New Data on the Stratigraphy and Pedology of the Clovis and

QUATERNARY
RESEARCH
23, 388-402 (1985)
New Data on the Stratigraphy and Pedology of the Clovis and
Plainview Sites, Southern High Plains
VANCE T. HOLLIDAY
Received February 24? I984
The well known Clovis and Plainview archaeological sites of New Mexico and Texas have
yielded new data on regional late Quaternary geologic. paleoclimalic, and pedologic histories.
EoIian sedimentation at the Clovis site from about lU.000 to less than 8500 yr B.P was followed
by the formation of a cutn~lic soil between 8500 and 5000 yr B.P Episodic eolian and slope wash
deposition then culminated in massive eolian sedimentauon about 5000 yr B.P. after which a
Haplustalf formed then ~2s subsequently buried by part of a dune system within the last IO00 yr.
At the Plainview site. a basal stream gravel contains Plainview cultural material tea. IO.000 yr
B.P.), which is followed by a localized early Holocene iacustrine deposit. two eolian deposits (the
younger dating to about 5000 yr B.P.), and a marsh deposit which slowly accreted as an Argiustoh
formed in the younger eolian unit. The data indicate that on the Southern High Plains t I J between
12,000 and 8500 yr B.P. sedimentation varied from site to site. (2) there was a regional climate
change toward warming and drying in the early Holocene. (3) two episodes of severe drought
apparently occurred in the middle Holocene (6500 to 4500 yr B.P.). (4) between 4500 yr B.P. and
the present an essentially modern climate existed. but with several shifts toward aridity within the
last 1000 yr, (5) argillic horizons have developed in late Holocene soils. (6) clay illuviation can
occur in calcareous soils, and (7) long-distance corretation of Holocene stratigraphy in the region
is possible. particularly with the aid of soil morphology.
,< I985 Univervty of W&hingw
ations during the Holocene
(Holhday,
1982a). The sites also commonly contain
buried and nonburied soils that formed in
the Holocene and allow investigation of the
rates of development
of Holocene soils.
The signiticance or complexities of these
soils have not always been recognized by
previous investigators and the nature and
rates of pedogenesis of soils formed in the
Holocene are known only from a few specific sites such as Lubbock Lake (Holliday,
1982b, 1983) and one locality in the Sandhills region (Gile, 1979). An understanding
of Holocene pedogenesis in the region can
aid in making stratigraphic correlations and
in dating Holocene sediments.
In this paper new sedimentological, geochronological, and pedological information
from the Clovis and Plainview sites (Fig. 1)
is assessed. The new data, combined with
information
previousiy pubished, are significant for reconstructing the late Quater-
INTRODUCTION
The Southern High Plains of northwest
Texas and eastern New Mexico contain
some of the best known Paleo-Indian sites
in North America. Many are the type sites
for Paleo-Indian projectile points and most
have thick and multiple strata.
Geologic research has recently been carried out at some of these sites for several
reasons. Most of the previous investigations focused on the stratigraphy and depositional environments
of the Paleo-Indian-age deposits (greater than about 8000
yr B.P.) but these sites commonly contain
younger sediments and soils. The paleoenvironments of the Southern High Plains for
the past 8000 yr are poorly known, although
a few sites, such as Lubbock Lake (HOIliday er ul., 1983), have provided information on this period, and there are indications of significant, regional climatic vari388
0033-5894/85 $3.00
HIGH
PLAINS
PALEO-INDIAN
nary paleoenvironments
of the localities
and for building a regional late Quaternary
history for the Southern High Plains.
Several subdivisions of the late Quaternary of the Southern High Plains are used
in this paper. The late Quaternary, as used
here, is divided into the latest Pleistocene
(12,000 to 10,000 yr B.P.) and the Holocene
(10,000 yr B.P. to the present). The Holocene is subdivided into early (10,000 to 6500
yr B.P.), middle (6500 to 4500 yr B.P.), and
late (4500 yr B.P. to present).
REGIONAL
SETTING
The Southern High Plains, or Llano Estacado, is an extensive plateau covering
about 130,000 km?. The area is an almost
featureless plain. Thousands of small lake
basins provide slight topographic relief in
addition to several northwest- to southeasttrending drainages (with no permanent running water) and several large dune fields.
The flat surface of the region is constructional, formed by deposition of extensive
sheets of eolian sediment during the Pleistocene (Blackwater
Draw Formation)
(Fenneman,
1931: Hawley et al., 1976:
Reeves, 1976).
The relatively uniform climate of the
Llano Estacado is a dry, midlatitude, semidesert (Strahler and Strahler, 1983, Plate
C2). Mean annual precipitation for Clovis
(1305 m; 4280 ft) is 42.3 cm (16.67 in.)
(NOAA, 1982a): for Plainview (1027 m;
3370 ft) it is 48.1 cm (18.93 in.) (NOAA,
1982b). However, interannual variability is
high.
METHODS
Available exposures at the Clovis and
Plainview sites were examined and one profile at each site was described and measured (Table 1). Samples from each profile
were analyzed for particle-size distribution,
sorting characteristics, CaC03 and organiccarbon content, and bulk density (Table 2).
Variations in parent material were detected
by calculating fine sand/very fine sand and
very find sand/silt ratios (Evans, 1978, p.
389
SITES
363) and an “index of similarity” (Buol et
ul., 1980, p. 14). Because variations
in
parent material were detected in many profiles, clay movement could be measured
only between horizons with similar parent
material. For any horizon that exhibited
clay illuviation (suggested by this procedure), an examination of a thin section was
used to confirm the presence of such clay.
However, this approach could determine
only minimal amounts of translocated clay.
Clay minerals were identified for selected
horizons using X-ray diffraction analysis.
Thin sections were also prepared for selected horizons and described following the
terminology of Brewer (1976).
