847.—

Suess, Wohns., I, p. 41 (223), 1859.
Crania rostrata Deshayes, Hist. An. s. Vert., Ed. II, VII, p. 302, No. 7,

1836 ; partly (+ anomala and rostrata part).
Not Crania rostrata ILexinghaus, Mon. Cram, p. 3, No. 3, fig. 3, 1828. — Rve.,

Icon., pi. 1, fig. 3, 1862.
Hab. Mediterranean.

C. var. alba Jeffreys.

Crania anomala var. alba Jeffreys, Brit. Conch., V, p. 165, 1869.
Hab. Shetland, Hebrides.

The shell of Crania anomala is rounded, with a slight tendency toward
a squarish form. The posterior border of the valves is nearly straight,
and Barrett, who examined living specimens, asserts that the two valves
open and shut on this edge, like the sides of a hinge. Upper valve sub-
conical or depressed, with the apex not prominent and rather posterior.
External surface smooth in normal specimens, or slightly marked with
concentric lines of growth. Internally rather smooth, with coarse and
conspicuous punctation. The margin of the valves is rough, and pre-

MUSEUM OF COMPARATIVE ZOOLOGY. 35

scnts under a glass conspicuous calcareous prisms, radiating from the centre
of the shell. The muscular impressions are very variable in shape and
position as well as prominence. Ihe color is usually a livid reddish-
brown, with occasional white rays. The extreme nucleus of the shell is
mammillated. The lower valve varies in thickness according to the object
upon which it rests. If the latter be smooth and level, it is often very
thin and almost imperceptible, so that Miiller was not without justification
in overlooking it. The margin is usually rough or tuberculose, and the
muscular impressions vary as 'in the upper valve.

The variety alba of Jeffreys is pure white, or occasionally with a few
radiating brown lines, but does not differ otherwise from the normal form.
From specimen figures and descriptions of C. turbinala, I have been
unable to discover any characters which are not common to varieties of
C. anomala. I agree with Mr. Jeffreys in thinking C. ringens Hoeninghaus,
to be synonymous with anomala on general considerations, but I have
seen no typical specimens of ringens.

The few specimens of Crania dredged by the United States Coast
Survey Expedition (off the Sambos, Florida, in 116 fathoms, and off
the Sand Key in 105 fathoms) offer some apparently constant differ-
ences from C. anomala. They are somewhat distorted, very transverse,
and have pbscure indications of radiating i-ugosities. The shells are smaller
than C. anomala, have a strong concentric foliation caused by the imbri-
cation of the lines of growth. The color is much the same as in anomala ;
one white specimen with a few radiating brown lines was dredged on a
stone in 126 fathoms, off Sand Key, by M. de Pourtales. The interior of
the lower valves was of a green color. The posterior muscular impres-
sions are smaller and closer together than in C. anomala. It is very pos-
sibly, however, a strongly marked variety of that species ; but in case the
collection of a larger number of specimens should prove its distinctness, I
would propose for it the name of C. Pourtalesii.

Note. — Not having personally been able to examine Poli's Test. Utriusq.
Sicilian, I have been indebted to the kindness of Mr. George W. Tryon, Jr.,
for examining the work for me. It is evident to any one who appreciates
the binomial system of nomenclature, that Poli was in no sense binomial.
He named the animal generically and specifically, while the shell received
two additional names, making four in all, if we take them together, involv-
ing the absurdity of the animal being a different genus and species from
its shell.

The references of Poli given below, from Mr. Tryon's notes, are as fol-
lows : —

Vol. I, p. 34. " Genus 15, Criopus," description of animal as follows :
" Habitat in Anomia imperforata."

36 BULLETIN OF THE

Vol. II, p. 189. " Anomia turbinata," description of shell follows, and to
it is added a description of the animal as " Criopus firribriatus"

Vol. II, p. 255, a Table of Genera contains : —

Genus 18. Criopus (animal). Genus 18. Crwpoderma (shell).

Species 1. Criopus jimbriatus, " Habitat m Anomia truncata et capite ser-
pentis Lin. ; in Anomia turbinata."

Vol. II, p. 261. List of species : —

1. Crwpoderma turbinatum. Anomia turbinata.

2. Criopoderma truncation. Anomia Iruncata.

3. Criopoderma caput serpentis. Anomia caput serpentis.
Plate 30, fig. 15, Anomia turbinata.

Poli evidently considers Anomia as a synonyme, and only uses it by way
of explanation. It is evident that such a system of nomenclature as the
above can never be fairly squared with the binomial system.

Family DISCINdTLE.

Syn. = Discinidce Gray, Syn. Brit. Mus., I, p. 155, 1840* Ibid., p. 88, 1842. P.
Z. S., p. 202, 1847. Ann. Mag. Nat. Hist., II, p. 439, 1848. — Da-
vidson, Int. Class. Brach., pp. 51, 125, 1853. — Woodward, Man. Rec.
and Fos. Sh., p. 237, 1854. — Davidson, Me'm. Lin. Soc. Norm., X, p.
84, 1856. — H. and A. Adams, Gen. Rec. Moll., II, p. 584, 1858.—
Mas. Gray's Moll., IV, p. 202, 1859. — Suess, Wohns.,X p. 42 (224),
1859. — Carpenter, Lect. Moll., Smithsonian Rep., p. 276, 1860. —
Bronn, Klass. Ord. Thierr., Ill, I Abth. p. 301, 1862.

= Discinide'es Davidson, Me'm. Lin. Soc. Norm., X, p. 231, 1856.

= Orbiculacea Anton, Verzeichn., p. 21, 1839 * (fide Herrm.).

= Orbicul idee King, Ann. Mag. Nat. Hist., XVIII, p. 28, 1846. — Owen,
Anat. Inv., p. 503, 1855. — Chenu, Man. de Conchyl., II, p. 231,
1862.

«=: Orbicidina Agassiz, Nomend. Index, p. 757, corr. praec. 1848 (not Lam.,
gen. Rhizop.).

-< Orbiculidce D'Orbigny, Cours Ele'm. Pal., II, p. 89, 1849.

~<Les Orbicules Ccvier, Lecons dAnat. Comp., I, t. v, 1798* (fide Gray).
— Regno An., II, p. 505, 1817.

= Orbicules Deshayes, Encyc. Me'th., II, table, 1830. Ibid., Lam., Hist.
Nat. An. s. Vert., Ed. II, Vol. VII, p. 309, 1836.

= Orbiculidie McCoy, Carb. fos. Ireland, p. 104, 1844.

-=: Orbkulce Herrmannsen, Ind, II, p. 156, 1847, as of Desh.

-= Craniadct Forbes and Hanley, Brit. Moll., II, p. 364, 1853.

-<Les Cranks Ferussac, Tabl. Syst., folio 38, 1819. — Rang, Man. Moll., p.
262, 1829.

^< Craniacece Menke, Syn., Ed. I, p. 56, 1828, olim*

-=: Craniacea Menke, Syn., Ed. II, p. 96, 1830. — Agassiz, Nomencl. Fasc.,
IX, p. 31, 1846. — Moerch, Cat. Yoldi, p. 64, 1852.

MUSEUM OF COMPARATIVE ZOOLOGY. 37

-= Cranke Herrmannsen, Ind. Gen. Mai., I, p. 315, 1846 (as of Rang., Fer.,

and Desh. ).
-=; Brachiopea Rafinesque, Anal. Nat., p. 148, 1815.

— =: Palliobranches a cog. non symmetrigues Blainville, Man. Mai., p. 515, 1825.
«< Fixivalcia Latreille, Syst. Regn. An., p. 205, 1825 (Ed. Berth, p. 196).
-=: Terebratulidea G. B. Sby., Trans. Lin. Soc, XIII, p. 469, 1822.

Characters. Shell structure permeated with very minute tubuli. Shell
attached to foreign bodies by a pedicel passing through the neural valve,
inarticulated. Valves suborbicular, with a subcentral apex. Animal with
free spiral arms, with the apices of the spires directed toward the neural
valve. Mantle extremely vascular, fringed with long chitinous setae fur-
nished with setellaj.

Synopsis of the Family.

Genus Discina Lam. Type D. striata Schumacher sp. 1817.

Subgenus Discina, sensu stricto. Shell with subequal externally convex valves,
with subcentral apices. Lower valve with a small subtriangular longitu-
dinal septum or prominence in the centre, with a minute circular orifice be-
neath it, for the peduncle, from which an impressed line or furrow extends
on the inside, posteriorly, for a short distance. Shell of rather solid tex-
ture, impunctate ; perforated by very minute tubuli (?). Type D. striata
Schtjm., = D. radiosa Gld., -f- D. Evansii Dav., -+- D. norvegica Sby., +
D. ostreoidcs Lam.

Subgenus Orbiculoidea D'Orbigny,* = Schizotreta KuTORGA.t Shell similar
to the last, but with the perforation at the posterior end of the furrow,
which last is impressed from the outside, instead of from the inside as in
Discina. Type Orbiculoidea elliptica Kutorga. Dav., Int., p. 129, pi. ix,
figs. 253-255, 1852.

Subgenus Discinisca D all, = Discina Auct. Lower valve more or less flat-
tened, concave or compressed. Upper valve more convex ; apices of both
subcentral or subposterior. Lower valve with a small septum as in Discina,
behind which is an impressed disk or area, externally concave, and inter-
nally elevated. This is perforated by a longitudinal fissure, extending
from a short distance behind the septum nearly to the posterior margin,
which is often slightly indented behind it. Shell more or less horny in
texture, minutely tubulous. Type Discina lamellosa Brod. Rve., Conch.
Icon., pi. i, fig. 4, 1862.

Genus Trematis Sharp.J — Orbicclki D'Orbigny. § Shell with the upper valve
with a posterior apex and small false area. Lower valve flattened, with a

* Comptes Rendus, XXV, p. 269, 1847.

t Verh. Kais. Min. Ges., 1847.

J Quart. Journ. Geol. Soc., No. 13. Vol. IV., p. 66, June, 1847.

§ Comptes Rendus, XXV, p. 269, August, 1847.

33 BULLETIN OF THE

large foramen extending nearly to the posterior border. Shell structure
in two layers, the outer calcareous and sculptured with a peculiar netlike
sculpture resembling perforations ; inner layer horny, minutely tubulous,
as in Discina. Type Orbicula terminalis Conrad, in Nat. Hist. New York,
Part IV, Geology (Emmons), p. 395, fig. 4, 1842.

I am not prepared at present to admit Siplionotreia and Acrotreta into
this family, but am inclined to think, with Kutorga, that they form a
peculiar group by themselves.

Keyserlingia Pander (Bull. Ac. Sci. St. Petersb., Ill, p. 46, 1861,
Type K. reversa Pand., pi. ii, fig. 1, a-g), appears to have relations with
Trimerella or Gotlandia, or perhaps with Siphonolreta, but its position is at
present doubtful.

Rarity of specimens and errors in identifying types are probably the
reasons of the confusion of forms existing in this unfortunate family.

Genus DISCINA Lam.

Syn. = Discina Gray, Ann. Phil., XXVI (New Scr., X), p. 244, 1825. Ibid., transl.
Isis, p. 494, 1834. Syn. Brit. Mus., p. 155, 1840 * and p. 88, 1842.

— Davidson, Int. Class. Brach., pp. 51, 126, 1853/- — Woodward,
Man., p. 237, 1 854. — Davidson, Mem. Soc. Lin. Norm., X, p. 84, tabic,
p. 232, 1856. — H. and A. Adams, Gen. Rec. Moll., II, p. 584, 1858. —
Mrs. Gray's Moll., IV, p. 202, 1859. — Suess, Wohns., I, p. 42 (224),
1859. — Carpenter, Lect. Moll., Smithsonian Rep., p. 276, 1860. —
Hall, 13th Regent's Rep., p. 77, 1861. — Bronx, Klass. Oidn. Thicrr.,
Ill, 1st Abtb., p. 301, 1S62. — Hall, 14th Regent's Rep., p. 130, 1864.

— Pal. N. Y. Vol. IV, Part I, p. 15, 1870.

=- Discina Lamarck, Hist. An. s. Vert., VI, p. 236, 1819; Ed. II, VII, p.
296, 1836. — Rang, Man. Moll., p. 263, 1829. — Cuvier, Regne An.
Ed. Voigt, III, p. 602.* — Thomas Brown, Conch. Textb., Ed. V,
p. 108, 1839. — Macgillivray, Ibid., Ed. IX, p. 124, n. d.

= Crania /3 Schumacher, Essai, p. 102, 1817.

= Crania (sp.) Gould, Moll. U. S. Expl. Exped., p. 465, 1852.

= Orbicula Sowerby, Min. Conch., VI, p. 4, pi. 506, 1830. — Desii ayes,
Enc. Me'th. vers, II, tab. accph., 1830 (not Ibid., Ill, p. 668, 1832). —
G. B. Sowerby, Jr., Conch. Man., Ed. II, p. 209, 1842. — Morris,
Cat. Brit, foss., p. 123, 1843. — King, An. Mag. Nat. Hist., XVIII,
p. 28, 1846. — Sby., Thes., I, p. 365, 1847. — Mcerch, Cat. Yoldi, p.
64, 1852. — Owen, Anat. Invert., p. 503, 1855. — Chenu, Man.
Conchyl., II, p. 231, 1862.

= Orbicula (sp.) Lamarck, Hist. An. s. Vert., Ed. II, pp. 317, 318, 1836.—
Eichwald, Urwelt Rossi., II, p. 76, 1842.
;=--< Orbicula Lamarck, Phil. Zool. (Ed. 1830, p. 317), Ed. I, 1809.* — Sower-
by, Gen. Sh. fasc. XIII, n. d. (1821). Trans. Lin. Soc, XIII, p. 466,

MUSEUM OF COMPARATIVE ZOOLOGY. 39

1822. — Blainville, Diet. Sci. Nat., XXXII, p. 304, 1824. Ibid.,
XXXVI, p. 291, 1825. — Defrance, Ibid., XXXVI, p. 293, 1825.

= Patellites (sp.) Schlotheim, Petref., I, p. 114 ; II, p. 108, 1820-1823.

= Patella (sp.) Brongniart, Tabl. des Terr., p. 419, 1829.

= Calyptrcea (sp.) Goldfuss, Alberti, Betr., Mori. Trias , pp. 54, 93, 1831 *
(fideBRONN, Ind. Pal.).

= Terebratala pars, Schweig., Naturg., p. 690, 1820 * (fide Gray, An. Phil.).

Not Orbicula Cuvier, Tabl. Ele'm. R. An., p. 435, 1798. — Lamarck, Hist.
An. s. Vert., I, VI, Part I, p. 242, 1819. — Deshayes, Enc. Me'th. Vers,
III, p. 668, 1832. — Schumacher, Essai, p. 55, 1817. — Thomas
Brown, Conch. Textb., p. 107, Ed. V, 1839, nor Macgillivray,
Ibid., Ed. IX, p. 123, n. d. (= Crania).

Subgenus DISCINA (Lam.) Dall.

Shell of rather solid texture, with a considerable amount of calcareous
matter iu it ; no signs of punctation to be seen with a half-inch objective.
Valves convex, the lower valve varying in amount of convexity with its
habitat, but always more or less inflated. A small, sharp, longitudinal
septum rises from the centre of the lower valve, of a subtriangular shape,
covering and hiding a small tubular perforation of the apex of the shell.
This perforation is very oblique, and from its internal opening a groove
extends backward nearly half-way to the posterior border of the shell
inside. The anterior muscular scars meet in front of the septum and form
a semilunar elevation with the points directed backward. The posterior
scars in the lower valve are small and widely separated. On the external
surface the foramen appears nearly in the middle of the shell, and the fur-
row is continued anteriorly for a short distance. (There is no furrow in
my specimens outside behind the foramen, which is the only point of differ-
ence from Sowerby's figures.)

Upper valve convex, apex subcentral ; a slight median longitudinal
callus internally. There is no strongly impressed disk about the foramen
as in Discinisca, though slight traces of a differentiated area exist there.

Type Discina striata Schcm. sp.

Syn. Crania (/3) striata Schumacher, Essai, p. 102, pi. xx, figs. 1 a-f, 1817 (not
of Defrance). Habitat?
Crania radiosa Gld., Moll. U. S. Expl. Exped., p. 465, figs. 480, a-c, 1852.
Hab. Cape Palmas, Liberia, not Rio.

** Not Orbicula striata Sby., in Murch. Silurian Syst., tab. v, fig. 21, 1S39, and Siluria,
pi. xx, fig. 3, 1859. This species is perhaps identical with Discina Verneuilii Dav., 1848;
but if it should prove distinct, it must have anew name, as that of Schumacher has many
years' priority. It occurs in the upper Ludlow rocks of Shropshire, and the D. Ver-
neuilii in the Wenlock limestone of England.

40 BULLETIN OF THE

Orbicula striata Sowerby,** Thes. Conch., I, p. 366, pi. Ixxiii, fig. 9, 1847.

— Forbes and Hanley, Brit. Moll., II, p. 368, 1853. — Davidson, An.

Mag. Nat. Hist., IX, p. 376, 1852. Habitat 1
Orbicula Evansii Davidson, P. Z. S., 1852, p. 81, No. 12, pi. xiv, figs. 32 - 34.

An. Mag. Nat. Hist, IX, p. 376, 1852. — Suess, Wolms., I, p. 44 (226),

1859. Hab. Bodegas, Cal., in error.
Orbicula norvegica Sowerby, Trans. Lin. Soc, XIII, p. 468, pi. xxvi, fig. 2,

1822. Syn. exclus. ; Gen. Shells, fasc., XIII, figs. 3-5, n. d. (1821 ?),

(not. of Lam.). Ballast, North Africa.
Orbicula (s. g. Discina) norvegica Blainville, Diet. Sci. Nat., XXXII, p. 304,

1824. Man. Mai., p. 515, pi. lv;, fig. 5, 1825 (not of Lamarck).
Orbicula (s. g. Discina) ostreoides Bang, Man. Moll., p. 263, 1829.
Orbicula ostreoides Reeve, Conch. Icon. Mon. Orb., No. 7, pi. 1, figs, la, lb,

1862.
Discina ostreoides Lamarck, Hist. An. s. Vert., Ed. I, VI, p. 237, 1819. Ibid.,

Ed. II, VII, p. 297, 1836 (no description). — Thomas Brown, Conch,
extb., Ed. V, p. 108, pi. xiv, fig. 8, 1839. — Macgillivray, Ibid., Ed.

IX, p. 124, pi. xiv, fig. 8, n. d.
Not Discina ostreoides Ttjrton, Dith. Brit., p. 238, 1822 (= Crania anomala).
Discina norvegica Crouch, Int. Lam. Conch., pi. xiii, fig. 2 * (fide Forbes and

Hanley, II, p. 368).
Patella anomala Sowerby, Trans. Lin. Soc., XIII, p. 468, 1822, in synon. (not

of Muller).

When changes in nomenclature depend upon the identification of types
described by the earlier authors, the work is one of great difficulty, and
requires the utmost caution, lest fresh confusion be the result. In many
cases an approximation to a determination alone can be arrived at, and
authors may conscientiously differ as to the decision, and its bearings on
nomenclature. In the present case, however, there is but little difficulty,
as the species under consideration has been well described and carcfully
figured by the describers, though under several names ; the history of tbe
type specimens is very clear, and was put on record at the time.

Lamarck constituted the genus Discina to receive a shell which he called
D. ostreoides, but of which he did not give any figure or specific description.
The specimen was received from Mr. J. Sowerby, and is the same species
and from the same lot of specimens, as the shell described by Mr. G. B.
Sowerby in the Lin. Transactions, and well figured by him there, under
the name of Orbicula norvegica. His very excellent figure enables mo to
speak with positiveness in saying that it is identical with Crania radiosa
Gould, of which the type specimens are before me. The figures of Schu-
macher are sufficiently exact to allow of identifying the species with his
Crania striata. The figures given by Reeve and Davidson are excellent,
and almost certainly represent the same species, though this is a matter of

MUSEUM OF COMPARATIVE ZOOLOGY. 41

little consequence, the main point being the identification of Sowerby's
shell with the specimens before me, which may be regarded as certain.
The habitat of the species is undoubtedly African, the localities " Rio,"
" Bodegas," etc., being erroneous.

The fact of the type being settled, only one course remains, — to rear-
range the genus in accordance with the facts. Objections may, and probably
will, be raised against such rectifications, but accuracy being the aim and
basis of all science, nothing else is worth regarding, and rectifications,
however long delayed, are inevitable at last.

The species has been well described by various authors, and there is
nothing further in regard to it for me to add. The catalogue number of
Dr. Gould's types, in the Smithsonian Cabinet, is 5962. They were ob-
tained at Cape Palmas, West Africa. The exceedingly minute foramen
hidden beneath the septum might well excuse Dr. Gould for calling it a
Crania. I am not aware of any other species of true. Discina in a recent
state, but there are several species usually denominated Discina, with
which I am unacquainted autoptical])'.

There are no species of Orbiculoidea or Trematis known in a recent
state.

The following species of Discinisca have been found living. I have ex-
amined only those after which an exclamation-point is placed.

Discinisca stella! Gould, Proc. Bost. Soc. Nat. Hist., VII, p. 323, Sept., 1860.

— Reeve, Conch. Icon., pi. 1, fig. 1, 1862.
Singapore and Phillipines. Cumixg. China Seas. Stimpson.

Discinisca lamcllosa ! Broderip, P. Z. S., 1833, p. 124. — Reeve, Conch. Icon ,

pi. 1, fig. 3, 1862.
Panama to Peru.

Type of the subgenus. I have examined an immense number of speci-
mens from Panama, and find that they exhibit many varieties. The apex
of very young shells is circular, whitish, and of a different texture from
the remainder of the shell. This circumstance was first pointed out by
Mr. E. S. Morse. The nucleus is probably the remains of the embryonic
shell. The species has no radiating stria?, and is a thinner shell than
l&vis.

Discinisca tenuis Sby., Thcs., I, p. 366, pi. 73, figs. 4, 5. (not Reeve, Conch.

Icon., pi. i, fig. 5, 1862).
Hah.?

Reeve's figures of " Orbiculis tenuis " do not represent Sowerby's species,
but agree very well with some of the varieties of D. lamellosa. I have seen
no specimens of either of the former species, but the figures exhibit dis-
crepancies too great to be reconciled. Sowerby gives no localities, and

42 BULLETIN OF THE

Reeve's localities, or one of them, probably refer to the form which he
figures. It can hardly, however, be found both in Chili and South Aus-
tralia, and the double habitat is probably due to an error in labelling or
identification.

Discinisca Icevis! Sbt., Trans. Lin. Soc, XIII, p. 468, pi. xxvi, figs, i a-d,

1822. — Reeve, Conch. Icon., pi. i, figs. 4, a, b.
Concepcion, Chili. Cuming.

A specimen of this species was received from Peru through Mr. Cuming,
labelled strigata Broderip.

Discinisca Cnmingii ! Brod., P. Z. S., 1833, p. 124. — Reeve, Conch. Icon.,

pi. i, fig. 6, 1862. = D. strigata Brod., teste Reeve.
Cape St. Lucas to Panama.

Mr. Reeve's figure offers no characters by which it might be distin-
guished from the last species. The specimens received under- this name
from Mus. Cuming cannot be distinguished from D. Stella Gould, by con-
stant characters.

Discinisca (?) antillarum D'Orbignt, Moll. Cuba, p. %8, pi. xxviii, figs. 34 - 36,

1853. — Reeve, Conch. Icon., pi. i, fig. 2, 1862.
' Cuba, Martinique. Cuming.

I have never seen this species, which is stated by Reeve to resemble D.
slella. It has relations with D. Cumingii, and a series should be compared.
This species was not obtained by M. de Pourtales.

Discina (Discitiisca ?) atlantica Jeffreys, MSS.
Northeast Atlantic.

I am indebted to Mr. Jeffreys for information in regard to this species,
which is on the point of being published in the Proceedings of the Zoologi-
cal Society of London.

The following species of Brachiopods, which were not obtained by the
Coast Survey Expedition, are known to inhabit the Caribbean province.

Cistella Woodwardiana Davidson, P. Z. S., Feb. 1866, p. 103, pi. xii, fig. 4,

a, b, c.
Northeast coast of Jamaica in 60 fathoms, Barrett.
Theadium Barrettii Woodward. Davidson, Geol. Mag., I, pi. ii, figs. 1-3,

1864. P. Z. S., 1866, p. 104.
With the last-mentioned species.

Glottidia (?) antillarum Reeve, Conch. Icon., pi. ii, fig. 8, 1861.
Martinique. Cuming.

MUSEUM OF COMPARATIVE ZOOLOGY. 43

REFERENCES TO PLATES.

Plate I.

Fig. 1. Diagram of Terebratula vitrea, natural size.

2. Diagram of Terebratula cubensis, natural size.

3. Diagram of Waldheimia floridana, natural size.

4. Diagram of Waldheimia septigera, natural size.

5. Diagram of Cistella lutea, much enlarged, showing the loop, septum,
and disk-like muscular attachments.

5 a. Side view or section of the same, showing the ribbed septum.

6. Diagram of Cistella rubrolincta, much enlarged, showing the interior
and the transverse plate of the septum.

6 a. Side view of the same.

7 a, b. Crania (? anomala var.) Pourtalesii, inside and outside of the
upper valve much enlarged.

8. Fry of Terebratula cubensis, very much magnified : a, nucleus of the
neural valve with its two pores (b) ; c, constricted neck of the valve beyond
the nucleus ; d, thin septum closing the upper part of the foramen ; /, fora-
men; h, nucleus of the hsemal valve with pores.

It will be observed that, in the closed apical termination, the posterior
lamina partly closing the foramen, and the lateral grooves or areas on earh
side of the latter, this immature form presents characters entirely analogous
to those (upon which Professor King has lately based a genus, Agulhasia,)
exhibited by a species which I am inclined to regard as an immature Tere-
bratulina.

9. Soft parts of the same : a, horseshoe-shaped membrane ; b,f, muscles ;
c, mouth and bag-shaped stomach, with (d) hepatic lobules.

10. Spicule from adult Terebratula cubensis, much magnified.

11. Coecal termination of the intestine of the same, enlarged.

12. Oyaria of Terebratula cubensis, enlarged, showing the position with
relation to the muscles of the neural valve.

13. Same, in the hamial valve.

14. Diagram of the parts about the mouth from the side, enlarged : A,
nasiform body ; B, intercorporeal groove ; C, supra-cesophageal body ; D,
excavation below it and behind the brachia ; E, brachia ; F, mouth ; H,
dorsal adjustor muscles (retractors of Owen) ; I, hepatic lobules ; J, heart
with vessel ; K, side of stomach and intestine, from which the hepatic
lobules have been removed, showing the two ducts, by which the lobules
communicate with the cavity of the stomach.

15. Diagram of the supra-cesophageal portion of the body from in front,

44 BULLETIN OF THE

enlarged : A, dorsal adjusters ; B, frilled portion of oviducts, seen as if
through the transparent tissues ; C, anterior tubular portion of oviducts,
terminating externally in (E) the oblique genital foramen ; D, D', upper
and lower portions of the nasiform body ; F, intercorporeal groove ; II,
supra-cesophageal body ; I, space between the last and the brachia ; J,
stumps of brachia, cut off to expose the other parts; K, oral groove
between the superior and inferior labia ; L, mouth, with the median prom-
inence of the superior labium above it ; M, median notch or sulcus of the
inferior labium (N).

1G. Attachments of the free end of the frilled portion of the oviduct,
much enlarged : A, free end of right oviduct ; B, attachment of the same
to the mesentery (E) ; C, apparent foramen ; D, secondary attachment,
which appears to be of a tendinous consistency and carries a bloodvessel
which enters the oviduct, to all appearance.

Plate II.

Fig. 1. Diagram of the soft parts of Waldheimia floridana, much -en-
larged. In this figure the pedicel, the peduncular muscles, and all the
other muscles except the occlusors, have been removed. The posterior
parts of the mantle lobes are not represented, as they would cover the parts
which it is desired to exhibit. The neural lobe is above and the haemal
lobe below. The pallial sinuses are represented with much more promi-
nence than they actually exhibit. They are really almost invisible and are
exceedingly difficult to trace even under a high power, so that their out-
lines as here given must be regarded as provisional, though they are prob-
ably sufficiently accurate. The parts are represented as they appear (with
the calcareous matter removed by acid), floating in water, with the pedun-
cular end towards the observer. The small genitalia are suspended in the
posterior part of the sinuses. The posterior end of the stomach, with the
heart and hepatic lobules, appear between the occlusors. Below them is
the fold in which the septum of the hamial valve extends half-way to the
margin between the median sinuses. The oviducts are seen in their
proper position. The broad brachial disk, with its appendages so widely
separated, is seen through the transparent membranes. The pallial lobes
are fringed by the very short close-set seta?, inside of which runs the
slender circumpallial muscle.

2. Represents the stomach with its appendages from behind. The
hepatic digitations, obscurely divided into lobes, cover the lower portion of
the stomach, and above them is seen the heart at the junction of the intes-
tine and stomach. On each side are seen the edges of the oral groove or
labia in front.

3. Side view of the same, with the hepatic lobules removed to show the

MUSEUM OF COMPARATIVE ZOOLOGY. 45

openings of the ducts. The position of the brachia and labia with regard
to the esophagus, and of the heart, are shown.

4. Diagram of Cistella var. lutea, representing the animal in the shell,
with the long brachial cirrhi behind the oral groove turned back to show
the groove and position of the mouth. The other cirrhi are curled up in
their natural position.

5. Side view of the stomach, intestine, and hepatic lobules of the same
species, with a section of the oesophagus, showing the position of the cirrhi
and labia. The heart is seen below, just behind the hepatic lobules.

6. Same, with the lobules removed, showing the openings of the ducts,
the position of the heart, and the shape of the mesentery in which the
intestine is suspended.

7 Genitalia of the same, showing how they are suspended from a rib-
bon-like lamella, with the ova in various stages upon the edge of it.

8. Rudimentary folds and simple opening of the oviducts of the same.

9. Diagram of the pallial sinuses of Waldheimia sepligera, with the
ovaria, from a dry specimen. The haemal lobe is above.

Cambridge, May, 1871.

JSull.M.C.Z .Vol III.No.1

PI. I.

■

'•i

^

/ /

•

^£L

1 •;

S.a

6 ^

U

mr

//V

• •

y^

\v

M

«£#

i ]2

\ •

'. k ■ * :

f^

.; :

<v- '

«®

I i -

f'r •>>

to." 8

Nj

13

14

' •

15 J'\— —

16.

-i

'.

BRACHIOPODA

Bull.M C Z.Vol in x ;

PI. II.

-■

X;

/

I

^ i ft- ' '/-r

'

m

■■

I R iCHIOPODA

^0 2. — Application of Photography to Illustrations of Natural
History. With two figures printed by the Albert and Wood-
bury Processes.

No attempts have thus far been made to apply the comparatively
new processes of carbon printing to general illustrations of Natural His-
tory, though excellent figures of microscopic preparations have been
printed by carbon processes by Deane, "Woodward, and others. The
lithographic plate3 of many memoirs on natural history have been
made up with the assistance of photographs, but the want of perma-
nence of the common photographic prints has prevented their use be-
yond that of auxiliaries to lithography. The rapid progress made in
carbon printing by the Woodbury and Albert processes promises to
furnish us, within a short time, the means for direct application of pho-
tography to illustrations of natural history, and these new methods of
printing are likely to replace to a great extent the ordinary litho-
graphic plates.

The accuracy of a photographic illustration is of course far beyond
that of an engraving or lithograph, and as soon as a few practical
difficulties of printing the separate figures of a plate at one impression
are overcome, we shall be able to illustrate our memoirs accurately and
economically, and give figures with an amount of detail which the great
expense of engraving or lithographing would usually make impossible,
even were it mechanically practicable.

The accompanying plates are fair specimens of what can be obtained
by the Albert and by the Woodbury processes, and give a good idea
of the character of the two modes of printing; the one imitating a
common photographic print, the other a lithograph. These figures are
taken from the illustrations of my forthcoming " Revision of the Echini."
The one is a figure (natural size) of Echinocidaris punctidata Desml.
from South Carolina, printed by the Albert process ; the other, a sec-
tion (slightly reduced) of Laganum decagonalis Less, from Hong Kong,

48 BULLETIN OF THE ZOOLOGICAL MUSEUM.

printed by the Woodbury process ; both negatives, of course, being taken
from nature.

Mr. Jno. Carbutt, of Philadelphia, is the superintendent of the Ameri-
can Photo-relief Printing Company (Woodbury process).

Mr. E. Bierstadt, of New York, is in charge of the Photo-plate
Printing Company (Albert process).

Alexander Agassiz.
Cambridge, November 30, 1871.

WOODBURYTYPE A.P.R. PTC. CO., PHILA.

PHOTO-PLATE PRINTING CO.

ALBERT TYPE PROCESS

E. BIERSTADT SUPT. , N.Y.

No. 3. — A Letter concerning Deep-Sea Dredgings, addressed
to Professor Benjamin Peirce, Superintendent United States
Coast Survey, by Louis Agassiz.

Cambridge, Mass., December 2, 1871.
My dear Friend, —

On the point of starting for the Deep- Sea Dredging expedition, for
which you have so fully provided, and which I trust may prove to be
one of the best rewards for your devotion to the interests of the Coast
Survey, I am desirous to leave in your hands a document which may
be very compromising for me, but which I nevertheless am determined
to write in the hope of showing within what limits Natural History has
advanced toward that point of maturity when science may anticipate
the discovery of facts.

If there is, as I believe to be the case, a plan according to which the
affinities among animals and the order of their succession in time were
determined from the beginning, and if that plan is reflected in the mode
of growth, and in the geographical distribution of all living beings ; or,
in other words, if this world of ours is the work of intelligence, and not
merely the product of force and matter, the human mind, as a part of
the whole, should so chime with it, that, from what is known, it may
reach the unknown ; and if this be so, the amount of information thus
far gathered should, within the limits of errors which the imperfection
of our knowledge renders unavoidable, be sufficient to foretell what we
are likely to find in the deepest abysses of the sea, from which thus far
nothing has been secured.

I will not undertake to lay down the line of argument upon which I
base my statement, beyond what is suggested in the few words pre-
ceding, namely, that there is a correlation between the gradation of
animals in the complication of their structure, their order of succession
in geological times, their mode of development from the egg, and their
geographical distribution upon the surface of the globe. If that be so,
and if the animal world designed from the beginning has been the
motive for the physical changes which our globe has undergone, and if,
as I also believe to be the case, these changes have not been the cause
of the diversity now observed among organized beings, then we may

VOL. III. 4

50 BULLETIN OF THE

expect from the greater depth of the ocean representatives resembling
those types of animals which were prominent in earlier geological
periods, or bear a closer resemblance to younger stages of the higher
members of the same types, or to the lower forms which take their
place nowadays. And to leave no doubt that I have a distinct per-
ception of what I may anticipate, I make the following specific statement.

It lies in the very nature of these animals that, among vertebrates'
neither Mammalia nor Birds can exist in deep waters, and if any Rep-
tiles exist there, it could only be such as are related to the extinct
types of the Jurassic periods, the Ichthyosauri, Plesiosauri, and Ptero-
dactyles ; but even of these there is very little probability that any of
their representatives are still alive. Among the fishes, however, I ex-
pect to discover some marine representatives of the order of Ganoids
of both the principal types known from the secondary zoological period,
such as Lepidoids, Sauroids, Pycnodonts, Coclacauthes, Amioids, and
Glyptolepis-like species may even be looked for. Among Selachians
some new representatives of Cestraciontes or Hybodontes may be
forthcoming, connecting the latter more closely to Odontaspis. I also
look forward to finding species allied to Corax, or connecting this genus
with Notidanus, perhaps also Jurassic-like forms. Among Chimreroids
we may expect some new genera more closely related to the extinct
types of that family than those now living. Among ordinary fishes 7
take it for granted that Beryx genera may be added to our list, ap-
proaching perhaps Acanus, or rather Sphenocephaly ; also types allied
to Istieus, to Anenchelum, and to Osmeroides, Elops, and Argentina.
Dercetis and Blochius may also come up. Species of all classes of the
animal kingdom which have been very rarely met with by fishermen
and naturalists are likely to be found in the deepest waters, in which
neither hooks nor nets are generally lowered. Nothing is known con-
cerning the greatest depth at which fishes may live. Upon this point
I hope to obtain positive data.

The Mollusks will, no doubt, afford a rich harvest of novelties, among
which some may be of the deepest zoological interest. It stands to
reason that a variety of Nautiloid Cephalopods may be discovered
when Nautilus proper and Spirula are so rarely found alive, and
among new forms there may be those combining characters of Argo-
nauts with features of Nautilus ; some may even be coiled up like Tur-
rilites. Belemnitic Squids would appear natural. Among Gasteropoda

MUSEUM OF COMPARATIVE ZOOLOGY. 51

we may look for high spired Natica-like types, for representatives of
Acteonella, Avellana, and the like; for small Volutoids of the tertiary
and cretaceous type?, for Rostellarias, even for Nerineas, and more
particularly for forms intermediate between Firula and Cyprea. Among
Acephala I would expect a variety of Myacea approaching those1 de-
scribed in my monographs of that family from the Jurassic and cre-
taceous formations, such as Ceromya, Corimya, Circomya, Goniomya,
Myopsis, etc., with Panorpa and Pholadomya, and others recalling per-
haps also Cardinia, Gresslya, or Cardiacea more closely related to Cono-
cardium than the living species, perhaps leading to Opis, or Trigonue
of extinct types akin to Myophoria, with Pachymya, Diceras Gram-
misia, Inoceramus, Pterinea, Monotis, and Posidonia. Kudistcs should
take the place of oysters, and the harvest of Bz'achiopods should bo
large.

Among Crustacea it is natural to suppose that genera may be dis-
covered reminding us of Eryon or of Pemphyx Gampsonyx, or some
Amphipods, and Isopods aping still more closely the Trilobites than
Scrolls, or Limuloids approaching that extinct family. The classifica-
tion, embryology, and order of succession of Echinoderras is now so
veil known, that it is perhaps still more easy to anticipate the char-
acter of discoveries in this branch of the animal kingdom than in any
other. I expect confidently, to find Spatangoids approaching Holaster,
Toxaster, Ananchytes, Ilemipneustes or Metaporbinus, and others akin
to Eysaster, Echinolamps approaching Pygurus, Xacleolites tending to
Clypeus, Galerites, like Pyrina or Globator, etc., etc., and again Cida-
rids akin to C. glandifera and clavigera with Glypticus-like species,
and Codiopsis, Coelopleurus, Cyphosoma, and Salenia.

Among Starfishes the types of Gcniaster and Luidia are likely to
prevail, with simple rayed Euryaloid genera, and among Crinoids a
variety of genera reminding us of Pentremites, Marsupites, Penta-
crinus, Apiocrinus, and Eugeniacrinus.

The question of the affinities of Millepora will probably receive ad-
ditional evidence, and genera connecting more closely the Bugosa and
Tabulata with one another, and with (he Acalephs may be expected in
the shapes of branching Ileliopores and the like.

With the monograph of Pourtales upon the deep-sea corals before
me, it would be sheer pretence to say anything concerning the prospect
of discovering new representatives of this or that type. His *
point them out already.

52 BULLETIN OF THE

But, there is a subject of great interest likely to be elucidated by our
investigation, — the contrast of the deep-sea faunae of the northern with
those of the southern hemisphere. Judging from what Australia has
already brought us, we may expect to find that the animal world of the
southern hemisphere has a more antique character, in the same way as
North America may be contrasted witli Europe, on the ground of the
occurrence in the United States of animals and plants now living here,
the types of which are only found fossil in Europe.

A few more words, upon another subject. During the first three
decades of this century, the scientific world believed that the erratic
boulders, which form so prominent a feature of the surface geology of
Europe, had been transported by currents arising from the rupture of
the barriers of great lakes among the Alps, or started from the north
by earthquake waves.

Shepherds first started the idea that within the valleys of Switzer-
land these huge boulders had been carried forward by glaciers, and
Swiss geologists, Venetz and Charpentier foremost among them, very
soon proved that this had been the case. This view, however, remained
confined to the vicinity of the Alps in its application, until I suggested
that the phenomenon might have a cosmic importance, which was
proved when I discovered, in 1810, unmistakable traces of glaciers in
Scotland, England, and Ireland, in regions which could have had no
connection whatever with the elevation of the Alps. Since that time
the glacial period has been considered by geologists as a fixed fact,
whatever may have been the discrepancies among them as to the ex-
tent of these continental masses of ice, their origin, and their mode of
action.

There is, however, one kind of evidence wanting to remove every
possible doubt that the greater extension of glaciers in former ages was
connected with cosmic changes in the physical condition of our globe.
All the phenomena related to the glacial period must be found in the
southern hemisphere with the same characteristic features as in the
north, with this essential difference, that everything must be reversed;
that is, the trend of the glacial abrasion must be from the south north-
ward ; the lee side of abraded rocks must be on the north side of hills
and mountain ranges, and the boulders must have been derived from
rocky exposures lying to the south of their present position. Whether
this is so or not has not yet been ascertained by direct observation. I

MUSEUM OF COMPARATIVE ZOOLOGY. 53

expect to find it so throughout the temperate and cold zones of the
southern hemisphere, with the sole exception of the present glaciers of
Tierra del Fuego and Patagonia, which may have transported houlders
in every direction. Even in Europe, geologists have not yet sufficiently
discriminated between local glaciers and the phenomena connected with
their different degrees of successive retreat on one hand, and the facts
indicating the action of an expansive and continuous sheet of ice moving
over the whole continent from north to south. Unquestionably, the
abrasion of the summits of the mountains of Great Britain, especially
noticeable upon Schiehallion, is owing to the action of the great Euro-
pean ice-sheet during the maximum extension of the glacial phenomena
in Europe, and has nothing to do with the local glaciers of the British
Isles.

Among the facts already known from the southern hemisphere are
the so-called rivers of stone of the Falkland Islands, which attracted
the attention of Darwin during his cruise with Captain Fitzroy, and
which have remained an enigma to this day. I believe it will not be
difficult to explain their origin in the light of the glacial theory, and I
fancy now they may turn out to be nothing but ground moraines, similar
to the " Horsebacks " of Maine.

You may ask what the question of drift has to do with deep-sea
dredging? The connection is closer than may at first appear. If drift
is not of glacial origin, but the product of marine currents, its formation
at once becomes a matter for the Coast Survey to investigate, and, I
believe, it will be found in the end, that, so far from being accumulated
by the sea, the drift of the lowlands of Patagonia has been worn away
to its present extent by the continued encroachment of the ocean in the
same manner as the northern shores of South America and of Brazil
have been

Hoping some, at least, of my anticipations may prove true,
I remain, ever truly yours,

LOUIS AGASSIZ.
Prof. Benjamin Peirce,
Superintendent Coast Survey, Washington, D. C.

No. 4. — Preliminary Notice of a few Species of Echini,
by Alexander Agassiz.

The following species are briefly noticed in advance of the fuller
descriptions of the Illustrated Catalogue to complete the synonymy of
the species of Echini in the collection of the Museum.

Strongylocentrotrus armiger.*

Test thin, flattened above, regularly arched below ; is at once distin-
guished from its congeners by its peculiar short, blunt, stout spines,
swollen in the middle of the shaft, resembling those of Echinometra
oblonga. The large primary tubercles, arranged in two principal verti-
cal rows, in both areas, flanked by indistinct vertical rows of secondaries,
are all placed nearer the abactinal pole on the depressed portion of the
test, decreasing rapidly in size towards the actinostome. Genital pores
large. Poriferous zone broad, pores arranged in arcs of from six to
seven pairs.

Free Public Museum (Liverpool) ; Brit. Museum; Mus. Comp. Zool.
— Australia.

Sphserechinus Australise. t

The Museum has received at various times, from Mr. Henry Edvvards,
a sea-urchin intermediate between Psammechinus and Strongylocen-
trotrus, remarkable for the thickness of its test, its compact abactinal
system, the uniform size of the primary tubercles ; the secondary tuber-
cles closely packed and filling completely the whole space between the
primaries, both in the ambulacral and interambulacral areas. The test
is globular, resembling, in outline, somewhat Amblypneustes, has no am-
bitus, a very small actinostome, with short, sharp, narrow actinal cuts.
The poriferous zone divided by secondary tubercles into an inner verti-
cal row of pairs of pores far above one another, and an outer part with
pores arranged in from three to four pairs, all the pores deeply sunken
between the secondaries. Actinal membrane covered by large elliptical

* The species included in this genus by Brandt are, S. chlorocentrotus = Echinus
Drbbachiensis Mull (teste Grube), and the specie* of section D of Blainville, including
K. lividus, parvituberculatus, .... ; in a different section lie places E. tuberculatus
Blainv. V

t Cryptopora Australia1 Michelin MS. Kcole des Mines, Paris.

56 BULLETIN OF THE

plates moderately closely packed. The spines are short ; in dried speci-
mens, violet at base tipped with green, test violet with greenish tubercles.
Ecole des Mines; Mus. Comp. Zool. — Australia.

Amblypneustes pentagonus.

Unlike the other species of Amblypneustes, the outline of the test

from above is pentagonal ; the ambulacra projecting considerably beyond

the. concave interambulacra. The coronal plates are high, not half as

numerous as in other species of the genus of the same size. There is

but a single primary vertical row of tubercles both in the ambulacral

and in the interambulacral spaces ; secondary tubercles few in number,

very irregularly scattered ; sutural pores small, limited to angle of

plates; test thin, high, remarkable for the great size of the primary

spines. Abactinal system delicate, and not prominent and stout as in

other allied species.

Mauritius ?

Amblypneustes inflatus.*

Test moderately stout, nearly spherical ; poriferous zone more than
equalling in width the median ambulacral region, where the tubercles
form irregular horizontal rows of from two to three tubercles for each
plate. In the interambulacral space there is for each plate a larger
primary, forming a distinct vertical row, and from one to six smaller
tubercles on each side of the median interambulacral line forming
horizontal rows, and very indistinct vertical rows. No bare median
ambulacral or interambulacral spaces.

New Holland.

This as well as the following species belong to the section of Ambly-
pneustes (Holopneustes),t in which the poriferous zones are arranged
in three vertical rows of pores, the middle row often extremely sporadic,
as in H. porosissimus, and having no sutural pores.

Amblypneustes purpurascens.*
Actinal and abactinal diameter equal, poriferous zone equal in width
to the corresponding tuberculiferous ambulacral space. In large speci-

* Liitl-cn MS. in lift.

t It is possible that these two species may turn out to be only different stages of
growth of II. porosissimus. The variation in the other species of Amblypneustes
is very great, but here is confined to the poriferous zone, which makes it difficult to
decide the point for want of material.

MUSEUM OF COMPARATIVE ZOOLOGY. 57

mens the middle row of pores is very irregular, forming zigzag lines,
while in younger specimens the middle row is nearly as uniform as the
inner and outer rows; the latter is characterized hy the greater size of
the inner pore, and the distance separating the pores of a pair. The
tubercles of both areas form most regular horizontal rows, as many as
ten tubercles of nearly uniform size on each side of the median line,
and from one to three for the ambulacral spaces, forming no distinct
vertical rows.

Mus. Copenhague; Mus. Hamburg; Mus. Comp. Zool. — New Hol-
land.

Spatangus Liitkeni.*

Closely allied to Spatangus purpureus, test relatively more convex
and abactinal diameter greater. Seen in profile test nearly semi-
circular, somewhat more curved anteriorly ; anal plane lower and more
oblique than in S. purpureus, the subanal fasciole belonging properly
to the actinal surface. The shape and proportions of the ambulacra, as
well as the distribution of the large tubercles between the ambulacra, is
not materially different from S. purpureus. The shape of the subanal
fasciole (having three pores on each side) is very different from that of
S. purpureus (having but two pores) ; it is in the latter species about
three times broader than long, while in S. Liitkeni it is only one and a
half times broader than long, in specimens of about the same size. The
color in alcohol is dark violet.

Mus. Copenhague ; Mus. Comp. Zool. — China Seas (Salmin).

Lovenia cordiformis.t

Differs from the other species of the genus by the great convexity of
tile test, the anterior vertex, compared to L. hystrix, the relatively
smaller number of primary tubercles in the anterior interanibulacfal
spaces, the narrow band of large tubercles upon the edge of the actinal
surface, the small size of the purses of these tubercles, the large size of
the tubercles within the peripetalous fa>ciole on each side of the anterior
ambulacrum near vertex, the flatness of the test round mouth not
forming a prominent posterior lip.

Mus. Comp. Zool. Guayamas ; Smith. Coll. Gulf of California.

* Sp. altus Liltken MS. in litt.
f Liitken MS. in litt.

58 BULLETIN OF THE ZOOLOGICAL MUSEUM

Mcera stygia.*

In the British Museum and Copenhague Museum are specimens of a
Moera allied to M. atropos, holding to it the same relation which Echi-
nocardium gibbosum holds to E. cordatum. The test has a short, lon-
gitudinal diameter, is remarkable for the great height of the abactinal
axis, the sharply truncated anterior extremity, the narrow lateral am-
bulacra, the elongate anal system, and the prolongation of the anal
plastron in a sharp keel.

A specimen in the Mus. Comp. Zool. is said to have come from the
Red Sea.

Rhynobrissus pyramidalis.

I am indebted to Mr. Thomas J. Moore for a specimen of this remark-
able Spatangoid. It is allied to Brissopsis, having like it a peripetalous
fasciole (narrow), a subanal and anal fasciole (broad) ; it has, however,
no anterior groove, the test being uniformly convex anteriorly, the vertex
is posterior slightly in advance of the bevelled posterior extremity. The
anal plastron is heart-shaped, somewhat as in Metalia, projecting like a
keel beyond the outline. Seen from above, the outline of test is dia-
mond-shaped, rounded anteriorly. The ambulacral petals but slightly
sunken, resembling those of Faorina. The actinostome is crescent-
shape, very narrow, extends well across the test, immediately sur-
rounded by a broad, bare space, which forms rapidly narrowing bare
ambulacral spaces on each side of the narrow elongate actinal plastron.
The spines of the lower surface are long, curved, and slender, while
those of the rest of the test are short, hair-like ; their coloring is light
violet. The anal fasciole is open above the anal system, but a secondary
subanal plastron is formed independent of the principal one by a broad
band passing below the anal system, slightly above the origin of the
anal fasciole, — a feature which thus far has not been noticed in any
other genus of Spatangoids, finding its parallel only in the double branch
of the anterior part of the peripetalous fasciole of Faorina.

Free Public Museum (Liverpool) ; Mus. Comp. Zool. — Linguin.
China Seas.

Cambridge, January 10, 1872.

* Liitken MS. in litt.

No. 5. — Fossil Cephalopods of the Museum of Comparative
Zoology. Embryology, by Alpheus Hyatt.

The researches recorded in the following pages were originally un-
dertaken in order to ascertain the limits of the embryological period
among the typical Ammonites.

During this period, which begins with the ovisac, the different species
possess a common form, and are very similar in the characteristics of
their septa, siphons and shells. It was at first proposed to give the
descriptions only in conjunction with the different groups to which
the young belonged ; but the intimate connection and importance of the
facts, elicited by a general comparison of the young of Nautilus, Goni-
atites, and Ammonites appeared to demand a separate publication.

The necessary illustrations have been furnished with unstinted liber-
ality by the Director of the Museum, Professor Louis Agassiz. The
collections also, which are extremely rich, have been placed entirely at
my disposal, and I have been permitted to break up whatever speci-
mens were considered suitable for the present purpose.

As this is my second formal publication upon the Cephalopods in
the Bulletin, it is only becoming to correct certain errors which are to
be found in the first (No. 5, Vol. I, Bulletin of the Museum of Comp.
Zoology), under the same general title as this number.

Proper credit was not given in the preface to Professor Edward
Suess for having been the first to publish the fact, that the typical forms
of Ammonites were capable of generic division, and two of his names,
Lytoceras and Phylloceras, should supersede two of those given in that
number of the Bulletin, namely Thysanoceras and Rhacoceras.

I have been rather severely criticised by Laube and Zittel for giving
Professor Agassiz the credit of having been the first to perceive that
the Ammonites were divisible into distinct families and genera, but it will
be noticed that this is given to him as a personal matter between an
instructor and his student. This I must be excused from withdrawing.
But I did know, however, that ever since Professor Agassiz published
his French translation of Sowerby's Mineral Conchology, he has re-
garded the Ammonites, not as a family, as Suess does, but as a large
group, perhaps equivalent to a sub-order, and composed not of a few
genera, but of several families containing many genera.

60 BULLETIN OF THE

The fact that two of the genera were precisely equivalent to those
of Professor Edward Suess ought to have secured some credit, at least
among his admirers, for the twenty-odd other precisely equivalent
genera established by Professor Agassiz and myself; but such has not
been the fact.

Either we knew of Suess' publications and patterned after them,
only increasing the number of genera, or else the investigation was in-
dependent. Besides the fact, well-known to palaeontologists in this
country, that these investigations began some years before Suess had
published anything, there was no object in concealing our acquaintance
with his researches. On the contrary a prompt acknowledgment
would have been of great advantage, since it would have been easy to
show, that, in our much more complete classification, the genera were
founded upon the same system of characteristics as those used by
Suess to distinguish Lytoceras from Phylloceras.

There has been nothing besides the above in the criticisms advanced
except general statements of disapproval, which, of course, cannot be
dealt with specifically and are of no imj)ortance.

I will say, in conclusion, that I have nearly finished a very thorough
review of the same groups without being able to effect any very
material alterations.

The same divisions are found to be sharply defined divergent series,
and whatever name or value may be given them, they are natural
groups ; this being so, my object is attained.

This object is, as in the present publication, to obtain some faint in-
sight into the laws of descent of these forms one from another, by
means of such indications as may be afforded by a close study of the
developmental and adult characteristics, corrected or verified by the
observed geological positions of the species.

Every one who has studied the coiled shells of Cephalopods and
Gasteropods, is aware that they retain in the interior of the umbilicus
the younger whorls, which are necessarily more or less covered up and
protected by those of later growth. By breaking away the older
whorls, one can therefore eventually arrive at those portions of the
involved cone which represent the very youngest periods of growth.

This in Ammonites and Goniatites has been shown to be a minute
globular sac. In Nautilus, however, this sac is not retained, but traces
of its former existence are apparent on the apex of the first whorl, in

MUSEUM OF COMPARATIVE ZOOLOGY. 61

the form of a scar or cicatrix. The shell of the neck of the ovisac, in
Ammonites and Goniatites, has the same form as the shell of the first
whorl, whereas in Nautilus it is evident that the shell of the ovisac
roofed over the living chamber, somewhat as in Gomphoceras. The
opening into the apex or neck of the ovisac, where it joined the first
whorl, was, as shown by the cicatrix, a narrow vertical slit bearing a slight
resemblance to the opening of the living chamber in Gomphoceras.

All of the animals of these three groups undergo a true metamor-
phosis in passing from the globular ovisac into the first whorl.

The first septum occurs at the junction of the ovisac and the first
whorl, in both Goniatites and Ammonites, and has an entire abdominal
cell with two simple lateral lobes as in Nautilus.

The second septum in the Jurassic species of Ammonites, which
alone were examined,* had no positive resemblance to the adult of any
species of Goniatites, but the entire, simple sutures and simple ventral'
lobes have a general resemblance to that type, though the superior lat-
eral lobes are present and undeniably Ammonitic even at this early age.

The second septum of the closely-coiled Goniatites has a much
shallower ventral lobe than in Ammonites, and no superior laterals.
The sutures resemble those of the adults of the Silurian Goniatites,
but the ventral lobe is shallower and broader. The superior lateral
lobes which are first seen in the third septum have the sharp outlines
which distinguish the Goniatites.

This first whorl, or apical portion of the shell, is closely coiled in
Ammonites, more or less involute, and resembles the adults of the typi-
cal Goniatites in form.

The form of the first whorl in Goniatites as well as the development
of the septa varies excessively in the different species, or even in
different varieties of the same species. In some the form and septa
must be very similar to those of the straight Nautiloids, such as Ortho-
ceras or Endoceras, while in others we have representatives of the
arcuate Cyrtoceras, and finally those which are closely coiled, involute,
and hardly distinguishable in external appearance from the young of
Ammonites. This variability is, so far as we know, greater in the
earlier or Silurian, than in the later or Devonian and Carboniferous

* Species of simpler adult structure and earlier occurrence, among the Ceratites or
Clydonites, would probably present an aspect more like some of the adult Goniatites in
their young.

62 BULLETIN OF THE

species, and prepares the observer to find a common fixed type of form
in the young of the Jurassic species of Ammonites, •which belong to
the same order and are intimately connected by transitional genera.

Everywhere the closest correlation is traceable between the amount
of coiling, the amount of involution or envelopment of the whorls,
the form of the whorl, and the development of the septa, — the simpler
sutures being invariably associated with shallow concave slowly chang-
ing septa, an elliptical form, and open coils. The more complicated
sutures, whose characteristics change more cpjickly in course of growth,
are combined always with a more or less crescent-shaped whorl, and
closely embracing or involving coils, whether found among Nautiloids
or Ammonoids.

This agrees with the correlations of the structure and morphology
of the adults as worked out by Bronn, Barrande, and others, among
Nautiloids and Ammonoids, from the straight Orthoceras to the coiled
Nautilus, and inversely, among Ammonoids, from the closely coiled
Goniatites and Ammonites to the straight Baculites ; the general
morphology being readily and accurately expressed as a coiling up
of a straight cone and the subsequent uncoiling of the same at later
stages of the earth's history. The shells are almost universally classi-
fied in accordance wTith this coiling and uncoiling, with which also the
structure of the siphon and septa are more or less correlative.

The siphon in Nautilus, Goniatites and Ammonites terminates, or
rather begins, with a blind sac, or siphonal coecum.

In Nautilus, this occupies a central position somewhat nearer the
ventral than the dorsal side, and is enclosed by the shell at the apex
of the first whorl. In Goniatites and Ammonites the siphonal coecum
has a cone-like prolongation, which appears to open into the bottom
of the siphonal coecum in Goniatites. The siphon is developed ear-
lier in the last two than in Nautilus, since it is found within the neck
of the ovisac.

The ovishell consists, in Ammonites and probably in Goniatites, of an
inner lining layer similar to that of tic chambers of the whorls, and an
outer thicker layer. They both extend entirely around the ovisac, but
the outer layer is very much thicker on the sides and abdomen. The
inner layer bends inward and forms part of the under side of the first
septum. The outer layer overlaps the inner and outer layers of the
abdomen and sides of the first whorl, showing that here was the mouth

MUSEUM OF COMPARATIVE ZOOLOGY. 63

of the ovisac, and that the outer layer of the ovisac corresponded with
both the inner and outer layers of the first whorl. Both of the latter
are continuous with the outer layer of the ovishell, in the same sense
that they are subsequently continuous with each other whenever the
growth is interrupted, and an imbricated suture or break is formed.

The shell of the young and mature whorls is composed of an inner
and an outer layer, the latter being divided into two portions, the
external or colored stratum, and the internal or white stratum. In
Nautilus both the layers extend entirely around the shell in the young,
but as soon as one revolution of the whorl is completed, the black
excretion of the hood replaces the outer layer to a considerable extent
on the dorsal or involved side. The outer layer is still present, but is
very thin.

In Goniatites and Ammonites the same layers are present, but do
not extend around the dorsal side, except in the loosely coiled young
of certain species.

In Nautilus, Goniatites, and Ammonites a thin layer lines the interior
of each chamber and coats the exterior of the siphon. This alone
appears upon the dorsal or involved portion of the whorls in the last
two groups.

All the layers, with the exception of the black layer of the hood and
lining layer, betray everywhere an imbricated structure which shows
that they were deposited by the edge of the mantle from within. The
lining layer, and the black deposit of the hood are continuous through-
out wherever they occur.

The siphonal coecum of Nautilus is formed by the funnel of the septum,
which projects posteriorly until it comes in contact with the shell of the
apex. The funnels of the younger septa are longer than those of the
older septa, and the sheath of the siphon which springs from the funnel
of the second septum extends posteriorly and lines the interior of the
siphonal coecum, forming a second siphonal coecum within the first.

In Goniatites and Ammonites the coecum is formed in a similar
manner, but the latter is not lined, the siphonal funnel and sheath of
the second septum extending only far enough to close the opening in
the first septum.

Subsequently the siphon is composed of the funnel and the sheath
which is continuous with it, but has a looser and more porous texture.
The sheath is discontinued at the opening of the funnel of the next

G4 BULLETIN OF THE

younger septum, into which it closely fits and thus forms the outer wall
of the siphon. Internally is found the corneous layer, which is con-
tinuous throughout.

In Nautilus and Goniatites, and in the youngest stages of Ammonites,
we find only the parts described above, but in the further development
of Ammonites, another is added. At first only a portion, and lastly,
nearly the entire thickness of the septum bends in the opposite or an-
terior direction, forming a loose collar around the siphon, but not built
into it, as is the shorter funnel below.

The siphonal coecum is close to the abdomen in Goniatites and Am-
monites, but the siphon, which springs from the neck of the coecum as
it passes through the second septum, diverges nearer to the centre of
the whorl.

This position of the. siphon, its structure, more especially in the
young Nautilus and the prevalence of shallow concave septa, the Nau-
tilian character of the fust septum in Goniatites and Ammonites even
in the closely coiled young, the prevalence of simply arcuate forms in
the young of several Devonian, and the straightened first whorl of the
varieties of two Silurian species, the continuity of the layers and their
thickness on the dorsal side of the young, are all characteristics which
appear to converge towards the structure and form of the siphon and
shell in the Vaginati, especially Endoceras (Orthoceras) duplex of De
Verneuil and Keyserling.

It is in this group, therefore, or in some closely associated genus that
we must look for the ancestors of the Tetrabranchiate Cephalopods.

I select Endoceras rather than Beatricea, which I have always
regarded as the prototype of all the Nautiloids, because there still
remain several characteristics in the structure of this form which must
be more thoroughly worked up, before the affinities can be settled
beyond a doubt.

The structure of the septa, and of the siphon in Endoceras is not so
irregular or vesicular as those of the shell in Beatricea, but there is a
wonderful resemblance of the cup-like internal chambers of Beatricea
to a line of siphonal cceca. The young of Endoceras, if this view
obtain, would be expected to resemble, or to show some approximation
to, Beatricea. It ought either to be alone composed of thick conical
septa of the adult siphon, and each of these terminating at the apex in
a closed sac, or, at least, the siphon should be larger proportionately in

MUSEUM OF COMPARATIVE ZOOLOGY. G5

the young, and the true septa have a much smaller comparative area
than in the adult.

The outlines and coarser shading of all the figures were drawn with
the aid of the Camera Lucida hy me, and these were worked over, and
artistically finished with the greatest care by Mr. Konopicky, Draughts-
man to the Museum. These outlines were often repeated and verified
in obscure or doubtful cases, and the camera drawing invariably cor-
rected by the careful study of the more minute structural details. The
statement of the number of diameters to which a specimen has been
magnified does not show that this power alone was used ; frequently in
working out the minor details much higher powers were employed, the
results being noted in the general outlines already drawn with a lower
power.

The objectives employed were made by Tolles & Wales, in their
best manner, the glass most relied upon being a one-half inch of re-
markably good definition, and working distance made by the first
named.

The enlargement of the figures was measured by a direct compari-
son of the Camera Lucida image, at a distance of ten inches from the
eye-piece with the size of the object on the stage as shown by a gradu-
ated ruler, or in the higher powers by a stage micrometer.

As others may possibly desire to repeat these observations, I will
conclude this preface with a few words upon the preparation of speci-
mens intended for examination.

Those specimens in which the shell is replaced by iron pyrites or in
which the matrix is stained with iron, are seldom well suited for the
observation of the very earliest period. The centre does not generally,
in my experience, break out as well as in those which are fossilized
by the ordinary Carbonate of Lime, and they are too opaque. Speci-
mens with which I have been most successful are found in the hard
dark blue limestones. The centres of these are usually filled with
translucent carbonate of lime even when the older parts of the whorls
are opaque.

The first process of reduction may be generally effected by breaking
away the whorls with chisels of suitable size-, care being taken not to
strike too near the centre; then the young shell should be bedded in
Canada balsam which has not been overheated. If scorched, the
balsam is apt to give way in large splinters and fly off the slide, carry-

VOL. III. 5

G6 BULLETIN OF THE

ing the minute ovisac with it. The removal of the last three or four
whorls is best accomplished under a dissecting microscope, with the aid
of the long-handled instruments used by dentists. These are chisels
of various forms, bent at the ends so that they can be held horizontally;
and almost every desirable point can be purchased, or obtained by
grinding down those commonly used.

Great care must be taken in grinding down specimens for trans-
parent sections not to get them too thin. After a certain thinness is
attained, the carbonate of lime is apt to crack in a very irregular
manner, obscuring the structure and seriously enhancing the difficulty
of making accurate observations.

The siphon is hardly ever at all points in the same plane, so that the
best practice in grinding the first side is not to pass beyond the outer
limits of this organ. Then, when the specimen is turned and receinent-
ed to a clean slide, the reduction of the exposed portion may be carried
forward to any desired level. The whorls are very apt to break off or
loosen during this last process, and the delicate crystals of lime on the
edges to become fractured, and therefore in most cases it is better to
allow a number of the earlier whorls to remain attached to the ovisac
as a support for those which are to be examined.

EMBRYO.

It has been customary to consider that one of the differences existing
between Goniatites, Ammonites, and Nautilus was to be found in the
absence of a globular ovisac at the beginning of the whorls in the two
latter. This view is very plainly expressed by D'Orbigny, Guido and
Fridolin Sandberger,* and Barrande ; but how this mistake could have
been made among Ammonites it is difficult to understand, unless indeed
the specimens the last-named authors examined, Amm. Icevigatus and
comphnatus Rein., had really lost their ovisacs.

* " Bei der Unterscheidung der Gattung Goniatites von Ammonites ist dieses Merk-
raal nicht unwesentlich. Es scbeint niimlich, class den Ammoniten einc in dieser ab-
gesclinlirten Kugelgestalt sich erhaltende Anfangsellc nicht ziikommc. Mehrere
wohlerhaltene Arten, unter anderen Ammonites kevigatus und complanatus Rein, aus den
untercn Oolith von Thurnau zeigten schon von der Anfangselle aus regelmassig kegel-
formiges Anwachsen ohne jene Abschnurang." — Dk Versltinerung von Nassau. G. und
F. Sandberger. Wiesbaden, 1S50-5G, p. 59.

MUSEUM OF COMPARATIVE ZOOLOGY. 67

Saemann, in Dunker and Meyer's Palajontographica,* has shown
that Lytoceras Jimbriatus had a distinct globular ovisac, and figured the
young of this species in comparison with the pointed young of Nautilus
atratus.

D'Orbigny describes the aspect of the apex of the whorl in Nautilus
as an obtuse cone, and in Nautilus lineatus of the Jura, writes of this
cone as resembling a Patella. Barrande, in criticising this view, re-
marks very justly, that the cone must have been at one time the living
chamber of the animal, and as this must have extended for some dis-
tance, there are no grounds for comparing it with the flattened cone of
Patella.

When observed at the bottom of the umbilicus, the ovisac of the
Ammonites appears as an oval body, generally more or less denuded
of the shell, which breaks away with the matrix.f When the whorls
which encompass it are removed, the ovisac is seen to be much larger
than these exposed lateral areas, which are merely extreme portions
of the narrow sides of the embryonic shell or ovisac* The entire
form, when seen from the side, is that of a very broad symmetrical
oval, flattened considerably on the abdomen, § and lenticular when
viewed from the abdominal or dorsal sides. If

These outlines may vary considerably in the same species. In some
specimens of Deroceras planicosta the abdomen is flatter than in others,
and often depressed as it nears the sides, giving the latter a remarkably
bulging aspect.**

Among the Arietidge, the species examined, Arnioceras semicosta-
tum,ft and Asteroceras obtusum \\ presented no considerable variation,
except such as would naturally result from difference of size in the
species.

Even in the abnormal forms, Scaphites and Crioceras, the ovisac fills
the fundus of the umbilicus, or, in other words, is closely enveloped by
the first whorl. Whether the young of the still more uncoiled genera,

* Vol. Ill, p. 158, pi. 19, Figs. B, C.

f Plate I, Figs. 7, 8. Plate II, Fig. 9.

% Plate I, Fig. 6, A. Plate II, Fig. 7.

§ Plate I, Fig. 4.

1F Plate I, Figs. 1, 2.

** Compare Fig. 2 with 5, and Fig. 6 with 1 on Plate I.

ft Plate II, Figs. 8, 9.

%X Plate II, Fig. 11.

GS BULLETIN OF THE

especially Baculites, have young which are straight or closely coiled, I
could not ascertain hy direct observation.

The ovisac of Goniatites does not, in the majority of the species,
differ very materially from that of Ammonites. It is evidently an
elliptic or globose body more or less flattened on the abdominal side as
in the typical Ammonites* It however presents very remarkable
differences in the relations of the ovisac to the first whorl and to the
umbilicus, and in its variability of form in the same -and different
species.

Dr. Guido Sandberger figured several species, all of which are re-
produced in the first Plate of this paper,f and from these and others,
also examined but not figured, he inferred that the species themselves
might be distinguished by the differences of the embryos. The
figures certainly appear to justify this remark, and my own observa-
tions, as well as those of Barrande, though apparently contradictory,
are really strongly confirmatory. Barrande found in the silurian of
Bohemia Goniatites fecundus, and Goniatites plebius with two well-
marked varieties, and was so fortunate as to obtain also the very
youngest stages of growth in each variety of the former, j These are
also reproduced on Plate I of this work.§ According to Barrande
the ellipticity of the orbit in the adult of one variety is due to the
straightness of the first portion of the first whorl, and the regular
spiral of the other variety to the close coiling of the young. He also
observed and figured the same elliptical form at all stages of growth.
This would seem to be sufficient to establish the two varieties as dis-
tinct species ; but the presence of all intermediate varieties between
the two produced here, the similarities of the adult shells, and the
occurrence of the same variations in Goniatites plebius forbids their
separation.

Goniates crenistria, in one so-called variety, has a decidedly rounded
abdomen, and the ovisac fills the fundus of the umbilicus, while in the

* Plate III, Figs. 3 4.

f'Diese ist stets stark aufgebliiht und wie die Figuren 26 bis 33 (PI. I, Figs. 11-
18) beweisen, bei don verschiedenen Arten oft von sehr characteristischer Gestalt, so das
sie in manchen Fallen ein zur Unterscheidung der Species geeignetes Merkmal abgiebf."
— Guido Sandberger, Beob. Uber Gonialilen, Jahrbuch der Nassau Verein, 1851.

J Syst. Sil. de BohGme. Cephalopods. Vol. II, pp. 32 and 37. Plates 7, 10, 11, 17
and 5, G, 7,241, 242.

§ Plate I, Figs. 9, 10.

MUSEUM OF COMPARATIVE ZOOLOGY. 69

other, with a flatter abdomen, it lies against one side of the first whorl,
occupying but a very sm^ill portion of the entire space.* Here there
can be but little doubt of the specific value of the differences which are
manifested, not only in the youngest stage, but throughout life, and
affect not the symmetry of the orbit of revolution, but the form of the
whorl itself.

In all of the seven species observed by Sandberger, as may be seen
by his figures f and in the four distinct species observed by me, count-
ing Gon. crenistria as two good species, the differences of the ovisac!
are very distinct, whereas in several individuals of the same species,
Goniatites crenistria, the so-called variety with a flat abdomen, which I
have observed, and in several of Go. atratus, no such differences were
observed between the different individuals. The slight differences in
the amount of coiling between the young of varieties latidorsalis and
ealeuliformis of Go. lamed, figured by Sandberger,} were probably
therefore unusual among the Devonian and Carboniferous species.

The succession also indicated by the foregoing facts is precisely
what might have been anticipated from the general morphology of the
Tetrabranchiate Cephalopods.

The range of form has been among the Nautiloids from the straight
Orthoceratite through intermediate arcuate genera, to the partially
coiled Lituites, and finally the closely coiled Nautilus. Such being the
case, if there is any truth in the doctrine of evolution, we must expect
to find some reference to the peculiarities of the parent Nautiloid stock
in the earlier stages of development among the Ammonoids. And
further, as a direct and unavoidable corollary of the above, we ought to
find this reference more distinct in the young of the earlier species of
Ammonoids, the Goniatites of the Silurian, and less noticeable in the
Goniatites of the Devonian and Carboniferous, and, finally, almost
obliterated, or, at any rate, still less distinct in the typical Ammonite
of the Jura.

So far as the facts have been ascertained, they all point in this direc-
tion. The simple Nautiloid-l'.ke. Goniatites of the Silurian may ex-
hibit an Orthoceratiticor straight form, or be closely coiled in the young
of different varieties of two distinct species. A species therefore, on
this horizon, may have a range of variation in form, during the earlier

* Plate III, Fig. 7. f Piute I, Figs. 11, 18. J Plate I, Figs. 15, 1G.

70 BULLETIN OF THE

stages of development, equivalent to that occurring among the adult
forms of Nautiloids from Orthoceras to Lituites.

The Goniatites of the Devonian, however, even in such simple
species as Go. compressus,* exhibit only arcuate whorls, and in the
majority of cases are more or less completely coiled. No elliptical
varieties have been observed in the adults, and the variations of form
in the young of the same species probably do not exceed what has
already been observed. |

Among the typical Ammonites of the Jura, not only is the young of
the same species invariably similar, so far as the coiling and form of
the ovisac is concerned, but the young of all the closely coiled or
normally formed species are also closely coiled and involute or en-
veloped. I

We find in this a perceptible acceleration in the development of the
young precisely proportionate to the estrangement, either in time, or in
adult organization, of the Ammonoids from their supposed parent stock.
There are evidently two tendencies at variance with each other: one
strongly reversionary, appearing in the frequency with which the ear-
lier Goniatites repeat the parent form in certain isolated instances in
the young of varieties, and in the different species of the later Gonia-
tites manifesting itself in the arcuate cone of the young of Goniatites
compressus and others of the Nautilini, and in the closer, though non-
involute coiling of the young of other forms. Evidently this tendency
is losing its power to affect and modify the organization, or, in other
words, its prepotency. The other tendency, which is expressed in the
closer coiling of the whorls, and finally in their increasing involution, is
decidedly progressive, increasing in power to the final and ultimate
extinction of all reference to the ancestral type, except in the internal
organization. Here, as will be shown, the siphon for a limited time
remains central in the first whorl, and the first septum has a large
entire abdominal cell, and simply concave lateral lobes, as in the Nau-
tiloids.

The form, however, of the first whorl of the Ammonoid is like
Goniatites, the shell similar, and the second septum has the invariable
abdominal lobe, superior lateral cells, and lobes of the simpler adult

* Plate I, Fig. 11.

t As already noticed between Figs. 15 ard 16 of Goniatites lamed. Plate I.

t Plate I, Figs. 2, 4, 5, 6, 7, 8. Plate II, Figs. 7 - 9, 11.

MUSEUM OF COMPARATIVE ZOOLOGY. 71

Goniatites, but not by any means of the simplest Goniatites. The
simplest adult Goniatites have no proper lateral cells, but only broad
lateral simple curves to the septa, as if the first septum of the Ammonite
was modified or broken by a small abrupt lobe on the abdominal
side. Contrast this with the development of the septa, and their
gradual change in Goniatites compressus, and we see at once that the
development of the same parts is very much quickened or accelerated
in the typical Ammonite.

That this acceleration of development is due to the prepotency of
the same progressive tendency as the closer and closer coiling, and
final involution of the ovisac, by the first whorl, can hardly be doubted.
Thus, not only in the whole series of Nautiloids are the forms more
or less completely coiled and finally enveloping, but in the young
Ammonoids this process is repeated, but only as a reversionary ten-
dency of individuals and species, or at most, perhaps, by the group of
Nautilini.

In the Arietida?, and many other groups of typical closely coiled
Ammonites, the same process is repeated to a greater or less degree in
nearly every series of species, the progress being from a non-involute,
or slightly involute, to a more involute form, and even in varieties of
species there is occasionally a marked difference in the degree of in-
volution of the adult whorls. Everywhere, throughout the order of
Ammonoids, we meet with this constant repetition without reference to
the geographical position or distribution of the species.

This increase of involution is, of course, due to the extension of the
sides of the whorls inwardly, and is invariably accompanied by a
decrease in the lateral or transverse diameter of the whorl, flattening
of the sides, and a corresponding elevation of the abdomen.

In those series, however, such as the Dactyloida? and Cycloceratidoe,
in which the amount of involution remains the same throughout, the
highest species, such as Dactylioceras Braunianum, and Cycloceras
iEgion, or Masseanum, have flattened sides and acute abdomens.
These modifications, being the same as those which are correlative with
the increasing involution of the species in other" series, produce, also,
mimetic forms which only need one characteristic, that of involution, to
become closely representative of the deeply involved species.

Thus, all the typical Ammonites may be resolved into natural series,
in which the different forms in each series are related to each other

72 BULLETIN OF THE

very much as Lytoceras fimbriatum and others are related to the more
involute Pliylloceras, and the different series contain more or less rep-
resentative or mimetic forms due to the resemblance occasioned by the
amount of the involution or the characteristics which are usually correla-
tive with the amount of involution. The differences between the series
are found in the development of the young, and the structural peculiari-
ties of the shell and septa.

"When, however, the organization of the group no longer progresses,
but retrogrades by the uncoiling of the wborls in Scaphites Ancycloceras,
and Baculites, repeating — as shown by several authors, but notably by
Barrande — the earlier forms of the Nautiloids in inverse order, these,
though strictly mimetic, are produced by the encroachment of senile
characteristics. These are observed in the old age of such species as
Amm. Humphriesianus, where the old whorl becomes smaller, more
cylindrical, and if growth was continued, must eventually strike off
from the regular spiral as in Crioceras or.Lituites. This irregular-
ity is found at earlier and earlier stages of growth, and finally affects
the whole form as in the completely straightened Baculites. Direct
inheritance of senile characteristics is not claimed, but merely that the
retrogression of the individual in old age and the retrogression of the
group are similar, and both due probably to the same cause, exhaustion
of the powers of growth.

There is, however, a notable exception, which can be accounted
for only by Professor Cope's law of "retardation," and which, to me,
was inexplicable until the appearance of his essay on the Origin of
Genera ; the two Gasteropod-like genera, Turrilites among Ammonoids,
and Trochoceras among Nautiloids. With regard to the latter, there
are no certain data ; but the former are produced at first in varieties of
species, which have, according to Quenstedt's, Oppel's, and my observa-
tions, simply prolonged into the adult an individual variation common
enough in the young shells. The young of several species of typical
Ammonites often assume the spiral, although this is entirely suppressed
at a later stage, and the succeeding whorls resume the normal mode
of growth and revolve in the same plane. When, therefore, the
normal mode of development is " retarded," we find even in the adult
this Turrilites-like condition of the young, which is as truly reversional
as the Orthoceratitic young of Goniatites fecundus. This happens
occasionally in the lower Jura, and finally, after the progressive stage

MUSEUM OF COMPARATIVE ZOOLOGY. 7-3

of the whole order passes its climax in the lower and middle Jura, we
find the development of a whole group affected by this retardation, and
the spiral is common to several generic forms.

The ovisac of Goniatites crenistria differs from that of the Ammo-
nites in the greater breadth proportionally of the abdomino-dorsal axis.
The ovisac of G. primordialis, G. retrorsus var., and G. diadema do
not seem to differ in this respect. Even the two depressions and the
bulging of the sides are as well marked as in the ovisac of Deroceras
planicosta.*

The ovisac of Nautilus was not present in any of the seven speci-
mens examined by me. The form of the mouth of this, however, can
be inferred from the oval ridge on the ajiex of the first whorl, and
the central scar which marks the former aperture through which the
animal probably passed into the fundus of the first whorl.f The outer
limits of the area thus marked out are flask-shaped. The lower portion
or abdomen, if it were extended laterally, would correspond to the
broader abdomen of the Ammonoidal ovisac, and the depressions on
either side of the dorsum to the embryonal umbilici. The apex of the
whorl rises in a well-defined ridge which marks out this area on the
dorsum and the sides. At the abdominal extremity the defining ridge
is hardly distinguishable, and the shell rises directly to the cicatrix,
which is here the most elevated portion of the apex. The striae of
growth and the longitudinal furrows both cross this area, but are
arrested at the edge of the scar. The ridge, -when seen from the side,
is found to be accompanied by a narrow, shallow, slightly concave
band, which at the dorsal end is particularly well marked. The edges
of the cicatrix, which is a flattened, corrugated, elongated narrow space,
are tumid and more or less elevated.

The ridge appears to mark the line along which the extreme outer
edge of the mouth of the ovisac abutted against the apex of the whorl.
The lips of the embryo shell were probably inflected as in Gompho-
ceras, but instead of being convex as in that genus, were probably
concave. The projecting edge of this concave area would then have
fitted neatly into the shallow channel or area running around the inner
side of the ridge, and the concavity have been the mould upon which

* Plate III, Figs. 3, 4, 5. The ovisac differs more when viewed laterally, since the
outline of a section through the centre, as in Fig. 5, is more decidedly circular,
t Plate III, Fig. 1, and ideal section last page of the text.

74 BULLETIN OF THE

the convexity of the central portion of the scar was formed, the
cicatrix itself being left vacant for the passage of the animal.

The abrupt and broken character of the border of the external
layer on the edge of the cicatrix, and' its crenulated aspect indicate that
here is the true location of the former junction of the ovisac and the
shell of the first whorl. When we consider how narrow and vertical
are the apertures of some of the arcuate Nautiloids of the Silurian
epoch, and how closely they approximate to the simplicity of the outline
of the cicatrix, this view acquires additional probability, and it seems
to be the only one which can reconcile the continuity of the external
layer and its markings over this region.

It still remains difficult, however, to account for the passage of the
large body of the embryo through the narrow aperture thus made,
and future investigations upon the embryology of Nautilus are much
needed, in order to settle this interesting question, as well as the true
affinities of the form and structure of the embryo.

The cicatrix occupies, as has been described, the true apex of the
whorl, as determined by the structure of the shell, but only the lower
end, which is curved dorsally, occupies the actual apex; the remainder
runs along what appears to be the inner or dorsal side, though this
really begins higher up at the dorsal border of the cicatrix.

The absence of the ovisac is due either to its delicacy and the readi-
ness with which it could be broken away from its attachment, or to the
advance of the mantle, which in course of growth strikes the cicatrix a
little inside of the extreme abdominal end, and then bends up over it,
and either absorbs or pushes the ovisac away. Whatever may be
the ultimate resolution of the question, one fact is very evident : the
embryo of Nautilus differs not only in its form, which is a vertical oval,
from the Ammonoids, which is a horizontal oval, but in the mode of
its passage into the first whorl.

The whole aperture, or lip, of the ovisac in the Goniatites and
Ammonites is united and continuous with the shell of the first whorl,
which opens into it at the apex. The siphonal ccecum also has the
peculiar pointed cone-like prolongation extending into the ovisac,
through the first septum, which shows that the important organ which
secreted it differed not only in comparative size, but in shape, and in
the earlier period at which it was developed.* The siphonal coecuin

* Plate III, Figs. 2, 5, 6.

MUSEUM OF COMPARATIVE ZOOLOGY. 75

of Nautilus was not formed until after the animal had passed its first
stage of growth and occupied the first whorl sufficiently long to build
the first true septum. Even then this organ had not the size and im-
portance to which it subsequently developed. I have examined the
apices of many fossil Nautili without succeeding in finding any suffi-
ciently well preserved to show the original condition of the external
shell. One fine specimen of Nautilus Koninckii, from Tournay, had
apparently a smooth termination ; the longitudinal plications which
cover the young shell of the ornamented Carboniferous Nautili reached
only a little beyond the second septum. The whorl was here a rapidly
increasing cone, the abdomen, however, quite as gibbous as the dorsum,
whereas in the adult the latter is the more prominent, the abdomen
becoming deeply inflected. The termination of the whorl was very
much flattened, so that from the side it had quite a pointed aspect,
whereas an abdominal view showed it to be rounded at the extremity.*

Nautilus Koninckii, it will be remembered, is a Carboniferous species
with a very large umbilical perforation. In fact, the whorls do not
even touch at first. The tip of the cone is free for some distance
before the involution brings the whorls in contact. No marks of a
cicatrix were discernible.

Saemann's original specimen of Nautilus atratus is the finest I have
ever seen, and yet this is only a cast. The apex, however, is formed
by a cake of iron, which has a rough, lumpy surface, difficult to ac-
count for on the supposition that it was the cast of the smooth interior
of an unbroken shell.f The area between the first and second
septa is smooth, the abdomen flattened, and a faint median depression
is noticeable near the suture of the first septum. This, and the second
septum, incline towards the umbilicus at a greater angle than any of
the succeeding septa, j The point of the external shell is the corner
which it makes on the abdominal side, as it passes around the angle of
suture. This has been habitually mistaken for the apex, whereas the
organic apex is really further inward, and nearly parallel to the first
septum, and, in some Nautili, such as Nautilus lineatus Sow., is an
almost flattened area apparently on the dorsal side. § Nothing ap-
proaching a cicatrix was actually discovered, and yet, besides the general
form, which is similar to that of N. Pompilius, the aspect of Saemann's

* Plate IV, Figs. 7-9. J Plate IV, Fig. 5.

t Plate tv pig. g. 5 Plate IV, Fig. 10.

76 BULLETIN OF THE

cast, and the sections of N. lineatus, which I have seen, indicate that it
is present. One of these casts has a very thick tumid shell over the
apical portion, and the other, which has a thin shell, exhibits a trans-
verse depression just inside of the siphonal ccecura. This shows that
the shell not only differs in thickness on the apex, but is more or less
corrugated also, as if by a scar.

The oval outline of the area of the cicatrix, slightly flattened on the
ventral side, is singularly like the adult of the Nautilus Bohemicus,
and others of the Silurian Nautili described by Barrande.* The
regular ellipse of the young of the latter, and the flatter cone of the
young of the Carboniferous Nautili, is not represented at all in the
young of N. lineatus and N. atratus of the Jura.

UMBILICUS.

As the embryo of the typical Ammonites, and the closely coiled
Goniatites, such as G. diadema, approaches the beginning of the first
whorl, its flattened dorsum becomes depressed or concave on either side
as previously described, and when the apex of the first whorl bends
dorsally, hollows are formed on either side, closed at the centre. These
are the embryonal umbilici.

They do not exist in Nautilus, but their homologues are probably
found in the ovisac, as previously pointed out, though they can form no
umbilici properly so called. The close coiling of the first whorl forms
the umbilici by enclosing these spaces in Goniatites and Ammonites.
In Nautilus they can never be so enclosed, owing to the loose coiling
of the first whorl. The umbilicus of Nautilus Pompilius penetrates
entirely through the whorls, as it does in the group of Nautilini among
the Goniatites, where the ovisac does not fill up the centre. The
lateral depressions or sinuosities observable on the sides of the outer
rim of the scar in N. Pompilius, the homologues of the umbilici of
Ammonites, were not observable in N. atratus, since, as previously
stated, no well-defined scar was observable among the fossils which
came under my observation.

WHORLS.

The whorl of the typical Ammonites, and the closely coiled Goni-
atites is at first as broad as the ovisac, then rapidly contracts, becoming

* Plate IV, Fig. 6.

MUSEUM OF COMPARATIVE ZOOLOGY. 77

considerably narrower.* Before the completion of the first revolution,
estimating from the neck of the ovisac until the whorl again touches
this point of origin, it reassumes the normal rate of increase.

The increase in the dorso-abdominal diameter appears to be very-
marked at first, and this gives a peculiarly broad aspect to the sides
of the whorl, just beyond the embryonal umbilicus ; f subsequently the
increase is constant and invariable in all the diameters.

The form of the whorl is notably distinct from what it auerward
becomes, and is identical with that of a typical adult Goniatite, and an
equally close representative of the young of G. diadema and other
closely coiled Goniatites. There is the same broad, even, somewhat
flattened curve to the abdomen, and abrupt sides. The retention of
this Goniatitic outline is greater or less in different species, and it is
necessary to be cautious in estimating the duration of this period of
growth. The septa give an accurate measure of the time during which
the young animal may be said with truthfulness to have resembled an
adult Goniatite in some of its characteristics.

The types I have examined are evidently too far removed from the
Goniatites in the structure of the adults, and their development con-
sequently too much accelerated, for any very extended or exact ref-
erence, in all their characteristics, at any one period of growth. Such
precise identity of the young with the adults of the parent type can
only be expected in the Clydonites, or some of the simpler transitional
groups, where the Goniatitic stage must necessarily be of longer con-
tinuance.

The outline and septal structure of the young are almost identical
between the Goniatites and Nautiloids, more especially the arcuate
forms of the young with plain concave septa, such as G. compressus,
when compared with the arcuate forms such as Cyrtoceras. Sand-
bergei*'s figures show in section the young whorl of this species, which
is a regular ellipse, and has an outline and septa very like Nautilus
Bohemicus ; the siphon also being at an earlier period undoubtedly
abdomino-central, we have here at one stage a close approximation in
the general characteristics of the structure.

