BIOLOGIC RELATIONS BETWEEN PLANTS AND ANTS. 429 



the holes in it do not exist there naturally, and seem to be made by those 

 insects. This, too, it would seem, is a rough draft of a myrmecophilous 

 feature. (See PL XVII.) 



Usually lodging plants offer to ants a well-closed cavity formed at 

 the expense of one of their organs. Such is the case with Acacia 

 comigera, a small tree of Central America, about six or seven meters 

 high, having at the base of each leaf two strong spines, representing 

 modified stipules. It is generally allowed that it was Belt who first, in 

 Nicaragua, studied the relations of this tree to ants. It had, however, 

 already been mentioned and figured by earlier observers: Hernandez 

 (1651), Hermann (16S9), Commelin (1G98), Plukinet (1691). The spines 

 are strong and have been compared with some exactitude to bull's 

 horns. They are hollow within, the cavities of the two contiguous 

 spines intercommunicating. The leaves are bipinnate, and at the base 

 of each pair on the median nervure there is found a crateriform gland 

 which in young leaves secretes a honey-like liquid. These foliary nec- 

 taries attract a great number of ants, which are constantly running 

 about from one gland to another. 



But this is not the only aliment offered to these insects. There are 

 nectaries of another kind. At the extremity of each of the small 

 divisions of a compound leaflet there is formed a little yellow fruit- 

 like body, attached to the leaf at a single point. When the leaf first 

 unfolds these little pears are not quite ripe, and the ants are continually 

 employed going from one to the other examining them. When an ant 

 finds one sufficiently advanced it bites the small point of attachment. 

 Then, bending down the fruit-like body, it breaks it off and bears it 

 away in triumph to the nest. The ants are therefore found continually 

 upon the plant occupied in harvesting these glandules, which ripen 

 successively. Since these organs are attractive to ants as dainties, 

 Francis Darwin gave them the name of " food bodies." We may per- 

 haps attribute the production of the "food bodies of the Acacia corni- 

 gera to the ants themselves. The ancestors of the plant must have pos- 

 sessed leaves that secreted at their borders a mucilaginous liquid in 

 greater or less abundance. The ants, attracted by this liquid, began 

 to nibble the secreting edge of these leaves, and thus produced an irri- 

 tation resulting in a more abundant secretion of the liquid and an 

 hypertrophy of the secreting parts. We have already spoken of the 

 theory that accounts for the primary differentiation of the perifoliary 

 nectaries by the irritation caused by the suction and bites of insects 

 seeking mucilaginous or sugary liquids. 



Belt has observed that two species of ants visit the spines of the 

 Acacia. The most frequent visitor is the Pseudomyrma bicolor. The 

 other is a species of Crematogaster. The two never inhabit at the 

 same time the same tree, nor do they perforate the same spines at the 

 same place, the former penetrating them near the apex, the latter 

 about midway of their length. 



