University of Virginia Library

THE SOUTH ANNEX

Despite the fact that their dimensions and fittings are recorded in
Appendixes B and C, I desire briefly to direct attention to these seven

Fig. 39.—Alterations at northwest corner of Kiva I enclosure.

rooms, both individually and as a group (pl. 2, lower). In floor area
they average about 56 square feet; in ceiling height, 5 feet 3 inches.
In each instance the principal beam or beams lay north and south and
had been seated in the buttressed wall by breaking out proportionate
sections of its masonry veneer. Three beams were utilized in Room 1;
one only in the other six.

Ceiling construction was most clearly illustrated in Room 2. Here
a single main beam, fitted into the north wall 4 feet 5 inches above the
floor, carried 5 secondary east-west timbers. These latter averaged
4 inches in diameter and, with lesser poles between, presumably had
supported a layer of brush or cedar shakes and several inches of adobe.
We took sections from the more promising of the five but, to our disappointment,
their annual growth rings were too uniform for successful
dating. Logs of comparable size had been inserted longitudinally
into the ends of the buttresses on either side of the room and embedded
in the added stonework to serve as wall plates for the southernmost
timbers. Posts as beam props near the weak south wall were
noted in Rooms 2 and 4.

The masonry necessary to complete the seven rooms was, in each
instance, appreciably thinner and less substantial than that of the
buttresses. While these latter vary in thickness from 12 to 18 inches,
the added stonework averages only 10.

In preparation for Room 7, the 10-inch layer of constructional
waste that had accumulated here was first removed. Then the face
of that debris as it underlay the two buttresses was concealed by a
coat of wall plaster that covered the buttress masonry and continued
down to round off with the adobe floor. For Room 3 a deeper excavation
was made since here the floor is 21 inches below ground level.
The individuality of these seven dwellings is further indicated by
the fact that the 3-inch-wide ledge marking floor level in second-story
Rooms 9B-I to 9B-III, next on the north, is 7 feet 3 inches above the
floor of Room 1, 7 feet 7 inches above that in Room 2, but only
6 feet 4 inches above the flooring in Room 4 which had been spread
directly upon the constructional debris.

Mud plaster still adheres to wall masonry in five of the seven rooms,
and this plaster is noticeably smoke stained in three of them, Rooms 5,
6, and 7. Small closets or niches, plastered inside, appear in the north
and east walls, respectively, of Rooms 4 and 5 (pl. 48, left). Fireplaces
are present in all rooms except 2 and 4. Three of these fireplaces
are slab lined; one is lined with clay, and the fifth, with a
combination of masonry and slab fragments on edge.

Fig. 40.—The McElmo Tower and adjacent structures. (From the original
survey by Oscar B. Walsh.)

The fireplace in Room 3 is situated in the northwest corner; it is
lined with slab fragments and was packed with wood ash when found.
In the southwest corner of the room a neatly constructed ventilator,
6½ inches wide by 19 high and 6 inches above the floor, opens through
the south wall to connect with an external, masonry-lined shaft 15
inches deep. The top of this shaft, at ground level, was capped by an
inverted metate that had been worn through and the resultant hole
subsequently enlarged to an oval 7½ by 12 inches (pl. 50, B). Within
the room, against the west wall and midway between ventilator and
fireplace, is a deflector that consists of a thin sandstone slab on end
augmented by a discarded metate braced at the back by two slab fragments
embedded in the floor (pl. 46, B).

A rectangular fireplace occupies the middle floor of Room 1, and a
few inches distant two slightly worn tabular metates on end formed a
protecting screen. This latter was required to deflect drafts created
by the north door and an improvised ventilator, a small irregular hole
broken through the south wall at floor level.

In Room 5 sandstone slabs formed a bin against the south wall to
enclose a very un-Chaco-like metate (pl. 48, right). The mill has a flat
over-all grinding surface 8 inches wide by 19 inches long and was
accompanied by a sandstone mano measuring 7 by 4 inches. The
lower end of the mill rested 2 inches below floor level while the raised
end, where the miller knelt, was 5½ inches above.

Interior doors connected Rooms 4 and 5 and, at one time, 5 and 6.
The former, with a sill height of only 20 inches on its Room 4 side,
was provided with a step formed by a rounded mass of adobe 5 inches
wide and 20 inches long. Also, the door had been reduced to a width
of 17 inches when secondary adobe jambs were added to receive a
doorslab positioned from Room 5. The floor in Room 5 is 2 inches
above that in 4. Outside doors had originally been provided for
Rooms 4 and 6, but that in 6 was subsequently sealed. Entrance
to the other five rooms, therefore, must have been through ceiling
hatchways.