The umbilical perforation is very contracted and small in Nautilui
Pompilius, when compared with the fossil, and differs also in form from

* Plate I, Fig. 6. Plate II, Figs. 2, 11. t Plate I, Fig. 7.

78 BULLETIN OF THE

the Paleozoic species. The Nautili of the Silurian, such as Nautilus
Bohemicus, Nautilus Sternbergii, Nautilus tyrannus, Nautilus Sacheri,
figured by Barrande, have very large umbilical perforations. The
Devonian Nautilir like those of the Silurian and Carboniferous, all
belong to the group of Imperfecti, and have large umbilici, showing the
entire spire.' What is the exact size of the umbilical perforations I
cannot state, but doubtless it is in the majority of the species large, and
the young whorls rounded as in the Carboniferous and Silurian mem-
bers of the same group.

The Carboniferous forms are distinguished by their highly orna-
mented and varied adult shells, as well as by their exceedingly large
umbilici. Nearly all of these DeKoninck* has pointed out are charac-
terized by a large umbilicus, showing the entire spire. Even the
Nautilus oxystomus, a species in which the whorls are considerably invo-
lute, has an umbilicus, and at first a rounded whorl, which subsequently
becomes hexagonal and then lanceolate in transverse sections, and
involute. The umbilical perforation in this species is much smaller
than in those species with non-involute whorls. The Museum has
among other treasures the entire collection of DeKoninck, and I have
been able to verify these observations.

The species of the Trias and Permian I have been unable to
examine. But of the two species figured by King,f N. Freislebeni,
and N. Bowerbankianus, one has a closed, or what is generally called a
closed, umbilicus, L e. with probably, as in the modern N. Pompilius,
a small perforation. N. bidorsatus of the Muschelkalk also has a form
like many of the Jurassic species, and a much smaller umbilical per-
foration than is common in Carboniferous species. Other species of
the Muschelkalk, such as N. mesodiscus, N. Sauperi, N. reticulatus,
are as completely coiled, and as involute as any of the Jurassic and
succeeding formations. It is probable, therefore, that the earliest gen-
eral change in the size of the "^bilical perforation will be found to
take place in this formation.

The Nautili of the Jura, as pointed out by Pictet, j approximate
closely to the existing species in the adult stage, and, as DeKoninck
has shown, their umbilici are comparatively closed. The young of two

* Animaux Fossiles, p. 544.

t Permian Fossils of England, p. 219.

4 Pictet,Traite (ilthnentaire de Pal^ontologie, 1845, Vol. II.

MUSEUM OF COMPARATIVE ZOOLOGY. 79

species are figured by D'Orbigny, and these are decidedly in advance
of the circular and very slowly increasing whorls of the young of the
Carboniferous and Silurian Nautili, and the umbilical perforations are
much smaller. Barrande says of the umbilical perforation, " Au centre
de la spire, il existe une vide ou perforation, qui est remarquable par sa
Constance, et son etendue dans les Nautiles paleozoiques. Nous re-
trouvons cette perforation, quoique tres-reduite, dans les especes fossiles
des epoques posterieures et meme dans les Nautiles qui vivent dans
nos mers." The Jurassic shells belong almost wholly to the Striati,
a group with longitudinal ridges or plications on the whorls.

The young of Nautilus lineatus of the Jura presents an umbilical
perforation quite as small as that of the modern Nautilus. The whorl,
however, does not make the graceful curve of Nautilus Pompilius, but
bends inward more abruptly, and instead of touching the apex of the
whorl first, it strikes the dorsal ridge of the area of the cicatrix, fitting
itself to its flattened surface.* The result is an irregularity in the
curvature of the dorsum at this point, which appears to be abnormal,
but is probably characteristic at le?st of the species, since I found it
in two different specimens.

This fact, however, as well as the figures of the young given by
various authors, shows that in the" Jura the whorling probably becomes,
in several species, as close, and perhaps closer than in the modern Nau-
tilus ; certainly, in all those forms which, like Nautilus lineatus, are
very involute in the adult.

The Radiati of the Cretaceous are, as a whole, more deeply involute
than the Striati of the Jura, though not differing as respects the size
of the umbilical perforation. The earlier age at which the involution
begins is particularly noticeable, and the consequent prevalence of
forms which increase in size more quickly than the majority of Jurassic
species may be assumed with confidence, though the material at my
command does not enable me to substantiate this statement by actual
observation made upon the uncovered young.

It is founded, however, upon the fact which appears to be universal
among Ammonoids and Nautiloids, that the earlier a species begins to
become involute, the quicker must be its increase in size. Involution,
indeed, is only one method of expressing the expansion of the shell in-
wardly by the growth of the sides over the umbilical area, and it is

* Plate IV, Fig. 10.

80 BULLETIN OF THE

evident that this, when it occurs early in life, must, as in the modern
Nautilus, occasion a more rapid spreading of the sides at the apex
than is to be found in the Striati of the Jura. This, of course, does not
exclude the effect of the spreading of the sides independently of involu-
tion, as in Nautilus excavatus Sow. of the Jura; but this is not gen-
erally so well marked in the young as in this species, and has no
bearing upon the question, which concerns only the prevalence and
early development in time of involution, as it may be observed in full-
grown specimens.

The Loevigati of the Tertiary appear to come no nearer to the
existing Nautili than the Radiati, except in their smoother shells. The
group of Aganites or Nautili, with deep lateral lobes like those of
Clymenia, form a distinct genetic series, but they are no exception to
the rule. Nautilus aganiticus, gravesianus, and sinuatus of the Jura
are all less involved than either Nautilus aturi, zic-zac, lingulatus, or
Morrisii, and others of the Tertiary.

Another peculiarity, the concavity of the dorsal side, is earlier de-
veloped in the closely coiled young of Jurassic, Tertiary, and existing
Nautili than in the Palaeozoic forms. Barrande, among the Nautilini
of the Silurian, treats this characteristic as one which is due to the in-
volution of the whorls. His figures show that it is very slight, and
arises at a comparatively late period, and only after the whorls come in
contact. It does not exist at all in Nautilus vetustus, which is a very
loosely coiled whorl. The conclusion here also seems to be inevitable
that a characteristic, at first fluctuating, and pertaining exclusively to
the older periods of growth, becomes more embryonic in the later
species of the same genetic series. In this instance also we have, as
the result of this law of acceleration, a characteristic which is at first
dependent upon another characteristic, the involution of the whorls,
becoming incorporated in the organization, and finally manifesting itself
independently, in the growth of each individual, before the whorls
touch each other. It is certainly universally present at a very early
period in the young cf the Jurassic and succeeding periods, though
absent in many of the Carboniferous forms.

These exceptions may be said to prove the rule, for species, such as
N. Koninckii, in which it is absent, have therr backs convex instead of
concave, because the abdomens of the whorls which they involve are
concave instead of being convex.

MUSEUM OF COMPARATIVE ZOOLOGY. 81

The dorsal concavity in the young of Nautilus atratus is marked at
the very commencement of the whorls by the flattening of the elliptical
outline of the second septum.* In Nautilus Pompilius, however, the
dorsum is flattened at an earlier period, the concavity affecting the con-
formation of the first septum.

Sandberger f and Richard Owen J both allude to the Goniatitic
stage, and also to a subsequent Ceratitic period. Sandberger's recog-
nition of these periods of development, though not quoted by Owen,
has priority in point of time, and is much fuller and more comprehen-
sive. He also shows that Hauer really first called attention to them
by his figures of Amm. floridus. § Sandberger, however, failed in
seeing their full significance, since he most emphatically denies that
these transformations show any affinity between Goniatites and Nau-
tilus in the following words: " Obgleich ich die letztere Thatsache '*
(the simple Nautilian characters of the septa of young Goniatites)
" keineswegs fiir einen weiteren Nachweis vewandschaftlicher Verhalt-
nisse von Goniatites und Nautilus geltend zu machen gedenke und
angesehen wissen will." Owen describes the transformations of an
Ammonite, only in order to show that they were simpler in the young
than in the adult, and that the young of the same species at different
periods of growth had been by different authors referred respectively
to Goniatites, Ceratites, and Ammonites.

The so-called Ceratitic stage exists only in the septal sutures, and
will be referred to further on.

After the Goniatitic stage is completed among the typical Ammonites
the outlines of the whorls assume no general form, but vary according
to the group or genus in which the shell is found.

SEPTA.

The first septum of the typical Ammonite is situated at the junction
of the first whorl and the ovisac. It has deep, entire, simple lateral
lobes on either side, and a prominent abdominal cell as in Nautilus. ||

The dorsal side was not so accurately observed. It is probable,

* Plate IV, Fig. 6.

t Oberhessische Gesellschaft fiir Natur-und-Heilkunde, 1858.

J Palaeontology : Second edition, p. 99.

§ Cephalopod, von Bleiberg. Plate I, Fig. 14.

|| Plate I, Figs. 2, 5.

VOL. III. 6

82 BULLETIN OF THE

however, that the lines " X " delineated in the interior of the specimen
of Deroceras planicosta, Plate I, Fig. 6, show the junction of the
first septum with the dorsal side of the whorl. They are in the proper
position with relation to the external suture, and it is difficult to imagine
any other structure to which they could belong. I failed to find any-
thing similar in other specimens, and the original was accidentally lost
after the figure had been drawn, while transferring it from one glass
slide to another.

The huge size and depth of the abdominal cells of the first septum
has no exact parallel, so far as I am aware, in any known adult
Cephalopod. The pouch-shaped lateral lobes also appear to be
peculiar. The sutures, however, vary considerably in the young of
the same species. In some specimens inferior lateral cells* appear on
the sides, and the lateral lobes are not so high, the abdominal cell, also,
varying proportionally in breadth and depth. These lateral cells, how-
ever, do not exceed the strict Nautilian limits, and remind the observer
of the slight inferior lateral cells of Nautilus Pompilius, both in position
and in outline. Often, however, they occur so exactly on the edge of
the embryonal umbilicus, and the descent into the lateral lobes is so
abrupt that the observer confounds the suture with the projecting edge
of the first whorl.f When such a specimen is viewed from the side by
transmitted light, and the ovisac is transparent, as it frequently is, the
shell appears to be pointed as in Nautilus, and the ovisac absent.
When, however, a few specimens are completely broken down, and
the ovisac and first septum laid bare, the error is easily corrected.

In section, the first septum appears to blend with the wall of the si-
phonal coecum.t The enclosed transparent spaces, however, as indicated
at 1 e, 2 e, Plate II, Fig. 1, between the siphonal wall and the inner
side of the shell proper, probably represent all of this septum which ob-
tains along the median line. In other specimens which I have studied
this cannot be so plainly made out, and I should still be in doubt if
these transparent spaces had not been enclosed by the lining layer of
the shell. They can be compared in other respects also with the suc-
ceeding septa, which they resemble in position, size, and the peculiar
spreading or fan-shaped sections of the sutural portions or junctions.
The texture of the septum seems to be very distinct, but the drawing

* Plate I, Figs. 2, 5. J Plate II, Fig. 1, 1 e.

t Plate II, Fig. 7. Plate I, Fig. 4.

MUSEUM OF COMPARATIVE ZOOLOGY. 83

necessarily exaggerates its transparency as well as the opacity of the
siphonal wall. They cannot really be distinguished at their borders,
the one from the other, but in every case the true calcareous septum
blends with the brown of the wall of the siphonal coecum. This may
be the effect of fossilization, since, in other instances, at a later age, there
was no difficulty in distinguishing the two at their junctior

The second septum* has a very distinct, deep, broad ventral lobe, di-
vided by a ventral cell and siphonal fissure, deep superior lateral cells
with shallow superior lateral lobes and inferior lateral cells equally slight,
the inferior lateral cells beginning to appear upon the border of the
whorl. Thus, when seen from the front, this septum has an appearance
which leads the observer to suppose that it is interrupted-! An abdom-
inal view,! however, readily corrects this mistake, especially in section
where the continuity of the suture is readily defined, and a lateral view
of the succeeding septa shows the divided ridges of the ventral cell very
plainly. § These two ridges normally form one cell. The fissure is
probably due to the prolongation of the siphonal funnel. This is a
collar-like inflection of the septum, directed posteriorly, the circular ex-
tremity resting upon the sheath of the siphon. This funnel has a shal-
low channel on the ventral side, which is found in Nautilus also. When,
'herefore, the siphon is near enough to the shell, this channel divides the
central cell ; otherwise, it ceases to depress the septum before arriving
at the edges, and leaves the suture entire. || These facts also demon-
strate another, of some importance, that the ventral cell is no adventi-
tious product of the near approach of the siphon to the ventral side in
Goniatites and Ammonites, but an independent characteristic, constantly
present in the latter, but absent from many species of the former. This
conclusion is substantiated by a specimen, two septa of which are figured
below. ^[ No abdominal channel or gutter was present, and the abdominal
cell was left entire.

The species in which the gutter is absent, among the Goniatites, have
a simple ventral lobe, as many of the Nautilini, Goniatites discoideus
Hall, and Goniatites Patersoni Hall. In some, however, as pointed out
by Barrande in Goniatites plebius, the ventral prolongation of the fun-
nel deepens the ventral lobe itself. This is due, perhaps, to the extensive

* Plate I, Figs. 4, 5. Plate II, Figs. 7, 8. § Plate II, Fig. 8.

t Plate I, Figs. 1, 0. || Plate II, Figs. 10, 0.

I Plate I, Fig. 2. ^ Woodcut on last page of text.

84 BULLETIN OF THE

development of the funnel in Goniatites, which contrasts with the slighter
proportional development of this part in Ammonites.

Leaving the funnel to be more fully described in treating of the struc-
ture of the siphon, we can proceed to the dorsal side of the second sep-
tum. This suture is precisely similar to the ventral side in the number
of the lobes and cells, though these are shallower and narrower, as might
be expected from the more confined limits within which they are neces-
sarily distributed. The young Ammonite in the second septum, therefore,
really possesses, like the simpler forms of Goniatites, — the Nautilini,
which almost exclusively occupy the Silurian epoch, — only large lateral
lobes, which, with their duplicate dorsals, make four in all, together with
one dorsal, one ventral, and two inferior lateral lobes.

These last occur upon the edge of the whorl, and are therefore fdled,
at first, only by single lateral projections of the animal's body. They,
by growth, become duplicated on the ventral and posterior sides ; but, at
this stage, they are single. The sutures of the second septum, there-
fore, possess one pair of median lobes, the ventral and dorsal ; two pairs
of superior lateral lobes, and one pair of inferior lateral lobes, making
eight in all.

Six is the number which has always been attributed to the young
Ammonites, and also the normal forms of Goniatites, though differently
counted. The external lateral lobes were counted by D'Orbigny and
others separately from their duplicatures on the dorsal side ; very in-
consistently, however, the dorsal lobe, which is plainly only a duplication
of the ventral, was enumerated as a separate lobe. In this way observ-
ers have found only two external inferior, and two external superior
lateral, with one external, and one internal median lobe, making six in
all. D'Orbigny was strictly correct, so far as he went, in assigning six
lobes to the young Ammonite, though at fault, with the exception of the
ventral lobe, in not observing their counterparts on the dorsal side.

The third and succeeding septa* on the first volution differ only in
the fuller development of these same lobes and cells ; on the early part
of the second volution in Deroceras planicosta, and on the latter part of
the first whorl in Arnioceras semicostatum, however, the increase in
breadth of the whorls exceeds the normal increment of growth of the
lobes and cells. When this takes place, the inferior lateral lobes no
longer occupy their old position on the umbilical border, but retire with-

* Plate I, Fig. 4. Plate II, Figs. 7, 8.

MUSEUM OF COMPARATIVE ZOOLOGY. 85

in, and the inferior lateral cells appear. * The whorl continues to en-
large, and finally these also retire to make way for the first auxiliary
lobes. This occurs in Deroceras planicosta upon the last quarter of the
second volution ; but in Arnioceras semicostatum the development is
much accelerated, the first auxiliary lobe appearing on the first cpjarter of
the second volution. The first auxiliary cell succeeds these lobes in the
latter species, certainly as early as the second quarter of the third whorl,
and in the former they are delayed until the second quarter of the
fourth.

The peculiar minor lobe dividing the superior lateral cells, the first
decisively Ammonitic characteristic assumed by the sutures, is fully ex-
pressed in Arnioceras semicostatum, on the latter part of the third vo-
lution, while in Deroceras planicosta the minor lobes are hardly
apparent on the latter part of the fourth.

Further than this it is not my present purpose to carry the history of
the development of the sutures. From this point, or rather after the
first two whorls, the sutures should be studied in connection with special
series and groups of series.

D'Orbigny, who studied more thoroughly than any other author the
mode of apparition of the auxiliary lobes upon the umbilical side, decid-
ed that they were due to the increase in breadth of the whorl, and the
facts here recorded confirm the accuracy of his statements.

My observations do not lead me to recognize the so-called Ceratitic
period of other writers. If there is anything peculiar to the Ceratites,
it consists in the presence of an indefinite number of small, club-shaped,
serrated lobes. No corresponding features exist in the young Ammonite,
not even in Amm. floridus, as figured by Hauer. The smooth and deeply
sinuous sutures of the young Ammonite, such as Arnioceras semicosta-
tum, have a faint resemblance to those of the adult of Goniatites Hyas,
from Rockford, Indiana, and to some adult Clydonites or young Cera-
tites. These are mere similarities, however, which do not impress the
observer as anything more than mimetic changes, such as are observable
everywhere between the structures and forms of distinct, but genetically
connected series. The fundamental characteristics of the simpler and
earlier-occurring Xautiloids and Goniatites are very distinctly repeated
in the young Ammonite, but beyond this comparisons can only be safely
made by tracing series downward. I have succeeded, by the aid of the

* Piute I, Fig. 8. Plate II, Fig. 9.

86 BULLETIN OF THE

splendid material accumulated by Professor Agassiz, in doing this with
considerable certainty for the Arietidae, Liparoceratidre, Cycloceratida;,
and somewhat less completely for several other families of typical
Ammonites.

Goniatites has the first septum very similar to that of the typical
Ammonites, but on the ventral side a deep siphonal fissure occurs, ow-
ing to the near approach of the conical prolongation of the siphonal
ccecum to the ventral side.* The sutures of the first septum follow
the cone in two specimens of Goniatites diadema, but in both the neces-
sarily violent removal of the shell probably carried away the outer side
of the siphon.

Barrande's observations show that this frequently occurs in the adult,
when the whorls of the Silurian Goniatites are torn apart in order to ex-
pose the ventral side. The broken edges of the upper part or neck of
the siphonal coecum, in both the specimens alluded to, as in the one fig-
ured,f make continuous sutures along the abdomen from the first to the
second septum, and this could hardly be accounted for under any other
conditions.

A comparison of sections of this part of Deroceras planicosta and
Goniatites diadema J show that in both the same conical prolongation
exists, and has about the same relation to the shell. In the former,
however, the cone ceases to be an integral portion of the siphonal coecum,
and the invariable continuity of the suture of the ventral cell is due to
the manner in which the first septum bends inward to form the floor of
this part only, instead of being more deeply inflected to form the cone,
as in Goniatites.§ Both are normally entire, but in Ammonites alone,
owing to the structure of the cone, is this apparent in all specimens.
The suture of the second septum in Goniatites diadema is continuous ;
the neck of the siphonal coecum not being so close to the shell, it is
not easily obliterated, and the dark line of the suture is readily traced
across the siphonal area. This septum has entire outlines, a shallow
broad ventral lobe, and superior lateral cells, with narrower, but still
very well marked lateral lobes near the embryonal umbilicus. || The
dorsal sides of neither of these septa were observable on account of
the opacity of the specimens.

* Plate III, Figs. 3, 5, 6. § Plate II, Fig. 1, 1 e.

t Plate III, Figs. 3, 5, 6. || Plate II, Figs. 3, 4.

t Plate I, Figs. 3, 4. Plate III, Figs. 5, 6.

MUSEUM OF COMPARATIVE ZOOLOGY. 87

Comparing the first septum with the corresponding septum of Dero-
ceras planicosta, it is noticeable that the ventral cell rind the lateral
lobes are deeper and narrower in the latter ; a characteristic liable,
probably, to considerable variation, since it accords with the larger size
of the neck of the ovisac, and the consequently greater spread of the
sutures in Goniatites diadema.

The suture of the second septum differs far more from its representa-
tive in the young Ammonite. The ventral lobe of the latter, with its
siphonal fissure, is represented by an entire shallow lobe ; the superior
lateral cells have their counterparts in wide, shallow, evenly curved
sutures, unbroken by any superior lateral lobes. This comparison, and
the abrupt introduction of the superior lateral lobes in the third septum
of Goniatites diadema,* also show plainly that the superior lateral
lobes do not arise, as do the inferior lateral and auxiliary lobes, from
projections of the body, which appear first as single lobes upon the
edge of the umbilicus, subsequently becoming double (i. e. divided into
dorsal and ventral, vis-a-vis,) by the spreading of the sides of the
whorls. On the contrary, they are derived from the direct growth of a
lobe which bisects the large entire superior lateral cell, and divides it
into superior and inferior lateral cells. The inferior lateral lobes are
present in both. The outline of this suture in Goniatites diadema is
very like that of the adults of Nautilus atratus and similar species
among the Nautiloids which have sweeping lateral and shallow abdom-
inal curves.

The next, or third septum of the young Goniatites diadema intro-
duces the superior lateral lobes as a minute angular indentation, not at
all similar to the rounded superior laterals of the young Ammonites.
The development of the sutures is more accelerated in the growth of
the latter than of the former, since here we have two septa with Nau-
tilian characteristics, in place of one >among the later-occurring and
more complicated Ammonites. When the young Goniatites diadema
is contrasted also with those of its own group, we find some interesting
resemblances. The Nautilini never develop superior lateral lobes, the
Uctrorsus group, Magnosellares of Sandberger, not until a very late
stage of growth, and in some varieties, as shown by the masterly
researches of the last-named author, these lobes are very slight, even
in the adults. I have not examined the young of other species and

* Flate III, Fig. 4.

88 BULLETIN OF THE

groups with more complicated sutures. This additional evidence is not
necessary, however, to show that within the Goniatites there is still
greater acceleration displayed between the young of different groups in
the development of the lobes and cells, than can be found between the
young of the latter and the typical Ammonites. This also corresponds
with the scope of the modifications observable in the adults of the
entire division between the simple Nautilini, and the Genuifracti or
Serrati, or the range of variations in the adults of the same species,
as exhibited by Goniates retrorsus and others. No such variability
of the sutures exists among typical Ammonites, excluding the allied
divisions of Ceratites. Thus, in the adults as well as in the form of
the young and the amount of involution, we find greater fixity resulting
from the acceleration of the development as we rise higher in the type.
This continues until senility or the decline of the type again produces
something similar among the aberrant uncoiled genera of the upper
Jura and the Cretaceous. The fourth, and succeeding septa of the
young Goniatites merely increase the divergence from the common
type, the rotund ovisac, and first septum, and the history of these
later stages will undoubtedly be found to be characteristic of, and
highly useful in the definition of special groups.

The first true septum of Nautilus has exceedingly shallow lateral in-
flections of the sutures, and a ventral cell of corresponding curvature.
These could hardly be termed lobes and cells, so slight are they, if it
were not for the greater intensity of their expression in the second and
succeeding septa. A very important feature of this first and all the
succeeding septa, is the dorsal lobe. This has been previously noticed
by De Montfort * who founded his genus Bisiphites upon this feature,
and by other authors, as an adult characteristic of many of the fossil
and recent Nautili. Its developmental history has not, however, been
followed out, except by Quenstedt.f This observer described only its
later stages, and its disappearance in the adult. He appears also not
to have noticed that, in the adult, a slight inflection remains in the
suture after the lobe in the septa ceases to exist. Barrande describes
the sutural lobe and the corresponding conical depression of the septa,
but considers that they may be independent of each other. They are,
however, so closely connected in the extreme young, that the sutural

* Conchyl. Syst. 1808.

t Petrefactenkunde Deutschlands, Ceph. p. 55.

MUSEUM OF COMPARATIVE ZOOLOGY. 89

inflection, which is alone retained in the adult, should be regarded as a
remnant of the more complete lobe of the young. In the Aganites
group, the lobed and its accompanying sutural inflection are present
whenever the siphon funnel is not too close to the dorsal side. When
this occurs, as in Nautilus zic-zac, the lobes of the septa are not
developed, but the sutural inflection appeared to be distinctly marked.
"Whether this was really persistent, or whether the depressions I saw
were due to the violent removal of a portion of the funnel, was not
satisfactorily determined. The dorsal lobe of Nautilus Pompilius im-
pinges against the internal portion of the upper end of the cicatrix,
just as the siphonal ccecum does against that of the lower end, and is
apparently the result of the passage of some organ or part from the
ovisac into the shell, but the entire outline of the first septum in the
young of Nautilus atratus, N. Koninckii, is conclusive against such a
supposition.* It is evidently only another of the same kind as those
characteristics already cited, which are developed earlier in later
species of .the same genetic series, according to the law of acceler-
ation of development. When compared with the small area of the
first septum of Nautilus Pompilius, it is seen to be a large and very
well marked lobe. The size, however, does not increase propor-
tionally with the growth of the shell and the area of the septa, and
thus in the adult it becomes comparatively insignificant. During the
younger stages, also, it modifies not only the sutures, but forms a
decided conical depression in the septa themselves, so that the latter
bend posteriorly at their junctions with the dorsal side of the shell.
The lobe in the suture, however, is formed higher up, just under the
cone of the next older septum. This is due to the great extension of
the septa of Nautilus on the dorsal side, which reach so far that they
are overlapping or imbricated. This cone disappears on the first
quarter of the third whorl, and the septa, instead of bending posteriorly
into a lobe, are simply rectangular with a minute depression on the
surface. This, however, speedily disappears, and a slight ridge, evidently
marking the trace of some organ or part, makes its appearance. This
is found first bisecting the sutural lobe, and seems to extend con-
tinuously underneath the septa built upon the inner surface of the
dorsal side of the shell. The septa, after the appearance of the ritlge,
lose their rectangularity, and become evenly concave, the narrow,

* Piute IV, Figs. 5-9.

90 BULLETIN OF THE

shallow depression in the edges of the suture being the only remnant
of the dorsal lobe.

The second and succeeding septa of the young Nautilus Pompilius
differ in the possession of a narrow ventral cell upon the median line.
This subsequently loses its prominence, but is not wholly lost in the
general expansion of the cell ; even in the adult there is a slight rising
of the suture as it approaches the median line of the abdomen on
either side. The first septa of Nautilus Pompilius, as might be supposed
from the form of the whorl, are remarkably broad from side to side,
and slightly inflected on the dorsum on either side of the posterior in-
flection or deep dorsal lobe previously described.

In the Clymenia3 the dorsal lobe exists, and it may be observed at
an early stage, very large and distinct even in the simple sutures of
Clymenia laevigata. Sandberger's researches show that it is a common
characteristic of all the species of this Devonian group. The siphonal
funnel of the Clymenia?, which occupies the bottom of this lobe, is
evidently entirely independent of the lobe itself, as in the case of the
ventral lobe of Goniatites and Ammonites. The homology of the
dorsal lobe in the young Nautilus with that of the adult Clymeniae, can
hardly, therefore, be considered doubtful. It occupies the same position,
and has a similar conical form, with an accompanying broad, deep, and
rounded sutural inflection. This is especially marked in the Aganites
group of the Tertiary, where the siphon is variable in position, and
sometimes close to the dorsum occupying the area of the dorsal lobe, as
in Nautilus zic-zac, and in other species more central. Here we have
an additional reason, besides that previously found in the earlier
occurrence of the Goniatites, for separating the Clymenia? including also
Aganites as a distinct genetic series from the Ammonoids, notwith-
standing the ventral lobe of Clymenia pseudogoniatites.

The young Nautili of the Silurian, as figured by Barrande, have
shallow concave septa, and a more or less shallow outline in accordance
with the elliptical or circular shape of a section of the young. This is
especially noticeable in the figure of Nautilus ellipticus by Barrande.*

At a more advanced age all the species except Nautilus vetustus have
an elliptical outline more or less flattened on the abdomen. They also
possess, in the adult, a narrow ventral cell, like that of the young Nau-
tilus Pompilius, on the third and succeeding septa. The Carboniferous

* Op, cit., Plate XXXIT, Fig. 1.

MUSEUM OF COMPARATIVE ZOOLOGY. 91

species, which have similar cone-like young, have also similar septa.
Nautilus Koninckii * at least, has septa, which differ from the more
advanced stages of the young of Nautilus Bohemicus, possessing a
faint dorsal cell, or rather subangular outline to the dorsal side of the
septum. The suture of the first septum was observed upon one side,
though obscured on the other, and this evidently possesses decided,
though very shallow ventral and dorsal cells, with correspondingly slight
lateral lobes. The septa were elliptical in outline, with the dorso-
ventral axis longest, as in the Silurian Nautili, though they speedily
reverse this in course of growth, the transverse axis becoming the
longest in the adult. The dorsal lobe appears first in Nautilus Bar-
randei, Hauer,f or at least has not been described or seen in any
species occurring earlier, though it is a characteristic not likely to
escape observation, and was well known to Barrande, Sandberger, and
others who have worked up the Nautili of the earlier formations. The
developmental history of the septa of the Jurassic species shows a
decided acceleration in the septal characteristics. Saemann's original
specimen, when thoroughly cleansed,! showed that the outline of the
first septum was an ellipse, slightly flattened on the ventral side, cor-
respondingly with the flatness of the external outline, and nearly
parallel with the area of the cicatrix.§ Sections of two different
specimens of Nautilus lineatus show the same characteristics. The
first septum of Saemann's cast has an entire cell on the dorsal side, as
in the young of Nautilus Koninckii of the Carboniferous, though in
accordance with the shape of the dorsum it is much broader, and this
cell is still more prominent in the second septum, and broken by a
dorsal lobe. The second septum crosses the central axis of the spiral
at a different angle from the first septum, and is not, therefore, parallel
with the area of the cicatrix. The dorsal lobe is very distinct in the
third septum, and is supplemented by a decided though shallow inflec-
tion of the large dorsal cell forming a shallow supplementary lobe, such
as has been already described in Nautilus Pompilus. The concavity
of the dorsum becomes apparent between this and the second septum.

* Plate IV, Figs. 7, 9.

t Haidinger's Naturw. Abhandl. Ceph. von Hallstadt. By Hauer. Vol. III.

J It was covered with a thick coating of iron-rust, which obscured the sutures, and
entirely concealed the first septum, which was consequently omitted in Saemann's
figure of this specimen.

§ Plate IV, Figs. 5, C.

92 BULLETIN OF THE

The lateral lobes are only very faintly marked, as is also the ventral
cell. In the next three septa all these features are intensified, but
in the sixth and seventh septa a change takes place. The ventral cell
becomes flattened, and forms a transition to the slightly concave ventral
cell of the eighth septum. After this period the concavity becomes
deeper, and the shell, by growth, increases the depth and size of the
other lobes and cells which have been described, but otherwise does
not materially change them.

Thus the Jurassic Nautili, with a shallow lobe instead of a cell on the
abdomen, -do not possess this characteristic in the young, but have the
usual projecting cell of the Silurian Nautili. Their septa also resemble,
at the earliest period, those of the same forms when considerably older,
but this outline changes in the second septum. The development is also
accelerated in another way than by the earlier appearance of the Silurian
characteristics. The increased involution, as previously shown in the
small size of the umbilical perforation, and tendency of the whorls to
spread inward, causes the second septum to advance its outer or ventral
edge, and make a sharp angle with the first septum. The antero-poste-
rior axis of the whorl, in other words, curves suddenly instead of slowly,
as in the Silurian and Carboniferous Nautili, and the first septa are not
consequently parallel either with the mouth of the ovisac, as indicated
by the area of the cicatrix, or with each other.* The second septum
resembles the first, however, in its shallowness, and is a transition to the
third, which crosses the whorl at the same angle as the older septa, and
also resembles them in its deeper concavity, and the shape of the dorsal
cell and lobe. The acceleration of the development shows itself very
decidedly in the modern Nautilus. The outline of the first septum of
Nautilus Pompilius is broader, but otherwise the suture is identical with
that of the third septum of Nautilus atratus, and like that has the same
trend as the older septa, and is deeply concave. Its breadth, also, is due
to the accelerated rate of growth of the young, as are also the absence of
any stages corresponding to the first and second septum of Nautilus
atratus. The outline of the area of the cicatrix is very similar to that
of the first septum in Nautilus atratus, and this shows that, in all
probability, the neck of the ovisac resembled in shape the apex of
the first whorl in the Jurassic Nautili.

* Plate IV, Fig. 10.

MUSEUM OF COMPARATIVE ZOOLOGY". 93

SIPHON.

The siphon has been so much studied and so fully described that at
first sight it seems impossible to add anything to our knowledge in this
direction. The structure, however, in many of the most essential fea-
tures has been either misinterpreted, or only partially understood, on
account of the disuse of the microscope in palasontologieal researches.
The siphon in Nautilus Pompilius * begins with a closed coecal prolon-
gation of the first septum. This is circular in outline, apparently flat-
tened at the bottom, and rests directly upon the lower end of the cicatrix.
The ventral side is somewhat inclined, the dorsal more or less abrupt,
the ccecum swelling out below the septum in the specimen examined.
Above the first septum the siphon expands, becoming much larger.
The walls, as well as the bottom of the ccecum, are lined internally by
an extension of the sheath, which is continuous with the siphonal funnel
of the second septum. There is also in the figure on Plate IV a layer
lying between the internal corneous layer of the siphon and the sheath.
The walls of the third septum are continuous beyond the mouth of the
siphonal funnel of the second septum, and seem also to coat the inside
of the ccecum, but I have been unable to verify this observation upon
other specimens, and prefer to consider it doubtful.

The structure of the outer wall, or sheath of the siphon, differs from
that of the septum itself, in being of a looser and more granular texture.
This, in the specimen examined, was more marked upon the dorsal than
upon the ventral side, between the apex and the first septum, as well as
between the latter and the second. At the third septum, however, this
difference is not so noticeable ; the denser portion of this septum, or the
siphonal funnel, extends posteriorly nearly as far on the dorsal as on
the ventral side. As previously described, the siphonal funnel rests
upon the looser-textured wall, and is really continuous with it. The
difference in texture merely results from the quicker formation of the
latter, while the animal is passing from one septum to the next in a«;e,
and the quieter deposition of the former while the animal is taking its
periodic rest, and building up a septum.

The siphonal funnel is always spoken of as belonging to the septum,
and not to the siphon proper. This is true of the adult, but not of the

* Plate IV, Fig. 4, R.

94 BULLETIN OF THE

young. The siphonal coecum is entirely made up of the funnel of the
first septum. The funnels of the second and third septa are excessively
long, but in the fourth a decided decrease in this respect is noticeable,
and on the fifth septum it assumes very nearly the short aspect of the
same part in the full-grown shell. The lips of the funnel incline inward,
resting upon, and surrounded by, the looser-textured wall of the siphon,
which reaches exteriorly considerably beyond the lips of the funnel in
the young, and in the adult sometimes to the bottom of the septum it-
self. Barrande describes the length of the funnel as exceedingly vari-
able in the Silurian Nautili; but this does not appear to be true among
the modern forms. I was unable to determine whether the funnel in the
Jurassic Nautili was longer in the young than in the adult. It is proba-
ble, however, that, as in other characteristics, this will be found to have
acquired greater constancy, as we proceed in time ; in fact, the funnels
of the young Nautilus Pompilius, invariably longer than those of the
adult, show that the length has not only become more constant, but has
even been reduced, in accordance with the law of acceleration, to an
embryonic characteristic.

Sandberger and Barrande, who have studied the siphonal funnel more
than other authors, use a somewhat different terminology, the former
speaking of it as a funnel and the latter as a conical neck. I prefer
Sandberger's term, because it seems to me to express the form quite as
well as the latter. In all the species that I have examined, whether
Nautili, Goniatites, or Ammonites, the aperture is always somewhat
wider than the bottom, and one side, the ventral, inclined, the opposite
being more abrupt.

In other Nautili, however, the funnel form is more distinctly expressed,
as in Nautilus zic-zac. Here the structure which I have described, tin;
continuity of the outer layer of the siphon or sheath with the septum,
is most, plainly expressed. The siphonal funnels in the adult extend
posteriorly, as in the young Nautilus, to the opening of the next younger
funnel. They become narrower posteriorly, until near the junction, and
then a tumid band plugs up completely the siphon, some distance outside
of, or posterior to, the flaring sides of the funnel. In other words, the
funnels set into, and not upon, each other, like a pile of the necks of
broken bottles, when the septa are fractured in order to expose them to
view. The shell of the funnel throughout is of the same clo-e, pearly
texture, except the swollen band at the junction. "When this is more

MUSEUM OF COMPARATIVE ZOOLOGY. 95

closely examined, it is seen to be composed of two portions, the tumid
external sheath of looser granular texture, and the narrowing neck of
the funnel. Thus the tumid band is really, though so narrow, the repre-
sentative of the external sheath of the siphon in Nautilus Pompilius, and
performs the same function of uniting the siphonal funnels, and protect-
ing the continuous horny layer of the interior. In other Nautili, such as
Nautilus lineatus of the Jurassic, the siphonal funnel becomes a ring or
section of a cylinder, flaring or inclined outward both above and below
like an eyelet.

Inside of the funnel and the sheath in Nautilus Pompilius is a
layer, which is also of a loose texture similar to the sheath, and inside
of this, the continuous dark corneous layer. The intermediate layer
has not been previously observed, and I have unsuccessfully endeav-
ored to find it in the siphon of the adult. The condition of the
specimen was such that I cannot now be sure that it is really a distinct
layer.

The sheath in the adult is not simply of a loose granular texture, but
looks more like the rough surface of a sponge pierced with holes,
which are visible with an ordinary magnifier in desiccated specimens.
Externally, the sheath is covered by the lining membrane of the cham-
bers, and has a pearly aspect in many specimens. The thickness of the
sheath and its color vary greatly. In one specimen it was so exces-
sively thin and porous, even in the adult, that the color of the corneous
layer shone through. At an older stage, however, it assumed the usual
opaque aspect. Viewed from the interior, it is usually smoother, pearl v,
and shows broad bands, probably bands of growth. If so, the animal
must progress neither quickly from septum to septum, as supposed by
Owen, nor slowly and constantly, as described by Valenciennes, but
probably with many intermediate periods of arrest marked by the
deposition of these bands.

The siphonal ccccum of Nautilus lineatus* occupies a different position
from that of Nautilus Pompilius. The coecum lies against the inner side
of the area of the cicatrix near the centre in Nautilus lineatus, and both
.-ides swell out under the first septum as on the dorsal side of the same
organ in Nautilus Pompilius. The siphon between the first and second
sepia makes a curve like the central portion of the letter S, first inclin-
ing ventrally, and then bending again towards the centre. The position

* Plate IV, Fig. H>.

96 BULLETIN OF THE

of the coccum nearer the centre accords with the situation of the first
septum, and its parallelism to the area of the cicatrix, but it is still, as in
the recent Nautilus, considerably nearer to the ventral than to the dorsal
side. Owing, however, to the position of the first septum with relation
to the angle or apex made by the bending of the external shell around
the ventral edge of the first septum, the coecum is not so near the so-
called apex of the whorl as in Nautilus Pompilius. In other words, it
lies nearer the centre of the actual apex, the centre of the cicatrix, and
farther away from the apparent apex, or angular termination of the
outline, than in Nautilus Pompilius. The peculiar abrupt curves which
it subsequently makes are due partly to the abruptness of the ventral
side of the coccum, and partly to the angle at which the first septum lies
with relation to the second. All of these characteristics are thus shown
to be dependent upon the altered or accelerated development of the
septa in the young of Nautilus Pompilius, which causes the first septum
to assume a position and other characteristics similar to those of the
third septum of Nautilus atratus and lineatus. Another fact in the
same direction is the ventral position of the coecum in the Jurassic and
existing species, as contrasted with the central siphon of the young of
the Silurian and Carboniferous Nautili. The extreme variability of
position of this organ among the adult forms of the Silurian of course
renders this characteristic somewhat doubtful, but it is a curious fact that
we should find it so strictly accordant or correlative with the other char-
acteristics already described. It must evidently be added to our other
list of characteristics, which though at first variable, become in course of
time fixed and invariable, through the action of the law of acceleration.
The siphon of Goniatites differs in a remarkable manner from that of
Nautilus. It has a long conical termination which penetrates the first
septum, and lies so close to the abdomen as to form a very decided fis-
sure, probably due, however, as previously explained, to the removal of
the shell. Otherwise the siphonal coecum would be open on the ab-
dominal side, which it does not seem to be when seen from the side, and
covered by the shell. The coecum above is flask-shaped, the neck of
the flask lying between first and second septa. The first septum forms
the round bottom of the flask and the closed conical prolongation ; the
second septum, the neck' and part of the body of the upper portion.
The neck, however, continues to decrease in size until it reaches the
third septum. The siphonal funnel is apparent even in the first septum,

MUSEUM OF COMPARATIVE ZOOLOGY. 07

as may be seen in the ventral view of this part in Goniatites diadema,
on Plate IV, where the sutures incline posteriorly at their junction with
the ccecurn. Thus, instead of being entirely out of the ovisac as in
Nautilus, whether recent or fossil, the siphonal ccecum is to a considerable
extent developed within the ovisac. In other words, this part is devel-
oped and completed later, as is also the first true septum, of which it is
an integral part, in the existing Nautilus, than in the fossil Goniatites,
certainly as early as the Carboniferous and perhaps earlier. I should
expect, however, to find some change in this respect in the Silurian
forms of Goniatites.

The ccecum is made up of the same elements, if we consider that
the internal corneous element has probably been destroyed in the course
of fossilization.

The siphonal funnel at a later age becomes more distinct, and is not
confounded with the sheath of the siphon as it is in the first and second
septa, and perhaps the third also, but this last fact could not be deter-
mined satisfactorily. At a later stage the funnel is very distinctly seen,
and, as described by Guido Sandberger, is continuous on the ventral
side, though it lies almost its entire length against the shell. Barrande
has described this occurring in the same manner in the Goniatites of
the Silurian, and thus accounts for the appearance which deceived Von
Buch. This author supposed that one of the main distinctions between
Goniatites and Nautili lay in different positions of the siphon. He
removed the shells of certain Goniatites and Ammonites, and apparently
exposed the siphon lying in immediate contact with the shell, whereas
he had really torn off the ventral lip of the siphonal funnel with the
shell. Barrande, on removing the shells of Goniatites, found that this
frequently occurred, and thus established the truth of Guido Sandberg-
er's observations.

The posterior edges of the funnel * rest in and upon the siphonal
sheath very nearly as in Nautilus Pompilius, and this sheath, also, as
in the latter, connects the siphonal funnels of adjacent septa, as de-
scribed in Nautilus.

The siphonal ccecum of Ammonites possesses very nearly the same
form as in Goniatites, but is, in the species examined, perhaps somewhat
flatter dorso-abdominally. It has the same position, but. yet no difficulty
was experienced in removing the shell without disturbing the ccecum,

* Plate III, Fijr. 8, P.

98 BULLETIN OF THE

or breaking the continuity of the suture of the first septum on the ab-
domen. The prolongation into the ovisac was not discernible on the
abdomen, except in one specimen, and from the side not so distinct
as in Goniatites.* The figures of this portion, owing partly to the great
thickness of the specimen, and to its more attenuated structure, were
completed with great difficulty, and only after repeated observations.
By the use of dark ground illuminations and a powerful condenser, the
coccum extending below the first septum was seen from the abdominal
side in Fig. 5, Plate I, though the walls seemed broken and partly de-
stroyed; the cone, however, Was not made out. Longitudinal sections
of several specimens gave substantially the same results as the single
one figured. The coecum is formed as in Goniatites by the siplional
funnel of the first septum, but the conical prolongation does not open
into the ccecum,t and its interior is filled cither with a succession of
other cones, or by a number of pillars stretching across its interior. The
peculiar aspect of this portion in the specimens examined has suggested
an explanation of this remarkable modification. When the first septum
was formed, it may have been composed entirely of the thin membra-
neous layer on the ventral side of the coecum, and the first thick layer at
y in the figure ; as the siplional sheath was built up or thickened, it was
carried forward on the ventral side, the different successive layers or
partial septa from y upwards, marking the resting-places in this transit
until at 1 e the first septum was completed. Whether this be so or not,
the conical part of the coccum has not, as far as my observations go,
any decided connection with the coecum when closely examined in
section, though when seen from the side, as in Plate I, Figs. 3, 4, a
connection appeared probable.

Von Buch pointed out the anterior direction of the so-called siplional
funnel of Ammonites, and has been followed by all authors since his
time. The use of the microscope, however, readily detects here a very
curious error. The anterior siplional funnel of Ammonites is not iden-
tical with the true funnel of Goniatites and Nautilus, but an additional
organ. { In the adult nearly the whole thickness of the septum bends
in an anterior direction, reversing the shape of the lips of the funnel.§
In the }<nnig, only a small part of the septum bends anteriorly, and the
funnel is partially maintained, while in the very youngest stages no such

* Pie, I, Figs 3,4. t Hate n, Fig. 4,

i [Mate II. n- 1 v. § Plato II. Fig. 5.

MUSEUM OF COMPARATIVE ZOOLOGY. 99

anterior extension is apparent.* In all cases this anterior deflection of
the septum forms an open collar surrounding the sheath. The lining
membrane of the chambers, of course, surrounds it as well as the sheath,
but otherwise it cannot be considered as connected with the sheath. The
difference between Ammonites and Goniatites consists, then, so far as
the adult siphon is concerned, in the possession by the former of a more
comjdetely rounded siphonal coccum and the siphonal collar. Besides
the earlier development and different form of the siphonal ccecum in the
Goniatites and Ammonites, as compared with Nautilus, we find extensive
differences in the formation of the siphonal funnel. This, instead of
lining the interior of the co3cum by its extension from the second sep-
tum, as in Nautilus, only reaches the mouth of the opening through the
first septum, and in the succeeding septa, instead of reducing the length
gradually, the funnel becomes at once, in the third septum, very short
and distinct from the sheath. They all three, Nautilus, Ammonites,
and Goniatites, agree, however, in having, during the earlier stages of
development, a siphon formed by a ccecal prolongation of the first sep-
tum. The large size of this ccecum, as compared with the area of the
first and second septa, is also an important fact in this connection.

If there is any truth in the application of embryology to the solution
of the problem of evolution, or even of the relative rank of forms, it is
evident that in either sense the true prototype of the Cephalopods
must have these characteristics, namely, a large siphon, composed
of circular prolongations of the septa, or siphonal funnels closed at
their posterior ends. Consulting the development of the simpler Nau-
tilus, we see also that these siphonal funnels should set one into another,
like a pile of cups, or cones, and the septa be concave and very
shallow, as in the Jurassic and Carboniferous species. The conditions
are partially fulfilled by the genus Endoceras, whose septa are shallow
and concave, and the siphon consists of a series of cones, placed one
within another, and closed at the posterior extremity. These cones are
not strict siphonal funnels ; the funnel portion really only extends from
the opening of one septum to that of the next, and the cones, which
are evidently the homologues of the sheath of Nautilus, are built
against the continuous wall thus formed, as partitions in the siphon
itself. These are the characteristics of the adult only, and it can be
reasonably anticipated that the young of Endoceras would exhibit a

* Plate II, Fig. 1.

100 BULLETIN OF THE

siphon composed wholly of coccal prolongations of the septa them-
selves.* The discovery of the adult corresponding to the young of
Endoceras is yet to be made, and even the young of Endoceras has, I
believe, never been described. A fragment of an Endoceras, the
Orthoceras duplex of Verneuil and Keyserling, can be even more
closely compared with the young of Nautilus, than our American
species. In this the septa bend posteriorly, forming extraordinarily
long funnels. Instead of simply connecting adjoining septa, as in the
ordinary forms, the funnel overlaps and extends just twice as far, to a
point opposite to the second younger septum, posterior to that from
which it takes its rise.f This compares with the tendency of Nautilus
to increase the length of the funnel in the young, and the slighter dis-
tinction which exists between the texture of the sheath and the siphonal
funnel itself, until at the earliest period, the siphonal ccecum is wholly
composed of the siphonal funnel closed at the posterior end. Evidently
a similar method of development is to be anticipated in Orthoceras
duplex, and as the funnels extend so far, we ought also to find more
than one ccecum. That is just as in the young Nautilus Pompilius we
find that the sheath of the second septum extends into the siphonal
coecum and is closed, forming really a second ccecum ; so, in the young
of 0. duplex the siphonal funnel of the second and third septa at least
will probably be found fitting into the siphonal ccecum, and closed at
the bottom.

I have industriously examined the young of Orthoceras in order to
ascertain whether they too had any close resemblance to the adult of
Endoceras ; so far, however, my search has been fruitless, probably
owing to the unfitness of the forms which have come into my hands
for microscopical examinations. In the group of the Vaginati to which
Endoceras belongs, it is common to find the huge siphon filled to a
greater or less extent by calcareous deposits, differing considerably
from those filling the living chamber and the septal chambers. Bar-
rande, with his masterly grasp of facts, has demonstrated that these
deposits are made by a posterior prolongation of the body of the
visceral sac, and are not the result of fossilization. Barrande, also, con-
siders the cones of Endoceras to have been deposited in a similar

* If, indeed, any septa remain in the extreme young.

t De Verneuil and Comte de Keyserling, Russie et l'Oural. Vol. II, Plate XXIV,
Fig. 7.

MUSEUM OF COMPARATIVE ZOOLOGY. 101

manner, and attributes the spaces between them to the suddenness with
which the animal arose from one resting-place to another, which did
not permit the secreting surface to fill up the entire tube. Whether
this is the true explanation or not does not here signify, since the
main points of structure are the same, and the siphon is closed, accord-
ing to both authorities.

Barrande has also, in his immortal work on the Silurian Cephalopoda
of Bohemia, given, with his customary fulness and accuracy, a com-
plete analysis of the elements of form and structure among the Nau-
tiloids. He has, however, settled upon Ascoceras as the prototype,
regarding the Vaginati as the nearest allies of Ascoceras. In describ-
ing the structure of Ascoceras, this author acknowledges the existence
in the young of simple concave shallow septa, pierced by a true siphon,
which opens into the bottom of the living chamber, as usual in all the
Tetrabranchiata.

On one side of the living chamber a series of septa are built up,
whose sutures reach only partially around its circumference, and the
septa themselves in the interior are equally incomplete. The imper-
fect septal chambers thus formed do not open into the living chamber,
but are closed by the bent edges of the septa, whose free internal
borders bend posteriorly until they come in contact with each other.
These posterior prolongations are, in Barrande's opinion, the equivalents
of the siphonal funnels of the Vaginati, many of which group have
siphons open on one side, and constructed not Unlike this large posterior
prolongation of the living chamber in Ascoceras. There would be not
the slightest hesitation in accepting this opinion, if it were not for the
siphon and perfect septa existing in the young. This, according to the
laws of development, as they are now understood, would constitute the
Ascoceras a degraded type, one which like the Cirripeds among
Crustacea, had developed into a structure simpler and of a lower
zoological rank, than if the growth had been arrested at a compara-
tively early period. This would meet one of Barrande's principal,
reasons for considering Ascoceras as lower than the Orthoceratites,
which is, that it certainly could not be placed above them, and really
possessed, in its immense, incomplete, adult siphon, or posterior pro-
longation of the living chamber just described, a much simpler struc-
ture than any of the other Nautiloids. There is probably but little
doubt that the Ascoceratites, as claimed by Barrande, have all the

102 BULLETIN OF THE

elements of structure found in subsequent forms besides their greater
simplicity, but this can be accounted for best by considering them as
forms having a truly retrogressive development, what has usually been
described as a retrograde metamorphosis.

The types of Ascoceras which I have seen are much too imperfectly
preserved to enable me to speak from my own knowledge of the nicer
points of structure described by Barrande, but that author's unsur-
passed and minutely accurate figures supply all deficiencies. One
peculiarity of the structui'e appears to be very unfavorable to the view
here presented and is so considered even by the author himself; this is
the want of connection between the young siphon and the supposed
larger siphon of the adult. The first of the imperfect septa bends
posteriorly, as do the others, but is not, however, discontinued when it
reaches the last of the entire septa of the young. It is prolonged over
the surface, and is really a complete septum, not pierced by any si-
phonal opening. We cannot imagine any normal progressive mode of
growth by which the minute so-called siphon of the young Ascoceras
could be changed into the huge visceral prolongation of the living
chamber of the adult, without some of the intermediate steps of this
change being visible, and some connection maintained with the siphon
of the young. Barrande also states that in no instance has he found
any more than one of the perfect elliptical septa of the young pre-
served ; that over this he has observed the striations and markings of
the external shell ; and further, that the so-called minute siphon, or
elliptical funnel, penetrates this septum. Now this condition is precisely
what we should expect to find, if a portion of the exposed lower end
of the Ascoceras, or area of the first septum represented the scar left
by the ovisac on the apex of the whorl, as in Nautilius Pompilius.
This would not only account readily for the presence of the strirc of
the shell upon the exterior, but also for the projecting end of the so-
called siphon. The latter would then represent the siphonal coecum,
which, as in Goniatites and Ammonites, penetrated the first septum,
and by the removal of the ovisac had been left exposed.

Of course this explanation can only be considered as a suggestion
calling for a re-examination of some of Barrande's fine specimens,
and is quite as likely to be overthrown as to be confirmed by such a
process.

Whether this be so or not, the young Ascoceras was evidently, as

MUSEUM OF COMPARATIVE ZOOLOGY. 103

described by Barrande, in possession of a siphon, or the representative
of one in the young, and occupied the entire cavity of the living-
chamber; whereas, it subsequently lost this siphon, and built up a
portion of the living-chamber with imperfect septa, leaving a cylindrical
hollow, which was evidently occupied by the visceral sac, thus plainly
retrograding in structure, and undergoing retrograde metamorpbosis,
like Orthoceras in the young, and perhaps somewhat similar to Endo-
ceras, or others of the Vaginati in the adult.

THE SHELL.

The shell of the young Nautilus, at the apex of the first whorl, consists
of two layers, an imbricated internal nacreous layer, and a layer of
denser texture.* The internal layer is at first a single plate deposited
at the apex. The zones subsequently secreted overlie this internally
on both the dorsal and ventral surfaces, and from this point the imbri-
cated structure is maintained throughout. This shows that the internal
layer is entirely deposited from within, and probably by the border of
the mantle. Whether seen in transverse or longitudinal sections, it
presents the crenulated aspect common to nacreous shells. It is con-
siderably thicker on the dorsal side, owing to the longer time which
the mantle must take in passing a given point on the dorsal or inner
side of the spiral. This thickness is increased by the manner in which
the septa extend and overlap their borders on the same side, though in
the figure this is not shown, because the cut extended through the side
of the dorsal lobe. Subsequently, as the whorl grows larger, this
difference between the absolute thickness of the shell decreases, and in
the adult they are about the same on either side.

The external layer is quite thin on both surfaces in the young, but
becomes also rapidly thicker on the dorsal surface, and assumes a
denser and more opaque structure than the interior. A break in this
layer on the dorsal surface just before it completed the first revolution,
and came in contact with the apex,f shows that this layer was probably
also deposited by the border of the mantle. At this point the layer is

* Plate IV, Fig. 4.

t The break is marked by a very faint line in the figure, near the angle of the
umbilical perforation. This has accidentally been made too slight in the drawing;
the break is much thicker and more decided.

104 BULLETIN OF THE

imbricated, and the direction of zone the same as those of the internal
layer.

"When the first revolution is completed, the hood begins to deposit
the dark layer on the external surface of the ventral side of the first
whorl, and the true external layer becomes very thin, and is often
hardly definable, appearing in the section as a very faint band next to
the black deposit of the hood. On the ventral side, however, it steadily
increases in thickness. It is this layer also which attains such ar
excessive development, and composes the principal part of the shell
close to the umbilical border. The brownish color of the exterior does
not seem to be distinct, but rather to be intimately blended with this
layer. A faint tinge is visible before the completion of the first
volution, probably beginning somewhere between the fourth and fifth
septa, which gradually extends in breadth and thickness, until it
assumes the aspect which distinguishes the adult. The bands on the
sides, however, were not observed in the young.

Valenciennes has fully described these three parts in the adult, dis-
tinguishing the internal and external layers, and also the outer colored
portion of the last, which, however, he described as a distinct layer.
He regarded the two last as confined to the exposed sides and ventrum
of the adult, and for this reason attributed their production to the
ventral arms of the animal. I am at a loss, however, to account for
their imbricated structure, precisely similar to the internal layer, or for
the presence of these layers upon the dorsal side of the young, unless
they have been deposited from the interior by the edge of the mantle,
as were the zones of the internal layer. The replacement of the
mantle-edge by the hood when the first revolution was completed, as
shown by the appearance of the black deposit, would, if this view were
accepted, account for the very slight traces of the external layer left
on the dorsal side after the hood came in contact with the apex of the
first whorl. A fine specimen of Nautilus Pompilius in the possession
of the Museum, with the contained animal in an excellent state of
preservation, has been examined, and the structure of the mantle-edge
confirms the view here taken. The entire mantle-border is thickened,
as described by previous authors, but the edge on the sides and ventrum
has a somewhat different structure from that of the dorsum. The edge
is tumid and divided, as in the Lamellibranchiata, into lips, containing
in the channel between them a brownish substance, probably the re-

MUSEUM OF COMPARATIVE ZOOLOGY. 105

mains of the animal matter which colors the external layer. This
brownish matter is friable and not a consistent horny membrane or
epidermis. The mantle-edge on the dorsal side, however, is thin and
lies posterior to the hood described by Owen and Valenciennes which
secretes the black layer. That on the ventrum is much thicker, with
tumid border and one channel. Farther down on each side, and cor-
responding to the increased thickness of the shell, the border becomes
exceedingly thick with two channels, an outer and an inner, each still
partly filled with soft matter, evidently unformed shelly excretions,
whereas the dorsal border has no such material left in its channel.

Barrande states that the dark layer of the hood is wanting in the
Silurian Nautili, and this accords admirably with the continuity of all
the layers on the dorsal side in these species, and also in the slightly
involute Carboniferous Nautili, which I have examined. Like the size
of the umbilical perforation, and the concavity of the dorsum, the
presence or the absence of the dark layer and the extent of the exter-
nal layers depend upon the greater or less closeness and increasing
constancy of the involution of the whorls. It could not be inferred from
this, however, that the fold of the mantle or hood was absent or present
in the same proportion. The black deposit simply indicates, so far as
we know, the secreting power of the hood, and shows that this organ came
in contact with the whorls, but nothing more. The external layer is
visible with a common hand lens in Nautilus lineatus of the Jura, and
exhibits characteristics similar to those of Nautilus Pompilius.

Lining all of the chambers, and the exterior of the siphon, is an
exceedingly thin membrane, which upon the septa, and the sides of
the shell proper, though not upon the siphon, is generally, but not in-
variably, connected with an equally thin layer of nacre, and in the desic-
cated specimens ; they may be observed peeling off together. Even in
Jurassic fossil shells, it is distinct and easily traced, probably owing to
its connection with the nacreous deposit described above.

The same layers are traceable in the ventral and lateral sides of the
shell of Goniatites. All the layers * consist of imbricated plates, laid
on internally. There is in many specimens a dark-colored layer,
equivalent, however, to the external layer of Nautilus, and like that,
invested by a layer which corresponds to the smooth, colored layer of
Nautili Pompilius. In one specimen, figured on Plate IV., I was for-

Plate IV, Fig. 11.

106 BULLETIN OF THE

tunate enough to find in section all these parts, where a transient
mouth had been formed, and the deposits continued from this break.
Here all of these layers are decidedly imbricated ; even the thin exter-
nal colored zones show their imbrications in its external portion as
zones of growth, evidently due to internal deposition. Keyserling has
described and Santlberger figured what they have called the wrinkled
layer, lying between the involved ventral and the dorsal side of the shell
of the next whorl. This I have not seen, nor succeeded in detecting
anything analogous in the specimens I have examined. The figures of
Sandberger, however, are known to be very accurate, and the wrinkled
layer is plainly shown, limited to the involved portion of the whorl,
and not extending outside of these limits. The specimens I have
examined with success are all young, and it may bo that this layer is
absent from the younger periods of Goniatites. Besides the two layers
there is, as in Nautilus, an internal lining, layer, or fossilized membrane,
visible in some specimens, and this, with the prolonged edges of the
septa, appears to be all that is deposited on the dorsal side. Not only
the external layer is wanting on the involving dorsal side, as in Nau-
tilus, but the internal also.

The shell of Ammonites also consists of two layers * besides the in-
ternal lining layer, which both here and in Goniatites, may be often
seen to pass between the septa and the shell, as it really does also in
some instances in Nautilus. f It is evidently due to the mantle, which
deposits a film upon the surface whenever it rests for a sufficiently
long period in any one chamber of the whorl. I was also unable, in
the young, to observe any deposit similar to the wrinkled or black layer
of the Nautilus or Goniatite. J Only the internal lining layer of the
chambers thickened by the edges of the septa is deposited on the
enveloped region or dorsal side as in Goniatites.

The shell in Ammonites, as in Goniatites, also extends over the
ovisac on the abdomen and the sides. This, at first, led me to suppose
that here, at least, it was deposited by the arms of the animal, as in
Argonauta, and as supposed by Valenciennes, for the external layer in
Nautilus. But nothing in the structure of the shell itself confirms this

* Plate IV, Figs. 1, 2, 3.
t Plate II, Fig. 1.

J The wrinkled layer undoubtedly exists in the adults of several species, especially
in Amaltheus margaritatus, as described by Sandberger.

MUSEUM OF COMPARATIVE ZOOLOGY. 107

view. The ovishell of Deroceras planicosta, as figured below, * is com-
posed of two layers, — the internal lining layer, and an external thicker
layer, which is not continuous with any of the layers in the shell of
the first whorl. The internal lining layer bends internally, and coats
the inner side of the first septum. The external layer is very thin
upon the dorsal side, and was perceptible only by the aid of a fif-
teenth. The increase in thickness of the external layer is cpiite abrupt
upon the abdomen, but not so abrupt as upon the sides, especially at
the edges of the embryonal umbilici. f The external layer reaches
beyond the neck of the ovisac, overlapping the edges of the layers
which compose the apex or beginning of the first whorl. These are
evidently the product of the mantle-edge, and show that the external
layer of the ovisac must have been secreted from the interior by the
body of the embryo. The shell of the whorls consists at first of two
layers, besides the internal lining layer of the chambers. J These,
unlike the external layer of the ovisac, to which they collectively
correspond, are absent from the dorsum. The presence of the external
layer on the dorsum, and its extreme thinness there, as compared with
what it is upon the abdomen and sides of the ovisac, is a characteristic
also of the adult of Nautilus.

The absence of all the external layers, however, from the dorsal side
in Goniatites and Ammonites, appears to begin at the earliest period
in the closely involved species. The septa upon that side in the latter,
from the first to the sixth inclusive, extend their borders considerably,
and build up quite a dense thick layer, which, however, ceases with
the seventh septa. § After this, only the lining layer, probably accom-
panied by a thin nacreous deposit, is found on the dorsal side. A
close examination of the layers showed everywhere indications of arrests
of growth, and the formation of more or less permanent mouths, || or
zones of growth, always, however, imbricated as in Goniatites and
Nautilus. The last formed, beginning inside of the edges of the earlier
deposited zones, a structure entirely at variance with the supposition
that any portion of the shell was laid on from the exterior.

Comparisons have also been made with the shell of Argonauta, which

* See woodcuts on last page of the text.

+ Plate II, Fig. 3.

J See last page of the text.

§ Plate II. Fig. 1, and last page of text.

|| Plate IV, Fig. 1.

108 BULLETIN OF THE

is known to have been, partially at least, deposited by the enveloping
arms of the animal. The shell consists of three layers ; the inner and
outer layers have a similar structure. These two are composed of
minor plates or zones of variable lengths, but these are never im-
bricated, they are simply laid one upon another.* The median layer
is very irregular when seen in longitudinal section, showing thicker
and thinner portions, the thicker being somewhat more opacpue than
the thinner parts. In a transverse section, this layer presents a similar
aspect, but the structure varies from a smooth, even shade, to a granular
aspect. When viewed from above, this layer presents a reticulated
aspect, due to the presence of numerous white opaque thread-like lines
of growth. These are parallel with the lips of the shell, notwithstand-
ing their minor irregularities, and may be, in some instances, followed
for a considerable distance across the whorl. Especially when they
represent the former edge of some one of the numerous mouths,
marking the periodical arrest and renewal of the growth. They
bend posteriorly on the abdomen, and anteriorly on the sides, and are
covered everywhere by the internal and external layers which possess
no such marks of growth. There are no marks of an imbricated
structure, and the absence of lines of growth from the external and in-
ternal layers, as well as their presence in the median layers, shows also
that the structure is entirely distinct everywhere from the shells of
either Nautilus, Goniatites, or Ammonites. It is evident that the in-
ternal layer is deposited by the body of the animal internally, the
median layer by either the mantle-edge, or by the anterior edge of the
abdominal arms, externally, in successively thicker or thinner, but more
or less irregular zones, and lastly, the external layer probably by the
inner side of the expanded portion of the arms.

* Plate IV, Figs. 12-16.
Cambridge, June 5, 1872.

MUSEUM OF COMPARATIVE ZOOLOGY. 100

Figure I.

Deroceras plantcosta. Two septa showing the siphonal funnel, P;
the dorsal lobes, H ; and the ventral cell, J.

Figure II.

Deroceras planicosta. Shell of embryo, d', lining layer; X, external
layer ; d", median layers of the apex of first whorl, an internal transparent
and a thicker darker layer, both overlapped by X; d'", external layer of first
whorl next to dark layer of d" ; d, additional layers deposited by the dorsal
portion of the body of the animal in passing from neck of ovisac to seventh
septum ; le, first septum.

Figure III.

Nautilus Pompilius. An ideal outline of the ovisac and apex of first
whorl. A, ovisac; 1c, first whorl; W, shoulder formed by the external
ridge of the cicatrix ; X shows where the shell of first whorl connected with
the shell of the embryo, and also where it is always found broken. The two
dotted lines from X show the area of the scar itself as seen from the side ; T
is the line of involution showing the extent to which the apex is enveloped at
the first revolution of the shell.

110

BULLETIN OF THE

Fig. II.

Fig in.

MUSEUM OF COMPARATIVE ZOOLOGY.

Ill

EXPLANATION OF SIGNS USED IN THE PLATES.

A. Ovisac.

B. Ovishell.
C Whorls.

lc. First whorl.
2c. Second whorl.

D. Shell.

d\ Lining layer.

d". Median layer.

d'". External layer.

d+. Black deposit or layer of the hood.

E. Septum.

le. First septum.
2e. Second septum.

F. Abdominal lobe.

G. Lateral lobes.

g. Superior lateral lobes,
g". Inferior lateral lobes,
g'". Auxiliary lateral lobes.
H. Dorsal lobe,
h'. Superior dorsal lobe,
h". Inferior dorsal lobe,
h''. Auxiliary dorsal lobes,
h. Minor lobes.

J. Abdominal cell.

K. Lateral cells.

k'. Superior lateral cells.

k'\ Inferior lateral cells.

k'". Auxiliary lateral cells.

L. Dorsal cell.

1'. Superior dorsal cells.

1". Inferior dorsal cells.

I'". Auxiliary dorsal cells.

M. Marginal lobes.

N. Marginal cells.

O. Siphonal fissure.

P. Siphonal collar.

P. Siphonal funnel.

Pi. Siphonal coecum.

S. Siphon.

s'. Horny layer of siphon.

s". Accessary layers of siphon.

s'". Sheath.

U. Embryonal umbilicus.

y. Cone of the siphonal coecum.

X. General mark.

Plate I.

Deroceras planicosta.

Fig. 1. Front, or dorsal view of ovisac, with a portion of first whorl, magni-
fied -f- 80 diameters. A, ovisac ; U, umbilicus ; le, first septum ;
2e, second septum ; R, siphonal coecum ; O, siphonal fissure.

" 2. Abdominal side of same specimen.

" 3. Section of the same, the ovisac worn down somewhat, and the fir6t
whorl split off still lower, giving section of siphonal coecum. D,
Shell ; S, beginning of true siphon.

<( 4. Side view of the same, showing also the third septum 3e. Magni-
fied 123 diameters.

" 5. Dorsal view of another specimen, magnified -f- 80 diameters.

" 6. Abdominal side of another specimen; lc, first whorl; X, lines in
the interior, probably indicating sutures made by the intersection
of first septum, or siphonal coecum with the shell of the first whorl.
Magnified -|- 80 diameters.

" 7. Same specimen before being so completely reduced ; only the sides
of the older whorls are worn down, admitting the light to the ex-
posed portion of the ovisac, and the embryonal umbilicus. Magni-
fied-80 diameters.

" 8. Another specimen with the septal outlines for four and a quarter
whorls. Magnified 21 diameters.
All of the above, from 1 to 8 inclusive, were taken from the same
block of dark blue limestone. Loc. Wiltshire, England.

GOXIATITES FECUNDUS.

" 9. Young of circular variety. Barrande. Ceph. Bohemia. Plate

XI. Fig. 2.
" 10. Young of elliptical variety. Ibid., Plate

XI. Fig. 4.

GONIATITES AFTER SANDBERGER'S FIGURES.

Fig. 11. Young Gon. compressus. Sandberger, Jahrb. d. Nass. Verein. 1851,

p. 15, PL III.
" 12. Young Gon. subnautilinus. Sandberger, Jahrb. d. Nass. Verein.

1851, Fig. 27.
" 13. Young Gon. canaliculars, var, gracilis. Sandberger, Jahrb. d. Nass.

Verein. 1851, Fig. 28.
« 14. Young Gon. sublamellosus. Sandberger, Jahrb. d. Nass. Verein.

1851, Fig. 29.
" 15. Young Gon. lamed, var. latidorsalis. Sandberger, Jahrb. d. Nass.

Verein. 1851, Fig. 30.
" 16. Young Gon. lamed, var. calculiformis. Sandberger, Jahrb. d. Nass.

Verein. 1851, Fig. 31.
"17. Young Gon. planorbis. Sandberger, Jahrb. d. Nass. Verein. 1851,

Fig. 32.
" 18. Young Gon. diadema. Sandberger, Jahrb. d. Nass. Verein. 1851,
Fig. 33.

A Hyatt. i

>

J/eiv-FiB'

Plate II.

Deroceras planicosta.

Fig. 1. Section of siphon and siphonal coecum. B, ovishell ; 6V, lining layer
of the shell ; d", median layer ; d'", external layer ; d, dorsal
layers of the young; le, first septum; 2e, second septum, and so
on ; R, siphonal coecum ; s', remnant of the horny layer of the
siphon ; s", accessory layer, lining the interior ; s'", sheath.
Magnified 317 diameters.

" 2. Section at right angles with siphon * showing the form of the whorls.
^Magnified 21.5 diameters.

" 3. Centre of fig. 2 enlarged. D, shell; U, embryonal umbilicus. Magni-
fied 70 diameters.

" 4. Segment of the siphon from Plate IV, Fig. 1, on first quarter of the
whorl ; D, shell on the abdominal side ; E, septum ; P, the
siphonal funnel ; P, siphonal collar. Magnified 158.5 diameters.

" 5. Section of same specimen of nearly the same age. Amplification
same.

" 6. Enlarged drawing of one of the sections of the siphon in Fig. 2 ; O
marks the bottom of the siphonal fissure ; J, the lateral wings
of the abdominal cell divided by the fissure, as seen from behind ;
F, the two truncated branches of the abdominal lobe. Magnified
104 diameters.

" 7. Young, with nine first septa delineated, and the amount of the in-
volution of the first whorl around the ovisac is shown by the
shaded line reaching from the broken edge of the whorl. Mag-
nified 80 diameters.
All the specimens described above are from the same block of Wilt-
shire limestone as those given in Plate I, Fig. 1-8.

Arxioceras semicostatum.

Fig. 8. Side view of the young with eleven septa, magnified -f- 10 diame-
ters. F, ventral lobe ; H, dorsal lobe ; g', superior lateral lobe ;
h', superior dorsal lobe ; g", inferior lateral lobe ; k', superior
lateral cell ; 1', superior dorsal cell ; k", inferior lateral cell ; 1",
inferior dorsal cell.

" 9. Side of three and a quarter whorls of an older specimen from the
same locality; magnified 21 diameters.

" 10. Front view or projection of the same. O, siphonal fissure ; J, ab-
dominal cell ; F. abdominal lobe.

ASTEROCERAS OBTUSUM.

Fig. 11. Section of first two whorls and ovisac, magnified 49 diameters.
The left side only of this specimen is perfect.

* This section is not exactly at right angles to the siphon, though the variation is
not perceptible except when Fig. 3 is drawn. Then the embryonal umbilicus is seen
to be absent from one side. This, however, is too slight a deflection to affect materially
the truthfulness of the general outline.

.

I Hyutt Dei

E Koitopicky.ltih

Yew£ng/.itfi C° Bostt

Plate III.

Nautilus Pompilius.

Fig. 1. Front view of the young, showing the cicatrix, and area of attach-
ment of the ovisac, the siphon, portions of the fourth and fifth
septa, the siphonal funnel with its abdominal channel or gutter,
and part of tbe side and dorsum of the continuation of the first
whorl, which has been left attached to the apex. The shadow
cast by the broken edges of the fifth septum on the upper part
of the siphon gives the exact outline of the abdominal cell,
which is bidden behind the siphon. Magnified 8 diameters.

GONIATITES ATRATU3.

Fig. 2. Section through the centre of the siphonal ccecum and siphon. Loc.
Choquier. Magnified 21 diameters.

GONIATITES DIADEMA.

Fig. 3. Front view of the ovisac, with a portion of the first whorl, exhibiting
the false abdominal lobe of the first septum, which is formed by
the breaking away of the ventral side of the cone of the siphonal
coecum. lc, first whorl ; le, first septum ; 2e, second septum.
" 4. Side view of the same species. Loc. Choquier. Magnified 74
diameters.

GONIATITES LlSTERI.

" 5. Section of the ovisac, which has also cut through a portion of the
siphonal ccecum. The approximation of the siphonal coecum and
cone to the shell below the first septum is clearly exhibited, and
a fragment of the second septum on the abdominal side. Magni-
fied 102.5 diameters.

" 6. The siphonal ccecum after a further reduction of the same section.
The neck of the ccecum and remnant of the second septum have
broken away, leaving only that part of the ccecum which was
formed by the siphonal funnel of the first septum. The cone
opens freely into the siphonal ccecum. Magnified 317 diameters.
Loc. Choquier.

GONIATITES CRENISTRIA PllUl.

Fig. 7. Section of six whorls from a specimen of the variety having a flat
abdomen. This shows very plainly the eccentricity of the siphon
which is cut by the plane of the section at different levels in dif-
ferent parts of the specimen, and entirely cut away in the young.
Magnified 15.5 diameters.
" 8. Enlarged view of a siphonal funnel from the same, d', lining layer
of shell ; d", d'", median and external layers of the shell ; P,
siphonal funnel; s'", siphonal sheath. Magnified 104 diameters.
Loc. Rudesheim.

BullM.C.Z. Vol.1

Plate III

A Hyatt Del.

£ Konopicki/'. Lith

JfewEnfflith .C? Boston

Plate IV.

Deroceras planicosta.
Fig. 1. Section showing the imbricated structure of the median and exter-
nal layers of the shell. The lining layer is not represented.
Loc. Wiltshire. From the same block as the specimens of this
species previously figured, d", median layer of the shell ; d'",
external layer. Magnified 40 diameters.

" 2. Section of shell transverse to the siphon at the juncture of the side
of one whorl with the abdomen of another, from same specimen
as PI. II, Fig. 2. d", median layer; d"', outer layer; d', lining
layer. Magnified 104 diameters.

" 3. Enlarged portion from the abdominal side of same section. Magni-
fied 317 diameters.

Nautilus Pompilius.

Fig. 4. Section of young through the centre of embryonal umbilicus and
the siphonal coecum. d, lining layer of the shell ; d", median
layer ; d"', external layer ; d -}-, dark layer deposited by the
hood ; X, the plate which closes the apex of the whorl. Mag-
nified 15 diameters.

Nautilus atratus.
Fig. 5. Side view of Saemann's original specimen, le, first septum.
" 6. Dorsal side of the same, showing the large dorsal lobe of the second
and succeeding septa.

Nautilus Kontnckii.
Figs. 7,8,9. Ventral, dorsal, and side views, exhibiting three septa, and absence
of the rlutings on the apex of the young shell.

Nautilus lineatus.

Fig. 10. Section of the young. The external layer of the apex d'" is too
deeply shaded, and consequently seems to fuse with the dark
layer of the hood, which begins at d+. whereas it here becomes
thinner and passes inside of it, as in N. Pompilius. Fig. 4.

GONIATITES LlSTERI,

Fig. 11. Section of shell, which passes through the edge of a former aper-
ture at the end of the first whorl, exhibiting the imbricated struc-
ture of the layers. Magnified 40 diameters.

Argonauta Argo.
Fig. 1 2. Longitudinal section of shell, but transverse to the fibres of the

BullM.C.Z VoLlll.No}

A Hyatt I 'el

E Konopicky, Lith .

New Fng IitA C° Boston

median layer, d', inner layer; d", fibrous median layer ; d'", ex-
ternal layer. Magnified 104 diameters.

13. Transverse section of the shell, but parallel with the fibres of the

median layer. Magnified 104 diameters.

14. Portion of the last, enlarged 317 diameters.

15. Median layer of the same enlarged to 317 diameters, showing the

granular aspect of another portion of the same specimen, where
the section has cut obliquely through one of the bundles of fibres.

16. View of the median fibrous layer from above. Magnified 21

diameters.

No. 6. — Notes of an Ornithological Reconnaissance of Por-
tions of Kansas, Colorado, Wyoming, and Utah. By J. A.

Allen.

In the following pages are indicated some of the results of field work on
the Plains and in the central portions of the Rocky Mountains ; these re-
sults including more or less complete annotated lists of the birds of nine
quite widely separated localities, with a general summary of the whole.
Although the region in question presents by no means a new field, the
faunal lists here offered form the first special reports that have been
made upon the ornithology of any restricted locality within the region
referred to at the head of this article. The observations on which the
following lists are based were made in 1871 (from May 1 to January 15,
1872), during an expedition sent out by the Museum of Comparative
Zoology to the Rocky Mountains to obtain specimens of the vertebrated
animals of the plains and mountains of the West. The expedition
commenced its work at the Missouri River, in the vicinity of Fort
Leavenworth, and collected at intervals thence westward to the Great
Salt Lake Valley. Mr. Richard Bliss, Jr., of the Museum, accom-
panied the expedition as ichthyologist ; and Mr. C. W. Bennett of
Springfield, Massachusetts, as taxidermist. Both of these gentlemen
rendered important aid in the ornithological work, Mr. Bennett add-
ing to his zeal the qualifications of an experienced collector and a
skilful sportsman. Among the acquisitions of the expedition are over
fifteen hundred birds, representing about two hundred species, besides
large suites of nearly all the mammals of the region visited, including
all the large herbivorous species, large collections of fishes, many rep-
tiles and insects.

An opportunity was thus afforded me of studying many species pf
birds in the field which I had previously seen only as dried skins, and
of examining large series of fresh specimens of many of the puzzling
forms of the middle region of the continent. From this has resulted a
confirmation of all the general conclusions arrived at in my recent
paper on the " Winter Birds of East Florida," * and the discovery of
several well-marked geographical races not previously chronicled. In
the woodlands of Eastern Kansas a decided general tendency to a

* Bull. Mus. Comp. Zool., Vol. II, No. 3, April, 1871.

114 BULLETIN OF THE

greater intensity of color than at the northward was noticed, in accord-
ance with the law of the increase in intensity of color to the southward,*
which in several species was especially marked. The males of the
common indigo-bird ( Cyanospiza cyaned) were not only much more
than ordinarily lustrous, but the females shared the blue tint of the
males to an unusual degree. There was here found also a thick -billed
race of the cardinal (Cardinalis virginianus), which in the size and
form of the bill makes a decided approach to the thick-billed race of
this bird found in Lower California (C igneus auct.). The hairy
woodpecker begins to noticeably resemble the darker (Picks Harrisii)
race of the Rocky Mountains, and Colaptes auratus has quite commonly
a greater or less number of red feathers mixed with the black ones
forming the maxillary patches, thus clearly showing a marked ten-
dency to a differentiation towards the C. mexicanus of the western half
of the continent, at a point some six hundred miles east of the habitat of
that species. A well-marked variation is also noticeable in the Icterus
Baltimore, through the paler colors of the middle wing-coverts, which
in Middle Kansas become either pure white or are only faintly tinged
with pale yellowish, instead of being orange, as in the eastern form.
With this gradual change in the color of the coverts the white edgings
of the remiges become greatly broadened, as in the so-called Par us
septentrionalis, the latter being here less strongly marked in this respect
than further westward. The bill of the Icterus Baltimore is here slen-
derer and rather more decurved than in northern specimens. Pro-
fessor Baird has also recorded a specimen of Pipilo erythrophthahnus
from Fort Leavenworth, which " has a few white spots on the scapulars
only, the wing-coverts without them, exhibiting an approach to P.
arcticus" f and remarks that other western specimens have more than
the usual amount of white on the wings. The Fort Leavenworth speci-
men he regards as "probably a hybrid," between P. erythrophthalmus
and P. arcticus. While only one of our Fort Leavenworth specimens
thus approached P. arcticus, all resembled it in the enlarged hind claw,
and in a more than the usual amount of white on the wing.|

* See Bull. Mus. Comp. ZooL, Vol. II, p. 233.

t Birds of North America, p. 513.

\ In respect to the claws, I find that in the Fipilos of this group there is a decided
enlargement of the claws to the southward, along the Atlantic coast, as well as in the
interior and on the Pacific coast, this enlargement reaching its maximum in Lower
California, in Pipilo megalonyx of this group. Florida specimens have larger claws —

MUSEUM OF COMPARATIVE ZOOLOGY. 115

Passing to the Plains proper, the faded aspect of all the birds is
strikingly noticeable, especially in the species that range across the
continent. The well-known " neglecta " type of Stumella ludoviciana,
the " Henryi " type of Chordeiles popetue, the " rufa " type of Eremo-
phlla alpestris, the " Cassinii " type of Peuccea cestivalis, the " Park-
manni " type of Troglodytes aedo?i, the " septentrionalis " type of
Parus atricapillus, are not only prevalent forms, but corresponding
pallid forms are equally marked in Coturniculus passerinus, Spizella
socialis, Falco sparverius, ^gialitis vociferus, and others ; these pallid
races prevailing throughout the arid plains to the westward. The
same tendency is manifest in the mountains of Colorado, where the
Sitta " pygmcea " forms a similar pale race of Sitta pusilla ; * Zono-
triclda leucophrys, through the greater amount of ashy white on the
lores, passes into Z. " Gambeli " / Geothlypis Macgilllvrayi permanently
retains white spots on the eyelids, which appear in G. Philadelphia
only in the young and in the females. Pipilo erythrophthalmus, through
an acquisition of white streaks on the back and wings, becomes P.

more noticeably that of the hallux — than those from Massachusetts; and in those from
Eastern Kansas they are fully as large as in Florida specimens, while the "P. arcticus "
in Colorado has them still larger. On the Pacific coast the specimens from Oregon
have small claws, the size increasing southward to Lower California, where they be-
come excessively enlarged. This increase in the size of the claws to the southward I
have traced in several other genera, it corresponding with the increase in the size of
the bill in the warmer latitudes, and is doubtless due to a similar climatic cause. See
Bull. Mus. Comp. Zool., Vol. II, pp. 230, 239.

* Sitta " pi/gnuEd" differs from S. pusilla in being everywhere lighter colored; the
head is greenish ashy brown instead of pale hair brown, the back is less deeply blue,
and the white markings on the tail and wings are broader and purer. In <S. pusilla the
middle tail-feathers are generally only slightly paler at their bases, but are sometimes
distinctly white, as they almost always are in S. pygmcea. The oblique white bar on the
other tail-feathers is also much broader and more strongly white in S. pygmcea. In S.
pusilla the edge of the wing is generally pale grayish white, but sometimes distinctly
white, as are also the basal portions of the inner webs of the greater primary coverts.
In S. pygmcea the white on the edge of the wing is not only more strongly marked, but
covers also a larger portion of the inner vanes of the greater primary coverts, and the
concealed basal portion of the primaries also shares the white. The outer edges of the
primaries are also more broadly bordered with white than those in S. pusilla. The style
of markings in the two forms is identical, only that the white is more pronounced and
the general tints paler in S. pygmaa than in S. pusilla, apparently establishing it as a
paler race of the latter, co-ordinate with so many other similar examples of pallid races
in the interior of the continent.

In like manner the western race of Sitta carolinensis (S. aculeata Cass.) has less black
on the inner secondaries than has the eastern form of this species.

116 BULLETIN OF THE

" arcticus " ; and Picoides " americanus " becomes P. dorsalls, through
a somewhat similar increase of white in the dorsal plumage. All the
Vireos of the Rocky Mountain plateau are paler races of species that
range across the continent, the difference in some of them being so
great as to give them the character of strongly marked geographical
varieties. Most of the Einpidonaces are here also similarly represented ;
and farther southward and westward occur pallid forms of Myiarchus
and Tyrannies; everywhere establishing the law of pallid races in arid
regions, which there represent the brighter conspecific forms of the
contiguous moistcr districts. The differences in color between the
conspecific forms of arid and of comparatively moist regions is much
greater, as a rule, towards the end of the breeding season, or just
before the autumnal moult, than after this moult, or in spring speci-
mens, or than is observed between young birds of the two forms ;
showing most unmistakably the direct influence of the intensely heated
dry winds and strongly reflected light upon the colors of birds in semi-
desert regions.

Recent investigations show a rather greater tendency to an enlarge-
ment of the bill to the southward along the Pacific slope of the continent
than that pointed out in my paper on the Florida Birds as existing to
so marked a degree among the birds of the Atlantic States.* Instances
are seen in the southern forms of the Chrysomitris psaltria group, in the
Carpodacus purpureus group, in the Cardinalis virgmianus group, in
( 'urrirostra " americana" in the rostratus form of Passerculus, in the
western forms of Melospiza melodia, in Passerella " schistacea" and in
the Pyranga cestiva and P. ludoviciana groups ; it is also well illus-
trated by Certhia familiaris, Mniotilta varia, and almost constantly in
the J'ireonidce, as well as in numerous other families.

From the valley of the Columbia River a comparatively narrow belt
extends northward along the Pacific coast, where the annual rainfall is
nearly double that of any other portion of the continent ; and here the
birds (and mammals also), as a general rule, not only reassume the
brighter colors of the region east of the Great Plains, but in many
cases present a depth of color unequalled eastward in the same lati-
tudes, frequently taking on a peculiar deep plumbeous or dusky brown
in replacement of ashy or rufous, with a partial obsolescence of spots and
streaks, especially marked in several of the fringilline genera.

* See Bull. Mus. Comp. Zobl., Vol. II, p. 230.

MUSEUM OF COMPARATIVE ZOOLOGY. 117

In respect to the rank and relationships of a great number of forms
among North American birds, to which at first was accorded the rank
of species, the gradual passage of one form into another, through whole
groups of forms that, as it were, cluster about a common type, is an
interesting and suggestive fact. Every acquisition of new material from
the middle and western portions of the continent but the more fully
shows the complete and gradual coalescence of widely differing forms,
which reach their typical or maximum development at particular locali-
ties, characterized by special climatic conditions, but which intergrade
at intermediate points, where the conditions of environment are also of
an intermediate character. In illustration of this, the genera Pipilo,
Junco, Melospiza, Passerella, Carpodacus, Colaptes, and Picus may be
cited from among the numerous and more strongly marked examples
of longitudinal variation.* In respect to the Pipilos of the United
Siates, the P. erythrophthahnus of the East passes southward into a well-
marked form in Florida, differing from the northern race in having the
white on the wings and tail much more restricted, in its smaller size,
larger claws, and longer tail. To the westward it begins at the Mis-
souri River to pass into the P. arcticus, through the occasional accession
of white streaks on the scapulars and interscapulars, and its larger
claws; these characters — especially the development of white in the
dorsal plumage — reach their maximum on the dry plateau of the in-
terior ; but westward P. arcticus merges into another form, P. oregonas,
towards and on the Pacific coast, in which the white on the wings be-
comes again reduced, the white streaks on the back (though generally
still retained) become narrower and fewer, and at times are either
almost or entirely obsolete, and the claws become considerably smaller.
To the southward the two forms, in the interior, run into each other,
both culminating in Lower California in the P. megalonyx, in which
the claws have become enormously developed, and the white spotting
varies from obsoleteness to the large amount that typically characterizes
P. arcticus. Further southward, in Mexico, P. megalonyx is well-
known to grade through P. macronyx into P. maculatus, which are
more or less olivaceous. Junco presents three strongly marked forms
or " species," that in a similar manner inosculate ; J. hyemalis being the
eastern form, J. oregonus the western, and J. caniceps occupying an in-
termediate region at the southward, between the habitats of the others ;

* See also Bull. Mus. Comp. Zool., Vol. II, p. 237.

