University of Virginia Library

II. RESIDENTIAL QUARTERS

In the pages that follow I shall describe a number of rooms that
seem significant for one reason or another. Some are former dwellings,
some had served for storage, some are of interest from the
architectural point of view, and some because of what we found in
them. To avoid the monotony of repetition, dimensions and room
fittings are listed in Appendixes B and C.

With the exception of Rooms 1-7 which are intrusives and, as such,
will be reserved for Chapter IV, Pueblo del Arroyo dwellings show a
surprising uniformity in size. This, despite the disparity between
Room 35, the smallest, measuring 8 feet by 12 feet 8 inches, and
Room 55, which is 13 feet 2 inches wide by 25 feet 9 inches long. The
average of those we excavated, 44 in number, is about 11 by 13½ feet.

Room 8 stands at the extreme southwestern corner of the pueblo
as originally planned, that is, before Kiva B and its associated rooms
were added.

A few years prior to my 1920 visit, to approximate time by the
amount of blown sand that had settled within its walls, the west half
of the second-story room had been cleared by treasure hunters. They
had broken a hole through the southwest corner of its floor (pl. 4, B)
and thus gained access to the lower room which they found in an
excellent state of preservation and relatively free from accumulated
rubbish. If the original inhabitants abandoned anything here, the
record has been lost for we recovered nothing but a single bone bead
and part of an awl.

A door in the north wall of this lower room, 8A, connects with
Room 16 and thence with 24 and 25. The lintel of that door consists
of eight pine poles about 3 inches in diameter; 4 inches below them
and 5 inches back from the wall face, a single secondary lintel pole
had served to support a doorslab or curtain. On the west, flush with
the northwest corner and 6 feet above the floor, a ventilator admitted
fresh air from the outside. At the east end of the room our treasure
hunters had forced their way through an apparently sealed door only
to be dissuaded from going farther by ceiling-high debris in Room 9A.
The same intruders had also pried a number of stones out of the south
wall revealing two pine logs, 6 inches in diameter, laid in horizontally
as longitudinal tie beams or stays completely enclosed by the masonry.
We learned subsequently that this method of strengthening walls was

in common use at Pueblo del Arroyo. Except for the hole broken
through from above, the ceiling of 8A was intact.

Rooms 8B-I and B-II. For some undiscovered reason the room
over 8A had been divided into two unequal parts by a masonry partition
(pl. 4, B). The eastern part, 8B-I, is the larger of the two, and
its south wall is 4 inches longer than the north. The partition, 8 inches
thick and composed of irregular blocks of sandstone strengthened by
two built-in pine posts, left 8B-II completely isolated except for a
possible hatchway to the third story. That the partition was a late
introduction is suggested by the presence of two south doors in 8B-I,
both giving access to the flat roof of Kiva B. The westernmost of these
two may have replaced the other when a shallow, adobe-rimmed fireplace
was built directly below its east jamb. When we uncovered it
the fireplace was still filled with wood ash, and plaster below the doorsill
was dark with soot. An unusual feature of this particular hearth
was the presence of two circular depressions, each about 5 inches in
diameter and an inch deep, at the northeast and northwest corners,
presumably as pot rests. Precautionary measures against the possibility
of fire are evidenced by an unusually thick adobe flooring about
the fireplace and over the underlying layers of bark and split cedar.

Between the north door, which had been closed with masonry, and
the northeast corner two sandstone slabs had been embedded in the
floor and rimmed with adobe mud to create a shallow basin somewhat
resembling those designed for grinding meal (pl. 4, A). But in this
instance the basin abuts the wall, and its floor slabs show no evidence
of grinding. Through the east wall a T-shaped door that formerly
opened into Room 9B-III had been partly blocked to leave a 12-inchdeep
recess on the 8B-I side.

Smoke-stained plaster still adhered to all four walls of 8B-I. The
room had been abandoned and a small amount of sandy debris had
collected before its ceiling collapsed. Broken ceiling poles, split-cedar
shakes, and 5-inch-thick chunks of adobe flooring provide clues to
construction. Some of the poles were slightly charred but destruction
by fire is not indicated. In and under this wreckage we found
pieces of tanned but unidentifiable skins, yucca-fiber cord, a drilled
bone awl, a small quantity of human hair, and fragments of two
sandals, one plaited (fig. 3) and the other woven. This latter, apparently
of apocynum fiber, bears a design in color on the upper surface
and, on the sole, a raised pattern produced by knotted threads. It is
of a type commonly attributed to an earlier civilization than that
represented by Pueblo del Arroyo.

Fig. 3.—Sandal fragment from Room
8B-I. (Drawn by Hashime Murayama.)

On the floor in the southeast
corner we found a piece of selenite
an inch and a half long
and a cube of lead ore, both
unworked. Elsewhere in the
room were a couple of hammerstones,
half a mano, a
handful of potsherds, a doorslab
that had been briefly used
as a metate (shown leaning
against the wall in plate 4, A),
and part of a sandstoneabraded
plank 7 inches wide, 4
inches thick, and about 2 feet
long. The doorslab, at least,
appeared to have fallen from
the third-story room.

Room 9A, adjoining Number 8 on the east, is noteworthy for its
exceptional length. Originally 111 feet long, it was subsequently
reduced to 58 feet 3 inches upon installation of the partitions creating
Rooms 10 and 11. The reduced room measures 5 feet 10 inches wide
at the east end; 6 feet, at the west. Since we cleared only the west
end of it, Room 11 may be divided by a possible third partition.

Whatever the idea behind it, Room 9 as originally planned was an
architectural mistake, a fact its builders soon discovered. Its south
wall began to settle outward even while under construction and, in an
effort to correct the error, eight low external buttresses were hastily
erected. Less than a foot of constructional debris—stone spalls and
mortar droppings—had collected on the surface when those buttresses
were installed. That the wall was leaning even before it reached ceiling
height is obvious from the fact that the partitions setting off
Rooms 10 and 11 are 5 inches wider at the top than at floor level.
The north wall likewise leaned southward but to a lesser degree (pl. 5,
right).

Blown sand and debris of occupation filled 9A almost to its ceiling.
Beneath and among this debris we found such typical tools and discards
as pieces of worked wood, bone, shell, and stone, 3 sandstone
disks or jar covers, 11 hammerstones, 10 manos, 1,357 miscellaneous
potsherds for study, and fragments of a willow screen. On the floor
close in the northwest corner lay a handful of turquoise and shell
chips from some jeweler's workbench.

At the west end of the room we came upon the partially disarticulated
skeleton of an adult male (U.S.N.M. No. 327141). The skull
had been crushed and some of the long bones thrown to one side. The
left half of the upper jaw was recovered several feet farther east,
about midway of the room and a foot above its floor. To our surprise
that fragment was the remainder of an upper jaw (field No. 89) we
had found 6 weeks earlier among broken masonry a foot and a half
above floor level in the middle of Room 3. Thirty-three inches of
solid masonry separates the two rooms.

Doors connect 9A with Room 15 on the north, Room 8 on the
west, and Room 1 on the south. The latter, like Room 3, is one of
several rude dwellings built against the outer south wall of the pueblo
some years after its completion, and the connecting doorway came
still later. The west door, which we restored, is believed to have
been blocked with masonry, perhaps at the time the adult male was
buried, for the treasure hunters who broke through from 8A had
destroyed most of the frame in their enthusiasm. A clay-lined hearth
lies in the middle of the floor, 24 feet 3 inches from its east end.

Absence of facing stones from an irregular area about 15 feet from
the northeast corner suggests an intended north door to connect with
Room 13. The missing stones were not present in the debris below,
so we brought in others and refaced the area as a security measure.

Rooms 9, 10, and 11 resulted from the partitioning of an exceptionally
long room. Their second stories include, not three, but six
rooms, and each of the five masonry walls separating them rests upon
paired beams at second-story floor level rather than upon first-story
stonework. Three of the six rooms overlie most of 9A and have been
designated, beginning with the easternmost, 9B-I, 9B-II, and 9B-III.
Figure 4 illustrates the relationship of the group. Neither the east
nor the west wall of Room 10A supports a second-story partition.

Rooms 9B-I to 9B-III connect with those adjoining on the north
but not with each other. All three had been lived in, for smokestained
plaster still adheres to their walls. The floors of 9B-I and
9B II had collapsed under the weight of masonry fallen from above,
but a remnant survived in the southeast corner of 9B-II and here
we found a number of discards, some of them partially embedded in
the adobe flooring: the reworked handle of a dipper (U.S.N.M.
334678), 2 bone awls (No. 334906), a clay figurine (fig. 37, a), 2 reworked
pieces of wooden tablets (pl. 38, i; No. 334702), a scrap of
cotton cloth (No. 334715), and a bit of kaolin. The very diversity of
these items suggests that they were among household rubbish dumped

Fig. 4.—Floor plan and profile of Rooms 8-11, first and second stories. (From the original survey by Oscar B. Walsh.)

upon a rain-soaked pavement and pressed into it by the overburden.
Also present was part of a sandstone mano that had been used in
preparing azurite pellets for blue paint.

Portions of third-story walls still stand here, and it is barely possible
there was once a fourth story.

Room 9B-III is best preserved of the three. Perhaps because its
supporting poles were not firmly seated, the floor next to the east
wall had settled 4 inches at one time, causing an unevenness that was
corrected by spreading a new layer of mud upon the old, thus bringing
the east third up to level.

A rectangular door opened into Room 15B, and another, this one
T-shaped, connected with 8B-I, next on the west. A second T-shaped
door gave access to the roof of Room 1B. The lower part of this
latter door once extended to the floor, but subsequently the sill was
raised to a height of 18 inches and then, for someone's convenience,
an adobe step, 24 inches wide, 3 inches high, and with an 8-inch tread,
was built immediately below.

Between this doorstep and the southwest corner of the room is a
semicircular fireplace, rimmed with adobe mud. Opposite, abutting
the north wall, partially sunk into the floor and extending out into
the room 33 inches, is a double receptacle—two shallow basins, one
before the other, and separated by a 2-inch-thick adobe partition. The
southernmost basin, paved with an oval sandstone slab, is enclosed on
three sides by an adobe rim, 5 inches thick and 4 inches high, bearing
imprints of juniper splints and inch-thick poles.

This double basin, the fireplace against the south wall, and the
smoke-blackened plaster unite in identifying Room 9B-III as family
living quarters. A sandstone metate, 16 inches long and 15 inches
wide, lay upon the floor in the northeast quarter. While clearing the
room we also noted 2 worn hammerstones, a modified flint flake,
4 worked sticks and part of a spindle shaft, a scrap of mammal skin,
part of a blue macaw feather, and an unidentifiable quill with a bit
of yucca cord attached.

Because the floor of 9B-III is part of the original ceiling of
Room 9A and in an excellent state of preservation (pl. 5, left), it
seems desirable to examine it closely since it illustrates the predominant
method of ceiling construction at Pueblo del Arroyo.

When the builders had raised the first-story walls to the height
desired, in this instance 7 feet above the floor, they placed paired
beams in position across the width of the room. The surviving beams
average 8 inches in diameter, and the spacing between pairs averages

49 inches. These beams were laid directly upon the sandstone-andmud
masonry without use of wall plates. Next, paired poles about
3 inches in diameter at the butt were laid upon the beams and at right
angles to them, that is, lengthwise of the room. Since such poles are
plainly visible in any ceiling construction seen from below, I prefer
to describe them as "ceiling poles." In this particular instance they
average about 7 feet long and their diameters decrease from 3 to
between 1 and 2 inches at the lesser end. To equalize this difference
the builders placed each pair butt to tip and the pairs about 6 inches
apart. At the west end of the room the poles are tenoned several
inches into the masonry.

Immediately upon the ceiling poles and at right angles to them
is a layer of juniper splints, each piece being about 2 feet long and
from 1 to 2 inches wide. Like the poles supporting them, many of
these splints have an end built into the stonework, ample proof that
wall construction proceeded as the several parts of the ceiling were
being brought together. To hold the layer in place, single splints were
laid crosswise over it at intervals corresponding to the paired ceiling
poles beneath and were lashed tightly to the latter with yucca-leaf
thongs. Then followed a layer of juniper bark and, finally, about 5
inches of adobe mud, packed and smoothed on top. Thus, the western
part of the ceiling of Room 9A became the floor of 9B-III. Excluding
the 8-inch supporting beams, the whole assemblage here is 10 inches
thick, an approximate average.

Upper rooms are usually a trifle larger than those directly beneath.
9B-III, for example, is about 10 inches wider than 9A owing
to the presence of two floor-level ledges or offsets that average,
respectively, 4 inches wide on the north and 6 inches wide on the
south. Apparently the layers of cedar splints and bark overlying the
ceiling poles so concealed the face of the lower wall that the masons
were not always able to keep its upward extension in alignment—if
they wished to do so. Less frequently an upper wall overhangs the
lower by an inch or two. There may be offsets at the ends of a room
as well as at the sides.

