86 BRAINERD: HYBRIDS OF THE PALMATA GROUP 
amous flowers ovoid-conical, 5-7 mm. long, or half as long as the 
normal capsule in either parent, containing in the 48 capsules 
examined an average of 43 seeds to the capsule. (PLATE 5, Aa.) 
In June 1906 Mr. W. DeWitte Miller sent me from Plain- 
field, N. J., a plant that he considered this hybrid. Three years 
after transplanting, the rootstock was much branched and 8-10 
inches broad; in the spring of 1909 part of it was broken up and 
ten new plants obtained, which in a year or two became so luxuri- 
ant that half of them were purposely destroyed. From the original 
plant 23 offspring have been raised. Of these, 6 had deeply 
lobed leaves like those of V. palmata; 4 had uncut leaves like 
those of V. papilionacea; and the remaining 13 had leaves some- 
what lobed as in the hybrid. (See PLATE 6.) In I909 numerous 
offspring were raised from 8 of the above 23 plants, with the 
following results: (1) The seedlings from a plant with uncut 
leaves (in one instance 18) had always uncut leaves like the parent. 
(2) The seedlings from a plant with deeply cut leaves (in one 
instance 19) had always deeply cut leaves like the parent. (3) 
The seedlings of plants with leaves somewhat lobed as in the 
original hybrid, in the five instances tested, always gave plants 
of the three forms given by the original hybrid. In more technical 
language, the form of the hybrid leaf displays no dominance of 
the form of either parent but is a compromise between the two 
forms. But when plants having this hybrid form of leaf repro- 
duce, the offspring present variously the two extreme forms of the 
original species and the compromise form of the hybrid, in this 
conforming to the Mendelian law of ‘segregation.’ 
There is also approximate conformity to the Mendelian ratio 
of 1:1:2 in the number of each of the three kinds of offspring. 
The total progeny raised from plants having the hybrid form of 
leaf was 49; of these, 13 had deeply cut leaves (form A); 12 had 
uncut leaves (form a); and 24 had the compromise leaf (form Aa). 
This is as near to the above ratio as is practically possible with 
a total of 49 plants.* The further Mendelian principle is illus- 
trated: that when an offspring of a hybrid reverts to a pure 
character of either parent species, that character continues pure 
in succeeding generations if the plant is self-fertilized. In this 
* The theoretical ratio 12} : 12} : 24} cannot be concretely realized, as the 
number of each kind of hybrid forms must necessarily be a whole number. 
