No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Stratigraphic interpretation of the Kulu Formation (Early Miocene, Rusinga Island, Kenya) and its implications for primate evolution
Abstract
Early Miocene fossils from Rusinga Island, Kenya, provide some of the best evidence for catarrhine evolution and diversification, and, together with more than eighty-five other mammalian species, form an important comparative reference for understanding faunal succession in East Africa. While there is consensus over the stratigraphic position of most of Rusinga's volcaniclastic deposits, the lacustrine Kulu Formation has been placed in various parts of the geological sequence by different researchers. To resolve this discrepancy, we conducted detailed geological analyses which indicate that the Kulu Formation was formed in the Early Miocene during a period of volcanic inactivity and subsidence following the early, mainly explosive hyper-alkaline phase of the Kisingiri complex and prior to the final eruptions of nephelinitic lavas. The underlying Hiwegi and older formations were locally deformed and deeply eroded before sedimentation began in the Kulu basin, so that the Kulu sediments may be significantly younger than the 17.8 Ma Hiwegi Formation and not much older than the overlying Kiangata Agglomerata-Lunene Lava series, loosely dated to ca. 15 Ma. The overall similarities between Kulu and Hiwegi faunas imply long-term ecological stability in this region. Our stratigraphic interpretation suggests that the Kulu fauna is contemporaneous with faunas from West Turkana, implying that differences between these assemblages-particularly in the primate communities--reflect paleobiogeographic and/or paleocological differences. Finally, the position of the Kulu Formation restricts the time frame during which the substantial faunal turnover seen in the differences between the primate and mammalian communities of Rusinga and Maboko Islands could have occurred.
... Inset map shows the location of Rusinga Island within Africa. to youngest): the Wayando Formation, Kiahera Formation, Rusinga Agglomerate, Hiwegi Formation, and Kulu Formation (e.g., Van Couvering, 1972;Drake et al., 1988;Peppe et al., 2009; but see Bestland, 1991). The Kisingiri Group contains the Kiangata Agglomerate and Lunene Lavas (Van Couvering, 1972). ...
... He also suggested the possibility of older sediments on Rusinga's so-called Gumba Peninsula (Fig. 1, points 5-8). Kent (1944: 89) characterized the Basal Agglomerate as an "altered (Bestland and Retallack, 1993;Drake et al., 1988;Kent, 1944;Peppe et al., 2009;Pickford, 1984Pickford, , 1986Shackleton, 1951;Van Couvering, 1972;Whitworth, 1953) nephelinite block in a similar matrix" with frequent staining. Importantly, his classification of the Agglomerate based on rock type seems to have resulted from conflating the younger Rusinga Agglomerate deposits with conglomerates in the Kiahera Formation, as evidenced by his description of idiomorphic melanite garnetscharacteristic of the Rusinga Agglomerate (Van Couvering, 1972) -found in the matrix, but only in some places. ...
... Subsequent stratigraphic work on Rusinga Island mostly focused on other units, with some reorganization or correlation to Drake et al.'s (1988) sequence of formations (Bestland, 1991;Michel et al., 2014Michel et al., , 2020Peppe et al., 2009;Retallack et al., 1995). The most recent study of the Kiahera Formation sedimentology was done by Bestland and Retallack (1993) who examined the type locality at Kiahera Hill. ...
... The early Miocene sedimentary succession on Rusinga is composed of the Rusinga and Kisingiri groups (Fig. 1), with the majority of fossils coming from the older Rusinga Group. The Rusinga Group succession consists of pyroclastic, fluvial, lacustrine and lahar deposits that record active volcanism and periodic landscape stability (Van Couvering & Miller, 1969;Van Couvering, 1972;Bestland et al., 1995;Bestland & Krull, 1999;Peppe et al., 2009) and is divided into (from oldest to youngest): the Wayando Formation, Kiahera Formation, Rusinga Agglomerate, Hiwegi Formation and Kulu Formation (Van Couvering, 1972;Peppe et al., 2009). Apropos the current study, the Hiwegi Formation has been further subdivided into the Kaswanga Point, Grit, Fossil Bed and Kibanga members ( Fig. 1; Van Couvering, 1972;Drake et al., 1988). ...
... The early Miocene sedimentary succession on Rusinga is composed of the Rusinga and Kisingiri groups (Fig. 1), with the majority of fossils coming from the older Rusinga Group. The Rusinga Group succession consists of pyroclastic, fluvial, lacustrine and lahar deposits that record active volcanism and periodic landscape stability (Van Couvering & Miller, 1969;Van Couvering, 1972;Bestland et al., 1995;Bestland & Krull, 1999;Peppe et al., 2009) and is divided into (from oldest to youngest): the Wayando Formation, Kiahera Formation, Rusinga Agglomerate, Hiwegi Formation and Kulu Formation (Van Couvering, 1972;Peppe et al., 2009). Apropos the current study, the Hiwegi Formation has been further subdivided into the Kaswanga Point, Grit, Fossil Bed and Kibanga members ( Fig. 1; Van Couvering, 1972;Drake et al., 1988). ...
... Although there has been some debate about the order of the stratigraphic units on the island (e.g. Van Couvering, 1972;Drake et al., 1988;Bestland, 1991;Peppe et al., 2009;Michel et al., 2014), recent stratigraphic and palaeomagnetic work (Peppe et al., 2009(Peppe et al., , 2017b confirmed the basic sequence of formations mapped by Van Couvering (1972) and Drake et al. (1988), and made some important changes to the stratigraphic positions of specific fossil sites Michel et al., 2014;Peppe et al., 2017b). Drake et al. (1988) suggested that the Kiahera, Rusinga Agglomerate and Hiwegi formations record the early explosive eruptive history of the Kisingiri Volcano, followed by a period of volcanic quiescence during the deposition of the Kulu Formation, and then a thick series of extrusive volcanics in which (on Rusinga Island) the Kiangata Agglomerate is capped by the multiple lava flows of the Lunene Lavas. ...
Palaeontological deposits on Rusinga Island, Lake Victoria, Kenya, provide a rich record of floral and faunal evolution in the early Neogene of East Africa. Yet, despite a wealth of available fossil material, previous palaeoenvironmental reconstructions from Rusinga have resulted in widely divergent results, ranging from closed forest to open woodland environments. Presented here is a detailed study of the sedimentology and fauna of the early Miocene Hiwegi Formation at Waregi Hill on Rusinga Island, Kenya. New sedimentological analyses demonstrate that the Hiwegi Formation records an environmental transition from the bottom to the top of the formation. Lower in the Hiwegi Formation, satin‐spar calcite after gypsum in siltstone deposits are interpreted as evidence for open hypersaline lakes. Moving up‐section, carbonate deposits – interpreted previously as evidence of aridity – are actually diagenetic calcite cements, which preserve root systems of trees, suggesting a more closed environment; further up‐section, the uppermost palaeosol layer contains abundant root traces and tree‐stump casts, previously reported by as evidence of a closed‐canopy forest. These newly interpreted environmental differences are reflected by differences in faunal composition and abundance data from Hiwegi Formation fossil sites R1 and R3. Taken together, this work suggests that divergent palaeoenvironmental reconstructions in previous studies may have been informed by time‐averaging across multiple environments. Further, results demonstrate that during the early Miocene local or regional habitat heterogeneity already existed. Rusinga’s Hiwegi Formation varied both spatially and temporally, which challenges the interpretation that a broad forested environment stretched across the African continent during the early Neogene, transitioning later to predominately open landscapes that characterize the region today. This result has important implications for interpretations of the selective pressures faced by early Miocene fauna, including Rusinga Island’s well‐preserved fossil primates.
... Further studies have since demonstrated that faunas across East Africa were not stable throughout the early Miocene. The discovery of early Miocene localities in the Turkana Basin (Arambourg, 1933;Savage and Williamson, 1978;Boschetto et al., 1992;Leakey et al., 2011), as well as additional fossil collecting at Kisingiri and Tinderet in Western Kenya, and at Napak and Moroto in Uganda, have provided unequivocal evidence for faunal evolution throughout the early Miocene ( Fig. 1; Pickford and Andrews, 1981;Pickford, 1986b;Pickford et al., 1986a, b;Drake et al., 1988;Cote, 2008;Peppe et al., 2009). Unfortunately, this perceived faunal change remains untethered to accurate absolute ages at several localities. ...
... More recently, Peppe et al. (2011Peppe et al. ( , 2016Peppe et al. ( , 2017) have begun to reassess the age of the Kisingiri sequence using updated dating techniques. Their preliminary dating results using 40 Ar/ 39 Ar and paleomagnetism indicate that the oldest levels (Wayondo Formation) may be as old as 20 Ma, and that the top of the Hiwegi Formation is approximately 18 Ma (Peppe et al., 2009(Peppe et al., , 2011(Peppe et al., , 2016(Peppe et al., , 2017McCollum et al., 2012). This indicates that deposition of the Kisingiri sequence may have occurred over a substantially longer period of time than suggested by Drake et al. (1988) and that previous analyses of the Hiwegi fossil faunas that have assumed nearcontemporaneity of Hiwegi Formation localities may be inaccurate. ...
... This is three million years younger than previously reported K-Ar dates, but is supported by mammalian biostratigraphic correlations with other fossil localities in the region. Faunal similarities between Bukwa and Kisingiri corroborate preliminary findings that the fossil localities at Rusinga and Mfangano range in age from~20 to 18 Ma and are not all~1 7.8 Ma, as was previously thought (Peppe et al., 2009(Peppe et al., , 2011(Peppe et al., , 2016(Peppe et al., , 2017McCollum et al., 2012). Based on our faunal comparisons, we anticipate that re-dating of the Tinderet deposits using 40 Ar/ 39 Ar, and additional 40 Ar/ 39 Ar dating to confirm our initial age for Napak (MacLatchy et al., 2006), will confirm that these two sequences arẽ 20 Ma. ...
Field expeditions to Bukwa in the late 1960s and early 1970s established that the site had a small but diverse early Miocene fauna, including the catarrhine primate Limnopithecus legetet. Initial potassium-argon radiometric dating indicated that Bukwa was 22 Ma, making it the oldest of the East African early Miocene fossil localities known at the time. In contrast, the fauna collected from Bukwa was similar to other fossil localities in the region that were several million years younger. This discrepancy was never resolved, and due to the paucity of primate remains at the site, little subsequent research took place.
