.N 
 
 
 ■.^^ 
 
 t*. 
 
 0? 
 
 *< 
 
 .0' 
 
 *°-^ «5°* i?^ 
 
 
 
 
 
 
 
 
 
 -A < ' *o , » * O * . . s * S\ <. 'o . » * .0 O < . . S 
 
 ^fv 
 
 vP c,^ 5 
 
 
 
 r * 
 
 •** 
 
 <?■ o n o ' 
 
 O > 
 
 -° 
 
 
 
 
 0^ 
 
■5 ^ 
 
 
 ' ,^ ^ v^ 
 
 i-o ^ > 
 
 ^ ^ .' * 
 
 V *>-. 
 
 ' <$> V *•' 
 
 ^6* 
 
 o V 
 
 o • » .0 o /\ <r 
 
 O 
 
 V 
 
 * V ^ 
 
 ,0 O ' . . 5 s • "\ 
 
 0^ o • " • , ^o ^ 
 
 <* 
 
 ■^0 
 
 
 •o V^ 
 
 
 
 ■<* 
 
 ,v ^, •. 
 
 ,V 
 
 .o v 
 
 >bK 
 
 
 * ^ 
 
 o 
 
 Z- t 
 

 
 * . 
 
 
 
 > 
 
 
 
 #: 
 
 V 
 
 
 
 ?V * 
 
 ** 
 

%1 
 
 s? 
 
 ■■ 
 
 A FURTHER STUDY OF PREHISTORIC SMALL 
 HOUSE RUINS IN THE SAX JUAN 
 WATERSHED 
 
 BY 
 
 T. MITCHELL PRUDDEX 
 
 ,AN W 1919 
 

 PRESS OF 
 
 THE NEW ERA PRINTING COMPANV 
 
 LANCASTER, PA. 
 
A FURTHER STUDY OF PREHISTORIC SMALL HOUSE 
 RUINS IN THE SAN JUAN WATERSHED 
 
 By T. MITCHELL PRUDDEN 
 
 IN an earlier paper on the "Prehistoric Ruins of the San Juan 
 Watershed," ' it was noted by the writer that among the several 
 types of ruins widely scattered over this extensive area, the 
 most abundant and uniform in character is the small pueblo, 
 standing in the open country; now usually reduced to a jumbled 
 heap of roughly trimmed, much-weathered stones — the "Mounds" 
 of the regional vernacular — and mostly grass or sage grown. With 
 this small house ruin are almost invariably associated the site of a 
 circular ceremonial chamber and a burial mound, both as a rule 
 lying to the southward of the pueblo. This habitation complex 
 was called the "Unit Type" for reasons set forth, together with a 
 description of its details and variants, in the paper just referred to 
 and in a subsequent study.- The abundance and apparent uni- 
 formity of this type of small open country ruin, and its topographic 
 and cultural relation to other and more complex examples of pre- 
 historic ruins in this region, both in the open and in various forms 
 of cliff shelters; as well as its interest as an obvious expression of 
 family, clan, or other social relationships among these primitive 
 house-building people, led the writer to undertake to secure, 
 through such excavations and other studies as were practicable, as 
 precise a conception as might be of the life and culture of these 
 early folk who made the little ranch houses in the open country. 
 For the question of the relationship of these people to those who 
 also in this district made the more complex, picturesque, and often 
 stately communal dwellings, is one whose solution would seem to 
 
 1 "The Prehistoric Ruins of the San Juan Watershed, etc.," American Anthro- 
 pologist (N. s.), vol. 5. (April-June, 1903)- Pp- 224-228. 
 
 2 "The Circular Kivas of Small Ruins in the San Juan Watershed." Ibid., vol. 
 16. (Jan. -March, 1914)- Pp- 33~58- 
 
4 AMERICAN ANTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 depend upon a detailed independent study of each type of archi- 
 tecture and of such associated artifacts, belongings, and conditions 
 of life of each group as may have survived the vicissitudes of time 
 and exposure. 
 
 The first year of this special study— 1913— was directed par- 
 ticularly to the underground kivas, which in the four ruins then 
 examined were found to exist under the shallow depressed area 
 close at the southern side of the pueblos; partly filled by the wreck- 
 age of their own roofs and walls and of the front walls of their adja- 
 cenl houses, and buried by the sand drift. 
 
 These underground, circular, stone-walled and pilastered cham- 
 bers all resembled kivas already known in some of the larger ruins 
 of this district, in their recesses and pilasters, while those com- 
 pletely exposed showed a ventilator shaft, a deflector, a fire pit, 
 and a door and underground passage leading upward from the kiva 
 into the adjacent pueblo. The various bone and stone implements 
 discovered were of the forms common in the larger and smaller 
 ruins of other types in this district, while the pottery conformed to 
 the type outlined by Fewkes as characteristic of the San Juan area. 1 
 Details and illustrations of this study may be found in the paper 
 referred to. 2 
 
 As it seemed desirable to gather additional data over a still wider 
 range, further field studies were made in the summer of 191 5 which 
 it is the purpose of this paper briefly to record. 3 Three more ruins 
 of the unit type were examined, lying many miles apart but all in 
 sagebrush openings upon the great pinon, juniper, and sage clad, 
 canyon-scored mesa, which on the borders of Colorado and Utah, 
 slopes upward toward the McElmo canyon and Ute mountain. 
 
 There are many ruins, large and small, upon this rugged upland, 
 from the vicinity of Cortez and Dolores westward to the great mesa 
 tongues which fall away toward the San Juan river from the foot of 
 
 l. is, "Antiquities of the Mesa Verde National Park Spruce Tree House," 
 Bulletin 41. Bureau of American Ethnology (1909), p. 35. 
 
 American Anthropologist (n. s.), vol. 16 (Jan. -March, 1914). 
 1 was fortunate tins season, as in so many previous summers of field study, in 
 having tin i nnperation of Clayton Wetherill, to whose knowledge of the country and 
 energy and ^kill in search and excavation I am greatly indebted. 
 
PRL-DDENJ PREHISTORIC SMALL HOUSE RUIXS 5 
 
 the Blue mountain (Sierra Abajo). With the exception of the 
 Goodman point ruin, these are little known save to the cattlemen, 
 to whom their significance lies largely in the probability of some 
 accessible water source near by. In the canyon bottoms of the 
 larger stream courses, which, cutting deep across this uplift, carry 
 the Blue mountain water to the San Juan — namely Montezuma 
 Recapture and Cottonwood creeks, and the Butler and Comb washes 
 — there are also many small ruins of the unit type. Of the several 
 hundreds of these known to the writer from his earlier reconnais- 
 sance survey of this northern San Juan district, it seemed that the 
 most instructive, in completing the picture of the builder of these 
 simple homesteads in the open country, might be those which had 
 given the impression of relatively great age. This is especially the 
 case with many of the small ruins among the pinons on this great 
 plateau. For many of them show excessive weathering and disinte- 
 gration of such of the pueblo building stones as are exposed above 
 the sand drift, by which and the sage, in theirstate of surface denuda- 
 tion, they are often largely covered and sometimes almost com- 
 pletely concealed. Fragments of standing wall are seldom if ever 
 visible, while here, as elsewhere, the kivas are almost completely 
 filled with wall and roof wreckage and sand drift. The writer has 
 now in mind relative, not absolute, age, and is not unmindful of the 
 fallacies to which in either problem one may yield, if he be not dis- 
 creet in judging of age in these prehistoric structures without due 
 allowance for rapid changes in appearance, which may be wrought 
 by erosion and the sweep and fill of the sand drift and the action of 
 the rain and the frost in these lofty open regions. But since some of 
 the ruins of our type already examined had shown better surface 
 preservation than many of those in other regions, it seemed wise to 
 go well afield and include both forms in the attempt to determine 
 the characters of the type, without immediate regard to the question 
 of either relative or absolute age as indicated by the state of physical 
 disintegration. 
 
 The results of this further series of excavations in their larger 
 features are practically identical with those of the earlier series, 
 and would seem to establish the uniformity of structural type in the 
 
6 AMERICAN AXTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 dwellings which these ruins denote, and the identity of the culture 
 of their makers over a wide area in the northern San Juan country. 
 
 But in order to record the data from which a proper summary 
 can be framed, it will be necessary, in spite of repetitions, to recount 
 briefly the details of each of these excavations of the present series. 
 
 Finally the attempt will be made to summarize the structural 
 characteristics of the ruins of this type on the basis of the combined 
 field studies and to make such a sketch of the life and culture of this 
 primitive housebuilder as the various observations may seem to 
 justify. 
 
 Owing to the striking uniformity disclosed in orientation and construction, 
 we have found that we can most economically lay out the plan for excavation in 
 these ruins, by first determining the exterior lines of the back and end walls of the 
 pueblo. This in most instances may be readily done by starting at the outer 
 edge of the stones in the northern throw of the fallen walls, and working inward 
 at a right angle to the long axis of the mound. Thus usually one soon 
 comes upon the remnant of the north wall in place, which may be followed 
 to and along the ends of the house. With the length of the pueblo and 
 the direction of its long axis thus established, one can start a trench 
 across the kiva pit at a right angle to the pueblo at its middle segment, 
 with a reasonable assurance that a trench 4 feet wide will include the ventilator 
 opening and shaft, the deep southern recess; the fire pit and deflector; the northern 
 recess and the door from it to the passage into the pueblo from the kiva; the pas- 
 sage itself and the manhole in the floor of the front room. Such a trench will 
 probably include also, if they are present, the sipapu and one of the niches in the 
 northern wall of the kiva. This plan of operation is of value in the economy of 
 labor, because the exact situation of the pueblo is not usually evident in the undis- 
 turbed stone heap, and the center of the kiva depression (so-called "kiva-pit") 
 on the surface is not apt to correspond very closely to the center of the buried kiva 
 itself. The excavation, especially of the deeper parts of the kivas, is frequently a 
 toilsome process, since toward the bottom the building stones and the adobe soil 
 and drift are apt to be cemented into a solid mass, requiring the constant use of 
 tin pick and trowel. The determination of the floor level of these kivas in the 
 hard mass is not always easy, and we have found that by a generous wetting of 
 tin- deeper layers, as the bottom is approached, the last two or three inches of 
 filling often peels off from the floor with ease, and with much greater safety to 
 artifacts King upon the floor, than is secured by even the most careful manipula- 
 tion of the pick and trowel and brush in the hard dry material. 
 
prudden] PREHISTORIC SMALL HOUSE RUINS ~ 
 
 Ruin No. V 
 
 The first ruin here to be described, which in the entire series 
 may be called Ruin no. v is on the Cohon mesa in western Colorado, 
 and lies about three quarters of a mile eastward of Ruin no. iv of 
 1913. 1 
 
 It is about one quarter of a mile west-southwest of the point at 
 which the trail crossing the Cohon mesa from the Picket corral at 
 the edge of Ruin canyon 2 goes over the rim of the Hovenweep can- 
 yon to Risley's spring at the bottom. I estimate from the location 
 of his nearest monument, kindly furnished to me by Prof. R. C. 
 Coffin, that this ruin is approximately i6}4 miles northwest of Ute 
 mountain. It is in a sage-grown opening in the pihons and junipers, 
 about 50 yards across. There are many ruins of our type widely 
 scattered on the Cohon mesa, as well as numerous larger and more 
 complex forms along the edge of Ruin canyon, which itself shelters 
 interesting cliff houses. 
 
 This ruin, no. v, was selected, first, because I wished to determine 
 more fully than was practicable by the partial excavation of no. IV 
 in 1913, the details ol structure in the Cohon group; and second, 
 because it presented the simplest form of a compact isolated house 
 which it seemed interesting to compare with some of the units set 
 in rows which were in the season's prospect further west. 
 
 There are several small ruins of the same general character as 
 this no. v within the radius of a mile on this segment of the Cohon 
 mesa, the nearest about 250 yards southwest. Water is accessible 
 to them in a small spring under the rim rock of Ruin canyon to the 
 west, while to the east is the more abundant Risley's spring, at the 
 bottom of a small offset of the Hovenweep. 
 
