Primate bearing Plio-Pleistocene cave deposits of Humpata, Southern Angola

HUMAN EVOLUTION Vol. 7 - N. I (17-33)- 1992

M. Pickford Chaire de Paldoanthropologie et de Prdhistoire, Coll~ge de France et Institut de Palrontologie 8, rue Buffon 75005 Paris

P. Mein Drpartement des Sciences de la Terre Universit6 Claude Bernard, Lyon I 27 - 43, Boulevard du 11 Novembre 1918 69621 Villeurbanne

B. Senut Laboratoire d'Anthropologie et Institut de Paldontologie Musdum National d'Histoire Naturelle 8, rue Buffon 75005 Paris

Key Words: Plio-Pleistocene, Cercopithecidae, Cave breccias, Angola, Geological context.

Primate bearing Plio-Pleistocene cave deposits of Humpata, Southern Angola

Numerous fossil primates have been discovered in cave breccias of the Humpata Plateau in Southern Angola. Associated faunal remains suggest an age for the sediments near the Plio-Pleistocene boundary. This report deals principally with the geological context of the cave breccias, their ages and their palaeontology.

Historical Background

The first mention of the existence of fossiliferous fissure fillings in Southern Angola dates from 1933, when Beetz described the economic potential of the fossiliferous travertines of the Humpata Plateau. Fossil primates were first recorded from the region by MourA (1950), by DA~T (1950), and by AP, AMBOtPaG & MOUTA (1952), although they were evidently known to occur there many years before (DART, 1950), one of the specimens having apparently been donated to Harvard University by the Rev. M. Ennis (Mn~KOW, 1972), some time before 1950.

The fossil primates collected by Mouta and originally studied by Arambourg (Ae, AMBOUR6 & MOUTA, 1952) are now housed in the Universidade Nova da Lisboa, Portugal, but most of the specimens have yet to be described.

During the 1960' s, a great deal of excavation was carried out in the Cangalongue area by Dr. A. Machado Cruz, b u t h e died before publishing anything about his discoveries. Many of the fossils that he unearthed have been lost, although a few, including some primates, were preserved in storage that he built near the excavation site and looked after by his African assistant Sr. Shekulu Tchihonga and are now housed at the Museu Regional da Huilfi. FRANfA (1964) exca- vated some primates from a cave at "Leba", but the specimens that he obtained are perhaps only subfossils, not to be confused with the Plio-Pleistocene assemblages (ERvEDOSA, 1980; B~Un. & ALMEIDA, 1964). Similar subfossil primate remains were recovered from unconsolidated cave deposits at Cangalongue, along with numerous other mammalian remains, which are stored in the Cangalongue huts built by Cruz.

18 PICKFORD, MEIN and SENUT

Surveys carried out in the latter part of the 1960's and early 1970's appear to have been less successful at locating fossil primates (ANrtrN~S, 1965; MASON, 1976).

Because of the obvious palaeontological potential of the Humpata Plateau, the authors organised the Angola Palaeontology Expedition. During the first field season (PIc~ORD et al., 1990), two of the old localities were relocated and two new ones discovered, which indicated that the area was worth further attention. Consequently, a long term research expedition was planned. This report provides the results of the second field season carried out in July-August 1990.

Results

During 1990, numerous new fossil localities were found in the Humpata Plateau, and many superbly preserved mammalian remains were collected. The Tchiua quarry, where fossils were first reported from the Humpata Plateau (MourA, 1950; DART, 1950) yielded abundant cercopithecid remains including several complete skulls and mandibles of diverse ontogenetic ages, as well as quantities of postcranial bones. Other sites of particular importance are Malola, Ufefua and Cangalongue. Several less richly fossiliferous sites were located, including Tchivinguiro and accessory sites at Ufefua, Cangalongue and Malola. There are undoubtedly numerous other fossil occurrences in the region.

The main fossil sites were mapped (Figure 1 - 8) and where in situ strata occurred, the stratigraphic sequence was recorded.

Geology and Stratigraphy

The Humpata Plateau (FABER, 1926), on average about 1750 metres above sea level, termi- nates westwards in an immense escarpment, the Serra da Chela, more than 1000 metres high. The plateau itself is relatively flat lying, except where a few streams have incised themselves deeply into the Chela dolomites and the underlying sandstones. There are some 200 sq. km. of dolomite outcrops in the region, which are up to 200 metres thick in places (AMAm~a~, 1973; VALE et al., 1968).

