710 
entering), and the large door in front is roughly 
closed by a well-riddled moose or caribou skin, 
or occasionally a piece of canvas. The roof is of 
skins battened down by spruce poles. The row 
of houses is so close to the scarp of the bank, 
that the ‘ street’ in front is a narrow path, where 
two persons can hardly pass without stepping 
indoors or down the hill, and, when I visited the 
village, was so monopolized by scratching dogs 
that I could hardly force my way through. A 
fire is built on the dirt-floor in the centre of the 
habitation, and the smoke left to get out the 
best way it can. As the occupants are gener- 
ally sitting flat on the floor, they are in a stra- 
tum of air comparatively clear; for the smoke 
can find air-holes through the cracks of the 
house-walls, while that which is retained under 
the skins of the roof is utilized to smoke the 
salmon which are hung up in this space. 
It was at this village that to me the most 
wonderful and striking performance ever given 
by any natives we encountered on the whole 
trip was displayed, and in this I refer to their 
method of fishing for salmon. I have spoken 
of the extreme muddiness of the Yukon from 
the mouth of the White River; and this spot, 
of course, is no exception. I believe I do not 
exaggerate in the least when I say, that, if 
any ordinary pint tin cup was filled with it, 
nothing could be seen at the bottom until the 
sediment settled. The water is from eight to 
twelve feet deep on the banks in front of 
their houses, where they fish with their nets ; 
or at least that is nearly the length of the 
poles to which the nets are attached. The 
salmon that I saw them secure were caught 
about two hundred and fifty or three hundred 
yards out from the bank, directly in front of 
the houses. Standing in front of this row of 
cabins, some person, generally an old squaw, 
possibly on duty for that purpose, would an- 
nounce that a salmon was coming up the river, 
when some man, identifying its position, would 
run down to the beach, and pick up his canoe, 
paddle, and net, and start out into the river 
rapidly ; the net lying on the canoe’s deck in 
front of him, his movements being guided by 
his own sight and that of a half-dozen others 
on the beach and bank, all shouting to him at 
the same time. Evidently, in the canoe he 
could not judge well at a distance: for he 
seemed to rely on the advice from shore until 
the fish was near him, when, with one or two 
dexterous and powerful strokes with both 
hands, he shot the canoe to the position he 
wished ; regulating its finer movements by the 
paddle in his left hand, while with his right 
he plunged the net the whole length of its pole 
SCIENCE. 
a 
[Vou. III., No. 71. 
to the bottom of the river, from eight to twelve 
feet ; often leaning well over, and thrusting his 
arm deeply into the water, so as to adjust the 
mouth of the net (covering about two square 
feet) directly over the course of the salmon. 
Of seven attempts, at intervals covering 
three hours, two were successful, salmon 
being caught weighing about fifteen pounds. 
How these Indians at this great distance can 
see isolated running salmon on the bottom of 
an eight or ten feet deep river, and determine 
their position near enough to catch them in the 
narrow mouth of a small net, when under the 
eye a vessel holding that many inches of water 
from the river completely obscures an object 
at its bottom, is a marvellous problem that I 
will not attempt to unravel; it of course de- 
pends in some way on the motion of the fish. 
In vain they attempted to show members of the 
party the coming fish. I feel perfectly satis- 
fied that none of the white men saw the least 
traces that the natives tried to show. In their 
houses and on the scaffoldings were several 
hundred that had been caught in this way. 
The only respectable theory that I could evolve, 
was that the salmon came along near the top 
of the water, so as to show, or nearly show, 
the dorsal fin (for it borders on the marvel- 
lous that they could be seen at the bottom, or 
that any motion of theirs could be detected 
from the top when they were on the bottom, 
among the ripples of the swift muddy stream), 
and that when they neared the canoe, the 
sight of it, or more likely some slight noise, 
probably made on purpose, sent them to the 
bottom without any considerable lateral devia- 
tion, and that they were thus directed into the 
net; but my interpreter told me that this 
superficial swimming did not take place, but 
that the motion of 
the fish was com- 
municated clear to 
the top from the 
bottom. The nets 
used I have par- 
tially described al- 
ready. ‘'Themouth 
is held open by a 
light wooden frame 
of a reniform shape, 
as shown in fig. 14 ; 
and, as one will readily see, this is of great 
advantage in securing the handle firmly to the 
rim of the net’s mouth, and is undoubtedly the 
object aimed at. I might state here, that far- 
ther down the river (that is, in the ‘ lower ram- 
parts ’) the reniform rim becomes circular, of 
course increasing the chances of catching the 
