THE WORLD'S COTTONS — HUTCHINSON 499 



now growing wild in natural vegetation. J. B. Hutchinson [13] has 

 given an account of the '"''dlgodon hrujo'''' of the Salinas de Cabo Kojo 

 in Puerto Rico, and O. F. Cook and J. W. Hubbard [5] recorded the 

 existence of the plant they described as G. inorriUl on coastal dunes 

 beyond the limits of agriculture in northwestern ISIexico. The success 

 of primitive types of cotton in establishing themselves on waste land 

 is such that it would be expected that they would sometimes spread 

 into undisturbed natural vegetation, and Stepliens [17] has given an 

 account of the transition that is to be observed in Yucatan from 

 domestic cottons to the ecotype that is established on the coastal sand 

 dunes of the Progreso area; there, considerable specialization to the 

 exposed maritime situation has taken place. 



In South America also, accounts by E. K. Balls [19] indicate that 

 the cottons found growing outside the present limits of cultivation are 

 escapes and not truly wild plants. Even in the Galapagos Islands, 

 the evidence from both herbarium material and from accessions es- 

 tablished in the living collection is that there is a continuous range of 

 form, from those closely similar to cultivated types in western South 

 America to types evidently closely adapted to survival in the v\'ild. 



The evidence for the view that types of cotton now found wild are 

 escapes from cultivation is so widespread, and the number of forms 

 for which an origin by escape from cultivation cannot be demonstrated 

 is so small, that in 1947 J. B. Hutchinson, E. A. Silow, and S. G. 

 Stephens [15] set out to account for the origin and development of the 

 cottons on the assumption that the whole process had gone on under 

 domestication. This extreme view is no longer tenable [18], but truly 

 wild cottons are so rare that the differentiation of the modern culti- 

 vated cottons is best considered in terms of the cottons that are now, 

 or recently have been, cultivated. Consideration of these early culti- 

 vated cottons leads directly to the study of the origin and spread of 

 the annual cottons. In fact, the success of the cottons as crop plants 

 has depended very largely on the comparative ease with which annual 

 types have been selected from the original perennial shrubby forms. 

 The branching pattern of the cotton plant, with vegetative branches 

 arising from the lower part of the stem and fruiting branches from 

 the upper, lends itself to the development of annual forms. By selec- 

 tion of plants giving rise to fruiting branches low on the main stem, 

 the vegetative period of the plant can be reduced, and flowering 

 brought forward far enough to allow a crop to be matured on a spring- 

 sown plant before the onset of cold autumn weather. 



COTTONS OF AFRICA AND ASIA 



Perennial cottons are confined to areas where there is no winter frost, 

 since all cottons are frost-susceptible. Though they can survive a long 

 dry season, the development of extensive conunercial crops in areas of 



