Urocystis Cepulae. T. Whitehead. 69 
sporidia of Uvocystis Anemones. An attempt to determine the 
relative importance of the sporidial germ tube and promycelium 
in securing infection was made by the writer in the following 
way. Young roots severed from healthy seedlings were fixed to 
coverslips with Canada Balsam. Chlamydospores germinating 
in water were then sown on the roots and the coverslips inverted 
over Van Tiegham cells. The promycelia produced by the 
chlamydospores grew over and amongst the root hairs but in 
no case were they observed to penetrate the walls. As, however, 
it proved to be impossible to see whether sporidia were produced 
on the promycelia, the attempt failed in its main object; only 
succeeding, in fact, in demonstrating that infection does not 
invariably occur when a promycelium comes in contact with 
an onion root. 
GROWTH IN Host PLANT. 
Once inside the host plant the hyphae develop intercellularly 
at approximately the same rate as the cotyledon grows in 
length. As in other members of the Ustilagineae the mycelium 
is evanescent and is difficult to trace far from the region of 
sporulation. The hyphae are sparsely septate and branching 
occurs at relatively infrequent intervals. These branches usually 
turn sharply forward and grow along the longitudinal walls, 
so producing a mycelium characterised by parallel hyphae. At 
intervals, which are long in the non-sporulating region and 
short in the part of the leaf in which sporulation occurs, hyphal 
branches pass along the end walls of the host cells and anasto- 
mose with hyphae growing at a different tangential or radial 
level. Branched haustoria resembling the intercellular hyphae 
are produced but do not form a prominent feature of the 
mycelium. As the spore forming region is approached the 
hyphae branch much more frequently until a compact mycelium 
is developed in which individual hyphae are difficult to trace. 
A pseudoparenchyma, however, is not produced nor is there 
any gelatinisation of the hyphal walls as described in other 
species of Urocystis (6). 
The sporogenous mycelium is practically confined to the 
mesophyll lying between the vascular bundles. Only rarely is 
the phloem invaded and spores have never been found in the 
outermost layers of the mesophyll or in the epidermis. 
DEVELOPMENT OF CHLAMYDOSPORE. 
The development of the chlamydospore or spore ball is not 
easy to follow owing to the denseness of the mycelium in which 
the spores are produced. Most success has been obtained by 
fixing in Carnoy’s fluid young seedlings in which sporulation 
had just commenced and cutting thick (12-18 ») longitudinal 