 The preliminary inspection of no. v reveals a densely sage- 
 grown stone heap about 40 feet long and 20 feet wide whose top 
 rises about 2}4 feet above the general surface hereabouts, its long 
 axis running east and west. The kiva site, close to the south, is 
 
 1 Loc. oil., American Anthropologist, vol. 16 (Jan. -March 1914), p. 56. 
 
 2 There is another "Ruin canyon," a lew miles southward, one of the tributaries 
 of the Yellow Jacket, and containing well-preserved houses and towers, to which visitors 
 were frequently conducted in the early days by the Wetherills by way of the McElmo 
 canyon and the mouth of the Yellow Jacket. 
 
s AMERICAN ANTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 depressed at its center about 15 inches. The whole area of depres- 
 sion is about 25 feel in diameter. 
 
 The nearest edge of the burial mound is about 60 feet south- 
 easterly from the center of the kiva pit; it is slightly raised above 
 the general level; its soil is darker in color than the surrounding 
 natural earth and it has been completely despoiled by pot-hunters. 
 
 On the northerly side of the ruin, and about 60 feet away, is a 
 semicircular mound of earth about 50 feet long, now about 2 feet 
 above the general level along its axis, and about 5 feet wide at its 
 base, and extending on either side beyond the ends of the pueblo. 
 The space between this earth mound and the pueblo, open at the 
 ends, is crossed by three radially placed incomplete rows of large 
 trimmed stones laid in a single tier, dividing it into segments. 
 Such lines of stones radiating from the pueblo, with and without 
 the partially enclosing mound of earth at its back, are to be seen in 
 some of the other ruins of this type on the Cohon mesa. 
 
 The Pueblo. — The pueblo proved to be about 30 feet long and 
 16 feet wide externally, and consisted of two rows of rooms, three 
 rooms in each row; the rooms being of unequal size and shape (see 
 figure 1). 
 
 The front wall stood about 8 feet north of the inner circle of the 
 kiva. The yet standing walls of the house were from 10 to 12 
 inches thick, laid up of roughly trimmed sandstone of the region. 
 The surface of some of the building stones was fairly smooth, while 
 others were deeply indented with pit-like or linear marks of trimming 
 utensils. We completely cleared only the middle room of the 
 south row, revealing, close to its front wall, the manhole to the kiva 
 passageway. 
 
 The center longitudinal wall was standing about 30 inches high. 
 The others were more thrown down, but with the lower tiers in 
 place. Only in the center of the front wall of the middle room the 
 entire structure had slumped, together with the wall of the south 
 segmenl of the manhole, into what proved to be the collapsed tunnel 
 to the kiva. The manhole to this tunnel (see ground plan, figure 1) 
 .il" mt 24 inches in transverse diameter, was edged on its northern 
 intacl segment, with flat stones forming the surface of the floor 
 
EnzziTi^iniHioxnr] 
 
 IxiiEEBrEES-ir rrF^m 
 
 ^aHxcraBZHxcH 
 
 Scale 
 
 4- Feet 
 
 Fig. i. — Ground plan of Ruin no. v. on Cohon mesa, a, Manhole into tunnel to 
 kiva. b, Underground tunnel from front room of pueblo to kiva. c, Cubby-hole or 
 niche in north wall of kiva beneath banquette, d. Pilasters for the support of the roof 
 timbers, e, Recesses between the pilasters with banquettes about 33 inches above the 
 level of the kiva floor. /, Deep southern recess and banquette, g, Horizontal portion 
 of the ventilator flue, opening within at the floor level, h, Perpendicular shaft of the 
 ventilator with the opening at the ground level outside the kiva wall and protected by 
 flat stones now displaced. *, Sipapu in floor of kiva. k. Fire pit in floor. 
 
[0 
 
 AMERICAN ANTHROPOLOGICAL ASS0CIATI0X [memoirs, 5 
 
 hereabouts, while below, the gradually widen'ng, and here nearly 
 perpendicular passage was lined with thicker rough-edged flat 
 stones, well set. The floor "I this center front room was in part 
 plastered, in part surfaced with packed earth, and lay as did those 
 ol the other rooms, about 10 to 12 inches above the level of the 
 general surface of the .mound about the pueblo. 
 
 The Kiva. — (Figures 1 and 
 2.) The excavation of the kiva 
 revealed such structural fea- 
 t ures as I have previously found 
 in this type of ruin. The di- 
 ameter of the inner circle was 
 about 13^ feet. The ban- 
 quettes averaged about 53 
 inches in length along their 
 inner edges; were about 33 
 inches above the floor level; 
 and were from 18 to 20 inches 
 deep. Their free edges were 
 set with flat stones laid in 
 adobe, the rest of the bench 
 being surfaced with the latter. 
 The backs of the recesses, ex- 
 cept the deep southern one, 
 were plastered upon the earth. 
 The south recess w r as about 6 
 feet deep, with the sides and 
 back stoned. The back of the 
 north recess, lying tow r ard the 
 pueblo, was, like all of the 
 others, nearly perpendicular, 
 i xcept that in this case the up- 
 per part sloped sharply backward, and like all the others its surface 
 w .1- deeply smoked. 
 
 The pilasters were from 26 to 27 inches wide upon the face, and 
 the top ol those besl preserved rose about 21 to 24 inches above the 
 
 Fig. 2.— Northerly segment of kiva. 
 Ruin no. v. This shows two pilastei th< 
 
 ing of the vaulted tunnel through the 
 
 flooi ol the front room of 
 
 the pueblo which starts at the level of the 
 
 northern banquette; the cubby-hole in the 
 
 wall "i the kiva below the I anqui tt< ; the 
 
 • u in the flo< tween 1 he fire-pit and 
 tlu nertli wall. 
 
pruddex] PREHISTORIC SMALL HOUSE RUINS II 
 
 level of the banquette. The well-faced stones forming the inner 
 circle of the kiva were on the average about 3 inches thick and 4 
 inches long, well set in adobe, and were smaller, as is usually the 
 case, than those forming the face and sides of the pilasters. Both 
 were thickly plastered; the plaster, however, readily flaked off on 
 exposure and its fracture showed six layers. The floor of the kiva 
 was about g}4 feet below the general floor levels of the pueblo. 
 
 Near the center of the back of the north recess we found an 
 arched doorway from the banquette level (figure 2) about 20 inches 
 wide and 23 inches high in the middle. Beyond this a similarly 
 arched passageway, now filled with a densely packed mass of build- 
 ing stones and earth, led directly with a slightly rising floor, to the 
 line of the manhole in the middle front room (figure 1). The walls 
 of this passage and its doorway and floor were formed of adobe 
 plastered on the earth, and the surfaces were densely smoked, 
 7 inches below the middle segment of the northern banquette is a 
 niche or cubby-hole 1 1 inches wide, 8 inches high and 10 inches deep, 
 lined with flat stones and adobe (figure 2). 
 
 In the floor of the kiva, at about 7 feet 6 inches from its north 
 lower wall, is the northern edge of a circular fire pit, 27 inches in its 
 north and south, and 30 inches in its east and west diameters. The 
 edges of this fire pit are in part formed by flat stones set in adobe, 
 in part by the adobe floor carried over the edge. It was about 12 
 inches deep and full of clear ashes with few charcoal fragments. 
 
 Nearly midway between the fire pit and the north wall is a well- 
 constructed si pa pit (figure 2), the opening of which is 3 inches in 
 diameter. This is formed by the neck of an olla irregularly frac- 
 tured from its body. The flare of the rim of the olla is about a half 
 inch, it is set flush with the plastered floor of the kiva and is decor- 
 ated with four symmetrical clusters of three painted bands crossing 
 the flare of the rim. About 2 inches below the top of the rim the 
 sipapii widens to about four inches in diameter and from the opening 
 is about 5 inches deep. The flask-like cavity is lined with small flat 
 stones and irregular pieces of broken pottery set in adobe, with 
 which the bottom is formed. The sipapu was filled with sand and 
 was without a cover. 
 
12 IMERICAN ANTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 Beneath the middle portion of the southern banquette is a door- 
 wax- to the horizontal tunnel of the ventilator, 22 inches high and 
 15 inches wide, with a flat stone lintel and sill. The bottom of this 
 tunnel is on a level with the floor of the kiva, sloping slightly up- 
 ward as it recedes. Outside and close behind the south wall of the 
 deep recess, about 18 inches below the general surface of the ground, 
 we found the square stoned opening of the perpendicular shaft of 
 the ventilator (see figure 3) about a foot in diameter. This shaft 
 was followed down through several tiers of flat stones in place, but 
 we did not have time to complete its connection with the horizontal 
 portion. 
 
 Between the kival opening to the ventilator and the south side 
 of the fire pit lay two rough stone slabs, approximately 23 inches 
 
 high, 15 inches wide, and 3^ 
 inches thick, fallen over the 
 south side of the fire pit. 
 These I conjecture formed a 
 deflector when standing up- 
 right. The floor of the kiva 
 was smoothly plastered. 
 
 In this ruin a north and 
 south compass line, drawn at 
 
 Fig. 3. — Exterior opening of the venti- 
 lator, Ruin no. v. rl § ht angles to the face of the 
 
 pueblo at its middle point, 
 passes nearly over the center of the manhole in the front middle 
 room, traverses the passageway to the kiva, runs along the east 
 side ot the niche, bisects the sipapn and the fire pit, and nearly 
 coincides with the western side of the horizontal tunnel of the 
 ventilator. 
 
 Few artifacts w r ere found in this ruin. In the kiva, on the floor 
 mar the sipapu, Avas a large well-formed and finished tcamahias l or 
 -tone scraper; and in the debris near the floor were a few turkey- 
 bone awls and unworked turkey bones; a rough flint arrowpoint; 
 .1 pottery polishing stone; a roughly flaked flint celt about 5 inches 
 
 kes, "Antiquities of the Mesa Verde National Park." Bulletin 41. Bureau 
 merican Ethnology (1909), p. 39. 
 
PREHISTORIC SMALL HOUSE RUIXS 
 
 13 
 
 some chips of flint. The few fragments of pottery found 
 )rrugated and smooth, thick decorated ware, not sufn- 
 undant and differentiated from that presently to be de- 
 >m other similar ruins to justify a separate consideration 
 s rooms of the pueblo revealed nothing noteworthy in the 
 tifacts. 
 
 Ruin Xo. VI 
 
 xt ruin group examined, no. vi, lies several miles west of 
 mesa, upon the great sage and pinon upland between the 
 :hes of Cross canyon and Dove creek on the east and 
 ivon on the west. This is known as Squaw mesa, and the 
 w Point ruin, lies toward its southern end. I cannot give 
 •cation, because there is as yet no accurate published map 
 trict. The ruin is most conveniently reached from the 
 ement called Dove creek, on the Dolores-Monticello road. 
 l store and post office kept by the Stokes Brothers, who 
 - ed to furnish grain and fodder and provision for those 
 this way, and who know the country hereabouts. The 
 iched from Dove creek by traveling southward over a 
 on road about 6 miles and then for some 9 miles, without 
 ck along the sagebrush sags and ridges, 
 ttlemen of this region know these ruins well on account 
 ient artificial reservoir belonging to them, which is still 
 rvice during certain seasons in this waterless land. The 
 
 1 an irregular sagebrush opening in the pinons covering 
 acres, in which a series of shallow washes from various 
 
 unite to form one of the small tributaries of Cross canyon, 
 here are of two distinct forms. First, those of the unit 
 Dying a long sage-clad ridge on the western side of the 
 and some small outlying spurs; and second, on the eastern 
 
 2 opening, on an elevated tongue of rock between two 
 ishes, is a great compact mass formed of the still half- 
 rails of many square and circular rooms, the whole some 
 
 3 feet across. The trim, unweathered stones, the many 
 walls; and the well-preserved timbers of this great com- 
 
 
 . 
 
