133 



are perfect adepts at utilising the thorns as dwellings, and the 

 nectar and Beltian bodies as food. The simplest explanation, 

 therefore, is that these ants were formerly pandendrophiloiis, like 

 the vast majority of Pseudomyrma species, but that they long 

 ago discovered the greater advantage of living on the acacias, and 

 have since confined their attention exclusively to these plants. 

 That this adaptation may have been very easily and quickly 

 established is shown by Ps. gracilis and Camponotus pianatus, 

 both widely distributed, and pandendrophik)us ants, which in 

 certain regions have become as completely adapted to the acacias 

 as the obligatory Pseudomyrmas. On this view there is nothing 

 any more remarkable in the predilection of particular ants for 

 particular species of plants than there is in the predilection of 

 particular phytophagous insects for particular host plants. 

 Indeed, this view can be rejected onl\' by those who are un- 

 familiar with the ant-life of the tropics, who have never been 

 impressed by the vast numbers of these insects perpetually ex- 

 ploring the surfaces of the rank and varied vegetation in their 

 eager search for food and habitations. No suitable cavitv in the 

 plant body, no sweet exudation, no particle of accessible food 

 escapes their attention, and any plant that furnishes one or more 

 of these desiderata is at once appropriated and becomes " myr- 

 mecophilous." And although it must be admitted that some of 

 these dendrophilous ants {Pseudomyrma, Azteca) sting and bite 

 severely, and may therefore defend the plants, this is, of course, 

 merely a coincidence or by-product, as it were, of the true defence 

 which the ants exercise in behalf of their own bodies and their 

 brood. I believe, therefore, that we may adopt vox Ihering's 

 point of view, and say that Acacia cornígera, hindsii, and sphcero- 

 cephala have no more need of their ants than dogs have of their 

 ifeas. If this is true, the relation between the ants and })lants 

 is not one of symbiosis, but one of parasitism. 



Notes on Central American Species of Cecropia and 



Triplaris. 



The case of the acacias is, indeed, much more like that of the 

 Cecropias than is generally supposed. Some years ago (1907) 



