SEASONAL BEARING OF PERENNIAL VARIETIES. 23 



SEASONAL BEARING OF PERENNIAL VARIETIES. 



The continued existence of jDerennial cottons in weevil-infested 

 countries, like Guatemala, proves the presence in these also of means 

 of protection. One of the most important is, doubtless, the produc- 

 tion of an annual crop at a definite season, leaving the weevils with- 

 out opportunity to breed in the intervening months, thus greatly 

 reducing their numbers. 



The popular impression that troi^ical plants take advantage of the 

 continuous summer climate and blossom continuously is correct only 

 for a small minority. Where there are definite wet and dry seasons 

 many tropical plants have alternating periods of growth and rest 

 almost as pronounced as in temperate climates, and even in regions 

 of continuous humidity there are some species which shed their leaves 

 annually and rest for a time. 



A further general reason for a simultaneous annual blossoming of 



all the flowers of a species is undoubtedly to be found in the greatly 



^ increased opportunities of cross-fertilization, just as many insects 



swarm and many birds and mammals collect in flocks before the 



breeding season. Simultaneous flowering is carried to a remarkable 



extreme among the bamboos, where whole species grow for long series 



l, of years without flowering, and then flower and die at once over long 



W distances and in spite of local diversity of conditions which might be 



expected to advance or retard maturity. 



Accordingly, while it would not be reasonable to insist that peren- 



hnial varieties of cotton have adopted the habit of annual flowering 

 pnly because of the boll weevil, the analogy of other plants may be 

 mvoked to show that such a character can be brought about by select- 

 • ive influence. The weevil could certainly assist in the development 

 of such a tendency, especially if there Avere a season of the year in 

 which the insects were less numerous, from climatic or other ext^'rnal 

 causes as yet unknown. 



The tropical varieties of cotton are, as is well known, mostly peren- 

 nial, and some of them develop into trees of considerable size, the 

 trunk attaining a diameter of G or 8 inches, and the main branches a 

 length of 15 or '20 feet. The existence in Mexico of tree cotton 

 immune to the weevils has been reported, but as yet this has not been 

 substantiated. Possibly the weevil has not yet penetrated some of the 

 remote and arid parts of the republic. In eastern Gvuitemala, at 

 least, the tree cottons appear to enjoy no immunity from the weevil, 

 and at the time of the visit of the writer it was often impossible 

 to secure uninjured ])olls, even as samples of the varieties. The 

 native cottons of the island of Cuba, according to Mr. E. A. Schwarz, 

 also have the habit of annual blossoming, in the intervals of Avhich 

 the number of the weevils becomes greatly reduced. The cutting back 



