6 RAY LANKESTER, FOSSIL FISHES OF SPITZBERGEN. 



submitted to me any scales of anotlier type sucli as the rhomboid triangulär scales 

 which are stated by D:r Nathorst to occur in the same beds. 



The scales under description are clearly those of a Glyptodipterine fish — and 

 most nearly resemble those of one of the Carboniferous Rhizodonts — especially Strep- 

 sodus. Their very slight thickness and the character of their superficial markings 

 (well-shewn in the drawings figs 7, 9, 10, 11) is in favour of their reference to a 

 Carboniferous Rhizodont rather than to an upper Devonian Holoptychius. 



At the same time there would be notliing very remai-kable in the occurrence of 

 such a Rhizodont as that indicated by these scales in upper Devonian strata. 



2. TEETH. 



In figs 13, 14, 15, 18 are figured several specimens of conical teeth from the same 

 beds as those which furnish the scales. Those drawn in fig 18 are smaller and diffe- 

 rent in curvature from the other three specimens and probably belong to a distinct 

 genus. There is nothing in the surface structure of the other three (figs 13, 14, 15) to 

 forbid our associating them with the Strepsodus-like scales: and it seems highly pro- 

 bable that the scales and teeth belong to one and the same organism. I have not made 

 a microscopical examination of the structure of the teeth which might throw some liglit 

 on their affinities. 



The fact that the fluting of the teeth drawn in figs. 13, 14 and 15 is carried up 

 to the apex of the tooth is not favourable to their reference to a Carboniferous Rhi- 

 zodont, the fluting of the teeth in those forms being confined to the base of the tooth. 

 The fluting of the present teeth is similar to that of the Devonian genus Dendrodus. 

 Much weight, however, cannot be attached to this character. The existence of largcr 

 and smaller ridges in connection with the fluting (well shewn in fig. 14) is a feature 

 which may assist in generic Identification, though any such Identification based merely 

 on the superficial characters of a simple form of tooth can have little value. 



3. BONES Avith TUBERCULAR ORNAMENTATION. 



In figs 16 and 17 are drawn two bony fragments presenting a rich tuber- 

 cular ornamentation of the surface which was in life immediately subjacent to the 

 epidermis. 



They are probably fragments of cranial bones of a large Rhizodont fish — possibly 

 identical with that to which the scales and teeth belonged — possibly distinct. Be- 

 yond this suggestion, the fragmentary state of the specimens does not enable me to go. 



Fragmentary as these remains are — both those from the lower and those from 

 the higher horizon — yet they are worth figuring and recording, since it is not im- 

 probable that either in the same locality or in some other not widely remote, new 

 fragments will be discovered and identified with these, tending to complete our 

 knowledge of the zoological position of the fishes in question and even more certainly 

 tending to fix with assurance the geological horizon of the beds in which they occur. 



