tort] PRITCHARD—DISSEMINATION OF PUCCINIA 183 
the beginning of the experiment, about 10 per cent of the uredo- 
spores were viable, but by the end of September this had dwindled 
to 2 per cent; and only an occasional uredospore germinated in 
October and none whatever after November 15. 
Some confusion arose at first over a fungus whose hyphae 
emerged from the germ pores, but without showing any con- 
spicuous evidence of its entrance. This fungus proved to be an 
Alternaria, which parasitized many uredospores. 
Repeated attempts were made to germinate uredospores from 
Hordeum jubatum and occasionally from Agropyron repens which 
were buried under ice and snow, but always without success. 
Old uredospores can be obtained in abundance all winter and in 
early spring on Hordeum jubatum, lying between the stem and 
sheath, but it is practically impossible to find them here on other 
grasses in winter, although the plants may be buried under ice in 
low places, as they drop off before winter and are replaced by 
teleutospores. Hence no uredo pustules were available from 
other grasses except Agropyron repens, whose few winter-borne 
uredospores would not germinate. 
The germination tests show that during the winter of 1904-1905 
in North Dakota all or practically all the uredospores of P. graminis 
probably lost their viability, and hence were not the cause of the 
large amount of black rust in the state the following summer. 
The annual reappearance of P. graminis in Kansas, Nebraska, 
and the Dakotas has often been explained by assuming that it 
passes the winter in Texas and spreads north by means of the 
wind and growing crops. To obtain data on this point an endeavor 
was made to catch uredospores of P. graminis from the air before 
any pustules appeared upon the wheat. A post 5 feet high was 
set in the edge of a wheat plot and a soup dish 7 inches in diameter, 
containing a small quantity of distilled water, just enough to fill 
one tube of the centrifuge, was exposed at its top 30-40 minutes. 
The whole inner surface of the dish was rinsed with the water 
to collect any spores adhering to its sides. The water was then 
poured into a tube provided with a tapering bottom, and the 
débris precipitated by means of a centrifuge. All the particle- 
bearing liquid was removed from the narrow end of the tube by 
