518 Wisconsin Academy of Sciences, Arts, and Letters. 
I am indebted again to Prof. Arthur for my material and its 
identification. Portions of the host (Potentilla canadensis) 
bearing the rust were put into Flemming’s strong fixing fluid at 
Lafayette, Indiana, and mailed to the author at Menomonie, 
Wisconsin. The subsequent treatment, including the staining, 
was the same as that described in a previous paper (3). 
In the formation of the primary uredospores, the hyphae 
mass beneath the epidermis of the host. Certain cells, which 
may nearly always be shown to be end cells of longer or shorter 
branches, assume a position perpendicular to the epidermis, and, 
by elongating, separate it from the tissue beneath. These cells 
now enlarge greatly, and divide unequally into a smaller distal 
cell and a larger one just beneath it, which is of about three 
times its size (PL XXIX, fig. 1). 
The cytoplasm of the two cells is at first similar. The nu¬ 
cleus of the smaller cell is also small, and, in general, appears 
as if it had never fully organized after the preceding nuclear 
division. The nucleole is not to be seen, and the nuclear mem¬ 
brane is very imperfect. This whole distal cell now dwindles 
and disappears. 
In the meantime, the larger cell beneath it grows somewhat, 
and comes to lie in close contact with the neighboring cells. At 
the point of contact between each pair of cells the walls dis¬ 
appear, and the two nuclei come to lie in a common cell (fig. 
2 )- 
Up to this point, the process is identical with that described 
in my previous work on Phragmidium speciosum Fr. and 
Caeoma nitens S. It seems that the walls separating the two 
fusing cells are generally more completely dissolved away in the 
case of P. potentillae canadensis Diet, than in the case of either 
of the other two forms just mentioned, since in later stages one 
does not so commonly find the condition which I have described 
and figured, in which the basal portion of the fusion cell is di¬ 
vided by the remnant of the gamete walls. The cells fusing are, 
as far as can be seen, equal, and the process is, at least appar¬ 
ently, a fusion of equal gametes, rather than the fertilization 
of an egg by the entrance of a nucleus from some other cell. 