THE CLOVIS SITE
Setting and Previous
Work
The Clovis site (Roosevelt County, New
Mexico), also known as Blackwater Draw
locality 1, is one of the best known archaeological sites in North America. Its fame
rests primarily on the abundant Paleo-Indian material it contains. In particular, it is
the type site for the Clovis point and was
the first site where the stratigraphic relationship of Clovis and Folsom cultures was
documented (Wormington, 1957). The site,
located between Clovis and Portales, lies in
a small basin immediately
north of and
draining into Blackwater Draw (Fig. 2), a
tributary of the Brazes River (Fig. 1).
The site was discovered during gravel
mining operations. Hester (1972) has summarized all archaeological
work through
1963, and Stevens (1973) has done so for
1963 through 1972. Mining has now destroyed most of the site (Fig. 2); however,
a portion of the southwest part of the site
remains (the “South Bank” of Stevens,
1973, and Agogino et ul., 1976). The South
Bank lies along the basin margin (Fig. 2) at
about the interface between the basin sediments and dune sands on the uplands (part
of the Sandhills, an east-west-trending
belt
of dunes that extends far into the Texas
Panhandle; Figs. 1 and 3). The section ex-
390
VANCE
TABLE
Unit
G
Depth
(cm)
Horizon
A
,-
E
5
4
3
1
FIELD
DESCRIPTIONS
Munsell
DRY
OF
THE CLOVIS
Moist
Clovis
1.5YR 414
7.5YR 514
Texture
siteb
fs
t-s
1OYR 513
1.5YR Sl4
Abl
Btbl
Alb2
A2b2
A3lBwb2
48-59
59-81
81-91
91-118
118-138
7.5YR 414
7.SYR 514
7.5YR 414
1OYR 513
1OYR 513
7.5YR 414
7.5YR 415
7.SYR 414
7.5YR 414
1OYR 414
to 7.SYR 414
2Ab3
2Aklb3
3Ak2b3
138-148
148- 185
185-195
IOYR 513
7.5YR 612
1OYR 612
IOYR 313
1OYR 3/3
1OYR 3/2
Plainview
Fill
Al
A2
A3
2ABkl
2ABk2lAb
2Btklbl
O-26
26-35
35-55
55-68
68-82
82- 102
102-130
1OYR 4/3
1OYR 412
?.5YR 412
1OYR 412
7.5YR 412
1OYR 314
1OYR 212
1OYR 212
1OYR 212
1OYR 2/2
7.5YR 3/2
7,SYR 312
CL
CL
CL
CL
CL
CL
3Btk2bl
3Btk3bl
3BCbl
130-145
145-200
2OG228
1OYR 413
1OYR 5/3
1OYR 5/3
7.5YR 314
7.5YR 314
7SYR 414
4Clbl
228-250
1OYR 7/2
4C2bl
250-274
1OYR 613
5C3bl
(no sample)
274-276
6K2b2
276 +
1
AND PLAINVIEW
SITES’
color
O-20
20-48
L
F
I.
T. HOLLIDAY
IOYR 513
to 7.5YR 414
1OYR 513
to 7.5YR 414
structure
DRY
consistence
m
m
sh
L fS
LtX
fSL
fSL
fSL
lcsbk
lcsbk
lmsbk
lmsbk
lmsbk
fSL
SCL
L
site<
lmsbk
m
m
C&Ox
Boundary
non
non
Ch
sh
h
sh
sh
sh
non
cs
non
non
non
CW
sh
h
h
e
cw
t?S
CW
es
lmsbk
2msbk
2msbk
2msbk
2msbk
lmpr & 2csbk
vh
vh
vh
vh
vh
vh
e
es
es
es
es
es
cw
SCL
SCL
SCL
lmpr & lcsbk
2mpr & lmsbk
lmpr & lcsbk
vh
h
h
ev
es
es
cs
SL
m
sh
f2S
cs
LS
m
sh
es
aw
IO
& lcpr
& lcpr
& lcpr
& lcpr
CW
CS
w
CW
CW
CW
cs
CS
cs
CS
cs
aw
Not described
‘Abbreviations
used: Texture-f
= fine; S = sand: LS = loamy sand: SCL = sandy clay loam; SL = sandy loam;
loam; CL = clay loam; g = gravelly. Structure-Grade:
1 = weak; 2 = moderate: Class: m = medium; c = coarse:
sbk = subangular blocky; pr = prismatic:
m = massive. Dry consistence-lo
= loose; sh = slightly hard: h = hard;
very hard. CaCOs (reaction to dilute HCl)-e
= moderately
efferevescent:
es = strongly efferevescent:
ev = vitilently
vescent: non = noncalcareous.
Boundary-Distinctness:
g = gradual: c = clear; a = abrupt; Topography:
s = smooth:
L =
Type:
vh =
efferw =
WWY.
’ South Bank, Roosevelt County, New Mexico.
c Protile 1, Pit 2, Hale County, Texas.
amined as part of this study is at the South
Bank (Fig. 2).
Haynes and Agogino (1966) and Haynes
(1975) present the most recent and comprehensive review of the stratigraphy of the
Clovis site (Fig. 3) and discuss most of the
reported radiocarbon ages. Hester (1972)
provides a brief review of all previously reported radiocarbon ages from the site. The
following summary of the stratigraphy and
geochronology of the site follows the descriptions and radiocarbon ages of these investigators and the terminology of Haynes
(1975).
