1904 ] CURRENT LITERATURE 483 
divisions appeared in the embryo sac mother-cell. Cold weather followed 
and all stages of development in the nucellus were suspended until the end 
of April. The mother-cell gives rise to four megaspores arranged in a 
row, though other arrangements are sometimes found, due presumably to 
the shape of the mother-cell. The four megaspores correspond to the four 
Usually o 
megaspore mother-cells reaches the synapsis stage. The mitotic figure 
‘at the first division of the mother-cell is plainly heterotypic and 
the number of chromosome pairs seems to be eight. The expected 
sixteen chromosomes were counted repeatedly in vegetative tissues, but 
a smaller number was sometimes found. The second division is homotypic. 
The lowest of the four megaspores is almost ee the one to germinate, 
but any of the others may develop, and in one case two contiguous mega- 
spores of one row were observed with several nate nuclei. Frequently 
more than one megaspore in the nucellus may begin to develop. The 
mitotic figure at the first division in the germinating megaspore is heter- 
otypic, and generally eight chromosomes could be counted. This number 
was also counted at the second division. When the eight-nucleate stage 
is reached the nuclei have begun to assume a peripheral position in the sac, 
but there is no polar grouping as in angiosperm sacs 
Strasburger calls attention to the fact that Ceratozamia mexicana also has 
eight to sixteen chromosomes, while other gymnosperms, so far as determined, 
have twelve and twenty-four. Many embryo sacs nearly filled with tissues 
be compared with the fusion of polar nuclei in embryo sacs, but no fusions 
were found. The archegonium initials appear as soon as the embryo sac is 
filled with tissue. 
After discussing the development of prothallia and fertilization in 
Gnetum, the conclusion is reached that it is not possible to derive the 
embryo sac of angiosperms from that of Gnetum, the resemblances being 
only phylogenetic parallels. The densely staining areas in the embryo sac 
mother-cells of Casuarina and of various gymnosperms were studied in 
axus. Though denser, they have the same general structure as 
trophoplasm. It is suggested that these structures may be useful in deter- 
mining the position of Casuarina. 
The nuclear fusions which initiate the formation of endosperm in angio- 
sperms are not found in gymnosperms and show no indication of their origin. 
They are not to be regarded as sexual fusions, although, under some 
conditions, the fusion of sister nuclei may be a process of fertilization.— C. J. 
HAMBERLAIN. 
