THE

QUARTERLY JOURNAL

OF

THE GEOLOGICAL SOCIETY OF LONDON.

PROCEEDINGS

OF

THE GEOLOGICAL SOCIETY.

DECEMBER 19, 1866.

Theodore Cooke, Esq., Principal of the Engineering College, Poonah; and John Starkie Gardner, Esq., Park House, St. John's Wood, N.W., were elected Fellows.

The following communications were read :—

1. On a NEW SPECIMEN of TELERPETON ELGINENSE. By Professor HUXLEY, LL.D., F.R.S., F.G.S.

I AM indebted to my friend the Rev. Dr. Gordon, of Birnie by Elgin, for the opportunity of examining the very beautiful and important specimen of Telerpeton Elginense, of which I propose to give a description in the present paper. It is the property of Mr. James Grant, General Assembly Teacher, in Lossiemouth, Elgin, who has been good enough to entrust it to Dr. Gordon for transmission to me; and it was obtained from the well-known reptiliferous beds of Lossiemouth, along with some highly interesting fragments of Stagonolepis and Hyperodapedon.

The fine-grained sandstone in which the fossil has been imbedded is broken irregularly into five pieces; and the several bones are represented by sharp and well-defined casts, the original osseous substance having disappeared, or being represented only by pulverulent bone-earth, or by oxide of iron. The body is curved towards the right side, and the head and neck are bent dorsad in a plane different from that of the trunk.

The length of the skull is 1.65 in., that of the vertebral column,

VOL. XXIII.-PART I.

H

from the atlas to the anterior margin of the sacrum, is 4.5 in. From the front margin of the sacrum to the end of the undisturbed part of the tail is 2 inches. To this total of 8.15 in. it is probable that 2 inches more, at least, must be added for the distal moiety of the tail, which would give the entire animal a length of not less than between ten and eleven inches.

The fore limb had a length of not less than 2 inches, while the hind limb, when extended, must have measured fully 3 inches.

Figs. A-E.-Outline-sketches of parts of the Skeleton of Telerpeton Elginense, Mantell. (Natural size.)

A

E

A. A side view of the skull.

B. The anterior teeth of the upper and lower jaws.
C. The left hind foot.

D. The right scapula and coracoid.

E. The left half of the pelvis.

The skull is broad, its occipital margin measuring 1.5 in. The cranium retains this width for about half its length, and then rapidly narrows to the snout, with a contour which is at first convex, and then bcomes concave, until it terminates in the roundly truncated snout, which exhibits the remains of two cylindrical incisor teeth, placed side by side and close together (fig. A).

The posterior margin of the skull is nearly straight, but presents an obtuse angle directed backwards in the middle line. The postero-lateral, or parietal, angles of the skull are produced directly outwards, instead of backwards, into short and broad processes, which become connected with the strongly curved squamo-jugal arcades, the inner faces of which were concave and directed somewhat downwards.

The counterpart of the fragment which contains the greater part of the skull exhibits the impressions of part of the roof of the skull with the left orbit and left supratemporal fossa. It proves that the squamosal bone was large, thick, and slightly elongated posteriorly and externally, while its outer edge seems to have been undulated. There is no evidence that any postfrontal bone separated the temporal and the orbital fossæ, the two forming one oval space about 0·7 in. long and 0·4 in. wide. The roof of the skull can be traced forwards, narrowing gradually for a distance of 0.65 in., and then seems to have suddenly contracted to form the interorbital region. Here, however, it is completely hidden by the matrix.

In the prefrontal region it widens out again; and a curious, perfectly separate, sandstone cast of the interior of this part of the skull has been formed. This cast is shaped somewhat like an ace of spades, with a truncated apex and a trilobed base, and presents a few traces of bony matter upon its upper surface. The rest of its exterior is stained, for the most part, of a reddish colour, as if by oxide of iron.

The middle basal lobe presents a truncated, uncoloured surface, where it has broken off from the matrix which lies behind it. The anterior end is similarly fractured and unstained; and there are two oval, uncoloured elevated spaces on the under surface of the cast, which answer to the posterior nares. All the rest of the cast has been enveloped in bone, which must have been furnished, at the sides, by the prefrontals and below by the vomers or the palatine bones. The impression of the upper surface of the vomers has left a strong median groove along the under surface of the cast. The surface of the matrix upon which this cast fits shows the remains of the oral faces of the facial bones, the bony matter itself appearing to be, for the most part, replaced by oxide of iron.

Each of the teeth, already mentioned, which are implanted in the præmaxillæ, is rather less than 0.1 in. wide, and rather more than 0.1 in. long. Their apices appear to have been rounded (fig. B).

The maxillæ are strong, and send up a process behind the external nostril. The roots of several teeth, placed in a single series from before backwards, occupy the alveolar surface of the thick anterior moiety of each. The palatine bones meet, and appear to be completely united in the middle line, nor does any posterior palatine space appear to be left between them and the transverse and maxillary bones. The very imperfectly preserved bony matter which remains is dotted over with rows of red spots of oxide of iron, the arrangement of which forcibly reminds me of that of the palatine teeth in Hyperodapedon, though I cannot make sure that these spots really represent teeth. Posteriorly the palatine bones meet the pterygoids, which diverge and pass backwards, to become connected with the quadrate bones in the ordinary way.

In the interspace between these the remains of the basi-sphenoid are visible.

