394 
M. G. Sykes 
2) The number of pairs of kuots or aggregations in the reticu- 
lum far exceeds the number of pairs of chromosomes and it is con- 
cluded that in Furikia it is inadvisable to call these structures »pro- 
cliromosomes«. 
3) Occasional contact between the pairs of knots is seen in 
synapsis but no definite and clear cases of fusion were recorded. 
The number of knots which can be seen in synapsis is often twenty- 
four, but this is not regularlv the case. 
4) A double spireme is spun out from the reticulum during 
synapsis. This stage corresponds to the »split spireme« of Farmer 
and Moore aud Miss Sargent, but is due to the paired arrange- 
ment of the constituents of the nucleus. 
5) This stage gaves rise by fusion, as was suggested by Wini- 
warter, Schreiner, Dixon & Allen 1 ) and later confirmed by Stras- 
burger’s school, to a single spireme. This by Splitting along the 
line of fusion, and at the same time by Segmentation, gives rise to 
the double Segments which are to form the chromosomes. 
6) These Segments are generally broken off as straiglit pieces 
and bend later to form loops. No definite looped stage of the 
spireme was observed; it was found impossible to consider any spe- 
cial curved Segment of spireme as the original of one of these loop- 
shaped pairs of chromosomes. 
7) The Segments gradually coutract and the fission in thern 
becomes less clear but, in preparations which show the distinction 
of chromatin and linin, the paired chromatin masses are clearly seen. 
8) The heterotype division takes place along this fission, through 
the whole length of the loop, thus giving rise to two thinner loops. 
9) In each of the daughter loops a second split is soon visible, 
again running longitudinallv along the whole length of the loop. 
This split can be seen in the reconstructed daughter nuclei and is 
the split along which the homotype division takes place. Thus in 
neither case is there a transverse split across the bend of the loop 
and the whole process reealls strongly Miss Sargent’s 2 ) description 
in Lilium , saving for the fact missed by her that the first split is 
along the line of a previous fusion and that tlierefore the heterotype 
division is a true reduction division. 
10) The reticulum and knots iu the nuclei of the pollen grains 
are unpaired throughout. 
*) Allen. 1904 & 1905. 
2) E. Sargent. 1896 & 1897. 