Extensive erosion occurred in the basin
prior to about 11,000 yr B.P. Deposition of
spring sediments (Unit C) occurred during
Clovis time (I 1,500 to 11,000 yr B.P.) and
was followed by deposition of diatomaceous lacustrine sediments (Unit D) from
I 1,000 to 10,000 yr BP Some soil formation in and erosion of Unit D occurred
about 10,000 yr B.P. Silty, carbonaceous lacustrine sediments were deposited in the
basin, along with the accumulation
of
sandy, eolian sediments along the basin
margin (Unit E), from about 10,000 to
sometime after 8500 yr B.P Soil formation
and extensive erosion began sometime after
8500 yr B.P. and ended prior to about 5000
HIGH
TABLE
PLAINS
2. LABORATORY
Particle-size
PALEO-INDIAN
DATA
FOR THE CLOVIS
distribution
Unit
Horizon
Coarse
Medium
AND PLAINVIEW
SITES
(%)
Sand
very
coarse
391
SITES
Total”
76
OrganG
carbon
Fine
Very
tine
Sand
Silt
Clay
site’
94.6
92.4
3.0
4.1
2.6
3.5
0.2
0.2
%
CaCO:
Bulkd
density
&/cm’)
-
-
0.80
0.65
1.74
1.70
1.72
I.17
1.70
0.80
I.05
I .20
I.00
I.13
I.12
I .oo
1.00
Sorting’
-
2.1
1.0
14.7
II.0
55.2
58.5
Clovis
22.6
21.9
Abl
Btbl
AIb2
A2b2
A3lBwb
-
3.2
6.3
6.0
6.4
1I.3
IO.1
14.7
15.2
17.2
24.8
45.3
37.3
30.8
26.0
29. I
25.7
18.7
19.7
13.2
II.9
84.3
77.0
71.7
62.8
77.1
9.8
13.2
19.3
25.2
14.1
5.9
9.8
9.0
12.0
8.8
0.2
0.2
0.1
0.1
0.1
2Ab3
2Aklb3
3Ak2b3
I.1
0.8
-
13.1
IO.1
5.4
15.7
12.0
6.7
36.9
36.8
25.6
l0.Y
14.0
15.7
17.7
73.7
54.4
14.0
17.3
40.0
8.3
9.0
5.6
0.1
0.2
0.2
0.4
6.3
IO.3
I .75
I .56
I.28
Al
A2
A3
2ABkl
2ABk2/Abl
2Btklbl
-
0.5
0.2
0.3
0.8
4.4
I.4
I.1
2.1
2.9
4.0
Il.6
5.5
4.8
1.3
8.6
9.0
Plainview
17.2
16.8
14.4
7.6
7.8
9.1
siteR
33.7
23.7
20.3
17.2
19.6
22.9
34.4
35.7
34.1
34.5
36.6
30.6
31.9
41.1
45.6
48.3
43.8
46.5
I.3
1.2
0.9
0.9
I.3
0.5
0.3
4.1
4.6
2.6
7.2
IO.4
1.71
1.63
1.71
I.60
I s7
Il.51
3BtkZbl
3B3tkbl
3BCbl
-
0.8
0.8
0.8
9.5
9.9
9.7
30.0
30.7
33.9
22.7
23.5
25.1
63.0
64.9
70.1
IS.3
20.3
17.4
18.7
14.8
12.5
0.6
0.5
0.3
2.7
5.1
6.5
I.80
1.61
1.56
0.90
0.90
0.90
4Clbl
4C2b I
-
0.8
1.2
9.1
43.0
44.7
32.0
30.5
85.2
85.1
9.1
8.2
5.7
6.7
0.4
0.3
8.6
6.2
1.60
I.53
0.75
0.75
-
-
’ Silt and clay by pipet (Day. 1965) on organic matter- and CaCOJ-free
basiq: sands sepamted by wet sieving
’ Walkley-Black
technique (Allison,
1965).
’ By Chittick apparatus as described by Dreimanis (1962) and modified by Bachman and Machettc (1977).
” Clod method (Blake. 1965~.
’ Sorting characteristics
following Inman 11952) and Folk (1974).
’ South Bank.
x Prolile I? Pit 2.
yr B.P. Eolian sand (Unit F) was deposited
about 5000 yr B.P. A soil formed in Unit F
in the late Holocene and was followed by
deposition of more eolian sand (Unit G) and
development of a weak soil in that sand.
The work at the South Bank was in upper
Unit E and Units F and G (Fig. 3). Upper
Unit E has a thick, cumulic A horizon
whose texture varies from fine sandy loam
to loam (Table l), suggesting that the surface of the unit was slowly aggrading with
variations in sedimentation. Unit E in the
South Bank is moderately to poorly sorted
(Table 21. This, and the proximity of the
section to the edge of the basin, suggests
that reworking,
by slope wash, of the
valley-margin eolian facies of Unit E and
-
then dry sieved.
probably eolian sedimentation were part of
the aggradation process. Unit E is also calcareous with a distinct zone of secondary
CaC03 in the Aklb3 and Ak2b3 horizons.
A radiocarbon sample from the top of the
A horizon in upper Unit E yielded an age
of 4855 L 90 yr B.P. (SI-4.585; NaOH-insoluble fraction). This suggests that pedogenesis in and some erosion of Unit E occurred between 8500 and 5000 yr B.P. and
that Unit F was deposited around 5000 yr
B.P. This is also indicated by Haynes (1975,
p. 64), who notes a ‘.strongly developed pedocalcic paleosol”
in Unit E and that
“along the north rim this paleosol contains
the strongest development of calcium carbonate of any observed in the younger
group of sediments (Units B to G).” The
soil in Unit E at the South Bank is not as
392
.I.-
32’
OwOkm
0
w
50
’
rni
FIG. 1. The Llano Estacado (Southern High Plaim) showing the locations of the (Zlovis site
I ). the Plainvl iew site, and other localities and physiographic features referred to in fhe text.
well developed because of cumulization of
the profile.
The transition from Unit E to F is marked
by a weak cumulic soil, indicating a continuation of the episodic sedimentation of Unit
E. There is a clay bulge in the A2b2 horizon
(Fig. 5), but its position within the A horizon and the lack of field evidence for a B
horizon indicates that the clay may not be
illuvial. Alternatively,
it may be that dustderived clay was mechanically infiltrated
into the soil as the surface aggraded, then
clay translocation was halted upon burial.
WD-
This process occurs elsewhere in the region
(Holliday, 1982b). As in Unit E, the sorting
of the sand fraction in the transition zone
is moderate to poor, suggestive of slope
wash additions (Table 2).
Unit F is considered to be principally
eolian, as interpreted by Haynes (1975), although it is not particularly
well sorted
(Table 2). An eolian origin is indicated by
the position of the South Bank profile high
on the valley wall, the absence of coarse
elastics in the South Bank, or any exposures described by Haynes (19X), and the
HIGH PLAINS
PALEO-INDIAN
SITES
393
FIG. 2. Map of the Clovis site showing the topography of the original (premining) basin. approximate extent of the gravel quarry, and the location of the South Bank section (contours approximate).
