ABNORMALITIES IN OENOTHERA. 179 



Hilversum. I will add some observations which extend the 

 range of variation in number of parts, and shall then suggest 

 what appears to me a probable explanation of the phenome- 

 non as it occurs in most of the races of Oenothera. All the 

 cases in 1909, with the exception of two, occurred in hybrid 

 0. Lamarckiana from various sources, and these two were 

 O. brevistylis plants from a cross with 0. Lamarckiana. 

 Whether this is of significance as indicating greater variabil- 

 ity in plants derived from a cross, I cannot say, but it seems 

 not improbable that this is the case. Using signs for the 

 flower parts I shall now give the formulae for the flowers 

 recorded in 1909, in which K=sepal, C=petal, S=stamen, 

 N=stigma lobe. 



1 K 7 C?Si5Ni3 6 K5C5S10N 



2 K^CeSisNn 7 KeCeSuN* 



3 K7C7S16N13 8 K7C7S14N9 



4 K5+C7S14N12 9 K 7 C7Si6N 1 74-5 



5 KsCsSeN 10 KsCsSeN 



11 KsCsSeN 



This list of flowers, while shorter than that of DeVries, 

 extends in both directions the range of variations observed 

 by him. The highest number of stamens recorded by De- 

 Vries in a flower is 14, while I observed one remarkable 

 flower with 16 stamens and two styles which were separate 

 at the top, in contact below, but terete and easily pulled 

 apart. The hypanthium was also considerably flattened and 

 much thicker than usual and even had a longitudinal groove 

 down its center. This flower gave me the clue to the explana- 

 tion of these cases of polymery. 



DeVries states (1909, p. 483) in regard to his cultures of 

 0. Lamarckiana forms, that "trimerous flowers are certainly 

 not present," but he has observed them, though very rarely, 

 in 0. biennis and in hybrid cultures. In the season of 1909 



I observed three such cases, in O. Lamarckiana of various 

 descent. The flowers were normal in every way except that 

 the parts were in threes, which made the flowers smaller, 

 though the individual organs were not reduced in size. 



1 This flower had two independent stigmas, and styles which were 

 merely in contact in the calyx tube. 



