i spk Hits * a Pua ae _ SMITHSON IAN. INSTITUTION - BUREAU ae AMERICAN ETHNOLOGY ees _ BULLETIN 96 2 : tas Finis RUINS IN THE PIEDRA DISTRICT _ SOUTHWESTERN COLORADO By FRANK H, Hi. ROBERTS, Jr Fe SECs rR ee HAR ; 3) ) en ss if> a aiid SMITHSONIAN INSTITUTION BUREAU OF AMERICAN ETHNOLOGY BULLETIN 96 EARLY PUEBLO RUINS IN THE PIEDRA DISTRICT, SOUTH- WESTERN COLORADO BY FRANK H. H. ROBERTS, Jr. UNITED STATES GOVERNMENT PRINTING OFFICE WASHINGTON : 1930 For sale by the Superintendent of Documents, Washington, D.C, = - - - - Price 75 cents : i - Brann a 7 v ic i ‘oD a he ms | a yee ay Ae nt | “0 nine # eM AME, A. 10 +P y ” i he : — — — cs : : per ae ’ Bei) Wi A TLE it re ” as ia i a = tek Ts: ic re oe i : - - 7 : | Sa 7 a Be 2 ah ATURE Bi Att, : " . | oa 7 sS . - — i r : >) a ee 1 : aa 7 ‘ _ 5 e i. _ 7 - (T.. : tT ae ‘ ’ De bs Dee DD Fahne oud’ won ail Imad wer thier es : . a 5 ti / : S— ‘ / x re F »| ee ligt ford { | HOUSE GROUP NO 2 , HOUSE GROUP NOJ FA = DEPRESSION : He Di = tas a “oN A = a ‘\ - HOUSE GROUP NO 6 a \ . Eee DEPRESS/ON 3 sas BURIAL > : @ MOUND ee teat Z HOUSE GROUP NO 5 REFUSE AND BURIAL MOUND ‘mM MAP OF THE A VILLAGE ON STOL.LSTEIMER MESA BUREAU OF AMERICAN ETHNOLOGY BULLETIN 96 PLATE 4 a, Rows of stones to support base logs in type A dwelling b, Excavating Houses I and J, group 2, of the A Village HOUSE REMAINS ROBERTS ] PUEBLO RUINS IN COLORADO IA clay. In general, it may be said that the dwellings were grouped in small clusters of from 3 to 15 houses, built quite close together but rarely touching. Each such unit tended to a crescent shape, due to the fact that the structures were erected around the borders of a circular depression, the remains of the pit from which the adobe used in the construction of the houses was taken. The usefulness of the depressions did not end there, however. Some of them served in the construction of underground ceremonial rooms or kivas, others functioned as reservoirs, and still others as dance courts or plazas. The location of such former village units is shown on the surface, in most instances, by a low, greatly eroded mound with the circular depression either at one side or in the center. Where the houses had been destroyed by fire the mound is usually covered with large blocks of the burned adobe, as well as with potsherds and broken stone implements. In some cases no pottery fragments are found on the surface because of the long period of time during which the remains have been exposed to the action of the elements. In others the elevation has been washed over until there is but the slightest suggestion of a depression or of rooms about the depression. In practically all cases a heavy growth of sagebrush covers the sites. pel. 2, a, 0.) Evidence secured in the excavation of the house remains shows that while all belonged to the general type which is called jacal, the various structures may be classified under three subheadings. This classification is based upon certain characteristic features in their construction. To simplify discussion the various forms will be listed as Class A, Class B, and Class C. The first comprises a group of houses which had sloping walls; the second, dwellings with perpen- dicular walls; and the third, a combination of perpendicular walled houses and small masonry structures. In all probability the forms evolved in the order listed. The C group unquestionably was later than the A form and the B style would seem to bridge the gap between them. It is a well-known fact, however, that aboriginal house forms, as well as other factors of the material culture of a people, did not always develop along what, under present conditions, would appear to be the most logical lines. Consequently, it is essential that there be some tangible evidence to show that a certain form bore the suggested relationship to the others. That it might be the natural step leading from one to another, according to the modern point of view, is not sufficient. The evidence upon which rests the belief that the house types appeared in the order named will be considered at some length at the end of the following description and discussion of the three forms, j { 22, BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 Crass A Hovsss Over half of the ruins excavated belonged to this group. Most: of them were located at the northern end of Stollsteimer Mesa along the edge of the bluff, which rises 100 feet (30.48 m.) above the Piedra River at that point. They were gathered together in what might be termed a village composed of a number of units, each of which consisted of several dwellings bordering one of the depressions mentioned in a previous paragraph. The major portion of the village is indicated on the map. (Pl. 3.) Additional units were investigated at some distance east and south along the edge of the bluff, but inasmuch as they were too far removed from the others to make their inclusion in the map practicable they have been omitted. House Group 3 was not complete and its original extent can not be given. The eastern portion of the unit had been worked out at some previous date and a portion of it had been plowed, so that it was impossible to determine the size and positions of the houses or to trace the limits of the refuse and burial mound which lay to the southeast. At five other locations on the mesa-top isolated units of the Class A group, probably constituting single family centers or one-clan vil- lages, were excavated. Inasmuch as all were typical and since the similarity between them and the units included in the village here represented was so marked, it has not been deemed necessary to print the additional plans. Indications were that a village could have consisted of a single unit or have been made up of several as in the case of the community at the northern end of the mesa. There were certain outstanding characteristics common to all of the houses, although there was some individual variation in minor features. The structures were, with a single exception, one-roomed and quadrilateral in form. Some closely approached the rectangu- lar in outline but the majority were too asymmetrical to be so desig- nated. Rarely did the corners form right angles. Most of the dwellings were erected over a shallow pit so that the floor level, which was the bottom of the pit, was somewhat below the general ground level. ‘The depth of these pits varied from house to house but all fall within a range of from 6 inches (15.24 cm.) to 2 feet (60.96 cm.), with the majority averaging 1 foot (30.48 cm.). The floor was usually hard-packed earth, although occasionally it was plastered with adobe. In not one instance was stone paving in evidence. The superstructure was supported on posts set in the floor at a distance of from 12 to 18 inches (30.48 to 45.72 em.) from the corners of the room. Some of the larger houses had two additional posts. In such instances the extra supports were placed at about the ROBERTS |] PUEBLO RUINS IN COLORADO 23 center of each side wall and at approximately the same distance from the side wall as the corner posts. The main pillars probably were crotched at their upper ends and carried stringers which in turn supported the smaller roof beams and also the slightly sloping side walls. The latter had a framework of small poles, from 2 to 4 inches (5.08 to 10.16 cm.) in diameter, placed from 4 to 6 inches (10.16 to 15.24 cm.) apart. These sloped from the walls of the pit to the stringers on the main posts. The wall poles were not em- bedded in the earth at their lower ends, but rested on the floor and were held in position against the walls of the pit by heavy logs. The method in which the latter were used was characteristic. Those at the ends of the room were lying on the floor, while the longer ones which held the framework for the side walls were placed on top of the end logs and supported at intervals along their length by large stones. In some structures only three or four stones were used while in others a whole row of them was deemed essential. (PI. 4, a.) Some of the individual variations in house construction were observed in connection with the placing of the base logs along the walls. Where the main support posts for the superstructure were only a short distance from the wall the base beams were wedged tight between them and the wall poles. When the builders had mis- calculated this space, however, they were forced to place large stones between the support posts and the base logs to hold the latter in position. The heavy roof beams were overspread with small poles, brush, and leaves. The entire wooden framework of the structure was then covered with a 6 to 8 inch (15.24 to 30.32 cm.) layer of plaster. One rather curious feature observed in connection with this plaster covering was that the adobe was applied only to the outside of the framework, where it was carefully smoothed down. ‘Traces of the builders’ hands are frequently observed on the burned clay, and the wall poles were completely hidden from view. Only such mud as worked its way through between the poles when it was being applied to the exterior appeared on the inside walls. This was smoothed down, however, so that here and there the timbers were covered with a thin coating of it, but a greater part of the poles must have been exposed. It was, perhaps, this very feature which made pos- sible such a complete burning of the buildings. Had the poles been embedded entirely in adobe, fire would have had little effect on them. The ceilings of the structures would have been quite combustible, but not the walls. Entrance to the dwelling seems, in most cases, to have been gained through small doorways in the sides of the houses. Actual $8242 °—30——3 24. BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 doorways, with two exceptions, were not present in the ruins, but the large stone slabs which were used to close them were frequently found in positions which indicated lateral openings. These stones were generally lying on top of the wall débris with one end resting either on the ground at the edge of the pit or on the charred remains of the base logs. Indications were that in a majority of cases the doorways opened toward the depression. A few of the house remains suggested that the smoke hole in the center of the room had served as an entrance. On the basis of the information gained from a careful study of the débris which filled all of the house pits it has been possible to reconstruct the dwellings as outlined in preceding paragraphs. In Yi GROUND LEVEL Yy 2 POLE AND PLASTER WALLS , H/ MAIN SUPPORT POSTS BRACE LOGS AT BASE OF WALLS i; )) MAIN ROOF BEAMS Y SMALL ROOF TIMBERS POLE BRUSH AND PLASTER ROOF 8 SMOKE HOLE 9 DOORWAY 10 STONES SUPPORTING BRACE LOG Ficure 4.—Postulated type of construction in the Class A houses order to illustrate the house type more clearly, however, the accom- panying postulated reconstruction has been prepared. (Fig. 4.) The view presented is that which would have been apparent had the back wall of one of the structures been removed. The really ques- tionable feature in the drawing is that of the doorway. Evidence obtained was not such as to definitely establish the height of the sill above the floor nor the exact dimensions of the opening. An average of the size of the stone covers found was used in determin- ing the latter, while the former is based on parts of walls in two dwellings which indicated that the opening had been from 6 to 10 inches (15.24 to 25.4 em.) above the ground level. The average size of the stone covers, which were oval in shape, suggests that the doorways ranged between 2 and 214 feet (60.96 and 76.2 cm.) in width and 8 and 31% feet (91.44 cm. and 1.0668 m.) in height. ROBERTS | PUEBLO RUINS IN COLORADO 25 While this may seem to have been small, such doorways would com- pare quite favorably in size with many in the stone structures of later periods. The general reader must bear in mind the fact that the customary practice of the Pueblo builders was to make the open- ings into their dwellings only as large as was felt to be absolutely necessary. Inasmuch as they were restricted to the use of stone slabs, skin, or textile hangmgs for doorway coverings, small openings were an advantageous feature. The method of timber arrangement at the base of the walls was indicated in so many of the pits that there can be no question as to its correctness. The roof construction was clearly shown in the re- mains of three structures—Group 1, A and F and Group 4, A—where the whole mass had dropped into the pit as a unit and could be un- covered, turned over, and fitted together so that the impressions in the plaster showed that small poles and brush had rested upon the heavy timbers whose charred remnants were lying on the floor. Fragmentary sections of the roofs of other dwellings showed that they, too, had had similar construction. The ceiling height no doubt varied, but measurements of the wall material and a few support posts which had not been completely consumed by the flames indi- cated an average “of 6 feet 6 inches (1.9812 m.). There also must have been some variation in the size of the smoke hole. Evidence in the roof plaster indicated a range between 2 feet 6 inches and 3 feet (76.2 and 91.44 cm.) for the sides of the opening. In many of the structures the edges of the smoke hole had been reinforced with a coping of stones placed around it on the roof. There was practically nothing in the way of interior furnishings in the houses. Some of them had fire pits in the floor near the center of the room. In others there was no special provision for the fire, which seemed to have been built directly on the floor under the smoke hole. Where pits were in evidence they were either roughly circular in form or tended to the rectangular. Sometimes they were lined with stones and occasionally they were simply plastered. ‘The circular forms ranged between 1 foot 6 inches and 2 feet (45.72 and 60.96 cm.) in diameter. Measurements for the other forms showed about the same average size. Rectangular pits were from 1 foot to 1 toot 6 inches (30.48 to 45.72 cm.) in width and 2 feet to 2 feet 6 inches (60.96 to 76.2 cm.) in length. It is interesting to note in connection with this reference to the fire pits in the houses that many had been built in the open, but in locations where the houses would more or less serve as windbreaks. This is a very common feature in south- western ruins and indicates that when the weather permitted con- siderable cooking was done outside of the houses. 26 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 Not a single structure of the Class A group had a corner storage bin. There were no storage holes in the floor, which is in contrast to many dwellings of earlier and later periods, where the feature is not at all uncommon. Their absence may be accounted for in some meas- ure by the fact that the people seem to have made use of the space between the side wall base logs and the floor for storage purposes. Many of the bowls and smaller jars which were recovered from the ruins were found sitting on the floor against the wall of the pit in positions which indicated that they had been placed under the raised side logs. No provision was made, apparently, for mealing bins. All of the grinding stones or metates found in the remains of the dwellings were on the floor along one of the side or end walls, and in not a single instance was there the slightest indication of an inclosing bin of any of the forms so frequently reported by investigators working in other types of ruins. There were a few of the dwellings in the village at the northern end of the mesa which call for special consideration, The major- ity of them conformed so closely to the average house type already described that a detailed discussion and description of each would be superfluous. The combination of structures at the northern end of house group 2 (pl. 3) was so unique, however, that some addi- tional paragraphs may be profitably devoted to it. When work was first started on the mound which covered the débris filled pits (pl. 4, 0) it was thought that a connected group of four houses would be uncovered, but as the excavations progressed it became apparent that such was not the case, and that there had been in reality but two houses. Each of these was different, however, from all of the others investigated in that it had a shed at one end. The two houses, I and J (fig. 5), were of the characteristic Class A style of construction. They differed to a certain extent, however, in that they had doorways in one of the end walls. These apertures were so placed that they were directly opposite, and formed in a broad sense a single opening between the two rooms. Contrary to what seems to have been the general plan in doorway construction, their sills were on the floor level. No cover slabs were found, and it is suggested that they were left open so that the structures func- tioned as a two-roomed house rather than as distinct and separate dwellings. This raises a question which, unfortunately, can only be answered by speculation, Was any effort made to cover the openings so that a short passage resulted or were they left open to the weather? If the walls of the houses had been straight instead of slanting, no such provision would have been necessary. As it was, however, the two doorways barely touched at one side, the ROBERTS | PUEBLO RUINS IN COLORADO 20 STORAGE Pir ens TRAITUABITS tS anurans ea | ty oll i N a 1s v4 a I z | % | | we “iy ww hu May, | Pcad Mw udu iiyl muon ow) ' o 8 40FT | eee er earenren) ass _ SECTION ON AN EAST*WEST LINE THROUGH STRUCTURES / \Y a j wry = - PRESENT SURFACE-~ -- Y | . j F Ms igpgere moe j Fe 7 / Z y, _ . H]/ Yj SECTION ON NORTH-SOUTH LINE Fieurp 5.—Detail ground plan of portion of House Group 2 in the A Village 28 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 south, and then only at the bottom of the wall. Consequently, un- less some measures were taken to roof over the openings or to build a short passage, the inhabitants must have been subjected to no little inconvenience, if not actual discomfort. Every time it rained water would have run down the slopes of the walls, through the doorways, and into the houses. This would have been particularly troublesome for the occupants of I, as its floor was several inches lower than that of J. It seems logical to suppose that if the door- ways were left exposed the people would have made some efforts to prevent an inflow of water; that they would have constructed the usual type of entrance with raised sill. Since they did not it seems plausible to believe that there was some sort of covering, possibly one made of poles, brush, and plaster, which connected the two. Unfortunately, at the time when the particular débris which might have contained the desired evidence was removed, the writer was engaged in the problem of the storage pit in the floor of shed H. The workmen did not call attention to the doorways between the two structures until it was too late to make a careful study of the débris at that point. The fact that no evidence of such construction was found might be taken arbitrarily to show that none existed and the question thus summarily dismissed. Considering all things, how- ever, it seems more in keeping with other features, which indicate that the people had a fairly well-developed discernment in house building, to suggest that some means must have been taken to in- close the doorways, although the manner in which it was done can not be known. ‘The only dimensions which could be secured for these openings were the width of each and the wall thickness at that part of each house. ‘There was only a small variation in door- yay width. ‘The opening in I measured 1 foot 814 inches (52.07 cm.), while that in J was 1 foot 914 inches (54.61 em.). The aver- age wall thickness at the sides of the doorway in I was 1 foot (30.48 cm.), while in J the average was 1024 inches (27.3 cm.). House I had another doorway, in the south wall, which opened toward the circular depression of the house Group 2 unit. House J had no additional opening, at least no traces of one were found, and its only means of access to the exterior seems to have been through house I. The suggestion that the two structures might have served as a two-roomed dwelling instead of being separate domiciles devel- oped out of this feature. The second doorway in I was of the char- acteristic form indicated in the other house remains throughout the village. Its height could not be determined, but the width at the sill was 1 foot 6 inches (45.72 cm.). Both structures had interior fire pits. They were of the rectangu- lar type and had been lined with small stone slabs. Neither was ROBERTS | PUEBLO RUINS IN COLORADO 29 very large. That in I measured 1 foot by 2 feet (30.48 by 60.96 cm.) and was 1 foot (30.48 cm.) in depth. The one in J was 1 foot 6 inches (45.72 em.) by 1 foot (30.48 cm.), with a total average depth of 914 inches (24.13 cm.). Sheds did not play a very prominent part in the building activities of the dwellers on Stollsteimer Mesa because the only examples of such structures found during the course of the summer’s work were the two associated with the I and J house combination. They were of comparatively simple construction and indications were that they had not been erected until after the completion of the houses. Three upright posts supported their roofs at the outer ends, while the inner probably rested on the house tops. There were no traces of support posts along the ends of the houses. The walls were slightly different in form from those found in the dwellings. This was no doubt due to the fact that the floors were on the ground level and not below it. (Pl. 5, a.) The wall poles did not have heavy base logs to hold them in position. Instead, their lower ends were embedded in the earth. Since the structures were lean-tos, it was only necessary to erect two walls for each, the ends of the houses furnishing the third, and the fourth, the southern side, was left open. It is very probable that a greater part of the fair weather activity around the dwellings took place in these lean-tos or sheds. Judging from the practices of the present-day Indians in the Southwest, the families probably lived during the entire summer season in the sheds, retiring to the houses only when cold weather set in. Greatest interest centers in the shed marked “H” (fig. 5) because of the fact that within its confines a large storage pit had been lo- cated. This pit had been dug into the soil of the mesa top. Its walls were carefully plastered with adobe mud. As will be seen from the section of the house group (fig. 5), the pit was somewhat similar in shape to some of the large pottery jars which the people made. The narrow necklike opening at the top had been inclosed by a ring of stones placed, on the floor (Pl 5, 6.) ) there was no indication of communicating doorways between the rooms. A person wishing to go from B to C would have had to go outside to doso. What was of even greater interest, though, was the suggestion that A had had no opening in its walls, but that the only entrance was in the roof. It is possible that the room could be entered only through the passage which led from the kiva, but this does not seem likely. B and C were evidently living rooms, because each had a fire pit. That in B was rectangular in shape while the one in C was of the circular form. Both were simple pits in the floor. Their walls had not been lined with stones but merely covered with plaster. Further evidence of the secular character of the rooms was observed in the fact that B contained seven metates or milling stones and three were found in C. In the fire pit of C were a number of slightly charred deer bones. Another deer bone was found in one corner. There is a deep groove down one side of it which shows that some member of the household had intended to cut it into two pieces, probably to make some implement needed in their daily life, but the bone was cast aside before the work was completed. BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 NS b> 85 OR a SS van es Cin 39 KOPF; 0) thas SN SSS S SAS S Ss SQ Ss SETS QO =) Ray END AS; ne ) Lt sity Sere AL | es age PASSAGE i r/ VAS . UM YY Yf 1, Main support posts in Fieurn 8.—House and kiva combination in B Village. kiva; 2, Fire pit; 3, Ventilator passage; 4, Ventilator opening on ground level The remains of the kiva, or underground ceremonial chamber, were discovered in an unexpected way. During the removal of the débris from the interior of the shallow pit which formed the lower This portion of room A a hole was found in the southeast corner. BUREAU OF AMERICAN ETHNOLOGY BULLETIN 96 PLATE 7 a, Charred posts in adobe wall, illustrating method of construction b, Doorway in wall between rooms B and B~-1 of the B Village HOUSE REMAINS BUREAU OF AMERICAN ETHNOLOGY BULLETIN 96 PLATE 8 a, During excavation. Workman standing on floor of passage from House A b, Steps at end of passage from House A. Same view as a after débris was removed from pit B VILLAGE KIVA BUREAU OF AMERICAN ETHNOLOGY BULLETIN 96 PLATE 9 a, Ventilator opening at southeast side. Charred butts of main support posts still in floor b, Fire pit in kiva. Portions of original wall plaster may be seen in background B VILLAGE KIVA TIVM LNOYS NI Gasn SMOILS ONIOYOANIAY AO SGN Y3addN ONIMOHS YOLVTIILNAA AO FYNLVSA IWNOILONYLSNOD Ol ALWTd 96 NILATIAG ADOTONHLA NVOIMAWY JO OVaeNE ROBERTS] PUEBLO RUINS IN COLORADO 43 was at first thought to be the remains of a storage pit in the floor but, as work progressed, it soon developed that it was the entrance to what had formerly been an underground passage. The latter was followed until it ended near the edge of a circular depression. The débris at that point indicated that there had been some sort of a room into which the passage had given access. The débris consisted in large measure of adobe earth which had been washed in, settled, and packed by water until it was incredibly hard. Its removal was a long and laborious task. Tracing the walls of the inclosure at the outer end of the passageway proved the most difficult part of the entire summer’s investigations. (Pl. 8, a.) Throughout most of the excavation work the only indication of a possible wall was a slight difference in the color of the earth. The fill contained a small amount of charcoal and to find: the walls it was necessary to follow the hazy line separating the ordinary yellow earth from that contain- ing an occasional bit of burned wood. Not until a considerable part of the débris had been removed was actual wall plaster uncovered. From that time on the work was comparatively easy. At its com- pletion there stood exposed to view all that remained of an early form of kiva. (PI. 9, a, 0.) The pit portion cf the underground chamber had been roughly circular inform. (Fig.8.) The walls were uneven and at the north- east side there was a considerable bulge. On the whole, however, it may be said that there was a conscious effort to make it circular, a form which is characteristic of the ceremonial rooms of most of the San Juan peoples in the Pueblo periods. Some of the late pre- historic ruins in the Kayenta district have rectangular ones, a form which is to be seen in modern Hopi and Zufi villages. Adobe plas- ter placed on the earth walls and floor of the pit had been the only finish applied to that part of the chamber. The superstructure had been supported on four posts placed in the floor at some distance from the wall. They had been securely embedded in the earth at an average depth of 3 feet (91.44 cm.). The butts of the posts were wedged tight by cobblestones which had been tamped into the dirt which filled the holes around them. The charred ends of the uprights were still in position when the débris was removed from the floor of the kiva. They indicated an average diameter of 10 inches (25.4 cm.). Three of them had been set approximately 2 feet (60.96 cm.) from the wall, while the fourth was 4 feet (1.2192 m.). This discrepancy may have been due to the fourth one having been set into the room in such a position as not to interfere with the entrance to the passage leading from room A. Fortunately, the entire superstructure had not been consumed by the fire which destroyed the kiva. There was sufficient material in the débris on the floor to give a fairly accurate picture of the method 44 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 used in its construction. On top of the upright posts, which were probably forked, two heavy stringers had been placed. These sup- ported a series of four crossbeams upon which the flat portion of the roof rested. The sloping portion, around the sides, was built of smaller poles which slanted from the main framework to a small bench at the top of the kiva wall. (Fig. 9.) The framework of poles was then covered with quantities of brush and leaves upon which earth was heaped. es Sy, % My % 4 z DEPRESSION = e Wi = SN % S BS z s S Zz s “ty, eo = SZ Sa tng, wr S € ao Ze “4 “AHH uusyyuresitse S = f=" z= REFU. > 5, ¢ = = f MOUND” & oO s 40FT Saas aa a SS Ro woes saseel ne = SO am seer be bogs : : Y ff if Yj // BS == > monial form frequently found associated with one-clan houses of the Pueblo II period in the region farther west, the McElmo dis- trict (fig. 1), but clearly a secular building. Mr. Jeancon explains the structure as having been a lookout tower, occupied in turn by in- dividuals or families to whom the duties of watchers had been as- signed. It was located at the north end of a jacal village on a high 57 58 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 point overlooking the valley. The position which it occupied at the edge of a drop in the mesa top had necessitated, as in the case of house D in the C-3 unit, the building of some sort of terrace. There was nothing of interest in house F beyond the fact that such a structure had once existed. For some reason there was no fire pit in the room and no indications of an exterior provision for a fire. Near the center of the floor, where such a pit would be, there was a fairly large burned area which suggested that fires had been built directly on the floor, a trait suggested in a few others as well. Houses A and D had good fire pits. That in A was circular in form and had consisted of a simple pit, the walls of which were lined with adobe plaster. The D pit was also circular in form but had been finished with stone slabs. Both were of average size, their diameters being approximately 2 feet 6 inches (76.2 cm.). The main difference between the stone structure in this group and those of the C-1 and C-2 units was in the fact that the two rooms had not been erected as a single building, but that room C was subsequent to room B. The east and west walls of C abutted against the north wall of B, showing distinctly that they were of later construction. The joining of the west wall of C and the north- west corner of B had been well done, but the workmanship on the east wall was not so good. There was a miscalculation on the part of the masons and the end of the east wall almost missed the north- east corner of B. There were many things about the C-3 unit which suggested that it may have represented a stage in the transition between the C and B classes. There is no way of proving such to be the case, but the presence of a depressed floor, the only one in a C group house; the even greater crudity in the stone work, with the second room appar- ently an afterthought; the lack of connection between jacal and stone construction; suggest a step not quite as advanced as that of the first two, C-1 and C-2, previously discussed. The circular depression of the unit presented the same features mentioned in connection with those of C-1 and C-2. Over half of the débris which filled it was completely removed in order to make doubly certain that there could be no mistake in the conclusion that it had served as a reservoir. The face of the dirt fill of the unex- cavated portion showed beyond doubt that the same conditions had prevailed in this unit as in the others. One point observed in the study of all of the depressions in the C units was that there was no refuse in the pits. The only débris, above the silt and sand strata at the bottom, was material which had washed in after the destruction of the houses. This is a clear indication that the people had definite ROBERTS ] PUEBLO RUINS IN COLORADO 59 use for the pits. Otherwise they undoubtedly would have been filled with house refuse. In the C-3 group, for example, it would have been much easier to have dumped the waste matter from the dwell- ings into the pit than to have deposited it at the side farthest away from the domiciles. Previous to these investigations the writer had always been skeptical over the suggestions that some of the circular depressions had served as reservoirs, but after the evidence obtained through hard and tedious excavation it does not seem possible to escape such a conclusion. The refuse mound for this unit was located at the southeast side of the depression on the slope of a small ravine. It had the char- acteristics noted in the other mounds, even to the lack of burials. Although no burials were discovered during the investigations at C-8, one skeleton was found. The individual had not been buried but had been burned in house A when it was destroyed. The bones were lying on the floor of the room, near the northern wall, in posi- tions indicating considerable distortion of the body. Most of the small bones had been entirely consumed by the flames and the skull was in a very poor state of preservation. It was not possible to tell whether the individual, an adult, had lost his life in the fire, whether he had been wounded and thus trapped because of his inability to escape, or whether death had preceded the destruction of the house. Nothing is known, of course, of the practices of the Indians of that period and it has been suggested that on occasions, like the Navaho, they burned the house in which a person had died. In view of the elaborate burial customs of the Pueblos and considering the attitude of the modern descendants of the old people it does not seem likely that they followed that practice. The probabilities are that the per- son perished as the result of some factor which was closely associated with the destruction of the dwelling, but how or under what condi- tions it is impossible to tell. Hovusr Tyrer SuMMARY AND CONCLUSIONS The characteristic features of the house forms of the three groups discussed in the preceding pages may be briefly summarized. Features common to all were: Jacal walls; flat roofs with smoke holes near the center; lateral doorways; the grouping of houses around the pit from which the earth used in the plaster construction was taken, such combinations being considered to represent a unit, possibly a single clan group. Some of the dwellings had interior fire pits, while others did not. Class A dwellings were, with one exception, one-roomed; quadri- lateral in form; had a pole and brush superstructure erected over a 60 BUREAU OF AMERICAN ETHNOLOGY [BULL, 96 pit, the bottom of which was the floor of the house. The roof and sloping side walls were supported by a framework of logs which rested on upright posts set in the floor some distance from the cor- ners. The wall poles were held in position against the sides of the floor depression by heavy base logs. The wooden framework was covered with a thick layer of adobe mud applied to the exterior. The inside walls were not plastered and the poles were covered only by such of the adobe as worked its way through between the timbers and then was smoothed down. The houses ranged from 4 to 12 in each unit. Class B dwellings approached more closely to the rectangular form and had perpendicular walls in which the main supporting framework was incorporated. The base ends of the wall poles were placed in shallow, narrow trenches, where they were held in position either by quantities of tamped adobe clay or wedged-in stones. The poles were completely incased in the mud plaster which consti- tuted a greater part of the wall material. The heavy roof timbers crossed the short way of the room instead of paralleling the long walls. The pit portion of the structure was much shallower, and as a consequence the floor was not as far below the ground level as in the A domiciles. The class C houses were closely allied to those of the B group, their chief and characteristic difference being the lack of a depressed fioor or pit. Associated with the C houses were small two-room stone structures which seem to have been granaries. Four of the five units excavated showed a combination of stone and jacal buildings, one abutting another, while in the fifth they stood as separate and dis- tinct structures. ‘The masonry of the stone buildings was extremely crude. No attempt was made to dress the stones, and discrepancies in size and shape were compensated for by the use of large quanti-. ties of mud mortar. There were fewer dwellings per unit in this group than in the preceding ones and several units were not com- bined to form a single village. In the introduction to the subject of house remains it was stated that the three forms were thought to have developed in the order listed. This conclusion is based on several distinct bits of evidence. That the C form was subsequent to the A was definitely shown in the superposition of remains at house Group 8 in the A village on the mesa top. This was the only actual stratigraphic evidence found during the 1928 work, but Mr. Jeancon noted a similar condition in his 1925 investigations, previously referred to. His notes on the work, kindly loaned for comparative purposes, give clear indications that the small houses of stone which he excavated rested upon the remains of jacal structures. ‘There was still another important factor ROBERTS ] PUEBLO RUINS IN COLORADO 61 in the 1928 results, however, which shows a later development for the C group. This was in connection with pottery forms.’ Inasmuch as ceramics are discussed at some length in a following section of this paper, an extensive consideration of them is not germane to the present problem, but to make clear the chronological sequence of the two houses it 1s necessary to mention briefly one form, the cooking jars. One of the outstanding characteristics of the latter in Pueblo II and following periods was the indented corrugation of their ex- terior surfaces. The comparable pots in the Pueblo I period had no indentations, merely plain coils around their necks. That the latter preceded the former has been shown by stratigraphic evidence at many places in the Southwest. The same is true for the question of the relationship of the pottery forms to the periods named. The factor of significance in the Piedra jacal ruins was that the only potsherds from vessels of the indented corrugated ware were found in the remains of C houses. Not a single fragment of the type was observed in the vast amount of pottery and potsherds from the A sites. With two such conclusive bits of evidence, superposition of house remains and pottery specimens with an established sequence, the later horizon for the Class C dwellings is clearly demonstrated. The problem of the relationship existing between the B houses and the other forms is somewhat more difficult. It can be answered only by a careful interpretation of the conditions under which they were found. There was no clear-cut evidence of the kind just dis- cussed for the A and C groups. Investigation showed that two of the A class houses were associated and contemporary with the B structures in the village at the foot of the southern end of Stoll- steimer Mesa. None of the B group was found on the mesa top dur- ing the 1928 work, but the remains of three were uncovered in the preliminary investigations of 1923 and 1924.4° A thorough inspec- tion of their remains showed that they unquestionably were contem- porary with the A structures comprising the major part of the settlement. The dwellings of the C group were found to be similar to those of the B class; in fact, with some minor modifications, were so similar to them that they might be considered as a single form were it not for the association of the C form with stone structures. The situa- tion found to exist, then, was that A and B forms occurred together, while B and C were closely related. A was not found in combination with C but was shown to have preceded it. Furthermore, none of the B houses contained any of the indented corrugated potsherds, al- though they did yield ax heads, a form of implement associated with C but not found in any of the A structures. On such grounds it 46 Roberts, 1925, p. 41, 62 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 seems quite evident that the B class occupied the intermediate posi- tion and forms A, B, and C developed in the order named. On the basis of the foregoing it is possible to trace quite conclu- sively the growth and change in house types in the Piedra district. The earliest form of which there is evidence, the A, shows some re- semblances to late Basket Maker houses in the sloping side walls, flat roof, and interior support posts, but its general features are suffi- cient to set it apart as representative of a different horizon. The B and C classes are distinctly different from the dwellings of the former culture and may be considered as showing definite Pueblo characteristics. Although the house forms discussed would tend to show that there was an abrupt change from support posts set out from the wall to those incorporated in it, such was not the case. During the work of 1922, and also Mr. Jeancon’s 1925 investigations, rectangular struc- tures were found in which the main roof posts had been set against but not included in the walls.** The development of the perpendicular walled building was a fea- ture of decided importance because it made possible the grouping of houses together to form a single building. ‘This permitted consider- able economy in material and labor because one wall could serve for two buildings. In this connection there is a point of considerable significance which has already been mentioned but which should again be stressed, namely, that in this district, at least, the rectangu- lar form of structure was developed first and then the step toward consolidating a number of them into a single building was taken. A similar feature was observed in the late Basket Maker village in the Chaco Canyon in which connection it was pointed out that con- trary to common belief rectangular structures in the Chaco pre- ceded and made possible the compact house type instead of resulting | from efforts to join circular or oval dwellings together in a single structure.** It is interesting to note that the one-roomed dwellings prevailed for some time after the feasibility of contiguous grouping was discovered. No doubt the tendency to conservatism was just as operative in the acceptance of the development of such a feature as it had been in the shift from A to B construction methods. Just how the building of the small masonry structures began is not known. Late Basket Maker ruins in some localities show occasional bits of wall with horizontally laid stones and it is possible that the idea survived from that period. On the other hand, it may have been that the occasional use of large bowlders here and there in a wall which was predominantly of adobe may have suggested that more stone and less mud would make a good wall. Regardless of 47 Jeancon and Roberts, 1923, p. 31. 48 Roberts, 1929, p. 147. ROBERTS ] PUEBLO RUINS IN COLORADO 63 how it developed, in the course of time there appeared a combina- tion of jacal houses with granaries of stone construction, and even- tually, although the present study does not show it, the jacal struc- tures disappeared, the granaries were enlarged, more rooms were added, and dwellings of stone became the rule. The unit-type houses on the lower Chimney Rock Mesa represent the latter stage. There seems to be considerable probability that the jacal structures per- sisted longer in the Piedra district than in some other sections of the Southwest where they are found to have existed. This was possibly due to the fact that there was an abundance of timber for use in such structures as well as to the difficulty of obtaining suitable build- ing stone on the lower benches along the river. On the upper mesas where there was an outcropping of rock it would have been a com- paratively simple task to secure stone for use in walls. There seems to be no question but what the house-group units with their row of one-roomed dwellings grouped about a depression defi- nitely foreshadowed the characteristic unit type of the Pueblo IT period. All that was necessary to the development of the latter was the straightening of the house row and the pushing of the structures together, so to speak. The elimination of the crescent shape, which was due to the erection of houses around the rim of the pit where the adobe was puddled, was a natural development in the shift to stone construction and the building of contiguous rooms. A straight wall of stone, comparatively speaking, would be easier to lay than a curved one, hence its development. The B village went even further and furnished the precursor of the underground passage from house to kiva, a feature which is such an integral part of the Pueblo II unit-house complex. The depression of the house-group units became, in the later stone structures, the kiva, another feature which was being developed in the stage represented along the Piedra. The depressions, it will be recalled, were found to have been used for reservoirs, dance courts, and in two instances for kivas. The apparent lack of kivas in so many of the units raises an inter- esting question. Was the feature fully established at that time or was it still in the process of development? Because of the many indications of the instability of the period, that there was a plasticity in numerous parts of the culture, it seems logical to suppose that the specialized ceremonial room was still in an embryonic state. For years it has been considered axiomatic that the absence of a kiva in a Pueblo ruin indicated a temporary habitation, perhaps a farming settlement. That such was true in the Piedra villages seems improba- ble. The evidences of continued occupation of the sites are too marked to warrant the belief that they were but temporary communi- ties. The fact that in some instances a kiva was present also argues 64 BUREAU OF AMERICAN ETHNOLOGY (BULL. 96 against such conclusions. Furthermore, there are no sites in the region which could be construed as being the remains of the perma- nent villages to which the residents withdrew at the end of the harvest season. To postulate the location of permanent dwellings 60 or 70 miles (96.561 or 112.654 k.) down the river is going too far afield. The presence of storage facilities shows that the harvest was not carried away, and it does not seem probable that they left their food supply behind at a time when it would be most needed. It seems more in keeping with the condition of the period to believe that the kiva had not yet become a component part of the Pueblo com- plex; that it was still in its incipient stage. The influx of new peoples, the commingling of traits, no doubt had a marked effect on this as well as other features of the culture. The development of the kiva was considered at some length in the report on the Chaco Canyon Late Basket Maker village. The discussion presented the suggestion that the lesser ceremonial observ- ances of that period were held in the dwellings and that the small- clan kivas of the later Pueblo periods represented in a somewhat modified form the old, original dwelling which survived because of the tenacity with which primitive peoples cling to matters pertaining to their religious customs.*® This would indicate a decided Basket Maker infiuence on the religious practices of the later peoples, but one which probably did not assert itself immediately in the Pueblo I period. It possibly remained dormant until conditions became more stabilized. Then because of the vast difference between the type of dwelling in vogue and that which had been the scene of the simple family ceremonies in the older horizon, it was found neces- sary to construct chambers embodying the essential features of the old domiciles. Perhaps it was at this point that the custom of having but one ceremonial chamber to a group, the clan, appeared. Due to the disturbed state of the period it is quite reasonable to sup- pose that some of the clan units had structures for ceremonial ob- servances while others did not. Possibly one to a village was thought sufficient, then as the culture unfolded old beliefs began to assert themselves until the ultimate end was that each clan con- structed and maintained its own ceremonial chamber. Under such circumstances an occasional unit without a kiva would not be sur- prising, as a matter of fact might well be expected. The kiva in the B village showed in its type of roof construction a distinct survival of the superstructure of the Basket Maker III houses. It was so like the coverings of the Chaco Canyon houses of that period that it might just as well have been found there as * Roberts, 1929, pp. 81-90. ROBERTS] PUEBLO RUINS IN COLORADO 65 along the Piedra. This feature was somewhat disappointing when first uncovered, because it had been hoped that some clue would be found to suggest the origin of the cribbed type of kiva roof so char- acteristic of later Pueblo periods, but its demonstration of the sur- vival of the preceding period’s form of construction is of value. It shows that the cribbed roof was not an immediate development of the Pueblo peoples. The designations given the various forms of house remains in this paper are different from those suggested by Mr. Jeancon in his first report on the work of the Colorado institutions. For the benefit of the student and those who are familiar with the earlier paper, it has been deemed advisable to correlate the two. Mr. Jeancon’s classifi- cation was only a tentative one, based on rather meager information for the older forms, but with the knowledge gained in the recent in- vestigations it may be clarified and fitted into the chronological sequence for southwestern sites which is at present followed by the archeologists working in the area. Mr. Jeancon’s classification was as follows: “First. Pit houses: These are houses which were semisubterranean with the plastering appled to the native earth and with a double slant roof. There are thousands of this type scattered over Archu- leta County. “Second. Pit houses with cobblestone walls. While these are not in a class by themselves they plainly indicate a step in the sequence of house building. “Third. Pit houses with cobblestone walls and the paving of floors with slabs and cobbles. The first use of horizontal slabs laid above the cobblestone walls. Here is probably where the first type of flat roof occurred. “Fourth. Single rooms, all above the ground, without any excava- tion, using a few cobbles for the foundation and horizontally laid slabs of stone above these for the remainder of the walls. “Fifth. Small groups of from three to six or eight rooms built in a similar manner to those of group 4. “Sixth. The large pueblo or concentration of several houses of the type of group 5.” °° The first type undoubtedly corresponds to the A dwellings. It seems unlikely that any of the structures had a double slant roof. The traces of sloping poles found in the earlier work were no doubt the remains of the slanting walls from a superstructure similar to that of the A houses. The remains examined in 1921 were of houses which had been more nearly consumed by the fire which destroyed 50 Jeancon, 1922, pp. 5-6. 66 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 them than had those of the 1928 investigations. Hence it is highly probable that all indications of the flat portion of the roof had been obliterated. The other features of the first type houses correspond and agree with those of the A class. The second, third, and fourth stages represent variations of the stone structures of the C class. The writer reviewed the situation in such remains during the 1928 season and found that in practically every case they had been associated with some form of jacal construction. The fifth type is that of the unit houses or one-clan structures which are so characteristic of the main phase of the Pueblo II period. The sixth type, the large pueblo, is the typical Pueblo III vil- lage or large community house. No example of the B houses was found during the earlier investi- gations, but such a condition is not surprising in view of the fact that only a small amount of work was done in the so-called pit house or jacal ruins. The A and B dwellings are definitely Pueblo I in their horizon. The class C structures probably mark the end of the first period and the beginning of Pueblo II. Consequently, as stated in the introduction, the Piedra district clearly shows that it was occupied during Pueblo I, I, and III. DISTRIBUTION OF HOUSE FORMS The remains of jacal structures comparable to those described in preceding paragraphs are to be found in widely scattered sections of the Southwest. Those in closest proximity to the Piedra district might be considered, in general, as belonging to the same regional group. On both sides of the San Juan River, east from the con- fluence with the Piedra, for a distance of 25 miles (40.235 k.), are almost countless numbers of mounds and depressions indicating some form of the jacal unit cluster. Along each tributary canyon for a distance of several miles from the main stream are additional house sites belonging in the same category. The region east and west of the San Juan extending down into New Mexico, the Frances, Burns, and Governador Canyons, contains similar remains. Although there has been no excavation in the vast majority of them, surface indica- tions are sufficient to show their general character. West from Arboles, in the vicinity of Allison (fig. 2), there are many jacal sites. The latter fall into two general groups, one located along the Middle Mesa, skirting the Colorado-New Mexico boundary line, the other on the north side of the valley following the tops of the lower benches which slope down from the Piedra Peaks still farther north and east. No indications of ruins were found on the floor of the valley, although there may have been isolated units there ROBERTS ] PUEBLO RUINS IN COLORADO 67 at one time. This is one of the best agricultural sections in south- western Colorado, and because of the extensive cultivation which the valley has undergone any traces of ancient occupation which may have remained until recent times have been obliterated. On both sides of Spring Creek, still farther west, are ruin sites. They extend down the stream to its juncture with Pine River just below the little town of La Boca. Pine River is the next large tributary of the San Juan to the west of the Piedra. The district was studied and several ruin sites exam- ined a number of years ago by Dr. Albert B. Reagan. In describing one of the house types which he found Reagan wrote: They were made of poles stood apparently in upright position and adobe mortar plastered on both sides of these to make the walls. In several other instances a form seems to have been made of poles and the adobe poured into it and let dry. The adobe walls were all made by the puddling process. The roofs were made of cedar poles over which brush and probably rushes were placed and on top of this adobe mud was placed. Many of the ruins were destroyed by fire as is attested by the adobe being burned to a_ brick constituency.” Irom the description in the foregoing paragraph it seems quite evident that there was a form of the B type dwelling in the Pine River Valley. The presence of walls made by pouring adobe mud into frames made from poles is unique as far as the northeastern San Juan region is concerned and it is debatable that such actually occurred. Careful inspection of many sites in the valley failed to reveal any such evidence to the writer. It is of interest to note that in this section, also, the houses had been destroyed by fire. Along the Animas River, still farther west, are additional sites revealing remains of the characteristic jacal ruins. They occur sporadically from just north of the city of Durango down as far as the New Mexico line. From that point on south they are more numerous. It is only a comparatively short distance from the State line to the large group of ruins at and in the vicinity of Aztec, N. Mex., the subcenter of the eastern San Juan area discussed in the introduction to this paper. The most interesting district, from the standpoint of early house remains, west of the Piedra is that of the La Plata. (Fig. 1.) In that region are vast numbers of jacal ruins and it was the investiga- tions of Mr. Earl Morris in those remains that first called attention to the type. His findings are of especial value as a check on the results obtained by the recent excavations along the Piedra. It is rather curious:that the majority of the ruins examined by Mr. Mor- ris showed long rows of contiguous chambers, the realization of the 51 Reagan, A. B., 1919, p. 171, 68 BUREAU OF AMERICAN ETHNOLOGY [BULL 96 many-roomed structure, but none of the isolated single-roomed dwellings such as constituted the major part of the Piedra phase. Mr. Morris summarized the characteristics of the remains which he uncovered as follows: The inhabitants of the mesas were an agricultural people whose domiciles were one-storied aggregations of cell-like chambers, usually grouped to form a rectangie. Generally speaking, the rooms extended down into the earth, and with few exceptions the sections of the walls above ground were con- structed of upright poles covered with plaster.” A more complete and detailed description of one of the houses which he excavated shows unquestionably that it was of the type which has been called the Class B dwelling. Shallow trenches were dug where it was desired to place the walls. In these poles averaging about 4 inches (10.16 ecm.) in diameter were set side by side, and held upright by stones wedged inio the trenches on both sides of their butts. The poles were then coated with mud till they were almost, if not quite, hidden, and a strong wall superficially resembling one of adobe was formed. It is probable that the roof consisted of beams, twigs, and bark eovered with clay. The presence of the charred stumps of the poles still rest- ing in the trenches between the rows of stones, and the large quantities of plaster burned to a bricklike consistency, smooth on one surface and bearing upon the other the distinct imprints of poles, twigs, and knots, with the finger- prints of the primitive masons, shows these mesa dwellings to have been the structural analogues of the modern post houses of the Mexicans.” The B form of dwelling was not the only one represented in the La Plata district, for, judging from Mr. Morris’s report, there was a variation of the C style. In the description of one group of ruins it is observed that— The more easterly of these structures was built entirely of poles and mud, while parts of the other are of stone. There is not enough fallen masonry to indicate that the stone sections of the walls were more than 8 or 4 feet (91.44 cm. or 1.2192 m.) in height.™ One feature found in many of the house remains of this region was not encountered in any of the sites along the Piedra. The La Plata peoples frequently lined the walls of the excavated portion of the dwelling with large stone slabs. This led to the adoption of the term “slabhouses,” and many of the older reports on southwestern archeology refer to the Slabhouse people. Such a designation was found to be confusing, however, as later investigations revealed the fact that the late Basket Makers had a similar constructional feature in their houses, and the name slabhouse might indicate a dwelling belonging to one or the other of two different peoples and periods. The use of the word has been discontinued at the present time, although it is recognized that slab-lined pits may be the remains = Morris, 1919 b, p. 202. 53 Morris, 1919 b, p. 187. 54 Morrks, 1919 b, p. 188. ROBERTS ] PUEBLO RUINS IN COLORADO 69 either of Basket Maker or early Pueblo structures. The presence of slabs is not an essential characteristic, however, of the dwellings of either period, as many of the houses did not have them. The absence of slab-lined pits in the houses along the Piedra may be attributed to an environmental factor rather than to any out- standing cultural difference between the people of that district and those of the La Plata region. Nowhere in the immediate vicinity of the Piedra settlements was there a sufficient outcropping of stone suitable to an extensive use of slabs in construction work. Occasional large slabs could be obtained, but it would have been difficult to secure any great number of them without transporting the material a considerable distance. As a matter of fact, the large slabs used to cover the doorways in the houses must have represented a consider- able amount of effort on the part of the builders. The house remains of the La Plata in many instances exhibited a feature which has been commented upon in the discussion of both the Piedra and Pine River Valley ruins, namely, destruction by fire. With reference to this condition in one of the ruins which he ex- amined, Mr. Morris wrote: Fire destroyed the building and its contents. To judge from the large quan- tities of corn and the many vessels sitting about in the rooms, the conflagra- tion must have been sudden and catastrophic. Whether it started from wind- fanned sparks or was caused by lightning or by enemies is purely a matter of conjecture.” There is probably a factor of some significance in the widespread evidence of the burning of buildings. That it was sudden and unexpected is shown by the destruction of food and personal prop- erty. As Mr. Morris points out, the cause can only be suggested by speculation, but it seems unlikely that fortuitous sparks were responsible for such widespread conflagrations, especially the blot- ting out of entire villages. The occasional burning of a single house might be expected, but not that of whole communities. The writer is more inclined to think that, in most cases, conflict was responsible. The buildings may not have been fired in every instance by enemies, but by the owners who in precipitate flight after a defeat sought to destroy all that they could not carry with them, and thus pre- vent its falling into the hands of the plunderers. It seems more likely that at this particular time there was intervillage, possibly interregional, strife growing out of the commingling of Pueblo and Basket Maker peoples, with the attendant transition from one form of culture to another, than that nomadic groups had begun their depredations. This would in large measure explain the almost count- less remains of burned villages scattered throughout the eastern San 55 Morris, 1919 b, p. 191. 70 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 Juan area. It may well have been that after the attackers had moved on the survivors from a destroyed village returned to the site of their former homes and built anew. The changes and improve- ments in house construction may have arisen from just such hap- penings. The absence of kivas in villages of this period was also noted in the La Plata region. At the conclusion of his earlier work in the area Mr. Morris stated that no kiva had been found in or connected with a jacal dwelling,® but since that time he has discovered such chambers. He did ascertain, however, that some of the circular de- pressions gave evidence of having been large circular pit rooms, and suggested that they may have represented the prototype of the kiva. There is nothing to show that such was the case, but the writer sus- pects that there was something analogous between the latter and the so-called dance plaza, which was uncovered by Mr. Jeancon along the Piedra. The explanation for both may possibly be drawn from the late Basket Maker village in the Chaco Canyon, where the remains of a large circular structure suggestive of the early type of the great kivas of the Chaco cultures was found.®* The latter are distinct from the smaller clan kivas discussed in an earher para- graph, and are believed to have been the structures in which the major ceremonies of the entire village were performed, as contrasted with the minor clan observances held in the small kivas which rep- resented the old, original dwelling. If the suggestion that the large structure in the Chaco was an example of the forerunner of the great kivas is correct, there should be some evidence of the exist- ence of a comparable feature in the intervening horizons. Such evi- dence is possibly to be found in the large circular remains uncovered by Morris and Jeancon. Thus far there is nothing conclusive about the available data, and the suggestion is offered as an explanation and not as a demonstrated fact. The investigations by Morris developed, in addition to the forms of house construction, other features which show that there was a definite relationship to the cultural horizon represented by the Piedra jacal remains. One of the most outstanding of the latter is in the ceramic industry. This factor belongs to the consideration of pottery, however, and will be reserved for that section of the report. Another point of significance is that the skeletal remains in the refuse mounds accompanying the villages referred to showed an occipital deformation of the skull which is a condition compar- able to most of the human remains found in the Piedra district. 56 Morris, 1919 b, p. 202. "7 Roberts, 1929, pp. 80-81, 90. For an extensive treatment of great kivas and their possible significance in ruins belonging to the Chaco Cultures see Morris (1921) and Kidder (1924, pp. 50-51). ROBERTS | PUEBLO RUINS IN COLORADO ral Morris mentions that 11 undeformed skulls were found at a site in Long Hollow, but these were unquestionably from a late Basket Maker village and are not to be confused with those from the Pueblo I ruins. Before passing to the consideration of comparable examples of house construction in still other districts of the southwestern area the La Plata situation may briefly be summed up by saying that in the ruins described by Morris are to be found variations of the B and C class of dwellings of the Piedra region; that the structure of many contiguous rooms prevailed, while the isolated, single dwell- ings were missing. The sites show the same relation of houses to depressions, although none of the latter are believed by Morris to have been reservoirs. The closest approximations to the A type dwellings thus far re- ported from the Southwest were found at Beaver, Willard, and Para- gonah, in southeastern Utah, by Mr. N. M. Judd, who uncovered the remains of rather crude structures belonging to an early Pueblo horizon. These houses did not bear a striking similarity to the ones in southwestern Colorado in many of their details, but they did have the sloping-walled, flat-roofed superstructure supported on four posts. Mr. Judd’s description of the type as he found it at Willard follows: The four upright and perhaps notched posts supported crosspieces upon which lay lighter poles and split timbers. Over these were spread, in succession, layers of willows, reeds, or grass, and mud, the latter fitting closely about the two former and retaining their impressions perfectly after fire or natural decay had removed all other traces of the vegetable matter. The relatively small, flat portion of the roof between the uprights probably contained a smoke vent for the fireplace directly beneath. That the sides of this lodge sloped from the ground to the crosspieces sup- ported by the four central posts seems obvious from close examination of the floor. About the fire pit and within the square formed by the four pillars the earth had been tramped hard and smooth through constant use. Elsewhere the floor was traceable in direct proportion to the distance from its center, and in no place could it be followed with certainty more than 7 feet (2.1336 m.) from the rim of the fireplace. Further, the two pits previously described were so situated as to be well under a sloping roof, and therefore of least incon- venience to the inhabitants of the hut. Although no marks were found which might indicate the former resting place of inclining wall timbers, the mere fact that the floor, so distinct in the middle, became less plain toward its borders, suggests that this ancient structure was circular in ground plan, and in outward appearance not unlike the earth lodges of certain western tribes, or, for example, the familiar winter hogan of the Navaho.” Houses of this type were grouped in small communities but stood as separate units. They seem to have been followed by rectangular I FS ee a se lee ee ee ee *8 Judd, N. M., 1926 b, p. 8. 