36 INJURIOUS EFFECTS OF PREMATURE POLLINATION. 



CONCLUSION. 



The following summary of results obtained with tobacco, cotton, 

 orange, and tomato flowers contrasts the effects of immature and 

 mature pollinations. The orange work does not lend itself well to 

 this comparison, for it is unsafe to say that the } r oung pistils were not 

 receptive when pollinated: 



To determine whether cotton bolls would set without pollination, or 

 by the substitution of other substances for pollen was not attempted, 

 but a summary of this work with the other three flowers is as follows: 



Much has been said in apology for the lack of success of attempted 

 hand pollinations, advocating perseverance, and regarding it as a great 

 accomplishment if even 5 or 10 per cent of the flowers set seed. But 

 this is erroneous, for if one is working with plants that naturally 

 cross readily and is not trying to make a cross between two very 

 different plants, the results should certainly not be considered suc- 

 cessful unless a greater percentage of fruits is obtained than set on 

 the same plants naturally without artificial pollination. We increase 

 the production of fruits by giving attention to plants in other respects, 

 and likewise by a careful study of the flowers upon which we work 

 we can obtain greater percentages of fruits bj r hand pollination, 

 including emasculation, than set naturally. After the behavior of 

 tobacco, cotton, and tomato flowers was quite well understood, and 

 they were correctly operated upon for obtaining the best results, the 

 following percentages of fruit were obtained: 



Flowers baud pollinated. 



40 tobacco (experiments 7. 16, 20, and 36) 



40 cotton ( experiments 2, 4, 7, and 10) 



60 tomato (experiments 3, 4, 7, 9, 13, and 17) . 



Fruits 

 obtained. 



Percent. 



95 

 82 

 86 