113 BULLETIN OF THE

but they more or less mix up during winter, and specimens are of fre-
quent occurrence that, from their not being referable to either form, have
been assigned to the series of " hybrids.'" J. caniceps is the most strong-
ly marked form, in its having the middle of the back reddish, forming
a restricted, well-defined patch. J. oregonus has the back also reddish,
this color occupying a larger area than in J. caniceps, more diffused,
and involving the secondaries ; but its extent and intensity varies greatly
in different individuals. The sides are also tinged with a pinkish rufous
tint, and the slate of the anterior half of the body is darker than in the
other. J. hyemalis has the rufous tint present only in young or au-
tumnal specimens, which sometimes strongly approach J. oregonus. J.
oregonus, on the one hand, inosculates with J. caniceps, and on the
other, with J. hyemalis; caniceps and oregonus both apparently merg-
ing into J. cinereus of Mexico, through the scarcely distinguishable
J. dorsalis. Melospiza melodia is represented in the interior by a race
(JLfallax) paler than the eastern, and on the Pacific coast by a darker
race, which again divides into a northern (M. insignis) and a southern
(JL Heermanni et Gouldii, etc.), all of which so intergrade as to be but
unsatisfactorily definable, though in their extreme stages they present
strong points of difference. Few congeneric species, it would seem,
need be more distinct than Colaptes auratus and G. mexicanus, the one
occupying the eastern and the other the western side of the continent.
Yet a mixed race has been long known to exist in the region where their
habitats adjoin, in which every possible combination of the characters
of the two birds is presented, and which shade off gradually on the one
side into C. auratus, and on the other into C. mexicanus ; these, as it
were, engrafted characters not entirely fading out in either direction for
a distance of several hundred miles; while to the southwestward is a
smaller synthetic race (C. chrgsoides) partaking mainly of the charac-
ters of O. auratus.

"When but comparatively few instances were known, in which speci-
mens combined in various degrees the characters of two quite distinct
species, their synthetic character was generally explained by the theory
of hybridity ; but the irrefragibility of the evidence now at hand in
proof of the gradual intergradation of such forms over large areas, — the
transition being so gradual as to occupy hundreds of miles in the pas-
sage,— and also coincident with a similarly gradual change in the con-
ditions of environment, together with the demonstrable evidence of the

MUSEUM OF COMPARATIVE ZOOLOGY. 119

power of climatic influence, seems to furnish a far more satisfactory
explanation of these perplexing phenomena. But an advocate of the
theory of hybridity might still assume that this gradual transition over a
wide area is no objection to the theory, since the gradual fading out of
the impression of contact in either direction from the line of junction of
the respective habitats of two forms is just the result that would be
anticipated from such a sexual intermingling of the forms in question.
But the real objection to the theory — granting the possibility- of hy-
bridization on such a gigantic scale, which seems really improbable —
is, that widely different forms occur also at different points in latitude,
between which each successive stage of gradual differentiation can be
readily traced, where hybridity can scarcely be supposed to account for
the gradual change. Furthermore, a differentiation is now known in
so many cases that it amounts to the demonstration of climatic varia-
tion as a general law, by means of which a species may be safely pre-
dicted to take on a given character under certain specific climatic
conditions. If the theory of hybridity be urged to account for the in-
tergradation of forms occurring at localities differently situated in respect
to latitude, as has been sometimes done, it evidently falls under the
weight it has to support; and yet there seems to be little better evi-
dence in its hehalf in cases where the intcrgrading forms happen to he
differently situated in respect to longitude.

In regard to how these well-marked geographical forms shall be rec-
ognized, there may he just grounds for a diversity of opinion. Evi-
dently in cases where they are slightly marked or somewhat inconstant,
no great harm would result if they were nominally ignored. Practi-
cally, most naturalists recognize as species such groups of individuals as
are not known to graduate by nearly imperceptible stages into any
other similar group ; and as varieties, such groups of individuals as
occur at certain localities, or over certain area-, which differ more or less
from other groups of individuals inhabiting other (general!)' contiguous)
localities, with which there is evidence that they do, more or less fully,
intergrade. Convenience seems to demand such a course, in order to
enable the naturalist to specify what particular phase or race of a spe-
cies inhabits a given section of country ; the first specific name used for
any part of the group being, of course, retained for the longest known
form, and the other races, when of such prominence as to rend i naming
them advisable, being designated by additional varietal names; as, tor

120 BULLETIN OF THE

example, A b, var. c. This method is, indeed, already in more or less
common use.

The division of the middle and western portions of North America
into faunal areas is still attended with many difficulties, partly from
the absence of data, and partly from the peculiarly varied character
of the surface. The presence or absence of forests, directly resulting
from the peculiarities of climate, seems to be among the most effective
influences in the modification of the range of species. If a nearly un-
broken forest extended from the Atlantic to the Pacific, with the nearly
uniform conditions of humidity that would naturally follow, undoubtedly
many species would range across the continent, or at least to the base
of the Rocky Mountains, that now extend westward from the Atlantic
coast only to the edge of the Great Plains ; the western slope of the
continent would differ less in its animal life from the eastern than it
now does, and the middle region would lack the widely different zoologi-
cal aspect it now presents from that of either the Atlantic or the Pacific
coast regions. With the present elevation of the interior, and the
resulting climatic conditions, nearly all the woodland species of the East
not only range westward to the treeless districts of the interior, but
extend up the rivers that descend from the central plateau as far as
these streams are skirted to any considerable degree with trees ; a few
not only reach the base of the Rocky Mountains, but pass around the
higher elevations of the chain in Colorado, by means of the northern
valleys, and occur on the Pacific coast, while others reach the same
coast by gaps in the mountains at the southward. On the other hand,
most of the field and prairie species, or those which are but slightly
dependent upon woodlands for shelter or sustenance, do not disappear at
the edge of the plains, as do the strictly woodland species, but range
not only over the plains of the middle region, but also over the plains to
the westward of the main chain of the Rocky Mountains, and thence
generally to the Pacific coast. In the wooded parts of the Rocky
Mountains a few species occur that are peculiar to that region, but the
greater part arc either but slightly modified forms of eastern species, or
forms that, while they differ widely from both the eastern and western,
still freely " hybridize " or intergrade with them. These strictly western
forms, unless of alpine or subalpine distribution, also generally occur
along the streams of the western edge of the Plains, as far as the streams
are bordered with trees.

MUSEUM OF COMPARATIVE ZOOLOGY. 121

The observations made the past summer (given below in detail)
establish the occurrence of a number of eastern species at points several
hundred miles to the westward of the westernmost point from which
they have been previously recorded ; and in like manner other western
species were found occurring at points considerably to the eastward of
points from which they were before known. Northern species were also
found at localities considerably south of their previously known range,
both Anthus ludovicianus and Leucosticte tephrocotis being found breed-
ing above timber-line in the mountains of Middle Colorado. A more
extensive overlapping of the habitats of eastern and western species is
thus established than there was previous evidence of, which may tend
to modify the currently received boundary between the Eastern and
Middle Provinces of the North American Region. This boundary has
generally been considered as running in the United States near the
100th meridian, or " at the edge of the sterile plains." But the dis-
tinctively " Plains species " are nearly all found now to range eastward
over the prairies, the others first appearing somewhat to the westward.
Tims of about twenty species that are distinctively characteristic of the
Plains, fully one third occur on the prairies of Illinois and Wisconsin,
another third are met with as far east as Missouri, and the others range
more or less regularly into Eastern Kansas. In other words, all the
species of the Plains occur in Kansas at points from two hundred to
three hundred miles to the eastward of the 100th meridian, and
most of the others extend to the woodland districts eastward of the
Mississippi. On the other hand, many eastern species follow up the
rivers to the most western limit of trees, sometimes to a distance of
three hundred miles west of their formerly supposed western limit,
where they mix with western species not commonly supposed to occur
much to the eastward of the eastern base of the Rocky Mountains.

The fauna? of the middle and western portions of the continent pre-
sent peculiarly broken and irregular areas, in consequence of the great
irregularity of the surface of the country. The more southern fauna),
while occupying the lower table lands, extend also up into the lower
mountain valleys, to a limit varying with latitude and the peculiar local
conditions of the valleys themselves. Above this basal zone occur sev-
eral other zones, which an; continuous for considerable distances along
the main chains, but also embrace distant insular patches in the more
isolated groups of mountains. The higher zones arc still less regular

122 BULLETIN OF THE

in their continuity and in their respective areas, the highest having an
arctic character and occupying only the partially snow-covered summits
that rise above the limit of tree-growth. But at present, the data at
hand are too few for a satisfactory attempt at an analysis of the char-
acters and limits of the several avian faunae of the Middle and Western
Provinces.

I. List of Birds observed at Leavenworth, Kansas, from May 2 to
Mmj 11, and at Topeka, Kansas, from May 11 to May 24,
I<s71 ; with Annotations.

The following list embraces one hundred and twenty-one species, of
which specimens of nearly all were actually collected. Though an
incomplete list of the birds of Eastern Kansas, and based on observations
made when many of the species were migrating, it is believed to con-
tain many facts of value, especially since no report has as yet been
made of the ornithology of this section of country.* Our collections at
Leavenworth were made principally in the heavy timber on the East
Leavenworth side of the Missouri River, opposite Fort Leavenworth.
A few specimens were collected on the west bank of the river, on tin;
military reservation between the fort and the city, where is also con-
siderable timber. Most of the water birds were obtained about a lagoon
on the Missouri side. In the forests the birds were excessively abun-
dant, both in species and individuals. Among them such southern forms
as Helminthophaga pinus, Oporornis furmosus, Wilsonia (= Myiodi-
octcs) mitrata, Thryothorus ludovicianiis, Icteria virens, Cardinalis vir-
ffinianus, and Lophophanes bicolor were conspicuously numerous, the
fauna being emphatically Carolinian.! Although the vegetation was
as far advanced the 1st of May as it usually is in Southern New
England the 1st of June, very few birds had commenced nesting, and
some of the later-arriving species had not yet appeared. By the 10th
of May nearly all the trees were in full leaf, and most of them were
leafing by the first of the month. The only nests found were tho.>e of

* Since the above was written there has appeared a "Catalogue of the Birds of
Kansas," bj '.'.< fessor F. H. Snow, of Lawrence, Kansas. The list contains the name ol
239 species, yet some of the most characteristic the western half of the State

are omitl <1. The author has attempted to indicate those "known to bree.l in the
State," but it h in this respect very imperfect, though still not without much value :i
st fauna! list. ! Sec Am. Nat, Vol. VI, LS72, p.

: See Hull. Mik V. >mp 7 I., W II, \\ M3, 1*71

MUSEUM OF COMPARATIVE ZOOLOGY. 123

Harporhynchus rufus, Pipilo erythrophthalmus, and Cardinality virginia-
nus, but fledged young of T/n-yotliorus ludovicianus were shot May 3d.

At Topeka, about the same number of species were observed as at
Leavenworth. Some of those obtained at Topeka, however, were not
seen at Leavenworth, and others that were common at Leavenworth
were not noticed at Topeka. Yet the general character of the fauna at
the two places is quite similar, as would naturally be expected; Topeka
being only about sixty miles from Leavenworth, in a southwesterly
direction. At Topeka our excursions were mainly confined to the
timbered bottom lands of the Kaw River. Though most of the larger
trees had been removed, the locality was still tolerably well wooded,
and in many places there was a dense undergrowth of hazels, sumachs,
and other shrubs. The adjoining prairies were visited a few times,
and one excursion was made to the Wakarusa, ten miles to the south-
ward. Here, however, only one species (Polioptila ccerulea) was taken
that was not also seen at Topeka.

At Topeka the birds were even more numerous than at Leavenworth.
In the course of half an hour, on the day of our arrival (May 12th), I
saw or heard thirty species of birds, by actual count, and in most cases
observed a number of individuals of each. This, however, seems to be
a feature more or less common to prairie regions, where the timber is
restricted to narrow belts along the streams. Especially does this seem
to be the case during the season of migration ; but it was also observed
later in the season at Fort Hays.

Although the forests were in full leaf on our arrival, we noticed that
several species of birds became common during the last days of our
stay there, that were not met with at first. Among them were such
species as Seiurus aurocapillus, which usually arrives in New England
at the time of the first leafing of the trees. Ripe wild strawberries
were abundant as early as May loth, and the weather was as hot, from
this date till we left Topeka, as it usually is in Southern New England
in July, the maximum temperature daily increasing from 84° to 94° F.
in the shade.

When no locality is mentioned in the remarks that follow, it is to be
understood that the species was observed at both Leavenworth and
Topeka in about the same numbers. A star is prefixed to the names
of those known to breed in Eastern Kansas, whether from person:) 1
observation or from their known breeding range including the localities
in que- tion.

124 BULLETIN OF THE

TURDIDiE.

1. *Turdus migratorius. Only two were seen. Said to be a scarce
resident.

2. *Turdus mustelinus. Exceedingly abundant. Quite unsuspicious
and apparently not yet nesting. Song less melodious and less varied than
in the Eastern States. Colors considerably brighter.

3. Turdus Swainsoni. Common at Topeka. All of the eight or ten
specimens taken were females, the males probably having already gone
north. These specimens were also all strongly suffused with rufous. One
was shot at Leavenworth, May 8th, and a few others seen.

4. Turdus Pallasi. A single female, with the plumage excessively
worn and faded, was taken at Topeka, May ISth.

5. *Harporhynchus rufus. Abundant. Nest and three eggs ob-
tained May 3d, at Leavenworth. Nest placed in bushes, several feet from
the ground. A nest nearly finished was also found May 19th at Topeka,
placed on the ground, under a small bush in an open field.

6. *Mimus carolinensis. Abundant.

SAXICOLIDJE.

7. * Sialia sialis. Common. Said to be resident.

SYLVIIDjE.

8. *Polioptila casrula. Three specimens were seen and two taken
May 22d near Topeka. These were the only ones observed.

SYLVICOLIDjE.

9. Parula americana. Common.

10. * Helminthophaga pinus. Not uncommon.

11. Helminthophaga celata. Common at Leavenworth.

12. Helminthophaga ruficapilla. Common at Leavenworth.

13. * Dendrceca eestiva. Moderately common. The streaks on the
breast, in the few specimens taken, were very broad and conspicuous, much
broader and the colors generally much brighter than they are often seen
in specimens from the Eastern States.

14. Dendrceca pennsylvanica. Not common. One specimen shot
at Leavenworth, and three or four others seen. Not observed at Topeka.

15. Dendroeca Blackburniee. One specimen shot at Leavenworth,
May 4th, — the only one seen.

16. * Dendrceca caerulea. One specimen taken at Leavenworth, and
a number of others seen. Apparently rather common in the forests of the
Missouri bottom.

17. Dendrceca coronata. One specimen seen May 3d at Leaven-
worth, the only one observed.

MUSEUM OF COMPARATIVE ZOOLOGY. 125

18. *Dendrceca discolor. Rather frequent.

19. Mniotilta varia. Not common. Two specimens obtained at Leav-
enworth, and two or three otbers seen.

20. *Seiurus aurocapillus. First observed May 15th; afterwards
common.

21. * Geothlypis trichas. Common at Topeka ; only a few were seen
at Leavenworth, where it arrived about May 8th.

22. Geothlypis Philadelphia. One specimen obtained May lGth at
Topeka, where others were seen later.

23. * Oporornis formosus. Common. Nest found nearly completed
May 16th.

24. * Wilsonia mitrata. Rather common at Leavenworth ; less so at
Topeka.

25. Setophaga ruticilla. Common, but only males were taken or
observed.

20. * Icteria virens. Rather common at Leavenworth. Abundant at
Topeka, where three or four males were often seen hovering in the air and
singing at the same time.

TANAGRIDiE.

27. *Pyranga rubra. Very abundant. The colors of those obtained
were unusually intense, as compared with northern specimens.

PARIDJE.

28. *Parus atricapillus. Abundant.

29. *Lophophanes bicolor. Abundant. One of the most numer-
ously represented and most noisy species met with at Leavenworth ; not so
abundant at Topeka.

SITTTD-2E.

30. Sitta carolinensis. Common at Leavenworth.

TROGLODYTID^.

31. * Troglodytes aedon. Common.

32. * Thryothorus ludovicianus. Common at Leavenworth. Not
seen at Topeka. They apparently breed very early, as we shot a young
one fully fledged May 3d.

HIRUNDINIDJE.

33. *Hirundo horreorum. Moderately common.

34. Hirundo bicolor. Common, especially at Leavenworth.

35. * Hirundo lunifrons. Common at Leavenworth ; less numerous at
Topeka. At the latter locality several pairs were seen along the bluffs of

126 BULLETIN OF THE

the Kaw River, in company with Cotyle riparia, entering the holes in the
bank in company with that species, and also sitting in the mouths of the
holes. One was shot as it left a hole, so that there is no reason for
doubting the observation. They had the same appearance of breeding
in the banks as Cotyle riparia themselves.

3G. * Cotyle riparia. Exceedingly abundant, especially at Topcka.
Hundreds of them were excavating their holes in the bluffs of the Kaw
River, May 15th to '20th, but had not yet commenced to lay. At least no
eggs were found in any of a considerable number of nests examined.

3 7. * Cotyle serripemiis. Common. They appear to breed either
singly, or a few pairs together, and not in large colonies like Cotyle riparia.
They were excavating their holes, but had not yet laid.

38. *Progne subis. Common at Topeka, and abundant at Leaven-
worth, breeding in boxes provided for their use.

VIREONIDJE.
30. * Vireo olivaceous. Common.

40. Vireo gilvus. Common.

41. Vireo flavifrons. A single specimen was shot at Topeka, and
several others were seen.

42. * Vireo noveboracensis. Common.

43. * Vireo Belli. Exceedingly abundant after May 15th at Topeka;
not seen earlier. Commenced pairing immediately after their arrival, and
were one of the most numerous and conspicuous species of the smaller
birds.

AMPELIDJE.

44. * Ampelis cedrornm. Several small roving flocks were seen at
Topeka, May 20th and later.

LANIIDJE.

45. * Collurio ludovicianus. Said to be moderately frequent, but
seen only at Leavenworth.*

* Messrs. Dresser and Sharpe, in a paper on " Lanius excuhitor and its Allies " (Proc.
Zool. Soc, 1870, p. 595), combine C. exciilitoroides and C- elegans with C. ludovicianus,
their conclusion being based upon an examination of specimens of each of these so-
called species. I am glad to find my own opinion on this point (first partially expressed
in Ainer. Nat., 1869, p. 579, and more fully reiterated in Bull. Mus. Comp. Zool., Vol.
II, p. 270, April, 1871) thus confirmed.

The original specimen of the C- elegans (Lanius elegans of Swainson), now in the
British Museum, these gentlemen refer to the L. laktora of Northeastern Africa and Asia,
presuming the specimen to have come from some other locality than North America, or
that the /.. lahtora may occur in North America as n straggler from Northern Siberia.
In this connection I may add that I have been long impressed with the close resemblance

MUSEUM OF COMPARATIVE ZOOLOGY. 127

ALAUDIDjE.

46. *Eremophila alpestris. Common on the prairies.

FRINGILID.EJ.

47. * Chrysomitris tristis. Common.

48. * Coturniculus passerinus. Common.

49. * Chondestes grammaca. Moderately frequent.

50. Zonotrichia leucophrys. One specimen seen May 8th at Leaven-
worth.

51. Zonotrichia querula. Exceedingly abundant at Leavenworth.
Found almost exclusively in the forests, and generally in company with
Z. alhicollis, which it resembles in habits and somewhat in song.

52. Zonotrichia albicollis. Common. Fully as numerous May 11th
at Leavenworth as at any time previously. Less numerous at Topeka.

53. * Spizella pusilla. Common.

54. * Spizella socialis. Obtained one or two at Topeka still in im-
mature plumage, — a condition in which I have never seen this species in
the Northern States at this season of the year, although I have handled
hundreds of specimens taken in spring at northern localities.

55. * Spizella pallida. Common at Topeka. Greatly resembles the
specimens of Spizella socialis in immature plumage, taken at the same
locality, with which they were associated.

56. Melospiza melodia. Not common. Only one specimen was ob-
served.

5 7. Melospiza palustris. Not common.

58. Melospiza Lincolnii. Common.

59. *Euspiza americana. The males were excessively numerous, but
only a few females were seen. Not yet breeding.

60. * Goniaphea ludoviciana. Only a few observed, which were
nearly all males.

61. * Cyanospiza cyanea. Common. Not seen till May 8th, but was
afterwards abundant. Both sexes unusually brightly colored. One of the
females taken at Topeka had a strong shade of blue over the whole throat
and breast, and other females were similarly more or less tinged with blue.

62. * Cardinalis virginianus. Exceedingly abundant. Young a week
old were found May 10th. At the same date other nests were found con-
taining three eggs each, as well as several unfinished nests. All of the

western specimens of Collurio (or Lanius) ludovicianus bear to certain forms of Lanius
from Northern Africa. On recently comparing two specimens of shrikes, one from Cali-
fornia and the other from Algeria, contained in the Lafresnaye collection in the Museum
of the Boston Society of Natural History, I was unable to distinguish the Algerian one
from the California!!.

128 BULLETIN OF THE

half-dozen specimens of this species taken in Eastern Kansas differ from
any I have seen from the Atlantic States in having a much larger and
more swollen beak. It is a little smaller than that of the Cape St. Lucas
form (C. "igneus "), in this respect being about half-way between the lat-
ter and the race of the Atlantic States. The color of the males is not quite
so deep as in specimens from Florida.

03. * Pipilo erythrophthalmus. Abundant. Nests with eggs were
found about May Gtk and later. The song of this species was generally
very different from that of the eastern bird, though occasionally it was
indistinguishable from that of eastern individuals. Iiather mure white
on the wings than in eastern specimens.

ICTERIDJE.

C4. *Molothrus pecoris. Very abundant. Generally seen lurking
among the bushes in search of bird's nests in which to deposit its eggs.
Plumage appreciably darker than at the north.

65. *Agelaeus phceniceus. Common.

CG. * Xanthocephalus icterocephalus. Several times seen around
the prairie marshes at Topeka, where it was said to be common.

67. *Sturnella ludoviciana. Common. Several very pale-colored
specimens were taken. It has here the song and generally the plumage of
the so-called S. neglecta.*

08. * Icterus Baltimore. Common ; chiefly frequenting the forests.
The notes of the Baltimores here are very peculiar, many of them being
entirely unlike any of those of their eastern representative.

69. * Icterus spurius. Abundant.

70. * Quiscalus purpureus. Abundant.

CORVIDJE.

71. *Corvus americanus. Common. Young full grown taken at
Topeka, May 23 d.

* Dr. Otto Finsch, in the Proceedings of the Zoological Society (1870, p. 573), in
speaking of the species of Sturnella, says: " The separation of the Stui-nellce into five
localized species, as Dr. Sclater endeavored to set forth (Ibis, 1861, p. 179), in which he
was followed by Mr. Cassin (Proc. Ac. Phil., I860, pp. 23, 24), seems to me to be in-
admissible; nobody can distinguish the so-called species from the short diagnoses given

as above cited Dr. Cabanis (J. f. Orn., 1850, p. 14, et 1861, p. 10), after having

examined specimens from North America, Cuba, Costa Rica, Venezuela, and Guiana,
comes to the conclusion that there is only one species; and I believe this opinion is quite
right." These remarks of Dr. Finsch antedate by a few months my revision of this
group published in April, 1871 (Bull. Mus. Comp. Zool., Vol. II, pp. 288-291), in which
I came to the same conclusion. The part of the Proceedings of the Zoological Society
containing Dr. Finsch's article had not then reached this country, and I am gratified to
find that my own opinions on this point coincide with those of such high ornithological
authorities.

MUSEUM OF COMPARATIVE ZOOLOGY. 129

72. * Corvus corax. Frequently seen ; apparently common.

73. *Cyanura cristata. Abundant. One of the most numerous species
met with. It has here a variety of notes I never noticed in the varied vo-
cabulary of the representatives of this species elsewhere.

TYRANNTD^.

74. * Tyrannus carolinensis. Abundant.

75. * Myiarchus crinitus. Abundant.

7G. *Sayornis fuscus. Common at Leavenworth. Darker colored
than at the north.

7 7. * Contopus virens. Common.

78. * Empidonax Traillii. Not common.

79. Empidonax minimus. Not common.

ALCEDINID-ai.

80. *Ceryle alcyon. Common.

CAPRIMULGIDJE.

81. *Chordeiles popetue. Common.

82. * Antrostomus vociferus. A few heard at Leavenworth.

83. * Antrostomus Nuttallii. Common at Topeka.

CYPSELID^.

84. * Chaetura pelasgia. Abundant. Breeds chiefly in hollow trees.

PICIDJE.

85. *Picus pubescens. Common. Darker colored than further north,
in this respect resembling Florida specimens, and approaching the so-called
Picas " Gairdneri" of the Rocky Mountains.

86. Picus villosus. Probably more or less common, but only one was
observed.

87. * Centurus carolinus. Common. Those taken were very intensely
colored. Some of the males had the whole throat bright red.

88. *Melanerpes erythrocephalus. Abundant.

80. * Colaptes auratus. Abundant. Several specimens were taken,
with the black maxillary patch more or less tinged with red, through the
mixture of red feathers with the black ones, thus already showing a ten-
dency to the coloration of C. mexicanus, six hundred miles east of the habitat
of that species.*

* Since the above was written, a specimen with red feathers in the black maxillary
patch has been found in the Florida collection. I have also learned of the capture of
a well-marked example of the so-called C " hybridus" at Topeka, February 13, 1872,
by Mr. 0. S. George. Mr. Edwin A. Papenoe informs me that in this specimen the
qirlls are "orange red," and that the feathers of the maxillary patch are tipped with
" <1 trk blood red," with the other characters nearly as in C. auratus.
VOL. II. 9

130 BULLETIN OF THE

ARID^J.

The Conurus carolinensis, Dr. C. A. Logan informed me, was formerly
common here, but hud not been recently observed.

FALCONID-E.

90. *Falco spaverius. Abundant.

91. *Buteo borealis. Common.

92. *Buteo lineatus. Common.

93. * Circus cyaneus, var. hudsonius. Common.

94. *Nauclerus furcatus. Several pairs seen at Topeka, where it
arrived about May 15th.

CATHARTICS.

95. * Cathartes aura. Common.

COLUMBIDJE.

9G. * Zenaedura carolinensis. Abundant.

TETRAONIDjE.

97. *Cupidonia cupido. Abundant on the prairies.

PERDICIDJE.

98. * Ortyx virginianus. Abundant.

CHARADRIIDJE.

99. * -aDgialitis vociferus. Common.

SCOLOPACIDJE.

100. Actodromas maculata. Common about the lagoons at Leaven-
worth.

101. Ereunetes pusillus. Common on the sand-bars in the Kaw River,
at Topeka.

102. Gambetta flavipes. Numerous about the lagoons. Ova in the
females quite large. Probably breeds. Males considerably darker than
the females, with the transverse bars of black broader and much more con-
spicuous than in the females.

103. Rhyacophilus solitarius. Rather common. Probably breeds.

104. *Tringoides macularius. Abundant.

105. *Actiturus Bartramius. Common on the prairies.

106. Limosa fedoa. A few seen.

107. *Numenius longirostris. Common on the prairies.

GRUIDJE.

108. Grns americanus. Two individuals were seen on a sand-bar in
the Kaw River at Topeka.

MUSEUM OF COMPARATIVE ZOOLOGY. 131

ARDEIDJE.

109. *Ardea herodias. One individual seen.

110. *Butorides virescens. Common.

111. *Botaurus lentiginosus. Common.

RALLIDiE.

112. *Fulica americana. Common.

113. * Rallus virginianus. Apparently common.

114. *Porzana Carolina. Probably common, though but fewwerc seen.

A3NTATIDJE.

115. * Aix sponsa. Common.

116. * Querquedula discors. Abundant.

117. * Anas boschas. Abundant.

1 18. * Fulix marila. A single female was killed at Topeka, — the only
representative of the species seen.

Two other undetermined species of ducks were seen, but not taken.

PODICIPIDJE.

119. Podilymbus podiceps. Common.

LARID-ffiJ.

120. *Hydrochelidon fissipes. Several seen.

PELECANIDJE.

121. Pelecanus erythrorhynchus. Said to be common. Saw a speci-
men which was killed about May 10th, which had the crest on the upper
mandible remarkably high and thick.

II. — List of Birds observed in the Vicinity of Fort Hays, Kansas, from
May 2G to July 3, 1871 ; with Annotations.

The subjoined list of sixty-one species of birds, observed in June
at Fort Hays and viciuity, indicates the general character of the sum-
mer avian fauna of the eastern border of the Great Plains. The next
following list of twenty-five species, observed during three weeks in
midwinter, somewhat to the westward of Fort Hays, embraces all the
more characteristic species of winter. Many not mentioned in these
lists occur in fall and spring, chiefly swimming and wading birds.

Fort Hays is situated on Big Creek, three hundred miles west of the
Missouri River, about ten south of the Saline River, and about the
same, distance north of the Smoky River. The timber here is not only

132 BULLETIN OF THE

confined to the immediate vicinity of the streams, often to their beds,
but generally occurs in thin, irregular belts or scattered clumps, and
ceases entirely a few miles to the westward. The Smoky is already
quite destitute of trees as far west as Fort Hays, and they soon disap-
pear from the Saline. The observations on which the following notes
are based are the result of about thirty-five days spent consecutively in
the field, during which time an area of country of from fifteen to thirty
miles' radius was quite thoroughly explored. The belt of timber along
Big Creek, preserved on the Military Reservation at Fort Hays,
afforded by far the richest field, though some species were obtained on
the Saline, and during a single day's hunt on Big Timber Creek, that
were not met with on Big Creek. A longer time spent on Big Timber
would dojibtless have added several other woodland species to the list
here given. In further description of the locality, it may be added that
the trees consist mainly of the white and red elms, the ash-leaved
maple, cottonwoods, black-walnut, and ash. Most of these trees assume
a spreading form, and grow to a large size. There is little under-
growth, except where the first growth has been removed, as it has been
to a large extent on most of the streams within fifteen to twenty miles
of the post. The undergrowth consists mainly of sumach, dwarf-plum,
and Amorpha fruticosa. In proportion to the amount of timber, the
tree-nesting species are very abundant, and their nests are easily found,
frequently half a dozen pairs of nearly as many species breeding in a
single tree.

The " Plains " are here, as usual, somewhat rolling broad level
plateaus, being separated by low ridges, or broken by sharp ravines and
moist hollows. They are covered with short grass, usually but two or
three inches high, except in the hollows and near the streams, where it
often grows to the height of one or two feet. On the plateaus and
ridges, in consequence of the excessive heat and scanty fall of rain, the
grass becomes parched and dry during the latter half of June, and for
the rest of the year the landscape wears an arid and forbidding aspect,
relieved only by the deep green foliage of the trees along the streams.
During May and much of June, however, the fresh young grass is
thickly dotted with a variety of showy flowers, which vary the land-
scape with their respective tints. They are mainly social plants, and,
growing thickly, their bright colors are conspicuous, giving their several
hues to lanrc areas. Most characteristic amon"r them are Malvastrum

MUSEUM OF COMPARATIVE ZOOLOGY. 133

coccineum, and one or two other malvaceous species, Verbena aubletia,
a Lippia, a Scutellaria, and an aster-like composite plant, — all low forms
and very prolific of large showy flowers. Among the coarser herbs are
Amorpha canescens, Echinacea, angustifolia, Delphinium azureum, a
Lepachys, a Dalea, two species of Linum, Onosmodium carolinianum,
and Verbena hastata, all common in their respective localities, but gen-
erally of dwarfed stature as compared with their size on the moister
prairies to the eastward. The sensitive brier (Schrankia uncinata) was
also abundant, and Rosa lucida was agreeably frequent along the
streams. Two species of Melocactus and an Opuntia attest by their
abundance the dryness of the climate.

The birds found here fall naturally into two groups, in accordance
with the situations they most affect, — those of the timber and those of
the Plains proper. The former class is much the more numerous in
species, only about six being confined strictly to the Plains ; these
latter are, however, among the most characteristic, being by far the
most numerously represented, and almost the only kinds that inhabit
the treeless belt which extends thence westward to the Rocky Moun-
tains. They are the horned lark {Eremophila alpestris), the chest-
nut-colored bunting {Plectrophanes ornatus), the lark finch (Chon-
deslcs grammaca), the lark bunting (Calamospiza bicolor), the yellow-
winged sparrow (Coturnicidus passerinus), and the meadow lark (Stur-
nella ludoviciana). The Carolina dove (Zencedura carolinensis) and the
night-hawk (Chordeiles popetue) are most numerous about the timber,
but are also everywhere common on the open plains, where the dove
nests on the ground as readily as it does in trees at the eastward. The
killdeer and mountain plovers {JEgialitis vociferus and JE. montanus),
and Bartram's tattler or field plover (Actiturus Bartramius) frequent
the plains, chiefly near moist hollows, as well as the neighborhood of
streams. About one fifth of the species were strictly western, not reg-
ularly occurring east of the Missouri River. Several others, however,
as Chordeiles popetue, Sturnella ludoviciana, Peuccea cestivalis, Troglo-
dytes a'edon, etc., have received distinctive names, owing to the faded
appearance they here exhibit, and others might be thus separated with
equal propriety. The bleaching of the plumage is evidently the result
of the excessive dryness of the climate, and the lack of shelter from the
intense rays of the sun, and in some degree, perhaps, of the wearing
off of the edges of the feathers by the almost incessant heavy winds.

134 BULLETIN OF THE

During our five weeks stay at Fort Hays, the maximum daily tempera-
ture in the shade usually ranged from 90° to 108° F. This tempera-
ture is frequently accompanied by parching winds, especially later in
the season. The most striking feature of the avian fauna here is the
great abundance of more or less strictly woodland species, considering
the scantiness of the forest vegetation.

TURDIDiE.

1. Harporhynchus rufus. Common in the narrow timber belts which
border the streams.

The habits of this species, in respect to the location of its nest, indicate
how greatly it is governed by circumstances. In dry, sandy localities, it is
well known to commonly nest on the ground, and to place its nest in low
bushes, where the soil is damp and clayey. Along Big Creek, near Fort
Hays, we found it nesting in low bushes, and also in trees sixteen to
twenty feet from the gi'ound. Big Creek is subject in summer to sudden
freshets, the stream, flowing between abrupt banks, sometimes rising ten or
twelve feet in a single night, half submerging the trees that grow along its
narrow bed. It was under the latter circumstances that the nests of this
species were found placed twenty feet above the ground, while but a few
yards distant other nests.were found in low bushes, the bushes, however,
growing on the bluffs, several feet above high-water mark. Other species
that generally nest near the ground were also found to place their nests at
a similar elevation, when breeding in the trees that grew along the bed of
Big Creek. The several species seemed to be well aware of the peculiarities
of the stream, and hence placed their nests above the high-water line.

2. Mimus polyglottus. Common in the timber along Big Creek. Nest
placed in trees fifteen feet from the ground. Fresh eggs obtained June 6th.

3. Mimus carolinensis. One or two seen on Big Timber Creek. Not

common.

SAXICOLIDJE.

4. Sialia sialis. Not uncommon along the timbered streams.

PARIDJE.

5. Parus atricapillus. Frequent in the timber along the streams.

TROGLODYTIDJE.

6. Troglodytes aedon. Abundant, nesting in the hollows of trees.
Seven fresh eggs taken from one nest June 7th.

In respect to plumage, this species has here all the essential characters
of the so-called T. " Parhnanni" the colors being appreciably paler than in
specimens from the Atlantic States.

MUSEUM OF COMPARATIVE ZOOLOGY. 135

SYLVTCOLID^.

7. Icteria virens. Common on Big Timber Creek.

Probably Dendrceca cesliva and D. discolor occur sparsely along the Big
Timber, but none were observed during a day's hunt along that stream.

HIRUNDINIDjE.

8. Hiruudo lunifrons. Common at localities. Large colonies breed
on the cliffs bordering the Saline River.

9. Cotyle serripennis. Not uncommon along the streams, in the banks
of which it nests. Nests examined June 7th were not yet completed.

10. Progne subis. A few pairs were seen in the vicinity of Fort Hays,
where they were breeding in boxes erected for their accommodation.

VIREONIDJE.

11. Vireo gilvus. Rather common in the timber on the " Reservation "
at Fort Hays, and along the Saline and Big Timber.

12. Vireo Belli. Common along Big Timber, and doubtless more or less
frequent along the better timbered portions of the other streams.

ALAUDID-S1.

13. Eremophila alpestris. Abundant ; as frequent on the high divides
as elsewhere. Very unsuspicious; in this regard its habits contrasting
strongly with those of most of the other prairie species, especially Calamo-
spiza bicolor and Plectroplmnes ornatus. It was decidedly the most numer-
ous species in the vicinity of Fort Hays. Resident, breeding very early,
and apparently twice in the season. The first brood was fully fledged in
May, and before the end of June the young birds were* already gathering
into flocks. June 11th, we found young in the nest half grown, and the fol-
lowing day young that, although they had left the nest, were still unable
to fly. No nests were found containing eggs, the species being a close
sitter, and the nest very difficult to find.

The plumage of this species was very much bleached, a large proportion
of the specimens observed having the throat either distinctly white, as also
the superciliary stripes, or with only the faintest trace of yellow, and the
other tints were correspondingly pale.

FRINGILLIDjE.

14. Chrysomitris sp. ? A Chrysomilris was frequently heard, but all our
efforts to procure a specimen were fruitless. It had the restless habits and
the notes of C. pinus, but this species is not known to frequent so southern
a locality in the breeding season. It is hence more likely to have been
C. psaltria.

15. Plectrophanes ornatus. Common out on the plains almost every-

136 BULLETIN OF THE

where, it being one of the most interesting and characteristic species of the
Plains. It has a short, shrill, but very sweet song, which is often uttered
while on the wing. It is very wary for so small a bird, and has the habit
of circling round the observer when disturbed for several minutes together,
approaching tantalizingly near, with feints of a nearer approach, but gen-
erally keeping well out of range. The nest is a very neat, though slight
structure, placed of course on the ground, and is composed of dry fine
jrass and rootlets. The eggs are generally five, blotched and streaked
with rusty on a white ground. Full sets of freshly laid eggs were first
found about June 3d.

The plumage varies greatly in color in different individuals of even the
same sex, the variation being generally in respect to the purity and inten-
sity of the colors. The most highly colored males have the breast and
middle of the abdomen more or less strongly tinged with very bright ferrugi-
neous ; others have these parts pure black ; while in others still the black
is obscured by the feathers having brownish-white margins. The lesser
coverts vary from gray to black. The red tinge on the abdomen seems
merely indicative of a high state of plumage ; those thus marked also having
the lesser coverts black ; but they are also black in some specimens that
are not tinged with red. Plectrophanes melanomas Baird, is merely the
ferrugineous phase of this species, and not even a local race. The highest
colored female (the sex determined by dissection) was nearly as brightly
colored as the paler colored males, having the chestnut collar, and the
black on the breast nearly as distinct as some of the males. It was also
nearly as large, and, until dissected, was supposed to be an immature male.
Thirty specimens of the bird were obtained, and three full sets of eggs.

1G. Coturniculus passerinus. Abundant everywhere on the Plains.
Several nests, with full sets of (usually five) fresh eggs were found between
June 3d and 10th. In notes and habits it does not differ from the eastern
birds, but is paler colored.

On comparing Florida specimens (of which I have thirty before me,
from Miami, Florida, collected by Messrs. Maynard and Henshaw) with
northern ones, the former are found to be far more brightly colored than
the latter. Between northern and southern specimens of the same species
greater differences in color are rarely observable than in this, the differ-
ences being far greater than occur between many con?pecific geographical
races to which has been awarded specific rank. The difference consists
in the much brighter and blacker tints of the southern form. Massachu-
setts specimens, though lighter than Florida ones, are still much darker
than those from the Plains.

17. Chondestes grammaca. Common. Most numerous in the moist
ravines and near the streams. Forms a very slight nest on the ground,

MUSEUM OF COMPARATIVE ZOOLOGY. 137

about June 1st. The first full set of eggs was found June 3 J, and in one
instance half-grown young were found June 6th. Generally, however,
they appeared to commence laying about June 5th. Quite unsuspicious,
and lias the most elaborate song of any bird on the Plains.

] 8. Peucaea aestivalis, var. Cassinii. Rather common along the
streams, where its low but peculiarly sweet song is heard at morning and
evening, beginning with the first approach of dawn, and continuing at
evening considerably after nightfall. It is very retiring, and it was
only after several attempts that I discovered the author of the sweet notes
that at these still hours added greatly to the pleasures of camping on the
plains. The plumage is very much paler than that of Florida specimens,
agreeing with that of the so-called P. "Cassinii."

19. Calamospiza bicolor. Common here and there on the plains, liv-
ing apparently in scattered colonies. Females were obtained from June 5th
to 10th, that had evidently commenced incubation, but our long searches
for the nest of this species proved always fruitless. The birds are very
wary and difficult to shoot. Like most birds of the Plains, they are very
tenacious of life, and when shot through vital parts, will generally fly sev-
eral hundred yards before falling, finally dropping dead. It is a bird of
powerful flight, delighting in the strongest gales, which force most other
species to lie sheltered in the grass. It has habits that strongly recall the
yellow-breasted chat, singing generally on the wing, hovering in the same
manner as that bird, while its notes are so similar to those of the chat as to
be scarcely distinguishable from them. Hence while collecting, we natu-
rally applied to it the cognomen of the " Black Chat." The plumage of
the males varies considerably in color, some being entirely black, except
the white wing-patches, while others have the plumage more or less skirted
with brownish-white, and in others there is an intermixture of feathers
wholly brownish. After the moulting season the males assume the plumage
of the female, the change in color being similar to that of the males of
Dolichonyx oryzivora.

20. Euspiza americana. Abundant on Big Timber Creek, and some
were seen along the Saline.

21. Goniaphea melanocephala. Several pairs seen along Big Creek
near Fort Hays. A nest with half-grown young was obtained June ljth.
Another nest built by the same pair was found with eggs about June 27th.
The song of this species so much resembles that of G. ludoviciana that at
first we mistook the species for that bird, and were only undeceived by
shooting specimens.

No representatives of the genera Cyanospiza, Spizetta, or Melospiza were
observed during our five weeks' stay at this locality.

138 BULLETIN OF THE

ICTERID^I.

22. Molothrus pecoris. Common in the timber, and frequent on the
plains ten to fifteen miles from the nearest trees.

23. Xanthocephalus icteroceplialus. A small flock seen at intervals
about the corral at Fort Hays during our whole stay there. They probably
bred in the vicinity.

24. Quiscalus purpureus. Abundant along Big Creek at Fort Hays.
Nests with newly hatched young were found June 1st, and others with
fresh eggs as late as June 12th. A nest was found in an old woodpecker's
hole, the top of which had been broken off", June 8th, containing two eggs,
and two young just hatched. A few twigs and rootlets had been laid on
the rotten wood to serve for a nest. Mr. William Brewster informs me
he has known this species to breed in a woodpecker's hole in Maine, — a
rather strange departure from its usual habits, considering its long tail,
which would seem to be an impediment to such a mode of nesting.

25. Icterus Baltimore. Common in the timber. All the specimens
obtained on Big Creek had much more white on the edges of the quills
than eastern birds, the middle coverts in the males being entirely white or
only faintly stained with yellow, instead of deep yellow or orange as in the
eastern birds. The specimens of this bird collected at Topeka and Leaven-
worth are in this respect about half-way between the Fort Hays specimens
and those from the Eastern States. All the Kansas specimens are smaller
than average New England ones, and have the bill relatively longer,
slenderer, and more decurved. The females were also uniformly without
black on the throat and head.

26. Icterus spurius. Abundant in the timber on the Reservation at
Fort Hays. Full sets of fresh eggs were taken every day from June 6th

to 10th.

27. Sturnella ludoviciana, var. neglecta. Abundant. The eight
specimens taken were all very pale, or of the S. neglecta type. Song
shorter, the notes more guttural and less ringing than those of this bird
are on the prairies of Iowa, Northern Missouri, and Eastern Kansas, over
which regions the neglecta type of plumage also prevails. A single nest
found May 30th. It was open at the top, and rather slovenly made.

CORVIDiE.

28. Corvus corax. Only a few pairs seen, though reputed to be
common.

29. Cyanura cristata. Abundant in the timber.

TYRANNID^.

30. Tyrannus carolinensis. Abundant in the vicinity of the timbered
streams.

MUSEUM OF COMPARATIVE ZOOLOGY. 139

31. Tyrannus verticalis. Very numerous along the timbered streams.
It has much the same habits as the T. carolinensis, nesting in the same
manner. The eggs are so much like those of that bird, that they are
sometimes actually indistinguishable from them. Nests with fresh eggs
obtained June 4th to 10th.

32. Myiarchus crinitus. One specimen obtained, which was the only
one seen.

No species of Empidonax, Sayomis, or Contopus was observed.

ALCEDINID^J.

33. Ceryle alcyon. Common.

CITCULIDJE.

34. Coccygus americanus. Common.

PICIDJE.

35. Picus pubescens. A single individual seen.

36. Melanerpss erythrocephalus. Abundant wherever there was
timber, and no less inquisitive and irrepressible than at the East.

37. Colaptes auratus. Rather rare, and very wary. One was several
times seen at a distance that seemed nearly red enough to be C. mexicanus.
Those taken had red in the cheek-patches, as at Leavenworth.

CAPRIMULGID^J.

38. Chordeiles popetue. Abundant. Most of those taken wera very
light colored, corresponding with the so-called C. " Henryi," but some
were nearly as dark as the average eastern bird.*

STRIGID^.

39. Bubo virginianus. Not seen alive, but a dried carcass was found
near Fort Hays.

40. Athene hypogaea. A large colony observed near the post, and sev-
eral small colonies elsewhere, living in the burrows of the prairie-dogs
(Cynomys ludovicianus).

Different specimens vary greatly in color and in the amount of feathering
on the feet. Some have the tarsi densely feathered, while in other speci-
mens the tarsi are nearly bare, a large series presenting every degree of
variation between these extremes. The A. hypoga:a, formerly supposed to
be confined to the region east of the Rocky Mountains, as distinguished
from the A. " cunicidaria " of the western half of the continent, seems to
have been based on specimens with the tarsi quite fully clothed, and hence

* In the synonymy of C- popetue given in Bull. Mus. Comp. Zool., Vol. II, pnge 300,
">ot-note, C. tcxcnsis was inadvertently included.

140 BULLETIN OF THE

mainly on individual variation of this character. The Rocky Mountain
form, to which the name of cunicularia lias generally been restricted, is a
little larger than the birds from the Plains, their elevated habitat cor-
responding to a more northern locality. Specimens were collected the past
summer at intervals from Fort Hays to the Salt Lake Basin. After a com-
parison of these with authentic specimens of both A. " cunicularia " and A.
hypogoza of authors, I find no difference that is constant, except the rather
larger size of the Rocky Mountain form, — a diiference that would a priori
be expected.

FALCONIDJE.

41. Falco peregrinus. A pair of these birds were found breeding on
a high cliff near the Saline River, May 29th, the young being then half
grown. The nest was placed on an inaccessible shelf of the cliff, and was
composed of sticks. The only other instance that has come to my knowl-
edge in which this bird has used any other nest than the bare ground is
that mentioned recently by the Rev. William Jarvis in the " American Nat-
uralist,"* where he speaks of a nest found by him in the White Mountains,
which was " made of a few dry sticks placed round a hollow on a shelf of
the cliff."

42. Falco sparverius. A few pairs observed nesting in hollow trees.

43. Buteo borealis. A few pairs seen, and a nest found June 1st, con-
taining three fresh eggs.

44. Circus cyaneus, var. hudsonius. Rather common.

CATHARTID^h

45. Cathartes aura. Moderately common. Usually seen in small
parties of from three or four to a dozen, about the carcasses of recently
killed buffaloes. A considerable number were apparently breeding in the
vicinity of some high cliffs on the Saline, but a careful search for their
nests was unsuccessful.

COLUMBIDJE.

46. Zenaeduia carolinensis. Common everywhere, but most numer-
ous in the vicinity of timber. Very commonly met with in pairs, many
miles from the nearest timber. Many nests were found at Fort Hays, in
the timber along Big Creek. Most of them were built in the usual way,
forming such slight structures that the eggs could be readily seen through
them from the ground. Several pairs, however, were found occupying de-
serted nests of the purple grackle, which they had slightly repaired. In one
case a nest with two eggs was found on /In' ground, only a few yards from
shrubs. From the frequency with which I had seen pairs of these birds far
out on the plains in the nesting season, T was led to anticipate this method

* Am. Nut., Vol. V, [>. CG2.

MUSEUM OF COMPARATIVE ZOOLOGY. 141

of breeding. I afterwards learned that further west, where the prairies were
entirely destitute of timber, and where this bird was very common, they
always nested on the ground, as from necessity of course they must. The
fact, however, is interesting as showing how readily the bird greatly modi-
fies its breeding habits to suit its surroundings, while other tree-nesting
species disappear entirely in regions where there are no trees. The
present species, however, seems everywhere but slightly dependent upon
trees, as it seeks its food in fields, and not in forests.*

MELEAGRIDJE.

47. Meleagris gallopavo. Common along the timbered portions of the
streams, which here form its western limit.

TETRAONIDjE.

48. Cupidonia cupido. Rare. It is every year, however, advancing
westward. Was first seen in the vicinity of Fort Hays about two years
since, and is apparently fast becoming common.

49. Pedicecetes phasianellus, var. columbianus. Common along
the streams. It is here called the " grouse," in distinction from the prairie-
hen ; but further west, beyond the range of the true prairie-hen, it is almost
universally called " prairie-hen " or " prairie-chicken."

PERDICIDJE.

50. Ortyx virginianus. Occasional, but every year is becoming more
common. Like the prairie-hen, it is quite rapidly working westward, fol-
lowing the settlers.

CHARADRIID^I.

51. .ffigialitis vociferus. Common everywhere. To the collector an
unmitigated nuisance, from their incessant screaming about his head wher-
ever he goes.

52. iEgialitis montanus. Moderately common. Unlike the preceding
species, they are quite unsuspicious and retiring, and nearly always silent.

SCOLOPACIDJE.

53 Tringoides macularius. Common.

54 Actiturus Bartramius. Not common.

55 Numenius longirostris. A few pairs were observed near Fort
Hays, where they were breeding.

56. Numenius hudsonicus. A single specimen was seen and shot
June 15th.

* Since the above was written I have been informed by Professor 0. C. Marsh that he
has often found the eggs and young of this species on the ground in Western Kansas
and in Colorado. He says (in a letter): " Once I flushed a female who was covering a
couple of very young birds on the ground, not in a nest, but in a small depression on the
g^ouwH."

142 BULLETIN OF THE

ARDEID-ffi.

57. Butorides virescens. Occasional.

58. Nyctiardea grisea, var. neevia. A single specimen was seen flying
along Big Creek.

RALLIDjE.

59. Fulica americana. A single specimen was shot June 8th. Said,
however, to be common.

ANATID.2E.

60. Aix sponsa. Not common.

61. Querquedula discors. More or less frequent throughout the
summer.

III. List of Birds observed in Northwestern Kansas, December 25,
1871, to January 12, 1872 ; with Annotations.

The following list is based on observations covering a period of
nearly three weeks, made during a wagon journey of over two hundred
and fifty miles. The area traversed was nearly fifty miles square,
extending westward from Park's Fort Station, on the Kansas Pacific
Railway, to Grinnell, and from the Smoky River on the south to the
head-waters of the Solomon on the north. The opportunity was hence
unusually favorable for observing the birds that inhabit the Plains in
winter.

The locality does not differ essentially from the country about Fort
Hays, except in the greater scarcity of timber, which is limited to a
few scanty clumps of bushes and scattered trees on the Saline and
Solomon Forks, opposite Coyote Station. The small number of species
observed under such favorable circumstances indicates the poverty of
the winter avian fauna of the Plains. The only species really numerous
were Eremophila alpestris, which was met with everywhere, and roving
locks of two species of Plectrophanes (P. nivalis and P. Maccownii).
As our halts near the timber were necessarily short, a longer stay at
hese points might have added a few other species to the list of those
observed.

1. Parus atricapillus. A few were seen in the shrubs along the
streams.

MUSEUM OF COMPARATIVE ZOOLOGY. 143

ALAUDIDJE.
2. Eremophila alpestris. Abundant everywhere, but especially nu-
merous along the railroad and near the settlements. Though so numerous,
they appeared to suffer considerably from the unusual severity of the
■winter, as they were frequently found frozen. A number were also
obtained that had maimed themselves by flying against the telegraph-wires
at Coyote Station.

FRINGILLID^l.
:i. Chrysomitris tristis. One small flock seen.

4. Plectrophanes nivalis. Flocks, sometimes of large size, were seen
wheeling about over the plains nearly every day, in their usual restless
manner.

5. Plectrophanes Maccownii. Common in small flocks. Easily ap-
proached, and far less erratic in their movements than the preceding
species.

C. Plectrophanes ornatus. More or less frequent in small flocks, but
far less numerous than the preceding, or than they were in summer at
Fort Hays.

7. Spizella monticola. Frequent along the wooded parts of the
streams.

CORVIDJE.

8. Corvus corax. Four or five were seen feeding on some buffalo car-
casses on the divide between the North and South Forks of the Solomon,
fifteen miles from the nearest timber.

ALCEDINID.3E.

9. Ceryle alcyon. One was seen on the Saline, north of Coyote
Station.

PICID./E.

10. Picus pubescens. One was seen in some timber on the Saline,
and one on Big Creek, near Fort Hays.

11. Colaptes auratus. Two were observed on the Saline, north of
Coyote Station.

FALCONIDiE.

12. Falco peregrinus. Not common.

13. Falco columbarius. Occasional.

14. Astur atricapillus. A single individual observed.

15. Buteo lineatus. Frequent.

1G. Archibuteo lagopus. Common. Most numerous of the rapacious
birds.

1 7. Aquila chrysaetos. Frequent.

144 BULLETIN OF THE

18. Haliaetus leucocephalus. Common.

19. Circus cyaneus, var. hudsonius. A single individual seen.

STRIGIDiE.
20.? Otus " Wilsonianus." An owl was beard at one of our camps on
the Solomon, supposed to be of this species.

21. Athene hypogaea. Several were seen just at nightfall near Buffalo
Station. Said to be more or less frequently observed in mild weather
throughout the winter.

MELEAGRIDJE.

22. Meleagris gallopavo. Said to be common on the streams as far
west as the timber extends.

TETRAONIDJ3.

23. Cupidonia cupido. A few occur as far west as Coyote, where
they have recently made their appearance from the east.

24. Pedicecetes phasianellus, var. columbianus. Common along

the streams.

PERDICID^.

25. Ortyx virginianus. Not yet common west of Fort Hays, though
said to have been observed at Coyote.

IV. List of Birds observed at C/tri/enne, Wyoming Territory, from
August 1G to August 28, 1S71 ; with Annotations.

Cheyenne, from its situation in the midst of the Plains, form- a
locality possessing peculiar interest ornithologically. Its elevation
above the level of the sea is said to be G,041 feet. The nearest timber
is twenty miles distant, but along the bed of Crow Creek — a small
stream near the town, consisting, at this season, of little more than a
chain of slight pools — were scattered clumps of rosebushes and low-
willows. The latter were rarely more than three to six feet in height,
grew very much scattered, and were nearly destitute of foliage, their
leaves having been devoured by cattle. Although but forty-one species
were obtained or observed here, it is probable that even a number con-
siderably less than tins would include all that regularly breed here.
The abundance of the Tyrannidce found here at this season is one of
the most interesting ornithological features of the locality, since they
would hardly be expected in very great number or variety at points so
remote from timber, xllthough tin' greater part were young bird-, and
may have come from woodlands, probably the greater number and per-

MUSEUM OF COMPARATIVE ZOOLOGY. 145

haps all the commonly wood-inhabiting species enumerated below, breed
sparingly among the low willows that grow along Crow Creek.

TROGLODYTIDjE.

1. Troglodytes aedon. Frequent.

SYLVICOLID.Eh

2. Dendrceca aestiva. Two or three specimens obtained. Not
common.

3. Wilsonia pusilla. Several specimens obtained. Rather more fre-
quent than the last.

4. Icteria virens. One specimen obtained, which was the only one seen.

HIRUNDINID^J.

5. Hirundo horreorum. Frequent near the town.

6. Hirundo lunifrons. Moderately common.

7. Collurio ludovicianus. Moderately common.

ALAUDID.2B.

8. Eremophila alpestris. Common.

FRINGILLID.E.

9. Chrysomitris tristis. Several small flocks seen flying oyer.

10. Plectrophanes Maccownii. Abundant. In its notes and mode of
flight not readily distinguishable from P. ornatus, for which species we at
first mistook it. The latter was not observed at this locality.

11. Chondestes grammaca. Common.

12. Pocecetes gramineus. Abundant. All the specimens obtained
were very palely colored, the young of the year as well as the adult.

13. Passerculus savanna. Common.

14. Spizella socialis. Abundant. Very faintly colored, the young
especially, and hardly distinguishable from S. pallida.

15. Spizella pallida. Common.
10. Spizella pusilla. Common.

17. Calamospiza bicolor. Common.

18. Cyanospiza amcena. Not common.

19. Goniaphea melanocephala. Not common.

ICTERIDJE.

20. Molothrus pecoris. Rather common, associating with Xantlio-
cephalus icterocephalus.

21. Xanthocephalus icterocephalus. Several small flocks met
with along Crow Creek.

VOL. III. 10

14G BULLETIN OF THE

22. Sturnella ludoviciana, var. neglecta. Abundant.

23. Icterus Baltimore. Frequent. Only young birds seen, which
were very palely colored.

TYRANNIDiE.

24. Tyrannus carolinensis. One specimen obtained, and a few others
seen.

25. Tyrannus verticalis. Very abundant, and somewhat gregarious.
Chiefly young birds seen, associating in loose flocks of several dozens.
Kept exclusively in the valley of Crow Creek.

26. Contopus virens, var. Richardsouii. Common.

27. Sayornis Sayus. Common.

28. Ernpidonax flaviventris, var. diflicilis. Common.

TROCHILIDJE.

29. Selasphorus platycercus. Common.

STRIGID^.

30. Athene hypogaea. One small colony observed.

FALCONIDJE.

31. Falco peregrinus. A single individual was seen August 20th.

32. Falco sparverius. Common.

33. Buteo sp. ? A very light colored large species of Buteo was com-
mon, but none were obtained.

34. Circus cyaneus, var. hudsonius. Abundant. Nearly all seen
were birds of the year, in which the plumage was very red, much more
so than in eastern specimens of corresponding age.

CATHARTID^.

35. Cathartes aura. Frequent. Six were seen at one time feeding on
the carcass of a dog.

COLUMBID^J.

36. Zenaedura carolinensis. Common.

CHARADRIIDJE.

37. iEgialitis vociferus. Common.

SCOLQPACID^.

38. Actodromas Bairdii. Common along Crow Creek.

39. Actodromas minutilla. Common along Crow Creek.

40. Rhyacophilus solitarius. Common with the preceding.

hajjIlibje.

41. Porzana Carolina. A single individual observed. Probably not
frequent.

MUSEUM OF COMPARATIVE ZOOLOGY. 147

V. List of Birds observed at the Eastern Base of the Rocky Mountains
in Colorado Territory, between Colorado City and Denver, in July
and August* 1871 ; with Annotations.

The list given below is based on observations made on a journey
from Colorado City to Denver, during the first two weeks of August,
supplemented by a few notes made at Denver during the first week of
July. Probably five sixths of the species breed at the localities where
they were observed. Four days were spent in the neighborhood of
Colorado City, two at Lake Pass, and about ten days at Denver. The
distance between Colorado City and Denver is nearly one hundred
miles. The highest point is at Lake Pass, on the divide between the
Arkansas and South Platte Rivers, which is said to be about 7,000 feet
above the sea. Though really on the Plains, our road passed quite
near the foot-hills, and along the streams there was considerable timber.
We found here, as would be naturally anticipated, a fauna in many
respects peculiar, — a blending of that of the mountains with that of
the Plains. Most of the species found on the Plains extend to the
foot-hills, and even into the valleys between them. On the other hand,
many, belonging properly to the wooded region of the mountains, fol-
low the timber belts along the streams for some distance into the
Plains. We hence have here a far richer bird fauna, through the ad-
dition of the mountain species, than is met with on the Plains proper.

TURDIDJE.

1. Turdus migratorius. Tolerably common, especially along the
creeks in the foot-hills.

2. Turdus Pallasi. Only observed on Monument Creek, at Lake
Pass ; altitude of the locality about 7,000 feet.

3. Oreoscoptes montanus. Common on Dry Creek, ten miles south
of Denver.

4. Mimus carolinensis. Common along the Fontaine-qui-bouit, near
Colorado City. Also seen in the Garden of the Gods, at Lake Pass, at
Denver, and on Bear Creek, about fifteen miles southwest of Denver,
behind the first foot-hills.

5. Harporhynchus rufus. Observed near Colorado City, and obtained
on Bear Creek, in the foot-hills southwest of Denver.

* From July 4th to 8th, and August 1st to 13th.

148 BULLETIN OF THE

CINCLID^I.

6. Cinclus mexicanus. Seen on the Fontaiue-qui-bouit, at Colorado
City, August 1st.

SAXICOLIDJE.

7. Sialia mexicana. One pair observed in the foot-hills west of Denver.
There is also a specimen in the Museum, collected near Denver. It is
from the late Mr. Cassin's collection, and bears the following label : " Sialia
mexicana, Clear Creek, Rocky Mts., K. T., July, 1859. W. S. Wood, Jr."

8. Sialia arctica. Many seen, and three shot, a few miles north of
Colorado City.

SITTID.S2.

9. Sitta carolinensis. A single specimen was seen about twenty miles
north of Colorado City, on the Monument.

TROGLODYTIDJE.

10. Salpinctes obsoletus. Obtained in the foot-hills southwest of
Denver, on Bear Creek.

11. Catherpes mexicanus. Common in the Garden of the Gods,
near Colorado City. Seen only on the bare rocks. The vertical sandstone
cliffs of the Garden of the Gods seemed to afford them a favorite haunt, over
which they flitted to the highest points of the naked cliffs. Their shrill,
ringing notes reverberated among the cliffs with almost incredible loudness,
it seeming almost impossible that so small a bird should be able to produce
such penetrating and startling echoes.

12. Troglodytes aedon. Common everywhere.

A bird supposed to be Chamcea fasciuta was observed in the foot-hills near

Colorado City. Although no specimens were obtained, it was seyeral times

seen, and watched at a distance of only a few yards, and I feel confident it

was that species, though previously known only from localities as distant as

Lower California.

SYLVICOLIDJE.

13. Icteria virens. Common near Colorado City, and also observed
near Denver.

14. Dendrceca Auduboni. Common along the streams at the foot of
the mountains from Colorado City to Denver. Properly a bird of the moun-
tain fauna.

15. Dendrceca aestiva. Occasional from Colorado City to Denver.

16. Setophaga ruticilla. Common in the foot-hills west of Denver,
the first week in July, and also seen at Colorado City.

HIRUNDINIDJE.

17. Hirundo horreorum. Generally dispersed, but not numerous.

18. Hirundo lunifrons. Abundant at Denver, common at Colorado
City, and frequently seen between these points.

MUSEUM OF COMPARATIVE ZOOLOGY. 140

19. Hirundo thalassina. Common at the Garden of the Gods, and
about Castle Rock, at Lake Pass, breeding in holes in the rocks, instead of in
hollow trees, as is its usual custom.

20. Cotyle serripennis. A few seen along the South Platte at Denver.

VIREONIDJE.

21. Vireo gilvus, var. Swainsoni. One was shot on Kettle Creek,
near its junction with the Monument, where also others were seen. Paler
than eastern specimens, and pertaining to V. Swainsoni Baird, which may
be recognized as the western paler race of V. gilvus.

22. Viero solitarius, var. plurnbeus. Two shot at the same locality
as the last, the only point where they were met with. Paler than eastern
specimens, with barely a trace of olive above and on the sides, but appears
to be merely the pale western race of V. solitarius. Other specimens are
in the Museum of Comparative Zoology from Colorado.

LANIIDJE.

23. Collurio ludovicianus. Rather common in the vicinity of Denver.

TANAGRIDjE.

24. Pyranga ludoviciana. A single specimen was shot near Colorado

City, the only one seen.

ALAUDID^l.

25. Eremophila alpestris. Common.

FRINGILLIDJE.

26. Chrysomitris tristis. Common at Denver and Colorado City, and
seen at intervals between these points.

27. Chrysomitris pinus. Common at the Soda Springs, near Colorado
City, in August, and also observed near Denver. Probably breeds in the
mountains, which are here but a few miles distant.

28. Chondestes grammaca. Common. Occasionally seen in con-
siderable Hocks in company with Calamospiza bicolor.

29. Fooecetes gramineus. Common.

30. Spizella socialis. Common. Seen in large flocks the first week
in August.

31. Spizella pallida. More or less frequent, associating with S. socialis,
from which, in nestling plumage, it is scarcely distinguishable.

32. Melospiza melodia. A few observed near Colorado City.

33. Calamospiza bicolor. Abundant. Moulting the first week in
August, wben the males were curiously mottled with irregular patches
of brown and black.

34. Euspiza americana. Frequent near Colorado City.

35. Goniaphea melanocephala. Observed at Colorado City and at
Denver.

150 BULLETIN OF THE

3G. Cyanospiza amceria. Common at Colorado City, and also ob-
served at Denver. Common in the foot-hills southwest of Denver.

37. Pipilo erythrophthalmus, var. oregonus. Numerous among the
foot-hills, and more or less frequent along the streams tor ten or twenty-
miles to the eastward.

38. Pipilo chlorurus. Common along the streams to some distance east
of the foot-hills, though it probably breeds only in the mountains.

ICTERIDvE.

39. Molothrus pecoris. More or less frequent.

40. Agelaeus phoeniceus. Common about Denver.

41. Sturnella ludoviciana, var. neglecta. Abundant.

42. Icterus Baltimore. Moderately common along the timbered
streams. All the specimens examined presented an exceedingly bleached
and weathered appearance.

43. Icterus spurius. Common in the vicinity of Denver.

44. Scolecophagus cyanocephalus. Common near the streams.

CORVIDJE.

45. Corvus corax. Common along the Platte near Denver, and ob-
served at intervals along Plum Creek.

46. Pica caudata, var. hudsonica. Seen at intervals along the
streams.

4 7. Cyanurus Stelleri, var. macrolophus. The form called macro-
loplius was common along the streams.

48. Aphelocoma floridana, var. Woodhousei. A single pair was
obtained near Colorado City, the only individuals seen.

49. ? Picicorvus columbianus. A small party, probably of this
species, seen near Colorado City, but no specimens were obtained. This is
the species already referred to as a probably undeseribed species of wood-
pecker.* The colors of this species correspond very closely with the sup-
posed woodpecker, and having since learned that the habits of Picicorvus
columbianus so closely resembles those of Melanerpes torquatus as to render
it easily mistaken for a woodpecker, it seems more probable that it may
have been this bird than that a large species of woodpecker inhabiting
this region should have thus far been overlooked.

TYRANNIDiE.

50. Tyrannus carolinensis. Moderately frequent from the Soda
Springs northward to Denver, ranging to the base of the mountains.

51. Tyrannus verticalis. Common at Denver, and occasionally south-
ward to Colorado City. Not seen in the mountains, nor in South Park.

* Am. Nat., Vol. VI, p. 350.

MUSEUM OF COMPARATIVE ZOOLOGY. 151

52. Contopus virens, var. Richardsoni. Tolerably frequent.

53. Sayornis Sayus. A single individual shot near Colorado City,
and one other seen.

54. Empidonax obscurus. More or less frequent in the bushes along
the streams.

ALCEDINIDJE.

55. Ceryle alcyon. Seen occasionally along most of the creeks.

CAPRIMULGID^l.

56. Antrostomus Nuttallii. Heard great numbers at our camp near
the Garden of the Gods.

57. Chordeiles popetue. The paler form, called " Henryi" of this
species was everywhere common.

CYPSELIDiE.

58. Panyptila melanoleuca. Observed only at the Garden of the
Gods, where many pairs were breeding, though sought for at Castle Rocks
and other similar places. They breed in holes and crevices in the rocks,
usually far above gun-shot. They seemed very shy, and flew mostly near
the tops of the highest rocks. Upon ascending the rocks most frequented
by them they moved to other points, and thus managed to keep generally
out of range. By spending a considerable part of two days, we procured only
four specimens, though several others were killed, which fell in inaccessible
places. They fly with great velocity and are very tenacious of life. As
they swoop down to enter their nests, the rushing sound produced by their
winois can be heard to a considerable distance. Hirundo lhalassina was
also breeding here in similar situations.

TROCHILIDJE.

50. Selasphorus platycercus. Common and quite generally dis-
tributed.

PICIDJ3.

CO. Picus villosus, var. Harrisii. Shot a single specimen at the (Jar-
den of the God s.

61. Melanerpes erythrocephalus. Common at Denver, and fre-
quent southward along Plum Creek and elsewhere where there were many
trees.

G2. Melanerpes torquatus. First seen at Monument Park. Common
in the timber along Plum ('reek.

C3. Colaptes mexicanus. Common.

STRIGIDJE.

64. Athene hypogeea. Common near Denver, and also seen near
!!hke's Mills, on Plum Creek.

152 BULLETIN OF THE

FALCONID^E.

65. Falco peregrinus. Seen at the' Garden of the Gods, Castle Rocks,
and on Bear Creek, in the foot-hills southwest of Denver.

Cij. Falco sparverius. Abundant everywhere. Very numerous in the
Garden of the Gods, where they appear to nest in holes in the rocks. The
old birds were seen to enter holes in the cliffs, and several broods of newly
fledged young seen there were evidently raised in the vicinity, although
there were no trees within several miles in which they could have nested.
The remarkable pinnacles of rock, rising vertically to a height of from 100
to 300 feet, which occur at this point, abound in holes admirably suited for
nesting-sites for these and other birds, while the only timber in the vicinity
consists of dwarfed piiions, pines, and cedars, with here and there a cotton-
wood along the neighboring creek.

67. Buteo borealis. A large red-tailed hawk was frequent every-
where between Colorado City and Denver.

G8. Circus cyaneus, var. hudsonius. Common and generally dis-
persed. Next to the sparrow-hawk, the most numerous species of Fal-
conidce observed.

CATHARTIDJE.

CD. Cathartes aura. Seen at intervals. Not apparently abundant.

COLUMBIDJE.

70. Zenasdura carolinensis. Common.

TETRAONIDiE.

71. Pedicecetes phasianellus, var. columbianus. Said to be
abundant, especially near Lake Pass.

CHAEADRIID^S.

72. .SJgialitis vociferus. Frequent at Summit Lake (Lake Pass),
and common generally along the streams.

73. ^Jgialitis montanus. Not numerous.

SCOLOPACIDJE.

71. Actodromas Bairdii. Common at Summit Lake.

75. Gambetta flavipes. A single specimen was shot at Summit Lake,
August 5th, — the only one seen.

7(1. Gambetta melanoleuca. A single specimen was shot at Summit
Lake, August -r>th. No others were seen.

7 7. Rhyacophilus solitarius. Numerous at Summit Lake.

78. Tringoides macularius. Common at Summit Lake, and along the
streams.

79. Actiturus Bartramius. Frequently observed flying over.

MUSEUM OF COMPARATIVE ZOOLOGY. 153

ARDEID^J.
80. Demiegretta ? sp. ? A single heron was seen on the South Platte,
apparently a Demiegretta.

ANATIDiE.

81.? Anas boschas. A single specimen of apparently this species was
observed at Summit Lake.

VI. List of Birds observed in South Park, Park County, Colorado
Territory, in July, 1S71 ; with Annotations.

South Park is an elevated plateau enclosed in the mountains of
Colorado Territory, occupying nearly its geographical centre. Its
average elevation is a little more than nine thousand feet, its area
nearly two thousand square miles. It is situated between the 35th
and 36th parallels, about fifty miles west of the eastern base of the
Rocky Mountains, and has a length of about sixty miles by a breadth of
about thirty. The surface of the Park is somewhat diversified, low nearly
parallel ridges running through it in a northwest and southeast direc-
tion, dividing it somewhat irregularly into a series of valleys, through
which flow the South Platte River and its tributaries. Most of the
ridges are scantily covered with pines and aspens, especially their
northern declivities, and the streams arc fringed with various species
of willow and cottonwood. The ''bunch grass " generally grows luxu-
riantly, especially in the vicinity of the streams, but considerable por-
tions of the Park are arid and alkaline, particularly to the eastward,
where the vegetation strongly resembles that of the more barren por-
tions of the plains. Here the prevailing plants are low artemisia-like
forms, rising to but a few inches, and a Utw species of Cactus.

The avian fauna of South Park is far from rich in species ; the
greater part of which are woodland birds, the remainder being such as
typically characterize the Plains. During a reconnoissance of two
weeks in duly, only fifty-four species of birds were observed, many of
which were seen but once or twice, and less than half of which were
very common. Of the numerous family Sylricolidcz, but two species
were observed; of the Tyrannidce, only representatives of two cenora
[Contopns and Empidonax) ; while swimming and wading birds were
almost wholly absent. A few others occur in close proximity to the
Park, the most of which doubtless frequent to some extent the belts of
timber that intersect it.

154 BULLETIN OF THE

From the great elevation of the Park, its fauna has a decidedly
northern aspect, and may be regarded as at least subalpine, and as
representative of the Canadian fauna of the Eastern Province of the
continent. The nights are cool even in midsummer, the average sun-
rise temperature for the summer months being probably little, if any
above 40° F. The midsummer showers are generally accompanied
with hail and a great reduction of the temperature. The maximum
temperature frequently reaches 80° in the shade, the heat at midday
being usually quite oppressive. July 14th may perhaps be taken as
an average day, when the temperature at sunrise was 38° ; at 2 r. jr.
78° in the shade; and at sunset, G0°. Although most of the birds are of
a northern type, one or two species, as, fur example, Sturnella ludovici-
ana and Zencedura carolinensis, are more or less frequent that barely
reach the Canadian fauna in the Atlantic States.

In this connection a word or two may be added in respect to the
country lying between South Park and the Plains. At Denver
(altitude 5,100 feet) the avian fauna is analogous to the Carolinian
of the Eastern Province, and extends even into the valleys among the
foot-hills. From the base of the mountains up to about 7,500 feet
the fauna is more analogous to the Alleghanian, or to that of Southern
New England. Thence upward to about 10,500 feet is a zone more
resembling the Canadian fauna of the East, or that of Northern New
England. From this point upward to the timber-line in the Snowy
Range the fauna is more nearly representative of the Hudsonian, or
that of the shores of Hudson's Bay and the valley of the McKenzie
River. Above this, in the Snowy Range, is a region dotted with snow-
fields, where are found several essentially arctic forms.

Following up Turkey Creek, by the stage-road leading from Denver
to South Park, we find along this stream the most varied fauna and
flora of the middle portion of the Rocky Mountains. Here the rain-
fall is evidently the greatest, and the vegetation accordingly the most
luxuriant. Pines and spruces thickly clothe the slope of the moun-
tains ; the streams are densely enclosed with willows, alders, cotton-
woods, and other deciduous trees and shrubs, and rosaceous and ranun-
culaceous plants predominate, giving a flora of a cold-temperate or sub-
alpine type not met with elsewhere between the Rocky Mountains and
the Appalachians, and as different from that of the Plains as if it grew
on another continent. A profusion of flowers of bright tints meet the

MUSEUM OF COMPARATIVE ZOOLOGY. 155

traveller at every step, constantly changing in species with the increase
of altitude. Gradually, as one approaches the Park, the variety of
species diminishes, the timber becomes more scanty, and on every hand
there are evidences of increasing aridity in the climate. The birds
also decrease in number and in species, till finally we enter South Park
at its northern extremity by a pass having an elevation of about 10,500
feet, and an alpine fauna and flora. A few species of birds * were last
seen as we entered the mountains, and others disappeared higher up,
where still others were for the first time observed.

Between South Park aud Pike's Peak the country is much drier than
that portion of the mountains between South Park and Denver. Leav-
ing South Park at its southeastern edge, the road thence eastward to
Colorado City, at the eastern base of Pike's Peak, passes through a
succession of open park-like tracts of country, covered with short grass.
The hills are low and rather scantily timbered, and the whole aspect
more or less arid and forbidding. The flora and fauna are far from
rich, the birds being mainly such as are found in South Park itself,
and the herbaceous vegetation also much the same.

TURDIDJE.

1. Turdus migratorius. Common everywhere.

2. Turdus Pallasi. Frequent about Fairplay, and also observed at other
points.

The Veery (Turdus fuscescens), although not observed in or about the
Park, was met with at several points between Denver and the Park, espe-
cially along the North Fork of the South Platte.

SAXICOLID^I.

3. Sialia arctica. Common everywhere.

PARIDJE.

4. Parus atricapillus. A small party were met with at Fairplay, rep-
resenting of course the septentrionalis race, characterized mainly by lighter
colors, and more especially by a broader edging of white on the quills.

TROGLODYTID^J.

5. Troglodytes aedon. Common.

SYLVICOLIDJE.

6. Dendrceca Auduboni. Common along the streams and timbered
ridges.

* Pipiln eryikrophthalmu8, var. arcticus, Harporhynchus rufus, Mimus carolinensis,
Icteria virens, etc.

15G BULLETIN OF THE

7. Wilsonia pusilla. Abundant in the willows along the streams. Saw
sometimes a dozen pairs during a morning's hunt.

The preceding two species were the only Sylvicolidce seen. Geothlypis
Philadelphia, var. Macgulivrayi, however, is doubtless more or less frequent, as
it was common everywhere in the mountains to the eastward.

HIRUNDINIDJE.

8. Hirundo horreorum. More or less generally distributed throughout
the Park, but most numerous in the vicinity of Fairplay.

9. Hirundo bicolor. Generally distributed, breeding in woodpecker's
holes.

10. Hirundo lunifrons. Common at intervals throughout the Park.
Found a large colony at Fairplay, nesting under the eaves of buildings.
Thirty-eight nests were observed on one house, all within a space of twenty
feet.

VIREONIDJE.

11. Vireo gilvus. Rather common.

12. Vireo solitarius, var. plumbeus. One shot and others seen at
Fairplay.

ALAUDIDJE.

13. Eremophila alpestris. Common throughout the Park.

FRINGILLIDiE.

14. Chrysomitris sp. ? A species of Cltrysomitris was frequently noticed
at a distance. C. pinus occurs throughout the adjoining region, and doubt-
less this was the species observed in the Park. C. tristis was not met with
after entering the mountains.

15. Carpodacus purpureus. A few pairs were seen at Fairplay.

16. Passerculus savanna. Common along the streams, but far more
numerous near the mountains. Very numerous at our camp on Jefferson
Creek (July 14), where we found nests with eggs and with young. The
numerous specimens obtained here presented great variations in size and
in markings, but no decided general differences from summer specimens
from Massachusetts.*

17. Pocecetes gramineus. More or less common.

18. Chondestes grammaca. Not numerous.

1 9. Zonotrichia leucophrys. Exceedingly numerous at Fairplay,
and everywhere more or less common. The large number of specimens
obtained were, with few exceptions, typically of the leucophrys type.

* For remarks on the numerous supposed species of this group attributed, to the
middle and western portions of North America, see Bull. Mus. Comp. Zool., Vol. II, pp.
272-278, April, 1871.

MUSEUM OF COMPARATIVE ZOOLOGY. 157

The others were intermediate between this form and the so-called' Z.
Gambeli.*

20. Junco "caniceps." Abundant at Fairplay, and generally common
near the borders of the Park.

21. Melospiza melodia. Occasionally seen along the streams, but no-
where very common. Song undistinguishable from that of the eastern
bird. Nest and eggs similar.

22. Melospiza Lincolnii. Abundant along the streams, and especially
numerous near the mountains.

23. Spizella socialis. Not common. But one specimen obtained, and
but few observed. The one obtained is scarcely distinguishable from east-
ern examples, except in being a little lighter colored^

24. Calamospiza bicolor. But few seen in the Park, and only near
its eastern border. Numerous at one or two points on the road from South
Park to Colorado City.

25. Pipilo chlorurus. Common near the streams.

ICTERIDJE.

2G. Molothrus pecoris. Moderately frequent.

27. Sturnella ludoviciana. Abundant. One of the characteristic
species of the open portions of the Park.

28. Scolecophagus cyanocephalus. Abundant, keeping mainly near

the streams. Young full-fledged, and most of the old birds moulting before

the middle of July.

CORVIDJE.

29. Corvus corax. A few pairs observed near Fairplay.

30. Pica caudata, var. hudsonica. Frequent along the streams.

* The sole difference which has been supposed to constantly separate Z. Gambeli from
Z. leucophrys consists in the superciliary stripe being continuous to the bill in Z. Gam-
bell, while in Z. leucophrys it terminates at the anterior canthus of the eye, being cut off
at this point by a black line running from the eye to the black stripe on the head, or by
the black extending down, so as to cover the lores. The extension of the black over the
lores is, however, quite variable, especially in specimens from the Rocky Moun-
tains. In individuals referable to Z. leucophrys, the black covers the lores completely;
in others it extends only as low as the middle of the eye, and in others again not as low
as the eye, the ash in front of the eye being cut off from the superciliary stripe by a
short narrow black line running upward from the anterior canthus. This line is some-
times quite broad and distinct, or it is dotted with ashy feathers, or is reduced to a few
black feathers separated by ashy ones;. in which case it is difficult to say whether the
specimen should be called Z. leucophrys, with the black line reduced to scattered feath-
ers, or Z- Gambeli, with a few black feathers in the superciliary stripe. Notwithstanding
this, the ruses are few in which the specimen cannot be referred by this criterion to one
or the other series. Yet the irregularity of the extension of the black over the lores in
either race, and the evident transition between them, seems to render their specific sep-
aration questionable; their true relations being more those of geographical races.

158 BULLETIN OF THE

TYRANNIDiE.

31. Contopus borealis. Occasional in the limber along the ridges,
but more plentiful in the mountains to the eastward of the Park.

32. Contopus virens, var. Richardsoni. Common. Found several
nests with half-grown young, July 18 to 23, at Fairplay. Nest placed in
the fork of small branches, and quite different from that of C. virens in
the eastern States, as are also its notes.*

33. Empidonax " obscurus." Common in the thick willows' near the
streams. Found a nest July 20th containing young but a day or two old,
and another the same day with young nearly ready to fly. Nests placed in
the forks of branches in dense willow clumps, and much resembled the
ordinary nest of Dendroeca cextiua. Bird very shy, hiding in the bushes,
thus rendering it very hard to shoot. Rarely seen even when but a few
yards distant, often stealing away without coming into view.

CAPRIMULGID.E.

34. Chordeiles popetue. Abundant.

TROCHILIDJE.

35. Selasphorus platycercus. Abundant.

PICID^J.

36. Sphyrapicus varius, var. nuchalis. Common in the mountains
near the eastern border of the Park.

3 7. Sphyrapicus Williamsoni. A few seen, chiefly at the eastern
side of the Park. Quite common in the mountains further eastward.

38. Melanerpes erytkrocephalus. Not common. A few pairs ob-
served at Fairplay.

39. Colaptes mexicanus. Abundant.

FALCONID.5!.

40. Falco peregrinus. One specimen was shot at Fairplay, a young
bird that came about our camp in pursuit of blackbirds.

* Specimens of Conlvpus virens from the Rocky Mountains are considerably darker
throughout than those from the Atlantic States. They generally lack the white edge to
the outer vane of the first primary, usually seen in the latter, though eastern specimens
are often without it. The western specimens are less strongly tinged with yellow be-
neath, and the axillaries are considerably darker. The greater coverts and secondaries
are less broadly edged with white. The lower mandible, instead of being yellow as in
eastern summer specimens, appears to be always dusky, — black towards the tip and
yellow only at the base. But this feature is also frequently shared by autumnal speci-
mens at the East. The whole •difference between the two hence seems to consist in the
darker tints of the western form.

The variety Richardsoni was the only form seen at Denver and Cheyenne, as well as
in the mountains, while the specimens from Eastern Kansas were of the eastern type,
differing from Massachusetts specimens only iu being somewhat more olivaceous.

MUSEUM OF COMPARATIVE ZOOLOGY. 159

41. Falco sparverius. Not common.

42. Buteo sp. '? A large Buleo was occasionally seen, but none were
procured.

43. Aquila chrysaetos. Occasionally observed.

44. Circus cyaneus, var. hudsonius. Seen occasionally throughout
the Park. Shot a pair at Fairplay.

CATHARTID^.

45. Cathartes aura. Not frequent, and seen only at Fairplay.

COLUMBIDiE.

4G. Zenaedura carolinensis. Not common. A few pairs were seen at

intervals.

TETRAONIDJE.

4 7. Tetrao obscurus. Apparently not common.

CHARADRIID-S!.

48. .ffigialitis vociferus. Common.

49. iEgialitis montanus. Not common. Saw newly hatched young
July 28th, and full-grown young the day preceding.

SCOLOPACID.S1.

50. Gambetta melanoleuca. A single specimen was shot on the
Platte, near the eastern edge of the Park, — the only one seen.

51. Rhyacophilus solitarius. A single pair was seen near Hamilton.

52. Tringoides macularius. Common along the streams.

ANATIDJE.

53. Chaulelasmus streperus. A single female was shot July 28th on
the Platte, near the eastern edge of the Park, — the only one seen.

54. Querquedula sp. ? A few pairs were seen along the streams, and
at some brackish lakes near Hamilton, probably Q. cyanoptera.

VII. List of Birds ohserved in (lie Vicinity of Mount Linclon, Park
County, Colorado, from July I'dtli to July 2Gth, 1871 ; with Annota-
tions.

The birds mentioned in the following list were all observed during a
week spent in the vicinity of Montgomery, at the northeastern base of
Mount Lincoln, at the head of the South Platte River. Doubtless all
the species mentioned breed at, or near, where they were observed ;
the list also probably includes nearly all that occur there in summer.
The region is strictly alpine in its features. Our camp was in the

ICO BULLETIN OF THE

valley of the Platte, at an altitude of about 12,000 feet, from whence
excursions were made every day by some of the party to the region
above the timber line, which is here about 13,000 feet above the
sea-level. One excursion was made to the top of Mount Lincoln,
the height of which is usually given as a little over 14,000 feet.
Three species (Anthus ludovicianus, Leucosticte tephrocoiis, Lagopus
leucurus) were obtained above timber line that are truly arctic in their
summer distribution, and nearly all the others are known to range to
high northern latitudes. Snow remains throughout the year in the
gorges nearly down to the forest line, and frosts are of almost nightly
occurrence at points considerably below Montgomery, the temperature
in July frequently falling to below 30° F. The showers of rain, which
were of almost daily occurrence during our visit, are generally attended
with heavy thunder, hail, and sleet. Ice is said to form every night at
the mining camp on Quandary Peak, about 13,500 feet above the sea.

The timber which thickly covers the lower slopes consists almost ex-
clusively of a single kind of spruce, with here and there representatives
of two species of Populus. No other conifer was observed higher up
than at a point in the Platte valley about five miles below Montgomery,
or much above 11,500 feet. Two or three kinds of willow and a small
Betula occur abundantly in the upper valley of the Platte, and on the
declivities of the mountains in places unoccupied by the heavier forest,
up to 300 to 500 feet above the limit of trees, becoming more and
more diminutive towards their upper limit. For some distance above
the forest line are beautiful grassy slopes, and a variety of herbaceous
plants, most of them producing a profusion of large, brightly tinted
flowers. Many of the species are peculiar to this elevated region,
while some are dwarfed or otherwise modified forms of species met
with much lower down. Even anions the snow-tilled jrorsies are
extensive flower-sprinkled grass-plats of great beauty, a variety of di-
minutive and exquisitely pretty plants ranging even to the summit of
Mount Lincoln, wherever there is soil enough to afford them a foothold.

Among the other arctic animals observed are the Little Chief Hare
(Lagomys prhiceps), abundant among the loose rocks from a little
below timber-line to far above it, and several alpine butterflies.

Nearly all the birds mentioned below were met with as high as the
timber line, and many ranged above it. Wilsonia pusilla, Zonotricliia
leucophrys, and Mclospiza Lincolni were nowhere more abundant than

MUSEUM OF COMPARATIVE ZOOLOGY. 161

among the dwarfed willows and birches just above the general limit of
tnes. Three species (Anthus ludovicianus, Leucosticte tephrocotis, and
Lagopus leucurus) were met with exclusively above the timber line.

From about 11,500 feet altitude up to the tree limit the fauna appears
to be strictly representative of the Hudsonian fauna of the Eastern
Province, while that above the tree limit more resembles that of the
American arctic fauna.

