1907.] Wheeler, Fungus-growing Ants of North America. 741 



auditory sensations. This result agrees also with the aceounts of others 

 who have investigated the pereeption of vibrations in insects. 



Of all ants the Attii would seem, at first thought, to offer in the great 

 sponge-like masses of decomposing vegetable matter of their fungus gardens 

 the most favorable of resorts for all kinds of myrmeeophiles and synoeketes. 

 But the number of such animals hitherto observed in the nests of these 

 ants is very small. This is probably due to the excjuisitc care and diligence 

 with which the ants patrol and cultivate all parts of their gardens to 

 prevent the growth of the aerial hyph?e, alien fungi and bacteria, for under 

 such circumstances any intruder might be easily detected and ruthlessly 

 destroyed. Nevertheless a few animals have managed to secure a foothold 

 in the nests, but so far as known, only in those belonging to species of Afta 

 s. str. and Acromynnex. I have never seen any traces of myrmeeophiles 

 in the many nests of Trachymyrmex, M ycetosorifes and Cyphomyrmex which 

 I have examined. Bates (1892) and Brent (1886) state that certain Amphis- 

 bfenian lizards manage to live in the Afta nests of Brazil and Trinidad. It 

 is probable that these reptiles feed on the ants. Belt (1874) mentions a 

 large Staph yJinid beetle as occurring on the Atta nests of Nicaragua, and 

 Wasmann (1900) concludes that this beetle, which he identifies as Smilax 

 pilosa Fabr., is probably a true myrmecophile, because it so closely resembles 

 the large Atta workers in its dark brown color and abundant pile. The 

 same author (1S94, 1895) mentions several Histerid beetles (Philister 

 rufulus Lev>'is, Hister (?) costatus ]Mars, Reninus salvini Lewis and Car- 

 cinops (?) multistriata Lewis) as having been taken from the nests of Atta 

 mexicana, and three Staphylinidae belonging to the genera Aleochara and 

 Athefa from the nests of .4. scxdens. These are probably all not true 

 guests but synoeketes. To the same group belong also a number of speci- 

 mens of the myriopod Scutlgera which I found running about in the galleries 

 of an .4. texana nest. 



The only myrmeeophiles known to live in intimate relations with Attiine 

 ants are the small and al)errant cockroaches of the genus Attaphila of which 

 I described the first species (.4. fungicola, PI. LIII, Figs. 47-49) from Texas 

 (1900). This insect, which is verv^ common in the fungus gardens of .4. 

 texana, measures only 3-3.5 mm. in length. It is yellowish brown and has 

 ver\' small eyes, one-jointed cerci, and ]:>eculiar antenna\ consisting of a few 

 cylindrical joints. The females are wingless, the males have vestigial tvg- 

 mina and hind wings. The antennae are always imperfect, their terminal 

 joints having been bitten off, in all jirobability, while the ants an' clij)j)ing 

 the fungus mycelium. The .structure of the remaining antennal joints is so 

 unlike that of all other Blattidae that Attaphila niu.st be regarded as the type 

 of a distinct subfamily, the Attaphilina^. Since publishing my description of 



