BRANCHES AND LEAF-NUMBER. 



115 



value is actually slightly in excess of the observed. In the other three 

 series the small excess of the observed mean is entirely due, I believe, 

 to the fact that in their influence the main-stem and secondary-branch 

 whorls do not exactly balance each other, but on the contrary the main 

 stem preponderates. It has been shown above (table 25, p. 49) that the 

 number of primary branches greatly exceeds the number of tertiaries. 



Table 59. — Observed and predicted leaf-number in "branch-origin" whorls. 



Every primary branch means, of course, an entry of one main-stem 

 whorl in the frequency distributions of table 54, while every tertiary 

 branch means the entry of one secondary-branch whorl (excepting, 

 naturally, in both cases mutilated whorls). Hence if primary branches 

 exceed tertiaries in number we should expect a preponderant effect of 

 main-stem whorls (with high leaf-number) over secondary-branch whorls 

 (with low leaf -number) in the means of table 55. 



Taking all these points into account I think we may safely conclude 

 that the presence of a branch originating a.t a particular node is luithout 

 any influence on the number of leaves in the whorl belonging to that node. 

 This conclusion is confirmed by the data for whorls at which two 

 branches originate, but it hardly seems worth while reproducing the 

 evidence in detail. Strasburger ( :02) states that the same thing is true 

 with regard to the influence of flowers on the leaf -number in whorls. 

 He says (p. 486) : ' 'Die in einem Wirtel vertretenen Bliithen beeinflussen 

 nicht die Zahl der Blatter. ' ' 



