PLANTING 51 



would involve too much labour in practice, but equally good 

 results might, it was thought, follow, if the same length of stem 

 were buried in the ground in a slanting direction, so that none of 

 it should be below the usual depth. This, however, did not prove 

 to be the case. A comparison throughout eight years of stocks 

 worked after they had been planted (a) at a depth of 6 inches, 

 (b) at a depth of 24 inches, and (c) at a depth of 6 inches, but 

 with 24 inches of stem buried, showed a very great advantage 

 in favour of deep planting, but not so great a one in favour of 

 shallow planting with the stem buried slantingly, though during 

 the first three years these latter trees had been somewhat ahead 

 of those with the stems buried vertically. 



Relative growth . 100 I 4 I 9 947 



In the first set of trees, it should be mentioned, the stocks 

 were budded at a height of 24 inches above the roots, so as to 

 be strictly comparable with the other sets. 



Similar experiments with pears on quince stocks gave very 

 different results, for, as had been ascertained in previous experi- 

 ments (p. 38), the burying of the roots of this stock at a great 

 depth leads to very bad results, consequently the trees with the 

 2 feet of stem buried at a depth of only 6 inches did much 

 better than those with the 2 feet buried vertically, but they 

 showed no advantage over those with only 6 inches of stem in 

 the ground. The values were 



Relative growth . 100 34 98 



WORKING STOCKS ABOVE OR BELOW GROUND LEVEL 



In connection with this experiment the question of the relative 

 advantage of working stocks above or below the ground level 

 was investigated. The results were confined to pears on the 

 quince and pear stocks, for in the case of apples, which had been 

 similarly treated, there were too many failures to admit of any 

 conclusions being drawn. The effect of the different methods 

 of working the stocks was measured by lifting and weighing 

 the trees after ten years. There were six trees of each of six 

 different varieties in each plot, making 216 trees altogether, 

 and the results were 



