The Bulletin 
7 
the stigmatic surface. The stamens have undergone a marked trans¬ 
formation in both their form and usefulness. The filaments no longer 
stand truly upright, as do those in the normal flower, hut are hiorei or 
less recurved or reflexed. According to Dorsey (2), this is due to the 
unsymmetrical development about their longitudinal axes. In length 
and in diameter they have suffered material degeneration. Very often 
their length is considerably less than the greater diameter of the anthers, 
and their transverse diameter has been so reduced that the filaments 
bear very much of a resemblance to fine, silken threads. The pollen 
that is produced by these stamens, instead of being plump and smooth, 
appears shrunken under the microscope [Reimer and Detjen (1)], and, 
when seen in quantity, very much resembles a heap of dented and col¬ 
lapsed tennis balls. It is needless to say that such pollen is unviable 
under the best of conditions, and therefore absolutely worthless. Hed¬ 
rick and Anthony (3) state that many varieties of grapes with the 
reflexed type of stamen. produce some normal viable pollen grains 
among the usual misshapen and abortive kind. Such mixed pollen may 
yet be found with some varieties of rotundifolia grapes; but, to date, 
many varieties and seedlings of this species have been examined and 
no normal pollen has ever been found associated with the reflexed type 
of stamen. The nectar glands that are found in the flowers with the 
reflexed type of stamen are well developed; they secrete nectar and emit 
a very fragrant odor which is attractive to many insects. 
Such, in general, are the characteristics of the three types of flowers 
in the genus Vitis. Intergrading types are often found, but these will 
receive special consideration elsewhere. 
III. Introduction of Hermaphrodite Vines With 
Upright Stamens 
In the Summer of 1910, while the study of self-sterility of Vitis 
rotundifolia was in progress, Professor F. C. Beimer (1) discovered a 
wild vine with upright stamens growing in a plum thicket about one 
mile from the city of Raleigh, V. C. At the time of the discovery the 
vine bore a fairly good crop of fruit and, although late in the season, 
a few flower clusters. It was the flowers on these clusters that com¬ 
manded attention. 
Bagging experiments conducted immediately upon the discovery of 
the vine gave results indicative of self-sterility. The more extended 
work of the following two seasons gave similar results. 
Although the data obtained in the work of these three seasons pointed 
very strongly toward self-sterility, a belief in the self-fertility of the 
vine persisted and evidence secured in later years substantiated the 
correctness of this supposition. 
The strongest evidence that was available in favor of the self-fertility 
of the vine was the fact that when it was discovered it bore a fairly good 
crop of fruit. While none of the flower clusters that were bagged in 
