Wars and Slavery in the Animal Kingdom. 77 



between the number of masters and of slaves found 

 in colonies of F. sanguined. "The correlation," he 

 says, "existing between the numbers of masters and 

 slaves is as little owing to psychic processes, as the 

 numerical correlation existing between mice and 

 buzzards, or between certain butterflies and cuckoos." 

 Yet, it is hard to see what is proved by this compari- 

 son, unless it be the very contrary of what Bethe 

 pretends to prove; for the more mice there are, the 

 more buzzards will come to the spot, and the more 

 butterflies, the more cuckoos ; however, in the case of 

 sanguined colonies just the reverse takes place, namely, 

 the more masters there are, the fewer slaves they have 

 in their colonies! 



As the sanguineas accommodate themselves to 

 given circumstances regarding the number of their 

 slaves, so also regarding their species. Their favorite 

 slave species is F. fusca. This black ant is found as 

 auxiliary in the greater number of the above men- 

 tioned 410 sanguined colonies near Exaten. In 25 

 colonies the place of F. fusca is taken by a different 

 species, viz., F. rufibarbis; 17 colonies have both 

 species. Near Feldkirch, in Vorarlberg (Austria), I 

 found side by side with colonies which had the above 

 mentioned slaves, others with F. cinerea, or with F. 

 fusca and cinerea. The latter species does not occur in 

 Dutch Limburg, and for this reason no cinereas are 

 found there as slaves in sanguinea nests. Yet, the fact 

 that the sanguineas occasionally invade weak colonies 

 of the large hill ants (F. rufa and pratensis) to rob 

 their pupae and to rear them as auxiliaries, proves 

 that their* "blind instinct" does not force them to rob 