TURDID^J.

1. Turdus migratorius. Abundant. Frequently met with far above
timber line. Found a nest containing newly batched young within three
hundred feet of the upper limit of trees.

2. Turdus Pallasi. Common ; ranging upward to the timber line.
Its song was heard at all hours of the day at our camp, near Montgomery.

3. Myiadestes Townsendii. Several were observed at an altitude
of over 12,000 feet, or near the timber line.

CINCLID^.

4. Cinclus mexicanus. Common on the Platte above Montgomery.

SAXICOLID.S3.

5. Sialia arctica. Abundant. More numerous here than we found it
at any other point. It was seen by Mr. Bennett on the top of Mount
Lincoln, and it breeds up to the limit of trees. Saw a brood of newly
Hedged young at the extreme upper edge of the timber.

SYLVIID.E.

6. Regulus calendulus. Common as high as the timber line. Shot a
female feeding her newly Hedged young.

PARID.33.

7. Parus montanus. Common. Collected full-grown young, July 23d.

SITTID.ffi.

8. Sitta carolinensis, var. aculeata. A single individual.

TROGLODYTIDjE.

9. Salpinctes obsoletus. Observed several pairs among the rocks
near the timber line.

MOTACILLID^.

10. Anthus ludovicianus. Common among the snow-fields above
timber line. Saw young birds scarcely able to fly, July 20th.

VOL. III. 11

162 BULLETIN OF THE

SYLVICOLID^J.

11. Dendrceca Auduboni. Common up to the limit of trees.

12. Wilsonia pusilla. Abundant. Most numerous among the low
willows above the limit of trees.

HIRUNDINIDjE.

13. Hirundo horreorum. Common at Montgomery. In clear weather
flies to the tops of the mountains.

14. Hirundo bicolor. Common about Montgomery, and seen far
above timber-line.

15. Hirundo lunifrons. A few seen in company with the preceding.

FRINGILLID^.

1G. Leucosticte tephrocotis.* Common above timber line on Mount
Lincoln, breeding among the snow-fields. The common form of L.
tephrocotis appears to be abundant in winter throughout the mountains
of Colorado, whence I have seen specimens collected near Denver. I
also met with it in December on the plains of Wyoming Territory, near
the Medicine Bow Mountains.

17. Carpodacus purpureus. Common at Montgomery.

18. Chrysomitris pinus. Common up to the limit of trees.

19. Passerculus savanna. Common in the valley of the Platte, and
also numerous on the mountains above timber line.

20. Pocecetes gramineus. Common, ranging considerably above
timber line.

* The specimens (4 d 2 ? ) of Leucosticte obtained on Mount Lincoln differ very much in
color from winter specimens of Leucosticte tephrocotis, as well as from any figure or
description of any form of Leucosticte I have seen. Whether they represent more than
the breeding plumage of L. tephrocotis or a well-marked southern form of that species I
am at present uncertain, being without summer specimens of that species. The follow-
ing is a description of the Mount Lincoln specimens : Male. Bill entirely black, or in some
specimens with a faint trace of yellow at the base of the lower mandible. Nasal feathers
whitish ; front of head black, fading to sooty brown on the mentum; no ashy nuchal
collar as in winter specimens of L. tejjhrocotis; above umber brown, each feather
broadly edged with bright red, fading to rosaceous on the rump and upper tail-coverts;
throat sooty brown, tinged slightly with red ; breast umber brown ; rest of under parts
crimson, fading to bright rosaceous posteriorly; wings and tail dusky, tinged with crim-
son, especially on the basal portions; lesser wing-coverts bright rosaceous. Length
6.25 to 6.75; alar extent, 11.75 to 12.60; wing, 4.05 to 4.20; tail, 2. GO to 2.87. Female
similar, but duller colored and smaller. Different specimens vary considerably in the
intensity and amount of red. Besides wanting the gray nuchal collar, these specimens
have the rosaceous of winter specimens replaced by bright red, and the bill black
instead of yellow.

Since writing the above, I have had an opportunity of examining several specimens
of Leucosticte killed at Central City, Colorado, in March, 1S69, by Mr. F. E. Everett, and

MUSEUM OF COMPARATIVE ZOOLOGY. 163

21. Zonotrichia leucophrys. Exceedingly numerous, even to a con-
siderable distance above timber line.

22. Junco "caniceps." Common, ranging considerably above timber
line.

23. Spizella socialis. Frequent about Montgomery.

24. Melospiza LincoLuii. Very numerous along the Platte, and also
common lor some distance above timber line.

ICTERID.E.

25. Scolecophagus cyanocephalus. Common at Montgomery, and
ranges to the tops of the mountains.

CORVIDJE.

26. Perisoreus canadensis. Common.

27. Cyanura Stelleri, var. macroloplia. More or less frequent.

TROCHILID^I.

28. Selasphorus platycercus. Common. Repeatedly seen about
the flowers far above timber line.

PICID-ffil.

20. Picus villosus. The variety Harrisii was more or less frequent
up to the timber line.

30. Picoides americanus, var. dorsalis. Common up to the timber
line.

31. Colaptes mexicanus. Common up to the timber line.

FALCONID^I.

32. Buteo sp. ? A large Buteo was frequently observed.

by him presented to the Boston Society of Natural History. Of three males, one
(marked "young male ") differs but little from the Mount Lincoln specimens, it having
no ash on the head. Another corresponds very nearly in color with the so-called L.
griseinucha, and another nearly as well with the so-called L. liitoralis. Although these
birds may have been born at widely separated localities, it seems probable that some
of the differences whereon certain species of Leucosticte have been founded may be
only individual variations. It is to be noticed, however, that the amount of ash on the
head, and the intensity of the colors, vary with locality from the north southward; the
most southerly form having no ash on the head, the bill black instead of yellow, and
the red of a brighter tint than those from more northern localities. The type of L.
tephrocotis was a male, killed on the Saskatchewan in May (see Faun. Bor.-Am., II,
p. 266), in which the ash formed a narrow nuchal band. In L. griseinuclia, a more
northern form, the gray involves nearly the whole head and the throat; and in L. liitoralis
and campestris there is more gray on the head than in tephrocotis, and they also appear
to be more northern in their distribution. In view of these facts, it s?ems probable that
the Mount Lincoln specimens above described represent the smaller, brighter-colored
southern race, in which the ash on the head has entirely disappeared.

164 BULLETIN OF THE

TETRAONID^.

33. Tetrao obscurus. Apparently more or less common. One was
shot in the upper edge of the timber.

34. Lagopus leucurus. Common above timber line. Said to descend
into the timber in winter, when many are killed by the miners for food.

SCOLOPACID^l.

35. Tringoides macularius. Common along the Platte to its source.
Its nest and eggs were found at Montgomery, July 24th,

ANATID^S.

36. Mergus merganser. A pair seen at Montgomery.

VIII. List of Birds collected in the Vicinity of Ogden, Utah Terri-
tory, from September 1st to October St/i, 1871 ; with Annotations.

The region to which the following remarks refer embraces the north-
eastern portion of the valley of the Great Salt Lake, including a por-
tion of the lake shore near the mouth of the Weber River, the lower
portion of the Weber valley, Ogden Canon, and the mountains north of
Ogden. It thus includes a great variety of surface, including the sage-
covered plains that form the eastern border of the lake, the numerous
lagoons and marshes about the mouth of the Weber River, the thickets
of willow and cottonwood that border both this river and the Ogden,
and the scantily wooded or almost naked slopes of the adjoining
mountains. From the lateness of the season, a few of the summer
residents had already migrated, and many others that pass the summer
in the mountains or to the northwards had become common. The
higher parts of the Wahsatch Range, which bounds the Great Salt Lake
Valley to the eastward, doubtless form the summer haunts of most of
the land birds observed here, which leave the valley in summer, since
the higher peaks of the range rise to the snow line.

The great abundance of aquatic birds that frequent the vicinity of
the Great Salt Lake has long been known as one of the characteristic
ornithological features of this interesting locality, especially the abun-
dance of pelicans, gulls, cormorants, avocets, stilts, ducks, and other
wading and swimming birds, many of which regularly pass the summer
here and breed on the islands of the lake. Yet the bird fauna of this
peculiar region has not as yet been fully explored, and hence still offers a
highly attractive and promising field to the collector. The only special

MUSEUM OF COMPARATIVE ZOOLOGY. 165

report that has as yet appeared respecting the ornithology of the Great
Salt Lake Valley is a list of thirty-one species contained in Stansbury's
Report,* prepared by Professor Baird from the collections made by
Captain Stansbury during his admirable survey of this region in 1849
and 1850. Since that time the region immediately under consideration
has undergone important changes. Through the industry and energy
of the Mormon emigrants, large portions of the arid plains that sur-
round the lake have been transformed, by irrigation, from a desert to
productive farms, abundantly provided with orchards and shade-trees.
So great a modification of the flora has, of course, induced a correspond-
ing change in the fauna, so that now many birds are common that were
formerly rare, especially among the granivorous and fruit-eating kinds.
At the same time the water-fowl have greatly decreased, and have
acquired the wildness characteristic of their tribe in the older settled
portions of the country. Ducks are still abundant, but, being subject to
constant persecution from juvenile and other sportsmen, their numbers
are said to be annually appreciably decreasing.

The present list contains one hundred and thirty-seven species col-
lected or observed by Mr. Bennett and myself during five weeks
spent almost constantly ornithologizing. As almost every excursion
added to our collection one or more species we had not previously seen,
it is presumably more or less deficient for even the single month (Sep-
tember) during which most of our observations were made.

TURDIDJS.

1. Turdus migratorius. Common. Said to have been very rare in
the Great Salt Lake Valley when it was first settled, but having been
carefully protected, it has gradually increased in numbers till it is now a
common bird. Doubtless the successful cultivation of the smaller fruits
has also done much towards increasing its numbers, by attracting them
from less favoring localities. Many are said to remain here all the year,
though it is much more numerous in fall and spring than during either
summer or winter. Most of the specimens taken by us are much paler
than the robin of the East, some of them presenting an exceedingly pallid
aspect.

2. Turdus Pallasi. One shot September 11th.

3. Miraus carolinensis. Common. As numerous in the dense thickets
along the Weber River as I ever saw it at the East.

* Stansbury's Exploration and Survey of the Valley of the Great Salt Lake of Utah,
1S52, pp. 314-325.

166 BULLETIN OF THE

4. Oreoscoptes montanus. Common. Called " Gray-Bird," and
also " Mocking-Bird." Very destructive to the fruit, even the peach not
escaping its rapacity.

SAXICOLIDJ3.

5. Sialia arctica. Common in spring and fall.

CINCLID.E.

6. Cinclus mexicanus. Common along the Ogden and Weber Rivers.
Shot fourteen in Ogden Canon in the course of an hour or two, October
2d, and saw several others.

SYLVIIDJE.

7. Regulus calendulus. Shot September 11th and later. Probably
not uncommon in fall and spring.

PARIDJE.

8. Paras atricapillus, var. septentrionalis. Abundant.

TEOGLODYTID33.

9. Salpinctes obsoletus. Abundant in the Wahsatch Mountains as
far down as the first " bench." About the first of October we saw them
several times on the shore of Salt Lake, near the mouth of the Weber
River, twelve or fifteen miles from the mountains. Its preference for rocks
even here was manifested by one which had chosen a heap of stones as a
temporary resting-place.

10. Cistothorus stellaris. Abundant in the marshes everywhere.

MOTACILLID.E.

11. Anthus ludovicianus. Abundant after September 15th.

SYLVICOLID^I.

12. Geothlypis trichas. Common.

13. Geothlypis Philadelphia, var. Macgillivrayi. Apparently not
uncommon.

14. Icteria virens. Moderately common.

15. Helminthophaga ruficapilla. One shot September 20th, and
others seen. Apparently common.

1G. Helminthophaga celata. First seen September 11th. Common
later.

17. Dendroeca Auduboni. Common after the 15th of September.

18. Dendroeca Blackburniae. Not common. A few specimens ob-
tained.

19 ? Dendroeca nigrescens. A bluish warbler was once or twice seen
which was probably (if this species.

MUSEUM OF COMPARATIVE ZOOLOGY. 167

20. Dendrceca aestiva. Common.

21. Wilsonia pusilla. Common.

22. Setophaga ruticilla. One seen September 8th.

HURUNDINID.SJ.

23. Hirundo horreorum. Common.

24. Hirundo lunifrons. But few were seen, owing probably to the
lateness of the season. Their nests were very numerous on the cliffs of
Weber and Echo Canons.

25. Hirundo thalassina. A few seen September 11th.

26. Cotyle serripennis. Moderately common.

VIREONID-SJ.

27. Vireo olivaceus. More or less common.

28. Vireo gilvus. Rather common.

29. Vireo solitarius. Rather frequent. Somewhat paler colored than

in the Eastern States, but much brighter than those obtained in July in

Colorado.

AMPELIDjE.

30. Ampelis cedrorum. Rather common.

LANIIDJE.

31. Collurio ludovicianus. Quite common. Said to breed.

TANAGRID.E.

32. Pyranga ludoviciana. Frequent.

ALAUDID^.

33. Eremophila alpestris. Common.

FRINGILLIDJE.

34. Carpodacus purpureus. Not numerous.

35. Chrysomitris tristis. Abundant.

36. Chrysomitris psaltria. Apparently common, associating with the
preceding.

37. Passer domesticus. Recently introduced and apparently flour-
ishing.

38. Passerculus savanna. Common.

39. Pocecetes gramineus. Abundant.

40. Coturniculus passerinus. Common.

41. Zonotrichia leucophrys, var. Gambeli. Abundant. Mainly of
the form called " Gambeli," but a typical leucophrys was also taken.

42. Junco " oregonus." Common after about October 1st. The
specimens taken scarcely differ from fall specimens of J. lujemalis, except
in the rufous on the sides.

168 BULLETIN OF THE

43. Poospiza Belli. Common on the Plains among the sage-brush.

44. Spizella socialis. Abundant.

45. Spizella pallida. Abundant.

46. Melospiza melodia. Very abundant. A little paler than speci-
mens from the Atlantic States, but not so markedly so as is the case in
some other species. From the locality it should, however, be referable to
the M. fallax, a name applied to the paler form of AT. melodia from the
Iiocky Mountain region.

47. Melospiza Lincolnii. Exceedingly abundant.

48. Passerella iliaca, var. scliistacea. Not numerous. First ob-
tained September 10th.

49. Goniaphea melanocephala. Common in summer. Leaves about
the first week of September. Called " Pea Bird," it being very fond of
young peas, and is hence regarded as obnoxious.

50. Cyanospiza amcena. Not common.

51. Pipilo erythrophthalmus, var. arcticus. Common.

52. Pipilo chlorurus. Common after September 20th.

ICTERIDJE.

53. Dolichonyx oryzivorus. Common. Said to breed here.

54. Agelaeus phceniceus. Exceedingly abundant. Flocks of thou-
sands seen about the marshes. In color, especially in respect to the
shoulder-patch, it closely resembles the form prevailing in the Atlantic
and Gulf States. A few specimens were taken that had very small spots of
black on the ends of the middle coverts, but in none were they so well devel-
oped as to typically represent the so-called A. gubernator. One specimen
was taken which had the exposed portions of the greater coverts of the
same color as the middle ones, thus forming a very broad conspicuous
brownish -yellow patch on each wing.

55. Xanthocephalus icterocephalus. Abundant, occurring in large
flocks about the marshes, associating more or less with the preceding. The
colors vary exceedingly in different individuals, from some young females
that are only tinged with pale yellowish on the throat, to some males that
have the whole throat and breast intense orange red.

56. Sturnella ludoviciana, var. neglecta. Abundant, typically rep-
resenting the so-called -S. neglecta. The notes of some individuals, how-
ever, were scarcely different from those of the eastern bird, though gen-
erally the song is dissimilar and much richer.

57. Icterus Bullockii. Said to be common in summer. Saw only
stuffed specimens in collections at Salt Lake City, said to have been taken
in the vicinity.

58. Scolecophagus cyanocephalus. Very abundant.

MUSEUM OF COMPARATIVE ZOOLOGY. 169

CORVID^J.

59. Corvus corax. Common.

GO. ? Corvus americanus. None were seen, but it was said to be
more or less frequent.

61. Pica caudata, var. hudsonica. Common.

G2. Cyanura Stelleri, var. macrolopha. Common in the moun-
tains, and occasional in the thickets along the streams ; called " Mountain
Jay."

63. Aphelocoma floridaua, var Woodhousei. Common.

TYRA.NNID.E.

64. Tyrannus carolinensis. One obtained and two or three others
seen.

65. Tyrannus verticalis. Not common.

66. Contopus borealis. Not common. Several seen among the cot-
ton woods along Weber River.

6 7. Contopus virens, var. Richardsonii. Not common ; seen in the
same localities as the last.

68. Empidonax flaviventris, var. difficilis. Not uncommon in
favorable localities.

69. Empidonax "obscurus." Common.

ALCEDINID.E3.

70. Ceryle alcyon. Common.

CAPRIMULGIDJE.

71. Antrostomus Nuttallii. Abundant in the mountains near Ogden.
One was seen October 7th, half-way up the mountains, during a severe
snow-storm, the snow being already several inches deep.

72. Chordeiles popetue. Abundant till nearly October 1st.

TROCHILIDJE.

73. Selasphorus platycercus. Common.

TICIBM.

74. Picus sp. ? A small spotted woodpecker was seen a few times,
near the mouth of Ogden Canon, which was apparently P. puboxcens.

75. Colaptes mexicanus. Rather common along the Ogden and
Weber Rivers. Tliey were accustomed to frequent a high clayey bank
near the town, in which were a number of holes. These the woodpeckers
entered, and we repeatedly saw them sitting in them, with their heads
thrust out at the entrances. About a dozen individuals were quite regu-
larly seen around the place, especially early in the morning. We found

170 BULLETIN OF THE

on examining these holes that they entered horizotnally but a few inches,
and then turned abruptly downward, having about the size and form of the
holes the Colaples aural us is accustomed to make in decayed trees. They
appeared much attached to the spot, and from all the circumstances we
were led to believe that they had nested in the holes which we saw them
frequenting, as there were no trees within several miles that could have
served them as nesting sites.

STRIGIDJE.

76. Otus vulgaris, var. Wilsonianus. One specimen taken.

77. Bubo virginianus. Occasional. Said to be quite frequent in winter.

78. Athene hypogaea. Several colonies were met with in the vicinity
of Ogden, living chiefly in holes made by coyotes and other Canidce, no

pecies of Cynomys occurring here.

FALCONIDJE.

79. Falco peregrinus. Common about the marshes of Salt Lake
preying upon the water-fowl.

80. Falco columbarius. Moderately frequent.

81. Falco sparverius. Exceedingly numerous. Along the Weber
River, below Weber Canon, a dozen or more were frequently seen in the
air at the same instant, catching grasshoppers.

82. Buteo borealis. Not uncommon.

83. Circus cyaneus, var. hudsonius. Abundant. Especially nu-
merous about the marshes, where they were constantly seen pursuing
the blackbirds, but apparently with indifferent success.

84. Aquila chrysaetos. Said to be occasionally killed.

85. Haliaetus leucocephalus. More or less frequent.

86. Pandion haliaetus. A few seen. Said to be common in summer.

CATHARTIDJE.

87. Cathartes aura. Common.

COLUMBID.E.

88. Zenaedura carolinensis. Abundant. Said to breed generally on

the ground.

TETRAONIDJE.

89. Tetrao obscurus. Reported to be common in the mountains.

90. Centrocercus urophasianus. Common. Much less numerous
than formerly. A few years ago it is represented to have been exceed-
ingly abundant, but less than a dozen were seen by us during a quite
thorough reeonnoissance of the northeastern portion of the Great Salt Lake
Valley, they having never been known to be so scarce there before.

91. Pedioecetes phasianellus, var. columbianus. Common, but said

MUSEUM OF COMPARATIVE ZOOLOGY. 171

to be much less numerous the present season than usually. In the Great
Salt Lake Valley we have of course the paler or light-colored form of the
Plains, to which Mr. D. G. Elliot has restricted the name P. columbianus*
the darker form of the more heavily wooded regions of the north being con-
sidered by him to be the true P. phasianellus (=P. Kennicotlii Suckley).f

92. Bonasa umbellus. The ruffed grouse is said to occur sparingly
in the mountains.

PERDICLIXE.

93. Ortyx virginianus. A few pairs of the common quail were intro-
duced from the East last year, and are said to have each raised a brood of
young.

94. Lophortyx californicus. A few pairs were introduced a short
time since, and are said to have raised young the present year (1871).

CHARADRIID.SJ.

95. iEgialitis vociferus. Abundant.

SCOLOPACID^.

96. Gallinago Wilsoni. Very abundant.

97. Macrorhampus griseus. Abundant September 25th and later;
perhaps breeds here.

98. Pelidna americana. Common.

99. Actodromas minutilla. Not common.

100. Gambetta melanoleuca. Abundant.

101. Gambetta flavipes. Not common.

102. Rhyacophilus solitarius. Not common.

103. Tringoides macularius. Not common. Only two nr three were
seen.

PHALAROPODID^I.

104. Phalaropus "Wilsoni. Abundant. Said to breed in great num-
bers on the islands in the lake.

RECURVIROSTRID^l.

105 Recurvirostra americana. Very abundant. Flocks of several
thousands seen on the shores of the lake. Said to breed on the islands.

106. Himantopus nigricollis. Common, but far less numerous than
the preceding, with which it freely associates. Both this and the pre-
ceding species are known locally as " White Snipes."

GRUID^l.

107. Grus canadensis. Occasional in fall and spring.

* Proc. Acad. Nat. Sci. Philad., 1862, p. 403.
t Ibid ,1861, p. 361.

172 BULLETIN OF THE

TANTALIDJE.

108. Ibis falcinellus, var. Ordii. Common in summer. Leaves early
in September. Obtained five specimens out of the seven we saw. Called
" Black Snipe." Said to have become numerous only during the last two
or three years.

109. Ibis alba. Only a few seen. Said to be frequent in summer.

ARDEIDJE.

110. Ardea herodias. Rather common.

111. Botaurus lentiginosus. Common.

11 2. Nyctiardea grisea, var. nasvia. Common.

RALLID^3.

113. Rallus virginianus. Common.

114. Rallus crepitans. Said to be common.

115. Porzana Carolina. Common.
11G. Fulica arnericana. Abundant.

ANATIDJE.

117. Anser hyperboreus. Common in fall and winter, appearing
about October 1st. Called " White Brant."

118. Bernicla canadensis. Abundant. Exceedingly numerous in
the fall. A few said to breed.

119. Anas boschas. Common.

120. Dafila acuta. Abundant.

121. Nettion carolinensis. Abundant.

122. Querquedula cyanoptera. Abundant.

123. Spatula clypeata. Common.

124. Chaulelasmus streperus. Abundant.

125. Mareca arnericana. Common.

126. Aix sponsa. Common.

127. Fulix marila. Common.

128. Aythya ferina, var. arnericana. Abundant.

129. Erisniatura rubida. Common.

130. Mergus merganser. Common.

PELECANIDJE.

131. Pelecanus erythrorhynchus. Loaves about September 1st.
Stansbury found it breeding' on the islands in immense numbers.*

GRACULID^J.

132. Graculus dilophus. Common. Called " Black Swan " !

* See Stansbury's Report, pp. 179 and 188.

MUSEUM OF COMPARATIVE ZOOLOGY. 173

LARID.2E.

133. Larus delawarensis. Abundant. Has the singular habit of
feeding on grasshoppers, which it captures in the air. Breeds in large
numbers on the islands.

134. Chroecocephalus Philadelphia. First seen about October 1st.
Three specimens were taken, which differ from average eastern examples
in having a much shorter, relatively thicker, and less decurved bill. The
difference is so great as to be quite striking.

135. Xema Sabini. A single specimen was taken September 28th,
the only one seen.

PODICTPIDJE.

136. Podiceps cornutus. One specimen taken. Perhaps common.

137. Fodilymbus podiceps. Common.

IX. Summary List of Birds observed in Kansas, Colorado, Wyoming,
and Utah, in 1871.

The preceding lists being more especially a record of observations
respecting the character of the avian fauna of particular localities, the
following general summary is appended to show at a glance the area
over which the different species were respectively met with, and to
indicate the vertical range of those observed in the mountains of
Colorado. It is to be understood that the species are summer resi-
dents at the localities named, unless otherwise stated. Some were seen
only during migration, and unless known to breed at the localities
where they were observed, the season of observation is stated. The
altitudes given are of course only approximate, though believed to be
in the main correct. The list has some negative value, as it is proba-
ble that at most of the localities cited few, if any, other species were
characteristically common.

TURDID-S3.

1. Turdus mustelinus. Eastern Kansas.

2. Turdus Pallasi. Topeka, Kansas (one specimen) ; mountains of
Colorado, from about 8,000 feet to timber line ; Ogden, Utah.

3. Turdus Swainsoni. Eastern Kansas (May).

4. Turdus fuscescens. Mountains of Colorado, up to about 8,500
feet.

5. Turdus migratorius. Eastern Kansas (rare) ; mountains of Col-
orado up to timber line ; Ogden, Utah.

174 BULLETIN OF THE

6. Harporhynchus rufus. Eastern Kansas ; western edge of the
Plains of Colorado, and in the mountains up to about 7,500 feet.

7. Oreoscoptes montanus. Western edge of the Plains of Colorado ;
Ogden, Utah.

8 Mimus polyglottus. Eastern and Middle Kansas.

9. Mimus carolinensis. Eastern and Middle Kansas ; western edge
of the Plains in Colorado, and in the foot-hills up to about 7,500 feet ;
Ogden, Utah.

CINCLID^J.

10. Cinclus mexicanus. Mountains of Colorado from the Plains up
to timber line ; Ogden, Utah.

SAXICOLIDiE.

11. Sialia sialis. Eastern Kansas, west to Fort Hays.

12. Sialia arctica. Mountains of Colorado, from the Plains up to
timber line ; Ogden, Utah.

13. Sialia mexicana. Foot-hills west of Denver, Colorado.

SYLVIID.S2.

14. Regulus calenaulus. Mountains of Colorado, from about 8,000
feet to timber line ; Wahsatch Mountains, near Ogden, Utah.

15. Polioptila caerulea. Eastern Kansas.

CHAMMD51.
16.? Chameea fasciata. Mountains of Colorado, west of Colorado City-

PARID^I.

17. Parus atricapillus, var. septentrionalis. Eastern and Middle
Kansas; mountains of Colorado, up to about 11,000 feet; Green River,
Wyoming Territory ; Ogden, Utah.

18. Parus montanus. Mountains of Colorado, from 8,000 feet up to
timber line.

19. Lophophanes bicolor. Eastern Kansas.

SITTID^J.

20. Sitta carolinensis. Eastern Kansas; var. aculeata, mountains
of Colorado up to timber line.

21. Sitta pusilla, var. pygmsea. Mountains of Colorado, up to about
8,000 feet.

TROGLODYTID.52.

22. Troglodytes aedon. Eastern and Middle Kansas; Cheyenne;
western edge of the Plains in Colorado, and in the mountains of Colorado,
up to about 10,000 feet.

23. Salpinctes obsoletus. Mountains of Colorado, from the Plains
up to timber line ; Ogden, Utah.

MUSEUM OF COMPARATIVE ZOOLOGY. 175

24. Catherpes mexicanus. Near Colorado City.

25. Cistothorus palustris. Ogden, Utah.

2G. Thryothorus ludoviciauus. Eastern Kansas.

MOTACILLID^.

27. Anthus ludovicianus. Mountains of Colorado, above timber

line; Ogden, Utah (Sept.) ; YVahsatch Mountains, probably breeding above

timber line.

SYLVICOLID.E.

28. Mniotilta varia. Eastern Kansas (May).

29. Parula americana. Eastern Kansas (May).

30. Helminthophaga pinus. Eastern Kansas.

31. Helminthophaga celata. Eastern Kansas (May) ; Ogden, Utah
(Sept.) ; Wahsatch Mts.

32. Helminthophaga ruficapilla. Eastern Kansas (May) ; Ogden,
Utah (Sept.); Wahsatch Mts.

33. Dendrceca coronata. Eastern Kansas (May).

34. Dendroeca Auduboni. Mountains of Colorado, from the Plains
up to timber line ; Ogden, Utah.

35. Dendrceca Blackburniae. Eastern Kansas (May) ; Ogden, Utah
(Sept.).

36. Dendrceca pennsylvanica. Eastern Kansas (May).

37. Dendroeca caerulea. Eastern Kansas.

38. Dendrceca discolor. Eastern Kansas.

39. Dendrceca aestiva. Eastern Kansas ; Denver, and western edge
of the Plains, Colorado; Cheyenne, Wyoming; Ogden, Utah.

40. Seiurus aurocapillus. Eastern Kansas.

41. Geothlypis trichas. Eastern Kansas ; Ogden, Utah.

42. Geothlypis Philadelphia. Eastern Kansas ; var. Macgillivrayi,
mountains of Colorado, below 9,000 feet.

43. Oporornis formosus. Eastern Kansas.

44. Icteria virens. Eastern and Middle Kansas; Cheyenne; Colora-
do, at the base of the mountains ; Ogden, Utah.

45. Wilsonia mitrata. Eastern Kansas.

46. Wilsonia pusilla. Mountains of Colorado, from the Plains to above
timber line; Cheyenne ; Green River, Wyoming (Oct.) ; Ogden, Utah.

47. Setophaga ruticilla. Eastern Kansas; mountains of Colorado,
up to about 8,000 feet ; Ogden, Utah.

TANAGRID^3.

48. Pyranga rubra. Eastern Kansas.

4 9 Pyranga ludoviciana. Mountains of Colorado, from the Plains
up to about 8,000 feet.

176 BULLETIN OF THE

HIRUNDINID^J.

50. Hirundo horreorum. Eastern Kansas to Ogden, Utah; in the
mountains of Colorado up to timber line.

51. Hirundo lunifrons. Eastern Kansas ; Ogden, Utah; in the moun-
tains of Colorado up to timber line.

52 Hirundo bicolor. Eastern Kansas (May) ; in the mountains of
Colorado up to timber line.

53. Hirundo thalassina. Mountains of Colorado, from the Plains up
to about 8,000 feet ; perhaps to timber line.

54. Cotyle riparia. Eastern Kansas.

55. Cotyle serripennis. Eastern Kansas ; western edge of the Plains ;
Ogden, Utah.

56. Progne subis. Eastern Kansas.

VIREONID-SI.

57. Vireo olivaceus. Eastern Kansas; mountains of Colorado up to
11,000 feet ; Ogilen, Utah.

58. Vireo gilvus. Eastern Kansas ; western edge of the Plains of
Colorado ; Ogden, Utah.

59. Vireo solitarius, var. plumbeus. Western edge of the Plains
of Colorado, and in the mountains up to about 10,000 feet; Ogden, Utah.

60. Vireo noveboracensis. Eastern Kansas.

61. Vireo Belli. Eastern and Middle Kansas.

AMPELIDjE.

62. Ampelis cedrorum. Eastern Kansas.

63. Myiadestes Townsendii. Mountains of Colorado, up to timber

line.

LANIIDJE.

64. Collurio ludovicianus. Eastern Kansas; western edge of the
Plains of Colorado ; Cheyenne ; Ogden, Utah.

ALAUDIDJE.

65. Eremophila alpestris. Eastern, Middle and Western Kansas ;
South Park, Colorado ; Laramie Plains ; Ogden, Utah.

FRINGILLIDjE.

66. Carpodacus purpureus. South Park, Colorado; Ogden, Utah.
6 7. Chrysomitris tristis. Eastern Kansas; western edge of the

Plains of Colorado ; Ogden, Utah.

68. Chrysomitris pinus. Mountains of Colorado, from the Plains up
to timber line.

69. Chrysomitris psaltria. Middle Kansas (?) ; Ogden, Utah.

MUSEUM OF COMPARATIVE ZOOLOGY. 177

70 Leucosticte tephrocotis. Mountains of Colorado, above timber
line ; Carbon County, Wyoming (in December).

71. Plectrophanes nivalis. Western Kansas (winter).

72. Plectrophanes ornatus. Middle Kansas.

7.5. Plectrophanes Maccownii. Cheyenne ; Western Kansas (in
winter).

7 4. Passer domesticus. Great Salt Lake Valley (introduced).

75. Passerculus savanna. Western edge of the Plains to above tim-
ber line.

7G. Pooecetes gramineus. Western edge of the Plains to above tim-
ber line.

77. Coturniculus passerinus. Kansas; western edge of the Plains;
Ogden, Utah.

78. Chondestes grammaca. Kansas ; South Park ; Cheyenne ; Lara-
mie Plains ; Ogden, Utah.

79. Zonotrichia querula. Fort Leavenworth, Kansas.

80. Zonotrichia albicollis. Eastern Kansas (May).

81. Zonotrichia leucophrys. Eastern Kansas (May) ; mountains of
Colorado up to above timber line; var. Gambeli, Ogden, Utah (Sept.);
Wahsatch Mountains.

82. Junco " caniceps." Mountains of Colorado, from about 7,500 feet
up to above timber line.

8.3. Junco " oregonus." Ogden, Utah (Sept.) ; Wahsatch Mountains.

84. Poospiza Belli. Ogden, Utah

85. Spizella monticola. Western Kansas (in winter).

8G. Spizella socialis. Eastern Kansas; western edge of the Plains;
mountains of Colorado up to timber line; Ogden, Utah.

87. Spizella pallida. Eastern Kansas; western edge of the Plains;
Ogden, Utah.

88. Spizella pusilla. Eastern Kansas ; western edge of the Plains.
80. Melospiza melodia. Eastern Kansas; western edge of the Plains;

South Park ; Ogden, Utah.

90. Melospiza palustris. Eastern Kansas (May).

91. Melospiza Lincolnii. Eastern Kansas (May); mountains of
Colorado, from about 8,000 feet to above timber line ; Ogden, Utah.

92. Peucaea aestivalis, var. Cassinii. Middle Kansas;

93. Passerella iliaca, var. schistacea. Ogden, Utah (Sept.) ;
Wahsatch Mountains.

94. Calamospiza bicolor. Middle Kansas ; western edge of the
Plains ; South Park, and mountains to the eastward ; Cheyenne.

95. Euspiza americana. Kansas; Colorado City.
9G. Goniaphea ludoviciana. Eastern Kansas.

VOL. III. J2

ITS BULLETIN OF THE

9 7. Goniaphea mclanocspliala. Middle Kansas; Denver; Chey-
enne; Ogden, Urali.

98. Cyanospiza cyanea. Eastern Kansas.

00. Cyanospiza amoena. Western edge of the Plains, and mountains
of Colorado up to about 8,000 feet ; Cheyenne ; Ogden, Utah.

100. Cardinalis virginianus. Eastern Kansas.

101. Pipilo erythrophthalmus. Eastern Kansas ; var. arcticus, west-
ern edge of the Plains of Colorado, and in the mountains up to about <s,000
feet ; Ogden, Utah.

102. Pipilo clilorurus. Mountains of Colorado from about 8,000 feet
up to timber line ; Ogden, Utah.

ICTERIDjE.

103. Dolichonyx oryzivorus. Ogden, Utah.

104. Molothrus pecoris. Kansas ; mountains of Colorado up to
ll,O0() feet; Cheyenne; Ogden, Utah.

10"). Agelaeus phoeniceus. Eastern Kansas; mountains and western
edge of the Plains of Colorado : Ogden, Utah.

106. Xanthocephalus icterocephalus. Eastern and Middle Kansas ;
western edge of the Plains; Ogden, Utah.

107. Sturnella ludoviciana, var. neglecta. Kansas; Plains of Colo
rado and Wyoming ; South Park ; Ogden, Utah.

108. Icterus Baltimore. Kansas; Cheyenne; western edge of the
Plains in Colorado.

109. Icterus Bullockii. Ogden, Utah.

110. Icterus spurius. Kansas; western edge of the Plains in Colo-
rado.

111. Scolecophagus cyanocephalus. Western edge of the Plains;
mountains of Colorado, to above timber line ; Ogden, Utah.

112. Quiscalus purpureus. Eastern and Middle Kansas.

CORVIDJE.
11.3. Corvus corax. Kansas; Colorado; Wyoming; Utah.

114. Corvus americanus. Eastern and Middle Kansas; Utah?

115. Picicorvus columbianus. Near Colorado City ? Medicine Bow
Mountains, Wyoming.

116. Pica caudata, var! hudsonica. Western Kansas; Plains and
mountains of Colorado, up to at least 11,000 feet ; Wyoming ; Utah.

117. Cyanura cristata. Eastern Kansas.

118. Cyanura stelleri, var. macrolopha. Mountains of Colorado,
up to timber line; Medicine Bow Mountains, Wyoming; Wahsatch Moun-
tains, Utah.

MUSEUM OF COMPARATIVE ZOOLOGY. 179

1 1 9. Aphelocoma floridana, var. Woodhousei. Near Colorado
City ; Ogden, Utah.

120. Perisoreus canadensis. Mountains of Colorado, above 12,000
feet; Medicine l>ow Mountains; Wahsatch Mountains.

TYRANNIDiE.

121. Tyrannus carolinensis. Eastern and Middle Kansas ; Denver;
Cheyenne ; Ogden, Utah.

122. Tyrannus verticalis. Middle Kansas; Denver; Chevenne ;
Ogden, Utah.

123. Myiarclius crinitus. Eastern Kansas.

124. Sayornis fuscus. Eastern Kansas.

125. Sayornis Say us. Western edge of Plains in Colorado ; Cheyenne.

126. Contopus borealis. Mountains of Colorado up to 12,000 feet;
Ogden, Utah (Sept.) ; Wahsatch Mountains.

127. Contopus virens. Eastern Kansas; var. Richardsoni, west-
ern edge of Plains, and mountains of Colorado, up to about 12,000 feet;
Ogden, Utah.

128. Empidonax minimus. Eastern Kansas (May).

129. Empidonax Traillii Eastern Kansas.

130. Empidonax "obscurus." Mountains of Colorado up to 12,000
feet ; Ogden, Utah.

131. Empidonax " Hammondi." * Fort Fred. Steele, Wyoming
(Oct.) ; Ogden, Utah (Sept.).

132. Empidonax flaviventris, var. difficilis. Ogden, Utah.

ALCEDINICffi.

133. Ceryle alcyon. Kansas ; Plains of Colorado and mountains up
to 9,000 feet ; Piaius of Wyoming ; Ogden, Utah.

CAPRIMULGID^E.

134. Antrostomus Nuttallii. Eastern Kansas; mountains of Col-
orado, up to about 8,000 feet ; Ogden, Utah.

135. Antrostomus vociferus. Eastern Kansas.

136. Chordeiles popetue. Eastern Kansas; var. Henryi, Middle
Kansas, and generally westward to Utah ; in the mountains of Colorado up
to 1 2,000 feet.

* Dr. Elliott Coues being at present engaged in .a revision of the Tyrannidce, all the
specimens of Empidonax collected during the present expedition have been sent to him
for examiration, and the names here provisionally a lopted are given with his approval.
Since the preceding pages were put in type, I have learned that specimens of the so-
called Empidonax "Hammondi" were among those collected at Ogden, Utah, and at
Fort Fred. Steele, Wyoming.

180 BULLETIN OF THE

CYPSELIDJE.

137. Chaetura pelasgia. Eastern Kansas.

138. Panyptila melanoleuca. Garden of tbe Gods, near Colorado

City.

TROCHILID^J.

139. Trochilus colubris. Eastern Kansas.

140. Selasphorus platycercus. Western edge of the Plains, and
mountains of Colorado to above timber line ; Cheyenne ; Ogden, Utah.

CUCULID^.

141. Coccygus americanus. Eastern Kansas.

PICID.E.

142. Picus villosus. Eastern Kansas ; var. Harrisii, mountains of
Colorado.

143. Picus pubescens. Eastern and Middle Kansas.

144. Picoides americanus, var. dorsalis. Mountains of Colorado,
from about 8,000 feet to timber line.

145. Sphyrapicus varius, var. nuchalis. Mountains of Colorado,
from 7,000 to 12,000 feet.

146. Sphyrapicus Williamsoni. South Park and the mountains to
the eastward.

147. Centurus carolinus. Eastern Kansas.

148. Melanerpes erythrocephalus. Kansas ; mountains of Colo-
rado, from the Plains to about 11,000 feet.

149. Melanerpes torquatus. Western edge of the Plains up to about
10,000 feet.

150. Colaptes auratus. Eastern Kansas.

151. Colaptes mexicanus, Western edge of the Plains and moun-
tains of Colorado up to timber line ; Ogden, Utah.

STRIGID-S3.

152 Otus vulgaris, var. Wilsonianus. Ogden, Utah.

153. Athene hypogaea. Middle Kansas; western edge of the Plains;
Cheyenne ; Ogden, Utah.

FALCONID^J.

154. Falco peregrinus. Middle Kansas; South Park; Plains of
Wyoming ; Ogden, Utah.

155. Falco colurnbarius. Ogden, Utah (Sept.) ; Wahsatch Moun-
tains.

15G. Falco sparverius. Eastern and Middle Kansas ; western edge
of the Plains, ami mountains of Colorado ; Ogden, Utah.

157. Buteo borealis. Eastern and Middle Kansas ; Colorado.

MUSEUM OK COMPARATIVE ZOOLOGY. 181

158. Archibuteo lagopus, var. Sancti-Johannis. Western and
Middle Kansas (winter) : Carton County, Wyoming (winter, abundant).

159. Nauclerus furcatus. Eastern Kansas.

ICO. Circus cyaneus, var. hudsonius. Kansas ; Colorado ; Wyoming >
Utah.

161. Aquila chrysaetos. Colorado; Wyoming; Utah.

162. Haliaetus leucocephalus. Kansas; Colorado; Wyoming; Utah.

CATHARTIDJE.

1C3. Cathartes aura. Kansas; Plains of Colorado; South Park;
Wyoming; Utah.

COLUMBIDiE.

164. Zenaedura carolinensis. Kansas; western edge of the Plains;

mountains of Colorado up to about 11,000 feet; Laramie Plains; Ogden,

Utah.

MELEAGRIDJE.

1G5. Meleagris gallopavo. Eastern and Middle Kansas; along the

streams as far as timbered ; not much to the westward of Middle. Kansas.

r

TETRAONIDJE.

16G. Tetrao obscurus. Mountains of Colorado (up to timber line);
Wyoming and Utah.

107. Centrocercus urophasianus. Laramie Plains; Carbon County,
Wyoming : Salt, Lake Valley.

168. Pedicecetes phasianellus, var. columbianus. Middle and West-
ern Kansas ; Plains of Colorado and Wyoming ; Great Salt Lake Valley.

169. Cupidonia cupido. Eastern and Middle Kansas, spreading west-
ward.

170. Bonasa umbellus. Mountains of Colorado, Wyoming, and Utah.

171. Lagopus leucurus. Mountains of Colorado, above timber line.

PERDICID2E.

1 72. Ortyx virginianus. Eastern and Middle Kansas, spreading west-
ward ; Great Salt Lake Valley (introduced).

1 73. Lophortyx californicus. Great Salt Lake Valley (introduced).

CHARA.DRIIDJE.

174. ^Ggialitis vociferus. Kansas; Plains of Colorado and Wyo-
ming ; South Park ; Great Salt Lake Valley.

1 75. iEgialitis montanus. Middle Kansas ; Plains of Colorado and
Wyoming ; South Park.

SCOLOPACID^.

176. Gallingo Wilsoni. Ogden, Utah (Sept.).

177. Macrorhampus griseus. Shores of Great Salt Lake (Sept.).

182 BULLETIN OF THE

178. Pelidna alpina, var. americana. Great Salt Lake Valley (Sept.).

179. Ereunetes pusillus. Eastern Kansas (May).

180. Actodromas maculata. Fort Leavenworth, Kansas (May).

181. Actodromas Bairdii. Western edge ot' the Plains (August).

182. Actodromas minutilla. Colorado (August 3d) ; Cheyenne
(August); Laramie Plains (August); Ogden, Utah (September).

183. Gambetta melanoleuca. Eastern Kansas (May); Lake Pass,
Colorado (August 3) ; Cheyenne (August) ; Laramie Plains (August) ;
Great Salt Lake. (September).

184. Gambetta flavipes. Lake Pass, Colorado (August) ; Ogden,
Utah (September).

185. Rhyacophilus solitarius. Eastern and Middle Kansas ; Colo-
rado (August) ; Wyoming (August).

18G. Tringoides macularius. Kansas; Colorado, up to 13,000 feet;
Wyoming ; Utah.

187. Actiturus Bartramius. Kansas; Colorado.

188. Nnmenius longirostris. Kansas.

189. Numenius borealis. Middle Kansas (June 20th, one individual).

PI1ALAROPODID.S3.

190. Phalaropus Wilsoni. Great Salt Lake.

RECURVIROSTRID.SJ.

191. Recurvirostra americana. Laramie Plains ; Great Salt Lake.

192. Himantopus nigricollis. Great Salt Lake.

GRUIDJE.

193. Grus americanus. Eastern Kansas (May).

TANTALIDJE.

194. Ibis falcinellus, var. Ordii. Great Salt Lake.

195. Ibis alba. Great Salt Lake.

ARDEID.S3.

196. Ardea herodias. Eastern Kansas; Great Salt Lake Valley.

197. Ardetta exilis. Eastern Kansas.

198. Botaurus lentiginosus. Eastern Kansas; Great Salt Lake
Valley.

199. Bntorides virescens. Eastern Kansas.

200. Nyctiardea grisea, var. naevia. Eastern Kansas ; Great Salt
Lake Valley.

RALLIDJE.

201. Rallus crepitans. Great Salt Lake Valley.

202. Rallns virginianus. Eastern Kansas.

MUSEUM OF COMPARATIVE ZOOLOGY. 183

203. Porzana Carolina. Eaetem Kansas; Great Salt Lake Valley.

204. Fulica americana. Eastern and Middle Kansas; Great Salt

Lake Valley.

ANATIDJE.

205. Anser hyperboreus. Great Salt, Lake (October).

206. Bernicla canadensis. Great Salt Lake Valley.

207. Anas boschas. Eastern Kansas; Great Salt Lake Valley.

208. Dafila acuta. Great Salt Lake Valley (September).

209. Nettion carolinensis. Kansas. Great Salt Lake Valley.

210. Querquedula discors. Eastern Kansas.

211. Querquedula cyanoptera. Great Salt Lake.

212. Spatula clypeata. Great Salt Lake.

213. Chaulelasnius streperus. South Park, Col. Great Salt Lake.

214. Mareca americana. Great Salt Lake.

215. Aix sponsa. Kansas; Great Salt Lake.

21G. Fulix marila. Eastern Kansas (May); Great Salt Lake (Sept.).

217. Aythya ferina, var. americana. (beat Salt Lake.

218. Erismatura rubida. Great Salt Lake (September).

219. Mergus merganser. Great Salt Lake (September); Wahsateh
Mountains; Eort Fred. Steele, Wyoming (October).

220. Lophodytes cucullatus. Fort Fred. Steele, Wyoming (Oct.).

PELECANID^E.

221. Pelecanus erythrorhyncbus. Eastern Kansas. Great Salt Lake.

GRACULIDJE.

222. Graculus dilophus. Great Salt Lake.

LARIDJE.

223. Larus delawarensis. Great Salt Lake.

224. Chraecocephalus philadelpbia. Great Salt Lake (October).

225. Xema Sabini. Great Salt Lake (October).

22G. Hydrocbelidon fissipes. Eastern Kansas (May).

PODICIPID.E2.

227. Podiceps cornutus. Great Salt Lake Valley.

228. Podilymbus podiceps. Great Salt Lake Valley.

Cambridge, July 10, 1872

MUSEUM OF COMPARATIVE ZOOLOGY. 185

No. 7. — Interim Report of the Hydroids collected by L. F. de
Pourtales during the Gulf Stream Exploration of the United
Slates Coast Survey. By George T. Allman.

(Published by permission of Prof. B. Peirce, Supt. U. S. Coast Survey.)

Setting 'aside a very few whose imperfect state of preservation ren-
dered their determination impossible, the whole of the specimens forming
the collection may be distributed among seventy-three species. Of these
species sixty-three are undescribed, two have been already described by
M. de Pourtales, and, like the undescribed species, have not yet been ob-
tained beyond the area of the Exploration ; while the remaining eight, so
far as the diagnosis of specimens in almost every casedestitute of gonosome
can be relied on, occur also on the eastern side of the Atlantic. It
must be borne in mind, however, that the identification of specimens in
which the gonosome is wanting cannot be regarded as absolute.

Eleven species belong to Gjmnoblastic genera. Of these one has
been already described by M. de Pourtales. Of the remaining ten,
seven have their hydranths sufficiently well preserved for description ;
while the remaining three, retaining nothing but the tubular perisarc with
at most the included coenosarc, can be referred only provisionally to
definite species or even to definite genera.

Only one of the Gymnoblastic Hydroids admits of being referred
with any degree of probability to a species occurring on the eastern
shores of the Atlantic. This is a large, simple-stemmed form, which
cannot be distinguished from Tubularia indivisa ; but as nothing re-
mains of it beyond the stems, it would be rash to assert positively that
it is identical with the Tubularia indivisa of the European shores.

Of the Calyptoblastic form-; there are sixty-two species in all. Of
these fifteen belong to the Campanularinaj, while forty-seven must be
referred to the Sertularinaj.

Of the fifteen species referable to the Campanularinse one has been
already described by M. de Pourtales, while- two others cannot be dis-
tinguished from the Filellum immersum and the Halecium muricatum of
the European shores. In the absence of hydranths and gonosome,

186 BULLETIN OF THE

however, it would be unsafe to insist on the identity of these with the
European species.

Among the Sertularina;, nineteen species belong to the Sertularidae
and twenty-eight to the Plumularidae. Of the nineteen species of
Sertularidae -ixteen have not yet been met with beyond the area inves-
tigated by the Explorers, while three are undistinguishable from the
following European species: Sertularella Gayii and its variety, rohusla ;
Sertularella polyzonias ; and Sertularella tricuspidata.

The collection is especially rich in the Plumularida?, no less than
twenty-eight out of the seventy-three species of which the collection is
composed belonging to this beautiful family. Two of these would seem
to be common to both sides of the Atlantic ; for though it is true that no
gonosome is present in either of them, their trophosomes are undistin-
guishable from those of the Antennularia ramosa and Plumvlaria catk-
arina of the European shores.

Among the Gymnoblastea contained in the collections, the gonosome
is present in a large proportion of species. So also a large proportion
of the Plumularida: are provided with their gonosomes and present some
interesting and beautiful modifications of this part of the hydroid colony.
It is remarkable, however, that the gonosome is absent in almost every
instance from the other Calyptoblastie forms. Among the fifteen species
of Campanularinae the gonosome occurs in only one ; while among the
nineteen species of Sertularidae it is present in only two, unless a remark-
able structure, which will be described in the detailed report, is rightly
referable to this part of the colony, and then a third species must be
included among those Sertularidas in which the gonosome is preserved.

The two species of Sertularidae in which the gonosome is undoubt-
edly present are common to both sides of the Atlantic, while not oidy
those, but the single example of the Campanularime, in which it also
occurs, are from some of the deepest dredgings made.

January, 1873,

MUSEUM OF COMPARATIVE ZOOLOGY. 18'

No. 8. — The Echini collected on the Hassler Expedition. By
Alexander Agassiz.

(Published by permission of Prof. Peirce, Supt. of the U. S. Coast

Survey.)

The following preliminary notice of the Sea-urchins collected by Pro-
fessor Agassiz and Mr. Pourtales during the voyage of the Hassler,
contains a description of the most interesting of the new species col-
lected.

The most valuable collection was made off the Barbadoes, at a depth
of one hundred fathoms : a very fine specimen of Asthenosoma hystrix
was dredged at that point, three young specimens of Coclopleurus flori-
danus, — the anal plates of two of the specimens of Coclopleurus were
preserved, (there are four of them, as in the Arbaciadre,) — and a re-
markable Spatangoid ( Paleopneustes cristatus A. Ag.), noticed in one
of the letters of Professor Agassiz to Professor Peirce ; this proves
to be another of these remarkable cretaceous types already noticed in
the Report of the Echini of the Straits of Florida. A very good series
of specimens of Ilemiaster Philippii and of Echinus margaritaceus were
collected on the east coast of Patagonia. In the Straits of Magellan
the common species are Arbacia Dufresnii and Echinus Magellanicus ;
Schizaster Philippii was also found. No Goniocidaris was collected.
Along the west coast of South America no novelties were found, and
nothing of special importance was brought home except a very fine
specimen of Astropyga pulvinata collected at Panama by Lieutenant
Cutts. During the stay at the Galapagos a few species of Echini
were collected, which leave no doubt that the Galapagos form a part
of the Panamic District. The following species were collected there :
Cidaris Thouarsii, Strongylocentrotus gibbosus, Toxopneustes semi-
tuberculatus, Encope micropora, Ilhynehopygus pacificus. None of the
East Indian types often credited to those islands were found at the
Galapagos, namely, no Amblypneustes nor Temnopleurus. But we
must be careful not to judge from negative evidence, as, notwithstanding
the Hassler visited so many parts of the west coast of South America,

188 BULLETIN OF THE

it is quite remarkable that only single specimens of the two most
common species of Echini of that coast, Strongylocentrotus albus and
Arbacia nigra, were brought home.

No species hitherto not noticed from the west coast of Central
America, Mexico, or the Gulf of California, or San Francisco, were
collected.

During the visit made to Juan Fernandez a few Echini were col-
lected ; among them specimens of a small Spatangoid (Nacospatangus
gracilis A. Ac.) allied to Micraster and Spatangus. Small specimens
of Echinus margaritaceus were also collected there ; other specimens
of the same species were obtained off Cape Dos Bahias, on the east
coast of Patagonia. A fine specimen was in the Museum collection
previously, from Cape Horn.

A full description and figures of the most interesting species will be
given in the Zoological Results of the Hassler Expedition.

Paleopneustes cristatus A. Ag. (nov. g-r-n. ctsp.)

Seen in profile this species has a remarkable resemblance to some of the
forms of Ananchytes ovata. Like it, it has neither peripetalous nor anal
or subanal fascioles. The actinal surface is nearly flat, the anterior ex-
tremity rounded, with no trace of indentation for the anterior ambu-
lacrum, which consists, as in Ananchytes, of slightly diverging zones, with a
pair of pores piercing the centre of each ambulacral plate. The abactinal
svstem is, however, compact, as in Spatangus, and the lateral ambulacra
form imperfect petals, diverging, extending half-way from apex to edge of
the test, terminating abruptly, and very slightly depressed below the level
of the test. The vertex corresponds to the apical system, and is nearly
central. The anal system is large, circular, and placed close to the edge
of the test in the truncated posterior extremity of the test. The lateral
petals are continued by distant pairs of pores in the centre of the ambu-
lacral plates to the actinostome. The actinal ambulacra form broad avenues
on each side of the triangular elongate plastron. The actinostome is trans-
verse, narrow, with a very prominent posterior lip. The upper part of the
test is covered by distant tubercles of uniform size, arranged in regular

MUSEUM OF COMPARATIVE ZOOLOGY. 189

horizontal lines; they are more closely packed and larger on the actinal
surface. On the upper part of the test the spines are straight, short,
comparatively stout. They are large, slightly curved at the base, and
spatulate at the actinal sides The apical part of the anterior interambu-
lacrum carries a cluster of longer spines closely packed together, forming a
sort of tuft. The spines of the abactinal part of the test resemble at first
glance much more those of the regular Echini than those of the Spatan-
goids, standing out from the test in all possible directions, and not having
a general direction as is usually the case in Spatangoids. The color when
alive was of a dark violet.

This genus differs from Asterostoma only in the absence of actinal am-
bulacral furrows and in having a labiate actinostome instead of the pen-
tagonal sunken mouth represented in the poorly preserved specimens of
Asterostoma.

100 fathoms off Barbadoes.

Nacospatangus gracilis A. Ac. (now gen. et sp.)

Off' Juan Fernandez in 65 fathoms was dredged a small Spatangoid
(17 m.m. long), -which will form the type of a subgenus of Spatangus inter-
mediate between Maretia and Micraster. It has the high test of Maretia
alta, vertex posterior; in the anterior lateral ambulacral petal the abactinal
part of the anterior poriferous zone is obliterated, the posterior petals are
as in Maretia. The anal extremity of the test resembles that of Spatan-
gus proper, but the whole test is covered, as in Micraster, with uniform tu-
bercles ; the genus resembles those species of Micraster which have a sub-
anal fasciole. In this genus the subanal fasciole is heart-shaped, and
sends off an anal branch. The whole surface of the test is covered by
silvery-gray short curved spines, somewhat longer on the actinal side, espe-
cially towards the posterior extremity of the actinal plastron.

A very fine series of Ilemiaster Philippii A. Ag. (Loven's Abatns
Philippii MS. taken by Kinberg off La Plata). This is extremely inter-
esting, as showing how with increasing age the lateral ambulacra, which
in young stages (18 mm. long) are slightly depressed, gradually become
more and more sunken, til! in specimens measuring 40 mm. in length the
lateral ambulacra, especially (lie anterior, are deeply sunken, much as in
Tripylus. The peripetalous fasciole does not vary greatly in outline or
breadth, and this sjiecies is readily distinguished from II. australis by
the short posterior lateral ambulacra, and the narrow peripetalous fasciole.

190 BULLETIN OF THE

Very small specimens show this species to go through changes similar to
those of Brissopsis. They are quite cylindrical when measuring about 75.
mm. in length.

Lat. 37° 42' S. Long. 56° 20' W. 44 fathoms.

Lat. 51 26 " 68 56 55 fathoms.

Off' Cape Dos Bahias 55 fathoms. E. coast of Patagonia.

Cambridge, January, 1873.

No. 9. — Catalogue of the Terrestrial Air-breathing 3Iolhi8ks of
North America, with Notes on their Geographical Range. By
W. G. BlNNEY.

In connection with ray friend, Mr. Thomas Bland, I have collected
many facts relating to the North American Land Shells since the pub-
lication of our Monograph.* The following pages give a synopsis of
the more important of these, especially such as throw light upon their
classification, synonymy, and geographical distribution.

As regards classification, I have followed the arrangement proposed
by Mr. Bland and myself in the Annals of the Lyceum of Natural
History of New York, Vol. X. p. 158.

As regards synonymy, I have followed our Monograph referred to,
except in a very few cases where more ample opportunities of study
have caused me to reconsider our decisions. Future study will, no
doubt, eliminate as synonymes several species of the present catalogue.
I have added the species described since our Monograph was printed.
Happily but two synonymes have been added to our list since then, —
Vertigo tridentata and Helix ptycophora.

As regards geographical distribution, it must be borne in mind that
the data are very imperfect on which I base my views. Future re-
search will, no doubt, greatly modify them. I have omitted the species
of Lower California as belonging to the fauna of Mexico. San Diego
is the lowest point from which we have truly Californian species.

PULMONATA GEOPHILA.

OLEACINIDiE.

Glandina Vanuxemensis, Lea. Glandina decussata, Desh.
truncata, Gmel. bullata, Gld.

parallela, W. G. B. Texasiana, Pfr.

* Land and Freshwater Shells of North America, Part I. Pulmonata Geophila.
By W. G. Binuey and Thomas Bland, Washington, Smithsonian Institution, 1869.

102

BULLETIN OF THE

HELICIDiE.

(

Macrocyclis Vancouverensls,

sportella, Gld.

concava, Say.

Voyana, Note.

Duranti, Newc.
Zouites kopnodes, W. G. B.
fuliginosus, Griff.
friabilis, W. G. B.
caducus, Pfr.
lasvigatus, Pfr.
demissus, Bum.
ligerus, Say.
intertextus, Binn.
subplanus, Binn.
inornatus, Say.
sculptilis, Bland.
Elliotti, Red/.
cerinoideus, Anth.
cellarius, Mull.
Wkitneyi, Newc.
nitidus, Mull.
arboreus, Say.
viridulus, Mke.
indentatus, Say.
limatulus, Ward.
minusculus, Binn.

Vitrinince.)

Lea. Zorfites milium, Morse.

Binneyanus, Morse.

ferreus, Morse.

conspectus, Bland.

exiguus, Stimpson.

chersinellus, Dull.

capsella, Gld.

fulvus, Drap.

Fabricii, Beck.

Gundlachi, Pfr.
*

gularis, Say.

suppressus, Say.

lasmodon, Phillips.

significans, Bland.

interims, Say.

multidentatus, Binn.
Vitrina limpida, Gould.

Angelicee, Beck.

Pfeifferi, Newc.

exilis, Mor.
Limax maximus, Lin.
flavus, Lin.
agrestis, Mull.
campestris, Binn.
Hewstoni, /. G. Cooper.

(Heliciiuc.)

Arion fuscus, Mull. Patula

Andersoni, J. G. Coop.

foliolatus, Gld.
Ariolimax Columbinaus, Gld.

Californicus, /. G. Coop.

niger, ,/. G. Coop.
Prophysaon Hemphilli, Bl. & Binn.
Binneia notabilis, J. G. Coop.
Hemphillia glandulosa, Bl. & Binn.
Patula solitdria, Say.

strigosa, Gld.
Hemphilli, Newc.
Cooperi, llr. G. B.
Idahoensis, Newc.
Haydeni, Gabb.
alternata, Say.
Cumberlandiana, Lea.
tenuistriata, Binn.
perspectiva, Say.
striatella, Anth.

MUSEUM OF COMPARATIVE ZOOLOGY.

193

Patula pauper, Mor.

Horni, Gabb.

asteriscus, Morse.

incrustata, Pfr.

vortex, Pfr.
Helix lineata, Say.
*
labyrinthica, Say.
Hubbardi, Brown.

*
Yatesii, J. G. Coop.

*
auriculata, Say.
uvulifera, Shuttl.
auriformis, Bid.
Postelliana, Bid.
espiloca, Rav.
avara, Say.
ventrosula, Pfr.
Hindsi, Pfr.
Texasiana, Moricand.
triodontoides, Bid.
Mooreana, W. G. Binn.
tholus, W. G. Binn.
hippocrepis, Pfr.
fastigans, L. W. Say.
Jacks oni, Bid.
Troostiana, Lea.
Hazardi, Bid.
oppilata, Moricand.
Dorfeuilliana, Lea.
Ariadnae, Pfr.
septemvolva, Say.
cereolus, Muhlf.
Carpenteriaua, Bid.
Febigeri, Bid.
pustula, Fer.
pustuloides, Bid.
leporina, Gld.
Harfordiana, /. G. Coop.

Helix poly gyrella, Bid. & J. G. Coop
*
spinosa, Lea.
labrosa, Bid.
Edgariana, Lea.
Edvardsi, Bid.
barbigera, Redf.
stenotrema, Fer.
hirsuta, Say.
maxillata, Gld.
monodon, Rack.

palliata, Say.
obstricta, Say.
appressa, Say.
inflecta, Say.
Rugeli, Shuttl.
tridentata, Say.
Mullani, Bid. & J. G. Cup.
fallax, Say.
introferens, Bid.
Hopetonensis, Shuttl.
vultuosa, Gld.
loricata, Gld.

*
major, Binn.
albolabris, Say.
divesta, Gld.
multilineata Say.
Pennsylvanica, Green.
Mitchelliana, Lea.
elevata, Say.
Clarki, Lea.
Christyi, Bid.
exoleta, Binn.
Wheatleyi, Bid.
dentifera, Binn.
Roemeri, Pfr.
thyroides, Say.
clausa, Say.
Columbiana, L,ea.

VOL. III.

13

194

BULLETIN OF THE

Helix germ ana, Gld.

Downieana, Bid.
jejuna, Say.
devia, Gld.
profunda, Say.
Sayii, Binn.

*
harpa, Say.

pulchella, Mull.

*
hispida, L.
rufescens, Penn.

*
Berlandieriana, Mor.
griseola, Pfr.

*
fidelis, Gray.
infumata, Gld.
Hillebrandi, Newc.

*
arrosa, Gld.
Townsendiana, Lea.
tudiculata, Binn.
Nickliniana, Lea.
Ayresiana, Neicc.
redimita, IK. G. Binn.
intercisa, IF. G. Binn.
exarata, Pfr.
ramentosa, Gld.
Californiensis, Lea.
Carpenteri, Newc.
Mormonum, Pfr.
sequoicola, J. G. Coop.
Diabloensis, J. G. Coop.
Traski, Newc.
Dupetithouarsi, Desk.
ruficincta, Newc.
facta, Newc.
Gabbi, Newc.

*

Helix Newberryana, W. G. Binn.
*
Kelletti, Fbs.
Tryoni, Newc.

*
hortensis, Mull.

*
aspersa, Mull.

*
varians, Mice.
Holospira Roemeri, Pfr.

Goldfussi, Pfr.
Cylindrella Poeyana, Pfr.

jejuna, Gld.
Macroceramus Kieneri, Pfr.

Gossei, /'/'"•
Bulimulus multilineatus, Say.
Dormani, W. G. B.
Marielinus, Pfr.
Floridanus, Pfr.
patriarcha, W. G. B.
alternatus, Say.
Schiedeanus, Pfr.
dealbatus, .Say.
Cionella subcylindrica, L.

acicula, Midi.
Stenogyra decollata, L.
subula, Pfr.
octonoides, Ad.
gracillima, Pfr.
Pupa muscoruni, L.
Blandi, Morse.
Hoppii, Miill.
variolosa, Gld.
pentodon, Say.
decora, Gld.
corpulenta, Morse.
Rowelli, Newc.
Californica, Rowell.
fallax, Say.
niodica, Gld.

MUSEUM OF COMPARATIVE ZOOLOGY.

195

Pupa Arizonensis, Gabb.
hordeacea, Gabb.
armifera, Say.
contracta, Say.
rupicola, Say.
corticaria, Say.
pellucida, Pfr.
borealis, Mor.

Liguus fasciatus, Mull.
Orthalicus zebra, Miill.

Strophia incana, Blnn.
Vertigo Gouldi, JJinn.

Bollesiana, Morse.

milium, Gld.

ovata, Say.

ventricosa, Morse.

simplex, Gld.

(Orthalicina?.)

Orthalicus uudatus, Brug.
Punctum minutissimum, Lea.

(S

Succinea Haydeni, W. G. B.
retusa, Lea.
Sillimani, Bid.
ovalis, Gld., not Say.
Higginsi, Bid.
Haleana, Lea.
Mooresiana, Lea.
Grosvenori, Lea.
Wilsoni, Lea.
Concordialis, Gld.
luteola, Gld.
lineata, W. G. Blnn.
avara, Say.

uccininos.)

Succinea Stretchiana, Bid.
Verrilli, Bid.
aurea, Lea.
Groenlaudica, Beck.
obliqua, Say.
Totteniana, Lea.
campestris, Say.
Hawkinsi, Bd.
rusticana, Gld.
Nuttalliana, Lea.
Oregonensis, Lea.
effusa, Shuttl.
Salleana, Pfr.

PHILOMYCIDiE.

Tebennophorus Caroliniensis, Bosc.
Pallifera dorsalis, Blnn.

VERONICELLIDiE.

Veronicella Floridana, Binn. Veronicella olivacea, Steams.

During the many years in which my attention has been devoted to
the Terrestrial Pulmonata of North America, I have lost no opportu-
nity of adding to the knowledge of their geographical distribution, pre-
sented by my father in the first volume of his work.* The result of

* A. Binney, Terr. Moll. U. S., I. Chaps. V.- IX.

196 BULLETIN OF THE

this accumulation of facts is here briefly offered. It must be studied
in connection with the chapters referred to, and also with the text-book
of our land shells, prepared for the Smithsonian Institution by my
friend, Mr. Bland, and myself.*

The regions west of the Rocky Mountains have become known since
the publication of my father's work, and much additional information
received from the eastern portion of the continent. It becomes neces-
sary, therefore, somewhat to modify the limits of the sections which he
indicated. I have already suggested f that, in regard to the geograph-
ical distribution of the Terrestrial Pulmonata of North America, there
appear to be three distinct faunas, which I have called, —

I. The Pacific Province.

II. The Central Province.

III. The Eastern Province.

The boundaries of these provinces and the subdivisions which appear
to exist in them will be given below, as well as lists of their peculiar
species. It must be distinctly understood, however, that future re-
searches, especially at the South and Southwest, may greatly modify the
views here presented.

I. The Pacific Province comprises a narrow strip between the
Sierra Nevada and Cascade Mountains on the east, and the Pacific
( >cean on the west. Its southern limit is San Diego, from whence it
extends northerly into Alaska.

Over this province the following species range : —

Macrocyclis Vancouverensis. Ariolimax Columbianus.

sportella. Prophysaon Hemphilli.

Helix Columbiana. Succinea rusticana.

germana. Oregonensis.

tudiculata. Nuttalliana.

Over the whole of this province we find also the following species,
common to Eastern North America. They also extend over the whole

northern portion of the continent, where the mountains have ceased to

* Land ami Freshwater Shells of North America, Part I. Pulmonata Geophila.
By W. G. Binney aid Tin. mas Bland, Washington, 1869.
t Proc. Bo>t. Sue. Nat. Hi-:., IX. 177, 1863.

MUSEUM OF COMPARATIVE ZOOLOGY. 197

be barriers to •1'-i~'K'ition. It is, no doubt, from these regions that
they have spread through the Pacific Province, and not westward over
the Rocky Mountains. Had other eastern species extended over the
boreal regions, we should, no doubt, have found tbem also spreading into
the Pacific States. They are especially found along the Sierra Nevada.

Zonites arboreus. Limax campestris.

indentatus. Patula striatella.

minusculus. Helix liueata.

milium. Punctum minutissimum.

In the Pacific Province we also find several species common to the
circumpolar regions of Asia, Europe, and America. They have like-
wise spread southward along the Sierra Nevada and on either side of it.
They have also spread southward over the Central and Eastern Prov-
inces, and now inhabit most, if not all, of North America. They are

Zonites fulvus. Cionella subcylindrica.

Other species will probably be added to this list by further search ;
among them Helix jndchella.

In dealing with the species from the North in Eastern North Amer-
ica (see below, p. 204), the question of their distribution will be more
fully discussed.

In addition to the species already enumerated as common to the whole
Pacific Province, there are many more restricted in their range. It
appears that the Pacific Province is divided into two regions, (a) the
Oregonian and (b) Californian, the two intermingling slightly or over-
lapping in the extreme north of California, near Humboldt Bay. The
faunas of these regions are nearly allied.

(a.) The Oregon Region lies between the Cascade Mountains and
the Pacific Ocean, extending northerly through British Columbia hito
Alaska. The following species are peculiar to it : — *

Helix devia. Arion foliolatus.

fidelis. Hemphillia ^landulosa.

Townsendiana. Succinea Hawkinsi.

There seems to be here some overlapping of the Pacific and Central
Provinces, as Helix Townsendiana, Helix devia, and Macroryclis Van-

* I omit Onchidella borealis, Dull, from Sitka, being doubtful whether the geuus
should be treated as American.

198

BULLETIN OF THE

couverensis extend along the mountains southeasterly into Idaho and
Montana. The former two become much dwarfed in size at their most
eastern range.

(b.) The Californian Region extends from Humboldt Bay to San
Diego, between the Sierra Nevada and Cascade Mountains on the east,
and the Pacific Ocean on the west.

The following are the species peculiar to it : —

Macrocyclis Voyana.
Duranti.
Vitrina Pfeifferi.
Zonites Whitneyi.
conspectus,
chersinellus.
Limax Hewstoni.
Binneia notabilis.
Ariolimax Californicus.

niger.
Arion ? Andersoni.
Helix Yatesii.

Harfordiana.

loricata.

infumata.

Hillebrandi.

arrosa.

Nickliniana.

Ayresiana.

redimita.

intercisa.

Helix exarata.

ramentosa.

Californiensis.

Diabloensis.

Carpenteri.

Mormonum.

sequoicola.

Traski.

Dupetithouarsi.

ruficincta.

Gabbi.

facta.

Kelletti.

Tryoni.

Newberryana.
Pupa corpulenta.

Rowelli.

Californica.
Succinea Sillimani.

Stretchiana.

Of the above, several species extend beyond the limits of the region.
Thus, Vitrina Pfeifferi, Zonites Whitneyi, Pupa corpulenta, Succinea
Sillimani, Succinea Stretchiana, and S. rusticana are found also on
the western slope of the Sierra Nevada in the Central Province. Helix
infumata and Macrocyclis Voyana are also found outside the bounds of
the Region, in the Oregonian Region.

.With the fauna of Lower California there seems no connection, unless
Helix Stearnsiana proves identical with H. Kelletti, in which case the
former must be considered as belonging to the Lower California fauna
rather than to that of our Pacific Province. Another species, H. Car-
penteri, is included in the above list, having been cpioted from San Diego

MUSEUM OF COMPARATIVE ZOOLOGY. 199

and Tulare Valley, California. It may, however, belong rather to the
Lower California fauna,* having been described from that region under
the name of H. Remondi, and from Guaymas. Veronicella olivacea,
Stearns, a Nicaraguan species, is also said to extend into California.

From the list of California species are omitted Columna Californica,
actually collected at Marmato, New Granada, by Mr. Bland, and

* The peninsula of Lower California forms a distinct molluscous province of it-
self, extending nearly to San Diego. The following species are peculiar to it : —

Ccelocentrum irregulare, Gabb. Bulimulus pallidior, Sowerby.
Helix Stearnsiana, Newc. (Kelktti, excelsus, Gould.

Fbs. ?) inscendens, W. G. Binn.

areolata, Sowb.( Veitchii, Tryon sufflatus, Gould.

Pandorae, Forbes. pilula, W. G. Binn.

levis, Pfr. proteus, Brod.

Rowelli, Newc. (Lohri, Gabb.) Xantusi, W. G. Binn.

Berendtia Taylori, Pfr. artemisia, W. G. Binn.

Bulimus spirifer, Gabb. Onchidium Carpenteri, \V. G. Binn.
Gabbi, Crosse.

Veronicella olivacea, Stearns, a Nicaraguan species, is also found in Lower Califor-
nia. Of the above list one only has been found near San Diego, //. Stearnsiana.
Another, //. Roivelli, has been referred to Arizona, but with doubtful accuracy. //.
Pandorce and areolata have also erroneously been referred to California. Helix Re-
mondi (Carpenteri) is omitted from the list, as it also occurs in the California
Region. It is the only species common to the peninsula and mainland of Mexico.
The most interesting fact in the fauna of Lower California is trft presence of Bulimulus
proteus and B. pallidior, — species described originally from South America, the former
from Chili. Such facts can only be accounted for by a theory of former connection
of the two points.

Though still more remotely connected with the subject of this paper, it will be
interesting to add here a list of species found at and north of Mazatian, on the
Pacific coast of Mexico.

Glandina turris, Pfr. Helix acutedentata, IP. G. Binn.

Aibersi, Pfr. ventrosula, Pfr.

Holospira Remondi, Gabb. Bulimulus Ziegleri, Pfr.
Helix Mazatlanica, Pfr. Californicus, Rce. ?

Carpenteri, Newc. Orthalicus uudatus, Bnuj.

anilis, Gabb. Pupa chordata, Pfr.

Behri, Gabb. Succinea cingulata, Forbes.

Of the above, II. Mazatlanica has lately been quoted from San Francisco, con-
founded probably with some allied species.

//. Mormonum is omitted from this list, its presence in Sonora not having been
confirmed, although asserted, doubtfully, by Messrs. Fischer and Crosse.

200

BULLETIN OF THE

Zonites cultellatus, probably an accidentally introduced European shell.
BuKmns Californicus is also omitted, belonging, no doubt, to the region
of Mazatlan. Also Glandina Albersi, which we know to live in the
Sierra Madre.

Separate lists of species peculiar to the several regions of the Pacific
Province are given above. There now follows a complete list of all
the species hitherto observed in the entire Province.

Macrocyclis Vancouverensis.
sportella.
Voyana.
Duranti.
Zonites Whitneyi.
arboreus.
indentatus.
minusculus.
milium,
conspectus,
chersinellus.
fulvus.
Vitrina Pfeifferi.
Limax campestris.
Hewstoni.
Prophysaon Hemphilli.
Ariolimax ColunAianus.
Californicus.
niger.
Arion? follolatus.

Andersoni.
Binneia notabilis.
Hemphillia glandulosa.
Patula striatella.
Helix lineata.

Yatesii, /. G. C, not Pfr.

Harfordiana.

loricata.

Columbiana.

germana.

devia.

fidelis.

infumata.

Hillebrandi.

Helix arrosa.

Townsendiana.
tudiculata.
Nickliuiana.
Ayresiana.
redimita.
intercisa.
exarata.
ramentosa.
Californiensis
Carpenteri.
Mormonum.
sequoicola.
Diabloensis.
Traski.

Dupetithouarsi
ruficincta.
facta.
Gabbi.
Kelletti.
Tryoni.
Newberry ana.
Cionella subcylindrica.
Pupa Rowelli.

Californica.
corpulenta.
Succinea Sillimani.

Stretchiana.

Hawkinsi.

rusticana.

Nuttalliana.

Oregoneusis.
Punctum minutissimum.
Veronicella olivacea.

MUSEUM OF COMPARATIVE ZOOLOGY. 201

Several of the above will eventually prove to be synonymes, but the
total number of species is small in comparison with the great size*
of the Pacific Province. An equal extent of territory in the Mis-
sissippi Valley, or even on the Atlantic coast, would show a larger
number ; and the comparatively small regions of Texas, Florida, and
the Cumberland Mountains would each show an equal number of
species peculiar to itself, independent of what they have in common
with the rest of Eastern North America. This disparity in number is
still more plainly shown in the separate region of Oregon. Thus it ap-
pears that the Pacific Province is not rich in the number of its species,
but it is peculiarly favored in their size and beauty, — in this respect
strikingly in contrast with the Central Province and Eastern Province.

From the Central Province the Pacific Province is quite distinct. A
few species have been shown above to inhabit both slopes of the Sierra
Nevada, and a few of the Oregon species have passed the barrier of the
Cascade Mountains on the north, but the peculiar Pacific forms, such
as Arlonta and Aglaia, are unknown in the Central Province. On the
other hand, the only form which has any development in the Central
Province, Palula, is scarcely known in tile Pacific Province.

Compared with Eastern North America, or the Eastern Province, as
it is designated below, the Pacific Province is remarkable for the absence
of all the larger Zonites. The presence of the smaller species, also, may
perhaps be accounted for by migration from the north, so that the genus
Zonites cannot be considered as characteristic of the Province. The
genus Pupa is less common. The genera Tebennophorus and Pallifera,
so universally distributed in Eastern North America, are unknown, and so
are the southern genera Glandina, and Bulimulus. On the other hand,
we find the genus Macrocyclis much more developed, and meet several
genera unknown in the Eastern Province, such as Ariolimax, Binneia,
Prophysaon, and Hemphillia. The genus Helix is proportionally more
developed in the Pacific Region, and is represented by quite dissimilar
subgenera. The sections so peculiar to the Eastern Province, Polygyra,
Stenotrema, Triodopsis, Mesodon, are scarcely represented. In their
place we find Aglaia and Arionta, forms unknown in the Eastern Prov-
ince. The latter, though feebly represented in Europe, is the form of
Helix characteristic of California. It is prolific of species and also of
varieties to a degree which has caused some confusion in the synonymy.
The section to which Helix Newberryana belongs, Glyptostoma, is also
peculiar to California.

202 BULLETIN OF THE

From Lower California and Mexico the Pacific Region has been
•shown to be equally distinct, wanting entirely the Holospira, Glandina,
Bulimulus, and Zonites of those regions.

Failing on the north, east, and south, the west alone is left fo us from
whence to trace the pulmonate fauna of the Pacific Region, and here
the secret of its origin lies buried under the Pacific Ocean.*

II. The Central Province extends from Mexico to the British
Possessions, between the Rocky Mountains on the east, and the Sierra
Nevada and Cascade Mountains on the west.

The following are the species peculiar to the province: —

Patula strigosa. Patula Horni.

Cooperi. Helix polygyrella.
Haydeni. 'Mullani.

Idahoensis. Pupa Arizonensis.
Hemphilli. hordeacea.

The first two of these species, perhaps identical, are also found on
the eastern slope of the Rocky Mountains, in Wyoming and Dakota.

To the above must be added, as inhabiting the province, but not pe-
culiar-to it, the following species from the Pacific Province, inhabiting
either slope of the Sierra Nevada : Vitrina Pfeifferi, Zonites Whit-
•neyi, Pvpa corpidenta, Succinea SiUimani, and Succinea Stretchiana.
The following, also, from the Oregonian Region of the Pacific Province,
Helix devia, Helix Toivnsendia?ia, and Macrocyclis Vancouverensis, are
found at its most northern point, though the former two species are re-
duced in size. We find, also, over the Central Province the following
species, whose derivation can readily be traced to the north ; f Zonites

* A subsidence of eight hundred feet in tire continent of North America would
leave on its eastern shore a strip of land of about equal size of our Pacific Region,
equally distinct in its terrestrial mollusca from the balance of the continent. In
this case, however, we should have a distant island of the Appalachian chain on
which we should find all the species of the eastern coast of the mainland. This
would give us a proof of what we can now only suspect as regards the Pacific Prov-
ince,— of former more wide distribution of its pulmonate fauna. From wherever
the fauna may have originated, we can easily explain its present condition. The
physical and climatic features of the Pacific Region are such as readily to account
for its richness in terrestrial mollusks in comparison with the less favored Central
Province, and even with the Eastern Province.

t See remarks on the distribution of these species over Eastern North America,
p. 204.

MUSEUM OF COMPARATIVE ZOOLOGY. 203

minusculus, fulrus, indebitatus, Helix pulchella, H Hneata, Patula
striatella, Cionella subcyHndrica.

Helix Rowelli, a Lower California species, is omitted from the list, its
presence in Arizona not being well authenticated.

The fauna of the Central Province is quite distinct from that of the
Pacific Province, but is nearly allied to that of the Eastern Province,
its genera and subgenera being the same, excepting Polyyyrella, its only
peculiar subgenus of Helix. It may therefore be of the same origin
as the fauna of the Eastern Province.

The paucity of species over this large province is owing to the nature
of its climate and soil, — causes in equal force on the western border
of the Eastern Province.

In order to avoid mistakes in the study of the geographical distribu-
tion of North American Land Shells one must constantly bear in mind
the changes in the names and boundaries of the trans-Mississippi States
and Territories.*

III. The Eastern Province comprises the remaining portions of
the continent north of Mexico. The species by which it is inhabited
have been derived partly from the north, partly from the interior, and
partly from the south. It may, therefore, be divided into ;he (a)
Northern Region, (b) the Interior Region, and (c) the Southern Region.

(a.) The Northern Region f .comprises the whole northern portion
of the continent, including Greenland and Alaska. Its southern
boundary is not perfectly known, and probably not exactly marked ; it
may, however, be indicated in general terms as the same with the polit-
ical division between the British Possessions and the United States to
the northeast corner of New York, where it runs southwesterly along
the Appalachian chain of mountains to Chesapeake Bay, thus including
all New England, and the portions of New York, New Jersey, Penn-
sylvania, and Maryland lying east of those mountains. Into this

* Thus, Helix Mullani was described in Land and Freshwater Shells of North
America, I. 131, from points in Washington Territory and Oregon. Both localities
are now in Idaho.

t F'or a description of this Region, see A. Binncy, 1. c. pp. 124, 125, under sections
5 and 6. The American land shells, especially those of the Interior Region, are
forest species ; they become rare towards the northern region of the continent as the
deciduous trees become rare.

204 BULLETIN OF THE

southern extension of the Region we find the Interior Region overlap-
ping, as will be shown below while treating of the Interior fauna. At
other points in the Region, also, have been found species from the In-
terior Region,* especially small Zonites, which are able to bear the
severe climate of the north.

The following are the species of the Northern Region : —

Vitrina limpida. Helix pulchella.

Angelicae. Cionella subcylindrica.

exilis. Pupa muscorum.
Zonites fulvus. Blandi.

nitidus. Hoppii.

viridulus. decora.

Fabricii. borealis.

milium. Vertigo Gouldi.

Binneyanus. Bollesiana.

ferreus. simplex.

exiguus. Punctum minutissimum.

multidentatus. Succinea Haydeni.
Patula striatella. Verrilli.

asteriscus. Higginsi.

pauper. Groenlandica.

Helix harpa. Totteniana.

Of the above, several are circumpolar species, common to the three
continents of Europe, Asia, and America. There being no mountain-
barriers in these regions, they are not restricted in their range across
America. In their progress southward, also, they have met with no
transverse mountain-barriers, but have spread equally on the east and
west of the Rocky Mountains and Sierra Nevada. Hence we find them
common to the whole of North America-t Such are : —

* Sec Proc. Phila. Acad., N. S., 1861, p. 330, for the northern range of species
from the Interior Region.

t In the same way we can account for the distribution of the small eastern species
over the Central and Pacific Provinces. They have not crossed the mountain-bar-
riers, but spread southward from their wider range in the north. Such are: —

Zonites arboreus. Limax campestris.

indentatus. Patula striatella.

minusculus. Helix lineata.

milium. Punctum minutissimum.

These northern species, both indigenous and circumpolar, may have been assisted

MUSEUM OF COMPARATIVE ZOOLOGY. 205

Zonites viridulus. Helix pulchella.

fulvus. Ciouella subcylindrica.

nitidus. Pupa muscorum.
Helix harpa.

This list will be increased should it be proved that Mr. Gwyn Jef-
freys * is correct in referring the following American species to those of
Europe. Vitrina linipida = V. pellucida, Punctum minutissimum =
Helix pygmasa, Drap., Limax campestris = L. Ia3vis, Mull., Vertigo
Gouldii = V. alpestris, Aid., Vertigo Bollesiana = V. pygmrea, Drap.,
V. ovata = V. antivergo, Drap., V. ventricosa = V. Moulinsiana, V.
simplex = V. edentula, Drap., Succinea ovalis = S. elegans, Risso, S.
Totteniana = S. putris, Drap. var.

From Asia have come into Alaska the following : Vitrina exilis,
Patula pauper, Papa borealis.

The species peculiar to Greenland are Vitrina Angelica, Zonites
Fabricii, Pupa Hoppii, and Succinea Groenlandica. Of these, Pupa
Hoppii has, however, also been found on Anticosti Island.

Into this Northern Region have also been introduced by commerce
from Europe the following : Zonites cellarius, at most of, if not at all
of, the ports from New York to Halifax; Limax flavus, L. agi'estis, and
Arion fiiscus, which follow man over the whole United States, living
around his habitations; and L. maximus, also around human habitations,
but noticed only in Newport, R. L, New York City, and Philadelphia ;
Helix hispida at Halifax, Helix rufescens at Quebec, Helix hortensis on
the islands off the coast of New England and the British Provinces,
and on the mainland in Canada and Greenland.

Of the species referred above to the Northern Region, several have
spread beyond its limits. Vitrina limpida has been found in Central
New York ; Zonites viridulus extends to Mexico ; Z. milium to San
Francisco ; Z. fulvus and Helix pulchella all over the United States ;
Zonites nitidus, Z. multidentatus, and Punctum minutissimum to Ohio,
the last to Texas and to California ; Cionella subcylindrica to the States
south of the Great Lakes and into California and New Mexico ; Patula
striatella to Virginia, as well as into Oregon and Nevada.

in their migration southward by glacial agencies. There is a wide field for specu-
lation here.

* Ann. and Mag. N. H., 1872, 245, 246.

206 BULLETIN OF THE

The Northern Region does not differ in the characteristics of its fauna
from that lying south of it, but its climate is too severe for any but the
more hardy forms. Thus, we find only the small Zonites and Helix,
with the genus Vitrina. Compared with the balance of North America,
the Region is peculiar for the great distribution of its species east and
west, owing to the mountain-ranges having here lost the great elevation
which they have farther south, and thus ceasing to be barriers to distri-
bution. The Region is also interesting as being the source from whence
have spread southward over the whole continent seyeral small species
now found in Florida and Texas, and even in Mexico and the West
Indies.

(b.) The Interior Region lies to the south of the Northern Region,
but extends only as far as the Rocky Mountains * on the west. South-
erly it extends to the alluvial regions of the Atlantic and Gulf coasts,
the dividing line here not being sharply defined.

This is the only portion of the continent where we have evidence of
the origin of our land mollusks in former geological times. In the
Post-pleiocene deposits along the Ohio and Mississippi Rivers are found
immense beds of shells, " proving that our existing species were living at
a period which, though recent in a geological sense, was anterior to the
last geological revolution, when the surface of this portion of the earth
was brought to its present condition, and to the existence of the higher
order of animals which now inhabit it, and even to that of the extinct
mammalians which are known only by their gigantic remains." t

From the evidence gathered from these deposits, it appears that the
fauna of this Region can be traced to Indiana and Ohio. From this
centre the species have extended over the Region ; some of them also
have passed the barrier of the Appalachian chain into the Northern
Region, and some have spread, with the enlargement of the continent,
into the Southern Region.

The following species have actually been found fossil in the Post-
pleiocene deposits: — f

Zonites arbor eus. Zonites intertextus.

fuliginosus. ligerus.

inornatus. gularis.

