310 PROCEEDINGS OF THE ACADEMY OF [April, 



SO carefully by Britton and Michaux. In order to correlate the 

 different varieties with one another, it is necessary to ask tAVO ques- 

 tions: Is the leaf form an arrested one, or does it represent an 

 advanced condition of growth ? I believe that all the forms known 

 can be classified either as arrested, evolved or reverted forms. 

 Before, however, making this classification, it is necessary to state 

 the fact that the oldest and youngest leaf on the same branch 

 show an entirely different form from the intermediate ones, of 

 which the form with four-lobed leaf may be taken as the normal 

 one for our Lirlodendron tuUpiJera^ The fact that the oldest 

 and youngest leaf on the same branch can differ so much from the 

 other ones seems to be almost constant for the full-grown tree. It 

 must also be emphasized that the intermediate leaves have, in- 

 stead of four lobes, sometimes six or even eight lobes as teeth. 



Arrested Leaves. — The oldest and youngest leaves which have a 

 shape somewhat like those of the seedling plant are evidently 

 arrested ones. The primordium of the youngest leaf of a normal 

 branch has been arrested in its development at a certain stage, and 

 therefore the leaf exhibits an evident often extremely different 

 configuration. 



Reverted Leaves. — The gigantic leaves from the sprouts (meas- 

 ures below) evidently belong to this category, and are in shape 

 like leaf No. 5 of the seedling tree. 



Evolved Leaves. — The four-lobed leaves, whether provided with 

 deep or shallow sinuses, and the six to eight-lobed leaves referred 

 to above have acquired their different character by passing through 

 a further transformation. In other cases where this rough classi- 

 fication does not apply, the form of the leaf may be explained by 

 the persistence or duration of the juvenile form, which produces 

 leaves scarcely less variable than the others mentioned above. All 

 of these facts have been taken into consideration in making the 

 msasurements. 



In Table III are presented the measurements of two terminal 

 normal branches, the leaves being counted from the base in an 

 ascending direction. The amount of variation is shown by com- 

 paring the leaves of the same position on the two shoots. The 

 statistical study of the youngest, oldest and intermediate leaves of 

 the normal branch brings out quantitatively the effect which the 

 light exercises upon the development of the leaves. That light is 



*Holm, I. c. 



