PLANTING 33 



years on two plantations containing 100 trees, no reduction 

 occurred in the superiority (amounting to 50 per cent, in growth) 

 of the rammed trees. Nor did this excess of growth seem to be 

 accompanied by any loss of cropping power, even during the 

 early years after planting, for, in the plantations in question, the 

 crop produced by the rammed trees was in excess of those from 

 the unrammed ones to the extent of 22 per cent, in one case, 

 and over 100 per cent, in the other; whilst in a third series, 

 wherein records of the fruiting were kept for ten years, there was 

 an excess of 13 per cent. With gooseberries, currants and rasp- 

 berries equally good results were obtained as to fruiting; the 

 crops from the rammed bushes of the first two species showing 

 in the second season an excess of 100 per cent, over those of 

 their unrammed fellows (IX, 25). 



In these trials by independent growers the special item in 

 rough planting to which attention was drawn was the ramming 

 of the ground after the roots had been covered with the soil, 

 but the roots were not purposely illtreated by being broken or 

 unduly huddled together, since other experiments, which were 

 in progress at the farm, had already indicated that it was this 

 ramming which was responsible for the benefit, and that root- 

 injury, unless carried to an excessive extent, was an unimportant 

 factor in the case. 



In these other experiments the three main items of rough 

 planting were examined separately : 



(1) Ramming the trees forcibly into the soil, a heavy rammer 

 being used for the purpose, and the process continued till the soil 

 was puddled, and shook like a jelly. 



(2) Leaving the roots untrimmed, or injuring them abnormally 

 by hacking them with a spade. 



(3) Huddling the roots together into a small hole, with the 

 majority of them pointing downwards. 



The first two sets of experiments on these points (V, 68) did 

 not yield satisfactory results; they agreed in showing that a 

 considerable benefit resulted from ramming, but as to the other 

 points there was too great a divergence to lead to any safe con- 

 clusions, the probable indications being that neither root-injury 

 nor the huddling together of the roots had any appreciable 

 effect. A more extensive series of experiments was, there- 

 fore, initiated (IX, 38). In the first place the behaviour of 

 two small plantations was examined, one of which had been 

 planted carefully, the other carelessly, but without either pur- 

 posely injuring the tree-root, and also without ramming. The 

 D 



