SALMONID.E. 637 



the character vague, or even unsafe, so that some zoologists 

 have rejected it entirely as unreliable. However, when a 



Fig. 290. — Vomerine teeth of Fig. 291. — Vomerine teetli of 



Salmo fario, lower view. a Cliarr, side view. 



greater number of individuals really belonging to the same 

 species are examined, a pretty safe conclusion may be arrived 

 at as regards the arrangement of the teeth, 



5. The form of the caudal fin m specimens of a given size, 

 age, and sexual development. 



6. A great development of the peetoral fins, when constant 

 in individuals from the same locality. 



7. The size of the scales, as indicated by the number of 

 transverse rows above the lateral line : one of the most con- 

 stant characters. 



8. The number of vertcbrw. — Considering the great number 

 of vertebrae in Salmonoids the constancy of this character is 

 truly surprising. An excess or a diminution of the normal 

 number by two, is of rare occurrence, and generally to be ex- 

 plained by the fact that one vertebra has been abnormally 

 divided into two, two such vertebrae being considerably 

 smaller than the others ; or, on the other hand, that two have 

 merged into one centrum, which is then unusually large, and 

 provided with two neural spines. We have seen one case 

 only, in which three vertebra? were united. The number of 

 vertebrse can be easily ascertained in specimens destined for 

 preservation in spirits, by an incision made along one side of 

 the fish, a little above the lateral line, 



9. The number of pyloric appendages. — There can be no 

 doubt that this character may materially assist in fixing a 

 species. We shall see that in some species it varies from 30 

 to 50 ; but in others, as in the Salmon and Charr, it has been 