88242°—30——6 72 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 structures of adobe, without timber reinforcements in the walls, which formed single-roomed dwellings or were combined into build- ings of several rooms, a stage which in its broadest aspects corre- sponds to the general features of the B period along the Piedra. There was one factor in such house groups that is of particular inter- est and that was the association with the adobe buildings of an arbor or court shelter in which the cooking or other household tasks were performed.®® It will be recalled that a similar bit of construction was found at the northern end of the main house row in the B village on the Piedra. Mr. Judd observes, in another part of the same report, that all of the cooking in the adobe villages was done in the open or under such arborlike structures. The many outdoor fire pits in the southern Colorado villages suggest that a somewhat similar condition prevailed there, although it was probably not as extensively practiced as in the Utah communities where none of the houses had an interior fire pit, excepting, of course, the earlier jacal- like domiciles. The remains of a culture which suggests some affiliations with that found by Judd in Utah were developed along the Muddy and Virgin Valleys in southeastern Nevada by M. R. Harrington. Some of the dwellings in his Pueblo Grande de Nevada appear to have had a form of the jacal construction very similar to that found by Judd. In one there were indications that the pole-and-mud walls had sloped in- ward, and it is quite possible that it may have been a form analogous to the A houses, but from the available information it is difficult to ascertain its exact status. One feature about the Nevada settlement which is strikingly reminiscent of the La Plata and Piedra villages is that the houses, of the many-roomed variety, were grouped about circular courtyards.°° Harrington also observes that most of the rectangular rooms with solidly built walls were so small that they could have been used only as granaries. This suggests a condition similar to that of the stone structures in the C group.* One other characteristic in the Nevada village of interest to this study is that there appeared to have been a use of arborlike shelters such as was reported by Judd and as found in the B village on the Piedra. This use of open structures was apparently a rather common practice in the early Pueblo periods. The position of Pueblo Grande in the chronology of the Southwest has been in some dispute, but it seems to the writer that it began in Pueblo I and continued into early Pueblo II. There are clear evi- dences of late Basket Maker survivals in some of the remains, but 59 Judd, N. M., 1926 b, p. 29. 6 Harrington, M. R., 1927 b, p. 268. ®t Harrington, M. R., 1927 b, p. 267. ROBERTS] PUEBLO RUINS IN COLORADO °- 73 such may be expected in a site on the outlying margins of a culture area in a period which witnessed the mixing of two peoples and the development of a new culture out of an older one. To conclude, it may be stated that the three house forms studied in the Piedra district show certain general similarities to types in other regions. The A dwellings have no close counterparts, but slightly analogous structures have been found in the region north of the Rio Colorado, especially at Willard, Beaver, and Paragonah, Utah; and in the Muddy River Valley in southeastern Nevada. The B dwellings may be duplicated in the Pine River Valley and varia- tions of both the B and C forms are to be found in abundance in the La Plata district. A form of structure comparable to the stone por- tions of the C units was noted in the Nevada section. On the basis of such comparisons it would appear ‘that the jacal houses described in detail in this report represent an Early Pueblo phase which is typical of the northeastern San Juan area, but which have in general only a superficial similarity to the more widely scattered buildings of that cultural horizon. LESSER OBJECTS OF MATERIAL CULTURE A major element, the dwellings, in the material culture of the early inhabitants of the Stollsteimer Mesa section of the Piedra district has been discussed in preceding pages, but a number of smaller fac- tors which played an equally important role in the day-to-day life still remain to be considered. It is not possible to give an accurate, detailed picture of this phase of the culture because excavations in sites such as those along the Piedra yield specimens of one group only, those made from imperishable substances. There undoubtedly were many objects which, because of inherent qualities in the mate- rials from which they were fashioned, have not survived. Even though they may have escaped destruction in the fires which con- sumed the houses, they long since have decayed and become mixed with the earth which covered them. Although it may seem to be needless repetition, the fact that objects obtained from investigations of an archeological site located in the open, exposed to all the vagaries of the weather, generally tell but half a story is so frequently forgotten or disregarded that it should again be emphasized. Too many times a certain stage or culture is thought of only in the terms of stone tools and pottery, and the other elements which contributed in no small degree to making the complex a rounded whole are entirely ignored. It is true that a thing which is not present can not be described, but that such probably did exist must not be forgotten. 74. BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 The group of objects which is not represented in the specimens collected from the Piedra sites but which were an important part of the material culture of the region includes baskets, sandals, other articles of dress, textiles, wooden implements, and weapons. That the people made and used baskets is shown by the impressions on the bottoms of many pottery containers; that they had planting and digging sticks may be assumed from the fact that they cultivated corn and beans; the existence of the bow and arrow is indicated by the many unusually fine arrowheads recovered; that their axes and mauls were handled is suggested by the presence of grooves for their hafting; that the people had clothing may be inferred from the find- ing of such objects in cave sites of the period, in other sections of the area, where dry sand has preserved them; but, with the exception of a single small fragment of charred basketry, not one scrap of the actual material in this group has come down to the present day in the Piedra district. All that remains upon which an appreciation of the skill and development of the people in the lesser factors of their material culture can be based are the pottery objects, bone and stone implements, articles of personal adornment made from stone and shell, and small fossils and curiously shaped stones which were collected and saved for unknown reasons. Porrery Among the objects left by a prehistoric people one of the most im- portant upon which the archeologist can base his studies is pottery. The material from which the vessels were made, coupled with the fact that they were fixed in form by being baked in a fire, makes for a permanence which is lacking in the vast majority of the objects fashioned by man for use in his daily life. The firing of the bowls and jars also made for a brittleness, a liability to breakage, which in itself is of value to the student of early cultures. The fact that many were broken, that there was the necessity for constant replac- ing of the damaged pieces by new ones, provided a quantity of mate- rial that is invaluable. Each vessel made presented a new problem to the potter and with the solution of each one more step was taken toward a mastery of technique. Changes appeared from time to time in the shape, ornamentation, and composition, and these changes were — faithfully recorded in the objects themselves. From this record it is possible to note the cultural growth of a people, the development of their artistic ideas, and the progress of their craftsmanship. From the standpoint of early Pueblo pottery the Piedra sites fur- nished information of considerable value. The large number of specimens obtained show clearly that the ceramic industry was in the full exuberance of its youth, the hands and fancy of the potters were ROBERTS | PUEBLO RUINS IN COLORADO 10 unhampered by convention, by long-established style and technique, there was a plasticity of shape and decoration, a multiplicity of in- vention and trying out of new methods not to be equaled in any following period. True pottery first made its appearance in the late Basket Maker horizon but did not assume a role of major impor- tance until the Pueblo I period, when it definitely usurped the place held by basketry in earlier stages of the prehistoric sedentary south- western cultures. It was pointed out in the general introduction to this report that unfired clay containers were probably known in the closing days of the main Basket Makers, period II, but the ceramic idea did not become fully developed until the following stage. The complete realization of its possibilities, however, was not attained until the first Pueblo period, which saw the culmination of many features prophesied in the Basket Maker III wares, in addition to the appearance of a variety of new ones. There was no lack of material in the San Juan area from which the potters could fashion their vessels, for suitable clay is to be found in fairly close proximity to practically every ruin. The clay in itself was not sufficient, however, as some sort of binder or tempering mate- rial was necessary to prevent its cracking in the process of drying in the sun and later firing. In the unfired prototypes of the closing days of the Basket Maker II era cedar bark furnished the binder in the clay.°2 When the firing of vessels was developed it was necessary to replace the bark by a noncombustible substance and the custom of mixing sand with the clay, true sand-tempered pottery, appeared. The earliest specimens in this group show a large content of sand, the potters not yet having learned that a smaller amount of tempering and a better kneading of the clay would give it a harder, more com- pact texture. The later forms show a marked reduction in the amount of sand used. This in turn was replaced by ground or pow- dered rock which varied in nature from district to district, but in general being either dark igneous particles or a light-colored quartz- like substance. Still later periods saw the use of ground potsherds, a practice in vogue even at the present time among some of the modern Pueblo potters. Regardless of the material it was necessary to take creat care not to use too much tempering, because under such condi- tions the cohesiveness of the clay would be lost and the vessel would tend to crack or collapse. It is impossible to know the exact manner in which the materials were treated and how they were fashioned into bowls and jars, but a careful study of the vessels themselves and of fragments shows many things. An idea of the stages involved in the making of such an object may be gained through a knowledge of the methods employed aac al rte ee ee + Morris, E. H., 1927, p. 198. 76 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 by modern potters of the region. There is no doubt but that the vessels were made by hand and without the aid of a potter’s wheel. Supports were used, as is shown by the impressions of baskets on the bottoms of some of the specimens. In some of the ruins of later periods bases formed from the cut-down tops of large jars have been found. Even at the present time some of the Pueblo women make use of similar objects.®* It has been suggested that the latter objects might eventually have led to the discovery of the potter’s wheel, since they were turned from time to time as work progressed in order that the surface occupying the maker’s attention might be brought closer to her, but such is rather doubtful because even the advent of the white man and his suggested improvements has failed to make any material change in the basic Pueblo methods. The earliest vessels, those of the mud-pottery group, appear to have been made by a combination method. The bottoms were molded in baskets and the sides built up by means of strips of clay rolled out to the desired size and then looped around the circumference, each loop making but a single circuit of the perimeter and being pressed or welded to the one immediately below. In the vessels with con- stricted openings longer loops were used where a swelling out was desired, shorter ones where a drawing in of the sides was deemed advisable. This method seems to have lasted on down through the fired wares of the late Basket Maker and early Pueblo periods, when a new method was evolved. The latter consisted of the use of longer loops of clay which were carried around the growing vessel several times before being fastened at the end. This was the beginning of the coil. Its culmination was reached in the full Pueblo periods, IT and Ili, continuing through to the present, when the entire vessel was made from a long spiral of clay, each shorter strip being welded at its end to the one preceding, thus forming one long continuous strand. The additions were made as needed during the growth of the jar or bowl. This difference in the technique of manufacture is an important one because it marks the boundary line between charac- teristic Pueblo I and later vessels. The short-looped ones are of the early period while those with the spiral coils belong to the following horizons. A great deal of attention was paid to the surfaces of the wares. In the early types all signs of the coils or loops of clay used in the building up of the walls were, as far as possible, obliterated. This was accomplished through the use of some implement, a corncob, a piece of gourd, a stick, bone, fragment from a broken pot, or a stone. In later periods, when the appreciation of decoration had reached a higher development, the coils came to play an important ® Stevenson, J., 1883, p. 329; Guthe, C. B., 1925, p. 27. ROBERTS } PUEBLO RUINS IN COLORADO owe part in the type of vessel which was used for culinary purposes. A painted design would soon become obscure from soot and smoke from the fire, and perhaps for this reason it became the custom to allow the coils to remain on the exterior and by various ways of pinching them together and of indenting with the finger nail or some sharp- pointed instrument a pleasing decoration was obtained. The sur- faces of pots destined for food bowls, storage and water jars, and other nonculinary purposes were generally smoothed down. As indicated in the foregoing, one classification for pottery in the Southwest, irrespective of period, is based primarily on function, a feature of considerable importance not only with respect to the form and general character of the vessel but in its quality as well. Thus far no more satisfactory grouping has been developed than that of the twofold, culinary and nonculinary, grouping. The culi- nary vessels, as mentioned above, fall into three general classes according to the period. ‘The earliest, those of Basket Maker ITI, were smooth surfaced; the Pueblo I variety were those with the banded necks; then came the indented corrugated pots of Periods II and III, whose entire exterior surfaces showed the coils from which they were made.** In general it may be said that the outside of these vessels is almost always black, due to their use over fires, but now and then one is found which had not been so used or, as in the case of some of the specimens from the Piedra, the soot had been burned off by fires which destroyed the houses, so that the original color of a varying shade of gray is observable. Characteristic features of the nonculinary group in the San Juan area, irrespective of the period to which it belongs, may be gen- eralized to the extent of stating that the surfaces of the vessels were generally smooth and usually ornamented with some form of painted decoration. This group is composed of one major and two minor forms. The major class is that of the well-known black-on-white pottery, while the two minor forms are the red with black decora- tions and vessels with a polished black interior, the extent of the polish depending largely upon the period in which the latter were made. The earliest forms show scarcely any polish while those of Pueblo III times have a very high gloss. Efforts to make too limited a definition for the black-on-white pottery, light-colored vessels with painted black designs, have led to confusion and it has been found that hard-and-fast rules as to the interpretation of the term are not advisable. The light-colored surface may vary from a dull, dark gray to an almost chalky white. 6 For a synoptic illustration of the ancient corrugated wares see introduction to Guthe, 1925, pl. 2, by A. V. Kidder. The term Pre-Pueblo is the older name for the present Pueblo I, Early Pueblo is Pueblo II, and the Classic Pueblo examples are of the Pueblo III period in the classification used in this paper. 78 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 This depended to some extent on the question of material, proper mixing, proper firing, and other variable elements, so that the breaking up of the group into black-on-gray and black-on-white forms is not practicable. The same potter might frequently have both types in a single batch of pots. The decorations show similar traits in that the pigment may be any hue in the range from brownish-red to lampblack. This was in great part due to the manner in which the vessel was fired and is not to be attributed to a difference in the make-up of the paint. A properly fired vessel would have a good black decoration, a slight amount of overfiring would give a brownish-black and a marked overfiring the brownish- red tone. In this connection it may be noted that the tendency to brownish-red and brownish-black hues was more marked in earlier periods than after the technique of firing had become better devel- oped and more thoroughly understood. The red vessels with black decoration show distinct and separate characteristics for most of the periods, but inasmuch as they played a small part in the ceramics of the earlier periods, rarely does the proportion of black on red to black on white exceed 2 per cent; any extensive consideration of the various stages is not essential to the requirements of this paper. It will be sufficient to point out the main characteristics of the red wares of the Basket Maker III and Pueblo I periods. The earliest red vessels were the result of an intentional overfiring of bowls, jars, and other forms which other- wise would have been of the black-on-white variety. This over- firing imparted a light orange-red tone to the surface, although, due to a certain lack of control in the firing, shades may be found varying from a yellowish-brown to an almost good red. This form of red ware first appeared quite late in the Basket Maker III horizon and extended over into the Pueblo I period. After the latter stage had gotten under way a certain amount of coloring matter seems to have been mixed with the clay. Whether this was an intentional or natural mixing is not known, but the probabilities are that the raw material contained the necessary pigment. Modern potters along the Rio Grande depend on clays containing the red pigment and not on the ordinary type which has been treated with red coloring matter. A still later development was that of the application of a liquidlike wash of clay made from the coloring matter to the surfaces of the vessel which it was desired to make red. This feature was appar- ently a late discovery of the Pueblo I horizon and was one which became characteristic of the red wares of later periods. The use of the “liquid” clay, called the slip, on vessels of the black-on-white group developed at about the same time, although it probably pre- % Morris, BE. H., 1927, pp. 186-187. ROBERTS ] PUEBLO RUINS IN COLORADO 79 ceded that of the red wares, because pottery with a true slip is first found in the Pueblo I period. The vessels of the shiny black interior subgroup are of two forms. One consists of the bowls with gray to grayish-brown exterior and the other those with a brownish-red exterior. The latter seems to have been an outgrowth of the former and both are undoubtedly the prototype of the polished black interior vessels with red exteriors, which are present in fairly large numbers in later periods. What methods the early potters used to obtain the black finish can never be known with certainty, but it is very likely that they employed methods quite comparable to those of the modern Pueblo Indians in certain villages along the Rio Grande, where shiny black vessels are a favorite form of pottery. The latter use a smothered fire to obtain the desired effect.°° The smothered fire produces a dense smoke, which penetrates the paste of the vessel and leaves a carbon deposit which gives the desired black. This black would disappear if the pots were reburned in an open flame. The prehistoric potters were not as adept in the smoking process as their modern descendants, because large, irregular black splotches on the exterior of such vessels were of frequent occurrence. The carbon deposits on the interior of bowls of this class penetrate but a short distance into the paste so that cross sections of fragments show but a thin black line. The manufacture of pottery by the sedentary southwestern peoples was a general household industry, not a specialized craft restricted to skilled workers. As a consequence occasional forms are found which can not be fitted into any general classification. These indi- vidual variations and off forms should be considered as such, not as examples of defects in the method of systematization. A good illus- tration of the latter is apparent in the pottery from the Piedra. There are two groups, or rather subgroups, in the nonculinary wares which in one sense should be classified as separate forms but which in another should be considered simply as variants of one of the major divisions. There are a large number of vessels which have every characteristic of the black-on-white wares except that they have no decorations, no painted ornamentation. Some investigators and students might feel that they should be placed in a separate group, but the drawing of such fine distinctions on the part of earlier workers has led to so much confusion that it is thought that for the sake of simplicity they should be regarded merely as aberrant forms of the black-on-white group, which indeed they are. Many times the lack of a decoration can not be attributed to an original omission on the part of the maker but to the weathering away of the decoration since the vessel was made. Careful study of many specimens which rE aL 6 Guthe, C. E., 1925, pp. 74—-Td. 80 BUREAU OF AMERICAN ETHNOLOGY [BULL, 96 suggest the entire absence of painted ornamentation shows that origi- nally they had been decorated, but time has practically obliterated all traces of the designs. The second debatable class is that of the miniature vessels which might well be grouped under the heading of toys. They could have been of no practical use and must have been the playthings of the children. Even to-day a small girl occasionally may be seen at one of the pueblos making little vessels for her own amusement. There has been some tendency on the part of a few investigators to assign these objects to the group of ceremonial offerings, but such a cate- gory seems a little far-fetched. In all respects, except that of size and usefulness, the miniature pots correspond to the general charac- teristics of one of the major groups, culinary or nonculinary, of vessels. The general consideration of features of the prehistoric pottery of the San Juan area briefly sketched in preceding paragraphs has been deemed essential to a proper understanding of the ceramics of the Piedra sites because many characteristics evidenced by the lat- ter can not be appreciated without such a discussion. As a matter of fact the pottery from the latter district presents such a variety of features that a proper examination of it can not be made without reference to the general aspects of the early development of the industry. The major part of the pottery which is described and discussed in the following pages belongs in the A and B house grouping. Only a brief consideration will be given to the ceramics of the C group. Since only fragmentary specimens were obtained from those sites, a lengthy study is out of the question. Also, the C houses bordered so closely on the Pueblo IT period that an extensive dis- sertation on the lesser objects of material culture accompanying them is not practicable in view of the fact that the bulk of this report deals with material from the Pueblo I horizon. Considering the pottery from the A and B villages and houses as a unit, regardless of function, ornamentation, or shapes, there are certain features which may be considered as characteristic. The paste, the mixture of clay and tempering material which formed the substance from which the vessel was made and which in the finished object composes the body of the walls and bottom as contrasted with their surfaces, shows considerable variation. The latter is most marked in the tempering material, which may consist of white sand, small, irregular patches of igneous rock or of a light-colored stone. In general the clay used was a fine-grained type which was capable of being kneaded and mixed to a uniform consistency which would produce a hard, compact structure in the paste. The full possibili- ROBERTS J PUEBLO RUINS IN COLORADO 81 ties of the raw material were seldom realized, however, because a majority of the vessels exhibit a rather granular, checked condition in their internal structure. Under proper firing conditions the paste becomes a light shade of gray or even a fairly clear white, although examples are found which are a dull, slate-gray hue. When the vessels were underfired a distinct dark streak remained in the center of the paste. The hardness of the paste, also dependent upon the mixing and firing, varies greatly. Some of the pots are hard while others are quite friable. Surfaces on most of the vessels were smooth. That is, the loops of clay from which they were fashioned were obliterated by rubbing when the pot was still moist. The necks of the culinary jars and some of the large water and storage vessels constitute the chief ex- ception. On the latter the bands were allowed to remain on the necks and form a simple decorative effect. The actual surface finish varies greatly but has a distinct correlation with paste quality. Con- tainers made from well-reduced, finely-tempered clay have a com- paratively smooth exterior texture, but those of the sand-paste group have rather coarse, rough surfaces. Although rubbed to a greater or less degree, none of the vessels show the careful smoothing and polishing which marks the wares of the late Pueblo II and Pueblo III periods, in fact, no polishing stones of the forms so frequently found in ruins of those periods have thus far been found in late Basket Maker or Pueblo I sites. The vessel surfaces of these early stages indicate that the smoothing implement was either a corncob, piece of gourd, stick, bone, or, possibly, a rather rough stone. As mentioned in a preceding paragraph, the practice of covering the surface of a rubbed-down vessel with a coating of “liquid” clay to which some coloring substance, generally kaolin in the white wares and ocher in the red, had been added first appeared in this period. A number of vessels from the Piedra sites have a true slip while many have no slip at all and present the characteristic aspect of the black-on-white wares of the late Basket Makers. Many of the latter suggest the application of the clay wash but in reality have only a pseudo slip, a creamy film of clay brought to the surface in the process of smoothing and rubbing the vessel. ‘The clay brought to the sur- face in this manner did not completely cover the coarse grains of tempering material and the protruding particles give to it a slightly bespeckled appearance which is typical of that period. Where a slip is present on the Piedra vessels it suggests an element of ceramic manufacture still in the developmental stages. Its characteristics are difficult to describe because it is so variable. On the same vessel it is both thick and thin. This may be considered a good indi- cation that the clay and water were not thoroughly mixed. Possibly 82 BUREAU OF AMERICAN ETHNOLOGY [EULL. 96 too much solid substance was used, and as a consequence all that in excess of the amount which would go into solution settled to the bottom of the container. In dipping up the “liquid” clay to apply it to the surface of the vessel the potter may occasionally have gone deep enough to get some of the sediment. This would account for the spottiness of the slip which in places is so thick as to be almost lumpy and in others so thin that the gray of the underlying paste shows through. On rare pieces the tendency to a thick slip is consistent over the entire vessel, and where such is the case the surfaces have a smooth, soft texture which is almost enamel-like in its quality. Many of the vessels clearly indicate that the slip was applied with a textile mop; the impression of a rubbing rag is unmistakable. The modern Pueblo potters use folded cloths to apply the slip to their vessels, and it is quite possible that this is a survival in technique from the earliest days of the industry.®* The slip is rarely a dead white but has a faint suggestion of a pearly gray tone, a slight yellowish cast, or occasionally an almost imperceptible pinkish hue which imparts a feeling of color warmth entirely lack- ing in the chalky tinted vessels. A majority of the ceramic objects in the Piedra collection bearing a slip are of the bowl form, but this is in part attributable to the greater number of bowls, the percentage in the various groups being quite consistent. A rather interesting contrast between the bowls and other forms, however, is that the latter when decorated generally have a slip while the former has a large number without the extra surface treatment. There is a point of some significance in this factor when it is recalled that in the preceding period, when no slips were present, ceramic decoration was confined largely to bowls. The appearance of the slip and the development of jar and pitcher ornamentation are features of the Pueblo I period, and under such circumstances it seems fitting that they so frequently occur together. The collection of pottery shows 46.3 per cent of the vessels with a shp, excepting of course the culinary pots which rarely were so treated. The difference between this group and specimens of the nonculinary vessels from sites of later periods is quite marked. The percentage of nonslipped vessels in the latter is so small that it is practically negligible. Here, then, is an interesting contrast. The late Basket Maker pottery is characterized by a total absence of the slip, in the Pueblo I wares slightly less than half have it, while in the later periods of the Pueblo cycle it is present in practically all cases, There was still another feature in the treatment of the exterior sur- faces of vessels. This was the application of a red wash. The latter SEGutive, ‘C,H, 1925, p. 57. ROBERTS | PUEBLO RUINS IN COLORADO 83 was not fired into the surface and made permanent; hence it is very indistinct or not apparent at all on many of the fragments and speci- mens found. Where potsherds or vessels were exposed to the weather all traces of it have been washed away, but whenever they are found in protected places the red is quite noticeable. The feature was missed entirely by many of the earlier investigators in the region be- cause they neglected to study their finds before they were washed and cleaned. The impermanent nature of the red wash led to the adoption of the term “ fugitive red” by southwestern archeologists. Whenever that term is used reference is made to surface treatment of the kind just described. It is a common characteristic of the late Basket Maker pottery °* and is to be observed on many of the Piedra specimens. The practice does not seem to have survived to any great extent in the following Pueblo periods. In fact, its discontinuance may be associated in some way with the application of a true slip to the various pottery forms. Just what the relationship between the two features is has not been ascertained, but the one did not last for any great length of time after the appearance of the other. VESSEL FORMS At no other period in the prehistoric Southwest was there such a diversity of form in ceramic objects. The variety of shapes is so nu- merous, the lack of rigid conventions so marked, that it makes classi- fication of form even more difficult than in the ceramics of the later periods where archeologists have not found it easy to obtain suitable names for the vessels. In general, however, it may be said that the vessel shapes as illustrated by the collection include full-bodied jars with long tapering necks and constricted orifices; full-bodied jars with short, cylindrical necks and constricted orifices; globular-bodied jars with short, squat necks and wide orifices; full-bodied vessels of an elongated spherical or ovoid shape with short necks and wide ori- fices; globular or spherical pots with a small, circular opening at the top; globular pots with a depressed or flattened top and small circu- lar orifice, the so-called seed jar forms; bowls; pitchers of many varieties; ladles of various forms; bird-shaped vessels; cylindrical jars; double-lobed jars; egg-shaped vessels; and eccentric forms. Although the 1928 work along the Piedra did not yield a specimen of the effigy type, animal forms must be included in the general list. The writer found, in his previous work in this district, an unusually fine duck and a mountain sheep.