404 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1937 



very useful in the identification of individual chromosomes as they 

 pass from one parent to another in hybrids (pi. 30, fig. 1). 



In teosinte, knobs are found only on the ends of the chromosomes, 

 or, as in the perennial species {Euchlaena perennis), they may be so 

 small and indistinct as to be practically nonexistent. In maize, these 

 knobs occur at various locations along the chromosome but they are 

 not found on the ends. 



These two closely related genera, then, are differentiated by the 

 one having terminal knobs and the other having internal knobs on 

 their chromosomes. Now there is no known mechanism whereby 

 the chromosomes of com can mate with those of teosinte, generation 

 after generation, without exchanging segments. If such exchanges 

 have taken place over the years, then no such differences in chromo- 

 some morphology as have been observed should be found. The con- 

 clusion must be reached that no such widespread interbreeding has 

 been taking place unless it is assumed that all of the qualities that 

 constitute a teosinte plant as distinct from those that make com are 

 located in or very close to the knobs. To be even remotely reason- 

 able this assumption would require far more knobs than are found. 

 Further, such repeated interbreeding even with strict selection would, 

 in time, have made the members of the two genera much more alike 

 than they are. 



The third theory of the origin of maize through the hybridization 

 of teosinte and some unspecified grass having some of the character- 

 istics of the sorghum tribe composes some of the conflicts met in the 

 other theories of origin, but raises additional ones. It derives it8 

 support from the obvious fact that the nearest known wild relatives 

 of maize possess only some of the features of maize, lack others en- 

 tirely, and are more highly specialized in stiJl other characteristics. 

 Thus the seed spike of teosinte with its single instead of paired spike- 

 lets and its hardened shell-like outer glume protecting the seed hidden 

 in the hollowed stem, or rachis, is in these respects more highly 

 specialized than the ear of com. Then, too, maize flowers in the long 

 days of summer and is insensitive to alterations in the length of day, 

 whereas teosinte is extremely responsive to day length and flowers 

 promptly when subjected to a short period of daylight. 



Given a parent which would bring into a hybrid some of the char- 

 acters possessed by maize and not by teosinte it would be possible to 

 derive maize within a period of time such that the interfertility of 

 the hybrid derivative with its parents was not lost. It would also 

 be possible to account for some of the chromosome morphology and 

 would explain the great variation and unstable germ plasm of com. 

 Undoubtedly com possesses many characters that relate it to the 

 sorghum tribe, many more in fact than any of its relatives, but the 

 possession of these obvious aflSnities with the Andropogoneae does not 



