ArticlePDF Available

Pleistocene Cascade Tufas of Kaokoland, Namibia. Communications of the Geological Survey of Namibia


Abstract and Figures

Cascade tufas are common in Kaokoland, especially in the region north of Oruvandje, and there are impressive examples at Otjitamei and Ojtikondavirongo, yet they have not previously been reported in the literature. The Damaraland tufas further south at Ongongo, near Warmquelle, were described recently, and proved to be of interest on account of their fossil content (plants, gastropods, and a possible frog skeleton). The Kaokoland tufas are more numerous than those in Damaraland and are also highly fossiliferous, containing not only abundant plants and gastropods, but also vertebrates (fish, snakes, lizards, birds, mammals). The mammals are important because they indicate that some of the breccia infilling cavities in the tufas are probably of Late Pliocene and Early Pleistocene age, the first time that the age of Namibian tufas has been reasonably well determined. The geomorphological relationships of the tufa lobes reveal that they span a considerable period of time, some of the older eroded lobes probably being of Late Miocene age, overlain by Pliocene and Pleistocene tufas. Some of the breccias contain large mammal bones and teeth associated with primitive stone tools. The aims of this paper are to document the impressive tufa lobes in Kaokoland, to put on record the discovery of fossil invertebrates and vertebrates within them and to discuss the significance of the fossils for bio-chronology and palaeoecology.
Content may be subject to copyright.
Fossiliferous Plio-Pleistocene Cascade Tufas of Kaokoland, Namibia
Martin Pickford
, Helke Mocke
, Brigitte Senut
, Loïc Ségalen
and Pierre Mein
1. CR2P/UMR 7207 – Sorbonne Universités – MNHN, CNRS, UPMC, Université Paris VI, T.46-56,
E.5, case 104, 4 Place Jussieu, 75252 Paris cedex 05, France (e-mail: <>)
2. Geological Survey of Namibia, Aviation Road, Windhoek. Namibia
(e-mail: <>)
3. 114 bis, rue Hénon, 69004, Lyon, France (e-mail : <>).
Abstract: Cascade tufas are common in Kaokoland, especially in the region north of Oruvandje, and
there are impressive examples at Otjitamei and Ojtikondavirongo, yet they have not previously been
reported in the literature. The Damaraland tufas further south at Ongongo, near Warmquelle, were de-
scribed recently, and proved to be of interest on account of their fossil content (plants, gastropods, and
a possible frog skeleton). The Kaokoland tufas are more numerous than those in Damaraland and are
also highly fossiliferous, containing not only abundant plants and gastropods, but also vertebrates
(fish, snakes, lizards, birds, mammals). The mammals are important because they indicate that some of
the breccia infilling cavities in the tufas are probably of Late Pliocene and Early Pleistocene age, the
first time that the age of Namibian tufas has been reasonably well determined. The geomorphological
relationships of the tufa lobes reveal that they span a considerable period of time, some of the older
eroded lobes probably being of Late Miocene age, overlain by Pliocene and Pleistocene tufas. Some of
the breccias contain large mammal bones and teeth associated with primitive stone tools. The aims of
this paper are to document the impressive tufa lobes in Kaokoland, to put on record the discovery of
fossil invertebrates and vertebrates within them and to discuss the significance of the fossils for bio-
chronology and palaeoecology.
Key Words: Tufa Lobes; Breccia; Plants; Gastropods; Mammals; Plio-Pleistocene; Namibia; Kaokoland.
To cite this article: Pickford, M. Mocke, H. Senut, B. Ségalen, L. & Mein, P. 2016. Fossiliferous Plio-
Pleistocene Cascade Tufas of Kaokoland, Namibia. Communications of the Geological Survey of Namibia, 17,
Submitted December 2015
Tufa deposits have long been known to
occur in the vicinity of Sesfontein and
Warmquelle in Damaraland, Namibia, alt-
hough not a great deal has been published
about them (Korn & Martin, 1937, 1955;
Mocke, 2014). In April, 2013, MP was alerted
to the presence of « peculiar » rocks near
Oruvandje further north in Kaokoland, by Dr
E. Freyer, Windhoek. A preliminary survey
was carried out by MP and HM in November,
2013, which resulted in the identification of an
abundance of cascade tufa deposits east and
northwest of Oruvandje. All the deposits ex-
amined proved to be fossiliferous, notably con-
taining plant remains and land snails, but at
Okongwe, east of Okozonduno, vertebrate-
bearing sandy limestone was found infilling a
cavity in the tufa. Extraction of the fossils from
2 kg of breccia resulted in a crop of several
dozen rodent teeth, as well as a few bat teeth,
frog bones, snake vertebrae and bird bones.
The rodents indicate a Late Pliocene age for
the breccia, but it is clear from the field rela-
tions that tufa deposition took place over a
considerable period of time, with many car-
bonate lobes being formed at various altitudes
and at various times within each valley. Cavi-
ties within the waterfall tufas often contain
speleothems as well as silty, sandy and pebbly
breccia infillings of diverse ages. This find was
sufficiently interesting to prompt a further sur-
This article reports the results obtained
so far, but it is stressed that the quantity of tufa
deposits is so vast that it will take many years
of research to study them all in detail.
Figure 1. Relief map of Namibia showing the area of Kaokoland Tufa deposits (black rectangle at Oruvandje on
the southern flanks of the Joubert Mountains) west of Etosha Pan (dotted ovoid outline) and in the Naukluft
Mountains. Note the positions of the tufa deposits at high altitudes along the edge of the «Great Escarpment»
which separates the deeply incised coastal strip, from the flatter interior of the continent.
Material and Methods
Prior to field survey, identification of
tufa deposits in the Oruvandje area was carried
out using Google Earth. The main tufa deposits
are easily spotted using satellite imagery be-
cause of the clearly unconformable relation-
ship that they have with the underlying, heavi-
ly folded Precambrian strata. By this means,
the cliff-forming carbonate lobes at Omatapati,
Okongwe, Okapika, Orutjene and Okovanatje
were located at the mouths of streams flowing
generally southwards from the Omatapati-
Otjozongombe Carbonate Massif. Visits to
Omatapati, Okongwe and Okapiku in 2013
confirmed their tufa nature. Further searches
using Google Earth resulted in the identifica-
tion of additional deposits at Otjitamei and
Otjikondavirongo as well as near Ongongo,
north of Warmquelle, all confirmed as tufa
lobes by ground control.
Field surveys comprised clambering
over the tufa lobes searching for micromam-
malian remains and gastropods. By this means
several mammal-bearing breccia occurrences
were discovered at Okongwe, Omatapati and
Otjitamei. Gastropods and plants remains are
present at all the deposits surveyed. At Omata-
pati, some of the breccias are rich in large
mammal remains associated with stone tools of
various sorts.
Each fossil occurrence found was lo-
calised using GPS set to WGS 84. Sample
blocks for palaeontological analysis were kept
separate from each other in order to avoid mix-
ing faunas.
