246 THE ZOOLOGIST. 



most of them) the selection is not natural but unnatural, and 

 (as in the case of the long-legged game-fowl, which will drop 

 back in one generation to the short-legged type) these variations 

 do not, therefore, become properly fixed but naturally tend to 

 variate back at once to a normal condition. Most of these forms 

 are monstrosities, and only come into being through unnatural 

 interbreeding (with subsequent sterility), against the tendency 

 through natural evolution to revert back from the monstrous to 

 the correct type. 



If unnatural evolution can cause a variant to thus pass on 

 its characteristics, it is not perhaps unreasonable to suppose 

 that selection according to the requirements of Nature (whether 

 Lamarckian or Wallacian) may obtain much better results, and 

 pass on its characteristic as unvaryingly as any mutant. 



I do not know if Darwin is right in stating that variation 

 gives rise slowly to species. I cannot see any reason why it 

 should not. The matter appears to be by no means settled. 

 Because species have been proved to arise through mutations, 

 this is no evidence that they may not arise by variation. 



Some authorities, among them Professor MacBride,* appear 

 by no means to accept the theory of mutation in toto, and state 

 that the celebrated example of De Vries, the evening primrose, 

 and on which the chief arguments in favour of his theory are 

 based, may be an introduced form and is strongly suspected of 

 being a hybrid, and thus appear to discredit the theory of 

 evolution by mutation. 



As far as I can gather there do appear to be mutants, causing 

 not merely new varieties, such as the white-bellied type of Mas 

 rattus, Long-haired Cavies, &c, but even more, even indeed a 

 species (?), at one bound, infertile with the parent form. 



Dr. K. E. Lloyd,t writing on the researches of Mr. W. L. 

 Tower regarding the Potato Beetle of America, states: "Here 

 the origin of a species was actually observed (Leptinotarsa 

 pallida) " ; and later : "A number of L. multitceniata were allowed 

 to breed in captivity under conditions which were nearly normal. 

 The great majority of the offspring raised were of the kind 

 multitceniata, like their parents, but both L. rubicunda and 



* ' Zoology, the Study of Animal Life.' 

 f ' The Growth of Groups.' 



