1918] JON ES—HETEROSIS 325 
seeds, pollen from the same plant or another plant of the same 
variety was used. These seeds were then either selfed, or crossed 
with a closely related plant. The colored seeds, however, were the 
result of a cross with a different variety. The two kinds of seeds 
were separated and weighed. It was found that in all the 11 cases 
the out-crossed seeds exceeded the others in weight by percentages 
ranging from 3 to 21. Since the two genetically different kinds of 
seeds developed side by side in the same inflorescence, under as 
nearly the same conditions as it was possible to obtain, such 
increases in weight are surely significant. 
That the increase in weight was a manifestation of heterosis 
and not merely the result of crossing a large seeded plant on a 
small seeded plant, was shown by the fact that where two varieties 
‘were used as pollen parents which differed in size of seed, one having 
seeds twice as large as the other, the crosses involving the large 
seeded plant showed no greater increases than the crosses in which 
the small seeded plant was used as pollen parent. In fact, the 
latter crosses gave rather greater increases. From this CoLtins 
and Kempton conclude that ‘the rate of increase bears no direct 
relation to the size of seed in the variety used as the source of 
pollen” (Joc. cit. p. 11). 
In the experiments of Coxttins and Kempton, reciprocal 
crosses were not made. Although the fact of increased endosperm 
development resulting from cross-fertilization is shown by the 
results reported, still more conclusive evidence has been obtained 
by the writer from reciprocal crosses in maize by the use of similar 
pollen mixtures. A number of crosses were made between types of 
maize previously selfed from 3 to 6 generations. These inbred 
Strains were quite uniform and were derived originally from different 
cultivated varieties. Reciprocal crosses were made, not between 
individual plants, but between the different strains. All of the 
plants of each line, however, were descended from individual 
- plants in the preceding generation and were genetically nearly 
identical. 
Some of the strains had yellow, others white endosperm. 
Either way the cross was made the heterozygous seeds immediately 
resulting from pollination were light yellow, with a more or less 