The elongate, topographic low in the south-central portion of the map is the upper end of the drainageway into Blackwater Draw (ca. 2 km south of the South Bank). The inset shows the location of
the site relative to the draw and the city of Portales- New Mexico. Based on Evans (1951, Fig. 1).
Hester (1972. Fig. 15). and Haynes (1975. Fig. 4-2).
units blanket-like distribution mantling the
lowlands and uplands, based on Haynes
(1975). There were probably some slope
wash additions, again a function of the
proximity of the profile to the basin margin.
Lower Unit F has produced a radiocarbon age of 4950 2 150 yr B.P. (O-lS7;
Brannon et a/., 1957; Haynes, 1968, 1975).
Haynes (1968, p. 621) suggests that this age
may be too young because of the nature of
the material dated, which was charred
bison bone. However, the new radiocarbon
age from the top of Unit E indicates that O157 is reliable.
Unit F exhibits the best expressed buried
soil (bl) at the South Bank and it is classified as a Haplustalf. The soil has an ochric
epipedon and structure in the B horizon is
weakly developed. Haynes (1975, pp. 6465) suggests stronger structural development in the soil in other exposures at the
site and mentions that “At most exposures
394
VANCE
--0
T. HOLLIDAY
IZOrn
“.e c-3 15x
FIG. 3. Stratigraphy of the Clovis site. Top. Sketch of the section described at the South Bank,
including locations of new radiocarbon ages from the mid-Holocene units reported in the text and
correlation of soils and stratigraphic units with previous investigations. Bottom, Generalized northsouth geologic cross section of the site, based on Haynes (l975* Figs. 4-3. 4-4, 4-7: 1983, personal
communication).
this sand, Unit F, contains a coarse, irregular, blocky structure that is responsible for
its being referred to as the ‘jointed sands’ ”
and “Along the north rim . . . the soil profile developed in Unit F is more strongly
developed than elsewhere at the site. Here
it contains a brownish-red zone grading into
a reddish-brown zone of fme to medium,
irregular, blocky structure in the lower half.
The unit is weakly cemented by secondary
carbonate. . . .”
Several thin sections were taken from the
bl soil. The Btbl horizon has argillasepic
fabric with common free grain argillans and
occasional embedded grain argillans; outer
boundaries of argillans are usually sharp.
Distribution is intertextic to granular, with
very common interconnected vugs (simple
packing voids). Orientation is flecked to
moderately striated (Fig. 4). The thin sections and particle-size distribution (Fig. 5)
demonstrate that sufficient illuvial clay is
present to qualify the B horizons as argillic.
In the South Bank section the bl soil and
most of the section is noncalcareous. Clay
minerals are illite, some mixed-layer illitesmectite clays, and minor amounts of kaolinite. The sand fractions of the soil are
almost entirely quartz.
Unit G overlies Unit F, and is a noncalcareous sand representing dune sediments
of the Sandhills. At some localities around
the Clovis site, Haynes (1975) subdivided
Unit G into Gl (older) and G2 (younger),
based on stratigraphic,
sedimentological,
and pedological
characteristics.
At the
South Bank exposure only one deposit was
apparent.
HIGH
PLAINS
PALEO-INDIAN
SITES
395
FIG. 4. Photomicrographs
of thin sections
(under
crossed
Nicois)
from the Clovis
(a, Btbl)
and
Plainview
sites (b, 3Btk3bl).
Free- and embedded-grain
argillans
(e.g., arrows)
are common
and
distinct,
coating
quartz
grains in (a). Embedded
grain argillans
(e.g., arrows)
are common.
though
less distinct
in (b), as is some secondary
CaCOJ (SC). The bar is IO brn.
The soil formed in Unit G is the local
surface soil. It is weakly developed with an
ochric epipedon and no diagnostic subsurface horizons. There is some suggestion of
translocated clay in the C horizon (Fig. 5),
but the difference in clay content between
the A and C horizons may be less than the
analytical error. The soil in Unit G is classified as an Ustipsamment.
Stratigraphy, pedology, and the new ra-
diocarbon age, discussed above and those
discussed by Haynes (1975) and Hester
(1972), suggest that Units E and F represent
a period of episodic eolian deposition. This
period began perhaps 10,000 yr B.P. and
terminated by 4500 to 4000 yr B.P.
Soil formation in Unit F and deposition
of and soil formation in Unit G are poorly
dated. A radiocarbon
sample was taken
from the top of the buried A horizon in Unit
VANCE
396
P/oinview
% CIOY
Clovis
% c/w
0
0
r
T, HOLLIDAY
i
A
!I
C
2 50
:
Abl
Etbl
2
Fl
AlbZ
100
A2b2
I
A3/Bwb2
3BCbl
FIG. 5. Illuvial clay content for the surface, bl, and
b2 soils at the Clovis site and the entire section at the
Plainview site, calculated using procedures discussed
in the text.
F, but the resulting age was “modern” (SI4584). This may be due to the small sample
size after pretreatment and because overlying Unit G is quite permeable and may
have allowed contamination
by throughflowing water. It may also be that the age
of burial was quite recent and not within
the resolving power of the radiocarbon
technique.
Haynes (1975, Table 4-3) believed that Unit F was bured by Unit G
around 4000 yr B.I?, but the evidence was
not discussed. Very late prehistoric and historic archaeological
material recovered
from Unit G suggests that deposition took
place quite recently, probably within the
last 500 yr or less, although the exact provenience of the artifacts in the deposits is
unknown (F. M. Nials, 1980, personal communication). This interpretation of the age
of Unit G is also supported by the welldeveloped nature of the Unit F soil and the
very weakly developed soil in Unit G.
THE PLAINVIEW
Setting and Previous
SITE
Work
The Plainview site is a significant archaeological locality within the city limits of
Plainview, Hale County, Texas (Fig. I), in
Running Water Draw. The site was discovered during caliche quarrying along the
south margin of the draw and consisted of
an extensive bone bed of extinct bison
(Bison anticpus) associated with lithic material (Sellards et al., 1947; Guffee, 1979).
