228 
TABLE XV. 
Average No. 
of Dorsal Spines 
and Rays. 
BVICCLINGHS Wit: SOLaxTaN PANG, . soy oo. «io.c-s aos omhg ih clea etormeutiaieay= SLES 28.50 
SPecimens Wall Qvanalirays, 355... csc 5 toe tisnerstetolle ier fee eke ee 29.20 
SHECLINERS With: lr AHAL.TAYS \. fo...) 1t0k ween} otek eens tee 29.53 
Dpeermens. With. La aMal TAYS, 2. 0's. .acc. ul erce sto sm atepialnec sie sie, os 30.24 
Specimens with ld aoal rays. eu eeies wow ts iawgee sche Sam ae 30.85 
From the latter half of Table XV the correlation between the spinous and 
soft dorsals may be obtained. Here it is seen that when the one dorsal increases, 
the other decreases. This condition shows that within certain limits these two 
kind of structures are convertible into each other and may replace each other. 
5 
SUMMARY. 
1. In Etheostoma caprodes the males are more variable than the females in 
the ratio of .507: .468. In Etheostoma nigrum the females are more variable than 
the males in the ratio of .402: .454. 
2. The specimens of both species in Turkey Lake differ from those in Tippe- 
canoe Lake in every structure examined. ’ 
3. The variation in the two species is determinate for the lake, that is, both 
species are modified in the same way by the same lake with but one exception. 
4. This difference is not the result of selective influence, but apparently the 
direct effect of the environment. 
5. The successive broods vary with the varying conditions of the year in 
which they are born. 
6. The variations in the fins are correlated as follows: 
a. When the dorsal spines increase in number the dorsal rays decrease in 
niumber. 
b. When the anal rays increase the dorsal spines, the dorsal rays and the 
sum of the elements in the two dorsals increase. 