14 AMERICAN ANTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 pact congeries of rooms, tier above tier, would seem to mark this 
 ruin as of a later period than the fallen, disintegrated, often largely 
 obliterated structures of the former group upon the ridge near by. 
 Between the two groups of ruins is a large slope of bare rock, across 
 the lower end of which in prehistoric times a broad, low, earthern 
 dam had been thrown, forming a reservoir which is capable when 
 full of holding a large quantity of water gathering from the long 
 slopes on three sides. At the time of our visit the reservoir was 
 partly filled with water, upon which we and our horses and mules 
 and a varying floating population of range horses subsisted during 
 our two weeks' time here, the loss of water during this period, in 
 early July, being largely from evaporation. 
 
 The long sagebrush ridge on which the ruins which especially 
 concern us lie, runs approximately north and south; the southern 
 end, facing across the McElmo valley toward Ute mountain, swings 
 a little to the southeast. The ruins lie across the axis of this ridge, 
 in irregular parallel rows, extending one behind the other along its 
 summit and down its eastern slope. There are a few single-kiva 
 ruins in the group, but most of them are composed of two or three 
 kivas with their corresponding pueblos placed end to end and facing 
 southward. Altogether there are at least twenty-four of these 
 single, duplex, and triplex forms well defined on this ridge and its 
 outlying spurs. The burial mounds are regularly placed south or 
 southeast of the rows of kivas, and in most instances there is a 
 separate burial mound for each of the houses, although these are 
 frequently not more than 50 or 60 feet apart. The sites of the 
 pueblos are all low mounds, the surface stones so broken and 
 weathered as hardly to show they have been shaped and dressed 
 and with no marks of walls visible. The pueblo sites, as well as 
 the usually well-defined kiva pits, are now all densely sage grown. 
 There are among the others numerous slight mounds with associated 
 depressions so ill defined that, though suggesting ruin sites, they 
 will require excavation for the determination of their nature. 
 
 The particular ruin of this group selected for excavation lies at 
 the -out hern end of the ridge as it falls off into a pinon-covered sag. 
 Here are three kivas in a row with separate low pueblo mounds 
 
prudden] PREHISTORIC SMALL HOUSE RUINS 1 5 
 
 north of two of them. The middle one of these kivas with its house 
 was selected. The mound, as we found it, including its slopes, was 
 about 58 feet long, about 18 feet wide, and rose some 4 feet above the 
 general level of the ground (figure 4). The nearest edge of the 
 burial mound was about 72 feet southeast of the pueblo. 
 
 The Pueblo. — This pueblo shows well a common feature of these 
 ruins, namely an irregularity of outline and a rough careless con- 
 struction of the masonry, in striking contrast to the uniform pro- 
 
 FiG. 4. — Ruin no. V. on .Squaw point, before excavation. Showing the elongated 
 heap or "mound" of disintegrated stones, with the slightly depressed kiva site in the 
 foreground between the shovel handle and the "mound." 
 
 portions, the well-laid masonry, and the painstaking finish and 
 plastering of the kiva and its accessories. Our estimate of the 
 amount of trimmed stone fallen from the pueblo walls, both within 
 and without the rooms and in the kiva, indicates that these could 
 not have been more than 6 or 7 feet in height, and hence that this 
 pueblo, like all the others of this type which I have examined in 
 detail, was a one-story structure. 
 
 On excavation it was found that the pueblo in this ruin is not, 
 as is commonly the case, symmetrically placed in relation to the 
 north and south axis of the kiva, for it forms a slight angle with 
 this and its larger part lies west of this line (see ground plan, figure 
 5). Its front wall stands about 8 feet north of the inner circle of 
 the kiva. 
 
X TTTTT^s izrrT"rnzro 
 
 § 
 
 j^aznxsD crn 
 
 t.ii i i nrrnt 
 
 pzoiEairiL_xj_j-fp=n=ESD: 
 
 iiiixisnxxiiuHzaxn 
 
 mrrrrr^ rrTH n 
 
 
 is/ 
 
 Scale 
 
 1 ii,. 5. — Ground plan of Ruin no. VI. on Squaw point, a, Manhole into tunnel to 
 kiva. /', Underground tunnel from front room of pueblo to kiva. c, Cubby hole or 
 niche in the north wall of kiva beneath banquette, d. Pilasters for the support of the 
 roof timbers, e. Recesses between the pilasters with banquettes about 33 inches above 
 the kiva floor level; the backs of these recesses, unlike those of fig. 1, are set with stones 
 1 >.. 1 1 it in the southern recess. /, deep southern recess and banquette, g. Horizontal 
 portion oi the ventilatoi Hue opening within at the floor level. Perpendicular shaft of 
 tin ventilator opening at the ground level, i, Sipapu in kiva floor, k. Fire pit in the 
 floor, m, Stone wall deflector between the fire-pit and the ventilator opening, n, 
 Stoni l" n. lies possibly for mealing purposes. 
 
prudden] PREHISTORIC SMALL HOUSE RUINS 1 7 
 
 It is approximately 39 feet long and from 14^ to 15 >£ feet wide 
 exteriorly, its north side being a little longer than the south and its 
 eastern end a little wider than the western. Its walls are from 
 9 to 12 inches thick and it contains six rooms of the relative size 
 shown in figure 5. The buried walls were standing in places 3 feet 
 high, in others not more than a foot above the level of the ground, 
 and their tops were covered from 6 to 18 inches with debris and 
 drift and fallen building stones. There were no evidences of ex- 
 terior doorways, but several interior doors from room to room were 
 found as indicated in the ground plan (figure 5 and in plate 1, d). 
 The sills of these doors were from 18 to 24 inches above the floor 
 levels; one was 18 inches, two 14 inches wide. The tops of all the 
 doorways had fallen with the walls. The floors of the room were 
 not plastered but formed of trodden earth, and those in the north 
 tier were at different levels, all lower than those of the south tier, 
 and some of them were a foot or more below the bottom of the walls. 
 
 The top of the manhole leading to the underground passage to 
 the kiva had partly caved in, so that its exact size and shape were 
 not determinable, but its situation in the middle front room (figure 
 5) was revealed by a number of flat stones irregularly distributed 
 about it. Digging down at the center of these stones we found a 
 bell-shaped circular chamber with level floor, whose walls were well 
 defined by the smoked plaster upon them. This chamber at the 
 bottom was about 3 feet 6 inches in the north and south diameter, 
 and about 4 feet in the east and west diameter, narrowing at first 
 very slightly from the bottom, then drawing in sharply to a diameter 
 of about 20 inches, 4 feet above the level of the floor. The whole 
 depth of the chamber from the floor to the level of the floor of the 
 room above in which the manhole opens is about 5 feet 3 inches. 
 
 A curved passageway about 18 inches wide led from this bell- 
 shaped chamber into the kiva, where it opened upon the north 
 banquette close to the pilaster (see figure 5 and plate 1, a and b). 
 The passage was caved in and completely filled with trimmed stones 
 from the front pueblo wall, several thin stone slabs, and debris. 
 Its outlines were readily made out, however, because several large, 
 thin, densely smoked stone slabs were still in place along its sides. 
 
jg AMERICAN iNTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 Lbout 2 feet high, and this was probably the approxi- 
 mate- height of the p; iy. The floor of the passage sloped 
 irregularly upward from the level of the banquette at which it 
 .] beneath the front wall of the pueblo (see plate 1, a 
 and b) and ended at a step oi about 12 inches down to the floor of 
 the chamber. 
 
 This larger «>r smaller bulge or chamber in the passageway from 
 the kiva to the pueblo is a regular feature in these ruins, giving 
 e for the individual making his way on hands and knees from 
 the kiva to the pueblo, as he must through the horizontal part of 
 the passage, to mil tend before raising himself up through the man- 
 hole. 
 
 The n>«>ms of the pueblo (plate 1, d) were largely filled to the 
 height of the yet standing walls with trimmed stones fallen from 
 tin- masonry, all closely packed with soil, sand drift, and rock frag- 
 ment-. In tlie\ corners of the middle and east front rooms near the 
 door -. e figure 5 and plate 1, d) were roughly constructed elevated 
 platforms of stone and adobe, about 4 feet long, from 12 to 18 
 inches wide and from 6 to 12 inches high. From the presence of 
 large thin stone slabs near these narrow platforms and of several 
 hand -rinding stones — "mullers," in these rooms, it is conjectured 
 that these platform- were remnants of corn-grinding places. 
 
 There were everywhere in the upper layers of the room debris, 
 small fragments of pottery, mostly corrugated and smooth with 
 black-banded decoration. But these have n6 obvious relationship 
 to the original occupants, since they were apparently deposited 
 there after the house had long been in ruins. Near the floor level, 
 
 - usual in these ruins, one found a mass of larger and smaller 
 ments of hard adobe which had been laid on and between the 
 surfaces of the fallen roof timbers and bore the marks of their sur- 
 s, while the parts which had evidently been pressed through 
 between the timbers were densely smoked. The tracings upon 
 the-e fragments of adobe showed that the roof timbers, here as in 
 the other ruin- of this type, for the most part had been split or 
 peeled before having been laid and plastered. In one of the rooms 
 the charred remnants of several roof timbers lay upon the floor, 
 
AMERICAN ANTHROPOLOGICAL ASSOCIATION 
 
 MEMOIRS, 5, PLATE I 
 
 - _-._„ 
 
 RUIN NO. VI. ON SQUAW POINT AFTER EXCAVATION 
 
 a. General view of the pueblo and kiva with underground passage partly caved in. b. Kiva, 
 showing pilasters, banquettes, deflector fire pit and sipapu. c. Sipapu in floor and niche in north 
 wall. d. End view of excavated pueblo showing doors between rooms. 
 
prudden] PREHISTORIC SMALL HOUSE RUINS 1 9 
 
 their outlines so well preserved, though little of the wood remained 
 that one could infer that, when in place, they lay north and south 
 from wall to wall. 
 
 At the floor levels in all the rooms there were a few fragments of 
 rough corrugated ware, a few of smooth ware finished with white 
 slip with banded and linear decorations in black. There were 
 found at or near the floor level, two stone axes, several polishing 
 stones, some partly chipped flint, a few turkey-bone awls and un- 
 worked turkey bones, a small piece of hematite with ground sur- 
 faces, part of a metate, a granite hammer and a granite discoidal 
 rubbing stone, nine spheroidal pounding stones from 2 to 3 inches 
 in diameter, and several mullers. Most of the latter were formed 
 of sandstone, but several were of conglomerate rock, whose firmly 
 cemented constituents, including many flint pebbles from one eighth 
 to a quarter of an inch in diameter, must have made them very 
 effective instruments. 
 
 In three of the rooms there was much scattered charcoal in small 
 fragments and small heaps of ashes at the floor level. In the corner 
 of one room about 18 inches below the top of the mound and about 
 as far above the floor level we found among the fallen stones in the 
 debris a large compact mass of ashes and charcoal fragments held 
 in place by a few -tones. This was evidently an improvised fire- 
 place in use long after the early floor level had been abandoned, 
 whether by the original owners, or, as is also possible, by others at 
 a much later time and when the house was a ruin with barely shel- 
 tering walls still standing, is not evident. There seemed to be no 
 difference in the fragments of pottery found at this level from those 
 exposed near the floor. This seems of some importance, because 
 it reminds one of the difference in significance which may attach to 
 artifacts found in the upper layers of the debris in such ruins and 
 those at or near the floor levels. 
 
 The Kiva. — The kiva was filled for about 3 feet below the sur- 
 face with closely packed, fine, reddish sand laid in layers of from 
 one to three inches thick with many dark to black layers between. 
 This material is clearly drift, and one may conjecture that these 
 dark streaks mark successive fires to which in the long processes of 
 
20 AMERICAN ANTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 filling up ol the kiva the district has been subjected. The rest of 
 the kiva to the bottom was so densely packed with sand, adobe, 
 and dressed building stones, fallen in from the pilasters and the 
 adjacenl pueblo, that the constant use of the pick was necessary. 
 These stones were most abundant in the northern segment where 
 the front walls of the pueblo had fallen, while several slabs of various 
 sizes lay about the middle. Save for numerous small fragments of 
 pottery, corrugated, painted, and plain gray, which were scattered 
 through the mass from top to bottom, there were few artifacts in 
 this kiva. At or near the floor level and on the benches were a few 
 turkey-bone awls, a stone axe, fragments of corrugated and smooth 
 pottery, a broken corrugated pot, a few mullers and polishing stones, 
 one spheroidal pounding stone, and numerous flint chips. 
 
