NATURAL CONTROL. 143 



Careful studies have proved that the 29 species of parasites are all 

 derived from native hosts, which are mainly weevils breeding in 

 weeds and other plants growing normally around the cotton fields. 

 Some of these parasites have lived for many generations on certain 

 common weevils until suddenly some abnormal condition has deci- 

 mated the numbers of the normal hosts or freed the parasites of their 

 normal control, thus upsetting the natural equilibrium between them 

 and their hosts. In this way the parasites have been compelled to 

 seek new hosts, and the presence of the boll weevil in large numbers 

 has led them to attack it. This has been demonstrated by the 

 sudden adaptation of several species of parasites at Victoria, Tex. 

 These were normally enemies of the huisache pod weevil, but were 

 unable to attack this insect in 1907 because of the failure of the hui- 

 sache trees to fruit. Another demonstration of sudden adaptation 

 was found in the sudden increase of attack by Eurytoma tykaermatis 

 following the cutting of a number of weeds in which this parasite was 

 attacking a native weevil. Such adaptability of course suggests the 

 advisability of keeping the weeds in the vicinity of cotton fields cut 

 down during the summer in order to force the parasites to attack the 

 boll weevil. 



After a parasite has once attacked a new host it becomes com- 

 paratively easy for succeeding generations to repeat the attack. In 

 this manner many species of weevil parasites have increased their 

 range of hosts until they have obtained a complete series extending 

 throughout the }'ear. A rotation of hosts has, therefore, been estab- 

 lished. As many as f7 different wee\ils are attacked by one of the 

 species of boll-weevil parasites. To illustrate the value of this rota- 

 tion we have but to quote one of the commonest examples. The 

 strawberry and blackberry bud weevil is very common in the South 

 in the 1 latter half of March and until June. Two of the species of the 

 boll-weevil parasites attack this weevil as soon as it begins to breed, 

 and they are able to develop at least one generation before the boll 

 weevil can begin its attack on the cotton squares. These parasites 

 are found attacking the boll weevil throughout the summer. In the 

 fall they begin to attack certain stem weevils, such as the potato 

 or Solanum stem weevils, and produce a generation during the winter 

 which emerges in time to attack the strawberry weevil. Thus, two 

 generations are developed while the boll weevil is in hibernation. To 

 obtain a practical benefit from this rotation of hosts it is only neces- 

 sary to have a hedge of dewberries or blackberries along the fences. 



It is of extreme importance to know that no matter what exigencies 

 reduce the boll weevils, the parasites, though also reduced in numbers, 

 will still be conserved on their native hosts and will attack the boll 

 weevils again as soon as they become sufficiently numerous. 



The location of the developing stages of the weevil is of much 

 importance to the insect enemies. (See PI. XV.) It has been found 

 that cotton varieties display two distinct tendencies in their response 

 to injury to their fruit. Certain varieties, such as King, Simpkins, 

 and Shine, are distinguished by a transverse ring at the base of the 

 pedicel of the square and bolL When the square or boll is badly 

 injured, the plant immediately cuts off circulation by forming a 

 corky layer at this ring, and the injured member falls to the ground. 

 Other varieties, such as Dickson, Rublee, and in general the cluster 



