486 ants' nests. 



ably the same as that of the earthen mounds in general. It is extraor- 

 dinary that the little paving- stones are placed side by side with great 

 regularity like a street pavement, while the interior of the cui)ola contains 

 no stones whatever. MacCook has even seen upon these mounds stones 

 containing fossil remains and native gold. Mr. Henry de Saussure, of 

 Geneva, made similar observations before MacCooh among the genuine 

 Pogonomyrmex barbatus i. sp., Smith, but did not publish them. 



3._W00D NESTS. 



There are also woodcutters among the ants, and in not a feAv cases 

 the same species knows how to make earthen structures and how to 

 hollow outwjod, as, for instance, our Gampouotus lujniperdus Latr. 



The best woodcutters are those species of the genus GaDiponotm 

 Mayr, which have a short, broad head, rounded ofl'in front, especially 

 the subgenus Colobopsis Mayr. 



These ants frequently bore with their short, powerful jaws into the 

 very hardest wood, and construct secure and elegant labyrinths for 

 themselves in it. This is the case with Camponoius pubeficcns in Wallis 

 and Tessiii, and Camponotns marginatus. The latter bores into the 

 softer layers of the wood when they are somewhat decayed and lets 

 the harder part remain, so that its nests are more concentric around 

 the center of the bough or trunk in their arrangement. I have noticed 

 them in cherry trees and Paulowiiias. 



The smaller and very timid species of Colobopsis build themselves 

 nests in the hardest wood. These nests open outward by only a very 

 few small apertures, which are concealed by the irregularities of the 

 bark of the tree. These apertures are kept closed by the head of a 

 "soldier'' sentinel, who permits only friends to enter. The soldier's 

 head is broadened and rounded off in front, evidently for this very use. 

 The rounded surface (front view in iig. 11, magnified ten times) is 

 rough, of a dull-brown color; the feelers are planted back of the 

 rounded surface, so that the latter presents no hold and blocks uj) the 

 entrance to the nest like a living stopper. I first observed this fact 

 among our Colobopsis fruncofa Spin, at Yaux, Canton Yaud (fig. 13, 

 drawn four-thirds of the natural size), but the similar structure of the 

 head and the habit of living in trees, which characterize the other 

 species of Colobopsis, lead us to infer that they live in the same way. 



Fig. 13 represents a portion of the original piece of a nest of CoJo- 

 bopsis truncata discovered by me in a very hard, dead bough of a pear 

 tree. B is the bark of the pear tree; Ch is the chambers and pas- 

 sages of the nest; O is the exterior opening of the nest; behind it, in 

 the gallery of egress of the nest, stands a Colobopsis "soldier" as a 

 sentinel, keeping the door closed with his head. At W are seen two 

 Colobopsis workers, one hastening toward the door from the outside, the 

 other standing in the nest. The soldier will go back into the nest for a 

 moment in order to let the first worker come in (I have noticed this 