* This is the extreme limit, but before reaching it the land shells have become
very rare, owing to the nature of the soil. For a description see A. Binney, 1. c.
t A. Binney, Terr. Moll. U. S., I. 185.

MUSEUM OF COMPARATIVE ZOOLOGY.

207

Macrocyclis concava.
Patula solitaria.

alternata.

perspectiva.
Helix lineata.

labyrinthica.

auriformis.

stenottema.

hirsuta.

monodon.

palliata.

obstricta.

appressa.

Helix inflecta.

albolabris.

elevata.

exoleta.

tkyroides.

clausa.

profunda.
Pupa armifera.

contracta.
Succinea obliqua.
Helicina* orbiculata.
occulta.

Of the above all are now living and are equally numerous, excepting
Helicina occulta, a species most abundant in Post-pleiocene days, but
now almost extinet.f The other species of Helicina is now confined
to more southern limits.

In addition to the above, the following species, now living in the In-
terior Province, probably had their origin in Post-pleiocene times, and
will, no doubt, be found fossil in the " bluffs " : —

Zonites friabilis.

leevigatus.

suppressus.

indentatus.

internus.

minusculus.

limatulus.
Helix Dorfeuilliana.
leporina.
multilineata.
Pennsylvanica.
Mitchelliana.
dentifera.

Helix bucculenta.
Sayii.
tridentata.
fall ax.
Pupa pentodon.
fallax.
rupicola.
corticaria.
Vertigo milium.

ovata.
Succinea avara.
ovalis.

* Though not Pulmonata, these two species are strictly terrestrial in their habits,
and arc here introduced from their value on the question of the permanence of the
Post-pleiocene species. One of them is almost extinct, the other more restricted in
its range at present.

t Sec A. Binney, Terr. Moll. U. S., I. 183, 184; Bland and Binuey, Ann. Lye.
N. II. of N. Y., IX. 289.

208 BULLETIN OF THE

Tebennophorus Caroliniensis, Pallifera dorsalis, and Limax campestris
probably have also come down from Post-pleiocene times. From their
nature they could leave no record of their presence in the " bluffs."

There are also found in the Interior Region several forms of Succinea
which have been described as

Succinea retusa. Succinea aurea.

Grosvenori. Mooresiana.

lineata.

The following is a complete list of those species of Jhe Interior Region
which have, spread beyond it by passing the barriers of the Appalachian
chain, and are now found over New England and the whole southern ex-
tension of the Northern Region, described on p. 203, as well as over the
whole Southern Region. They may, therefore, be said to inhabit all
of the Eastern Province.

Macrocyclis concava. Helix fallax.
Zonites fuliginosus. albolabris.

inomatus. thyroides.

suppressus. Pupa pentodon.
indentatus. fallax.

arboreus. armifera.

minusculus. contracta.

Limax campestris. rupicola.

Patula alternata. corticaria.

Helix lineata. Vertigo milium,
labyrinthica. ovata.

hirsuta. Succinea avara.
monodon. obliqua.

palliata. Tebennophorus Caroliniensis.

tridentata. Pallifera dorsalis.

Helix Sayii and Helix dentifera have spread into New England only
from the Interior Region. They have not been found in more southern
latitudes on the Atlantic slopes of the Appalachian chain, nor in the
Southern Region.

The geographical range of these species is very great, forming one
of the most striking features of the North American fauna. Still more
widely distributed are those minute species which have been mentioned
above as spreading southwardly from the Northern Region equally on
both sides of the Sierra Nevada and Rocky Mountains. These species

MUSEUM OF COMPARATIVE ZOOLOGY. 209

may be said to inhabit the whole continent of North America as far
south as Mexico. The range of some is still greater. Thus, Zonites
minusculus has been found from British Columbia to Labrador on the
north to Yucatan and Florida on the south, and still farther in Cuba,
Jamaica, Porto Rico, and Bermuda. .Helix labyrinthica also is found
over all Eastern North America, and perhaps in Mexico (as H. Strebeli,
see Fischer and Crosse, Moll. Mex. et Guat., 267). It is also by some
considered identical with an Eocene fossil of France and England.
(See Land and Fr,esh\v. Shells N. A., I. 84.) Zonites arboreus ranges
from Labrador to New Mexico, and in Nevada and California, and
from British Columbia to Florida, Cuba, and Guadaloupe. Vertigo
ovata is found from Maine to Mexico and in Cuba.

The character of the soil and climate, with, perhaps, the gradual ele-
vation, is such as to render the land shells rare, if not quite extinct,
before the Rocky Mountains are reached, the western boundary of the
Interior Region. But one species, Patula solitaria, seems to have
passed this mountain-barrier into the Central Province. This is found
with P. Cooperi in Montana and Idaho, very difficult to distinguish
from forms of the last species. It is, however, oyiparous (from Salmon
River, Idaho),, while P. strigosa, Cooperi, Hemphilli, and Idahoensis
are viviparous. It has been suggested by Dr. H. Dohrn that this
characteristic is connected with the fact of the great dryness of the
soil in the Central Province. The young shell is ready to protect itself
from the moment of its birth, while, if deposited as an egg by the parent,
it might perish from drought.

The following list contains the names of all the species inhabiting the
Interior Region, including those which have spread into it from the
Northern Region : —

Macrocyclis concava. Zonites indentatus.

Zonites fuliginosus. limatulus.

friabilis. minuscules.

laevigatus. fulvus.

ligerus. gularis.

intertextus. suppressus.

inornatus. internus.

nitidus. Limax campestris.

arboreus. Patula solitaria.

viridulus. alternata.

VOL. III. 14

210

BULLETIN OF THE

Patula perspectiva.

striatella.
Helix lineata.

labyrinthica.

Dorfeuilliana.

leporina.

auriformis.

stenotrema.

hirsuta.

monodon.

palliata.

obstricta,

appressa.

inflecta.

trident ata.

fallax.

albolabria.

multilineata.

Pennsylvanica.

Mitchelliaua.

elevata.

exoleta.

dentifera.

thyroides.

clausa.

Helix profunda.
Sayii.
harpa.
pulchella.
Pupa muscorum.
pentodon.
fallax.
armifera.
contracta.
rupicola.
corticaria.
Vertigo milium.

ovata.
Succinea retusa.

Grosvenori.
Mooresiana.
ovalis.
lineata.
avara.
aurea.
obliqua.
Totteniana.
Tebennophorus Caroliniensis.
Pallifera dorsalis.

The above list shows the Interior Region to be remarkable for the
development of the section of Zonites familiar by the European Z. oli-
vetorum (Mesomplux of Alb. ed. 2). In the genus Helix the section or
subgenus Mesodon is most developed. This is almost exclusively a
North American subgenus, as is also Triodopsis, which is also greatly
developed in the Interior Region.

In addition to the species included in the above list as inhabiting all
of the Interior Region, there is a large group of species found within
its limits, but having a more restricted range. They are found in what
may be called the Cumberland Sub-Region. This is comprised in the
southern portion of the Appalachian chain, situated in Eastern Tennes-
see and the adjoining counties of North Carolina, with an offshoot into
the mountains of West Virginia.*

* For a description of its physical ami climatic characters, see Dr. A. Binney,
Terr. Moll. U. S., I. 122. It is there designated as the Southern Interior Section,
and is given a wider western range.

MUSEUM OF COMPARATIVE ZOOLOGY. 211

Tlie following species are peculiar to this Sub-Region : —

Zonites kopnodes. Helix spinosa.

subplanus. labrosa.

sculptilis. Edgariana.

Elliotti. Edvardsi.

demissus. barbigera.

capsella. maxillata.

lasmodon. Rugeli.

Patula Cumberlandiana. introferens.

tenuistriata. ? Clarki.

Helix fastigans. Christyi.

Troostiana. Wheatleyi.

Hazardi. Downieana.

Of these, several have spread beyond the limits given above for
the Sub-Region. Thus, Zonites lasmodon and Helix spinosa have been
found in Northern Alabama. Helix Hazardi has also spread into North-
ern Alabama, and equally into Georgia and Kentucky. Helix labrosa.
and Helix Edgariana in Alabama, and in one case have been collected in
Arkansas. Helix barbigera, Helix maxillata, and Zonites kopnodes
have found their way into Alabama and Georgia; Helix Clarki into
Georgia. Zonites subplanus has been found even in Pennsylvania, having,
no doubt, crept along the mountain chain ; but no other of the species of
the Cumberland Sub- Region has been found as far north, excepting Z.
demissus. This last-named species is found in a highly developed state
in Eastern Tennessee, and has extended into Western Pennsylvania,
North Carolina, Georgia, Alabama (near Mobile), and Arkansas in a
much dwarfed condition.

If to the twenty-four species catalogued above as peculiar to the Sub-
Region are added the sixty-six species which inhabit it as a portion of
the Interior Region (see p. 209), it will be seen that in the Cumberland
Sub-Region we find the largest number of species of any other portion of
North America. The Sub-Region is equally prolific in individuals, and
the individuals are highly developed. These facts are partially explained
by the nature of the country. Low mountains, thickly shaded, well
watered, and with a genial climate and proper soil, offer in their thickets
and ravines innumerable safe breeding-grounds for the land shells.*

■© o*

* Sec A. Binney, Terr. Moll. U. S , I. pp. 122, 123. Being less adapted for culti-
vation than the balance of Eastern North America, we may hope for the preservation

212 BULLETIN OF THE

There seems also to be in this Sub-Region conditions peculiarly condu-
cive to testaceous variation. Six (or twenty-five per cent) of its pe-
culiar species are carinated, and here also the following species of the
Interior Region show the same tendency to carination, — Zonites Ugerus,
intertextus, Patula alternata, Helix appressa and palliata. Here, also,
we first notice the variation of Patula alternata towards heavy ribs upon
its shell ; which is still more apparent as the species extends towards
the southwest.*

The Cumberland Sub-Region is peculiar for the development of
Zonites, and in the genus Helix for the development of the section or
subgenus Stenotrema, almost peculiar to these narrow limits.

(c.) The Southern Region comprises the peninsula of Florida, with
the adjacent islands, together with the alluvial regions of the Atlantic
and Gulf coasts. It includes, therefore, the eastern portion of North
Carolina, South Carolina, Georgia, all of Florida, the southern part of
Alabama, Mississippi, Louisiana, extending into Texas.f Its boundaries,
however, are but imperfectly known, and probably not accurately de-
fined. Many of the species from the Interior Region and Cumberland
Sub-Region have spread into its northern portion, and the following
have extended over the larger portion of it : — -

Macrocyclis concava. Helix fallax.
Zonites fuliginosus. albolabris.

inornatus. thyroides.

suppressus. Pupa pentodon.
indentatus. fallax.

arboreus. armifera.

minusculus. contracta.

Limax campestris. rupicola.

Patula alternata. corticaria.

Helix lineata. Vertigo milium,
labyrinthica. ovata.

hirsuta. Succinea avara.
monodon. obliqua.

palliata. Tebennophorus Caroliniensis.

tridentata. Pallifera dorsalis.

of our land shells in this Region, while they decrease rapidly before the advance of
•civilization elsewhere. See Ibid., pp. 132, 133.

* This heavily ribbed form was common in Post-pleiocene days.

t Sec A. Binney, Terr. Moll. U. S., I. 120, for a description of the Region.

MUSEUM OF COMPARATIVE ZOOLOGY. 213

Equally wide over the Region has been the distribution of those
minute species whose origin has been traced to circumpolar regions
(see p. 205). Such are : Zonites viridulus, /ulcus, and Helix pul-
chella.

In addition to these species derived from the north, are found the
following species peculiar to the Region, whose origin can be traced to
the south, in* the peninsula of Florida, from whence, indeed, many of
them have not yet spread over the whole Region : —

Glandina truncata. Helix Hopetonensis.
Zonites cerinoideus. major.

Helix auric ulata. jejuna.

uvulifera. Bulimulus Floridanus.

Postelliana. Dormani.

espiloca. dealbatus.

avara. Cylindrella jejuna.

cereolus. Pupa variolosa.

septemvolva. modica.

Carpenteriana. Succinea effusa.

Febigeri. campestris.

pustula. Wilsoni.

pustuloides. Veronicella Floridana.

Of the more widely spread species, Helix septc7nvolva is represented
by various forms over the whole southern littoral region, both of the At-
lantic and Gulf. So is Glandina truncata, Helix jejuna, pustula, pustu-
loides, and Pupa modica. Helix Hopetonensis and major extend only
along the Atlantic alluvial Region. Bulimulus dealbatus is also dis-
tributed over the whole Region, from North Carolina to Texas, and has
spread northward to Arkansas and Kentucky. Succinea camjiestris ex-
tends along the Atlantic coast as far as South Carolina, as does also
Zonites cerinoideus, even into Nortli Carolina. Helix espiloca and Pos-
telliana have been noticed thus far in the southeastern corner of Georgia.
The former also at New Orleans and Indianola. Succinea Wilsoni,
at Darien, Ga.

The following European species have been introduced by commerce
into this Region, and still exist at the points named : Slenogyra decollata,
Lin., and Helix aspersa, Miill., at Charleston, S. C. ; Acicula acicida,
Mull , Florida.

From the list of species peculiar to the Southern Region it will be

214 BULLETIN OF THE

seen that the prevailing form is Polygyra, a group or subgenus pecu-
liarly American, represented in the Interior Region indeed, but meet-
ing its greatest development here. The presence of Glandina and
Veronicella shows, also, the more southern character of land-shell fauna.
But the Region, and especially that portion of it from whence the fauna
was distributed, i. e. the southern extremity of Florida, is still more
peculiar in showing the connection between the land shells of the con-
tinent of North America and those of the West India Islands and the
Spanish Main. Of the species given above (p. 213), CylindrcUa jejuna
was, perhaps, introduced from Cuba, and Bulimulus Dormani may prove
identical with B. maculatus, Lea, of Carthagena. The following species
have evidently been introduced* from the West India fauna: — f

Zonites Gundlachi, Cuba, etc. Bulimulus Marielinus, Cuba.

Patula vortex, Cuba, etc. Strophia incana, Cuba.

Helix varians, New Providence. Stenogyra subula, Cuba, etc.
Cylindrella Poeyana, Cuba. gracillima, Cuba, etc.

Macroceramus Kieneri, Cuba. Liguus fasciatus, Cuba.

Gossei, Cuba. Orthalicus undatus, Cuba.

From Yucatan one species has been introduced, Helix oppilata.
Orthalicus zebra, found in several of the Florida keys, is, no doubt, of
foreign origin, though from what point introduced it is difficult to say.
It has been found in Mexico in the Sierra Madre, and in Maranhon.
Bulimulus multilineatus was introduced \ from the continent of South
America, where it has been found at St. Martha, N. Granada, and at
Maracaibo and Pto. Cabello in Venezuela.

Florida has not only received several of its species from the West
Indies, but also from its southern extremity it has contributed in return
to the fauna of those islands. From hence, no doubt, Zonites arboreus
has passed into Cuba and Guadaloupe ; Zonites minusculus to Cuba,
Jamaica, Porto Rico, (Bermuda?) Pupa fallax to Cuba; Vertigo ovata
to Cuba ; Zonites indentatns to San Domingo?

* Either by oceanic currents since the formation of the peninsula of Florida, or
else from some island of the West India group, now enclosed in the peninsula. It
is interesting in this connection to refer to the discovery, by Mr. Conrad, of a Ter-
tiary fossil at Tampa Bay, Bulimus Floridanus, Cojir. See also below, p. 217.

t Also several non-pulmonate species, as Helicina subglobulosa, Cuba ; Ctevopoma
rugulosum, Cuba ; Chondropoma dentatum, Cuba.

t See note t to p. 21G.

MUSEUM OF COMPARATIVE ZOOLOGY.

215

From the various sources indicated above the southern extremity of
Florida has become inhabited by about seventy species of land shells, a
number small in comparison with those found in the Cumberland Sub-
Region (see p. 211), but large when compared with those found in the
great Interior Region.

In addition to those species apparently originating in the peninsul;
of Florida and thence spreading over the whole Southern Region, there
is found within its limits a number of species confined to its southwest-
ern portion. These seem restricted to the southern part of Texas, which
may be considered an offshoot of the Mexican fauna as shown by the
presence of the genera characteristic of that country, such as Holospira,
Bulimulus, and Glandina. "Within the region, however, are many
species peculiar to it, but belonging to the subgenera characteristic of
North America, such as Polygyra and Mesodon. It seems, therefore,
best to consider Texas as belonging equally to the fauna of North America
and of Mexico, being the point where the two overlap. As the limits
of the region are ill defined, several species extralimital to the State of
Texas are included in the following catalogue of the Texan Region: —

Glandina Vanuxemensis.
decussata.
parallela.
bullata.
Texasiana.
Zonites significans.

caducus.
Fatula incrustata.
Helix Hubbardi.
ventrosula.
Hindsi.
Texasiana.
triodontoides.
Mooreana.
tholus.
hippocrepis.
Jacksoni.
Ariadne.

Helix vultuosa.
divesta.
Roemeri.
Berlandieriana.
griseola.
Bulimulus patriarcha.
alternatus.
Schiedeanus.
Macroceramus Gossei.
Holospira Goldfussi.
Roemeri.
Stenogyra octonoides.
Pupa pellucida.
Succinea Haleana.

concordialis.

luteola.

Salleana.

Of the above Helix Jacksoni and Zonites significans are included
with great hesitation. They are found at Fort Gibson, in Indian Ter-

216 BULLETIN OF THE

ritory.* They are more related to the fauna of the Cumberland Sub-
Region than that of Texas.

Besides the species characteristic of the North American fauna which
Texas has as a portion of the Southern Region of the great Eastern
Province, we find in the' above list two species peculiar to it of the
characteristic American subgenus Mesodon, Helix lioemeri and JL,
divesta.'f

Several species on the list have been introduced from other regions,!
such as Helix Hubbardi,% a Jamaica species, as well as Macroceramus
Gossei, a Cuban species, which is also found on the Florida Keys. Patula
incrustata from Cuba, as well as Pupa pellucida and Stenogyra octo-
noides.

Of the remaining species on the list, sixteen have actually been found
in Mexico ; probably all will be, as there seems no well-defined boundary
here between the North American and Mexican fauna.

Bulimidus serperastrus, Say, although actually found in Texas, is evi-
dently a member of the Mexican fauna, and is therefore omitted from
my list.

The characteristic of Texas appears to be the great preponderance of
the subgenus Polygyra, of the type of H. Texasiana, while the type of
Florida, the septemvolva, is quite wanting. The great abundance of indi-
viduals is also remarkable, showing the Region to be peculiarly adapted
to pulmonate life. In the number of its species, also, the Texas Region
is favored ; by adding to the above list of peculiar species those which
it has in common with all of the Eastern Province, and also those of the
Southern Region, we find a total of seventy species, the same number as
found in Florida.

* A. Binncy, Terr. Moll , 1. 122, gives the limits of the corresponding " Southern
Interior Section " such as would include these species. Several of the species of East
Tennessee, also, have been found in Arkansas, — a fact also favoring a wider limit
to the Cumberland Sub-Region.

t This species has not actually been found within the limits of the State of Texas,
but in the neighboring State of Arkansas and in Mississippi. To it may be applied
the remarks on Zonitrs mynificans and Helix Juchsoni on p. 21.";.

\ Either by commerce, by oceanic current-, or from some former molluscous fauna
of which these now isolated localities were offshoots.

ij Since the above was written, this species has been found by Dr. Newcomb near
Savannah, Ga. It may therefore prove a widely distributed American species. In
Jamaica it is known as If. Yendramtana, Glovne.

MUSEUM OF COMPARATIVE ZOOLOGY. 217

On the accompanying map the Pacific Province is colored pink,
the Central Province * blue ; the Eastern Province (of which the
northern portions are not shown) is uncolored. The subdivisions, or
regions, of the Eastern Province are also indicated by colored lines.
The red line marks the division between the Northern and Interior
Regions. From this line the last-named region extends (its Sub-Region
of the Cumberland shown by green lines) to the brown and yellow
lines, which, taken together, mark the northern boundary of the South-
ern Region, the yellow separately indicating the Texan Sub-Region,
the brown the Floridan Sub-Region.

In the above pages I have simply stated the facts now known regard-
ing the actual distribution of our land shells, scarcely attempting to
explain it. I will here venture to make a few suggestions on this sub-
ject.

Even at the present stage of our knowledge, we are justified in
believing that North America has received a group of small species
from the eircumpnlar regions. These species are common to the three
continents, Europe, Asia, and America. A great duration of time has
been required to effect their wide distribution over the continent, even
into Mexico. I believe, therefore, that they are no recent acquisition
to our fauna. They may even antedate the creation of our strictly
American species. f

Again, in the Southern Region we have evidence of immigration
from other faunas ; Florida possessing West Indian and South American
species, Texas many from Mexico. We have, however, at the same
time, equal evidence of a distinct creation for a large portion of the
fauna of our Southern Region, so peculiar is it to the region.

By a distinct creation only can I account for the origin of our pecu-
liarly American fauna of the Eastern Province. I have traced it to

* On the map the dividing line between the Centra] and Eastern Provinces is
carried more easterly, above Lat. 40°, than described in the text. This is done on
account of the Central Province overlapping the Eastern at this point, as indicated
by the distribution of Patula strigosa.

t Of these minute species common to the three continents, I find two at least,
Helix pulchella and Cionella subcylindrica, giving somewhat similar proof of great
antiquity by their distribution in Europe and Asia, and especially by their presence
in Madeira.

218 BULLETIN OF THE

Post-pleiocene times,* since when scarcely any change has occurred in
the fauna.

Of the origin of the fauna of the Central Province, little can be said
with our present knowledge. The same applies to the Pacific Prov-
ince, though the peculiarity of its species surely indicates a distinct
creation of its fauna.

Finally, we have in the list of American land shells several species,
purely local in their distribution, imported through the more or less
direct agency of man. Of these, Helix aspersct was no doubt intro-
duced as an article of food by foreign residents of Charleston, S. C,
and seems to have established a hold there. f Zonites cellarius was
introduced by foreign shipping, probably around water-casks. It is
also well known to have been introduced into other countries. The
Limaces are found around human habitations; they seem to follow the
English to all their colonies. The other foreign species mentioned on
pp. 205,213 have probably been introduced around the roots of plants,
as have been other species which are from time to time sent me from
greenhouses, gardens, etc. They are only local except Helix hortensis,
which may have been accidentally introduced in some other manner,

* I suggested on p. 206 that the region of Ohio and Indiana was the point from
whence the fauna was distributed. Another theory might suggest that the Cum-
berland Sub-Region was the point of origin of all the species, those still restricted
to that sub-region not being adapted to the wider distribution which the other
species have obtained. Any one familiar with the habits of snails is well aware
how much they differ in this respect. Some are much more disposed to migrate
than others. Thus, Helix appressa is content to remain within a radius of a few feet
under a decaying log. Helix thjroidus is more restless, travels much, and climbs
trees. Helix nemoralis has no local attachments, migrating far and wide. These
facts I haye verified in my own garden during many years. The IT. Xapp7-essa spoken
of are descendants of Illinois specimens given me fifteen years ago by the lamented
Kennicott.

t I have been asked what authority I have for this opinion, so think it worthy of
statement that Charleston specimens belonging to the cabinet of the late General
Totten still retain a strong odor of the garlic which seasoned them for the foreign
palate. I have myself had specimens given me by French residents of the town
where I reside, who had bought them as food in Philadelphia. The species has
also been imported into Havana, Rio Janeiro, St. Iago, Chili, and other ports as an
article of food. I received living specimens from a garden in Charleston, S. C,
within the last twelve years, and in 1871, Professor Fcatherman sent me specimens
from Baton Rouge.

MUSEUM OF COMPARATIVE ZOOLOGY. 219

since the discovery of America by European?, and owe its present dis-
tribution in the northeast to its being peculiarly adapted to colonization.
I have elsewhere related my successful attempt to colonize the allied
H. nemoralis*
Burlington, N. J., June, 1873.

* Before closing I will continue the note to p. 202, which suggests some of the
changes which would be caused in the pulmonale fauna of Eastern North America
by a subsidence in the continent of eight hundred feet. In the Southern Begion
the change would be still greater. All the species peculiar to it, catalogued on p.
213, would perish, excepting Bulimulus alter natus. This species would still be found
in Kentucky, restricted to a small area; all record of its former wide distribution
being at the same time destroyed.

The West Indian and South American species, catalogued on p. 214, would no
longer be found on the North American Continent, nor would any record be pre-
served of the former connection of the regions. Indeed, no one would then suspect
that the tropical genera Gland'ma, Veronicella, and Cylindrella had ever been repre-
sented on this continent.

The West India Islands being much more widely separated from North America,
the presence among them of the small American species (catalogued on p. 214)
would be still more difficult to explain.

Again, the supposed subsidence would destroy most of the species peculiar to the
Sub-Begion of Texas (see p. 215), and remove the evidence of the present inter-
mingling of the North American and Mexican faunas in that Sub-Begion.

Another effect would be to remove from our reach all evidence of the origin of
our species in Bost-pleiocene days, the fossil deposits in the bluffs being rendered in-
accessible. Thus one would not be able to have correct impressions of the origin
and distribution of certain species. The non-pulmonate Hclicinat give the best in-
stance of this. Finding Helicina orbiculata and occulta confined to the narrow limits
of the Appalachian Island, one would have no reason to suspect their past history
has been so much more interesting than that of many of the species of Stenotrema,
etc., found with them, which never had had a larger distribution. It would be im-
possible to know that Helicina orbiculata and occulta flourished greatly in Bost-pleio-
cene times ; that later, one of them, occulta, became comparatively rare and restricted
in range, while orbiculata became very numerous in individuals over a vast extent
of territory; and finally, that our supposed subsidence gradually restricted them to
the Appalachian Island.

This supposition of subsidence might be carried still further, till we should have
in certain islands of the Appalachian chain the sole resting-places of the now widely-
distributed Eastern North American fauna. The more southern of these islands
would alone retain the species of the present Cumberland Sub-Begion, and thus be
much richer in species than the more northern islands. On the other hand, these
more northern islands would possess species derived from the present northern
regions which would not be found in the southern islands.

220 BULLETIN OF THE ZOOLOGICAL MUSEUM.

Still more instructive is the supposition of a subsidence in Eastern North Amer-
ica which would leave above the level of the sea only two groups of islands formed
by the White Mountains of New Hampshire, and Mount Mitchell and Black Moun-
'tain of North Carolina. On the latter we may suppose would be preserved all the
species given in the lists on pp. 209, 210, 211. Of these species all would he pecu-
liar to the island, except such as are named in the list on p. 208, which would all
be found also in the White Mountains, where we should also find the following
species peculiar to the islands, Helix Sayii, dentifera, Vitrina limpida, Zonites milium,
Binneyanus, ferreus, exiguus, midtidentatus, Patula striatella, asteriscus, Pupa decora,
Vertigo Gouldi, Bollesiana, simplex, Succinea Totteniana. Of the former distribution
of these species nothing could be known, hut a former connection of the two groups
of islands would be surely indicated by the presence of so large a proportion of
species common to each. A former connection of the two groups of islands with
Europe and Asia would lie as surely indicated by the presence on each of Zoiiitcs
fulvus, nitidus, viridulus, Helix harpa, pulchella, Cionella subcylindrica, and Pupa mus-
corum. Nor could it escape the attention of conchologists that these and other small
species, Z. arboreus, &c. (see p. 204, note,) proved that a former connection must
have existed between these groups of islands and the far-off Central and Pacific
Provinces.

MUSEUM OF COMPARATIVE ZOULOGY.

221

Xo. 10. — Ophiuridce and Astrophytidcc, Old and New. By
Theodore Lyman.

Ophiopeza Peters.

Ophiopeza fallax Pet. is well distinguished from 0. Yold'd Ltk., as I
satisfied myself by examination of the originals at Berlin and Copenhagen.
Two specimens of the same size differed as follows : 0. fallax: diameter of
disk !) mm . ; grains on disk, 2G in a mm. long ; eighteen mouth-papilla; to
each angle; eight nearly equal, crowded arm-spines. 0. Yold'd: diameter
of disk, 9 mm.; grains on disk, 13 in a mm. long; five spaced arm-spines,
the middle one longest. Ophiopeza does not differ from Pectinura Fbs., ex-
cept in having small supplementary mouth-shields, and O. fadax even has
such pieces as an occasional accident.

Pectinura Forbes.

Liitken (Addit. ad Hist. Oph., III. pp. 31 and 10.r>) correctly separated
from Ophiarachna Midi, and Tr. those species which had spines arranged
along the outer edge of the. side arm-plates, and placed them with Pectinura
(Forbes, Linn. -Trans., XIX. 143). The original P. vestita of Forbes is only
known by the figure (PI. XIII., Figs. 1 -7), which is apparently that of a
young animal. In Ophiarachna only two species arc left, O. inerassata, the
largest known Ophiuran, having four arm-spines, and pores between the
under arm-plates nearly to the tip of the arm ; and 0. affinis, with six arm-
spines and pores only between the first and second under arm-plates. With
Pectinura should he included Ophiopezella Ljn. and Ophiochaxma Grube,
which are only extreme forms of this genus. Its species may then be tabu-
lated as follows: —

Xo pores between
under arm-plates.

f Radial shields, granulated, 13-15 1 „ T

arm-spines ) R *Pmosa L>"m-

Disk cnv- .
ere J, under |
its g r a n u - {
lation, with
coarse scales,
or swollen
plates.

Radial shields naked : also some I

other disk-plates; 9 arm-spines

P. in/ernalis Ltk.

f Arms cylindrical at their inser- I p ■ ttk.

tion in the disk, winch is puffed. )

Pores hetwcn first
and second under -
[ arm-plates.

Arms widened ) 10-11 arm-spines. P. mannornta Lym.
at their insertion I
in the disk, which f
is flat. J 5-6 arm-spines. P. stellata Lkt.*

* By comparing the originals I found Crube's Ophiochasma a'hpersitm was the Ophiarachna
stellata of Ljungman.

222

BULLETIN OF THE

f 10 thin equal arm-
spines ; underarm-
plates encroached
nn by side arm-
plates.

Disk, under
its granula-
tion, covered
with niiiiut
and smooth
scales

P. i-estita Fbs.

Pores only be-
tween first and sec-
ond under arm'
h I Plates.

P. maculata Vll.

Pores between the
under arm-plates
continued for some
distance along the
arm .

7- S conical arm-spines, the lowest It, • ,,,

one a little the longest ) P "Ptemspinosa Ltk.

8- 9 flat, pointed arm-spines: the )
lowest one very long and flat, often [ P. rigida Lym.*
equal to two joints in length. )

Pectinura (OpMarachna M. T.) infernalis. The original at Leyden is
lost: but among the unsorted specimens of the Museum Gotleffrov at Ham-
burg I found a specimen from near .Sumatra. It was unmistakable, and is
figured, Plate VII., Fig 1. I found another specimen from the Philippines,
by Semper.

Pectinura septemspinosa. The original of Midler and Troschel from
the Moluccas is yet at Leyden, and remains unitpie. Diameter of disk 25
mm. Arms stiff, thick, cylindrical. Disk closely and evenly granulated,
with a smooth surface not indicating the scales below ; six grains in a mm.
long. Radial shields small, oval, brown and very distinct. Usually seven
arm-spines, but, on the inner joints, eight; they are conical, not so long as a
side arm-plate ; the lowest one a little the longest, and having its base cov-
ered by one of the tentacle-scales. The under ami-plates within the disk
have pores between them. Supplementary mouth-shield small. Color, yel-
low-brown. On Plate VI., Figs. 10-13, are shown the peculiar broken
upper arm-plates of this species ; an angle of the mouth ; and a row of arm-
spines.

Pectinura marmorata sp. nov.

riate V., Figs. 1-7.

Special Marks. — Pores only between the first and second under arm-
plates. Arms somewhat widened at their insertion in the disk. Eleven
arm-spines.

Description of a Specimen. — Diameter of disk 20 mm. Length of arm
about 105 mm. Width of arm at disk 4.7 nun. Height of the same 4.7 mm.
Fourteen small, close-set, tooth-like mouth-papillffl to each mouth-angle, of

* Mr. F. W. Hutton (Catalogue of the Echinodermata of New Zealand, 1872) has described an
Ophium ( Ophiarachna ') cylindrica. It is earnestly to be desired that general zoologists, who have
not large collections for comparison, should abstain from describing species. For such persons, thus
situated, to give useful diagnoses is simply impossible. They only add to the confusion already
existing. Zoology, so far as concerns genera and species, has now passed into the hands of special-
ists ; and they alone can treat such subjects.

MUSEUM OF COMPARATIVE ZOOLOGY. 223

which the outermost one is widest. Outside this, and continuous in the row,
is a long papilla which stretches upward into the mouth-slit, and embraces
the second mouth-tentacle ; its base rests on the side mouth-shield and on
the first under arm-plate. This same piece is seen in other genera, Ophio-
coma, Ophiura, etc., and is usually reckoned among the true mouth-papilla?.
In Ophioglyplia it takes on a great development, and is the piece which car-
ries the scales of the mouth-tentacle, and which gives the forked look to the
mouth-slit. In the species under consideration it may cany a sort of lobe or
articulated scale. Five short, flat, rounded teeth, the uppermost one a little
longer and sharper. Mouth-shields rounded heart-shape ; rather wider than
long, standing close to the mouth-papilhe ; length to breadth 2.5: 3. Sup-
plementary mouth-shields semicircular and about half the size of the mouth-
shields. Side mouth-shields very small and wedged between the first under
arm-plate and the mouth-shield. Under arm-plates strongly overlapping,
bounded without by a broken curve and by a re-entering curve within ;
length to breadth, within disk, .G : .8. The first plate has a broad diamond
shape, and, between it and the second, are two large pores. The side arm-
plates cover nearly one half the, height of the arm at its base, the rest being
occupied by the upper arm-plates, which there are considerably arched, have
a wavy outer side, and a length to breadth as .6 : 3. A little farther out, the
upper arm-plates are lower and less arched. Disk flat and round, but embrac-
ing the base of each arm by a forked projection. With the exception of the
radial shields and shields of the mouth, it is closely granulated; about eight
grains in the length of a mm. This granulation is, however, easily rubbed
off, and then the coarse swollen scales of the disk maybe seen; among
which there are along the edge of the disk, in each interbrachial space, about
six imbricated plates, making a close row between outer ends of the radial
shields, which are large and conspicuous, oval in shape, and having a length
to breadth of 4 : 2. Arm-spines near disk, ten or eleven, small, slightly ta-
pering, somewhat flattened, a little rounded, about two thirds of the length
of a side arm-plate ; the lowest one scarcely longer than the rest. Two flat,
oval tentacle-scales ; the one which lies at the base of the lowest arm-spine
is smaller, as usual, but has its end rounded and not cut square off.

Color, in alcohol, light yellowish brown above, with, lines and patches of
darker, and with darker specks on the upper arm-plates ; below, the under
arm-plates and mouth-region are white.

This species, by its conspicuous radial shields and its arms widened next
the disk, stands near P. stellata, from which it is easily distinguished by
more numerous arm-spines and differently shaped under arm-plates.

C. Semper ; Philippines.

224 BULLETIN OF THE

Peetinura rigida sp. nov.

Special Marl*. — Very large, with cylindrical arms; disk closely granu-
lated and smooth. Aery small sunken radial shields; pores between under
arm-plates far out on arm. Nine or ten much flattened arm-spines, the
lowest one much the longest and largest.

Description of a Specimen. — Diameter of disk 36 mm. Length of arm
175 mm. Width of arm near disk 5.5 mm. ; height 6 mm. Ten flat, close-
set, rough-edged mouth-papillse to each mouth-angle, of which the two
under the teeth are a little higher than the rest. There is also a small ad-
ditional papilla, nearly covered by the outermost mouth-papilla, which
stretches upward and embraces the second mouth-tentacle. Six short,
broad, Hat teeth with a curved cutting edge ; the lowest one often split in two.
Under arm-plates, within the disk broader than long, bounded without by
a curve, and within and on the sides by re-entering curves ; length to breadth
1.8 : 2.8. Farther out on the arm they are as broad as long, with rounded
corners, and they everywhere are thick and even swollen, and have pores
between them, nearly to the end of the arm. Side arm-plates flat, and occu-
pving nearly the whole height of arm. Upper arm-plates regular, not
broken, and with a wavy outer margin; they occupy most of the upper sur-
face; length to breadth 2 : 4. Mouth-shields large and lying close to the
mouth-papillae, roundish with a small point within, and a re-entering curve
without ; length to breadth 4 : 4.5. Supplementary mouth-shields small,
roundish, about 2 mm. in length and width. Side mouth-shields minute and
wedged between the first under arm-plate and the mouth-shield. Disk, ex-
cept small sunken radial shields about 2 mm. long, and the shields of the
mouth, closely and smoothly granulated with about eight grains to 1 mm.
Ion?;. The underlying scale-coat is of delicate, smooth scales, not more than
.4 mm. wide, and difficult to distinguish. Arm-spines nine or ten, much
flattened, about three fourths the length of the side arm-plate, except the
lowest, which is much stouter, and whose length equals two underarm-plates.
The upper spines are somewhat shorter and notably wider than the lower,
and all taper to a blunt point. Two tentacle-scales, of a rounded oval
shape ; the one which covers the base of the lowest arm-spine often larger
than the other.

Color, in alcohol, purplish brown, with black radial shields.

Variations. — Another specimen, of about tin' same size, had the lowest
arm-spine even longer, and often with a thickened end.

This species stands nearest to P. seplemspinosa, from which it is distin-
guished by more numerous and more flattened arm-spiues, and by the great
length of the lowest one.

Zanzibar; Mr. Cooke.

MUSEUM OF COMPARATIVE ZOOLOGY. 225

Ophiocoma brevipes Peters, insularia Lym., and ternispina v. Mart.

I have examined great numbers of these, especially in the Museum Godef-
frov and the Museum of Comparative Zoology, and do confess myself much
puzzled. They ail agree in being completely and closely granulated, above
and below, and in having very regular, cleanly cut arm-plates. The variation
in color is from dark brown, through gray, and reticulated in patterns, to
pure white. It would seem that 0. brecipes (originals at Berlin) has five
spines on the first eight joints, and then four, and that the upper are the
longest. 0. insularia has four spines at the base of the arm, and the upper
are the shortest; the disk granulation is coarser, only six or seven in the
length of a mm. 0. ternispina has but three spines, which are tapering,
cylindrical, and often bent. A number of specimens, supposed to be 0. bre-
vipes, from the Philippines, had only four arm-spines at the base of the arm;
there were from nine to twelve grains in the length of a mm. On present
evidence it will not do to bring these three under one head. As to O.
squamala M. T. the original at Paris is lost, and nobody can now tell what
it was, though it might have been 0. brecipes.

Ophiocoma alternans (von Martens, Oph. Ind. Oc, p. 251). I found,
by the Berlin original, that this is only the young of some species, probably
O. scolupendrina.

Ophiarthrum pictum.

Ophiocoma picta Mull, and Trosch., Syst. der Asteridcn.
Plate VII., Figs. 2-4.

Special Marks. — Disk ornamented with meandrine brown lines ; a dark
line along the upper arm. Three ringed arm-spines, the upper one lomnx.

Description of an Individual. — Diameter of disk 15 mm. Width of arm
■without spines 3 mm. Distance from outer side of mouth-shield to inner
points of mouth-papillae to that between outer corners of mouth-slits, 3 : 4.
Mouth-papillae four (rarely three) on each side, the outer one tapering and
pointed; the next wider than long and rounded; the two innermost as broad
as long, and bead-like. Tooth-papillae (including as such all those under the
teeth) fifteen to seventeen, of nearly equal size, bead-like ; arranged in two
outer vertical rows of four or five each, with a more irregular central row,
and two or three odd papillae below. Teeth four, upper one largest and
broadest; all stout and thick, with rounded corners and a straight cutting
edge. Mouth-shields nearly round, faintly pointed within; length to breadth
2 : 2. Side mouth-shields thick, nearly joining their neighbor of the next
mouth-shield; within, running to a point, but not meeting. Under arm-
plates about as broad as long, clearly defined and regular, with outer side
slightly curved and lateral sides re-enteringly curved ; length to breadth
VOL. III. 15

226 BULLETIN OF THE

(4th plate) 1.4 : 1.4. The first plate is minute, being nearly crowded out by
the side mouth-shields. Side arm-plates forming only a small ridge for the
arm-spines. Upper arm-plates hexagonal, with the outer and inner sides
'shorter than the others; length to breadth 1.4:2. Skin of disk quite
smooth, above and below. Arm-spines three, cylindrical, stout, gently
tapering, blunt; lengths to that of upper arm-plate (7th joint) 5. 3, 2.7 : 1.4.
They are essentially smooth, although, under the microscope, the points look
minutely thorny. One huge, tentacle-scale, longer than broad, and ilat.
Ground-color of disk and upper arms purplish brown. Disk, above and be-
low, ornamented with patterns in dark brown lines. Above, there are
straight lines from arms to centre, and, in interbrachial spaces, above and
below, meandrine patterns, which are tilled with white, yellowish, or purplish
brown. A dark stripe along upper arm; arm-spines with three to five dark
rings.

The original specimen of Miiller and Troschel at Leyden, brought from
Java by Kuhl and Van Ilasselt, remained for a long time a unique speci-
men. Within a few years the Leyden Museum has obtained others from
islands near New Guinea, and Professor Semper also got very fine ones from
the Philippines and the Pelews. This species, by its well-marked upper
arm-plates and regular arm-spines, approaches nearer than 0. elegans to the
genus Opldocoma, from which it differs only by its naked disk.

Ophiomastix flaceida sp. nov.

I'late VI., Tigs. 14, 15.

Special Marl?. — Xo tentacle-scales. l>isk smooth, except a few short
spines above. Upper and side arm-plates obscured by thick skin.

Description of an Individual. — Diameter of disk 15 mm. Length of arm
100 mm. "Width of arm 3 nun. Height of arm 2 nun. Distance from
outer side of mouth-shield to inner points of tooth-papilla?, compared with
that between outer corners of mouth-slits, 3 : 2.5. Mouth-papilla? four on
each side, small and bead-like, except the outer one. which is pointed.
Tooth-papilla? similar, usually nine, arranged in three regular rows, of which
the lowest is on a level with the mouth-papilla?. Teeth five or six, the
tipper ones longest, with a square cutting edge ; the lowermost smaller and
rounded, or partly split. Mouth-shields small, of a much rounded heart-
shape, length to breadth 2 : 1.8. They are close against the outer mouth-
papilla?, leaving little space for the minute side mouth-shields, which do not
meet within, and are obscured by thick skin. Under arm-plates squarish,
with rounded corner.-; length to breadth 1.5 : 1.5. Side arm-plates obscured
by thick skin, as also are the upper one-, except in dry specimen-, where
their form is seen to be rounded hexagonal. Disk wholly covered by smooth

MUSEUM OF COMPARATIVE ZOOLOGY. 227

skin, sparsely set above with short, slender, rounded spines, about 1.2 mm.
long. Arm-spines three ; the two lowest slender, rounded, pointed, often
bent, and in length about 1.5 mm. On every second or fourth joint, and on
alternating sides, the upper spine is much thickened, and has the clove-shape
characteristic of the genus ; its length is about 2.5 mm. Towards the end
of arm there are no such spines, but the upper one is usually the longest and
stoutest. No tentacle-scale ; the tentacles are rather long and thick and
without papillae. Color, in alcohol, disk dull greenish brown ; arms yellow-
ish brown.

Philippines ; C. Semper.

Ophioplocus Esmarki sp. nov.

Plate V., Figs. 12 - 14 ; Plate VI., Fig, 6.

Special Marks. — Color, in alcohol, uniform dull purplish above, and light
brown below. Arms decidedly flattened, having the width to the height
3.5 : 2. Arm-spines not tapering, and of nearly equal length.

Description of a Specimen. — Diameter of disk 16 mm. Length of arm
53 mm. Width of arm 3.5 mm. Distance from outer side of mouth-shield to
inner point of teeth, to that between outer corners of mouth-slits, 3 : 3.2.
Mouth-papillae five on each side, or rarely six, stout, irregular, or squarish;
the two outer ones rest on the side mouth-shield, the others on the mouth-
frames. Teeth five or six, the upper ones short, stout, broader than long,
with a curved cutting edge ; the lowest one smaller, nearly on a level with
mouth-papilla?, and resembling them. Mouth-shields heart-shaped, with a
curve without and an angle within; length to breadth 1.2 : 1.2. Side
mouth-shields long triangular, somewhat swelled, not quite meeting within ;
length to breadth 1.3 : 1. Under arm-plates broad pentagonal, with the odd
angle inward; length to breadth, close to disk, 1 : 1.5. The upper arm-
plate is simple only at the tip joint of the arm ; at the next joint a longitudi-
nal crease of division appears ; at the next the upper arm-plate is completely
divided, and there appear two supplementary pieces on the median line, of
which the outer one partly separates the- two halves of the plate. Farther
down the arm the two halves are wedged apart by the supplementary pieces,
whose number has increased to six or seven. Close to the disk most of the
upper surface is occupied by thirteen supplementary pieces, of which three,
of an irregular hexagonal form, lie on the median line. This development
of pieces is quite as in O. imbricatus. Side arm-plates small, and crowded
mostly to the under side of the arm. Scales of disk, above, irregularly im-
bricated and thickened; the longest 1 mm. long; below they are thinner,
finer, and more regular. Along edge of disk run half a dozen larger
rounded, swelled scales. Radial shields small, about 2 mm. long. Genital

228 BULLETIN OF THE

slits 3 mm. long and starting close to mouth-shields. Arm-spines three,
nearly of equal lengths, stout, rounded, blunt, and not tapering. Lengths
to that of under arm-plate, near disk, .8, .9, 1:1. Tentacle-scales two, flat,
nearly oval, and placed closely side by side. On outer side of tentacle-pore
is a semicircular rim or lip. Color, in alcohol, dull uniform purplish above.

This speeies is of high interest as the first from the Pacific coast of North
America, which represents a genus peculiarly characteristic of the great
ocean. Oldoplocus imbricalus, described thirty years ago by Midler and
Troscht'l. has been brought in plenty from such diverse localities as Zanzi-
bar, Mauritius, New Caledonia, the Philippines, and the Kingsmill Islands.
It long remained the sole representative of a genus which was confined to
these faunal limits. 0. Exmurki differs from it in the precise way in which,
on the theory of Agassiz, species should differ within their genus, namely,
in the proportion of their parts. It will be therefore proper to briefly note
these differences. 0. oiihricatus is a thicker animal ; not only is the disk
thicker, but the arm is higher, having a width to height of 3..") : 3, from
which it follows that the arm-spines are less crowded, having more room ;
they also taper more, and the undermost is much the longest, having a
length of 1.3 : 1 as compared with the under arm-plate. The scaling of the
interbrachial space below is coarser and more uneven, having four or five
scales in the length of a mm., while in (>. Eamarki the smallest number is
five or six. The genital slit begins at some distance from the month-shield,
and is very short, while in 0. Esmarki it begins close to the mouth-shield,
and is longer. The minute lip or rim outside the tentacle-pore is barely to
be seen, while in the American species it is conspicuous. These differences
are all the more instructive from the fact that O. imbricatus is a singularly
steady species, and shows none of the variations offered by Ophiocoma sco-
hijti inlriiKi and 0. erinaci us in the same fauna. In examining great numbers
of individuals from many and distant localities, 1 have found no essential va-
riations in 0. imhriratiw, except three specimens from .Japan and one from
Java, in the Leyden Museum, in which the genital opening ran as far as the
mouth-shield.

Professor Esmark found numerous specimens at San Diego, California,
among stones in shallow water.

Amphiura (.1 mph i/>hnl!s) planispina v. Martens (Monatsbor. Kbnig.
Akad., Berlin, 180 7, p. 317). This species is originally described, perhaps
by a misprint, as having; only om mouth-papilla on each side, like Jlnniphn-
I/s. The originals, however, show three month-papilla? on each side. In
color if is just like OpJiinphmt/ivus Wurdemani, which, however, has a mar-
ginal fence of jointed scales round the disk, and its innermost mouth-papillae
are thickened, and run upwards towards the teeth.

MUSEUM OF COMPARATIVE ZOOLOGY. 229

Amphiura lsevis sp. nov.

Plate IV., Tigs. 18-21.

Special Marks. — Disk flat ami thin, nut lobed, covered with small thin
scales. Radial shields narrow and closely joined. Eight mouth-papillae to
each angle. A light line along upper side of arm.

Description of an Individual. — Diameter of disk G.5 mm. The arms
(which were broken) were flattened, not very wide (1.2 mm), gradually
tapering, and seemed to have been not less than eight times the diameter of
disk. Eight mouth-papilke to each angle, of which the two outer on either
side are scale-like and larger than the others, which are stout and blunt; the
outer papilla rests on the side mouth-shield ; the others are supported by the
mouth-frames. Mouth-shields spear-head shaped ; length to breadth .(J : .4.
Side mouth-shields long triangular, meeting within, and closely joined to
mouth-shields. Under arm-plates wide pentangular, with the odd angle in-
ward ; slightly separated by side plates; length to breadth (8th plate) .4 : .G.
Side arm-plates meeting below and encroaching above ; forming but a slight
crest laterally. Upper arm-plates wider than long ; bounded by a wide
curve within, but nearly straight without ; length to breadth (2d plate)
.5 : .9.. Disk unusually thin and flat, with a fine line of juncture along its
rim, between the scaling of upper and lower surfaces. Scales very thin and
fine, without central rosette of primary plates ; near edge of disk nine or
ten scales in 1 mm. long ; half-way to centre, four; and in the interbrachial
spaces below, seventeen. Radial shields rather long, narrow, closely joined
except at the inner tips, which are separated by three little scales ; length to
breadth of each 1.2 : .3. Arm-spines three, nearly equal, cylindrical, taper-
ing, rather slender ; lengths to that of an upper arm-plate .4, .5, .4 : .5.
Tentacle-scales two, wide and thin with a curved free edge ; they arc set at
right angles, one on the side arm-plate, the other on the side of the under
arm-plate, beyond which it projects. Color, in alcohol, brown-gray, with a
light line along upper arm, and obscurer light cross-lines between joints.

This species approaches nearest to A. gracillima, suii/ifis, Januarii, and
pulchella, but differs in several respects, especially in having eight mouth-
papillae to each angle.

Philippines ; C. Semper.

Ophionephthys phalerata sp. nov.

Plate VI., Figs. 7 - 9.
Special Marks. — Arm-spine, next the lowest one. much stouter than the
rest and with thorns at the end. Disk naked, except a belt of fine scales
round each pair of radial shields. No tentacle-scales.

Description of a Specimen. — Diameter of disk 9 mm. Arms very long.

230 BULLETIN OF THE

One which was broken near the end was 145 mm. "Width of arm 1.3 mm.
Distance from outer side of mouth-shield to inner point of mouth-papilla*
compared with that between outer corners of mouth-slits 1.4 : 1.4. Mouth-
papillae four to each angle ; namely, two large wedge-shaped ones, just
under the teeth, and one, small and slender, sitting on each side mouth-shield.
Mouth-shields small, broader than long, nearly oval ; length to breadth
.5 :.7. Side mouth-shields curved and united at their ends so as to form
a continuous ring. Under arm-plates broader than long, squarish, with a
peak within and a re-entering angle without; length to breadth, within the
disk, .4 : .5. Side arm-plates nearly meeting above and making only a
slight ridge, on which stand the spines. Upper arm-plates irregular oval,
rather small, not covering the upper side of arm; length to breadth .4 : .7.
The plates next the disk are narrower and more irregular. Radial shields
long and narrow ; joined for about half their length, which is to their breadth
as 2 : .6. Their free sides are edged, with two or three parallel rows of
minute scales. The rest of the disk is puffed, deeply wrinkled, and quite
naked. Arm-spines on the basal and middle portions of the arm five,
of which four are short, cylindrical, and tapering; but the one next the
lowest is flattened, much thickened, and thorny at its end. Its length is
about .4 mm. No tentacle-scales. Color, in alcohol, brownish gray.

The true OpMonephthys of Liitken, which includes but one species as yet
(0. limicola), has lines of fine scales not only round the radial shields, but
thence extending towards the centre of the disk and along its edges ; also
the mouth-papilla? may be' either four or six to each angle of the mouth. I
prefer, however, to leave the new species in this genus, because the groups
which centre about Amphiura are not yet very clearly established.

One specimen from Philippines ; C. Semper.

Ophiocnida echinata?

Ophiophragmus ccldnatus 1 Ljungman.
Ophiocnida longipeda Lym. M. S.
Plate IV., Figs. 22, 23.
Special Marks. — Upper disk, and a triangular patch in each interbrachial
space below, beset with minute sharp spines. Radial shields narrow, wholly
separated, naked. Three arm-spines, except close to disk. Four mouth-
papilla? on each side. Arms very long.

Description of an Individual. — Diameter of disk 10 mm. Length of arm
200 mm. Width of arm 2.2 mm. Four mouth-papilla? on each side, of which
the innermost is bead-like, the two next of about the same size but more
flattened, and the outer one smallest and tooth-like. Distance from outer
side of mouth-shield to inner point of mouth-papilla? to that between outer

MUSEUM OF COMPARATIVE ZOOLOGY. 231

corners of mouth-slits 2.2 : 2.2. Mouth-shields oval, with outer end cut off ;
length to breadth 1.2 : .8. Side mouth-shields broad and thick, -widely
touching within; they directly join the neighboring side mouth-shields; a
continuous ring is thus formed, and the usual rudimentary under arm-plates
are not interposed. Under arm-plates squarish with clearly rounded outer
corners ; length to breadth (2d plate) 1 : . 7. Side arm-plates only slightly
prominent. Upper arm-plates much broader than long, four-sided, bounded
without by a gentle curve, within by a shorter and re-entering curve, and on
the sides by straight converging lines; length to breadth (2d plate) .7 : 1.8.
])i>k flat, but soft and a little wrinkled ; pretty evenly beset above with
minute, short, tapering spines, the longest about .2 mm. : below, a triangular
patch of similar spines. Radial shields long, narrow, pointed within ; di-
verging and completely separated : length to breadth 2.8 : .G. Under the
spines, the scaling of the disk is obscurely visible, especially between each
pair of radial shields. Arm-spines short, cylindrical, tapering; the lowest
usually longest : on the first four or five joints outside disk, usually four ; on
all beyond, only three: lengths to that of an upper arm-plate .6, .7, .8 : .7.
Tentacle-scales two (on first joint only one), small, not over one third as
long as an under arm-plate, and disposed at right angles to each other.
Color, in alcohol, pale gray. A young one with a disk of 4 mm. had the
primary plates in the centre of the disk, naked ; and often only three mouth-
papilla) on a side.

I deem it best to leave this species under its present name, although it dif-
fers from Ophiophragmus echinatus Ljn. as follows : 0. longipeda has no
spines on radial shields, which also are wholly separated ; mouth-shields
oval and not four-sided ; four mouth-papilla? on each side : three arm-spines,
except on a few first joints. Vide Ljn. Oph. Yiven., p. 316.

Philippines ; C. Semper.

Ophiopsammium * gen. nov.

Teeth : tooth-papilla) numerous and arranged in a vertical, oval clump,
as in Ophiotlirix. Xo mouth-papillae. Disk and arms naked below, but
closely granulated above. Arm-spines stout and thorny, mounted on a crest-
like side arm-plate, as in Ophiotlirix. Tentacles long, covered with papilla?,
and issuing,, not from the under surface, but from the side of the arm.

This genus is nearest Ophiothela, but differs in having the whole upper
surface closely granulated, as also in the side arm-plates.

* 6&IS, a snake; ^a^jj-iov , a granule.

232 BULLETIN OF THE

Ophiopsamrnium Semperi sp. nov.

Plate IV., Figs. 11- 17.

Special Marks. — Six or seven stumpy arm-spines, of which the lowest is
flattened and hooked : two or three little spines on the edge of the disk in
the interbraehial space : granulation of upper surface fine and even ; finer
on the disk.

Description of an Individual. — Diameter of disk G.5 nun. Length of arm
2:5 mm. Width of arm, without spines, 1.7 nun. Mouth-shields small,
rounded, broader than lung, much obscured by being closely soldered with
surrounding parts and covered by thick .-kin. Tooth-pa] >ilhc about twenty-
five, (if nearly equal size and length, crowded into a short oval (dump. Teeth
three: the uppermost longest, the lowest touching the mouth-pnpilhe. Under
surface, and part of the sides, of arms, covered by smooth skin, through
which may be obscurely seen the outlines of the under arm-plates, which,
near base of arm, are wide, short triangles with the apex inward ; and, near
point of arm, have an angular heart-shape. Side arm-plates covered with
thick skin, and forming a well-marked crest, at right angles to the arm, con-
spicuous from below, hut little projecting as seen from above. I pper side
of arm densely covered by a granulation, of about fourteen grains in the
length of a mm. This granulation descends partly down the side of arm.
The terminal joints are naked, with a minute rudimentary tipper arm-plate :
those which follow have a few scattered grains, and these grow soon into a
continuous stratum. Disk, below and on. sides, covered with smooth skin;
above, with a continuous granulation, finer than that of arms: on edge of
disk between the arms, a pair of sharp spines, about .0 mm. long. Genital
scales wide, with their outer edge showing on each side of the arm. Arm-
spines, near disk, usually sevon ; short, stout, blunt, minutely thorny, often
much swelled at the base; the lowest one flattened and armed on one edge
with a double hook, the next one or two also flattened. Length of third and
longest spine .8 mm.: length of lowest and shortest .4 mm. At tip of arm,
three spines, of which the upper is smallest and pointed; the second pointed
or forked : and the third is a sharp double hook. The Ion-' tentacles, of an
elongated club-shape, issue from the side of the arm just below, and in front
of, the middle arm-spine; there is no tentacle-scale. Color, in alcohol, of
arms, straw yellow : disk more gray.

A younger specimen had a disk of 4.5 mm., and arms of lfi mm.; there
were only six arm-spines.

This singular species has rather flat and little tapering arms, which have
a tendency to roll on themselves, as in Ojihiuchondrus.

Philippines ; C. Semper.

MUSEUM OF COMPARATIVE ZOOLOGY. 233

Ophiomaza cacaotica Lym., and Ophiocnemis obscura L.JN., which
have been considered as perhaps the same, arc. on comparing the originals,
quite different. The two specimens of 0. obscura at Stockholm have the
interbrachial spaces on the back of the disk, covered by four or five very
irregular radiating lines of elongated scales, covered by a thick skin so as to
obliterate their outlines ; in the centre, a great number of similar but rounded
scales. The upper arm-plates are as wide at the base of the arm as beyond,
ami there is scarcely any notch in the margin above the arm. The specimen
at Copenhagen, which Dr. Liitken took to be 0. obscura (Addit. ad Hist.
Oph. 40), is really 0. cacaotica, which has in each interbrachial space above,
only one. or at most two, radiating rows of plates, which are naked, swollen,
and clearly denned. Furthermore, however, there is in the Museum Godef-
froy ( Xi>. 6258) a dried specimen of Ophiomaza, considerably like 0. cacaotica,
but with a fine scaling i nhricated on the interbrachial spaces below. It may
he this scaling always exists, but is usually hidden by the extremely thick
skin. As to Ophiomaza being separated generieally from Ophiocnemis by the
absence of granulation on the disk, I can only say that the same distinction
is allowed to separate Ophiarthrum from Ophiocoma.

Ophiocnemis marmorata, see Ophiothrix clypeata.

Ophiothrix comata Mi'i.r.. & Tk. has not been since found and
is generally ignored. The originals in the Vienna Museum are dry and
nearly ruined by time. It is therefore wed to note that the species belongs
to the division of the u'<'iius which ha.- long needle-like spine-', a thin disk,
and slender rounded arms. 0. Suensonii is its type, and 0. comata resembles
it considerably. The sketches (Plate IV.. Figs. 27, 28) show that the shapes
both of upper and under arm-plates are different; and whereas 0. Suensonii
has a purple arm-stripe above and below, 0. comata has, along the upper
arm, a central white stripe bordered by a purple line on each side, and no
Stripe at all below.

Ophiothrix fumaria Mull. & Tk. (Plate IV.. Figs. 33-3(1). Originals at
the Garden of Plants; dry. and in bad condition. Diameter of disk 9.5 mm.
Length of arm 42 nun. Disk-scales conspicuous and bearing thorny cylinders
(Fig. 3ii), which are scattered over the disk, as shown in Fig. 33. Arm-
spines seven, rounded, the second and third longest, "2..") mm, and slightly
cluli-ended. They are opaque and feebly thorny (Fig. 3.~>). Under arm-
plates wider than long, of ;ui angular oval shape (I'd-. 34). It resembles <>.
aspidota, but has coarser disk-stumps ami much shorter arms. I have an
Ophiuran from Banka Strait which resembles it closely but has narrower
under arm-plates.

Ophiothrix ciliaris Mull. & Tr. (Plate IV.. Figs. 29-32). Origi-
nals at the Garden of Plant-: belonging with 0. fumaria, and in similar
condition. Diameter of disk 5 mm. ; length of arm 3.) mm. Seven or eight

234 BULLETIN OF THE

flattened, delicate, glassy arm-spines with sharp thorns, along their edges
only (Fig. 32); the second, third, and fourth are longest, and about equal.
Disk closely beset with minute stumps (Fig. 29), which are larger and tinted
near the edges of the disk and below (Fig. 30). The radial shields have a
less number, and are small. The under arm-plates (Fig. 31) are bounded
by a curve without, and have sides which converge. The upper arm-plates
are broader than long, overlapping, diamond-shape, and slightly keeled.
Color pink. (J. ciliuris and 0. fun/aria are likewise originals of Lamarck.
With them I found Ophiomaza cacaotlca, which marks their locality as the
region of the Indian Ocean. No. 12S, Museum Godeffroy, seems to be 0.
fumaria.

Ophiothrix aspidota Mull. & Tn. Original at Berlin, Xo. 1,108,
Fast Indies, by Schoenlein. Diameter of disk 10 mm. Radial shields na-
ked ; rest of disk closely beset with minute scarcely thorny conical stumps,
which are smaller below. Upper arm-plates have a microscopically granular
surface. Arm-spines glassy with very feeble thorns. Dr. von Martens
agrees that No. 1,9G6, from Makassar, is this species. I also have it from
the Celibes ; and there are young ones — often with a few stumps on the
radial shields — in the Museum Godeffroy.

Ophiothrix propinqua Ly.m. The Ophiurans described by Dr. Liitken
(Addit. ad Hist. Oph., III. 50) as the young of 0. lonr/ipeda are not that, but
the adult <>. propinqua, and have naked radial shields. The tendency of
young Ophiothrices is to have thorny radial shields, even when these are
naked in the full grown.

(Ophiothrix) clypeata L.tx., is the young of an Opltiocnemis, almost
certainly of 0. marmorata. In the Garden of Plants are specimens brought
in 1842 by Ilombron and Jaquinot from Trincomalee, Ceylon.

Ophiothrix Martensi sp. nov.

Plate IV., Figs. 9, 10.

Special Marl:*. — Seven stout arm-spines; the upper ones with thorny,
club-shaped ends ; the lowest having the shape of a triple or quadruple
hook. Disk naked above, with short, conical, scattered spines on the edge
and in intcrbrachial spaces below. Color, above, bright indigo, with a
darker blue line along the arm, ami, along the under side of arm, a white
line.

Description of an Individual. — Diameter of disk 14 mm. Length of arm
63 mm. Width of arm, without spines, 2 mm. Four thick teeth ; below
them, two pairs >>f stout tooth-papilla? ; and, below these, the usual oval of
small, crowded, tooth-papillae about thirty-two in number, of which about

MUSEUM OF COMPARATIVE ZOOLOGY. 235

seventeen form the margin, and the rest are arranged in two or three irregu-
lar lines. Mouth-shields rather broader than long, of a rounded diamond
shape. Under arm-plates covered by a thick skin ; squarish, with rounded
corners. Side arm-plates rather small; although, below, they make a well-
marked rid^e for the arm-spines. Upper arm-plates much wider than long,
six-sided but with much rounded angles; the lateral corners pointed ; length
to breadth, near disk, .7 : 1.7. Close to tip of arm, the shape is wholly dif-
ferent ; longer than broad, wider without than within, the outer side curved;
length to breadth .5 : .4. Disk covered by thick skin which obscures the
outline of the radial shields, and nearly hides the scaling. Near the edge
are a few scattered spines, about .5 mm. long, stout and conical, and which
are more numerous on the. under side. Radial shields quite smooth ; length
to breadth 3.5 : 2.5. Arm-spines near the disk, seven; the upper ones
short, stout, with a nearly smooth shaft, and a club-shaped end bearing
strongly curved thorns ; the lowest one keeps the form of a triple or quad-
ruple hook quite to the base of the arm. Lengths, to that of an upper arm-
plate, 1.8, 1.9, 1.8, 1, .G, .4, .4, : .7. Close to tip of arm only three spines,
of which the uppermost is longest; the lowest is a hook, the other two hav-
ing a rounded shaft and about six curved thorns at the end. Tentacle-scales
small, scale-like, usually serrated. Color, above, bright indigo, with a
darker line along the arm, bounded by a lighter one on either side; below,
paler indigo, with a white line along the arm.

A smaller specimen with disk of G.5 mm. had a thinner skin, so that the
various plates were better defined, and the ends of the arm-spines seemed
more thorny. In the. interbrachial spaces between the radial shields could
be distinguished four irregular radiating rows of oval scales. Below, the
short disk-spines were more numerous than in the full grown, and were a lit-
tle thorny at their ends. The blue lines along the arms were, continued to
the centre of the disk, but were not margined by lighter lines. Under arm-
plates thick and conspicuous; squarish, with rounded corners; length to
breadth (7th plate) .6 : .G. Chewing apparatus the same as in the adult,
except that the oval of small tooth-papillas has only twenty-one, of which
eight papilla? form an irregular double or single line.

Philippines; by Semper.

Specimens also in Garden of Plants; at Berlin, and at Stockholm.

Ophiothrix pusilla sp. now

mate III., Figs. 21 - 30.

Special Marl;*. — Upper arm-plates swelled, narrow, covering only part of
the arm. Side arm-plates with a projecting point running toward the upper

236 BULLETIN OF THE

arm-plate. About nine short arm-spines, of which the lowest continues as a
hook quite to base of arm. Disk densely beset, radial shields and all, with
very short, even, forked stumps.

Description of an Indlcldual. — Diameter of disk 5 mm. Length of arm
1G mm. Width of arm 1 mm. Tooth-papillae of usual form, arranged in an
oval, whose border is made up of about eleven papilla? with a central, single
line of three or four papillae. Mouth-shields small, indistinct ; of a rounded
diamond shape. Under arm-plates as long as broad, squarish, with a notch
in their outer side ; they are obscured by thick skin, and are somewhat sep-
arate'. Side arm-plates stout and pretty strongly projecting ; each one has
a triangular piece projecting towards and near the upper arm-plate. Upper
arm-plates swelled and very distinct, longer than broad, narrow, of an oval
diamond shape. Near the base of arm they touch, but near the end they
are separated, and. the side-plates being very projecting, there appear con-
siderable naked spaces on the arm. Disk densely and evenly beset, above
and below, with minute, forked stumps about .4 mm. high, so that no trace
of radial shields is seen. These stumps are very short, and consist of a little
trunk with a diverging crown of three, four, or live sharp, slender prongs.
Those of the interbrachial spaces below are somewhat longer in the trunk,
and a few, towards the mouth, have only short blunt thorns. Genital scales
at outer end of openings very broad, and nearly meeting over the arm.
Arm-spines short ; near base of arm, nine ; of which the lowest is a triple
hook. Lengths, to that of an upper arm-plate, .7, 1, 1, 1, ,8, .6, .5, .4, .3 : .6.
The five upper are tapering, rather slender and glassy, with three to six
sharp, not very long, thorns on each side. The three lower are blunt, with
two to five blunt thorns at their tip. Close to tip of arm there is only a
larii'e triple hook, above which is a short and very feeble spine. Tentaclc-
scale none, or so feeble as to be scarcely visible. Color pale blue, in alcohol.

Philippines ; by Semper. Specimens also in Garden of Hants.

Ophiothrix exigua sp. nov.

Plate IV., Figs. 24 - 2G.

Special Marks.- — Upper arm-plates thick, distinct: as broad as long;
curved without. About seven short arm-spines, of which the lowest con-
tinues as a hook to the base of the arm. Disk above densely beset, radial
shields and all, with very short, even, forked stumps, but naked below.

Description of an Indlcldual. — Diameter of disk 5 mm. Length of arm
17 mm. Width of arm 1.1 mm. Tooth papilla? arranged in an oval whose
border is composed of about eleven papilla', with a central line of three.
They are more pointed and less crowded than in 0. pusllla. Mouth-shields
much wider than long, and of a diamond shape, with rounded angles. Under

MUSEUM OF COMPARATIVE ZOOLOGY. 237

arm-plates about as long as broad, with a strong notch on their outer side and
a curve within; they are somewhat separated. Side arm-plates stout, not
projecting, and completely covering the space between upper and under
plates, so that the arm, without its spines, has a pretty regular, rounded form.
Upper arm- plates slightly broader than long; length to breadth, near the
disk, .•! :.7 : distinct, rather thick, of a much rounded diamond-shape, with
the outer side curved, and the two laterals converging inward nearly to a
point. Disk, above, densely beset with very short, minute stumps, each
bearing a crown of two, three, or rarely four, sharp, diverging prongs, each
stump about .4 nun. high, on the average. Below, the interbrachial spaces
are naked. Genital scales, at outer end of openings, very broad, and arching
partly over the arm. Seven rather short arm-spines flattened, not much
tapering; the longest with seven to nine rather strong thorns on each edge
and three blunt thorns at the tip ; the lowest has a hooked form even at the
base of arm. Lengths to that of an upper plate (7th joint) .8, 1.2, 1.2,
.5, .3, .2 : .6. Tentacle-scales minute, pointed. Color, in alcohol, pale blue,
with darker markings on upper arm.

It differs from 0. pusilla in having the disk nearly or quite naked below,
in different upper arm-plates, and in wider arm-spines.

Philippines, by Semper. Specimen also in Garden of Plants.

Ophiothrix stelligera sp. nov.

Plate III., Figs. 15 -SO.

Special Marks. — Disk, above, closely beset with minute, stout stumps,
each with a crown of five or six short, not much diverging prongs ; there are
also a few short spines. Below, interbrachial spaces, near mouth, naked ;
but near edge of di<k, beset with small cylindrical spines. Under arm-plates
with a clean curve, without.

Description of a Specimen. — Diameter of disk 5.5 mm. Length of arm
28 mm. Width of arm 1.5 mm. About twenty-one tooth-papillae, of which
fifteen form the border of the oval, and six are arranged in an irregular line
in the centre. .Mouth-shields as broad as Jong, of a rounded diamond-shape.
Underarm-plates a little broader than long; 7th plate, length to breadth,
.6 : .8, bounded, without, by a gentle curve, and by laterals which converge
moderately, giving the whole plate a nearly oval look. Side arm-plates only
moderately projecting, covering completely the space between the upper and
under plates. Upper arm-plates thickened and with a slight longitudinal
ridge; a little broader than long.; length to breadth .0 : .8, near di>k ;
bounded without by a curve; on the sides by re-entering curves, which con-
verge to a blunt point. Disk, above, closely beset with minute, stout stumps,
each with a crown of five or six short, not much diverging prongs ; the out-

238 BULLETIN OF THE

line of the radial shields may be distinguished, because they bear fewer
stumps. Under the microscope the disk has a luuk uf being sprinkled with
little glass stars, among which appears, very rarely, a short spine. Below,
the interbrachial spaces are naked at their inner angle, but towards the mar-
gin are beset with cylindrical spines which are about ^ mm. lone, while the
stumps are not over i mm. Genital plates wide, at outer end of the open-
ings. Ann-spines about eight, short, glassy, flattened, and rather blunt,
about as in 0. exigua ; the longer ones have six to nine well-marked thorns
on each edge, and three minute thorns at the tip ; lengths to that of an upper
arm-plate (7th joint) 1.8, 2, 2, 1.8, 1, .8, .6, .5 : .0. The lowest spine retains
the form of a triple hook quite to base of arm. Tentacle-scales blunt, small
but distinct. Color, above, disk indigo round margin, paler towards centre ;
upper arm-plates mottled blue and white, with an ill-marked central line :
lower surface paler.

This stands near O. carlnata v. Mart., but the disk-stumps with their star-
like crowns, the disk-spines below, the more numerous arm-spines, and dif-
ferent upper arm-plates, will separate it from tins, as well as from 0. exigua
and 0. purilla.

Philippines, by Semper ; also from Borneo, in the Hamburg Museum.

Ophiothrix plana sp. nov.

Plate IV., Figs. 1 - 8.

Special Marks. — Bisk smooth and covered by a thick epidermis, through
which the scaling and the outlines of radial shields are seen vaguely. Near
edge and on interbrachial spares below, short thorns covered with so thick
an epidermis as often to be bead-like in form. Color of disk pale blue,
with minute darker rings. Seven arm-spines, all. usually, with widened
ends, except the upper one, and covered with a thick epidermis.

Description of an Individual. — Diameter of disk 6.5 nun. Length of arm
32 mm. Width of arm 1.5 mm. Tooth-papilla; eighteen, of which fifteen
form the bonier of the oval, and three much shorter ones a single line in the
centre; all the papillae, except lowest, quite stout. Mouth-shields small,
oval, with a peak within, and closely soldered with surrounding parts.
Under arm-plates squarish, with rounded corners and a re-entering curve
without ; length to breadth (8th plate) .0 : .7. Side arm-plates well marked,
but small, and lying close to arm. Upper arm-plates small and swelled,
bounded without by a wide curve; on the sides by slightly re-entering curves
which converge strongly, leaving the inner side very short ; length to breadth
(8th plate) .G : 1. At the very tip of arm, the side arm-plates meet above,
but not below. Bisk nearly smooth, with radial shields rather small and on
a level with the rest of the surface ; the centre occupied by a rosette of thin

MUSEUM OF COMPARATIVE ZOOLOGY. 239

scales, nearly obscured by epidermis ; a large, round, primary scale in the
centre, surrounded by two and part of a -third rather irregular concentric
riivs of smaller ones ; from this central rosette radiate interbrachial bands,
each consisting of a line of three or four larger elongated scales, with a line
of about five narrower and smaller ones on each side ; the whole variable,
irregular, and much obscured by a thick epidermis. Along edge of disk,
outside radial shields, a cross line of about five large scales. Radial shields
smooth and fiat ; separated by a line of two or three scales ; length to
breadth 2:1. Interbrachial spaces below beset with minute spines, which
are encased in such a thick epidermis as sometimes to have a bead-like form:
they continue to the edges of the upper surface, but there are none in the
centre. Arm-spines seven ; a little out on the arm, six ; flattened, glassy :
the upper one slender, tapering, and with five or six thorns on each edge;
the lowest one formed like a stout double hook ; the rest with a smooth-
shaft much flattened at the end, where it has five to seven long thorns on
each edge, giving a wide brush-like shape to the tip, which is increased by the
thick epidermis investing the whole ; lengths to that of an upper arm-plate
(4th joint) 1.8, 1.8, 1, .G, .5, .4 : .6. At the very tip of arm, only three spines,
of which two are needle-like and the lowest one is a slender double hook.
Tentacle-scale well marked, with a sharp point. Color, above, pale purplish
blue with minute spots, or rings, of darker on the disk.

Another specimen with disk of 7 mm. had the arm of 27 mm. The disk-
thorns were confined to the interbrachial spaces below.

This species is easily recognized by its smooth, shiny disk, with little dark
rinirs, and its broad-ended arm-spines.

Philippines, by Semper. Specimens also in Garden of Plants ; and, from
Makassar, at Berlin.

Ophiothrix rudis sp. nov.

Plate III., Figs. 11-14.

Special Mart*. — Disk, above, except radial shields, closely sot with short,
very thick cylinders, jointed at their baseband having rounded ends. Seven
stunt, rounded, blunt, nearly smooth arm-spines ; the two upper ones much
the longest.

Description of a Specimen. — Diameter of disk 9 mm. Length of arm
62 mm. Width of arm 2.2 mm. About thirty-five crowded tooth-papilla?,
whereof twenty-one form the border of tin- oval, and are much more project-
ing than the fourteen central ones. Mouth-shields wider than long, of a
broad, rounded diamond shape. Under arm-plates closely soldered to each
other; near base of arm. broader than long, of an irregular oval shape, with
a decided depression or re-entering curve within; length to breadth (8th

240 BULLETIN OF THE

plate) 1 : .8. Side arm-plates stout and well marked, but not very promi-
nent. Upper arm-plates, near base of arm, wider than long, with sharp lat-
eral angles, and bounded without by a strong curve, on the sides by
re-entering curves converging inwards, and within by a very short line •
length to breadth (8th joint) .8 : 1.2. Disk naked below, but the upper sur-
face is closely set with short, even, very stout, nearly smooth cylinders hav-
ing rounded ends. These cylinders are .5 mm. high, jointed at their base,
so that they can lie flat to the disk, and their ends under the microscope are
seen to be slightly thorny. Radial shields completely naked, small, sepa-
rated partially by a single line of little cylinders; length to breadth 3 : 1.5.
Seven stout, gently tapering, blunt, rounded, slightly flattened, nearly
smooth arm-spines, which are wholly opaque, and only present feeble termi-
nal thorns under the microscope. The two upper ones are much the longest,
and usually a little crooked ; lengths to that of an upper arm-plate (8th
joint) 2.6, 2.8, 1.2, 1, .6, .3 : .8. Tentacle-scales none, or very minute and
rounded. Color dull indigo, lighter below, with arm-spines mottled with
yellowish.

San Diego, California. Many specimens from shallow water, by Professor
Esmark. 0. magnifca is somewhat near, but has very thorny arm-spines.

On the Species of Ophiothrix from the Waters of Western Europe and of
the Mediterranean.

The genus Ophiothrix is usually the most perplexing in respect of
the identification of its species. Those of Europe make no exception,
as may be seen by the following list of names taken from the princi-
pal writers on echinoderms : —

Stella seolopendroidps : Rosida scolopendroides, Linck de Stel. Mar., p. 52,

PL XX VI., Fig. 42, 1733.
Asterias fragilis Abildgaard (O. F. Midler), Zool. Dan., p. 28, PL XCVUX,

1789.
Asterias pentaphyllum : varia ; aculeata ; hastata ; fssa ; nigra Pennant,

Brit. Zool, IV. 54, 55, 1812.
Ophiura fragilis Lamk., Hist, des An. sans VcrtcL., II. 546, 1S16.

" tricolor " " " " « "

Asterias tricolor Delle Ciiiaje, Memorie, III. 78, PL XXXIV., Fig. 9, 1823.
" ecldnata " " " " "

" pentagona " " " " "

" Cuvieri " " '• " 70.

" Fcrussacii " " '• " "

" quinquemaculata " " IV. 197.

MUSEUM OF COMPARATIVE ZOOLOGY. 241

Ophiocoma rosula Forbes, Brit. Starfishes, 60, 1841.
Ophiothrix fragilis Mull. & Troscil, Syst. d. Asterid., 110, 1842.
« echinata " " " "111, "

" alopecurus " " " " " "

tricolor " " " " 112, "

Ferussacii
" quinquemaculata "

" Rammelsbergii " 113,

" fragilis (yar. tenuispi?ia) Sars, Middelhavets Littoral Fauna, II.

p. 74, 18.3 7.
" alba (?) Grube, AViegmann's Archiv., 185 7, 344.
" rosula Lyman, Illust. Catalogue, 154, 18G5.

echinata Litkex, Addit. ad Hist. Oph., III. 52 and 104, 1869.
" quinquemaculata " "

" fragilis " "

" " Ljungman, Vestindiska oeh Atlantiska'Oph., 623, 1871.

" lusitanica " " " " " 625, "

« pentaphyllum " " " " " 622, "

rubra ' " " " " " C24, "

« maculata " " " " " fi-'3, "

" echinata « " " " " 653, "

" Lutkeni, Wyv. Thomson, Depths of the Sea, 100, 1873.

Do all these names refer to one species or to many? I have tried to

throw some light on this question by bringing together for study as many

originals and as great a number of specimens as possible.
In this way I had, side by side, in Paris : —

Ophiothrix echinata. Original of Mull. & Trosch. ; by courtesy of Professor
Peters.

Ophiothrix Rammelsbergii. Original of Mull. & Trosch. ; by courtesy of
Professor Peters.

Ophiothrix alba. Original of Grube ; by courtesy of Professor Grube.

Ophiothrix fragilis. Originals, as identified by Ljungman with Abild-
gaard's Asterias ; by courtesy of Professor Love'n.

Ophiothrix Lusitanica. Original of Ljungman ; by courtesy of Professor Loven.

Ophiothrix pentaphyllum. Originals as identified by Ljungman with Pen-
nant's Asterias ; by courtesy of Professor Loven.

Besides these were great quantities of specimens from the coasts of Den-
mark and Sweden, the Isle of Wight, the northwest coast of France, coast of
Portugal, Madeira (?), Algeria, Spezia, Naples, the northern Adriatic, and
the coast of Egypt.

VOL. III. 16

242 BULLETIN OF THE

At Stockholm, Professor Loven permitted me to examine the whole of
that great collection, and Mr. Ljungman showed me the originals of his
O. rubra and 0. maculata.

At Copenhagen, Dr. Liitken personally set hefore me the very rich collec-
tion of the University, and showed me the original of Wyville Thomson's
O, Liitkeni. I received like marks of kindness from Mr. Schmeltz of the
Museum Godeffroy at Hamburg ; from Professor Richiardi at Pisa ; Profes-
sor Panceri at Naples ; Sig. Trois at Venice ; Dr. von Martens at Berlin,
and from Professor Schlegel at Leyden.*

All these Ophiothrices (with one exception presently to be mentioned) have
certain features in common. They have arms rather short, narrow, and
rounded, being from four to six times the diameter of the disk ; their lower
arm-plates have a notch or re-entering curve in their outer side ; the upper
arm-plates are more or less broad diamond-shaped or rhomboidal, and over-
lap one another. The arm-spines, eight or nine in number, stout, blunt, but
never club-ended, somewhat rounded, feebly translucent, the side thorns not
very distinct and from eleven to seventeen on each edge ; the disk beset with
grains, stumps, or short spines, except the radial shields, which are usually
naked, but sometimes have a few stumps which are merely the persistence of
a character of the young.

Among these characters the plates of the arms furnish no clew whatsoever;
nor yet do the arm-spines, which present a wide range as to length, thickness,
and the number of thorns on their edges. A careful microscopic study of
the armature of the disk, its grains, stumps, or spines, is the best guide to
the specific differences ; to which may be added a consideration of the abso-
lute size attained by these specimens. Some hints, too, may be gained from
the pattern of color : but this must be taken with great caution. It is true
that writers, notably Grube, have successfully distinguished Ophiothrices
almost wholly by their coloration ; but this coloration is that which appears
after immersion in alcohol, and is not to be confounded with that of the living
animal. Thus Ophiothrix angulata has the most varied sets of hues,f but in
alcohol all these turn to pale blue. The pigment patches are arranged in
certain lines or patterns, which are brought out by the alcohol, and which
usually are characteristic. In this way, a stripe along the arm {0. Suensonii)
or regular spots on the disk (0. plana) are good guides to species. In the
European species this guide is often dubious, because many of the specimens,

* To Professor Deshayes a special acknowledgment is due. Whenever I have
been in Paris he has taken me into his own laboratory at the Jardin des Plantes,
given me free access to the collections, and treated me, not as a stranger, but as
one who had a claim on his attention.

t Illustrated Catalogue I., PI. II., Figs. 1, 2,3.

MUSEUM OF COMPARATIVE ZOOLOGY. 243

although brilliant in life, fade to a dull straw-color in alcohol, while others
present a mottled and varied look without any special pattern.

Beginning with northern waters, we find, on the shores of Denmark and
Sweden, a type which is without question the Asterias fragilis of Abildgaard.
It is large, and has a fleshy disk swelling into lobes in the interbrachial
spaces; a diameter of 15 or 10 mm. is a common size, and specimens with a
disk of 9 mm. have still young characters. Here it may be observed that
the upper arm-plates, at the base of the arm, furnish the best test of the ani-
mal having taken on its adult form. For example, in the group under dis-
cussion, these plates have a length to breadth about as 2:3; whenever,
therefore, the length of these plates, near the disk, is greater than, or equal
to, the breadth, it is safe to assume that the creature is still in its young
stage. The arm-spines of 0. fragilis vary more than those of any other
European type ; they may be found very short, and so thickened that the
edge thorns, of which there are only 14, are nearly obliterated; or quite
long, flattened, and glassy, with as many as 22 well-marked thorns on each
edge. Often, or perhaps generally, the upper arm-plates, instead of being
decidedly angular, have the outer side cleanly curved. The radial shields
are somewhat sunken in the swelled disk, and are a little bent near their
small end. The armature of the disk consists of a great variety of short,
stout, stumps, accompanied or not by a few slender cylindrical spines (Plate
II., Fig. 42). A very common form of stump is with a crown of four thorns,
either cylindrical or else spread like a partly opened fan (Fig. 38) ; some, too,
may have only three thorns, whereof the middle one may be flattened and
elongated (Fig. 39). There may be near the centre of the disk a quantity of
conical grains covered with minute points (Fig. 43), with which the preceding
shapes mingle, especially near the edge. In the young we get the simpler
shapes from which are produced the full-grown stumps. Thus, a little one
having a disk only 5.5 mm. in diameter exhibited minute, very stout stumps
which were simply clavate, or else had a crown of three strong thorns (Fig. 41).
"When the disk was increased to 7 mm. the same shapes presented themselves,
but more elongated (Figs. 37, 40); and there were added a few true spines
(Fig. 4 2), slender, cylindrical, and with very feeble thorns at the tip and along
flic sides. The same shapes again occurred on a larger disk, 9 mm., with an
increased number of spines; and, near the edge, a much elongated modifica-
tion of the clavate stump (Fig. 44).

OpJiiolhrix Rammelsbergii, Midi, and Trosch., is a half-grown specimen
of 0. fragilis, in which the disk stumps were encased in thick wrappers of
skin, giving them the look of smooth papilla;. It is a variety always to he
borne in mind while studying this genus, and is illustrated in Figs. 10, 11 ; in
one of which the skin-bag is left on, and in the other partly torn off.

I could not find in 0. alba Grube any characters to distinguish it from a

244 BULLETIN OF THE

half-grown 0. fragilis, although it is said to have been brought from the
Pacific by Escholtz.

It will be best next to speak of the European OpMothrix most unlike that
just described. In Naples, Dr. Gasco showed me a large number of living
Ophiothrices, and called my attention to their great differences. One sort was
blue with a swelled body, short arms, and rather stout arm-spines; the other
was reddish-brown, with a flat rounded disk and slender spines on it, and long
arms, by whose rapid worm-like motion the animal slid briskly over the bot-
tom of a basin tilled with sea-water. This latter is the species identified by
Liitken as Asterias quinquemaculata of Delle Chiaje. Here it is proper to say
that the descriptions and plates of that Italian author, so far as concerns
Ophiurans, are utterly unrecognizable ; the figures, in fact, portray animals
that do not exist anywhere. However, to avoid multiplying names, there is
no objection to taking the nomenclature of Delle Chiaje and applying it arbi-
trarily to Mediterranean species. Unlike most Ophiurans, this species is
better marked as young than as adult, when it bears some resemblance to O.
pentaphyllum (to be described further on). With a disk of 3.5 mm. the arm
was already 25 mm. long, or in proportion of 7:1, which is greater than I
have usually found, even in the full grown of other European species. The
disk was fiat and circular, little lobed, and bad its surface regularly spriukled
with minute, equal, slender, trifid stumps, with sometimes none on the radial
shields and sometimes as many as four (PI. II., Eig. 4G), also there were long,
thin, cylindrical spines, as long nearly as those of the arm, having very small
thorns on the sides and tip ; these spines were all articulated on little
mamelons, on which they have a free motion (Fig. 4 7), a character I have
not observed in the other European species. There were but six arm-spines,
the two upper ones longest, viz. 2.7, 2.5 mm., with a glassy look and nine-
teen thorns on each edge. A specimen, whose disk was 7 5 mm. in diameter,
had stouter forked and trifid stumps on the centre (Fig. 52) ; stouter articu-
lated spines ; and, on the edge of the disk, much elongated forked stumps
(Fig. 53). An individual with a disk of 9 mm. resembled the young one first
mentioned, except that the disk spines were of several sizes, and the upper
arm-plates, of course, proportionally wider. There were six or seven arm-
spines, whereof the longest was 4.5 mm., with as many as twenty-five or even
twenty-seven thorns on each edge. A large specimen had the disk 14 mm.
in diameter, flat and circular, with large smooth radial shields, resembling in
these respects 0. pentaphyllum ; besides the long articulated spines (Fig. 51),
there were in the centre numerous short, thick stumps, with crowns of three
or four long thorns (Figs. 49, 50), or conical grains with thorns on top (Fig.
48) ; while near the edge and below were very elongated clavate stumps (Fig.
54). This was the usual armature in a large number of adults, some with
the disk as large as 16 mm.; the differences were in the proportion and

MUSEUM OF COMPARATIVE ZOuLOGY. 245

closeness of the various stumps and spines mentioned above ; there also was
sometimes a much elongated form of 49 (Fig. 55). Among the young, the
arm is from five to seven times the disk ; among the adult, from six and a
half to ten times.

