PTR ee oe ir ee A 
1876.) Botany. 491 
In Michaux’s Flora is the note “ Flores omnes in nonnullis individuis 
abortivi,” and botanists are generally aware that fruit is seldom met with. 
The flowers have been said to be unisexual (dicecious) ; but all appear 
to have well formed ovary and ovules, although some individuals were 
known to want the stamens. Professor Goodale, knowing a station in 
Maine in which Æpigæa year after year sets fruit, kindly procured from 
thence a large number of fresh specimens; and these I have now ex- 
amined in regard to stamens and pistil. They show the following hetero- 
morphous condition of things. 
(1.) About ten per cent. of the specimens have a style considerably 
longer than the stamens, raising the stigmas a little out of the throat of 
e corolla, in which the anthers are included: the stigmas are cylin- 
draceous, radiate like the spokes of a wheel, half a line in length, there- 
fore strongly projecting, moist and glutinous, and evidently in good con- 
dition for fertilization. The anthers in these flowers are slender, com- 
monly withering without dehiscence, and containing few, yet perhaps 
well-formed, pollen-grains. The fruiting specimens gathered at the same 
station in former years all evidently belong to this form, as the persistent 
style and long stigmas show. One or two specimens of this form mani- 
fest a disposition to convert their anthers into petals; but this is occa- 
sionally seen in other forms. ; 
_ (2.) A smaller number of specimens show the stigmas of the preced- 
mg on a shorter style, sometimes so short as to place the radiating 
stigmas as low as the middle of the tube of the corolla, sometimes bring- 
ing it nearly up to the throat. In one instance a short-styled flower 
Was detected in a cluster of flowers otherwise of the character of No. 1. 
These short-styled blossoms, instead of having more conspicuous Or 
higher anthers than in the long-styled, bear them either at the same 
proportional height and in the same condition, or bear mere rudiments 
u anthers, or not rarely none at all, and even the filaments are smaller, 
abortive, or occasionally altogether wanting. This sometimes happens 
m No. 1 also. 
: (3.) The larger number of flowers, perhaps three-fourths of the spec- 
mens under examination, have the long style of No. 1, an ovary equally 
well-formed and ovuliferous, but either rather smaller or not going on to 
stow; but the stigmas are short, only slightly projecting beyond the 
lobes of the cup to which they adhere, in all stages erect, and compara- 
tively smooth and dry., Their tips, however, appear somewhat papillose 
T a strong lens, and grains of pollen placed thereon incline slightly 
adhere, yet not so much as upon the surface of the style far below, 
og Sets well covered with pollen from the contiguous anthers. The 
rence between these stigmas and those of the foregoing forms is 
_‘Sttiking and constant, no gradations between them having been detected. 
Ry anthers abound with pollen, and are dehiscent at or a little before 
Opening of the corolla. 
