Soloman River Phase
Donald J. Blakeslee 2008

Table of Contents
Settlement Pattern
Kinds of Sites
Kinds of Features
Cache Pits
Activity Area
Chipped Stone
Ground Stone
Bone, Antler and Shell
External Relationships
Native Exchange

Solomon River is a local expression of the Central Plains Mosaic (01), formerly called the Central Plains Tradition. The reason for the change is that the duration of the complex was not long enough to allow the definition of sequential phases. Instead, the phases that are well dated are more or less contemporaneous with one another. The Central Plains Mosaic, in turn, is a regional expression of the Plains Horticultural Tradition, which was previously called the Plains Village Tradition. The reason for this latter name change is that none of the sites of the Central Plains Mosaic were villages (Blakeslee 1999:31-38).

Solomon River phase sites occur in the Glen Elder locality – around the junction of the North and South Solomon Rivers (02). Sites of this phase were brought to light prior to the construction of Glen Elder dam and Waconda Lake. To the east is the Smoky Hill phase, and to the north one finds the Upper Republican phase, while to the south only fairly isolated Central Plains Mosaic sites have been reported. To the west are blufftop sites of the poorly-understood High Plains Upper Republican. No clear-cut boundary lines have been drawn between these units, and given the nature of Central Plains Mosaic adaptations, none are likely to exist.

Solomon River phase sites appear at about A.D. 1030 and appear to disappear by AD 1300 (Blakeslee 1999:44-46). This assessment is based on 13 dates run by Beta Analytic Laboratory corrected by the most recent calibration curve (03) and on a few early dates that have not yet been published (Richard Krause, personal communication). Dates run earlier by Gakushuin are now known to be unreliable (Blakeslee 1994), but the early literature on the Solomon River sites depended on those early and misleading dates. The sample of currently dated sites may be a biased sample of the total number of sites in the locality, as all of the excavated sites lie near the main rivers, and there are many other sites upstream on tributary creeks (Latham 1996) (04).

Settlement Pattern
The people who created the Solomon River phase appear to have lived in relatively isolated extended family homesteads within a scattered community of some sort (Blakeslee 2002). Analysis of the details of house construction indicates that many were built by two teams of people, each using their own traditional methods (Blakeslee 2005).(05). Homesteads were located near springs, groves of hardwood trees and easily-worked soils that retain moisture well (Latham 1996).

A debate about whether the sites were occupied year-round (Lippincott 1976) or only during the planting and harvesting seasons (Falk 1969: 51; Morey 1982) appears to have been resolved in favor of the year-round hypothesis (Blakeslee 1999:43-44; Dorsey 2000).
An older debate considered whether or not there was a definable sequence of different kinds of habitation sites. Originally proposed by Krause (1969), it was critiqued by Lippincott (1976, 1978), with a reply by Krause (1982) and further responses by a variety of authors (Blakeslee et al. 1982). Blakeslee (1999:22-30) reviews the debate and provides a recent perspective.

The size and distribution of the scattered communities is not known. There are localized traditions of house architecture within the Glen Elder locality (see below), and analysis of ceramics from ossuary suggests that the people who contributed offerings there came from a wider region than just the adjacent habitation site but not from just across the river (Blakeslee 2002:179-182).

Details of ceramic styles also indicate the presence of communities larger than single sites. At Waconda Lake, Blakeslee (1999:115-120) and his students detected four such ceramic dialects in the Glen Elder locality. The extent to which this patterning in space reflects movements of one or two families from one site to another is not known. The chronological control is not adequate to differentiate sequential sites within the short period of occupation of the locality.

Kinds of sites
Like the habitation sites in other phases of the Central Plains Mosaic, those of the Solomon River phase appear to be accumulations of the remains of houses that were occupied sequentially rather than simultaneously. They contain both houses and external activity areas (06), and only sites located in fairly extensive groves of trees have more than one or two houses. No kill sites have been reported.