Breccia was dissolved in formic acid at 7%
concentration, buffered by calcium triphos-
phate. After extraction, the insoluble residues
were examined under a binocular microsope,
and fossils found were consolidated using a
dilute solution of glyptol dissolved in acetone.
Images of fossils were captured using a
digital camera with the lense positioned over
the eyepieces of the microscope. Images were
enhanced using Photoshop Elements 03.
Geological and Geomorphological Settings of the Cascade Tufas of Kaokoland
Figure 2. Distribution of the Otavi Dolomites and other Proterozoic carbonates (black areas) in Na-
mibia and Angola. The southern flanks of the Joubert Mountains and the Naukluft Mountains were the
sites of extensive tufa deposition during the Late Neogene and Quaternary.
Figure 3. Oblique relief map of Namibia showing the distribution of dolomites (light blue) and exten-
sive tufa deposits (dark blue). The tufa deposits formed preferentially at high altitudes along the edges
of the Great Escarpment indicating the essential roles that climate and bedrock substrate played in
their genesis.
Figure 4. The main areas in Kaokoland, north and east of Sesfontein, at which cascade tufas have been mapped
(Map modified from Google Earth).
The cascade tufas of Kaokoland are
spatially associated with Precambrian dolo-
mites and limestones of the Otavi Group. They
generally occur at the mouths of streams that
flow from the upland dolomitic areas to low-
lying country beneath. Several of the occur-
rences show three or more tufa masses at dif-
ferent altitudes in the same valley, and most of
them show that, at each level, two or more
lobes of carbonate were built up as the lime-
rich waters were diverted from one lobe to start
building up another one nearby.
Figure 5. Stereo oblique views of cascade tufa lobes in Kaokoland showing the characteristic geomorphology
that they produce. A) Okapiku (3 lobes), B) Okongwe (4 lobes), C) Orutjene (2 lobes) and Okovanatje (1 lobe),
D) Omatapati (2 lobes of very different ages), E) Otjimatei (1 huge lobe), F) Okongotirwa (2 lobes), and G)
Otjikondavirongo (2 lobes of different ages). The blue dots follow the contact between carbonate rocks above
and relatively impervious sericitic rock below. The blue ovals encircle resurgences upstream from the tufa lobes.
The dimensions, co-ordinates and altitudes of the lobes are provided in Table 1 (images modified from Google
A common theme of the Kaokoland tu-
fa lobes is that they occur near the exposed
contact between impervious rocks beneath (se-
ricitic, fine-grained rocks at Omatapati,
Okongwe and Okapika, for example, compris-
ing the Swakop Schist) which are overlain by
folded, well-bedded carbonate rocks (Otavi
Dolomite) (van der Merwe, 1983). A further
common theme is that a short distance up-
stream from several of the occurrences, there
appears to be an ancient resurgence (some-
times still active) where lime-charged waters
emerged to the surface. It is thus likely that the
ancient water table at the time of lobe growth,
was considerably higher than it is today, and
that phreatic water was the main agent which
dissolved the Proterozoic carbonate, transport-
ed it in solution, and then brought it to the an-
cient land surface where much of it was precip-
itated as a byproduct of aquatic plant photo-
synthesis (mainly mosses) and as a result of
exposure to warm air. Under the present-day
water table regime, the lobes are undergoing
erosion, or only very slight growth compared
to the intensity of activity that occurred when
the water table was high. On the basis of the
observation that there are two or three genera-
tions of extinct lobes at some of the sites
(Okongwe, Okapiku, Omatapati) it is deduced
that there occurred alternating phases of active
lobe construction on the one hand and of lobe
destruction, erosion or absence of growth, on
the other. From this it is concluded that the
ancient water table fluctuated in height de-
pending on local to regional climatic change.
The position of the tufa system at high alti-
tudes along the edges of the «Great Escarp-
ment» was probably crucial from the palaeo-
climatic point of view, with alternating phases
of lobe growth or lobe destruction being large-
ly controlled by changes in rainfall and/or
Table 1. Surface dimensions of cascade tufa lobes in Kaokoland, Damaraland and the Naukluft Mountains, Na-
mibia (+ means that the lobe is partly covered by a later lobe or has been eroded) (Thicknesses are not provided
but most lobes are 15 30 metres thick along the cliff exposures, sometimes well over 50 metres, thinning up-
stream and laterally. Latitude, Longitude and Altitude are from Google Earth).
Local name Lobe N° Latitude S Longitude E Altitude (m) Width across (m) Length (m)
Okapiku 1 18°53’30.8’’ 14°03’18.6’’ 1218 220 140
---------- 2 18°53’21.6’’ 14°03’23.6’’ 1317 320 250
---------- 3 18°53’17.0’’ 14°03’25.5’’ 1359 490 370
Okongwe 1 18°53’43.0’’ 14°04’32.1’’ 1350 200 50
----------- 2 18°53’40.5’’ 14°04’23.6’’ 1338 320 180
----------- 3 18°53’35.9’’ 14°04’17.1’’ 1321 120 80
----------- 4 18°53’51.6’’ 14°04’16.1’’ 1221 330 160
Orutjene 1 18°51’38.6’’ 14°02’27.6’’ 1346 230 170+
---------- 2 18°51’34.2’’ 14°02’29.1’’ 1363 190 180
Okovanatje 1 18°52’15.5’’ 14°02’43.5’’ 1319 220 130
Omatapati 1 18°54’01.8’’ 14°08’50.2’’ 1137 700 670
------------- 2 18°53’35.3’’ 14°08’53.7’’ 1220 400 340
Otjitamei 1 18°51’53.9’’ 13°44’56.9’’ 1306 810 600
Okongotirwa 1 18°49’31.1’’ 14°14’48.8’’ 1190 110 60
---------------- 2 18°49’27.5’’ 14°14’31.9’’ 1220 460 300
Otjikondavirongo 1 18°44’43.8’’ 13°33’10.5’’ 1141 130+ 190+
--------------------- 2 18°44’37.1’’ 13°33’08.0’’ 1165 310 370
Ongongo SW 1 19°09’14.0’’ 13°49’04.1’’ 772 230 150
Ongongo SE 1 19°09’15.4’’ 13°49’22.9’’ 797 350 120
Bleskranz 1 24°08’01.5’’ 16°14’00.0’’ 1443 300 300
Under the current water table condi-
tions, there is only very minor active tufa dep-
osition at some sites (Okovanatje, Okapiku,
Otjikondavirongo) which is related more to
surface waters (following the rains) locally
redistributing carbonates derived from the
lobes or upstream Neogene carbonate-rich val-
ley infillings, than to recharge from subterra-
nean waters flowing through Proterozoic coun-
try rock. A result of this superficial activity is
the development of epikarst features on the
surface of the lobes (rillenkarren, clints, grikes)
(Pickford & Senut, 2010).
