University of Virginia Library

II. Part II

THE SITE AND THE EXCAVATIONS

By Frank C. Hibben

The ruin mound is but one of many which lie in a cove immediately
to the east of the Research Station of the University of New Mexico
and the School of American Research. It was chosen not only because
of its accessibility but also because superficial shard evidence seemed to
indicate a typical structure belonging to that class which has been
designated as the Chaco small house. This site, because of the richness
of the shard evidence, seemed to indicate special possibilities, both for
archaeological research, insomuch as but little is known of this type
of structure, and also for the instruction of students. This particular
ruin had been designated as the fiftieth in the series of ruins, large and
small, which had been enumerated in the charting of the Chaco Canyon
during a survey carried out by Dr. Reginald Fisher. In accordance
with the cataloging system of the Museum of Anthropology of the
University of New Mexico, this ruin is designated as Bc50, B being the
initial to indicate the archaeological area of the State of New Mexico,
c designating the specific area of Chaco Canyon, and 50 the ruin itself.
Following the custom of other excavations in the Chaco, this one was
given the additional name of Tseh So, meaning "windows" or "openings,"[1]
a name which was applied to it by some of the Navajo laborers,
possibly because of the large number of apertures which appeared in
the walls after the digging had progressed.

The mound, before excavation, presented no features of especial
interest (Plate Ia). It was a roughly oval pile with the long axis lying
north and south. Its apex, approximately over the center, was some
seven feet above the level of the surrounding wash. This level is only
approximate insomuch as the wash from the cliff talus to the south
has filled in the adjacent end of the mound to raise the level several
feet above the land immediately to the north of the ruin. This mound
is adjoined by another one of similar size and similar shape, some fifty
feet to the east, which has been given the number of Bc51. In between
these two house mounds was a pile of refuse which seemed to offer
exceptional possibilities because of the appearance of human bones at
several places on its surface. During the summer's excavation of 1936,
Bc50 was 80 to 90 per cent excavated, stratigraphic test trenches were
carried into the refuse mound just mentioned, and one room in Bc51

was completely examined. Insomuch as Bc51 seems to offer possibilities
for further research, this report will confine itself to Bc50, or Tseh So,
and the refuse mound which lies between the two ruins.

Excavation Methods

The entire month of August, 1936, was consumed by this excavation.
The students of the 1936 Chaco Field Session were employed as
junior archaeologists, overseers in charge of minor projects and general
assistants under the supervision of the staff of the Field Session.
The laborers used were some ten to sixteen Navajos, the number being
inversely proportionate to the proximity and duration of "squaw
dances," "sings," and horse races.

The students, assisted by the laborers, first traced out all of the
walls possible on the surface of the mound (Plate IIa). A room was
assigned to each two students as a special responsibility. Room numbers
were assigned to kiva enclosures also, as these had been used for
living purposes in every case. The walls of the rooms on the west tier
were distinguishable on the surface before excavation. Many of the
other walls, and those especially of the kivas, were from one to three
feet below the surface. In the southeast corner of each room, a column
one meter square was left as a permanent record of the stratigraphy
within that room. The rest of the room fill was excavated in foot
levels, the artifacts from each level being sacked and examined separately
later in the field museum. All artifacts were numbered with the
Museum of Anthropology numbers immediately upon their acquisition.
Burials, when encountered, were taken out by members of the staff
(Plate IIb). The somatological material first was photographed with
the permanent museum numbers noted, and then was prepared and
sacked in the field. All wood and perishable objects were preserved in
the field with appropriate materials and removed to the field museum.
The kivas presented a special problem but were treated in general as
were the rooms, a "strat" column being left from the original surface
down to the first floor.

After excavation had proceeded in the superstructure down to the
first floor, these "strat" columns were scraped clean, photographed, described
and graphed. A regular photographic crew was constantly
prepared to photograph any new objects or subjects of interest which
came to light. Throughout the excavations and the subsequent examination
of the stratigraphy columns, the shard evidence was carefully
noted and percentages computed.

In addition to the above, a trench was carried completely around
the ruin to serve as a stratigraphy test on the fill surrounding the
exterior. Additional trenches were cut in at right angles to the encircling

trench at the four points of the compass and carried out
twenty-five feet from the walls.

All excavated earth was examined twice for content—from trowel
to shovel, and from shovel to wheel barrow, according to each level as
mentioned before. This dirt was then deposited in the low flat to the
north of the mound, well out of the way and below the drainage level
of the excavation. Drainage from the denudational slope above the
mound was accomplished by a V-shaped cut to carry the run-off water
to both sides of the mound.