In the Leba and Cangalongue outcrops, the dolomites are separated from the underlying sandstones by a dolerite sill of post-Permian age (VAL~ et al., 1968) which has evidently had some effect on the formation of cave systems within the dolomites, presumably because the ancient water table was perched on it (Figure 4).

There are numerous quarries in the region, which fall into two categories, "Pedreira do cal preto" and "Pedreira do cal hydraulico" (I~EMME, 1955). It is the latter that are of interest to vertebrate palaeontology, because they comprise excavations for sinter deposits which are often associated with various kinds of breccia and other cave fillings, the latter being rejected by the miners as unsuitable for quick lime production.

Palaeontologically the most productive areas of the Humpata Plateau are the Cudeje and Cangalongue Valleys. In the Cudeje Valley (Figure 2) (CuNJE & MOUTA, 1950) there are several fossiliferous sites on both the north and south flanks. Ufefua and Tchiua occur in the cliffs above streams which flow southwards into the Cudeje, and Malola occurs on its south bank. Other affluants of the Cudeje cut into the dolomites, but have not yet been surveyed for their fossil potential.

The north-south Cangalongue Valley (Figure 3) contains fossiliferous breccia in fissures and

PRIMATE BEARING PLIO-PLEISTOCENE CAVE DEPOSITS OF HUMPATA, SOUTHERN ANGOLA 19

former caves at various places along its west bank. Much of the valley has not yet been examined. Near Tchivinguiro Spring (Figure 2) fossiliferous pink breccia outcrops in the walls of small

caves above the spring. In other areas where sinter has been mined, such as extensive quarries near the Fazenda da

Leba (Leba Farm) very few fossils were seen. The breccia at these northern quarries is extremely coarse, in effect being chunks of dolomite supported by sinter, with very tittle clastic material derived from the ancient land surface. It is likely that erosion along the northern margin of the dolomite outcrop has proceded further than in the Cujede Valley, and that only the lower roots of the cave systems are preserved.

In contrast, at Malola (Figure 7), Tchiua (Figure 6) and Ufefua (Figure 8), there are large volumes of pink breccia, some of which contain indications of its neat" surface origins. Termite foraging tunnels and root holes can be observed at Malola and Tchiua respectively.

In both the Cangalongue and Cudeje Valley cave systems, the broad developmental and sedimentary histories appear to be similar, although they are unlikely to have been contemporane- ous.

Following the development of the cave systems, there was a period of sinter deposition, followed by erosion, followed by deposition of coarse breccia, often containing fragments of cave sinter, followed by a period during which finer grained pink to red breccias accumulated. These were then eroded to their present configm'ations.

The main exception is Tchiua, where in addition to the usual developmental sequence, there is a fissure filling which consists of grey-yellow very fine grained breccia containing abundant primate remains. The latter is considered to represent a calcified bat guano. In places it has been eroded and fissures in it are filled with red breccia suggesting that the Tchiua grey breccia is older than the red one.

Fossil mammals also occur in coarse breccia comprised predominantly of sinter, but are usually poorly preserved, as for example at Cangalongue Quarry 3 (Figure 5). Fossil mammal assemblages which occur in pink and red breccia are much richer and better preserved, as for example at MaMa (Figure 7), Cangalongue Quarries 1 and 2, and at Ufefua Quarry 2 (Figure 8). The only other fossiliferous sediment, which occurs at Tchiua, is interpreted as a calcified bat guano with little clastic contamination from the ancient land surface.

Faunal Remains

The faunal content of the grey-yellow breccia at Tchiua is peculiar in that it is dominated by cercopithecids, many of them young individuals~ At this site, there are relatively few micromammal remains compared with pink breccia, and there is a limited number of reptiles, including an articulated snake skeleton.

Bovids, suids, equids and other large mammals which are rare or absent at Tchiua, are commoner at Cangalongue, Malola and Ufefua, but at these sites large mammals are always outnumbered by microfaunal remains. Medium sized mammals, including hyracoids, lagomorphs and small carnivores, are relatively well represented in these faunas.

