40 Proceedings of the Royal Physical Society. 



kiefer." Is it not possible that the exceeding closeness of 

 their anterior margins to the edge of the cranial shield in 

 Whiteaves's figure may be due to a slight forward displace- 

 ment, sucli as often occurs in Pterichthys to a much greater 

 extent ? In a specimen of B. canadensis in the Edinburgh 

 Museum remains of these plates occur, which evidently are 

 so displaced, as they are shoved forwards quite over the edge 

 of the cranial shield. 



I have not seen the small median plate which Whiteaves 

 (same figure, No. 18) represents immediately behind the two 

 last mentioned, and concerning which he remarks, "Judging 

 by analogies with the Asterolepis of Hugh Miller, but not of 

 Pander, this may have been the hyoid plate." Unfortunately 

 for this comparison, the "hyoid" plate of Hugh Miller's 

 Asterolepis {—Homostens) was, thirty years ago, determined 

 to be the median dorsal plate of its carapace (7, p. 76). 



Body-carapace. — This is more depressed in Bothriolepis 

 than in Pterichthys, has a dorso-lateral angulated margin as 

 well as a ventro-lateral one, and the dorsal surface is broader 

 than the ventral one. The median dorsal plates are not so 

 acutely elevated mesially as in Pterichthys ; in some species 

 they are only gently convex on the upper aspect. The ante- 

 rior median dorsal, usually rather wide in its shape, articulates 

 as in Pterichthys (but not as in Asterolepis)^ its antero-lateral 

 margin overlapping the anterior dorso-lateral, while the 

 postero-lateral margin is, on the other hand, overlapped by 

 the posterior dorso-lateral. The inner surface of this ante- 

 rior median dorsal (PI. II., Pig. 3) shows a sharp median 

 ridge, from which anteriorly two short branches are seen to 

 diverge at acute angles forwards and outwards. On the 

 inferior surface of the body the anterior ventro-laterals 

 (PL II., Pig. 5) show a peculiarity in shape which dis- 

 tinguishes them from the corresponding plates in Pterichthys 

 and Asterolepis in not exhibiting in front the prominent 

 emargination for the semilunar plates seen in those genera. 

 In fact, no precisely similar semilunar plates exist, though 

 these seem to be represented by a single small triangular one 

 occupying the median notch at the union of the two anterior 

 ventro-laterals. This is figured by Whiteaves in B. cana- 



