288 THE AMERICAS Ml'SEl'M .lorUSAL 



ingly on July 28, farewells were .said, the " hallylioo," tlie " corials" 

 (eanoes), and the woodskins were loaded and ofV we went np the river; 

 late ill the at'ternoon of .Inly ;'>(), we arri\«'(l at ( 'lieiiai)o\\ ii. 'V\\v river 

 travelinji; was o\v\ tor the time, and now imagination ran ahead aionj^ the 

 distant way thrf)n^h the forests and across the savamiahs to Roraima, a 

 (Hstanee of one hiuidred and ten miles that had to he traversed entirely on 

 foot. One day's halt was necessary for the orj:;anization of tiie provision 

 loads and for the en^a<;ement of additional h<'ar(>rs; and then on August 

 first, the line of twenty-six natives fihd olV into the forest. 



We reached Sa\eritik on the Ireng Kiver, wiiich forms the Brazilian 

 boundary, at about noon of August ."), with the first third of the long walk 

 successfully accomphshed. Taking account of the reiiiaiiiiiig provisions 

 and of the axailable time, it was obvious that a crisis had been reached. 

 Slow traveling anfl \exatious delays had so reduced supplies that they were 

 insufficient for the journey to Roraiuia atid back to Saveritik. Two courses 

 were open: we could go on within the mai-gin of safety, penetrate for some 

 distance into Brazil, and study the life of the savannahs at their northern 

 limit, or we could still make the effort to reach Roraima in the hoj)e and 

 expectation of obtaining food somewhere l)eyond. Tiie \\\yv of the famous 

 momitain made the decision, and accordingly the immber of bearers was 

 reduced to seventeen, while four men were sent back to Kaieteur for ad- 

 ditional supplies to l)e at hanfl (m our return to that point. On August 6 

 we safely passed the ri\er, and it was not without some emotion that I 

 stepped out into the forests of Brazil, a country that will always hold the 

 interest of scientists on account of the work of such men as Bates, Wallace, 

 Agassiz, and Darwin. 



After a steep climb througli the forests u]) to a l)ar()metric level of 3600 

 feet, we emerged on to the great savannahs. Before us rolled great grassy 

 plains marked here and there by the deeper greens of the forest trees along 

 the watercourses. Occasionally an outcrop of reddened soil or gray 

 clay added its contrasting color. From this time on, we lived amid such 

 scenes, camping for the noonday meal in a patch of forest by a wayside 

 stream, or sometimes on the open savannah, unprotected from the fierce 

 direct rays of the sun at its zenith. Gathering speciiuens by the way we 

 would at length come to a favorable site for the night camp where the 

 bundles of equipment would be set down, the natives would receive their 

 rations, and the smoke from the camp fires would rid us of the sandflies 

 which were always present in immense numbers. 



After several days of such traveling we came out upon the great head- 

 land overlooking the beautiful valley of the Kwating River whose wide 

 plain spread loUO feet below, and at last we could see flat-tojjped and cloud- 



