6 MASS. EXPERIMENT STATION BULLFITIN 306 



total rainfall exceeds 6 inches for any one month the beneficial effects of fertil- 

 izers, especially nitrogenous materials, are likely to be reduced. Trends in these 

 directions are indicated particularly by the results from Worcester County. 

 The total precipitation for the period May to September inclusive for the three 

 \ears under consideration was as follows: 1930, 16.21 inches; 1931, 24.62 inches; 

 1932, 14.75 inches. 



Presentation of the Results 



Worcester Experiment 



In Table 3 are the yield figures for the Worcester experiments. The mean 

 increase over the check for each treatment is given in Table 4, with a comparison 

 of four of the experiments from which the most data were secured. 



Table 3. — Yield from Worcester Experiments 



(Pounds dry matter per .-icre) 



First, it may be noted that with the exception of phosphorus each additional 

 nutrient produced some increase in yield, if the mean of all experiments for the 

 three years is considered; that is, the fertilizer effects were mainly cumulative. 

 If individual years and experiments are considered, it will be seen that there were 

 some negative effects from phosphorus. These negative differences are probably 

 within experimental error, which is admittedly high in this type of work. The 

 mean increase due to phosphorus in both combinations was 20 pounds; to potas- 

 sium 343 pounds; to nitrogen, alone and in combinations, 556 pounds; and to 

 calcium in the form of lime, 230 pounds. Certain individual differences in re- 

 sponse are discussed later. 



