Bulletin of the Museum of Comparative Zoology at Harvard College

m
w

HARVARD UNIVERSITY

LIBRARY-

OF THE

Museum of Comparative Zoology

BULLETIN

OF THE

MUSEUM OF COMPARATIVE ZOOLOGY

AT

HARVARD COLLEGE, IN CAMBRIDGE.

VOL. xn.

CAMBRIDGE, MASS., U.S.A.
1885-1886.

Reprinted with the permission of the original publisher

KRAUS REPRINT CORPORATION

New York

1967

Printed in U.S.A.

CONTENTS.

Page
No. 1. — Clilaniydoselaohus anguineus Garm. — A Living Species of Clado-
dont Shark. By S. Garman. (20 Plates.) 1

No. 2. — Report on tlie Results of Dredging by the United States Coast
Survey Steamer " Blake." XXVII. Report on the Specimens of Bottom
Deposits. By John Murray 37

No. 3. — Observations on the Development of Agelena nasvia. By William
A. LocY. (12 Plates.) 63

No. 4. — Studies from the Newport Marine Laboratory. XVII. Preliminary
Observations on the Development of Ophiopholis and Echinarachnius. By
J. Walter Fewkes. (8 Plates.) 105

No. 5. — Report on the Results of Dredging by the United States Coast
Survey Steamer " Blake." XXVIII. Description of thirteen Species and
two Genera of Fishes from the " Blake " Collection. By G. Browne
GooDE and Tarleton H. Bean 153

No 6. — Report on the Results of Dredging by the United States Coast
Survey Steamer " Blake." XXIX. Report on the Mollusca. Part I.
Brachiopoda and Pelecypoda. By W. H. Dall. (9 Plates.) . . . 171

No. 1. — Chlamydoselaclius anguineus Garm. — A living Species

of Cladodont Shark.

By S. Garman.

Description.

Measurements. — Total length 59.5 inches. Snout to angle of mouth 4.5, to
back of skull 4.25, to occipital pores 3.9, to end of gill-covers 7, to base of pecto-
rals 8.5, to end of pectoral 14.25, to vent 35.5, to base of ventrals 32, to end of
ventrals 38.6, to base of anal 39.75, to end of anal 47.6, to base of dorsal 42.25, to
end of dorsal 47.75, to base of caudal 48.5 ; distance from bases of pectorals to bases
of ventrals 23 ; greatest width (across ventrals) 7, width across caudal 5, width
across dorsal and anal 6.5, width of head across eyes 3.5, width of the largest tooth
between the ends of the outer prongs 0.25, length of the longest cusp 0.17 ; and
greatest circumference 11.5 inches.

Rows of teeth, ^^f • i^ • if •

Rays on first branchial arch (hyomandibular and ceratohyal) 22, on second 15,
third 14, fourth 12, fifth 9, sixth 6, and seventh none.

Hab. — Japan.

The length of the specimen described is not far from fifteen times its
diameter, or a little more than five times its greatest circumference.
An elongate body, a long subtriangular and flattened head, an ante-
rior month, a most extensive gape, jaws bristling with sharp subconi-
cal hooked teeth, and a sinister look about the eyes, give it a remote
resemblance to certain ophidia ; and the narrow isthmus between the
gills crossed by the free mantle or flap of the first gill-cover is strongly
suggestive of certain fishes. The resemblances to snakes and fishes are
only remote ; the shagreen, the fins, the teeth, the gill-openings, the
cartilaginous skeleton, etc., show the animal at once to be a Selachian,
one of the Sharks.

The single small dorsal, and the large ventrals, anal, and caudal, have
the appearance of being bunched together ; they are placed so far back
as to leave a space of almost two feet of the length entirely unrelieved
by fins, which contributes considerably toward an eel-like appearance.

The skull is short, and, jaws and suspensorium (hyomandibular) being
very long and loosely articulated, the hinder portion of the head spreads
easily till its width equals its length, and the outline from above resem-
bles an equilateral triangle, or, better, an arrow-head with barbs.

VOL. XII — NO. 1. 1

2 BULLETIN OF THE

The gape is wide. The structure of the mouth and throat is such as
to permit the creature to swallow with ease others whose bodies have
diameters as great as its own, or even greater. Both mouth and throat
are lined with shagreen. On the inner edges of the gill arches the scales
are larger. At the angles of the jaws there are neither labial folds nor
labial cartilages.

The eye is moderately large ; it is on the side of the head, over the
middle of the length of the moiith, and, from the sharp rather prominent
brow, has a savage look. The pupil is horizontally oblong. Around the
pupil the skin covering the eyeball is rough with small scales. There is
no trace of a nictitating membrane.

The slightncss of the convexity of the top of the head makes the angle
formed with the sides, in front of the eyes and around the snout, some-
what sharp. The snout extends but little in advance of the mouth.

The nostrils are lateral ; they are placed about half-way from the
eyes to the end of the snout. Each nostril is vertically elongate, and so
constructed that the upper half opens forward and the lower half back-
ward. Internally the nasal chamber is not divided. During forward
motion the water enters through the upper section of the nostril, passes
downward behind the partition and out again through the lower section.
Backward motion reverses the current. The partition divides the open-
ing, but not the chamber ; it is formed by a sharp fold pushing back-
ward from the middle of the front wall to meet a similar fold from the
opposite side. In the Notidanidae the structure is similar. Commonly
among Selachians the anterior fold takes the form of a flap partially cov-
ering the nostril.

The gill-openings are large ; the fii'st, when extended, will admit an
object of four inches or more, and the last will take one of two inches in
width. A vertical from the upper angle of the fifth touches the front
edge of the pectoral, and a third part of the sixth opening passes back
above the same fin. The arches are quite .slender. The blade-like folds
of the membrane are free for a considerable extent of their length at the
outer end. Plate V. gives the appearance in the fourth opening on the
right side. Sharp points on the edges of the gill-covers indicate the
ends of the branchial rays. The opercular flap, or first gill-cnver, is
broad and free around the neck, except for a short space behind the
occiput. A thin inner fold descending from a point in front of and be-
neath the first branchial cartilage copnects the flap with the isthmus.

As is to be expected in connection with large branchial apertures, the
S2:)iracles are very small.

MUSEUM OF COMPARATIVE ZOOLOGY, 3

An open canal, the lateral line, extends on each side from the back of
the skull to the end of the tail. Other open canals, branches of the
same system, are seen beneath and on the side of the head (Plates I. and
IV.). On the skull the canals are covered and appear as lines of pores
(Plate III.). In addition to the main lines indicated by the artist, a
transverse branch extends to the post-orbital process, where it makes a
short backward turn, then descends on the side of the face to join a line
parallel with the mouth and extending forward from the angles of the
jaws. Behind the post-orbital process, between it and the spiracle, there
are short lines and groups of small pores. The line above the mouth con-
tinues to the tip of the snout ; in front of the eye a branch passes above
the nostril, and, a short distance in front of the latter, appears on the
upper surface and turns backward as the main branch. Smaller pores
are numerous over various parts of the head.

From the bases of the pectorals to those of the ventrals is about
twenty-two inches. This section of the body is slightly compressed ; its
depth in life was probably in the neighborhood of four, and its width
somewhat near three inches. A prominent doubled or grooved keel along
the median line of the belly adds considerably to the depth. Toward
the pectoral arch and at the pelvis the keel loses its prominence ; it is
largest near the middle of the total length, where it projects three quar-
ters of an inch, and the groove has a depth of one third as much (Plate
XX., A, B). At first, the specimen being a fertile female, the promi-
nence of the keel and its folds was looked upon as a possible sexual
development, appearing while the young were carried. Study of the
structure and failure in a search for similar growths in other sharks
cause a change of opinion. From their position, shape, and extent, it is
evident the folds will furnish support to one of the theories of the origin
of paired fins. The muscle of the inside of the keel corresponds to the
rectus abdominis of other vertebrates. It differs somewhat from the other
muscle of the abdomen, as will be seen from the description given below.

Situated behind the middle of the body, on the tapering portion, and
being large, the posterior fins give the specimen the appearance of being
more uniform in size from end to end than it really is. None of the
fins are at all rigid, but, on the contrary, all are very soft, and, like the
body itself, extremely flexible. They are covered with shagreen except
near the outer edges, which are very thin and membranous. There is a
single comparatively small dorsal. This fin begins above the origin of
the anal, as is indicated by the peculiar armature of the upper edge, and
gradually rises backward to terminate in an acute point about opposite

4 BULLETIN OF THE

a similar point on the anal fin. Both upper and lower (anterior and
posterior) margins are curves, which meet in the apex (Plate XIII.).

Pectorals, ventrals, anal and caudal are large. The pectorals are
moderately long; both front and hinder margins are curved — the lat-
ter most strongly, and the curves meet in a blunt angle at the end of
the fin, which is nearer the front edge. The ventrals are placed some
distance behind the middle of the total length. They are a little larger
than the pectorals, the reverse of what is usual among sharks. Each is
broadly rounded, being about twice as long as wide, and ends in an
acute point behind the vent (Plates I. and XII.). In length the anal fin
approaches eight inches, and in depth it is close upon three. The curve
of the lower margin is tolerably regular and sharp. An acute angle is
formed by the posterior extremity.

The tail is without a pit at its root, and the fin is not divided into
lobes by a notch in its lower border. Rising very gradually from a point
opposite the beginning of the lower part, the upper portion of the caudal
fin reaches in its greatest height not more than three eighths of an inch,
and is continued downward behind the end of the vertebral column to
form more than half of the posterior border, as is proved by the changes
in direction in fibre, shagreen, and armature on the edge. The lower
portion of the fin lacks little of three and a half inches in its greatest
width, and, with the filamentary extremity, is not far from twelve inches
in length. Not a trace of the caudal notch is to be found. At its
widest, the tail is a little less than half as wide as long. Its shape is
better shown in Plate XIV. than in Plate I.

The Tceih.

Plate VI. Figs. J - 8.

As there are fifty-one rows and six teeth in each row, the whole
number of teeth in function at once amounts to three hundred and six.
In this the soft one at the inner end of each row is not counted. In a
general way a tooth may be described as three long, slender, very sharp,
subconical cusps, separated by a pair of rudimentary denticles or but-
tons, on a broad backward extended base. Variation according to posi-
tion on the jaw makes it necessary to modify the description for teeth
of the different series. All of the teeth are small ; the largest of them
is hardly a quarter of an inch in width across the tips of the cusps, and
the smallest is less than one sixteenth. On the upper jaw there are
thirteen rows on each side ; on the lower, there are twelve on each side

MUSEUM OF COMPARATIVE ZOOLOGY. 5

and one on the symphysis. Behind the teeth proper, on each jaw there
is a patch of scales similar to those on the lips at the angle of the mouth.
The anterior row iu each of these patches, being somewhat regular, was
wrongly counted as teeth in the preliminary description. Backward
the size of the teeth decreases. There is also some change in the shape,
but the change from teeth with broad base, three cusps, and two but-
tons, to scales with a single cusp, is sudden and decided ; i. e. they do
not grade into each other. A strong lens, however, is necessary to dis-
tinguish them, since in the hinder row each cusp looks much like a single
scale. In the front teeth the median cusp is but little longer than the
others ; it curves directly backward, and does not extend much beyond
the prongs of the base. The lateral cusps of the same tooth incline
laterally and curve backward. The points are slightly bent upward.
Between the cusps, on the inside of each of the lateral, and on both sides
of the median, a slight ridge runs from the base toward the apex ; it
also connects with the button. On the outside of the laterals this keel
or ridge is obsolete, except very near the base. Striations do not appear
on the first rows of teeth. At the junction of cusp and base the enamel
is inflated or swollen into a low ridge or collar around the base of the
cusp ; this ridge is marked by slight prominences and hollows, as if folds
once existing in the enamel had disappeared, leaving only these traces- of
their presence. Anteriorly the cusps are greatly bent back toward the
base ; posteriorly they are nearly or quite erect. The base is broad and
long. On its upper side a ridge runs backward behind each of the but-
tons. These ridges end in a pair of prongs, which extend beneath the
base of the next tooth in the row. In front of the prongs, between their
bases, a small pore marks the opening of a vessel which, descending for-
ward, passes to the lower side to reappear in the anterior portion of the
tooth's base. Except at the opening of this vessel, the groove, from the
notch between the prongs and forward under the base, is not open as
figured in Plate VI. ; its covering, however, is translucent, very thin,
and easily carried away. On each side of the ridge in which this groove
lies there is a concavity for the reception of the basal prongs of the pre-
ceding tooth. Outside of each of these indentations there is a rounded
prominence which is situated beneath the base of a lateral cusp. About
a third of the length of the base of each tooth extends under that of the
next behind it in its row.

Backward the characters of the teeth change. In the sixth and
seventh rows the little prominences around the base of a cusp have
become shallow plications or foldings in the enamel, which iu the ninth

6 BULLETIN OF THE

row extend half-way to the apex. In the eleventh row the folds are very-
distinct. In front the teeth are symmetrical ; those farther back have
lost some of the symmetry. Their bases looli as if pulled to one side
(backward) by the prongs. Gradually the lateral cusps become shorter,
until in the twelfth row they are hardly more than half as long as the
median. The cusps have become nearly erect and the striation is very
distinct in the hinder rows. Besides the keel at each side, a similar one
marks the front of each cusp in these rows. On these teeth the prongs
of the base are so short as to be scarcely noticeable, only a shallow in-
dentation remaining of the notch between them. Here the buttons are
merged in the ridges till they appear as projections on the sides of the
cusps, and the cusps themselves have become stouter, shorter, and more
like the scales. The changes appearing gradually in the lateral rows
have culminated in the last row, where the tooth has plicated enamel,
nearly straight cusps, a median cusp twice as long as the laterals, and
a broad rounded base without prongs or concavities and but slightly
notched in the posterior margin. For a description of a tooth of the
twelfth or thirteenth row, that of Cladodus viirabilis Ag. is not far out
of tlie way ; in fact, it agrees so well that, if consideration was limited to
that particular tooth, one could have little hesitation in naming the new
species Cladodus anguineus. Possibly the bases of the teeth of C. mira-
hilis might not accord so well. Pta-nodus springeri and P. arviatus (Pris-
ticladodus springeri and var. armatus St. J, & W.) present forms of bases
which are intermediate between those of Chlamydoselachus and Cladodus,
as shown in the numerous species figured by St, John and Worthen.

TJie Scales.

Plate VI. Figs. 9-13.

Over the entire body the scales are small and irregular in size and
shape. On the flank and belly they are polygonal plates, or depressed
lumps (figs. 9, 10), surmounted by one, two, or three sharp promi-
nences, the median of which is the stronger, in places becoming a keel.
On the tail this keel is produced beyond the base as a spine (figs, 10,
11), This spine is very sharp, has three longitudinal ridges, and
is excavated slightly or flattened beneath. About the mouth and in
particular around its angles the spines are larger, more conical, and more
erect, — more like teeth (fig. 12), Each of a few of these scales has a
small cusp on one side near its base. In the mouth, just behind the last
row of teeth, there are spines which are more slender, and which have

MUSEUM OF COMPAJIATIVE ZOOLOGY. 7

broader bases. These resemble the teeth of certain fossil species -which
have single cusps. They are hardly one fourth as large as the teeth
immediately in front of them. Where they have been worn, on the top
of the head or on the belly, the scales are not so harsh to the touch.
From each side of the lateral line elongate scales with chisel-shaped or
truncate ends reach out to meet similar ones from the other side, thus
forming a cover or protection for the canal (fig. 10). The upper edge
of the tail and its posterior border, to the lateral line, are armed by a
sharp edge of scales. The edge is formed of two rows — one from each
side — of broad, thin, subquadrangular scales, which have met on the
median line and become so closely applied as to appear a single ridge.
Each scale entering into the construction of the edging is opposed to
two others, in this manner imbricating or breaking joints. Near their
bases these plates are striated ; their distal halves are smooth. Similar
scales guard the front or upper edge of the dorsal.

Comparing the scales with those of a very young Heptahranchias pec-
torosus, we find that in the latter the shapes and sizes are much more
regular, that all are three-cusped, and that on the upper edges of tail
and dorsal there are placed side by side three series of enlarged and de-
pressed scales. On a specimen of Heptahranchias cinereus, thirty-seven
inches in length, the upper edges of dorsal and tail are covered with
slightly enlarged scales, which differ little from those of the sides, and
the scales on the lateral line — which has a dermal cover — are not en-
larged or different in any way from those of other parts of the body.
The line itself ends, in that species, before reaching the notch in the
caudal ; it is only to be traced by its pores. On a large //. maculatus
the lateral line of the scapular region is alternately open or closed for
irregular distances. The skin being thin, the canal is shallow or near
the surface. Along the edges of the open portions the scales differ little
from the others.

TJie Skull.

Plates vn. and VIII.

Between Chlamydoselachus and its nearest allies there are internal
differences which are quite as numerous and striking as the external.
The comparative length of the skull, the length of the jaws, and the
position in which the latter are suspended, again present a remote re-
semblance to the serpents. From the marked similarity in the brain,
branchialia, and in other respects, one would not expect great differences
in the skulls of this genus and the Notidanidae, yet from the skulls

8 BULLETIN OF THE

alone it is doubtful whether close affinities would be suspected. In the
Notidanidae the articulations of the jaws are as far back as in any of the
Galei, but even in them the jaws pass little behind the skull, while in
the majoi'ity of the other Selachia the suspensorium, or hyomandibular,
is directed downward, outward, or forward.

The skull of the frilled shark is suggestive of immaturity ; the thin
walls, soft cartilage, and large pores and foramina with thin edges around
them, seem to be those of a young, rather than an adult specimen.
Compared with that of Heptabranchias it agrees better with an embr^^o
than an adult. Looking at it from above, its shape may be likened to
that of the body of a guitar, the vertebral column answering to the neck
of the instrument, and the narrow section between the orbits to the
middle of its box. Across the nasal capsules the width is nearly two
thirds, and across the interorbital space nearly two fifths of the length.
The walls are very thin. In longitudinal section the thickness of floor
and roof is comparatively uniform. There is a marked contrast in this
respect if compared with skulls of Hexanchus and Heptabranchias, in
which these portions are thick and irregular (see Gegenbaur, Das Kopfs-
kelet der Selachier, PI. IV. figs. 1 and 2). The roof is not very convex,
nor is it to be called very irregular. Behind the front teeth the floor
makes a sharp bend upward, which allows the jaws and teeth to rest
nearly at the level of the bottom of the skull. The chamber is large,
and the brain small. The rostrum (a) is broad, thin, scoop-shaped, reg-
ularly rounded in front, and notched (v) at the side in front of the nasal
sac (d). Behind it the broad anterior foramen (i) extends quite to the
interorbital space. On this space a second foramen (p), open in younger
specimens, is indicated by a narrow depression or gash, not reaching
through the cartilage. A little farther back there is a rounded space in
which the surface is rugose (o). From the parietal fossa («) there are
two pores on each side, as in Heptabranchias. Professor Gegenbaur
figures four in Hexanchus also, but places them in a transverse series.
Behind the fossa a low occipital crest extends to the vertebrae. The
preorbital process (/) is moderately stout ; its outer end is unseg-
mented and rests close upon the pterygo-quadrate at its outer edge.
Above the eyes the expansions are thin and prominent. Of the supra-
orbital foramina (h) the anterior is the opening for the ramus ophthal-
micus and in front of the latter is the upper opening of the ethmoidal
or preorbital canal (e). The postorbital process is of irregular shape and
moderate breadth. There are three latei-al processes on the occipital
region. The first is seen in the paroccipital region. The second {t) is

MUSEUM OF COMPARATIVE ZOOLOGY. 9

a postparietal or epiotic process ; this process, with a pore (k) iu front
of it, recalls a similar arrangement on the skull of Rhina. The third (l),
pterotic, is formed by the backward extension of the edge of the cup
receiving the end of the hyomandibular. This pi'ocess is quite promi-
nent ; its height is more than half that of the skull, and the upper angle
is produced. The articular depression, extended by this process, meas-
ures about five eighths of an inch in length by one fourth of an inch in
height. The articulation permits the hyomandibular to lie against the
side of the skull, or to be turned sidewise at an angle of forty-five de-
grees. On the side of the skull appear the openings for the second to
the seventh pairs of nerves (2 to 7), the nostrils (c), and the orbito-nasal
canal (x). In the orbit the articular facet for the palatal or trabecular
process of the quadrato-pterygoid (tp) extends nearly to the top of the
skull, and is produced below the floor in a prominent ridge. Immedi-
ately in front of the deep cavity in which is found the opening for the
fifth pair of nerves (5) is the short style (os) connecting with the eye-
ball. In the section (Plate VII.) the opening for the optic nerve (2) is
shown to be farther forward and upward than in the Notidanidse ; the
olfactory lobe passes out but a short distance in front of it. Between
and below the openings for the fourth and fifth pairs there are carti-
laginous bridges (iv, r). The pituitary cavity (joy) is shallow, and
around the opening for the internal carotid the wall is thin. The
thread-like termination of the notochord lies close to the lower face
of the wall. Around it, in the parachordal region, the cartilage is hard
or granular.

There are no spiracular cartilages (metapterygoids).

TJie Jaws and Hyoid Cartilages.

Plates VII. -IX.

Labial cartilages at the angles of the mouth and along the jaws have
not been found.

The suspensorium, hyomandibular (km), is about three inches long,
depressed, curved, and tapers in the hinder third of its length, where
applied to the pterygo-quadrate. Its thickness is about one half of its
width. The curvature is but moderate, certainly not more than that
of a circle with a diameter of six inches. The end articulating with
the ceratohyal is small. Anteriorly the articulating surface is oblique,
forming an angle of about forty -five degrees with the shaft ; posteriorly
the shaft has tapered to less than half its width. The inner (hinder)
edge is sharper, and bears nine branchial rays (br-?-).

10 BULLETIN OF THE

The upper jaw, quadrato-pterygoid {q-pg), is a broad, thin, twisted
blade, a trifle over five inches in length. Anteriorly it is spatulate, near
seven tenths of an inch in width, and is twisted so as to rest obliquely
under the skull, the lower edge being turned outward. Posteriorly it is
more than an inch in width, strongly convex on the inner side, deeply
concave on the outer, and at the articulation with the meckelian is
twisted half-way around so as to bring the side of the extremity in
contact with the latter. An inch and three quarters back from its front
end it bears a strong, compressed, palatal or trabecular process {tp).
This process is a little more than half an inch in height, a little less in
width, and is rounded on its upper edge, where it is attached by ligament
to the skull near the top of the orbital cavity. Its outer and inner faces
are convex ; the latter being received in a concave articular depression
in the skull. At the sides and beneath, this concavity has been enlarged
by ridges of cartilage. Half an inch from the hinder end the pterygo-
quadrate bears a small hyal process, which is overlaid by a similar one
on the hyomandibular. Two inches from the same end there is a larger
process, quadratic or otic {qp), formed by a short bend in the thin
upper edge. Some of the most prominent differences between Chlamy-
doselachus and the Xotidanidso are to be seen in the attachments and
articulations of this cartilage.

The lower jaw, MeckeVs cartilage (mk), is a strong, broad, twisted
cartilage. It is broad posteriorly, and tapers gradually forward to near
the end, where it decreases in size rapidly aud presents but a small sur-
face to its fellow from the opposite side at the symphysis. It is twisted
in front to bring the sharp upper edge outward, where the cartilage
bends inward toward the middle of the snout. In the posterior two
thirds of the length the lower border is wide, with a sort of flange ; for-
ward the outer edge of this flange fades into the middle of the blade,
while the inner continues as the inner edge. Backward the lower jaw
is convex on the inner side, and deeply concave on the outer. Bending
outward at the end, the articulation of the pterygo-quadrate with this
cartilage has the appearance of taking place on the upper edge.

The Ceratohyals {chy) are moderately slender, curved, club-shaped or
bilotate anteriorly, and tapering in the posterior two fifths of their
length, where applied aud attached by ligament to the meckelians.
Though the thick front lobe articulates with the side of the basihyal,
the outer and thinner lobe underlies the hinder angle of that cartilage.
The latter reaches back above the ceratohyal to articulate with the first
ceratobranchial. In the figure the lobe has been brought too far back,
and too near its fellow on the opposite side of the basihyal.

MUSEITM OF COMPARATIVE ZOOLOGY. 11

The Basihyal (hhj/) is elongate, tongue-shaped, and posteriorly, at its
■widest, is half as wide as long. Behind the more slender anterior por-
tion, glossohyal, there is a deep excavation or hole. Behind this, be-
tween the ends of the ceratohyals, in the proper basihyal portion, it
becomes much thicker, bulges downward, and has a couple of longitu-
dinal ridges near the middle. The narrower anterior portion is convex ;
posteriorly the border is concave, and with the sides forms two angles,
the hypohyals, which are produced to meet the first pair of cerato-
branchials.

The basihyal of Hexanchus is broad and short, and the hypohyals are
not distinct. That of Heptabranchias is pointed in front and has small
hypohyals, according to Gegenbaur. Heptabranchias maculatus, how-
ever, 1 find to be nearly as in Hexanchus. The articulations in both
these cases diff"er from those described above, having more resemblance
to each other.

Tlie Branchial Cartilages.

Plate rS.

Extrabranchials are not present.

Basibranchials (h-hr). — The first or anterior of the series is distinct.
It is closely connected with the first, and loosely attached by ligament
to the second pair of hypobranchials. The second of the series is also
distinct, but closely joined with both second and third pairs of hypo-
branchials. The thii-d is fused with the corresponding pair of hypos,
has an oblique and indistinct longitudinal division, and is closely joined
with the fourth, which in turn is fused with the fifth. The pairs of
hypobranchials corresponding to the last two are mere processes on their
sides. As in other Selachia, the last of the series ends in a long spine-
like process. In transverse section the anterior four are triangular, flat
above, keeled beneath.

A complete series of basibranchials in unsegmented condition would
indicate lower rank than such as that figured by Gegenbaur (PI. XVIII.
fig. 1) in Heptabranchias. Excepting the last of the series, higher rank
of a genus is apparently accompanied by reduction or loss of basibran-
chials (see Spinax, Scijllinm, Galens, and others).

Professor Gegenbaur's figure of Heptabranchias is not to be taken as
characterizing the genus, since in any of the genera there are hardly two
species alike in respect to these cartilages. In most cases the estimate
of the value of these and other portions of the skeleton, in connection

12 BULLETIN OF THE

■with generic diagnoses, is subject to important modifications on account
of individual and specific variation. A specimen of //. xj^ctorosus before
me has the first and second fused with the hypobrancliials, and longi-
tudinally divided or partly divided through the middle, the third fused
with one of the hypobranchials, the fourth distinct but divided through
the middle, and the fifth normal. A specimen of H. maculatus has the
first appearing as a small lozenge-shaped lump, the second fused with the
right hypobranchial, the third fused with both of the corresponding pair
of hypos, and the fourth either suppressed or fused and divided.

Hypobranchials (Ji-hr). — Anteriorly these cartilages are distinct and
moderately slender ; posteriorly they become mere processes on the
sides of the basibranchials. As the articulation with the ceratobran-
chial takes place on the upper side of the posterior lobe of the inner
extremity of the preceding ceratobranchial, each hypo- is really articu-
lated with two ceratobranchials.

Ce7'atobra7ichials (c-br). — These are slender and long. The first pair
articulate with the posterior angles of the basiliyal, hypohyals. Each
succeeding articulates with the hypobrancliials on the upper side of the
posterior lobe of the broadened inner extremity of the one immediately
preceding. The sixth pair is much stouter, articulates directly with the
last basibranchial, without hypos, and bears a downward-inflated margin
on its outer extremity.

The epibranchials {e-br) are slender. The upper ends are thicker and
broader and the pharyngobranchials articulate against the outer side.

The pharyngobranchials (p-br) are slender and elongate, with the ex-
ception of the sixth pair. The latter are short and flexible ; they are
fused with the epibranchials.

The Vertebral Column.

Plate X.

The notochord is persistent. For a short distance back of the head
there are vertebral constrictions ; this condition only obtains in the few
vertebrse that are somewhat calcified (fig. 3). Here the condition is sim-
ilar to that in Centrophorus, as figured by Kolliker, or in the posterior
vertebra; of Heptabranchias. Behind this, in the much greater portion
of the column, tlie vei'tebrse are much less distinct, and the notochord
maintains a imiform diameter (fig. 2). Forward the vertebrae can be
distinguished with readiness, but in the middle of the body, though ex-
ternally the different segments are well marked (fig. 1), a longitudinal

MUSEUM OF comparatrt; zoology. 13

section shows so much uniformity of appearance that it is only by means
of the hypophyses, or external marks, that the segments can be detected
(fig. 2). In the neural canal (nc) and between the interneurals {in)
the segments are tolerably distinct. With the exception of a small lump
in the mouth of the nerve aperture (fig. 5), there is no evidence of cal-
cification in the vertebrae of more than half of the total. Both neurapo-
physes and interneurals are perforated for the nerves. The interneurals
are rather thick and strong. Small wedge-shaped interspinous pro-
cesses (is) extend about half-way through the crest. The ridge formed
by the interneurals and the interspinous processes is surmounted by a
very strong elastic cord of fibrous tissue. Over the abdominal cavity
the hsemal processes bear short flexible unsegmented ribs, and be-
tween the hsemapophyses are the small triangular interhsemals (ih).
Above the anterior portion of the anal fin the haemapophyses begin to
take on a downward extension, gradually becoming blade-like, and over
the posterior portion of the same fin they are supplemented by small
pieces of cartilage (PI. XIII. c-c), which farther back become the ra-
dials of the caudal. The radials of the upper part of the caudal begin
independently, or separate from the vertebrae; farther back they join
the latter and continue to the end as if part of the column. The lines
of separation between the vertebrae and accessory cartilages are very
indistinct ; in places, it is only with diflBculty they can be traced. The
column ends abruptly ; the terminal segment resembles a slice taken
from the front of a following vertebra. The canals are visible at the
end, but seem to be stopped by soft cartilage.

The anterior vertebrae of a large Heptabranchias at hand differ from
the anterior of those described above in having the notochord segmented
by thick membranous septa, each of which has a very small central
perforation. Towards the tail there are vertebz'al constrictions, and the
column tapers to a point.

Th^ Pectoral Cartilages,

Plate XI.

Coraco-scapulars, fig. 2. — These are strong, moderately slender, and
taper to a point at each end. The coracoid is irregularly triangular,
flattened below, broadened backward, and in the anterior third of its
length — which is turned upward, like the runner of a sled, toward the
binder basibranchial — tapers to a point. The bent portions are an inch
in length and nearly straight on their lower faces. They are applied to

14 BULLETIN OF THE

each other and united by ligament, except at the extreme end, where
there is a small space in which the cartilages appear to have coalesced.
They approach each other in an angle of about forty-five degrees. The
scapula is less angular and more slender ; at its upper end there is an
elongate flexible segment.

The metapterygium (mtp) of the pectoral is about half as long as the
fin itself; it is slender, broader backward, tapered in front and articu-
lates by a small surface with the coracoid. It is in two segments, and
bears two radials at the end and ten on the side.

The mempterygium (msp) is a large triangular plate extending upward
in an angle between the propterygium and the radials, but not sepa-
rating the metapterygium from the coracoid, as in the sketch; it articu-
lates with the process in the articular surface of the coracoid and with
the propterygium.

The propterygium (prp) is small, subtriangular or oblong, and —
articulating in the cavity of the articular surface of the coracoid — fur-
nishes a convex facet for the mcsopterygium. It is not fused with the
coracoid, as might be understood from the figure. The articulation is
similar to that of Heptabrauchias, as figured by Alivart, Fins of Elasmo-
branchs, PI. LXXV. fig. 3.

The radials are in three series, of which those of the outer are small
and short, and those of the inner elongate. Of the latter, three or four
of the anterior have coalesced in an irregular plate.

The Cartilages of the Pelvis and Ventrals.

Plate XI. Figr. 1, Plate XII.

The pelvis is a broad comparatively thin plate of cartilage about twice
as long as wide. On the upper surface it is concave, and has a ridge
along each sid^. Below it is convex and has a median ridge which bifur-
cates forward. Toward the vent the border is concave ; in front the
margin is convex. The anterior twelve radials articulate directly with
the side of the pelvis. Several of the foremost of these rays are only
partly distinct from each other.

The peculiar shape of this pelvis suggests an embryonic character of
other sharks. In embryos the pelvis is longer than in the adult, in
comparison with the transverse measurement. An embryo of Hepta-
brauchias before me has it half as long as wide, proportions which
are intermediate between those of the adult and an adult Chlamy-
doselachus.

MUSEUM OF COMPARATIVE ZOOLOGY. 15

The hasipterygium of the ventral is elongate and triangular, the iliac
ridge being continued along its upper side. At its extremity there is a
series of three radials. The majority of the radials are in three series,
but some of the posterior form a single one. Thirteen radials articulate
with the side of the hasipterygium.

The Cartilages of the Anal Fin.

Plate XIII. h-b.

These cartilages occupy an extent of three fourths of an inch at the
widest by about seven inches in length. The fin itself is less than half
an inch longer. The band of cartilages is widest forward and tapers
toward the tail ; it is widely separated from the hsemapophyses. The
radials are thin, and have not a great deal of rigidity. Segmentation
has been very irregular in its operation : in the lower half of the band,
the radials are directed backward in the usual manner ; in the upper
half, nearer the vertebrae, a few segments agree with the former, but
the majority are directed obliquely forward. The directions of the lines
of separation in either case are those separating the interneurals and
the neurapophyses, and are probably determined by similar causes, —
movements of the body in particular directions, or the directions of
the forces exerted by the muscles.

The Cartilages of the Dorsal Fin.

Plate XIII. Fig. a -a.

The series of cartilages in the base of the dorsal is three fourths of an
inch in width by four and three fourths inches in length. It is widest
backward and tapers gradually in front. It is not near the vertebrae ;
its only connection with the latter is by membrane. That portion of
the base extending in front of the fin has its radials directed obliquely
forward ; the part beneath the fin, though irregular, has them directed
backward.

The great extent of the band compared with the size of the fin, and
the manner in which it dwindles toward the front, taken in connection
with the fact of the continuation of the peculiar scales of the fin border
some two inches in front of the cartilages, show that in ancestral forms
of this animal the dorsal fin was much larger, and corresponded more
nearly in proportions with the anal.

16 BULLETIN OF THE

The Cartilages of the Tail.

Plate XIV. Fig:. 1.

Below the vertebrae the series of radials of the caudal fin begins con-
siderably in advance of the fin itself, over the hinder portion of the anal.
(PI. XIII. c-c.) Posteriorly they have the appearance of being formed
by downward growth of the hsemapophyses, separation or segmentation
from which only obtains in twenty-one of the anterior. The lines in the
sketch, which might be taken as separations in the others, only serve to
show the course of the vessel enclosed by the transparent cartilage.

The upper series of the radials of the caudal begins under the extrem-
ity of the dorsal fin. For a short distance in front, not shown in the
sketch, the series is separated by a space from the neural intercalaria,
as if the radials had originated like those of the dorsal and anal, inde-
pendently, and afterward through downward growth had in the greater
portion of the extent come in contact with the neural processes. These
radials and the interneurals are not fused like the radials and haeraa-
pophyses. They retain a considerable size at the end of the vertebral
column.

TJie Brain.

Plates XV. and XVI.

The brain is very small. Comparatively the amount of fore-brain is
much smaller than in the higher sharks, Carcharias, Zygsena, and others.
In outlines and proportions there is great similarity between this brain
and that of the Notidanidae. In both of the genera of that family
the brain is equally elongate, and the disposition of the nerves is not
greatly different ; the differences are mainly in details rather than in
general build. Owing to the softness of the mass, when removed from
the skull, it collapsed and spread out so that the figures sketched are a
trifle more broad and flattened than is natural. The condition prevented
such a removal of the envelopes as was desirable. The olfactory lobe
is shorter than that of Hexanchus (compare Maclay, Das Gehirn der
Selachier, Plate II.). The olfactory bulb is similar in shape in these
genera ; it is a club-shaped expansion with lobules at the end from which
the nerve distribution takes place. Being broader in front, the hemi-
spheres taper more toward the hypophysis than is the case in Hexanchus.
As in the latter, the optic lobes are rounded above and in front, and are
— when viewed from above — about half exposed.

MUSEUxM OF COMPARATIVE ZOOLOGY. 17

The cerebellum is of medium size, rather smooth ou its upper surface,
rounded in front, and presents an acute angle — with blunted apex —
between the corpora restiformia. On the upper surface the longitudinal
depressions are partly due to the uneven floor of the ventricle, on which
the upper walls rest. There are three moderate transverse depressions.
In the cerebellum the amount of plication is greater than that in Hex-
anchus as figured by Maclay. There is some likelihood that his figure
is taken from a young specimen, and that a large one will be marked by
greater complication. In Maclay's figure of Hexanchus the folds are
represented by a simple upward line with a transverse bar on the top,
like a letter T. To represent the same section in the new shark, we
shall have to place another X on each end of the transverse bar. jVIaclay
figures a longitudinal section of the cerebellum of a young Mustelus,
which shows a pretty close agreement. An adult Mustelus, which is a
great deal more complex, is also figured.

The corpora restiformia are comparatively large ; they approach each
other behind the cerebellum till there is but a small space between
them.

The medulla is large, somewhat larger than the same portion in the
Notidanida). The waved appearance in the sinus rhomboidalis, fourth
ventricle, is caused by the transverse bands of fibres in its membranous
roof.

The nerves of the third pair (^oculo-motorius) emerge from the lower
surface of the brain close behind the hypophysis, and a little farther
from the median line than the outer margins of the latter.

Back of the optic lobes, beneath the cerebellum, are the roots of the
fourth pair (trochlearis) ; these nerves are very small.

Not far from the middle line on the ventral surface, near the
anterior extremity of the medulla, are the roots of the sixth pair
(abducens) .

In a group at the side of each of the corpora restiformia are four
roots, three of which go to make up the nerve of the fifth pair (trigemi-
nus) and the fourth root apparently forms the seventh and eighth
(facialis and auditorius). One of these roots emerges at the side, two
on the inferior surface, and one at the side of the fourth ventricle,
■whence it passes outward by the side of the restiform bodies.

The nerves of the ninth pair (glosso-jjharyngeus') emerge quite near
the roots of the vagus, between the first of the series and the sixth pair,
on the lower side.

The tenth pair (vagus) is somewhat asymmetrical, having eight roots

VOL. XII. — NO. I. 2

18 BULLETIN OF THE

on one side and twelve on the other. There are also four pairs of ven-
tral roots rising nearer the median line.

The close similarity existing between the brains of Chlamydoselachus
and the Notidanidse is a strong point in favor of genetic relationship.

The Heart.

Plates XVII. and XVm.

Departing considerably from the conventional form of heart, this genus
presents a shape that is somewhat peculiar. Seen from below, it has a
small subquadrangular ventricle, a large auricle, and a long bulbus arte-
riosus. The ventricle measures nearly three quarters of an inch in
either width or length. When filled, the auricle is subtriangular, and
measures on each side an inch and a half The bulbus is almost twice
as long as the ventricle. Behind the auricle, and above and beliind the
ventricle, lies the sinus, which has a capacity that nearly equals the
bulk of the ventricle. From it the opening into the auricle is guarded
by a pair of valves that are without chordtc. The auriculo-ventricular
opening is furnished with a pair of valves provided with chordae tendinese.
In the ventricle the cavity or chamber is small ; its outlines in longitu-
dinal section resemble those of a pipe with a short stem, the stem being
directed toward the left upper side and the bowl toward the bulbus.
Along the inside of the passage (PI. XVIII. fig. B), the muscles lie in
bands [cdumnce) loosely laid one upon another, those in the posterior
section, or stem of the pipe, running transversely, and those of the an-
terior section being longitudinal.

The bulbus contains six rows of valves, or seven if we count the single
valve nearest the ventricle as a row. Two or three of the posterior
series have chordae teudineae.

Generally among sharks the bulbus is short, and the rows of valves
are less numerous, ranging from two to five. Professor Owen says
(Anat. Vert., I. 474) that Hexanchus and Heptabranchias have each four
rows of valves. From facts that have come to my notice during this
study I reach the conclusion that the number certainly varies among
the species of a genus, and that it probably varies among individuals of
a species, or even in different periods of the life of an individual speci-
men. A young Heptabranchias pectorosus at hand has five rows. A
large IT. maculahis has only three ; there are in this case, however,
traces in the middle of the bulbus as of two rows that have become
obsolete. Of other sharks that have been examined Pristiurus melanos-

MUSEUM OF COMPARATIVE ZOOLOGY. 19

tomus has a very short bulbus and two rows, and Heterodontus francisci
has three rows in a bulbus httle if any more than half as long as the
ventricle. Selache has three rows in a bulbus equally short (see Pavesi,
Del Genere Selache, PI. III., aS*. rostrata). A specimen of Somniosus
microcephalus has a short bulbus and four rows of valves.

Behind the ventricle, in the partition, between the peritoneum and the
pericardium, there is a spongy mass of dark tissue an eighth of an inch
in thickness.

A Cardiac Parasite, Tetrarhynchus wardii, sp. n.

The worm figured .on Plate XVIII. figs. 8-10, was found within the
cardiac chamber, attached to the lower side of the auricle, between it
and the ventricle and bulbus. Dr. E. L. Mark, to whom it was sub-
mitted, pronounces it a Tetrarhynchus, and thinks it possesses characters
which will not admit of placing it in any of the described species of the
genus. The head is large, subelliptical in transverse section, and sub-
quadrangular or oblong in longitudinal outline. Against the scalpel or
needle it is as hard as bone. The groove on each side extends back half
the length of the head. When the hardened mucus is removed, the sur-
face is seen to be covered with small papillae. Some of the teeth are
much curved, forming arcs of ninety degrees ; others are a little nearer
straight. All seem to be compressed, and the base extends under the
cusp or claw nearly the length of the latter.

The slender portion, or tail, in a measure resembles the flattened tails
of certain angle-worms. It does not show marks of division into seg-
ments. Entrance into the cardiac chamber must have been effected when
the worm was small, for at present the walls seem entirely closed against
intruders.

Being indebted to Professor Ward for our knowledge of it, we have
taken the liberty of introducing the species in his name.

Ovaries and Oviducts.

Plate XIX. Fig. 1.

Fortunately for us, when the captor tore the viscera from the speci-
men he left several important pieces. A section of some twelve inches
in length of the ovaries and oviducts is represented in the sketch. The
ovaries had been badly preserved and were much torn. Three inches
from the anterior opening of one of the oviducts it bore a nidamental
gland ; the gland of the other tube was an inch farther back. A piece

20 BULLETIN OF THE

left at the cloaca showed one of the ducts greatly distended, possibly
with young that had hatched within it. Only one of the tubes had
been in use. In Fig. 2, Plate XIX., the oviduct that had not been
expanded is shown at one side (ov), the other having been slit open with
the cloaca to show internal arrangement.

The Nidamcntal Gland.

Plate XX. Fig. C.

The gland consists, in appearance, of two thick plates of laminated
structure. The plates are longer and thicker in the middle, and
shorter and thinner at each side. The short sides have been applied
and united ; this leaves an acute point descending from the thicker por-
tion on the inside of the tube. The insides of the walls are crossed by
minute stria), between the laminaj, which appear transverse, but in
reality are spiral and ultimately — following the outlines of the anterior
or posterior borders — terminate, forward or backward, in the longitu-
dinal folds of the tube itself. The inner edges of the lamina) are set
with minute pores. Near the middle of its length there is a deeper
transverse groove. This is crossed by the laminae without change in
their directions on its account. The plates are not distinct from each
other through the whole of their length ; branches frequently cross
obliquely from one to the other. The bottoms of the grooves between
them have closely-set transverse partitions. The walls of the gland
are thicker anteriorly ; tliey begin abruptly, or even extend a little in
front of their points of attachment to the tube. The appearance is
such as would result from twisting the inside walls of the duct very
closely for a short distance. In this we have a hint as to the origin of
the gland. The distended condition of the oviduct is the only reason
for supposing the eggs to be hatched before extrusion, after the shells
have been supplied.

The Intestines.

Plate XrX. Figs. 3, 3.

What remnants of the intestines were left show that the shark pos-
sesses a spiral valve in the intestine and a cajcal pouch behind the valve.
The intestine {int) opens into the cloaca {cl) behind the openings of the
oviducts. The ureters unite before reaching the cloaca, into which they
empty by a single aperture (ua), Plates XII. and XIX. In this speci-

MUSEUM OF COMPARATIVE ZOOLOGY. 21

men there is no appearance of a urethral papilla ; the anterior border of
the opening is inflated into a flap or valve, which closes the opening
against objects passing outward through the cloaca, or, better, which is
made to close it by the objects themselves. The mouth of each of the
abdominal pores is inflated in a similar manner into a broader flap, by
which the pores ai'e hidden (ahp), Plates XII. and XIX.

TJie Abdominal Folds.

Plate XX. Figs. A, B.

As represented on the plate, section B is of natural size. The folds
become less prominent near the pectorals and toward the pelvis. The
section was made eight inches in front of the latter. In each figure
the inner wall of the belly has been stretched to its utmost, so that
the prominence of the folds is not overdrawn. As stated above, the
folds hang three quarters of an inch lower than the skin of the body
at their outer sides, and are separated below by a groove. One of the
folds is seen to hang below each of the large abdominal vessels. The
vessels are parallel or nearly so. Between them there are two mus-
cular bands, one to each fold. Each band is nearly an inch in width,
very thin at its lower edge, and near one fifth of an inch thick toward
the rounded upper edge, between the veins. The fibre in these tropeic
(rpoVcajs, the keel of a ship) or keel muscles differs from that in the
walls of the flank in being coarser in the bundles and plates, and more
loosely put together. Apparently the keel muscle corresponds to the
rectus abdominis of other vertebrates.

History.

Chlamydoselachus anguineus Gamaan, 1884, Jan. 17, Bulletin of the Essex Institute,
Vol. XVI., —Science, Feb. 1, p. 116, — Science, Mar. 21, p. 345, — Science, Nov.
28, p. 484 ; Gill, 1884, Science, Mar. 21, p. 346, —Science, Apr. 11, p. 429, — Sci-
ence, Dec. 12, p. 524.

Didijnwdus anguineus Cope, 1884, Mar. 7, Science, p. 275 (change of name only), —
American Naturalist, April No., p. 412, —Science, May 30, p. 645, —Pal. Bull.,
No. 38, printed July 1, pp. 572, 588, 589, and Proc. Amer. Phil. Soc, same article.

The specimen from which the description and figures given below have
been taken is the only one of this sharlrof which we have any knowl-
edge at present. It was purchased in a miscellaneous lot of alcoholic
specimens by the Museum from Prof. H. A. Ward, who had secured it

22 BULLETIN OF THE

in Japan. A portion of one side was damaged /rem lying against the
side of the cask in which the lot was kept, the brain was softened, and
— with the exception of short pieces of the oviducts, stomach, and intes-
tine — the viscera had been torn away by the fisherman ; otherwise it
was in a tolerable state of preservation. The dissections have been
made from one side, and in such a manner as to be sewed up again to
leave the specimen apparently intact. A preliminary description, with
outlines of the body and teeth, was given in the Bulletin of the Essex
Institute, Vol. XVI., 1884, and separate copies of the article were pub-
lished in January of the same year. In this paper species and genus
were described as new, and recognized as belonging to a new family,
ChlamydoselachidcB. This was followed, on the 1st of February, by an
article entitled " A Peculiar Selachian," also with outlines, in the weekly
journal "Science," in which ordinal characters were noticed, and, to dis-
tinguish from other Galei, the name Selachophichthyoidi was applied.
In each of these notices resemblance of the teeth to those of Cladodus
was pointed out. Science of March 7th contained the following note
from Professor Cope : —

" A Carboniferous Genus of Sharks still living. — I observe that in a late num-
ber of Science, Mr. Garman describes a new genus of sharks from the Japanese
seas, under the name of Chlamydoselacbus. The figure of the teeth which he
gives shows the animal characterized by Mr. Garman to be a species of the
genua Didyniodus (Cope, Proceedings Philadelphia Academy, 1883, p. 108,
equal to Diplodus Agass. Poiss. fossiles, pre-occupied in recent fishes), which
has hitherto been supposed to be confined to the carboniferous and Permian
periods. The species possess two, three, or four denticles. Material in my
possession enables me to fix the position of this genus, which I will endeavor
to explain in the next (April) number of the American naturalist. Didymodus
becomes by this discovery the oldest living type of vertebrata."

In the number of Science for March 21, in a note headed " The oldest
living Type of Vertebrata, Chlamydoselacbus," I pointed out differences
between this genus and Didymodus which would not admit of placing
them together, and again noted the resemblances to Cladodus, remark-
ing at the same time that, " if the new selachian was to have been placed
in either of the fossil genera mentioned [Hybodus, Sphenonchus, Diplo-
dus (Didymodus Cope) and Cladodus], it should have been Cladodus."

Professor Gill published a letter in the same number of the journal, in
which, in the course of comments on Cope's note, he accepts my conclu-
sion that Chlamydoselacbus represents a very distinct family, and also
makes it a distinct suborder at least. He objects to the name I had

MUSEUM OF COMPAKATTVE ZOOLOGY. 23

suggested for the order, and gives Pternodonta as a substitute. The
opinion advanced by me in regard to the propriety of placing the genus
nearer than any other of the recent sharks to the fishes, he accepts with-
out hesitation. He dissents emphatically, however, in regard to the
relations to extinct types, basing his objections on Dr. Traquair's dis-
covery of Ctenacanthus costellatus, in which he says the Doctor has
proved beyond a doubt that the Cladodont dentition and the ctenacan-
thoid spines coexisted in the same fish. Cladodus, he says, was obvi-
ously not at all related to Chlamydoselachus, and adds that it did not
have the essential dentition of that genus. Agreeing to some extent
with Cope, he asserts that Chlamydoselachus did have a representative
in the carboniferous genus Diplodus Agass. (Didymodus Cope), but
doubts that the two can be congeneric. In this letter the sharks are
arranged to include the new type. The arrangement given places
Hybodus, Cladodus, Ctenacanthus, etc., the Hybodontidse, in the Lipo-
spondyli ; and Chlamydoselachus and Didymodus, which he calls Chla-
mydoselachidse, in the Selachophichthyoidi. It is also suggested in the
note, that the Hybodonti may not have been Squali at all, but may be
more nearly related to the Holocephali, the primitive form from which
both diverged being theoretically like Ctenacanthus.

The next publication on the subject was that of ]\[r. Cope in the
American Naturalist, April, 1884, p. 412 : —

" Tlie Skull of a still living Shark of the Coal Measures. — The genus Didymo-
dus is a well-known form of Elasmobranchi of the Coal Measures, and I have
reported it as occurring also in the Permian. Mr. S. Garman has recently
published an account of a shark supposed to have been taken off the coast of
Japan, which he names CJilaviydoselachus anguineus, referring it to a new genus
and family. He figures the teeth, and these are, as I have pointed out, identi-
cal with those of the genus above-named. The species should then he called
Didymodus anguineus."

After disposing of the genus Chlamydoselachus, this writer in the same
article proceeds to give a description of the skull and teeth of Didymo-
dus, which we take occasion to quote and discuss below, p. 28.

Science of April 11 contains a letter from Professor Gill on "The
Eolations of Didymodus or Diplodus," in which, commenting on Cope's
note, he says : —

"A resume oi Professor Cope's observations has just appeared, as promised,
in the American Naturalist for April (XVIIl. 412), and we are therefore in
a position to test his utterances. Notwithstanding the reverence and confidence
that I have expressed, I can but think now that for once Professor Cope has

24 BULLETIN OF THE

been too hasty, and tripped. I am convinced, not only that Didymodus has no
generic nor even family relations with Chlamydoselachus, but that it represents
even a different order."

This is followed by the history of Diplodus as worked out by Kner in
1867, with the addition of a more recent notice, the substitution of the
name Didymodus. The letter contains also expressions of doubt in
regard to resemblance between Thrinacodus and Chlamydoselachus.

From a letter in Science of May 30, by Mr. Cope, in reply to the
foregoing, the following quotation is taken. The title of the letter is
"Pleuracanthus and Didymodus." After stating Gill's position the
author remarks : —

" L There is no generic difference to be detected, in my opinion, between
the teeth which are typical of Diplodus Agas. and Thrinacodus St. J. & W.
and the recent Chlamydoselachus. Differences there are, but apparently not of
generic value

"3. Diplodus being regarded as a synonym of Pleuracanthus, it follows that
Chlamydoselachus Garm. is distinct on account of the different structure of the
dorsal fin, which is single and elongate in Pleuracanthus, according to Geinitz
and Kner. The presence of the nuchal spine is also probably a character of
distinction, although we do not yet know whether such a spine is concealed in

Chlamydoselachus or not I suspect that the skulls I describe represent a

different genus from Pleuracanthus proper. This genus will not differ from
Chlamydoselachus Garm. so far as we know the latter ; but the button indicates
another species

" 5. Of course a study of the anatomy of Chlamydoselachus, which I hope
]\rr. Garman may soon give us, may reveal differences between that genus and
Didymodus ; but of these we know nothing as yet.

The next publication on the subject is that of Mr. Cope in his Palse-
ontological Bulletin, No. 38, printed July 1, "On the Structure of the
Skull in the Elasmobranch Genus Didymodus." This bulletin consists
of pages 503 to 590 of the Proceedings of the American Philosophical
Society of Philadelphia for 1884. In the article there are several para-
graphs relating, with more or less directness, to the frilled shark, the
substance of the most of which has been indicated above. The following
forms the opening paragraph of the paper : —

" The genus Diplodus was described by Agassiz from specimens of teeth
from the European Coal Measures. In America, Newberry and Worthen have
described four species from the Carboniferous of Illinois and Ohio ; and I have
reported two species from the Permian beds of Illinois and Texas. Recently
Mr. Samuel Garman has described a shark, said to have been taken in the
Japanese seas, under the name of Chlamydoselachus anguineus, whose teeth, as

MUSEUM OF COMPAEATIVE ZOOLOGY. 25

represented, do not differ generically from those of Diplodus. This is an inter-
esting discovery, indicating that this genus, and not Ceratodus, is the oldest
type of vertebrate now known in the living state."

Near the end of the paper the quotations given above, under 1, 3, and
5, are repeated.

In a letter to Science of November 28, on " The oldest living Type of
Vertebrates," I called attention to the fact that, when my paper on the
recent discovery was read before the American Association for the Ad-
vancement of Science, Philadelphia meeting, Sept. 4, 1884, Professor
Cope abandoned his position concerning the affinities of Didymodus
and Chlamydoselachus, and agreed with me in the conclusions that the
two genera were very different, and that, from all the data we had for
comparison, the nearest known allies of Chlamydoselachus were Clado-
donts of the Subcarboniferous and Middle Devonian. The same letter
pointed out certain necessary changes in the classification, such as the
separation of the Cladodonti from the Hybodonti, and their inclusion,
with Chlamydoselachus, in a separate group characterized by "vertebrae
partially or imperfectly developed, a persistent notochord, and teeth
with broad backward-expanded bases." The communication also ad-
vanced the idea that the connection of the trabecular process of the
pterygo-quadrate with the skull, in such genera as Chlamydoselachus,
Rhina, and others, though it is of secondary origin, is none the less a
true articulation.

Up to date, the last word in the discussion is that of Professor Gill, in
reply to my note of November 28 ; it is in the same journal, issue of
December 12, and has the same heading. The several statements con-
tained in it, in regard to which we should disagree most seriously, are
the following : —

" The differences between us now are fictitious rather than real, or better,
perhaps, they are chiefly differences of expression." " The palato-pterygoid
not articulated with the skull is a true character of the typical sharks and
Rhinse." " I must dissent from the opinion that the Cladodontidse are related
to the Chlamj^doselachidae rather than the Hybodontidoe."

The characters of genus and fiimily assigned in the preliminary de-
scription were essentially as given below.

ChlamydoselacMdcB.

Body elongate, slender. Head broad, depressed. Eyes lateral, with-
out a nictitating membrane. Nasal cavity separate from that of the

26 BULLETIN OF THE

mouth. Mouth anterior. Teeth with broad, backward-extended bases
and slender cusps. Spiracles present. One dorsal, without a spine.
An anal fin. Caudal without a pit at its root. The first gill-cover free
across the isthmus. Intestine with a spiral valve. Anterior basibran-
chial cartilages present.

Chlamydoselachus.

Six gill openings. Opercular flap, first gill-cover, broad. Teeth sim-
ilar in both jaws ; each with three slender, curved, subconical cusps, sep-
arated by a pair of rudimentary denticles, on a broad base. No median
upper series of teeth in front ; a series on the symphysis below. Mouth
wide, without labial folds at the angles. Pupil horizontally elongate.
Fins broad ; caudal without a notch. Basihyal elongate. The name is
derived from X'^/Jti^'s, a mantle, or frill, and o-eA.a;(os, a shark-

The position of Chlamydoselachus in the system of recent sharks is
not difficult to determine. Six gill-openings and the structure of the
brain at once remove it from the others, and place the genus near the
outlying genera Ilexanchus and Heptabranchias. As it differs more
than they do from other sharks, it lies farther from the main body of
the Galei. The shape of the body, position of the mouth, articulations
of the jaws, dentition, squamation, lateral line, pelvis, tail, and tropeic
folds furnish characters sufficient to establish the distinctness of both
genus and family. By such features as the chondrification, the unseg-
mented notochord, the elongate bulbus, numerous cardiac valves, open
lateral line, and the squamation, its rank is detennined to be somewhat
lower than that of the Notidanidse. Possessing, as in the shagreen and
certain cephalic peculiarities, more in common with Rhina (Squatina),
though not at all closely related, it naturally falls into place in our lists
between the latter and the mentioned family.

What gives the new type a far greater importance than its standing
among recent forms, however, is found in its affinities to some of the
earliest known sharks, those of the Middle Devonian. Close affinity to
the genus Cladodus makes it in present knowledge " the oldest living
type of vertebrate."

In connection with its relationship to the early Selachia a number of
perplexing questions present themselves for answer. Further accumula-
tions of fossil material will be needed to determine how far success has
attended the attempts here made to solve some of the problems.

MUSEUM OF compaeatht: zoology. 27

It will be remembered that, from my first mention of the genus, I have
insisted on its resemblance to the Cladodonti. A further study of both
extinct and recent forms enables me to speak still more positively in
asserting that Chlamydoselachus is a Cladodont. As shown in the
descriptions above, some of its teeth are so characterized as to make it
imperative, if these teeth alone were considered, that the species should
be placed in the genus Cladodus of Agassiz, and nearer than almost any
of the fossil forms to his type C. mirabilis. It is only the fact that
others of the teeth differ in base or cusps, or both, from those of any
of the discovered species of that genus, that prevents the new shark
from being placed in Cladodus.

"What were the shapes of the Cladodonts ? is a question that has been
asked a great many times by palaeontologists, but so far it has not re-
ceived a satisfactory answer. Opinions generally have inclined toward
the conclusion that the teeth of Cladodus belonged with the spines of
Ctenacanthus. If we could say positively that the teeth of the former
really belong with the spines of the latter, it would be a long step toward
restoring the shape of the animal that bore them. The conclusion has
been advocated by Thomson, Romanowsky, Hancock and Atthey, Barkas,
and more recently by Dr. Traquair. Romanowsky has gone so far as to
describe a species of Cladodus (C. tenuistriatus) from a spine alone. He
does not state that he found spine and teeth associated directly, but
that the discovery of the spine was made in a locality in which teeth of
Cladodus mirabilis were numerous. According to Dr. Traquair, Barkas
proposed to unite Cladodus, Hybodus, and Ctenacanthus. In the publi-
cation cited, no reasons are given for the changes. Up to the time of Dr.
Traquair's publication, (Geol. Mag., Jan., 1884,) the relations of these
genera have been merely matter of personal opinion, conjecture, sug-
gested, as the Doctor puts it, by the obvious general resemblance of
teeth of Cladodus and spines of Ctenacanthus to teeth and spines known
to belong together in Hybodus. Others have held it probable that
Psammodus or Orodus would prove the dentition of Ctenacanthus. Of
all the contributions toward answering the question as to the dentition
of Ctenacanthus, that of Dr. Traquair is the most important. In it he
describes a new species, C. costellatus, from a nearly entire fossil bearing
the spines of that genus and teeth which certainly much resemble those
of some Cladodonts. Only one of the teeth is sufficiently visible to give
an idea of its shape, and this is a most unsatisfactory one. It has a
smooth, pointed, conical cusp on a broadish base which " looks as if it
i7iight support lateral denticles." The lateral and posterior portions of

28 BULLETIN OF THE

the base are not seen, and there seem to be no rudimentary denticles.
What is exposed favors to some degree the conclusion of its discoverer ;
but we do not know that the tooth has the lateral cusps, the backward-
expanded base, or the rudimentary denticles of the typical species of
Cladodus.

The probability is that Cladodus, known only from dentition, — which
would include teeth of Rhina or the upper front teeth of Heptabranchias,

— contains species that, if living to-day, would be distributed among a
number of genera, or even among different families. This being the case,
it might not be out of the way for Dr. Traquair to claim that Ctenacan-
thus — with a short body, two dorsals each with a spine, a mouth simi-
lar to that of Heterodontus, and teeth with one cusp and no buttons —
is a Cladodont; while Chlamydoselachus — with elongate body, a spine-
less dorsal, an ophidian mouth, and teeth with lateral cusps and buttons

— is claimed to represent Cladodus itself,

Chlamydoselachus, however, has not been an undisputed Cladodont.
In several publications Professor Cope has asserted its identity with
Diplodus of Agassiz (renamed Didymodus by Cope, but later admitted to
be identical with Xcnacanthus). As he has since abandoned his posi-
tion, it would be unnecessaiy to consider the subject further, if it were
not that he has made no publication of his change of opinion, and that
matter in one or two of the communications may be used to throw
light on the affinities of Chlamydoselachus and allied sharks. From
the American Naturalist of April, 1884, page 412, we quote the fol-
lowing description of skulls said by its writer to belong to the genus
Diplodus Ag., renamed Didymodus, in which Mr. Cope claimed the
frilled shark must be placed (see page 22).

*' The palatopterygoid arch is suspended to the postorbital process of the cra-
nium, as ill the existing Flexanchidae. The genus would then be referred to the
suborder Opistharthri of Gill, but for the following peculiarities : The skull is
segmented, so that cartilage-frontals, parietals, and occipitals can be distin-
guished, together with an element which has the position of the intercalare.
The occipital supports a large vertebral cotylus. There are membrane bones
extending from the nose over tlie orbits, which are either supraorbitals or
frontals. The tissue of the bones is granular, which leads to the belief that the
granular ossification which covers the chondrocranium in recent sharks, pene-
trated the entire chondrocranium in this genus. Hence the basicranial axis
consists of the sphenoid and presphenoid bones. One at least of the nares is
on the superior face of the muzzle. The frontal cartilage-bones are elevated
and fissured at the posterior extremity, each apex projecting freely upwards and
backwards, presenting a certain resemblance to the structure seen in the Lepi-

MUSEUM OF COMPARATIVE ZOOLOGY. 29

dosirenidae. The structure points to the type from which the true fishes
(Hyopomata) diverged from the sharks. The characters are thought to define
an order of the subclass Elasmobrauclii, equivalent to all other known forms.
To these two divisions were given the names Ichthyotomi and Selachii."

In the Proceedings of the American Philosophical Society of Phila-
delphia, 1844, the skull is figured, and from a study of the illustrations
conclusions are reached which differ somewhat from those embodied in
the above description. A comparison with the plate in the Proceedings,
or Pal. Bull., No. 38, will show whether they can be justified.

1. As in aged specimens of some recent sharks, the cartilage of the
skull is highly charged with calcareous matter.

2. " The penetration of the granular ossification " does not distinguish
these skulls from those of certain Galei.

3. The skull is unsegmented ; the lines of segmentation, so called, are
partly accidentals, which are not alike on the two sides of the skull.

4. What is called one of the nares on the superior face of the muzzle
is not a nostril ; it is behind the nasal sac, and is the opening of the
preorbital, or ethmoidal, canal.

5. The fronto-parietal fontanelle, so called, is the parietal fossa, from
which pass the aqueducts of the vestibule, common to all Selachians.

6. Figure 4 of the plate should be reversed in direction, the prolonged
anterior portion, in the figure, should be turned backward from the in-
terorbital region ; thus bringing what in the figure serve as orbits behind
the postorbital processes.

7. The Ichthyotomi, as based on these skulls, have not been separated
from the Selachii.

8. From these skulls alone the genus to which they belong cannot be
separated from the Opistharthri of Gill.

9. The skulls belong to a genus that represents a line of descent
closely allied to that in which we find the Notidanidse, and is not "the
type from which the true fishes (Hyopomata) diverged from the sharks."

10. The genus to which these skulls belong is probably neither
Diplodus Ag. {Didymodus by Cope), Xenacanthus Beyr., nor Pleuracan-
thus Ag., but a new one.

It was a mistake to consider the genus identical with Chlamydosela-
chus ; it was another to make it identical with Xenacanthus Beyr. We
know the genus only from the skull and teeth. The dentition affords
a means of comparison, which places the genus in the Cladodonti with
Thrinacodus and Cladodus. There is nothing in the skull, as far as we
know it from the description, that will place it elsewhere.

30

BULLETIN Oi* THE

The name Didymodus having been proposed as a substitute for
Diplodus Ag., it passes out of use as a synonym for Xenacanthus Beyr.
Such being the case, it is not available as a name for a new genus.
Professor Cope suspects that the skull he has described represents a
different genus from Pleuracaathus proper (Xenacanthus Beyr.), which
" will not differ from Chlamydoselachus Garm." As it does differ very
decidedly from the latter genus, confusiou will be obviated by allow-
ing the name Didymodus to remain a synonym, and employing a new
name, say Diacranodus, for the new genus. The species will then stand
D. compressus and D. platypternus. The genus may be distinguished
by the attachment of the pterygo-quadrate to the post-orbital process of
the cranium, and by the teeth : cusps two, diverging, subconical, slender,
and separated by a median rudimentary denticle or button on the base;
bases extended backward, thinner and rounded posteriorly.

By reference to page 25, it will be seen that, while on some points we
accord well, in others there is considerable variance between Professor
Gill's views and my own. Against his assertion that Cladodus is not at
all related to Chlamydoselachus, and that it did not have the essential
dentition of the latter, I am compelled to maintain that these genera
have essentially the same deutition, that they are very closely related,
and that, in fact, the recent genus is the living representative of the
fossil Cladodus. As the letters and replies noted in the history, pages
22-25, discuss the question, it is not necessary to give it further atten-
tion here. In regard to the recent affinities our diffei-ences are slight.
The arrangement given below expresses the conclusions reached as nearly,
perhaps, as any at present made. A comparison with Professor Gill's
letter in Science, March 21, 1884, will show the extent of our agree-
ments and differences.

Hyhodonti: Hybodus and allies.

Cladodonti: Cladodus and allies ;
Chlamydoselachus.

Notidani: Hexanchus and Hepta-
branchias.

Heterodonti: Heterodontus (Ces-
tracion Cuv.).

Lavmoe: Most living sharks.
\.Ehinoe (Squatinse).

fSELACHII^

Elasmobranchii ■

Galei

Batoidei.

IHOLOCEPHALI.

Without resorting to the "genealogical tree," a scheme that shall not
separate groups that belong to the same line of descent is difficult to
an-ange. In the above the Hybodonti were probably the progenitors of
the Heterodonti and a portion of the Lamnse. The Ehinse may have

MUSEUM OF COMPARATIVE ZOOLOGY. 31

come from an intei'mediate between the Hybodonti and the Cladodonti,
a branch from the latter. From the Cladodonti we have the Notidani
and a portion of the Lamnse. Probably the line leading to the Notida-
nidse had separated from that of Cladodus before the Carboniferous;
but that it belongs to that line appears probable, in view of the affini-
ties to Chlamydoselachus, and the presence in the upper jaws of Cla-
dodont teeth and intermediates between them and the most modified
ones in the lower jaws.

A great diff"erence between the teeth of Cladodonts and others is to
be seen in the bases. In Cladodus the teeth are braced posteriorly, by
a backward prolongation of the base, most often extended under the fol-
lowing tooth in the same row. As soon as teeth of this character emerge
from the thecal fold under which they are formed, they come into func-
tion, on the inner side of the jaws, and continue in use even after passing
to the outer side.

Another plan is to be seen in many of the blade- or chisel-like cutting
teeth, Scymnus and others, or in the broad teeth of the JSTotidanidse.
Here the bracing is done by a forward prolongation of the base, the
teeth coming into function only after the extension has passed to the
outer side of the jaws. In Pristicladodus of McCoy {P. dentatus), and
its possible descendants, Carcharodon and Carcharias, or in Lamna, there
is also a slight backward extension, which has the effect of bringing the
tooth into function sooner by raising the apices of the cusps.

In Hybodus pi'oper, the teeth are intermediates in character of base,
being braced neither forward nor backward, or but slightly in either or
both directions. Such teeth come into use somewhat early, and become
useless soon after turning to the outer side of the jaw. Between the
Cladodonts and the Hybodonts there are many intermediates, some of
which are evidently out of place as now situated in Hybodus.

Before the discovery of Chlamydoselachus it was supposed the line of
the Cladodonti had become extinct in the carboniferous. Now, with
Cladodus at the farther end, their line is extended from the Sabcarbon-
iferous to the present ; or, taking Pternodus (Pristicladodus St. John
and Worthen, not of McCoy, type P. springeri St. J. & W.) — a genus
allied to both of the preceding genera — as our point of departure, it
extends from the Middle Devonian.

How much we are justified in treating Chlamydoselachus as a per-
sistent type, palseontological research will ultimately determine. The
reasons for so doing are found in the identity of dental characteristics
of Cladodus and its recent representative, and in the possession of char-

32 BULLETIN OF THE

acters by the ndnlt of the recent form that belong to the embryonic of
the lowest of the other recent sharks.

Whether Chlamydoselachus gives a fair idea of the shapes of the De-
vonian Galei is a question we may not be able to answer satisfactorily
at present. The genus bears evidence of having been considerably modi-
fied in more recent times. But, being of lower rank through possession
of characters comparatively more or less embr}'onic, it affords us a safer
starting-point for an estimate of ancestry than do the others, which have
in attaining higher rank experienced considerably more modification.

Starting from the specimen, then, its less remote ancestors differed
from it somewhat as follows : their teeth were less slender and more stri-
ate, more like the scales at the angle of the mouth ; the teeth not being
so much hooked, their jaws and the suspensorium were shorter; their
branchial laminae were more free at the outer ends, — may have pro-
truded ; their scales in general were more like those of the flank or belly ;
and in them the dorsal resembled the anal in size and shape, or at an
earlier period both may have been confluent with the caudal.

If we were to hazard a conjecture as to Cladodus, we should make the
body elongate; the mouth anterior; the jaws and suspensorium but
moderately long; the scales flattened and irregular in shape, but, judg-
ing from the teeth, to some extent possessing keels or spines ; the oper-
cular flap broad and free across the isthmus, as in certain larval Batrachia
before coalescence with the pectoral region ; the branchial apertures six
or more in number; the eye without a nictitating membrane ; the noto-
chord persistent and unconstricted ; the vertebrte imperfect or the col-
umn unsegmented ; the bulbus elongate and many-valved ; the pelvis a
broad elongate plate ; the lateral line an open groove; the dorsal large;
and, possibly, the tail diphycercal, the abdomen with tropeic folds.

As we see them by the aid of Chlamydoselachus, it appears that the
Cladodonti of the Middle Devonian, though low in rank, were true
Sharks, and that the primitive form connecting them with the Fishes is
to be sought farther back, in the earlier Devonian or in the Silurian.

Cambridge, July 4, 1885.

MUSEUM OF COMPARATIVE ZOOLOGY. 33

DESCRIPTION OF THE PLATES.

PLATE I.

Entire figure, ^\ natural size ; side of liead, about | nat. ; ventral fins from
beneath.

PLATE II.

Front view of head, with open moutli.

Upper view of head.
Head as seen from beneath.
The fourth gill-opening.

PLATE III.
PLATE IV.

PLATE V.
PLATE VI.

Teeth, six times nat. Figs. 1 and 5, upper view ; Figs. 2 and 6, as seen from
behind ; Fig. 3, from the side ; Fig. 4, from beneath ; and Figs. 7 and 8, from
the front.

Scales. Fig. 9, from middle of the flank; Fig. 10, from the side of the tail on
the lateral line ; Fig. 11, upper edge of tail ; Fig. 12, at angle of month ; Fig. 13,
from middle of belly. Fig. 9 seven times, and Figs. 10-13 five times nat.

PLATE VII.

A. Skull, hyomandibular, and pterygo-quadrate from above, natural size.
a, rostrum ; b., fontanelle ; r, nostril ; d, nasal sac ; <?, preorbital canal ; f, preor-
bital process; g, postorbital process ; h, supraorbital pores; k, epiotic or postpari-
etal pore ; /, backward extension of articular facet for liyomandibular or pterotic
process ; m, vertebrae ; n, parietal fossa and aqueducts of the vestibule ; o, rugose
space on frontal region; p, frontal slit; hm, hyomandibular; q-pff, quadrato-
pterygoid; q-p, quadratic process of pterygo-quadrate ; r, rostral sinus; t, epiotic
process.

B. Skull in longitudinal section (reversed in drawing). </, aqueducts of the
vestibule ; so, supraoccipital crest ; fm, foramen magnum ; nc, notochord ; bo,
basioccipital ; ic, internal carotid ; pj/, pituitary excavation ; to and r, bridges.
Numbers 2-10, second to tenth pairs of nerves.

VOL. XII. — NO. 1. 3

34 BULLETIN OF THE

PLATE Vin.

A. Side view of skull, quadrato-pterygoid, hyomandibular, and Meckel's carti-
lage, mk, natural size ; lettering as in Plate VII. ; br-r, branchial rays ; c-hy, cera-
tohyal; tp, trabecular process of pterygo-quadrate ; y, ligament attaching trabecular
process to the skull; q-p, quadratic process of pterygo-quadrate j xx, orbito-nasal
pores.

B. Side view of quadrato-pterygoid, hyomaudibular, etc., at rest in position
against the skull. All natural size.

PLATE IX.

Branchial cartilages and meckelian, lower view, size of nature, mk, Meckel's
cartilage ; b-hy^ basiiiyal ; c-hy, ceratohyal ; br-r, branchial rays ; b-br, basibran-
chials ; h-br, hypobranchials ; c-br, ceratobranchials ; e-br, epibranchials ; p-br,
pharyngobrauchials.

PLATE X.

Vertebrae : 1, side view of section from middle of body ; 2, longitudinal, and 4,
transverse section of same ; 3, longitudinal section from back of iiead (reversed),
all twice uat. size. 5, nerve aperture showing the calcareous lump at the side;
n, neurapophysis ; is, iuterspinous process ; nc, neural canal ; ch, notochord.

PLATE XL

1. Pelvis, upper view, ^^g- nat. />«, pubic; eV, iliac ridge ; (5/?, basipterygium.
(Reversed.)

2. Pectoral cartilages, cr, coraco-scapular ; prp, propterygium ; msp, meso-
pterygium ; mtp, metapterygium.

PLATE XII.

Pelvis and ventral cartilages from beneath : cl, cloaca ; ab-p, abdominal pores ;
ua, urethral aperture. Size \\ nat.

PLATE XIII.

Cartilages of dorsal and anal fins, \ nat. : aa, radials of dorsal ; bh, radials of
anal ; cc, anterior radials of caudal.

PLATE XIV.

1. Tail with cartilages exposed, i\ nat.

2. Tip of tail, \ uat.

MUSEUM OF COMPARATIVE ZOOLOGY. 35

PLATE XV.

Brain, upper view, -f nat., and transverse sections. 1, olfactory lobe;
3, oculo-motorius ; 4, trochlearis ; 5, trigeminus ; 7 and 8, facialis and acusticus ;
10, vagus.

PLATE XVL

Brain : A from beneath, B from the side, and C in longitudinal section. Num-
bers as in Plate XV. ; 2, optic nerve ; 6, abducens; 9, glosso-pbaryugeus.

PLATE XVIL

Lower view of heart, | nat. 1, auricle; 2, ventricle; 3, bulbus ; 4, sinus;
5, dark tissue between cardiac and abdominal chambers.

PLATE XVni

B. Heart in longitudinal section, showing cavity in ventricle, 6, and valves in
bulbus, 7- -§ nat.

C. Parasite, Tetrarhynchus icardii, f nat. Figs. 8 and 10 from the sides,
and Pig. 9 from the front.

PLATE XLS.

1. Ovaries and oviducts, \ nat. o, ovary ; ow, oviduct; ng, nidamental gland.

2. Longitudinal-seetiou through cloaca and oviduct, nat. size ; showing ov, ovi-
ducts ; int, intestine ; iia, urethral aperture ; cl, cloaca ; ab-p, abdominal pore ;
jD, caecal pouch. Reversed in transfer.

3. Section of intestine showing spiral valves.

PLATE XX.

A. Transverse section of abdomen showing the tropeic folds, | nat.

B. Same section, natural size.

C. Longitudinal section of nidamental gland, \ natural size.

Plate I.

f'ln-AMVnoSELAOnUS ANGUINEUS Gakman.

Plate II.

Plate III.

Plate IV

Pl.ATK V.

Plate VI.

Plate YIII.

Plate X.

Plate XI.

Plate XII.

Plate XIII.

Plate XIV.

Plate XV.

Plate XVI.

~ «

\l

4^

OQ

Plate XVII.

.,<v

^ 3

.y^

''/Uj^<^»-'

Plate XVIII.

Plate XIX.

Plate XX,

!imm

No. 2. — Reports on the Results of Dredging, under the Supervision
of Alexander Agassiz, in the Gulf of Mexico (1877-78), in the
Caribbean (1878-79), and along the Atlantic Coast of the United
States, during the Summer of 1880, bij tJie U. S. Coast Survey
Steamer "Blake," Lieutenant-Commander C. D. Sigsbee, U.S.K,
and Commander J. R. Bartlett, U. S. N., Commanding.

(Published by permission of Caklile P. Patterson and J. E. Hilgard, Super-
intendents of the U. S. Coast aud Geodetic Survey. )

XXVII.

Report on the Specimens (f Bottom Deposits. By John Murray.

BLAKE DEPOSITS.!

1. Specimens of deposits procured in the Gulf of Maine and along the
Coast of North America between the Gulf of Maine and Cape Hatteras
in 1880 {Stations 301-312, and 330-347) and in the Gulf of Maine in
1875.

These deposits consist of blue or gray colored muds and sands, the
latter being found only in depths less than 100 fathoms. They lie be-
tween the coast and the inner edge of the Gulf Stream. The greatest
depths are 1394 and 1186 fathoms, situated between 30 and 40 miles
outside the 100-fathom line. These deposits are chiefly made up of the
debris of the land of the North American continent, the mineral par-
ticles and clayey matter making up usually from 80 to 85 per cent of
the whole deposit.

1 Mr. John Murray, to whom the specimens of bottom deposits collected by the
"Blake" were sent for examination, has looked over the whole and selected some
typical specimens. These have been described in detail, and he has added some gen-
eral notes on the specimens characteristic, 1. of the Coast between the Gulf of Maine
and Cape Hatteras ; 2. of tlie coast between Cape Hatteras and Lat. 31° 48' N. ; 3.
of the coasts around the greater and lesser Antilles ; and, finally, of the Gulf of
Mexico and Straits of Florida.

Alexander Agassiz.

MusKUM OF Comparative Zoology,

Cambridge, July 10, 1885.

VOL. XII. — NO. 2.

38 BULLETIN OF THE

The mineral particles consist of fragments of ancient rocks, quartz,
monoclinic and triclinia felspars, magnetite, hornblende, augite, mica,
tourmaline, and occasionally glauconitic grains.

In 1240 fathoms, and Lat. 38° 34' N. oflf this coast, the "Challenger"
dredged many rounded and angular pebbles of milky and hyaline quartz,
fine-grained quartzites, felspathic quartzites, mica schists, serpentine
rocks, and compact limestones. These fragments were not larger than
6 or 7 centimetres in diameter. The "Blake," in 1241 fathoms and
Lat. 39° 43' N., dredged similar, but much larger, fragments of the same
rocks, some of which were glaciated. In Lat. 41° 14' N. and in a
depth of 1340 fathoms, the "Challenger" again dredged similar rock
fragments, and one block of syenite weighing 5 cwt. These deposits
being all within the influence of the Labrador Current, these rocks may
be regarded as chiefly ice-borne.

The carbonate of lime in these deposits consists of coccoliths and
coccospheres, of pelagic and other Foraminifera, and of fragments of
Echinoderms, Polyzoa, Ostracodes, and Mollusks. The pelagic Forami-
nifera shells and coccospheres are more abundant in the deeper deposits
far from the land than in those from shallower water near the coast.

The siliceous remains of Diatoms, Radiolarians, and Sponges, together
with arenaceous Foraminifera, and glauconitic casts of calcareous Foram-
inifera make up sometimes 4 or 5 per cent of the deposit.

The following are descriptions of some of the typical deposits : ^ —

Station 305. — Lat. 41° 13' 53" N. Long. GS** 57' 25" W. Deptli, 810 fath-
oms. Surf. temp. 56:^°. Bot. temp. 39°. Gray mud, browu when wet, earthy,
plastic, dries into hard lumps. Mixed with the mud were some few piunulse of
Crinoids, also a few rock fragments (sandstone, diorite, and diabase) measuring
from 10 to 30 milUmetres in diameter.

Carbonate of Calcium, 5.08 per cent, consists of coocoliths and coccospheres,
fragments of Echinoderms, and the following foraminifera : —

^ The methods employed in the examination of these deposits are the same as those
adopted by Messrs. Murray and Renard for the Challenger deposits. The carbonate
of calcium was determined by estimating the carbonic acid, weak and cold hydro-
chloric acid being used for the purpose. The part insoluble in the acid is designated
"residue," which by washing, decantation, and microscopic in.spection is separated
into three parts : (a) Minerals, the contraction m. di. indicating their mean diameter
in millimetres ; {h) Siliceo-us Organisms, including the glauconitic casts of forami-
nifera and other calcareous organisms ; (c) Fine Washings, including those particles
which, resting in suspension, pass with the first decantation. The numbers in
brackets indicate the percentage of the whole deposit.

MUSEUM OF COMPAEATIVE ZOOLOGY.

39

Qlohigerina bulloides "
O. inflata
G. dutertrei
Pulcinulina menardii
P. micheliniana
P. canariensis

Y

Uaplophragmium canariensis "

Textularia sp.
Pelagic Bulimina marginata

species. Uvigerina pi/gnuea

Truncatulina lobatula

Pulvinulii'M elegans

Bottom-
living
species.

Residue, 94-92 per cent, dark brown, consists of Minerals [75.00], m. di
0.5 mm., quartz, mica, felspar, hornblende. Siliceous organisms [5.00], Diatoms,
Radiolarians, and Sponge spicules. Fine washings [14.92], argillaceous matter,
fine mineral particles, fragments of Diatoms and siliceous spicules.

Station 308. — Lat 41« 34' 45" N. Long. 65" 35' 30" W. Depth, 1242
fathoms. Surf. temp. 65". Bot. temp. 38". A dark gray mud, plastic, pul-
verulent, granular, dries into hard lumps.

Carbonate of Calcium, 7-27 per cent, consists of Echinoderm fragments, many
coccoliths and coccospheres : the following Foraminifera were observed : — '

Orbulina universa, rare.
Globigerina bulloides, common.
Q. infata, common.
Q. conglobata, few.
G. dubia, few.
Puloinuuna menardii, few.
P. canariensis, few.
Pullenia obliquiloculafa, rare.

Uvigerina pygmcea^ few.
U. pygnuea, var. angulosa few.
Bulimina marginata, few.
Lagena jimbriata, rare.
Diseorbina sp., few.
Rotatia repanda, few.
Pleurostomella sp., rare.
Gristellaria cultrata, rare.

Residue, 92.73 per cent, dark brown, consists of Minerals [75.00], m. di.
0.3 mm., quartz, monoclinic and triclinic felspars, magnetite', mica, hornblende,
tourmaline, glauconite, and glassy fragments. Siliceous organisms [4.00], Sponge
spicules, Radiolarians, and Diatoms. Fine washings [13.73], argillaceous matter,
and many angular, fiue mineral particles.

Station 312. —Lat. 39® 50' 45" N. Long. 70" 11' W. Depth, 466 fathoms.
Surf. temp. 7H°- Bot. temp. 40". A gray mud.

Carbonate of Calcium, 3.46 per cent, consists of a few Echinoderm fragments,
coccoliths, and the following Foraminifera • —

Reophax fusiformis, few.
B,. scorpiurus, few.
Uaplophragmium fontinense ?
Ammodiscus incertus, few.
A. gordialis, rare.
Clavulina communis, few.
Cyclammina pusella, rare.
Bulimina marginata, rare.

Gristellaria cultrata, rare.
Uvigerina pygmaa, rare,
few. U. pygmeea, var. angulosa, rare.

Globigerina inflata^ common.
G. dutertrei, few.

Pulvinulina menardii, var. tumida, rare.
Cassidulina crassa, rare.
Polystomella sp., rare.

40 BULLETIN OF THE

Residue, 96.54 per cent, gray, consists of Minerals [80.00], m. di. 0.4 mm.,
fragments of milky and hyaline quartz 1 to 2 mm. in diameter, felspar, horn-
blende, mica, glaucouite, augite, fragments of ancient rocks, and fragments of
serpentine rocks much decomposed. Siliceous organisms [6.00], Sponge spicules,
a few Radiolarians and Diatoms. Fine washings [10.54], green argillaceous mat-
ter with glauconitic particles, fine minerals, and fragments of Sponge spicules and
Diatoms.

Station 340. — Lat. 39° 25' 30" N. Long. 70° 58' 40" W. Depth, 1394
fathoms. Surf. temp. 76|°. Bot. temp. 38°. A gray mud, coherent, plastic,
dries into hard lumps.

Carbonate of Calcium, 16.81 per cent, consists of coccoliths and coccospheres,
otoliths of fish, fragments of Dentalium and Echinoderms, and the following
Foraminifera : —

Glohigerina. bulloides, few. Rotalia repanda, rare.

G. inflata, few. Truncatulina lobatula, few.

G. dubia, few. Uvigerina pi/gmaa, few.

G. rubra, few. Bulimina marginata, rare.

Puhinulina menardii (dwarfed), rare. Nonionina umbilicatula, rare.

P. micheliniana, rare. Biloculina ringens (dwarfed), rare.

P. elegans, rare.

Residue, 83.19 per cent, dark brown, consists of Minerals [40.00], m. di.
0.3 mm.," quartz, felspar, mica, hornblende, magnetite, olivine, glauconite, glassy
fragments. Siliceous organisms [5.00], Diatoms, Radiolarians, and Sponge spi-
cules. Fin^ washings [38.19], argillaceous matter, fine mineral particles, and
fragments of siliceous organisms.

2. Specimens of deposits procured off the Coast of the United States
between Cape Hatteras and Lat. 31° 48' N.

These deposits are green muds or sands. They are with two excep-
tions under 1,000 fathoms, and are mostly under the waters of the Gulf
Stream, or along its inner margin. The mineral particles are much
the same as those in the deposits north of Cape Hatteras, but are all
very much smaller, and have evidently not been transported by ice.
The mineral particles, with the exception of the concretions formed at
the bottom, seldom exceed 0.4 mm. in diameter, and consist of quartz,
felspars, augite, hornblende, magnetite, and a few fragments of glassy
rocks. Glauconitic grains and casts are frequently very abundant, as
are also grains of manganese peroxide.

The carbonate of lime makes up usually over 50 per cent of the
whole deposit, and consists chiefly of the dead shells of pelagic Forami-
nifera, along with shells of pelagic Mollusks, fragments of Echinoderms,

MQSEUM OF COMPARATIVE ZOOLOGY.

41

Polyzoa and coccoliths. All the tropical species of pelagic Forami-
nifera are abundant in these deposits, while they are relatively rare in
the deposits along the coast to the north of Cape Hatteras.

The remains of siliceous organisms, such as Diatoms. Radiolarians,
Sponge spicules, and glauconitic casts of Foraminifera and other organ-
isms, make up usually 10 or 12 per cent of the deposit.

The finer washings of these deposits are of a greenish color, which
seems to be chiefly due to the presence of some organic substance, the
nature of which has not yet been determined. A similar greenish mat-
ter was met with by the " Challenger " in deposits from the same depths
off the coasts of Africa, Australia, Japan, and China.

Phosphate of lime and manganese concretions are present in all the
deposits, and one remarkable concretion of these substances is described
in detail from Station 317, in a depth of 333 fathoms, immediately
under the waters of the Gulf Stream.

Many of these deposits might equally well be called Globigerina oozes.

SfatwniU.—Jjat 32° 2i' N. Long. 78° W W. Depth, 142 fathoms.
Surf. temp. 81°. Bot. temp. 56|°. A greeuish gray saud, granular, very slightly
coherent.

Carbonate of Calcium, 47-G4 per cent, consists of shells of Gasteropods, Lamel-
libranchs, Pteropods, and Ostracodes, fragments of Echiuoderins, coccoliths, and
the following pelagic and other Foraminifera : —

Biloculina ringens, few.

Globigerina bulloides, common. '

G. dubia, common.

G. infiata, common.

G. rubra, common.

G. conglobata, few.

G. sacculifera, few.

G. {OrbuUnd) universa, few.

Spk^roi'Jiiia dehiscens, few.

Puhinulina menardii, common.

P. menardii, var. tumida, com-
mon.

P. michelinian/j, few.

Ptillenia obliquiloculata, com-
mon.

Planispirina calata, few.
Miliolina agglutinans, rare.
M. seminulum, rare.
M. venusta, conmion.
Vernevilina triquetra ? rare.
Texfularia conica, few.
Bulimina marginata, few.
Pela"'ic Nodosaria communis, rare.

Bottom-

species

Cristellaria cultrata, common. V livin

C. rotulata, rare.

C. obtusata, rare.

C. calcar, rare.

C. sp. few.

TJvigerina pygm/pa, few.

TruncatuUna lobatula, few.

Puhinulina elegans, rare.

Rot alia sp.

Nonionina umbilicatula, rare.

species.

42 BULLETIN OF THE

Residue, 52.36 per cent, a green sand, consists of Minerals [40.00], m. di.
0.3 mm., many glauconitic grains, quartz, mica, felspars, hornblende, magne-
tite, augite, phosphatic grains. Siliceous organisms [8.00J, Sponge spicules. Dia-
toms, Kadiolarians, and many fine glauconitic casts of Poraminifera. Fine
Koshings [4 36], traces of argillaceous matter, fine mineral particles, fragments
of Diatoms, and much green amorphous matter.

Station 317. — Lat. 31° 57' N. Long. 78° 18' 35" W. Depth, 333
fathoms. From this place, where the ground was said to be hard, there
was procured a very remarkable concretion that appears to have been
formed in the position from which it was dredged.

This was irregular in form, the greatest diameter being about nine
inches, and of a mottled black, red, and brown color. The surface was
somewhat irregular, and presented many ovoid, smooth projections, the
largest of which were about one centimetre in diameter. The whole
mass was overgrown with sponges, corals, and annelids. Imbedded in
the concretion were two sharks' teeth, resembling Lamna, the largest
being 2\ inches in length and one inch across the base. This tooth is
similar to many found by the '•Challenger" in great numbers in the
greater depths of the Central Pacific, frequently forming the centres of
manganese nodules. In the specimens from the deep water of the
Pacific the interior of the tooth had been in every instance completely
removed, only the hard outer dentine remaining. In the tooth imbedded
in this concretion, on the contrary, the vaso-dentine of the interior of
the tooth is well preserved, in this respect resembling the sharks' teeth
of the same kind found in various tertiary deposits, as for instance in
South Carolina and in the Island of Malta. The vessels of the tooth are
infiltrated with peroxide of iron and manganese and phosphate of lime.

The whole mass has a breccia-like appearance, the several fragments
being cemented by deposits of carbonate of lime and manganese per-
oxide. When thin sections are pi'epared and examined with the micro-
scope, the preparation has a variegated appearance ; all the grains being
closely cemented together. There are numerous sections of pelagic and
other calcareous Foraminifera, of Pteropods, and fragments of Echino-
derras. The interior of the Foraminifera is sometimes completely filled
with calcite, and the same crystals are found cementing many of the
fragments of which the rock is composed. Small fragments of quartz, of
felspars, and of zolene are also seen in the sections. But the most char-
acteristic element is formed by small rounded gi'ains of a brownish or
yellow-green color, having much the aspect of glauconite, which is also
present. Chemical reactions show that these grains are phosphatic.

