1888.] Evolution as illustrated by the Geogr. Distrih. of Plants. 283 



Man, selecting for his own purposes, has probably not succeeded in 

 producing a single new species of plant or animal, and probably not a 

 single naturally permanent variety which might be regarded as a step 

 towards a new species. 



Unless kept under man's incessant supervision our cultivated plants 

 and domestic animals revert to their ancestral form. Pigeons allowed 

 to breed freely among each other revert to the rock pigeon form, dogs 

 mongrelize towards the " dingo " type. 



Not only is this the case, but in proportion as the varieties produced 

 by man's selection deviate more and more from the parent type, so the 

 breed as a rule becomes increasingly difficult to keep up, on account of 

 increased delicacy of constitution and often also diminished fertility. 



The explanation of this is that man tries to produce variations 

 which are not m the direction in which the speuies tends to vary, and 

 the further he proceeds with these variations, which we may call 

 un-natural, the more difficult his task becomes. Not that variations 

 occur less often, for the reverse is the case, but that the variant forms 

 have not those qualities which ensure permanency in the ordinary 

 conditions of life. 



Another experience of horticulturists also bears on the same point,. 



It is well known what difficulty there is to get some wild plants to 

 vary when first taken into cultivation. Some have resisted for 

 generations every attempt to produce desired varieties. At last some 

 lucky or more observant gardener detects in some individual a slight 

 tendency to vary in the desired way. Carefully separating this- 

 individual and breeding from itj he finds among some of its descendants 

 a further variation in the required direction ; after this the task is 

 comparatively easy. The descendants of those plants which for two or 

 three generations have varied in one direction vary freely, at first in 

 the same direction, and afterwards owing probably to correlation of 

 organs in other ways also. Thus illustrating the fact that the 

 tendency to variation is hereditary and produces like effects in mani^ 

 individuals. 



(3.) Another observation tending to the same conclusion is the^ 

 frequent occurrence of well-marked varieties among wild plants 

 growing side by side with apparently identical environment, and the 

 fact that the same varieties also occur independently and in places 

 remote from each other with different environments. 



The collection and classification of marked varieties, especially 

 extreme varieties, sometimes called monstrosities, is at present occupy- 



