The Mode of Inheritance of Semi-Sterility in the Offspring of Certain Hybrid Plants. 325 



were observed to fall from the early bunches than from those blooming 

 later. This accounts for the increasing proportion of aborted ovules 

 in the later pods. 



If, to compare with the sections of ovaries, we reckon the aborted 

 ovules in the first five seed-places of these 50 pods, we have: 



Numbers of aborted Observed numbers Calculated numbers 



ovules of pods of pods 



1 7 7-8 



1 1-6 



It is, I think, obvious that, as compared with the ovaries in the flowers, 

 the numbers of pods with more than 2 aborted ovules have diminished 

 excessively. There are 14 pods above, and 36 below the mean. That 

 this is due to selective dropping of ovaries there is little doubt, although 

 it is not possible to verify it by direct observation, on account of the 

 normal dropping of flowers mentioned above. 



The totals were: 



] 2 3 4 .5 



Seeds 34 27 2.5 30 27 



Aborted ovules . . 16 23 25 20 23 



This shows a decrease of aborted ovules at the apex and base of the pod. 

 (If we treat in the same way the 92 pods which were examined 

 from one plant of the Velvet -Yokohama cross, VY 4, we have, for the 

 first five ovules in the pod: 



Aborted ovules Observed pods Calculated pods 



4 2, 14-4 



3 24 28-8 



2 50 28-8 



1 15 14-4 



This shows a similar diminution of pods with more than two aborted 

 ovules in the first five positions, and doubtless for the same reason. 

 There are 26 pods above, and 66 below the mean. 

 The totals of seeds and aborted ovules were: 



12 3 4 5 6 



Seeds 56 GO 47 47 55 39 



Aborted ovules . . 36 32 45 4.5 37 49 