MUSEUM OF COMPAEATIVE ZOOLOGY. 43

They are similar to the grains found in phosphatic nodules dredged off
the Cape of Good Hope and elsewhere by the " Challenger," and iden-
tical in their physical and chemical properties to the phosphatic grains
in cretaceous rocks.

The manganese is infiltrated through the whole mass of the concretion,
appearing in the microscopic sections in the form of dendrites or concre-
tions, sometimes opaque, sometimes black-brown, and slightly transparent.
The phosphatic grains are sometimes enclosed in the manganese.

The " Challenger " dredged on several occasions, especially off the
Cape of Good Hope, concretionary masses like that above described, but
very much smaller. Phosphatic nodules were always found in the
deposits in depths less than 1,500 fathoms, near continental shores, but
never in the deeper deposits far removed from land.

An analysis of a portion of the above concretion by M. Element,
Brussels, gave as follows : —

Phosphoric acid (PjO.) 23.53

Carbonic " (CO.) 15.56

Sulphuric " (SO3) 2.29

Fluorine 2.28

Chlorine 0.16

Lime (CaO) 52.15

Magnesia (MgO) 1.01

Insoluble residue 0.52

Loss on ignition 3.15

100.65

Oxygen corresponding to Fluorine — 0.96

" corresponding to Chlorine — 0.04

99.65

Atomic Ratios.

P2O5 997

CO2 707

SO3 57

Fl 120

CI 5

CaO 1864

MgO 50

1866

1914

The substance analyzed also contained traces of silica, of iron, of
alumina, and of manganese.

44

BULLETIN OF THE

Siaiion^i^. — 'Lut. 33° 19' N. Long. 76° 12' 30" W. Depth, 457 fath-
oms. Surf. temp. 83°. Bot. temp. 40°. Greeu mud, slightly coherent, granular.

Carbonate of Calcium, 59.43 per cent, chiefly made up of pelagic and other
Foraminifera, as in tiie following list, shells of Pteropods, Gasteropods, and Os-
tracodes, Ecliinoderm fragments, and coccoliths.

G. (Orbulincf) universa, common.
Globigerina bulloid^s, common.
G. conglobata, few.
G. bulloides var. triloba, common.
G. aquilaterali.1, few,
G. sacculifera, few.
G. dubia, common.
G. rubra, common.
Candeina nilida, common.
Sphteroidina dehiscens, few.
Pullenia obliquiloculata, common.
Pulvinulina menardii, abundant.
P. menardii, var. tumida, common.
P. menardii, var. Jimbriata, com-
mon.
P. micheliniana, few.
P. canariensis, few.

>

-a
n

Biloculina rhigens, rare.
Miliolina neminulum, few.
Bulimina marginata, rare.
Poli/morphina sp., rare.
Uvigerina pygmcea, rare.
Sphceroidina bulloides^ common.
Pullenia spheeroides, few.
Truncatulina lobatula, few.
T. sp., rare.

Nonionina umbilicatula, few.
Nodosa ria communis, rare.
N. Uevigata, rare.

o

o
3

Residue, 40.57 per cent, greenish brown, consists of Minerals, [20.00], m. di.
0.1 mm. quartz, liornblende, felspars, glauconite, and glassy fragments. Sili-
ceous organisms [5.00], "Diatoms, Radiolarians, and Sponge spicules, and casts
of many of the organisms mentioned above. Fine washings [15.57], argillaceous
and green amorphous matter, fragments of Diatoms, siliceous spicules, and fine
mineral particles.

3. Specimens of deposits procured around the Shores of the Greater and'

Lesser Antilles.

The specimens are chief!}' from depths between 100 and 1,000 fathoms,
although a few are in depths less than 100 fathoms and a few are over
2,000 fathoms. They are all in more or less close proximity to the
coasts. The mineral particles are chiefly fragments of volcanic rocks or
crystals derived from these, such as monoclinic and triclinic felspars,
hornblende, augite, olivine, magnetic iron, and pumice ; along with a few
fragments from ancient rocks, as quartz, tourmaline, mica, and epidote.
Glauconitic grains were rare in these deposits, and phosphatic grains
were likewise rare. In the deposits farthest from land the size of the

MUSEUM OF COMPARATIVE ZOOLOGY. 45

mineral particles seldom exceeded 0.1 mm. in diameter, but near shore
they were very much larger, and fragments of rucks and pebbles were
frequently dredged. Altered fragments of plagioclase, basalts, and dia-
base were rather frequent.

The percentage of carbonate of lime in these deposits was usually
very high, being frequently 70 or 80 per cent, and in the case of a chalk
rock 90.24 per cent. Where the shores were composed of volcanic or
other rocks not calcareous, the debris of these made up the larger part
of the deposits, which might be called volcanic muds. But the majority
of the deposits should be termed Pteropod or Globigerina oozes, owing
to the large number of these organisms present in them. It should be
remembered, however, that both in the size of the mineral particles and
in the nature of a large number of the calcareous particles, these de-
posits differ considerably from similar deposits found far away from land
in the open ocean and called also Pteropod and Globigerina oozes.

The siliceous organisms never make up more than four or five per
cent of the whole deposit, and consist of Radiolaria, Sponge spicules, and
a few Diatoms.

Fragment of White Chalk. — From 994 fathoms, off Nuevitas, Cuba,
there was obtained a fragment of white chalk coated on the surface with
streaks of peroxide of manganese. This chalk contained 90.24 per cent
of carbonate of lime. The sections showed the rock to be composed of
crystalline grains of carbonate of lime, which however were not the
result of precipitation. A few sections of Globigerina and Textularia
were observed, but no other organisms could be recognized. After dis-
solving away a considerable quantity, small fragments of quartz and
hornblende, Sponge spicules and Radiolarians were observed in the resi-
due. It is impossible to be certain that this rock was formed in the
position from which it was dredged, though there are reasons for sup-
posing that it was. The ooze which came up from the same place was
of a reddish or brownish tinge, and contained an immense number of
Pteropods, Heteropods, and pelagic Foraminifera ; the percentage of
lime was not so high as in the white chalk rock, and the residue was
much darker in color.

Concretions. — Off the Birbadoes in 221 fathoms (St. 280) a very
hard calcareous concretion was obtained, which showed perfectly how
the rock was formed by crystallization of cai'bonate of lime around the
shells of Foraminifera and other centres. A zone is seen around the
shells, composed of fibro-radiate calcite ; the crystals of calcite, coming
from the various centres, abut against each other, and frequently leave

46

BULLETIN OF THE

an empty space between. When these spaces are filled by a further
deposition of lime, the whole becomes very compact and massive.

The centres of the Foraminifera are frequently filled with a gray or
yellowish substance which does not, however, give the reactions of
pliosphate of lime.

The mineral particles were very few in number, among them frag-
ments of quartz and plagioclase being observed. This concretion was
about two inches in diameter and had a rough areolar surface on which
Serpulae and Polyzoa were growing.

A similar and somewhat larger concretion from 200 fathoms (St.
291) was also obtained off' the Barbadoes, which was much more over-
grown with organisms, and on its upper surface had a large cavity in
which a hermit-crab had lived. (Polycheles Agassizii, see Bulletin VIII.
No. 1.)

Off" the north coast of San Domingo, in 772 fathoms (No. VI.), there
were obtained several small manganese Nodules and a few fragments of
a Corallium coated with manganese, precisely similar to that dredged by
the "Challenger" in 1,52.5 fathoms near the Cape Vcrdcs (see Narrative
of the Voyage, page 125). The interior of the nodules were of a liglit
brownish color and were composed in all cases chiefly of a mass of pelagic
Foraminifera. The largest of these nodules had a diameter of about
two inches. Microscopic sections of the nodules and concretions were
easily made and showed with great distinctness the structure of the
mass, composed chiefly of pelagic Foraminifera cemented together as
above stated,

Stcdion 103. — Old Baliama Channel. Depth, 438 fathoms. Surf. temp. 79°
Bot. temp. 49 1°. A Ptcropod ooze or -wliite coral mud, slightly cohereut when,
dry, chalky.

Carbonate of Calcium, 87.06 per cent, consists of Gasteropod, LamelHbranch,
Ostracode, Ptcropod and Heteropod shells, calcareous Alga?, Echinoderm frag-
ments, Polyzoa, Alcyouium spicules, coccoliths and rhabdoUths, and the following
Foraminifera : —

Glohigerina dubia

G. rubra

G. hirsuta

G. (pquilateralis

G. {Orbulinu) wiiversa

Tiihinulina menardii

P. menardii, var. tumida

P. mirhdiniana

Pullenia obliquiloculata

• Pelagic
species.

Ci/mbalopora buUoides
Milioli/ia seminulum
M. Uiuifpana
M. bicornis
M. arjglutinans
Biloculina cornuta
Pulrinulina sp.
Cassidnlina crassa
Textularia turris

MUSEUM OF COMPARATIVE ZOOLOGY. 47'

Discorhina sp. Cristellaria cultrata

Truncatulina sp. Vertebralina striata

Poli/trema rubra Articulina conico-articulata

Carpenteria sp. Bulimina margiiiata

Orbiculina adunca Nodosaria costalata
Orbitolites marginalis

Residue, 12.94 per cent, light brown, consists of Minerah [3.00], m. di.
0.1 mm., quartz, liorubleude, magnetite, mica, olivine, and a few glassy frag-
ments. Siliceous organisms [3.00], Sponge spicules, Diatoms, and a few casts.
Fine washings [6.94], argillaceous matter, fine mineral particles, and fragments
of siliceous organisms.

Station ll'i.—'W. of Navassa Bank, 19 Dec, 1S78. Depth, 1050 fathoms.
Surf. temp. 82°. Bot. temp. 39|°. A light brown Globigerina ooze, slightly
coherent, pulverulent, granular ; dries into lumps, wliich break easily between the
fingers.

Carbonate of Calcium, 62.38 per cent, consists of Lamellibranch, Pteropod, and
Heteropod shells, coccoliths and rhabdoliths, and the following Foraminifera : —

Globigerina bulloides P. menardii, vzx . fimbriata

G. rubra P. micheliniana

G. aquilateralis P. canariensis

G. dubia Pullenia obliqniloculata

G. hirsuta Biloculina depressa

G. sacculifera B. sphara

G. {prbulind) unitersa Cassidulina sp.

Sphteroidina dehiscens Webbina clatala

Candeina nitida Truncatulina lobatula

Palcinulina menardii TJcigerina sp.

P. menardii, var. tumida,

Residue, 37.62 per cent, red, consists of Minerals [15.00], m. di. 0.07 mm.,
(angular) felspars, quartz, hornblende, mica, magnetite, many glassy fragments.
Siliceous organisms [4.00], Sponge spicules, Badiolarians, and a few casts. Fine
tcashings [18. G2], argillaceous matter, fine mineral particles, and fragments of
siliceous orgranisms.

'D"

Station 117. — Lat. \T 47' 20" N. Long. 67° 3' 20" W. Off Porto Rico.
Depth, 874 fathoms. Surf. temp. 82|°. Bot. temp. 40°. A coral mud or Ptero-
pod ooze, slight coherent, granular. Also, a small quantity of larger material,
which appears to have been washed from the dredge, consisting of Gasteropod,
Lamellibranch, Ostracode, Pteropod, and Heteropod shells, Echinoderm frag-
ments. Coral, Polyzoa, and Serpula tubes.

Carbonate of Calcium, 70.6G per cent, consists of Pteropods, Hetcropods, frag-

48

BULLETIN OF THE

meats of Echinoderms aud Gasteropod and Lamellibrauch sliells, calcareous Algaj,
coccoliths, aud the followiug Foraminifera : —

Globlgerina rubra

G. dubia

Sphmroidina bulloides

G. hirsuta

Truncatulina lobatula

G. sacculifcra

T. sp.

G. cequilateralis

Rupertia sp.

G. conglobata

Rotalia sp.

G. {Orbulim) unicersa

Cristellaria cultrata

Sphmroidina dehiscens

Pelagic

Ijugena squamata

Pullenia obliquiloculata

species.

Textularia biculeata

Pulviiiulina menardii

Clavidina ci/lindrica

P. menardii, var. tumida

Gaudri/iiia rugosa

P. menardii, ^^x.fimbriata

Biloculina depressa

P. michelviiana

B. ringens

P. canariensis

B. sphara

P. sp.

Hcsidue, 29.31. per cent, dirty brown, consists of Minerah [10.00], m. di.
0.05 mm., (angular) quartz, hornblende, mica, felspar, olivine, scoriee, small
fragments of rocks. Siliceous organisms [7-00], Sponge spicules and Radiola-
riaus. Fine washings [12.34], argillaceous matter, fine mineral particles, and
fragments of siliceous organisms.

Station 138. — 0(T Santa Cruz, January 7, 1879. Depth, 2,375 fathoms. Surf,
temp. 7r)?/. Bot. temp. 3S^°. A light brown Globigeriua ooze, slightly coherent,
pulverulent.

Carbonate of Calcium, 63.54 per cent, consists of Gasteropod and Lamelli-
brauch shells (larval forms), Ostracode, Pteropod, and Heteropod shells, Alcy-
onium spicules, Eeliinoderm fragments, coccoliths aud rhabdoUths, aud the
following Foraminifera : —

Globigerina rubra

G. dubia

G. conglobata

G. sacculifera

G. bulloides, var. triloba

G. (^Orbulind) universa

Pulrinulina menardii
Pulcinulina michcliniana
P. canariensis
Planorbulina sp.
Miliolina bicornis
M. circularis

Residue, 36.46 per cent, red, consists of Minerals [20.00], m. di. 0.2 mm.,
several fragments of mica scliist 3 to 5 mm. in diameter, felspars, quartz, mica,
hornblende, magnetite. Siliceous organisms [5.00], Sponge spicules. Fine
washings [11.46], amorphous clayey matter, fine mineral particles, and frag-
ments of siliceous spicules.

MUSEUM OF COMPARATIVE ZOOLOGY, 49

Station 182. —Off Dominica. Depth, 1,131 fathoms. Surf. temp. 81°. Bot.
temp. 392°. -^ ^'o^* browu volcanic mud (dark when wet), coherent, plastic,
earthy, slightly granular.

Carbonate of Calcium, 13.78 per cent, consists of Pteropods, Echiuoderm
fragments, coccoliths, and the following Foraminifera : —

Globigerina rubra P- micheliniana

G. dubia Spharoidina bulloidea

G. co>if/lobata Fullenia quinqudoba

G. scicculifera Truncatulina lobatula

G. {Orbalina) universa Foli/morphina sp.

Spharoidina dehiscens Uvigerina asperula

Fullenia obliquiloculata Vaginulina sp.

Fuhinulina menardii Cassidulina crassa

P. menardii, ysLV.Jimbriata Biloculina, fragments.

Residue, 86.23, brown, consists Qi Minerals [35.00], m. di. 0.3 mm. (angular),
quartz, liornblende, magnetite, felspar, olivine, augite, a few glassy fragments,
fragments of scoria;. Siliceous organisms [2.00], Sponge spicules. Fitie wash-
ings [49.22], argillaceous matter, fine mineral particles, and fragments of sili-
ceous spicules.

Station 197. — Off Martinique. Dopfh, 1,224 fathoms. Surf. temp. 80°.
Bot. temp. 39°. A light brown volcanic mud, coherent, plastic, earthy, slightly
granular.

Carbonate of Calcium, 13.41 per cent, consists of otoliths of fish, Pteropods,
Echinoderra fragments, coccoliths, and Foraminifera as follows : —
Globigerina rubra Fullenia obliquiloculata

G. dubia Fulcinulina elegans

G. conglobata Fullenia quinqueloba

6. sacculifera Truncatulina lobatula

G. injlata T. robertsoniana

G. bulloides, van triloba Lagena sp.

G. {Orbulina') unioersa Cassidulina crassa

Fulcinulina menardii Haplophragmium globigeriniformis

P. menardii, var. tuinida Trochammina ringens

P. micheliniana Reophax nodulosa, fragments.

Residue, 86.59 per cent, brown, consists of Minerals [60.00], m. di. 0.5 mm.
(angular), felspar, magnetite, olivine, augite, quartz, hornblende, palagonite, and
fragments of pumice from 1 to 2 mm. in diameter. Siliceous organisms [3.00],
Radiolarians, Diatoms, and Sponge spicules. Fine tcashings [23.59], argillaceous
matter, fine mineral particles, and fragments of siliceous organisms.

Station 241. — Off Grenadines. Depth, 163 fathoms. Surf. temp. 80°. Bot.
temp. 53°. A yellowish brown Pteropod ooze, has a greenish tinge when wet,
slightly coherent, pulverulent, granular.

VOL. XII. NO. 2. 4

50 BULLETIN OF THE

Carbonate of Calcium, 76.20 per cent, consists of otoliths of fish, Serpula tubes,
Ostracode, Pteropod, and Heteropod shells, fragments of Polyzoa, Echiuoderms,
calcareous Algae, and the following Poraniinifera : —

Qlobigerina bulloides Cristellaria sp.

G. buLoideSy var. triloba Textularia conica

G. rubra T. agglutinans

G. infiata Cassidulina crassa

G. conglobala Clavulina parisiciisis

G. sacculifera Verneuilina spinulosa

G. ifirbulind) universa Haplostiche soldanii

Pulvinulina menardii Nonionina umbilicatula

P. menardii, var. tumida Amphistegina mamillata

P. micheliniatia Orbiculina adunca

P. sp. Articulina sagra

Sphceroidina bulloides Planispirina celata

Poli/trema rubra Spiroloculina limbala

Planorbulina mediterranensis Miliolina seminulum

Discorbina sp. M. macilenta

Truncatulina lobatula M. linneeuna

T. sp. M. aggludnans

Polymorphina sp. Biloculina ringens (very small).

Residue, 23.80 per cent, yellowish green, consists of Mitterals [10.00], ni. di.
0.25 mm. (angular), quartz, hornblende, felspar, magnetite, augite, olivine. Sili-
ceous organisms [5.00], Diatoms, Radiolarians, Sponge spicules, and a few pale
glauconitic casts. Fine washings [8.80], argillaceous matter, fine mineral par-
ticles, fragments of siliceous organisms, and greenish organic matter.

Station 275. — Off Barbadoes. Depth, 218 fathoms. Surf. temp. 80^. Bot.
temp. 52^°. A Pteropod ooze or Toraminiferal sand, somewhat coherent, pul-
verulent, granular, dries into lumps which are easily broken by the pressure of
the fingers.

Carbonate of Calcium, 38.09 per cent, consists of otoliths of fish, Gasteropod,
Lamellibranch, Pteropod, Heteropod, and Ostracode shells, fragments of Echiuo-
derms and Polyzoa, Alcyonium spicules, coccoliths, and the following Foram-
inifera : —

Globigerina rubra Candeina nitida

G. dubia Sphesroidina dehiscens

G. infata Pullenia obliquilorulata

G. conglobata Pulvinulina menardii

G. sacculifera P. menardii v&x.Jimbriata

G. (equilateralis P. micheliniana

G. bulloides var. triloba Biloculina ringens

G. (jOrbulina) universa B. depressa

MUSEUM OF COMPARATIVE ZOOLOGY. 51

Millolina seminulum Cristellana cultrata

Spiroculinu impressa C calcar

Vertebralina striata Sagrina columnella

ClavuUna communis Uvigerina pygmaa

C. parisiensis Truncatulina lobatula

Textularia coiiica Planorbulina sp.

T. lucidenta Nonionina timbilicatula.
T. agglutinans

Residue, 61.91 per cent, yellowish brown, consists of Minerals [25.00], m. di.
0.2 to 0.3 mm. magnetite felspar, quartz, hornblende, and a 'i^sy glassy fragments.
Siliceous organisms [25.00], many Sponge spicules, a few Diatoms, one or two
Radiolarians, and glauconitic casts of the calcareous organisms. Fine washings
[11.91], amorphous clayey matter, with fragments of casts, fine mmerals, and
siliceous particles.

4. Specimens of deposits procured in the Gulf of Mexico and in the
Florida Strait.

During the years 1875, 1876, 1877, and 1878, very extensive series
of soundings wei'e obtained at all depths, and in all parts of the above
areas.

There is a very great variety in the shallow water deposits under 100
fathoms. Near the coasts of the North American continent, where
rivers enter, and where there are few coral reefs, the deposits are either
sands or fine clayey muds, formed of detrital matter brought down from
the land. Where the shores are lined by coral reefs, the deposits are
chiefly made up of coral debris, the shells of pelagic Foramiuifera and
Mollusks and other calcareous organisms.

The character of the deposits in depths greater than 100 fathoms is
likewise largely determined by the greater or less proximity to the
embouchure of rivers or to coral reefs.

In all the deeper deposits in the Gulf of Mexico and Strait of Florida,
the crystalline mineral particles are very small, rarely exceeding one-
tenth of a millimetre in diameter. They consist principally of small
rounded grains of quartz, with fragments of felspars, mica, horn-
blende, augite, magnetite, and rarely tourmaline. In a few places there
were fragments of pumice, and glauconitic particles were occasionally
noticed. The mineral particles and fine clayey matter appear to be
almost wholly derived from North American rivers.

The carbonate of lime in the deposits of these regions is mostly made
up of the shells of pelagic Foramiuifera and Mollusks. In depths greater

62 BULLETIN OF THE

than 2,000 fathoms the Pteropod and Heteropod shells appear to be
nearly, if not quite, absent, — the carbonate of lime then consisting of
the shells of pelagic Foraminifera ; in less depths the Pteropod and
Heteropod shells are present, and in depths varying from 200 to 500
fathoms they make up the bulk of the deposits in many places. In
several of the deposits, where the percentage of carbonate of lime is very
high, the whole has a very chalk-like appearanee ; it appears, indeed, as
if it were in the process of transformation to true chalk.

The siliceous organisms consist of Radiolarians and Sponge spicules,
with a few Diatoms, but these seldom make up more than three or four
per cent of the whole deposit.

Phosphatic Concretions. — The phosphatic concretions in the dredg-
ings in Florida Strait are very interesting. In a great many deep-sea
deposits there is usually a small percentage of phosphate of lime, but
near the shore, in some instances, the quantity is very considerable.
Sharpies, who analysed the ooze of the Gulf Stream, found —

Carbonate of Lime 85.62

" of Magnesium 4.26

Silica 1.32

Alumina

Oxide of iron 0.31

PuospHATE OF Lime 0.18

Loss on ignition 8.15

100.04

In certain concretions found by the " Blake " in the Florida Strait,
and by the " Challenger " in various parts of the world near land, the
quantity of phosphate of lime is very much greater than in the deposits.
These concretions appear always to be associated in an intimate way
with organisms.

In 125 fathoms S. W. of Land Key, Florida, a fragment of bone was
obtained several centimetres in diameter. It was of a dirty brown
color, of great hardness, and had a conchoidal fracture. A microscopic
examination of thin sections showed that the bone structure was per-
fectly preserved.

The following is the result of an analysis of this specimen by M.
Klement : —

Phosphoric acid (PjO^) 33.42

Carbonic " (COj) ...... 5.80

MUSEUM OF COMPARATIVE ZOOLOGY. 53

Suiphuric acid (S Ogj 2.74

Fluorine 1.21

Lime (Ca 0) 51.90

Magnesia (Mg 0) 0.70

Iron and Alumina. ....... 1.56

Insoluble residue 0.21

Loss on ignition 2.16

99.70
Oxygen corresponding to Fluorine . . — 0.51

99.19

There were also traces of Silica and Chlorine.

Atomic Ratios.
P2O5 1417

1814.

CO2 264 I

s O3 69 r

Fl 64 >'

Ca 0 1853

. ^ . 1888

Mff 0 35

■^o

}

At the same place and depth there was a concretion of a brown color
consisting of an aggregation of calcareous organisms cemented by a
brownish yellow matter, often showing concentric rings after the man-
ner of agate. This yellowish brown matter is isotropic, between crossed
nicols only the calcite and the shells of the Foraminifera brighten np ;
the calcite lies crystallized in the interior of the Foraminifera. In
treating the brown or yellow parts under the microscope with molyb-
date of ammonium and nitric acid, there is an abundant yellow precipi-
tate characteristic of phosphoric acid.

At other stations small phosphatic concretions were also obtained by
the " Blake," all more or less resembling those described above. There
are difficulties in understanding how phosphate of lime and carbonate of
lime are deposited at the bottom of the sea, yet there is no doubt that
8uch a deposition does take place nnder some special circumstances.
Their solution is, however, an almost universal phenomenon in the
ocean.

Specimen 60, Li>ie P'. — Lat. 24° 50' N. Lon^. 84° 50' 45" W. 15 May,
1875. Depth, 200S fathoms. A reddish brown Globigeriiia ooze dries into
slightly coherent lumps.

54

BULLETIN OF THE

Carbonate of Calcium, 47.87 per cent, consists of coccoliths, rhabdolitlis, and
the folio wiuof Foramiiiifcra : —

Globlgerina codglohala

G. but laid es

G. bulloides, var. triloba

G. sacculifera

G. cequilateralis

G. rubra

G. dubia

G. {Orbulina) universa

Candeina nitlda
Pullenia obliquiloculata
PulcinuUna menardii
P. menardii, var. tumida
P. canariensis
P. elegans

Truncatulina lobatula
Nonioniiia umbilicatula

Residue, 52.13 per cent, reddish brown, consists of Minerals [20.00], m. di.
0.05 mm., quartz, mica, felspar, hornblende, magnetite, palagonite, glauconite.
Siliceous organisms [5.00], Sponge spicules, glauconitic or other casts. Fine
washings [27.13], amorphous clayey matter, with fine mineral particles and frag-
ments of siliceous spicules.

Specimen 4, Line P. — Lat. 26° 40' N. Long. 96" 01' W. 29 January, 1877.
Depth, 4S9 fathoms. A brown mud, coherent, plastic. This deposit resembles
very much a fine river clay, mixed with a very few pelagic Foraminifera ; it would
seem, judging from its position, to be derived from the fine detrital matter carried
down by the rivers.

Carbonate of Calcium, 2.76 per cent, consists of one or two coccoliths along
with the following Foraminifera : —

Globlgerina bulloides
G. dubia
G. rubra
G conglobata
Pullenia obliquiloculata
PulcinuUna menardii
P. menardii, var. tumida
P. micheliniana

Pelagic
species.

Biloculina ringens
Ammodiscus charoides
BoUvina a^nariensis
Bulimina rostrata
B. oculata
Nodosaria raphanus
Ucigerina asperula
U. asperula, var. aiiberiana
Spharoidina bulloides
Truncatulina lobatula

Bottom-
living
species.

PulcinuUna elegans

Residue, 97.24 per cent, of a light slaty-brown color, consists of Minerals
[25.00], m. di. 0.01 mm., quartz, magnetite, mica, felspars, augite, hornblende,
and several small red particles. Siliceous organisms [1.00], siliceous spicules and
fragments of Radiolarians. Fine washings [71.24], amorphous clayey matter.

Specimen 21, Line BE. — Lat. 20° 59' N. Long. 96° 39' W. 25 May, 1877.
Depth, 511 fathoms. Volcanic mud, very coherent, clayey.

Carbonate of Calcium, 15.14 per cent, consists of Echinoderm fragments, fish
teeth, and Foraminifera as follows : —

MUSEUM OF COMPARATIVE ZOOLOGY.

55

Globif/erina rubra

Planispirina celata

G. duhia

Bolicina a-nariensis

G. injtata

Nonionina umbilicatula

G. con/jlobata

Lagena squamosa

G. bulloides

Pelagic
species.

Ammodisrus charoides

G. bulloides, var. triloba

Uvif/erina asperula

G. {Orbulina) unicersa

Cassidulina crassa

Fullcnia obliquiloculata

Bulimina marginata

Puki/iulina menardii

Truneatulina lobatula

P. micheliniana

Puhinulina elegans

Bottom-
living
species.

Residue, 8i.86 per cent, chocolate color, consists of Minerals [50.00], m. di.
0.1 mm., quartz, pumice fragments, magnetite, hornblende, tourmaline, glaucoiiite,
mica, many glassy fragments. Siliceous organisms [3 GO], Radiolariaus and
Sponge spicules. Fine washings [31.86], argillaceous matter, fine mineral par-
ticles, and a few fragments of siliceous spicules.

Specimen 23, Litie DD— Lat. 22° 00' N. Long. 92° 13' W. 22 May, 1877.
Depth, 353 fathoms. A Uglit greenish gray fine calcareous mud, coherent.

Carbonate of Calcium, 67.81 per cent, consists of Echinoderm fragments,
Pteropod, Ostracode, Gasteropod, and Lamellibranch shells, and the following
Foraminifera : —

Globigerina rubra

G. dubia

G. conglobata

G. infata

G. bulloides, var. triloba

Pullenia obliquiloculata

Puhinulina menardii

P. canariensis

Miliolina seminulum

M. sp.

Bulimina margin-ata
B. aculeata
Bolivina nobilis
B. (snariensis
Truneatulina lobatula
Uvigeri/ia pi/gmcea
Nodosaria hispida
Textularia conica
T. sp.

Residue, 32.19 per cent, consists of Minerals [3.00], m. di. 0.05 mm., quartz,
felspar, hornblende, magnetite, glauconite, glassy fragments, and a few red
particles. Siliceous organisms [10.00], Geodia and other Sponge spicules,
Diatoms and Radiolariaus. Fine washings [19.19], argillaceous matter, fine
mineral particles, and fragments of siliceous organisms.

Specimen 51, Line P' . —Lat. 25° OS' 15" N. Long. 87° 12' 50" W. 14 May,
1875. Depth, 2119 fathoms. A brown Globigerina ooze, shghtly coherent.

Carbonate of Calcium, 41.86 percent, consists of a few coccoliths and rhab-
doliths, Ostracode valves, Echinoderm fragments, and the iollowing Forami-
nifera : —

Ob

BULLETIN OF THE

P. michellniana
P. canariensis
Truncatulina lohatula
Puhinulina elegans
Biloculina depressa
Haplophrar/mium fflobif/eriiiiformis
Hi/iieramml/ia vacjans
Ammodiscus charoides
Noniotiina umhilicatula
N. pompilioides
Vcifjcrina afiperida
Clamdiiia communis
Rcophax (fragments).

Globigerina iiijtata

G. rubra

G. dubia

G. (eqiiilateralis

G. sacciilifera

G. conglobata

G. buUoides, var. triloba

G. (Orbulina) unicersa

Candeinu nitida

PuUenia obliqidloculata

Bphccrulditia dehisce ns

Pulcinulina menardii

P. menardii, var. tumida

P. menardii, \'A\\Jimbriata

Residue, SS.H per cent, light brown, consists of Minerals [30.00], m. di.
O.L mm. (mostly rounded), quartz, felspar, mica, hornblende, glanconite, magne-
tite, tourmaline. Siliceous organisms [3.00], Sponge spicules and lladiolarians.
Fine washings [25.14], argillaceous matter, fine mineral particles, and fragments
of siliceous organisms.

Specimen 15, Line F'. — Lat. 27° 55' N. Long. 89° 53' W. 17 March, 1875.
Depth, 407 fathoms. A gray mud, clayey, coherent, plastic.

Carbonate of Calcium, 10.27 per cent, consists of otoliths of fish, Pteropod
fragments, and the following Foraminifera : —

Pulvinulina pawperata
P. elegans

Globigerina rubra

G. dubia

G. buUoides

G. (equilaieralis

G. sacculifera

G. {Orbulina) unicersa

Pulvinidina menardii

P. menardii, var. tumida

P. micheliniana

PuUenia obliqidloculata

Biloculina ringens

Planispirina celala

PuUenia spharoides

^ Pelagic
species.

Haplophragmium globigeriniformis

Chilostomella ovoidea

Bolicina anariensis

Bulimina marginata

Sagrina columnella

Virgulina subsquamosa

Truncatulina lobatula

Uvigerina ])ggma;a

U. asperida

Lagena orbignyana

L. sp.

Sphceroidina buUoides

Residue, 89.73 per cent, light brown, consists of Minerals [10.00], m. di.
0.05 mm., quartz, angite, magnetite, felspars, hornblende, and a few small red
particles. Siliceous organisms [3.00], casts of Foraminifera, Sponge spicules, and
Radiolarians. Fine icashings [70.73], amorphous clayey matter, and fragments
of siliceous organisms.

MUSEUM OF COMPARATIVE ZOOLOGY.

57

Specimen 40, Line F . — Lat. 25° 31' 45" N. Long. 90° 28' W. 13 May,
1875. Depth, 1,922 fathoms. A dark brown Globigerina ooze, coherent, plastic.

Carbonate of Calcium, 36.54 per cent, consists of Echini spines, Ostracode
valves, coccoliths, and the following Foramiuifera : —

Biloculina depressa

Miliolina sp.

Truncatulina lobatula

Nonionina pompilioides

Globigerina rubra

G. dubia

G. conglobata

G. sacculifera

G. bulloides, var. triloba

Pullenia obliquiloeulata
Bottom-living Sphcproidina dehiscem
species. Candeina nitida

Pulvinulina menardii
P. menardii, var. tumida
P. menardii, \d.r . Jimbriata
P. micheliniana
P. canariensis

Residue, 63.46 per cent, reddish, consists of Minerals [30.00], m. di. 0.07 mm.,
quartz, mica, felspar, augite, plagioclase, glauconite, and red palagonite-like par-
ticles. Siliceous organisms [5.00], Radiolarians, Sponge spicules, and brown
flexible casts of Foramiuifera. Fine washings [28.46], amorphous clayey matter,
with fiue minerals and fragments of siliceous spicules.

Specimen 30, Line G C. — Lat. 23° 23' N. Long. 94° 39' W. May 17, 1877.
Depth, 2,057 fathoms. A reddish Globigerina ooze, coherent, clayey, with lus-
trous streak.

Carbonate of Calcium, 32.12 per cent, consists of a very few coccoliths and
rliabdoliths, and the following Foramiuifera : —

Globigerina dubia

G. rubra

G. sacculifera

G. conglobata

G. helicina

G. bulloides, var. triloba

G. several irregularly growing forms.

G. iOrbulinci) unicersa

Pullenia obliquiloeulata

Sphceroidina dehiscens

Pulvinulina menardii
P. menardii, var. tumida
P. micheliniana
P. canariensis
Truncatulina lobatula
Nonionina umbilicalula
N. pompilioides
Pulcinulina elegans
Bolivina textilarioides
Miliolina cultrata

Bottom-
living
species.

Residue, 67.88 per cent, red, consists of Minerals [15.00], m. di. 0.05 ram.,
quartz, felspars, magnetite, augite, hornblende, a few red particles, glassy frag-
ments, and fragments of scoriiE. Siliceous organisms [3.00], Sponge spicules,
and fragments of Radiolarians. Fine washings [49.88], argillaceous matter, fine
mineral particles, and a few fragments of siliceous spicules.

Specimen 21, Line C C. —Lat. 23° 18' iM Long. 92° 03' W. Depth 2,080
fathoms. A light brown Globigerina ooze, reddish when wet, coherent, clayey.
Carbonate of Calcium, 35.52 per cent, chiefly made up of pelagic Foramiuifera,

58

BULLETIN OF THE

along with Ostracode shells, fragments of Ecliinoderras, coccoliths, and rhabdo-
liths. The following is a list of the Foraminifera : —

Globigerina bulloides, few, small.

G. bulloides, var. triloba, common.

G. dubia, common, large.

G. eequilateralis, few.

G. rubra, abundant.

G. conglobata, common.

G. sacculifera, common.

G. {Orbulina) universa, abundant.

Candeina nitida, few.

Pullenia obliquiloculata, abundant.

Spheeroidina dehiscens, few.

Puloi/iulina menardii, abundant.
P. menardii, var. tumida, abundant
P. menardii, y^x.fimbriata, few.
P. micheliniana, abundant.
P. canariensis, few.
Truncatulina lobatula, few.
Nonionina pompilioides,{tv/ .
Rotalia soldanii, rare.
Bolivina sp., rare.
Biloculina ringens, rare.

Bottom-
\ living
species.

Miliolina sp., rare.

Residue, 64.48 per cent, reddish, consists oi Minerals [3.00], m. di. 0.05 mm.,
felspars, quartz, magnetite, augite, hornblende, glassy fragments. Siliceous or-
ganisms [3.00], Sponge spicules, Diatoms, Radiolarians, oasis of Foraminifera.
Fine washings [58. 4S], amorphous clayey matter, fine mineral particles, and frag-
ments of siliceous organisms.

Station 4.

temp. 39^0

- Off Morro Light. Depth, 936 fathoms. Surf. temp. 77|°. Bot.
A Pteropod ooze, of a grayish white color, chiefly composed of
Pteropods, with many pelagic Foraminifci-a, slightly coherent.

Carbonate of Calcium, 68.84 per cent, consists of otolitlis of fish, Gasteropod,
LameUibranch, Ostracode, Pteropod, and Heteropod shells, Echiuoderm frag-
ments, coccoliths and rhabdoliths, ana Foraminifera as follows : —

Globigerina bulloides

G. rubra

G. dubia

G. cequilateralis

G. sacculifera

G. conglobata

G. {Orbulina) universa

Candeina nitida

Spheeroidina dehiscens

Pullenia obliquiloculata

Pultinulina menardii

Pelagic
species.

Biloculina sphara
B. depressa
Miliolina sp.
Planispirina celata
Hyperammina ramosa
H. vagans
H. subnodosa
Ammodiscus incertus
Gaudrjfina pupoides
G. rugosa
Cassidulina crassa
Truncatulina lobatula

Bottom-living
species.

Spheeroidina bulloides _
Residue, 31.16 per cent, grayish brown, consists of Minerals [10.00], m. di.
0.07 mm., quartz, hornblende, felspars, plagioclase, orthoclase, mica. Siliceous
organisms [15.00], Radiolarians, Diatoms, and Sponge spicules. Fine washings
[6.16], argillaceous matter, fine minerals, fragments of siliceous organisms, and
greenish organic matter.
Note. — Fragments of an areolar tufaceous rock were obtained in the dredging.

MUSEUM OF COMPARATIVE ZOOLOGY.

59

Long. 830 49' w. Depth, 392 fathoms. Surf.

A grayish green coral tnud, pulverulent and

Station 27. — Lat. 24° 30' N.
temp. 73°. Bot. temp. 44|°.
granular.

Carbonate of Calcium, 82.06 per cent, consists of otoliths of fish, Gasteropod,
Lamellibranch, Ostracode, Pteropod, and Heteropod shells, Echinoderm frag-
ments, coccoliths and rhabdoliths, and Foraminifera as follows : —

Glohigerina rubra
G. dubia
G. conglobata
G. bulloides

G. {Orbulind) universa
PuUenia obliquiloculata
Pulvinulina menardii
P. micheliniana
Sphceroidina bulloides
Miliolina venusta
M. seminulum
Cassidulina crassa
BoHvina dilatata

Textularia sp.
Bulimina aculeata
Nodosaria hispida
TJvigerina usperula
Cristellaria variabilis
Discorbina obtusa
D. allomorphinoides
Truncatulina lobatula
T. ungeriana
T. rosea

Rotalia soldanii
Polystomella crispa
P. striatopunctata
Nonionina ■umbilicafula.

Bigenerina sp.

All the Foraminifera in this deposit appear very small (dwarfed).

Residue, 17. 9i per cent, dark green, consists of Minerals [5.00], m. di. 0.1 mm.,
quartz, felspars, hornblende, magnetite, plagioclase, mica, many glassy fragments.
Siliceous organisms [10 00], Sponge spicules, Radiolarians, Diatoms, and a few
casts of Foraminifera. Fine washings [2.94], argillaceous and green flocculent
matter, fine mineral particles, and fragments of siliceous organisms.

Station 33. —Lat. 24° 1' N. Long. 88° 58' W. Depth 1,568 fathoms. Surf,
temp. 72^°. Bot. temp. 40|°. A light brown Glohigerina ooze, with a rosy
tinge, dark brown when wet, coherent, pulverulent, granular.

Carbonate of Calcium, 72.21 per cent, consists of otoliths of fish, Pteropod
and Ostracode shells, Echinoderm fragments, coccoliths and rhabdoliths, and the
following Foraminifera : —

Globigerina rubra

Miliolina seminulum "^

G. dubia

Biloculina dejjressa

G. conglobata

B. tubulosa

G. sacculifern

Cassidulina crassa

Bottom-

G. {Orbulina) universa

Pelagic

Lagena hispida

- living

PuUenia obliquiloculata

species.

TJvigerina asperula

species.

Spharoidina dekiscens

Pulvinulina elegans

Pulvinulina menardii

Truncatulina lobatula

P. menardii, v&r.^mbriata

T. ungeriana

P micheliniana j

60

BULLETIN OF THE

Residue, 27.79 per cent, reddisU brown, consists of Minerals [6.00], m. di.
0.15 mm., quartz, hornblende, magnetite, felspar, glassy fragments. Siliceous
organisms [10.00], Sponge spicules, Radiolarians, Diat-oms. Fine washings
[11.79], argillaceous and flocculent matter, fine mineral particles, and fragments
of siliceous organisms.

Station 41. — Lat. 23° 42' N. Long. 83* 13' W. Depth, 860 fathoms. Surf,
temp. 73°. Bot. temp. 39^*'. A white chalky Pteropod ooze, granular ; with
several hard chalky concretions, which are perforated by worms, and in parts
showing deposits of manganese.

Carbonate of Calcium, 83.67 per cent, consists of otoliths of fish, Pteropod and
Heteropod shells, coccoliths, rhabdoliths, and Foraminifera as follows : —

Glohigerina rubra
G. inflata
G. sacrulifera
G. conglobata
G. dubia

G. bulloides, var. triloba
G. {Orbulina) universa
Spharoidina dehiscens
Candeina nitida
FulvinuliAa menardii
p. menardii var. tuniida
P. menardii ^vc . fimhriata

1

Pelagic
species.

Biloculina depressa
Miliolina seminulum
M. circularis
Planispirina celata
Rhabdammina discreta
Hyperamina ramosa
Bulimina marginata
JJtigerina oculata
Spheeroidina bulloides
Truncatulina rosea
T. lobatula
Pulcinulina pauperata

Bottom-
living
species.

P. micheliniana

Residue, 16.33 per cent, light brown, consists of Minerals [4.00], m. di. 0.08
mm., quartz, magnetite, felspar, hornblende, and a few glassy fragments. Sili-
ceous organisms'\j[ .QQi\ many Radiolarians, Sponge spicules, and Diatoms. Fine
icashings [5.33], light brown flocculent and argillaceous matter, with fine min-
eral particles and fragments of «''iceous organisms.

Station ^%. — l.2.t. 28° 47' 30" N. Long. 88° 41' 30" W. Depth, 533
fathoms. Surf. temp. 66°. Bot. temp. 4U°. Mud (river), of a light brown
color, dark with a greenish tinge when wet, showing Gasteropod shells imbed-
ded, very coherent, clayey streak, dries into very hard lumps.

Carbonate of Calcium, 6.43 per cent, consists of a few Gasteropod shells, cocco-
liths, and the following Foraminifera : —

Pullenia obliquiloculata

Globigerina inflata

G. conglobata

G. bulloides

G. dubia

G. rubra

Q. {Orbulina) universa, fragments.

Pulvinulina menardii
P. menardii, var. tumida
P. micheliniana
Miliolina seminulum
Bulimina marginata

MUSEUM OF COMPARATIVE ZOOLOGY. 61

Lagena gracillinia Pulvinulina elegans

Cristellaria gibba Sphceroidina bulloides

Uvigerina pygmcEa

Residue, 93.57 per cent, brown, consists oi Minerals [25.00], m. di. 0.05 mm.,
quartz, feldspars, hornblende, fragments of coal. Siliceous organisms [3.00],
fragments of Radiolarians. Fine washings [65.57], argillaceous matter and fine
mineral particles, with a few fine siliceous fragments.

In the examination and description of these deposits I was assisted by the
abbe Reuard, who determined many of the mineral particles. I have also to
acknowledge the services rendered by my assistants, Mr. James Chumley and
Mr. Frederick Pearcey.

John Murray.

]SJ"o. 3. — OhservatioTis on the Development of Agelena ncevia. —

By Wm. a. Locy.*

Several memoirs have been published on the development of the
Araneina, but the residts attained are still unsatisfactory on account of
the disagreement of authorities, and the limited extent to which the
method of sectioning has been employed in studying the subject. Up to
the present time only a single memoir, illustrated by figures of actual
sections, has appeared.

Valuable as were the works of the earlier writers, Herold ('24), Eathke
('42), and Von Wittich ('45 and '49), they now are principally of his-
torical importance, since their labors were performed either before the
announcement of the cell theory, or before it had gained general recog-
nition, and before embryology had attained its pre-eminence among mor-
phological studies.

Claparede ('62) made extended observations on the external features
of development, but did not discuss the preblastodermic period nor the
period of the revolution of the embryo.

Salensky ('71) published in Eussian a memoir, the figures illustrating
which show critical observations on the external features of development.
He was the first to figure the " rudimentary terga " of the period of revo-
lution, and also the development of the procephalic lobes.

In a short paper on the development of Pholcus, Emerton ('72) con-
fines his observations to the external features of development. He figures
the polygonal areas of the blastema, and erroneously concludes that they
are blastodermic cells without distinct nuclei. The relation of the primi-
tive cumulus to the ventral plate is well figured.

Balbiani ('73) has produced one of the most satisfactory memoirs yet
written ; he figures and describes in detail the external features of the
early stages of development up to the period of the formation of the
appendages.

Ludwig's ('76) observations were confined to the formation of the blas-
toderm, and are at variance with Balbiani's, mainly in denying the exist-

* Contributions from the Embrvological Laboratory of the Museum of Com-
parative Zoology at Harvard College, under the direction of E. L. Mark. No. VIII.

VOL. XII. — NO. 3.

64 BULLETIN OF THE

ence of the peripheral layer of protoplasm that is divided into polygonal
areas prior to the appearance of the blastoderm.

Barrois (78) added to what was already known an extended descrip-
tion, with figures, of his so-called limuloid stage, and gave notes, without
figures, on the development of the germinal layers.

Balfour ('80) was the first to produce figures of actual sections to illus-
trate the history of the germinal layers. Unfortunately, he had no
material for the preblastodermic period.

Sabatier ('81) contributes notes on the formation of the blastoderm,
and also on the yolk nucleus of spiders' eggs.

Schimkewitsch ('8-1) offers the latest contribution to the subject in a
preliminary notice in the Zoologischer Ameiger for August 18, 1884,
which embraces notes on the entire development.

I. -The Egg.

The eggs of Agelena 7i(evia are very abundant in the autumn. Those
for the present study were obtained near Cambridge, ^Mass., from Sep-
tember 15 to October 15. They exist in cocoons of white silk attached
to the underside of fence boards or loosened bark, and in other sheltered
places. This species, as well as others, continues to deposit eggs in
captivity, thus furnishing a ready means of obtaining freshly laid
material.

Treatment. — For observations on fresh material the long-used method
of immersing the eggs in oil is indispensable. The oil should be per-
fectly clear and scentless. In hardened eggs the external features can be
studied to great advantage by mounting in alcohol after they have been
shelled and stained ; the structures previously obscured by the chorion
thus become properly exposed. Before using this method I was unable
to trace the " rudimentary terga." Another valuable method for surface
study consists in clearing the already stained egg in clove oil. I have
found this especially appHcable in determining, by means of optical
sections, the thickness of the blastoderm on entire eggs.

In the important work of preparing eggs for cutting, experiments were
made with several reagents. The most satisfactory method of treatment
proved to be the very simple one already long in use. The eggs were
heated in water to about 80° C, and cooled slowly, after which they were
passed successively from weak to stronger grades of alcohol.

Good results were also obtained with Perenyi's fluid, which renders the
yolk less brittle, but at the same time changes somewhat its characteris-

MUSEUM OF COMPARATIVE ZOOLOGY. 65

tic appearance, and therefore can be used only in connection with other
methods.

Corrosive sublimate, either cold or hot, renders the eggs too brittle.

On account of the thickness of the chorion neither chromic acid nor
acid alcohol can be entirely extracted, and osmic acid will not penetrate.

Borax carmine (Grenacher's alcoholic) has proved to be, on the whole,
the best staining fluid. It is difficult to make any stain penetrate the
material of the later embryonic stages and those subsequent to hatch-
ing, on account of the development of the cuticula. This difficulty was
at length overcome by prolonged immersion in the staining fluid. In
some cases seventy-two hours were required to obtain an adequate stain.
Owing to the weak grade of alcohol used in making the stain, the eggs,
to prevent maceration, were left in the stain only twenty-four hours at a
time. They were then re-hardened, and after an interval immersed again
in the staining fluid.

The brittleness of the yolk of spiders' eggs constantly produces ci am-
bling of the sections. I have found that the yolk of eggs treated with
Perenyi's fluid may be cut satisfactorily ; in other cases I have used suc-
cessfully Mason's collodion method.*

Composition of the Egg, — The composition of the freshly deposited
egg has already been described with accuracy in most particulars by
Balbiani and others. In certain points, however, there has been neither
agreement in descriptions nor great accuracy.

To make clear the subsequent account I shall describe briefly the con-
stituent parts of the egg. It is enveloped by two membranes in contact
with each other. The outer, or so-called chorion, is tough and homo-
geneous, with its external surface covered by granules, which vary in
size and abundance in eggs of diff'erent species. In Agelena noevia they
are arranged in a single layer, and do not offer any serious impediment
to observations ; in some species (e. g. Epeira diadema), however, they
are several layers deep as well as very large, and must be removed to
allow accurate observations. On removing these granules the chorion
presents a finely punctate appearance, which is perhaps due, as Balbiani
has suggested, to the impressions left by the granules. This membrane,
unlike the chorion of insects, is added to the egg while i<" is passing
through the oviduct, and like the egg-sheU of Apus would fall into the
category of "secondary egg membranes," as defined by Ludwig ('74).

* See E. L. Mark, " Notes on Section Cutting," in the American Naturalist,
June, 1885, p. 628.

VOL. XII. — NO. 3.

5

66 BULLETIN OF THE

"Within the " chorion " is the structureless vitelline membrane which
closely invests the substance of the egg. It is thinner than the chorion,
from whicli it is easily separable after maceration. This membrane in-
vests mature eggs before they leave the ovarian follicle, and is doubtless
a product of the vitellus itself.

In making sections portions of these membranes were often cut. The
vitelline membrane stains faintly in Borax carmine ; the chorion retains
its layer of outer granules, which are not dissolved in alcohol. There is,
however, no trace of the areal arrangement of these granules, such as has
been figured by Ludwig ('76) for Philodromus limbatus.

The egg is composed of finely granular protoplasm, in which is accu-
mulated a large amount of nutritive material in the form of albuminoid
yolk corpuscles, and minute fixt globules. The albuminoid material is so
distributed as to give the protoplasm a characteristic arrangement. The
latter consists of a central mass enveloping the nucleus, a peripheral
layer, and a coarse network connecting the two.

The peripheral layer (couclie germinative of Balbiani) is the most strik-
ing feature in the arrangement of the protoplasm. It is in immediate
contact with the inner surface of the vitelline membrane, and is so
crowded with fat globules that Balbiani concluded erroneously that it is
composed exclusively of such globules.

The central mass of protoplasm forms around the nucleus an irregu-
larly limited, spheroidal envelope, containing neither yolk corpuscles nor
the fatty globules which are so characteristic of the peripheral layer. Its
outer portion is continuous with branching protoplasmic strands, which
form a coarse network around the yolk coi-puscles.

According to the observations of Balbiani the " yolk nucleus " persists
during a part, at least, of the embryonic development, and should there-
fore be mentioned as one of the constituents of the egg. There is also to
be included the perivitelline fluid, which makes its appearance during
the contraction of the vitellus. I have no positive information concern-
ing the source of this fluid, but having found no evidence of its existence
in a definite morphological condition before the contraction takes place,
I rest upon the assumption that up to this time it is uniformly distrib-
uted through the formative portion of the vitellus.

MUSEUM OF COMPARATR^E ZOOLOGY. 67

II. — The Embryo.

For convenience in describing the development of the embryo, the
following periods may be recognized : —

(1.) The preblastodermic period, in which are embraced the changes in the
mature egg up to and including the formation of the blastoderm.

(2.) The period from the completion of the blastoderm to the formation of the
rudimentary appendages, embracing, (a) the invagination, (6) the stage
of the primitive cumulus, (c) the formation of the ventral plate, and (d)
the division into protozonites.

(3.) The period from the appearance of the appendages to the reversion of the
embryo.

(4.) The period of reversion.

(5.) The period from the reversion to the hatching of the embryo.

1. Preblastodermic period. — The superficial and internal changes, al-
though going on simultaneously, may be more easily described if consid-
ered separately. The surface changes can be watched on the living egg,
and have been already thoroughly studied; but it is impossible to under-
stand fully these changes without that knowledge of the internal pheno-
mena which is to be obtained only by the aid of sections. The lack of
this method of study has led several previous observers into errors of
interpretation.

My earliest observations on the eggs of Agelena noevia were made a
few (probably not more than three or four) hours after their deposit. At
this time the polarity of the egg is very apparent ; one hemisphere is
characterized by small yolk corpuscles packed closely together, though
not joined in masses, and the other by agglomerations of larger yolk
corpuscles. The irregular masses thus formed are separated by spaces in
which are found smaller isolated corpuscles like those which distinguish
the opposite hemisphere.

Balbiani (73) was the first to give an adequate account of the surface
changes which occur during this period, in which the peripheral layer of
protoplasm is principally concerned. We shall see by following the his-
tory of this protoplasmic layer, that it is the equivalent of the blastema
observed in the eggs of many insects (Diptera, Phryganids, etc.), and
Crustacea (crab, etc.), and I shall so designate it hereafter.

Ludwig (76) and Barrois (78) have both called in question the accu-
racy of Balbiani's observations as to the peculiarities of the blastema.
My own observations are more in agreement with those of Balbiani,
which they serve in a measure to confirm.

68 BULLETIN OF THE

In freshly laid eggs this layer is in contact with the vitelline mem-
brane, but early becomes separated from it by the contraction of the
vitellus. The perivitelline fluid which makes its appearance during this
process is coagulable by heat and is also stainable. At first the contrac-
tion of the egg is uniform on all sides, but soon it takes place more
rapidly on one side, thus giving rise to a flattened surface (compare
Fig. 5), upon which the ventral plate is afterwards established. Through
the pressure of this contraction the blastema is moulded upon the periph-
eral yolk corpuscles into regions that correspond in position and size
with the underlying corpuscles. Owing to mutual pressure these regions
become regular hexagonal areas, (PI. V. figs. 24, 25,) and resemble the
subsequently formed cells of the blastoderm. The absence of nuclei is
the fundamental feature that at once distinguishes them from the blasto-
dermic cells, though they have frequently been mistaken for such on the
supposition that the nuclei were obscured.

The division of the blastema into areas as described above is a very
early phenomenon. At the time of my first observations a number of
faintly marked areas had already made their appearance at the more
active (animal) pole. At this time they could not be detected upon the
opposite hemisphere ; but after a short interval they also made their
appearance there in isolated patches ; finally they covered the entire sur-
face of the egg. At the outset the boundary lines of the areas are very
faint, but they become more distinct as the contraction of the vitellus
continues. In some places the yolk corpuscles become separated from
the blastema by a more rapid contraction of the interior protoplasm, and
then the polygonal areas in such regions remain only partially outlined
and incomplete, as described and figured by Balbiani ('73, Fig. 2).

After the areas are definitely formed the yolk corpuscles sometimes
shift their original positions, and thus cease to coincide with the areas,
since the latter do not at the same time undergo corresponding changes.

The next alteration in the surface makes its appearance only after the
lapse of a considerable interval (twelve to forty-eight hours) ; this led
Balbiani to assert, erroneously, that the egg is undergoing a period of
rest. Sections show on the contrary, that the interim is one of great
internal activity, during which repeated divisions of the nuclear sub-
stance lead to the formation of numerous cells which migrate towards the
periphery. The appearance of some of these at the surface marks
the beginning of new surface changes. The cells thus emerging from
the yolk constitute the primary blastoderm ; they first appear in the
interspaces between the yolk corpuscles, but often migrate afterwards to

MUSEUM OF COMPARATIVE ZOOLOGY. 69

positions directly over the yolk corpuscles. Each cell embraces a large,
clear, oval nucleus, which is surrounded by an irregularly radiating mass
of protoplasm.

The influence of these nuclei upon the protoplasm of the blastema
soon makes itself evident; a period of rearrangement supervenes in
which the boundaries of the polygonal areas described above are gradu-
ally effaced, and the protoplasm of the blastema, as M'ell as that which
accompanies the migrating nuclei, is grouped into new masses with the
nuclei as centres. The cells formed in this manner are at first large,
irregular, and very unequal in size (Fig. 26), but by repeated divisions
they become smaller, polygonal, and of more nearly uniform dimensions
(Fig. 27). They ultimately form a continuous layer — the blastoderm
— in the production of which the whole of the protoplasm of the blas-
tema has been employed.

I now turn to a consideration of the internal changes which accom-
pany the external features already described.

The structural and other peculiarities of the blastema in the eggs of
spiders have been subjects of considerable discussion, and therefore
deserve especial attention.

Balbiani (73) was the first to carefully study this layer,* and to
describe its division into areas.

Ludwig (76) denied its existence, and located the polygonal areas
described by Balbiani on the outside of the chorion, they being due, in
his opinion, to a peculiar arrangement of the granules covering the outer
surface of that membrane.

Barrois (78) admitted the existence of the blastema as a partial laj'^er,
but denied its division into areas ; the latter, according to his view, are
due to intersecting lines of granules located between the chorion and the
vitelline membrane.

Sabatier ('81) agrees substantially with Balbiani.

Thus the four observers who have discussed this topic have given
three irreconcilable explanations of the polygonal areas that Balbiani
referred to the peripheral layer of protoplasm.

Sections of eggs during this period afford decisive evidence on the
points under consideration. In the eggs of Agelena noevia, at least,
there can be doubt neither as to the existence of this layer, nor as to its
division into areas. Figure 28 is from a section of an egg containing
the first segmentation-nucleus {nl.), in which the blastema {bV.) is seen

* It had been mentioned by earlier writers, Rathke ('37), ClaparMe ('62), and
Emerton ('72), but they confounded it with the blastoderm.

70 BULLETIN OF THE

to be of considerable thickness, and to envelop closely the peripheral
layer of yolk corpuscles. An enlarged view of the same, given in
Figure 30, PL VI., shows the finely granular structure and the vesiculated
condition of the hardened blastema. It also shows how the blastema fits
over the yolk corpuscles, and dips down between them. It is to these
depressed regions that the boundary lines of the polygonal areas are due.
Figui'es 31-33 are enlarged views of separate yolk corpuscles with the
accompanying blastema.

The protoplasm of the blastema has a very characteristic appearance.
In addition to the common character of being very finely granular, the
protoplasm is throughout finely vesicular or spongy. The latter charac-
teristic is especially marked in eggs heated in water to coagulate the pro-
toplasm, and arises, I think, in the following manner. The fat globules
described as filling the protoplasm of the blastema in the fresh egg are
dissolved in the alcohol used for hardening purposes, and consequently
leave in the protoplasm spheroidal spaces of nearly uniform size, which
constitute the interstices. A discussion of the cause of the division of
the blastema into areas will be found under general considerations at the
end of the paper.

I have been unable, for the want of material, to trace the final changes
in the germinative vesicle. In the earliest condition of the deposited
egg that I have been able to procure there is a single central nucleus
(Figs. 28, 29 nl.)^ which is doubtless the descendant of the germina-
tive vesicle. This is the first segmentation-nucleus ; it is large, oval,
very finely granular, and surrounded by a spheroidal mass of protoplasm.
The latter is in immediate continuity with the network of protoi^lasm,
Avhich extends throughout the egg. The yolk corpuscles in the vicinity
of the protoplasm, which envelopes the nucleus, are much broken and
become successively smaller in approaching the nucleus, and at length
appear to merge into the finely granular protoplasm.

In the succi3eding stage the central nucleus divides into two of equal
size, which occupy a sub -central position (Fig. 34, nl, nl'). These nuclei
have essentially the same character as the one already described. Fig. 36,
from a two-cell stage of another egg, shows one of the nuclei with a cen-
tral vacuole [vL). The yolk is rudely divided at the same time, and
having been previously arranged in radiating branched columns {Deuto-
plasniasdulen of Ludwig), now forms two groups of such columns (Fig. 34).
It is probable that each of the two nuclei is divided into two others,
and that each resulting therefrom is similarly divided, but I have not
seen the four-cell stage. The next stage sectioned is one with eight

MUSEUM OF COMPARATIVE ZOOLOGY. 71

nuclei, all of which are nearer the centre than the surface of the egg. In
an egg still further advanced, containing at least thirty nuclei, none of
the cells have as yet emerged at the surface.

These internal cells are, however, continually migrating towards the
periphery, and, as might be expected from other evidences of the bipolar
condition of the egg, make their appearance first in that region which I
have already designated as the animal pole. The further history of these
cells after they have emerged at the surface has already been described
under the head of surface changes.

The problastodermic period, then, so far as I have been able to study
it, begins with the incomplete separation of the protoplasm into two
masses: one forming a thin layer at the surface — the "blastema" —
and the other concentrated around a nuclear structure inferred to be a
derivative of the germinative vesicle. The division of this nucleus is
accompanied by a corresponding division of the central mass of proto-
plasm ; a repetition of this process of division results in the formation of
a number of cells which, migrating to the surface, appropriate the contig-
uous portions of the blastema until the latter ceases to exist as a separate
layer ; there is no evidence that the nuclei of any of these cells arise in
any other way than by the repeated divisions of this single, central, first
segmentation-nucleus ; finally, the peripheral cells continue to subdivide
as well as to receive accessions from more tardily migrating elements until
a continuous single layer of cells — the blastoderm — envelops the egg.

2. The second period includes the changes from the formation of the
blastoderm to the appearance of rudimentary appendages.

In the eggs of Agelena juevia the blastoderm was established on the
third day of development, the temperature being about 23° C during the
day, and 19° to 20° C at night. Within certain limits* the tempera-
ture has a marked influence on the rapidity of the development, and one
can hasten or retard the growth by elevating or lowering the temperature.
For a day or two the blastodermic cells undergo rapid division, and are,
as a consequence, much reduced in size. There is a condition of the
blastoderm intermediate between those shown in Figs. 26 and 27, in
which the cells are regularly polygonal, but much larger than in Fig. 27.

My observations on the next surface change are not entirely satisfac-
tory, as I have seen it in only one instance. It appeared late on the
third day of development, and consisted of a depression at one pole simi-
lar to the depression in the surface of a peach at its stem end. This is

* The eggs are killed by a temperature higher than 30° C.

72 BULLETIN OF THE

probably the same pheuomenon that Salensky ('71) described as a pro-
cess of invagination, but to what extent it is comparable to a true in-
vagination I am not at present able to say. Although a direct connection
between this depression and the primitive cumulus has not yet been
traced, it is certain that in point of time the depression is the forerunner
of the cumulus, and the circumstantial evidence of their similar positions
on the egg indicates a connection between the two.

The external feature just spoken of as the primitive cumulus origi-
nates as a thickening of the blastoderm, at one end of the flattened sur-
face of the egg, and usually terminates in the production of a low conical
elevation. In surface aspect the cumulus is ovoid, with its more pointed
end directed towards the centre of the flattened surface, and it often
shows a tendency to elongate in that direction. This patch of cells being
rather opaque, appears whitish by reflected light, and dark by transmit-
ted light. In some specimens it is considerably elevated above the sur-
face of the egg, but in other cases it is only slightly or not at all raised.
Upon hardened eggs the surface of the cumulus is often depressed by a
median longitudinal furrow from which two or three smaller irregular
furrows radiate towards its margin (PI. I. fig. 4).

A second thickening, which I shall call the caudal thickening (c dn.
ca.), soon makes its appearance on the flattened surface of the egg, at a
distance of about 80° from the cumulus (Fig. 2.) It increases rapidly
in size, spreading out most in the direction of the cumulus, and ulti-
mately becomes shield-shaped. In the region between these two struc-
tures the ventral plate is gradually formed by a blastodermic thickening,
which is not at first continuous with the two terminal thickenings. In-
dications of the existence of a ventral-plate thickening, which appears
lighter by reflected light, are to be seen in the surface view shown in
Fig. 2, PI. I.

Immediately following the stage just described, the whole ventral
surface of the egg becomes divided by a series of transverse ridges and
furrows into protozonites (PI. II. fig. 6). I have not the material to
determine all the steps in the process, for the time involved in passing
from the stage of the primitive cumulus to the protozonite stage is a com-
paratively short one. The earliest condition in the latter stage which I
have examined shows three zonites and the cephalic plate. At this time
the latter is only faintly outlined. It is a broad thickening, rounded
towards the dorsal region of the egg, and fading into the protozonites on
the ventral surface. The caudal plate does not become visible until two
or three more zonites are established. It is similar in outline to the

MUSEUM OF COMPARATIVE ZOOLOGY. 73

cephalic plate. The addition of new zonites to those already existing
goes on raj)idly ; the two anterior ones (those of the chelicerae and the
pedipalpi) are cnt off from the posterior end of the cephalic plate. They
are late in making their appearance, and, as Balfour puts it, "lag behind"
the others in their development. The other zonites are developed from
the caudal plate.

Soon after the protozonites are first established they form ridges which
reach nearly around the egg, and thus appear to radiate from the dorsal
region. (Compare Emerton, '72, Figs. 8, 9.) They soon undergo con-
centration which so shortens the thickened ridges, that together they
form a band about 45° wide on the ventral surface of the egg — the
embryonic band. Fig. 6, PI. II., gives a side view of an egg in which
this concentration is well advanced but not yet completed. At the same
time the embryonic band increases in length, thus extending in an antero-
posterior direction further and further around the egg. When at length
seven or eight protozonites are fully established, the band embraces
approximately two-thirds of its circumference. At about this time also
the rudimentary appendages begin to appear ; these mark the commence-
ment of the third period of growth.

The internal condition of the egg during the second period can be
satisfactorily studied only by means of sections. I have made sections
passing through the primitive cumulus in two directions, sagittal and
transverse. In sagittal sections two features are conspicuous : (1) The
ventral surface of the egg is clearly differentiated from the dorsal surface
by the condition of the cells along its entire length (PI. VII. fig. 41).
(2) The cells in the region of the cumulus are arranged in several irregu-
lar layers. A thickening of the blastoderm has also arisen at the caudal
eminence, and there is a tendency to thicken along the ventral region
embraced between these two structures.

Figure 41 is from a sagittal section of the egg represented in PI. I,
Fig. 3 ; the cells of the ventral side are large and rounded or oval,
while those of the dorsal side are much flattened. The cells of the
primitive cumulus (cum. pr.') are conspicuous for their size ; they are
loosely arranged in layers. In some cases (PI. VI. fig. 39) they are four
layers deep.

Sections of eggs a little more advanced show a large number of cells
along the ventral-plate region, and also at the caudal thickening.

Balfour's figure (1. c, Fig. 11) of this stage cannot be compared criti-
cally with my own, as he was uncertain about the direction in which the
egg was cut ; but from its close resemblance to my sections, I think it

74 BULLETIN OF THE

safe to infer that he was wrong in supposing the larger accumulation of
cells shown in his figure to represent the caudal thickening rather than
the primitive cumulus. Fig. 39 is a transverse section through the
primitive cumulus in the region of its greater width.

In radial sections of the egg during this stage the cells of the unmodi-
fied blastoderm appear lens-shaped, the deep surface being more convex
than the outer, and contain each a single large nucleus, that is usually
central in position (Fig. 40). They are frequently preserved in the pro-
cess of division, their nuclei exhibiting the customary dumb-bell shaped
figure (Figs. 42, 44).

The " interzonal filaments " are quite persistent, remaining distinguish-
able even after the formation of the dividing cell wall (Fig. 44).

The nuclei in nearly all the sections which were stained in borax car-
mine are in a condition very favorable for study. The filaments of
chromatine are deeply stained, the nucleoplasm only faintly. The ar-
rangement of the chromatic substance in the nuclei varies from a condi-
tion in which it is concentrated into a ball at the centre of the nucleus
(Fig. 43), to one in which it forms a hollow shell near the surface of the
latter.

Sections during the protozonite stage show that the blastoderm of the
embryonic region consists of two distinct cell layers — the ectoderm and
the mesoderm (Figs. 49, 45). The cells of the outer layer (ectoderm)
are columnar (Fig. 45), and their nuclei, which are smaller than in pre-
vious stages, are very close together and much nearer the superficial
than the deep ends of the cells. The cells of the inner layer (meso-
derm) are not columnar but rounded cuboidal, and in general are much
less regularly arranged than the ectodermic cells; their nuclei, which
occupy the centres of the cells, do not at this stage present any other
characteristic differences from the nuclei of the ectoderm. At a later
period the nuclear elements of the mesoderm become spindle-shaped, and
thereby can be readily distinguished from those of the ectoderm. As
in the preceding stage, the cells of the non-embryonic or dorsal region
of the blastoderm are much flattened, even more than previously, and
only a single layer deep.

The cellular elements of the mesoderm are not everywhere definitely
arranged, and the deep margin of the layer especially is irregular in out-
line ; it partly envelops the yolk corpuscles, which are reduced to small
fragments on the surfaces adjacent to the protoplasm, but it does not at
this time form an uninterrupted layer.

The yolk corpuscles of this and succeeding stages are not absolutely

MUSEUM OF COMPARATIVE ZOOLOGY. 75

alike in constitution ; some of them are stained deeply and appear homo-
geneous, while others are stained lighter and appear granular. During
the whole of this period there continue to remain in the yolk mass a
large number of cells, which are distributed through its substance at
tolerably regular intervals. There is often a comparatively small amount
of protoplasm enveloping the large angular nuclei of these yolk-cells, and
about them the yolk corpuscles are more or less definitely grouped.

3. At the beginning of the third period the embryo still has a trans-
versely banded appearance as in the protozonite stage ; the concentration
from the sides is completed, and about six zonites are distinguishable
between the head- and tail-lobes. The zonites now begin to grow thin-
ner in the ventral median line, and at the same time their ends become
gradually more prominent and rounded. The small knob-like promi-
nences at the ends of the zonites are the rudiments of the appendages, and
in about two days after their first appearance (at the temperature stated)
the six cephalo-thoracic appendages are fully established as represented
in PI. II. fig. 7. The two anterior pairs of appendages are much
smaller than the four succeeding pairs, the latter being about equal in
size. The appendages thus established correspond to the chelicerse, the
pedipalpi, and the four pairs of ambulatory appendages of the adult.

Simultaneous with the growth of the appendages new zonites, derived
from the tail-lobe, make their first appearance ; the four anterior of
these are very prominent, and a little later they bear four pairs of pro-
visional appendages (PL IV. fig. 20, ^r. app.). In this first part of the
third period the head plate is faintly bilobed; the tail-lobe is broad and
rounded.

A ventral view (PI. IV. fig. 19) of the same egg (PI. TI. fig. 7)
shows a faint median furrow, which marks the thinning out of the ecto-
derm in the median plane after the separation of the lateral halves of the
underlying mesoderm. There are slight elevations just inside the bases
of the limbs, best seeu in optical section along the upper margin of the
figure ; they are the beginnings of the nervous ganglia.

At first the appendages grow out perpendicular to the axis of the
body (PL II. fig. 7), but as they increase in length they curve towards
the median line, as shown in Fig. 8. They are now indistinctly four-
jointed. The central lumen, which can be observed readily in optical
sections of the leg, is shown by actual sections to be a prolongation of
the cavity of the corresponding mesodermic somite.

At the present stage — the last part of the third period — the head
plate has become distinctly bilobed, a prominent upper lip composed of

76 BULLETIN OF THE

two lateral elements has been developed, and the stomodreum has be-
come faintly marked (PI. III. fig. 16, PI. IV. fig. 23). The four pairs
of provisional appendages are now fully established, and the embryo has
increased in length till the head- and tail-lobes are nearly in contact ;
the dorsal region is, as a consequence, much reduced. Behind the
somites which bear the provisional appendages the tail-lobe has given
rise to at least six indistinct additional somites ; the terminal end of the
tail is much narrowed and is becoming more pointed. The swellings
produced by the rudimentary ganglia, at the bases of the appendages, are
further developed, and the median ventral furrow has increased both in
depth and in width.

Balfour has given good figures and descriptions of the germinal layers
during the formation of the appendages. The mesoderm is of especial
interest at this time. Early in the protozonite stage it forms a con-
tinuous band, about as wide as the embryo, composed of a single layer of
cells extending the whole length of the embryonic band. About the
time the appendages begin to appear the mesoderm splits along the
median ventral line, thus forming two parallel bands, which remain
united, however, in the head and tail regions. The division of the
mesoderm into lateral halves is followed by an increase in the tliickness
of the resulting bands, each of which becomes split into a somatic and a
splanchnic layer. It is also at this stage that the mesoderm is divided by
transverse constructions into somites, each of Avhich contains a central
lumen. I am unable to determine from my specimens whether its divis-
ion into successive blocks precedes or fullows the appearance of the
lumen. In the growth of the appendages the somatic layer of the meso-
derm accompanies the outgrowing ectoderm, and forms a continuous
lining to its cavity.

During this period the ectoderm has also increased in thickness, but
along the median ventral line it remains thinner ; from this it results
that there are two bands of thickened ectoderm corresponding to the two
deep bands of mesoderm. The ventral median depression previously
mentioned is at first due to the relative thinness of the ectoderm in this
region ; it is afterwards made more conspicuous by the further separa-
tion of the mesodermic bands. From the ectodermic bands are formed
the nervous ganglia. They are developed first in the thoracic region in
the form of swellings at the bases of the appendages, but by the time the
stage represented in Fig. 8 has been reached, they have also been formed
in the abdominal region.

As already correctly maintained by Balfour, the segment of the cheli-

MUSEUM OF COMPARATIVE ZOOLOGY. 77

cerse has a separate pair of ganglia which ultimately disappear, serving
only to aid in the formation cf the circuraoesophageal commissure. At
an early stage, then, the nervous elements consist of two rows of ganglia,
a pair of ganglia for each somite, which are widely separated except in the
head lobe and the tail lobe, where they are continuous in the median
line.

Another important growth on the part of the ectoderm leads to the
formation of the stomodseum, which arises as a simple tubular infolding
between the ganglionic thickenings of the cheliceral somites, and imme-
diately below the ventral margin of the cephalic plate. It becomes ex-
panded at its deep end into a sort of pocket, but it has only a small
external opening. The walls of the stomodaeum are composed of cells,
two or three rows deep, which are elongated and somewhat wedge-shaped
rather than distinctly columnar.

4. The period of reversion is marked by the origin of^ many important
organs : proctodaeum, heart, lungs, tracheae, spinning glands, muscles,
etc. The embryo undergoes great changes in external form, gradually
passing from the condition represented in PI. II. fig. 8, where the ven-
tral surface of the embryo is uniformly convex, and occupies an arc of
about 300°, to a form (PI. II. fig. 11) in which the ventral surface is
folded upon itself.

As a prelude to reversion the tail-lobe of the embryo becomes promi-
nent, being raised from the surface of the egg. The early steps in the
process of reversion will be best understood from the examination of a
series of dorsal views. Fig. 13 (PI. III.) presents the dorsal aspect at
the beginning of reversion, and Fig. 8 (PI. II.) a side view at nearly the
same stage. The tail-lobe has lost its broad rounded character, and is
being changed into a more distinctively caudate structure. It still re-
mains nearly in contact with the cephalic lobe. Tlie dorsal element?
("terga" of Barrois) have begun their upward growth, and appear in
the figure as four pairs of prominent lateral elevations. A corres])onding
growth of the abdominal segments is also in progress ; the dorsal elements
growing upward finally meet in the median line of the back.

Each of the lobes of the procephalic plate has a semilunar form, and is
composed of a central area, apparently separated from a marginal rim by
means of a deep fold (PI. IV. fig. 23). The prominent upper lip (/r.)
is apparently an outgrowth of the ventral border of the cephalic plate,
and overhangs the entrance to the stomodreum (sd.).

The chelicerte (1 ap]}.) and the pedipalpi (2 «/?/>.) both appear as post-
oral structures. The prominent ganglia (gji.) belunging to the cheli-

78 BULLETIN OF THE

ceral segment lie just in front of the bases of the chelicerse, and are like-
wise post-oral structures, as claimed by Balfour.

The next stage in the process is represented in Pi. III. fig. 14, in
which the tail-lobe is much narrower and more clearly circumscribed ; a
considerable interval now separates it from the procephalic plate. Five
pairs of dorsal (tergal) elements belonging to the abdominal segments
are now visible ; the four anterior pairs belong to the segments bearing
provisional appendages, and a fifth, smaller pair, has been interpolated
between these and the tail-lobe. The tail-lobe is apparently split in the
median line into two bands that, in passing forwards, diverge rapidly.
These are the two bands of ectoderm which, as before mentioned, join
each other in the head and the tail-lobes. Between these divergent
bands of ectoderm is to be seen a part of the yolk mass covered by only
a thin layer of ectoderm. The legs have increased in length until they
nearly meet in the median plane (PI, IV. fig. 22).

In the next stage (PI. III. fig. 15) the dorsal region is much elon-
gated owing to the retrogression of the tail-lobe, and the rudimentary
terga extend much further dorsad. Up to this time the only dorsal ele-
ments developed were the five pairs belonging to the abdominal somites,
but during this stage the dorsal elements of the limb-bearing somites
begin a more rapid growth. The dorsal elements of the somite bearing
the fourth pair of legs grow much more rapidly than the others.

In a dorsal view of a somewhat later stage (PI. III. fig. 16) the tip of
the tail is just visible at the posterior margin of the embryo, the dorsal
region having increased proportionately in extent. The procephalic lobes
are closing together in the median plane. The dorsal elements of the
somites now nearly meet in the median line of the back. In the figure
some of the provisional appendages {pr. app.) are visible along the sides
of the body.

In a slightly older embryo (PI. III. fig. 17) the tail-lobe is no longer
visible from above ; the cephalic lobes have become fused, and the dorsal
elements of the somites have met in the median line. Along this line a
narrow slightly elevated ridge indicates externally the position of the
heart.

The much reduced caudal lobe is to be seen from below (PI. IV.
fig. 21) and, diverging from it in two lines, the provisional appendages.
Owing to the wide separation of the neural bands the legs of each pair
are far apart. Between them a part of the yolk (not the whole, as stated
by Barrois) protrudes, forming a sort of ventral yolk sack. The rapid
appropriation of this store of yolk causes the disappearance of the sack ;

MUSEUM OF COMPARATIVE ZOOLOGY. 79

the embryo becomes more folded upon itself ventrally, as shown in
PI. II. fig. 10, and the legs, increasing in length, gradually approach and
finally overlap each other in the median line. The embryo has no\y
acquired a strong ventral flexure — the reversion is completed.

During this period the bases of the cheliceras in growing have moved
forwards and met in the median plane, so that they appear as pre-oral
appendages. There has also appeared between their bases a prominent
outgrowth to form the rostrum.

Balfour ('80, p. 180) endeavors to account for the process of rever-
sion as the result of a rapid " elongation of the dorsal region, that is, the
region on the dorsal surface between the anal and the procephalic lobes."
I understand by this that it is to the growth of the ectodermic cells of
the dorsal region that he would ascribe the elongation of the dorsal surface.
I shall endeavor to show presently that this explanation is not sufficient
to account for the changes which actually take place during reversion.

The growth of the derivatives from the ectoderm during the period
under consideration is very great. At the beginning of the period the
stomodseum forms a pocket-shaped invagination with a small external
opening. Its calibre diminishes, except at its anterior end ; it continues
to grow inwardly, and at length forms an arched tubular organ, with its
free end directed backward, and projecting some distance into the yolk.
Near the close of the period its deep end becomes somewhat enlarged to
form the rudiment of the sucking stomach. To the latter are attached a
vertical muscle (mu. vrt. PI. IX. fig. 62) extending to the dorsal wall of
the embryo, and two lateral muscles (mu. lat.).

The proctodajum is a later formation, which makes its appearance as an
infolding at the tip of the tail-lobe some time after the beginning of thi's
period. The relation of the tail-lobe to the rest of the body is best ap-
preciated from sections, since it is not always evident from surface views
that there is a deep fold which serves to separate it from the underlying
portion of the dorsal surface. The prominence which it attains and the
changes which it undergoes are readily traceable in a series of figures
from successive stages during reversion (PI. VIII. figs. 50-54). The
strong resemblance of the condition shown in Fig. 50 to that which
Bobretzky ('74, Fig. 15) has figured for Oniscus at an apparently similar
stage of development, misled me into the supposition that I should find
the proctodaeum of Agelena developing in the manner described by him
for Oniscus. But such is not the case. In Agelena the tip of the lobe
is the tip of the tail — the morphological end of tlie body, and the
depression which separates this lobe from the neighboring portion of the

80 BULLETIN OF THE

embryo is not the proctodseum, but simply a fold in the dorsal wall of
the embryo. The pocket resulting from this fold is flattened in a plane
perpendicular to the sagittal plane, and is not a tubular infolding like
the real proctodajum. This pocket is lined with ectodermic cells, which
subsequently form a part of the epidermis at the posterior end of the
dorsum. By the traction exerted along the median ventral line of the
body during reversion the tail is drawn downwards and greatly short-
ened, thus obliterating the pocket. When in the progress of its rever-
sion the embryo has reached about the stage represented in Figs. 10,
16, the proctodeum is formed as an invagination just ventral to its tip.
At this early period it has the appearance shown in the sagittal section,
n. VIII. fig. 54. The tail-lobe is now a short thick prominence, and
the dorsal fold has nearly disappeared.

At an early period the proctodaeum is enlarged by the outgrowth of
its dorsal wall into the form of a capacious pocket, which is retained by
the embryo throughout its development. This diverticulum (br. stc.
PI. VIII. figs. 55, 56) is the so-called stercoral pocket of the adult. The
walls of the rectum and the stercoral pocket are composed of columnar
epithelium, and are closely invested by mesodermic elements.

The nervous system is characterized during this period by the wide
separation of the nerve bands and a gradual concentration of their sul>
stance headwards. The distance between the bands is greatly increased
by the passage of the yolk from the dorsal to the ventral side through
the aperture left by their separation. At the period of their great-
est separation they occupy curved lines along the lateral walls of the
yolk sack, separated from each other by its diameter. During reversion
also the actual length of the nerve cords is somewhat decreased. At the
beginning of this period they reach nearly around the egg from the head-
to the tail-lobe (PI. XII. fig. 77), but during reversion they pass through
the stages of shortening represented in Pis. XI. XII. figs. 72, 71, 70, 78.
Their connection with the tail-lobe is severed, and the nerve cords grad-
ually move forwards ; with the absorption of the yolk mass this lateral
separation is diminished until they are in contact along the ventral line.

After the process of reversion is well advanced certain cells in the
bases of the chelicerse become conspicuous from their enlarged condition
and spongy appearance, which serve to distinguish them sharply from
surrounding cells. They are the rudiments of the poison glands, and
although I have not been able to trace an external outlet until a later
period, it is probable that these cells are derived from an infolding of the
ectoderm at the point where later an outlet is discernible.

MUSEUM OF COMPARATIVE ZOOLOGY. 81

The spinning glands are not yet definitely established, but in the anal
region on the ventral side of the proctod?eum there is a large accumula-
tion of ectodermic cells (PI. XI. fig. 70) from which they are subse-
quently developed.

Late in this period the infoldings for the lungs arise. There appear a
pair of large oval masses of cells, the nuclei of which are arranged in par-
allel lines (PI. XI. fig. 73). From these cells the respiratory lamelLne of
the lungs are finally formed.

The mesoderm likewise has been growing rapidly during reversion.
In the previous period it was confined to the ventral portion of the em-
bryo, but during the present period it grows upward on either side until
it reaches the dorsal median line, thus forming a continuous layer be-
neath the ectoderm, as well as an investment for all organs, which -arise
as outgrowths of either ectoderm or entoderm.

The dorsal growth of the rudimentary terga, already spoken of as
external features, is followed by this underlying layer of mesoderm.
Early in the formation of the dorsal elements tliis niesodermic layer is
divided into corresponding somites. Balfour ('80, p. 181) concluded
that the cells out of which are formed the dorsal somites of the meso-
derm " are not derived from prolongations of the somatic and splanchnic
layers of the already formed [ventral] somites, but are new formations
derived from the yolk." My sections, however, indicate that there is a
direct continuity between the two (PI. IX. figs. 59, 61), and that the
dorsal mesoderm is an outgrowth from the previously established ventral
niesodermic somites.

It is during this period also that the heart is formed. "While I have
been unable to arrive at an entirely satisfactory understanding of the
details of its formation, I am convinced that it is not, as Balfour states,
developed from a solid cord of cells, but from the dorsal limb of the up-
growing mesoderm, and tliat its dorsal wall is closed first, while the
ventral wall — the floor — remains for a time widely o])en below, thus
communicating freely with the yolk. My sections also sliow that at a
later period the aorta is formed, by means of a Ci^nstriction, from the
mesenteron. This agrees with the recent observations of Schimkewitsch,
('84").

A layer of characteristic cells, to which Balfour alludes in speaking of
the formation uf the dorsal mesoblast, precedes the formation of the
heart in tlic dorsal region. These are what have been called " primary
entoderm " cells, and are sharply distinguished from the surrounding cells
by their large size, their large, oval nuclei, and their yellowish tint. Tliese

VOL. XII. — NO. 3. G

82 BULLETIN OF THE

cells are derivatives from the yolk-cells, and first appear just before the
reversion of the embryo begins. They are abundant along tlie sides of
the body, and about the oesaphagus as well as in the dorsal region.

The yolk during this period is somewhat changed from its early char-
acteristics. The corpuscles are undergoing disintegration, and are much
vacuolated, which gives them in certain regions a spongy appearance.
The nuclei of the yolk-cells, while they have increased in number, are
smaller and angular (often triangular) in outline.

5. The 2^eriod from reversion to hatching. — Few surface changes of
importance arc necessary to convert the embryo of the period just de-
scribed into the adult. The following are the most obvious : The embryo
becomes more closely flexed upon itself (PL II. fig. 11), and the constric-
tion which separates the abdomen and the cephalo-thorax is formed. At
least two pairs of provisional appendages are modified into as many
large spinning mammilla?.* In addition to these two large pairs there
is a pair of .smaller median mammilhe, the origin of which I have not
traced. The remnant of the tail persists for some time as a post-anal
knob ; ui)on the ventral surface appear the infoldings, from wliich are
formed the trachefB, and also those of the generative organs ; upon the
head the eyes make their appearance. Two or three days before hatch-
ing the embryo begins to unroll, and undergoes a moult ; at the time of
hatching it is completely straightened.

I shall now proceed, after this general account of the more important
embryonic stages, to the consideration of the development of separate
organs and sets of organs.

III. — Organogeny.

In the present paper only the following organs will receive attention :
(1) the alimentary tract, including stomod;x;um, pharynx, stomach, mid-
intestine, stercoral pocket and rectum; (2) the eyes; and (3) the lungs.

* Balfour ('80, p. 183) lias stated: "The four rudimentary appendages have
disappeared, unless, which seems to me in the highest degree improbable, they
remain as the spinning mammilla;." Notwithstanding his doubt, I think I have
traced the development of two pairs directly into the mammillae. The mammilla?,
thorefori', are appendages of abdominal somites, homodynamic with the cephalo-
thoracic appendages, and there are consequently six somites condensed into tlie
space between the posterior pair of mammillce and the anus. Upon the ventral face
the evidences of this are early obliterated, but upon the dorsal surface the poste-
rior somites are recognizable by the arrangement of the longitudinal muscles, at
least as late as the stage represented in Fig. 70, PI. XL

MUSEUM OF COxMPARATIVE ZOOLOGY. 83

The portion of the alimentary canal first to appear — the stomodseum
— arises as an invagination of ectoderm just before the beginning of
the third period of development, and therefore after the establishment
of a mesodermic layer in the region in which the invagination occurs.
As already stated, it grows rapidly during the third period, and in the
fourth period it acquires certain muscular attachments, developed out
of mesodermic cells. After the reversion of the embryo is completed,
a tube of about the same calibre as the stomodsum arises behind the
stomach, and, extending through the cephalothorax, opens widely, by
a bell-shaped expansion, into the yolk of the abdomen (PI. XII. fig.
78). This post-gastric portion of the canal .was evidently overlooked by
Balfour, as he (1. c, p. 187) states that he was unable to find "any trace
of an anterior part of the mesenteron adjoining the stomodseum." An-
teriorly it apparently does not open into the sucking stomach during
embryonic stages, but is so plugged with cells that its relations are
obscured.

At the time of hatching the intestinal tract is still incomplete, the
epithelial wall of the mesenteron being largely or altogether w^anting.
There may be distinguished in the anterior portion of the tract the fol-
lowing parts : pharynx, oesophagus, sucking stomach, and post-gastric
tube.

The pharynx passes from the mouth obliquely upwards and backwards,
and, turning at nearly a right angle, is continued into the oesophagus.
The latter is of uniform calibre and extends backwards with a slightly
downward curve, terminating in the enlarged sucking stomach. A
muscle arising from the dorsal wall of the cephalothorax just in front of
the anterior margin of the brain, is inserted at the angle of the pharynx
(PL XL fig. 10, mu.). From the sucking stomach three distinct muscles
extend to the body wall : a vertical muscle {imt. vrt.), lying in the sagittal
plane and extending downward from the cephalothorax a little behind
the brain to be inserted along the dorsal wall of the stomach ; a pair of
lateral muscles (mu. lat.), which arise from the sternal plate and ascend
obliquely towards the sagittal plane to be inserted into the lower half of
the lateral walls of the stomach (PI. IX. fig. 62 ; PI. XI. fig. 70). A
few fibres arising with the vertical muscle join the fibres of the lateral
muscles without having a distinct attachment to the stomach (Fig. 62).

I am in doubt concerning the origin of the post-gastric tube already
alluded to. Its anterior end, which lies just beneath the stomach
(PI. XII. fig. 78), is rounded and plugged with cells, and I have been
able to trace an enveloping layer of mesodermic elements nearly across

84 BULLETIN OF THE

its anterior surface. These facts afford strong evidence that this portion
of the alimentary tract is derived from the entoderm rather than from the
stomodseal infolding of the ectoderm. The cellular elements of which it is
composed do not, however, differ enough from those of the stomodseum
to add anything to the reasons just given for supposing an entodermic
origin. But if, as I believe, this is not an outgrowth of the storaodgeum,
it must be the first-formed portion of the mesenteron, the walls of which,
as we shall subsequently see, are begun at both ends and completed by
the gradual advance and ultimate meeting of the two separate formations.

On each side of the stomach are given off cseca, which extend into the
bases of the limbs. The cellular elements composing the walls of these
tubes are flattened.

The walls of the anterior or stomodseal portion of the alimentary canal
are composed of three layers : the cuticular, the epitlielial, and the peri-
toneal. The pharynx, the oesophagus, and the sucking stomach are all
lined with a cuticular layer which is continuous at the mouth with the
cuticular covering of the body. In the pharynx it is thickened and cor-
rugated by tootli-like projections, but in the oesophagus and the stomach
it is much thinner and not roughened. I have not been able to make
out satisfactorily wliether this layer extends into the post-gastric portion
or not. If it does, this would be an argument in favor of the ectodermic
origin of this portion of the canal.

The posterior part of the alimentary canal — the proctodaeum — does
not begin until the reversit)n of the embryo is well advanced. Its exter-
nal orifice is minute and leads directly into an expanded portion, which
becomes the stercoral pocket. This enlargement is present at an early
stage of the invagination, and presents in sagittal section a triangular
outline (PI. VIII. fig. 54). Its walls are at this time thick and com-
posed of large ectodermic cells, which are, however, only a single layer
deep. The invagination forming the proctodaeum pushes before it an
enveloping layer of the already formed mesoderm. The invagination is
gradually differentiated into two parts : a straight narrow tube (the ter-
minal portion of the rectum), and the stercoral pocket. The wall of the
proctodaeum is composed of columnar epithehum, the large spherical
nuclei of which are placed close to the inner ends of the cells, which
almost meet, and thereby nearly obliterate the lumen of the tube. This
epithelium is enveloped externally by the usual layer of flattened meso-
dermic cells. The stercoral pocket increases rapidly in size, and becomes
pear-shaped in outline. The cellular elements of its walls change from a
columnar to a flattened epithelium. At the time of hatching its wall is

MUSEUM OF COMPARATIVE ZOOLOGY. 85

composed of tKree layers : an internal epithelial layer, composed of the
flattened ectodermic cells with oval nuclei ; a middle layer, composed of
very much flattened cells, recognizable only by their long very narrow
nuclei ; and an external covering of mesodermic elements (PI. IX.
fig. 57).

