322 A. J. MANGELSDORF 



use has already been made of the wild forms. The important varieties today 

 are almost without exception complex hybrids that include in their ancestry 

 representatives of both the S. officinarum and the S. Barberi groups of cul- 

 tivated varieties, together with representatives of one or both of the wild 

 species. 



Thus the sugar cane breeder has been exploiting, to the best of his ability, 

 the advantages that heterosis has to offer. He is, however, acutely aware that 

 a better understanding of the genetic basis of heterosis is prerequisite to its 

 more effective utilization. Since he suffers the disadvantage of isolation from 

 the centers of research, he cherishes such rare opportunities as he may have 

 to peer over the shoulder of the research worker, to whom he must look for 

 new facts that may lead to a better understanding of the mechanism of gene 

 action and thus, of heterosis. 



Recently some of us who are engaged in sugar cane breeding in Hawaii 

 formulated a number of postulates with the object of providing a basis for 

 discussing heterosis and related matters. These postulates have been ex- 

 cerpted or inferred from the published literature and from correspondence 

 with workers engaged in genetic research, whose helpful suggestions are 

 gratefully acknowledged. 



Although the evidence supporting these postulates is sometimes meager, 

 and sometimes capable of other interpretations, we have deliberately phrased 

 them in a categorical vein in the belief that they might thus better serve 

 their primary purpose — that of provoking a free exchange of ideas. 



POSTULATES RELATING TO INCIDENCE OF LESS FAVORABLE ALLELES 



1. Naturally self-fertilized populations tend to keep their chromosomes 

 purged of all alleles other than those which in the homozygous condition 

 interact to best advantage with the remainder of the genotype and with the 

 existing environment^ to promote the result favored by natural selection (or 

 by human selection). This does not imply that any single population will con- 

 tain all of the best alleles existing in the species. Selection can make a choice 

 only between the alleles present in the population. 



2. In addition to their prevailing (normal, plus, or wild type) alleles, cross- 

 fertilized organisms such as corn and sugar cane carry in the heterozygous 

 condition, at many loci, recessive alleles which in the homozygous condition 

 would be inferior in their action to that of their normal or prevailing partners. 



3. These less favorable alleles may be thought of as belonging to one of two 

 classes, which, although differing in their past history, may have similar 

 physiological consequences: (a) fortuitous, resulting from sporadic mutation, 

 and representing the errors in the "trial and error" of the evolutionary proc- 

 ess; or {b) relic, representing the residue of what were once the prevailing 



1. The term environment is here used in a broad sense to mean the sum-total of the ex- 

 ternal influences acting upon the organism, including its nutrition. 



