338 Mottier . — The Development of the Heterotypic 
17, 18). The contracted mass loosens up as the regular spirem, in which 
may be recognized the series of chromomeres held in the linin matrix. 
These chromomeres are for the most part paired, which I think is due to the 
longitudinal splitting of single chromomeres. The paired chromomeres 
vary somewhat in size, but on the whole there is much uniformity. Some 
of the chromomeres are, however, not paired, at least so far as can be made 
out, and many of the unpaired ones are smaller than the average. Now 
the question is, what happens in synapsis? According to Strasburger 
(’ 05 , p. 37), there takes place in the synaptic mass a rearrangement by 
which homologous pangens are brought together. ‘ Meine Ansicht ist 
jetzt also die, dass durch die Wechselwirkung der beiden Gamosomen ihre 
Pangene eine bestimmte Orientierung erfahren, so dass sie bei der darauf 
folgendenStreckung dieser Gamosomen eine ubereinstimmende Aufeinander- 
folge erhalten. Dieser vorbereitende Schritt hatte somit das Ergebniss, 
die homologen Pangene an der gestreckten Gamosomen einander un- 
mittelbar zu nahern.* The bringing together of homologous pangens means 
that every maternal pangen is brought near to its corresponding paternal 
pangen. Since Strasburger regards the longitudinal fission of the spirem 
as being due to a union of two parental threads rather than a fission, one 
naturally infers that he understands that the homologous groups of pangens 
are brought together side by side. 
We cannot, of course, determine what takes place in synapsis, but it 
may be possible that some kind of a rearrangement occurs. In those cases 
in which the larger masses of chromatin are present just prior to the 
contraction, it is certain that these larger masses are broken up into much 
smaller ones before the loosening up process begins. If any rearrange- 
ment takes place, this must happen either during the balled-up condition or 
earlier; for it is less probable that a rearrangement and an exchange of 
pangens should be brought about in the longitudinally split spirem. In 
view of the orientation of the parental chromosomes in the spirem, that is, 
end-to-end, as set forth in this paper, it seems to me that, if some sort of 
rearrangement or interchange takes place in synapsis, this must be different 
from that held by those who regard the parental chromosomes placed side 
by side in the spirem. To me it seems that in synapsis or earlier the 
pangens of like affinity, or those bearing like characters, are brought 
together into chromomeres. These chromomeres are then organized into 
chromosomes. Each chromosome is accordingly composed chiefly of 
homologous chromomeres, which become arranged end-to-end, and which 
split lengthwise to form the paired chromomeres. A longitudinal fission 
means an equal distribution of homologous chromosomes, and this we 
know is accomplished during the second or homotypic mitosis in all plants 
thus far thoroughly investigated. 
If it is assumed that all chromosomes are not alike in every quality, 