We have collected new fossils at Bukwa, reanalyzed the existing fossil collections, and provided new radiometric dating. ⁴⁰Ar/³⁹Ar incremental heating ages on lavas bracketing the site indicate that the Bukwa fossils were deposited ∼19 Ma, roughly 3 Ma younger than the original radiometric age. Our radiometric dating results are corroborated by a thorough reanalysis of the faunal assemblage. Bukwa shares taxa with both stratigraphically older localities (Tinderet, Napak) and with stratigraphically younger localities (Kisingiri, Turkana Basin) perfectly corresponding to our revised radiometric age.
This revised age for Bukwa is important because it indicates that significant faunal turnover may have occurred in East Africa between 20 and 19 Ma. Bukwa samples immigrant taxa such as large suids, large ruminants, and ochotonids that are absent from stratigraphically older but well-sampled localities in the region, such as Tinderet (∼20 Ma) and Napak (20 Ma). Further age refinements for Bukwa and the entire East African early Miocene sequence will help to constrain the timing of this faunal turnover event, of particular importance in paleoanthropology since this temporal sequence also provides us with what is currently our best window into the early evolution of cercopithecoid and hominoid primates.
... The Kisingiri Volcano, which formed at the western extent of Kenya's Nyanza Rift, is associated with some of the richest fossil assemblages from the early Miocene in East Africa ( Fig. 1; Pickford, 1986;Drake et al., 1988;Peppe et al., 2009;Michel et al., 2014;Driese et al., 2016). The best-known Kisingiri locality, Rusinga Island, has been the focus of extensive paleontological and geological research, spanning more than eight decades and producing tens of thousands of fossil remains. ...
... More recent analyses, however, indicate that Rusinga's mammalian faunas were deposited over as much as a few million years, beginning »20 Ma in the Wayando Formation until ca. 17 Ma in the Kulu Formation (Peppe et al., 2009;McCollum et al., 2013). It is important to note that these new age constraints are still preliminary, but they indicate a much longer depositional time during the early Miocene of Rusinga. ...
... The Gumba Beds are likely younger than the Kiahera Formation. Rusinga stratigraphy from Peppe et al. (2009). Gray star indicates stratigraphic position of an ash dated to ca. 18 Ma (McCollum et al., 2013); D, stratigraphy for the Karungu Miocene localities. ...
We describe new material of Rhinocerotidae recently collected in western Kenya. A skull from Karungu is one of the best-preserved Miocene skulls in Africa. It differs substantially from that of Rusingaceros leakeyi, the only other relatively well-known rhino from this region and age, in its degree of brachycephaly, possession of a deep nasal notch, and long nasal bones that probably carried a horn of moderate size. Miocene African rhinos are still too poorly known to resolve their phylogenetic relationships, but we tentatively assign this skull to a new species of Victoriaceros, a genus whose type species comes from the younger site of Maboko, although the Karungu skull has a much smaller nasal horn. A parsimony analysis resolves them as sister species within the Elasmotheriini, close to the other African genera Turkanatherium and Chilotheridium, but we consider this result debatable, as Victoriaceros differs considerably from them. Still, they might all be descended from European forms. A partial skull from Gumba is assigned to the Aceratheriini, making it one of the earliest representatives of this group and suggesting that the origin of this tribe could be African.
http://zoobank.org/urn:lsid:zoobank.org:pub:2B1E8135-CCD4-43EB-826B-6DF7176DC74E
SUPPLEMENTAL DATA—Supplemental materials are available for this article for free at www.tandfonline.com/UJVP
Citation for this article: Geraads, D., T. Lehmann, D. J. Peppe, and K. P. McNulty. 2016. New Rhinocerotidae from the Kisingiri localities (lower Miocene of western Kenya). Journal of Vertebrate Paleontology. DOI: 10.1080/02724634.2016.1103247.
... (= Mfangano) Island (approx. 0°28´S, 34°01´E) in Lake Victoria, southwestern Kenya (Drake et al., 1988; Peppe et al., 2009). Horizon and age.—Rusinga ...
... Horizon and age.—Rusinga Group (Formation is unknown) (Figure 3; Drake et al., 1988; Peppe et al., 2009Drake et al., 1988; Pickford, 2007a; Peppe et al., 2009 Peppe et al., , 2011). These two islands are approximately 10 km apart. ...
... Horizon and age.—Rusinga Group (Formation is unknown) (Figure 3; Drake et al., 1988; Peppe et al., 2009Drake et al., 1988; Pickford, 2007a; Peppe et al., 2009 Peppe et al., , 2011). These two islands are approximately 10 km apart. ...
A trigonid of a lower molar of a primitive, large hippopotamus from the upper lower Miocene of Mfwangano Island in southwestern Kenya is described. The molar trigonid is similar in size to that of living hippopotamuses, and is comparable in morphology to that of kenyapotamine hippopotamids (Mammalia, Cetartiodactyla) in having a brachyodont crown, bunodont cusps, an M-like structure on the distal trigonid wall, a single-ridged premetacristid, and a buccolingually bifurcate mesial root, and in lacking a paraconid. On the basis of its size and morphology, the specimen appears to be assignable to Kulutherium, which is a putative kenyapotamine previously known from the upper lower Miocene of Kenya and is so far represented only by the upper dentition. The present specimen provides additional evidence that a hippopotamus-sized, large hippopotamid was already living during the early Miocene. If it proves to be Kulutherium, it provides additional evidence that Kulutherium should be assigned to the Kenyapotaminae.
... This means we can expect sites that are closest in age to have more similar community composition as long as they are in roughly similar habitats. Similarly, as geographical distance increases, the dispersal propensity of species lessens, possibly because of an increased chance of encountering geographic barriers or unsuitable habitat, which affects the similarity of community References for habitat reconstructions: Andrews 1992 Andrews , 1996; Andrews and Van Couvering 1975; Andrews and Walker 1976; Andrews et al. 1979 Andrews et al. , 1981 Andrews et al. , 1997 Behrensmeyer et al. 2002; Cerling et al. 1991 Cerling et al. , 1992 Cerling et al. , 1997 Cote 2008; Gentry 1970; Grossman 2008; Hill et al. 2013; Kappelman 1991; Kortlandt 1983; Leakey et al. 2011; Madden 1972; Miller 1999; Michel et al. 2014; Miller and Wood 2010; Nesbit Evans et al. 1981; Peppe et al. 2009; Pickford 1981 Pickford , 1983 Pickford , 1987 Pickford , 1995 Pickford and Andrews 1981; Pickford and Mein 2006; Pickford et al. 2003; Retallack 1991 Retallack , 1992a Retallack et al. 2002; Shipman 1986; Shipman et al. 1981 composition (Beaudrot and Marshall 2011; Beaudrot et al. 2014; Kamilar 2009; Soininen et al. 2007). Thus, identifying the mammalian communities of Early and Middle Miocene catarrhine-bearing sites and comparing them across time and space will provide important information about the forces that shape the taxonomic composition of early catarrhine communities. ...
... 3). Comparing the two Rusinga assemblages, only one genus, Turkanatherium acutirostratum, is reportedly present at Kulu (Peppe et al. 2009) but is not present in the Hiwegi assemblage. However, Geraads (2010) argues that Turkanatherium acutirostratum cannot be identified anywhere but Moruorot (he had no access to the Kalodirr fossils at that time), which would then make the two Rusinga assemblages identical at the genus level. ...
... The similarities among the two sites may be a result of sampling size (Kulu = 26 genera; Hiwegi = 68 genera), but may also be the result of historically treating the many localities at Rusinga Island as a single time-averaged fauna (Michel et al. 2014). It seems unlikely that two faunal assemblages a million or more years apart (Peppe et al. 2009Peppe et al. , 2011 will be identical and perhaps there is a need for reanalysis of the Rusinga faunas, especially given the importance of these localities to understanding of the Early Miocene, as demonstrated by Michel et al. (2014). The small size of the Loperot assemblage (19 genera) may affect the analysis. ...
The site of Loperot in West Turkana, Kenya, is usually assigned to the Early Miocene. Recent discoveries at Loperot, including catarrhine primates, led to a revision of its mammalian fauna. Our revision of the fauna at Loperot shows an unusual taxonomic composition of the catarrhine community as well as several other unique mammalian taxa. Loperot shares two non-cercopithecoid catarrhine taxa with Early Miocene sites near Lake Victoria, e.g., Songhor and the Hiwegi Formation of Rusinga Island, but Loperot shares a cercopithecoid, Noropithecus, with Buluk (Surgei Plateau, near Lake Chew Bahir). We use Simpson's Faunal Resemblance Index (Simpson's FRI), a cluster analysis, and two partial Mantel tests, to compare Loperot to 10 other localities in East Africa representing several time divisions within the Early and Middle Miocene. Simpson's FRI of mammalian communities indicates that Loperot is most similar in its taxonomic composition to the Hiwegi Formation of Rusinga Island, suggesting a similarity in age (≥18 Ma) that implies that Loperot is geographically distant from its contemporaries, i.e.
... Lists of taxa per fossil fauna were compiled from primary literature of Australia (Marshall, 1975;Merrilees, 1975;Hope, 1978;Rich, 1991;Long et al., 2002;Megirian et al., 2004;, Kenya (Madden, 1972;Isaac, 1978;Savage and Willamson, 1978;Pickford and Andrews, 1981;Thomas, 1981;Hill et al., 1985Hill et al., , 2001Pickford, 1986Pickford, , 2000Pickford, , 2001aPickford, , 2001bPickford, , 2007Retallack, 1991a;Nakaya, 1994;Leakey et al., 1995;Retallack et al., 1995;Hill, 1996;Behrensmeyer et al., 2002;Retallack et al., 2002;Leakey and Harris, Fig. 4. Australian soils of dry woodland mallee and desert shrubland examined (A-J: Table 3), and early Eocene paleosols from the Willwood Formation, near Worland, Wyoming, U.S.A. (Retallack, 1998) and a Cretaceous (Albian) paleosol from the Mussentuchit Member, Cedar Mountain Formation, near Cedar Mountain, Utah, U.S.A. (Retallack, 2009a(Retallack, ). et al. (2006, Ambrose et al. (2007), and Peppe et al. (2009). Ages of Australian sites are from Metzger and Retallack (2010) and of Rocky Mountain sites from Retallack (2007bRetallack ( , 2009a. ...
