competing ecological groups. Such an apparent evasion of competition 

 by near relatives is well illustrated by our observations on the top- 

 minnows (Fig. 25) very small fishes of the killifish family, of which 

 we have three species in Illinois. Two of these species are distributed 

 throughout the state, but the third is southern in its distribution, lapping 

 over on the Illinois area of the other two only in the southern part of 

 the state. Now it was to me an extremely interesting fact that we found 

 this southern species much more frequently in company with the two 

 more widely distributed top-minnows than these two were with each 

 other. It was as if those which occupy the same area conjointly had found 

 themselves compelled to an ecological division of it, such as would keep 

 them largely out of each other's way, while those of an essentially un- 

 like geographical distribution had found no such mutual avoidance nec- 

 essary. It seems quite possible that this intra-local separation of com- 

 petitive groups, this effective isolation of forms inhabiting the same ter- 

 ritory, may be one of the early steps sometimes the very first step, 

 perhaps in the differentiation and fixation of species. A difference in 

 respect to choice of breeding grounds especially, by preventing the inter- 

 breeding of two diverging groups, would separate them as effectually 

 as an impassable mountain chain running through the area of their 

 original distribution. 



This principle of an evasion of competition seems to apply to asso- 

 ciations also, as well as to species, and to explain in part the composi- 

 tion of neighboring associations. Similarly endowed species, similarly 

 disposed towards their environment, would profit mutually by a geo- 

 graphical separation, which should give to each a range not entered by 

 the other; and adjacent associations might thus be formed, alike in their 

 ecological make-up but different in their species. It is in some such 

 way that we may perhaps explain a few otherwise unexplained limita- 

 tions of the distribution of our Illinois fishes. Six of our one hundred 

 and fifty Illinois species are so definitely limited to the Wabash drainage 

 as to suggest that there must be some ecological barrier against their 

 spread, since there is certainly no geographical one, namely: brindled 

 stonecat, Schilbcodes miurus (Fig. 26 and Map LIX) ; green-sided darter, 

 Diplesion blennioides (Fig. 27 and Map LXXXIX) ; Notropis ille'ce- 

 brosus (Fig. 28 and Map XXXVII) ; silver-mouthed minnow, Ericymba 

 buccata (Map XLVI) ; long-eared sunfish, Lepomis megalotis (Fig. 29 

 and Map LXXVI) ; and Boleichthys fusiformis (Fig. 30 and Map 

 XCVIII). 



