320 Journal of Agricultural Research voi. xii. no. 6 



The fine sandy loams had an even more varied past history. Three 

 of these soils were from southern California. One of these was from a 

 Redlands orange grove which had been treated in the past with a great 

 variety of fertilizers. A second was from a peach orchard of the San 

 Fernando Valley, and the third from the Riverside area had been de- 

 voted to field crops. 



Three fine sandy loams were obtained in the San Joaquin Valley. Two 

 from the vicinity of Kearney Park, in oranges and Sultanina (Thomp- 

 son Seedless) grapes, had an excellent record for past production. 

 The third from the foothills near Oakdale was purposely chosen from a 

 body of Arnold fine sandy loam where the growth of oats was unusually 

 small, though the soil was normal in depth and drainage. The last two fine 

 sandy loams were virgin soils. One of these (No. 13) came from the coast 

 of Mendocino County, and was an extremely poor, shallow soil. In 

 fact, it later developed that this soil was very acid; therefore no de- 

 ductions have been drawn from the results obtained with it. The other 

 virgin sample was obtained from the Honey Lake area. This soil was 

 considered to be rather in the class of desert soil, but was not especially 

 deficient in organic matter. 



The soils of both types were selected to include as many past treat- 

 ments and crops as possible. They had, of course, been exposed to very 

 different climatic influences. This is an unavoidable factor in Cali- 

 fornia, with its great diversity of local conditions. It is believed that 

 these various influences have been largely neutralized by the cropping 

 of the first season, 191 5, under uniform conditions. Greater emphasis 

 is therefore attached to the results of the past year (1916). 



The attempt was also made to choose soils which, though of uniform 

 physical texture, would have a very different crop-producing power. 

 In this way it was hoped that some light might be thrown on the relation 

 between yield and water-soluble nutrients. 



In Table II are given the moisture equivalents, hygroscopic coeffi- 

 cients, specific heat, and mechanical analyses of the 14 soils. Moisture 

 equivalents are obtained by the method of Briggs and Shantz (<?); 

 hygroscopic coefficients by the method of Hilgard (26). Specific heat 

 was determined by heating 20 gm. of soil in a tin-foil container until it 

 reached a temperature of approximately 100° C. It was then lowered 

 into an insulated tank of water of known temperature and volume. The 

 amount of heat contained in the soil was measured by the increase of 

 temperature of the water. The mechanical analyses were performed 

 by the method of the Bureau of Soils (9). It should be noted that the 

 analyses as reported were performed on the same samples which were 

 used for the chemical analyses. These samples had been passed through 

 a i-mm. sieve. This was done in order that the chemical and physical 

 analyses should be made on identical portions. The percentage of gravel 

 by this screening process is also stated, so that the mechanical analyses 



