June i6 ,1933 
Pucciniastrum americanum 
891 
orange stain more deeply than^the others, therefore this layer is easily 
recognized in sections. Disorganization, followed by collapse, occurs, so 
that these interstitial cells are represented finally by more lines of degen¬ 
eration products. 
METHODS OF SPORE FORMATION 
The formation of uredospores in Puccinia from a definite layer of basal 
cells which send out buds (spore initials) is well understood. Nuclear 
division is followed by cell division, cutting off from the spore initial 
the stalk cell below and the spore above. Such spores are thus not 
borne in chains, even though the basal cells may continue to bud and 
give rise to a number of spores. 
Tiro (jo, p. 4^0, 492^ 4gj) states that the uredospores of Melamp- 
sorella cerastn and M. fetirichii are borne in caeomalike sori, that is, in 
chains. Magnus {12), while confirming Liro regarding the first species, 
disagrees with him as to the second, figuring stalked spores in a sorus of 
M. jeurichii, and further holds that it would be impossible for a spore 
to be formed out of such stalk cells. He creates the new genus Milesina, 
especially characterized by stalked uredospores, leaving in Melamp- 
sorella M. cerastii, which he and Liro agree has catenulate spores. 
Ludwig and Rees (jj), who made a study of the uredinium in P. 
agrimoniae, find that the apical cells of the primordium elongate and 
their contents stain less deeply. The terminal cells become the first 
peridial cells, and the other cells of the chains are the uredospores, no 
intercalary cells being formed. These authors further state that the 
spore chains arise from a layer of basal cells just above a “hyphal plate 
layer,” which is a tissue extending across the bg.se of the sorus and con¬ 
necting at the margins with the peridium. Stages in spore formation 
are not shown, although they follow Magnus in their explanation of the 
probable method. Since intercalary cells were not found, they suggest 
that the peridium is formed in a way analogous to that in the ordinary 
aecidium. They also suggest that the genera now included in the Pucci- 
niastratae may be separated into two subgroups, the fern rusts having 
pedicellate spores in one group, and Pucciniastrum, Melampsorella, Melam- 
psoridium, characterized by spores borne in chains with each chain 
maturing but one spore at a time, in the other. 
The writer has not made an extensive study of the order of cell division 
in spore formation in P. americanum^ but there can be no question that 
the uredospores are frequently borne singly on pedicels. On the other 
hand, if the very important fact that intercalary cells are present is dis¬ 
regarded, there is considerable evidence showing that the first spores, 
especially in deep-seated sori on petioles and canes, are borne somewhat 
as Magnus describes for Melampsorella. In other words, in P. amer¬ 
icanum the first spore initials arise as the result of tangential divisions 
of basal cells and not by budding. Any claim that the upper daughter 
cell may sometimes become a spore at once, the sister cell remaining as a 
basal cell, would be difficult to disprove. 
FORMATION OF THE OSTIOLE 
The formation of the ostiole and the peristomal cells in P. americanum 
can be followed very readily by studying sections of sori which are not 
very deeply seated. About the time the first spores appear in the cavity 
beneath the central parLof the buffer peridium, it will be seen that a few 
