Grain sorghums, because of their superior drought-resistance, largely replace corn in the Southwest where the average annual precipitation is less than 25 inches. Sorghum leaves have more stomata per unit of area and smaller stomata than corn leaves, sorghums in each case showing typical xerophytic characters. Sorghum leaves and stems have a waxy and cutinized epidermis which reduces evaporation. Sorghum plants are less leafy than corn plants and have more secondary roots. The osmotic concentration of leaf juices is lower and that of the stalk, crown, and root juices is higher in sorghums than in corn. Sorghum stalks have a lower moisture content than corn stalks. Sorghum has a lower transpiration ratio than corn under conditions of high evaporation. Sorghum leaves and stalks wilt and dry more slowly than those of corn. The slower drying permits sorghum plants to withstand drought longer than corn and they may subsequently recover when moisture becomes available. The waxy cuticle apparently is largely responsible for the slow drying of sorghums. It's believed that the sorghum stalk recovers from a dormancy induced by drought chiefly because it has not wilted beyond recovery. Sorghum plants can send up tillers and still produce a good crop if rains come after the main stalks have been killed by drought. Ear]y types of grain sorghum sometimes partly escape drought and produce some grain before the soil moisture becomes exhausted when kafir is a complete failure.

1. The percentage juice and the total sugars of juice extracted from sorghum plants in the process of crushing for syrup manufacture were determined on plants given the following treatments: (a) Cut without stripping leaves at time of crushing; (b) cut with stripping leaves at time of crushing; (c) stripped, cut, and left to lie in a pile for various intervals before crushing; and (d) unstripped and left lie in piles before crushing. 2. It is clearly shown that there is a proper time for harvesting sorghum in order to obtain high percentage of juice and of total sugar (Fig. 5). The highest total sugar was obtained from October 1 and October 4 cuttings; less from the September 25 cutting and later dates in October. 3. There is some difference in the value of internodes from different parts of the sorghum stalks, the central having the highest, the lower ones an intermediate, and the uppermost the least value (Fig. 2 and Table 1). 4. Heavy pressure during the processes of crushing the sorghum stalks usually gave the higher amounts of sugar and juice (Table 3). 5. By repeating the crushings three times and by increasing the pressure of the rollers in each case, the respective crushings yielded total sugars in ratio of 2.71:2.4:1.26, or in percentage, approximately 43:40:17. 6. Stripping of leaves from the plant three to four days previous to cutting and crushing caused a slight depression in amounts of total sugars obtained on various dates (Table 2). 7. Allowing the stalks to remain in piles several days after cutting increased the sugar content of the juice probably due to two reasons, namely, evaporation, thereby making the sap more concentrated; and hydrolyses of starch and dextrins to sugars. 8. Allowing leaves to remain on stalks in each instance reduced the percentage of juice and the total amount of sugars (Table 2).

The reduction in probable error of grain sorghum yields is not great after the plat has been increased in size to 1/25 acre. Long narrow plats seem the most reliable for varietal trials and certainly are the most practical from the standpoint of economy. The probable error is slightly lower for long narrow plats than for square plats of the same area. When protected by borders, 2- to 4-row plats 8 rods long having an area from 1/50 to 1/25 acre are desirable and convenient units. Since the average yields of grain sorghums on the field stations in the Great Plains are less than 30 bushels per acre, two to four replications of a variety should usually reduce the probable error to 1 to 3 bushels per acre when 1/50 to 1/25 acre plats are used. For dependable results, it is also necessary to grow varieties for a sufficient number of years to sample the seasons which vary greatly in the Great Plains area.