ANT COMMUNITIES 



advantage thereof, but doubtless the ant quite under- 

 stands. The other closer is seen reaching up and ad- 

 justing pellets from within. At 7.21 the outside closer 

 goes in at the small opening between the top of the gate 

 and the inlying gravel. At 7.27 I still see, by the mo- 

 tion of the pellets of soil and agitation of the pebbles 

 and occasional glimpse of the tips of antenna? and 

 mandibles, that the final sealing-up of the communal 

 walls is being accomplished within. In a moment all 

 is quiet, and the gate of the emmet city is shut for the 

 night. Externally it now seems to be a simple semi- 

 circular or triangular depression in the gravel armor of 

 the mound. 



The other illustration of the gate-closing habit among 

 ants is drawn from the cutting ants (A tta fervens) of 

 Texas, observed in 1877. [McC. 10, pp. 33-40.] One 

 immense commune was assembled around the trunk of 

 a live-oak tree (Quercus rirens) on a road-side. The ex- 

 cavations from the interior had gathered into a mound 

 twenty-one feet long and about four feet high. This 

 accumulation, called by the natives a "bed," was evi- 

 dently one of many years' standing, and when sub- 

 sequently opened was found to be inhabited by legions 

 of ants from the size of a bumblebee to that of a small 

 garden ant. It seemed incredible that such hosts of 

 living creatures could dwell within such a narrow com- 

 pass and all find nourishment. And to think that 

 they live on mushroom gardening! as we shall pres- 

 ently see. 



My first view of the mound was a disappointment. 

 It was in broad daylight, and not a sign of life appeared. 

 Could it be an abandoned nest? Having satisfied my- 

 self that the mound was inhabited, I arranged for an 



72 



