LUTHER BURBANK 



cross-fertilized and the strains rendered impure. 

 So of course the plant developer has no difficulty 

 in effecting almost any cross he may wish. It is 

 only necessary to take pollen from one flower and 

 deposit on the pistil of another to have reasonable 

 assurance that the cross will be effected. 



But the results of such hybridizing are usually 

 altogether disconcerting. The hybrid progeny 

 seem to branch in every conceivable direction. A 

 gardener of mine declares that hybridized 

 squashes "go crazy", so widely varying are their 

 forms and so little subject to prediction. More- 

 over, it is exceedingly difficult to fix any new type 

 thus developed or to restore an old type thus 

 disturbed by crossing. 



Even if the hybrids do not vary greatly in the 

 first generation they may become entirely chaotic 

 in the second. 



A classical illustration of this is fiu-nished by 

 some experiments of Prof. L. H. Bailey, who de- 

 veloped a variety by crossing that seemed to come 

 reasonably true to type one year. Thinking the 

 variety fixed he sold the seed to a prominent seeds- 

 man, and it was said that the following year no 

 two specimens of the entire lot bore any close 

 resemblance to each other. 



This happened a good many years ago, and was 

 so disconcerting as to lead Prof. Bailey for a time 



[54] 



