573 



found where coffee borders on cacao, or where these plants are inter- 

 mingled. In coffee gardens the ant also apparently feeds on resinous 

 exudations of a vegetable nature, besides searching for the secretions 

 of Coccus viridis, which it aids, though to a less degree than does 

 Phgiolepis lotigipes [see this Review, Ser. A, v, p. 274J, so that it is 

 harmful, though not the cause of much injury. It also greatly hinders 

 the labourers harvesting the coffee berries. D. hituberculutus is 

 partly responsible for infestation by the so-called " Robusta " cater- 

 pillars, wliich are attracted by the excreta of Coccus viridis and 

 Pseudococciis citri, tended by this and other ants, and as it does not 

 appear to play any useful role in coffee plantations, its presence there 

 is distinctly harmful. 



For establishing D. bituberculatus in cacao plantations the presence 

 of adequate shade is necessary. Other ants must be combated, 

 especially Phgiolepis longipes [see this Review, Ser. A, iii, p. 664 ; 

 V, p. 275]. The introduction of Pseudococciis crotonis is also very 

 necessary, for it is only where this scale occurs that D. bituberculatus 

 can maintain itself for a long period. The scale is best introduced on 

 the empty shells of infested cacao pods after they have been carefully 

 opened at harvesting time. As a supplementary measure such trees 

 as Tephrosia Candida, randoe {Ceiba pentayulra) and Deguelia micro- 

 phylla may be planted among the cacao. To introduce D. bituberculatus 

 leaf-nests are hung up in places where the ant abounds, and when they 

 are well infested they are removed to the cacao trees. The description 

 given of the nests of D. bituberculatus shows that protected situations 

 at some height above ground are preferred. Such places are not often 

 met with in a well-kept plantation, but the difficulty has been solved 

 by arranging artificial nests, of which many forms are described. One 

 type is made of the woody bracts of bamboo, bent and tied down so as 

 to form a case which is filled with cacao or banana leaves ; another 

 consists of a section of bamboo stem, also filled with leaves, and a 

 third is a conical nest of leaves secured together. The nests must be 

 hung on young branches near the pods, and cacao plants not yet in 

 bearing should also be protected in this way. 



The cost of this efficacious means of protecting cacao against 

 Helopelfis is greater than that of the direct methods of collection or 

 burning with torches. The following costs were worked out on an 

 area of about 52 acres, each of 500 plants, and do not include a charge 

 of about 6s. per acre, which might be required to free the ground from 

 Phgiolepis lon/jipes. If six nests per tree are necessary to begin with, 

 the manufacture of 3,000 nests per acre wll cost about Qs., filling them 

 wdth leaves and hanging them up will cost about 3.9., and the 

 introduction of Pseudococcus crotonis, about Is., or about 10s. in all for 

 the first year. In the following years the annual cost may amount to 

 about 4s. per acre, assuming that four nests per plant have to be 

 renewed. The cost of direct work against Helopeltis, such as collection 

 and burning with torches, varies considerably, the average figuras 

 ranging from Is. 8d. to 4s. '2d. per acre. This is considerably lower than 

 that of introducing and maintaining D. bitubercidatus, but the latter 

 method requires less attention and labour and is therefore the best for 

 combating Helopeltis on cacao. 



A supplement to this paper deals fully with Pseudococcus crotonis, 

 particulars being given of its systematic position, developmental foi'ms. 



