312 NOTES OF A BOTANIST 
the height was found to be 1635 feet; adding to 
which the height of Sao Gabriel above the sea, by 
other observations, the total height may be 
estimated at about 1800 feet, with a probable error 
of 50 feet. Spruce gives his full calculations.] 
Mode of obtaining Salsaparilla 
March 23. — There is a small plantation of Salsa in a tabocal 
(bamboo grove) a little way down the falls, whither I went this 
day with the owner to witness the mode of taking up the roots. 
The plant selected had five stems from the crown, and the 
numerous radiating roots extended about 3 yards on every side. 
The roots were first bared, and had the Salsa been the only plant 
occupying the ground, the task would have been easy, but they 
are often difficult to trace among the intricate mass of roots of 
other plants, which require to be cut through with a knife or small 
cutlass. The earth, which is only a thin covering, is scraped away 
by the hand or by a pointed stick. The roots being at length all 
laid bare (in this case it was the work of half a day, but of large 
virgin plants it sometimes takes up a whole day), they are cut 
asunder near their base, a few of the more slender being left in 
order to stay the plant in its place. A well-grown plant will yield 
at the first cutting from one to even two arr6bas. In a couple of 
years it may be cut again, but the yield is much less — the roots 
slenderer and yielding (say the Indians) less starch. 
An Indian Festival 
April 17 and 18 I was present at a Dabocurf (or festa of Barre 
Indians) on an island near the base of the falls, a little above the 
ancient village of Camanaos. The house was pleasantly situated 
on rising ground, the walk up to it fringed with Coffee trees 
laden with berries, amongst which were three or four clusters of 
Pupunha palms and here and there a Cocura tree. A flat, semi- 
circular space of hard sand in front of the house had been clean 
swept to prepare it for the dancers. This space was skirted by 
spreading Ingas, under shade of which benches had been put up 
with backs to them, the seats of strips of Paxiilba palm laid close 
together. 
There had been prepared beforehand a quantity of cauim (dis- 
tilled from sugar-cane) ; two flageolets of Paxiilba about 6 feet 
long (made by Indians on the river I^anna) and three or four 
smaller ones ; a number of gaitas of a single internode of the 
slenderer branches of some Cecropia, with a wind-hole cut on one 
