July, 1922] DODGE — STUDIES IN THE GENUS GYMNOSPORANGIUM 
Prof. Arthur writes that he thinks my use of the term subcuticular is 
misleading and suggests the term super epidermal as m.ore accurately de- 
scribing the location of the telium of this species. After examining a num- 
ber of species of rusts having ''subcuticular'' sori I am convinced that much 
further careful work along this line is desirable. It is possible that the in- 
troduction of the new term will be found advantageous and even necessary. 
For example, preparations of young material of Peridermium Peckii, P. 
balsameum, P. Hydrangeae, and Caeoma abietis-canadensis , loaned by Dr. 
Adams, seem to show that the pycnia of these forms arise in the cuticular 
layer of the host in the same way that the telium of Gymnosporangium 
clavipes is developed, and that in the first two forms the epidermis entirely 
disappears beneath the mature pycnium. The pycnia of these forms have 
always been described as subcuticular,^ and it is in this broad sense that 
the term has been used in this paper. 
It should have been noted that as the sorus of G. clavipes ages the walls 
of the epidermal cells frequently disappear and the massing of hyphae con- 
tinues until one can find gaps in sections where no epidermal cells separate 
the sorus from the mesophyll. This condition is due either to the gradual 
absorption or crushing down of the original cells or to the wide separation 
of them as noted previously in connection with figure 6 at A. 
The Origin of the Teliospore 
It was found that the teliospore buds of G. clavipes on trunks or cork- 
covered branches arise from subterminal cells of the basal primordium, 
agreeing in this respect with the other species previously studied by the 
writer. It then became a matter of interest to learn how the spores arise 
in the primary or first sori, which always occur on leaf blades or on young 
stems still unprovided with cork. The rapid maturation of spore initials, 
the mucilaginous degeneration products surrounding cells and the cuticu- 
larized layer, and the natural tendency of the spore walls to swell, all 
contribute to make it somewhat difficult to obtain sections showing clearly 
a series of stages in the origin of the spores. 
If one follows the development of the terminal or buffer cells in a very 
small leaf sorus, he will see that the cuticle is apparently very much stretched 
and remains unbroken for some time. Fully expanded buffer cells can be 
found in a sorus where other terminal cells are just beginning to disorganize. 
Half of a small sorus at the margin of a leaf base on a young stem is shown 
in figure 7. At A the cuticle with the clinging, disorganized cuticularized 
layer (shaded) is pushed up. Buffer cells are degenerating, and a teliospore 
bud is arising from a subterminal cell. The remains of the nuclei in the ter- 
minal cell at B show as deeply stained masses, but the cell still contains some 
cytoplasm. A young teliospore adjacent to an old buffer cell is shown 
^ Ludwig, C. A. Notes on some North American rusts with Caeoma-Hke sori. Phyto- 
path. 5 : 278. 1915. 
