THE ARGENTINE ANT IN RELATION TO CITRUS GROVES. 25 



comparatively free from other infesting insects. Several experi- 

 ments were conducted in that county in the summer of 1916 which 

 bring out pretty well the varying effects of ant attendance on the 

 mealybugs under different conditions. 



Experiment I. 



The subject of Experiment I was an orange tree whose 6 main 

 branches had been cut back to stubs about 2 to 3 feet long. Three 

 of the stubs, with 28 new shoots, were banded to exclude the ants, 

 while the other 3, with 27 shoots, were left free to the ants. Mealy- 

 bug infestation, prevalence of mealybug enemies, ant attendance, 

 and vigor of tree were noted at intervals from the beginning of the 

 experiment, April 14, to its conclusion, September 2, 1916. The re- 

 sults are summarized in Tables III and IV. 



Table III. — Effect of the Argentine ant on abundance of mealybugs on orange. 

 Los Angeles County, Cal., 1916. 



Date. 



Ants present. 



Mealybug infestation. 



Number 

 of mealy- 

 bug ene- 

 mies seen 



Ants excluded. 



Mealybug infestation. 



Number 

 of mealy- 

 bug ene- 

 mies seen. 



Apr. 14 



May 3 



May 17 



July 6 



July 17 



74 clusters and groups 



106 groups 



361 groups of 10 to 150 bugs each 



112 groups, 10 to 50 ovipositing fe- 

 males with egg masses 



As on July 6, but more young scat- 

 tered over leaves 



73 clusters and groups 



83 groups 



45 groups of 10 to 30 bugs each 



12 ovipositing females only with 

 egg masses; 9 masses of destroyed 



mealybug material 



No living mealybugs 



The larger groups or clusters of mealybugs at first occurred on 

 the main branches, where they had passed the winter, but the mi- 

 grating young formed smaller but populous groups at the bases of 

 the smaller branches and of the leaves. It will be noted that at the 

 outset of this experiment there was nearly complete uniformity in 

 the amount of infestation between the branches from which ants 

 ware excluded and those to which ants had access. Substantial 

 uniformity of infestation persisted up to May 3, when there was a 

 somewhat greater number of groups of mealybugs and more scat- 

 tered individuals on ant-invaded branches than on those kept free 

 from ants. Between May 3 and July 17 the mealybugs rapidly 

 diminished to complete disappearance on the branches from which 

 ants were excluded, whereas on those to which ants had access mealy- 

 bugs continued to increase rapidly for a time, reaching the high 

 point of infestation on May 17. Thereafter the infestation de- 

 creased on these branches also, but much more slowly than on those 