Two 8-inch logs rest upon the second-floor offsets and carry the
weight of the east wall of 9B-III. The masonry comes down on both
sides completely to conceal the two logs. Similarly, paired timbers
supported the other four second-story partitions in this room series.
Like those employed in ceiling construction, wall-supporting beams
usually extend entirely through the masonry on either side of a room
and end flush with the opposite wall face. It seems likely each log

was measured for correct length before being positioned. Rarely is
there any excess. One exception: an extra 13 inches on one of the
logs under the east end of 9B-I extended through the north side and
was hidden in the wall between Rooms 12B and 13B. A flint chip was
often used to draw a ring around a log—a line to which the builders
then hewed with their stone axes. More surprising still, the ax cuts
were customarily erased with sandstone abraders.

Timbers supporting these second-story partitions provided tree-ring
dates that help fix the time when Pueblo del Arroyo was building.
Several beams had decayed, and in one instance the annual growth
rings were of such uniform thickness they could not be read, but six
others gave the following cutting dates: A.D. 1097+x, 1100 (two),
1101, 1102, and 1103. With steel-ax-hewn logs from the Smith Lake
region south of Crownpoint, we replaced the rotted beams under the
partition between 9B-1 and 9B-II and rebuilt the wall above (pl. 5,
left).

Room 10A, as previously stated, came into being when two partitions
were introduced into an abnormally long room, the original 9A.
Like the latter, 10A was utilized for a time as living quarters, for
there is a slab-lined fireplace in the middle east half of the floor and
the wall plaster is sooted from fires that once burned in it. The lining
slabs stand upon an earlier floor 7 inches lower and the space between
is packed with adobe spalls and sand.

A remodeled doorway of more than usual interest pierces the south
wall 7 feet 4 inches from the southeast corner. As we interpret its
puzzling features, the opening was originally 25 inches high, 18½ inches
wide at the lintels, and 20 inches wide at the sill which remains 17
inches above the later floor. Presumably after construction of Room 7
this opening was neatly closed from the outside with stonework matching
the exterior masonry but leaving a 24-inch-deep recess, full
height in Room 10A. Then a larger, substitute doorway was cut
through above, giving access to the roof of Room 7, and two of the
former lintel poles were repositioned to serve, together with the sill
of the recess, as steps to the substitute opening with its sill at a
height of 4 feet 4 inches (fig. 5).

Under household debris in the northeast quarter of 10A, between
2 and 6 inches above the floor, we recovered the disarticulated skeleton
of an adult male (U.S.N.M. No. 327139). Like that in Room 9A,
its skull had been crushed and there was no accompanying burial
furniture. From the waste surrounding the Room 10 skeleton we
recovered 447 miscellaneous potsherds, 7 manos, 14 hammers, and a
few lesser artifacts.

With two and possibly three stories above them, Rooms 9 and 10
normally would have been utilized for storage. But both had been
plastered and provided with fireplaces, fixtures not typical of storerooms,
and their walls are still smoke blackened. Although dark and
ill ventilated, the fact that these two had been lived in reminds one of
an observation recorded by Mindeleff (1891, p. 103), namely, that the
Hopi formerly moved from the upper stories to ground-floor rooms
each fall in order to economize on fuel consumption during the winter.

Fig. 5.—Remodeled south doorway of Room 10A.

Room 11A was not excavated. We merely cleared a narrow strip
at the west end in order to note floor level and wall relationships.
In this limited area, however, we unearthed the skeleton of an adolescent
(field No. 154), sex not determined, three bone awls (U.S.N.M.
No. 334890), and a bone scraper (pl. 37, s). Here, as with the two
adults buried in Rooms 9 and 10, there was no recognizable burial
offering.

Rooms 12-24. Adjoining Rooms 8 to 11 on the north, a series of
paired rooms extends lengthwise of the south wing. With the possible
exception of the easternmost, which we did not explore, each pair
has a connecting doorway; in addition, the two pairs at the west end
of the series also connect with rooms both north and south. Ceiling
height varies from 6 feet 1 inch in Room 13 to 6 feet 10 inches in
Room 16, and averages 6 feet 5 inches. There were no subfloor walls.

The masonry of this group is among the best in the village but is
rather mixed as to type. Laminate sandstone predominates but there
is wide use, in some walls more than in others, of dressed blocks of
friable sandstone. Chinking with small chips is at a minimum. The
uniformity of first-story rooms was largely repeated in those of the
second and third stories. Many of the upper walls had fallen, in
whole or in part, but the quantity of fallen masonry does not, in my
opinion, indicate a fourth story for the entire group.

We excavated 9 of the 12 ground-floor rooms in this series. None
had been utilized as a dwelling; none had plastered walls; one only,
Room 20, boasted a fireplace and that is an 11-inch-square box built
into the east end of an oval depression, 5 feet 4 inches long by 57
inches wide by 11 inches deep, burned but containing no ash, and
subsequently floored over except for the 11-inch-square box mentioned.
Only one room, 24, had been provided with a ventilator and
that in the west, or outer, wall 5 feet 10 inches above the floor. It
apparently was short-lived for it had been neatly sealed from the outside
with matching masonry. A T-shaped door in the same wall had
likewise been blocked but with no attempt toward disguising.

Parts of crushed ceilings were present in Rooms 12, 16, 20, and 23.
The customary sequence of beams, poles, juniper shakes, bark, and
adobe mud was varied only in Room 20, where the bark had been
omitted. At the east end of this room the ceiling had settled but without
destroying a corner hatchway (pl. 6, A). Opposite, at the west
end, about 4 feet of ceiling remained intact, forming an open space
beneath. Pot hunters had discovered this hollow and from it had
cut a hole into Room 12—damage that we repaired.

The hatchway in the southeast corner measured 25 by 37 inches;
its western end was formed by the eastern main beam, 37 inches from
the wall, and its northern margin by one of the ceiling poles. On the
east the opening was bordered by two pine poles about 4 feet long.
Since their square-cut ends abutted the south wall instead of being
tenoned into it, and were not supported from below (we propped them
for the photograph), the two poles must have been held in place
solely by the weight of the 5-inch layer of adobe flooring that covered
the north half of them. Similar hatchways doubtless were far more
common at Pueblo del Arroyo than our data indicate. One probably
connected the first and second stories of Room 21 whose surroundings
are identical with those of 20.

Room 20 was largely filled with blown sand that apparently had
been carried in purposefully and dumped through the second-story

hatchway. Other than a hatful of miscellaneous potsherds, no artifacts
were present. But on the floor and under the sand lay the
skeleton of a puppy.

A youth of 15 or 16 had died on his sleeping mat in the middle of
Room 13A. Before decomposition was complete vandals had entered
the room, kicked the right pelvis to one side, and thrown the left leg
against the south wall, 4 feet away (pl. 13, A). Shortly thereafter
sand began to accumulate in the southwest corner, doubtless washed in
through a hole in the ceiling, and as the pile grew it spread across the
floor to fill the body cavity of the skeleton and half cover the skull.

Under the torso a mat of plaited rushes had left its imprint in the
adobe floor; the form of a coiled basket 4¾ inches in diameter was preserved
in the sand near the skull; at the left of the body and paralleling
it was the partial impression of a long, unidentifiable object made
of willows and reeds. Floor sweepings, including wood ashes and a
small quantity of potsherds, had been dumped through the north door
until they formed a considerable pile. Subsequently wall masonry and
flooring had crashed through from above, half filling the lower room.
From among this wreckage we recovered a single artifact, a crude
sandstone disk 4 inches in diameter.

Parts of an adult female skeleton were scattered among fallen
masonry or debris of reconstruction throughout the eastern half of
Room 21A and from 2 to 3 feet above the floor.

Room 15 was of particular interest to us because of its diversified
contents, mostly fallen from the second story. These are:

                 

	Plate	Figures
3 bowls	22	d-f
1 pitcher	28	f
2 ollas	30	a, b
4 cylindrical vases	24	f
	55	a-c
1 "feather box"	24	j
2 seed jars	29	g, h
1 corrugated pot	32	d

In addition to the vessels, we unearthed two small hammerstones,
three worked flakes (U.S.N.M. No. 334792), part of a shell pendant
(Conus interruptus Brod.) from the west coast of Mexico (No.
334734), 406 tabulated potsherds of which 8.9 percent were ChacoSan
Juan, and fragments of two human femurs (field No. 118).

Room 16 lies north of 8 and west of 15. Its ceiling, though broken,
was a typical one: two pine beams across the width of the room, their
ends firmly seated in the north and south walls 6 feet 10 inches above

the floor; 16 selected pine poles lying east-west upon the beams; at
right angles to the poles, a layer of juniper shakes, then juniper bark
and, finally, a layer of adobe mud to provide a floor for the room
above. When the second- and third-story west walls gave way, much
of their stonework crashed through the second-story floor into the
lower room and forced to one side a squared pine timber that may
have propped a previously cracked beam.

That timber is unique in Southwestern archeology as far as I know.
It measures 5 feet 10¾ inches long by 8½ inches wide by 5½ inches
thick and thus closely approximates the ax-hewn railway crossties of
50 years ago. Indeed, more than one camp visitor mistook it for a
crosstie before attention was directed to its stone-ax-cut ends and
sandstone-abraded sides (pl. 39).

As found, the timber leaned against the south wall in a way that
brought its lower end near, but not directly under, the break in the
western ceiling beam (pls. 7, A; 8, A). Although our initial thought
was of a prop for a beam already cracked and sagging in the middle,
there was no imprint of the squared end in the adobe floor and we
observed no stone identifiable as a pedestal. Actually, the lower end
was 1½ inches above the floor, completely surrounded and weighted
down by the broken masonry and debris fallen from above. Nevertheless,
our initial guess may have been correct, for in length the timber
is just a foot less than the height of the beam-end seatings.

The only other items found in the wreckage of Room 16 were parts
of a decayed willow screen, five sections of peeled willows each
14 inches in length, flint cut and broken at both ends (pl. 38, a),
pieces of yucca-fiber cord, and a number of adult human bones. The
screen, originally about 28 by 37 inches, consisted of dressed willows
sewed together at intervals of 4¼ inches (pl. 7, B).

Above the broken second-story floor we collected part of a spindle
shaft; part of a squared and abraded pine plank 2¼ inches wide by
1⅜ inches thick and still 32 inches long; and what appears to be a
section cut from a bow made of mountain mahogany (pl. 38, d).
Smoothed with fine-grained sandstone, the piece is 20 inches long and
⅞ inch in diameter at the rounded butt. Here minute nicks left by a
flint flake show how the shaft was girdled preparatory to breaking.
An irregularity at the opposite end has been smoothed but there is no
bowstring nock.

Room 23, like 15 next on the south, was practically empty when its
ceiling collapsed under weight of masonry fallen from above. Indeed,
at the west end of the room ceiling poles and their overlying layer of

split juniper still stood, pressed flat against the wall as our examination
began. Blown sand had accumulated among and upon the broken
timbers and fragments of adobe flooring. In this mixture, clearly
fallen from one of the upper stories, were five remarkable sandstone
tablets and several other artifacts.

The five tablets are remarkable for their uniform thinness (average,
⅜ inch) and for the perfection of their workmanship. Four are shown
on plate 42, d-g; the fifth, more shattered than the others, was not
photographed. All are rectangular (average 15[fraction 1 by 16] by 8[fraction 9 by 16] inches) and
of very fine-grained calcareous sandstone (or siltstone); all were
broken in falling. Two are slightly discolored by smoke and one bears
the stain of twilled matting. The opposite side of this latter tablet
is as clean as its companions.

With these five tablets was one of dark gray carboniferous limestone,
discoidal in shape and weighing 7⅛ pounds in its present fragmentary
condition (pl. 42, i). A sixth rectangular tablet (pl. 42, j),
from adjoining Room 27, is made of this same slaty rock which is
not known to occur between the Zuñi uplift, in the Fort Wingate section
of New Mexico, and Durango, Colo., an air-line distance in
either case of over 50 miles from Pueblo del Arroyo.

These seven stone tablets evidence skill and boundless patience.
They were reduced to their present form solely through abrasion.
None exhibits any mark that might suggest a clue to its original use.
Since those of sandstone, especially, were too fragile for any conceivable
utilitarian purpose, it is our guess that all seven were employed
in ritualistic observances of the unnamed clan that occupied
the suite of rooms connecting with Room 23.

Our surmise is strengthened by the number of other unusual stone
and earthenware artifacts recovered from these same rooms—artifacts
not to be confused with ordinary culinary utensils. As is well
known, Pueblo Indians still store the ceremonial paraphernalia peculiar
to each society in dark, interior rooms of the house recognized as the
ancestral home of that society. Fewkes (1904, pp. 104-106) reports
the use of painted slabs on Hopi altars and the finding of similar
slabs in prehistoric ruins; Morris (1919, p. 24) describes polished
slabs from Aztec Ruin that seem entirely comparable to those before
us. Of our seven, however, only one (pl. 42, d) bears any trace of
paint and that is a yellow wash applied to one side only.

