45 



August 5 and the taro planted on August 15. To plat No. 1 

 superphosphate and sulphate of potash were added ; to No. 2, 

 nitrate of soda, superphosphate, and sulphate of potash ; to No. 

 3 no fertilizer was added ; No. 4, ammonium sulphate, superphos- 

 phate, and sulphate of potash ; and No. 5 received ammonium sul- 

 phate only. The same fertilizer was applied to plats 6 and 7 as 

 to No. 4 except that in the case of No. 6 it was applied on Sep- 

 tember 16, just before planting, while on No. 7 it was applied two 

 months after planting in order to determine if the time of appli- 

 cation would cause any great variation in results. 



"On August 29th the plants in plat No. 2 appeared to be about 

 twice as large as the check plat ; in plat 1 they were smaller, 

 while in 4 and 5 they were about the same size as in the check 

 plat. On September 12 the plants in plat 1 had assumed a yel- 

 lowish color and were still smaller than those in the check plat. 

 In plat 2 they were still larger than the check plat, but also very 

 yellow. Plats 4 and 5 at this date showed great improvement, 

 and the plants were larger than in any of the plats and of better 

 color, No. 5 appearing especially green. The final results of the 

 experiments are shown in detail in the following table : 



EFFECT OF VARIOUS FERTILIZERS ON TARO. 



Amount 



No. of per Acre. 



Plat. Fertilizer Applied. Pounds. 



^ 5 Superphosphate 450^ 



^Sulphate of potash.. 450^ 

 [Nitrate of soda 400] 



2 "I Superphosphate 450 }- 



[ Sulphate of potash . 400 J 



3 No fertilizer 



[" Ammonium sulphate 300 ] 



4 -{ Superphosphate 450 [■ 



[ Sulphate of potash . 400 J 



5 Ammonium sulphate 300 



"These experiments, while they were carried through only one 

 season, indicate the economic value to be derived from the appli- 

 cation of ammonium sulphate, superphosphate, and sulphate of 

 potash to taro. In the second column are shown the weights in 

 pounds per acre, the plats used being only one-twentieth of an 

 acre in size ; hence only one-twentieth of these weights were ap- 

 plied on each plat. The yields of taro and poi are given in 

 pounds per plat and not in acreage figures. The weight of taro 

 does not include leaves, but only the marketable root bulbs. 

 Column 5 represents figures obtained in factory. 



"It is clearly evident from these results that the effect of nitrate 

 of soda is to produce a bulb of greater weight and volimie than 

 any of the other fertilizers, but in so doing it causes a decrease in 