... In cases where mammals are preserved in calcareous paleosols (Bown, 1979;Retallack, 1991a), a more appropriate comparison of both paleosols and mammal faunas is with Australian mallee. Mallee-like mammal faunas and soils identified in this paper include the following: mid-Cretaceous (98 Ma) Mussentuchit Member of Cedar Mountain Formation in Utah ( Fig. 4M: Cifelli et al., 1999;Retallack, 2009a), early Eocene (54 Ma) Willwood Formation in Wyoming (Fig. 4K-L: Bown, 1979;Bown et al., 1994;Retallack, 1998;Kraus and Riggins, 2007), early Miocene (18 Ma) Hiwegi Formation on Rusinga Island Kenya (Retallack et al., 1995;Peppe et al., 2009), and early Miocene (22 Ma) Ulta Formation near Kangaroo Well, Northern Territory, Australia, and middle Miocene (16 Ma) uppermost Etadunna Formation at Lake Palankarinna, South Australia (Metzger and Retallack, 2010). These were not rainforest mammal faunas, and do not demand extreme hydrological and paleoclimatic changes in the past compared with the present (Herold et al., 2011). ...
... 2003; Pickford and Kunimatsu, 2005;Pickford and Senut, 2005; Tsujikawa, 2005a,b;Ambrose et al., 2007;Peppe et al., 2009) and the United States (Douglass, 1903;MacDonald, 1949;Black, 1961;Munthe, 1988;Rasmussen, 1989;Emry, 1990;Stucky et al., 1990;Bown et al., 1994;Gunnell, 1994;McDonald et al., 1996;Prothero and Emry, 1996;Cross and Yi, 1997;Pinsof, 1998;Cifelli et al., 1999;Rasmussen et al., 1999;Reynolds and Lindsay, 1999;Turner and Peterson, 1999;Hill, 2001;Tabrum et al., 2001;Ackersten et al., 2002;Sankey, 2002;Foster, 2003;Gingerich, 2003;Kielan-Jaworowska et al., 2004; Eaton, 2006a,b;Barnosky et al., 2007;Retallack, 2007b). ...
Mallee is an endemic Australian woodland and shrubland of semi-arid, summer–dry regions between dry woodland and desert shrubland. In other parts of the world, such as Africa and the Americas, such climatic regions support grassland ecosystems. Using Australian and African climofunctions and models gives very different reconstructions of paleoclimate (subhumid versus perhumid) and paleovegetation (woodland versus rainforest) for North American fossil mammal faunas before Cenozoic evolution of grassland ecosystems. Modern mammal faunas of Africa and Australia have different ecological spectra of taxonomic units, body size, feeding, and locomotion of species of mammals on precipitation gradients. Gradients in proportions of such categories yield transfer functions for mean annual precipitation from percent species of Artiodactyla or Macropodidae, percent species of moderately large animals (45–180 kg), percent species of arboreal mammals, and percent species of grazers. These transfer functions can be applied to fossil mammal faunas to estimate paleoprecipitation in Africa, Australia and North America. Modern transfer functions match well paleoprecipitation estimates based on depth to calcic horizons in paleosols at the same localities in Kenya and inland Australia back through the Miocene. For fossil mammal faunas of the Rocky Mountain region of North America, African transfer functions fail, but Australian transfer functions predict paleoprecipitation back to the Cretaceous–Tertiary boundary (66 Ma). Furthermore, modern mallee soils investigated in this study closely match Cretaceous to Eocene paleosols of the Rocky Mountains. Extinct mallee-like vegetation, such as pori woodlands of Kenya and cunhaka woodlands (newly defined) of the Rocky Mountains better explains the dominance of small, nocturnal, insectivorous, arboreal mammals of Paleogene and Mesozoic mammal faunas, than comparisons with African grassland or rainforest faunas.
... Most of these have been assigned to three distinct taxa, Kalyptochromis hamulodentis, Nderechromis cichloides, and Palaeofulu kuluensis (Van Couvering, 1982). The Kulu Formation can be no older than 17.8 Ma and is generally thought to have been deposited between 17 and 15 Ma (Peppe et al., 2009). Articulated fossil cichlids have also been found in the Arabian Peninsula in the Baid Formation, which is likely early Miocene in age (Madden et al., 1983), although it has also been considered Oligocene (e.g., Micklich and Roscher, 1990). ...
... Van Couvering (1982) noted that the fishes from the Bukwa locality did not indicate that this might have been an alkaline lake deposit; presumably, her observation was to contrast Bukwa with some other East African Miocene sites in the Kenya rift that have a hyperalkaline chemistry (Peppe et al., 2009). Modern alkaline lakes do contain some cichlid species (e.g., Trewavas, 1983), but the diversity in such environments is very low. ...
Renewed research at the early Miocene fossil site of Bukwa in northeastern Uganda has resulted in new fossil finds, including fish, with representatives of two families, Cichlidae and Alestidae. Although the two families were previously briefly reported from Bukwa, we here give a more detailed account of the fishes based on newly collected material. The cichlid material, mainly composed of vertebrae, can be tentatively assigned to one or more species of Pseudocrenilabrinae. The alestid material, comprising a diversity of teeth, likely represents several different species of Alestes, Brycinus, and/or Bryconaethiops. Although the ichthyofaunal diversity of Bukwa is low, the fishes are important for indicating the paleoenvironment and hydrographic connections of Bukwa. The early Miocene was a critical time for African faunas, because it was during this time that the Afro-Arabian and Eurasian plates came into contact with one another, ending the long isolation of Africa, which, along with rifting in East Africa, created new terrestrial and hydrological connections allowing faunal interchanges. Bukwa is one of only a few African early Miocene localities known that sample fish and, based on these fish, the site probably represents an area of interconnected lakes and large rivers, including floodplains.
SUPPLEMENTAL DATA—Supplemental materials are available for this article for free at www.tandfonline.com/UJVP
Citation for this article: Murray, A. M., T. Argyriou, S. Cote, and L. MacLatchy. 2017. The fishes of Bukwa, Uganda, a lower Miocene (Burdigalian) locality of East Africa. Journal of Vertebrate Paleontology. DOI: 10.1080/02724634.2017.1324460.
... The early Miocene fossil localities along the northeastern margin of Lake Victoria, Kenya are among the richest and best collected early Miocene paleontological sites in eastern Africa (e.g., Pickford, 1986;Drake et al., 1988;Peppe et al., 2009;Michel et al., 2014;Lehmann et al., 2014). These fossil sites on Rusinga and Mfangano Islands and near Karungu, mainland Kenya are all associated with the Kisingiri volcano and the western-most extent of Kenya's Nyanza Rift (Fig. 1). ...
... Oswald (1914), who conducted the earliest, and only extensive, geological study of the Karungu fossil sites, interpreted the deposits to be lacustrine and representing a deepening lake succession. Historical (e.g., Andrews, 1911Andrews, , 1914Newton, 1914;Oswald, 1914), and more recent paleontological collections by our research team (e.g., Lehmann et al., 2014) indicate an abundance of aquatic vertebrate taxa, such as crocodiles, turtles, and fish, which broadly agrees with Oswald's (1914) environmental interpretation and contrasts with the more terrestrialdominated vertebrate collections from the similarly-aged, nearby Miocene fossil sites at Rusinga and Mfangano Islands (e.g., Pickford, 1986;Peppe et al., 2009;Ungar et al., 2012;Michel et al., 2014). ...
... Fossil collections from the early Miocene deposits on Rusinga Island, Lake Victoria, Kenya (Figure 1) provide some of the best evidence of East African paleocommunities immediately following the connection of Africa to Eurasia (e.g., Savage, 1965; Pickford, 1986 Pickford, , 2004 Schmidt-Kittler, 1987; Cote et al., 2007; Drake et al., 1988; Peppe et al., 2009; Peppe et al., 2011). Rusinga Island is particularly well known for its abundant well-preserved fossil catarrhine primates, Proconsul, Nyanzapithecus , Limnopithecus, and Dendropithecus (e.g., MacInnes, 1943; Le Gros Clark and Leakey, 1950; Andrews and Simons, 1977; Walker et al., 1993). ...
... K-Ar dates published by Drake et al. (1988) suggested that the Hiwegi Formation was deposited ~17.9 Ma, and that the entire fossiliferous Rusinga Group sequence (Figure 1) was deposited in less than a half million years. More recent analyses using 40 Ar/ 39 Ar dates, magnetostratigraphy, and lithostratgiraphy demonstrate that the fossiliferous strata on Rusinga were deposited over a much longer time interval, between ~17-20 Ma (Peppe et al., 2009; Peppe et al., 2011; McCollum et al., 2012). ...
Early Miocene deposits on Rusinga Island (Lake Victoria, Kenya) contain an abundance of faunal and floral remains. Despite the attention that has historically been given to the early Miocene fauna from Rusinga Island, little attention has been given to the early Miocene fossil floras and to date no studies have described fossil leaf morphotypes from Rusinga Island. Here, we present a morphotype catalog of fossil leaves collected from the Grit Member of the Hiwegi Formation on Rusinga Island. We describe 14 morphotypes, comprised of 12 dicotyledonous angiosperms and two monocotyledonous angiosperms, as well as two distinct dicotyledonous angiosperm leaf fragments. Characteristics of the flora and sedimentological evidence, coupled with previous research, suggest that the local paleoenvironment was a riparian habitat within a patchwork of woodland and forested biomes in what was likely a warm climate. This work represents an important first step in understanding the early Miocene vegetation of Rusinga Island, and highlights both the need and potential for future research on these early Miocene floras.
... The early Miocene fossil localities along the northeastern margin of Lake Victoria, Kenya are among the richest and best collected early Miocene paleontological sites in eastern Africa (e.g., Pickford, 1986;Drake et al., 1988;Peppe et al., 2009;Michel et al., 2014;Lehmann et al., 2014). These fossil sites on Rusinga and Mfangano Islands and near Karungu, mainland Kenya are all associated with the Kisingiri volcano and the western-most extent of Kenya's Nyanza Rift (Fig. 1). ...