It is noteworthy that although 13 pottery vessels were recovered
from Room 15, adjoining, none was found in the wreckage of Room
23. However, in addition to the tablets described above, we unearthed

half a sandstone doorslab on which yellow, then red, ocher
had been ground for paint; a sandal-shaped tablet (pl. 42, h); a
grooved stone ax made from a quartzite cobble, flat on one side, still
pitted by the hammerstone with which it was shaped, but spalled at
the bit through carelessness (pl. 41, b); a sandstone jar cover and
parts of three others; a plug of sandy clay, burned but still friable,
for a storage jar whose orifice was exactly 3 inches in diameter
(pl. 40, w); a single white flint arrowhead (field No. 112); two crude
handstones for smoothing adobe floors; an unfinished metate
(U.S.N.M. No. 334856); a couple of manos and two metates that are
much narrower than those characteristic of Chaco Canyon.

The smaller of these two milling stones measured 13½ inches long
by 1¾ inches thick, by 6 inches wide at one end and 7 inches at the
other. Its mano groove was 10½ inches long, 4½ inches wide, and
½ inch deep. The second metate measured 16 inches long, 1 inch
thick, 6 inches wide at one end and 10½ at the other. We left these
in the room where found and, with them, a "lap stone" 14 by 9½ inches
by 1½ inches thick. Tool marks and worn areas on the sides of this
specimen identify it as a sort of portable work table.

Our observations in this series of ground-floor rooms, 12 to 24,
suggest they were utilized primarily for storage. None had plastered
walls; only one, Room 20, was provided with a fireplace and this was
plainly of later introduction. The north doors in Rooms 12 and 16 still
preserved secondary lintels for support of doorslabs. A youth had
died on the floor of Room 13; an adult female apparently had been
buried on a partial fill in Room 21 since her scattered skeleton was
found from 2 to 3 feet above the floor. A third interment may have
been made in front of the partially blocked south door in Room 24
because here we found a number of miscellaneous human bones. Other
bones, perhaps from the same skeleton, were recovered in Room 16
and parts of two femurs in the fill of Room 15.

Despite the presence of these burials, shallowly covered with household
rubbish and debris of reconstruction, families continued to occupy
the rooms above. Practically all the artifacts we retrieved from
this section of the ruin clearly had fallen from upper rooms. Excepting
the intentional fill in Room 20, relatively little blown sand had
collected in these ground-floor rooms prior to collapse of their ceilings.
In only one room, 24, were evidences of burning conspicuous,
and here fallen wall material and blown sand together measured
33 inches in depth before the conflagration left charred timbers and
a blackened east wall. A layer of burned wood, building stones, and
adobe flooring on the outer west side of Rooms 8 and 16 and about

2 feet above their floor levels is doubtless wreckage from the same
conflagration.

Reference to the ground plan (fig. 2) will show that my consideration
of this series of paired rooms as a unit has no real significance.
A more logical grouping would have joined the three western pairs
with the four rooms adjoining on the north, for all 10 are connected
by doors and these doors are repeated not only in the second story
but, where walls still stand, in the third story also. Presence of
Kiva C prohibited north doorways in the first and second stories of
Rooms 20 and 21.

Room 27, in construction at least, is representative of the best at
Pueblo del Arroyo. It included three, probably four, stories and each
superposed room, like that at ground level, connected with rooms
adjoining on the east, west, and south. The north wall for each story
was blank. We feel reasonably confident of a fourth tier because
much of the third was still intact and the quantity of fallen masonry
was such that it filled the basement room and came to door level in
the second story. Many of the broken beams and ceiling-pole fragments
in this debris had been charred to some extent but not enough
to warrant belief that the place had been destroyed by fire.

After having been in use for an unknown length of time, both east
and west first-story doors were carefully blocked. The first was sealed
from Room 27 in such manner as to leave a recess on the Room 28
side; the second was closed from Room 26 to leave a 15-inch-deep
recess in 27. Thereafter the only entrance to this room, excepting a
possible hatchway of which we saw no evidence, was the south door
leading to Room 23 and thence to 15 and 9. Each of these basement
rooms was equally dark and poorly ventilated; no ray of sunlight ever
pierced their enclosing walls. It seems unlikely that any one of the
four was ever used for anything but storage.

Pueblo del Arroyo families lived in the second, third, and fourth
stories. Here the stone walls were invariably plastered and, although
it may have weathered from exposed surfaces, plaster is usually to
be seen in corners even now. The outermost room in any unit was
the living room, lightest of all. The occupying family, or families, in
accord with deep-rooted Pueblo custom, doubtless folded their sleeping
mats and blankets when not in use and stacked them at one side;
their extra clothing, together with dried herbs and other foodstuffs,
hung from suspended shelves in the inner rooms. Cooking utensils
and jars of precious water were ranged along the walls, safe from
careless feet. A hearth was usually present and mills for the grinding
of maize.

While removing the constructional debris that completely filled the
first story of Room 27 we came upon the following earthenware
vessels:

                             

	Plate	Figures
4 bowls	22	g-i
	24	c
1 bird-shaped bowl	24	h
5 ladles and fragments	26	a
	27	b, f-h
1 pitcher	28	g
6 ollas	30	c-f
	31	a
	32	b
2 seed jars	29	d-e
1 small coiled jar	24	i
3 corrugated pots	33	g-i
2 bifurcated vessels	35	a, b
1 small seed jar	29	b

All were broken except the little bird-shaped bowl which was found
close in the southwest corner. The jumble of building materials was
so complete we could not separate it into its successive levels. Most
of the broken vessels were recovered between 2 and 3 feet above the
floor.

In addition to the pottery there were recovered from the Room 27
fill the double paint mortar seen in plate 41, k, a rectangular slate
tablet (pl. 42, j), a stone disk 1⅝ inches in diameter (U.S.N.M.
No. 334807), a piece of turquoise squared for mosaic (No. 334741),
an antler wedge fragment (No. 334934), and, left in the room, a sandstone
tablet measuring 7 by 15 by ¾ inches, 11 cobblestones, 3 manos,
1 hammerstone, 6 small rubbing stones, 6 smoothing stones averaging
3 by 6 by ½ inches, and 191 tabulated potsherds.

A glance at the illustrations will show the character of this earthenware
assemblage. It is surprising to find only four bowls in the lot
and each larger than the Pueblo Bonito average. Fragments of the two
representations of bifurcated baskets and those of the threefold, redware
ladle were recovered about 2 feet above the floor. None of these
three items was designed for everyday use and yet their fragments
were intermingled with those of water jars and corrugated pots.

The polished slate tablet (pl. 42, j) and the double paint mortar
(pl. 41, k), the latter found intact in the southwest corner 3 feet
above the floor, likewise may have been connected with activities not
purely domestic. Paints were indispensable in the preparation of
prayer sticks and altar paraphernalia, and red paint is still a necessity
in completing the individual toilet, male or female. Our Zuñi workmen

coveted every scrap of red ocher not wanted for the collections
and forthwith made a rouge to brighten cheek or brow.

Our notes on excavation of Room 27 report, also, the finding of
divers other artifacts: a sandstone disk 1⅝ inches in diameter by
⅛ inch thick (U.S.N.M. No. 334807), a bit of turquoise squared for
inlay, the fragment of an antler wedge (No. 334934), 6 well-worked
rectangular smoothing stones (average 6 by 3 by ½ inches) and as
many smaller ones, a crude hammerstone, 3 manos, part of a sandstone
tablet measuring 15 by 7 by ¾ inches, and 11 rather large,
waterworn cobbles. Most of these items were left in the room after
completion of our examination.

Room 28 is a large room, almost 250 square feet in floor area. Its
north and east walls are blank. South and west doors once gave access
to adjoining rooms but the west door eventually was blocked from
Room 27, leaving in Room 28 an 18-inch-deep recess, 26 inches wide
by 44 inches high. Three large pine beams supported the ceiling and
the floor of the room above. These were not strong enough, however,
for occupants of the second story had propped each beam with a
sturdy post.

The second-story east and north walls, fallen long ago, had been
firmly bonded with those of the south and west. A possible thirdstory
east wall, or, more likely a buttress, is suggested at the southeast
corner where there had been union with the outer third-story
masonry of Rooms 21 and 22. There was no corresponding union
at the northwest corner. Hence, if Room 28 lacked a third story the
inhabitants of 22C and 27C stepped out upon its flat roof and utilized
it as an open-air living room.

Collapsing masonry had crashed down, filling Room 28 to a point
above its second-story floor level, a total of 9 feet. In this fill we
encountered a layer of charred wood sloping from ceiling height in
the southwest corner to half that in the northeast quarter. Above that
layer were several feet of broken stonework and blown sand. Below
it and chiefly throughout the northen half of the room was a dump
from which we recovered 18 hammerstones, 3 smoothing stones, 9
manos and parts of 8 others, a copper bell (U.S.N.M. No. 334763),
part of an obsidian blade (pl. 40, n), and 8 restorable earthenware
vessels, including:

           

	Plate	Figures
2 bowls	22	c
	23	a
1 pitcher	28	j
2 ollas	31	d, e
1 canteen	28	d

Plate 2

Upper: Beginning excavations, outer south side of Pueblo del Arroyo. The second wheelbarrow
is loading in Room 7. Jackson's "old arroyo" at left.

(Photograph by O. C. Havens, 1923.)

Lower: Rooms 1 to 7 after excavation. In the left foreground, the unnumbered room east
of 7. Lighter areas in the original walls identify Expedition repairs.

(Photograph by O. C. Havens, 1925.)

Plate 3

Upper: The unnumbered second-story room at the northeast corner of the south wing had
been roofed and utilized for storage. Chicken yard at left.

(Photograph by Charles Martin, 1920.)

Lower: Northeast corner of the south wing, showing 1925 repairs. Salvaged stones for
wall repair piled at left.

(Photographed by O. C. Havens, 1925.)

Room 32 adjoins 28 on the north and is almost exactly the same
size. Its three east-west ceiling beams had also been propped in the
middle by posts. In this case, however, the holes in which the posts
were seated had been lined with pieces of sandstone slabs. Traces of
adobe plaster persist on the north wall, not on the others. A north
ventilator 4 feet 8 inches above the floor opens into Room 35 and a
similar one in the second story, now blocked, formerly opened into
35B. We detected no positive indication of a third story.

The only known means of access to Room 32A was the door in its
west wall. That door led into Room 31 and thence into 30 and 29.
Either or all of these four rooms might have been provided with a
ceiling hatchway but neither connects with the rooms adjoining on the
north and south. The north and south walls of all four are doorless,
a condition repeated in the second story and, where present, in the
third also. Here was another isolated block of rooms!

There is a north-wall ventilator in 32A. North ventilators are to
be seen, also, in 30B, 31B, and 32B, although the last named had been
sealed. Room 29A has a west ventilator but none in the north wall.

The artifacts we recovered from the wreckage of Room 28 gave
evidence of domestic activities pursued in the room or rooms above
and this was equally true of our finds in 32. Here we unearthed 5
metates and parts of 4 others, 2 whole manos and 2 fragments, 2
smoothing stones, 6 hammers, a single small polishing stone, and a
rather large piece of obsidian. We left all these in the room. One of
the metates, measuring 21 by 16 inches, had a mano trough on each
side. A number of lesser artifacts will be presented in chapter V but
I cannot resist directing attention at this time to the broken obsidian
blade illustrated as figure n, plate 40. The lower half of it was found
in Room 32, the tip in 28, and no door connects the two.

Unlike those in the south wing, rooms of the central section evidence
repeated alteration and revision of the basic plan. In none of
the rooms heretofore described were subfloor walls disclosed by our
testing, while in those next to be considered substitution and replacement
occurred frequently. For example, the floor of Room 34 is
18 inches lower than that of 30, next on the south (fig. 6), and its
three beams lay lengthwise of the room rather than across its shorter
dimension. Kiva D was built in a former dwelling, thus blocking
doors to Rooms 35, 39, and 40. Plastered walls identify Rooms 36
and 37 as living quarters occupying part of the site originally intended
for a kiva.

Sometime after its east door was closed, Room 39A became a
neighborhood dump. Approximately 4 feet of household waste and

Fig. 6.—Profile showing difference in floor
levels in adjoining Rooms 30 and 34.

blown sand had collected
here before the upper walls
gave way. From this mixture
of sand and sweepings
we recovered a number of
bone and stone implements,
shell ornaments and fragments
of ornaments, 2 tabular
metates, 5 manos, 2
thin sandstone slabs with
chipped edges, a discoidal
stone 11 inches in diameter
by 3 inches thick, pecked on
one side but rough on the
other, and 10 restorable
earthenware vessels. In accord
with my instructions,
the larger stone implements
were left in the room.

