Castle A originally contained five stories, with possibly a balcony level atop the fifth floor. A calculation as to the number of rooms it contained can be, beyond a certain conservative estimate, only hypothetical, for the cataclysm which resulted in its destruction so completely tore away the front rooms of the third, fourth, and fifth floors as to leave only the slightest trace, if any, of certain rooms. When the building fell it simply peeled off from the face of the cliff, almost as if a great knife had made a diagonal slice from the cliff wall at the top to the front base. (Fig. 1, p. 10.)
At present there are positive indications of 26 rooms, some showing a complete floor outline, others only a corner or wall trace. Beam sockets in the cliff face, and evidence on ceilings which were excavated from the two lower room tiers, indicate conservatively a room total of 44 or 45. A dozen more rooms could easily have existed.
The floor remnants and caves of the third, fourth, and fifth floors, and part of Rooms 1 and 2 on the second floor were excavated or cleaned by George Boundey, former custodian of Tumacacori National Monument, in the spring of 1927. No report on his work is available. The rooms he excavated seem consistent with the construction type employed in the later type rooms which this work has uncovered, and so will be covered by that description.
The rooms uncovered in this project are numbered in the order of excavation. The second, or upper tier, was cleaned from east to west and includes Rooms 1, 2, 3, 4, and 5, and was completely finished before the first, or lower, tier was begun. Rooms la, 2a, 3a, and 4a were found to lie immediately in front of their corresponding numeral indications of the former group. (See ground plan on folded map.) The caves which were cleaned will be described in conjunction with the rooms which they adjoin.
Structural features typical of the whole building, insofar as our incomplete evidence reveals, include: use of masonry walls, and ceiling dependent upon two master beams running transversely to the long axis of the room.
The walls, of uniformly crude construction, were built of unfaced limestone blocks set in thick courses of caliche mud. No knowledge of staggering, or breaking of joints, was had by the builders. Room walls were double, averaging between 18 and 24 inches in thickness, without rubble fill, although small stones were used for chinkers. Walls of storage chambers and small caves were of single construction, averaging about 1 foot in thickness.
Room 1. This room was previously excavated by Boundey. Nothing was found, in cleaning. There were no ceiling indications left, and no signs on the cliff wall forming its north side of beam sockets for a second story. The southeast corner of the room was built over fill, and has sloughed off, together with the east and south walls. The remaining floor section is cavate and is separated from Room 2 on the west by the fragment of a double wall partition. A few inches above the floor level, in the north side, are two large cavate storage chambers.
Room 2. This room was also excavated in part by Boundey. Cleaning away of debris revealed a complete floor, of the same style as in Room 1, save that the floor had been plastered with a reddish clay. In the north side two small natural eaves had been partially walled up from below, to serve possibly as storage pits. Wall fragments were found on the east, south, and west sides. The south wall, being the longest, was expected to show signs of a side doorway, but was so badly crumbled nothing could be determined. A clay ridge, running north and south 6 feet east of the west wall, marked a 6-inch drop in the west third of the floor. In the west side of this ridge, slightly south of the floor center, was found the charred stump of a large post upright, which had served as a support under one of two large master beams. Sockets in the north wall, 11 inches in diameter and 8 feet above the floor, had supported the north ends of the beams, while the south ends had rested on top of masonry wall. Other beam sockets, higher in the cliff, show definitely the original presence of two stories above Room 2.
Room 3. The floor is cavate, while the north wall is cliff. The south end of the floor was built upon fill, as the natural ledge graded off sharply before the end was reached. (See floors in Fig. 1.) The firebox, in the south center, was lined with upright basalt slabs, making a pit 6 inches deep with edges protruding 2 inches above the floor. West of the firebox under a slightly elevated portion of the floor, were two shallow pits which contained three baby burials. (Photo 10.) (See Human Burials.)
The burned ceiling rested directly on the floor (Photos 8-9), and, because of a large metate which rested above it, we believe the roof to have marked a second story of occupation. No post holes were found in this floor. Master beams ran north and south. Above the closely spaced Sycamore rafters were sticks 1/2 to 1-1/4 inches in diameter, then a transverse layer of Reeds, then smaller transverse sticks, then a 4- or 5-inch layer of clay.
