5 Palate he 
424 
many cases heavy fruits are produced which, if not attached to a strong, 
woody stem, would be easily broken off. Such is the case in many 
Meliacew, where the fruits often become large, and, when grouped on one 
stem, very heavy. Ficus minahasse (see Pl. XXVI) has a fruit stalk 
which often is 3 meters in length and thickly covered with fruit. No 
small twig-could possibly support such a weight. An enumeration of the 
plants of the Lamao Reserve which possess cauliflory would include species 
in many of the large families. However, instances where the trunk bears 
the flowers or fruit. are not common. Among those observed were 
Eugenia whitfordii, Ixora macrophylla, and a number of species of Ficus. 
Aw undetermined species of the anonaceous Goniothalamus had the short 
flower-stalk springing from a root at the base of tree. Very numerous 
are the trees which exhibit an attachment of the flowers to the smaller 
and larger branches. Another significance ascribed to cauliflory is pro- 
tection of both the flower and the fruit from too much heat and rain. 
Clearings in forest.—Clearings made in the midst of the Anisoptera- 
Strombosia formation often by Negritos, usually: fill up rapidly with 
climbing bamboo and are similar in character to those described under the 
Dinochloa parang. However, no extensive tracts have ever been cleared. 
In Plate X is shown a river terrace from which the second-story growth 
has been taken. Plate XX VII is a view of a portion of this same terrace 
about two years later. The forest was opened sufficiently so that lght- 
loving species could thrive. Many weeds (see p. 395) had gained a foot- 
hold and other species enumerated in connection with the plate have 
invaded this region from an open, rocky terrace, near by. That the 
clearing will quickly recover its normal condition is evidenced by the 
fact that species of the standing trees already have seedlings in the new 
growth. . 
Seedlings.—As most of the data for this paper were collected during 
the dry season, there has been little opportunity to study the early con- 
dition of the seedling process. The tables in which are given the actual 
number of young trees under 4 meters in height show some results. For 
instance, Anisoptera vidaliana in the open places on the rocky terrace 
(Table XIII) show many more young than old trees, while on the more 
closed adjoining hillside (Table XIV) there are fewer young than old 
ones. In Parkia roxburghii the difference is still more striking. ‘The con- 
clusion is that these two trees are exceedingly intolerant of shade. This is 
especially true of Parkia. After the rainy season sets in, many seedlings 
of Anisoptera were observed in rather shady forests, though they were 
always more plentiful in open places. On the other hand, Dipterocarpus 
vernicifluus shows a large number of young trees in the closed forests, 
and corresponding with a less number of larger trees in the open places, 
there are fewer seedlings. During the rainy season many seedlings of 
this tree were found in very dense shade. The large, winged seeds 
germinate lying on top of the ground. In open places, even well stocked 