The tufa masses that resulted from the
build-up of carbonate lobes are not homogene-
ous, but are full of cavities of various dimen-
sions. Occasionally, masses of tufa (bryophyte
curtains, for example) broke off from the lobe
fronts under their own weight and would ac-
cumulate at the base of the lobe, adding to the
complexity of the depositional succession. Va-
dose water seeping underground through the
lobes resulted in their thorough cementation by
calcite and the deposition of speleothems in
cavities. By this means huge volumes of dense-
ly cemented tufa were formed.
Cavities in the tufa acted like caves,
and frequently show classic speleothem depo-
sition of stalagmites, stalactites, flowstone and
cave pearls in gour pools. If the cavities were
open to the surface, then they provided roost-
ing places for bats and owls, and if accessible
were (and still are) used as lairs by carnivores,
baboons, klipspringers, dassies and other ani-
mals (land gastropods, agama lizards, snakes).
The carbonate lobes were vegetated, as
shown by the quantities of leaves, stems and
root systems that they preserve. Most of the
tufas show that mosses were the dominant
plants associated with their growth, producing
phytoherms (cushions, curtains) with layered
structures, each layer of which has a cellular
fabric (Pedley et al. 2003). Other plants asso-
ciated with the tufas are sedges, reeds and trees
of various sorts, including Figs (Ficus sycomo-
rus) and Mopane (Colophospermum mopane)
(Mocke, 2014).
The tufa lobes tended to block the
streams in which they were growing, and up-
stream from the resulting «dams» or «barrag-
es» there are masses of calcified conglomer-
ates, sands and silts forming lobe-top terraces
and valley infillings. This gives rise to a char-
acteristic geomorphological signature, in
which the tufa lobe front is a cliff facing
downstream, behind which is a flat or gently
sloping, roughly triangular terrace-like area,
broad at the downstream end and narrowing
The presence of cobbles and pebbles in
the lobe front, often caught up in tufa, attests to
the fact that from time to time particularly en-
ergetic surface water flow occurred, which
transported cobbles and boulders over the
lobes and into the downstream sectors of the
systems where they were often calcified, on
occasion forming impressive deposits of ce-
mented colluvial and fluvial gravels. There is
evidence of phases of active cut and fill up-
stream and downstream from the cascade tufa
lobes as well as within the lobes themselves.
The tufa lobes are always accompanied
upstream and downstream by calcified fluvial
deposits in the valley bottoms, sometimes ex-
tending right across the valleys, sometimes
eroded to form terrace-like cliffs through
which the present day streams flow for a few
days after rain. Fans often formed at the ends
of valleys where the streams debouch onto
plains and these are generally calcified into
cohesive aprons, sometimes incised by subse-
quent stream activity. Further downstream,
paludal tufas have been deposited, for exam-
ple, near the Camel Camp at Sesfontein. At
Otjomatemba, on the terrace upstream from the
Otjitamei Tufa Lobe there is an extensive cov-
er of paludal tufa half a metre thick, underlain
by poorly consolidated silt. At both sites the
paludal tufas are porous and rich in plant re-
mains, and are intercalated in soft marls and
silts. In general though, the paludal tufas close
to the cascade lobes were intensively calcified
and are thus densely cemented.
Figure 6. Locations of the main tufa lobes along the southern flanks of the dolomite massive exposed between
Omatapati and Orutjene (image modified from Google Earth).
Close examination of the carbonate
massif north of Oruvandje, between Omatapati
and Otjozongombe, reveals that there are many
small occurrences of tufa in almost all the val-
leys. These could represent remnants of an-
cient tufa lobes that have been eroded, or may
have been lobes that for one reason or another
(climatic change, not enough carbonate-rich
water, their confined position within the val-
ley) stopped growing while they were still
small. They are comparable to the « Barrage
Tufas » recorded in the Naukluft Mountains
(Stone et al. 2010; Viles et al. 2007; Goudie &
Viles, 2015). Each of these occurrences, of
which there are many dozens, need to be inves-
tigated to determine their content and origins.
There are karstic features such as
caves, dolines, sink holes and rillenkarren in
the Precambrian dolomites and limestones of
the area, but the larger karst features are not
numerous, nor very obvious, apart from a cave
near Robbie’s Pass, 9 km north of Otjikondavi-
rongo (18°40’05.6’’S : 13°32’13.4’’E, alti-
tude : 1562 m). The Plio-Pleistocene tufa
lobes, in contrast, show an abundance of local-
ised epikarst activity such as klints, grikes and
rillenkarren, and where they are eroded or in-
cised, subterranean karst features such as spe-
leothem deposition (stalagmites, flowstone)
and cave pearls which formed in gour pools,
can be observed.
The Kaokoland Tufa Lobes
The Okozonduno-Oruvandje area
Figure 7. The Okongwe Tufa Lobe Cluster viewed from the northwest, showing the development of lobes at
different altitudes. Lobe N° 1 is associated with a small spring to the south of the main valley in which the other
three lobes formed. Fossil mammals were found in breccia infilling former cavities in lobe N° 4 (the base of
which is obscured from view by the trees).
Figure 8. The lowermost tufa lobe (N° 4) in the Okongwe cluster, with the collapsed cave, viewed from the
southwest. Fossil mammals occur in breccia inside the collapsed cave as well as in subsidiary infillings of other
caves on the north flank of the lobe (left in the image).
Figure 9. A) A large fragment of bryophyte curtain tufa at the foot of Okongwe Lobe 4, near the cave. Note the
alternating layers of structured and massive tufa (2-5 cm thick) and the cellular structure within the layers that
are most resistant to erosion. B) Speleothem covering a block of intensely calcified tufa at the Okongwe 4 cave
Figure 10. The Okapiku Tufa Lobe viewed from the west.
Figure 11. The Okovanatje Tufa Lobe with its intermittently active bryophyte curtains, viewed from the west.
Figure 12. Okovanatje Tufa Lobe viewed from the north. The stream feeding the system divides into two chutes,
each of which is the site of moss growth and tufa deposition from the top of the cliffs to their base. The main
growth trajectory is thus sideways out into the neighbouring valley but there is also a minor vertical growth tra-
jectory which maintains the height of the structure.
Figure 12. Okovanatje Tufa Lobe, showing well developed but currently inactive bryophyte curtains (arrows)
growing out over hollows in the tufa lobe front.
Figure 13. Cave and canyon eroded into the older of the two tufa lobes at Omatapati. This lobe could well be of
Late Miocene or Early Pliocene age judging from its stage of erosion.
Figure 14. Rillenkarren developed in densely cemented tufa exposed on the surface of the eroded lower tufa
lobe at Omatapati.
Figure 15. Eroded speleothems and cave breccias in the lower tufa lobe at Omatapati.