®? Both unquestionably belong to the same class and period as the pottery found in the recent investi- gations. There are some forms which are found in sites of this period 68 Morris, E. H., 1927, pp. 176-177; Roberts, F. IH. H., jr., 1929, pp. 110-111. ® Roberts, 1925, pl. 20. 84 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 in other sections of the San Juan which are not represented in the collections from the Piedra district. Whether they existed in the latter or not can not be stated. They possibly were absent and again it may be that examples simply have not been found. h FicurRE 15.—Culinary vessel shapes There was not as great a variety in the shapes of the culinary vessels as in the nonculinary group. That there were some differ- ences, however, is apparent in the outline drawings of the common types. (Fig. 15.) The shapes occurring most frequently seem to have been those with roughly globular or ovoid bodies (fig. 15, a, b; BUREAU OF AMERICAN ETHNOLOGY BULLETIN 96 PLATE 13 BANDED-NECK CULINARY JAR BUREAU OF AMERICAN ETHNOLOGY BULLETIN 96 PLATE 14 SMALL CULINARY VESSELS WITH BANDED NECKS BUREAU OF AMERICAN ETHNOLOGY BULLETIN 96 PLATE 15 SMALL CULINARY VESSELS WITH SMOOTH EXTERIORS ONIMOOD NI G4ASN SIMOEG 39DNV1 SI ALVTd 9 NILATING ADOTONH.LA NVOIYAINY JO NVAYNE ROBERTS] PUEBLO RUINS IN COLORADO 85 pls. 18, 14, 15), with necks whose sides were vertical, sloped slightly inward, or had a tendency to curve outward. A great majority of the necks on these vessels show traces of the bands of clay from which the pot was built up. The latter were entirely obliterated on the body but only hghtly rubbed on the neck portions; in rare cases (pl. 14, 6, c, for example) they were not rubbed at all. The presence of slightly blurred neckbands is of such frequent occurrence that they may be considered as one of the characteristic features of the jars of this class. Practically identical specimens belonging to the culinary group were found in the La Plata district by Mr. Morris.” There were many of the completely smoothed forms, but they con- stituted a smaller and much more variable group. (Fig. 15, ¢, d, e, 7; pl. 15.) The latter are quite reminiscent of some of the smooth- surfaced culinary vessels of the late Basket Makers.*+ This is true not only of the form and surface treatment but in their appearance as well. Quite a large proportion of them have the rough, bespeckled quality described for the earlier vessels and no doubt represent a survival of such forms. As a matter of fact, even the banded neck groups suggest considerable relationship in form to some of the plain- surfaced jars of the preceding horizon. One of the curious forms in the culinary group and one which apparently did not survive in later periods is that of the tublike bowls. (Fig. 15,2; pl. 16.) The manufacture of bowls for cooking purposes was not common in the Pueblo cultures and up to the pres- ent time has been reported from only one late Basket Maker site.” These vessels differ from the ordinary type of bowl in that they have a slightly incurved wall just below the rim and in the proportion of the depth to the diameter. They correspond more closely to the body sections of the culinary jars or even to those of some of the seed jars of the nonculinary group. There is a possibility that they may represent a developed phase of one type of culinary vessel found in the Basket Maker III horizon which is practically absent in the Pueblo I. This is the form which Mr. Morris has called the de- pressed spherical vessels without necks.”* All that would be needed to convert such forms into the tublike vessels would be to enlarge the orifice and add handles. The suggested but debatable prototype of the latter has not been included in the outline drawings because only two such vessels are represented, both by fragments, in the collection from the Piedra. They constitute so small a percentage of the culi- nary vessels that they can not be considered an average form. They 70 Morris, E. H., 1919 b, pl. 68, 0; pl. 68, a, b, c, d, e, f; pl. 70, d. i Morris, E.'H., 1927, p. 165. iZiRoberts,,H. tH. +H.,, Jt-,, L929) :p. 112; 7 Morris, E. H., 1927, pp. 163-164; Roberts, F. H. H., jr., 1929, fig. 29, p. 112; Kidder, A. V., and Guernsey, S. J., 1919, pl. 59, a. 86 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 could very well be a survival of actual objects from the preceding period. The tublike vessels present still further complications in that they are capable of being classed in either the culinary or nonculinary categories. They are an example of one of the difficulties encountered in any attempt to establish hard and fast rules for ceramic classifi- cation. They illustrate one of the objections frequently made to the functional consideration of pottery. Enough of three vessels was secured to make possible their restoration, although fragments from many more were found, and of the three, two had definitely been used for culinary purposes. They were found in the fire pits of the houses from which they came, showed blackening by smoke, and one con- tained charred residue from some food substance which had boiled dry init. The third shows no signs of having been placed over a fire and in addition has a slip and a decoration on the interior. The paste of all three is of the coarse, rather sandy type characteristic of the culinary vessels. This would suggest that the decorated one was the aberrant form rather than that the other two had been put to a use for which they were not primarily intended. It may be remarked further, that of all the fragments from such vessels which were noted during the progress of the work no other indication of a painted decoration was observed. Pitcher forms in the culinary group seem in general to have been of two varieties. One had a rather high neck and constricted orifice (fig. 15, 7) rising from a spherical body. On the pitchers of this form the handle extended from the rim to the shoulder. The second form (fig. 15, 4; pl. 14, @) was characteristically of the general cooking-pot shape; the body form is exactly that of the large banded- neck vessels. The only distinction to be made is that one type has a handle while the other does not. The handle did not extend from the rim, however, but was joined to the wall of the neck a short distance below it. The orifice at the top was much larger in this form of culinary pitcher and the vessel probably more service- able for that reason. That this was recognized is indicated by the fact that it was the more prevalent form. The latter is found in practically every early Pueblo site. It is not confined to the Pueblo I period but survivals are found extending down into later periods. The survivals in some cases are probably representative of the form only, having been made in the later stages, and in others no doubt are vessels which actually passed down to subsequent horizons. Morris obtained examples in his La Plata work, Fewkes found them in the Navaho National Monument district, Kidder and Guernsey like- wise secured the form there and they have been reported from other ROBERTS ] PUEBLO RUINS IN COLORADO Qi districts.** Many of the latter, ROwey. er, are from ruins subsequent to the Pueblo I period. There was a marked gradation in the size of the vessels, even when the miniature forms are excepted. The jars of the globular-bodied form (fig. 15, a) range from a height of 3% inches (9.9 cm.) to 1314 inches (34.3 cm.). Body diameters in the group grade from those of 344 inches (9.4 cm.) to those measuring 111% inches (28.3 cm.). Neck heights—that is, the banded portions of the jars—vary from 114 inches (3. 2 cm.) to 34} inches (8.5 cm.). There was not as great a difference in the ddiances of the walls of the various vessels as in their other measurements. The range in average wall thickness was only from five thirty-seconds of an inch (4 mm.) to one-fourth inch (6.4 mm.). The bottoms have a greater average thickness than do the walls of the jars. They range from one-fourth of an inch (6.4 mm.) to five-eighths of an inch (1.6 cm.). An occasional specimen will show a bottom thickness of three-fourths of an inch (1.9 cm.). The cooking jars with the ovoid-shaped bodies and banded necks (fig. 15, 6) show a range of height from 51% inches (14 cm.) to 1514 inches (38.7 cm.). Their body diameters are from 334 inches (9.54 cm. ) to 124, inches (31.6 cm.). Neck heights, from the lip of the rim to the bottom of the lowest band, grade from 27% inches (6.3 em.) to 544 inches (14.5 cm.). There is no appreciable differ- ence in the average thickness of the vessel walls in this group. All approach so closely to three-sixteenths of an inch (5 mm.) that it may be considered as the normal measurement for that feature. Bottoms show a slight variation with a range of from one-fourth of an inch (6.4 mm.) to three-eighths of an inch (9.6 mm.). Some of the larger vessels possibly have an even greater bottom thickness but it was not possible to measure them. Such specimens would con- stitute a very small percentage of the group, however. The pitchers of the typical cooking-pot form with handle attached (fig. 15, 2) were rather small. Their height variation ranges from 414 inches (11.45 cm.) to 8 inches (20.82 cm.). The body diameters grade from 4,5 inches (10.3 cm.) to 614 inches (16.51 cm.). Neck heights are 13% inches (4 cm.) to 314 inches (8.89 cm.). The aver- age thickness of their bottoms and walls show such slight differences that they are practically negligible, being from five thirty-seconds to three-sixteenths of an inch (4 to 5 mm.). Pitchers with spherical bodies and inward sloping necks with con- stricted orifices (fig. 15, g) are as a group somewhat larger than the preceding form. Heights are from 714 inches (19.1 em.) to 11 74 Morris, E. H., 1919 b, pl. 70, c; Fewkes, J. W., 1911 a, pl. 18, UJ; Kidder, A. V., 1924, pl. 34, g. $8242°—30——7 88 BUREAU OF AMERICAN ETHNOLOGY [BULL, 96 inches (27.94 cm.). Body diameters grade from 54% inches (15.1 em.) to 1214 inches (31.75 em.). The banded portions vary from 3 inches (7.62 cm.) to 514 inches (13.34 cm.). The average thick- ness is five thirty-seconds inch (4 mm.) to one-fourth inch (6.4 mm.). The smallest of the tublike vessels has an average height of 9 inches (22.86 cm.). The circumference tends to the elliptical rather than the circular form with diameters of 117% inches (30.18 cm.) and 1234 inches (32.4 cm.). The average wall thickness is one-fourth inch (6.4 mm.). The intermediate form has an average height of 9,5, inches (23.8 cm.) and an average diameter, it more closely ap- proaching the circular form, of 1434 inches (36.5 cm.). Its walls average one-fourth inch (6.4 mm.) in thickness. The latter increases in the vicinity of the handles, and the bottom reaches a maximum of five-eighths of an inch (1.6 cm.). The largest of the group has a height of 9%; inches (23.4 cm.), an average diameter of 15 inches (38.1 cm.), and an average thickness of five-sixteenths inch (8 mm.). The bottom has a seven-sixteenths inch (1.2 cm.) thickness. Handles on the tublike bowls are placed well down on the sides of the vessels. Those on the first are 414 inches (11.43 cm.) below the rim. The second one has them 45g inches (11.8 cm.) below the rim and the third one 6 inches (15.24 cm.) down on the side. The forms of the handles themselves are as distinct as their measurements. Those on the first or smallest vessel are of the flat, shelf-like type. They measure 1 inch (2.54 em.) in length, 114 inches (8.81 cm.) in width and are one-eighth of an inch (8 mm.) thick. The second one has handles of the “ door-knob” type. They are 2 inches (5.1 cm.) in diameter and extend out from the sides of the vessel 114 inches (8.81 cm.). The largest vessel has handles of the nubbin type. ‘They are 13% inches (3.5 cm.) in diameter at the base and project from the vessel walls 114 inches (8.18 cm.). The small vessels with smooth surfaces which represent a survival from the preceding period range in size from those with a body diameter of 344 inches (9.4 cm.) and height of 3% inches (9.9 cm.) to those with a body diameter of 6 inches (15.24 cm.) and a height of 614 inches (15.9 cm.). The average wall thickness ranges from five thirty-seconds of an inch (4 mm.) to three-sixteenths of an inch (5 mm.). Vessels comprising the nonculinary group show a much greater variety of forms. This may in part be attributed to the fact that their functional requisites were less restricted than were those of the cooking pots where efficiency was the prime consideration, and in some degree to a striving for containers which would be both useful and pleasing to the eye. The nonculinary vessels were the forms ROBERTS ] PUEBLO RUINS IN COLORADO 89 which furnished the background upon which designs were painted, and the desire for artistic expression exemplified by the decoration seems to have been reflected in the vessels themselves. A badly formed jar, pitcher, or bowl would scarcely be compatible with a well-conceived design, and although the potters of the period fell far short of perfection in both shape and ornamentation there is evidence of a conscious striving toward that end. Although the black-on-white, black-on-red, black interior, and un- painted vessels might well be considered as separate groups, there is so pronounced an overlapping of shapes that the discussion will be greatly simplified by considering the nonculinary vessels as a unit. The black-on-red and black interior wares constitute so small a per- centage that attention may be called to the forms peculiar to them without devoting a separate section to their description. The un- painted vessels, as already mentioned, have such close relationship to the black-on-white group that they should be classified with them. As will become apparent in the ensuing discussion, there is a certain constancy of shape in some classes, the bowls for example, while others, especially the jars and pitchers, exhibit considerable diversity. The large water or storage jars have marked individual variations which make them rather difficult to classify in a study of shapes, but there are certain forms which may be considered as typical and representative of the period. Those illustrated in the outline draw- ings (fig. 16) may be considered as characteristic of the four, how- ever, a and d being the most typical. The body shape, from the line of greatest diameter to the bottom, is generally between a half-oval and a hemispherical form. The bot- toms are either rounded or slightly flattened and none have the small concavity which is apparent on so many of the large jars of later periods in the San Juan area. Quite a number show clearly the imprints of the baskets in which they rested while being made. The upper parts of the vessels show the greatest variation. They were generally slightly depressed or flattened and then turned upward to form the neck. The main difference in many of the vessels lies in the extent of the flattening of the upper zone. The forming of the neck by continuing the vessel walls upward is a feature similar to that found in the jars of the Basket Maker IIT wares where the neck was also a part of the vessel wall. One group, however, had a charac- teristic which is found to prevail in the later horizons. The neck was made as a separate piece and then joined to the body. ‘This form is represented in Figure 16, c, and Plate 18, b. The latter suggests quite strongly the globular-bodied, short-necked jars of the following periods; perhaps represents their prototype. The egg- shaped body illustrated in Figure 16, b, and Plate 18, a, is not as 00 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 characteristic of the Piedra wares as the other forms (pls, 17; 19, 6), but still may be considered as a type in that more than one example is represented in the collection. It suggests a variation of the c form but fits into the general group. Its chief difference is that the upper portion is more rounded than in the other shapes. 36 OO Fiaure 16.—Storage and water jar shapes The necks on the various vessels are either smooth and decorated or the bands of clay from which they were built up have been left unrubbed as in the case of the culinary jars. Examples of both treat- ments have been found in all of the shapes. The remainder of the surfaces on the jars were smoothed down. The paste and surface quality, however, is as variable as the shapes. Some of the vessels | BUREAU OF AMERICAN ETHNOLOGY BULLETIN 96 PLATE 17 BANDED-NECK STORAGE OR WATER JAR SYVFYALVM YO FDOVYHOLS AYVNITNONON q D 81 ALVId 96 NILATING SO TONH.LA NVOINANYV JO NVANNE SYVFYALVM YO 439DVYOLS GALVYOOAG q 61 ALW1d 96 NILATIONG ADOTONH.LA NVOIMANY AO NVAYN BULLETIN 96 PLATE 20 BUREAU OF AMERICAN ETHNOLOGY DIFFERENT FORMS'OF SEEDIJARS ‘ROBERTS ] PUEBLO RUINS IN COLORADO QO] have coarse, sandy paste quite like that of the culinary wares, while others are characterized by fine, hard walls and comparatively smooth exteriors. Both slipped and unslipped examples are in evi- dence. Some are decorated and many are not. The greatest variation in the vessels of this group is in the han- dies. There are almost as many different forms of the latter as there are jars. The individual fancy of the potter seems to have found an outlet in that portion of the pots which the tendency to conform to style in shapes denied. In some cases the handles are merely stubs or nubbins of clay protruding from, the side of the jar, in others they are a single loop of clay, flat or twisted. Still others consist of double loops of clay, bent upward or downward and occasionally twisted. Sometimes the handle was bifurcated and the ends shaped in a semblance of horns. On other jars they are merely broad, flat protuberances suggestive of the stubby tail of a bird. Occasionally an effigy form is found. The latter represent either birds or animals. The modeling is of the impressionistic kind, however, and the bird or animal represented not easily identifiable. That jars of the types described in the preceding paragraphs are characteristic of the period and the culture is shown by their marked similarity to specimens from the Pueblo I sites in the La Plata region. Mr. Morris found many examples that can not be distin- guished from the Piedra vessels.7° As a matter of fact, all four shapes are represented in his collection from that district. It is possible that comparable forms have been found in other early sites, but a careful check of the reports on various investigations has failed to show that such is the case. Only one of the group illustrated by Figure 16, a, and Plate 17 has been restored from the fragments brought in from the field, so that the range in size can not be given, but from the remaining shat- tered vessels capable of being assembled it can be stated that they ‘approximate closely the measurements of the example which is illus- trated. The specimen figured has a height of 201% inches (52.1 cm.) and a body diameter of 1414 inches (36.2 cm.). The neck height is 5 inches (12.7 em.) and the diameter of the orifice 4,7‘; inches (10.3 cm.). The average thickness of the walls of the vessel is seven thirty-seconds of an inch (6 mm.). The bottom averages five-eighths of an inch in thickness (1.6 em.). The handles, which are of the single-loop, twisted variety, measure 4,5 inches (10.3 cm.) in length, are 2%, inches (2.8 cm.) in width and thickness and extend from the side of the jar three-eighths of an inch (1 cm.). Jar 6, Figure 16, Plate 18, a, has a total height of 173g inches (44.2 cm.) and a body diameter of 12;'s inches (30.7 cm.). Its neck ® Morris, E. H., 1919 b, pl. 67. 92 BUREAU OF AMERICAN ETHNOLOGY [ BULL, 96 a is 811 inches (9.4 cm.) in height and the diameter at the orifice is 4 inches (10.16 cm.). The average thickness of the walls is nine thirty-seconds of an inch (7 mm.). The bottom has a one-half inch (1.27 cm.) average thickness. The handles, which are of the simple double-loop variety, have an average length of 47% inches (11.3 cm.), are 1 inch (2.54 cm.) wide and one-half inch (1.27 cm.) thick. They project from the wall eleven-sixteenths of an inch (1.8 cm.). Jar c, Figure 16, Plate 18, 0, has a total height of 1534 inches (40 cm.) and a body diameter of 13% inches (34 cm.). The neck height is 314 inches (8.3 cm.) and the diameter at the orifice is 444 inches (10.8 cm.). The average wall thickness is nine thirty-seconds of an inch (7 mm.) and the bottom five-eighths of an inch (1.6 cm.). The handles, of the simple one-loop type, have an average length of 314 inches (8.3 cm.). The width averages 1;'; inches (2.8 cm.) and the thickness eleven-sixteenths of an inch (1.8 cm.). Their inner surfaces are one-half inch (1.27 cm.) from the sides of the jar. The jar figured as d, Figure 16, and Plate 19, 6, is 153% inches (39.5 cm.) in height and has an average body diameter of 137 inches (35.3 cm.). Jars of the same group with a much larger body diam- eter are represented in the collection, but on most of the latter por- tions of the necks are missing, so that the height can not be deter- mined. The largest body diameter noted is 15 inches (38.1 em.). The neck height of jar @ is 614 inches (16.5 cm.). The neck is oval in contour and has diameters at the orifice of 334 and 35¢ inches (8.6 and 9.2 cm.). The average thickness of the walls is seven thirty-seconds of an inch (6 mm.) and of the bottom one-half inch (1.27 cm.). The average length of the handles, which in this case are of the double-loop variety with a pronounced upward curve, is the same as that for the preceding jar, 314 inches (8.3 cm.). The average width is 114 inches (8.2 cm.) and the average thickness five- eighths of an inch (1.6 cm.). The handles project nine thirty- seconds of an inch (7 mm.) from the vessel walls. The group of vessels called seed jars, in some reports referred to as heart-shaped bowls, also shows marked variations in form. The simplest and the earliest were those with a pronounced globular body and small orifice. (Fig. 17, a.) This form occurs both with and without handles. When the latter are present they generally take the form of a simple knob or cone of clay as illustrated in Figure 17, b; Plate 20, a. These lugs, usu- ally two in number, are placed on opposite sides of the opening. They frequently have small vertical holes passing through them. Small cords passed through the holes would have enabled the owner to hang the vessel from the ceiling of the house or other convenient place where it would be out of the way. Some of the modern Indians. ROBERTS ] PUEBLO RUINS IN COLORADO 93 however, explain the presence of the holes on a different basis. They claim that the perforations were so placed in order that bits of feathers could be placed in them and the jars used as ceremonial objects on the clan altars. Such a thing is, of course, possible, but whether such actually was the case can not be known. The globular shape was a common one in the so-called seed jars of the Basket Maker ITI period and it continued to be quite popular during the early stages of Pueblo I. An interesting feature in the technique of manufacture makes it possible to distinguish these ves- sels from the standpoint of period. Those of the Basket Maker ITT —— \ a b fe d —— é na £ — hk i Ficurn 17.—Seed jar shapes horizon had a characteristic treatment of the lip of the orifice which sets them apart from the following group. The Late Basket Maker potters when completing their vessels smoothed the opening by rub- bing upward and outward from the interior. As a result of this practice a slightly perceptible ridge remained around the orifice. This was frequently rubbed but not completely obliterated. The comparable Pueblo I jars show that just the opposite procedure was followed. The opening was smoothed from the outside toward the interior and to find the slight ridge resulting it is necessary to feel inside the pot. This practice seems to have been an early develop- ment of the period as all of the Piedra specimens show it. 94 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96° A slight variation of the globular form with a moderately depressed top and larger orifice is quite common. (Fig. 17, ¢.) Seed jars of this form frequently have small handles placed vertically on the sides in a manner suggestive of ears. In practically all cases these are perforated by small holes which pass through the clay on a horizontal line instead of being perpendicular as noted for the 6 forms?) (P1520;'0:) The globular body eventually gave way to one which was more oval in form. There are examples in this group, both with and with- out handles. (Fig. 17, d, e.) The handles show great variation. They may be of the “ door knob,” nubbin, effigy, or loop variety. It: is of interest to note that the body shape illustrated in d is the one most frequently found in the black-on-red group. From the oval shapes it was a natural step to the next forms. ‘The latter may be considered as characteristic of the Pueblo I period Where the globular body was typical of the Basket Maker TI], ceramics the flattened top and hemispherical or half-egg-shapec body predominates in the first Pueblo horizon. In the Piedra dis. trict there was a marked tendency toward the half-egg shape with flattened top (fig. 17, 7, 7, 2), although the hemispherical body form 2, was present in fairly large numbers. These vessels in a large ma: jority of cases have handles of one form or another. Most of ther are of the broad, flat, shelf-lke form which is suggestive of a stubby bird tail. (Pl. 20, c, d.) Loop forms are present in fairly large numbers, however, and there is an occasional example of an effigy usually some indeterminate animal. One of the jars has well: modeled likenesses of human mammae for handles. (Pl. 20, e.) Such representations are not uncommon in the canteen shapes of later pericds,’® but this is the first example thus far observed in the wares from the earliest Pueblo stage. It definitely sets at rest the conten- tion of many that the representations of the mammae on pottery was a development growing out of contact with the white man. Seed-jar forms prevail throughout the following periods in the San Juan and are found to some extent. in other cultural areas, but in no case do they seem to have been present in such abundance ot in as great a variety of shapes as in the Piedra sites. This feature is so marked that it may be considered as one of the characteristics of the early Pueblo ceramics in the Piedra region. Sizes vary to an even greater degree than the vessel shapes. The globular forms, with the exception of the miniature group which is not included in any of the measurements, range from an average diameter of 4 inches (10.16 em.) to 81% inches (21.6 cm.). The com- 7 Cushing, F. H., 1886, pp. 512-513. ROBERTS | PUEBLO RUINS IN COLORADO 95 nonest size falls in the range between 5 and 7 inches (12.7 and 17.8 m.). The diameter of the orifices varies from 2 inches (5.1 cm.) to 314 inches (8.9 cm.). The oval body group on the whole is more consistent, ranging from theight of 234 inches (7 cm.) to 53 inches (13.7 cm.) and a diameter 34 inches (8.9 cm.) to 7% inches (20 cm.). The orifices grade rom a diameter of 314 inches (8.25 cm.) to 3% inches (9.8 em.) The flattened top forms vary from a body height of 3,%, inches 8.45 cm.) to 8; inches (21.4 cm.); and from a body diameter of 4 inches (12.1 cm.) to 1275 inches (30.7 cm.). The orifices grade rom a 24-inch (7 cm.) diameter to one of 5 inches (12.7 cm.). The emispherical and half-ege-shaped body forms in this group are. so losely allied in the matter of measurements that the figures given are haracteristic of both. The average wall thickness for the entire seed-jar group ranges rom three-sixteenths of an inch (5 mm.) to nine thirty-seconds of ninch (7 mm.). The bottoms vary from an average of one-fourth fan inch (6.4 mm.) to one-half inch (1.27 cm.) in thickness. It is f interest to note in this connection that the wall thicknesses of all f the vessels, regardless of kind, show approximately the same meas- rements. Practically the entire collection falls within the three- xteenths to nine thirty-seconds of an inch (5 to 7 mm.) group. This ould indicate that the potters rolled out their loops of clay to a ither definite size, irrespective of the kind or size of vessel which ley were planning to make. The pitchers of the nonculinary wares also show a variety of 1apes. Typical forms are illustrated in the group of outline draw- igs, Figure 18, and in Plates 22 and 23. The body shapes of tese vessels, from the line of greatest diameter to the bottom, ce quite similar to those of the storage and water jars. Some are E the hemispherical form while others tend to the semiovoid shape. n certain vessels there is a pronounced shoulder at the line of great- it diameter (fig. 18, 6, 2), while others are more rounding (a, c, e, /). ody walls in most cases merge rather gradually into the necks, but i certain forms (fig. 18, /, g) there is a definite line of demarcation tween the two. The bottoms are either rounded or flat, the rounded rms predominating. There were no pitchers with slightly incurved * concave bottoms. The greatest differences in the pitchers are in the necks. Some ‘ the latter have slightly incurved walls with just a faint suggestion ‘an outward flare at the rim (fig. 18, a) ; in others the sides slope ‘adually inward from the body to the orifice (fig. 18, 5, c, e); still hers have practically vertical walls (fig. 18, d,7,g,h). The shape 96 BUREAU OF AMERICAN ETHNOLOGY [BULL 96 illustrated in outline fi is practically the same as that noted for one of the culinary vessel pitcher forms. The shapes illustrated in Figure 18, 7, j, are clearly representa- tions of gourds. The neck portions are continued and bent down- ward to form hollow handles for the vessels. On the smaller of the JEG Ok SOA6G8 PicukE 18.—Varieties of pitcher forms two forms the neck or handle curves down until the tip almost bu not quite touches the incurving side just above the shoulder of th: vessel; on the larger it actually touches. Such pitchers could be mad by cutting a circular opening in the top or neck of a gourd o that type. An interesting point is raised by these objects. Do they represen the old, original form of pitcher, the prototype patterned after vege IASSANHHSMGEVGNY Avy mGadsss IZALVTd 96 NILATING ADOTIONHLA NYOINAWY JO NVAENE BUREAU OF AMERICAN ETHNOLOGY BULLETIN 96 PLATE 22 PITCHER FORMS BUREAU OF AMERICAN ETHNOLOGY BULLETIN 96 PLATE 23 Decorated Undecorated TYPES OF SMALL PITCHERS BUREAU OF AMERICAN ETHNOLOGY BULLETIN 96 PLATE 24 BLACK-ON-RED SEED JARS ROBERTS | PUEBLO RUINS IN COLORADO 97 tal models, as has long been held by some students of southwestern ceramics, or are they a special form appearing only after the manu- facture of pottery was well established? Present evidence suggests that the latter was the case. Pitchers had been made for a consider- able period of time before the actual gourd forms were modeled. The latter so far have been reported only from Pueblo I and later sites. Superficially the types of pitchers found in both the late Basket Maker and early Pueblo periods resemble gourds with the stem-bearing tips removed and it is possible that they were so sug- gested. On the other hand, the influence may have been an indirect one. There is a marked similarity between many of the southwestern basketry water bottles of early periods and the pitcher forms. The basket containers are, in many instances, thought to have been pat- terned after vegetal models, and inasmuch as basketry had so decided an influence on ceramics the pitcher shapes may well be a survival of those forms rather than the result of an actual copying of gourd forms.7 This would illustrate the transference of an established functional form from one medium to another. This would seem to be a more plausible explanation for the origin of the shapes than to hold that immediately upon the discovery of the pottery-making art the ceramists, in casting about for something to pattern their vessels after, searched other fields for their models. In any case, the con- ventional form was the first to appear, and then came the true imitation. Handles on the pitchers are generally of the vertical loop variety, roughly circular or oval in cross section. Sometimes the fillets of clay were twisted and again two loops placed side by side were used. An occasional handle of the strap form is to be observed. The handles were attached to the neck either at the rim or slightly below it. Their juncture with the body is usually at a point just above the line of greatest diameter or shoulder. On some of the pitchers the handle is not fastened to the body but only touches it (fig. 18, 7, g); possibly reflecting some influence from the actual gourd forms. As a whole the pitchers of the Pueblo I period are more graceful in form and show an improvement in symmetry over those of the preceding period. Their surface finish is much better and is com- parable to that of the other vessels. This is in decided contrast to the late Basket Maker pitchers which are characterized by the extreme roughness of their surfaces. Pitcher sizes vary to the same extent as do those of the other vessel forms. The group illustrated in Figure 18 is so characteristic of the general size range that measurements of each one represented ™ Cushing, F. H., 1886, pp. 482-493. 