1g1t] PRITCHARD—DISSEMINATION OF PUCCINIA 179 
dant in this region. Either the aecidiospores are not borne as 
great distances by the wind as formerly supposed, or their ger- 
minative power is remarkably low. There is some additional 
evidence in support of the latter view. 
Plots of small wheat plants in the experimental garden were 
sprayed repeatedly with aecidiospores in the spring of 1905, yet 
scarcely any rust appeared until the plants were nearly two feet 
high, a fact commonly observed here in the field every spring, 
although volunteer wheat plants barely out of the ground in the 
fall are often covered with rust. 
Two series of infection experiments were made to obtain further 
data with reference to the spreading of P. graminis by means of 
its aecidiospores. From 98 aecidial pustules, taken at random 
in 1906, a total of 368 plants were inoculated. Plants of wheat, 
rye, oats, barley, and usually Avena fatua, Agropyron tenerum, 
A. repens, and Hordeum jubatum were inoculated from each aecidial 
pustule and covered with bell jars 24-48 hours. Germination 
tests of the spores, made by placing them in water and on wet 
filter paper kept in a moist chamber at 18-20° C., showed a via- 
bility of about 8 per cent. Rust appeared only on Avena fatua, 
Agropyron repens, rye, oats, and Hordeum jubatum. No plants 
of barley or wheat were infected. These experiments were repeated 
in 1907 by inoculating 247 plants from 13 pustules of known 
origin, the original host species always being included in each 
group. Tests of the spores showed about the same percentage 
of germination as those used in the former experiments. No 
barley plants were infected, and the only wheat plants which 
developed rust were those inoculated with a form which came 
originally from wheat. The aecidiospores of only 9 pustules, 
however, of the rrr used in the two series of experiments caused 
infection. This relatively low number of infections agrees with the 
results obtained by repeatedly spraying the wheat plots with 
aecidiospores in 1905, and may partly account for the confining 
of the rust to the immediate vicinity of the barberry hedge as 
observed for the three successive years 1905-1907. It is also 
in harmony with the very limited spreading observed by both 
MARSHALL and ScHOLER when they set barberry bushes in the 
. 
