11 



tables for this year — have, of course, the explanation already given ; 

 viz., that the trees set very full in spring and early cast a great 

 quantity of their surplus apples. 



A curious drop in the ratios of injury of both check and ex- 

 perimental trees for the observation of September 8, probably 

 marks an interval between second and third broods similar to 

 that already noticed between the first and second of last year.* 

 The same indication is apparent on Tables lY., VI., and IX. 



Of the apples borne by these trees, we examined 2,060 from the 

 check, and 1,580 from the tree sprayed with poison. The number 

 on these and several other of the trees was so great that towards 

 the end of the season we did not handle them all, but stopped 

 when we had demonstrated that we had a fair average of the 

 whole. 



The increased benefit from a second spraying is made evident 

 by a comparison of Table IV. with the preceding one. Although the 

 ratio of benefit to fallen fruit is somewhat less, the fifty-one per 

 cent, benefit to the picked apples once sprayed (Table III.) be- 

 comes eighty-four percent, after the second spraying (Table IV.). 

 Curiously, the third spraying (Table V.) seems to have been 

 without important effect, the benefit to fallen fruit being a little 

 greater and that to picked fruit a little less on this table than on 

 the preceding. The record from two other trees not tabulated 

 here, gives a similar result. 



The respective advantages of spraying once and twice are more 

 fully represented on Table VI. The benefit from the second spray- 

 ing begins to be conspicuous August 31, and continues quite to 

 the end of the season, October 4. The total effect on the fallen 

 fruit is apparently slight; but this is due to the fact that much 

 the larger part of the apples fell from general causes early in the 

 season. The ratios relating to the picked fruit are not disturbed 

 by tiiis superabundance of early windfalls, and more correctly ex- 

 hibit the relative effects of spraying once and twice. By this the 

 damage to picked fruit after twice spraying was only about one 

 seventh that after spraying but once. 



The most remarkable exhibition of benefit is that of Table VII., 

 where a single tree twice sprayed — the Vandevere variety — 

 is compared with its selected check. The damage to the latter 

 ranges, as will be seen, from naught to sixty-two per cent., the 

 general average for the fallen fruit being forty-eight per cent, and 

 for the picked apples forty-three per cent.; while that to the treated 

 tree varies' only from zero to six per cent., the "total fallen" being 



"On thaVnt Aiij?iist t.h« larv;e were reported by the assistants engaged ontheworkas "mostly 

 full grown," and September 8. as mostly full grown, but with specimens occasionally occur/ ing 

 that were not more than half size. Septernber 2i>, on the other hand, when 4,14G apples "were exam- 

 ined, a very large perwntage (three fourths to five sixths) of the apples contained very small 

 larvif, many of them near the blossom end and apparently jn^t entering, the others being' all full 

 grown October ^, finally, many of the apples contained very young larvae, although the majority 

 were one third to two thirds grown. The notes of the assistants are supported by alcoholic speci- 

 mens. 