98 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 will be sufficient to show the gradation in the specimens of the collection. Figure 18, a, has a total height of 127; inches (30.6 cm.) and a body diameter of 9;%; inches (24cm.). The neck height is 31% inches (8 cm.) and its diameter at the orifice 334 inches (9.6 cm.). The handle is of the strap variety and averages three-eighths of an inch (1 em.) in thickness and 2 inches (5.1 cm.) in width. The total height for 6 is 10 inches (25.4 cm.) and its body diameter 83°, inches (21.8 cm.). The neck height is 33% inches (9.1 cm.) and the diameter of the orifice is 342 inches (10 cm.). The handle joins the neck seven-sixteenths of an inch (1.1 cm.) below the rim. The handle is of the strap variety, three-sixteenths of an inch (5 mm.) thick and 14% inches (3.7 cm.) in width. The height of ¢ is 914 inches (23.4 cm.) and its body diameter 834 inches (22.2 em.). The neck height is 314 inches (8.3 cm.) and the diameter of the orifice 814 inches (8 cm.). The handle, which is of the oval type, is three-eighths of an inch (1 cm.) thick and 1 inch (2.54 cm.) wide. It was attached to the neck at the rim. One of the largest pitchers found was Figure 18, d. Its height is 1042 inches (27.8 cm.) and the average diameter of the body 975 inches (24.3 cm.). The neck height is 334 inches (9.6 cm.) and the average diameter of the orifice 43g (11.1 cm.). The handle, which is of the single loop twisted variety, joins the neck practically at the rim. Jt is nine-sixteenths of an inch (1.5 cm.) thick and three- fourths of an inch (1.9 cm.) wide. The pitcher outlined in e has a height of 77% inches (20 em.) and a body diameter of 64; inches (15.4 em.). The neck height is 2 inches (5.1 cm.) and the orifice diameter 23% inches (6 cm.). The handle is of the oval form and is attached to the neck five-eighths of an inch (1.6 cm.) below the rim of the orifice. The handle thick- ness is seven-sixteenths of an inch (1.1 cm.) and the width seven- eighths of an inch (2.25 cm.). The long-necked pitcher, f, has a height of 934; inches (28.3 cm.) and a body diameter of 714 inches (18.1 cm.). Its neck height is’ 34% inches (9.4 cm.) and the diameter of the orifice 214 inches (5.7 em.). ‘The handle is three-eighths of an inch (1 cm.) below the rim, is thirteen-sixteenths of an inch (2.1 cm.) thick, and eleven- sixteenths of an inch (1.8 cm.) wide. The smaller pitcher of the same form, g, is 54% inches (15.1 cm.) in height and the body diameter averages 34% inches (9.7 cm.). The neck height measures 2,°; inches (5.5 cm.) and the diameter of the orifice is 114 inches (3.2 cm.). The handle is attached to the neck at the rim. It measures one-fourth of an inch (6.4 mm.) in thickness and nine-sixteenths of an inch (1.4 em.) in width. ROBERTS ] PUEBLO RUINS IN COLORADO 99 The straight-necked form, A, is 73% inches (19.2 cm.) high and has a body diameter of 6 inches (15.24 cm.). The neck height measures 2;% inches (6.5 cm.) and the diameter of the orifice is 314 inches (8.9 cm.). The handle is joined to the neck seven-six- teenths of an inch (1.1 cm.) below the rim. The handle is nine- sixteenths of an inch (1.4 em.) thick and 1 inch (2.54 em.) in width. The smallest of the gourd forms, 7, is 334 inches (9.5 cm.) in height and has a diameter in the body portion of 3%; inches (8.1 em.). The neck is difficult to determine, but it has an approximate length of 23g inches (6.1 cm.). The diameter of the orifice is 1 inch (2.54 cm.). The handle is so tapering that measurements of it are not practical. The larger of the gourd forms, 7, has a 75g inches (19.4 cm.) height. The average body diameter is 6,4; inches (15.4 cm.). The neck height is more easily determined on this specimen. It is 3% inches (8.6 em.). The orifice diameter is 214 inches (5.7 cm.). The hollow handle on this pitcher is also tapering and measurements are difficult to determine. At the center it is thirteen-sixteenths of an inch (2.1 cm.) thick and 1 inch (2.54 em.) wide. Other examples of this form of pitcher range between the two extremes in size here given. Bowls in the nonculinary group are not characterized by a great variety of shapes. In general it may be said that there are three forms. The great majority of the vessels of this class are fairly deep in proportion to their diameter and closely border on the hemi- spherical shape. (Fig. 19, a.) Most of them have rounded bottoms, many showing traces of the basket in which they rested while in the process of manufacture, while the remainder have slightly flattened bases. The second form, 0, is too deep to be considered a hemispheri- cal type. Some of the vessels in this group are, in fact, almost straight sided. Bowls of this form constitute but a small percentage of the group, however. The third form, c, is of a shallow variety, the depth of the bowl being markedly less than half of the diameter. Most of the black-on-red bowls fall into this group, although the black-on-white ware is well represented in it. Regardless of shape, there are certain features which are charac- teristic of the bowls. There is a greater amount of asymmetry in these vessels than in any other form. For some reason there was a more pronounced tendency for the bowls to warp, possibly due to a faulty mixing of the paste. Hence, the containers are rarely a good circular form. They tend more to elliptical and oval or egg-shaped contours. In a few instances the warping resulted in what is known as the conventional heart-shape. This asymmetry is so apparent in 100 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 the illustrations of the bowls (pls. 25-81), however, that further dis- cussion of it is not necessary. The rims on the vessels are in most cases rather uneven or wavy. They may be considered as being of the direct variety, for outcurved or incurved rims have not been found. The lips of the rims are either rounded or slightly flattened. The a b 5 e€ ve or h 2 Oe v4 / ” 0 Pp q Ficurn 19.—Additional nonculinary vessel shapes vessel walls are somewhat thin for a short distance below the rim in a majority of the specimens; in fact they may be said to taper in to the lip. The interior of the bowls characteristically is more care- fully smocthed than the exterior. This may be accounted for on the grounds that the field for decoration occupied the attention of the maker to the exclusion of the other portions of the vessel. BUREAU OF AMERICAN ETHNOLOGY BULLETIN 96 PLATE 25 BLACK-ON-WHITE BOWLS SHOWING INTERIOR DECORATION BUREAU OF AMERICAN ETHNOLOGY BULLETIN 96 PLATE 26 BOWL INTERIORS,BLACK-ON-WHITE GROUP BUREAU OF AMERICAN ETHNOLOGY BULLETIN 96 PLATE 27 BLACK-ON-WHITE BOWLS BUREAU OF AMERICAN ETHNOLOGY BULLETIN 96 PLATE 28 BOWL INTERIORS, BLACK-ON-WHITE VESSELS BUREAU OF AMERICAN ETHNOLOGY BULLETIN 96 PLATE 29 EXAMPLES OF BAND DESIGNS ON BOWLS OF THE BLACK-ON-WHITE GROUP BUREAU OF AMERICAN ETHNOLOGY BULLETIN 96 PLATE 30 BOWLS AND LADLE e and f black-on-red vessels. SYOIY4SLX3A IMOG NO GALNIVd SANIT AO SAIdWV X43 l€ ALVTd 96 NILATING ASOTONHLA NYOINANY AO NVAYNN BUREAU OF AMERICAN ETHNOLOGY BULLETIN 96 PLATE 32 BIRD-FORM VESSEES ROBERTS] PUEBLO RUINS IN COLORADO 101 An occasional specimen is found with handles, but this is a com- paratively rare feature. One example has a single hooklike lug at- tached to one side slightly below the rim; another has two small vertically perforated lugs on opposite sides close to the bottom; some exhibit the small shelf-like projections suggestive of stubby bird tails; others have simple cone-shaped knobs. Bowl sizes show a considerable range. The smallest serviceable forms have diameters of 414 inches (11.43 cm.) and a depth of 214 to 21% inches (5.7 to 6.85 cm.) and the largest, excepting the tublike forms discussed in the culinary group where their measurements were given, a diameter of 954 inches (24.5 cm.) and a depth of 53; inches (13.5 em.). The great bulk of the bowls fall into the diameter range of 6 to 8 inches (15.2 to 20.3 cm.) and a depth variation of from 23/4 to 314 inches (7 to 8.9 cm.). The average wall thickness of the entire group shows but a slight fluctuation from three-six- teenths of an inch (5 mm.). Dippers and ladles show some differences in form, but they con- stitute a very small proportion of the nonculinary wares. The simplest shape is that of a shallow scoop. (Fig. 19, d.) This shape may be likened to a half melon or gourd. The horizontal contour is decidedly egg-shaped. This type of ladle is generally found in association with the early Pueblo horizons. The second form is characteristically that of a half gourd of the long-necked variety. (Fig. 19, e; pl. 30, 6.) There is no division or break between the bowl and handle portions. The third form is a modification of the second. It consists of a bowl and a handle, but the two are separated by a thin partition wall. (Fig. 19, 7.) The fourth form, g, has a solid handle. Ladles of this group suggest the preceding form, with the additional filling in of the handle portion so that it becomes a solid object. The form is distinct from the bowl with handle- attached types found in later periods. The handle lengths on both / and g may vary considerably. Some specimens will show the reverse of the conditions here illustrated. Several have been found with the short, stubby, paddlelike handle solid instead of being a continuation of the bowl separated from it by a thin partition, while others have a long handle of the g style which is partitioned off from the bowl by an intervening strip of clay which completes the circumference of the cup portion. An unusual double-bowled ladle with a short, solid bifurcated handle was found on the site of the A village in the preliminary investiga- tions during 1923.7° A good example of the concave handle, parti- tion form, from the district is illustrated in the report on the Colo- 7 Roberts, F. H. H., jr., 1925, p. 46, pl. 21. 102 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 rado institution’s first year’s investigations.”® Dippers and ladles on the whole occur so seldom in the pottery forms of the Pueblo I period along the Piedra that the lack of them might paradoxically be said to be one of their characteristics. Fragments from the d forms, Figure 19, indicate that they have a length range of from 41% to 51% inches (11.48 to 14 cm.). The diame- ter of the bowls varies from 2 to 3 inches (5.1 to 7.62 cm.). The range in bowl depth is between 1 inch (2.54 cm.) and 21% inches (6.35 cm.). The half-gourd-shaped ladle (fig. 19, e) has a total length of 934 inches (24.8 cm.) and a bowl diameter of 514 inches (14 cm.). The depth is 214 inches (6.8 cm.). Other examples of the form are too fragmentary to measure. The third type, 7, was in general quite small. The bowl portions range from 2% inches (7.3 cm.) to 314 inches (8.9 cm.) in diameter and from 1%; inches (3 cm.) to 18@ inches (3.5 cm.) in depth. Handle lengths vary from 1,3; inches (3 cm.) to 25g inches (6.7 cm.). Widths vary from nine-sixteenths of an inch (1.4 cm.) to 15¢ inches (4.1 cem.). The average handle thickness is five-sixteenths of an inch (8 mm.). The last form illustrated, g, has a 614-inch (16.5 em.) length. Of this total 4 inches (10.16 cm.) is handle. The average bow] diameter is 21% inches (6.35 cm.) and the depth 15¢ inches (4.2 cm.). Bird-form vessels furnish an interesting feature in the ceramic group. They first appeared in the closing days cf the late Basket Maker period and seem to have attained their greatest prehistoric prominence in that of Pueblo I. They are rare in Basket Maker III and although found in later horizons apparently did not occur to as great an extent as in the initial Pueblo stage. In the modern period a vast number of bird-shaped vessels have been made. This is particularly true of Zuni, but because of the kinds of birds repre- sented the feature seems to be largely the result of post-Spanish influences. Prehistoric specimens from the region show approxi- mately the same proportion of the bird forms as found elsewhere. The shapes in the Piedra collections vary from those which merely indicate the plump, oval body of a bird (fig. 19, h, 2) to actual effigies of them (fig. 19, %). Between these two extremes are many vessels with various features suggestive of the bird. On some there are the wings and the tail, on others just a tail. (Fig. 19, ¢.) Still others have bodies which are characteristically of the seed-jar form but which have a head, stubby wings, and tail of a bird. (Fig. £95(9; pl. 21,0.) 7 Jeancon, J. A., 1922, pl. xx1; d. ROBERTS ] PUEBLO RUINS IN COLORADO 103 Most of these forms from the Piedra district are extremely sugges- tive of ducks. The efligy vessel unquestionably represents that fowl. Just what significance they may have had is unknown, but it is possible that they were of ceremonial importance. When the effigy was found it was filled with finely ground corn, possibly sacred meal such as is used in many of the modern ceremonials. The seed jar with head, wings, and tail illustrated in Plate 21, 6, was filled with quartz crystals. The small number of vessels of this kind in propor- tion to the other shapes indicates that they were not in common everyday use. A suggestion as to a possible explanation for the form is to be found in one of the myths of the Zuni. The story is one phase of the tale which describes the wanderings of the people and their tribulations before they finally settled in the valley where their present village is located. Too much reliance should not be placed on the myths and traditions of the modern Pueblos in seek- ing explanations for the ancient cultures, but in the present instance at least an interesting sidelight is thrown upon the possible signifi- cance of the bird-form pottery. The myth mentioned relates that one of their important ances- tral personages was lost in the land of snow and cold in the far northwest; and, as he wandered blindly about, calling loudly for assistance, a duck heard him and came to his rescue. Eventually, after a series of misfortunes and with the aid of other creatures, the fowl led him first to the Spirit Lake, sometimes called the Lake of the Dead, where those who had been lost in the migration of the people lived. Thence he was carried, still accompanied by the duck, to the village where the Zuni had settled.8° The journey of this mythological personage from the Spirit Lake to the Pueblo is re- enacted every four years, when the young boys are initiated into the rites which assure their joining, after death, the ancestral group in the Lake of the Dead. During the ceremonies the story of the per- sonage who was lost in the north and of the duck which befriended him is related.* A part of the ceremonial paraphernalia used in the observances includes a stuffed duck. The latter fowl is considered by the Zufi as the wisest and most knowing of all creatures in the subject of travel routes. It may well be that the pottery representa- tions of that particular bird found in the beginnings of the Pueblo cultures symbolized some such mythological event. The vessel illustrated in Figure 19, 7, and Plate 32, 6, is of con- siderable interest. The general outlines are characteristically of the plump body of a bird. There is a short neck with a small orifice where the neck of the bird would be. At what would be the pos- 50 Cushing, F. H., 1896, pp. 407-414. 81 Stevenson, M. C., 1887, pp. 544-545. 88242°—30——_8 104 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 terior or tail end of the body in the approximate position of the cloaca there is a perforation through the side of the vessel. Mr. Morris found a number of vessels of this type in his work farther west and south, and he suggests that they represent a modified form of the vessels with lateral spouts, which are found only in the Basket Maker III period.*? He further mentions the possibility that bone tubes were set in the openings, with pitch, to take the place of clay spouts on the prototypes. Although there is probably no signifi- cance attached to it, these forms have the peculiar attribute of emitting a shrill whistling sound when placed in such a position that the wind blows against the lateral opening. Because of this char- acteristic they have occasionally been referred to as whistling jars. The most consistent form in the group, from the standpoint of dis- tribution, appears to be that of 7, Figure 19; Plate 32, @. Mr. Morris found them in the La Plata Valley and Navaho Reservation sites,** and Doctor Fewkes obtained one from Pipe Shrine House on the Mesa Verde.*! They are quite frequent in slightly modified forms in later periods.®® An occasional variation of 7 is secured and h also is found in later horizons. One example of the latter in a late pre- historic site was found by Doctor Fewkes at Chevlon Ruin in the upper Little Colorado region.’* Numerous other bird forms are found in the upper Gila region *7 and along the Rio Grande,** but all are from ruins of periods subsequent to that of the Piedra-La Plata specimens. Sizes vary. The jars with head, tail, and wings (fig. 19, 7; pl. 21, 6) are approximately the same size. The body height for 7 is 9 inches (22.86 cm.) and the diameter of the body 10 inches (25.4 cm.). The head rises 214 inches (6.35 cm.) above the top of the jar. Other measurements are not available. The specimen, Plate 21, 0, is 12 inches (380.48 cm.) from breast to tail and 81% inches (21.6 cm.) from wing to wing. From wing tip to wing tip, however, is 1014 inches (26.11 cm.) The body is 714 inches (19.1 cm.) in height. The head rises 2 inches (5.1 em.) above the top of the jar. The head, in its present condition with the beak broken off, measures 214 inches (6.85 cm.) in length. The wings are 2 inches (5.1 cm.) long and seven-eighths of an inch (2.3 em.) wide. They average one-fourth of an inch (6.4 mm.) in thickness. The orifice in the top of the jar is 814 inches (8.9 cm.) in diameter. ®2 Morris, H. H., 1927, p. 180. 83 Morris, E. H., 1927, p. 165; 1919 b, pl. 71, a. % Fewkes, J. W., 1923 a, fig. 107, e. 8 Holmes, W. H., 1886, figs. 354~358. st Fewkes, J. W., 1904, p. 67. 57 Hough, W., 1907, pl. vir. 88 Hewett, E. L., 1906, pl. x1v, c; Jeancon, J. A., 1923 b, pl. 38. ROBERTS | PUEBLO RUINS IN COLORADO 105 The form h has a length of 414 inches (10.5 em.), a width of 212 inches (7.2 cm.), and a height of 234 inches (7 cm.). The orifice is oval in contour with diameters of 1 inch (2.54 cm.) and five-eighths of an inch (1.6 cm.). (PI. 32, ¢.) The total length of 2 is 33g inches (8.6 cm.) and the height 21% inches (5.8 cm.). The orifice has a diameter of five-eighths of an inch (1.6 cm.). The effigy form, %, stands 5 inches (12.7 cm.) high and measures 414, inches (11.43 cm.) from breast to tail. Additional measure- ments for this vessel are not available, as it is in the State Museum at Denver, Colo. The last of the group, Z, has a total height of 37% inches (9.8 em.). The breast to tail measurement is 4% inches (10.9 cm.) and from side to side the body is 35£ inches (9.3 cm.). The orifice is seven- eighths of an inch (2.2 cm.) and the perforation at the end three- eighths of an inch (1 cm.) in diameter. Among the unusual forms is a group of ellipsoid-shaped vessels. (Fig. 19, m.) What their function may have been is not known. They were not numerous, only a half dozen having been found in the section, and none are very large. The body is long and nar- row, the ends tapering and rounded, and there is a small oval orifice in the top. Now and then a specimen has a low, squat neck, but these forms are rare. The form in general occurs sporadically throughout the San Juan area in about the same proportion as observed along the Piedra. The only specimen which is sufficiently intact to warrant measurements has a length of 514 inches (13 cm.) and a body diameter, through the center, of 2,%; inches (5.5 cm.). The orifice has its long axis coinciding with the long axis of the vessel. It measures 114 inches (3.2 cm.). The short axis is three- fourths of an inch (1.9 cm.). The average wall thickness is one- fourth of an inch (6.4 mm.). One of the most unusual forms found during the investigations and one which thus far is unique in the pottery of the San Juan area was that of a large black-on-red jar with a lateral orifice. (Fig. 19, n; pl. 83, 6.) What the purpose of this vessel may have been is purely conjectural. At first glance it suggests the so-called feather boxes of the modern Pueblos, and also the analogous pre- historic form,®® made from cottonwood. It differs from the latter, however, in that it has a marked tapering to the body and appar- ently was made to stand upright, as indicated in the drawing. Furthermore, the opening in the side is smaller than that usually found in the “ feather boxes.” The general shape and size of the 89 Nusbaum, J. L., 1922, p. 117. 106 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 pottery jar is very similar to that of many of the cylindrical vases from the Chaco Canyon. Here, again, there is a difference in that the openings of the latter are in the top, which in this specimen is closed. There seems little question but what this jar is related in some degree to the Chaco forms. The latter have been considered as being used in ceremonial observances and may have been made for the express purpose of holding ceremonial sticks as a part of the altar paraphernalia.®® The present example may well represent a modified and perfected form of the wooden feather box. The vases for holding prayer-feather sticks would, functionally, be closely associated with such a form. The latter vases might possibly be the outgrowth of such objects. They will be considered in more detail in connection with the next specimen (fig. 19, 0), which is charac- teristically of that form. The jar with the lateral opening has a height of 97% inches (25.1 cm.). The diameter at the top is 3% inches (8.6 cm.) and at the bottom 41% inches (10.5 cm.). The opening in the side also tapers from the bottom toward the top. Its width at the bottom is 134 inches (4.5 cm.) and at the top 1,4; inches (4 cm.). The length of the orifice is 242 inches (7.5 cm.). The walls of the jar are unusually thin, one-eighth of an inch (3 mm.), but the bottom and top average one-fourth of an inch (6.4 mm.). The discovery of a small cylindrical vase (fig. 19, 0; pl. 38, @) constitutes one of the high lights of the ceramic contribution from the region. Cylindrical vases have hitherto been considered a form peculiar to the Chaco Canyon, Pueblo Bonito especially, as well as a late development in pottery shapes.*t To find an example in a Pueblo I site at so great a distance from the locale of the type is of special interest. It shows definitely that such a form was an early Pueblo development. There can be no question of the con- temporaneity of this specimen and the early Piedra sites, as it not only has the characteristic ceramic features of the period but was found in house C of the Group 1 unit in the A village. The occur- rence of such a form at so great a distance from the main center is not as disturbing as it might seem, since the Piedra section gives such strong indications of affiliations with the Chaco. Whether the form originated in the northeastern portion of the range of the Chaco cultures and then spread to the main center, where it became a prominent feature, or whether it was the result of an influence from the latter can not be determined from present evi- dence. The situation as it now stands would suggest that the form was late in the Chaco, since it has been found only under conditions s0'Pepper,. G. H., 1920, ‘p. 377; *1 Pepper, G. H., 1920, pls. 2-6; pp. 117-122. ROBERTS] PUEBLO RUINS IN COLORADO 107 which would place it in the Pueblo III horizon. No example has come from a Pueblo I or II site in the canyon, but the lack of in- vestigations of a comprehensive nature in Pueblo I remains makes it inadvisable to draw definite conclusions at this time. The possi- bility of its development in the Piedra district was first suggested by Mr. Jeancon.*? Among the corrugated culinary vessels of the Pueblo II period obtained by the expeditions of the Colorado institutions are a number of the same general shape as the Chaco vases. They differ, however, in that they have rounded bottoms, although some of the Chaco specimens have slightly similar bases, and are of the culinary variety. Since the finding of the black-on-white vase and the quite similarly shaped black-on-red jar, the suggestion of such an origin for the type seems even more plausible. On the other hand, the origin of the form has quite generally been attributed to the Mexican area rather than to the Southwest. The Chaco forms are particularly comparable to the types of cylin- drical vases found in the highlands of Guatemala and many have thought that the southwestern forms were the result of a cultural influence from that portion of the higher culture center to the south. That there was commerce, an exchange of objects between the Pueblos and the Mexican area, is shown by the finding of copper bells and macaw skeletons in Pueblo Bonito.°* With this actual trading of objects it is possible that there may have been an ex- change of ideas and that the cylindrical vases were an outgrowth of that factor. The evidence of intercourse between the areas is of so much later a date than that represented by the Piedra forms, how- ever, that considering the problem as a whole it seems more likely that the cylindrical vases, like the making of pottery, were an in- digenous development in the San Juan culture area. As to the center of its origin in that range, conclusions must await further develop- ments in the Chaco Canyon itself. The Piedra specimen is not as large as most of those from the Chaco, but only slightly smaller than the smallest of that group. It is 514 inches (13.3 cm.) in height and has a diameter of 41% inches (10.5 cm.). The sides are straight, not incurved or tapering as in many of the larger and later examples. The average thickness of the walls is three-sixteenths of an inch (5 mm.). The double-lobed jar outlined in Figure 19, p, and illustrated in Plate 33, d, represents a form which is not unique but which is comparatively rare in collections. Several specimens have been found in the Piedra district and elsewhere as well, but this is the earliest stage in which it appears. Many similar examples are found 2 Jeancon, J. A., and Roberts, F. H. H., jr., 1928, p. 261. 93 Pepper, G. H., 1920, pp. 194-195. 108 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 in later periods. Even the potters of the modern villages make them occasionally, but they simply represent a survival of the shape. Such a jar really represents one globular vessel placed on top of another with the bottom of the upper omitted so that the vessel is actually one container. The present example is rather rough in finish and undecorated. It stands 6°; inches (16.7 cm.) in height. The upper body has a diameter 342 inches (9.7 cm.) and the lower 4; inches (11.8 cm.). The orifice diameter is 2 inches (5.1 cm.). The mountain sheep effigy (fig. 19, g) was found by the writer and his brother during one of the earlier expeditions in the region. Frag- ments from a number of similar vessels have been found, but thus far only the one complete specimen has been secured.** The figure is hollow and at one time had a handle attached to the rear of the head. The handle projected out over the body of the animal. At some time during the period when the vessel was in use the handle was broken off, the remaining stubs were rubbed down and the object continued in service. When found it was serving in the capacity of a mortuary offering. The vessel has a total height of 814 inches (20.9 em.), a length of 614 inches (16.5 cm.), and measures 414 inches (11.48 cm.) across the breast. The specimen is in the Colorado State Museum at Denver. A curious feature of the effigy forms is that the superficial portions, such as the eyes, horns, wings, and tails of the birds, were not molded as a part of the vessel but were stuck on after its completion. Fre- quently the pellet of clay representing the eyes was incised, some form of sharp instrument having been used to cut into the clay, to make them appear more lifelike. Although there is probably no con- nection whatever, it is interesting to note that this is the same treat- ment found in the features on the archaic figurines from the Valley of Mexico. Pottery effigies were fairly prominent in later Chacoan periods. Some are wholly unidentifiable, but others unquestionably represent deer, frogs, birds, and human beings. Very few complete specimens have been found, but many fragments are in the various potsherd collections from the large ruins. There is only one vessel found along the Piedra which has not been described and which is not included in the outline drawings. It is a unique specimen and a decidedly eccentric form. (PI. 33, ¢.) It was molded, not coiled, and has a very crude finish. Possibly it was the work of a beginner in pottery making; perhaps represents the efforts of a child. The form is more suggestive of a modern candle- stick than anything else. Its chief value is as a curiosity. The object *! The duck and sheep effigies are illustrated in the report for the investigations of 1923. See Roberts, 1925, pl. 20. et ALVId 96 NILATING *poyetooapun ‘p pu d Spor WO-Yovyq ‘g Soy M-UO-yory_ ‘Dd P P I p Iq I la STASSAA WYHOS-DIYLN3AD0F ANV AVSDINYGNITAD ASO TIONHLA NYOLYAWNY 40 NVAYNE BULLETIN 96 PLATE 34 BUREAU OF AMERICAN ETHNOLOGY MINIATURE POTTERY FORMS BUREAU OF AMERICAN ETHNOLOGY BULLETIN 96 PLATE 35 MINIATURE VESSELS ROBERTS ] PUEBLO RUINS IN COLORADO 109 stands 65g inches (16.8 cm.) high and the diameter of the body is 436 inches (11.1 cm.). The forms of the miniature vessels are so lke those of the larger containers that there is no necessity for a detailed discussion of them. The forms include bowls, jars, pitchers, and ladles, and an occasicnal bottlelike shape. (Pls. 34, 35.) The majority of them are rough in finish and do not give indications of the coiling or band technique. They were merely molded out of clay. The majority of those illus- trated came from burials, but an occasional one was found in a house pit or in the refuse mound of the unit or village. The only factor of special interest in connection with this group is the large numbers of them found. The proportion of such objects in the Piedra sites is much higher than ordinarily is found to be the case. DESIGNS ON THE PAINTED POTTERY Decorations on vessels made during the Pueblo I period along the Piedra River were placed on the interior of bowls and ladles and on the exterior of the various forms of the nonculinary jars, pitchers, bird shapes, and eccentric forms. Rarely is any ornamentation found on the exterior of bowls. This is in contrast to the vessels of this class found in some sections of the San Juan in later periods, particu- larly Pueblo III, where the decoration was frequently extended to the exterior surface. The use of designs on pitchers, seed jars, large storage and water jars, bird forms, vases, and eccentric shapes sets this period off distinctly from the preceding—Basket Maker IJJ— where painted ornamentation was restricted mainly to bowl interiors and occasional ladles. It was not until quite late, practically the very end of the Basket Maker cycle, that a few meager and experi- mental decorations were applied to pitchers and other nonculinary forms.®* Even in the Pueblo I stage the ornamentation of such vessels was not as elaborate, on the whole, as that on bowls, and in the majority of cases the designs do not exhibit the dash and freedom from restraint which more or less marks those on the latter group of containers. As in the designs on the Basket Maker ITI wares, the true character of the decorations of the period is more typically illus- trated by the bowls than the other forms, although a few of the latter do have characteristic ornamentation. There is one rather striking feature to be noted in the decorations on pottery. The influence which basketry had upon the ceramic in- dustry has already been mentioned in the discussion of the develop- ment of pottery forms. This factor is even more marked in the mat- ter of design. The earliest of the decorations show the geometrical characteristics imposed by textiles in the weaving of their patterns. % Morris. E. H., 1927, p. 189. 110 BUREAU OF AMERICAN ETHNOLOGY [BULL, 96 This is not the most significant factor, however. What impresses the psychology of primitive man upon the student is the manner in which the designs were used. In the late Basket Maker period they are essentially textilelike in character, but in addition are of the kind found only on baskets. The potters of this period had great re- sources for design elements in the elaborate patterns woven in the sandals and twined bags, headbands, and other objects of the textile art, but did not use them.