34 



This would not mean that the game in the district was exter- 

 minated, bnt that it was reduced so low that the amount the 

 tribe was capable of securing was not great enough to support 

 it. With the decrease or migration of the tribe the game would 

 increase. Another tribe with higher capacity for hunting, or 

 with better w^eapons, could have become still more numerous 

 and have reduced the game to a lower number before being 

 compelled to migrate. Even white man with his superior arms 

 never exterminated game when he was wholly dependent on it 

 for existence; it is only when he has other sources of supply 

 that he can pursue it to extermination. 



We do not expect uncivilized men to exterminate their game, 

 so we should not expect insect parasites to entirely destroy their 

 hosts. 



AcTiox OF Predators. 



Predators of many kinds attack the host at every stage of 

 its existence ; they also attack the parasitized and unparasitized 

 in their relative proportions, so that they do not gTeatly disturb 

 the balance. Such predators as mammals, birds and lizards 

 are fairly liberal in their choice of food and seldom show a 

 choice for one particular species. With those species on which 

 they do feed they follow the line of least resistance and take 

 them as they come, the most common forming the larger por- 

 tion of their food. 



Owing to their power of locomotion, especially birds, they 

 cover large areas in search of food; as soon as their food in 

 one district becomes scarce they move off to another. Bates has 

 described how flocks composed of several species of insectivor- 

 ous birds move about the country in Brazil, and I have observed 

 the same thing in Africa and the Malay Islands. Thus preda- 

 tors act more as a movable death-factor ; where the egg-para- 

 sites have been scarce there will they gather together to feed 

 off the larvae, and where the larva-parasites have not been ef- 

 fective there will they feed off the adult. 



In the following table I have confined the action of the pred- 

 ators to the adult stage, but this action would be felt on larva 

 and pupa, but the results would be the same. 



Death Factors Actixg Upon Closely Allied Speciks 



In studying the death factors of two or more closely allied 

 species it is often very difficult to say why one should be com- 



