Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
INTRODUCTION Recent studies by workers of the Rockefeller Foundation in Mexico and Colombia and of the Committee on Preservation of Indigenous Strains of Maize of the National Academy of Sci- ences- National Research Council elsewhere in Latin America have begun to demonstrate a wealth of diversity in maize not fully appreciated by earlier students. Utilization of this material in plant breeding and other crop improvement work has been aided by systems of classification emphasizing natural relation- ships. The most notable of such investigations is the monograph of Wellhausen et al. ( 45), in which twenty-five races of Mexican maize are described. An important contribution of this study was the demonstration that a natural classification leads immediately to considerations of origin and phylogeny, and that such problems, although difficult, can be analyzed by the techniques of the geneticist, the cytologist, and the agronomist. The success of the Mexican studies has raised the hope that all the New World's maize may be susceptible to similar analysis. In Cuba, as in Mexico, a natural classification of indigenous strains of maize collected for plant breeding work led to the present study of the races of Cuban maize. In 1949 the Atkins Garden and Research Laboratory of Harvard University, near Cienfuegos, initiated a program of corn improvement as part of its continuing interest in local agricultural research. Exploration and testing constituted the primary phases of the Atkins Garden's work with maize. In February and March, I. D. Clement, now director of the Garden, and E. Hernandez X., who had participated in the Mexican program of the Rockefeller Foundation, made fifty- seven collections of maize at nineteen Cuban localities. Her- nandez, in his report to the local director of the Rockefeller Foundation in Mexico (24), briefly described six races of Cuban corn and stated that five of these appeared to be related to Mexican types. In 1952 W. L. Brown made collections of maize on eleven
2 RACES OF MAIZE Caribbean islands for the National Research Council's gene-bank program. His report on the maize of the West Indies ( 8) con- tained descriptions of eight races, five of which occurred in Cuba. At the suggestion of Professor P. C. Mangelsdorf, of Harvard University, the writer began studies of Cuban maize in November 1952. His original purpose was simply to supplement Hernandez' report with rather complete descriptions, additional collections, and perhaps some minor changes in taxonomy. The appearance of Brown's paper, the conclusions of which differed from those of Hernandez in important respects, led to a reconsideration of these objectives. One basis for these differences appeared to be a somewhat vague concept of race in maize which had developed in large part from studies in Mexico, where geographical isolation had permitted the evolution of well defined entities. In Cuba geographical barriers to hybridization, apparently never strong, had been destroyed to a considerable extent by the construction of railroads and highways in the thirty years preceding this study. Thus populations of Cuban maize consisted to a large extent of probable hybrids between races, the pure forms of which had become rare. The problem therefore became one of separating putative pure types from a mass of mongrels, and unless reason- able rules for making such separations were established, the process would be purely arbitrary. The rule proposed in the present paper consists in supplement- ing the morphological concept of race developed in the Mexican studies with that of the breeding population. The general pro- cedure followed has been to examine large collections of maize for extreme forms which could have given rise by hybridization to the mass of intermediate types prevalent in Cuba today. Having determined in this manner the appearance of the pure forms of putative ancestral races, breeding populations of such types were sought in the field. Races consist of such populations, the progeny of which breed true to type in experimental plantings. In case no such populations are encountered, the extreme forms are considered chance segregates. In the present study, therefore, emphasis was placed on studies in the field. Simultaneously, attempts were made to check the hypothesis that recent miscegenation among races has been im-