The stratigraphy test in the refuse mound was carried on at the
same time as the main excavations. Also, a surveying crew kept in
constant touch with the developments of the excavation as they
progressed.

As the mound, as it first appeared, became complicated in its lower
levels with earlier horizons and different periods, the process of excavation
was renewed on the old lines after the first floor was reached. A
series of substructures occurred on the western edge of the mound,
located by a test trench in that direction. These extended under the
whole superstructure, as was later ascertained. Therefore, after the
primary excavation of the superstructure, and the examination of room
fill and stratigraphy columns, the process was repeated beneath these
rooms. The floor of each top room was dug away, and the contents
of the exposed interior room noted in levels as before. A stratigraphy
column was left as before, the top of which was a portion of the room
floor of the superstructure. The substructure, which may from this
time on safely be called Pueblo I, did not in each case coincide with the
walls of the pueblo which had been built upon it. (Fig. 1.) Tests were
completed below the Pueblo I level, and a large number of Basket
Maker shards and a portion of a pithouse indicated yet another level of
occupancy. Thus, this small but remarkable ruin, Tseh So, presented
features of a gigantic stratigraphy test in itself. No better could have
been procured for the initial education of the students involved. Full
realization of the possibilities which this ruin presented made the careful
examination of all possible stratigraphy tests essential. Complete
notes on the cubic content of every portion of the ruin made possible a
rather detailed description of the fill and architectural features of every
portion of it and a number of interesting deductions derived thereform.

General Features of Superstructure (Pueblo II)

The superstructure or Pueblo II structure gives the essential character
to the mound. This structure was composed of twenty-four rooms
of which four on the east side, Numbers 12, 13, 23 and 24, were not true

Map IV. Plat of Excavations

(As of August 20, 1936)

rooms but kiva enclosures. The twenty rooms remaining were arranged
in a double row north and south with five additional rooms placed
on the east side to form a structure roughly T shaped in plan. The
kivas, four in number, were placed on the east side of the structure.[2]

Masonry. The room walls of the Pueblo II pueblo were of poorly
constructed stone masonry of single thickness, having no rubble or
stone core. The essential peculiarities of this type of masonry are: the
use of rather large building blocks, all of sandstone, roughly squared,
most of them naturally so, a few of them pecked and shaped, and nonbanded.
These comparatively large blocks (average dimensions
14″ x 6″ x 8″) were interespersed with much adobe plaster studded with
many small spalls of sandstone chips, water-washed pebbles, and potsherds.
This type of masonry, although varying slightly from room to
room, is the type throughout the Pueblo II structure (Plates VI, VIIa,
b). It was quite evident that Rooms 8, 18, and 19, were added on to the
rest of the pueblo after the building of the original structure. The
character of the walls and viga supports, and the amount of debris give
evidence of a second story over Rooms 1, 2, 3, 4, 22, and possibly 14.
Of the nature of this, little may be said except that apparently the
masonry was of the same general type as that below. The single wall
thickness does not hold up well and any slight sinking of the ground,
such as that which occurred in several places because of a settling of
the Pueblo I structure underneath, causes bad distorting and buckling
of these thin Pueblo II walls. In several places, in order to get down
to the Pueblo I and Basket Maker levels with safety to the students and
workmen, it was necessary to remove portions of the badly sagging
and unsafe walls.

Openings. These walls were pierced in many places by numerous
"windows" or doorways, some open and some sealed, the sealing being
done with the same types of masonry and the same shard spalling as
accompanied the walls of the rooms. Most of these doorways were provided
with a well-worn sill and lintel, and, in one case, wooden uprights
on the sides. The lintels also were of wooden pieces. These doorways
are rather small according to modern standards, as is usual in
such pueblos. Measurements averaged eighteen by thirty-six inches
for those openings complete enough to measure.

Plaster. The plaster yet adhered to the interior of many of these
walls. It was of sandy adobe smeared on with the hand and varying in
thickness from an eighth of an inch to an inch. The plaster on the south
side of Room 1 was especially well-preserved and showed four distinct
smoke-blackened layers, each exhibiting many finger and hand impressions.
As far as could be ascertained from the fragments of plaster

preserved, there was no decoration or coloring on these interiors. The
plaster of certain of the kivas was quite otherwise, as will be mentioned
in the description of those structures.

