728 THE AMERICAN NATURALIST. [VoL. XXXVI. 
This augmentation, as has often been shown, takes place in 
Acacias in the passage from the customary seedling leaf to the 
majority of bipinnate adults, a large number of pinnz being 
added before the full-sized leaf is reached. But this mere addi- 
tion of pinnz scarcely seems to be the right interpretation of 
the increase in complexity if one compares the seedlings of 
Liriodendron and Acacia. What corresponds to the increase 
of lobes in the former is not so much the addition of pinnz in 
the latter as the increase in degree 
of pinnation. Starting with a singly 
pinnate leaf, one arrives at a bipin- 
nate form of remarkably fine struc- 
ture in such species as A. decurrens 
Willd. But may there not exist a 
tendency toward triple pinnation 
in some of the forms? This ques- 
tion, suggesting itself, led to a 
rather careful search for evidence 
in its favor, with the result that in 
A. decurrens not a few leaves were 
discovered in which such a tend- 
ency was manifest. Of one of these 
leaves a shadow print is given 
(Fig. 3), showing clearly the third 
degree of pinnation on some of the 
Fic. 1. — Seedling of A. Zerosa, showing basal leaflets of the pinne. How 
abnormal third leaf. (Natural size.) 

much of a prophecy this is of a fully 
tripinnate leaf in certain Acacias of some future epoch no one 
can really tell; the tendency, however, cannot be without 
significance. 
In his paper, which deals with animals as well as with plants, 
Jackson carries the same principle so far involved one step in 
advance in point of application. As a leaf develops base and 
tip first of all, he reasons that any change in shape which may 
add to its complexity will not appear at those points, but rather 
upon the proximal portions of the lamina. This he shows to 
be the case in Liriodendron, citing also several other examples 
in substantiation of his view. 
