10 
can be distinguished. The fragmentary detached head (No. c ), which cannot 
lie ascribed to P. crassus with certainty, exhibits a coarse but close tubercular 
ornament both on the mandible and on the branehiostegal rays, which are 
eighteen in number. The same fossil bears traces of clustered small and 
slender conical teeth, with a row of somewhat larger teeth at wide intervals. 
Vertebral Axis . — As shown in PI. II, Fig. 4, the neural spines (n.s.) 
in the region anterior to the dorsal fin are very long and slender. They 
exhibit a central cavity in the fossil, and must thus have been Incompletely 
calcified. Obscure traces of them may also be seen just behind the head in the 
type specimen. 
Appendicular Skeleton . — As usual in Palceoniscus, all the fins are 
relatively small. The rather stout fin-rays are closely articulated to the base, 
and they are ornamented with longitudinal flutings, of which there are traces 
in all the fins of the type specimen. Small and delicate fulcra occur on the 
anterior border of the dorsal fin in both specimens, and they would doubtless 
be seen on the other tins if these were sufficiently well preserved. There are 
indications of the clavicle (cl.) and the forwardly extended infraclavicle (icl.) 
in the type specimen, and the pectoral fin ( pet .) is imperfect. A pelvic fin 
(plv.) is nearly complete, and is distinctly deeper than long, with about twenty 
rays articulated quite to the base. Its foremost five or six rays gradually 
increase in length to the longest. The dorsal fin (d.), which arises immediately 
opposite the hinder end of the pelvic pair, is also somewhat deeper than 
long, with about thirty rays, of which the foremost few gradually increase in 
length to the longest. The anal fin (a.) is incomplete in the fossil, but must 
have been smaller than the dorsal fin, which is well in advance. Of the caudal 
fin (<?.) only fragments remain. 
Squamation. — There are about forty-live transverse series of scales 
between the occiput and the dorsal fin ; and those of the caudal region are 
probably not more numerous. The principal scales of the flank are distinctly 
deeper than long, while those towards the ventral margin are seen to be 
somewdiat longer than deep. The squamation, however, is much obscured by 
the reticulating fissures already mentioned as traversing the head-bones. 
Traces of the pcg-and-socket articulation are observable in some parts, and 
fine oblique sculpturing can be seen on the scales immediately above and 
behind the head ; but it is uncertain whether the hinder border of the scales 
was serrated, There is obviously no series of very large ridge-scales, but 
