I02 PSYCHE [June 



spines or prominent tubercles; scutellum continued back as a pair of flat, triangular projec- 

 tions. Spines on the epinotum very robust. Wings fully 5 mm. long ; extending far beyond 

 the tip of the gaster, yellowish white, opaque, like ground glass; veins brownish, insertions 

 black. 



I have taken this new fungus-growing ant in the following localities in Texas : 

 Fort Stockton, Pecos County ; Del Rio, and Langtry, Valverde County ; Marfa, 

 Presidio County; Marble Falls, Burnet County and in many places in the vicinity 

 of Austin, Travis County. It is nowhere very abundant and is easily overlooked 

 on account of its extremely retiring disposition. Its nests and mushroom gardens, 

 which I hope to describe at length in another place, are similar to those of Atta 

 septentrionalis McCook. The entrance to the nest is very often surmounted by a 

 peculiar turret, sometimes i^ inches high and built of little twigs, leaves, etc. 



Atta turrifex is a well-marked species. The worker differs from the worker 

 of A. septentrionalis and the Mexican A. saussurei Forel in the following characters : 

 The antennal scape is much shorter, hardly reaching beyond the posterior angle of 

 the head, the lobes of the frontal carinae are broader and more concave, the poste- 

 rior angles 'of the head are more acute, the sides of the head are straight and not 

 Founded. The spines on the pro- and mesonotum are longer and of a different 

 conformation, the postpetiole is conspicuously shorter, the general color of the body 

 is darker, and there is no dorsal black band on the gaster. The characters of the 

 head alone, as shown in the accompanying camera sketches of A. turrifex and A. 

 septentrionalis^ will suffice to separate the species at a glance. There are corre- 

 sponding differences between the females of the two species. 



As I have at last discovered A. septentrionalis in Texas (at Milano and Den- 

 ton), it is clear that Buckley's name A. tardigrada, usually applied to this species, 

 should be given up, since there is no way of deciding to which of the two species, 

 his very inadequate description refers. 



7. Brachymvrmex nanellus, sp. nov. 



Worker. Length i mm. 



Mandibles 5-toothed, median tooth minute. Head about as broad as long, clypeus 

 broadly rounded in front, its anterior border with a sinuous impression on either side. 

 Antennal scape reaching to the posterior corner of the head, funiculus rather short and 

 thick, joints 2-6 not longer than broad. Eyes with rather large ommatidia of which there 

 are only about six in the maximum diameter. Maxillary palpi short, the three terminal 

 joints less than one and a half times as long as broad. Promesonotal and mesoepinotal 

 sutures distinct, the latter deeper, very conspicuous, and constricted. Petiole seen from 

 behind, oblong but little higher than broad. 



Whole body smooth and shining, gaster somewliat more opaque. 



