252 Journal Xew York Extomological Society. [^'o'- ^i^ 



" In habits T. obsciirior scarcely differs from the other species 

 of the subgenus. The workers are sluggish in their movements and 

 ' play possum ' or ' feign death ' like their congeners. Caterpillar 

 excrement is used for the substratum of the fungus-gardens. At the 

 beginning of the season (]\Iay and June) work is carried on both day 

 and night, but later the ants come forth only at night, except on cloudy 

 days, after a rain the night before, when a few individuals may 

 occasionally be seen outside the nest. (August 29 e. g.) On July 

 24 at 8 130 A. M. I saw a few ants at a single nest, and these all seemed 

 to be coming in. August 5 at 9 P. M. I made the rounds of five nests 

 and found individuals abroad at three of them. At one they were 

 out in large numbers. The light of my lantern threw them into great 

 excitement. 



"During 191 1 the marriage flight took place in June. On July 

 22 I found nest 8 which I believe had been excavated by a queen 

 fecundated during this summer. Very little excavating was done after 

 July I. 



" The surface portion of the ohscurior nest is typically a cres- 

 centic crater, several inches high at its highest point, with the 

 entrance corresponding to the center of a circle of which the crescent 

 is an arc. Nest no. 4, which had a circular crater, and nest no. 3, 

 with a simple conical crater, were exceptions, or rather variations 

 from the type. The entrance is usually concealed under vegetable 

 debris, as is often the case in nests of other fungus-growing ants. 

 The number, shape and size of the chambers and the length, direc- 

 tion and method of branching of the galleries arc very variable, as 

 will be seen from the accompanying figures. I give herewith a table 

 of dimensions of the chambers (length, breadth and thickness) and 

 of the galleries (length), together with the depth of the floor of the 

 lowermost chamber below the surface. The chambers and galleries 

 are numbered in sequence as in Wheeler's paper, " The Fungus-grow- 

 ing Ants of North America." My measurements in the field were 

 recorded in the English instead of the metric system, but in the table 

 these measurements have been reduced to millimeters, so that they 

 may be readily compared with those in Wheeler's table. The follow- 

 ing notes on the individual nests are added as an aid in interpreting 

 the figures of the plate: 



"Nests I and 2. — (May 31.) At the foot of a sandy knoll. The 



