KIDDER-GUERNSEY] ARCHEOLOGICAL EXPLORATIONS IN ARIZONA 51 
of the top of the loop shows such wear as would have been caused 
by the play of a string or rope running through it. It is possible 
that ¢ and d also once contained loops, and with e formed part of 
a series of loopholes such as were later found in other kivas. 
On the wall seven or eight thin coats of plaster may be counted, 
each one blackened with smoke. This renewing of the plaster in 
thin layers seems to be characteristic of all kivas, not only here but 
in Canyon de Chelly, Mesa Verde, Chaco Canyon, and the Rio 
Grande. There are no recesses or cubby-holes in this kiva; the single 
large niche is edged and flagged with flat pieces of sandstone (pl. 
15). In the masonry at the top of the back wall of the niche, and 
2 feet 9 inches above its floor, there is a smooth, level surface about 
3 feet long; while not distinctly defined at either end, it is conceivable 
that this smooth area may indicate the former presence of an entry- 
way at this place. The smooth appearance may, however, be quite 
secidental, and it is much more probable that entrance was gained 
by a hatchway in the ceiling. No sign of roofing was visible, 
although the wall on the southeast side stands 7 feet high. In the 
masonry on the northwest side, 6 feet from the floor, are sockets left 
by two sticks, 2 inches in diameter, 8 inches apart; being too small 
for roof beams, these probably represent the former location of 
wall pegs. 
What seems to have been a second kiva has, unfortunately, been 
almost entirely destroyed by water falling from the cliff above. 
There remain only a segment of curved wall and traces of a niche or 
bench. 
Between the large rock mentioned as having formed part of the 
back of the vertical shaft of kiva 1 and another bowlder lying to 
the northwest of it is a cist or inclosure of large sandstone slabs 
(fig. 20, No. 16). When first uncovered it was thought that it might 
have some connection with the adjacent kiva, but no relation between 
the two could be made out, and it proved on excavation to be filled 
with earth, rubbish, and a quantity of turkey droppings. 
The two open spaces or terraces occupied most of our time, as the 
deposit in them was 2 feet to 3 feet deep and was so filled with speci- 
mens of all kinds as to require excavation almost entirely with a 
trowel. It consisted of the usual cliff-dweller débris—corn husks, 
cobs, leaves, sticks of wood, bits of cloth, bone, turkey droppings, and 
ashes. In it were two fairly well defined living surfaces, one at 14 
inches, the other at 18 inches above the cave floor. They were dis- 
tinguishable by the compact character of the general refuse, and 
were covered in some places with a thin coating of adobe and ashes. 
The levels were not constant throughout the mass, having been 
*In Mesa. Verde kivas short pegs are regularly found protruding from the front of 
the pilasters near their tops, 