 ( )n clearing, the kiva (figure 5 and plate I, a and b) is found to be 
 a six-pilastered, recessed, stoned structure with a deeper southern re- 
 1 ess, a ventilator, deflector, fiie pit, sipapu, and a doorway leading to 
 the pueblo. The following measurements should be recorded. The 
 inner circle is approximately 13 feet in diameter, flaring very slightly 
 from the floor. The banquettes are 33 inches above the floor, are 
 approximately 5 feet long on their inner faces and vary in depth 
 from [8 to 21 inches. The southern banquette, however, is about 
 4 feet deep. The wall of the lower circle is formed of well-trimmed 
 stones from 4 to 5 inches long and 2 to 3 inches thick, and upon this, 
 set a little back of flush, rest the pilasters, formed of larger stones 
 sel . a- is usual, without breaking joints. 
 
 The pilasters are from 21 to 24 inches wide across the face, and 
 th< ise which are best preserved stand from 32 to 35 inches above the 
 level of the banquettes. Their tops are thus from 65 to 68 inches 
 above the level of the floor. The banquettes are formed at the 
 front by Hat, smooth stones about 2 inches thick set in adobe. 
 These extend back from 6 to 10 inches, the remainder of the bench 
 being formed of adobe. The backs of the shallower recesses, about 
 30 inches high, an- irregularly set with roughly trimmed stones, 
 and, while in general nearly perpendicular, in places flare outward 
 at the top some 3 inches. The back of the deeper southern recess 
 was unstoned, being plastered directly upon the earth. Its sides 
 
prudden] PREHISTORIC SMALL HOUSE RUINS 21 
 
 were stoned as far back as the pilasters extended. The masonry 
 of this, as of the other kivas examined, is formed of single walls of 
 stones set against the earth, except the pilasters, which above the 
 bench form a cubical mass of larger stones. The plaster on the 
 lower walls of the kiva was largely in place, showing in its thicker 
 portions ten layers, but this as well as the plaster facing the pilasters, 
 readily peeled off on exposure. The surfaces of all the upper parts 
 of the kiva walls, especially above the banquettes, were densely 
 smoked. 
 
 At the western end of the northern banquette, close beside the 
 pilaster, is a square doorway to the curving passage above described, 
 leading to the pueblo (see plate I, a and b). About 4 inches below 
 this banquette is a cubby or niche in the lower wall 6 inches wide 
 and 7 inches high (plate I, c). About 5 inches east of the door to 
 the passage, and about 5 inches above the level of the banquette, a 
 roughly formed spheroidal pounding stone 2}4 inches in diameter is 
 set flush in the back wall of the recess. It is loose and readily taken 
 out of the cavity in the masonry, which it exactly fits. I have no 
 clue to its meaning or purpose. The floor of the kiva was about 
 7 feet below the floor level of the front row of rooms of the pueblo. 
 A north and south compass line through the center of the kiva 
 nearly bisects the ventilator opening, the deflector, the fire p ; t, the 
 sipapu, and the niche in the north wall. 
 
 In the well-plastered floor of the kiva, about 33 inches from the 
 north lower wall, is the sipapu (see plate 1, b, c). This is formed 
 at the top for 2 inches in depth, from the neck of a broken olla 3>£ 
 inches in internal diameter, with a flare of five eighths of an inch, 
 the top of which is set flush with the floor. Below the neck the 
 cavity bulges slightly, without special lining, and its hard bottom 
 is 9 inches deep. It was filled with sand and was uncovered. 
 
 About 39 inches south of the edge of the rim of the sipapu is the 
 northern side of the oblong fire pit (plate I, b). This is 28 inches 
 in the north and south diameter; 33 inches in the east and west 
 diameter; 18 inches deep at its center; the bottom is plastered and 
 the rim is formed in part by a few flat stones set in adobe, but mostly 
 by adobe continuous with that forming the kiva floor. It war- 
 nearly full of clear ashes. 
 
AMERICAN A XTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 About [3 inches south of the southern edge of the fire pit is a wall 
 of trimmed stone forming the deflector (figure 5 and plate 1,6), about 
 inches wide, 25^ inches high and 93^ inches thick. The stones 
 are well laid in adobe. About 30 inches south of the deflector, the 
 inner wall oi the kiva beneath the deep southern banquette, is re- 
 cessed about 3 inches and at the center of this recess is the opening 
 of the ventilator, about 26 inches wide, the bottom flush with the 
 floor of the kiva. The top of the ventilator shaft was caved in at 
 this part, but the opening, judging from the lines of stones still in 
 place, appeared to have been slightly higher than its width. About 
 5 feet south of the back ot the deep southern recess and one foot 
 below the surface of the ground, we found the partially stoned per- 
 pendicular shaft of the ventilator measuring about 8 by 12 inches. 
 The Burial Mound. — About 72 feet southeast of the kiva is the 
 nearest edge of the burial mound. It lies at the southern end of 
 
 the ridge on which this group 
 of ruins is situated and extends 
 a few feet down the slope. It 
 is about 40 feet across north and 
 south, and some 30 feet east and 
 west. The dark soil of which 
 it is composed varies from 2 to 
 4 feet in depth, thinning out 
 ' irregularlyat its borders. Small 
 
 pelvis in the burial mound of Ruin no. VI. 
 
 fragments of pottery are every- 
 where abundant in the soil. It did not appear to have been dis- 
 turbed. Beneath the dark soil one comes abruptly upon the com- 
 pact reddish-yellow natural earth of the region. The entire area 
 was turned over, the men working both ways along the sides of 
 exploratory trenches. 
 
 Nine burials were found, three of them young children; five near 
 the southern end of the mound area, four near its center. Their 
 depth varied from 6 inches to 2 feet below the surface. The placing 
 oi the bodies seemed to have been without reference to the points of 
 the ( ompass. Those in which the determination could be made were 
 placed in the flexed position, lour King on the right side, one on the 
 
pruddex] PREHISTORIC SMALL HOUSE RUINS 23 
 
 left. In some there had been considerable displacement of the 
 bones in the soil. In only three cases were the bones fairly intact 
 and all were very friable. 
 
 Pottery — ladles, bowls or ollas — either whole or in pieces, were 
 found either near or upon the bones, in six of the nine burials. In 
 one case the inverted top of a large smooth ware olla with a bone 
 awl lying in the sand and soil which it contained was placed over 
 the pelvis (see figure 6) ; in another (see figure 7) a bowl lay near the 
 head. Views of the skull of figure 6 are seen in figure 8, showing 
 the occipital flattening and other features found in the three of the 
 skulls from this mound sufficiently intact for the determination. 
 The teeth of this and other 
 intact skulls were worn down 
 and flat at their free edges. 
 In one case a fragment of 
 much decayed wood about 
 an inch thick and five inches 
 long was found with the pot- 
 tery near the remnants of the 
 bones of the children. No 
 other artifacts except the pot- 
 
 , , . . Fig. 7. — Skeleton and bowl near head in 
 
 tery and bone awl above men- , ■ , j c D • 
 
 burial mound of Ruin no. VI. 
 
 tioned and a square conglom- 
 erate grinding stone were found with these bodies. But scattered 
 through the soil of the mound were found a small dumbbell-shaped 
 object, perhaps a pendant, of polished hematite about three-fourths 
 i>\ an inch long and one-fourth of an inch in diameter; a bone flint 
 flaker; part of a deer-bone scraper; a turke\ 7 -bone awl; several pot- 
 tery polishing stones; four spheroidal pounding stones from 2 to 3 
 inches in diameter; and a small ball of unburnt clay. 
 
 The absence of pottery from some of the burials might well have 
 been due to their nearness to the surface of the ground, favoring its 
 destruction by the elements or its annexation by modern passers-by. 
 Judging from the great number and variety of the fragments of pot- 
 tery which the soil of this mound contained it would seem to have 
 served for a long period either as a burial place or as a rubbish heap. 
 
24 AMERICAN ANTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 None of the bodies were covered by stone slabs, as is the ca^e in 
 main- burial mounds in this region, and none of the bodies were 
 placed in shallow pits below the general level of the natural surface 
 of the ground, as was also frequently done when the accumulation 
 of waste forming the mound was meager; but in most of these 
 burials which were fairly intact, small flat stones from 5 to 7 inches 
 across were found over the bodies, usually near the head. 
 
 The approximately intact pieces of pottery from this mound are 
 shown in plate II. The most common fragments were either corru- 
 gated ware or smooth decorated food bowls. The smooth ware, 
 both the whole pieces and the fragments, was commonly thick, 
 sometimes gray, but usually finished with a white slip and bore a 
 great variety of rather simple decorative designs in black, largely 
 
 Fig. 8. — Skull from burial shown in fig. 6. 
 
 limited to the inside. A few of these food bowl designs are repro- 
 duced in figure 9. 1 Their thick edges frequently showed coarse 
 dots or short cross bands which were often grouped, leaving plain 
 spaces between. Occasionally a simple isolated geometrical figure 
 was made on the outside of the bowls just below the rim. There 
 were in this mound many, and in the kiva a few, small fragments of 
 red bowls, with well-polished surfaces, and decorations in black; 
 and this red ware was of finer texture and finish than most of the 
 white and gray. I found no whole pieces of it. Many of the frag- 
 ment- of the red ware as well as some of the better types of the white 
 had been drilled for purposes of repair. 
 
 1 Drawn from specimens by Elizabeth B. Prudden. 
 
( SlILt , 
 
W lei m 
 
 DD 
 
 m m 
 
 
 mm 
 
 Fig. 9. — Decorative designs from fragments of food bowls in the burial moundof 
 
 Ruin nc. VI. 
 
 25 
 
AMERICAN ANTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 Ruin No. VII 
 
 The lasl ruin examined — no vn oi the series — is one of a group 
 several miles west of the last, on the great upland called "Bug 
 mesa," situated between Squaw canyon and Monument canyon 
 farther west. 1 This Bug mesa ruin lies somewhat over six miles 
 southward of Hug spring, a locally well-known watering place and 
 cattle ranch; and is most conveniently reached from Dove creek by 
 a rough wagon road which passes the spring. 
 
 This group of ruins in main respects resembles the last described. 
 In an irregular opening in the pinons, several acres in area, is a 
 series of shallow washes converging to form a rocky stream bed, 
 which rapidly deepens to a gorge. Rising above the right bank of 
 the wash is a great irregular mass covering an area of approximately 
 400 by 200 feet and formed by a congeries of partly upright, partly 
 fallen walls, showing the outlines of main - rows of square rooms and 
 circular structures close to the southward of them. The whole 
 ruin mas- appears to be above the level of the rock surface on which 
 it stands. Across the stream bed, a few yards away, is a straggling 
 series of irregular fallen structures composed of large and small 
 buildings, courts, and passageways. A rough, partly demolished 
 wall crosses the stream bed between the two parts of the ruin. 
 The building stones of this ruin are well dressed; some of the walls 
 are standing 10 or 12 feet above their base, with indications of two 
 or more stories. Fragments of timbers are still well preserved; and 
 this, together with the absence ot marked weathering of the building 
 stone.- and the freedom of the ruin in general from sand drift and 
 soil accumulation, conveys the impression that it is not of great age 
 as compared with the adjacent ruins of the unit type which especially 
 concern u> here. 
 
 Westward less than 100 yards from this big ruin is a short sage- 
 
 1 < >n the reconnaissance map of this region published by the writer in 1903 {Ameri- 
 can AnthrO] s.), vol. 5, April-June, 1903) the canyon now known as Monu- 
 ment canyon 1- called Bur canyon, and the main Squaw canyon was called Pierson 
 canyon, these being the local designation at that time for these imposing but officially 
 nameless gorges. It is now clear that some parts oi the course of these canyons is not 
 
 accurately Bhown on this map. since the details were only determined as well as 
 mi^ht be from a general horseback and compass survey reinforced by the data and views, 
 
 [ways harmonious, ol local cattlemen who at that time ranged the region. 
 
