4 BULLETIN 1018, U. S. DEPARTMENT OF AGRICULTURE. 
Beginning on April 22, soil samples for moisture determinations 
were taken each week in foot sections to a depth of 6 feet. 
Immediately after thinning, 20 normal plants were selected in each 
plat and designated as observation plants. To minimize the error 
which might result from soil variations, 5 plants were selected 29 
feet apart in each alternate row, excluding the outside rows, thus 
locating the 20 plants on four -alternate inside rows, so that each 
represented an area of approximately one two-hundredths of an acre. 
Measurements of the growth of the main stem of these plants were 
recorded each week on the same day that moisture samples were 
taken. This was accomplished by measuring from an india-ink 
mark near the ground to the tip of the terminal bud. In planning 
the experiments it was decided that a detailed study of a few indi- 
vidual plants selected in the above-described manner and produced 
under different conditions of soil moisture and available food supply 
would be of greater service in providing a better knowledge of the 
water requirements of the crop than a cursory inspection of a large 
number of plants studied collectively. 
Beginning with the appearance of the initial flower about June 
30, the flowers on each of the selected record plants were marked with 
the date of opening, in order to determine the length of the develop- 
ment period of the boll and to obtain information in regard to 
shedding. 
ADAPTATION AND LENGTH OF THE DEVELOPMENT PERIOD. 
The Pima variety, which is the only kind of cotton now grown in 
the Salt Kh-er Valley, seems in many ways admirably adapted to'the 
climate of that district. Thus far in the history of the industry there 
have been no very destructive insect pests or diseases to affect the 
yield on a large scale, and storm damages have been infrequent. 
Yields of a bale and more per acre have not been uncommon, yet the 
average yield for the whole producing area for several years has 
been less than one-half bale per acre. Some of the growers would 
attribute this, low average to the comparatively long development 
period required by the Pima Egyptian crop. While other factors, 
suph as the location of a part of the crop on poorly adapted soils 
and in the hands of inexperienced growers, undoubtedly are largely 
responsible for the low average yields, yet the length of season re- 
quired is of no little importance in the greater part of the Salt Eiver 
Valley, where comparatively early frosts have been responsible for 
serious damage to the late crop of bolls. This has been the case 
especially where heavy soils or delayed preparation made planting 
impossible before the middle of April. It has been estimated by some 
of the growers that the heavy frost of November 1, 1919, was re- 
sponsible for destroying from 15 to 20 per cent of the bolls on about 
