THEORIES OF DEVELOPMENT 



in plant characters is of the utmost practical im- 

 portance for the plant developer as an aid in fix- 

 ing characters so that they will breed true, often 

 a task of great difficulty. 



This does not matter, of course, in the case of 

 orchard fruits, which are propagated by grafting, 

 and of various bulbous and other plants that are 

 propagated by root division. But in case of an- 

 nuals grown from the seed it is highly important 

 that a new desirable character should be fixed in 

 such a way that it will be reproduced in the 

 progeny. 



Now according to the Mendelian formula, in its 

 simplest terms, where any pair of antagonistic 

 characters that Mendelize are in question, the re- 

 cessive character which disappears absolutely in 

 the first filial generation will reappear tangibly 

 in one in four of the offspring of the second gen- 

 eration, and will be submerged in the germ-plasm 

 of two others of each group of four, the remaining 

 member of the group being a pure dominant. 



To illustrate from Mendel's careful experi- 

 ments, when a tall and a short variety of garden 

 pea are crossed, all the progeny are tall; but in 

 the next generation one specimen in four is short 

 and the other two specimens, while individually 

 tall, have the factors for shortness submerged in 

 their germ-plasm. The short specimen being 

 purely recessive will breed true to shortness ; but 

 the two tall specimens that are mixed will not 

 breed true. 



The same principle holds for any pair of an- 

 [39] 



