MESA VERDE Big Juniper House of Mesa Verde, Colorado Wetherill Mesa Studies

KIVAS

It is fairly certain that the three kivas excavated at Big Juniper House were the only ones at the site. All three were oriented northwest-southeast, as determined by an axis drawn through the ventilator, hearth, sipapu, and niche in the face of the north banquette. Kiva A showed definite evidence of remodeling, and Kivas A and B had curious structures built over them after they were abandoned. The structures are referred to as "later walls over kivas" and are indicated by the shading in the site plan (fig. 5). The later walls and the compound retaining wall north of Rooms 4, 5, and 24 are probably the latest features at Big Juniper House and belong to Component E.

Evidently, all three kivas were in use during the Component D occupation. Parts of two vessels occurred on the floors of different kivas. In analyzing the pottery, we discovered that parts of a vessel from the floor of Kiva B fitted a partially restorable vessel from the floor of Kiva A, and part of a vessel from the Kiva B floor fitted sections of a vessel from the floor of Kiva C.

The vessels did not seem to have been discarded on the kiva floors as trash, nor did they appear to have drifted in from surrounding areas. Rather, their presence seemed a deliberate result of breaking the pots and then dividing the fragments between the kivas. Perhaps it was a "ceremonial killing" of the vessels related to the deliberate abandonment of the kivas, or possibly a "killing" of the kivas to obtain roof beams and other wood for building new rooms or a pueblo elsewhere.

All three kivas were later used as trash dumps. The probability that all of them were abandoned at about the same time, the presence of trash in the kivas, and the fact that later structures were built over two of them indicate that Big Juniper House continued to be occupied with out kivas. The lack of definite Component E rooms suggests either that the occupation was small or that any structures that were built (other than the upper, later walls and the compound wall) were impermanent.

It is also possible that many Component D rooms continued to be occupied after the kivas were abandoned. Evidence for this is the general scarcity of trash in most of the rooms, especially in the East House Mound. It is common at Anasazi sites for rooms as well as kivas to be secondary trash dumps.

Kiva A

Dimensions. About 15 feet in diameter, although it is somewhat shorter north and south (figs. 23 and 24). The floor is approximately 7.7 to 8.4 feet below the present surface. The inside dimensions of the later wall over Kiva A are approximately 14.2 feet north-south by about 17.5 feet east-west. The top of the later wall was exposed on the present surface.

Fig. 23 Kiva A, looking east (toward top of picture). Later wall over kiva on west, east, and south sides. Dark area on floor east of hearth is a subfloor pit.

Fig. 24 Plan and sections of Kiva A. (click on image for an enlargement in a new window)

Banquette. The banquette face in Kiva A, and in the other kivas, was constructed by using a sandstone slab base, between and over which was coursed masonry of some what smaller stones than those used for the surface rooms. The simple walls were made primarily of chipped-edge masonry and some scabbled masonry. The courses were often chinked with rows of small stones or pottery sherds. The latter were not found in the walls of surface rooms. The height of the banquette above the floor of the kiva averaged about 3 feet and varied from 2.9 to 3.5 feet. The width of the banquette, upon which the pilasters were built, averages about 2 feet, with a variation of 1.5 feet on the north between the third and fourth pilasters (numbered clockwise from the first pilaster west of the ventilator) to 2.2 feet between the fourth and fifth pilasters and between the sixth and first pilasters. There is no deepened interpilaster space (southern recess) between the sixth and first pilasters in Kiva A or in the other Big Juniper House kivas.

Plaster remained on several parts of the liner and on the banquette, with a maximum thickness of about 0.2 foot. Where the plaster was thickest, there was evidence of at least six coats. The first coat was a thick layer of yellow-brown adobe applied directly to the stones. The second was a thin layer of white plaster, the third was a thick layer of yellow-brown adobe like the first coat, the fourth was again a thin white layer, the fifth was a thick yellow, sandy adobe or plaster, and the sixth was again a thin layer of white plaster.

The liner was constructed directly on the floor and against the raw earth of the hole when the Indians built the kivas. Parts of the liner were torn down to see if earlier walls stood behind it, but only sterile earth was encountered.

There was no liner at the back of the banquette, a feature usually seen in Pueblo III kivas in the Mesa Verde area. Above the banquette the native earth served as the kiva wall. Perhaps the earth was originally plastered, but no evidence of plaster remained.

Niche. In the north face of the banquette, on the kiva axis, is a small square niche about 0.4 foot below the top of the banquette, with dimensions of 0.4 foot on the sides and about 0.7 foot deep. Nothing was found inside of it.