The south wall of the room was entirely gone, save for a pile of stones entirely below and in front of the floor level, which may have served as a rear masonry wall for Room 3a and as a front wall for Room 3. The east wall, adjoining Room 2, yet stands against the cliff to a height of 8 feet. The west wall, resting on an arm of the native limestone, is fragmentary. No sign of door opening was found.
Artifacts found on the floor consisted of: three three-quarter grooved axes (one double-bitted); five metates, one with mano in position; 12 manos, and parts of several others; one tiny, molded shallow bowl, 1-3/4 inches in diameter and 3/4-inch deep; two bone awls; one bone dagger; one bone needle having an eye (a rare feature); half of a spindle whorl of bark, 3 inches in diameter; one obsidian arrowhead; several fragments of stone hoes; many fragments of crude, plain pottery.
Two caves adjoined Room 3 in the north cliff face. The easternmost had never been inhabited, but was used solely for burial purposes (see Human Burials). The westernmost cave was of good size and showed much evidence of occupation, having considerable charcoal and broken pottery on the floor and a long-used circular firepit in the center of the floor. A storage pit in the north wall, partially sealed with a single stone wall, had served as a granary, as it yielded many fragments of corn stalks and leaves.
Room 4. This room was constructed in the same manner as Room 3, with one important exception: it had the south end of the floor built over fill and had the same arrangement of ceiling material, at least insofar as the badly broken charcoal fragments indicated. The floor showed no apparent layer of clay, although small patches were scattered over it, which may have belonged to the ceiling clay. The ceiling, on the north, or higher, side of the floor, contacted the bare rock, while the south side rested on stratified sand and lime dirt which had been washed in. The firebox, on the south center of the floor, was as in Room 3, only the slabs were of limestone instead of basalt.
This was the first room to reveal an unusual feature which appeared also in three of the others later excavated. There were 10 post holes in the floor, arranged into two groups of four each at the ends, joined by two larger holes nearer the center of the room. (Fig. 2b, Photo 11.) Post holes averaged 2-1/2 to 5 inches in diameter, corner posts and center posts being the larger. This use of supports for ceilings is very typical of the earlier pit house structures found in the Gila Valley and other parts of the Southwest, in which the posts form the sole support for the roof, in lieu of masonry walls. However, where such arrangements were used in this building we know there were walls, and hence no structural need for so many posts unless great weight was to be borne by the roof. That a second story existed here was shown by the discovery of a large metate, several manos, and a three-quarter grooved stone axe above the ceiling. On the floor of Room 4 were found four metates, several manos, and much broken plain pottery.
Eight feet above the floor of Room 4 was found the mouth to a cave room, opening onto the second story level. This room was entirely cavate, with a single-thickness masonry wall on the south. In the center of this wall was a narrow doorway, 15 inches wide, 29 inches high, the upper ledge forming the lintel, the bottom being 1 foot above the floor. The walls of this room showed no smoke, but the floor, plastered with clay, showed much blackening. One crude stone axe and a broken plain olla were found. Any perishable artifacts would have rotted, as the dirt which filled the cave was all rainwashed.
Room 5. As in Room 3, the front wall was entirely gone, and the floor sloped off with the ledge. The floor was entirely cavate, with cliff on the north and cliff slope on the west, and a high arm of limestone on the east, capped with fragmentary masonry wall. Stratified sand and lime soil, mixed, were found all over the floor to a depth of 3 inches, with gravel and larger material in layers on the top. Only the slightest traces of a burned ceiling had been preserved from action of water, although part of the clay roof material found on the floor showed the closely set, parallel impressions of Willow stems about 3/4 of an inch in diameter.
In the south-central part of the room (Fig. 2a) was an excellent firebox of lime slabs, mudded on the outside and plastered on the inside (Photo 12). Cutting off the western third of the room, running north and south, is the standing clay remnant of what was once a wattle-wall partition, with an entryway near the north end. This wall stands about 1 foot high, and 3 to 5 inches thick, presenting a smooth face on the east; the west side shows imprints of a vertical row of sticks, placed close together, between which the soft mud had oozed. The sticks have long since rotted away but the imprints convinced us they were willows, about five-eighths of an inch in diameter. Whether cross sticks interlaced the vertical ones when the wall stood to a greater height could not be determined.
