This paper reports the results of experiments on the rate of absorption of hygroscopic moisture by seeds of flax and wheat exposed to atmospheres of different relative humidities, the rate of drying of wet seeds, the effect of an air current on the rate of drying, and the rate of absorption of free water by wheat. The absorption and loss of moisture by standing grain is of particular significance in relation to combine harvesting. Coleman and Fellows have shown that the final moisture content of the seeds of cereals and flax varies greatly when exposed for a long period (30 days or more) to atmospheres of different relative humidities (Fig. 1). Dry flax seeds absorb hygroscopic moisture somewhat more rapidly at first than wheat seeds (Fig. 2), and both flax and wheat seeds absorb moisture much more rapidly than do alfalfa seeds. Whole flax bolls absorb moisture approximately twice as rapidly as do the seeds (Fig. 3). The rapid absorption of hygroscopic moisture by the bolls probably explains why dry flax becomes difficult to thresh after a short period of high humidity. The rate of absorption of hygroscopic moisture by dry seeds (flax, wheat, and corn) varies with the temperature, at least within a certain range of temperature (Fig. 4). The rate of absorption is approximately twice as rapid at 30 degrees C as it is at 10 degrees C (Fig. 5). At 40 degrees C the rate of absorption is about the same as at 30 degrees C for a time, but later becomes slower, probably because the high temperature affects adversely the enzyme activity of the seeds. Wheat seeds containing 21% of moisture, based on dry weight, and flax seeds containing 14% will either absorb more moisture or lose moisture, depending upon the humidity of the air to which they are exposed (Fig. 7). In an atmosphere of 90% relative humidity such seeds absorbed moisture slowly, whereas at 75% relative humidity they lost moisture slowly. According to Coleman and Fellows, wheat will come to equilibrium at 17 to 18% of moisture and flax seeds at about 11.5% in an atmosphere of 75% relative humidity. The same results were obtained by the writer. Seeds of wheat of 38% and flax of 25% moisture content still absorbed moisture in a saturated atmosphere, but lost moisture in an atmosphere of lower relative humidity (Fig. 8). The seeds lost moisture rapidly in an atmosphere of 45% relative humidity. The curves are hyperbolic in form, flattening out as the moisture content approaches equilibrium. At 75% relative humidity the rate of drying of wheat was comparatively slow below 18% of moisture. The moisture content was reduced only 1%, that is, from 18.2 to 17.2%, from the 5th to the 7th days. This moisture content of 17.2%, based on dry weight, is equivalent to 14.7% based on wet weight, and the sample, therefore, was still too wet for safe storage. It is evident, therefore, that a relative humidity lower than 75% is necessary to insure the rapid drying of wet grain. An air current of about 500 cc per minute drawn through a U-tube containing wet wheat and flax seeds approximately doubled the rate of drying during a period of 8 hours, as compared with still air in a desiccator, both of 45% relative humidity. The absorption of free water by wheat is extremely rapid as compared with the absorption of hygroscopic moisture. The absorption of water at 25 degrees C is about 20 to 45% more rapid than at 10 degrees C (Table 4). The experiments reported indicate the significance of humidity in relation to the moisture content of grain which is freely exposed to the air. The results suggest a few points of practical importance, as follows: 1. It is certain that wet grain will not dry in an atmosphere of high relative humidity. 2. A relative humidity well below 75% appears to be necessary for effective drying. 3. Movement of the air, as by a breeze or wind, increases the rate of drying, provided, however, that the relative humidity is low enough to permit drying. 4. In air of high relative hu

The data given above show that on certain soil types in Iowa increases in the yield of wheat may be expected from top dressings with sodium nitrate. From these limited data it is difficult to determine which treatment may be the more profitable. The later applications were somewhat more efficient in increasing the protein content of the grain and approximately equal in increasing the yield when compared to early treatments. However, there is more difficultly connected with applying fertilizer to wheat at the later stages of growth. A top dressing of 100 or 200 pounds per acre of sodium nitrate to winter wheat in the early spring can be easily applied and might be very profitable. The increase in the weight per bushel due to the fertilizer treatment was not sufficient to be significant on either soil type. The increase in yield produced by the fertilizer was of more significance than the increase in protein content. The returns secured from the use of superhosphate were not appreciable on either soil type.

The possibility of the prediction of the quality (protein content, weight per unit volume, quality index, etc.) of the wheat crop of the individual districts of a large geographic area must depend on the existence of significant correlations between these variables in different years. Statistical analysis of the data for average weight per hectolitre, protein content, and quality index of the Roumanian wheats tabled by Zaharia for the nine year period 1900 to 1908 shows average values of +0.3653 for weight per unit volume, +0.4652 for protein content, and +0.5447 for quality index. The inter-annual correlations for rainfall are low, averaging only +0.1767 for May and +0.0869 for June. It is significant to note that, while these correlations represent interrelationships covering a range of nine crop years, the magnitudes of the correlations are not greatly affected by the period of separation of the two variables. With inter-annual coefficients of the magnitude of those determined herein, the prediction of future quality of wheat crops should be possible within a reasonable degree of accuracy. While this paper is based on Roumanian data, the primary consideration is that of the principle involved. There can be no reasonable doubt that similar results will be found for districts of other geographic areas. The magnitude of the correlation will depend upon the extent of differentiation of the districts in soil and climatic conditions.

The relative value of broadcasting fertilizer as compared with drilling in the row with wheat was studied during the years 1926-29 at the Kansas Agricultural Experiment Station at Manhattan. Small plats 1/500 acre in area were used. These, were replicated three to four times. The plats were handled by hand. In addition to these, drill-width plats were put in with a fertilizer drill in 1928-29. Comparisons were made here also with broadcasting and with no fertilizer. Superphosphate (16% P2O5) was compared with 2-12-2 fertilizer. The mixed fertilizer was made from sodium nitrate, superphosphate, and potassium chloride. The fertilizers were applied in all cases at the rate of 175 pounds per acre. The results show that the average increase from superphosphate when applied broadcast was 9.03 bushels an acre, while the average increase where the fertilizer was applied in the row with the seed was 16.4 bushels an acre. In other words, there was almost as much difference between row application and broadcasting as between broadcasting and no treatment. On this soil superphosphate showed slightly greater increases than 2-12-2, when used at the same rate of 175 pounds per acre. Applying the fertilizer in the row with the seed gave slightly higher yields than when applied either above or below the seed. The results obtained in drill-width plats in 1929 confirmed the results obtained in the small plats 1927-29, showing that under the conditions of these experiments the use of fertilizer in the row with wheat is decidedly superior to broadcasting on yields of both grain and straw.

1. Wilting coefficients of some soils were studied using some important crops grown in western Oklahoma as indicators. 2. The wax-seal and unpervious-pot method of Briggs and Shantz was employed. 3. Soil texture showed its usual effects on retaining moisture against plant usage. 4. The formula for wilting coefficient secured by the writers was as follows: (...) which is not much different than that secured by Briggs and Shantz. 5. Crops differ somewhat in their ability to reduce soil moisture. 6. Of the small grains studied, including oats, wheat, and barley, oats reduced the moisture content to the lowest percentage. 7. Silvermine corn was the least effective of the crops studied in reducing soil moisture.