MUTATION. 159 



cases they show a mutual repulsion. This shows that such a 

 mutual repulsion is not given by the nature of the genes them- 

 selves, but by some cause outside them. When anything happens 

 which disturbs this repulsion, a male-giving egg may be produ- 

 ced without this gene, and with the birth of a male heterozy- 

 gote, individuals of the new recessive form must necessarily be 

 produced, as all the females are heterozygous. 



No amount of inbreeding, no long series of generations dur- 

 ing which a family of animals has been purely bred, suffice 

 at all to show, that the production of an individual with a new 

 recessive character is caused by a mutation. On the other hand, 

 the fact that at least two cases have been observed in animals 

 of the spontaneous loss of a gene from a gamete produced by a 

 homozygote, make it probable that real mutation, real spon- 

 taneous loss of a gene is a phenomenon, which we can hope to 

 observe now and then in favorable circumstances. 



Some authors seem to think, that in plants it is always easy 

 to distinguish cases of real mutation from instances in which a 

 recessive is produced by a heterozygote. But we must remem- 

 ber that, just as we have a series of instances in animals in 

 which a family may contain heterozygotes, without producing 

 the corresponding recessives, we know a series of instances in 

 plants in which individuals may be heterozygous for even a 

 number of factors without the production of the corresponding 

 recessives. These are those instances in which ordinary mo- 

 nocious plants seem to consist not of one individual but of two 

 individuals rolled into one, which two may each have their 

 own genotype. 



In such cases the pollen may be genetically different from 

 the ovules, so that in crossing experiments the same plant may 

 prove to be very different genotypicaUy according to whether 

 we use it as the male or as the female parent. The best-studied 

 examples are to be found in the carefully planned work of 

 Miss Saunders with Matthiolla. Other examples, which are 

 important for an insight into the mutation question are found 

 in the work of de Vries with Oenothera. De Vries studied cros- 



