34 MISS L. S. GIBBS ON THE FLORA AND PLANT FORMATIONS 
the ombrophobous stage, with the abandoned padi houses above them—proof 
that this area had evidently been worked since his time. I conclude, there- 
fore, that the forest is cleared in a definite rotation in appreciable communal 
areas on a well-considered plan, regulated by the needs of the community. 
This view is held by Whitehead (16. 112), who confirms the definite 
ownership of land and the appreciation of its economic possibilities under 
varying conditions, mentioning, in a specific case, the value of old rice-land 
that had returned to forest during the minority of an heiress, when the 
rotans had not been cut for seven years. 
The primary forest begins at about 3500’, where the party halted for the 
Dusuns to cut water-vessels at a fine clump of bamboo, the last met with on 
the ascent. Entering the welcome shade of the high forest, the beautiful 
spaced trees showed a substage of younger ones, with herbaceous under- 
growth averaging in parts a quarter to over a metre in height. 
Of the trees, Gontothalamus roseus, judging from the number of fallen pink 
corollas, was one of the finest and commonest, whilst Laportea stimulans, 
Tarenna Gibbsie, and the pink-flowered Ardisia oxyphylla, the latter as an 
undershrub, were equally frequent. 
The undergrowth varied from areas covered with the Arum-like foliage 
of Arisema simplicifolia, about a third of a metre high, the quaint 
spathes green splashed with red, and green and white at the throat, to 
others with Ophiopogon malayanus and Schismatoglottis caulescens, the latter 
with a pretty little white spathe and yellow spadix. Farther on, after 
crossing a stream, this undergrowth increased in height, averaging 1 metre, 
and mainly made up of the white-flowered Argostemma hamelievfolium, the 
graceful recurved fronds: of Hlatostema thalictroides, the familiar pink 
Impatiens platyphylla and Hedyotis pulchella. I also saw one beautiful 
group of an acaulescent palm, unfortunately sterile, the large fronds with 
interrupted pinnz being 3-4 metres long. The bed of the Kadamaian 
proved disappointing, as the huge boulders and rocky sides sheltered very 
little growth, no doubt owing to the terrific force of the torrent. But the 
absence of conspicuous bryophytic growth, both by the water and under 
the trees, was as striking as that of epiphytic foliaceous lichens. 
On leaving the Kadamaian, the invisible track we followed passed, still 
rising, over a ridge to the Lobang torrent, where the herbaceous undergrowth, 
chiefly urticaceous in character ( Elatostema rubro-stipulatum being common), 
was not at all thick. On following the Lobang, several enormous squarish 
blocks, like erratic boulders, were passed, Finally working up the bed of 
this torrent, which fell over a smooth greenish rock, apparently serpentine, 
we reached the well-known Lobang rock, another cyclopean block, the inclined 
face of which, blackened in parts by many camp-fires, showed the remains of a 
pudding-stone vein, with rounded pebbles embedded in a cement which almost 
suggested some human agency. This vein of conglomerate would explain 