Room 40 is one of four
adjoining dwellings on the
site of a kiva that apparently
was never completed.
Room 40's smoke-stained
walls had been plastered at
least twice. Four large
north-south beams at a
height of 7 feet 4 inches carried 29 ceiling poles. The west ends of
these poles instead of being tenoned in the masonry had rested upon
the westernmost beam, seated close against the wall. Below the poles
two anciently repaired areas suggest that the ceiling timbers of the
room next on the west had been removed before construction of
Kiva D. Both dressed and undressed blocks of sandstone went into
the work of repair.

There were two clay-lined fireplaces in Room 40, one in the northeast
corner and the other against the south wall. An irregular, claylined
bin in the southeast corner, over 3 feet square and 10 inches
deep, was filled with wood ashes. Between this feature and the northeast
fireplace is a clay-lined hole, somewhat irregular but about
8 inches in diameter by 16 inches deep. Its size and depth suggest
the seating of a former beam prop.

Room 40 ceased to function as a dwelling when a young man was

buried on the floor in the southwest corner. Thereafter all doors not
previously blocked were sealed with coarse stonework. The body lay
on its right side, face to the west, and legs flexed (pl. 12, A). Two
pieces of split cedar a couple of feet long lay just beyond the head and
longer slabs at the back, with irregular blocks of sandstone piled
on top. The presence of these sandstone blocks is unusual, in our
Chaco Canyon experience. A few casual spalls were under the body
but the larger pieces had been placed to surround it, although haphazardly.
As burial offerings a bowl and a pitcher had been placed
beyond the head and outside the encircling stones. These two vessels
may be seen to better advantage on plate 24, figures d, e.

Blown sand and debris of occupation had been carried in to complete
the interment. In this debris, or covered by it, were 3 large tabular
metates, 3 smoothing stones, a sandstone disk, 5 hammerstones, and
3,883 miscellaneous potsherds for analysis after all recognizable duplicates
had been put aside. Alone on the floor in the northeast corner
of the room was the occipital bone of an infant.

An earlier floor at a depth of 13 inches is that of a Chaco-type
kiva that appears to have been abandoned even before it was well
started. Our test trenches did not disclose the bench curve, but the
end of a typical subfloor ventilator duct, 17 inches wide by 26 inches
deep, masonry lined and clay floored, lies 26 inches from the middle
of the south wall. A foot beyond is the accompanying fireplace, 20
inches in diameter, clay lined, and filled with ashes. I am sure I am
correct in the recollection that every Chaco Canyon kiva we cleared,
whether it represented the beginning of construction or abandonment
after long use, had a fireplace filled with wood ash.

The ventilator duct of the razed kiva passes under the south wall
of Room 40 at a point 8 feet 5 inches from its southwest corner,
continues 4 feet 8 inches into Room 36, and then turns abruptly to
the east. From the north wall to the abrupt angle the duct had been
roofed with small poles overlain by thin sandstone slabs. Presumably,
the kiva plans were discarded at this stage for, reduced to a width of
13 inches and a depth of 20, the unfinished duct was crossed by the
east wall foundation and we discovered no extension of it in Room 37.
Furthermore, four larger poles, each about 6 feet long, had been laid
across the covered duct between the abrupt angle and the north wall
as though to provide additional support for a sandstone-and-abode
pavement, 56 inches wide and 9 inches thick, that crosses Room 36
in a northwest-southeast direction and underlies its north, east, and
west walls. The south edge of this "pavement" was left unfinished,

or had been razed, 4½ feet from the east wall of the room and 2 feet
10 inches from the south.

A similar feature appeared in Room 40, immediately underlying
the floor. In this case, however, the pavement, 5 inches thick instead
of 9, and 5 feet 2 inches wide at the west, abuts the middle west-wall
masonry above its foundation and extends thence eastward about 13
feet where it ends, somewhat irregularly, with a width of 4 feet
5 inches. Neither here nor in Room 36 did we locate the perimeter of
the intended kiva nor did we discover the significance of the overlying
"pavements."

Room 41, adjoining 40 on the east, had undergone repeated alteration.
East and west across the middle of it is a series of demolished
mealing bins, apparently 3 in number (pl. 11, A). Both the metates
and the slabstones that once enclosed them had been removed by the
former occupants. At the west end of the series a rectangular, slablined
basin measuring 16 by 27 inches by 6 inches deep was identified
by our Zuñi workmen as a receptacle for ground meal, while an oval,
clay-lined, and ash-filled depression adjoining the easternmost bin
was described as a place for live coals to provide warmth for those
at the mills. Another clay-lined depression nearby, 18 inches in diameter,
was ash free although reddened by fire.

There had once been a doorway through each of the four walls but,
for some reason, each had been wholly or partially blocked. The
original north door, 5 feet 8 inches from the northwest corner of the
room, clearly was closed in consequence of the construction of Kiva J,
for a substitute was soon provided. This latter has its sill 30 inches
above the floor, a height that necessitated an assist of some sort. The
sill is 14 inches deep and, from it, four built-in steps gave access to
the court at the kiva roof level (fig. 7).

The first step rises 15 inches to an 11-inch tread; the second step
has a 12-inch riser and a depth of 10½ inches. At this level both jambs
had been torn out and replaced with a new masonry facing that extends
northward an additional 14 inches to enclose the third step. This
latter has a 17-inch-high riser that slopes to the rear and reduces its
tread to 9 inches. The fourth step, only 6 inches high, reached the
Kiva J roof.

A second blocked north-wall opening measures 45 inches wide by
28 inches high with a sill 31 inches above the floor. The dimensions
suggest a former cupboard or recess. Large, unworked pieces of
sandstone had gone into the closing of it.

A curious and quite incomprehensible figure, composed of more or

less parallel lines with crosswise or lengthwise hatching, had been
scratched into the wall plaster between the sealed recess and the
stepped door. It is a less impressive example of Chaco Canyon art
than the sandal figure on the north wall of Room 44 (pl. 14, B).

Room 41 had not long been vacated, if at all, when its ceiling collapsed.
Some of the timbers were partially burned, especially those

in the northeast quarter, and among them blown sand had subsquently
collected. Between and beneath the charred timbers we recovered a
miscellany of artifacts, including 5 bone awls and part of a scraper
made from the shoulder blade of, apparently, an elk, 2 arrowheads,
a tubular bone bead, a stone ax (pl. 41, d), 7 stone jar covers, 3 small
sandstone tablets at least one of which had been used in the manufacture
of beads, a large turquoise pendant, burned and broken
(fig. 8), 8 stone hammers, half a beaver's incisor (Castor canadensis)
with a knifelike edge, and 2 large tabular metates each with its associated
mano or hand stone. If the reader can mentally substitute
steel implements for those of bone and stone, this varied assortment
is just what one might see in the living room of a present-day Pueblo
home.

Room 43 lies between 44, a ground-floor dwelling, and two storerooms,
42 and 45. The four clearly comprised a family suite. We
made no observations in the two storerooms, but someone, prior to
Mindeleff's 1887 visit, had done a little prospecting here (pl. 15, b)
and thrown the excavated material, including a single human femur,
upon the accumulation in Room 43B.

Fig. 8.—Turquoise pendant from Room 41.

Although Room 43 is not an unusually large room, two pairs of
beams, 10 to 11 inches in diameter, were installed to carry the weight
of its ceiling. But these proved insufficient. The south beam in the
northern pair had to be braced at both ends and a 6-inch post was
placed under the east end of the south beam in the other pair. Mud
applied to the walls when the room was last plastered filled the angles
behind these three posts and curved out on their sides.

A door connecting with Room 44 measures 4 feet 8 inches high, sill
to lintel poles. Although the sill is only 17 inches above the floor, a
height not at all out of the ordinary, it must have handicapped some
elderly member of the household, for a block of sandstone had been
embedded in the floor below to provide a 10-inch step (pl. 8, B).

West doors connect with the two storerooms, 42 and 45. Both
doors had been equipped with a second set of lintels and secondary
jambs having an outward batter for retention of doorslabs. Ruppert's
sketch, reproduced as figure 9, leaves further description unnecessary.
It is unusual to find, in the same wall, doors as unequal in size as these
two.

Three feet of windblown sand, earth, and floor sweepings had
gathered in Room 43 before its ceiling fell. From that accumulation
we removed a number of bone awls and shell beads, numerous chips

of flint, chert, and obsidian, pumpkin seeds and pinyon nuts, metate
and mano fragments, a sandstone slab with white paint, and 11 hammerstones
all but one of which were of either flint or silicified wood.
Two metates as well as other utensils had fallen from the second story.
Upon the sand and earth were the broken beams, poles, and juniper
splints of the ceiling and, above them, shattered masonry from the
upper walls. That the ceiling remained intact for some time after
abandonment of the room is further established by the presence of
several mouse and rat skeletons, and quantities of droppings, in the
lower 8 inches of the fill.

Room 44 held our attention for several reasons: It was a first-story
room that had been lived in and altered at least twice during occupancy;
its ceiling was partly intact and a built-in stairway led to the
roof level of Kiva F. Finally, neighbors had dumped in a lot of household
trash, and trash piles are often archeologically rewarding.

The ceiling was of customary local construction: First, two great
pine logs laid east-west, then pine poles, juniper splints, bark, and

adobe mud. In this case, however, a thin layer of adobe was spread
directly upon the splints; next came the bark and, upon that, more
adobe mud. Although the second-story east wall had fallen and broken
through, much of the second-story floor remained intact. Sometime
after its completion, however, four of its secondary supporting poles
had been cut off to allow for a small hatchway, 12 by 17 inches, in
the southeast corner. The opening was not a complete success, however,
for it had subsequently been closed with a couple of sticks and
a layer of reeds and then sealed by three successive floor surfaces,
each from ½ to ¾ inch thick.

Reconstruction in the ground-floor room was evidenced by the fact
that the two main beams, each 11½ inches in diameter, were only 5
feet 2 inches above the floor. Despite their size, both beams had
required bracing to help support the weight they carried. The south
beam, now with a fairly fresh-looking crack, was propped at the east
end only while its companion had been braced at both ends by 8-inch
posts set close against the walls. At least one of these posts stands
upon a sandstone slab 4 feet 4 inches below the floor or 9½ feet beneath
the beams. Since finished masonry of the south wall continues to a
point 9 feet 3 inches below the beams, there to rest upon a single
course of foundation stones, it seems likely the three props were installed
at the time of, or shortly following, initial construction of the
room.

The latest pavement in Room 44 was only 5 feet 2 inches below the
principal beams but test holes revealed earlier floors at depths of
17 and 32½ inches. If any architectural features are associated with
the lowermost floor they were not disclosed by our limited testing to
that depth. For reasons not immediately apparent, that lower floor
had been abandoned, and windblown sand and waste from the village
dump had been carried in and spread upon it. Then the fill itself was
covered by a layer of adobe mud, creating a new surface 15½ inches
above the older at the west side of the room and 2 inches higher at
the east side.

On this second floor, which is somewhat uneven, exploratory
trenches discovered a built-in stairway, a circular, slab-lined fireplace,
and two puzzling fixtures: a masonry "bin" in the southeast corner
and a neatly walled trench under the north side of the room. The
bin, finished on the outside only, is 44 inches square by 17 inches high.
It had been filled with sand and debris of reconstruction (among
which we found three discarded hammerstones, fragments of three
dressed sandstone slabs, and miscellaneous late postherds) and eventually
was concealed by the third and final floor.

The second fixture, a "trench" of small-stone masonry with concave
end and sides 18 inches apart, underlying the north wall 5 feet 4 inches
from its northwest corner, seems less puzzling in retrospect than at
the time we exposed it. The concave end is only 10 inches from the
north wall, but north-wall masonry extends down between the sides
of the structure for a like distance while the sides themselves continue
an additional 12 inches, a total of 22, to a point 6½ inches below
the earlier floor, that at a depth of 32½ inches.

This "trench" can only be the end of the subfloor ventilator tunnel
connected with the partially razed kiva beneath Room 47B. If my
surmise is correct, the vertical shaft that formerly stood upon the tunnel
end was razed and both the present north wall of Room 44 and
the subfloor at a depth of 17 inches were introduced at the same time.
Therefore, the lower floor, that at 32½ inches, is probably the original,
and the original north wall of 44 was probably torn out when the kiva
was built in 47A. Similarly, a 3-inch-deep concavity in the middle of
the east wall must mark a former door, closed and plastered over
when Kiva F was built.

Intrusion of Kiva F into the house group fronting Room 44 not
only blocked the former east door but prompted the opening of a substitute.
This substitute, 32 inches north of the old one, is 26 inches
wide and 35 inches high; its lintels are on a level with the ceiling
poles and its sill is 4 feet 2 inches above the contemporary floor. To
enter and leave by this elevated door, four steps were provided. They
are part of a block of solid masonry, 47 inches wide and extending out
into the room 52 inches (pl. 9, A). This masonry block abuts an
earlier coat of plaster on the north and east walls. The four steps,
repeatedly resurfaced, led to the doorsill 50 inches above the floor;
thence, five more steps continued through the door and a 28-inch-long
passageway to the roof level of Kiva F. That passageway, of masonry
finished on the inside only, is the same width as the door, 26 inches,
and abuts the outer east wall of Room 44.

