129 



have been studied by Keller (1892), and more recently and 

 somewhat more closely by Sjöstedt (1908). These acacias are 

 A. fistula Schweinf., zanzibarica Taub., drepanolohium Harms, seyal 

 Del., and bussei Harms. Their stipular thorns exhibit a much 

 greater variety of shape and a much more extraordinary enlarge- 

 ment at the base than is found in any of the American species. 

 So extreme is this enlargement, in fact, that Sjöstedt regards 

 them as galls, and believes that they may owe their development 

 to the stings of Diptera or Hymenoptcra or, more probably, to 

 the irritation produced by certain Coccids which he found on the 

 very young twigs. Both Keller and Sjöstedt, however, are 

 certain that the ants have nothing to do with the modification 

 of the thorns, since they found them attaining their full size 

 before being entered by the ants, and also on bushes that har- 

 boured none of these insects. The galls are at first solid and 

 are hollowed out by the ants. These belong mostly to the den- 

 drophilous genus Crematogaster. The species taken in the 

 thorns of A. fistula by Keller and identified by Forel (1892) 

 were C. chiarinii Emer\', acacice Forel, and ruspolii Forel. 

 Sjöstedt took C. chiarinii in the thorns of A. zanzibarica ; C. 

 admota Mayr, C. sjöstcdti Mayr, and Sima penzegi Mayr in those 

 of A. drepanolobiufn and C. solenopsidis Emery var. flavida 

 Mayr and Cataulacus intrudens F. Sm. in the thorns of .-1. bîissci. 

 The various Crematogaster s do not perforate the tips of the 

 thorns, like the Pseudomyrmas, but make one or more roimd 

 holes in the dilated basal or gall portion. It may be noticed 

 in passing that the Crematogaster which Belt found nesting in 

 A. cornigera has the same habit. The African acacias bear no 

 Beltian bodies, although they are furnished with crater-shaped 

 nectaries on the leaf-petioles. Sjöstedt failed to observe the 

 ants in the act of visiting these nectaries, but he admits that 

 they may do so when the leaves are very young. One infers 

 from his description that the ants obtain their food largel}' from 

 the numerous Coccids [Dactylopius coccineus Newst.) and larval 

 Membracids which infest the plants. Both Keller and Sjöstedt 

 believe that these acacias of the African plains may be protected 

 from the antelopes, goats, and camels by their ants, although 

 these are certainly far less vicious than the American Pseudo- 

 myrmas. It is, however, by no means clear that the plants are 



17 



