1907] GATES—POLLEN DEVELOPMENT AND MUTATION 95 
observed a few times, while the heterochromosomes are very common 
although they are certainly absent from some pollen mother cells. 
Figs. 29, 30, 31, and 32 are sections of the equatorial plate of the 
O. Lamarckiana hybrid, showing one or more heterochromosomes. 
The chromosomes are so densely grouped at this time that it is often 
impossible to make out the individuals. This renders some of the 
figures unsightly, but they are intended to be merely accurate repre- 
sentations of what is seen on the spindle. 
After passing into the cytoplasm, where they are generally found 
in metaphase, the heterochromosomes apparently degenerate and 
disappear. They gradually become smaller and stain less deeply, 
though retaining their ring form, as in fig. 32. Finally they are 
so small and inconspicuous as to be scarcely distinguishable from the 
granules (microsomes) of the cytoplasm. It is probable that they 
usually disappear before the telophase of the heterotypic mitosis. 
Several other points in regard to their history have yet tobedetermined. 
Figs. 34 and 35 are equatorial views. Fig. 34 shows several 
chromosomes together with a heterochromosome, and also a very 
large or giant ring chromosome, which apparently contains enough 
chromatin for several chromosomes. Fig. 35 shows another; and 
these two cases were the only ones observed. They are apparently 
quite distinct from the heterochromosomes already described, although 
they are ring-shaped. No other observations have been made in 
regard to them. 
Fig. 35 is a polar view of the equatorial plate of the heterotypic 
mitosis in the O. Lamarckiana hybrid. The apparently unequal 
pairs are probably due to foreshortening, owing to one chromosome 
lying in a lower plane. Figs. 36 and 37 are early anaphases in the 
O. Lamarckiana hybrid. Figs. 38 and 39 are late anaphases. Fig. 
38 shows clearly the premature splitting of the chromosomes for the 
homotypic division. This splitting may occasionally be observed in 
late metaphase, and in the late anaphase the separation of the halves 
is often complete. V-shaped chromosomes were not observed, the 
split apparently taking place simultaneously throughout. Thus it is 
clear that the heterotypic mitosis distributes whole chromosomes, 
which were arranged in pairs (though not at first in contact) in the 
Prophase; while the homotypic mitosis is a typical division, separating 