The jambs of this elevated door are 27 inches wide, and the decayed
remains of their associated lintel poles lie 35 inches above the doorsill.
We could trace these remains only 15 inches, the outer face of
the wall having fallen. Since two 8-inch-high steps were built within
the door we assume the missing lintels were raised 8 to 12 inches
above their fellows to allow an equal headway.

A glance at Ruppert's drawing of this door (fig. 10) and the flight
of steps leading to and through it shows that the architects of Pueblo
del Arroyo were not bound by inflexible rules. To them a stairway

was not so much an expression of their art as a means toward an end.
It mattered little that the 10 steps varied in height from 6 to 15 inches
and in depth from 11 to 14. An average for the 10 risers is 8.5 inches;
for the 9 treads, 11.8 inches.

After this stairway had served its purpose for a time, some circumstance
brought about further alterations. Debris from the village
dump was again carried in and spread throughout the room in a layer
varying from 15 inches at the foot of the stairway, the height of the

bottom step, to 17 inches against the west wall. Then a new floor was
laid upon the debris, covering the masonry-walled bin in the southeast
corner and rounding off with a fresh coat of wall plaster.

But this third and final floor forced alterations at the west door.
Here the original floor was 18½ inches below the doorsill; the second
floor reduced that height to 3 inches, a fact that offered no obstacle to
passage. However, when 17 inches of waste was brought in and a
third floor laid, remodeling of the door was unavoidable. The change
was made from Room 43 by raising the doorsill 14 inches but leaving
10 inches of the old one bare on the opposite side. The final floor in
Room 44, therefore, is on a level with the remodeled sill of the west
door.

Despite its lowered ceiling, Room 44 was still habitable. Elimination
of the southeast corner bin allowed more space, but the middle
of the floor was again occupied by a fireplace, this one nearly square

instead of circular and lined with sandstone slabs that stood up unevenly
a couple of inches. In the north wall at the convenient height
of 47 inches, a small recess was provided, a neatly plastered repository
whose lower corners are rounded while those above are squared. Five
inches east of this pocket is the figure of a sandal, drawn by some
former resident upon the brown plaster with a bit of kaolin (pl. 14,
B).

A hole torn through the south wall is of recent origin, as evidenced
by marks of a steel pick upon the stones. Nearby, the name "C. F.
Jones" carved in the adobe plaster may identify the wielder of that
pick. As double insurance against oblivion the name was repeated
on the west end of the south beam, together with the indistinct date,
18—. To save the north beam at least temporarily we sawed off its
decayed east end and propped the remainder with an 8-inch post at
the outer corner of the stairway.

Eventually the occupants of Room 44A moved out and their home
became a dumping place for neighborhood rubbish. From this we
salvaged a number of discarded artifacts, some of which are illustrated
hereinafter, and several restorable pieces of pottery. Among
these latter none is more intriguing to an archeologist of the southwestern
United States than the bowl illustrated in plate 25. It is a
brown bowl with polished black interior ornamented with a linear
design in matte paint and is more fully described in chapter VI.

Another interesting find from Room 44 is the incomplete skeleton
of a macaw, one of those gorgeous red-blue-and-yellow parrots
(Ara macao) from tropical Mexico and south. The skeleton is especially
interesting to me because the sternal apparatus had been fractured
by a blow and subsequently healed. Presumably the bird had
bitten a careless finger somewhere on the long trail to Chaco Canyon,
or after arrival there, and had been felled by a stick in angry hands
with the result more fully described on page 127.

Foodstuffs represented among the household rubbish include deer
and antelope bones, pumpkin rind, stems, and seeds (C. pepo), pinyon
nuts, shells of the canyon walnut (Juglans major), seeds of prickly
pear (Opuntia sp.) and, most surprising of all, seven scales of the
Rio Grande gar (Lepisosteus tristoechus?; U.S.N.M. No. 334958).

In the second-story room, 44B, plaster still adhered to all four
walls when we bared them. At least three of the four doors originally
provided for 44B had later been blocked, including one of T shape
overlooking Kiva D.

Room 46 is of interest chiefly in connection with its neighbor, 47.

As represented on figure 2, both are portions of former structures
partially razed to satisfy ceremonial requirements. The original Room
47 was commandeered first and a kiva was built inside it. Eventually
this ceremonial chamber was replaced by Kiva E.

Room 46 likewise was appropriated, although only the north half
of it was actually required. Since the remainder of this former house
was thus rendered useless for all practical purposes it was abandoned
and half filled with debris of demolition. However, a section of ceiling
at the south end was left in place. The beam supporting it, 10 inches
in diameter, lies 8 feet 1 inch above its contemporary floor, and each
end is propped by a post close against the wall. We dug down and
found the base of the western post resting upon a sandstone slab
4½ inches below the floor. The reuse of salvaged timbers is proved
once more by the fact that this original south beam is dated A. D. 1070
while the post bracing it was felled 18 years earlier.

Sometime later, as though to utilize that ceiling remnant, a new
north wall was built in upon the fill together with an accompanying
floor 3 feet 8 inches below the old ceiling poles. Five of these latter
were tenoned in the new wall but the other eight were cut off a few
inches short. The space beneath, low and cramped as it must have
been, was suitable only for storage and this is the purpose to which
it had been put. We noted many pumpkin seeds and pinyon nuts on
the floor and, at the west end, a pile of soft white sandstone of the
sort sometimes used to whiten wall plaster. Access to this improvised
storage place was by means of a 19-by-22-inch hatchway 2 feet from
the southeast corner. This feature obviously belonged to the original
room since the lone ceiling pole supporting its north edge, while
tenoned into the new north wall, had been severed 2 feet from the
south.

The new north wall not only abutted the upper 4 feet of the plastered
first-story masonry but rose to block two doors in the second
story, 46B. One of these two was T-shaped and had formerly opened
into Room 47B, next on the east. In figure 2, therefore, the indicated
north wall of Room 46 is really at the second-story level and so, too,
is that of Room 47.

Room 47 had been subjected to earlier and even more extensive
alterations than its neighbor. Its smoke-blackened walls testify to
occupancy as a dwelling and the presence of an open hearth. After
an unknown interval the room had been preempted by some priestly
group that wished to erect a kiva on the site. The kiva was built and
then it, in turn, had been abandoned and partially razed. What remained

was filled with the debris of demolition and other waste; a
new floor was spread upon that fill, and a new north wall was built in
to complete a remodeled, second-story house. It is this later, secondstory
wall that is represented on our ground plan of Pueblo del
Arroyo (fig. 2).

We found the remodeled, second-story room an intriguing one. In
the first place, its north wall, which still stands to a height of 4½ feet,
was erected upon a shallow foundation that rests directly upon the
first-story fill. Secondly, that wall abuts from both sides the plastered
remnant of a former north-south wall, 25 inches thick, that had
stood upon paired 9-inch beams at second-story floor level and thus
formed two narrow but unequal compartments in the space we are
describing as Room 47B. The western compartment measured 44
inches wide at the north end; the eastern, 5 feet 8 inches.

Masonry walls built upon paired beams bridging a room below were
no novelty at Pueblo del Arroyo but, in the present instance, it is
clear that the original north-end support of the two logs had been
withdrawn, for here the pair had settled 10 inches under the weight
of the masonry they formerly carried. It was this settling, no doubt,
that brought about removal of their masonry wall and union of the
space on either side. The beams were left in place but the wall was
entirely razed except a veneering at each side of the paired beams
and a plastered remnant abutted by the built-in north wall. The near
end of this remnant, broken in line with abutting stonework on either
side, had been muddied over when the north wall was last plastered.

Instead of hiding the paired logs under an all-over floor, the builders
adzed down their upper surfaces and covered their sunken north ends
with a single layer of 1-inch slab fragments and a coating of adobe
mud. Although the intervening portion had been flattened on top, the
ends of the pair as embedded in the north and south walls retain their
full circumference. At the south, where no evidence of reconstruction
is to be seen, the pair rests upon the second-story floor offset and
extends through the masonry to terminate, square cut and abraded,
flush with the face of the north wall in Room 44B.

The two beams and the veneering at their sides remained as a
minor obstacle in reconstructed Room 47B, for the rebuilt floor,
spread upon the debris fill of the room beneath, was at two levels.
That east of the two logs lay 5 inches below the second-story floor
offset, and the western portion about 3 inches lower. Here the built-in
north wall abuts a blocked T-shaped door that once gave access to

Room 46B. In the corner below this blocked door is a masonry-lined
bin, unburned but ash filled (fig. 11).

In the larger, east half of the room blocked east and south doors
are to be seen, the latter only 5 inches from the one-time partition.
Below it is a 9-inch-deep fireplace, its east and west sides of slabs
and its north side of clay, with two burned sandstone firedogs lying
on the bottom (pl. 9, B). A second rectangular fireplace abuts the

middle of the north wall and between it and the first one is a circular
depression, 2 feet in diameter and 5 inches deep, clay lined and ash
filled. There were no other fixtures in the rebuilt, second-story room,
but fires lighted on the open floor had burned through the adobe in a
couple of places to scorch the paired beams of the old partition.

Beneath the divided floor was a purposeful fill consisting mostly of
constructional waste from razed walls. We put down two test pits,
one on either side of the paired logs, to learn what lay concealed.
Although somewhat handicapped for working space by the architectural
features already described, our pits reached the floor of the firststory
room and disclosed an interesting sequence in utilization.

Room 47A was rectangular, as are most rooms in Pueblo del
Arroyo, and it had been long occupied as a dwelling, for its walls were
not only plastered but thickly smoke stained. Such use was interrupted,
however, when the room was appropriated and a circular kiva,

or ceremonial chamber, built within its four walls. Because the north
wall of 47B stands upon a loose fill we made no search for the
northern ceiling beam of the original lower room, but that at the south,
a magnificent timber 12 inches in diameter, had been chopped off with
stone axes, leaving both ends embedded in the masonry. The west
stub, which gave us a tree-ring date of A. D. 1066, projects 9 inches
from the wall, and a former supporting post still stands beneath it.
One of the paired beams in the rebuilt second-story room was felled
11 years later, a fact that only adds to the tangled history of this
section of the ruin.

Ceremonial chambers in Chaco Canyon were traditionally cylindrical,
their domed ceilings of cribbed logs resting on 6 or 8 pilasters
equally spaced upon an encircling bench. Hence, when priestly builders
set out to construct a cylindrical kiva within the narrow confines of
rectangular Room 47A they were forced to compromise between ingrained
custom and solid stonework. At the south end of the former
dwelling the kiva masonry still stands 9 feet high, to within 14 inches
of the second-story floor offset. Its concave face curves evenly to the
west, half encompassing the post left standing under the beam end,
and then merges with the straight west side of the one-time residence.

A greater obstacle intervened on the east side. Here, to realize the
diameter previously decided upon, the kiva builders tore out just
enough of the old wall facing to allow for construction of the domed
ceiling—a section concave both laterally and vertically and with maximum
depth of 8 inches at bench level. To support the undisturbed
masonry above this concavity three lintel-like poles were laid in horizontally
49 inches above the bench and their ends embedded in the
kiva stonework. To tie this new stonework to the older, three lesser
poles were inserted at right angles to the kiva wall, their outer ends
flush with its concave face.

Our test pits disclosed a bench, 15 inches wide by 24½ inches high,
and two masonry pilasters thickly plastered and smoke stained. One
pilaster measured 14 inches long; the other, 12. They averaged
20 inches wide by 20 inches high, and each supported four 3-inch
poles, side by side, instead of the single pilaster log we expected to
find. The four poles rested on top of the stonework with a thick pad
of mud above. We removed two and found them to be 29 and 30
inches long, respectively, all embedded in the wall masonry except the
portion full length of the pilasters.

In the adobe mud above one set of pilaster poles we noted the
imprint of a horizontal log, clearly a lower member of the cribbed

ceiling. We noted, also, that the wall facing for a couple of feet above
each pilaster had been dislodged—evidence that the ceiling logs had
been wrenched free for use elsewhere. A south bench recess between
the two pilasters may be presumed. That which was previously described
as a "masonry-lined trench" extending from under the north
wall into Room 44 is undoubtedly part of the ventilating system of
this Room 47A kiva.

When this ceremonial chamber had served its purpose, the north
half of it and the north wall of the one-time dwelling in which it
stood were both removed to provide space for a larger kiva. This
latter and its relation to the whole will be described as I discuss the
sequence of events leading up to construction of Kiva E.

Room 51 was a two-story house occupied, both upstairs and down,
as living quarters. Two east-west beams had spanned the lower room
at a height of 8 feet 3 inches; the ceiling-pole level lay at 9 feet 1 inch
and the second-story floor offset, 10 inches higher. All four walls
had been plastered and the corners tied. A ceiling of customary local
composition is suggested by split-cedar strips in the wreckage.