Ventilator. On the south face of the banquette is the ventilator tunnel, opening about 0.5 foot above floor level. The floor may have been level with the bottom of the ventilator, but had slumped in at this point due to its construction over the subfloor ventilator, which was filled with trash and soft earth. The rectangular hole measured about 1 foot wide and 2 feet high. The top of the tunnel was about 0.8 foot below the banquette. Evidently the floor-level ventilator was masonry-lined on the sides and roof and probably also in the shaft section, but the shaft had collapsed to just above the base, to a height of 2 feet. The roof of the tunnel was also lined with slabs supported by wooden lintels, the decayed remains of which were found during the excavation.

The ventilator shaft seems to have remained in the same location for the two types of tunnels. The subfloor ventilator tunnel began at the base of the shaft and inclined steeply downward, ending abruptly at a vertical earth face 0.3 foot in front of the hearth. Although we found no remains, the subfloor tunnel was probably roofed inside the kiva, with an opening into the rounded portion of the subfloor tunnel before the hearth. The vertical face provided a built-in deflector and served to deflect the air up into the kiva.

Later, the subfloor tunnel was filled with trash and dirt, and then a floor-level ventilator was constructed. The floor-level tunnel was somewhat narrower than the subfloor one, as the opening into the banquette face was smaller than the underlying walls of the earlier ventilator.

Pilasters. Kiva A had six pilasters. They were designated pilaster 1 through 6, counting counterclockwise from a point on the south axis of the kiva. We had reason to believe that Kiva A was originally constructed as a four-pilaster kiva, like Kivas B and C, and that the two narrow pilasters (second and fifth) were installed later, perhaps during a major remodeling of the kiva. The spaces between the first and third pilasters and the fourth and sixth pilasters were a good deal larger than between the first and sixth and the third and fourth. It is thus possible that the four pilasters were inadequate and that the two smaller pilasters were constructed to support a sagging roof. The narrowness of the two pilasters and features of their construction also point to a different time of construction.

The four wide pilasters were all constructed of chipped edge masonry with abundant, small sandstone spalls and sherds used for chinking material. These pilasters were also not so well made or so regular in appearance as the two narrow pilasters. The latter were made of pecked face masonry with squared corners. Pecked facing is a later masonry style than chipped-edge shaping, and is further evidence that the narrow pilasters were built at a later time than the four wide pilasters. Chinking was sparsely used in the narrow pilasters, because the evenness of pecked-face stones did not need spalls or chinking to level the courses.

All of the pilasters were set flush with the edge of the banquette and were rectangular in shape. In these two characteristics the pilasters of Kiva A differed from those in Kivas B and C and probably indicate (along with other evidence presented later) that Kiva A was constructed earlier than the other kivas. It is a common feature of Pueblo III kivas to have pilasters that flare from front to rear and are set back from the edge of the banquette.

The height of the pilasters above the banquette ranged from 1.8 feet (third) to 2.4 feet (fourth). It is probable that all the pilasters were about the same height during the occupation of the kiva, and that their varying heights upon excavation were due to differences in preservation.

A possible secondary roof support, 3.3 feet east of pilaster 1, was imbedded in the dirt wall of the kiva about 0.65 foot above the banquette. It was about 1 foot high by 0.9 foot wide and was composed of nine courses of very small stones set in two columns.

Floor. About 0.2 foot of adobe mud on sterile earth covered most of the floor area. The subfloor ventilator was filled with trash and loose dirt and then plastered over. A cist, discussed below, was also filled in the same manner and then plastered. The floor was basin-shaped; it sloped up to the walls and reached its lowest point near the hearth.

The hearth was a circular pit with raised edges, about 0.6 foot deep, and filled with white ash. Pieces of unworked sandstone were mudded in at the edges of the firepit. Several sherds (table 3) were in the firepit fill.

A small cylindrical hole, about 0.5 foot deep, was in the floor between the hearth and the north banquet face, in the general line of the kiva axis. The hole, or sipapu, was filled with soft dirt, part of the general kiva fill, and was probably open at the end of the Kiva A occupation.

An oval pit east of the hearth was later filled in and plastered over. It was about 0.5 foot deep. In the pit fill were sherds (table 3) and a hammerstone (table 9).

No definite defector was found. A whole, plain-faced metate (fig. 97f) was partly imbedded in the upper fill of the subfloor ventilator and may have served as a deflector during the last stage of Kiva A occupation, when the floor-level ventilator was in use.

