FARR: QUADRIPARTITION IN SISYRINCHIUM 55 
Commelinaceae, Orchidaceae, and Cyperaceae. The last-named 
is included because of the work of Juel on Carex, of which it 
it may be questioned whether the term quadripartition is 
applicable. In 1918 the same authors (24) published again, 
adding another species of Monocotyledons to this list, namely, 
Dioscorea quinquefolia. This brings the total number of genera 
up to nineteen or twenty. In Canna they report a tetrahadral. 
arrangement of the microspores but do not take this as evidence 
of quadripartition. 
Guignard (12) states that in four species of Iris there is 
occasionally a suggestion of a partition after the heterotypic 
nuclear division but this is always ephemeral. After the 
homoeotypic mitosis in Sisyrinchium, Antholyza, Freesia, Ixia, 
and Monbretia, when the fibers of the central spindle are formed 
between the four nuclei, it is stated that rather frequently there 
is noticeable a faint thickening on the internal face of the mother 
wall at the places which are to become the points of insertion 
of the partitions. Further than this no details of quadripartition 
are given, nor are drawings or photomicrographs shown to give’ 
evidence as to whether the process is accomplished by cell-plates 
or by constriction furrows. It is not unlikely that the ephemera! 
equatorial differentiations which Guignard found in these 
Iridaceae, and which Tackholm and Séderbeg reported for 
Aristolochia Sipho are orange zones, such as the writer (5) has 
shown to occur in Magnolia. Tackholm and Séderberg (24) 
do not describe the process of quadripartition in Dvzoscorea 
except to state that it is accomplished by equatorial plates 
being formed on the spindles. 
It thus appears desirable that a study be made of the details 
of the process of quadripartition in a Monocotyledon. The 
writer has discussed at some length the literature on this subject 
in his first paper in 1916 (4). At that time he presented a 
description and drawings of quadripartition in Nicotiana and 
other Dicotyledons which led to the conclusion that no cell-plates 
are formed, but that division is accomplished by furrows very 
much as in animal cells. Mrs. Wanda K. Farr (6) has more 
recently shown that the same type of cytokinesis occurs in 
Cobaea scandens as was found in Nicotiana. In 1918, the writer 
(5) supplemented the work of Guignard on Magnolia, in which 
the latter showed incipient furrows during interkinesis, but did 
not present the stages following the origin of the tetranuclear 
