GREAT LAKES COREGONIDS 



309 



ranges of a morphologically distinct race likewise may be interrupted. Where these 

 areas of similarity are not widely separated geographically or, geologically speaking, 

 in time, the distinctive races may be considered subspecies, as usually their relation- 

 ships with other members of the species group is clear even though they are so sepa- 

 rated from them that there is no possibility of finding intermediate forms, which 

 ideally is the criterion of a subspecies. But where like areas are widely separated in 

 space and time, even though the forms in each may be nearly identical in structure 

 and habit, taxonomists generally have preferred to consider them species rather than 

 subspecies. 



In aquatic habitats such zoogeographical islands also may occur; in fact, lakes 

 are particularly good examples of isolated habitats. Though the types of lakes vary 

 within certain limits, it is also true that aquatic habitats are simpler, in general, than 

 terrestrial ones; they are influenced by fewer variables. Land habitats vary more 

 because of differences in humidity, temperature, light, soil, elevation, etc. In 

 aquatic environments humidity is not a variable, and temperature is limited in 

 temperate regions between 0° and 25° or 30° C. At depths temperature differences 

 are even eliminated, as are those of illumination. There remain differences in the 

 chemistry of the water, depending on the soil of the basin, in depths, and exposure to 

 wind. Where variables like temperature are not involved, the ordinary effect of 

 latitude, which is so important in the distribution of land animals, is minimized; and, 

 of course, where factors are few, the chances of finding them frequently in like com- 

 binations are greatest. It is thus possible to find in a lake in Indiana, as far as certain 

 species are concerned, the same sort of habitat as in a lake in Canada 500 miles 

 farther north; and it is likewise possible that two lakes in the same township may be 

 so totally different in their physical conditions that their fish populations are very 

 dissimilar. Now, in a given species the same mutation has a tendency to recur 

 with a somewhat definite frequency. If it marks a higher degree of habitat adapta- 

 tion than its parent in one place, and therefore tends to supersede its parent, it is 

 only natural to expect the same outcome in another location where the environment 

 is the same. It should not be surprising, then, to find varieties of a species of fish 

 distributed according to the type of habitat rather than according to geographic 

 zones. 



The forms of the Great Lakes whitefishes thus appear to be distributed. The 

 deep-bodied type of herring (Leucichthys artedi) is distributed here and there in lakes 

 between New York and Manitoba, while in other lakes in this area the other extreme 

 in development possible to this species may occur. Where two morphologically 

 distinct forms of a species occur in the same lake, both extremes may be found in the 

 area of intergradation, whether as a result of migration or of Mendelian segregation 

 of interbred characters. 



Botanists are confronted regularly with the irregular distribution of morpho- 

 logically distinct individuals in the case of certain species of plants and find it con- 

 venient to introduce the terms "variety" and "form" in their nomenclature as units 

 ranking less than a subspecies. In the case of the whitefishes it might be desirable, 

 for certain reasons, to follow botanical practice; but, on the other hand, it is also 

 desirable to keep the question of zoological nomenclature as simple as possible, and 

 it is already sufficiently complicated by the use of trinomials. There seems to be on 



