8 BULLETIN OF THE NATURAL HISTORY SOCIETY. 
set, with alley-ways between, in some cases only four feet wide; 
but the spaces between the huts further back are greater. 
Such at least was the arrangement of the huts in the later 
years during which this community existed. Not only were 
the huts more scattered in the rear part of the village, but 
anything like a systematic arrangement in rows entirely dis- 
appears after passing the third row from the front of the 
village. 
Of the depressions marking the hut bottoms of .the front 
row, which as I have said, were more closely placed than those 
further back, one was much longer and deeper than the rest. 
This, at first sight, gave the impression of a long communal 
dwelling, and here we determined to begin our exploration. 
In digging a trench through a part of this depression we 
struck an ancient fire-place, which was made the centre of ex- 
ploration for several feet around. It was found that at this 
point the deposit in the hut bottom was about two feet deep, 
but its fire-place rested on an older kitchen-midden, or refuse- 
heap beneath. 
HUT BOTTOM NO. 1.— ITS ANTIQUITY. 
The hut bottom of which the first named fire-place is the 
centre is distinguished on the plan as hut bottom A. The older 
shell-heap beneath (marked No. 1) differs in several respects 
from that of hut bottom A; there was much less charcoal 
mingled with the shells, and fish-bones were more plentiful. 
When also we had traced this lower kitchen-midden in different 
directions beneath and beyond hut bottom A, we found stone 
chips or flakes, which had been struck off in the manufacture 
of weapons, differing in kind from any that were met with at 
the higher levels. The weapons found in the lower kitchen- 
midden were also larger, coarser, and of a different form from 
those exhumed from the hut bottom and waste heaps of A. 
Fragments of bone harpoons were more plentiful at the lower 
horizon, the pieces of bone more fragile, and the stone chips 
more abundantly coated with carbonate of lime than at the 
higher levels. There were also differences in the patterns of 
the pottery found in the two deposits. The ornamentation of 
the fragments of pottery found in this lower kitchen-midden 