Entry. We found no definite evidence of an entry, such as ladder holes near the firepit, but it is assumed that the entry was in the roof, above the hearth. This opening would have served also as the smoke hole.

Later wall over kiva. The later wall built over the kiva was constructed after the kiva had been partially filled in. The bottom of the later wall was at about the height of the top of the first, second, and third pilasters on the western perimeter of the kiva (1.65 to 2 feet above the banquette) and was never more than 2.3 feet or less than 1.65 feet above the banquette around the rest of the kiva. The wall was stepped back from the bottom course to the top on the west and east sides and more or less vertical on the north and south. It followed fairly closely the shape of the kiva on the south, east, and west sides but diverged sharply on the north, cutting across the inside from the third to the fifth pilasters. On the north side, the wall was constructed on fill composed primarily of building stones that extended almost to the floor of the kiva. About 25 percent of the stones in the upper wall was pecked-face masonry. The remaining stones were scabbled and chipped-edge masonry. The height of the upper wall ranged from 2.3 to 3 feet.

The upper wall was primarily a compound wall of five to six courses in height. Part of the southern perimeter of the upper wall had a slab wall base; the rest was coursed masonry.

Other features. None.

Artifacts on floor. Sherds of a Mancos Black-on-white jar (fig. 57b) were found on the floor and floor fill and in the ventilator tunnel; most of them came from the ventilator area. Sherds of another Mancos Black-on-white jar were found in the floor fill and on the floor (fig. 57c), but they were so scattered that we were not able to tell where the jar was originally located. It is probable that when the kiva roof fell, or when the kiva was filled, the jar shattered and the sherds "exploded" over a wide area. Fragments of a third Mancos Black-on-white jar were found on the floor and floor fill in Kiva A, and a part of the same jar was found on the floor of Kiva B, in front of the ventilator tunnel.

A conically shaped sandstone block (fig. 135) was found on the floor of the floor-level ventilator shaft, resting on its side and apparently placed there intentionally. It may have been a ceremonial object.

A burned, bird-shaped concretion, possibly used as a firedog or pot support, was on the floor near the hearth (fig. 141, right). Unfortunately, we did not record its exact location in the kiva. The plain-faced metate in the subfloor ventilator fill was discussed previously.

Several artifacts placed intentionally on the kiva banquette included a plain/troughed metate (fig. 97k), in the first interpilaster space next to pilaster 1; a possible anvil (fig. 132, bottom), in the fill on the south part of the banquette between the first and sixth pilasters; a possible door slab (fig. 136a), midway between the fifth and sixth pilasters on the banquette; a troughed metate part on the south banquette; two Subtype 1A manos (one in the fill) on the south banquette (fig. 98a); a side-notched hammer (fig. 112d), in the fill of the south banquette; two Type 2 hammerstones in the fill of the south banquette; a plane (scraper) (fig. 122f), in the fill just above the north banquette; and, on the south banquette, a subcircular travertine pendant with a central perforation (fig. 133g). Also, various sherds were found (table 3) in the fill just above the banquette.

Fill. In general, the earth fill of the kiva was darker and softer toward the top and more orange and compact in the lower parts. No artifact changes were noted. About 5.6 to 7.6 feet below the surface, numerous building stones composed most of the fill. It is probable that this was deliberate fill upon which the upper wall was constructed over the kiva. Also in this level we encountered sections of a human vertebral column and other scattered human bones. Articulated portions of a human leg and arm bones were labeled Burial 22 (fig. 177). Several of the sherds of the restored Mancos Black-on-white jar shown in figure 57b were near the burial and may have been associated with it.

The large number of artifacts found in the fill indicated that Kiva A was used as a trash dump after its abandonment. Several artifacts are worth mentioning—restorable vessels found in the fill that are not listed in table 3. A partially restorable Cortez Black-on-white jar (fig. 46d) and a partially restorable plain bowl, sliped and polished on the interior but without painted design, were found in the upper fill of the kiva, and about half of a small McElmo Black-on-white pitcher or jar was found in the fill of the subfloor ventilator tunnel. Several artifacts from the floor fill may have been associated with the use of the kiva. Two azurite beads were found together close to the floor, as well as various other artifacts and refuse material listed in tables 3, 6, 9, and 10.

Dates. Several pieces of charcoal were found in the fill, but none provided dates. It is likely that Kiva A was constructed during the Component C occupation of Big Juniper House and therefore would have been associated with the jacal structures underlying Rooms 1a, 1b, and 10, and with the other features of this component mentioned previously. If this were the case (reasons for this inference are discussed below), Kiva A was probably constructed around A.D. 1050. It undoubtedly was used through the Component D occupation and probably abandoned somewhat later than 1100 but before 1150. About A.D. 1150, the Component E "later wall over the kiva" was built.