Up to now, more than 50 cranial remains along with numerous postcranial elements of cercopithecids have been found in the Tchiua breccia by the Angola Palaeontology Expedition. Most of the crania and mandibles are more or less complete, suggesting that at least 25 individuals are present in the sample. The final count could be considerably greater than this however, because only 10 cubic metres of breccia have been examined and there remain hundreds of cubic

2 0 PICKFORD, MEIN and SENUT

metres to be sorted through. It should be pointed out that the fossil content of the breccia is highly variable, some blocks being rich in fossils, while others appear to be devoid of them.

Until the fossils have been extracted from their breccia blocks, we can only provide an incomplete list of fauna (Table 1). Fossil lizards, amphibians and birds are common at Malola, Cangalongue and Tchiua. The micromammal fauna is dominated by rodents, evidently the prey remains of owls, since most of the species are nocturnal, only two diurnal rodents being found (Otomys, Dasymys) among more than 19 genera present in the breccias. The only common diurnal micromammals consist of Macroscelididae. Four genera of Microchiroptera occur in the breccias, some remains of which consist of complete skulls and mandibles, and, on occasion, articulated thoracic cages. Up to four species of soricids occur in some of the sites, many specimens comprising mandibles and maxilla with complete dentitions. Table 1 provides a list of the genera so far recognised in the Humpata breccias. It will take some time to determine the fauna at the species level because most of the material appears to be specifically distinct from extant species of Africa. Medium sized mammals are reasonably well represented, especially hyracoids (Procavia, Gigantohyrax) which evidently lived in the caves and fissures. We have found several small to medium-sized carnivores (mustelids, viverrids, canids) and primates (Parapapio, Dinopithecus, Cercopithecoides). The diversity of large mammals is restricted, only a few large bovids - Connochaetes, Hippotragini (Thomas, pers. comm.), suids (Metridiochoerus andrewsi (stage II), and carnivores (hyaenid, Petter pers. comm.) having been found.

Biostratigraphy

At present, few biostratigraphic data are available, mainly because the fauna has yet to be studied in detail. Of the 19 genera of rodents so far recorded at Humpata, most are represented by new species. There are at least four species of bats, but this group is not particularly valuable for biostratigraphy. There are four species of Soricidae and several Macroscelididae, but their affinities are not yet clear.

If the determination of Serengetilagus is correct, then this might indicate affinities with the upper Pliocene of East Africa. A similar indication is yielded by the presence of Gigantohyrax at Malola. The best evidence concerning the age of the deposits is provided by three specimens of Metridiochoerus andrewsi stage II, all from Cangalongue (quarries 2 and 3) which correlate best with specimens from the time period 1.8 to 1.3 Ma in East Africa (HARMS & WHrrE, 1979). A similar age estimate could be implied from the occurrence of Dinopithecus and Cercopithecoides at Tchiua. At the moment we feel reasonably confident in proposing that the Humpata fissure fillings date from the end of the Pliocene and the beginning of the Pleistocene. Further collections may yield more refined age estimates.

Palaeoecology

Rodents and the other microfauna are proving to be useful for determining African Plio- Pleistocene palaeoenvironments. This is mainly because these small mammals appear to have evolved rather slowly in Africa, but shifted their ranges as climates and vegetation patterns changed.

Among the 23 species of rodents so far known to occur at Humpata, only two are diurnal,


- ~ , ~ Hurnpato

0

Plio-Pleistocene fossil breccia

Post-Permian dolerite sill

U ~ Cambro-Silariam Chela Dolomite

Carnbro-Siluriam Sandstones

Pre-Cambrian granite

C Geological boundary J . f Contour linez

S Fault . r ~Stream line "-'.-. Road ~" ~ Track

�9 Cave

0~,, ~5 Km

Figure 1 - Locality map of the Humpata Plateau, Southern Angola, showing outcrops of the Chela Dolomite and the location of major fossil sites (Dolomite: squares; fossil sites: crosses). Figure 2 - Geological sketch map of the Leba-Tchivinguiro Area.


r ' ~ ' d �9 . s :.~\L~ , / . " , 1 , - i / ' 7 ' r T ' > ' L ~ ~ ~ C _ Y , ' ~ ~ ' , ~

�9 * k~ , . . . . . d ill f , _ .