Several features in these two rooms deserve fuller consideration
than is possible in Appendixes B and C. First, a rectangular fireplace
sunk into the floor in the middle east half of 51A, lined and paved
with sandstone slabs, is unusual in that it has attached to the west side
of it a semicircular, shelflike appendage 5 inches from front to back
and 5 inches above the bottom of the hearth. Its concave side was
lined with mud rather than slab fragments.

Ventilators do not regularly occur in the ground-floor rooms of
Pueblo del Arroyo, but a 12-by-15-inch recess on the west side of
51A may have been such a one although only 40 inches above the floor.
The opening had been blocked from Room 50 which remains unexcavated.
An indubitable former ventilator on the east side of 51A,
7 feet 5 inches from its southeast corner, is noteworthy in that it once
opened into Room 52B as well as 52A. We judge it to have been
15 inches wide by 17½ inches high—rather on the large side; its sill
slab lay 8 feet 2 inches above the floor and its wooden lintels a few
inches above the ceiling poles of 51A, or approximately at floor level
of 52B. Both because the facing masonry hereabout had fallen and
because the one-time opening had been closed when Room 52A was
abandoned and filled, our examination was not conclusive. Nevertheless
I believe this to have been a local example of the oblique ventilator
we disclosed in Pueblo Bonito and which Mindeleff (1891,
p. 207) described as a recurrent feature of old Zuñi architecture.

A. Slabs embedded in the floor of Room 8B-I and rimmed with mud provided a shallow
basin. The leaning door slab, slightly used as a metate, was found elsewhere in the
room.

Plate 4

B. A built-in partition separated Rooms 8B-I and 8B-II (at left). In left foreground,
hole broken through into first-floor room.

(Photographs by O. C. Havens, 1923.)

Plate 5

Left: Room 9A
from the east. In
upper middle, above
the ends of decayed
ceiling beams, a
new partition on
steel-ax-hewn logs
separates Rooms
9B-I and 9B-II. At
far end, the repaired
door to
Room 8A and part
of the original 9A
ceiling under 9BIII.

(Photograph by
O. C. Havens,
1925.)

Right: Room 9A
was filled with
blown sand and debris
of occupation.
Above the fill, decayed
beams and
a ceiling remnant
support the east
wall of Room 9B-I.
View from the
west.

(Photograph by
O. C. Havens,
1923.)

Plate 6

A. Hatchway in southeast corner of Room 20B.

(Photograph by O. C. Havens, 1923.)

B. Beam and cross poles under the floor of Room 46B. The
workman squats in the southeast corner hatchway.

(Photograph by O. C. Havens, 1925.)

A. The west wall of Room 16B had fallen, crashing through the floor and
into the room below.

Plate 7

B. Part of a willow screen under the broken ceiling of Room 16A.

(Photographs by O. C. Havens, 1923.)

Plate 8

A. The broken ceiling of Room 16A and the squared timber that
presumably had braced it.

(Photograph by O. C. Havens, 1923.)

B. Partially blocked east door in Room 43A with slab step and,
at either side, props for ceiling beams.

(Photograph by O. C. Havens, 1925.)

Plate 9

A. Stairway from Room 44A to roof of Kiva F. At right, post
bracing a ceiling beam.

(Photographs by O. C. Havens, 1925.)

B. Fireplaces in Room 47B and, at right, paired logs that had
supported a former second-story partition.

Plate 10

Upper: Supporting posts built into the south wall of Room 52.

(Photograph by O. C. Havens, 1925.)

Lower: Ceiling poles roofing the northeast corner of the Kiva C square. At midleft, refilled
Room 63 and south face of the north wing.

(Photograph by Neil M. Judd, 1926.)

A. Dismantled mealing bins in Room 41.

Plate 11

B. Blocked east doors and reconstructed mealing bins in Room 55.

(Photographs by O. C. Havens, 1925.)

Doorways connected 51A with rooms adjoining on the north and
west. In both cases the lintel poles had decayed and brought about
collapse of the overlying masonry—damage we repaired as excavations
progressed. A former east door, 28 inches wide by 4½ feet high,
6 feet 8 inches from the southeast corner of the room and 16½ inches
above the floor, had been blocked and converted into a recess at a
height of 44 inches when Room 52A was filled.

A substitute for this latter doorway was provided in the form of
a narrow, vertical passage that gave access to the adjoining, secondstory
room, 52B (pl. 13, B). Since the sill of this substitute is 30
inches above the floor, a step was necessary—in this instance a
10-inch-high masonry platform built against the plastered east wall.
Measured on its sill, the passageway is 26 inches wide and 23 inches
deep (6 inches less than the thickness of the wall); part of its 11 lintels
were in line with the ceiling poles of 51A, 9 feet 1 inch above
the floor, but those on the opposite side are 6 inches higher and at floor
level of Room 52B. This 6-inch difference in elevation clearly was
intended to facilitate use of the passageway. As a further aid a
recessed landing, 23 inches wide by 21 inches deep, was provided
4 feet 8 inches above the doorsill or approximately 27 inches below
the floor of 52B. One would expect to find protruding stones or
built-in wooden steps between sill and landing but we observed none.

The coarser stonework within this narrow passageway and on
either side of it at landing level is in marked contrast to that of the
original wall and reflects haste or carelessness when alternations were
made, presumably in conjunction with the filling of Room 52A. In
time this inferior masonry bulged and buckled; the upper south wall,
lacking renewed corner ties, settled outwardly and thus explains why
the near jamb of the door is 18½ inches from the corner at the top but
only 10 inches at the bottom.

That portion of the second-story east wall which had not previously
fallen collapsed during our excavation of Room 51A. Before this,
however, we had noted several constructional features that must be
recorded. On the east side, 23 inches above the second-story floor
offset and flush with the south wall, was part of a plastered recess or
cupboard, 16 inches wide, 10 inches deep, and 18 inches high. Its
north side had fallen, but lintel-pole sockets survived in the opposite
jamb. Twenty-one inches north of this recess a rather crudely
finished corner was to be seen, still 57 inches high. We guessed it to
be the south jamb of a second-story door provided at the time this
wall was rebuilt.

In the northwest corner of 51A an 18-inch-high section of masonry
remained above the second-story floor offset. Embedded in this section
were the ends of two parallel logs that, apparently, had once spanned
Room 53A to support a partition between 50B and 53B. The two
ends, unevenly decayed instead of being square-cut and abraded, continued
half a foot beyond the corner and thus into the north wall of
51B. Here, resting directly upon the beam ends, was a horseshoeshaped
piece of stonework, 10 inches high and 14 inches across, open
end to the east. Part of a sandstone metate, trough down, covered the
south log end and the whole assemblage had been concealed in the
masonry where the four second-story rooms met. I surmise that this
unique feature was a repository for prayer plumes or other offerings
placed there at the time of construction.

That the occupants of Room 51A eventually wearied of their remodeled
home and moved elsewhere and that the neighbors promptly
appropriated it as a convenient dumping place is evident from the
quantities of household rubbish we removed. Among the debris were
numbers of broken or discarded artifacts including three large trough
metates one of which measured 25 by 12 by 8 inches and had two
mano grooves. In addition there were parts of two large, thin metates
of the type we ascribe to the Pueblo II culture; 30 manos, 5 of them
but little used; fragments of four flat, polished lap stones; a large
stone ax, its blade broken; 50 outworn hammerstones; two discoidal
sandstone jar covers, and other objects. We left most of these in
the room, but a number of items retained for the national collections
will be described in the pages to follow. We should note, also, the
skull of a mountain sheep (Ovis canadensis), broken for extraction
of the brain, and the butt end of a war club made from the antler of
an elk (Cervus canadensis). Only four restorable earthenware vessels
were recovered from the debris, but one of these (pl. 31, c) is the
largest found during our Chaco Canyon explorations.

Room 52B is a second-story house whose east wall has fallen to
within 20 inches of its associated floor and whose south wall, of rather
crude construction and heavily plastered, is braced by 4 recessed posts.
These posts average 5 inches in diameter and stand 11½ inches, 47
inches, 7 feet 5 inches, and 11 feet, respectively, from the southeast
corner (pl. 10, upper). The first one is based upon a sandstone slab 20
inches below the floor. We did not disturb the other three. Clearly
this wall is of secondary construction and for the second story only
since, with a 4-inch-wide offset 9 inches below the floor, it rests upon
a roughly built, 5-foot-high foundation that abuts both the east and
west sides of the first-story room.

In the middle of the east wall, 8 inches above the floor, is the sill
of a door that had been closed with masonry. Seven inches in front
of that door a block of red cedar, 7 by 12 by 29 inches, stands half
embedded in the floor. A test pit to discover the significance of that
block revealed only that it stands upon the lower of two east-wall
offsets, 8 inches wide and 14 inches below the floor. It is my guess
that the cedar block was placed there to serve as a 15-inch-high step
to the east door after a new sill had been installed more than a foot
above the old one.

A lesser offset, 5 inches wide, lies at floor level and ends, in line
with the lower, 6½ feet from the southeast corner. At this point the
east-wall foundation abuts a 16-inch-wide wall, its top at the level of
the lower offset, that continues across the room in a southwesterly
direction. Now only a foot high, this wall rests upon a loose fill of
sandstone spalls and adobe, but 25 inches lower and a foot to the
south is the concave face of a partially razed kiva, some of its plaster
still adhering. We followed this latter to a depth of 7 feet 9 inches
and since it is of Type III masonry and apparently without a bench,
I judge it to be part of the unfinished ceremonial chamber we found
under the northeast quarter of Kiva E—this despite the fact that the
two arcs, as exposed in our narrow trenches, do not coincide. I suspect
a slight error in our bearings on that under Room 52B.

Room 55 likewise exhibits a number of alterations that reflect
external pressures. Successive doors connecting with the residence
next on the east had been sealed before, or coincident with, appropriation
of that house as a site for Kiva H. Nevertheless, Room 55 apparently
continued in use, as a mealing room if nothing more. Its
walls had been plastered twice and on the second coat some artistically
minded member of the family had drawn with white kaolin three outlines
of the left hand and two crude human figures. During our
examination of Pueblo del Arroyo we observed only two other instances
of casual wall decoration, in Rooms 41 and 44. There was no
hearth in 55.

A length of 26 feet is greater than usual, and three beams rather
than the customary two were installed to support the ceiling. Beam
seatings are 6½ feet above the floor, but an earlier, original floor lies
5 inches lower. The second-story floor offset is 8 feet 7 inches above
this earlier floor.

The east wall of 55A had known a succession of doorways prior to
construction of Kiva H. What I judge to be the oldest is 4 feet 7
inches from the southeast corner, its sill 8 inches above the earlier
floor. As first constructed, this door measured 4½ feet high and 28

inches wide, but subsequently a new sill was introduced 13 inches
above the old one and the width was reduced to 19 inches when
secondary jambs were added to support a doorslab positioned from
the former house next on the east. Surplus mud from these alterations
was smeared over the surrounding masonry and used to round
off the new jambs. Still later the reduced opening was itself blocked
and the blocking half concealed beneath a second coating of excess
mud.

Another blocked door is seen in the middle of the east wall, its sill
at a height of 39 inches. Irregular stonework below and to the left
suggests ancient repairs. Between this sealed doorway and that first
described is a puzzling break in the masonry, 44 inches high and plumb
as a well-finished jamb.

A third former door in the same wall, 5 feet from the northeast
corner, gives access to storage space improvised between Kiva H and
the old house walls enclosing it. Presumably the door is older than
the kiva but, in any case, secondary jambs had been provided for retention
of a doorslab from Room 55.

The successor to these three doorways is at the south end of the
east wall, 16 inches from the corner. It is 22 inches wide by 38 inches
high and its sill lies 35 inches from the floor. Such a height naturally
called for an assist and, however the need may have been met at first,
a large block of sandstone, slightly used as a metate, was later leaned
against the wall to provide a foot-high, makeshift step. Above, between
the jambs, are two built-in masonry steps, each 5 inches high
but with treads of 4 and 6 inches, respectively. They were the means
of communication between Room 55A and the rooftop of Kiva H at
the second-story level.

Beginning 4 feet from the south end of the room and paralleling
the east wall at a distance of 19 inches is a series of five dismantled
mealing bins. All the metates and most of the bin slabs had been
removed but their former positions gave the series a total length of
12 feet 8 inches. Young women at their daily grinding knelt in the
19-inch space between the bins and the east wall, for the metate
seatings were on that side, their east ends at floor level and their
opposite ends a couple of inches lower. These seatings averaged 23
inches in length and their lower edges overhung by 2 or 3 inches the
slab-paved floors of their respective meal basins. A hewn pine board,
5 by 25 inches and half an inch thick, had been incorporated in the
floor of the northernmost basin. At the opposite end of the series a
broad, thin metate of P. I-II type had been included in the slab
lining of the south basin.