The single instance of an ossuary/cemetery consisted of two ossuary pits (one of which was excavated) surrounded by numerous small pits. Many of the latter were empty, but some contained a few human bones and one held a complete skeleton. The mortuary sequence seems to have involved individual pit burials followed by removal of the remains from the individual pits for re-interment in the ossuary. Whether the individual burials were primary or followed an interval of scaffold burial is not known (Lippincott 1976; Blakeslee 1999:145-147)..

Kinds of features

Solomon River phase houses were built after stripping the sod from the ground surface, creating a very shallow pit. Four or more centrally located support posts were then erected, and the walls of the house were made from smaller posts. Cross beams between the tall central posts and along the lower wall lines supported rafters or cribbed poles (06). This layer was then covered to make the roof, but the precise nature of the covering has not been determined; rather it has always been assumed to have included an outer layer of sod. The nature of the covering of the vertical walls is also not known. An extended entryway issued from one wall of the house.

There are localized variations in the details of the houses (07). On the South Solomon, they were longer than wide with the entry coming from one of the short walls. There is no clear pattern of central support posts, and the fireplaces are offset slightly toward the entryway. They contain from four to ten internal cache pits. (08). At the mouth of Oak Creek, the houses are fairly square with four substantial central support posts. One lacks a hearth; the hearth in the other is central to the support posts. They (N=2) contain five and seven internal caches (09). Also on Oak Creek are two square houses with many wall posts and many small central posts. Both have central hearths, but one lacks internal caches while the other contains only two (06). At 14OB27, the most of the houses have trapezoidal shapes, and the pattern of center post in them is also trapezoidal and not parallel to the walls. Hearths tend not to be located centrally among the center posts, and the entryways are neither centered on the walls not perpendicular to them (10).

At the eastern end of the locality, on Limestone Creek, only one house has been excavated, but it differs from all of the other styles. It is nearly square, with numerous wall posts, a clearly defined pattern of central posts and a centrally located hearth. It contains four internal cache pits (11).
Like the houses in the Glenwood locality of the Nebraska phase and in the Medicine Creek locality of the Upper Republican phase, at least some Solomon River phase houses appear to have been built by two teams of people responsible for different walls. In some instances, the front of the house differs from the back; in others the differences are between the two sides. Blakeslee (2005) interprets the patterning as reflecting different family traditions of architecture. Since some of the smallest houses exhibit this pattern, there is reason to think that houses were always built by two teams, with the houses built entirely in one style being the products of two closely related teams.

The hearths, both internal and external, consist of shallow unlined basins. Their contents usually consisted of ash and charcoal mixed with bits of bone and fragments of artifacts. Often the hearths exhibit a fairly thick layer of reddish baked earth.

cache pits
Cache pits were made in a variety of forms. The largest are bell-shaped, although these are usually less than a meter deep (Lippincott 1976: Appendix I).. Cylindrical caches tend to be smaller, with about one sixth or less the capacity of the bell shaped features. Small to medium sized basins that contain trash may not have been intended for storage. Many of the caches also ended up filled with trash or ashes from a hearth.

In many sites, piles of mussel shells occur on the surface or in the fill of cache pits. These were analyzed by Ron Dorsey, who found that they reflected clambakes held at about five year intervals, always at the end of the warm season. Much smaller accumulations of shells also occur, and these reflect mussels taken at all seasons of the year and at closer intervals of time than the large clambakes (Dorsey 1998, 2000).

Activity Areas
Adjacent to the houses in many of the sites, excavation uncovered scattered post molds, cache pits and hearths that did not seem to derive from houses. (Blakeslee 1999: 66-69, 73-74). No two of the activity areas looked much alike (10). A few may be the remains of poorly-preserved houses, but most exhibit patterns of placement of features that is totally unlike those of the various house forms.

The people who created the Solomon River phase sites had a broad subsistence pattern consisting of horticulture, hunting, fishing and gathering. Crops included maize, sunflower, squash and marshelder, and the presence of tobacco pipes suggests that tobacco was grown. Since the sites were excavated mostly without the use of flotation, other crops may have been missed, and the data on the frequency and ubiquity of plant remains has not been calculated. Blakeslee (1999: 89-92) argues that the form of horticulture here and elsewhere on the plins was a form of swidden. Large clambakes that occurred at about five year intervals and always at the end of the warm season and organized hunts of certain rodent species (for which we do not know either the season nor the temporal spacing) may reflect the cyclic failure of swidden gardens (Blakeslee 2002:xx).