-~ ~ . - , ~ - 5 ~ - ~ . ~ F _ .44 / ~ ~ ~ ~ ~ ]111

Plio-Pleistocene fossil breccia ([~ Geological boundary ~ Contour linez

~ Post-Permian dolerite sill . Fault

~ Cambro-Siluriam Chela Dolomite t:_ ~. Stream line - " ~ ' ~i -~ Road

Cambro-Siluriam Sandstones ~ ~" -- Track

�9 Cave

"1800

'1700

11600

1500

.1400

11300 m.

U f e f u a 2 Tctl iua Ho lo la

tl, < 0 I 2 3 4 5

Km.

1900 ~ N Tchiv ingui ro Haigta

, 00 2 Do(erite silt Sands tone Gran i te

Figure 3 - Geological sketch map of the Cangalongue Area. Figure 4 - East-West and North-South profiles in the Humpata Plateau, Southern Angola. (Cross- hatching: Chela dolomite; Black: dolerite; Dots: Chela sandstones; Squiggle: granite).

P R I M A T E B E A R I N G PLIO-PLEISTOCENE C A V E DEPOSITS OF I-IUMPATA, S O U T H E R N A N G O L A 23

o c . r " ~ '~, ~ - ~ m - " ~ . - ' -

0 ~= ._ = , ~

/

ug' ~ !

\

7

. _ r - l - . _

r " l 0 r 0,, ' OJ U. . 0 f - ~ O 0 . . ~

c o

i_ . - r

E

t-,

o

g �9

,...1 1


Remnant o f-~7-~ l 'ettow ,:tl oreccia on~-~ II dolomite -'u~" watt 2.Sin +'~{~,7

Pink breccid~ with stalogS~ mite frogs ~ on dotomite~ waLt 3m X, i

Fiowsfone on doLomite ,, wot[ (,m ~

j s

- ' < L~ r

Figure 6 - Tchiua Quarries.

2O ___J es

0 20

..~Store ~bKiin ,

Cercopithecids ~ ] . ' j

~ ' ~ /

/ / /

'i o m

"I trench ~ " ,"

-/-..,t,o+.,, / ~<+~

/ @ / k . . �9

Figure 7 - Locality map of Malola Kiln.

/- i ,J I/ 7----

~<'"~<""?'~ W. - y.o." /I ,'

"+ ....... " : '7 . FOSS'[S, Cove ' - ~ / , " / ' / / I

-~- '=;'?,. ;;" ,/i- cove I

(-~-/..: ..... :" & j' h i ....... - V . . " <+.<..,; , . - - ' . . - -

--r . !...~-.~" o 300 ,.~(.I "~''~

Figure 8 - Sketch map of the Ufefua Valley showing the locations of quarries, caves and fossil sites.

PRIMATE BEARING PL10-PLEISTOCENE CAVE DEPOSITS OF HUMPATA, SOUTHERN ANGOLA 25

Amphibia Reptilia

Chelonia Lacertidae OpLlidea

Ayes

Macroscelidiade Soricidae

Crocidura Rodentia Muridae

Uranomys Acomys Dasymys Aethomys i � 9 Thallomys Zelotomys .os I: Pelomys Malacomys Praomys Grammomys

Dendromuridae Dendromus Steatomys eetromyscus

Gerbillidae Tatera

Otomyinae Otomys

Bathyergidae Cryptomys

Gliridae Graphiurus

Hystricidae Hystrix

Chiroptera Rhinolophus Miniopterus Nycteris

Lagomorpha Serengetilagus

Primates Cercopithecoides Parapapio Dinopithecus

Carnivora Mustelidae Vivveridae Canldae Byaenidae

Hyracoidea Procavia Gigantohyrax

Rhinocerotidae Equidae Suidae

Metridlochoerus Bovldae

Hippotragini Connochaetes

Mollusca Achatina Edouardia Thapsla Subulinidae

~1 BI cl DI EI'FI ("l E} I I-J]

KEY Cangalongue 1

2 Cangalongue 2 3 Cangalongue 3 4 Cangalongue 4 5 Malola 6 Tchiua.(plnk breccia) 7 Tchiua (grey breccia) 8 Ufefua 2 9 Molo

A Malola Kiln B Malola 2S C Malola 2 l~Cangalongue 2 E Cangalongue I F Ufefua I G Ufefua S H Ufefua 8/9 I Molo J Cangalongue North

�9 Present o cf 2 2 species p~resent

4 species present

Table 1 - Fossil fauna of the Humpata Plateau, Southern Angola.

