SALMONlDyE. 55 



the common trout, including the Gillaroo, which variety he recorded having met 

 with in most fresh water races ; (5) the great lake trout. 



Dr. Giinther, in a catalogue of fishes in the British Museum, described in 1866 

 the following anadromous forms: (1) Salmo salar, Vertebra?, 59 (60), Ccecal 

 appendages, 51 to 77 ; (2) 8. argenteus, Ccec. pyl. 61-67 ; (3) S. trutta, Vert. 

 59-60, Ccec. pyl. 43-61; (4) 8. orcadensis, Vert. 56-57, Ccec. pyl. 50; (5) 

 8. brachypoma, Vert. 59, Ccec. pyl. 45-47 ; (6) 8. cambricus, Vert. 59, Ccec. pyl. 

 33-52. And of the non-migratory fresh water forms, admitted (7) 8. levenensis, 

 Vert. 57-59, Ccec. pyl. 49-90; (8) 8.fario, variety Gaimardi, Vert. 59-60, Ccec. pyl. 

 33-46, variety ausonii, Vert. 57-58, Ccec. pyl. 38-47; (9) 8. ferox, Vert. 56-57, 

 Ccec. pyl. 43-49; (10) 8. stomachicus, Vert. 59-60, Ccec. pyl. 44; (11) S. galli- 

 vensis, Vert. 59, Ccec. pyl. 44 ; (12) 8. nigrvpinnis, Vert. 57-59, Ccec. pyl. 36-42. 



From the foregoing we find that Donovan, Turton, Fleming, Jenyns, Yarrell, 

 Parnell, Jardine, Thompson, and White admitted the existence of five or six 

 British species of true salmons in our waters, which Giinther increased to twelve, as 

 well as two varieties of the common trout. 



A thorough investigation into these various forms possesses more than a 

 passing interest, for if we have many species of true salmon and trout in our 

 waters and they interbreed, it first becomes a consideration as to what are the 

 probabilities of sterility occurring in the offspring ? On the other hand, should 

 trout from two apparently distinct species be crossed and no unusual phenomenon 

 occur, except improvement in the breed, while signs of sterility do not follow, the 

 supposition must be raised that we are dealing with local races and not with 

 different species, and that the young are mongrels and not hybrids. 



Here 1 must shortly allude to how I propose discriminating between local races 

 or varieties and species, for what one naturalist considers a variety another looks 

 upon as a species. I shall consider species among the true salmons to be an 

 assemblage of individuals which agree together in their structure and in the 

 development of the sexes, but differ in some structural character from all other 

 fishes. Also that functionally they are capable of fertile union together, but not to 

 an equal extent or not at all with the members of any other group of fishes. Even 

 within the limits of a single species we find no two exactly similar, but a tendency 

 to divergence from the original type appears to exist, which power of divergence 

 is most likely to preserve and accumulate useful variations. For it has been 

 abundantly proved that artificially, by judicious selection and breeding from 

 individuals which are possessed of some desired variation, such may become 

 permanent in future generations ; while natural selection (perhaps assisted by 

 some unknown factor) would similarly tend to favour the continuation in a wild 

 state of such forms as possess variation favourable to the life of the animal, 

 and thus produce and continue certain local varieties or races. If, however, the 

 variations from the type, present in certain examples, are not of a persistent 

 character, nor exceed the differences between the limits laid down for a species, 

 these cannot be considered as indicating a distinct species ; for to render such 

 valid, we must have a permanence of variation from the original type (see vol. i, 

 p. 245). Thus, among the sticklebacks we find in the ten-spined form some, 

 due to local causes, possessing ventral fins, othei-s destitute of them ; but their 

 difference not being permanent merely resolves itself into a local race or variety. 

 The number of vertebras and the ccecal appendages have been asserted to be 

 constant characters which may materially assist in fixing a species among the 

 Salmonidce, and I propose investigating in detail some of these various structural 

 and functional differences that have been brought forward for the purpose of 

 establishing species among the true salmon, Salmones. Dr. Giinther has pointed 

 out nine constant characters of variation in his elaborate treatment of these fishes, 

 and which I shall commence alluding to as my conclusions differ from those he 

 has so elaborately set forth. 



1. The number of vertebrae. The constancy of this character we are told "is 

 truly surprising," as an excess over the normal number by two, or a decrease to the 

 same extent, is of rare occurrence. But the inquirer finds that in the recorded 

 number of these bones in the migratoiy species, the variation has been restricted 