The existence of McElmo Black-on-white sherds and the partially restorable vessel in the deliberate fill of the subfloor ventilator tunnel indicates this remodeling took place around 1080 to 1100, about the time that McElmo Black-on-white began to be made in the Mesa Verde.

Remarks. When Kiva A was constructed, it had a subfloor ventilator, probably four pilasters, and a floor cist or pit next to the hearth. In the Mesa Verde area, kivas with four pilasters that are rectangular in shape and set flush with the banquette face are typical of Pueblo II. Kivas of this period do not normally have southern recesses or a masonry lining above the banquette. Subfloor ventilators in Mesa Verde kivas have been most commonly found in Pueblo III kivas, although it is a rare trait in that period. However, more sites, and many more kivas, of Pueblo III than of earlier periods have been excavated in the Mesa Verde area. As a result, the subfloor ventilator data may be "overweighted" for comparative purposes. Remodeling of subfloor ventilators to floor-level ventilators, as in Kiva A, is quite typical of Pueblo III kivas.

It has also been suggested that four-pilastered kivas did not have cribbed roofs, but rather followed the old roofing pattern of the four-post kivas or "proto-kivas" and pithouses of earlier times (Lancaster and Pinkley, 1954, pp. 55-56). The roofs in these are assumed to consist of four horizontal beams placed on the roof supports. In the case of Kiva A, the primary beams rested upon the four masonry pilasters over which smaller poles spanned the rectangular space to form the top of the roof. Other small construction members were placed from the four primary beams to the top of the earthen kiva walls to form the sides of the roof. Shakes, bark, reed, or other material, and a coat of adobe finished off the roof. It is not a hard and fast rule, but this generally appears to be the pattern. It is almost a certainty that six- and eight-pilaster kivas, primarily confined to Pueblo III, had cribbed roofs and therefore they reflect a basic change in roof construction.

It is probable, then, that Kiva A underwent a major remodeling that included a change from a subfloor to floor-level ventilator, and a four-pilaster "old" kiva roof design to a six-pilaster, cribbed-roof of "modern" design. The pit next to the hearth was probably filled and plastered over at this time, too.

The question remains as to why Kiva A was remodeled. If the kiva had been in good condition there would have been little reason to tear down the roof, install two new pilasters, and build a new roof. If Kiva A was extensively changed, as has been suggested, it must have fallen into a state of disrepair and perhaps the roof had already collapsed. Although there is little evidence at Big Juniper House to suggest abandonment of the site between Components C and D, there may have been a time when Kiva A was not in use. Perhaps there was insufficient manpower to keep it in good repair. If, as seems likely, the population increased at the beginning of the Component D occupation, with new styles in architecture including most of the masonry rooms at the site, the reoccupation of an older kiva with appropriate changes in its design would probably have been easier than building a completely new one. It is also probable that sometime after Component D began, Kivas B and C were constructed to accommodate other groups.

From its relationship to the surface rooms at Big Juniper House, Kiva A may be considered an integral part of an occupation unit. This has been mentioned in my remarks on Rooms 10 and 11. During the Component D occupation, the roof of Kiva A was probably part of an outdoor courtyard or work surface that included Room 10 and Area D, and possibly Rooms 21 and 22. It was probably the focus of religious and social activities for the people of the occupation unit as well.

Later, Kiva A was used as a trash dump and a wall was built over it. The purpose of this wall is a mystery, but it appears to be a feature restricted to kivas. One other example is Kiva B, and similar later walls over kivas were found at Two Raven House and at Sites 1230 and 1253.

Kiva B

Dimensions. About 13 feet in diameter, varying between 12.8 and 13.2 feet (figs. 25 and 26). Floor is approximately 8.3 feet below present ground surface. The inside dimensions of the later wall constructed over Kiva B are 17.1 feet long north-south and 14.2 feet wide east-west.

Fig. 25 Kiva B, looking southwest (top of picture). Later wall over kiva is on the southern part, in front of the ventilator shaft.

Fig. 26 Plan and sections of Kiva B. (click on image for an enlargement in a new window)

Banquette. Kiva B followed the same style of banquette facing as in Kiva A—a base of slabs with intermixed and overlying coursed masonry (fig. 27). Sherds and sandstone spalls were also used for chinking as in Kiva A. Several of the stones in the banquette face were dressed by pecked facing, a situation that did not occur in Kiva A. The remainder of the stones were primarily scabbled, with a few chipped-edge. The banquette averaged about 2.8 feet above the floor; it was plastered on top with about 0.2 foot of adobe mud. There was also some plaster remaining on the banquette face, but not enough to determine the presence of multiple coats. The width of the banquette averaged about 1.5 feet, with a range of 1 to 2 feet.

