40 British Uredinee and Ustilaginee. 
when it occurs on various species of Hordeum, cons 
almost entirely of mesospores (=P. anomala, Rost). 
American species produces bicellular spores in the sumr 
(= P. vexans, Farl.), and subsequently mesospores o 
(= Uromyces brandeget, Peck). The brown colouratior 
the membrane of P. graminzs is sometimes abnormally c 
fined to the pedicel, the spore remaining colourless. 1 
condition I first observed in some specimens sent me by | 
W. Marshall, from near Ely. I have also seen it in A 
tralian specimens. Upon the other hand, instead of spc 
being bicellular, they are sometimes variously compl 
triseptate, or even quadriseptate; or the upper or lo 
cell may be longitudinally or obliquely septate, which 
the normal condition of the spores of Triphragmit 
Greville * figures this condition as P. variabilis. T 
septation is purely accidental, occurring in one spore-b 
but not in any of the others on the same plant. Mr. V 
and Professor Trail have met with this condition in 
bullata, and the Rev. Dr. Keith in P. lychnidearum. 
have seen it in one of the graminaceous species on Loli, 
perenne, and Mr. Grove in Puccinia betonice. 
The teleutospores of Phragmidium and Xenodocl 
differ from those of Puccinia in being composed o 
greater number of elementary spores, which are develo 
from above downwards (basipetal formation). The dila: 
upper end of the hypha becomes full of coarsely granu 
protoplasm; delicate septa appear from above dov 
wards. The protoplasm thus differentiated is soon s¢ 
to be surrounded by a cell-wall (endospore), which 
this stage of its development is considerably smal 
than the compartment in which it is situated; so that : 
young spores are obviously, in structure, individually « 
tinct from one another. The uppermost spore is usua 
* Greville, ‘ Scot. Crypt. Flora,” t. 75. 
