INDIAN TRIBES OF BRAZIL. 369 



the .shallower places. About noon Senhor Joao concluded that we 

 must make the rest of the way by boat. Just then two naked Indians 

 in a curious bark canoe came paddling down through the Hooded 

 woods. These were Hypurinas. Pedro Bom, and his brother. Climb- 

 ing into their boat, we were paddled down to the river to a large open 

 shed of palm thatch where Pedro Rom was living. His wile, as 

 scantily clothed as he, was swinging a siek and crying child in an old 

 bark hammock. Another hammock of three long strips of bark (the 

 primitive type of hammock?) was hanging under the shed. These. 

 with a rough clay pot for cooking, seemed to comprise his household 

 goods. There seemed to be nothing to eat about the premises but a 

 bunch of palm fruits which he had just brought in the canoe; but 

 Pedro climbed to the roof of the barrack and brought down four fish 

 (Matrinchao, an abundant fruit-eating fish of the Purus and its 

 branches). These had been slowly roasted and smoked and dried until 

 they would keep without salt. This process is called "moquiar" on 

 the Amazon, and is probably used by all the tribes. Each of us fin- 

 ished a fish, which with a little farina made us a good dinner, the 

 smoke in the fish seeming to take the place of salt. 



Several bark canoes were drawn up on the bank near the shed. 

 These were rude affairs, tied into shape and then dried over a fire; 

 the ends were open and raised but little above the water level. 

 They were made of a single piece of bark taken from the standing 

 tree. The bark of the jutahy is said to be most commonly used. 

 Pedro was easily persuaded to accompany us up the river, and seemed 

 to think nothing of leaving his wife and sick baby alone in the woods. 

 Our party, now numbering live, embarked in one of these canoes, all 

 sitting flat on the bottom, and Pedro finding we were so deep that the 

 Water ran in at the open ends tilled them with clay. 



The river had now diminished to a channel of 20 or 30 feet in 

 width, but rapid and deep and full of fallen timber. Some of this lav 

 just above the water, and we would pull our boat beneath it, all lying- 

 down as we passed under; then a log would be just at the surface, and 

 we would pull the light canoe up and over it, all climbing out .mi the 

 log as we did so. 



After three hours of this we reached so much fallen timber that we 

 could follow the river no farther, so we drew our boat to land and 

 went on afoot. Just here were drawn up the canoes of the Jainamadi 

 village. They were also of bark, but much better made than those of 

 the Hypurinas, the ends being raised up and sewed together, forming 

 a hollow beak. 



The trail now passed over high land and through large tracts of old 

 clearings of the Jamamadi. Some of these had grown up to timber 

 100 feet in height, while in others there were remains of the cultivated 

 plants, bananas, pineapples, and pupunya palms. Their method of 



NAT MUS 1901— 21 