Fig. 27 Detail of Kiva B banquette face.

Liner above banquette. Kiva B was the only kiva at Big Juniper House with a liner above the banquette, a common feature of Pueblo III kivas in the Mesa Verde area. The liner was built only on the north part of the banquette, between the second and third pilasters, and was similar in style to the banquette face, having a slab base topped by coursed masonry of small stones (fig. 28). However, smaller stones were used in the liner than on the banquette. It was probably about the same height, 2.3 feet, as the second pilaster. The remainder of the wall above the banquette was the dirt wall of the original excavation and was probably plastered.

Fig. 28 Detail of north section of banquette face with niche, Kiva B.

Later wall over kiva. The later wall built over Kiva B was constructed on fill at the level of the tops of the pilasters, and extended down lower on the west side to slightly above the level of the banquette. The later wall over Kiva B was similar in most respects to the wall over Kiva A; the one difference is the existence of a cross wall or partitioning wall that was bonded to the west part of the later wall and extended east (fig. 29). The height of the later wall ranged from 3 to 4.1 feet and averaged about 3.5 feet. The highest point was at the western side near the area of the cross wall. Portions of the later wall were exposed at the present surface. As in the Kiva A later wall, approximately 25 percent of the building stone was pecked faced, the rest scabbled or chipped-edge. We found no occupation surface to indicate whether there was a room or structure inside the later wall and the cross wall. The later wall was most clearly defined on the north and west sides where it was carefully stepped back from bottom to top, to a maximum displacement of 1.9 feet on the north side.

Fig. 29 Later walls over Kiva B on north (top of picture) and west sides, and part of cross wall in the center.

Niches. There were two niches. One was in the usual place on the kiva axis in the northern portion of the banquette face (fig. 28). It was about a 0.3-foot-square opening with the top inner edge 0.4 foot below the top of the banquette. It extended into the banquette 1.3 feet. The other niche was in the southern banquette face about 0.6 foot east of the first pilaster. Its dimensions were 0.45 foot wide, 0.2 foot high, and 0.7 foot deep. Nothing was found in either niche.

The north niche was difficult to find, having been filled and plastered over. The other niche was probably open.

Ventilator. The ventilator was a floor-level type, probably originally masonry lined throughout, and the shaft was built of both slabs and coursed masonry. Slabs, probably supported by wooden lintels, made up the tunnel roof, and the sides of the tunnel were coursed masonry. The tunnel floor was native earth.

Pilasters. Kiva B had four pilasters, all flaring from front to rear and set back from the edge of the banquette 0.15 foot. They ranged in height above the banquette from 1.6 feet (fourth pilaster) to 2.3 feet (second pilaster). The other two pilasters were nearly 2 feet high. The fourth had partially fallen, and it is probable that it, too, was about 2 feet high when the kiva was used.

The pilasters had pecked-face stones with the corners nicely squared. They were decorated with bands of small sandstone spall- and sherd-chinking between the courses of the larger stones.

Floor. The basin-shaped floor was formed by about 0.3 foot of adobe applied on the sterile earth.

The hearth was a D-shaped pit, 0.7 feet in maximum depth, with several small sandstone slabs and rocks mudded into its sides. It was filled with fine white ash and contained no artifacts.

No sipapu was found.

Two cists were in the floor: Cist 1 below the fourth pilaster, and Cist 2 between the first and second pilasters. Both had been filled with trash and plastered even with the floor before the kiva was abandoned.

Cist 1 has a flat floor and deeply undercut sides, extending about 1.1 feet under the banquette on the east. Its depth is approximately 2.6 feet. In the cist fill we found eight sherds (table 3), two Type 1A manos, one Type 2 hammerstone (table 9), and a miniature ladle (fig. 90). We also found an unfired teardrop-shaped clay pendant (fig. 96, left) and a number of unworked bones of turkey, black-tailed jackrabbit, deer, and unidentified mammals (tables 12 and 13).

Cist 2 had straight sides and a flat floor, and was approximately 0.8 feet deep. Four sherds were the only artifacts found in the fill (table 3).

