LIFE ON THE EARTH. 103 



Fishes commencing, as already stated, in the 

 upper Silurian Strata, become from that point highly 

 important in geological history more important than 

 even their numerous remains at first seem to indicate ; 

 for though teeth, scales, and fin-rays of these animals 

 are not scarce in the strata, these, the most conservable 

 of the hard parts of fishes, are scattered irregularly, 

 and, until studied after the method of Agassiz, give 

 but slight information. Under the hands of this great 

 naturalist and his disciples, Egerton and Enniskillen, 

 we have seen the history of fossil fishes grow to em- 

 brace many hundreds of distinct forms, very interesting 

 in physiology, always very valuable in geological rea- 

 soning. The method of Agassiz is no doubt in some 

 degree conventional, and specially framed for the study 

 of fossils, yet the characters derived from the dermal 

 covering are always of high value in the classification 

 of the vertebrata, and specially influential in fishes. 



Two great orders of fishes have enamelled scales 

 e. g. Placoid and Ganoid Fishes the latter have ena- 

 mel externally, bone internally for each scale, and 

 the scales so closely packed as to constitute a real 

 dermo-skeleton. These orders occur in all the strata 

 above the Silurians, and still exist : they are the 

 only orders which occur in the Palaeozoic rocks ; in 

 the existing ocean and in the Tertiary Strata they 

 are comparatively the least abundant. 



