104 SCIENCE AND FRUIT GROWING 



whole tree, though in most other cases the roots are accountable 

 for a larger proportion of the total, the averages from the 

 number of instances indicated below being 



Apples (461 instances). Roots = 22-9 per cent, of total. 



Damsons (6 instances). ,, = 25*2 ,, ,, 



Pears (15 instances). ,, = 23*5 



Plums (44 instances). ,, = 28-3 ,, ,, 



The trees examined varied much in age and size weighing 

 from 4 to 148 kilos per tree but no connection was observable 

 between such variations and the existence of higher or lower 

 values for the proportional weight of the roots. 



The results of the chemical analyses were 



Branches p Whole Green Ripe 



and stem. tree. fruit. fruit. 



Apples. 



Water . . 45-0 497 45*8 86-4 



Ash . . 1-126 1-515 ri88 0-282 



Potash. . 0*192 0*184 0-191 0-123 



Gooseberries. 



Water . . 53'o 89-8 84-3 



Ash . . 2-133 0-384 0-354 



Potash. . 0-218 0*198 0*148 



These values, combined with a knowledge of weights of the 

 trees, the prunings and the fruit crop, led to the conclusion that 

 the average annual production per acre during twenty-two years 

 had been, in the case of Bramley 



Undried wood . . 2,045 kilos j _ Potash 



Dried wood . . 1,110 ,, J 



Fruit . . . 3,090 = 3-61 kilos. 1 



The total potash removed thus amounts to 7*52 kilos per 

 annum, a quantity considerably in excess of the 4*3 kilos, which 

 the Rothamsted experiments on wheat would lead us to suppose 

 the soil could supply without the addition of manure. There 

 is, however, in the case of trees, a factor which increases the 

 capability of the soil to support growth namely, the formation 

 of leaves : this, it is true, constitutes an extra drain on the 



1 Cf. Van Slyke, Taylor and Andrews, " Plant-food constituents used 

 by bearing fruit trees," New York State Agric. Expt. Sta., Bull., 265, 1905 ; 

 also P. Collier, " Analysis of Fruit Trees," New York State Ann. Rep., 1892, 

 173-209. 



