336 BOTANICAL GAZETTE [MAY 
area. Measurements should therefore be made only on nuclei 
whose membrane is distinct; and of course where nuclei are cut 
by the knife, only sections should be measured which pass through 
their greatest diameter. 
The fact that, at the beginning of the so-called synapsis period, 
a conspicuous growth of the nucleus of the pollen mother cell takes 
place, without an appreciable growth in the size of the cell, means 
that there must be at this period a readjustment in the Kern- 
plasma relation (HERTWic). In another paper (6) considerable 
attention was devoted to this matter of the Kernplasma relation 
in O. gigas. 
It is known that in all higher plants the pollen mother cells 
first undergo a large amount of growth, in which the nucleus and 
cytoplasm share simultaneously. In Oenothera, as here stated, 
and probably also in many other plant and animal forms, the 
nucleus then undergoes further growth, while the cytoplasm remains 
stationary. The earlier growth of the nucleus of the pollen mother 
cell is accompanied by a corresponding growth in its chromatic 
content, so that the reticulum continues to fill the nuclear cavity; 
but its later growth is due merely to an increase in the karyolymph, 
while the amount of chromatin ceases to grow, and the reticulum 
therefore ceases to occupy the whole of the nuclear cavity. During 
this later nuclear expansion, the chromatin begins the series of 
rearrangements which change the reticulum to the spirem condi- 
tion and finally to diakinesis, and which do not necessarily differ 
in any fundamental particular from the prophases of any somatic 
mitosis. The peculiar appearances at this time, as compared with 
somatic prophases, are partly the result of the fact that the re- 
arrangements go on in a much larger cavity, which allows the 
chromatic materials to be more loosely distributed; while the pecul- 
larities of the diakinesis and the heterotypic gemini may be partly 
accounted for by the fact that the members of the pairs, instead - 
lying parallel, usually occupy a great variety of positions relative 
toeach other. Furthermore, in many forms the attraction between 
homologous chromosomes is probably greater at this time than ™ 
somatic mitoses, for the heterotypic gemini are often more closely 
paired than the chromosomes during the sporophytic divisions. 
