232 
AMERICAN JOURNAL OF BOTANY 
[Vol. 9 
attained by plants receiving the same treatment, and that the larger plants 
were less susceptible to rust infection — increased resistance being par- 
ticularly marked in the sub-group including the largest of the plants. 
The seed emplo3^ed was a commercial ''Swedish Select" oats in which 
we should expect a mixture of strains as regards rate of growth, speed of 
maturity, and susceptibility to rust. In view of the uncertainty as to 
the varietal purity of the seed employed, the differences in incidence of 
rust infection on plants receiving the same treatment and showing differences 
in vigor of growth are probably indicative of constitutional differences in 
susceptibility to rust which may be correlated with similar constitutional 
differences in speed of growth ; and so may be considered as not necessarily 
bearing on the main problem I am considering, which is concerned with 
the effect on rust susceptibility of externally induced variations in the 
vegetative vigor of the host. The establishment of an inverse relation 
between susceptibility to rust and speed of growth in oat varieties would, 
however, lend new significance to the practical injunction of the agrono- 
mists to plant early-maturing varieties of oats in order to escape loss from 
rust, indicating that selection of rapidly growing and early-maturing strains 
of oats automatically implies selection for rust resistance as well. 
The Possibility of a Direct Relation between Environmental 
Conditions and Rust Resistance 
Groups D, £, and F in experiments IV and V were intended as tests for 
a possible direct effect on rust susceptibility of specific nutrient substances — 
that is, an effect independent of variations in the health and vigor of the 
host plant. A potash fertilizer was applied to the plants of group D\ a 
phosphate fertilizer to those of group E\ and the plants of group F were 
richly fed with a nitrogen salt. 
The infection observed in these groups is in no instance so far different 
from that on plants of similar weight in the groups not treated with any 
special fertilizer as to justify the inference that the fertihzing chemicals 
were exerting any influence on the rust resistance of the host other than is 
implied in their effect on the general condition and vigor of the plant. 
In experiment IV the potash and phosphate applications proved ex- 
cessive, and the growth of the plants was appreciably retarded as compared 
with the plants of group C; in experiment V the potash and phosphate 
fertilizers had no effect on the growth of the plants. In both experiments 
the potash- and phosphate-fertilized plants show a somewhat higher in- 
cidence of infection than plants of similar weight not treated with special 
fertilizers; a tendency at variance with the statements of Bolley (1889, p. 18) 
and Spinks (1913, p. 247) that these fertilizers give increased rust resistance. 
In group F the stimulating action of the nitrate fertilizers on the growth of 
the host was so marked that there can be no hesitation in referring the 
increased susceptibility observed to this effect rather than to any direct 
