20 RAMSAY H. TRAQUAIR’S 
Family SAURODIPTERID. 
Genus Megalichthys, Agassiz. 
(Agassiz, Poissons Fossiles, vol. ii. part 2, p. 89.) 
Fragmentary remains, consisting both of scales and head plates, referable 
to the genus Megalichthys, have occurred both at Tweeden Burn and near 
Glencartholm. They are far too imperfect to justify any specific deter- 
mination. 
Family CCLACANTHID&, 
Genus Colacanthus, Agassiz, 1843. 
(Agassiz, Poissons Fossiles, vol. ii. pt. 2, p. 170; Huxley, Dec. Geol. Survey xii. 1866, p. 8.) 
Ceelacanthus lepturus, Agassiz. 
A few scales which I cannot distinguish from those of the common 
Ceelacanthus lepturus of the Coal Measures (Huxley, op. cit. pl, ii. figs. 3 and 4, 
pl. iii. fig. 1a), occur upon a bit of shale from Tweeden Burn, along with scales 
of Eurynotus and Megalichthys. 
Celacanthus Huxleyi, sp. nov. Traquair. 
Pl. I. figs. 1-4. 
Among the specimens from Glencartholm, Eskdale, are five of a 
Ceelacanthus, which seems undoubtedly different from any hitherto described 
member of the genus. They are hardly so perfect as most of the smaller fishes 
from these rocks, and it will therefore be advisable to notice each of the 
specimens in succession. 
No. 1 (fig. 2) is the most perfect as regards figure, but its details are consider- 
ably obscured by a thin and utterly irremovable layer of matrix which adheres 
to its surface. It shows a small fish 14 inch in length, and } inch in depth at 
the first dorsal fin. The head, concerning whose structure nothing more can be 
said, save that it displays the general contour of that of a Calacanthus, is con- 
tained nearly four times in the total, the terminal appendage, or secondary caudal 
fin, being however absent. Both dorsal fins, as well as the principal caudal, are 
tolerably plainly exhibited, all showing the form and structure characteristic of 
the genus, but of the other fins there are no certain traces; and as to scales, only 
the merest “shadows,” as it were, can be distinguished. The internal skeleton 
is also dimly visible, and from the crowded appearance of the vetebral spines it 
seems possible that the short stumpy contour of the fish is to some extent due 
to post-mortem shortening up. Interesting as this little specimen is, it is 
hardly possible to found much upon it beyond the generic diagnosis. 
