180 POMOLOGY 



from several points of view, viz. : (1) The soil in some of the 

 experimental orchards has been exceedingly variable and yet 

 no account has been taken of that fact in recording the ef- 

 fects of the various fertilizer treatments. (2) Fruit-trees, 

 particularly large apple trees, vary exceedingly in the size 

 of the crop they produce, and averages for any given plot 

 may be misleading. (3) When buffer rows have not been 

 maintained between the plots, there has not infrequently 

 been cross feeding which would seriously modify the re- 

 sults. (4) Missing trees in an orchard may give certain 

 advantages to those adjacent to the open spaces, and also 

 diseased or subnormal individuals may not give a repre- 

 sentative result of the treatments used in an orchard. (5) 

 Differences of topography which may give an advantage 

 to certain plots over others because of unequal frost action 

 and drainage have often played a large part in results se- 

 cured, without allowances being made for the inequalities, 

 often without mention of them. (6) Unwarranted con- 

 clusions have been drawn of the value of a single element 

 by subtracting the performance of a two-element plot from 

 that of a three-element plot. Many other suggestions or 

 criticisms might be enumerated, all of which would be jus- 

 tifiable in critically examining this problem. 



Without question it is more difficult to select uniform 

 conditions for an orchard experiment than for field tests of 

 the farm crops, and the available data are open to criticism 

 on many grounds. Nevertheless, this field work has been 

 a valuable, if not a necessary, forerumier of the more tech- 

 nical studies of a physiological nature that must follow. A 

 number of valuable economic questions have already been 

 settled and much of the previous confusion in regard to or- 

 chard fertilization has been cleared up. 



149. Fertility removed by fruit-trees. — Proceeding from 

 a chemical view-point, the amount of fertility removed by 



