212 THE LAW OF JOHANNSEN. 



bearing upon the evolution question. Through selection-exper- 

 iments with pure lines there has been discovered what we 

 have called "Johannsen's law," which states that the genes 

 themselves are qualitatively stable. The validity or otherwise 

 of this law, is certainly of the utmost importance for a solution 

 of the evolution-problem. For, if we need not take into account 

 the change of the genes themselves for evolutionary processes, 

 we can limit our research to those processes which are compat- 

 ible with this qualitative stability. For one thing, this does 

 not exclude selection. We know, that ordinarily selection does 

 not affect a pure line, but this does not mean that it may not 

 affect other groupings of individuals. The work with pure lines 

 has taught us valuable things about the nature of genes, and 

 incidentally about pure lines. 



Johannsen showed, that many at first-sight pure species, in 

 reality consisted of mixtures of pure lines. But it does not 

 seem logical, that those authors who have accepted as proved 

 the law of the stability of genes, as demonstrated by the inef- 

 fectiveness of selection in material which is geno- typically pure, 

 and so long as it is pure, should continually be exposed to crit- 

 icisms which are directed against their supposed view, that 

 all species are composed of stable, pure lines. So far as we know, 

 the only author who has proposed a theory of evolution based 

 upon this supposition, is Lotsy. 



The existence or non-existence in nature of "pure lines" is 

 immaterial to the law of Johannsen. We happen to know, that 

 in certain organisms such pure lines do occur in nature, but 

 even if they would not have eixsted in nature, it would have 

 been possible to discover and formulate the law. Even in 

 those plants where no pure lines exist, selection-experiments 

 in artificially self-fertilized series would show a diminishing 

 effectiveness of selection, going parallel with homozygosis, 

 and it would have been possible to deduce Johannsen's law 

 from such experiments. 



The best demonstration of Newton's law is the experiment 

 in which diverse bodies are made to fall in a vacuum. The fact 



