Content uploaded by Nick Blegen
Author content
All content in this area was uploaded by Nick Blegen on Dec 04, 2018
Content may be subject to copyright.
The Middle Stone Age After 50,000 Years Ago: New Evidence From the
Late Pleistocene Sediments of the Eastern Lake Victoria Basin, Western Kenya
ABSTRACT
Here we report tephra correlations, lithic artifacts, obsidian sourcing data, and fauna from nine Late Pleistocene
localities of the eastern Lake Victoria basin of western Kenya, as well as new excavations from the 49–36 ka site
of Nyamita Main on Rusinga Island. The Late Pleistocene of Africa is an important period for the evolution and
dispersals of Homo sapiens. A conspicuous behavioral feature of this period is the replacement of Middle Stone
Age (MSA) technologies by Later Stone Age (LSA) technologies. Current research shows this process is complex
with the LSA appearing and the MSA disappearing at dierent times in dierent places across Africa. Account-
ing for this paern requires a precise chronology, detailed evidence of past human behavior and environmental
reconstructions of the appropriate scale. Data presented here provide this detail. Tephra correlations improve the
regional chronology and expand the lateral area of Late Pleistocene eastern Lake Victoria basin exposures from
~650km2 to >2500km2. Lithic artifacts show MSA technology is present younger than 36 ka in western Kenya,
25–35 kyr younger than the rst appearance of early LSA technology elsewhere in equatorial East Africa. Obsid-
ian sourcing data presented here shows the use of the same raw material sources by MSA and LSA populations
through long periods of time from >100 ka through <36 ka. The methods employed here provide the temporal
resolution and appropriate geographic scale to address modern human behavioral evolution.
INTRODUCTION
Late Pleistocene Africa is an important time period for
understanding the evolution and dispersal of Homo sa-
piens. Fossil evidence of Homo sapiens is present in East Af-
rica by 195 ka (McDougall et al. 2005). This region also was
likely signicant to human dispersals across and out of Af-
rica in the Late Pleistocene between 130–50 ka (Beyin 2006;
Groucu et al. 2015; Rito et al. 2013). A conspicuous aspect
of hominin behavior in Late Pleistocene Africa is the disap-
pearance of Middle Stone Age (MSA) technologies, dened
by points and characterized by prepared disc cores, and the
replacement of these stone tool industries by Later Stone
Age (LSA) technologies, dened by backed geometric mi-
croliths (Ambrose et al. 2002) and often associated with
the bipolar core reduction (Diez-Martín et al. 2009), some-
time between ~60–30 ka (Ambrose 1998; Tryon and Faith
2013). The longevity and adaptability of MSA technology
throughout Africa make the reasons for its replacement by
LSA technology a maer of interest and debate.
THE MIDDLE TO LATE STONE AGE
TRANSITION ACROSS AFRICA
The latest occurrences of MSA technologies and earliest oc-
currences of the LSA technologies are staggered through-
PaleoAnthropology 2017: 139−169. © 2017 PaleoAnthropology Society. All rights reserved. ISSN 1545-0031
doi:10.4207/PA.2017.ART108
NICK BLEGEN
Max Planck Institute for the Science of Human History, Kahlaische Straße 10, 07745 Jena, GERMANY; and, Department of Anthropology, Har-
vard University, Peabody Museum of Archaeology and Ethnology, 11 Divinity Avenue, Cambridge, MA 02138, USA; blegen@shh.mpg.de; and,
nick_blegen@fas.harvard.edu
J. TYLER FAITH
Natural History Museum of Utah & Department of Anthropology, University of Utah, Salt Lake City, UT 84108, USA;
tyler.faith@anthro.utah.edu
ALISON MANT-MELVILLE
Department of Anthropology, University of Connecticut, Storrs, CT 06269, USA; Alison.melville@uconn.edu
DANIEL J. PEPPE
Terrestrial Paleoclimatology Research Group, Department of Geosciences, Baylor University, Waco, TX 76706, USA; daniel_peppe@baylor.edu
CHRISTIAN A. TRYON
Department of Anthropology, Harvard University, Cambridge, MA 02138, USA; christiantryon@fas.harvard.edu
submied: 14 October 2016; revised 19 August 2017; accepted 10 October 2017
140 • PaleoAnthropology 2017
electron spin resonance (ESR) performed on equid teeth
found in an archaeological excavation, although ESR on
another tooth from a dierent location within the Naisiu-
siu beds produced an age of 39±5 ka (Skinner et al. 2003).
Together, these dates suggest LSA artifacts in the Naisiusiu
beds date somewhere between 67–34 ka, and thus predate
the LGM.
At Enkapune Ya Muto in the central Kenyan rift, the
Endingi industry contains faceed platform triangular
points and akes with radial scar paerns characteristic of
MSA technologies, as well as three backed microliths char-
acteristic of the LSA (Ambrose 1998). This level has been
dated (lab number QL-4259) to 44.9±0.64 ka cal BP (cali-
brated using IntCal13 in OxCal; (Ramsey 2009; Reimer et
al. 2013)). The Nasampolai and Sakutiek industries, both
LSA industries containing backed microliths, overlie the
Endingi Industry. Radiocarbon ages of the Sakutiek indus-
try, the younger of these two LSA industries, range from
44.24±1.5–19.98±1.2 ka cal BP (Ambrose 1998). Based on a
combination of radiocarbon, obsidian hydration, and sedi-
mentation rates, the earliest LSA of Enkapune Ya Muto is
considered to have begun by at least ~50 ka and possibly
earlier (Ambrose 1998). The LSA materials at several sites
of Lukenya Hill in south-central Kenya are considered to
be <20 ka based on bone collagen and apatite 14C dates
(Kusimba 1999, 2001). The Late Pleistocene cultural layers
of GvJm-22 ‘Occurrence E’ and ‘Occurrence F’ were origi-
nally called LSA (Gramly and Rightmire 1973). However,
recent work suggests the lower of these two cultural layers,
‘Occurrence F,’ is MSA with Levallois cores and trimmed
points dating between 26 ka and >46 ka (Tryon et al. 2015).
The broad overlap of the MSA and LSA as well as
geographic variation in the timing of this transition across
Africa does not agree with models of behavioral evolution
positing a single, sudden event (Klein 2008). The complex
chronological and geographic circumstances of the MSA/
LSA transition demand a more detailed archaeological re-
cord accompanied by high resolution chronological and
paleoenvironmental records on expansive deposits of a
geographic scale that reasonably approximate the size of
the ancient landscape across which Pleistocene humans
would have ranged. In light of this, the Late Pleistocene
sediments of the eastern Lake Victoria Basin (eLVB) pro-
vide a valuable venue in which to address the MSA/LSA
transition. Tephrostratigraphy provides precise chrono-
metric dating and lateral stratigraphic control of geo-
graphically expansive deposits. Combined with ongoing
environmental reconstructions, this provides the ability to
assess the character of past landscapes over 100s to 1000s
of square kilometers. The rich history of archaeological re-
search combined with new data from ongoing eldwork
in the eLVB allows us to characterize hominin technologies
and behavior in the eLVB and across equatorial Africa. We
also examine previous and ongoing obsidian raw material
sourcing studies to gauge the scale of past hominin interac-
tions. Finally, we combine new geological and archeologi-
cal data from the eLVB with previously published research
from eastern Africa to provide insights into the MSA-LSA
out Africa with aspects of MSA technology occurring <30
ka regardless of broad dierences in raw materials, site
type, or dating technique employed. In southern Africa,
MSA technology persists as late as 28–27 ka at Rose Cot-
tage Cave based on charcoal 14C AMS dates (Wadley 2001),
while the MSA/LSA transition is dated between 49–45 ka
by AMS 14C methods on charcoal at Border Cave (Villa et
al. 2012). In southeastern Africa, cosmogenic nuclide dates
on sediments provide an age of ~29 ka for MSA materials
in Mozambique (Mercader et al. 2012). In northern Malawi,
MSA artifacts at Mwanganda’s Village site are OSL dated
to between 42–22 ka (Wright et al. 2014) and 47–30 ka at
Chaminade II (Wright et al. 2017). In western Africa, MSA
technologies are found in Pleistocene sediments of the
Senegal River Valley and OSL dated as late as 30–11.5 ka
(Chevrier et al. 2016; Scerri et al. 2015) and in the Nile Val-
ley of northern Sudan, OSL dates suggest MSA technology
persists until ~15 ka (Osypiński and Osypińska 2016). In
the southeastern Ethiopian cave of Goda Buticha, charac-
teristically MSA material is dated to ~34 cal BP by 14C with
aspects of MSA technology such as trimmed points and
a preferential Levallois core found in a middle Holocene
layer dated as late as ~7 ka (Tribolo et al. 2017).
In equatorial East Africa, the MSA/LSA transition is
conventionally thought to occur earlier. The Kisele indus-
try (MSA) from Bed VI of Mumba rockshelter in northern
Tanzania is dated to >65 ka based on U-series dates on bone
from the overlying Bed V and 73.6±3.8–63.4±5.7 ka based on
quar and feldspar OSL dates from Bed VI-A (Gliganic et
al. 2012; Mehlman 1989). The Mumba industry from over-
lying Bed V is variably characterized as “transitional” be-
tween MSA and LSA technologies, because of the presence
of backed geometric microliths, bifacial points, and pre-
pared disc cores (Marks and Conard 2008; Mehlman 1989),
or as an LSA industry because of the presence of backed
microliths and bipolar aking (Diez-Martín et al. 2009; Eren
et al. 2013), with the MSA elements incorporated into the
assemblage due to the excavation of vertically thick, arbi-
trary excavation spits (Prendergast et al. 2007). Bed V and
the Mumba industry are dated to between 65–45 ka based
on amino acid racemization (AAR) of ostrich eggshell frag-
ments and 56.9±4.8–49.1±4.3 ka based on quar grain OSL
dates (Gliganic et al. 2012; McBrearty and Brooks 2000). A
similar technological succession is found at Nasera rock-
shelter in northern Tanzania with strata containing the
Kisele industry dated to a minimum age of ~56 ka by U-
series on bone (Mehlman 1977, 1989; Tryon and Faith 2016).
The Naisiusiu beds of Olduvai Gorge contain LSA mi-
croliths (Leakey et al. 1972). Mehlman (1977, 1989) suggests
an age equivalent to the Last Glacial Maximum (LGM) for
these deposits based on archaeological similarities with the
LSA Lemuta industry at Nasera. Conventional radiocarbon
dates on bone from the Naisiusiu beds suggest very late
to terminal Pleistocene ages between 17.5±1.0–10.4±0.6 ka
(Hay 1976; Leakey et al. 1972), but more recent AAR dates
on ostrich eggshell of >42 ka and 40Ar/39Ar dates of 42.0±10.0
ka on tephra suggest a considerably older age (Manega
1993, 1995). Recent ages of 62±5 ka are reported based on
New MSA Evidence from the Eastern Lake Victoria Basin• 141
CHRONOLOGY AND STRATIGRAPHY
The base of the Late Pleistocene eLVB exposures known so
far is estimated to date to ~100 ka (see Figure 2; Beverly et
al. 2015b). The phonolitic Wakondo Tu was hypothesized
by Tryon et al. (2010) to derive from Late Pleistocene erup-
tions of Longonot or Suswa dated to 100±10 ka (Baker et
al. 1988). This tu is near the base of the Nyamita Valley
stratigraphic sequence and has been used to correlate these
exposures to other deposits of the Wasiriya beds across
Rusinga Island and to Karungu on the Kenyan mainland
(Blegen et al. 2015; Tryon et al. 2010; Van Plantinga 2011).
U-series dates of 94–111 ka on a barrage tufa in the Nyam-
ita Valley into which glass from the Wakondo Tu is incor-
porated provide additional support for this ~100 ka age es-
timate (Beverly et al. 2015). The most common stratigraphic
marker in the Nyamita Valley and throughout the eLVB is
the Nyamita Tu (see Figure 2; Beverly et al. 2015a; 2015b).
At section Nyamita 2 (see Figure 2), optically stimulated lu-
minescence (OSL) dates on sands bracketing the Nyamita
Tu provide ages of 50±4 ka and 46±4 ka below and above
the tu respectively, indicating this tu was deposited
~49 ka (Blegen et al. 2015; Tryon et al. 2010; Van Plantinga
2011). At least two stratigraphically higher tus are widely
encountered throughout the Late Pleistocene exposures
of the eLVB (Blegen et al. 2015). The uppermost of these
tus, originally referred to as the ‘Bimodal Trachyphono-
litic Tu’ and abbreviated to ‘BTPT’ (Blegen et al. 2015) has
recently been correlated over an area >115,000km2 across
much of Kenya and to its source Menengai Crater where it
is 40Ar/39Ar dated to 35.62±0.26 ka (Blegen et al. 2016). The
BTPT, now properly termed the Menengai Tu, provides
the minimum 40Ar/39Ar age for the Late Pleistocene eLVB
tephrostratigraphy. This minimum age is corroborated by
>35–33 ka AMS 14C dates on gastropod shells that post-dep-
ositionally burrowed into the sediments at the Nyamita 2
and Nyamita 3 localities (see Figure 2; Blegen et al. 2015;
Tryon et al. 2010; 2012).
