12 



You are all familiar, no doubt, with the idea of animal associa- 

 tions groups of species habitually associated in the same environments ; 

 and these are as recognizable among fishes as among the animals of the 

 land ; and here again, substantially as in the other cases, we find animals 

 which, taken by themselves, may be seen to form associate groups 

 with a large extension, covering many distinguishable kinds of situations, 

 and others which are rather narrowly limited to a single sort of habitat, 

 characterized by quite special conditions. There is a group of creek 

 fishes, for example, made up of species which may be found almost any- 

 where in a creek and in almost any kind of a creek, and another group, 

 like certain of the darters and the stone-roller, which are thoroughly 

 at home only in rocky streams with a relatively rapid flow of at least 

 fairly clear water. If we analyze the aquatic environment into all the 

 situations clearly distinguishable, we shall find, perhaps, a group of cer- 

 tain species distinctive for each situation, but for other species our 

 analysis will be seen to have gone too much into detail ; it will distinguish 

 differences of condition to which they are indifferent. Even the distinc- 

 tion between small river and creek, and between river and lake, is too 

 narrow for some fishes, which are found with almost equal frequency in 

 both. The grass pickerel (Fig. 21), which we have taken in one hundred 

 and eleven Illinois collections, is almost equally abundant in creeks and 

 in ponds; and the river-chub (Fig. 22) has been found about equally 

 common in creeks and in small rivers, and virtually absent from the larger 

 rivers and from lakes. Nevertheless, the distinction of animal associa- 

 tions is very helpful to our grasp and understanding of the system of 

 living nature, and will become much more so as our knowledge becomes 

 both more comprehensive and more precise. 



An organization of the animal population of a region into associa- 

 tions may be approached either from the side of the environment or from 

 that of the animal inhabitants; either by an analysis of the environment 

 into habitats and situations, and a critical survey of the inhabitants of 

 each, or by an analysis of the animal population into groups of most fre- 

 quent associates, and a study of the local and ecological distribution of 

 each such group. By the first method, spatial units so-called units of 

 environment are first distinguished and delimited, and the animals con- 

 tained in each such unit are then identified, listed, and enumerated. By 

 the second method, the associate groups are first distinguished, defined, 

 and analyzed, and the territory under examination is mapped in a way to 

 mark the area of distribution of each such group. The first method is 



