Mr Connell on the Chemical Nature of Fossil Scales. 305 



inent of the animal matter has evidently been made chiefly by 

 carbonate of lime, of which the matrix consists ; and if we suj> 

 pose that a little less than one-half of the original animal mat- 

 ter has given place to calcareous and a small quantity of sili- 

 ceous matter, whilst the larger proportion has been dissipated 

 without substitution, the scale in its original state would have 

 a constitution very analogous to the others ; the apparently 

 larger proportion of phosphate of lime being in this way ac- 

 counted for. To what extent the analogy may have gone in 

 other respects between the Megalichthys of the carboniferous 

 strata of Burdiehouse, and the fish of the far more recent beds 

 of Tilgate, is a point which must be left to the decision of those 

 better qualified for the task, upon a due consideration of all the 

 accompanying fossils. 



There seems every reason to believe that the occurrence of 

 true scales of saurian animals, in a fossil state, is very rare, 

 compared to that of fish scales, — a fact which the chemical na- 

 ture of the former fully prepares us to anticipate ; and I have 

 never had any opportunity of examining, or indeed of even see- 

 ing, an undoubted saurian fossil scale. In the Ossemens Fos- 

 siles of Cuvier, I observe only two instances of such remains, 

 notwithstanding the numerous fossil animals of this kind de- 

 scribed in that work. One of these occurs in a fine calcareous 

 slab containing vertebrae and rii)s of the fossil Gavial of Caen, 

 in Normandy, in which several rows of scales occur in the pre- 

 cise relative position which they must have occupied in life *, 

 and as to which, therefore, not a doubt can exist that they are 

 of true saurian origin. The account given of them by Cuvier 

 is, that " they differ from those of living crocodiles more than 

 any other part of the skeleton, and this crocodile of Caen was 

 undoubtedly the best mailed of the whole genus. The scales 

 are very thick, rectangular, thinner towards the edge, and their 

 whole external surface is hollowed into little hemispherical ca- 

 vities, of the size of a lentil or pea, and pressed against one 

 another." If the original chemical composition of these scales 

 was the same as those of living reptiles, which there is every 

 reason to anticipate, they ought to contain very little phosj)hate 

 of lime, and will in all likelihood be found now to consist prin- 



• Oss. I'os. torn. v. jiart 2, p. 139. PI. vii. fig. 14. 



