On certain portions of the skeleton of Protostega gigas

•f %VJ*p^V;^.:

»^%H'¥'Sv;
^^BSdm

^^v;Vv: , ^

^«fc;iS^^

BS^HK9|

%^-S-r^

jf taming anb jt'alior.

LIBRARY

I University "of Illinois.

m?;

1&:. -g&K

mT"

y r»

r * •

^-v^wuiiii uiio w\ji\. mi \ji UCiUl'T^ tile ™

* Latest Date stamped below. A
i charge is made on all overdue

b°°kS- U. of I. Library

FEE 22 19^

i

FEB 5H

79

\j^
JAN2S

v^/vy^

1984

17625-S

i *+ .

t

FIELD COLUMBIAN MUSEUM

PUBLICATION 7.
ZOOLOGICAL SERIES. VOL. i, No. 2.

ON CERTAIN PORTIONS OF
THE SKELETON

OF

PROTOSTEGA GIG AS.

BY

O. P. HAY, PH. D.,
Assistant Curator of Ichthyology.

D. G. ELLIOT, F. R. S. E., Curator of Department.

CHICAGO, U. S. A.
November 21, 1895,

ON CERTAIN PORTIONS OF THE SKELETON OF PROTOS-
TEGA GIGAS COPE.— O. P. HAY.

The Dermochelyoid turtle, Protostega gigas, was first described
by Professor E. D. Cope in Proc. Amer. Phil. Soc., 1-871, page 172,
and again in the same publication in 1872, page 403. In 1875, m n^s
''Cretaceous Vertebrata," pp. 99-113, pis. IX-XIII, the same writer
more fully described and illustrated the structure of this remarkable
reptile.

The materials which were in Professor Cope's hands consisted of
a number of vertebrae, ten ribs, some marginal bones, certain por-
tions of the skull, some limb bones, and some large plates. Of the lat-
ter there were what the describer regarded as two entire and parts
of one or two others. These plates he considered as belonging to the
carapace, and this was supposed to be free from the ribs, as the pecu-
liar carapace otDermochelys is free from the ribs of that turtle. In this
conclusion he was undoubtedly wrong, as was later shown by Dr. G.
Baur (Biolog. Central blatt, vol. 9, p. 190). This author pointed out that
the plates were components of the plastron, an opinion that finds
abundant confirmation in the materials here to be described. These
consist of a large portion of the plastron of a large individual whose
remains were entombed in the Cretaceous deposits of Butte Creek,
Kansas. As shown in Plate IV, there are present the hyoplastron
and the hypoplastron of the left side almost complete. There are
also portions of the same bones belonging to the right side. These
parts of the plastron were also accompanied by the nuchal.

The length of _the hyoplastron and the hypoplastron taken to-
gether amounts to 1.2 metres, including the estimated length of a
piece missing from near the hinder end of the hypoplastron. These
two bones are united by suture, which may be seen immediately in
front of the fracture produced in excavating the fossil. The relation
of these two bones is therefore unmistakably indicated. The suture
between the two bones is a very short one, in comparison with that

of Thalassochelys.

67

58 FIELD COLUMBIAN MUSEUM — ZOOLOGY, VOL. i.

The length of the hyoplastron is 6icm; its width 52.5 cm. The
extreme width of the hypoplastron is somewhat less than that of the
hyoplastron, but it cannot be accurately determined. The latter
bone is thickest just behind and somewhat mesiad of the excavation
for the fore limb, and here the thickness amounts to 45mm. The
hypoplastron is not so thick, but still quite thick and solid. The cor-
responding bones in Professor Cope's possession were not more than
half an inch in thickness, at the most. This condition was in all
probability due to the pressure to which they had been subjected.

As will be observed, the anterior inner angle of the hyoplastron
is extensively developed, surpassing in this respect that of Thalasso-
chelys, in which again the plastron is more developed than in
Chelonia. As usual in all the recent marine turtles, this angle ex-
tends further forward than does the outer one. To that border of
this angle which lies next to the fore limb was attached the hinder
end of the epiplastron. Neither of the epiplastra was secured. In
Thalassochelys the anterior ends of the epiplastra extend in front of a
line joining the bottoms of the excavations for the fore limbs a dis-
tance equal to that from the bottom of the excavations for the fore
limbs to those for the hind limbs. This, in the Protostega plastron
before me, amounts to 84 cm. The xiphiplastra of Thalassochelys
extend behind the excavations for the hinder limbs as far as do the
epiplastra from the anterior excavations. If these proportions hold
good for Protostega, the whole length of the plastron would amount
to at least 2.4 metres.