The deflector was indicated by a mud line of a different color from the surrounding floor (fig. 25). It is assumed the deflector was a short jacal wall—a type of deflector found occasionally in Mesa Verde kivas. It was situated about 1 foot from the south edge of the hearth and about 2.1 feet from the ventilator opening. The deflector was about 3.4 feet long and 0.6 foot wide, with a possible extension, on the east side that curved south.

Entry. No entry was indicated, but it is assumed that there was a hatchway in the roof above the hearth.

Other features. None.

Artifacts on floor. The plan and photograph of Kiva B show the location of artifacts found on the floor (figs. 25 and 26). Pottery consisted of a restorable Mesa Verde Corrugated jar (Pot 1), close to the east side of the hearth (fig. 44b); a partially restorable Mancos Corrugated jar (Pot 2), southwest of the hearth; the base of a black-on-white jar (Pot 3), north of the Mesa Verde Corrugated jar; a partially restorable Mancos Corrugated jar (Pot 4), close to Pot 3; a partially restorable Mummy Lake Gray jar (Pot 5), north of the ventilator opening (fig. 36); a Mancos Black-on-white bowl (fig. 56f), in the same area as Pot 5; and a Mancos Black-on-white jar, also in the same general area as Pot 5. Sherds of the last two were found, respectively, on the floor of Kiva C and on the floor of Kiva A.

Stone artifacts included a Type 2 hammerstone, next to Pot 1; a Type 2 hammerstone, next to Pot 2; a Type 1 hammerstone, northeast of Cist 2; a pitted hammerstone (fig. 116f), over the east side of Cist 2; a Type 1A mano, in the Pot 2 area; a Type 1A mano, close to the north banquette face between pilaster 2 and the north banquette face niche; a full-grooved, unfinished ax (fig. 118a), over the west side of Cist 2; a bird-shaped concretion (fig. 141, left), possibly a pot support west of the hearth; a Type 1 unspecialized milling stone, to the north of the hearth; a Type 4 unspecialized milling stone, over the southeast side of Cist 2, next to the pitted hammerstone; an unmodified pebble showing possible slight use as a polishing stone, east of the pitted hammerstone and Cist 2; and a large, unmodified flake, over the north side Cist 2. Also on the floor below the third pilaster were two skeletons, one of rabbit and one of kangaroo rat. A Type 1 unspecialized milling stone was just under the floor and below the second pilaster.

One of the most interesting finds at the site was an apparent cache of eight mammal bone awls and a mammal bone scraper (fig. 149, ch. 5). These were found 3 feet east of the ventilator, in the fill next to the southern portion of the banquette. They had probably fallen from the banquette.

The following artifacts were on the banquette: a large Mancos Black-on-white jar sherd (fig. 62l); a Type 3 deer bone awl and a Type 2 hanmerstone, on the western portion of the banquette; a combination-grooved abrader and three utilized flakes, on the north part of the banquette; and a roughly rounded slab fragment, on the south banquette. A whole Mancos Black-on-white miniature ladle (fig. 90), found in the fill near the second pilaster, might have fallen there from the banquette. Pottery sherds taken from the fill just above the banquette are listed in table 3 as Kiva B, Banquette Fill.

Fill. There was no discernible stratigraphy in the fill. The kiva was excavated in artificial levels to the floor, and no significant change was noted in the artifact types recovered. In the lower fill and floor fill we found a restorable McElmo Black-on-white bowl with a corrugated exterior (fig. 73a). Stone and bone artifacts, sherds and worked sherds found in Kiva B fill are listed in tables 3, 6, 8, 9, and 10. The quantity of artifacts and refuse material, mostly in the upper fill of the kiva, indicates that it was used primarily as a trash dump after its abandonment.

Dates. The following charcoal specimens, all juniper, were dated. All came from the fill; two, MV—1659 and MV—1688, came from Level 3 (6.1 to 7.8 feet below the surface) on the west side of the kiva in the vicinity of the second pilaster. The third, MV—1692, from the floor fill, gave the latest date at Big Juniper House.

The "B" after the 1130 date indicates bark is present and that 1130 is very likely a cutting date—indeed the only reasonably certain cutting date derived from our wood and charcoal specimens. Kiva B does not show evidence of burning and it is likely that the specimen is not part of a structural feature in the kiva, but rather part of the fill deposited after the kiva's abandonment. However, the fill in which MV—1692 occurred was deposited before the Component E later wall of the kiva was constructed, and therefore would have been earlier. Based on the evidence from the artifacts found associated with the floor of the kiva and in the kiva fill, it is probable that the 1130 date is quite close to the actual time of abandonment of the kiva.

