ME. B. SPllUCE ON INSECT-MIGEATIONS IK SOUTH AMEEICA. 361 



grey of the evening and morning, i. e. before betaking themselves 

 to their eyry, and before resuming their journey on the following 

 day. The scarcity of fish in rivers of clear or black water is well 

 known ; and even were they more abundant, this very clearness of 

 the water would render it difficult for fish-eating fowls to catch 

 them, unless when there was little light ; hence, perhaps, the Ibis's 

 choice of hours for fishing ; and the turbid water poured into the 

 Rio Negro by the Casiquiari dulls its transparency at that point, 

 which makes it eligible for a fishing-station, leaving probably only 

 a single day's stage for the travellers to reach the Orinoco. The 

 Ibises, however, did not, as one might have supposed, turn up the 

 Casiquiari, but held right on to the north, crossing the isthmus of 

 Pimichin, and descending the Atabapo to the Orinoco. Some of 

 them, I was told, would halt on the Guaviare, whose turbid waters, 

 alligators, turtles, &c. quite assimilate it to the Solimoes or Upper 

 Amazon ; and others push on to the Apure ; the former lot, how- 

 ever, are said to travel chiefly by way of the Japura from the 

 Amazon. Those that frequent the Upper Orinoco return in May ; 

 and their halting-place near San Carlos is not at the mouth of the 

 Casiquiari, but on islands a day's journey below the village, so 

 that they are at that season less persecuted by the Indians. If 

 they went all the way down the Eio Negro in May, they would 

 reach the Amazon long before its beaches began to be exposed ; 

 but it has been ascertained that they sojourn awhile on the Eio 

 Branco, whose beaches are earlier uncovered. Plocks of Wild 

 Ducks sometimes accompany the Ibises ; and it is quite possible 

 that some of the smaller aquatic and riparial fowls make similar 

 migrations. 



When the Ibises are roosting, a shot or two from a gun is 

 enough to make the whole caravan take to flight and remove to 

 some distance ; but the Indians of San Carlos know better than 

 to scare them away with fire-arms. They get into their canoes a 

 little after midnight, creep silently up the river, and under cover 

 of the night disembark beneath the trees where the Ibises are 

 roosting. Then, when at break of day the birds wake up, and 

 begin to stir and to be visible, the Indians pick them ofi" wdth 

 poisoned darts from their blowing-canes, in great numbers, before 

 the bulk of the flock takes alarm ; so that they mostly return to 

 the village with great piles of dead Ibises ; and although this lasts 

 only three or four days, the quantity killed is so great that, what 

 with fresh and what with barbacued game, everybody feasts royally 

 for a fortnight ; whereas throughout the rest of the year, the 



