516 Varietäten, Descendenz, Hybriden. 



sterility. The flowers of these hybrids had uniformly one half ot 

 their poUen-grains quite empty and collapsed, and one-half of their 

 ovules had no embryosac. Counts gave 3917 perfect pollengrains 

 to 3388 empty grains. (Some of the empty grains are often hidden 

 under the füll ones, or swept aside by the liquid). Counts of ovules 

 gave 50 ovules with embryo-sac to 49 ovules with aborted embryo- 

 sacs. The empty poUen-grains were uniformly mixed with the füll 

 grains in the anther; and the sterile ovules were distributed at 

 random in the ovaries. The sterile ovules do not grow in size, and 

 the place where they are can usually be seen as a constriction from 

 the outside of the pod. Only about one or two percent of the pods 

 of these hybrids have no aborted ovules; some pods have several 

 aborted ovules, some have few, but the commonest number are 

 two or three. 



In the second generation half the plants have perfect pollen- 

 grains, and the other half have a mixture of equal numbers of füll 

 and empty grains in all their flowers. The plants with perfect poUen 

 have also perfect ovules; the plants with semi-sterile pollen have 

 also half their ovules sterile. In the Fo-generation of the Velvet- Yoko- 

 hama cross, which was thoroughly investigated, there were 180 

 fertile plants and 195 semi-sterile plants. In the F^ all the descen- 

 dants of fertile plants showed to be fertile; the progeny of the 

 semi-sterile plants again separates into equal numbers of fertile and 

 semi-sterile plants. The semi-sterile plants have not been tested 

 beyond this third generation; fertile plants remain fertile in pollen 

 and ovules in F4 and F5. 



The writers conclusions is as foUows: "The random abortion of 

 half the poUen-grains and half the embryo sacs, and the Splitting 

 of the progeny of the semi-sterile plants into half semi-sterile and 

 half fertile, as well as the constancy of the fertile plants, agree in 

 all details with a simple Mendelian hypothesis, in which the pollen- 

 grains and embryo-sacs, not the zygotes, are the individuals affec- 

 ted by segregation. If the Velvet been has a factor whose absence 

 stops the development of those poUen-grains and embryo-sacs, 

 which lack it, and the other three allied beans have another similar 

 factor segregating independently, then those pollen-grains and 

 embryo-sacs of the hybrid which have both factors will be abnor- 

 mal, because unlike the Zygote, they normally have Single and not 

 double factors. Hence those pollen-grains and embryo-sacs which 

 lack both factors, and those which posses both factors, alike fail to 

 develop. This hypothesis can be verified or disproved by appropriate 

 crosses between different fertile lines, which the writer hopes to 

 undertake". M. J. Sirks (Haarlem). 



Castle, W. E., An apple chimera. (Journ. of Heredity. V. p. 

 200-202. 1914.) 



The writher gives in this paper some interesting photographs 

 and descriptions of a peculiar apple chimera; Golden Russet and 

 Boston Stripe being combined in the same fruit, as the result of a 

 graft. Trees producing these apples bear only a few fruits of this 

 composition; the rest of the crop belongs entirely to one or other 

 of the two varieties concerned. The stem end of the apple was in 

 each case of "russet" and without stripes; the blossom end was 

 smooth skinned of a light red color striped with yellow or green. 

 The line of demarcation on the surface was Sharp; there was gene- 