A few days before hatching a short tube is to be seen extending for-
wards from the ventral face of the stercoral pocket to which it is joined.
It is continuous behind with the part of the proctodaeum which I have
called the rectum, and in front it spreads out into a trumpet-shaped
expansion which embraces the posterior portion of the yolk mass
(PI. VIII. figs. 55, 56 ; PI. XI. fig. 70). The cells composing the epi-
thelial lining of this portion of the intestine resemble more those of the
stercoral pocket than those of the rectum. They are invested externally
by a layer of flattened mesodermic elements continuous behind with those
which envelop the stercoral pocket and the rectum, and in front with
the mesodermic layer which invests the yolk.

Schimkewitsch claims for the hind part of the alimentary canal in
Epeira a very thin cuticular lining ; I have not been able to demonstrate
its presence in sections of Agelena.

From the dorsal wall of the pre-stercoral tube — just where it becomes
confluent with the antero-inferior face of the stercoral pocket — the two
malpighian tubes take their origin. The position of these tubes furnishes
the only evidence that I have concerning the source of the pre-stercoral
tube, and leads to the conclusion that it is of entodermic origin, and
therefore a part of the mesenteron.

I have not as yet traced the alimentary canal to its adult condition.
At the time of hatching it is still incomplete, being composed of an
anterior and a posterior portion, the inner extremities of which open
towards each other by wide expansions, which abut directly upon the
yolk. In the latest developmental condition that I have examined ■•—
about eight or ten days after hatching — the mesenteron is greatly ex-
tended, and appears to be continuous at its sides with the yolk compart-
ments of the abdomen. It doubtless is functionally active previous to
this time, since there was a considerable amount of effete matter in the
stercoral pocket.

2. The Eyes. — It seems somewhat remarkable that up to the present
time the development of the sense organs in the Araneina has been
hardly more than touched upon. The meagre description by Claparede *

* Claparede ('62, pp. 56, 67) is the only one, T believe, who has written anything
about the development of the eyes, and he has given only an account of the exter-

86 BULLETIN OF THE

of the external features in the development of the eyes is, I believe, all
that has heen published upon that subject. I have been able, by means
of sections, to trace the formation of the eyes, which begins at a compara-
tively late stage of development, through the most important changes.
I shall confine my descriptions at first to the median anterior pair, which
differ in some important respects from the remaining three pairs. The
first step in their formation consists in local thickenings of the "hypo-
dermis" (ectoderm) in the frontal region. Each thickening at first
causes the deep surface of the hypodermis to bulge, while the outer sur-
face retains its original direction. The thickening is soon followed by
an extensive invagination, which begins just in front of the thickened

nal appearances in the region where the eyes are developed. Relative to Pholcus,
he says (1. c, p. 50) : "'Les yeux n'apparaissent qu'a la fin de la vie embryonnaire,
plus tard dans tous les cas que chez les Aoarides, si j'en juge par lea observations
de Mr. Van Beneden sur V Atax Ypsilophora. Leur apparition est prcce'dee par la
formation de quatre pelita sillons que j'appellerai les sillons opthalmiques. Les
quatre sillons sont disposes par paires, deux d'entre eux appartenant au cote droit
et deux au cote' gauclie. Ce sont de petites depressions transversales, arquees, dont
la convexitd est dorsale. Les deux sillons de chaque cote vont en divergeant du
cote' externe. lis repondent a la double rangee de yeux des Pholques. Avant
que les yeux eux-memes apparaissent sous la forme de petits globes dans les sillons
opthalmiques, ceux-ci se colorent par le depot d'une petite quantite de pigment
(v. fig. 25 en o)."

With regard to the formation of the eyes in Lycosa, he adds, (1. c, pp. 67, 68) :
"De meme que chez les Pholques, nous voyons chez les Lycoses, les yeux n'ap-
paraitre que fort tard. Les pieds et les palpcs ont de'ja une grande partie de leurs
articulations, lorsque les sillons ophthalmiques, au nombre de six, formant deux
groupes symme'triques de trois, se raontrent comme pre'curseurs des yeux. Ces
sillons ne tardent pas a se colorer par le de'pot d'un pigment sombre (fig. 45, PI. V,
en o). Plus tard le pigment, qui devient d'un noir rougeatre, se groupe dans les
sillons ophthalmiques en masses distinctes qui font le'gerement saillie a la surface
de la tete. De chaque cote' de la tete deux de ces amas de pigment plus petits que
les autres appartiennent au sillon inferieur, un appartient au sillon median, un au
sillon supc'rieur. Les huit yeux de I'araigne'e se montrent done dans I'origine sous
la forme de simples amas pigmentaires. Toutefois a I'epoque oii la Lycose quitte
I'ceuf, on aper(;oit dcja au sein de chacun de ces amas de pigment un corps re'frin-
gent, le cristallin, sur la formation duquel je n'ai rien pu constater de pre'cis. Ces
yeux offrent alors la disposition indique'e dans la figure 50 (PI. VI), disposition qui
s'eloigne encore notablement de celle de I'adulte (v. fig. 51). Les quatre yeux de
la range'e inf^rieure sont, au moment de la naissance, beaucoup plus petits que les
autres. lis le restent d'ailleurs toute la vie durant. Ceux de la ligne me'diane
sont les plus gros. On reconnait facilenient, tout au moins pour les quatre gros
yeux, que chaque globe oculaire est pyriforme, se terminant en une pointe qui
regarde I'arriere. Sans doute cette pointe n'est que I'extre'mite' pe'riphe'rique du
nerf optique, dont je n'ai pu cependant suivre le cours jusqu'au centre nerveux."

MUSEUM OF COMPARATIVE ZOOLOGY. 87

area. By this process of infolding the whole region of the thickening
becomes inverted, and tinalh' lies under that portion of the still unmodi-
fied hypodermis, which was at first just behind the thickening (PL X.
figs. 63, 64), so that what was its external surface becomes its deep sur-
face, and what was the deep surface lies relatively nearer the exterior.
As a result of this infolding the region of the eye at this stage is com-
posed of three distinct layers : an external (Figs. 64, 66, h d), a middle
(rtn'.) and internal (rt/i".) layer. The external portion is composed of a
single layer of cells, which at the outset do not differ from the hypoder-
mis cells, with which they are continuous. The middle layer is the orig-
inally thickened portion, and is composed of elongated cells, all having
the same general inclination ; their nuclei are oval, being elongated in
the direction of the long axis of the cells, and are arranged in three or
four superimposed irregular rows. The internal layer, like the outer
one, is composed of a single row of cells.

The pocket of the invagination is not very broad, as is best to be seen
on frontal sections (PI. X. fig. 65). After a time the orifice of the inva-
gination is closed by a fusion of its lips, and the retinal bulb, formed by
the middle ajid internal layers, becomes separated from the hypodermis ;
thus all direct evidence of its mode of origin is obliterated.

After the invagination is completed the cells of the external layer
begin to elongate ; they are now so closely crowded together that their
nuclei are almost in contact, and the cell boundaries are not easily dis-
tinguishable. Their nuclei also become lengthened, without losing much
in thickness, until they are three or four times as long as broad. An
accumulation of homogeneous faintly stainable substance appears be-
tween the thin cuticula, which everywhere covers the hypodermis, and
the free ends of these elongated cells ; it is not at first sharply defined
from the latter (PI. X. fig. 66). This accumulation of substance finally
causes an elevation of the surface, and just before hatching it has as-
sumed a tolerably lenticular shape (PI. X. fig. 68, Ins.). It gradually
l)ecomes more refractive, and a few days after hatching assumes nearly
the form of the cuticular lens of the adult (PI. X. fig. 69, Ins.). The
lens is evidently produced by the secretive activity of the underlying
elongated cells ; these have, in the meantime, continued to elongate, and
their boundaries have become sharply defined ; during this period the
nuclei do not seem to share in the process of elongation, for after the
formation of the lens they are seen to occupy the deep ends of the cells,
and to be only a little larger than the nuclei of the adjacent hypodermis
(PI. X. fig. 69, k d). This layer of cells now constitutes the so-called

88 BULLETIX OF THE

vitreous body of the eye, and is, as just shown, a modified portion of the
hypodermis, with Avhich it lias never ceased to be continuous.

The cells of the middle (inverted) layer undoubtedly form all of the
retinal elements. I am unable to assert positively what becomes of the
" inner layer." Certain stages show that the nuclei of this layer have
become considerably flattened in the direction of radii to the optic bulb,
so thai it is possible they ultimately constitute a kind of enveloping tunic
to the deep surface of the bulb. That I have not mistaken mesodermic
elements for this posterior layer, is evident from the great size of the
nuclei, and the successive stages exhibited in the conditions of the layer.
It, therefore, seems to me at present doubtful if the inner layer really
shares in the formation of the retina proper. After the comjjletion of
the infolding the cells of the middle layer elongate and the layer thus
becomes thicker, especially its central portion, and the whole invaginated
mass, therefore, assumes a more nearly spherical form. This elongation
of the cells appears to result from an outgrowth of their anterior ends,
since the nuclei are crowded into the deeper portions of the layer, while
the anterior part becomes translucent. The formation of the bacilli in
the extreme anterior ends of the cells soon makes this part of the layer
more highly refractive. The bacilli increase in length from before back-
wards. A little later a zone of pigment granules makes its appearance in
the retinal cells between the forming bacilli and the more anterior of the
nuclei. These granules are limited to near the surfaces of the cells.
About the same time a layer of pigment cells is observable along the
deep surface of the eye-bulb. They are probably the cells of the "inner
layer" of the involution, for they do not appear to be elongated in the
direction of the axis of the bidb, as are the nuclei of the true retinal or
nerve-end cells. As the pigmentation increases it appears in some of the
retinal cells in a position posterior to the prenuclear zone above men-
tioned ; these patches of pigment seem to correspond in height with the
elongated nuclei of the cells, and are not always continuous with the
prenuclear zone of pigment.

It is probable, therefore, that not only the posterior ends of the bacilli
are jiractically ensheathed by a layer of pigment, but also that the nuclei
of the nerve-end cells become more or less enveloped in pigment, and
that the two zones are not at first continuous with each other, nor with
the still more posterior pigmentation of the inner layer. Tlie pigment in
all cases belongs to cells of the originally involuted ectoderm, and there
are no interstitial cells between the nerve-end cells of the retina ; at least
satisfactorily stained specimens show only nuclei of two sorts : namely.

MUSEUM OF COMPAKATIVE ZOOLOGY. 89

the large elongated nuclei of the nerve-end cells and the flattened nuclei
of the inner layer. In this stage (PI. X. fig. 69) the essential features of
the eye are established, and it is possible to affirm positively that the
anterior median eyes in Agelena ncevia belong to the type in which the
nuclei of the retinal cells are post-hadllar.

The three remaining pairs of eyes originate somewhat later, but in sub-
stantially the same way as the pair just described ; a hypodermic thick-
ening, a backward directed infolding which inverts the thickened region
and carries in beneath it a thin layer of hypodermis, the closure of the
orifice of involution, and the detachment of the involuted mass from the
hypoderm. The lens is also produced from modified hypodermic cells
resembling, though shorter than, those forming the lens of the median
anterior pair. But the two layers of the infolded mass do not undergo
the same changes as do the corresponding layers in the pair of eyes pre-
viously described. In the first place, the two layers remain permanently
(up to my latest stage, ten days after hatching) separated by the devel-
opment of a (in hardened specimens) much folded chitinous layer, which
is probably homologous with the cuticular covering of the body, with
which in the earlier stages it appears to be continuous. Secondly, while
the retina is developed as in the anterior eyes, from the cells of tlie
inverted portion of the infolded region, the bacilli do not arise in the
ends of the cells which adjoin the vitreous body, but at the opposite or
posterior ends. They are, therefore, found in the immediate vicinity of
the chitinous substance. The nuclei, in the latest stages examined, still
continue to occupy the anterior portion of the layer. Whether they are
vdtiraately displaced to the margin of the retina, I am not at present able
to say. Clearly, however, the retina is developed out of the middle layer,
as in the previous case, but the nuclei of the retinal cells are pre-hacillar
in position. About the time of hatching nerve filaments grow out from
the brain, and thus connect the cerebral ganglia with the retinal portion
of the eye.

3. The lungs arise as a pair of extensive invaginations at about the
same time as the proctodseum. In sagittal sections of early stages the
lungs appear as oblong plates of cells, the large oval nuclei of which are
arranged in parallel rows (PI. XI. fig. 73). The cells forming the ven-
tral wall of the floor over the lung sacks, liowever, are several layers
deep, and their nuclei are not arranged in parallel rows as the other
nuclei are. The nuclei of the parallel rows undergo a change of form,
becoming flattened on one side and very convex on the other. In each
single row the convex faces look in the same direction, but the rows are

90 BULLETIN OF THE

SO arranged in pairs that the convex surfaces of all of the nuclei in one
row are directly opposite the convex surfaces of the same number of
nuclei of an adjacent row. The cells whose nuclei constitute such a pair
of rows form the two walls of a thin flat hollow sack, a respiratory
lamella. The outer surface of each lamella is covered with a continuous
thin chitinous secretion from these cells. The flattened surfaces of the
nuclei are turned outwards as regards the lumen of the sack, and the
convex faces are turned inwards. Ultimately the cells corresponding to
each pair of nuclei, which thus face each other, come in contact, and are
apparently fused together, thus forming pillars of protoplasmic substance
joining the walls of the lamella. The posterior borders of the lamellae
are free, and in most of the sections a pair of these cells with large nuclei
are found at these free ends. In the later stages (six or eight days after
hatching) the marginal pairs of cells become pigmented like the " hypo-
dermis " cells generally. The other cells remain up to this time without
pigment.

The surface of each lamella presents, as has been said, two chitinous
limiting membranes, one forming its dorsal surface, the other its ventral
surface ; these are continuous with each other at the free (posterior)
rounded margins of the lamellae, and at their anterior limit with the cor-
responding membranes of the lamella next above and below, respectively.
There is a constant difi"erence between these chitinous coverings : that of
the ventral surface is smooth and of uniform thickness, that of the dor-
sal surface is early characterized in sagittal sections by the presence of
fine, close-set faintly expressed tooth-like markings. Tliese markings
may be traced over the free edge of the lamella, but do not extend on to
its ventral surface.

The space embraced between the two chitinous layers of each lamella
is interrupted at intervals by the short 2-cell columns described above,
each with a single nucleus, which is so large as to touch its neighbor.
The protoplasrii enveloping the nuclei is exceedingly scanty, and thins
out at the flat margins of the nuclei into a layer which it is difficult to
trace as a lining to the chitinous membrane. It occasionally presents
slight irregular elevations, which project into the common cavity.
Through the anterior attachments of the lamellae, the cavities of which
communicate directly with the body-cavity, the blood has free access to
the cavity of each respiratory lamella ; blood corpuscles are conse-
quently to be seen in sections (PL XII. fig. 76, cp. kce.), and the
coagulated plasma of the blood often fills the lamellar cavities more or
less completely.

MUSEUM OF COMPARATIVE ZOOLOGY. 91

While these changes in the cells composing the lamellae have been
taking place, those which form the ventral wall of the body covering tlie
lungs have become detinitely arranged in two layers — an outer one,
which is distinctly continuous with the hypodermis of the surrounding
regions of the body, and like it is composed of a single layer of close-set
pigmented cells, with large elongated nuclei, and a deeper one composed of
elongated, unpigmented cells, with smaller nuclei. These two layers are
continuous with each other at the anterior lip of the pulmonary opening.

The two layers of cells are connected by cellular strands, probably of a
muscular nature, which run obliquely downwards and forwards from the
inner to the outer layer. They appear to be simple elongations of the
body of the ceUs of the inner layer. The remaining space between these
two layers forms a part of the body cavity, and like the true lamellae is
traversed by the blood. The darsal chitinous covering of the inner layer
is toothed like the dorsal surface of the true lamellae. According to the
figures of the adult structure given by MacLeod ('84, PL I. fig. 3), one
would expect to find the chitinous layer of this surface plain rather than
complicated.

I have not yet succeeded in demonstrating the existence of any mus-
cular differentiations in the 2-cell columns, such as is described and
figured by MacLeod for the adult.

IV. — General Considerations relative to some of the

Phenomena.

The discussion concerning the nature of the peripheral layer of proto-
plasm — or blastema — in the early condition of the laid egg, has al-
ready been referred to. Sabatier ('81) is the only author who has
attempted to assign a cause for the division of this layer into definite
polygonal areas. He concluded that the cause resides in the movements
of the internal protoplasm which migrates towards the periphery. This
hyaline protoplasm gushing forth from between the yolk corpuscles at the
surface produces, in his opinion, certain lines of division in the blastema.
The areas thus formed naturally correspond in size and position \n.ih. tlie
yolk corpuscles.

In view of the facts learned from sections of this stage this explanation
seems to me inadequate. If the markings were produced by a centrifu-
gally directed force the dividing lines wo\ild be at least " flush " with the
surface of the blastema, if not slightly elevated above it. As a matter

92 BULLETIN OF THE

of fact they are depressed (PI. V. fig. 28 ; PI. VI. fig. 30), and, there-
fore, can be accounted for more reasonably on the supposition of a centri-
petal force. It has already been indicated that these areas are formed
during the contraction of the vitellus ; surface views and sections com-
bined show that the very plastic protoplasm of the blastema is moulded
to the surface of the peripheral yolk corpuscles, to which it at fir.st
forms caps, all of the caps being joined at their margins. Owing to mu-
tual pressure these subsequently appear as polygonal areas. The cause,
then, producing this surface phenomenon is mechanical, and depends
upon the contraction of the protoplasm of the egg. But under what
influence does this contraction transpire 1 As the phenomenon takes
place during the stage which is characterized by the existence of the first
segmentation-nucleus, it is more than probable that the cause if. resident
in this central nucleus, which exerts its attractive influence on all the
protoplasm of the egg, but finds its external manifestation at this period
principally in the blastema. Thus, it is probable that the same cause
which produces in these eggs contraction of the vitellus, also induces the
division of the blastema into areas.

The maimer in which the protoplasm acts upon the yolk in the assimi-
lation of its substance is clearly indicated in the blastema stage. That
portion of the surface of the corpuscles which is in contact with the pro-
toplasm of the blastema appears deeply eroded (PL VI. figs. 30-33), and
the fine fragments into which the detached yolk substance is broken
gradually merge into the still more finely granular protoplasm. A some-
what similar fate overtakes the yolk corpuscles in the vicinity of the
central nuclei (PI. V. figs. 28, 29; PI. VI. fig. 34). Here, however,
there is a very gradual transition from the larger corpuscles to the much
smaller ones which immediately surround the finely granular protoplasm
of these central cells, — a process of fragmentation appears to precede the
erosion, and thereby a much greater surface of yolk substance is exposed
to the action of the protoplasm. In the former case there is no total
fragmentation of the yolk corpuscles, and the erosion proceeds from one
side only, leaving the opposite side with a sharp, more or less even out-
line. The preliminary fragmentation of the yolk in the vicinity of the
rapidly proliferating cells is, without question, correlated with the rapid
growth of the latter.

ClaparMe ('62), Barrois (78), and Balfour ('80), have each given
explanations of the reversion of the ewhryo. Balfour's is the simplest.

MUSEUM OF COMPARATIVE ZOOLOGY. 93

According to his conclusions the reversion is produced by the longitu-
dinal expansion of the dorsal region. It should be borne in mind that
previous to reversion the ventral plate extends nearly around the egg,
bringing the head end and the tail end near together on the dorsal side,
and that the narrow region separating these two structures represents
properly the whole of the dorsum. By expansion of the dorsum the
head and tail are removed further and further apart, and, according to
his notion, the embryo naturally bends upon itself ventrally, since it
cannot straighten out, and since, as he erroneously states, it does not
become shorter.

I have already shown (PL VIII. figs. 50-52) that the tail-lobe be-
comes separated from the rest of the body during reversion, a condition
that has not hitherto been recognized, but which is very apparent in
sagittal sections. This fact alone is sufficient to show that Balfour's
explanation as stated is untenable, for were reversion produced by a sim-
ple expansion of the dorsal region, — unaccompanied by shortening of
the embryo, — at the end of the process the tail would still exist as an
elongated conical appendage, instead of being shortened almost to oblit-
eration. In reality, however, a shortening of the ventral band does
take place, which is at least equal to the pre-existing tail-fold, and the
tail is in consequence drawn forward ventrally. This shortening would
tend to make the bands assume the position which would make the dis-
tance between head and tail least. There are further to be accounted
for in this period, the wide divarication of the nerve bands and the ven-
trad movement of the passive yolk mass. The cause for the latter must
be found in the relative pressures exerted upon it by the dorsal area on
the one hand and the ventral area on the other hand : that area which is
increasing most rapidly in extent would exert a constantly diminishing
pressure ; however, the shape of the area is of the utmost importance.
Although the dorsal region is changing its proportions most, it changes
rapidly from a very hroad and short condition to a lovg and narrow one.
The principal force, then, that pulls the nerve bands away from the ven-
tral surface is the one which tends to reduce the width (not as Balfour
will, the one that increases the length) of the dorsal region. The evi-
dent cause for this reduction in width is the dorsad concentration of the
ectodermic elements which accompanies the formation of the so-called
terga, and this is also the cause for the descent of the yolk mass, for the
cells that are brought close together to form the thickened ectoderm of
the tergal region put the remaining cells of the dorsal region to a tension,
the force of which is exerted upon the yolk mass. It will be remem-

94 BULLETIN OF THE

bered that at this time the ventral area is composed of a thin layer of
ectodermic cells ; these cells offer the least resistance to the movement of
the yolk mass, which, therefore, takes a ventrad direction.

During the period of reversion Agelena presents an interesting resem-
blance to certain conditions in the development of Oniscus. But accord-
ing to Bobretzky ('74, Fig. 15, hd.) the proctodeeum in Oniscus arises
some time before the formation of a tail-fold (1. c, Fig. 17, rf.), which,
moreover, is never conspicuously indicated. In Agelena the tail-lobe is
very prominent, and the fold which results in its formation appears long
before the proctodseal invagination. The appearance of this tail-fold, as
seen in sagittal sections, is so like that presented by the first stage in the
formation of the proctodceum in Oniscus (1. c, Fig. 15) as to suggest the
possibility that the infolding in the latter case is really a tail-fold and
not the proctodeeum, in which event hd. of Fig. 15 would correspond to
rf. of Fig. 16, and the proctodaeum in the latter figure would be a new
invagination. The principal objection to this view, aside from the
author's reputation for accurate observation, lies in the closeness of the
stages of Figs. 15 and 17, which would not seem to allow time for such
radical changes. Another and perhaps sufficient objection is, that the
invagination in question {Jul. Fig. 15) is lined with columnar epithelium
like the proctodteum of the succeeding stage, and that the slight tail-fold
is lined with flat cells. But whatever may be the truth with regard to
Oniscus, I am certain that in Agelena the tip of what I have called the
tail-lobe becomes the morphological end of the body, and that the proc-
todseum pierces the tip of this lobe after the reversion of the embryo is
nearly completed, and the tail-lobe has become much shortened.

One fundamental difference supposed to exist between the eyes of
Arthropods and those of A^ertebrates, relates to the direction in which the
light traverses the retinal elements. In the vertebrate eye the light
passes through the cells from their deep to their outer (genetically con-
sidered) ends. In the arthropod eye the light was supposed to have the
reverse direction ; but that this difference does not exist in the eyes of
Agelena is rendered apparent from its manner of development already
described. If the proliferation of cells which precedes the invagination
led directly to the formation of the eyes, the light would then traverse
the percipient elements from their outer to their deep ends ; and it is
probable that an ancestral eye of this kind prevailed. In the process of
invagination, however, this thickened portion — from which are formed
the retinal elements — is completely inverted, and as a consequence the

MUSEUM OF COMPARATIVE ZOOLOGY. 95

light must traverse the cells from their deep to their outer (genetically
cousidered) ends. There is, therefore, a striking analogy between the
condition which obtains in the eye of the spider and that which prevails
in the case of all vertebrates.

Saint Clocd, Minn., Dec. 1, 1885.

LITEEATUEE.

Ayers, H.

'84. Ou the Development of Oecanthus niveus and its Parasite Teleas.
Mem. Best. See. Nat. Hist., Vol. III. No. 8, pp. 225-281, PI. XV IH.-
XXV. Jan. 1884.
Balbiani, E. G.

'73. Meraoire sur le Developpement des Araneides. Ann. des Sci. Nat., 5°^
ser., Zool., Tom. XVm. Jan. 1873. Art. No. 1, pp. 1-91, PI. I.-XV.
Balfour, F. M.

'80. Notes on the Development of the Araneina. Quart. Jour. Micr. Sci.,
n. ser.. Vol. XX. pp. 167-201, PI. XIX.-XXI. AprD, 1880.
Barrois, J.

'78. Recherches sur le Developpement des Araignees. Jour, de I'Anat. et
de la Physiol, Tom. XIV. pp. 529-5i7, PI. XXXIV. 1878.
Bertkau, P.

'72. Ueber die Respiratious-organe der Araneen. Arch. f. Naturg. Jahrg.
XXXVIIL, Bd. I. Heft 2, pp. 208-233, Taf. VII. 1872.
Blanchard.

'57. Observations relatives a la generation des Arachnides. Comp. rend.
Paris, Tom. XLIV. No. U, pp. 741-742. AprH 6, 1857.
Bobretzky, N.

'74. Zur Embryologie des Oniscus murarius. Zeitschr. f. wiss. Zool., Bd.

XXIV., Heft 2, pp. 179-203, Taf. XXL, XXII. April 13, 1874.
'78. TJeber die Bildung des Blastoderms uud der Keimblatter bei den Insec-
ten. Zeitschr. f. wiss. Zool., Bd. XXXI. Heft 2, pp. 195-215, Taf. XIV.
Sept. 6, 1878.
Clapardde, E.

'62. Recherclies sur 1' Evolution des Araignees. Natuurk. Verhandel.,
Deel I. Stuk I. Utrecht, 1862. 92 pp. 8 pi.
Croneberg, A.

'80. Ueber die Mundtheile der Arachnideu. Arch. f. Naturg. Jahrg. XLVL,
Bd. I. Heft 3, pp. 285-300, Taf. XIV.-XVI. 1880.

96 BULLETIN OF THE

Emerton, J. H.

'72. Observations on the development of Pholcus. (Dec. 20,1871.) Proc.
Bost. See. Nat. Hist., Yol. XIV. pp. 393-395, PI. 2. June, 1S72.

Graber, V.

'79. Ueber das unicorneale Tracheatcn- und speciell das Araclmoideen- und
Myriopoden-Auge. Arch. f. mikr. Auat., Bd. XVII. Heft 1, pp. 58-93,
Taf. V.-VII. 1879.
Grenacher, H.

'79. Untersuchungen iiberdas Sehorgan der Arthropodcu. Gottingcn, 1879.
8 + 188 pp., 11 Taf.
Hatschek, B.

'77. Beitriige zur Entwickclungsgcschichte der Lepidopteren. Jenaische
Zeitschr., Bd. XI. Heft 1, pp. 115-li8, Taf. VII.-IX. Mar. 20, 1877.
Herold, M.

'24. Untersuchungen iiber die Bildungsgeschichte der wirbellosen Thiere im
Eie. Erster Theil. Von der Erzeuguiig der Spinnen im Eie. Marburg,
1824. 10+ 63 pp., 2 Taf. Folio.

Lankester, E. R. and A. G. Bourne.

'83. Lateral and Central Eyes of Scorpio and Limulus. Quart. Jour.

Micr. Sci., Vol. XXIII. pp. 177-212, PI. X.-XII. Jan. 1883.
Ludwig, H.

'74. Ueber die Eibildung im Thierreiche. Arbeiten a. d. zoolog.-zootom.

lustitut in Wiirzburg, Bd. I. pp. 287-510, Taf. XIII.-XV. 1874.
'76. Ueber die Bildung dcs Blastoderms bei den Spiimen. Zeitschr. f. wiss.

Zool., Bd. XXVI. Heft 4, pp. 470^85. Mar. 6, 1876.
MacLeod, J.

'82. Recherches sur la structure et la signification de I'appareil respiratoire

des Arachnides. Bull. Acad. roy. Belg., S-^^ ser., Tom. III. Ko. 6,

pp. 779-792. June 3, 1882.
'84. Same title. Arch, de Biologic, Tom. V. fasc. 1, pp. 1-34, PI. I., II. 1834.

Mark, E. L.

'81. Maturation, Fecundation and Segmentation of Limax campestris. Bull.
Mus. Camp. Zool, Cambridge, Mass., Vol. VI. Pt. II. No. 12, pp. 173-
625, 5 pi. 1881.

MetschnikofF, E.

'71. Embryologie des- Scorpions. Zeitschr. f. wiss. Zool., Bd. XXI. Heft 2,

pp. 204^2P,2, Taf. XIV.-XVII. 15 June, 1871.
'74. Embryologie der doppeltfussigen Myrlapoden (Chilognatha). Zeitschr.
f. wiss. Zool., Bd. XXIV. Heft 3, pp. 253-283, Taf. XXIV.-XXVII.
Sept. 16, 1874.
Plateau, F.

'77. Recherches sur la structure de I'appareil digestif et sur les phenomenes
de la digestion chez les Araneides dipneumones. Bull, de I'Acad. de Bel-

MUSEUM OF COMPARATIVE ZOOLOGY. 97

gique, 2e ser., Tom. XLIV. pp. 129-181, 323-355, 477-531, PL I.-III.
1877.
Rathke, H.

'42. Eutwickeluugsgeschiclite der Lycosa saccata. Froriep's Neue Notizen,
Bd. 24, No. 517, pp. 165-168. 1842.

Sabatier, A.

'81. rormatiou du blastoderme chez les Araneides. Comp. rend., Paris,
Tom. XCII. No. 4, pp. 200-202. Jau. 24, 1881.
Also abstract in English :
'81a. Ou the formation of the blastoderm in the Araneida. Ann. Mag. Nat.
Hist., 5th ser.. Vol. VII. pp. 277-279. Mar. 1881.

Salensky, W.

'71. Embryology of the Araneidae. Mem. (Sapisky) Kieff Soc. of Natural-
ists, Vol. II. Part 1, pp. 1-72, PI. I.-III. 1871. (Russian.) Abstract in
Jahresb. iiber Auat. u. Physiol. (Hofmann u. Schwalbe). Bd. II. (1873),
pp. 323-325. 1875.
Schimkewitsch, W.

'84. Etude sur I'Anatomie de I'lilpeire. Ann. des Sci. Nat. 6™* ser. Zool.,
Tom. XVII. Art. No. 1, 94 pp., PI. I.-VIII. Jan. 1S84.

'84*. Zur Eutw'ickluugsgeschichte der Araneen. Zool. Anzeig., Jahrg. VII.
No. 174, pp. 451-453. Aug. 18, 1884.

Schiitz, J.

'82. Ueber den Dotterkeni, dessen Eutstehung, Structur, Vorkommen und
Bcdeutuug. Inaugural-Dissertation. Bouu, 18S2, 21 pp., 1 Taf.

Sograff, N.

'82. Zur Embryologie der Chilopoden. Zool. Anzeig. Jahrg. V. No. 124,
pp. 582-585. Nov. 6, 1882.
Weismann, A.

'82. Beitrage zur Kenntniss der ersten EntTvicklungsvorgange im Insek-
tenei. Bonn, 1882. 32 pp., 3 Taf. From "Beitrage zur Anat. u.
Erabryol. J. Heule als Festgabe z. 4. Apr. 1882 dargebracht von semen
Scliiileru."

Whitman, C. O.

'78. The Embryology of Clepsine. Quart. Jour. Micr. Sci., n. ser.. No. 71,
Vol. XVIII. pp. 215-315, PI. XII.-XV. July, 1878.
Von Wittich, W. H.

'45. Observationes qusedam de Aranearura ex ovo evolutione. Halis, 1845,

20 pp., 1 Tab.
'49. Die Eutstehung des Arachnideneies im Eierstocke ; die ersten Vorgange
in demselben nach seinem Verlassen des Mutterkorpcrs. Miiller's Archiv.
Jahrg. 1849, pp. 113-150, Taf. III.

VOL. XII. — NO. 3. 7

98

BULLETIN OF THE

DESCRIPTIONS OF FIGUEES.

LIST OF ABBREVIATIONS.

A.

an.

anus.

L. lab.

labium.

ao.

aorta.

Ins.

lens.

1. app.

1st pair of appendages.

lob. ca.

caudal lobe.

2. app.

2nd pair of appendages.

lob. ce.

cephalic lobe.

6. app.

6th pair of appendages.

Ir.

labrura.

B.

b.

brain.

lu.

lumen.

bV.

blastema.

M. ms d.

mesoderm.

bid.

blastoderm.

VIS d. so.

mesodermic somite.

br. sic.

stercoral pocket.

mu.

muscle.

C.

cd. ab.

abdominal portion of

mu. lat.

lateral muscle.

nerve cord.

mu. vrt.

vertical muscle.

cd. n.

nerve cord.

N. nl.

nucleire.

c dn. ca.

caudal thickening.

0. ce.

oesophagus.

ch.

chorion.

ocl.

ocellus.

cr.

heart.

P. phx.

pharynx.

eta.

cuticula.

ppl.

protoplasm.

cum. pr.

primitive cumulus.

pr. app.

provisional appendages.

D.

d.

dorsum.

pr d.

proctodaeum.

E.

ec.

ectoderm.

pr-stc.

pre-stercoral tube.

ed.

epidermis.

pr z.

protozonite.

en.

large . >called ento-

p-sd.

post-gastric tube.

dermic cells.

R. rtn'.

1st (inverted) layer of op-

G.

gl.

gland.

tic invagination.

gl. src.

spinning gland.

tin".

2nd (non-inverted) layer.

gn.

ganglion.

S. sd.

stomodaeum.

gra.

granular ("puncti-

T. T.

tail, tip of body.

forra " ) substance.

trg.

tergite.

H.

hd.

hypodermis.

V. vit.

" vitreous body."

I.

iv.

invagination, to form

vl.

vacuole.

the pulmonary sac.

Y. yh.

yolk.

The figures of Plates I.-IV. are surface views of eggs by reflected light, each
magnified about sixty diameters.

Fig.

PLATE I.

1. An egg showing the primitive cumulus.

2. A more advanced egg, showing the primitive cumulus, the caudal thick-

ening (c dn. ca.), and between the two, traces of the forming ventral
plate.

MUSEUM OF COMPARATIVE ZOOLOGY. 99

Fig. 3. The same view of another egg.
" 4. End view of an egg at the stage of the primitive cumulus.
" 6. View upon the anterior end of the egg from which Fig. 2 was drawn.

The larger polygons are boundaries of yolk masses, the smaller those

of blastoderm cells.

PLATE II.

Fig. 6-11 are side views of eggs, and are arranged in the order of their sequence
in development.

■" 6. Left side of the embryo :it the stage which exhibits six protozonites.

" 7. View of the right side at the stage marked by the beginning of the
appendages.

" 8. A more advanced embryo (left side) with incurved permanent appendages
and four pairs of provisional appendages.

" 9. A somewhat oblique view of the right side during the period of reversion,
showing the rudimentary terga (trg.), also an increase in the distance
between the cephalic and caudal lobes.

" lO. Embryo (left side) wlien reversion is nearly completed, showing the per-
sistence of the two posterior pairs of provisional appendages, and
also their change in position to the posterior part of the body.

" 11. Embryo after the reversion is completed ; tlie last two pairs of provisional
appendages are being modified to form the spinning mammillae.

PLATE III.

Fig. 12. The head and tail lobes at the stage when the appendages begin to
appear.

" 13-17 form a series of dorsal views during reversion.

" 13. Dorsal view of an embryo at the beginning of reversion.

" 14. An embryo a little further advanced, to show the separation of the head-
and the tail-lobes, and also the increase in the dorsal extension of the
tergites.

" 15. A similar view of an embryo a few hours older than the preceding.

" 16. Dorsal aspect of a still older embryo, in whicli the tail-lobe is just disap-
pearing from tlie dorsal surface.

" 17. An embryo at about the stage represented in Fig. 11.

PLATE IV.

Fig. 18-23 present a series of ventral aspects.
" 18. An embryo at the stage which shows six protozonites.
" 19. An embryo showing the beginning of the appendages ; it is of the same

age as that shown in Fig. 7.
" 20. View of the caudal lobe at about tlie same stage as is represented in

Fig. 8.
" 21. Ventral view during tlie period of reversion. The embryo corresponds

nearly with the stage represented in Figs. 10 and 16.
" 22 Ventral aspect of the embryo from which Fig. 14 was drawn.

100 BULLETIN OF THE

Fig. 23. Au enlarged view of the head region, at about the same stage as the pre-
ceding.

PLATE V.

Fig. 24. A portion of the surface of a living egg of Agelena nctvia, after the division
of the blastema into polygonal areas. Several of the yolk corpuscles
have shifted from their original positions, and therefore no longer
coincide with the areas. X 163.

" 25. Polygonal areas of the blastema and underlying yolk globules more
highly magnified; from a living egg. X 440.

" 26. Primary blastodermic cells before they become regular in form and size;
from a living egg. X 163.

" 27. A portion of the blastoderm on the third day (temperature 23° C) of de-
velopment ; from a living egg. X 163.

" 28. A little more than one half of the section of an egg, containing only one
nucleus, the first segmentation nucleus; siiowing blastema {hi'.),
nucleus (nl.), and yolk corpuscles (yk.)- X 110.

" 29. First segmentation-nucleus with the surrounding protoplasm highly mag-
nified, showing also a rapid diminution in the size of the yolk cor-
puscles in its vicinity. X 330.

PLATE VI.

Fig. 30 Enlarged view of a portion of Fig. 28, showing the blastema (W.) and

underlying y^lk corpuscles (yk.)-
' 31, 32. Isolated peripheral yolk corpuscles, to which portions of the blastema

are attached.
" 33. Isolated yolk corpuscle with a vacuole, which in turn contains a rounded

yolk globule.
" 34. Section through the nuclei of an egg in the two-cell stage, showing the

two groups of yolk columns (Deutoplasmasaulen). X 110.
" 35. One of the deep internal cells, surrounded by yolk.
" 36. A nucleus containing a central vacuole ; from an egg in the two-cell

stage.
" 37. A migrating cell that has just reached the periphery, abutting on tlie

blastema (W.).
" 38. Detached portion of the blastema viewed from within, showing depres-
sions into which the yolk corpuscles fit.
•' 39. Section of an egg passing transversely through the primitive cumulus in

the region of its greatest width. X 110.

PLATE VIL

Fig. 40. Radial section of two blastodermic cells.
" 41. Section passing sagittally through the primitive cumulus. X 110.
" 42. A blastodermic cell in the process of division, with " interzonal filaments."

MUSEUM OF COMPARATIVE ZOOLOGY. 101

Fig. 43, 46, 47. Blastodermic cells ; to show some of the conditions presented by
the chromatine and nucleoplasm of their nuclei.

" 44. A cell in the process of division, further advanced than the one repre-
sented in Fig. 42.

" 45. A portion of Fig. 49 highly magnified to show the columnar nature of the
ectodermic cells and the complete differentiation of the mesoderm.

" 48. Ectodermic cells with two nuclei from a late stage, during the infolding,
to form the ovary.

" 49. Sagittal section through an embryo in the protozonite stage, X 110.

Note. — In cutting and mounting, the section was artificially ruptured in
two places, but none of the blastoderm has fallen away.

PLATE VIII.

Fig. 50. Sagittal section through an embryo during reversion, showing stomo-

dffium, tail-lobe, etc. X 110.
" 51, 52. Views of sagittal sections of the posterior region during reversion, to

show the condition of tlie tail-fold.
" 63. Section of the tail region and apart of the dorsal region, to show the

mesodermic somites of the rudimentary terga. The section is cut

obliquely to the median plane. X 110.
" 5t. Sagittal section of the morphological tip (T) of the body near the close

of reversion, to show the early condition of the proctodaeura. X 110.
" 65. Sagittal section of the hind pan of the body, to show the stercoral pocket

(br. stc.) and the pre-stercoral tube {pr-stc).
" 56. Sagittal section of the hind part of the body, to show the trumpet-shaped

condition of the pre-stercoral tube and the somatization of the body as

indicated by the segmental grouping of the muscles {mu. 1 — mu. 6).

PLATE IX.

Fig. 57. A nearly horizontal section of the proctodaeura and the stercoral pocket
{br. stc), about three days before hatching, showing the columnar
epithelium and the narrow lumen of the proctodaeum. X 310.

" 58. Sagittal sections of anus and stercoral pocket, about eight days after
hatching. X 110.

" 59. About one half of a transverse section in the region of the " rudimentary
terga," to show tlie mesodermic somites of these dorsal elements and
their connection with the ventral portion of the mesoderm. X 100.

" 60. Transverse section of an embryo near the beginning of reversion, pass-
ing through the stomodaeum and the 2nd pairs of legs. X 100. Com-
pare the separated nerve bands {fin.) with those of Fig. 62.

" 61. A section from the same embryo as Fig. 59, showing entodermic cells
(en.) in the region of the tergal elements.

" 62. Transverse section after completed reversion, through that pnrt of the
stomodasum wliich becomes the sucking stomacli. If shows the mus-
cles attached to sucking stomach -[ww. vrt.,mu. lat.), and the approxi-
mation of the nervous bands {(jn.). X 110.

102 BULLETIN OF THE

PLATE X.

Sections illustrating the development of the eye8.

The four pairs of eyes are called according to their positions : anterior lateral,
anterior median, posterior lateral, posterior median.

Fig. 63. A sagittal section showing an early condition of an anterior median eye,
— a thickened mass of " hypodermis " cells with the beginning of an
invagination. X 430.

" 64. Sagittal section of an anterior median eye after the invagination is fully
established, three or four days before hatching. X 430.

" 65. A frontal section through the anterior median pair of eyes, showing the
narrow lumen of the Invagination and its limited lateral extension.
X about 300.

" 66. An older stage (one to two days before hatching) showing an elongation
on the part of the " hypodermis " cells which constitute the " vitreous
body," and also the closure of tlie invagination. X 430.

" 67. A sagittal section passing through an anterior and a posterior eye of the
same side, two days after hatching. X 430.

" 68. Sagittal section through an anterior median eye one day before hatching.
X430.

" 69. Sagittal section through an anterior median eye, eight days after hatch-
ing ; the retinal portion has not yet reached its full devolopment.
X about 360.

PLATE XI.

Fig. 70-72, 77, 78, show the gradual headward concentration of the nerve bands.
Fig. 77 represents the earliest stage, in which the nerve bands reach
nearly around the egg; Fig. 72 shows the nervous elements con-
tracted so as to occupy only the folded ventral region ; in Fig. 71, the
abdominal cord is shortened considerably more ; Fig. 70 represents
the ventral ganglia concentrated within the thorax ; Fig. 78 shows
the condition of the brain and ventral ganglia at the time of hatching.

70. Sagittal section approximately in the median plane, from an embryo
about two days before hatching.

71. Sagittal section through the brain and nervous ganglia at about the com-
pletion of reversion. X 110.

72. Sagittal section through the nervous system at the stage of the formation
of the proctodaeum. X 100.

73-76 show four successive stages in the formation of the lungs, all magnified
about 300 diameters.

73. View of the right-hand surface of a sagittal section of the lungs in an
early condition (about the middle of the period of reversion), showing
the nuclei arranged in parallel rows.

MUSEUM OF COMPATIATIVE ZOOLOGY. 103

PLATE XII.

Fig. 74. Sagittal section (left-hand surface) from an embryo somewhat older than
the preceding.

" 75. Sagittal section of the lungs at about the time of hatching.

" 76. Sagittal section of the lungs five or six days after hatching ; the upper
and lower walls of each lamella are connected by the union of the
nuclei (n/.).

" 77. Sagittal section showing the nervous system at the beginning of reversion.

" 78. Sagittal section of the cephalothorax at the time of hatching, to show the
post-stomodaeal portion of the alimentary canal, and also the concen-
trated condition of the ventral ganglionic mass.

LOCY

Development OF Agelena

PlI

rum pr.

CUfn.pf-.

.J^-

"C <fn.at.

I
Ik

cu/rt pr

c dn CO,-

cufn . py.

U'A I.OCV if\

LOCY

Development of Agelena

pLir

app

\

y

■r'

lob.ce

^^

'app.

r J

'^'

v

loh ce...

V

'V/'

\

loi. ra .

pr. app

lob en

try

&

•I'P

r

U<b n

10

pr.app.

LOCY.

/hVi-:; : PMLM' jf «gelexa

/oA

lob- rn

try.

12

13

'W

"P/r

•V'P

'^iifeiaaai**^

'--.^'^

/"■ ap/i

1*

15

&■

pr. upp

1 firt^i

16

17

LOCY.

Devklopme.nt of Agelena

PlIV

,.>

2 op,.

J

,^-

18

*y affp

19

•app.

■ app

prapp

3 app

J '

20

loh-ca

21

/■

•f'P

li>b ce.

sd

23

app.

prapp

22

' *. ,. " ''•' ' ■ >

..U

)/

24

25

27

28

29

r

30

31

32

A

•■■■.»

vl

33

^^^'^■..

36

.i '.'•ISti ..c

--.JJJg^SS^'

:^-^r^-

«

^^^-

-*

#

«?&f.

■*■ >■

#

4&

<S't,

f

/■■

-^^*a,g^

Pi.vin.

^-

^-

,/

-•>

_-'•-«'

so

.'•'I

'•ij- (/ ^o

.0^

■"N

53

en.

"■-, ^-

t

■^'^"'*''*-

■V

hi- -sir

9

(»

•m

_-,'- (»V

.0^ -"

&6

'■>:

^<^

•O »

9

^*

■"'• •'/!

■, '•' ''♦

UJ

7< */r

/,r.ii-

55

WA Locv-. o«l

B Meisci l.:t.

PlIX

.&"

0

... r

0

I, -^

t c i^

< ciP/ , ,,

cSfK- "?

^ Sf V -^^r^l

S8

57

or'.-"'

=3 e

r;c:-tfc -^

t /e

Hi '

V V

o

S^, ^ - ^J its'

^^s^ :^ ' ^"^z

'"%.

\^

so

nut.

J

, c

fc«o^ ""

f«\vf.*e„

,«

r •*

.»c

~ . vV wi,' d. «»

( 7^ tc

^^

' "'-' '

i^

..„>-■

%

^>.
^J«

o .

^J^ tiuJat.

^<£J'

*^..

62

LOCY.

DEVELOPMENT CF AgELENA

-^.0.^%^

#-!

?^

■'^

65

X.;iJW^V

>'/•

WJi.

;^;tv^^^?:-;

.•/-■. :'C<.v

n>t»

'*'.

66

'/

■>..

■•■?

^- »■.' jjf -.■t--''^

67

d

■ s^^^i^^M\

v:?'

vJ#^

i^

i%-

68

69

tt-.A

r..-n

;Jl

/«*^
_^j;..

'e*

*^:

<^\

•0

i-.-y-'

■v.r-

4

*"■ ^« t^

JS

«■■,

i

v.

3

%.

'%.=

•^•'>

>?!«

!/

Hi-

*J:i:vi^:

k^

j^

""^X

a^*

**--

-f-

70

« . ■'cl

pkr

\ \

a

* K'!

•^.z

■

'--^-'

4

'*"j. ^

r,l

■ ^

0

•^^x^C'

^

^

^^c-"

o

»

'J

"|\o-

#

i^

Arf

%:

..rx

X

.*'

72

U'A f.ocy. c«l

LOCY.

P. V!
i .. All.

'2* - • - ■■>' T I I/.L-V (J

U

I

%

U \>'-

k

KfiJ ,,^

\j.

m^'

."3^'
*i-?^

'■■.^•5^

^.:-

.»*-»?-r?iJ!353t**» «-»-;«. a

1 :«

7S

■ # .

'<is«

^^

78

No. 4. — Studies from the Newport Marine Zoological Laboratory.
Communicated hy Alexander Agassiz.

XVII.

Preliminary Observations on the Development of Ophiopholis and Echin-
arachnius. By J. Walter Fewkes.

The following paper considers the development of our common Ophi-
uran, Ophiopholis aculeata, Gray, and of the Clypeastroid, Echinarachnius
parma, Gray. All the observations were made last summer, those on
the former at Eastport, Maine, those on the latter at the Newport Labora-
tory, Newport, R. I.

OPHIOPHOLIS ACULEATA, Gray.

Few observations have been published on the metamorphosis of our
common Ophiopholis. The eggs of 0. acxdeata (hellis, Lym.), according
to A. Agassiz,* are laid in bunches, and the young develop without pass-
ing through a free plutean stage. He gives two figures of the young
Ophiopholis made by L. Agassiz in 1848.

In " Sea-Side Studies " the young Ophiopholis is said to be carried in
a pouch, in which the first stages of development occur.f A figure, one
of the two mentioned above, is doubtfully identified as a drawing of the
young Ophiopholis in the second number of the Embryological Mono-
graphs. J Packard § states that in Ophiopholis the development is direct
and without metamorphosis.

* Embryology of the Echinoderms. Mem. Amer. Acad., Vol. IX., pp. 18 and 22.
The pluteus referred to Amphiura squamata in the " Embryology of the Echino-
derms," and doubtfully to Amphiura in " Embryological Monographs," may be a
pluteus of Ophiopholis. Amphiura squamata, Sars, is viviparous, and has no free
pluteus.

t Sea-Side Studies in Natural History. Marine Animals of Massachusetts Bay,
p. 137.

t Mem. Mus. Comp. Zool, Vol. IX. No. 2, PI. III. Fig. 20.

§ Zoology for Students and General Readers, p. 110. As nothing is said of
direct observation, it is probable that this statement is a compilation probably
from those already quoted.

VOL. XII. — NO. 4.

106 BULLETIN OF THE

These are the most important references whicli have been found to the
embryology of this, one of our most common Ophiurans. My observa-
tions differ radically from the statements quoted.

The eggs of Ophiopholis are cast free in the water, and the young pass
through a metamorphosis, in which a larva commonly called the pluteus
is formed. The mode of development of this pluteus is different from
that of any Ophiuran which has yet been described. It is most closely
allied to that of Ophiothrix, but in the mode of formation of a gastrula
differs widely from the account of a species of Opbiothrix. " 0. versicolor"
traced by Apostolides.*

The formation of the gastrula in Ophiurans has been very unsatisfac-
torily studied. An invaginated gastrula has never been figured in this
group. Balfour f in a short notice states that he has observed in Ophio-
thrix that the gastrula stomach is formed as in other Echinoderms by au
invagination (of the blastoderm). The same mode had previously been
suggested as probable by many embryologists, and had found its way
into all the more important text books. It is not accepted by Aposto-
lides, one of the latest students of the development of these animals.

Apostolides \ strongly combats the explanation of the method of for-
mation of the gastrula by invagination, and brings forward new observa-
tions on Ophiothrix, the same genus studied by Balfour, to show that
no invagination of the blastoderm occurs, and that the hypoblast of the
stomach is formed from cells in the inside of the blastosphere. To these
observations he brings as aids his studies of Amphiura to prove that in
Ophiurans the normal method of invaginated gastrulae does not exist.

The observations, therefore, which I have made, are thought to have a
morphological importance as supporting the a priori views of most em-
bryologists, and the direct observations of Balfour on another genus, of
the method of formation of the stomach of the pluteus of Ophiurans by a
primitive invagination of the blastoderm. I have never observed the gas-
trula of Ophiothrix, and can speak with confidence of Ophiopholis only,
as far as this point is concerned. The differences between Apostolides'

* le Th^se. Anatomie et D^veioppement des Ophiures. Arch. d. Zool. Exp. rf
Geu. X. Apostolides does not seem to hare suflBciently studied the descriptions of
the various species of Ophiothrix in the writings of Ljungnrian (Oph. Of Kong. Akiid.
p. 625, 1871. Description of 0. Lusitanka), and Lyman {Bull. Mus. Comp. ZoSi,
Vol. III. Part 10, pp. 240-249). The " 0. versicolor," Apostolides, is probably, as
has been suggested to me by Mr. Lyman, the same as 0. Lusitanka, Ljn.

t A Treatise on Comparative Zoology.

X Op. cit., pp. 192 and 207.

MUSEUM OF COMPARATIVE ZOOLOGY. 107

and Balfour's observations on Ophiothrix may be settled by those whose
good fortune it may be to study the embryology of this genus, but
since the archenteron is shown in the account of observations here pub-
lished to be formed by an invagination in Ophiopholis, we may still
adhere to our acceptance for some genera of brittle-stars of a general law
of Echinoderm development, known to apply to the gastrulte of some
genera of the Holothurioidea, Echinoidea, and Asteroidea. "While, how-
ever, my observations are believed to show that in at least this genus an
invaginated gastrula occurs, they do not prove that the opening into the
primitive infolding becomes the anus of the pluteus.

Our most accurate knowledge of the young stages of Ophiurans relates
to a viviparous genus, Amphiura.

Although the development of Amphiura has been studied by several
observers, we find in their accounts of the subject so many discrepancies,
that a call is made for a new study of the first stages of this and related
genera. ]\[etschnikoff* supposes that in Amphiura the stomach, " Ver-
dauungsapparat," is formed by invagination. According to Apostolides
tlie endoderm is formed by delamination, and there is no such invagina-
tion, although he describes a primitive opening in the lai-va, which he
considers the anus.t "Why lie should give this name to the opening in
question does not appear, and if he has gi'ounds for such an interpreta-
tion he does not make them evident in his account.

Another opening into the digestive tract, of the origin of which he is
equally reticent, he calls the mouth. The endoderm or wall of the diges-
tive cavity, according to this author, is formed in Amphiura by delamina-
tion, and not by invagination of the blastoderm.

As bearing upon the question of whether the primitive opening of
the larva becomes a mouth or not in Ophiurans, an observation of Sir
"Wyville Thomson on Ophiacanfha j,'ivipara, Ljn., is important. He
says : J " Although I had not an opportunity of working the matter out

* Studien iiber die Entwickelung der Echinoderraen und Nemertinen. M^m,
de I'Acad. Imp. des Sci. St. Pctersh., VII. Ser., XIV. 8, p. 14.

t Metschnikoff, in a later publication {Zeit.f. Wiss. Zool, XXXVII., p. 307,) ex-
presses an opinion against the idea of Apostolides that the endoderm is formed by
delamination in Ophiothrix, and explains the error into which he supposes Apos-
tolides has fallen, by the supposition that he (Apostolides) has confounded the
mesoderm with the entoderm. In a note on the same page he takes occasion,
however, to express his agreement with Apostolides that an intestine and anus is
present in the embryo of A. squamata.

X Notice of Peculiarities in the Mode of Propagation of certain Echinoderms of
the Southern Sea. Journ. Linn. Soc. XII., pp. 77 and 78. Here mentioned as

108 BULLETIN OF THE

with the care and completeness I could have wished, I feel satisfied from
the examination of several of the young, at a very early period, that in
this case no provisional mouth, and no pseudembryonic appendages what-
ever are formed, and that the primary aperture of the gastrula remains as
the common mouth and excretory opening of the mature form." In a
larva of an unknown Ophiuran, Krohn * finds the first infolding, " Ver-
tiefung," in the position later occupied by the mouth of the adult.

I am unable to quote any direct observations on the gastrula of Ophi-
urans to show that the primary opening or gastrula mouth becomes a
plutean anus. An anus is wanting in the adult Ophiuran.

Although Apostolides criticises the explanation given by others of the
method of formation of the openings into the internal cavity (stomach)
of the gastrula by an invagination, he does not show how mouth or
anus is in reality formed. As he does not show the old view to be erro-
neous, and suggests nothing better, we must at present adhere to the
commonly accepted explanation. The interpretation of Metschnikoff,
who regards the first formed opening as a mouth, seems more reasonable
than to suppose with Apostolides that it is an anus. Whatever it may
eventually become, Metschnikoff 's suggestion, that it is formed by an
invagination, conforms with what I have observed in the gastrula of
Ophiopholis.

In a short notice of the development of Ophiophragma, Professor
Kachtrieb t refers to a blastopore, and a stomach " enteron " iii its gas-
trula. No infolding of the blastoderm to form this enteron is recorded,
but the recognition of the primitive opening as a blastopore in another
Ophiuran genus is worthy of notice.^ I believe the gastrula stomach of
Ophiophragma will be found to be formed by invagination as in Ophio-
pholis. Professor Nachtrieb also studied the development of Ophiothrix,
but his mention is too short to give me any information as to how he
regards the gastrula stomach as formed. From what he does give it
is supposed that the stomach is developed in the same way as that of
Ophiopholis.

Ophiocoma didelpht/s, Wyv. Th. See also general results of the voyage of the
" Challenger," by the same author, p. 241 et seq., Fig. 60.

* Ueber einen neuen Entwickelungmodus der Ophiuren. Arch. f. Anat. Physiol.
«. Wiss. Med. 1857.

t Johns Hopkins University Circular, March, 1885.

J By a comparison of Apostolides' figures of Ophiothrix it will be seen that the
pluteus has pushed out the lateral arms to double the diameter of the body before
a mouth or any external opening into the cavity of the pluteus is formed.

MUSEUM OF COMPARATIVE ZOOLOGY. 109

Observations.

Plate I. Tigs. 1-93.

The material upon which my observations were made was collected at
Eastport, Maine, in the summer of 1885. Adult Ophiopholes were
dredged off Friar's Head, Campobello. Great numbers were taken just
below the line of low tide on Clarke's Ledge, near Eastport.

The following observations on the development of the egg were begun
after its fertilization, and after it had been laid.

Ova were voluntarily cast by the female on the 1 7th of August. They
were found in multitudes at the bottom of the glass dishes in which the
adults were confined, forming a greenish or yellowish cloud discoloring
the water. A white fluid of spermatozoa was also found in another dish
containing males. As both elements are cast in the water it is probable
that in this species fecundation occurs outside the body,* as is generally
the case among Echinoderms. The ova of Ophiopholis, like the adults,
appear to be very hardy, and very little care is necessary to keep them
alive. The contact of sperm and ova was not observed. The white fluid
containing spermatozoa was mingled with fluid containing ova, and it is
thought that artificial fecundation was thus effected.f The ova began
to develop soon after. Fecundated ova were also found in water in
which many Ophiopholes were living.

Each egg, PI. I. fig. 1, is enclosed in a transparent capsule .13 mm. in
diameter. This capsule in the first stages observed was not thick as in
the viviparous genus, Amphiura, but very thin. It is thought to be
homologous with the outer layer, m c, mentioned and figured in Metsch-
nikofi"'s account J of the development of A. squamata. Its thickness
may have been greater in younger stages. The eggs are not laid in
bunches, masses, nor were they observed to be cemented together. They
were not observed to develop in pouches, although pouch-like parts of
the genital glands, ovaries, are sometimes squeezed out through the geni-
tal slits as in the genus, Gorgonocephalus. When the ova were first
examined segmentation had not begun, but no germinative nucleus was
seen. Each egg in the youngest stage, PI. I. fig. 1, has the yolk com-

* In Amphiura fecundation takes place in the body, teste Apostolides, Metsch-
nikoff, and others.

t Metschnikoff {Zeit. f. Wiss. ZoSL, XLIL p. 664), artificially fertilized Ophio-
thrix fragilis.

X Op cit., p. 14. Plate III. Fig. 3.

.110 BULLETIN OF THE

posed of a slightly opaque greenish centre, vt, the diameter of which is
two-thirds that of the egg-capsule, and a superficial transparent layer t,
which may be a thickened envelope, part or the whole of which later
becomes a vitelline membrane. Both these elements of the ova are
affected by segmentation.

The transparent plasmic envelope of the vitellus of Ophiopholis is
thought to be identical with a similar layer described by others in star-
fishes and sea-urchins. Selenka,* for example, has described a similar
layer in Toxopneustes variegatus, and the question of the origin and fate
of the vitelline membrane (?) has been discussed by Giard, t Fol, %
Perez § and Selenka. || The discussion of the intricate complications of
the question which the origin and fate of the cortical layer of the Echino-
derm egg necessitate, must be passed over at present, as most of the
phenomena considered by the above-mentioned authors antedate a stage
of the Echinoderm egg corresponding with the youngest Ophiopholis
which I have studied. The observations here recorded are supposed to
have a value in indicating the existence of the cortical transparent layer
in the Ophiurans where, possibly with the exception of Amphiura, it has
been overlooked by other embryologists.

The outer or superficial layer of the yolk is believed to be the same as
the " Structurlosen geblichen HUlle," described by Metschnikoff (op. cit.
p. 14), in Amphiura. The same layer is thought to be figured by him in
PI. III. figs. 3-6, mv. as the " Dotterhaut."

While the outer membrane of the Amphiura egg is conspicuous in
Metschnikoff's figures of the youngest eggs, in older stages it suddenly
disappears. The same thing occurs in the capsule of the Ophiopholis
egg, which leads me to suspect that they are the same structures.

According to Apostolides ^ a part of the protoplasm seems to condense

* Beobachtungen uoer die Befruchtung und Theilung des Eies von Toxopneus-
tes variegatus. Sitsungsb. d. phys.-med. Gesells. z. Erlangen. X. pp. 1-7. Zoologische
Studien. I. Befruchtung des Eies von Toxopneustes variegatus. Leipzig, 1878.

t Compt. Rend. LXXXIV. 7 (Trans.) Ann. Mag. Nat. Hist. [4] XLX. 113, pp. 434-
436.

t Compt Rend. LXXXIV. No. 14 ; id. LXXXV. No. 4 ; id. No. 14.

§ Op. cit.

II Op. cit. The external transparent region mentioned above may correspond
with the "superficial hyaline membrane" described by Fol (Ann. Mag. Nat.
Hist. [4] XX.) over the vitellus of the starfish ovum. See also Perez, Sur la
Fecundation de I'CEuf chez I'Oursin. Compt. Rend. Acad. Sci., LXXXIV. p. 620;
LXXXV. p. 353. Ann. Mag. Nat. Hist. [4] XX. pp. 156-158.

IT Op. cit.

MUSEUM OF COMPARATIVE ZOOLOGY. Ill

at the centre of the yolk in the first condition of tne egg. It there be-
comes more dense and divides. Of the two masses of unequal size,
" Toujours," he says, " une des deux pr^sente des proportions conside-
rables par rapport h I'autre." In fig. 3 PL XI., to which he refers for
this phenomenon, this difference in size of the two is not well shown,
and the condensed central region is not separated from the superficial by
as strong a line of demarkation as in Ophiopholis. It is supposed that
the condensed central part which is referred to in his description is the
same as the slightly opaque or greenish centre of the Ophiopholis egg,
but such an interpretation is open to doubt. The transparent superficial
layer is not homogeneous throughout. At one pole on the outer surface
of the transparent layer of the yolk, still fastened to it or not sepa-
rated from the yolk cells, a single globule was observed, PI. I. fig. 2, c?,
possibly in the process of forming. This globule forms a slight elevation
on the surface of the transparent layer, and a corresponding conical ele-
vation was observed under it on the denser part of the vitellus.* Later
in time a globule separated from the yolk was observed, and in later
stages of development an additional globule is formed, PI. II. fig. 3.
The largest number of free polar globules observed was two. Polar glob-
ules are not figured or mentioned in Ophiothrix by Apostolides. He
speaks of them, however, in Amphiura. The superficial layer of the
yolk appears to surround both cells, and in the contiguous surfaces of
the two cells this layer is undivided, corresponding in its position with
the plane of the first cleavage, PI. I. fig. 3. This coincidence causes the
two cells of the 2-cell stage to appear separated by a transparent layer,
which at the same time unites them,t Fig. 3, 1 cl. pi.

The formation of the 2-ceU stage does not occur immediately after tht
sperm is added to the glass containing the ova. As in Echinarachnius a
considerable time elapses after the mixture of the two elements before
^jhe formation of a 2-cell stage. The indications are that the first changes
go on more rapidly in Ophiopholis than in Ophiothrix, as observed by

• Thi8 conical elevation may be connecieu with the " Dotterhiigel " of Fol and
others. It has, however, no existence on the outer surface of the plasmic or corti-
cal layer. A more acute histological examination of the single globule on the sur-
face of the latter above the conical elevation is necessary before it can be stated
whether it is a polar globule or a spermatozoon. I regard it as the first polar
globule.

t A characteristic connection of the two blastomeres is mentioned in the egg of
Ophiophragma by Professor Nachtrieb. Whether it has a likeness near or remote
to this condition in Ophiopholis, it is impossible for me to say. From hit short
description I am unable to compare the two genera in this regard.

112 BULLETIN OF THE

Apostolides. No change was detected in the ovum of the latter seven
hours after fecundation. On the seventh hour after the contact of ova
and sperm, according to Apostolides, the first indications of the segmen-
tation of the egg appear in Ophiothrix. Seven hours after the capture
of Ophiopholis some of the ova were found in the 4-cell stage, and a
little over three hours after the egg was dropped by the female they were
found in the 2-cell stage. Is this discrepancy the result of a difference
of temperature in the water 1 *

The second segmentation stage, the 4-cell stage, PL I. figs. 6, 7, is
brought about by a formation of a second cleavage plane, 2 cl pi, at
right angles to the first. As in the first the two cells of the 2-cell stage
are separated by a transparent layer, in the same manner each of the two
ceUs of the 4-cell stage are divided by a similar plane. A nucleus is
observed in each of the spheres of the 4-cell stage. The division of the
2-cell stage into four cells is regular,t and all the cells are of the same
size up to the 4-ceU stage.

Cleavage.

The first external change in form of the spherical egg in its segmenta-
tion is the formation of the first cleavage-plane. A constriction or annular
groove, destined to divide the egg into hemispheres, forms about the egg
in the same way as in other Echinoderms. It was not observed whether
a collar-like J extension of the superficial portion of the yolk sinks into
the denser central region, or whether the two cells form in some other
way, but in a well-formed 2-cell stage the two cells are separated by a
transparent walL The transparent wall is identical with that which
covers the denser part of the ovum in its undivided condition.

The 4-cell stage is formed from the 2-cell stage by a cleavage-plane,
2 cl pi, at right angles to the first. As in the 2-cell stage the two cells
or blastomeres are separated by a transparent layer, so in the 4-cell stag€

* Metschnikotf {Zeit. /. yfiss. Zodl. XLIL, p. 665) has recorded a great differ-
ence in the time occupied by the early development of the eggs of the same Echino-
derm from two localities. Tliese differences are probably due to temperature.
The influence of temperature on the rate of development of the ova of Echino-
derms is a subject which would repay an extended investigation.

t If any irregularity in size exists the difference is very small.

t It may be supposed tliat the superficial layer following a constriction of
the denser part of the ovum, bisects the latter by a centripetal growth- This
would correspond with the mode of formation of the 2-cell stage in other Echino-
derms.

MUSEUM OF COMPARATIVE ZOOLOGY. 113

the cells are likewise separated by a transparent wall of the same general
appearance. Each of the cells of the 4-cell stage has a nucleus.

The 8-cell stage, Figs. 8, 9, follows close upon the 4-cell, the addi-
tional cells forming by a subdivision of those already existing. This
division is generally regular, all the cells being of like size. In some ova
smaller cells were observed with larger in the 8-cell stage. The cells
have a centrifugal tendency, and a central unoccupied cavity, cav, can be
seen enclosed by them. This cavity, which is the cavity of the blasto-
sphere, grows in size as the larva matures. There is no solid morula stage,
but a segmentation-cavity can be recognized in eggs as young as the
4-cell stage. The transparent layer t, which envelops the 8-cell stage,
and which is thought to be the same as the thin superficial layer of ear-
liest stages, is somewhat diminished in thickness. It is seen to be spread
over the surface" of the cells, and to separate the spheres of segmentation
from each other.* A day after the eggs were laid they had developed
into free swimming spheres, PI. I. fig. 10, covered externally with cilia.
These larvse were found in great numbers free in the jars. The egg has
developed into a larval stage, which has burst the imprisonment of the
capsule, and the blastomeres have arranged themselves on the periphery
of a hollow sphere. The superficial layer of cells, still more transparent
than the profound, bears long vibratile cilia ; the larva moves readily from
place to place.

At one pole of this larva the blastoderm, or that shell of cells which
encloses the cavity, is slightly thickened and more deeply colored than
the remainder. This pole is the pole where the invagination to form
the archenteron takes place, and it may be said at once that this pole
is the seat of the next important change in the growth of the larva.
At this point, PI. I. fig. 11, ach., the blastoderm begins to fold inward,
forming an invagination, which later becomes the stomach. The posi-
tion where this infolding begins is the pole at which the mouth of the
gastrula, or, is later situated. At this time in the career of the young
Ophiopholis it begins to depart widely in form from that of the genus
Ophiothrix, as figured and described by Apostolides. The segmentation
of the egg is very similar in the two genera, but the form of the blasto-
pheres is somewhat different. The blastodermic cells are very much

* This condition of the plasmic cortical layer in stages of cleavage as old as the
8-cell condition, is thought to indicate that the layer may be something more than
a vitelline membrane, although the vitelline membrane may be formed from some
portion of it. It is not possible for me to arrive at any good interpretation of the
homology of this structure.

VOL. XII. — NO. 4. 8

114 BULLETIN OF THE

more elongated and conical in Ophiothrix than in Ophiopholis, and the
thin superficial layer of cells bearing the cilia is not represented in
Apostolides' figures.' The cavity of both is hollow. In Apostol ides' ♦
figure of Ophiothrix we have in the middle, cell-like structures lettered,
es. He does not explain the lettering, but from the fact that he speaks
of the cavity as " creux," it is supposed that this region is a cavity, the
segmentation cavity. In a copy of this figure in Embryological Mono-
graphs t A. Agassiz letters the cells of the blastoderm ; e, ectoderm, and y
'• yolk cells." The structures y are the same as es.

In a comparison of our figures of a blastosphere, PL I. figs. 10, 11;
with that of Apostolides, we see in both a slight infolding of the blasto-
derm, which is here regarded as the beginning of the invagination in
both cases. Apostolides does not so consider it in Ophiothrix, but he
ascribes to Balfour the mistake of considering it an infolding. He says :
" C'est peut-etre ce point que M. Balfour, qui n'a pas pousst^ tres loin ses
observations, a pris pour un commencement (^invagination. II n'en est
pourtant rien, la suite prouvera que ce point n'est que le premier indice
de la formation des bras du pluteus." It is a significant fact that just
between this stage (his fig. 9) and the stage which he figures in fig. 10,
when cal6areous rods are developed, is the time when the process of
invagination occurs. I find no stages of Ophiopholis which resemble in
shape his figures 10 and 11 of Ophiothrix.

Apostolides says : " Peut-etre M. Balfour a-t-il obtenu des f^conda-
tions de VOphiotrix rosula, qui est plus abondante en Angleterre, et
chez laqueUe les choses se passent peut-etre autrement que dans I'espfece
que nous avons souraise k I'observation." It would be an interesting
fact if one species of Ophiothrix forms a gastrula stomach by infolding,
and another in the way described by Apostolides. J Closely related star-
fishes, sometimes regarded as simply different species, however, have a
wide difference in their development. A. vulgaris has a brachiolaria,
while Leptasterias has young without nomadic stages. The gastrula of
the latter may or may not develop as that of the former. There is noth-
ing to show that it is exceptional.

The " plan general " of the development of the gastrula of Echinoderms
is more widely soread among Echinoderms than the following quotation

* Op. cit, PI. XI. fig. 9.

t Mem. Mua. Comp. ZoUl, Vol. IX. No. 2.

X We are here brought face to face with one serious defect in Apostolides' and
Balfour's observations, namely, the difllculty of knowing exactly the species which
both studied.

MUSEUM OF COMPARATIVE ZOOLOGY. 115

from Apostolides' paper would seem to indicate : ** Le jugement de M.
Balfour repose sur de simples probabilites de ressemblance avec le type
Holothurie, dont I'embryologie lui sert comme plan general de tons leg
Echinodermes." Since we know that the formation of a gastrula has
been observed also in starfishes and sea-urchins, it would have at least
been more just to Balfour to have inserted these types after that of the
Holothurians in the above quotation.

It was noticed at the close of the first day that the thickened blasto-
derm begins to fold inward at one pole, and at the same time that the
blastoderm at that point becomes more densely pigmented. The larva,
PI. I. fig. 13, is now pear-shaped, slightly flattened on one side and trun-
cated at the pigmented pole. The flattening on one side is the first
indications of the ventral surface, and one of the first expressions of a
bilateral symmetry which later becomes so well marked by the growth of
mesoblastic cells. The internal surface of the cells at tlie truncated pole
bulge somewhat into the cavity of the blastosphere, and from it meso-
blastic or amoeboid, spherical, and star-shaped cells, ad, begin to bud.
These cells form in two lateral* clusters, PL I. fig. 14, and indicate at
once the position of the infolded archenteron. They are the beginning
of a middle layer, and from them many important structures form. The
least diameter of the larva is .11 mm. ; the greatest .13 mm.

The same irregular triangular form, and the clustering of pigment
about the blastopore seems to be found in the gastrula of Ophiophragraa.
It is the presence of this pigment on each side of the gastrula mouth
which has been of assistance in the identification of the lateral arms, /, in
later stages as compared with the blastopore. A clustering of pigmented
cells at the lower extremity of the stomach has rendered it extremely
difficult for me to study the changes which go on in the formation of the
water tubes and other structures in this region of the embryo. The
walls of the stomach are yellow and green. Metschnikofi't found it
very difficult to observe the " Mesoderm formation " in Ophiothrix fra-
gilis, which he was able to artificially fertilize.

It is supposed that our embryo can be compared with that of Am-

* There is already a considerable literature on the question of whether in
Echinuderms the " Mesodermkeiin " or " Mesoderm cells " arise in a bilaterally
symmetrical manner as regards a " spaltartige Rinne " of the gastrula, by which
the symmetry, is early indicated. Selenka and others hold that they do; Metsch-
nikoff, that they do not. My observations show such a symmetry in the mesoblas-
tic cells of Ophiopholis.

t Zeit.f. Wiss. Zool., XLII. p. 664.

116 BULLETIN OF THE

phiura by Metschnikoff (op. cit. PL III. fig. 6),* and that the two struc-
tures v., supposed by him to be water tubes, correspond in position to
the chisters of cells on each side of the invagination. These clusters in
Ophiopholis were quite dense, and the vesicles, if they existed here,
would be difficult to see. Metschnikoff says that in Amphiura these
bodies are also difficult to see through the " Cutiszellen " (mesoderm
ceUs), and that later in normal development one is lost. He was able to
observe that one of these bodies in Amphiura develops into the water
tubes of the adult. It is not wholly certain that similar bodies do not
exist in Ophiopholis, PI. I. fig. 14, a cl., where clusters of amoeboid cells
make observation on live material somewhat difficult at these points.

The bilateral arrangement of tlie budding cells in the cavity of the
blastosphere and the shape of the larva give to it a marked bilateral sym-
metry even at this early stage. The pole of invagination may be called
an anterior pole, while the cells on each side indicate the sides of the
larva. One hemisphere of the gastrula is flattened ; the opposite is
more rounded. The former may be called the ventral, the latter the
dorsal surface.

At seven o'clock on the day following the spawning the invagination,
which forms the archenteron, has extended about half way down the
cavity of the blastosphere, PI. I. fig. 15. Almost the whole of the
second twenty-four hours is occupied by the changes which accompany
the infolding of the archenteron. f

The pole of the infolding slowly sinks into the cavity, carrying with it
at this point the shell of cells, or that part of the blastoderm which is to
form the wall of a digestive canal. The larva has become very much
flattened on the ventral side, so that when seen from the pole of invagina-
tion the lateral diameter is twice that at right angles to it in the same
plane. As we have arbitrarily called the longest diameter, when seen
from the pole of the blastopore, the lateral, a name which seems appro-
priate, not only on account of the bilateral symmetry which the larva at
this early age h^ 3, but also from the fact that from its extremities form
the two calcareous rods and fleshy arms, known as the lateral arms, we
may speak of the other diameter as the dorso- ventral. The dorso- ventral
diameter connects the dorsal and ventral side of the larva, which are

* Fig. 6 is a little older. The mode of origin of these vesicles was not observed
by MetschnikoflF. Their position relatively to the mouth of the larva is somewhat
exceptional.

t The time occupied to form the gastrula of Ophiophragma is about the saiae as
in Ophiopholis. Cf. Nachtrieb, op. cit.

MUSEUM OF COMPARATIVE ZOOLOGY. 117

readily distinguished from each other. The ventral side is quite flat, the
dorsal more couvex.

Looking through th'i larva with its infolded outer layer of cells or
hypoblast, from the ventral side, we notice that the infolding has pro-
ceeded about two thirds the axial length of the larva, and formed a fun-
nel-like tube. This tube is the hypoblast, the primitive stomach, and
at the pole of infolding is situated a mouth, or. The whole larva, PI. I.
fig. 1 6, is now in the gastrula stage.

At the pole of ^invagination in the region of the coeloraa, between the
infolded walls and the external crust of cells, epiblast and hypoblast, two
masses of cells, a cl, are situated, one on each side, which are the
mesoblastic cells already spoken of. These cells are spherical, stellate,
branched, or elongate. The walls of the anterior pole of the gastrula are
more densely pigmented than the remaining parts of the larva. The
pigmentation is most dense on each side of the mouth. When the same
gastrula is seen from one side, PI. I. fig. 15, it is noticed that the infolded
archenteron does not hang exactly in the longer axis of the larva, but that
the closed end approaches the ventral side. Its extremity has a tendency
from the very first to draw near the ventral wall. It approaches so near
that it may be supposed to be met by a second infolding, through which
an opening may be formed. I have not observed this second invagina-
tion, or this opening to be formed ; although the general law of Echino-
derm development would call for such an occurrence. I did not observe
a second opening to be formed in the larvae of Ophiopholis.*

On the second day, PI. I. fig. 16, after the fecundation of the Ophio-
pholis, it was observed that the invaginated end of the stomach becomes
somewhat inflated, PI. I. fig. \Q, g a, by an enlargement of the cavity.
Although this inflation has not been traced farther, and water tubes
were not seen to arise from it, as we know takes place in the course of
Echinoderra development, up to this point the modifications in this
region of the archenteron closely resemble similar formations observed
by others in the echinoid pluteus. The origin of the water tubes from
the primary invagination is yet to be observed in Opliiurans, notwith-
standing from a priori grounds we suppose such to be the case. All
embryologists, however, do not accept such an explanation. According
to Apostolides,t who has written the last important work on the devel-

* The clustering of cells in the cavity of the larva made accurate observations
in regard to the changes wliich occur at iliLs point very difficult. Nachtrieb seems
to have had a similar difficulty in the genus, Ophiophragma.

t Op. cii., p. 199.

118 BULLETIN OF THE

opment of Ophiothrix, " Ces deux masses cellulaires ne sont pas, comma
I'a suppose M. Balfour, dues k des diverticulum de I'archenteron, sem-
blables aux cavites vaso-periton^ales des Holothurins, mais elles sout
des produits d'une formation directe, comme cela a lieu pour I'estomac."

PI. I. fig. 18, shows a larva slightly older than the gastrula last
described. If we look at this larva from the flat or ventral side, we
notice on each side of a single opening * small pigmented protuberances.
These prominences in the future growth of the larva become more and
more extended, and even in their earliest form give evidence that they
develop into the lateral arms of the pluteus. The larva is now three
days old, and has begun to assume a form like the youngest Ophiuran
pluteus described by others. The longitudinal axis is .18 mm. ; the dis-
tance from the tip of one lateral prominence to the opposite is .16 mm.

Tne anal lobe has grown more pointed than in the larva just described,
the body of which is about spherical. The oral lobe is smaller than the
anal, although similar ro it in form. It is as yet undivided. A mouth
leading into a cavity opens on the upper pole on the ventral side of the
oral lobe, and a broad band of cilia extending along the lateral arms
surround the mouth, the oral lobe, and the ventral region of the body.
The opening thus surrounded by a ciliated band is easily seen. Its lips
are richly ciliated.

The limestone rods have already been formed in the body, PI. I.
fig. 19. There are two centres of formation of these bodies; but these
centres of calcification are at first not joined. The limestone rods, sp,
originate as spicules with three prongs. One prong extends into the
lateral rod, another in the direction of the anal lobe, and a third into
the oral lobe. Later, a fourth process is formed from the common
union of the three already mentioned, which extends to the middle line
of the dorsal side.

The amoeboid or mesoblastic cells are formed throughout the region
of the larva, between the epiblast and hypoblast. They are spherical,
sometimes branched, forming suspensoria connecting the wall of the

* Still further observation is necessary to show whether the mouth of the gas-
trula of Ophiurans becomes the mouth of the pluteus, or whether, as in Holothu-
rians (Cucumaria,) Selenka, the Starfishes (Asterias vulgaris,) A. Agassiz, (Aster-
ina,) Ludwig, and in some Echinoids (Strongylocentrotus) according to Krohn and
A. Agassiz, the gastrula mouth becomes a vent. I have not observed an anus in tlie
pluteus of Ophiopholis. The single opening is, therefore, supposed to be the gas-
trula mouth. Whether, as in some other Echinoderms, a new opening is formed,
and the gastrula mouth becomes an anus or not in later stages, was not observed.

MUSEUM OF COMPARATIVE ZOOLOGY. 119

stomach and ttie outer wall of the pluteus. The oral ciliated band is
not as transparent as the anal lobe of the pluteus. Stomach walls and
oral ciliated belt have a yellowish-green color.

In the oldest plutei which we shall mention, figs. 21-23, the larva
has assumed a triangular profile when seen from the ventral side and
the two lateral arms, II, have pushed out on each side. The anal lobe is
slightly pointed ; the oral, ol, well developed, undivided, and rounded.
The whole external surface is ciliated. The oral band of cilia is indi-
cated by a closer approximation of the cells of the middle layer. The
distal ends of the posterior rods are pigmented. The body of the plu-
teus is surrounded by a superficial transparent layer of cells. The
mesoblastic walls of the arms are crowded with granules.*

The rods which form the supports of the lateral arms have length-
ened to keep pace with the growth of the arms. These rods are not
latticed. The rods of the anal lobe are bow-shaped, and at the apex of
the anal lobe they bifurcate, the larger division extending to the apex
of the lobe.

The anterior rods are smooth, and extend half way down the lobe, or
in some cases to the ciliated oral baud. The mouth, oesophagus, and
stomach are well differentiated from each other. The hypoblastic walls
of the latter, ga, are green and yellow.

The oldest pluteus is a little more than tnree days old. On the
fourth day I left Eastport, and all my plutei died from want of care.
It is probable, however, that they are hardy, and can be easily raised,
and the young Ophiopholis traced from them to its adult.

The following summary of the preceding observations may be
made : —

1. OjoAe'ojoAo^w acM^ra^a has a development with metamorphosis, pass-
ing through a larval stage called the pluteus.

2. The ova are laid in the surrounding water. The yolk has a cen-
tral and a peripheral region, which is distinguished in the 8-cell and
previous stages of segmentation. The cleavage is like that of other
Echiuoderms.

3. A gastrula is formed by an mvagination of the blastoderm, and
consequently the stomach of the pluteus is an infolded wall of the blas-
toderm, and not formed by delamination from the cells in the cavity.

4. The mesoderm cells originate in two lateral clusters.

* MetschnikofE accurately represents, op. ck., PI. V. fig. 2, an Ophiuran pluteus
which has the cells " cutis " crowded in the lateral arms in the same way as in
Ophiopholis.

120 BULLETIN OF THE

ECHINARACHNIUS PARMA Gbay.
General Notice.

Our knowledge of the development of Echinarachnius is small.
Johannes Miiller* long ago described a pluteus which he referred to
Echinocyamus. From its likeness to the pluteus described by Miiller,
■which is a very characteristic one, A. Agassiz suggested t that the com-
mon Newport pluteus is the young of Echinarachnius. The pluteus of
Arbacia is known, that of Strongylocentrotus X is characteristic, and
Mr. Agassiz was led to refer a pluteus, which is neither of these, and
which is found in great numbers in Narragansett Bay, to the young
of Echinarachnius. No one has up to the present brought forward any
observations bearing on this suggestion. I have raised the egg of Echin-
arachnius into a pluteus, which is closely allied to his, and have raised
plutei which are identical into a young stage of Echinarachnius. The
plutei described by A. Agassiz are not mature. A. Agassiz has also
figured § the young stages of Echinarachnius after the absorption of the
pluteus. In a paper on the embryology of the genus Arbacia, I have
described II the peculiar pigmentation on the viscous covering of the egg
of the Echinarachnius while in the ovary.

These contributions constitute the greater part of our knowledge of
the development of Echinarachnius.

The development of the pluteus of the " sand-cake " or " sand-dol-
lar," IT E. parma, resembles in many respects that of Arbacia.** The

* Ueber die Gattungen der Seeigellarven. Siebente Abliandlung iiberdie Meta-
morphose der Echinodermen. Abh. k. preus. Akad. d. Wiss. Berlin, 1855.

t Revision of the Echini. Mem. Miis. Comp. ZdoL, III. p.~730.

t The pluteus of Strongylocentrotus must be rare at Newport. I have not
recognized it in friy fishing there in several summers.

§ Op. cit., PL XII. Embryological Monographs, No. 2. Mem. Mus. Comp. ZooL,
Vol. IX. No. 2.

II On the Development of the Pluteus of Arbacia. Mem. Peabody, Acad. Sci.,
L 6, 1881.

IT Many genera of Clypeastroids, besides Echinarachnius, are also called sand-
dollars from the shape of the adult. In the South Mellita bears that name.
Eciiinarachnius is sometimes called the sand-cake, in New England coast towns.

** For a history of the development of Arbada see A. Agassiz, Revision of the
Echini, pp. 729, 733-735. E. Selenka, Keimblatter u. Organenlage der Echini-
den. Zeit. f. Wissensch. ZooL XXXIII. PI. VII., Figs. 34-37. J. Walter Fewkes,
On the Development of the Pluteus of Arbacia. Mem. Peabodi/ Acad. Sci. L

MUSEUM OF COMPAKATIVE ZOOLOGY. 121

adult plutens is very different from that of either Arbacia or Strongylo-
centrotus, and most closely resembles the pluteus described by Miiller
as that of the genus Echiuocyamus.

Artificial Fertilization.

The sexes of Echinarachnius are distinct, the male and female organs
being found iu different individuals. Although the colors of the adult
of different specimens vary, and in some instances it was jjossible for
me to tell the sex without dissection, this could not be done in all
cases. The colors of the ripe glands, ovaries and spermaries, can easily
be distinguished. The former are commonly dark-red or purple ; the
latter orange or yellow.

Derbes* was not able to distinguish the male from the female of
E. esculentus by external characters. The sperm according to him has
a milliy white color, and the ova are orange or brown.

The males and females of »S'. druhachiensis, according to A. Agassiz.f
are distinguished by a " more vivid coloring of the spines of the latter,
which are of a violet' tinge, while those of the males are more yellowish-
green." The ova and sperm of Strongylocentrotus, he says, resemble
in color that of E. esculentus as described by Derbes.

My method of procedure in artificial fecundation is as follows : The
apical portion of the aboral region is incised through the test by a ring-
shaped cut, with a radius equal to that of the petaloid openings. This
dissection is carried on with the sea-urchin under water. The incised
part is turned over, and transferred to a glass dish with water, and the
remainder of the animal is placed in pure sea-water.

Upon the inner surface of the incised part fragments of the ovaries
will be found, if the specimen is a female, and spermaries if a male. In
the former case, if the eggs are mature, small transparent globules will
be found to float away from the glands, especially if the organ is slightly
washed with a pipette. If a white fluid exudes from the glands the
specimen, if alive, is probably a male, and the white fluid is colored by

No. 6. H. Garman and B P. Colton, some Notes on the Development of Arhncia
punctulata Lam. Studies Biol. Lab. Johns Hopkins Unit-. IL, pp. 247-255, and W. K.
Brooks, Handbook of Invertebrate ZoiJlogy for Laboratories and Sea-Side Work,
figs. 78-83.

* Observations sur le Mecanisme et les Phcnomenes qui accompagnent la For-
mation de rEmbryon cliez I'Oursin comestible. Ann. Sci. Nat. [3] VIIL 1847.

t Revision of the Echini, p. 708.

122 BULLETIN OF THE

sperm. Many eggs can be washed out of the undissected sea-urchin
with a pipette introduced through the aboral region. The floating eggs
and the milky sperm are mixed by simply pouring the water from one
jar to the other. A better method of artificial fecundation is to collect
a watch glass crystal full of eggs, leaving enough water for them to
float, and then to drop a few drops of water charged with sperm among
them. The contents are then gently stirred, and after a short time
evidences of the success of the process may be looked for. I have
found that chopping up the two glands together, although in some cases
to be recommeuded, in most instances, and especially in the case of
Echinarachnius, the egg of which is delicate, leaves so much decaying
matter that the ova are killed. It is well not to put too much water
with the ova, as repeated dilution renders the collecting of the ova for
study difficult. I took no precautions about the temperature of the
water, and did not find it necessary to change the water until after
segmentation was finished.* Artificial fecundation was accomplished
from the middle of July to the end of August.

Each ovum, PI. II. fig. 1, is visible to the naked eye. It is sur-
rounded by a viscous 1 layer in which are beautiful, spherical or some-
times angular, red pigment spots, pig, which are supposed to correspond
to the "clouded areas," described by A. Agassiz t in the star-fish egg.
The viscous layer of the egg of E. esculentus is described by Derbes.
A. Agassiz describes a ** thick homogeneous structureless shell " in
Strongylocentrotus. The pigment spots are conspicuous on the outei
surface of the viscid capsule of the egg of Echinarachnius. After fer
tilization the ova sometimes sink and sometimes remain floating. Theii
specific gravity is about that of the water.

The diameter of the yolk, t;<, is .13 mm. The diameter of the viscid
covering is from .22 to .25 mm. The yolk is yellow ; the envelop trans-
parent. The yolk was not observed in the free egg to fill its capsule in
any stage or segmentation.

A nucleus and nucleolus were observed in ovarian eggs. These
structures were difficult -to see in free eggs.

The spermatozoa immediately after the mingling of the two sexual

* Selenka and others have already pointed out refined ways of fertilizing sea-
urchin eggs. See Selenka, " Keimblatter und Organenlage der Echinodermen."
Zeit.f. Wiss. Zool, XXXIH. p. 40.

t Similar pigment spots are found according to Nachtrieb in the egg of Mellita.
These spots on the ovum of Echinarachnius were first described in my paper on the
development of Arbacia. Mem. Peahody Acad., I. 6.

MUSEUM OF COMPARATIVE ZOOLOGY. 123

elements were observed crowded upon the outer surface of the viscid
layer, with heads partially buried in it. None were observed to pene-
trate to the yolk. The egg was seen to be slightly jerked about, possi-
bly by the combined movement of the many spermatozoa on its surface.
In no case was the movement very great. No polar globules were
observed.*

Cleavage.

The segmentation of the egg of Echinaracbnius is regular, and the
first formed segment spheres are of the same size. After the formation
of the 8-cell stage from the 4-cell an inequality in size of the blastomeres
is noticed. As in Strongylocentrotus one of the first changes after the
disappearance of the nucleus is the drawing away of the yolk from the
shell. From an hour to two hours after the ova and sperm have been
artificially brought together, the first cleavage furrow, p, is noticed
encircling the egg.

In some eggs this furrow, PI. III. fig. 1, is limited to one pole, and
the indentation gradually deepens until the e^,g is divided into two
hemispheres connected at the pole opposite that at which the fur-
row first appears. Folds which recall similar plications observed by
Metschnikoff in the Epibulia egg, and by myself in the egg of Agalma,
appear on each side of this primitive furrow, PI. III. fig. 2. These
wrinkles are supposed to be the " Faltenkranzen." This method of seg-
mentation reminds me of what we have in the egg of the Sinhonophore.
It was not traced beyond the 2-cell stage.

In most cases the primitive furrow is not limited to one pole, but
girts the ovum. Four cells were, however, observed in a 4-cell stage, in
each of which the furrow, which is to form a new cleavage plane, is
limited to one pole of the cell. PI. III. figs. 6, 7.

In those ova in which the primitive furrow girts the eg^,, the con-
striction deepens uniformly on all sides, until the ovum is divided into
two equal spheres, PI. II. fig. 3, united by flat faces with each other.
In each of the two cells a nucleus can be seen. The blastomeres of the
2-cell stage are never seen widely separated from each other.

* According to Nachtrieb no polar globules were observed by him in the closely
allied genus Mellita. I suspect, as is well known in some other Echinoids, that
the polar globules are formed while the egg is in the ovary.

t More than one method of cleavage has been observed in the Oyster by Brooks,
and in Reniila by Wilson. It is not improbable that the segmentation of Echin-
aracbnius mentioned above is a second kind of cleavage.