PALEOENVIRONMENTS
Fauna from the Late Pleistocene beds of the eLVB consti-
tute the largest and most diverse Late Pleistocene faunal as-
semblages in East Africa. The majority of specimens belong
to taxa indicating open, semi-arid grasslands (e.g., equids
and alcelaphin antelopes) distinct from the evergreen bush-
lands, woodlands, and forests historically found in the re-
gion (Lillesø et al. 2011; Robertshaw et al. 1983; White 1983).
Isotopic data from tooth enamel indicate that the majority
of the large-bodied mammalian fauna recovered from the
Wasiriya and Waware beds consumed a predominantly C4
diet indicating the presence of an extensive C4 grassland
in the eLVB (Garret et al. 2015). Exceptions to this general
trend include the area of Nyamita Main. In addition to taxa
with open-habitat preferences such as Rusingoryx atopocra-
nion, Damaliscus hypsodon, and Grevy’s zebra (Equus grevyi)
found throughout the Nyamita landscape (and elsewhere
in eLVB), Nyamita Main has produced taxa associated with
dense cover or freestanding water, including reedbuck
(Redunca redunca), duiker (Sylvicapra grimmia), bushbuck
transition in equatorial Africa.
THE EASTERN LAKE VICTORIA BASIN
OF EAST AFRICA
GEOLOGY
Tectonic activity in the East African Rift System (EARS) has
produced a series of rift basins in which rapid sedimentation
buried paleoanthropological materials, with extensional
faulting subsequently re-exposing these materials. The Vic-
toria basin is formed in the depression between the eastern
and western branches of the EARS, probably within the last
few million years (e.g., Danley et al. 2012). During the Late
Pleistocene, the eLVB formed a repository for sediments, in-
cluding volcaniclastic deposits of volcanic eruptions from
>200km east in the central Kenyan Rift (Tryon et al. 2016).
While no formal geological formation has yet been dened,
the Late Pleistocene exposures of the eLVB are currently
known to include the Wasiriya beds of Rusinga Island, the
Waware beds of Mfangano Island, and the Late Pleistocene
exposures of Karungu on the Kenyan mainland (Figure 1c).
These sediments were informally named by Pickford (1984)
based on previous mapping and descriptions (Kent 1942;
Van Couvering 1972). The rst measured sections and sedi-
mentary descriptions of the Wasiriya beds were reported by
Tryon et al. (2010) and expanded in recent studies (Figure
2; Beverly et al. 2015b; Blegen et al. 2015; Garre et al. 2015;
Van Plantinga 2011). Sediments are exposed in sections up
to ~18m thick and are primarily comprised of four distinct
lithologies: 1) poorly sorted coarse sand and gravel chan-
nels cemented by carbonate representing episodic channel
erosion and deposition; 2) ne grained mudstone, siltstone,
and silty sandstone preserving evidence of incipient soil
development indicating a more stable landscape; 3) tephra
preserved as both primary fall-out deposits and tus that
have undergone varying amounts of reworking and incipi-
ent pedogenesis; and, 4) tufa deposits made primarily of
calcium carbonate indicating the presence of springs and
small ponded areas on the landscape (Figure 3; Beverly et
al. 2015b; Tryon et al. 2010). Barrage tufa deriving from the
Nyamita Valley spring ceased continuously forming in the
Late Pleistocene, but small springs still exist today (see Fig-
ure 3) and have likely existed intermiently from the Late
Pleistocene through the present, providing a source of fresh
water to local fauna.
Laterally discontinuous deposits are connected strati-
graphically and chronologically by tephrostratigraphy.
Late Pleistocene tephras are correlated over a ~60km north-
south transect between Rusinga Island, Mfangano Island,
and Karungu (Blegen et al. 2015). This tephrostratigraphy,
as well as lithostratigraphic and chemical characterization
of the paleosols between the tephras (Beverly et al. 2015a;
2015b; 2017), indicate the same stratigraphic sequence is
discontinuously preserved over an area of at least ~120km2.
Thus, the eLVB constitutes a laterally extensive stratigraph-
ic sequence of Late Pleistocene exposures preserving dier-
ent depositional environments.
142 • PaleoAnthropology 2017
Figure 1. A) Map of Africa showing sites mentioned; B) map of East Africa focused on Lake Victoria and eastern (Kenyan) rift region;
C) map of approximate area of the eastern Lake Victoria Basin (eLVB) around Winam Gulf referenced in this study. This includes
Rusinga Island, Mfangano Island, and Karungu, the southern shore of the Winam Gulf, and the Uyoma Peninsula on the north shore
of the Winam Gulf.
New MSA Evidence from the Eastern Lake Victoria Basin• 143
more dense woody cover than the reconstructions for other
parts of the Wasiriya and Waware beds in the same time pe-
riod (Garre et al. 2015). Together, the isotopic and mam-
malian fossil data indicate that the Nyamita Valley prob-
ably represented a patch of more dense riverine woodland
with standing water within an otherwise dry C4 grassland
ecosystem.
HISTORY OF ARCHAEOLOGY
Archaeological exploration of Late Pleistocene sediments of
the eLVB extends to the early twentieth century, with early
(Tragelaphus scriptus), impala (Aepyceros sp. nov.), and hip-
po (Hippopotamus amphibius) (Faith 2014; Faith et al. 2012;
2014; 2015; Garre et al. 2015; O’Brien et al. 2016; Tryon
et al. 2012). The presence of Hippopotamus indicates stand-
ing water, likely related to the spring system that probably
served as a magnet for water-dependent grazers and sup-
ported dense vegetation cover that favors taxa rarely found
in other parts of Nyamita Valley or the eLVB. Additionally,
isotopic reconstructions from Wasiriya beds paleosols in
the Nyamita Valley indicate this region was characterized
by ~46–82 % leafy cover in the Late Pleistocene. This is
Figure 2. A) Map of Rusinga Island showing the extent of Pleistocene outcrops of the Wasiriya beds as well as archaeological and pale-
ontological localities discussed in the text (after Tryon et al. 2010, 2012); B) map of Nyamita valley indicating relevant archaeological
and geological localities (after Beverly et al. 2015; Van Plantinga 2011); C) stratigraphic columns of measured and sampled sections,
arranged west (left) to east (right). Lithologies are indicated for all units. Tus are color-coded to chemically correlated group (see
Blegen et al. 2015). Tuaceous units not chemically characterized and assigned are shown in grey.
144 • PaleoAnthropology 2017
istic of the Late Pleistocene MSA in East Africa. Levallois
cores (recurrent and preferential) and Levallois points, as
well as bifacial and unifacial trimmed points, are found in
the Wasiriya beds of Rusinga Island (Tryon et al. 2010). Col-
lection of artifacts (n=12) from the surface of the Waware
beds on Mfangano Island lack Levallois technology in this
small sample, but contain bifacial points and can be also
aributed to the MSA (Tryon et al. 2012). The archaeology
of Karungu is indistinguishable from that of the Wasiriya
beds of Rusinga Island, including Levallois technology, as
well as bifacial and unifacial points (Faith et al. 2015; Tryon
et al. 2016).
Four excavations have previously been conducted at
three localities within our study area in the Late Pleistocene
sediments of the eLVB. One of these, Aringo 3, is a small
1x3m2 test excavation at the locality of Aringo, Karungu
(Faith et al. 2015). At Aringo 3, artifacts were recovered
from a gravel channel underlain by a tufa-cemented con-
glomerate indicative of the presence of a paleo-spring and
overlain by a ~2m thick paleosol capped by the Nyamita
Tu. Based on its stratigraphic position, the Aringo 3 site is
study of MSA materials by Archdeacon W.E. Owen (Owen
1937, 1938, 1939), at times in collaboration with L.S.B.
Leakey (Leakey and Owen 1945). Creighton Gabel (1969)
excavated a series of rockshelters along the north of the Wi-
nam Gulf, one of which, Randhore, sampled undated strata
aributed to the MSA (Tryon et al. 2016). Sally McBrearty
conducted excavations of Sangoan-Lupemban and MSA
materials at Muguruk (McBrearty 1986) and MSA material
in the Pleistocene sediments of Songhor (McBrearty 1981).
Laura Basell (2007) conducted work at Rambogo rockshel-
ter as part of a review of Sangoan-Lupemban industries in
eastern Africa (see Figure 1c). The authors’ current research
project began in 2008 as part of collaborative research on
the early Miocene and Late Pleistocene sediments of Rus-
inga Island (Peppe et al. 2009; Tryon et al. 2010).
RECENT FIELD ARCHAEOLOGY AND
EXCAVATIONS
All the fresh and unweathered stone artifacts found by our
project on Rusinga Island, Mfangano Island, and Karungu
since 2008 are typologically and technologically character-
Figure 3: A) Late Pleistocene barrage tufa at base of Nyamita Valley; B) modern spring in Nyamita Valley; C) close-up of spring
showing presence of perennial fresh water; D) photograph (looking ~southwest) at 2009 excavations at Nyamita Main. Note the Late
Pleistocene uvial sediments unconformably overlying more consolidated, variegated, and tilted early Miocene sediments.
New MSA Evidence from the Eastern Lake Victoria Basin• 145
SUMMARY OF PRIOR RESEARCH IN THE
LAKE VICTORIA BASIN
Research reviewed above emphasizes that the Late Pleisto-
cene exposures of the eLVB provide the chronological con-
trol, geographically expansive deposits, and archaeological
materials necessary to address human behavioral evolu-
tion in Late Pleistocene Africa.
Original work presented here further provides: 1) re-
ned chronological and stratigraphic control for 50–36
ka deposits through identication and correlation of new
tephras; 2) a greatly expanded (>2500km2) lateral area of
known Late Pleistocene eLVB exposures through identi-
cation, survey, and tephra-correlation of new geological
deposits; and, 3) improved characterization of MSA tech-
nology and hominin behavior through systematic surface-
collection (following Tryon et al. 2012) and expanded ar-
chaeological excavations at the site of Nyamita Main.
MATERIALS AND METHODS
TUFF SAMPLING STRATEGY
All 24 tu samples analyzed for this study were collected
from or can be stratigraphically linked to a series of 12 sec-
tions (>0.50m to <10m thick) measured from Late Pleis-
tocene outcrops in the eLVB between 2014 and 2016. The
geographic locations of surveyed localities introduced here
are provided in Figure 1c and GPS coordinates for tephra
samples at each of these localities are provided in Table 1.
Schematic stratigraphic sections with lithologies are pro-
vided in Figure 4. Whenever possible, tus were sampled
from sections with multiple tephra deposits exposed in
stratigraphic succession. Field correlations were made by
walking exposures and by using a Jacob’s sta and Abney
level to establish the stratigraphic equivalence between
exposed tus. Both eld and laboratory methods of cor-
relation are necessary as exposures are discontinuous and
tephra deposits in the eLVB vary widely in their thickness,
amount of subsequent soil development, and amount and/
or size of natural glass.
EXCAVATION
Excavations at Nyamita Main were carried out over three
weeks in August of 2013. Spatial data was collected in three
dimensions using a total station connected to a PC laptop
operating EDM for windows excavation software (McPher-
ron and Dibble 2002). Prior to excavation, the topography
of the locality was mapped in 0.40m intervals and all ar-
tifacts and fossils on the surface or exposed in situ were
recorded and collected. A 1x1m2 grid was laid out over
the intended area of excavation. Excavations were aligned
north-south along the western edge of Tryon’s 2009 trench
and carried out with arbitrary 10cm levels because of the
absence of any natural vertical stratigraphy within the ar-
tifact and fossil-bearing matrix. All in situ stone material
and fossil fauna with a maximum dimension >2cm were
piece-ploed. All material with a discernable long axis (a
ration of length/ width approximately ≥1.5) was piece-plot-
ted with a point at either end. All excavated sediment was
<100 ka and >49 ka. The excavated sample includes a Leval-
lois point and preferential and recurrent Levallois cores. It
also includes a ~2cm obsidian ake.
Two excavations have been conducted at Wakondo
(Jenkins et al. 2017; Tryon et al. 2010). A 4m2 trench dug
to an average depth of 0.7m recovered nine artifacts in situ
(Tryon et al. 2010). At Wakondo Bovid Hill, the larger of the
two excavations, a series of trenches totaling 19m2 with an
average depth of 0.50m were excavated into small and shal-
low braided stream channels cuing into a paleosol over-
lying the Wakondo Tu (Jenkins et al. 2017; Tryon et al.
2010). These excavations targeted a bone-bed of the extinct
alcelaphin Rusingoryx atopocranion deposited in sediments
OSL dated to 68 ka (Blegen et al. 2015). Stone artifacts from
the Bovid Hill excavation are few (n=78 with 9 in situ), but
these lithic artifacts are laminar, including those produced
by Levallois methods, and are consistent with an MSA ari-
bution. Cut-marked fauna indicates an association between
the bones and stone artifacts (Jenkins et al. 2017; Tryon et
al. 2010).
Tryon excavated 4m2 at Nyamita Main in December
2008–January 2009 with the goal of establishing the source
of the artifacts found on the surface. Raw material types
and technology suggest that the surface and in situ mate-
rial sample the same assemblage. The excavation samples
strata ~4–5m above the Nyamita Tu, and thus is <49 ka.
