May 19,1923 
Statistical Study of Biologic Forms 
555 
This characteristic variation is brought out more clearly by an analysis 
of the differences in the means than by a mere consideration of the con¬ 
stants. The means of this biologic form are 40.30i0.40X i4.64±o.i2/x. 
The difference in the means between the agrostis and the tritici forms is 
significant for both dimensions. Between agrostis and secalis, however, 
the difference in the means is significant for length only. Again, between 
the agrostis and avenae forms, the difference is significant for both 
length and width. The difference in the length of teliospore means of 
the agrostis and phleipratensis forms has no value as it is only 1.96 times 
greater than its probable error. The difference in the widtJi, however, 
appears to be very significant because it exceeds its probable error by 
6.18 times. 
EFFECT OF ENVIRONMENTAE CONDITIONS ON THE MORPHOEOGY OF 
RUST SPORES 
It has been reported {14) that resistant host plants and other unfavor¬ 
able environmental conditions affecting the normal development and 
vigor of the rust fungus tend to change appreciably the size of its uredinio- 
spores. However, when normal conditions are reestablished, the uredin- 
iospores of the following generation are of the original dimensions. 
This experiment was elaborated and extended to the aeciospores and 
teliospores. Curves for the distribution of the spores studied were 
plotted into classes; constants were calculated; and differences in the 
means of varying conditions established. Tables IX and X represent 
the spore-size variations for length and width, respectively, as well as 
the constants together with their probable errors, for the several spore 
types as affected by different host plants and varying physical factors. 
The classes in these tables differ in Ube same order as those in Tables III 
and IV. The curves in figures 7 to 12 represent the distribution into 
their respective classes of the spores of the various biologic forms, grown 
under different conditions. The probable errors of the differences in 
the means of the spore measurements, together with the ratios obtained 
from a division of the differences by their probable errors, and the differ¬ 
ences in the means themselves are all given in Tables XI and XII. 
HOST PLANTS 
Species and varieties which are congenial to a certain biologic form of 
P, graminis can not, under favorable cultural conditions, exert any 
perceptible influence on the spore morphology of that biologic form. 
However, even under identically the same environmental conditions, the 
spores of any biologic form will be reduced in size when cultured on 
resistant varieties. This is substantiated by measurements made of 
spores grown simultaneously on susceptible and resistant hosts. 
SUSCEPTIBLE HOSTS 
Aeciospores of P. graminis secalis were produced on Berberis vulgaris 
and P. sieboldii Miq., two equally susceptible species. The inoculations 
were made with teliospores of the same origin. The difference in the size 
of the two sets of aeciospores was practically negligible and highly 
insignificant (fig. 7 and 8). The means of the spores on B. vulgaris 
were 17.10 ± 0.19X 13.46 ± o.ii iij as compared with i7.44±o.i5X 
13*30 ±0.09 n for the means on B. sieboldii, 
39364—23 - 2 
