Cannon: StTuprEs IN PLANT Hysrips 169 
tions, and these go to the opposite poles of the spindle and there 
build the tetrad-nucleus. 
Although in many regards the description of the mitoses just 
given must be considered as a meager one, I think that enough 
has been said to show that they are the exact homologues of these 
nuclear divisions in pure races of plants. That is, in the cotton 
hybrid the first maturation is a heterotypic one and the second is 
homotypic. 
My material was not of sufficient amount or kind to permit me 
to study the subsequent nuclear division of the tetrad, or the 
germination of the spore. 
The division of the cell does not take place until the daughter 
nuclei are formed and completely reconstructed. When the di- 
vision occurs the planes of division are at a sharp angle with one 
another and pass between the daughter nuclei, and the result is an 
equal apportionment of the cytoplasm among the four nuclei of 
the group. Indentations appear in the periphery of the cytoplasm 
and midway between the nuclei, which deepen into constrictions, 
and finally accomplish the separation of the nuclei and the forma- 
tion of the tetrads. 
A cell-wall is not organized until the tetrads are built and the 
formation of the wall takes place as follows: colorless accretions 
appear on the periphery of the tetrad ; later a delicate membrane 
may be distinguished connecting them, Both the thickened por- 
tions and the membrane are evidently produced by secretion and 
respond to the test for cellulose. Somewhat later the spore-wall 
_ becomes relatively thick by the addition of the cuticularized exine ; 
the places where the cellulose accretions first appear mark the lo- 
cation of the germinal pores. 
Abnormal Nuclear Divisions. —The foregoing account of the 
tetrad formation in hybrid cotton is based on the study of anthers 
which were among the first to be put up, namely October 29, 
and were taken from the form known as “344.” A study of ma- 
terial collected in November and December showed that most of 
the male cells were perfectly normal. The flowers gathered in 
January were unfortunately older, and did not therefore show the 
division stages, but the appearance of the cytoplasm as well as of 
the nuclei was such as to indicate that normal mitoses had taken 
