376 Ever-sporting Varieties 



cation between them and the thalamus fails, so 

 that they cannot be thrown off when the flower 

 fades away, but remain as small stumps around 

 the base of the more fully converted filaments. 

 This fact would frequently render the enumera- 

 tign of the altered organs quite unreliable. 



For these reasons I have chosen a group of 

 arbitrary stages in order to express the degree 

 of deviation for a given lot of plants. The 

 limits were chosen so as to be sufficiently trust- 

 worthy and easy to ascertain. In each group 

 the members could be counted, and a series of 

 figures was reached by this means which al- 

 lowed of a further comparison of the competing 

 sets of plants. 



It should be stated that in such experiments 

 and especially in the case of such a showy crite- 

 rion as the pistilloid heads afford after the time 

 of flowering is over, the inspection of the con- 

 trolling beds at once indicates the result of the 

 experiment. Even a hasty survey is in most 

 cases sufficient to get a definite conclusion. 

 Where this is not the case, the counting of the 

 individuals of the various groups often does not 

 add to the evidence, and the result remains un- 

 certain. On the other hand, the impression 

 made by the groups of plants on the experi- 

 menter and on his casual visitors, cannot well 

 be conveyed to the readers of his account by 