2 6 PICKFORD, MEIN and SENUT

the remainder being nocturnal or fossorial. The predominance of nocturnal species over diurnal ones, probably reflects the accumulation of fossils as owl pellets.

Most of the genera from Humpata are either widespread and tolerant of varied habitat conditions, or are restricted to sandy substrates in semi- arid steppes or wooded and grassland savannahs. Three species occur preferentially in swampy vegetation or along stream banks, and two genera are arboreal. Several of the Humpata species today inhabit rocky areas or arid country, whereas two of them occur today in much more humid areas in Eastern Zimbabwe and Mozambique (STUART • STUART, 1988).

The picture that emerges is that the fissures probably span a longish period of time, and that conditions which were initially somewhat humid began to dry out, with the result that different fissures contain faunas of slightly different aspect. However, a great deal of further work is needed on this aspect of the history of Humpata.

From a sedimentological and geomorphological perspective, there is little doubt that the Pliocene was a period of cavern and fissure formation, in which deposition was predominantly of sinter. A change in environment occurred near the Plio-Pleistocene boundary, after which pink sandy breccias were deposited, predominantly, though not exclusively, of aeolian origin. The onset of arid conditions appears to have been quite abrupt, and more or less contemporary over wide parts of southern Africa, for example in Botswana (PLc~ORO & MZ~N, 1988) and the Transvaal.

Fossil localities: Humpata region, Angola

Tchiua

13016 , 10"E : 15~ Two main quarries (Figure 6) oriented approximately north-south occur on opposite sides of

a ridge at Tchiua. The northern quarry is the one that has yielded most of the fossils, including all the primates. Most cercopithecid specimens occur in a distinctive grey-yellow indurated marl, mostly massive, but sometimes bedded, but also in bedded travertine rich in micromammals. Fossils tend to be black to mottled grey-black, and teeth are usually pale yellow. In addition, there are a few fossils in travertine and red breccia. The southern quarry is less fossiliferous, but yields blocks relatively rich in micromammals.

This site has sometimes been called "Leba", and this designation has, in the past, caused some confusion. Antunes (ms) for example, thought that the site of "Leba" was on the north side of the Leba Plateau, near the Fazenda da Leba, but as originally described by MotrrA (1950) and ARAMBOURG & MotrrA (1952), it is 5 km south of the fazenda.

Previous collections from this site consist of 19 cercopithecid specimens and several other mammal remains. During the 1990 field season, 200 blocks of fossiliferous sediment were collected, of which 36 contain cercopithecid skulls and mandibles. The remainder contain cercopithecid postcranial elements representing most of the skeleton.

Judging from the size of the quarry dumps, there is still a great deal of material awaiting collection.

Ufefua 1

13015 ' 15"E : 15~ Ufefua 1 (Fig. 8) is the largest of a series of 11 quarries in the Ufefua (or Mulola) Valley. It


contains much sinter, but has relatively little breccia. Only a few fossil gastropods (Achatina) were found at this site.

Ufefua 2

13~ : 15~ Ufefua 2 (Fig. 8) consists of a small quarry high up on a ridge on the opposite side of the

valley from Ufefua 1. It contains appreciable quantities of red breccia, which contains a variety of micromammals. This site yielded three specimens of unidentified animals in anatomical connec- tion, including a partial skeleton of a canid.

Tchivinguiro

13~ : 15~ This is the site of a spring (nascente) west of the Agricultural School at Tchivinguiro. Above

the spring are two small caves, the walls of which are partly comprised of pink fossiliferous breccia. Nothing was collected from this site.

Malola Kiln

13o18'10"E : 15o07'20"S Malola Quarry (Figure 7) from which much pink breccia has been removed, is oriented at

about 45 ~ magnetic. The fissure in which the sediment accumulated is evidently part of an extensive cavern system in the vicinity, of which a 30 metre deep cave and several subsidiary fissures occur in the neighbourhood.

Malola Kiln is notable for its exceptional concentrations of micromammals, but it yields relatively few large mammals. Among the latter, however, are Gigantohyrax, Procavia, a hyaenid and a papionine cercopithecid. At this prolific site, a great deal of breccia remains to be examined.

Half a kilometer downstream from Malola Kiln, is a second large quarry with red breccia, but relatively few fossils. A block of red breccia contained Achatina.

