THE IXSTIXCTIl'E BEHAVIOR OF ANTS. 527 



the colony has been in existence for some years. A decline in the food 

 supply, continuing after the elimination of the larvae leads, as I have 

 observed in artificial nests, to a suppression, first, of the soldier forms, 

 and then of the males and supernumerary females. Thereupon, the 

 smaller workers gradually die off, leaving only the queen surviving as 

 the most resistent and most important individual in the colony. These 

 facts indicate that there is an instinctive regulation of the personnel of 

 the colony, but there are others which point in the same direction. In 

 colonies infested with the Lomechusine beetles, if we accept Wasmann's 

 interpretation, there is an endeavor on the part of the ants to replace 

 the workers, which the parasites have destroyed, by converting female 

 larvae into workers. Like many form regulations, this effort fails, 

 since it results in the nonviable pseudogynes, but it is, nevertheless, 

 sufficiently successful to indicate the presence of regulation. Another 

 phase of regulation is seen when the queen of the colony disappears. 

 When this happens one of the workers may become gynaecoid and 

 assume her functions, as Wasmann and I have found in Polyergus 

 colonies. In a few ants (Lcftogcnys and perhaps Diacaiinna and 

 ChampsomynncA'} this condition has become permanent, so that winged 

 queens are no longer produced. Another case which shows the resem- 

 blance between instinct regulation on the one hand and form regulation 

 and regeneration on the other has been observed by Janet. He found 

 that if the workers of a very young colony be removed, the queen, 

 instead of lapsing into the impassive, egg-laying stage characteristic 

 of her sex in old colonies, proceeds at once to produce and rear another 

 brood, thus restoring or regenerating the lost part of the colony, just 

 as many mutilated animals and plants restore their missing organs. 

 Many cases of parasitism among ants probably depend on the regulation 

 of instincts. The killing of the queen Monomorium by her own work- 

 ers in colonies that have been entered by Wheclericlla queens (Chapter 

 XXVII), and the elimination of the host queen in many other cases, 

 may depend on a tendency to preserve the individual that is able to 

 reproduce on the smallest amount of food. This, I take it, is the sig- 

 nificance of Forel's hypothesis that the hosts prefer the small parasitic 

 to their own much larger queens. Simpler examples of regulation in 

 instinct are seen in the rebuilding in typical form of the disturbed nests 

 and fungus-gardens of ants. In such cases a part of the nest is 

 repaired with reference to the whole, just as if it were part of a living 

 body undergoing regeneration. 



Among recent writers Driesch (1903, 1908) is one of the few who 

 have given some thought to the nature of the stimuli which set the 

 instinct actions going. He distinguishes two kinds of stimuli, the 