Near the east wall of the room is a circular clay wall built on the rock floor; its walls average 4 inches thick and its diameter is 30 inches, with a height of about 8 inches. It may once have been a low dome, but the top, if such, has fallen and disintegrated. No sign of firing is visible within this enclosure, which fact precludes the probability that it was used as a firing or baking oven. Its function has yet to be understood.
The floor of Room 5, as in Room 4, is pitted with a number of post holes. (Fig. 2a). These 12 holes (Ed. note: 9?) are arranged much on the order of those previously described, and represent the same convention in building. Whether they represent a roof structure which was used before any masonry walls were constructed in front, or whether they were built along with the masonry wall, is problematical. If there were a shortage of long or strong timbers available it is logical to assume that the pithouse type of roof supports would have been of good secondary assistance to the walls. However, the repeated presence of this style of building, coupled with the older-type wattle partition wall, proves pretty conclusively that the builders had not forgotten pit house days.
On the floor of Room 5 were found: three large metates; four manos; two partial ollas, one of which had rested on the west end of the roof; one stone pounder; two large plain ollas, one resting against the firing oven (?) and one against the firebox, where some woman must have conceived the idea of keeping hot water at all times. (See Photos 12-14.)
Near one metate which was found in position were found two small holes in the floor, 2 inches deep and five-eighths of an inch in diameter, one 6 inches from the northeast corner, one 6 inches from the northwest corner. Two similar holes were found north of the large olla, one 8 inches northeast of it, one 6 inches northwest of it. The purpose of these holes is not known.
Because of the numerous interesting features about Room 5, the fact that it had been undisturbed save by water, and that there appeared to be no second story, it was decided to restore the walls and roof. This we did, making the walls of fallen rocks found in the debris and using the lime dirt from our excavation, mixed with water, for the mud. The roof was restored of Sycamore timbers, as that is the tree growth which abounds here and which was originally used. Over the Sycamore was placed a closely fitting layer of Willows, above which were transverse layers of light Willow and Cottonwood sticks, on top of which was placed a 5-inch layer of mud. (Fig. 3.) The roof had a slight pitch to give drainage. We knew the exact height of the roof against the cliff, because of the old beam sockets there, but had to guess at the height of the front. A side entrance was made in the center of the south wall, considered the logical entrance to a ground-floor room.
Room 1a. This room is at the extreme east end of the base ledge for Castle A, 11 feet above the river terrace level, and 9 feet below Room 1. To the north and west is solid limestone. No wall remains on the south at all and only a slight fragment abuts the north wall at the east end. Only the cavate floor outline, which is slightly lower than the rest of the ledge, indicated the boundaries of the room. Two to 4 inches of sand and lime dirt covered it, and this layer in turn had become so hard packed as to suggest another level of occupation with a new floor. No remnants of ceiling material were found, and nothing to suggest a second story.
In the center of the floor was a storage well, 2 feet wide and 2 feet deep, which could have held water. A few inches from the edge of the well was the single large post hole of the room, pointing to probable use of only one supporting master beam for the ceiling. Two storage pits in the north wall, one walled half way up from the bottom, produced only trash and broken pottery, none complete.
In the fill above the floor were found: two large metates; three manos, and numerous broken sections; parts of several plain ollas which had rested on the ledge outside the south wall of the house; half of a corrugated olla, and part of three Black-on-white bowls; one spindle whorl; one stone pottery molder; and one three-quarter grooved axe.
Room 2a. This is a large cavate room, directly in front of Room 2, and 10 feet below it, with cliff face forming the north side, and lime rock extensions from the cliff forming bases for fragmentary remaining masonry walls at the east and west ends; the south wall stands a few inches above the floor but presents no sign of a doorway.
A layer of scattered and charred ceiling fragments lay about a foot above the floor, while in the west third of the room other fragments touched the floor. Proof of a second story was lacking, for the broken pottery throughout an 8-foot thickness of debris could have fallen from any of the floor levels higher up the cliff, but it is believed to have existed. Two to 4 inches of water-washed sand and lime dirt covered the floor, and patches of clay which may have been a covering or merely parts of ceiling clay fallen through the destroyed ceiling, were scattered about, touching the limestone base.
The customary firebox of limestone slabs was found south of the center of the room. Six feet east of the west wall was a low clay ridge running north and south the width of the room. Extending from the north wall along this ridge for 7 feet, 6 inches, and then turning at right angles west to join the west wall, was the remnant of a clay wattle-work partition wall, (Photo 15) similar to that in Room 5. It was very low, and no positive doorway was found.
