94 



VARIATION AND DIFFERENTIATION IN CERATOPHYLLUM. 



on for smooth results. The correlations are all positive, with the excep- 

 tion of that for the ninth whorl, which is negative, but so lov/ as to be 

 entirely insignificant in comparison with its probable error. The data 

 from Series IV lead to essentially the same results as the Series I, II, 

 and III material, and consequently it does not seem necessary to take 

 the space to reproduce the figures in detail. 



Table 47. — Correlation between the first and each of the following nine whorls on 

 primary branches in respect to leaf-number. Series I, II, and III combined. 



We may safely conclude, then, that there is a distinct tendency for 

 branches which show an excess (or defect) from the average in the number 

 of leaves in the first whorl, to have the succeeding tvhorls greater (or less) 

 than their respective averages. The longer the brayich grows the weaker 

 this tendency becomes, till finally, when the ninth or tenth whorl is reached, 

 the number of leaves which it bears is entirely independent of the number 

 in the first tvhorl. Branches which start large maintain an excess over 

 the average for roughly 6 to 8 whorls. So far as our data go they show 

 the same kind of relation for the other axial divisions of the plant, viz, 



