Page:Through the Brazilian Wilderness by Theodore Roosevelt (1925).djvu/160

 pletely devoured by the ravenous fish. It was a further illustration of the uncertainty of temper and behavior of these ferocious little monsters. In other lagoons they had again and again left us and our dogs unmolested. They vary locally in aggressiveness just as sharks and crocodiles in different seas and rivers vary.

On the morning of January 9th we started out for a tapir-hunt. Tapirs are hunted with canoes, as they dwell in thick jungle and take to the water when hounds follow them. In this region there were extensive papyrus-swamps and big lagoons, back from the river, and often the tapirs fled to these for refuge, throwing off the hounds. In these places it was exceedingly difficult to get them; our best chance was to keep to the river in canoes, and paddle toward the spot in the direction of which the hounds, by the noise, seemed to be heading. We started in four canoes. Three of them were Indian dugouts, very low in the water. The fourth was our Canadian canoe, a beauty; light, safe, roomy, made of thin slats of wood and cement-covered canvas. Colonel Rondon, Fiala with his camera, and I went in this canoe, together with two paddlers. The paddlers were natives of the poorer class. They were good men. The bowsman was of nearly pure white blood; the steersman was of nearly pure negro blood, and was evidently the stronger character and better man of the two. The other canoes carried a couple of fazendeiros, ranchmen, who had come up from Cáceres with their dogs. These dugouts were manned by Indian and half-caste paddlers, and the fazendeiros, who were of nearly pure white blood, also at times paddled vigorously. All were dressed