GRAIN SORGHUMS IN THE TEXAS PANHANDLE. 
41 
Table XXII shows that the conditions under which the crop is 
grown have much more effect on its composition than the source 
from which the seed is obtained. For instance, Dwarf milo grown 
at the Arhngton Experimental Farm, Rossl}^!, Va., during the five 
years from 1915 to 1919, inclusive, shows only very slight variation 
in chemical composition from seed produced the previous year at 
Arlington and at Amarillo. Results are available from crops grown 
at xVrlington from Chico seed in only three of the five years, and this 
naturally causes some variation from the averages of the crops 
grown from Arlington and Amarillo seed, but in general the composi- 
tion is practically the same. In the same way milo grown at Amarillo 
from seed from each of the three points is very similar in composi- 
tion, but is lower in water content and in ash and considerably higher 
in protein and fat than milo grown from the same seed at Arlington. 
Milo grown at Chico from seed from the three sources shows rather 
more variation than that grown at Arlington and Amarillo. The 
grain grown at Chico shows a slightly higher water content than that 
grown at Amarillo, but considerably less than that grown at Arlington. 
The ash and protein content of the Chico milo is less than that grown 
at Arlington and decidedly less than that grown at Amarillo. Milo 
grown at Chico has about the same percentage of fat as that grown 
at Amarillo and is slightly lower in fiber. 
In general, the same observations may be made with regard to 
feterita and Dawn kafir grown at the three stations. The variation 
between crops grown from the same seed at the three stations is 
greater than that between crops grown at any one of the stations 
from seed from the three sources. The moisture content of the seed 
grown at Chico is intermediate between that grown at Amarillo and 
at Arlington. The Amarillo seed is materially higher in protein in 
each case, and is also higher in fiber. The variations in ash and fat 
are not marked. 
SUMMARY. 
The data on the date of seeding, spacing, and environmental 
experiments with grain sorghums, as presented in this bulletin, may 
be summarized briefly as follows: 
(1) The yields are influenced by seasonal conditions to such an 
extent that no one date of seeding is best for all years. The average 
yield in a series of years is the one safe basis for practice. 
(2) All the varieties did not give the highest average yield from the 
same date of seeding. Some yield better from early seeding than 
others. 
(3) Dwarf milo produced the best average yields from sowing on 
the normal date, about May 23; Dawn kafir from the early date, 
May 10; and feterita and Manchu kaoliang from the late date, about 
June 10. 
