268 BOTANICAL GAZETTE [APRIL 
put into them by him at Lafayette, Indiana, and mailed at once to 
Madison. The usual time for fixation was from two to three days. 
Flemming’s triple stain gave the best results. Good preparations 
were also obtained by the use of Heidenhain’s haematoxylin. 
In the case of Phragmidium, material was found of all ages, 
so that the entire history from the uninucleated mycelial cells to the 
fully formed binucleated spores could be easily followed. It was 
found that in the young pustule the hyphae form a layer just beneath 
the epidermal cells (fig. 1). This layer, in which the direction of 
the individual hyphae is almost lost, is usually from one to three 
cells thick. The mycelial cells here are much thicker and shorter 
than those found in the vegetative hyphae. Each cell has a very 
dense, finely granular cytoplasm and a single large nucleus. The 
nucleus occupies a central position in the cell. There is a well-defined 
nuclear membrane and one large nucleole., The chromatin is always 
-more or less massed at this stage, making a ragged net with the 
strands very irregular in thickness. Thus a great deal of clear space 
is left in the nucleus, which is filled with nuclear sap. The chromatin 
is regularly stained by the violet of the triple stain and the nucleole 
stains a deep red. 
Certain cells now become elongated in a direction perpendicular 
to the epidermis, forming a continuous series and raising the epl 
dermis (fig. 2). The exact origin of the cells of this series is hard to 
determine; many of them can be seen to be end cells of short hyphal 
branches. In thick sections, long branches may often be traced 
between the subepidermal cells of the host, which terminate in one 
of these cells. In the cacoma the cells are not crowded; there 's 
even considerable intercellular space. Very often it may be seen 
that cells standing side by side arise from different hyphal branches, 
or at least from distant parts of the same hypha (jig. 2; 4 and 6). 
The single nucleus of the cell now apparently divides and the cell 
elongates into a rather narrower upper part which is cut off as 4 
small distal cell from the larger basal cell. The cell division 1s very 
unequal. The end cell as a result is only about one-third ee 
of the cell beneath it (fig. 3). The protoplasm of the end cell is at 
first dense and appears quite normal. It soon becomes very eee 
however, and finally appears quite clear, with only a few strands 0 | 
