Page:Popular Science Monthly Volume 82.djvu/47

Rh a race $$Y$$ by three characters, $$i. e.,$$ by the gens $$ABC$$ (for example, $$A$$ $$=$$ long head; $$B$$ $$=$$ awned glumes; $$C$$ $$=$$ rust resistance), to which the race $$Y$$ opposes $$a b c.$$ These are antagonistic characters ($$a$$ $$=$$ short head; $$b$$ $$=$$ awnless glumes; $$c$$ $$=$$ capacity for rust infection). $$A$$ is the antagonist of $$a, B$$ of $$b,$$ etc. But $$A$$ is not antagonistic to $$b$$ or $$c,$$ nor $$B$$ to $$a$$ and $$c$$.

As long as the plant is self-fertilized, the mosaic of its characters is maintained. But if it is fertilized by a distinct race several cases can arise in the course of successive generations. The product called a hybrid ($$F_{1} =$$ filius 1) is evidently the sum of the two parents ($$X + Y$$); if forms not closely related to each other are crossed, the hybrid generally takes a form intermediate between the two parents. We shall not speak of these hybrids here, for they are generally sterile and practically useless for cereal culture. If, on the other hand, closely related forms are fused in the hybrid ($$F_{1}$$) the characters of the father or the mother exclude those of the other parent; one of the parents seems to have been absorbed by the other. Then we say that the character of the father or of the mother dominates or vice versa. Let us take two parents $$X$$ and $$Y,$$ differing in the antagonistic characters $$A\ B\ C$$ for $$X$$ and $$a\ b\ c$$ for $$Y$$. The hybrid ($$F_{1} = X + Y$$) will have the appearance $$A, B, C,$$ if the total gens of $$X$$ dominate those of $$Y,$$ or the appearance $$a, b, c$$ in the contrary case. In other words, one of the parents may seem to be absorbed by the other. But it often happens that if $$A$$ dominates $$a, b$$ dominates $$B, c$$ dominates $$C.$$

But if this hybrid ($$F_{1}$$) is allowed to fertilize itself, its direct descendants, $$i. e.,$$ the second generation ($$F_{2}$$), show that the character or characters which had disappeared reappear in a proportion which can be predicted with almost mathematical certainty. I can not take the time to explain to you the details of this phenomenon. But the most astonishing thing is that among the descendants of the second generation ($$F_{2}$$) (that is, the descendants of the hybrid by self-fertilization) there are (1) those resembling the father exclusively ($$X$$), or the mother ($$Y$$); (2) new forms, $$i. e.,$$ those in which a part of the paternal and maternal characters are combined in a new mosaic.

To choose a very simple example, if the two parents differed by their two pairs of characters $$A B$$ and $$a b,$$ the hybrid of the first generation ($$F_{1}$$) would bear the apparent characters $$A B$$ or $$a b,$$ that is, it would resemble the father or the mother exclusively, according to the predominance; that of the generation ($$F_{2}$$) would comprise individuals of different sorts: $$AB,\ Ab\, Ba,\ ab.$$ The two combinations $$Ab$$ and $$Ba$$ are new.

If, in a second case, the antagonistic gens are $$ABC$$ for ($$X$$) and $$a\ b\ c$$ for ($$Y$$), the first generation might be $$A\ B\ C,$$ but in the second