124 BULLETIN OF THE

The cleavage plane, 1 dpi, whicti divides the ovum into two segments,
may be called a meridional plane. It is the first cleavage plane. In
the formation of the 4-cell stage the two segments already formed are
divided by a plane at right angles to this, and the mode of division in
the two is identical. The division of the two cells which form the
2-cell stage begins by a slight constriction, girting the spheres which
later bisects them, forming four smaller nucleated spheres or blas-
tomeres, all of the same size. There is no 3-cell stage in this kind of
cleavage. The second plane of cleavage divides both cells of the 2-cell
stage.

The formation of the second cleavage plane will thus be seen to differ
from that of Asterina, as described and figured by Ludwig.* In Asterina
the two cells of the 2-cell stage are of unequal size. The smaller of
these divides first, so that we have a 3-cell stage, fig. 2 (op. cit.).
In Echinarachnius both the cells were observed to divide at the same
time and form the 4-cell stage. The cleavage plane which forms the
4-cell stage (2nd cleavage plane) in Echinarachnius is at right angles to
the first, and identical in its position in each cell of the 2-cell stage.
Ludwig, p. 6, op. cit., says of Asterina : " Die Theilungsebene der
beiden Zellen II. (larger cell of 2-cell stage) ist aber nicht etwa die
auf die Zelle II. iibergreifende Theilungsebene der Zellen I. (smaller
cell of the 2-cell stage), sondern bildet mit letzterer, so wie auch mit
der Theilungsebene der beiden ersten Furchungskugeln einen rechten
"Winkel." Three cells were observed abnormally formed in the ovum of
of Echinarachnius, and their mode of formation is traced below.

In Strongylocentrotus, according to A. Agas8iz,t after the yolk sepa-
rates from the inner wall of the outer envelope, it is slightly depressed
on one side, and a similar change soon after occurs on the opposite pole.
After these depressions in the poles of the yolk of Strongylocentrotus
occur, a slit is formed, according to A. Agassiz, which divides the egg
into two large elliptical masses.

In the egg of Echinarachnius in normal cases a constriction was
observed, PI. II. fig. 2, girting the yolk, similar to fig. 23, p. 709, of
the work last mentioned.t This constriction deepens uniformly on all
sides until the 2-cell stage is formed. In several eggs of Echinarach-
nius, PI. III. figs. 1, 2, 3, the 2-cell stage is formed in another way.

* Entwickelungsgeschichte der Asterina gibbosa, Forbes. Zeit. f. Wi&s. Zool.,
XXXVII. pp. 6, 7.

t Revision of the Echini, p. 710.

MUSEUM OF COMPARATIVE ZOOLOGY. 125

A furrow appears at one pole. This furrow penetrates into the yolk,
forming in profile a slit-like structure, which in this way divides tlie
yolk into the 2-cell stage. In Strongylocentrotus, figured by A. Agassiz,
we notice that a flattening of each cell of the 4-cell stage occurs prepara-
tory to the passage into the 8-cell stage. This flattening occurs on one
side at first (p. 710, fig. 27). Several eggs of Echinarachnius, PI. III.
fig. 6, were taken in a similar condition. In many others, however,
each of the four cells of the 4-cell stage is divided from the very first by
a constriction reaching wholly around the cell, PI. II. figs. 4, 8.

In several ova the following modification of development was observed
after the 4-cell stage. Au egg was found in the 4-cell stage apparently
normally formed. Pi. II. fig. 9. Immediately after two of the spheres
begin to fuse, and the wall of the cleavage plane separating them is
broken down. In this way we pass by retrogression from an egg with
four, PL III. fig. 9, into one with three segment spheres, PI. III. fig. 12.
Whether the many eggs in a 3-cell stage which were observed were all
formed in this manner or not, cannot be stated. It was not observed
how the 4-cell stage in this abnormal mode of development is formed
from the 2-cell stage. Segmented ova with three segmentation spheres
are quite common in some trials for artificial fecundation.

An egg fertilized at noon was found in the 2-cell stage at 1.30 p. m.,
and passed into the 4-cell stage at 2 p. m. At 3 p. m. it was in the
8-cell stage. We can, therefore, roughly say that the formation of a
fresh cleavage plane occupies approximately an hour's time. By a
comparison with the rate of growth of the stai'fish it will be seen that
the rate of development of Echinarachnius is more rapid. The water
in which my eggs were kept was evidently warmer than that in which
Strongylocentrotus was reared.

The mode of formation of the 8-cell stage from the 4-cell does not
differ from that of the 4-cell from the 2-cell, The segments of the
4-cell stage are, however, not always bisected, and here appears the
first indication of an unequal segmentation. The spheres of the egg
even in the 8-cell stage have a peripheral tendency. In the 8-cell stage
it will be noticed, PI. II. fig. 11, that the eight spheres cannot be so
brought together as to touch each other on adjacent sides. A recess,
cav, is thus early left, which later forms in the interior of the ovum a
" segmentation cavity." This cavity increases in size as the size of the
segmentation spheres diminishes in the progress of segmentation. An
egg in the 32-cell stage was found four hours after impregnation, PI. II.
fig. 12.

126 BULLETIN OF THE

The whole process of cleavage occupies about ten hours.* A rota-
tion of the spheres of segmentation according to A. Agassiz occurs in
Strongylocentrotus. This was not observed in Echinarachnius. Through-
out all the changes the egg is enclosed in the capsule, cap, which has
been mentioned in the unsegmented egg.

Shortly after the end of the first half day after fecundation, the blas-
tomeres arrange themselves superficially about the segmentation cavity,
forming a hollow sphere, which is the blastosphere, PI. II. fig. 14. Mi-
nute cilia, which ai-e long and fine, appear over its surface, and the egg
begins to rotate and fret against the sides of the envelope or egg capsule,
which closes it in. There is no solid morula stage; but a true blastula
is immediately formed. At this time a thickening of the blastoderm at
one pole takes place, the outline becomes more pyriform, PI. II. fig. 15,
and at the truncated pole a collection of pigment of deeper color than
in the remainder of the ovum congregates. This increase in thickness
of the cells at one pole is indicative of the formation of a gastrula
mouth at that pole. Immediately after the thickening of the blasto-
derm an infolding begins to take place at this pole, PI. II. fig. 1 6. By
this infolding, ga, the layer of cells which form the walls of the cavity,
or the blastoderm, are infolded, and form the hypoblastic layer, or walls
of a gastrula stomach. The infolding is at first very slight, but the
increasing age of the embryo carries the walls deeper and deeper into
the cavity.

With the first indication of an ingrowth of the gastrula stomach, or
archeuteron, we find budding off into the segmentation cavity certain
cells, a cl, which from their form, position, and other characters, are
called the amoeboid or mesoblastic cells. They give rise to important
structures, which later appear in the embryo, between epiblast and
hypoblast, and which belong to the middle layer or mesoblast. Prouho t
finds in Dorocidaris that these cells are not all the same. When his
paper came into my hands it was too late to verify in Echinarachnius
what he finds in Dorocidaris. At the time my observations were made
all the so-called amoeboid or mesoblastic cells were regarded as the same
in character, and although I supposed that they did not all form the
same structures, their differentiation in form was thought to take place
much later than in the gastrula stage. These cells form on each side of
an axis, passing through the gastrula mouth or primitive infolding.
Their bilateral arrangement was not as marked as in Ophiopholis. They

* Kate of growth in water of unrecorded temperature.
t Comp. Rendus, ci. pp. 386-388.

MUSEUM OF COMPARATIVE ZOOLOGY. 127

form among other structures the calcareous rods and the euspensoria,
filiform bodies which connect the hypoblast and epiblast. In Echinus
miliaris and Toxopneustes, according to Selenka (op. cit, p. 46), they
also form certain muscles of the stomach and intestine.* I cannot
at present say whether these mesoblastic cells originate from the hypo-
blast alone, or from the epiblast as well, in the genus Echinarachnius.
It seemed to me that they arose from a neutral zone on the region of
the blastopore. This zone or region, from its position at this time, is
either epiblast or hypoblast, or both. As, however, the hypoblast is
formed of infolded cells, which elsewhere are later epiblast, we might
say that cells originating from this neutral zone are strictly derived
from the epiblast. The observations of several naturalists are at vari-
ance on this point, as far as the gastrula of other Echinoids is concerned.
Selenka holds that iu Echinus microtuherculatus, Sphcerechinns granvr
laris and Arbacia pu^ttdosa the mesoderm cells spring from the hypo-
blast. Other naturalists, as Greef, MetschnikofT and Bergh, derive
them from the epiblast as well, in these and other genera.

Ludwig, who has reviewed the diflferent observations of the embry-
ologists who have studied the question, concludes that in general the
mesoderm cells arise from- the hypoblast, but that " auch aus dem Ekto-
derm sich Zelleu abschniiren und zu Mesodermzellen werden konnen."

No special observations were made on the character of the contents
of the segmentation cavity, and the space between hypoblast and epi-
blast in the gastrula. Ludwig f regards it as filled with a liquid
through which the mesoblastic cells can move in Asterina. This seems
more natural than to regard it with Hensen X as occupied by " Gallert-
kern " or any fixed gelatinous structure.

Gastrula.

As the infolded blastoderm or hypoblast pushes its way in the form
of a pouch into the segmentation cavity, it changes its form from a
simple infolding to a funnel-shaped tube, the parts of which are at first
undifferentiated. The primitive opening, blastopore, or gastral mouth,
PI. IV. fig. 1, gm, would seem to serve as both mouth and anus, since
there is no other communication with the outside water. Krohn says

* Tlie question what structures in the Echinoderm pluteus these cells form is
a complicated one, and has been variously answered.
t Op. cit. p. 14.
t Arch./. Xaturg. 1863.

128 BULLETIN OF THE

that the gastral mouth serves for reception of food until the formation
of the real mouth (second opening). Salenka* says of the view of
Krohn : " Ich kaun dieser Ansicht nicht beitreten die nach aussen
schlagenden Geisselendes Urdarms scheiueu den Eiutritt von fremden
Stoifeu durchaus zu verhindern."

A. Agassiz says that in the starfish and Strongylocentrotus gastrula
currents of watef enter the mouth, pass into the stomach, and pass out
through the same opening. The gastrula mouth in these instances cer-
tainly serves as both mouth and anus.

Food was not seen to enter the mouth of the gastrula of Echinarach-
nius, and no observations were made on currents of water. The open-
ing of the blastopore has probably the same function as the homologous
opening in Asterias and Strongylocentrotus.

We find that the infolded funnel now becomes enlarged at the base
into a chamber, and is attached to the outer wall of the embryo by sus-
pensoria or filamentous bodies derived from the mesoblastic cells. Ex-
teriorly the larva is truncated, flat on one side, more rounded in the
diametrically opposite region. It is ciliated with fine long cilia, those
on the pole opposite the blastopore being prominent. These longer cilia
may be the same as the tuft opposite the blastopore mentioned by Pro-
fessor Xachtrieb in jSIellita, and by Prouho f in the gastrula of Doro-
cidaris papillata. The morphological importance of these cilia has been
magnified, although they may indicate one more likeness between the
well known pilidium and the Echinoderm larva. The invaginated cells
of the hypoblast are cylindrical, ciliated, and not yet differentiated into
the walls of the oesophagus, stomach, and intestine.

In a gastrula one day old, PI. IV. figs. 1, 2, we observe that the
invaginated pouch has extended to the opposite pole of the larva, and
as it lengthens in this direction its free end slowly approaches the flat
side of the gastrula, which side is that known as the ventral. It now
beuds still more to this region, and is met by a corresponding infolding
from the ventral surface. The walls of this infolding bi-eak away, and
form the future anus, v, of the stages immediately following the gas-
trula, and probably the mouth of the pluteus.

In a gastrula in which the opening had not broken through, PI. IV.
fig. 2, it was observed that the gastrula stomach, ach, sends out two
horn-shaped coeca, which are similar to structures in other genera
known as "water-tubes," " Enterocoelen" or "laterale Scheiben." In

* Op. cit. p. 48.

t Sur la forme larvaire du Dorocidaris papillata. Comp. Rend. ci. pp. 386-388.

MUSEUM OF COMPARATIVE ZOOLOGY. 129

PI. V. fig. 2, vp, one * of these " vasn-peritoneal vesicles," now con-
stricted from the enteron, is seen as a closed sac on one side of the
gastrula stomach. The beginnings of the formation of pouches, which
probably form the water-tubes, were observed in Echinarachnius, but
I have not traced them in their later stages of growth. I have ob-
served only one of these constricted off from the gastrula stomach.
The vesicle is separated from the enteron before the "mouth opening"
is formed. The wall of the infolded pouch now begins to differentiate
itself by constrictions into three regions, corresponding with the oeso-
phagus, stomach, and intestine, PI. IV. fig. 5, of the fully grown plu-
teus. At about the same time, also, the limestone rods or calcareous
framework of the pluteus first appears, PI. IV. fig. 3, sp.

The calcareous rods appear on each side of one of the openings into
the gastrula stomach. In Echinus, according to Krohn, the primitive
invagination, or the blastopore, becomes the vent of the pluteus. A.
Agassiz says the same of the gastrula of Strong}'locentrotus. I have no
observation on this point in the gastrula of Echinarachnius, and noth-
ing to show that there is any difference in this genus from what is
recorded in Strongylocentrotus and other Echinoids.

The formation of the two limestone spicules which characterize the
pluteus at this age, takes place in the cluster of mesoblastic or amoe-
boid cells, acl, on each side of the opening, which henceforth serves as
the mouth of the pluteus. The first appearance of the limestone rods
was detected in a gastrula one day old, PI. IV. fig. 3. As in Oj)luo-
pholis, these structures arise in the bilateral masses of mesoblastic
cells,t one on each side of the blastopore or oral end of the stomach.
They are at first disconnected, branched, or stellate, and trifid, resem-
bling small sponge spicules. Around them are clusters of the amoeboid
cells, from which they form.f The neighboring epiblastic wall of the
gastrula is reddish and yellow. It was also noticed that at the lowest
point of the infolded pouch the same color is prominent. At the last
mentioned position the aggregation of cells and pigment renders it very
difficvdt to observe the formation of the external opening. In one
specimen, PI. V. fig. 2, cl, I noticed an infolding of the ventral wall
opposite the lower end of the invaginated stomach of the gastrula,

* According to Selenka, a sincrle vaso-peritoneal sac constricts from the stomach
of tlie gastrula. This sac later divides into the right and left vesieles.

t A. Agassiz has already called attention to the fact tliat the limestone rods are
first deposited in the midst of similar cells, to which he gives the name "yolk
cells." Revision of the Ecliini, p. 712.

VOL. XII. — NO. 4. 9

130 BULLETIN OF THE

which was thought to be the infolding to form the proctodseum of the
last stages of the gastrula. The epiblastic cells, although becoming
thinner by the growth of the mesoblast, are still large and prominent,
while the hypoblastic are still cylindrical in shape. Both are ciliated.
The former layer furnishes the motor cilia of the body ; the latter the
ciliated lining of the stomach.

Immediately after the first stage in the formation of the calcareous
spicules in the larva, which from now on ceases to be a gastrula, the
anterior lobe, al, and the two prominences which form later the poste-
rior arms, pr, begin to push out, and the region in which they form has
a resemblance to the three legs of a tripod. The larva when seen from
the oral or ventral side, PI. V. fig. 5, has on each side of the mouth, in
a plane in advance of this opening, a small prominence, pr, thickly pig-
mented, especially on its distal end, into which extends a rod from the
stellate calcareous rod of each side. In the interval between these rods
a large undivided lobe, ol; bearing the mouth, appears. This undivided
lobe is the anterior or oral lobe, and on its ventral surface is a circular
ciliated opening, or, the mouth. As this lobe grows, the mouth opening
is carried up with it on its side. When seen from one side, so that the
length of the two prominences, posterior arms, and oral lobe can be
compared, it will be found that the anterior lobe is more prominent
than the two posterior arms. In this stage the body of the larva is
nearly spherical in form, and as it rests on a tripod formed of the two
posterior arms and the single anterior lobe, the intervals between these
prominences are easily seen. The anal pole of the larva is pigmented,
and filled with numerous amoeboid and mesoblastic cells. When seen
from the oral pole, we notice that the ventro-dorsal diameter is about
the same as the lateral. The mesoblast is much thicker than either
the hypoblast or the epiblast. A broad band of cilia surrounds the
mouth.

In a pluteus two days old, PI. V. fig. 6, raised from the eg^, we have
the two posterior rods still more developed than before, while the ante-
rior lobe is still undivided. Seen from the ventral side the distal ends
of the two posterior rods diverge in a V-shape, while the posterior face
of the anterior lobe appears rectangular. The opening of mouth and
anus are well seen.

In the interior of the pluteus we notice that the calcareous rods
which support the posterior arms are double, and have not joined to
become latticed. From a point in the body of the pluteus on a level
with the anus these rods join the system of rods of the body. One

MUSEUM OF COMPAKATRE ZOOLOGY. 131

division extends to the apex of the anal lobe on the posterior side.
At the apex of this lobe it subdivides and interdigitates with other cal-
careous arms. At about this time or a little later a strong muscular
band, well seen in adult plutei, connects the anal rods near the anal
pole on the dorsal side. A simple not latticed calcareous rod, ar, bends
downward on each side from the common origin of the posterior rods and
the most posterior anal body rod, and is continued into the oral lobe.
Seen from one side the course of this rod is at first horizontal, until it
is about opposite the junction of the oesophagus and stomach, where it
sends into the anal lobe a stout anterior anal branch, which extends into
the apex of the anal lobe. It there bifurcates, and the divisions inter-
digitate with the corresponding divisions of the other rods. The cal-
careous rod of the oral lobe, or, is smooth and solid. A straight rod
arises from the union of the posterior anal rod with the posterior rod,
and extends to the medial line of the dorsal side, ending near the anal
opening. Orange and yellow pigment is found in patches at the extrem-
ity of the posterior rods, in the anal lobe, and in the anterior lobe. The
larva was at times observed to rest on the glass standing on the two
posterior arms and the anterior lobe.* At about this time a strong
muscular band first appears, which later is very easily seen connecting
the anal calcareous rods near the apex of the anal lobe. The object ot
this muscle is probably to move the two posterior arms, although the
rigid union of these two structures by interdigitation would seem to
make any considerable motion impossible. As the larva matures, the
body becomes more and more helmet-shaped, approaching the form of
the Echinocyamus larva figured by Miiller.

We are struck, in considering the external form of a pluteus, PI. V.
fig. 11, following the last in age, with the size of a protuberance of the
dorsal surface, and the inflation of that region upon which the anus is
situated. The posterior arms, pr, are weU formed, and the oral lobe is
not yet differentiated into the two oral arms. In a larva four days old,
PI. V. fig. 7, we see that the anterior or oral lobe has become notched
at each corner of the free side of the lobe, and two oral or anterior arms
have begun to form. In the stage just before this it was obsen-ed that
the posterior arms are longer than the oral lobe. Now, however, the
oral lobe has increased in length by the growth of the anterior lobe, the
length of which has about equalled that of the posterior rods. With
the growth of the anterior lobe the form of the oesophagus, oe, has kept
pace. The last mentioned organ is now flask-shaped.

* This attitude was probably taken on account of the shallowness of the water.

132 BULLETIN OF THE

It is at about this time, PL V. fig. 12, ia the growth of the pluteus
that we see signs of the formation of two additional pairs of rods, one
of which is known as the antero- lateral, air, the other the antero-inter-
nal. These two pairs of rods have certain points in common, as far as
their mode of formation goes. Neither of them arises from the primi-
tive centres of calcification from which the rods already formed invari-
ably spring, and to which they are fused. Of these two pairs of rods
the anterolateral arise from separated lateral centres, and are latticed,
PI. VI. fig. 2, air. While the antero-internal, PL VI. fig. 4, air, origi-
nate on a medial line forming at a single centre. The rods of the two
arms, air, are simple, smooth, sometimes with small lateral spurs or
teeth and not latticed.

The antero-internal spicule or rod is well formed in the body of the
pluteus before any projection on the surface at the point where it later
appears can be noticed. It arises as a trifid spicule in the basal region
of the oral lobe. As it grows it becomes crescentic, the convexity
turned to the oral lobe, and the two horns extend about parallel with
the anterior rods. A slight spur or median tooth arises from the con-
vex side of the crescent. The first appearance of the antero-internal
arms is marked on the surface of the larva by a projection on each side
of the anterior lobe within the anterior rods. Into these projections,
as they increase in size, the extremity of the two horns of the crescentic
spicule extend.

The antero-lateral rods, al, PL VI. figs. 3, 4, PL VIII. fig. 5, originate
in a different way from the single median centre of calcification of the
antero-internal. They arise, not from a single centre of calcification, but
from two lateral centres. Just above the anterior lobe, in the interval
between it and the base of the posterior arms, a projection forms on the
edge of the plutean body. This projection raises with it a section of the
circumoral ciliated band, and in its interior there forms a pair of rods
which become joined and form a latticed rod, resembling the posterior
rod. The antero-lateral rods are not fused with the other rods, and as
by the growth of the antero-lateral rods little by little the arm equals
in length that of the posterior, pari passu the rod grows without joining
the remaining spicules. It is this freedom of the two systems which
renders it possible for these arms to be moved by the muscles of the
pluteus. The antero-internal system of rods also does not join the
other rods, and is likewise movable, while the muscular fibers which
accomplish this are easily seen near its junction with the anterior arms.

The pluteus is now, PL VI. fig. 4, in about the same stage as that

MUSEUM OF COMPARATIVE ZOOLOGY. 133

figured by A. Agassiz* It is immature, and an important growth takes
place before it acquires the adult form. The pluteus which A. Agassiz
has figured is, according to my observations, about a week old. Eggs
artificially fertilized on July 16 developed into the pluteus, with the
antero lateral and the antero-internal arms just beginning to form, on
July 23. It was not easy to raise these plutei into older conditions,
but in the month of September there was fished from the surface of the
bay with the dip-net a complete series of plutei, which connects the
pluteus figured by A. Agassiz with the adult as here described, in which
all the four pairs of arms are of equal length. As the preceding plutei
were obtained by artificial fecundation, it is not to be supposed that the
fact that mature plutei are found in the middle of September, indicates
that these plutei are a month old. When artificially fertilized, the eggs,
however, were just ready to be laid. If, as A. Agassiz f says is the case
of Strongylocentrotus, the female Echinarachnius lays her eggs, or the
eggs can be fertilized at all seasons of the year, it would be very difficult
to determine the age at which the adult pluteus is attained from noma-
dic larvcC fished at random from the sea.

A larval pluteus of Strongylocentrotus (Toxopneustes) fig. 52, f is
very similar to the stage of a pluteus of Echinarachnius at this age.
In Echinarachnius as in Strongylocentrotus, the antero-internal arms
are just beginning to appear, and although the antero-internal crescentic
spicules have already formed, the arms corresponding to these rods are
still quite small. This larva which was raised from the egg of Strongj--
locentrotus is twenty-three days old according to A. Agassiz.t It would
thus be about two weeks older than my Echinarachnius of similar form,
also reared from the egg.

The adult pluteus, PI. VII. figs. 1, 2, of Echinarachnius, first ap-
peared in great numbers at K'ewport in 1885, on September IG. In
former years they have been found earlier in the season. The older
stages were captured with a dip-net on the surface of the water, both by
night fishing and in the day-time. For a number of years I have kept
a record of the dates when our marine larvpe first appear in numbers,
and find that the adult pluteus of Echinarachnius is most common at

* Revision of the Echini, p. 727.

t Our common sea-urcliin (S. Drobachiensis) matures its genital organs in winter,
according to A. Agassiz. (Revision of tlie Ecliini, p. 709.) February is the month
wlien he onlinarily succeeded witli artificial fecundation. " Tlie sea-urchins spawn
during the wliole year." Op. cit. p. 719.

t Op. cit. p 719.

134 BULLETIN OF THE

Newport, R. I., in the last of September. As the plutei undoubtedly
develop from ova laid by adults, which live within a short distance of
the laboratory, their date of appearance is not dependent upon those
causes to which we very probably ascribe the marked periodicity in
the times when Salpae and those animals which live in the high seas
revisit Narragansett Bay.

Although it is not known whether Echinarachnius lays its eggs in all
months, or can be fertilized at all times of the year, it can be said that
in the last' five years in which I have kept my record of the times when
marine larvae appear at Newport, the greatest number of larval Echin-
arachnius appear in September. None were collected in June, in July
they are sporadic, and the largest number usually came in September.
Every one acquainted with pelagic fishing knows how much variation
there is in the time when pelagic larvae appear, and these statements indi-
cate only approximation. I suggest that they point to the end of August
as the probable time of ovulation at Newport of the Echinarachnius.

The body of the adult pluteus is elongated, rounded on its anal
pole. On the opposite end it is continued into four pairs of arms, all of
which have calcareous rods, w^hile two pairs, pr and air, are latticed.
The latticed rods are the stoutest, and are known as the antero-lateral,
ah, and the posterior, /)r. The latter arise from the posterior side of
the body, and are fused with the system of rods which extends through
the body. A very prominent cluster of dark red pigment cells, ing, is
found near the distal ends of all the rods. Pigmentation of the same
color occurs in small granules along the length of the arms, and on the
body walls. The arms are skirted by a ciliated baud, on the edges of
which are small granulations. The ciliated band connects the two pos-
terior arms on the median ventral line. Laterally from these arms the
same band passes to the antero-lateral rods.

No ciliated epaulettes were observed. In some specimens it was
noticed that the ciliated band in the connection between the two pos-
terior arras on the ventral side was so folded that we have a median and
two lateral lobes in the region of the band placed between the two pos-
terior arms. Something similar is figured in the pluteus of Echinocya-
mus by Miiller, but as Miiller says that ciliated epaulettes are not found
in his pluteus of Echinocyamus, we may conclude that the last men-
tioned bodies are not epaulettes in the closely allied Echinarachnius.
The antero-lateral arms, air, closely resemble the posterior in pigmen-
tation, latticed axes, ciliation, and size. The ciliated band continued
on them from the posterior arms is carried thence to the edge of the

MUSEUM OF COMPARATIVE ZOOLOGY. loo

anterior lobe, passing to the anterior arms. The four arras which arise
from the anterior lobe are called the anterior, ar, and the antero-inter-
nal, air. Both are furnished with a solid non-latticed central axis or
calcareous rod. The anterior rods are listened to the general cal-
careous framework of the body ; the antero-internal are free, united to
the former by muscular attachments. The larva may be compared to
the parts of a chair, the antero-internal rods forming the front legs, the
anterior the hind legs, and the oral lobe the back by which these rods
are joined to the body of the pluteus. The distal ends of the anterior
rods are more widely separated than the antero-internal. The anterior
rods are skirted by a ciliated band continued from the antero-lateral as
mentioned above. They bear a red pigment spot at a short distance
from their distal end.

The last pair of arms to be mentioned, the antero-internal arms, air,
are joined together at one end by the ciliated band which passes upon
their rim from the anterior arms, and is joined between them, just as
the other end of the circuit is joined on the medial line between the
posterior rods. One edge of the junction of the two antero-internal
arms forms one of the so-called lips of the mouth, PI. VII. fig. 2. The
oral opening, or, is placed between this structure, and the ciliated band
joining the two posterior rods or arms. The two antero-internal arms
commonly want the prominent pigmentation of the distal ends of the
other arms. In one or two specimens, however, pigmentation was ob-
served on the distal end of the antero-internal arms. As the calcareous
rods which support the antero-internal rods have a separate origin from
the rods of the body system, they are capable of movement, and are not
fused with the other rods. A number of muscular threads by which
this movement is brought about can be seen in the inner angles formed
by the rod and the neighboring anterior rods, PI. VII. fig. 18. In the
interior of the body we find that the larger part of the body of the plu-
teus is occupied by the stomach. Between the stomach and the epi-
blastic layer which covers the body of the pluteus many cells are found,
some of which are yellow in color. Filaments or threads connect the
hypoblastic wall of the stomach and superficial epiblast.*

* While studying the Embryology of Agalma similar threads were noticed
passing from the epiblast of the primitive hydrophyllium to the hypoblastic lining
of the same. Similar threads are known in many worm larvas. If these struc-
tures are the same as the so-called suspensoria (Selenka) of tlie gastrula, we may
reasonably doubt whether Selenka is right in supposing them derived from the
mesoderm cells. As far as observation has gone there is no reason to doubt that

136 BULLETIN OF THE

The apex of the anal lobe is densely pigmented, and the walls are
supported by highly ramified divisions of the anal rods (posterior and
anterior), which interdigitate and form an intricate network. An anus
is present, and the intestine is elongate or flask-shaped. The mouth is
commonly widely open, circular, with ciliated lips. The CESophagus is
densely ciliated on the interior walls. The pluteus moves from place to
place easily but not rapidly, and is just visible to the naked eye. Tlie
length is .85 mm. ; diameter of the body .20 mm.

One of the most striking differences between the adult pluteus of Echi-
narachnius and A. Agassiz's drawing of that of Strongylocentrotus of the
same age, is the existence of large pigment spots near the distal end of
each arm, while the anterior and antero-internal arms of Strongylocen-
trotus have no 3uch spots, at least of the great prominence which we
find in Echinarachnius. The pluteus of Strongylocentrotus, according
to A. Agassiz, has ciliated epaulettes. These structures are not repre-
sented in the pluteus of Echinarachnius. Like the pluteus of Echino-
cyamus, as figured by M tiller, Echinarachnius has no ciliated epaulettes.
The reseml)lauce of the pluteus of Echinarachnius to that ascribed to
Echinocyarnus is very great. If we compare the figures given by Miiller
and those of the Echinarachnius pluteus here figured, we notice one or
two marked differences between them. The arms of the pluteus are
much longer and larger in Echinarachnius than in Echinocyamus. The
posterior and antero-lateral rods of both genera are latticed. The an-
tero-lateral and antero-internal in both are not connected with the body
system of rods. The anterior lobe before the origin of the antero-
internal arms is longer in Echinocyamus than in Echinarachnius. The
characteristic pigment spots of the ends of the arms of Echinarachnius
are not found in Echinocyamus. The difference of the young from the
adult sea-urchin formed from the pluteus of Echinocyamus has at-
tracted the attention of Miiller. The young Echinarachnius raised
from the pluteus is somewhat different from Miiller's figures of the
young Echinocyamus.*

the snspensoria are mesodermic, as Selenka says. In Agalma these structures
appear to be epiblastic. It may be said, however, that they originate from the
epiblast, just as the mesoblastic cells themselves may originate as simple exten-
sions and buds. The homology, therefore, of the suspensoria and the filament in
the primitive Iiydropliyllium cannot be made out at present. It may be said that
the likeness between the two is great. (For filaments in the primitive larva of
Agalma, see Embryology of Agalma, Bull. Afus. Comp. Zool., XI. No. 11.)

* It is taken for granted that tlie pluteus described by Miiller is an Echinocy-
amus, although he did not raise it in the egg.

MUSEUM OF COMPARATIVE ZOOLOGY. 137

Formation of the Young Echinarachnius.

The growth of the young Echinarachnius from its pluteus is not easy
to trace on account of the condensation of pigment upon its walls as it
matures. This formation of pigment renders it very difficult to study
the sequence of the appearance of the plates, and obscures the internal
changes which accompany the maturation of the larva into the adult.
The contour of the young sand-dollar after it absorbs the pluteus is very
different from that of the adult. No one would recognize both as be-
longing to one and the same Echinoid. The whole of the pluteus is
absorbed into the growing Echinarachnius.

A vesicle, the vaso-peritoneal vesicle, on the left hand side of the
stomach (see figures) appears in the very earliest stages of the growth
of the sea-urchin from the pluteus to enlarge, and was observed to have
the form of a retort-shaped structure, with an external opening on the
dorsal side of the body, near the posterior arms, PI. VII. fig. 3. It was
not possible for me to determine whether the left " water-tube " sends
out a prolongation which forces its way to the surface, opening through
a dorsal pore, as A. Agassiz has described in Strongylocentrotus, or
not. In the earliest stage in which I began to study the growth of the
young sand-dollar, the dorsal opening had already formed, communi-
cating through a tubular body with the water-tube. Consequently,
the growth of the tube through the body was not observed or studied.
In the pluteus in which this external opening had formed, the arms of
the pluteus were all of the same length, and consequently the pluteus
was regarded as adult. In the pluteus of Strongylocentrotus, accord-
ing to A.. Agassiz, the young sea-urchin first appears in a young or im-
mature pluteus, in which the arms are not of the same length, judging
from his fig. 52, in "Revision of the Echini, Embryology," p. 717. In
this figure the autero-internal arms had not begun to push out from
the oral lobe, and the antero-lateral rods were just formed. This plu-
teus appears to be immature as far as the appendages go, since they
are not fully formed. The beginning of the young Echinarachnius on
the left water-tube was not traced in a pluteus as young as this pluteus
of Strongylocentrotus.

Balfour* in his account of this figure (fig. 52) gives an interpretation
to the structure, t, difiereut from A. Agassiz. The latter author says,
*' On the left water-tube we notice a very prominent loop, t, which, from

♦ Op. cit. pp. 472, 473.

138 BULLETIN OF THE

its resemblance to the tentacular loops of Brachiolaria, and from its
position ou the water-tube connecting with the water pore, I have no
hesitation in considering to be the first tentacular loop formed."
Balfour considers this structure an invagination of the external surface
of the larva, an infolding which later is to form the ventral region of
the Echinoderm, Metschnikoff * ascribes to A. Agassiz the discovery of
an invagination of the outer skin of the pluteus to form the body of the
future echinus. I have not been able to find in A. Agassiz's works,
quoted by Metschnikoff, that he has made such a " discovery," and cer-
tainly he does not give to the lettering of his figures the same interpre-
tation which Balfour does, when he says that the structure in question, t,
is a tentacular loop.

PI. VII. fig. 3, represents the young of Echinarachnius formed on a
pluteus of the same general form as that figured in PI. VII. fig. 2. The
left water-tube has here formed the " rcsette " of five radial tubes,
which are seen in profile in the figure. The whole body of the pluteus
is not represented, but a portion of the edge of the stomach of the plu-
teus is seen on the right-hand side of the figure. The figui'e is a
representation from the dorsal side.

The five radial bodies of the " rosette," one of which is lettered, wt,
are the water-tubes or the ambulacral divisions which, when seen from
one side, would form a five-rayed water system derived from the left
water vesicle. The system communicates with a tube which passes
through the mesoderm ic layer of the plutean body, and opens externally
by an opening, ma, which is supposed to be the madreporic opening.
This opening is at first situated near the base or proximal end of the
posterior rods. Its communication with the water-tubes has a retort
form, especially in older stages. The retort-shaped vesicle was observed
in a stage a little older, PI. VII. fig. 4, than the last to be in direct
commimication with the rosette-shaped system of five ambulacral ves-
sels. Each of the five radiating ambulacral tubes which form the
rosette extends outward from a central region, which is in direct com-
munication with the retort-shaped body. These radial ambulacral tubes
are at first simple, without lateral branches, coeca-like folds or loops, f
with transparent walls, the outlines of which can be plainly seen

«

* Op. cit. p. 41.

t These five loops are supposed to be homologous with the " odd ocular tenta-
cles " of Arbacia and Strongrjiocentrotus. See A. Agassiz, Report on the " Chal-
lenger " Echinoidea, p. 8. They are supposed to be the same as the " Fiihler," F.
in Asterina. (See Ludwig, op. cit., PL VII. figs. 96, 97.)

MUSEUM OF COMPARATIVE ZOOLOGY. 139

through the wall of the pluteus. The line of the edge of the future
Echinarachuius cau be traced on the dorsal side of the stomach of the
gluteus, while the " dorsal pore," ma, lies near the edge of the dish
turned in the direction of the oral region of the pluteus. The growth
of this opening is a migration from the vicinity of the posterior arm
towards the middle line of the dorsal side of the body. As it grows in
this direction it works at the same time to the anal apex of the pluteus,
never, however, reaching that position. Unlike the figure of a Spatan-
goid pluteus, PL VIII. fig. 13, by Metschnikoff, the retort-shaped vesi-
cle before division into the rosette does not extend so that the dorsal
pore lies in the median line. In the figures which we have of the
young Spatangoid, the line bounding the wall of the growing Clypeas-
troid is always recognized on the dorsal surftice of the body in stages as
old as fig. 8, PI. VIII., of Metschnikoff 's paper on the development of
Echinoderms. I have given a series of figures to illustrate the relative
changes in position of the dorsal pore, ma, from very early conditions, up
to a stage when the deposit of pigments renders observation impossible,
lu the progress of this migration of the madreporic body or dorsal pore
it will be observed that the length of the ambulacral tubes increases,
and additional feet form as diverticula, while interesting calcareous
deposits occur, PI. VII. fig. 9. It was not observed whether these feet
bud from the five primary tubes or not. There is no reason to doubt
that they do. The appearance of pigment spots on the body of the
forming sea-urchin takes place at about the same time as that of the
tritid rods which they later obscure. The first limestone formation
which was observed is a trifid spicule in the wall of the body of the
growing sea-urchin. In its very first form this trifi^d spicule is spheri-
cal in contour. Later, it assumes a trifid shape, and seems to be en-
closed in a transparent sac, the outer wall of which is believed to be
formed of epiblast, the calcareous body being formed possibly in meso-
blast. This transparent sac and its enclosed calcareous body of tripod
shape resembles the structures, cc, in E. lividus, as figured by Metsch-
nikoff.* If these bodies are morphologically the same in Schizaster
and Echinarachuius, we have a likeness hitherto unrecorded between
the young Spatangoid and the immature Clypeastroid.

Metschnikoff figures, PI. VI. (fig. 10, op. cit.,) in the Ophiuran plu-
teus a similar calcareous body, to which he gives the name of " Hohl-
kehlen," already used by Miiller.

Ludwig has already remarked on the resemblance of similar cal-

* Op. cit., PI. VIII. fig. 9.

140 BULLETIN OF THE

careous bodies (fig. 100, op. cit. pp. 67, 68) to the " Basalplatte," of the
so-called " Stiihlchen," in the skin of the Holothurians, and to the "Rad-
chen " of the Chirodotae. He considers that the " Chirodotaradchen
den Basalplatten der Seesternstachel gleichzusetzen sind, gewissermas-
sen rudimentare Stachel darstellen, bei denen sich die ganze Ausbildung
auf die Entwicklung einer Basalplatte reducirt hat."

A. Agassiz has called my attention to the resemblance of similar
bodies in Echinarachnius to the calcareous wheels in the Holothurians.
It seems probable that the stellate bodies in the young Echinarachnius
are the same as the '* Basalplatte " of the spine of Asterina.* It was not
observed that these bodies, as they first appear in Echinarachnius, bear
a definite relation either to the ambulacral tubes or the intermediate
intervals which we may suppose are the interambulacral regions. Al-
though a large number of plutei were examined, the number of these
bodies was not found to be uniform. Some plutei have five, some
one, others three, and many more than five, of these trifid calcareous
bodies. As the echinus grows older the ends of the three spurs of the
trifid spicule became divided or bifurcated, and even subdivided, while
in some cases these bodies were again subdivided. In all these cases
they are still enclosed in a transparent cyst or cell, similar to that
figured by Metschnikofi" for the " Hohlkehlen." This sheath or capsule
is supposed to be the outer enveloping layer, epiblast, of the spine. It
was of course my first impression that these rods were the beginnings
either of ocular or genital plates, and I turned to A. Agassiz's memoir
on the development of the starfish, where similar calcareous bodies are
found, to see if it were not possible to homologize them with the plates
which first appear in the Asteroidea. It was not possible to satisfac-
torily compare these structures, and I was then led to inquire whether
these structures are the beginnings of plates or of other parts of the
echinoid body. My observations at present have not gone far enough
to answer these questions satisfactorily. If these trifid bodies are the
beginnings of plates it cannot be stated at present whether they are
ocular or genital plates, and there is a doubt in my mind whether they
are plates of the test or spines of the same.

PI. VII. fig. 16, is an instructive stage in the development of the sea-
urchin within the body of the pluteus. On the right-hand side of the
figure we see the ambulacral feet, am, of which there are more than
five, the additional having probably formed by lateral budding. On

* Compare fig. 10, p. 69 (Ludwigop. cit.) with the trifid bodies of Echinarach-
nius. See also fig. 100, op. cit., a, h, c.

MUSEUM OF COMPAEATIVE ZOOLOGY. 141

the left-hand side we see the plates of the test of the future sea-urchin.
In the middle of the figure, a little to one side (left), we notice a cen-
tral plate, " centrale," c pi, of pentagonal outline, around which are
arranged a ring of five plates, pi, closely fitting to the central plate.
Outside, or peripherally to these, we see other pentagonal bodies, three
below in the right-hand lower comer, and one above adjoining the upper
letters, pi.

In all the peripheral system of plates we have a reticulation of cal-
careous nature.

In the five plates which surround the central plate we have two
kinds of calcification, one of which forms plates of the test, the other
probably the spines of the plate. The calcareous deposit of the plate
forms an irregular network or reticulation of no regular form, while
the calcareous deposit of the spine has a circular wheel-shaped or stel-
late form, from the rim of which there spring prolongations, in our
figure drawn in a fan shape. The circular portion is ftie base of the
spine ; the fan-shaped continuations or extensions, the beginning of the
shaft. In some of the plates which are more peripherally arranged as
regards the five plates described, we find fan-shaped calcareous forma-
tions, and no reticulated or lace-formed calcifications corresponding to
them.

In the stage which we are considering, the centre of calcification
which is supposed to form the plate of the test (reticulated calcification)
and that which later forms the spine (stellate calcification) are not
joined.

The development of the spine in sea-urchins and starfishes has been
traced by A. Agassiz. He says : * " The shell of a sea-urchin is made
up of an irregular network of limestone cells, which makes its appear-
ance in the early pluteus stage ; with increasing size this network
becomes closed at certain points, and sends off upright shanks, which
little by little form very irregular fan-shaped spines. In our common
sea-urchins these spines are immovable, forming at that stage part of
the test itself. As the spines grow they become more pointed, but are
still immovable. In somewhat more advanced stages a slight constric-
tion is formed at the base of the spine, and very soon after that, below
the constriction, a tubercle is formed, upon which the spine is articu-
lated, and is then capable of a certain amount of motion, etc." (I have
omitted his reference to plates and figures.) It would seem, then, look-

• Revision of the Echini, pp. 667-669.

142 BULLETIN OF THE

ing at fig. 16, PI. VII., if the stellate bodies iu it are spines, and the
reticulated network plates of the test of Echinarachnius, and if this
genus resembles that described by A. Agassiz in the way the spines are
formed, that the stellate cells must have arisen from the reticulations,
and been constricted from them. There is nothing to show that in
Echinarachnius stellate and reticulated rods arise one from the other ;
for these two centres of calcification are distinct in early stages, and we
sometimes have stellate rods without the corresponding lace-work rods.
This then would throw a doubt on the interpretation which I have
given to the stellate calcification as immature spines, unless Echinar-
achnius is very different* from the sea-urchins described in Agassiz's
account, as far as the growth of the first-formed spines is concerned.

The question whether the stellate bodies are spines or pedicellariee
is a very difficult one. If the spines of the genus Echinarachnius form,
like those of Strongylocentrotus, from the reticulated plate of the test,
as recorded by A. Agassiz, we cannot regard the stellate bodies as
true spines. According to Metschnikoff,t the pedicellaria is one of the
first structures to appear in the echinus of E. lividus. PI. VII. fig. 7, of
the last-mentioned work, shows a young pedicellaria, which has a very
close likeness to some of the stellate calcareous bodies of the young
Echinarachnius. The growth of the stellate calcareous body was not
traced into a pedicellaria in Echinarachnius. The homology of the five
plates surrounding the central plate in PI. VII. fig. 16, has not been satis-
factorily made out. It may be conjectured that they correspond with
either the genital or ocular plates of the adult, but they were not traced
to these plates, and such an interpretation would be conjectural. The
formation of new plates, according to A. Agassiz, takes place in Strongy-
locentrotus in a spiral manner. The new plates of Echinarachnius are
thought to form in the same way as those of other Echinoids, but I
have been unable to trace them, on account of the great deposit of pig-
ment, and the consequent opacity of the forming test. In stages older
than PI. VII. fig. 16, the sea-urchin was nearly opaque. In fig. 17,
PI. VI I.J the different plates which compose the test could not be

* The formation of the stellate bodies which have been identified as spines in
Echinarachnius resemble in their growth the growth of the spines of Asterina, as
figured by Ludwig (Zeit.f. Wiss. Zool. XXXVIL pp. 67-70, fig. 100), more closely
than they do those of Asteroidea and Echinoidea figured and described by Agassiz.
Ludwig makes no mention of A. Agassiz's accounts of the development of the
spines in starfishes and sea-urchins.

t Op. cit., PI. Vn. fig. 6.

MUSEUM OF COMPARATIVE ZOOLOGY. 143

identified, on account of pigmentation, although a single opening, ma,
which is thought to be the madreporic opening, was clearly observed.
This interpretation of the opening, ma, is conjectural ; for, with the
exception of the single fact that it occupies the same position as the
opening, ma, of previous stages, there is nothing to show that it is
the madreporic opening. Its communication with the water system
could not be traced.

The oldest stages, PI. VIII. figs. 15, 16, of the young Echinarachnius
here considered, were taken by dredging in the shallow w^aters on a
sandy bottom, where these Echinoderms live. The dredge brought up
a large number of very small sand cakes which were free in its meshes,
while many of the younger specimens were washed out of the sand and
"roots" of Laminaria from the bottom. These young Echinarachnii
are regarded as developed from plutei hatched the past summer. They
were dredged near the end of September. A young Echinarachnius,
older than that here (PI. VIII. fig. 15) described, is figured by A.
Agassiz.* My figures represent stages between that which he has
given and the young Echinarachnius, just after it has absorbed the
pluteus. The form of the young sea-urchin in this stage is spherical,
elongated, plump, more like a Spatangoid or some " round sea-urchin "
than a Clypeastroid. A. Agassiz has compared it to that of the genus
Echinometra. The larger diameter is 1 mm. ; the smaller .8 mm. The
young are almost completely opaque, on account of the formation of
spines, pigment, calcareous rods, and plates.

The spines are relatively larger and more prominent than in the
adult. In many of these structures the superficial layer of the spine
closely hugs its calcareous centre forming the shank, while in others,
mostly younger, the thickness of the outer transparent layer is percep-
tible in lateral profile. The sea-urchin, when seen from the abactinal
area, is found to be oblong, a diameter passing through the anus being a
third longer than that at right angles to it. The anus is slightly excen-
tric, and has the form of a crescentic slit, which is formed by a circular
plate, " centrale," ap, almost closing the circular opening, leaving a
crescentic orifice with concavity towards the apex. The ambulacral
areas are distinguishable from the interambulacral at the apex of the
body, while near the periphery of the test (seen from the aboral region)
their discovery and separation is more difficult. The spines, sp, around
the rim of the body are large and long. The ambulacral feet, am, are
widely extended.

• Loc. cit.

144 BULLETIN OF THE

Seen from the actinal side, PI. VII. fig. 16, it will be noticed that the
oral opening is very large, and that the rim, ed, of the test surrounding
this opening is notched. The diameter of the opening is about one half
that of the test of the sea-urchin. The larger part of this opening is
closed by a muscular wall, in which are imbedded the teeth and dental
apparatus of the Aristotle's lantern, Ian. The five teeth of the last-
mentioned structure are all well developed. The young of Echinocy-
amus, figured by MUller, is probably in about the same stage as the
young Echinarachnius just described. As far as the form of the spines
and their position on the test goes there is little question that the
young Echinoid ascribed by MUller to Echinocyamus has a somewhat
different form after the absorption of the pluteus from that of Echinar-
achnius, although the differences are slight. In both Echinarachnius
and Echinocyamus we seem to have at first the spines arranged in a
single row about the margin of the test, an approach to the arrangement
of spines in Arbacia and some other genera. The young Echinarachnius
is less spherical than that of Echinocyamus. The forms of the ambulac-
ra] feet are alike. The spines are movable, but their motion is small.
The motion of the spines in the young Echinarachnius is observed long
before the absorption of the pluteus. The external changes subsequent
to the stage last described, passing from the young Echinarachnius into
that described by A. Agassiz, consist in a diminution in size of the ver-
tical axis and a migration of the anal opening, ap, more towards the
margin of the disk.

The existence of large spines in the young Echinarachnius and their
subsequent diminution in size in the adult may show a likeness of the
young Echinarachnius to certain embryonic types where the spines
attain a relatively large size. The primary position of the centrale and
subsequent migration of the anal opening from its normal position seem
to indicate a likeness in the young flat sea-urchin to round forms like
those which have an apically placed anus.

The following summary may be made of the preceding observations :

1. The egg of Echinarachnius can be artificially fertilized, and resem-
bles that of other Echinoderms in its mode of segmentation. It has no
polar globules, while the egg is free in the water.

2. A gastrula is formed by invagination, as in some other Echinoderms.

3. The pluteus referred to Echinarachnius by A. Agassiz is an imma-
ture pluteus of Echinarachnius.

4. The development of the young Echinarachnius on the water-tube
of the pluteus resembles that of other sea-urchins. The rosette form of

MUSEUM OF COMPAEATIVE ZOOLOGY. 145

the water-tubes described in other Echinoderms is likewise found in
Echinarachnius.

5. The first-formed calcareous deposits of the test in the young Echin-
arachnius are trifid in shape, varying in number in different specimens.
The extremity of each trifid division later in its growth bifurcates, and
the calcareous body thus formed appears to be enclosed in a transparent
wall, which has a spherical outline.

6. Spines are very early formed, and are proportionally very large as
compared with those of the adult, as in other Echinoderms.

Cambbidgb, March, 1886.

VOL. XII. — NO. 4. 10

EXPLANATION OF THE PLATES.

a. Space between the vitellus and the egg capsule.

ach. Archenteron.

a d. Amoeboid cells, mesoblastic cells.

air. Antero-internal calcareous rod.

td. Anterior lobe.

dr. Antero-lateral rod.

am. Ambulacral tube, or member of primitive rosette.

op. Anal plate, " centrale."

or. Anterior rod.

c. Transparent body, in which is contained a branched calcareous rod.
cap. Capsule.

cav. Cavity.

d. Cell.

d pi. Cleavage plane.

1 d pi. First cleavage plane.

2 d pi. Second cleavage plane.

c pi. Central plate, " centrale " ?

cr. Calcareous rod.

d. Polar globule. In PI. I. fig. 2, polar globule ?

e. Body of young Echinarachnius.

ed. Edge of the test and junction of the muscular oral structure.

/ Muscular filament connecting the wall of the stomach with the epiblast.

g. Abnormal segmentation sphere.

ga. Stomach.

gm. Mouth of gastrula, blastopore.

». Intestine.

Ian. Aristotle's lantern.

/, Lateral arm.

Ir. Lateral rod.

ma. Madreporic opening ; dorsal pore.

mm. Large segmentation spheres, macromeres.

0. Thickened floor of a retort-shaped structure. (This is the same as that

which Metschnikoff calls the invagination of the wall of the pluteus.)

oe. Oesophagus.

d. Oral lobe.

or. Mouth.

p. Primitive furrow. In PI. I. fig. 1, cortical hyaline layer.

pig. Pigment.

MUSEUM OF COMPAEATIVE ZOOLOGY. 147

pi. Calcareous plate.

pr. Posterior rod and posterior arm.

r. Smaller segmentatiou spheres, micromeres.

rd. Anal rod.

s. Stellate calcareous Dody.

sp. Spine. In Pis. I. and IV. the trifid calcareous rod of the pluijus.

spi. Spicules of the two posterior rods after absorption of the body of the pla-
tens by the growing urchin.

t. Transparent layer.

V. Anus.

vp. Vaso-peritoneal vesicle, water-tube.

vt. Vitellus.

wt. Water-tube.

PLATE I.

Figures drawn with camera lucida, Obj. D. D. eyepiece 2, Zeiss. Reduced one
third in photography.
Fig. 1. Ovum of Ophiopholis aculeata, found free in water after? fertilization.

p, Superficial cortical layer of the yolk.
" 2. Same, showing a single polar globule (?) at d.
" 3. Egg in the 2-cell stage with polar globules, d and first plane of cleavage,

1 dpi.
" 4. The same without polar globules.
" 5. Egg in the 2-cell stage with the first cleavage plane, so turned as to show

tlie thickness of the transparent plasmic region following the firt

plane of cleavage.
" 6. Egg in 4-cell stage.
" 7. The same.
" 8. Egg in 8-cell stage.

" 9. The same showing the segmentation cavity, cav. Twelve hours old.
" 10. Blastosphere. The cluster of cells in the middle of the figure are not

cells in the blastocoelic cavity, but cells of the blastoderm with granu-
lations. All the blastodermic cells have the same appearance. One

day old.
" 11. Blastosphere, showing the beginning of an invagination of the archen-

teron, ach, forming the blastopore.
" 12. The same seen from the side. Amoeboid, mesoblastic cells at a cl.
" 13. A little older larval stage, from the ventral side. A small cluster of cells

represented with granulations near the right-hand lower corner.
" 14. Older gastrula from ventral side.
" 15. The same from a lateral view.
" 16. An older gastrula from ventral side. Some of the epiblastic cells in the

lower left-hand side are granulated. This character is not confined to

the cells of this region.
" 17. An older gastrula in which the lateral prominences, which later form the

lateral arms, are shown. The prominence in the medial line above is

the oral lobe, ol, of later figures.

148 BULLETIN OF THE

Fig. 18. An older larva with circumoral filiated band, and prominent anterior

lobe. The prominences on the sides are lateral arms.
" 19. An older larva, in which the calcareous rods, sp, are found in the body.

Thirty-six hours old.
" 20. Still older larva, one side outlined, in which the length of the calcareous

rods has greatly increased. Three days old.
" 21. An older pluteus from the dorsal side.
" 22. The same from ventral side.
" 23. Dorsal view of the oldest pluteus of Ophiopholis which was studied.

Mouth, or, seen through the oral lobe.

PLATE IL

All figures drawn to a scale with camera lucida. Obj. B. B. Eye-piece 2, Zeiss.
Reduced one half in photography.
Fig. 1. Ovum of Echinarachnius parma in its capsule.

" 2. Same ovum with the primitive constriction, p, which forms the first plane
of cleavage. The external pigmented layer of the capsule is not
represented.

" 3. Ovum in 2-ceU stage.

" 4. Ovum in which the second plane of cleavage has begun to divide each of
the two cells of the 2-cell stage.

" 5. A stage somewhat older.

" 6. Outline of the four cells of the 4-cell stage, showing the two planes of
cleavage at right angles to each other.

" 7. Ovum in 4-cell stage.

" 8. Ovum in which constrictions have begun to form new planes of cleavage,
which later divide the four cells of the 4-cell stage to form the 8-cell
stage.

" 9. Ovum in the 8-cell stage.

" 10. The same seen in a plane at right angles to the last.

" 11. Ovum in 8-cell stage showing the segmentation cavity, cav.

" 12. Segmented ovum of a stage with more cells than the 32-cell stage. Two
of the cells are represented with nuclei.

" 18. The same, older, with segment spheres more angular and segmentation
cavity shown.

" 14. Blastosphere, "planula stage." The cells have taken the form of a
hollow sphere. The larva has left the egg-capsule.

" 15. The same showing the flattening at one pole preparatory to an invagina-
tion. (Optical section, in which the cells are not shown in tlie hemi-
sphere turned to the observer.)

" 16. Young gastrula with partially infolded blastoderm, forming the stomach,
(/a.

" 17. Older gastrula. Outlines of blastodermic cells not represented.

MUSEUM OF COMPARATIVE ZOOLOGY. 149

PLATE in.

Fig. 1. Abnormal? egg of fJ. /xir/na, in which the groove-like constriction which
generally encompasses the ovum and forms the first cleavage plane,
is limited to a furrow at one pole, p.

" 2. Ovum in which this furrow, p, has deepened, forming a slit.

" 3. Same egg in an older stage.

" 4. Ovum in 2-cell stage, the original connection between the two cells in
the undivided part of the original ovum turned from the observer.
This egg is seen at right angles to the plane in which Figs. 1-3 are
drawn.

" 5. An ovum in 4-cell stage, in which we have two large segment spheres
and two small. This condition is thought to be uncommon.

" 6. Ovum in 4-cell stage, in which the beginning of the furrow destined to
divide each of the four cells is found on one side, inner side, of all
cells, and does not take the form of a groove-like constriction reaching
wholly about the blastomere.
7. Ovum in 4-cell stage in which each of the four blastomeres Is divided
later into two of unequal size.
8-12, Formation of the 3-cell stage. They first represent the formation of a
4-cell from a 2-cell stage, and then the subsequent breaking away of a
part of one segmentation or cleavage plane, so that two of the blasto-
meres are reduced to one. This is thought to be abnormal, patholo-
gical, or at all events unusual.

PLATE IV.

Fig. 1 Gastrula of Echinarachnius, in which the archenteron, ach, has made its
way to the ventral side of the body. Lateral view.

2. The same, ventral view.

3. An older stage in which the limestone rods, sp, have begun to form.
Lateral view.

4. View of the last from anterior pole.
6. An older larva, showing the differentiation of the anterior lobe oud the

posterior rods or arms. Ventral view.

6. A slightly younger larva seen from one side.

7. An older larva seen from one side.

8. A larva older than the last, seen from the side.

PLATE V.

Fig. 1. Gastrula of E. parma. Lateral view. (The cilia on the body are too
faintly photographed.)
" 2. Older gastrula showing the " water-tube " at vp. Lateral view. As the
larva is under slight pressure, the spicule or calcareous rod, spi, is
slightly thrown out of position.

150 BULLETIN OF THE

Fig. 3. Youngest pluteus with well developed posterior arms. Lateral view.

" 4. The same, older. Lateral view.

" 6. An older pluteus seen from the ventral side. The posterior arms are
well developed ; the anterior lobe is not divided.

" 6. Ventral view of a pluteus, of about the same age as the last.

" 7. View of a somewhat older pluteus in which the two anterior rods, ar, are
formed from the anterior or oral lobe. Ventral view.

" 8. Anal lobe of a pluteus about as old as fig. 6, seen from the ventral side.
The appendages are not figured, c/, cell nuclei 7 Filaments,/, con-
necting the epiblast and hypoblast.

" 9 Lateral view of a pluteus, a little younger than the last.

" 10. View of a pluteus from opposite (lateral) side. This stage is a little
younger than the last.

" 11. Side view of a pluteus of approximately the same age as the last.

" 12. Pluteus in which the antero-lateral arms have begun to form. Ventral
view.

PLATE VL

Fig. 1. Lateral view of a pluteus just before the formation of the antero-lateral

rods.
" 2. A figure showing the relation of the antero-lateral rod, air, when first

formed, to the posterior, pr.
" 8. An older pluteus seen from the dorsal side and laterally.
" 4. A still more mature pluteus, seen from the ventral side.

PLATE Vn.

Fig. 1. Adult pluteus of E. parma, showing the young echinus forming at e.
Dorsal view. Camera lucida, obj. B. B., eye-piece 2, Zeiss. Reduced
one third in photography. Calcareous rods on the right-hand appen-
dages not represented in the arms.

" 2. Adult pluteus. Ventral view. Camera B. B., eyepiece 2, Zeiss. Re-
duced about one third in photography. Right-hand rods not repre-
sented.

" 3-17. Stages in the development of the young sea-urchin. All drawn with
Camera D. D., eye-piece 2, Zeiss. Reduced one third. All from
dorsal side of the pluteus, except Fig. 16.

" 3. Young sea-urchin with five ambulacral tubes, ict, and a single external
opening, ma. Formed on the left-hand water-tube, e, in Fig. 1.

" 4. The same, older.

" 6. Still older stage, slightly elevated from the wall of the stomach of the
pluteus, which it closely hugs in Figs. 3, 4. Pigment spots of den-
dritic shape have appeared. A spherical calcareous body is seen at pi.

" 6. The same, older.

" 7. Older stage with five ambulacral tubes, seen in profile, an external open-
ing, ma, and two trifid limestone formations, pi.

" 8. The satoe, older.

MUSEUM OF COMPARATIVE ZOOLOGY. 151

Fig. 9. The same, still older.

" 10. An older stage, with numerous limestone bodies of dendritic shapes.

" 11, An older condition of the sea-urchin, in which the ambulacral tubes
have developed very considerably, and the dendritic calcareous body
is enclosed in a transparent " ceU," pi, resembling " Hohlkehlen,"
described by Johannes. Mliller.

" 12, 13. Still older stages, similar to Fig. 11.

" 14. A young stage, in which the pluteus is so twisted that a central body,
cp^, "centrale," and five peripheral bodies, pi, are shown. At am
are the ambulacral tubes. The view is at right angles to that of
Figs. 3-13. ma is turned out of sight. The bodies, pi, may be five
radial water-tubes.

" 15. A sea-urchin of about the same age as Fig. 13, showing its relation to the
anal rods, rd, of the pluteus.

" IG. View of a young sea-urchin from the ventral? side, submitted to slight
pressure. At cpl there is a central plate without calcareous deposits.
Around this plate is a ring of five polyhedral plates, in which, pt, there
is a deeper stellate calcareous system, " stellate cells," and superfi-
cial, " lace-work cells or rods." If this is a ventral view, and we are
looking at the plates from below, the stellate rods would be superfi-
cially placed on the test, and the " lace-work " rods more profound.
The lace-work of rods would then be the beginning of the plates of
the test of the sea-urchin.

" 17. A sea-urchin older than the last, with attachment to its pluteus.

" 18. The mouth and adjacent region of the anterior rods of a pluteus of aoout
the same age as Fig. 2, showing the muscular fibres at the end of
the dotted line without letters. The antero-internal rods are moved
in part by these muscles. Free-hand drawing.

PLATE VIII.

Fig. 1. View of the surface of the pluteus between the posterior rods, pr, and the
anterior rod, showing a structure similar to the so-called loop, t,
described by A. Agassiz in Strongylocentrotus.

" 2. The same " loop ' with the external opening partially closed, and the
whole structure more retort-shaped.

" 3. Relation of the " loop "-like structure to the posterior rod, pr.

" 4. The relation of the same structure, " loop," to the water-tube, wt.

" 5. A view of the same very much reduced in size, with the orifice almost
closed.

" 6. The arrangement of the calcareous rods in the body of the pluteus. Soft
parts removed. Lateral view. Anal pole at left of the sketch.

" 7. The relation of tlie infolded part of the external surface to the water-
tube.

" 8. A well formed sea-urchin with spines. Dorsal view of pluteus. Arms
of pluteus removed.

" 9. An older sea-urcliin. Ventral view. Arms of pluteus removed.

152 BULLETIN OF THE MUSEUM OF COMPAEATIVE ZOOLOGY.

Fig. 10. Lateral view of a sea-urchin of about the same age, showing alternating
ambulacra! tubes and spines.

11. A young sea-urchin, under slight pressure, showing a central and five
peripheral plates in its apical region.

12. A young sea-urchin before the absorption of the pluteus, showing two
ambulacral zones. Oral view.

13. Three contiguous plates from a very young sea-urchin, showing charac-
teristic double calcification.

14. A young Echinarachnius raised from a pluteus. The sea-urchin has
absorbed all the soft parts of the pluteus. Three limestone rods, spi,
still remain unabsorbed. Lateral view. Oral region below.

15. Aboral view of a young Echinarachnius obtained by dredging.

16. Oral view of an older stage also obtained by dredging. The animal from
which Figs. 15 and 16 were drawn do not differ much in size.

Id. pi

2cl. pi

Id. pi

I

j0^Mljmf"i%.- ^

\

<fltT\w"' ■ a cl
cav
17 J. W. F. del.

08 V

10

Plate III.

Plate IV.

I

ga

sp-

8

J.W.F. del.

/

9

gm

i

Mr

C

fet

ach

cl

"' t'n

9*

g«

Plate V.

J.W.F. del

Plate VI.

ar

ga % ]\

a cl

J:^^^^f^

« •

^

J.W.F. del.

Plate VM

J W F. del

/

13

am

^
Q

Plate Vlil.

am

J. W. F. del.

i;ro. 5. — Reports on the Results of Dredging, under the Supervision
0/ Alexander Agassiz, in the Gulf of Mexico {1^11-1^) and in
the Caribbean Sea (1879-80), hj the U. S. Coast Survey Steamer
" Blake" Lieut.-Commander C. D. Sigsbee, U. S. N., arid Com-
mander J. E. Bartlett, U. S. N., commanding.

(Published by Permission of Cablile P. Patterson and J. E. Hilgabd, Super-
intendents of the U. S. Coast and Geodetic Survey.)

XXVIII.

Description of Thirteen Species arid Two Genera of Fishes from the
Blake Collection. By G. Brown Goode and Tarleton H. Bean.

The following descriptions relate to species first brought to light by
Mr. Agassiz during his earlier " Blake " explorations in the Gulf of Mexico
and the Caribbean Sea, in 1877, 1878, and 1879. Several other species
have been diagnosed, and their characters will soon be published.

Aphoristia marginata, n. sp. ^

The species is described from a specimen collected by the steamer " Blake "
at Station CLXXi. with a specimen from Albatross Station 2376 as a collateral
type.

The extreme length of the type is 102 miUimeters.

The body is slender lanceolate in form, its greatest height contained 4| times
in the extreme length. The scales are moderate, strongly and sharply denticu-
late, the surface ornamented with many lines and striations, which are so
arranged as to form a semblance of median furrows ; 88 to 90 scales in a longi-
tudinal series, 34 in a transverse series.

Jaws and snout covered with scales.

The length of the head is contained 5| times in total length. The length of
the snout in that of the head 4-^ times, and equal to the diameter of the upper
eye.

The eyes are moderate, close together, the upper very slightly in advance.
The nostril, in a long, slender tube, nearly midway between lower eye and tip
of snout.

Mouth moderate, oblique, curved, its posterior angle beneath the anterior
margin of the pupil of the upper eye ; its length of gape in that of head 4^
times, and 5 times in greatest height of body. Dentition feeble. /''

VOL. XII. — NO. 5. '

154 BULLETIN OF THE

The dorsal fin begins at a point over the posterior margin of the upper pupil.
It is composed of 96 to 100 rays, those about the middle of the body the longest,
and contained about 2^ times in the height of the body.

The anal origin is separated from the snout by a distance equal to four
times the length of the snout. It has 86-87 rays, and its height is slightly
less than that of the dorsal.

The median caudal rays are short, included 85 times in total length.

The distance of the ventral from the snout is contained 5-| times in the total
length ; its distance from the anal, 1 J times the diameter of the eye ; its rays,
four in number, the longest contained 25 times in head.

Color in life, reddish gray much speckled with brown. Belly bluish gray.
Bases and membrane covering fin-rays dark brown. Dorsal and anal fins very
dark on their last tenth. Caudal pale, in marked contrast with the dark area
of dorsal and anal. Tips of dorsal and anal rays, and some of the membrane
covering caudal rays, vermilion.

Color in alcohol, uniform grayish brown, lighter below, with a dark brown
line marking margin between the body and the base of the vertical fins, with a
lighter line or stripe, as wide as eye, inside.

Radial formula : D. 96-100 ; A. 86-87 ; V. 4 ; P. none. Scales, 88 to 90-34.

SPECIMENS.

Sta.

Lat.

Long.

Fnu.

No. Spec,

CLXXXI.

28° 42'

88° 40'

321

1

2376

29° 3'

88° 16'

324

1

XXVII.

Off St.

Vincent.

94

13

1154

I

Aphoristia pigra, n. sp.

This species is described from a specimen obtained by the steamer " Blake "
from dredging off St. Kitts, at a depth of 250 fathoms, with the Fish Commis-
sion specimens from Stations 2318 (4) and 2405 (2) as collateral types.

It is distinguished by its abbreviated form, and its large, rough, strongly
pectinate scales. The extreme length of the type is 98 millimeters.

The body is shorter than in the congeneric Atlantic species ; its greatest
height is contained 3^ times in its total length, or three times in total without
snout.

The scales are large, very rough, with strong horizontal striae and stoutly
denticulated margins, and rather loosely fixed to the skin ; about 65 in a
horizontal series, 34 in a transverse series. The jaws and snout are covered
with small scales. The length of the head is contained 4^ times in the total
length. The length of the snout is contained 4| times in that of the head.
The eyes are moderate in size, very close together, with no scales between ;
the upper is very slightly in advance, and is distant from the dorsal outline a

MUSEUM OF COMPAKATIVE ZOOLOGY. 155

space equal to its own short diamecer. The greatest diameter of the eye is
contained six times in the length of the head. The mouth is oblique, curved,
its posterior angle directly beneath the middle of the lower eye. Length of
gape in that of head four times. The teeth are feeble, closely placed, a little
stronger on the colored side.

The nostril tubular, a little nearer to the lower eye than to the tip of the
snout.

The dorsal fin begins at a point over the middle of the upper eye, and con-
tains about 90 rays to the middle of the base of the caudal. The rays about
the middle of the fin are the longest, their height being a little more than
^ that of the body. The distance of the anal fin from the snout is contained
3| times in total length. The longest anal rays are about the middle of the
body ; their length is equal to that of the longest in the dorsal The anal ia
connate with the caudal, and consists of 69-75 rays.

The length of the median caudal rays is contained nearly seven times in the
total length. The distance of the ventral from the snout is contained 4§ times
in the total length. It is separated from the anal by a distance equal to the
long diameter of the eye. The number of ventral rays is four ; the longest ray
is 3^ times as long as head.

Color grayish or brownish, with a sub-metallic lustre upon the scales when
examined separately. The denticulations of the scales are dark and promi-
nent, giving a clouded general aspect. Some of the smaller specimens (from
Station 2318) have a few large irregular brownish blotches above and a dark
subcircular blotch near the root of the tail, its diameter twice that of the eye.
Colorless below.

Radial formula : D. 90 ; A. 69-75 ; V. 4 ; P. none. L. lat. 65.

Specimens : " Blake " Station xxiii. ; 250 fms., off St. Kitts, W. I. " Alba-
tross " Stations 2318, 2425, 2405, 2374. Off Key "West, Fla., and between
Delta of Mississippi and Cedar Keys, Fla.

Monolene atrimana, n. sp. ^/

The length of the specimen described to base of caudal is 114 millimeters
^^xvi. " Blake," off Barbadoes, 288 fathoms).

The height of the body (37 mm.) is one third of the total length without
the snout, and equals four times the long diameter of the eye ; it also equals
If times the distance of the ventral origin from the snout. The height at the
origin of the ventrals (29 mm.) equals about three times the length of the lower
eye (9 mm.). The least height at the base of the tail (8 mm.) equals f of
the length of the mandible (10 mm.). The body is thin, its greatest width
(4^ mm.) equalling half the length of the eye.

Scales ovate, or oblong, smaller than in M. sessilicauda, and without evident
pectinations. The head is everywhere scaly, except on the lips and the ante-
rior half of the snout. The scaling of the fins is essentially the same as in

156 BULLETIN OF THE

M. sessilicauda. There are 30 rows of scales above and 32 below the lateral
line on the colored side.

The lateral line of the colored side is strongly arched in its anterior part over
the base and anterior third of the pectoral fin. The arc of the curved portion
of the lateral line (10 ram.) equals 2^ times the height of the curve (4 mm.).
The curve is entirely similar to that of M. sessilicauda. The lateral line of the
blind side is nearly straight, very slightly ascending anteriorly. There are 105
scales in the lateral line to caudal base, 18 of these in the curved portion.

The length of the head (24 mm.) equals | of the standard length and 2f
times the diameter of the eye. The distance from the snout to the front of
the upper eye (6 mm.) is much greater than the distance to the lower eye
(3 mm.). The inter-orbital area is a mere narrow ridge, whose width (1 mm.)
equals only ^ of the length of the eye. The length of the maxilla (8 mm.)
equals J the length of the head, and on account of its oblique position its hind
margin does not extend much beyond the vertical through the front margin
of the lower eye.

The length of the mandible (10 mm.) equals ^ of the head's length. The
teeth are uniserial, and well developed on both sides. The nostrils are in very
short tubes, in the same line with the interorbital ridge, the posterior one being
slightly less distant from the lower eye than the anterior is from the tip of the
snout. A concavity above the snout.

The dorsal fin begins upon the snout on the blind side in the perpendicular
through the front of the lower eye. It contains 124 simple rays, the longest
rays being in the posterior fourth of the fin, and half as long as the head. The
anal fin begins between the tips of the ventrals and under the origin of the
pectoral. The vent is not on the ventral outline, but on the blind side and
close to the beginning of the anal fin. The anal is composed of 100 simple
rays, the longest (14 mm.) being behind the middle of the fin and slightly
longer than the longest of the dorsal (13 mm.). The caudal is sessile, rounded,
the middle rays (20 mm.) about ^ of standard length of body.- The pectoral
is present only on the colored side, is inserted close to the edge of the oper-
culum, its length (27 mm.) exceeding that of the head, and contained 4^ times
in the standard length. The ventral of the colored side is nearly on the ridge
of the abdomen, while that on the blind side is mostly lateral and slightly
larger than its fellow. The length of the left ventral (7 mm.) is contained
about 3^ times in length of head.

The color on the left side is light brownish gi-ay ; the fins are mostly dusky,
except the right ventral, which is pale ; the pectoral and the eyelids are
black.

D. 124 ; A. 100 ; V. 6 ; P. 12. Scales, 30-105-32 (18 in curved portion of
lateral line).

A single specimen (xvi.) was taken by the "Blake" off Barbadoes, in 288
fathoms, and another one (xvii.) in the same locality, at a depth of 218
fathoms.

MUSEUM OF COMPAEATIVE ZOOLOGY. 157

Cithariclithys dinoceros, n. sp.

The type is a specimen 92 mm. long to base of caudal, obtained by the
steamer "Blake " at Station xxi., off Guadaloupe, 175 fathoms.

The greatest height of the body (40 mm.) is contained 2.3 times in the total
length, and equals about four times the least height of the tail.

Scales thin, deciduous, cycloid, large, 48 in the lateral line, which is slightly
curved over the pectoral; 14 above and 16 below the lateral line.

The length of the head (27 mm.) is contained 3^ times in the total length,
and equals about 3 J times the diameter of the eye (8 mm.). The interorbital
space is very narrow, its \^dth less than \ diameter of eye ; ridge rather
prominent, narrow, sharp.

The upper eye distant from profile by a space (2 mm.) about J of the orbital
diameter.

The length of the maxillary (12 mm.) is le^sthan half the length of the head ;
that of the mandible (16 mm.) more than half, and twice the diameter of the eye.

The teeth uniserial in both jaws, those in the front much the largest. A
strong spine upon the snout overhanging the upper lip (much lower than in
C. unicornis). Above this there is a second, shorter spine.

The dorsal fin begins upon the snout in advance of eye upon blind side. It
is composed of 91 rays, the longest somewhat behind the middle of the fin
(its height 13 mm.), about equal to half the length of the head.

The anal fin originates about under the origin of the pectoral ; its distance
from the snout (30 mm.) equals J of the total length. It is composed of 73
rays, and is as high as the dorsal.

Caudal subsessile, pointed, its length (17 mm.) contained about 5j times in
total length, and equalling twice the diameter of the orbit.

The pectorals originate immediately behind the branchial opening, far below
the lateral line. The third and fourth rays of the fin upon the eyed side elon-
gated. Its length (38 mm.) is contained 2-^ times in total length. This fin has
10 rays ; that of the blind side contains 6 rays ; its length (12 mm.) is less
than J that of its mate, and is less than half the length of the head.

The ventral on the eyed side originates upon the ventral ridge at a distance
from the snout (27 mm.) equal to the length of the head; it contains 5 rays,
the length of the first (6^ mm.) contained four times in length of the head.
The ventral of the blind side has 6 rays ; its length is contained 2§ times in
length of the head.

Radial formula : D. 91 ; A. 73 ; P. 10/6 ; V. 5. L. lat. 48, 14/16.

Color, grayish brown above, white below.

XXL Off Martinique. xix. Off Barbadoea.

XXVI. Off St. Lucie. xxix. " "

XXVIII. Off Barbadoes.

158 BULLETIN OF THE

BATHYGADUS Gthr.

A genus of Macruridae with large terminal mouth, prominent nape, no teeth,
lanceolate gill-rakers, free notched branchiostegal membrane, high vertical fins,
first dorsal composed largely of branched rays, anal fin set far back. Head
large, fleshy, without prominent ridges, spiny armatures, or external depres-
sions. Nape elevated, hump-like.

Snout broad, obtuse, not produced. Mouth terminal, very large. Suborbital
ridge very low, not joined to the angle of the preoperculum. The maxillary
may be received entirely within a groove under the prefrontal and suborbital
bones, its tip narrowed and blade-like. Intermaxillaries protractile downwards,
separated anteriorly, rib-shaped, compressed vertically, very broad and without
true teeth, and provided posteriorly with a short flange which is received under-
neath the maxiUa. Mandible received within the intermaxillary bones, without
true teeth, but with minute asperities, similar to those in the intermaxillaries.
A barbel. Vomer and palatines toothless.

No pseudobranchiae. Gill-rakers numerdus, moderate, lanceolate, with mi-
nute denticulations along the inner edges. Branchiostegal membrane free from
the isthmus, deeply cleft; branchiostegals 7, very stijff. Gill-opening very wide.
Operculum with a blunt spine-like prominence at its angle. Ventrals below
the pectorals, many rayed, the anterior rays produced.

Dorsal consisting for the most part of branched rays.

Scales cycloid, plain : lateral line strongly arched over the pectoral

Bathygadus arcuatus, n. sp. ^

The type is a specimen, 325 mm., obtained by the steamer " Blake " at Sta-
tion Lxxxix., off' Martinique, at a depth of 334 fathoms. A much larger
specimen, 580 mm., was taken by the Fish Commission at Station 2394. This
specimen is referred to as a collateral type.

The body is shaped much as in Chalinura Simula, but the nape is still more
convex. Its greatest height (57 mm.) is 5; in its total length. The back is
gibbous, the dorsal outline rising rapidly from the interorbital region to the
origin of the first dorsal, whence it descends gradually to the end of the tail.

The scales are moderate, cycloid, subovate, without armature ; those of the
abdominal region and those above the pectorals the largest. The lateral line
is .=(trongly arched over the pectorals, the length of the arched portion con-
tained about 3| times in the straight portion ; the greatest height of the arch
is about J of the length of its chord. The number of scales in the lateral line
is about 140, eight rows of scales between the origin of the dorsal and the arch
of the lateral line, 13 or 14 rows of scales between the vent and the lateral
line counting backwards, 22 counting forwards. Scales covpp all parts of the
head except the jaws and chin.

The length of the heaxi is contained 5 times in total. Interorbital area flat,

MUSEUM OF COMPARATIVE ZOOLOGY. 159

its width (11 mm.) equal to ^ length of head. Postorbital portion of head about
2| times diameter of eye. The operculum terminates in a flat obtuse spine., its
length, including the flap, about equal to diameter of eye. Preoperculum
entire, with a prominent ridge in advance of its posterior edge. The orbit is
rounded, the least diameter of the eye equal to the length of the snout, and
contained 4^ times in length of head (slightly less in the larger specimen).

Snout very broad, obtuse, the intermaxillaries extending beyond it, its width
at the nostrils equal to about twice the length of the eye. Posterior extremi-
ties of the intermaxillary processes elevated, producing a decided hump upon
the top of the snout. The ridge formed by the prefrontal and suborbital bones
terminates very slightly behind the posterior margin of the orbit, and is not
connected with the angle of the preoperculum.

Nostrils immediately in front of the lower part of the eye, not tubular, the
anterior one very small, pore-like, only about ^ as large as the posterior one.
Distance of anterior nostril from tip of snout about | length of eye. Length
of barbel (51 mm.) 6f in length of body, and equal to length of head without
snout (in the larger specimen the barbel is as long as the mandible), more than
3 times as long as the eye.

There are no true teeth, the intermaxillaries and mandible being broad
plates, covered with minute asperities. A naked space at the symphysis of
the intermaxillaries.

Distance of first dorsal from snout (77 mm.) nearly 3^ times length of its
base ; the fin contains 2 spinous and 10 or 11 branched rays ; the first spine
is minute, the second (in the types) somewhat mutilated, its length nearly 3
in length of head.* It is not stouter than the branched rays, and is entirely
smooth.

The second dorsal is separated from the first by a very short interspace,
equal to about ^ of the length of the eye. Its rays are long, subequal, the
first slightly the longest, its length equal to that of the base of the first dorsal.

The anal is much lower than the dorsal, the longest rays being in front, its
third ray about half as long as the first ray of the second dorsal ; this fin is
inserted under the seventh ray of the second dorsal. About three of the termi-
nal rays might be considered caudal rays.

Pectoral inserted slightly in advance of the ventral, which is in about the
same vertical with the origin of the first dorsal. The second ray of the pecto-
ral is slightly produced. The length of the fin equal to that of the head
without the snout.

Ventral insertion distant from the tip of the snout a distance equal to that
of first dorsal from snout. The first and second rays are filamentous, the latter
slightly the longer, and extending to the fifteenth (or eighteenth in larger
specimen) ray of the anal fin.

* Judging from the larger specimen, this spine in a usual state would be consid-
erably longer.

160 BULLETIN OF THE

Radial formula: D. li. 9-10 (135); A. (120); P. 25; V. 8.
Color, brown ; vertical fins, bluish or black ; peritoneum, black ; inside of
gill-covers and roof of mouth, bluish.

2394.

2374. 2 juv.

Lxxxviii. OS Martinique. 476 fathoms. 1 "
Lxxxix. " " 334 "

Bathygadus favosus, n. sp.

The type is a specimen, 350 millimeters in length, obtained by the " Blake "
at Station lxxx., off Martinique, at a depth of 472 fathoms, with the Fish
Commission specimens catalogued below as collateral tj'pes.

The body is heavy, stout ; its greatest height, at origin of first dorsal (57 mm.),
is contained a little more than six times in the total length. The profile of
the body descends gradually and in a slight curve from the first dorsal to the
snout.

The scales are small, deciduous, cycloid, without armature, about 135 in the
lateral line, about 10 above and 16 below the lateral line, the latter series
counted from the vent.

The length of the head (65 mm.) is contained about 5| times in total length.
The interorbital area is slightly convex ; its greatest width (22 mm.) equals
about ^ of the length of the head. The postorbital part of the head is 2| times
as long as the eye, which is nearly round, its diameter equal to \ the length of
the head. The snout is broad, oblique, its width at the nostrils (23 mm.) a
little more than the width of interorbital area; its length (17 mm.) slightly
more than ^ that of the head. The nostrils are close to and in front of the
middle of the eye, the posterior one somewhat the larger. No barbel.

The teeth in both jaws in villiform bands ; a naked space at the symphysis
of the intermaxillaries. The intermaxillary bands are more than twice as wide
as those on the mandible. Vomer and palatine toothless. Tlie longest gill-
raker on the anterior arch is slightly more than half as long as the eye. The
number of gill-rakers on this arch is 25, 20 being below the angle.

Pseudobranchise present, very rudimentary in some individuals, in others
wanting or present only upon one side.

The first dorsal is distant from snout (68 mm.), which is slightly more than
length of the head ; the length of its base (24 mm.) is about equal to width of
the snout at the nostrils. The fin consists of 2 spines, the first of which is mi-
nute, and 9 branched rays. The length of the longest spine, which is armed,
is contained twice in that of the head (specimens examined imperfect). The
second dorsal begins immediately behind the first, the membrane being con-
tinuous. The anterior rays are longest (apparently about j the length of the
head).

MUSEUM OF COMPAEAXrVE ZOOLOGY. 161

The anal is lower than the second dorsal; its distance from the snout (112 mm.)
is about equal to J of the total length.

The pectoral is inserted under the anterior rays of the first dorsal, and very
slightly in advance of the origin of the ventraL Its length is more than half
that of the head.*

The distance of the ventral from the snout (69 mm.) is contained 5 times in
the total. This fin is inserted nearly under the base of the pectoral ; the first
ray is somewhat produced ; f its tip reaches to the fourth ray of the anal fin.

Radial formula : D. li. 9, 125 ; A. 110 ; V. 9 ; P. 14 ; B. 7.

Color, blaish brown, darkest upon head and abdomen, especially in Museum
specimens.

Lxxx. Off Martinique. 472 fathoms. " Blake."

34,911. N. Lat. 15° 24' 40", W. Long. 63° 31' 40". "Albatross."

34.909. "

34.910. N. Lat. 15" 24' 40", W. Long. 63° 31' 40".

«

34,920.
34,918.
(2392).
Ilxxxii. 1 juv.
(2394). 1 "

€t

Neobythites robustus, n. sp.

V

The type is from "Blake" Station xciv., off Moro Castle, Cuba. 250-400
fathoms. Length, 88 mm.

With specimen (No. 29,057) from " Fish-Hawk " Station 1043, Lat. 38° 39',
Long. 73° ir, 130 fathoms, as a collateral type.

Body rather short and deep, its greatest height (19 mm.) nearly 4| in total
lengtli and about equal to length of head. The interorbital area is convex ;
its width (6 mm.) is greater than the diameter of the circular eye (5 mm.)
and li times the length of snout (4 mm.). The length of the head (19 mm.)
is about 4 times the diameter of the eye. The mouth is moderate, the max-
illa extending to the vertical through the posterior margin of the eye, the
mandible a little beyond, its length (10 mm.) equal to that of postorbital part
of head. Teeth in villiform bands in the jaws, and on the palatines. Vo-
merine teeth bunched in a circular patch. Gill-rakers moderate, the longest
a little more than twice in diameter of eye, four above angle of first arch,
eleven below. Pseudobranchise rudimentary. Gill-opening wide, the mem-
brane deeply cleft, behind free from the isthmus. A pair of short flat spines

• In one of the "Albatross" specimens the pectoral extends to the vertical
from the eighth ray of the second dorsal.

t Its length in one of the " Albatross " specimens is equal to that of the head
without snout.

VOL. XII. — NO. 5. 11

162 BULLETIN OF THE

upon the anterior portion of the operculum, extending backward nearly to
its posterior edge.

The nostrils are small, the anterior as close to the snout as the posterior
ones are to the eyes. No apparent cirri. The scales are minute ; the lateral
line is obsolete on the last fourth of the length of the body.

The dorsal origin is behind that of the ventral and pe(;toral ; its distance
from the snout (24 mm.) is contained 3§ times in length of the body. The
height of the fin is moderate; the longest ray is contained about 3 times in
the length of the head.

The anal origin is under the eighteenth ray of the dorsal ; the height of the
fin about equals that of the dorsal. The vertical fins are not connate with the
caudal, which consists of 12 or 13 very slender rays, its length nearly equal
to half that of head.

The pectoral with a broad base, close to gill-opening, irs length nearly § that
of the head.

The ventral a single bifid ray, inserted slightly in advance of the vertical
through the base of the pectorals, and not far from the humeral symphysis.
It reaches nearly half-way to the vent, the distance of which from the origin
of the ventral is equal to the length of the head.

Color yellowish brown.

Neobythites marginatus, n. sp. \/

The type is from " Blake " Station lxxix., off Barbadoes, 209 fathoms.

Body compressed, somewhat elongate ; its height (18 mm.) contained 5 J times
in total length, and less than the length of the head. Interorbital area convex,
its width (5^ ram.) greater than the diameter of the circular eye, which is 4^ mm.
The length of the head (22 mm.) is contained 4§ times in that of the body.
Mouth large, the maxilla extending considerably behind vertical through pos-
terior margin of orbit ; its length equals half that of the head. The length
of the mandible (13 mm.) is slightly more than § of height of body.

The teeth as in N. gillii.

Gill-rakers slightly longer than half the diameter of the eye, 7 and 3 rudi-
ments below the angle of the anterior arch. Pseudobranchiae absent. Gill-
openings as in N. rohustxis. A long flat spine upon the upper edge of the
operculum, extending back nearly to its margin. Two short flat spines upon
the angle of the preoperculum. Nostrils as in N. gillii.

The scales small, very closely imbricated, in about 123 rows, 7 above and 29
below the lateral line.

The lateral line obsolete in its posterior half.

The dorsal is composed of 101 rays ; its distance from the snout is contained
4 times in total length.

The anal originates under the fourteenth doi'sal ray at a distance from the
snout contained more than 2§ times in total length.

MUSEUM OF COMPARATIVE ZOOLOGY. 163

The caudal consists of about 8 or 9 rays very closely placed ; its length is
contained about 10^ times in the total length.

The pectoral is placed as in N. robustus ; its length about equal to 2^ times
that of the head, extending to vertical througn the vent.

The ventral, a bifid ray inserted in advance of .base of pectoral, not reaching
to the vent ; its length (14 mm.) considerably less than the height of body.
The distance from its origin to the vent (19 mm.) slightly more than the
height of the body.

Color light yellowish brown, an obscure narrow band of darker brown com-
mencing on the snout, interrupted by the eye, and extending backward f of
the distance to the tail, another beginning on the snout, extending ove.' the eye
and back as far as the first described, interrupted posteriorly. Dorsal fin milky
white at base in its anterior third, above this a blackish band extending the
whole length of the fin. A narrow white margin above.

Aphyonus mollis, n. sp.

The type is a specimen obtained at " Blake " Station ccxxi., Lat. 24" 36' N.,
Long. 84° 5' W., at a depth of 955 fathoms. 85 + x mm.

This species is closely allied to Aphyonus gelatinosus, Gthr., obtained by
H. M. S. " Challenger."

The body is much compressed, its greatest height (14 mm.), 6 in its total
length. Head thicker than body, its height (15 mm.) slightly greater. Length
of head (20 mm.) about 4j in total ; width (11 mm.) over half its length.
Snout, 3J in length of head. Eye not externally visible. Diameter of orbit, as
seen through the skin, about ^ length of head. Maxilla extends to vertical
through posterior margin of orbit, the mandible somewhat farther back, its
length (13 mm.) nearly equal to height of body. A few weak teeth on vomer
and palatines, mandible, and very rudimentary ones in maxillary ; not visible
to the eye, but appreciable to the touch. Gill-laminse on the fourth and rudi-
mentary gill-rakers, 8 rudiments and 4 developed below the angle. Dorsal
origin almost over posterior edge of operculum, its distance from the snout J of
total length ; fin-rays, more than 110 well developed, the longest 3 in head.
Anal origin slightly nearer base of caudal than to tip of snout, its rays shorter
than those in the dorsal. Pectoral with a fleshy base ; its origin somewhat
behind that of the dorsal, its length equal to width of head. Ventral origin in
advance of that of pectoral, close to humeral symphysis ; the fin is a single
simple ray, whose length (11 ram.) equals that of the pectoral : its tip does
not reach the vent, by a space equal to height of head.

Skin not loose. Texture of body rather firm, not transparent, whitish.

164 BULLETIN OF THE

BARATHRONUS, n. gen.

Head stout, body and tail compressed, covered closely by skin, scaleless.
Vent far bahind pectoral, included in a cleft. Mouth wide, oblique, the lower
jaw projecting. Intermaxillary teeth rudimentary ; several fang-like teeth
on the heat! of the vomer, none on palatines. A few rather large recurved,
separated teeth in the mandible. Nostrils close together and small. Eye
visible through the skin, partly upon the top of the head, with or without
dark pigment in the iris. Barbel none. Gill-rakers very numerous and slender,
and rather long. Gill-laminoe well developed on all the arches. No pseudo-
branchiae. Head full of muciferous channels. Gill-raembranes not united, but
covered by a fold of skin. Ventrals reduced to single simple rays, placed in
advance of the pectorals and close to the humeral symphysis. Dorsal and anal
placed far back.

Caudal scarcely differentiated, composed of rather numerous very slender
rays upon a somewhat narrow base.

Baxathronus bicolor, n. sp.

The type is an individual, 120 mm. long, from "Blake" Station lxxl, oflF
Guadaloupe, at a depth of 769 fathoms.

Body much compressed, its greatest height (19 mm.) contained 6J times in
the total length. Head much thicker than body, its greatest width equal to f
of its length (23 mm.), which is contained 5^ times in the total length. Eye
concealed by the skin ; diameter of orbit about equal to width of interorbital
area, and contained 4| times in length of the head. Maxilla extends slightly
beyond the perpendicular through posterior margin of orbit ; it is almost
entirely concealed under the preorbital, and is much expanded at the tip,
where its width is rather greater than that of the eye. Intermaxilla very
thin, broad, and slightly protractile.

