mates are conservative, since the sample for 

 grading was random. Multiple pickings would 

 have resulted in greater percents in the highest 

 grade. Additionally, no summer pruning was 

 performed in our trellis treatment, and fruit 

 quahty clearly would have benefitted greatly 

 fi*om summer pruning. 



Fruit size also varied significantly 

 among the treatments (Table 2). The average 

 fruit from trees on M.7 or M.9/MM.111 were 

 140-count or smaller. Fruit from trees on M.9 

 averaged nearly 120-count in size, and those 

 from trees on M.26 were somewhat smaller than 

 120-count. 



When size and grade are considered, 

 along with yield, crop value can be estimated 

 (Table 2). Accomiting for crop value and costs, 

 Table 2 presents the net returns possible from 

 these treatments. The two M.9 and the M.26 

 treatments produced similar returns, with the 



M.9 trellis treatment giving approximately five 

 percent more, and the M.26 treatment giving 

 approximately six percent less than the M.9 

 post. The M.7A treatment netted only 53 per- 

 cent of what the M. 9 trellis netted. M. 9/MM. Ill 

 was slightly less profitable than M.7A. 



When evaluating different rootstocks 

 and training systems, it is necessary to assess 

 many different characteristics. Costs of estab- 

 hshment, training characteristics, jrield poten- 

 tial, fruit grade, fruit size, and costs of manage- 

 ment must all be considered before selecting an 

 appropriate combination. The best system is 

 the one that can be managed within the con- 

 straints of a particular grower and that provides 

 the best net returns to the orchard. In this 

 study, trees on M.9 and on M.26 were the most 

 profitable over the first ten years, and clearly 

 would be better choices than trees on either 

 M.7AorM.9/MM.lll. 



^f^ %f^ %{« «£# %t« 

 0^ rj% rj% •Y* •Y* 



12 



Fruit Notes, Winter, 1993 



