33 



ing on field B. For the years 1902 and 1905, where the experiments 

 were made upon the same kind of land, the average yield on field A 

 was 13.65 tons per acre and on field B 21.83 tons per acre. The 

 natural fertility of field B apparently caused the fertilizer to produce 

 a larger effect than when applied to field A, w r hich was poor land. 

 The average percentage of sucrose was again 1 per cent greater in 

 field A than in field B. 



A test was made with a mixture of 720 pounds of acid phosphate 

 and 240 pounds of sodium nitrate. The comparison is made on a 

 yield of three years. For field A the average was 17.67 tons per acre 

 and for field B 12.52 tons per acre. The cane grown on field A was 

 one-half per cent richer in sucrose than that grown on field B. The 

 yield on field A was greater and the yield on field B less than in the 

 case where acid phosphate and potassium chlorid were used. 



A like comparison was made with 720 pounds of acid phosphate 

 and 480 pounds of cotton-seed meal. The average for the three 

 years on field A was 15.17 tons per acre and on field B 12.45 tons per 

 acre. The percentage of sucrose in the canes of field A was 1 per cent 

 greater than on field B. These data show no very striking differ- 

 ence between the action of cotton-seed meal and sodium nitrate. 



The use of 720 pounds of acid phosphate and 480 pounds of kainit 

 resulted in a slightly greater tonnage than in the case of cotton-seed 

 meal i. 6^16.42 tons and 14.12 tons on fields A and B, respectively. 

 The content of sucrose obtained from field B was higher. 



A comparison was also made with a mixture of sodium nitrate and 

 potassium chlorid without any addition of acid phosphate. The 

 average yield on field A was 16.06 tons per acre and on field B 14.01 

 jtons per acre. These data show that the use of 720 pounds of acid 

 phosphate in connection with the other fertilizers produced no greater 

 yield than the use of a mixture of sodium nitrate and potassium chlo- 

 i id alone. There is a marked difference, however, when a very large 

 {uaiitity of acid phosphate is used in connection with potassium chlo- 

 rid and sodium nitrate. In this case it is seen that the average yield 

 m iield A was 22.70 tons per acre and on field B 20.49 tons per acre. 

 The average percentage of sucrose in field B is more than 1 per cent 

 greater than in field A. When the acid phosphate in the application 

 vvas reduced by one-half, and the potassium chlorid was doubled, as in 

 he next comparison, the average yield on field A was 19.83 tons per 

 tcre and on field B 19.55 tons per acre. Again the percentage of 

 lie-rose in field B was 1 per cent greater than in field A. 



These data show that the increase of the acid phosphate to 1,440 

 )ounds per acre gave 2.87 tons more per acre in field A and 0.94 ton 

 >er acre more in field B. When the quantity of potassium chlorid 

 ised in the last experiment was diminished by one-half and the quan- 