... Oswald (1914), who conducted the earliest, and only extensive, geological study of the Karungu fossil sites, interpreted the deposits to be lacustrine and representing a deepening lake succession. Historical (e.g., Andrews, 1911Andrews, , 1914Newton, 1914;Oswald, 1914), and more recent paleontological collections by our research team (e.g., Lehmann et al., 2014) indicate an abundance of aquatic vertebrate taxa, such as crocodiles, turtles, and fish, which broadly agrees with Oswald's (1914) environmental interpretation and contrasts with the more terrestrialdominated vertebrate collections from the similarly-aged, nearby Miocene fossil sites at Rusinga and Mfangano Islands (e.g., Pickford, 1986;Peppe et al., 2009;Ungar et al., 2012;Michel et al., 2014). ...
A 50 m thick stratigraphic section at Ngira, near Karungu on the shore of Lake Victoria in western Kenya, documents the early Miocene paleoenvironments of the area. The basal Ngira paleosol is a 7.6 m thick, oxisolic Vertisol that formed during a prolonged period of pedogenesis; it began as a smectite-dominated Vertisol that was later overprinted through polypedogenesis to become a kaolinitic paleosol highly depleted of all base cations, with abundant Fe concentrations and depletions, and complexly variegated color mottle patterns that reflect extensive ferruginization. Paleoenvironmental reconstructions using bulk geochemistry indicate warm and wet conditions during development of the Ngira paleosol that probably supported a tropical seasonal forest on a stable upland surface for 10s to 100s of thousands of years. Following this long-lived stable landscape, rapid subsidence, perhaps associated with slip on a high-angle fault associated with the onset or progression of the Nyanza Rift and/or the development and eruptive history of the nearby Kisingiri volcano, buried the paleosol and formed a nascent lake basin that experienced multiple transgressions and regressions. During one interval of regression, fluvial sandstones and conglomerates were deposited along with fluvio-lacustrine sandstones and claystones that include weakly developed paleosols. These weakly-developed paleosols indicate a relatively dry paleoenvironment with seasonal precipitation, and probably a shrubland/bushland or riparian forest habitat. Important terrestrial and aquatic vertebrate fossils are primarily preserved within fluvial and fluvio-lacustrine deposits, indicating that the terrestrial Karungu fauna lived in a relatively dry and open habitat. This study demonstrates polypedogenesis and inferences regarding onset of abrupt tectonic activity in the early Miocene in equatorial eastern Africa, and emphasizes the contrasts between landscape stability of the Ngira paleosol and the poorly developed soils in the fluvio-lacustrine facies.
... Likewise, Pickford (1986) recognized only P. nyanzae from Rusinga and Mfangano, notably highlighting the distinct morphology of Kisingiri Proconsul when compared to Tinderet and Ugandan specimens. Despite these and other objections, the presence of P. nyanzae and P. heseloni at the Kisingiri localities has come to be universally accepted (Begun et al., 1994;Leakey et al., 1995;Rafferty et al., 1995;Ward et al., 1995;MacLatchy and Bossert, 1996;Harrison, 2002Harrison, , 2010Smith et al., 2003;Ishida et al., 2004;Nakatsukasa et al., 2007;Deane, 2009;Harrison and Andrews, 2009;Peppe et al., 2009;Pickford et al., 2009;Michel et al., 2014). ...
... Rather than being younger than all of the Tinderet sites, the Kisingiri stratigraphic sequence is substantially longer than previously thought, with the oldest fossil strata contemporaneous with those from Koru and Legetet Hill at approximately 20e19 Ma (McCollum et al., 2013). The youngest Miocene fauna from Rusinga is not associated with radiometric dates, but magnetostratigraphic analysis suggests an age near 17 Ma or younger (Peppe et al., 2009). Hence, differences between species of Proconsul, and in particular between assemblages from Tinderet and Kisingiri, cannot be simply explained by chronology. ...
The Gregory Rift Valley is known for many Plio-Pleistocene sites that have yielded abundant fossil hominins, most of which lie on the Rift Valley floor. Here we report a new Pliocene site, Kantis, on the shoulder of the Gregory Rift Valley, which extends the geographical range of Australopithecus afarensis to the highlands of Kenya. This species, known from sites in Ethiopia, Tanzania, and possibly Kenya, is believed to be adapted to a wide spectrum of habitats, from open grassland to woodland. The Kantis fauna is generally similar to that reported from other contemporaneous Au.afarensis sites on the Rift Valleyfloor. However, its composition and stable carbon isotopic data from dental enamel suggest a higher component ofC4 vegetationthan in thesesites. Although the Gregory Rift Valley has been the focus of paleontologists’ attention for many years, surveys of the Rift shoulder may provide new perspective on African Pliocene mammal and hominin evolution.
... Likewise, Pickford (1986) recognized only P. nyanzae from Rusinga and Mfangano, notably highlighting the distinct morphology of Kisingiri Proconsul when compared to Tinderet and Ugandan specimens. Despite these and other objections, the presence of P. nyanzae and P. heseloni at the Kisingiri localities has come to be universally accepted (Begun et al., 1994;Leakey et al., 1995;Rafferty et al., 1995;Ward et al., 1995;MacLatchy and Bossert, 1996;Harrison, 2002Harrison, , 2010Smith et al., 2003;Ishida et al., 2004;Nakatsukasa et al., 2007;Deane, 2009;Harrison and Andrews, 2009;Peppe et al., 2009;Pickford et al., 2009;Michel et al., 2014). ...
... Rather than being younger than all of the Tinderet sites, the Kisingiri stratigraphic sequence is substantially longer than previously thought, with the oldest fossil strata contemporaneous with those from Koru and Legetet Hill at approximately 20e19 Ma (McCollum et al., 2013). The youngest Miocene fauna from Rusinga is not associated with radiometric dates, but magnetostratigraphic analysis suggests an age near 17 Ma or younger (Peppe et al., 2009). Hence, differences between species of Proconsul, and in particular between assemblages from Tinderet and Kisingiri, cannot be simply explained by chronology. ...
... The timeline charts major splitting events (large bullets) between ape lineages, which are taken from Raaum et al. (2005) and references therein. Fossil apes featured in the text are also plotted with lines approximating their temporal ranges in the fossil record (Begun 2007;Benefit and McCrossin 1995;Heizmann and Begun 2001;MacLatchy et al. 2006;Peppe et al. 2009). Shaded regions correspond to the phylogeographic groups discussed in the text, and their placement across major splitting events represents both current debates over the relationships of their constituent taxa as well as, in some cases, the likelihood that these convenient geotemporal groups comprise members of multiple evolutionary lineages. ...
... The best known of these STEMs is the genus Proconsul ( Fig. 2), known primarily from sites in Uganda and Kenya dated between 20 and 17 Ma (Bishop et al. 1969;MacLatchy et al. 2006;Peppe et al. 2009). Kenyan sites on Rusinga and Mfangano Islands have provided much of the material attributed to Proconsul, and discoveries there of more than a dozen partial skeletons and hundreds of isolated fossils have documented nearly every aspect of its skeletal anatomy. ...
The evolutionary history of humans comprises an important but small branch on the larger tree of ape evolution. Today’s hominoids—gibbons,
orangutans, gorillas, chimpanzees, and humans—are a meager representation of the ape diversity that characterized the Old
World from 23–5million years ago. In this paper, I briefly review this evolutionary history focusing on features important
for understanding modern ape and human origins. As the full complexity of ape evolution is beyond this review, I characterize
major geographic, temporal, and phylogenetic groups using a few flagship taxa. Improving our knowledge of hominoid evolution
both complicates and clarifies studies of human origins. On one hand, features thought to be unique to the human lineage find
parallels in some fossil ape species, reducing their usefulness for identifying fossil humans. On the other hand, the Miocene
record of fossil apes provides an important source for generating hypotheses about the ancestral human condition; this is
particularly true given the dearth of fossils representing our closest living relatives: chimpanzees and gorillas.
KeywordsHominoid evolution-Apes-Miocene-Human origins
... minimum age ranges for MN6 in central Europe; van der Meulen et al., 2011); S. enjiessi, 17.2 Ma (average of the ages of Kalodirr [16.7 Ma] and Moruorot [17.7 Ma];Rose et al., 1992); D. macinnesi, 17.8 Ma (age of locality R3a from the Hiwegi Formation of Rusinga Island;Drake et al., 1988;Senut, 1989;Peppe et al., 2009); Ek. heseloni, 17.8 Ma (age of the Kulu Formation of Rusinga Island; ...
Pliopithecoids are a diverse group of Miocene catarrhine primates from Eurasia. Their positional behavior is still unknown, and many species are known exclusively from dentognathic remains. Here, we describe a proximal radius (IPS66267) from the late Miocene of Castell de Barberà (Vallès-Penedès Basin, NE Iberian Peninsula) that represents the first postcranial specimen of the pliopithecoid Barberapithecus huerzeleri. A body mass estimate based on the radius is compared with dental estimates, and its morphology is compared with that of extant and fossil anthropoids by qualitative means as well as by landmark-based three-dimensional geometric morphometrics. The estimated body mass of ∼5 kg for IPS66267 closely matches the dental estimates for the (female) holotype, thereby discounting an alternative attribution to the large-bodied hominoid recorded at Castell de Barberà. In multiple features (oval and moderately tilted head with a pronounced lateral lip and a restricted articular area for the capitulum; proximodistally expanded proximal radioulnar joint; and short, robust, and anteroposteriorly compressed neck), the specimen differs from hominoids and resembles instead extant nonateline monkeys and stem catarrhines. The results of the morphometric analysis further indicate that the Barberapithecus proximal radius shows closer similarities with nonsuspensory arboreal cercopithecoids and the dendropithecid Simiolus. From a locomotor viewpoint, the radius of Barberapithecus lacks most of the features functionally related to climbing and/or suspensory behaviors and displays instead a proximal radioulnar joint that would have been particularly stable under pronation. On the other hand, the Barberapithecus radius differs from other stem catarrhines in the less anteroposteriorly compressed and less tilted radial head with a deeper capitular fovea, suggesting a somewhat enhanced mobility at the elbow joint. We conclude that pronograde arboreal quadrupedalism was the main component of the locomotor repertoire of Barberapithecus but that, similar to other crouzeliids, it might have displayed better climbing abilities than pliopithecids.
