1. 8ALM0. 6 



know of one example only (of S. cambricus) where the usual number 

 of the rays of the anal fin is increased by two, and we are still un- 

 certain in what light this variation ought to be considered. Although 

 some species appear to be distinguished by a conspicuously low dorsal 

 and anal fin, yet the proportion of the height of these fins to their 

 length is a rather uncertain character. In most of the species the fin- 

 rays arfe longer during the earlier stages of growth or development, 

 but in a few we have observed that mature individuals have had the 

 fins comparatively longer than immature ones. The caudal fin espe- 

 cially undergoes considerable changes with age, and dependently upon 

 the sexual development. Young specimens of all species have this fin 

 more or less deeply excised, so that ihe young of a species which has 

 the caudal emarginate throughout life is distinguished by a deeper 

 incision of the fin from the yoiing of a species which has it truncate 

 in the adult state. We shall subsequently see that the individuals 

 of a species do not all attain to maturity at the same age and at the 

 same size. Hr. Widegren first pointed out that mature individuals 

 generally have the caudal less deeply excised than immature ones of 

 the same age and size. It is therefore evident that the variations 

 in the form of the caudal are considerable and numerous, and that 

 it is a very fallacious chaf9,cter if due regard be not paid to the age 

 and sexual development of the fish. Further, species inhabiting 

 rapid streams as weU as still waters show considerable variations 

 in the form and length of all the fins ; those individuals which live 

 in rapid streams, being in almost constant motion and wearing off 

 the delicate extremities of the fins, have the fin-rays comparatively 

 shorter and stouter, and the fins of a more rounded form, parti- 

 cularly at the corners, than individuals inhabiting ponds or lakes 

 (see p. 149) ; moreover one and the same individual may pass a part 

 of its life in a lake, and enter a river at certain periods, thus changing 

 the form of its fins almost periodically. 



Finally, to complete our enumeration of these variable characters, 

 we must mention that in old males, during and after the spawning- 

 season, the skin on the back becomes thickened and spongy, so that 

 the scales are quite invisible, being imbedded in the skin. 



After this cursory review of variable characters, we pass on to 

 those which we have found to be constant in numbers of individuals, 

 and in which it is difficult to perceive signs of modification due to 

 external circumstances. Such characters, according to the views of 

 the zoologists of the present age, are sufficient for the definition of 

 species : at all events, in every description, they ought to be noticed, 

 and the confused and unsatisfactory state of our knowledge of Sal- 

 monoids is chiefly caused by authors having paid attention to the 

 more conspicuous, but unreliable characters, and but rarely noted 

 one of those which are eniunerated here : — 



1. The form of the prceoperculum of the adult fish. The prseoper- 

 culum is composed of a vertical (posterior) and horizontal (lower) 

 part (limb), both meeting at a more or less rounded angle. The 

 development of the lower limb is a very constant character : in some 

 species (as in the Salmon) it is long, in others {S. ferox, S. brachif 