Vomer very close to intermaxillary symphysis, its head somewhat raised and
bearing three fang-like teeth (two of which are on one side and one on the
other, in the type separated by a moderately wide interspace). The mandible
has five enlarged, separate, recurved teeth upon each side, which increase in
size posteriorly ; its upper edge, posteriorly, is produced above the level of the
tooth-bearing surface, and is received under the expanded maxilla. The long-
est gill-raker is about as long as the eye. The dorsal origin is distant from the
snout (54 mm.), which is contained slightly less than twice in the total length.
Its rays are well developed, numerous, long and slender, about 70 in number ;
the longest contained about 3 times in the length of head.

The anal originates in vertical from fourteenth dorsal ray, equidistant be-
tween eye and base of caudal. It contains 57 rays, about as long as those in
the dorsal.

The pectoral with a fleshy base^ its length (18 mm.) a little less than height
of body.

MUSEUM OF COMPARATIVE ZOOLOGY. 165

The ventral well in advance of pectoral, close to humeral symphysis, the
rays being placed very close together at their origin, the length of the fin
(13 mm.) contained about 9 times in the total length, about 3 times in distance
from its origin to the vent.

The caudal has about 10 rays ; its length is contained about 8 times in the
total length.

Color, yellowish white, with a broad vertical band of black from the origin
of ventral nearly to the vent, another sinular and narrower band above it upon
each side.

Bregmaceros atlanticus, n. sp. v

Specimens were obtained by the " Blake " at the following stations : —

xcix. Off Granada. 90 fathoms. 3 spec.

cv. ? 11 "

cxiii. OffNeris. 305 fathoms. 1 «

CLXXXV. Lat. 25° 33' N. Long. 84" 21' W. 101 fathoms. 1 «

The species agrees very closely with the only other known species of the
genus, B. macclellandii, Thompson (= Calloptilum mirum, Richards.), from the
Western Pacific, from which, however, it differs in the lesser number of rays
in the first anal, and in the greater height of the vertical fins (judging from
figures.

The type (cv.) is 46 mm. long to base of caudal. Foiin compressed, mod-
erately elougate. Body height (6 mm.) 7f in its length. Interorbital area
convex, its width (2^ mm.) greater than diameter of eye (2 mm.), which is four
in length of head (8 mm.). Length of head 5| in total. Jaws even in front.

Maxilla reaches to vertical through middle of eye ; the mandible to vertical
through its posterior margin.

Teeth on intermaxillary minute, apparently in a single series, mandibulary
teeth biserial, the inner teeth enlarged.

Scales large, about 10 in a transverse series, about 65 in a longitudinal
series.

Cephalic appendage reaches nearly to base of first dorsal, its length (10 mm.)
4^ in total.

Distance of dorsal from snout (17 mm.) 2J in total ; that of anal the same.

The dorsal and anal fins received in a groove formed by the scales along
their bases.

Anterior portion of second dorsal and second anal less elevated than in
B. macclellandii. The differentiations between the developed and undeveloped
rays of the anal are so slight that the limits of the so-called anterior and pos-
terior sections of the fin cannot be determined.

Length of the longest anal ray (22 mm.) about 2 in body length.

D. 1 + 15-16. A. 15-16 -h X (7 or 8) -^- 21-22.

166 BULLETIN OF THE

/'

Peristedium longispatha, n. sp.
LViii. Off Santa Cruz. 314 fathoms.

Body high anteriorly, its greatest height (39 mm.) contained 4^ times in
total length. The length of the head, without prolongations, is contained 2^
times in the total length ; with prolongations, 2 times. The crown of the
Lead is flat, separated from the nuchal plate by a deep furrow, which is convex
forwartl. The interorbital space is deeply concave, the supraorbital margins
being swollen ; its width (16 mm.) equal to the long diameter of the orbit.
No protuberance on the forehead, which is much depressed, its outline descend-
ing abruptly and rapidly in front of the eyes. A ridge, but no spine, beneath
the eye. The length of the snout (49 mm.), including the preorbital extension,
is more than half the length of the head; the preorbital extension equals
about half the length of the snout. The processes are flat, rounded anteriorly,
and covered with minute granulations ; they diverge considerably, the distance
of the tips apart (36 mm.) being nearly twice that at their bases (19 mm.)-
A ridge arises at the base of the preorbital process and extends to the angle
of the preoperculum, and its width at the angle (8 mm.) is contained twice
in the diameter of the orbit A narrow inconspicuous and interrupted ridge
helow. A ridge on the operculum, ending in a sharp spine at the angle ; its
length is equal to the diameter of the eye.

The jaws are feeble, toothless ; the lower jaw with 2 long, much fringed
barbels, and 14 shorter ones. The length of the long barbels (32 mm.) twice
the diameter of the eye.

The maxilla does not quite reach tne vertical through the anterior margin of
the eye. The diameter of the eye (16 mm.) is contained four times in the
length of the head without its prolongations. The greatest width of the head
over the preopercular ridge (61 mm.) is contained three times in the total
length.

The dorsal origin is directly in a line with the upper angle of the gill-open-
ing. The longest spine (18 mm.) slightly longer than the width of interorbital
space. The fin has 8 -j- 19 rays.

The anal origin is under that of the second dorsal. The fin has 19 rays.
Caudal small, slightly emarginate, the length of its middle rays (23 mm.)
equals 1^ times the diameter of the eye.

Ventrals slightly in advance of the pectorals and extending farther back,
reaching slightly beyond vent, and to vertical through seventh row of scales.

Pectoral short, extending to vertical from fifth scale of the lateral line, the
longest detached ray to the sixth. Twenty-nine rows of scales.

Color in life, bright roseate ; a black blotch ne&r the tip of the pectoral.
Dorsal with narrow dark margin ; tip of caudal black.

The elongation of the preorbital' extension is noticeable in the smallest
examples.

MUSEUM OF COMPARATIVE ZOOLOGY.

167

LVIII.

OflF Santa Cruz.

(Type.)

« Blake."

LXII.

Off Barbadoes.

209 fathoms.

«

LXIII.

« u

«

((

2397.

"Albatross."

2376.

«

2407.

«

2358.

«

Peristedium platycephalum, n. sp.
The length of the type to tip of snout, without prolongations, is 145 mm.

Type LX. Off Barbadoes.

123 fathoms.

y

LIX.

288^ "

2299 yg. ?

r

Body much depressed, its greatest height (23 mm.) 6^ in body length, 6| in
total.

Length of head without prolongations (47 mm.), twice the height of body, 3^
in its length, with prolongations i\ in body length. Interorbital space deeply
concave, the supraorbital margin being swollen, its width (14 mm.) equal to the
long diameter of the eye. No protuberance on the forehead, which is much
depressed, its outline descending abruptly and rapidly in front of the eyes. A
ridge below the eye, not armed; a small vertical spine behind each nostril.
Stout spines upon operculum and several upon the vertex. The length of the
snout with its extensions (29 mm.) is half the length of the head, its processes
(10 mm.) about 3 in its own length. The processes are flat, triangular, di-
verging slightly, the distance apart of their tips 2-2^ that at their bases. A
ridge extends backwards from the base of each process along the lower edge of
the preoperculum, ending behind in a sharp flat spine, the greatest width of the
expanded portion, on the preoperculum, only \ as wide as the eye. Beneath
this is another less conspicuous ridge with minutely serrated edge, which is
double in front and single behind, the two portions separated by a slight notch.
Jaws normal, the two tentacles much fringed, their length (16 mm.) not
much exceeding the diameter of the eye ; between them, and placed about equi-
distant from each other, are two bunches of short tentacles, about four in each.
Chin with numerous short tentacles, some of them as long as the eye, arranged
for the most part in bunches of four-
Maxilla does not reach to the anterior margin of orbit.

Diameter of eye (13 mm.) nearly four in greatestlength of head, and exactly
half total length of snout. Greatest width of head, over the preopercular
ridges (43 mm.), nearly equal to its own length without the processes. Dorsal
origin over the upper angle of gill-opening. The fin has 8+17 rays. The
length of the longest spine (18 mm.) is equal to that of postorbital portion
of head.

168 BULLETIN OF THE

Anal origin about under origin of second dorsal, a trifle farther back, and in
the vertical through the space between the seventh and eighth lateral scutes.
The fin has 17 rays. It is about as high as the dorsal.

Caudal small, slightly eniarginate, with tips slightly produced, the length
of the middle rays (18 mm.) equal to that of the dorsal.

Ventral origin in advance of the axil of the pectorals ; the fin extends slightly
beyond the vent, but not quite to the origin of the anal ; its length (35 mm.)
about twice the length of the dorsal.

Pectoral father long, extending to the ninth scute of the lateral line, and
past the vertical through tlft origin of the anal. Twenty-nine rows of scutes.

Color red. Body and fins mottled and blotched with darker.

BENTHOSAURUS, n. gen., Synodontid.

Body long, somewhat compressed, tapering into a slender elongate caudal
peduncle. Scales cycloid, of moderate size. Head slightly depressed ; cleft
of mouth wide, horizontal, the lower jaw projecting at its extremity and
anteriorly at the sides. The maxilla is long, not stout, dilated posteriorly ;
the intermaxillary very long, styliform, tapering, immovable. The inter-
maxillary and mandible with bands of small teeth, of uniform size, inter-
rupted at the symphysis. A short oblong band of similar teeth on each
side of the vomer, separated by a rather wide interspace. Palate and tongue
smooth. Eye very small, inconspicuous. Gill-opening extremely wide, the
branchiostegal membrane free from the isthmus. Gill-rakers long and slender,
numerous, about twice as many below the angle as above. Pseudobranchiae
absent. Branchiostegals eleven. All the fins well developed ; no adipose
dorsal. Dorsal fin median, anal post-median. Caudal forked, with lower
lobe produced. Ventral seven-rayed, inserted opposite the interspace between
pectoral and dorsal, the outer ray produced.

Benthosaunis is closely allied to Bathysaurus and Bathypterois, resembling
the latter in nearly every particular save in the structure of the pectoral fins.

Benthosaurus grallator, n. sp.

V

Body elongate, somewhat compressed, depressed slightly forward, tapering
behind into a long slender tail; its greatest height contained 7^ times in its
standard length, and equalling half the length of the head, its greatest width
^ the length of the head ; its height at the origin of the anal, | of its greatest
height. Least height of tail half the height of the body at the ventrals.
Length of caudal peduncle 6^ times its least height.

Scales very thin, cycloid, leathery, deciduous ; oval in form, except at the
base of the dorsal and anal fins, where they become more elongate ; the hori-
zontal diameter of a scale in the lateral line equals twice the diameter of the
eye. The lateral line is straight, above the median line anteriorly, becoming

MUSEUM OF COMPARATIVE ZOOLOGY. 169

median on the caudal peduncle, the tube-bearing scales being prominent, and
about 55 in number. Between the dorsal fin and the lateral line are about
nine rows of scales ; between the latter and the anal fin, eight or nine rows.

Head twice as long as the greatest height of the body, its length contained
a little less than four times in the standard body length, considerably depressed,
scaleless except on the vertex and the preoperculum. Operculum, perhaps
accidentally, denuded.

The snout is much produced, almost equal to the width of the interorbital
space, which is convex. The maxilla extends far behind the posterior margin
of the eye, its length equalling that of the postorbital part of the head. The
mandible projects beyond the upper jaw to a distance slightly more than the
diameter of the orbit, and receives the snout within its extremity when the
jaws are closed. The teeth have been fully described in the generic diagnosis.
The mandible has a series of seven large pores on its lower surface. There are
several similar pores under the eye. The nostrils are situated about midway
between the eye and the extremity of the snout, small, slit-like, the posterior
about twice as large as the anterior one in each pair.

The dorsal fin contains eleven rays, and is inserted midway between the tip
of the snout and the base of the middle caudal rays. The fin is highest in
front, the length of the rays diminishing rapidly posteriorly. There is ap-
parently no adipose dorsal.

The anal fin contains twelve rays and is similar in shape to the dorsal, the
anterior rays being the longest, and about equal in length to the mandible ; its
distance from the snout is about three times the length of its longest ray.

The caudal is forked, its middle rays f as long as those in the upper caudal
lobe ; the lower lobe is much prolonged, the lower ray being more than 4 times
as long as the middle rays. Its extremity is broken off in our specimen, but
apparently it must have been nearly twice as long as the stump which now
remains.

The pectoral fin is normal, composed of 9 rays, and is inserted close to the
opercular flap ; its length is slightly greater than that of the head (though
mutilated), extending beyond the origin of the dorsal.

The ventral is composed of 7 rays, and its base is entirely in advance of the
perpendicular from the origin of the dorsal ; the inner rays reach to the vent,
while its outer ray is enormously prolonged, extending far beyond the extrem-
ity of the upper caudal lobe ; the length of the prolonged ray is fully 4 times
that of the head. The two ventrals are close together.

Radial formula : D. 11 ; A. 12 ; P. 9; V. 7 ; B. 11. Scales, 9-55-8 or 9.

Color brown, the roof of the mouth and inside of the branchiostegal flap
black, as well as the operculum and branchiostegal membrane.

A single specimen, 392 mm. (15^ inches) long to the tips of the prolonged
ventral rays, was taken at a depth of 1850 fathoms, at Station CLXXiv., in
Lat 24° 33' N., Long. 84° 23' W.

A second example of the same fish, and" oT nearly the same size, was taken

170 BULLETIN OF THE MUSEUM OF COMPARATIVE ZOOLOGY.

by the steamer "Albatross," September 6, 1884, in Lat. 39° 3' 15" N. and
Long. 70° SCy 45" W., at a depth of 1537 fathoms. This is well preserved, and
throws additional light on the external characters of the species ; the fins,
especially, are more nearly perfect. Measurements are given alongside of those
taken from the " Blake " specimen.

Benthosaurug grallator.

Current number of specimen clxxiv. Blake.

Length to base of middle caudal rays 275 mm.

Body, — Greatest height 37 "

Greatest width 24 "

Height at ventrals 32 "

Least height of tail . . . ' 16 "

Length of caudal peduncle 105 "

Head. — Greatest length 73 '•

Greatest width 28 "

Width of interorbital area 20 "

Length of snout 19 "

Length of upper jaw 48 "

Length of mandible 56 "

Distance from snout to orbit 21 "

Diameter of orbit 2.5 "

Dorsal. — Distance from snout 137 "

Length of base 35 "

Length of longest ray (first) 49+"

Length of last ray 11(?)"

Anal. — Distance from snout 152 "

Length of base 34 "

Length of longest ray (first) 55 "

Length of last ray 12(?)"

Caudal. — Length of middle rays 23 "

Length of external rays, upper lobe 69 "

Length of external rays, lower lobe 100-J- "

Pectoral. — Distance from snout 67 "

Length 77-f- "

"Ventral. — Distance from snout 103 "

Length 285+"

Branchiostegals 11 "

Dorsal 11 "

Anal 12 "

Pectoral 9 "

Ventral 7 "

Number of scales in lateral line 55

Number of transverse rows above lateral line 9

Number of transverse rows below lateral line 8 or 9

35.651.

267

mil

35

(t

20

ti

33

n

16

((

95

«

70

it

26

«

18

((

18

n

48

«

63

«

2.6"

123

«

46

n

u

17

ti

146

l<

33

it

50

a

16

it

19

it

50

it

221

it

62

it

84

it

99

it

240

a

11

«

13

it

9

ti

7

a

60

9

9

ITo. 6. — Reports on the Results of Dredging, under tJie Supervision
of Alexander Agassiz, in the Gulf of Mexico (1877-78) and in
the Caribbean Sea (1879-80), hy the U. S. Coast Survey Steamer
"Blake" Lieut.-Commander C. D. Sigsbee, U. S. K, and Com-
mander J. E. JBartlett, U. S. N., Commanding.

(Published by permission of (Jarlile P. Patterson and J. E. Hilgapd, Superin-
tendents of tlie U. S. Coast and Geodetic Survey.)

XXIX.

Report on the MoUitsca, hy W. H. Dall. — Part I. Brachiopoda and

Pelecypoda.

In 1878, the moilusks of the " Blake " were intrusted to me for exami-
nation and report, and a Preliminary Note upon them appeared in August
of that year (Bull, Mus. Comp. Zool., Vol. V. No. 6, pp. 60-62).

The following season a second instalment was received, and, in Feb-
ruary, 1880, after a cursory examination of the material, I was enabled
to furnish Prof, Agassiz with a short resume of the general conclusions
which seemed to result from the data obtained by that examination
(Bull. M, C. Z., Vol. VI, No. 3, pp. 85-93).

My time during ordinary working hours being absorbed by official
duties, and the entire period from February, 1880, to January, 1881,
being occupied by field-work on the Pacific coast, progress has neces-
sarily been slow ; the more so, as nearly all the material consists of
specimens so small as to require reiterated scrutiny under a glass to
determine their characters. To separate and label the specimens con-
tained in nearly two hundred different lots, to select specimens for
figuring and to scrutinize and revise the drawings, to search the lit-
erature relating to moilusks for the scattered data in relation to such
as are native to the region in question, and to accurately describe such
species as seemed to be new, has been the task before me, to be carried
out in the scanty leisure afforded by such evenings and holidays as were
not necessarily otherwise employed. The delay in completing the work,
it will be seen, has been inevitable under the circumstances, and my
thanks are due to Prof, Agassiz, and others interested, for the patience
with which they have kept these circumstances in mind.

VOL. XII. — NO. 6, 11

172 BULLETIN OF THE

In 1881, however, 1 was able to prepare preliminary descriptions of
Bome of the more striking novelties (BuU. Mus. Comp. Zool., Vol. IX.
No. 2,* pp. 33-144) and enumerate some of the more remarkable forms
in the collection which had been described already. The dredgings of
the U. S. Fish Commission having produced a number of deep-water
species of limpets and chitons, — which were generously submitted to
me for study by Prof A. E. Verrill in charge of that material, — the
investigation was facilitated by the possession of the Blake collection, and
the study of the whole brought about the publication, in 1882, f of some
extremely interesting facts in regard to these groups of mollusks.

Lastly, the investigation of the literature necessary for this work, and
for the determination of the Tertiary fossils of the Southeastern United
States, culminated in the preparation, under my supervision, of a general
index to the species reported from the coast and islands of the region
between Cape Hatteras, North Carolina, and Cape San Eoque at the
northeastern extremity of South America, including the Bermudas and
West Indies. This is the first attempt to bring together the names of
the alleged species said to inhabit this region, though there have been
a number of excellent local catalogues. The list was found so useful for
both biological and paleontological purposes, that it was printed by the
U. S. Geological Survey as one of its Bulletins. t A short article on the
characters of Dimya, based on the study of the soft parts, first collected
by the Blake expedition, was printed in Science (No. 2, Feb. IG, 1882,
p. 51).

* This Bulletin was first published in signatures, distributed as soon as printed
to those most interested, as follows : pp. 33-48, July 12, 1881 ; pp. 49-64, Aug. 12,
1881 ; pp. 65-80, Aug. 25, 1881 ; pp. 81-96, Sept. 26, 1881 ; pp. 97-112, Oct. 31, 1881 ;
pp. 113-128, Nov. 26, 1881 ; and the remaining pages and index, Dec. 5, 1881. By
the great kindness of the Rev. R. Boog Watson, who supplied nae promptly with
his preliminary descriptions of the Challenger gastropods, of Dr. Paul Fischer, and
of the late Drs.^Thomas Davidson and J. Gwyn Jeffreys, who, working on deep-
sea material, were equally considerate, I was enabled to complete this preliminary
work without clashing in the matter of priority; all the descriptions of particular
groups in the Bulletin alluded to being either intentionally subsequent or clearly
prior to the work of the above-mentioned gentlemen on the same groups. It need
hardly be said, that, when they led the way I was greatly the gainer from the ac-
knowledged experience and ability with which their tasks were performed, and
which greatly lessened my own labors.

t " On certain Limpets and Chitons from the deep Waters off the Eastern Coast
of the United States, by W. H. Dall." Proc. U. S. Nat Mus., Vol. IV. pp. 400-
414, April, 1882.

t Bulletin U. S. Geological Survey, No. 24, 336 pp., 8vo. Washington, Govern-
ment Printing Office, 1885.

MUSEUM OF COMPARATIVE ZOOLOGY. 173

Having enumerated the publications directly or indirectly related to
work on the " Blake " moUuscan collection, or portions of it, it remains
to characterize the final report, of which this is the first part, and to
make acknowledgment of the courtesies which have been extended to
me by various naturalists.

Owing to the confused state of the Antillean fauna, mentioned in
my Preliminary Report, and the wide distribution of many of the abyssal
species, the work of identifying species already described, or deciding
that they were not described, has required an unusual amount of labor,
altogether disproportionate to the apparent result. The existence of
quite a number of unfigured yet described species has rendered it proba-
ble that among those described some will eventually be found synony-
mous with forms previously known. This, however, must be expected
in any work covering so large a number of little known forms from an
imperfectly studied fauna. Those who have attempted similar work
will best understand and excuse such involuntary errors. The investi-
gation of the soft parts (in the small proportion of the collection in
which I found those preserved) has added some important facts, and en-
abled a better judgment to be formed of the value of certain anatomical
features, especially the gills, in general classification. I believe students
will find especial profit in considering the new data in the groups repre-
sented by Cuspidaria, Verticordia, Meiocardia, Dimya, and Pecten. It is
my impression, long since avowed, that, in the Pelecypods, no character
yet fixed upon for the division of the group into Orders is suflSciently
well defined to warrant its use for that purpose. They form a remarka-
bly homogeneous assembly, in which the characters fade out gradually,
or are imperceptibly modified in the transition from one minor group
to another. The use of the adductor muscles has been by common
consent of the best systematists practically abandoned. My friend, Dr.
Paul Fischer, in his admirable Manual, now in process of publication, has
essayed the use of the characteristics afforded by the gills for ordinal
distinctions. The data in the present paper will, I think, show that
this attempt can be no more successful than those which have preceded
it. In various publications during the last twenty years, especially on
the genera Siphonaria, Gadinia, Chiton, the true limpets, the Cocculini-
doe, and their allies, Dimya and Neoera {= Ctispidaria), I have shown the
extreme mutability of the branchiae within naiTow systematic limits ; that
they are organs which may exist or not exist in nearly allied genera ;
may be paired or unpaired structures ; may be found coincidently with
the presence of a lung, or in any stage of development from mere cuticu-

174 BULLETIN OF THE

lar wrinkles to compound and very complex lamellar organs. If ordinal
distinctions exist in the Pelecypods their fundamental basis has yet to
be made clear.

The present paper is prefaced with some general observations on
abyssal mollusks, the essential principles of which have appeared in
several scattered articles, at various dates, but which I have thought
it would be well to bring together and elaborate a little on the present
occasion. I have added a systematic table of the species referred to
in this part of my Report, deferring a complete and alphabetical index
until the whole shall be printed.

I have included the Brachiopods under the general title of Mollusca,
because I believe that, on the whole, the characters they present are
those of animals most nearly allied to Polyzoa and Mollusca, and that
there is nothing to be gained by splitting up the sub-kingdom thus con-
stituted, however clearly we may recognize its subdivisions. The dis-
memberment which has been proposed by various authors is more a
matter of phrase than of biological distinction. The intimate relation
of the Mollusks, as a group, to the Worms, is indicated by many embryo-
logical and histological characteristics. The indebtedness of all the in-
vertebrates to the vermian stock would bankrupt them to pay. To say
that the Brachiopods are Worms, in any proper or literal sense of the
words, appears to me as absurd as it would to assert that Vertebrates
are Ascidians. It is a clear case of including the greater in the less.
The relations are there, and should be fully recognized ; but the subject
should not be clouded by the miscomprehension of systematic values,
or the misuse of systematic terras. It should not be forgotten that our
knowledge of the development and even the adult anatomy of the
Mollusca is trifling compared with the field which remains unexplored.
Until more is known, we can well afford to acknowledge the inadequacy
of the basis for any comprehensive statement of relations which may be
termed conclusive.

During the progress of my studies I have had the privilege of con-
tinuous and friendly consultation with two veteran naturalists, Dr.
Thomas Davidson and Dr. John Gwyn Jeffreys, who now rest from
their labors. In the latter case, I have also had the advantage of be-
ing able to consult the original collection of Dr. Jeflfreys now forming
part of the U. S. National Museum.

To Prof, Spencer F. Baird, Director, and Mr. G. Brown Goode, Assistant
Director of the National Museum, I am indebted for the opportunity for
study of the collections made by the U. S. Fish Commission steamer

MUSEUM OF COMPAEATTVE ZOOLOGY. 175

*' Albatross" in the Antilles and on the eastern coast of the United States
south of Cape Hatteras. These collections, to which frequent reference
will be found in the following pages, though less extensive than those
of the " Blake," often supplemented the latter in a very helpful manner,
without which this report would have been in many cases less full and
accurate. They also contained many novelties which will form the sub-
ject of future study, and are occasionally noticed here when they tend
to throw special light on the subject in hand.

To the Eev. R. Boog Watson I am especially indebted for advice, criti-
cism, early copies of liis papers on the Challenger gastropods, and advance
proofs of some of his plates to appear in his final report. To Dr. Paul
Fischer, conchologist to the French expeditions on the " Talisman " and
" Travailleur," and to Mr. Edgar A. Smith of the British Museum, reporter
on the Challenger pelecypods, I am also under serious obligations.

To Dr. J. C. McConnell, whose pen drawings of shells for the process
adopted in illustrating this paper speak for themselves, every reader will
appreciate my indebtedness. It is proper to say, however, that this pro-
cess does not lend itself like lithography to the reproduction of texture
or surface, and that the details of description are in all cases to be taken
as conclusive, even when the minor characters mentioned are not fully
presented by the figures, or in the case of any supposed discrepancy.

The types of the species described will be found in the Museum of
Comparative Zoology at Cambridge, and in the U. S. National Museum.

It has not been thought necessary to reprint the descriptions published
in 1881, but, for the convenience of the student, the maximum length
of the specimen figured, given in millimeters, follows the references to
figures in the description of the plates.

The names adopted for species, etc., although conformed to Latin
construction and whatever their resemblances, are not to be taken aa
derived from any classical language. The ravages of the purists upon
our nomenclature, already disastrous, must be checked if possible, and
I know no other way of doing it than to declare the above-mentioned
names absolutely without meaning, whatever reminiscences they may
awaken. Compare the observations of Adanson, more than a century
ago, on this topic, in the prelude to his " Histoire Naturelle du Senegal."

The arrangement of tables of distribution, in area and depth, is de-
feiTed until the second part of this Report shall be printed. For a satis-
factory account of the faunae of the deep sea the data are wanting, and
can hardly be gathered in many years to come. The anatomical plates,
which will probably have to be lithographed, are also deferred. Some

176 BULLETIN OF THE

interesting species, taken up too late to have figures of them included
in the present set of plates, will be illustrated at the completion of the
work.

General Considerations.

In any account of deep-sea Mollusca it is advisable to premise, first,
that our knowledge of them is far from thorough or complete, and con-
sequently our conclusions about them must not be considered as final
in all cases. Secondly, the conclusions drawn from a study of the Mol-
lusca, with their special modes of life and reproduction, are frequently
quite different from the results which would follow from a study of
other auimals, such as fishes or sea-urchins, whose modes of life and
reproduction are widely different from those of moUusks. In short, in
drawing general conclusions we cannot include all classes of deep-sea
animals as if they formed a homogeneous population.

There are of course certain features in regard to which general rules
apply to all the inhabitants of the deeps, but they are few and liable to
modification with greater knowledge.

In discussing the Blake collections the work done by other expeditions
is often important for the proper understanding of the facts developed,
and consequently will occasionally be referred to for that purpose.

The "Challenger" and "Albatross" have both dredged in close prox-
imity to some of the Blake stations ; probably the richest haul on the
whole Challenger voyage was that obtained near St. Thomas in the
West Indies.

The collection of Mollusca obtained by the parties on the "Blake"
was notable in several respects beside those which may be reasonably
ascribed to the methods used in collecting. To the latter we may refer
the absence or rarity in the collection of very minute forms, which are
only accidentally preserved in the contents of a trawl net, even from com-
paratively shallow water ; while it is hardly to be expected that, in the
long period of washing and straining which the contents of a trawl un-
dergo while being hauled in from deep water, anything small enough
to go through the meshes of the net should be retained.

On the other hand, large shells appear to be rare in the great depths,
and when found are usually of great fragility ; so that their destruction
or serious fracture is almost inevitable. For these or other reasons,
deep-sea dredging has afforded few specimens of even moderately large
size, judged by the standard of shells living in shallow water or along
the shores. Among sheU-les's moUusks several which were of unusual

MUSEUM OF COMPARATIVE ZOOLOGY. 177

size have been found by different expeditions, one by the " Challenger,"
belonging to the DorididcB, being as large as an orange. All of these,
however, were of a peculiarly loose and gelatinous consistency. It
would seem as if a certain looseness of texture is required by the con-
ditions of great pressure which exist in the depths, in order to afford
that thorough permeation of the tissues by water necessary to equalize
the pressure. Whether this, as seems most probable, or the expansion
due to removal of the pressure on being carried to the surface, is the
cause of the looseness referred to, is uncertain, but that the deep-sea
animals of this group, as well as the fishes, exhibit such a state is cer-
tain. The shells almost without exception are extremely thin and light,
often reminding one of the delicate dwellings of some of the tropical
land snails ; to which a curious resemblance in form and texture may
frequently be noted.

The colors of the abyssal shells are almost always faint, or delicate,
though often very attractive from their very delicacy. The iridescence
or pearly character of the shell, in many groups, is often of peculiar
brilliancy and beauty, and it seems as if the texture of many shells not
intrinsically pearly was nevertheless of such a character as to give out
a sort of sheen in the abyssal species which is wanting in their shallow-
water relatives, and may be compared to pearliness.

While we do not find in any of the deep-sea species those sturdy
knobs and stout varices which ornament the turbinellas and conchs of
shallow water, and have made the great group of rock-purples, or Murices,
so attractive to collectors, there are nevertheless many abyssal shells
which have a delicate, and sometimes profuse sculpture, even more
elegant. The surface is frequently etched with a sort of shagreen pat-
tern, varied in detail and hardly perceptible except by a microscope, but
extremely pretty. In some the entire surface is adorned with profuse
arborescent prickles ; in others, pustulated with the most delicate shelly
blisters, systematically arranged, and which perish with a touch. In
most representatives from deep water of the family of scallops (Pecten),
the shell is as thin as a sheet of mica, its constituent prisms large
enough to be seen with the naked eye ; translucent, strengthened within
by delicate shelly riblets radiating from the hinge and often picked out
externally with delicate dots and splashes of orange, scarlet, or maroon.
Some of the family of top-shells {Trochidoe) are variegated with lovely
colors. In one form, those dredged in deep water by the Fish Com-
mission in the latitude of New York are stout, tall, and brightly varie-
gated with yellow and red-brown. In the specimens obtained from deep

VOL. XII. — NO. 6. 12

178 BULLETIN OF THE

water on the coast of Florida by Pourtal^s, and by the " Blake " in the
West Indies, the form is more depressed, the shell far more delicate,
the colors pale pearly tints of lemon and pink. It seems as if difi'er-
ences of temperature and nutriment, as between the north and the
tropics, were indicated in very similar ways, both by the dwellers in
the deep sea and those which inhabit the laud.

It might be thought that in the abysses, of whatever latitude, the
conditions would be so similar that we should find the same animal
presenting few, if any differences, from whatever part of the ocean it
might come. This is to some extent true of the great oceanic deeps
away from the continental shores and archipelagos. There the water
is always cold, and a certain and not very profuse mollusk fauna has
been found widely spread ; having apparently migrated from the polar
regions, and perhaps especially from the south polar regions, into the
deeps of both hemispheres. It is very necessary, in considering the dis-
tribution of the deep-sea mollusks, to bear in mind the different values
which the expression " deep sea " has had, and which, if confounded,
would give rise to serious errors.

Formerly, when dredging with the usual appliances in small boats,
one hundred fathoms was considered extremely deep, and specimens
from even half that depth were considered as having come from deep
water. This was proper enough when the collections were compared
with those from the shore between tides, or even from the adjacent
region below tide-marks, but which supported a growth of algae, either
ordinary sea-weeds, or the solid calcareous kinds known as corallines.
But when naturalists began to investigate at much greater depths, the
old terms lost their meaning.

For present purposes deep-sea mollusks may be taken to include all
those living at depths too great to allow algce of any sort to flourish,
the limit depending somewhat on the locality. Those living only above
that limit would form the littoral fauna, which, roughly speaking, may
be said to extend from the shores to about one hundred fathoms in
depth. With them in suitable places would be mixed many deep-water
forms, which extend their range to shallow water without being charac-
teristic of it.

The remainder of the sea would naturally be divided rather by tem-
perature than depth. But the temperature itself is somewhat dependent
upon the depth, the influence of the great warm currents of the ocean
rarely extending below seven or eight hundred fathoms, and this depth
corresponds roughly to a temperature of about forty degrees Fahren-

MUSEUM OF COMPAEATIVE ZOOLOGY. 179

heit. Below this it diminishes to the freezing point at the rate of about
one tenth of a degree to one hundred fathoms, forming the area which
will here be called the abyssal or benthal region. The area between
the abyssal and the littoral regions, chiefly on the slopes of the conti-
nental platforms, may be called the archibenthal area.* In the abyssal
areas the temperature at the bottom is known to be quite uniformly
cold, the supply of food sinking from the surface cannot vary much in
kind or quantity, and the distribution of life is comparatively sparse and
uniform, as might be expected.

But it is not in the abysses that the chiefest treasures of the dredger
are to be found, nor the richest abundance of species and individuals.
For these we must look to the archibenthal region skirting the conti-
nental shores or islands, where strong currents bring abundant food and
change of water, especially on relatively steep slopes which descend
from the hundred-fathom line toward the deeps ; there it is that the
richest harvest comes up in the trawl. Such spots were found by Pour-
tal6s near the Florida reefs ; by the " Blake" near Cape San Antonio and
off Grenada; by the "Challenger" near St. Thomas; and by the Fish
Commission off Martha's Vineyard. This increase is due to a variety of
causes. In the first place it is certain that warm waters are more favor-
able to a diversity of development and increase of individuals than cold
ones. They are more stimulating to the organization both ot the mol-
lusk and of the creatures which form its food, and both multiply in con-
cert. Secondly, the mollusk fauna of such regions, beside its population
derived by migration from the abysses, is made up in great part of forms
related to and connected with those which have developed along the
shores, which are constantly being carried by tide and other agencies into
deeper water than that in which they originated. There a certain pro-
portion of them continue to flourish, probably become more or less modi-
fied by change of food and environment, and so contribute to the variety
and number of the fauna. It is not always, perhaps not often, that
the species of the archibenthal region originally derived from the shores
are to be found on the shores immediately adjacent to the spot where
they are dredged. Often the littoral and adjacent archibenthal mollusk
faun£e are entirely, or almost entirely, dissimilar. This is the case off
the coast of Africa, or off the coast of New England, as observed by the
naturalists of the U. S. Fish Commission and the French expedition on
the " Talisman." But either in the far north or in the tropics we

* These areas have been generally recognized and called by yarious names.
Prof. A. Agassiz has termed the archibenthal area tlie " continental region."

180 BULLETIN OF TflE

shall find in shallow water of the appropriate temperature the species
in question. Drawing a line from Hatteras to Madeira, and consider-
ing the species dredged from the Atlantic Ocean north of this line, by
all expeditions up to 1883, in water more than one thousand fathoms
deep, we find that more than forty-two per cent of all the species of
mollusks are found somewhere or other living in water less than one
hundred fathoms deep. If we knew the littoral fauna of the tropics
better, it is probable that the percentage would be much increased. A
similar result has followed the study of the Blake collections, though
the exact figures are not ready to be given.

if, on the other hand, we consider the larger groups, such as genera
or families of mollusks, we shall find that the percentage of those pecu-
liar to the archibenthal and abyssal regions is extremely small, though
future researches are likely to enlarge it. We must regard the species
which have extended their range so far beyond their littoral area of
origin as having taken advantage of the uniform conditions of food and
temperature offered by the deep sea. In this connection, it should
be observed that the temperature limits of many species are more
sharply defined on the side of cold than on that of heat. The differ-
ence between 45° and 40° F. may absolutely check the distribution of
a species which would find no inconvenience in a rise of temperature
from 45° to 80°. It is probable that this is connected with the de-
velopment of the young, rather than the resisting powers of the adult
tnollusk, since it has been shown by Brooks and Eyder that a fall of
a very few degrees in temperature of the water was fatal to all the
floating embryos of the American oyster. A much greater rise would
probably only have hastened the development of the embryos.

It is quite within the limits of probability that archibenthal species
might rise to the littoral zone in some far distant locality, and by a
change in the direction or temperature of an ocean current all the in-
tervening deep-water individuals might perish, leaving two widely sepa-
rated colonies of the same littoral species. The weight of probability,
however, is greatly in favor of the continuous uniformity of the deep
sea as compared with the shores, and it is probable that they are ma-
terially modified only by physical changes of great importance, such
as raised the Isthmus of Panama above the sea.

On the steep slopes above referred to, the currents bring a great
variety and amount of material, which sinks to the bottom and furnishes
food or protection to the creatures which live there. Often the most
diverse elements enter into the accumulations. In one haul made by

MUSEUM OF COMPARATIVE ZOOLOGY. 181

Sigsbee near Havana, but in over four hundred fathoms, quite a large
number of common Cuban land shells were found, beside quantities of
marsh grass, bits of rattan, bamboo, sugar-cane, dead leaves, etc., all of
which were in good condition. If fossilized with the living sea shells
dredged with them, the deposit, as observed by Prof. Agassiz (Bull.
M. C. Z., Vol. V. p. 295), might sorely puzzle paleontologists of a future
century seeking to determine the circumstances under which it was
formed.

When we consider the great uniformity of texture of the deposits
forming the floor of the oceanic deeps, it would seem as if the envi-
ronment offered attractions for only a limited variety of forms. The
bottom is generally composed of extremely fine impalpable m\xd, and in
many portions of the abyssal area offers no stones or rugose inorganic
objects for sedentary mollusks to perch upon. It is not quite destitute
of such stations, however, and all are utilized by the abyssal popula-
tion. In the absence of stones, many unusual selections are made.
The chitinous tubes of hydroids and the irregular leathery dwellings
of tubicolous annelids are occupied, after their original owners are dead
or dispossessed, by various little limpets, such as Lepetella and Cocculiiia.
The long spines of the abyssal sea-urchins or echini offer a welcome
perch for species of Capulus, which, when they grow too large to find
a satisfactory foothold, secrete a shelly pedestal which serves them for
life. The carbonic acid in the water rapidly destroys the shells of such
mollusks as die in the great depths, so that they do not form gravelly
accumulations or " coquina " rock, as in shallower waters. A bivalve,
Modiola polita, related to the ordinary mussel of Northern seas, spins a
sort of nest of stout bygsal threads, in which it is completely concealed,
and which protects in its meshes, not only the young fry of the maker,
but various little commensal animals of different orders, such as mol-
lusks, worms, and Crustacea.

In the evolution of animal life two classes may be recognized : those
which maintain successfully the struggle for existence by facility in vary-
ing their superficial characters to meet the exigencies of their environ-
ment, — in short, by their facile plasticity ; and a smaller group, which
seem to have an innate strength of constitution which resists the influ-
ence of changes in the environment better by a dogged persistence in
their original form. These respond little, if at all, by external variation,
to the ordinary fluctuations of the physical world about them. This
has been noted by Darwin in birds, in his comparison between the vari-
ations of pigeons and the ** inflexible organization " of the goose. But

182 BULLETIN OF THE

it does not seem to have been realized among naturalists that natural
selection may act, in certain cases, as successfully by confirming the
inflexibility of a particular stock, as it does in others by seizing the
favorable variations of the vast majority of living beings which vary
indefinitely in all directions. Yet the former method may explain the
long persistence with but slight modification of certain organic forms
through immense periods of time and vast areas of distribution. The
few mollusks whfch have been recognized as wellnigh world-wide in
their spread, owe their uniformity, it is likely, to some such cause as
this. Those mollusks which live on algse and other vegetable matters,
and are ordinarily called phytophagous, are almost absolutely wanting
in the depths of the sea, where vegetation except as a sediment from
near the surface does not exist. "We have, then, at the bottom of the
ocean, a fauna almost exclusively of animal feeders, who receive their
sustenance chiefly from a constant gentle rain of dead or dying animals
whose normal existence is passed near the surface of the sea. For this
reason, the flesh-eaters of the deep sea, among mollusks at least, are not
obliged to prey upon each other to the same extent as the shallow-water
forms. The latter have to take part in a fierce struggle for existence,
among the vicissitudes of tidal and storm waves, variation in elevation
of land, and a vastly denser population of all sorts. In proportion to
the whole number, comparatively few of the shells dredged from deep
water show the drill-holes of enemies of their own kind, or the frac-
tures and injuries so common in shells from littoral dredgings.

It will be borne in mind, that the influence of natural selection on
variations in external characters, the conditions remaining about the
same, is toward the production of a stable equilibrium in specific charac-
ters in any species, and tlie more so when the characters presented for
its action are salient. For instance, if a few strong, long, sharp spines
protect a certain species against the attacks of fishes, this character
tends to be preserved in the species, and as a rule — confirmed by ob-
servation I may add — there will be little variation in the position and
number of the spines in question. In another case, where the same
end has been attained by the production of a profusion of similar spines,
the presence or absence or exact position of any one or more of the
spines is less important to the animal, is therefore less sharply restricted
by natural selection, and the tendency to vary within a certain range is
less affected, and persists. For these and perhaps other reasons also, it
may be stated as a general law in animal structures, that the greater
the number of similar parts in any member of an organic individual,

MUSEUM OF COMPARATIVE ZOOLOGY. 183

or of similar members, the greater the tendency to vary, first, in the
minor features of these parts or members as compared with each other,
and, secondly, in the number of similar parts or members in any indi-
vidual as compared with the average number characteristic of the
species,* or presented by any other individual of the species.

What is true of minor details in a complex series — where the com-
plexity relieves the detail of its importance as a part of the total, con-
sidered as a subject of selective action — is true of individuals of any
species, if we suppose the conditions uniform, and of such a kind as to
bear but lightly on specific characteristics. The latter, in mollusca, are
chiefly features of external form, color, and sculpture. Now, if the
form, color, and sculpture are unimportant in the struggle for existence,
in any given case, it follows that selective action will cease to affect
them, except so far as they may be indirectly dependent on other
characters which remain important and continue to be selected. Such
correlation has not been shown to be frequent in mollusks, if even its
existence can be said to have been demonstrated. I believe it to be
an important factor in a certain sort of cases, not however those we are
considering. The deep sea is doubtless very dark, if not absolutely
destitute of light. The water must be very quiet, the character of the
bottom almost uniformly soft and level. Most of the enemies of mol-
lusks there are blind, or at any rate can have little power of vision for
objects not luminous. The absence of violent motion in the water
removes from the category of modifying influences any mechanical ef-
fects of that medium upon the shell-fish contained in it. So it is evident
that the factors which would affect the restriction of " tendencies to
vary " in the above-mentioned characteristics, are almost eliminated
from the environment, especially if it be compared with that of littoral
species. The logical result therefore is, that we may expect in the
deep sea a very wide range of variation in form and sculpture within
the specific limits of the "flexible" species, and an almost complete
uniformity over very wide areas of the forms which we may consider
as " inflexible " species.

This is what, according to my judgment, is actually found. With

* This has been to some extent recognized by Owen in his discussion of vege-
tative repetition, and is illustrated by the variations of number in the teeth and
phalanges of cetaceans as compared with seals or other mammals; in the number
and variation of segments and segmental appendages in worms ; in the teeth of
the IlelicidcE, the coils of the shell in Polygyra, and in the spiny processes of certain
Muricida among mollusks. See American Naturalist for Sept., 1881, pp. 711, 712.

184 BULLETIN OF THE

few exceptions, — which may be assumed to belong to the " inflexible "
group, — in those cases where a considerable number of individuals of
one species were obtained by the " Blake," the variation in form and
sculpture is very wide, much more so than in most littoral forms. Owing
to the absence of light, color in abyssal mollusks is almost wanting ; but
in the species which possess it, as in some of the Pectens and Callio-
stomas, the range and variety of coloration within the species is very
wide. The tints are chiefly browns, pinks, and shades of yellow. The
sheen and play of colored light presented by the pearly species are re-
markably brilliant and fine. Among the archibenthal forms a notable
number are characterized by squarish red-brown spots on a light-colored
ground. I suspect that the abyssal mollusks are less active and ener-
getic than their congeners of the shorea This is indicated by the
looseness of the tissues, less favorable to prompt and violent motion
than more compact muscular apparatus would be. The tenacious
character of the mud forming the ocean floor, noticed by all explorers
of the deeps, would also tend to make motion through it slow and diffi-
cult. The delicacy of the shells, the extreme fragility and tenuity
which mark them, are inconsistent with liability to constant friction and
collision, either from the motions of the animal itself or of the waters
in which it lives. An exception may be noted in favor of the swimming
mollusks, such as the squids and cuttlefishes, but the deep-sea repre-
sentatives of these groups are far softer and less muscular than their
shallow-water relatives.

Much of the sculpture which is presented by the deep-sea species is
particularly beautiful from its delicacy. There seems to be an especial
tendency to strings of bead-like knobs, revolving striae and threads,
and delicate transverse waves. It is particularly notable that many of
the deep-sea forms, among all sorts of groups indifferently, have a row
of knobs or pustules following the line of the suture and immediately
in front of it. The representatives of the rock-purples, or Murices, a
group which, in shallow water, frequent the rocks and stony places, and
are there strongly knobbed or spinous, retain a similar character in the
deeps, but the processes in question are extremely delicate or foliaceous,
instead of being stout and strong. This is probably a reminiscence
of the time when their distant progenitors were shallow-water animals.

The groups which subsist upon other animals with a hard covering,
so that they have to bore or break their way to their food, are much less
numerous in the deep sea than those which feed upon soft tissues, or
kill their living prey by bites with poisonous fangs. The latter, Toxo-

MUSEUM OF COMPARATIVE ZOOLOGY. 185

glossa, as represented chiefly by the PleurotomidcB, outnumber any other
single group of raoUusks in the abyssal fauna.

The groups of less specialized character, such as the tooth-shells
(Dentalium), are rather abundant in species, more so than those of a
medium character which intervene between them and the highly spe-
cialized Pleurotomidce, but oiir knowledge of the deep-sea MoUusca is
yet too imperfect to afford any important generalizations on this score.
So far as yet determined, the groups systematically lowest in the scale,
such as the Ckitonidce, or mail-shells, are rare in deep water, yet the
representatives of this family found there belong to the more archaic
sections of their class. Some very interesting forms of the molluscoid
Brachiopoda are found in the abyssal region, among them some of the
largest known species ; but as a general rule the number of species is
small, and bears no comparison to that afforded by the archibenthal
area. In the early days of deep-sea exploration it was more or less
confidently anticipated that the deeps would afford specimens of ani-
mals characteristic of remote geological ages, which might have been
preserved there, little changed, while their shallow-water relatives
had perished from the earth. This expectation has been disappointed.
While there are numerous representatives of forms first made known
from Tertiary strata and hitherto unknown from shallow water, there
are not enough of these to characterize the abyssal mollusk fauna as
archaic in type, — not more, perhaps, than still exist in comparatively
shallow water ; none so remarkable as the Trigonia of austral seas, the
Pleurotomaria of the Antilles, or the Navtilus of the Spice Islands.
There is no relation of abyssal species with fossil species of moUusks
which compares with th"at between the land and fresh-water faunae of
to-day and those of the Carboniferous and Jurassic strata, whose Unios,
Physas, and Pupas are hardly more than specifically distinct from still
existing members of the same genera. I am impelled to insist more
forcibly on these facts from realizing that, in the reports on the raollusks
collected by the " Blake," as in the lists of those found by the Fish Com-
mission and by foreign dredging expeditions, many species find a place,
and attract general attention from intrinsic interest, which are not to
be counted as true abyssal species. Such are the Pleurotomaria, just
mentioned, of which two species were found by the " Blake " in 69-200
fathoms, and which belong to a group going back almost unchanged to
the earliest fossiliferous rocks, such as the Cambrian formation. One
gi'eat value of the Blake collection consists in the fact that it contains
representatives of animals from aU depths ;n the same general area,

186 BULLETIN OF THE

beginning near the shores and extending to the abysses, while most
deep-sea dredging parties have ceased work as soon as they came into
comparatively shallow water, for fear of confounding what were sup-
posed to be two sharply differentiated faunse. We learn from the work
of the " Blake " that the differentiation is much less marked than would
be anticipated, and that, in addition to the species found widely dis-
tributed over the floor of ocean, there is an important contingent of
species which are probably derived from the adjacent litorale, as well
as a tolerable number which are found in water of all depths, from a few
fathoms on the Florida coast to two thousand fathoms iu the adjacent
deeps, without affecting their external characters. Further exploration
in other seas will probably prove that there are local faunae in the ar-
chibenthal areas, as there are on the shores, a conclusion which would
accord well with what we learn from paleontology.

One point has been brought out by the study of the Blake collections
which was foreshadowed by Pourtales in his study of the deep-sea corals
dredged by him in the vicinity of the Florida reefs. It is being con-
firmed by present study of the mollusk fauna of our southern coast in
connection with the tertiary and quaternary fossils of the Atlantfc and
Gulf slopes. It is that a large proportion of the tertiary shells which
have been called Pliocene, or even Miocene, in this country and in
Sicily, still exist in a living condition near our shores. The tertiaries of
Calabria and of localities in the South of Italy having been pretty fully
studied, Pourtales was able to identify many of his corals with those
found by Italian paleontologists. Had our own tertiaries been half as
well known, or had he had a good collection of the shells of the southern
and West Indian tertiaries, he would have been able to recognize their
relations with his dredgings as being equally close. At least this is the
case with the molluscan fauna, if not with other invertebrate groups.
His dredgings, it should be clearly understood, were in the archibenthal,
and not the abyssal region, which last his operations never reached.
There is not enough known, so far, of the strictly abyssal mollusk
fauna, to afford a safe basis for generalization in connection with these
tertiaries. I may observe, however, that from middle Louisiana, ou
the edge of the Eocene beds, I have recently received certain fossils
which present every appearance of being a deep-water (archibenthal 1)
deposit, including Limopsis and several other characteristic forms. The
data which have been received relating to the circumstances under
which the fossils are found are as yet insufficient for a satisfactory
discussion of the subject.

MUSEUM OF COMPARATIVE ZOOLOGY. 187

SYSTEMATIC LIST OF THE SPECIES.

A. MOLLUSCOIDEA.
Class BRACHIOPODA.
Order ARTHROPOMATA.

Family TEREBRATULID^.

TEREBRATULA (Auct.).

Terebratula cubensis Pourtales.
Terebratula Moseleyi Davidson.
Terebratula Bartletti Dall.
Terebratula incerta Davidson.

TEREBRATULINA D'Orbigny.
Terebratulina Cailleti Crosse.

Family EUDESIID^.
EUDESIA King.
Eudesia .*loridana Pourtales.

Famly MEGATHYRID^.

MEGATHYRIS D'Orbigny.

Megathyris (Cistella) Barrettiana, var. ruhrotincta Dall.
Megathyris (Cistella) Barrettiana, var. ? Schrammi C. & F.
Megathyris (Cistella) lutea Dall.

Family PLATIDIID^.

PLATIDIA Costa.
Platidia anomioides Scacchi, var. radiafu, DaU.

Family THECIDIID^.

THECIDIUM Sowerby.

Thecidium mediterraneum Sowerby?
Thecidium Barretti Woodward.

188 BULLETIN OF THE

Order LYOPOMATA.

Family CRANIIDiE.

CRANIA RetziiM.
Crania Fourtaleoii Dall.

B. MOLLUiJiCA VERA.

Class PELECYPODA.
Family PECTINIDJE.

PECTEN Miiller.

Subgenus JANIRA Schumacher.
Janlra hemicyclica Ravenel.

Subgenus AMUSIUM Schumacher.
Amusitun Dalli Smith.

Section Pkopeamcsicm De Gregorio.

Amusiuin Poortalesianum Dall.

Amusitun Pourtalesianum, var. striatulum DalL

Amusium Pourtalesisinum, var. marmoratum DalL

Amusium cancellatum Smith.

Amusium Hoskynsi Forbes.*

Amusium Holmesii Dall.

Amusium Sayanum Dall.

Amusium alaskensis Dall.*

Subgenus PECTEN s. a

Pecten magellanicus Gmelin.
Pecten caurinus Gould.
Pecten nucleus Bom.
Pecten dislocatua Say.
Pecten phrygium Dall.
Pecten ezasperatus Sowerby.
Pecten ornatus Lamarck.
Pecten antillarum Rdcluz.
Pecten effluens Dall.

MUSEUM OF COMPAEATIVE ZOOLOGY. 189

Section PsBCDAMnsiuM H. and A. Adams.

Fecten imbrifer Loven.*
Pecten reticalns Dall.
Fecten thalassinus DalL
Feoten Sigsbeei DalL

HINNITBS Defranoe.
Hinnites Adamai Dall.

Fahilt LIMID^.

LIMA Bragifere.

Iiima squamosa Lamarck.
Iiima tenera Sowerby.
Itima inflata Lamarck.
Lima Mans Gmelin.
Iiima albicoma DalL

LIMATULA S. Wood.
Iiimatola setifera DalL

LIM.iEA Bronn.

Iiimaea Bronniana Dall.
Limaea Bronniana, var. lata Dall.

Family SPONDYLID^.

SPONDYLUS Linn€.
Spondylns Gussoni Costa.

PLICATULA Lamarck.
Plicatula spondyloidea Meuschen.

Family DIMYID^.

DIMYA Rooault.
Dimya argentea Dall.

190 BULLETIN OF THE

Family AVICULIDiE.

AVICULA Lamarck.
Avicula atlantica Lamarck.

Family MYTILID^.

MYTILUS Linne.
MytiluB ezustus Linue.

MODIOLA Lamarck.

Modiola polita Yerrill and Smith.
Modiola opifez Say.

CRENELLA Brown.
Crenella decussata Montague.

MODIOLARIA Beck.
Modiolaria lateralis Say.

Family ARCID^.

LIMOPSIS Sassy.

Limopsis minuta Philippi.
Limopsis tenella Jeffreys.
Limopsis antillensis DaU.
Limopsis cristata Jeffreys.
Limopsis aurita Brocchi.

PECTUNCULUS Lamarck.

Pectunculus undatus Linne.
Pectunculus undatus, var. scriptus Bom.
Pectunculus pectinatus Gmelin.
Pectunculus pectinatus, var. carinatus Dall.

ARCA Linn^.

Area pectunculoides Scacchi.

Area pectunculoides, var. orbiculata Dall.

MUSEUM OF COMPARATIVE ZOOLOGY. 191

Area polycyma Dall.

Area glomerula DalL

Area auriculata Lamarck.

Area lienosa Say.

Area retieulata Chemnitz.

Area Adamsi Shuttleworth.

Area Noae Linne, var. occidentalis PhilippL

Area umbonata Lamarck.

Area ectoeomata DalL

Area barbata Linng.

MACRODON Lycett.

Maerodon asperula DalL
Maerodon sagrinata DalL

Family NUCULID^.

NUCULA Lamarck.

Nucula aegeensia Forbes.
Nucula eymella Dall.
Nucula erenulata A. Adams.
Nueula erenulata, var. obliterata Dall.
Nueiila Verrillii DalL*

FAiiiLY LEDID^.

LEDA Schumacher.
Subgenus YOLDIA Morch.

Yoldia solenoides Dall.
Yoldia liorhina Dall.

Subgenus LEDA s. 8

Leda Carpenter! DalL

Leda messanensis Seguenza.

Leda solidula Smith.

Leda vitrea D'Orbigny, var. cerata Dall.

Leda eoneentriea Say.*

Leda acuta Conrad.

Leda solidifacta Dall.

Leda Verrilliana Dall.*

192 BULLETIN OF THE

Leda Bushiana Verrill.*
Leda subaequilatera Jeffreys.
Leda hebes Smith.

Section Satubkia Segoenza.

Leda pusio Pbilippi.
Leda quadrangiilaria Dall.

Section Neii,onei;.la DalL
Leda corpulenta Dall.

MALLETIA Desmoulins.
Section Tindabia BellardL

Malletia cytherea Dall.
Malletia Smithii Dall.*
Malletia dilatata PhiUppi.

Family CARDITID^.

CARDITA Brugifere.
Cardita domingensis D'Orbigny.

Family CRASSATELLID^.

CRASSATELLA Lamarck.
Crassatella floridana Dall.

Subgenus ERIPHYLA Gabb (em.).
Eriphyla parva C. £. Adams.

Family ASTARTID^.

ASTARTE J. Sowerby.

Astarte Smithii Dall.

Astarte Smithii, var. globula DalL

Astarte nana Dall.

CIRCE Schumacher.
Circe (Gouldia) cerina C B. Adams.

MUSEUM OF COMPARATIVE ZOOLOGY. 193

Family UNGULINID^.

DIPLODONTA Bronn.

Diplodonta tiirgida Verrill and Smith.
Diplodonta venezuelensia Duuker.

Family LUCINID^.

LUCINiA. Brugi^

Lncina antillarmn Reeve.
Lucina sombrereuaia DalL
Lncina lencocyma Dall.
Lncina funiculata Reeve,
Lncina lenticnla Reeve.
Lncina scabra Lamarck.
Lncina sagrinata Dall.
Lncina qoadrisulcata lyOrbigny;

LORIPES PolL

Loripes compresBa Dall.
Loripes lena Verrill and Smith.

CRTPTODON Turton.

Cryptodon orbicnlatns Seguenza.
Cryptodon pyriformis Dall.
Cryptodon flezuosns Montague.

Family CHAMID^.

CHAMA BrugiSre.

Chama lactuca DalL
Chama sarda Reeve.

Family CARDIID^.

CARDIUM Linn^.

Cardinm ceraniid\im Dall.
Cardium medinm Linn^.
Cardinm peramabilis Dall.
Cardinm muricatum Linn^.
Cardium laevigatum Linn^.
Cardium serratum Linne.

VOL. XII. — NO. 6. 13

194 BULLETIN OF THE

Family ISOCARDIID^.

ISOCARDIA Lamarck.
Subgenus MEIOCARDLV H. and A. A flams.
Meiocaxdia Agassizii Dall.*

CALLOCARDIA A. Adams.

Subgenus VESICOMYA Dall.

Vesicomya subquadrata Jeflfreys.*
Vesicomya atlantica Smith.
Vesicomya pilula Dall.
Vesicomya venusta Dall.

Family VENERID^.

CYTHEREA Lamarck.

Subgenus DIONE Megerle.

Dione hebraea Lamarck.
Clone albida Qmelin.

Section Veneriglossa Dall.
Dione (Veneriglossa) vesica Dall.

VENUS (Linne') Deshayes.
Venus pilula Reeve.

Subgenus CHIONE Megerle.
Chione pygmaea J amarck.
CUone Ccincellata Lamar..k.

Family PETRICOLID^.

PETRICOLA Lamarck.
Petricola diveiricata Chemnitz.

Family TELLINIDiE.

TELLINA Linne.

Tellina Antoni Philippi.
Tellina squamifera Deshayes.

MUSEUM OF COMPARA.TIVE ZOOLOGY. 195

Tellina sybaritica Dall.
TeJlina tenera Say.
Tellina? plectrum Hanlej.
Tellina Gouldii Hanley.

Family SEMELID^.

ABRA (Leach) Risso.

Abra longicallis Scacchi.
Abra lioica Dall.

ERVILIA Turton.
Ervilia nitens Montague.

CUMINGIA Sowerby.
Cumingia tellinoides Conrad.

SEMELE Schumacher.

Bemele obliqua Wood.
Semele cancellata D'Orbigny,

Family POROMYID^.

POROMYA Forbes.

Poromya granulata Nyst and Westendorp.

Section Cetoconcha Dall.
Poromya (Cetoconcha) albida Dall.
Poromya (Cetoconcha) elongata Dall.
Poromya (Cetoconcha) bulla Dall.
Poromya (Cetoconcha) margarita Dall.

Family VERTICORDIID^.

VERTICORDIA Wood.

Verticordia acuticostata Philippi.
Verticordia Woodii Smith.
Verticordia perversa Dall.
Verticordia Seguenzae Dall.

Subgenus TRIGONULINA D'Orbigny.
Trigonulina ornata D'Orbigny.

196 BULLETIN OF THB

Section Euciboa DalL
Verticordia (Euciroa) clegantlaslma DalL

Subgenus PECCHIOLIA Meneghini.
Peochiolia argentca MaritL*

Subgenus HALIRIS DalL

Hallria Fisoheriana Dall.

MYTILIMERIA Conrad.
Mytilimeria Nuttallii Conrad •

LYONSIELLA Saw.
Lyonsiella insctUpta Jeffreys.*

Family CUSPIDARIID^.

CUSPIDARIA Nardr..

Subgenus CUSPIDARIA s. s.

Cnspidarla rostrata Spengler.

Cuspidaria roatrata (? var.) microrhina Dall.

Cuapidauia JeSreyal Dall.

Cuapidarla obeaa Loven.

Cuspidaria? arouata DalL

Subgenus CARDIOMYA A. Adams.

Cardiomya caUfortiica DalL*
Cardiomya perroatrata DalL
Cardiomya costellata Deshayes.
Cardiomya coatellata. var. curta Jeffreys.
Cardiomya coatellata, var. corpulenta DalL
Cardiomya striata Jeffreys.

Subgenus LEIOMYA A.Adams.
Leiomya adnnoa Ciould.*

Section Yulcanomta Dali*
(? Leiomya) Vnloanomya Smithli Dall.*

Section Plectodon Carpenter.
Iieiomya (Plectodon) scaber Carpenter.*
Leiomya (Plectodon) granulata Dall.
Leiomya (l^leotodon) granulata, var. velvetina Dall.

MUSEUM OF COMPARATIVE ZOOLOGY. 197

Section Rhinoclama DaU and Smith.
Leiomya (RMnoclama) halimera DalL*

Subgenus TROPIDOMYA DaU and Smith.
Tropidomya abbreviata Forbes.*

Subgenus HALONYMPHA Dall and Smith.
Halonympha claviculata DalL

(Genua?) MYONBRA DaUandSmiti.

Myonera paucistriata Dall
Myonera undata Verrill.
Myonera lamellifera Dall.
Myonera limatnla Dall.
Myonera laticella Dall.*

Family ANATINID.^.
PERIPLOMA Schumacher.

Periploma fragilis Totten.*
Periploma papyracea Say.

THRACIA Leach.

Thracia Stimpsoni Dall.*
Thracia corbuloidea Blainville.*
Thracia distorta Montague.*
Thracia pheuseoUna Lamarck.

ASTHBNOTH^RUS Carpenter.
AsthenothaeruB Hemphillii DalL

Subgenus BUSHIA DaU.
Bushla elegems DalL

Family PANDORID^
PANDORA Hwass.

Subgenus CLIDIOPHORA Carpente*.
Clldlophora carolinensis Bush.
Clidiophora trilineata Say.*
Clidiophora Qouldiana DalL*

Subgenus PANDORA s. s.
Pandora (Kennerlia) Bnshiana Dall.*

198 BULLETIN OF THE

Family CORBULID^.

CORBULA Brugifere.

Corbula cubaniana D'Orbigny.
Corbula Baxrattiana C. B. Adams.
Corbula Swiftiana C. B. Adams.
Corbula Dietziana C. B. Adams.
Corbula disparilis D'Orbigny.
Corbula (Taeniodon ?) cymella Dall.
Corbula Krebsiana C. B. Adams.*
Corbula Chittyana C. B. Adams.*
Corbula Kjaeriana C. B. Adams.*

BASTEROTIA Mayer.
Basterotia quadrata Hinds, var. granatina Dall.

Family SAXICAVID^.

SAXICAVA F. de Bellevue.
Saxicava azaria Dall.

Family PITOLADID^.

XYLOPHAGA Turton.
? Xylophaga aby^orum DalL

This Report contains twelve new subgenera or sections, and eighty-one new
species. The species marked by an asterisk are introduced for purposes of
illustration, etc., and were not collected by the " Blake." The total amounts to
thirteen species and varieties of Brachiopods and two hundred and fourteen
species and varieties of Pelecypods obtained by the " Blake," beside the thirty
or more species casually mentioned but not collected.

MUSEUM OF COMPARATIVE ZOOLOGY. 199

MOLLUSCOIDEA.
Class BRACHIOPODA.
Order ARTHROPOMATA.

Family TEREBRATULID^.
Genus TEREBRATULA adctorum.

Terebratula cubensis PouEXALfes.

Terebratula cubensis, Pourtales, Bulletin M. C. Z., I. p. 109, 1867 ; Dall, ibid., III.
p. 3, pi. i. figs. 2, 8-16, 1871 ; ibid., IX. p. 103, 1881.

Habitat. Station 45, 101 fms. ; Station 16, 292 fms. ; Sigsbee, oflF Ha-
vana, 175 and 400 fms.; Lat. 26° 31', Lon. 85° 3', 119 fms. ; Barbados, 100
fms.; Stations 231 and 232, St. Vincent, 95 and 88 fms.; Stations 193 and
202, Martinique, 169 and 210 fms. ; station 155, Montserrat, 88 fms., bottom
temperature 69°.0 F. ; Station 167, Guadalupe, 175 fms. ; Stations 249, 253,
and 254, near Grenada, in 262, 92, and 164 fms. ; and Stations 273, 276, 282,
293, 296, and 300, about Barbados, in 103, 94, 154, 82, 125, and 82 fms., re-
spectively. In general, at a depth of 80-400 fms., sandy or stony bottom, with,
the temperature ranging from 50° to 70° and averaging about 58°.5 F.

This species has been fully described, figured, and discussed by me in the
papers referred to, especially volume third of this Bulletin, and nothing more
can be added to the data there accumulated except the additional localities
here recorded.

Its distinctness from T. vitrea may be considered as fully established.

Terebratula Moseleyi Davidson.
Terebratula Moseleyi Dav. Chall. Rep. Brach., p. 30, pi. xi. figs. 12-14, 1880.

A specimen sent to Mr. Davidson was identified by him as this species. It
was obtained at Station 193, off Martinique, in 169 fms., sand, shell, and dark
mud, the bottom temperature being 51°.0 F. The Challenger specimens were
dredged west of Kerguelen Island in the Southern Ocean, at Station 148, lati-
tude 46° 471 south, and longitude 51° 37' east of Greenwich, on a rocky bottom
in 210 fms.

200 BULLETIN OF THE

Terebratxila Bartletti Dall.

Terebratula Bartletti Dall, Am. Nat., Nov. 1882, p. 885.

Plato TI. F1k8< 4 a-c.

Shell whitish or often with a delicate madder-brown tinge, moderately thin,
ovoid, inflated, polished, with occasional traces of delicate evanescent ex-
tremely fine radiating lines, especially on the sides near the hinge line ; apex
of the neural valve rather attenuated, curving over and closely appressed to
the apex of the haemal valve ; foramen complete, small, its lower margin pro-
duced into a sharp point lying over the apex of the haemal valve and conceal-
ing it ; area short, very wide, triangular, bounded by a sharp carina on each
side, concave, with a median slightly impressed line, posterior margin a little
arched in a posterior direction; it is entirely concealed in the living shell,
being as before mentioned closely appressed to the outer surface of the other
valve; cardinal border rather pointedly arched, teeth small but stout; margin
of the valve smooth, flexuous ; it falls away a little from a point immediately
in front of the teeth, then continuing forward is emarginated and its front
border strongly squarely produced upward and forming two well-marked
comers between which the front margin is nearly straight ; outer surface of
the valve roundly convex. Haemal valve with the margin correspondingly
flexuous, generally rounded but with a more or less obtuse ridge extending
toward the beak from the inner angles of the anterior flexuosity: beak rather
pointed, incurved cardinal process small, semicircular, fimbriated in all cases,
showing six to eight anteriorly pointing irregular denticulations ; cardinal
plate divided, its lateral platforms wide, deeply concave ; tooth sockets small,
narrow, close to the margin of the beak ; loop large, very square, curved up-
ward, proportionally wider and shorter than in T. cubensisy with a less convex-
ity in the median line, and without the lateral notches and median prominence
of T. cubensis. Interior of valves smooth except for the muscular impressions
and certain ridges due to their changes in the development of the individual ;
in the haemal valve there is an obtuse ridge (seen through the shell it resem-
bles a septum as in Waldheimia) between the abductor scars, in the neural
valve there is a Well-marked groove in the same place : in T. cubensis the ante-
rior margin of the adductor scars is underneath and behind the anterior margin
of the loop ; in this species (as in T. vitrea) they are considerably in advance
of it, a circumstance resulting from the greater bi;lk of the soft parts in the
latter species, compared with the size of the shell. The measurements in an
adult individual are as follows. Lon. of neural valve 40.0, of haemal do. 38.0,
lat. 31.5, lat. of anterior flexuosity 22.0; beak to anterior edge of loop 8.0, to
points of crura 5.5, width of anterior margin of loop 6.5; diameter 26.75 mm.
The greatest width of the shell is behind its middle in T. vitrea, as already
pointed out by me (Bull. Mus. Comp. Zobl., III. No. 1, p. 3, 1871) ; in the
present species it is anterior to the middle of the shell.

MUSEUM OF COMPARATIVE ZOOLOGY. 201

.Habitat. Stations 290, Barbados, 73 fms. ; 232, St. Vincent, 88 fms.; 155,
Montserrat, 88 fms. ; 253, Grenada, 92 fms. ; 273, Barbados, coral and shells,
103 fms. ; 45, in Lat. 25° 33' N., and Lon. 84° 21' W. Gr., 101 fms.; 177,
Dominica, sand and shells, 118 fms. ; 157, Montserrat, sand and stones, 120
fms. ; 297, Barbados, stones, 123 fms. ; 258 and 254, Grenada, 159 and 164
fms. ; 193, Martinique, 169 fms. ; 291, Barbados, 200 fms. ; 139, Santa Cruz,
sand and gravel, 218 fms. ; 147, St. Kitts, 250 fms. Its location, therefore,
appears to be between seventy and two hundred and fifty fathoms, in water
varying from 51°.0 to 69°.0 Fahrenheit in temperature.

The relations of this form appear to be with T. vitrea, T. cuhensis, T. sphe-
noidea, and T. scillm. Its assemblage of characters does not appear to be shared
by any of those forms. The rather large number of specimens of all ages, col-
lected as above, show its range of variation very well. Those who would unite
all the above-mentioned species under one name, would doubtless include the
present form within that limit, and logically so. I do not see my way clear,
however, whatever may be thought to be the value of a " species," to ignore
what appear to be constant diflferences in the organisms under consideration.
It is probable that there are too many specific names in the group of Terebra-
tvla of which T. vitrea is an example, a number of additions having been re-
cently made to the list. The present form is certainly more difi"erentiated
from either vitrea or cuhensis than several which have been named and are
generally accepted. The form of the loop resembles closely that of T. siracu-
sana Seguenza (Bull. Malac. Ital., IV., tab. 4, fig. 13), its general form is more
like T. scillcB Seg. (1. c, tab. 3, fig. 8), at least like the variety mentioned.
Other discriminating characters may be found mentioned in the preceding
description, which, with the figures, will be a sufficient means for identification.
The anterior flexuosity is often, though not usually, as strong relatively in the
young as in the adult. The appressed neural apex is very constant.

Terebratula incerta Davidson.
Megerlia incerta Davidson, Challenger Brach., p. 49, pi. xi. figs. 17, 18, 1880.

Plate TI. Figs. 6, 6 a.

Habitat. Challenger Expedition, Mid- Atlantic, Lat. 1° 47' N., Lon. 24° 26'
W., 1850 fathoms. Blake Expedition, Stations 235 and 236, in 1507 and 1591
fms., oozy bottom, off Bequia, bottom temperature 39°.0 F. ; and Station 16,
292 fms., off Morro Light, Havana, Cuba, bottom temperature 55°. 6 F., one
specimen only.

This species was obtained of adult size at the stations cited. It is readily
recognizable from Mr. Davidson's excellent figures (by his kindness I com-
pared specimens) and the peculiar and characteristic radiating filaments which
surround the base of the peduncle. In all the specimens examined the loop is
incomplete, or rather the crura are not Timted. but the soft parts, the setse, and

202 BULLETIN OF THE

the shell canals are those of Terebratulina. It, and perhaps T. Murrayi Dav.,
may be considered Terebratulinas in which the crura do not unite.

Genus TEREBRATULINA D'Orbignt.

Terebratulina Cailleti Crosse.

Terebratulina Cailleti Crosse, Journ. de Concliyl., XIII. p. 27, pi. i. figs. 1-3, 1865;
Dall, Bull. M. C. Z., III. p. 10, 1871 ; IX. p. 103.

Habitat. Barbados, 100 fms. ; Sigsbee, off Havana, in 80, 119, 127, 240, and
450 fms. ; Yucatan Strait, 640 fms. ; Station 2, 805 fms. ; West Florida, 30
fms. ; Station 16, 292 fms.; Station 20, 220 fms. ; Station 44, 539 fms. ; Sta-
tion 45, 101 fms. ; off Morro Light, Station 16, 292 fins. ; Santa Lucia, Sta-
tions 216 and 218, 154 and 164 fms. ; St. Vincent, Stations 224, 231, and 232,
114, 95, and 88 fms. ; Dominica, Station 177, 18 fms.; Montserrat, Stations
154, 155, and 156, in 298, 88, and 88 fms. ; Grenada, Stations 246, 247, 253,
and 254, in 154, 170, 92, and 164 fms. ; off the Grenadines, Station 238, in 127
fms. ; Barbados, Stations 272, 273, 276, 278, 281, 282, 290, 291, 292, 296, and
298, in 76, 103, 94, 69, 288, 154, 70, 200, 56, 84, and 120 fms. respectively.

As this series of localities proves, this little species is abundant and widely
distributed in the Antillean region, frwn which it extends southward to the
vicinity of Pernambuco and Rio de Janeiro. It occupies for this fauna the
place taken by T. caputserpentis L. and its varieties in the north. It ranges
between 30 fms. and 805 fms. in depth, and exists in water the temperature of
which may be 45°. 0 to 75®.0 F. Its favorite location, however, appears to be
at a depth of between 100 and 200 fms., and in water of the temperature of
60° .0. It has been fully discussed in the papers cited.

Family EUDESIID^.

Genus EUDESIA Kixo.

Eudesia floridana PouRXALfes.

Waldheimia floridana Pourtales, Bull. M. C. Z., I. p. 127 ; Dall, Ibid., HI. p. 12,
pi. i. fig. 3, pi. ii. figs. 1-3, 1871 ; IX. p. 103.

Habitat. Off Sand Key, 125 fms. ; Sigsbee, off Havana, 175 fms. ; Lat.
26° 31', Lon. 85° 3', 119 fms.; Station 45, 101 fms.; Station 5, 229 fms.;
Station 19, 310 fms. ; Station 291, 200 fms., Barbados.

The generic name Waldheimia being preoccupied for a genus of insects, as
heretofore pointed out, Eudesia, King, is the next in order of priority, and
should therefore be adopted, although in its original significance it was merely
a synonym of Waldheimia King.,

MUSEUM OF COMPAKATIVE ZOOLOGY. 203

Family MEGATHYRID^.

Genus MEGATHYRIS D'Orbignt.

Subgenus CISTELLA Gray.

Cistella Baxrettiana Davidson, var. rubrotincta Dall.

Cistella Barrettiana Dall, Bull. M. C. Z., IX. pp. 103, 104.

Argiope Barrettiana Davidson, P. Z. S., Feb. 1866, p. 103, pi. xii. fig. 3.

Argiope antiUarum Crosse & Fischer, Journ. de Cbnchyl., XIV., July, 1866, p. 270,

pi. viii. fig. 6.
Cistella (Schrammi var. 1) rubrotincta Dall, Bull. M. C. Z., III. p. 19, pi. i. fig. 6, 1874.

Habitat. Sand Key, 80 fms. ; Station 2, 805 fms.; Yucatan Strait, 640
fins. ; Station 45, 101 fms.; Station 20, 220 fnis, ; Barbados, 100 fms.; Sigs-
bee, off Havana, 450 fms. ; Station 276, 94 fms. ; Station 231, St. Vincent, 95
fms. ; Tortugas, 43 fms.; Station 297, 170 fms., off Grenada; Station 132, 115
fms., Santa Cruz ; Station 155, 88 fms., near Montserrat, W. I.

Tbis pretty little species bas about the same range in depth and temperature
as T. Cailleti. The above s;ynionymy represents the conclusions of Mr. David-
son and myself, after several years of correspondence and the study of quite
abundant material.

Cistella (Barrettiana var.?) Schrammi Crosse & Fischer.

Cistella {Barrettiana (?) var.) Schrammi, Bull. M. C. Z., IX. p. 104.
Argiope Schrammi, Crosse & Fischer, 1. c, p. 269, pi. viii. fig. 6, 1866.

Habitat. Station 45, 101 fms. ; Barbados, 100 fms.

There is much doubt as to the distinctness of this form from C. Barrettiana,
which seems very variable in sculpture and color.

Cistella lutea Dall.

Cistella lutea Dall, Bull. M. C. Z., III. p. 20, pi. i. fig. 5, pi. ii. figs. 4-8, 1871 ;

Ibid., IX. p. 103.

Habitat. Sig-*bee, off Havana, 80 to 127 fms. ; Barbados, 100 fms. ; Station
21, 287 fms. ; Tortugas, 30 fms.

Owing to the differences in the form of the shell and especially of the sep-
tum, Mr. Davidson was inclined to regard this as a valid species, and so stated
in his last communication on this subject. Jt^may, however, be only an
extreme form of Barrettiana, though intermediate specimens are still wanting.

204 BULLETIN OF THE

Family PLATIDIID^.
Genus PLATIDIA Costa.

Platidia anomoides Scacchi.

Platidia anomoides Dall, Bull. M. C. Z., IX. p. 104.

Terebratula anomioides Scacchi, Philippi, Moll. Sicil., IL p. 69, pi. xviii. fig. 9, 1844.

Habitat. Near Morro Light, Cuba, Station 16, 292 fms. ; Station 253, 92
fms. ; Barbados, Station 280, 221 fras. ; Grenada, Station 260, 291 fma.; St.
Vincent, Station 232, 88 fms. Also San Diego, California, Orcutt; and off
the coast of North Carolina in 16 fms. by the U. S. Fish Commission, 1885.

Variety radiata Dall, Proc. U. S. Nat. Mus., 1885, p. 661.

Shell radiately ribbed with small irregular ribs, apex of the dorsal valve
not notched but even with a trace of flattened area ; haemal valve deeply
notched ; mai^in with rather prominent setae lying in the grooves correspond-
ing to the ridges ; labia as usual with short brachial membrane and fringe
behind them ; a broad smooth area of membrane in front of them ; about
25-30 single brachial processes on each lobe turned down and curled under ;
the point of the septum projects in front of the broad membranous area ; the
anterior labium and perhaps both of them, somewhat reinforced by chitine ;
size the same as the typical form.

Habitat. Station 139, off Santa Cruz, in 218 fms., bottom temperature
51''.0; sessile on smooth Terebratula. The Califpmian specimens are also
of this variety.

This specimen was sacrificed to get at the soft parts. This is the only form
in which "^he brachia are turned downward, and the only living form in which
the haemal v2^e is notched in the adult state, as far as known to me.

Family THECIDIID^.
Gends THECIDIUM Sowebbt.

Thecidium mediterraneum Sowekbt.
? Thecidium mediterraneum Sowerby, Dall, Bull. M. C. Z., IX. p. 104.

Habitat. Station 241, 163 fms.

The specimens being loose dorsal valves, it is possible that they may belong
to the preceding species ; but Mr. Davidson thought not. No complete speci-
mens were found in the Blake collection.

MUSEUM OF COMPAEATIVE ZOOLOGY. 205

Thecidium Barretti Woodwakd.

Thecidium Barretti, Woodward, Dall, Bull. M. C. Z., IX. p. 104. Dayidson, Geol.
Mag., I. pi. ii figs. 1-3, 1864; P. Z. S., 1866, p. 104.

Plate TI. Fix. 2>

Habitat- Barbados, 100 £ms. ; Station 232, St. Vincent, 88 Tms. ; Station
115, Montserrat, 88 fhis.

This rare species was identified by Mr. Davidson after •comparison with his
type. It is here satisfactorily figured, so far as the interior is concemedj for
the first time, acccording to that eminent authority.

Order LYOPOMATA Owen.
Family CRANIID^ Gray.

Gbnus crania Ketzius.

Crania Pourtalesii Dall.

Crania Pourtalesii DaU, Bull. M. C. Z., IX. p. 104 ; Ibid., III. p. 35, pL i. fig. 7, 1871.

Habitat. St. Vincent, W. I., 88 fnis., Station 232; Sand Key, Fla., 105
fms. ; off the Sambos, 116 fms. (single valves).

This species is probably abundant in suitable places, but no satisfactory haul
of them has yet been made.

206 BULLETIN OF THE

MOLLUSCA VERA.

Class PELECYPODA Goldfuss.

Family PECTINID^.

Genus PECTEN Mulleb.

Pecten Miiller, Prodr. ZooL Dan., p. xxxi, 1776. Type Pecten {Ostrea) maxi'

mus L., L c, p. 248.

This ancient genus has been cut wp into many sections, most of which shade
into one another by imperceptible gradations, or interchange characters, or
would belong to different sections at different stages of post-embryonic growth.
For purposes of convenience and usefulness most of these sections were better
discarded, as a name without any essential characters is merely an incumbrance
to workers and a stumbling-block for learners. For my own purposes I find
the following arrangement convenient : 1. Pecten, with the subgenera Janira ;
Amusium and section Propeamusium ; Pseudamusium and section Camptonedes ;
Pecten typical and the sections Pallium and Lyropecten ; 2. Neithea ; 3. Hemi-
pecten; 4. Ilinnites,

In form of shell and characters of hinge, Dimya is related to Pecten, and by
its habit to Ilinnites ; in its shell structure, it is nearer the Aviculidoe and
Ostreidos ; in its anatomical peculiarities it is archaic, foreshadowing the pearl-
shells, the oysters, and the scallops in different degrees. It is well entitled to
family rank, and for present purposes I prefer to arrange it between the Pecti-
nidce and the Aviculidoe, though no linear arrangement will express all its
relations.

The form of the foot in typical Pecten is recorded as cylindrical, with or with-
out the posterior margin grooved. In P. caurinus the groove is deep, the stem
calibre uniform, the distal end a little swollen, with a minute slit and radiated
aperture on the posterior median line, the whole extremely phallic in appear-
ance ; in P. antillarum the foot is grooved, subcylindrical and worm-like, with
no perceptible slit at the tip, and that of P. nucleus Born is much the same ;
P. irradians has a beginning of a sucker-slit and hardly expanded tip ; P. ma-
gellanicus has the tip much enlarged, solid, with a large sucker ; when we get
to Amusium pleuronectes we have a spade-shaped tip and well-developed sucker,
with moderate stem ; and, finally, in A. Dalli the sucker is large, hood-shaped,
thin-walled and darkly pigmented, with a broad base abruptly enlarged from
a very slender stem. Similar modifications appear in the anal extremity,
which from elongate and free varies to the usual appressed type of most bi-

MUSEUM OF COMPARATIYE ZOOLOGY. 207

valves. Other characters seem equally interchangeable, such as the armature
of the Ups, which may be internally striate or smooth, externally smooth, papil-
lose, or arborescent.

All these facts confirm me in my belief that the subdivisions of the group
may advantageously be limited to a comparatively small number.

SCBGENUS JANIEA SCHUMACHEB.

Pecten (Janira) hemicyclica Ravbnel.

Janira hemicyclica Tuomey & Holmes, Miocene Fos. S. Car., p. 25, pL viii. figs.

1-4, 1855.
Pecten hemicyclicus Ravenel, fide T. & H. 1. c.

Plate VI. Tig. S.

Two lower valves of this species were dredged on the west coast of Florida
by the Bache in 19 fathoms. It is found not very rarely on the east and west
coasts of South Florida, and often identified as P. ziczac. The ribs on the flat
valve difl'er greatly in diflferent specimens, being sometimes obsolete and some-
times very strong. The color of this valve is much as in P. ziczac. The color
of the convex valve and its sculpture are quite different from those of P. ziczac,
which grows to a considerably larger size at present. The fossil specimens
of hemicyclica, as figured, are larger than any recent ones I have yet heard of,
The very young of this species are externally indistinguishable from the fry of
P . magellanicus Gmelin and Amusium pleuronectes. The transverse rugosities
or grooves of the hinge-line referred to in Pseudamusium thalassinum are well
marked in the fry of this species, and very evident traces of them are visible
in the adult. In the young they occupy a lanceolate area on each side of the
cartilage pit, and are shown in the figure, considerably magnified, on Plate VI.
These shells and some other young fry are not to be distinguished from young
Pecten similis Laskey, of most collectors. I find fully half the " P. similis " of
the Jeffreys collection to be of this character. Many of them might have
grown to be that species, but many probably might not. Unusual localities,
such as Korea or Jamaica, quoted for P. similis (genuine) on the authority of
Dr. Jeffreys, should be suspected or held for more information.

Subgenus AMUSIUM (Bolten) Schumachek.

Historical Synonymy.

Amusium Rumphius, Amboinische Rariteitkamer, pp. 144, 188, pi. xlv. figs. A, B,
1705. Klein, Tent. Meth. Ostrac, p. 134, 1753. Martini, Verzeichn.
Samml. Nat., 1774.

Synonymy Proper.

Amusium Bolten, Mus. Boltenianum, ed. i. p. 165, 1798 ; Pecten pleuronectes auct. (no
description or type mentioned).

208 BULLETIN OF THE

Amusium Megerle von Muhlfeld, Entwurf. (etc.) Mag. d. Gesellschaft f. Natnrh.
Freunde zu Berlin, V. i. p. 69, 1811.
Bolten, M'u8. Bolt., ed. ii. p. 116, 1819 (name only).
Schumacher, Essai, p. 117, 1817; P. pleuronectea (full description).
xAinium b, Link, Beschr. Rostock Samml., part 3, p. 156, 1807 ; P. japoniatm.
Amusium Herrmannsen, Ind. Gen. Mai., L p. 47, I8i6 ; = Amusium Klein corr.
H. & A. Adams, Gen. Rec. Moll., II. p. 564, 1858. JeEEreys, Annals
and Mag. Nat Hist., Nov. 1876, p. 424 ; P. Z. S. 1879, p. 661.
Pleuronedia Swainson, Malacol., p. 388, 1840, P. pleuronectes (description).

Chenu, Man. de Conchyl., II. p. 187, 1862 ; P. japonica. Jeffreys, in
Wy ville-Thomson, Depths of the Sea, p. 464, 1873.
Amusium Woodward, Manual, ed. ii. p. 412, 1866. Stoiiczka, Pal. Indica, IIL Cret.
Pelecypoda, p. 426, 1871.

Shell smooth or very slightly sculptured externally ; valves gaping at the
sides, nearly equally convex, with radiating internal ribs ; ears subequal, small ;
notch obsolete or none ; hinge line straight ; margin entire ; shell free (byssif-
erous ?). Type Pecten pleuronectes L.

The name Amusium is of uncertain meaning or origin, but appears to
have been in use colloquially at least two hundred years ago to denominate
the " compass shell " or " flounder scallop." It was used by Rumpf in his
Treasury of Rarities from Amboyna, as pointed out by Dr. Jeffreys, and
probably here made its first entry into print. It was adopted by Klein, in his
curious and very unequal work on shells, ^br one of the groups in which he
placed the Pectens of Lamarck and later authors ; it was referred to by Martini,
and doubtless by other non-binomial writers, whom it would be profitless to
search out.

Its first entry into binomial scientific literature (if an auctioneer's sale cata-
logue without figures or descriptions may be so called) was in the obscure
pamphlet usually known as the Museum Boltenianum, of which a new edition
was publishtd in 1819. The first place where the name Amusium received a
description entitling it to recognition was in Schumacher's Essai, in 1817,
though Link had characterized the group as a section of his genus Pedinium
(= Pecten) ten years previously. Apparently in ignorance of Schumacher's
work, Swainson described it as a new genus in 1840, under the name Pleuro-
tiectia, which was adopted later by Chenu. Herrmannsen and others have
suggested that the name should be spelled Amussium, but the uniformity of
previous usage and the uncertainty in regard to its derivation seem to render
this inadvisable.

The characters which separate this group from the typical genus are chiefly
cnnchological. The byssus (if any exists, for so far I have not been able to
find any) passes between the gaping valves, and the notch, which usually exists
in the very young, is not found in the adult form, which would seem to have
discarded the byssus entirely, and supplied its place by using the terminal
sucker of the foot, which is large and expanded. The group frequents deep

MUSEUM OF COMPARATIVE ZOOLOGY. 209

and temperate waters for the most part, and the prismatic structure of the
shell is especially evident in the abyssal species, which in other characters
differ from the type, and form a transition toward Pseudamusium and the more
typical scallops.

A few species of Amusium are reported from the Cretaceous, but it appears
to be rather a modern member of the Pectinidce.

A living specimen of the type species, previously only known from the
eastern Asiatic seas, was dredged in the Gulf of Mexico by the U. S. Fish Com.
steamer " Albatross" in the winter of 188-4-85, at Station 2388, in 35 fms. sand,
Lat. 29° 24', Lon. 88° 1' W. Gr., and dead fragments at Station 2404, in
60 fms., Lat. 28° 44', Lon. 85° 16' W. Gr.,both on a line between the delta of
the Mississippi and Cedar Keys, Florida.

Amusium Dalli E. A. Smith.

Amussium Dalli Smith, Challenger Rep. Lamellibranchiata, p. 308, pi. xxii., figs.

7 a-c, 1886. (Off Bermudas, 4.35 fms.)
Amussium lucidum Jeffreys, var. striata, in part? (P. Z. S., 1879, p. 562.)

Plate IV. Figs. 1 a, I b.

Valves nearly equal, the right slightly more convex ; the adults gaping at
the sides ; the young closed or almost closed ; diversely sculptured ; right
valve nearly smooth except for growth lines, the internal lirse (7-9) marked
by obscure radiating ridges of the outer surface ; prismatic structure in a
radiating sense, distinctly marked, visible to the naked eye ; auri.;ies sculp-
tured only with growth lines, their upper edge denticulate in the very young,
arched internally, almost exactly equal, very small ; hinge line very short and
straight; left valve with somewhat irregular sharpish concentric waves, hardly
raised above the surface and more distant toward the periphery; prismatic
structure reticulate, the prisms almost separable at the extreme margin becom-
ing effaced toward the umbo with age ; auricles flat, subequal, without byssal
notch or fascicle, smooth or with faint growth-lines ; interior glassy, lirae 9-10,
usually 9, stouter longer and more opaquely white (in adults) in this valve
than in the other ; auricular crura very prominent, strong, forming the feet
of a stout arch of which the cartilage pit represents the keystone ; color trans-
lucent white near the margins, fuliginous in the central part which covers the
viscera. Alt. 62.0, lon. 59.0, max. diam. 6.0 mm., but reaching a larger size
as indicated by fragments. The shell is extremely thin and fragile, or rather
brittle.

Obtained at Station 41 in 860 fms. in the Gulf of Mexico ; Station 117, in
874 fms., Lat. 17° 47', Lon. 67° 3' W. Gr. in the Caribbean Sea; Station 147,
off St. Kitts, in 250 fms. (bottom temperature 52°.5 F.); Station 150, between
St. Kitts and Nevis, in 375 fms.; Station 151, in 356 fms., off Nevis; Station 153,
in 303 fms., off Montserrat (bottom temperature 48°. 75); Stations 161, 162, 163,
and 173, off Guadalupe, in 583, 734, 769, and 734 fms.; Stations 227 and 228,

VOL. XII. — NO. 6. 14

210 BULLETIN OF THE

oflF St. Vincent, in 573 and 785 fms. ; Station 236, off Bequia, in 1591 fms., soft
mud; Stations 245 and 268, off Grenada, in 1058 and 955 fms.; Station 275, off
Barbados, in 218 fms. sand, bottom temperature 52°. 5 F.

The bottom was, in all cases, sand, ooze, or mud, and the temperatures, except
those above cited, varied from 39° to 47°. 5 F., averaging about 41°.0 F.

Amusium meridionale Smith would appear from the figures and description
closely to resemble the young of this species. Mr. Smith kindly informs me
that the form differs, and the sculpture of the deeper valve is not identical ; in
A. Dalli the valve is much more glossy and the radiating lirse are hardly ap-
parent. Mr. Smith thinks A. meridionale does not attain a large size.

This elegant species was obtained by the "Challenger" as well as the
" Blake." It is evidently a true inhabitant of the deeps, although its range
is nearly 1400 fms. It is of extreme tenuity, and all the specimens obtained
were more or less broken about the margin. The adult valves are convex
nearly or quite to their edges, but the lower one while young has a concave
margination, as in the species of Propeamusium. Notes in regard to the
synonymy will be found under the head of Amusium Fourtalcsianum.

