6 PLANT-BREEDING 



their average. The processes by which new qualities are 

 produced must be studied separately. Under the assump- 

 tion that these processes are neither slow nor invisible, but 

 consist in leaps and jumps such as are popularly indicated 

 by the name of sports, they are now called mutations, and 

 this great subdivision of the phenomena of variability is 

 designated, in consequence thereof, as mutability. 



Darwin was well aware of the existence of different cases 

 of variability, and of the possibility of their bearing on the 

 theory of evolution. He considered the assumption of an 

 origin of species in nature by leaps and sports, such as were 

 observed to occur among horticultural plants. He pointed 

 out that the affinity of closely allied species can be explained 

 on this assumption as well as by slow changes. If we con- 

 sider all the varieties and subspecies of apples, or beets, or 

 of one of the cereals, and assume thousands of years for 

 their production, the changes may have been brought about 

 by rare sports as well as by long continued changes; the 

 effect, at the present time, would be the same. Darwin 

 agreed that this possibility could not be denied and that it 

 was a very weak point in his hypothesis of slow evolution. 



The mutations must not be assumed to be considerable 

 changes. From a study of the differences among small 

 species, we may form some conclusion as to their probable 

 size. Common observation shows the difference between 

 allied species, ordinarily, to be quite striking; but a little 

 discussion and a closer inspection will easily prove that, in 

 such cases, the differences are due to more than one, and 

 often to numerous, characters. In groups (such as bram- 

 bles, roses, buttercups, willows, and many others), where 

 large numbers of species are closely allied, the differences 

 between any two of them become smaller, and, the number 

 of distinct forms increasing, the distinction, in the end, may 

 become reduced to one single differential mark for each two 