Part of a hatchway was noted in Room 3 where several of the ceiling
timbers had survived. The opening was situated directly above the
deflector that stands at the west end of the room, halfway between
fireplace and ventilator. Here one of the secondary east-west timbers,
22 inches from the south wall, marked the north edge of the passage,
and a pair of 2-inch-diameter poles extending from the timber to the
southwest corner formed its west margin. Two similar poles, upon
and at right angles to the first pair, suggest both the thickness of the

ceiling construction and the depth of the hatchway frame. There was
no other means of entering or leaving Room 3. Presence of this opening
could have had no relationship to the fact that the ceiling was
4 inches lower along its south side.

Room 2 lies between 3 and 1, and potsherds as well as sandstone
chips were used in chinking its outside wall. Several of its lesser
ceiling timbers continue through the west side and into Room 1—evidence
that 2 was completed before the area of Room 1 was enclosed
and roofed. Without lateral doors, Room 2 could have been entered
only through its ceiling, but the location of the hatchway is not apparent
in this instance. A main-beam seating in the north wall 9 feet
10 inches above the floor is that of a second story whose ceiling height
we estimate at 5 feet. Portions of plastered second-story masonry
still stand on both the east and west sides, and in the former there
remains one jamb of a door that had opened upon the flat roof of
Room 3. A door from Room 9B-II likewise gave access to the Room 3
roof. The flat roofs of Pueblo homes have always been utilized for
fair-weather household activities.

Sometime after construction of Room 1 a passage had been broken
through its north wall—the 33-inch-thick wall whose threatened collapse
had prompted erection of the eight buttresses—to connect with
Room 9A. At the opposite end of that exceptionally long room,
before or after partitions were introduced to create Room 10, another
door had been cut through. This provided a passageway from Room
10 to Room 7 or the site 7 now occupies. We know 7 was built later
than Room 6 because the south half of its west wall abuts the outer
southeast corner of 6. We believe the connecting door antedated construction
of 7 because it had been neatly blocked with masonry of the
same type as that on either side—close-fitting, laminate sandstone
masonry for which the builders of Room 7 and its like had no patience.
At the time of this blocking, a 24-inch-deep recess was left on the
Room 10 side, and a substitute door, with steps, was opened immediately
above—an awkward but effective example of replacement illustrated
in our description of Room 10 (p. 15).

A noticeable feature in connection with Rooms 1-7 is their surprisingly
low ceilings. Holes broken in the north-wall masonry for reception
of their main beams vary in height above the floor from 4 feet
5 inches (Room 2) to 5 feet 6 inches (Room 7). Assuming these
beams to have been 6 to 8 inches in diameter, we may estimate ceiling
heights for the series at from about 5 feet to a little over 6. The
adult occupants must have moved habitually in a stooped position

when indoors. Above roof level the butt ends of great 11-inch beams
extend through from Rooms 8-11 to end flush with the outside wall.

Kiva B (fig. 41).—Raised above Room 1 at the west end of this
annexed group, and likewise built against the outer south wall of
Pueblo del Arroyo, is a small ceremonial chamber. We designated
it "Kiva B" because it was the first of its kind we happened upon at
this site and "A" was being reserved for a Great Kiva we anticipated

within the pueblo but never actually sought. Kiva B is circular and it
had been enclosed by straight masonry walls to effect the subterranean
positions required of Chaco Canyon kivas, but otherwise it is as
un-Chaco-like as the house group with which it is associated. Like
that in the houses, its masonry is a haphazard composition of both
laminate and friable sandstone, worked and unworked (pl. 45, lower).
Remnants of its enclosing square survive on either side, but the south
wall presumably was undercut and lost with enlargement of Jackson's
"old arroyo" (pl. 44, B; see also Introduction, p. 2).

Kiva B is 11½ feet in diameter at the floor. Encircling the floor is
an earthen bench that averages 12 inches wide and 34 inches high.
Above bench level the kiva is masonry lined, but the bench itself is
entirely of earth except at the south recess. Clearly the bench was
left when the kiva pit was dug down into the canyon alluvium. Sooted

plaster adheres both to the stonework of the upper wall and to the bare
earth of the bench.

