220 THE ORIGIN OF SPECIES [pt. ii 



from another having taken place gradually, whether by in- 

 finitesimal stages, or by small mutations, unless there be some 

 at present inscrutable law that determines that such shall be 

 the case. It will be much safer for the present, at any rate, to 

 leave out of account such a su])position, and to work upon the 

 idea that the whole distii\ction of a species may appear at once. 



Now the ncAv and distinct forms that ha\-e come into existence 

 range from the minute varieties of the Drabas and Hieraciums 

 of northern Europe to differences of well-marked Linnean- 

 specific rank, and one must therefore suppose that mutations 

 giving rise to such forms may be of similar variation in size. 



It must not be supposed that this is being laid down as an 

 absolute rule, but it would seem probable that it is a very 

 general one. Individual forms may owe their origin to many 

 causes, but in most cases it would seem to have been due (im- 

 mediately) to a mutation small or large, which differentiated the 

 new form from its predecessor, but there seems no reason to 

 suppose that the new form is necessarily better adapted than its 

 predecessor, and Avill kill it out in competition. The widety dis- 

 tributed, and presumably parental, Ranunculi of New Zealand 

 are just as common in the south of South Island, where there is 

 such a mass of endemics (fig. on p. 156). 



Natural selection comes in, not as a causative and positive 

 agent, but as a destructive and negative one. The new form will 

 instantly have a most strenuous struggle for existence, so that, 

 if not perfectly suited to the conditions that obtain upon the 

 spot where it is born, and at the moment of its birth, it will be 

 remorselessly killed out. If it passes successfulh^ through this 

 competition, it may be regarded as eminently suited to that spot 

 and those conditions, and may then spread as long as it can find 

 suitable conditions into which to travel. Not infrequently it will 

 meet with conditions that suit it even more perfectly than those 

 to which it was born, and we shall be liable to imagine it specially 

 adapted to them, when really it is only they that are suited to 

 it. Actual experience of the great changes in climatic conditions 

 that go on from year to year shows that most species are really 

 suited to a somcAvhat wide range of conditions. This being so, 

 there is little reason Avhy the child should suppress the parent in 

 competition. The latter will have proA^ed its suitability to the 

 conditions, and will probably have a much wider range, and the 

 chance of a direct and severe struggle between the two is but 

 small. Even if the child should suj3press the parent in portions 

 of its range, it will not be likely to overtake it o\'er the whole, 



