TRENCHING 27 



The observations on the behaviour of the trees extended in 

 some cases over a considerable number of years, though in others 

 they applied only to the first few seasons after planting; but, 

 of course, the results during these first seasons are of the 

 greatest importance, as it is during them that the effect of the 

 previous trenching would be most prominent. In one instance 

 it was possible to investigate the effect of trenching the same 

 piece of ground at dates fifteen years apart, and it was found 

 that, whereas on the first occasion the trenching had produced 

 no good result, it had done so to an appreciable extent on the 

 second (Ridgmont loam, first and second series). It would appear 

 from this that the effect may vary with the nature of the seasons 

 immediately following the operation ; and this is, certainly, a 

 very probable proposition. It also appears, especially from the 

 results at Harpenden, that, even in the same season, different 

 plants may be affected to very different extents, and even in 

 opposite directions, by the trenching. 



The bulk of the experiments extended over the four seasons 

 from March 1909 to the end of 1912, a period which included 

 the unusually hot, dry summer of 1911, the cold, wet summers 

 of 1909 and 1912, and the season 1910, which was of inter- 

 mediate character. 



Samples of soil were periodically taken for determinations of 

 moisture and nitrate, and observations were made on the growth 

 of fruit trees in the plots. The following is a summary of the 

 results, the sign + being used where the result was in favour of 

 trenching, and where it was against trenching. 



Speaking generally, the values for tree-growth and for the soil 

 factors' are fairly closely connected. In the sandy soil (first 

 series), where the water supply is apt to be a limiting factor, we 

 find a drier subsoil and a reduced tree-growth to result from 

 trenching (though the results of the second series make it more 

 than doubtful whether the trenching was really the cause of the 

 extra dryness of the soil) : in the loams, trenching has had, in 

 most cases, only a small effect on the water contents, and 

 none on the nitrates; and the effect on plant-growth has, on 

 the average, been nil: on the clay, where the nitrogen supply 

 assumes considerable importance, there is a small gain in nitrate 

 on the trenched plot, and a small gain in growth also. Thus, 

 trenching is seen to have had no greater effect on the develop- 

 ment of the trees than it has had on the moisture and nitrate 

 content of the soil ; indeed, if a general mean is taken of all the 

 plant data above, it is found to be exactly nil. Evidently it is 