The quadrate bone, 0.7 in. long, is strong; and its anterior aspect

is convex forwards, and from above downwards. It consists of an outer and an inner lamella, which pass into one another along the ridge-like anterior convexity of the bone, and enclose a deep cavity posteriorly. The distal end of the bone was provided with a transversely elongated convex condyle. The angle of the ramus of the mandible, which is 1.35 in. long, projects not more than 0.15 in. behind its articular surface (fig. A).

At the articular surface, the jaw is not more than 1.5 in. thick ; but in front of this point it rapidly rises, and, at 0-65 in. from the posterior extremity, forms a coronoid process, the summit of which is 0.4 in. distant from the lower edge of the ramus. It then declines in height, and, at 0.75 in. from the angular end, begins to bear teeth. Of these teeth the three posterior occupy a space of Q.25 in., and each has a conical crown 0.1 in. high. Three teeth in the upper jaw, of similar size and form, interlock with them, the hindermost maxillary tooth being posterior to the hindermost mandibular tooth. The next two teeth, forwards, in the ramus of the mandible are somewhat smaller than those just mentioned; but the most anterior tooth of all is a curved tusk, twice as long as any of the others, and having its concave side outwards, its convex side inwards and towards its fellow. These teeth bite behind the two long teeth lodged in the præmaxilla (fig. B).

Three teeth in the upper jaw answer to the foregoing; and the anterior of these passes externally to the mandibular tusk when the mouth is closed. Thus there appear to be six teeth below, and seven teeth above, on each side of the upper and lower jaws; but it is possible that additional posterior teeth may not be visible.

I have carefully examined into the mode of implantation of those teeth, and I have been unable to satisfy myself that they are lodged in true alveoli. They appear to be anchylosed to the edges of the jaw-bones, as in many modern Lizards with a so-called" acrodont" dentition.

Each tooth contains a proportionally large pulp-cavity.

The vertebral column is broken in the middle of the dorsal region, and it is not practicable to ascertain the number of vertebræ with precision; but it may be safely assumed that the cervicodorsal series contains not fewer than twenty, and not more than twenty-two vertebræ. There are certainly not more than two sacral vertebræ. Eleven caudal vertebræ, belonging to the proximal half of the tail, lie in undisturbed relation to one another. There were probably as many more in the broken-up part of the tail.

The casts of these vertebræ show that they had completely ossified centra, very slightly concave at each end; large neural canals, and stout neural arches, running out into broad oblique processes or zygapophyses. The spines were very low narrow crests. The transverse processes must have been represented by mere tubercles.

Five long and slender vertebral ribs are visible on the right side of the anterior dorsal region, in connexion with a similar number of anterior dorsal vertebræ. The largest of these, though its distal end is not entire, measures 1·15 in. in length, but is nowhere more than

0-05 in. broad. These ribs are somewhat expanded at their proximal ends, but show no traces of a division into distinct capitula and tubercula.

On the left side the remains of several vertebral ribs, and a few slender grooves apparently produced by sternal ribs, are to be seen. The ribs of the four or five posterior dorsal vertebræ are exceedingly short; but there appear to be no proper lumbar vertebræ, in the sense of præsacral vertebræ with anchylosed or abortive ribs.

The cast of the only sacral vertebra which is visible shows it to have been possessed of a stout lateral process, 0·16 in. long, and 0-07 in. thick, which abutted against the ilium.

The ilium occupies such a position as to hide any other sacral vertebra which may have existed; but there could not have been more than one additional vertebra in this region of the spine. The eleven anterior caudal vertebræ occupy a space of 1.6 in., which gives rather more than 0.14 in. for the length of each vertebra. The anterior five or six possess slightly curved transverse processes, which taper to their extremities, and attain a length of 0.3 in.

The neural spines and subvertebral bones of the caudal vertebræ are not clearly exhibited.

Three vertebræ, following the tenth caudal, are represented by cylindroidal holes in the matrix, as much as 0.25 in. deep; and beside these lie imperfect impressions of yet two other elongated vertebral centres.

The left shoulder-girdle is displayed on the upper surfaces of the first and second fragments, the line of breakage between which has in fact passed through this system of bones (fig. D).

The scapula was 0.8 in. long; and the cast proves that its proximal, or glenoidal, end was thick and prismatic, and slightly expanded anteriorly. Distally, or dorsally, it passes into a broad and flattened blade, not more than 0.2 in. wide, which is abruptly truncated at its vertebral end.

The coracoid must have been a very stout bone, nearly 0.9 in. in antero-posterior, and 0.5 in. in transverse measurement. Of its three margins the internal presents a convex, the anterior and the posterior a concave curvature. It appears to have possessed a considerable fenestra in the inner moiety of its anterior half.

An obscure impression, leading to a triangular hole, which is the transverse section of a cavity at least 0.2 in. deep, in the matrix, I believe to represent the cast of a clavicle. There is a corresponding hole and cavity in the counterpart fragment of the fossil, whence I conclude that the clavicle must have been about half an inch long.

The impression of the head of the humerus lies in its natural relation to the scapula and coronoid. It was 0.4 in. wide, and had a strongly marked deltoid ridge, or outer tuberosity, which projected downwards. In the counterpart the rest of the cast of the humerus is preserved, and proves that bone to have had a length of about 0-85 in. The middle of its shaft is not more than 0.2 in. wide, but its distal end expands to the width of the head.