INHERITANCE OF MENTAL DEFECTS AND DISEASE 51 
parents is a neuropathic person (RR). On looking through the 
charts I find that only three of the 19 cases fulfill this condition. 
If one of the parents has a brother, sister or other near relative 
who is neuropathic, the assumption that this parent is heterozy- 
gous is only probable. In going over the charts for cases of this 
kind I find a record in the alleged DRX RR matings of only five 
instances. In all the other cases the conclusion is apparently 
based on no evidence at all beyond the fact that it is necessary to 
assume it in order to make the facts come out in accordance with 
the hypothesis. 
The third class of cases discussed, the matings of a homozygous 
normal with a double recessive, DD X RR is represented according 
to the authors, by “five matings with a total of 18 offspring. 
Theoretically all the offspring of such matings should be normal, 
but capable of transmitting the neuropathic make-up to their 
progeny. The charts show: 8 normal with neuropathic offspring, 
7 normal with normal offspring, 2 normal without offspring, and 
1 died in childhood.” The assumption that one parent is a 
homozygous dominant is naturally somewhat unsafe. From the 
nature of the case we can never know that this is correct, but 
from what has just been quoted it may be inferred that this 
assumption is made because all the children are normal, and some 
of the grandchildren neuropathic. Of course some of these cases 
cited may have been DRX RR matings which happened to have 
only normal (DR) children. What the authors have done is to 
divide up the cases in which normal and neuropathic mate into 
DD XRR and DRX RR in such a way as to best make the results 
fall into line with the theoretical expectations. That other 
interpretations are not improbable is evident from what has 
previously been said. 
The alleged DRXDR matings turn out more in accordance 
with expectations since seven matings with 54 offspring yielded 
12 neuropathic, and 34 normal individuals, and 8 who died in 
childhood. 
A subsequent paper by Rosanoff and Orr deals in much the 
same way with a larger amount of data, represented by 73 pedi- 
