244 PLANT-BREEDING 



the young seedlings of the stocks to the shape of the flowers 

 is only an instance of the general phenomenon of association 

 and so it is with the" marks which enable Burbank and Nils- 

 son to make their selections. 



These coincidences are called technically, correlations, 

 and it is these which I have chosen for the subject of this 

 chapter. 



Some of them are of a simple and obvious nature, and 

 their internal cause may easily be understood as soon as the 

 association has been pointed out. Such instances may serve 

 as a basis for further discussions, and become a guide into 

 the more intricate cases. 



It is a common experience that many color-varieties of 

 plants may be distinguished by their seeds. This rule holds 

 good for garden plants as well as for large crops. A dark 

 color of the seed indicates a bright flower, a pale seed is 

 usually associated with white or pale corollas. Stocks have 

 grayish, brown, or bluish seeds according to the color of their 

 petals. Among the lupins the red and blue forms may often 

 be distinguished by their dark seeds, but the white variety 

 has white seeds (Lupinus angustifolius). White vetches 

 have, as a rule, yellow or greenish seeds, in contrast to the 

 dark seeds of the common species. The true opium poppy 

 has white flowers and pale seeds, and many other garden 

 varieties of this plant differ in their flowers and seeds in a 

 more or less corresponding degree. 



In all such cases there can be no doubt, that the coinci- 

 dence is a real correlation, and that the cause, which darkens 

 the flowers is the same as that which is active in the seeds. 



Berries often show the same correlation to the flowers. 

 If a species has red or blue corollas combined with dark 

 berries, its white variety will often show pale or even white 

 fruits. One of the most interesting instances 'of this rule is 

 the pale variety of the belladonna, a very poisonous plant, 