The locality is the type site for the Plainview point. Continued quarrying resulted in
the excavation of three deep pits (Fig. 6).
The pit with the bone bed (Pit 1, Fig. 6) is
now tilled but the other two pits provide
long, continuous exposures of the fill in
Running Water Draw. The section discussed in this paper (Fig. 7) is located toward the center of the valley (Fig. 6) and is
representative
of the valley-axis stratigraphy.
Running Water Draw has cut through the
Blackwater Draw Formation and onto the
Pliocene Ogallala Formation
(the Panhandle Formation of Sellards et af., 1947).
Bedded carbonate gravel and sand (Unit I),
varying from a few centimeters to more
than 1 m thick, comprise the basal late Quaternary fill. The Plainview bone bed rested
on top of Unit I. Discussion of the site stratigraphy by Sellards et al. (1947) and Guffee
(1979) and an examination of blocks from
the bone bed indicate that the kill took
place in an abandoned meander channel
and the bones were quickly buried by sandy
loam fluvial deposits. The age of Plainview
points is about 10,000 yr B.P. (Johnson and
Holliday,
1980: Holliday
and Johnson
1981), providing an age for the end of Unit
1 deposition.
New Results
In the described section (Table l), Unit 1
is very thin and the sandy loam that buried
the bone bed is missing. An organic-rich,
lacustrine clay (Unit 2) overlies Unit 1 in
some exposures. Unit 2 is not present at
the described section. There, Unit 1 is
buried by almost 1.5 m of sandy sediment
(Units 3 and 4). The lower 50 cm (Unit 3)
and the upper 1 m (Unit 4) are both sandy
loams, but the particle-size data indicate a
HIGH PLAINS
PALEO-INDIAN
SITES
397
FIG. 6. Geologic sketch map of the area of the Plainview site in Running Water Draw (no adequate
topographic base is available) showing the locations of the cahche pit that contains the archaeological
remains (Pit I. now filled in)+ the present-day caliche pits (2 and 3), the profile studied as part of this
investigation. and the cross section for Figure 7 (A-A’). Based on Sellards er (I/. (1947. Fig. 2) and
Blakely and Koos (1974. sheets 19, 26, and 27).
subtle lithologic break between the two
units (Table 2). In other exposures, the Unit
3 sandy loam interfaces with a highly calcareous, apparently lacustrine deposit of
marl. Unit 3, especially the marl facies, occasionally exhibits a weak soil at the top.
The sand fraction of Units 3 and 4 is only
moderately sorted (Table 2), which is not
particularly indicative of eohan sedimentation. An eolian origin is indicated by the
massive nature of both units throughout the
area of the site, the complete absence of
coarse clasts, and the blanket-like distri-
bution of both units across the draw. The
lack of sorting of Units 3 and 4 may be due
to a short transportation
distance for the
sediments; they were derived from the
Blackwater Draw Formation immediately
adjacent to the draw.
The eolian sediments are buried by an
organic-rich, clayey, lowland marsh deposit
(Unit 5), which is the present suficial deposit in the draw in the site area. In the
described section Unit 5 exhibits subtle
lithologic variation (Table 2) and in exposures near the valley margin a sandy, eolian
398
VANCE
Sellards
o-
et 01 ,I947
ym
T. HOLLIDAY
T
h!s
Study
“e
co ,5x
FIG. 7. Stratigraphy of the Plainview site. Top. Sketch of the described section at the Plainview
site, including correlation of soils and stratigraphic units with previous investigations and the location
of the radiocarbon sample. Bottom. General north-south geologic cross section of Running Water
Draw through the Plainview site. Based on Sellards ef u/. (1947. Fig. 3).
The micromorphology
of the soil at Plainfacies is apparent. The organic-rich mateview
is
generally
uniform
throughout.
rial from lower Unit 5 yielded a radiocarbon
age of 3325 2 65 yr B.P. W-4820; NaOHFabric is argillasepic with common, thin,
insoluble fraction).
embedded-gran
argillans and occasional
free-grain
argillans,
which usually have
A well-developed soil formed in Units 4
sharp
outer
boundaries.
Distribution is agand 5. The surface horizons qualify as
glomeroplasmic
to
intertextic.
Vugs are
pachic, mollic epipedons. The mollic colors
common
and
occasionally
interconnected
extend into the 2Btklbl horizon but the organic carbon content there is below 0.6%. with occasional, thin, calcitans along their
The 2Btklbl horizon is considerably finer sides (Fig. 4b). The fabric in the zone of
secondary carbonate accumulation is probgrained than the lower B and C horizons
is flecked to
and probably represents the initial Unit ably crystic. Orientation
and
5 sediments and original A horizon of weakly striated. The micromorphology
particle-size distribution
(Fig. 5) indicate
the soil. However, the entire B horizon
the presence of significant amounts of illuhas a well-developed structure. Secondary
CaC03 accumulation occurs in the lower A vial silicate clay. The B horizon, then, qualhorizon and upper and middle B horizon as ifies as argillic.
Clay minerals of the Plainview soil are
films, threads. and coatings on ped faces.
and
This zone of CaCOX accumulation is Stage illite and mixed-layer illite-smectite
I of Gile et ul. (1966), but does not qualify
some kaolinite. Sand-size minerals in the
as a calcic horizon (Soil Survey Staff,
soil are almost entirely quartz with some
1975). The soil is classified as an Argiustoll.
feldspars.
HlGH
PLAINS
PALEO-INDIAN
SITES
399
Lacustrine sedimentation
was discontinuous and considerably less extensive where
aliuviation continued until about 10,000 yr
B.P. (Plainview).
Increased eolian sedimentation and precipitation of CaCOj followed noncalcareous lacustrine deposition. At Clovis this
occurred sometime after 8500 yr B.P. At
Plainview the event is not dated, but a similar change in similar stratigraphic position
occurred around 6500 yr B.P. at Lubbock
Lake (Holliday ef ul., 1983). At Lubbock
1983).
