246 
TRAVELS ON THE RIO NEGRO. 
\_Fehrmry, 
Barra than at Para, — a circumstance which shows the 
total inapplicability of that instrument to determine 
small heights at very great distances, I cannot there- 
fore think that Sao Carlos is more than four hundred, or 
at the outside five hundred feet, above the level of the 
sea. Should, as I suspect, the mean pressure of the 
atmosphere in the interior and on the coasts of South 
America differ from other causes than the elevation, it 
will be a difficult point ever accurately to ascertain the 
levels of the interior of this great continent, for the dis- 
tances are too vast and the forests too impenetrable to 
allow a line of levels to be carried across it. 
When my Indians returned with the roots of timbo, 
we all set to work beating it on the rocks with hard 
pieces of wood, till we had reduced it to fibres. 
It was then placed in a small canoe, filled with water 
and clay, and well mixed and squeezed, till all the juice | 
had come out of it. This being done, it was carried a ( 
\ 
little way up the stream, and gradually tilted in, and 
mixed with the water. It soon began to produce its | 
effects : small fish jumped up out of the water, turned | 
and twisted about on the surface, or even lay on their | 
backs and sides. The Indians were in the stream with | 
) 
baskets, hooking out all that came in the way, and diving ■ 
and swimming after any larger one that appeared at all 
affected. In this way, we got in an hour or two a bas- 
ketful of fish, mostly small ones, but containing many 
curious species I had not before met with. Numbers 
escaped, as we had no weir across the stream ; and the 
next day several were found entangled at the sides, and 
already putrefying. I now had plenty to do. I selected 
