298 AMERICAN JOURNAL OF BOTANY [Vol. 9, 
at all is developed after the heterotypic karyokinesis, though they report 
in extremely rare instances that a hyaline line may be seen across the equator. 
In Nymphaea alba they find a transitory cell plate appearing in the late 
anaphase of the homoeotypic mitosis, and their figure 51 substantiates this 
observation. No such transitory plate is found in Nuphar luteum. Their 
figures 53 and 52 show the tetranucleate stage of these two species respec- 
tively, with no partitions or cell plates present. Their figure 54 is of a 
pollen mother cell of Nymphaea alba after the partitions are completely 
formed separating the four cells. No further figures are given of cyto- 
kinesis. They describe the 'formation of the partitions as follows: 
Se forment brusquement les plaques cellulaires. Ce phenomene se produit tres 
rapidement et on n'apergoit, sur les coupes, que les plaques entierement formees. 
They interpret the disappearance of the transitory cell plate as due to the 
fact that the spindle fibers upon which it is formed are of nuclear origin, 
and that the material of which they are composed is used again in the 
formation of the linin thread and nucleoli of the new nuclei. The cell plate, 
which they believe is formed later and finally accomplishes division, is con- 
sidered as being formed on spindle fibers of cytoplasmic origin appearing 
after the disappearance of the transitory cell plate. 
Several considerations make it seem advisable to investigate further 
the cytokinesis of the pollen mother cells of some of the Nymphaeaceae. 
According to the work of Lubimenko and Maige (13), it would seem that 
Nymphaea alba resembles Magnolia in the existence of a transitory cell 
plate after the heterotypic nuclear division, but that it differs from Mag- 
nolia in that no incipient furrow is developed at this stage. Nuphar luteum, 
on the other hand, seems to correspond exactly to other flowering plants 
having quadripartition in so far as the events immediately following the 
first division of the nucleus are concerned. Furthermore, the instance of 
the disappearance of a transitory cell plate and the later formation of a 
permanent cell plate on the same spindle reported by these authors for 
Nymphaea apparently stands alone in the literature as the only case of 
such a behavior, and in itself would warrant further investigation. 
For these reasons I collected pollen mother cells of Nelumbo lutea (Willd.) 
Pers. on July 9, 1921, at the Amana colonies in Iowa County, Iowa, where 
they grow luxuriantly in a large pond (text fig. i). On this date large 
numbers of the flowers were in full bloom, but the blooming season extended 
on into August so that it is likely that the flowers used in this study would 
have bloomed in about mid-season. Reduction divisions occur in the 
pollen mother cells of this species when the buds are about one inch in 
diameter and about one and one half inches long. At this time the bud is 
between one and two feet above the water, the peduncle elongating only a 
few inches more before its growth ceases. In this, as well as in the forms 
which I have previously studied, it has been found that the stages of the 
