INHERITANCE OF WAXY ENDOSPERM IN MAIZE. 63 
There were 110 ears of the progeny of both the hybrids that were 
expected to be segregating in a 1 to 1 ratio. These ears had 47,872 
seeds, with 49.2 per cent white. Here again the seemingly close ap- 
proximation of the observed percentage to that expected by theory 
is misleading. The deviation of 0.8 per cent is 5.10 times the prob- 
able error and is evidently too large to be ascribed to chance. Only 
4 of the 110 ears deviated from the expected percentage in excess of 
three times the probable error, and with 110 ears 5 such ears would 
be expected. 
The deviation in this case can not be explained by assuming that 
colored seeds are being classified as white, as the colored seeds are in 
excess. It is hardly reasonable to assume that white seeds were being 
included in the colored class. The second-generation plants from the 
colored groups failed to produce any all-white ears, which would be 
the case if the excess of colored seeds resulted from mistakes in classi- 
fication. 
TABLE XXIX.—IJnheritance of aleurone color in all the maize ears expected to 
have 50 per cent of the seeds white. 
Number of Number of seeds. 
ears 
Per- 
Nature of cross. AG z D+E. centage 
elow x= 2 ; ; 
oie ex- | Total. | White.| pected Devige white. 
*|pected. white. 5 
Progeny of hybrid Dh 234: 
TUTTE DGC ALU a eae 35 20 | 138,986] 6,794 | 6,994] —200 5.01 48.6 
a0 La, Ll ee rrr 25 16 | 10,154] 4,907| 5,077| —170| 5.00 48.3 
Progeny of hybrid Dh 237: 
IDG Lt, 19 2 yee eee eee 24 14} 10,929) 5,488 | 5,463) — 25 71 49.7 
MTINOUUION. - se. ose soles. 26 12) 12,803 | 6,457 | 6,401 56 1.46 50.4 
Riptalner cn. 2. ost cosas cides 110 62 | 47,872 | 23,596 | 23, 937 —341 5.10 49.2 
{ 
One parent of each of the ears having equal numbers of white and 
colored seeds must be homozygous for white aleurone, and is, there- 
fore, making white gametes only. When these gametes unite with 
gametes produced by plants heterozygous for colored aleurone, the 
resulting seeds are expected to be white and colored in equal pro- 
portions, but this has not proved to be the case. A deficit too large 
to be due to chance occurs in the number of white seeds. It becomes 
of interest to note that the largest deviations below the expected per- 
centage occur in the progeny of the hybrid Dh 234. This is all the 
more remarkable, since the hybrid in question has been shown to be 
producing an excess of white seeds on the first-generation ears as 
well as on the second-generation ears, expected to have but 25 per 
cent of their seeds white. 
Since the variation in the percentage of white seeds, noted for the 
47,872 seeds forming the total for the 110 ears which were expected 
