LIXXEAX SOCIETY OF LONDON. 8 1 



mutation by loss of a factor which fulfil this requirement, viz., 

 two in the case of Antirrhinum, two in Avena, two in the mouse, 

 and one in the case of the fly Drosophila. This number certainly 

 is exceedingly small in comparison with the large number of cases 

 reported, and even in these cases I am not yet convinced that 

 tliey cannot be explained by an impurity of the material in 

 respect to this factor. But, as I say, I acknowledge the possi- 

 bility, and, as the question is not vital to my theory, I think w hat 

 I say suffices in consideration of the short time at my disposal. 



With the above restriction, I therefore come to the conclusion 

 that species in the sense I define them are stable for an indefinite 

 time, and that inheritable variation does not exist. 



What then is the 



Reason of the Apparent Variahility of Species. 



Although of a complicated kind, this reason can now be ex- 

 plained in a few words. In the first place, we now know froni 

 Jordan's work that a Linnean species is no unit, but contains a 

 great many micro-species. Consequently, as Jordan points out, 

 the Linnean species is a collection of specific forms which in 

 future ought to constitute subgenera or genera in a new and 

 more scientific classification. 



As such, it can also, without objection in future, for con- 

 venience' sake be retained in systematics ; and I see no objection 

 to continue to speak of, say. Antirrhinum majus, if we only keep 

 in mind that we mean by this an aggregate of micro-species and 

 their hybrid derivatives. 



For evolutionary questions, however, the Linnean species is of 

 no use and its apparent variability a snare. 



I quite agree with Bateson when he says : — 



" When this variability is sorted out, and is seen to be in part 

 a result of hybridisation, in part a consequence of the persistence 

 of hybi'ids by parthenogenetic reproduction, a polymorphism due 

 to the continued presence of individuals representmg various 

 combinations of Mendelian allelomorphs, partly also the transient 

 effect of alteration in external circumstances, we see how cautious 

 we must be in drawing inferences as to the indefiniteness of 

 specific limits from a bare knowledge that intermediates exist." 



It is, indeed, my firm conviction that, with the possible exception 

 of mutation by loss of a factor, no transmissible variation exists. 



One could yet make one objection against this view, based on 

 the occurrence of 



Bud Variation. 



It was this point especially which led Darwin to decide against 

 Pallas's idea that all variability was due to an original cross. 

 My position is that bud-variation cannot serve as such, because 

 there is no proof available that bud-variation ever occurs in a 

 pure species ; all cases reported seem to be explainable on the 

 assumption that they took place in a hybrid, by segregation in 



LINX. SOC. PROCEEUIXG.^. — SESSION 1913-1914. >J 



