140 
A l)0LPHiN Escapes Our GuM, 
us as it was to the Indians. In spite of tlie real platoon-tiring directed 
upon its momentarily-exposed back, into the broad surface of Avhich 
several bullets made their way, we did not succeed in getting possession 
of it. 
312. The boats had to be emptied and the baggage transported along 
the liaidcs. Between the fissures in the rocky dykes we noticed a numlier 
of dead turtle which, probably unabU^ to \\ ithstand the strength of the 
current at high water, had been driven betweeen the crevices, jammed in 
there, ami died of hunger. \\'hile our crew s were busily engaged hauling 
up the corials 1 hastened td the foot of the I'iriwai to see it'/tliere was 
anything new in the l)otanical line, and hardly had i made my way across 
the river bank which was overgrown with the pretty (Jaa iitoclit coccinca 
Moeuch., and (k'coratcd with thousands of beautifully brilliant purple- 
red l)lossoms, and reached the adjacent oi>eu savanuah, than the glori- 
(Mis azure blue flowers of the lovely I (xmica crol niloidc.s Morlc. and 
Jdvqucinonfia liirsKta Chois. lighted upon my gaze. Their immense number 
lent an endless charui to the extensive i)laiu, but the wildly romantic sur- 
]f»undings of i'iriwai's peculiar iriouiitaiu-base sui prised nie more. Above 
the large-leaved Potlios and TUhindsUic there towered blocks of a coarse- 
grained granite often 50 to (ib feet high, likewise covered with the rocky 
\rgetation already often desci'ibed, auu)Ugst which 1 also discovered a 
new species cd' hitrijchno.s, t^lri/chnos Schoiitbiirgk't Klot/.sch, from the 
bark of which the Macusis make their urari jtoison. To climb this protec- 
tive zone of rocky boulders irregularly piled one on top of the other so as 
to reach tlie two gloomy watch-towers would have required more time 
than I could afford : I could only gaze at them from below and see a 
lonely bird of prey circling round them. 
343. AThen I returned to our people, 1 fouiul that the cataract had 
lieen overcome and the tired chaps busy over their lu'cakfast which con- 
sisted of several large Sihini-s that had been caught during my absence. 
After we had doubled round Piriwai along a sheet of water continually 
broken by pi'ojecting rocky tops, we saw that the mountain had turned 
the stream some considerable distance to the south-east. Maikang]>ati, 
which rises at some little distance from Piriwai on tlie western^ shore, 
gives rise, like it, to a few, if only insignilicant, rapids. Close to its north- 
ern base, the little stream Zunona empties into the Cotinga. 
344. It was already almost 14 days since we had left the settlement at 
Warami, and from what the Indians told us. we could hardly reach the 
next village under six. The supply of cassava bread and fariuha was 
already markedly diminished, and the rations had to be reduced to half, 
although our troubles were increasing practically every hour because 
in front of us was an almost impassable cataract and not a single moun- 
tain to the right or left, but only the flat savannah vStretcliing along both 
sides of the bank. Since the tapir no game had been seen : the huntsmen 
who strolled along the banks into the savannah, returned in the evening 
just as empty-handed as they had started in the morning. Napoleon and 
others apparently as greedy eaters as himself, began to growl, but this 
did not help them much. 
345. The 3rd and 4th October nndoulitedly proved the most fatiguing 
days we experienced during the whole of oui- ( 'otinga trip. In spite of 