Small animals dominate the faunal assemblages in spite of the fact that neither flotation nor water screening were used. In Central Plains Mosaic sites where waterscreening was practiced, deer are the third most common species, bison is fifth, pronghorn twelfth and elk eighteenth (of 26 species and two genera) (Blakeslee 1999: Table 30). Non-migratory large game may have been hunted out quickly around the year-round habitations, and garden hunting seems to have been important both to protect the gardens from pests and as a way off adding meat to the diet (Blakeslee 1999:77-85).

Fishing, gathering of a wide variety of wild plants, especially nuts and the prairie turnip, along with taking an occasional bird rounded out the diet. The winter food supplies of some rodents may have been taken, at least episodically when the swidden gardens failed.


Most Solomon River pottery is in the form of utilitarian jars, with a few bowls and miniatures also present. All but the miniatures were formed by the paddle and anvil technique, which is obvious from the cord roughening on vessel bodies (13). All jars are utilitarian, and some sloppy products are present (14), but most vessels, although quite plain, are well made (15). The cord roughening on lower bodies is never removed by smoothing, and shoulders are rarely smoothed. Rims are sometimes smoothed and lips frequently are, although some cord roughened lips (16a) and collar faces (16b) are present. Decoration on rims is sometimes applied directly over cord-roughening (17), a trait rare in other localities in the Central Plains Mosaic.
Blakeslee (1999:103-108) defines a Carr Creek ware with seven constituent types based on rim form and decoration. The Tipton group consists of jars with direct rims (21), while the Downs group is made up of jars with collared rims (18). Tipton types include Tipton plain (19), Tipton decorated lip (20), Tipton modified Edge (21), and Tipton decorated face (22). Downs types are Downs plain (23), Downs modified base (24), and Downs decorated face (25).

Decoration, when present, consists of punctations (26a), incising or tool impressions on the lip, outer lip edge, rim face, and collar face, and pinching or tool impressions on collar bases. Collar face motifs are geometric (26b). Shoulders are not decorated.
Temper is mostly grog or sand, and the sand is either well rounded or highly angular, with the latter derived from local bedrock sources. A few shell tempered specimens have been found, some of which are identical in paste to the grog and sand tempered vessels.
Bowls are small compared to the jars and usually constrict toward the orifice (27). The bowls and miniature jars occur only in small numbers. They tend to differ from the jars in details of paste and temper. Most of the miniatures were formed by pinching (28a), but some nevertheless were finished with a cord-wrapped paddle (28b).

Slightly differing styles of ceramics can be discerned in different parts of the locality (29).
Sites at the west end of the locality contained pots with tall, fairly vertical rims, including a relatively large number of the Tipton decorated face type. The only motif on the Downs decorated face rims were horizontal lines. Midrim fillets and wavy lips also occur.
In sites along the South Solomon, decoration is rare except at 14ML17, which has a fair number of Tipton decorated face vessels. Motifs on collared rims include both horizontal lines and zig-zags. Tipton ware vessels from the eastern sites have flaring rather than vertical rims, and a high proportion of Tipton modified edge vessels are present. Decoration created by striking the damp clay with a pointed stick is present at both of the easternmost sites but is missing in all of the others in the locality.

The sites along Oak Creek in the center of the locality, with two exceptions, affiliate with the western (1), southern (1) and easter dialects (3). The exceptions, at the mouth of Oak Creek, contain a high proportion of Downs decorated face vessels including the only ones in the locality that have a band of diagonal lines along the top of the collar face.

chipped stone tools
Solomon River phase people obtained the bulk of their stone from northern Kansas (30), with only Flattop chalcedony and Alibates coming from outside the state (31). The bulk of the chipped stone in is comprised of Smoky Hill jasper (32) and gray Permian chert (33). Some of the latter is clearly from the northern portion of the Flint Hills (34) and all of it may be. On average, 70% of the stone is Smoky Hill jasper, of which about one third was (probably) local upland lag gravel; the rest must have come from bedrock sources 50 or more kilometers away. Slightly less than one fifth of the chipped stone is gray Permian chert.