%* Due to the similarity of purpose, to the fact that ceramics replaced to a considerable degree the baskets, the pottery was decorated only with the basketry designs. In the minds of the earliest potters there presumably could be no interchange of factors between groups of unrelated usage. In the following period, the one with which this paper is most concerned, Pueblo I, the cera- mists began to cast about for new elements, and designs from other textile forms came into use. Perhaps this is one of the factors attrib- utable to the Pueblo peoples. At any rate, there is the record of the old conservative practice of certain things for certain purposes giving way to a wider point of view and the adopting of new ideas. The discussion of designs which is to follow will attempt no in- terpretation of the so-called symbolism of decoration. No effort will be made to discover what esoteric, mythological, or mystical factors might conceivably be concealed in the various combinations of ele- ments. Such a treatment of prehistoric ornamentation would be so largely a matter of pure guesswork that it is not deemed advisable or essential. A study of that nature belongs to the field of speculative art, not that of practical archeology where stylistic features and gen- eral characteristics are sufficient to an understanding of period and cultural differences. The ornamentation on the vessels is predominantly of the recti- linear geometric style, but curvilinear and realistic figures do occur. As a matter of fact, life forms are present more frequently on vessels from the Basket Maker III and early Pueblo I stages than in any of the later pre-Spanish phases with the exception of the outstanding examples of such figures found on the vessels from the Mimbres Val- ley in southern New Mexico. The main point of emphasis with re- spect to the realistic figures in the designs should be placed on the fact that from the very beginning of the manufacture of painted ves- sels both forms were used in the decorations, and they existed side by side, so to speak. It has previously been pointed out that one did not grow out of the other or result from it in the early development of southwestern pottery designs,®” but this is so contrary to the general widespread belief that attention should again be called to it. Certain life or realistic forms did become conventionalized in very late 2’ Morris, W,. H.,. 1927, p: 197. 7 Roberts, H.W. H., jr. 1929, p:, 122: ROBERTS ] PUEBLO RUINS IN COLORADO 1h periods, but this was only after many centuries of experiment and development in the field of pottery decoration. The main elements in the designs on the vessels from the Piedra villages, and this is also true for the period in general, consist of dots; stepped or zigzag and squiggled lines; ticked or dotted lines; solid triangular figures; dotted triangles; triangles with rectilinear, or rarely curvilinear, tips, elements of the fret decoration; circles; and checkerboarding. ‘These comparatively simple elements were used in an almost countless number of combinations and, although the designs fall into certain definite stylistic groups, no two decora- tions are exactly alike. The field for decoration on bowls, the concave inner surface, was treated in a number of ways. One group of designs consists of a single panel passing through the center bisecting the bowl. In another group the field is marked by a tripartite division usually ‘ obtained by the use of rectilinear or curvilinear spiral radii. A third form has a quadrant treatment. The latter was obtained from the first form by passing a second panel, series of lines or single line, across the field at right angles to the first figure. In both the tri- partite and quadrate forms the sectors may contain additional fig- ures. Another form or plan of decoration comprises the group of vessels with figures which are pendent from the rim but do not cross the center of the field; do not go to the bottom of the bowl. In this group are two panels placed at opposite sides, three figures occurring at fairly regular intervals, or rarely four panels. Figures of this form may be either rectilinear or curvilinear, at right angles to the rim or cutting obliquely across the field for decoration. Another arrangement consists of two or four figures placed on opposite sides at some distance below the rim. The last major group consists of an encircling band extending all the way around the sides of the bowl. The band may be just below the rim, midway down the sides of the vessel, or practically circling the bottom. There are certain varia- tions and combinations of the several plans described and occasion- ally there is what might be termed an all-over decoration. one which practically covers the entire surface of the bowl. On the jar and pitcher forms there are two zones of decoration, one around the shoulder or line of greatest diameter, and the other on the neck. The designs in general in this group consist of an encircling band made up from elements already mentioned. On an occasional form the decoration consists simply of a single figure repeated on opposite sides of the vessel. Some of the water jars have two panels placed on opposite sides of the neck. The seed-jar forms have the decoration placed on the flattened top. The field for design on these vessels is quite comparable to that of the bowls, except that it is slightly convex instead of concave. Interestingly | 3 BUREAU OF AMERICAN ETHNOLOGY [BULL 96 enough, it was treated in much the same fashion as the interior of the bowls. BLACK-ON-WHITE WARES A number of characteristic examples of each of the several types or styles of decoration mentioned in the preceding paragraphs will be considered in some detail. A lengthy discussion or description of all of the designs represented in the collection is beyond the require- ments of the present paper, but a sufficient number will be presented to show the typical forms. Because of the predominance of the black-on-white wares, they will be considered first. The color of the pigment in this group, as previously mentioned, varies from a brownish-red to a good black, practically the whole range between these two extremes being represented. The major- ity tends to a brownish tinge, the result of a lack of control in the firing of the vessels. Several of the vessels in the Piedra group exhibit a rather inter- esting feature in the pigment aside from that of color variation. On certain specimens there is a slight vitrification, a glazing, so to speak, in some of the lines of the decorations. 'Vhis vitrification was no doubt due to some element in the material from which the paint was made. The heat of the fire in which the pots were baked melted and fused some of the mineral, producing a glasslike sur- face on certain portions of the decoration. In most cases this glaze has a slightly greenish tint to it. The effect was unquestionably accidental, but is of interest because of the early stage which the vessels represented. Some investigators have held that glazing in the designs was unknown before the arrival of the Spaniards. It is true that it did not come into widespread use until late in the Pueblo cycle, Period IV, but here is direct evidence that the possibility of such a feature was apparent at the very beginnings of the culture. The decorations on many of the vessels are too indistinct to show well in photographs, and for that reason it has been deemed essential to reproduce them in the black and white drawings. The text fig- ures err, for the sake of convenience in reproduction, in the bowl and jar top outlines, which in rare cases only approach the accurate circular form indicated. The actual designs, however, follow closely the originals. The general tendency in illustrating pottery designs is to idealize them, and as a result much of the actual Indian char- acter is lost. Even though the temptation to improve was great, the accompanying text figures reproduce exactly the irregularities and mistakes, the crudity of line, present in the actual decorations. Strange as it may seem, it is harder to copy the irregularities than it is to make a more perfect drawing. ROBERTS ] PUEBLO RUINS IN COLORADO 113 The group of designs consisting of a single panel or figure bisect- ing the field for decoration has a great many variations. The figure may be quite simple or very elaborate. One of the plainest designs falling in this group and one which is quite characteristic not only of this period but also of the late Basket Maker ornamenta- tion is illustrated in Figure 20, a. As will be seen from the drawing, a zigzag figure consisting of four more or less parallel lines cuts its way across the center of the concave surface. The brushwork was rather crude and the angles are more rounding than sharp. ‘There is no question but what the undulating or squiggled line which is so frequent in the designs of later periods originated in just such a fashion from the rectilinear form which had been imposed by the basketry technique. The pigment in the design had been applied directly to the smoothed surface of the vessel. There was no slip. a FIGURE 20.—Zigzag line elements in bisecting panel designs The color of the pigment is a decided brown. The whole aspect of the vessel is such that under ordinary circumstances it would readily be taken for a Basket Maker III bowl. Another simple variation of the use of zigzag lines in a panel figure is illustrated in Plate 25, a. The second bowl decoration on Figure 20, 6, does not strictly belong in the group under consideration because it has two figures instead of one. Since it represents an amplification and elaboration of the single zigzag element, it is included at this point for com- parative purposes. Here there are two zigzag elements, each com- posed of seven parallel stepped lines, cutting across the center of the bowl and almost filling the field for decoration. This design represents a later development and is one which on the whole may be considered quite characteristic of the early Pueblo period. The lines in the decoration show better drawing, a surer touch than those of the preceding example. In addition the pigment is a better black 114 BUREAU OF AMERICAN ETHNOLOGY (BULL. 96 and the bowl had a thin slip. The vessel was materially blackened during the burning of the house in which it was found and at the present time the lines in the design are rather hard to trace, al- though there is no doubt concerning the form of the ornamentation. Slight variations of this type of design are frequently encountered on vessels of this and the following period. The form is even found on an occasional red-ware bowl. Two very crude attempts at decoration are illustrated in the designs Figure 21, a, 6. Both are examples of the division of the field for decoration into halves by a single figure or panel cutting across the center of the bowl. The first, a, has an additional treat- ment in that the extra figures were placed on the opposite walls of the divided interior. The bisecting element is simple in the ex- a b FIGURE 21. Simple forms of the bisecting panel plan of decoration treme, consisting as it does of a single broad line with a circle at the center or bottom of the bowl. This circle at the bottom of the bowl is a feature which is one of the characteristic elements of the Basket Maker IIT designs * and one which unquestionably is a sur- vival from basket ornamentation. Both the zigzag line and circle are to be found on the baskets from Basket Maker sites.°° What significance, if any, the small circle may have had is not known, but it has frequently been suggested that it may have symbolized the same thing as the sipapu in the kiva—namely, the mythical place of emergence of the ancestral peoples. The two figures on the walls of the bowl are simple broad lines with a series of dots along the lower edge of each. On one of the lines there are seven dots while the other has eight. It does not seem that any special significance can be attached to the numbers. 98 Morris, BE. H., 1927, p. 193; Roberts, F. H. H., jr., 1929, p. 119. ® Guernsey, S. J., and Kidder, A. V., 1921, pl. 24; Morris, E. H., 1927, fig. 42,*as 43,-a; 0; Cc, d. ROBERTS ] PUEBLO RUINS IN COLORADO 115 The effect in each case is the same regardless of the difference in number. The pigment of both the central and side figures is a fairly good black with a slight suggestion of a glaze. The surface of the bowl did not have a slip. The second design (fig. 21, 6) is a single panel composed of five approximately parallel lines passing through the center of the bowl. The outer lines have additional figures in the form of the slightly leaf-like elements. The latter were probably not intended to rep- resent leaves, however, but are more likely what resulted when the potter tried to paint a series of triangular figures with tips, the so-called bird symbols, of the form so frequently found on the vessels of this period. ‘The specimen itself gives the impres- sion that too hasty or especially unskilled brushwork was responsible for the form which the elements took. The pigment in this decora- a b FIGURE 22.—Bisecting panel and quartered deSigns tion is a rather faded black. Before it was applied to the interior of the bowl the bowl was given a coating of “ liquid” clay—a slip. Two very similar designs are illustrated in Plate 25. One of them, ¢, is quite well done and shows a fair amount of skill with the brush. The other, 7, has practically the same characteristics illustrated in the example depicted in the drawing. There was an amplification in the design illustrated in the photograph, how- ever, in that it has the border lines, with the stepped treatment, around the wall just below the rim. The bisecting panel is quite comparable to the one just described—in fact, suggests that it might have been made by the same potter. Both of the latter bowls (pl. 25) were slipped. One especially well illustrates the thick and thin feature described in the discussion of the slip in general. It is ¢ of Plate 25. Still additional examples of the same crude style of painting are illustrated in Figure 22. The figure in a may be considered either 116 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 as a single panel cutting across the center of the field for decora- tion or as two similar panels meeting a circle at the center or bottom of the bowl. The design is one which is so characteristic of the Basket Maker JII period that it might have come from such a site rather than a Pueblo I house along the Piedra. A closely com- parable decoration is on a bowl which came from a late Basket Maker site in the Chaco Canyon.t. The main difference between the two is in the series of zigzag dots which fill the panels in the Chaco specimen. The pigment of the design in the present example (fig. 22, a) is a fairly good black and the bowl was treated with a slip before the decoration was painted on it. The second design (fig. 22, 6) rightfully belongs in the group of quadrate decorations but has been included at this point because of its general similarity in technique to the forms just described. As will be noted, it also has the small central circle from which the other figures radiate to the rim. What the latter may have been intended to represent it is not possible to determine. They definitely show a departure from strict basketry forms of design, however. The de- lineation is far from good. Most of the brushwork in this group (figs. 21, 22; pl. 25, b, f) suggests that the artist was using too thin a pigment and had too much paint on the brush when the painting was done. They are in distinct contrast to some of the other deco- rations which show a careful and quite precise handling of both the brush and the pigment. To a certain extent the crudity of these specimens may be attributed to the possibility that the potter was a beginner and had not yet mastered the technique of decoration. The pigment in the design Figure 22, 6, is thin and grayish black in tone. This characteristic may be due in large measure to the fact that no slip was placed on the vessel and the pigment was so thin that a part of it sank into the paste of the vessel. The latter feature is fre- quently found on vessels of the Pueblo I period. It is so marked im some sections that it has been considered one of the characteristics of the period. ‘wo examples of the use of a more complicated device in the form of the bisecting panel are illustrated in Figure 23. Both illustrate characteristics which are typical of the normal phase of the Pueblo { pottery designs. The main elements are parallel stepped or zig- zag lines, dots, bird figures, and in one the use of the squiggled or wriggly lines. The central figure in @ might well be considered to consist of two panels pendent from the rim of the bowl and extending down the sides to the center. Both are alike in that they are made up of a series of nine parallel lines, are stepped, and are bordered 1 Roberts, F. H. H., jr., 1929, pl. 16, a. ROBERTS ] PUEBLO RUINS IN COLORADO 117 along one side and the lower end by rows of dots. The band of tipped triangular figures, of the so-called bird-symbol form, around the walls of the bowl just below the rim are decidedly characteristic. They constitute one of the chief elements in the designs of the de- veloped phase of the Pueblo I period. Rarely is one found on a late Basket Maker vessel, and although they are of quite frequent occur- rence in later stages they can not be considered as typical of those horizons as they are of the early part of the Pueblo era. The partic- ular bowl on which this design was painted was found in one of the burned houses and as a result of the terrific heat to which it had been subjected in the destruction of the building the pigment was burned to a good red color. Because of this the bow] is quite striking in its appearance, but the pigment probably was originally a varia- tion of black. In some of the minor rites preceding the return of the Katcinas it is used to reflect a ray of sunlight through the kiva hatchway and onto the sun symbol on the altar, as well as to reflect sunlight into the medicine bowls.7¢ “4 White, L. A., 1928, p. 564. *6 Fewkes, J. W., 1920, p. 501. 26 Wewkes, J. W., 1920, p. 517. ROBERTS | PUEBLO RUINS IN COLORADO 155 At Zui three crystals are placed on the altar of the Little Fire Fraternity.** It is possible, but of course the question is purely conjectural, that the above uses of the crystal may represent an ultimate development of features which had their beginnings at the very dawn of the Pueblo cultures; that in the specimens from the Piedra graves we have some of the earliest examples of the cere- monial quartz crystal. SHELL ORNAMENTS The only objects of shell obtained were in the ferm of four bracelets (pl. 54) which were on the left arm of one of the skeletons. This form of bracelet is very widespread in its distribution through- out the Southwest. It becomes more and more common toward the south and west, in fact usually is found in considerable numbers in sites throughout Arizona. Examples are only occasionally found as far north and east as the Piedra district. The bracelets were made from glycymeris shell (Glycymeris giganteus Reeve, 1843), which is found in the waters of the Gulf of California from Magdalena Bay, Lower California, to Mazatlan, Mexico.*® They are another example of trade relations. They may have been bartered several times and passed through a number of hands before reaching their ultimate owner, or they may have passed directly from an Indian of the Gulf of California region to the Piedra resident on whose skeletal remains they were found. Their exact story, unfortunately, can never be known. It would be en- tirely in keeping with Indian nature and with what is actually known of recent peoples to postulate that their owner obtained them while on a journey to the western ocean. On the other hand, it is just as possible that they passed from Indian to Indian until, after many exchanges, they finally reached southwestern Colorado. HUMAN BURIALS The early Pueblo peoples along the Piedra River interred their dead, in a great majority of cases, in the refuse mound of the house- group unit or of the village. Out of the 104 burials investigated only two were in other locations. One of these was on the edge of the bluff overlooking the river at the north side of an A unit on Stollsteimer Mesa and the other was beneath a large stone slab under the floor of house A in group 3 of the large village. ‘There were certain features about the burials as a group which are of interest in so far as they show a common treatment, variations in the form 27 Stevenson, M. C., 1904, p. 551, pl. cxxvIl. °3'The shell was identified and information furnished by W. B. Marshall, assistant curator of mollusks, U. S. National Museum, 156 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 of interment, or a lack of definite practice. In general it may be said that the body was placed in a contracted position, lying on the side, and accompanied by mortuary offerings of pottery, and occasionally stone and bone implements. The form of contraction observed was not so much that of the so-called natal position, the knees drawn up to the breast, as a folding of the legs. The lower limbs were drawn up tight against the upper ones, which extended at approximately right angles to the body line. Now and then the contraction of the lower limbs was not particularly marked, but in not one example were the thigh bones found in a position indicat- ing that they had been drawn up against the body. In contrast to other sections, the skeletons were only occasionally covered with large stone slabs. The whole group showed the use of such slabs in only 5.8 per cent of the interments. No great significance can be attached to the question of the orien- tation of bodies. If any one direction was more favored than the others in the placing of the head it was the north. Also, those interred with the head toward some other point of the compass in a fairly large number of cases had had the face turned in a northerly direction. This might be considered as suggesting that the north had a certain symbolical importance in the minds of the group. Some students of the region and of the mythology of the Pueblos hold that the north is a direction of great significance to that people because of their traditions of having originated and once lived there before migrating southward to the region they now occupy. If such was true in the Piedra settlements the importance could not have been particularly marked because of the number buried with the head and face toward other regions. The burials showed 33.7 per cent with the head toward the north, 21.1 per cent toward the east, 13.4 per cent toward the south, and 13.4 per cent toward the west; not determinable, 18.4 per cent. The latter were burials which had been disturbed by badgers or other burrowing ani- mals, and for that reason the head direction could not be determined. Of the group with heads placed in other directions 42.2 per cent had the face turned toward the north, in 30.7 per cent of cases it was toward the south, in 19.3 per cent the head faced the east, and in 7.8 per cent it was toward the west. The directions used in working out these percentages were not the absolute points of the compass, but the range from northwest to northeast stood for north, northeast to southeast for east, and so on. The side on which the body was most frequently placed shows about the same variation as observed in the head orientation. The skeleton was lying on the left side in 41.7 per cent_of the burials, on the right side in 33.7 per cent, on the face in 2.9 per cent, on the ROBERTS ] PUEBLO RUINS IN COLORADO 157 back in 2.9 per cent, and in 18.8 per cent the position could not be determined because of disturbance. The total number of skeletons showed 17.3 per cent children, 32.7 per cent adult males, 34.4 per cent adult females, and 15.6 per cent adult remains which could not be sexed. The percentage of children seems rather small in comparison to that of the adults, but it is quite probable that the bones of many of the infants did not with- stand the agents of decay as well as those of the older children and adults did. Infant mortality is usually quite high among peoples of lower cultures, and it would seem that there should be a larger proportion of such remains than actually was found. That the number is small is more likely due to the cause just mentioned than to a lower death rate among the infants in this particular group. The skulls of all but six individuals exhibited the characteristic occipital deformation of the Pueblo peoples. The flattening did not occur as regularly on the back portion of the head as is the case in later periods, but was observed more frequently on one side or the other. Perhaps the cradling practice which produced such re- sults was not fully perfected at this stage and as a consequence the flattening was not as well controlled. Another point which sug- gests itself, but one which can neither be proved nor disproved at the present time, is that possibly many of the inhabitants were of the old long-headed strain and because of a more pronounced occip- ital protuberance the head was more inclined to roll to one side than to le flat, as would be the tendency with the rounder forms. If such was the case, that is a possible explanation for the side flattening. The displacement and modification caused by the de- formation made it impossible to ascertain the original head form, however. Some of the skulls gave a distinct impression of being long-headed despite the deformation. The six undeformed crania were characteristically of the long, high crested form which super- ficially could not be distinguished from Basket Maker skulls. There unquestionably was a mixed group in these villages. The in- dications point strongly to an amalgamation of Pueblo and Basket Maker peoples. Not only is this suggested by the skeletal remains but, as shown in the discussion of pottery forms, by cultural fea- tures as well. The combination burial and refuse mounds were so similar in their general characteristics and the interments as a whole were so typical that a detailed discussion of each is wholly unnecessary. So that the reader may have a clear understanding of the nature of the mounds and the position of the burials in them, however, one typical example will be considered. It was the refuse heap for one of the A house group units. 158 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 Before investigation there was little indication on the surface that any considerable amount of refuse had been placed there. As a matter of fact the edges of the deposit blended so imperceptibly with the surrounding ground level that the average person would have passed over it without noticing even the slight rise near the center. The top was somewhat darker in color than the natural earth, due to the ash content, and a few potsherds and stone flakes were scattered here and there. As soon as trenching activities were begun the reason for the apparent lack of depth in the mound was discovered. The refuse from the houses had been deposited in a small natural gully and most of the material was below the sur- rounding ground level. (Fig. 40.) The deposit may originally have risen above the ground level to a greater extent than noted at the time of the excavations and have been packed down or washed away by surface water, but even if it had not considerable refuse had been deposited there. By tracing out the edges, through the slight difference in earth color, it was found that the mound was slightly oval in shape with a north-south axis of 65 feet (19.812 m.) and an east-west of 87 feet (26.52 m.). The depth near the center, in the range of the gully bottom, was 4 feet 6 inches (1.3716 m.). The content of the mound was no different from that in any of the deposits mentioned in the discussion of the house-group units. It was chiefly sand, ashes, broken stones, bones, and potsherds. The burials, with one exception, were clustered in the central portion of the mound. (Fig. 40.) This feature was found to hold true, in the main, in all of those examined. This can be explained on the basis that the dead were interred in the deepest part of the mound. There is no good explanation for the isolation of the burial No. 1 in Figure 40. Any number of reasons might be postu- jated to account for the exclusiveness of the particular individual, but none could tell, as far as we know, definitely why the body was so placed. The characteristics of the various interments may be briefly summarized. No. 1, an adult male. Lying on the left side with the head toward the east, face to the south, 1 foot (380.48 cm.) below the surface. The lower limbs were drawn up against the upper, the femurs at approxi- mately right angles to the body line. The arms were bent at the elbows, lying side by side in front of the body. At the head were two seed jars and a pitcher. One of the seed jars was black on red. It contained a large assortment of stone flakes, small fossils, bits of copper ore, a large spearhead, and a stone pipe. The collec- tion suggests that of a medicine man. This may account for the isolation of the burial. ROBERTS | PUEBLO RUINS IN COLORADO a ay i} Mp, we ny! 9 ee 4 & SS &, & fo 50 FEET TO ZB HOUSE GROUP zS Z \\ pageetl™ AHA gga 2 > 0% // € (73 4 /4 = “hs ge 16 2 £5. 2 . & * . = 26 + £ S “se 2 jt \\ i se < inst % Y, wr a | ( r av” aan bey Manes apy lt MM YY Yf ]j/ SECTION OF MOUND THROUGH GRAVES 6,6, 7 AND 8. Ficgur® 40.—Distribution of burials in typical refuse mound Tea 159 160 BUREAU OF AMERICAN ETHNOLOGY fecny, 96 No. 2, an adult male. Buried 2 feet 6 inches (76.2 cm.) beneath the surface. Jying on the right side with the head to the east and face to the north. The legs were flexed, the left more tightly than the right. The right arm was bent close to the body, the hand almost touching the face, while the left was extended at approxi- mately a right angle to the body line. No mortuary offerings ac- companied the skeleton. No. 3, an adult female. Buried 2 feet (60.96 cm.) below the sur- face. ying on the right side, head to the east and face to the north. Lower limbs tightly flexed, femurs extended at right angles to the body. Both arms bent at the elbow and drawn against the body, the right hand beneath the face and the left in front of it. Bones in an advanced stage of decay. Mortuary offerings consisted of a small culinary vessel placed in front of the face and a bowl at the back of the head. No. 4, an adult female. Skeleton 1 foot (80.48 cm.) below the top of the mound. Body had been covered with a layer of river bowl- ders. Lying on the left side with the head slightly south of east, face toward the south. Lower limbs moderately flexed. Arms slightly bent, lying side by side and extended in front of the body. Zones in a poor state of preservation. Five vessels had been placed in front of the knees—two bowls, one small culinary jar, a seed jar, and a pitcher. No. 5, an adult male. Bones 18 inches (45.72 cm.) beneath the surface. An undeformed long head. Lying on the left side, head to the southeast. ‘The legs were moderately flexed, the femurs ex- tended at an angle of about 60° from the body line. The left arm was extended along the body, the hand lying under the left femur. The right arm was bent at the elbow and extended in front of the body. A mortuary bowl was placed just back of the pelvis or hips. No. 6, an adult, probably female. Was 2 feet (60.96 cm.) below the surface. A disturbed burial, leg and body bones all piled to- gether. Head was toward the east, face to the north. Accom- panied by one bowl placed near the head. No. 7, an adult male. Interred 1 foot 6 inches (45.72 cm.) be- neath the surface. Lying on right side with the head to the north- west, facing the southwest. Leg bones had been disturbed but body and arms were articulated. The right arm was bent at right angles io the body, the left only slightly extended in front of the leg bones. No mortuary offerings accompanied the remains. The disturbance in this burial and that of No. 6 might be attributed to burial No. 12, which is in close proximity. If the latter had been interred after Nos. 6 and 7, the latter might have been disturbed by the gravedig- ROBERTS | PUEBLO RUINS IN COLORADO 161 gers preparing the pit for No. 12. Indications strongly pointed to an earlier deposition for No. 12, however, so that there is no ex- planation for the condition of Nos. 6 and 7. There were no traces of badger burrows in that part of the cemetery and no indications on the surface that the mound had ever been dug into. It is, of course, possible that in prehistoric times some one may have dug into the graves, disturbed the bones, and then re-covered them, the time elapsing between that occurrence and the recent exhumation being sufficient to obliterate all traces of the earlier work. No. 8, an adult male. Was buried 2 feet 6 inches (76.2 cm.) be- low the surface and was on the old ground level. Lying on the right side with the head toward the northwest, face to the south- west. Legs were flexed, but the left femur was drawn up to a greater extent than the right, so that the lower left leg rested on both the upper and lower right leg bones. The right arm was bent at the elbow so that the lower arm was extended in front of the body. The right was only slightly bent and the hand was just above the left knee. The skull was covered by a small stone slab, a pot lid, and a larger one was over the body portion of the skeleton. A single broken bowl placed in the space between the left knee and the pelvis and the left arm constituted the only mortuary accompaniment. No. 9, an adolescent child. Buried 1 foot (80.48 cm.) beneath the surface. Lying on the left side with the head to the northwest, face to the north. A badger or similar burrowing animal had dug through the legs of the skeleton, disturbing the bones. The right arm was bent at the elbow and the forearm extended practically at a right angle to the body line. The left arm had been extended along the left side. No mortuary offerings accompanied the remains. No. 10, an adult male. Interred 1 foot (30.48 cm.) below the sur- face. Skeleton was covered with river bowlders. Lying on left side with the head to the east, face to the south. Lower legs tightly flexed against femurs. Upper legs at right angles to the body line. Left arm bent at elbow and forearm extended at right angle to body. Right arm not flexed to so great an extent. Two bowls accompanied the burial. The smailer one was placed upside down inside the larger. Possibly contained food which was covered to protect it from the dirt. Both bowls were placed at the back of the head. No. 11, an adult male. At a depth of 2 feet (60.96 cm.) be- neath the surface. Was lying on the natural soil at the bottom of the refuse. On left side, head to the north, face to the east. Lower legs not as tightly drawn up as in most burials. Left femur was flexed so that the knee was drawn up, the right extended to 162 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 form a right angle with the body. Arms bent at elbows, hands in front of thorax. Bones in an advanced stage of