G2 BOTANICAL GAZETTE [JANUARY 
actual nature of the immediate product of the cross is not known. 
The parental species have undoubtedly sustained similar distribu- 
tional relations to each other for uncounted hundreds or thousands 
of years, and there is no reason to suppose that hybridization may not 
have taken place many generations ago. On this account it is not 
possible to say whether the tree from which the germinated acorns 
were taken was the immediate product of the cross or whether it is 
the nth generation of its progeny. 
In balanced crosses in which the parents show a large number 
of dissimilar characters, the first generation rarely offers the spec- 
tacle of pure dominance of the characters derived from one parent, 
and recessiveness of those from the other parent. It is only when the 
parents differ by a point or two only that such total dominance 
is seen, and the first generation or the immediate product of the cross 
resembles one parent or the other, and its progeny.split in the next 
generation. In cases such as that under discussion, and which is also 
illustrated by the walnuts, the first generation shows a mixed domi- 
nance as well as a possible fractionization of some meristic qualities, 
so that the hybrid appears as an intermediate between the two parents, 
toward both of which its relative position may be variously estimated. 
In the second generation the movements of the recessives coupled 
with the range of fluctuating variability should give a wide diversity 
of types, varying in number with the number of differentiating points 
of the parents of the cross, which may include both parents, the type 
of the first generation, and an intricately interwoven connecting 
series of forms. 
By reason of the number of dissimilar characters involved in such 
a cross, the probability of deriving an individual composed entirely 
of recessive characters, or of the particular combination characteristic 
of either ancestor is very small. A progeny of hundreds of thousands 
of species would be necessary to furnish a series inclusive of both 
ancestors and intermediate combinations. ; 
While it may not be said that any of the plantlets of the progeny 
under observation are reconstituted ancestral forms, yet some of the 
individuals include so many of the qualities of the red oak and willow 
oak that the evidence is overwhelmingly in favor of the conclusion 
that the origin of Q. heterophylia is to be attributed to the hybridiza- 