»' " „>■• ' l • 
 
 
 
 
 
 Of 
 
 % 
 
 8$q»'"" 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 i is 
 
 
 \%i 
 
 ' c 2 
 
 3 ~ 
 
 s "3 
 
 o g 
 
 j 5 
 
 
pruddex] PREHISTORIC SMALL HOUSE Rl 2~ 
 
 grown ridge on which are closely clustered at least twelve ruins ol 
 the unit type, mostly with two or three kiva pits and long incon- 
 spicuous pueblo mounds. A little farther westward, across a 
 shallow wash, is another longer and wider sagebrush ridge over 
 which at least twenty similar ruins are irregularly scattered, some 
 with a single kiva pit, others with two and three. The burial 
 mounds lying close to the southward of apparent ly nearly all of 
 these small ruins had been thoroughly ravaged. The wash between 
 these two ruin-covered ridges widens out northward into a bare slope 
 of rock across the lower end of which, before the white men came, an 
 earth dam had been thrown affording a reservoir of considerable 
 capacity (plate in, a). 
 
 I was not able to form an opinion as to whether the artificial 
 reservoirs at the Squaw point and at the Bug point ruins probably 
 belonged to the one or the other types of ruins near by or may have 
 been used by the people of both at their apparently different periods 
 of occupancy. The cattle men in recent times have made a short 
 dam below the original one at the Bug ruins to supplement the water 
 supply on this range. 
 
 The small ruin selected for special examination lies at the south- 
 ern end of the ridge nearest the big ruin. There were in this two 
 well-defined kiva pits about 27 feet across and about 2 feet below the 
 general level at their centers. No walls were visible in or about 
 them. The long pueblo mound close to the north of the kiva pits, 
 showing only a few broken and weathered stones at its flat sum- 
 mit, was about 18 inches above the general level. The burial 
 mound was a short distance southeast of the kiva sites. After 
 establishing the exterior dimensions of the pueblo we selected its 
 eastern end and the corresponding kiva for such study as might 
 be permitted by time and the meager water supply rapidly dwindling 
 under the July sun. 
 
 The Pueblo.— The exterior of the entire pueblo was 92 feet long 
 and about io>^ feet wide; its walls were from 9 to 11 inches thick, 
 and it consisted of a single row of rectangular rooms, varying, some- 
 what in size (see figure 10 and plate III, b). The walls were standing 
 from 18 to 30 inches above the floor levels, which were about the 
 
28 AMERICAN AXTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 same as that of the surrounding ground surface. No doorways 
 were found between the rooms, the walls in most places having 
 fallen to below the level at which they are commonly placed in these 
 ruins. 
 
 While most of the pueblos of this type which we have excavated 
 have contained few artifacts, nearly all of the rooms in this house 
 which we had time to dig out revealed many conditions and articles 
 of interest. In the first place the whole eastern end of the pueblo 
 had been burned, the roof had fallen in upon the possessions of its 
 thrifty occupants, and such objects in the rooms as the fire and time 
 had spared lay undisturbed beneath the wreckage of later fallen 
 walls and the accumulated soil and sand drift. The fallen timbers 
 within the rooms were represented only by some small charred frag- 
 ments, but the masses of adobe mud, bearing the impress of the 
 split timbers upon and between which it had been plastered, served 
 to locate the position of the fallen roof in the debris, and showed 
 that while most of the artifacts found here were within the rooms 
 at the time of the catastrophe, some of the broken pots and several 
 flat rocks had probably come down with the roof upon the upper 
 layers of whose wreckage they lay. 
 
 Room 1 (see figure 10) presented little of interest; a few frag- 
 ments of food bowls, a bone awl, a muller, and a stone hammer were 
 found in the mass of fallen building stones and drift with which the 
 room was filled. 
 
 Room 11 was noteworthy for the large number of pieces of pottery 
 of various forms, broken and intact, which lay upon or near the 
 plastered floor; and for the large amount of charred corn which was 
 heaped upon the floor, or was contained in large ollas, or lay in 
 scattered masses with the fragments of the broken ollas which 
 evidently had contained it. In the southwest corner of this room 
 was a shallow bin formed of stone and adobe (plate in, c), filled with 
 shelled charred corn, which was also heaped up against the west 
 wall. Altogether there was not less than a bushel of charred corn 
 in this room, scattered and in heaps and in jars. Most of this corn 
 was shelled, but some was still on the ear, all charred, and in a few 
 instances the inner layers of the husk were still in place (figure 11). 
 
>Cj O oj JO 
 
 « •« 
 
 c S 
 
 — c 
 
 ~ 5 m 
 
 J3 <U w O 
 
 c w > "■- 
 
 ai rt 5 ■- 
 
 S o = is 
 
 . is a .> 
 
 a a -a •* 
 
 ■S >- OJ 
 
 _; P O ,G 
 
 53 W 
 
 {* -M ^ 
 
 5 c « .e 
 
 t: e 
 
 c •*: .2 
 
 fe o 
 
 29 
 
30 
 
 AMERICAN ANTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 
 
 Fig. 11. — Corn on the ear, charred. On 
 the right, one ear is in the husk. 
 
 All of the ears of corn found in this ruin, as well as the many free 
 cobs scattered about, were short, averaging not more than 3 inches 
 in length. One of the smaller pots contained a handful of charred 
 grass seed. 
 
 Thirteen pieces of pottery were found in this. room. Several of 
 the smaller pieces and two of the larger ones were nearly intact, but 
 
 most of them had been over- 
 turned, or smashed where they 
 lay, all the pieces being found 
 together. One large broken cor- 
 rugated olla lay above the stra- 
 tum of roof mud and presumably 
 had fallen with the roof. Upon 
 or near the floor were several 
 large corrugated pots, but the 
 more abundant pieces were 
 smooth ware, gray or with white 
 slip and linear or band decora- 
 tions in black. Represented in the smooth ware were ollas, food 
 bowls, dippers, mugs, and canteen forms (see plate iv and figures 
 12, 13). Several lids to the ollas and pots were found beside them. 
 Some of these were made of thin slabs of sandstone roughly 
 rounded; but the more abundant were crudely fashioned from 
 a< lobe mud which apparently had been shaped and pressed while soft 
 into place when the jars were full, as they all bore upon their under 
 surfaces the impress of corn cobs with which the shelled corn in the 
 jars appears to have been covered (see figure 14). J Most of the 
 pottery showed no trace of the fire which destroyed the pueblo, 
 but a mug, a food bowl, and parts of a large corrugated pot were 
 baked to a dull red color, with the partial fading of the painted 
 decoration of such as bore it. 
 
 1 I am informed by Clayton Wetherill that the stopping with adobe of the mouths 
 of jars filled with corn was common in several of the cliff houses which he and his 
 brothers have excavated in Colorado, Utah, and Arizona. But in his experience the 
 corn in the jars was protected by a layer of cedar bark which left its impress upon 
 the under surface of th 'mil -1 v;r;rs which in our specimens show marks of the 
 obs. 
 
AMERICAN ANTHROPOLOGICAL ASSOCIATION 
 
 MEMOIRS, 5, PLATE IV 
 
 > 
 
 
 POTTERY FROM ROOM II OF RUIN NO. VII ON BUG MESA 
 
PRUDDEX 
 
 PREHISTORIC SMALL HOUSE RUINS 
 
 3' 
 
 In Room in, near the west wall, about S inches below the surfa< e 
 of the mound and above the stratum of roof mud, were found in an 
 
 Fig. 12. — Small olla, canteen and mugs from Room n of Ruin no. vn on Bug mesa. 
 
 area of less than a square yard, 45 flint pebbles from 2 to 3 inches 
 in diameter, nearly all of which had been more or less flaked (see 
 
 
 Fig. 13. — Corrugated jars from Room n of Ruin no. vn on Bug mesa. 
 
 figure 15). There were here numerous flint flakes. All these 
 flints were in such a position in the debris above the roof mud as to 
 
32 
 
 AMERICAN AXTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 Fig. 14. — Lids of ollas formed of adobe 
 pressed down upon corncobs apparently 
 laid on the top of the shelled corn within. 
 
 suggest that they had been upon the roof when it burned and fell in. 
 On clearing away the roof mud and the charcoal mixed with it, we 
 found upon the floor of this room a large quantity of charred corn, 
 mostly shelled, but in part still on the cob; here also were a large 
 shattered corrugated pot and a thick food bowl. The floor of this 
 
 room was roughly set with flat 
 stones (see plate 111, d), which 
 near the center, toward which 
 the floor slightly sloped, form- 
 ed a rectangular eighteen-inch 
 manhole to the tunnel leading 
 to the kiva. 
 
 Across the eastern end of 
 Room iv was a rough stone 
 and adobe bench about 6 inches 
 high and 18 inches wide, while from its southern wall projected an 
 irregular, shallow, partly enclosed bin formed of flat stones and 
 adobe. This was filled with charred corn, mostly shelled, which 
 was also piled elsewhere upon the floor beneath the masses of roof 
 mud and mingled with pieces of wood charcoal. Altogether there 
 was more than a bushel of corn 
 in this room also. Part of a 
 thick food bowl lay upon the 
 floor here. On the low bench 
 at the eastern end was found 
 the charred spatulate tip appa- 
 rently of a wooden utensil, and 
 a plaited ring about 4 inches in 
 diameter (see figure 16). Here 
 also were found a matted mass 
 of coarse black hair, evidently 
 human; and a small bundle of partly charred thin narrow strips of 
 the bark, apparently of a shrub, its inner layer striated, its outer 
 finely punctate. 
 
 There were in these rooms, in addition to the articles above 
 mentioned, eight sandstone mullers, much worn; two conglomerate 
 
 Fig. 15.— Flaked flint pebbles and 
 partly shaped flint utensil from the Bug 
 mesa ruin. 
 
PREHISTORIC SMALL HOUSE RUINS 
 
 33 
 
 mullers worn smooth; one smooth flat sandal pattern stone; tour 
 small hand grinding stones; three stone axes; three spheroidal 
 pounding stones; two stone pot covers; part of a sandstone met ate 
 and a stone mortar about 6 inches in diameter. 
 
 Room v, at the eastern end of the pueblo, was narrower than the 
 others (see figure 10), and its walls were standing about 18 inches. 
 Its floor, about one foot above the floor levels of the other rooms, 
 was formed at the western side, of closely set flat stones, the re- 
 mainder being made of a thick 
 layer of adobe. The floor 
 stones, as well as the layer of 
 earth beneath them, about 5 
 inches thick, had evidently 
 been exposed to heat and were 
 very hard and red. Below this 
 layer of baked earth was about 
 4 inches of black organic ma- 
 terial resting upon a mass of 
 white clay such as we found 
 near by in the excavation of 
 the kiva. The presence of 
 these materials beneath the floor accounts for its elevation above 
 the others in the pueblo, and with the structure of the walls give 
 the impression that th<s room was rather an annex than an integral 
 part of the house. The parts of the wall of this room still standing 
 « were thickly plastered within with adobe. This was in places 
 deeply smoked, while other parts were red from intense heating. 
 
 The side of this room adjacent to the pueblo is filled with a solid 
 heap about 19 inches high composed of large and small irregular- 
 shaped gray masses of minutely porous tufa-like material, in part — 
 especially where the conical heap lies against the west wall — of a 
 slag-like substance. The latter is partly porous, forming contorted 
 masses with shiny greenish surfaces, or is in the form of reddish 
 friable spongy masses or of excrescences upon the surfaces of the 
 other material. Embedded in and fused to this slag-like stuff are 
 the curled and otherwise contorted parts of thick-walled pots and 
 
 Fig. 
 
 16. — Plaited ring, partly charred, 
 from the Bug mesa ruin. 
 
34 AMERICAN ANTHROPOLOGICAL ASSOCIATION [memoirs, 
 
 large and small fragments of baked sandstone rock. This heap of 
 fused material forms a coherent mass covering more than half of 
 the floor of the room, and it is the parts of the floor and walls nearest 
 to this that are the most hardened and reddened by heat. A con- 
 siderable amount of charred corn, shelled and on the cob, lay on 
 the south side of this slag heap and on the floor beside it. Here 
 also were a roasted and reddened food bowl and a mug without its 
 handle. 
 
