SELECTIVE FERTILIZATION. 173 



on the stigmas of each. Only ten hybrids were obtained from 

 330 seeds resulting from the mixed pollen. Likewise Heribert- 

 Nilsson (1920) found that the pollen tubes of CEnothera gigas 

 grow slower in the styles of O. Lamarckiana than do the latter's 

 own pollen tubes. These results were obtained by cutting off the 

 styles close to the ovary at different times after pollination and 

 noting the shortest time after pollination necessary for seeds to 

 set. 



A somewhat different kind of selective fertilization has been 

 demonstrated by Correns (1920) with the dicecious plant Melan- 

 driitm. The staminate plants of the species worked with are 

 heterogamous, and the pistillate plants homogamous, correspond- 

 ing to the sex conditions in animals of the Drosophila type. The 

 pistillate-determining pollen tubes apparently grow faster than 

 those which result in staminate plants. This was demonstrated 

 in the following way. The mature seed-pods were divided into 

 upper and lower halves and the seeds of each grown separately. 

 The seeds from the upper portion resulting from the shorter 

 lengths of pollen tubes gave 68 per cent, pistillate plants, while 

 the seeds from the lower portion gave 56 per cent, of the same 

 type. It should be noted that the conditions are reversed in the 

 case of Melandrium and Zea. In the former the pollen tubes 

 enter the ovary at a common point and are free to fertilize the 

 first ovules they reach. The tubes which grow fastest therefore 

 fertilize the ovules in the upper part of the ovary, leaving the 

 slower-growing tubes to pass on down to the lower part. In 

 Zea each ovule has a separate style so that, the longer the distance 

 to traverse is, the less chance will the slower-growing tubes have 

 of reaching the goal first. 



Correns also found that when pollen was applied in large excess 

 as compared to a moderate application, the pistillate plants result- 

 ing were always in greater proportion. But even when a deficient 

 amount of pollen was applied there was still a small excess of 

 pistillate plants showing that some other selective factor than 

 differences in rate of pollen-tube growth was operating in addi- 

 tion. 



Heribert-Nilsson also obtained aberrant ratios from self-fertil- 

 ized and back-crossed heterozygous red-nerved plants of CEno- 