This prehistoric counterpart of mealing bins in their own homes
so intrigued our Zuñi workmen they undertook a voluntary restoration.
The results (pl. 11, B) give the general idea but are inaccurate
as to details, and I neglected to make corrections before the season
closed. Actually, the five metates are bedded too high. Each should
rest with its upper end at floor level and the lower, a couple of inches
below; each should be in an individual compartment walled with slabs
on edge, as is the first of the series. Recumbent slab fragments or
compacted mud should lie both at the raised end and at the sides
between mill and bin walls, and the lower, open end of each mill
should be 2 or 3 inches above its own section of a partitioned, slaband
clay-lined trench averaging 13 inches wide. Cornmeal ground on
the several metates fell into this segmented trench and was transferred
thence to waiting bowls or baskets.

Another error I failed to correct concerns the milling stones themselves.
The last occupants of Room 55 had taken their metates with
them, leaving only the bin outlines and imprints of the five mills. In
their enthusiasm our Zuñi gathered the metates nearest at hand but
did not differentiate between the troughed P. III variety that really
belonged here at Pueblo del Arroyo and those of earlier vintage.
Numbers 2 and 3 are of the thin, broad-margined type we found in
Chaco Canyon P. I pit houses and in the older section of Pueblo
Bonito (Judd, 1954, p. 137), while number 1 has an almost end-to-end
mano trough and narrower walls than is typical of local metates.
Furthermore, the mills should have been arranged according to their
texture, the coarsest at one end of the series and the finest at the
other.

In addition to the metate seatings there was another noteworthy
feature in Room 55—a rectangular, clay-lined depression that abutted
the south wall 42 inches from the southwest corner (fig. 12). It
measured 38 inches long by 18 inches wide and was constructed upon
the earlier floor, at a depth of 5 inches. A single sandstone slab was
incorporated in the clay lining of the east side. Two inches beyond
the north end of this curious feature is a narrower extension, 4 feet
5 inches long, 14½ inches wide at the south end and 12 inches at the
opposite end. This extension, an inch shallower than the first section,
is paved throughout with sandstone slabs but, for some reason, its
south half is lined partly with clay and partly with slab fragments
while the other half is lined with clay only.

Exploratory tests at the north end of Room 55 revealed not only
the earlier floor already mentioned, but also, sunk into it, an 11-inchwide
trench 3½ inches deep. Almost but not quite paralleling the north

Fig. 12.—Floor plan of Room 55.

wall, this trench is floored with a single pine plank 12 feet long,
10½ inches wide, and averaging 1 inch in thickness. Adobe mud lining
the trench rounds off onto the plank. A contemporary fireplace somewhere
on the earlier floor is indicated by smoke-stained plaster on its
associated walls. Our test also revealed the fact that finished masonry
of the east wall continues 17 inches below the earlier floor and, further,
that at a depth of 35 inches a 1-inch layer of shale lies within the
foundation arc of a razed kiva—the only recorded instance at Pueblo
del Arroyo of shale under a kiva floor. We discovered no connection
between this razed kiva and the plank-paved trench above.

Rooms 56-58 appear to be no more than alcoves built to utilize
otherwise wasted space following construction of Kiva H. Their sides
are of relatively crude stonework abutting the plastered south wall.
Despite their shallowness (Room 56 is a mere 32 inches, floor to
single E.-W. beam), Rooms 56 and 57 were once connected by a
door 30 inches high and 18 inches wide, subsequently blocked to leave
a recess in Room 57. The sill is a wooden plank. From Room 56 a
stepped doorway led down into ground-floor 55A.

Room 58, larger and more angular than the other two, probably was
inhabited. Its south and east walls, part of the former dwelling in
which Kiva H was built, are tied at the corner and plastered. Smoke
stains on the plaster evidence an open hearth. A test hole against
the east wall disclosed a well-laid floor at a depth of 58 inches and,
12 inches above it, the sill of a blocked T-shaped door. The lower
portion of this door is 33 inches high; 29 inches of the upper part
remains. Together, they formed a T-shaped doorway more than 6
feet in height connecting two ground-floor rooms that subsequently
were preempted by the priesthood as sites for Kivas H and I.

Room 59 is an inexplicable structure at the northwest corner of
the square enclosing Kiva G. Two of its three walls, concave and
joined, abut the third which is the north side of the enclosing quadrangle
(fig. 18). All three walls were plastered and the plaster
rounded off with the adobe floor. The latter had been spread upon a
fill of loose sand. We saw no trace of beam or ceiling-pole seatings.

The westernmost of the two concave walls rises to a height of 68
inches; its companion stands 54 inches high. Where the former abuts
the north wall it includes a straight section 38½ inches long and 41½
inches high. In the face of this section, 24 inches above the floor, a
single stone projects to form a step, 3 by 8 inches. The top of the
section provides a second step, 40 inches long with a 6½-inch tread.
Following a 2-inch rise the section retreats 14 inches to create what
may be a third step, although it looks more like an open shelf, 30 inches

wide in front, 34 inches wide at the rear, with sides 29 inches high.
If a projecting stone step was necessary in the face of the 2-foot-high
section below, a similar step would appear equally desirable in the
back wall of this upper, shelflike portion. There were no artifacts and
no potsherds in the sand filling this queer triangular structure.

Rooms 60-61 are closetlike and were built upon the fill at the west
side of Kiva G. Their north and east walls consist of salvaged blocks
of sandstone and were erected with little thought of stability. An
oval fireplace lies against the west side of Room 60. In the southwest
corner of 61 a single post hole, 4 inches in diameter and 10 inches
deep, marks the position of a former ceiling prop.

The unroofed space north of these closets provided an outdoor
workroom whose pavement, at the same level as the floors in 60 and
61, lies a mere 15 inches above the floor of Kiva E on the west but
11 feet 5 inches above that of Kiva G. An open hearth, 12 inches
square and 7 inches deep, lined and paved with fragments of sandstone
slabs, lies 11 feet 8 inches north of Room 60 and 28 inches from
the west wall.

Room 65 is one of two small structures occupying leftover space
following construction of Kiva J. Its north, east, and south walls
were constructed almost wholly of large, dressed blocks of friable
sandstone and sparsely chinked, while the west wall was built of laminate
sandstone and is perhaps superior to the average of Pueblo del
Arroyo masonry. This west wall was clearly erected to complete the
enclosure of Kiva F, although its builders inexplicably made a corner
1 foot from the previously erected north wall instead of abutting the
latter. Adobe plaster neatly rounded that corner before the occupants
of Room 65 filled the 12-inch space with their own stonework. Thereafter
all four walls were plastered; the highest, on the south, still
stands 6 feet 9 inches with no visible beam or ceiling-pole seatings.
Surprisingly, the wall plaster remains unsmoked.

Despite its plastered walls, customarily the mark of a dwelling,
Room 65 had been utilized for storage purposes. On the floor, arranged
along the east half of the north wall, were five Corrugated-coil
cooking pots; seven more stood against the west half of the south wall.
Some had been placed on their rims, bottom up. All were crushed as
they stood when the ceiling collapsed upon them (pl. 14, A). In the
southeast corner, its orifice turned to the wall, was a canteen, the only
nonculinary vessel in the room (pl. 28, a).

After the accident there was nothing to salvage here but the ceiling
timbers. These appear to have been pulled out one by one, leaving
only chunks of roofing abode with impressions of reeds, bark, and

split cedar. Upon this debris neighboring housewives soon began to
dump floor sweepings and kitchen waste. From this rubbish we recovered
six more restorable cooking pots (pl. 33), a number of bone
and stone artifacts, beads, paints, pumpkin seeds, and shells of pinyon
nuts, charred corncobs, flint and obsidian chips left by a maker of
arrowheads, and like discards. Among larger objects we noted but
left in the room are: a tabular metate measuring 17 by 14 by 1½ inches
with a mano trough 14 inches long and 7½ inches wide, fragments of
5 other metates, 18 manos and 3 fragments, 10 outworn hammerstones,
and 12 fragments of sandstone tablets and door slabs.

A broken area on the east side of the floor, caused by the burrowing
of a rodent, invited deeper exploration. An earlier, and obviously the
original, floor was exposed at a depth of 4 inches and the finished
masonry of the north wall continues 3 inches farther, there to rest
upon a 7-inch-high foundation. The east-wall foundation rises a
couple of inches higher. At a depth of 20 inches, beneath a purposeful
fill of constructional debris, we came upon an ash-blackened pavement
that antedates Room 65.

Opposite, a second test hole against the middle west wall revealed
finished masonry extending 3 inches below the original floor and resting
upon a 10-inch-high foundation. At the point where it was first
exposed that foundation stood upon burned sand and pieces of sandstone
mixed with charcoal and partly filling a large firepit. The
foundation apparently bisected the pit, for that portion of it we laid
bare measured 37 inches long and 28 inches deep. Its north side,
11 inches thick, passes under the west wall 9 inches below the original
floor and 4 feet from the northwest corner. The inside of the pit had
been burned to a depth of 1½ inches without fusing its half-inch clay
lining.

Despite the similarity of their stonework, the north wall appears
considerably older than the east and south walls, and that on the west
side must have followed the north by a short interval since its foundation
is practically at the same level. The south and east walls abut
the west and north, respectively. The better to protect its masonry
from further deterioration, Room 65 was partially refilled upon conclusion
of our study.

For reasons previously stated, our studies inside the main walls of
Pueblo del Arroyo were concluded with the excavation of Room 65.
We cleared 44 of the 58 ground-floor rooms numbered on figure 2 and,
in addition, 7 kivas and various external accretions to be considered

in chapters III and IV. All pertinent architectural data recorded during
our investigations are presented in Appendixes B and C. Nevertheless,
before proceeding, a few generalizations seem desirable.

Evidences of constructional change, of alteration, of removal and
replacement, were encountered most frequently in the middle section,
between the north and south wings. The more deeply buried floors and
wall remnants lie here. Major changes, as when one or more dwellings
were relinquished to permit erection of a ceremonial chamber, would
seem to have been forced without regard to domestic considerations.
The wishes of kiva groups obviously took precedence. Ceremonialism
was a powerful influence here.

Despite such evidence, Pueblo del Arroyo was a planned community.
Its masonry is predominantly of the variety we called Type III at
Pueblo Bonito (Judd, 1954, p. 19) but with generally larger, less
carefully selected tablets of laminate sandstone and with less banding.
There is good substantial stonework in Pueblo del Arroyo and there
is poor stonework. Soft, friable sandstone that absorbs water and
readily disintegrates was used almost exclusively in construction of
Kivas I and J, for example, and these two have survived the passing
centuries less successfully than their neighbors.

Some building practices were rather formalized; others were haphazard.
Foundations, even for a four-story structure, might be a
single course of sandstone and mud or several courses. When walls
were razed the useful stones were saved, but spalls and chunks of
adobe mortar were left where they fell. Replacement walls were built
on top of such accumulations without thought of possible settling.
Foundations are usually 2 to 6 inches wider than the walls they support,
and the differences may be all on one side or more or less
equally divided. So, too, with storied walls: each is usually a few
inches thinner than that next below.

It is my conviction that the builders of Pueblo del Arroyo completed
each storied room, both walls and ceiling, as their work advanced.
When the desired wall height had been attained, the main
beams were positioned, their ends flush with the opposite side of the
wall, square cut and abraded. Ceiling poles were laid upon the beams
and at right angles to them, often with several inches of the butt
tenoned in the wall masonry. Cedar splints, locally preferred in ceilings,
were next laid upon and at right angles to the poles and were
bound to them by yucca-leaf thongs; splint ends were commonly
anchored by a course of thicker, heavier stones. Since individual
splints, less frequently strips of bark, were sometimes incased an inch
or more in the masonry, I reason local masons worked more or less

blindly while building walls at ceiling level. Hence the possible irregularity,
or even the absence, of what we called the "floor offset" or
ledge.

An upper wall was customarily set off, or set back, an inch or
more from the face of that below. This setback invariably occurs in
line with the adobe surface that completes a floor assemblage and is
an almost constant feature of wall interiors but rarely occurs on the
outside. An offset may be poorly indicated or wider than normal; one
at the end of a given room may be, but rarely is, several inches
higher or lower than those on either side. Whatever the motive behind
it, the offset increased the floor dimensions of each upper room over
that next below and reduced the thickness of intervening walls.

As is clearly illustrated in Ruppert's sketch of the wall separating
Rooms 30 and 34 (fig. 6), floors in adjoining houses are not necessarily
at the same level, and floor offsets are variable. Posts installed
to brace first-story ceilings were invariably set upon base slabs in dug
holes. The holes were sometimes, but not always, lined with slab fragments
on edge; sometimes, but not always, shale was packed between
post and slab lining.

In large measure, ground-floor rooms were utilized for storage;
residential quarters were on the upper floors. Customarily, storeroom
walls were not plastered but those in dwellings were. Also, the principal
living room was usually provided with a fireplace. Eight of
the 44 ground-floor rooms we excavated in Pueblo del Arroyo were
equipped with fireplaces, and one, number 40, had two hearths, one
semicircular and the second rectangular.