Figure 16. Brown tufa overlying pale, densely cemented fluvial gravels and sands at the Omatapati Lower Tufa
Figure 17. The upper tufa lobe at Omatapati. This lobe has advanced more than 100 metres over a pre-existing
flat boulder-covered river terrace, part of which is exposed in the foreground.
Figure 18. Otjitamei Tufa Lobe showing the cliff front and the pool at its base (in shadow), at present fed by
underground seepage of vadose water. Mammal fossils occur in breccia exposed at the top of the cliff.
Figure 19. Otjikondavirongo Tufa Lobes, comprising an older, somewhat eroded lobe (the sloping ground to the
right of the image), and a younger, still active lobe forming the steep cliffs with vegetated waterfall to its left.
The latter lobe has grown out from its parent valley, forming a huge bluff in the topography (vehicles in trees at
left for scale).
Figure 20. At the foot of the old eroded tufa lobe at Otjikondavirongo, tufa is intercalated with fluvial conglom-
erate and sand which has been densely cemented. The sands contain land snails (Sculptaria).
Figure 21. The base of the actively growing part of the Otjikondavirongo Tufa Lobe. Note the abundance of
moss and other plants growing on the cliffside and on exposed boulders at the foot of the cliff, and the algae pro-
liferating in the pool. Note in particular the carbonate precipitating onto the bryophyte curtain, behind which
there is a substantial cave.
The riverside tufas at Ongongo were
described by Mocke (2014). The tufas are well
exposed in a canyon which has cut through the
carbonate deposits, revealing the interior of a
large tufa system, rich in plant remains, fresh-
water gastropods and land snails.
South of Ongongo there are two fur-
ther occurrences of tufa at a higher altitude
than the riverside tufas. Called the Ongongo
South Tufa Lobes, they crop out either side of
a north-south trending ridge of Basement
quartzite. Judging from their topographic posi-
tion, they must have been emplaced prior to
the downcutting of the present day valleys ei-
ther side of the ridge.
Figure 22. Ongongo South Tufa Lobes positioned either side of a Basement quartzite ridge south of Ongongo
Valley, viewed from the north. These tufa lobes, perched high above the plains, attest to the former presence of a
radically different drainage network from that of today, because, as they stand, there is no obvious source for the
carbonate of which they are built, nor is there a water supply to carry the carbonate in solution.
Figure 23. Paludal tufa overlying marls near the Camel Camp at Sesfontein. The tufas cover a large area, but are
Figure 24. Porous paludal tufa at the Camel camp, Sesfontein, is rich in natural molds of sedges and other plants
and also contains land snails (Achatina). These tufas are probably Late Pleistocene or Holocene.
Figure 25. Examples of leaf imprints in tufa at Ongongo Gorge. A) Colophospermum (?), B-D) Ficus, E) Salva-
Freshwater snails and land snails have
been preserved in most of the Kaokoland tufas.
Mocke (2014) illustrated an imprint of Sculp-
taria, a land snail, at Ongongo. This paper rec-
ords the presence of Bulinus and a planorbid at
Omatapati Lower Tufa Lobe, in association
with the land snail Succinea, Sculptaria at
Otjomatemba (near Otjitamei) and Otjikonda-
virongo, where Achatina also occurs, and
Xerocerastus at Okongwe. Finally, the Omata-
pati Upper Tufa Lobe yielded an elongated
Subulinidae, probably Opeas.
All these taxa occur in the region to-
day, and they attest to a Summer rainfall re-
gime, in strong contrast to the gastropod fauna
of the Winter rainfall zone in southwestern
Namibia (Pickford, 2008).
Figure 26. Freshwater and land snails in Kaokoland tufas and breccias. A-B) Omatapati Lower Tufa Lobe, A)
damaged shell of a planorbid, B) complete sinistral shell of Bulinus (top) and dextral shell of Succinea (below);
C) Impressions of Sculptaria shells in densely cemented fine sands at Otjomatemba, upstream from Otjitamei;
D) Sculptaria, preserved in pink sandy breccia associated with the old tufa lobe at Otjikondavirongo. The snail is
ca 15 mm in diameter; E) A sectioned shell of Xerocerastus, a dry country land snail common today in Kaoko-
land and Otavi, fossilised in pink, sandy breccia at Okongwe Tufa Lobe 4. The shell is ca 2 cm tall; F) Subulini-
dae shell in pink, sandy breccia associated with speleothems (bottom right corner of image) at Omatapati Upper
Tufa Lobe. The shell is ca 20 mm tall.
Figure 27. Micromammalian remains in situ in grey sandy breccia at the top of the Otjitamei Tufa Lobe.
Figure 28. Fossil bat jaw (possibly rhinolophid identified by V. Rossina) in breccia at the top of the cliffs at
Otjitamei Tufa Lobe.
Figure 29. Fossiliferous Late Pliocene breccia in situ in the northern flank of the Okongwe Tufa Lobe N° 4, rich
in the remains of rodents and other micromammals as well as fish, frogs and snakes.
Figure 30. A small sample of the rodent fauna from sandy breccia associated with the palaeocave in Okongwe
Tufa Lobe N° 4, to show part of the diversity (length x breadth measurements (in mm) in brackets). A)
Zelotomys woosnami, right M3/ (0.99 x 0.97); B) Aethomys chrysophilus left M1/ (2.56 x 1.83); C) Aethomys
namaquensis right M1/ (2.49 x 1.70); D) Aethomys sp. broken right M1/ (ca 2.47 x 1.75); E) Graphiurus
murinus, right upper molar (0.85 x 1.23); F) Steatomys krebsi ? left M1/ (1.77 x 1.20); G) Acomys spinosissimus
right M1/ (1.78 x 1.09); H) Zelotomys woosnami right M2/ (1.0 x 1.77); I) Aethomys namaquensis right M3/
(1.28 x 1.30); J) Stenodontomys darti left m/2 (1.27 x 1.06); K) Parotomys littledalei right m/2 (2.22 x 2.23).
Figure 31. Pedetes (Springhaas) mandible in situ in pinkish-grey sandy breccia at Omatapati Upper Tufa Lobe.
Figure 32. Small ruminant mandible and isolated teeth (probably klipspringer, Oreotragus) in situ in grey-pink
sandy breccia at the top of Otjitamei tufa cliff.
All the Kaokoland tufa lobes examined
show evidence of ancient and not so old human
activity in the vicinity. Most of the stone arte-
facts observed were lying on the present day
surface, or were partly buried in loose sedi-
ment, and are thus of little interest for dating
the tufas. However, at Omatapati Upper Tufa
Lobe, there are important masses of well-
cemented breccia rich in the remains of large
mammals (jaws, teeth, post-cranial bones) as-
sociated with stone tools of what looks like a
primitive flake technology. The age of this
breccia is unknown, but the mammal remains,
when extracted and studied, may yield infor-
mation of biochronological significance.
Figure 33. Stone tools and mammalian bones and teeth in breccia at Omatapati Upper Tufa Lobe.
