20 EVENINGS AT THE MICROSCOPE. 



of the belly, and the last (/) from the throat. Thus we 

 see there is considerable variety in form presented by the 

 scales even of the same individual fish. They all, how- 

 ever, differ from those of the Perch, in this respect ; — that 

 their free overlapping edges are entire, or destitute of the 

 crystalline points which we saw in the former examples ; 

 while they agree in having the front edges, by which they 

 are during life imbedded in the skin, cut into waves or 

 sinuosities. The lower part, as we now look at them, is 

 the free portion of each, which alone is visible in the living 

 fish, the other parts being concealed by the three neigh- 

 bouring scales that overlap it, — above, in front, and below. 



In those from the lateral line, the tube already referred 

 to is seen to pervade each, running through it longitudi- 

 nally, so that it opens behind on the outer surface, and in 

 front on the inner or under surface of the scale. In the 

 scales near the front of the line, just behind the head, the 

 tube is large and prominent (a), while in the scales at the 

 opposite extremity it becomes slender ; diminishing, in the 

 very last scale, viz., at the commencement of the tail-fin, to 

 a mere groove. 



The whole surface of each scale, when viewed under a 

 lens of low power, is seen to be covered with concentric 

 lines, following the irregular sinuosities of the outline. 

 These lines are the edges of the successive layers of which 

 the scale is believed to be composed, each layer being 

 added in the process of growth to the under surface, and 

 each being a little larger every way than its predecessor ; 

 thus the scale is a very depressed cone, of which the centre 

 is the apex. There is a marked difference (indicated in 

 the figures) between that part of the surface which is ex- 

 posed, and that which is covered by the other scales ; the 

 concentric marks in the former are much coarser and less 

 regular, often being interrupted, and seeming to run into 

 each other, and frequently swelling into oval scars. This 

 may, perhaps, be owing to the surface having been par- 



