486 EXPERIMENT STATION RECORD. 



riiilly affecting the nitrogen content of the soil, provided clover was retained. 

 Rotation maintained the carbon and hnnnis content of the soil and in some 

 cases slightly increased it. The substitution of timothy and other nonlegumi- 

 uous plants for clover caused a loss of nitrogen, but the carbon and humus con- 

 tent was kept up. The conservation of the humus prevented rapid nitrification 

 and the loss of soil nitrogen only slightly exceeded that removed by the crops. 



The i)otash compounds of the soil were less soluble at the close of the 10-year 

 period than at the beginning. The formation of soluble potash compounds did 

 not keej> pace with the amount removed as plant food and that lost in the drain 

 waters. 



Report of the work at the Holly Springs Station for 1907, C. T. Ames 

 {Missis.'fipiri 8ta. Bui. 110, pp. 7). — This bulletin reports briefly the results of 

 fertilizer experiments on cotton, corn, and cowpeas, with notes on grasses and 

 forage crops. 



The work of the San Antonio experiment farm in 1907, F. B. Headley and 

 S. H. Hastings {U. S. Dept. Agr., Bur. Plant Indus. Virc. 13, pp. 3-16).— This 

 circular contains a description of the farm and the lines of work undertaken. 



In experiments with cotton and corn in 1907, 4 plats plowed and subsoiled 

 to a depth of 12 in. or more yielded more in every case than an equal number 

 of plats not subsoiled. In October, 1907, part of a field of sorghum v.'as plowed 

 and the other part left unplowed until December. In the meantime, 7 in. of 

 rain fell and soil moisture determinations made in December showed 24.9 per 

 cent of water in the first 3 ft. of the land plowed in October, as compared with 

 22.6 per cent for the land plowed in December, or a difference equivalent to 1.1 

 in. of rain. 



It is pointed out that on the heavy soil of San Antonio a light rain of 0.1 to 

 0.4 in. may reduce rather than increase the total amount of soil water. Soil 

 moisture determinations were made in an orchard and a cornfield cultivated 

 on June 5 and 6, 1907, respectively, to illustrate the effect of early tillage after 

 rain. On May 29, 1.2 in, of rain fell and 7 days after, in the absence of a dust 

 mulch, the moisture content of the first 3 ft. in the orchard was 0.6 per cent 

 less than it had been before the rain. In the near-by cornfield, which was culti- 

 vated 2 days earlier, an increase of 0.6 per cent of moisture over what was in 

 the soil May 28 was determined. 



Soil moisture determinations were also made on ^Slarch 12, 1908, on two fields, 

 one of which was fall plowed and well tilled throughout the winter and the 

 other plowed in February, The upper foot of soil of the winter tilled plat re- 

 tained 3,4 per cent more moisture than the plat left without cultivation until 

 February. 



Two methods of eradicating Johnson grass were tried. In September, 1906, 

 a 6-acre Johnson grass meadow was plowed 4 to 5 in. deep with a disk plow. 

 Two acres were left fallow during the winter, cultivation was given during 

 I'ebruary, March, and April at a cost of $10.84 per acre and in May, when the 

 eradication was nearly complete, the field was sown to German millet. The 

 <tther 4 acres were sown to oats in November. The growth of Johnson grass in 

 the spring in this field was weak and but few plants headed out before harvest- 

 ing early in May. The cost of cultivation to eradicate Johnson grass after 

 plowing was only $1.12 per acre. 



In 1907 a field of sorghum sown broadcast yielded 1.68 tons per acre, while 

 the same variety drilled in rows 4 ft. apart in an adjoining plat yielded at the 

 rate of 2.9 tons of cured fodder. A plat of 1.28 acres planted in alfalfa A-arieties 

 in double-drilled rows 2 ft. apart to permit cultivation survived the severe 

 drought of the summer of 1907 and in March. 1908, yielded from the first cut- 

 ting 1,236 lbs. of cured hay from the field, equivalent to 957 lbs. per acre. Ad- 



