forms) is common; on the numerous sandstone hills, es- 
pecially on Cerro Chiribiquete and Cerro de la Campana, 
FI, viridis var. toxicodendroides is extraordinarily abun- 
dant; HZ. viridis is found only in the vicinity of Jerijeri- 
mo. The affuents of the upper and middle basins —the 
Ajaju, Macaya, Macayari, ’acunema and Cananari—are 
similarly populated. Inthe lower basin, from the Cachi- 
vera (rapids) de Jerijerimo to the Cachivera de Yayacopi, 
the composition of the Hevea flora resembles that of the 
upper and middle basins; the large affuent, the Pirapar- 
ana, is also similar. From Yayacopi downstream to the 
mouth of the Apaporis, however, Hevea Benthamiana 
predominates along the banks, while the HZ. guianensis- 
complex is relegated to the higher hinterland mesas; HZ. 
viridis is found, but apparently not in abundance, in the 
vicinity of the Cachivera de Yayacopi. (The lowest afflu- 
ent of the Apaporis, the Taraira, is said by ‘‘caucheros”’ 
to have practically no Hevea along its banks. If this be 
true, it curiously coincides with conditions which Allen 
found on the Rio Papuri, an affluent of the lower Vaupés 
in the same general region as the Taraira. ) 
It is interesting to note here that the results of a mile- 
by-mile census of Hevea which I made along a 62 kilo- 
meter path connecting the middle Vaupés and the middle 
Apaporis basins indicates that Hevea Benthamiana is 
almost completely absent on the high hinterlands and 
that the only representatives of the genus encountered 
are H. guianensis and H. guianensis var. lutea, the latter 
by far the more abundant. Similar results were obtained 
on a 86 kilometer traverse which I studied between the 
Itilla and the Macaya. 
CaQuEra River Basin: Hevea guianensis var. lutea 
and, to a lesser extent, H. guianensis are found in abun- 
dance along the Caqueta upstream from a point about 
[ 13 ] 
