230 
HARLEY HARRIS BARTLETT 
from individuals of two different species, growing at the same locality, 
as is so frequently the case with closely related species of the subgenus 
Onagra. This explanation of the source of the two species which 
De Vries differentiated at Amsterdam is amply verified by the fact 
that wild rosettes from the same locality as the seeds proved to belong 
to the same two species. If it were not for a suspicion that two of 
MacDougal's types were in reality non-hereditary seasonal phases of 
the same species, one might almost as readily believe that there were 
three species in the mixed seed collection as that there were two, and 
thus explain the supposed mutation. MacDougal said of the third 
type that all the individuals of it completed their seasonal develop- 
ment much earlier than the other two. It seems not impossible 
that the set of plants which developed early in the season may have 
appeared so different from later plants of the same form as to be mis- 
taken for a mutation. Several times in the writer's experience with 
the Oenotheras, a culture has consisted of two types of plants, those 
which went through the life-cycle of a biennial, forming a dense rosette 
which persisted for weeks, even though the plants, by virtue of being 
started early in the spring under glass, completed their development 
in one season, and those which went through a strictly annual life- 
cycle, forming only a transitory rosette of a few leaves before they 
become cauline. Since the annual type is lower and less robust, 
due probably to its not having elaborated a reserve food supply for 
use during the period of rapid growth, one might easily be misled into 
interpreting it as a mutation. 
After MacDougal gave up his work on Oenothera, his cultures were 
continued by Dr. Geo. Harrison Shull. Shull, however, cannot supply 
seeds of the questionable third form of ''Oe. cruciata'' and writes that 
he has never cultivated it. Of three lots of seed which were originally 
turned over to him, two yielded the same type. 
Even if we assume that MacDougal's third form was really distinct, 
there is no reason to believe that he correctly identified it with Doctor 
Robinson's New Hampshire type. De Vries characterized the latter 
as having a strikingly longer hypanthium than either of the Lake 
George types, whereas MacDougal figured his dubious third type as 
having a shorter hypanthium than one of the Lake George forms. 
Moreover, it appears that MacDougal never cultivated the New 
Hampshire type, but made his identification merely from de Vries's 
very brief account of it. 
