40 a 



EXPERIMENT STATION, 



[Jan. 



In case of the rowen there are two years in which neither the 

 sulfate nor the nitrate produced as large a crop as the check 

 plot. For the entire period of the experiment the increase on 

 the nitrate plot is slightly better than that on the sulfate plot. 



The following table shows the increase in cost per acre due 

 to the addition of sulfate of ammonia and nitrate of soda to 

 the mixture : — 



Year. 



Cost per Acre. 



190S, . 



1909, . 



1910, . 



1911, . 



1912, . 



1913, . 



1914, . 

 Average.s, 



6.39 



6.13 



From the tables it will be seen that nitrate of soda has pro- 



duced the larger increase and at a lower cost. 



Lime Expeeiment. 



An experiment to study the relative value of different sources 

 of lime on the basis of equal applications of combined calcium 

 and magnesium oxides was begun this year. The field selected 

 for this experiment is the one on which for so many years we 

 studied the effects of spring and winter applications of manure. 

 The jDlots have not received any manure since 1911. 



There were five pairs of plots in the manure experiment, 

 and since there are four different kinds of lime to be tested, 

 one pair of plots is given up to each kind of lime and one pair 

 is used as a check plot, receiving no lime. No manure or fer- 

 tilizer of any kind was applied this year. 



The crop grown was medium green soy beans, which were 

 cut green and put in the silo with corn. The following table 

 gives the yields per acre of hay in 1913 before liming, and of 