... The Miocene deposits on Rusinga Island and the abundant fossil assemblages, which include multiple early hominoid taxa, attracted the vast majority of paleoanthropological attention from the 1930s through the early 2000s (e.g., Le Gros Clark and Leakey, 1951;Whitworth, 1953;Savage, 1965;Pickford, 1984;Walker and Teaford, 1988;Teaford et al., 1988;Walker et al., 1993;Peppe et al., 2009). However, Rusinga Island also preserves an important Late Pleistocene record. ...
In 2010, a hominin right humerus fragment (KNM-RU 58330) was surface collected in a small gully at Nyamita North in the Late Pleistocene Wasiriya Beds of Rusinga Island, Kenya. A combination of stratigraphic and geochronological evidence suggests the specimen is likely between ∼49 and 36 ka in age. The associated fauna is diverse and dominated by semiarid grassland taxa. The small sample of associated Middle Stone Age artifacts includes Levallois flakes, cores, and retouched points. The 139 mm humeral fragment preserves the shaft from distal to the lesser tubercle to 14 mm below the distal end of the weakly projecting deltoid tuberosity. Key morphological features include a narrow and weakly marked pectoralis major insertion and a distinctive medial bend in the diaphysis at the deltoid insertion. This bend is unusual among recent human humeri but occurs in a few Late Pleistocene humeri. The dimensions of the distal end of the fragment predict a length of 317.9 ± 16.4 mm based on recent samples of African ancestry. A novel method of predicting humeral length from the distance between the middle of the pectoralis major and the bottom of the deltoid insertion predicts a length of 317.3 mm ± 17.6 mm. Cross-sectional geometry at the midshaft shows a relatively high percentage of cortical bone and a moderate degree of flattening of the shaft. The Nyamita humerus is anatomically modern in its morphology and adds to the small sample of hominins from the Late Pleistocene associated with Middle Stone Age artifacts known from East Africa. It may sample a population closely related to the people of the out-of-Africa migration.
... For comparison we used Sivanasua viverroides (Schlosser, 1916) from Rothenstein 1/13 described by Fejfar et al. (1997), including a cast of specimen 0030 stored at the MNCN, and the P4-P3 from Chêne de Nav ere, France, MN4, cast stored in the MNHN, figured by Ginsburg & Morales (1999), Sivanasua antiqua (Crusafont-Pair o, 1959) (Peppe et al. 2009) housed at the NHMUK; Herpestides antiquus (Blainville, 1842), from Saint-G erand-le-Puy, France, MN2a, housed at the MNHN and described by Viret (1929) and Beaumont (1967); Protictitherium crassum (Dep eret, 1892) from the Vallesian of Spain from the sites of Los Valles de Fuentidueña, MN9 and Cerro de los Batallones, MN10 (Fraile 2017). The taxa Lophocyon carpathicus Fejfar et al., 1987from Kosice-Bankov, Slovakia, MN5, Legetetia nandii Schmidt-Kittler (1987 from Legetet, Kenya, lower Miocene 19 Ma (Werdelin 2010), and Protictis spp. ...
Lophocyonids are one of the more enigmatic groups of Carnivora in the European Miocene fossil record. Lophocyonids are clearly distinguished from other Feliformia by their peculiar lophodont dental morphology. For this reason, the systematic relationships of the family have been controversial. Here we describe and interpret dental remains from the early Miocene of Sabuncubeli, Turkey, which we attribute to a new genus and species: Izmirictis cani. The phylogenetic analysis allows us to include Izmirictis within a monophyletic group, Lophocyonidae new rank, characterized by the molarization of the anterior premolars (P3 and p4), the lophodont adaptation of the molar dentition and the complex morphology of the incisors. The phylogenetic analysis shows a close relationship between Izmirictis and primitive hyaenids (represented herein by Protictitherium). The divergence between Lophocyonidae and Hyaenidae is estimated by biochronological data to have occurred during the early Miocene (MN2). Dental microwear analysis, although limited by poor dental enamel preservation, indicates that the pronounced lophodonty in Izmirictis cani could be connected to a herbivorous feeding habit.
http://zoobank.org/urn:lsid:zoobank.org:pub:8D75382D-6C22-41C1-A47F-BA3EAF766881
... Rebekkachromis (see Figs. 4e,6e,f,7d,e). A further fossiliferous deposit from Kenya is the lower Miocene Kulu Formation (17-15 MYA, see Peppe et al., 2009Peppe et al., , 2011. Four species have been reported (Van Couvering, 1982): Kalyptochromis hamulodentis Van Couvering, 1982 was described based on a single specimen; it has two supraneurals, 30 vertebrae and cycloid scales. ...
The fossil record of cichlids is sparse, and every new discovery can provide new insights into the evolutionary history of this speciose freshwater fish family. In this article, we describe †Rebekkachromis gen. nov. from the middle-to-late Miocene (c. 11 MYA) of the Central Kenya Rift within the East African Rift system. †Rebekkachromis is represented by two species that differ from all other fossil and extant African cichlids, except Etia, in possessing the unique character combination of two supraneural bones and a set of robust tricuspid oral teeth in the outer row of the dentition. Furthermore, †Rebekkachromis exhibits the only proposed morphological synapomorphy of the Haplotilapiini, namely the presence of tricuspid teeth in the inner row of the oral dentition. We show that †Rebekkachromis constitutes the oldest reliably identified fossil record of a haplotilapiine. The evolution of cichlid fishes possessing tricuspid teeth in the rivers and lakes of the Central Kenya Rift during the middle-to-late Miocene could have been facilitated by volcanic activity, as repeated ash falls may well have fostered the growth of algae and in particular diatoms. These fishes could thus have had a major advantage, because they could exploit the newly available, rich food resources.
... Creighton Gabel (1969) excavated a series of rockshelters along the north of the Winam Gulf, one of which, Randhore, sampled undated strata attributed to the MSA Basell (2007) conducted work at Rambogo rockshelter 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 Rusinga Island (Peppe et al. 2009;Tryon et al. 2010). ...
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 different times in different places across Africa. Accounting for this pattern 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 ~650km 2 to >2500km 2. Lithic artifacts show MSA technology is present younger than 36 ka in western Kenya, 25-35 kyr younger than the first 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.
... Our recent work on Rusinga's Pleistocene localities (primarily Wakondo, Nyamita, and Nyamsingula; Fig. 1B) is part of a broader research program reconstructing paleoenvironments, faunal communities, and hominin landscape use around the eastern shores of Lake Victoria (Tryon et al., 2010(Tryon et al., , 2012Faith et al., 2011Faith et al., , 2015Faith et al., , 2016Blegen et al., 2015;Beverly et al., 2015a, b;Garrett et al., 2015). The research in Wakondo's Pleistocene deposits was initiated through limited surface collections and rescue excavations conducted by Miocene paleontological researchers (McNulty et al., 2007;Peppe et al., 2009), who discovered three partial bovid skeletons eroding from sediments at the sub-locality at Wakondo referred to as "Bovid Hill" (Fig. 1D). From these 2007 collections we identified the first recorded cut marks from the Wasiriya Beds (Tryon et al., 2010), and the Wakondo locality then became the focus of systematic surface collections in 2009 and 2010, which increased the sample of MSA lithic artifacts and confirmed high densities of Rusingoryx fossils at the Bovid Hill sub-locality. ...
The foraging behaviors of Middle Stone Age (MSA) early modern humans have largely been based on evidence from well-stratified cave sites in South Africa. Whereas these sites have provided an abundance of data for behavioral reconstruction that are unmatched elsewhere in Africa, they are unlikely to preserve evidence of the diversity of foraging strategies employed by MSA hunters who lived in a variety of ecological and landscape settings across the African continent. Here we describe the results of recent excavations at the open-air site of Bovid Hill at Wakondo, Rusinga Island, Kenya, which yielded 24 in situ MSA artifacts within an assemblage of bones comprised exclusively of the extinct alcelaphin bovid Rusingoryx atopocranion. The excavated faunal assemblage is characterized by a prime-age-dominated mortality profile and includes cut-marked specimens and an associated MSA Levallois blade-based artifact industry recovered from a channel deposit dated to 68 ± 5 ka by optically stimulated lumines-cence. Taphonomic, geologic, and faunal evidence points to mass exploitation of Rusingoryx by humans at Bovid Hill, which likely represents an initial processing site that was altered post-depositionally by fluvial processes. This site highlights the importance of rivers and streams for mass procurement in an open and seasonal landscape, and provides important new insights into MSA behavioral variability with respect to environmental conditions, site function, and tactical foraging strategies in eastern Africa. Bovid Hill thus joins a growing number of MSA and Middle Paleolithic localities that are suggestive of tactical hunting behaviors and mass capture of gregarious ungulate prey.
... The Lake Victoria basin formed in the depression between the eastern and western branches of the EARS, probably within the last few million years (reviewed in Danley et al., 2012). The northeastern part of the Lake Victoria Basin (Fig. 1c) is unlikely to have been volcanically active since the cessation of rifting and volcanism associated with the failed Nyanza Rift in the Early Miocene (e.g., Peppe et al., 2009;Van Couvering, 1972). During the Pleistocene, the eLVB formed a repository for sediments, including volcaniclastic deposits from eruptions originating from sources outside of the basin. ...