Lake, Holliday (in press) suggests that inCONCLUSIONS
creased temperature produced cakareous
The data from the Clovis and Plainview
sediments and reduced effective moisture
sites provide additional and significant in- led to increased eolian deposition. The data
formation to the growing body of data on suggest a regional trend toward warming
the late Quaternary geology, paleoclimates,
and drying.
The middle Holocene records at Clovis
and pedology of the Southern High Plains.
At Clovis, Haynes and Agogino (1966) and and Plainview document two cycles of
Haynes (1975) show that spring discharge eolian activity separated by a brief interval
produced localized alluviation
between
of nondeposition and formation of weakly
[I,500 and 11,000 yr BP Between I 1,000 developed soil. Similar records are reand 10,000 yr B.P. the springs fed ponds,
ported from Lubbock Lake (Holliday er u/. ,
producing diatomaceous
lacustrine sedi- 1983) and the Sandhills east of Clovis (Fig.
ments. At Plainview, however, ahuviation
I) (Gile, 1979, 1983). These correlations
continued until 10,000 yr B.P. These dif- (Fig. 8) suggest two periods of prolonged,
ferent sedimentation histories are probably
severe drought throughout at least the cenrelated to Fluctuation in or the presence or tral Llano Estacado, resulting in loss of soil
absence of spring discharge and the change moisture, reduction in vegetation cover,
in climate toward less effective moisture
and deflation
by wind. A similar twoat about 11,000 yr B.P. documented
drought ‘41tithermal has been suggested by
throughout the southern Great Plains (e.g., Benedict and Olson (1978) and Benedict
Johnson, 1976; Lundelius, 1974; Lundehus
(1979) for the Central Plains and by Gaylord
et al.,
1983). Wind may have reworked al- (1982) for dune fields in south-central Wyluvium on the floors of the draws and oming.
dammed the relatively sinuous and narrow
The fate Holocene at Clovis and Plainreaches of upper Blackwater Draw (with
view is primarily a period of landscape stathe Clovis site) and lower Yellowhouse
bility, also documented
in the Sandhills
Draw (with Lubbock Lake). The broader (Gile 1979, 1983) and at Lubbock Lake
and less sinuous reaches of middle Running
(Holliday et al., 1983). Minor eolian acWater Draw (with the Plainview site) re- tivity occurred at all of these sites within
mained open (Holliday, 1984).
the last 1000 yr (Fig. 8). Because conditions
Lacustrine sediments were deposited at of landscape stability prevail today, the late
Clovis and Plainview in the earJy Holocene.
Holocene climate of the region is interExtensive and continuous iacustrine sedi- preted as having been essentially like that
mentation occurred until at least 8500 yr of today with minor shifts toward aridity
B.P. in settings where alluviation halted at within the past 1000 yr. It is possible that
1I-000 yr B.P. (Ciovis and Lubbock Lake). climatic changes toward increased moisture
The evidence indicates that the profile is
a composite. The A horizon aggraded as
fine-grained lacustrine sediment (Unit Sl
slowly tilled the draw. Pedogenesis in the
underlying, sandier material began at this
time. Eventually the A horizon aggraded to
the point that the B horizon began to develop in the bottom of what was the original
A horizon (102-130 cm). This same phenomenon has been observed along Yellowhouse Draw to the south (IIolliday, 1982b,
400
VANCE
T. HOLLIDAY
Clovis
Bailey
Co. “Sandhills”
Fairvtew
O-
Fq33l-K
warmer,
drier
?
* = 14C do+ed
FIG. 8. Correlation of late Quaternary depositional and pedological events at the Clovis and Plainview sites (as described in this paper) with events at the Lubbock Lake site (Hoiliday cf lt/., 1983)
and the Sandhills (Gibe. 1979. 19831 and an interpretation of general climatic trends for the Southern
High Plains, relative to the modern climate.
occurred (as suggested by Hall (1982) for
the period 2000 to 1000 yr B.P.) but these
events are not evident in the geologic
record.
The pedologic comparisons for the two
sites are between the soil in Unit F at
Clovis and throughout
Units 4 and 5 at
Plainview. Both soils formed at about the
same time and, although there are some
variations in setting and parent material
texture and carbonate content, they both
exhibit similar development.
This is best
seen in the amount of translocated clay in
the B horizon in the thin section and from
laboratory data. This soil development is
comparable to that in middle Holocene
eolian sediments (Stratum 4) at the Lubbock Lake site (Holliday, 1982b).
The co-occurrence of Nuvial clay with
secondary CaCOX at Plainview (2Btk I bl ,
3Btk2b1, and 3Btk3bl horizons, Fig. 7) is
significant. In other situations where this
is documented (e.g., Allen and Goss, 1974),
it has been demonstrated or suggeted that
the argillic horizons formed in decalcified
material and were subsequently recalcified.
There are no data from any locality in the
region that indicate that this process occurred in the late Holocene. On the contrary, data from Lubbock Lake (Holliday,
in press), a site near Amarillo (Fig. 1) (Goss
et al., 1973). and other regions (Aguilar et
al., 1983), clearly demonstrate that clay can
be translocated into a calcareous medium.
The data from the Plainview site further
support these conclusions.
The above information suggests that the
timing and nature of Holocene deposition
on the Southern High Plains were similar.
This allows for making long-distance correlations through the area with some degree
of confidence. Additionally, the data demonstrate that soil morphology (macro- and
micro-) is a useful tool for correlating Holocene sediments in the region and that
many of the sites can provide significant
information
concerning the formation of
Holocene soils in a semiarid environment.
HIGH PLAINS
PALEO-INDIAN
SITES
401
study of man. mountains, and the two-drought Altitherma], Soufh\~~estern Lore 45(3). I- 12.
Special thanks to Robert Stuckenrath (Smithsonian
Benedict, J. B., and Olson. B. L. (1978). “The Mount
Institution) for dating the radiocarbon samples from
Albion Complex: A Study of Prehistoric Man and
the sites. Access to and orientation at the Clovis site
the Altithermal:
Research Report No. 1. Center for
was kindly provided by Fred Nials (Eastern New
Mountain Archaeology, Ward, Cola.