The micromammal concentrations at
Okongwe and Otjitamei occur in discrete lay-
ers and masses suggesting that they represent
accumulations of owl pellets on the floors of
cavities in the tufa lobes. Some specimens rep-
resent the remains of animals that lived (and
died) on the lobes. The Omatapati Upper Tufa
Lobe is of interest on account of the associa-
tion of large mammal remains with stone tools
found there, indicative of early human habita-
tion in the area.
Biochronology of the Kaokoland Tufa Lobes
The only fossils found so far in Kaoko-
land tufas that yield biochronological data are
the micromammals. The small assemblage re-
covered from Okongwe Lobe 4 comprises mi-
cromammals, mostly rodents, but also mac-
roscelidids and bats. Most of the rodent species
are extant, but there is one genus, Stenodonto-
mys Pocock, 1987, which is extinct. The type
locality of this genus is Makapansgat, South
Africa, dated ca 3 Ma (3.0-2.7 Ma, Pickford,
2006; or 3.03-2.58 Ma, Herries et al. 2013).
The Omatapati Upper Tufa Lobe contains Pe-
detes (Springhaas) and other microfauna and
macrofauna, but it has not yet been studied in
detail. This means that the Okongwe Tufa
Lobe breccias could be about the same age as
the Thabaseek Tufas near Taung, South Africa,
which are 1,500 km to the southeast and which
yielded the type specimen of the early hominid
Australopithecus africanus Dart, 1925 (Pea-
body, 1954; Butzer et al. 1978). A similar Late
Pliocene age has been inferred for the traver-
tine domes at Kaukausib in the Sperrgebiet
(Pickford, 2000), so it would appear that the
genesis of the Kaokoland Tufas could well be
related to the same palaeoclimatic events that
led to superficial carbonate accumulations over
much of the sub-continent.
Table 2. Mammal fauna from Okongwe Tufa Lobe 4, Kaokoland, Namibia.
Vespertilionidae (Vesper Bat)
Acomys spinosissimus (spiny mouse)
Zelotomys woosnami (broad-headed stink mouse)
Stenodontomys darti (extinct rodent)
Gerbilliscus winkleri (bushveld gerbil)
Parotomys littledalei (whistling rat)
Graphiurus murinus (dormouse)
Aethomys namaquensis (rock rat)
Several other murids (mice)
Macroscelididae (Sengis)
The geomorphological settings of the
various subaerial carbonate units in the
Oruvandje, Warmquelle and Sesfontein re-
gions indicate that they span a considerable
period of geological time. Because the
Okongwe cave fill has yielded a diverse rodent
fauna including the extinct genus Stenodonto-
mys originally described from Makapansgat
(Middle Pliocene) and Langebaanweg (Early
Pliocene) the deposits are likely to be Pliocene
or older, whereas the valley bottom Swamp-
land Tufas at Sesfontein and hard pans (indu-
rated marls) are Late Pleistocene to Holocene,
on the basis of the presence in them of planor-
bids and Bulinus. Some of the tufa at Omata-
pati Upper Tufa Lobe contains the remains of
large mammals associated with stone tools
which suggest an Early Pleistocene age, but
further research is required to confirm this es-
Palaeoclimatic implications of the Kaoko-
land carbonate deposits
The Cascade Tufas at Oruvandje and
north of Warmquelle attest to the existence of a
more humid palaeoclimate than the present day
arid to hyper-arid regime that typifies the re-
gion. The travertine masses are vast, they oc-
cur in most of the valleys in the region, they
are located high above the extant valley floors
and in cases such as Otjikondavirongo, their
fronts have grown significant distances out-
wards over the neighbouring low country. In
the Oruvandje area, there are three main levels
at which the travertine masses formed. Today,
the tufa masses at Oruvandje and Warmquelle
are perched well above the valleys into which
their respective drainages flow. The travertine
cliffs at Ongongo South are located at the up-
per edges of a long North-South ridge of
quartzite, and they could not have formed
where they are under the present day topo-
graphic conditions. They attest to a considera-
ble amount of erosion on both sides of the
Quartzite Ridge since the period of growth,
which, at the time of accumulation of the trav-
ertine, was traversed by a river.
The land snail fauna (Succinea, Sculp-
taria, Xerocerastus, Opeas (?) and Achatina)
indicate that, at the time of tufa growth,
Kaokoland lay within the Summer rainfall
The micromammals from Okongwe
Lobe 4 comprise taxa that today occur prefer-
entially in semi-arid to sub-humid areas. Based
on the ecological requirements of extant ro-
dents of Southern Africa (De Graaf, 1981) the
fauna from the Okongwe Lobe 4 deposit and
Omatapati indicate a somewhat more humid
and probably cooler Pliocene palaeoclimate
than the present day arid regime that character-
ises the region, which supports Mopane wood-
land. The Pliocene vegetation at Okongwe may
have been equivalent to wooded savannah,
such as Miombo woodland. The land snails
from the Kaokoland tufas are like those that
occur in the region today, but they also resem-
ble those of the more humid Otavi Mountains,
for example.
Preliminary palaeontological survey of
parts of Kaokoland during the 1990’s resulted
in the discovery of a few unidentified gastro-
pod specimens in epikarst deposits at Otjo-
matemba (Pickford & Senut, 2010), Robbie’s
Pass Cave and Erova, and some rodents and
dassies from internal karst deposits at Ondera,
Rocky and Tim’s Cave. These occurrences
were rather poor in fossils, but they already
revealed that most of the sites were Pleistocene
in age, and that the sediments accumulated
under a Summer rainfall climatic regime (Pick-
ford & Senut, 2010).
The new findings are much more com-
prehensive, and they indicate the existence of a
vast archive of fossiliferous breccia at several
loci in Kaokoland, preserving plants, inverte-
brates and vertebrates. Most of the fossils ob-
served comprise the remains of microverte-
brates that were brought into caves and fissures
in the growing tufa lobes by owls and other
raptors, but there are also the remains of small
birds, snakes, frogs and the occasional fish that
represent animals that lived in the vicinity of,
or even on the lobes. There are also rare spec-
imens of larger mammals such as small bovids
and dassies, which, like the extant Klipspringer
(Oreotragus) and hyraxes (Procavia), also
probably lived on or near the tufa lobes.
Finally, at one of the sites, Omatapati
Upper Tufa Lobe, there are well-indurated
breccias rich in the remains of large mammals
associated with stone tools. All this translates
into a major palaeontological resource in
Kaokoland which will take many years of con-
certed effort to prospect and study.
There have been previous attempts to
date the main periods of tufa growth in the
Naukluft Mountains, Namibia (Viles et al.
2007; Stone et al. 2010; Goudie & Viles, 2015)
but without a great deal of success, due to the
open geochemical nature of the depositional
systems typical of porous tufa deposits (Rich et
al. 2003).