Floors. The floors were of hard-packed but unburned adobe
throughout, varied in some instances, such as Room 14, with slabs or
scales of carbonaceous shale, gathered presumably from the outcroppings
in the cliff immediately behind the ruin. The sandy character of
the adobe floors in some cases rendered them difficult of perception and
of distinction from the hard-packed sandy aeolian fill and adobe washed
from the walls, which in each case constituted the layer immediately
upon the floor. Several floor layers were detected in some of the rooms,
as in Rooms 8, 9 and 10, where two successive floor levels were found,
one laid directly upon the other at an interval of about three inches.
The floors in the kivas presented a different problem.

Roofs. The roofs of the Pueblo II structure were of the usual
pueblo type, supported on vigas averaging about eight inches through.
Fragments of some of these were found in place. In Room 4, a large
section of the roof was found intact, fallen, after the breaking of the
central viga running in an east-west direction, upon a quantity of
debris which had collected in the room before the collapse (Plate V).

Cross members of piñon and juniper, averaging three inches in
diameter, had been laid at right angles to the viga, i. e., north and south
in this case, and at intervals of a foot. At right angles to these last
and covering the entire surface, were split slabs or slivers of juniper.
On top of these, and again at right angles, that is, parallel with the
piñon cross members, was a matting of, in this case, horsetail reed
(Equisetum) bound together so as to form a compact whole by twisted
strands of yucca fibre every six inches along their length. On this last
was placed, evidently, a final covering of rubble, leaves, brush and adobe.

It must be mentioned in this connection that from the prints of
the roof covering from other rooms, the construction was essentially
the same but the horsetail reed was varied with sacaton grass bound
together in the same way. Also, the primary covering of juniper slabs
did not seem to be constant, although it also occurred in Room 1.
Prints of the roof preserved in the adobe lumps from Rooms 9 and 16
show that fragments of matting, and occasionally fragments of basketry,
and even of fabric, were used either to patch holes in the roof
or to supplement the reed cover.

Cists and Bins. A limited number of cists and bins were encountered
in several of the rooms in the south end of the pueblo. Room 19
contained three such bins set immediately on the floor and outlined by
sandstone slabs (Plate IIIa). These measured a foot in each diameter
and about six inches deep, with the side toward the center of the room

made at an intentional angle. The cracks between the stones were filled
with an abundance of adobe plaster. No evidence of fire was present.

Room 17, adjoining, contained a cist in its southeastern corner, as
did also Room 16 in its southeastern corner. These cists were in each
case two feet in diameter and two and a half feet deep. They were
supplied with an adobe coating which differentiated them from the
adobe floor surrounding and were lined with adobe plaster to about one
half their depth. Their use is problematical insomuch as they were not
furnished with any grass or matting lining which might indicate storage
cists. The cists in the kivas were of a different nature.

Fireplaces. Fireplaces were found in several of the rooms, being
located only in those which were on the outer edges of the pueblo where
the structure was very obviously of a single story. Rooms 9, 14, and
16 were furnished with firepits, these being of two distinct types. That
which occurred in Room 14 represents perhaps the most usual type.
This is a rectangular depression, ten by twelve inches, let into the floor
to a depth of four inches and outlined with four slabs on edge. Two
broken manos, projecting in from opposing corners of the rectangle,
served evidently as pot rests. The fire pits in Rooms 9 and 16 were
centrally located and made of adobe directly in the adobe floor. These
were circular, some fifteen inches in diameter and eight inches deep.
A low coping of adobe separated them from the adobe floor as in the
case of the adobe cists mentioned above.

Special Structures. Under the head of special structural features
in the Pueblo II unit may be grouped certain innovations such as the
use of polishing and sharpening stones built directly into the wall, frequent
use of trough metates built in the wall, and two examples of
protruding stones beneath doorways used as convenient steps for entrance
and exit. On the north wall of Room 1 was a large sandstone
piece which had evidently been used for sharpening of bone implements
and possibly for those of stone also. The lintel of a doorway between
Rooms 8 and 9 had evidently been used as a sharpening medium for
stone implements with broad-bladed or broad-bitted cutting edges.
Steps or projecting stones were noted on the west side of Room 7 and on
the east side of Room 10 to accommodate the doorway between these
two rooms. Metates were used liberally as building material, and in at
least two instances may have served as wall receptacles.

