212 LUTHER BURBANK 



of necessity complementary to the capacity to 

 produce tubers. For at least once in my experi- 

 ence a potato plant that by rare exception pro- 

 duced seed developed at the same time some 

 of the largest tubers that I have ever seen. 



Nevertheless, there is an association between 

 seed production and development of the root 

 system, as we have seen illustrated. And it is 

 not unlikely that development of the root of the 

 horse-radish may have had an influence on its 

 seed-bearing capacity. It may be recalled that 

 the carrot and parsnip which produce roots 

 somewhat suggestive of that of the horse-radish 

 in shape and relative size, are biennials, and do 

 not take on the functions of root and of seed 

 development in the same season. 



The roots are formed in the first year partly 

 at least to supply nourishment for the develop- 

 ment of the stem and flowers and seeds in the 

 ensuing season. 



Whatever the relation between the root of the 

 horse-radish and its lack of fertility, the fact 

 remains that the plant is propagated solely by 

 division, and that hence there is no opportunity 

 for the development of new varieties or the 

 improvement of old ones. Each horse-radish 

 root is in effect a part of an original plant now 

 endlessly divided, and the variation in different 



