336 SCIENCE PROGRESS 



Prof. Thomson gives a full and excellent account of the facts 

 that led to the formulation of the mutation hypothesis, and here 

 we find the explanation of its origin. I cannot find an instance 

 of an established " mutation " except in domesticated races. 

 De Vries' original case of a " mutating form " was the evening 

 primrose. It was introduced into Europe from America prob- 

 ably during the eighteenth century, so there is no doubt as to 

 its having been subjected to selection by man. All the other 

 instances are similar, and when large variations in wild species 

 are taken and bred from by man precisely the same criticism 

 applies. No one denies that large variations do sometimes 

 occur in races which have not been selected by man, though de 

 Vries was not able to find any among the hundreds of wild 

 plants he investigated. These large variations must throw the 

 organism in which they occur so much out of adaptation to its 

 environment that they must as a rule end in elimination, though 

 it is conceivable that there might be some sudden change in the 

 environment occasionally which would favour the preservation 

 of a large variation in a particular direction should it occur. 

 Changes in the environment are, however, almost invariably 

 very gradual. But as I have already pointed out, man has 

 always selected variability in the animals and plants he has 

 domesticated. He has done more than this. He has selected 

 the character of producing large variations, as large variations 

 have been most easily selected by him, and as he has substituted 

 himself for many other factors in the environment, he has 

 removed that check upon the constant production of large varia- 

 tions which must usually be present under natural conditions. 

 It is thus not surprising that de Vries found large variations in 

 the first domesticated plant with which he experimented. But 

 the selection of large variations by man will not be constant. 

 When he has reached a certain standard he will in certain cases 

 do no more than try to keep up this standard, and he will then 

 reject large variations to some extent. Thus a particular 

 organism will exhibit a tendency to produce large or small 

 variations according to whether it has been recently selected for 

 one or the other character. This may very possibly account for 

 the origin of de Vries' hypothesis that " mutations " appear in 

 considerable numbers in a given race at intervals but that be- 

 tween these " mutating" periods the race remains stationary. 



The application of all these facts and theories about heredity 



