GREAT LAKES COREGONIDS 329 



in most of the species this variation is possibly no greater than would be found in 

 other unrelated species if the same number of individuals of these were studied in 

 the same way. This variation is exhibited by individuals presumably intimately 

 related — that is, by individuals of the same school in one locality. In many species 

 it has been found that geographically separated races have developed peculiar 

 characters even within a lake, and it is probable, furthermore, that different schools 

 in one locality would show peculiarities. To cite the most conspicuous examples of 

 differentiation where a species has been segregated definitely, geographically, we have 

 the dymondi subspecies of reighardi in Nipigon and Superior, the cyanopterus, regalis, 

 and prognathus subspecies of nigripinnis in Superior, Nipigon, and Ontario, respec- 

 tively, and the orientalis subspecies of Tciyi in Ontario. Within a lake we have the 

 conspicuously differentiated forms of artedi — manitoulinus in Huron and albus in Erie 

 and Superior and the unnamed deep-water variant of Jioyi in Huron. 



These various forms probably have arisen through isolation and to some degree 

 may be the result of different environmental conditions operating in each generation. 

 It is not clear, however, what environmental factors might operate to develop the 

 various forms that occur in separated lakes. There seems to be no definite direction 

 of variation expressed by the forms in any lake. Thus, while nigripinnis in Superior 

 has fewer gill rakers than any of the known races in other lakes, zenithicus in that 

 lake seems to have a few more than its relatives elsewhere. Similarly, though Jioyi 

 and ~kiyi in Lake Ontario have more gill rakers, it is not true of other coregonids in 

 the lake; and in Nipigon hoyi seems to have somewhat more lateral line scales than 

 the hoyi elsewhere, while all the other forms of Leucichthys seem to have fewer than 

 their relatives in other lakes. 



In the case of the varieties of artedi there seems to be some clue as to the causes 

 operating to produce certain characteristics of development, but until a study has 

 been made of the forms of artedi known to occur in the inland lakes tributary to the 

 Great Lakes any statement regarding the manner in which environment influences 

 the direction of variation is purely hypothetical. We do know, however, that the 

 forms of albus and manitoulinus both exhibit the same sort of variations — both have 

 deep, abbreviated bodies with relatively longer paired fins and few lateral-line scales. 

 They are not alike, however, in certain other peculiarities, such as the relatively 

 large head and eye and the dark color of the latter. These varieties always occur 

 in the warmest waters of the territory available for occupation. Lake Erie, which 

 is the shallowest and most southerly of the Great Lakes, presumably is the warmest; 

 and it has been shown in Table 13 that Black Bay, in Lake Superior, is much warmer 

 than the open lake. The same probably is true of Cutler Bay, where manitoulinus- 

 occurs. 



It would be expected that, if temperature had a part in this variation, an oppo- 

 site type of development would result where temperature conditions were reversed, 

 and this appears to be true. The habitat of the Lake Superior herring is almost 

 certainly the coldest in the Great Lakes occupied by the species throughout the 

 year. The artedi of the lake are the slenderest, most elongate forms, and they have 

 the most lateral-line scales, though the paired fins are not conspicuously different 

 from those of Michigan and Huron specimens. It is noteworthy, also, that the 

 clupeajormis of Lake Erie and Black Bay of Lake Superior also are known to be 