OBSIDIAN SOURCING: THE VICTORIA
BASIN AND EAST AFRICA
Geochemical sourcing data on Pleistocene obsidian arti-
facts in the eLVB has long been known from the MSA sites
of Songhor and Muguruk (Merrick and Brown 1984), and
more recently from the sites of Kisaaka and Aringo 3 (Faith
et al. 2015). The presence of obsidian chemically sourced
to Lake Naivasha at all these sites suggests persistent long
distance contact between Late Pleistocene populations
making MSA technologies in the eLVB and the central Ke-
nyan Rift 140–250km to the east. Obsidian sourcing from
Mumba and Nasera rockshelters in northern Tanzania
show MSA and early LSA producing hominins had con-
tinued contact with these same Naivasha obsidian sources
240–305km from these sites from at least ~60 ka through
the Holocene (Mehlman 1989; Merrick et al. 1994). Obsid-
ian LSA artifacts of the Naisiusiu beds of Olduvai Gorge
also derive ~250km away from the Sonanchi and Mundui
sources of the Naivasha basin. At Lukenya Hill, propor-
tions of obsidian appear to increase throughout MSA layers
at GvJm-16, and at least some of the non-local obsidians of
these assemblages appear to derived from Naivasha sourc-
es (Merrick and Brown 1984). The same is true of the LSA
layer, Occurrence E, of GvJm-22 at Lukenya Hill (Gramly
1976; Merrick and Brown 1984). Thus, not only do the Late
Pleistocene eLVB sites aest to the scale of hominin interac-
tions through raw material transport distances, these sites
provide important evidence that MSA hominins in western
Kenya and LSA hominins in northern Tanzania were using
the same resources in the Naivasha basin during the same
time periods.
146 • PaleoAnthropology 2017
TABLE 1. LOCATIONAL GPS INFORMATION (decimal degrees in WGS 1984)
FOR ALL LOCALITIES WITH TUFF SAMPLES NEWLY REPORTED IN THIS STUDY.
Tuff Sample
Locality
Section
GPS locality
(WGS 1984)
Chemical
Composition
LVP2014-18
Gode Ariyo
Sec.2015.A
S-0.503500°
E34.338417°
Wakondo
LVP2014-30
Homa Peninsula
Luanda
not available
Wakondo
LVP2015-48
Gode Ariyo
Sec.2015.A
S-0.503500°
E34.338417°
Wakondo
LVP2015-49 Gode Ariyo Sec.2015.WPT017
S-0.505267°
E34.339600° Wakondo
LVP2015-51
Gode Ariyo
Sec.2015.WPT018
S-0.506300°
E34.339383°
Wakondo
LVP2015-73
Maguna South
near Maguna
Point Site #2
S-0.762733°
E34.081217°
Wakondo
LVP2014-19
Gode Ariyo
Sec.2015.A
S-0.503500°
E34.338417°
Nyamita
LVP2014-20
Gode Ariyo East
Sec.2015.B
S-0.507733°
E34.353033°
Nyamita
LVP2014-27
God Bura
Elephant Site
S-0.730117°
E34.081717°
Nyamita
LVP2015-50
Gode Ariyo
Sec.2015.WPT017
S-0.505267°
E34.339600°
Nyamita
LVP2015-52
Gode Ariyo
Sec.2015.WPT018
S-0.506300°
E34.339383°
Wakondo
LVP2015-54
Gode Ariyo
near
Sec.2015.WPT018
S-0.506717°
E34.339416°
Nyamita
LVP2015-56
Gode Ariyo East
near Gode Ariyo
Sec2014.B
S-0.512583°
E34.35425°
Nyamita
LVP2015-57
Gode Ariyo East
Sec.2015.B
S-0.507733°
E34.353033°
Nyamita
LVP2015-61
Kajiei
WPT029
S-0.38775°
E34.74678°
Nyamita
LVP2015-63
Maguna
Maguna Point
Site #1
S-0.74628°
E34.08945°
Nyamita
LVP2015-65
Maguna
Maguna Point
Site #1
S-0.74628°
E34.08945°
Nyamita
LVP2015-67
Maguna
Maguna Point
Site #1
S-0.74628°
E34.08945° Nyamita
LVP2015-81
God Bura
Molly's Site
S-0.73178°
E34.19016°
Nyamita
LVP2015-86
Kachuku
WPT055
S-0.86383°
E34.20126°
Nyamita
LVP2014-29
Rangoye
Sec.2014.WPT 312
S-0.334833°
E34.294600°
Nyamsingula
LVP2015-55
Adam's Site
WPT021
S-0.494167°
E34.412567°
Nyamsingula
LVP2015-58
Gode Ariyo East
Sec.2015.B
S-0.507733°
E34.353033°
Nyamsingula
LVP2015-72
Maguna South
WPT 040
S-0.759716°
E34.08345°
Other
New MSA Evidence from the Eastern Lake Victoria Basin• 147
and methodologies outlined in Blegen et al. (2015) for teph-
ra analysis and Brown et al. (2013) for obsidian analysis.
EPMA of major element oxides on volcanic glass is the fa-
vored method of obsidian sourcing because this methodol-
ogy has been tested and proven valid for obsidians from
Stone Age assemblages in East Africa. A large comparative
dataset of geological obsidian sources characterized by
EPMA is available (Brown et al. 2013; Merrick and Brown
1984).
TEPHRA CORRELATION
Correlations rely on glass composition and are based on
overlap of major element oxides following Brown and
Nash (2014). Samples in this study include both fresh vitric
tephra deposits, as well as tus subsequently reworked by
uvial processes or variably overprinted by pedogenesis.
Lithic and crystal phases from an eruption cannot always
be reliably distinguished from those found in detritral
sediments, and thus our correlations rely on chemical com-
position of the glass component for all samples following
Brown and Nash (2014). All correlations are supported by
similarity coecients (SCs) to quantify similarity between
screened through 1/8” mesh and sieved material collected
in bags specic to date, square and level. All materials are
permanently housed in the collections of the Archaeology
Department of the National Museums of Kenya, Nairobi.
LITHIC ANALYSIS
No single, standardized framework of lithic analysis exists
in the study of MSA archaeology in Africa. Lithic analysis
and terminology follow Tryon et al. (2005). For data report-
ing, we use tables similar in format to Shea (2008), but de-
part slightly from his scheme in classifying ‘debris’ as ake
fragments and angular waste (aked material that cannot
be oriented).
GEOCHEMICAL CHARARTERIZATION OF
VOLCANIC GLASS
The geochemical analysis of tephras and obsidians focuses
on electron probe microanalysis (EPMA) of eleven major el-
ement oxide proportions in volcanic glass from tuaceous
deposits, sediments, and obsidian artifacts. Preparation
protocols for microprobe analysis of tephra and obsidian
follow University of Utah Microprobe facilities equipment
Figure 4. Schematic stratigraphic sections of tephra correlations between localities of the eastern Lake Victoria Basin. Tephra samples
with the ID number (example: ‘72’) are those samples newly presented in this study. Samples with an ‘*’ after the ID number (ex-
ample: 76*) are those recently published in a study of the Menengai Tu (Blegen et al. 2016).
148 • PaleoAnthropology 2017
Site, Luanda Homa Peninsula, and Kajiei on the south shore
of the Winam Gulf, and Rangoye on the Uyoma Peninsula
north of the Winam Gulf (see Figures 1c, 4). These tephra
correlations roughly quadruple the area over which Late
Pleistocene eLVB exposures are known, from ~650km2 to
>2500km2, and show that they occur both north and south
of the Winam Gulf (see Figure 1c).
Geology of Pleistocene Localities
In the three new Karungu localities, the sedimentary pack-
ages and depositional environment above, below, and be-
tween the tephras appear similar in thickness and charac-
ter to the paleosols described for nearby stratigraphically
equivalent sites of Karungu (Beverly et al. 2015a). The sedi-
mentary sequences Gode Ariyo and Gode Ariyo East on the
south shore of the Winam Gulf (see Figure 1c) also preserve
a sequence of paleosols between known tephra. The expo-
sures at Gode Ariyo are unique in the Late Pleistocene beds
of the eLVB in preserving a >2m thick and well-developed
paleo-Vertisol below the Wakondo Tu (see Figure 4; Figure
5a). This lower stratigraphic interval is very poorly exposed
in the Late Pleistocene beds of the eLVB, and indicates the
presence of lower stratigraphic intervals within the same
Late Pleistocene eLVB tephra sequence further northeast
up the Winam Gulf and Nyanza Rift. The sedimentary se-
quence at Kajiei near the southeast corner of the Winam
Gulf is distinctive. The character of the Nyamita Tu here is
typical, but there are no paleosols. Instead the strata found
above and below the Nyamita Tu at Kajiei are comprised
of coarse sands locally derived from uvially redeposited
weathered granitic basement rock.
ARCHAEOLOGY OF LATE PLEISTOCENE
SITES
Much of the archaeological material collected at the new
eLVB localities is typologically and technologically arib-
utable to the MSA, characterized by Levallois points as well
as uni- and bifacially trimmed pointed pieces (Figure 6).
Prepared cores spanning a wide range of sizes, raw mate-
rials, and styles of preparation have also been recovered
(Table 4; Figures 6–9) similar to what has been observed
on Rusinga Island, Mfangano Island, and Karungu in the
eLVB (Tryon et al. 2016).
A few localities oer archaeological materials typo-
logically or stratigraphically dierent from what has been
previously observed. At Maguna Point Site #2 at the local-
ity of Maguna South in the Karungu area, MSA bifacial
trimmed points (see Figure 6d, e) and Levallois cores (see
Figure 8a, c) are found associated with and directly above
primary air-fall exposures of the 35.62±0.26 ka Menengai
Tu (see Figures 4, 5b). The MSA material collected above
the Menengai Tu at Maguna South makes this material
demonstrably the youngest MSA material in the eLVB.
The exposures at God Bura also provide an assemblage of
lithic artifacts from a controlled surface collection conduct-
ed above the Nyamita Tu at the God Bura Elephant Site.
MSA material collected in association with bones and a tufa
deposit consists of three Levallois points (see Figures 6j and
means of glass analyses from samples after which a tephra
is named the ‘type sample’ and the mean value of all other
modes in our dataset following the methodology of Blegen
et al. (2015). We use ten elements and/or element oxides:
SiO2, TiO2, Al2O3, FeO, MnO, MgO, CaO, Na2O, K2O, and
Cl in computation of SCs. In this methodology, for any
two-tephra comparison, SCs are the ratios obtained by di-
viding pairs of sample means (with the larger value of the
two samples always the denominator such that the ratio is
always ≤1.0) element by element as dened by Borchardt
et al. (1972). Resulting SCs range between 0 (complete dis-
similarity) to 1 (perfect similarity). Previous studies have
proposed arbitrary cutos for interpreting SCs in terms of
potential correlation. Kuehn and Foit (2006) proposed a
value of ≥0.95 for denitive correlation, and Frogga (1992)
recognizes that values ≥0.92 are typically accepted for cor-
relations. As in Blegen et al (2015), this study employs ran-
domization procedures to develop empirically informed
SC cutos for accepting or rejecting potential correlations.
In this methodology, for each of the tephra type samples
published in Blegen et al. (2015), the means and standard
deviations of each element oxide are used to generate 5000
random, normally–distributed samples using Microsoft Ex-
cel statistical package (Rochowicz, Jr. 2010); this eectively
represents 5000 replicates of the type tephra. We then cal-
culate SCs between each of the 5000 replicates and the type
sample that are used to generate a frequency distribution
of expected SC values when comparing two samples of the
same tu. From this distribution, we determine the lower
SC limit that encompasses the upper 95% of observations.
We use this value as the lower limit cuto for rejecting po-
tential correlations. For example, an SC value between an
unknown tu and a type sample that falls below the 95%
cuto is excluded for consideration as a potential correlate.
To increase the stringency of our protocol, we also require
that the ten element oxides of the unknown tu considered
in our analysis overlap within two standard deviations of
the mean of the type sample. This is because an unknown
sample that is very similar in composition for most element
oxides (e.g., 9 of 10) to a type sample will record a rela-
tively high SC value, even if one oxide is distinct and out-
side the range of expected values. The SCs included in this
analysis were used as a data exploration and conrmation
technique. All correlations were investigated in more detail
utilizing the known stratigraphy of a site and visual inspec-
tion of the tephra datasets.
