INHERITANCE OF WAXY ENDOSPERM IN MAIZE. a7 
ated in excess of three times the probable error being no more than 
could be expected in a group of 28 ears. 
With 12,400 seeds the probable error for a 25 per cent ratio is 
0.082 per cent. We may, therefore, feel reasonably certain that if 
the real ratio deviated from the expected by as much as three-tenths 
of 1 per cent the deviation could have been detected. Furthermore, 
the distribution of the percentages is as uniform as could be expected. 
Measured by the method proposed by Yule (16), the range is really 
smaller than the expected. In other and even larger groups it has 
been shown that there is a deviation of approximately 1 per cent; 
and since in these groups the deviation from the observed means was 
greater than the expected, it becomes almost certain that the devia- 
tions are due to the abnormal behavior of some of the individuals 
rather than to a small and consistent deviation. It would thus 
appear that the present group differed from the preceding by includ- 
ing none or very few of the aberrant individuals. 
SUMMARY OF THE PROGENY OF DH 234. 
The total number of ears representing the progeny of the hybrid 
Dh 234 and expected to have 50 per cent of the seeds waxy is 66. 
These 66 ears had 27,029 seeds with 50.4 per cent waxy, the deviation 
in this case being about 1.9 times the probable error. 
While the observed percentage of waxy seeds for the 66 ears is a 
very close approximation to the expected percentage, we have seen 
that the two major groups, waxy X< horny and horny X waxy, that 
make up this total, differ from each other by an amount that would 
be expected to occur as the result of chance but once in 116 times. 
Had the crosses all been made in one direction—i. e., waxy by 
horny—there would have been a significant deviation below the ex- 
pected number of waxy seeds. Had the crosses been confined to 
horny by waxy, the number of waxy seeds would have been in 
excess. Since the reciprocal crosses were made in approximately the 
same numbers, the fit of the observed percentage of waxy seeds for 
the total to the expected percentage appears good. 
THIRD GENERATION OF THE HYBRID DH 237. 
Four ears were selected for planting from the hybrid Dh 287. 
These four ears are Nos. 1129, 1130, 1131, and 1134 (Table I). 
The last two ears represent crosses between Dh 237 and Dh 234, 
and their progenies are considered separately following the analysis 
of the progeny of ears Nos. 1129 and 1130. These last-mentioned 
ears were borne on the same plant. Ear No, 1129 was the result of 
self-pollinating a plant of Dh 237, while ear No. 1130 was the result 
of pollinating this same plant by mothe plant of the same hybrid. 
89356 ° —19-———2 