The blue species, already mentioned as living beside 0. quinquemaculata,
is that identified by Miiller & Troschel with A sterias echlnata Dell. Ch. One
of their originals at Berlin lias the disk 11 mm. and the arm about 40 mm.
The upper arm-plates are rhomboidal, overlapping, and with a slightly
thickened lobe without; length to breadth .9 : 1.2. There are nine short,
stout, little, tapering arm-spines, the longest 2.1 mm., and with a dozen
blunt, feeble thorns on each edge. The disk is evenly set with larger and
smaller trilid stumps (PI. II., Figs. 2, 3), with very few small cylindrical
spines (Fig. 1). This is as large a specimen as I have seen, for the species
is small ; it has a puffed disk, in which the radial shields' are somewhat
sunken, and arc therefore not conspicuous. The arms are always short ; in
five adult specimens the average of the arm to the disk was as 5 : 1. Already,
with a disk of 7 mm., the adult characters are taken on ; the upper arm-
plates are broader than long, as 1 : .G ; while a specimen of O. quiitquemacu-
lata, of the same size, had them of equal dimensions, .9 : .9. The arm-spines
vary little ; they are even, short, stout, and little tapering, and are from
seven to nine in number. The smallest specimen (Naples) had a disk of
2.5 mm., and the arm 13 mm.; the scaling was very distinct, each scale
usually bearing a slender, trifid stump (PI. II., Figs. 12, 13); on the upper
surface their character was the same, but some were simply forked ; on the
radial shields were a few similar but smaller stumps, which, in the adult,
wholly or nearly disappear. Another had a disk of 7 mm., which carried on
its upper surface, evenly set, two and three forked stumps (Figs. 5, 9) and a
few short spines (Fig. 4), there being, on the radial shields, some little trifid
ones (Fig. G). The lower interbrachial spaces were closely set with long
stumps and short spines (Figs. 7, 8) ; between the radial shields of the same
pair there was a single line of stumps. Another specimen — disk 10 mm. —
was evenly set with little forked stumps, covered with a thick envelope of
skin (Figs. 10, 11). An individual from Algeria had a few forked stumps so
elongated as more properly to lie called spines (Fig. 14), but the greater part
of the armature consisted of forms similar to G, 11, and 12. Four specimens
from the Adriatic and one from Egypt presented no new features. It will be
noted of this species : first, that it is small ; secondly, that the disk-stumps
are fine, and never have more than two or three thorns as a crown, while
longer spines are wanting, or very rare ; thirdly, that the arms are short,
being from three and a half to six times the disk.

llelier* has recalled attention to the Adriatic species 0. alopecurus Mull.

* Zooph. und Echinod. des Adriatischcn Moeres, 63, 1868.

246 BULLETIN OF THE

& Trosch.,* but without recognizing any specimens, and in his description
seems to have included tins and O. echinata under the name of O. fragilis.
In the collections of Professors Grube and Richiardi, and at the museum of
the Palazzo Ducale in Venice, I found nearly a dozen specimens, collected
chiefly near Trieste ; and there are others in the museums at Copenhagen
and at Stockholm. They present pretty uniform characters,, and attract
attention by the close-set crop of long, glassy, or silvery spines, by which the
disk is almost hidden ; closer examination shows that there are no stumps at
all on the disk, and that the radial shields are either absolutely naked, or
have only a few extremely minute spines. Moreover, the animal is distin-
guished usually by its dark green color, with which the spines contrast, like
spun glass. A specimen with a disk 12 mm. in diameter, and arms 96 mm.
long, was dark green, with lighter mottlings on the radial shields, which
were naked. The upper arm-plates were rhoinboidal, overlapping, and with
a well-marked lobe without. The arm-spines were nine, flat and regular ;
longer, more tapering, and more glassy than in O. echinata, and with sixteen
or seventeen thorns on an edge. The longest one (usually the second) was
to the upper arm-plate as 3.2 : 1. The disk-spines were uniform in shape,
being stout at the base, tapering, somewhat flattened, forked at the tip, and
with a few very minute thorns on their sides ; their length on the back of the
disk was 1.2 mm., near the edge longer. Other specimens did not vary
essentially from this, except in the comparative length and slenderness of the.
disk and arm-spines; the former having sometimes a regular fluted form
(PL III., Fi". 1) and a maximum length of 3 mm. The radial shields are
usually wholly naked, but may have a few minute spines, not over .2 mm.
long (PI. III., Fi<*s. 2, 3). One specimen had a light-colored disk, with a
black spot on each radial shield, and dark upper arm-plates.

There seems no question as to the distinction of the four species just
noticed ; there are now to be considered some whose claims are less clear.
The Asterias pentaphyllum of Pennant is an inhabitant of the English coast;
Lutken considers it a variety of 0. fragilis, while Ljungman regards it as a
good species. Those I have seen were from the Isle of Wight and from Ma-
deira (?) ; they were readily distinguished from 0. fragilis, but, as there were
no young forms, I am unable to speak with a full knowledge. The disk is
flat and round, and not puffed; radial shields naked and conspicuous; upper
arm-plates with a well-developed peak on the outer side. The disk-stumps
and spines seem in their young state to be thorny grains (PI. II., Fig. 30) and
not forked stumps, as is usual; from this form develop larger grains and
thick stumps (Figs. 31-33) and even columnar spines (Figs. 34, 35); the
more pointed spine (Fig. 3G) was found only on two specimens, said to come

* The original examined by Trosche] no longer exists atLeydcn, but the present
ideatification of 0. alopecurus seems a reliable one.

MUSEUM OF COMPARATIVE ZOOLOGY. 247

from Madeira. As to arrangement, the thick stumps and thorny grains, of
which the most numerous are 32, 33, are concentrated in the centre of the
disk, while the coarse spines are sparsely distributed from the centre awards
the periphery.

Next comes 0. lusitanica, the commonest species, or variety, of some parts
of the coasts of France and Portugal, and even extending into the Mediter-
ranean. Somewhat larger than 0. echinata, it is at once distinguished by
the greater absolute size and thickness of the disk-stumps, and by the more
numerous thorns at their ends ; for, whereas 0. echinata commonly has stumps
simply forked, and never with more than three terminal thorns, 0. lusitanica
has them with a crown of four, five, and even six thorns (PI. II., Figs. 15, 16,
22). A young one whose disk was 5.5 mm. in diameter, had clavate or trilid
stumps, and some with four tei*minal thorns (Figs. 16-18); a still smaller
one, with a disk of 5 mm., had clavate stumps (Fig. 23) also on the radial
shields, where they are scarcely ever seen in the adult. On a specimen from
Naples, disk 8 mm., the stumps were chiefly stout cylinders with a crown of
five thorns (Fig. 15) ; but there were, besides, a few stout spines, some colum-
nar with two side thorns and six terminal (Fig. 19), others with a swelled
cluster of numerous thorns at the tip (Fig. 20). It should here be added that
spines are so rare in this species as almost to be accidents ; while the stumps
are of so even a height and so closely set as to give to the upper disk the
figure of a regular five-rayed star, in which the radial shields are sharply
defined. Of many examined, one from St. Va-est la Ilouge, France, had an
occasional spine of a type like that found in O. echinata and O.frayilis, but
much thicker (Fig. 29 : compare Figs. 1,4 2). To continue among rare forms,
an individual from the lies Chauses had small toothed grains and others
more elongated, derived from them (Figs. 25, 26) ; and 26 may be still further
elongated by the shooting up of the three central thorns, and thus take on
the character of a small spine ; this was found in one specimen from the lies
Chauses (Fig. 27). Among stumps, Figs. 21, 24, are rare eccentric shapes
derived from 15. Fig. 28 is an elongated clavate stump, comparable
to 9 in 0. echinata, 44 in 0. fragilis, and 53 in O. alopecurus, and like
them found near the edge of the disk. Of eighteen specimens examined
with special care, viz. two from Naples, one Portugal, eight St. Va-est la
Ilouge, seven lies Chauses (France), ranging from 5 to 13 mm. in diameter
of the disk, there were five specimens in which the. disk-stumps had not
developed beyond the two or three forked forms, but, as before mentioned,
larger and stouter than in 0. echinata of the same size-. The other thirteen
had more or less cylindrical stumps with crowns of four to seven thorns, and
two of them had also thorny grains (Figs. 25, 26). Among five adult speci-
mens, the average of the arm to the disk was as 5 : 1. The upper arm-
plates are decidedly angular ; the arm-spines about as in 0. echinata,, little
variable, and with thirteen to seventeen thorns on an ed-e.

248 BULLETIN OF THE

Of Ophiothrix rubra Ljn., there exists only a single specimen, from near
Lisbon, at Stockholm. The disk is 7 mm. in diameter, and is crisp, with
thorny stumps on it, as well as on the radial shields. Arm-spines stout and
thick. Without more material to judge from, I am not satisfied that this
differs from 0. lusitanica. Dr. Ljungman thinks it may be 0. ecJiinata.

There are at Stockholm two specimens of Ophiothrix maculata Ljn., from
120 fathoms on the Josephine bank, near the coast of Portugal. They have
disks of 12 mm. diameter and somewhat resemble 0. penlaphyllum, but have
only seven arm-spines, and bear a reddish spot on each upper arm-plate.
The fewness of arm-spines is important, though I have seen an 0. quinque-
maculata with a disk of 0 mm. that had no more. The red spots are not of
so much consequence, since 0. pentaphyllum is variously mottled in alcohol,
0. alopecurus has sometimes banded anus or spots on its radial shields, and
O. echinata occasionally carries a large live-sided patch on the back of the
disk. Dr. Ljungman considers it a well-defined species, and it should at least
be provisionally admitted on such excellent authority.

Ophiothrix Liitkeni is a deep-sea form, dredged by Wyville Thomson in 374:
fathoms S.W. of Ireland. I examined the original at Copenhagen. It differs
from others of which 0. fragiiis is the type by having high rounded arms,
short, thin arm-spines, and minute spines on the upper arm-plates. The indi-
vidual was large, and, in its dried state, was light-colored with red mottlings.

This completes the list; and it remains to consider its proper divisions.
Dr. Liitken* admits (1) a northern species, Asterias fragiiis Abgd., of
which he has specimens from Iceland, North Norway, Denmark, North Sea,
Spithead, and British Channel. With this he includes, as a variety, 0. penla-
phyllum ; and as a variety of the young, O. echinata Mull. & Trosch. from
Naples. (2) Ophiothrix quinquemaculata, also from Naples, the same which
has just been noticed. (3) Ophiothrix alopecurus (echinata Ltk.), from
Trieste, also noticed and illustrated above. (4) Ophiothrix Liitkeni, dredged
by Professor Thomson in 374 fathoms, off Ireland. On 0. lusitanica, maculata,
and rubra Dr. Liitken has no opinion to offer. In support of his view that
all the northern forms are one species, he says pertinently. '"That the animals
living in the North Sea should afford such differences — that one form should
belong to Scandinavia, the other to Britannica — is hardly credible. That
would be in opposition to all analogy from all other inhabitants of that sea,
which are of course the same on both sides."

I have but one objection to this division: it is not possible to include
0. echinata as a variety of 0. fragiiis. An examination of the upper arm-
plates in two specimens of equal size will satisfy the observer that 0. echinata
is an adult, while an O. fragiiis of the same diameter is not even half grown.

Dr Ljungman, who is naturally more inclined to see specific differences,

* In a letter, December, 1S72.

MUSEUM OF COMPARATIVE ZOOLOGY. 249

recognizes a greater number, as follows:* (1) Ophiotkrix fragilis, the same
as understood by Liitken, but in a narrower limit ; Denmark, Norway, etc.

(2) Ophiothrix quinquemaculata, as distinguished by Liitkea ; Naples.

(3) Ophiothrix alopecurus Mull. & Trosch. from Trieste, which Mr. Ljung-
man identifies with 0. fragilis M. T., tenuispina Sars, and echinata Ltk.

(4) Ophiothrix metadata Ljn. ; Josephine Bank, 120 fathoms. (5) Ophio-
thrix pentaphyllum, common on south coast of England and west coast of
France; identified with Asterias pentaphyllum of Pennant, and included by
Liitken in his 0. fragilis. (6) Ophiothrix lusitanica Ljn. ; as already de-
scribed above ; Portugal. Dr. Ljungman had not seen 0. Lutkeni, and
therefore does not refer to it. It is sufficient to prove the difficulty of the
subject, that, of the two European authors best qualified to judge, one recog-
nizes four and the other seven species.

Combining the information from these authorities with personal observa-
tion, I am inclined to divide the species as follows : —

1. Ophiothrix fragilis Dub. & Kor.
A sterias fragilis Ablg.

Ophiothrix Rammelsbergii Mull. & Trosch.

Ophiothrix alba (?) Grube.

Ophiothrix fragilis Sars,Ltk. (pars). Denmark, Norway, Iceland, Faro Isl's.

2. Ophiothrix quinquemaculata Mull & Trosch.
Asterias quinquemaculata Dell. Ch.

Ophiothrix quinquemaculata Ltk.

Ophiothrix echinata (?) Ljn. (non Mull. & Trosch. nee Ltk.). West Coast
of Italy.

3. Ophiothrix echinata Mull. & Trosch. (non Ltk. nee Ljn.).
Asterias echinata Dell. Ch.

Ophiothrix fragilis (car. mediterranea, juv.) Ltk.

Ophiothrix rubra (?) Ljn. Algeria, west coast of Italy, Adriatic, Egypt.

4. Ophiothrix alopecurus Mull. & Trosch.
Ophiothrix fragilis (?) Mull. & Trosch. {non Asterias Ablg.).
Ophiothrix fragilis (car. tenuispina) .Sars.

Ophiothrix echinata Ltk. {non Mull. & Trosch. nee Ljn.). North Adriatic.

5. Ophiothrix pentaphyllum L.jx.

Asterias pentaphyllum ; varia ; aculcata ; haslata ; fssa ; nigra; Pennant

{teste Ljungman).
Ophiothrix rosula Forbes.
Ophiothrix fragilis Ltk. (pars). South coast of England, west and north

coast of France, Madeira (?).

6. Ophiothrix lusitanica Ljn.

Ophiothrix rubra (?) Ljn. Northwest coast of France, Portugal, Naples.
* Letter, March, 1873.

250 BULLETIN OF THE

7. Ophiothrix maculata Ljn\ Josephine Bank, 120 fathom?, off Portugal.

8. Ophiothrix Lutkeni Wvv. Tiioms. Southwest of Ireland, 374 fathoms.
Of these species, I consider the first four and the last as well marked ; the re-
maining three, perhaps, need to be illustrated by more specimens and localities.

It may be proper to add, that the foregoing critique is simply intended as a
guide to the naturalist who has sufficient specimens of these species for study.

Astrophyton cacaoticum sp. now

Plate VI., Figs. 1-3.
Special Marls. — Disk and arms essentially smooth with only a few micro-
scopic grains. Radial ribs high, narrow, and well-marked. Arms unusually
slender and forking close to the disk; with hook-bearing ridges feebly raised
above the general surface. Five madroporic bodies.

Description of a Specimen (dried). — Diameter of disk 30 mm. Arm forked
close to disk. Width of arm C mm. ; of each fork 3 mm. Each then con-
tinues as a slender, slowly tapering main trunk, throwing out side branches
on alternate sides, and these branches in like manner throw out side twigs.
Distance from first. fork to second 1 7 mm.

" " second

" " third

" " fourth

" " fifth

" " sixth

" " seventh

" " eighth

" " ninth

" " tenth

" " eleventh

" « twelfth

" " thirteenth

Total, 286 "

Along the upper surface of the arm are scattered microscopic granules,
which are still fewer and more minute than those of the disk ; on an alco-
holic specimen these granules would doubtless be invisible. Beginning near
the mouth there are two pointed tentacle-spines, 5 mm. long, on each pore,
beyond the first fork usually three; and these continue nearly to the tips of
the arm, where they are replaced by the hook-bearing ridges, here composed
of a double ring of large prominent grains encircling the arm, and with but
a short space from one ridge to the next. Each grain bears a minute simple

] k. Except at the tip of arm, the hook-bearing ridges are small and

low, and separated from each other by a considerable smooth space. The

" third

20 "

" fourth

21 «

" fifth

20 "

" sixth

20 "

" seventh

25 «

" eighth

25 "

'• ninth

25 "

" tenth

27 "

" eleventh

20 "

" twelfth

22 "

" thirteenth

12 "

" end

32 "

MUSEUM OF COMPARATIVE ZOOLOGY.

251

hooks are found in greater or less number quite to the base of arm. Disk
absolutely naked in the interbrachial spaces, except along its edge above,
where, as well as on the high narrow radial ribs, there are minute grains,
about five in the length of a mm. In the space round the mouth, below,
there are a few scattered microscopic granules, which in an alcoholic speci-
men must be invisible. Five small, narrow madrcporic bodies, placed one in
the inner angle of each interbrachial space and close against the line of sep-
aration between the under and upper surfaces. Mouth and tooth-papillae
sharp and spiniform, arranged in an irregular clump, above which appear
three or four teeth, irregularly superimposed, with a striated surface and a
rather wide curved cutting edge. Genital openings 2.5 mm. long, situated
1.5 mm. inside the outer end of the radial rib. Color, chocolate-brown.

Guadeloupe ; 20 fathoms.* A specimen in the Garden of Plants and an-
other, by exchange, in the Museum of Comparative Zoology.

Astrophyton nudum sp. nov.

Plate VI , Figs. 4-5.

Special Marks. — Xo tentacle-scales on pores. Disk and arms quite
smooth ; the latter ringed with faint lines, which, magnified, are seen to be
rows of minute conical papilla?. One large madrcporic body.

Description of a Specimen. — Diameter of disk 44 mm. Length of arm
about 375 mm., as follows :

fork to second 10 mm.

" " third 12 "

Distance from first
" " second

third " " fourth 16 "

fourth " " fifth 16 "

fifth " " sixth 17 "

sixth " " seventh 20 "

seventh " " eighth 20 "

eighth " " ninth 19 "

ninth " " tenth 22 "

tenth " <: eleventh 20 "

eleventh " " twelfth 20 "

twelfth " " thirteenth 20 "

thirteenth " " fourteenth 20 "

fourteenth " " fifteenth 19 "

fifteenth " " sixteenth 20 "

sixteenth " " seventeenth 20 "

seventeenth " " eighteenth 20 "

eighteenth " " nineteenth 14 "

nineteenth " " twentieth 1 7 (tip broken).

Total,

312

252 BULLETIN OF THE

Width of arm at disk 16 mm. ; width of first fork at its base 6 mm. ; at
the fifth branch 3.3 mm. ; at the fifteenth branch 1.7 mm. ; at the twentieth
branch 1.2 mm. General surface of arm smooth, without spines or grains;
a lens shows a fine network of cross-lines which make an ill-defined mosaic.
Each joint is marked by a minute ridge placed in a sunken line and running
over the top of the arm and ending on either side near the tentacle-pore.
These ridges (about 2 mm. apart at base of arm) consist of minute papilla?
not more than .3 mm. long, sometimes in a single row, but more often in zig-
zag or alternating order, so as to make an almost double row. Each papilla
consists of a hook covered with a thick sheath of skin, and either mounted on
a very small base, or else sitting directly on the arm. Approaching the tip
of arm, the hook-rows become more and more annular, till, on the fine twigs,
they completely encircle the arm in a single row. In this way one or two
hooks nearest the tentacle-pores often are prominent, so as to appear like
tentacle-scales, though really there are none. Disk essentially naked,
although, when partially dry, scattered microscopic grains are seen. Radial
ribs regular, rather high, with the outer end cleanly cut off; they are covered
with a smooth, close, fine granulation ; about eight grains in the length of a
mm. Mouth-papilla?, teeth, and tooth-papilla? all flat, spiniform, and similar;
about twenty-one in all ; those that represent teeth are about three, and are
longer than the others. One large madreporic body at the inner angle of
interbrachial space, close against the line of separation between upper and
lower surfaces. Genital slits 5 mm. long, and lying under outer ends of
radial ribs. Color, in alcohol, above, yellowish brown ; below and at ends
of twigs, much lighter.

A specimen from Philippines, by Semper.

Catalogue of the OrmuRiDiE and Astrophytid.e, collected by Prof. C.
Semper, and now belonging to the Museum of Comparative Zoology.

Ophiomastix annulosa Mull. Tr.

" mixta Ltk.

" fiaccida sp. nov.

Ophiocoma scolopendrina Agas.

" erinaceus Mull. Tr.
" brevipes ? Peters.
<« (i

Ophioplocus imbricatus Lym.

Number
of specimens.

5 Philippines.
2 Pelews.

1

Pelews.

13
4

Philippines.
Pelews.

1

Pelews ?

14

8

Philippines.
Philippines and Pelews.

4

5

Philippines.
Pelews.

9

Philippines.

MUSEUM OF COMPARATIVE ZOOLOGY.

253

Ophiarachna incrassata Miill. Tr.

Pectinura marmorata sp. nov.
" stellata Ltk.
" infernalis Ltk.
<< i< <<

" gorgonia Ltk.

" spinosa Lvm.
Ophiolcpis annulosa Miill. Tr.

cincta Mull. Tr.
Ophiothrix Galatea? Ltk.

" " var. ?

" longipeda Miill. Tr.

(i <<

" " var. ?

" " var. ?

" hirsuta Miill. Tr.

aspidota Miill. Tr.

" Martensi sp. nov.

" cataphracta v. Martens

" purpurea v. Martens

" exigua sp. nov.

" striolata Grube

" plana sp. nov.

stelligera sp. nov.

" pusilla sp. nov.

" elegans ? Ltk.

" triloba ? v. Martens

Ophiogymna elegans Ljn.
Ophiarthrum elegans Peters
" pictum Lym.

It u

Ophiocnemis marmorata Miill. Tr.
Ophiactis sexradia Ltk.

Ophionephthys phalerata sp. nov.
Amphiura (Amphipholis) depressa Ljn

" laevis sp. nov.
Ophiocnida (Ophiophragmus Ljn.) echinata ?
Ophiopeza fallax Peters

Number

of specimens.

1

Pelews.

2

Philippines.

5

Philippines.

2

Philippines.

1

Philippines.

1

Pelews.

2

Pelews.

1

Philippines.

1

Philippines.

1

Pelews.

1

Philippines.

1

Philippines.

7

Philippines.

1

Pelews.

3

Pelews.

1

Philippines.

1

Philippines.

1

Philippines.

1

Philippines.

15

Philippines.

2

Philippines.

1

Philippines.

12

Philippines.

9

Philippines.

4

Philippines.

1

Philippines.

2

Philippines.

1

Pelews.

1

Philippines.

13

Philippines.

4

Pelews and Philippines.

3

Pelews.

o

Philippines.

13

Philippines.

3S

Philippines.

1

Pelews.

1

Philippines.

8

Philippines.

7

Philippines.

;a? 4

Philippines.

1

Philippines.

Number
of specimens.

10

Philippines.

2
12

20
1
3

Philippines.

Philippines.

Philippines.

Singapore.

Philippines.

1

Philippines.

254 BULLETIN OF THE

Ophioglypha sinensis, Lym. var. ?

Ophionereis dubia ? Lym.

Ophiopsammium Semperi sp. nov.

Ophiothela isidicola Ltk. (young)

Astrophyton asperum Agas. (young)
<< it

' ' nudum sp. nov.

In all, forty-five species, whereof eleven are new. These possess great
value as illustrating the fauna of the shallower waters about the Philippine
group, because Professor Semper passed several years in that region, and
searched diligently for animals of all sorts. On the whole, this collection
faithfully represents the fauna of the great ocean, although some rather com-
mon species, notably Ophiocoma Valencies and 0. pica, are missing. This
well may be, because species are often thus lacking in corners of faunal
regions, e. g. Ophiura brevicauda, abundant at St. Thomas, is almost wanting
in Florida. Of Astrophyton, besides .1. asperum, there is a new and beauti-
ful species, A. nudum, which, with A. clavatum and A. verrucosinn, make
four for the great ocean. To these may perhaps be added A. exiguam, in the
Garden of Plants, brought by Peron and Lesueur, in 1803, from the South
Sea. It is apparently a young one, having a disk of only 8 mm. in diameter,
which, with the upper surface of the arm, is granulated finely, and has larger
rounded grains among the smaller. Dr. von Martens need have no doubt
as to the occurrence of this genus in the limits of the Indian Ocean. We
have A. asperum not only from China and the Philippines, but also from the
Straits of Malacca, brought to the Garden of Plants by Eydoux, in 1832.
A. verrucosinn rests on the authority of a specimen from "Indian seas," in
the Garden of Plants, which may well be good, since such species as
Opldocnemis marmorata are found from Port Natal on the south to the Philip-
pines on the north. It seems in every way probable that deep dredging will
bring up not only plenty of these species, and of others like Trichaster
jMlmiferus, but also additional forms, in the neighborhood of Astromorpha.

Homologies of Chewing Apparatus in Opiiiut.id;e.

The skeleton of an Ophiuran within the circle of the disk (PI. VII.) con-
sists of the line of arm-bones, jointed one on the other, like vertebra? ; the
genital plates (Fig. 13, o) ; the radial shields (/) ; of certain irregular pieces
arranged along the margin of the disk (Figs. 5 and IS, s) ; and, finally, of the
strong forked pieces (Figs. 5, 11, 13, 18,/) which form the five angles of the
mouth, support the teeth, and thus make up the chewing apparatus. It is

MUSEUM OF COMPARATIVE ZOOLOGY.

255

agreed that these forked pieces (/*) are in some way made from the division
of an arm-bone on its median line, and the swinging of each half sideways
till it meets and is soldered with the corresponding half of the neighboring

arm-bone. To understand this, an arm-bone must be described in some
detail. Each one then is, near the base of the arm, essentially the same as
its fellows. Its inner surface (Fig. A) has, above, a broad umbo (1), below

JTj.C.

which are two smaller knobs (2) standing on each side of a socket (3) ; still
lower, and quite on the sides, are two large depressions (w) for attachment
of muscles; above (T) and below (/) are notches for the upper and lowei

25G

BULLETIN OF THE

canals, of which the latter has been more studied and is known to carry a
nerve and a water-tube. The outer surface of the arm-bone (Fig. B) pre-
sents, above, a hollow (4) in which rests the umbo (1) of the next piece be-
yond it; below are two depressions (5) into which fit the knobs (2), and
between which is a peg (G) fitting in the socket (3). On each side of the lower
edge is a triangular swelling (/•), which is the outer wall of the tentacle-socket.
Seen from above (Fig. C), the upper longitudinal canal (l') divides the piece
in two, leaving on either side an elongated triangular surface on which rest
the upper arm-plates. On the outer side may be seen the upper surface of
the hollow (4) and the articulating peg below it (6) ; and within is the upper
surface of the articulating umbo (1). A view from below (Fig. D) shows the
lower longitudinal canal (/) ; then without is the articulating peg (0) and the
two sockets of the tentacles (r) ; within are the great lower muscle-fields («■),
the two articulating knobs (2), and the socket (3) for the articulating p< :g.
(For detailed views see PL VII., Figs. 7-10.) From the way in which the

joint is held by the umbo above and the peg below, a vertical motion of the
arm upward is difficult where these parts are well developed, while the
lateral motion is comparatively a free one. As a fact, the chief motion in
the living is a lateral one, and only certain species roll their arms in a verti-
cal plane, and this rolling is downward and not upward; the umbo must then
slip outward and downward, while the peg must press deeper in its socket
(Fig. G, lettered like the others).

A great modification is to be seen in the bones near the tip of the arm,
which are much elongated, and are quite different in detail of structure.
Along the upper surface runs a very wide and deep longitudinal canal
(Fig. E, t') ; from the outer end projects a forked process, which is the ar-
ticulating peg (6) ; at the inner end may be recognized the articulating knobs
(2), and the socket (3). On the lower surface may be seen the same parts
(Fig. F) and the corresponding canal (/), which is slightly marked. (See
also Fl. TIL, Figs. 1G, 17).

MUSEUM OF COMPARATIVE ZOOLOGY. 257

To return now to the chewing apparatus, each angle of the mouth has a
supporting skeleton (Fig. C) in the form of a V, at whose apex is the jaw-
plate (e). As has already been said, each side of this V is composed (wholly
or in part) of the halves of one or more arm-bones greatly modified, and
called collectively mouth-frames (PL VII., Fig. 5,/). It has generally been
assumed that there was only one modified arm-bone in each half of a
mouth-frame, but plainly there must be two, because there are two tentacle-
sockets (r, r'), in which are lodged the so-called mouth-tentacles ; and in
no Ophiuran or Astrophyton is there ever more than one tentacle, on
each side, to every joint or arm-bone. When the mouth-frames are care-
fully examined, especially if boiled in potash, there is seen to be a line
or suture between the wide outer part (/) and the narrow inner point (c).
The suture runs nearly vertically through, or a little outside, the hollow
for the nerve-ring ('0. and, in some genera, as Ophioglypha, this inner
point (r), called the jaw, is easily detached from the outer portion (/),
which is more properly the mouth-frame.* This jaw has no tentacle, and
is regarded by Midler as an interambulacral piece, which is soldered with
its fellow from the side ; and on the angle or point thus made is fixed
the jaw-plate (c), which belongs to the skin formation, and which in turn
supports the teeth (<l"). It is in Ophiolhrix that the homology of mouth-
frames with the innermost arm-bone may most clearly be seen. In Fig. C,
which is a diagram of the innermost arm-bone and of the mouth-frames seen
from above, it is evident that the former is split nearly to its outer edge, and
that its halves arc turned sideways to meet their fellows from the next arm.
The angles 7, 8, 9, correspond in the two pieces. This upper portion of the
mouth-frame must be considered the first arm-bone having its own tentacle-
socket (Fl. VII., Fig. 13,V). The second arm-bone must be placed directly
below the first, and so intimately soldered with it as to form one ; it is pro-
vided with its tentacle (?•), which is the second mouth-tentacle. On this
view, each side of a mouth-frame would consist of three pieces, to wit, the
first and second arm-bones and an interambulacral piece. All, however, is a
theory, based on the position of the tentacles, and needs demonstration from
embryology.

The skin formation remains to be considered. It is usual to make two
distinct divisions, namely, the skin proper, which includes the arm-plates and
the plating or scaling of the disk; and the skin appendages, which arc spines,
grains, and stumps. Great weight is therefore given, in classification, to
these parts ; but, morphologically, they are all the same, — a fact which may be

* See J. Miiller iiber den Bau der Echinodermen, 1854, Plate VII., Pig. 6,/.
This paper, with that of Gaudrv, Pieces Solides ehez les Stellerides, Annales des
Scion. Nat., 1S51, p. 339, are the must important for the subject.

vor.. in. 17

258 BULLETIN OF THE

illustrated by the arm-plates and spines. If we examine the broken end of
an arm which is repairing, and where new joints arc rapidly forming, we shall
see that the tip is a mere tube (Plate V., Figs. 1, 2, 3, 4. Compare, also,
•the figures of young Ophiurans given by Miiller). This tube is a calcareous
network, filled and covered by the secreting tissue, or, as it may be termed,
the skin. There seems, then, to be no beginning of an internal arm-bone, —
nothing hut this open calcareous tube. Immediately, however, there appear
annular strictures round the arm, marking the future joints, with a sunken
longitudinal line, or even s'it, above and below, dividing the tube into two
side arm-plates. This embryonic stage is partly persistent in some genera ;
e. g. species of Ophiomusium, which have no under arm-plates on most of
the joints. Then appear on the central point of juncture, above, clusters of
grains, which, in time, grow into upper arm-plates, and a similar process fol-
lows for the lower arm-plates. On the lower surface may lie seen (Fig. 1),
on the terminal joints, a little flap on each side ; this flap grows more acute
and rounded, becomes separated from the side arm-plate, and ends as a true
arm-spine. So that, in this species of Pcctiiutra, beginning with a tube
of calcareous network, covered by its secreting skin, we end, at the base of
the arm, by the complex assemblage shown in figures 5, G, 7 ; and all these
parts are merely different growths and divisions of this same network.

The same is true of the various divisions of the upper arm-plates and their
supplementary pieces, explained in Plate V. The network may send its
branches from its edges or from its upper surface, and these branches ma)
remain connected, and thus enlarge the plate; or they may be separated, and
make spines, tentacle-scales, and supplementary plates. It is the same with
other Eehinodermata, and the process is simply illustrated by the growth of
a young spine in Ophiothr'uc (PI. III., Figs. 4-7). The mouth-parts make no
exception. The jaw-plate (PI. I., Fi'_r. 4, e) is a skin-plate, and supports
another skin-plate, the tooth ('/"), which has been separated from it. The
peculiar papilla; of Ophior/li/pJia, which embrace the second mouth-tentacle
(7"), are, on one side, carried by a peculiar piece attached to the innermost
under arm-plate. This piece is only an enlarged outer mouth-papilla (so
called), on whose edge have formed these additional papilhe (sec the mouth-
papilhe described under Pectinura marmoratci). A similar process gives all
the variety in the lamelhe of the stony Polyps. "What is true of the arm-
plates is true of the skin of the disk, whose scales may lie traced from the six
primary plates first formed on the hack. On these scales may be developed
spines grains, or stumps, just as on the arm-plates. The strict morphologi-
cal connection of all these parts should warn us not to distinguish them too
emphatically, and not to give them too great a value in generic distinctions,
especially those minute papilla; which form the armature of the mouth. As
to what is provisionally called the skeleton, it is well to remark that Gaudry

MUSEUM OF COMPARATIVE ZOOLOGY. 259

is rifht in considering the arm-bones or disks as parts whose homology is
obscure ; certainly they are not properly ambulacral plates, as Job. Midler
thought, because the tentacles, with their water-system, lie above the ambu-
lacral plates in starfishes, whereas in Ophiurans they lie below the arm-bones
and above the under arm-plates, which latter are plainly the true ambulacral
plates, not only from their position, but from their early formation, which
corresponds to that of the same parts in the starfishes.

2G0 BULLETIN OF THE

EXPLANATION OF PLATES.

Note. — Of these seven plates, all printed by photolithography, the two first are by the Helio-
type process of Boston ; the five la^ by the Alberttype, of New York. Although they have the
advantage of giving exactly the outlines of the original india-ink drawings, they are very inferior to
the drawings themselves ; and not only those, but all similar plates I have seen, are faulty in their
blurred outlines and their uneven and spotted shading Their general effect is that of a lithograph
from a worn stone. Doubtless the process will, before long, be perfected ; but at present it lacks
much.

PLATE I.

Diagrams.

To make clear the terms commonly used in describing Ophiurans, and to show
the varied forms of the parts to which these terms apply, there are given two dia-
grams, Figs. 1 and 2, representing the under and the upper surface of a disk, with
the bases of arms. Each surface is divided into live equal sections, exhibiting the
types of as many genera; while the oases of the arms may either correspond, or
may belong to others ; so that, in the two- diagrams, there are nine genera, as fol-
lows : A, Ophiura ; 15, Ophiocoma ; C, Qphiomyxa ; 1), Ophiothrixj E, Ophioyly-
])ha ; F, disk of Amphiura; G, arm of Ophiopsammium ; II, arm of Hcmicuryale ;
I, arm of Ophiomusium.

Fig. 3 is a diagram of one of the five angles of the mouth, seen diagonally from
below, to show the relations of the chewing apparatus, mouth-tentacles, mouth-
shields, and jaws.

Fig. i is a diagram of the mouth parts in Ophioghjplia, showing half of an an-
gle, with the under arm-plates of two joints.

To these figures the same lettering is applied.

a. Scutum buccal V ; mouth-shield ;' nitindschild ; plaque buccale. This plate
is always present, though sometimes quite shrouded by a thick skin (Ophiomyxa).
Sometimes it takes on a great development, running far out into the interbrachial
space (some species of Ophioghjplia) ; in Ophiarachna it has a small supplement-
ary piece lying outside of it. One of the five shields is the madreporic, and is
connected with the stone-canal.

b. Scutclla adoralia ; side mouth-shields. These are lettered in Figs. 8, 4, and
will be seen inside the point of the mouth-shield in Fig. 1. Often they may be
covered either by thick skin (Ophiomyxa) or by granulation (Oj)hiura). Their
size is considerable in Ophioglyjrfia (Fig. 1, E), but in' Ophiothrix they are small
and narrow (Fig. 1, ]>). Usually their outer end rests against the innermost side
arm-plate ; but in some species of Ophiactis they extend farther, and touch their
outer ends at the outer corner of each mouth-slit ; thus forming an unbroken ring
round the mouth.

c. Scutclla urn] in ; jaws ; mundeckstiick (sometimes included in mouth -frames).
These are the only pieces of the skeleton not covered by the tegument or its

MUSEUM OF COMPARATIVE ZOOLOGY. 261

plates. They are lettered in Figs. 3, 4, and may be seen in Fig. 1, within the
side mouth-shields, against which they usually press, though in Ophiothrix they
are separated by an indentation. Sometimes they are covered by the skin or by
granulation (Opkiomyxa, Ophiura).

d. Papillce orales vel marginales; saumpapillen ; papilles calcaires de labouche ;
mouth-papillae. These run along the lower edge of each mouth-angle, and the
greater part, or the whole, rest directly on the mouth-frames ; in a whole group
or genera they are wanting (Ophiothrix, Ophiocnemis, etc.). Ophiomyxa has them
under the form of little comb-like lobes. They are close and numerous in some
genera (Ophiocow.a), and confined to a single papilla on each side in others (Hcmi-
pholis). They run diagonally upward in some species of Ophioglypha. See also
under tentacle-scales.

d". Papilhc dentales ; zahnpapillen ; tooth-papilla?. (Fig. 3, d', and the point
of the mouth-angle of Ophiothrix in Fig. 1.) This group of papilla?, lying just
under the teeth, is most developed in Ophiothrix, well marked in Ophiocoma, but
cpuite wanting in Ophiura.

d". Dcntcs ; zahne ; teeth. (Figs. 3, and 4 d", and the point of the mouth-angle
in Ophioglyplia and Ophiura in Fig. 1.) These, like the tooth -papilla?, are always
carried by the jaw-plate, and are never wanting among true Ophiurans. In some
genera, they descend to the level of the lower margin of the jaw-frames (Ophio-
glypha, Ophiura, Ophiopeza), in others they are supplemented by tooth-papilke
(Ophiocoma, Ophiomasti.r, Ophiothrix)', and in Ophiomyxa they have the form of
comb-like lobes, resembling the rest of the chewing apparatus.

e. Tones angularis j maxiller ; jaw-plate. (Figs. 3 and 4.) This is a narrow
calcareous plate, running vertically along the inner point of the jaws, with little
hollows in its surface, to which the teeth are bound by small muscles. It is
a part which always exists, and is made up of several pieces, which are often
separable.

h. Scutella ventralia; bauchschilder ; plaques ventrales du bras; under arm-
plates. (Figs. 1 and 4, //.) They differ extremely in size and form ; being always
minute and more elongated at the tin of the arm, and in some genera they con-
tinue subordinate (Ophioglyplia, Ophiomusium), while in others they widen and
make a broad continuous strip (Ophiura, Ophiarachna, Ophiocoma). Some genera
have them covered by a thick skin (Ophiopsammium, (Jphioscolcx), as also Ophio-
myxa, where they divided, lengthwise, in two.

i. Scutella lateralia : plaques laterales du bras ; side arm-plates. These may
be considered the fundamental covering of the arm, for they alone surround
it at the tip; and in one genus (Ophiomusium) they so continue nearly to its
base, almost wholly excluding the upper and under plates (Fig. 2, I i). Other
genera, like Ophioglyplia (Fig. 1, F i), have them persistent on the lower sur-
face of the arm, but replaced above by the upper arm-plates. Coming to forms
like Ophiocoyna, they are widely separated (Fig. 1, B) above and below. Fi-
nally,, in the extreme case of Ilcmicuryalc (Fig. 2, H i) the side arm-plates are

262 BULLETIN OF THE

reduced to small bead-like projections. Then there are two different forms, —
the ridge, which stands out free of its neighbors, and bears the spines nearly
at right angles to the axis of the arm (Fig. 1, B D), (Ophiocoma, Ophiothrix,
Ophiacantha) ; and the flat, which clings close to the surface and overlaps the
next plate beyond (Fig. 1, A E ; Fig. 2, A I), (Ophiura, Ophioglypha, Pccti-
nura). ' These bear the spines on their outer edge, and lying close to and par-
allel with the arm. In consequence of two such modes of structure, Ophiurans
are divided into supple-armed and stiff-armed. The ridge-like plates give space
for a more or less free lateral motion, while the flat and overlapping ones impart
rigidity. If a living Ophiothrix be placed side by side with an Ophiura, the
former will wriggle briskly along the bottom, while the latter may lie quite torpid,
or only slightly bend its rigid arms. Dr. Graeffe told me that one of the most
singular spectacles he had seen was an Ophiothrix longipeda swimming free, and
with its five immensely long arms in rapid and perplexing motion. Ophioglypha
has more mobility than Ophiura, but its way of lifting itself along by two of its
arms, as described by Professor Mobius,*' is very different from the lively squirm-
ing of the arms of an 0})hiothrix. Even in genera which are clothed by a thick
skin, such as Ophiomyxa, the side arm-plates will be found well developed under-
neath.

j. Scutella dorsal ia ; riickenschilder ; plaques dorsales du bras; upper arm-
plates. Like those of the lower surface, they may either remain very small, as
when they first appear at the arm-tip (Ophiomusium, Fig. 2, I/), or may develop
into wide scales, covering the whole upper surface (Ophiura, Fig. 2, A/; Ophi-
arachna). It is these plates that are specially liable to multiplication, either by
breaking up mechanically, or by the addition, in various ways, of supplementary
pieces (see p. 267). The extremest case is furnished by Eemieuryale, where the
original plate becomes lost in a mosaic of additional pieces (Fig. 2, H j).

I. Scutella radialia ; radialschilder ; plaques radiales ; radial shields. Not
properly a part of the general scale covering of the disk, and belonging rather to
the interior skeleton through their connection with the genital plate, these shields
are an exceptional feature, and one that never is wanting, although sometimes hid-
den by thick skin {Ophiomyxa, Fig. 2, C), or by scales and granules (Ophiura,
Fig. 2, A) ; sometimes very large, with their inner portion buried by scales (Ophi-
othrix, Fig. 2, D; Ophioglypha, Fig. 2, E), or, again small and narrow (Amphi-
ura, Fig. 2, F).

771. Radial scales sometimes exist as large scales just next the outer end of the
radial shield. In OphioaJypha they serve to support an arm-comb of small papil-
lae (Fig. 2, E m).

n. Genital scales are largely developed in some genera, where they bound a part
of the body wall of the genital opening (Ophioglypha, Fig. 1, E n), and even
pass upwards and arch over the base of the arm (Ophiothrix, Fig. 1, D n).

p. Spina; brachialcs; arm stacheln ; piquants ; arm-spines. It has already been

* Schriften des Naturw. Vcreins fur Schleswig-Holstein, 1. 179, "1873.

MUSEUM OF COMPARATIVE ZOOLOGY. 263

shown that those spines may stand either on ridges at right angles to the length
of the arm (Fig. 1, B C D) or on the outer edges of the side arm-plates, parallel
to it (Fig. 1, A E). In the former case the wider diameters of the spines are
upward, like the paddles on a wheel {Ophiothrix) ; in the latter, the spine has its
edge upward as if it had heen revolved 90° on its own axis. The variations in
these little organs are almost endless ; extremely long, thin, glassy, and thorny
[Ophiothrix Sucnsonii), very small, thick, opaque, and smooth (Ophiomusium
iburneum) ; rounded and tapering (Ophiocnemis m.armorata) ; flat and of even width
{Opliiura cinerea); covered with thick skin {Ophiomyxa flaccida) ; naked (Ophio-
coma echinata); club-ended {Ophiomastix annulosa); armed with hooks (under
spine in some species of Ophiothrix). The importance of the arm-spines depends
partly on their length, thickness, and number, and partly on the extent of the side
arm-plates. When these last occupy a large part of the circumference, and the
spines also are long and numerous, the arm proper is almost hidden and resembles
a round bristle-brush {Ophiomyces frutectosus).

q. Papillce ambulacrales ; tentakel schuppen ; papilles tentaculaires ; tentacle-
scales. Although these small organs are strictly homologous with, and even some-
times similar to, the arm-spines {Ophioglypha), they nevertheless have a different
function, to wit, that of covering the tentacle drawn in ; and usually they differ
in form and position from the nearest arm-spine. Thus in Ophiothrix, while the
lowest spine is often hooked, the tentacle-scale is minute and tooth-like ; in
Ophiopsila it is like a spatula ; Ophionereis has a single circular one ; Ophiocoma
one, or two, of a shape more or less oval. In Opliiura the upper scale laps over
the base of the lowest arm-spine, though in most genera the two are separated.
One of the tentacle-scales is often carried by the under arm-plate, either on its lat-
eral edge {Amphiura) or on its surface, making a continuation of the line of arm-
spines {Ophiomyces). Although some species of Amphiura have two scales, one
on the edge of the under arm-plate and the other at right angles on the edge of
the side arm-plate, others have no scale at all, — a want shared by many species
{Ophioscolcx ylacialis, Ophiomyxa, Ophiopsammium). The two pairs of mouth-
tentacles are not neglected in this respect (Figs. 3, 4, q' q") ; but, in almost all
genera, are furnished with one or more tentacle-scales (q q). That of the first,
or upper, tentacle sits on a little ridge of the jaw. It is this, when largely devel-
oped (as in some species of Amphiura), that has been described as a peculiar
papilla situated high up in the mouth. Ophiothrix is one of the few genera that
lacks this scale, which exists even in Ophiomyxa. While these scales are closely
homologous, on the one side, with arm-spines, they also are, on the other, with
certain mouth-papilla?. Of this there is an illustration in Ophioglypha (Fig. 4),
where the tentacle has a row of scales on either side of it, one row carried on the
edge of the side arm -plate (i), the other on that of the under arm -plate (some-
what displaced in the figure). It is a greater development of what occurs in the
Amphiura: with two scales (PI. IV. Fig. 20). The homologue of the side arm-
plate which belongs to the innermost under plate is the side mouth-shield ; and

264 BULLETIN OF THE

this not only in OphiogJi/pha (5), but in most other genera, bears one or more
tentacle-scales, under the form of outer mouth-papillae ; while what may exactly
be called mouth-papillae (d) are apt to stand on the mouth-frames. This distinc-
tion, however, is only one of convenience as applied to parts really homologous.
Thus in Ophioglypha the two rows of scales belonging to the second pair of
mouth-tentacles stand respectively on a plate which is really an overgrown outer
mouth-papilla (see under Pectinura marmorata), and on the side mouth-shield
(partly, also, on the mouth-frames ?). The first pair of mouth-tentacles, in the
same genus, are duly furnished with their own scales, which stand, as usual, on a
ridge of the jaw. All the complex arrangement just described may be most
clearly seen in 0. Lymani.

The integument which encloses the disk is, properly speaking, covered, or beset,
with calcareous plates, which, in the young, are six in number, to wit, the largest
in the centre and the other live in a close circle round it, one opposite the base of
each arm. In some genera the secretion of lime is stopped in the disk integument
at an early period ; and, in the adult, nothing is to be found but a few minute
grains buried in the skin, while the general surface of the disk is smooth and
fleshy. Such genera are said to be naked (Ophiomyxa, Ophiarthrum, Figs. 1, 2,
C). Others are equally called naked where a thick skin covers a regular scaly
coat (Ophiopsila). In many genera the disk is plainly covered with plates, which
may be finely imbricated scales (Amphiura, Fig. 2, F) or coarser and thicker
ones irregularly arranged (Ophioglypha, Figs. 1, 2, E), or thick angular pieces
set side by side on the same level, like a mosaic (Ophiolcpis, Ophiomusium). Then
there are genera in which the scale coat is beset, or even hidden, by appendages.
These may be spines (Ophiothrix, Figs. 1, 2, D) ; or scattered grains (Ophiocoma ,
Fig. 1, B) ; or grains so closely set as to completely hide the disk, radial shields
and all, except the mouth-shields (some species of OpJiiura, Fi^s. 1, 2, A). There
is one genus (Ophiopsa7nmium) in which the disk and upper surface of the arms
are covered, first by a smooth integument, and this again by a close granulation.
The radial shields are to be seen in all species in which the scale coat is visible,
and the mouth-shields are never hidden except in a few species which have a very
thick, naked integument.

PLATE II.

Spines and Stumps of the Disk of Ophiothkix.

All the figures are enlarged about thirty diameters.
Figs. 1-14 Oph loth rix cch inata.

" 15-29 " lusitanica.

" 30-36 " pentaphyllum.

" 37-44 " fragilis.

" 45-55 " quinqucmaculata.

1, 2, 3, a spine and two stumps, 0. echinata, original of Mull, k Trosch. 4-9,

MUSEUM OF COMPARATIVE ZOOLOGY. 2G5

Naples. 10-11, showing a stump covered by skin, and another with the skin
partly torn off: Naples. 12-13, stumps from lower interbrachial space of a
young, having a disk of 2.5 mm. The scales of the disk are then visible, and
Fig. 13 shows how one stump stands on each scale : Naples. 14, a short spine
from the interbrachial space below : Algeria.

15-20, (J. lusltanica, Naples; 16- IS being from the young; 15, 19, 20, from
an adult ; the last (20) is a very rare spine. 21 -29, N. W. coast of France ; of
these 22 is the common form ; the rest are more or less rare ; 28 is found near the
edge of the disk.

30-35, O. pcntaphyllum, Isle of Wight ; of these 30-33 are thelbharacteristic
grains and stumps of the centre of the disk ; 34, 35, are its thick columnar spines ;
36, a disk spine : Madeira ?

O. fragilis. 41, stump from a young, with a disk of 5.5 mm. 37, 40, 42, two
stumps and a spine from a disk 7 mm. in diameter. 44, an elongated stump from
the edge of a disk 9 mm. in diameter. 38, 39, stumps from a large specimen
(disk 16 mm.) ; 39 is rare : Denmark. 43, rough grain from centre of disk ; large
specimen : Sweden.

O. qvdnquemaculata. 46, 47, stump and articulated spine from a young, with a
disk of 3.5 mm. : Naples. 52, 53, short and elongated stumps from a disk
7.5 mm. in diameter ; the latter from the edge. 51, articulated spine from a large
specimen. 45, irregular, thickened spine from an adult. 48, 49, 50, short
stumps from the centre of a large disk. 54, an elongated stump from the edge of
the same. 55, a large stump like 49, much elongated : Spezia.

PLATE III.

Growth of Spines, Hooks, and Stumps.

Ojihiofhrix alopcciirus Mull. & Trosch. Ophiothrix rudis sp. nov. Ophiothrix
stclligcra sp. nov. Ophiothrix pusilla sp. nov.

Fig. 1. The fluted form of the long spines which cover the upper surface of the
disk, except the radial shields, in O. alopecurus. These usually are more slender,
with smaller side thorns and only slight appearance of fluting; a]a.

Figs. 2, 3. Minute spines, simple and forked, which sometimes are found sparsely
on the radial shields ; *p-.

Fig. 4. A young arm-spine from near tip of arm, showing a central shaft with
thorns forming on either side, and the holes, along the two lines of juncture, not
yet filled up ; *£.

Fig. 5. A similar spine with the side thorns broken off to show more distinctly
the central shaft ; 7i°.

Fig. 6. Tip of a similar spine more magnified, to show the holes along the lines
of juncture, holes of growth; 1:jQ.

Fig 7. Fragment of a very young spine much magnified, to show the soldering
of a thorn with the central shaft, and the holes of growth.

266 BULLETIN OF THE

Fig. 8. Base of an adult arm-spine, showing the development and close solder-
ing of the side thorns to each other and to the central shaft; sy1.

Fig. 9. Tip of an adult spine magnified, to show the central shaft and side
thorns beginning to form.

Fig. 10. Hook from near tip of arm, corresponding to Fig. 4. At its base are
seen the holes of growth ; ^. For its position with its three arm-spines, see
PL VII., Fig. 16.

Fig. 11. Ophiothrix rudis. A disk-stump; ^

Figs. 12, 13. Under and upper arm-spines from the eighth joint; the latter has
a blue belt round the middle ; -lp-.

Fig. 14. Under arm-plate of the eighth joint; ^.

Fig. 15. Ophiothrix stclligcra. Under arm-plates of the seventh, eighth, and
ninth joints, with one of the hooks ; -y-.

Fig. 16. Upper arm-plate, near tin' disk; xp-.

Fig. 17. Common form of disk-stump; -4"-

Fig. IS. A similar one from above; xfa.

Fig. 19. Rare form of disk-stump; Ha.

Fig. 20. Spine from the interbrachial spare below; x"2.

Figs. 21, 23, 25. Ophiothrix pusilla. Unusual forms of disk-stumps; ^Q.

Fig. 22. The common form of stump on the back of the disk; L\-.

Fig. 24. Stump from interbrachial space below; hp-.

Fig. 26. Stump from interbrachial space near mouth ; —p-.

Fig. 27. Side arm-plate of seventh joint with its eight spines and hook ; -j1.

Fig. 28. Part of an under and a side arm-plate close' to tip of arm, showing the
one small arm-spine and the large hook; much magnified.

Fig. 29. Upper arm-plates of sixth, seventh, and eighth joints, with two side
arm-plates, to show the triangular projection from the latter; \".

Fig. 30. Upper arm-plates, near end of arm, with spines and side arm-plates,
showing the triangular projection from the latter.

PLATE IV.

Ophiothrix plana sp. nov. Ophiothrix Martcnsi sp. nov. Ophiopsammium Scm-
pcri gen. et sp. nov. Amphiura lozvis sp. nov. Ophiocnida (Ophiopihragmus
Ljn.) echinata (?) (longipeda Lym. MS.). Ophiothrix cxigua sp. nov. Ophiothrix
comata Mull. & Trosch. Ophiothrix ciliaris Mull. & Trosch. Ophiothrix
fumaria Mull. & Trosch.

Fig. 1. Ophiothrix plana. Portion of upper surface of disk; nt the centre and
left the skin is partly dried to exhibit the underlying scales which on the right
are covered by the epidermis ; \.

Figs. 2, 4. Different forms of the third arm-spine. Fig. 4 has its epidermis ;
s f

Fig. 3. An upper arm-spine ; ^r.

MUSEUM OF COMPARATIVE ZOOLOGY. 267

Fig. 5. A short disk-stump covered by a thick epidermis ; *^.

Fig. 6. Three disk-stumps, one with and two without epidermis ; ^Q.

Fig. 7. Edge of a side arm-plate near tip, carrying two spines and a double
hook ; *)&.

Fig. 3. Under arm-jilates ; $.

Fig. 9. Ophiothrix Martensi. Hook and arm-spine next above it, with its
epidermis ; sjL.

Fig. 10. Upper arm-spine and upper arm-plates close to disk ; showing the
central blue line with a white line on each side.

Fig. 11. Ophiopsammium Sempcri. Twelfth joint, seen diagonally from below,
showing, on the further side, an extended tentacle and the lowermost arm-spines ;
and, on the nearer portion, a side arm-plate and its spines, foreshortened, and
with tin.' tentacle omitted ; -2/1.

Fig. 12. Part of upper surface of disk, showing the base of an arm, and the
pairs of spines in the interbrachial space ; -f.

Fig. 13. Tooth-papillae and teeth seen from within ; \".

Fig. 14. Longest arm-spine (third) near base of arm ; \".

Fig. 15. Under arm-plates, as they may be distinguished near tip of arm ; 1,Q.

Fig. 16. Joints, seen from above, close to end of arm, showing the beginning
of the granulation, and the embryonic upper arm-plates ; each side arm-plate is
furnished only with two little arm-spines and a strong double hook ; ^

Fig. 17. Granulation of the upper arm near its base ; ■21Q.

Fig. 13. Amphiura lavis. A portion of the disk, from above, with the base
of an arm ; x^-.

Fig. 19. An angle of the mouth, showing the four mouth-papilk~e on each side,
with mouth-shield and side mouth-shields ; -^

Fig. 20. Two joints of arm, near the disk, seen from below ; xy.

Fig. 21. Ophiocnida echinata ? Part of a side arm-plate with its spines ; *-.

Fig. 22. A portion of the disk, from above, with the base of an arm ; ^-.

Fig. 23. An angle of the mouth, with a mouth-shield and the first two under
arm-plates ; f.

Fig. 21. Op7iiot7irix cxigua. Upper arm-plates and bases of the arm-spines,
near the disk ; -\^.

Fiur. 25. The common form of disk-stump ; -"-.

Fig. 20. Hook, standing below the arm-spines : sf-.

Fig. 27. Ophiothrix comata (from the original in the Vienna Museum). An
under arm-plate, with its tentacle-scale ; -]".

Fig. 2^. Two upper arm-plates, with the liases of arm-spines ; J,°.

Fig. 29. OphiotJirix ciliaris (from the original at the Garden of Plants). A
portion of upper surface of disk and two upper arm-plates, showing the distribu-
tion of the fine disk-stumps ; ?.

Fig. 30. A disk-stump, from near the edge ; *£.

Fig. 31. Two under arm-plates.

2G8 BULLETIN OF THE

Fig. 32. Second arm-spine ; 1ih-.

Fig. 33. Ophiothrix fumaria (from the original in the Garden of Plants). A
portion of upper surface of disk, with some upper arm-plates. On either side of

the points of the radial shields appear the upper corners of the genital scales ; \.

Fig. 34. Under arm-plates, with tentacle-scales.

Fig. 35. Arm-spines, near base of arm ; f .

Fig. 36. A disk-stump on its scale, much magnified.

PLATE V.

Formation of Arm-spines, Arm-plates, and Supplementary Pieces.

Pcdinura marmorata sp. nov. ; Hemieuryale pustulata v. Mart. ; Ophioplocus
Esmarki sp. nov. ; Ophiura squamosissima Lym. ; Ophiura cincrca Lym. ;
Ophiopholis aculcata Mull. & Trosch. ; Ophioncrcis dicbia Lym.

Fig. 1. Pectinura marmorata. The tip of an arm undergoing repair. The
point is only a tube of calcareous network, covered by the secreting membrane.
Farther in, this tube is widened and cut transversely by furrows into joints, which
have lobes at their outer edge ; and these lobes grow more sharp and rounded, and
finally become the lowermost arm-spine, p. There are no under arm-plates at this
stage of growth ; i1Q.

Fig. 2. The same from above. Almost the entire surface is occupied by the
side arm-plates, at whose central point of juncture appear little collections of
granules, j, which are to be separated later as upper arm-plates ; ^-.

Fig. 3. The same seen from the side. One joint has two partly formed arm-
spines, p, and the next joints have one each ; ^p.

Fig. 4. The broken arm under repair, showing the old portion, and eighteen
new joints ; of which the rast five are represented in Figs. 1, 2, 3 ; -"".

Figs. 5, 6, 7. Two joints, close to the disk, seen from below, from above, and
from the side, to compare the perfect plates and spines with the young ; \n-.

Fig. 8. Hemieuryale pustulata. Tip of the arm, magnified, showing the tubu-
lar point, the side arm-plate, i, and the young upper arm-plate, j.

Fig. 9. A joint near the tip, seen from above. j>, arm-spine ; i, side arm-plate ;
j, upper arm-plate ; k, supplementary piece.

Fig. 10. Joint farther inward. i, side arm-plate ; j, upper arm-plate ; I; sup-
plementary piece, in addition to which there are now numerous others, smaller.

Fig. 11. Joint about one third out on the arm. k, the great supplementary
piece, which now is larger than the side arm-plate, i, and forms the little cushion
characteristic of the genus. The upper arm-plate can no longer be distinguished
among the numerous irregular supplementary pieces. This is an instance of great
changes in the relative importance of parts in the process of growth.

Fig. 12. Ophioplocus Esmarki. Tip of arm showing the very short tubular
point, followed by a joint which has a small upper arm-plate. The corresponding
plate of the next joint has a longitudinal furrow ; the same is fairly separated in

MUSEUM OF COMPARATIVE ZOOLOGY. 269

two parts on the succeeding joint, and lias a couple of supplementary pieces with-
in. The two parts are still mure diverging on the next joint, j; the large side
arm-plate, ;', comes up toward the median line, and the upper arm-spine p is promi-
nent ; \"-.

Fig. 13. A joint near the middle of the arm. The two parts of the upper arm-
plate, j, arc now completely separated by the intrusion of supplementary pieces,
which have increased to seven ; p, arm-spine ; i, side arm-plate ; \a.

Fig. 14. A joint at the base of the arm. The two parts of the upper arm-plate
j, instead of occupying the top, have been wedged apart by supplementary pieces,
until they are on each margin. The side arm-plates, i, have become small as com-
pared with the supplementary pieces, and the arm-spines,^?, a#no longer con-
spicuous ; X£:

Fig. 15. Opinion squamosissima. Piece of arm near its tip, enlarged. The
outer joint bears a small, simple upper ami-plate ; the innermost one has also a
simple plate, j, outside which are two supplementary pieces, k ; on either side is
a large side arm-plate, ?', bearing a short spine.

Fig. 16. A joint near the base of the arm. The upper arm-plate still occupies
the median line, j ; but the two supplementary pieces, k, have moved from a
position beyond, to one on either side of the upper arm-plate, and still another sup-
plementary piece has been formed in a line with the rest and lying on the margin.
The side arm-plate has become comparatively small, i, and has been crowded
clown on the side of the arm.

Fig. 17. Opliiura cincrca. A joint near base of arm, showing the upper arm-
plate broken in several pieces, as is usual ; f.

Fig. 18. OpTi iopholis aculcata. Two joints near end of arm showing the upper
arm-plate, j, with a row of grains or small supplementary pieces, only along their
outer margin ; f.

Fig. 19. Two joints near the base, showing the same plates, ;', completely en-
circled by a close row of small pieces, which may be double on the sides ; f.

Fig. 20. OpMont reis dubia. A joint near end of arm showing the simple up-
per arm-plate, j : -cr-

Fig. 21. A joint near base of arm, with a small supplementary piece, k, on
each side of the upper arm-plate, j ; 21Q.

PLATE VI.

Astmplnjton cacaoticum sp. nov. ; Astrophyton nudum sp. nov. ; Ophioplocus Es-
inarki sp. nov.; Opliioneplithys pludcrata sp. nov. ; Pcclinura scpi spi-
iiosa Mull. & Trosch. ; Opldomastix ftaccida sp. nov.

Fig. 1. Aatropilvjton cacaoticum (from a dried specimen). A part of the
disk, fmm below, with outline of one fork of the arm, showing the distance of the
branches with approximate accuracy ; \.

270 BULLETIN OF THE

Fig. 2. Tart of upper side 01 disk, showing the slender radial ribs, and the
first fork of the arm ; j.

Fig. 3. The head of a radial rib and part of the interbrachial margin, enlarged
to show the granulation along the edge ; -}.

Fig. 4. Astrophyton nudum. Tip of a twig, seen from below, showing the
continuous line of honks covered by a thick skin;- and, in front of them, two
short, conical tentacles ; -j0-.

Fig. 5. Three hooks, from the double alternating rows of the branches near
the base of the arm. From one, the skin has been stripped, showing the naked
hook ; A,a.

Fig. 6. Ophioplocus Esmarki. Two joints near base of arm, seen from the
side ; 11-.

Fig. 7. Ophioncphthys plialcrctta. Two angles of the mouth, the base of the
arm, and a part of two interbrachial spaces ; \.

Fig. 8. Portion of upper surface of disk, with base of an arm, and radial shields
surrounded by their peculiar wreath of scales ; ].

Fig. 9. An upper arm-plate, with a side arm-plate and its spines, showing the
peculiar thickened spine ; z^.

Fig. 10. Pedinura scptemspinosa (from the original at Leyden). Upper
arm-plates and spines, near disk, showing how the former are usually broken ; *.

Fig. 11. An angle of the mouth enlarged, exhibiting the mouth-shield with its
small supplementary piece and the surrounding granulation.

Fig. 12. Arm-spines, near disk, with the lowest one somewhat longest, and
having its base covered by a tentacle-scale ; j.

Fig. 13. Two under arm-plates within the disk, with the pairs of pores between
them ; -}.

Fig. 14. OpMomastix flaccidci. An angle of the disk and base of an arm. seen
from above, where are seen the peculiar upper spines thickened in different de-
grees; [h

Fig. 15. Two joints from below, showing two tentacles having no tentacle-
scales ; x-

PLATE VII.

Pcdinura infcrnalis. Mull, and Trosch. ; Ophiartlirum piditm~Lym.; Homolo-
gies of the skeleton of the arm and mouth-parts in Ojiltiotltrix, Ophiura, Opliio-
glypha, and Opliiomyxa.

Fig. 1. Pcdinura infcrnalis. Part of upper surface of disk, to show the
characteristic arrangement of the naked plates ; ;.

Fig. 2. Ophiarthrum pidnm. A corner of the disk and three upper Tin -plates,
with the patti rn lines and arm-stri] e ; {'. Specimen from Pelew Islands.

Fig. 3. A tentacle and its scale; i{'.

MUSEUM OF COMPARATIVE ZOOLOGY. 271

Fig. 4. Side arm-plate (third joint from the disk) with its spines, and a corner
of the upper arm-plate ; 5-

Fig. 5. Ophiura hvvis (Mediterranean). The skeleton of the mouth-parts,
and of the arm as far as the edge of the disk, with the genital plates, o, o, in po-
sition, seen from above; f. c, jaws; e, jaw-plate; d", teeth; u, circular canal for
the nerve-ring of the mouth ; r, r', sockets of the second and first pairs of tenta-
cles; q, tentacle-scale of the first pair; /, mouth-frames; r, r (on the arm-bones),
places for the tentacles ; s, one of the supplementary pieces lying along the margin
of the disk, under the skin ; V upper arm-canal. On the left are a few scales of
the disk, in position, with their coat of grains. The radial shields are removed.

Fig. 6. Jaws, one mouth-frame, and two arm-bones in profile, the latter are
separated and turned so as to show part of the top ; lettering as above ; j.

Fig. 7. Inner side of first arm-bone, showing its points, which articulate with
the outer side of the mouth-frames (compare Fig. 6) ; t, lower arm-canal ; h, under
arm-] date ; u; lower muscle-field ; ^.

Fig. 8. Fifth arm-bone, outer side, with tentacle-sockets, r, upper and lower
canal, articulating peg, and depression to receive the umbo of the next arm-
bone; y.

Fig. 9. Inner side of same bone, with the lower muscle-field, w, the articulating
umbo, and the depression below it to admit the articulating peg ; j.

Fig. 10. Same bone from below, with the muscle-field, v, on the inner side,
and the tentacle-socket, r, on the outer, where also is the articulating peg; on the
median line runs the lower arm-canal, t ; j.

Fig. 11. Ophioglypha cileata (Mediterranean). Skeleton of arm and mouth-
parts, to edge of disk, seen from above, with a genital plate, o, in position on one
side, and on the other a radial shield, J, turned aside to show the underlying parts.
On the middle line is the first upper arm-plate, j; on one side of it is the lower
arm-comb, in' ; and, on the other, the upper arm-comb or radial scale, m. Other
letters as above ; y.

Fig. 12. Three innermost side arm-plates, i, forming one wall of the genital
opening. The upper corner of the side mouth-shield, b, supports the inner end
of the genital plate, o, here seen in profile, with the under arm-comb, m! (com-
pare Fig. 11), q, tentacle-scales; f.

Fig. 13. Ophiothrix quinqueinaculata (Mediterranean). Skeleton of mouth-
parts and arm, as far as edge of disk, witli a radial shield and genital plate ar-
ranged and lettered as in Fig. 11, except that the mouth-frames are separated
from the innermost arm-bone ; •}.

Fig. 14. Outer articulating surface of mouth-frames (compare Fig. 13) ; ^.

Fig. 15. Inner articulating surface of t he arm-bone next tie' mouth-frames : ->

Fig. 16. Ophiothrix alopecurus (Adriatic). An arm-bone close to the tip of
the ami, seen from above, with a side arm-plate in position, and bearing a hook
ami the bases of three young spines ; 3f-.

Fig 17. The same bone from below ; 3f-.

272 BULLETIN OF THE

Fig. IS. Ophiomyxa pentagona (Mediterranean). Skeleton of mouth-parts and
arm, as far as edge of disk, seen from above. The radial shields, 7, are turned
upwards and outwards, so as to expose the peculiar forked genital plates, o. v,
stout triangular pieces covering the trench of the nerve-ring. These in Ophi-
oglypha ciliata appear only as thin plates, j, little rudimentary upper arm-plates,
split in two. s, supplementary pieces lying within the skin, along the disk
margin ; {■.

Cambridge, February 15, 1S74.

Ftetei

Piatell.

i 1 3 9 M

1-" 7 8 9

■

PLATE III.

*w

LYMAN .1 DEYKOLLE AD. NAT.

Place IV.

evil S-\; ft*\V»"JTC

0 **
IP

^|I»

^

Lyman &.R.oeiler ad ml.

Plate V.

fllPI Ills ^P» J

Lyman & Ueyr-olle ad n*-t.

I

Plate VII.

a ' %

Lyinan fcRoettor ad naX.

No. 11. — Exploration of Lake Titicaca, hy Alexander Agassiz
and S. W. Garman.

I. Fishes and Reptiles. By S. W. Garman.

FISHES.

The fishes of Lake Titicaca present but little variety. Only two
genera are represented. A month of search and inquiry discovered
but one species of Trichomycterus, a siluroid, and five of Orestias, a
cyprinodont. Though found in great numbers and easily taken, on
account of their inferior cpiality these fishes form but a small propor-
tion of the food of the people living on the shores. The successful
introduction of some superior food-fish would be of very great advan-
tage to the inhabitants of the valley. A more inviting opportunity
for a grand experiment in fish-culture can hardly be imagined than is
offered here in this extensive sheet of pure water, with its numerous
mountain streams, complete isolation, abundance of vegetation, and
vast numbers of animals. The siluroids ai-e more prized by the na-
tives and are tolerably common in the markets. They are too oily to
be considered good. Specimens that weigh a pound are large. As is
usual with the Siluridee, they are carnivorous and most active by night.
The cyprinodonts are much more numerous, but have little to recom-
mend them for the table except the absence of better. Their small
size and numerous strong bones make them very unsatisfactory arti-
cles of food. It is their worthlessness, no doubt, that has fastened upon
several species the epithet Carache as a common name. Immense
schools of these fishes are found in their feeding-places, the great
fields of rushes ("totora") and the beds of weeds which cover the
bottom in the shallower portions of the lake. When young, they are
preyed upon by other fishes and the thousands of birds frequenting
their places of resort ; as they grow older their structure and size
give them immunity. The many bones and scales brought up in the
dredge from the deeper waters along these localities were those of
adults. All the species of this genus are small ; we saw none weigh-
ing more than eight ounces.

Angling and spearing are not practised. The fishing is done with

274 BULLETIN OF THE

pounds, nets, and pots. When compared with the complicated struc-
tures on the coast of the United States, these primitive traps hardly
deserve the name of pounds. They are mere fences, with an opening
near the middle for the pocket, placed across the course travelled by
the fishes. The fence is made of rushes tied close together by one
end along a line of the required length ; this line is fastened on the
bottom by means of stakes, and the buoyancy of the rushes keeps
them upright.

The pot is a short cylinder of open basket-work with one end
rounded and closed, and with a gate in the other, like that of the lob-
ster-pot, which admits the fishes but prevents their egress. Consid-
erable ingenuity is displayed in the structure of these baskets. The
warp is of single stems of a smooth, stiff, wiry grass ; the woof is
made by wrapping several small stems with split straws, making rolls
which are bound to the stems of the warp, on the outside, by passing
one of the straws which bind the roll around each stem at its proper
distance from each other. The spaces in the warp are determined by
the size of the fishes desired ; those in the woof by the strength of the
materials. Such traps are used as are lobster-pots.

The net in common use is a small dip-net with a long handle.
Armed with this the Indian glides back and forth along the beach,
late in the evening when the hungry siluroids come close to the
water's edge to feed, occasionally dropping the net quietly down so as
to cut off its retreat and then with a jerk throwing an unwary fish far
out of the water. It is said that these nets are also used in fishing
by torchlight from balsas.

Eight species of the genus Orestias were described by Cuvier and
Valenciennes (Hist. Nat. d. Poiss., Vol. XVIII, p. 221) as occurring
in tins lake. Of these Cuvieri and Humboldti are synonymous, the
latter being the young of the former, as shown by Dr. G anther ; Mulleri
and luteus are doubtless to be referred to one species ; and, from the
variations in proportions exhibited by our specimens, it appears likely
that one of the stouter forms of Agassizii served as the type of Jussiei.
The description and figure of Agassizii given by Cuv. and Val., from
the common form of a half-grown animal, represents the species most
accurately; for this reason the name is retained and Jussiei placed
as a synonyme. Here also is the place for the 0. Owenii of Giinther.

MUSEUM OF COMPARATIVE ZOOLOGY. 275

0. Owenii of Cuv. and Val. agrees with albus, in its large muzzle and
colors, rather than with Agassizii ; the description is very indefinite;
the type was taken in Lake Urcos.

In the following list of the fishes obtained notes in reference to
value, abundance, etc. are given under the name of each species.

Trichoniycterus dispar Gunth.

Called Suche or Maure by the natives.

It is found in all parts of the lake and the streams entering it. In the
stomachs of a number dissected small molluscs, Crustacea, and fishes were
found. One specimen of fourteen inches contained another of six. The
color is generally brown, reticulated with many narrow irregular white lines ;
sometimes with spots like trout or uniform.

Adults of less than seven inches were taken ; the largest specimen meas-
ured sixteen. The Indians distinguish between small and large, calling the
former Maure, the latter Suche. They are higher priced in the market than
the other fishes of the lake.

Orestias Cuvieri Cuv. et Val.

The full grown is called Omanto and the young Peje Rev. Its market
value is less than that of O. Pentlandi. The bare space on each side of the
dorsal row of scales occurs in all adults.

Dead ones with the stomach everted were frequently met with on the
shores of Titicaca Island. Ten and a half inches is the measurement of the
largest specimen.

O. Pentlandi Cuv. et Val.

Apparently not so abundant as the preceding. This species ranks next
to the Suche for the table. With the bare space on each side of the dor-
sum. One specimen shows the entire top of the head covered by a single
large scale ; otbers have the vertex bare. The largest secured was nine
and a quarter inches in length. Bova is the common name of this fish.

O. Agassizii Cuv. et Val.

One of the several species to which the name Carache is applied. The
most abundant fish in the lake. It is valued less than either of the pre-
ceding.

There is much variation in the proportions of different individuals ; the
depth of body varies from a third to a fourth of the length, exclusive of the
caudal fin. In the adult the colors are uniform dark In-own above, lighter
below ; the young are spotted or banded. Longest specimen eight inches.

276 BULLETIN OF THE

The names Bova and Carache suggest a possible likening of the fishes to
cattle. A scabby sheep is called Carache, though the term is properly the
name of the scab itself.

O. Mulleri Cuv. et Yal.

Common name Carache. Not so plentiful as the preceding. Its angular
head and shoulders, the great width of the latter compared with that of the
body, and the strongly granulated scales serve to distinguish this sjiecies
from Agassizii, and the narrow muzzle and small mouth from albus.

In some individuals the angles of the head are less pronounced than in
others. The colors in different specimens vary from quite light to very
dark. None obtained were longer than six inches and a half.

O. albus Cuv. et Val.
»

Less numerous than the other Caraches. The muzzle is wider, the mouth
larger, and the head longer than in the species preceding. The head has the
aspect of that of Cuvieri. Color light brown, often reddish or yellowish.
None of the few examples taken exceeded six inches in length. In the young
of all the species the striation of the scales is plainly visible ; the granula-
tion on the anterior part of the body in several appears later. A number of
small fishes, less than half grown, which are considered the young of this
species were obtained at Moho. These differ somewhat from the adult ; the
back has a dorsal and two lateral series of spots ; the belly is of salmon
color and is closely covered with scales. The Indian from whom they were
purchased called them Silgo.

The collection from which the foregoing list is made was said by the fish-
ermen to contain specimens of each of the different fishes found in the lake.

Batrachiaxs.

Cyclorhamphus marmoratus Dun. et Bibr.
Specimens were taken at Arequipa and at Vincocaya on the summit,
14,538 feet above the sea-level. The adult from the higher regions is about
half the size of that from nearer the coast, though similar in all other re-
spects.

Cyclorhamphus culeus (nov. sp.) PI. I.

Body large, depressed. Head rounded in front, wide, flattened. Skin
smooth, very loose and baggy, of the entire upper surface glandular, giving
off, when the animal is held in the hand for a little time, a secretion similar
to that of the Amblystomre. Fingers free. Toes rather more than half-
webbed. Frist cuneiform bone forming a slight jn'ominence. Tympanum

MUSEUM OF COMPARATIVE ZOOLOGY. 27/

hidden. Eustachian tubes not apparent. No vocal sac. Tongue broader
forward, free posteriorly. Vomerine teeth, generally present, in two very
small groups between the inner nares. Internal nostrils large, anterior
very small. Color varying in different individuals from uniform olive brown
to brown irregularly spotted with white or so thickly sown with white dots
as to appear gray.

Average length of body about three inches. Two very large specimens
were secured; the bodies of these measure a little more than four inches, from
snout to anus. One of them is smooth, as is usual ; the other lias numerous
prominent nipple-like warts on the flanks from the tympanum backward.

The young of this species and those of marmoratus are very similar. The
adult, however, instead of being erect and trim with prominent palmar and
plantar tubercles as the latter, is much depressed and flattened, and carries
itself more like Pipa ; it has a great baggy skin which hangs in loose folds
about it and the tubercles are slightly developed or absent.

The vomerine teeth are reduced to a minimum ; often they are absent on
one side or altogether.

As might be expected from the exclusively aquatic habits of culeus, its
skeleton is weaker and less perfectly ossified than that of marmoratus. In
the latter the skull and its processes are strong and the foramina and fonta-
nel very small ; in the other the skull is thin and shell-like, the processes are
more slender and the foramina and fontanel larger. Large specimens of
culeus are here compared with much smaller ones of the other species.

These animals are very abundant in the extensive beds of weeds which
occur on the bottom of Lake Titicaca. They feed on the molluscs, Crus-
tacea, worms, etc., and are fed upon by the birds and fishes. Marmoratus
was found in little creeks and marshy places, in situations indicating habits
similar to the common Ranse ; during the two months of the observations
culeus was only to be found in the lake, crawling lazily about among the
weeds or half hidden by them, watching for prey. The latter was the only
one found in this vicinity ; the former was secured on the summit and the
western slope. These animals are able to remain under water for great
lengths of time without coming up for air ; hours of watching in clear
water, where many could be seen, failed to detect any approaching the sur-
face. It is possible they are more lively at night, when their enemies are
less active. Numbers were brought up in the trawl at more than four
miles from the shore. None were found on the land. The natives were
positive they never left the water. All stages of the animal are represented
by the specimens in the collection. On PL I. are given figures from above
and in profile (natural size).

Pleurodema Bibroni Tschudi.

Abundant at Puno, on the lake. Colors varying.

278 BULLETIN OF THE MUSEUM.

Leiuperus marmoratus Dum. ct Bibr.

From Vincocaya and Puno.

Leiuperus sagittifer Sciimidt.

Many specimens. From Arequipa, Juliaca, and Puno. From these ex-
amples a regular series leading from the one species to the other may he
selected. The existence of so many intermediates suggests that sagittifer
may be no more than a variety.

Bufo chilensis Tscnunr.

Many toads were collected under the impression that several species were
represented ; a careful examination and comparison includes them all in one.
The toad of the Titicaca Valley and the mountains differs from that of the
coast ; those of these localities are smaller, and have the black blotches nar-
rowed down to mere spots which makes them appear much whiter. There
is much variation in the coloration ; some have few spots and small, others
many, and others are so dark as to render the marks obsolete. Many are
rough with small spines, others are smooth. In some cases the warts are
greatly developed, in others the skin is quite free and glossy. "With the
cutaneous fold on the tarsus ; the very young are dark colored and have
yellow feet. From Arequipa, Vincocaya, Puno, and Carapata.

REPTILES.

Saukians.

Fhyllodactylus gymnopygus Dum. et Bibr.
Taken from beneath rocks on the plains near Arequipa, Peru.

Leiolaemus signifer Dum. et Bibr.

Very common on the plains' around Vincocaya and Colca, at an altitude

of 14,500 feet.

L. maculatus Gray.

Abundant on the hills around Lake Titicaca. None were found on the

summits, where signifer was so very plentiful.

L. Belli Gray.
Two examples presented by Col. E. A. Flint, were from the Cuzco Valley.

Ophidians.
Tachymenis peruviana Tschudi.
We are indebted to Col. E. A. Flint for one young and three adults
from the Cuzco Valley. This species is quite variable. These examples
have eight upper labials, the eye over the fourth and fifth, and in three of
them there are two anteorbitals instead of one. The colors are of the usual
pattern and very distinct.

Cambridge, November 26, 1875.

Bulletin, Vol. T::.. No. 11.

i

No. 12. — Exploration of Lake Titicaca, by Alexander Agassiz
and S. W. Garman.

II. Notice of the Palotozoic Fossils by Orville A. Derby, M. S.
With Notes by Alexander Agassiz.

Sperifera camerata Morton.

This well-known species occurs at Yampopata and the island of Titicaca.
The specimens are well preserved, and their identity with the North Ameri-
can and Brazilian forms cannot he doubted. Like some of the latter, they
show concentric lamellae like those represented by d'Orbigny on his S. con-
dor.* In my paper on the Brazilian Carboniferous Brachiopods I have
united with doubt S. condor with S. cmnerata. Mr. Agassiz's specimens,
which come from near d'Orbigny's locality, and are associated with the
same forms, appear to clearly establish their identity. As in North America,
this species enjoys a wide range. D'Orbigny and Salter f identified it from
various localities in the vicinity of Lake Titicaca. Toula % describes it as a
variety of S. striatus from Cochabamba on the eastern slope of the Andes,
and I found it in a specimen brought by Professor James Orton from the
Pichis River, Peru.

Athyris subtilita Hall sp.

This species occurs in the collection from Yampopata, d'Orbigny describes
it under the name Spirifer Roissyi from Yarbichambi, and Salter and Toula
have identified it from the Isthmus of Copacabana and Cochabamba.

Rhynchonella sp.

A species of this genus is represented by fragments on a slab from Yam-
popata, Though not sufficiently well preserved for positive identification,
there are reasons for believing it the same as the form figured by Salter from
Santa Cruz, and by Toula from Cochabamba as R. pleurodon Phillips.
This form is doubtfully identified by Meek and Toula with R. uta Marcou
(R. osagensis Swallow), to which it is certainly much more nearly allied than
to Phillips's species. A similar form occurs at the Pichis River locality, in
Peru.

* "Voyage dans 1'Amerique Meridionale, Paleontologie, p. 46, PI. V. Fig. 11 - 14.

t Quart. Jour. Geol. Soc, Vol. XVII. p. 50.

t LIX. Bde. d. Sitzb. d. Kais. Akad. d. Wissensch. I. Abth., 1869.

280 BULLETIN OF THE

Streptorhynchus sp.
A single fragment of a dorsal valve represents this genus. Part of the
. cardinal process is exposed, showing that it belongs to the group character-
ized by a high ventral area and a long cardinal process, for which S. Halli-
anus Derby may be taken as the type. It is marked by a distinct median
depression, which, if constant, will distinguish it from any species known to
me.

Chonetes glabra Geinitz.

Chonetes glabra Geinitz Carb. und Dyas in Nebraska, p. 60, PI. IV. Fi^.
15-18.

This characteristic species of the North American and Brazilian Coal
Measures is represented by several specimens on a slab from Yampopata.
Toula has also identified it from Cochabamba.

Productus costatus? Sowerby.

This form appears to be the most abundant fossil at Yampopata. It is
identical with the common form of the North American Coal Measures, as a
careful comparison of numerous well-preserved specimens proves. I follow
Hall and Meek in questioning its identity with the European P. costatus.

Judging from d'Orbigny's and De Koninck's figures, P. Inca d'Orb. is
identical with this, differing only in being a little larger than any of Mr.
Agassiz's specimens. European palaeontologists have referred P. Inca
to P. semireticulatus Martin, but it is certainly much more like P. costa-
tus ? than the North American and Brazilian forms referred to Martin's
species. These last, when well preserved, are covered with very numerous,
closely set hair-like spines quite unlike the few scattered coarse spines char-
acteristic of P. costatus. P. Inca appears to have had a spiny covering of the
latter kind, as had also many of the European forms of P. semireticulatus
figured by Davidson and De Koninck. As specimens usually occur with the
spines denuded and the scars more or less obscured by weathering, the two
forms of the so-called P. semireticulatus are almost indistinguishable, though
appearing very different when perfect. To unite them, as is usually done, is
to ignore one of the best specific characters among shells of this class.

The collection contains a few casts in red sandstone from the island of
Titicaca, that appear identical with the Brazilian forms of P. semireticulatus.

Productus Chandlessii Derby.

Productus Chandlessii Derby, Bull. Cornell Univ., Vol. I. No. 2, p. 51 , PI. IV.
Compare P. uolivicnsis d'Orb.

A single specimen from Yampopata is identical with the Brazilian form to

MUSEUM OF COMPARATIVE ZOOLOGY. 281

which I have given the above name. Since seeing it I am led to suspect
that d'Orbigny's P. bolieiensis is the same species, though his figures repre-
sent a very different shell, with some characters probably never seen on a
species of this genus. De Koninck's figures are much more like our shell,
though, if strictly accurate, the two would appear to be distinct. Aside from
some differences in form, the front of the shell in his figures is quite strongly
ribbed, while among hundreds of specimens from Brazil it is uniformly
smooth. The differences are sufficiently important to warrant the retention
of my name till the original types of P. boliviensis can be examined.

Productus Cora d'Orbigny.

Productus Cora d'Orbigny (op. fit.), p. 55, PL V. Fig. 8-10.

There are several fragments on a slab from Yampopata which most proba-
bly belong to this species, the type of which was from the neighboring
locality of Yarbichambi. It is unfortunate that we have no really good
specimens from Bolivian localities to show clearly the characters of the
species. D'Orbigny's figures of it are useless for purposes of identification,
and Toula, who had specimens from Cochabamba, thinks that the European
specimens figured under the name by De Koninck, and which are now re-
ferred to as the type of the species, are distinct. If it should prove so, a
thorough revision of the peculiar group of Producti to which this belongs
will be necessary.

The Brazilian forms referred with a question to this species are appar-
ently identical with those collected by Mr. Agassiz.

Euomphalus antiquus d'Orbigny.

Solarium antiquum d'Orbigny (op. cit.), p. 42, PI. III. Fig. 1-3.

The specimens referred here are in the state of partial casts, occurring
both at Yampopata and the island of Titicaca. They agree in the most
essential particulars with d'Orbigny's type, though differing somewhat from
his figures and from each other in the prominence of the spire, the form of
the whorls, and#the development of the thin carinas characteristic of the
species. It is possible that these differences may prove to be of specific im-
portance when larger collections are examined.

This species is named by Meek as one of the types of his subgenus Ompha-
lotrochus proposed in the Geol. Surv. of California, Vol. II. p. 16.

Euomphalus sp.
A cast of a species of Euomphalus with more angular whorls than the pre-
ceding and apparently with only a median carina occurs in the collection
from Yampopata, associated with a fragment of a much larger species with
smooth evenly rounded whorls.

282 BULLETIN OF THE

Tropidoleptus carinatus Coxrad sp.

Slrophomena carinatus Conrad. Ann. Geol. Kept, of N. Y., 1839, p. 34.

Several slabs* of coarse gray sandstone from the island of Coati, two miles
southeast of the island of Titicaca, are covered by well-preserved casts of
this species undistinguishable from the forms so abundant in the Hamilton
group of New York and the Devonian sandstone of Erere, Brazil (Rathburn,
Bull, of Buffalo Soc. Nat. Sci., Jan. 1874).

The Coati specimens are of large size, with very heavy dental plates and
septum.

The undetermined species of Orthis figured by Salter (Quart. Jour. Geol.
Soc, Vol. XVII., PI. IV. Fig. 7) appears to me to belong to this species.

Vitulina pustulosa Hall.

Vitulina pustulosa Hall, 13th Kept. State Cab.; Pal. of N. Y., Vol. IV.
p. 410, PI.

Associated with the preceding are casts differing in no respect from those
of this species so abundant at Erere, Brazil.