The soft parts of this species present some features of interest. The ocular
papilla; or ocelli are present, but devoid of pigment. The mantle is slightly
tinged with purple. The gills are long, single on each side, and furnished
with long separate filaments much as in Dimya. There are no branchial
palpi, but the lips are produced to a very unusual length, forming an arch
over the space below the mouth, both upper and lower lips being equally pro-
longed and applied to each other in a sort of horseshoe-shaped manner. They
are internally concentrically rugose in the specimen, which may be due to
contraction caused by the alcohol. The ovary projects from tlie body between
the gills in the form of a legume ; from its anterior end spring'' the stalk of
the foot, which is slender, the groove being well marked; the distal end of the
foot is greatly enlarged, looking like the end of an Anatifa without a shell ; it
is dark purple, the only part of the animal so strongly pigmented ; the enlarge-
ment or "cornet" is hollow, the aperture, with a stout margin, looking for-
ward and downward; internally it is domed and radiately striate, being in fact
an exaggerated and efficient sucker, by means of which the animal should be
able to hold on to any flat surface, or (by expanding and contracting it like
the foot of Yoldia) to move about on the semifluid mud of the bottom. The
anua does not project from the surface to an appreciable extent.

Section PROPEAMUSIUM De Gregorio (em.), 1883.

Shell small, thin, vitreous, smooth or sculptured, the lower valve u-sually
concentrically waved and with a byssal notch, but no pectinium or byssal ser-
rations ; when adult internally lirate ; the upper valve smooth or sculptured,
but usually, if sculptured, with the radiating sculpture prominent; valves
closed, the lower one convex over the internal lirations, then angulated and

MUSEUM OF COMPARATIVE ZOOLOGY. 211

applied to the internal surface of the upper valve, thus forming in the adult
and perfect shell a concave area about the distal margin of the inferior valve.
Type Amusium fenestratum Forbes.

The species of this section are found in deep waters, -widely distributed,
except in the arctic seas. It should be noted that in this group, as in many
other Pectens, there are often a pair of ridges or lirse, sometimes very promi-
nently elevated, on the inside, nearly parallel with the margin of the body
of the valve and situated at or on the prominence inside which is adjacent
to the auricular sulcus outside. These are not peculiar to either section of
Amusium, and are not counted by me in enumerating the internal lirse of
species of Propeamitsium. I notice that Smith in the Challenger Report has
counted them as lirse; so in the same species, when they are present, the num-
ber of lirse by my enumeration would always be two less than his. I have
called them the auricular crura for distinction's sake. They are found in
species of Pseudaviusium as well as of Propeaviusium proper, and are some-
times absent in species of either group.

Amusium (Propeamusium) Pourtalesianum Dall.

Amiisshim lucidum Dall, Bull. M. C. Z., IX. p. 117, 1881.

< Pleuronectia lucida Jeffr., Depths of the Sea, p. 464, fig. 78 b, 1873.

<.Amussium lucidum Jeffr., Ann. Mag. Xat. Hist., XVIII. p. 425, Xov. 1876 ; P. Z. S.,

1879, p. 562. E. A. Smith, Challenger Rep. Lamellibranchs, p. 317, pi. xxiv.

figs. 2 a-c.

Plate IV. Fig. 3. Plate V. Figr. 13.

Taken at Station 2, off Morro Light, in 805 fms.; Station 5, in 229 fms. ; Sta-
tion 19, in 310 fms. ; Station 21, in 287 fms. ; Station 35, in 804 fms. ; Station
44, in 539 fms.; all in the Gulf of Me.xico near Cuba. Also in the Gulf west of
Florida in 30 fms., and at Charlotte Harbor, "West Florida, living in 13 fms.;
Station 47, in 337 fms.; Stations 50, 60, and 100, off Havana, in 119, 480, and
400 fms.; Stations 162, 163, and 167, near Guadalupe, in 734, 878, and 175
fms.; Stations 176 and 177, near Dominica, in 391 and 118 fms. ; at Barbados,
in 100 and 154 fms.; Station 227, near St. Vincent, in 573 fms.; station 262,
in 92 fms., near Grenada. The depth at winch it has been found living varies
from 805 to 13 fms., aud the bottom temperature from 39°. 5 F. to 82°. 5 F.
This is a remarkable range.

Tliis species was included under the name of lucidum by Dr. Jeffreys in
1876, and had been well figured in 1873. The figures " a " and " b " of luci-
dum represent what are now considered as two different forms. At that early
date in deep-sea work, minute discrinunation as to species was less practicable,
and therefore less usual, than at present, when the amount of material is so
much greater.

Figure " a " of the " Depths of the Sea " is magnified ne{jrly four times
linear, and comes from the Eastern Atlantic. The other figure was taken from

212 BULLETIN OF THE

a specimen dredged by Pourtales in the Straits of Florida, now in the National
Museum, and was only magnified some two and a half times, linear. From
among the forms at first confounded under the name of lucidum, and several
of which have been carefully worked out by Mr. Edgar A. Smith of the British
Museum in the Challenger Keport, the following may be discriminated : —

1. P. lucidum (Jeflfr. em.) Smith. Fig. "a" of 1873, E. Atlantic.

2. P. Pourtalcsianum Dall. Fig. " b " of 1873, West Indian region.

3. P. meridionale Smith 1. c. {A. Iv^idumJeSi., var. striata Jeffr., according
to localities and specimens cited P. Z. S., 1879, p. 562).

2 a. P. Pourtalesianum, var. striatulum Dall (agrees with Jeffreys' descrip-
tion of his var. striata, but not with his specimens).

2 b. P. Pourtalesianum, var. mxirTnoratum Dall. West Indian region.

Mr. Smith has discriminated the typical P. lucidum (Chall. Rep. Lamellibr.,
p. 317, pi. xxiv. figs. 2 a-c), from which the form here called Pourtalesianum
differs in being more oblique, longer, and of larger size when adult. The
series of A. Dalli Smith, obtained by the " Blake," indicate that A. meridionale
Smith is possibly the young of Dalli, as the diflerences of gape and of sculpture
correspond fairly well to stages of growth observed in the Blake series. Mr.
Smith, however, is confident that it is distinct. It is at all events perfectly
distinct from lucidum. My impression is that Dr. Jeffreys derived his idea
of his variety striata from specimens of J. Pourtalesianum, var. striatulum Dall,
sent him by Pourtales ; but that when he came to describe it he cited Chal-
lenger localities and specimens, which on a cursory examination he took to be
the same thing, and omitted to mention his West Indian ones. However this
may be, the var. striatulum above mentioned agrees perfectly with Dr. Jeffreys'
rather brief description.

No specimens of the typical A. lucidum were obtained by the " Blake," or
have been received from the Fish Commission dredgings on the American or
Gulf coasts.

The A. Pourtalesianum,yaT. striatulum,v!-as obtained by the " Blake" at Station
128, near Frederikstadt, in 180 fms.; Station 192, off Dominica, in 138 fms. ;
and Station 226, off St. Vincents, in 424 fms. The bottom temperature at
these stations varied from 42°.5 to 64°. 0 F.

The variety marmoratum (Dall, Bull. M. C. Z., IX. p. 117, 1881), with its
brilliant mottling of orange, scarlet, brown, bright yellow, and opaque white
dots or flecks, appeared indifferently at most of the stations in company with
the pale translucent typical form.

I haA'e examined one hundred and forty valves of this species, 64 per cent of
which had ten internal ribs ; 5.7 per cent nine ribs ; and 21.5 per cent eleven
ribs ; not counting the crural callosities. Six specimens had twelve ribs, and
one each had eight, thirteen, fourteen, fifteen, and twenty-one ribs. The extra
ribs usually appear as intercalary knobs or lirse near the outer edge of the
ribbed area, and seldom extend into the body of the shell like the normal
ones.

MUSEUM OF COMPARATIVE ZOOLOGY. 213

Amusium (Propeamusimn) cancellatum Smith.

Amussium cancellatum Smith, Challenger Rep. Lamellibranchs, p. 315, pi. xxiii.

figs. 8 a- 8c. 1886.
? Amussium fenestratum, var. cancellatum, Jeffreys, P. Z. S., 1879, p. 561. (Porcupine

Exp. 1869, Station 37, 2435 fms.)

Plate T. Figs. 1, 1 a, 2.

Obtained by the Coast Survey steamer " Bache " in Charlotte Harbor, W.
Florida, in 13 fnis., bottom temperature 82°.0 F.; by the "Blake," off Cape San
Antonio, in 424 fms.; in Yucatan Strait, in 640 fms.; at Station 19, in 310 fms.;
Stations 51 and 61, near Havana, in 450 and 243 fms.; Station 128, in 180 fms. ;
Stations 130 and 136, in 451 and 508 fms., near Santa Cruz ; Stations 150 and
151, near Nevis, in 373 and 356 fms., bottom temperature 45°.0; Station 176,
in 391 fms., and Station 188, in 372 fms., near Dominica ; Stations 221 and
222, near Santa Lucia, in 423 and 422 fms.; Stations 226 and 230, near St.
Vincent, in 424 and 464 fms. ; Station 236, off Beqnia, in 1591 fms., bottom
temperature 39°. 0 F. With the exception of the first and last localities, the
bottom temperature at the above stations ranged from 41°. 5 to 60°. 5 F.

This fine species was found by the " Challenger " off Bermuda in 1075 fms.,
also off Culebra and St. Thomas. Many of the Blake specimens are more
finely developed than those figured by Smith in the Challenger Report. It
seems to have a wide geographical and bathymetrical range, and to be almost
independent of temperature limits. The valves are tightly closed, the soft
parts pale yellowish, with a narrow brown marginal line on the mantle edge ;
tliere are no pigmented ocelli, the lips are wrinkled, the foot very short, deeply
incised, without differentiated stem, and elongated instead of hood-shaped.
The distal end of the intestine is free for a short distance. No byssus was
observed. The young resemble the genuine fenestratum Fbs., but are more
regular. Of those examined, 75 per cent had eleven lirse, and the remainder
ten to fourteen lirae internally.

Amusium (Propeamusium) Hoskynsi Forbes.

Pecten Hoskynsi Forbes, Rep. ^gean Sea, p. 192, 1843 ; Jeffreys (ex parte), Light-
ning and Porcupine Moll., P. Z. S., 1879, p. 502.

Pecten concentricus Forbes, 1. c. (ex parte ; lower valve ?).

Pecten Jimhriatus et antiquatus Philippi, Moll. Sic, II. p. 01, pi. xvi. figs. 5, 6
(upper and lower valves), 1844.

Not P. Iloskynsi of G. 0. Sars, Leche, et al., of northern seas.

This form has been obtained in the Mediterranean and the adjacent parts of
the Atlantic. No authentic specimens are in the Jeffreys collection from the
vicinity of America, the specimens so considered, mostly very young or imper-
fect, appear to belu-ig to A. cnncdlatuvi Smith (uou JefFreys). The ribs are
less clearly develope 1 than in most of the species, until the individual is fully

214 BULLETIN OF THE

adult; in very young specimens they are absent, and the shell can hardly be
distinguished from P. imhrifer, which has commonly been confounded with it,
as mentioned under that species. The present one is characterized by the very
large shelly bubbles, generally worn away, and leaving their semicircular basal
edges plainly visible in fifteen to eighteen series, with intercalary ones near
the margin. The umbo of the upper valve has sharp elevated concentric
ridges without pustulations. There are inside near the margin of adult shells
17-21 short liroe, thickest at their distal ends usually and not extending into
the hollow of the shell, except the marginal ones, and these not always. The
shell is very thin, vitreous, and pale yellowish in color.

Amusium (Propeamusium) Holmesii Dall.

Plato V. Figs. 5, 11.

Shell small, somewhat obli([ue, thin, brightly colored (like A. var. ma.rmo-
ratum), the convex valve with twenty to twenty-eight faint radiating ridges,
absent toward the beaks and with intercalary ridges toward the margin ; inter-
spaces between the ribs polished, mostly smooth or with traces of concentric
lamellae which near and over the radiating ridges rise to arched scales, which
in some cases are closed in front, forming spindle-shaped pustules, with the
longer axis in the concentric line to which they belong ; anterior ear the larger,
pectinately scaled on the margin, with obscure radiating or lamellar sculpture ;
concave valve finely concentrically waved, the waves becoming crowded and
scaly toward the margin ; anterior ear with three or four radiations and strong
lamellae, posterior ear lamellate, but not radiated ; anterior margin rounded,
posterior ditto, produced. Alt. 12.0, Ion. of shell 12.0, of hinge line 6.0,
max. diam. 2.0 mm. Internal liraj eleven to fourteen, fine, with a tendency
to pair, and falling considerably short of the margin in the completely adult
shell.

Dredged at Station 273 in 10,3 fms., and in 100 fms. at Barbados, living.
The bottom was yellow coral and broken shell, and the temperature 59°.5 F.

This pretty shell is nearest A. Hoskynsi Forbes, from which it differs by the
polished umbonal region, the larger number of ribs, the smaller size and differ-
ent shape of the pustular scales, by its larger size, differently proportioned
and pectinated ears, and by the strength, number, and character of its internal
lirse. Its form is nearest that of A. Pourtalesianum, var. marmoratum, which
is also brightly colored. It is named in honor of Dr. Holmes, the author of
the "Pliocene Fossils of South Carolina."

Amusium (Propeamiusium) Sayanum Dall.

Plate V. Figs. 3, 9.

Shell compressed, nearly equivalve, somewhat inequilateral ; white ; ^nth
dissimilarly sculptured valves ; right valve with (near the middle) about

MUSEUM OF COMPAKATIVE ZOOLOGY. 215

twelve radiating ridges between which toward the base intercalary ridges rap-
idly appear, so that at the base, in one specimen, there are thirty-two ridges
and beginnings of ridges ; over the whole are set closely declining concentri-
cally continuous laniellas, uniform, when unbroken, over ridges and intervals ;
the ears are nearly equal, similarly sculptured to the rest, and with three to
five radiating ridges ; the cardinal line straight and the umbo not prominent ;
the left valve is a little smaller, smooth or not ridged, and covered with simi-
lar but less elevated and closer lamellae, which swell up in obsolete radiating
lines in sympathy with the ridges of the other valve, but which swellings do
not cover any genuine ridges ; ears subequal, similarly sculptured ; byssal notch
not deep ; interior of valves with 10 to 16 radiating lirse (beside the crura) of
which about 10 run home to the body of the valve ; ligamental pit moderate,
hinge line simple and without transversely rugose areas ; max. alt. 15.5, Ion.
15.0, hinge line 7.0, diam. 3.0 mm.

Soft parts yellowish white except the liver; "ocular" tubercles without
pigment, but present.

Dredged off Morro Light, Havana, at Stations 16 and 100, in 250-400 fms.,
living, bottom temperature 55''.6 F. A single valve was obtained at Station
143, off Saba Bank, in 150 fms., bottom temperature 63°. 5 F.

This pretty species is nearest to A. squamigerum E. A. Smith, having the
same type of concentric sculpture on the right valve, but differs in the strong
rectilinear character of the radiating ribs, in being proportionally wider and
more inequilateral, having a wider and less oblique hinge line, and being
apparently somewhat flatter. These differences, however, though strong as
regards the figured specimens, might be less evident for a large series. Still,
they are so marked for what we have, that I have thought it perhaps better to
give the present form a separate name than to assume the existence of inter-
mediate varieties, without any intermediate material.

Amusiiun (Propeamusiuin) Alaskensis Dall.

Pecten (Pseudamussium ?) alaskensis Dall, Am. Journ. Conch., 1871, p. 155,

PI. xvi. fig. 4 a, b.

Plate V. Figs. 7, 7 a.

In order to bring together the available material relating to this group, this,
the largest species yet known, and having the most numerous ribs, has been
included in the figures. So far it has occurred only on the Alaskan coast, and
no other species is known from there ; but there are several in the waters of
Japan and Korea,

216 BULLETIN OF THE

Subgenus PECTEN s. s.
Pecten magellanicus Gmelin.

P. magellanicus Gmelin {Ostrrn), S. N., p. 3317, 1788.

P. Clintonius Say ; P. prlncipoides Emmons, and P. tenuicostatus Mighels, auctorum.

This species was not actually obtained by the " Blake," but it was found
with many of the Blake species off the Carolina coast by the U. S. Fish Com-
mission. It was thought well to note here that this species by the character
of the foot and of the young shell forms a very complete link between Amu-
sium and such Pectens as P. caurinus, etc. It has precisely the foot of typical
Amusinm.

Pecten caurinus Gould.

Peclen caurinus Gould, Moll. U. S. Expl. Exp., Proc. Best. See. Nat. Hist., III.

p. 345, Dec. 1850.

Plato V. Fig. 4.

The study of the group of Pectens included in Amusium and Pseudamusium
was made to utilize all available material, and in several cases resulted in the
identification of small specimens, supposed at first to be distinct, with the young
of forms which as adult would be classified in other sections of the genus
Pecten. Among these immature shells was the young of P. caurinus, which has
a striking resemblance to some species of Pseudamusium. It is figured here-
with as a matter of interest to those engaged in a study of the development
of the group. The specimen was dredged at Sitka, Alaska, in 1865. This spe-
cies grows to a large size and is found on both sides of the North Pacific. It
has no internal lirag, ami is not externally like Amusium, but in the charac-
ter of the foot it stands midway between the sucker-footed and the typical
Pectens. The foot is cylindrical, with the usual groove behind ; the tip is very
slightly enlarged, glandiform, with a small incision behind, which does not
look as if it could perform the office of a sucker with much efficiency.
The mantle margin is broad and thick ; set, in both valves, with a profusion
of well-developed ocelli and tentacular processes. The lips are arborescent at
their outer margin and radiately finely wrinkled inwardly.

Pecten nucleus Born.
Pecten nucleus Born, Test. Mas., pi. vii. f. 2, 1780.

Valves of this very neat little Pecten were found in 80-127 fms., off" Havana,
by Sigsbee.

The validity of this species cannot be considered settled. A dwarf form of
P. dislocatus from southern waters seems to lead up to it by imperceptible
gradations. The name of Born is of course prior to that of Say, but it is said
that there are even earlier names. A larger supply of P. nucleus is needed
before the identification can be considered conclusive, and so for the present
I leave the names imdisturbed.

MUSEUM OF COMPARATIVE ZOOLOGY. 217

Pecten dislocatus Sat.

Pecten dislocatus Say, Journ. Acad. Nat. Sci. Phil., ii. p. 260, 1824.

Collected eight miles S. S. W. of Sand Key light, in 125 fms., by the steamer
" Bache," in 1872, at Station 36, in 84 fms., and by Sigsbee, off Havana, in
182 fms. The specimens obtained were dead, and perhaps disgorged by some
fish, as this species is known to prefer water of moderate depth. No living
specimens were obtained. The relations of this species to P. nucleus Born are
rather close, and it is a question I have not time at present to investigate
whether Say's name is the first which has been given to this shell. Though
allied to F. i^urpuratus Lam., it seems distinct from it.

Pecten phrygium, n. s.

Shell of the general form of P. asper Sby., flattish, oblique, both valves
similarly sculptured ; left valve with about eighteen ribs with nearly equally
wide interspaces ; these ribs have three sharp thin keels upon them, a median
and two lateral ones, which project on the distal margin of the valve, and, with
the similarly projecting points of four intercalary smaller keels, fimbriate the
margin in a remarkable manner ; when perfect there are rounded scallops
thrown from keel to keel close together, like the edges of a roll of coins, which
hide all the keels except the extreme edge of the median keel of the ribs,
which, without projecting much, forms a raphe, connecting the scallojied sur-
face ; the material of the scallops is very fragile, and when worn away leaves
a totally different surface, which in this case shows a multitude of sharp thin
scales (the bases of the scallops) zigzagging from keel to keel and imbricating
the keels ; the ears are moderate and subequal, there are about five distant
narrow ridges on the anterior and two or three on the posterior ear of the left
valve, with finer intercalary thread.s ; the right valve has the ridges on the
posterior ear scaly, and the byssal fasciole transversely concavely ridged ; there
are four or five free spines to the pectinium ; the cardinal margin is straight
and internally strongly vertically striate, as in very young specimens of
Janira or Amusium. Height of largest valve, 36.5 ; width, 36.5 ; cardinal
margin, 19.0 mm. Umbonal angle about 100°.

Dredged living at Station 32, in 95 fms., north of Yucatan Banks, in N. Lat.
23° 32' and W. Lon. 88° 5'; bottom temperature not noted, but probably about
60° F.

Dead valves were found at Station 45, in 101 fms., bottom temperature
61°.75 F. ; at a depth of 127 fms., off Havima, by Sigsbee; and off Grenada, at
Station 244, in 792 fms. It is probably a dweller in about 100 fms. when
living.

The very remarkable sculpture of this species sufficiently distinguishes it
from any other, and specimens not fully adult may be recognized at once by
the fimbriated basal margin of the valves. It is related to P. Philippii Recluz

218 BULLETIN OF THE

and P. noronhensis Smith, differing from both in having more numerous ribs,
flatter and similar valves (the two valves are differently sculptured in the
others), and in details of form and sculpture of surface and ears.

In one specimen, apparently quite adult, though not as large as one of the
dead valves, there are on the interior of the valves, especially the left one,
narrow lira) corresponding to grooves bounding the ribs externally, but which
are hidden on the .outside by the imbricated sculpture. These lira are very
distinct, and are raised at the ends into a little white prominence like the lira)
of ^. alaskensis; one more instance of the interchangeability of characters in
this group.

The soft parts are streaked with purple in dots and dashes ; there is an
ocellus for each rib on the margin, except near the anal opening, where the
mantle margin is without them, and is folded in such a way as irresistibly to
suggest that it is the first step toward siphonation; there are here two very
peculiar large crimson color marks on the mantle edge not duplicated elsewhere.
The ocelli are of different sizes, some mucli larger than others. The remainder
of the superficial anatomy calls for no sjjecial remark.

Pecten exasperatus Sowerbt.
Pecten exasperatus Shy., Thes. ConchyL, Pecten, p. 64, pi. xviii. figs. 183-186, 1846.

Valves of j'oung specimens were found in 13-19 fms., Charlotte Harbor,
W. Florida, and in 640 fms., Yucatan Strait ; the latter fresh, but not original
to that depth, in all probability.

Tliis species is very closely related to, if not identical with, P. fusco-purpureus
of Conrad, M'hich name would, if the species be identical, fall into synonymy.

The adult hinge line in perfect examples usually shows the transverse corru-
gations (referred to imder species of Pseudaviusium) with great distinctness.

Pecten ornatus Lamarck.

Pecten ornatus Lam., An. s. Vert., VI. p. 176, 1819.

Young specimens of this common West Indian species were obtained at
Station 11 in 37 fms., off Havana in 80-182 fms., on the western coast of
Florida in 50 fms., and a single valve, doubtless drifted but fresh, off Santa
Lucia, at Station 220, in 116 fms.

Pecten antillarum Recluz.
P. antillarum Ee'cluz, Journ. de Conchy!., IV. p. 53, pi. v, fig. 1, 1853.

Dead valves were found by Sigsbee off Havana, in 127 fms.

This species lives in a few fathoms of water about the Florida Keys, where it
has been abundantly collected by Hemphill and others. The soft parts are
about the same color as the shell, the foot vermiform, simple, and quite small.

MUSEUM OF COMPARATIVE ZOOLOGY. 219

Pecten eflQuens, n. s.

Shell small, high, flattened, covered all over with very fine striae radiating
from the umbo, but diverging from the middle line of the valve without refer-
ence to the other sculpture; left valve with about ten Little elevated poorly
defined ribs, which are most distinct in the young, rounded, smooth, separated
by wider shallow interspaces, in which are from three to seven minute sub-
equal regular threads, with small hardly elevated, regularly spaced, transverse
scales on them ; anterior ear very small, obliquely cut off; posterior ear high,
short, with about ten obscure radiating threads ; cardinal margin straight,
simple ; right valve ^ith faint radiating ridges most prominent near the mar-
gin, and tending to pair ; ears similar to those of the opposite valve, byssal
notch small, fasciole very narrow, with four pectinium-spines beyond the edge
of the ear, and a series of them overhanging the fasciole within it ; cartilage
pit rather small, inner cardinal border nearly smooth ; color pale orange.
lemon-yellow toward the umbo ; height of largest valve, 26.0. width 22.0 mm.;
umbonal angle about 85°.

Valves were dredged in 127 fms. off Havana, by Sigsbee.

This shell seems nearest P./urtivus Loven, but has smaller anterior ears,
finer strite, and altogether different coarse sculpture, especially on the left
valve. Both forms have the Camptonectes striation.

Section PSEUDAMUSIUM H. & A. Adams.

" Pseudamusium Klein, Tent. Meth. Ostracol., p. 134, 1753." Stoliczka, Pal. Indica,

Cret. Pelec, III. p. 426, 1871.
Pseudamussium H. & A. Adams, Gen Rec. Moll., II. p. 553, 1858 (no type mentioned).

Chenu, Man. de Conchyl., II. p. 184, 1862; P. dispar. Jeffreys,

Ann. & Mag. Nat. Hist., Nov. 1876, p. 424.
Syncydonema Meek, Cret. & Jur. Foss. Smithsonian Check List, 1864, p. 31 ; P. rigida

H. & M. Stoliczka, 1. c, p. 426, 1871.

Shell thin, smooth or delicately sculptured, small; valves subequal, closed
except at the byssal foramen; ears unequal, the posterior ones often hardly
differentiated from the body of the shell; hinge line straight ; notch distinct;
margin entire; interior destitute of radiating ribs; shell free, byssiferous.
Types Pecten dispar and P. pseudamussium Lam.

The name Pseudamusium is due to Klein, but was first introduced into
regular nomenclature and defined by H. & A. Adams, who, however, did not
name a type, an omission which was supjdied by Chenu. There do not ap-
pear to be any sound characters by which Sxjncydonema can be differentiated
from the other species here included, and it is very probable that Camptonectes
(Agassiz Ms.) Meek, 1864, and its synonym Ehurneopecten Conrad, 1867, should
also be combined with it. These species, however, have a somewhat peculiar

220 BULLETIN OF THE

sculpture, and there seems to be no serious objection, considering the laT^re
number of species, to retaining the name Camptonectes in a sectional sense, as
has been done by Stoliczka, provided it be understood that the division is not
known to represent any fundamental diagnostic characters.

The peculiar sculpture, upon which alone Camptonectes is founded, is not, as
was supposed by Stoliczka, singular to Mesozoic species, but may be found on
living forms, like P. furtivus Lov^n, and many others. It may exist in un-
ribbed species or in those with ribs; in the latter case being supplemental to
the other ornamentation.

Pecten (PseudamuFium) imbrifer Loviiw.

Pecten imbrifer Loven, Ind. Moll. Scand., p. 31, 1846.

Pecten ninmmiUatus M, Sars (ined.) fide G. 0. Sars.

Pecten Hoskynsi, var. pustulosus Verrill, Trans. Conn. Acad., V. p. 681, pi. xlii. figs.

22, 22 a, PI. xliv. fig. 11, July, 1882.
Pecten pnstulosux Verrill, 1. c, VI. p. 261, 1884.
Pecten Hoskynsi Jeffreys (ex parte) ; G. O. Sars, Moll. Reg. Arct. Norv., p. 20, pi. 2,

figs. 1 a-c, 1878; Leche et al., not of Forbes, 1843

Plate IV. Figs. 4 a, 4 b.

Valves slightly convex; left one least so, slightly concave at its distal margin;
valves diversely sculptured, right valve without perceptible prismatic sculpture,
surface smoothish, with radiating series of (larger or smaller) hemispherical
punctate bubbles arranged on the slightly raised concentric growth-margins;
radiating sculpture of similar nature on the auricles ; left valve with concentric
sharp equidistant raised lamina>, wider near the margins and showing more
or less prismatic texture ; auricles well defined, the anterior very small, the
posterior much larger with strong concentric and faint radiating sculpture ;
byssal sulcus very small and fasciole very narrow, passing straight along the
body margin ; color vitreous white often with a grayish discoloration. Alt. 12.5,
Ion. 12.0, max. diam. 3.3 mm.

Arctic seas and cold waters north of Europe, the Atlantic, and along the
northeastern coast of the United States.

When the valves are worn, as is often the case, the sculpture on the right
valve is represented by more than hemispherical loop-like lines connected by
sections of the concentric lines very much as in P. Hoskynsi, which is, however,
a smaller species. The pustules vary much in size in different specimens, and
have a dotted or cellular surface.

This fine species was first described by Loven, and is destitute, at any stage,
of the internal lirae of Propeamusium, and has a more vitreous and translucent
texture than that of P. Hoskynsi, with which it has been widely confounded.

The range of variation of the external sculpture is very much as in P. Hos-
kynsi, and it is difficult to separate young and depauperated specimens of the
latter from young irahrifer; especially when the external sculpture is worn.

MUSEUM OF COMPAKATIVE ZOOLOGY. 221

After a careful study of the specimens in the Jeffreys collection, I am not
surprised that he should have united them, the majority of his examples of
P. Hoskynsi being very young and imperfect, while he had only two or three
specimens of P. imhrifer. The latter is a cold-water species, reaching its
finest development in arctic or subarctic seas; it is doubtful if it reaches as far
south as the coast of France on that side of the Atlantic, unless in very cold
and deep water. On the other hand, no species of Propeamusium has been
found in the arctic seas. I have not seen P. leptalea Verrill, but the diagnosis
reads much like a description of one of the more finely sculptured forms of
imbrifer.

Pecten (Pseudamusium) reticTilus Dall.

Plate V. Fies. 8, 10.

Left valve less convex and smaller, valves diversely sculptured; right valve
with solid uniformly elevated concentric laminae crossing thread-like rather
distant radiating riblets ; where the lamina crosses a thread, especially near the
margin, it rises into a minute grooved spine ; auricles similarly sculptured ;
surface showing the prismatic texture in a very delicate manner ; left valve
also prismatic, with some strong radiating sculpture on the auricles, but the
body of the valve marked with fine concentric, uniform, wavelike undula-
tions ; auricles well marked, the anterior the smaller ; byssal notch rather
deep, fascicle narrow, close to the border of the valve. Alt. 7.0 ; Ion.
7.25 mm.

Obtained in 82-123 fms. at Barbados. At Station 297, where the specimens
were living, the bottom was stony, and the bottom temperature 56° 5 F.

This species is among the Pseudamusiums what A. cancellatum is among the
Propeamusiums. It is differentiated from the following species by the char-
acters mentioned under the latter, and appears to be always pure white. There
were six strongly pigmented, proportionally very large, ocelli on the mantle-
edge of the left valve. In the very young the reticulation in a concentric sense
is sometimes looped, which at first gives it a different aspect. By accidents
of growth the radiating sculpture and its spines are sometimes not rectilinear
from the umbo, which also gives it for a moment an unfamiliar aspect.

Pecten (Pseudamusium) thalassinus Dall.

Amussium fenestratvm Verrill, Trans. Conn. Acad., V. p. 582, 1882.
AmussiuTA sp. Verrill, Ibid., VI. p. 261, 1884.

Left valve less convex and slightly smaller ; right valve sculptured much
as in reticulus, but less pronounced and without spines, sometimes nearly
smooth except near the margin, where traces of the radiating sculpture are
always visible ; auricles as in reticulus, but less strongly sculptured ; left valve
with concentric sculpture coarser than in reticulus, notch similar ; prismatic

222 BULLETIN OF THE

structure barely visible on extreme of tbe left valve only; interior of adult
marked by a flat thickened margination, especially in the left valve, extending
parallel with the basal margin ; auricular crura elevated into u low ridge or
blunt tooth on each side ; hinge margin furnished with a lanceolate area on
each side of the cartilage pit which is finely deeply closely grooved in a direc-
tion vertical to the hinge-line, the projections between the grooves interlock-
ing so strongly as to cause the hinge-line of one valve sometimes to break off
bodily, rather than separate from the opposite valve, when one is trying to
open a pair, even when the soft parts are absent. The shell is prettily varie-
gated with mottlings of red, brown, and yellow on a creamy ground, both
valves participating, but the upper one rather the brighter in most cases.
Alt. 8.5 ; Ion. 9.0 ; max. diam. 2.0 mm.

This shell has been dredged by Professor Verrill in 80 to 317 fnis. off
Martha's Vineyard. Specimens were obtained by Sigsbee off Havana in 450
fnis., and at Station 36 in 84 fms.

The figure on Plate VI. fig. 5. representing the interior of the young fry
of Janira hemicyclica, equally Avell represents the very young of this species,
which at this stage can hardly be distinguished from the other except by the
external sculpture, which is often absent, or by the rug* of the hinge.

Because the internal lira; occasionally fail, or are late in being deposited in
certain individuals of Propeamufshim, Dr. Jeffreys formed the opinion that they
were of no importance as a specific character. In this manner he was led into
several errors, from judging merely by the exterior of some of these small shells.
In the present case the external sculpture is much like that of Amusium fenes-
tratum Forbes, but finer and more regular. Both have similar varieties of
color. Dr. Jeffreys was thus led to unite them, although they belong to differ-
ent sections of the genus, as also in the case of A. Hoskynsii and Pseudamnsium
pustulosnmYerviW, an error entailing some serious results in the matter of areal
distribution. Professor Verrill followed Dr. Jeffreys until a special study of
the gi-oup enabled me to call his attention to the discrepancies existing between
the two species.

The peculiar arrangement of grooved and comigated areas on the hinge-line
is of interest in connection with the development of hinge characters. It is
very common in the fry of various Pectens. It is well developed in the very
young P. magellanicus. I have found traces of it in many of the minute
specimens oi Pecten I have examined, though often very faint, and in very few
does it reach the development attained in the present species. It was the adult
and peraianent characteristic of the extinct genus Neithea, and traces of it are
even recognizable in some .species of Janira or Pecten in their adult condition.
When I first discovered it in the present species, I naturally supposed I had
come upon the type of a new generic division, but I have since found it in the
young of Pectens of nearly all sections of the genus in its widest sense. On
the other hand, in some species even the very young have hardly a trace
of it.

MUSEUM OF COMPARATIVE ZOOLOGY. 223

Pecten (Pseudaxausitun) Sigsbeei Dall.

Plate IV. Fig. 2.

Valves rather convex, the left cue most so ; both apparently polished, but
with microscopic silky concentric stria ; no radiating sculpture, no prismatic
markings ; anterior auricles well marked, very small, oblique; posterior auri-
cles larger, with a broad shallow byssal sulcus but no fasciole or pectinium,
the markings only of concentric growth lines ; color brownish with opaque
white splashes. Alt. 11.5; Ion. 9.1 ; diam. max. 3.75 mm.

Two \'alves were obtained by Sigsbee in 158 fms., Lat. 22° 10', W. Lon.
82° 20', near Havana, Cuba. This little sjjecies is very recognizable by its
plump oval shape, like an apricot stone, and its smooth surface destitute of
radiating sculpture.

Genus HINNITES Defrancb.
Hinnites Adamsi, n. s.

Plate V. Fig. 6.

Shell thin, ashy white externally, internally semi-nacreous ; rounded with a
comparatively short straight hinge-line ; attached valve unknown ; upper valve
indistinctly auriculate, rather flat, irregular toward the margin with a small
pointed but not prominent apex, a little to the right of the middle of the hinge-
line ; sculpture composed of somewhat irregular radiating costse, not bifur-
cating but increasing by intercalation toward the margin, where they are much
crowded; these costae are formed by crowded overlapping rounded scales, like
biscuit piled one over another, and showing sharp edges only where worn ;
there are about forty with a somewhat smaller number of intercalary ones ;
the concentric sculpture is composed of ill-defined lines of growth, and the
whole surface is microscopically gT^nulose; interior polished, silvery, repro-
ducing the external rugulosities ; muscular impressions in\isible ; cartilage pit
triangulai', distinct, hinge-line smooth, margin nearly simple. Lon. of shell,
28.0; of hinge-line, 13.0 ; height of shell, 30.0 mm.

Station. 227, off St. Vincent, in 573 fms., fine sand and gray ooze ; the bottom
temperature 40°.5 Fahrenheit.

This shell has an unmistakably abyssal facics and seems to belong to the
genus Hinnites. It is named in honor of Prof. Charles B. Adams, of Amherst,
to whom so much of our knowledge of the fauna of the West Indies and
Panama is due, and who was among the first of American naturalists to recog-
nize the variability of what we call species, and the close relations which exist
in nature between forms admitted by naturalists to be of " specific " value, or,
in other words, which have obtained a temporary equilibrium of characters
which they transmit to their descendants.

224 BULLETIN OF THE

Family LIMIDiE.

Genus LIMA BRUGifeEE.

Lima squamosa Lamarck.

Lima squamosa Lam., An. s. Vert., VI. p. 156. Sowerby, Thes. Conchyl., I. p. 84,

pi. xxi. figs. 1, 18.

Young specimens, in the condition called L. carihhcea by D'Orbigny, were
dred;.,'ed in 80 I'nis. off Havana; in 100 fms. at Barbados; at Station 210, near
Martinique, in 191 I'nis. ; and in 640 fms., Yucatan Strait ; all single valves ;
one living specimen at Station 292, in 56 fms., sand, off Barbados, bottom
temperature 74°. 5 F. It has been found in Florida by Hemphill, and is
widely dispersed over the world.

Lima tenera Sowerbt.
Lima tenera Sby., Thes. Conchyl., I. p. 84, pi. xxi. figs. 2, 3, 10, 11, 12, 1846.

Obtained by Sigsbee in 80 and 127 fms. off Havana; at Station 21, in 287
fms. ; and at Barbados in 100 fms.; all the specimens being separate valves.

A variety which may talce the name of planulata was obtained at the last-
mentioned station. It is distinguished from the ordinary varieties by being
more elongated and compressed, Ijy finer and smoother sculpture, which wants
the little spurs or spines on the radiating ribs, which are flattened and sepa-
rated by narrow dee])ly punctate channels. In full-grown specimens the
inner basal margin i.s smooth and thickened all round, wliile the margin of
the gape is pouting and more marked than in the type. This variety is flatter
than any form of Lima which has been figuretl, and at first was naturally sup-
posed to be a distinct species. There seem, however, to be intermediate speci-
mens, and I think it is quite certainly an extreme form of L. tenera.

Lima inflata Lamarck.

Lima inflata Lam., An. s. Vert., VI. p. 156.

Lima fasciata Sowerby, Thes. Conchyl., I. p. 85, pi. xxi. figs. 16, 17 ; not Ostrea
fasciata Linnc.

Valves of this species were dredged on the west coast of Florida in 19 fms.,
and off Sand Key in 128 fms. Fine specimens have been obtained by Hemp-
hill at Cedar Keys and Key West on the reefs, and it extends along the
Atlantic coast northward nearly or quite to Cape Hatteras.

MUSEUM OF COMPARATIVE ZOOLOGY. 225

Lima hians Gmelin.

Ostrea hians Giuel. S. N. 3333.

Lima frayilis Montague, Test. Brit. Suppl., p. 62.

Lima tenera Turton, Zool. Journ., V. p. 362, pi. xiii. fig. 2.

Lima aperta Sowerby, Thes. Conch., I. p. 87, pi. xxii. figs. 26-29.

One valve was obtained near Santa Cruz, at Station 127, in 38 fms.

Lima albicoma, n. s.

Shell small, short, waxen gray, very inequilateral ; sharply truncated above
and roundly produced below, anteriorly ; hinge-line short, cardinal area trian-
gular with a narrow long cartilage in a shallow sulcus overhung at the outer
end by the acutely pointed umbo ; anterior edges of the cardinal and truncated
areas strongly carinate ; between the carinae the truncature is concave with one
strong and two or three fine radiating threads parallel with and near to the
longer carina, the rest of this area striate with lines of growth ; the anterior
angle of the hinge margin hidden in the concavity, when the shell is viewed
from the side; the posterior angle is visible, but not prominent, though sharp ;
from this angle to the outer end of the anterior carina the base describes two
thirds of a circle ; exterior smooth, finely regularly grooved with very numer-
ous punctate grooves, at the basal margin about ten to the millimeter; the.se
grooves radiate primarily from two imaginary lines, one of which (somewhat
as in L. tenera and L. scabra) is median to the umbo and the base. The
other is nearly similarly situated with regard to the former, posteriorly, as
the margin of the truncature is, anteriorly ; hinge-line smooth. Max. alt.
8.00; max. Ion. 8.00; max. diam. 4.00; hinge-line 2.75 ; Ion. of truncature
6.00 mm.

A valve of this interesting species was dredged at Barbados, in 100 fms. ; the
type was dredged by the Fish Commi-ssion steamer "Albatross" at Station
2322, near Havana, in 115 fms.

The peculiarities of sculpture, ajjart from all the other characters, sufficiently
separate this from any other described species.

Genus LIMATULA S. Wood.

Limatula setifera, n. s.

iLimattda ovata Dall, Bull. M. C. Z., IX. p. 118, 1881 (not of Wood or Jeffreys).

Shell ovate, inflated, white, with about thirty-four radiating acute ribs,
strongest in the middle of tlie valve, with a fine sharp thread in the channel
between each pair of ribs ; both ribs and threads sharp, thin, and produced
into a series of sharp spines, which in perfect specimens are nearly as high as

VOL. XII. — NO. 6. 15

226 BULLETIN OF THE

the ribs which seem to connect and bisect their bases ; the intercalary threads
appear only in the basal half of the valve and would be wanting in the young ;
toward the sides, or rather ends, of the valve, the spines tend to widen into
scales concentric to the umbo ; the anterior edges are slightly produced, and
the ribs slightly oblique, so that the rib wRich is median to the base when
traced to the umbo is a little posterior to the median line there ; there is no
median sulcus ; cardinal area and ligamental sulcus wide antero-posteriorly and
short in the opposite direction ; cardinal margin simple, pit very shallow ; inte-
rior grooved and basal margin dentate, especially toward the middle, to corre-
spond with the ribs; shell nearly equilateral. Alt. 9.00; Ion. 5.75; diam.
6.00 ; hinge-line 3.50 mm.

Viilves were obtained by Sic^bee, near Havana, in 450 fms. ; Station 272,
near Barbados, in 76 fms.; and by the U. S. Fish Commission at Station 2612,
31 miles S. E. by S. | S. of Cape Lookout, North Carolina, in 52 fms., coarse
white sand, bottom temperature 67° F.

The sculpture separates this form from L. confusa E. A. Smith (uvata Jeffreys
non Wood), and from the fossil L. ovata Wood, which is usually much smaller.
It is perhaps the prettiest species yet described.

Genus LIM.^A Bronn.
Limsea Bronniana, n. s.

Shell small, white, stout, ovate, rather short and swollen ; with 14-16
smooth subequal rounded regular radiating ribs, with nearly equal impressed
channels between them ; tliere is a nearly smooth anterior and a smaller smooth
posterior area without ribs ; the concentric lines of growth are usually faint ;
shell sightly produced at the anterior base ; interior smooth, grooved for the
ribs, strongly dentate at the basal margin ; cardinal area narrow, longest in
front of the large and strong cartilage pit ; the middle part of the hinge
smooth, the angles each with four or five strong teeth, obliquely set ; alt. 3.5 ;
Ion. 3.1 ; hinge-line 2.0; diam. 2.5 mm.

Yar. lata. Shell larger, thinner, less inflated, sculpture less prominent
radially, the lines of growth more prominent, the ribs less elevated and
numbering 20-26, with the posterior area radiately striate, the interspaces less
channelled, the umbo nearly smooth. Alt. 5.3; Ion. 5.2; hinge-line 2.5;
diam. 3.1 mm.

The typical form was taken by the U. S Fish Commission off the coast of
the Carolinas, at Stations 2596, 2612, and 2619, in 15-52 fms. ; by the " Blake,"
in 100 fms., near Barbados. The variety lata was obtained by the " Blake" off
Havana, in 452 fms. ; at Station 2, in 804 fms.; and at Station 100, near
Havana, in 400 fms.

This species is immediately distinguishable from L. Sarsi by its sculpture,
and by the obliquity of the teeth on the angles of the hinge margin. The

MUSEUM OF COMPARATIVE ZOOLOGY. 227

variety is evidently a more southern form and looks specifically distinct, but
having only a few separate valves it seems better to wait for more material,
which may prove more nearly intermediate. If distinct, it may be called
L. lata.

Family SPONDYLID^E.

Genus SPONDYLUS Linn^.

Spondylus Gussoni Cqsta.

Spondylus Gussoni Costa, Cat. Sist., p. xlii, 1829 ; Phil., Moll. Sicil., I. p. 87, pL v.

f. 16, 1836.

Found at Station 150, near Xevis, in 375 fms. ; at Station 193, in 169 fms. ;
at Station 278, Barbados, in 69 fms.; at Station 262, Grenada, in 92 fms. ; and
in 640 fms., Yucatan Strait.

The specimens have been compared with authentic European examples, and
agree precisely.

Genus PLICATULA Lamarck.
Plicatiila spondyloidea Meuschen.

Ostrea spondyloidea Meuschen, Mus. Gronovianum, 1778.

Plicalula ramosa Lamarck, An. s. Vert., ed. i., VI. p. 184, 1819.

P. marginata Say, Journ. Acad. Nat. Sci. Phil., IV. p. 136, pi. ix. f. 4.

Found in 36 fms. at Station 12 ; in 54 fms. off Sombrero Island; and in
640 fms. in Yucatan Strait; in the last case probably ejected by a fish.

This extremely variable shell has been erected into a number of species. It
varies from nearly smooth to strongly plicate; from greenish white to closely
streaked with brown. The young are rather pretty, the old ones coarse and
rude. The number of ribs is most numerous when the shell is attached to a
flattish surface ; the ribs are fewest and most coarse when the shells grow on
one another like reef oysters.

Family DIMYID.^.

Genus DIMYA Rouault.

Dimya Rouault, Mem. See. Geologique de France, 2me ser., III. p. 471, 1848.

Stoliczka, Cret. Pelec. of S. India, Pal. Indica, III., ser. vi., pp. xxii, 397, 1871 ;
{Aviculidce, subfamily Vulsellina). Dall, Science, Feb. 16, 1883, p. 51.

Woodward, Man. Moll., 2d ed. by Tate, p. 408 {Oslreida), 1871.
Margariona (Dall, Ms. 1882), Kobelt, Naclir. Mai. Ges., p. 186, 1882.
Type Dimya Deshayesiana Rouault.

228 BULLETIN OF THE

Shell inequilateral, inequivalve, closed ; upper or left valve slightly smaller,
lower or right valve attached to some extraneous object; external layers
nacreous ; inner surface porcellanous ; epidermis none or very little ; liga-
ment linear, minute; cartilage inserted in a triangular pit in the cavity of
the beak; hinge-line short, straight; inner margins radiatingly wrinkled ; pallial
line simple. Mantle completely open, margin papillose without ocelli; gills
single, one on each side composed of a single row of long filaments, palpi
none; anal end of intestine produced, free; sexes separate; foot none; anterior
adductor single, distinct ; posterior adductor double, and leaving a pair of
closely approximated subequal impressions on the shell ; mouth with distinct
lips ; visceral mass smalL

Dimya Deshayesiana RonACLT.

Dimya Deshayesiuna Rouault, 1. c, p. 471, pi. xv. figs. 3, 3 a, 8 b, 1848. Eocene of
Bos d'Arros, France, equivalent in age to the Paris Basin eocene.
Stoliczka, 1. c, p. 397, 1871.
Anomia intustriata D'Archiac, Mem. See. G^ologique de France, 2me ser., III. p. 441,

pi. xiii. figs. (9 a, 10 a?), 11, 1848.*
Dimya Deshayesiana Tate, Woodward, Man. Moll, 2d ed., p. 408, 1871.

Dimya argentea Dall.

Plate IT. FigB. 5 a, 6 b.

Shell white, micaceous silvery outside, opaque brilliant porcellanous white
inside ; irregular, laterally compressed, attached by the beak of the right valve
(to a dead echinus-test, etc.), which is deeper and larger than the other ; ex-
terior obscurely finely radiately striate ; outline irregularly ovate, broader be-
hind ; hinge-line short, straight, without notch or auricles ; in well-developed
sxamples it has a leaf-shaped wrinkled area on each side of a small impressed
triangiilar area, below and partly under which is a small, deep, subtriangular
pit for the brown homy cartilage ; ligament hardly perceptible, linear, nearly
as long as the hinge-line; interior with an impressed area bounded by the

* It se^ms very probable that the shells described as Anomia by D'Archiac were
all of the same species as that subsequently described by Rouault. Fig. 11 cer-
tainly represents the same shell, and it looks as if Figs. 9a and Ida represented
attached valves of Dimya which had been worn through at the point of attach-
ment of the lower valve, and the resulting accidental perforation taken to be
normal by D'Archiac. The interior markings are precisely similar, as far as can
be judged from the figures. In the absence of specimens, however, it is safer to
preserve the specific name of Rouault, which certainly relates exclusively to the
species as we now understand it. Rouault's paper was read in 1847, but seems to
have been published in 1848, while .the volume appeared as a whole in 1849.

MUSEUM OF COMPARATIVE ZOOLOGY. 229

pallial line, the region of which is slightly raised, the shell falling away from
it toward the centre and the periphery ; outside of the pallial line smooth or
marked by irregularities of growth ; inside with fine radiating wrinkles (due to
the papillae on the mantle-edge which are strongest at their distal ends, and
which there form a narrow band of elevated waves and grooves which borders
the body cavity of the shell and is strongest near the hinge); muscular impres-
sions distinct, the double impression of the adductor more duplex in the right
valve than in the other, the two parts of the muscle being slightly twisted one
upon the other ; the anterior adductor slightly nearer the beak than the other ;
margin partaking of the irregularities of the object to which the shell is at-
tached, usually sharp, simple and very thin, as in a young oyster. Lon. of
shell 10.5, lat. 12.0 mm.

Soft Parts. Mantle open throughout; its margin thickened, smooth, white,
rather distantly studded with a single line of small (in their contracted state
strawberry-shaped) papillae, more elongated toward the middle; there are no
ocelli, but the central portions of the papillae often show a pronounced dusky
tinge ; within the line of the papillae a thin smooth " curtain " or band of tissue
at right angles to the plane of the shell extends toward the opposite valve,
with a width, in its contracted state, about equal to the thickened marginal
band; inner portions of the mantle thin translucent and studded with aggre-
gations of more opaque whitish cloudy dots or nebulous markings ; the -mar-
gin of the mantle has its extreme edge brownish near the anal end, and is
somewhat firmly attached to the shell. The adductors are composed of strong,
stiff', unstriated, easily separable fibrillae, of a greenish cream-color and very
polished surface, and are provided with a thin but stout investing fascia; as
before stated, the posterior adductor is composed of two subequal rounded
portions slightly twisted on each other as if to admit of a slight lateral motion
of the upper valve upon the lower one ; they are quite round except where in
contact ; the anterior adductor, with an area equal to one of the halves of the
posterior muscle, is more elongated and pointed toward its outer end ; the ali-
mentary canal is small, apparently simple, with one certainly and possibly
two coils buried in the liver; the oral aperture, just below and behind the
anterior adductor, presents a curious resemblance in its form to the epistome
of a polyzoan ; the form is not absolutely the same (owing to the alcohol ?) in
diff'erent individuals, and in the largest one examined was much more pro-
nounced than in smaller examples ; there is first a soft ovoid prominence at
whose apex projects from a depression a linguiform lip or projection over
which is a waved fold with a deep sulcus between ; from the lateral ends of
the " lip " and fold a wrinkle extends forward and then backward, the two
wrinkles on each side parallel with a shallow depression continuous with the
sulcus between them; the oral aperture appears at the bottom of the sulcus
and will receive the end of a probe as large as a good-sized pin, but the course
of the oesophagus, being lacerated, could not be continuously traced; the lip
and fold in the oldest example dissected appeared to have a hard, almost

230 BULLETIN OF THE

cartilaginous consistency, in others they were soft like ordinary tissue; the
other end of the intestine passes over the posterior adductors firmly knitted
to the fascia by connective tissue and having two thin broad bands of muscle
parallel with it which seem adapted to give some voluntary motion to the
free end of the intestine, which projects 10.0 mm. beyond its attachments, near
its end is deeply constricted, and ends in a small round centrally pierced but-
ton-like expansion. The diameter of the terminal button is about 0.37, of the
constriction 0.25, of the internal tube about 0.25, and the average of the free
part of the intestine 0.5-0.7 mm. The diameter of the tube is somewhat
irregular, and it contained, in the specimens examined, more or less dark gray
fsecal matter. In the structure of the two ends of the alimentary canal this
mollusk resembles Trigonia and some Fectinidce. The liver presented differ-
ent appearances in different specimens ; in younger but pretty well developed
examples it presented the appearance of a number of free simple deep olive-
green tubules radiating from a comparatively small plexus as a centre, through
which passed the intestine ; in older specimens it had actj^uired a more solid
and compact consistency, the single tubules were no longer recognizable, the
interior was brownish when cut, and the outer surface was ornamented with
a few whitish dendritic branches extending over it from the generative glands
below the liver ; the mass of the ovaries or sperm glands occupy part of an
irregular space behind the mouth and in front of the posterior adductors, the
liver projects into it beneath its surface in the median line, its lower external
surface is keeled, but a foot is not developed; the substance of the ovary is
disposed in, proportionally, rather large ramifications, with acorn-shaped*
lobules containing granular cream-colored matter irregularly distributed on
the inner surface ; the supposed male gland was similar, but the contents were
of a finer consistency and of a more greenish color. The texture of all the
viscera is extremely loose and delicate, and their disposition differed in minor
details in all the animals examined. This might be partly owing to the vio-
lence necessary, even with the greatest care, to open the valves so that the
soft parts could be examined, and the extreme delicacy of the visceral tissues.
The organ of Bojanus was not distinctly recognized. The gills are of a very
primitive type, resembling the palpi of some Acephala in form, though not in
construction ; there are no true palpi ; the gills originate above the mouth
and behind the anterior adductor ; their bases extend backward in a nearly
right line to the lower surface of the posterior adductors, behind which two
strong bands, one to each gill, anchor the main stem of the gill to the thick-
ened margin of the mantle above and behind the adductors ; beyond the point
where these muscular bands are attached, the bases of the gills curve downward,
hanging free, and terminate in a point reaching to the lower posterior edge of
the mantle. The aforesaid bases are broad stout bands carrying each a large
vessel and sustained by a rather stiff (chitinous) framework, or fundamental
tissue. From them proceed the gill-filaments, each of which is extremely long,

* Like the acorn without its cup.

MUSEUM OF COMPARATIVE ZOOLOGY. 231

slender, and composed of a central more solid rod with a tube on each side of
it, and with its distal end enlarged in a somewhat hoof-shaped manner. From
the blood-vessel in the base a single tube issues to the root of each rod and
after continuing a short distance divides, one branch passing down one side
of the rod, the other crossing to the opposite side and continuing to join the
first again at the distal extremity ; it appeared as if the fluid passed down on
one side and back by the other. The exterior of the filaments is abundantly
ciliated, and though there is no organic connection, there is something in the
character of the surface of the knobs at the ends of the filaments which makes
them adhere tenaciously to each other or to any other part of the gill they
may touch; I could see nothing peculiar, but the mechanical effect showed
that something was there to produce it. The longest separate filaments were
5.0 mm. in length, the diameter of the stalk of the rods measured about 0.025,
and the knobs at the end 0.050 to 0.075 mm. The general outline of the gill,
as traced by the terminations of the filaments, differs in different individuals
to some extent.

Habitat. Station 231 of Bartlett, in 1878-79, 95 fms., coarse .sand and rock,
off St. Vincent, bottom temperature 61°.5 F. Specimens on dead sea-urchin,
tests and on the thin marginal expansion of the shell of Phorus; Barbados, in
73 fms., coarse coral sand and broken shelLs, bottom temperature 70°.7 F.;
Station 134, off Frederikstadt, Santa Cruz, in 248 fms., coarse sand and broken
shells, bottom temperature 54°. 5 F. ; and Station 238, off the Grenadines, in
127 fms., fine coral sand, bottom temperature 56°.0 F. A fresh valve was
dredged by the U. S. Fish Commission, 36 miles S. ^ W. of Cape Hatteras
(Station 2601), in 107 fms., bottom temperature 64°.4 F. It will be seen from
the above figures that the species inhabits the warmer area.

The form argentea above described appears closely similar to the figures of
D. Deshayesiana, given by Rouault. So nearly identical are they, that, until
I have had an opportunity of comparing the recent and fossil forms, I feel
barely justified in separating them. The figure of Ostrea temdplkata of
Seguenza * resembled Dimija so much in its exterior that I requested Pro-
fessor Seguenza to examine the interior and inform me of the character of the
muscular impressions. He did so, and also most kindly sent me two valves
which determine the correctness of my suspicions. The shell is an undoubted
Dlmya. It differs from Rouault's figures and from the recent argentea in
having a much stronger and coarser sculpture of divaricating radii, in its
shorter hinge-line and proportionally larger cartilage pit ; the visceral area is
smooth or slightly dotted, while in the recent form it is striated ; the relative
position of the muscular impressions and their form also differ somewhat in
the two shells. For these reasons, unless a considerable individual variation
may be hereafter demonstrated, it would seem that the two species may be
regarded as distinct.

* Form. Terz. n. prov. Reggio, p. 123, t. xii. figs. 1, 1 a, 16, 1880. Miocene,
Aquitaniano, to Pliocene, Astiano, Calabria.

232 BULLETIN OF THE

The anatomical characters above described indicate an organization of ancient
and rather primitive type. The gills are especially notable. For this reason
it would seem probable that, among the multitude of oysters described from
strata of the Carboniferous period to those of recent seas, numerous species of
Dimya might be discovered by a more critical examination of the interior and
muscular impressions.

The systematic position of this remarkable mollusk is difficult to detennine
in existing classifications. Woodward, from Rouault's description, places it in
the Ostreidce, suggesting that the anterior adductor scar is paralleled by a
small anterior scar seen in some species of Peden* Stoliczka says : " Its form
and structure resemble Flacuna or Placenta, but there are no hinge teeth
present ; the two muscular scars separate it from all Ostreacea, and as there is
an anterior muscular scar indicated in most of the Mytilacea, the classification
of the genus may be more correct in this place. If this should not be the
case, tlie only other classification admissible would be near M^ochama in the
Anatinidce."

The genus is peculiar in having but one single gill on each side, nearly all
others with which it can be said to have relations being provided with two,
though one of these may be nearly obsolete ; nor does any genus occur to me
as having gills composed of rod-like filaments free from organic connection
except at their base. The free lamella; of Pecten are perhaps the nearest ana-
logue. The mantle, except in the absence of ocelli, resembles that of Pecten ;
from which, however, the nacreous shell, absence of the foot, and many details
of structure strongly separate it. "We are too ignorant, however, of the adult
anatomy of mollusks in general (though the fact is very generally ignored), to
dogmatize on assumptions which the discoveries of twenty-four hours may
overthrow. Two things, however, appear reasonably certain : first, that the
genus Dimya occupies a sort of middle place between the Mytilacea and Os-
treacea without being admissible into the families of either group as at present
constituted ; secondly, that the total rejection is necessary of tlie ordinal
groups founded on the number of muscles (i. e. Monomyaria, Heteromyaria,
and Dimyaria), which have been so long in vogue. Stoliczka's remarks, in his
introduction to the Cretaceous Pelecypoda of India, are worthy of note in this
connection, and" appear to the writer to be full of sound common-sense. Even
the proposition by Gill of the order Heteromyaria, in 1871, was an indication
of the crumbling of the old-fashioned classification, which can onl}"^ be replaced
in a satisfactory manner by a great advance in our knowledge of the anatomy
of animals which have been carelessly lumped together on the unwarranted
assumption that the characteristics of the soft parts of one would suffice to
classify several hundred others by their shells.

Since the above was written, Dr. Paul Fischer, in his excellent Manuel de
Conchyliologie, finding, as I have done, that the features heretofore taken as
bases for ordinal subdivisions of the Pelecypods are insufficiently important for

* This is, however, due to the mantle, not to an adductor muscle.

MUSEUM OF COMPARATIVE ZOOLOGY. 233

such a purpose, has proposed a division based on the number of branchiae,
those with two branchiae on each side composing his order Tetrabranchia, as
opposed to the Dibranchia with one gill on each side. But to this arrange-
ment Dimya does not lend itself ; Ostrea, Mytilus, etc., to which it is certainly-
most nearly allied, having four branchial leaves, to say nothing of the addi-
tional accessory plates which may be taken as representing a third pair.
Moreover, some species of Area (ex. A. ectocomata Dall) have but a single
pair, while others have two or three. All the evidence points to the conclu-
sion that the Pelecyjjoda comprise but a single order, knit closely together by
inter-ramifying characters.

The genus Dimyodon Munier Chalmas (1886), of the great Oolite, appears
to differ from Dimya by the projection of the wrinkled hinge-areas so as to
form striated teeth, recalling those of Plicatida, and by its single posterior
adductor scar. It has not been reported in a recent state.

Family AVICULID^.

Genus AVICULA Lamarck.

Avicula atlantica Lamarck.
Avicula sp. indet. Dall, Bull. M. C. Z., IX. p. 117.

Habitat. Station 26, 116 fms.

Two very young living specimens were obtained here, and worn fragments
were found from other localities. It is spread over the whole Antillean region,
and northward in suitable localities at least as far as Hatteras.

Family MYTILTD.^.

Genus MYTJLUS Linn:^.

Mytilus exustus Lixne.
Myiilus exustus Linn^, Dall, Bull. M. C. Z., IX. p. 117.

Habitat. Sigsbee, off Havana, 158 fms. ; Barbados, 100 fms., etc.

This is spread over all the shores of the Antilles, and the specimens obtained
from more than a few fathoms are drifted or disgorged by fishes into the deeper
water. The species does not live in deep water.

234 BULLETIN OF THE

Genus MODIOLA Lamarck.
Modiola polita Verrill and Smith.

Plate VI. Fig. 3.

Modiola polita V. & S., Am. Journ. Sci., XX. pp. 392, 400, Oct. 25, 1880. Verrill,
Proc. U. S. Nat. Mu8., IIL p. 402, 1881 ; Dall, Bull. M. C. Z., IX,
p. 116, 1881 ; Verrill, Trans. Conn. Acad. Sci., V. p. 578, July, 1882.

Mytilus luteus Jeffreys, Rep. Brit. Assoc. Adv. Sci., 1880 (name only), Ann. Mag,
Nat. Hist, Oct. 1880, p. 315 (name only).

Modiola lutea Fischer, Journ. de Conchyl., Jan. 1882, p. 52 (described).

Modiotaria polita Verrill, Trans. Conn. Acad., VI. p. 281, pi. xxx. f. 12, 1884.

Modiella polita Monterosato, Nora. Conch. Medit., p. 12, 1884.

Modiola {Ami/gdalum) lutea Fischer, Man. Conchyl., p. 968, 1885.

Habitat. North Atlantic. I\Iediterranean (Monterosato) ; Gulf of Gascony,
and Marocco coast, Bay of Biscay (Fisclier) ; New En<,'land coast (Verrill) ;
Gulf of Mexico and Antilles (Blake Exp.), Station 4.3, 339 fins., off Tortugas,
bottom temperature, 45° .0 F. ; Station 47, 321 fms,, bottom temperature, 46°,5;
on the European side to over 1000 fms. The young were obtained rather
abundantly at Station 2644, near Cape Florida, in 193 fms,, by the U. S, Fish
Commission.

This very elegant mollusk attains a length of 50.0 mm. (2 inches) without
marked change of proportions from the very young to the adult .stage. The
smallest are waxen translucent; as they grow older, some of them may be
prettily maculated with sagittate opaque white spots, radiating in a reticulate
manner from the beak. As they attain maturity, tliey take on a magnificent
golden brown color, especially deep toward the anterior end. The shell is
pearly white, all these colors being situate in the epidermis, which, usually
very polished and smooth, in rare cases may be somewhat irregularly radiately
striate with the finest striae.

The soft parts of this species are delicate, and contain but little solid matter
compared with the capacity of the valves. The four labial palpi are moderate
and lamellated ; the gills, two on each side, extend the whole length of the
animal, the inner lamina on each side being somewhat shorter than the outer
one ; the mantle margin is thin, smooth, and simple ; the single branchial
opening has (in alcohol) one hardly perceptible row of inconspicuous rounded
papillae ; the mantle behind the commissure of the branchial orifice is com-
pletely open; the muscles are slender, the posterior adductor the largest, then
the pedal and the anterior adductors, in that order; the foot is nearly half as
long as the shell, longitudinally wrinkled, digitiform, very slender ; the pointed
tip is grooved, the byssal socket at the base is strongly marked. This long and
slender foot is well adapted for nest-weaving, in which this species excels.

MUSEUM OF COMPARATIVE ZOOLOGY. 235

When dredged, the washed contents of the trawl may present several
bodies looking like wads of fine flax soaked in mud, and having various dead
shells or worm-tubes entangled therein. In this unpromising nidus is hidden
our gem of the sea. Long continued gentle washing under a stream from the
wat«r-cock finally removes most of the mud. Immersed in water, we see that
the nest is composed of the finest and most silky threads, inextricably inter-
laced and of great strength. Among them the young nestle until they are
ready to spin for themselves. Many of the threads centre at and are con-
nected with the byssal sinus, from which much force is necessary to detach
them.

It will be seen from the notes on the soft parts that this mollusk is most
nearly related to Modiola, and not to Modiolaria, as before examination I sus-
pected. I have compared it with the chief types, and there is no doubt of this.
If we separate the polished species from the bearded mussels, this species, ac-
cording to Fischer, may be referred to Amygdalum Megerle (1811), from which
it hardly differs. Monterosato proposed the name Modiella for it ; but this
had been used a year earlier by James Hall (1883) for a different group.

Modiola opifex Sat.

Modida opt/ex Say, Journ. Acad. Nat. Sci. Phil., IV. p. 369, pi. xix. figs. 2, 2a, 2 b,
1825; Phil. Abbild. und Beschr. n. Conchyl., III. Modiola, p. 21, t. ii. fig. 7.

One valve was dredged from 640 fms. in Yucatan Strait, a depth which it
doubtless reached in some accidental manner. This species was described by
Say as attached to Fecten nodosus, and found in a mass of sand grains of its
own collecting. Kroyer had it from Brazil, and the U. S. Fish Commission
has dredged it to within a few miles of Cape Hatteras, but only as separated
valves. It forms a transition, conchologically, between Modiolaria and the
group typified by Modiola semen, sometimes called Botula.

Genus CRENELiLiA Brown.

Crenella decussata Montagu.

Crenella decussata Montagu, Dall, Bull. M. C. Z., IX. p. 116.
Nuculocardia divaricata D'Orbigny, II. p. 311, pi. xxvii. figs. 56-59, 1845.

Habitat. Barbados, 100 fms. (Alaska, California, New England, British
seas, Norway, etc.).

This little shell is proportionately a little more solid and strong than north-
em specimens, and the crenulations which exist in both, and from which the
group takes its name, partake of this difference. I have seen nothing, how-
ever, in the few specimens I have been able to examine, which would authorize
the separation of the southern form from the northern one.

236 BULLETIN OF THE

Genus MODIOLARIA Beck.

Modiolaria lateralis Sat.

Modiolanasp. indet., Dall, Bull. M. C. Z., IX. p. 117.

Mytilus lateralis Say, Journ. Acad. Nat. Sci. Phil, II. p. 264, 1822.

Flate VI. Flsrs. 7, 8.

Habitat. West Florida, 30 fms., living. East coast of the United States,
from Florida neariy to Cape Hatteras, at various depths, but mostly in compara-
tively shallow water.

Another southern species of Modiolaria, but which is not reported from our
coasts as far as the books indicate, is M. lignea Reeve, which I have received
from C. W. Johnson, of St. Augustine, and from Charles T. Simpson, who col-
lected it at Tampa, thus occurring on both coasts of the peninsula. It is nota-
ble for having no radiating striae, and for its rich chestnut-color, bluish black
on the umbones and toward the margins. It grows over an inch in length
and spins a fibrous nest. M. cinnamomea Chemn. is another Florida species
which almost or quite reaches the latitude of Cape Hatteras.

Family ARCIDJE.

Genus LIMOPSIS Sassy.

Limopsis minuta Philippi,

Limopsis minuta Philippi, Dall, Bull. M. C. Z., IX. p. 119.

Habitat. Gulf of Mexico, west of Florida, 30 fms. ; Station 36, 84 fms. ;
Barbados, 100 fms.; Bache, April 22, 1872, Lat. 21° 14' N., 100 fms.; Sigsbee,
off Cuba, 119 fms.; Starion 20, 220 fms.; off Morro Light, 292 fms.; Station
19, 310 fms. ; Sigsbee, ofif Havana, 450 fms.; Yucatan Strait, 640 fms.; Sta-
tion 2, 805 fms. ; Station 253, near Grenada, in 92 fms.

This species is named minuta on the authority of Dr. Jeffreys, it having
been impracticable for the writer to compare with the various fossil forms of
south Europe. It seems to agree sufficiently well with the material I have
been able to examine, and which has been named minuta by other naturalists.

Limopsis tenella Jeffreys.

Limopsis tenella Jeffreys, Dall, Bull. M. C Z., IX. p. 118.

Habitat. Station 44, 539 fms.; Station 41, 860 fms.; Station 56, 888 fms.;
Station 33, 1568 fms.

MUSEUM OF COMPAKATIVE ZOOLOGY. 237

This seemg to be an excellent species. The width of the area varies more,
however, than one would suppose from Jeffreys' description. It is sometimes
wider than in L. aurita, but the hinge-line is longer and the comers more
nearly rectangular than in that species.

Limopsis antillensis Dall.

Limopsis antillensis Dall Bull. M. C. Z., IX. p. 119.

Plate VIII. FlKB. 7, 7 a.

Habitat. Sigsbee, off Havana, 80 fms.

No more specimens of this species have been found in the collection. I sus-
pect it to attain a considerably larger size when adult.

Limopsis cristata Jeffbbts.

Limopsis cristata Jeffreys, Dall, Bull. M. C. Z., IX. p. 119.

Limopsis minuta var.?

Habitat. Yucatan Strait, 640 fms.

On further study I am somewhat in doubt whether these three poor valves
do not belong to a young stage of L. minuta. The specimens in the Jeffreys
collection are all so small, or in such poor condition, that I cannot regard them
as affording suflScient evidence of a species different from minuta^ though per-
haps those in the British Museum may be better preserved.

Limopsis aurita Brocchi.

Limopsis aurita Brocchi, Dall, Bull. M. C. Z., IX. p. 118, 1881.

Habitat. Gulf of Mexico, west of Florida, 30 fms.; Station 36, 84 fms.;
Station 20, 220 fms.; Gulf Stream, 447 fms.; Stations 253 and 264, in 92 and
416 fms., near Grenada ; Station 269, near St. Vincent, in 124 fms.; Station
176, off Dominica, in 391 fms.; and Station 16.3, off Guadelupe, in 769-878
fms. The bottom temperatures varied from 39° to 70° F., the average being
about 55° F.

Var. paucidentata. Shell thinner, smoother, with narrower margin, smaller
and fewer (4 -|- 4) teeth, ends of valves less expanded, less oblique ; radiating
sculpture red viced to rows of small scars ; concentric sculpture obsolete. Alt.
9.0 ; max. lat. 9.0 mm.

Two valves at Station 117, in 874 fms., gray ooze, near Jamaica; bottom
temperature 40°.0 F.

238 BULLETIN OF THE

Genus PECTUNCULUS Lamarck.

Pectunculus undatus Linne.

Area undata Linne, S. N., ed. xii., p. 1142; Hanley, Shells of Lin., p. 97

Pectunculus undulatus Lam., An. s. Vert., VI. p. 50, 1819.

Pectunculus lineatus Reeve, Conch. Icon., pi. v. fig. 25, 1843.

Pectunculus scriptus (Born) Reeve (young shells).

Pectunculus hirtus Phil. Zeitschr. fUr Mai., 1846, p. 191.

"? Pectunculus angulatus Lam.

? Pectunculus pennaceus Lam. (^ decussatus Linn.).

Pectunculus passus Conrad, Tert. Foss. U. S., p. 64, pi. xxxv. fig. 3, 1844 ; Tuomey &

Holmes, Plioc. Fos. S. C, p. 48, pi. xvii. fig. 3 (good), 1855.
Pectunculus quinquerugatus Conrad, Ann. Journ. Sci., IV. p. 346; T. & H., 1. c, p. 49,

pi. xvii. fig. 4, 1855.
Pectunculus carolinensis Holmes, Post Pliocene Fos. S. C, p. 15, pi. iii. fig. 4, 1858.

(Not of Conrad " 1838 " = 1844.)
Pectunculus tricenarius Conrad, Tert. Fos. U. S., p. 63, pi. xxxv. fig. 1, 1844.
Pectunculus parilis Conrad, 1. c, p. 64, pi. xxxvi. fig. 2, 1844.

A single valve of the scriptus variety at Station 127, Santa Cruz Island, in
38 fms.

I have been unable to examine any specimen of P. pennaceus Lamarck (= F.
decussatus L. sp.), or at lea.st none of the specimens so named which I have
been able to examine have had at one end of the beaks the heart-shaped area
described by Lamarck and Hanley. I am therefore unable to say whether it
and its synonyms should find a place here, as suggested by D'Orbigny and
Krebs, both good judges. For the rest, it is evident that an absurd number
of species have been made of this group, especially in fossils, where a man is
allowed to describe a species from one valve without adverse comment.

A careful examination of a large number of good specimens of this species of
Pectunculus will show any competent observer, — 1st, that the reticulated sculp-
ture is always present on the umbones of a perfect shell, and its greater or less
extension and uniformity over the valves in the adult varies with the individ-
ual; 2d, that the hinge in this group is very mutable within certain limits, and
undergoes great changes with age, and the number of teeth is greater in the
adult than in the young; 3d, that a certain lateral expansion and angulation,
which are very marked in ' some .specimens, are variable characters ; 4th, that,
as one goes south, the shells of this species become more brightly colored, more
inflated, more purely porcellanous, and show a tendency to equalizing the
strength of the radiating and concentric sculpture, thereby producing reticu-
lation, which governs the disposition of the pilose epidermis and alters the
aspect accordingly; they are also rather smaller when full grown. I have
come to this opinion through the study of a large series collected by the U. S.
Fish Commission, and another existing in the Jeffreys collection, covering

MUSEUM OF COMPARATIVE ZOOLOGY. 239

P. undatus and P. pilosus. The carolinensis form of Holmes is little cancel-
lated, and grows very large ; 10 cm. is not an uncommon length. It is abun-
dant in rather deep water as far north as Cape Hatteras, but not common
inshore. In the Antilles the lineatus form is not uncommon, and averages
smaller than the preceding. The strength of the radiating undulations is very-
variable ; they may be very evident, numerous, and even a little keeled in the
middle, or flat, rounded, and strongest in the middle part of the valve, or alto-
gether obsolete ; they are rarely discernible in cancellated specimens, but many
which are not cancellated are entirely without them.

Pectuncultis pectinatus Gmelin.

Pectunculus variegatus? D'Orb., Bull. M. C. Z., IX. p. 118.

Area pectinata Gmel., S. N. 3313, 1790.

Pectunculus pectinatus Lara., Hanley, Rec. Sh. p. 165. Reeve, Conch. Icon., fig. 28.

Area pectunculus minor Chemnitz, VII. t. 58, figs. 570, 571, 1784.

? Pectunculus oculatus Reeve, Conch. Icon., fig. 38, 1843.

Pectunculus pectiniformis D'Orblgny, Moll. Cuba, p. 313 (? not of Lamarck).

P. aratus Conrad, Tuomey & Holmes, Pliocene Fos., p. 50, pi. xvii. figs. 6 a, 6 b.

P. charkstonensis Holmes, Post Pliocene Fos., p. 16, pi. iiL fig. 6, 1858.

Habitat. Gulf of Mexico, west of Florida, 30 fms. ; Charlotte Harbor,
Florida, 13 fms.; off Sombrero in 54 fms.; Sigsbee, off Havana, 80-119 fms.;
off Gordon Key in 68 fms.; Station 10, off Cuba, in 37 fms.; Station 32, Lat.
23° 32' and Lon. 85° 5' W. Gr., in 95 fms.; Stations 36 and 45, off Cuba, in
84 and 101 fms.; Station 56, in 175 fms., off Havana; Station 117, 874 fms.
(one valve); and Station 278, in 69 fms., Barbados.

Var. carinatus Dall, at Station 247, near Grenada, in 170 fms., ooze, bottom
temperature 53.°5 F. (living), and at Barbados in 100 fms.

The shell before us is with certainty the pectinatus of the best authorities ;
one of its varieties seems to have been identified with the East Indian Area
pectunculus of Linn6 (Pectunculus pectiniformis Lam., not D'Orb.), which
I have not foimd authenticated from the "West Indies. It is probably the
oculatus of Reeve, and certainly the aratus of Conrad. The different forms
observed by me are three.

The first one has fewer ribs (about 20-30), about sixteen plications of the
basal margin inside ; very square channels between the ribs ; close set, elegant
concentric wrinkles over the whole ; and in many specimens a (sexual ?) pe-
culiar truncation of the shell behind the hinge-line with a consequent angu-
lation more or less pronounced ; the colors pink or rosy, or white with pinkish
blotches, with dark pink or brown blotches, or, oftener, variegated tracery of
lines. The above is most like the Linnean species, and is probably what has
been so named.

The second or typical form has the ribs' either more numerous (30-40) or
wider with shallow hardly channelled interspaces ; is whitish with brown

240 BULLETIN OF THE

tracery, usually "white inside, but occasionally very dark brown, this character
being apparently local; the concentric wrinkles are close but less strong, a
little wear makes them seem absent, and the shell smooth ribbed ; the same
dilFerences exist as to truncation ; this character is probably sexual.

The variety carinatv^ has the same number of ribs as the typical form, but
they are carrnated, and the interspaces toward the margin, owing to impressed
radiating lines, seem to have several small threads in them between the ribs ;
the concentric wrinkles are more distant, and take a lamellate aspect, forming,
with the ribs, a reticulation which seems very characteristic ; the shell is a
little more globose than the ordinary form, but not much ; otherwise it seems
precisely the same, and all the gradations, from flat wrinkled ribs to keeled
and reticulated ones, may be seen in the series before me. A single one taken
by itself would certainly appear distinct from the ordinary form, and this gives
us a hint of what we may expect when large numbers of specimens come to be
studied scientifically and with due regard to their geographical distribution.

Genus ARCA Linn^.

Area pectvmcTiloides Scacchi, var. orbiculata.

Area pectunculoides Scacchi, var. orbiculata, Dall, Bull. M. C. Z., IX. p. 121.

Plate VIII. Fig. 5.

Habitat of the variety : Gulf of Mexico, Station 33, 1568 fms. (one valve).

Typical form : Sigsbee, oflF Havana, in 480 fms.; Station 16, near Havana,
in 292 fms., living, bottom temperature, 66° ; Station 176, near Dominica, in
391 fms.; Station 211, near Martinique, in 357 fms.; and Station 230, off St.
Vincent, living, in 464 fms., bottom temperature 41°.5 F.

Examuiation of a large number of specimens in the Jefireys collection has
convinced me that the single valve described as variety orbiculata is merely
an extreme variety of the typical pectunculoides, and not distinct, as I suspected
then. It is, however, certain that all the American specimens are shorter and
rounder than those from farther east in the Atlantic sea-bed and the Norwegian
and arctic seas.

Area grenophia Eisso may be this species, but it was not figured, and the
description is quite insuflicient. Area pectunculoides, var. crenulata Verrill,
appears to have the form of var. orbiculata, the teeth of the Gulf specimens
above mentioned, the marginal crenulations of glomerula, and the sculpture
of the type of pectunculoides. I have only seen one right valve of crenuloM,
but both valves seem to be sculptured alike.

By a slip of the pen, in treating of Area glacialis Gray, Prof. Verrill (Trans.
Conn. Acad., V. 576, 1882) represents me as recording A. glacialis from the Gulf
of Mexico. This is an error ; as, in mentioning it in the Blake Preliminary

MUSEUM OF COxMPARATIVE ZOOLOGY. 241

Report (1. c, p. 121), I reported the pectunculoides (with which Prof, Verrill
had seemed disposed to unite glacialis) from the Gulf, but expressly objected
to its' identification with glacialis, which I do not know from that region.

Area polycyma Dall.

Arcapolycyma Dall, Bull. M. C. Z., IX. p. 122.
Plate Till. Figs. 3, 3 a.

Habitat. Barbados, 100 fms., three valves ; a single valve at Station 262,
near Grenada, in 92 fms.

Only one more valve of this interesting Little species has turned up since the
original specimens were described.

Area glomerula Dall.

Area glomerula Dall, Bull. M. C. Z., IX. p. 121, 1881.

Area (Scapharca ?) incequisculpta E. A. Smith, Chall. Rep. Lam., p. 267, pi. xvii. figs.
8 a- 8 c, 1885.

Plate VIII., Figs. 9, 9 a.

Habitat. Barbados, 100 fms. ; Bache, April 22, 1872, Lat. 21" 14', 100 fms.;
Station 20, 220 fms. ; Station 19, 310 fms. ; Sigsbee, oflf Havana, 450-480 fms.;
Station 100, off Havana, in 400 fms. ; Stations 206 and 211, in 170 and 357 fms.
off Martinique, bottom temperature 49°.0 F. to 62°.0 F. The Challenger
specimens were obtained from off Culebra Island, West Indies, at Station 24, in
390 fms., pteropod ooze.

The specimens described by me in 1881 were separated valves, and che
differences of sculpture, noted at the time, were set down to individual varia-
tion. Mr. Smith has, however, shown that the difference is between the two
valves of the same specimen. There is generally a single more prominent rib
on the posterior slope of the right valve, but nothing of the kind in the left
valve.

Area atirietilata Lamarck.

Area auriculata Lam., An. s. Vert., VI. p. 43, 1819.

Habitat. Station 142, in. 27 has., Flannegan's Passage; and at Station 12,
oflF Havana, in 36 fms.

A single living specimen of this species was obtained in each case. The
foot is of good size and deeply grooved, the byssus small. A sort of bridle of
tissue from below the mouth passes under the anterior adductors and thence
to the interior of the umbones, where it is strongly attached and then sw^eeps

VOL. XII. — NO. 6. 16

242 BULLETIN OF THE

back toward the lower edge of the posterior adductor. On this band between
the adductors are the gills, two on each side. There are three long filaments
and several granulations and pigmented dots on the mantle edge near the anal
orifice. The rest of the edge is nearly smooth. On each side of the anus is
a prominent whitish lobule, from which a tube seems to pass back over the
adductor and a shorter one toward or into the anal tube near its orifice.

Area lienosa Sat.
Area lienosa Say, Am. Conch., IV. pi. xxxvi. fig. 1, 1832.

One young specimen was dredged in 19 fms., west of Florida. The Fish
Commission has dredged in deeper water dead valves of this species measuring
115.0 mm. long., 65.0 mm. high, and 35.0 mm. in diameter (or 70.0 mm. for the
whole shell). These had about 38 ribs, narrow and sharply grooved on top
except in the older third where they were entire and uniformly closely trans-
versely waved. The epidermis is soft, profuse, moderately long, and dark
brown. The teeth are small, vertical, uniform and close set. The young are
sometimes sharply auriculate. The anterior outer margin of the area is not
covered with the black cartilage, which creeps up more and more in the middle
line, as the shell grows; thus producing a marked difference between young
and old.

Area reticulata Chemnitz.

Area auricidata Chemnitz, Conch. Cab., VII. p. 193, t. 64, f. 540, 1784 ; Gmelin, S. N.,
p. 3311, 1788; Dillwyn, Cat., I. p. 237, 1817 ; Lamarck, An. s. Vert., 2d ed., VI.
p. 475.

Area squamosa Lam., An. 8. Vert., 1st ed., VI. p. 45, 1819.

Area domingensis Lam., 1. c, p. 40, 1819 ; E. A. Smith, Chall. Rep., p. 265.

Area elathrata Lam., 1. c, p. 46, 1819, fide Deshayes.

Area elathrata Defrance, 1816, fide Nyst, Cat. Arc, 1848.

Area gradata Broderip & Sowerby, Zool. Journ., IV. p. 365, 1829.

"iArea congenita E. A. Smith, Chall. Rep., p. 265, pi. xvii. figs. 6, 6a.

Byssoarca divarieata Sowerby, P. Z. S., 183.3, p. 18 ; Reeve, Conch. Icon., pi. xvi.
fig. 108.

1 Area donaciformis Reeve, Conch. Icon., pi. xvi. fig. 104, 1844.

Dreaged at Stations 65 and 66, ofi" Havana, in 80-127 fms. ; at Station 21, off
Cuba, in 287 fms., dead ; at Station 32, in 95 fms., in the Gulf of Mexico, living ;
and at Station 262, near Grenada, in 92 fms., fine sand, bottom temperature 62°.
It has not been found living from more than 100 fms. This well-known species,
usually named gradatu or domingensis, appears quite variable in outline, espe-
cially in the young. Some of my specimens approach so closely to the figure
of A. congenita that it has suggested the idea that that may be merely an ex-

MUSEUM OF COMPARATIVE ZOOLOGY. 243

treme form of reticulata. In the absence of specimens for comparison, how-
ever, the question cannot be fairly settled. It is a shallow-water species, and
the material obtained by the " Blake " was all immature or dead.

Area Adaznsi Shuttlewoeth.

Area lactea C. B. Adams, non Linne.

Area ccelata Conrad, Tert. Form. U. S., p. 61, pi. xxxii. fig. 2 (1844), not of Reeve.

Dredged by Sigsbee oflf Havana in 80 fms. ; and at Station 220, near Santa
Lucia, in 116 fms.

This species is very common in shallow water throughout the "West Indies,
and extends northward nearly or quite to Cape Hatteras. Its simulated ribs
of trailing blisters give it a remarkably similar appearance to Area lactea, which
however has real ribs. There is a dwarf, very short squarish variety, which
from its greater proportional diameter (though not otherwise different) would
at first be separated as distinct, and which may be called Area Adamsi var.
Conradiana.

Area Noae Likne.

Area harhadensis D'Orbigny, II. p. 321, as of Petiver.
Area occidentalis Philippi, and C. B. Adams.

A valve of this common form appears in the collection from Charlotte Harbor,
Florida, in 13 fms. It is common in shallow water throughout the West Indies.
It is possible that the Antillean form may be separable from that of the Medi-
terranean, but I have not been able to examine the matter critically as yet.

Area umbonata Lamarck.

"What appears to be a dead valve of this species was dredged at Station 282,
near Barbados, in 154 fms. It may have been disgorged by a fish.

Area ectoeomata, n. s.

Plate VI. Fiffs. 9, 10.

Shell white, compressed, elongate, equivalve, very inequilateral ; covered with
a long, soft, silky red-brown epidermis projecting in ribbon-like strips, which
may be broken up into narrow fiat filaments, and project especially at the
lower posterior angle of the shell ; valves gaping slightly for the large stout
byssus ; external sculpture of narrow, somewhat irregular, minutely nodulous
concentric waves ; the interspaces sparsely radiately striate; these stria) and
little nodules correspond to thickened radii in the ribbon-like epidermis which
are seated on them ; these radii in old shells remain after the flattened web
which connected them is worn away, and so give to the older shells the aspect

244 BULLETIN OF THE

of having a different kind of epidermis ; shell extremely inequivalve, not one
sixth of its surface being anterior to the beaks ; hinge line straight ; area very-
narrow, its f;ection forming a V-shaped figure, the black part of the ligament
entirely posterior, generally beginning to show about as far behind the beaks as
the beaks are behind the anterior end of the hinge line ; hinge peculiar, teeth
transversely grooved, anterior end of the hinge with a few (4-6) teeth, irregu-
lar or tending to trend with the longest axis of the valve ; posterior end with
four or five elongated teeth nearly parallel with the hinge line; between these
the hinge is edentulous or faintly irregularly granulous; Ion. of figured type
26.0; max. alt. of do. 14.0; diam. 9.5 mm. A specimen obtained by the Fish
Commission measures 46.0 nmi. in length exclusive of the epidermis.

Dredged living at Station 193, near Martinique, in 169 fms., sand, bottom
temperature 51°; and at Station 300, off Barbados, in 82 fms., bottom temper-
ature 60" F.

Foot small, split in the median line; byssal groove large and deep; palpi
none; a single gill on each side with the broad margin of insertion curled
downward into a spiral at its posterior end; mantle margin thick, smooth,
plain, dotted with black posteriorly, otherwise (in spirit) whitish. The dots
are single, at regular intervals, and look much like ocelli.

This fine species does not closely resemble any I find described ; it is most
like a form I find in the collection named Area (Barbatia) Listeri Phil., but
which is dark colored, very much smaller, less compressed, and otherwise differ-
ent in various details. The present species belongs to the subgenus Barbatia,
and for those who give this group a generic value its name would be Barbatia
ectocomata.

Area barbata Linne.

Area barbata Linne, Gmelin, Syst. Nat., p. 3306, 1788.

Young specimens of this well known form were dredged near Barbados, at
Stations 290 and 292, in 56-73 fms.; and by Sigsbee off Havana in 127 fms.
(dead valves). It is a shaUow-water species and probably does not live in
more than 100 fms.

Genus MACRODON Ltcett.

Macrodon asperula Dall.

Macrodon asperula Dall, Bull. M. C. Z., IX. pp. 20, 1881.

Plate VIII. Figs. 4, 4 a.

Habitat. Station 33, 1568 fms., in the Gulf of Mexico, bottom temperature
40°. 5 F.; Station 19, 310 fms., off Cuba ; Cape San Antonio, 1002 fms., this
specimen too young to be certain of the identification.

MUSEUM OF COMPARATIVE ZOOLOGY. 245

Area (Bo/rhatia) pteroessa E. A. Smith seems very similar externally to our
shell, but the hinge is diflferent and the shell more produced behind ; the
manner in. which the black ligament is placed would seem to be similar in
both.

An allied species with an outline almost precisely similar to A. culebrensis
Smith (Chall. Rep., pi. xvii. fig. 9 a) was obtained (a single valve) N. W. of
the N. W. end of Cuba in 80 fms. by the U. S. steamer " Albatross " in 1885.
It has a hinge much like that of 31. asperula, but its external surface is
entirely different; there are numerous concentric grooves, with wider inter-
spaces covered everywhere with an oblique shagreened ornamentation ; beside
this there are obsolete radiating series of minute scales, probably stronger in
some specimens, and on and behind the ridge from the umbo to the posterior
angle of the margin are four well defined and two or three obsolete nodulous
radii. The valve is about six millimeters long and quite inflated ; the um-
bones must nearly touch in perfect specimens, as the area is extremely narrow
and the beak well developed. It may take the name of M. sagrinata.

Professor Yerrill's Area profundicola, though not very characteristically
figured, is, from a typical specimen, more finely striated, the lower posterior
region less patulous and its hinge margin not so high. The front teeth are
more, and the hind teeth less, oblique than in Macrodon. It may be observed
that the gap between Macrodon and certain forms of Barbatia is not very wide.

Family NUCULID^.
Genus NUCULA Lamarck.

Nucula Lamarck, Prodrome d'une Nouv. cl. des Coquilles, p. 87, no. 104, 1799.

Type Area nucleus, L.
NxKulana Link, Beschr. Rost. Samml., p. 155, 1807.

I take this opportunity of mentioning, for students who cannot get access to
the rare work of Link, that his Nuculana is an exact synonym of Nucula
Lamarck, and was intended merely as a modification of that word; while the
diagnosis, " shell smooth, closed all round," will not apply to the group sepa-
rated by Schumacher, afterward, under the name of Leda. That the only
species of the group in the collection was N. rostrata was merely an accident,
and it was evidently not intended as a type, for it does not agree with his
diagnosis.

246 BULLETIN OF THE

A. With smooth margin.

Nucula segeensis (Forbes) Jeffreys.

Nucula tenuis Montagu, Dall, Bull. M. C. Z., IX. p. 123.
Niicula (Bgeensis Jeffreys, P. Z. S., 1879, p. 681.

Habitat. Sigsbee, off Havana, 175 and 450 fnis. ; off Morro Light, 292
fms.; Station 20, 220 fms.; Station 3, 450 fnis.; and Station 230, near St.
Vincent, in 464 fms. ; all mostly dead valves.