There are no pilasters in Kiva B. Its flat roof was supported by
beams extending across from wall to wall at a height of 7 feet 9 inches.
An empty beam socket may be seen below and to the right of the
seated figure in plate 45, lower. At approximately waist level of this
same figure paired ceiling beams from Room 8 end flush with the
wall's exterior, a few inches below the ledge that identifies floor level
in second-story Room 8B-I. That ledge lies 10½ feet above the kiva
floor and, as previously noted, 7 feet 3 inches above the floor in
Room 1, adjoining.

The masonry-lined south bench recess in Kiva B is 47 inches wide
and 6 inches deep. In the middle of it, on the floor and abutting its
rear wall, is a rude masonry platform 7 inches high, 29 inches long,
29 inches wide at the back and 24 inches in front. From the top of
this platform a ventilator, 10 inches wide by 21 inches high, opens
into an air duct or tunnel that extends southward 44 inches to meet a
masonry-lined shaft. This latter is 14 inches square and presumably
once rose to kiva roof level. The tunnel had been covered with small
poles, split cedar, and a layer of sandstone slabs. That this roofing
had collapsed at some time and been replaced is suggested by the misalined
stones above (pl. 49, B). The ventilator opening apparently
was narrowed at the same time for, as rebuilt, it was not carefully
centered in the recess wall. Through shaft and tunnel fresh air was
drawn into the kiva as air heated by the midfloor fireplace escaped
through the ceiling hatchway.

The fireplace is a stone-lined box, 16 by 20 inches, sunk 5 inches
into the floor. Because a deflector is lacking I presume the 7-inchhigh
platform, 18 inches distant, in some way shielded the flame from
ventilator drafts. The expected sipapu, between fireplace and north
bench, was not discovered.

Built upon the Kiva B roof and against the outside of Room 8 are
two wall fragments and a buttress (pl. 45, lower; fig. 41). The latter
is square ended directly above the inside curve of the kiva, but the two
fragments clearly represent one-time, roof-level enclosures: a small
room at the northeast corner of the square and an alcove opposite.
Close in the far corner of this alcove is a fireplace, 11 by 19 inches
by 6 inches deep. Beneath the alcove floor and extending southward
between the convex kiva curve and its enclosing wall is a long, narrow
space that had been paved 4½ feet below the level of the kiva
ceiling and thereafter filled with household sweepings. From this

rubbish we tabulated 1,902 potsherds of which 99 were of types I
described as "Old Bonitian" when discussing the material culture of
Pueblo Bonito (Judd, 1954, p. 21) and 798 were "Late Bonitian."
Of these latter 40 were decorated in straight-line oblique hachure, 196
were of a variety we called "Chaco-San Juan," and 562 were fragments
of Corrugated-coil culinary ware.

The narrow space opposite, between the kiva and its east enclosing
wall likewise had been intentionally filled with blown sand and debris
of occupation. Here we recovered a single square-stemmed, sidenotched
arrowhead, a knife made from a flint flake, two bone awls,
and a small needle (U.S.N.M. No. 334921). Also, a bone flaker for
chipping arrowheads and knives, an oval, flat-bottomed but undrilled
lignite button measuring 1[fraction 5 by 16] by 1 by [fraction 7 by 16] inches, a handful of unworked
lignite fragments, and a number of miscellaneous potsherds.

Kiva B itself was filled mostly with masonry fallen from the second
and third stories of Room 8. Among this wreckage we found the
customary assortment of potsherds, a bone awl, a couple of hammerstones,
and a chert flake chipped on both edges for use as a knife.
In addition there was a ⅝-inch-thick section of a globular concretion
that may have been shaped as a jar cover but that had last served as a
palette in the preparation of yellow paint (U.S.N.M. No. 334839)
and a stone artifact whose like I have never seen in another collection
from the Southwest. It is a discoidal made from a waterworn cobble
of very hard reddish conglomeratic quartzite and its periphery reduced
by pecking to a uniform ½-inch width—as beautifully symmetrical
as any Chunkee stone ever found in Georgia. Ours measures
2⅞ inches in diameter by 1⅝ inches thick (pl. 40, v), and I am told,
with considerable hesitation, that cobbles of similar composition might
be found in the San Juan Mountains. Traces of fugitive red paint
are to be seen on both faces.

