486 GENETICS 



than one character, it is easy to see that great variation can occur 

 in a heteroz3'gous population. Recombinations of genes are not, 

 however, changes in genes. Gradual progress or evolution could 

 not continue by that method. The possibilities of gene combina- 

 tions are limited by chance, and even if certain types were better 

 suited to the environment, the sorting of genes would always give 

 again those less well adapted. 



In the case of a pure line, the individuals are homozygous and 

 variation does not result from new assortments of genes. Differ- 

 ences between individuals in a pure line are the result of the 

 environment in which the individuals develop. When the fact 

 that the individuals of a pure line are homozygous is understood, 

 the failure of selection becomes very clear. Environment does 

 not change the genes when it limits or obscures their presence, as 

 in this kind of variation. These variations that are not inherited 

 are known as fluctuations. 



More and more is becoming known about another kind of varia- 

 tion, which is called a mutation. A mutation is a variation that is 

 apparently the result of a change in the genes, or Mendelian units, 

 because it is inherited. Many cases of mutation have been care- 

 fully studied in Drosophila and other organisms. They are inher- 

 ited according to the expected segregation and recombination 

 of genes in the germplasm. While a variation of this kind may be 

 particularly conspicuous with reference to a certain character, it 

 is often found to affect other characters as well. Mutated genes 

 may be dominant, recessive, or without dominance, as are other 

 genes. It is obvious that mutations are the kind of variations by 

 which evolution can occur. 



The question of how genes become changed to produce mutations 

 naturally arises. There is some evidence that mutations may 

 result from changes in the position of a gene in the chromosomal 

 group. Again, atypical distribution of chromosomes is often 

 associated with mutations, as when an organism has one more 

 chromosome or twice as many as the normal specimen. These 

 facts make it appear as if the unusual appearance of inherited 

 characters were closely bound up with a particular arrangement 

 of the interacting set of genes. Any disturbance in location or 

 amount upsets the end result. These disturbances are, however, 

 of a permanent kind. Whatever produces them may duplicate 

 the effect manv times. That is, in a normal stock the same muta- 



