254 The Edinburgh New Philosophical Journal. 



of different sizes, but all retaining the shape of the scales. These 

 leaflets are disposed above each other in such a manner, that the 

 smallest occupy the centre of the scale and form its exterior part, 

 while the largest, with their margin projecting beyond the preceding, 

 are soldered successively to their lower surface. We thus see evidently 

 that the concentric lines visible on the surface of scales are simply the 

 edge of the leaflets composing them.' The fact related by M. Agassiz 

 can refer only to the leaflets of the lower bed, which are separated 

 by maceration ; his conclusion, therefore, is founded on a false 

 interpretation, which, in other respects, without the aid of the micros- 

 cope, cannot be correct. All the modifications observed in the form 

 and nature of the surface in scales, arise from the form of the augment- 

 ing leaflets, and the manner in which they are placed above each 

 other. Layers of enamel are deposited on the outside of some scales 

 (as among the Ganoides). 



"With regard to the longitudinal canals, M. Agassiz calls them 

 furrows. ' There are grooves at the margin of their outer surface, 

 which correspond to each other in the different leaflets, and multiply 

 during the growth of the scale.' In describing the different external 

 forms which the contour of scales presents to the naked eye, 

 M. Agassiz likewise mentions the lobes, and continues thus : ' When 

 these lobes are hardened in the form of small teeth or very sharp 

 serratures, and are only found in the last leaflet (the preceding suc- 

 cessively disappearing as they become blunter), scales are then pro- 

 duced having a simple serrated edge; but when they are found on 

 many consecutive leaflets, the edge of the scale is covered with nume- 

 rous rows of points, and it is then very rough to the touch.' 



" We shall afterwards see that our explanation of these points differs 

 materially from that given by M. Agassiz; that we find them to be 

 distinctly organized formations, composed of an envelope and a 

 dentiform body, which presents roots, different surfaces, many degrees 

 of developments according to position, and different forms in different 

 families of fishes. 



" Lastly, M. Agassiz establishes the principal divisions of the class 

 of fishes according to the form of the scales. ' I believe that I have 

 found,' he says, 'in the differences presented by scales, a means of 

 tracing more exactly the natural affinities of all kinds of fishes. It 

 cannot at least be disputed that the animals of this class possess, in 

 their scaly integuments, a character peculiar to themselves, and not 

 found to exist in any other class. The following are the orders and 

 the names of the principal families : 1st Order : The Placoides. — Thus 

 named on account of the irregularity of the solid parts of their integu- 



