DODGE: STUDIES IN THE GENUS GYMNOSPORANGIUM 135 
leaves have been fixed at a time when the sorus is fully matured. The 
mycelium is confined to the leaf bearing the sorus and does not ordi- 
narily invade the stem at any point along the line of attachment. 
The large cells surrounding the leaf vein appear to prevent the my- 
celium from entering the phloem of the stem. 
At the point where the sorus is to be developed, we find a well- 
defined pseudo-parenchyma, the cell walls taking the orange stain. 
The upper cells of this pseudo-parenchyma are somewhat enlarged 
and elongated. These are likewise binucleated. They soon begin 
to swell, lose their cytoplasm, and the nuclei degenerate. In order to 
show these upper cells in this condition the material must be fixed at 
the earliest possible time that an infected leaf can be distinguished, 
Fic. 1. Section of a sorus of G. transformans on a leaf of the southern 
white cedar at the narrowest portion of the young sorus. The epidermis is broken 
up On either side, only traces of the cuticle and fragments of the cell walls being 
visible. At the center epidermal cells are still visible. A number of buffer cells 
in various stages of degeneration can be seen, and binucleated teleutospore buds 
growing through the buffer cells are common. 
that is, when a spot appears as a slight, waxy, translucent, light orange 
blister. In such cases the epidermis may not have been ruptured 
- and fixation of the mycelium is not apt to be of the best, unless the 
leaf is cut through. The upper cells mentioned become mere bladdery 
sacs and during this process of swelling the inner walls of the epidermal 
cells, and hypodermal cells when present, are broken down either by 
enzyme action or by actual pressure, and the heavily cutinized epi- 
dermis is lifted up and split open (Text-fig. 1). The splitting usually 
occurs in a line along one side of the leaf, but very often the split 
runs longitudinally down through the middle. Sometimes two sori 
develop side by side on the same leaf. The bladdery cells evidently 
function as buffer cells to disrupt the epidermis. These buffer cells 
perhaps represent simply the first series of teleutospore mother cells 
