434 
markably and persistently turbid, never clearing themselves spon- 
taneously. This is owing in part to the extremely fine division of the 
soil, and in part to its generally acid character and the consequent 
lack of “granulation,” or cohesion of its ultimate particles in gran- 
ules, such as occurs in the alkaline souls of the other geological areas 
of the state. The surface waters of the district are soft and slightly 
alkaline, but contain much silica, and much solid matter in suspen- 
sion which it is extremely difficult to remove completely by any 
ordinary filtering or precipitation process. The inference is plain 
that it is to this condition of the waters—due to the geological his- 
tory of the soil of this region—that the unequal distribution of these 
fishes is largely to be attributed. 
6. In consequence of another clearly recognizable inequality of 
distribution, partly coincident with the two preceding and partly in- 
dependent of them, two additional groups may be distinguished; one 
of 8 species, distributed in this state mainly through the Ohio and 
Wabash drainage, and the other of 27 species, distributed through 
the Mississippi and its more northerly tributaries. The general dis- 
tribution throughout the country at large of each of these two groups 
of species is quite varied, and offers no hint of a reason for these dif- 
ferences in Illinois. Two hypothetical explanations are suggested— 
the first presupposing different centers of population outside the 
state, from and towards which these species move, into and out of 
Illinois streams, with the spring rise, summer recession, and winter 
cooling of the waters, one of these centers to the west and north, and 
one to the east and south; and the second presupposing an organiza- 
tion of the fish population into more or less distinct communities of 
mutually well-adjusted species, each community so adapted to its 
environment that members of adjacent communities can not success- | 
fully intrude upon its territory. 
7. An analysis of our statistical data of ecological distribution 
gives uS many instances of a marked difference in preference of 
situation between nearly related species inhabiting the same area, 
the effect of which is to break the force of a competition between 
these species such as would prevail if they were similarly distrib- 
uted ecologically as well as geographically. Closely related species 
are, as a consequence, often found much less frequently associated 
in their common territory than either is with widely unlike species 
of the same geographical range. Exceptions to this rule are found 
