2 NUJIBEB or SPECIES. 



thousand species are known. Even this large number 

 18 certainly far short of those actually in existence.' 



I have kept in captivity about half of our British 

 species of ants, as well as a considerable number of 

 foreign forms, and for the last few years have generallj 

 had from thirty to forty communities under observation. 

 After trying various plans, I found the most con- 

 venient method was to keep them in nests consisting 

 of two plates of common window glass, about ten 

 inches square, and at a distance apart of from -^ to 

 I of an inch (in fact just sufBciently deep to allow 

 the ants freedom of motion), with slips of wood round 

 the edges, the intermediate space being filled up with 

 fine earth. If the interval between the glass plates 

 was too great, the ants were partly hidden by the 

 earth, but when the distance between the plates of 

 glass was properly regulated with reference to the 

 size of the ants, they were open to close obser- 

 vation, and had no opportunity of concealing them- 

 selves. Ants, however, very much dislike light in 

 their nests, probably because it makes them think 

 themselves insecure, and I always therefore kept 

 the nests covered over, except when under actual 



' I have had some doubt whether I should append descriptions 

 o£ the British species. On the whole, however, I have not thought 

 it necessary to do so. They are well given in various entomological 

 works: for instance, in Smith's Catalogue of British Fossorial 

 Hymettoptera ; Saunders' Synoims of British Heterogyna ; and in 

 Mayr's Die Ewropdiichen Formioiden, all of which are cheap and 

 easily procurable. I have, however, given figures of the principal 

 •peoies with which I have worked. 



