In view of a widespread belief1 that the mounds lying southwest of Alto, Texas, mark the site of an historic Neches village and the second location of the mission San Francisco de los Tejas, and since the archaeological evidence denies such an identification, it is necessary to review the early history of this region, pointing out certain discrepancies in the literature.

The Neches was one of nine main trihes forming the Hasinai Confederation. Detailed accounts of their names, approximate locations, and political structure have been published by Bolton, Swanton, and others. Numerous names, with variations in spelling, have been given to this group of tribes. The French called them Cenis; the Spaniards, Hasinai or Tejas. These names were usually applied to the confederacy as a whole, but occasionally to a specific tribe or village. It is sufficient here to mention only two: the Nabedache, the most southwesterly group, among whom the first East Texas mission was established in 1690; and the Neches, who lived just across a river (presumably the Neches River) from them.

When the heavy growth of weeds, trees, and underbrush had been removed (Fig. 3, A) and the site examined preparatory to staking, it was found that either a north-south or an east-west axis would cut the mound at an odd angle, so the 0-00 line was laid down the long axis of the mound and the base line at right angles to it at Stake 0 (Fig. 4). The 0-00 line was 54° 30′ west from magnetic north. An alidade tower was built and permanently centered over a stake 5 feet back from Stake 0 in line with 0-00, and a datum plane established approximately 5 feet above the highest point of the mound. A telescopic alidade and stadia rod were used to determine all elevations, with steel tapes and hardwood rules 6 feet long used for other measurements. A portable photographic tower was also built and used whereever elevation was necessary to secure a complete photographic record. Approximately 250 photographs were taken during the excavation.

In general the same techniques were used in the excavation of all house outlines in the site, with, of course, modifications and adjustments to meet any unusual conditions. Whenever postholes were struck the immediate area wa's cleaned off and any suggested line of postholes followed until the entire outline was exposed or, as often happened, the postholes proved not to be part of a definite outline. The practice of profiling all postholes was soon instituted because the soil in them was often so nearly the color of the surrounding soil that their exact placement was impossible. These cross sections also gave exact depths and diameters, data otherwise impossible to obtain accurately.

When an outline had been roughly established, the top of each posthole was troweled out slightly and photographs were taken. A trench was then dug around, and just outside of, the outline and its inner wall cut back by thin vertical slicing until all postholes were exposed in profile (Fig. 16, E). Measurements and plots were then made. Fireplaces and centerposts, if present, were also profiled.

The accompanying Figure 26 was made by the writer from Newell's field notes and profiles. These profiles summarize the potential stratigraphy within the excavated leg of the mound in terms of the actual sequence of construction. (All remarks hereafter on the construction of “the mound” must be taken to apply only to the excavated eastern leg, subject to possible correction if and when the other leg is examined.) It will be shown that not all these stages of construction have determinable cultural significance; that is, some show no apparent differences in the associated cultural material and can be grouped into what we shall call “phases” of occupation.

The reader should first examine NewelFs contour map of the site (Fig. 4), showing the peculiar L-shape of the complete mound, the grid system of numbered squares, and the position of the numerous house outlines.

In the preceding section, a number of “zones” were established in the mound and surrounding fields, providing a framework within which to reconstruct cultural development at the site. We can now try to view the history of the site in somewhat broader terms, organizing the zones into stages of development, marked by such events as changing tastes and the appearance of new traits. The sequence of construction of the great mound may first be briefly reviewed.

The first occupation (always excepting such information as might be revealed in the unexcavated wing) can be determined by following backward the mound zones discussed above. The secondary mound (Zones 10 to 12) constituted a single unit of construction, completely overlying the broad primary mound platform, and lapping over its slopes. The platform itself had to be built up in a certain order because its fills slanted over one another in consistent manner.

In the two preceding sections, a structural framework for the site has been given which will enable the reader to visualize the manner in which artifacts were correlated with the growth of the mound.

With this in mind, it should be clear in the discussion below what I mean by experimentation in establishing valid types. That is to say, it was necessary to create trial “types” first, plot them against the physical strata of the site, first in one way, then in others. It was not until a picture of consistency developed, first within the site, and second in related sites, that I began to feel that the typology was on firm ground. After that it was necessary to re-group and re-plot the specimens against the strata, and finally to re-total the findings in terms of the “phases.” The cultural stratigraphy, then, is the result of a very long-drawn process of typological experimenting and plotting in a number of different ways, the final purpose of which was to find combinations consistently falling together and contrasting with other combinations.

The various potteries supposedly made at the Davis site are described below, as (1) types, and (2) descriptive groups of uncertain typological affiliation. The statistics cited are taken from Tables 3-11 and the summaries in Tables 13-14.

Descriptions of each major type have been made in the approved outline form of the Southeastern Archaeological Conference and are on file in the Anthropology Department of the University of Texas. They are not published here because I regard the pottery typology of this site, and the whole lower Mississippi region in general, as still in a state of experimentation. There may be many changes before we can be satisfied with the full ranges of variation (especially local variations) in many present types; there may be considerable shifting in the composition of known types, in conferences for that purpose or by other means, And, no matter how good and complete the formal outline descriptions are in any area, one cannot positively identify pottery types with them.

