MANURES 103 



of a very different order of magnitude, and they are evidently 

 altogether exceptional in their behaviour, though we must 

 remember the possibility (see p. 97) of other bush fruits showing 

 a similar effect if the experiments on them were sufficiently 

 prolonged. 



Dealing, first, with the absence of effect of manures on apple 

 trees, a comparison of the Ridgmont soil with that at Harpenden 

 (XVI, 62), which latter may be taken as representing a soil of 

 average fertility, shows to the disadvantage of the former, and 

 renders it impossible to explain the Ridgmont results on the 

 ground that the soil there is exceptionally rich. The next 

 question which arises, therefore, is whether the material removed 

 annually from the soil by growing apple trees in it is less 

 than the material which becomes available for growth through 

 the gradual decomposition of the insoluble mineral matter in 

 the soil and subsoil. The Rothamsted experiments 1 show that, 

 as regards wheat, a crop which requires only 4*3 kilos, or less, 

 of potash a year, might conceivably be grown for an indefinite 

 time without the addition of manure, potash to this amount 

 becoming available from that in the subsoil ; and Ebermayers' 2 

 results show that, in the case of forest trees, even when these 

 are regularly cropped, the removal of potash is considerably 

 less than this (r6 to 2' 8 kilos), so that land under forest must 

 tend to increase in richness, as, indeed, would be inferred from 

 the fertility exhibited by virgin soil, and by land from which 

 plantations of forest trees have been cleared. 



But many points of uncertainty arise in applying data from 

 forest plantations or from the growing of wheat, to the case of 

 apple trees : it appears fairly clear, however, that the latter 

 remove considerably more material from the soil than forest, 

 trees do, and that it must generally amount to more than the 

 4-3 kilos of potash required as a minimum by wheat without the 

 soil becoming depleted. 



To enable some estimate to be formed, determinations of the 

 potash in the branches, roots and fruit of apple trees, and also 

 of gooseberry bushes, were made : further, to apply these results, 

 it was necessary to ascertain what proportions the roots bore to 

 the branches. For Bramley's Seedling on paradise stock, the 

 roots were found to represent 16 per cent, of the weight of the 



1 Dyer, Bernard : " Results of Investigations of the Rothamsted Soils," 

 U. S. Dept. of Agric., Bull., 106, 1902. 



2 " Waldstreu," p. 116, and " Physiologische Chem. der Pflanzen," pp. 709, 

 782 : see also Warington, Chemistry of the Farm, i8th ed., pp. 74, 81, and 

 The Woburn Report, XVI, p. 69. 