 From the condition of the floor, walls, and contents of this room 
 it is evident that an intense heat has been developed here, cither 
 before or at the time of the burning out of the pueblo. I have sub- 
 mitted specimens ol the slag, pottery, and roasted rock to Pro- 
 fessor W. Campbell, professor of metallurgy in Columbia University, 
 and wish here to express my thanks to him for his courtesy in making 
 the examination and for permitting me to report his opinion that 
 the degree of temperature required to produce the appearances 
 presented by the specimens submitted to him might readily be 
 secured with the wood and other inflammable material which this 
 early housebuilder possessed; provided that the conditions of firing 
 were such, either by chance or design, as to concentrate, furnace- 
 like, the available heat. 
 
 The finding of slag-like material in the burned-out rooms of 
 prehistoric pueblos is, I am informed, not unique; and while it is 
 not very profitable to speculate upon the various possibilities in 
 this instance, it is not difficult to conceive of such conditions, and 
 the presence here of such an amount of piiion or other timber — 
 perhaps from a burning and fallen roof — and of such a store of corn, 
 also under certain conditions a lusty heat producer, as might readily 
 convert this small room, with its non-conducting floor and its thick 
 stone walls, into a veritable furnace. 
 
 The Kiva. — The kiva was found about 9 feet south of the front 
 wall of the pueblo, its lower circle being about 21 feet in diameter. 
 This is considerably larger than any kiva in my series of excavations. 
 Its northern segment only was dug out, owing to the lack of time 
 and the waning water supply. Its plastered floor was about 11 
 feet below the general floor levels of the pueblo. The lower circle 
 
prudden] 
 
 PREHISTORIC SMALL HOUSE RUINS 
 
 35 
 
 was walled with trimmed stones roughly set. A shallow bench, 
 from 6 to 9 inches deep, formed of flat stones was aboul 42 inches 
 above the smooth adobe-plastered floor; above the bench rose aboul 
 28 inches of smoked earth-plastered wall. At the poinl in the cir- 
 cumference in the outer circle nearest to the pueblo and to the man- 
 hole in the floor of Room no. in, was an arched doorway in the smoked 
 
 earth forming the back of the shallow recess 18 inches wide and aboul 
 
 19 inches high, the sill being about 5 inches above the banquette 
 level. This leads to a passage which slightly widens after it leaves 
 the kiva, the floor and the arch gradually rising. The doorway and 
 the tunnel are formed by adobe plastered against the soil and the 
 heavily smoked surfaces make it easily practicable to follow the 
 passage through the debris and stones with which it is filled. 
 
 Until the floor was reached 
 no noteworthy artifacts were 
 found in the kiva. But on that 
 portion of the floor exposed 
 there lay scattered three 
 pointed and two flattened 
 pieces of deer horn such as 
 might have been used in chip- 
 ping flint, and a cylindrical seg- 
 ment of deer horn about an 
 inch in diameter and 3 inches 
 long, rounded at the ends (see 
 figure 17). There were also 
 one turkey-bone awl; a well-formed spearhead about 3 inches long 
 and one inch wide; and a lump of unbaked clay about 5 inches in 
 diameter, with a groove around it apparently to facilitate its trans- 
 portation by suspension. 1 
 
 1 It is interesting to note that while the pottery makers of this house had apparently 
 transported the gray clay which they used in their pottery for some distance, there was 
 at hand, unused, close beside their kiva, a bed of excellent kaolin, the adaptability of 
 which to pottery making the writer was moved to test by fashioning a small bowl from 
 a specimen of this kaolin brought to New York. This achievement is not recorded in a 
 spirit of boastfulness, since the lack of recognition of the obvious is a trait not monopo- 
 lized by prehistoric people. 
 
 Fig. 17. — Flint flaking utensils from the 
 kiva floor of the Bug mesa ruin. 
 
36 AMERICAN ANTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 There were no pilasters in the 16-foot segment of the wall which 
 we exposed, but a small jog in the wall at one side might indicate 
 the site of a fallen pillar or some other form of roof support. 
 
 The kiva was not sufficiently cleared to secure any data relating 
 to a sipapu, fire pit, deflector or ventilator. A line of wall about 
 4 feet long and one foot high projected from the front of the pueblo, 
 between Rooms I and 11, but there was no evidence of any enclosure. 
 In the approximately level space in front of Room iv two stone slabs 
 were set upright in the ground in such relations as to suggest a corn- 
 grinding place in the court in front of the pueblo. 
 
 The Burial Mound, lying about 30 yards southeast of the kiva, 
 had been completely ravaged and we failed to find the trace of an 
 undisturbed burial here. 
 
 We may now summarize some of the details of these small 
 habitations and their associated artifacts, on the basis of our special 
 study, recorded in this and in the previous paper. 
 
 The pueblos, as a rule, were rather roughly constructed of 
 trimmed stones laid in mortar, the bottom tier resting directly upon 
 the ground. Some were plastered within, others apparently not; 
 I could find no evidence of external plaster, but owing to the severe 
 weathering conditions this might readily have wholly disappeared. 
 
 Our estimates of the amount of building stone in these ruins 
 derived from building up segments of wall indicate that the pueblos 
 were one-storied and not more than 7 or 8 feet high. There were 
 commonly from three to six or eight rooms in the pueblos, de- 
 pending upon whether there was one or two rows. They varied 
 considerably in size, averaging about 7 feet by 9 and, as will be 
 seen by the ground plans, were but crudely squared. The floor 
 levels were commonly about that of the general ground level, but 
 some were from a few inches to a foot below. Some of the floors 
 were plastered with adobe, a few were laid with flat stones, but more 
 often they seem to have been formed of the tramped earth and the 
 surfaces had apparently been at different levels. 
 
 We found no marks or traces of ladders or notched sticks giving 
 access to the roofs. There were no fireplaces in the rooms, but 
 
— 
 
 prudden] PREHISTORIC SMALL HOUSE RUINS 37 
 
 occasionally small heaps of ashes lay along or near the side walls. 
 In Ruin no. vn there were roughly made low narrow stone and 
 adobe platforms in the corners of two rooms which may have been 
 permanent mealing places. Small shallow stone and adobe bins 
 had been made at the sides of two rooms in the same ruin for the 
 storage of corn which we found filling them and spreading out upon 
 the floor. 
 
 The circular or square manholes leading to the kiva in the floor 
 of the center rooms — of the front row when two rows were present — 
 were bordered with flat stones at the top of a shallow lining wall. 
 This protected the upper part of the perpendicular portion of the 
 passageway to the kiva. Below the stone-protected opening the 
 passage bulged in various degrees to form a turning space for the 
 body of the kiva habituate as he came out through the horizontal 
 part of the passageway from the kiva. This passage was always so. 
 low and narrow as to require a crawling posture and such of the 
 jambs and sills of the kiva doorways as were preserved were smooth 
 and greasy. The passage was sometimes tunneled through the 
 earth with an arched top, its sides being sometimes plastered with 
 adobe, often left unplastered and in all cases more or less densely 
 smoked. But the roof of the tunnel was sometimes supported with 
 wood or stone and in one instance large stone slabs were set along 
 the sides. 
 
 The roofs of the pueblos were apparently supported by cross- 
 beams, for I have repeatedly found upon the floors, lines of small 
 remnants of much-decayed or of charred wood, judged by the growth 
 rings to have been parts of beams of considerable size, and lying 
 crosswise of the rooms. 
 
 With these wood remnants were always found large and small 
 masses of usually elongated smoked adobe, bearing on their sides 
 the impress of split or unsplit peeled timbers, apparently both pinon 
 and cedar. Though carefully sought for, I have never found on 
 these lumps of smoked adobe marks of small sticks or twigs or 
 withes or cedar bark, with which kiva roofs of the Mesa Verde and 
 elsewhere were chinked before the laying on of the mud. I conclude 
 therefore that in the construction of the roofs of neither the pueblos 
 
38 AMERICAN ANTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 nor the kivas were these used, but that the mud was laid in a thick 
 layer directly on to the supporting and covering roof timbers. 1 
 
 Many of our pueblos apparently had been very small at first, 
 consisting of only one short row of rooms, or of a single or a pair of 
 rooms, and were enlarged by the apposition of a new line or of single 
 rooms at the ends. These later additions were marked by the ab- 
 sence of any tying together of the lines of the masonry by stones 
 from one section to another; often by a difference in the character 
 of the stones, the trim of the laying of the wall; and sometimes by 
 the sagging away of the untied addition from the original structure. 
 
 When it became necessary to have houses considerably larger 
 than the common form with one or two short rows of rooms, it is 
 usual to find added at one end a second unit with its house and kiva; 
 the burial mound serving for both. Thus in Ruin no. vn on Bug 
 point, Utah, the entire pueblo was 92 feet long, which is about 
 double the usual length, forming one continuous building, with the 
 two kivas in front of the middle of each half, and here there was a 
 single burial mound. 
 
 This is a frequent form in the groups of old ruins on the Bug and 
 Squaw mesas. Less frequently one finds, as a variant of the usual 
 form, short wings, consisting of one or two rooms extending south- 
 ward beside the kiva. Such winged houses are not uncommon and 
 I think most frequent in the close communities. But my excava- 
 tions have shown that the superficial appearance of short one-room 
 wings may be created by the falling of the front walls of one or two 
 of the middle front rooms into the kiva, as usually happens in the 
 process of disintegration when the latter is close to the pueblo. 
 Thus the undisturbed ruin gives the appearance of a recessed front 
 partly enclosing the contiguous side of the shallow kiva pit into 
 
 1 Kidder, American Journal of Archaeology, Second Series; Journal of Archaeological 
 Instituteof America, vol. XIV, no. 3 (1910), p. 342 describes the roof of one of the rooms 
 of a small exposed cliff-dwelling in Utah which he found well preserved, as supported 
 in the' middle by a cedar beam eight inches in diameter, running lengthwise of the room, 
 on which rested at right angles, four two-inch beams, while upon these and parallel to 
 the first were laid slabs of split cedar, directly upon which lay some three inches of 
 adobe. This construction would accord very well with the evidence derived from the 
 uniform appearances in all the roof vestiges of the pueblos of our study; and in the kiva 
 roofs also the adobe was plastered directly upon the timbers. 
 
prudden] PREHISTORIC SMALL HOUSE RUINS 39 
 
 which the front walls have disappeared and been long buried under 
 the sand drift. 
 
 In the kivas, in marked contrast to the crude character of the 
 pueblos, no pains were commonly spared in construction and finish. 
 The stones of the masonry were usually carefully dressed and laid 
 in adobe mortar in a single tier against the earth before plastering. 
 The size, proportions, orientation, and relation to the pueblo were 
 singularly uniform. The inner edge of the upper kiva circle was in 
 one instance close beneath the front wall of the pueblo; in one it was 
 9 feet away requiring a longer connecting underground passa 
 the others showed distances between these extremes. 
 
 Most of the kivas were about 13 feet in diameter, and the height 
 of the pilasters, that is the outer portions of the roofs, about 6 feet 
 above the floor. It is interesting to note in this connection that 
 the average diameter of sixteen kivas described by Fewkes ' in the 
 Cliff Palace of the Mesa Verde is about 12 feet and the height of the 
 pilasters about 6 feet. The Bug ruin kiva was larger and deeper 
 and I regret, especially on account of the apparently great age of 
 that whole group, not having had time to determine more fully its 
 other characters. 
 
 It is clear from the circular kivas which have been found in 
 other larger types of ruins in this district with the roofs partly or 
 quite intact, that the presence of stone pilasters in the walls of these 
 underground chambers, indicates that the roofs were of similar con- 
 struction, that is, that they were laid up of timbers from the top of 
 one pilaster to another around the circle, drawing these in with each 
 superimposed row; the central part being formed of cross timbers 
 in which was a smoke hole or possibly an entrance about the middle, 
 over the fire pit. This presumption is sustained in all our kivas by 
 the position of the tops of the intact pilasters sufficiently below the 
 general level of the ground to allow for a court or plaza over the 
 vaulted roof. 
 