Only three post boles were in this floor, and they had undoubtedly been placed for much-needed supports to help hold the considerable span of roof over the east end of the room.
Found on the floor were: one excellent large metate; one small round basin metate; eight good manos, and parts of several broken ones; a chunk of yellow ochre; two broken stone hoes, and one complete; a thin, circular sandstone slab, painted red (probably used as a grinder for iron oxide paints); parts of three plain ollas, and a large portion of a broken Brown-on-yellow bowl.
Room 3a. This was the only room excavated having the longest axis running north and south. It lies directly in front of and 8 feet below Room 3. The north wall is now solid limestone, although a jumbled pile of rocks here probably formed at one time the base wall to the south end of Room 3. Masonry walls were built on the east, south, and west sides, although those to the east and south are practically demolished. That Room 3a was built later than Room 3 is indicated by the juncture of the west wall with a wall fragment parallel to the cliff; this is the only wall juncture showing positive chronology of building to postdate construction of the lower tier of rooms in relation to the second tier.
The floor is solid limestone, although the south end was built over dirt fill, and it had never been flooded as had the other ground floor rooms. Fragments of one charred ceiling were directly on the floor, while the second-floor ceiling, also charred, was about 15 inches above. No accurate details of either ceiling could be made out. Because of the shape of this room, the master beams probably ran east and west instead of north and south.
The customary firebox, contrary to form, was placed to the north center of the room. (Fig. 4.) Four feet south of the firebox a crack in the ledge had provided a sizeable hole (Photo 16), in which were found most of the pieces of a large plain olla. In the northeast corner was found a large olla, shattered but complete. By the side of this olla was a large metate, one end propped up on a stone, in position as it had evidently been used.
Northeast of the firebox, supine and extended on the floor, with feet to the north, was a skeleton of an adult, believed an elderly male (Photo 17). Because of doubt as to whether this individual was buried here, the circumstances should be described now. The skull was twisted sharply to the left and was too badly decayed to determine any sign of injury before death. The left arm was missing, but may have so decayed as to be unrecognizable. The bones were all in proper position, underneath a several-inch thickness of scattered ceiling fragments, but not burned; and directly underneath them, in contact with the floor, were more pieces of the same charred material. No burial artifacts were found, and no indication that a grave had been dug for the body. Whether this body was buried on the floor after the ceiling fell, or was lying on the floor at the time of the fire, or was buried under the clay floor of the second story cannot be determined.
Artifacts found on the floor, in addition to those already described, were: one round basin metate; one antler, badly rotted; three manos; two bone awls; parts of several stone hoes. From the second story came one grooved stone pick and one grooved axe. Near the west wall on the bottom floor was found a large section of charred and rotten basketry of a coiled weave strikingly similar to that of the modern Papago granary baskets, made of a grass rod foundation, and bound with yucca fiber, each binding strip piercing the adjacent bundle and holding some of the grass stems.
Room 4a. This room, also cavate, is in front of and below Room 4, separated by an elevation of 8 feet, 5 inches. The east wall, adjoining Room 3a, stood 8 feet high. The west and south walls were fragmentary, and there was a fragment of north wall against the ledge which probably served as the base to the front wall of Room 4. The floor was unevenly covered with a thin layer of clay, and the south end was overdirt fill. A low clay ridge 6 feet, 6 inches, east of the west wall ran north and south across the width of the room but was broken down at the south end. The customary firebox was slightly south of the center of the floor. There were nine post holes, four in a square on the east side, the others scattered (Photos 18-19).
The burned ceiling material, badly disintegrated, was 2-1/2 feet above the floor, while the stratified sand and lime dirt beneath was rain or flood-washed; the natural supposition of a second story found no support, for there was no sign of a lower ceiling nearer the floor. This unusual depth of fill beneath the superstructure suggests, if we assume the building burned all at one time, which is the most reasonable thought, that Room 4a had been abandoned long before any of the other rooms on the same level.
