8 WEALDEN AND PURBECK FOSSIL FISHES. 



proves that each ramus in l)oth jaws bears ten or eleven transverse rows of teeth; 

 while one skull in the British Museum (no. P. 3172 a) seems to exhibit an unpaired 

 symphjsial row in the lower jaw. In all the teeth the principal cusp is high and 

 narrow, compressed to two sharp lateral edges, with the incurved apex smooth and 

 the expanded base vertically striated. The one or two pairs of well-defined 

 lateral denticles are striated to the apex. From the symphysis backwards to the 

 middle of each ramus the teeth are highest and about equally elevated ; but in 

 this series those at and near the symphysis have a less extended base than those 

 further back, with less space for the lateral denticles, which are usually in two 

 pairs (the outer very small), but may be flanked by a third minute cusp (PI. I, 

 fig. 1 b). In the hinder half of each ramus the teeth rapidly diminish in size and 

 elevation, Avith the principal cusp curving sharply backwards (Text-fig. 4). There 

 is no essential difference between the teeth of the upper and lower jaws. 



The ceratohyals are massive cartilages seen in several specimens, and the 



,^X.\. ^v\%, .rJ%-4, .\ /I 



Fig. 4. — Hybochts hasamis,'Egevton; four upper and lower tcerli Irom iiiiider lialf of jaws, nat. size. — 



Weald Clay; Pevensey Bay, Sussex. 



basih3"al is also large, somewhat broader than long, as already described in Catal. 

 Foss. Fishes, Brit. Mus., pt. i (1889), p. 274, pi. xii, fig. 2. The branchial arches 

 are only five in number, as shown by the ceratobranchials preserved in series in a 

 specimen already described, loc. cit., p. 274, pi. xii, fig. 8, and as still better seen 

 in another head in the Beckles Collection (B. M. no. P. 11872). The hindmost 

 or fifth arch is comparatively small. Each ceratobranchial is expanded and 

 sharply truncated at its lower end, where it would articulate with the hypo- 

 branchial; but the cartilages of this lower series remain undiscovered. 



The trunk is known only by fragments, of which the best is represented in 

 Text-fig. 5. The notochord must have been persistent, but the neural arches and 

 spines (n.s.) are well calcified in the usual granular form. They are narrow bands 

 of cartilage arranged in close series. Below the space for the notochord in the 

 abdominal region there are also traces of comparatively slender haemal elements 

 or ribs in a specimen described in Proc. Yorks. Geol. Polyt. Soc, vol. xii (1891), 

 p. 65, pi. ii, fig. 1 . 



Of the fins, only parts of the dorsals have hitherto been discovered in the 

 original of Text-fig. 5. In this specimen the anterior dorsal (d 1) probably 

 remains in its natural position, Ixit the second dorsal (r/. 2) is accidentally 