Superstructure Pueblo II

The Kivas

The kivas of Tseh So were four in number. Two of these were kivas
"with a southern extension," or, as they are popularly called in the
Chaco, "Keyhole Kivas." These four kivas were placed, as was noted

before, on the east side of the pueblo, three in a line along the main tier
of rooms and the fourth at the end of the room extension formed by
Rooms 3, 4, 5, and 6. Two of these kivas, those toward the north end,
were plain pit kivas with straight sides and a rectangular bench; and
the kiva farthest to the south (Kiva 3) exhibited San Juan type pilasters.
The masonry in all four kivas was essentially the same as the
masonry for the rest of the pueblo, although as a whole, the walls were
better formed and better laid. The building blocks of sandstone which
formed the pilasters, benches, and parapets were, as a rule, well-shaped
and carefully laid to form a comparatively smooth wall with but few of
the sandstone spalls noted before. Pottery spalls were liberally used
to stud the plaster between the stone courses. All four of the kivas
were fitted with fire pits, fire screens, and ventilators, in the regulation
manner. Each of the four was oriented towards the south, i. e., the
ventilator shaft was oriented to the south. One of the most interesting
features is the series of turkey skeletons which were found back of the
fire screens in all of the kivas. In each case the deposit consisted of an
entire turkey skeleton spread out in the space immediately in front of
the ventilator shaft. In Kiva 4 there were two skeletons. These are
all of young female turkeys, and in all five skeletons the head is missing.
The meaning of these offerings is purely conjectural.

Kiva 1 (See Plate VIII and Fig. 2a)

Kiva 1 is located to the east of Rooms 7 and 8, and to the south of
Rooms 3 and 5. The enclosure surrounding this kiva is designated as
Room 12. The kiva proper is nine and one-half feet in diameter and
the same distance in depth, with a bench running around its circumference
six feet from the floor level and seventeen inches wide. Kiva 1
was provided with a southern extension in the form of a platform two
feet by four feet ten inches in dimensions, and at a depth of four
feet from the surface. Underneath this platform runs the ventilator
shaft, fourteen by eighteen inches in dimensions from its entrance or
opening in the kiva, below the edge of the platform, to its exit at the
surface through a chimney of masonry a foot square immediately to
the south of the platform extension. This ventilator shaft was covered
over in its horizontal extension under the platform with overturned
metates supported on wooden cross members which are still in place.
At its end on the kiva floor it was also provided with wooden uprights
plastered in with adobe on each side of the opening to form a rounded
frame. This rounded framework on the floor of the kiva opens out onto
a small semi-circular platform, a foot by two feet in width, and some
three inches high, placed upon the kiva floor as a sort of threshold for
the ventilator entrance. This small platform just mentioned, and the
opening to the ventilator shaft, were separated from the rest of the

kiva by a low wall of masonry eighteen inches high, which ran in a
semi-circular direction from the wall on the one side of the ventilator
shaft to the wall on the other. This last named partition wall served
as a fire screen for the fire pit directly behind it to the north. However,
it is most unusual to find the fire screen or partition wall joining
the walls of the kiva both to the east and to the west. The fire pit
immediately to the north of this screen wall is still slightly to the south
of the center of the kiva. It was well made, sunk into the kiva floor,
lined with slabs of sandstone, and was well filled with ash when found.
A small hole three inches in diameter, eighteen inches east of this fire
pit, may have been the sipapu. No other depression which might have
been used for this purpose is in evidence.

The shard evidence showed the Escavada, Exuberant Corrugated,
and Gallup combination predominating, as was true in the rooms. Of
several hundred shards pried from the plaster of the walls, Escavada
ran as high as 80 per cent, Lino and Red Mesa were next, with Gallup
running a poor last.

A most interesting situation was found on the north side of Kiva 1
where a large section of the kiva was cracking and apparently sinking
into a soft space beneath, along with the north wall. A small test hole,
dug through the side of the kiva in this direction, produced large numbers
of Red Mesa shards, as might have been expected. A test hole dug
through the floor of the kiva produced Lino gray ware.

Evidently Kiva 1 had been excavated in Pueblo II times down
through, or at least to the edge of, the original Pueblo I mound and
possibly extended into the Basket Maker levels. The caving observed
on the northern side of Kiva 1 may indicate a subsequent settling or
decaying out of large volumes of animal and vegetable matter in the
Pueblo I structure on that side.

Kiva 2 (See Plate IX)

Kiva 2 is the northernmost of the kivas of Tseh So, besides being
the largest in diameter and the deepest. The kiva proper is twelve feet
in diameter north and south, thirteen feet in diameter east and west,
and is at a depth of ten feet from the surface adjacent to the wall of
Room 14. The usual ledge or bench runs around the circumference of
the kiva, with the exception of that portion immediately above the
screen wall and the ventilator opening. This bench varies from six
inches in width on the west wall to twelve inches on the east wall, and
is thirty-two inches from the floor. Kiva 2 was not provided with a
southern or "keyhole" extension, but the south side rose vertically to
the surface with no offset or shelf. The ventilator shaft in this case
rose immediately behind and to the south of the south wall of the kiva,
coming out in an opening approximately a foot square, of rough

