NOTES. 309 



Here matters of fact seem very much to rest. In re^^ard 

 to our present knowledge of tlie Cephalaspis, its classification 

 and analogies are involved in very considerable obscurity. 

 In a paper read before the Geological Society on 1 8th May 

 1859, by Sir Philip Grey Egerton, entitled " Paloeichthyologic 

 Notes," I find the following paragraph. It serves to show 

 the more recent attempts at classification as respects the As- 

 terolepis, as well as the Cephalaspis. 



FAMILY PLACODERMATA.* 

 " This family definition was first proposed by Professor M'Coy for 

 the reception of all the genera included in the Cephalaspidae (Agas- 

 bIz), except Cephalaspis, together with some other genera assigned by 

 Agassiz to the Celacanths. Professor Pander has adopted this family 

 term, and includes in it the following five genera : — Pterichthys, Agass. 

 (Asterolepis, Eichen.) ; Coccosteus, Agass. ; Asterolepis, Agass. (Ho- 

 raosteus, Asmus, — 'this is the Asterolepis of the text') ; Heterosteus, 

 Asmus ; and Chelyophorus, Agass. The distinction, as I (Su: P. Eger- 

 ton) understand it, between this family and the Cephalaspidae, is ('that 

 is, is meant to be by those who make this classification'), that whereas 

 in the latter the head only is encased, in the former the thorax is also 

 invested with bony plates. Cephalaspis, Pteraspis, and Auchenaspis, 

 would consequently constitute the Cephalaspid family,— Pterichthys and 

 Coccosteus being the types of the Placodermy. Chelyophorus is probably 

 a member of the same family. "Whether Asterolepis and Heterosteus 

 belong to it, must depend upon further investigation. Hugh Miller de- 

 scribes the plates of Asterolepis as the homologues of true cranial bones, 

 and assigns to this fish the scales figured at page 68 of the * Footprints' 

 [present edition]. Pander, on the contrary, maintains that the small ante- 

 rior plates only are true cranial plates, and that the larger plates are homo- 

 logues to the thoracic plates of the Coccosteus. The plate figured by 

 Hugh Miller at p. 88 as a hyoid plate. Pander assumes to be the poste-. 

 rior dorsal plate, homologous to the large cuspidate plate of Coccosteus. 

 It is clear, then, that the family afl&nity of this genus must depend upon 

 the solution of this moot point. I regret I have no materials to thro\w 

 light on this subject." — {Proceedings of the Geological Society, Egerton, 

 Old Red Fishes:) 



By this mode of classification the Cephalaspidae, — includ- 

 * Professor Owen objects to this term, as being synonymous with Pla- 

 coidei. 



