44 PERFECT SOCIETIES OF INSECTS. 



rior, and bore a vast number of holes and irregular 

 galleries. That part of the wood appears moist, and 

 is covered with little gelatinous particles, not unlike 

 gum-arabic. These insects seem to be furnished with 

 an acid of a very penetrating odour, which perhaps is 

 useful to them for softening- the wood''. The soldiers 

 in these societies are as about one to twenty-five of the 

 labourers^. The anonymous author of the observa- 

 tions on the Termites of Ceylon seems to have disco- 

 vered a sentry-box in his nests. " I found," says he, 

 " in a very small cell in the middle of the solid mass, 

 (a cell about half an inch in Iieiglit, and very narrow,) 

 a larva with an enormous head. — Two of these indivi- 

 duals were in the same cell : — one of the two seemed 

 placed as sentinel at the entrance of the cell. I amused 

 myself by forcing the door two or three times ; — the sen- 

 tinel immediately appeared, and only retreated when 

 the door was on the point to be stopped up, which was 

 done in three minutes by the labourers." 



I hope this account has reconciled you in some de- 

 gree to the destructive Termites : — I shall next intro- 

 duce you to social insects, concerning most of which 

 you have probably conceived a more favourable opi-* 

 nion ; — T mean those which constitute the second class 

 of perfect societies, whose workers are not larva*, but 

 neuters. These all belong to the H^menoptera order 

 of Linne : — there are four kinds of insects in this order, 

 (which you will find as fertile in the instructors of man* 

 kind, as you have seen it to be in our benefactors,) that, 

 varying considerably from each other in their proceed-t 



* Latr. ///*/. Nat. xiii. 64. " Diet. Hist. NaL xxii. 57, 53, 