Masonry toppled from the second and third stories of Rooms 9
and 10, plus the ever-present windblown sand of Chaco Canyon, likewise
filled the seven deserted homes associated with Kiva B. None of
these dwellings contained a recognizable trash pile but each sheltered
a few artifacts, or fragments of artifacts, abandoned or overlooked
on moving day. Room 6 held the fewest, two manos and four hammerstones;
Room 3 the most. Here we found the following:

• 1 tubular bone bead, ½ inch long

• 2 bone awls

• 1 discoidal potsherd, 1⅝ inches diameter, edges beveled, 3 biconical drillings

• 1 metate, 12 x 17 x 3 inches

• 2 manos and 2 fragments

---

105





• 1 manolike stone shaped on 3 edges but not used as a mano

• 1 hammerstone

• 1 small smoothing stone

• 2 corrugated pots (pl. 47, c, d)

• 1 coyote skull (field No. 88)

• 1 right half of a human upper jaw (field No. 89)

The last item, half an upper jaw found among fallen masonry about
18 inches above the floor, is of peculiar interest since the left half
of the same jaw (field No. 92) was recovered a week later in Room 9.
There is no means of direct communication between the two rooms,
and I have no idea as to how the fragments became separated. They
presumably belong to a disarticulated adult male skeleton (U.S.N.M.
No. 327141), the major portion of which lay at the west end of 9.

The two culinary vessels from Room 3 (pl. 47, c, d) are thoroughly
typical of the early phase of Pueblo III as it developed throughout the
San Juan drainage. Each has the characteristic egg-shaped body,
constricted neck, and outflaring rim, but the coils on d are narrower
(5½ to the inch) and less boldly indented than those on c. Neither
form nor workmanship fixes precisely the district in which these two
were made, but there can be no doubt as regards the canteen illustrated
on plate 28, c, restored from fragments recovered in Rooms 2
and 3. Its gray surface was smoothed and decorated with organic
paint prior to polishing. The paint itself is mostly a smoky gray in
color, without relief, and burnished in places by the polishing stone.
Shape and a slightly concave base mark the vessel as of Pueblo III
age; its decorative elements are familiar ones in southwestern Colorado
and southeastern Utah where organic paint was favored in
Pueblo III times.

I am less confident of my judgment as regards two bowls found
crushed in Room 5 (pl. 47, a, b). The first, unslipped and inexpertly
made, has a rim that is rounded in part, [fraction 3 by 16] to ¼ inch thick, partially
incurved, and irregularly ticked with carbon paint. Externally the
vessel has been hand smoothed without entirely obliterating the
structural coils. The second bowl is thinner, with a flattened, unornamented
rim, a flattened bottom, and a decorative band of opposed rectangular
scrolls, plus some solid fill-in, drawn in a blue-gray pigment
that lacks the relief of a mineral paint but, nevertheless, exhibits the
tendency of a low-grade mineral paint to rub off. Perhaps we have
here one of those infrequent examples in which the two pigments
were mixed. An occasional line extended past its rightful ending
might reflect the carelessness of early Chaco (Transitional) potters,
or simply lack of experience.

There is but little else in the way of cultural material from these
seven outside rooms that merits description. Stone implements such
as hammers, manos, and metates, and slab fragments were left in the
rooms where they were found; bird and mammal bones and awl fragments
were noted and discarded. A bone awl from Room 1
(U.S.N.M. No. 334922) is noteworthy because of its 9-inch length
and the drilled hole at the butt. Like that shown on plate 37, f, it
might have served as a dagger, although we lack evidence of such a
weapon among the historic and prehistoric Pueblos.

In addition to fragments of the canteen (pl. 28, c), we found in
Room 2 a thin red claystone pendant ⅞ inch in diameter (U.S.N.M.
No. 334749), a couple of squared sandstone slabs each ⅜ inch thick
but one measuring 10½ by 12 inches and the other 14 by 15, and a
cherty sandstone concretion 6¼ inches in diameter by 5½ inches thick,
unmodified except for an encircling groove at the waist. It is doubtless
only fortuitous that this grooved concretion lay on the floor beside
the unfinished ax shown in plate 41, c.

The binned metate in Room 5 (pl. 48, right) is foreign to Chaco
Canyon. Its outstanding feature is a transversely flat, full-length,
over-all grinding surface. On this type of mill a hand stone as wide
as or wider than the grinding surface ordinarily would be employed,
but in this particular instance the accompanying mano was only 7
inches long. Metates of this type, mounted in a bin, came into use as
early as Pueblo II (Brew, 1946, p. 240). Thereafter they were
favored in certain areas but not in all; they are the preferred type in
Hopi homes today (Bartlett, 1933, p. 17). The type has been variously
described but most frequently as "flat," although the grinding
surface is invariably more or less concave longitudinally.

