28 



THE AMERICAN NATURALIST [Vol. XLIV 



The grand total shows that there is a slight empirical 

 mode on four and a conspicuous one on six laminae, as 

 compared with the very prominent modes on these two 

 grades in 1905. The distributions from the three classes 

 of inflorescences shows the origin of these modes very 

 clearly. The mode on four is due entirely to the tendency 

 to the production of one ternately compound and one 

 simple leaf. Both of these types of leaves have been 

 shown to be much more frequent than those divided into 

 two leaflets. The mode on six is due to both of the in- 

 volucral leaves bearing the modal (three) number of 

 leaflets. 



TABLE V 



Showing the Frequency and Mode of Origin of the Different Total 

 Numbers of Lamina per Inflorescence 

 Laminae of Individual Leaves 



In the 1905 series inflorescences with seven leaflets 

 formed only 2 per cent, of the population; in the 1909 lot 

 they are over 17 per cent, of the total number. There is 

 no way of determining absolutely why there is such a 

 difference, but it seems quite logical to suppose that in 

 the Cold Spring Harbor series inflorescences with three 

 leaves were much more abundant than in the Meramec 

 Highlands lot, for the frequency of seven lamina? in the 

 Cold Spring Harbor material is entirely due to flower- 

 ing stalks with three leaves of which two are ternately 

 compound and the third undivided. 



