62 RAMSAY H. TRAQUAIR’S 
Tarrasius problematicus, sp. nov. Traquair. 
Pl. IV. figs 4-6. 
The specimen from Tarras Foot.—This is a fragment (Pl. IV. figs. 4, 5) display- 
ing what is evidently the posterior or caudal part of a small fish, cut off both in 
front and behind, and measuring 14 inch. The shape of the portion of the body 
here shown is gently tapering, the depth in front being 7% inch, and + inch 
where it is cut off behind ; the dorsal and ventral margins are nearly straight, 
being only very slightly convex. The whole surface is covered with regularly 
arranged, minute, but proportionally thick, rhombic, and apparently non-over- 
lapping scales, each of which shows on its external brilliant surface a small 
shailow depression or sulcus. At the anterior part of the fragment the 
internal skeleton is clearly displayed by the removal of the scales next the 
eye of the observer. There are no vertebral bodies visible, but four neural (?) 
arches are seen, from which proceed obliquely upwards and backwards as 
many neural(?) spines, in front of which two others are seen, whose support- 
ing arches are not included in the specimen. Above these spines comes a 
series of slender interspinous elements, distally enlarged and laterally flattened, 
while proximally they pass for a little way between the extremities of the 
neural(?) spines, after the manner of modern fishes. Appended to the 
extremities of the last described elements, and set at a slight angle, there 
seems to me to be a second set of interspinous elements, minute, short, and 
somewhat hour-glass-shaped, but owing to the minuteness of the parts it is 
not easy to distinguish them accurately from the proximal extremities of the 
suceeding fin-rays. 
The whole of the dorsal(?) margin exhibited in the specimen is bordered 
by a continuous fin, the depth of which is equal to two thirds of that of the 
part of the body to which it is appended. This fin consists of imnumerable 
closely set rays, distinctly articulated, and tapering distally to fine points, 
but so far as can be observed, not dichotomising. As in the continuous dorso- 
caudal fin of Lepidosiren and Ceratodus, their direction becomes posteriorly 
more and more oblique, until at the posterior end of the fragment they are in 
fact nearly horizontal. From this there can be hardly a doubt but that it is 
the tail of the fish with which we have to deal, that the caudal fin was diphy- 
cercal, and continuous with the dorsal and anal. 
On the hemal(?) aspect of the vertebral axis no arches or spines are 
distinctly exposed, but their presence is betrayed by oblique elevations of the 
scaly surface, exactly symmetrical with the spines of the opposite aspect. 
Along the ventral(?) margin also the impressions of a set of interspinous bones 
are seen, exactly corresponding with those which follow on the neural (?) spines 
opposite, so that although the fin itself is unfortunately lost, we may very 
