201 
has apparently been published. The description of the species, 
as given in the revised edition of Gray's Manual, is as follows: 
“ Root-leaves very large, oblong, crenate, rarely pinnatifid, those 
of the stem lanceolate; fruiting pedicels ascending ; pods globular 
(seldom formed); style very short. Roots large and long.” 
It has been pointed out already that the pinnatifid leaves are 
not “rarely” but normally present in the spring and fall. In 
addition, the quoted description should be amended by mention 
of the fact that the lower stem leaves are invariably pinnately dis- 
sected, and gradually pass through pinnatifid and notched forms 
until near the top they are simply lanceolate with crenate and 
toothed margins. Of hundreds of floweri ng stems examined, all had 
this peculiar arrangement of stem leaves, and extended observation 
has tended to confirm the opinion that the plant was originally 
semi-aquatic, which, at the time of flowering, was accustomed to 
stand knee-deep, as it were, in water. As a further proof of the — 
value of structural peculiarities as indices of the former habits of 
plants, it may be well to note that very vigorous flowering plants of 
the species under consideration have been observed growing in 
ditches and on the margins of ponds, which, though full in the 
spring and fall, are dry or nearly so in the summer. These 
plants, at the time the notes were made, stood in water which was 
of such depth that the dissected leaves were nearly all covered. 
By far the most interesting characters of this plant, how- 
ver, are its provisions for propagation and the preservation 
of the species. As noted in the description quoted from the 
Manual, the pods are seldom formed. When formed they ane 
without seeds and soon disappear, apparently falling off, leav- 
ing the pedicels fast to the axis of inflorescence. The cause 
of this failure to mature seed does not appear. The flowers 
are freely visited by insects, pollen is present in the anthers and 
seems abundant, and the ovary and ovules are apparently entirely | 
normal. A possible cause of the lack of seeds suggested itself 
when the stigma was studied. This organ is covered with numer- 
©us Comparatively long hairs, which may prevent the pollen from 
reaching the stigmatic surface, or it may be that. the pollen, in 
order to fertilize the ovules, has to fall directly on the tips of these 
hairs. In either case it would seem highly probable that now and 