As shown by the figure, the hinder end of the hypoplastron is
prolonged backward and somewhat inward as a long process. Mesiad
of this process there has been another and, judging from the example
of Chelonia and Thalassochelys, a longer process. A portion of this
process is missing, but the bone, where the fracture has occurred is
still 21 mm. thick. This missing process was also evidently directed
somewhat toward the middle line of the body, as well as backward.
Between the two processes has been received the forked end of the
xiphiplastron of that side. The upper end of the inner border of the
outer process has been chamfered off where it forms a suture with
the xiphiplastron. This chamfering of the bone continues beyond
the point of union of the two processes and is then carried backward
on the inner process as far as this remains. The upper side of the
outer border of the outer process has also entered into sutural union
with the xiphiplastron. The whole structure is here extremely simi-
lar to that seen in Chelonia and Thalassochelys.

Nov. 2 1, 1895. SKELETON OF Protostega gigas — HAY. 59

Had the breadth of the body of Protostega possessed the same
ratio to the length that we find existing in Thalassochelys, the lower side
of the animal would have been about 2.2 metres wide. The positions
of the surfaces for union with the epiplastra and xiphiplastra, and the
location of the axis of strongest development of the two plastral bones
of each side make it evident that the outer border of the bony plas-
tron was at a considerable distance from the outer edge of the body.
This is shown too by measuring outward from the excavation for the
arm a distance proportional to that found in Thalassochelys. The tips
of the digitations of the plastral bones must have lacked as much as
30 cm. of reaching the marginal bones. This will leave a space of
about 120 cm. from the bottom of the excavation for one arm to that for
the other. When we come to compare the distance from the hinder to
the front excavations, in the restoration of Protosphargis by Capellini
(Mem. Ac. dei Lincei, 1884, 3 ser., vol. 18), with the distance of
the two anterior excavations apart, I find that the two measure-
ments have almost exactly the same ratio that I have given them in
Protostega.

If we have placed the plastral bones aright, there is left between
them a great fontanelle. Where the hyoplastra are widest this is
about 43 cm. in width ; and opposite the union of the hyo- and hypo-
plastron, about 90 cm. This is somewhat smaller, however, than the
fontanelle found in Protosphargis, and much smaller than that of
Dermochelys.

The nearest relative of Protostega is undoubtedly Protosphargis \
but when we come to compare the two plastra, we find abundant dif-
ferences. That of Protosphargis is considerably less developed than
that of Protostega. Notwithstanding this, there was on the front of
the hyoplastron of Protosphargis a long slender process which ran for-
ward and inward to connect with the epiplastron. In Protostega the
corresponding angle of the hyoplastron is broad, rounded off, and
digitated. In Protosphargis again there is a broad notch in the ante-
rior and outer border of the hypoplastron, but none in Protostega.

It appears to be quite evident that Capellini's restoration of Pro-
tosphargis is in one respect not wholly accurate. The epiplastra ap-
pear to be too short and to converge too rapidly, thus making the
plastron too short.

Accompanying the plastral bones here described is another bone
which must he regarded as the nuchal. Considerable portions of it
are wanting at each lateral extremity ; and the tip of the process which
projects backward toward the first dorsal neural arch is also broken
away. The portion of the bone remaining projects outward on each

60 FIELD COLUMBIAN MUSEUM — ZOOLOGY, VOL. t.

side of the middle line less than 18 cm. If the length of the bone had
the same ratio to the remainder of the carapace of Protostega that we
find in Chelonia, it should extend laterally about 40 cm. That it had
this length so as to reach the first marginal, is quite probable. If the
anterio-posterior extent of the bone were equal to that of Chclonia, it
would be about 30 cm. at the narrowest part; but it is only 6 cm. In-
deed, the portion remaining appears to represent little more than the
median, backwardly projecting process and the anterior thickened
border of the nuchal of Chelonia. The reduction in the anterio-pos-
terior direction really appears to have gone further than in Dermochc-
lys even. In the latter, however, the anterior border of the bone has
been removed, so that it, as well as the other borders, are jagged
and thin. In Protostega it is the hinder border of the bone which has
been removed.

