108 THE CLIFF RUINS OF CANYON DE CHELLY  [ers.ann.16 
a date some time subsequent to the completion of the kiva structure, 
as the wall on the south, now some 3 feet above the level mentioned, 
does not conform to the lower exterior wall on which it was placed. 
On the western side there is another fragment of the upper inclosing 
wall. Both this wall and the one on the south are less than 15 inches 
in thickness. 
West of the kiva there are remains of other stone walls whieh differ 
in character from those on the east. They are now usually less than 
3 feet high; they were 12 to 15 inches thick, and the lines are very 
irregular. South of the kiva, in the center of the ruin, there are other 
stone walls even thinner and more irregularly placed than those on the 
west, but most of the walls here are of adobe. As the use of adobe 
blocks is not an aboriginal feature, the occurrence of these walls is a 
matter of much interest, especially as they are so intimately associated 
with the stonework that it is not always an easy matter to separate 
them. 
The occurrence and distribution of adobe walls is shown on the 
ground plan. They are not found as subordinate walls, dividing larger 
rooms, except perhaps in one instance; but apparently this method of 
construction was employed when it was desired to add new rooms to 
those already constructed. No room with walls constructed wholly 
of adobe can be made out, but walls of this character closing one side 
of a room are common, and rooms with two or even three sides of adobe 
are not uncommon. There are some instances in which part of a wall 
is stone and part adobe, and also instances in which the lower wall, 
complete in itself, is of stone, while the upper part, evidently a later 
addition, is of adobe; such, for example, is the cross wall in the eastern 
tier, about 50 feet from the cliff. 
The mere occurrence of adobe here is evidence of the occupancy of 
this site at a period subsequent to the sixteenth century—we might 
almost say subsequent to the middle of the seventeenth; but its oceur- 
rence in this way and in such intimate association with the stone walls 
indicates that the occupancy was continuous from a time prior to the 
introduction of adobe construction to a period some time subsequent 
to it. This hypothesis is supported by other evidence, which will 
appear later. Attention may here be directed to the fact that there 
are four chimney-like structures in the lower ruin, all of adobe, and all, 
except the one which pertains to the kiva, attached to adobe walls. 
On the western margin of the ruin, and nowhere else within it, there 
are traces of another kind of construction which was not found else- 
where within the canyon. This method is known to the Mexicans as 
“jacal,” and much used by them. It consists of a row of sticks or 
thin poles set vertically in the ground and heavily plastered with mud. 
At present not one of these walls remains to a height of 6 inches above 
the ground, but the lines of poles broken off at the ground level are 
still visible. The ground at this point is but 3 or 4 feet above the gen- 
eral level of the bottom. The ground plan shows the occurrence of 
