406 BOTANICAL GAZETTE [NOVEMBER 
instances, however, do the cells produced look as if they might be 
normal. The cytoplasm is much vacuolated, and the nuclei are 
not distinct, even in the best of preparations (figs. 67, 68, 72, 73). 
HYPERTROPHY OF POLLEN GRAIN.—Those grains that appear 
most nearly normal have little or no starch (fig. 67). Pollen 
grains that have proceeded to full wall development and genera- 
tive cell formation begin to enlarge until about twice normal 
volume (fig. 68). In all such there is considerable accumulation of 
starch grains, and in some the cell is tightly packed with relatively 
enormous grains (fig. 69). Whenever the nuclei show at all, they 
are irregular in outline, poorly defined, and almost uniform in 
texture. Later, the starch is dissolved, and the contents of the 
pollen grain become more and more homogeneous, until in the 
end the grain is filled with a granular substance staining deeply 
and uniformly throughout. In the best preparations, some of 
these cells still showed the remains of the nuclei; but there is 
little doubt that they are destined to go to pieces (figs. 70-73). 
In every case the tube nucleus was irregular and difficult to trace, 
but often the two male nuclei, although very small, still held their 
shape and showed separate chromatin masses. Usually, however, 
only deeply and characteristically staining masses indicate the 
remains of the nuclei (figs. 71, 72, 75). 
“POLLEN TUBE’? FORMATION.—Hypertrophy of the pollen 
grains may be due to high osmotic pressure set up in the disinte- 
grating contents. The starch formation undoubtedly is patho- 
logical. The conditions in the cell that induce starch formation 
probably also cause other changes, which lead to greatly increased 
osmotic pressure, which is further increased by the solution of the 
starch. Very frequently these large pollen grains put out pollen 
tubes through all three of the germ pores. These tubes may remain 
mere bulbs (fig. 72), or they may become many times the length 
of the cell producing them, and are always almost solidly filled 
with the dense homogeneous contents of the mother cell (figs. 74, 
75). Often the disorganizing nuclei migrate into the tubes, and the 
pollen grain may be almost entirely emptied (fig. 74). The tubes 
often become inflated at the end, as if under pressure, and it 
appears that they sometimes burst. 