RESULTS
TEPHROSTRATIGRAPHY AND GEOLOGY OF
LATE PLEISTOCENE SITES
Tephra Correlations
Chemical characterizations of 328 individual shards from
24 tu samples (Tables 2 and 3) incorporate nine new lo-
calities into the Late Pleistocene tephrostratigraphic frame-
work of the eLVB: God Bura, Maguna, and Kachuku in the
region of Karungu, Gode Ariyo, Gode Ariyo East, Adam’s
New MSA Evidence from the Eastern Lake Victoria Basin• 149
TABLE 2. MEAN MAJOR AND MINOR ELEMENT OXIDES AND ELEMENTS BY WEIGHT PERCENT.*
Sample M,N No. SiO2 TiO2 ZrO2 Al2O3 FeO MnO MgO CaO Na2O K2O F Cl Sum
O
Sum
less
O
H
2
O
TOTAL
LVP2014-18
1,1
20
59.96
0.53
0.12
15.42
7.91
0.35
0.33
1.02
8.71
4.85
0.51
0.30
100.00
0.28
99.72
1.68
100.52
1.47
0.03
0.04
0.30
0.18
0.03
0.01
0.04
0.33
0.19
0.04
0.02
2.21
0.02
2.21
1.00
1.76
LVP2014-30
1,1
25
59.23
0.54
0.11
15.25
7.79
0.35
0.32
0.99
8.46
4.76
0.55
0.31
98.65
0.30
98.35
2.41
99.89
1.77
0.04
0.05
0.49
0.25
0.03
0.01
0.07
0.82
0.26
0.15
0.02
3.34
0.07
3.33
1.76
2.24
LVP2015-48
1,1
14
60.80
0.59
0.12
15.27
7.91
0.36
0.26
0.99
8.75
4.93
0.30
0.30
100.58
0.19
100.39
1.46
100.96
0.98
0.03
0.06
0.25
0.17
0.03
0.02
0.03
0.56
0.17
0.04
0.02
1.91
0.02
1.92
1.13
1.18
LVP2015-49
1,1
6
60.75
0.58
0.12
15.20
7.90
0.35
0.32
1.01
8.59
5.07
0.36
0.29
100.54
0.22
100.33
1.57
101.02
0.56
0.01
0.03
0.12
0.12
0.02
0.02
0.02
0.38
0.19
0.04
0.02
1.01
0.02
1.01
1.14
0.37
LVP2015-51
1,1
15
60.89
0.59
0.11
15.28
7.85
0.35
0.40
0.99
8.68
5.02
0.35
0.30
100.82
0.21
100.60
1.56
101.29
0.57
0.02
0.05
0.14
0.10
0.03
0.02
0.02
0.48
0.13
0.05
0.02
1.22
0.02
1.22
1.23
0.41
LVP2015-73
1,1
15
60.37
0.61
0.10
15.53
7.55
0.34
0.29
1.10
8.05
4.92
0.29
0.27
99.43
0.18
99.24
2.30
100.70
0.42
0.06
0.05
0.26
0.62
0.04
0.05
0.20
0.84
0.15
0.08
0.06
1.47
0.05
1.43
0.73
1.28
LVP2014-19
1,1
20
61.12
0.57
0.09
15.56
6.55
0.28
0.32
1.04
7.33
4.67
0.37
0.16
98.34
0.19
98.15
3.54
100.97
1.42
0.15
0.05
1.61
1.66
0.07
0.09
0.29
1.81
1.21
0.15
0.05
2.55
0.07
2.55
2.33
1.31
LVP2014-20
1,1
18
61.26
0.58
0.10
15.55
6.63
0.30
0.34
1.11
7.47
4.99
0.39
0.17
98.89
0.20
98.68
2.91
100.86
0.97
0.04
0.03
0.23
0.13
0.02
0.01
0.03
0.25
0.26
0.04
0.01
1.63
0.02
1.64
1.10
0.81
LVP2014-27
1,1
8
58.81
0.56
0.11
15.10
6.65
0.29
0.41
1.11
7.06
4.86
0.40
0.18
95.54
0.21
95.33
2.78
97.37
2.42
0.05
0.04
0.60
0.41
0.03
0.11
0.08
0.57
0.44
0.09
0.03
4.08
0.04
4.08
1.10
4.41
LVP2015-50
1,1
15
61.78
0.64
0.10
15.25
6.64
0.28
0.41
1.12
7.34
5.13
0.13
0.16
98.99
0.09
98.90
2.72
100.89
0.85
0.04
0.04
0.29
0.16
0.04
0.03
0.07
0.39
0.26
0.08
0.01
1.53
0.03
1.53
1.34
0.66
LVP2015-52
1,1
15
62.02
0.64
0.10
15.53
6.66
0.28
0.31
1.11
7.36
5.34
0.20
0.16
99.72
0.12
99.59
2.96
101.81
0.63
0.02
0.04
0.18
0.12
0.02
0.03
0.05
0.32
0.29
0.05
0.02
1.19
0.02
1.17
1.24
0.71
LVP2015-54
1,2
8
61.85
0.64
0.09
15.31
6.65
0.28
0.38
1.12
7.55
5.16
0.22
0.16
99.42
0.13
99.29
3.07
101.62
0.89
0.04
0.05
0.17
0.17
0.01
0.01
0.04
0.47
0.30
0.04
0.01
1.54
0.02
1.54
1.41
0.70
150 • PaleoAnthropology 2017
TABLE 2. MEAN MAJOR AND MINOR ELEMENT OXIDES AND ELEMENTS BY WEIGHT PERCENT (continued).*
Sample M,N No. SiO2 TiO2 ZrO2 Al2O3 FeO MnO MgO CaO Na2O K2O F Cl Sum
O
Sum
less
O
H
2
O
TOTAL
LVP2015-54
2,2
2
64.95
0.59
0.18
10.22
9.06
0.43
0.17
0.64
5.86
4.49
0.54
0.32
97.44
0.30
97.14
2.77
98.90
0.94
0.03
0.01
0.16
0.09
0.01
0.01
0.01
1.57
0.18
0.03
0.00
0.26
0.01
0.24
0.48
0.73
LVP2015-56
1,1
15
62.58
0.65
0.08
15.39
6.77
0.30
0.35
1.10
7.76
5.32
0.19
0.16
100.67
0.12
100.55
1.20
101.00
0.42
0.03
0.03
0.18
0.08
0.02
0.03
0.03
0.17
0.19
0.04
0.01
0.67
0.01
0.67
0.74
0.41
LVP2015-57
1,1
14
62.28
0.66
0.08
15.36
6.76
0.31
0.34
1.09
7.54
5.34
0.19
0.16
100.10
0.12
99.98
1.79
101.02
0.69
0.05
0.04
0.18
0.16
0.04
0.02
0.06
0.27
0.26
0.06
0.01
1.18
0.03
1.18
1.21
0.76
LVP2015-61
1,1
14
62.03
0.67
0.08
15.67
6.73
0.29
0.33
1.12
7.70
5.11
0.19
0.16
100.36
0.12
100.25
0.19
99.68
0.40
0.04
0.03
0.06
0.08
0.02
0.02
0.01
0.15
0.04
0.06
0.01
0.33
0.03
0.33
0.28
0.58
LVP2015-63
1,1
5
61.57
0.67
0.11
15.74
6.67
0.31
0.33
1.11
7.57
5.07
0.15
0.17
99.45
0.10
99.35
1.59
100.21
1.01
0.05
0.04
0.30
0.10
0.03
0.02
0.03
0.23
0.32
0.04
0.01
1.48
0.01
1.49
0.97
0.63
LVP2015-65
1,1
14
61.31
0.66
0.07
15.69
6.61
0.32
0.33
1.10
7.34
5.03
0.21
0.17
98.92
0.12
98.79
1.66
99.72
0.68
0.04
0.02
0.18
0.07
0.02
0.02
0.03
0.40
0.23
0.04
0.00
1.39
0.02
1.40
1.21
0.49
LVP2015-67
1,1
2
62.26
0.68
0.10
15.84
6.77
0.29
0.33
1.11
7.65
5.15
0.23
0.17
100.57
0.13
100.43
0.02
99.66
0.48
0.01
0.07
0.16
0.10
0.02
0.03
0.02
0.02
0.19
0.02
0.00
0.66
0.01
0.65
1.01
0.37
LVP2015-81
1,1
15
61.55
0.64
0.08
15.86
6.29
0.27
0.35
1.07
7.43
5.18
0.21
0.16
99.08
0.12
98.96
2.33
100.59
1.17
0.10
0.03
0.49
0.91
0.05
0.06
0.11
0.31
0.27
0.07
0.03
1.83
0.03
1.82
1.32
1.04
LVP2015-86
1,1
15
61.33
0.67
0.11
15.63
6.59
0.30
0.40
1.12
7.35
5.10
0.20
0.17
98.97
0.12
98.84
2.70
100.81
1.12
0.03
0.05
0.26
0.15
0.02
0.03
0.03
0.31
0.21
0.04
0.01
1.89
0.02
1.89
1.38
1.79
LVP2014-29
1,1
14
58.44
0.41
0.19
15.48
7.20
0.33
0.22
0.77
7.83
4.63
0.61
0.41
96.51
0.35
96.16
3.97
99.33
1.71
0.11
0.05
0.69
0.77
0.06
0.07
0.12
1.10
0.27
0.10
0.09
3.16
0.06
3.20
1.76
2.06
LVP2015-55
1,1
13
60.77
0.52
0.22
16.10
6.54
0.29
0.37
1.03
8.28
5.32
0.46
0.35
100.24
0.27
99.97
2.30
101.55
0.92
0.05
0.05
0.23
0.30
0.02
0.03
0.06
0.55
0.18
0.10
0.05
1.89
0.05
1.86
1.24
0.86
LVP2015-58
1,1
15
60.78
0.44
0.23
15.52
7.27
0.33
0.19
0.98
8.77
4.96
0.58
0.41
100.46
0.34
100.13
1.79
101.11
0.99
0.04
0.05
0.28
0.12
0.02
0.02
0.07
0.70
0.22
0.09
0.06
1.93
0.05
1.94
1.45
0.91
LVP2015-72
1,1
11
61.44
0.58
0.11
14.35
7.55
0.31
0.15
1.06
7.33
4.75
0.25
0.23
98.12
0.16
97.96
2.64
99.76
1.04
0.05
0.04
0.43
0.13
0.03
0.10
0.13
0.42
0.23
0.07
0.05
1.54
0.04
1.54
0.99
1.00
*Sample listed on left (M,N=mode number, number of modes; No.=number of analyses). One standard deviation from the mean listed below each element oxide mean. Voltage=10keV,
beam current=25nA, spot size=10mm.
New MSA Evidence from the Eastern Lake Victoria Basin• 151
implying that it is not likely derived from the Pleistocene
sediments at the site.
OBSIDIAN
A ~3cm obsidian medial blade fragment from the 68 ka site
of Wakondo Bovid Hill (Jenkins et al. 2017) was recovered
in 2011 and analyzed for this study (see Figure 6i). It is
chemically indistinguishable from the source of Sonanchi
crater in the Naivasha basin of the central Kenyan rift val-
ley, ~250km east of the Wakondo Bovid Hill site. Another
obsidian core fragment (~3cm) collected from above a unit
aributed to the Nyamita Tu at Onge, Karungu and ana-
lyzed for this study is from this same Sonanchi source. A
single piece of obsidian angular waste (~1cm) surface col-
lected from above the Nyamita Tu at Kajiei is sourced to
south Naivasha basin (Table 5).
6n for examples) as well as a large bifacial pointed tool (see
Figure 9c). Previously, the only large bifacial artifacts (n=2)
collected by our team from the Pleistocene exposures of the
eLVB are heavily weathered, rolled, and clearly redeposited
(Faith et al. 2015; Tryon et al. 2012). The large bifacial tool
at God Bura Elephant Site could also have been reworked
from (currently unknown) sediments much older than the
Nyamita Tu (49 ka), but the artifact is not heavily rolled or
weathered. The locality of Kajiei on the southeast extremity
of the Winam Gulf preserves Levallois points (see Figure
6k, l) and a very small (1.9cm) Levallois core (see Figure
6c) collected above the Nyamita Tu. Exposures above the
Nyamita Tu at Kajiei included a backed piece (see Figure
6h), the only LSA artifact recovered in our surveys of Late
Pleistocene eLVB exposures. The artifact in question was
found on a deation surface in association with poery,
TABLE 3. SIMILARITY COEFFICIENTS (SCs) FOR ALL DISTINCT MODES
OF SAMPLES BASED ON TEN MAJOR-ELEMENT LIST
(samples reported here listed vertically [left] and compared with the type samples [top])*.
Sample
Wakondo Tuff
CAT09-05
(0.92)
Nyamita Tuff
CAT09-21
(0.93)
Nyamsingula Tuff
CAT10-03
(0.91)
LVP2014-18
0.98
0.87
0.89
LVP2014-30
0.98
0.87
0.90
LVP2015-48
0.95
0.86
0.90
LVP2015-49
0.96
0.88
0.88
LVP2015-51
0.95
0.88
0.87
LVP2015-73
0.93
0.90
0.88
LVP2014-19
0.88
0.95
0.85
LVP2014-20
0.88
0.98
0.86
LVP2014-27
0.86
0.95
0.84
LVP2015-50
0.84
0.96
0.82
LVP2015-52
0.86
0.95
0.84
LVP2015-54
0.85
0.96
0.83
LVP2015-54
0.79
0.73
0.73
LVP2015-56
0.86
0.97
0.84
LVP2015-57
0.86
0.96
0.84
LVP2015-61
0.86
0.97
0.85
LVP2015-63
0.87
0.96
0.85
LVP2015-65
0.87
0.96
0.84
LVP2015-67
0.86
0.96
0.84
LVP2015-81
0.85
0.95
0.83
LVP2015-86
0.85
0.96
0.83
LVP2014-29
0.86
0.80
0.91
LVP2015-55
0.91
0.90
0.92
LVP2015-58
0.89
0.82
0.94
LVP2015-72
0.86
0.87
0.82
*Type samples from Blegen et al. (2015) and Lower 95% confidence limit for 10 element SCs (in
parentheses) for each type sample/mode in top row.
