268 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. [Jan. 6, 



head " under the " scales," and he particularly mentions and figures 

 the radiating direction of these " fibres ;" but in the * Monograph of 

 the Old Red Sandstone Fishes ' I find the following general remarks 

 applied to the whole of the Cephalaspidce : — 



" It would appear from the condition of the specimens preserved, 

 that all the cranial bones were only protecting plates, which covered 

 a cartilaginous cranium similar to that of the Sturgeons ; at least 

 I have never been able to discover any cranial bones deprived of 

 that characteristic granulation, which indicates that the plates were 

 in direct relation with the integument. Therefore, I think there can 

 be no doubt that all these granular plates rested by their inner and 

 smooth surface on a cranial cartilage, such as is found in cartilaginous 

 fishes and in the embryos of osseous fishes." — Monog. Gres Rouge, 

 p. 3. 



Nevertheless, in speaking of the genus Cephalaspis, a few pages 

 further on, Prof. Agassiz states that he has nothing to add to his 

 previous account of the genus ; so that I am puzzled to know what 

 view I ought to ascribe to him at present. We shall see by-and- 

 by that the last-quoted is the only one warranted by the facts of 

 the case. 



The disk of Cephalaspis Lloydii is said to consist of an external 

 striated enamel, of a middle layer " composed of granules similar to 

 those of the bones of Chondropterygious fishes," and of an internal 

 layer made up of superimposed lamellae. Prof. Agassiz considers that 

 this structure " singularly recalls that of the test of the Crustacea" 



Notwithstanding these, partly real and partly imaginary, differences 

 between his different species of Cephalaspis, Prof. Agassiz found in 

 Cephalaspis rostratus (a species which I have had no opportunity 

 of observing) a form and structure of so transitional a character that 

 he included them all under the same genus. 



That so close an affinity obtains between all the species of Cepha- 

 laspis has, however, been disputed latterly by M. Rudolph Kner, 

 who in 1847 published a memoir in Haidinger's ' Naturwissenschaft- 

 liche Abhandlungen * for the purpose of proving that C. Lloydii 

 and C. Lewisii are not piscine remains at all, but that they are the 

 internal shells of a Cephalopod allied to Sepia, for which he pro- 

 posed the name of Pteraspis. 



M. Kner's reasoning is based upon his examination of the structure 

 of a fossil (evidently closely allied to C. Lloydii) from the Silurian 

 rocks of Gallicia. The form of this fossil, says M. Kner, is very 

 similar to that of C. Lloydii ; but it is larger, having a length of about 

 four inches by a width of two. It consists of three layers. The 

 innermost is shining, bluish-green, enamel-like, and presents four or 

 five distinct lamellae. This layer forms one continuous surface 

 marked in the centre by a longitudinal depression, smaller at one 

 end than at the other, and by obscure radiating lines. The upper 

 part of the conical depression is covered with minute pores or 

 depressions, which are visible in the deeper as well as in the more 

 superficial layers, but become evanescent in its lower part. 



Between the layer of enamel and the prismatic layer which 



