208 x MR. SPRUCE'S EXCURSION ON THE AMAZON. 
opposite bank, along which we are coasting. Jauarí palms are 
abundant, especially on the island, while the slender lofty trunk and 
white bark of the Tazizeira are everywhere visible. A little way in 
is another island, between the first island and the right bank. The 
channel between the islands is, towards its further extremity, nearly 
choked with the pretty grass above-mentioned. Our men pushed 
through with poles. After threading about among islands in narrow 
channels, walled in by lofty trees and tapestried by an impenetrable 
mass of twiners, among which a Bignoniacea, with large panicles 
of purple foxglove-like flowers, is most conspicuous, we emerge into 
the main stream about 12 o'clock, joining the other arm before spoken 
of. Their united course is first north-west by west; breadth from 
500 to 800 yards. A little way up, the course trends rather more 
northerly, and a very long island separates the river into two chan- 
nels. That to the right, which we are now entering, is very shallow, 
with here and there sandbanks peeping out—resorts of tartaruga. 
At a short distance within it an igarapé enters from a lake without 
name. The channel grows more and more obstructed with sand- 
banks, until it is at length nearly choked, leaving only a narrow 
strip of water on the western side, along which our canoe could pass. 
. Towards night we drew up about midway of the bank on our left, the 
first great tartaruga-bed on the river, and appearing about 200 
yards broad. Our men immediately dispersed in search of turtles' 
eggs, now become very scarce. They found a few of two species of 
turtle, the larger of which they called tartaruga, and the smaller 
| traeajá. 
| This day, a little after entering the Aripecurá, we e: two 
small Indian sitios, the last human habitations we were to meet with. 
On the Rio Trombétas there are said to be only three sitios above its 
junction with the Aripecurü. 
(To be continued.) 
