1907.] Wheeler, Fungus-growing Ants of North Atyierica. 769 



of C. wheeleri and entail a corresponding contrast in habits. All the forms 

 of rimosus that have come under my observation live in the shade of trees 

 and bushes in rather moist, black soil. These ants are, in fact, restricted 

 to such localities on account of the material they require for constructing 

 their gardens and the peculiarities of the fungus which they cultivate. The 

 habits of the subspecies minutus Avhich I have had abundant opportunities 

 of observing in the Florida Keys, Bahamas, Culebra and Porto Rico, and 

 those of the subspecies dentatus which I first found in the lovely barrancas 

 about Cuernavaca, Mexico, resemble so closely the habits of the var. comalen- 

 sis at New Braunfels, Texas, that I may confine my remarks very largely to 

 this form. 



At New Braunfels a number of beautiful springs, the sources of the Comal 

 River, gush forth from the foot of Mission ^Mountain, one of the limestone 

 hills that constitute the Grand Prairie escarpment (Fig. 31) . The volume and 

 temperature of these springs is practically constant during the entire year. 

 They nourish an exuberant vegetation consisting of ash-trees, live-oaks 

 and shittim wood (Bumelia lycioides) and a dense undergro'ni;!! of sub- 

 tropical shrubs and herbaceous plants too numerous to mention. The 

 entomologist who enters this undergrowth must be prepared to endure the 

 fiery torments of the "red-bugs" or " coloradillos " ("Leptus" irritans) and 

 exercise some care lest he tread on a water moccasin. But, if he be in 

 search of ants he will be rewarded by finding a number of interesting sub- 

 tropical species, among others three species of Pseudomyrma (pallida, 

 hrunnea and flavida), a singular little Strumigem/s (S. margaritoe Forel) 

 hitherto known only from the island of St. Vincent, besides the fungus-grow- 

 ing ant with which we are here concerned. 



This ant, owing to the close agreement between its color and the black 

 soil over which it moves, is more difficult to detect than any of the other 

 small Attii described in the preceding pages. Single workers wander about 

 slowly in the damp shade of the plants in search of the caterpillar excrement 

 with which they construct their gardens. As soon as one of the short, cylin- 

 drical, ribbed pellets is found, the ant seizes it in her jaws, raises it above 

 her head like a man shouldering a cask and returns home with accelerated 

 pace. The slightest touch causes the ant to drop her load, draw up her legs 

 and antennffi and "feign death." And he must have exceptionally good 

 eye-sight who can distinguish her rough, opaque and inert body from the 

 particles of earth among which it falls. 



The colonies of C. comalensis are larger than those of C. tclieclcri, some- 

 times comprising a hundred or more workers and from one to three dealated 

 queens. The nests are under rather small flat stones or pieces of wood, 

 with the entrance sometimes nearly a cm. in diameter, at the periphery. On 



