14 BOTANICAL GAZETTE [JANUARY 
While the formation of the pollen tube was not studied in 
detail, it seems to take place in the usual way. Microspores 
germinating on the stigma show the male cells still distinct 
g. 69). Inno observed instance did all the spores of a group 
develop tubes; usually only one or two members of the tetrad 
germinate, the others being held back from the stigmatic secre- 
_ tions. The tubes show very prominently in cross-sections of the 
flower. They pass down the central conducting strand; occa- 
sionally one may be seen in one of the air chambers, but serial 
sections generally show it to be only a loop put out from the 
central tissue. 
Having reached the upper end of the ovary, the pollen tubes 
pass with singular directness down through the ovarian cavity to 
the upturned micropyles of the ovules. One can but remark the 
efficiency of the chemotropism and the precision with which the 
tubes pass to the openings in the ovules (fig. 75). Few to many 
curved, tangled, and feebly staining pollen tubes are usually 
present among the fertilized ovules in the cavity. These seem 
to have come into the ovary too late; at any rate, the function- 
ing pollen tubes are always much straighter and stain more 
deeply than the ones which have failed to enter the ovules. 
It not infrequently happens that one of these functionless 
pollen tubes, having failed to enter a micropyle, may swell up at 
its tip and terminate its development in a cyst-like enlargement 
(figs. 70-74). Scores of these growths were encountered in 
material of all stages of development after fertilization, collected 
through two seasons from various stations, and killed in different 
ways. Their occurrence, therefore, is so uniform as to merit 
some attention, especially as they do not seem to have been dis- 
cussed in connection with any other plant. These cystoids may 
be found lying anywhere in the cavity of the ovary, but seem 
never to occur in the style nor in any tissue through which the 
tube passes. They may lie along the walls of the cavity or in 
contact with the ovules, but usually are free in cavity of the 
ovary. In form they vary from spherical to oblong; some may 
be irregular in outline or even lobed, while those in contact with 
the wall or ovule are often much flattened or elongated (fig. 74)- 
