384 FISHES. 



condition, nor lengths of this other chain, just as fanciful, of simul- 

 taneous and slightly varied forms; these chains have no existence 

 whatever, save only in the imagination of some naturalists, much greater 

 poets than observers, but that they belong to this real chain of co- 

 existing beings which are indispensable to one another and to the 

 whole, and which, by their reciprocal influence, sustain the order and 

 harmony of the universe; a chain, of which no link could have existed 

 without all the others, and the foldings of which, being incessantly 

 brought near each other or separated, comprehend the whole globe in 

 its windings. 



Methodical distribution of Fishes into natural Families and higher 



Divisions. 



It will be obvious from all that precedes, that the distinctions be- 

 tween the external and internal peculiar organs which characterize 

 fishes, are not less numerous than they are striking. There are, 

 indeed, few classes of animals in which it is more easy to detect the 

 genera and natural families, and from them to arrange the species. 

 Under the slightest examination each is in a condition to exhibit the 

 relations which place it amongst the herrings for example, or the 

 anchovies, the megalopes, the elopes or chirocentres ; or to the eels, 

 the murcena, symbranchise, or the cecilia. Nor are we Less struck 

 with the affinity of the countless tribes of the cyprins, the silures, 

 salmons, perches, and those which resemble them. But to arrange 

 these genera and families in an orderly manner, it is essential that we 

 should fix on a small number of important characters, from which 

 some great divisions would result, and these, without any interruption 

 of natural relations, to be sufficiently exact so that no doubt could be 

 allowed to exist as to the position of each fish in the classification. 

 And this is the point to which we have not yet come in a sufficiently 

 detailed manner. 



To say the truth, the numerous peculiar characters of the chondrop- 

 terygians or fishes with a true cartilaginous skeleton, or, to speak still 

 more accurately, fishes with the granulated periosteum, were always too 

 prominent not to have been properly estimated by every methodical 

 mind. Every ichthyologist, therefore, has constituted these fishes 

 into a separate order, but almost every one of them has violated the 

 just division which they made in blending with it some fishes which 

 had no other resemblance to them than in the softness of their skeleton. 



Still these latter fishes should not be indiscriminately excluded with 

 the crowd : there are some, such as lophius, and the lump-fish, which, 

 with this softness, almost differ in nothing from the ordinary fishes, 

 and cannot be separated from them ; but there are, also, others with 

 peculiar characters in their integuments, teeth, and especially in the 

 disposition of their skeleton and head. The tetrodons, diodons, 

 coffres, and even the balistes, are of this number. The syngnathes 

 have also in their branchia? distinctive characters of very great import- 

 ance. The remarkable external aspect of these genera has required very 

 much the attention of naturalists in separating them from others ; but 

 no great success has been attained in the fixing of their characters. 



Artedi, for example, not only united them to the lophius and the lump- 

 fish in the order of branchiostegal fishes, but he established the whole 



