CLASSIFICATION 



47 



pectoral, pelvic and caudal fins, the number of scales in a longitudinal and in a trans- 

 verse series, the number of pores in the lateral line, and the number of vertebrae. Of 

 these, the number of rays in the dorsal and anal fins is perhaps the most important, and, 

 since this is correlated with the number of vertebrae and is subject to the same fluctua- 

 tions, the two characters may be considered together. Jordan (1893) has shown that, 

 as a general rule. Flatfishes from tropical seas have fewer vertebrae than those from 

 temperate or arctic seas, at least as far as the " Flounders " are concerned. Further, 



. 28. — Intestine and " pyloric " appendages of a, Psettodcs erumei ; b, Limanda limanda ; 

 c, PseudopUuronectes yokohamtB ; D, Pleuronectes ptatessa ; E, Microstomus kitt ; f, Glypto- 

 cephalus cynoglossus ; G, Platichthys flesus ; H, Zebrias zebra. Diagrammatic, d and h 

 from ocular side, others from blind side, [a and h after Wii.] 



as has been pointed out by CoUett (1S80, p. 148), within the limits of certain species 

 having a wide distribution in arctic and temperate regions the average number of 

 vertebrie and fin-rays is markedly higher in samples from more northerly localities 

 than in those from regions further south. This has been tested by me in Paralichthys 

 olivaceus and Hippoglossoides platessoides limandoides, and the same phenomenon has 

 been recorded in other fishes.' The well-known work of Johannes Schmidt on Zoarces 

 viviparus^ has shown that the differences in the number of vertebrae in samples from 

 different localities may be correlated with environmental factors, particularly with 



See Hubbs (1921, i9.;5). 



For references see Hubbs {1926). 



