32 



Characters commonly used to separate the sexes in the family Cur- 

 culionidse are not distinctive in this species. As a rule the antennae 

 are inserted nearer the tip of the snout in the male than in the female. 

 This character is variable among boll weevils; and though a large 

 number of accurate measurements might show that a slight difference 

 generally exists, it is too inconspicuous a character to be of general 

 use. With most species the top of the rostrum of the male is rougher 

 than is that of the female. However it may be with other species, 

 there is but little if any difference in this respect between the young 

 adults of the boll weevil. As the individuals become older the greater 

 activity of the females serves to wear the roughness from the top of 

 the rostrum, and thus gradually, as a result of different habits, this 

 character becomes more distinctive. In less than half of the boll 

 weevils, however, is this character sufficiently noticeable to separate 

 the sexes. The terminal segment of the abdomen shows no external 

 difference in either sex, although in many weevils important charac- 

 ters are there found. 



PROPORTIONS OF THE SEXES. 



No reliable secondary sexual characters having as yet been discov- 

 ered, the certain determination of sex therefore rests solely upon the 

 primary characters, thus requiring a certain amount of dissection in 

 each case. Such determinations have been made upon large numbers 

 of weevils taken in the held and upon many bred in the laboratory at 

 various reasons of the year. The results are briefly summarized in 

 Table VI. 



Table VI. — Proportions of the sexes. 



Number X ^^ r 



Season of 1902. both bred and from field - 



240 



260 



Hibemarerl weevils. 1902-3 



269 



174 



First generation. 1903 _. 



Bred weevils. 191 S 



43 



45 



32 



33 



Field weevils, midsummer, 1903 



52 



59 



Total 



649 



558 







From these 1,207 determinations it appears that males are somewhat 

 more numerous than females, the percentage being nearly 54 of males 

 to 46 of females. It is noticeable, however, that the only season at 

 which a preponderance of males occurs is during late fall. If we 

 exclude the figures for hibernated weevils for a moment, we find that 

 the totals for the balance of the season are remarkably close for the 

 two sexes, being 380 males and 8^4 females. It seems safe to say, 

 therefore, that the sexes are practically equal in numbers except that 

 more males than females seem to be found among hibernating weevils. 

 It may be that the retardation of development due to approaching 



