38 THE AMERICAN NATURALIST. 
geneticaly related, we may deduce some interesting conclu- 
sions from the preceding data: 
I. The variation here recorded is saltatory. 
2. It is bilateral, without reaching perfect bilateral correlation. 
This makes it probable that primarily the variation was 
introduced or induced by changes on one side. Judging 
from its greater variability, the left side was primarily 
concerned. 
3. It is improbable that the variation originated independently 
in each of the specimens. 
4. The variation probably arose in one of the ancestors of the 
specimens. 
5. Admitting 4, the saltatory variation arising in an ancestor 
was prepotent to a very high degree. 
There are entirely too many probabilities in these conclu- 
sions, in spite of the fact that the probabilities all verge nearer 
certainty than uncertainty. The matter of the prepotency of 
the variation can readily be determined by breeding some 
specimens from the cave with normal specimens, which will 
be attempted. 
