23 



tlio Bureau growing- under various conditions of soil, climate, and 

 elevation. The most signiticant conditions were found at San Bartolo, 

 State of San Luis Potosi, Mexico, at the hacienda of Espinosa y Cuevas 

 Hnos., this being the onl}^ locality where tree cotton was found grow- 

 ing for which accurate temperature records were available. A com- 

 parison of these records with the United States AVeather Bureau records 

 at Brownsville, Tex, — which point represents the mildest climate of 

 the cotton belt of the United States — shows that both the minimum 

 and daily mean temperatures of the two places are very nearly alike 

 during the winter months. At the Mexican hacienda referred to, the 

 owners state that the tree cotton was injured by the light frosts dur- 

 ing the winter of 1902-3 to the same extent as was the American 

 upland cotton growing there. An examination of many squares of 

 the tree cotton plants showed that fully two-thirds pf them were 

 infested by the boll weevil. At Cuernavaca, State of Morelos, Mexico, 

 the squares of a variety of cotton known among the natives as Algodon 

 Arbol (cotton tree) were found to ])e badly infested by the boll weevil. 

 At all places where Mexican tree cotton was found entirely free from 

 the boll wee\'il it was undoubtedly due to the nonexistence of the 

 insect in that section. 



The observations mentioned in the foregoing paragraphs lead to the 

 conclusion that there is no variety of cotton in Mexico which is inmiune 

 to the boll weevil. This conclusion is borne out b}- e:^periments con- 

 ducted at Victoria, Tex., with cotton plants g-rown from the seed of a 

 large number of varieties procured in Mexico and Cuba. 



FUTILE METHODS SUGGESTED FOE CONTEOL. 



In some quarters of Texas and Louisiana there is still considerable 

 misunderstanding about the "habits of the boll Aveevil, and many fal- 

 lacious suggestions are proposed from time to time. The supposition 

 exists in many quarters that the boll weevil is attracted to lights. A 

 number of machines based upon this idea have been constructed. The 

 possibility of attracting the boll weevil to lights was one of the lirst 

 matters relating to the pest to be investigated by entomologists. 

 During September, 1897, Mr. J. D. Mitchell, of Victoria, Tex., 

 a naturalist and cotton planter, set out trap-lanterns in cotton fields 

 in Victoria County for one night. The insects captured were sent 

 to the Bureau of Entomology' for examination. In all 2'1,49!2 speci- 

 mens were taken, representing approximately 328 species. • Divided 

 according to habit, whether injurious or beneficial, the result w^s: 

 Injurious species 13,113 specimens, beneficial species 8,262 specimens, 

 of a neuti*al character 3,117. The interesting point in connection 

 "with this experiment was the fact that not a single specimen of 

 the boll weevil was found, although the lights were placed in the 



216 



