tgtt] CURRENT LITERATURE 7i 
believes will go far toward solving the anomalous hereditary behavior of the 
oenotheras. As reported in Die Mutationstheorie (2:471), reciprocal hybrids 
among the oenotheras are Peer unlike, being usually similar to the type 
of the pollen parent. To over the significance of this phenomenon the 
author crossed together the anal hybrids, thus (AX B)F;X(BXA)F,:. 
The results of such a cross he calls “double reciprocal hybrids.” For the 
study of these double reciprocal hybrids he has used chiefly O. biennis and 
O. muricata. It will be seen that there are two possible combinations of the 
same reciprocal hybrids, e.g. (biennisX muricata) X (muricata X biennis) and 
(muricataX biennis) X (biennisX muricata). In the first case the cea 
occupies the middle place in the formula and biennis the extremes, while in 
second case the position of the two parent species is reversed. The sheers 
and unexpected result of these crosses is the complete disappearance of the 
characters of the species occupying the middle position in these formulae. 
only of biennis, while the alternative arrangement of the parents results in a 
progeny of pure O. muricata, the biennis having completely disappeared. 
This remarkable result does not belong only to the biennis-muricata cross, but 
six other combinations in which O. biennis entered as one of the parental 
types followed always the same law. These six combinations involved a small- 
flowering ‘‘O. biennis’’ from Chicago, cruciata, strigosa, Hookeri, Lamarckiana 
laeta, and Lamarckiana velutina. The same principle holds when one of these 
F, hybrids is crossed with one of the parental types, e.g. (AX B)F:XA, the result 
being the same as if the middle parent had not entered into the breeding (in 
the example the offspring are all pure A). The progeny of such a cross the 
author calls ‘‘sesquireciprocal hybrids.” 
From these results it is obvious that the eggs and sperms carry different 
morphological potentialities. The cross AXB results in an AB heterozygote 
which produces A eggs and B sperms. The reciprocal cross produces only 
B eggs and A sperms. DeVries offers the explanation that both A and B 
eggs and A and B sperms are produced, but that the B eggs and A sperms fail, 
an assumption which is in accord with the observation that about half the 
ovules and half the pollen grains are abortive. 
DeVries has gone further and discovered just what — are 
carried by each sex. In both pure-bred O. biennis and in its crosses w 
muricata, the pollen-borne type is epistatic to the seed-borne type, so ace the 
latter is never seen; but in a series of crosses of O. biennis with “ biennis Chi- 
cago,” cruciata, Fisclen. strigosa, and Lamarckiana, the seed-borne type of 
O. biennis is epistatic to the pollen-borne type of the other species, thus allow- 
ing it to become visible. The F, offspring between biennis S and all these 
other species resemble O. biennis, which therefore represents the pollen-borne 
type. The F, between diennis 2 and all of the species mentioned gives a new 
form unlike biennis and unlike any of the pollen parents, but essentially 
identical in all the crosses. This new form is the seed-borne type of O. biennis. 
