No. 520] 



MENDELIAN PHENOMENA 



l>i:» 



Natural selection may have a very important influence 

 on variations of this kind, since the variations are heredi- 

 tary. The recognition of variations of this type must 

 of course be attributed to Mendel, but de Vries, Correns, 

 Von Tschermak and perhaps some others must be recog- 

 nized as rediscoverers of this principle. 



In the second place, after we have eliminated all varia- 

 tion caused by Mendelian recombination we still have a 

 type of variation which is more and more coming to be 

 called fluctuation, and which is due wholly to environ- 

 ment. Such variations are now believed not to be hered- 

 itary and therefore not amenable to the action of natural 

 selection. In this connection we need only mention the 

 important investigations of Johanssen on beans and bar- 

 ley, 3 the work of Nilssen and his able staff at Svalof on 

 wheat, oats and other species, and the recent classic work 

 of Jennings on Paramecium.* The work of these inves- 

 tigators indicates strongly that selection is without effect 

 on fluctuations due to environment. Dr. E. M. East, in 

 Illinois Experiment Station Bulletin No. 127, reaches 

 the conclusion that no effect of selection has been proven 

 in clonal varieties of potatoes. 



In the third place, de Vries found a type of variation 

 which the cytological work of Gates, 6 Rosenberg 6 and 

 Gager, 7 indicates to be due to irregularity in the distri- 

 bution of chromosomes in mitosis. 



Gates has shown that in the species with which de Vries 

 worked there are irregularities in chromosome distribu- 

 tion. Furthermore, de Vries has shown that his mutants 

 differ from each other in almost every detail, just as we 

 should expect them to do if each of the chromosomes is 

 more or less responsible for the whole process of develop- 

 ment. 



Jena, Fisher, 1903'; Eep. 3d Int. Con. on Genetics, pp. 98-113. 



•B. R. Gates, Science, January 31, 1908; Science, February 12, 1909, and 