152 • PaleoAnthropology 2017
Figure 5. Photographs of Gode Ariyo and Maguna South sites sampled for tephra analysis in July 2015: A) photograph of Gode Ariyo
near measured section ‘Sec.2015WPT 017’ showing the large paleo-Vertisol exposed beneath the Wakondo Tu at this locality. This
lower stratigraphic interval is very poorly exposed at other known localities in the Late Pleistocene beds of the eLVB; B) photograph of
Maguna Point Site #2 collection at Maguna South. Note MSA artifacts are coming from above the Menengai Tu at Maguna Point
Site #2; this upper stratigraphic interval is not exposed at other known localities in the Late Pleistocene beds of the eLVB. Artifacts
collected from Maguna South: Maguna Point Site #2 are pictured in Figures 6 and 7.
New MSA Evidence from the Eastern Lake Victoria Basin• 153
Figure 6. Selection of pointed pieces and technologically diagnostic akes from new localities in the eLVB: A) bifacial point, chert
Maguna Point Site #1; B) tip (broken) of a thick biface, chert, Aringo; C) bifacial point (tip broken), chert, Aringo; D) bifacial point,
ne-grained lava, Magana South WPT 040; E) bifacial point, chert Maguna Point Site #2; F) Levallois ake, retouched distally and
ventrally ne-grained lava, Kisaaka; G) bifacial point (tip broken) black chert, Maguna Point Site #1; H) backed microlith, chert,
Kajiei above Nyamita Tu; I) medial blade fragment, obsidian, Wakondo Bovid Hill; J) Levallois point, coarse-grained lava, God Bura
Elephant Site above Nyamita Tu; K) Levallois point, lava, Kajiei above the Nyamita Tu; L) Levallois point, red lava, Kajiei above the
Nyamita Tu; M) Levallois point (tip broken), lava, Maguna Point Site #1; N) Levallois point, red lava, God Bura Elephant Site above
Nyamita Tu; O) Levallois point, lava, Gode Ariyo, above Nyamita Tu; P) Levallois point, lava, Maguna South; Q) bifacial worked
piece, (point?), obsidian with mae patina, Kisaaka.
154 • PaleoAnthropology 2017
TABLE 4. SURFACE COLLECTED ARTIFACTS FROM EASTERN LAKE VICTORIA BASIN SITES IN 2015.
Artifact Type
Nyamita
Rangoye
Kajiei
Kachuku
Wakondo-
Bovid Hill
Aringo
God
Bura
God
Bura
Elephant
site
God
Bura
Molly
Site
Gode
Ariyo
Gode
Ariyo
East
Gode
Ariyo
West
Kisaaka
Maguna
Maguna
Point
Site 1
Maguna
Point
Site 2
Maguna
South
Obware
Onge
Total
Complete flakes
Initial cortica l flake
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
1
Residual cortica l flake
0
1
0
0
0
0
0
1
0
0
0
0
0
1
0
3
0
0
0
6
Levallois
0
0
0
0
0
1
2
2
0
1
0
0
0
0
0
0
0
1
0
7
Levallois point
1
1
0
0
0
0
3
3
0
2
1
0
0
0
1
0
2
0
0
14
Non-cortical flake
1
9
3
0
0
4
3
3
0
1
0
2
1
2
9
27
3
0
2
70
Blade
0
0
1
0
0
1
0
0
0
0
0
0
0
0
2
0
0
0
0
4
Proximal flake fra gments
Levallois flake
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
1
0
3
Non-Levallois flake
0
0
0
0
0
1
1
2
0
2
0
0
0
2
3
17
0
0
1
29
Non-Levallois blade
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
4
0
1
1
6
Flake fragment other 0 0 1 0 0 0 0 0 1 0 0 0 0 0 8 24 4 2 1 41
Angular waste 0 0 0 0 0 1 0 0 1 1 0 0 0 1 18 50 0 0 0 72
Tools
Biface
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
Bifacial point
0
0
0
0
0
6
0
0
0
0
0
0
1
1
1
1
1
1
0
12
Point
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
End scraper
0
1
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
2
Retouched flake or flake
frag
0
0
0
1
0
4
0
0
0
2
0
1
0
0
1
1
0
0
1
11
Retouched blade or
blade frag
0
0
1
0
0
1
0
0
0
1
0
0
0
0
0
0
0
0
0
3
Microlith
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
New MSA Evidence from the Eastern Lake Victoria Basin• 155
TABLE 4. SURFACE COLLECTED ARTIFACTS FROM EASTERN LAKE VICTORIA BASIN SITES IN 2015 (continued).
Artifact Type
Nyamita
Rangoye
Kajiei
Kachuku
Wakondo-
Bovid
Hill
Aringo
God
Bura
God
Bura
Elephant
site
God
Bura
Molly
Site
Gode
Ariyo
Gode
Ariyo
East
Gode
Ariyo
West
Kisaaka
Maguna
Maguna
Point
Site 1
Maguna
Point
Site 2
Maguna
South
Obware
Onge
Total
Cores, core fragments
Casual 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 2
Multiplatform
0
0
1
0
0
0
1
0
0
0
1
0
0
0
0
0
0
0
1
4
Levallois preferential
0
1
0
0
1
2
0
1
0
1
2
0
0
0
0
0
0
0
0
8
Levallois recurrent
0
2
1
0
0
4
1
0
0
2
2
0
2
0
0
1
1
0
1
17
Blade
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
1
Discoidal
0
0
0
0
0
1
0
0
0
2
0
0
0
0
1
1
0
0
0
5
Core on flake
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
Core frag 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 1 3
Indeterminate
1
0
0
0
0
4
0
0
0
0
0
0
0
0
0
0
1
0
0
6
Total
3
17
9
1
1
32
12
13
2
16
6
3
5
7
46
130
13
6
9
331
Raw Material (all
artifacts)
Lava 1 15 3 1 1 14 10 12 1 11 4 2 4 3 39 105 9 3 3 241
Chert-White/Red
2
0
1
0
0
1
0
0
0
1
0
0
0
0
0
0
0
0
0
5
Chert- Br own/Ta n
0
1
0
0
0
5
0
0
0
1
2
0
0
0
1
3
0
0
0
13
Chert - Bla ck
0
0
2
0
0
2
0
0
0
0
0
0
0
1
0
8
0
0
0
13
Chert ( type not specified)
0
1
2
0
0
5
1
1
1
2
0
0
0
2
1
4
3
2
2
27
Chert - other
0
0
0
0
0
3
0
0
0
1
0
1
1
1
1
3
0
0
1
12
Obsidian
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
2
Quartz 0 0 0 0 0 1 1 0 0 0 0 0 0 0 3 7 1 1 1
15
Quartzite
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
1
3
Total
3
17
9
1
1
32
12
13
2
16
6
3
5
7
46
130
13
6
9
331
156 • PaleoAnthropology 2017
Figure 7. Selection of small cores from new localities in the eLVB: A) Levallois core, green chert, above Nyamita, Kisaaka; B) Levallois
preferential ake core, white chert, Gode Ariyo; C) Levallois core, chert, Kajiei, above Nyamita Tu; D) Levallois core, lava, Kisaaka;
E) discoid core, red quarite, Onge, above Nyamita tu.
New MSA Evidence from the Eastern Lake Victoria Basin• 157
Figure 8: Selection of medium sized cores from new localities in the eLVB: A) Levallois core, lava Maguna Point Site #2, Maguna
South above Menengai Tu; B) Levallois core, coarse-grained two-toned brown lava, Rangoye, near Nyamsingula Tu; C) Levallois
core, lava, God Bura above Nyamita Tu.
158 • PaleoAnthropology 2017
Figure 9. Selection of large sized cores from new localities in the eLVB: A) Levallois core, coarse-grained green lava, Gode Ariyo above
Nyamita Tu; B) Levallois core, coarse-grained two-toned brown lava, Aringo; C) large bifacial pointed tool, lava, God Bura Elephant
Site, above Nyamita Tu.
New MSA Evidence from the Eastern Lake Victoria Basin• 159
cavations (n=388) are made of the moled chert (n=204) or
lava (n=184) and whole ake measurements by raw mate-
rial group are provided in Table 7. The moled chert source
is found on the eastern side of Rusinga Island and the apha-
nitic lava likely is derived from the Lunene Lavas, or simi-
larly aged lavas from Kisingiri, which are found in place at
the top of the ridges on Rusinga Island and as cobbles in
drainages o of the ridges (Tryon et al. 2014). The quar
and quarite have been observed in Pleistocene gravels at
Nyamita, and as large clasts within granodiorite volcanic
bombs in the early Miocene Kiahera Formation. All of these
raw materials are local. The source of the green and black
cherts has not been observed on Rusinga Island and these
materials are likely non-local.
Taphonomic Context of Lithic Materials from Nyamita
Main
Lithic artifacts excavated from the site represent the prod-
ucts of on-site knapping in a minimally winnowed context.
Both the moled chert and the lava raw materials, the only
raw materials represented by suciently large numbers to
assess their taphonomic integrity, conform to experimental
models for stone tools produced on-site and preserved in a
primary context (Schick 1986). Size class analysis of piece-
ploed and sieve-recovered moled chert artifacts shows
that 80.0% of all artifacts are ≤ 2cm in maximum dimension
(Figure 12). The lava raw material is in a slightly winnowed
context, as lava pieces recovered from the excavation show
slightly lower proportions of size class 1 akes (< 1cm in
maximum dimension). Accordingly, higher proportions of
the akes are from larger size classes 3–5, indicating some
uvial action sorted and removed smaller pieces. However,
the presence of both lava cores and akes in the same part
of the site, suggests the material was knapped on-site and
has been subject to minimal winnowing removing some of
the smaller pieces.
Technology of Lithic Materials
Like all lithic artifacts known from the eLVB, lithic artifacts
from Nyamita Main are typologically and technologically
characteristic of the MSA (Figure 13). Flaked pieces include
bifacial points (n=2) and unifacial points (n=1) collected
EXCAVATIONS OF NYAMITA MAIN
In August 2013, an excavation area of 52m2 was opened on
the crest and eastern slope of a small hill forming the lo-
cality of Nyamita Main. The excavations uncovered sand-
to-gravel-bearing channels cross-cuing a silt-to-clay-rich
paleosol unit (Figures 10 and 11).
The main trench of the 2013 excavation was oriented
approximately east–west and covered an area of 26m2 to
an average depth of 0.70m below surface. It extended to
Tryon’s 2009 4m2 trench to recover a larger sample of lithic
artifact material in situ (see Figure 10). The north trench
was oriented north–south along the crest of the hill and an
area of 26m2 was excavated to an average depth of 0.85m
to recover faunal material in situ and establish its context.
Sediments encountered in the Nyamita Main excava-
tions are variants of the rst two general lithological cate-
gories found in the Wasiriya beds. These are: 1) dark brown
clay paleosols with varying amounts of silt and sand; and,
2) moderately well-sorted medium to coarse sand with
sub-round to angular lava and carbonate grains (see Figure
11a-c).
Lithic Assemblage of Nyamita Main
A total of 449 surface and in situ artifacts were recovered
in 2008/2009 and 2013 at Nyamita Main (Table 6). Tryon’s
excavations recovered 26 of the in situ pieces and 95 of the
surface collected pieces. The 2013 excavations recovered an
additional 295 artifacts in situ (133 were piece-ploed and
162 were recovered from the sieve) and 33 from the surface,
for a total of 321 in situ pieces and 128 from controlled sur-
face collections. Most of the in situ artifacts were recovered
from a small sand-gravel-lled channel in the main trench
of the excavation.
Lithic Raw Materials at Nyamita Main
Eight raw materials are recognized from the Nyamita Main
excavations: 1) a red and white moled chert; 2) a ne-
grained (aphanitic) lava; 3) a coarse-grained lava; 4) quar;
5) quarite; 6) a green chert; 7) a black chert; and, 8) ‘other’
small pieces of angular debris that cannot be condently
aributed to any of the above categories (see Table 6). The
vast majority of all artifacts from the 2008/2009 and 2013 ex-
TABLE 5. INDIVIDUAL OBSIDIAN ARTIFACTS ANALYZED FOR SOURCING.*
Sample No. SiO2 TiO2 ZrO2 Al2O3 FeO MnO MgO CaO Na2O K2O F Cl Sum
LVP2015-89
7 75.68 0.12 0.05 12.18 1.89 0.05 0.08 0.39 4.70 5.03 0.43 0.15 100.53
Wakondo
Bovid Hill
0.28
0.03
0.04
0.15
0.07
0.01
0.01
0.03
0.21
0.09
0.10
0.01
0.39
LVP2015-90
7
70.44
0.31
0.24
8.49
7.83
0.26
0.06
0.39
7.08
4.63
0.54
0.39
99.78
Kajiei
0.30
0.04
0.05
0.08
0.10
0.04
0.01
0.06
0.26
0.12
0.22
0.01
0.74
LVP2015-91
7
75.96
0.11
0.04
12.10
1.79
0.03
0.06
0.35
4.70
5.18
0.51
0.17
100.53
Onge
0.20
0.04
0.06
0.09
0.05
0.01
0.01
0.05
0.22
0.07
0.21
0.01
0.75
*Mean major and minor element oxides by weight percent. Sample listed on left (No.=number of analyses). One standard deviation from the
mean listed below each element oxide mean.
