796 
Journal of Agricultural Research 
Vol. XXIII, No. io 
Plate 8, B, shows the November 18 sowing as it appeared on May 21, 
1920. While very little tillering took place on the plants in this plot, 
a very good quality of wheat was produced at the rate of 22 bushels 
per acre, with a bushel weight of 60.75 pounds. From the agronomic 
standpoint, this is remarkable for winter wheat sown on such a late 
date. For spring-sown winter wheat no such result is obtained. The 
resulting plants tiller vigorously, without shooting or heading, and pro¬ 
duce no crop. The physiological basis for the behavior of the November 
18 sowing affords a very interesting problem. Doubtless the effect of 
soil temperatures in relation to the stages of plant development is 
largely responsible. 
The date-of-seeding experiment was repeated in the fall of 1920, 
using Harvest Queen (white-chaffed Red Cross) seed. The plots were 
the same width as those used the previous year, but they were only 2 
rods long. Sowings were made on September 21, October 4 and ii # 
and November 4 and 19. 
Table X. —Influence of seeding date upon the percentage of rosette disease developed in 
Harvest Queen (white-chaffed Red Cross) wheat sown in plots 54 inches by 2 rods on five 
different dates in IQ20 
Date of seeding. 
Sept. 21. 
Oct. 4. 
Oct. 11. 
Nov. 4. 
Nov. 19. 
Percentage of disease. 
93-6 
85.6 
96. 0 
36. 6 
39-4 
In connection with this date-of-seeding series it should be noted that 
owing to the mild winter the last two sowings made in November, 1920, 
emerged between Christmas and New Year's and a good stand resulted 
in the early spring. 
The results from this series are shown in Table X. While a considerable 
percentage of rosette disease developed in the two latest sowings, it 
should be noted that the symptoms differed considerably from those mani¬ 
fested by plants in the early sown plots where normal fall emergence 
took place. The disease did not appear until fully six weeks after it had 
developed in the early sown plots. At this time the plants were from 12 
to 16 inches in height. The first indications consisted in a retarding of 
certain plants followed by the dark blue-green coloration found in typical 
plants. Only slight excessive tillering resulted, and but few diseased 
plants developed normal heads. These same modifications in symptoms 
take place under conditions of reduced temperature in the greenhouse. 
Just why a greater percentage of disease developed in the last two sow¬ 
ings made in 1920 than in the last sowing made the year previous can not 
be explained satisfactorily at this time unless it is for the reason that a 
steadier growth followed emergence in the sowing made on November 18, 
1919, than with the last two sowings made in 1920. The latter sowings 
were subjected to a rather limited dormant period after emergence, thus 
simulating more nearly the conditions surrounding early fall-sown wheat. 
It will be noted from Table X that there was an increase in the per¬ 
centage of disease which developed in the sowing of October 11, 1920, 
compared with the two earlier sowings, thus causing a decided rise in the 
curve. It will be noted also from Table IX that there was a constant 
decrease in the percentage of disease in the 1919 plots from successively 
