Feb., 1927] PROGRESS OF AGRICULTURAL EXPERIMENTS 



27 



■^'ided into nine i)l()ts, each approximately 1-20 acre. Immediately after 

 the plants were set, nitrate of soda was applied to two plots at the 

 rate of 200 i)onnds per acre; to two at the rate of 300 pounds per acre; 

 and to two at the rate of 400 pounds per acre. On three jilots no fer- 

 tilizer was api)lied. 



The berries were harvested between June 25 and July 11, 1926. 

 Yields of each plot were kept separate, but the variation between plots 

 was such that no conclusions can be drawn as to the effect of nitrate 

 at the different rates of ajiplication. However, the average yield from 

 the six plots receiving nitrate is distinctly less than that from those 

 plots which did not receive the application. The vields are shown in 

 Table XII. 



Table XII — Yield of 1/20 acre strawberry plots 



The average decrease in yield on the fertilized plots is about 32 per 

 cent, but the decrease for the first half of the season is approximately 

 22 per cent and for the last half about 46 per cent. Apparently the 

 application of nitrate shortens the season. 



The season of 1926, in wliich these results were obtained, was very 

 dry, there being little rain during the first half of the season and none 

 at all during the last half. It is possil)le that under more favorable 

 conditions of moisture a different result might have been obtained. 



In the s})ring of 1926 i)lants of Howard 17 strawberries were set in 

 the permanent garden soil fertility plots in a field of sandy loam which 

 originally contained only a small amount of organic matter. For the 

 past eight years these plots have received treatments as listed in Table 

 XIII. The exiieriment, therefore, gives an opportunity to measure 

 the cumulative effect of fertilizer ap])lications on the strawberry plants. 

 Previous exjieriments with this fruit at this station have dealt with the 

 effect of application of fertilizer either in the year the plants were set 

 or the year in which they fruited, on soil previously under uniform 

 tieatment. The fertilizers in this experiment are broadcast and 

 worked into the soil before the crop is sown or i)lanted. The data are 

 given in the table. 



There were 25 rows in the limed half and 25 in the unlimed half of 

 each plot. By recording separately the data for each row it is possible 

 to compute a probable error for the result from the variation recorded 

 in the different rows under similar treatment. 



In the main, the differences between limed and unlimed halves of 

 plots are not significant when the probable errors are considered, but 

 in two instances there are large differences in mortality of mother 

 plants. In Plot 4 the per cent of mother plants surviving is consider- 

 ably higher on the limed half than on the unlimed, even taking the 

 j)robable error into consideration. However, on Plot 6 with a treat- 