This extremely interesting collection adds considerably to our knowl-
edge of the geology of the Andes and the relation of the fossils occur-
ring there to those of other parts of the world. Of the nine Carbonif-
erous species, all but Euomphalus antiquus are represented in the Coal
Measures of North America and Brazil by identical or closely allied
forms. In a discussion of the relations of the South American Car-
boniferous fossils thus far known, I have shown that all the beds yet
examined belong to this division of the Carboniferous age.t These
beds are now known at a number of widely separated points in Bolivia
and Peru, showing a very extended range of the Carboniferous deposits.
Besides the numerous localities in and about Lake Titicaca, they are
known in the provinces of Arque and Oruro, south of the lake, at Santa
Cruz and Cochabamba, on the eastern slope of the Andes, and on the
Pichis River, one of the upper tributaries of the Ucayali, in Southern
Peru. More recently I have recognized in Professor Orton's collection
specimens of a Productus and a Streptorhynchus from near Moyabamba,
in Northern Peru. It is interesting to note in this connection that

* These slabs were not found in place, and may have been brought to the island for
building purposes by the Incas (A. Agassiz).

f Bulletin of the Cornell University, Vol. I. No. 2, p. 60.

MUSEUM OF COMPARATIVE ZOOLOGY. 283

Squier reports having seen a specimen of coal said to have come from
Lake Titicaca.

It is rather surprising that as yet no Subcarboniferous has been
found, though the Devonian was found by Mr. Agassiz close alongside
of the Coal Measures, the island of Coati being only two or three miles
distant from Titicaca.* Very extensive beds in Bolivia have been
referred to the Devonian by d'Orbigny and Forbes, but as yet the
palaiontological evidence is quite defective. Salter states that of the
fossils referred by d'Orbigny to the Devonian, only two are certainly
supra-Silurian, and those are as likely to be Carboniferous as Devonian.
Of the fossils referred to the Devonian by Salter only one, Phacops
latifrons, was considered as undoubtedly of that age, and this loses
much of its value in being in a rolled pebble from an uncertain locality.

The following notes explain in a general way the structure of the
Carboniferous system of the elevated plateau of Peru (A. Agassiz) : —

In addition to the fossils described already by Professor Derby a
number of Fusilinse were also found in the Carboniferous beds of
Lake Titicaca. They have been sent to Mr. Brady for identification.
The Carboniferous beds from which the fossils above noticed are
derived form the southern terminus of the Carboniferous system as
marked on the map of Bolivia and Peru by David Forbes. But this
system is not limited to the comparatively small area assigned to it
by Forbes. It consists, for the whole area which I have examined, of
a series of rather limited elongated basins, the longitudinal axis
extending, as noticed by Forbes, in a general northwestern to south-
eastern direction. The strata generally form a series of short folds
often quite sharp, so that at comparatively small distances the rocks
dip in very different directions. In the Straits of Tiquina, as men-
tioned by Forbes, we have strata thrown up nearly vertically dipping
east on the west side, and west on the east side ; it is by a series of
such faults more or less prominent, that the successive basins of the
Carboniferous system of the elevated plateau of Peru have been sepa-
rated. Near Copacabana the metamorphic rocks underlying the
denuded sandstones dipping to the east and west are plainly seen.
Another such protruded mass I have observed near Puno. Another
forms the hill of dark red trachyte to the west of Juliaca, rising directly

* See Note on page 282 (A. Agassiz).

284 BULLETIN OF THE

from the alluvial plain, the former bed of Lake Titicaca. To the
westward of Puno we find at Vilque, a distance of about fifteen miles,
another basin. Again at Santa Lucia, near Maravilles, a third basin
still farther to the westward, resting upon underlying conglomerates
and metamorphic rocks, in which the strata dip to the eastward, and
finally still a fourth basin at Sumbay Station, on the Puno and Are-
quipa Railroad, at a distance of about fifty miles east of Arequipa.
Between the Santa Lucia and the Sumbay basins there is an extended
volcanic deposit forming the elevated plain of Vincocaya, at a height
of nearly 15,000 feet. Professor Orton also mentions Compuerta as
a Jurassic locality, but as he did not collect the fossils himself, the
presence of a Jurassic outlier between the basins may be questioned
for the present. Of course I do not wish to be understood as stating
that there are only four such parallel basins, — a far greater number
will undoubtedly be observed hereafter. At all these points there
has been bituminous coal found and a number of beds opened, on
which more or less exploring has been done. On the northeast end
of the peninsula of Copacabana, at Yampopata, a small coal-mine has
been opened, capable, with the limited number of men employed, of
producing about thirty tons a day of a fair quality of coal, which
has repeatedly been used by the two small steamers plying at inter-
vals on the lake. The coal is a grateful substitute for the use of dry
llama-dung, for which you are frequently compelled to wait until a
sufficient supply can be collected. The works of the coal-mine at
Yampopata consist of a tunnel driven in at right angles to the bed
at the water level of the lake, giving a back of some six hundred feet.
From the tunnel a gangway has been driven a tolerable distance both
ways. But all the appliances for mining are so crude, and the knowl-
edge of coal-mining so defective, that with the present system no
great results can be accomplished. Yet this mine is situated on the
shore of a lake some one hundred and twenty miles long and forty
miles wide, round which is gathered a population of no less than
200,000 Indians living in a region entirely destitute of fuel, unless
the few bushes sold at a fabulous price for cooking purposes can be
dignified by that name. Even this source of fuel is becoming annu-
ally more difficult of access, so that whole villages are compelled to
use dry llama-dung and other manure as their only combustible. The

MUSEUM OF COMPARATIVE ZOOLOGY. 285

bed of coal on which the mine is worked dips to eastward, and is about
three feet in thickness. Smaller beds have been traced as outcrops to
the westward of the Yampopata mine ; the same bed has also been
opened on the east coast of the peninsula of Copacabana. Near Vilque
quite a number of coal-beds are found, in one of which a small pit has
been opened and coal of an excellent quality discovered. The bed dips
to the west. At Santa Lucia three small beds have been tested, and
considerable prospecting without satisfactory results has been done
on some of the beds. The coal appears of a rather purer quality,
but is mixed with layers of shale. At Sumbay, though there has not
been much mining done, the facilities for mining are good. Several
beds have been tested to a considerable depth (90 feet), and a good
quality of coal developed. As the mines are only about two miles
from the railroad, the position is, from its excellent grade to the road,
admirable. The railroad from Sumbay to Puno runs directly across
the width of the Carboniferous system of the elevated plateau to
the west of Lake Titicaca, and the nature of the successive basins,
all running more or less northwest and southeast, can readily be
traced. Although my observations show a far greater extent of
the Carboniferous series than was supposed to exist, it is evident
from information gathered in the country that its northern terminus
is by no means ascertained. Colonel Flint 'informs me that in the
Cuzco Valley along the railroad from Juliaca to Cuzco there are
beds of coal identical with those I had the opportunity to visit. Mr.
J. J. Thomson, on his map of the Department of Puno, marks the
district to the west of Pomata as Carboniferous. Mr. Orrego, in a
sketch of the minerals occurring in the Departments of Arequipa and
Puno, mentions also as Carboniferous the localities which I visited
along the line of the Arequipa and Puno Ptailroad. Mr. Orrego also
states that this Carboniferous system extends as far north as Caylloma.
He subsequently mentioned to me that they form a series of basins
having the same general structure and position relatively to inter-
calated metamorphic rocks noticed above, which at the points of con-
tact have highly metamorphosed the shales and sandstones in the
vicinity. Professor James Orton has also found, in the extension of
the same general line as the axis of Lake Titicaca, Carboniferous
fossils at the head-waters of the Amazonas (Pichis River), while he

286 BULLETIN OF THE ZOOLOGICAL MUSEUM.

states that Professor Raimondi of Lima has traced Carboniferous
series at a height of 14,000 feet, on the Apurimac, at a locality
intermediate between the Pichis River and the Cuzco Valley. I shall
return to the general structure of the shores of Lake Titicaca in
a subsequent notice, but -would only call attention to the fact that
Forbes and d'Orbigny assign the rocks along the east shore of the
lake to the Devonian. I have been unable to find fossils at any of
the localities where I landed, — Vilquechico, Moho, Conima, Carabuco,
— and can only state that the rocks (dipping to eastward or west-
ward) are not, except occasionally, the colored sandstones and shales
separated by metamorphic rocks, so characteristic of the Carbonifer-
ous series of the west shore of the lake and the territory beyond.

No. 13. — Recent Corals from Tilibiche, Peru, by Alexander
Agassiz and L. F. Pourtales.

The corals described here by Mr. Pourtales were collected in a
ravine about two miles east of Tilibiche, in the valley of Berenguela.
Tilibiche is on the northern edge of the Nitrate Basin to the rear of
Pisagua, Peru. These corals are interesting, coming as they do from a
height of 2,900 to 3,000 feet above the level of the sea, at a distance
in n straight line from the Pacific Ocean of twenty miles. The ra-
vine where they were found is about 450 feet below the general level
of the great Nitrate Basin of Peru, on the eastern side of the ridge,
parallel to the coast which divides the so-called Pampa de Tamarugal
from the lower narrow pampas extending from the summit of the
coast terrace (at a height of about 1,100 feet) to its western base.
The height of the base of the second parallel chain ranging from
2,500 to 3,000 feet. The river flowing through the valley of Be-
renguela has cut a deep canon not only through the comparatively
soft deposits underlying the Nitrate Basin, but also through the
Jurassic beds which constitute the greater part of the chain form-
ing the eastern edge of the Nitrate Basin. The corals were found
attached to the surface of the rocks in the interstices between adjoin-
ing masses, growing much as they would at the present day in similar
circumstances.

From the general features of the country along the Pacific coast
of Peru it requires but little imagination to reconstruct the former
internal sea formed by the Coast Range, which must have, within
comparatively recent geological times, covered the whole of the Ni-
trate Basin, and which has gradually been elevated to its present
position. At one time (the older period) this inland sea was con-
nected with the Pacific through the breaks of the Coast Range form-
ing the quebradas of Vitor, Camarones, Pisagua, Loa, etc., and
subsequently became an inland salt lake disconnected from the
Pacific, to be eventually drained by the breaking through of the
barriers at the old points of connection with the Pacific. This inland
salt lake was thus gradually changed to a lagoon, and finally entirely
drained as soon as the rivers flowing through it, forming the above-

288 BULLETIN OF THE

mentioned valleys, had cut their way as canons through the strata
underlying it. It seems therefore possible that even if the former
extension of the Pacific Ocean over the tract occupied by the nitrate
beds cannot account entirely for the deposition of the salt and ni-
trates, it must at any rate have played an important part in their
formation. The rivers, taking their rise higher up in the Andes to
the eastward, flowing through the basins, are all fresh. The water
near the general surface of the basin is saline, but as we go down
we soon reach a stratum of absolutely fresh water, showing that if
the saline matter were brought down from the mountains it must
have all been washed out at the present day, and that the main
cause to which the formation of the nitrates has been assigned is no
longer active. Certainly neither the number nor size of the extinct
and actual river-beds crossing the Nitrate Basin favors the presump-
tion that they could have been a sufficient cause for the accumulation
of the immense deposits of salt extending over the large area covered
by the nitrate and other saline beds.

From the careful observations made by Darwin on the elevation of
the west coast of South America, the positive proof of the recent
elevation of the continent (at certain points) to a height of 800 feet is
placed beyond doubt, it can, judging from terraces and other some-
what less positive proofs, be considered as reasonably certain that this
elevation extended to a height of 1,300 feet, while the presence of
corals at Tilibiche would seem to leave but little doubt that the
continent has gradually been raised within a comparatively recent
period to a height of at least 2,900 feet. The presence of extensive
saline basins on the west slope of the Andes, at a height of over
7,000 feet, flanked on their western edge by low ridges, may be due
to a similar cause. But, however this may be, we might almost be
tempted to claim that the elevation of the continent can be traced
to a still greater height, judging from the presence of eight species
of Allorchestes, a genus belonging to a truly marine family of Crus-
tacea (Orchestiadae) in Lake Titicaca, at a depth of 66 fathoms, and
thus attempt to establish the former connection with the sea of the
lake now at a height of 12,500 feet above the level of the Racine.
Only eight, and two of these are probably identical species, out of
eighty-one known species of this family, as I am informed by Mr.

MUSEUM OF COMPARATIVE ZOOLOGY. 289

Faxon, inhabit either fresh water or live inland in moist localities.
It is, however, quite remarkable that in none of the other fresh-water
lakes in which marine forms have been found, Lakes Superior, Michi-
gan, Lake Wetter, or in Lake Baikal, and other lakes supposed
formerly to have had a connection with the sea, has this family as
yet been discovered, though a closely allied species of Allorchestes
has an extensive geographical distribution in the rivers of the North-
ern United States. It must also be remembered that we have four
species of Orchestias which are land inhabitants, living under damp
leaves at a considerable distance from the sea.

There were a few other interesting specimens found in this locality
which unfortunately have been lost. One species of Millepora, very
closely allied to M. alcicornis, and a species of a Crustacean, closely
allied to a large Aega, which was found in a pool near Tilibiche.

In the saline pools there were numerous specimens of fresh-water
Gastropods (Hydrobinise). Diptera and Neuroptera larvce were found
in abundance.

These corals are fossilized into a compact crystalline limestone,
the crystals having generally destroyed the internal structure ; they
are impregnated with salt, which effloresces after washing.

Isophyllia duplicata n. sp. (Plate, figs. 1, 2, 3.)

Rounded masses about 10 cm. in diameter; the lower surface is not pre-
served, so that the absence or presence of an epitheca remains undetermined.
Calicles coalescing in very sinuous series, containing sometimes six or seven
centres which remain, however, always very distinct. The adjacent walls
remain always separated by a furrow, across which the costse are frequently
continuous. The latter are thick and appear to have borne blunt spines.
The septa are thick, with blunt equal teeth ; four cycles, with occasional
rudiments of a fifth, the septa of the first, second, and sometimes third, not
very different, and reaching to the centre. No paliform lobes, but sometimes
a slight thickening of the septa in their place. Columella generally absent ;
occasionally one or two obscure papilla? represent it. Width of calicinar
valleys 5 to G mm., of mural furrows 2 to 3 mm.

This genus and its nearest allies (Symphyllia we do not think can be
separated from it) is not represented in any lower strata than the Tertiary,

290 BULLETIN OF THE MUSEUM.

and they have their fullest development in recent seas. There are none
living now, however, on the Pacific coast of America, hut Symphylliae,
Myeetophylliae, Manicinse, and other genera of the Lithophylliacees me'an-
droides of Milne-Edwards and Haime are very abundant and characteristic
of the West Indian Fauna.

Convexastraea ? peruviana n. sp. (Plate, figs. 4 and 5.)

Mostly in small spheroidal masses, from 1 to 8 cm. in diameter. Calicles
crowded, small (2 mm. in diameter), deep. Septa thick, with apparently
smooth edges, in six regular systems and three cycles, the primary septa
alone reaching the centre, but leaving a small space between the ends.
Septa of adjacent calicles sometimes coalescing. No distinct Avail or furrow
perceptible between the calicles. No pali or columella visible, although a
small space in the centre looks as if it had been occupied by the latter.

This fossil, which at first sight reminds one of a Porites, on account of
the size of the calicles and general aspect, comes nearest the genus Convex-
astraea, and particularly Convexastra\i Waltoni Edw. & PI. Still I am not
quite satisfied with this identification. The species of that genus described
thus far belong to the Jurassic and Triassic formations.

Explanation of the Plate.

Fig. 1. Isophyllia duplicata n. sp. Nat. size.

" 2. Magnified portion of the same.

" 3. Single calicle nearly circumscribed, magnified.

" 4. Convexastra?a ? peruviana n. sp. Nat. size.

" 5. Calicles of the same magnified.

Bulletin, Vol. III., No. 13.

I • IER, lh

No. 14. — The Development of Salpa, ly Wm. K. Brooks, Ph. D.

Sketch of Adult Structure.

The accounts of the general anatomy of Salpa, published by Sars,*
Krohn,t Huxley,:}: Vogt,§ Muller,|| and Leuckart,1T leave little to be
done upon this subject ; but since these papers cannot be procured
easily in this country, and are absolutely inaccessible to the majority
of American students, it will not be out of place to give, as briefly as
possible, a description of the structure of this aberrant and highly
interesting genus. This seems the more important since many parts
of the developmental history must be unintelligible to those who are
not familiar with the more important peculiarities of the adult struc-
ture.

The animals classed together as Tunicata agree very closely, as far
as general plan of structure is concerned ; and although they vary
greatly in form, and in the relative size, position, and degree of de-
velopment of the various organs, all the important structures found
in one may usually be found in the others ; and although the genus
at present under discussion departs very widely from what may be
considered as the typical form, those who are acquainted with the
structure of one of the fixed ascidians will find no difficulty in tracing
its homology with Salpa.

The chief difficulty in gaining a clear conception of the relations of

* Fauna littoralis Norvegia?. VII. 63-85.

t Observations sur le generation et le developpement des Biphores. Ann. de Sc. Nat.,
VI. 1846, 110.

X Observations upon the Anatomy and Physiology of Salpa and Pyrosoma. Phil.
Trans. 1851.

§ Recherches sur les animaux Inferieurs deja Mediterranee. Mem. de 1'Institut
National Genevois, 1854, II. 1-62.

I! Verhand. der phys. med. Gesellschaft in Wurzburg. Bd. III. p. 57.

H Salpa und Vervvandte. Zoologische Untersuchungen, II.

The literature of our subject is so extensive that a complete historical sketch would
enlarge our account of the adult structure beyond reasonable limits; accordingly
authorities will be referred to, in this chapter, only in those special cases which have a
peculiar interest. The chapters which treat of the development will contain references
to the various writers, giving a history of the progress of our knowledge of this sub-
ject; and very complete historical sketches of our knowledge of Salpa in general may
be found in the papers by Huxley and Leuckart, above referred to.

292 BULLETIN OF THE

the various parts of a tunicate animal lies in the fact that it is com-
posed of a number of nearly concentric tunics or sacs, which are con-
nected with each other in a rather complicated manner ; and the
names given to these parts by the various authors are not always used
in the same way; the names used by one being applied by another to
totally different organs, so that it is difficult to understand what part
a name is intended to designate unless the terms used are defined
carefully. The precise term used is of minor importance, provided its
meaning is clearly stated, and since Huxley, in his paper on Pyro-
soma,* has defined, with great clearness and exactness, all the terms
which he uses, I shall employ his nomenclature as far as possible.

The Test. — The outer wall of a Tunicate is the "cellulose test"
(Figs. 1 and 3, a). This is a sac with two openings, — the "branchial
aperture " e, and the " atrial aperture " g.

The Outer Tunic. — Within the cavity of the test, and united to the
latter at the two openings, is the "outer tunic" (Figs. 1, 2, and 3, b).
This also is a sac with two openings, and usually conforms to the
shape of the inside of the test, to which it may or may not be united ;
whenever the two are separated over a considerable area, there is, of
course, a chamber between them, but as this is not one of the true
cavities of the body and is of no homological importance, the test may
be regarded as enclosing no especial cavity. The outer tunic is usually
more or less muscular, and is often spoken of as the "muscular
tunic " ; it is the " second tunic," of most writers. Its cavity — that is,
the space between its inner surface and the outer surfaces of the
tunics within it — is the true "body cavity," and since all the blood-
channels of Salpa and the forms allied to it are parts of this "body
cavity," more or less shut off by the union of portions of the outer
tunic to those within, it is often convenient to speak of it as the
" sinus cavity," or " sinus system." Within the outer tunic are the
"branchial sac," with its diverticulum, the digestive organs; and the
" atrial tunic."

The Branchial Sac- — During the earlier stages of development the
branchial sac is an entirely closed chamber surrounded by a tunic,
which is in turn entirely surrounded by the body cavity, by which it

* On the Anatomy and Development of Tyrosoma. Trans. Linn. Soc.; 1S60, XXIII.
'pp. 19C-250.

MUSEUM OF COMPARATIVE ZOOLOGY. 293

is separated from the outer tunic ; at an early stage of development,
however, the outer and branchial tunics unite at the anterior extrem-
ity, and the central portion of the area thus united disappears by
absorption, so that an opening is formed through which the cavity of
the branchial sac communicates with the outer water ; this opening is
the branchial aperture. Since the test, the outer tunic, and the
branchial sac are all united around its circumference, the free edges
are composed of all three of these. Upon the inner surface of the
outer tunic around the aperture there is a set of muscles, by the con-
traction of which the opening may be entirely closed. The interior
of the branchial sac bears several structures which are very constant
in form and position throughout the group. Upon the hsemal side
there are two long parallel folds which project towards the ventral
axis of the cavity, and form the boundaries of a deep longitudinal
furrow (Figs. 1, 3, and 24, m), which projects as a vertical ridge into
that portion of the body cavity which forms the haomal sinus, but
remains in free communication with the branchial cavity by a cleft
upon its neural side. In consequence of the thickness and opacity
of the epithelium which lines the fundus of this fold, it appears (es-
pecially in the transparent Tunicata in a fresh state) like a strong
hollow rod, mounted upon a thin ridge-like plate, and has been called
the " endostyle."

The bottom of the furrow is richly supplied with very long cilia,
and its sides are glandular, and secrete an adhesive slime, which
serves to entangle the particles of food which are carried with the
respired water into the branchial cavity. Upon each side of the
endostyle and parallel to it, there is a prominent line of cilia (Fig.
24, /, also 1, 3, 32 and 33, I) to which the name " epipharyngeal ridge "
has been given ; these two ridges are continued backward beyond the
posterior termination of the endostyle, where they unite to form the
"posterior epipharyngeal ridge," which passes backward along the
middle line of the posterior wall of the branchial sac to the mouth
(Fig. 1, o), before reaching which they again separate in Salpa, and
pass, one on each side of the tongue-shaped organ shown in Fig. 1, o.
In front of the anterior end of the endostyle the two epipharyn-
geal ridges diverge from each other so as to pass around the branchial
sac, near its anterior end, and thus form the " peripharyngeal ridges "
(Figs. 2, I; and 33, h). Upon the neural median line of the branchial

294 BULLETIN OF THE

sac, and therefore opposite the endostyle, there is a row of tongue-
shaped organs projecting into the respiratory cavity ; these are the
" languettes." The number of these is very variable ; in Salpa there
is only one (Figs. 1 and 3, w), in Pyrosoma there are eight, and in
most of the fixed ascidians the number is much greater.

At the posterior end or base of the branchial sac, its wall is directly
continued into that of the oesophagus (Fig. 24, o',) which is short and
opens into a somewhat dilated stomach (Figs. 24, a", and 4 o"). The
cavities of these organs are continuous with the branchial cavity, and
since the branchial sac is shown, by its development, to be nothing
more than the anterior end of the digestive tract, Huxley has called
it the " pharynx" ; but it will be much more convenient, for our pur-
pose, to retain the name which indicates its functional but not its
morphological relations. The anterior opening, already described as
the " branchial aperture " is undoubtedly the true homological
mouth, but it is much more convenient to make use of the name
which indicates its function, and to apply the term " mouth " to the
aperture of ingestion, at the point where the oesophagus leaves the
base of the branchial sac. The intestine bends around the stomach
and then runs forward nearly parallel with the oesophagus (Fig. 24),
and usually terminates at a point which is anterior to the mouth ;
the anus opens into the cavity of the " atrium," which is now to be
described.

The Atrial Tunic. — Since the form and relations of this structure
are very variable in the different Tunicata, a clear conception of it as
presented in the typical forms is essential to a correct appreciation of
tunicate structure and development in general, and, owing to a lack of
such a clear conception, many of the published accounts of these ani-
mals are of very little value. In the perfectly transparent genus
Perophora, which presents exceptionally favorable conditions for the
study of the atrium, the two openings of the external tunic are at the
same end of the body ; one of these, the branchial aperture, is formed
by the union of the branchial sac and outer tunic, as already de-
scribed, while the other, the atrial aperture, is formed in a similar way
by union of the outer and atrial tunics. The latter is a large bag,
parallel to the branchial sac, and upon that side of it which bears the
languettes ; at the bottom of this sac, which Huxley has called the
" mid-atrium," are the external openings of the intestine and repro-

MUSEUM OF COMPARATIVE ZOOLOGY. 295

ductive organs. On each side of the body the atrium is, as it were,
tucked into the space between the branchial and the outer tunic, thus
forming two long and broad but very shallow " lateral atria," the
cavities of which are continuous with that of the mid-atrium ; these
lateral atria are extended upon each side of the branchial sac until
they almost meet, and thus cover the whole surface of the sac except
a narrow line over and parallel with the endostyle. In a transverse
section we should have the test and outer tunic as two concentric
circles, and, within these, sections of the branchial and mid-atrial
chambers side by side and separated from the outer tunic and from
each other by the sinus cavity ; the cross-sections of the lateral atria
would be shown as two long parallel-walled diverticula from the mid-
atrium, curving around the branchial sac, and almost meeting upon
its opposite. side. The outer wall of each lateral atrium is separated
from the inner surface of the outer tunic, and the inner wall from the
outer surface of the branchial sac by blood sinuses ; but the separa-
tion from the branchial sac is not complete, for at certain points this
is united to the atrial tunic so that the branchial sinus is divided up
into a network of longitudinal and transverse vessels or blood-chan-
nels, crossing each other at right angles. In each of the parallelo-
grams thus formed the two tunics are absorbed, thus forming a " bran-
chial slit," through which the respired water passes from the branchial
chamber into the cavity of the lateral atrium, through which it is
driven into the mid-atrium, from whence it escapes through the atrial
aperture. It is difficult to give a clear account of the relations of the
branchial sac and atrium without the aid of diagrams, but an illustra-
tion may help to a conception of this somewhat complicated subject.
In the middle of a long glass tube blow a bulb to represent the stom-
ach, and enlarge one end of the tube to represent the branchial sac, and
the other to represent the mid-atrium ; the small tube uniting the
branchial sac to the stomach will represent the oesophagus, and that
which connects the stomach to the atrium will represent the intestine.
Now bend the intestine around the stomach so that the branchial
sac and atrium shall lie side by side, flatten out the latter, and
wrap it around the branchial sac, and the whole will form a pretty
correct model of these organs in a typical Tunicate. This illustration
is open to one objection, inasmuch as it seems to imply that the bran-
chial sac and atrium are serially homologous, while, in fact, the latter is

296

BULLETIN OF THE

not a part of the digestive tract, but originates independently. Such
a model will represent only the condition of the organs in the adult,

la

>. c

A "£,

•5 « 3

•£ "3 to

*> C .-

r- " ~

£.3 ..

,«. - c

2. rt '"

'P ^ ~

t? "* =

« t» O

■5 "w> s"

and although the branchial sac may be regarded as a pharynx, tne
atrium is in no sense a rectum.

The various tunics above described, with their diverticula and

MUSEUM OF COMPARATIVE ZOOLOGY.

297

appendages include all the organs of a tunicate animal, with the ex-
ception of the nervous system, the heart, and the reproductive organs.
Nervous System. — The nervous system is situated within the body
cavity, and is attached to the inside of the outer tunic between the
two external openings (Figs. 1, 3, 24, and 33, v).

Auult solitary Salpa ; haemal view : a, test; 6, outer tunic; c, wall of branchial sac ; e, bran-
chial aperture; /, muscular girdles ; ft., branchial cavity; k, peripharyngeal ridges j m, en-
dostyle ; n, gill ; nil, nucleus ; r, heart ; u, chain of young males.

The Heart. — The position of the heart varies greatly, but it is usually
found near the digestive organs (Figs. 1 and 3, r), and in all Tunicates
its motion is periodically reversed.

The Reproductive Organs.- — These are usually placed upon or near
the intestine, and their external openings are near the anus.

We are now prepared to consider the special modifications presented
in the organization of Salpa.

298 BULLETIN OF THE

The test (Figs. 1, 2, and 3, a) is very thick, and, in our species, so per-
fectly transparent that its outline can be traced only when the animal is
seen strongly illuminated upon a dark background. The branchial aper-
ture is at the anterior end of the body of the solitary form (Figs. 1, 2, e),
although the test of the chain-Salpa (Fig. 3) projects some distance in
front of it, upon the haemal side ; in both the test extends posteriorly
beyond the atrial aperture g. In the young of both forms the latter open-
ing is much nearer the branchial than in the adults, and in this respect
the young Salpa resembles the adult of the ordinary fixed ascidians. The
shape of the test varies greatly in the diffei'ent species, as well as in
the two forms of the same species. The solitary form of our species,*
when seen from above or from below (Fig. 2), is barrel-shaped in out-
line, with the posterior extremity obtusely pointed and the anterior
truncated. At the sides of the posterior extremity the test is pro-
longed so as to form two long, slightly curved processes, each of which
contains a coecum of the outeriunic, with a cavity which is continuous
with the sinus system. When seen in profile (Fig. 1), it is truncated
at both ends, from the neural to the haemal side ; the extremities are
slightly convex ; the posterior truncating plane is more inclined than
the anterior. Besides the two large posterior processes there are six
much smaller ones, two on the median line and two pairs, all of which
are composed of the test only, with no inner chamber. The anterior
opening for the admission of water (Figs. 1, 2, 3, e) is by far the
larger ; it occupies the whole width of the body, while the posterior
one (<j) through which the water is expelled is much narrower, and
placed at a short distance from the posterior extremity, at the base of
the truncating plane, on the neural side ; the anterior opening being
nearer the haemal side. The lips which close these openings are quite
prominent, and can be thrown considerably beyond the general out-
line, either when drawing in water or forcing it out.

Since the Chain-salpce are normally united into a chain composed

* This species is very abundant along the southern shore of New England, and was
described by Desor (Proc. Boston Soc. Nat. Hist., III. 184S, p. 75) as Salpa Caboti;
and subsequently figured ami described by A. Agassiz (op. cit. XI. p. 17) under the
same name. It seems to agree, in all respects, with the Salpa spinosa, Otto, figured
and described by Sars (Fauna littoralis Norvegine, 1S46, p. 85, Tab. 10, Figs. 1, 2. and 9).
The Salpa spinosa of Otto is stated to be the same as the S. mucronata-democratica of
Forskfd, but as I have not been able to see Forskal's figures, I am unable to tell from
his description whether the American species is the same.

MUSEUM OF COMPARATIVE ZOOLOGY".

299

of two rows of alternating zooids, the tests are flattened at the
points where they come into contact with each other, and the surface
is therefore composed in part of flat facets and in part of curved sur-

faces ; and at the point where two zooids are in contact a spur, or
diverticulum, from the outer tunic of each projects through the test,

300 BULLETIN OF THE

and meets and is joined to a corresponding spur from the outer
tunic of its neighbor. These diverticula are hollow, and the blood
passes into and out of them ; but the cavity of each is separated by a
partition from that of the one to which it is joined, so that there is
no communication between the sinus systems of adjacent zooids.
Each is furnished with eight of these spurs, by two of which it is
attached to the neighbor in front of it on the same side of the chain ;
two serve to join it to the one directly behind it ; two unite it to the
neighbor obliquely in front on the opposite side of the chain ; and the
remaining two connect it with the one obliquely behind it. When
the chain is quite young, the test is thin and the surface curved at
all points, and the spurs project some distance beyond it, and thus
keep the animals apart, and at the same time bind them together;
but as they grow larger the tests thicken, embrace, and gradually
cover up the spurs, and at last the tests of the two adjacent zooids
meet and become flattened by mutual pressure, and tend to force
the animals apart, and the chain now falls apart at the slightest dis-
turbance; so that a full-grown chain can be found only in water
which has been unusually still for some days. Although the motion
of the separated zooids is not as active as that of a united chain, in
which all the components act together to effect locomotion, they live
and flourish when separated, and all traces of the spurs disappear, and
the body again becomes rounded and presents the form shown in Fig.
3, the outline of which differs in several prominent features from that
of the solitary form. The long terminal processes are wanting, and
the posterior end of the body is prolonged into a broad, bluntly
pointed cone, which contains the digestive and reproductive organs,
and is the so-called "nucleus."

We may state here that the solitary form, which is the female, is
hatched from an egg which is carried within the bod}' of the Chain-
salpa ; and the Chain-salpse are the males, and are produced by a
process of budding from the body of the female, and a single egg
passes into the body of each male before birth.

The outer tunic (Figs. ], 2, and 3, b) conforms to the inner surface
of the test, and the two are usually in contact, although they are
united only around the edges of the apertures, and may easily be sep-
arated from each other. Upon the inner surface of the outer tunic,
the ganglion (v) and the muscular girdles (/) are attached.

MUSEUM OF COMPARATIVE ZOOLOGY. 301

The branchial sac presents all the characteristics already described
as peculiar to this structure. It is attached to the outer tunic
around the edges of the branchial aperture (e), the lips of which are
reflected inward and provided with a complicated system of muscles
(Figs. 3, 24, and 33, /') ; the whole forming a valve which usually
prevents the water from passing outward, although it may be so
arranged, by the contraction of the muscles, that the contents of the
branchial sac may be violently expelled through it.

Owing to the transparency of Salpa, the endostyle is very promi-
nent, and has the appearance of a solid rod in the adult, although
earlier stages show that it is simply a furrow between two longitu-
dinal folds. There are only two branchial slits, one on each side ; but
these are very large, and cover almost the whole surface of the
branchial sac, except the median dorsal and haemal lines. On the
neural side, the branchial slit opens directly into the atrium, and
the line where the two tunics unite is marked by the so-called "gill"
(??); the posterior border of the slit is seen in Figs. 4, 32, and 33 as a
curved line passing by the anus and mouth, and then bending forward
to become continuous with the epipharyngeal fold (I), which bounds
the slit on the haemal side. Its anterior boundary is the peripharyn-
geal ridge, and it will be seen that the branchial sac is a complete
cylinder only in that short portion which lies anterior to this ridge ;
while from this backward to the mouth, its sides are entirely want-
ing, and the branchial cavity opens directly into the atrium.

Although the inside of the branchial sac is supplied with cilia, as
in all Tunicates, these do not seem to be of as much functional impor-
tance as in the remaining members of the group, since respiration is
effected entirely through the action of the muscular girdles, which
also assist deglutition and are the organs of locomotion. These con-
tract rythmically, with great regularity, and at each contraction
the water is expelled from the branchial sac through the atrial
aperture ; and when the muscles are relaxed, the elasticity of the
test distends the chamber, and a fresh supply is drawn in through
the branchial aperture, the lips of which readily admit its passage in
this direction, while a similar set of valves allow its passage out of
the atrial aperture, but prevent its return. As the result of this
rythmical discharge of water, the animal is impelled forward with a
motion which is pretty uniform in the case of a perfect chain ;

502

BULLETIN OF THE

although the solitary. individuals, and those which have been set free
by the breaking up of a chain, move by jerks. Since the blood circu-
lates in the sinus system which surrounds the branchial sac and atrium,
and also penetrates the cavity of the gill, which is simply part of the
body cavity shut off by the union of the branchial and atrial tunics,
its aeration is amply provided for.

If a little carmine is added to water containing Salpse, the manner
in which the food is conveyed to the mouth can be distinctly seen. The
carmine, drawn in with the water, adheres to the inner surface of the
branchial sac, anterior to the epipharyngeal ridges, and is then rolled
along by the cilia until it reaches these ridges, the cilia of which are
so set that they change its direction and convey it to the anterior end
of the endostyle, the cilia of which gradually carry it backward to-
ward the mouth. The contraction of the muscles now becomes more
vigorous, and at each contraction the body is so compressed that the
epipharyngeal ridges come into contact with the haemal or branchial
surface of the gill, and the water is forced along the tube thus formed,
driving the food before it to the mouth.

Digestive organs of very young solitary Salpa, figured with the neural side below : b, outer
tunic ; d, wall of atrium ; g, atrial aperture ; h, cavity of branchial sac ; i, cavity of atrium ; I,
epipharyngeal ridge; m, endostyle ; n, gill ; o, mouth ; o', oesophagus; o", stomach ; </",
intestine ; o"", aims.

The Digestive Organs. — In the adult these are so obscured by the
organs which overlie them that it is almost impossible to trace their
course, which must therefore be studied in the young. Fig. 4 shows
the atrial and part of the branchial chamber, and the digestive organs
of a very young solitary Salpa, in which the cavity of the intestine is

MUSEUM OF COMPARATIVE ZOOLOGY.

303

not yet in communication with the stomach, and is still imperforated
at the anal end. The mouth (o) opens at the base of the branchial
sac, on the lucmal side of the posterior end of the gill («), and is joined
by a short curved oesophagus (o') to the stomach (o"); the intestine (ow)
is parallel to the long axis of the stomach, and the anus {o"") is close
to the mouth, but opens, not into the branchial, but into the atrial
chamber, the posterior boundary of which is indicated by the line I.

Fig. 5.

~niu

Nucleus of adult Cliain-salpa, — male; figured with the neural side uppermost : a, test ; b,
outer tunic ; 1, external aperture of testis ; 2, anus ; 3, mouth ; nu, nucleus.

The whole digestive tract is immovable, and without muscles, and the
food is driven through the permanently distended cavity by means of
the cilia, with which its entire inner surface is lined. The great pos-
terior sinus surrounds the digestive system on all sides, and the
mitriment is absorbed directly from its surface by the blood. In the
young a layer of large dark-colored cells may be seen, covering the
posterior portion of the stomach and intestine (Fig. 33) ; these

304 BULLETIN OF THE

seem to be the first traces of the " tubular hepatic system " ; a layer
of anastimosing tubes which, in the adult, covers the outer surface of
the stomach and intestine, and opens into the stomach at its anterior
end. The function of this organ is much disputed, but as nothing
definite is known upon the subject, a history of the discussion is not
necessary here.*

The Testis. — In the adult Chain-salpa, the digestive organs are cov-
ered, outside the " hepatic organ," by the glandular organ shown in
Fig. 5. This is a layer of arborescent folicles opening into the atrium
by two apertures (1), one on each side of the anus (2) ; and a micro-
scopic examination shows that it is the testis. It is found only in the
Chain-salpa, which is therefore a male ; while the ovary is developed
within the body of the solitary Salpa, and will be described at length
in connection with the development of the chain.

Fig. 6.

Spermatozoa, from the testis of an adult and from the branchial sac of an immature male.

The Ganglion and Sense Organs. — As an adaptation to its loco-
motive life, the nervous system of Salpa is more highly developed
than in most of the adult Tunicata, and is provided with highly spe-
cialized sense-organs, which are supposed to be those of sight and
hearing. The structure of these organs is described at length in the
papers by Vogt, Leuckart, and H. Miiller already referred to.

The Heart and the Circulation. — Since the discovery by Van Hasselt
in 1824, that the circulation of Salpa is subject to periodical and
somewhat regular reversal of its direction, the heart has been made
the subject of especial study by numerous observers ; but owing to its
delicacy and transparency and to its rapid pulsation, its structure and
mode of action have never been correctly described, and can be made
out only by studying the living animal after it has been treated
with ether until the contractions have grown very feeble and slow.

It consists of three concentric saddle-shaped portions, of which the
upper (Fig. 7, 4) and the lower (3) are thick and inflexible, while the
middle one (-:") is thin and composed of parallel rows of muscles, which

* Leuckart, pp. 33 -3S, devotes considerable space to this subject.

MUSEUM OF COMPARATIVE ZOOLOGY.

305

extend transversely from one side to the other. The muscular saddle
is larger than the upper, and is attached to it around its entire edge,
so that there is a closed empty chamber between the two ; and the
lower edges of all three are united, thus forming a crescent-shaped
passage between the lower and muscular saddles. When the muscles
of the middle saddle are relaxed, as in death by etherization, this
layer is drawn up against the upper one, apparently by atmospheric
pressure, and the upper closed chamber is no longer visible, while the

Heart and proximal portion of stolon of young solitary Salpa : b, outer tunic, continued into
the outer wall of the stolou ; 1, sinus chambers of stolon; 2, prolongation of pericardium;
4, forming the partition of the stolon ; 3, inner saddle of pericardium ; 4, outer saddle of
pericardium ; 5, muscular saddle ; G, arrow showing the course of the current through the
cavity of the heart.

lower crescent-shaped channel is open throughout its entire length ;
as the heart lies in the sinus system, this channel is of course filled
with blood. Now it is plain that if one of the transverse muscular
bands which compose the middle layer be contracted, the latter must
be drawn down at this point until it comes into contact with the
upper surface of the lower saddle, as seen in the figure, in which
two such points of contraction are shown. This will of course divide

306 BULLETIN OF THE

the channel through the heart, so that there will no longer he an un-
obstructed passage.

The muscles contract successively in order, from one end of the
heart to the other, and the middle layer is thus thrown into a wave,
which sweeps along the roof of the lower layer and drives the blood
before it, and before the first wave has travelled from one end of the
heart to the other a second follows it, and so on. After this has gone
on for some time the muscles begin to contract in the reverse order,
and the waves start from the other end and the direction of the cur-
rent is reversed. The cause of this change or the meaning of the
reversal is as yet by no means clear, although after the pulsations
have continued for some time in one direction numbers of blood-
globules may be seen crowded together in certain organs, such as the
elseoblast ; and these block up the sinuses until they are set free by
the change in the direction of the current.

In our species the blood-channels are in all cases sinuses, which arc
parts of the body cavity and have no special walls, although several
writers insist that in certain parts of the bodies of other species the
blood Circulates in true vessels lined with epithelium.

Embryology of the Solitary Salpa, — Female.

We are now prepared to enter upon the history of the development
of Salpa. Tli is is rendered somewhat complicated by the fact that
the two forms, besides differing considerably in structure, are devel-
oped in totally different ways; and before the embryo of the second
generation has completed its development within the body of its
parent, the formation of the third generation begins within it: accord-
ingly at certain stages we are compelled to study an embryo going
through one series of changes, and within this another embryo differing
in form, and undergoing an entirely distinct form of development.

On account of the manner in which the two forms of development
overlap, it is somewhat difficult to select a point at which to begin
our account. The fertilization of the egg seems to be the best point
of departure, although some parts of the description cannot be clearly
understood until the account of the development of the male has been
read.

The Egg: its Fertilization and Segmentation. — At the time when the

MUSEUM OF COMPARATIVE ZOOLOGY.

307

S;ilpa chain is discharged from the body of the solitary form each
" zooid " * contains a single egg,t which is situated upon or very near
the median plane of the neural side of the animal, within the sinus

Fig. 8.

Fig. 10

Fig. 8, egg before impregnation. Figs. 9 and 10, egg during impregnation. Figs. 11 and 12,
changes following impregnation : 6, outer tunic of male ; c, wall of branchial sac of male ;
h, cavity of branchial sac of male ; 1, sinus system of male ; 2, blood corpuscles of male ;
3 gubernaculum ; 4, yolk ; 5, germinative vesicle ; 6, spermatic filaments ; 7, capsule of egg ;
8, nucleus.

system, and midway between the atrial orifice and the stomach (see
P'igs. 31, 32, 33, and 34, s). There is no observable trace of a vitel-
line membrane ; the yolk is transparent without granules, and the
germinative vesicle does not contain a germinative dot (see Fig. 8). J

* The use of the word "zooid" in this connection must not be understood to imply-
any opinion as to what is or is not a zoological individual; it is used merely as a con-
venient word to designate one of the males which compose a chain.

t Leuekart says (Ueber Salpen, p. 49) that some species contain more than one egg.

J Huxley, who studied what appears, from his description, to be the species here
described, says (Salpa and Pyrosoma, p. 577) that there is a germinative dot occasion-
ally. Leuekart (p. 51) gives the germinative dot as one of the characteristics of the
egg, but I have not been able to find it in any instance.

308 BULLETIN OF THE

The egg is inclosed in a capsule of small epithelial cells (see Fig.
10), which are in direct contact with the surface of the yolk, and upon
the lower surface of this capsule there is a long stem or gubernacu-
lum, which passes down the side of the body of the zooid in the
sinus between the branchial sac and outer tunic, and is attached to
the wall of the branchial sac, upon the right side, near the heart.
According to the observations of H. Midler (Ueber Salpen. Zeitsch.
f. wiss. zool. IV. 3), " the stalked capsule which, in all new-born
Chain-salpas, encloses the egg, is an evagination from the wall of the
branchial sac, and the epithelium of the latter is directly continued
up the stem and over the yolk " ; but I was unable to trace the epi-
thelium on to the guberuaculum, which seems to be a solid rod of
protoplasm, passing through the wall of the branchial sac and pro-
jecting into its cavity. At the point where the gubernaculum joins
the wall of the branchial sac, the latter is slightly depressed, so as to
form a cup with its convex surface turned towards the branchial cav-
ity, so that the cavity of the cup, which is to form the " brood-sac,"
is a diverticulum from the sinus system ; and blood-corpuscles may
usually be found within it, as well as upon the sides of the gubernacu-
lum, adhering together so as to form irregular clusters, as shown in
Fig. 9.*

Very soon after the chain is discharged into the water, the egg
undergoes impregnation, which, however, is not effected by the sper-
matic fluid of the zooid which contains the egg, nor by that of any
other zooids in the same chain. The testis, at this time, is in a
very rudimentary state (see Figs. 33 and 34, /), and does not become
developed until after the solitary embryo into which the egg is devel-
oped has been discharged from the body. Wherever Salpa is found
at all, it is very abundant, and individuals at all stages of growth
occur together, so that some portion of the fluid discharged into the
water from the testis of an adult male readily finds its way into the
cavity of the branchial sac of the young male which carries the unim-
pregnated ovum, and numbers of actively moving spermatic filaments
may be found within this cavity at this time. These filaments seem
to be drawn to the exposed tip of the gubernaculum by some attrac-

* Leurkart (p. 47) describes this cup as a solid organ developed upon the wall of the
branchial sac; although he refers to Vogt's correct description of it (Bilder aus dem
Thierleben, S"J).

MUSEUM OF COMPARATIVE ZOOLOGY,

309

tivo force, since they swarm about it ; and great numbers become
attached to it, as well as to the wall of the branchial sac in its vicin-
ity, as shown in Figs. 9, 10, 11, and 12. A few of them penetrate
the gubernaculum, and work their way up towards the egg (Fig. 10) ;
and, although none were actually seen which had travelled over much
more than half the distance between the tip of the gubernaculum and
the egg, there seems to be no obstacle in their way, and they prob-

-4.

Fig. 14.

7.

c

%.

CO/

rig.

16.

\8&

a

its

,70.

Successive stages of segmentation : c, branchial tunic of male : A, cavity of branchial sac of male;
i, sinus system of male ; 4, yolk ; S, food-yolk ; 9, germ-yolk ; 10, orifice of brood-sac ; 11,
invagination orifice.

ably reach the yolk ; but the subsequent changes follow so rapidly
that no egg was found showing them so situated.

Immediately after impregnation the germinative vesicle disappears,
but as the actual process was not observed, nothing can be stated
as to the manner in which the disappearance takes place. Soon
after this is lost sight of, the gubernaculum becomes irregularly
swollen and shortened, as shown in Figs. 11 and 12, and the yolk

310 BULLETIN OF THE

dow seems to be prolonged into its cavity, although at an earlier
stage the epithelium of the capsule appeared to surround the entire
surface of the yolk. The swelling and shortening go on rapidly, and
in a short time the egg presents the appearance shown in Fig. 12. A
single nucleus can now be seen at the point previously occupied by
the germinative vesicle, and the egg, nourished by the blood which
bathes it, begins to grow, and is already somewhat larger than before
impregnation. The shortening of the gubernaculum continues until
the egg is drawn down from the median neural surface of the nurse
to the point upon the right side of the lower or haemal surface of the
branchial sac, where the gubernaculum is joined to the latter, as
already described. The brood-sac, or cup-like depression of the
branchial sac, has meanwhile increased in size, and now forms a
nearly hemispherical cup, large enough to contain the egg, which is
soon entirely withdrawn into it, as shown in Figs. 13 to 18. Since
the cavity of the brood-sac is a diverticulum from the sinus system
of the nurse, the blood has free access to it, and bathes the egg on all
sides. The latter is perceptibly larger at this time than it was dur-
ing the stage last described, and this process of growth continues
during the whole of the subsequent development ; so that the
embryo, at the time when the first traces of organs make their
appearance, is many times larger .than the unimpregnated ovum,
and when the solitary embryo escapes from the body of the Chain-
salpa it is two or three times as large as the latter itself was at the
time when the egg was fertilized. This remarkable growth is men-
tioned here, in advance, as it will not be referred to in the subsequent
description, although it must be understood as going on at all stages.

The egg is now pear-shaped, and is attached by its broad end to the
floor of the brood-sac, at the point where the gubernaculum origi-
nally joined the latter (see Fig. 13); the nucleus is now divided into
two, all traces of the epithelial capsule have disappeared, and nothing
more is known about it.

Of the two nuclei now present the lower, with the portion of yolk
which surrounds it, is destined to form the " germ-yolk " of the em-
bryo, and soon divides again, as shown in Fig. 14, and at a stage a
little later, Fig. 15, it is composed of a mass of minute segments.
The upper nucleus of Fig. 13 with its portion of yolk forms the so-
called "food-yolk," and segments much more slowly. In Fig. 15 it

MUSEUM OF COMPARATIVE ZOOLOGY. 311

is not at all divided, and in Fig. 1G it is divided into four large
spherules, enclosed in a common membrane. The food-yolk now
gradually becomes invaginated into the germ-yolk, as shown in Fig. 17,
and is soon entirely surrounded by the latter, forming the symmetri-
cal vase shaped "gastrula," shown in Fig. 18.* The cavity of the
gastrula opens directly into the sinus system of the nurse, and the
blood now circulates into and out of the primitive digestive cavity, as
well as around the outside of the embryo, which grows rapidly, and
soon fills the brood-sac, so that its outer surface comes into contact with
the latter, which soon ceases to be visible as a separate covering (Figs.
19 and 20), and of course the blood no longer bathes the outside of
the embryo, although it continues to pass into and out of the primi-
tive digestive cavity. The germ-yolk now becomes finely segmented,
although the spherules are still somewhat larger than the more trans-
parent ectoderm cells outside them. The invagination cavity or
"primitive digestive cavity" becomes separated into two portions; the
outer remains in free communication with the sinus system of the
nurse and forms the inner chamber (Fig. 19, 10) of the placenta, and
its opening becomes the orifice of the placenta. This persistence and
functional importance of the "orifice of Rusconi" are very remarkable,
and seem to have no parallel among the other Tunicata, or in any of
the various groups of animals with which it has been proposed to as-
sociate them. The gastrula of Salpa seems to be a special adaptation
to the very anomalous mode of development of the embryo. The lower
portion of the primitive digestive cavity now becomes entirely sur-
rounded by the endoderm, and soon becomes obliterated, as shown in
Fig. 18. Very soon a cavity reappears in this portion M the embryo,
and persists and forms the cavity of the branchial sac (Fig. 20, 13). A
constriction now appears upon the outside of the embryo, separating
the placenta from the embryo proper, and soon a body cavity becomes
visible, separating the branchial sac from the outer wall, and also ex-
tending up around the placenta, to form its outer chamber (Fig. 22, 12).
The placenta, therefore, consists of an inner chamber communicat-
ing with the sinus of the nurse, and having no communication with

* A number of eggs at all stages between the two represented in Figs. 17 and 18 were
found, but the two here shown seem to be all that are necessary for clearly represent-
ing the process. Huxley (Plate XVI. Fig. 7) gives a figure of a stage in which the
invagination is about "naif completed, but he does not refer to ii in the text.

;i2

BULLETIN OF THE

any of the cavities of the embryo; with a cavity which is part of the
original "cavity of invagination," and is surrounded by a wall of cells
derived from the endoderm : and an.outer chamber bounded on the in-
side by the cells of the endoderm, and on the outside by the ectoderm,

Fig. 19.

-73

Fig. 21.

Successive stamps in the formation of the embryo ; b, outer tunic of male ; c, wall of branchial
sac of male ; li, branchial cavity of male : 1. sinus system of male; 10, opening of inner cham-
ber of placenta ; 11, club-shapi il organ ; 12, outer chamber <>f placenta ; 13, branchial cavity
of embryo; 11, nucleus of embryo.

having no communication with the sinus svstem or other cavities of
the nurse, but being directly continued into the body cavity of the
embryo.

The Placenta. — For the sake of clearness we will neglect for the
present the changes which are now taking place in the body of the

MUSEUM OF COMPARATIVE ZOOLOGY. 313

Embryo a little more advanced : 15, lateral atrium ; i, mid-atrium ; e, branchial aperture ; k,
peripharyngeal ridge ; I, epipharyngeal ridge ; v, ganglion.

embryo, and will describe at once the subsequent development of the
placenta. The inner cavity of this is at first one undivided chamber
into and out of which the blood of the nurse is constantly passing ;
but soon a singular club- or stump-shaped structure (Fig. 18, 11) ap-
pears in its opening, dividing this, at first incompletely, into two, and
projecting into the sinus system of the nurse as well as into the cavity
of the placenta, and serving to more effectually divert the blood into
the latter. Appearances seem to indicate that this organ is not derived
from any of the parts of the nurse or embryo, but is formed directly
from the blood, by the aggregation and fusion of its corpuscles. Ob-
servations upon this point were made so frequently and with such
uniform results that there seems to bo little doubt that it does origi-
nate in this way. No traces of a cellular structure were observed
in it. It is at first entirely free from all the adjacent parts, but
very fine threads are soon visible extending from it, and some of
these soon unite to the neck of the placenta and serve to anchor the
two together ; the terminations of the remaining threads were not
traced, but they seemed to float in the blood of the sinus.

The lower end of this organ now extends downward, and joins the
bottom of the placenta, and soon becomes divided into a number of
root like portions (Figs. 19, 21, and 22,11), which separate the cavity
into irregular intercommunicating lacuna?, through which the blood

314 BULLETIN OF THE

now passes, and among the meshes of which the corpuscles are often
entangled, so that the circulation is much less rapid than in the sinus
outside. Before this stage is reached the blood of the foetus also be-
gins to circulate, and as the outer chamber of the placenta is part of
its sinus system, the small corpuscles contained in its blood can be
seen passing around the outside of the inner chamber, and thus
coming into very close proximity to that of the nurses. Huxley's ac-
count of the placental circulation at this period is so good that it will
not be out of place here, although he adds very little to the accounts
previously published by Sars and Krohn. He says (p. 575), that the
placenta " contains two perfectly distinct cavities or sacs ; of these
the outer is concave and cup-shaped and envelops the inner, which is
subspherical. Now the outer sac is in free communication by a nar-
row neck, divided into two channels by a partition, with the dorsal
sinus of the foetus ; and the inner sac is in equally free communication
by a neck similarly divided, with a short sinus arising immediately
behind the heart ; and as there is no communication between the two
sacs, it follows that the current of blood in each is perfectly distinct
from and independent of that in the other. A more beautiful sight,
indeed, can hardly be afforded to the eye of the microscopic observer
than the circulation in this organ. The blood-corpuscles of the parent
may be readily traced entering the inner sac on one side of the parti-
tion, coursing round it, and finally re-entering the parental circulation
on the other side of the partition ; while the foetal blood-corpuscles,
of a different size from those of the parent, enter the outer sac, cir-
culate round it at a different rate, and leave it to enter into the general
circulation of the dorsal sinus. More obvious still does the inde-
pendence of the two circulations become when the circulation of either
mother or fcetus is reversed."

The following historical sketch of our knowledge of this structure
is also taken from Huxley's paper, page 592 : "Cuvier* speaks of
finding a fcetus attached to the parent by a pedicle ; and, referring to
a figure, he says : ' Ce corps rond [evidently the placenta] seroit-il un
organe servant uniquement pendant le temps de la gestation pour
etablir l'union entre la mere et son petit et qui s'effaceroit ensuiteT

Chamissot calls the pedicle of attachment ' pediculus umbilicalis ' ;
the placenta, ' globulus opacus.'

* Ann. Mus. d'hist. nat. 1804, IV.

t Nova Acta, 1832, XVI. pp. 3G2 - 422.

MUSEUM OF COMPARATIVE ZOOLOGY. 315

" Meyen was the first to give this structure the name of placenta,
and his account of it is so very clear and precise that it is wonderful
it should have been subsequently forgotten or overlooked. He says:
' Wir haben bei ganz jungen Individ uen den Verlauf der Blut-beweg-
ung selbst by 200-maliger Vergrosserung beobachten konnen. Der
Muttertheil der Placenta hat nur wenig Gefasse, um so rnebr aber der
Fotus-theil, in dem sich ein ausserordentliches Convolut von Gefassen
befindet, das sich in einem Stamme endigt, der sich in das grosse
Bauchgefasse ganz in der Nahe des Herzens ergiesst. Ein unrnit-
telbares Uebergehen der Blutgefasse aus dem Muttertheil in den
Fotus-theil haben wir nicht sehen konnen. Hat der Fotus die hinlan-
gliche Ausbildung im Leibe der Mutter erreicht, so verwiichst das
grosse Blut-gefass und die Placenta fiillt ab.' — Page 440."

Krohn, Huxley, Leuckart, and Vogt subsequently investigated the
structure of this organ, and have correctly described its more impor-
tant features as they exist when fully formed, but its origin and the
nature and structure of the club-shaped core are here described for
the first time.*

The embryo, at the stage last described (Fig. 20), consists of the
outer tunic derived from the outer layer of the gastrula, the body
cavity, and the branchial sac, the wall of which is formed from a por-
tion of the cells of the inner layer, while the remainder form a large
mass (14), at what is to be the posterior end of the body. The subse-
quent development of these parts, as well as the formation of those
which subsequently make their appearance, will now be described.

The Outer Tunic. — Little need be said of this ; it is not at first
covered by the test, which makes its appearance later, apparently by
excretion from the surface of the outer tunic ; the latter now covers
all of the embryo except the part in contact with the placenta, and
is reflected on this to form the outer wall of the outer or foetal
chamber, as already described. It is entirely separated, at first, from
the branchial sac by the body cavity, but at a later stage an invagi-
nation appears at the anterior end (Fig. 22, e), which unites with the
anterior end of the branchial sac, and a perforation appears which
forms the incurrent opening (e, Figs. 1, 3, 23, 24, 25, 33, 34). The

* Leuckart says (p. 53), that the residual yolk becomes the placenta, hut we have
seen that this becomes invaginated to form the digestive layer of the germ, and that the
placenta is not formed until the close of the gastrule stage.

316

BULLETIN OF THE

atrial opening (g) in the same figures is formed in a similar manner,
but somewhat later, by coalescence with the walls of the atrial
chamber. The muscular bands wliich adhere to the inside of the
outer tunic of the adult are derived from the atrial tunic, and will be
described in connection with that organ.

The Body Cavity. — During the early stages the whole surface of
the branchial sac, except the region next the placenta, is surrounded
by the body cavity, the connection of which with the outer chamber
we have already described. As development progresses, the outer
tunic and the wall of the branchial sac unite at various points, and
the atrial tunic is formed between them and has regions of attachment

Embryo more advanced than the one shown in Fig. 22, but less highly magnified : the muscu-
lar girdles are partially separated : 1, sinus system of nurse ; 10. opening of placenta : 12t
outer or foetal chamber of placenta •, c, branchial sac of nurse ; h, branchial cavity of nurse ; hi
outer tunic; c'. wall of branchial sac; e, branchial aperture; h\ branchial cavity; i, mid-
atrium; w, gill; m, endostyle ; /, muscles; v, ganglion; x, elaeoblast; d i, digestive or'
gans.

to both, so that we no longer have a single body cavity, but instead
of it a sinus system.

Branchial Sac and Digestive Organs. — The branchial sac is at first
a single, nearly oval cavity (13, Fig. 20), occupying a little more than
the anterior half of the embryo, while the posterior half is filled with
a mass of cells (14, Fig. 20), which are to give rise to the various
organs of the nucleus. The cavity of the branchial sac soon becomes
lengthened backward, so as to occupy two thirds or three fourths
of the body of the embryo (Figs. 22, IS; 23 and 24, h), and the
nucleus is divided, unequally, into two portions, of which the smaller
(Fig. 23, di), which is upon the neural side, gives rise to the wall of
the digestive organs, while the larger or hamial portion (Fig. 23, x)

MUSEUM OF COMPARATIVE ZOOLOGY. 317

becomes the eltcoblast. The cavity of the oesophagus soon becomes
visible as a diverticulum from the posterior end of the cavity of the
branchial sac, and is in direct communication with the latter at the
earliest stage observed. Two diverticula are now formed upon the
lower or haemal surface of the oesophagus, side by side and parallel
with each other (Fig. 4, o") ; as these grow they gradually unite and
form a single large diverticulum, which is bent forward towards the
mouth so that it lies nearly parallel with the oesophagus, as shown in
Fig. 24, o". Since the cavity of the stomach is derived from that of
the branchial sac, it must, like the latter, be regarded as a portion of
the primitive digestive cavity of the gastrula. The cavity of the in-
testine, however, first appears as a closed chamber, parallel with and
on the haemal side of the stomach, and having no communication
with the cavity of the latter (Fig. 4, o").

The partition which separates the two soon disappears, and they be-
come continuous, although the anal end of the intestine, which is close
to the mouth, is still closed. After the chamber of the mid-atrium or
cloaca has been formed the anus unites with this, the partition dis-
appears, and the digestive organs now form an U-shaped tube, con-
necting the branchial with the atrial cavity. As nothing was learned
of the development of the so-called " hepatic organ " of the solitary
Salpa, I will now give what little was learned about it in the chain-
salpa, in order that this description of the mode of formation of the
various organs may be as complete as possible. In the young chain-
salpa, at the time it is discharged into the water, the posterior end of the
stomach and the side of the intestine are covered with a single layer
of large, nucleated, dark-colored cells (Figs. 33 and 34), which present
the characteristics of liver cells, and agree in appearance, and approxi-
mately in position also, with the liver cells upon the wall of the upper
portion of the stomach of a Polyzoon.

At the time that the observations upon Salpa were made a species of
Appendicularia was very abundant. The wall of the stomach of this
bore, upon the inside, two large clusters of cells which were arranged
in a single layer, and agreed, in appearance, color, &c, with those found
upon the outside of the stomach of Salpa, which are undoubtedly liver
cells, and the rudimentary " tubular hepatic organ."

The study of this organ was extremely difficult, as the eloeoblast of
the solitary Salpa obscures this portion of the digestive organs at all

318

BULLETIN OF THE

stages, while, during all stages but the earliest, it is hidden in the
chain-sal pa by the testis. The second -portion of the nucleus, which
is much larger than the portion which forms the digestive organs is

Fig. 24.

Embryo at the time the first traces of the stolon appear: 1, sinus system of rmrse ; b, outer
tunic of nurse; c, branchial tunic of nurse ; h, branchial cavity of nurse ; 6', outer tunic ;
d, atrial tunic ; e, branchial aperture ; /, muscles ; g, atrial aperture ; 7s', branchial cavity ; k
heart ; I, epipharyngeal ridges ; 7?!., endostyle ; o, mouth ; o', oesophagus ; o", stomach ; x,
elaeoblast. Through an error the ganglion, as well as the heart, is marked it.

situated upon the hsemal side of the body, as already stated (see Fig.
23, x), and soon becomes divided up into large transparent cells, filled
with oil globules, or with a body of oil-like appearance and refractive
power, but not divided into globules. These cells soon become polyg-
onal by mutual pressure, and form a large spherical mass which in-
creases much more rapidly than the remaining organs of the body
(Fig. 24, x), and at the time the Salpa chain begins to be formed
within the body, the mass of cells is nearly as large as all the other
portions of the embryo. The blood circulates freely among the cells
and over the outside of the mass, which is not separated from the
sinus system by any membrane, and the blood globules, and the oil
which escapes from the older cells, often accumulate in large masses
near the digestive organs and heart.

MUSEUM OF COMPARATIVE ZOOLOGY. 319

This organ, for which Krohn has proposed the name " elceoblast,"
attains its greatest development in the embryo, and begins to disap-
pear as soon as the chain begins to form within the body, and is there-
fore provisional, as pointed out by Krohn, and has no permanent place
or function in the adult. Meyen regarded it as a yolk (p. 401), and
there seems to be little doubt that it is, if not strictly yolk, at least
something having a very -similar use, and supplying the material which
is to be employed in the formation of the zooids of the chain. Leuckart
states (p. 57), that it is found in both forms of Salpa, but this is not
the case. It is always present in the solitary Salpa during the em-
bryonic development, and the nucleus of the young chain-salpa di-
vides, like that of the solitary form, into an upper smaller portion,
which gives rise to the digestive organs, and a lower, larger portion,
which thus agrees with the elaxiblast in its mode of origin, and, during
the earliest stages, in appearance also (Fig. 32, t). It never attains
a great size, as compared with the other organs, and soon forms
a compact globular mass of cells (Fig. 34, t), which in the adult
chain-salpa forms the testis (Fig. 5). Since the testis of the chain-
salpa agrees so closely with the elaeoblast of the solitary one, it seems
very probable that this is also a reproductive organ, and that Mej-en
was right in calling it a yolk-deposit. It is not a true ovary., for this
originates in a different way, as will be shown further on.*

The Atrium. — The adult Salpa is composed of three principal tunics,
each enclosing a special cavity. The formation of the-outer one of
these, the outer tunic, has already been described, and we have seen
that it is derived from the outer layer of the gastrula ; its cavity has

* Few organs have had their functions more disputed than this elaeoblast. Many of
the earlier observers considered it a liver, and Sars repeats this error, although Meyen
had previously determined its connection, in some way, with development. Krohn
discovered that it was transitory and provisional, but does not commit himself in regard
to its function. Huxley described the true liver, and, although acquainted with all that
had been written upon the elaeoblast, says (p. 571), "There would seem to be no clew
either to the homology or to the function of the elceoblast. Without hazarding a con-
jecture, it may be remarked, as a curious fact, that these animals, so remarkable for
possessing in the foetal state a true though rudimentary placental circulation, possess
an organ which in structure and duration somewhat calls to mind the thymus gland."
Leuckart (p. 57) concludes that it is a depot of nutriment stored up for future use, but
he does not state whether it is to lie used in the formation of the embryo or in the de-
velopment of the chain. We will return to the subject of its homology with the testicle
further on.

320 BULLETIN OF THE

also been described as the body cavity or sinus system. The inner
tunic, or branchial sac, has been shown to he part of the inner or
digestive layer of the gastrula. The third tunic, now to be described,
is the atrial tunic, and its cavity the atrial chamber or cloaca.
The earliest stages in the formation of this tunic were not traced
in either form of Salpa, but as it presented, when first observed,
an appearance very similar to that described by Kowalevskv,* as
presented during the early stages in the development of Pyro-
soma, it is possible that it originates in the same way ; although
in the absence of any observations upon its origin, the fact that
Kowalevsky's observations upon" Pyrosoma are here referred to
must not be regarded as implying anything more than a belief that
the similarity which is known to subsist between the two genera, in
many other respects, may extend to the way in which the atrium Is
formed. In the egg-embryo of Pyrosoma two circular depressions
appear upon the neural side of the outer tunic, near the anterior end,
and these deepen so as to form tubes derived from the outer layer,
witli cavities which open externally ; these tubes lengthen and pene-
trate the body cavity upon each side of the branchial sac ; their exter-
nal openings close, and they become separated from the outer tunic,
and form the " lateral atria " of Huxley. f According to Huxley
(pp. 215, 21G), ■ — and Kowalevsky's account corroborates that of
Huxley in every particular, — these at first " arc very small and thick-
walled ; they soon become larger and the walls proportionally thinner,
and the sacs themselves are both absolutely and relatively larger.

" In Fig. 29 they are very much larger and thinner, and their rela-
tions to other organs are especially worthy of attention. The outer
layer of each is applied to the outer tunic of its side, leaving a small
interspace, which communicates freely with the great posterior sinus,
in which the intestine and genatilia are disposed, and with the ante-
rior sinus which lies between the pharyngeal wall [branchial sac],
and the external tunic. This interspace is, in fact, the parietal sinus.
The internal layer, continuous with the outer anteriorly and poste-
riorly, but separated from it by a wide chamber fur the rest of its
length, is applied against the wall of the pharynx [branchial sac], for

* Ueber die Entwickelungsgescliiclite cler Pyrosoma. Von A. Kowalevsky. Arcliiv.
fur Mikr. Anat., XII., 1S75, p. 597.

t Huxley, Anatomy and Development of Tyrosoma, p. 205. Trans. Linn. Soc,
XX 1 1 1., 1800, p. 193.

MUSEUM OF COMPARATIVE ZOOLOGY. 321

four fifths of the extent of the latter, and then coats the lateral por-
tions of the gastro-intestinal tract, forming the anterolateral boundary
of the great posterior sinus. The space between the wall of the pharynx
and the inner layer of the sac communicates anteriorly with the ante-
rior sinus, posteriorly with the posterior sinus, and it is interrupted
at several points by the union of the pharynx and inner layer with
one another. It represents the system of branchial sinus." In
Salpa this \inion of the inner wall of the lateral atrium with the
branchial sac does not take place. "In side views it is not easy to
make out the boundaries of the lateral sacs ; but it is most important
to observe that, as has been already mentioned, in the middle of the
lateral face of the pharynx, and therefore also in the middle of the
lateral face of the inner wall of the sac, a series of opake rings with
clear centres, the rudiments of the branchial stigmata, make their
appearance." In this respect also Salpa differs from Pyrosoma ; no
branchial clefts are ever formed in connection with the lateral atria.
" These correspond with the points of union of the pharynx and the
inner wall of the sac. They are at first small, round, and very in-
distinct, but by degrees they elongate in a direction perpendicular to
the long axis of the pharynx, and their real nature becomes apparent.
Hence it is clear that these stigmata must eventually open into the
lateral sacs, as indeed they may be seen to do in such buds as that
represented in Fig. 30 ; and hence also it follows that the lateral sacs
are the rudiments of the lateral atria.

"At first the lateral atria appear to be perfectly distinct from one
another, and no atrial aperture is discernible. In buds, such as that
represented in Fig. 29, again, they do not extend, posterioi'ly, fur-
ther than the sides of the alimentary canal ; but in more advanced
buds they are produced backward on each side until they pass beyond
the level of the posterior margin of the stomach, so that they now
constitute the entire lateral boundaries of the great posterior sinus.
The longitudinal section of a somewhat smaller bud shows that in
this condition the atria are no longer distinct, but are united together
below [on the neural side of] the stomach, by a comparatively nar-
row and short canal, which is the mid-atrium.

"I have not traced out all the details of the process of coalescence
of the lateral atria ; but I suppose that each branchio-parietal portion
of the atrium, at first a distinct sac, is prolonged downwards and in-

322 BULLETIN OF THE

wards, under the stomach, and that the opposed walls of the pro-
longation become applied to one another, coalesce, and then become
perforated.* At any rate, the mid-atrium is now surrounded by a
membranous wall, continuous on all sides with the lining of the lat-
eral atria, and applied superiorly and anteriorly against the stomach
and oesophagus, posteriorly and inferiorly against the external tunic,
but not touching either of these parts, except for a small space on tin;
floor of its chamber, where it becomes united with the external tunic
to allow of the formation of the atrial aperture. This aperture is
situated on the neural side of the body, in front of the posterior end,
which is chiefly occupied by the genitalia ; but as development goes
on the mid-atrium increases, disproportionately, and encroaches upon
the other organs, upwards and forwards, in such a manner that its
anterior wall invests the whole posterior and lateral faces of the gas-
trointestinal division of the alimentary canal The facts which

I have detailed are exceedingly important for the comprehension of as-
cidian structure in general." This account of the form and connections
of the atrial chamber applies to those zooids of Pyrcsoma which are
produced by budding, as well as to the egg-embryo, but according to
the observations of Kowalevsky the diverticula which give rise to the
lateral atria of the bud-zooid are derived, not as in the egg-embryo,
from the outer tunic, but from the branchial sac, or inner tunic, and
the cavities are therefore diverticula from the cavity of the latter. t

The atrium of Salpa, when first observed (Fig. 22, 15), was com-
posed of two broad lateral atria within the body cavity, one on each
side of the branchial sac, and a very small mid-atrium (i) ; so that we
have all the essential parts of this structure, occupying similar posi-
tions to those of Pyrosoma, and bearing almost identical relations to
the surrounding parts. The lateral atria do not, however, as in most
Tunicata, remain connected with the mid-atrium, and unite with the
wall of the branchial sac to form the branchial slits, but soon be-
come entirely separated (Fig. 23,/), and the two walls of each unite so
as to form a broad solid sheet of tissue, which soon splits up to form
the muscular bands of the branchial sac, of which there are six in the
solitary form, and five in the chain-sal pa.

* This conjecture is fully proved by the observations of Kowalevsky.

t Compare also the manner in which the atrium is formed in the bnd-zooids of
Amauricium and Dideniinum. Ueber die Knospung der Ascidien. von A. Kowalevsky.
Archiv. fur Mik. Anat. X ,iS74, p. 441.

MUSEUM OF COMPARATIVE ZOOLOGY.

!23

The Muscles. — The bands of the two sides of the body are at first
entirely distinct from each other, as noticed by Krohn, and all of those
upon one side are, at first, united above and below, as in Fig. 23 ; the
haemal ends of the first, second, third, and sixth, in the solitary Salpa,
soon become free, as shown in Fig. 24, and the neural ends separate
into two bundles, composed of the first, second, third, fourth, fifth,
and sixth, respectively (Fig. 24). The first now separates at both
ends (Fig. 25), and after a time the haemal ends of the fourth and
fifth also become free, but the amount of separation is not uniform
in specimens of the same age, and the union between the fourth
and fifth may persist until the animal is nearly full grown (Figs.
1, 2, and 23); and to this the discrepancy in the various descrip-
tions of the species is due. The neural ends of all the corresponding
bands upon the two sides of both forms of Salpa soon unite, as well
as the haemal ends of the first, second, third, fourth, and fifth of the
solitary Salpa, which thus form closed muscular belts or girdles en-
circling the body ; while the luemal ends of the sixth pair of the
solitary Salpa, and of all in the chain-salpa, are permanently free.
Soon after the muscular layer begins this process of division, it ad-
Fig. 25.

Solitary Salpa, — female, at the time when it escapes from the body of the male, viewed from the
side, with the neural surface uppermost: a, test ; h, outer tunic; d, digestive organs; c,
branchial aperture ; g, atrial aperture ; h, branchial cavity; i, atrial cavity; I, epipharyngeal
fold ; m, endostyle ; n, gill ; p, placenta ; u, stolon ; v, ganglion ; w, languette ; x, elaeoblast.

heres to the inside of the outer tunic, in which situation the muscles
of the adult are always found ; and this tunic has accordingly been
called the muscular tunic, although, as we have seen, the muscles
really belong to the atrial portion of the body. The muscles of the

124

BULLETIN OF THE

branchial and cloacal apertures do not appear to be derived from the
atrial tunic, as they make their appearance in the positions which
they subsequently occupy. See Figs. 24 and 33.

The Branchial Slits. — During the changes which have been de»
scribed as taking place in the lateral atria, the mid-atrium has
increased in size (Figs. 23, 24, and 25, i), and at last extends almost
from the stomach to the ganglion (v). "When seen in section (Figs. 21
and 33, i), it presents an irregularly triangular outline, with its base

Fig. 26.

Fig. 2C>, from Huxley: 6, outer tunic; !, epipharyngeal ridge; m, enclostyle ; r, heart; v,
stolon ; 2, central tube of stolon, which in Huxley's figure stops short, ami does not reach the
outer wall of the heart ; but as its connection with the latter is unmistakable, the figure lias
been altered accordingly.

parallel to the posterior neural surface of the branchial sac, which
surface, like the corresponding one of the atrium, is flattened so
that they would be shown as parallel by a section at right angles
to the one represented in Fig. 33. The branchial and atrial tunics
now unite upon each side, so that the sinus (u, Fig 4) is converted
into a tube which communicates, at its posterior end, with the heart
and periviscereal sinus (Fig. 24), and at the anterior end with the
neural sinus. This tube is the gill or " hypopharyngeal band," and it
is evident that its cavity is part of the primitive body cavity, and its
walls are derived from or rather are parts of the branchial and atrial
tunics. The centres of the two regions upon the sides of the gill,
where these two tunics have become united, are now absorbed, so that
a single long and narrow branchial slit is produced upon each side of
the gill. The branchial cavity is thus thrown into communication
w'ith that of the atrium, and the upper surface of the latter now

MUSEUM OF COMPARATIVE ZOOLOGY. 325

unites with the outer tunic, and the external atrial opening is formed
by absorption. See Figs 1, 24, 25; 3, 38, and 34, g*

The Heart. — I was prevented, by lack of time, from making a.
btiies of observations upon the manner in which the heart originates,
and accordingly quote what Leuckart (p. 55) gives upon this subject :
" The heart presents, at the earliest stages, an oval form, and lies in
the space between the nucleus and the ventral surface above the
placenta." See Fig. 24, k, of the present paper. " One end is di-
rected obliquely upward and backward, and the other forward and
downward. At first, as already stated, it is a solid mass of cells, with-
in which a cavity gradually appears [according to Vogt, p. 84, the
heart is hollow from the first] and thus becomes transformed into a
pouch, whose walls quickly become thin, and also, very early exhibits
a pericardium. The first faint pulsations are separated from each
other by long intervals, but are to be seen at a. period when the ends
of the heart seem to be closed. A circulation becomes visible only
after most of the other organs are formed ; but as the blood contains
no globules during the earliest stages, the absence of a circulation
cannot be absolutely affirmed."

The Nervous System. — The ganglion, when first observed (Fig. 22,
v), was a hollow oval chamber, with indications of an opening at
the anterior end. As it was impossible to devote much time to the

* Most of the writers upon Salpahave entirely mistaken the nature of the branchial
sac, and its relation to the typical form among the Tunicata, and Leuckart seems to be
the only one who has recognized the fact that the so-called "gill" is simply the sinus
between two large branchial slits. His statement is clear, and, as far as it goes, cor-
rect. He says (p. 56) : "Kurz nach der Aushohlung der Ganglionkapsel beobachtet
man in der Riickenwand des Embryos eine neue Bildung. (Tab. 11, Fig. 6.) Es
entsteht hier in der Mitte, zwichen der Ganglionkapsel und der Wurzel des Nucleus
wie fruher im Innern des Embryonalkorpers, eine lichte Stelle, die sich allmahlig'in
einen langlichcn Hohlraum verwandelt, und jederseits durch die Wand der Athem-
hohle hindurchbricht. Die Innenlage der Riickenwand, die Anfangs beide Hohlen von
einander trennte, wird durch diesen Durchbruch in einen cylindrischen Strang verwan-
delt, der von der Wurzel des Nucleus nach dem spatern Nervenknoten hinzieht, und
naturlicher Weise nichts Anderes, als die erste Anlage der Kieme sein kann. Die
Hohle durch welche die Kieme von der Kurperwand abgetrennt wird, ist die Kloak-
hohle, die also auch bei den Salpen, als ein eigncr, von der Athemhohle (im engeren
Sinne) verschiedener Hohlraum ihren Ursprung nimmt."

As far as this goes it is correct, but he makes no mention of a special tunic around
the cloacal chamber, and says nothing of the lateral atria, or of the origin of the mus-
cular bands.

326 BULLETIN OF THE

study of this organ, Leuckart's account is again referred to, in order
to furnish as complete an account as possible of all that is known
of the evolution of all the organs of the foetus.

According to Leuckart (page 56), "The mass of cells, which is the
first indication of the ganglion, and which lies upon the anterior end
of the branchial sac, diagonally opposite the heart, to which it is not
inferior in size, soon shows traces of a cavity, surrounded by very
thick walls, and it continues in this condition for some time. After
the remaining organs have gradually been developed, and histological
differentiation begins, the cavity of this organ becomes filled up, thus
transforming it into a solid mass of cells, which is easily recognized
by its great size, and is surrounded by a capsule. The ganglion
proper does not seem to exist before this stage is reached, for the
primitive mass of cells with the cavity which appears within it are
to be regarded, not as the ganglion, but as the capsule within which
the ganglion is to be formed."

I have now described the way in which all the organs of the soli-
tary Salpa are formed, with the exception of the ovary, and the stolon
which gives rise to the chain ; and these can be best treated in con-
nection with the history of the development of the chain-salpa.

Development of the Salpa-Chain.

When the solitary embryo has reached the stage of development
shown in Figs. 24, 25, 33, s, at which time it is about one thirtieth
of an inch long, the tube which is to give rise to the chain appears
within its body, and is at first simply a cupdike protrusion of the outer
tunic into the cellulose test which now surrounds the embryo. This
cup (Fig. 25, u) is situated midway between the nucleus and the
placenta, upon the hamial side of the body and directly opposite
the heart, and its cavity is a diverticulum from the sinus system,
into and out of which the blood passes. Since the wall of the cup
is derived from the outer wall of the body cavity, while the heart
is upon the inner or branchial tunic, it is plain that the cavity
of the cup is separated from the pericardium by the width of the
sinus system. A small bud-like protrusion now appears upon the
surface of the pericardium, and lengthens so as to form a long rod,
which extends across the sinus and projects into the cavity of the
cup, as shown in Huxley's Fig. 4, Plate XVI., which is copied as Fig.

MUSEUM OF COMPARATIVE ZOOLOGY. 327

26 of the present paper. This rod, crossing the blood current at
right angles, more effectually diverts this into the cup, so that a
steady stream now passes into and out of the latter, which rapidly
lengthens so as to form a tube projecting from the outer tunic into
the cellulose test (Fig. 27).

The prolongation from the pericardium (Fig. 27, 2, and Fig. 7, 2)
also lengthens, and reaches almost to the tip or blind end of the tube,
and soon shows traces of a central longitudinal cavity (Fig. 27, 3),
which appears to be entirely closed at both ends. In a cross-section
at this stage we should have : first, the outer tube, thin-walled and
derived from the outer tunic ; within this a chamber continuous with
the sinus system, and within this a second tube, derived from the
pericardium, with very thick walls and a cavity without connection
with any of the pre-existing cavities of the embryo. This inner tube
now becomes flattened until its edges unite with the inner wall of the
outer tube, the cavity of which thus becomes " divided by a partition
into two canals, which are distinct for the whole length of the tube,
except at its very extremity, where they communicate just as the two
scalar of the cochlea do ; and it thence happens that, in the living
animal, a constant current passes up on one side of the partition and
down on the other, the direction of the two currents being generally,
but not always, reversed with the reversal of the general circulation"
(Huxley, p. 573).*

* Huxley's figures of the early stages in the formation of the chain are very correct,
and exhibit the relations of the various parts correctly ; hut the passage above cpioted
includes nearly all of his description, which is correct as far as it goes, hut very brief.
Vogt also (Sur les Tuniciers nageants de la Mer de Nice, p. 36) gives a correct account
of the tube, but Leuckart, although he was acquainted with and refers to Huxley's
description, disputes its correctness. It is so easy to make observations at this period,
and their result is so satisfactory, that it is hard to understand how such a difference
of opinion could arise; but as Leuckart's statement is made with the greatest con-
fidence, it cannot be passed without notice, and is accordingly quoted here: "Man
hat behauptet, dass der rohrenformige Keimstock der Salpen aus mehreren ubereinan-
der gelegenen Hiiuten bestehe. Ich habe indessen — abgesehen natiirlich von der
iiussern Cellulosescheide, die sich bei der Entwickelung der Knospen in keinerlei Weise
betheiligt — vergeblich versucht, diese beiden Haute darzustellen. Das Keimrohr der
Salpen zeigt nur eine einzige Substanzlage, unci hat eine einfach-zellige Besehaffenheit.
.... Der Hohlraum, den die Keimrohre einschliest, communicirt, wie wir schon
fruker beschrieben haben, mit dem Lacunen-system des miitterlichen Leibes. Man
sieht auf das Deutlichste, wie die Blutku'rperchen an der einen Seitenwand der Keim-
rohre emporsteigen und spat&r an der entgegengesetzten Wand wiederum in den Kries-
lauf des miitterlichen Korpers zuriickkehren " (page 69).

328

BULLETIN OF THE
Fig. 27.

Very young stolon : b, outer tunic ; c, wall of pericardium ; r, heart ; 1, sinus cavity ; 2, central
tube of stolon ; 3, cavity of central tube ; 4, constrictions iu outer wall of stolon ; y, y, ovaries.

At about this period a long club-shaped mass of protoplasm
(Figs. 27 and 28, y) appears within each of the sinus chambers of
the tube; and soon after the constrictions (4) make their appear-
ance upon the outer wall. We now have all the parts which are
to take part in the development of the chain, and it will be best to
state here the future history of each, and to give an outline of the
development, before entering into the details.

The wall of the outer tube, which, as we have seen, is part of the
outer tunic of the solitary form, is destined to form the outer tunics

Fig. 28.

Proximal portion of stolon a little more advanced, and rotated 90° from the position represented
in Fig. 27 : 6, c, r, y, 1, 2, and 4 as in Fig. 27 ; 3, thickened edges of inner tube ; s, germina-
tive vesicles.

MUSEUM OF COMPARATIVE ZOOLOGY. 329

of the zooids, and the constrictions upon it indicate the bodies of
the latter. Cy the deepening of these constrictions, each of the sinus
chambers, which are diverticula from the body cavity of the solitary
form, becomes divided up to form the body cavities of the zooids
upon one side of the chain.

The central partition gives rise, near one edge, to a row of bud like
protrusions upon each side, which become the branchial and digestive
organs of the zooids of each side of the chain ; while a similar
double row, upon the other edge, give rise to the ganglia. It is prob-
able that the cavities of the branchial sacs and gauglia originate as
lateral diverticula from the tubular chamber of the partition into
these buds; but this, for reasons which will be stated presently, could
not be determined with certainty. The club-shaped organs within the
sinus chambers become divided up into single rows of eggs, one of
which passes into the body cavity of each zooid at a very early
period of development.

It will be seen from this account that, in Salpa, as well as in Pyroso-
ma, " gemmation takes place, not, as in so many of the lower animals
[e. g. the Hydrozoa and Polyzoa], by the outgrowth of a process of
the body wail whose primarily wholly indifferent parietes become
differentiated into the organs of the bud ; but, from the first, several
components, derived from as many distinct parts of the parental organ-
ism, are distinguishable in it, and each component is the source of cer-
tain parts of the new being, and of these only." * While these changes
are in progress the tube lengthens, so that it at length encircles the
nucleus, as shown in Figs. 1 and 2, u. The constrictions upon its
surface deepen, and the wall protrudes between them, and each is soon
seen to mark off, on each side of the stolon, the body of a young Salpa,
and these soon become large enough to be visible to the unaided eye.

They do not increase in size gradually, from one end of the tube to
the other, but develop in sets of from thirty to fifty each, and the de-
velopment of all which are embraced within a set progresses uniformly.
There are usually three of these sets upon the tube of an adult soli-
tary Salpa, but sometimes there are four, two, or only one.

* Huxley, on the Anatomy and Development of Pyrosoma (p. 211).

I hr.ve taken the liberty of slightly changing this quotation, since the portion in
brackets reads, in the original, "the Hydrozoa and Polyzoa, or Salpa and Clavelina
among the Ascidians."

330 BULLETIN OF THE

The zooid3 which compose the set nearest the proximal end of the
tube are very much smaller than those of the second, which again are
sharply distinguished from those of the third, and so on ; and at a
time when the first are very rudimentary indeed the last present
all the organs of the adult, with the exception of the testis, and are
ready to be discharged into the water. The carefully drawn, but
somewhat rudely engraved figure of the fully developed chain, given
by Eschricht,* gives an excellent idea of the position and relations of the
various parts at this stage, and is, by far, the best figure which has ever
been published. The position of the zooids is uniform throughout all
parts of the stolon. The incurrent openings of all are upon the con-
vex, and the excurrent upon the concave or inner side of the tube.
The zooids of the opposite sides alternate with each other, and
their h&mial surfaces face inwards. They are not perfectly parallel,
for the visceral ends of their bodies are nearer each other than the
neural ends, which incline outward. Fig. 34 shows a few zooids from
a set which is nearly large enough to be discharged. The young
chain, when first set free, is about half an inch long, and the single
zooids, of which there are from twenty to thirty on each side, meas-
ure about one tenth of an inch in length. They grow very rapidly
and soon reach their full size, when the chains "are often a foot or
even a foot and a half long, and contain two rows of individuals, which
are united together in such a way that they stand obliquely to the axis
of the chain; the branchial openings are all on the upper side of t lie
chain as it floats in the water, while the posterior openings are all
on the lower side of the chain, close to the edge. Each individual is
connected both with its mate on the right or left side, and to those
immediately in front and behind on the same side. The succeeding
individuals in the chain overlap considerably. The chains do not
appear to break up spontaneously, but when broken apart by accident,

the individuals are capable of living separately for several days

The individuals composing the chain, when full giuwn, are about three
epiarters of an inch long."t I do not think that the separation of the
zooids makes the least difference in the length of their life, and the
full-grown chains fall apart at the slightest touch, and, unless the
water has been perfectly still for several days, chains more than four

* Undersogelser over Salperne, Tali. IV.

f Verrill, Invertebrate Animals of Vineyard Sound (p. 447;.

MUSEUM OF COMPARATIVE ZOOLOGY.

331

or five inches long are not met with, and those a foot or more in
length are very rare indeed, even when the water has been still for
some time, and are only fonnd in sheltered places, although full-
grown chain-zooids are very abundant.

