Mr Connell o« the Chemical Nature of Fossil Scales. 301 



riments as have since been made, are confirmatory to a great ex- 

 tent of these views. Chevreul found the scales of various re- 

 cent osseous fishes to contain large quantities of phosphate and 

 carbonate of lime, and to be very analogous to bone. I have 

 made the same observation in regard to cartilaginous fishes, ha- 

 ving obtained from the dorsal scale of a sturgeon (Acipenser Stu- 

 rio), 53.85 per cent, of bone-earth. * I have also examined the 

 scales of one or two recent crocodiles, principally with the view 

 of ascertaining what proportion of indestructible earthy matter 

 entered into their constitution, and have found that its amount 

 was very trifling. A scale from the thorax of a Nile crocodile 

 about eight feet long left, when completely incinerated, little 

 more than one per cent, of incombustible matter, which consisted 

 of phosphate and carbonate of lime. The carinated dorsal 

 scales of crocodiles appear to contain a rather larger proportion 

 of bone earths. In one of them, taken from the back of a small 

 Nile crocodile, four feet long, I found that a kind of bony axis, 

 consisting principally of phosphate, and carbonate of lime, tra- 

 versed the scale longitudinally ; but still the proportion of 

 earthy matter in the whole scale, including this bony axis, did 

 not much exceed 3 per cent. From the general appearance of 

 the carinated scales of larger crocodiles, it seems likely that 

 the proportion will be found to be larger ; but I have not had 

 any opportunity of examining them. The flat scales of croco- 

 diles, however, such as those on the sides and belly, there seems 

 no reason to doubt, consist almost entirely of perishable animal 

 matter, with only such a minute portion of bone-earth as above 

 stated ; and therefore differ essentially from the flat scales of 

 recent fish, which contain about half their weight of bone-earth, 

 and may be considered as true bone. 



It is evident that these considerations afford a clew to the so- 

 lution of the proposed question. When a fish becomes fossilised, 

 the bone-earth of its scales will survive like that of its other 

 bones, whilst the other animal matter entering into their consti- 



• It thus appears, that although the internal skeleton of a fish may be en- 

 tirely cartilaginous, the same animal may nevertheless have a portion of 

 its external covering composed of true bone; a circumstance which seems to 

 increase the probability, that the fossil Gyracantlius of Burdiehouse, al- 

 though possessed of external bony fin rays, may have been a cartilaginous 

 flah. See the Memoirs, Edin. Trans, vol. xiii. 



VOL. XIX NO. I.XXVIII. OCTOBER 1835. X 



