in.] ANTS AND PLANTS. 97 



setting free her terrible foe. In this manner a com- 

 paratively small force of Polyergus will fearlessly 

 attack much larger armies of other species, suffering 

 itself scarcely any loss. 



I have elsewhere discussed the relations of flowers 

 to insects, especially to bees, and particularly the 

 mode in which flowers have been modified, so that 

 the bees might transfer the pollen from one to another. 

 Ants are also, as mentioned in the preceding lecture, 

 of considerable importance to plants, especially in keep- 

 ing down the number of insects which feed on them. 

 So far as I know, however, there are no plants which 

 are specially modified, in order to be fertilized by 

 ants ; and, indeed, even to those small flowers which 

 any little insect might fertilize, the visits of winged 

 insects are much more advantageous ; because, as Mr. 

 Darwin has shown in his excellent work on cross- 

 and self-fertilization of plants, it is important that 

 the pollen should be brought, not only from a different 

 flower, but also from a different plant, while creeping 

 insects, such as ants, would naturally pass from flower 

 to flower of the same plant. 



Under these circumstances, it is important to plants 

 that ants should not obtain access to the flowers, 

 which they would otherwise rob of their honey, with- 

 out conferring on them any compensating advantage. 

 Accordingly, we not only find in flowers various modes 

 of attracting bees, but also of excluding ants ; and 

 in this way ants have exercised more influence on 

 the vegetable kingdom than might be supposed. Some- 

 times, for instance, flowers are protected by chevaux 

 de frise of spines and fine hairs pointing downwards 



H 