Preliminary assessment of the rodent
fauna from Okongwe Tufa Lobe N° 4, indi-
cates a Late Pliocene age for tufa deposition
there, but it is clear from the field relations of
the various tufa lobes examined, their state of
preservation and their degree of erosion, that
there are older and younger occurrences in the
region. The lower lobe at Omatapati could well
be Early Pliocene or Late Miocene in age, as
could the older of the two lobes at Otjikondavi-
rongo. In contrast, the discovery of early stone
tools in the Omatapati Upper Tufa Lobe sug-
gests an Early Pleistocene age for some of the
activity there. Finally, the paludal tufa fields at
Sesfontein Camel Camp and on top of the
Otjitamei Tufa Lobe at Otjomatemba, are
probably Late Pleistocene to Recent.
The discovery of richly fossiliferous
Late Neogene to Pleistocene and Holocene tufa
deposits in Kaokoland is of great interest for
African Neogene and Quaternary palaeontolo-
gy. The tufa lobes occur in a part of Africa that
has previously yielded very little palaeontolog-
ical information. The nearest richly fossilifer-
ous areas are at Etosha and Ekuma (Late Mio-
cene to Late Pliocene) on the northwestern
edge of Etosha Pan (Pickford et al. 2014,
2016) over 200 km to the east, in the Otavi
Mountains 470 km to the east (Middle Mio-
cene to Recent) (Pickford & Senut, 2010) and
the Humpata Plateau in Angola, 440 km to the
north (Plio-Pleistocene) (Pickford et al. 1990,
1992, 1994).
Preliminary studies of the Kaokoland
fossils indicate that during phases of active
tufa growth, the region was somewhat more
humid and probably slightly cooler than it is
today, probably comprising Miombo wood-
land, or the transition between Miombo and
Mopane vegetation types, in contrast to the
very arid climate with Mopane Woodland that
currently characterises the area.
The cascade and paludal tufas of both
Kaokoland and the Naukluft Mountains
formed at the high edge of the Great Escarp-
ment, in the vicinity of dolomite and limestone
outcrops of the Otavi Group in the case of
Kaokoland, and of the carbonate-rich Naukluft
nappes in the case of the latter occurrences.
Mosses and other plants played a sig-
nificant role in the tufa formation, as shown by
the quantity of phytoherms and bryophyte cur-
tains and cushions preserved. The tufas are
associated with densely cemented gravels,
sands and silts in the bottoms of the valleys
both upstream and downstream from the tufa
lobes, attesting to the presence of important
quantities of carbonate in the groundwaters of
the region. There are frequent plant remains
(leaves, root imprints), gastropods and verte-
brates preserved in sandy to pebbly breccias
associated with the lobes.
The micromammalian remains in the
tufa lobes appear to result from the activity of
owls and other raptors that roosted in crannies
and caves in the tufa lobes, regurgitating bone-
rich pellets which accumulated on the floors of
the cavities which then became cemented by
calcite. Bats also lived in these cavities, as
shown by the presence of jaws and teeth in the
breccias. The fossils of other vertebrates such
as lizards and snakes (colubrids (Rage, pers.
comm.)), dassies and small bovids, may well
represent the remains of animals that lived on
the tufa lobes. Finally, there is evidence of ear-
ly human activity preserved in some of the
lobes, where well-cemented breccia contains
large mammal remains associated with primi-
tive stone tools.
The NPE thanks the French Embassy
in Namibia (His Excellency, J.-L. Zoël), the
Cooperation Service of the French Embassy in
Windhoek (J.-P. Martin, P. Portes-Gagnol), the
Muséum National d'Histoire Naturelle, Paris,
UMR 7207 and CR2P (CNRS, MNHN) (S.
Crasquin) for their administrative and diplo-
matic help.
In Windhoek, Dr Gabi Schneider, Di-
rector of the Geological Survey of Namibia,
provided essential help and encouragement.
We also thank Mrs Jane Eiseb for administra-
tive help. The existence of ancient tufa depos-
its in the Oruvandje area was indicated to us by
Dr Eckhart Freyer.
We thank Alma Nankela of the Namib-
ian National Heritage Council for arranging
authorisation to carry out research in Namibia.
We were helped in the field by Thom Pita,
Abel Tjamburo, Jemau Rutjani Moriri, Nangu-
la Vatillva, Gerald Baba, Herunga Nkahima
and Nancy Mathee and we appreciate the hos-
pitality and help given to us by Chief Elias
Tjizembisa of Oruvandje. Thanks to Valentina
Rossina for the preliminary identification of
the Otjitamei bat fossil and Jean-Claude Rage
for identifying the snake fossils from
Butzer, K.W. Stuckenrath, R. Bruzewicz, A.J.
Helgren, D.M. 1978. Late Cenozoic paleo-
climates of the Gaap Escarpment, Kalahari.
Quaternary Research, 10 (3), 310-339.
Dart, R. 1925. Australopithecus africanus: The
man-ape of South Africa. Nature, 115, 195-
De Graaf, G. 1981. The Rodents of Southern
Africa. Durban, Pretoria, Butterworths, 267
Goudie, A. & Viles, H. 2015. The Naukluft
Mountains and their Tufa Casacades. In:
World Geomorphological Landscapes, Part
2, Dordrecht, Springer Science+Business
Media, pp. 133-136.
Herries, A.I.R. Pickering, R. Adams, J.W.
Curnoe, D. Warr, G. Latham, A.G. & Shaw,
J. 2013. A multi-disciplinary perspective on
the age of Australopithecus in Southern Af-
rica. In: Reed, K. Fleagle, J. & Leakey, R.
(Eds) The Paleobiology of Australopithecus.
Dordrecht, Springer Science+Business Me-
dia, pp. 21-40.
Korn, H. & Martin, H. 1937. Die jüngere geol-
ogische und klimatische Geschichte
Südwestafrikas. Zentralblatt für Mineralo-
gie, Geologie und Paläontologie, B11, 456-
Korn, H. & Martin, H. 1955. The Pleistocene
in South West Africa. Proceedings of the
3rd Pan-African Congress on Prehistory.
Livingstone, pp. 14-22.
Mocke, H. 2014. Note on the fossil fauna and
flora in tufa at Ongongo Springs, Damara-
land, Namibia. Communications of the Geo-
logical Survey of Namibia, 15, 134-141.
Peabody, F. 1954. Travertines and cave depos-
its of the Kaap escarpment of South Africa,
and the type locality of Australopithecus af-
ricanus Dart. Bulletin of the Geological So-
ciety of America, 65, 671-706.
Pedley, M. Martin, J.A.G. Delgado, S.O. &
Cura, M.A.G.D. 2003. Sedimentology of
Quaternary perched springline and paludal
tufas: criteria for recognition, with examples
from Guadalajara Province, Spain. Sedimen-
tology, 50 (1), 23-44.
