MAKING NEW KINDS OF PLANTS 



431 



with the attached anthers, as shown in Fig. 246. By 



comparing with Fig. 161, one can see just what part 



of the flower is removed. The pistils and 



stigmas are left exposed and uninjured. 



This is done before the buds are open. Mr. 



Burbank finds that the time when the hum 



of the bees is first heard in the trees is the 



best for pollination, as the stigma then 



seems to be in a receptive condition. The 



pollen is applied by simply dipping the 



finger into it and touching it to the stigmas. 



The tree is then left to itself: the bees do 



not visit the flowers : the petals are not 



present to attract them, and there is no 



foothold for them should they come. So 



the manner in 



there is little danger that any other pollen which the pet- 



J 



will be brought. (Where it is important to 

 know the exact parentage, the flowers may 



i ^ A^ ^ T -i T atory to polli- 



be covered with bags, as described on page nation. 

 289: for practical purposes this seems to be, as a 

 rule, unnecessary.) 



The application of pollen may or may not result 

 in the setting of fruit. If the parents are of very 

 dissimilar species, fruit will not set at all: or it may 

 do so occasionally, say once in a thousand times. 

 Sometimes fruit appears to set, but on ripening it 

 is found to contain no seeds: again, seeds are pro- 

 duced which look well but are incapable of germina- 



als, together 



