1904] COKER—SPORES OF CONIFERAE 213 
the megaspore mother-cells, nor do their divisions give rise to cells 
resembling the megaspores. There seems no doubt that Srras- 
BURGER and JAGER were wrong in giving the number of megaspores 
as generally three. 
In addition to Taxus baccata and Thuja orientalis (as here 
shown) there are but three other conifers in which four potential mega- 
spores have been definitely described. They are Larix sibirica, 
Pinus Laricio, and Sequoia sempervirens.’ 
As mentioned above, there is no spongy tissue surrounding the 
megaspore mother-cell in Taxus, and as the lower megaspore develops 
in sprouting, it crushes aside and destroys the cells immediately adjoin- 
ing it, eating its way gradually into the more distant layers (fig. 23). 
HoFMEISTER observed many years ago that more than one embryo- 
sac may occur in Taxus, and this I have found to be not uncommon 
in Taxus canadensis. Here one of the two prothallia, when two are 
present, is usually much smaller than the other. It is interesting to 
find that both prothallia may bear archegonia (fig. 24, a). The 
archegonia in the upper prothallium face towards the lower prothal- 
lium and not toward the micropyle, showing that the upper prothal- 
lium is inverted. The pollen tube does not stop at the upper end of 
the nearést prothallium, but growing past it presses in between the 
two to reach the archegonia. In all such cases it was the archegonia 
of the lower prothallium only that were fertilized (fig. 24). 
In conclusion I wish to thank Professor STRASBURGER for his 
Valuable advice and unfailing kindness.® 
7 See my paper on Taxodium for literature. 
8 . 
The drawings have been inked in by Mr. H. A. Allard under my direction. 
