Structure of the Scales of Fishes. 279 



of hollow tubes, -which may be truly called canals. These canals tra- 

 verse the scale longitudinally ; they tend towards a focus which, as we 

 shall afterwards shew, is a centre of nutrition, a place where the tissue is 

 found in its development. Such of them as are situate in the basal field, 

 are in direct relation with the skin ; they do not exist in the terminal 

 and lateral fields but when these are covered by the skin. Would it be 

 too rasli, then, to infer from these premises, that the canals serve to 

 transport the nutritive juices of the skin towards the centre of nutrition 

 and development, and that they have an intimate connection with the 

 organization of the scale ? At the same time, we wish to advance no- 

 thing before having confirmed this opinion, by the study of scales in 

 their development ; but to us, there appears, even at this stage of our 

 investigation, no doubt that these canals perform the functions of true 

 nutritious vessels. 



The longitudinal lines do not always reach the focus, and after leav- 

 ing that point, they do not alwaj's extend to the periphery. 



Cellular lines. — The greatest difficulty we have experienced in study- 

 ing the intimate structure of scales, was the explanation of the nume- 

 rous lines which run along them in a direction more or less parallel to 

 their edges. We cannot, for many reasons, admit of the explanation 

 given by authors, who have seen nothing more in them than the edges of 

 the secreted layers. 



A single microscopical inspection refutes this opinion. If we examine 

 one of the scales, even the most perfectly formed of Corvhia nigra, (pi. III. 

 fig. 9), for example, we perceive that these lines rise above the surface, 

 and that they are placed on a separate base, which is different from the 

 under layer of the scale, and the corpuscles. 



These lines are very short, and as it were divided into small pieces, in 

 the middle of the scale ; they are interrupted by the longitudinal vessels. 

 We often find, in scales preserved among spirits of wine, large spaces de- 

 prived of these striae ; by rubbing the surface, it is easy to remove any 

 quantity. How can we find in all these circumstances, any thing to fa- 

 vour the opinion that these lines are the edges of the homogeneous laj'crs 

 of the scale ? How can we fail, on the contrary-, to regard them as sub- 

 versive of the notion hitlicrto held by authors .'' 



We ought, therefore, at the very outset, to reject these ideas, and 

 seek for an explanation in the comparative study of scales. This wo 

 liave found by observing scales which may be said to be membranous, 

 and much less advanced in development and complication. By these 

 means, we have arrived at the certainty that these lilies owe their origin 

 to the cells which, in the first instance, are formed in the upper layer of 

 the scale, placed on a base ; that, by slow degrees, these cells become 

 filled, elongated, and at last represent lines of variable width, which, at 

 the most, by an unequal edge, revc;il their primitive nature. We shall 

 now give examples. 



Let us first examine the scales of Ophidium barbutum, which are very 



