NATURAL CONTROL. 133 



and 1906. The results were published in Bulletin 59 of the Bureau 

 of Entomology from which the following statements are abstracted. 



For the present purposes proliferation may be defined as the devel- 

 opment of numerous cells from the parts of the bud or boll of the 

 cotton plant which are injured by the weevil. It is clearly a mani- 

 festation of an inherent tendency on the part of the plant to counter- 

 act irritation by the growth of large numbers of new cells. This 

 growth usually begins in the layer of cells adjoining the lining of the 

 boll or in the stamina! column of the undeveloped bloom. Part of 

 the formation may project through the rupture made by the weevil 

 or may form a hemispherical mass protruding from the inner side of 

 the carpel of the boll and pressing into the lock. The react ion on the 

 part of the plant begins generally before the egg hatches. In some 

 cases the egg itself may be moved a considerable distance by the 

 growth. In other instances the egg becomes enveloped and the larva 

 emerges in the proliferous mass. Under such circumstances it may 

 be destroyed early in life, although it often makes its way through 

 the mass into portions of the fruit which have not been affected. As 

 the larva feeds it continues and increases the irritation, and the 

 response of the plant is immediate. In this way it often happens 

 that the space the larva has eaten out becomes filled by the proliferous 

 mass, and the pressure becomes so strong that eventually the larva 

 or the resulting pupa or adult is crushed. It is clear from the observa- 

 tions made that it is this crushing effect that destroys the weevil. 

 (See PI. XIV, b.) A number of experiments in which weevil larvae 

 were placed in proliferous tissues showed that they could develop 

 normally upon this modification of their natural food. 



The frequency of the occurrence of proliferation was determined by 

 the examination of 1,870 squares and 2,042 bolls of a large number 

 of American and several foreign cottons. In the case of squares, it 

 was found that in the averages for all seasons and localities proliferous 

 growth followed feeding punctures in 48 per cent of the cases. The 

 highest percentage, 75, was in the case of the Jannovitch, an Egyptian 

 variety. In the case of bolls, proliferation followed in SI per cent of 

 the cases of feeding punctures. It is consequently apparent that 

 proliferation occurs more frequently as a result of feeding punctures 

 m bolls than in the case of punctures in squares. 



No very satisfactory results followed a study of the effect of climatic 

 conditions upon the frequency with which proliferation follows the 

 attack of the weevil. The observations included a number of 

 varieties growing in two localities during two seasons, but there seems 

 to be no special relation between the locality and the season and the 

 number of cases in which proliferation was found. In fact, the 

 maximum percentage of formation of proliferation in bolls and the 

 minimum in squares occurred at the same time in the same locality 

 and with the same variety. 



Table LXII shows the weevil mortality due to proliferation in 

 squares and bolls under natural conditions. 