In examining specimens dredged oflf the Carolinas by the Fish Comnussion,
I was struck by the fact that only N. proximo, and not a single N. tenuis had
been secured. This led me to review the specimens identified for me by Dr.
Jeflfreys as N. tenuis, and so named in the preliminary report, as above. I
have compared them with all the varieties of tenuis, and with all the specimens
of (Bgeensis in the Jeffreys collection. The Blake specimens, nearly all being
separated valves, agree in form and general appearance with the flatter forms
of tenuis, the only external differences -being that the former are a little more
pointed and pinched dorsally behind, and that two moderately distinct ridges
enclose a very narrow elongated area along the upper posterior margin.
Symptoms of such an area were visible occasionally in individuals of genuine
tenuis, but not so clearly defined. Inside, the hinge line of tenuis is narrower,
the teeth more delicate and perhaps fewer, the cartilage pit a little smaller.
These are just the differences which separate tenuis from N. cegeensis, and it is
probable that the Blake specimens should be referred to the latter species. It
is by no means clear to me that cegeensis is anything more than a geographical
race of tenuis; but so far, though the hinge characters are slight, I have not
found many intermediate specimens. The West Indian specimens are larger
than those from the Mediterranean, and consequently the number of teeth is
greater, but the proportion is about the same in all. The largest specimen
measures 10.7 max. Ion., 8.0 max. alt., and 4.7 mm. max, diam., with 8 anterior
and 15 posterior teeth.

Nucula C3miella, n. s.

Shell small, white, rather thick, rounded, triangular, moderately convex;
beaks somewhat anterior, not prominent; exterior sculptured with evenly
distributed concentric waves or narrow lirae separated by wider interspaces;
an obscure flexuosity in front of the beaks indicates, without sharply defin-
ing, a lunule over which the concentric sculpture passes; a faint ridge ex-
tends backward from the beaks half as far as the teeth, but becomes obsolete
without enclosing an area ; inside, the shell is brightly pearly, the margin
plain, with seven anterior and eight or nine posterior rather stout teeth, sepa-
rated by a good-sized cartilage pit, directed vertically downward from the

MUSEUM OF COMPARATIVE ZOOLOGY. 247

beaks, not, as usual, oblique ; both ends rounded, the anterior rather the more
angular. Max. Ion. 5.1; max. alt. 4.0; max. diameter about 3.9 mm.

Habitat. Yucatan Strait, in 540 fms., two right valves, one of which was a
little more triangular than the other.

This little shell, which probably grows to a larger size, resembles a little
Corbicula or Astarte in its concentric, without any radiating sculpture. I have
not been able to find anything to which it might be referred, and, though the
material is scanty, have concluded to give it a name.

B. Margin crenulated.
Nucula crenulata A. Adams.

Nucula crenulata A. Adams, Dall, Bull. M. C. Z., IX. p. 123, 1881.
Nucula culebrensis E. A. Smith, Chall. Rep. Lam., 1. c, p. 228, pL xviii. figs. 11, 11 a,
1885.

Plate VII. Fig. 2.

Habitat. 20 miles west of the Florida coast, in 30 fms.; Station 36, 84
fms.; Barbados, 100 fms.; Sigsbee, off Havana, 158, 182, and 450 fms.; Station
20, 220 fms.; Station 19, 310 fms.; Yucatan Strait, 640 fms.

Nucula crenxolata A. Adams, var. obliterata Dall.
Nucula crenulata, var. obliterata Dall, Bull. M. C. Z., IX. p. 123, 1881.

Plate vni. Fig. 2.

Station 44, 539 fms. ; Yucatan Strait, 640 fms. ; Station 2, 805 fms. ; Station
226, 424 fms., near St. Vincent; Station 236, near Bequia, in 1591 fms.; and
at Station 262, near Grenada, in 92 fms. ; bottom temperatures ranging from
39°. 0 to 62°. 0 F.

This species is very variable, and presents sometimes an almost smooth sur-
face (as in the var. obliterata), and at others either a series of regular concen-
tric waves or a more or less broken and irregular concentric sculpture, the
whole being imited by every variety of transitional features. The most nearly
allied species is N. sulcata Bronn (not A. Adams), which is however less
trigonal, grows much larger, and yet has a finer and more irregular sculpture,
in which the concentric element is less dominant. The N. culebrensis of Smith
agrees so well with young, regularly sculptured specimens of crenulata, that,
taking the locality into consideration, I feel quite confident of their practical
identity. The ordinary adult and many young crenulata are more coarsely
and roughly sculptured, but this is not invariable, and the large number of
specimens I have examined have given an excellent opportunity for com-
parison.

248 BULLETIN OF THE

The var. obliterata is as a rule more trigonal and more compressed than the
typical form. In some specimens the beaks are very prominent vertically.
Its faint sculpture will always enable it to be distinguished from jV. Verrillii
(^N. trigone Verrill, Trans. Conn. Acad., VI. p. 438, 1885, not of Bronn or
Seguenza, 1877), which has a smooth margin, while the smoothest obliterata
always show minute crenulations. Extreme specimens of the type and vari-
ety would be taken by any one as distinct species without the connecting
series. N. cancellata JeflFreys is more globose, smaller, and more delicately
sculptured.

Family LEDID^

Genus LEDA Schumacher.

subgends yoldia morch.

Yoldia solenoides Dall.

Yoldia solenoides Dall, Bull. M. C. Z., IX. p. 127, 1881.

Plate IX. Fisrs. 2, 2 a.

Habitat. Station 49, 118 fms. Lat. 28° 61'.5 and W. Lon. 89° 1'.5, in the
Gulf of Mexico, no temperature noted.
No additional specimens have turned up.

Yoldia liorhina Dall.
Yoldia liorhina Dall, Bull. M. C. Z., IX. p. 127, 1881.
Plate IX. Figs. 1,1a.

Habitat. Sigsbee, off Havana, 182 fms. ; Station 23, 190 fms,; Station 33,
1568 fms.

The cartilage is large and black, and inserted on a wide triangular space
directly below the beaks, but in the dead valve from Station 33 the place of
the cartilage is very small, though the shell is otherwise identical with the
others. The only living specimen, from 182 fms., shows no external liga-
ment, but the dead valve referred to might almost be taken for a Solenella or
Malletia.

MUSEUM OF COMPARATIVE ZOOLOGY. 249

Subgenus LEDA Schumacher (s. s.).
Leda Carpenteri Dall.

Leda Carpenteri Dall. Bull. M. C. Z., IX. p. 125, 1881.
Plate \1TI.. rig. 11 ; Plate IX. Fig. 3.

Habitat. Barbados, 100 fms. ; Station 5, 229 fms. ; Station 9, 111 fms., bot-
tom temperature 55°.0 F. ; Station 21, 287 fms.; Station 128, 180 fms., off
Frederikstadt ; off the Carolina coast, U. S. Fish Commission, 1885.

Since describing this species I have been able to compare it with specimens
dredged by the Fish Commission in some abundance farther north, and with
Leda clavata Calcara, a Sicilian fossil which is its nearest relative. L. Carpen-
teri differs from clavata in its greater smoothness and in having the hinge-line
narrower, the teeth smaller, more delicate, and less numerous, especially the
anterior series ; the ligamental pit is much smaller, and the series of teeth
are straighter and with much less margin between them and the edge of the
dorsal crest. The raised line inside the rostrum is in clavata nearly in the
middle of the shell ; in Carpenteri it is invariably nearer the dorsal edge, thus
making the dorsal channel distinctly narrower than the ventral one.

In fresh specimens of L. Carpenteri, especially youngish shells, the pale green
epidermis is marked by a beautiful radiating series of arched striae, only visi-
ble with a glass except in very marked cases, or near the ventral edges of the
valves where the striation is strongest. It is absent in decorticated specimens,
and so would appear to be purely a character of the epidermis.

Leda clavata has been erroneously united with L. cuspidata, which differs
both in shape and sculpture. I have not seen any recent specimens of
clavata or cuspidata. Some marked as such in the Jeffreys collection were
L. Carpenteri.

Leda messanensis Seguenza.
Leda messanensis Seguenza, Dall, Bull. M. C. Z., IX. p. 124, 1881.

Habitat. Station 19, 310 fms. ; Station 20, 220 fms. ; Barbados, 100 fms.;
Sigsbee, off Havana, 450 fms.

This species, which I have compared with specimens received from the
author, varies in sculpture much like the others, being sometimes almost
wholly smooth, and at others with well developed concentric sculpture; it also
varies remarkably in proportional length, some specimens being rery short
and high. In considering these variations, one cannot help surmising that the
present number of nominal species of these little shells will eventually require
to be diminished.

250 BULLETIN OF THE

Leda solidula E. A. Smith.

Leda solidula E. A. Smith, Chall. Rep. Lam., p. 233, pi. xix. figs. 6, 6 a, 1886.

One valve was found from 1002 fms., near Cape San Antonio; another from
640 fms., near by, in Yucatan Strait ; both were inadvertently included among
the varieties of L. messanensis at the time the preliminary examination was
made. The type was dredged by the Challenger expedition at Station 120,
off Pernambuco, in 675 fms., red mud.

Leda vitrea D'Orbignt, var. cerata Dall.
Leda vitrea (?) U'Orb. 1846, var. cerata Dall, Bull. M. C. Z., IX. p. 126, 1881.

Plate Till. Fig8. 12. 12 a.

Habitat. Barbados, 100 fms. ; Sigsbee, off Havana, 450 fms. ; Station 206,
near Martinique, in 170 fms.

Among the species of Ledida from our southern coast, recent and fossil, are
several closely allied to each other and to foreign forms, which have been in
a state of more or less confusion. These are as follows, in order of publication.

Leda (Nuaila) concentrica Say, Journ. Acad. Nat. Sci. Phil., IV. 141, pi. x. fig. 6,

1824.
Leda (Nucula) acuta (Say ?) Conrad, Am. Mar. Conch., p. 32, pi. vi. fig. 3, 1831 ; Tert.

Fo8., p. 67, pi. XXX. fig. 2, 1845.
Leda cuneata Sowerby, P. Z. S., 1832, p. 198; Thes., p. 128, fig. 92.
Leda commutata Philippi, Zeitschr. Mai., p. 101, Jan. 1844.
Leda vitrea D'Orbigny, Moll. Cuba, II 262, pi. xxvi. figs. 27-29, 1846.
Leda jamaicensis D'Orbigny, 1. c, p. 263, pi. xxvi. figs. 30-32, 1846 (= acuta + cuneata).
Leda (Nucula) eborea Conrad, Proc. Acad. Nat. Sci. Phil., III. p. 24, pi. i. fig. 4, 1848

(= concentrica Say).
Leda unca Gould, Proc. Post. Soc. Nat. Hist., VIII. p. 282, 1862.
Leda Bushiana Verrili, Trans. Conn. Acad., VI. 229, 1884.
Leda unca Verrili, 1. c, p. 260, 1884.

Leda concentrica Say, described as a fossil, is without doubt the same as the
recent elorea Conrad, which I have from Conrad's original collection. It is
distinguished by its strong sculpture and long straight rostrum. It ranges
from Florida to Texas.

Leda acuta was poorly described, and very badly figured. I have not been
able to compare with the figure in the Am. Marine Conchology, but his figure
in the " Fossils of the Tertiary Formation " is much more slender and recurved
than the species which American writers have regarded it as intended to
represent. This may probably be the fault of the figure, and it will save a
good deal of trouble, and give us a clear way out of the confusion, to adopt

MUSEUM OF COMPARATIVE ZOOLOGY. 251

Conrad's name as it has been traditionally applied. I find the next name in
order, to be Leda cuneata of Sowerby, which from specimens identified by
Hanley (and coming from Catalina Island, California) is quite evidently the
same as L. jamaicensis D'Orbigny. The latter author described and figured a
young specimen, so that the magnified figure he gives agrees only with speci-
mens of the same age; but, for them, it is very exact. I have examined a
large number of L. commutata Phil., and, while it is very similar, I cannot
convince myself that it is the same. The commutata generally has one very
strong anterior rib, and the acuta has a shallow groove bordered by two faint
ribs. This is the most obvious character, though there are others. The L. unca
of Gould was not figured and the description is brief. It is described as hav-
ing the dorsal area keeled and smooth, characters not appropriate to any of
the just mentioned forms, though shared by the proportionally more elongate
L. Bushiana Verrill, which is not " acutely rostrate." The vitrca, acuta,
and second unca of Verrill all have the dorsal area strongly sculptured, even
when worn; more so, generally, than the rest of the shell. None of these
therefore should be identified with unca Gld. Verrill's second unca (1. c.,p. 260),
which seems distinct from either vitrca or acuta, may take the name of Ver-
rilliana. The variety cerata is united with the typical vitrea by intermediate
forms.

Leda acuta Conrad.

Nucula acuta Conrad, Am. Mar. Conch., p. 32, pi. vi. fig. 3, 1831.
Leda jamaicensis D'Orbigny (1846), Dall, Bull. M. C. Z., IX. p. 124, 1881.
Leda cuneata Sowerby, P. Z. S., 1832, p. 198.
Leda inornata A. Adams, fide Hanley, from type.

Leda unca Verrill, Trans. Conn. Acad., V. p. 572, 1882, pi. Iviii. fig. 41 (not VI.
p. 260.)

Plate VII. Figs. 3 a. 3 b, 8.

Habitat. Sand Key, 80 fras. ; off Sombrero, 54 and 72 fms. ; Jamaica, Santo
Domingo (D'Orb.); off southern New England, 85-165 fms. (Verrill). Florida
(Hemphill).

The relations of this species to the others have been considered under the
preceding species. I have not been able to consult Conrad's original publica-
tion, but Binney (Bibl. N. Am. Conch.), citing from it, refers the species to Say.
In other places Conrad puts his own name after it.

The specimens from Yucatan Strait cited in the preliminary report under
this species, on further study, appear to be L. messanensis Seg. L. commutata
Phil., as before mentioned, appears to be diff"erent from this, though a closely
allied form. The frayilis of Cliemnitz, a badly figured and described shell, to
which Dr. Jefi'reys would refer L. commutata, is much larger than any known
commxdata, and is referred by Hanley to a Chinese species. Doubtless Chem-
nitz would have included commutata in his species. The Lembulus deltoideus

252 BULLETIN OF THE

of Risso, briefly described and supposed to be this species, is better figured, and,
if the identification could be confirmed, is the oldest stable name for L. com-
mutata, although the latter had been referred to L. minuta of Miiller in 1792.
Risso's figure and description, however, are hardly evidence enough taken with-
out corroborative information.

The L. acuta is abundant oflf the Carolina coast at moderate depths.

Leda solidifacta Dall.

Leda solida Dall, Bull. M. C. Z., IX. p. 126, 1881. (Nom. preoc. ?)
Plate VII. Fig8. 7 a, 7 b.

Habitat. Station 21, 287 fms.

This species is nearest L. concava Bronn, but is less rostrate, and has the
anterior side proportionally longer. The name solida is said to be preoccupied
in this group, though I have not been able to lay my finger on the place. If
this be scf, the specific name may be modified to solidifacta. No additional
specimens have been found.

Leda subaequilatera Jeffreys.

Leda subcequilatera Jeffreys, P. Z. S., 1879, p. 679, pi. Ivi. fig. 3.

A single valve of this small and rare species was dredged at Station 253,
near Grenada, in 92 fms. It agreed very well with Dr. Jefi'reys' types, with
which it has been compared.

Leda hebes E. A. Smith.

Leda intermedia Sars, Dall, Bull. M. C. Z., IX. p. 127, 1881.

Leda hebes E. A. Smith, Chall. Rep. Lam., p. 234, pi. xix. fig. 7, 1885.

Habitat. Station 2, 805 fms.

The opportunity of comparing the valves referred in my preliminary report
to L. intermedia, with authentic specimens of the latter, has shown that, though
similar, they are not identical. It would seem likely that they are adult speci-
mens of what Mr. Smith has described as L. hebes, from the same region. The
adults are more convex laterally and below, and somewhat more rostrated than
the young as figured by Smith. Than L. intermedia they are less inflated,
less rounded behind, less polished, and have more hinge-teeth, especially be-
fore the ligament pit. The striation confined to the middle part and basal
margin of the valves, and very distinct there, forms its most remarkable
characteristic.

MUSEUM OF COMPAKATIVE ZOOLOGY. 253

Section SATURNIA Seguenza.

Leda (Saturnia) pusio Philippi.

Nucula pusio Phil., Moll. Sic, II. p. 47, pi. xv. fig. 5, 1844.
Leda pusio Jeffreys, P. 2. S., 1879, p. 578.

Two specimens exactly agreeing with Jeffreys and Seguenza's specimens
were dredged dead at Station 236, near Bequia, in 1591 fnis.

This species has a gap in the tooth line, but no internal ligament. There is
a pit under the beaks, exterior to the line of the teeth, which may have had a
ligament in it. Mediterranean specimens show the same. Seguenza places it
in his section Saturnia as type. There is a gradual change from shells with
an internal cartilage set in a spoon-shaped process, and an outside ligament, to
those where the two seem to have come together, coalesced, and finally become
entirely external. It does not seem possible to draw hard and fast lines.
Yoldia and Malletia, Leda and Tindaria, approach each other by insensible, or
rather undefinable degrees. The extremes of the series are very distinct, the
passage from one to the other very gradual. I do not regard any of the divis-
ions of Leda as more than sectional ; at least, until more is known about the
soft parts, I prefer to regard them so.

Others may be able to decide definitely what constitutes a genus, a subgenus,
or a section, and measure all these groups by that standard. I find myself
unable to do more than point out relative values, as they appear to me, in a
single series, and even in this I find it often difficult to satisfy myself that the
correct proportion between them has been attained.

Leda (Saturnia) quadrangularis Dall.

ILeda (Jeffreysi Hidalgo, var. ?) qnadrannularis Dall, Bull. M. C. Z., IX. p. 124, 1881.

Plate VIII. Fig. 6.

Habitat. Station 33, 1568 fms.

This turns out on comparison with specimens to be entirely unlike L. Jef-
frey.ii, and I have not found anything like it. It is nearest pusio, and has the
same concentrically finely undulate surface, but the basal pout and longest
slope of hinge-line are posterior here, while, in pusio, both are anterior. I
have not been able to satisfy myself that there was any ligament pit inside.
There is a smooth interval between the two sets of teeth, but no pit, and no
evidence that any ligament was attached there. It would belong to Seguenza's
section Saturnia. The valve is 4.6 mm. in length, 4.0 mm. high, and the pair
were probably 3.0 mm. in diameter.

254 BULLETIN OF THE

Section NEILONELLA Dall.
Neilonella Dall, Bull. M. C. Z., IX. p. 125, 1881.

Shell like Tindaria Bellardi, with a single ligament directly between the
beaks, chiefly external, but its base dividing the hinge-line, while its upper
surface extends about equally before and behind the beaks. Type Leda
{Neilonella) corpulenta Dall.

This section is almost exactly intermediate between Leda, with an inner and
outer ligament, and Tindaria, with a purely external one.

Leda (Neilonella) corpulenta Dall.

Leda (Neilonella) corpulenta Dall, Bull. M. C. Z., IX. p. 125, 1881.

Plate VII. Figs, la, lb.

Habitat. Station 23, 190 fms., living, bottom temperature 64°.0 F.; Station
21, 287 fms. ; Station 47, 331 fms.; Sigsbee, off Havana, 450 fms.

No additional specimens of this interesting form have turned up in the col-
lection ; it probably lived at all the stations mentioned, though valves only
were obtained except at Station 23. There is nothingr like it, recer* or fossil,
in the Jeffreys collection.

Genus MALLETIA Desmoulins.

Section TINDARIA Bellardi.

Malletia (Tindaria) cytherea Dall.

Nucida (Tindaria?) cytherea Dall, Bull. M. C. Z., IX. p. 123, 1881.

Malletia veneriformis E. A. Smith, Chall. Rep., p. 246, pi. xx. figs. 9, 9 a, 1886.

Plate VIII. Figs. 1, 1 a.

Habitat. Off Cape San Antonio, 413-424 fms.; Yucatan Strait, 640 fms.;
Station 226, near St. Vincent, in 424 fms. ; and Station 2392 of the U. S. Fish
Commission in the Gulf of Mexico, Lat. 28° 45', Lon. 87° 30' W., in 724 fms.,
mud, living, bottom temperature 40°.7 F.

The original specimens from ■which this species was described were single
valves, subiridescent with decay. The supposed minute pit proves pathologi-
cal. The reception of two magnificent specimens from the Fish Commission
dredgings enables me to correct my erroneous reference of the species to
Nucula, which I regret the more since it may have led my friend Mr. Smith
into a redescription of the species.

MUSEUM OF COMPARATIVE ZOOLOGY. 255

The shell when living is of a brilliant white, covered with a fine smooth but
not polished straw-colored epidermis. The ligament is wholly external, deli-
cate, and nearly hidden in a groove just behind the beaks. There are twelve
anterior and twenty-eight posterior teeth, which dwindle to a spot just under
the beaks, below which is a little flat or subconcave space very like a shelf for
a cartilage, which, however, does not exist. The measurements of the fully
adult form are, max. Ion. 15.0 ; max. alt. 11.2 ; max. diam. 9.0 mm. There
is a polished space in front of the beaks where the concentric waves fade out,
faintly margined by an obsolete radius or two, but not otherwise differentiated ;
and immediately in front of and close to the beaks is a very small rounded area,
over which the epidermis is of a darker color than elsewhere, but apparently
not marked by sculpture. The pallial line is entirely simple, and the interior
of the shell brilliantly polished, with a tendency to iridescence, though not
pearly.

Malletia (Tindaria) Smithii Dall.

Malletia cuneata E. A. Smith, Chall. Rep. Lam., p. 247, pi. xx. figs. 10, 10 a, 1885.
Not M. cuneata Jeffreys (1876), P. Z. S., 1879, p. 586, pi. xlvi. fig. 10.

A dead valve of this species was dredged by Sigsbee in 450 fms., off Havana.
The Challenger specimens were taken in 390 fms., off Culebra Island. A
specimen was dredged by the U. S. Fish Commission at Station 2119, near
Grenada, in 1140 fms., bottom temperature 39°.5 F. This measured 7.75 mm.
in length, and had nine anterior and twenty-two posterior teeth, counting all
the small ones.

As my friend, Mr. E. A. Smith, in his valuable report on the Challenger
Lamellibranchs, has overlooked the prior use of his specific name by Jeffreys, it
gives me much pleasure to propose the name of Smithii for this very elegant
little shell.

Section NEILO Adams.
Malletia (Neilo?) dilatata Philippi.

Leda dilatata Philippi, Dall, Bull. M. C. Z., IX. p. 125, 1881.
Neilo dilatata Seguenza, Nucul. Terz., 1877, p. 1184.

Habitat. Off Morro Light in 292 fms., two right valves.

This agrees exactly with the Italian fos.sils. There is no cartilage pit, but a
wide subtriangular gap in the line of teeth, and a groove for an external liga-
ment. I cannot see that the hinge without the soft parts offers decisive evi-
dence of the place to which this species should be referred. It is probably a
Malletia, and belongs in the vicinity of M. arrouana Smith, in which the gap
in the line of teeth would seem to have become closed.

256 BULLETIN OF THE

Family CARDITIDiE.

GeNCS CARDITA BRUGltEE.

Cardita domingensis D'Orbignt.
C. Dominguensis D'Orbigny, Moll. Cuba, IL p. 291, pi. xxvii. figs. 27-29, 1845.

Habitat. Station 12, in 36 fnis. off Cuba ; off Sombrero, in 54 I'ms. Ex-
tends northward to the Carolina coast.

D'Orbigny's figure is of a very young shell ; adult specimens are twice as
large and have more ribs.

Family CRASSATELLIDiE.

Genus CRASSATELLA Lamabck.

Crassatella floridana Dall.

Crassatelia antillarum (?) Reeve, ysLT.Jiondana, Dall, Bull. M. C. Z., IX. p. 131, 1881.

Plate VI. Fig. 12.

Habitat. Gulf of Mexico, west of the Florida coast, 30 fms.

The single young specimen obtained as above, and represented by the figure
(11.0 X 6.75 mm.), is the only one in the Blake collection. The U. S.
Fish Commission have since dredged off the southeastern coast of the United
States and in the Gulf of Mexico a considerable number of adult valves of the
same species, the description of which I am thus enabled to complete. The
largest of these valves measured 78.0 mm. in length and 57.0 mm. in height,
the complete shell must have had a diameter of 31.0 mm. When fresh it is
covered with a fine bright brown epidermis, which becomes fibrous after death
and maceration, or in very aged specimens ; the whole shell in front of the
anterior rostral carina is covered with rather even concentric grooves, about
1.0 mm. wide. The figure gives a good idea of the somewhat flattened tip
of the beaks ; the anterior and posterior areas are depressed, smooth, narrow,
and subequal ; the anterior is larger in the left, and the posterior in the right
valve; the grooves do not continue behind the flexuosity which marks off the
rostrum, tUe area between that and the dorsal area or corselet is merely con-
centrically striated ; the interior is pinkish chocolate, pink, or white, darker
behind; the muscular scars are rounded, strong, but rather small; the pedal
scar is close behind the upper corner of the anterior adductor, and is strongly
marked.

MUSEUM OF COMPARATIVE ZOOLOGY. 257

When 1 first received these valves I supposed that they would turn out to
be identical with some one of Conrad's Tertiary species; but after comparing
with them all, I found that none of them agreed sufficiently well with the re-
cent species to render it desirable to refer it to either of them. The nearest
of the fossil forms to the C. fioridana is the C. undulata Conrad (not Sowerby),
of the variety figured by him on Plate XI. of his Fossils of the Tertiary For-
mations of the United States, which (though dated 1838 on the title-page), ex-
cepting the first few pages, was not issued until 1845. From this C. fioridana
differs in being more pointed anteriorly and less so behind; in having flatter
and less pointed beaks ; in having a more pronounced flexure below the ros-
trum, and the latter proportionately shorter, higher, and more ridged above ;
the cardinal teeth are more oblique, and the anterior lateral does not run up
in front of the cardinals, but ceases near their lower extremity. I find these
differences to hold good through a large series, and consequently conclude that
the recent species is distinct. It is entirely different from the C. antillarum,
until now the only recent species of Crassatella proper known to inhabit the
Antilles.

The margins of C. fioridana are smooth at all stages, but the outside grooving
in aged specimens becomes obsolete near the margin.

Subgenus ERIPHTLA Gabb.

Eriphyla Gabb. Pal. Cal., I. p. 180, 1864 ; Stoliczka, Pal. Indica, III. p. 156, 1871
(but not pp. 181, 182, pi. \i.; = Dozia Bosquet, 1868). Type E. umbonata
Gabb.

Eriphylopsis Meek, Inv. Pal. Upper Missouri, p. 125, 1876. Type E. gregaria Meek
and Hayden.

The genus Eriphyla of Gabb was poorly figured, and hastily, or at least
imperfectly, described by its author, for whom, however, allowance should be
made on account of his isolated position in California, far from well-equipped
museums or libraries. Meek, who was one of the most careful and exact
paleontologists, examined into the subject, and found that there could be little
doubt that the differences between the type of Eriphyla and the small Crassa-
telloids formerly included under Gouldia, and best known by that name (and
for his purposes best typified by C. mactracea Linsley), were essentially these.
The teeth appeared to be reversed as regards the valves, and there was a little
furrow behind the beaks which by Gabb and himself was supposed to indicate
the presence of an external ligament, the internal cartilage when absent, as
in dead valves or fossils, leaving no evidence of its existence. In 1871 Sto-
liczka complicated the problem by referring to Dozia lenticularis Goldfuss as
the type of Eriphyla ; and by describing that group from the peculiarities
of the aforesaid Dozia (which probably belongs near Dosinia). This error
has been copied from Stoliczka into Tryon's Structural and Systematic Con-

VOL. XII. — NO. 6. 17

258 BULLETIN OF THE

chology, Vol. III. p. 226, under Eriphyla, and of course gives an entirely wrong
idea of Eriphyla, which has no pallial sinus, or at least none has ever been
sho\^^l ; and in the E. gregaria there is a perfectly simple pallial line, as in the
recent species I have referred to.

Now it is well known that in Astarte it occasionally happens that the teeth
may be reversed with regard to the valves. In the allied Eriphyla it appears
to be a common occurrence. I find the Antillean shells presenting absolutely
the 8ame arrangement of teeth as the E. gregaria or E. umbonata. E. mac-
tracea, however, seems to have the teeth the other way generally ; but not
invariably, if I have correctly identified some valves from the Florida coast.
A little groove behind the beaks is often there, too, but it does not carry any
external ligament, and as the existence of an external ligament was based
merely on the presence of this feature (which varies more or less between differ-
ent specimens), it is evident that there is no warrant for claiming an external
ligament for Eriphyla any longer. Meek, both in his publications and in con-
versation, was confident of the identity of Eriphyla with the so-called Gouldia,
if it could be shown that the teeth in the latter were reversible ; but at that
time, just before his death, we had but a few specimens of the recent forms
which did not seem conclusive, as they were all of the C. mactracea. So, in
his last revision of his Paleontology, he suggested that, if the Californian and
Missouri fossils did not agree, the latter might take the name of Eriphylopsis.
The recent Antillean forms, as I have said, agree perfectly with Eriphylopsis,
and there is every reason to think that they agree with the original Eriphyla;
which, until a difference is definitely shown by nmewed observations, I prefer
to retain. Should any differences be found, the recent forms would follow the
Missouri fossil and be included in the subgenus Eriphylopsis.

That these little shells present a recognizable facies sufficient to enable one
to decide instantly whether any one of the species is an Eriphyla or a typical
Crassatella is, I think, undeniable. Whether this facies — of which the im-
portant features are the small size, triangular form, inequality of the valves,
absence of rostration, and the angulated posterior extremity — is sufficient to
entitle the group to a name, I am quite willing to leave to others to decide for
themselves. It seems to me they are, and that the distinctions are just as clear
between Eriphxjla and say Crassatella nana, as if one of the larger Crassatellas
had been chosen.

The fact of the inequality of the valves has been questioned, but I have
never seen a perfect pair in which, looking forward over the beaks, the right
valve did not advance above the other ; the contrary being the case in looking
the other way, though not so well marked. In convexity they are about equal.
This is also true, but much less perceptible, in Crassatella proper.

I have gone into the matter at this length because, it seems, I was insuffi-
ciently detailed in my previous statement ; not making myself fully under-
stood by some, who were unfamiliar with the errors of Gabb and Stoliczka.

MUSEUM OF COMPARATIVE ZOOLOGY. 259

Crassatella (Eriphyla) parva C. B. Adams.

Crassatella (Eriphyla) parva (C. B. Adams, 1845), Dall, Bull. M. C. Z., IX. p. 131,

1881.
C, Martinicensis D'Orbigny + C. guadalupensis, D'Orbigny, 1846.

Habitat. Martinique, Jamaica, St. Domingo, Cuba, St. Thomas (D'Orb.) ;
Cuba (Pfr.) ; Jamaica (Adams) ; Station 21, 287 fms. (Blake exp.).

After the examination of a great many specimens from all paits of the An-
tilles, I am driven to the conclusion that both of D'Orbigny's species are
identical with the present one, the distinctions being entirely within the range
of its variation. Krebs, an excellent observer, came to the conclusion, a good
many years ago, that the two species of the Mollusq^ues de Cuba were the
same.

Family ASTARTID^.

GeNCS ASTARTE J. SOWERBT.

There are several species referable to this genus in the Gulf of Mexico and
adjacent waters, mostly quite small, and having a tendency to coloration in the
inside of the valves. The viviparous subgenus Parastarte is also indigenous
to the shallower waters of this latitude. It too is brightly colored, and has a
vernicose epidermis.

Two species or forms were obtained by the " Blake," one abundantly and at
various depths; the other, in but one haul, and only one or two specimens. Of
the latter, however, the U. S. Fish Commission has obtained valves at numer-
ous stations, and not any of the other species ; so curiously checkered is the
luck of the dredger. In connection with the identification of the species I
have carefully examined the large series in the Jeffreys collection, and have
had the advantage of the criticism of Mr. E. A. Smith, who recently mono-
graphed this difficult group. I have decided to give names to these Gulf
forms, not because I am certain that they represent permanent immutable
entities, if such things exist, but because they differ in a diagnosable way from
anything I can find named. The most hardened believer in the immutability
of species, after an encounter with a large collection of Astartes, would prob-
ably be content with permission to retire in good order from the field, with
bag and baggage, without any request that drums or fifes should announce his
movements to the rest of mankind.

Astarte Smithii, n. s.

Plate \"II. Fiers. 5 a, 6 b.

Shell small, belonging to the group of A. sulcata, having a squarish globose
form, crenated margin, and pale brownish epidermis. The exterior is concen-

260 BULLETIN OF THE

trically sculptured with (in the adult) usually 15-20 ribs, rather narrower than
their interspaces, and generally with, toward the middle of the shell, a dupli-
cated appearance, caused by a faint wave immediately above the main one ;
the ribs in all cases fail about the beginning of the last third of the shell,
which portion is merely striated or even smooth ; in some specimens the
whole surface is nearly smooth, or has about double the usual number of faint
subequal close-set ribs over the anterior two-thirds ; in these cases it some-
times happens that the fine ribbing will extend over the greater part of the
area usually smooth, but, after comparing all the specimens, I am unable to
regard these difl'erences as more than varietal ; the lunule is lanceolate, some-
times subcordate, smooth, somewhat depressed and bounded rather by the
change in the sculpture than jy any line of demarcation ; the ligament is
short, immediately behind the beaks ; the posterior area is elongated, bounded
by two faint ridges, from which the surface slopes to the hinge-margin ; the
interior is smooth, with the muscular scars small and situated rather close to
the margin ; the crenulations of the edge are rounded, minute and close-set;
they are noticeable at all ages ; the right valve has one strong cardinal tooth
with a pit on each side of it, the anterior hinge margin slightly grooved, the
posterior sharp-edged ; the left valve has two strong teeth and the anterior
margin sharp, while there is a long groove in the posterior margin to receive
the edge of the right valve. Lon. of shell 7.0 ; alt. 6.0 ; diam. 4.0 mm.

Habitat, Oflf Sombrero, 64 fms. ; Station 36, 84 fms. ; Sigsbee, off Havana,
100-450 fms.; Barbados, 100 fms.; Station 5, 229 fms.; Station 44, 539 fms.;
Station 274, Barbados, in 209 fms., sand, bottom temperature 53°.5 F.; Station
206, near Martinique, in 170 fms., bottom temperature 49°.0 ; Station 132,
near Santa Cruz, in 115 fms., bottom temperature 65°.0 ; Station 33, in 1568
fms., one valve, perhaps drifted. Dredged in 200 fms., on Campeche Bank, by
Dr. W. H. Rush, U. S. N.

The strongly sculptured form, which may be taken as the type of the
species, has a shorter and more cordate lunule, a much more sunken and
sharply defined dorsal area, and a shorter ligament, than the variety with less
pronounced sculpture, which may take the name of Astarte Smithii, var. glo-
bula. The two varieties occur indifferently together, the type, however, being
much the more numerous. I need hardly add that the specific name is given
in honor of Mr. Edgar A. Smith, of the British Museum, who has monographed
this genus, and to whom I am indebted for many useful criticisms and kindly
furnished bits of information.

The figure, drawn before the specimens had been finally studied, does not show
the apparent duplication of the riblets in the middle part as well as many of the
specimens do, but it is a fair representation of the one from which it was made.

A small species of Gouldia, which I took to be Venus cubaniana D'Orb., being
mixed with these Astartes, they were hastily taken to be all one species, caus-
ing some confusion of localities in the preliminary report. This species is
related to Astarte lens Stm., which is referred to by Jeffreys as a variety of

MUSEUM OF COMPARATIVE ZOOLOGY. 261

.4. crenata. But the lens, or crenata, of the same size as A. Smithii, is longer,
aiuch flatter, and usually not crenate ; the waves or ribs are of a diff'erent
form, and the color is a more ruddy brown.

Astarte nana, n. s.?

1 Astarte nana Jeffreys, Smith, Obs. on the Genus Astarte, Leeds Journal of Con-
chology, p. 213, 1881. (Gulf of Florida, 60 fms., Pourtales.)

Plate VII. Figs. 6 a, 6 b.

I have not been able to find, as yet, in the Jeffreys collection, any specimens
of his Astarte nana ; nor have I seen anything more in the way of description
than the four and a half lines given by Mr. Smith. The locality is suggestive,
the specimens were collected by Pourtales, and the features mentioned by
Smith, as far as they go, agree with the present form, though insufficient for
identification. I prefer to use the name nana, and if hereafter it should prove
that it is not Jefl'reys' nana, another name can be applied to it. The shell is
well represented by the figure ; it is about the same color as A. Smithii, but
somewhat larger, flatter, with the beaks more erect and more prominent ; it
has about thirty uniform concentric ribs separated by equal intervals and cover-
ing the entire shell except the lunule : the latter is smooth, but not circum-
scribed by a line ; there is a depvession along the dorsum, but hardly a dorsal
area as distinguished from the rest of the shell. The inner margins are smooth
at all ages observed ; the muscular scars are proportionally larger, and the
pallial line further from the margin than in A. Smithii; the lunular region
is longer and not so deep ; the teeth, though larger, are the same as in that
species. Lon. of shell 8.2 ; alt. 7.8 ; diameter 4.1 mm.

Habitat. Sombrero, 54 fms.; Station 36, Gulf of Mexico, Lat. 23° 13', Lon.
89° 16' W., 84 fms., bottom temperature 60°.0 F. OS' the Carolina coast nearly
to Cape Hatteras, valves at various depths, U. S. Fish Commission.

This shell rnay be crenulate at some age ; it is not, however, like the pre-
ceding species, crenulate at all ages. Some of the Fish Commission specimens,
apparently of the same species, have the interior of a rose pink or light yel-
lowish brown color.

Genus CIRCE Schcmachek.
ScBGENus GOULDIA C. B. Adams.

<GouIdia C. B. Adams, Cat. Coll., p. 20, 1847.

< Thetis C. B. Adams, Proc. Bost. Soc. Nat. Hist., p. 9, 1845, non Sowerby.

Lioconcha Morch, Cat. Yoldi, pt. ii. p. 26, 1853.

Gouldia Dall, P. Z. S., 1879, p. 131 ; Bull. M. C. Z., IX. p. 128, 1881.

<Circe E. A. Smith, Chall. Rep., pp. 221-223, 1885; P. Z. S., 1881, p. 489.

In 1879, in a discussion of the claims of the name Gouldia to retention,
I showed, while two forms were included by Prof. Adams in his genus, that

262 BULLETIN OF THE

G. cerina was his first species ; that a species altogether similar was cited as an
example by the brothers Adams in their revision of the genera of recent Mol-
lusca and adopted as a type by Stoliczka ; that the only other reviser of the
genus, Dr. Carpenter, took a similar view, and postulated the elimination of
the incongruous element of the genus typified by Prof. Adams's second species ;
that a group admitted by all to be separable from the genus Circe (in a sec-
tional or subgeneric sense at least capable of retaining a name) had been sepa-
rated bj' Morch under the name of Lioconcha and generally adopted; that
this group Lioconcha was essentially similar to Gouldia as revised by H. & A.
Adams, Stoliczka, and Carpenter; that Gouldia, having been properly de-
fined by Prof. Adams some eight years before the publication of M6rch'.s unde-
fined name, was therefore entitled by all the laws and usage of biological
nomenclature to take precedence of Lioconcha if they be considered (as they
are) practically synonymous. I referred to Gouldia as a genus, a proceed-
ing which my friend Smith of the British Museum has objected to in a
lively manner, and which, after due consideration, I do not feel disposed to in-
sist upon. The group is closely related to Circe, as typified by C. scripta and
C. divaricata. The differences in the soft parts are, that in Circe proper the
narrow branchiae hang between the dome of the shell and the adductor, while
in Gouldia cerina they are suspended between the two adductors ; also that
Circe has short but distinct siphons, while in Gouldia there are only orifices
between which the mantle edge is tacked together. These characters, like the
conchological ones, are evident enough, but probably in a long series of species
would pass by insensible degrees from one to another. But the general ac-
ceptance which Gouldia has received under the name of Lioconcha indicates
sufficiently that it represents for the majority an a.sseinblage of characters
sufficiently recognizable. The value of the group, as in other cases, will de-
pend upon the view of the individual naturalist. I shall be quite content to
regard it as merely a subgenus. But Mr. Smith would go further, and, dis-
regarding the work of the revisers and the obligations of the nomenclature,
would overthrow Gouldia altogether by a plan which would practically result
in putting an undefined subsequent name in the place of a properly defined
prior one. In this I cannot follow him. In the case of a compound genus
not revised by its author, it is a sound rule to hold by the revision of the first
reviser, when not on other grovmds objectionable. As to following the work-
ings of an author's mind beyond the point where he has seen fit to publish
them, I think it will be as well to wait until the theosophists have their
machineiy in better working order.

Some other notes on this subject will be found under the head of Crassatella.

As to the place which Circe should occupy, I naturally was disposed to
accept without question the views of M. Deshayes, who has studied the Pele-
cypoda so long and well. But on examining the soft parts of Circe I found my-
self obliged to difi"er from his verdict that they were essentially those of Meretrix.
On the other hand they are quite as near Astarte, if not closer, and the shell

MUSEUM OF COMPARATIVE ZOOLOGY. 263

is certainly nearer Astarte than it is to Meretrix, deduction being made of
heterogeneous species. I have therefore, awaiting further information, followed
the acute and accurate Woodward in referring Circe and its subdivisions to the
Astartidce, where they seem to me more at home than in the position assigned
them by the learned French malacologist.

Gouldia cerina C. B. Adams.

(Jotddia cerina C. B. Adams, Dall, Bull. M. C. Z, IX. p. 130, 1881.

Plate VII. Figs. 4a,4li.

Habitat. Charlotte Harbor, Florida, 13 fms.; Barbados, 100 fms.; Station 5,
229 fms. U. S. Fish Commission at various depths northward to Hatteras,
abundantly.

Gouldia (Circe) bermudensis E. A. Smith is more globose, the hinge is differ-
ent, and the lunule shorter, but the sculpture is essentially the same in both, at
least so far as reticulation is concerned. C. cerina is variable, and some speci-
mens are faintly and others very strongly reticulated. C bermudensis is very
much like C. metastriata Conrad, a tertiary fossil.

Family UNGULINID^.

Genus DIPLODONTA Bronn.

Diplodonta venezuelensis Dunker.

Dlplodonta venezuelensis Dunker, Dall, Bull. M. C. Z., IX. p. 136, 1881.

Habitat. Yucatan Strait, 640 fms., one valve and fragments ; Sigsbee, off
Havana, 80 fms. ; West Florida, 19 fms. ; all disunited valves.

Diplodonta turgida Verrill & Smith.
D. turgida Verrill & Smith, Trans. Conn. Acad., V. p. 569, pi. Iviii. fig. 42 (1881).

Habitat. Station 247, in 170 fms., off Grenada, one fre.sh -valve.

This differs from the preceding species by its much greater inflation ; the
hinge teeth are also much more delicate, longer, and of a somewhat different
shape.

264 BULLETIN OF THE

Family LUCINID^.

Genus LUCINA BrcgiIire.

Lucina antillarxun Reeve.

Lucina antiUantm Reeve, Dall, Bull. M. C. Z., IX. p. 136, 1881.

Habitat. Charlotte Harbor, Florida, 13 fms.; Sigsbee, off Havana, in 182
and 450 fms. ; Yucatan Strait, 640 fms.

This species is closely related by some of its varieties to Z. costata Conrad,
of the Florida coast.

Lucina sombrerensis, n. s.

Shell white, stout, nearly equilateral, very globular, small, covered with
sharp elevated thin concentric lamellae, separated by wider interspaces and
becoming crowded and less prominent toward the basal margin ; beaks promi-
nent, full but not inflated ; lunule very small, wider than long, situated di-
rectly under the front of the beaks and bounded by a fine groove; posterior
flexuosity present but inconspicuous, and not modifying the sculpture, which
is the same over the whole shell except as above specified; outer surface not
polished, interior the same; lateral teeth, especially the anterior one, promi-
nent ; cardinal teeth small, nearer the anterior lateral, two in each valve ; lig-
ament in a deep groove above the hinge-line, which groove extends nearly to
the posterior lateral; interior of the margin finely crenulate. Lon. 6.5 ; alt,
6.5; diam. 6.0 mm.

Habitat. Off Sombrero in 72 fms., two valves; West Florida, 50 fms., ©ne
small valve.

This little shell has been known to me for some years from various parts of
the Antilles, and as, after most thorough search, I have been able to find noth-
ing like it described, I am driven to the conclusion that it is still without a
name.

Lucina leucocyma, n. s.

Shell in size, form, and concentric sculpture strongly recalling L. sombre-
rensis, from which it differs in being shorter and higher, in having the concen-
tric lamellae thicker and closer together, and especially five broad radiating
sulcations, sharper toward the beak and becoming less marked toward the
margin, except the anterior one which is strong throughout ; these have
the effect of producing on the surface four broad rounded and gradually
widening ribs which give the shell an unmistakable character. There is no
other radiating sculpture ; the shell is pure white, the beaks considerably
enrolled and bent forward with a minute lunule under them; the interior is
white, with a very finely crenulated margin, with two anterior and one poste-

MUSEUM OF COMPARATIVE ZOOLOGY. 265

rior flexure due to the sulcations ; the hinge is veiy strong, the lateral teeth,
especially the anterior one, strong ; the ligament in a deep groove must be
entirely concealed. Lon. of shell 5.6, alt. 6.5, diam. 6.0 mm.

Habitat. One valve from oft' Sombrero in 72 fms. ; others were collected
in South Florida by Henry Hemphill, and in 6 fms., living, off Turtle Harbor,
Bahamas, by Dr. W. H. Rush, U. S. N.

The strong and salient characters of this small species render it recognizable
at once. A somewhat similarly sculptured species is found at Cape St. Lucas,
but that one has a very deep cavity for the lunule, projecting into the interior.

Lucina funiculata Reeve.
Lucina funkulata Reeve. Dall, Bull. M. C Z., IX. p. 136, 1881.

Habitat. Station 2, 805 fms., one valve.

There is strong reason for doubting the distinctness of this form from L.ja-
maicensis. The thinner character of the shell and the more delicate sculpture
seem to be the differences.

Lucina lenticula Reeve.

Loripes icterica Dall, Bull. M. C. Z., IX. p. 135, 1881.

Lucina lenticula Reeve, Conch. Icon. Lucina, pi. xi. fig. 67, 1850.

Habitat. Station 21, 287 fms.; Yucatan Strait, 640 fms.; Station 36,84
fms., Gulf of Mexico, living, bottom temp. 60°.0 F.; Barbados, 100 fms. ; Sigs-
bee, off Havana, 127 fms. ; Station 220, near Santa Lucia, in 116 fms., bottom
temp. 58°.5 F. ; Station 264, 92 fms., near Grenada, bottom temp. 42^.5 F.

On more careful study these detached valves seem more likely to prove a true
Lucina, and probably Reeve's L. lenticula, than to belong to the genus Loripes,
which in most characters they very much resemble. The material is too poor
and insufficient for a satisfactory determination, at any rate. It may be that
the species should properly be called L. Candeana D'Orbigny, but that is
referred to by Guppy as a Diplodonta.

Lucina scabra Lamarck.
Lucina scabra Lamarck, Reeve, Conch. Icon. Lucina, pi. viii. fig. 45, 1850.

Habitat. Sigsbee, off Havana, in 182 fms.

Lucina sagrinata, n. a.

Shell small, white, subovate, inequilateral, compressed, sculptured with
numerous not very close concentric moderately elevated sharp laminse, be-

266 BULLETIN OF THE

tween which are radiating flutings, not continuous, l)ut each set between each
pair of laminae, independent of those preceding or following it, thus giving a
very pretty shagreened effect to the sculpture ; the flutings are fine and little
raised, not as high as the laminae; beaks not prominent, somewhat posterior;
ends of the shell rounded, anterior slope depressed by the narrow lanceolate
smooth lunule bounded by ridges slightly scalloped by the ends of the laminae;
dorsal slope convex, dorsal area narrower and longer than the lunule, but
otherwise similar to it; teeth stout; inner margins smooth; muscular scars
large, elongated. Lon. of shell 7.6, alt. 5.4, diam. 4,0 mm.

Habitat. Sigsbee, off Havana, in 182 fms. ; Station 21, in 287 fms., Gulf
of Mexico, one valve.

Lucina (Divaricella) quadrisulcata D'Orbignt.

Lucina quadrisulcata D'Orbigny, Voy. Am. Mer., p. 584, 1846 ; Moll. Cuba, II.

p. 294, pi. xxvii. figs. 34-36, 1853.
Lucina americana C. B. Adams, Contr. Conch., p. 243, 1852.

Habitat. Station 36, 84 fms., Gulf of Mexico ; Station 137, 38 fms., near
Santa Cruz.

Gends LORIPES Poll

Loripes compressa Dall.

Loripes compressa Dall, Bull. M. C. Z., IX. p. 135, 1881.
Plate XIV. Fig. 2.

Habitat. Off Cape San Antonio, 413 and 424 fms.; off Sombrero, one valve,
72 fms.

After describing this species I was for a time much in doubt as to the value of
it. Since then I have had an oppportunity of examining the large series com-
prised in the Jeffreys collection, which shows that none of the varieties of L.
lacteus at all nearly approach it. L. lens of Verrill seems to me perfectly dis-
tinct from L. lacteusy and bears much the same relation to it that Lucina filosa
Stm. does to the British L. borealis.

Lioripes lens Verrill & Smith.
Loripes lens Verrill & Smith, Trans. Conn. Acad., V. p. 569, VI. p. 259 (1880).

Habitat. Station 47, in the Gulf of Mexico, 321 fras. (living), bottom temp.
46°.75 F.; Station 230, 464 fms., off St. Vincent, bottom temp. 410.5 F.;
Station 256, 370 fms., near Grenada, bottom temp. 44°.5 F. ; Station 264,

MUSEUM OF COMPARATIVE ZOOLOGY. 267

416 fms., near Grenada, bottom temp. 42°.5 F. ; and many stations of the
U. S. Fish Commission off the coast of New England.

The specimens are smaller, on the whole, and somewhat less rude, but other-
wise do not differ from those from more northern stations.

Genus CRYPTODON Torton.

Cryptodon orbiculatus SEonENZA.

Verticordia orbiculata Seguenza, Men. Vert., p. 9, 1876.
Axinus orbiculatus Jeffreys, P. Z. S., 1881, p. 703, pi. Ixi. fig. 5.

A shell (one valve) which seems to agree with Seguenza's description was
dredged at Station 220, in 116 fms., near Santa Lucia, but I cannot make out
such a sculpture as is figured by Jeffreys on the plate referred to. There are
extremely fine radiating rows of dots, and a powdery surface over them, but I
cannot make out riblets and pores such as are figured.

Cryptodon flexuosus Moxtagu.
One valve occurred at Station 262, in 92 fms., near Grenada.

Cryptodon pyriformis, n. s.
Cryptodon ? obesus Verrill, Dall, Bull. M. C. Z., IX. p. 136, 1881.

Shell thin, white or flesh-color, subtranslucent, not very convex, when
fresh with an appearance as of white dust on the exterior surface, beaks high,
narrow, rather pointed, more or less recurved, with a depressed lanceolate
lunule in front of them; anterior dorsal slope concave, steep, terminating
about half-way between the umbo and the base in a rounded angle; posterior
slope shorter, slightly convex, then inflexed to meet the posterior rib which
has its steepest side posterior and anteriorly passes into a wider flexure of the
surface which lies between the rib and the middle part of the shell; the base
is rounded and produced in the middle line, the curve extending from the rib
to the anterior angle; the outer surface is marked by faint lines of growth
and obscure malleations; the interior is smooth, exhibiting the flexuosities;
the hinge-line is narrow and flattened under the beaks, perfectly edentulous;
the muscular impression faint and elongated. Lon. 11.2, alt. 14.0, diam.
6.5 mm. A larger but imperfect specimen must have been 17.0 mm. high.

Habitat. Yucatan Strait, 640 fms. (broken valves) ; also at the Fish Com-
mission Stations 2646 and 2678, off the Florida and Carolina coasts, in 85 and
731 fms.

This fine species was doubtfully referred to C. obesus in my preliminary
report. Since then I have had the opportunity of examining an unparal-
leled series of this genus comprised in the Jeffreys collection, besides a good

268 BULLETIN OF THE

series of the C. obesv^, and find nothing closely resembling it, either among
the specimens or in the literature. It is very much flutter and thinner than
C. obesus; its texture is of a less earthy and solid character; the base is more
produced in the middle and less evenly rounded. The flattish high and pointed
beaks are also noteworthy. It is nearer C. obesus than to any other form, and
consequently other comparisons are hardly needed. I had formed the idea
that this genus was marked by great variability, but ray study of the Jeffreys
series has convinced me that it is much less so than I had supposed. The
species do vary in breadth and in the sharpness of their flexures, but the
identification of the species is not especially difficult.

Family CHAMID^.

Genus CHAMA BRUGitRB.

Chama lactuca, n.. s.

Shell attached usually by the left valve ; valves diff'erently sculptured ; free
valve orbicular, moderately convex, tip not greatly enrolled ; sculptured with
radiating and concentric series of very small short spines, each grooved under-
neath, generally only the marginal series raised so as to appear spiny and these
only slightly so, the rest look like little radiating noduhitions of which the
radii are discontinuous with each new period of growth ; attached valve in-
flated, smooth, polished, gyrate like a much enrolled Capulus, with indistinct
lines of growth and a succession of flat, wide, very thin, sharp concentric
lamellce, separated by rather wide and gradually increasing interspaces ; the
Lamellae nearly complete the circuit of the valve, and are interrupted only near
the dorsal margin, are slightly recurved, their margins usually irregular from
small fractures ; their anterior or, rather, distal faces, are microscopically ra-
diately shagreened, and just in front of each lamella is a narrow band with
stronger and more distant radiations ; interior smooth, the cavity of the left
valve extending to the tip of the enrolled beak ; the color is usually white,. or
marked with concentric rings of pale livid brown corresponding to periods
of growth ; margin smooth ; muscular impressions narrow, marginal ; hinge
weak, of two lamellar teeth in each valve, the anterior the larger; ligament
hidden in a deep groove so as to be practically internal ; longest diameter
about 25.0, shorter about 15.0 mm.

Habitat. Barbados, 80-100 fms., dredged by the " Hassler."
The apical portion of the valve is shaped like Tellimya, polished and claret-
colored. It may be well to call attention to the fact that the very young
Clxama (macrophylla) has a shell shaped like Cypricardia, with similar hinge
teeth and a simple pallial line ; the adults have very similar characters in the
soft parts, except such as are more or less dependent on the habitus. There
can be no doubt of the near relation of the two.

MUSEUM OF COMPARATIVE ZOOLOGY. 269

Chama sarda Reeve.
One specimen from 38 fms., at Station 127, near Santa Cruz, living.

Family CARDIID^.

Genus CARDIUM Linn^.

Cardium ceramidum, n. s.

Cardlum sp. indet. Ball, BuU. M. C. Z., IX. p. 132, 1881.

Plate rv. Fig. 6.

Shell related to and doubtless the descendant of Cardium haitense Sowerby,*
from the Miocene of Jamaica and Santa Domingo, but much smaller ; with
eighteen ribs instead of twenty-four; the four middle ribs much larger in
proportion to the others ; the granules on the ribs smaller; the anterior
slope fuller and rounder; the posterior more oblique and less elevated; the
shell not so high in proportion to its length; the hinge-margin narrower;
the teeth more delicate, and the beaks not so elevated. Alt. of largest valve
8.2; Ion. do. 8.2; double diameter of same 8.0 mm.

Habitat. Off Havana, Sigsbee, in 182 fms. ; Samana Bay, Dominica, Cou-
thouy; St. Thomas, living near the shore, U. S. Fish Commission steamer
" Albatross," in 1884.

This lovely little shell is yellowish; the foot is extremely long and sub-
cylindrical with a very narrow serrated margin behind ; the palpi are large and
lamellate, the gills broad, the mantle near the orifices at the posterior end,
furnished with a multitude of long stout tentacular processes.

Cardium medium Linne.

Cardium medium Linne', Dall, BuU. M. C. Z., IX. p. 132, 1881.

Habitat. Sigsbee, off Havana, 80 fms.; Barbados, valves in 100 fms.
This common West Indian shell probably inhabits shallow water, and the
valves dredged as above were drifted or disgorged by fishes.

Cardium (Fulvia) peramabilis Ball.

Cardium (Fulvia) peramabilis Dall, BuU. M. C. Z., IX. p. 132, 1881.

Plate IV. Fig. 7.

Habitat. Sigsbee, Station 50, 119 fms.; U. S. S. Bache, April 22, 1872,
100 fms.; Barbados 76-100 fms.; Station 9, 111 fms.; Station 177, 18 fms., oflF

* Quarterly Journal Geol. Society, Vol. VI. p. 52, pi. x. figs. 11 a, 11 b, 1849.

270 BULLETIN OF THE

Dominica; Station 36, 84 fms.; Lat. 23° 18', Lon. 89° 10', 84 fms.; off Som-
brero, 54-72 fms. ; west of Florida, 50 fms. ; off Sand Key, 80 fms. ; Station 132,
in 115 fms., off Santa Cruz; Station 220, in 116 fms., off Santa Lucia; Station
154, 164 fms., off Grenada.

Var. tinctum. Stations 272 and 287, in 7^ to 76 fms., sand and coral, also in
100 fms., all near Barbados, and living specimens; also at Station 127, off Fred-
rikstadt, Santa Cruz, in 38 fms.

The U. S. Fish Commission has dredged this species off the New England
coast, at Station 861, in 115 fms., and thence southward at many intervening
stations to the West Indies, the deepest being off Hatteras, in 124 fms., living.

This extremely lovely shell seems to live in from 50-125 fms., and in water
at a temperature ranging from 40° to 80° Fahrenheit. It is related to and per-
haps descended from the Eocene C. Nicoletii Conrad, which attains a vastly
greater size. Cardium parile and semiasperum Deshayes, of the Paris Basin, are
similar in their general features while differing in detail. The Fulvia modesta
of Adams and Reeve, a North Pacific (Japan to California) species is, so far as
I can learn, its nearest living relative.

A more exquisitely beautiful shell than a perfect specimen of the variety
tinctum I have never seen ; figures can give no adequate idea of its delicacy,
its color, or its elegance.

The small spines are rarely perfectly preserved and often gone entirely,
which makes quite a different-looking shell of it.

Cardium muricatum LinnI:.

A few valves of this well-known Antillean species were dredged by Sigsbee
in 187 fms., off Havana.

Cardium laevigatum Linn6.

Young valves apparently of this species were dredged off Sand Key, in
80 fms. ; off Sombrero, in 72 fms. ; and at Station 2, in 805 fms. As with the
preceding species, they were probably drifted from shallower water, or dropped
by fishes, since the species is known to live at comparatively moderate depths.

Cardium serratum Linne.

Cardium serratum Linne', Dall, Bull. M. C. Z., IX. p. 131, 1881.

Habitat. West of Florida, living, in 30 fms., Bache; Barbados, 100 fms.;
Sigsbee, off Havana, 127 fms.; and, living, in Havana Passage at Station 152,
in 27 fms.

This extremely common shell lives to about 100 fms. in depth; the genuine
deep-water specimens are pale, or with a few pink flecks, without any of the

MUSEUM OF COMPARATIVE ZOOLOGY. 271

usual brown or yellow markings ; have the umbones of a deep pink, fading oflF
on the dome of the valves, and a little more elevated than in specimens from
shallow water; they do not exceed 15.0 mm. in height, and may be considered
as forming a variety syhariticum.

Family ISOCARDIID^.

Gends ISOCARDIA Lamakck.

Subgenus MEIOCARDIA H. & A. Adams.

Meiocardia Agassizii, n. s.

Shell subquadrate, widest at the gently arched base, polished, moderately
inflated, whitish or yellowish; beaks small, incoiled away from the hinge
margin, with a depressed but not circumscribed area in front of them; from
the beak proceeds a strong but moderately sharp-edged keel to the lower
posterior end of the shell, which is produced in a rather sharp angle ; on the
posterior area thus separated near its inner edge is another keel, less sharp,
which extends from the beak to the somewhat rounded upper posterior angle
of the shell, from which the margin obliquely descends with a slight con-
cavity first, and then a little convex, to the lower angle; between the two
keels and nearer the anterior one is a single rounded rib extending to the
margin, over which the concentric sculpture is continuous and unmodified;
the entire surface is sculptured with fine even concentric rugae, somewhat obso-
lete toward the umbo and a little stronger behind the principal keel; toward
the margin are occasional impressed lines indicative of changes in the rate of
growth and presumably varying with the individual; the anterior end of the
shell descends toward the base, where it is somewhat pointedly rounded, and
from which the base extends in an easy outward curve to the posterior angu-
lation; pallial line faint, simple, margin of the shell plain; hinge much as in
M. vulgaris Keeve, the cardinal tooth in the right valve rounded behind, gently
indented on each side ; ligament external, in a groove with raised edges, con-
tinued under the beak as in other species. Lon. of shell 22.0; max. alt. 17.0;
diara. of right valve 8.0; of whole shell, probably 16.0 mm.

One right valve dredged off Trinidad by the U. S. Fish Commission in 117
fms., bottom temperature 64^.5 F.

The great interest attaching to this species, not only on account of its ap-
pearance in a new faunal region for the genus, but on account of the very
small number of recent species known, has led me to insert it here, though not
obtained by the " Blake." * It is dedicated to the memory of my teacher, the
late Prof. Louis Agassiz, whose work in the molluscan subkingdom is familiar
to alL " Cypricardia" isocardioides of Deshayes (Inv. Bas. Paris) would, from
the figure, fall into line with M. Agdssizii. It is certainly not a Cypricardia.

272 BULLETIN OF THE

'5ENU3 CALLOCARDIA A. Adams.
SoBGENus VESICOMYA Dall.

Shell small, smooth or concentrically striate; hinge of Meiocardia but with-
out lateral teeth; epidermis polished, umbones moderately prominent; lunule
circumscribed by a groove; otherwise as in Meiocardia. Type Callocardia
atlantica Smith (Chall. Rep. Lam., p. 156, pi. vi. fig. 8).

In his excellent work on the Lamellibranchiates of the Challenger expe-
dition, Mr. Edgar A. Smith has pointed out that the dentition of these shells
differs from the single valve of Callocardia known, and, while retaining the
name, calls attention to similarities with Kelliella. I have carefully studied
the hinge of Kelliella, using specimens received from Prof. G. 0. Sars, and also
the hinge of C. atlantica and of Pecchiolia subquadrata Jeffreys. They are very
difficult objects, owing to their minuteness, shape, and fragility, but I have
been able fully to confirm the accuracy of the excellent figure of the hinge of
Kelliella given by Professor Sars Jn his Moll. Reg. Arct. Norvegiae (pi. 19,
Sg. 15). The hinge of the Callocardia atlantica, if I have rightly identified
my little shell, of which I feel pretty confident, is destitute of the angular
arrangement noticeable in Kelliella miliaris, and resembles that of Meiocardia,
deduction being made of the posterior (and only true ?) lateral tooth existing
in that genus and in Bucardium or Isocardia (cor). In the Jeffreys collection
I find two lots of specimens labelled Pecchiolia subquadrata. One comprises
two small and mutilated valves; the other, three fresh specimens, all of which
were obtained by the Porcupine expedition. Judging by these, the figure
(P. Z. S., 1881, pi. Ixx. fig. 3) of this species is poor; the shell is usually
higher in proportion to its length, more as in Lyonsiella abyssicola Sars. The
two largest valves measure 4.0 and 4.0 mm. high against 4.5 and 5.0 mm.
long. Only one approximates to the figure, and the cause is evidently patho-
logical. Dr. Jeffreys calls attention to the thickness of the hinge-line, com-
pared with the size of the shell, and describes it a.« edentulous. This probably
arose from the fact that the extremely thin and fragile lamellar teeth snap off
even with the hinge-line if the shell be forced open after drying with valves
closed, or at most but one tooth remains.

To get the dentition, which I saw was mutilated in the specimen which
seems to have served Dr. Jeffreys for his descrijJtion, I sacrificed the best
perfect specimen, breaking away the ventral margins without opening the
valves, and in this way found it perfect. When closed the left short cardinal
is in front of and above the right short cardinal, and the left long cardinal in
front of and below the right-hand equivalent tooth, as in Isocardia. The ex-
ternal ligament is visible inside when the valves are closed, for there is a slight
gape under its posterior end, but its attachments and position are strictly
marginal and external. The yellowish suffusion of the surface is a little more

MUSEUM OF COMPAEATIVE ZOOLOGY. 273

darkly clouded on the upper posterior part, and is divided here by a pale ray
from the umbones to the lower posterior margin. The sparsely set micro-
scopic tubercles can only be observed with a magnifier; to the eye the surface
looks shining and smooth, not unlike that of Kellia suborbicularis. Taking it
for granted that the specimens labelled by Dr. Jeffreys are really the same as
his type in the British Museum, and authentic, it follows that Lyonsidla or
Pecchiolia subquadrata Jeffreys is a congener of the Callocardiae of Smith,
though, as Smith himself indicates, not of the original Callocardia. They
have a hinge much nearer the original Isocardia than Kelliella has, but differ-
ent from either ; and, if they were two inches in diameter, would be unani-
mously accorded separate names. As the element of size can hardly, on
reflection, be considered in systematic work as opposed to definite characters,
I have taken the responsibility of separating them, leaving to my more
"conservative" friends the usual option of disregarding the distinctions if
they prefer.

Through the extreme courtesy of Mr. Smith, I have received from him an
enlarged drawing of the type of Callocardia showing the hinge. It is impos-
sible without a figure to exhibit clearly the marked differences which exist
between the hinges of Callocardia and Vesicovvja. This I hope, later, to
supply.

The species then comprise Callocardia (^Vesicomya) subquadrata (Jeffr.) Dall,
and C. (F.) atlantica, C. {V.) pacijlca, and C. {V.) Adamsi Smith; C. (F.)
pilula and venusta Dall.

Lest there should be a question as to the species, I retain the description
I had prepared before the publication of the Challenger report, and which I
should have otherwise (as in other cases) suppressed in this paper.

Callocardia (Vesicomya) atlantica Smith.
Callocardia (?) atlantica Smith, Chall. Rep. Lam., p. 157, pi. vi. fig. 8, 1885.

Shell small, extremely fragile, rotund, polished, whitish, ^^dth a pale filmy
epidermis ; sculptured evenly all over with fine close-set concentric rounded
lines or threads; the margin of the valves is nearly circular, above which rise
the small prominent, inflated beaks; they are nearly median as regards the
ends of the valves, but extend a little forward of the median line; both ends
and the base are subequaUy rounded ; a ridge extends (as in Cetoconclid)
from the beaks backward, including between itself and the hinge margin a
narrow inbent area; the ligament is wholly external, long linear, black; there
is no internal cartilage ; in each valve is a thin short rectangular lamellar
cardinal tooth parallel with the hinge-line and immediately contiguous to a
lamellar elongated tooth, between which and the margin is a sharply cut
groove extending far behind the tooth ; interior polished, white ; muscular
Bears invisible. Lon. of shell 4.5, alt. 4.5, diameter about 4.0 mm.

VOL. XII. — NO. 6. 18

274 BULLETIN OF THE

A single specimen was obtained living, at Station 236, off Bequia, in 1591
fms., fine ooze, bottom temperature 39°. 0 F.

The specimen agrees well in every way with Mr. Smith's descriptions and
figures.

Callocardia (Vesicomya) pilula Dall.

Diplodonta pilula Dall, Bull. M. C. Z., IX. p. 136, 1881.
Plate VIII. Fis. 13.

Habitat. Station 43, 339 fms., one specimen.

This little shell when submitted to higher powers, and especially when com-
pared with the V. atlantica, was recognized as a Vesicomya. Its very minute
size and the difficulty of opening it without breaking it led to its reference in
the preliminary work to Diplodonta, while the doubt attending the reference
was at the same time expressed. I am of the opinion that it is really a young
specimen of the species afterward named atlantica by Mr. Smith, but I do not
feel certain of it. Should it turn out to be so, I should prefer to waive my
name in favor of Mr. Smith's, for the reason that there was not information
enough in my description to enable any one to recognize it as a Vesicomya.

Callocardia (Vesicomya) venusta, n. s.

Shell pale straw-color, elongate-ovoid, inflated, equivalve, inequilateral, thin,
chalky in consistency, fragile ; beaks full, near but not torching each other,
lunula large, marked by a sharply cut groove or line ; outer surface polished,
uniformly concentrically sculptured with fine not very regular wrinkles ; in-
terior polished ; anterior scar high, rather narrow, rounded below and pointed
above ; posterior scar much shorter, rounder and broader; pedal scar small,
round, strongly marked, under the anterior tooth; hinge that of the subgenus,
teeth flat and thin, the anterior cardinal of the left valve the larger, its edge
waved ; inside margin of the valves, close to the edge, sculptured with a few
not quite parallel sharp grooves, much as in Transennella ; anterior ends of the
valves rounded, posterior more pointed especially toward the lower posterior
part of the margin. Max. Ion. of shell 19.0; max. alt. 14.0: double diameter
of largest valve 11.5 mm.

Habitat. One valve dredged at Station 1, in 801 fms., mud, bottom temper-
ature 39°.5 F., off Havana. Also several valves by the U. S. Fish Commission
at Station 2678, in 731 fms., oflf Cape Fear, North Carolina, bottom temperature
38°.7 F.

The presence of a very young valve shows that the elongated shape is con-
stant. The shell in shape a good deal resembles Meiocardia Agassizii, but the
sculpture is less regular, there are no lateral teeth or radiating keels or ridges,
and the substance of the shell is more earthy.

MUSEUM OF COMPARATIVE ZOOLOGY. 275

Family VENERID^.

Genus CYTHEREA Lamarck.

Subgenus DIONE Megekle v. Muhlfeldt.

Cytherea (Dione) hebraea Lamarck.

Ci/therea hebroea Lamarck, An. s. Vert., VL p. 308, 1818.

Habitat. West of Florida 30 fms. ; off Gordou Key 68 fms. ; Barbados,
100 fms.

These specimens are all very young, and yet seem to show the characters of
this species sufhciently. Most of them show traces, outside of the smooth
colored surface, of a chalky layer which is very soon worn off and leaves no
trace in the adult.

Cytherea (Dione) albida Gmelin.

Dione albida Reeve, Conch. Icon. Dione, pi. x. fig. 39.

A number of very small and immature valves, dredged at Stations 247 and
262, near Grenada, in 92-170 fms., may belong to this or some allied species.
They are not in a condition to be accurately determined.

Cytherea (Veneriglossa) vesica, n. s.

Shell thin, inflated, rounded ovate, white, uniformly concentrically grooved,
polished; no differentiated dorsal area; luuule wide, short, marked by a fine
inscribed line ; beaks tumid, involved, as in Isocardia, twisted away from the
hinge-line so that their tips are widely separated; margins thin, simple; hinge
with the teeth arranged much as in Cytherea Sayana Conrad, but with the
depressions prolonged into pits, the ends of the teeth sharp and pointed, and
the ventral margin of the hinge-shelf upturned; ligament long, in a deep
groove, passing away from the hinge-line under the beaks as in Isocardia ;
muscular impressions small, near the margin ; pallial line with a shallow
wide wave just before the posterior adductor scar. Lon. of shell 22.0 ; alt. 21.0;
diam. 17.0 mm.

Habitat. Station 36, 84 fms., in the Gulf of Mexico ; Station 167, near Gua-
delupe, in 175 fms.; Barbados, 100 fms., by the "Hassler" ; all dead valves.

This is a very singular shell. In the absence of the soft parts I am at a loss
to place it. If it were not for the slight wave in the pallial line, I should, in
spite of its Venerid teeth, have placed it in the Isocardiidre. The very young
shells, though more elongated and less tumid, resemble Vesicomya atlantica
Smith; the adults are more like it on a larger scale. The dentition is alto-

276 BULLETIN OF THE

gether different from Circe, and those forms of Cytherea which have somewhat
similar teeth have the beaks and ligament different, and an angular pallial
sinus. It seems to be worthy of a section to itself.

Genus VENUS Linne.

Venus pilllla Reeve.

V. pilula Reeve, Conch. Icon. Venus, pi. xv. fig. 58, 1863.

Valves were obtained by Sigsbee, off Havana, in 80 fms., and at Station 272,
in 76 fms., Barbados, which appear to be referable to this species.

Subgenus CHIONE Megerle v. Muhlfeldt.

Venus (Chione) pygmsea Lamarck.

Venus pygmcea Lam., An. s. Vert., 2me ed., VI. p. 337. Desh. Cat. Biv. Brit. Mus.,

p. 129.
Venus incequivalvis D'Orbigny, Moll. Cuba, p. 277, pi. xxvi. figs. 38-40, 1846.
Venus irapezoidalis Kurtz, Cat. N. and S. Car., p. 5, 18G0 (fide Stm.).

Habitat. Station 26, 110 fms., in the Gulf of Mexico ; Yucatan Strait, 640
fms. (valves).

The shell is found living in Charleston Harbor and southward.

Venus (Chione) cancellata Lamarck.

V. cancellata Lam., An. s. Vert., V. p. 588, 1818. (1 = V. dysera Linne.)

Habitat. Gordon Key, 68 fms. ; off Havana, in 127 fms., drifted valves.
It is abundant on the Floridian and Antillean coasts in moderate depths of
water.

Family PETRICOLIDiE.

Genus PETRICOLA Lamarck.

Petricola divaricata Chemnitz.

Petrkola divaricata (Chemn.) D'Orbigny, Moll. Cuba, II. p. 265, 1853.

One valve was obtained by the "Bache" in 68 fms., near Gordon Ke3% The
species is not rare, in proper places, in South Florida and through the whole
Antillean region.

MUSEUM OF COMPARATIVE ZOOLOGY. 277

Family TELLINID^.

Genus TELLINA Linne.

Tellina Antoni Philippi.

Tellina Antoni Philippi, Dall, Bull. M. C. Z., IX. p. 134, 1881.

Habitat. West of Florida, 19 fms. [Bache]; Carolina coast, young valves,
U. S. rish Commission.

Tellina squamifera Deshates.

T. squamifera Deshayes, P. Z. S., 1854, p. 365 ; Reeve, Conch. Icon. Tellina, pi. Iv.

fig. 325, 1869.

Habitat. Off Sombrero, valves, in 54-72 fms.

Only one valve of this elegant little shell appears to have been known to
Reeve.

Tellina sybaritica Dall.

Tellina sybaritica Dall, Bull. M. C. Z., IX. p. 134, 1881.

Plate VI. Fig. 11.

Hibitat. Yucatan Strait, 640 fms., one valve.

I have not been able to find any more specimens of this verj' lovely little
shell, but I doubt if it lives in the depth of water from which it was dredged.
Its brilliant crimson color would be rather anomalous in that depth of water,
and the shell seems very solid for such a habitat.

Tellina tenera Say.

T. tenera Say, Journ. Acad. Nat. Sci. Piiil., II. p. 303, 1822. Tryon, Am. Marine

Conch., p. 148, fig. 349.

Habitat. East coast of the United States and the Antilles ; off Sand Key,
80 fms. ; off Sombrero, 72 fms. ; west of Florida, in 30 fms. ; and at Station 287,
near Barbados, 7-50 fms.

The southern specimens have a tendency to be brighter colored.

Tellina plectrum C?) Hanley.

Tellina plectrum {■?) Hanley, Dall, Bull. M. C. Z., IX. p. 134, 1881.

Habitat. Yucatan Strait, 640 fms., one valve.

I have not been able to learn anything more about this specimen.

278 BULLETIN OF THE

Tellina Gouldii Han let.
Tellina Gouldii Hanley, Dall, BulL M. C. Z., IX. p. 134, 1881.

Habitat. Yucatan Strait, 640 fms.

This is not Angulus Gouldii Carpenter, of the west coast of America, which
has the aspect of a small Macoma. Whether it is the true T. Gouldii of Hanley
I have no means of deciding, but it resembles the figure of that species. I
have it also from South Florida, collected by Hemphill.

Family SEMELIDtE.
Genus ABRA Risso.

Abra Risso, Hist. Nat. Eur. Mer., Moll., p. 370, 1826.
Syndosmi/a Recluz, 1843.
Abra (Leacli Mss.) Gray, 1852.

In using Syndosmya of Recluz for the following species I followed Deshayes,
who adduces some weighty reasons for not adopting the name Jbra (Leach
Mss.) quoted under the synonymy of certain species in 1818 by Lamarck, and
not published by or for Leach until 1852. But it seems that Risso used the
name in 1826, and, witliout having time thoroughly to investigate the question,
I have concluded to follow Adams and Verrill in the present case.

Abra longicallis Scacchi.
Syndosmya longicallis Scacchi, Dall, Bull. M. C. Z., IX. p. 133.

Habitat. Station 41, G\ilf of Mexico, Lat. 23° 42', Lon. 83° 13' W., living in
860 fms. ; Stations 136 and 137, near Santa Cruz, in 508 to 625 fms., bottom
temperature 42°. 5 F.; also Stations 161, 163, 221, 227, 228, 230, and 264, in
416 to 769 fms.

This is a really abyssal species, and ranges from moderate depths to at least
1500 fms. It is very widely distributed.

Abra lioica Dall.

Syndosmya lioica Dall, Bull. M. C. Z., IX. p. 133, 1881.

Plate IV. Fig. 8.

Habitat. 20 miles west of Florida, 30 fms. ; off Sombrero in 54 and 72 fms. ;
off Sand Key, 30 fms.; Station 36, 84 fms.; Station 9, 111 fms.; Station 5,
229 fms.; Station 2, 805 fms. Stations 128, 167, 206, and 247, in 170 to 180
fms. ; valves, etc.

MUSEUM OF COMPARATIVE ZOOLOGY. 279

This species is more oblique and inequilateral, more rostrated and quad-
rangular, than A. ceqiuilis Say, as figured and described by bim. It does not
gape behind, as that is said to do, nor is its palUal sinus as irregular as in
A. (equalis, but it is more ovate and does not rise as high ; the teeth also appear
to be stouter and wider in A. lioica. It has been obtained at many stations of
the Fish Commission off the Carolinas, and even to the New England coast.
It does not seem to live in more than 180 fms.; all found at greater depths
were dead valves.

Genus ERVILIA Tcrton.

Ervilia nitens Montagu.

One valve of tbis common West Indian shell was dredged in 7 to 50 fms., at
Station 287, near Barbados.

Genus CUMINGIA Soweebt.

Cumingia tellinoides Conrad.

Cumingia tellinoides Conrad, Journ. Phil. Acad. Nat. Sci., VI. p. 258, pi. ix. figs. 2, 3,

1830.
Lavignon Petitiana D'Orbigny, Moll. Cuba, II. p. 236, pi. xxv. figs. 33-35, 1846.
Lavignon antillarum D'Orbigny, 1. c, p. 236, pi. xxv. figs. 36-38, 18-16.

Habitat. Station 10, in 37 fms., Gulf of Mexico, living.

The species is variable in form. The artist has reversed one of D'Orbigny's
figures, so that his two figures look more unlike than they would otherwise.
It is a connnon shell on the coast of the United States in comparatively shoal
water, and from its nestling habit is generally deformed.

Genus SEMELE Schumacher.

Semele obliqua Wood.

Tellina obliqua Wood, Gen. Conch., t. 41, figs. 1-2, 1815.

Amphidesma variegata Lamarck, An. s. Vert., lere ed., V. p. 490, 1818.

One valve dredged by the "Bache," May 13, 1872, in 63 fms. S.W. of Gar-
den Key.

Semele cancellata D'Orbigny.
Amphidesma cancellata D'Orbigny, Moll. Cuba, IL p. 241, pi. xxv. figs. 42-44, 1846.

Valves were obtained in 30 fms. west of Florida. It is not rare un the
shores of South Florida and the Antilles.

280 BULLETIN OF THE

Family POROMYID.E Dall.
Genus POROMYA Forbes.

Poromya Forbes, 1844 ; Emhla Loven, 1846. Type P. anatinoides Fbs. {= P. granu-

lata Nyst.)

Shell gaping a little behind, granulose externally under a thin epidermis, in-
ternally with an internal cartilage in a stout posteriorly directed fossette over
which a linear external ligament extends from under the beaks backward over
the cartilage to the posterior end of the hinge-line ; before the cartilage in the
right valve is a stout cardinal tooth, generally notched in front; in the left
valve is a small sunken triangular tooth in front of the fossette, and a long
distant posterior lateral tooth lies behind the beak. There is a very slight
indentation of the pallial line, the foot is long and cylindrical, the siphons
rather short, surrounded with a I'ringe of rather stout tentacles. There is no
ossicle. Gills as in Cetoconcha, with no free branchiae. The interior of the
shell is faintly pearly under a wash of non-perlaceous substance.

Section CETOCONCHA Dall.

Shell differing from Poromya proper by the cartilage being almost external
and the fossettes diminished in size and upturned, the external ligament con-
sequently nearly obsolete ; the dentition obsolete except the cardinal tooth of
the right valve, which itself is sometimes absent in the adult, though observ-
able in the young shells ; other shell characters miich as in Poromya, The foot
is compressed and hatchet-shaped, grooved behind ; the mouth has two large
superior palpi and two (or none) small inferior palpi not modified as gills.
The foot stands in a socket as in Verticordia and Cuspidaria. On the ventral
surface of the body, behind the foot, are two (sometimes four) rows of less than
semicircular lamellae closely adjacent to each other and firmly fixed to the sur-
face by the whole base of each lamina. There is one row on each side ■with a
shorter supplementary outer row in other cases. They radiate forward in a
curve from a point a little distance behind the foot, and may quite or not quite
meet at this point. In C elongata I found no inferior palpi, a state of things
perhaps due to injury, though the specimen seemed perfectly preserved; the
other species had them. In all there was a row of similar lamellae to those
above described, starting on each side from behind or under the inferior palpus
of that side, or the place in front of which it should have been, and extending
backward in such a curve as would, if prolonged, have joined its posterior end
to the anterior end of the row coming from behind the foot. The lamellae are
not connected by a raphe. These lamellae represent the branchiae of ordinary
Pelecypods, and if even these are absent, as seems possible, in Cuspidaria, it
is difficult to doubt that we have a progressive series: in Cuspidaria none ; in

MUSEUM OF COMPARATIVE ZOOLOGY. 281

Cetoconcha the ventral body wall externally gathered along a line into pinched-
up laminae which develop at opposite ends of a lateral line, with a supplemen-
tary second line corresponding to the branchial appendix ; finally, the posterior
consolidation of the series into a small gill free from the ventral surface except
at its proximal extremity, as in Lyonsiella and Verticordia.

The incurrent siphon is long, retractile into the cavity between the sides of
the mantle, the excurrent siphon much shorter ; around their bases is a series
of stout (in C. elongata arborescent ?) darkly pigmented tentacular appendages,
with smaller papillae inside from them, but no visible ocelli ; the mantle from
below the mouth backward is open for two thirds the distance to the siphons,
a marked distinction from Verticordia and its congeners ; its margin is plain
and not very stout ; the intestine passes through the heart, below which are
two glandular brown feather-shaped renal organs ; the liver, ovaries, and
muscles are well developed, but a large part of the body cavity is vacant, and
its walls are sustained by mesenteric bands or fibrilla; attached to the adduc-
tors or the dome of the shell. Type Lyonsia bulla Dall.

The remarkable characters of this group will be sufficiently evident to those
who have a fair knowledge of the macroscopic anatomy of the Pelecypods.
Especially do the gills attract attention, and enforce the lesson of the muta-
bility of these breathing organs, and their unfitness for use in fundamental
classification.

To Poromya as restricted belong P. gramdata Nyst, P. sublceins Verrill, P.
neceroides Seguenza, and P. australis and probably P. Icevis Smith.

To Cetoconcha belong C bulla Dall, C. tornata (Pecchiolia) Jeffreys, C. nitida
{Thracia) Verrill, and C elongata, albida, and margarita, new species.

The shells grouped by Deshayes, in his discussion of the moUuscan fossils
of the Paris basin, under the name of Poromya, form a very heterogeneous
assembly, which, in the absence of typical material, would be difficult to
assort properly.

Poromya granulata Nyst and "Westendorp.

Poromya granulata Nyst and West., Dall, Bull. M. C. Z., IX., p. 108, 1881.

Corbula gramdata Nyst and West., Nouvelles Res. dcs Coq. Foss. d'Anvers, p. 6,

pi. iii. fig. 3, 1839.
P. anatawides Forbes, ^gean Rep., 1844, p. 103.

Habitat. Sand Key, 15 fms. ; Station 36,84 fms.; Station 9, 111 1ms.;
Station 5, 229 fms.; off Sombrero, 72 fms.; temperatures at bottom 49°. 5 to
60°. 0 F.

Variety P. australis Smith.*

Habitat. Barbados, 100 fms.; off Sombrero, 54 fras.; Station 20, off Bahia
Honda, Cuba, in 220 fras., living, bottom temperature 62°. 0 F.; Station 202,
off Grenada, in 92 fms., sand, same temperature.

* Poromya australis Smith, Chall. Lam., p. 64, pi. xi. figs. 2 a,2 b, 1885.

282 BULLETIN OF THE

A further study of these specimens, together with those of the Jeffreys col-
lection, has confirmed me in the opinion I expressed in my preliminary report
as to the remarkable variability of this species, the modifications being so grad-
ual that I am in doubt as to whether more than one species exists in our seas,
unless the P. sublcevis Verrill be different, as from the figure would seem likely,
if it be normal. P. neceroides has a surface similar to that of P. sublcevis, but
is at the opposite extreme of form. P. rotundata has a sparse or close granu-
lation indifferently, if one may judge by the few specimens I have seen. I
find among the specimens collected some small, inflated, and triangular, com-
pared with the average of the others, in which the granulation is composed of
beautiful minute spheres, perfectly transparent and closely set in quincuncial
arrangement with the greatest regularity. They agree in most details with
the description given by Smith of his australis. The granulation is coarser
than in the average granulata, and the effect of the light upon the transparent
spherules, under a glass, gives them the appearance of little cups or tubes. I
cannot feel positive that they are the same as the form described by Mr. Smith
without a comparison of specimens ; but they agree too closely to warrant
giving any other name to them until a comparison can be made. I have seen
one alcoholic specimen of P. granulata in which the soft parts showed no essen-
tial differences from Cetoconcha. The lower palpi were present and the siphons
not very long.

Poromya (Cetoconcha) albida, n. s.

Shell not pearly externally, white, thin, punctate and polished toward the
beaks, toward the margin with rather sparee granules covered with a thickish,
wrinkled, straw-colored epidermis ; beaks slightly nearer the anterior end, not
contiguous, rather high, small, inflated, but less so than in C tornata; both
ends rounded, the posterior a little less inflated and more produced; base
evenly rounded ; interior strongly radiately striate ; muscular impressions
high, narrow, impressed ; right valve with the hinge-line obtusely arched, the
centre under the beaks with a solid triangular thickening ; the anterior end
of this supports a stout, short, round-topped cardinal tooth, behind which the
callus supports on its dorsal surface a stout triangular cartilage, the anterior
end of which probably appears between the anterior bifurcation of the external
ligament, but the posterior apex of which is internal and covered by the liga-
ment ; the ligament, as in all this group, turns away from the hinge-line and is
lost under the beaks; above its course is an elevated narrow ridge which ex-
tends posteriorly to the end of and very close to the hinge-line. Lon. 21.5 ;
alt. 19.5 ; diameter of the valve 8.5, and of the whole shell probably 17.0 mm.

A single right valve was obtained by the Fish Commission at Station 2159,
in 98 fms., near Havana, Cuba. The shell resembles Poromya suhlcevis Verrill,
but has twice or three times the size; otherwise, until the hinge is critically
examined it would pass for that species. It is, however, larger than any known
Poromya.

MUSEUM OF COMPARATIVE ZOOLOGY. 283

Poromya (Ceto concha) elongata, n. s.

Shell whitish, not pearly, somewhat resembling the preceding, but much
more elongated ; surface more densely and minutely granulate, tbe granules
being in even radiating series for the most part ; the lines of growth are
stronger and the surface not so smooth ; the epidermis is similar, but appar-
ently thinner ; the beaks are contiguous, and are less elevated, less spiral, and
less prominent ; the hinge-line, though longer, is thinner; the cardinal tooth
more acute and much smaller ; the shelf for the cartilage weaker, longer, and
narrower; the ridge extending backward from the beaks is not so near the
hinge-margin, and the area between is wider and obliquely cut oflF at its pos-
terior end, forming a more decided angle than in G. albida ; the anterior end
and base are elegantly rounded, but the posterior end is somewhat rostrated
with an obscure impression extending from the beaks to the lower posterior
rounded angle of the rostration ; the beaks are nearly equidistant fror.i the
ends, but probably a little behind the median line ; the hinge-margin in the
right valve is a little expanded before the beak. Lon. 22.5 ; alt. 17 ; diameter
of right valve 6.25 ; of shell, probably, 12.5 mm.

A single right valve was obtained by the " Blake" at Barbados, in 100 fms.;
and a living specimen by the U. S. Fish Commission at Station 2337, north-
west of Cuba, in 199 fms. It has somewhat the shape of Poromya neceroides
Seguenza, but the hinge differs. The soft parts are described in the sub-generic
diagnosis with some additional notes under the next species. The lower palpi
are absent, and the gill rows one on each side, adjacent, but not touching, at
the point of origin without any appendix.

Poromya (Cetoconcha) bulla Dall.

Lyonsia bulla Dall, Bull. M. C. Z., VI. p. 61, 1878 ; IX. p. 107, 1881.

(?) Thracia nitida Verrill, Trans. Conn. Acad., VI. p. 221, pi. xxxii. fig. 22, 1884.

Habitat. Station 31, Gulf of Mexico, in Lat. 24° 33' N., Lon. 84° 23' W.,
1920 fms., living, bottom temperature 39°. 5 F.; U. S. Fish Commission (as
Thracia nitida), off Chesapeake Bay, in 1917 fms.

The agreement between Professor Verrill's figure and description on the one
hand, and the Blake specimens on the other, is so close, that I can hardly
doubt they are the same species, though I have not examined specimens of his
shell. The soft parts of this species are much the same as in C. elongata, ex-
cept that the retractile siphon is proportionately longer, and the gill series
consists of two short rows (5-7 lamellse) on each side radiating forward from
a point immediately behind the foot. The lower palpi are present but not
branchial; but on the body surface near them are two short rows (8-10 sec-
tions or lamellse), one on each side, diverging backward, the anterior end of
each being under or behind the lower palpus of that side. The trend of these

284 BULLETIN OF THE

is such that, if continued, they might join endwise with the inner series, corre-
sponding to the gills coming from behind the foot, and form a single row.

The young shell has the teeth of Cetoconcha, the adult loses them entirely,
having only the usual enlargement of the hinge-line to support the now nearly
external cartilage, the linear filmy ligament proper outside of the former being
hardly perceptible, though present. It was on this account, even in the ab-
sence of an ossicle, that I was led provisionally to describe this as a Lyonsia,
and perhaps Professor Verrill to call it a Tliracia. On further study 1 found
in some cases, under the epidermal fibres noted in the original description,
calcareous granules, especially toward the ends of the shell, while in other
places there seemed to be no granules. The posterior slope shows more epi-
dermis than the rest; the outer surface of the shell is faintly iridescent where
polished, as is the interior. The tendency to rostration at the posterior end
seems more marked in the older than the younger shells, but differs in differ-
ent individuals. It is not very marked in any.

The measurements of the largest specimen in my possession are 13.0 mm.
long, 10.0 mm. high, and 9.0 mm. in maximum diameter. There seems to
have been an error in recording or in printing the dimensions of the specimen
used for the original description.

Poromya (Cetoconcha) margarita, n. s.

Plate VIII. Fig. 10.

Shell small, white, inflated, slightly inequivalve and inequilateral, sub-
rostrate posteriorly. The right valve a little the smaller ; the lateral outline
of the valves viewed from within recalls Poromya granulata, but the shell is
much more inflated ; granulations faint or obsolete except behind, as in the
last species ; teeth of the hinge obsolete in the adult ; anterior and basal mar-
gins rounded ; behind is a slight concave wave in the margin below, while the
upper posterior margin descends more rapidly than the anterior one, and is
subtruncate, and the rostration thus produced gapes slightly ; the beaks are
inflated, but do not rise very high above the hinge-line. Max. Ion. 7.3; alt.
5.5 ; diam. 6.6 mm.

Soft parts as in the last species, but the retractile siphon much shorter ; the
respiratory laminae nine on each side anteriorly, the posterior series about
eight, and the appendix with about six lamellae. The lower palpi present,
small ; the foot slender, grooved behind, and the other features as in the
description of the subgenus.

