40 MASS. EXPERIMENT STATION BULLETIN 196. 



is near the foot of the west side of a drumhn of moderate elevation. Ac- 

 cording to a soil survey made by the United States Department of Agri- 

 culture the soil is of that type for which the name Holyoke Stony Loam 

 was suggested. It is a type of soil which with minor variations is very 

 common througliout Massachusetts, and is known in the ordinary 

 language of the farm as a moderately strong gravelly loam, with good 

 capacity for retaining and conducting moisture, and somewhat affected, 

 more or less unevenly as is pointed out in another connection, by seepage 

 water which tends toward the surface from the higher portions of the 

 drumhn as it finds its way downwards. 



The land referred to had been used in an orchard experiment designed 

 to test the relative results of different systems of manuring continuously 

 followed from the year previous to the setting of the nursery stock. The 

 kinds of fruit which had been used in the orchard experiment were 

 peaches and pears. The land used was divided into five equal plots, each 

 of which was uniformly manured annually from 1889 (the year previous 

 to the setting of the trees) until 1897, both inclusive. As already in- 

 dicated, the trees were set in 1890. The land proved unsuited to the 

 peach and pear, a number of the trees died quite early in the experiment, 

 and, since the number lost in different plots differed widely, it was decided 

 to be inexpedient to continue the experiment with these kinds of fruit. 

 Both manure and fertilizers used while the land was in fruit were ap- 

 pUed broadcast in the early spring. 



Plan of the Experiment. 



The statements concerning the history of the area used in the experi- 

 ment to be described have made it apparent that we had available five 

 plots lying side by side upon a fairly even slope which had been respec- 

 tively subjected to a widely varying fertilizer treatment. It is at once 

 apparent that results on these plots could not be compared one with 

 the other in such a way as to throw any light upon the question of the 

 relative effects of different methods of appl}ang manure. Accordingly 

 each of the original orchard plots was divided in the middle by a line 

 running directly up and down the slope. The different original plots 

 were separated by unmanured or fertilized strips of land 14 feet wide, 

 while the two halves of each of the original plots as laid out for the ex- 

 periment under discussion were separated by a strip 7 feet in width. For 

 purposes of record it was decided to retain the original orchard plot 

 numbers, and to make the apphcation of manure to the north half of 

 each plot in winter, that to the south half in spring. The system followed 

 to insure as closely as possible absolute equality in amount and kind of 

 manure applied to north and south plots was as follows : — 



All the manure used on plots 1 to 4, both north and south halves, was 

 the product of the experiment station herd of milch cows. These cows 

 were liberally bedded with baled planer shavings. The floor upon which 

 the cows were kept was watertight, and the manure was removed at 



