272 MR. SPRUCE'S BOTANICAL EXCURSION 
December 31.—Last night was gloomy throughout. We started at 
six, and from half-past seven to near eleven we had heavy rain. At 
five minutes after eleven we passed the mouth of the Rio Ariramba on 
the left bank, a narrow winding furo, leading to the Rio Cumina, but 
said to be now impassable for our canoe, on account of sandbanks : 
this furo I had not noticed in ascending. 
Gloomy and exceedingly cold all day: temperature at midday, 
inside and outside of cabin, 75°. 
Yesterday afternoon we had heard the deep short baying of an 
onca on our right: Diogo imitated him with a cuyamboca, and he 
followed us for an hour, often coming near, but never allowing us to 
sight him. This morning, also, we heard an onça (probably the same) 
for an hour before sunrise, but he seemed at a great distance. 
Steep white banks, looking afar like limestone, proved to be indurated 
clay and sand, continued deep under water; the structure is rather 
oolitic, grains of quartz forming the nuclei. There are also some 
reddish-coloured and yellow masses. On these cliffs grew a magni- 
ficent Marcgraaviacea, with racemes of crowded flowers, twenty inches 
long, projecting beyond the mass of deep green foliage like so many 
plumes. The curious saccate bracts, above an inch long, as many as 
two hundred of them on one raceme, are each accompanied by a small 
purplish flower, resembling that of a Clusia in miniature. In the 
same place I was glad to get the Miira-piruna (i. e., Pao préto, black 
wood) in flower. It is a curious Leguminosa, coming near Swartzia, 
but scarcely referable to that genus; the branches, &c., covered with 
a brown felt; large pinnate leaves, with winged petioles ; flowers of 
one large petal growing in naked racemes from the bare stem and 
branches; calyx leathery, 4-cleft ; four groupes of stamens, one turned 
upwards and with elongate anthers, the other three downwards and 
with roundish anthers; the petal and the filaments yellow above, 
violet below. The tree attains a large size in favourable situations ; 
its timber is nearly black, hard, and durable, and much esteemed for 
cabinet-work. 
At five o'clock we reached the mouth of the Jarenucá, our rate of 
travelling having averaged perhaps a little less than three miles 
per hour. 
1850. January 1.—We quartered last night on the same sandbank 
as on the first night after entering the Aripecuri. Heavy and long- 