Pickford, M. 2000. Neogene and Quaternary
vertebrate Biochronology of the Sperrgebiet
and Otavi Mountainland, Namibia. Commu-
nications of the Geological Survey of Na-
mibia, 12, 359-365.
Pickford, M. 2006. Synopsis of the biochro-
nology of African Neogene and Quaternary
Suiformes. Transactions of the Royal Socie-
ty of South Africa, 61 (2), 51-62.
Pickford, M. 2008. Freshwater and Terrestrial
Mollusca from the Early Miocene deposits
of the northern Sperrgebiet, Namibia. Mem-
oir of the Geological Survey of Namibia, 20,
Pickford, M. Fernandes, T. & Aço, S. 1990.
Nouvelles découvertes de remplissages de
fissures à primates dans le "Planalto da
Humpata", Huilà, Sud de l'Angola. Comptes
Rendus de l’Académie des Sciences, Paris,
310, 843-848.
Pickford, M. Mein, P. & Senut, B. 1992. Pri-
mate bearing Plio-Pleistocene cave deposits
of Humpata, Southern Angola. Human Evo-
lution, 7, 17-33.
Pickford, M. Mein, P. & Senut, B. 1994. Fos-
siliferous Neogene karst fillings in Angola,
Botswana and Namibia. South African Jour-
nal of Science, 90, 227-230.
Pickford, M. Mocke, H. Ségalen, L. & Senut,
B. 2016. Update of the Pliocene fauna of the
Ekuma Valley, Etosha, Namibia. Communi-
cations of the Geological Survey of Namib-
ia, 17, 115-144.
Pickford, M. & Senut, B. 2010. Karst Geology
and Palaeobiology of Northern Namibia.
Memoir of the Geological Survey of Namib-
ia, 21, 1-74.
Pickford, M. Senut, B. Hipondoka, M. Person,
A. Ségalen, L. Plet, C. Jousse, H. Mein, P.
Guerin, C. Morales, J. & Mourer-Chauviré,
C. 2014. Mio-Plio-Pleistocene geology and
palaeobiology of Etosha Pan, Namibia.
Communications of the Geological Survey of
Namibia, 15, 16-68.
Pocock, T.N. 1987. Plio-Pleistocene fossil
mammalian microfauna of Southern Africa
a preliminary report including description
of two new fossil muroid genera (Mamma-
lia: Rodentia). Palaeontologia africana, 26,
Rich, J. Stokes, S. Wood, W. & Bailey, R.
2003. Optical dating of tufa via in situ aeoli-
an sand grains: a case example from the
Southern High Plains, USA. Quaternary
Science Reviews, 22, 1145-1152.
Stone, A. Viles, H.A. Thomas, L. & Calsteren,
P. 2010. Quaternary tufa deposition in the
Naukluft Mountains, Namibia. Journal of
Quaternary Science, 25 (8), 1360-1372.
Van der Merwe, J.H. (Ed.) 1983. National At-
las of South West Africa. Cape Town, Unifo-
to, 92 maps with accompanying texts.
Viles, H.A. Taylor, M.P. Nicoll, K. & Neu-
mann, S. 2007. Facies evidence of hydro-
climatic regime shifts in tufa depositional
sequences from the arid Naukluft Moun-
tains, Namibia. Sedimentary Geology, 195
(1-2), 39-53.
ResearchGate has not been able to resolve any citations for this publication.
Full-text available
Rescue excavations in the Ekuma River Valley, Etosha National Park, resulted in the collection of well preserved plants, invertebrates and vertebrate fossils of Pliocene age. The collections contain several new records for the Ekuma Delta Member, including a lung fish (Protopterus), a long-snouted crocodile (Euthecodon), an aardvark (Orycteropus), a hyaena and a gigantic suid (Noto-choerus capensis). This article updates the fossil record of the Ekuma Delta Member and discusses its palaeoenvironment.
In the fluvio-paludal deposits of the Northern Sperrgebiet, freshwater and terrestrial molluscs are generally rare but in certain localities or horizons they are quite abundant. The commonest terrestrial snails are Trigonephrus and Dorcasia whilst urocyclid slug plates are rare, having been found at one locality only. Four freshwater snail genera have been identified, Tomichia, Lymnaea, Bulinus and Succinea, the latter being semi-terrestrial. The terrestrial assemblage has a distinct southwestern African cachet to it, but apart from Tomichia the freshwater snails belong to more widespread groups. The Lymnaea from the Sperrgebiet are by far the oldest known in Africa, only represented in other parts of the continent during the Middle and Late Miocene, Late Pleistocene and Holocene. The Bulinus from Langental are the earliest known in Africa.
During the past eight years, palaeontologists participating in a programme of research called Paleokarst Afrique, have been examining karst fillings in Botswana, Angola and Namibia. Numerous richly fossiliferous occurrences have been discovered in all three countries, ranging in age from middle Miocene to Holocene. Preliminary analyses of the fossils indicates that the southern portion of Africa was home to primates throughout this time period, as well as to a host of micromammals and other vertebrates, many of which are new to science. -Authors
Analyses are presented of the mammalian component of rich microfaunal fossil breccia collections mainly of owl pellet origin from the Transvaal Plio-Pleistocene australopithecine sites Kromdraai, Sterkfontein and Makapansgat Limeworks, with briefer references to Swartkrans, Langebaanweg and the Makapansgat Cave of Hearths. Rodent incisors show Cryptomys robertsi to be a distinct extinct species occurring with C. hottentotus, and indicate a relationship between Mystromys, the Cricetomyidae and certain fossil Cricetodontidae. The fossil assemblages are generally similar to modern ones but elephant shrews (several species) and dormice are relatively commoner, and Mystromys has declined in favour of the murid Mastomys. Crocidura and Saccostomus are abstracted from the older fossil sites, appearing only in the more recent Cave of Hearths. Mystromys darti Lavocat has been rediscovered in abundance in in situ Rodent Corner breccia at Makapansgat, yet it is totally absent from other parts of the Limeworks deposit. It is referred to a new genus Stenodontomys, with a second species from Langebaanweg. Another extinct cricetid previously known under a manuscript name as Mystomys cookei, common to Makapansgat, Taung and the Krugersdorp district sites, is formally described for the first time also under a new generic name, Proodontomys. On microfaunal evidence Makapansgat is definitely older than the Krugersdorp sites, of which Kromdraai is perhaps the oldest and Swartkrans the youngest. Certain extinct fossils link Makapansgat to Langebaanweg (Stenodontomys), Kromdraai (Macroscelides proboscideus vagans) and Taung (Gypsorhychus). Suggestions that Taung is significantly younger than other australopithecine sites are not supported. - from Author
This paper presents a review of, and new data concerning, the age of Australopithecus in southern Africa. Current dating suggests that Makapansgat Limeworks is the oldest hominin deposit in southern Africa, with Australopithecus africanus dating to between 3.0 and 2.6 Ma. The Taung Child A. africanus fossil from Taung is most likely penecontemporary with the Makapansgat material between 3.0 and 2.6 Ma. A. africanus from Sterkfontein Member 4 is estimated to date to between 2.6 and 2.0 Ma, with the Sts 5 specimen dating to around 2.0 Ma. The A. africanus deposits from Gladysvale are most likely contemporaneous with the Sterkfontein group with an age between 2.4 and 2.0 Ma. The potential second species of Australopithecus, StW 573 from the Silberberg Grotto at Sterkfontein, is most likely dated to between 2.6 and 2.2 Ma. As such, StW 573 is contemporary with A. africanus fossils from Member 4 and suggest that two contemporary Australopithecus species occurred at Sterkfontein between ~2.6 and 2.0 Ma. Based on the presence of Equus the A. africanus fossils from Jacovec Cavern also likely date to Australopithecus sediba-bearing deposits of Malapa date to 1.98 Ma and suggests that three different species of Australopithecus occur in South Africa between 2.3 and 1.9 Ma. Given these dates, A. africanus represents the oldest southern African hominin species being found in two temporally distinct groups of sites, Makapansgat/Taung and Sterkfontein/Gladysvale, and A. sediba is the youngest species at ~1.98 Ma. However, if StW 53 is also Australopithecus, as some have suggested, then this genus survives to younger than 1.8 Ma in South Africa. Australopithecus thus lasted for a significant period of time in southern Africa after the genus is last seen in eastern Africa (Australopithecus garhi at ~2.5 Ma). This new dating indicates that the South African Australopithecus fossils are younger than previously suggested and are contemporary with the earliest suggested representatives of Homo (~2.3 Ma) and Paranthropus (2.7–2.5 Ma) in eastern Africa.
