PRINCIPLES AND METHODS OF PLANT-BREEDING 133 



This can best be explained by an example. A tall, green-seeded pea is 

 crossed with a dwarf, yellow-seeded variety. In the second generation 

 after the cross, not only does the expected segregation of tall greens and 

 dwarf yellows take place, but also novelties appear in the form of dwarf 

 greens and tall yellows. There has accordingly been a recombination of 

 the unit characters: tallness becomes linked with the yellow color; dwarf- 

 ness with the green color. From such recombinations the practical bene- 

 fits are derived. 



A large amount of work has been done on the problem of Men- 

 delism. The brief explanation presented aims to give some notion of 

 the intricate combinations that may arise from crosses. Ordinarily, 

 hybridization requires too much time and attention to be practiced 

 by the farmer. The production of new varieties by this process will 

 have to be left largely to scientific men. The practical man can accom- 

 plish most by the use of the second method of breeding, namely, 

 straight selection. 



Straight selection 



Variability the basis of selection. No two persons are exactly alike. 

 Although in general structure there is more or less similarity, indi- 

 vidual characteristics and personalities differ. We learn to recognize 

 persons by their differences, not by their likenesses. Such variability 

 is common to all plants and animals, and is the basis of selection. 

 Without variation nothing could be accomplished. In order to study 

 variation, individuals must, be observed. The differences are not appar- 

 ent by a casual observation of a general population. 



There are two kinds of variations, to which have been given the names 

 " fluctuations " and " mutations." 



Fluctuations. Variations of this kind are sometimes called quantitative 

 or continuous variations. They are called quantitative because they con- 

 form to certain mathematical principles. If an ear of corn that measures 

 exactly 10 inches in length is planted, the offspring are not all of that 

 length, but will range, perhaps, from 7 to 13 inches. If the kernels from 

 a head of wheat that measures 9 centimeters in length are planted, the 

 offspring do not all have heads 9 centimeters long, but may range in 

 length from 6 to 12 centimeters. 



The following tables (Figs. 76 and 77), called distribution tables, repre- 

 sent some of the practical results derived from such study. They furnish 

 an excellent means for explaining fluctuations, as well as other matters 

 that will be referred to later in this lesson. 



