350 



The Journal of Heredity 



The development of rhizomes was a 

 constant feature in all the teosinte at 

 Ciudad (niznian. (Fig. 5.) The plant 

 is lUKloiibtedK- jxTennial and ilistinct 

 from all the forms of teosinte previously 

 known. 



Further explorations are needed, 

 especially in Guatemala, before decid- 

 ing whether the Durango and Florida 

 forms should be placed in separate 

 species, but the perennial form is un- 

 doubtedly a new species. 



In the form of the seed the perennial 

 teosinte luore nearly resembles Florida 

 teosinte than that found at Durango 

 or Chalco. (Fig. 6). 



One plant was found with double 

 female alicoles, a persistant rachis, and 

 seeds protruding slightly beyond the 

 glumes, a combination of characters 

 not uncommon in the second genera- 

 tion of maize-teosinte hybrids. (Fig. 

 7.) The discovery of this plant show- 

 ing maize characters makes it seem 

 certain that this new teosinte will cross 

 with maize and produce fertile hybrids. 

 Although teosinte was growing among 

 the maize plants and the two species 

 had flowered simultaneously, this sin- 

 gle plant was the only evidence of 

 hybridization that could be found. 



Ii>' reason of its perennial nature, 

 this new species of Kuchlaena is less 

 maize-like than the annual forms of 

 teosinte and at once suggests that the 

 latter may have arisen through the 

 hybridization of a perennial teosinte 

 with maize. The evidence for this 

 view is altogether too meager to more 

 than raise the cjuestion, but it should 

 be recalled that in the only instance 

 that teosinte was found in a region 

 where the maize possesses distinctive 

 characters, these same characters were 

 found in the teosinte. 



Fortunately it should be possible 

 to gi^■e a definite answer to this ques- 

 tion in two or three years. If forms 

 closeK" resembling the annual t\pes of 

 teosinte can be diAeloped from hybrids 

 between perennial Icosintr and maize, 

 there would seem lillle reason to doubt 

 that wild types of amuial teosinte 

 have arisen in this maimer. On the 

 other hand, if the aiuuial character 



cannot be separated from the other 

 teosinte characters the theory must be 

 abandoned. 



CONCLUSIONS 



Teosinte was found growing wild 

 in three distinct localities: annual 

 forms in Durango and the State of 

 Mexico, and a perennial form in Jalisco. 

 Although occasionally cut for fodder, 

 teosinte was nowhere found to be cul- 

 tivated. On the other hand, it was not 

 found as a wild plant outside of culti- 

 \ated areas. 



The natural habitat of all forms ap- 

 pears to be level, alluvial lands, not 

 subject to extreme drought. The 

 grazing of animals restricts it to pro- 

 tected areas. 



None of the forms of teosinte are 

 common or widely distributed. In 

 the regions where they occur, they 

 attract little attention and then only as 

 a weed in maize fields. 



Although teosinte and maize were 

 growing in close proximit\', there is 

 little e\idence of natural hybridization. 



The name teosinte is frequently ap- 

 plied to species of Tripsacum — a fact 

 that has led to much confusion regard- 

 ing the natural distribution of teosinte. 



There is a perennial form of teosinte 

 that differs from all prc^iousK' known 

 types which are annual by the produc- 

 tion of rhizomes. 



In its seed characters the perennial 

 form of teosinte resembles the culti- 

 vated teosinte of Florida more closely 

 than it resembles the annual varieties 

 of Mexico. 



If all the forms of teosinte are 

 descended from a single stock, it seems 

 reasonable to belie\e that of the known 

 forms the perennial species most closely 

 approximates this common ancestor. 



Since_ the annual forms of teosinte 

 occup\- a position between perennial 

 teosinte and maize and since at least 

 one lorm of annual teosinte is dis- 

 tinguished from the others by charac- 

 ters which it shares with the maize of 

 the region in which it grows, it is sug- 

 gested that the annual types of teo- 

 sinte may have originated from hybrids 

 between perennial teosinte and maize. 