Remarks. Several architectural features of Kiva B point to its relative lateness when compared to Kiva A. The flaring pilasters, the pecked-face masonry used in the pilasters and scattered on the banquette face, the pilasters set back from the banquette edge, and the liner above the banquette are all features typical of Pueblo III kivas in the Mesa Verde area. These traits indicate that Kiva B probably was constructed no earlier than about A.D. 1100. In all respects, except for the number of pilasters, Kiva B is the latest kiva typologically at Big Juniper House. The four pilasters may indicate a holdover of an earlier style of roofing, or possibly the construction of a cribbed roof on four pilasters.

The physical relationship of Kiva B to the rooms is also an indication that it was the last kiva to be built. The integration of kivas within room blocks is a trait not well established before Pueblo III, but Kiva B was enclosed on three sides by rooms and is therefore more clearly integrated with rooms than either Kiva A or Kiva C. The usual Pueblo II pattern is reflected in Kiva A's propinquity to the rooms, clearly in front of them but not integrated into the room block. Kiva C presents a somewhat anomalous situation that will be discussed later.

Kiva B, from its probable dating, was a Component D occupation feature. The later wall over Kiva B is considered a Component E feature, probably constructed after 1130 but before 1150.

As discussed previously under Rooms 14, 19, and 20, Kiva B was likely a focal point for an occupation unit consisting of Rooms 13, 14, 15, 19, and 20. The roof of this kiva probably served as a courtyard-working area for this assumed occupation unit. Other rooms behind the main north wall of Rooms 13 to 15 may also have been part of this unit, but their relationship to Kiva B is not as clear as the above rooms.

Kiva C

Dimensions. Kiva C is less circular than Kivas A and B. The dimensions are approximately 10.4 feet wide north-south and 11.5 feet long east-west (figs. 30 and 31). The floor lies from 6.2 to 7.3 feet below the present surface.

Fig. 30 Kiva C, looking south (top of picture). East wall of Room 20 (at right) has slumped into the kiva.

Fig. 31 Plan and sections of Kiva C. (click on image for an enlargement in a new window)

Banquette. The same style of banquette facing that was observed in the other kivas—a base of slabs intermixed and overlain with coursed masonry (fig. 32) —was present in Kiva C. No pecked-face rocks were observed in the banquette face or the pilasters. The stones were primarily scabbled, with a few being chipped-edge.

Fig. 32 Detail of Kiva C banquette face and fourth pilaster.

The banquette was 3 feet above the floor and its width averaged 1.5 feet, with a range of 1.2 to 2 feet. The banquette is wider between the third and fourth pilasters than between the first and fourth and between the second and third.

There were traces of at least three layers of plaster on the banquette face, to a maximum thickness of 0.1 foot. A relatively thick layer of yellow, sandy plaster was overlain by two thin coats of brown plaster.

There was no liner above the banquette and the native earth of the kiva excavation served as the wall above it. Originally, this may have been plastered.

No later wall was built over the kiva as in Kivas A and B. The east wall of Room 20 skirted the kiva on the west side, but it was probably constructed over part of the kiva roof and consequently fell in after the roof collapsed, as shown in figure 30.

Niche. Kiva C had one niche, 0.3 foot square, which extended about 0.7 foot into the north portion of the banquette face on the kiva axis. The inside top of the niche was 0.6 foot below the top of the banquette. In the niche were a Type 2B worked sherd, a Mancos Black-on-white jar sherd, and two unmodified cores.

Ventilator. The ventilator had a floor-level tunnel and probably a masonry-lined vertical shaft. It appears that the shaft was constructed through an old cist which, in turn, cut through an older firepit. Because of the soft fill in the ventilator area, most of the stones that lined the shaft had collapsed and little remained of the original shaft lining. The ventilator tunnel did not appear to have been lined behind the banquette face. The native earth provided the walls, ceiling, and floor of the tunnel.

Pilasters. All of the Kiva C pilasters flared from front to rear and all were set back from the edge of the banquette an average of 0.1 foot. They were unlike Kiva B pilasters, however, for they were wider and lacked pecked face masonry. In fact, the fourth pilaster was so wide that it had to be shaped inward to conform to the curve of the banquette. The masonry of the pilasters was scabbled and chinked with small stones or spalls. Kiva C also differed from the other two kivas in the absence of sherd-chinking in the pilasters and the banquette face.

Floor. The floor in Kiva C was basin-shaped and slightly lower in its northern part. Approximately 0.2 foot of adobe applied to the sterile earth formed the floor. There were no subfloor features.

