218 E. A. ANDREWS 



of these did not fight much with one another and the termites 

 in 2 were especially mild, refusing to attack members of i at 

 all, while members of i would mildly attack members of 2. The 

 workers in i were small and light colored, those in 2 large and 

 dark colored; possibly the difterences are of specific value or 

 more than differences of age. 



But members of these 2 communities tested with members of 

 7 other communities gave various results, which all showed 

 that I was more vigorous and 2 less vigorous in attack and 

 defense ; but certain colonies were obnoxous to both colonies and 

 certain other colonies not attacked by either. 



Representing by the small numeral the introduced termite 

 and by the large numeral the bowl of aliens we found the fol- 

 lowing interactions of i with 2 — 9, and of 2 to i — 9 commu- 

 nities. 



That is, the termites i showed more hostility than did 2 ; some 

 colonies, 4 and 9, w^ere but little attacked by either i or 2 ; one 

 colony, 6, was violently attacked by both i and 2. 



When 9, which was only noticed and not attacked by i and 

 2, was put with 6 there was a violent fight, in fact one worker 

 of 9 killed two successive workers of 6 when pitted against 

 them isolated in a small arena. 6 was especially aggressive 

 to some others : thus individuals of 8, but slightly attacked 

 by I and 2, were seized by the throat in 6; again 6 attacked 

 9 though I and 2 did not. 



It thus appears that different communities have different 

 grades of hostility, variously directed towards other communities. 



Would not a complete knowledge of the interactions of a 

 group of communities enable one to trace their probable phylo- 

 genetic connections? 



