222 



THE AMERICAN NATURALIST [Vol. XL VII 



It is impossible to say whether pollen grains capable 

 of functioning ever result from those divisions where the 

 chromosomes are irregularly distributed. The irregu- 

 larities in reduction do not produce supernumerary 

 pollen grains, such as have been described by Juel and 

 Strasburger for Hemerocdllis, for in all tetrads mature 

 enough to show the separation of the pollen grains I have 

 always counted four. One might expect an increase in 

 number of pollen grains formed by one mother-cell, 

 judging from the grosser manifestations of this abnor- 

 mal factor. Functioning pollen is formed in quantity 

 and no trouble at all is found in securing plenty of selfed 

 seed of the abnormal strain. 



Reduction phenomena in the ovule of the abnormal 

 have so far been given only a superficial examination 

 and the observations are not complete enough to report. 

 Observations on the ripe capsules of selfed plants would 

 lead one to believe that here, as in the case of the pollen, 

 partial sterility is present, due to the abortive develop- 

 ment of the ovules, but the latter is only a supposition, 

 which further cytological study may or may not support. 



Cytological examination of the anthers of the three 

 classes of plants obtained from the abnormal X normal 

 (304 X 402) was made. The conditions in the reduction 

 divisions of the normal and homozygous abnormal segre- 

 gates are identical with those present in the two grand- 

 parents. The heterozygote differs from the pure ab- 

 normal in degree only, having fewer sterile anthers and 

 other abnormalities. Otherwise what has been said of 

 the pure abnormal (300-309) applies also to the hetero- 

 zygote. 



In connection with this cross, it is interesting to note 

 what bearing the nature of the reduction divisions in the 

 ovule (300-309) might have upon the F 2 ratio. The ab- 

 normal class, although within the probable error, is al- 

 ways deficient. This is true also in the fasciated peas 

 with which Mendel 27 worked and in one race of peas hav- 



