56 FARR: QUADRIPARTITION IN SISYRINCHIUM 
condition. It was found that after the disappearance of the 
ephemeral orange zone a furrow developed on the margin of the 
equator of the heterotypic spindle. However, the development 
of this furrow is arrested when it is only about one-fourth 
completed and the homoeotypic nuclear division ensues. The 
division of the cell is finally accomplished by the completion 
of this furrow and the formation of two other furrows at approxi- 
mately right angles to it. Although Nymphaea has not as yet 
been thoroughly investigated the work of Lubimenko and Maige 
(16) indicates that it is similar in this respect to Magnolia. 
The following study was made upon material of Sisyrinchium 
Bushii Bicknell, collected near College Station, Texas, and 
identified by Dr. A. S. Hitchcock. This species proved especi- 
ally fine for study, both because of the excellent fixation and 
because of the distribution of the mitotic figures in the anthers. 
In Nicotiana the mother-cells of a given anther are all in very 
nearly the same stage of division, and hence to secure an un- 
broken series of stages it is necessary to take material from several 
different anthers. In Magnolia the mother-cells within the 
anther were at different stages but there was no special arrange- 
ment of them. In either of these cases the objection might 
be raised that a cell-plate was overlooked due to the study of 
an incomplete series of stages. It was therefore highly desirable 
that a form be found in which the stages were arranged in order 
within the anther from one end to the other, so that by studying 
a number of anthers it could be determined with certainty that no 
steps have been omitted. Such a situation is that which Sisy- 
rinchium Bushii presents. The mother-cells at one end of the 
anther may be in metaphase of the heterotypic division and 
those at the other end in metaphase of the homoeotypic. This 
is the condition shown in Fic. 1.. In Fic. 2, the cells at the inner. 
end are in metaphase of the homoeotypic division and those at 
the outer end are in the tetranuclear stage. It is evident that 
the study of a number of such anthers is sure to reveal every 
stage, however transitory. It will be remembered that a similar 
condition exists in the testes of Batracoceps, and among plants 
it was found by Strasburger (22) in the anthers of Fritillaria and 
by D’Angremond (1) in the banana. 
e pollen-mother-cells of Sisyrinchium are smaller than 
those characteristic of many Monocotyledons. Miss Kliene- 
berger (15) has recently published a study of the size of nuclei 
