28 



Journal of Agricultural Research 



Vol. XIX, No. 1 



D 



^o 



I 



I 



^^ 



/// 



In pure teosinte there is, properly speaking, no central spike, since 

 the last division of the inflorescence gives two equal branches, each 

 bearing 4 rows of spikclets. In the F3 hybrid plants there were all stages 

 from the condition found in teosinte, which was recorded as 4-rowed, 

 to a well-formed central spike; and the number of rows is one of the 

 best measures of the differentiation into a central spike. In many plants 

 the number of rows was greater by 2 at the base of the spike than at the 

 top. In such instances the intermediate odd number was assigned, 

 The distribution (fig. 25) was slightly bimodal, the 

 modes being at 4 and 8, with the mean at 6.6. 



All the significant correlations with this character 

 are coherences, though none are very close. 



It is of interest that the only other tassel character 

 showing coherence with rows in central spike is sec- 

 ondary index. All other tassel measurements might 

 be expected to increase with increased vigor; and since 

 coherences would appear as negative correlations, any 

 tendency for rows of central spike to increase with size 

 would reduce the coherences. Two of the five alicole 

 characters show significant coherences. There is also 

 a significant coherence with number of alicoles. 



IE 



ROWS OF ALICOLES 



The number of rows of alicoles in the pistillate 

 inflorescence is one of the most striking differences 

 between teosinte and maize. In all varieties of teosinte 

 the number is 2. The lowest number in maize is 4. 

 as is characteristic of all 8-rowed varieties. In the 

 large-eared varieties the number reaches 18. All the 

 Fi plants had uniformly 2 rows of alicoles, indicating 

 the dominance of the teosinte character. 

 ^ , „ , ,. , In the second generation iii out of 123 plants also 



Fig. 26. — Rows of alicoles: _ ° _ . . 



frequency distribution had 2 rows (fig. _ 26). This nunibcr is 19 in excess of 

 of plants in F2. Class ^^^ number expccted if the character were behaving as 



value, one row. ^ _ ^ _ '-' 



a simple Mendelian unit. The uniformity of the Fj 

 plants with respect to this character made it impossible to determine 

 correlations, but of the 12 plants with more than 2 rows of alicoles all 

 but I had more than 4 rows in the central spike. 



\^l<v,^<<^(o 



Az^a^fPf cpy/fiafor' 