Mexico University) and by Eddie Guffee (Wayland
Blake. G, R. (1965). Bulk density. in .-Methods of Soil
Baptist University) and Roberta Speer (West Texas
Analysis’. (C. A, Black. Ed.). Mo~7ogrup~7 Series 9,
St&e University) for the Plainview site. Guffee also
pp. 374-390. American Society of Agronomy. Madcollected the radiocarbon
sample from Plainview.
ison, Wise.
B. L. Allen (Texas Tech University) assisted in de- Blakely, E. R,, and Koos. W. M. (1974). “Soil Survey
scribing the sections. Daniel R. Muhs (University of
of Hale County, Texas.*’ U.S. Department of AgriWisconsin) and Richard G. Reider (University of Wyculture. Soil Conservation Service.
oming) provided very helpful comments on the manu- Brannon- H. R., Jr., Daughtry. A. C., Perry. D.. Simscript and Peter W. Birkeland (University of Colamons. L. H.. Whitaker. W. W., and Williams. M.
rado). B. L, Allen, and C. Vance Haynes. Jr. (Uni(1957). Humble Oil radiocarbon dates. I. .Sc;er7ce
versity of Arizona), reviewed various parts and earlier
~wad7;77gfo~7,
D.C.) 125, 147-150.
versions of the manuscript. Haynes also kindly shared Brewer, R. (1976). “Fabric and Mineral Analysis of
his considerable knowledge of the Clovis site and area.
Soils” (reprint with supplement). Krieger. New
Eileen Johnson (Texas Tech University) has been a
York.
source of considerable support and encouragement for Buol. S. W.. Hole. F. D.? and McCracken, R. J.
this research. The Cartography laboratory of the Uni(1980). “Soil Genesis and Classification.”
Iowa
versity of Wisconsin aided in preparation of the figState Univ. Press, Ames.
ures. Sally Monogue (University of Wisconsin) typed Day. P. R. (1965). Particle fractionation and particlethe manuscript. This investigation is part of the consize analysis. In **Methods of Soil Analysis” (C. A.
tinuing research of the Lubbock Lake Project (The
Black, Ed.). Mor7ogtxp~7
Series 9, pp. 545-567.
Museum, Texas Tech University)
funded by the
American Smiety of Agronomy. Madison, Wise.
Nationai
Science
Foundation
(SOC75-14857:
Dreimdnis, A, (1962). Quantitative determination of
BNS7612006;
BNS76l2006-Aol:
BNS78-lll5S):
calcite and dolomite by using the Chittick apparatus.
Texas Tech University: The Museum of Texas Tech
Jo1w77~7~ qf Sedimentmy Petrdog)~ 32, 520-529.
University, the Department of Geological Science,
Evans. G. L. (1951). Prehistoric wells in eastern New
University of Colorado: and the Department of GeMexico. American Anriquiry 17, l-8.
ography. Universitq of Wisconsin.
Evans, L. J. (1978). Quantification
and pedological
process. ltr “Quaternary
Soils*’ (W. C. Mahaney.
F?EFERENCES
Ed.), pp. 361-378. Geo Abstracts. Norwich.
Agogino. G. A.> Patterson, D. K., and Patterson,
Fenneman, N. M. (1931). “Physiography
of the
D. E. (1976). Blackwater Draw locality No. 1. south
Western United States.” McGraw-Hill,
New York.
bank, report for the summer of 1975. p/&s A77fi7roFolk, R. L. (1974). “Petrology
of Sedimentary
poiogixi 21, 213-223.
Rocks.” Hemphill* Austin, Tex.
Aguilar. J.. Guardiola. J. L.. Barahona, E.. DorronGaylord, D. R. (1982). Geologic history of the Ferris
bore. C.. and Santos, F. (1983). Clay iliuviation in
Dune Field, south-central Wyoming. fn ‘*Interprecalcareous soils. [?I .‘Soil Micromorphology.”
Vol.
tations of Windflow Characteristics from Eolian
2. .&Soil Genesis” (P. Bullock and C. P. Murphy,
Landforms” (R. W. Marrs and K. E. Kolm, Eds.).
Eds.), pp. S41-5%. A. B. Academic. Berkhamsted.
pp. 56-82. Gedogical Society oj’ Ameriu~, Sped/
Herts.
Paper 192.
Allen. B. L., and Goss, D. W. (1974). MicromorGile. L. (1979). Holocene soils in eolian sediments of
phology of paleosoils from the semiarid Southern
Bailey County, Texas. Soil Science So~iezy o,f
High Plains of Texas. fti “Soil Microscopy” (G. K.
Americu .kmrrd 43, 994- 1003.
Rutherford. Ed.). pp. 511-525. Limestone Press. Gile, L. (1983). Holocene soils exposed in a Sandhills
Kingston, Ontario.
blowout on the Southern High Plains. In “GuideAllison. L. E. (1965). Organic carbon. lrz “Methods
book to the Central Llano Estacado” (V. T. Holof Soil Analysis” (C. A. Black, Ed.), Monogrup/z
liday. Ed.), pp. 135-165. Friends of the Pleistocene,
S~,r~~,.\9, pp. 1367-1378. American Society of
South-Central Cell Field Trip, ICASALS and The
Agronomy, Madison, Wise.
Museum, Texas Tech University. Lubbock.
Bachman, G. 0.. and Machette. M. N. (l977). “CaIcic
Gile, L., Peterson, F. F., and Grossman, R. B, (1966).
Soil\ and Calcretes in the Southwestern United
Morphological and genetic sequences of carbonate
States.” U.S. Geological Survey Open File Report,
accumulation in desert soils. Soi/ Science 101, 34771-794.
360.
Benedict. J. B. (1979). Getting away from it all: A Goss. D. W.* Smith, S. J.. and Stewart, B. A. (1973).
ACKNOWLEDGMENTS
402
VANCE
T. HOLLIDAY
Movement
of added clay through
calcareous
material. Geodermu
9, 97- 103,
Guffee.
E. J. (lY7Y).
“The Plainview
site: Relocation
and archaeological
investigation
of a late Paleo-Indian kill site in Hale County,
Texas.”