The later Middle through early Late Pleistocene (~100–400 ka) of East Africa is an important time and place for the evolution of our species. This period records the first appearance of Homo sapiens and spans significant technological changes including the decline of large handheld stone tools characteristic of the Acheulean, the development of stone tool technologies collectively known as the Middle Stone Age (MSA). These include diverse Levallois prepared core techniques and the manufacture and use of pointed weapons. It is in association with MSA technologies in sub-Saharan Africa that most of the behaviors characteristic of modern humans first appear. This doctoral dissertation provides new chronological and archaeological data relevant to hominin behavior associated with MSA technology in the Middle and Late Pleistocene of East Africa. Improved chronological resolution is achieved through tephrostratigraphy, the correlation of volcanic ashes, combined with chronometric dating in two regions: the Kapthurin Formation in the Rift Valley, Baringo, Kenya and the eastern Lake Victoria Basin of western Kenya. New data on hominin behavior is provided by archaeological excavations of two sites: 1) The 196-226 ka Sibilo School Road Site in the Kapthurin Formation. 2) The 33–49 ka site of Nyamita Main in the eastern Lake Victoria Basin. The archaeology of the Kapthurin Formation and the eastern Lake Victoria Basin are connected thematically by the presence of MSA technology. These basins are also connected stratigraphically and chronologically, as this study shows, by tephra correlations between them. Results of this work demonstrate: 1) Levallois prepared core techniques, important aspects of MSA technology, are shown to be >380 ka in the Kapthurin Formation, ~100 kyr older than previously estimated in East Africa. 2) Long distance transport (>166 km) of high quality obsidian for stone tool manufacture was a feature of hominin behavior associated with Middle Pleistocene MSA technology ~200 ka ago. 3) MSA technology persisted in East Africa later than 49 ka and perhaps later than 33 ka, after Later Stone Age technologies, often considered categorically superior, are documented in the region. By demonstrating both the early and late presence of various aspects of MSA technology and associated hominin behavior this work shows that tephrostratigraphy and the excavation of new archaeological material in East Africa are productive means of producing new and important data on the MSA and the evolution of human behavior.
... Rusinga Island fossils preserve an important snapshot of the East African early Miocene. Rusinga is most famous for its rich mammalian fossil record and particularly for its many primate species, including two species of Proconsul (Le Gros Clark, 1950;Andrews, 1974;Andrews and Simons, 1977;Walker et al., 1993;Lehmann, 2009;Peppe et al., 2009;and chapters in Werdelin and Sanders, 2010). In addition to the mammalian taxa, Rusinga Island also includes a diversity of reptile fossils that are important for understanding African paleoecology and biogeography. ...
... The Lake Victoria basin formed in the depression between the eastern and western branches of the EARS, probably within the last few million years (reviewed in Danley et al., 2012). The northeastern part of the Lake Victoria basin (Fig. 1c) is unlikely to have been volcanically active since the cessation of rifting and volcanism associated with the failed Nyanza Rift in the Early Miocene (e.g., Van Couvering, 1972;Peppe et al., 2009). During the Pleistocene, the eLVB formed a repository for sediments, including volcaniclastic deposits from eruptions originating from sources outside of the basin. ...
... The Lake Victoria basin formed in the depression between the eastern and western branches of the EARS, probably within the last few million years (reviewed in Danley et al., 2012). The northeastern part of the Lake Victoria basin (Fig. 1c) is unlikely to have been volcanically active since the cessation of rifting and volcanism associated with the failed Nyanza Rift in the Early Miocene (e.g., Van Couvering, 1972;Peppe et al., 2009). During the Pleistocene, the eLVB formed a repository for sediments, including volcaniclastic deposits from eruptions originating from sources outside of the basin. ...
The tephrostratigraphic framework for Pliocene and Early Pleistocene paleoanthropological sites in East Africa has been well established through nearly 50 years of research, but a similarly comprehensive framework is lacking for the Middle and particularly the Late Pleistocene. We provide the first detailed regional record of Late Pleistocene tephra deposits associated with artifacts or fossils from the Lake Victoria basin of western Kenya. Correlations of Late Pleistocene distal tephra deposits from the Wasiriya beds on Rusinga Island, the Waware beds on Mfangano Island and deposits near Karungu, mainland Kenya, are based on field stratigraphy coupled with 916 electron microprobe analyses of eleven major and minor element oxides from 50 samples. At least eight distinct distal tephra deposits are distinguished, four of which are found at multiple localities spanning >60 km over an approximately north to south transect. New optically stimulated luminescence dates help to constrain the Late Pleistocene depositional ages of these deposits. Our correlation and characterization of volcaniclastic deposits expand and refine the current stratigraphy of the eastern Lake Victoria basin. This provides the basis for relating fossil- and artifact-bearing sediments and a framework for ongoing geological, archaeological and paleontological studies of Late Pleistocene East Africa, a crucial time period for human evolution and dispersal within and out of Africa.
... Rusinga and Mfangano are remnants of the eruption and deposition of lavas and sediments of the Kisingiri volcano, inactive since the middle Miocene. Both islands share a similar bedrock lithology of Miocene lavas and volcaniclastics, unconformably overlain by poorly consolidated Pleistocene sediments (e.g., Peppe et al., 2009;Tryon et al., 2010). The Pleistocene Wasiriya Beds of Rusinga and the Waware Beds of Mfangano are primarily made up of tuffaceous alluvial and fluvial sediments and weakly developed paleosols suggestive of a relatively unstable landscape dominated by episodic depositional events (Figs. ...
... Rusinga Island fossils preserve an important snapshot of the East African early Miocene. Rusinga is most famous for its rich mammalian fossil record and particularly for its many primate species, including two species of Proconsul (Le Gros Clark, 1950;Andrews, 1974;Andrews and Simons, 1977;Walker et al., 1993;Lehmann, 2009;Peppe et al., 2009;and chapters in Werdelin and Sanders, 2010). In addition to the mammalian taxa, Rusinga Island also includes a diversity of reptile fossils that are important for understanding African paleoecology and biogeography. ...
‘Crocodylus’ pigotti is a relatively small crocodylid from the Miocene of Rusinga Island in Lake Victoria, Kenya. Known only from one relatively complete skull and limited, fragmentary, referred material, ‘Crocodylus’ pigotti lacks a detailed description. Moreover, recent analyses have shown ‘Crocodylus’ pigotti to be an osteolaemine crocodylid, more closely related to the extant dwarf crocodiles (Osteolaemus) than to true Crocodylus. Here, we describe numerous new remains of ‘Crocodylus’ pigotti recovered from localities within the Fossil Bed Member of the Hiwegi Formation at Kaswanga Point, Rusinga Island. We recovered parts of several individuals and report on previously unknown parts of the anatomy, provide an updated phylogenetic analysis, and reallocate the species ‘Crocodylus’ pigotti to a new genus, Brochuchus.SUPPLEMENTAL DATA—Supplemental materials are available for this article for free at www.tandfonline.com/UJVP
... The area today receives w800e1000 mm of rainfall per year, and prior to substantial recent human impacts, the islands were likely covered by variably dense woodlands (Andrews, 1973). Both islands are the remnants of deposition related to the eruptive history of the Kisingiri volcano during the Miocene, and share a similar bedrock lithology of Miocene lavas and volcaniclastic rocks (e.g., Van Couvering, 1972;Pickford, 1986;Drake et al., 1988;Peppe et al., 2009). On Rusinga and Mfangano Islands, poorly consolidated Pleistocene sediments accumulated at the base of each island's central uplands and unconformably overlie Miocene deposits (Fig.1). ...
Open-air archaeological sites record only a small fraction of the behavioral traces of mobile forager populations. Whereas caves and rockshelters were often occupied at least in part for protection from the elements, the reasons why human foragers occupied other places on the landscape (however briefly) are varied and not always readily recoverable. We develop a framework for interpreting human use of the landscape and modeling occupation of open-air sites using the archaeological and paleoenvironmental record of Middle Stone Age (MSA) sites from Rusinga and Mfangano Islands, located near the eastern margin of Lake Victoria. Paleoenvironmental reconstructions using fossil faunas suggest an arid grassland setting unlike the present. Paleoecological modeling of the habitats of extant and extinct bovids, combined with GIS-based reconstructions of lake level change, indicate that human occupation of these sites coincided with substantial declines in the level of Lake Victoria. During this time, both Rusinga and Mfangano would have been connected to the mainland and represented local topographic highs within an extensive grassland. Geological, ecological, and ethnobotanical observations suggest that these topographic high points would likely have been important sources of stone raw material, fresh water, and a variety of plant resources for food, fuel, and other purposes. In contrast, the grassy lowland plains were probably exploited primarily as a source of large game, which included numerous species of large gregarious grazers, several of which may have followed now extinct migration routes.
... The description of Williams (1954) is accompanied with very little locality data for the holotype of Pelusios rusingae, except that it originated from Rusinga Island, Kenya. Although most fossiliferous levels on this island are dated to be early Miocene, ashes are notably lacking from the upper portion of the section (Peppe et al., 2009). Until better geographic and geologic locality data has become available, we therefore conservatively date this node at the top of the Miocene at 5.3 Ma (5.332 Ma; Gradstein et al., 2004; Ogg et al., 2008). ...
Turtles have served as a model system for molecular divergence dating studies using fossil calibrations. However, because some parts of the fossil record of turtles are very well known, divergence age estimates from molecular phylogenies often do not differ greatly from those observed directly from the fossil record alone. Also, the phylogenetic position and age of turtle fossil calibrations used in previous studies have not been adequately justified. We provide the first explicitly justified minimum and soft maximum age constraints on 22 clades of turtles following best practice protocols. Using these data we undertook a Bayesian relaxed molecular clock analysis establishing a timescale for the evolution of crown Testudines that we exploit in attempting to address evolutionary questions that cannot be resolved with fossils alone. Some of these questions, such as whether the turtle crown originated in the Triassic or Jurassic, cannot be resolved by our analysis. However, our results generate novel age-of-origination estimates for clades within crown Testudines. Finally, we compare our fossil calibrations and posterior age estimates to those from other studies, revealing substantial differences in results and interpretation.
... apparent differences between older and younger sites. 38,80 In particular, the faunas from the older sites (Songhor, Koru, Legetet) are markedly different from those of the Turkana localities and share very few taxa. The younger fauna from Rusinga has some similarities to that from Kalo- dirr and Moruorot, although it is still more similar to the fauna from Song- hor than to that of the Turkana sites. ...
