344 
Journal of Agricultural Researcn voi. xxm, no. 5 
Section 3 of the table differs from sections 1 and 2 in that the spikes of 
a single day are of different ages. It was thought that daily weather 
variations might be overcome if samples of varying ages were taken on 
each day. Spikes marked A before August 4 flowered on July *6, 
all those marked B, and those marked A after August 4, on July 
7, and those marked C, D, and E, on July 9. In the figures showing the 
distribution by days from flowering and in the various averages, spikes 
60 
jo 
62 
48 
44 
40 
06 
I" 
^ 2 e 
^24 
20 
/e 
/2 
6 
4 
O 
\ 
4 
✓ 
s 
t 
f 
1 
\ 
\ 
/ 
/ 
t 
\ 
% 
J 
/ 
/ 
» 
/ 
§ 
\ 
\ 
9 
r 
4 
9 
9 
/ 
# 
l 
9 
i 
< 
P/ 
-Q 
$ 
f 
f 
Vf 
l 
I 
t 
/ 
/ 
! 
hz , 
7 ? 
0 ' 
r 
1 
V 
p 
/ 8 3 4 6 6 7 3 6 / 0 // /2 / 3/4 / 6 /<?/?/# /9 202 / 22232426262728 . 29303/32333430 
Fig. ii.—A verage dry matter in kernels of Baku barley from flowering to maturity at Aberdeen, Idaho, 
in 1919 and 1920. 
of the same age are grouped together. In computing data for figures, 
small abnormal tip and basal kernels were omitted where it was thought 
the abnormality was misleading. The data in the table have been plotted 
as frequencies in figures 12 to 15. 
The frequencies of the dry weights (fig. 12) are not much more easily 
interpreted than the table itself. This is due to the fact that the final 
weights of the kernels on the spikes vary over a considerable range. The 
tip kernels begin to ripen about the twenty-first to twenty-third day and 
are found distributed to the left of the general mass of kernels from 
weights of 275 mgm. to 375 mgm. 
