140 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



sown over their roots, as was the case in most of the experiments. 

 This fact, as well as the varying intensity of the effect of grass in 

 different soils and the greater or less susceptibility of certain varieties, 

 account for the effect of grass in commercial and farmhouse orchards 

 being generally less than that observed in the experimental plots (p. 3). 

 Practical experience derived in planting annually some 400 or 500 

 trees in the orchards of farm tenants in Devonshire leads the authors 

 to the unhesitating conclusion that the most essential item for the 

 welfare of orchard trees is the keeping of the grass away from their 

 roots (p. 4). Oases in which trees have not suffered to some extent 

 from being grassed over are rare, some trees at the farm which were 

 grassed over twelve years after planting beginning to suffer severely 

 the third or fourth season after planting (p. 19). 



No explanation of the difference in the grass-effect in different 

 soils can be traced to the depth of good soil available for root develop- 

 ment, and little difference is shown in its effects, whether the trees are 

 on surface- or deep-rooting stocks (p. 8). In certain soils which are 

 favourable to the practice the effect of grassing to within five or six 

 feet of the stems may result in heavier cropping and increased colour 

 in the fruit, and such results have been obtained at Eidgmont. The 

 practice cannot be generally recommended, however, except experi- 

 mentally, owing to the differences in soils (p. 11). The action of 

 the grass upon the well-being of the trees is felt as soon as ever any 

 of the tree roots come into the grassed area, and recovery begins 

 as soon as any of the roots get beyond, or even near, the limits of 

 the same, examination in one case where grass-effect was apparent 

 showing that the proportion of roots extending into the grassed soil i 

 was only -^-^—th. part of the weight of the whole tree (p. 12). 

 Forest trees appear to be affected by grass in the same way as fruit 

 trees, though in the case of conifers planted in light soil the effect 

 was much less than with other trees, and recovery occurred as time j 

 went on instead of the effect becoming intensified (pp. 41-4). 



Experiments were conducted with eighteen different grasses, the 

 action of all of which was considerable, the effect produced in each 

 case being proportionate to the strength of the grasses. Clovers 

 had a similar effect, but the lightness in the colour of the leaves was 

 absent (pp. 45-51). 



Further investigations strengthened the conclusions that grass- , 

 effect was not explainable by its affecting the aeration or temperature j 

 of the soil, or altering the content of carbonic acid, moisture 

 (pp. 52-68), or food (pp. 69-81). Trees in tilled soil did not exhibit i 

 the grass-effect in dry seasons, and it seems to be practically inde- I 

 pendent of rainfall. As grass crops, if properly manured, actually | 

 enrich the soil, and trees grown in soil taken from the grass plots 

 flourished better than in soil taken from the tilled plots, if grass- 

 effect is due to starvation it is in a land of plenty. Experiments 

 begun in 1909 seem to definitely prove that the results of cutting 

 and grazing the grass are identical, at least in the case of freshly 



