g2 BOTANICAL GAZETTE [FEBRUARY 
Fig. 16 probably shows the origin of one of them by the cutting off 
of a loop of the spirem. There can be no doubt that these bodies 
arise from the spirem thread previous to the breaking up of the 
remainder of the thread into chromosomes. The nucleoli at this time 
are large, globular, and deeply-staining. Fig. 20 shows a slightly 
later stage, in the prophase, where two of these ring bodies, which I 
shall call ‘‘heterochromosomes’’ on account of their later history, 
are present in addition to the nucleolus and several chromosomes. 
That these are not the ring-shaped chromosomes frequently found 
in the prophase of the heterotypic mitosis is shown by their later 
history; for the rings do not segment in metaphase, forming V- or U- 
shaped structures, but pass towards the poles of the spindle or into 
the cytoplasm undivided and retaining their ring form. The shape 
of the ordinary chromosomes, which are nearly globular or slightly 
elongated and pear-shaped, precludes the possibility of the pairs 
forming the X, Y, V, and 0 shapes which characterize the heterotypic 
mitosis when the chromosomes are long and rod-shaped. 
The heterochromosomes are not always formed, however, and 
hence if the original pollen mother cells were all alike in the early 
synapsis stages, two kinds of pollen cells must result; for, as will 
be shown later, the heterochromosomes probably represent a portion 
of the chromatin which is rejected and disappears in the cytoplasm, 
so that the daughter nuclei must differ in their chromatin content. 
Fig. 21 shows thirteen chromosomes in the prophase at a Ister stage, 
when they are condensed to a considerably smeller size. The adja- 
cent section shows a nucleolus and one chromosome, but no hetero- 
chromosomes are present. It has been determined from a number of 
counts in the prophase that the sporophyte number of chromosomes, 
exclusive of the heterochromosomes (or at least when the letter are 
not present), is fourteen. Whether the number is the same when the 
heterochromosomes are present has not yet been determined with 
3 The term was proposed by MonTGoMERY as a general term for the aberrant 
chromosomes in Hemiptera, and is used in preference to “idiochromosomes,”’ because 
the latter term has been applied to a certain type of heterochromosomes by WILSON 
- More recently MonrGoMERY he 20) has proposed the term “allosome’”’ as 4 
general term for all aberrant chrom and “‘monosome” for one which passes 
mes, a 
undivided to one of the daughter aot The latter term ot be most suitably 
applicable to the bodies described here. 