160 • PaleoAnthropology 2017
Figure 10. A) Plan map of Nyamita Main 2013 excavation with piece-ploed lithic artifacts sorted by raw materials, fossil bones/teeth,
and natural rock color-coded by category. B) Side cross-section of Nyamita Main 2013 excavation (looking west) with piece-ploed
lithic artifacts sorted as before. C) Front cross-section (looking north) of Nyamita Main 2013 excavation with piece-ploed lithic
artifacts sorted as before.
New MSA Evidence from the Eastern Lake Victoria Basin• 161
individuals and all but one of these, a carnassial aribut-
ed to Canis, are bovids (Table 8). The bovids represent at
least seven taxa from four tribes. Alcelaphin antelopes are
most common (62.5% of all large mammal specimens by
NISP) and include the extinct medium alcelaphin related
to wildebeest (Rusingoryx atopocranion), the extinct small al-
celaphin similar to modern blesbok (Damaliscus hypsodon),
extant hartebeest (Alcelaphus buselaphus), and at least two
individuals that could not be diagnosed to a more specic
taxonomic level. The presence of bushbuck in the excavat-
ed sample is consistent with the same taxon and other taxa
indicative of closed vegetation (common duiker) recovered
from the surface collections.
Although sample sizes are small, the relative abun-
dance of alcelaphin specimens among the bovids from the
Nyamita excavation is similar to large samples of faunal
material from the eLVB (Tryon et al. 2010, 2012, 2014), and
is matched in contemporary arid to semi-arid grassland en-
vironments (Alemseged 2003; Vrba 1980). The extinct bo-
vids at Nyamita Main, R. atopocranion and D. hypsodon, are
characterized by exceptional hypsodonty (Faith et al. 2011;
2012). This adaptation is interpreted as suited to consum-
ing grasses in dry and griy environments (Damuth and
Janis 2011; Faith et al. 2012; Marean 1992). However, the
presence of bushbuck (Tragelaphus scriptus) and reedbuck
(Redunca redunca and Redunca cf. arundinum), as well as hip-
popotamus fossils surface collected from the site (Tryon et
al. 2010; 2012; 2014) suggest a local component of closed
from the surface in 2008/2009 (Tryon et al. 2010; 2014) as
well as Levallois points (n=2) made in the red and white
moled chert and lava raw materials that dominate the
lithic assemblage from the 2013 excavations (see Figure 13).
Cores exhibit both the preferential and recurrent Levallois
methods of ake production (see Figure 13).
Fauna of Nyamita Main
Fossil fauna at Nyamita Main derives from and immedi-
ately adjacent to a single ~2m wide by ~1m deep channel
feature encountered along the north–south long axis of the
north trench (see Figure 10). The general shape of this chan-
nel can be observed in the north facing cross-sectional pro-
le of piece-ploed material in the north trench (red dots in
Figure 10c). The channels appear to drain south and west,
similar to the trend of the modern Nyamita Valley. Faunal
material comprised the majority of ploed material in this
trench and was relatively complete with many specimens
comprising entire elements such as hemi-mandibles. No
cutmarks or percussion marks were observed, but surface
preservation is poor. Very lile faunal material was recov-
ered in the Main Trench and what was recovered consisted
mostly of isolated tooth or bone fragments.
All of the vertebrate remains identiable below fam-
ily level are cranio-dental specimens. The taxonomically
identiable sample based on teeth consisted of 16 speci-
mens identiable to the level of tribe or lower (Table 8,
NISP=16, MNI=11). There are a minimum number of eleven
Figure 11. Photographs of Nyamita Main excavations in August of 2013: A) wide-angle photograph (looking east) of excavations
showing North trench on left and Main Trench to the right with total station set up to far right; B) photograph of north wall prole of
North trench showing the stream channel in cross section; C) main trench under excavation viewed looking north.
162 • PaleoAnthropology 2017
clay paleosols. The fauna cannot be directly related to the
artifacts at Nyamita Main. The majority of artifacts occur in
the main trench while almost all fauna was encountered in
the small channel feature in the north trench (see Figures
10a, c; 11b). Further, the fauna on the site cannot be con-
dently shown to be archaeological as no obvious cutmarks
or percussion marks were observed.
habitats and standing water respectively. This agrees with
evidence from tufa and isotopic analyses of paleosol car-
bonates and organic material from the Nyamita Valley
(Beverly et al. 2015a; 2015b; Garre et al. 2015).
The sand-to-gravel-lled channel from which fossil
fauna was recovered is variably cemented by carbonate
and is more resistant to erosion then the surrounding silt-
TABLE 6. SUMMARY OF ASSEMBLAGE COMPOSITION FOR NYAMITA MAIN LITHIC ARTIFACTS.
n
(subtotal)
%
Without
debris
Red/
White
Chert
Lava Coarse
Lava Quartz Quartzite Black
Chert Other
Core
Casual 5 1.85 1 2 1 - 1 - -
Multiplatform 1 0.37 - 1 - - - - -
Levallois
preferential
2 0.74 - 2 - - - - -
Levallois point 0 0.00 - - - - - - -
Levallois recurrent 2 0.74 1 1 - - - - -
Blade 0 0.00 - - - - - - -
Core on flake 0 0.00 - - - - - - -
Core fragment 1 0.37 1 - - - - - -
Core subtotal 11 4.07
Debitage
Initial coritcal flake 0 0.00 - - - - - - -
Residual cortical
flake
0 0.00 - - - - - - -
Levallois flake 2 0.74 - 2 - - - - -
Levallois point 3 1.11 - 2 - - - 1 -
non- cortical flake 71 26.30 32 32 - - 4 3 -
Flake fragment
proximal
81 30.00 38 40 - 1 1 1 -
Flake fragment
other
98 36.30 47 46 - 1 4 - -
Flake subtotal 255 94.44
Debris
Debris and
subtotal
179 83 55 1 6 4 1 29
Tool
point 2 0.74 1 1 - - - - -
side scraper 1 0.37 1 - - - - - -
bifacial point 1 0.37 - 1 - - - - -
Tool subtotal 4 1.48
TOTAL
449
205 185 2 8 14 6 29
n, % without debris 270 100.00
New MSA Evidence from the Eastern Lake Victoria Basin• 163
TABLE 7. SUMMARY STATISTICS OF DIMENSIONS (mm) OF WHOLE FLAKES
FOR THE TWO MOST COMMON RAW MATERIAL TYPES FROM NYAMITA MAIN.
Whole
Flakes
Mean
Flake
Length
(mm)
Mean
Flake
Width
(mm)
Mean
Flake
Thickness
(mm)
Mean
Flake
Platform
Width
(mm)
Mean
Flake
Platform
Thickness
(mm)
Flake
Frags +
AW
Total
Number
of Pieces
Total
Weight
(g)
Lava
27
37.30
27.69
9.11
18.46
7.38
101
128
746.89
±1 sd
22.33 12.36 4.95 9.48 4.99
Red/White
Mottled
Chert
27
18.01
17.10
4.68
10.65
3.79
134
161
306.67
±1 sd
9.98
8.50
3.96
5.44
2.17
Figure 12. Size class distributions of artifacts by raw material category. Experiment totals from Schick (1986).
164 • PaleoAnthropology 2017
Figure 13. Artifacts: A) elongated Levallois point (lava) retouched on both dorsal margins; B) proximal convergent ake or blade
(Levallois point?) made on red/white moled chert; C) Levallois core with overshot preferential ake made on lava (found on surface
on main trench 2013; D) recurrent Levallois or discoid core from Tryon’s 2009 excavation (also see illustration in Tryon et al. 2010);
E) red/ white moled chert Levallois core with recurrent unidirectional removals on debitage surface from Tryon’s 2009 excavations
on surface of main trench.
New MSA Evidence from the Eastern Lake Victoria Basin• 165
ing raw materials from the same sources at the same times.
This does not preclude the possibility that dierent hom-
inin populations maintained dierent technological tradi-
tions in spite of knowledge of alternatives. Such behavior
is ethnographically aested (Wiessner 1983), though on a
much smaller geographical and temporal scale. The data
presented here does suggest that, whatever the reason for
the persistence of the MSA in the eLVB, it was a choice
hominin populations made with likely knowledge of alter-
natives, and not because of ignorance due to geographic
isolation.
Our sample size of archaeological materials in the Late
Pleistocene eLVB remains small. However, it is noteworthy
that all lithic artifacts from excavated contexts in several
time intervals throughout the eLVB are characteristically
MSA. Also, while the sample of sourced obsidians pre-
sented in this and recent studies remains small, all sourc-
ing studies of Pleistocene obsidians from the eLVB over
the past 30 years including the sites of Muguruk, Songhor,
Aringo 3, Kisaaka, Wakondo Bovid Hill, Onge, and Kajiei
unilaterally show the source of archaeological obsidians in
the eLVB was the Naivasha basin (Brown et al. 2013; Faith
et al. 2015; Merrick and Brown 1984; Merrick et al. 1994).
At the very least, our assertion that MSA technologies
in the eLVB persist very late despite consistent contact with
early LSA-making populations, constitutes a reasonable
hypothesis, and one that can be tested by building on work
presented here.
Our research also shows the Late Pleistocene eLVB
stratigraphic sequence can be expanded laterally across an
area >2500km2 and extended chronologically to deposits
older than 100 ka and younger than 36 ka. The geographic
area incorporated by this expanded tephrostratigraphic
framework, encompassing most of the Winam Gulf, also
suggests that this framework will eventually subsume
some or all of the known nearby stratied Late Pleistocene
MSA sites including Rambogo, Randhore, and Muguruk
north of the Winam Gulf, as well as Late Pleistocene and
possibly Middle Pleistocene exposures of Simbi and Song-
DISCUSSION
THE LATE PLEISTOCENE ELVB AND THE
MSA/ LSA TRANSISTION
Data presented here showing MSA technology persisted
over a wide area in the Victoria basin as late as <36 ka
agrees with all prior evidence from these Late Pleistocene
exposures. In seven continuous seasons of survey and ex-
cavation, only a single diagnostic LSA artifact (see Figure
6h) has been recovered and this surface nd is probably
younger than the associated Pleistocene deposits. In con-
trast, hundreds of characteristically MSA tools have been
recovered on the surface and in situ from all chronostrati-
graphically dened intervals between ~100 and 36 ka (Try-
on et al. 2016).
While temporal variation in the end of the MSA and
appearance of the LSA is a phenomenon increasingly rec-
ognized across Africa, East Africa provides a unique venue
in which to study this phenomenon. The late persistence of
the MSA in western Kenya after 50 ka contrasts markedly
with the earliest appearances of LSA or MSA/LSA transi-
tional materials at Mumba and Nasera rock shelters and
the Naisiusiu beds of northern Tanzania as well as Enka-
pune Ya Muto in the central Kenyan rift (Ambrose 1998;
Leakey et al. 1972; Mehlman 1977, 1979, 1989). Additional-
ly, the obsidian sourcing evidence presented here from af-
ter 50 ka suggests MSA-making hominins of the eLVB and
the early LSA-making hominins of northern Tanzania and
Enkapune Ya Muto all shared access to at least one com-
mon area between 100 ka and 36 ka—the Naivasha basin
of the central Kenyan Rift. This is despite the presence of
high-quality obsidian sources closer to sites of the Victoria
basin and northern Tanzania (Brown et al. 2013).
Obsidians could have been transported by trade or
through direct procurement of highly mobile hunter-gath-
erer populations (Ambrose 2012). Regardless, the humans
making MSA or LSA assemblages from dierent parts of
equatorial East Africa would likely have had knowledge
of, or contact with, one another because they were procur-
TABLE 8. NYAMITA MAIN FAUNAL LIST INCLUDING NISP AND MNI COUNTS.
Taxon
NISP
MNI
Canis sp.
1
1
Tragelaphus scriptus
1
1
Redunca cf. redunca
1
1
Redunca cf. arundinum
2
1
Alcelaphus buselaphus
2
2
Damaliscus hypsodon
1
1
Rusingoryx atopocranion
1
1
Alcelaphini indet.
6
2
Ourebia ourebi
1
1
TOTAL
16
11
166 • PaleoAnthropology 2017
geochronology and volcano-tectonic evolution of the
Kedong–Naivasha–Kinangop region, Gregory Rift Val-
ley, Kenya. Journal of the Geological Society 145, 107-116.
Basell, L.S. 2007. Exploration of the Middle Stone Age of east-
ern Africa and excavations at Rambogo Rock Shelter, Kenya.
Ph.D. thesis, University of Cambridge.
Beverly, E.J., Driese, S.G., Peppe, D.J., Arellano, L.N., Ble-
gen, N., Faith, J.T., and Tryon, C.A., 2015a. Reconstruc-
tion of a semi-arid Late Pleistocene paleocatena from
the Lake Victoria region, Kenya. Quaternary Research.
84, 368–381.
Beverly, E.J., Driese, S.G., Peppe, D.J., Johnson, C.R., Mi-
chel, L.A., Faith, J.T., Tryon, C.A., and Sharp, W.D.
2015b. Recurrent spring-fed rivers in a Middle to Late
Pleistocene semi-arid grassland: Implications for envi-
ronments of early humans in the Lake Victoria Basin,
Kenya. Sedimentology 62, 1611–1635.