The pottery in the following sections is not considered to belong to the Alto Focus complex, but to occur with it at different points in the Davis site occupation by trade or other means. If the writer appears to vacillate over what is and what is not trade pottery here, it is due in part to the problem of separating what could have been produced at the site (as extreme variations of resident styles) from what probably was not (because of some distinctive attribute which would mark it as foreign). In certain cases of pronounced deviation, a foreign origin is obvious enough, particularly when the source areas are well known. But where the whole tradition is similar as in the clay-tempered pottery of the lower Mississippi Valley region, and a great range of decorative techniques was employed for long periods of time, the problem is not easy.

Nineteen ear spools are clay, probably tempered, polished, and fired in the same way as pottery vessels. Of these, one is part of a solid, bi-concave disk (Fig. 53, U); one is a small ring with slightly concave perimeter (Fig. 53, V); and the other 17 are parts of rings with strong flanges around both edges on the exterior (Fig. 53, W, X). The 17 fragments are from no less than 13 different rings, probably 15, but no definite pairs can be seen. Four are decorated with closely spaced engraved lines completely crossing the interior surface from one flange lip to the other (Fig. 53, W). Colors are like those of resident pottery: light brown, medium brown, reddish brown, and black.

One black fragment is from under the mound, beneath the floor of Feature 31 but above the floor of Feature 37. One other was in Phase 2 position, under mound wash; one is from Feature 9; one from the secondary mound; and 14, including the complete ring, are from the field in Zones 16 to 13. Although their numbers seem to have increased as time went by, this flanged form wasat least present before mound construction began.

With some exceptions, notably the projectile points, the various artifacts of stone, flint, and pigment are simple in nature and will be but briefly described. Figures 53-61 provide illustrations of nearly every descriptive grouping, specimens having been selected to show ranges of variation; the more variation —as in projectile point types—the more specimens are shown.

Tables 17-18 show the stratigraphic position of all stone artifacts and pigments found. Ordinary household artifacts such as milling stones, hammerstones, hones, knives, scrapers, and gravers were but sparsely represented in the mound, as might be expected. But since the mound provides our only sure stratigraphic control, the general dearth of utilitarian artifacts in it renders their occurrence in the three phases of occupation uncertain. That is, absence from one or more of the mound phases could be due to chance where only ten or a dozen (or fewer) specimens of a particular group came from the mound.

The vegetal remains, shells, and animal bones recovered from the site are, for the most part, evidence of the foods consumed here. They are arranged below by the phases and features in which they were found. Table 16 presents the associations with buildings, arranged according to position on floors, in subfloor pits, in postholes, or in fireplaces.

Table 15 summarizes the occurrences of pottery types and descriptive groupings in the buildings. As discussed in the section on Cultural Stratigraphy (pp. 66-70), some of the round houses under the mound were positively in Phase 1 because they were overlapped by the first mound fills, the four “small round mounds” called Zones 2 to 5. Features 10-18 were included in Phase 1 because they agreed with sub-mound houses in all particulars, and Newell’s notes stated that mound wash on the north side came down over them in part. Features 19, 29, 30, 32, 33, 35, and 41 were placed with Phase 2 in the artifact tables because there was some chance that they were not built until after the four “small mounds” were erected, or until the primary-mound platform was being extended laterally over the former village.

Below are given trait lists of the three mound phases, together constituting an Alto Focus. The “phase” framework is only a convenience for discussing what happened. At no time was there any pronounced change, it being quite evident that the three phases apply to a single cultural continuum undergoing subtle changes of two kinds: (1) modifications and innovations arising within the resident group, including the transfer of attributes from one type to another; and (2) acquisition of new traits by outside contacts, including the imitation of styles seen on “trade” objects.

The houses and non-pottery artifacts provide rather definite traits that can be listed in the conventional manner. Pottery, however, can be broken down into an infinite number of “traits” according to what the analyst wishes to present. Thus, various grades of temper, hardness, finish, design arrangement, even color tones, could all be listed as traits, as could any change in frequency in any feature at any given time. Naturally, these events cannot all be recorded, but the following phase lists attempt to show changes as well as patternizations during the life of the focus at this site; they should be more fruitful than plus-and-minus check lists.

The following discussions are necessarily abbreviated. The check-list method of showing trait presence or absence in other sites or cultures, while generally used to show “relationships,” will not be followed here. For one thing, these lists are almost always incomplete, the traits being followed only so far as the analyst has the time or the interest to follow them. For another, they are only inventories, with little if anything to do with culture processes such as selecting, development, and relative popularity; it is not the number of traits held in common by cultural entities, but the kind of traits, the way they cluster (or fail to) in the process of borrowing, and the use made of them that should occupy our attention. Third, a check-list made without thorough consideration of time factors is apt to throw together a miscellany of traits that show neither a direct relationship nor the direction of borrowing. Fourth, the mere number of traits in common between two groups is meaningless unless broken down into the more general or widely shared practices, and those special or specific traits which actually demonstrate a relationship between those two groups, and those alone. Thus, trait lists may be convenient for quick visualization of a given situation, but they are not studies in relationships.