12 BULLETIN 1260, U. S. DEPARTMENT OF AGRICULTURE. 
In most cases the sorghums were surface-planted on plowed ground 
with an ordinary 2-row corn planter in rows 33 to 44 inches apart . 
At Hays, Chillicothe, and Amarillo. large irregularities have been 
avoided bv thick seeding and by thinning the plants to the desired 
stand. The close -drilled plats were ordinarily seeded with a grain 
drill. 
HARVESTING METHODS. 
The row plats are harvested with a corn binder by first cutting 
out the guard rows along the roadways and sides of the individual 
plats. It is not wise to delay harvesting until every plant has be- 
come fully mature, because the plants that ripen earliest will by 
that time have lost leaves and in some varieties much of their seed 
either through shattering or the attacks of birds. When the imma- 
ture heads remain on the stalk there is usually sufficient sap left to 
complete the maturity of the seed. 
In forage-crop experiments it has been found that accurate yields 
of both fodder and seed may be obtained from the same plat if this 
plat is harvested when 90 per cent of the plants are mature. The 
fodder weights are practically the same whether the sorghum is har- 
vested at this time or at a somewhat earlier stage of maturity. The 
practice, therefore, has been to harvest at this stage and record the 
air-dry weight of the entire crop as the fodder yield. Later th<> 
heads are cut from the stalks and threshed to obtain the yield of 
seed. 
METHODS OF OBTAINING DATA. 
The row space per plant and stalk is obtained by an actual count 
of the plants and also the stalks on the plat or in every alternate 
row of the plat. r 
The length of the growing season is calculated as the time elapsing 
from the date the sorghum was seeded until 90 per cent of the plant- 
had ripened seed. 
The average height of the sorghum is determined at maturity by 
selecting several plants of average height by an ocular survey of the 
plat and then measuring these plants. The original method used 
was to measure 10 plants at different positions in the plat and take 
the average height. It was found, however, that the result was 
likely to be in error when this method was used, because while the 
operator is moving about within the plat he can not observe whether 
he is measuring undersized, average, or oversized plants in greater 
number. 
At Hays, Kans., Chillicothe, Tex., and Amarillo, Tex., the experi- 
ments were conducted by a member of the staff of the Office of 
Forage-Crop Investigations, and all forage weights were reduced to 
an air-dry basis. The field weights were corrected for moisture 
content by the sample method described in Bulletin 353 of the 
United States Department of Agriculture (/J). 4 
4 The need of correcting field weights by the sample method is especially acute in experiments with 
sorghums, because variel ies with large juicy stems cure so much more slowly than those With small pith} 
stalks. Several months of favorable curing weather are not sufficient to reduce all kinds of sorghum fodder 
to a uniform moisture content when it is standing in ^hoeks. The relatively high Yields of field-cured 
forage credited to late-maturing varieties of sorghums in comparison with early-maturing sorts are often 
due to their higher percentage of moisture, it is impossible to avoid such inequalities, except by reducing 
the moisture content in all varieties to a uniform percentage by means of samples. Short -season variel ies 
harvested in the latter pari of August or early in September may stand in shock for three months, the 
first half of which is usually very warm and dry. Late varieties, on the other hand, are usually not har- 
vested until the last of September or middle of October; often not until the first frost . Their curing period, 
besides being much shorter, is marked by lower temperatures and often greater humidity. Obviousl3 
these large, coarse-stemmed late sorghums can not be expected to reach the same degree of dryrn 
the smaller, finer-stemmed early varieties. 