Fig. 2. a—Kiva 1; b—Kiva 3

masonry, between the kiva and the north wall of Room 6. This ventilator
shaft was accommodated at its open end in the kiva by a coping of
wooden uprights and adobe to form an arcuated opening in the shape
of a "broken-sided" arch. The south portion of the kiva, immediately in
front of the ventilator opening, was screened off from the rest of the
kiva by a low single-thickness wall of masonry, as in Kiva 1. This
masonry extends from side to side of the kiva, leaving a space in front
of the ventilator shaft seventy-two by thirty-two inches. This was not
provided with a low threshold platform, as in Kiva 1. The fire pit lies
immediately to the north of this screen wall in essentially the same
plan as Kiva 1. The fire pit was lined with stone slabs sunk into the
floor, as before. A small hole to the east, six inches in diameter and
some twelve inches deep, possibly served as the sipapu.

Kiva 2 is of especial note because of several interesting cists
or recesses which occur in its walls. One of these is in the north
wall below the edge of the bench, some two feet from the floor. This
is a small recess floored with a single slab of sandstone eight inches
square, the whole recess being roughly five by five inches across, and
six inches deep. The largest of such receptacles occurs upon the east
side of the kiva, also in the side of the bench. This is a large receptacle
of roughly circular outline, some two feet in all dimensions, floored
with a piece of shale. When discovered, this recess was closed and
sealed by a carefully shaped slab of sandstone, rectangular with
rounded corners, which had been placed across the opening and plastered
in place with adobe. The recess, however, was empty when discovered.
Two other recesses or wall cists, occur on the south side of
the kiva immediately to the west of the ventilator opening just on top
of the bench. These openings, each about eight inches in depth, were
neither lined nor closed but merely open alcoves, possibly for the accommodation
of ritualistic paraphernalia.

The plaster of Kiva 2 could be observed in detail insomuch
as large quantities yet adhered to the wall. The plaster in most places
consisted of fourteen layers of pure adobe, each approximately a
sixteenth of an inch thick and superimposed upon the preceding layer.
The layers could be distinguished easily because of the smoke-deposited
soot on the surface of each layer, which formed a number of varves of
alternate dark and light colors. The plaster in this case was not
decorated and not colored.

Several vigas from this kiva were recovered for dendro-chronological
purposes.

Kiva 3 (See Fig. 2b)

Kiva 3 lies the southernmost of all the kivas, and flush with the
southernmost room of the pueblo, Room 19. This kiva also presents
a southern extension which makes it assume the "keyhole" form mentioned

in connection with Kiva 1. Kiva 3 has an additional feature in
the form of four pilasters, which adjuncts are reputed to be a San Juan
feature.[3] The kiva itself is almost exactly round, measuring ten feet
ten inches in diameter, north-south and east-west. It was provided
with the usual bench or ledge, ten to twelve inches wide and fifty to
fifty-four inches from the kiva floor. The floor of the kiva is nine feet
below the original surface and is especially featured by being paved
with large slabs of carbonaceous shale. The southern extension of the
kiva measures six feet five inches east and west, and six feet nine inches
north and south. The back or south wall of this extension is varied in
this case by a bulge or pilaster of plastered masonry to accommodate
the ventilator shaft which rises in that place.

The ventilator shaft itself arises behind the fire deflector, and measures
eleven inches by eighteen inches. The usual wooden lintel is in
place. This shaft passes horizontally underneath the southern extension
of the kiva and rises through the pilaster just mentioned.

The deflector differs from that in Kiva 1 in that it is not joined
to the sides of the kiva on each side of the southern extension, but is
a free-standing wall of masonry thirty-one inches long and sixteen
inches high, and of single stone width.

The fire pit lies directly behind the deflector and is augmented by
yet another upright slab on the deflector end which may have served as
an auxiliary deflector or fire screen. The fire pit measures twenty-one
by twenty inches and is eight inches in depth. It was entirely filled,
when found, with fine white ash.

A small hole three inches in diameter, seventeen inches from the
east wall, evidently formed the sipapu. Two other holes of about the
same size, and about three inches apart, were found to the north of the
fire pit midway from the pit to the wall. These may be diagnosed as
loom holes.

In addition to these holes, there occurred a shallow cist through
the carbonaceous shale floor to the west of the fire pit which contained
the entire skeleton of an ordinary dog (Canis familiaris).[4] This had
been slightly flexed to accommodate it to the oval contours of the cist
which lay only eight inches below the floor. The cist had been carefully
covered over with plastered adobe, the outlines of which were in
very marked contrast to the surrounding shale.

