COOK—HABITS OF CACAO AND PATASHTE, 611 
MORPHOLOGICAL AND ECOLOGICAL COMPARISONS. 
DIMORPHISM OF BRANCHES. 
The crown of the patashte tree, like that of the cacao, is made up 
of two entirely distinct kinds of branches. These may be distin- 
guished as uprights and laterals. The tree appears to consist of a 
main trunk bearing clusters of lateral branches, but when the stages 
of growth are observed it is seen that all of the primary lateral 
branches have at first a terminal position, standing at the end of an 
upright shoot. (See pls. 44-46.) 
Though the trunk increases in length by the growth of upright — 
shoots, these shoots do not form a continuous axis, but are strictly 
self-limiting. This is because each shoot, instead of carrying up a 
terminal bud to continue its growth, has the terminal bud replaced 
by acluster of buds, and these give rise to a whorl of lateral branches. 
The specialization is very definite. Each of the upright shoots ends 
with a whorl of lateral branches and no lateral branches are produced 
except in this way—in whorls at the ends of the upright shoots. The 
lateral branches are capable of subdivision, but the divisions are 
always of the nature of laterals, uprights never being produced from 
laterals. 
New upright shoots are formed only from dormant buds on the 
sides of the old uprights, below the terminal whorls of lateral branches. 
Thus the trunk is formed by a succession of upright shoots and is 
not only strictly sympodial but represents a very extreme type of 
sympodial structure. 
As a result of this peculiar method of growth the whorls of lateral 
branches, though always formed in terminal positions, are brought 
eventually into lateral positions and appear as lateral clusters of 
branches instead of as whorls. When a new upright develops just 
below a whorl of branches the thickening of the trunk, as it were, 
incorporates the whorl, which remains in its original horizontal posi- | 
tion, or nearly so, whereas when an upright starts several inches 
below the whorl the subsequent enlargement of the trunk throws the 
whorl over into an oblique position. Doubtless the result is influenced 
somewhat by the time when the new upright begins to grow. A 
whorl that had enlarged and formed a thickened woody base would 
be more difficult to push over into the oblique or lateral position. 
EXPLANATION OF PLATES 44-46.—PI. 44, end of an upright shoot of cacao, with a whorl of 5 lateral 
branches, PI. 45, uprights and lateral branches of cacao, the leaves of the uprights with long petioles 
and those of the laterals with short petioles, one of the uprights ending in a whor!l of 6 small lateral branches. 
Pl, 46, upright shoot of patashte ending in a whorl of 3 lateral branches; some of the leaves are removed 
to give a better view of the petioles. 