 For it seems certain that in our kivas as in some of the others of 
 the larger ruins, the roof was covered in with earth, forming a plaza 
 
 1 Fewkes, "Antiquities of the Mesa Verde National Park Cliff Palace," Bulletin 31, 
 Bureau of American Ethnology, (1911). 
 
40 AMERICAN ANTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 or court in front of the pueblo. The presence of several large flat 
 stones not apparently concerned in the masonry, in the deep layer 
 of debris filling the kivas, usually near the center, would suggest 
 that an opening existed in the roof and that it was bordered by 
 these stone slabs, just as it is in many of the underground ceremonial 
 chambers of the modern Pueblo Indians to-day. But in no in- 
 stance, though carefully sought, has any mark been found either on 
 the floor or on the deep southern recess, indicating a place where a 
 ladder of any form might have rested. 
 
 The assumption of a small flat outdoor court or plaza above the 
 roof of the kiva seems to be strengthened by the presence in nearly 
 all of the kivas excavated, beneath the sand drift and above the 
 fragments of roof mud, of a large amount of earth mixed with frag- 
 ments of pottery, broken stone implements, flint chips, turkey bones, 
 bone awls, etc. This accumulation of household debris mixed with 
 earth often forms a noteworthy part of the material filling the kiva 
 below the sand drift. And the assumption that a mass of earth 
 was formerly packed around the vault of the roof making a level 
 plaza, and that this slipped into the kiva when and as the roof fell, 
 together with the especially vulnerable construction of the pilasters 
 to which attention was called in the previous paper, 1 seems per- 
 fectly to account for the great uniformity in the appearance of the 
 kiva sites in the undisturbed ruins. Dr. Fewkes' illuminating 
 reconstructions at the Cliff Palace and at the Spruce Tree House 
 on the Mesa Verde, may be cited in support of this assumption in 
 our habitations of a flat plaza on top of the kiva at the sunny side 
 of the pueblo, and in this case near the middens which form the 
 burial mound. 
 
 The mode and apparent sequence of procedure in the construc- 
 tion of these dwellings seems clear. Placed as they regularly are 
 on unincumbered earth sites, the impulse to uniformity in perfor- 
 mance, method and result, so notable in folk in this stage of culture, 
 was here unhindered as it could not be in the larger houses built 
 on rock sites at the heads of gulches in recesses in the cliffs or in the 
 open. One may conjecture that the first operation was the digging 
 
 1 American Anthropologist, vol. 16, p. 49. 
 
prudden] PREHISTORIC SMALL HOUSE RUINS 41 
 
 of the hole for the kiva and the ditches at the bottom of which wen 
 the horizontal ventilator shaft, and in some case the passage to th< 
 pueblo. Theroofsof these shafts, and tunnelswere, insomeinstana - 
 at least, held up by cross pieces of wood, the remnants oi which we 
 found in place. In Ruin no. V (see figure 2) and in Ruin no. vn, how- 
 ever, the arched passage from the kiva to the pueblo seems to have 
 been made by tunneling through the undisturbed earth. It is inter- 
 esting to note as an advantage to the builders of these underground 
 chambers and passages, that the deeper parts of the adobe soil in 
 these regions is so firmly packed that it was not even necessary to 
 plaster with fresh mud the tunnels to the pueblo, though this was 
 sometimes done. 
 
 In our Kiva no. ill, which was crude in construction, the earth 
 sides of the lower circle before plastering were reinforced by driving 
 into them at intervals of a few inches small bits of rock which gave 
 a good holding surface. 
 
 The remarkable care with which the walls of the kivas were 
 usually laid up of a single tier of stones set against the earth, and 
 the heavier masonry of the pilasters was constructed, in both cases 
 without purposed breaking of joints, is shown by the fact that in 
 spite of the vicissitudes of disintegration and the work of the years, 
 the walls of the kivas and most of the pilasters, except the fronts of 
 their tops, had maintained their place and plumb; while in many the 
 thick mass of superimposed layers of plastering was when we found 
 it quite intact. 
 
 The burial mounds vary considerably in area but are seldom 
 more than 40 or 50 feet across; and so far as my observations go are 
 shallow and do not appear to have served for long periods either as 
 refuse heaps or burial mounds. For I have never found great 
 accumulations of debris, nor such differentiated layers of deposit 
 as are sometimes to be found in other types of larger ruins farther 
 south. 
 
 If, following the analogy of the still preserved roofs and tops of 
 the prehistoric stone houses in some of the more or less protected 
 cliff-cave dwellings of this region; and mindful of the fashion still 
 
42 AMERICAN AXTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 prevailing among the Pueblo Indians of today in their houses in the 
 open ; one assumes that these little box-like structures were flat at 
 the top with an entrance from the roof and with the edges of the 
 roof guarded by a tier or two of flat stones above the roof level for a 
 parapet; it is evident that these habitations were very inconspic- 
 uous feature's in the landscape, gray and doorless and meagerly if 
 at all windowed as their exterior was; against the dull backgrounds 
 of the earth, the rocks and the foliage of this arid country. 
 
 In plate v, a is reproduced the photograph 1 of a model of such a 
 pueblo with its plaza, which the writer has constructed from such 
 data as these studies have afforded, and under the above assump- 
 tion of its makers' kinship to the old cliff dweller and to the modern 
 Pueblo Indian. A vertical section through another model is also 
 shown in plate v, b. 
 
 Turning to the articles of food and minor artifacts which our 
 studies have disclosed, we find that these are mostly such as are of 
 durable inorganic character, objects fashioned of clay or stone. 
 Of foods we have found only corn and grass seed, so charred as to 
 be invulnerable to the action of time and the weather. 2 Other 
 organic objects were: fragments of charred timbers from the roofs; 
 rotten and disintegrating remnants of roof timbers and of the wooden 
 supports of underground tunnels and flues; the charred remains of 
 a small flat-tipped wooden implement; a bundle of partly charred 
 shreds of bark; a mass of black human hair; a small plaited ring. 
 
 That the reward of search for perishable objects should be so 
 meager is not surprising when one considers how complete has been 
 the exposure to the elements through a long period. The sites are 
 always open, the amount of protective covering of stone, debris, 
 soil, and sand drift is at best slight, and the climatic variations severe 
 upon the great plateau. On the other hand the conditions in many 
 of the cliff houses in these regions, whence a large part of our knowl- 
 
 1 The photographs of these models, as of the minor artifacts illustrating this study, 
 were made for me by my friend Dr. William P. Northrup. 
 
 -' We were not able to identify any of the erstwhile farming fields in the vicinity of 
 our ruins; nor did we make very particular search since most of the open country there- 
 abouts is constantly remoulded by sand drift and mantled with sagebrush and grease 
 wood. 
 
AMERICAN ANTHROPOLOGICAL ASSOCIATION 
 
 MEMOIRS, S, PLATE V 
 
 PHOTOGRAPHS OF MODELS SHOWING A RESTORATION OF THE "UNIT TYPE" HABITATION, 
 A. IN PERSPECTIVE. B. IN NORTH AND SOUTH SECTION 
 a In the court or plaza over the roof of the kiva south of the house, are seen th. 
 
 ed>ed openings of the ventilator and of the smoke hole (or possible entrai /ei the fire pit. 
 
 Stone slabs forming a probable mealing place are set near the house where in the Bug mesa run, 
 they were found. Notched timbers are placed as ladders to the roof and from here into the in- 
 terior of the house. -li- 
 fe This section shows many of the common structural features oi these dwellings, including 
 in the kiva. the ventilator, deep southern recess, deflector, fire pit. pilasters and banquette, vaulted 
 timber roof resting upon the pilasters, the underground passage to the house, with one,,, the 
 small doorways of the latter between the rooms. 
 
prudden] PREHISTORIC SMALL HOUSE RUINS 43 
 
 edge of the food, garments, utensils, and ceremonials of these early 
 folk has been derived, is wholly different. In the cliff houses, en- 
 tirely protected from the weather as objects of all kinds may lie, 
 the most delicate objects, fabrics, feathers, and the like, are often 
 found in as perfect form as on the day of their abandonment. E\ en 
 in the great open community houses, such as those on the Chaco 
 and the related ruin at Aztec on the Animas, where the amount of 
 covering debris is large and many rooms are still intact beneath, 
 the preservation of delicate objects is remarkable. Thus it is that 
 our meager contribution to the knowledge of many of the capac- 
 ities and ways of the masters of the simple prehistoric ranch houses 
 must be derived from an important but relatively limited group 
 of the more durable minor artifacts. 
 
 The stone and bone implements which were found were of 
 ordinary workmanship, some rude, some excellent. There were 
 axes, pounding grinding and polishing stones; mullers, metates, 
 and mortars; arrow and spear heads; stone scrapers {tchampais) 
 and sandal lasts; small pendants; pieces of hematite, worked and 
 unworked; bone awls and bone scrapers; deer-horn flint flakers of 
 various forms and sizes. There were many unworked turkey and 
 small mammal bones in nearly all the ruins. No specimens of 
 marine shells were found ; but several small fossil shells were scat- 
 tered in the debris of some of the ruins. No metal or metal objects 
 were found. 
 
 As to the kinds of pottery found little need be said in this sum- 
 mary, in addition to the details and illustrations elsewhere given of 
 the individual ruins which we have examined, save to note that in 
 form and decoration it conforms to the type found in many other 
 ruins of the Mesa Verde district as defined by Fewkes. 1 Some of 
 it is extremely crude in form, texture, and decoration; some is excel- 
 lent. In some of the crude corrugated pots the fill of broken quartz 
 is so coarse as to render them very rough and friable; but on the 
 other hand some of the smooth ware ollas are fine in texture, excel- 
 lent in temper, and are polished with the care which marks the 
 highest technical skill in this form of prehistoric pottery. I have 
 
 1 See references, p. 4- 
 
44 AMERICAN ANTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 never found the "life line"; nor have I seen depressions in the 
 rounded bottoms of bowls or ollas. 
 
 In a mug which had been exposed to the intense heat of a house 
 conflagration the handle had been detached, leaving a clean-cut 
 hole near the rim and one below, showing that to secure a firm 
 attachment of the handle a hole about one third of an inch in 
 diameter had been cut through the already shaped body of the mug 
 through which the clay cylinder which was to form the handle was 
 inserted to be then flattened and smoothed out in the interior and 
 shaped without. 
 
 The dippers found were mostly thick with square-edged rims; 
 the latter, however, in many cases, as is common in this region, 
 were beveled on the edge opposite the handle by scraping against 
 the rock when long used for dipping water from the pools. Some of 
 the dipper handles were flat and solid, but more were hollow cylin- 
 ders, closed and narrowed a little at the distal end, where there was 
 often some simple moulding of the clay to form small teat-like pro- 
 jections or crude suggestions of animal forms. Many of both the 
 solid and hollow dipper handles were pierced with a single or double 
 row of small holes or were decorated with stripes or cross bands of 
 black. 
 
 Most of the corrugated pots were smoked on the outside. In a 
 few some blunt-pointed hard object had been drawn across the coils 
 on the outside, after the shaping of the pot, making shallow, incised 
 lines and bands. We found none of the spiral or other figures which 
 in this region are not infrequent on the outer surface of corrugated 
 pots just below the rim. We found a few fragments of smooth 
 ware ollas on whose surfaces rude geometrical figures had been 
 incised in narrow shallow lines. In one specimen of smooth white 
 ware rows of shallow punctate depressions on the surface in uniform 
 parallel lines suggested the use of a roulette. We found only frag- 
 ments of red ware bowls decorated with black. 
 
