612 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. 
PRACTICAL SIGNIFICANCE OF DIMORPHIC BRANCHES. 
Attention has been called in a previous publication ‘ to the bearing 
of dimorphism upon cultural problems. After a further opportunity 
of studying the habits of branching it is still more apparent that the 
previous estimate of the practical importance of dimorphism is 
justified by the facts. 
The cacao tree takes widely different forms, and these can be 
controlled by pruning. Left to themselves, most of the trees will 
produce an open crown formed by a succession of upright shoots and 
whorls of lateral branches. But some of the trees show a thick, 
spreading crown of secondary lateral branches. Such trees are pre- 
ferred for cultural reasons, because they are much more productive. 
Though now their occurrence is determined entirely by accident, 
there seems to be no reason why they should not be produced regu- 
larly by design. Probably the best course would be to leave only one 
or two of the whorl branches, or at most three. But the experiment 
would be worth making with one, for the whorl joint is essentially 
weak and unsuited to more than temporary existence. With one 
whorl branch developing secondary laterals, as it might be encouraged 
to do by pinching off the terminal bud, a low rounded tree can be 
developed. The formation of uprights would need to be guarded 
against at first, but after a rounded top has been formed there seems 
to be little tendency to produce uprights. 
The still more striking specialization of branching habit in the 
patashte tree serves to emphasize the peculiarities of the cacao tree 
and to explain the nature of the whorl formation, illustrating also 
the contrasting forms of top obtainable by controUing the method of 
branching. But the caulocarpous habit of the cacao tree renders it 
entirely different from the patashte in many of its biological features. 
The classification of trees of such widely different habits as species of 
the same genus tends to obscure the significance of the facts. This 
difficulty is avoided by recognizing the patashte tree as representing 
a genus distinct from Theobroma. Unless such differences of habit 
are clearly recognized there is little hope of applying the facts in the 
solution of cultural problems. 
NUMBERS OF LATERAL BRANCHES. 
Although cacao and patashte have the same method of forming 
lateral branches in whorls at the ends of upright shoots, there is a 
definite difference in the number of branches in a whorl. In patashte 
there are always 3 branches in a whorl, in cacao always more than 3 
(pls. 44-46). The usual number in cacao is 5, but occasionally there 
are 4or6. If 6 branches were the normal number, it could be under- 
‘Cook, O. F. Dimorphic Branches in Tropical Plants: Cotton, Coffee, Cacao, the 
Central American Rubber Tree, and the Banana. U.S. Dept. Agr. Bur. Pl. Ind. 
Bull. 198. 1911. 