Beverly, E.J., Peppe, D.J., Driese, S.G., Blegen, N., Faith, J.T.,
Tryon, C.A., and Stinchcomb, G.E. 2017. Reconstruc-
tion of Late Pleistocene paleoenvironments using bulk
geochemistry of paleosols from the Lake Victoria Re-
gion. Frontiers in Earth Science 5, 1–12.
Beyin, A. 2006. The Bab al Mandab vs the Nile-Levant: an
appraisal of the two dispersal routes for early modern
humans out of Africa. African Archaeological Review 23,
5–30.
Blegen, N., Brown, F.H., Jicha, B.R., Binei, K.M., Faith, J.T.,
Ferraro, J.V., Gathogo, P.N., Richardson, J.L., and Try-
on, C.A. 2016. The Menengai Tu: a 36 ka widespread
tephra and its chronological relevance to Late Pleisto-
cene human evolution in East Africa. Quaternary Sci-
ence Reviews 152, 152–168.
Blegen, N., Tryon, C.A., Faith, J.T., Peppe, D.J., Beverly,
E.J., Li, B., and Jacobs, Z., 2015. Distal tephras of the
eastern Lake Victoria basin, equatorial East Africa: cor-
relations, chronology and a context for early modern
humans. Quaternary Science Reviews 122, 89–111.
Borchardt, G., Aruscavage, P., and Millard Jr, H. 1972. Cor-
relation of the Bishop Ash, a Pleistocene marker bed,
using instrumental neutron activation analysis. Journal
of Sedimentary Research 42, 301–306.
Brown, F.H. and Nash, B. 2014. Correlation: volcanic ash,
obsidian. In: Holland, H.D. and Turekian, K.K. (eds.),
Treatise of Geochemistry. Elsevier, Amersterdam, pp.
63–80.
Brown, F.H., Nash, B.P., Fernandez, D.P., Merrick, H.V.,
and Thomas, R.J. 2013. Geochemical composition of
source obsidians from Kenya. Journal of Archaeological
Science 40, 3233–3251.
Chevrier, B., Rasse, M., Lespez, L., Tribolo, C., Hajdas, I.,
Fígols, M.G., Lebrun, B., Leplongeon, A., Camara, A.,
and Huysecom, E. 2016. West African Palaeolithic his-
tory: new archaeological and chronostratigraphic data
from the Falémé valley, eastern Senegal. Quaternary In-
ternational 408, 33–52.
Damuth, J. and Janis, C.M. 2011. On the relationship be-
tween hypsodonty and feeding ecology in ungulate
mammals, and its utility in palaeoecology. Biological
hor in the Nyanza Rift for which tephras are reported (see
Figure 1; McBrearty 1981, 1986, 1991, 1992; Pickford 1982,
1986a, 1986b; Pickford and Thomas 1984). Establishing
chronostratigraphic context and ages for these sites will
provide a larger lateral area, longer timeframe, and larger
sample sizes of archaeological, paleontological, and geo-
logical material. Both the lateral and chronological expan-
sion of the Late Pleistocene eLVB provide improved control
and increased potential for investigating human behavior
though time and across space.
Tephrostratigraphy combined with archaeology has
been used to investigate the Acheulean to MSA transition
in the Middle Pleistocene of equatorial East Africa (Tryon
and McBrearty 2002, 2006), but similar programs either
within or between depositional basins have yet to be insti-
tuted for the MSA/LSA transition in the Late Pleistocene of
the same region. The combination of excavations of open-
air, low-density sites situated in a well-resolved tephro and
chronostratigraphic framework, presented here, suggests a
methodology to further improve our understanding of this
transition.
ACKNOWLEDGMENTS
Fieldwork was conducted under research permits NCST/
RCD/12B/012/2 issued to NB, NCST/5/002/R/576 issued to
CAT, NCST/RCD/ 12B/012/31 issued to JTF, and an explo-
ration and excavation license issued by the National Mu-
seums of Kenya (NMK). Our eldwork is made possible
through the support from the National Geographic Soci-
ety Commiee for Research and Exploration (9284-13 and
8762-10), the National Science Foundation (BCS-1013199,
BCS-1013108, and BCS-0841530), the Leakey Foundation,
Harvard University, the University of Queensland, Bay-
lor University, and the American School for Prehistoric
Research. We would also like to thank Tom Plummer and
the Homa Peninsula Project for donating a tu sample. We
thank Julio Mercader (reviewer) and an anonymous re-
viewer for their helpful comments on a previous version of
this manuscript.
REFERENCES
Alemseged, Z. 2003. An integrated approach to taphonomy
and faunal change in the Shungura Formation (Ethio-
pia) and its implication for hominid evolution. J. Hum.
Evol. 44, 451–478.
Ambrose, S.H. 1998. Chronology of the Later Stone Age
and food production in East Africa. Journal of Archaeo-
logical Science 25, 377–392.
Ambrose, S.H. 2012. Obsidian dating and source exploita-
tion studies in Africa: implications for the evolution of
human behavior. In: Liriis, I. and Stevenson, C. (ed.),
Obsidian and Ancient Manufactured Glasses. University
of New Mexico Press, Albuquerque, pp. 56-72.
Ambrose, S.H., Deino, A., Kyule, M.D., Steele, I., and Wil-
liams, M.A. 2002. The emergence of modern human be-
havior during the Late Middle Stone Age in the Kenya
Rift Valley. Journal of Human Evolution 42, A3–A4.
Baker, B., Mitchell, J., and Williams, L. 1988. Stratigraphy,
New MSA Evidence from the Eastern Lake Victoria Basin• 167
and Later Stone Age deposits at Mumba rockshelter,
Tanzania: optically stimulated luminescence dating of
quar and feldspar grains. Journal of Human Evolution
62, 533–547.
Gramly, R.M. 1976. Upper Pleistocene archaeological oc-
currences at site GvJm/22, Lukenya Hill, Kenya. Man
11, 319–344.
Gramly, R.M. and Rightmire, G.P. 1973. A fragmentary
cranium and dated Later Stone Age assemblage from
Lukenya Hill, Kenya. Man 8, 571–579.
Groucu, H.S., Petraglia, M.D., Bailey, G., Scerri, E.M.,
Parton, A., Clark-Balzan, L., Jennings, R.P., Lewis, L.,
Blinkhorn, J., and Drake, N.A. 2015. Rethinking the dis-
persal of Homo sapiens out of Africa. Evolutionary An-
thropology 24, 149–164.
Hay, R.L. 1976. Geology of the Olduvai Gorge: A Study of Sedi-
mentation in a Semiarid Basin. University of California
Press, Berkeley.
Jenkins, K.E., Nightingale, S., Faith, J.T., Peppe, D.J., Mi-
chel, L.A., Driese, S.G., McNulty, K.P., and Tryon, C.A.
2017. Evaluating the potential for tactical hunting in
the Middle Stone Age: insights from a bonebed of the
extinct bovid, Rusingoryx atopocranion. Journal of Human
Evolution 108, 72–91.
Kent, P.E. 1942. The Pleistocene beds of Kanam and Kanjera,
Kavirondo, Kenya. Geological Magazine 79, 117–132.
Klein, R.G. 2008. Out of Africa and the evolution of human
behavior. Evolutionary Anthropology 17, 267–281.
Kuehn, S.C. and Foit, F.F. 2006. Correlation of widespread
Holocene and Pleistocene tephra layers from New-
berry Volcano, Oregon, USA, using glass compositions
and numerical analysis. Quaternary International 148,
113–137.
Kusimba, S.B. 1999. Hunter–gatherer land use paerns in
Later Stone Age East Africa. Journal of Anthropological
Archaeology 18, 165–200.
Kusimba, S.B. 2001. The early Later Stone Age in East Af-
rica: excavations and lithic assemblages from Lukenya
Hill. African Archaeological Review 18, 77–123.
Leakey, L. and Owen, W. 1945. A Contribution to the Study of
the Tumbian Culture in East Africa. Coryndon Museum
Occasional Papers 1.
Leakey, M.D., Hay, R.L., Thurber, D., Protsch, R., and Berg-
er, R. 1972. Stratigraphy, archaeology, and age of the
Ndutu and Naisiusiu beds, Olduvai Gorge, Tanzania.
World Archaeology 3, 328–341.
Lillesø, J.-P.B., P. van Breugel, R. Kindt, M. Bingham, S.
Demmisew, C. Dudley, I. Friis, F. Gachathi, J. Kalema,
F. Mbago, V. Minani, H.N. Moshi, J. Mulumba, M. Na-
maganda, H.J. Ndangalasi, C.K. Ruo, R. Jamnadass,
and L. Graudal. 2011. Potential natural vegetation of East-
ern Africa (Ethiopia, Kenya, Malawi, Rwanda, Tanzania,
Uganda and Zambia). Volume 1: The Atlas. Working Pa-
per 61, Forest & Landscape, University of Copenhagen,
Denmark.
Manega, P. 1995. New geochronological results from the
Ndutu, Naisiusiu and Ngaloba Beds at Olduvai and
Laetoli in Northern Tanzania: their signicance for
Reviews 86, 733–758.
Danley, P.D., Husemann, M., Ding, B., DiPietro, L.M., Bev-
erly, E.J., and Peppe, D.J. 2012. The impact of the geo-
logic history and paleoclimate on the diversication of
East African cichlids. International Journal of Evolution-
ary Biology 2012, 574851.
Diez-Martín, F., Domínguez-Rodrigo, M., Sánchez, P.,
Mabulla, A.Z., Tarriño, A., Barba, R., Prendergast, M.E.,
and de Luque, L. 2009. The Middle to Later Stone Age
technological transition in East Africa. New data from
Mumba Rockshelter Bed V (Tanzania) and their impli-
cations for the origin of modern human behavior. Jour-
nal of African Archaeology 7, 147–173.
Eren, M.I., Diez-Martin, F., and Dominguez-Rodrigo, M.
2013. An empirical test of the relative frequency of bi-
polar reduction in Beds VI, V, and III at Mumba Rock-
shelter, Tanzania: implications for the East African
Middle to Late Stone Age transition. Journal of Archaeo-
logical Science 40, 248–256.
Faith, J.T. 2014. Late Pleistocene and Holocene mammal
extinctions on continental Africa. Earth-Science Reviews
128, 105–121.
Faith, J.T., Choiniere, J.N., Tryon, C.A., Peppe, D.J., and
Fox, D.L. 2011. Taxonomic status and paleoecology of
Rusingoryx atopocranion (Mammalia, Artiodactyla),
an extinct Pleistocene bovid from Rusinga Island, Ke-
nya. Quaternary Research 75, 697–707.
Faith, J.T., Pos, R., Plummer, T.W., Bishop, L.C., Marean,
C.W., and Tryon, C.A. 2012. New perspectives on mid-
dle Pleistocene change in the large mammal faunas of
East Africa: Damaliscus hypsodon sp. nov.(Mammalia,
Artiodactyla) from Lainyamok, Kenya. Palaeogeography,
Palaeoclimatology, Palaeoecology 361, 84–93.
Faith, J.T., Tryon, C.A., Peppe, D.J., Beverly, E.J., and Ble-
gen, N. 2014. Biogeographic and evolutionary impli-
cations of an extinct Late Pleistocene impala from the
Lake Victoria Basin, Kenya. Journal of Mammalian Evolu-
tion 21, 213–222.
Faith, J.T., Tryon, C.A., Peppe, D.J., Beverly, E.J., Blegen,
N., Blumenthal, S., Chri, K.L., Driese, S.G., and Pat-
terson, D. 2015. Paleoenvironmental context of the
Middle Stone Age record from Karungu, Lake Victoria
Basin, Kenya, and its implications for human and fau-
nal dispersals in East Africa. Journal of Human Evolution
83, 28–45.
Frogga, P. 1992. Standardization of the chemical analysis
of tephra deposits. Report of the ICCT working group.
Quaternary International 13, 93–96.
Gabel, C. 1969. Six rock shelters on the northern Kavirondo
shore of Lake Victoria. African Historical Studies 2, 205-
254.
Garre, N.D., Fox, D.L., McNulty, K.P., Faith, J.T., Peppe,
D.J., Van Plantinga, A., and Tryon, C.A. 2015. Stable
isotope paleoecology of Late Pleistocene Middle Stone
Age humans from the Lake Victoria basin, Kenya. Jour-
nal of Human Evolution 82, 1–14.
Gliganic, L.A., Jacobs, Z., Roberts, R.G., Domínguez-Rodri-
go, M., and Mabulla, A.Z. 2012. New ages for Middle
168 • PaleoAnthropology 2017
Merrick, H.V., Brown, F.H., and Nash, W.P. 1994. Use and
movement of obsidian in the Early and Middle Stone
Ages of Kenya and northern Tanzania. In Society, Cul-
ture, and Technology in Africa, S.T. Childs (ed.), MASCA
11 (supplement), pp. 29–44.
O’Brien, H.D., Faith, J.T., Jenkins, K.E., Peppe, D.J., Plum-
mer, T.W., Jacobs, Z.L., Li, B., Joannes-Boyau, R., Price,
G., Feng, Y. and Tryon, C. 2016. Unexpected conver-
gent evolution of nasal domes between Pleistocene
bovids and Cretaceous Hadrosaur dinosaurs. Current
Biology 26, 503–508.
