1883.] NATURAL SCIENCES OF PHILADELPHIA. 295 



River, in Missouri, Iowa and Minnesota ; tliat it avoids eastern, 

 wliile abounding in western Nebraslva, and is not found in Kansas 

 further east than Brookville, longitude 22° W. from Washington, 

 about 97° W. Greenwich, which is nearly' that of tlie sites reported 

 b}' Prof Todd. As Prof. Packard has reported the insect in 

 southern Montana, we ma,y now conclude that the entire western 

 part of the great valley of the Missouri (west of the river and 

 the above meridian) is inhabited b}' this ant and its closel}" allied 

 congener, P. barbatus, the Agricultural Ant. 



It is worthy of note that all the authentic reports which we 

 have of the latter insect also limit its eastern distribution to about 

 the same meridian. We have no account of it as inhabiting 

 southern Missouri, Arkansas and Louisiana, except a -note of 

 Nuttall'S in 1819, which appears to refer to one of these species. 

 Entomologists and naturalists generally in these States might do 

 good service b}' some attention to this point. It is a question of 

 profound interest, what natural cause has operated to establish this 

 eastern limit of distribution? The writer confesses his inability 

 to discover an}^ relation between the structure and econom}' of the 

 ant, and the physical condition of the country, that could throw 

 any light upon the question. 



The two species ver^- closelj' resemble each other, the worker 

 forms scarcely differing except in body-size; the worker-major of 

 Occidentalis corresponds almost exactly with the minor of bar- 

 batus. The chief differences in the sexual forms are of size and 

 color, but also a slight difference in venation. There are, however, 

 some marked differences in nidification and habit. The Agricul- 

 tural Ant occupies the southern section of the above marked 

 geographical district, and it seems scarcely possible to resist the 

 inference that it is a modified form of Occidentalis (or vice versa) 

 who inhabits the northern section. The local site of the nests in 

 Dakota is generally a sandy flat or bottom. 



2. Nidification. — From the observations of Prof Todd, it 

 further appears that the Dakota ants agree with those of Colorado 

 in the position of the gate, at one-half to one-third the distance 

 from the base ; in the general appearance of the mounds, which 

 are uniformly in the centre of a circular cleared area three or four 

 feet in diameter. In size they are smaller, being about six inches 

 high and about two feet in diameter. They are roofed in some 

 sites with small gravel stones of quartz, but in others, as at the 



