192 GENEKAL BOTANY 



must continually segregate in accordance with Mendel's laws 

 in future generations. Much of the great and pleasing variety 

 in nature would thus be produced, so that many of the so-called 

 varieties and species of wild plants have doubtless arisen by 

 these multifarious combinations and segregations of which we 

 have had a glimpse in the above sketch. 



PLANT IMPROVEMENT BY SELECTION 

 VARIATION AND SELECTION 



We have just learned how Burbank and other plant breeders 

 have been able to produce variations, or differences, in plants 



by crossing and hy- 

 bridizing, owing to 

 the fact that new 

 combinations of par- 

 ental characters are 

 wont to appear in the 

 FIG. 101. Variation in apples offspring of a cross. 



Four varieties of the Early Williams apple, selected Q u t of these differ- 

 from thirty-six varieties borne on seedlings. Modified . 



from photograph by Burbank ing offspring the CUl- 



tivator selects those 



plants which best suit his needs or his desires, and endeavors 

 to perpetuate their kind by vegetative means or by collecting 

 and sowing seed. 



This is the essential principle in all plant improvement by 

 the selection of variations, and the general method is familiar to 

 everyone. It is not always necessary, however, to resort to the 

 tedious process of artificially crossing plants in order to secure 

 the desired variations with which to start a new, improved race, 

 since all plants normally produce offspring which vary among 

 themselves in almost every conceivable manner. In a field of 

 corn, for instance, some plants are tall and others short ; in 

 some the ears are produced well up on the stalk, while others 

 are borne nearer the ground. The ears likewise vary in length, 

 in the number of rows of kernels on the ear, and in the nature 

 of the stored food in the kernel. Some of these differences in 



