484 Harper—Nuclear Phenomena in the Smuts. 
vides to furnish nuclei for the daughter buds. The resting 
nuclei here also show the characteristic elements, nucleole, 
chromatin mass, and nuclear membrane. 
The division is a typical karyokinetic division, although the 
conditions of size, etc., make a study of the details of the pro¬ 
cess very difficult. The equatorial plate stage is well defined 
(Fig. 14). The spindle is sharply bipolar though no polar 
asters can be made out. The nucleole can frequently be seen 
as a bright red granule somewhere in the neighborhood of the 
spindle (Figs. 14, 16). The separation of the chromosomes and 
their withdrawal to the poles is shown in Fig. 15, and the young 
daughter nuclei appearing as dense lumps of chromatin con¬ 
nected by the fibres of the old spindle, in Fig. 16. It is inter¬ 
esting that even in such minute nuclei the typical stages of 
nuclear division as found in other plants are all readily to be 
identified. The similarity of the nuclear divisions in these 
simpler plants with those seen in the higher forms is in strik¬ 
ing contrast with the relative simplicity of their external mor¬ 
phology and life histories. 
It is quite possible that these parasitic fungi are degenerate 
in their vegetative structure as a result of their parasitic mode 
of life. In that case, however, the nuclear divisions being en¬ 
tirely reproductive processes would have naturally remained at 
least at the stage of development they had reached at the time 
the plant began its parasitic habit, since reproduction is quite 
as important for the parasite as for the independent organism, 
though it can dispense to a considerable degree with complex¬ 
ity of vegetative development. 
If such conidial cultures of the anther smut be continued for 
several days and are then allowed to starve by failure to trans¬ 
fer them to fresh culture media, a very large per cent, of the 
spores will fuse in pairs by means of tubes pushed out and grow¬ 
ing together at their tips. This fusion is accompanied by a 
very considerable increase in volume, as seen by comparing 
Figs. 12 and 17. The tubes grow out from adjacent cells and 
are apparently directed by chemotactic stimuli, so that their ends 
meet exactly. The end walls are dissolved, and the tubes fuse 
firmly and evenly together so that the points of union cannot 
