PISHES. 



275 



admirably adapted for living forms, was neither sufficient nor 

 applicable in all cases to the fossil remains of this class. It 

 occurred, therefore, to Professor Agassiz, the great master 

 of this branch of Palaeontology, that, as the scaly appendages 

 of the tegumentary membrane were often preserved in a fossil 

 state when the other structures of the fish had perished, that 

 the scales would afford the palaeontologist the surest basis 

 for the division of the class into orders and families. More- 

 over, as the tegumentary covering of the vertebrated animals 

 harmonises so completely with the other parts of their 

 economy as to be in fact the outward reflection of the internal 

 organisation, the modifications in the structure of the exo- 

 skeleton were taken as the basis of the new classification. On 

 this principle, Agassiz divided the class into four orders : 

 I. The first order, or PLACOID, comprises those so termed 

 from 7rXa|, a broad plate; they have the skin irregularly 

 covered with plates of enamel, sometimes large, as in the 

 rays, sometimes reduced to small points, as in the sharks. 

 (Pig. 184.) 



FIG. 184. 



Fm. 185. 



II. The second order, or G-ANOID, so called from yavos, 

 splendour, because they are covered regularly with angular 

 scales, composed of horny or bony plates, coated externally 

 with a bright enamel. (Fig. 185.) 



FIG. 186. 



FIG. 187 



III. The third, or CTENOID, from KTCIS, a comb, includes 

 such as are characterised by scales, jagged or denticulated 

 on the posterior margin, as in the perch. (Pig. 186.) 



T 2 



