Dec. IS, 1914 Oil Content of Seeds and Nutrition of Plant 245 



Experiments similar to those with soy beans were made with the pea- 

 nut of the variety known as Spanish, and the results are given in Table 

 XI. The effects produced by the different soil types are of the same 

 general character as with soy beans, although the behavior of the two 

 speciesjander similar conditions is not always the same. In 1913 a series 

 of pot cultures with the sunflower were carried out at the Arlington 

 Farm in the same manner as described for soy beans, using a number 

 of different soils in the test. The weights in grams per 1,000 seeds as 

 grown in the Arlington clay, Norfolk sandy loam, Potomac Flats soil, 

 Norfolk sand, and the so-called Benning sand were 90.5, 73, 79.5, 

 56.4, and 52, respectively, while the corresponding percentages of oil in 

 the kernels were 51.25, 50.30, 55.70, 51.26, and 49.10. In this case soil 

 differences brought about very marked differences in size of the seed, but 

 the variations in relative oil content were less decided. 



OIL CONTENT OF SEED AS AFFECTED BY CLIMATE 



On comparing the data given in Table IX with those in Table XI it 

 becomes apparent that the variations in the size of seed and the oil con- 

 tent of soy beans attributable to differences in soil type are far less than 

 those observed when both soil and climate differ. The same relation- 

 ships are observed in cotton seed, as shown in Tables VIII and X. These 

 results are interpreted as indicating that under practical conditions 

 climate is a more potent factor than the soil in modifying the size of seed 

 and its oil content. The most probable explanation is that the atmos- 

 phere is subject to greater and more rapid variations in moisture and 

 particularly in temperature, and also that the "soil climate" is greatly 

 influenced by the weather conditions. Temperature and moisture differ- 

 ences of both soil and atmosphere are among the important factors of 

 environment which may influence the plant characters under study, and 

 this factor-complex must be at least partially analyzed before satisfactory 

 conclusions can be reached as to the principal external factors concerned 

 in oil formation in the plant. 



OIL CONTENT OF SEED AS AFFECTED BY FERTILIZERS ' 



The experiments with different soil types previously described have 

 included soils varying greatlj'^ in fertility, as indicated by the comparative 

 growth of the plants shown in Table XII, and the results as a whole 

 show that within the limits ordinarily met with in farm practice the 

 relative fertility of the soil does not very greatly influence the size of 

 the seed or its oil content. A large number of fertiUzer tests with cotton 

 were carried out at Lamar and Tinnnonsvillc, S. C, in 1909, 1910, and 

 191 1, to obtain more accurate infonnation as to the effects of fertilizers 

 on the size and oil content of the seed. The data are too voluminous 

 to present in detail, but in Table XII a summary of the results for 191 1 



