EXPERIMENT STATION BULLETINS ]81 



at all (lepends upon the cliaraeter of the soil and its wetness. If the land be clay it is 

 dangerous to use the roller in the spring. If it be Avet the roller is forbidden al- 

 together. With sand the case is different. Sandy soils are too open in texture as a 

 rule to hold moisture and produce the largest yields. The use of the roller is there- 

 fore indicated. 



Experiments in oat culture conducted by the Wisconsin Station are reported in 

 the Wisconsin Annual Report for 1890. It was found that rolling increased the 

 moisture in the upper two feet of soil, but samples taken from the surface to_ below 

 a depth of two feet showed a difference in favor of the unrolled ground. This 

 difference increased with the depth to which the samples were taken. 



A larger per cent of germination was found on the rolled ground and conse- 

 quently there was a better stand of oats. The oats on the rolled land yielded 61.12 

 bushels per acre by weight while those on the um-olled land yielded 58.89 bushels 

 per acre. The oats from the rolled plot were found to have larger kernels and to 

 weigh more per bushel. 



Other Stations have done considerable work on oat culture. The following have 

 secured results favoring fall plowing : Oklahoma, North Dakota, Kansas, Minne- 

 sota, while Kansas and North Dakota have found that spring plowing sometimes 

 gives the better results. 



Kansas, Minnesota, Illinois and Oklahoma have found results in favor of rolling 

 the seed bed for oats. Ohio found no advantage on heavy clay but on a sandy 

 soil found a marked advantage. Nebraska and Kansas have found that better 

 yields are secured with the press drill than when oats are sown broadcast or with 

 an ordinary hoe drill. 



The evidence seems almost uniformly in favor of fall plowing and of compacting 

 the seed bed, when the soil is loose in texture. 



CORN CULTIVATION. 



FREQUENT AND INFREQUENT, DEEP AND SHAULOW. 



This set of experiments was designed to supplement the experiment in the 

 cultivation of corn reported in Bulletin 154, page 268. The ground upon which the 

 tests were made was located in Field No. 6, north of the drive. Two strips of land 

 were used, one lying just west of the center of the field and the other near the west 

 side, the distance between being about twenty rods. The individual strips were 

 very uniform as to soil and quite level, but there was a marked difference in the 

 character of the soil between the east strip and the west strip. The east strip was a 

 sandy loam, inclined to be gravelly below a depth of one foot. The west strip 

 was not quite so sandy at the surface and had more clay below the first foot. In 

 taking soil samples the solid clay was usually reached in the third foot. No 

 mechanical analysis has been made of samples taken directly from these strips, 

 but in 1897 samples were taken across the field in two parallel lines, one .just west 

 of the east strip and the other just east of the west strip. The samples from each 

 line, five in number, were quite tmiform and were mixed to form a composite 

 sample. The results of the analyses of the air dried soil are given below: 



MECHANICAL, COMPOSITION OF SOIL FROM THE WEST HALF OF NO. 6. 



Organic Fine Coarse Medium Fine Vei"y fine Silt. Fine silt 

 East strip. Water, matter, gravel. suncl. sand. sand. sand. and clay. 



Soil 54 4.00 7.34 9.99 17.84 29.78 8.67 8.63 14.52 



Subsoil 14 1.68 9.79 11.92 16.03 31.85 9.02 5.63 13.66 



West strip. 



Soil 44 4.10 4.55 6.66 12.07 31.03 13.68 10.29 16.08 



Subsoil IS 1.82 6.25 8.41 15.12 33.93 14.63 6.75 13.22 



The above analyses show fairly well the difference in soil between the two strips, 

 except that in the third foot from which no sample for analysis was taken, th*^ 

 difference was most marked, there being a much larger per cent of clay in the west 

 strip than in the east strip at this depth. The two strips, each containing four plots, 

 were six hundred and two feet long. Each plot contained five rows of corn and 

 was eighteen and three-fourths feet wide. The object of having two strips was to 

 duplicate the work as far as possible in every detail and this plan was adhered to 



