THE MANNER OF OCCURRENCE OF MUTATION 461 
first generation that receives them. The valid evidence in 
regard to the origin of such factors is rather meager, for in 
many instances such mutants have been found in mass cultures, 
in which it cannot be determined how many normal individuals 
belong to the family that produced the mutant; often, too, the 
original occurrence of the mutant cannot be checked up, on 
account of its having been first observed in a stock that had not 
been carefully examined during the preceding generation. 
Let us suppose, however, that a census has actually been 
made which includes only those cases that are known to be 
-valid.2 Let us then assume that the data really show that in 
the great majority of cases the mutant factor, at its first appear- 
ance, appeared in only one member of a family. It is easy to 
show that this is just the result that would be expected even if 
mutation is equally likely to happen at any stage during devel- 
opment, and that it by no means indicates that the later stages 
of the germ tract.(gametes, cytes, and late gonia) are for some 
reason more inclined to mutate than the early stages (gonia, 
primordial germ cells). For, in the first place, since there are 
many more cells produced during the later stages than during 
the earlier ones, there are many more chances that a mutation 
should occur in the later stages, even if mutation is no more 
likely to oceur in one cell than in another. Secondly, unless we 
make the gratuitous assumption that mutations only occur dur- 
ing certain fixed periods in the life of the cell, which are of the 
same duration in all cells of an organism no matter how long the 
cell lives, then the fact that the cells usually exist much longer 
in the later stages (growth period, ete.) than in the early stages 
would again provide more opportunity for mutations to occur in 
these late stages, even though the cells at that time were not, 
per se, more mutable. Thirdly, unless approximately 100 per 
cent of the gonia and gametes produced from the mutated cell 
_ develop into viable individuals showing the mutant character, 
many of the cases in which more than one germ cell came to 
receive the mutant gene would be recorded as cases of single 
® See the paper of Bridges (’19), which has appeared and come to my attention 
since the present paper was written. 
