48 DorsEY: VARIATION IN FLORAL STRUCTURES OF ViTIs 
strength but being generally too weak to produce fruit.” Munson 
(’99) mentions that it is clear that the staminate vine can not 
bear since it has no pistil, ‘‘unless the vine changes its action from 
producing purely staminate flowers to bearing pistillate flowers, 
which in two or three instances only, in all my observation, I 
have known to occur.” 
In the instance mentioned by White, the stigma probably had 
a development similar to that shown in FIG. 10, PLATE 2, which 
isa photograph of two clusters from one cane of a Marian X 
Pocklington cross, which grew in the vineyard of the New York 
State Experiment Station. It will be seen from this photograph ~ 
that the flowers of this vine are functionally so nearly midway 
between the pistillate and perfect forms that on the same cluster 
some pistils have sufficient stigmatic tissue to permit of pollen 
germination, while others do not. During the three seasons in 
which the writer had the opportunity to observe this vine some 
clusters bore fruit each season. In PLATE 2, FIG. 15-18, there are 
shown photomicrographs of median sections of four pistils from 
this vine. Flowers with a similar stigmatic development have 
been observed by the writer on two wild vines of V. vulpina and 
also in a number of crosses at the New York State Experiment 
Station, in which V. Labrusca, V. bicolor, V.vulpina, and V. vinifera 
occurred. 
The series of median sections included (PLATES 2 and 3, 
FIG. 11-24) shows practically all gradations in stigmatic develop- 
ment between the truly perfect and the staminate forms. Even 
the purely staminate forms represented (PLATE 3, FIG. 19-24) 
in the sections show an abortive pistil with no stigmatic tissue, 
yet having carpels and rudimentary ovules. The statement of 
Booth (’o2) that all staminate flowers observed have small abortive 
pistils is corroborated. By following the series, then, it will be 
seen that the seed coats in the ovules are not fully developed 
in staminate forms and only partially so in some of the intermediate 
forms with small stigmatic surfaces. This fact is significant in 
that it indicates, that there is not a complete segregation of the 
pistillate and staminate forms but a suppression of pistil develop- 
ment in the case of the staminate flowers. 
Median sections through the pistils of some of the intermediate 
