64 NII^S HERIBERT-NILSSON 



ENGLISH SUMMARY. 



While investigating the genetic constitution of Oenothera La- 

 marckiana in 1910 I perceived the excellent opportunity offered by the 

 long style in this plant for experiments on the rate of growth of the 

 pollen tubes. In the experiments carried out the following method 

 was employed. The flowers were pollinated and the time noted. At 

 certain intervals the flowers were cut off at the base of the style 

 (see fig. 1). If the pollen tubes had not reached the cut at this time, 

 fertilization failed. The time required for growth through the whole 

 length of the style amounted to 19 hours in the middle of July, while 

 23 hours were needed at the end of the month. Hence it appears that 

 the temperature influences the rate of growth of the pollen tubes. 1 

 showed in another series of experiments that the pollen tubes of 

 O. gigas grew slower in the styles of O. Lamarckiana than O. La- 

 marckiana's owai pollen tubes. 



In 1915 when tabulating the segregation of red-nervedness' in 

 O. Lamarckiana I found that the reciprocal back crosses came out 

 differently with regard to the factor R. The plant is always hetero- 

 zygous as regards this factor, or Rr. Now, if the back cross Rr X rr 

 was carried out a typical mendelian segregation resulted, while the 

 cross rr X Rr always showed a deviation from the ratio expected (1:1) 

 exceeding three times the standard error. The red nerved plants were 

 always in excess of the rest. Assuming a difference in the rate of 

 growth between the ß-pollen tubes (rapid growth) and the r-poUen 

 tubes (slow growth) the results of the experiments become intelligible. 

 In the cross Rr X rr we have only one kind of pollen, r, while in the 

 cross rr X Rr two kinds of pollen, R and r, compete with each other. 

 The /?-tubes are the first ones to reach the ovary, and these tubes 

 fertilize a number of eggs before the r-tubes arrive. 



The fact that the gigas-tnhe^ grew slower in the style of 0. La- 

 marckiana than the Lamar ckiana-lnhes, seemed also to strengthen the 

 belief that the deviation discussed in the above is due to differences 

 in the rate of growth of the pollen tubes. But the whole gigas-p\Eint 

 develops slower and flowers later than the Lamarckiana-planl. Thus 

 the rhythm of development is seen to be the same for zygote and 

 gamete. I have showed that the red nerved plants develop much 

 more rapid than the white nerved; the assumption lies close at hand 

 that also the /?-gametes grow more rapid than the r-gametes. The 

 segregating rr X Rr cross gave the ratio 1,2 : 1. Thus this ratio does 



