968 Davis.—A Comparison of the Redttction Divisions of 
reduction processes in Oenothera. Some writers have interpreted this 
stage as one which brings the chromosomes into close association in pairs 
in preparation for their distribution by the heterotypic mitosis, or even 
results in their temporary fusion during the period of diakinesis. The 
second contraction in Oenothera gives no support to these views, since, 
as pointed out, there is no general pairing of the chromosomes (certain 
material of grandijiora excepted), and little or no opportunity for close 
associations. The period in Oenothera when the chromatic elements are 
in their most intimate relations to one another is that of mid-synapsis, when 
the thickened threads are so closely tangled and massed that contact would 
seem almost certain, although this need not of course involve the actual 
fusion of idioplasms. The second contraction in Oenothera may then have 
no further significance than the assembling of the segments (sporophytic 
chromosomes) of a spireme so complexly looped that its parts are necessarily 
drawn closely together by the continuation of the general process of con¬ 
densation which operates throughout the entire phase of synapsis and ends 
only when the mature chromosomes are ready for distribution on the 
equatorial plate of the heterotypic spindle. This interpretation on the 
evidence at hand seems to the author the most reasonable for this stage in 
the Oenotheras. 
Finally we must emphasize the conclusion, also stated in the author’s 
previous papers, that the sporophytic (somatic) chromosomes of Oenothera 
are formed end to end by the segmentation of univalent spiremes which 
follow synapsis. In this conclusion Gates, Geerts, and the writer are in full 
accord. There seems to be no possibility of interpreting the spireme as 
bivalent, i. e. composed of two spiremes (paternal and maternal) united side 
by side. Even should it be established that the spireme in earlier stages of 
developments shows indications of a doubling or lengthwise fission (a strep- 
sinema stage, of which the writer has seen no clear evidence), such a fission 
could have no relation to the hypothesis of the conjugation of parallel 
spiremes, since the chromosomes in the full diploid number are formed from 
as many segments of the spireme and are not formed in pairs from a haploid 
number of spireme segments. Such a doubling of the spireme (if present) 
might be related to the lengthwise fission of the chromosomes which 
appears during the anaphase of the heterotypic mitosis in preparation 
for the homotypic division, but it is difficult to see any other probable 
interpretation of importance. Thus it appears impossible to bring the 
facts into accord with the hypothesis of a side-by-side pairing of the 
chromosomes through the association of two parallel spiremes or two parallel 
sets of chromosomes as held by Gregoire, Allen, Rosenberg, Overton, 
Stomps (TO), and their followers. The results from the studies on biennis , 
grandijiora , Lamarckiana, rubrinervis , and gig as strongly support the view 
of Farmer and Moore, Mottier, Digby (TO), and others of an end-to-end 