Small amounts of Flattop chalcedony (35), Alibates chert (36), Ogallala silicified sediment (37) and petrified wood are present. Of these, the silicified sediment is the most ubiquitous, occurring in ten of fourteen sites. It is available in small amounts in gravel exposures all across the western part of Kansas and may have been obtained fairly locally. Alibates chert, from the Canadian River some xx kilometers away, was found in eight sites. Surprisingly, one fairly large core of Alibates chert was found.

Most of the Smoky Hill jasper was brought to the locality in the form of bifaces, as the raw material occurs naturally in thin slabs (38). Chipped stone tools include side notched arrow points and point preforms. The latter are more crudely flaked and somewhat larger than the notched points (39). The finished points include single and double side notched varieties with and without basal notches (40). Somewhat larger unnotched and notched points are found in small numbers (41).

Knifes include the alternately-beveled diamond-shaped type called Harahey (42), ovate (43) and fishbelly (44) forms. Informal cutting tools made from flakes are common. A few knives exhibit the heavy rounded wear that comes from cutting soft stones such as pipestone (45).
End scrapers are common. They are typical of Middle Ceramic Period in terms of size and proportions (46). The proximal ends lack evidence of haft wear, and most are large enough to have been held in the hand during use.

Chipped stone celts occur fairly regularly (47). Their primary use was probably for girdling trees for swidden horticulture, but they could have also been used as weapons as is the case with celts elsewhere in the world. They could also have been used in rough butchering of bison joints, but we lack direct evidence for such use.

ground stone tools
Stone tobacco pipes occur in a variety of forms. One has a prow at the base of the bowl like some Nebraska phase pipes, but the bowl is quite tall and meets the stem at less than 90 degrees, making the pipe quite distinct from those found in the Nebraska phase (51). Other elbow pipes lack the prow but have the slightly tilted and relatively tall bowl (52). The acute angle, tall bowl and short stem are shared with Great Bend pipes, but the Great Bend specimens have much taller bowls which bulge on the smoker's side giving the bowl the general appearance of the calf from a human leg. Biscuit pipes are also found (53). One pipe with a short, widely flaring bowl and a short stem that meets the bowl at an obtuse angle is quite distinct and may be an import (54).

Sandstone shaft smoothers (55), grooved abraders (56) and ungrooved abraders (57) are fairly common. Hammerstones occur, and the disk-shaped handstones with a cup in the center of each face may have been used as hammerstones (62). Nutting stones (59) with one or more pits have been found. Some are quite large; one from 14ML15 measures 22 by 22 cm and has five pits on one face and a single pit on another. Also quite large is a metate from the same site (60). Quite thick, they were clearly intended to be permanent furniture within the site. On the other hand, a carefully shaped thin grinding slab from 14ML8 appears to have been made as portable as possible (61).

Handstones are usually small and some exhibit central pits on both faces (62). The pits are generally ovoid and shallow and do not similar to the pits on nutting stones. The remainder of the faces are flat, indicating that the stones were used with a rotary motion on a flat rather than concave surface. What appears to be an artificial limestone ball was found at one site (63). The author has seen similar balls in sites in Nebraska, but their purpose is unknown.

bone, antler and shell
The most common bone artifacts are fragments of scapula hoes. The form of hoe made in these sites usually has an intact glenoid end, but sometimes it has been trimmed (71).A completed but unworn hoe has a convex bit (72), but the thin central portion of the blade wore away quickly, and most hoes that are found have an (often irregular) concave bit. Broken hoes were fashioned into a variety of useful items. One was used as a cutting board (73). Another bone horticultural tool was a broken bison rib used as a digging stick (74). Among the historic Arikara, such tools were used by men to cultivate plots of tobacco.

Awls are the next most common bone tool. Frequently, they were made from deer and antelope metapodials (75) by the groove and snap technique, and the byproducts of this technique are often found (76). Especially long specimens may have been used as hair or blanket pins (77). Splinter awls were also made (78).