The most striking feature of Kiva 3 is undoubtedly the very
good multiple coat of plaster which yet adheres to the walls below
the bench, and especially the series of designs which were incised into
this plaster (Plate X). These incisions take the form of at least seven
different groups of figures, for the most part on the north and west

walls of the kiva. However, others may have been destroyed by the
scaling away of the outer coats of plaster on the rest of the wall. The
figures were done in a casual, yet systematic manner. Among those
which may be tentatively identified are a seeming house design, a maze,
a fish, and a possible vegetable or tree design. These figures may be
studied in the accompanying illustration (Plate X). In addition to the
figures incised into the surface of the plaster, there are traces of paint
of two colors, black and white, on the north wall. No shapes seem to
be represented in the paint traces discernible at present.

Kiva 4

Kiva 4 is easternmost of the four kivas of Tseh So, and lies
adjacent to Rooms 5, 6, and 15. It most greatly resembles
Kiva 2 in general structure, in its pit-like appearance, and in
the absence of a southern extension and pilasters. Its walls are vertical,
well-formed, and of fair masonry of the same type as the rest of
the pueblo. A goodly amount of plain unpainted and unincised plaster,
of some seven distinct coats, yet adheres to the walls. The usual bench
encircles the kiva, twenty-five to thirty inches above the kiva floor, for
all its circumference with the exception of that portion immediately in
front of and adjacent to the ventilator and the fire screen. The bench
is especially wide at the north side of the kiva, and erratically narrows
to a point and blends into the wall on the east side. On the west it
ends in a well-rounded shoulder, as usual. The kiva floor lies at about
nine feet from the original surface.

The ventilator opening is eighteen inches high by twelve inches
wide and rectangular. It enters the south wall of the kiva for some
three feet, then turns at right angles to the surface as in Kiva 2.

The deflector is similar to that of Kiva 3 in that it does
not adjoin the walls of the kiva on either side of the fire pit. This deflector
lies thirty inches from the ventilator opening, is twenty-one
inches in length and eighteen inches high, and is of single stone thickness.
The fire pit lies just to the north of the deflector and in dimensions
is eighteen by fourteen by nine inches deep. It is lined with
squared sandstone slabs.

No sipapu was definitely located as such, but two receptacles or
cists occurred in the walls of the kiva. One of these is a semi-circular
opening twenty-four inches wide and fourteen inches high at its apex
which was let into the end of the west bench five inches above the floor.
This is a plain ungarnished opening, fashioned in the masonry and lined
with adobe. Another of these wall cists or openings occurred on the
west wall above this last named, although of somewhat smaller dimensions.
It presented the same features, and was lined with adobe, as
the previous one.

Evidence From Shards in the Superstructure

From the shard evidence found in the superstructure of Tseh So,
this structure is plainly of Pueblo II origin. On a table in back
(Table I) the three highest percentages of shards under each of
the headings are noted. Those wares which were found in especial
abundance at Tseh So are Escavada Black on White, Gallup Black on
White, Chaco Black on White, Mesa Verde Black on White, Red Mesa
Black on White, La Plata Black on Gray, Kana-a Black on Gray, Kana-a
neck-banded, Lino Gray Ware, Wingate Black on Red, Puerco Black on
Red, Tusayan Polychrome, Chaco Corrugated, and Exuberant Corrugated.
There were also represented by a few, or in some cases by a
single shard, several trade wares of interest which will be noted in
place. Those mentioned above constitute by far the greater bulk of the
shards found on the floor level. In seven out of twenty-four rooms (only
twenty rooms provided floor level data) Escavada displays the highest
percentage. Exuberant Corrugated leads next in six rooms, and Gallup
in five rooms. Of those second in importance in this category, Escavada
leads with seven instances out of the twenty-four, Exuberant
with six, Red Mesa with four, and Gallup with two. Of those types
third in importance on the floor level, Red Mesa occurs in six instances,
Exuberant Corrugated in four instances, Gallup also in four instances,
Escavada in two. From this it may be seen that the combination
which runs highest in percentages on the floor level is predominantly
Gallup Black on White, Escavada Black on White, Red Mesa Black on
White, and Exuberant Corrugated. There are a few erratic percentages
such as McElmo leading with 18 per cent in Room 22, but Escavada
runs a close second with 17 per cent. Sixteen per cent of the shards
on the floor of Room 4, and 14 per cent in Room 14, are Chaco Black on
White. With the exceptions just noted, the Escavada-Exuberant-Gallup-Red
Mesa combination reigns supreme. There is no need to mention
that this is a Pueblo II combination of perhaps early Pueblo II
date because of the large percentages of Red Mesa present.

