Feb. 23, 1924 
Teosinte Maize Hybrids 
563 
index, however, is a cause for some misgiving, since it indicates a ten¬ 
dency on the part of the observer to class short, wide-leafed plants as 
crinkly. With this reservation in mind, the biserial correlations of 
the various characters with plant type may be examined. Arranging the 
characters in the order of their degree of correlation with type of plant, 
the ranking shown in Table IV is obtained. 
Table IV. —Biserial correlations of crinkly type of plant with the characters named a 
Rank. 
Character. 
1 
2 
3 
4 
5 
6 
7 
8 
9 
Width index. 
Length of sheath. 
Length of leaf. 
Degree of lobing. 
Degree of color. 
Degree of crinkly. 
Length of longest tassel branch 
Height. 
Width of leaf. 
10 
11 
12 
*3 
14 
X S 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
26 
27 
28 
29 
30 
31 
32 
33 
34 
35 
36 
37 
38 
39 
Length of best inflorescence. 
Days to silk. 
Days to pollen. 
Height of tallest sucker. 
Length of branching space. 
Alicole index... 
Total number of leaves. 
Total sucker height. 
Number double $ alicoles. 
Tassel branch index. 
Rows in the central spike. 
Number of suckers. 
Number of tassel branches. 
Number of spikes in prophyllary. 
Length of central spike. 
Length of glume. 
Leaves above best inflorescence. 
Number of branches on best inflorescence... 
Length of ear stalk of best inflorescence. 
Position of longest leaf. 
Length 9 portion of best inflorescence. 
Number single 9 alicoles. 
Leaves above uppermost branch. 
Number rows of alicoles on best spike. 
Rows in terminal spike of best inflorescence 
Length of best spike... 
Position of best spike. 
Leaves on best inflorescence. 
Central spike index. 
Days pollen to silk. 
Correlation. 
928 
899 
882 
872 
756 
722 
7 11 
628 
619 
497 
495 
473 
458 
452 
448 
445 
444 
442 
426 
404 
334 
33i 
327 
3 l6 
300 
284 
257 
247 
224 
214 
204 
181 
131 
no 
099 
097 
082 
017 
012 
0 Coefficients above .186 are in excess of three times the error. Correlations 26, 32, 35, 36, and 39 are 
disherences. 
The last 8 coefficients in this table may be disregarded, since they are 
less than three times their error. Of the others, 16 are characters which 
enter either directly or clearly indirectly into the classification of type 
of plant. Thus, height of plant doubtless was a factor which influenced 
classification, and for this reason such characters as total sucker height 
and number of suckers (both of which are correlated with height and in 
a sense simply are other measures of height) also are expected to be cor¬ 
related with plant type. Eliminating such relationships as these, there 
are left 16 coefficients greater than three times their errors. A further 
reduction is possible with these 16, owing to the fact that in some instances 