Habitat. Station 44, off Tortugas, in 539 fms.; Station 22 1^^ near Santa
Lucia, in 423 fms., ooze ; and Station 176, off Dominica, in 391 fms., ooze;
living at all the stations; the bottom temperature ranged from 39°. 5 to 43. °5 F.
The fragility of the shell is such that nearly all the specimens were broken in
the trawl.

MUSEUM OF COMPARATIVE ZOOLOGY, 285

This little species was overlooked in my preliminary examination and taken
for the young of the preceding. A more careful study shows they are abso-
lutely distinct.

Family VERTICORDIID^ Seguenza.

Genus VERTICORDIA Wood.

Verticordia (Wood Ms. 1844) Sowerby, Min. Conch., pi. 639, Aug. 1844.

Verticordia (Wood Ms.?) Gray, Syn. Brit. Mus., 1840 (sine descr.).

Hippagus Philippi, Sowerby, not of Lea.

Iphigenia O. G. Costa, 1850, not of Schumacher, 1817.

Verticordia Seguenza, Journ. de Conehyl , 1860, p. 286; Fischer, 1. c, p. 295;

Seguenza, Rendic. R. Accad. delle Scienze, 1876. Ball, Bull. M. C. Z., IX.

p. 105, 1881.

Since my examination made in 1881 of the specimens of this group, I have
been able to examine alcoholic specimens of V. acuticostata and additional
specimens of other species, beside those contained in the Jeffreys and U. S.
Fish Commission collections. I have therefore reviewed the whole subject,
and have the pleasure of being able to add several facts of interest, and espe-
cially to determine positively the relations of the animal and the character of
the soft parts in the species referred to, and therefore probably for the whole
group. I have found also that the shells which have been referred to this
group differ among themselves in regard to characters of hinge and dorsal mar-
gin, so as to require separation into different subgenera or sections.

Verticordia (s. s.). Shell small, more or less convex, with a deeply im-
pressed lunule and a large, arched, bridge-like ossicle attached below the beaks
to an internal cartilage in each valve ; this ossicle is expanded outward at its
posterior end, and, in the most typical species, is much broader than long ;
the right valve has a strong conical tooth behind the internal convexity due to
the impressed lunule, and no lateral tooth ; the left valve has the lunular edge
produced to fit in front of the cardinal tooth of the right valve, and has the
upper surface of the posterior hinge-margin bevelled away so that that edge
may fit under the opposing edge of the right valve ; the cardinal tooth in
young specimens is grooved axially, but when adult is conical ; the line of the
external ligament is continued under the spiral of the beaks. Soft parts (in
V. acuticostata Phil.) having the mantle-edge thick and fleshy, corresponding
in form to the sulcations of the valves, but not fringed with papilljB ; united
on the ventral surface, with a simple very short slit opposite the foot ; with a
papillose siphonal opening posteriorly (the anal siphon probably present as in
Lyonsiella, but, on account of contraction from the spirits in which it had been
preserved, not clearly made out) with about four ranks of papillae, the inner-
most ones largest, but in the specimen much contracted. Mouth axially

286 BULLETIN OF THE

striate, opening below the anterior adductor, without palpi ; foot cylindrical,
large for the size of the animal, distinctly grooved behind ; laterally somewhat
compressed near the acute tip ; base set as it were in a socket, which, when cut
open, shows a chamber of considerable size well suited for a marsupium ; in
this specimen the aperture of this chamber fitted closely around the foot, which
stood like a stopper in a bottle ; on each side of the foot and attached to the
margin of this opening was a single lanceolate, small stout fleshy gill laterally
longitudinally sulcate, and very small for the size of the animal ; * the posterior
side of the foot was distinctly grooved, but no byssus was present.

It will be noted that these features agree essentially with the soft parts of
Lyonsiella abyssicola as described by Sars, and vindicate the judgment of
Jeffreys, who approximated the two groups.

The type of Verticordia as restricted is V. cardiiformis Sowerby (Min. Conch.,
pi. 639, 1844).

To this group I refer V. acuticostata Philippi ; V. Deshayesiana Fischer
{-\-Japonica A. Ad.) ; V. Woodii Smith; V. tomata Jefi'reys; V. flexuosa and
V. granifera Verrill, the former described as a Mytilimeria by that naturalist;
V. parisiensis Deshayes; V. perversa and V. Seguenzce, n. s.

The type of this group resembles Trigonulina in its narrow form and exter-
nal sculpture, but not in hinge characters.

Trigonulina (D'Orbigny, 1845, + Trigoniluna Chenu, 1662, -f Hippagus
(sp.) Adams and Reeve, not Lea).

Shell compressed laterally; ossicle long, narrow, rectangular, flat; right
valve as in Verticordia, but with a long lateral tooth; left valve as in Verti-
cordia. Type T. ornata D'Orbigny. Cuba.

To this group belong V. novemcostatus Ad. & Rve., and V. coelata Verrill;
all of which appear to be specifically identical with the type, which is found
in Japan, California, the West Indies, and on the eastern coast of the United
States. I have not been able to find the ossicle in all the specimens of this
species examined, even when the dry remains of the animal were in the shell.
But this may be accounted for by its form and extreme narrowness, which,
added to its position on a ligament which must be broken to open the shell,
render its loss extremely probable except under peculiarly favorable circum-
stances.

(?) EuciROA (Dall, 1878, Bull. M. C. Z., IX., 1881). Shell large, thick,
Cvthereae-form, with a true lunule and corselet; ossicle triangular, wider
behind; soft parts unknown. Type V. elegantissima Dall.

This species is separated chiefly by its form, the hinge being essentially like
that of Trigonulina, except as regards the ossicle, which from the arrangement
of the ligamentary scars must have been of a wide subtriangular shape. The

* These are probably what A. Adams took for palpi in his description of the
soft parts of V. Japonica or Deshayesiana.

MUSEUM OF COMPARATIVE ZOOLOGY. 287

fragments indicate that it must reach the largest size of any member of this
group.

Pecchiolia (Meneghini, 1851, Cons. suUa Geologia Stratigr. della Toscana,
p. 180). Shell destitute of lunule ; much inflated ; beaks much coiled and
widely separated ; line of the external ligament passing under and coiled with
the beaks ; ossicle (not seen, but must have been enormous and like that of
Verticordia, but flatter) ; right valve, hinge as in Verticordia except for the
depressed lunular space ; left valve ditto, but with an obtuse thickening of
the cardinal margin to fit behind the large cardinal tooth of the right valve.
Type P. argentea Mariti, fossil in the Vienna basin and Italian tertiaries. No
recent representative known.

This and other sections of Verticordia have been referred to Hipjjagus (Lea,
1834). But from a study of authentic specimens received froui Dr. Lea, I find
the type of Hippagus to belong to the Mytilidoe. The external sculpture, the
position of the muscular scars and ligament, and the general form and texture
of the shell in Hippagus agree perfectly with Crenella, from which Hippagus
difi"ers only in having the beaks slightly more twisted than the common recent
species of Crenella, and in having the shell thicker, so that the external sculp-
ture does not crenulate the edge. Compare the original figure of Hippagus
with Modiola cinnamomea Chemnitz, or a specimen with a shell of equal size
of Crenella decussata, and it is easy to come to a conclusion.

I have been able to study a fine series of Pecchiolia in the Jeff"reys collection,
and observe that one must be certain of the perfection of the specimen if he
would avoid being led into error, for the fossil shell scales off in such a way as
to alter the hinge entirely, and yet not appear defective. The scars of the in-
ternal cartilage are wonderfully large and deep, and if, as is probable, it bore
an ossicle , this must have been enormously large.

Haliris Dall. Shell globose, ossicle short, squarish ; lunule present, not
deep ; right valve with hinge teeth as in Trigonulina ; left valve with (in the
adult) a small but distinct cardinal tooth and a short stout lateral tooth near
the umbo ; lunule not produced ; adolescent or young shells with the dentition
obscure or imperfect. Type Verticordia Fischeriana Dall. Gulf of Mexico in
deep water.

To this group belong V. trapezoidea Seguenza, and perhaps his V. granulaia,
which I have not seen, and which has not been well figured. These have been
perhaps too hastily united by Dr. Jeff'reys.

The genus Mytilimeria, whose position has been somewhat debated, unques-
tionably belongs in this family. The mantle is closed except for the two large
siphonal openings, which are, on a large scale, similar to those of Lyonsiella,
and a minute slit for the very small phalliform foot, which has a shallow pos-
terior sulcus and a slightly enlarged glandiform tip. The mantle margin is
thick and plain except about the siphons. There are two rather good-sized

288 BULLETIN OF THE

branchiae, but the palpi art small, or perhaps obsolete ; on this point my speci-
men, which had been dried and soaked out, was not conclusive. The internal
cartilage bears a large ossicle as in Lyonsiella ; the beaks are subspiral, and the
external ligament coils under them away from the hinge-margin, as in Ceto-
concha and other Verticordice. The same uncircumscribed lunular depression
and puckering of the adjacent hinge-margin is observable in Mytilimeria, which
normally should have, from appearances, a tooth-like projection of the twisted
margin as in Phctodon, but which is usually masked by the ligament. The
sedentary habit of this mollusk, as in other cases of the kind, has resulted in
veiling some of its original characters. It is said to " burrow in sponge," but
the spongy substance in which it is found is of its own manufacture, and con-
sists of sand grains, etc., entangled in a solidified mucus, which is secreted by
the animal, and which, like the byssus, is not affected by water. It is proba-
ble that in life the foot is susceptible of much extension.

The surface of the valves in Mytilimeria is not granulated, but is covered by
an epidermis not unlike that of Cetoconcha. It may be recalled that sundry
species of Verticordia, Poromya, etc., have a habit of attaching sand grains to
themselves by a sort of mucous secretion. The pallial line is barely sinuated,
and the shell is thin and internally pearly, though the iridescence is veiled as
in Cetoconcha by a non-pearly stratum. It is quite certain that some of the
species referred to Mytilimeria would be better placed elsewhere.

Lyonsiella M. Sars (1872; Pecchiolia G. O. Sars after Jeffreys; Lcevicordia
Seguenza, 1876). Shell small, thin ; lunule faint or none; ossicle semicylin-
drical, forked behind; external ligament almost none; right valve edentulous,
lunular edge a little produced and thickened; left valve with an elongate ob-
scure thickening of the hinge-margin under the beak. Soft parts described by
Sars (Rem. Forma of Animal Life, p. 25, pi. iii. figs. 21-43, 1872). Type L.
abyssicola Sars. North Atlantic.

To this group belong L. insculpta Jeffreys (1875 -|- L. gemma Verrill, 1880),
and probably the following species, known to me only by single valves or by
description : L. angulafa Jeffreys (as Pecchiolia), L. mytiloides and azinoides
Seguenza (1876). The identification of L. gemma with the earlier described
insculpta is from authentic specimens; the figure of insculpta in P. Z. S., 1881,
is not very characteristic.

The following species form part of the Blake collection.

Verticordia acuticostata Philippi.

Hippagus acuticostatus Phil. Moll. Sicil., 1844.

Verticordia acuticostata Dall, Bull. M. C. Z., IX. p. 105, 1881.

Habitat. Station 31, in 84 fms.; living, at Station 5, in 152-229 fms., coral
ooze, off Cuba, in the Gulf of Mexico, bottom temperature 49°.5 F. ; Barba-
dos, 100 fms.

MUSEUM OF COMPAKATIVE ZOOLOGY. 289

Fine and very large specimens of this species have been dredged by the Fish
Commission in the Gull' of Mexico.

An account of the soft parts will be found under the discussion of the generic
characters.

Verticordia "Woodii Smith.

Venicordia Woodii Smith, Chall. Rep. Lam., p. 168, pi. xxv. figs. 7-7 b, 1885.

A fragment, probably belonging to this species, was dredged in 100 fms. at
Barbados by the " Hassler " in 1871.

Verticordia perversa Dall.

Shell translucent, thin, small, very much the shape of Area pednnculoides,
but wuth the beaks turned toward the larger end, equivalve, inequilateral,
waxen white ; surface covered "with excessively minute shining elongated
granules radiating from the beaks in single series with equal interspaces; there
are also between twenty-five and thirty radiating lines, on which extraneous
matter, especially sand grains, seems to stick, though I can perceive no epi-
dermis or special formation which should induce the sand to stick on these
lines and not between them; there are no ribs under them. All these lines
curve forward with an even sweep ; if we consider the hinge-line as horizontal
the anterior margin will be nearly vertical, and their junction evenly rounded,
with no trilobate appearance in the general outline; the highest part of the
shell is anterior; from the rounded anterior basal edge the posterior basal edge
rises at an angle of 45°, or so, toward the rounded junction with the hinge-
margin ; the beaks are rather low, the area about them is full but not inflated;
there are no keels on the shell; the lunule is small but indented, and its inner
edge in the right valve is convex and fits into a concavity in the opposite valve;
behind it in the right valve is a short but stout squarish tooth, on the other
side of which is the ossicle longer than wide, and indented behind ; there are
no teeth in the left valve ; the interior is polished, the granules showing
through the shell. Max. Ion. 5.0; max. alt. 5.0; diam. 3.0 mm.

Habitat. Dredged by the U. S. Fish Commission steamer "Albatross" at
Station 2678, oflf Cape Fear, North Carolina, in 731 fms., bottom temperature
38°. 7 F.

This little shell is remarkable for having its height and maximum diameter
thrown forward, as in Lyonsiella gemma Yerrill, but even more so, without
being lobed as are several species in which a similar tendency is indicated but
not carried out; it is so rounded withal that it looks like a short Modiola with
the beaks turned the wrong way. I have seen no other species which resembles
it at all.

VOL. III. — NO. 6. 19

290 BULLETIN OF THE

Verticordia Seguenzae Dall.

Shell having nearly the form of V. australiensis Smith (Chall. Lam., p. 167,
pi. XXV. figs. 6-6 b), with thin rather convex valves, greenish color, and about
forty radiating posteriorly convexly curved faint sulci, the interspaces between
which are gently rounded but little elevated, and hardly to be called ribs, and
have intercalary groovings toward the margin. The surface is covered with
minute glassy grains arranged with some regularity in radiating and concentric
series. As compared with V. australiensis the anterior dorsal margin is more
elevated and rounded up; the posterior margin is less curved. V. trapezoidea
Seguenza has the posterior dorsal margin much more curved, the anterior more
oblique, and the hinge is different, putting it in another section of the genus.
In V. SegxLenzcB the ossicle is very small, flatter than in the typical species,
rectangular, and wider behind; the length of the most perfect valve is 5.0, the
altitude 4.0, and the diameter, taken as twice that of the single valve, would
be 3.5 mm. A large dead valve, perhaps of this species, was found at Station
2602.

Habitat. Yucatan Strait, 640 fms., one valve; U. S. Fish Commission,
thirty-six miles south of Cape Hatteras, at Station 2602, in 124 fms., and about
the same distance southeast of Cape Lookout, N(;rth Carolina, at Station 2614,
in 168 fms., bottom in both cases sand, and bottom temperature about 61° F,
(three valves).

Although there is but little material, yet the species does not come very
close to any of those with which I could compare it, and it seemed worthy
of a name. The hinge, though delicate and with small teeth is that of the
typical Verticordia.

Verticordia (Trigonulina) ornata D'Orbignt.

Verhcm-dia ornata Dall, Bull. M. C. Z , IX. p. 105, 1881.

Trigonulina ornata D'Orb., Moll. Cuba, II , p. 292, pi. xxvii. figs. 30-33, 1846.

Verticordia ccelata Verrill, 1884 ; Trans. Conn. Acad., V. 566 ; VI., pi. xxx. figs. 9, 9 a.

Habitat. Barbados, 100 fms. ; Station 19, 310 fms. (Catalina Island, Cal.,
16 fms., Dall; Jamaica, W. I., D'Orb. ; China Seas, Adams ; east coast of the
United States off the Carolinas, and northward as far as Station 949, off Mar-
tha's Vinej'ard, in 100 fms., U. S. Fish Commission).

The sculpture of this elegant species -is composed of curved ribs, radiating
from the umbones and crowded in front with two or more gaps behind. There
may be a posterior rib forming the extreme margin, or the hindermost rib may
be within the margin, two cases figured by D'Orbigny; or the posterior ribs
may fail altogether, forming the variety ccelata Verrill. The ribs may all or in
part be grouped in pairs, or the pairs may resemble a wide rib deeply grooved
along its summit. The ribs may be high and strong, or low and uniform ; in

MUSEUM OF COMPAKATIVE ZOOLOGY. 291

the latter case they are usually more numerous ; the number of ribs may be
from eight to twelve ; I have not in over one hundred and fifty valves found
more or fewer. The interspaces may be irregularly granulose or finely radi-
ately striate. The degree of compression varies somewhat, and the posterior
end may be more or less angulated below. The extremes seem very unlike,
but are perfectly connected by intermediate specimens. The average specimen
has six to eight anterior ribs, a gap, one or two ribs, a wider gap, and one or
two more ribs near or at the posterior margin. The commonest form has
A 6, 0, 1, 0, 1 P, for its rib formula.

Verticordia (Euciroa) elegantissima Dall.

Verticordia elegantissima Dall, Bull. M. C. Z., IX. p. 106, 1881.
Plate II. Figs. 1 a, 1 b.

Habitat. Station 16, 292 fms. (fragments), oflf Havana, Cuba, and Station
18, in Lat. 23° 7' and Lon. 82° 43' 30" W., oflf Muriel, Cuba, 756 fms. (a single
valve) ; bottom temperature 55^.7 and 40°. 0 F. respectively.

A little more material in regard to this large and elegant species has come to
hand. The perfect valve represented in the figure was carelessly cracked by
the artist under the object-glass of his instrument. It has been tolerably re-
paired, but fresh and perfect specimens ought to be found by somebody. The
Fish Commission has dredged some imperfect valves (winter of 1885-86) at
Stations 2659 and 2660, off Cape Canaveral, in 509 and 504 fms., bottom tem-
perature 45°. 2 and 45°.7 F. respectively. The best of these is 40.0 long, 38.0
high, and -3^- mm. in diameter, showing a more rounded form than in the young
valve figured, and being the largest known species of the family. The interior
is strongly radiately striate toward the pallial line, and the scar of the pedal
muscle strongly marked.

Verticordia (Haliris) Fischeriana Dall.

Verticordia Fischeriana Dall, Bull. M. C. Z., IX. p. 106, 1881.
Plate II. Figs. 4 a, 4 b.

Habitat. Barbados, 100 fms.; Sigsbee, off Cuba, 119 fms.; and, living, at
Station 36, Lat. 23° 13' N., Lon. 89° 16' W. Gr., in the Gulf of Mexico, in
84 fms., bottom temperature 60°. 0 F. A single valve was dredged off Rinal-
do's Chair in 160 fms. by the Porcupine expedition, 1870, and was found
mixed with specimens of V. trapezoidea Seg., in the Jeffreys collection. It
has also been dredged by the U. S. Fish Commission at Stations 2600, 2601,
and 2602, in 87-124 fms., about thirty-six miles off Cape Hutteras.

The ribs of this species vary in sharpness, and when very sharp are serrate
by the granulations which on the keel of the ribs become more prominent than

292 BULLETIN OF THE

elsewhere. This would seem to be nearest to V. granulata Seguenza, in my
opinion unwisely combined by several authors with V. trapezoidea. V. granu-
lata is as yet but imperfectly known, but seems to have only about half as
many ribs as the present species.

Family CUSPIDAEIID^ Dall.
Genus CUSPIDARIA Nardo.

Cuspidaria Nardo, Revue Zoologique, Jan. 18i0, p. 30. (In report of the meeting

of the Congres Scientifique at Pisa, paper read Oct. 11, 1839.) Type Tellina

cuspidata Olivi.
Necera Gray, in Griffith's Cuvier, legend to plate xxii. fig. 5 (dated 1833, whole

volume issued in 1834. Type N. chinensis Gray, in Index, p. 698) ; Synops.

Brit. Museum, 1840, fide Gray in P. Z. S., 1847.
Not Neara Robineau-Desvoidy, Essai sur les Myodaires, 1830.

Valves with one or more teeth.

The name Necera being preoccupied in entomology, the next name, Cuspi-
daria, must necessfirily be adopted. The longer an untenable name is retained,
the more inconvenience results to science when it is, as it always will be,
eventually overthrown. Gray's name at any rate has a very slender claim to
priority, as the genus is not mentioned in the text or described anywhere, the
generic name is mis-spelled in the index of plates to Griffith's edition of Cuvier,
and we are left to conjecture who is its author. Gray himself, in Agassiz's
Nomenclator, only claimed it from the Synopsis?, which I have not been able to
examine, and which was quite likely subsequent to the publication of Nardo's
diagnosis in the January number of the Revue.

The group has been reviewed by Arthur Adams, who has proposed several
generic or subgeneric names and eliminated some incongruous species. Dr.
Jeffreys has also made a division into sections based on the sculpture of the
shell. Lastly, Mr. Edgar A. Smith has most carefully investigated the charac-
ters of a large number of species, especially with regard to the hinge, arranged
them in lettered sections pending further study, and tabulated the results. To
this I am much indebted for help. It will not be necessary for our purposes
to review the entire group, but merely those sections of it which contain
species represented in the Blake collection, or which are in some way aflFected
by this investigation.

The new subdivisions here instituted appear as proposed by Dall and Smith ;
a course taken with Mr. Smith's permission, and which I have felt to be due
to him, owing to the assistance I have derived from his valuable observations
on this group.

The shells of Cuspidaria possess an internal ligament, received in each valve
in a more or less differentiated groove or fossette, which may project from the

MUSEUM OF COMPARATIVE ZOOLOGY. .293

ixmbonal angle of the hinge-margin, or be more or less adherent to the anterior
or posterior slope of this angle. They may have one anterior and one posterior
cardinal and lateral tooth in each valve, any one of which (or all in the genus?
Myonera) may be entirely absent. Beside the teeth the hinge is reinforced in
many cases by a buttress extending in a direction vertical to the valve from
the hidden surface of the hinge-margin, posterior to the umbonal angle. This
buttress may consist of the vertical plate above mentioned and a thickened rib
curving round in front of the posterior muscular scar, and then directed pos-
teriorly, becoming almost immediately obsolete. Or the posterior muscular
insertion may be elongate and narrow, and the buttress take the form of a
" clavicle " or myophore, elongated, parallel with the posterior hinge-margin
and separating the two posterior muscular scars. The muscles are not always
inserted upon the buttress, but may be above and in front of it. Its purpose
would seem to be that of strengthening the valve, almost always thin and
fragile, against sudden contractions of the muscles, and to support the cardinal
border, and especially the strong posterior lateral tooth found in many species.
When this tooth is found in a species which has no posterior lateral in the
other valve, the valve which has a tooth shows the buttress stronger than the
other, indicating its function as a support for the tooth; but when elongated
and clavicular there is little difference between the buttresses of opposite valves,
indicating that in such cases the function is the strengthening of the valve
itself. The presence of the buttress is, in my opinion, important only in a
minor degree, except when it takes the clavicular form; as in different species of
the same group, and even in individuals of the same epecies, its size and promi-
nence vary very greatly. Adriatic specimens of the typical species, C. cuspi-
data, show a strong buttress; British specimens of the same species often show
it faintly or not at all, while otherwise well developed. The names Necera,
Ehinomya, Aulacophora, Spathophora, and Tropidophora, among those which
have been applied to members of this group, by Gray, Adams, and Jeffreys,
are all preoccupied in zoological nomenclature, some of them several times
over.

The characters of radiating and concentric sculpture in this group have no
more than a specific value; there are few species where they are not more or less
combined in the external ornamentation. The surface may be polished, smooth,
wrinkled, sulcate, or granulous. The anterior muscular scar is double or single,
the posterior scar double, in all the specimens I have seen where the scars could
be made out.

The outer part of the scar in each case is due to the adductor of that end of
the animal, the other part to the insertion of the sphincter-like muscular band
described under Myonera paucistriata, further on. The observations made on
the anatomy of several species will be found at the same place. If the writer
has not been misled by contraction of the parts under the action of alcohol, the
group comprising Cuspidaria and Myonera would seem to be destitute of gills
or palpi, at least in the normal form of such organs. This, however, may not

294 BULLETIN OF THE

be true of all species, for the gills of mollusks are largely mere modifications
of the cutis and very mutable structures.

Since the publication of ray preliminary paper on the Blake Mollusks, I
have had the advantage, for the Pelecypods especially, of being able person-
ally to examine the species comprised in the Jeffreys collection, including
specimens of the deep-water forms known to me previously oaly by description.
In this way, and by the aid of the valuable labors of Mr. E. A. Smith, pub-
lished in the Challenger report, I have been able considerably to improve upon
previous work, in a manner which would have been impracticable under other
circumstances. Otherwise, in many cases I coflld express opinions, on the
present group especially, only with reserve not unmixed with doubt.

The sections or subgenera represented in the Blake collection now follow.

SUBGENDS CUSPIDARIA, 8. 8.

Shell usually concentrically sculptured, with or without a buttress; fossette
posteriorly inclined and attached to the hinge-margin by its posterior edge;
one posterior lateral tooth in the right valve.

The following species in addition to those examined by Smith belong here:
C- arctica Sars, C. glacialis Sars, C. lamdlosa Sars, C. exigua Jeffreys, C. Jef-
freysi Dall. C. limatula Dall {contracta Jeffreys) has no teeth. C. arcuala
Dall, being described from a toothless left valve, is still in doubt.

A valve of C. glacialis Sai-s, dredged in 1467 fms. in the Gulf of Mexico by
the U. S. Fish Commission, is 45.0 mm. long, 28.0 mm. high, and the com-
plete shell must have been about 30.0 mm. in diameter.

Cuspidaria rostrata Spengleb.
NeoRra rostrata Spengler, Dall, Bull. M. C. Z., IX. p. Ill, 1881.

Habitat. Barbados, 100 fms. ; Station 36, 84 fms.; Sand Key, 80 fms.

No additional specimens of this species have turned up in the Blake collec-
tion since the above were noted. It has been dredged in from about 65-500
fms. in various localities from the West Indies northward. Among the Mu-
seum specimens are two very large dead and rather worn valves from Stations
2659 and 2660 of the U. S. Fish Commission, off Cape Canaveral, in 509 and
504 fms., bottom temperature 45°. 2 and 45°. 7 F. respectively. They differ
from the typical form in ha\'ing the rostrum proportionally more slender,
shorter, and curved, the dorsal margin being quite concave, and especially in
having the dorsal surface very wide and with a wide strongly marked de-
pressed area extending from the beak to the tip of the rostrum, smooth or
longitudinally striate. The surface of the valves was dark brown apparently,
and strongly concentrically striate. The valves are proportionally much more
compressed than in the usual form of C. rostrata. The dimensions of the best

MUSEUM OF COMPARATIVE ZOOLOGY. 295

preserved valve are as follows: — Lon. of shell 45.0; of rostrum 16.0; max. alt.
of shell at right angles to the average plane of the hinge-line 26.0; alt. of ros-
trum at its middle part 4.0; double diameter of valve 18.0; of rostrum 8.0;
of rostral area 4.5 mm. For this variety (which may well prove with better
material to be a distinct species) I would propose the name microrhina. The
valves are extremely thick and internally radiately striate, the pallial sinus,
muscular scars, and buttress well marked, and fossette strong, ovoid and nearly
vertically directed.

Cuspidaxia Jeffireysi Dall.

Necera Jeffreysi Dall, Bull. M. C Z., IX. p. Ill, i881.
Plate m. Fig. 9.

Habitat. Off Cape San Antonio, a fragment in 1002 fms.; Station 44, 539
fms. ; Yucatan Strait, 640 fms.; Station 136, in 508 fms., off Santa Cruz ; and
Station 230, off St. Vincent, in 464 fms., living, bottom temperature 41°.5 to
42°.0 F.

This species is characterized especially, and distinguished from rostrata and
its other close allies, by the straightness of the hinge-margin and the conse-
quent elevation of the anterior dorsum, by the minute fossette, and by the
greatest ventral expansion being almost directly under the beaks. The pro-
portional length of the rostrum varies in different specimens, but it is less than
in obesa, and usually less than in rostrata.

Cuspidaria obesa Lov^n.

Necera obesa Loven (1846), Verrill, Trans. Conn. Acad., V. p. 563, pi. xllv. fig. 10 c,

1882.
N. pellucida Stimpson, Inv. Grand Manan, 1853.

Habitat. Barbados, 100 fms.

A single specimen was found as above, and seems conspecific with the mori.
northern specimens dredged by the Fish Commission. I have also dredgec
this species in 16 fms., mud, near Catalina Island, off Santa Barbara County.
California : in companv with Plectodon scaler Cpr. and Cardiomya califomica
Dall.

The valve figured (Plate III. fig. 1) is supposed to belong to this species,
though showing some points of difference on which more material is needed to
decide.

296 BULLETIN OF THE

Cuspidaria(?) axcuata DaU.

Ne(Bra arcuata Dall, Bull M. C. Z., IX. p. 113, 1881.

Flate III. Fies. 3, 4.

Habitat. Yucatan Strait, 640 fms.

As only a single left valve of this species was obtained, it has been imprac-
ticable to determine to what section it belongs, but the general aspect is that
of Cuspidaria, or a Balonympha without a clavicle.

Subgenus CARDIOMYA A. Adams.

Similar to the preceding, but with radiating sculpture and the fossette more
vertical and prominent. Type Necera Gouldiana Hinds.

C. multicostata V. & S., C. perrostrata Dall, and C. ornatissima D'Orbigny
(-\- costata Bush) belong here, as do N. pectinata Cpr. (1865, + var. behring-
ensis Leche, 1883, from types), and C. califomica Dall.* (See Plate III.
fig. 6.)

Cuspidaria (Cardiomya) perrostrata Dall.

Necera (ornatissima D'Orb. var.?) perrostrata Dall, Bull. M. C. Z., IX. p. 110, 1881.
Necera perrostrata Verrill, Trans. Conn. Acad., V. p. 661, 1882.

Plate II. Figs. 3 a, 3 b.

Habitat. Station 43, 339 fms., off Tortugas, bottom temperature 45°.0 F.;
Station 264, in 416 fms., gray ooze, near Grenada, bottom temperature 42°.5 F.

Desiring to be as cautious as possible in describing new forms based on very
little material, I referred to this and D'Orbigny's C. ornatissima as possibly
identical, in my preliminary descriptions. That there is a good deal of varia-
tion in this group is clear, when plenty of material is accessible; but I am the
more confirmed in the belief that this one is specificallj' distinct by Professor
Verrill's opinion, and by the fact that the Fish Commission has obtained off
the Carolinas, and Miss Bush has described (as Necera costata), a form which
seems to be identical with D'Orbigny's, and is certainly distinct from the

• Cuspidaria (Cardiowyn) cnlifornica Dall. Shell differing from C. pectinata by its
smaller size and proportionally greater length ; larger number of ribs (16-20, wliiJe
pectinata averages 12-14) ; its straighter, longer rostrum with but two strong radi-
ating lirae extending to the lower extreme {pectinata has none, or only several fine
ones near the body of the valve) ; its less inflated shape and paler more delicate
epidermis. Lon. of sliell 7.0; of rostrum 2.5; alt. of shell 3.6; diam. 2.75 mm.
Color yellowish white ; ossicle as usual; buttress present in the right valve.

Habitat. Catalina Island, California, dredged in 16 fms., mud ; Dall, and pre-
viously Cooper, who confounded it, following Carpenter, with pectinata.

MUSEUM OF COMPARATIVE ZOOLOGY. 297

present one. Had Miss Bush in her excellent paper had more material, she
would probably have hesitated to give a name to the pretty species she has
called costata. Her distinctions from ornatissima are that the ribs are less
numerous, more curved, and the shell less convex in the Carolina specimens.
I find in her figure eight visible ribs; in seven valves from the Carolina coast
I find the ribs varying from five to seventeen; the strong ones extending to the
beaks number from five to eight; their curvature varies somewhat. The di-
ameter of D'Orbigny's figure relative to its height is as 11 : 14, while in Miss
Bush's specimens it is, she states, as 4 : 4, so that her specimens were really
more convex than D'Orbigny's, rather than less so. But his figures, made in
1840 or so, and much magnified, must not be construed too literally, as they
are on the face of them a little formal, though excellent for the time.

Cardiomya costellata Deshates.

Corhula costellata Desh. Expl. Sci. Morea, Geol., p. 86, pi. vii. figs. 1-3, 1837.
Necera costellata Jeffreys, Brit. Coneb., III. p. 49 ; V. p. 191, pi. xlix. fig. 3 ; P. Z. S ,

1881, p. 944.
Neoera curta Jeffreys (name, no description), Valorous Moll., Ann. Nat. Hist., 1876,

p. 495 ; P. Z. S., 1881, p. 943, pi. Ixxi. fig. 10.
Sphena alternata D'Orbigny, Moll. Cuba, II. p. 286, 1846 ; Atlas, pi. xxvii. figs.

17-20, 1845.
? Necera alternata (D'Orbigny) Dall, Bull. M. C. Z., IX. p. 110, 1881.

A fine series of specimens in the Jeffreys collection, especially from the
Mediterranean, is sufficient to convince the most sceptical of the great vari-
ability of this species. It varies from smooth, or with but two or three radi-
ating costse, to completely radiated all over; the rostrum varies in actual and
in relative length and direction; the amount of inflation, its direction, and
consequently the outline of the shell, vary considerably. The European speci-
mens sometimes have a smooth interval between the end of the rostrum and
the radiating sculpture of the body, and sometimes the whole is covered' with
radii. The most common form seems to be that in which there are compara-
tively few and rather strong radii on the posterior part of the shell, with the
rest smooth or faintly radiated, and the rostrum smooth, except a few radii on
its dorsal side, and rather long. This form has been collected by Hemphill in
two fathoms at Marco, Florida, and has been dredged by the U. S. Fish Com-
mission at Stations 2597, 2602, and 2614, off the Carolina coast. These are all
small, Jeffreys' finest British specimens being about 10 mm. long, and the
average length of those from all localities being about 6-7 mm. The form
named curta by JeS"reys (which may rank as a variety though connected by
indefinite gradations with the type) is also small, and has the rostrum short
and recurved, the striation strongest posteriorly but varying, as in the type.
Some of the specimens dredged by the U. S. Fish Commission at Station 2602
were of this variety.

298 BULLETIN OF THE

Cardiomya costellata, var. corpiilenta Dall.

Plate HI. Fie. 9.

The variety corpuUnta Dall is like a giant curta striated all over, and about
15.0 mm. long. A valve was dredged at Station 5, in 229 fnis., and a frag-
ment at Station 228, near St. Vincent, in 785 fms. It differs from C. striata
Jeffreys in the shorter and less differentiated rostrum and the alternate!}' larger
and smaller radii, which are also more distant and sharper, while the concen-
tric striae are much less evident. Still, in the type these characters intergrade,
as they might be seen to do here if we had specimens enough to compare.

Cardiomya striata Jeffreys.

Necera striata Jeffreys, Valorous Moll., Ann. Nat. Hist., Dec. 1876, p. 495 ; P. Z. S.,

Nov. 1881, p. 944, pi. Ixxi. fig. 11, 1882.
? NeoEra aJternata D'Orb., var., Dall, Bull. M. C. Z., IX. p. 110, 1881.
Necera multicostata Verrill and Smith, Trans. Conn. Acad. Sci., V. 659, pi. Iviii.

fig. 40 ; Proc. U. S. Nat. Mus., III. p. 398, 1880.

Plate m. me. lo.

Habitat.- Station 36, 84 fms.; Station 5, 152 fms.; U. S. Fish Commission,
off the Carolina coast, Station 2601; off Martha's Vineyard, Station 1038, etc.;
off Newport, Rhode Island, Station 874, and, others.

This fine shell differs from some of the varieties of C. costellata only in size.
It bears the same relation to them that the var. corpulenta does to the var.
curta. But taken by itself it seems so distinct that I have concluded to leave
it separate for the present. It should be stated that Dr. Jeffreys' remark as
to the radiation not being coarser posteriorly, is correct only for the one or two
specimens first obtained, and even in them it is only partially exact. The vast
majority have the sculpture decidedly stronger toward the rostrum. I may
also add, that none of the specimens in the Jeffreys collection at Washington
have the rostrum quite as straight as in the figure in the P. Z. S. It is a little
upturned in all of them, though the particular specimen figured may not have
had that peculiarity.

There is every probability of the correctness of Prof. Verrill's observation:
" Perhaps all these forms may eventually prove to be varieties of one species."
(Trans. Conn. Acad. Sci., V. p. 560, 1882 )

Subgenus LEIOMYA A. Adams.

Leiomya A. Adams, Ann. Mag. Nat. Hist., 1864, p. 208.

An anterior prominence or cardinal tooth in each valve, anterior and pos-
terior laterals in the right valve, left valve without laterals. Cartilage in a

MUSEUM OF COMPARATIVE ZOOLOGY. 299

posteriorly directed groove or fossette ; surface smooth or concentrically sculp-
tured. Type Necera adunca Gould.

This is equivalent to Smith's Section F. It would include, according to his
description, Neara Brazieri Smith. I have compared specimens of N. adunca
Gould, received from Drs. Arthur Adams ^d Gould, which agree perfectly with
Adams's description of Leiomya. The cardinal tooth in the right valve is bifid
at the tip and very small ; hence Adams in his diagnosis ascribed two cardinal
teeth to this valve, but I think they should be counted as one.

What the shell is, described by my friend Smith as type of his Section J,
under the name of Necera adunca Gould, I do not know. He has evidently
been misled by a wrongly named shell. It is certainly an entirely different
species and section from Leiomya. It has no cardinal teeth, a small central
fossette, a small thickish anterior and posterior lateral in the right valve, and
a similar anterior lateral (only) in the left valve; the surface is finely ridged.
It appears to be the only species with these characters, and I would sug-
gest the name of Vulcanomya Smithii for it in default of any other legitimate
designation. Its external characters and size closely resemble those of the
genuine N. adunca Gld., which would account for the error, in the absence
of types.

Mr. Smith kindly informs me that he has re-examined the specimens, and
finds nothing to change in his description of them. They were received at the
British Museum with Gould's name attached by some one unknown.

Section PLECTODON Carpenteb.

Plectodon Carpenter (Suppl. Rep. Brit. As., p. 638, Aug. 1864) is closely re-
lated to Leiormja. It diifers in the insertion of the cartilage behind and under
the beaks, instead of on the hinge- margin or in a fossette; in having, rather than
a true tooth upon the margin, a tooth-like prominence formed by the spiral
twisting under the beaks of the hinge-margin itself, upon and over which, in
P. scahcT, there is a minute external ligament; lastly, in Plectodon there is a
granulated surface much as in Poromya. The pallial sinus appears to be about
the same in both, and the tips of ^he siphons are protected, in both groups,
as in Schizothcerus, by a leathery ring, flattened and broadened at the sides.
Until recently only two right valves of Plectodon were known, but in 1873 I
dredged at Catalina Island, California, in 16 fms., mud, some half a dozen
living specimens, which have enabled me to make a careful comparison with
my Necera granulata. There can be no doubt of their generic identity, and
even considered as species they are very similar, the intwisting of the margin
being less marked in granulata and the supposed external ligament obsolete.
I regard Plectodon, therefore, as a mere section of Leiomya, which might also
include Rhinoclama, which is of about equal value with Plectodon.

300 BULLETIN OF THE

Leiomya (Plectodon) granulata Dall.

Neoera granulata Dall, Bull. M. C. Z., IX. p. Ill, 1881.
Plate III. FlB. 8.

Habitat. Off Sombrero, 54 and 72 fms. ; Barbados, 100 fms.; Station 177,
off Dominica, in 118 fms., sand, bottom temperature 65° .0 F., and a fragment
at Station 272, in 76 fms., coarse sand, off Barbados, bottom temperature
64°.75 F.

The longest specimen measured 18.0 mm., and is pure white. A variety
velvetina has finer and much closer set granules, more recurved rostrum, and
less sinuated posterior ventral margin; the concentric striae are also a little
more prominent.

The granules are arranged along the lines of growth, and sometimes are
elongated in that sense.

Section RHINOCLAMA Dall and Smith.

Like the last, but without cardinal teeth. Bhinomya A. Adams, not Des-
voidy or Geoffroy. Sections F and G, Smith. Type Necera philippincnsis
(A. Adams) E. A. Smith.

Necera notabilis Jeffreys and Necera semistrigosa Jeffreys, not lamclli/cra
Dall, belong here. The last, which is clo.sely simulated as to external char-
acters by semistrigosa, proves on comparison to have different hinge characters.

Necera teres Jeffreys, injiata Jeffreys, and gomerensis Smith, in so far as they
represent this type of hinge, belong hereabouts; though Dr. Jeffreys in his col-
lection had mixed these with other forms not closely related.

Leiomya (Rhinoclama) halimera Dall.

This species, represented by two right valves, is probably correctly placed
here. It may best be described comparatively.

Though larger than any specimen of Z. notabilis Jeffreys, in our collection,
it is of exactly the same shape and proportions; except that, instead of being
sculptured like L. notabilis, it has the finely concentrically striate and wrinkled
surface of L. teres Jeffreys. The hinge has the high and squarish laterals and
other features of L. teres, but the anterior lateral is a little longer, the space
between the laterals longer, the groove for the cartilage less marked, and the
buttress less strong. Lon. 10.0; alt. 6.5; diameter of largest valve about 3.0,
and of the valves when perfect probably 6.0 mm.

The specimens were dredged by the U. S. Fish Commission steamer '* Alba-
tross," at Station 2678, in 731 fms., ooze, off Cape Fear, North Carolina, bottom
temperature 38°. 7 F., in the winter of 1885-86.

MUSEUM OF COMPARATIVE ZOOLOGY. 301

The valves are of a yellowish white, and one of them shows remnants of a
thin but rather tough greenish epidermis.

SDBGEND8 TROPIDOMYA Dall and Smith.

An anterior cardinal tooth in each valve; no lateral teeth; cartilage as in
Leiomya. Tropidophora Jeffreys, not Troschel nor Thompson. T;ype Necera
abbreviata Forbes.

This is Section I of Smith. The type has the buttress fairly developed and
chiefly concentric sculpture.

Subgenus hALONYMPHA Dall and Smith.

An acute cardinal tooth in right valve; no other teeth in either valve; a
clavicular rib extending posteriorly in both valves, fossette small, central;
surface concentrically striate or smooth. Type Necera claviculata Dall. Sec-
tion K of Smith, who places here Necera injlata Jeff"reys and N. congenita
Smith.

The latter appears different from anything I have seen. N. injlata has been
in some confusion. The specimens so marked in the Jeffreys collection in
Washington are of two kinds. One valve from " Porcupine expedition, 1870,
St. 16, 17 a," is a left valve of Halonympha claviculata Dall, fitting almost ex-
actly the right valve which served as my type. Those from " off Gomera,
Chall. exp.," and " Porcupine exp. 1869, St. 39," are Rhinoclama teres Jeffreys.
Whether there is an injlata not represented in the Washington series I do not
know; the figures in P. Z. S., 1881, pi. Ixxi. figs. 2, 8, in which the differences
seem a little strained, might both have been made from varieties of teres in the
collection. Smith notes something of the same kind in his description of
N. teres (N. gomerensis of references to Plate X.) in the report on the Challenger
bivalves, p. 50.

Halonympha claviculata Dall.

Neara claviculata Dall, Bull. M. C. Z., IX. p. 112, 1881. Smith, Chall. Rep. Lam.,

p. 52, not pi. ix. figs. 8 - 8 b.
Neoera injlata, Jeffreys, P. Z. S., 1881, p. 942. 1882; (partly).

Plate II. Figs. 8, 2 a.

Habitat. Station 44, 539 fms., one valve; Sigsbee, off Havana, 450 fms.
(]) fragment; Porcupine expedition, Atlantic Ocean, Station 16, or 17 a, 1870;
Challenger expedition. Station 33, in 435 fiithoms, coral mud, near Bermuda.

In the Porcupine specimen it is clearly to be seen that the posterior muscle
was planted on the upper surface of the clavicle, which is therefore in this case
a myophore as well as a buttress.

302 BULLETIN OF THE

Smith in his text correctly describes the peculiarities of this species, but the
figures on Plate IX., referred to as Neoera claviculata, though intended as rep-
resentations of this species, have not a clavicle ! It seems that the " artist,"
though warned, took it upon himself to omit this feature (!), and succeeded, in
spite of Mr. Smith's instruction" to the contrary.

(Genus?) MYONERA Call and Smith.

Shell without cardinal or lateral teeth in either valve; with or without a
buttress; fossette vertical or posteriorly directed, attached to the hinge by
either edge; sculpture radiating or concentric. Type Necera paucistriata
Dall.

Necera sulcifera and angularis Jeffreys, lamellifera Dall, limatula Dall (-H
contratta Jeffreys), laticella Dall, undata Verrill, and fragilissima Smith, be-
long here. The group comprises Sections L and M of Smith, the difference
between which, judging by the figure of fragilissima, does not seem to be very
great.

The absence of gills or palpi, if confirmed by the study of fresh specimens,
ia a very remarkable feature of this and the preceding groups.

Myonera paucistriata Dall.
Necera paucistriata Dall, Bush, Trans. Conn. Acad., VI. p. 473, 1885.

Shell closely resembling Tropidomya abbreviata Forbes, externally, but
deeper, lai-ger, with the anterior concentric sculpture more pronounced, and
two strong keels posteriorly, one a little before and the other a little behind
the place where the single keel of abbreviata is placed. The concentric sculp-
ture ceases just before the anterior keel ; the space between and behind the
keels is smooth except for lines of growth or a few faint radiating striae. The
epidermis is pale, filmy, and polished, except on the dorsal margin and toward
the end of the gaping rostrum, where it seems to concentrate a little ; umbones
rather prominent, whole shell plump, thin, extremely fragile; interior polished;
rib faint; no distinct buttress; fossettes good-sized, extending obliquely pos-
teriorly; attached by most of the posterior edge to the margin above; ossicle
lingniform, smooth, narrow; hinge smooth and edentulous in both valves.
Max. Ion. lO.O; max. alt. 8.5; max diam. 6.5 mm.

Habitat. Stations 226 and 230, near St. Vincent, in 424 and 464 fms., sand,
bottom temperature 42°.5 and 41°.5 F. Specimens living, but smashed. Also
at Station 43, near Tortugas, in 339 fms., bottom temperature 45°.0 F., living
but broken.

This very lovely species resembles externally T. abbreviata, which has one
keel; Myonera angularis Jeffr., which has two keels more posterior than in
this species; and Myonera undata Verrill, which has no keels, and is much

MUSEUM OF COMPARATIVE ZOOLOGY, 303

larger. In M. angularis the posterior keel runs from the umbones to the pos-
terior ventral angle of the rostrum ; in paucisfriata the rostrum is posterior to
both the keels.

The shell of the present species is so fragile as to give way under the slight-
est pressure. The soft parts hardened by alcohol were stronger than the shell,
and oflfered some observations of interest. They were apparently in a perfect
state of preservation.

The outer edge of the mantle was plain, with a covering of epidermis as in
Mya. Around the siphonal opening, which externally is single, were numer-
ous tentacidar iilaments and several ocelli. The opening for the foot is very
small, a mere short slit without ornamentation. On looking from above at
the animal deprived of its shell, we see a globular body corresponding to the
cavity of the valves, divided by a membranous and fleshy horizontal partition
into upper and lower halves or subequal portions. The lower half constitutes
the peripedal chamber into which the pedal and siphonal orifices open. The
upper half contains the viscera, which, however, do not fill it, and the muscles.
From above we see the floor or septum between the two chambers surrounded
by a strong muscular band attached by its edge to the thin mantle and by
upward radiating fibres like a drumhead inside of a drum ; this muscular band
resembles a sphincter, and is produced to the ends of the shell, where it is
attached inside of each adductor ; the course of its roots being vertical, while
the adductors lie in a horizontal plane immediately outside of the former, so
that, when visible, the adductor scars and the others adjoin. In the middle
line of the back are visible the oesophagus and alimentary canal, passing as
usual through the heart, and through a small dark greenish liver-mass on whose
dorsal surface are two small bunches of oval tubules, perhaps genitalia, and a
whitish superficial subdendritic layer, probably the organ of Bojanus, From
the centre of the visceral mass a mesenteric band descends to the centre of the
floor or septum. In advance of this is the base of the foot, with a slender
pedal muscle.

Reversing the animal we see the septum has a sparsely tuberculous surface
(smooth in C. glacialis Sars). Anteriorly is the mouth, simple, without palpi
or gills, opening between two vertical mesenteric bands of tissue. Immedi-
ately behind the oral orifice is the foot, small, subcylindrical, set in an excava-
tion in the septum on a very short constricted peduncle and without any
byssal groove or byssus. Posteriorly is the cylindrical opening of the siphons
which are not separated from one another except by a delicate protrusile sep-
tum, pierced for the two openings and situated within the single orifice of the
mantle. No gills are visible anywhere unless the fleshy tuberculous ventral
surface of the horizontal septum fulfils that office. A similar state of things
in the main was observed in Cuspidaria glacialis Sars, and C obesa Loven, in
which, however, the foot was thorn-shaped, not cylindrical, and the visceral
mass filled or nearly filled the upper chamber.

304 L^LLETIN OF THE

Myonera undata Verhili.
Necera undata Verrill, Trans. Conn. Acad. Sci., VI. pp. 223, 277, June, 1884.

Habitat. Off Chesapeake Bay in 2221 fms. (Verrill); fragments were ob-
tained by the "Blake" in 450 fms. near Havana; at Station 175, near
Dominica, in 611 fms., ooze, bottom temperature 40®.0 F.; and at Station 230,
near St. Vincent, in 464 fms., bottom temperature 41°.5 F.

This species must attain a considerable size and remain extremely fragile.
The pieces obtained by the " Blake " indicate its place to be with Myonera,
but there is no buttress or appearance of an internal rib.

Myonera lamellifera Dall.

Ne(Bra lamelUftra Dall, Bull. M. C. Z., IX. p. 113, 1881.
Plate III. Fls. 7.

Habitat. Station 36, 84 fms. ; a single right valve.

A comparison of figures will show that Mr. Smith's Necera consociata is
wholly distinct from this species. N. semutrigosa Jeffreys, is less oblique in
form, and the concentric rugae are mere threads, not rising clear and sharp
from the shell as in M. lamellifera. Moreover they belong to entirely different
groups, semistrigosa being a typical Cuspidaria. The present species has a faint
rib leading to an obsolete buttress, but no teeth, while the fossette lies poste-
riorly directed in a narrow groove on the hinge margin.

Its nearest relative seems to be the next species.

Myonera limatula Dall.

NecBra limatula Dall, Bull. M. C. Z., IX. p. 112, 1881.

Necera contractu Jeffreys, P. Z. S., 1881, p. 941, pi. Ixxi. fig. 4, 1882.

Plate III. Fig. 5.

Habitat. Station 44, 539 fms., off Tortugas; bottom temperature 39°.5 F.
Two perfect and two broken valves.

None of the specimens of Necera contracta in the Jeffreys collection at
Washington show any tooth whatever in either valve. The posterior cardinal
margin of the right valve is bevelled off a little, to fit in under the edge of the
left valve, but it is not a tooth in any ordinary sense.

It differs from the preceding species in its smaller size with equal length,
finer and less elevated sculpture, and less evident posterior ventral sulcus. In
one right valve there is a point extending forward and downward from under-
neath the middle posterior hinge-line, but this seems clearly pathological.
There is no buttress or rib internally.

MUSEUM OF COMPARATIVE ZOOLOGY. 305

Myonera laticella, n. s.

Shell large, thin, inflated, rather short, white with a pale brown epidermis;
beaks full, prominent, their apices touching over the cartilage; right valve a
little the larger; sculpture of concentric, elevated, thin, but not sharp-edged
lamellae, more distant on the beaks, higher and more crowded toward the basal
margin, failing at the anterior boundary of the rostrum where they are repre-
sented by the fine incremental striae, if at all; radiating sculpture consisting
of somewhat irregular distant sharp elevated lines, which are most abundant
between the middle of the shell and the rostrum; these rise up under but do
not cut through the concentric lamellae, which by them are thrown into waves,
like loose cloth lying over several cords stretched taut, which waves grow
obsolete toward the base; there are also extremely fine radiating striations on
the smooth rostral areas, which are bounded by an imaginary line extending
obliquely from the beaks to the lower extreme of the rostrum ; these striations
bear elevated lines of epidermis ; there is a very narrow depressed area along
the cardinal margin behind the beaks; the margin of the right valve fits over
the posterior part of the rostral margin of the left valve; the rostrum itself is
short, slightly recurved and obliquely rounded oflF from below, and in a less
degree from above, to form a gaping rounded tip; the hinge-line is simple, with
a rather large posteriorly directed fossette for the cartilage in each valve, and
without buttresses ; the cartilage is reddish brown, and carries a subrectan-
gular ossicle; approximate length of shell (broken) 20.0; of young shell (taken
from lines of growth) Ion. 8.0, alt. 5.0, diam. (about) 5.0 mm. From the tip
of the rostrum to the beaks at the cartilage measures in the adult 13.5 mm.,
and the diameter when perfect must have been at least as much.

Habitat. Near Curacoa, at U. S. Fish Commission Station 2126, in 1701
fms., yellow mud, bottom temperature 39°.3 F. One living but broken speci-
men of which the base and anterior end were gone.

This very elegantly sculptured species is so difiFerent from any of those
described that I felt justified in characterizing it from the imperfect specimen.
Apart from its sculpture it somewhat resembles M. undata Verrill in general
form, though the rostrum in that species is shorter and the fossette propor-
tionally smaller.

Family ANATINID^.

Gencs PERIPLOMA Schumacher.

In the region covered by the investigations of the " Blake " several species of
this genus are indigenous. P. incequivalvi^, the type, is often found on sandy
beaches, but usually only the convex valve, destitute of its ossicle. The most
common species on our Southern coast is the P. anguli/era Phil., described

VOL. XII. — NO. 6. 20

306 BULLETIN OF THE

from Texas, but not rare in West Florida; the P. papyratia of Say (uot of
Gould's Inv.) is rare.

Two other small and apparently rare species exist on the coast, one of which,
p. fragilis of Totten, a northern form, has long been considered as Say's spe-
cies, although the shell in question does not agree with Say's in measurement,
in habitat, or with his description. But being the only one at all well known,
and Say's type apparently being lost, it seems to have been supposed that Say's
name must apply to it. This error was corrected by Conrad. The other spe-
cies (yet undescribed unless it be the undulata of Verrill, which I have not
seen) is close to papyratia in size, and chiefly differs in proportions.

Periploma papyracea (Say em.) Stimpbon.

Anatina papyratia Say, Journ. Acad. Nat. Sci. Phil., II. p. 314, 1822 ; Binney*8

Say, p. 104, 1858.
Periploma papyracea Conrad, Am. Journ. Conch., II. p. 7.0, pi. iv. fig. 9 (poor) ;

Ibid., p. 281, pi, XV. fig. 6, 1866.

Habitat. Station 128, otf Frederikstadt, Santa Cruz, in 180 fms. ; one liv-
ing specimen.

Totten's species is larger, flatter, more equivalve, rounder, aiid those I have
seen are destitute of the faint rib extending backward and downward from the
beaks. I should have been disposed to consider, from Say's description, that
he had under his eyes a very young P. angulifera, in which the discrepancy of
the two valves is greater and the rib is strong. But as Conrad has fixed upon
the shell which is in our hands, and figured it under Say's name, and there is
no means of absolutely settling the question by reference to a type, it seems
better to let Conrad's arrangement stand unmodified.

In this species the siphons are wholly disunited and retractile, the foot
very small, clavate; the labial palps enormous, lamellate, and far exceeding
in size the single gill on each side. In P. fragilis the palps are smaller, but
of similar character, while the gills are proportionally larger and the siphons
separate and unequal The prop to the fossette in fragilis is conchologically
a step toward Anatina proper ; but the others are without it.

Gends THRACIA (Leach Ms.) Blainville.

This genus is a synonym of Rupicola, Fleurian de Bellevue (1802), which is
nearly a quarter of a century older, but the name Rupicola was used by Bris-
son for a genus of Birds in 1760. Now Brisson did not use the binomial no-
menclature in the modern sense, and strictly speaking should have no standing.
Nevertheless his generic names are adopted by ornithologists, and on this
ground we may consider the name Rupicola preoccupied in a certain sense. It
should, however, be rejected entirely, and not used in a subgeneric or sectional

MUSEUM OF COMPARATIVE ZOOLOGY. 307

sense, as has been done by some concbologists; since, if it has the right to be
used at all, it is entitled to take precedence of Thracia as the primary name of
the group.

There are several species of Thracia, not including the plaited Cyathodonta
of Conrad, formerly referred to Thracia, which inhabit the southern and south-
eastern coasts of the United States and adjacent waters. Most of them have
not been definitely recognized hitherto. They are: —

Thracia Stimpsoni, u. s.

Thracia corbuloidea Blainville.

Thracia distorta Montagu.

Thracia phaseolina Lamarck.

Thracia Stimpsoni, n. s.

Thracia Stimpsoni is a magnificent species, nearly as large as T. Cunradi,
and of which but one right valve has yet been obtained, between Tampa and
Tortugas in 28 fms. in the Gulf of Mexico. It difl'ers from T. convexa Wood,
which is its nearest relative, in its whiter and much more coarsely granulose
surface, in its more horizontal posterior hinge-margin not rounded off at its
posterior end, and by two strong ribs, one of which extends parallel with the
posterior hinge-line, bounding a narrow smooth depressed marginal area, and
having a wider depressed broad ray on the outer and lower side; the other rib
extends from the beak to the lower posterior angle of the rostrum ; in front
of it the shell is much depressed, behind it elevated for a space extending to
the depressed ray above mentioned; the middle basal margin is more produced
than in T. convexa, and the pallial sinus is shallower and less angular. The
other proportions are about the same as in T. convexa.

The specimen in the National Museum is 65.0 mm. long, and is named in
honor of the late Dr. William Stimpson.

Thracia corbuloidea Blainvillb.

Thracia corbuloidea Blainville, comparing excellently well with the series
in the Jeffreys collection, has been obtained by the U. S. Fish Commission
in from 15 to 50 fms. off the coast of North and South Carolina, generally,
about twenty miles from shore. I have it also from Key West, collected by
Hemphill.

Thracia distorta Montagu.

Thracia distorta Montagu, has been sent me from Honduras by Mr. Charles
T. Simpson, who was formerly resident there. It does not exactly agree with
any particular specimen from British seas, but differs from most of them less
than they differ among themselves. It is possible that this may be the same

308 BULLETIN OF THE

shell described by D'Orbigny as T. rugosa " Conrad," but which Conrad had
never described. I have not seen D'Orbigny's figure.

Thracia phaseolina Kiener.

(?) Thracia phaseolina, Kiener, Dall, Bull. M. C. Z., IX. p. 110, 1881.

Habitat. Yucatan Strait, 640 fms., one valve.

The comparison of this specimen with the fine series of this species in the
Jeffreys collection has confirmed the original identification.

Genus ASTHENOTH^ERUS (Cpr. em.) Dall.
Asthenotharus Carpenter, Ann. Mag. Nat. Hist., XIII. p. 811, 1864.

Shell inequivalve, inequilateral, truncate and slightly gaping behind, resem-
bling Periploma in shape ; beaks not fissured ; no external ligament ; hinge
linear, toothless and without fossette; a wide X-shaped ossicle attached to the
posterior slopes of the domes of the beaks above and behind the hinge-margin.
Pallial sinus deep ; gills like Periploma, siphons separated / foot small. Type
A. villosior Cpr., Cape St. Lucas.

This group differs from Lyonsia in its Periploma-like shell, in having a trans-
Verse wide ossicle instead of a longitudinal narrow or triangular one; in being
anteriorly prolonged instead of posteriorly extended, and probably in the
character of the soft parts, which could not be well studied in the single dry
specimen available. It woiUd, indeed, seem to be a Periploma or Anatina,
destitute of the fossettes and their contained cartilage; in which the transverse
ossicle remains and the beaks are unfissured. The brown ligamentary basis on
which the divaricating feet of the bridge-like ossicle are planted, is visible on
each side through the shell, the brown lines simulating in position and appear-
ance, to a hasty glance, the fissures of Periploma. It is sufficiently separated
from Alicia by the edentulous hinge.

The original and not very clear diagnosis of Dr. Carpenter does not mention
the ossicle, though the latter is still adhering (though not in its place) to one
of the valves of the type in the National Museum. The '" spongy ligament "
he refers to, is the brown cementum which originally held the ossiculum.
The original publication was to be followed by detailed notes, which remained
unpublished at the time of the author's death, which took place all too soon
for science.

Asthenothserus Hemphillii, n. s.

Shell small, yellowish white, concentrically striate, with a filmy epidermis,
left valve slightly smaller than the right, subovate, posteriorly truncated and
slightly gaping; beaks in the posterior third of the shell, the anterior part
rounded like the small cud of an egg-oval; base rounded, rising toward the

MUSEUM VF COMPARATIVE ZOOLOGY. 309

truncation about as much as the posterior cardinal margin falls toward the
upper angle of the same; truncation vertical, but hardly angulated; a faint
ridge running from the beaks to the upper corner, more marked in the smaller
valve ; sculpture of fine not very regular concentric undulations, coinciding
with the lines of growth; surface finely granular, but appearing nearly smooth;
interior polished ; the sinus deep and rounded ; margins very thin, smooth, and
plain; not interrupted under the beaks; ossicle bridge-shaped, wide, short,
concave behind in the middle line, very large for the size of the shell ; points
of insertion not perceptibly raised; extreme length of shell 6.25; length of
anterior part 4.75; max. alt. 6.0; diameter 2.75 mm., of which the deeper
valve takes about 1.50 mm.

Habitat. West of Florida in 17 fms., one specimen; Marco, Florida, in 2
fms., H. Hemphill.

This unpretending little shell has the aspect of a very young Periploma or
Thracia. It is only when the interior is examined that its peculiarities become
manifest. It is possible that, among the innumerable fossil genera or sections
which have been proposed, some one may eventually turn out to include the
present form, but none of those I have been able to examine agree with it.
Gorimya Agassiz is perhaps the nearest, but has submedian beaks with one or
two internal ribs, the posterior cardinal margin slightly thickened as if for an
external ligament, and there is no evidence of a pallial sinus, or rather the po-
sition of the posterior adductor scar would indicate that there was no sinus.

The Californian species differs from the Floridian in having the beaks less
posterior, and, though a larger shell, in having a more slender and delicate
ossicle, which resembles, in some sort, a very long- winged butterfly with its
wings spread.

It is interesting to add to the links connecting the East and West American
faunae, and pleasant in so doing to be able to clear up the obscurity which has
somewhat interfered with the relegation of this genus to its proper place in the
system.

Subgenus BUSHIA Dall.

Shell inequi valve, inequilateral, truncated behind, but not gaping; porcel-
lanous; concentrically sculptured; hinge toothless, with a large U-shaped
ossiculum fitting in the apices of the beaks, which are filled with solid shelly
matter ; a strong external ligament, but for which the hinge-line is not bent or
thickened.

This differs from the typical Asthenothcerus in its porcellanous, instead of
earthy, shell-substance, destitute of granulations; its completely closed valves;
in possessing an external ligament ; and in the filling of the apices of the beaks
with a solid shelly mass on which the extraordinarily large and strong arched
ossiculum is planted, as on two pedestals.

It is dedicated to Miss Katharine J. Bush, of New Haven, whose excellent
work on Mollusca I have had frequent occasion to refer to.

310 BULLETIN OF THE

Bushia elegans, n. s.

Shell white, thin, inequilateral, the left valve a little the smaller, and the
basal edge of the right valve projecting beyond the other ; apices of the beaks
touching each other; shell posteriorly sharply truncate; anterior part of the
shell forming sixteen twenty-fifths of the whole length ; the anterior margin
rather pointedly rounded, and the extreme anterior point nearer to the level
of the base than to that of the hinge-line; beaks not much elevated but mod-
erately full; the surface evenly concentrically deeply grooved all over, about
three interspaces to the millimeter, the grooves narrower than the interspaces ;
a short external ligament behind the contiguous beaks; a keel extends from
the beak to the upper posterior angle of the truncation of the posterior side
(which is almost as abrupt as in Mya truncatu) parallel with the descend-
ing hinge-margin ; over this keel the raised interspaces form threadlike ribs;
within the keel is a narrow nearly smooth lanceolate depressed area, wider in
the left than in the right valve; the angle at the end of the keel where the
truncation begins is abrupt ; the basal angle is very bluntly rounded; interior
smooth, with some radiating stria;; the beak, inside its tip, is filled with a
dolid transparent deposit, on which the feet of the arched ossicle are' attached
by a layer of cartilage ; the hinge is toothless, thin, and weak; the imprint of
the mantle invisible; but the pallial sinus is moderately deep and rounded;
there is no visible epidermis; the surface is smooth, but not brilliant; the
posterior hinge-margin, looked at from above, is seen to be somewhat flexuous
laterally. Lon. of shell 12.5; alt. 10.0; diameter 6.0 mm.

Habitat. Station 272, near Barbados, in 76 fms., hard sand, bottom tem-
perature 64°.75 F. (one right valve). U. S. Fish Commission, Station 2639,
56 fms., in the Straits of Florida, one living specimen and one valve.

The possession of a living specimen in a good state of preservation has en-
abled me to fix the position of this elegant little shell, which from only the
single valve obtained by the " Blake " would have been a little doubtful.

The soft parts (in alcohol) afford the following notes. Siphons not very
long, entirely separated ; mantle closed, except in front of the foot ; ends of
both siphons papillose ; mantle simple, smooth along the edge ; gills large,
lamellse dorsally much crumpled, both sides united at the tips behind ; palpi
very small, narrow ; foot small, rather hatchet-shaped, not grooved behind ;
posterior adductor the larger ; the inner gill on each side much the shorter
and narrower of the two ; the gills together envelop the whole body except
the foot and a passage-way to the excurrent siphon. The ossicle forms a
U-shaped arch, its feet a little enlarged and divaricating backward ; the hinge
margin is normally entire; but, with the ossicle in place, it is impossible to
separate the valves without breaking a little notch, just below the beak where
the ossicle crosses, in the valve which does not retain the ossicle, or in both ;
the outer surfaces of the mantle and the soft parts in general, except the liver,
are not pigmented.

MUSEUM OF COMPARATIVE ZOOLOGY. 311

The peculiarity of the filling up of the tips of the beaks does not consist in
there being mere pedestals or sockets for the feet of the ossiculum. The
whole cavity seems evenly filled, and the ossicle stands, as it were, on a sort
of floor ; this is quite visible from without, through the translucent shell. It
is a common thing to find the early whorls of Gastropods filled solid with
shelly matter, but such cases are rare among the Pelecypods, if we leave out
of account the usual thickening due to growth.

Family PANDORID^.

Genus PANDORA Hwass.

Subgenus CLIDIOPHORA Carpenteb.

Of the PandoridoR the southern coasts and the Antilles have several species :
Clidiophora trilineata Say ; another form, of which one valve was described but
not named by Miss Bush ; Pandora (Kennerlia) glacialis Leach, which passes
Hatteras, its southern limit not yet known; P. carolinensis Bush, described
from near Hatteras, probably entering the Gulf of Mexico, and P. Bushiana,
received from West Florida. This group, being chiefly composed of shallow-
water species, is represented in the Blake dredgings only by worn left valves
of one species.

I may add, that in this genus, as in others, I regard anterior and posterior,
right and left, from the anatomical standpoint. A singular discrepancy exists
among authors in treating of this genus, as we find the rostrated or siphonal
end of the shell frequently treated as anterior. As a matter of fact, it is pos-
terior, as in other Pelecypods.

Pandora (Clidiophora) carolinensis Bush.

Pandora carolinensis Bush, Trans. Conn. Acad., VI. p. 474, 1885.
Pandora oblongaf Sowerby, Dall, Bull. M. C. Z., IX. p. 109, 1881.

Plate VIII. Figrs. 8, 8 a.

Habitat. Charlotte Harbor, Florida, 13 fms.; Yucatan Strait, 640 fms.,
detached valves only.

I presume that the valves above mentioned should rightly be referred to
Miss Bush's species. Whether both are referable to P. oblonga is a question
on which opinions may differ, as the type of oblonga is said to be lost.

They are not referable to P. trilineata Say (not Gould, etc.), which is a
much elongated, slender, narrowly rostrated species with the beaks more an-
terior even than P. hrevifrons Sby.; the base roundly arcuated, the posterior
cardinal margin concave, the anterior rounded from the beaks to the base, the

312 BULLETIN OF THE

impressed line of the left valve concave forward as it sweeps in a broad curve
from the small blunt beaks to about the middle of the base. Its surface is very
finely concentrically striated without radiating sculpture except the fine raised
lines which extend from the beaks to the tip of the rostrum, which is a little
deflected to the left. The shell is almost ilat and nearly equivalve, of a waxen
white, pearly within, and with a few radiating striae. The hinge is that oJ
Clidiophora, and it is a much more slender shell than P. carolinensis, being
only 8.0 mm. in height to 20.0 mm. in length, and less than 2.0 mm. in maxi-
mum diameter; the anterior part is only 3.5 mm. long. The anterior cardinal
area is linear, the posterior grooved out and bordered especially in the right
valve by a broad rib. This is the widest part of the whole shell. Numerous
valves were obtained at Stations 2592 and 2597, U. S. Fish Commission, oflf
Hatteras.*

This species was dredged alive in 6 fms. at Tampa, Florida, by Mr. Chas. T.
Simpson. With it was found a smaller species, belonging to the Pandora or
Kennerlia section of the genus. This, to which I propose to give the name of
P. Biishiatia, differs from all the known species of the group in having the
posterior cardinal margin convex, and the rostrum bent downward instead of
upward. The beaks are very anterior, and the anterior cardinal margin,
marked with a sharp keel setting off an almost linear area, descends from the
beaks in a straight line, the curve of the base commencing suddenly at a
rather obtuse angle, and following an even curve, is slightly inflexed only near
the posterior tip, below the short square-ended rostrum; the shell is very thin,
the left valve somewhat convex, the right one concave, both sculptured with
silky concentric striae; the margins of the two valves coincide; the beaks are
small, hardly rising above the long arch of the back; the right valve has a
strong keel on its upper posterior margin, and no other radiating sculpture;
the left valve has an impressed line from the beaks to the base a little behind
them, but which does not indent the basal margin ; there is also a sharp thread
from the beak to the lower angle of the rostrum; above this thread, as usual,

* As this species is clearly different from the form figured in Binney's Gould
under the name of trilineata, and generally so called by American concliologists, I
had thought it new, and intended to name it P. (C.) Jioridana ; hut referring to
Say's original description and figure, I found that the southern form which he
described and figured (poorly) is the one he named trilineata, and, as far as I can
discover, the northern form has had no name given to it which it can retain.
Specimens in the British Museum were labelled nasuta Sowerby, but they are not
the true naifuta of Sowerby according to Reeve, who had the advantage of Car-
penter's monographic determinations, and figures the genuine nasuta, which in any
case would be a synonym of the southern form. The tabacea of Meuschen (Mus.
Gronovianum) is known only by a very poor figure ; the P. depressa of Sowerby,
whicii has been identified with it, according to Deshayes, is a native of the Pacific
Ocean. I would therefore propose for the high, concentrically undulated New Eng-
land shell the name of Pandora ( Clidiophora) Gouldiana, in honor of the late Dr. A. A.
Gould.

MUSEUM OF COMPARATIVE ZOOLOGY. 31

o

the lines of growth are coarser. The shell is 11.5 mm. long ; 5.5 mm. high,
about 1.0 mm. in diameter, and with the anterior portion 2.0 mm. long.

Family COKBULID^.

Genus CORBULA BRuoiiRE.

The following notes on the species of Corbula are not as complete and final
as the study of a larger series and collection from a wider area would have
afforded.

The species of this group are very variable, and would doubtless be much
reduced in. number after a comprehensive examination of the recent and fossil
forms.

By the kindness of Prof. John Tyler, of Amherst, custodian of the Adams
collection, I was able to examine various types of the ten species of Corbula
from Jamaica, described by Prof. Adams in his Contributions to Conchology,
1852. As far as I know, they have remained unfigured up to the present time,
and it occurred to me that it would be desirable to figure them. So the figures
of the Corbulidce, with one exception, accompanying this paper, are camera-
lucida sketches from the original types of Prof. Adams, instead of specimens of
the same species collected by the " Blake." There are still a few of Adams's
species unfigured, and it would be of much use to science if some Amherst
student would avail himself of the opportunity to illustrate as many as possible
of the types contained in the Adams collection. Such expense as might be
connected with the task would be no more than a reasonable and proper tribute
to her devoted Professor from the College of which he was so great an ornament.

Corbula cubaniana D'Orbignt.

Corbula cubaniana D'Orb. (1846), Dall. Bull. M. C. Z., IX. p. 114, 1881.
Corbula Knoxiana C. B. Adams, Contr. to Conch., p. 238, 1852.

Plate I. Figs. 3, 3 a- 3 c.

Habitat. Sigsbee, off Havana, 100 fms., one valve.

The figures are drawn from C. B. Adams's type of C Knoxiana. Lon. of
shell 12.7 mm.

Corbula Barrattiana C. B. Adams.

Corbulr- Barrattiana C. B. Arlams, 1. c, p. 237, 1852 ; Dall, Bull. M. C. Z., IX. p.

114, 1881.

flate II. Figs. 7, 7 a - 7 o.

Habitat. West coast of Florida, 30 fms.; Station 21, 287 fms.
Identified and figured from the types. Lon. of shell 8.9 mm.

314 BULLETIN OF THE

Corbula Swiftiana C. B. Adams.

Corhula Swiftiana C. B. Adams, L c. p. 236, 1852 ; Dall, Bull. M. C. Z., IX. p. 114,

1881.

Plate II. FigB. 5 a - 5 c.

Habitat. Sigsbee, off Havana, 182 and 450 fnis., living; oflf Sombrero, 72
fms.

Identified and figured from the types. Lon. of shell 10 4 mm.

Corbula Dietziana C. B. Adams.

Corhula Dietziana C. B. Adams, 1. c, p. 2-35, 1852 ; Dall, Bull. M. C. Z., IX. p. 114.
Corbula Blandiana C. B. Adams, 1. c, p. 234 (= young stage of C. Dietziana, Ad.).

Plate I. Figs. 5, 5 a, 5 b.

Habitat. West coast of Florida, 30 fms. ; off Sombrero, 72 fms. ; Barbados,
100 fms. ; Gordon Key, 68 fms.
Identified and figured from the types. Lon. of shell 10.7 mm.

Corbula disparilis D'ORnioNT.

Corhula disparilis D'Orb., 1846; Dall, Bull. M. C. Z., IX. p. 115, 1881.
Corbula Philippii Smith, Chall. Rep., p. 33, pi. vii. figs. 4, 4 a, 4 b, 1885.
Corbula operculata Philippi, Zeitsch. Mai., V. p. 13, 1849.

Plate T. Figs. 4, 4 a, 4 b.

Habitat. Off the west coast of Florida, 30 and 50 fms. ; Station 12, 36 fms. ;
off Sombrero, 72 fms. ; Station 36, 84 fms. ; Barbados, Station 287, etc., 7^ to
100 fms. ; Sigsbee, off Havana, 127 and 450 fms.; Station 2, 805 fms.

This species closely resembles several exotic and fos.sil forms ; among the
former may be mentioned Corbula nucleus L. ; among the latter, C. oniscus
Conrad (Eocene of the United States) and C. parsura of Stoliczka, from the
Trichinopoly beds (Cretaceous of India), as well as some from the Danish
ligniferous strata.

Those who consult D'Orbigny's figures will observe that they differ from the
shell figured by my friend Smith in representing the valves as nearly equal,
and also in the absence of the epidermal radiations on the smaller valve and
the carina on the larger one. But I infer from D'Orbigny'.s remarks, that he
had only separated valves, and probably those which had lost their epidermis;
and it is probable that the artist represented two valves together which did not
belong together. The carina is a variable feature in this species, as in C. nu-
cleus. At all events, the specimens I have are certainly the same as G. Phi-
lippii Smith, and I believe them to be the species described by D'Orbigny.

MUSEUM OF COMPARATIVE ZOOLOGY. 315

The species extends northward to Cape Hatteras, and the smaller valve is fre-
quently of a pink color or pinkish brown. It reaches a length of 8.0 mm.,
and is very variable in its proportions and sculpture. I have no doubt that it
is the operculata of Philippi, but the C. Krebsiana of C. B. Adams is a different
Rnd more delicate species.

Corbiila (Taeniodon?) cymella Dall.

Corbula cymella Dall, Bull. M. C. Z., IX. p. 115, 1881.

Plate I. Fisrs. 7, 7 a.

Habitat. Gordon Key, 68 fnis., one living specimen, 13.5 mm. in length.

The accidental fracture by the draughtsman of one valve of the unique, speci-
men enabled me to get at the hinge. I found it very delicate, the right valve
with a single small slender tooth, behind which is a very small cartilage set in
a short groove in the hinge-margin, and continuous above with a darker-colored
linear substance, which may have been a bit of thicker epidermis than ordinary,
but looked like a linear external ligament covered only by the coil of the umbo.
The left valve had a smooth edentulous hinge margin, with the cartilage en-
tirely on top of the small thin horizontal process behind the beak.

Another feature omitted in the original diagnosis is, that the very fine radi-
ating lines, which cover the shell and are most noticeable on the posterior
supra-carinal area, are minutely granular. The thin and hardly unequal valves,
and the marginal, if not exposed, cartilage of this species, suggest a modification
in the direction of Tceniodon.

The following three species were not represented in the collection,
but, in view of their not having been figured and thus being placed in
doubt in the catalogues, it was thought worth while to include the
camera-lucida sketches of the types and a synopsis of the remarks of
Professor Adams in regard to each of them.

Corbula Krebsiana C. B. Adams.

a Krebsiana Ad. Contr. to Conch., p. 234, Oct. 1852.

Plate I. Fiffs. 1, la, lb.

Shell trigonal, very inequivalve, inequilateral, with the large valve ros-
trated; the ventral margin rounded anteriorly, nearly straight posteriorly;
white, often tinged with pink, except posteriorly; small valve finely concen-
trically striated; large valve finely and closely furrowed; beaks prominent,
much involuted, umbones very convex; with small posterior angles, one on
the small valve and two on the other: teeth small. Lon. 6.1; alt. 5.1; diam.
3.8 mm.

316 BULLETIN OF THE

Kingston, Jamaica, in three or four fms., mud, C. B. Adams. Probably
resembles C. operculata Philippi. [It is quite distinct from operculata (= dis-
parilis Oih.). — W.B.. D.]

Corbula Chittyana C. B. Adams.

C. Chittijana Ad. Contr. to Conch., p. 238, 1852.

Plate n. Figs. 6 a - 6 d.

This species resembles C. Barrattiana, but differs in being very thick and
solid, very wide, and in having two periods of growth, like C. Dietziana : it is
also more inequivalve. Lon. 8.5; alt. 5.75; diam. 5.5 mm.

Habitat. Kingston Harbor, Jamaica, in 4-5 fnis., mud, rare; Adams.

Corbula Kjoeriana C. B. Adams.
C. Kjoeriana Ad. Contr. to Conch., p. 237, 1852.
Plate I. Figs. 6, 6 a, 6 b.

This species differs from C- Swiftiana in being less distinctly rostrated
though usually a little more elongated behind ; the concentric ridges are
stouter and are continued into the lunule; both valves are sculptured alike ;
the umbonal angle is more acute and distinct, and is a little more distant
from the posterior dorsal margin. Lon. 12.0; alt. 7.5; diam. 4.5 mm.

Habitat. St. Thomas, Bland; Jamaica, 4-5 fms., mud, Adams.

Genus BASTEROTIA Matek.

Basterotia quadrata, var. granatina, Dall.

<iCorbula quadrata Hinds, Reeve, Conch. Icon , Corbula, fig. 40, 1843.

Poromya ? granatina Dall, Bull. M. C. Z., IX. p. 109, 1881.

? Basterotia corhuloides Mayer, Hurnes, "Wiener Beck., p. 40, pi. iii. fig. 11, 1866.

Plate I. Figs. Z, Z a, Z b.

Habitat. Yucatan Strait, 640 fms., one valve.

After further investigation it became evident that the shell above referred
to was only one of the rather numerous varietes of Corbula quadrata Hinds.
This form belongs to the genus Eucharis Recluz, 1850, not of Latreille in
1804, or of Peron or Eschscholtz of later, but still prior dates. I believe
Mayer's name is the first which has been applied to it which is valid. Mayer's
species appears (as he admits) hardly different from the living West Indian
and Pacific shell, which I have seen even from Korea!

MUSEUM OF COMPARATIVE ZOOLOGY. 317

Family SAXICAVID.E.
Genus SAXICAVA Fleurian de Bellevob.

Saxicava azaria Dall.

Saxlcava azaria Dall, Bull. M. C. Z., IX. p. 116, 1881.

Plate IV. Figs. 9 a, 9 b.

Habitat. Off Charlotte Harbor, Fla., 13 fms.; Station 39, Gulf of Mexico
sixteen miles north of the Jolbos Islands, in 14 fnis.

In spite of the variability of this group, I am pretty confident that this is
not one of the varieties of the northern S. rugosa. The second specimen shows
the surface more perfectly, and is covered Avith a closely adherent red-brown
epidermis under which the shell is finely wrinkled in a manner different from
any of the ordinary species I have been able to compare with it.

Family i^HOLADID.E.

Genus XYLOPHAGA Tcrton.

Xylophaga abyssorum, n. s.

Plate IX. Figs. 7, 7 a.

Shell minute, wedge-shaped and nearly closed behind, inflated and more than
half open in front; anterior area concentrically sharply striate, reflected at
the dorsal edge, and covering in somewhat less than half (on each side) of the
anterior face of the animal; beaks small, not prominent, but sharply defined,
giving rise to two somewhat nodulous keels, which extend to the opposite
edge of the valve, where they are doubled in forming a kidney-shaped callus
inside the edge; the anterior keel is less prominent than the second one, and
the space between them is slightly depressed and smooth except for incremental
lines; there is a narrow smoothish margin with oblique striations between the
first keel and the hinder edge of the anterior area; this widens out below the
lower edge of the anterior area and forms a sort of margin to the central keeled
part of the valves; it is rounded off at the ventral angle formed by the ventral
and anterior edges; behind the keels the posterior area is roundly arched be-
hind the two ends of the posterior keel; the surface is concentrically striate,
but less closely than is the anterior area; the posterior margins close perfectly;
the shell is pure white, and exhibits no accessory pieces or any place of attach-
ment for any. Lon. of shell 4.0; alt. 3.0; diam. 4.0 mm.

Habitat. One specimen in soft coral nodule at Station 215, in 226 fms.,
off Santa Lucia, bottom temperature 51°. 0 F.

318 BULLETIN OF THE MUSEUM OF COMPAKATIVE ZOOLOGY.

I am somewhat uncertain whether this shell properly belongs in the genus
Xylophacja, but it may be immature, and the accessory lamina; may be later in
developing. At all events the form and sculpture of the shell are so much
more nearly like that of Xylophaga, as figured by Chenu, than like any other
Pholad, that I prefer to place it here awaiting other information. Some
years ago what appeared to be exactly the same species was shown me burrow-
ing in the hempen covering of part of the first Atlantic cable; said to have
been recovered from the North Atlantic at a depth of over 1500 fms. The
siphons were simple, and it showed no accessory plates.

PLATE I.

Fig 1. Corhuhi Krebsiana C. B. Adams; 6.1. Page 316.
1 a.

1 b.

2. Basterotia qnadmta Y{\nAs ; 10 0 ; left valve. Page 316.

2 a. Same, hinge seen from above.

2 b. " " " below.

3. Corhula Knoxiana C. B. Adams ; 12.7; front. Page 313.

3 a. " " back of left valve.
3 b.

3 c.

4. Corbula disparilis D'Orbigny ; 9.0. Page 314.

4 a.

4 b.

5. Corbula Dietziana C. B. Adams ; 10.7. Page 314.
6 a.

5 b.

6. Corbvla Kjoeriana C. B. Adams ; 12. Page 316.

6 a.

6 b. "

7. Corbula a/mella Dall ; 13.5. Page 315.

7 a.

All the above, except figures 2 a, 2 b, and 4, 4 a, 4 b, are drawn from typical
specimens of the describer.

BLAKE MOLLUSCA.

PLATE

McConr.ell del

Photo. Lilh. Dy L. S Purjenon New Haven. Conr

PLATE II.

Fig. 1 a, 1 b. Verticordia (Euciroa) elegantissima Dall ; 13.25. Page 291.

" 2,2 a. Halonympha claviculata DaM ; 12.0. Page 301.

" 3 a, 3 b. Cardiomya perrostrata Dall ; 8.0. Page 296.

" 4 a, 4 b. Verticordia (Haliris) Fischeriana Dall ; 10.0. Tage 291.

" 6a-6c. Corbula Swijliana C. 3. Ad&ms, from type ; 10.4. Page 314.

" 6 a- 8d. Corbula Chittyana C. B. Adams, from type ; 8.6. Page 316.

" 7, 7 a- 7c. Corbula Barrattiana C.B. Adams, from typea; 8.9, Page 313.

BLAKE MOLLUSCA.

PLATE II

McCornell d»l

Photo. Lith. by L. S. Pun(l(>rson New H«ven, Conn

PLATE III.

Fig. 1. Cuspidaria obesa Loven, var.? 13.0. Page 295.

2. Cuspidaria Jefreysi T>iin ; 15.0. Page 295.

3. Cuspidaria arcuata Dall ; 12.5 ; inside. Page 296.

4. Same, outside.

5. M i/oneralimatula Ball ; ll.\5. Page 304.

G. Cardiomya pectinaia Cpr., var. berimjensis Leclie [N. W. coast of Amer-
ica], 6.0. Page 29G.

7. Myonera lamdlifera Dall ; 12.5. Page 304.

8. Leiomya (Plectodon) fjranulata Dall ; 11.0. Page 300.

9. Cardiomya costdlata var. corpulenta Dall ; 14.0. Page 298.
10. Cardiomya striata Jeffreys; 19 0. Page 298.

BLAKE MOLLUSCA.

PLATE

McConne'l del-

Photo. Lith, by L. S Punderson New Haven, Conn

PLATE IV.

Fig. 1 a. Pecten (Amusium) Dalli E. A. Smith; 62.0; inside of lower valve.

Page 209.
" 1 b. Same, inside of upper valve.

" 2. Pecten (Pseudamusium) SigsheeiDaW ; ll.b. Page 223.

" 3. Pecten (Propeamusium) Pourtalesianus Dall, var. marmoratus ; 13.5.

Page 211.
" 4 a, 4 b. Pecten (Pseudamitsium) iniMJer "Loyhn; 12.5. Page 220.
" 5 a, 5 b. Dimt/a argentea DnW ; 12.0. Page 228.
" 6. Cardium ceramidiiin Dall ; 8.2. Page 269.

" 7. Cardium peramahllis Dall ; 12.5. Page 269.

" 8. ^6ra/i0jca Dall; 8.1. Page 278.

" 9 a, 9b. Saxicava azaria Dall; 25.0. Page 317.

BLAKE MOLLUSCA.

PLATE IV.

McConrte'l del

Pt.oto. Lith, by L S Pundprcor* New Haven Conr.

PLATE V.

Fig. 1, 2. Pecten {Propeamusium) cancellatus E. A. Smith. Page 213.

1 a. Same ; bit of the sculpture enlarged ; 2G.0.

3. Pecten {Propeamusium) Sayanus Dall ; 15.5. Page 214.

4. Pecten caurinus Gould, young valve ; 6.0. Page 216.

5. Pecten (Propeamusium) Holmesii DaW ; 12.0. Page 214.

6. Hinnites Adamsi Dall ; upper valve ; 2S.0. Page 223.

7. 7 a. Pecten {Propeamusium) alaskensis Dall; 22.8. Page 215.

8. Pecten {Pseudamusium) reticulusTtaW; 7.0. Page 221.

9. Pecten {Propeamusium) Sayanus DaM ; 15.5. Page 214.

10. Pecten {PseudamusiuiT,) reticulus DaW ; 1 .0. Page 221.

11. Pecten {Propeamusium) Ilolmesii Dall ; 12.0. Page 214.

12. Pea.''n {Propeamusium) Pourlalesianus Dall; IZ.5. Page 211.

BLAKE MOLLUSCA.

PLATE V

UcConnell del

Photo Lilh. by L. S Pundcfson N«w Haven Conr.

Fig

.1.

H

2.

t<

3.

tl

4 a-4 c.

u

5.

<l

6.

«

6 a.

f<

7,8.

u

9, 10.

«*

11.

il

12.

PLATE VI.

Afagasella radiata Dall ; 6/1 [N. W. America].

Thecidium Barretti Davidson ex Woodward; 5.1. Page 205.

Modiola polita Verrill & Smith ; 42.5. Page 234.

Terebratula Bartlettii Dall ; 40.0. Page 200.

Pecten (Janira) hemicyclica Ravenel ; 4.0 ; inside upper valve of

young shell. Page 207.
Terebratula incerla Davidson ; 11.6 ; interior. Page 201.
Same, horizontal view of loop.
Modiolaria lateralis Say ; 7.6. Page 236.
Area ectocomata Dall ; 46.0. Page 243.
Tellina sybaritica Dall ; 7.0. Page 277.
Crassatella floridanaDaW; 11.0; young shell. Page 256.

BLAKE MOLLUSCA.

PLATE VI.

McConneM del

Photo. Lilh. by L. S Pundeison New Haven, Conn.

PLATE VII.

Fig. 1 a-b. Leda (Neilonella) corpidenta Dall ; 9.5. Page 254.

" 2. Nucula crenulata A. Adams ; 6.0. Page 247.

" 3 a-b. Leda acuta ConT&d ; 9.5. Page 251.

" 4 a-b. Gouldia cerina C. B. Adams ; 10.5, type. Page 263.

" 5 a-b. Astarte Smithii Dall ; 7.0. Page 259.

" 6 a-b. ^s^arte nana (? Jeffreys) Dall; 8.2. Page 261.

" 7 a-b. Leda soUdifacta Dall ; 12.5. Page 252.

" S. Leda acuta Conrad ; 9.5. Page 251.

BLAKE MOLLUSCA.

PLATE VII

McConnel! del

Photo. Lith, by L. S P'ind**'son New Haven, Conn-

PLATE VIII.

Fig. 1, 1 a. Tindaria cytherea Dall ; 8.6. Page 2-54.

" 2. Nucula var. obliterata Dall ; 6.0. Page 247.

" 3, 3 a. Area polycyma Dall ; 9.75. Page 241.

*' 4, 4 a. Macrodon asperula Dall ; 8.5. Page 244.

"- 5. Area pecttinculoides, var. orbieulata Dall; 8.0. Page 240.

" 6. Leda (Saturnia) qiiadranyularis Dall ; 4.G. Page 253.

" 7, 7 a. Limopsis antillensis Dall ; 4.25. Page 237.

" 8,8 a. Pandora {Clidiophora) caroUnensis 'Bush ; 14.2. Page 311.

" 9, 9 a. Area ylomeruJa Dall ; 5.75. Page 241.

" 10. Cetoconcfia maryarita Dall ; 7.3. Page 284.

" 11. Leda Carpenteri Dall ; 10.5. Page 249.

" 12, 12 a. Leda vitrea, var. eerata Dall ; 6.5. Page 250.

" 13. Vesicomya pilula Dall ; 2.6. Page 274.

BLAKE MOLLUSCA.

PLATE VI

McCoooell del

Photo. Lith. by L S Pundf'son New Haven, Conr

PLATE IX.

Fig. 1, 1 a. Yoldia liorhina Dall ; 13.1. Page 248.

" 2, 2 a. Yoldia solenoides Dall ; 12.5. Page 248.

" 3. Leda Carpenteri Dall ; 10.5. Page 249.

" 4. Pleurotoma serga Dall ; 9.0.

" 5. Pleurotoma (Mangilia) cilronella Dall; 4.0.

" 6. Pleurotoma (Mangilia) Pourtalesii DaW ; 17.0.

" 7, 7 a. Xylophaga abyssorum Dall ; 4.0. Page 317.

" 8. Conus Agassizii Dall ; 30.0 ; adult.

" 8 a. Same, young shell ; 9.0.

" 9. Pleurotoma (Dapknella) leucophlegmaDaW ; 10.26.

" 10. Plc'irotoma (Daphnelta) limacina T)a.\l ; 11.0.

BLAKE MOLLUSCA.

PLATE IX

McConnell del.

Photo. Lith. by L. S PunJetson New Haven. C"-

Harvard MCZ Libi

3 2044 066 302 910

Date Due