Faunal evolution over the last 65 million years of earth's history was dominated by mammalian radiations, but much of this era is poorly represented in Africa. Mammals first appeared early in the Mesozoic, living alongside dinosaurs for millions of years, but it was not until the extinction of dinosaurs 65 myr ago that the first major explosion of mammalian taxa took place. The Cenozoic (65 Ma to Recent) witnessed repeated and dynamic events involving the radiation, evolution, and extinction of mammalian faunas. Two of these events, each marking the extinction of one diverse fauna and subsequent establishment of another equally diverse fauna, both involving advanced catarrhine primates, are recorded in sites in the Turkana Basin, despite the poorly represented record of Cenozoic faunas elsewhere in sub-Saharan Africa. The first of these events occurred at the Oligocene-Miocene transition and the other at the Miocene-Pliocene transition.
... However, these younger beds have been examined primarily as an adjunct to work in the rich Miocene deposits (Le Gros Leakey, 1950, 1951;Napier and Davis, 1959;Andrews, 1981;Pickford, 1986;Walker and Teaford, 1988;Collinson et al., 2009). Renewed fieldwork on Rusinga initiated in 2006(McNulty et al., 2007Peppe et al., 2009) continued this trend, but in January 2009 was expanded to include the first formal archaeological surveys and excavations. While Pleistocene fossils have been reported across much of the island (Van Couvering, 1972;Pickford, 1986), our research thus far has focused on exposures on the southern portion of Rusinga, where both fossils and stone artifacts have been recovered, particularly from the Wakondo, Nyamita, and western Nyamsingula (Kakrigu) localities (Fig. 2). ...
Western Kenya is well known for abundant early Miocene hominoid fossils. However, the Wasiriya Beds of Rusinga Island, Kenya, preserve a Pleistocene sedimentary archive with radiocarbon age estimates of >33-45 ka that contains Middle Stone Age artifacts and abundant, well-preserved fossil fauna: a co-occurrence rare in eastern Africa, particularly in the region bounding Lake Victoria. Artifacts and fossils are associated with distal volcanic ash deposits that occur at multiple localities in the Wasiriya Beds, correlated on the basis of geochemical composition as determined by electron probe microanalysis. Sediment lithology and the fossil ungulates suggest a local fluvial system and associated riparian wooded habitat within a predominantly arid grassland setting that differs substantially from the modern environment, where local climate is strongly affected by moisture availability from Lake Victoria. In particular, the presence of oryx (Oryx gazella) and Grevy's zebra (Equus grevyi) suggest a pre-Last Glacial Maximum expansion of arid grasslands, an environmental reconstruction further supported by the presence of several extinct specialized grazers (Pelorovis antiquus, Megalotragus sp., and a small alcelaphine) that are unknown from Holocene deposits in eastern Africa. The combination of artifacts, a rich fossil fauna, and volcaniclastic sediments makes the Wasiriya Beds a key site for examining the Lake Victoria basin, a biogeographically important area for understanding the diversification and dispersal of Homo sapiens from Africa, whose pre-Last Glacial Maximum history remains poorly understood.
Understanding the biogeography and evolution of Miocene catarrhines relies on accurate specimen provenience. It has long been speculated that some catarrhine specimens among the early collections from Miocene sites in Kenya have incorrect provenience data. The provenience of one of these, the holotype of Equatorius africanus (NHM M16649), was previously revised based on x-ray fluorescence spectroscopy. Here we use nondestructive portable x-ray fluorescence spectroscopy to test the provenience of additional catarrhine specimens that, based hat two specimens purportedly from the Early Miocene site of Rusinga (KNM-RU 1681 and KNM-RU 1999) are instead from Maboko, three specimens purportedly from the Middle Miocene site of Fort Ternan (KNM-FT 8, KNM-FT 41, and KNM-FT 3318) are instead from Songhor, and one specimen accessioned as being from Songhor (KNM-SO 5352) is from that site. Elemental data reveal that two of the specimens (KNM-FT 3318 and KNM-RU 1681) are likely to have been collected at sites other than their museum-accessioned provenience, while two others (KNM-RU 1999, and KNM-FT 41) were confirmed to have correct provenience. Results for both KNM-FT 8 and KNM-SO 5352, while somewhat equivocal, are best interpreted as supporting their accessioned provenience. Our results have implications for the distribution of certain catarrhine species during the Miocene in Kenya. Confirmation of the provenience of the specimens also facilitates taxonomic attribution, and resulted in additions to the morphological characterizations of some species. The protocol presented here has potential for wider application to assessing questions of provenience for fossils from other locations and periods.
The lower Miocene of Rusinga Island (Lake Victoria, Kenya) is best known for its vertebrate fossil assemblage but the multiple stratigraphic intervals with well-preserved fossil leaves have received much less attention. The Hiwegi Formation has three fossil leaf-rich intervals, which span the entire formation from oldest to youngest: Kiahera Hill, R5, and R3. Here, we describe new fossil collections from Kiahera Hill and R3 and compared these floras to previous work from R5, as well as modern African floras. The oldest flora at Kiahera Hill was most similar to modern tropical rainforests or tropical seasonal forests and reconstructed as a warm and wet, closed forest. This was followed by a relatively dry and open environment at R5, which was reconstructed as a woodland to open tropical seasonal forest. The youngest flora at R3 was most similar to modern tropical seasonal forests and was reconstructed as a warm and wet spatially heterogenous forest. Floral composition of all three floras differed, but the Kiahera Hill and R3 floras were more similar to each other than either flora was to the R5 flora. The Kiahera Hill flora had few monocots or herbaceous taxa, was dominated by large leaves, and had higher species richness and greater evenness than the R3 flora. Our work, coupled with previous studies, suggests that the R3 landscape consisted of both closed forest areas and open areas with seasonal ponding. The absence of morphotypes from the R5 flora that were present in the Kiahera Hill and R3 floras provides evidence for local extirpation during the R5 time interval. Thus, this work indicates that the Hiwegi Formation on Rusinga Island samples multiple environments ranging from more closed tropical forests to more open woodlands in the Early Miocene and provides important context for the evolution and habitat preference of early apes.
Aim: The roles of geodynamics, climatic variability and landscape evolution in shaping aquatic biodiversity patterns on the African continent remains poorly understood. We studied the geographical origin and phylogenetic relationships of an Afrotropical freshwater snail genus to examine the role of drainage evolution on diversification and range evolution. The relevance of fish provinces in bio-regionalization of invertebrates was explored, as well as the evolution of habitat specificity.
Location: Africa including Madagascar.
Taxon: Lanistes (Gastropoda: Ampullariidae).
Methods: Based on a sampling covering the entire geographical range, we reconstructed a fossil-calibrated multi-locus molecular phylogeny using maximum likelihood and Bayesian inference approaches. After applying species delimitation methods. Additionally, we estimated ancestral areas and habitats and examined rates of diversification through time using lineage-through-time plots.
Results: RAxML and MrBayes analyses resulted in highly congruent topologies and a strongly supported phylogeny. Our BEAST analysis indicates that Lanistes probably originated in the Eocene about 50 Ma and the most recent common ancestor (MRCA) of all 23 Lanistes OTUs in our study may have inhabited an area including the Central African and adjacent Lower Guinean biogeographical regions. A steeper increase in species accumulation from the middle Miocene (c. 15 to 10 Ma), followed by a decrease towards the present was found. Sympatry and jump dispersal are the common cladogenetic events and only a single anagenetic dispersal event was detected. The biogeographical analyses further suggest that Madagascar was colonized from East Africa and that the Zambezi River was colonized at least twice independently. Seven species are confined to rivers and, three live exclusively in lakes. The estimation of ancestral habitats suggested that the MRCA of all Lanistes probably evolved in a riverine habitat.
Main conclusions: The diversification of Lanistes started in the Eocene and occurred at a constant pace apart from a possible climate-related increase in the Miocene. This study highlights the significance of temporal geographical isolation of river systems and subsequent reconnection in clade diversification and of jump dispersal in range evolution. More comparative analyses across various taxa are needed to obtain a better understanding of African freshwater biodiversity.
This chapter examines hypothesized dispersal events between Africa and Eurasia involving non-cercopithecoid catarrhines, particularly hominoid apes, and reviews the tectonic and climatic events that may have had a role in shaping them. All available evidence points to hominoid origins in Africa by the latest Oligocene, and the earliest evidence for apes outside of Africa in Eurasia occurs at ~17–16 Ma following a tectonic event at ~19 Ma that resulted in a landbridge between these continents through the Arabian Peninsula. Following their initial dispersal into Eurasia, the estimated number of subsequent dispersals between Africa, Asia, and Europe is dependent on the hypothesized phylogeny of these fossil apes. Here, we examined several recent phylogenetic hypotheses that suggest anywhere between one and four hominoid dispersal events between Africa and Eurasia, and a minimum of zero to two ape dispersals between Europe and Asia. The arrival of pongines, and possibly other apes, in Asia most likely occurred during or right after the Middle Miocene Climatic Optimum (~17–15 Ma), and the extinction of many Asian taxa was probably driven in part by Himalayan tectonic uplift in the Late Miocene (~9–8 Ma), which changed climatic patterns and resulted in the loss of preferred hominoid rainforest habitat. Similarly, climate change in Europe resulting in the loss of preferred habitat almost certainly played a role in European ape extinction.
The divergence of crown catarrhines—i.e., the split of cercopithecoids (Old World monkeys) from hominoids (apes and humans)—is a poorly understood phase in our shared evolutionary history with other primates. The two groups differ in the anatomy of the hip joint, a pattern that has been linked to their locomotor strategies: relatively restricted motion in cercopithecoids vs. more eclectic movements in hominoids. Here we take advantage of the first well-preserved proximal femur of the early Oligocene stem catarrhine Aegyptopithecus to investigate the evolution of this anatomical region using 3D morphometric and phylogenetically-informed evolutionary analyses. Our analyses reveal that cercopithecoids and hominoids have undergone divergent evolutionary transformations of the proximal femur from a similar ancestral morphology that is not seen in any living anthropoid, but is preserved in Aegyptopithecus, stem platyrrhines, and stem cercopithecoids. These results highlight the relevance of fossil evidence for illuminating key adaptive shifts in primate evolution. The proximal femur is key for understanding locomotion in primates. Here, the authors analyze the evolution of the proximal femur in catarrhines, including a new Aegyptopithecus fossil, and suggest that Old World monkeys and hominoids diverged from an ancestral state similar to Aegyptopithecus.
