iNsi-X'i" i-;.\i:.\i ii:s oi-- riii'. codm.xc moiii. 247 



down the trunks, while onK- a small stream of the insects was 

 travelling; n[) and down the trunks of the pear trees and throtigh 

 the bands. The ants were a])parently attracted to the pear 

 trees by Pseitdococciis capcusis, a mealy bug which was infesting 

 the pear fruits. The api)le trees were infested with woolly 

 aphis, which attracted the ants. The apple orchard was 

 irrigated during the whole of the fruit season, shallow ditches 

 leading the water from one tree to another. Water flowed 

 through these ditches at intervals of a few days. The remainder 

 of the time they were moist, during which period ants swarmed 

 in the ditches, apparently being attracted by the moisture, and 

 being led to the trees by means of connecting ditches. 



Table 5. recording the banding experiments, shows that in 

 the bands of the apple trees, over and under which many 

 ants constantly passed, only caie larva was collected from January 

 16 to February 13; 72 larvre were gathered in the same period 

 in the bands of those trees to which ants were prevented access 

 by a tanglefoot barrier. In the bands of the four pear trees, 

 over and tinder which comparatively few ants were constantly 

 passing, 62 larvae were captured from January 16 to February 2; 

 in those trees to which ants were ]:)revented access 104 larvae 

 were captured. 



Six large apple trees, loaded with fruit, in the Le Roux 

 orchard were banded during the whole season, and were examined 

 about every month. Only a total of four larvae was discovered 

 in these bands during the whole season. The owner of this 

 apple orchard claims and is reputed to have his fruit comparatively 

 free from codling infestation every year, although he sprays his 

 large trees only with small knapsack pumps. It was noted by 

 the writer, during 1916 and 1917. that this orchard was much 

 more free from codling than neighbouring ones, where irrigation 

 was not practised, and where Iridomyrmex humilis was not so 

 abundant. 



Results show that the banding of trees is not effective in 

 capturing codling larvae where the .Vrgentine ant visits the trees 

 constantly in large numbers. This ant appears to be of con- 

 siderable use in the control of Carporapsa pomonclla in orchards 

 where it is very abundant on the trees during the fruit season. 



Notwithstanding the effectiveness of Iridomyrmex humilis 

 in codling control under the above-mentioned conditions, it is 

 inadvisable to encourage the increase of this insect, which in 

 many respects causes injury, directly or indirectly, to the fruit- 

 grower. 

 Liogrylliis hiiiiacitlatiis ( Ieek as ax Fnemy of Codling L.^rv-I!. 



Crickets, identified by Dr. Peringuey, of the Capetown 

 3iluseum, as L. bimacitlatiis, made their first appearance of the 

 fruit season in the codling bands on January 8, and were present 

 from that date until the end of the fruit season, gradually 

 increasing in number and reaching a maximum during the last 