Huxley states of the species studied by him, which seems to be the
same as that found on our coast, that it " but rarely happened that
even two or three adhered together, and they never formed the re-
markable free-swimming chain of other species. Generally they were
found solitary, presenting only on their lateral faces traces of their
former adhesion" (p. 57-i).

The chain-salpa, like the solitary form, moves by means of the
stream of water which is continually discharged from the atrial aper-
ture, and since the apertures of all the zooids of a chain are turned
the same way, the current produced is quite powerful, and a chain,
two inches long, placed in a glass bowl, with a pint of water, will
soon set the whole of it in rotation. As we should expect, the motion
of a chain is much more uniform and rapid than that of a single
Salpa, and they usually move in nearly straight lines, although Mr.
Agassiz states (Proc. Boston Soc. Nat. Hist.) that they sometimes
change their course to escape capture.

Having now given a sketch of the fonnation of the chain, we will
go back and follow the development of the various parts more mi-
nutely.

The Oater Tunic. — Little need be said of the formation of this
portion, — the body wall of the zooid. It is derived directly from the
outer wall of the tube, and its growth keeps pace with that of the
contained organs, which are entirely separated from it, at all the earlier
stages, by a distinctly visible body cavity (Fig. 30). The branchial

Fig. 29.

Fig. 30.

Two stages in the development of the zonids upon the stolon : 4, constrictions which extern] in-
to tlic sinus chambers, and murk off the separate /""ids; 5, the ganglia ol the zooid ; '•',
primitive digestive organs ; s, a single ovum within the body cavity of each zooid.

332 BULLETIN OF THE

and atrial apertures are formed after the embryo is considerably ad-
vanced (Figs. 31 and 32), precisely as in the solitary form. Accord-
ing to Vogt,* this tunic splits into two layers, the outer becoming the
cellulose test, and the inner the outer tunic ; nothing of this kind
was observed, and appearances seemed to indicate that the test is an
excretion from the surface of the body, as stated by Huxley (p. 585),
and Leuckart ; but this must still be regarded as one of the unsettled
points in the history of the genus.

The Body Cavity. — The constrictions (Figs. 27, 28, 29, and 30, 4)
which mark out the zooids upon the surface of the tube gradually
extend inward, and enclose part of the sinus cavity, which thus
becomes converted into a body cavity, into and out of which it is
quite easy to watch the blood corpuscles of the solitary form make
their way. This direct connection of the sinus systems of the young
zooids with that of the solitary Salpa persists long after all the
principal organs of the former have attained their essential character-
istics; and Huxley states (p. 574) that he has seen "one of the large
blood corpuscles of the parent entangled in the heart (which was then
not more than one five-hundreth of an inch long) of a very young
foetus." This connection of the two circulations may be seen with
such perfect distinctness in our species that the only way in which
we can reconcile Vogt's observations upon the circulation of the chain-
zooid of S. pinnata with those here detailed is by assuming that
there is a very decided and remarkable difference between the
species in this respect.t

Digestive and Nervous Organs, and the Egg. — In Fig. 28 part of
the proximal portion of a stolon a little more advanced than the one

* Sur les Tuniciers nageants de la Mer de Nice (p. 42).

t According to Vogt (p. 47), the chain-salpa of S. pinnata is provided with an organ,
the " stolohlast," homologous in structure and function with the placenta of the soli-
tary emhryo. He says that this is situated near the nucleus and heart, and is com-
posed of two chandlers, into one of which the blood of the emhryo makes its way, and
thus comes into close contact with the blood circulating in the sinus system of the
parent, which has access to the second chamber. He says that there is no communi-
cation between the two, and that none of the blood of the parent gains access to the body
cavity of the zooid. Unfortunately his figures, and the letters of reference especially,
are so indefinite that very little can be made out by the study of them ; but the expla-
nation has suggested itself that the " stolohlast " may be nothing more than the rudi-
mentary testicle, as Vogt says that it begins to disappear at the same time that the
development of the latter begins, and disappears entirely after this is formed.

MUSEUM OF COMPARATIVE ZOOLOGY. 333

represented in Fig. 27 is shown, and, in order to exhibit the relations
of the various parts more clearly, it is figured in the position which
the latter would occupy if it were rotated 90° upon its axis. The
side, instead of the edge, of the inner partition is therefore seen, and
one of the sinus tubes (i), with its enclosed ovary (y), is above, and
the other below it. The edges of the partition soon begin to thicken
and spread out, so that a section would present nearly the shape of a
letter H, with a cavity in the cross-bar. The four flaring edges
now divide up into four rows of bud-like prominences, and two of
these extend into the space between the constrictions which mark out
the body cavity of each zooid, and lengthen tmtil they meet upon
the median line and surround the egg, which is now found in this
position, as shown in Fig. 29. In this figure the prominences as well
as the egg seem to be in contact with the outer tunic ; but as it was
necessary to use pressure in order to obtain a side view, this appear-
ance is to be explained as caused in this way, for at a stage only a
little later, but sufficiently advanced to be studied without pressure
(Fig. 30), a very distinct body cavity is seen separating them from
the body wall. In this figure the portion which surrounds the egg
(s) has now separated from the portions upon each side of, or more
strictly before and behind it, and forms the egg capsule. The two
remaining portions (5 and 6) have also separated from the partition,
and form oval masses, which are free within the body cavity of the
zooid. One of these (5) is destined to form the ganglion, and the
other (6) the branchial sac and digestive organs. They are very sim-
ilar in shape and size, and each contains a cavLy entirely surrounded
by a thick wall. These cavities are, without driiibt, at first divertic-
ula from the cavity of the inner tube or partition, and are invisible
in Fig. 29, on account of the pressure to which the specimen there
figured was subjected; but the early changes in these parts take place
so rapidly that no specimen was found at a stage which admitted of
the determination of this point, although hundreds were examined
with this end in view. The portion which is to form the digestive
organs increases in size much more rapidly than the ganglionic por-
tion, and extends forward under the egg and ganglion, to the anterior
end of the body, and thus gives rise to the branchial sac. The sub-
sequent development of these parts docs not differ essentially from
that of the same parts in the solitary embryo.

;34

BULLETIN OF THE

The Heart. — This, when first seen, was a granular body beneath
and close to the stomach ; and its position seems to confirm Vogt's
conjecture (p. 45) that it is derived from the digestive tract.

The Testis;. — This is at first (Figs. 33 and 34, t) a muss of cells,
under the digestive organs and behind the heart, and therefore, as has
been already pointed out, in the position occupied in the solitary
embryo by the ekeoblast, with which it agrees in appearance at first,
but it does not become excessively developed, as this does; and in
Fig. 34, t, it is shown as a compact globular mass of cells. As devel-
opment advances, it spreads out over the surface of the digestive
organs, and in the adult it presents the irregularly branched glandu-
lar appearance shown in Fig. 5.

The Ovary and E<j<js. — Soon after the partition grows out from
the pericardium and divides the tube into two chambers, and before
t he constrictions make their appearance upon the outer wall of the
tube, a long club-shaped organ (Figs. 27 and 28, y) is seen within
each chamber. These seem to lie free within the sinus cavity, and
they could not be seen to be derived from any of the pre-existing parts
of the embryo or of the tube; and the way in which they make their
appearance seems to indicate that they are formed directly from tb"
blood. The elseoblast begins to disappear at about the same time,
and as the blood which passes through the latter goes directly to the
tube, it is possible that some of its nuclei or smaller cells are thus
transported to the tube, and form the basis of the new organs. This,
however, is purely conjectural, as nothing was seen which indicated
that the club-shaped masses were thus formed.

Fig. 32.

Fig. 31, chain-salpa, before the formation of the intestine. Fig. ?,'2. chain-salpa, somewhat
older, showing the manner in which tin- muscular bands are formed : b, outer tunic ; r, wall
of branchial sac; </, atrial tunic; e, branchial aperture; /, muscular girdles; g, atrial aper-
ture ; //, branchial cavity ; I, atrial cavity ; d i, digestive cavity; tl i", wall of digestive cavity;
/.■. peripharyngeal ridges; I, epipharyngeal folds; m, endostyle ; », gill; ?•, heart; s, egg ; /,
testicle ; r, ganglion.

MUSEUM OF COMPARATIVE ZOOLOGY. 335

As the tube grows, these organs lengthen also, and soon a single
row of germinative vesicles is seen extending along each of them (Fig.
28) ; they are therefore the ovaries. At the time that the constric-
tions, which arc the first indications of the zooids, appear in the
outer wall of the tube, each ovary is seen to be made up of a row of
eggs, equal in number to the constrictions ; and as the zooids are
developed, and their body cavities are separated from the sinus cham-
bers of the tube, the chain of ova also divides, so that a single eg^
passes into the body cavity of each zooid (Figs. 29 to 34, s) and
becomes suspended there by a gubernaculum, by means of which it is
attached to the wall of the branchial sac, as already described.

Since the chain-salpa, at birth, always contains a single unimpreg-
nated egg, organically connected with its body, and since this egg and
the resulting embryo are nourished by the blood of the chain-salpa,
by means of a true placenta, and since no reproductive organs have
hitherto been described in the solitary Salpa,* it seems most natural

Fig. 33.
h.

<N.

1-

Side view of a single zooid from a chain at the stage shown in Fig. 34 ; the neural side is upper-
most : b, outer tunic ; d, wall of atrial chamber ; /, respiratory muscles ; /', muscles of bran-
chial aperture ; /", muscles of atrial aperture ; g, atrial aperture ; h, cavity of branchial sac ;
i, cavity of atrium ; k, peripharyngeal ridge ; I, epipharyngeal folds ; in, endostyle ; n, gill ; o,
mouth; o", stomach ; o''', intestine; o'"', anus j r, heart; s, egg ; t, testis ; v, ganglion ; v,\
languette.

* Although I made my observations without any knowledge that the origin of
the eggs within the solitary Salpa had ever been traced, I cannot claim original-
ity for the observation ; since Kowalevsky states incidentally, in his paper upon the
"Development of Pyrosoma," that this is the case in Salpa. He gives no description
or figures, and confines himself to a bare statement of the fact, which occupies not
quite one line. The passage containing the reference will be quoted farther on.

My observations were made in August and September, 1875, and Kowalevsky's paper
on Pyrosoma was printed in August, 1875, and readied the library of the Boston
Society, where I first saw it, on October 11.

336

BULLETIN OF THE

to accept the view which has been generally held since the time of
Chamisso's famous paper ; that is, that Salpa presents an instance of
" alternation of generations." This view, in its most modern form,
may be stated as follows : " It is now a settled fact that the repro-
ductive organs are found only in the aggregated individuals of Salpa,
while the solitary individuals, which are produced from the fertilized
eggs have, in place of sexual organs, a bud-stolon, and reproduce in
the asexual manner exclusively, by the formation of buds. Male and
female organs are, so far as we yet know, united in the Salprj in one
individual. The Salpce are hermaphrodite." (Leuckart. Sulpa und
Verwandten, pp. 46 and 47.) When, however, we trace backward the
history of one of the individuals which compose a chain, and find that
the egg is present at all stages of growth, and has exactly the same
size and appearance as at the time when it is impregnated ; when we
find one organ after another disappearing until at last we have noth-
ing but a faint constriction in the wall of the tube, indicating what
is to become the animal, the conclusion seems irresistible, that the
animal, which as yet has no existence, cannot be the parent of the
egg which is already fully formed.*

Fig. 34.

Seven zooids from a fully developed chain, immediately before its discharge from the body of the
female ; the references are the same as in Fig. 33.

* The development of 'the eggs in the body of one zooid, and their passage into
the body of another produced by budding from the first is not unusual among the

MUSEUM OF COMPARATIVE ZOOLOGY. 337

Summary and General Conclusions.

The life history of Salpa may be stated briefly, as follows :

The solitary Salpa — female — produces a chain of males by bud-
ding, and discharges an egg into the body of each of these before
birth.

These eggs are impregnated while the zooids of the chain are very
small, and sexually immature, and develop into females, which give
rise to other males in the same way.

Since both forms are the offspring of the female, the one by bud-
ding, and the other by true sexual reproduction, we have not an
instance of " alternation of generations," but a very remarkable
difference in the form and mode of origin of the two sexes.

After the fetus has been discharged from the body of the male the
latter grows up, becomes sexually mature, and discharges its sper-
matic fluid into the water, to, fertilize the eggs carried by other imma-
ture chains.

The fact that impregnation takes place, not, as we should expect,
within the body of the solitary, but within that of the chain-salpa is
no valid objection to the view that the latter is simply a male, for the
number of animals whose eggs undergo impregnation within the body of
the female is quite small, and in at least one genus, " Hippocampus,"
the unfertilized ova are received into a specialized brood-sac upon the
body of the male, and there impregnated.

We can also find an analogy for the singular fact that the eggs of
Salpa always develop females, while the males originate by budding.
The fertilized eggs of the bee always develop female embryos, while
the virgin bee, or one from the body of which the spermathecse have
been removed, produces only males, and Professor McCrady has pointed
out that the production of the male bees in this manner can be best
interpreted as a process of ovarian gemmation. The reproductive
process of the bee accordingly presents a very striking parallel to that
of Salpa. We cannot fail to associate the fact that in these two ani-

Tunicata. Huxley pointed out, in 1860 (Anatomy and Development of Pyrosoma, p.
212), that each bud carries away part of the ovary of its parent, and one fully devel-
oped ovum ; and he says : " It is not a little remarkable that the first recognizable part
of the new organism should be the foundation of that structure which will eventually
develop into a creature distinct from it." Other observers have described a precisely
similar occurrence in other Tunicata.

338 BULLETIN OF THE

ruals the males are developed through a process which falls short of
perfect sexual reproductiou, and are therefore lower than the female,
as far as their origin is concerned, with the well-known fact that
throughout the animal kingdom cases occur in which the male is to a
greater or less degi«ee supplemental to, and more or less degraded in
accordance with the degree to which it is subsidiary to the female.

The supplemental males of certain cirrhipeds, the male Argonaut,
with its parasitic hectocotylus arm, and the small males carried upon
the backs of certain female spiders are well-known instances of the
existence of such a relationship between the sexes. I am not pre-
pared at present to connect these two sets of facts in any way, but we
can hardly avoid the conclusion that there is a real connection be-
tween them, and that one may furnish the means for explaining the
other.

The fertilization of the eggs within the bodies of zooids, produced
by budding from the body of that whose ovary gave rise to the eggs,
:.s not unusual among the Tunicata.

As already mentioned, this was first observed in Pyrosoma by Hux-
ley, and has since been seen in Didemnium, Perophora, Amauricium,
Botryllus, and Salpa, and there is good reason for believing that it
will be found to occur in most tunicates. The zooids of these tuni-
cates are hermaphrodite, and develop eggs of their own, which, how-
ever, must pass into the bodies of the zooids of the next generation
before they can be impregnated, and the ova which are formed in the
ovaries of this generation must pass into the bodies of the third, and
so on.

The essential difference between this process and that in Salpa, a
difference which is here pointed out for the first time, is, that since
the sexes are here distinct, the chain-salpa contains no ovary, and
the process therefore comes to an end, while the zooids of the other
tunicates are hermaphrodite, and the process may therefore go on
indefinitely.

The way in which close interbreeding is prevented in these Tunicata
is worthy of notice, and shows that in cases where this would seem
likely to occur special arrangements may exist to render it impossible.
In most hermaphrodite animals self-fertilization is prevented by the
existence of a difference in the periods at which the two reproductive
elements ripen. In the hermaphrodite Tunicata just mentioned, self-

MUSEUM OF COMPARATIVE ZOOLOGY. 339

impregnation is impossible, since the egg must pass into the bud
before it can reach a position to which the spermatic fluid can gain
access ; but the zooid which contains the egg, and the embryo into
which the latter is to develop are the children of the same parent, and
very close interbreeding would be apt to occur. if the testis became
mature before the egg had been impregnated. In Salpa, which is not
hermaphrodite, the same provision against incest exists, and seems to
have been inherited from a hermaphrodite ancestral form, which in
turn may have inherited it from a still more remote ancestor before
the peculiar method of throwing off the eggs had been acquired, in
which its object was, as in most invertebrates, simply the prevention
of self-impregnation.

As the remainder of this paper will be mainly theoretical and
speculative, I must state here that while I fully realize the difference
between observations as to the way in which a phenomenon has been,
and speculations as to the way in which it may have been brought
about, it does not seem best to omit all theoretical discussion,
although the views here advanced may be very much modified or en-
tirely replaced by subsequent discoveries.

Exact observations are permanent additions to our stock of knowl-
edge, and although they may be supplemented they cannot bo super-
seded ; while any theoretical views which are reached in the present
imperfect state of our knowledge of zoology are liable to be entirely
set aside by the discovery of new facts, the history of our knowl-
edge of Salpa shows that a theoretical interpretation may be of the
greatest utility, and yet be entirely false. Chamisso's theory of the
" alternation of generations " in Salpa has resulted in the discovery of
the numerous and instructive instances of true " alternation," which
now form so large a chapter of zoological science, although the facts
detailed in this paper show that, as applied to Salpa, this theory is
absolutely without basis. I am encouraged by this to give my views
of the relationship of Salpa, and of the origin of the separation of the
sexes, although I have kept this discussion as distinct as possible
from my record of observations.

The free-swimming Tunicata have generally been regarded as the
lowest representatives of the group, and Huxley (Salpa and Pyrosoma),
gives a scries of diagrammatic sections, to show a gradual transition
from Appendicularia through Salpa and Doliolum to Pyrosoma and

340 BULLETIN OF THE

the ordinary Ascidians. Bronn (Klassen und Ordnungen) gives a
somewhat similar series, arranged in the same order, but beginning
with a Polyzoon and ending with a Lamellibranch and a Brachiopod.
It is, of course, unnecessary to enlarge, in this place, upon the fact
that the features which have been supposed to unite the Tunicata
with the Polyzoa, Brachiopods, aud Lamellibranchs are superficial and
absolutely without scientific value, as this is now recognized by ull
who are familiar with what is known of the embryology of these
various animals.

That the position assigned to Salpa, in the series above referred to,
rests upon a false conception of the nature of the " gill " and its rela-
tions to the branchial sac in the ordinary Tunicates, has never been
pointed out, but the history of the formation of this structure, as it
has been described in this paper, shows that it is not a rudimentary
but a specialized form of the branchial sac ; and this, as well as all the
other peculiarities of Salpa, seem to be special adaptations to its mode
of life, and seem to show that it occupies a very high position among
the Tunicata.

So little is known of the life history of Appendicularia and Doliolum
that their relation to the remaining members of the group cannot be
stated with any certainty ; but as far as we now know, Appendicularia
seems to be an adult representative of the " tadpole larva " stage, and
must be regarded, for the present at least, as a very low and embry-
onic form, while Doliolum appears to stand between Salpa and the
ordinary Tunicata as a transitional form.*

Salpa lacks the " tadpole larva " stage, and this is what we should
expect in a Tunicate which had become adapted to a locomotive life.
The power of locomotion during the earlier stages is very necessary
to those animals which subsequently become fixed, and we should ex-
pect, according to the theory of evolution, that, wherever these fixed
animals are descended from a free ancestor, they should retain, dur-
ing the early stages, the free locomotive form of their remote prede-
cessors.

* During my work on Salpa specimens of Appendicularia were frequently met with,
hut no thorough study of them was made, owing to lack of time. Although a constant
search for Doliolum was kept up, only two dead specimens were met with during the
summer, so that no opportunity for filling the gaps in our knowledge of this form was
afforded.

MUSEUM OF COMPARATIVE Z05LOGY. 341

The Cirrhipeds, for instance, pass through a" nauplius " stage, which
has been retained because it is necessary, but if a Cirrhiped should
now become adapted to a locomotive life, the nauplius stage would
lose its importance, and might, in time, disappear by the process
of acceleration of development. This seems to have taken place
in Salpa. The " tadpole larva " appears to represent a form similar
to Appendicularia, from which the Tunicata are descended ; and this
tailed stage has been retained by most of the fixed Ascidians, but
in Pyrosoma, where the embryo does not need special locomotive
power, the tail is not formed, although the embryo in its development
passes through this stage, which is represented by the so-called Cyatho-
zooid.

In Salpa, not only the tail but all traces of the larval stage have dis-
appeared, and the egg shapes itself at once and directly into the per-
fect animal. As Herbert Spencer would state it, the indirect method
of development has given place to the direct,* as is so often the case
in the higher forms of a group ; the Cephalopods, among the Mollusca,
for instance.

The relations of the branchial sac and atrium may be explained in
a similar way. Leuckart has pointed out (loc. cit.) that the " gill "
of Salpa is simply a sinus between two large branchial slits, and we
have seen that the atrium of the embryo has the two lateral chambers
upon the sides of the branchial sac, which in ordinary tunicates com-
municate with the branchial chamber by the formation of the bran-
chial slits. In Salpa it has been shown that no branchial slits are
ever formed upon the sides of the sac, and that the lateral atria be-
come converted into the respiratory muscular girdles. If we attempt
to follow out in imagination, the manner in which a tunicate might
become adapted to a free locomotive life, we can see that if an ordinary
ascidian were to be loosened from its attachment the branchial current
would give it more or less motion in the water, and if this free
life were advantageous natural selection might in time lead to the
separation of the branchial and atrial openings until they reached
opposite poles of the body, in which position the current of water
would be most efficient as a motive force. (Sars calls attention to the
fact that the two openings are nearer each other in the embryonic

* According to Krohn (Wiegem. Arch. 1852, XVIII., I., 53, Taf. 2), Doliolum passes
through a larval stage, and at this time is provided with a tail.

342 BULLETIN OP THE

than in the adult Salpa, and shows that in this respect the embryo
resembles the adults of other tunicates.) As soon as this locomotive
life had been established it would be of advantage to have as much
water as possible pass through those branchial slits which lie in the
straight line connecting the two external apertures, aud accordingly
those at the base of the branchial sac would become excessively devel-
oped at the expense of those upon the sides, and these latter, being
no longer of functional importance, would tend to disappear, and if
the wralls of the lateral atria were at all muscular, as they undoubt-
edly are in all Tunicata, we can easily understand how they might be
so modified as to violently expel the water from the body, and thus
still further assist in locomotion. If Salpa is descended from a form
having the ordinary branchial sac the presence of the lateral atria in
the embryo can be understood, but if we deny all evolution or consider
Salpa a low form, they are as meaningless as the aortic arches of an
embryonic mammal would be if we did not refer them to a gill-bearing
ancestor.

In one respect Salpa is more embryonic than the fixed Ascidians.
The structure of its ganglion and sense organs is much more special-
ized than in them, and resembles that of their larvae in this respect ;
but we can see that while a fixed animal, having little need of a
specialized nervous system would be likely to possess oue much more
rudimentary than that of its locomotive larva, a form which re-
mained free throughout life would be likely to retain it in its highly
developed form.

We come now to the question : Will the theory that Salpa has been
adapted to a locomotive life throw any light upon the separation of
the sexes 1

" It must have struck most naturalists as a strange anomaly, that
both with animals and plants, some species of the same family and
even of the same genus, though agreeing closely with each other in
their whole organization are hermaphrodites, and some unisexual."
In Salpa we have such a case, which a comparison with the other
Tunicata shows to have been originally composed of two hermaphro-
dites ; the male organ of the solitary form having been converted
into the elaeoblast, which becomes excessively developed, and sup-
plies the material for the formation of the chain, while in the chain-
salpa the ovary has entirely disappeared.

MUSEUM OF COMPARATIVE ZOOLOGY. 343

The life-history of a typical tunicate seems to be about as follows :
The tailed larva becomes converted into an animal, which may be-
come sexually mature, as in Ascidia, or may remain rudimentary, a3
in Pyrosoma, but which, in all cases, develop zooids by budding, and,
in most cases at least, discharges eggs into the bodies of these zooids,
which are hermaphrodite, and discharge their eggs into other buds in
the same way. The process of vegetative reproduction by budding
is well known to be antagonistic to true sexual reproduction, and
although it is very common among the lower representatives of most
of the larger groups of animals, it is not usually found to occur
among the higher forms, and is replaced by sexual reproduction,
unless there is some special necessity for its retention, as there is
in those forms which are fixed, as the Cirrhipeds and Ascidians.
Wherever these fixed animals are united into a colony, the power to
multiply by budding is of course essential, and however disadvanta-
geous it may be in other respects, it must be preserved, and we see that
in the Tunicata it has been so modified that instead of being antago-
nistic to, it has become to a certain degree accessory to reproduction
by eggs. Wherever a colony is arranged in a definite form, as in Py-
rosoma, there must be a point at which any further increase in size
would be of no advantage ; budding will therefore continue until this
point is reached, but we should expect that the last series of zooids
in each colony would gradually lose the tendencyto multiply in this
way. After this change had taken place, no more eggs could be dis-
charged from the body, and the ovary, being now of no functional im-
portance, would also tend to disappear. We should therefore have,
first, the tailed larval stage, then a variable but limited number of
hermaphrodite zooids, and finally, a zooid with the male organs only
developed, but containing an egg derived from the ovary of the one
next before it in the series.

Suppose, now, that this series of zooids should become adapted to a
free locomotive life. The larval stage would disappear, as already
shown, and we should now have a series of hermaphrodite zooids, end-
ing with an egg-bearing male. The necessity for budding would
cease to exist as soon as the solitary locomotive life was entered upon,
and as this process is known to be antagonistic to high evolution and
great specialization, it would gradually disappear until the series had
been reduced to two, an hermaphrodite zooid hatched from the egg,

344 BULLETIN OF THE

and an egg-bearing male produced by budding. The series could not
be still further reduced to one, for at the period when it had been use-
ful it had been made subservient to reproduction by eggs, and this
necessity for throwing off the eggs into the bodies off new zooids still
existing, one generation of buds must be formed to contain them.

This " syncopation of development " leads to a diminution in the
number of egg-producing zooids, and however advantageous it might
be in other respects, it would lead to the diminution and final extinc-
tion of the species, unless the number of eggs discharged from each
ovary could be in some way increased. If the solitary salpa produced
but one bud at a time, only one egg at a time could be placed under
the conditions necessary for development, and of course, the number
of new embryos hatched during a given period would be only a little
more than one tenth as great as at that time when there were ten
egg-producing zooids in the series ; and in order to allow the syncopa-
tion to take place, the number of egg-bearing buds produced by each
egg-producing zooid must increase inversely as the number of the latter
in the series is diminished. Although most Tunicata produce only a
single bud at a time, some, as Amauricium, form a long tube which
divides up into a series of five or six buds, which develop simultane-
ously;* and in Pyrosoma, the Cyathozooid forms a stolon almost
exactly like that of Salpa, and this becomes constricted so as to form
a chain of four Ascidiozooids.t If, as there seems to be so much
reason to suppose, Salpa has been derived through a form like Dolio-
lum, from one similar to Pyrosoma, it must have begun its solitary
life with a tendency to produce several buds»nd thus set free several
eggs at a time ; and we can readily understand that, as the series
was shortened, and it became necessary for the eggs to be discharged
more rapidly, this stolon might lengthen, and thus give rise to a
greater number of eggs at once. This would demand that, in some
way, a supply of nutriment should be provided to supply the mate-
rial used in their formation, and since the number of males would
now greatly exceed that of the hermaphrodites the fertilization of
the eggs would now be amply provided for, and the development
of the testis of the egg-producing form would no longer be necessary.
As this does not develop until late in life, it would exist, during the

* Kowalevsky. Knospung der Ascidian.

t Kowalevsky. Entwickelung der Pyrosoma.

MUSEUM OF COMPARATIVE ZOOLOGY. 345

formation of the chain, only as a rudimentary mass of formative
cells, which might easily he turned to another use, and some portion
of it would accordingly be devoted to the formation of the chain ; as
no injurious effect would follow, and as more buds could now be
produced, the process would go until the testis had been converted
into the elseoblast, and we should now have a solitary female hatched
from an egg, and giving rise by budding to a chain of egg-bearing
males.*

The number of theoretical questions which the development of Salpa
suggests is very great, and only a few of those which are the most
easily discussed have been noticed here. There are many others upon
which the development of Doliolum, when it is better known, may be
expected to shed light, and for this reason they are omitted here.

The vertebrate affinity of the Tunicate is now a subject of great in-
terest, but if Salpa is a highly specialized form, departing widely from

* Kowalevsky appears to entertain a somewhat similar opinion of the relation of
Salpa to Pyrosoma, as he concludes his paper upon the development of the latter as
follows : " Wir finden hier, in geschlechtlicher Beziehung, die beiden Salpen vereinigt.

" Bei den Salpen giebt es bekanntlich zwei Generationen, in der einen entwickelt sich
der aus vielen Eikeimen bestehende Eierstock, welcher in den Stolo hineingeht, und
sich hier zu je einem einzigen Eie vertheilt, sodann die einzelnen Knospen-resp. Ketten-
salpea, in welchen weiter aus diesem Eie ein Embryo entsteht, wieder mit einem aus
mehreren Eikeimen bestehenden Eierstock.

" Bei Pyrosoma, enthalt jede Knospe audi wie die Kettensalpa das einzige grosse Ei
zur unmittelbaren geschlechtlichen Vermehrung, und wie die Salpen-Amme, den Eier-
stock mit vielen Eikeimen zur Bildung der Geschlechtsorgane der kiinftigen Knospen.
.... Mochten wir diese Bildung der vier Ascidiozooide mit ahnlichen Vorgiingen bei
anderen Tunicaten vergleichen, so fallt uns besonders in die Augen die Aehnlichkeit
mit den Salpen, bei denen die aus dem Eie sich entwickelnde Salpe noch wahrend der
embryonalen Stadien schon den Stolo bildet auf dem auch die einznenlen Knospen ange-
deutet sind. Bei den Salpen geht aber die Bildung des Stolo langsamer vo>- sich als die
der Amrae selbst, und deshalb entwickelt sich die erste fruher, wird zu einen freileben-
den Thier und um wahrend der letzten Periode ihres Lebens entfaltet sich die Kette.

" Bei der Pyrosoma ist der ganze Vorgang ganz entgegengesetzt und namentlich die
Kettenindividuen resp. die Ascidiozooiden entwickeln sich schneller, dagegen wird der
Cyathozooi<l (resp. Amme) nie zu einem freilebenden Geschopfe, sondern bildet .<■. h nur
so weit aus, um in Stande zu sein, den schon angehauften Nahrungsdotter aufzulosen
und die ernahrende Flussigkeit den wachsenden Ascidiozooid zuzufuhren. Ist diese
Aufgabe erfiillt, so geht der Cyathozooid almahlig zu Grunde und bei dem Freiwerden
der aus vier individuen bestehenden jungen Colonie der Pyrosoma ist er ganz ver-
=chwunden." — Pp. 604 and 621. It will be noticed that Kowalevsky fails to see that the
formaf ion of the eggs within the body of the solitary Salpa proves this to be the female,
and the chain-salpa a male.

346 BULLETIN OF THE

the normal course of development and omitting the larval stage en-
tirely, we cannot expect it to throw any additional light upon this
question.

The presence of a placenta is of course only an analogy with the
Mammalia, since the resemblance is simply functional and not in any
sense morphological.*

* A memoir by Todarro (Sopra lo Soiluppo, e 1' Anatomia delle Salpe, del Dott. Fran-
cesco Todarro, Roma, 1875), on the development of Salpa, reached me after I had fin-
ished writing my own account. In this he refers to an abstract published by Kowalev-
sky in 1868 (Nachrichten von der K. Gesellschaft der Wissenschaften. Gottingen, 1868,
p. 407 - 415) ; as this is not referred to by Kowalevsky in any of his later papers on the
development of theTunicata, and is omitted from the index of the Nachrichten for that
year, it had escaped my notice until my attention was called to it by the reference in
Todarro's paper. It is simply a very brief and condensed preliminary abstract, without
figures, and although it agrees in general with my own observations, it seems to conflict
with them in several important particulars. The developmental history of Salpa is so
very complicated that it is hardly possible to decide, in the absence of figures, exactly
how much weight to attach to this apparent lack of agreement ; and although anything
upon the embryology of the Tunicata by this distinguished embryologist, to whom we
owe so large a part of our knowledge of this subject, must be regarded as having the
highest authority, it does not seem advisable to rewrite my paper in order to refer to
views which, from the condensed way in which they are stated, and from the lack of
illustrations, it is very possible that I may have misunderstood. In a few cases, however,
the want of agreement is too evident to be explained in this way. He says (pp. 109,
110) "that no direct communication, such as has been stated by some to exist, is to
be found between the cavity of the placenta and the body cavity of the mother.''

The animal which contains and gives birth to the embryo is, according to our view,
not the mother, but the male nurse. Regarding the existence of a communication be-
tween the body cavity of the nurse and the inner chamber of the placenta, I can only
reiterate my statement that, at all stages, from the first appearance of the cavity of the
gastrula until the embryo is fully formed, the blood of the nurse can be seen passing
into and out of the cavity of the placenta. It is possible that there is a difference be-
tween the various species in this respect, for Vogt and Todarro agree with Kowalevsky
in stating that at one period at least there is no such communication, while most of
the other writers are equally confident that the facts are as I have stated them.

His account of the formation of the stolon also differs somewhat from mine. He
says (pp. 412, 413) that this is made up of the following parts : (1) the outer wall, de-
rived from the outer tunic of the parent ; (2) the digestive tube, derived from the
intestine of the parent ; (3) two cloacal tubes, continuations of the posterior ends of
the cloaca of the parent ; (4) a bunch of cells, which gradually lengthens, becomes tubu-
lar, and gives rise to the ovaries of the chain-salpaa ; (5) a tube which becomes converted
into the nervous systems of the chain-salpae. According to my observations, the diges-
tive tube is not derived directly from the digestive organs of the parent, but from the
pericardium. The two cloacal tubes mentioned are without doubt the safne as those
which I have called the sinus-chambers of the stolon ; and the fact that the blood cir-

MUSEUM OF COMPARATIVE ZOOLOGY. 347

culates within, not around tliern, proves that their cavities are continuations, not of the
cloacal, or, as I have termed it after Huxley, the atrial cavity, but of the sinus system
or body cavity of the parent. Instead of one ovarian rod, I found two, and failed to
discover that they are hollow. In this paper Kowalevsky says that this tube gives rise
to the ovaries of the chain-salpae, although in his paper on Pyrosoma, published in 1S75,
and already quoted, he seems to agree with me in holding that they develop eggs before
the chain-salpa: are formed, and discharge a single egg into each one of these. He
fails to see, however, that this proves the solitary salpa to be the true female. His
fifth or nerve tube I did not find.

I was unable to homologize the placenta of Salpa with any organ of any other Ascid-
ian. Leuckart compares the foot of a Lamellibranch, the tail of Appendicularia, and
the placenta ; but this comparison seems to be rather fanciful. Kowalevsky suggests
that the placenta may be the homologue of the cyathozooid of Pyrosoma, and the first
generation of Doliolum. He says (p. 414) : " At the close of this communication I may
be permitted to call attention to the general analogy which is to be remarked between
the development of Salpa, Pyrosoma, and Doliolum. In Salpa the egg forms an em-
bryo which divides into two parts ; one forms the placenta, and the other the embryo
proper, which bears a dorsal [haemal] stolon from which the sexual individuals bud.
The egg of Pyrosoma forms a very rudimentary embryo, which produces, through
budding, four embryos, which now, in turn, develop four sexual individuals from a
dorsal stolon. The egg of Doliolum forms a perfect but sexless individual, which puts
forth a ventral stolon, from which are formed individuals with a dorsal stolon, out of
which the sexual individuals bud. From this comparison it is manifest that, between
the placenta of Salpa, the rudimentary embryo of Pyrosoma, and the free-swimming
stolon-bearing Doliolum an analogy exists ; that, in a word, we here have before our
eyes the different steps in the same developmental process."

I think there can be no doubt that Salpa has originated in a manner somewhat simi-
lar to that here pointed out ; although the fact that the solitary form or egg-embryo is
not sexless, but a female, would seem to indicate that the relation between the three
genera cannot be exactly as it is here described. In my account of the manner in
which the evolution of Salpa may be supposed to have taken place, I purposely refrained
from referring to Doliolum, as we know so little of its development. A complete his-
tory, by one observer, of all the stages of one species of this genus, from the egg
through all the alternations around to the egg, would be of the greatest interest. At
present our knowledge is made up of fragments by various observers of isolated
stages in the development of various species.

The Memoir by Todarro now remains to be noticed. This is an elaborately illustrated
quarto of 150 pages ; and the observations recorded, as well as the conclusions
reached, are so utterly at variance with all that has been done by previous observers
that it seems impossible to reconcile them. According to this writer, Salpa is the syn-
thetic type of all the Vertebrata, and presents, during its development, peculiarities
which are characteristic of each of the classes of this group, including the Mammalia.
It is an allantoi'dian vertebrate, developed in a true uterus, which is composed of a mus-
cular, a vascular, and a mucous layer ; and after impregnation the neck of the uterus
becomes closed by a plug of mucus, and the embryo forms an allantois, exactly as in
the higher Vertebrata. The sections which are represented in the figures are so strik-
ingly like those of the earlier stages of the higher vertebrates that T am unable to make
any comparison between them and my own observations The work seems to have been

348 BULLETIN OF THE MUSEUM.

done almost entirely upon sections of specimens hardened and treated with reagents,
and every ftmbryologist is well aware how untrustworthy results reached in this way
often are, unless they are carefully compared with those reached by a thorough study
of the living embryo, and also verified in every possible way. As I made no sections,
but confined my attention to the living animal, I am unable to offer any suggestion as
to the way in which what I believe to be the errors of interpretation in this paper are
to be explained, and am therefore compelled to confine myself to this very short and
unsatisfactory notice.

No. 15. — Exploration of Lake Titicaca, by Alexander Agassiz
and S. W. Garman.

III. List of Mammals and Birds. By J. A. Allen, with Field-
Notes by Mr. Garman.

The mammals and birds enumerated in the present paper were
collected incidentally by Mr. S. W. Garman, between January 1 and
March 5, 1875, while engaged in a general exploration of Lake Titi-
caca, Peru, under the direction of Mr. Alexander Agassiz. The col-
lection of birds, numbering sixty-nine species, represented by about
two hundred and thirty specimens, possesses especial interest as being
the first considerable collection ever made at this locality. Although <
but few new forms were obtained, the collection embraces a number
of the species recently described by Messrs. Sclater and Salvin, Caba-
nis and others, from the collections of Messrs. Bartlett, Whitely, Haux-
wcll, and Jelski, who during the last six or eight years have made
very large collections in Peru.* The resemblance of the bird-fauna
of Lake Titicaca to that of neighboring portions of the highlands not
far to the eastward, visited by Mr. Whitely, is shown by the fact that
of Mr. Whitely's small collection of forty-seven species, made at and
near Tinta, on the Yilcamayo, southeast of Cuzcot (11,000 feet
above sea-level), twenty-seven, or more than one half, are contained
in Mr. Carman's collection.

Mr. Garman collected almost exclusively about the immediate bor-
ders of the lake and on the lake itself, so that a large proportion of
the species are aquatic, j Several of the incessorial birds were met

* For a list of the pnpers referring to these collections published prior to 1873, see
Proc. Zool. Soc. Lond., 1873, pp. 252, 253. M. L. Taczanowski has since described
twenty new species, and published a list of 495 species collected in Central and West-
ern Peru by M. Jelski (Proc. Zool. Soc. Lond., 1874, pp. 501-505). Cabanis has also
described several new species from M. Jelski's collection (Journ. f. Orn., 1874, pp.
97-99), and Messrs. Sclater and Salvin have published additional papers on collections
made by Messrs. Whitely and Bartlett (Proc. Zool. Soc. Lond., 1873, 1874).

t See Proc. Zool. Soc., Lond., 18G9, p. 151.

| Six species are included in the list which were collected by Mr. Walter Davis at
Coroico, Bolivia, a few miles to the southwestward of Lake Titicaca, in a forest region.

350 BULLETIN OF THE

with mainly in the "tortora" fields which occupy the more shallow-
parts of the lake. The lake is flanked on the eastern and western
sides by high hills that rise quite abruptly from its shores, which are
scantily covered with short grasses, cacti, and other coarse spiny
plants. There are quite extensive plains at both the northern and
southern ends of the lake. Shrubs and low trees occur in some of
the ravines and on the island of Titicaca, but they are confined
mostly to the southern end of the lake.

Mr. Garman has been able to add interesting field-notes respecting
most of the species, which are given in quotation-marks, and are
taken mainly from his notedjooks. Several of the species arc left
undetermined, some of which may prove to be new ; but the only
ones I have ventured to describe as such are a Gallinule — a large
and strongly marked species — and an Ibis.

MAMMALS.

1. Conepatus nasutus (Bf.nnett) Guay. Three specimens were
obtained, — a female, and two young ones about half grown.

2. Auchenia Glama (Linn.) Desm. " The Llama is in this region the
beast of burden, but is never ridden. A hundred pounds is considered a
sufficient load. Should the animal be overloaded he refuses to go, and is to
be induced to move on only by lessening the amount of his load. A load
that he is willing to start with is generally one that he can carry the entire
day. "When travelling where there is a trail the herd march in single
file, the driver or drivers walking in front or behind, and urging them along
by an oft-repeated hissing, or lsJi — sh — sh.' They are never beaten and
seldom shouted at. The males are commonly used as the carriers, or " car-
gadores." To improve the wool of the Llama, which is inferior to that of
the Alpaca, the natives are in the habit of crossing the two races."

The collection embraces one skeleton and a skin.

". Auchenia Pacos (Linn.) Tseh. ''The Alpacas are raised for their
wool, and are not used as beasts of burden. As the coat is not shed each
year, if not sheared for several seasons the wool grows to be a foot to
eighteen inches in length. A year after shearing it is less than six inches.
The entire fleece is never taken off. great ragged-looking bunches being left
on the thorax and about the shoulders, 'to protect the animals.'

"The Alpaeas and Llamas form the chief wealth of the Indian. Knowing
his herd to be a source of continual profit, he is loath to part with any of it
for money, which ' must be buried in the ground.' Hence it was difficult to

MUSEUM OF COMPARATIVE ZOOLOGY. 351

secure specimens. Owing to this, and to the idea that they may sell their
luck, it took considerable bargaining to induce owners to supply us. After
a bargain is made the wife must agree to it, and, of course, she objects until
the price is raised. « It is my brother! ' 'It is my sister! ' ' It is one of my
family ! ' ' I cannot sell it ! ' ' Any other than that ! ' and dozens of similar
expressions, that are repeated over any other selected, are urged against
selling, until, thoroughly disgusted, one either throws down the price, kills
the "animal, and takes it, or goes to repeat the scene at some other hut."

Seven specimens were obtained, including a skeleton and six skins suitable
for mounting, representing both sexes, and both the black and the brown or
coffee-colored races.

4. Auchenia Vicugnia (Mol.) Desm. " The Vicunas, like the Guana-
cos, are wild, but the young are often captured and reared as pets. The
Vicuna is smaller than the Guanaco, and lacks the long hairs that in the
latter extend beyond the wool. It prefers the open pampa. When a herd
is alarmed they will run for a short distance, and then halt and turn about to
gaze at the object that alarmed them. If it approach they scamper on again,
to again repeat the manoeuvre. Their cry of warning or alarm is a peculiar,
rapidly repeated bleating bark,, somewhat resembling that of a large squirrel,
the last note being prolonged like the bleat of a young lamb, and may be
heard for more than half a mile. In March the young are born, at which
time the females are found among the foot-hills along the edge of the pampa.
We saw many females with young, but in no instance was there more than
one."

Five specimens were collected, including a skeleton and several skins suit-
able for mounting, the latter embracing both sexes and the young.

5. Auchenia Guanaco. " The Guanacos did not, like the Vicuna, trust
to their swiftness for safety, but took to the hills, where, running from sum-
mit to summit, it was very difficult to follow them. Their cry is much like
that of the Vicuna. These animals exhibit a great deal of curiosity. Hunt-
era on the plains often throw themselves on their backs, put up their feet
and hands, raise a blanket or poncho, or roll over and over to make the
game curious ; it then approaches gradually until within gunshot. Several
are thus often secured before the herd takes to flight. By concealing them-
selves in pits the Indians are able to draw them within reach of the bolus, —
a lariat, to the end of which are attached several shorter lines, each with a
weight, usually a rounded stone, at its end. The bolus is thrown in such a
manner as to twine around the necks and legs. One of the weights bought at
Juli was a fac-simile of the so-called ' sinker stones ' from Northern Europe.
It was about two and a half .inches in length by one and a half in diameter,
oval, with a groove cut around its middle for the cord. Though small, i'
was very heavy, being formed from ironstone, or meteorite, full of pyrites."

352. BULLETIN OF THE

Four specimens are contained in the collection, including a skeleton and
several skins suitable for mounting, of both the adult and young.

" Some of the habits of the Vicunas, Guanacos, Llamas, and Alpacas
deserve especial mention, particularly so since they do not seem to be gen-
erally known. One is that of lying down during the act of copulation, — a
habit common to but few if any other of the Ruminants, and one of rare
occurrence among other mammals. Another habit worthy of note is that of
their depositing their excrement in heaps, for which purpose they are said to
return to the same place day after day. In consequence, when travelling
through their haunts, one finds at short distances rounded mounds, where
for months, and perhaps for years, daily additions of excrement have been
made. Economically considered, this habit is of great importance to the
inhabitants of a region so scantily supplied with woody vegetation, the dung
of these animals being used by them for fuel. The greater part of the cook-
ing, such as it is, is done by its use, and even the government steamers on
Lake Titicaca are supplied with fuel from this source. If the pellets of
excrement were scattered about as by sheep, it would not be available, but
being deposited in mounds it is easily gathered. It is collected to such an
extent that it furnishes a profitable source of income, being sold for the use
of the steamers.

" The flesh of these animals is inferior to mutton ; the natives eat it, but
strangers use it under protest. That of the Vicunas and Guanacos is con-
sidered 'to be better than that of the Llamas and Alpacas."

6. Habrothrix sp. incog. One specimen (skin).

7. ? Reithrodon sp. incog. One specimen (skin).

8. Cavia boliviensis Waterhouse. Three specimens.

" ' Conejos.' Found in numbers near Puno, under walls and rocks. They
were to be taken only late in the evening when they had left their burrows,
or early in the morning before they had retired to them."

9. Dasyprocta Azarse Licht. Two specimens.

10. Lagidium Cuvieri Bennett. Six specimens are embraced in the
collection, which vary considerably in respect to the amount of fulvous suf-
fusion, and the distinctness of the dorsal stripe.

"'Vizcacha.' Very common in the rocky districts, but not met with on
the pampas. Invariably selects rocky places for its burrows, and builds no
mounds at their entrances, which are usually under or between large rocks.
Runs about and feeds during the day as well as at night. Is not very tena-
cious of life, being as easily killed as a rabbit. I sometimes found them a
mile away from their homes, and though closely pursued they would pass
burrow after burrow till they reached their own. The trails used by these
animals in passing to and fro are so well smoothed and worn as to be easily
traceable for long distances. Their long tails, which drag on the ground,

MUSEUM OF COMPARATIVE ZOOLOGY 353

effectually remove small obstructions as they swing from side to side in run-
ning. Their favorite haunts are among the precipitous and broken rocks on
the shores of the lake, the hillsides, and the banks of the streams. As is the
case with the beaver, new colonics are formed by the young males that are
driven from the paternal burrow by the old ones. After a young one has
established himself in a burrow of his own construction, usually a few rods
distant from the one from which he has been driven, he induces a female
to share it with him, and a new vizcacheria is founded. We found many
vizcacherias amongst the guacas or tombs. The Indians affect to believe
that they are the people (' los viejos ') who hundreds of years ago were
buried in these sepulchres and have been transformed."

BIRDS.

1. Turdus chiguanco Lafr. & D'Orb., Conitna and Moho. Five
specimens. " ' Chihuanco.' Common on the Southern shores. Notes,
nests, and eggs very much like those of T. migratorius."

2. Procnias occidentalis Sclat. One specimen, Coroico, Bolivia.
Collected by Mr. Walter Davis.

3. Tanagra palmarum Max. Two specimens, Coroico, Bolivia. Col-
lected by Mr. Walter Davis.

4. Ramphoccelus atrosericeus Lafr. & d'Orb. One specimen, Co-
roico, Bolivia. Collected by Mr. Walter Davis.

5. Basileuterus bivitatus (Lafr. & d'Orb.). One specimen, Coro-
ico, Bolivia. Collected by Mr. Walter Davis.

6. Hirundo andicola (Lafr. & d'Orb.). Two specimens ; Moho.

7. Atticora cinerea (Gm.). One specimen ; Moho.

8. Phrygilus fruticeti (Kittl.). "Abundant among the shrubbery
at the south end of the lake." Two £, one 9.

9. Phrygilus Gayi (Eyd. & Gerv.). " Common towards the southern
end of the lake." Moho and Tiquina. Two specimens.

10. Zonotrichia pileata (Bodd.). "Abundant about nearly all parts
of the lake." Coroico (Davis), Moho and Conima (Garman) ; seven
specimens.

11. Chrysomitris atrata (Lafr. & d'Orb.). Moho; four specimens.
" ' Silgarito.' Very common in the towns, inhabiting the thatch of the

eaves and gables of the houses. A very lively little fellow and a fine singer.
His song is quite as jolly as that of the Bobolink. A friend had one in a
cage that would throw himself back on his perch, partly spread his wings,
and sing as if in a perfect ecstasy whenever a handkerchief was shaken over
his head."

354 BULLETIN OF THE

12. Poospiza sp. incog. One specimen ( 9 ) ; Moho.

13. Sycalis luteola (Sparm.). "Common on the eastern side of the
lake." Four specimens ; Moho.

14. Sycalis lutea (Lafr. & d'Orb.). " Frequent on the eastern side
of the lake, occurring with S. luteola." One specimen ; Moho.

15. Sycalis uropygialis (Lafr. &. d'Orb.). "Eastern side of the
lake, associated with S. luteola." One specimen ; Moho.

16. Icterus cayauensis (Linn.). " Frequents the grassy marshes near
the shores, nesting in the reeds. Habits much like those of the Red-winged
Blackbird of the United States (Ayelceus phozniceus)." Five J, four 9 ;
Moho.

17. Cyanocorax violaceus Du Bus. One specimen, Coroico, Bo-
livia ; collected by Mr. Walter Davis.

18. Agiiornis maritima (Lair. & d'Orb.). "Rare. The only speci-
men seen was taken on Titicaca Island."

19. Muscisaxicola rubricapilla Fii. & Landb. "Common." Two
specimens; Moho and Vincoeaya.

20. Centrites oreas Scl. & Salv. " Quite common near the shores."
Moho and Conima ; six specimens.

21. Cyanotis Azaras (Naum.). Moho, two specimens. "'El Sieta
Color.' Though not uncommon, they are quite hard to secure, owing to
their habit of quickly and constantly darting about among the reeds."

22. Tyrannus melancholicus Vikill. One specimen, Coroico, Bo-
livia. Collected by Mr. Walter Davis.

23. Cinclodes fuscus (Vieill.). " Common among the cacti and spiny
vegetation of the hillsides." Moho; one specimen

24. Cynclodes bifasciata Sclat. " Common among the coarse vege-
tation of the hillsides." Two specimens ; Moho.

25. Upucerthia Jelskii (Cab.) Tacz. One specimen ; Conima.

26. Upucerthia sp. incog. One specimen; Moho.

27. Phleocryptes melanops (Vieill.). " Rather common, living in
the reeds." Moho ; five specimens.

28. Synallaxis sp. incog. (Probably 5. slictothorax, or a closely allied
species.) One specimen ; Moho.

29. Fatagona gigas Gould. "Apparently not common. One only
seen." Moho ; one specimen.

30. Oreotrochilus estellee (Lafr. & d'Orb.). Moho ; five specimens.
" Common. The nests were frequently found in the caves on the shore.
The young were ready to fly about February 10."

31. ?Ciccabis sp. incog. A specimen of an owl (probably C. melano-
nota) was shot by Mr. Garman, but was accidentally lost.

32. ? Pholeoptynx cuuicularia (Mol.). A specimen of a small owl

MUSEUM OF COMPARATIVE ZOOLOGY. 355

was obtained by Mr. Garman, and subsequently accidentally destroyed,
which, from his description of it and of its habits, must have been this
species.

33. Buteo erythronotus (King). "'Aigle.' Only a few were seen."
Two specimens ; Moho.

34. Milvago megalopterus (Meyen) = Ibycter mcgalopterus Sharp.
" ' Alcamarre.' Frequently seen on the plains, walking about in search of
food ; also visiting the shores of the lake to pick up dead fishes, it being not
particular whether its food be fresh or not. It has a strong buzzard-like
odor." One adult male and two birds of the year; Molio.

35. Circus cinereus Vieill. One specimen, an adult £ ; Moho.

3G. Sarcorhamphus gryphus (Linn.). " One specimen, taken near
Pampa de Arreiros ; altitude about 10,000 feet. One seen at Lake Titi-
caca."

37. Colaptes rupicola (Lafr. & d'Orr.). " ' Carpintero.' Com-
mon. Nests in holes in the rocks. Said by the natives to be able to bore
its way through the hardest rocks, by the aid, as they believe, of a certain
much-sought plant, which it holds in its bill." Four specimens, Tiquina and
Moho.

38. Bolborrhynchus aurifrens (Less.). "Very common about all
parts of the lake, and up to a height of 14,500 feet. Large flocks were
frequently met with in the fields. Pairs were often noticed around the
gables of the huts in the towns, but whether they built their nests in the
thatch was not ascertained." Five specimens; Moho.

39. Zenaida maculata (Vieili,.). " A few seen at the south end of the
lake ; apparently not common." One specimen ; Carapata.

40. Metriopelia aymara (Kxir. & Prev.). " Rather common about
some parts of the lake." Two specimens ; Vilquechico.

41. Metriopelia melanoptera (Mol.). " Rather common at the north-
eastern end of the lake." Four specimens ; Vilquechico.

42. Gymnopelia erythrothorax (Meyen). " Abundant towards the
north end of the lake, occurring in large flocks." Four specimens ; Moho.

43. Leptoptila rufaxilla (Rich. & Bern.). One specimen ; Coroico,
collected by Mr. Walter Davis.

44. Nothoprocta Branickii Tacz. " ' Perdiz.' Common. Habits
similar to those of the partridges of the United States (Orlyx virginianus)."

45. Falcinellus Ridgwayi sp. nov. Adult. — • Upper surface and sides
of the head, malar region, chin, and throat reddish-cinnamon, becoming
lighter anteriorly and darkening posteriorly into reddish-chestnut ; neck all
round dark purplish-chestnut ; whole dorsal surface dark resplendent green,
with strong purple reflections, in some specimens the purple predominating
in all lights ; wings bright shining lighter green, with bronzy and purple

356 BULLETIN OF THE

reflections; below wholly dark purple; thighs reddish; no chestnut on the
wings or body, nor white about the base of the bill, as in F. guarauna, nor
has it any dusky markings at the base of the bill, as in F. igneus ; bill dark
red; feet (in the dry skin) black, probably purplish-black in life; bare
tibial space dark red.

Young. — Wings, tail, interscapulars, and rump uniform dark glossy green,
with faint purple reflections, which in some specimens, however, are quite
strong; head and neck dull dusky brown, the feathers of the top and sides
of the head with very narrow whitish edgings; lower surface of the body
dusky, with a slight reflection of green ; bill and feet black.

Bill, 3.85 to 5.15 ; wing, 10.50 to 11.50 ; tail. 4.25 to 5.00; tarsus, 2.80 to
3.75 ; middle toe, 1.40 to 2.80.

In size this species agrees closely with F. guarauna. The young differ
but little from the young of F. thalassina.

Mr. Ridgway, in a recent revision of this group,* based on a large number
of specimens, has recognized three American species, under the names Jbis
falcincllus, I. guarauna, and /. thalassina. On showing him specimens of
the present species he at once recognized it as a fourth species, and one not
before described. It differs from 1. thalassina Ridgway in being much
larger, and in color ; in I. thalassina the reflections being " vivid bronzed
green" instead of mainly purple, as in the present species. In coloration it
also differs almost totally from both I. falcincllus and 7". guarauna.

Mr. Garman's collection embraces thirteen specimens of this species, seven
of which are adult and six immature. The series presents a considerable
range -of variation in color, size, and other features.

" ' Chihuanquira.' Not rare. Frequents the flats along the shores or
streams, wading in the shallow water, and searching in the mud with its
long bill for the worms, etc., upon which it feeds. Its flesh is very tender
and of excellent flavor."

Thirteen specimens ; Moho, Conima, and Vilquechico.

46. Theristicus melanopis (Gm.) Wagl. " ' Banduria.' Rare. Only
two were seen during the voyage. The flesh is very hard and tough. Our
cook, disgusted with his experience, declared that cooking only made it
worse. The specimen secured flew about two miles after receiving its
death-wound, finally dropping dead from mid-air."

One specimen ; Conima.

47. Vanellus resplendens (Tsch.). " ' Centinella,' Spanish; ' Leke-
leke,' Indian. Common on the flats and pampas. Considerably annoys
the hunter by warning the game of his approach. Keeping at a safe dis-
tance, it continually utters the peculiar cry from which it derives its Indian
name." Six specimens ; Moho.

* American Naturalist, Vol. VIII, p. 110, Feb., 1674.

MUSEUM OF COMPARATIVE ZOOLOGY. 357

48. Thinocorus Orbignyanus Geoff. & Li?ss. Five specimens.
" ' Pocoo-pocoo ' (Indian name, from its call). Common up to the height
of 17,000 feet. When flushed they fly swiftly for a short distance. The
female skulks and runs, hut the male boldly mounts a rock or other promi-
nence to keep watch. When started a second time he either runs or flies a
short distance to another high point near by."

49. Himantopus brasiliensis Bkeiim. " ' Kaitche-kaitche.' Named
from its cry, by the Indians. Only two were seen."

One specimen ; Juli.

50. Tringa Bairdii Coues. " Quite common at some localities." Five
specimens ; Moho.

51. Gambetta melanoleuca (Gm.). '• Not uncommon." Five speci-
mens ; Moho and Conima.

52. Gambetta flavipes (Gm.). ''Very few seen." One specimen;
Moho.

53. Nycticoraz obscurus Box. " ' Bobo." Numerous small flocks of
six or eight were seen." Five specimens; Moho.

54. Fulica gigaiitea Eyd. & Soul. " ' Ajoia.' The only locality in
which this bird was found is a mile or two north of Juli. It is said they are
much more rare than formerly, in consequence of the destruction of their
eggs by the natives. The pair in the collection had frequented the locality
in which they were secured for several years, but had been unable to in-
crease their number. Found in company with the ' Choca' (F. ardesiaca),
and has similar habits." Two specimens ( g and 9) '■> 3u\'\.

55. Fulica ardesiaca Tscn. " ' Choca ' is the common name given it
by the Indians, in imitation of its cry. One of the most common birds
about the lake. Wherever we found ' totora ' (reedy shallows), there we
found ' Chocas.' Many were so fat and heavy as to be unable to fly. While
skirting the totora in our boat, it frequently happened that, when nearing a
nest, the male bird would leave the cover and swim boldly out into the open
lake. If followed he would lead on until far from the nest, when he would
rise and fly into the reeds at some point quite distant from the nest. The
eggs and young are taken by the Indians, and the birds are easily domesti-
cated. Those seen among the. fowls about their huts seemed quite tame,
though not presenting so fine an appearance as the wild birds. Their flesh
is quite palatable when well cooked." Ten specimens ; Moho, Carapata,
and Aeherache.

56. Gallinula Garmani sp. nov. Similar to G. galeata, but much larger
and darker. Above very dark bluish-cinereous, darkest on the head and
rump. White markings on the wings, sides of the body, and lower tail-
coverts, the same as in G. galeata, but there is rather less white on the
abdomen. The brownish-olive on the back, rump, and secondaries, which is

358 BULLETIN OF THE

so marked a feature in G. galeala, is so much darker in the present species
as to be often clearly distinguishable only in favorable lights. It is most
marked on the interscapulars and secondaries. Rostral shield much larger
and of a darker red than in G. galeala, and extends further up on the
head. Bill, 1.07 ; head, 2.30 ; wing, 8.50 ; tail, 3.75; tarsus, 3.50 ; middle
toe, 3.30.

This species is readily distinguished from G. galeala, its nearest ally, by
its much darker colors and very much larger size. The difference in general
size between the two is indicated by a comparison of the length of the wing
and tarsus. While the folded wing in G. galeala averages about G.50 to
6.90, the same measurement in the present species ranges from 8.00 to 9.10;
while the tarsus averages lmt about 2.00 in G. galeala, in the present species
the same measurement is 3.00 to 3.75.

The collection embraces seven specimens, among which there is consider-
able variation, both in size and color; but the smallest specimens greatly
exceed in size the largest specimens of G. galeala. Some specimens present
a decided olive-brown tint over the middle portion of the dorsal surface,
while in others it is scarcely perceptible.

57. Bernicla melanoptera Eytox. " ' Uyato.' Not common. Diffi-
cult to approach, and very tenacious of life." Two specimens ( £) ; Moho.

58. Querquedula cyanoptera (Vikill.). " ' Pato Colorado.' Rather
rare." Four specimens (iwo J, two 9); Achecachc.

59. Querquedula flavirostris (Yieii.l.). '• Common about all parts
of the lake. Young obtained in February." Seven specimens; Conima,
Vilcruechico, and Guicha.

60. Querquedula puna (Tsch.). " Rather common. Young ones met
with in January." Four specimens (one young) ; Moho.

61. Dafila spinicauda (Vieill.). " One of several birds called by the
Spaniards ' Pato de las Cordillieras.' Rather rare, two or three pairs being
all we saw in two months." One specimen (<£) ; Carapata.

62. Erismatura ferruginea Eytox. " ' Pana.' Very common. Its
food consists of seeds of plants, mollusks, worms, etc. The young were less
than half grown at the first of February. They dive very quickly, and are
able to remain under water for a considerable length of time. Never try to
escape by Hying. Though exceedingly fat, they are quite tolerable for the
table." Fifteen specimens, including <£, 9> an(l young, ani* a^so several
skeletons ; Moho, Achecache, and Vilquechico.

63. Merganetta armata Goui.d. " Rare." One specimen ; Moho.

64. Phalacrocarax brasilianus (Gm.). '• ' Miji.' Abundant. They
sit much of the time on the rocks at particular localities, where the rocks
have become whitened with their excrement, whence they fly at quite regu-
lar intervals to their feeding-grounds." Four specimens ; Carapata.

MUSEUM OF COMPARATIVE ZOOLOGY. 359

65. Larus sen-anus Tscn. " ' Gaviota.' Common. Nest about the
first of February, on little knobs of earth that project above the water.
Though their nests are frequently robbed, they continue to frequent the
6ame locality." Four specimens ; Moho.

66. Rhynchops nigra Linn. " ' Aradora.' Rare ; perhaps accidental.
The single specimen taken was seen by many people, but none had seen it
on the lake before, and regarded it as a stranger. The sailors, who had
seen it before along the sea-coast, were sure it did not belong here."

67. Centropelma micropterum (Gould). <" Sondeador.' ' Samboul-
lidor.' Very common about all parts of the lake, where the water is at all
shallow. Feeds on fishes, batrachians, etc. In February younf were taken
about two-thirds grown. They are unable to rise from the water, but by
flapping their rudimentary wings and striking the water with their feet they
manage to progress quite rapidly for a considerable distance. They dive
quickly at the discharge of a gun, — so quickly that unless taken unawares
will dodge the shot, — and escape, often swimming a long distance under
water before reappearing." Thirteen specimens were obtained of this rare
species, which seems mainly confined to this lake.

68. Podiceps caliparseus Less. Two specimens ; Moho.

69. Podiceps Rollandi Quoy & Gaim. " ' Sondeador.' Common.
Found in company with Centropelma micropterum. Like that species it
never attempted to fly, but is very quick in diving, and hence is difficult to
shoot." Five specimens ; Carapata.

Cambridge, April, 1875,

No. 16. — Exploration of Lake Titicaca by Alexander Agassiz
and S. W. Garman.

IV. Crustacea. By Walter Faxon.

The crustacean fauna of Lake Titicaca, as indicated by the dredg-
ings of Mr. Agassiz, carried on with the assistance of Mr. S. W. Gar-
man, in January and February, 1875, is very meagre. Excepting a spe-
cies of Cypris, all the specimens collected belong to one amphipodous
genus, Allorchestes, which had hitherto afforded but one or two authen-
tic fresh-water species, ranging from Maine to Oregon and the Straits
of Magellan. Seven new species are described in this paper from Lake
Titicaca. Several of them are remarkable among the Orchestidce for
their abnormally developed epimeral and tergal spines. Some are
also noteworthy as comparatively deep-water forms of a family com-
monly regarded as pre-eminently littoral. I believe that no Orches-
tidce have heretofore been found at a depth so great as sixty-six
fathoms,* unless it be Orchestia (Talitrus) Brasiliensis Dana and Nicea
media (Dana), dredged in the harbor of Rio Janeiro (at what depth is
not specified) by the Wilkes Exploring Expedition. The marine spe-
cies usually inhabit the shore above low-water mark, and the previously
described fresh-water species are found in the shallow water of brooks,
pools, or edges of lakes. No strictly fresh-water Orchestidce have
been reported from the Eastern continent, although a few terrestrial
Orchestiac are described as inhabiting moist soil away from the sea.

* The greatest depth of the lake is 154 fathoms.

362 BULLETIN OF THE

Order AMPHIPODA.

Family ORCHESTIDjE.

Genus ALLORCHESTES.

Syn. 1849. Allorchcslcs (in part) Dana, Amer. Jour. Sci. [2], VIII. 136.

1852. Allorchcslcs (in part) Dana, Proc. Amer. Acad. Sci. Boston, II. 205.
1852. Allorchcslcs (in part) Dana, U. S. Explor. Exped. XIII. Crust. Pt.
II. 883.

1856. Allorchcslcs Bate, Rep. Brit. Assoc. 1855, p. 57 (no descrip.).

1857. Allorchcslcs Bate, Ann. Mag. Nat. Hist. [2], XIX. 136.

1861. Allorchcslcs Bate and Westwood, Brit. Sessile-eyed Crust. I. 38.

1862. Allorchcslcs Bate, Cat. Amphip. Crust. Brit. Mus. p. 34.

1866. Allorchcslcs Heller, Beitr. z. nah. Kennt. d. Amphip. d. adriat.-
Meeres, p. 4. Denkschr. d. Math.-Natur. Classe d. Akad. d.
Wissensch.

1874. Hyalclla Smith, Pep. U. S. Fish Comm. for 1872 and 1873, p. 645.

1874. Hyalclla Smith, Pep. U. S. Geolog. Geograph. Survey of Colorado
for 1873, p. 603,

First maxilla? with small uniavticulate palpi. Palpus of the maxillipeds
composed of four segments, the distal segment usually bearing a movable
spine at its apex. First antenna? shorter than the second antenna?, longer
than the peduncle of the second antenna?. First and second thoracic legs
subcheliform. Propodite of second pair larger than propodite of first
pair, and much larger in the male than in the female. Telson short and
entire.

Differs from Nicea Nicolet (as limited by Bate and Heller) in having the
telson single instead of double or cleft. The fourth segment of the palpus
of the maxillipeds is well developed, as in Nicea and Ga?nmanis, and, as in
these genera, is commonly unguiculiferous. Neither Dana, in describing
Allorchestes, nor Nicolet, in his description of Nicea * (published in the same
year), mentioned the form of the telson. The two names were therefore
synonymes. Bate, in a list of British Amphipoda, published in 1856 in the
Report of the British AssAciation for the Advancement of Science, indicates,
without describing, two genera, Allorchcslcs Dana and Galanlhis, gen. now,
which, as appears from his subsequent descriptions, were based upon the
trivial character of a different relative length of the first and second an-
tenna?, and a differently formed telson: Dana's name, Allorchestes, being
restricted to those species in which the first antenna? are (at least) as long

* Gay's Historia de Chile. Zoo!., III. 237, 1849.

MUSEUM OF COMPARATIVE ZOOLOGY. 3G3

as the peduncle of (he second antenna? and the telson entire, and his own
name Galanthis including the species with the two pairs of antennae subequal
and short and the telson cleft or double. In 1861 he suppressed the name
Galanthis in favor of Nieolet's Nicca. The proportion of the antennae and
the form of the telson brought together by Bate in his generic diagnoses are
not in reality always concomitant, and Heller for the first time properly dis-
tinguished the two genera by the character of the telson alone. Grube *
adopts the relative length of the two pairs of antenna? (at most a specific
character) as the generic distinction. All his species of Allorchestes have a
double telson, and should be transferred to Nicca.

Boeck,f apparently misled by the fact that Bate carelessly describes Nicea
Nilssonii with an entire telson, and places it under Allorchestcs,% would unite
the two genera, giving as a generic character " appendix caudalis brcvis,
crassa el Jissa." He furthermore considers both Allorchestes and Nicca
synonymous with Rathke's older Ilyale^ the type of which, II. Pontica, was
carefully described and figured with the posterior caudal stylets two-branched.
Boeck has not had access to Rathke's type, as far as I can learn ; but in a
specimen from the Mediterranean, " which is doubtless Rathke's species,"
he finds the last pair of saltatory appendages one-branched. This assump-
tion of identity, it seems to me, cannot outweigh the careful description and
illustration of the founder of the genus, unless confirmed by examination.of
the type of Ilyale Pontica.

In 1874 Professor S. I. Smith described a new amphipodous genus, Hya-
lella, from the fresh waters of the United States, differing from " Ilyale " in
having a styliform fifth segment to the palpus of the maxillipeds and an
entire telson. The so-called fifth segment may perhaps be more correctly
regarded as a movable spine, like those seen both lateral and terminal on
the caudal stylets, or like the unguis which tips the dactylopodite of the
thoracic legs. However this may be, it is quite as well developed in sev-
eral species of " Ilyale " (Nicca), and is not therefore a generic character.
Ilyalclla is then a synonyme of Allorchestes.

* Beitr. z. Kennt. d. istrisehen Amphipodenfaunn, Arch. f. Natur. 1866. pp. 382,
387.

f De Skandinaviske og Arktiske Amphipoder, beskrevne af Axel Boeck. Forste
Hefte. 1872. I am indebted to Dr. Hagen for a translation of Boeck's Norwegian.

\ Doubtless a large number of the species placed under Allorchestes by Bate in his
Catalogue of the Awphipoda in the British Museum have in reality a divided telson. In
fact, it would seem that the telson is cleft in most of the marine forms, and such prob-
ably formed the bulk of Dana's original genus Allorchestes. The only types of Dana's
species that I can discover are two specimens of A. media in the Museum of Compara-
tive Zoology. In these the telson is cleft to the base. This, however, will not affect the
synonymy as given above.

§ Zur Fauna der Krym, p. 87, PI. V. Figs. 20 - 28, 1836.

364

BULLETIN OF THE

Allorchestes armatus, sp. not.

Fig. 1.

Figs. 1-9. Allorchestes armatus: 1. Female, dorsal view (nat. s. 9mm.). - 2. Head. —3. 1st
maxilla. 4. 2d maxilla. —5. Mandible. — 6. Maxilliped. —7. Distal end of 4th segment of max-
illiped bearing a movable spine. — 8. 1st thoracic leg. — 9. 2d thoracic leg of male.

MUSEUM OF COMPARATIVE ZOOLOGY. 365

Fig. 10. FiS- n-

rig 12.

Fig 14.

Figs. 10-18. AlloTchestes armatus: 10. 2d thoracic leg of female, with epimeron, gill and incuba-
tory plate. —11. 3d thoracic leg. — 12. Section of body of female (4th thoracic segment) showing
the incubatory pouch with two eggs. — 13. 5th thoracic leg. —14. 7th thoracic leg. — 15. Abdomen,
side view. —16. 2d abdominal leg. —IT. Caudal stylets. — 18. Hairs on the integument under 625
diameters.

Body stout. Hind margin of the segments raised so as to form conspicu-
ous transverse ridges. Fore margin of the head produced into a point

366 BULLETIN OF THE

between the first antennae, and on each side in front of the eyes. Eyes
round. Epimera of the first four thoracic segments produced into prominent
spines. The spines of the first and second pair are of about the same length ;
the third somewhat longer ; the fourth longest, being about twice as long as
the third, and exceeding the breadth of the broadest segment of the body.
The first three pairs are directed downward and forward, while the fourth
project at nearly right angles to both the longitudinal and vertical axes of
the body. Telson broad, entire.

Peduncle of first antenna reaching the middle of the last segment of the
peduncle of second antenna? ; flagellum composed of twelve segments. Sec-
ond antenna much longer than first antenna ; basal segment clearly separated
from the head ; olfactory denticle prominent ; flagellum composed of thirteen
segments. Carpopodite of first pair of legs triangular, as broad as the pro-
podite, furnished with seta? on its distal margin ; palm of propodite slightly
concave, transverse ; dactylopodite curved. Second pair of legs in the male
very large; meropodite armed with prominent setae at the antero-inferior
an^le; carpopodite with a long process, setiferous at its extremity, projecting
downward and forward between the propodite and the meropodite ; propodite
lar^e, convex above and below, palm oblique, straight, with small setae ; dac-
tylopodite slender, curved. In the female, the second pair of legs are
smaller, the propodite similar to the corresponding segment of the first pair,
and not broader than the meropodite ; the palm nearly perpendicular to the
straight lower margin ; lower angle of the meropodite projecting under the
propodite as a blunt process, much shorter than the same process in the
male. Fifth pair of legs about as long as the fourth. Sixth and seventh of
about equal length, much longer than the fifth ; when extended backward
reaching considerably beyond the end of the longest caudal stylets. Hind
margin of the basipodites of the fifth, sixth, and seventh pair of legs slightly
serrate. Third pair of caudal stylets very small, curved upward, so as to
project but little beyond the telson.

The shell viewed under the microscope is furnished with rows of very
minute hairs, arranged as in Fig. 18.

Length from front of head to end of telson, 8mm- to 10rai»- Breadth from
tip to tip of fourth pair of epimeral spines, Cmm- to 10mm-

Collected at the following places in the lake: —

i^'Achacache, 11 fathoms, countless specimens.

llulfofPuno, 88 specimens.

Gulf of Desaguadero, 1 specimen.

i Chuquito, 40 fathoms, 4 specimens.

i-^uli, 60 fathoms, 25 specimens.

Between Taquili and Amantane, 66 fathoms, 2 specimens.

This seems to be the commonest crustacean of Lake Titicaca. The length

MUSEUM OF COMPARATIVE ZOOLOGY.

367

of the lateral spines is variable. In specimens from the deeper soundings
they are much longer than in those from shallower depths. The specimens
from 66 fathoms measure 10mm- between the tips of the fourth pair of spines;
length of a single spine, 4mm-; from front of head to end of telson, 8mra
Average specimens from Achacache, 11 fathoms, measure 7mm from tip to
tip of fourth pair of spines ; 9mm- from front of head to end of telson. The
former are also lighter colored and more transparent than the latter. Many
of the females are with eggs under the thorax.

Allorchestes echinus, sp. nov.

Fig. 20.

Fig. 10

Figs. 19 - 21. Allonhesus echinus : 19 Female (nat. 8. 6mm.). _ 20. Vertical section of body (4th
thoracic segment). — 21. 2d thoracic leg of male.

Body short and very stout, with four longitudinal rows of spines. One
row on each side of the median line of the back ; each spine of this row
arises from the hind margin of the terga of the first thoracic to the fourth
abdominal segment inclusive. Another row of eight smaller spines lower
down on each side of the body ; these arise from the terga of the first thora-
cic to the first abdominal segments inclusive, near their line of junction with
the epimera. Slight projection downward and forward from front margin
of the head between the first antennae ; a tubercle on each side below the
eyes. Eyes round, large, somewhat protuberant. Epimera of first four
thoracic segments large, triangular, their apices directed downward; a ridge
extends from the base down through the centre of each of these epimera to
the apex. The fourth epimeron has beside the ridge a small tubercle on the
hind margin. Fifth epimeron bilobed, with a tubercle on each lobe. Telson
entire.

Flagellum of first antenna composed of six to eight segments. Basal
joint of second antenna swollen and distinct from the head ; olfactory den-

368

BULLETIN OF THE

tide prominent ; flagellum composed of nine segments. Second pair of legs
large in the male ; inferior angle of carpopodite produced ; palm of propodite
concave, notched just above the lower angle and beset with setae. In the
female the carpopodite and propodite of the second pair of legs are of nearly
equal si?e ; lower angle of carpopodite produced as in the male. Sixth and
seventh pair of legs very long, — one third longer than the fifth. Basipc-
dites of fifth, sixth, and seventh legs serrate on their hind margin. Length
from front of head to end of telson, 5mm- to 7mm-

^Llampopata, 10-20 fathoms, 2 £ 1 £

.^Juli, 60 fathoms, 1 <J

Allorchestes longipes, sp. nov.
Fig. 22.

Fig. 23.

3

Figs. 22-25. Allorchestes longipes: 22. Female (uat. s. 10mm.).— 23. 2d thoracic leg of male —
24. End of abdomen, from above. — 25. Posterior dorsal margin of 4th abdominal segment of an-
other individual.

A longitudinal row of eleven spines along the middle of the back. The
first spine arises from the fore margin of the first thoracic segment ; the rest
from the hind margin of the first thoracic to the third abdominal segments.
The first five are short, the others long, — the eighth and ninth being the
longest. Eye round, protuberant. First to fourth pair of epimera long,
produced into a point at their lower extremities. Infero-posterior angles

MUSEUM OF COMPARATIVE ZOOLOGY.

569

of the first three abdominal segments slightly produced* Hind margin of
fourth abdominal segment in some examples has a median projection over
the telson, but in others this is reduced to the bold convexity of the whole
border (Figs. 24, 25). Telson entire. First antenna equal in length to the
distance from the eye to the fifth dorsal spine ; flagellum composed of thir-
teen segments. Second antenna equal to distance from the eye to the sixth
dorsal spine; basal segment distinct; olfactory denticle prominent ; fia<rel-
lum composed of fourteen segments. Carpopodite of second pair of legs in the
male produced at the lower angle ; propodite not very large (but larger in the
male than in the female) with a slight projection at the lower angle ; a few
setae on the lower margin. Sixth and seventh pair of legs very long, about
equal to the distance between the first and ninth dorsal spines. Third pair
of caudal stylets very short ; a few seta? on the hind margin of penultimate
segment, and one very slender seta at the tip of the terminal segment.
Length from front of head to end of telson, about 10mm-

Achacache, 11 fathoms, 12 specimens.

Gulf of Puno, 10 specimens.

Gulf of Desaguadero, 1 specimen.

Chucpiito, 40 fathoms, 1 specimen.

The specimen from Chuquito is a female with eggs. It is more transpar-
ent than the others, the first two dorsal spines longer and curved forward as
in A. lucifugax. The epimera of the first four thoracic segments are also
longer in this specimen than those obtained from other localities.

Allorchestes lucifugax, sp. nov.

Fig 26.

Allonnestes lucifuga.?, male (aid. s. llmni.)

A longitudinal row of eleven spines along the median line of the back.
The first spine arises from the fore margin of the first thoracic segment ; the

370

BULLETIN OF THE

others from the hind margin of the first thoracic to the third abdominal seg-
ments. The first spine of the series projects, almost parallel with the longi-
tudinal line of the body, as far as the front of the head. The six following
are curved forward. The last three are somewhat longer than the others.
Eye oval. First four pairs of epimera long, rounded at their lower ends.
Inferc^posterior angles of the first three abdominal segments slightly pro-
duced.

First and second antennas of nearly equal length. Propodite of second
pair of legs of male with a slight projection at the lower end of the palmary
edge ; carpopodite produced below. Sixth and seventh pairs of abdominal
legs very long, extending far beyond the telson. Telson entire. Length
from front of head to end of telson, llmm-

Juli, GO fathoms, 1 specimen.

Chufpiito, 40 fathoms, 1 specimen.

The two specimens of this species which were taken are males. The in-
'tegument is delicate and transparent, as in all the specimens dredged in deep
water.

This species resembles the last, but differs from it in the longer and pro-
curved anterior spines, and the shape of the four anterior pairs of epimera.

Allorchestes latimanus, sp. nov.

Fis 27.

Fig. 28.

Fig. 27.' Allorchestes lathnanus, male (nat. s. 12mm.). — 28. Iland of second thoracic legs more
enlarged.

MUSEUM OF COMPARATIVE ZOOLOGY.

371

Body thick. Hind margin of the sixth thoracic to the third abdominal
segments inclusive produced into a spiniform tooth on the median dorsal
line. Eye nearly round. Epimera of first four thoracic segments quadri-
lateral, their lower angles rounded. Infero-posterior angles of the first, sec-
ond, and third abdominal segments prolonged backward. Peduncle of first
antenna reaching the middle of the last segment of the peduncle of the
second antenna; length of entire first antenna two thirds the length of
second antenna. Second antenna equal to distance from front of the head
to fifth thoracic segment; basal segment and olfactory denticle conspicuous.
Carpopodite of second pair of legs in the male produced below ; propodite
broader than long, palm nearly straight, with a projection at its lower end.
(In the female the propodite is small, as usual in this sex.) Seventh pair of
thoracic legs of moderate length, not extending much beyond the telson when
stretched backward. The last pair of caudal stylets reach a little way be-
yond the telson, which is broad and entire. Length, exclusive of antennae
and caudal appendages, 7mm to 12mm-

Llampopata, 10-20 fathoms, 11 specimens.

Allorchestes longipalmus, sp. nov.

Fie. 29.

Fig. 30.

Figs. 29-31. Allorchestes longipalmus: 29. Female (nat. s. limn
■g of male. — 31. Shell seen under a high magnifying power.

Fig

31.

u

r

K-

'A

A

V

1

■30. Part of 2d thoracic

Hind margin of fifth thoracic to third abdominal segments produced into
spine-shaped teeth on the median line of the back. Eye round. Epimera
of the first four thoracic segments quadrilateral, their lower angles rounded.
Infero-postejifcr angles of the first three abdominal segments produced bo-

372 BULLETIN OF THE

hind. Telson entire, with a seta on each side of the hind margin. Pedun-
cle of first antenna about as long as the head and first two thoracic seg-
ments together ; flagellum composed of fifteen segments. Basal joint of
second antenna distinct ; olfactory denticle prominent ; distal segment of
peduncle much longer than the antecedent segment ; ilagelluin longer than
the flagellum of first antenna, composed of fifteen segments. Propodite of
second pair of legs in the male large and swollen ; palmary edge sinuous and
very long, encroaching upon the lower margin, armed with seta1 ; lower mar-
gin of propodite short ; inferior angle of carpopodite produced between the
meropodite and propodite. In the female, as usual in the genus, the second
pair of legs are weak, the propodite not larger than the meropodite, 'the
palm making nearly a right angle with the lower margin. Seventh abdom-
inal legs, when extended backward, reach the end of the caudal stylets.
The shell, seen under a high magnifying power, is furnished with small scat-
tered hairs, with here and there one of those cross-shaped figure s seen in
the integument of so many of the Orcliestidos. Length of body 9mm to 13mm
About two dozen individuals of this species were taken in the lake ; the ex-
act locality is not preserved, nor the depth of water. It is closely related to
the preceding species, but may be easily distinguished from it by the hand
of the second pair of legs.

Allorchestes cupreus, sp. nov.

Fir. 32.

Fig. 33.

Fie: 21

Fig3. 32-34. Allorchestes cupreus: 32. Female (nat.
1 Terminal segrneut of palpus of maxilliped.

10mm.) —33. 2d thoracic leg of male

MUSEUM OF COMPARATIVE ZOOLOGY. 373

Body smooth, without dorsal spines or teeth. Eyes nearly round. Epi-
mera of .first four thoracic segments quadrilateral, lower angles rounded.
Infero-posterior angles of first three abdominal segments hardly produced
backward. Telson entire.

About ten segments in flagellum of first antennae. Second antenna con-
siderably longer than first antenna, equal to about a third the distance from
forehead to the end of the abdominal stylets. Carpopodite of second pair of
legs produced downward between the meropodite and propodite ; propodite
swollen ; palm convex, setiferous, with a prominence at its base, against which
the tip of the dactylopodite closes ; dactylopodite curved, closing against
the palm throughout its whole length. Fifth, sixth, and seventh thoracic
legs short, with large basipodites lightly serrate on their hind border. Fifth
pair much shorter than the sixth and seventh, which are of about equal
length. The seventh pair, when stretched backward, reach to the end of
the telson. Many parts of the body display a coppery lustre.

In the female, the propodite of the second pair is long and narrow, not
broader than the carpopodite ; the carpopodite has but a short blunt process
at its lower angle in place of the long projection in the male. Length of
body, 9mm to llmm- About twenty-four specimens, particular locality not
preserved.

This is a stout species, resembling the last described, but differing in the
shape of the propodite of second thoracic legs, want of dorsal teeth, efcc.

Allorchestes dentatus, var. inermis.

Hyalclla dcntata, Smith, Rep. U. S. Fish Comm. for 1872 and 1873, p. 647,

1874.
Hyalclla inermis, Smith, Rep. U. S. Geogr. Geolog. Survey of Colorado for 1873,

p. 609, 1874.

Fig. 35.

Allorchestes dentatus, Tar. inermis, male (nat. s. 6mni.).
Six specimens were taken from the shallow water of the " Marjal," a marshy
tract on the western side of the lake, overflowed during a part of the year.

374 BULLETIN OF THE

They differ from specimens from the United States in having a firmer and
less transparent shell, and a little differently shaped propodite to the second
pair of thoracic legs in the male; hardly enough to warrant the establish-
ment of a new species when one considers the variability of the species
within the limits of the United States.

After an examination of a large number of Hyalclla dentata and H. iner-
mis from Utah, I am satisfied that they are but varieties of one species. The
form with dorsal teeth on the first and second abdominal segments is very
probably synonymous with Amphitoe aztecus Saussure * and Allorchestes
Kniclerbockeri Bate,f as pointed out by Professor Smith himself.

This species (var. inermis) was also collected by Mr. Agassiz at San An-
tonio, Peru, in saline water, 3,300 feet above the sea; nitrate district of
Pisagua. The specimens differ slightly from the type described from the
United States in having the fifth pair of thoracic legs a little shorter in pro-
portion to the sixth pair.

It may be well to announce here the discovery of this species during the
voyage of the "Hassler" at Puerto Bueno, Smyth Channel, Straits of Ma-
gellan. The specimens do not differ from var. inermis from the United
States. The ticket accompanying the specimens does not indicate their
fresh-water origin ; but Count Pourtales tells me that some animals were
collected at Puerto Bueno by Dr. Steindachner and himself in a fresh-water
pend and an outlet stream. The Allorchestes were probably among them.f

* Memoire sur clivers Crustaces nouveaux du Mexique et des Antilles, p. 58, PI. V.
Fig. 33, 1858.

t Catalogue of the Specimens of Amphipodous Crustacea in the Collection of the
British Museum, p. 36, PI. VI. Fig. 1, 1862.

J Among the Crustacea collected by the Thayer Expedition in Brazil are two species
of Allorchestes. One is represented by a unique female specimen taken from a canal

Allorchestes dentatus, var. gracilkornis, head.

at Campos by C. F. Hartt. It differs from A. dentatus, var. inermis, only in the second
pair of antennae, which are half as long as the body and twice as long as the first pair;

MUSEUM OF COMPARATIVE ZOOLOGY. 3/5

Order OSTRACODA.

Family CYPRIDID^.

Genus CYPRIS.

Cypris Donnetii?

? Cypris Donnetii, Baird, Proc. Zoolog. Soc. London, Pt. XVIII. p. 254, 1850.

A great many specimens of Cypris were collected by Mr. Garman among
the plants growing in the shallow water of the " Marjal." They answer to
Baird's description of C. Donnetii from fresh-water ponds, Coquimbo.

flagellum composed of thirteen segments. Length of body, 4mm- In the absence of
more specimens, I would consider this a variety {gracilicornis) of Allorchestes dtntaius.

The second species is represented by several specimens. It may be called Allorchestes
longistilus, sp. nov. Body smooth, long, and slender. Eyes nearly round. Epimera of

Fig. 37.

Allorchestes longistilus, male (nat. 8. 6""".).

first four thoracic segments quadrilateral. Infero-posterior angles of first three abdom-
inal segments produced. Telson entire, with two long seta? on the hind margin. First
antenna nearly as long as the second ; flagellum composed of thirteen segments. Car-
popodite of second thoracic legs produced below; propodite large, broadest afc distal
end; palm oblique, with large setas and a projection at the lower angle. Fifth, sixth,
and seventh thoracic legs subequal, the seventh, when extended backward, reaching a
short distance beyond the end of the telson. Last pair of caudal stylets very long, ex-
tending far beyond the tip of the telson, almost to the end of the second pair of stylets.
The female has shorter antennae and small, long, and narrow propodite to second pair of
legs. Length of body, 3mm- to 6mm- Swamp three mile3 south of Campos. Hartt.
Diners from A. dentatus, yar. inermis, in its slenderer body, longer antennas, and espe-
cially in the length of the third pair of caudal stylets.

June, 1876.

Harvard MCZ Libra

3 2044 066 303 199

Date Due

J

Sfl^H990

SEP 3 0 1997