Archaeological Research
Laboratory,
Llano Estacado
Museum.
Plainview!,
Tex.
Hall, S. A. (lY82). Late Holocene
paleoecology
of the
Southern
Plains. ~//~~?er~t~~~~ R~~.se~~rc/t 17, 3Y l-407.
Hawley,
J. W., Bachman,
G. 0.. and Manley.
K,
(lY76).
Quaternary
stratigraphy
in the Basin and
Range and Great Plains provinces,
New Mexico
and
western
Texas.
It7
“Quaternary
Stratigraphy
of
North
America”
(W. C, Mahaney.
Ed.)- pp. 235274. Dowden%
Hutchinson,
& Ross.
Stroudsburg,
Pa.
Haynes,
C. V. (lY68).
Geochronology
of late-Quaternary alluvium.
in “Means
of Correlation
of Quaternary
Successions”
(R. B. Morrison
and H. E.
Wright.
Jr., Eds.),
pp. 591-631.
Univ.
of Utah
Press. Salt Lake City.
Haynes,
C. V. (tY75).
Pleistocene
and recent
stratigraphy.
I?! “Late
Pleistocene
Environments
of the
Southern
High Plains”
(F. Wendorfand
J. J. Hester.
Eds.),
pp. 5Y-Y6,
Publication
9. Ft. Burgwin
Research Center.
Haynes,
C. V.. and Agogino,
G. A. (1966). Prehistoric
springs
and geochronology
of Blackwater
No. I locality,
New Mexico.
Arnerk~ut7
Antiquit!
31, 8l2821.
Hester.
J. J. (lY72).
“Blackwater
Draw Locality
No.
I: A Stratified
Early
Man Site in Eastern
New
Mexico.”
Publication
8. Ft. Burgwin
Research
Center.
Hohiday,
V. T. (1982a).
Mid-Holocene
climate
of the
southern
High Plains: Geologic
evidence.
AMQUA
Abstrucfs
7, 107.
Holliday.
V. T. ( l982b).
.‘Morphological
and Chemical
Trends in Holocene
Soils at the Lubbock
Lake Archaeological
Site, Texas.”
Ph.D. thesis.
University
of Colorado,
Boulder.
Holliday,
V. T, (Ed.) (lY83),
“Guidebook
to the Central Llano
Estacado.”
Friends
of the Pleistocene.
South-Central
Cell Field Trip. ICASALS
and the
Museum.
Texas Tech University.
Lubbock.
Holliday,
V, T. (1984).
Early
Holocene
soils at the
Lubbock
Lake archeological
site, Texs. Cutenu.
Holliday,
V. T., and Johnson,
E. (tY8l).
An update on
the Plainview
occupation
at the Lubbock
Lake site.
Plains Anihropologist
26, X1-253.
Holliday,
V. T.? Johnson,
E., Haas,
H., and Stuckenrath,
R. (tY83).
Radiocarbon
ages from the Lub-
bock Lake site. lY50- 1980: Framework
for cultural
and ecological
change on the Southern
High Plains.
P1uin.s Anfhropologist
28, l65- 182.
Inman,
D. L. (tY52).
Measures
for describing
the size
distribution
of sediments.
Joun7d
of ,Sedimen/og
Pewolog>j
22, l25- 145.
Johnson.
E. (lY76).
“Investigations
into the Zooarchaeology
of the Lubbock
Lake Site.”
Ph.D. dissertation.
Texas Tech University,
Lubbock.
Johnson,
E.% and Holliday.
V, T. (lY80).
A Plainview
kill/butchering
locale on the Llano
Estacado-the
Lubbock
Lake site. P/aim
Anfhropo/o,yist
25, 89111.
Lundelius,
E. L. (1974).
The last fifteen
thousand
years of fauna1 change in North
America,
It7 “History and Prehistory
of the Lubbock
Lake
site”
(C. C. Black,
Ed.), The M~~.wm
Jo7wt7d
(Lddx~d~
15, 141-160.
Lundelius,
E. L., Graham,
R. W.. Anderson.
E..
Guilday,
J., Holman.
J. A., Steadman%
D. W.. and
Webb. S. D. (lY83). Terrestrial
vertebrate
faunas. in
“Late
Quaternary
Environments
of the United
States-”
Vol. I. “The
Late Pleistocene”
(S. C.
Porter.
Ed.),
pp. 31 l-353.
Univ.
of Minnesota
Press. Minneapolis.
NOAA
(lY82a).
‘Climate
of New Mexico:
Climatography of the United
States, No, 60.” National
Climatic Data Center,
Asheville.
N.C.
NOAA
(lY82b).
“Climate
of Texas: Climatography
of
the United
States. No. 60.” National
Climatic
Data
Center.
Asheville,
N.C.
Reeves.
C. C.% Jr., (1976).
Quaternary
strdtigraphy
and geologic
history
of the Southern
High Plains.
Texas and New Mexico.
In “Quaternary
Stratigraphy
of North
America’*
(W. C. Mahaney.
Ed.),
pp. 213-234.
Dowden,
Hutchinson.
& Ross.
Stroudsburg.
PA.
Sellards,
E. H.. Evans,
G. L., and Meade.
G. E.
(1947).
Fossil
bison and associated
artifacts
from
Plainview,
Texas. Bulleiitl
of 717~ Geoiogicul
Societ.~
ofAtneric,u
58, Y27-954.
Soil Survey
Staff (1975). “Soil Taxonomy,”
Handbook
436. U.S. Department
of Agriculture,
Soil Conservation
Service.
Stevens,
D. (lY73).
*‘Blackwater
Draw Locality
No.
I, lY63-1972,
and Its Relevance
to the Firstview
Complex.”
M.S. thesis,
Eastern
New Mexico
University,
Portales.
Strahler.
A. N., and Strahler,
A. H. (1983). “Modern
Physical
Geography,”
2nd ed. Wiley,
New York.
Wormington.
H. M. (lY57).
‘*Ancient
Man in North
America.”
Denver
Museum
of Natural
History.