Awls with points to blunt to penetrate deer or antelope hide were probably used in making basketry (79). One awls has hack marks at the proximal end, and it may have been used as an eyeless needle (80).

Antler cylinders are found (81). Some show signs of use as flaking hammers, but others are merely smoothed. Antler tine pressure flaking tools are also present(82). Bone tubes of various sizes are found. The smaller ones may have been beads, but larger ones (83) may be fragments of whistles or shamans' sucking tubes. Beads were made from both bird and rodent bone (84). The occasional hollowed and perforated deer phalanx (toe bone) may be from the ring and pin game or from tinklers worn on clothing (85).

Shell tools include scrapers for shelling corn (86), shell spoons (87), and shell hoe blades (88), but most uses for shell were ornamental. Marine shell ornaments included gorgets (89) and columella pendants (90) of conch or whelk, marginella shell beads (91) and disk beads (92). Fairly large snail shells might represent natural accumulations, food remains or items intended to be made into ornaments (93).

External Relationships

Native Exchange
No obvious examples of trade wares occur in the ceramic assemblages. There is, however, more variability in the assemblage from the mortuary site 14ML16 than from any of the habitation sites. The people who contributed ceramic offerings there -- in the form of sherds rather than complete vessels -- may have included individuals from outside the locality.

In the chipped tone tool assemblages, there is no local material other than some of the Smoky Hill jasper and perhaps the Ogallala silicified sediment and petrified wood. The latter two make up only a small proportion of the stone at any site. The rest of the Smoky Hill jasper and all of the gray Permian cherts come from well outside the locality but may have been obtained by direct acquisition rather than exchange. The Alibates chert and Flattop chalcedony are better candidates for products of exchange, but there is no clear evidence that they were not also the products of occasional visits to the sources.

In the ground stone category, the unusual pipe with the flaring bowl (54) could well be a trade item, but its point of origin is not known. The marine shell items are clearly the products of exchange. Fifteen of the reported 21 marine shell specimens were found at 14ML16, the mortuary site. They include the gorget (89), a conch columella (90), and three different varieties of beads. The other six specimens, all beads, came from three of the 15 excavated or tested habitation sites. The strong association with mortuary ritual is typical of Kansas sites during the Early and Middle Ceramic periods (Blakeslee 1997). The precise nature of the exchange system that brought these items from distant oceans to the center of the continent is not known.

One isolated skull from 14ML15 and a modified skull fragment from 14ML5 suggest trophy taking. They are part of a relatively sparse but consistent pattern within the Central Plains Mosaic (Blakeslee 1999: 151-152) that indicates that inter-community raiding was fairly common.

The Central Plains Mosaic appears in numerous localities shortly after A.D. 1000. This statement is contra Roper (1995) who maps dates in various localities by century. The localities with the earliest dates are those in which the largest number of dates have been run, however, and a larger spread of dates is expected as the number of dates increases (see Radiocarbon dating stochastic effects on this web page).
There are Late Woodland sites just to the west of the Glen Elder locality, and the people who created them may have been in place when the shift in subsistence that generated the Central Plains Mosaic occurred. These sites have yet to be investigated. Krause (1995) has discussed the changes in ceramics that may have been involved.

The Glen Elder locality seems to have been abandoned in the period A.D. 1250-1300. This was part of a regional pattern that terminated most of the Central Plains Mosaic except perhaps along the northern fringes (Blakeslee 1997). In the same general time frame there is a movement of Central Plains populations into South Dakota to form the Initial Coalescent variant. Competing hypotheses regarding the cause of this movement include the intrusion of Oneota-derived populations (Ritterbush 2006:163-164), drought (e.g., Bryson et al. 1970), or over-exploitation of the environment (Blakeslee 1993). The hypothesis that some Upper Republican populations migrated south to form the Panhandle Aspect (Baerreis and Bryson 1966) is no longer tenable.


Baerreis, David A., and Reid A. Bryson
1966 Dating the Panhandle Aspect Cultures. Bulletin of the Oklahoma Archaeological Society 14:105-114.

