TRANSMISSION OF INDUCED EYE-DEFECTS yABl 
for expression. Moreover, the serum-treated females themselves 
which had yielded young with abnormal eyes were repeatedly 
bred to the same males after the serum treatments were stopped, 
yet, although many young were born, they never produced any 
more individuals with defective eyes. Inasmuch as the litters 
obtained in later experiments from a stock unrelated to our main 
strain contain individuals with imperfect eyes (fig. 3), we are 
at present engaged in establishing new defective lines. 
CONCLUSIONS 
After going through the literature of serology, the outstanding 
impression in the minds of the authors is the superabundance of 
theories which prevail and the scarcity of unambiguous facts. 
The field is an extremely intricate one and the pitfalls are many, 
not the least of which is the drawing of conclusions from the use 
of too few animals. For the literature certainly shows that even 
in the same species of animal there may be marked individual 
differences in serological behavior. 
In view of the confusion and uncertainty of interpretation 
which prevails in this field, the authors do not feel it incumbent 
upon them to attempt to supply a fully perfected theory which 
shall account for all the details of just what is taking place in the 
interior of the soma or the germ of the fetuses borne by the spe- 
cially treated mothers. The thing that interests them most is 
that a certain effect has been produced; and, what is of greater 
importance, that once established, the condition may be trans- 
mitted from generation to generation. In view of the fact that 
the defects have been carried into the sixth generation by breed- 
ing, without any subsequent treatments with the sensitized sera, 
and, above all, since the modifications have been extracted through 
the male line, thus eliminating all possibility of the condition in 
later generations having been due merely to placental transmis- 
sion from the blood of affected mothers, we feel that the evidence 
establishes a clear-cut case of inheritance of a specific modifica- 
tion produced by extrinsic factors. 
It is not entirely clear as to whether the result should be reck- 
oned primarily as an example of the inheritance of a somatic 