The hearth was a circular pit about 2 feet in diameter and 0.6 foot deep, slightly deeper on the south side. It had three sandstone concretions or "pot supports" mudded in the sides (fig. 140), and was filled with ash.

A sipapu was located on the kiva axis about midway between the hearth and the north banquette face. The cylindrical hole measured 0.4 foot in diameter and 0.7 foot in depth.

A small cist covered by a sandstone slab (fig. 136b) was next to the banquette, beneath the first pilaster. It is a cylindrical hole 0.8 foot in diameter and 0.85 foot deep. Nothing was found inside it.

No deflector was found. Three irregularly shaped slabs lay on the floor between the ventilator and the hearth, but their size and arrangement precluded their use as either ventilator covers or deflectors. Probably they were part of the collapsed roof.

Entry. As in the other kivas, no evidence of an entry was found. It is assumed that it was a hatch directly above the hearth.

Other features. None.

Artifacts on floor. The plan of Kiva C (fig. 31) shows the position of the various stone, bone, and pottery artifacts in place on the floor and on the banquette.

Pottery on the floor included: a Mancos Black-on-white bowl (fig. 56f), on the west side of the kiva near the cist; a partially restorable Mancos Black-on-white jar (fig. 57d), in the same area; a Type 10 worked sherd (fig. 83a), on the east side of the floor; and a Type 7 worked sherd (fig. 83c), on the south side, to the west of the ventilator. Sherds of the Mancos Black-on-white bowl were also found on the floor of Kiva B.

Stone artifacts on the floor were: a Type 2A mano (fig. 101b) on the east side of the floor below pilaster 4; a Type 1 unspecialized milling stone with a red paint stain on one grinding surface (fig. 103d) on the west side of the floor; a Type 2 unspecialized milling stone (fig. 104b) next to the west side of the hearth; a Type 4 unspecialized milling stone (fig. 104h) to the east of the sipapu; a "jar lid" (fig. 136g) to the northeast of the sipapu, which may have been used to cover this feature; the sandstone slab covering the cist, as mentioned previously; a handstone (fig. 105e) on the southeast portion of the floor beneath the fourth pilaster; three Type 2 hammerstones in the same area as the handstone; a utilized flake (fig. 126a) in the area of the Mancos Black-on-white bowl and jar; an unmodified flake in the same area; and the three pot supports already mentioned.

A bone artifact made of a mule deer rib (fig. 156b) was on the floor east of the hearth. It is possible that it was used as a knife or scraper.

Two artifacts were on the banquette: a plain-faced metate (fig. 97h) on the southern portion over the ventilator tunnel, and about half of a Type 1A mano on the east portion of the banquette next to the fourth pilaster.

Fill. We discovered no evidence of stratigraphy. About 4.5 feet below the surface, next to the second pilaster, we found the jumbled remains of a slab hearth and scattered ash that indicated the hearth was probably atop the roof and became part of the kiva fill when the roof collapsed.

Also in the fill was a partially restorable Mancos Black-on-white ladle (fig. 59d); a partially restorable Cortez Black-on-white bowl (fig. 46b); and a large Mancos Black-on-white bowl sherd that was probably used to mix pigment since the central area of the sherd was coated with a fugitive red paint.

It is interesting to note, though its significance is uncertain, that the fill of Kiva C produced by far the greatest number of worked sherds from a single provenience at Big Juniper House (table 6). Pottery sherds and stone and bone artifacts are listed in tables 3, 8, 9, and 10.

Dates. No wood or charcoal was recovered. Other evidence points to Kiva C as a probable Component D feature, or possibly a late Component C or an early Component D feature. The architectural style indicates it was built before Kiva B and later than the first construction in Kiva A, but earlier than the second construction or remodeling of Kiva A. The flared pilasters set back from the edge of the banquette are later features than Kiva A. However, the lack of pecked-face masonry and a masonry liner above the banquette of Kiva C are early features that indicate a later construction date for Kiva B and a later date for the construction of the two narrow pilasters with pecked-face stones in Kiva A.

If, as seems likely, Kiva C was constructed in late Component C times, it was built in the late 1000's. We have already seen that it continued to be used during the time the other kivas were in use and that it was probably abandoned at about the same time as the other kivas, or around 1130.

Remarks. Unlike Kivas A and B, Kiva C shows no clear relationship to surface rooms, with the possible exception of Room 16. Although the area in the Kiva C locale was stripped, there does not seem to be more than one room that would have been associated with the kiva. Kiva C has the appearance of an "outlier," perhaps used by the people living either in rooms north of the main wall of Rooms 13 to 16 or in the other two occupation units.