178 BOTANICAL GAZETTE [SEPTEMBER 
tables shows that under favorable conditions P. graminis passes 
readily to the barberry from wheat and certain grasses, viz., Agropy- 
ron tenerum, A. repens, Hordeum jubatum, and Elymus triticoides, 
confirming the general results of DE BArRy and others. 
TABLE Ii 
INOCULATIONS MADE UPON Berberis vulgaris WITH TELEUTOSPORES OF Puccinia 
graminis Pers. 
vical Bo eM | Source of material Method Results 
: ee, a ea March 29, ro67 | Agr. ten. Bell jar | Positive 
Pee ae eas March 29, 1907 | Agr. ten Bell jar Negative 
CR rs mmc ae ee March 29, 1907 | Agr. rep Bell jar | Positive 
Ys fieaeearas ee lacy oe March 29, 1907 | Agr. rep Bell jar | N 
Ree ols ee March 29, 1907 | Oats Bell jar | Negative 
Deon jh sess March 29, 1907 | Oats ell jar | Negative 
1 RAC See A SE March 29, 1907 | Oats Bell jar | Negative 
; Ee Pinar Gl March 30, 1907 | Wheat Bell jar | Positive 
is eens oss March 30, 1907 hea Uncovered = Negative 
IO March 30, 1907 | Hord. jub Bell jar Positive 
Sie eee March 30, 1907 | Hord. jub. Uncovered Negative 
LE Bea cena .....| April 4, 1907 eat _ Uncovered _ Negative 
PA a ee Oe / April 4, 1907 Wheat | Uncovered | Negative 
G | i 
Observations were made on the dissemination of rust from 
barberry bushes by taking note of the infection on the surrounding 
grasses. There was a small barberry hedge in Fargo very favorably 
located for this purpose, as it was surrounded on three sides by 
meadow and was heavily rusted every year. Careful observa- 
tions for three successive springs (1905-1907) furnished some sur- 
prising data. Early each year, the plants of Hordeum jubatum, 
Agropyron repens, and A. tenerum in the immediate vicinity of 
the hedge became thoroughly covered with the uredo stage, while 
Phleum pratense and Poa serotina were absolutely free from it, 
and Elymus virginicus bore only an occasional pustule. The rust 
was abundant within 25 yards of the barberry bushes, but practi- 
cally disappeared at a distance of 60 yards. The most persistent 
searching was required to discover a single pustule beyond 80 
yards, and in no one of the three springs at this early date, before 
rust had begun to spread from the uredospores, could I find fresh 
uredo pustules of P. graminis beyond 100 yards from the barberry 
hedge, notwithstanding the fact that in 1905 rust was fairly abun- 