Another peculiar feature about Room 4a was a group of 39 small holes in the floor, arranged to the west, east, and north of the firebox. These holes, like the four mentioned in Room 5, were small and shallow, averaging five-eighths of an inch in diameter by 1 inch in depth. They seem to have formed a design (Fig. 5) similar to two horseshoes placed together with the open ends apart, with the ends of the north shoe closed by a straight line. These holes were probably made to hold the lower ends of vertical sticks, which in turn could have supported a wattle-and-daub framework; or, the sticks could have been forked at the upper ends and supported light crosspieces for a table or shelf-like arrangement. But neither of these functions could be expected here. The wattle framework would not have been in the right position (in front of the firebox) to act as a wind deflector, but could have sufficed to prevent sparks flying all about the room. All of these reasonings seem weak, and the suggestion of ceremonial significance occurs; yet there is no other structural feature about the room to suggest a kiva or other type of ceremonial chamber.
Artifacts found in Room 4a included: one basalt metate; one broken sandstone metate; one round basin metate, found above the ceiling (probably fallen from Room 4), with a grinding surface on one side and a pounding surface on the other; three manos; two or three polishing pebbles; one sandstone pot surfacer; two hoes, one broken; one full grooved axe and one three-quarter grooved pick at ceiling level; one tiny three-quarter grooved axe on floor at west end; one large complete bowl, shattered; parts of two large plain ollas.
Ten test trenches were dug in the terrace level below the fallen wall material to the south, southwest, and southeast of Castle A in search of graves or former levels of occupation. (See map.) These trenches extended southward from the cliff base.
No graves were found, and no floors, although plain, utility-ware pottery sherds and some finer sherds with highly-burnished interior surfaces were found in abundance in Trench 6 and to a lesser extent in Trench 9. Trenches 1, 2, 3, 4, and 5 were shallow, striking shelving limestone ledge at an average depth of 5 feet, showing stratified river soil and lime dirt from the top to the ledge base beneath.
Trench 6 was 27 feet long, extending from the cliff 21 feet, 6 inches, to the point of greatest depth. Lime and river soil were mixed to 9 feet. Nine feet to 10 feet, three inches, was lime fill. Ten feet, 3 inches, to 11 feet, 10 inches, was mixed clay and river soil with chunks of lime rock. Scattered sherds were found from top to 10 feet, 3 inches, and charcoal was scattered from top to bottom. A crude three-quarter grooved axe was found in a pocket of river soil at 9 feet, 3 inches.
Trenches 7 and 8 were shallow, and did not contact the cliff. Trench 7 struck solid rock, at 4 feet, which was evidently fallen from the cliff. No sherds found here. Trench 8 revealed nothing and ended in scattered lime rocks and river soil at 5 feet.
Trench 9, at 10 feet from the cliff, reached a depth of 12 feet, 9 inches. Eighteen inches of lime fill covered the surface. From 18 inches to 12 feet was river soil. Twelve feet to 12 feet, 9 inches, was river soil mixed with limestone rocks. At 6 feet, 2 inches, was a well defined layer of charcoal, containing plain sherds, but no slightest indication of any clay floor or even packed hardpan. The layer was about 1 inch thick. At 5 feet, 6 inches, 6 feet, 2 inches, 8 feet, 7 inches, and 12 feet, 9 inches, were found sherds, all plain and showing no variation. At 11 feet, 2 inches, to 12 feet, 9 inches, were found numerous small fragments of charcoal.
Trench 10, at a depth of 6 feet showed pure river sand and dirt, with no sherds or charcoal. It was discontinued at this depth because of tree roots.
Taking trench work in front of Castle A as a whole, we learned little. Testing the river terrace level proved that at one time the stream had swept in against the cliff under the building at a depth of at least 12 feet below the present ground level at that point. The presence of pottery at such depths indicates the stream changed its course away from the cliff face during periods of human occupation of the upper ledges, and that sherds and refuse were thrown over the ledges at different times, covered by flood overflows of sand and soil, and the process repeated. Lack of house structures, human remains, or artifacts on the terrace level as at present seen through sections, precludes possibility of giving great significance to the fact that no decorated sherds were found. The plain wares found were identical in type with those found in conjunction with decorated types of the later occupation of Castle A, but changes in utility ware were very slow, as we can see all through the Southwest. Further test work here would be necessary, and more material found, if these sherds were to be more closely located in time sequence.
Last Updated: 04-Mar-2008
Copyrighted by Southwestern Monuments Association
Western National Parks Association