The one-end-open troughed metate was the prevailing type in Chaco
Canyon. Here, while the B.M. III and P. I cultures survived, metates
made from relatively thin slabs of sandstone predominated (Roberts,
1929, p. 132; Judd, 1924, p. 402, pl. 1, lower); those made from
thicker slabs became the fashion later. Both varieties were found at
Pueblo Bonito and the thinner I attribute to the P. II portion of the
population (Judd, 1954, pp. 133-137).

Metate bins are known from only two rooms at Pueblo Bonito, 90
and 291, and in both instances they had been dismantled. Although
Morris, excavating the "Annex" at Aztec Ruin (1924, pp. 235-236),
expresses doubt that troughed metates were ever binned, circumstantial
evidence from Pueblo Bonito points otherwise. In all our

digging there we found no trace of a metate, thick or thin, that was
not troughed, and Pepper reports none.[1]

Within the main walls of Pueblo del Arroyo we likewise observed
only two rooms equipped for milling, 41 and 55, and the bins in each
also had been stripped of their respective metates. Here, again, as at
Pueblo Bonito, we may not be positive that the missing mills were
troughed, but of all those reported in our field notes and catalog cards
only one is described as of the "flat" type with over-all grinding surface,
the binned example in Room 5. Since this is the exception, it
adds support to my conviction that Room 5 and its neighbors were
built by outsiders. A second, dubious specimen, recorded from Room
23 and described in my notes as 16 inches long by 1 inch thick, 10½
inches wide at one end and 6 inches at the other, may be one our Zuñi
installed in the Room 55 bin.

At Łeyit Kin, a small Chaco Canyon village apparently occupied
at the same time as the great pueblos, Bertha Dutton (1938, pp. 6768)
recovered 50 metates of which 5 were of the flat variety. From
a second small-house site about a mile to the west, Hibben (1937,
p. 90) reported 84 metates and metate fragments "all of a single type,
the open end trough." In the same ruin a year later Woodbury
(1939, p. 58) found 22 additional specimens, 5 with trough open at
both ends and 1 "of the plain surface (slab) type" without trough.
None was in a bin. Thus, by whatever adjective it is described, the
transversely flat, longitudinally concave metate with over-all grinding
surface is not a cultural trait of Pueblo del Arroyo, Pueblo Bonito,
and other major Chaco Canyon pueblos, although it may occur infrequently
in nearby contemporary small-house settlements.

As stated above, I believe Rooms 1-7 were built and occupied by
immigrants to Chaco Canyon. Everything about them looks alien:
careless masonry, small size and low ceilings, potsherd chinking in
walls, the above-floor ventilator in Kiva B. The binned metate in
Room 5 is not a local type, therefore its owner must have carried it
on her back, along with other possessions, from a former home. There
is nothing equally distinctive about the other artifacts recovered from
these rooms although some of the potsherds might also be regarded as
foreign. Potsherds were present in each room but not in rubbish-pile
quantities. As a matter of fact, only 1,559 nonduplicating sherds
were tabulated from the seven dwellings. Dr. Roberts will have more

to say about these fragments in his forthcoming monograph on the
ceramic remains of Chaco Canyon, and I only anticipate his analysis
by stating that, of the total, 67.6 percent belong in the four categories
I employed at Pueblo Bonito to approximate the proportion of pottery
manufactured by the Late Bonitians. One hundred sixty-three, or
10.4 percent, of the sherds were of the variety we called "Chaco-San
Juan."

Human skeletal remains were uncovered in the two adjoining rooms,
3 and 4. Upon approximately a foot and a half of constructional
debris or fallen masonry in the middle of Room 3 we found the right
half of an adult upper jaw (field No. 89). The remainder of that
same jaw (No. 92) was subsequently unearthed under like conditions
in Room 9, and there is no direct connection between 9 and 3. In
each case the fragment lay amidst broken stonework between 1 and 2
feet above the floor. There were no other human bones in Room 3,
but in addition to the fragment mentioned we removed from the west
end of Room 9 most of a disarticulated male skeleton (No. 91).

In the northeast corner of Room 4 windblown sand had collected
to a depth of 4 inches before the body of an infant (field No. 90),
wrapped in some sort of textile, was brought in and buried there, head
to the west. More sand was carried in to cover the little bundle and
to trail away in diminishing depth toward the west wall. The customary
debris of reconstruction had been dumped in upon the blown
sand, and among that debris, in the west half of the room, we found
the mandibles of a second infant and two adults. Fallen masonry had
thereafter collected wall high. Clearly Rooms 1-7 had been vacated
while families still dwelt in the village proper, on the opposite side
of the leaning wall against which 1-7 were built.