Three second-story rooms, 8B-1, 9B-III, and 47B, likewise had
fireplaces, and there can be no doubt that open hearths on upper levels,
both indoors and out upon the flat rooftops, were more frequent than
our data indicate. If there were a local preference as to hearth shape,
it was for the rectangular since we noted 6 of this kind, while 3 were
square and only 2 each of oval and semicircular form. Of the 13 fireplaces,
5 were lined with adobe mud, 3 were slab lined, 3 masonry
lined, and 2 not recorded. The rectangular fireplaces in 8B-I and 47B
each contained 2 stone firedogs.

Seatings for milling stones were present in Rooms 41 and 55. The
first apparently provided for three mills; the second, for five. In
both instances the series had been dismantled and metates removed,
together with most of the sandstone slabs that had floored and walled
the individual bins and meal trench. Because we recovered a number
of milling stones from among the wreckage of collapsed walls and
ceilings in first-story rooms, there is every reason to believe that those

mills had been installed, singly or in series, in the dwelling rooms
above.

A glance at figure 2 will show a positive grouping of certain rooms
as, for example, 12-13-20-21, 29-30-31-32, and 42-43-44-45. To some
anthropologists this might suggest clan occupancy of the suite, and I
am not prepared to argue the point. The first two groups lack external
doorways—a fact that means, of course, at least one room in each
unit was equipped with a hatchway connecting with the second story.
Just such a hatchway was present in the southeast corners of Rooms
20B (pl. 6, A), 44B, and 46B.

At Pueblo del Arroyo lateral doorways joining rooms on the same
level are usually rectangular or nearly so. Of 73 measured (counting
as 2 those examined from opposite sides of the same wall), size varies
in width and height from 15 by 19 inches (Room 10) to 44 by 57
inches (Room 40). Average dimensions of 52 ground-floor doors are
25 by 41 inches; only 7 are over 50 inches high. Of 20 second-story
doors, the average is 24½ by 39 inches and our single intact thirdstory
doorway (Room 23C) measures 26 by 41 inches.

Altogether, door size appears largely a matter of chance. In no
instance where two or more doors occur in a given room are measurements
equal. The nearest agreement is seen in Room 41 where two
doors, both 26 inches wide, are 48 and 49 inches high, respectively.
In Room 27 two 26-inch-wide doorways are 42 and 44 inches high,
and in Room 34 two doors also 26 inches wide are 56 and 58 inches
high. Occasionally an inverted tabular metate appears as a sill, but
the east door of Room 56 was equipped with one of wood.

Sill heights varied also and might be changed when the opening was
altered or when a new floor was laid. Steps, if needed, might be of
masonry, a slab on end, or a mud platform. Recessed stairways providing
access from a first-story room to the second-story level adjoining
are recorded for Rooms 41, 44, and 55. Storeroom doors were
usually equipped with secondary jambs for retention of a doorslab
placed from the outside, sometimes with a smaller lintel pole set into
the jambs 4 or 5 inches below the main lintels. Mindeleff (1891,
p. 182) says such a pole in Hopi houses of the middle 1880's was for
support of a blanket in cold weather. Where secondary lintels occurred
in Pueblo del Arroyo they were almost always overlain by
masonry.

Tau-shaped doors, known throughout the Plateau Province from
Pueblo II times to the present, have always intrigued Southwestern
archeologists. Mindeleff (ibid., p. 191) remarks that those he saw in
Hopi towns lacked the symmetry of T-shaped doors in ancient cliff

dwellings, one jamb being "stepped at a considerably greater height
than the other." Symmetry was the rule at Pueblo del Arroyo although
size and sill height varied.

We have record of eight such doors, three in first-story walls and
five in those of the second story. Four of the total (in 8B-I, 46B,
58A, and 62A) pierce the east wall; three (in 9B-III, 10B, and 44B),
the south, and one only (24A), the west wall. This latter, subsequently
closed with rude stonework, had once opened to the outside
at ground level—the only known passageway in the otherwise blank
west wall of Pueblo del Arroyo. In fact, our eight T doors all appear
to have been blocked, in whole or in part. That connecting 46B
and 47B had been sealed before the secondary north walls were introduced
on either side. From 8B-I a T-shaped door in the east wall
gave access to 9B-III which has a T door in its south wall. A large,
irregular opening broken through the south wall of 10B framed what
we believe to have been a T-shaped door and we restored it as such.

Ventilators occur with less regularity at Pueblo del Arroyo than
at Pueblo Bonito. Our notes record 19, 10 in ground-floor rooms and
9 in those of the second story. Again counting as individuals those
approached from opposite sides, four appear in the north wall, two
in the east, seven in the south, and six in the west. In only three
instances (Rooms 25A, 34B, and 35B) are ventilators paired in the
same wall. Single examples occur on the east side of Rooms 51 and
65 and in the otherwise blank north walls of 30B, 31B, 32A, and 32B.
Ventilators in outside walls were recorded in six rooms only (8,
9B-I, 10B, 24, 25, and 29) but we know from Mindeleff's photographs
that others were present in walls since fallen. As with doorways,
ventilators were sometimes blocked and occasionally the blocking was
placed from one side of the wall to leave a recess on the opposite side.

Mindeleff's few photographs enable us not only to visualize the ruin
as he saw it in 1887 but also to estimate the changes brought about
between then and 1920, the year of the National Geographic Society's
reconnaissance of Chaco Canyon. More upper walls were standing in
1887, and second-story ceiling beams could be seen here and there.
Five of them protruded from the outermost north wall, immediately
above four second-story ventilators, at least two of which had been
blocked (pl. 16, B). If those timbers were designed to carry an
external balcony, as is possible, it is to be noted that no door appears
in the third-story wall then surviving. Ten years earlier Jackson
(1878, p. 443) wrote that beams projected through the wall for about
5 feet "along its whole northern face the same as in the Pueblo Hungo
Pavie." Unfortunately passersby in temporary need of firewood have

since pulled out the five beams Mindeleff photographed and, at the
same time, toppled the overlying masonry (pl. 16, A).

Despite his implication to the contrary, Jackson probably saw no
more than the five projecting second-story beams Mindeleff photographed
10 years later. In Mindeleff's 1887 view (pl. 16, B) firststory
beam holes may be seen at ground level and, below them, a
sloping accumulation of scattered building stones fallen from the
upper third story. Standing masonry at the far right is part of that
same north wall, but there were no protruding second-story timbers
visible here at the time and Mindeleff's photograph lacks evidence
that the intervening portion had fallen during the preceding decade.
Holsinger, who made no mention of the five beams in his 1901 report
to the General Land Office, apparently assumed a missing outer tier
of rooms when he amended the north-wall statement printed on Jackson's
plan (fig. 45).

With the ruin as a whole in mind, Jackson wrote (1878, p. 443):
"Many of the vigas are still in place and perfectly smooth and straight
undressed logs of pine, averaging 10 inches in thickness; none of the
smaller beams or other wood-work now remains."

A ceiling beam spanning what may be Room 9B-I or 10B is to be
seen at the left in Mindeleff's general view from the southeast (pl. 1),
and the stub of another casts a shadow beneath the open third-story
door at the center of the picture. In this same view, near the right
margin, is a shadowed pear-shaped mass that looks like two-story
masonry. It is ghost masonry, however, because, while it seems to be
real enough on the original negative, it does not appear on any other
Mindeleff photograph. If not a phantom, the wall fragment would
stand near the easternmost, unexcavated kiva fronting the north wing,
and here all was rubble in 1887 as in 1920.

Since the days of Jackson and Mindeleff not only beams but lintel
poles over doorways and ventilators have been wrenched free and
used for fuel. Broken door jambs were weakened and their collapse
was accelerated as winter snows and summer rains softened the mud
that bound the stones together. Jagged holes marking former doorways
in the second-story south walls of Rooms 9 and 10 had more
than doubled in size between 1887 (pl. 15, a) and the beginning of
our investigations in 1923 (pl. 2, upper).

Loss of these old beams and lintels has forever erased tree-ring
dates that would have added greatly to the history of Pueblo del
Arroyo. The extent of that loss may be gaged from what has since
been learned. Of many hundred selected logs, large and small, that
went into the building of the pueblo, we collected only 45 samples.

These are sections cut from fragments unearthed during the course
of our investigations and borings taken from beams still in place.
Fourteen contributed nothing because their annual growth rings were
so uniform in width they could not be read. The others gave the following
dates:

                                                                 

Field No.	Location	Cutting date[1]
59	Room 12, beam	1102
60	Room 12, beam	1103
61	Room 12, beam	1103
62	Room 8B-I, beam	1105
63	Room 9B, beam under partition	1101
66	Room 16, east beam	1098
84	Room 9B, beam under partition	1102
85	Room 9B, beam under partition	1102
87	Room 9B, beam under partition	1100
88	Room 9B, beam under partition	1092
89	Room 9B, beam under partition	1100
100	Room 20, beam	1096
101	Room 20, beam	1095
102	Kiva C, pilaster 8	1067 [2] x
111	Beam, unnumbered room, NE. corner S. wing	1092 [3] 3
124	Room 39, recent dump (probably from Room 38)	1064 [2] x
125	Room 34, beam	1075 [2] x
126	Room 43, beam	1074
129	Room 44, north beam	1066
131	Room 46, south beam	1070
132	Room 46, prop for 131	1052[4]
133	Room 47, south beam	1066
134	Room 47B, one of paired beams under N.-S. partition	1077
137	Log in 2d-story wall, W. side of Court	1074
139	Log, W. side of Court, NE. of Kiva I	1091
146	Beam from unexcavated room next NE. corner Kiva C	1102
147	Beam, NE. corner Kiva C square	1103
148	Beam under 2d-story partition, 5th unnumbered room from W. end of N. tier	1117
149	Beam, 2d story, same room	1101 [3] 2
152	Beam, 1st story, same room	1101
153	Beam, unnumbered room N. of Room 53	1065 [2] x
154	Beam 7 ft. N. of No. 153	1075

Of these 32 dated timbers 17 were built into the south wing, 5 into
the north wing, and 10 into the section between. Of those in the first
group 6 were felled between 1067 and 1098; 11 between 1100 and
1105. The 10 timbers from the middle section were all cut between
1052 and 1091, and of those from the north wing 1 was cut in 1065,
1 in 1075, and 3 between 1101 and 1117, the latest recorded tree-ring
date for Pueblo del Arroyo (Smiley, 1951, p. 19). Together they
suggest the period during which the village was under construction.

Tree-ring dates are not always to be taken at face value, however,
for timbers in good condition were used again and again. We observed
a number of instances where both kivas and dwellings at the
time of abandonment had been stripped of their ceiling logs for reuse
elsewhere. A most illuminating example is in Room 46A where the
original south beam, with a cutting date of 1070, had later been
propped by part of a log felled in 1052. The beams and door lintels
destroyed between 1887 and 1923 would have added substantially to
our tree-ring record from Pueblo del Arroyo and might have enabled
us to fix the time of occupancy with greater assurance.

A. Burial, southwest corner of Room 40A.

Plate 12

B. Skeleton of a dog, on the floor of Kiva F.

(Photographs by O. C. Havens, 1925.)

Plate 13

A. The body of a youth, buried on the floor of Room 13A, had
been disturbed in prehistoric times.

(Photograph by O. C. Havens, 1923.)

B. Doorway in east wall of Room 51, connecting with secondstory
Room 52B.

(Photograph by O. C. Havens, 1925.)

A. Cooking pots in Room 65, crushed by collapse of the roof.

Plate 14

B. Sandal figure chalked on the north wall plaster of Room 44. At left, a small repository.

(Photographs by O. C. Havens, 1925.)

Plate 15

a, The outer south wall, from the southeast; b, south end of Room 45B(?);
c, east tier rooms, south wing.

(Photographs by Victor Mindeleff, 1887. Courtesy of Bureau of
American Ethnology.)

A. The north wing of Pueblo del Arroyo, from the northeast.

(Photograph by Charles Martin, 1920.)

Plate 16

B. Northeast corner of north wing, with second-story ceiling beams projecting from outer north wall.

(Photograph by Victor Mindeleff, 1887. Courtesy of Bureau of American Ethnology.)

Plate 17

Upper: Clearing the court at Pueblo del Arroyo. At center, second- and third-story south
walls of Rooms 19-21.

(Photograph by O. C. Havens, 1925.)

Lower: Beginning excavation at third-story floor level, northeast corner of Kiva C square.
Salvaged rocks for wall repair piled at lower left.

(Photograph by O. C. Havens, 1924.)

Plate 18

Left: Clearing
Kiva C. Above the
wheelbarrow, the
open south door of
Room 28B.

(Photograph by
O. C. Havens,
1924.)

Right: South
wall and bench recess,
Kiva C, with
wall restorations
above the workman.
At upper left, restored
south door
of Room 19C.

(Photograph by
Neil M. Judd,
1926.)

A. Fire screen, south bench recess, and rebuilt ventilator, Kiva G.

Plate 19

B. Southeast quarter of Kiva J showing ventilator, enlarged south bench recess, and
smoke-stained plaster above.

(Photographs by O. C. Havens, 1925.)