Cangalongue Cave

13~ : 15~ There are three main fossil quarries at Cangalongue Cave (Figure 5). All three fissure

systems are oriented at about 340 ~ magnetic. Despite their proximity, each quarry contains different kinds of breccia.

Quarry 1 contains pink breccia with termite holes overlying sinter deposits, and yields relatively few fossils.

Quarry 2 also contains pink breccia, but it appears to be waterlaid, and overlies a coarse breccia which contains broken stalagmites. This in turn overlies sinter. The pink breccia at this quarry is very rich in micromammals, but it also contains a variety of suids (Metridiochoerus andrewsi) and bovids (Connochaetes, Hippotragini). The skull and a tooth of a large mammal occur in situ in the wall of the quarry. Some of the large mammal fossils from this site have been gnawed by porcupines.

Quarry 3 is a large area of karstified breccia with abundant bone fragments. The breccia consists predominantly of broken stalagmites recemented by travertine, but there are small patches which are rich in micromammals. Many of the large mammal bones and teeth at this site are at an advanced weathering stage, akin to the "bone soup" that occurs at some of the Transvaal cave sites. A block of this breccia contains an edentulous mandible of a large cercopithecid, and a second block yielded a fragmentary M 3 of Metridiochoerus andrewsi.


PLATE I - Sub-adult cercopithecid skull f rom flowstone deposit at Tchiua, Angola: dorsal, right lateral and palatal views.


PLATE II - Top: Damaged/Cercopithecid snout with attached brain cast in calcified bat guamo, from grey fine- grained breccia, Tchiua, Angola: dorsal view. middle row left: infant cercopithecid skull fragment with deciduous dentition from grey fine-grained breccia and flowstone, Tchiua, Angola: right later view. middle row, right: infant cercopithecid snout with deciduous dentition, in block of grey breccia, Tchiua, Angola; left lateral view. botton row: same cercopithecid skull as in plate 1, anterior view.


PLA~ III - Top row: juvenile cercopithecid mandible with deciduous dentition in block of grey breccia, Tchiua, Angola, left lateral view. middle and bottom rows: Senile cercopithecid mandible, probably male, from pink breccia, Cangalongue, An- gola: right lateral and occlusal views.


PLATE IV - Top row: Cercopithecid scapula fragment from same flowstone block as the skull in plate 1: dorsal view. middle row: Cercopithecid distal femur in grey fine-grained breccia, Tchiua Angola. botton row: Metridiochoerus andrewsi, lower right M3, occlusal and lateral views, from Cangalongue, Angola.

32 PICKFORD, ME1N and SENUT

Cangalongue South (4)

13~ : 15~ About a kilometer south of Cangalongue Cave is a second vertical shaft cut through pink

breccia, and a subsidiary fissure, both of which are 10.5 metres deep. In the pink breccia occur micromammals and a few large mammal bones, including an equid metapodial. The pink breccia at this site appears to be wateflaid and closely resembles the breccia found in Quarry 2 further north.

Molo

13015 ' 15"E : 15~ About a kilometer west of the Leba Kiln is a large quarry facing the Namibe Road. It

contains large quantities of superficial pink breccia, but for the most part is poorly fossiliferous. However, several fossil snails, a few micromammals and large mammals, including equids, were found.

ACKNOWLEDGEMENTS - - Many people helped during various phases of the 1990 field season of the Angola Palaeontology Expedition. In France, Professor Yves Coppens and Professor Philippe Taquet provided funda- mental assistance. In Angola, we would like to thank Dr. Samuel A~O, the co-director of the expedition and Sr Joss Ferreira, Director of the Museu Regional da Huila. Others who were of assistance included Anna- Maria Oliveira, Manuela Batalha, Virgilio Coelho, Eleuterio Freire, Pedro Bondo, Rni Gomes, Ricardo Gomes, Ramos da Souza, Jofio Edouardo, Antonio Sakatengo, Shekulu Tchihonga, Junuario Katchatcha, and Tchipiqnita. Marcelina Gomes was also of assistance. We would also like to thank his excellency the Ambassador of France in Luanda, Mr. Stanislas Filliol and Mr. Etienne Roland (Conseiller de Coop6ration) and their staff for their interest in the project. We thank his Excellency, the Ambassador of The Popular Republic of Angola in Paris and his staff for their efficient and cheerful help. Finally, we are anxious to thank Mr Jack Blackshear, Director of CONOCO ANGOLA, and his staff for their help and hospitality.

