28 BULLETIN 754, U. S. DEPARTMENT OF AGRICULTURE. 
waxy, the deviation of 1.8 per cent from 25.2 per cent being no larger 
than could be ascribed to chance. ‘As long as the deficiency in any 
class of gametes in one sex is made up in the other sex, the discrep- - 
ancy would have to be relatively large before it could be detected in 
self-pollinated progeny. In an ear of 600 seeds, the deviation above 
and below 50 per cent would need to be as much as 38 per cent. 
Summing up the inheritance of waxy endosperm for the third gen- 
eration of the hybrids Dh 234 and Dh 287 and the crosses between 
them, we have 57 ears that are expected to have 25 per cent of the 
seeds waxy. The observed percentage for the total of 25,329 seeds is 
24.6 per cent. The deviation from 25 per cent is 2.09 times the prob- 
able error. . 
Fed, 1902 V1 77 AE 
Fic. 3.—Diagram showing the relations of ears Nos. 1902, 1917, and 1918. 
The second-generation ears of these bybrids, together with 45 ears 
previously published (7), determined the percentage of waxy seeds 
to be 23.7 per cent. The difference between the second and third 
generation ears is 0.9-+-0.21 per cent, which would be expected as the 
result of chance once in 142 times. 
There were seven groups of ears that made up the total of 25,329 
seeds (Table XTIT). Of these seven groups, five were below and two 
above 25 per cent. From this we may conclude that the percentage 
of waxy seeds for the third generation of the hybrids was signifi- 
cantly below 25 per cent, but even with the 25,000 seeds involved, it is 
not possible to determine whether the percentage observed is approxi- 
mating 23.7 per cent, the mean percentage found for the second- 
generation ears. 
There are 131 ears of the third generation that are expected to 
have equal proportions of waxy and horny seeds (Table XIV). The 
131 ears had 57,851 seeds, of which 49.8 per cent were waxy. The 
