190 American Fisheries Society 



southern limit of marine forms in the glacial period would 

 have pushed down to about 27° north latitude, or almost 

 exactly to the mouth of the Yaqui River in Mexico — a striking 

 coincidence. 



It is a well-known fact that, as a rule, northern fishes are 

 characterized by smaller scales and more numerous vertebrae 

 than those of the south. At least, so far as the scales are 

 concerned this is true of the Salmonidae, comprising the sal- 

 mons and trouts, with perhaps some exceptions which, after 

 all is known, may prove the rule. In our hypothetical trout 

 marine zone, it is only in the north that small scaled forms are 

 found in the sea or near the coast. In inland waters small 

 scaled forms are found in high altitudes far to the southward. 

 But, coastwise, in intermediate localities, trout with inter- 

 mediate sizes of scales occur, and in the sea and accessible 

 fresh water, large scaled forms occur far north. These facts 

 suggest that the routes which have been scheduled by Jordan 

 were not strictly correct. 



It does not seem possible that a large scaled fish entering the 

 Colorado, could ever become the small scaled species of the 

 Kern, or a small scaled form from any source could ever 

 directly change into a large scaled fish of another locality; 

 and Jordan's statement that "It is not the preservation of the 

 most useful features but of those which actually existed in the 

 ancestral individuals, which distinguished such species," is 

 an argument against the derivation of Durango trout from the 

 Missouri or Snake Rivers. Our zone theory hangs upon one 

 essential point, namely, that the conditions at present necessary 

 to the existence of the trouts indicate that the trouts were 

 evolved in and synchronously with the changes of environ- 

 mental conditions culminating in those of the present time. 

 In this zone of marine trout, having an extent of some 30 

 degrees, we may suppose were intermediate zones, occupied by 

 forms intermediate between those with a tendency to small 

 scales, associated with the northern limit, and those with a ten- 