The anterior border of the bone is relatively thick, 3 cm., and is
somewhat bevelled, so as to look downward and forward. On the
upper surface, near the anterior border on each side, is a broad shal-
low groove. The process which is seen to extend backward from the
body of the nuchal probably reached the first neural. It must then
have had a length of about 28 cm.

As in the case of other Cretaceous marine turtles, there is found
on the under surface of the nuchal no tubercle for articulation with
the last cervical vertebra.

As regards the presence of a dermal carapace of mosaic-like
plates, such -as is found in Dermochelys, the remains here described
afford no light. No evidence of its presence has been furnished by
any of the specimens of Protostega so far produced. It is neverthe-
less too early to assure ourselve that there was no such a structure,
considering how easily it could become detached from a carcass
which was being tumbled about by the waves and dragged by carni-
vorous lizards.

Professor Cope has described some of the vertebrae and ribs of
Protostega. The vertebras, like the remainder of the skeleton, had
been greatly flattened by pressure, and probably to this circumstance
is to be attributed their relatively great width. The true relation-
ships to the vertebral axis were thus rendered obscure. Notwith-
standing the possession of ball and socket articulatory surfaces, it was
thought that some of these vertebras belonged to the dorsal region.
Others were regarded as appertaining to the neck.

The length of the shortest cervical vertebra, the first behind the
axis, in a specimen of Chelonia with carapace 790 mm. long is 35 mm.
Professor Cope's specimen of Protostega had apparently close to three

Nov.2i,i895. SKELETON OF Protostega gigas — HAY. 61

times this length, and we might therefore infer that the shortest cer-
vical would have a length of about no mm. The longest vertebra in
his possession was only 60 mm. long and had at least one plane sur-
face. It is quite improbable therefore that it belonged to the ani-
mal's neck. The longest neck vertebra, the last but one, of a speci-
men as large as the one described by Professor Cope should have a
length of about 142 mm., and the longest dorsal vertebra, the third,
should have a length of near 270 mm. Professor Cope's account of
the longest vertebra in his possesion makes it not improbable that it
was the first sacral. The other vertebrae almost certainly belonged to
the tail. Their size and the form of their articular surfaces both
support this interpretation.

Ten ribs were in Professor Cope's hands. Each had a proximal
expansion, and it was evident that these ribs did not unite suturally so
as to form a carapace. But since the dorsal vertebras were regarded
as being so small, the conclusion was reached that either the expanded
proximal ends interfered with each other in the middle line or the
ribs must have been articulated to diapophyses. Since, however, the
dorsal vertebrae would have varied in length from 1 08 to 275 mm.,
and would have been proportionally wide, while the widest rib de-
scribed is 140 mm. at the proximal end, there is no necessity for be-
lieving that any rib touched either its fellow or its neighbors. The
second, third, fourth, and fifth vertebras probably ranged from 250
to 275 mm., and the next two or three were not much shorter. In
Dermochelys and Protosphargis the ribs in front of the fifth from the
last are little, if any, broader than this fifth. Hence we may safely
conclude that there were wide spaces between the ribs even near the
vertebral column. The ribs certainly lacked little of having reached
as advanced a stage of reduction as they have in Dermoehelys. Their
condition was probably much like that seen in Capellini's restoration
of Protosphargis.

Professor Cope estimated that the head of the individual which he
described had a length of 24^6 inches. However, basing my estimate
on the length of the maxillary bone as figured on plate X of "Cretace-
ous Vertebrata," and making the ratio of this maxillary to the length
of the skull the same as that found in Thalassochelys, I can make the
whole length of the skull, including the supraoccipital spine, only
about 1 8 inches, or 45 cm. The distance from the snout to the con-
dyle would be close to 13 inches, or 32 cm. Professor Cope's specimen,
judging from the size of the plastral bones in his possession, was not
much smaller than my own. Hence if we estimate at 32 cm the head,
measured to the condyle, we shall probably not make it too great.