Blakeslee, Donald J.
1993 Modeling the Abandonment of the Central Plains: Radiocarbon Dates and the Origin of the Initial Coaklescent. Plains Anthropologist Memoir 27: 199-214.

1994 Reassessment of Some Radiocarbon Dates from the Central Plains. Plains Anthropologist, 39 (148): 203-208.

1997 Assessing Sets of Radiocarbon Dates: An Example from the Central Plains. North American Archaeologist 5(1):3-9.

1999 Waconda Lake: Prehistoric Swidden-Foragers in the Central Plains. Central Plains Archaeology, 7(1).

2002 Fractal Archaeology: Intra-generational Cycles and the Matter of Scale, an Example from the Central Plains. In The Archaeology of Tribal Societies, edited by William Parkinson, pp. 173-199. Monographs in World Prehistory, Ann Arbor, MI..

Central Plains Earthlodges as the Products of Craft Specialists. In Earthlodges: Architecture and Symbolism, edited by Donna Roper. University of Alabama Press.

2005 Middle Ceramic Earthlodges as the Products of Craft Traditions. In D. C. Roper and E. P. Pauls (eds.), Plains Earthlodges: Ethnographic and Archaeological Perspectives, pp.83-110. Tuscaloosa: University of Alabama Press.

Blakeslee, Donald J., John Hotopp, Kerry Lippincott, John Ludwickson, and Thomas Witty
1982 Some Responses to Krause. Plains Anthropologist 27(95):83-90.

Bryson, Reid A., David A. Baerreis and Wayne M. Wendland
1970 The Character of Late-Glacial and Post-Glacial Climatic Changes. In Wakefield Dort, Jr., and K. Knox Jones, Jr., (eds.) Pleistocene and Recent Environments of the Central Great Plains, pp. 53-74. Lawrence: University Press of Kansas.

Dorsey, Ron
1998 Freshwater Mussels and Archaeological Sites on the Solomon River. M.A. thesis, Department of Anthropology, Wichita State University.
2000 Archaeological Interpretation of Freshwater Mussel Assemblages near the Solomon River, Kansas. Central Plains Archeology 8(1):13-24.

Falk, Carl R.
1969 The Faunal Evidenve. In Wood, W. R. (ed), Two House Sites in the Central Plains, p. 102. Plains Anthropologist Memoir 6.

Krause, Richard A.
1969 Correlation of Phases in Central Plains Prehistory. In W. R. Wood (ed.) Two House Sites in the Central Plains: An Experiment in Archaeology. Plains Anthropologist Memoir 6.

1982 The Central Plains Tradition Revisited: A Critical Review of Recent Interpretations. Plains Anthropologist 27(95):281-292.

1995 Attributes, Modes and Tenth Century Potting in North Central kansas. Plains Anthropologist 40(154):307-352.

Latham, Mark
1996 Sampling Strategies and the Site Distribution of the Solomon River Phase of the Glen Elder Locality of North-Central Kansas. M.A. thesis, Department of Anthropology, Wichita State University.

Lippincott, Kerry A.
1976 Settlement Ecology of the Solomon River Upper Republican Sites in North Central Kansas. Ph.D. dissertation, Department of Anthropology, University of Missouri, Columbia.

1978 Solomon River Upper Republican Settlement Ecology. In Donald J. Blakeslee (ed.) The Central Plains Tradition: Internal Development and External Relationships, pp. 81-93. Office of the State Archaeologist Report 11. Iowa City.

Morey, Darcy F.
1982 A Study of Subsistence and Seasonality in the Central Plains. technical Report 82-12. Department of Anthropology, University of Nebraska.

Ritterbush, Lauren
2006 Late Prehistoric Oneota in the Central Plains. In R. J. Hoard and W. E. Banks (eds.) Kansas Archaeology, pp. 151-164. Lawrence: University Press of Kansas.

Roper, Donna C.
1995 Spatial Dynamics and Historical Process in the Central Plains Tradition. Plains Anthropologist 40(153):203-221.