The shards which were used as spalls in the plaster of the general
masonry of the walls were of invaluable service in the dating of the
pueblo. Of these, only Rooms 4, 6, 7, 9, 11, and 14, exhibited enough
to gain any appreciable percentages. In these, the highest percentages
are found to be Escavada in two instances, and Exuberant Corrugated
in three instances. The second and third in importance tended to be
Red Mesa and Escavada, and in one case, in Room 9, 13 per cent of the
shards were found to be Lino Gray Ware. This shard combination is
found to be essentially the same as that diagnosed from the floors of the
rooms, although with Pueblo I leanings.

Rooms 2, 10, and 11 each contained a sealed doorway in which many
shards were used as spalls. In these Exuberant Corrugated, Escavada,
Gallup, Red Mesa, and Wingate predominated.

Dating. A viga beam (Museum No. Bc50 35/30) yielded the only
date available from the entire ruin. This is A. D. 922+, a non-bark date.
This specimen is derived from a floor in Room 15 associated with a
shard combination of Escavada, Exuberant Corrugated, and Red Mesa.
A number of other specimens gave probable dates, but this one is the
only one recorded as reliable. Dr. Douglass has examined this and the
other specimens as well, and substantiated only this date. The other
pieces are so decayed as to leave some doubt.

Substructure or Pueblo I

It was soon discovered after excavation had progressed to some
extent, that the main pueblo, which was designated as Pueblo II, had
been built on the mound of an earlier structure. In addition to this
general information, it was soon ascertained that the Pueblo II structure
did not cover entirely the mound beneath, but was built for the
most part on top of and to the east of this mound. That area lying to
the north and west of Rooms 2, 11, 20 and 22 of the superstructure
was found to contain several rooms of a building of a date
prior to that of the one which has been described above, and these
rooms were unencumbered by any later structure built upon them. It
is perhaps to be supposed that the growth of the Pueblo II structure was
from the central section and the east side towards the west, and that
this growth was never completed so as to cover all of the mound beneath.
All of these structures beneath the Pueblo II superstructure,
are here designated as substructures and numbered according to rooms.
Three such were excavated on the western and northern side of the
pueblo mound.

In addition to the substructures which appeared on the northwestern
side of the pueblo, it became evident that the entire Pueblo II
superstructure had been built upon the remains or mound of an earlier
edifice. This, of course, is shown along the western walls of Rooms
2, 20, and 22, where it appears that the walls of the Pueblo
II bear no relation to the walls of the substructure beneath and that
the substructure was evidently a mound with weathered walls and
filled-in rooms before the superstructure was ever built upon it. This
filling in of the substructure may or may not have been intentional.
The stratigraphical evidence from the fill of these rooms, however,
shows a combination of water-laid and aeolian fill mingled with the
crumbled debris of the peculiar masonry so as to indicate a quite
natural disintegration of this substructure, representing an unconformity
of some duration of time.

The walls of the substructure, immediately below the Pueblo II
structure, bore no relation to the walls of the superimposed structure,
and, furthermore, were of a much different type. It is also worthy of

note that the floor levels of the Pueblo II superstructure were by no
means on an even level. It would appear that the superimposed pueblo
was built upon an earlier one which was also a rounded mound with
the apex somewhere under Rooms 7 and 10. This high point
was somewhat to the south of the apex of the latter or final mound.
In the light of all this evidence, as well as the abundant shard material,
it may be assumed that this substructure is Pueblo I in date.

General Features of Substructure (Pueblo I)

Walls. The walls of the substructure of Tseh So were of a most
interesting variety. Character was given to this masonry by a series
of large upright sandstone slabs which formed the core of the wall,
especially at the base. Many of these slabs are two and a half feet in
each dimension, with a thickness of two to four inches. Several
of them show extended use as metates. These larger slabs were set in
single thickness, edge to edge, upright, in a manner which is reminiscent
of the Basket Maker pit house structures which may be seen on all
sides of this ruin. These uprights were embedded in the ground a sufficient
number of inches to hold them solidly in place, as well as to
support a considerable quantity of crude masonry above. The wall at
its base was rendered thicker, usually to a depth of about eight inches,
by the addition of thick layers of puddled adobe on both sides of the
sandstone slabs. On top of this rather precarious masonry base, the
wall was heightened by more puddled adobe, small slabs of sandstone
laid horizontally, and numerous spalls stuck into the adobe at random.
This wall was a very much weaker structure, in its upper portion, than
even the rather shaky masonry of the superimposed Pueblo II. However,
it seems to have served very well insomuch as it is still standing
in some places to a height of five feet above its original floor level.

