248 Journal of Agricultural Research voi. iii. No. 3 



in the percentage oil content . The different varieties of soy beans did 

 not respond alike to changes in seasonal conditions. 



In tests with several varieties of soy beans grown under a very wide 

 range of conditions there were found differences of more than 100 per 

 cent in the size of the beans and very large differences in oil con- 

 tent. Here, again, the different varieties were not affected alike by 

 changes in the environment. It was not practicable to grow cotton 

 under such diverse conditions, but the difference in oil content of the 

 seed as grown in the Coastal Plain and the Piedmont regions of the 

 South was greater than the varietal differences when grown in the 

 same environment. All varieties respond very much alike to changes 

 in the environment. 



Because of the interdependence of soil and climate with respect to 

 temperature and water supply it is difficult or impossible to develop 

 far-reaching generalizations as to the specific effects of either inde- 

 pendently of the other on plant development. Six Upland varieties 

 of cotton were grown three consecutive years on adjoining but contrasted 

 soil types in northern Georgia. Each year the clay soil gave heavier seed 

 than the sandy loam, but the relative oil content on the two soil types 

 varied from year to year. In experiments with several varieties of soy 

 beans only small differences were obtained in the size and oil content of 

 the seed grown on these two soil types. Similar results were obtained 

 with peanuts. Field experiments with soy beans and peanuts on sharply 

 contrasted soil types at Arlington Experiment Farm, Va., and vicinity 

 gave more decided differences in size and oil content of the seed. A 

 number of tests with soy beans, peanuts, and sunflower were carried 

 out also on different soil types under controlled conditions, using for 

 the purpose large earthen pots set into the soil. The various tests were 

 carried out under a wide range of soil types and climatic conditions, and 

 the results as a whole emphasize the fact that the relative effects of 

 different soil types are not specific and constant, but depend largely 

 on seasonal conditions. 



From the data in hand it is concluded that under practical conditions 

 climate is a more potent factor than soil type in controlling the size 

 of the seed and its oil content, probably because those conditions of the 

 atmosphere which constitute the climate largely control the corres- 

 ponding conditions of the soil. 



Within ordinary limits the relative fertility of the soil appears to be 

 a minor factor in influencing the size of the seed and its oil content. In 

 fertilizer tests with cotton the addition of a complete fertilizer to an 

 unproductive soil gave larger seed and a considerably higher percentage 

 of oil. Applications of nitrogen in increasing quantities did not affect 

 the size of the seed, but lowered the percentage of oil, while increasing 

 applications of phosphorus or potassium did not affect either character. 



