548 
Journal of Agricultural Research 
Vol. XXIV, No. 7 
P. GRAMINIS TRITICI 
The constants, together with their probable errors for the aeciospores 
of this biologic form, as given in the first section of Tables III and IV, 
respectively, show this form to have the largest arithmetical mean for 
both length and width, namely, 19.72io.19x 15.66i0.10 /4. 
P. GRAMINIS SECALIS 
The aeciospores of this biologic form differ considerably in both dimen¬ 
sions from those of the tritici form, the means being 17.10d1O.19 and 
13*46 ±0.11 /I, respectively. 
-/3/7-/C/ 
s/iosr. 
/ 
' - M' 
-»</■ 
fG - M 
\ 
/4.70±J0 
J 
\ 
\ 
/ 
« 
/ 
\ ■ "7 
\ / 
1 / 
W 
/ • 
/ 
// 
/ 
-4 _ 
\ 
- - J 
1 
( ! 
\ 
• 
\ 
t 
1 
© 
/ 
} 
/ 
// 
# 
/ 
/ 
I 
7 
i 
/ 
/ 
$ 
1 
-r 
% 
\ 
\ 
\ 
V 
\ 
/ 
y 
/ 
/ 1 
> /a 
r 
f X 
\ 
— 
\ 
N 
—\— 
\ 
> 
\ 
X 
An examination of Table V 
will show that the difference 
in the means of these two 
biologic forms is, without 
doubt, significant for length 
as well as for width. The 
difference in the means of 
length divided by the probable 
error of this difference is 9.71, 
while the difference in the 
means of the width is 14.66 
times greater than its probable 
error. The odds against the 
occurrence of such differences 
in random sampling are beyond 
comprehension. 
P. GRAMINIS A VENAE 
This biologic form differs 
quite markedly from either of 
•lu. 2.—i^inerences m wiams or aeciospores or Dioiogic 
forms of Puccinia graminis grown on congenial host the aOOVe, although tO a SOme- 
plants and under uniform cultural conditions. whatlesser extent. The differ¬ 
ences, nevertheless, are very significant, as can readily be seen from 
Table V. The means of the aeciospores of this form are i8.62d:0.iiX- 
i4.7OiO.IO IX. 
P. GRAMINIS PHLEIPRATENSIS 
Fig. 2.—Differences in widths of aeciospores of biologic 
Any attempt to produce aecial infection on different species and 
varieties of barberry with teliospores of this biologic form has invariably 
resulted in failure. A study of the aecial stage of this form therefore 
could not be made. 
P. GRAMINIS AGROSTIS 
As shown in Tables III and IV, this biologic form has aeciospores which 
are smaller in size than those of any of the other biologic forms. The 
means are 16.46±0.18X 12.98±0.09 ju. The differences in the means 
between this form and tritici on the one hand, and avenae on the other 
hand, as indicated in Table V, are certainly significant. The difference 
between agrostis and secalis may or may not be due to random sampling, 
for the diference in the means of spore length is only 2.46 times greater 
than its probable error and the difference in width does not exceed 3.43 
times the probable error. The odds against the normal occurrence of 
such differences are approximately 9 to i in the length differences, and 
about 45 to I in the width differences. It would be difficult therefore 
