THE VARIABILITY IN CULTIVATION OF HARDY FLOWERING PLANTS. 219 



The changes generally attributed to cultivation, of which I 

 propose to speak briefly, in order, are : — 



(1) Changes in stature and luxuriance and size of the 

 flower. 



(2) Changes in the colour of the flower. 



(3) Changes in structure and form. 



(4) Changes in seed-producing habit. 



(5) Double flowers. 



I must repeat that I am not going to discuss what are usually 

 called florists' flowers. It may perhaps be called the first step 

 towards florists' flowers of which I treat, and that step is often 

 the most difficult to make. After the first departure from the 

 type is made in the right direction, diligent gardening generally 

 does the rest. 



Before entering upon details, I wish briefly to notice common 

 sources of error if observations are made casually or carelessly. 

 We are often told of changes in form or colour of flower coming 

 on in a plant, which when investigated prove to be only apparent. 

 I am not speaking of exceptional sports, which may take place 

 from single to double, or from one colour to another, but of 

 alleged habitual changes in a plant itself. For instance, I have 

 spent much trouble during twenty years in satisfying myself 

 that the single wild Daffodil does not change by cultivation 

 into the large double Daffodil, and that the common yellow 

 Primrose does not alter the colour of its flower upon the same 

 plant. 



A common source of error arises from a seed falling and 

 growing up in the middle of a plant of some distinct variety, and 

 producing the type instead of the variety, which it gradually 

 supersedes by more vigorous growth. Thus I have seen 

 Geranium lancastrense seem to change into G. sanguineum, and 

 Campanula, turbinata into a pale blue C. carpatica. Such errors 

 are detected by careful examination. 



(1) To begin with changes in the growth and flowering of 

 an individual plant when brought into cultivation from a wild 

 source. The stature and luxuriance are often increased, but in 

 many instances, especially in plants from high elevations, there 

 is a deterioration of the flower. For example, Banunculus 

 Gouani, which in the high pastures of the Pyrenees is a dwarf 

 plant, the large flowers almost hiding the leaves, becomes in my 



l2 



