May 19, 1923 
Statistical Study of Biologic Forms 
551 
Table VII, bring out not only the variations of the urediniospore dimen¬ 
sions but also the significance of these variations whenever and wher¬ 
ever they are found. 
P. GRAMINIS TRITICI 
That this biologic form has the longest urediniospores is confirmed by 
the constants, together with their probable errors, given in the second 
section of Tables III and IV. In width, the urediniospores are practically 
the same as those of the oat-rust form. The means for P. graminis 
tritici fall at 32.40i0.19X 19.79i0.06M. 
P. GRAMINIS SECAUS 
The rye-rust urediniospores are both shorter and narrower than those 
of the wheat rust as shown by the means, 27.14io.14X 17.i9io.o6M, 
there being a difference of over 5 m in length and more than 3 m in width. 
There can be no doubt whatever concerning the significance of this 
difference, as pointed out in Table VII. 
P. GRAMINIS AVlSNAie 
Although the urediniospores of this biologic form are only a little more 
than I micron (m) longer than those of the secalis form, the difference, 
nevertheless, seems to be quite significant. The means of the uredinio¬ 
spore dimensions of P. graminis avenae are 28.50i0.15X i9.94±o.o7m. 
P. GRAMINIS PHLEIPRATENSIS 
The means of this biologic form are 23.95±0.12 X 16.88 ±o.o6m. The 
differences between the means for both width and length of this biolo^c 
form and those of P. graminis tritici and P. graminis avenae are quite 
significant. The difference in the means, of length only, of the timothy 
rust and the rye rust is also noteworthy. (Table VII). 
P. GRAMINIS AGROSTIS 
The urediniospores of this biologic form are without any doubt smaller 
than those of any of the other forms. This is demonstrated by the curves 
plotted in figures 3 and 4 and confirmed by the constants in Tables III 
and IV, and also by the differences in the means as shown in Table VII. 
Themeansof theredtop-rusturediniosporesare22.37±o.i2 X 15.68i0.05M. 
The differences in the means of the urediniospores of the biologic forms 
of P. graminis and the ratios between these differences and their probable 
errors are summarized in Table VII. 