The surface travertines, included cave deposits, and other associated phenomena along the Kaap escarpment (Campbell Rand) of the Union of South Africa are mapped and described with emphasis on chronology. The Thabaseek, Norlim, Oxland, and Blue Pool travertines (new names) at Buxton in the Taungs Native Reserve, type locality of Australopithecus africanus Dart, are of different ages and are related to each other and to water-cut channels in such a manner as to represent a long, seven-stage sequence of events, the first of which antedates Middle Pleistocene time. Several cave faunas, including the Australopithecus fauna, and archeological material of Middle and Late Stone Age of South Africa are correlated with the travertine sequence. Australopithecus occurs in the youngest part of the oldest travertine. Other travertines and associated phenomena along the Kaap escarpment are correlated with the Buxton sequence, and a tentative correlation is made with the diamond gravels of the Vaal River. Buxton travertines seem to have been formed during the waning of wet (pluvial) periods; on this basis, two earlier major travertines and two later minor travertines at Buxton correlate with the two earlier major and the two later minor pluvials interpreted from the gravels and terraces of the Vaal River by H. B. S. Cooke. Australopithecus and its associated fauna are considered no older than Lower Pleistocene on the basis of physical and biological data. There is no concrete evidence for a Pliocene age.
Precipitated carbonates (commonly termed tufas or travertines) maybe of considerable utility for palaeoenvironmental reconstruction. Their potential, however, for such reconstruction is commonly limited by difficulties associated with their absolute age control. Attempts to date such deposits via uranium series techniques have been complicated by their chemically open behaviour. Here we describe an alternative approach to date tufa deposits associated with ephemeral saline lake basins from the Southern High Plains, USA. We have optically dated sand grains of a mixed aeolian/fluvial (spring fed) origin as the integrating dosimeter. We assume that the grains are fully resetting prior to their incorporation into the tufa deposits and employ a time-dependent disequilibrium dosimetric model to account for the build-up of uranium series daughter products. The approach was applied to a set of four samples with known stratigraphic association. We obtained stratigraphically sensible optical ages ranging from 78±8 to 56±4ka. These data are consistent with existing palaeoenvironmental models of regional recharge.
This festschrift in honour of H.B.S. Cooke provides an opportunity to prepare a synopsis of the biochronology of African Suiformes, a group with which he worked for many years. Cooke realised early in his career that the extreme variation that he noticed among African Plio-Pleistocene suid dentitions was the result of rapid evolution. In early years he reasoned that the fossils could be used to correlate fossil sites with one another in a relative way. Later in his career, with the emergence of geophysical dating, he was able to convert his biostratigraphic scale to a biochronological one, and from then on could propose ages in millions of years for deposits yielding suids. Cooke seldom worked on fossils older than 7 Ma, partly because the fossil record was poor, but also because what had been found was rather monotonous from a morphometric aspect. This paper provides the known chronological ranges of all the Neogene and Quaternary suiform species of Africa in order to extend Cooke's pioneering biostratigraphic framework downwards to the base of the Neogene, taking into account recently described species of suines and hippopotamids. A brief discussion of the palaeoenvironmental changes that probably led to many of the rapid dental evolution completes the paper.
Outcrop-scale geometries and bed relationships of ambient temperature freshwater carbonates are poorly understood because many described tufas have been dismantled by erosion and present only part of a particular depositional model. At the field scale, four end-member models encompass the tufa continuum: (1) perched springline; (2) paludal; (3) fluvial; and (4) lacustrine tufas. Individual bed types can occur with variable dominance within several of these models, but one or more beds are characteristically dominant only within a single tufa model, so it can be differentiated from relatively isolated outcrop fragments. Two models (perched springline tufas and paludal tufas) are known in outline only in the literature despite being present within the Quaternary deposits of most karstic regions. Perched springline tufas generally form lobate, fan-shaped mound morphologies on hillslopes and develop from single or multiple spring resurgences. Mature deposits show a subhorizontal top and a steep face on the downflow side. The steep outer zones of tufa mounds may be developed into cascades with moss curtains or can be dominated by shallow rimstone pools according to face angle. Tufa deposits lying downslope of the mounds are usually detrital in nature, especially if some dismantling of the mound has occurred. The relatively thin subhorizontal lobe-top deposits commonly contain organic-rich deposits. Paludal tufas develop predominantly in waterlogged valley bottom situations, where line-sourced waters emerge from valley side and bottom aquifers. Lime mud precipitation predominates in these sites. Mud is deposited as subhorizontal laminites that thin towards the valley axis and downstream of resurgences. Tufa spring-mounds may form where lesser volumes of water are involved. Individual tussocks (phytoherm cushions) of grasses and rushes are the most diagnostic feature of the model, but sapropels and peats may be intercalated. Diagenesis in both models is rapid. Lithification of individual beds is virtually instantaneous and always occurs before the decay of the associated living vegetation. The resulting highly porous and permeable fabrics remain fresh in Holocene tufas, but aggrading neomorphism and partial spar infill of vegetation moulds is common in older deposits. Dissolution in many perched springline tufas is small scale. Many large cavities are primary, but with later coatings of speleothems. Early removal of organics from paludal tufas is responsible for autobrecciation and differential compaction.