332 Mottier . — 77 z<? Development of the Heterotypic 
results seem to show conclusively that there is no definite relation existing 
between the number of these lumps and the number of somatic chromo- 
somes in Podophyllum , L ilium, Tradescantia , and Galtonia , and conse- 
quently these plants lend no support to the prochromosome theory. 
In regard to the differentiation of the nuclear framework into chro- 
matin and linin, the writer believes that there can be little doubt in this, 
and he differs from the view expressed by Gregoire and Wygaerts (’ 03 ) 
that all is chromatin. In a later publication Gregoire (’ 06 ) has modified 
his view in this particular, admitting that such a differentiation may exist. 
It is true that in resting nuclei, in which the strands of the network 
connecting the chromatin masses are rather thick, these strands retain 
more of the stain, in which case the difference in appearance between 
chromatin and linin seems to be small indeed ; and, if the iron-haema- 
toxylin stain is used, frequently no difference is at all apparent. On the 
contrary, in such nuclei as Figs. 17, 19, and 40 of the presynaptic stage, 
and in Figs. 7 and 25 following synapsis, good staining by both the triple 
and iron-haematoxylin methods shows beyond doubt that we have two 
substances sufficiently distinct to justify the designation chromatin and 
linin. By this statement the writer does not intend to convey the idea 
that our methods of differential staining, however perfect they may seem, 
demonstrate completely the difference between chromatin and linin, but to 
indicate that such difference really exists, leaving the ultimate nature of 
the two substances to be determined by future research. 
This study has also shown that in the same plant the manner of the 
distribution of the chromatin may be different prior to the synaptic 
contraction (Figs. 22, 23, 19, and 20). In some cases the lumps remain 
relatively large, while in others a breaking-up into much finer granules 
obtains. In all cases, however, the thread on emerging from the balled-up 
mass is made up of chromomeres in which there is a general uniformity in 
shape and size. This shows conclusively, that, whatever else may happen 
within the synaptic mass, this is certain, that there does take place in cases 
as shown in Figs. 22 and 34 a further reticulation of the chromatin. To 
this phase of the subject we shall return beyond. 
As repeatedly asserted in the foregoing, the chromatin spirem emerges 
from synapsis, showing for the most part a double row of granules or 
chromomeres, which I interpret as the result of a longitudinal fission of the 
chromomeres, and that the entire thread is double seems to be indicated by 
the divergence in certain instances of the sister threads (Fig. 24). That we 
do not see all chromomeres paired or of the same size does not argue 
against the doctrine of a longitudinal fission more forcibly than against 
the union of two different spirems ; and in view of the manner in which the 
bivalent chromosomes are formed from the spirem, it is reasonable to con- 
clude that, if the longitudinally split thread is not the result of fission, there 
