314- 
Journal of Agricultural Research 
Vol. II, No. 4 
infection whatever was obtained from Puccinia andropogonis when its 
teliospores were sown on violets. 
The Pentstemon sp. inoculated with Puccinia ellisiana were infected 
about one-half as heavily as the Pentstemon sp. inoculated with Puccinia 
andropogonis , the usual Pentstemon rust, but no failure to infect suscepti¬ 
ble species of Pentstemons with either rust occurred in the experiments 
of 1914. 
A test was made with Puccinia ellisiana in the open. A small quan¬ 
tity of pedigreed teliosporic material of this rust grown under control 
conditions at the greenhouse was fastened among the leaves of a bunch 
of species of Pentstemon growing in open cold frames. The resulting 
natural infection was exceedingly vigorous and abundant. There were 
10 to 15 aerial sori on some of the larger leaves, while every leaf 
(30) which reached the susceptible period during the viability of the 
teliospores was infected. This proves conclusively that in nature the 
ordinary rust on Viola (Puccinia ellisiana) is able to infect Pentstemon 
under the conditions normally obtaining in the field, and that all that 
is necessary is to have Pentstemon plants intermixed with stools of 
Andropogon which are infected with Puccinia ellisiana . 
INCUBATION PERIODS 
Table V shows a very variable incubation period for each rust. This 
variability was to be expected, as it is well known to those mycologists 
who have done much culture work with rusts that the incubation period 
varies materially with the environment and host. It increases in length 
of time as the temperature rises in the greenhouses, and finally infection 
may cease entirely with extreme heat. But after making due allowances 
for these factors some interesting facts are seen when each year's cultures 
are compared. For instance, the incubation period of Puccinia ellisiana 
on violets for 1913 ranged from 13 to 25 days, with an average time of 
18 days, while the same rust on Pentstemon in 1913 ranged from 15 to 18 
days, with an average of 17 days. This shows a much greater variation 
in the range of the serial incubation stage of Puccinia ellisiana when on 
Viola than on Pentstemon, but nearly the same general average. The 
greater variation in range on species of Viola is probably due to the fact 
that several species of Viola were used in the inoculation experiments, 
while only one species of Pentstemon was infected, since the species of 
Viola used often seems to influence to a limited extent the incubation 
period. In the cultures of 1914, however, there is about the same amount 
of variation for Puccinia ellisiana on Pentstemon as on Viola. (See 
Table V.) 
