51 6 THE WONDER OF LIFE 



conditions of their life, liave some survival value. It is a 

 fact of observation that in many groups of organisms the 

 individuals fluctuate continually in various directions. It is 

 also a fact of observation that some of these variations 

 increase the survival value of their possessors. It is 

 inferred that the cumulative inheritance of these favour- 

 able variations, fostered by selection in any of its numerous 

 forms, and helped by the elimination — gradual or sudden 

 — of forms lacking the variations in the fit direction, or 

 having others relatively unfit, may lead to the estabhsh- 

 ment of new adaptations. The greatest difficulty in this 

 argument is to account for the origin of the fit variations, 

 and this has to be met by the accumulation of observational 

 and experimental data bearing on the origin and nature 

 of variations. It is also necessary to accumulate more 

 facts showing that selective processes — acting directively 

 on fluctuating variations — do really bring about the results 

 ascribed to them. 



(c) The work of recent years — notably that of Bateson 

 and De Vries — has made it plain that besides the con- 

 tinually occurring ' fluctuating variations,' there are ' dis- 

 continuous variations ' or ' mutations,' where a new char- 

 acter or group of characters not only appears suddenly, 

 but may come to stay from generation to generation. It 

 cannot be said that we understand the origin of these 

 mutations, in some of which the organism in many of its 

 parts seems suddenly to pass from one position of organic 

 equihbrium to another ; but that they do occur is indubit- 

 able, and their marked heritability is also certain. Mendel 

 has given at once a demonstration and a rationale of the 

 fact that certain mutations, when once they have arisen, 

 are not likely to be swamped, but are likely to persist. 



