Mr. Smith's Descriptions of Species of Brazilian Ants. 157 



size of tiie eyes ; the genus is fully characterized by Guerin in 

 the Iconographie du Rfegne Animal. 



The observation of the habits of these curious ants given by 

 Lund is, that they are to be found running on the trunks and 

 leaves of shrubs and trees; our indefatigable and observant coun- 

 tryman, Mr. H. W. Bates, sends me the following account of one 

 of the species, P. ocidata: "Its colonies I have hitherto found only 

 in the tumuli of different species of Termes ; in some instances 

 I found them in spacious elliptical chambers, in the outer walls 

 of the Termitaria ; one colony to each chamber ; the cham- 

 bers wide apart and having no connection with each other ; the 

 number of individuals few in each colony; the pupae are not en- 

 closed in cocoons. In some instances I have found them with 

 their larva and pupa? within the same chambers as the Termes, in 

 different parts of the Termitarium ; the workers are sometimes 

 found in numbers, coursing rapidly over trees and herbage. 

 Another species constructs its Formicarium in the pith tube of 

 dried twigs, the colonies are not numerous." We may from 

 these circumstances perceive that they are insects of varied habit, 

 and that, like those of the genera Formica and Myrmica found in 

 this country, some prefer to construct their habitations under 

 ground, others in decaying trees, whilst at least one species chooses 

 part of the same mound or tumuli, as a species of Termite; in the 

 same manner we find sjjecies of Myrmica scofermorfis occupying one 

 side of a little hillock, and Formica fava the other. 



I have a very strong suspicion that some of the species described 

 in this paper belong to the genus Condylodon, proposed by Lund, 

 whilst others would fall into that of Pseudomijrma ; the distinc- 

 tions between these being merely indicated by that author in his 

 communication to his friend Audouin ; but as the species which 

 presents the greatest disparity to the type (P. advcna) is one of 

 which I possess the winged female, and as 1 find the neuration 

 identical with that of the typical species, I retain them all in one 

 genus. 



Sp. 1. Pseudomyrma bicolor, Guer. 



Pseudomyrma bicolor, Guer., Icon. Reg. Anim. Ins., 437. 



Worker. — Length 5 lines. Obscure black ; shining and thinly 

 covered with a fine sericeous pile ; the mouth, anterior margin of 

 the face, the scape at the base and apex beneath, and the flagellum 

 beneath, rufo-testaceous ; the articulations of the joints of the 

 legs, the anterior tibiae and tarsi, rufo-testaceous. The first node, 



