INHERITANCE OF MENTAL DEFECTS AND DISEASE 37 
into clearly defined classes. This fact does not necessarily indi- 
cate, as Pearson and Heron imply, that the various kinds of men- 
tal defect are not transmitted according to Mendel’s law. It is 
not uncommon for segregation to occur in the usual Mendelian 
manner, although the character segregated may fluctuate so as to 
form a perfectly continuous series. Where the germinal factors 
manifest themselves somatically in characters that undergo a 
large amount of fluctuating variability, it naturally makes the 
demonstration of Mendelian segregation more difficult. Where, 
as in human beings, it is not feasible to employ experimental 
methods of analysis the difficulty of establishing Mendelian 
inheritance beyond cavil is greatly enhanced. One has to be 
guided by probabilities. The best that can be done is to select 
tentatively that hypothesis which gives the most plausible inter- 
pretation of the phenomena to be explained and is best in accord 
with what is known of the principles of inheritance followed in 
other fields. The very general occurrence of Mendelian inheri- 
tance among plants and animals of both primitive and highly 
organized types, and the remarkable success attained in explain- 
ing apparently non-conformable phenomena in terms of Mendel’s 
law, creates a very justifiable presumption in favor of the conclu- 
sion that mental defects are transmitted according to the same 
laws that prevail so widely in the plant and animal world. That 
inheritance in man obeys the laws followed by organisms in gen- 
eral is also indicated by the undoubted appearance of types of 
Mendelian inheritance among human characteristics. 
But while the general occurrence of Mendelian inheritance in 
the organic world creates a presumption in favor of the conclusion 
that mental traits in man are transmitted according to the same 
tule, it must be conceded that there are certain characters whose 
mode of transmission seems to present a clear exception to this 
type of inheritance. It is true that such cases are comparatively 
rare. But there is a much larger number of cases which may 
follow Mendel’s law, but in which it has never been proven that 
they actually do follow it. The successful extension of Mendelian 
analysis may justify us in shifting the burden of proof from the 
