3.5 



3.0 



" 2.5 



« 

 a 



m 



\- 

 «) 

 ■o 

 ■q. 



2.0 



-s 1-5 



o 



u 

 2 



1.0 



0.5 - 



0.0 



O herbicides 

 • no herbicides 



7/16 8/19 



Collection period 



8/31 9/15 



Figure 2. Effects of herbicide treatment of ground cover on the 

 mean number of spiders per tree at different times of the 

 season. This portion of the study was conducted in two blocks 

 at the University of Massachusetts Horticultural Research 

 Center which were under second-level IPM. Means within 

 each date accompanied by a different letter are significantly 

 different at odds of 19:1. 



(seven feet tall). On each sampling occasion (five 

 in all, starting Jvdy 1), spiders were collected 

 from 40 trees of each treatment. The five sec- 

 ond-level IPM blocks contained larger trees (10 

 to 13 feet tall). Forty to 65 of these in each 

 treatment were sampled on four different occa- 

 sions from each orchard. Sampling was carried 

 out by tapping the branches as described above. 

 In the five full second- level IPM blocks, her- 

 bicide treatments did not have any effect on the 

 number of spiders on trees. Mean numbers of 

 spiders per tree show the same seasonal trend 

 for herbicide as well as non-herbicide treat- 

 ments. In the two HRC blocks (Figure 2), 

 herbicide-treated trees contained significantly 



more spiders in August 

 than non-herbicide- 

 treated trees. There were 

 no significant differences 

 earlier and later in the 

 season. 



Lack of any difference 

 between herbicide- and 

 non-herbicide-treated 

 trees in the five second- 

 level IPM blocks might 

 have been due to the fact 

 that trees in these blocks 

 werematiire. Their cano- 

 pies reached well into the 

 vegetative region between 

 rows, diminishing the con- 

 trast between herbicide 

 and non-herbicide treat- 

 ments. 



In general, it can be 

 concluded that no nega- 

 tive effect of herbicide 

 treatment on numbers of 

 spiders per tree has been 

 demonstrated by these 

 data. Among smaller 

 trees there may be a slight 

 positive effect. Perhaps 

 when an understory cover 

 exists directly beneath the 

 trees, spiders may forage 

 there for prey and be di- 

 verted away from the 

 trees. They may move back into the tree cano- 

 pies when there are more insect prey to be foimd 

 there than in the ground cover. 



Conclusions 



Spiders were found to be significantly more 

 abundant in second-level than in first-level IPM 

 blocks. This result suggests that elimination of 

 insecticide use after early or mid-June allows an 

 increase in population of at least one group of 

 natural enemies. High toxicity of broad-spec- 

 trum insecticides to spiders, as revealed in our 

 laboratory tests, supports this suggestion, £is do 

 findings of Mansour et al. (1980), Madsen and 



22 



Fruit Notes, Winter, 1993 



