198 Double Flowers [CH. 



In many fully double flowers fertility is obviously im- 

 paired through the conversion of the reproductive organs 

 into vegetative parts. It must not however be assumed that 

 the sterility so often accompanying doubling is solely due to 

 this comparatively definite circumstance. In the hose-in-hose 

 Campanula, which has the sepals petaloid, the well-formed 

 anthers contain plenty of pollen (some may be petalodic), 

 but the female organs are in some way influenced by the 

 variation of the. sepals so that they are in some strains 

 sterile. In the double Arabis albida mentioned above 

 there are neither male nor female organs, but the tier- 

 upon-tier structure, which is here the form the doubleness 

 assumes, plainly shows that something more than a simple 

 homoeosis of the stamens and carpels has occurred. 



In the Stock two kinds of doubling occur which are of 

 quite distinct nature. The ordinary double Stocks much 

 used in gardens are fully double, possessing an immense 

 number of petals but no sexual organs, male or female*. 

 Such doubles must therefore, as we shall see, always be 

 bred from singles, a phenomenon which furnishes one of 

 the most curious problems that the study of heredity has 

 to elucidate. Besides these real doubles, plants are occasion- 

 ally seen with one or two extra petals, but the experience 

 of breeders makes it probable that these plants are not more 

 prone than ordinary singles to produce the real doubles. 



In Petunia another problem is presented. The doubles 

 have an immense mass of petals, apparently formed at the 

 expense of the stamens. A few anthers are nevertheless 

 formed which contain good pollen. The female organs on 

 the contrary are abortive and the double flowers set no 

 seedf- 



The case of Begonia offers some points of interest. The 

 plants are of course monoecious and it is only the male 

 flowers which are double as a rule. When the plants flower 

 for their first season the doubleness is frequently extreme, 

 no anthers being formed. But in their second season 

 such plants bear male flowers which contain good anthers, 

 especially, according to the experience of practical breeders, 



* Goebel states that rudiments of anthers are very rarely produced, 

 t See Vilmorin, Fleurs de pleine Terre, 1886, p. 669, and various 

 horticultural authorities. 