Examination beneath the floors of the superimposed structure
(Pueblo II) invariably revealed walls of this type running at angles
with no relation to the upper walls (Plates Ib, VIIb, c). In most cases,
due to the impossibility of completely removing the superstructure, it
was difficult to ascertain the exact dimensions of those substructure
rooms without endangering the students and the workmen. Enough
were followed, however, to gain a fair idea of their extent and general
plan. It was also determined that the kivas of Pueblo II had been sunk
down through and at the cost of the Pueblo I structure beneath. Several
old corners, especially beneath the floor of Room 15, gave
ample evidence of this. Needless to say, the walls which are here typical
of Pueblo I fall far below any registered wall sequence which has so
far been worked out for the Chaco Canyon. Even the masonry of the
Pueblo II structure is below the heretofore designated Type I for this

region. Also, these walls are remarkably like the Wingate Phase walls
which have been described by Mr. Harold Gladwin.[5]

Floors. The floors of the substructure are hard-packed adobe but
appear to be quite an accidental layer. They are in no wise the smooth
and evenly finished floors of the Pueblo II structure. In several places
in the substructure the floor appears to be composed of several layers
of adobe and refuse as though it were an accidental accumulation
packed down by hard use. Bones, and other material, embedded in the
floor, would indicate a surface of this origin. Also, at the edges of the
rooms there was no rounding of the floor into the wall plaster as had
been the case in the Pueblo II. Floors of this nature were revealed
over the entire area of the Pueblo I substructure.

Roofs. Little evidence of the roof structure of the Pueblo I was
secured. However, several lumps of adobe in the fill of the substructure
may serve as evidence for this part of the pueblo. These adobe lumps
seem to indicate a roof which was more nearly wattlework than any
other type of construction. Comparatively large branches or sticks,
some of them as much as an inch in diameter, were laid directly at
right angles upon larger ones which apparently did service as vigas.
The adobe was applied directly to these sticks just mentioned, and yet
bears the imprints of them to the extent that they may be identified as
piñon, cottonwood, and possibly some other species. It may not be determined
from the few adobe impressions recovered, whether these
branches were actually woven or merely laid across the vigas. No split
pieces were evidenced in this construction. It is impossible to state
how the coping of the roof was finished, nor, indeed, whether the Pueblo
I structure had a roof strong enough or enough debris represented to
indicate a second story.

Cists, Bins, Fireplaces. A small depression in Room 2 of
the substructure may have been used as a fireplace. This is a rudimentary,
shallow basin, adobe-lined and in no way separated from the
floor, in the northwest corner of this room. A few ashes and the slightly
blackened wall at the back of this pit indicate the presence of at least
one fire. The fact that no other fireplaces were discovered in the substructure
may be due either to: (1) the possibility that they may have
been missed as being those portions underlying the heavy walls of
Pueblo II and therefore not excavated, or (2) the possibility that they
may have been located outside of the pueblo proper.

Only one structure which may be termed a cist or a bin was discovered.
This is a polygonal bin outlined on the floor of Pueblo I in that
area beneath the floor of Room 4 of the superstructure. This
was composed of seven slabs of sandstone set on edge, embedded into

the floor so as to form a seven-sided receptacle. As stated in connection
with the fireplaces, other cists or bins may exist in those portions
which could not be excavated.

Special Structures. Under this heading may be mentioned a structure
of unknown use which occurred in that portion of the Pueblo I
pueblo which lies immediately under Room 10 of the superstructure.
In this place, within the enclosing walls of the substructure,
there occurred an enclosure composed of flat slabs on edge which may
or may not have been a cist. The slabs were so arranged as to enclose
an area approximately six feet long and a foot and a half wide, the
long axis extending in a north and south direction. The northerly end
of this peculiar enclosure lies beneath the south wall of Room
11. No indication of its use is evidenced nor does it recur in any other
part of the excavation.

Shard Evidence From Substructure

Throughout the fill of the substructure, and in greater numbers
on the floor, the shards display a predominant percentage of Red Mesa.
The Red Mesa is varied in some instances by Lino, Escavada, and
Kana-a neck-banded—Lino running the highest percentage of these
as may be seen from the accompanying table (Table I). Several more
or less complete Red Mesa bowls and one Kana-a pot were recovered
from the floors of the rooms in the west substructure. This evidence
is supplemented by abundant Red Mesa shards in the levels under the
room floors of the superstructure over the whole of the pueblo. This
evidence, coupled with the typical masonry, and the typical artifacts,
make the designation of the substructure as Pueblo I conclusive.