62 FIELD COLUMBIAN MUSEUM — ZOOLOGY, VOL. i.

On the other hand, a study of the figures of the parts of the
skull on plates X and XI of the work cited renders it highly probable
that the bone figured on plate X as the maxillary is not such, but the
postfrontal; while the figure on plate XI, said to represent the post-
frontal, really portrays the maxillary, prefrontal, vomer, and pala-
tine. In such case, the length of the skull would be about a fourth
greater, or 40 cm.

The length of the carapace of Chelonia has a ratio to the plastron
of about 31 to 24. Hence the length of the carapace of my specimen
must have been close to 3. i metres. The neck of our living marine
turtles projects beyond the front of the carapace a distance equal to
at least one-sixth of the length of the carapace. Hence, we are safe
in allowing 50 cm. for the neck outside of the shell. We have there-
fore for the length of this turtle the following figures :

Head 32 metres.

Neck beyond carapace 50

Carapace 3.10

Total 3.92 metres.

XM-UMBIAN V

ZOOLOGY, PL. VII.

SKULL OF CERVUS 7 — FEMALE.

62 Finn --;Y, VOL. i.

O study of the figures of the parts of the

sku'': renders it highly probable

th«tt • i , but the

' the post-
,iry, prefron.

skull would be al <urth

•Ionia has a ratio to the :
24. V. • >f the carapace of my

been close to 3.1 metres. .< of our living marine

;.s beyond the front of the carapace a distance equal to
At least on«?-&uth of the length of the carapace. Hence, we are safe
in allowing 50 cm. for the neck oui he shell. We have there-

fore for the length of this turtle tl iigures:

Head ..... r. ...» ...... .32 me ;

Neck beyond carapace .50

Carapace

PLATE VII.

Skull of Cervus ? — Female.

Fig. i. View from above.

2. " " side.

3. " " below.

FIELD COLUMBIAN MUSEUM.

ZOOLOGY, PL. VII.

SKULL OF CERVUS ? — FEMALE.

FIELD COLUMBIAN ML-

ZOOLOGY, PL. VIII.

HEAD OF CERVUS STEERII. — MALE.

PLATE VIII.
Head of Cervus steerii. — Male.

FIELD COLUMBIAN MUSEUM

ZOOLOGY, PL. VIII.

HEAD OF CERVUS STEERII. — MALE.

COLUMBIAN Mv

• .JY, P

SKULL OF CERVUS STEERII.— MALE.

PLATE IX.

Skull of Cervus steerii.— Male.

Fig. i. View from above.

2. " " side.

3. " " below.

FIELD COLUMBIAN MUSEUM.

ZOOLOGY, PL. IX.

SKULL OF CERVUS STEERII. — MALE.

FIELD COLUMBIAN MUSEUM.

^GV, PL. X.

'

PLATE X.

Skull of Cervus steerii. — Female.

Fig. i. View from above.

2. " " side.

3. " " below.

FIELD COLUMBIAN MUSEUM.

ZOOLOGY, PL. X.

SKULL OF CERVUS STEERII. — FEMALE.

FIELD COLUMBIAN MUSEUM,

ZOOLOG*, Ki.. XI.

SKULL of

NIGRICANS.— MALE

PLATE XI.

Skull of Tragulus nigricans. — Male.
Fig. i. View from above.

2. " " side.

3. " " below.

FIELD COLUMBIAN MUSEUM.

ZOOLOGY, PL. XI.

SKULL OF TRAGULUS NIGRICANS. — MALE.

.'OOLOGY, PL. XII.

... •

rntni ^IHBHBBBH

SKULL OF PTEROPUS AURI-NUCHALIS. — MALE.

PLATE XII.

Skull of Pteropus auri-nuchalis. — Male.
Fig. i. View from above.

2. " " side.

3. " " below.

4. " of lower jaw.

FIELD COLUMBIAN MUSEUM.

ZOOLOGY, PL. XII.

SKULL OF PTEROPUS AURI-NUCHALIS. — MALE.

FIELD COLUMBIAN Ml

7OOLCGV, PC Xlll.

:TA J'4

PLATE XIII.

Skull of Pteropus lucifer. — Male.

Fig. i. View from above.
2. " " side.
. " " below.

FIELD COLUMBIAN MUSEUM.

ZOOLOGY, PL XIII.

SKULL OF PTEROPUS LUCIFER. — MALE,

.