disintegration. Skull crushed by earth. Mortuary offerings consisted of a bowl, half of a nonculinary pitcher and a fragment from a culinary jar. No. 12, an adult female. Skeleton 3 feet (91.44 cm.) below the surface. Stone slab over head and body. Remains lying on left side, head to the northeast, face to the southeast. Lower legs only moderately flexed. Femurs formed an obtuse angle with the body line. Arms bent at elbows, forearms forming acute angles with body line. Small culinary jar in front of face only mortuary offering. No. 13, an adult female. Bones 1 foot (30.48 cm.) below the ground level. Lying on left side, head toward the northwest, face slightly east of north. Lower limbs loosely flexed. Femurs formed an obtuse angle with body line. Lower right leg bones crossed, lower left just above ankle. Arms slightly bent at elbows, left a little more than right. Bowl in front of face only offering. No. 14, an adolescent child. Remains were 2 feet (60.96 cm.) below the surface. Lying en the left side, head toward the east and the face toward the south. Lower limbs moderately flexed. Right knee rested on left tibia and fibula. Left femur was at right angle to body line, right formed a slightly obtuse angle. Arms bent at the elbows, hands in front of the chest. Right arm lying on top of left completely covering it. Two mortuary bowls had been placed in the grave with the body. They did not occupy positions similar to those noted in the majority of cases. Instead of being at one side or near the head, they rested on top of the body. No. 15, an adult female. Bones 1 foot 6 inches (45.72 cm.) below the surface. Lying on left side, head toward the west, face to the north. From just below the scaphoid bones, the shoulder blades, the skeleton was so disturbed as to make it impossible to tell the positions of the legs and arms. Pelvic bones and skull clearly in- dicated the sex. Grave furniture consisted of two bowls and a seed jar placed near the top of the head. No. 16, an adult male. Lying 2 feet (60.96 cm.) below top of mound. Covered with a worked slab of stone, possibly a door. Remains lying on right side, head to the southeast, face to the northeast. Lower legs tightly flexed against femurs. Latter ex- tended from pelvis at an obtuse angle. Back curved to a certain extent. Left arm sharply bent at elbow and hand placed under face. Right arm only slightly bent, forearm approximately paral- leling upper leg bones. Skull was of the undeformed, long-headed type. Skeleton was practically surrounded by mortuary offerings. Eighteen vessels—bowls, jars, pitchers, culinary pots. Several small ROBERTS ] PUEBLO RUINS IN COLORADO 163 jars contained pigment, one red ocher, another yellow ocher, a third pellets of azurite, and a fourth finely powdered charcoal. The in- dividual was certainly an important member of the community, judging by the possessions which accompanied him. Out of the group of 16 burials in this particular mound 7, or 43.75 per cent, were males; 6, or 37.5 per cent, were females; 2, or 12.5 per cent, were children; and 1, or 6.25 per cent, indefinite as to sex but probably a female skeleton. The figures run somewhat higher as to percentage of males and females and lower in the number of children and indeterminate remains for this mound than in the group as a whole. This is probably due to the fact that there had been less disturbance by animals with a consequent better pres- ervation of the bones than was found in some of those investigated. Hence, it was not as difficult to determine the sex of the skeleton, the position of interment, and other features associated with it. There were only two additional burials which warrant a more de- tailed consideration than was included in the general discussion. One of these was in a refuse mound accompanying a typical A unit east of the main village. It consisted of two male crania and a few cervical vertebrae only. The skulls rested on their bases in a small circular pit dug in the hardpan beneath the actual refuse. There were no traces of the rest of the bodies. As a matter of fact, the pit was not large enough to have contained a single body. The skulls were surrounded by four bowls, the latter filling all of the space between the crania and the walls of the pit. The burial was unquestionably that of heads only. An enticing field for the play of the imagination is opened by such a find, but the requirements of the present paper exclude its exploration. Why the heads of two individuals were removed and interred without the remainder of the bodies is not known. There are no grounds for believing that the people who inhabited the region could in any sense be considered head-hunters. There was nothing which might be interpreted as evidence of sacrificial rites. It is possible that they may have been the heads of men who had fallen in conflict or who had died while away on the hunt. Because of the difficulty of transportation, it was not feasible to carry the remains to home hearths for burial, so the heads were removed, taken back, and properly interred. This practice is not unknown among aboriginal peoples and may occasionally have been resorted to in this region. The careful burial of portions of the human body has been found in other localities as well. Mr. Morris uncovered what he termed “The Burial of the Hands” in a late Basket Maker cist in a cave in Canyon del Muerto in northeastern Arizona. There he found side by side on a bed of grass, palms upward, the ligament-bound 164 BUREAU OF AMERICAN ETHNOLOGY [BULD. 96 hands and forearms of an adult. They had been severed at the elbows. Indisputable evidence showed that this burial was com- plete, that the remainder of the body had never been included in the interment. The funerary offerings accompanying the hands were unusually rich, indicating some special significance. Any explana- tions for this find, as well as that of the Stollsteimer skulls, would probably fall far short, as Mr. Morris so aptly observed, of the truth of the mystery which lies behind it.?° The other burial of special interest was an isolated one on the edge of the bluff, overlooking the river, at the north side of an A unit at the northern end of Stollsteimer Mesa. The interment was called the Arrowmaker Burial because of the 22 fine arrow points, the quantity of stone flakes, the core from which they had been struck, and the bone-flaking tools which were found beside the skeleton. In addition to the points and the chipping material the grave contained 21 pottery objects, including the seed jar with head and wings (pl. 21, ®) and the black-on-red “ feather jar ” (pl. 33, 6), quantities of red ocher, and a pallet for the mixing of paint. The importance of the individual was still further emphasized by the fact that bones of a golden eagle (Aquila chrysaétos) were by his side. This unquestionably was the sepulchre of an outstanding member of the community. The body had been placed on the right side, the legs contracted, with the head to the east and the face toward the north. Possibly there was the thought that he could forever gaze in solitude upon the waters of the stream. That he was a master craftsman is shown by the fineness of the arrow points (pl. 51, upper row) which accompanied him on his last journey. GENERAL SUMMARY AND CONCLUSIONS Scattered along the lower branches and bluffs above the Piedra River in southwestern Colorado are the remains of many villages belonging to the earliest prehistoric Pueblo periods. The sites were first discovered by the writer during the summer of 1923 when he was conducting an archeological reconnaissance for the State His- torical and Natural History Society of Colorado, but an opportu- nity for an intensive investigation of them did not present itself until the 1928 field season. During the months of June, July, and August of that year several of these one-time communities were excavated and much valuable data on house types, as well as many specimens of the lesser objects of the material culture of the people, were ob- tained. The investigations were almost wholly restricted to sites ® Morris, H, H., 1925, p. 291, ROBERTS ] PUEBLO RUINS IN COLORADO 165 lying on the upper levels of Stollsteimer Mesa. The only exception was a small village on the first bench above the river at the foot of the southern end of the mesa. Evidence from the ruins of 80 houses, 2 kivas, 6 circular depressions, and 7 burial mounds makes possible quite definite conclusions regarding the forms of dwellings; pottery, bone and stone industries; and the general cultural status of the people. There are three kinds of houses, which for convenience were designated A, B, and C. All belonged to the general type which is called jacal, had flat roofs with smoke holes near the center, lateral doorways, and were grouped around the pit from which the earth used in their construction was taken. Each house group and pit combination was considered as a unit, possibly the home of a single family group or clan. These units are thought to represent the prototype of the unit houses of the Pueblo IT period. The A houses were characterized by a quadrilateral but rarely rectangular form, sloping side walls, and interior support posts for the superstructure. The B dwellings were more nearly rectangular, the walls were perpendicular, the main supporting posts were incor- porated in them, and the superstructure was completely incased in adobe mortar. The latter was in contrast to the A forms where the plaster was applied to the exterior only. Houses of the C class were closely related to those of the B form. Their characteristic differ- ence was in the absence of the depressed floor or pit and their asso- ciation and combination with small two-room structures of hori- zontally laid masonry which are thought to have been granaries. The house forms were found to have developed in the order in which they are named. The evidence for such a conclusion includes an actual stratigraphic relationship between certain A and C re- mains in which the latter rested upon the former, a fact which definitely shows their subsequent construction. That the B class occupied the intermediate position was demonstrated by their asso- ciation with the A forms and their close relationship to the C in structural characteristics. This was augmented by the finding of objects in the B dwellings which are typical of A and of C. Pottery peculiar to the A class was secured from B houses, while a ceramic form, the corrugated culinary ware, characteristic of C, was missing from both A and B, although one type of implement, the ax head, was found in both B and C but not in A. The A houses are thought to show a late Basket Maker inheritance in their sloping walls, flat roofs, and interior support posts, but their general features are considered characteristic of a later horizon. The B and C classes have little in common with the dwellings of the 166 BUREAU OF AMERICAN ETHNOLOGY [BULL, 96 preceding culture and may be said to illustrate definite Pueblo forms of the early period. The outstanding constructional development of the period was that of the rectangular, perpendicular walled building, a thing which made possible the joining of single-roomed houses into a compact communal dwelling of many contiguous rooms, one of the typical features of the true Pueblo cultural complex. The three house forms of the early Pueblo period in the Piedra district show some general similarities to types in other regions. The A dwellings seem to have no close counterparts, although it is quite possible that analogous structures were erected in the Muddy River Valley in southeastern Nevada. The B dwellings have their counterparts in the Pine River Valley and both the B and C forms are to be found in abundance in the La Plata district of southwestern Colorado. A form of structure comparable to the stone portions of the C units is to be found in Nevada. From the comparisons made it would appear that the jacal houses described in this report rep- resent an early Pueblo phase which is typical of the northeastern San Juan area, but which in general show only superficial likeness to known buildings of that cultural horizon in other subareas. Future work in the inconspicuous low mounds covering house re- mains of the Pueblo I stage, however, may bring more examples to light. The unspectacular nature of the ruins of communities of this type has been largely responsible for the lack of investigations in them and thus far there is very little information available con- cerning the characteristics of the houses. The circular depressions associated with the house-group units are no doubt the remains of the pits from which was taken the earth needed in mixing the mud plaster for the dwellings. The pits served secondary purposes, however, and it was ascertained that they had ultimately been used as reservoirs, as the subterranean por- tions of kivas and possibly, from information obtained by a previous expedition, as dance plazas. Ruins of only two kivas were uncovered. This is an unusually small number of specialized ceremonial rooms for so large a number of house-group units but is a feature comparable to that found in communities of the same period in the La Plata district farther west. This lack of kivas is not believed to indicate that the Piedra settle- ments were temporary habitations or farming centers, but rather as showing that the circular ceremonial chamber was still in a develop- mental stage, that it has not yet become an integral part of the Pueblo cultures. Prehistoric Pueblo kivas of the San Juan area have many features which are also characteristic of the late Basket Maker dwell- ings which would suggest a decided influence extending over from ROBERTS | PUEBLO RUINS IN COLORADO 167 the earlier phase to the religious practices of the later peoples. The influence was such, however, that it probably did not assert itself immediately in the Pueblo I period. When the older ceremonies began to be revived, after conditions became more stabilized, there was such a difference between the type of dwelling then in use and that which had been the scene of the simple family ceremonies in the older horizon that it was deemed essential to construct chambers embodying the main elements of the older domiciles. To simplify building problems, however, only one such structure was erected for the entire family group and the idea of the clan kiva developed. Due to the conditions resulting from the commingling of older and newer peoples, it is quite possible that some of the units contained more Basket Maker members, hence, made provisions for a ceremonial chamber, while others, predominantly Puebloan, did not. It was only as the various elements fused to form a new and characteristic culture that the entire group was permeated with old, surviving beliefs and each clan constructed and maintained its own ceremonial chamber. Under such circumstances house-group units without kivas might well be expected during this early stage. The lesser objects of the material culture of the people are repre- sented only by specimens made from imperishable materials. All that remains upon which an appreciation of their arts and industries can be based are the pottery objects, bone and stone implements, articles of personal adornment made from stone and shell, and small fossils and curiously shaped stones which were collected and saved, possibly for ritualistic or ceremonial purposes. The pottery shows clearly that the ceramic industry was in a transitional stage. Many features of the late Basket Maker period survived while new ones, which were to become characteristic of the Pueblo wares, appeared. Among the most significant of the new developments were: The banding on the necks of the culinary ves- sels; the shift from the banded to coiled pottery; variations in vessel shapes; the discovery of the slip or wash of “liquid” clay; the elaboration of painted decoration and the turning to new sources for design elements; the extending of painted decoration to the surfaces of large jars, pitchers, seed jars, and other vessel forms besides bowls and ladles. The designs on pottery containers are characterized by a boldness of conception with a marked lack of skill in their execution. The decorator was in no way hampered by convention or bound by long-established styles, but was free to follow the dictates of fancy, although the results show that the medium was still unmastered. Bone and stone implements were not abundant and consist of only a few general forms. Awls, punches, knife handles, an occa- 88242°—30-——12 168 BUREAU OF AMERICAN ETHNOLOGY [ BULL. 96 sional bead or whistle, and gaming pieces constitute the main objects fashioned from bone. Stone was used in the manufacture of mill- ing stones, cutting edges, hammers, projectile points, and ornaments. The projectile points, especially the arrowheads, show that consider- able skill had been developed in the art of stone chipping. The only shell was found in four bracelets fashioned from glycymeris. The latter could have come only from the Gulf of California and represents early trade relations between that area and the southwestern Colorado peoples. Burial customs were found to be quite similar to those practiced throughout the San Juan area. The cemeteries were in the refuse mounds of the house-group units or of the villages. In practically all cases the body was interred in a contracted position and with few exceptions was accompanied by mortuary offerings of pottery. No great importance can be attached to the orientation of bodies in the grave, as the heads were found directed to practically all pants of the compass. The north was somewhat more favored than the other directions, and it is possible that a factor of some symbolical significance was attached to that quarter, although it could not have been of outstanding importance. “ The refuse mounds did not occupy any definite position with respect to the village or house-group unit, as in later periods, but were located at the most convenient spot regardless of whether it was at the north, south, east, or west side of the community. Some- times a mound was found within the village boundaries. In char- acter, the refuse which comprised them was the same as that found throughout the Southwest. The writer believes that the data from the Piedra investigations give a clear and distinct picture of the life of the people of the Pueblo I period in that portion of the San Juan archeological area. There is clear evidence in the wholesale destruction of villages by fire that the stage was one of disturbance and instability. That there was a mixing of peoples is shown by the skeletal remains. The transition from late Basket Maker to Pueblo cultural features is demonstrated by changing house forms, the amplification and de- velopment of the ceramic industry, the appearance of new types of stone implements and an improvement in the technique of manu- facture of some of the latter. Although constituting a distinct horizon, which substantiates the general consensus of opinion among qualified students with respect to the chronological cultural periods in the development of southwestern sedentary peoples, the remains evidence survivals from the older culture at the same time that they present prototypes for the typical features of ensuing stages. The material culture shows that it was the result of a taking over and ROBERTS | PUEBLO RUINS IN COLORADO 169 adapting of already established factors, to which new ones were added. As time went on some things were changed and others com- pletely discarded. This, together with the two forms of crania in the skeletal material, indicates that there was no sharp break, al- though there is a distinguishable difference between the old and new cultures, and that the Basket Makers were to some extent, in the Piedra district at least, absorbed by the newer peoples not entirely blotted out or driven into other sections. That the Piedra villages properly belong in the Pueblo I period is shown by certain outstanding features which have been agreed upon by the majority of southwestern archeologists as characteristic. Among them are the banded-neck culinary vessels; the change in house types in which the crude one-room domiciles gave way to struc- tures with several contiguous rectangular rooms; the beginnings of true masonry; and the occipital deformation of the human head. ©. TABLE 1.—MEASUREMENTS OF STRUCTURES EXCAVATED ° [A—1-—A refers to class A, Group 1, house A, ete. ] Wall Measurements & Depth of pit East South West North House FI ee be ei i a ee =| ¢ g a|¢ & © |. 3 gs 2|¢g ria 8/8 & By a aq Ey w a i 1 By rae i Ww a AT Ae oS oS 14) 0 4, 2672 |} 16 | 5 5.0038 | 15 | 0 4,572 15] 0 | 4.572 1 0 0. 3048 PAc1 Syn he oe @ Nid 2. 2606 | 10 | 0 3. 048 10 3.175 10 | 5 3.175 1 0 3048 IAA Oe See ee 6|0 1.8288 | 10 | 0 3. 048 6 | 914} 2.0701 9 | 3%) 2. 8321 1 6 4572 Ati a 3 6/0 1.8288 | 10 | 5 3.175 6 | 0 1. 8288 | 14 | 0 4. 2672 1 0 . 3048 / NSS Oe Se Ae 6| 0 1. 8288 8 | 0 2. 4384 3,5 1. 0414 6| 5 | 1.9558 0 6 1524 pAead —oe See Lees 6| 0 1. 8288 | 12 | 0 3. 6576 6 | 0 1. 8288 5 |10 1.778 0 6 1524 JN Na ae aa 12) 0 3.6576 | 12 | 0 3. 6576 8} 0 2. 4384 | 11 | 314 3.4417 0 8 . 2032 D9 2) 2 eee 12) 0 3. 6576 75 2. 2606 9 | 914) 2. 9845 8) 5 2. 5654 1 0 3048 15-9 Oh Ee Seen sh 16 | 0 4. 8768 8 | 0 2. 4384 | 17 | 0 5.1816 9|0 2. 7432 0 6 1524 JACI D ei Ee 1410 4. 2672 9|0 2.7432 | 16 | 0 4. 8768 6|0 1, 8288 0 4 . 1016 WAR 2 Nye ote aoe 11|0 3. a028"| ln |) 0 3. 3528 9|;0 2.74321 10]0 | 3.048 0 6 1524 AR 2a ee oo Se ca 815 2. 5654 9] 5 2. 8702 6| 0 1. 8288 6 | 914) 2.0701 0} 10 254 JAG Cr SSS eras eee 13 | 0 3. 9624 8,5 2.5654 ; 15 | 0 4, 572 9,0 2. 7432 Z 0 . 8048 SACO 3 |S ee ees 13 | 0 3. 9624 0; 0 0.0000 | 12 | 0 3. 637 7A a) 6. 5278 0; 0 . 0000 JA a eee 13 | 0 3. 9624 | 18 | 6 5. 6388 | 14 | 0 4. 2672 | 19 | 0 5. 7912 1 0 . 3048 ee dete eee OO 3. 3528 | 18 | 6 5. 6388 | 12 | 0 3.6576 | 19 |10 6. 0542 0 9 . 2286 ASSIS amen 10 | 0 3. 048 0|0 0.0000 | 10 | 0 3. 048 20 | 0 | 6.096 0 0 . 0000 PAR OAM an ect Sess 12] 0 3. 6576 8/0 2. 4384 | 13 | 0 3. 9624 6| 5 1. 9558 0 6 . 1524 [CS 2 eee ee 14/0 4. 2672 9/0 2. 7432 | 14 | 0 4. 2672 hada IZ 2. 6162 0; 8 . 2032 js 6) ae eee ee 16|0 4.8768 | 11 | 0 3.3528 | 14] 0 4, 2672 8 | 6 2. 5908 0} 10 . 254 J.\=3 =) 0 eae ere nee 19 | 5 5.9182 | 10 | 0 3. 048 17 | 0 5. 1816 9/0 2. 7432 0 6 1524 JS FE eee 9/3 2. 8194 | 27 | 0 8, 2296 9|7 2. 921 28 0 | 8.5344 1 6 - 4572 (Ama Be ees hie 2 6 | 0 1. 8288 | 20 | 5 6. 223 10 | 0 3. 048 19 | 5 5. 9182 2 0 - 6096 Amp Aut ws Dek 2 9) 0 2.7432 | 10 | 0 3. 048 9] 0 2, 7432 | 11 | 0 3. 8528 1 6 4572 Am p= Bee 2 ease 10/0 3. 048 11 | 0 3.3528 | 10 | 0 3. 048 12; 0 3. 6576 1 6 . 4572 1 iS Ce 10 | 0 3. 048 20 | 0 6. 096 8/0 2. 4384 | 21 | 0 6. 4008 1 0 . 3048 AW5 Dee eeesaee 15/0 | 4,572 8|0 | 2.4384] 14|6 | 4.4196] 7/3 | 2.2008; 0| 10 . 254 170 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 TABLE 1.—MEASUREMENTS OF STRUCTURES ExcavATED—Continued Wall measurements East South West House a n ra 2) slZ| 8 |glZ] 2 |sld] 3 Ba feed ee eta Ne | ce Theat lal it Wiehe os lcce cts 2910 8. 8392 810 2. 4384 | 31 | 6 9. 5612 OE he at SMe eee 10| 0 | 3.048 9/0 | 2.7432] 11] 7 | 3.5306 Paes 24. 25 san4-< 9|7 2. 921 8|7 2.6162 | 10 | 7 3. 2258 10 el ee eee 9|5 | 2.8702} 8/0 | 2.4384) 9] 4 | 2.8448 J \ <3 epee a 22 | 914] 6.9469 | 8] 9 | 2.667 | 22] 0 | 6.7056 YAMBA ve oO ewe 8/0 | 2.4384] 18]0 | 5.4864} 9]0 | 2.7432 A-6-B__....---.-- 10} 0 | 3.048 9/7 | 2.921 8|0 | 2.4384 J ol Ae en 11 | 7 | 3.5306! 8 | 7 | 2.6162} 11 | 914] 3.5941 (A—G=D seks 9|5 | 2.8702} 7] 4 | 2.2352 7 | 1.7018 MaleAe os 9/6 2.8956 |} 15] 0 4. 572 9/6 2. 8956 Bales 20 | 0 6. 096 15] 6 4.7244 | 21 | 0 6. 4008 (Sh :t 3) 6 1. 0668 5 | 6 1. 6764 3] 5 1. 0414 B=1-C_.a2-2---222 11 | 0 | 3.3528] 11/0 | 3.3528] 11) 4 | 3.4544 B-1-D__.=..------ 9} 0 2. 7432 | 12] 6 3. 81 10/0 3. 048 123 ES Dee 20; 0 | 6.096 | 12/6 | 3.81 21/6 | 6.5532 B-1-F_____-_------ 15 | 0 4, 572 12] 0 3.6576 | 15 | 6 4. 7244 13) Ec ae! 16/0 | 4.8768] 11]0 | 3.3528} 17/0 | 5.1816 Bate oe ace ce 15|6 | 4.7244] 9]6 | 2.8956} 16/6 | 5.0292 1305 ES ene epee 9 }10 2.9972 | 24) 0 7.3152 | 11 | 0 3. 3528 (Baie eae 12|0 3.6576 | 15 | 0 4. 572 15 | 0 4. 572 BaleKe 222-22 9/0 | 2.7482] 8]6 | 2.5908] 11] 6 | 3.5052 5 os De ee 13 | 8 4.1656 | 32 | 0 9. 7536 | 14 | 0 4, 2672 B=I-M So. eee ae 20 | 6 6. 2484 1 11] 8 3. 556 21/0 6. 4008 IBSI=Ne se toes 11 | 4 3. 4544 T 186: 2. 286 11 | 5 3. 4798 G=I=A oe ce case 710 | 2.18386] 14]6 | 4.4196 914) 2.0701 GIB. =. 2 33-2.-2 6 | 914, 2.0701 6,0 1. 8288 ico 2. 1336 (C2) CC er 710 2. 1336 710 2. 1336 7/6 2. 286 C-2-A____.-_----- 12) 0 3. 6576 | 20 | 0 6. 096 11 |10 3. 6068 G=2-Bix oc acelecce 16 | 7 5. 0546 8 | 6 2.5908 | 16 | 334) 4. 9657 G=2-Ce ie ececeese 6 | 0 1. 8288 6) 6 1. 9812 6 | 314] 1.9177 (CAS) D eae Berra ee 5 | 6 1. 6764 6 | 0 1. 8288 5| 6 1. 6764 @-B-A.<.222-2-52 15 | 6 4, 7244 | 14] 0 4. 2672 | 19 | 6 5. 9436 G@-3-B.. = 2-5-2 6|6 | 1.9812] 7]|6 | 2.286 6|6 | 1.9812 G=3=@e io ee sce 5|0 | 1.524 8|6 | 2.5908} 5] 6 | 1.6764 @-3-D 2-2 - see 15 | 6 4.7244 | 10] 0 3. 048 14] 6 4. 4196 @-8= Bee connec set 6|6 | 1.9812] 8] 4 | 2.54 7|0 | 2.1386 (Oe i. err 8/8 2. 6416 | 14 | 214) 4.3317 9|7 2. 921 14 MOWFROTOAADAAAMOKATOOCCORCOCRBOCAnNPHRDBOWwWooCOENSOSCS | Inches Depth of pit SOC C TCO OC OOOO OH HB Bh eh eh eh ere PH OOO OHH OOH HE Hee | Feet a - ———————— —— — — — — - — — —) | Inches _ ao a RR) ROBERTS | PUEBLO RUINS IN COLORADO Lyra TABLE 2.—NUMBER, SIZE, AND PROVENIENCE OF OBJECTS ILLUSTRATED 1 Denver Museum. POTTERY iT Size Tlustration Height Diameter Provenience raenes| CO] oohes | Con MiP Sas Se REET 13% 34.3 11% 28.3 | B Village, house B___- [a eee en ee A eee 154 38. 7 T2AG |) Mdls6). | JR—2=Ji. es ae (GeO ES EO ee se Xe 8%6} 20.5 71346) 19.9 Buris le sees ee Rs aks ake EE 72340! 19.6 86} 20.5 | Burial 2-B-14________- (| OP a De ee 66 16.0 5% 13.3. | burial =f 2-2 ee 1 pleats DE EEE teers 41546) 12.6 54% 1330) || “BuriglkA—l_- Pee ew eee ee ero 7% 19.1 5156] 15.1 | Burial C-10____.------ acl eee ee 4% 11. 45 46 W.3 Nec ge Ogee ee ie has te. Se A 9 22. 86 12% BLO) Sao (6 0 eee Sth Se es eS igen Gasset. eS eon eee 20% 52.1 144 8652" Als 22 = oe | eee Oe ete at 1734 44.2 1246 30. 7 a= [er a es 5 P (re Ee ee eee ee ta 1534 40.0 1338 O45.0) b | AsO Ae ee rere (0 ene ee a ee 15%6 | 39.5 137% BOO | PAK o— a ee ee Ts ee ign iiagcn ==) ea 3146 8.8 5 12:7 Bal=Mie ee Dee eee SSeS 3%6 9.0 5% TSO) pl) AKS—= He 2 52 Se ees (8 Gg ee Ce eens 534 13. 7 7% 18.0 | Burial C-8.___..------ (6 (CAs eal Oe eae ee ee 414 10. 4 74 AGRA) [es 022 ns See eae le ee 454 11.8 7% 20.0 | Burial 2-B-1____-.---- | eee ee eS oe ie ees 346 8.4 61146} © 16::9' 9 |) A-5=B.. 5-2 ee ee 546 13.5 41346, 12.2 |____- doa ] 0 ee ate a ee ere 954 23.4 12% 31.1 Aad Ae 2 ee j Ci ee eee 84 21.0 115 4° 9a a lay NC fee eee ea 101 ee ES Cees ae ae ey eee ee 12%6| 30.6 SAG.) "2450.4 | dBA ee eee [Die eee et 10 25. 4 SiHiet 2158 | WAS At eo eee Cee Et re 914 23. 4 834 22.2 | Burial 2-B-9_-_...-.--- Cee ne oR ae 10156} 27.8 9%6 | 24.3 Bunia2 22 ee (ee ee ees See rad 71546] 20.1 66 15.4 Burial 2-B-12_____--_-- fos Se UNE 2 WaT ee 93%6 | 23.3 7346) | 1852" | Burial D=1-2 2 [2 eee earn 51546} 15.1 31346 9.7 | A-4-B_-..--.----=---- No eee eed er eens 76 19.2 6 15524) Burial'@=2-- 2 Se piece Sat SOOT 334 9.5 3346 Sil )#AK2=@s..52.--2<-esee8 Pane ees oka ee eS 75 19.5 6% 15. heq| Burial ©-8. 2-22. 2 a= M19 an eee ot 3% 8.3 634 Ute? 4 Burial |@=1_ | 0st Se. eae ae Aes 354 9. 2 6Hei} 16.0 |, Burial: C-2_-_...--...- (ES ee eee ee ee 3 Tal 7% 1940) Bi, AmonNe = 25-2 esccean es oA es Seine Se 2 ae ee 5% 14.0 3 CAL Mie I iepatte eeee C; ee eee ee ee 934 24.8 5% 1470) % | sBuriall222- 222 eee fice te oe es 16 3.0 274 i238) Wp ROUISO S25 oo =oe oa aoe ee aS ee ee 154 4.2 216 CGC EE 1 | stipe bee Se Sere erieta ee anes 1h ts Ss Oe eee ea 234 7.0 414 1025: *} Burial’ @-1022222_ 2222 - Ses Pee eee ae 21% 5.3 36 Sez tie teh (6 bo eee A ee Wneooc soe eee 9 22.8 10 if ty ee eee ea Kel ee ee 5 127. 416 00a Ae: a Gee ee ee | pea BL ere eee ete 3% 9.8 46 10.9 Be1-H > 52235e-4-- 5-22 Mire sos leseses ses * 548 13.0 2346 625» ||), A=6=K= 552 s2------esu= Ties et ne ee 97% | 25.1 334 8.6 | Burial 1.......---.---- 4% 10.5 Ose eee eee 5% 13.3 4g 10. 5 A-1=B 22 322253322 5=: Din cs tacit ee a Pe 6%6| 16.7 Blaha: COUT | | eens eeeee seen mer eee 4%6| 11.3 | Burial 2-B-10--------- [ees So Pee mee: 8% | 21.0 61,1 |- VIG SB: ||-o ee. kate eee ee ee AVG. 20s Oooo ence el Ss 314 8.8 66 16.5 | Burial 2-B-2-__-_--.--- De sva ieee eee cee 334 9.5 8 DOS) Pi Aang Leese eee eae 348402 348246 348155 348390 348463 348333 348485 348313 348312 348206 348232 348311 348230 348419 348308 348471 348465 348369 348320 348322 348273 348240 348394 348274 348194 348182 348386 348503 348268 348449 348216 348464 348444 348447 348221 (+) 348201 348505 (+) 348490 348491 () () 348413 348315 348170 348205 348382 (1) 348370 348222 172 BUREAU OF AMERICAN ETHNOLOGY [ BULL. 96 TABLE 2.—NUMBER, SIZE, AND PROVENIENCE OF OBJECTS ILLUSTRATED—Continued POTTERY—Continued Size Illustration Height Diameter Provenience I Centi- Centi- nehes | meters | 2¢hS | meters 1051 27 3 a, Nal ee a a na Se 34% 8.8 6% 15.9 | Refuse mound_---_-_-_- Petter Basset 316 8.8 73% 18;.7 | | Burial ‘©-8- 22-2220. = Petes Gee tote oak Le ea 334 9.5 7 17.8 ASQ RTS ee oat ee ee Det anceo Case 254 6.6 5% 13.0 | Refuse mound ----___- IPAS en kes elk nk Se 334 9.5 8 2053) |, A-3-B 2: 2 ese ac eees (0 lees eee eter Meet nd ea 334 9.5 814 2136 ‘|: Ano Ta as eee T3025 07 OY, hee ee 334 9.5 734 1988 | Bunalles S228 eee | Cee a ee eS 314 8.3 7 17.8 | Refuse mound-----__-- 1S OL: ee ea | (EP a (Ee ee el (eed | Ee, eee Restored from pot- sherds. Deis ete a ee | 416 11.4 914 24.1 Burial 2-B-10________- ea ef eee, | 4 10. 2 8 20.3 | Burial 2-B-5._....___- Pye Saeee 2 ee 234 q0 5 12, 7 IBS] gee ee eee ee POS Dye ee oe oo sd eae eed pepe Peers eee mR cel ree eee Restored from _ pot- sherds. | 2 eee aes em 344 8.3 74 1856; " | (O=Burial 1 22-2222 LOST ca ee ee 4 10. 2 814 21.6 | Burial C-10___..-....- sae ne ee 314 8.8 8 2053) |. Buriall= 2) =e V2. a en 334 9.5 8 20:3. | ‘Burial A—1*-<_-22-"- Deo ee ose eases 334 9.5 8 2023) |, Buntal'@—b22 eae 1 Oa Yar) ear ee en 3% 8.8 8% DUG Ame ieee eee eee | tet Oe aie ee ae 34 8.8 74% 19.0) Burials... -=ss2eee2 1 E525) LY ge 4% 11.4 9 22:90) Buriglige=-.22 eee | 1 rg RAE eee ne eee ee 4h6 11.4 914 24 ASB =O 2 Sono seco eee Wig OS: afc soos Ne aes 3 7.6 5% 14.0 | Burial C-5_______-__-- | Oar de ea Pe 334 9.5 8 202037 Bun el so. 2 eee LG a a ee 234 7.0 634 Lio wO-Burial—-l 2 oec eee se | ee an ieee wens ee ba ee ee eee Ee es Restored from pot- sherds. 10a ts) Se: ee 9 22.9 1215 17 | eAKb- He 2 oe eee | 6 ged en ee BN Mae (ie reer ado: A hese! ee ee ee oe Restored from _ pot- sherds. IN eS GR ee re 3 7.6 ai 1901}: BaI=Me 2c ee ieee ee oe eee 234 7.0 74 190U 20 s2 eee eee (eee Sen ere eel 334 9.5 7% 19:0) (Buriali4 2 = 222 see eee (6 Loe ete eee reat tes 334 9.5 845 D1-'Gi}|| sAj4 A: oe 2 kee ee HS: B0; 9 Beeson eee 84% 21.0 1154 2035" | Ace sean eeemennas | | ae oe ee a 7%6| 18.3 1154 29) 5) | ane OSs Sen CMe oe re ee 9346 | 23.4 12% S114) Apa AG Ss (6 eee 814 21.6 1214 Ole ft Bela AR ae ee ee AE 28 V/A Ma ee 10 25. 4 8%6 DEST ASA Ree ee ee Ds Sen UE SF eke 81546| 22.7 4%6 190, 2h Burial 12s 2 ke TN 1. TG: oe ee Re 34 8.3 74% 19,'0)}, Burial D-f.-22_- 22222 ] a Ee Se eee 3% 8.3 8 20:/3)|, Burial @-82 5-2. 2222s 336 8.6 |] : D0 -at)5 s e 97% 25.1 | 434 10.5 (Burial 1 Pee eS a edhy (2) 6 ee 15% 38. 7 12%6 S166)| Aq2=Jic 2225-22 eeceee led CY Vy Oe Wiis aes ae ae ee 454 11.8 46 LO, Ba Aso Ry 22 oe eee ] Oar a SOR ee et eae 454 11.8 44 10.8 | Burial A-1__-___-___-- (1 ee ea ee a eee 513%{6] 14.8 534 146) A-5=Wiss 22a ee (0 LARS ee 5% 14.0 31346 9.7 it A=2=Ge 2. foe eae PILenLOng 29 sass <8 5. coe 4146} 12.6 514% 13.0%) BurislAqi- 2202s cen | seer ee ee ee ae emt 434 iV AS I 45§ 1g) Uriel eee (Clee sere t= ane 6% 15.9 51546) 16:1 | O—Burial:2: 22222222 Heo weticatis. 6%6| 16.0 | 56) 13.8] Burial C-7......._--- 348356, 348470 348226 348361 348254 348225 348171 348365 348383 348373 348418 348511 348483 348173 348334 348451 348242 348186 348192 348259 348452 348164 348512 348312 348421 348420 348189 348276 348240 348245 348273 348398 348274 348160 348501 348469 348170 348246 348313 348331 348318 348215 348333 348185 348509 348463 ROBERTS ] PUEBLO RUINS IN COLORADO 173 TABLE 2.—NUMBER, SIZE, AND PROVENIENCE OF OBJECTS ILLUSTRATED—Continued POTTERY—Continued Size Illustration Height Diameter Provenience Inches | meters | 12S | meters RlateplG were sane. te 9%6 23.8 1434 S695: AS97283 202 tee ee on cn RR 9346 23.4 15 reas UR) VB] 3 So Fy. ge ana TEAL FN ly ee a a 204% 52.1 144 Leh OF] Mal 6 ee Platenlsiiae <2 1 ees 1738 44.2 126 BO: || PARTE ns ee Ae | By 2a Sea eee Gere Nee 1534 40.0 1334 3450) | A-S=B ee see platen Qa s = 22 2 ees 10%6 26.8 8%6 Ba | (ASA AS So ce ee | 6) sao Sh a eats et Se 15%6 | 39.5 13% Bona PASI al ase ee TeV EIO. UGS ae) pe ee eee er 5% 13.1 758 1953),|) Burial'@=1 22.5.2. | La ale eae Net a 534 13.7 76 18.0 | Burial C-8____________ (oh eet ees ees aS 8 8% 21.0 1154 D2OND | WALZ A) 8 dee ee ere Gees eee ee 746] 18.3 1154 2955) 522 ows 22. eee QR Ga2 A Sats eer 5% 15.0 9%6 2550 (Bb urialig? 229s fen Nertr 52 se ete 634 17.2 934 24 |, (A-$=B 23 oe (oe a eee ae 38%6 8.4 6146] 16.9 |_.-_- (6 (ape Be es | 0 Se Nia i a agian ates 434 11 56 1 Ra Soe | Pee dole Sas eee Plate 2a! — ak 714 18.1 11 27.9 | Burtal ©-10__.....2._- | 0) SER Sates Oe eae 86} 20.5 76 1853))) Burigl dace. 2 2 Ate 22 48-8 > ho = nd Tie} 19.5 614 16.5 | Burial ©-8. ......._..- eee ee ee 6%6 | 16.7 513{6) 14.8 | Burial 2-B-9_________- (jee Bk ee man heras Re is 71546} 20.1 66 15.4 | Burial 2-B-12_________ (i see hoe ee 93{6 | 23.3 THe 18,2: | Burial D=1-_----_- =. Obsees See ee no ten 958 23.5 834 22.2 | Burial 2-B-9__________ | Cl aan ee ee ee 12y6]| 30.6 96 24:10) | B=I-Ac 2 coe eae |) re a ee LENS 1014416] 27.8 9%6 243/3''| Burial(2s - 2-2 laters et hs ee 334 9.5 3346 830) MAS2=G Je | 0a eng SR Re alee 514 13.0 46 11.3 | Burial C-6____.______- Oe Doh hn ses ceetits 41346! 