Osypiński, P. and Osypińska, M. 2016. Optimal adjustment
or cultural backwardness? New data on the latest Lev-
allois industries in the Nile Valley. Quaternary Interna-
tional 408, 90–105.
Owen, A.W. 1939. An amateur eld collector in Kavirondo.
African Aairs 38, 220–226.
Owen, W. 1937. Draft of tentative preliminary report on the
July 1937 investigations at Ng’ira Karungu. Unpub-
lished manuscript housed in the archives of the Natu-
ral History Museum, London.
Owen, W. 1938. 218. The Kombewa Culture, Kenya Colony.
Man 38, 203–205.
Peppe, D.J., McNulty, K.P., Cote, S.M., Harcourt-Smith,
W.E., Dunsworth, H.M., and Van Couvering, J.A. 2009.
Stratigraphic interpretation of the Kulu Formation
(Early Miocene, Rusinga Island, Kenya) and its impli-
cations for primate evolution. Journal of Human Evolu-
tion 56, 447–461.
Pickford, M. 1982. The tectonics, volcanics and sediments
of the Nyanza Rift Valley, Kenya. Zeitschrift für Geomor-
phologie 42, 1–33.
Pickford, M. 1984. Kenya Palaeontology Gazeeer. National
Museums of Kenya, Department of Sites and Monu-
ments Documentation, Nairobi, Kenya.
Pickford, M. 1986a. Cainozoic paleontological sites of West-
ern Kenya. Münchner Geowissenschaftlich Abhandlungen
8, 1-151.
Pickford, M. 1986b. Sedimentation and fossil preservation
in the Nyanza Rift System, Kenya. Geological Society,
London, Special Publications 25, 345–362.
Pickford, M. and Thomas, H. 1984. An aberrant new bo-
vid (Mammalia) in subrecent deposits from Rusinga
island, Kenya, Proceedings of the Koninklke Nederlandse
Akademie van Wetenschappen. Series B. Palaeontology, Ge-
ology, Physics and Chemistry 87, pp. 441–452.
Prendergast, M.E., Luque, L., Domínguez-Rodrigo, M.,
Diez-Martín, F., Mabulla, A.Z., and Barba, R. 2007. New
excavations at Mumba rockshelter, Tanzania. Journal of
African Archaeology 5, 217–244.
Ramsey, C.B. 2009. Bayesian analysis of radiocarbon dates.
Radiocarbon 51, 337–360.
Reimer, P.J., Bard, E., Bayliss, A., Beck, J.W., Blackwell,
P.G., Ramsey, C.B., Buck, C.E., Cheng, H., Edwards,
R.L., and Friedrich, M. 2013. IntCal13 and Marine13 ra-
diocarbon age calibration curves 0–50,000 years cal BP.
Radiocarbon 55, 1869–1887.
Rito, T., Richards, M.B., Fernandes, V., Alshamali, F., Cerny,
evolution of modern humans. Preservation and Use of
Olduvai Gorge, Laetoli, Rock Art, and other Palaeoan-
thropological Resources in Tanzania, in Bellagio Con-
ference, Italy.
Manega, P.C. 1993. Geochronology, geochemistry and iso-
topic study of the Plio-Pleistocene hominid sites and
the Ngorongoro volcanic highland in northern Tanza-
nia. Ph.D. thesis. University of Colorado, Boulder.
Marean, C.W. 1992. Implications of late Quaternary mam-
malian fauna from Lukenya Hill (south-central Kenya)
for paleoenvironmental change and faunal extinctions.
Quaternary Research 37, 239–255.
Marks, A.E. and Conard, N.J. 2008. Technology vs. typol-
ogy: the case for and against a transition from the
MSA to the LSA at Mumba Cave, Tanzania. In Space
and Time: Which Diachronies, Which Synchronies, Which
Scales? Typology vs Technology. T. Aubry, F. Almeida, A.
C. Araujo, and M. Tiagom (eds). British Archaeologi-
cal Reports International Series 1831. Archaeopress,
Oxford, pp. 123–131.
McBrearty, S. 1981. Songhor: a Middle Stone Age site in
western Kenya. Quaternaria 23, 171–190.
McBrearty, S. 1986. The Archaeology of the Muguruk Site,
Western Kenya. Ph.D. thesis. University of Illinois.
McBrearty, S. 1991. Recent research in western Kenya and
its implications for the status of the Sangoan indus-
try. In Cultural Beginnings: Approaches to Understanding
Early Hominid Lifeways in the African Savanna, J.D. Clark
(ed.). Römisch-Germanisches Zentralmuseum, Forsc-
hunginstitut für Vor- und Frühgeschichte, Monogra-
phien 19, Bonn, pp. 159–176.
McBrearty, S. 1992. Sangoan technology and habitat at Sim-
bi. Nyame Akuma 38, 34–40.
McBrearty, S. and Brooks, A.S. 2000. The revolution that
wasn’t: a new interpretation of the origin of modern
human behavior. Journal of Human Evolution 39, 453–
563.
McDougall, I., Brown, F.H., and Fleagle, J.G. 2005. Strati-
graphic placement and age of modern humans from
Kibish, Ethiopia. Nature 433, 733–736.
McPherron, S.P. and Dibble, H.L. 2002. Using Computers in
Archaeology: A Practical Guide. McGraw-Hill, New York.
Mehlman, M.J. 1977. Excavations at Nasera Rock, Tanzania.
Azania 12, 111–118.
Mehlman, M.J. 1979. Mumba-Höhle revisited: the relevance
of a forgoen excavation to some current issues in east
African prehistory. World Archaeology 11, 80–94.
Mehlman, M.J. 1989. Later Quaternary Archaeological Se-
quences in Northern Tanzania. Ph.D. Dissertation. Uni-
versity of Illinois at Urbana-Champaign.
Mercader, J., Gosse, J.C., Benne, T., Hidy, A.J., and Rood,
D.H. 2012. Cosmogenic nuclide age constraints on
Middle Stone Age lithics from Niassa, Mozambique.
Quaternary Science Reviews 47, 116–130.
Merrick, H.V. and Brown, F.H. 1984. Obsidian sources and
paerns of source utilization in Kenya and northern
Tanzania: some initial ndings. African Archaeological
Review 2, 129–152.
New MSA Evidence from the Eastern Lake Victoria Basin• 169
Tryon, C.A. and McBrearty, S. 2002. Tephrostratigraphy
and the Acheulian to Middle Stone Age transition in
the Kapthurin formation, Kenya. Journal of Human Evo-
lution 42, 211–235.
Tryon, C.A. and McBrearty, S. 2006. Tephrostratigraphy of
the Bedded Tu Member (Kapthurin Formation, Ke-
nya) and the nature of archaeological change in the lat-
er Middle Pleistocene. Quaternary Research 65, 492–507.
Tryon, C.A., McBrearty, S., and Texier, P.-J. 2005. Levallois
lithic technology from the Kapthurin Formation, Ke-
nya: Acheulian origin and Middle Stone Age diversity.
African Archaeological Review 22, 199–229.
Tryon, C.A., Peppe, D.J., Faith, J.T., Van Plantinga, A.,
Nightingale, S., Ogondo, J., and Fox, D.L. 2012. Late
Pleistocene artefacts and fauna from Rusinga and
Mfangano islands, Lake Victoria, Kenya. Azania: Ar-
chaeological Research in Africa 47, 14–38.
Van Couvering, J.A. 1972. Geology of Rusinga Island and Cor-
relation of the Kenya Mid-Tertiary Fauna. Ph.D. thesis,
University of Cambridge.
Van Plantinga, A.A. 2011. Geology of the Late Pleistocene
artifact-bearing Wasiriya Beds at the Nyamita locality,
Rusinga Island, Kenya. M.A. thesis, Baylor University,
Waco, Texas.
Villa, P., Soriano, S., Tsanova, T., Degano, I., Higham, T.F.,
d’Errico, F., Backwell, L., Lucejko, J.J., Colombini, M.P.,
and Beaumont, P.B. 2012. Border Cave and the begin-
ning of the Later Stone Age in South Africa. Proceed-
ings of the National Academy of Sciences USA 109, 13208–
13213.
Vrba, E.S. 1980. 14. The signicance of bovid remains as in-
dicators of environment and predation paerns. In Fos-
sils in the Making: Vertebrate Taphonomy and Paleoecology,
A.K. Behrensmeyer and A.H. Hill (eds.). University of
Chicago Press, Chicago, pp. 247–271.
Wadley, L. 2001. What is cultural modernity? A general
view and a South African perspective from Rose Cot-
tage Cave. Cambridge Archaeological Journal 11, 201–221.
White, F. 1983. The Vegetation of Africa, a Descriptive Memoir
to Accompany the UNESCO/AETFAT/UNSO Vegetation
Map of Africa (3 Plates, Northwestern Africa, Northeast-
ern Africa, and Southern Africa, 1: 5,000,000). Unesco,
Paris.
Wiessner, P. 1983. Style and social information in Kalahari
San projectile points. American Antiquity 48, 253–276.
Wright, D.K., Thompson, J., Mackay, A., Welling, M., For-
man, S.L., Price, G., Zhao, J.X., Cohen, A.S., Malijani,
O., and Gomani-Chindebvu, E. 2014. Renewed geoar-
chaeological investigations of Mwanganda’s village
(elephant butchery site), Karonga, Malawi. Geoarchaeol-
ogy 29, 98–120.
Wright, D.K., Thompson, J.C., Schilt, F., Cohen, A.S., Choi,
J.-H., Mercader, J., Nightingale, S., Miller, C.E., Men-
er, S.M., and Walde, D. 2017. Approaches to Middle
Stone Age landscape archaeology in tropical Africa.
Journal of Archaeological Science 77, 64–77.
V., Pereira, L., and Soares, P. 2013. The rst modern hu-
man dispersals across Africa. PLoS One 8, e80031.
Robertshaw, P., Colle, D., Giord, D., and Mbae, N.B.
1983. Shell middens on the shores of Lake Victoria.
Azania: Journal of the British Institute in Eastern Africa 18,
1–43.
Rochowicz, Jr., J.A. 2010. Bootstrapping analysis, infer-
ential statistics and EXCEL. Spreadsheets in Education
(eJSiE) 4, 4.
Scerri, E.M., Blinkhorn, J., Groucu, H.S., and Niang, K.
2015. The Middle Stone Age archaeology of the Senegal
River Valley. Quaternary International 408, 16–32.
Schick, K.D. 1986. Stone Age Sites in the Making: Experiments
in the Formation and Transformation of Archaeological Oc-
currences. British Archaeological Reports, Oxford.
Shea, J.J. 2008. The Middle Stone Age archaeology of the
Lower Omo Valley Kibish Formation: excavations, lith-
ic assemblages, and inferred paerns of early Homo sa-
piens behavior. Journal of Human Evolution 55, 448–485.
Skinner, A., Hay, R., Masao, F., and Blackwell, B. 2003. Dat-
ing the Naisiusiu Beds, Olduvai Gorge, by electron spin
resonance. Quaternary Science Reviews 22, 1361–1366.
Tribolo, C., Asrat, A., Bahain, J.-J., Chapon, C., Douville, E.,
Fragnol, C., Hernandez, M., Hovers, E., Leplongeon,
A., and Martin, L. 2017. Across the gap: geochronologi-
cal and sedimentological analyses from the Late Pleis-
tocene-Holocene sequence of Goda Buticha, southeast-
ern Ethiopia. PloS One 12, e0169418.
Tryon, C.A., Crevecoeur, I., Faith, J.T., Ekshtain, R., Niv-
ens, J., Paerson, D., Mbua, E.N., and Spoor, F. 2015.
Late Pleistocene age and archaeological context for the
hominin calvaria from GvJm-22 (Lukenya Hill, Kenya).
Proceedings of the National Academy of Sciences USA 112,
2682–2687.
Tryon, C.A. and Faith, J.T. 2013. Variability in the Middle
Stone Age of Eastern Africa. Current Anthropology 54,
S234–S254.
Tryon, C.A. and Faith, J.T. 2016. A demographic perspec-
tive on the Middle to Later Stone Age transition from
Nasera rockshelter, Tanzania. Philosophical Transactions
of the Royal Society B 371, 20150238.
Tryon, C.A., Faith, J.T., Peppe, D.J., Beverly, E.J., Blegen, N.,
Blumenthal, S.A., Chri, K.L., Driese, S.G., Paerson,
D., and Sharp, W.D. 2016. The Pleistocene prehistory
of the Lake Victoria basin. Quaternary International 404,
100–114.
Tryon, C.A., Faith, J.T., Peppe, D.J., Fox, D.L., McNulty,
K.P., Jenkins, K., Dunsworth, H., and Harcourt-Smith,
W. 2010. The Pleistocene archaeology and environ-
ments of the Wasiriya Beds, Rusinga Island, Kenya.
Journal of Human Evolution 59, 657–671.
Tryon, C.A., Faith, J.T., Peppe, D.J., Keegan, W.F., Keegan,
K.N., Jenkins, K., Nightingale, S., Paerson, D., Van
Plantinga, A., and Driese, S.G. 2014. Sites on the land-
scape: paleoenvironmental context of late Pleistocene
archaeological sites from the Lake Victoria basin, equa-
torial East Africa. Quaternary International 331, 20–30.
