374 



the thickness of planting was constant, but the arrangement 

 of plants different, the figures at first view suggest a con- 

 tinuous increase in yield as the constant area devoted to 

 each plant approaches a perfect square in shape. However 

 the land was not perfectly uniform, as seemed the case when 

 the plots were located. There is a slight increase in fer- 

 tility towards the lower plots, which is apparent on compar- 

 ing the yields of plots 2 and 5, — plots which are practically 

 duplicates. 



Allowing for this natural advantage which their position 

 gives to the narrow rows, the yield becomes practically the 

 same for rows nearly 5 feet apart as for narrower rows less 

 thickly planted. For land of this character, high, sandy, 

 dry, and poor, 5 feet between rows is the minimum distance 

 that can be recommended, and on very poor land wide rows 

 are best. Rather wide rows are necessary to economy of 

 cultivation and to allow the planting of a row of cow peas in 

 the middle between the corn rows. 



In order to make this test as accurate as possible, our 

 usual custom of planting a row of cowpeas in each middle 

 was not followed in this experiment. As compensation for 

 this omission, crimson clover seed was sown broadcast soon 

 after the corn was gathered, and covered by using a Planet, 

 Jr., cultivator, supplied with five very small shovel points. 

 This seeding was made in September with the expectation 

 of plowing under the growth of crimson clover late in 

 March, 1897. 



VI. Cotton Seed Meal Versus Crushed Cotton Seed 



FOR Corn. 



An experiment to compare the fertilizing effect of nitrate 

 of soda and cotton seed meal and to test the effect of apply- 

 ing only half the cotton seed meal at planting time and the 

 other half later was located on a poor hill-side, having a 

 rather stiff soil. This spot suffered more from the pro- 