References

AMARAL I., 1973. Nota sobre o "karst" ou carso do Planalto da Humpata (Huila), no Sudoeste de Angola Gaacia de Orta Lisb6a ,1 (2): 29-36.

ANTtJt,~S M.T., 1965. Stir la faune de vertdbr~s du PIgistockne de Leba, Humpata (Angola). Actes du V ~ Cong. Panaf. Preh. Quat. Stud., TEr,~,WE, 127-128.

ARAMBOURO C. & MOUTA F., 1952. Les grottes etfentes ?t ossements du sud de l'Angola. Acres Cong. Panafr. Pr6hist., Algeria, 2nd session, Comm. 12: 301-304.

BEETZ P., 1933. Geology of South West Angola, between Cunene and Lunda axis. Trans. Geol. Soc. S. Aft.Johan- nesburg, 36 : 137-176.

BREUIL H. & ALMEIDA A., 1964. lntroduc~o a prd-historia de Angola. Mem. Junta Invest. Ultramar., 2nd Her., 50: 157-163.

DART R., 1950. A note on the limestone caverns of Leba, near Humpata (Angola). S. Aft. Archaeol. Bull., V (20): 149-151.

ERVEDOSA C., 1980. Arqueologia Angolana. Min. Educ. Rep. Pop. Angola. FABER F.J., 1926. Bijdrage tot de Geologie van Zuid-Angola (Afrika). Delft, Technische Boekhandel. FRAN~A J., 1964. Nota preliminar sobre urea gruta prg-hist6rica do Planalto da Humpata (Angola). Junta Invest.

Ultramar, 2 (50): 59-67. HARRIS J.M. & WHITE T.D., 1979. Evolution of the Plio-Pleistocene African Suidae. Trans. Am. Phil. Soc., 69: 1-

126. KITCHING J.W., 1965. A new giant hyracoid from the Limeworks Quarry, Makapansgat, Potgietersrus. Paleont.

Aft., 9: 91-96.

PRIMATE BEARING PLIO-PLEISTOCENE CAVE DEPOSITS OF HUMPATA, SOUTHERN ANGOLA 3 3

KLZMME M., 1955. Southern Angola and the limestone industry. Re1. C6pia Serv. Geol. e Minas Angola, pp. 5-44. MASON R., 1976. Exploration archaeology of the Kaokoveld and southern Angola and the potential australopithecine

sites in the Serra da Chela massif southern Angola. Ann. S. Afr. Mus., 71: 215-223. MrNKo~ E., 1972. A fossil baboon from Angola, with a note on Australopithecus. J. Palaeont., 46 : 836-844. MOUTA F., 1950. Sur la prdsence de Quaternaire ancien clans les hauts plateaux du sud de l'Angola (Humpata-

Leba). C. R. Saturn. S6anc. Sac. G6ol. France, 14: 261-262. MotrrA F., 1954. Noticia explicativa do esbofo geologica de Angola 1 : 2 000 000. PICr;FORD M., FEP, SANDES T. & Afo S., 1990. Nouvelles ddcouvertes de remplissages defissures ?t primates dans

le "Planalto da Humpata ", Hufla, Sud de l'Angola. C. R. Acad. Sci. Paris, 310 : 843-848. PlCWORD M. & MZIN P., 1988. The discovery of fossiliferous Plio-Pleistocene cave fillings in Ngamiland, Boswana.

C. R. Acad. Sci. Paris, 307 : 1681-1686. STUART C., • STUART T., 1988. Field Guide to the Mammals of Southern Africa. Cape Town Struik. VALE F., GONfALVES F. & SIMOES M., 1968. Noticia explicativa dafolha n ~ 355 (Humpata - Cainde). Serv. Geol. e

Minas Angola. VRBA E.S., 1982. Biostratigraphy and chronology, based particularly on Bovidae, of Southern African hominid-

associated assemblages. Act. W Cong. Intern. Pal6ont. Hum., 2: 707-752. H de Lumley & M.-A. de Lumley (eds.) U. I. S. P. P.

WELBOUP, NE R., 1976. Remarks on the fossil fauna of the Serra da Chela massif Southern Angola. Ann. S. Afr. Mus., 71 : 223-224.

Primate bearing Plio-Pleistocene cave deposits of Humpata, Southern Angola

Documents