 Now on the basis of the study of seven ruins of the unit type in 
 the northern San Juan district recorded in this and in the previous 
 paper, one would seem to have in hand sufficient data to justify 
 
pruddex] PREHISTORIC SMALL HOUSE RUINS 45 
 
 some general statements as to the character of these early homes 
 and the cultural status of their makers. For although only aboul 
 one third of the district has been covered in the widely distributed 
 examples selected for excavation, the uniformity in site grouping 
 and appearance, of the undisturbed ruins of this type over the whol< 
 district, as determined by the writer's earlier reconnaissance-; to- 
 gether with the almost complete similarity in structure of those 
 which have been dug out, would seem to indicate unity in structural 
 type. The sherds also, which owing to the uniform presence of a 
 burial mound, even in the smaller and more primitive of them, as 
 well as the many collections of whole pieces of pottery and other 
 artifacts made by many random delvers in the burial mounds which 
 the writer has had the opportunity of examining, suggest the uni- 
 formity of culture in these primitive house builders, over the whole 
 northern San Juan district from Pine river westward to Grand 
 gulch. Even the boulder sites, which are so common on the benches 
 bordering many reaches of the San Juan river, the Mancos, the 
 La Plata, the Animas and Pine river, may safely be presumed to 
 conform to the others, showing as they do when sufficiently pre- 
 served, the characteristics of the associated house, kiva pit, and 
 burial mounds. 
 
 Summary. — We may then venture to describe these little pre- 
 historic dwelling places in the open country of the northern San 
 Juan district, which conform to the unit type, and built on level or 
 slightly sloping earth sites, as follows. They commonly consist 
 of a square-cornered, flat-roofed, one-storied stone or stone and 
 adobe house or pueblo ; with usually one or two short rows of small 
 rooms; sometimes with side wings, without doors in the outer wall- 
 but in the roofs, and facing southward. An underground passage- 
 way leads from the floor of one of the center rooms of the pueblo to 
 the banquette level of a subterranean circular kiva either close to 
 or a few feet southward. 
 
 The kiva, almost always more carefully constructed than is the 
 pueblo, is usually, in part at least, stoned and plastered, with stone 
 pilasters for the support of the roof, recesses, and banquettes be- 
 tween them, the southern banquette being usually the deeper. 
 
46 AMERICAN AXTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 There are niches or cubby holes in the northern wall beneath the 
 banquette; a centrally placed fire pit in the plastered floor; a venti- 
 lator flue passing beneath the southern recesses at the floor level 
 and rising to the surface of the ground directly outside of the 
 southern kiva walls. A deflector is placed between the fire pit and 
 the ventilator opening. A sipapu, or a depression in the floor which 
 may represent this, is commonly present. On the top of the kiva 
 was probably a small level plaza or court close to the southerly 
 wall of the pueblo. 
 
 The best conception of the relationship of the kiva to the pueblo 
 is derived from a vertical central north and south section such as is 
 shown in the photo of the model (plate v). In such a section the 
 southern deep recess seems to assume an exaggerated share in the 
 subterranean space, since in fact it occupies only a short segment 
 of the circumference of the kiva above the banquette. 
 
 Thus these small ruins, which over a large district we have found 
 to be so uniform in structure and in the intimate relationship of the 
 kiva to the connecting group of domiciliary rooms, belong in the 
 middle period of pueblo development, as laid down by Fewkes. 1 
 This he believes followed a period in which the ceremonial struc- 
 tures — towers or kivas — and the secular living rooms were distinct; 
 and was succeeded by a period which developed various forms of 
 ceremonial chambers related to but not structurally connected with 
 the domiciles, as is also the case in several modern pueblos. 
 
 The artifacts which these studies have disclosed in the small 
 house ruins do not seem to be in any way different from those of 
 the builders of the larger houses of various types in the Mesa Verde 
 district. 
 
 A burial mound lies south or southerly, a few feet away from 
 each pueblo or small congeries of pueblos. In these the usually 
 flexed burials are sometimes with but more often without covering 
 stone slabs. With the remaining bones, commonly much disinte- 
 grated pottery or other mortuary offerings may be found. 
 
 While, as has been said, these little ruins are often widely scat- 
 
 1 Fewkes, " Archeological Investigations in New Mexico, Colorado, and Utah," 
 Smithsonian Miscellaneous Collections, vol. 68, no. 1 (191 7), p. 37. 
 
pruddex] PREHISTORIC SMALL HOUSE RUINS 47 
 
 tered over the summits of the uplands and through the valleys, I 
 wide and narrow, in the northern San Juan district, they are more 
 often in community groups upon favorable sites. This may be 
 on the reconnaissance map of the San Juan watershed accompanying 
 my first paper, 1 on which a single red spot in many instances re] 
 sents a whole compact community of houses. In such communities 
 one house is often larger than the others and stands commonly on 
 the most commanding part of the site. This is the rase in the 
 group near Cortez in the Montezuma valley in which my first 
 excavations were made. 2 Such a larger and more commanding 
 house in the community was also at the top of the ridge at Bug 
 point. 
 
 In the region which our excavations have covered, small ruins 
 of the unit type are in part close to as well as widely scattered main- 
 miles away from the great community houses on Squaw, Bug, and 
 Goodman points; and the less compact but still imposing groups at 
 the Yellow Jacket spring on the line of the old Spanish trail; and 
 many others at the heads of gulches or on commanding points on 
 the Yellow Jacket drainage, as at the Cannonball, the Hawkberries, 
 Ruin canyon, etc. 
 
 Similarly in the Montezuma valley, where some of our excava- 
 tions of 1913 were made; in addition to the hundreds of small ruins 
 of our type are the Buckhart ruin, sprawling over the rock escarp- 
 ments of a small branch of the upper McElmo; and the great Aztec 
 spring ruin at the eastern foot of Ute mountain. Among the gullies 
 at the western base of the Mesa Verde; along the bottoms of the 
 Mancos river, where it cuts through the rugged plateau; as well as 
 over the sage, pinon and cedar-clad summit of the mesa, small stone 
 heaps with their sunken kiva sites and their burial places are not far 
 away from the great picturesque houses in the cliffs. The great cliff 
 houses in Grand gulch, in Utah, and those along the western ridges of 
 
 1 American Anthropologist, vol. 5 (April-June, 1903). 
 
 2 In this larger and apparently two-storied house ruin, which has been much dug 
 over, Mr. Husted, the proprietor of the ranch on which the group is located, informed 
 me the interior door to one of the lower rooms was found walled up and the room con- 
 tained a skeleton beside which were some excellent pieces of pottery. The finding of 
 closed rooms in prehistoric pueblos used as mortuary chambers has been several times 
 recorded. 
 
48 AMERICAN ANTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 Butler wash, look out upon many small ruins apparently of our type, 
 which, except for their burial mounds, have as yet remained un- 
 touched. 
 
 While this study has been limited to the small house ruins of the 
 unit type, it should be noted that all over this district larger ruins 
 are to be found which show similar close relationship between cir- 
 cular kivas and domiciliary structures connected with them. 
 Furthermore many of the larger complex ruins in this district are 
 composed, in part at least, of a number of units; that is, a kiva with 
 its associated secular rooms, crowded together upon irregular sites 
 such as heads of gulches, mesa edges, cliff caverns, etc. In con- 
 nection with his illuminating studies of the Cliff Palace and the 
 Spruce Tree House on the Mesa Verde, Fewkes has called attention 
 to these congeries of "units" as composing larger ruins. 
 
 This early American who built these primitive houses so strik- 
 ingly alike over a large district in the northern San Juan watershed, 
 and apparently in many places south of the river, was first of all a 
 farmer and a hunter. But he was also an excellent mason, as good 
 a carpenter as might be with rocks and fire for tools, and a skillful 
 potter, in much of whose handiwork a genuine artistry finds quaint 
 expression. The structure of his kivas shows that he was a devotee 
 to religious ceremonial; that he had a heritage of tradition which he 
 shared with the dwellers in the cliffs and other kindred folks who 
 for some reason built bigger houses than he did in the open country 
 and that linked him to the modern Pueblo Indian. 
 
 He set the lines of his houses and of his ceremonial structures 
 according to the courses of the sun and the stars, and faithfully 
 provided food and utensils for the welfare of his dead. With a 
 ventilating system which would shame many modern house builders 
 he freshened the air of his darksome chambers underground where 
 he worked and played and kept in touch as his traditions bade 
 him, with Unseen Powers which shaped each detail of his life. I 
 guess he played and tinkered and gossipped on the flat roofs of his 
 box-like houses overlooking his neighbors and his neighborhood, 
 for that is what his surviving kinsmen do to-day at Zuni, at Acoma, 
 and in Tusayan, and in many other places. I know he must have 
 
 R D 1.4 8. 
 
PRUDDEN] PREHISTORIC SMALL HOUSE RUINS 
 
 found snug comfort in winter on the little plaza in the sunn) 
 of his pueblo, for he was quite human and none too generously clad, 
 for warm garments were hard to come by in those times and the 
 winds are keen which sweep over the bleak reaches of the uplands 
 in winter and down the valleys from the great mountains fringing 
 the northern borders of this land. It is not likely that he wandered 
 far from his home country, for one does not find among his scanty 
 belongings much evidence of barter with folks of other regions. 
 
 Attention in this study is not concentrated upon a particular 
 type of small habitations in the open country with the intent to dif- 
 ferentiate, either in time or place or culture, between their makers 
 and the builders of those other dwelling places of larger and more 
 complex types, which are widely distributed in the same districts 
 and often in close proximity. Whether the closely related people 
 who build the small isolated houses with which this study deals, 
 and the big communal dwellings in the same territory, occupied 
 them and the region at the same time or whether one followed the 
 other, and if so which were the pioneers, is still an open question 
 not to be settled until many more careful studies of many more 
 ruins of both types shall have been made. 
 
 Similarly, further studies must show whether the fashion of 
 separate habitations for individual social units first found expres- 
 sion, as seems probable, in the scattered homesteads, to be later 
 adapted to the exigencies of a closer communal life; or vice versa. 
 So far as the studies of both types have gone, there is evidently a 
 close relationship between the larger and the smaller forms, as well 
 as between the artifacts which the excavations of both have dis- 
 closed. The motive for the different types of building seems, in 
 part at least, to have been determined by the difference in site and 
 some as yet undisclosed community requirements. To what extent 
 the latter were of a defensive character is not altogether evident. 
 But apparently protection against human enemies did not enter 
 into the requirements of the time in the location and the making 
 of our little flimsy scattered open country houses. Even the dimin- 
 utive size of the houses and the inevitable mergence of their dull 
 gray walls into the kindred shades of the earth, the rocks and the 
 
50 AMERICAN ANTHROPOLOGICAL ASSOCIATION [memoirs, 5 
 
 foliage, would scarcely conceal them from the keen eyes of marauders 
 armed with the acute vision of men schooled in the open. 
 
 So one may safely conjecture that these scattered unprotected 
 farmer folk went about their tasks, practised their sports, and fol- 
 lowed the lead of their dominating traditions, for a long time at 
 least, quite unmolested. And while their houses, if snug, were un- 
 doubtedly stuffy, one may not forget that their lives were spent 
 largely out of doors and in a land which now as then lends itself, 
 austerely it is true, but still generously, to the love of living. 
 
■^ 
 
& *,-**~ ".» <?- 1 
 
 
 
 
 
 ,? 
 
 x ^ 
 
 
 
 
 \p9 
 
 
 
 A 
 
 
 > 
 
 
 
 
 
 -^.^ 
 
 c 
 
 u- x 
 
 -Q V 
 
 •V- ' 
 
 *^o* 
 
 < it/ 
 
 X Vj <£"°*<- "10 
 
 
° ^ 
 "by" 
 
 '«v > ^ / 
 
 
 
 ?•• ^ v % '■. ,/ V ^ «.*~*< 
 
 
 
 
 
 Y 
 
 ^ y 
 ■j 
 
 4 
 
 <*, ,^ V *^ 
 
 
 <*V 
 
 
 
 
 :■ 
 
 
 ^ : 
 
 ^ 
 ^ 
 
 -.- 
 
 
 A w ■*> '- 1 <V 
 
 
 ' . • * 
 
 s0 vV 
 
 
 . * 4 < 6 
 
 »>rf 
 
 o > 
 
 <0 -^ > 
 
 V 
 
 o 
 
 A 
 
 D0B6S ERGS. 
 
 LIBRARY BINDING 
 
 
 ST. AUGUGTINE 
 ^ FLA. 
 
 *W :g 
 
 ^ 
 
 .0' 
 
 J084 V 
 
 
 & .. 
 
 v v -^ 
 
 '-'\ 
 
 ,* \ . 
 
 V * ! 
 
 
LIBRARY OF CONGRESS 
 
 010 545 589 6