330 REPORT UPON COTTON INSECTS. 



In order that the significance of some of the examples given may be 

 fully understood it will be necessary to speak briefly of the habits of a 

 few insects. Ants, the most numerous of all the visitors of extrafloral 

 nectar glands, are of various habits. So far as I know all of the species 

 with which I had to do in Alabama are omnivorous, eating nectar and 

 other sweet substances, but largely feeding upon animal food. In pleasant 

 weather they may be found abroad night and day. But this is not true 

 of all ants. The leaf-cutting and umbrella ants, or Saiiba of Central and 

 South America (Occodoma), are entirely herbivorous. Excavating large 

 tunnels, and living in immense communities, they are the terror of gar- 

 deners in the hotter parts of our continent; for they have the habit of 

 marching in great armies which swarm over and defoliate every unpro- 

 tected plant, preferring cultivated plants, since they, as a rule, neither 

 possess properties rendering them unpleasant to the taste of the ants, 

 nor special provisions to secure a body-guard of protecting insects, and 

 one or the other of these means of defense is usually found in native 

 plants. Having reached the leaves or petals each ant snips out as large 

 a piece as he can carry and makes oif with it to the nest. In damp, 

 Avarm weather these ants forage at all hours, but when the air is hot and 

 dry they seem to realize that the leaves would dry up and become use- 

 less before they could get them to the nest, and so they hunt only dur- 

 ing the cooler hours of the day and at night.* Moggridge found that a 

 graminivorous ant of the south of France (Plieidole megaceplialu] works 

 mostly at night,t while McCook finds that the parasol-ants of Texas 

 forage only at night, visiting, then, the tops of the highest trees in their 

 leaf-collecting labor.} So great a pest are these ants in Central America 

 that it is found impossible, except by the most strenuous exertions, to 

 cultivate any but native plants. 



From this it appears that any plant not protected by an unpleasant 

 principle in its flowers and foliage is very liable to extinction where 

 these ants abound, unless it can secure a body-guard of some kind, and 

 this usually consists of nectar-loving ants. To give perfect protection 

 this force must reside constantly on the plant, finding their food, drink, 

 and lodging, which, it will be remembered, were all found on the Acacia 

 previously mentioned. A less perfect protection would be afforded by 

 ants attracted to the plant for some of their food, but residing else- 

 where ; but it is probable that so few of them would be on the plant at 

 any given moment that an army of the leaf-cutters would have no diffi- 

 culty in overrunning it in their sudden onslaught. Let us suppose a 

 case in which the attacking ants travel in small bands and only by 

 night ; then, evidently, a good protection would be afforded by a small 

 number of pugnacious, nectar-loving ants, called to the plant chiefly or 



* See, on these ants, Bates, Naturalist on the Amazous, and Belt, Naturalist in Nica- 

 ragua. 



t Ann. Nat. Hist., series , xiv, 1874, p. 92. 



t Quoted by Bettauy, Nature, Oct. 16, 1879, p. 583. 



