THE ANTS 33 



to a flat upright transverse scale (fig. 6,1) ; in the 

 stinging ants two segments at the base are reduced 

 to nodes (fig. 6, 3). There is an exception in the 



FIG. 6. 



case of one little rare genus, Ponera, which has 

 only the basal abdominal segment reduced to a 

 scale although a much thicker scale than in the 

 others (fig. 6, 2), and yet which has a distinct sting. 

 These arrangements give the body very free 

 movement so that the tail can be bent forward 

 till it reaches the head. Another curious distinc- 

 tion between the stingers and non-stingers is that 

 the larvae of the former spin cocoons and those 

 of the latter do not ; the larvae of Formica fusca 

 occasionally do not do so, but they are an excep- 

 tion to the rule. Cocoon spinning seems to 

 involve the larvae in some difficulties, as without 

 the help of the worker ants they are often unable 

 to extract themselves from their prison. This 

 is a condition which does not, I believe, exist 

 in other groups. In the stingless ants there 

 is a curious difference in habit between the 

 B.W.A. D 



