from the Shales of the Northumberland Coal-field. 355 



has been carefully removed from the upperside of the large 

 sphenoid bone of this species by which the size of the fish 

 was estimated. And now this interesting specimen reveals to 

 us the cranial bones of the occipital region in an undisturbed 

 and excellent state of preservation. The whole of the bones 

 of one side are almost perfect ; so that there is no difficulty in 

 restoring this portion of the cranium, the constituent bones of 

 which are arranged exactly as they are in the figure of the 

 "cranial buckler" oi Dipterus given by Hugh Miller in his 

 ' Footprints of the Creator.' 



The bones vary little in size, and, with the exception of the 

 central occipital and parietals, are mostly irregularly penta- 

 gonal. There are three occipitals : the central one is not much 

 larger than the lateral ; the former is nearly as wide as it is 

 long, and is seven-sided, with the anterior margin a little 

 pointed in the centre, and the posterior margin nearly straight. 

 The lateral occipitals are connected with the postero-lateral 

 margins of the central occipital, and, diverging in front, admit 

 a bone on each side, which is wedged in between them and 

 the antero-lateral borders of the central occipital and the ex- 

 ternal margins of the parietals. External to these bones, and 

 in connexion with their outer margins, are three other bones, 

 which form the lateral borders of the cranium. In all there 

 are five bones on each side of the central occipital and poste- 

 rior part of the parietals. Only a small portion of the left 

 parietal is preserved ; but enough is present to show that this 

 pair of bones are elongated, being widest apparently a little be- 

 hind their centre, and having their posterior margins slightly 

 divergent to receive the anterior angle of the central occipital. 



The surface of the bones is not ornamented with " waved 

 and bent lines," as those of DiiJterus are described to be by 

 Miller {ibid, p. 61), but is minutely granulated and punctate, 

 similar to that of the opercles described in the paper on Gte- 

 nodus already referred to, and here and there are indications 

 of the radial bone-structm'e beneath. 



The original estimate of the width of this head was nine 

 inches. It is now evident that it really was eight and a half 

 inches across the occipital region, without taking into account 

 a fragmentary bone, probably a portion of an operculum. 

 Were this added to the above measurement, the width would 

 be ten inches. 



The external characters of the palatal plates of the various 

 species of Gtenodus were described in the paper on that genus 

 mentioned above. Nothing, however, was said of the internal 

 structure, such matters of detail having been reserved for some 

 future occasion. But it is now perhaps desirable to give some 



