CORRESPONDENCE. 



171 



and finished lateral outline of the posterior head-plate, I think it extremely 

 improbable that this species ever possessed the lateral small plates forming 

 the cornua of Pteraspis rostratus ; certainly none of the many fragments in 

 our collections show any vestige of these. One of my specimens has the 

 occipital spine in situ, and in several of Mr. Mitchell's the spine is shown 

 detached ; thus differing from the occipital crest of Cephalaspis, which 

 forms an integral portion of the head plate. This spine is short, stout, 

 striated longitudinally, and is deeply inserted in the substance of the head 

 plate, in which it seems to have been inmovably fixed. In their composi- 

 tion the head-plates are quite similar to that of the English species, some 

 of Mr. Mitchell's specimens having, as noticed by him, the exterior stria- 

 tion, the internal reticulated markings, and the inner nacreous plates or 

 lamella) well preserved. From his specimens I have little doubt that the 

 perforations at b, in the figure given in your number for March last, are 

 indeed the eye orbits ; while those at a are too distinctly marked to have 

 been the result of accidental fracture, whatever may have been their nature. 

 As drawn in Mr. Mitchell's latter restoration, and in my figure (Geol. 

 Feb. 1863), the test consists of only two distinct plates, an anterior and 

 posterior, with a distinct spine. 



JSTo light has yet been thrown on the nature of the under surface of the 

 head, some of our many fragments may possibly belong to this part of its 

 body ; to me, however, they all seem mere broken fragments of the upper 

 cephalic plates. 



As to the oral appendages, until very recently I was of opinion that 

 these were of the nature indicated in Mr. Lankester's letter, both in this 

 genus and in Cephalaspis. This opinion was founded not only on negative 

 evidence, but also on the form of the plate protecting the under surface of 

 the head of the latter genus, and in my letter referred to (Feb. 1863) I 

 expressed this conviction pretty strongly. During the course of last au- 

 tumn, however, I had the good fortune to open out some magnificent spe- 

 cimens of Cephalaspis Lyelli, in which the position and character of the 

 mouth and teeth are distinctly exhibited. The mouth opened immediately 

 under the cephalic plate, the gape occupying about one-third of the entire 

 outer margin, the upper maxillae, or jaws, anchylosed with the cephalic 

 plate, forming an integral part of it, and are finished with a single row of 

 short, stout, slightly flattened teeth, which extend quite round to the cornua 

 or cusps. In one of my specimens, a portion of the lower jaw is preserved 

 with its single row of similar teeth. From the decided analogy between 

 Cephalaspis and Pteraspis, it is probable that the latter had been similarly 

 provided. One of Mr. Mitchell's specimens seems to bear this out, having, 

 as noticed by him, the anterior margin of the anterior plate turned down- 

 wards and inwards, as in all our moderately well preserved heads of Ce- 

 phalaspis. The analogy between these genera is further confirmed by Mr. 

 Lankester's most interesting discovery of the scales of Pteraspis, stated 

 by him to be similar to the dorsal series of Cephalaspis, — meaning, I pre- 

 sume, the bony rings covering the body of this creature. These, however, 

 my specimens show to have been again covered externally by scales similar 

 to those covering the cephalic plate. 



It seems to me that as yet the nature of the Cephalaspidae is very imper- 

 fectly understood. I strongly suspect the cephalic plate to consist of the 

 A r arious cranial bones anchylosed, while the bony rings protecting the body 

 equally appear to represent the vertebras and ribs ; all covered externally 

 with scales, or rather dermal scutes, thus indicating that this family may 

 have held among the fishes a place somewhat, although by no means ex- 



