1913] BURLINGAME—ARAUCARIA BRASILIENSIS 109 
could count the nuclei and be sure of having seen them all. These 
nuclei do increase markedly in size, and much of the starch which 
entered the tube at germination disappears before penetration of the 
tube into the nucellus. 
At the time that the tube enters the nucellus there are numerous 
nuclei and the body cell present. I was unable to distinguish the 
tube nucleus any longer from the others. At first only a few of 
these nuclei enter the portion of the tube in the nucellus. The body 
_ cell is usually in that portion just outside of the nucellus or just 
barely within it. There is apparently very little activity in the 
development of the tube from the time it first penetrates the nucellus 
for a short distance until spring. I did not succeed in finding the 
mitosis concerned in the division of the body cell, though the two 
male cells were found two or three times still imbedded in a common 
mass of dense cytoplasm. The two nuclei frequently differ in size 
markedly. The body cell and its nucleus have increased in volume 
after leaving the pollen grain some six or eight times by the time 
_ the pollen tube has penetrated the nucellus in October (plate fig. 
28). This increase continues on an even greater scale up to the 
time of division. After the division of the body cell, one, at least, 
of the pieces organizes itself into a large male cell or perhaps a real 
sperm. ‘These vary considerably in size and in the definiteness of 
their organization. They all agree in being very large and in being 
more or less completely delimited from the cytoplasm in which 
they are imbedded (plate fig. 29). Almost invariably the large 
nucleus is at the extreme end of the cytoplasm and the cytoplasm 
is very often evidently arranged in such a way as to lead one to 
surmise that the nucleus actively changes its position in reference 
to the cytoplasm. The nucleus is invariably at that extremity 
which would lead in the direction that the ‘‘sperm’’ would be 
inferred to be moving. The cytoplasmic body is sometimes bent, 
almost at a right angle in one instance observed, in such a way as to 
remind one strongly of the creeping movements of an amoeba. In 
one instance of two “sperms” lying in a common mass of cyto- 
plasm, apparently having just divided, there was present a peculiar 
fibrillar structure reminding one of a nuclear spindle without the 
_ chromosomes and with an aster at either pole. Whether this is 