12.3 4546 TAO) Burial 4z. o2-= 2 fo (Seas SS Beds pec 51546) 15.1 31346 ONT) | Ada oe Wlateid was bo kh 414 10.5 7% 18.1 | Burial C-8____________ |) ae See are 534 14.6 1016} 27.3 | Bracelet burial________ IBIRteZo tee == 2 ee 3%6 8.4 67% 17.4 | Burial C-1-.........-. | a) each lee OE ee ote ae 3%46 8.4 634 17.1 | Refuse mound -______- ean ae bees Lt. ad 234 7.0 634 17.01 \| B=1-—Burial d=. ---- 2-2. Chic eet et oN ae ee 21346 adh 616 16.5 | O-Buria) 1___________- (_ ae ae enue oe, Ds yea 354 9. 2 734 19:7 | Burial: 2-B-19--.. fate hak ee eS he 4 3% 8.9 T1%6 20.2 | Burial C-12_._________- DEA EN TOT RR ee oe eee 3% 8.2 Te 13;.3i)|/HO—Burial 2202-2222. | 6).0 Se ae ee 2146 6.5 53% 1307 |) Burial Ifo ee Cy eee TD, RS 334 9.5 8 20:3), Burial Ac: 22-2 22.2 les ee ae Pan 3%6| 8.7 854 2910) | GA=o Jia ee (7: Spa epee See BD 36 8.7 83g 9199) @Burialiiss 0. 2-2-2222 fot Se 2a et 354 9.3 834 O23) litAaeolaa eck oo oon eee plate: 2a ss 222 = oo 5 3e 7.8 6%6 16505) sB0rial) Bava. 225-2 == ] oN ee kt Oe 336 8.6 634 17s! | Burial A—2-- == 2-2 Ce eee 446 11.0 71Ko 19.5 | Burial 2-B-4...-..-+=- ob AR pepe eae aoe 334 9.5 8 20,3. | Burial! C-9- <- 222... (Sow BASE fee een ee EP 414 10.6 7146 19.5] Burial 2-B-4_________- |e ENS a eae re 3%6 8.4 634 17a) |pubiel B-725-2 5227 -.ns (Plate 28k 2 ee ae et Be 344 8.2 7%6 19/2) Buriali@-1022 22 == peas ee eee aeys 346 8.5 716 19.8 | Burial: C-15........... oe ey BANE ae 354 9.3 635 16.0 | Burial C-2._.........- (1 [Seeee El i eee Be See 34g 8.0 THe 1853) [iBiurialior ee (SSSR re RCs eee 35% 9:3 9%6 93:3, |, Burial\A—42_ 22 es <= fete ae re Se 3%6 8.5 7 19:1. |, Burial: C-Ve2- = -eee 2-3. Field No. Na- tional Mu- seum No. 348243 348397 348206 348232 348311 348272 348230 348446 348471 348240 348245 348187 348309 348320 348321 348450 348151 348464 348380 348386 348508 348381 348394 348182 348216 348453 348188 348268 348465 348349 348444 348356 348436 348513 348385 348495 348511 348162 348334 348242 348173 348225 348377 348336 348371 348473 348353 348378 348480 348499 348447 348198 348343 348445 174 BUREAU OF AMERICAN ETHNOLOGY (BULL, 96 TasLp 2—NUMBER, SIZE, AND PROVENIENCE OF OBJECTS ILLUSTRATED—Continued POTTERY—Continued Size Tllustration Height Diameter Provenience at Inches bss Inches Renee Piste 20 nee 2 Sean ee 3n6 8.8 734 19.8 | Bracelet burial_______- 143 11 eget a a Ae 34 8.2 78 18.2 | Refuse mound _______- 197 Cape ee ere 4 10. 2 9%6 2a.) || IASON 2 eas See oe 106 (5 Sa ee ee eee 354 9.3 834 2173" | Burial 2222522 eee ae 31 On 5 Se ee ee eee 3%6 9.1 834 2253n| (Buna tee eee 23 | (ope 31546] 10.0 9 22:9).| (AM5=B) oo cee cen 130 Plate 30,'a---..------.-.---..- 2% 7.4 66 16*4:\) Burial A222. 2s --2= 126 "4 ince Reel eee te ee i ; 5%6| 14.1 | Burial 12.-..........-- 155 re I 358 | 93 | 74 | 19.1 | Burial 2-B-15___-._.-- 189 (5 el ei eee 316} 9.4 THe 18.3 | Burial 2-B-7_________- 172 (Oe es eee 3 Vex; 7/6 1950) | (A—2=1- 2 Sooo ee 62 femeeee -5. eee = Ac 314 8.2 8 20/3) |B Ural eee eee 10 Plate sla 22 2 22a se 334 9.5 8 20.3 | Burial A-1__...__.._-- 117 | 0 eee Rn a eee 2% 7.4 66 15.4} Burial A-3...-.......- 126 Oe eae 21%6) 7.1 6% 16.5 | O-Burial 1___________- O-1¢ IPIAt6 S2.(a. 22.22 S22 ccc 21% 6.3 3n6 8.7 | Burial C-10-_________- 241 | 0). ee eee 374 9.8 4546 10/9) (BSL SR. 2.2 ee oe Be A-21 Cio ees ModE 234 7.0 44 10.4 | Burial C-10__________- 240 Piste 33; (84-8. Se Sheet 54 13.3 44 DONG), JAKIAB os neve eee sae = 47 ee 97% | 251 | pi bs : {burial eae ed 21 eee ee 654 16.8 438 Ds | ASbaB 22 ep cecccee es 103 (eS oo ee 6%6 16.7 31K 6 9.7 | Burial 2-B-10________- 206 Plate 841.0. 224-25 ec sc5e 31546} 10.0 254 OSG) | Aso xs - noone 2nceeee 57 [) en areas: © 4346 10.6 21\Yo 689) | 22 G0. 23k hoes 86 (Sane hae eee 354 9.3 3Y%o 7.8 | Burial C-10__________- 243 (eee eee 4 10. 2 316 S.0 Vase (6 (0 Yee 238 Pelstego,ihe...- 3-22 sscee-c5554 UH.) 4.9 146 4.9 | Burial 2-B-14________- 187 Dea sas oe ae oes 15% 4.1 1% 3219) A=4- Ben paan, ete ee 139 (Gases eee eee eee oe ee 158 4.1 24% §.8 | A-4-A 22-2 ee 94 die. sa. ee 14% 3.9 2346 See || Burial 122. 2S. 147 (See See 4 1.4 16 2.6 | B-1-Burial 3___.-___-- A-25 1 ER eee es ES 14 aaa 158 Ao | Burial i220. ok ee 195¢ PPA LOI SH See 3 = oe EE Re Eee eee ee Be Houses C-1 and C-2_- 298 Length 16d EGY Et: eae ee ee 314 8.0 1% 3.9 | Burial C-6___...-___-- 224 De ae Ae eee eee 1% 4.8 % 2.3 | Burial A-1_........_.- 17a (Ge Are Seer 148 4.8 1346 2.1 | Burial 2-B-14________-_ 186 (Se aS 248 5.4 % 2.3 | B-Refuse mound______| A-28 CE ee a a 1%4%6) 5.0 % 2.3. | Burial C-6.__._....-=- 224b | 2 et San PP 2 6a! 1346 2a) | O-Burialit 22-- 222--- 257 (i a ee 234 6.1 16 2.0 | Burial C-8:_ =... .2.222 212a 1 te ee ae ee ec: 31346 9.7 16 237 || (A-35B nee ee ee 161 11: ee Seay ed 1% 4.8 1546 2/4.) B Village:..-=.<---.-2- A-15 IPlate.88, (as-22= 225.4550 2.252 \8 bee ee. | OR NL Nl 3 ieee ee aS 137 ] 0 Sey eee Om bo ee) | ( Se S Je| Aen| [es Ro ae Ne 6 ae ere ene ee 54 ( ) ae ean ees peer SE) Ayer oe eS | ARG ae coed ee Lm 2 Yo ee ee Oe et 54 ee eee oe | he ea | oe ec On 2. eos ee oe A-13 (eae ee eerie) meet] FE S| gee bee Ee ane Sea Oe Be Fe (6 (see ee a 54 ) a ee | (cm Ce + | are OA Oc I a WE Onc Mego dues Boe alae 54 [5 Spe ee ne eer eae ae a pee a Oe Rl Ug Oi 232 one Sia 54 348352 348365 348317 348181 348172 348323 348341 348201 348368 348376 348221 348159 348334 348341 348513 348491 348413 348490 348205 348170 348314 348382 348236 348247 348493 348488 348389 348345 348267 348194 348437 348363 348439 348458 348335 348388 348434 348460 348507 348466 348256 348409 348302 348234 348234 348402 348234 348234 348234 ROBERTS | TABLE 3.—NUMBER, SIZE, AND PROVENIENCH OF OBJECTS ILLUSTRATED Plate 40, Plate 41, Plate 42, Plate 43, PUEBLO RUINS IN COLORADO BONE Plate No. Size Average thick- Length eee tnches | CoP | ncns | Cent 5% 13.0 16 1.8 6346 15.8 %e 1.5 416 10.5 34 1.9 516 13.0 %e 1.5 2546 5.9 %e 1.5 238 6.1 136 2.1 458 11.8 54 1.6 478 12.4 16 13 | 456 11.8 54 1.6 434 12.1 58 1.6 | 54% 1 130] Mey 18 | 41M%6| 12.0 He 132 36 7.8 34 1.9 2716 6.2 eo 125 376 8.8 1% 2.3 6346 15.8 146 1.8 4 10. 2 M6 1.8 3546 8.4 5% 1.6 434 a Ral %6 1.5 216 6.9 1K 2.3 46 10.3 58 16 258 6.7 34 1.9 416 al ess 146 1.8 614 16.5 134 4.5 76 18.9 1546 2.4 734 19.8 % 1.6 6% 16.5 re 1.5 84% 21.0 1% 3.9 234 6.1 34 2.0 | 2% 6.4 1346 yal 246 6. 2 146 1.8 334 9.5 146 3.4 26 5.6 56 1.6 24% 5.8 1 2.6 558 14.3 34 1.0 458 11.8 yy 1.3 a 2.6 % 1.0 1% 3.2 Yo 1.2 1136 4.6 %e 1.5 1% 3.5 % 13 1%6 Be iG) 12 % 2.3 34 1.0 34 1.9 6 0.8 Provenience ae Refuse mound A-1____ 44a Ses GOS ee ob a aos oe 44b Jams (0 (a) ee ee ee 44c B-1 refuse mound_-__- A-26 Refuse mound A-1___- la Burial! C=6: 222-22 = 223 Refuse mound A-1____ 44d 5 5 EO eee a eee A-15 aod dose ees | AS15p Refuse mound A-1___- 446 Burial C-6:2 ....=----- 223b Refuse mound __------ 285 B-1 refuse mound_-__- A-26 ea GOs. et 232522225], “AS28 B-) buriald.:-_.-..-.- A-24 Refuse mound A-1___- 44 Refuse mound C_____- 233 Burial ©C-6 2-22-22 =- 223 Bungee See ee 32a MEWS sense anaes 284a, B-1 refuse mound_---_- A-26 Retisey.---- == 2 os 284b Refuse mound A-1____ 44 Refuses ee1"s- 2. <2 286 Burial C7 aeee 244 Burial C20. ss 2-3 253 Refuse: 222 5os-=22222 287 Refuse mound A-1__-_- 44 WROMUISGh ee oe 288 Burial pea 5see 2 ee eS 33 Refuses 2. ete eeee 2 289 B-1 refuse mound_---- A-27 IBUTIAl 2eas 28 sae sees 32a Refuses: =. -82s=22. cee 290 B-1 refuse mound_-_-__ 291 oa GOs ee eA 2o a eae GQu es Os See 292 pS See G0s2 2a eee A) 244s Burial se2 ee 293 Refuse mound 4--_-_-_-- 158 NE slp ag ciel 51 Burial ©-sxesaw aes 212 oe COME 2 eeeeceaee 212 175 348203 348203 348203 348432 348203 348457 348203 348406 348408 348203 348457 348426 348432 348432 348432 348203 348442 348457 348183 348424 348432 348425 348203 348426 348461 348448 348426 348203 348427 348184 348427 348433 348183 348427 348428 348435 348429 348462 348429 348347 348220 348467 348467 176 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 TABLE 3.—NUMBER, SIZE, AND PROVENIENCE OF OBJECTS ILLUSTRATED—Continued STONE Size ; National Plate No. Length Breadth Thickness Proverierta Fed oe No. Inches Miaiels Inches A sale Inches ieee Plate 45, Aco... 76 | 17.9 558 | 14.3 3X6 | 10/0:| A-2-G_ 222-2 22s22_ 61 | 348217 | ee eee 7%6 | 19.2 51MY46] 14.5 358 O22! B-J-E = 222ereets 280 | 348412 OM hee ae 2%6| 6.2 1%6| 4.1 138 3.5 | B-1-Kiva_______-- A-23 | 348423 (0 [tae tee a 6% 16.0 4 10. 2 3 AG ARB Oe tt eee 85 | 348258 (ae Saree een 536 13.7 34 8.3 Q1S4 6) fee AR oi ees ee ee 53 348233 Plate: 46.)a.-..22.222- 8346 | 20.8 4% | 11.4 214 Dit | GAL D)s sete S558 42 |} 348211 Dee ae sas 75% 19.3 5 12:7 11346] 4.6 | A-4-B___________- 278 | 348303 (tA ee aa 10%6 | 25.5 31546] 10.1 214 Bet, | Aeon oeees sees 84 | 348252 (oka aml ee Seer 8 20.3 41346] 12.3 158 4,1.) B-l-D- .-222 22222 279 | 348410 Plate47, aot ts = 32 55% | 14.3 2% 7.4 134 Sao: Bal=Clsze 2 ee sees 277 | 348403 Dee ee | 6346 | 15.7 3146] 9.4 114 358i) (Bal-Gz ositece st 276 | 348414 Or ece tn | 71%6| 19.8] 436 | 11.1] 1% | 3.8| C-1-A___.__.._-.. 275 | 348438 0 ale, + See 534 | 14.7 3%6] 8.4 15 4.0 |('C=2-Be- 222 Se ee 198 | 348367 Flate 48, a________----- | 10% 26.0 4 10. 2 134 325: AS6-C oes aes 281 348329 | 0) Sac a | 1716} 45.0 248 5.3 146 3.4 | Refuse mound__-_- 167 348355 Plate 49, a______ Sede4es | 15% 1 ee apes eek pte | enn ed EN gage orig jn----- A-1-C___.-_ eee 43 348207 |e ee eee | SSiai | S20 Oneness | ee [eee ees SOE do2S Ss. eneeee 43 | 348207 (32 sa ee Hh Wa fe? bok jad een a ee eed |e Burial 22s ee anes 283 | 348510 (0 | Sapa e ee | 738 tb Se fal oe ge Josea=- | Pant epee eee ae IS 9 3 ees al 282 348415 | 1 Diameters only in this group. TABLE 4.—S1zE, NUMBER, AND PROVENIENCE OF OBJECTS ILLUSTRATED CHIPPED IMPLEMENTS AND ORNAMENTS Size —— Na- Illustration Length Breadth Provenience Field Hone = No. seum . : 0. Inches Sates Inches Tate Biateis0y asses 3 7.7| 16 CW yes ae ee RS 76 | 348250 | one 244 5.8] 1546 BIN [io oa fea ae cere oe | 296 | 348275 i Senge ate) 26 5.7] Me 9-8 || Burial G-8=__---222--2 76 | 348228 ri ee ee AereS 254 6.7] 1546 Dyas ea do3-- ae te 212d | 348468 Oe ee eel 274 7.4| 1% 2,9 | A-5 refuse.........-.---- 212c | 348468 ened Eepaaeaees 156 4.2| 1% a2 wee YG pe PS 233 | 348441 Plate'bl, ao... 25.5... 2%6 6. 5 | lo 1938] BUTI 1a eee oe 1 | 348150 Bate 8018. dc. se owes 15% 4.2] M6 2.4 | A refuse mound_..._-___ 297 | 348348 Plate 53, a_......----_-_-- 2% 5.4| 1% 3.9) | 51.3 dou: nee reer eee 250 | 348504 poe ee eee 1% 2.9| 1%. SHON wA=G— CL aie eee eee 295 | 348262 (Cero See 14 2.9 1% 2.9 | Burial C-8________._-... 212b | 348467 ibe tle epee Ne 3 13] 4% 0:6) "BEl= Oe. 5 eee A-5 | 348403 (ee ec 1% 4.8 | 146 1.8 | Refuse mound__-_______- 294 | 348430 fice sc ee een ee ore | aE [ate alll De es nee een LA 299 | 348431 Plate 54, 2a.._....._-____- 3%6 8.8) 3% 8.9 | Burial._.-.-.-...----_-- 141 | 348350 peieecoso een 314 8.9| 3% Segue: dots oe 141 | 348350 (ee eee 344 8.3 336 $37) | AG) 2. be ie SD 141 | 348350 ae eee 3146 8.8 | 336 8i7, | Sw 0c. Se ue aaa 141 | 348350 1 Diameter and thickness of this specimen. 3 The breadth in these specimens is the diameter through the hinge. BIBLIOGRAPHY Bancrort, H. H. 1890. History of Nevada, Colorado, and Wyoming. (Works of Hubert Howe Bancroft, vol. xxv.) San Francisco, 1890. CUMMINGS, BYRON S8. 1910. The Ancient Inhabitants of the San Juan Valley. Bull. Univ. Utah, vol. mr, no. 3, pt. 2, Second Archeological Number, Salt Lake City, 1910. 1915. Kivas of the San Juan Drainage. Amer. Anthrop., n. s. vol. 17, no. 2, pp. 272-282, Lancaster, Pa., 1915. CUSHING, FRANK HAMILTON. 1886. A study of Pueblo Pottery as Illustrative of Zuni Culture growth. Fourth Ann. Rept. Bur. Ethn., pp. 467-521, Washington, 1886. 1896. Outlines of Zuni Creation Myths. Thirteenth Ann. Rept. Bur. Ethn., pp. 321-447, Washington, 1896. DOMINGUEZ AND ESCALANTE. 1854. Diario y derrotero de los RR. PP. Fr. Franscisco Atanasio Dominguez (1777) y Fr. Silvestre Velez de Escalante para descubrir el camino desde el Presidio de Santa Fe del Nuevo-Mexico, al de Monterey, en la California Septentrional. Documentos para la Historia de Mexico, segunda serie, tomo 1, pp. 875-558, Mexico, 1854. Dovauass, A. E. 1921. Dating our Prehistoric Ruins. Natural History (Jour. Am. Mus. Nat. Hist.), vol. xx1, no. 1, pp. 27-30, New York, 1921. 1929. The Secret of the Southwest Solved by Talkative Tree Rings. National Geographic Magazine, vol. LvI, no. 6, pp. 737-770, Wash- ington, 1929. FEewkKEs, J. W. 1898. Archeological Expedition to Arizona in 1895. Seventeenth Ann. Rept. Bur. Amer. Ethn., pt. 2, pp. 519-742, Washington, 1898. 1904. Two Summers’ Work in Pueblo Ruins. Twerty-second Ann. Rept. Bur. Amer. Ethn., pt. 1, pp. 3-195, Washington, 1904. 1909. Antiquities of the Mesa Verde National Park: Spruce-tree House. Bull. 41, Bur. Amer. Ethn., Washington, 1909. 1911 a. Preliminary Report on a Visit to the Navaho National Monument, Arizona. Bull. 50, Bur. Amer. Ethn., Washington, 1911. b. Antiquities of the Mesa Verde National Park: Cliff Palace. Bull. 51, Bur. Amer. Ethn., Washington, 1911. 1912. Casa Grande, Arizona. Twenty-eighth Ann. Rept. Bur. Amer. Ethn., pp. 33-179, Washington, 1912. 1914. Archeology of the Lower Mimbres Valley, New Mexico, Smithson. Misc. Colls., vol. 68, no. 10, pp. 1-58, Washington, 1914. Lid, 178 BUREAV OF AMERICAN ETHNOLOGY [BULL. 96 FEwKES, J. W.—Continued. 1916. Animal Figures on Prehistoric Pottery from Mimbres Valley, New Mexico. Amer. Anthrop., n. s. vol. 18, no. 4, pp. 585-545, Lancaster, Pa., 1916. 1920. Sun Worship of the Hopi Indians. Smithson. Rept. for 1918, pp. 493-526, Washington, 1920. 1923 a. Archeological Field Work on the Mesa Verde National Park, Colo- rado. Explorations and Field Work of the Smithsonian Institution in 1922, Smithson. Misc. Colls., vol. 74, no. 5, pp. 89-115, Washing- ton, 1923. b. Designs on Prehistoric Pottery from the Mimbres Valley, New Mexico. Smithson. Misc. Colls., vol. 74, no. 6, Washington, 1923. GUERNSEY, S. J., AND KIDDER, A. V. 1921. Basket-Maker Caves of Northeastern Arizona. Papers Peabody Mus. Amer. Archaeol. and Ethn., Harvard Univ., vol. vir, no. 2, Cam- bridge, 1921. See Kidder, A. V., and Guernsey, S. J.; also Nusbaum, 1922. GuTHE, C. HE. 1917. The Pueblo Ruin at Rowe, New Mexico. El Palacio, vol. Iv, no. 4, pp. 83-389, Santa Fe, 1917. 1925. Pueblo Pottery Making: A Study at the Village of San Ildefonso. Dept. Archaeol., Phillips Acad., Andover, Mass., New Haven, 1925. HARRINGTON, M. R. 1927 a. On the Trail of the First Pueblos. The Outlook, vol. 145, no. 10, pp. 805-309, New York, 1927. b. A Primitive Pueblo City in Nevada. Amer, Anthrop., n. s. vol. 29, no. 3, pp. 262-277, Menasha, Wis., 1927. HEWETT, Epcar L. 1906. Antiquities of the Jemez Plateau, New Mexico. Bull. 82, Bur. Amer. Ethn., Washington, 1906, 1909 a. Archaeology of the Rio Grande Valley. Out West, vol. xxx1, no. 2, pp. 6938-719, Los Angeles, 1909. (Reprinted under the title “‘ Exca- vations at Puyé, New Mexico, in 1907,” in Papers of School of Amer, Archaeol., no. 4, 1909.) b. The excavations at Tyuonyi, New Mexico, in 1908. Amer. Anthrop., vol. 11, no. 3, pp. 484455, Lancaster, Pa., 1909. (Reprinted in Papers of School of Amer. Archaeol., no. 5, 1909.) 1921 a. The Chaco Canyon and its ancient monuments. Art and Archaeology, vol. x3, nos. 1-2, pp. 8-28, Washington, 1921. b. The excavation of Chettro Kettle, Chaco Canyon, 1920. Ibid., pp. 45-62. : 1922, The Chaco Canyon in 1921. Ibid., vol. xrv, no. 3, pp. 115-131, Wash- ington, 1922. Hopes, I. W. 1921. Turquoise Work of Hawikuh, New Mexico. Mus. Amer. Ind., Heye Foundation, Leaflet no. 2, New York, 1921. HoiMEs, WiLtiaM H. 1886. Pottery of the Ancient Pueblos. Fourth Ann. Rept. Bur. Ethn., pp. 257-860, Washington, 1886. ROBERTS | PUEBLO RUINS IN COLORADO 179 HoucH, WALTER. 1907. Antiquities of the Upper Gila and Salt River Valleys in Arizona and New Mexico. Bull. 35, Bur. Amer. Ethn., Washington, 1907. JEANCON, J. A. 1922. Archaeological Research in the Northeastern San Juan Basin of Colorado during the Summer of 1921. State Hist. and Nat. Hist. Soe. Colo. and Univ. of Denver, Denver, 1922. 1923. Excavations in the Chama Valley, New Mexico. Bull. 81, Bur. Amer. Ethn., Washington, 1928. AND Roserts, F. H. H., Jr. 1923. Further Archaeological Research in the Northeastern San Juan Basin of Colorado during the Summer of 1922. Colorado Magazine, vol. 1, Denver, 1923-1924. Jupp, N. M. 1922. Archeological Investigations at Pueblo Bonito, New Mexico. Ex- ploration and Field work of the Smithsonian Institution in 1921, Smithson. Misc. Colls., vol. 72, no. 15, pp. 106-117, Washington, 1922. 1923. Archeological Investigations at Pueblo Bonito, New Mexico. LHx- ploration and Field Work of the Smithsonian Institution in 1922, ibid. vol. 74, no. 5, pp. 184-148, Washington, 1923. 1924 a. Archeological Investigations at Pueblo Bonito, New Mexico. Ex- ploration and Field Work of the Smithsonian Institution in 1923, ibid., vol. 76, no. 10, pp. 71-77, Washington, 1924. b. Two Chaco Canyon Pit Houses. Smithson. Rept. for 1922, pp. 399- 413, Washington, 1924. 1925 a. Archeological Investigations at Pueblo Bonito, New Mexico. Ex- plorations and Field Work of the Smithsonian Institution in 1924, Smithson. Mise. Colls., vol. 77, no. 2, pp. 83-91, Washington, 1925. b. Everyday Life in Pueblo Bonito. National Geog. Mag., vol. XLVI, no. 3, pp. 227-262, Washington, 1925. 1926 a. Archeological Investigations at Pueblo Bonito and Pueblo del Ar- royo, New Mexico. Explorations and Field Work of the Smith- sonian Institution in 1925, Smithson. Misc. Colls., vol. 78, no. 1, pp. 80-88, Washington, 1926. b. Archeological Observations North of the Rio Colorado. Bull. 82, Bur. Amer. Ethn., Washington, 1926. 1927. Archeological Investigations in Chaco Canyon, New Mexico. Ex- plorations and Field Work of the Smithsonian Institution in 1926, Smithson. Misc. Colls., vol. 78, no. 7, pp. 158-168, Washing- ton, 1927. Kipper, A. V. 1915. Pottery of the Pajarito Plateau and of Some Adjacent Regions in New Mexico. Mem, Amer. Anthrop. Asso., vol. 1, pt. 6, Lancaster, Pa., 1915. 1924. An Introduction to the Study of Southwestern Archeology, with a Preliminary Account of the Excavations at Pecos. Department of Archaeology, Phillips Academy, Andover, Mass., New Haven, 1924, 180 BUREAU OF AMERICAN ETHNOLOGY [BULL, 96 Kipper, A, V.—Continued. 1927. Southwestern Archeological Conference. Science, vol. Lxvi, no. 1716, pp. 489-491, New York, 1927. See also Guernsey, S. J., and Kidder, A. V.; also Nusbaum, 1922, and Kidder, M. A., and A. V. Kipper, A. V., AND GUERNSEY, S. J. 1919. Archeological Investigations in Northeastern Arizona. Bull. 66, Bur. Amer. Ethn., Washington, 1919. Kipper, M. A. anD A. V. 1917. Notes on the Pottery of Pecos. Amer. Anthrop., n. s. vol. 19, no. 3, pp. 325-360, Lancaster, Pa., 1917. MarrTIN, Pavut 8. 1929. The 1928 Archeological Expedition of the State Historical and Natural History Society of Colorado. Colo. Mag., vol. vi, no. Ls pp. 1-385, Denver, 1929. Morris, Haru H. 1915. The Excavation of a Ruin Near Aztec, San Juan County, New Mexico. Amer, Anthrop., n. 8. vol. 17, no. 4, pp. 666-684, Lancaster, Pa., 1915. 1917. Discoveries at the Aztec Ruin. Journ. Amer. Mus. Nat. Hist., vol. 17, no. 3, pp. 169-179, New York, 1917. 1918. Further Discoveries at the Aztec Ruin. Journ. Amer, Mus. Nat. Hist., vol. 18, no. 7, pp. 603-610, New York, 1918. 1919 a. The Aztec Ruin. Anthrop. Papers Amer. Mus. Nat. Hist., vol. xxv1, pt. 1, New York, 1919. b. Preliminary Account of the Antiquities of the Region Between the Mancos and La Plata Rivers in Southwestern Colorado. Thirty- third Ann. Rept. Bur. Amer. Ethn., pp. 155-206, Washington, 1919. 1921. The House of the Great Kiva at the Aztec Ruin. Anthrop. Papers Amer. Mus. Nat. Hist., vol. xxv, pt. 2, New York, 1921. 1924 a. Burials in the Aztee Ruin. Ibid., pt. 3. b. The Aztec Ruin Annex. Ibid., pt. 4. 1925. Exploring in the Canyon of Death. Nat. Geog. Mag., vol. XLVIII, no. 3, pp. 263-3800, Washington, 1925. 1927. The Beginnings of Pottery Making in the San Juan Area; Unfired Prototypes and the Wares of the Harliest Ceramic Period. Anthrop. Papers Amer. Mus. Nat. Hist., vol. xxvii, pt. 2, New York, 1927. 1928. Notes on Excavations in the Aztec Ruin. Ibid., vol. xxv1, pt. 5, New York, 1928. NELSON, N. C. 1914. Pueblo Ruins of the Galisteo Basin, New Mexico. Ibid., vol. xv, pt. 1, New York, 1914. 1916. Chronology of the Tano Ruins, New Mexico. Amer. Anthrop., n. s. vol. 18, no. 2, pp. 159-180, Lancaster, Pa., 1916. 1917. Excavation of the Aztec Ruin. Journ. Amer. Mus. Nat. Hist., vol. XVII, no. 2, pp. 85-99, New York, 1917. NussBaum, J. L. 1922. A Basket-Maker Cave in Kane County, Utah, with Notes on the Artifacts by A. V. Kidder and 8S. J. Guernsey. Ind. Notes and Mono., Mus. Amer. Ind., Heye Foundation, New York, 1922. ROBERTS | PUEBLO RUINS IN COLORADO 181 PEPPER, G. H. 1902. The Ancient Basket Makers of Southeastern Utah. Suppl. Journ. Amer. Mus. Nat. Hist., vol. m1, no. 4 (Guide Leaflet no. 6), New York, 1902. 1920. Pueblo Bonito. Anthrop. Papers Amer. Mus. Nat. Hist., vol. xxvu, New York, 1920. PRUDDEN, T. MITCHELL. 1903. The Prehistoric Ruins of the San Juan Watershed in Utah, Arizona, Colorado, and New Mexico. Amer. Anthrop., n. 8. vol. 5, no. 2, pp. 224-288, Lancaster, Pa., 1903. 1914. The Circular Kivas of Small Ruins in the San Juan Watershed. Amer. Anthrop., n. s., vol. 16, no. 1, pp. 38-58, Lancaster, Pa., 1914. 1918. A Further Study of Prehistoric Small-House Ruins in the San Juan Watershed. Mem. Amer. Anthrop. Asso., vol. 5, no. 1, Lancaster, Pa., 1918. ReaGAn, A. B. 1919. The Ancient Ruins in Lower and Middle Pine River Valley, Colorado. El Palacio, vol. vir, nos. 9-12, pp. 171-176, Santa Fe, 1919. Roserts, F. H. H., Jr. 1925. Report on Archaeological Reconnaissance in Southwestern Colorado in the Summer of 1923. Colo. Mag., vol. 2, no. 2, Denver, 1925. 1929. Shabik’eshchee Village: A Late Basket-Maker Site in the Chaco Canyon, New Mexico. Bull. 92, Bur. Amer. Ethn., Washington, 1929. See also Jeancon, J. A., and Roberts, F. H. H., jr. ScHMIDT, Ericu F. 1927. A Stratigraphic Study in the Gila-Salt Region, Arizona. Proc. Nat. Acad. Sci., vol. 18, no. 5, pp. 291-298, Easton, Pa., 1927. 1928. Time Relations of Prehistoric Pettery Types in Southern Arizona. Anthrop. Papers Amer. Mus. Nat. Hist., vol. xxx, pt. 5, New York, 1928. STEVENSON, JAMES. 1883. Illustrated Catalogue of the Collections Obtained from the Indians of New Mexico and Arizona in 1879. Second Ann. Rept. Bur. Ethn., pp. 307-422, Washington, 1883. STEVENSON, MATILDA C. 1887. The Religious Life of the Zuni Child. Fifth Ann. Rept. Bur. Ethn., pp. 5389-555, Washington, 1887. 1904. The Zuhi Indians: Their Mythology, Esoteric Fraternities, and Cere- monies. Twenty-third Ann. Rept. Bur. Amer. Ethn., Washington, 1904. Tozzer, A. M. 1927. Time and American Archaeology. Natural History (Journ. Amer. Mus. Nat. Hist.), vol. xxvu, no. 3, pp. 210-221, New York, 1927. TWITCHELL, R. EH. 1911. Leading Facts of New Mexican History. Vols. r-1, Cedar Rapids, Iowa, 1911. 182 BUREAU OF AMERICAN ETHNOLOGY [BULL. 96 Wuits, L. A. 1928. Summary Report of Field Work at Acoma. Amer. Anthrop., n. s. vol. 30, no. 4, pp. 559-568, Menasha, Wis., 1928. WISSLER, CLARK. 1921. Dating Our Prehistoric Ruins. Natural History (Journ. Amer. Mus. Nat. Hist.), vol. xxi, no. 1, pp. 138-26, New York, 1921. Page ACKNOWLEDGMENTS. -___------- 1 ADOBE, structures of________-_- 72 AGRICULTURE, of the Basket Makerscs a. 33a e cl hr aes 5: ALL-OVER DESIGNS__-__---_-- ve 133 AMERICAN Museum oF Natura History, work of... =. 12 ANIMAS RIVER, ruins along_- -_-_ 67 ARBOR— associated with adobe build- 10 0 sas eye gees © oe ees eee 72 (EE COS KO Rees: ete eed 49 ARROWHEADS, description of _ __ 153 ARROWMAKER Bonist, _ inter- mentealled...' 22.22 21 3 164 ARTIFACTS— period identified by_------ 10 perishable, disappearance O}t eee he ee eee 73-74 AWATOBI, period of_____-_-_-- 8 Awis ol bone...2---2==-=..- 145-146 AX HEADS, description of__~__- 150 AZTEC— age of, compared with Pueblo Bonito___.------ 11 occupied by Mesa Verde Meople Sl 248 es Liss Soe 12 ruins in vicinity of __---_-- 67 AZTEC RUIN, period of___--_--- 7 Baker, Uutima, acknowledg- TCT OO se ee Bee i BAND DECORATIONS--_------- 1380-1382 Basker MAKER CULTURE, tran- sition from, to Pueblo------- 168 Basket MAKER RUINS, under Purell ees Ma bin ay Peas 9 Basket MAKER VILLAGE, skulls ER OUT ee ie exer at Ger) AS 2 | 71 Basket MakEeRs— absorbed by Pueblos_----- 169 agriculture adopted by - --- 3 approximate date of______ 12 characteristics of __-..-_-- 2-3 dwellings of 2-4. 2. oe = 3 PAUCTOL se ee ee ie 4 influence of, on later houses_ 165 periods in culture of__-_-~-~- 4 88242°—30 13 Page BASKEtRY, influence of, on ce- TAT CS a ee 97, 109-110 Braps— made of bone____.---___-_ 147 made of stone_______-----_ 154 Beaver, Urau, ruins uncovered Gere 2 eee ee eee 71 BETATAKIN RUINS, period of___ Bins. See MBALING BINS; STORAGE BINS. BirpD-FORM VESSELS— possible significance of _____ 103 SNAP CS Obs ee 102-104 SIZCS Ol- 2 ee _. 104-105 BLACK-ON-RED WARE— COLOTINS Of =e ee 78 decoration of__________- 138-139 BLACK-ON-WHITE DESIGNS, elab- orate patterns of ....-.----- 135 BLACK-ON-WHITE WARE_______ 77-78, 112-1388 BLANCO RUINS, period of_______ 8 BoNnE OBJECTS— provenience of__________- 175 8126) Ofiecaca%2224252 555528 175 Bones, record furnished by___- 144 BoONEWORK~____------------ 144-148 BowLs— black-on-red ware___-_--- 138-139 characteristic features of ___ 99 fOr COOKING = - 222 eee 85 shapes Of 2.2222 52-—=s2 99-100 SIZES Ol 3 ae ee 101 with handles______.------ 101 BuRIALS— church erected over_------ 50 detailed description of... 158-164 distribution of, in mound_ 158, 159 in refuse mounds__-__ 36, 155, 157 in? SLOLage DINSe 223s ea 3 investigation of____----- 155-164 miniature vessels found in-_- 109 of hands without a body. 163-164 of heads without bodies_-- 163 of supposed Arrowmaker--- 164 on edge of bluff___.----- 155, 164 orientation of bodies in_. 156, 168 183 184 INDEX BuriaLs—Continued. Page : Page position of body in________ 156 | CoLtor or pottery, affected by sex percentages of________ 163 PUT Be ec ee 78, 81, 112 suggesting that of medicine CoLoraDo RUINS, period of____- 8 PANS ae eel ak 2a. 158 | COMMUNAL STRUCTURES, devel- under floor of house______. 155 opment Of 42 Sosa. eee 5 Burns Canyon, remains in____ 66 | Conn, storage of........--__ 2 29-31 Casa GRANDE RUINS, period of __ 7 | Corron, introduction of____-_- 5 Casa GRANDE VILLAGES, prob- Court, formed by projecting ‘EETCa 6) ieee a el ae 8 wWallsstres ood. ease 52 CERAMIC INDUSTRY— aan CovERs, STONE, for pots____- 151-152 development of_..____- ___ 74-75 | Crystats. See QuaRTz cRYs- place of, in Southwest cul- TALS. TLE) 2 pee eae ee ee 3 | CYLINDRICAL VESSELS, black-on- transitional stage of ______- 167 red: Wares. es eee ee 139 CEREMONIALS, seasons for__-___ 32-33 | DANCE PLAZA, depression used Craco CANYON RUINS, period of - 7 as 2co5 oY PO ee eee 33, 70 CHALCEDONY, implements of__ 152 | Darss, tentative, of Pueblo pe- CHAVES PASS RUINS, period of __ 8 TIOdS*= 52s eee 11-12 CHECKERBOARD, element in dec- DeEcoraTIoN— OrTALION- = — 223. 118-119, 125-126 characteristic of period__-_ 109 CHEVLON RUINS, period of-_---- 8 field for___.._.._.._____.. 111 CuEvron, form of panels__-_ 125-126 of pottery_________.__- 109-139 CurmmuanuA Basin, abandon- on jar and pitcher forms_ 111-112 MAGN Ol oe ee 6 See also Designs; ORNa- CuiImNEY Rock Mrsa— MENTS. archeological work on aes 14 Ounce ein hg of Piedra Fr Absence OPlonie. eee 47 TYSISL gC (oar we eae ny is 6 CHIMNEY Rock RUIN, position PUGS Ole eo ae oi aren eee breed 18 DrEPRESSIONS— CHISELS, of bone___._.______- 146 position of. --~-~--------- 21 CaRonoLoax— suggestive of great kivas___ 70 Hedie Puchleg-cac = awe 4-8 used as dance plazas- £2 2,35, 00 relation of cultural objects used as reservoirs. ------ 33, a aa 10 53-54, 55-56, 58, 59 Scuthwestern, means of es- uses found for__-.--.----- 21 tablishing....-.-------- 9-12 | Destens— Cuurcu, Mexican, on Pueblo basketry, used on pottery-. 110 burial grounds____.._----_-- 50 black-on-white- ~~ ------ 112-138 CircLE DESIGN, possible sym- boldness cf conception of __ 167 bolism of...._.........__... 114 elementsim= 232 eaeee 111 CLAN— interpretation of symbolism group of structures occupied Ole ee as tee ee ee 110 eel arn Oa en ee 38 on painted pottery __---- 109-139 portion of village allotted placing of 2... 23 3S3 ee Itt Primate gard Ek he WES ents 35 realistic, contemporaneous Cuark, W. E., acknowledg- with geometric_-------- 110 mentor. re Wh awh 1 See also Decoration. CLoup BLOWERS— Dickens, W. F., acknowledg- gescribed= 222. 22-22 141-142 ment to2:.2t. 02s sae ee 1 period differences in_-__---- 142 | Dipprrs, described. -___-_---- 101-102 SIZCNO La es ce ae Se ee 142 | Domineurz, Francisco ATANA- RISCKOLe ee 4s a ee Sas i4] SIO, explorationster.- 2-2 ose 18 INDEX 185 DoorwayYys— Page Page between rooms. -_.-_-_-_-_- 41 | Game, abundant in Piedra dis- doubts concerning - - _----- 24-25 Tet 22h ae a 17 probably inclosed _-------- 93 | GAMING PIECES, made of bone__ 147 SIZOLO Meee ee ee ie eee 24-25 | Gira RIVER REGION, desertion Dots, used in band decora- OFS 2 eae hs ee his le ee 6 Hone eee eee ae ee 130-131 | Guazrna, on decorations-_------ 112 Dovatass, A. E— GOVERNADOR CANYON, remains method of, for dating struc- IN _~--~--~---~----~-~-~-------- 66 PAS + A he eS g | Grain. See Corn. study of growth rings by --_- 11 | GRAIN PITS— Ducx— SCATCIbY Obese St ee 29 vessels suggestive of__.-_-. 103 structures serving as_-~_-- 30 Zuii belief concerning_---_ 103 See also STORAGE. DweELLIncs— Hasitations. See DWELLINGS; erected over pits..._------ 29 Hovusss; PIT DWELLING. in groups__-------------- 91 | Hammuns, description of -_----- 149 of jacal type construction. 19, 20 | HANDLES OF VESSELS— Pueblo, development of __-- 5 fORtms?) O04 ae oases 91 three-room, deseribed_ __ _- 3¢ position;).of2_ 2. £2 2eee 88 See also Hovusss; Pit Harrineton, M. R.— DWELLING. archeological work of_--_--- 72 Erricy Forms, in vessels_-_-__- 108 mentiom of _-.-=----.--t 150 See also BrirRD-FORM VES- Hawikuy, period of___-------- 8 SELS. HEAD DEFORMATION— ELEMENTS IN DESIGNS— evidence of_...---.------ 10-71 eth Pee ie om Pe Tht theories concerning ------ - - 157 Mevative.s seen o olla ky 193. | Hons, described 22- 22 destroyed by fire-__------ 19 portions of, allotted to ClanSs2 432: 2S ee Seen 35 VITRIFICATION. See GUAZING. 190 INDEX ‘ W ALLs— Page change in, from slanting to perpendicular--_-------_- 36 construction of _____ 22-23, 37, 51 projecting, to form court _ _- 52 significance of change in__~_ 36 SUppOLbOl = =e. snes 22-23 timbers at base of - - - ----- 25 WATER JARS, typical forms of__ 89-92 Watts, GEORGE, acknowledg- WHISTLES, of bone =. <=2=4 WHISTLING JARS, origin of the WiLuarD, Uta, ruins at— described by Judd_______- similar to Nevada ruins-_ -___ ¢ e Page 104 Lk ee Ly bed 6,4 ath be! nian tae a a saeneed = 5 yp ; ining ie Pitas 46 1 abs alti ey if ‘ tay, i} \ et A ely ION LIBRARIES 4 atv ye ; BY tales ba i RLaepsyene gee fey ut haa tie f i, us) Pia 105 one vn ti ail He 4 iefigeh eh Ns Perk A * 5 SMITHSO! ccm aan Maem A Aa i hie stint 1 His weit I a Ad oat Hanahan Rrepaetee at Hi aM iit bp 1 haga ie F j ft Ass j ¢ i ei: of Parsti a te tint et Rote) nn De eds pine eth Arel'slet Ki vit te f 4 ru ay i Pome ie ie raaeectiet ty ak 4 jv erent epee He ie rab ithe oft het a) " quent Pye LE 4h a dae Daye gi8 bal - W *y OR ST eo eer reen est, bye Fb feet . pT scene hehe ile & en ORY ify Hiheryeone beset , , pel ? Felten hee +g { ERMINE Pal onaertty : tet Wty Bagh Leieivornyt th aie eye Hi f , ; j peiphpnrnieee : mr) Pia tauice : eoneden eety : + fe gimp ate Hea) 4b 4 H hee Mee rivet rele Bee haces pet bel f i , A mene ae : at Hie ee uli ge ibe ot Paes pret euhory) la ie the ; ] pete ot ; ROBOT Ee ie pi nteitce oak HF prs nlbey ont / ¢ Be bite iat it ny : it, AMEE fa \ } Fess “in ; ! 4 . i re f ‘ . 4 iy iad “ ' i f ' wually out 5 1 Seat ve Liueeak eae aired Path? Faihpe deren , (ue Pokal bet : : * eran Pewee Medan «yk feet beeeive MoH ‘ , ' Hehe if opateet ( ‘ Li faagy av ‘ i Ht Ji st é Te iteetey eu en DisAts ioe , pan senere q i ‘ { F fatgt of “ ¥ Baa ‘ : 1) Serva eeou ened ete! { ’ i } ‘ ry ite ‘ , ‘ ' t ' } { vated eat F : f ‘ i i ' ¢ i si } 5 f i € ‘ \ oi \ Aly BAR ier i va Neate ? baat ; y H ‘ a ‘ es Li \ i ‘ i ath Lense at ‘ ' f DRUPAL OY "He uf qe? \ ’ vi {QuisetOn suey ‘ by on rat soe y in , A eviitasen ee pict i ve ite rp ; d “ - } ; ptyey 45 Bu ri : eiLemee st 1 \ v hac ay Mid Bok i aie hey t stitute Anbu we yal we ‘ y i i stotenite any cturaras Leth ute ake use i ‘Gru 1, b PWPETL & it pale Mead if Aegathcy Risueuracaee wseet fa I ‘ RULER Py: fa ’ ) . i i ie ara : Saige “at ie Raat ‘ U AOU FAGSS TE at t see rg AP Hig i isesae Fin Sd aetaut cy zp uenty ie ‘ustial an i ast 4 ( } y aetieksp sie ey Tite Heh eg AYN : t ‘ {Fi $ ‘ Vue i eas ma EERIE ENS tay } i" 4 ng fix TR f rtG f Ye ’ Buhl Ge nsed K yeniel eye t Hebsh ; Dil aaa Lyrae Tagseuanea gi weve Meet Lhe shu aiit \ Prearoeuid iA th tEuren} \ Gteatiene tals ‘ Tasusarts v We ry yeUdNey® y i t a risk \ ‘ PRs ave i byt ‘ i Pte awn v! " teh ep Ns ees dish eeerd tt 1 eet § pashae sh i DE he’ R he VRE 4 Wee: ert 4-4 Ci ‘ / at i st Hotes aha see y Sete f april ; : REN eke is enor ietety ih “abl ‘ fas } Ay \ : OVE GAR $ , sy verataradt shaymnsueeegay x Bei i ih} aap MOeEY bus he Pays te Lt aes f ene A wy ; { Nats tieatisusisn’ Fea ns" se ig Pabive ; i : wo pel ‘ Oe bal vat ; ? { ey Meek ane ek? 83 H Aureus Pita Rete NaS 1 whe b meee parr Hit har i i bs hy Plant hit, Pern kad Bit i. hei sy) HyeLay cues jenn eusardss het vies ‘ eee AHL gi iied LA Useeverns sale VAL KO St } if Say : Ee jeer f | : yet ane f ey ravens Da ‘ ty AEP WEL pital anode AL ena j he: ! { he € y - D Ay i 8 t t ye ES ‘ a, bia ‘ LEC . \ eee he i Pomehgutereh el wt shh uM shoes h wnat bee mn ¢ ein ira eit i Wee HA cpeeutery) Yi bs Pe i sa me i": ' ve al hehe us fel \ Sua Hier abe) \ eM (ean ? ii" yard We neieae De Dial " ThA i is Ny Pee ieiishie§ wert’ eal Pen aye prieeevit he UR ett cer he Ht aa Sihieaht wr HEHE DEL cued eR ee ay Selva ir an I 4 Gin Ae Vey : Ryda abe Has { Pen ebs MOAN nie raha we st yey ely yt i Sats TH) aay ue DUH UBM ie AG tn i in al hy sancant rae 4 i it hess i wy \ ui mG tad ‘yale FUT BPA eons wit iin ia eta YH He Ara my ae ite thi con i i a Hi ce eeat CMa it Nhe Til ee iy oh i pea at at RRM ie ! at EAE BRE PC a PtSi i eat NS kL) nh