Substantial differences among the pelves of anthropoids have been central to interpretations of the selection pressures that shaped extant hominoids, yet the evolution of the hominoid pelvis has been poorly understood due to the scarcity of fossil material. A recently discovered partial hipbone attributed to the 10 million-year-old fossil ape Rudapithecus hungaricus from Rudabánya, Hungary, differs from the hipbones of cercopithecids and earlier apes in functionally significant ways. Comparisons were made to extant and other fossil anthropoids using combination of non-landmark-based and linear metrics. Measurements were taken on 3D polygonal models of hipbones collected using laser scans. These metrics capture functionally relevant morphology given the incomplete preservation of the Rudapithecus specimen. This fossil displays features that reflect changes in spinal musculature and torso structure found only in extant great and lesser apes among hominoids. Rudapithecus has an expanded cranial acetabular lunate surface related to orthograde positional behaviors, a shallow acetabulum and relatively short ischium like orangutans and hylobatids. It displays evidence of moderately coronally-oriented iliac blades as in all extant apes and Ateles, and flaring iliac blade shape of siamangs and great apes, associated with some level of spinal stiffness. However, this fossil lacks the long lower ilium that characterizes chimpanzees, gorillas and orangutans, associated with their reduction of the number of lumbar vertebrae. The R. hungaricus pelvis demonstrates that the extreme elongation of the lower ilium seen in extant great apes does not necessarily accompany adaptation to orthograde posture and forelimb-dominated arboreal locomotion in hominoid evolution. Lower iliac elongation appears to have occurred independently in each lineage of extant great apes, supporting the hypothesis that the last common ancestor of Pan and Homo may have been unlike extant great apes in pelvic length and lower back morphology.
Living hominoids (apes) are united by features related to their unusual locomotion, but few such traits are found in the earliest fossil forms. Which adaptations were likely present in the earliest hominoids?
New volumetric and mass flux estimates have been calculated for the Kenya Rift. Spatial and temporal histories for volcanic eruptions, lacustrine deposition, and hominin fossil sites are presented, aided by the compilation of a new digital geologic map. Distribution of volcanism over time indicates several periods of southward expansion followed by relative positional stasis. Volcanism occurs throughout the activated rift length, with no obvious abandonment as the rift system migrated. The main exception is a period of volcanic concentration around 10 Ma, when activity was constrained within 2° of the equator.
Volumes derived from seismic data indicate a total volume of c. 310,000 km3 (2.47 x 1010 kg/yr ), which is significantly more than the map-derived volumes found here or published previously. Map-based estimates are likely affected by a bias against recognizing small volume events in the older record. Such events are, however, the main driver of erupted volume over the last 5 Ma. A technique developed here to counter this bias results in convergence of the two volume estimation techniques.
Relative erupted composition over time is variable. Overall, the erupted material has a mafic to silicic ratio of 0.9:1. Basalts are distinctly more common in the Turkana region, which previously experienced Mesozoic rifting. Despite the near equal ratio of mafic to silicic products, the Kenya Rift otherwise fits the definition of a SLIP. It is proposed that the compositions would better fit the published definition if the Turkana region was not twice-rifted.
Lacustrine sedimentation post-dates initial volcanism by about 5 million years, and follows the same volcanic trends, showing south and eastward migration over time. This sedimentation delay is likely related to timing of fault displacements.
Evidence of hominin habitation is distinctly abundant in the northern and southern sections of the Kenya Rift, but there is an observed gap in the equatorial rift between 4 and 0.5 million years ago. After 0.5 Ma, sites appear to progress towards the equator. The pattern and timing of hominid site distributions suggests that the equatorial gap in habitation may be the result of active volcanic avoidance.
Despite the increasing rate of systematic research on scarabaeine dung beetles (Coleoptera: Scarabaeidae: Scarabaeinae), their fossil record has remained largely unrevised. In this paper, we review all 33 named scarabaeine fossils and describe two new species from Dominican amber (
Canthochilum alleni
sp.n.
,
Canthochilum philipsivieorum
sp.n.
). We provide a catalogue of all fossil Scarabaeinae and evaluate their assignment to this subfamily, based primarily on the original descriptions but also, where possible, by examining the type specimens. We suggest that only 21 fossil taxa can be reliably assigned to the Scarabaeinae, while the remaining 14 should be treated as doubtful Scarabaeinae. The doubtful scarabaeines include the two oldest dung beetle fossils known from the Cretaceous and we suggest excluding them from any assessments of the minimum age of scarabaeine dung beetles. The earliest reliably described scarabaeine fossil appears to be
Lobateuchus parisii
, known from Oise amber (France), which shifts the minimum age of the Scarabaeinae to the Eocene (53 Ma). We scored the best-preserved fossils, namely
Lobateuchus
and the two
Canthochilum
species described herein, into the character matrix used in a recent morphology-based study of dung beetles, and then inferred their phylogenetic relationships with Bayesian and parsimony methods. All analyses yielded consistent phylogenies where the two fossil
Canthochilum
are placed in a clade with the extant species of
Canthochilum
, and
Lobateuchus
is recovered in a clade with the extant genera
Ateuchus
and
Aphengium
. Additionally, we evaluated the distribution of dung beetle fossils in the light of current global dung beetle phylogenetic hypotheses, geological time and biogeography. The presence of only extant genera in the late Oligocene and all later records suggests that the main present-day dung beetle lineages had already been established by the late Oligocene–mid Miocene.
The opening and closing of the equatorial East African forest belt during the Quaternary is thought to have influenced the biogeographic histories of early modern humans and fauna, although precise details are scarce due to a lack of archaeological and paleontological records associated with paleoenvironmental data. With this in mind, we provide a description and paleoenvironmental reconstruction of the Late Pleistocene Middle Stone Age (MSA) artifact- and fossil-bearing sediments from Karungu, located along the shores of Lake Victoria in western Kenya. Artifacts recovered from surveys and controlled excavations are typologically MSA and include points, blades, and Levallois flakes and cores, as well as obsidian flakes similar in geochemical composition to documented sources near Lake Naivasha (250 km east). A combination of sedimentological, paleontological, and stable isotopic evidence indicates a semi-arid environment characterized by seasonal precipitation and the dominance of C4 grasslands, likely associated with a substantial reduction in Lake Victoria. The well-preserved fossil assemblage indicates that these conditions are associated with the convergence of historically allopatric ungulates from north and south of the equator, in agreement with predictions from genetic observations. Analysis of the East African MSA record reveals previously unrecognized north-south variation in assemblage composition that is consistent with episodes of population fragmentation during phases of limited dispersal potential. The grassland-associated MSA assemblages from Karungu and nearby Rusinga Island are characterized by a combination of artifact types that is more typical of northern sites. This may reflect the dispersal of behavioral repertoires-and perhaps human populations-during a paleoenvironmental phase dominated by grasslands.
Copyright © 2015 Elsevier Ltd. All rights reserved.
The lineage of apes and humans (Hominoidea) evolved and radiated across Afro-Arabia in the early Neogene during a time of global climatic changes and ongoing tectonic processes that formed the East African Rift. These changes probably created highly variable environments and introduced selective pressures influencing the diversification of early apes. However, interpreting the connection between environmental dynamics and adaptive evolution is hampered by difficulties in locating taxa within specific ecological contexts: time-averaged or reworked deposits may not faithfully represent individual palaeohabitats. Here we present multiproxy evidence from Early Miocene deposits on Rusinga Island, Kenya, which directly ties the early ape Proconsul to a widespread, dense, multistoried, closed-canopy tropical seasonal forest set in a warm and relatively wet, local climate. These results underscore the importance of forested environments in the evolution of early apes.
Environmental reconstructions of early Miocene sites are important for understanding the remarkable diversity and abundance of African mammals today. These provide essential context for the faunal interchange that occurred with the appearance of land bridges between Afro-Arabia and Eurasia. Tragulids, for example, were ecological precursors of some bovids, and an appreciation of their habitats during the early Miocene can provide insights into both their adaptive radiation and the environmental backdrop of the larger faunal turnover that followed. Here we reconstruct the diets of four tragulid species from early Miocene of Kenya, Dorcatherium parvum (n = 11), D. pigotti (n = 7), and D. chappuisi (n = 4) from Rusinga Island, and D. songhorensis (n = 13) from Songhor, using dental microwear texture analysis. Results indicate that all were likely mixed feeders, though there is variation in the sample. The Songhor species and D. chappuisi are inferred to have been variable grazers, D. pigotti is closer to browsers/generalists in microwear textures, and D. parvum is intermediate. This implies that, despite reconstructions of forested settings, especially at Songhor, at least some grass was available at both sites. It also suggests that the adaptive radiation of tragulids was more diverse in Africa in the past, and that we cannot assume an ancestral diet dominated by fruit and other browse items, as seen in chevrotains today.
A tag simulator is a device that can exchange data with an RFID reader in the same way as a real transponder. It may be of help for reader developers in the process of testing the device response and adjusting the communication algorithms. It may equally help system integrators in testing and tuning the performance of the complete RFID system that comprises the readers, the network to which they are connected and the database. The simulator described here is based on the Atmel AT91SAM7S64 micro-controller and covers the “transponder talk first” protocols of full duplex and half duplex types for the carrier frequencies of 125 KHz and 134.2 KHz. The simulator is a PC-configurable stand-alone device.
The affinities of the Hippopotamidae are at the core of the phylogeny of Cetartiodactyla (even-toed mammals: cetaceans, ruminants, camels, suoids, and hippos). Molecular phylogenies support Cetacea as sister group of the Hippopotamidae, implying a long ghost lineage between the earliest cetaceans (approximately 53 Ma) and the earliest hippopotamids (approximately 16 Ma). Morphological studies have proposed two different sister taxa for hippopotamids: suoids (notably palaeochoerids) or anthracotheriids. Evaluating these phylogenetic hypotheses requires substantiating the poorly known early history of the Hippopotamidae. Here, we undertake an original morphological phylogenetic analysis including several "suiform" families and previously unexamined early Miocene taxa to test previous conflicting hypotheses. According to our results, Morotochoerus ugandensis and Kulutherium rusingensis, until now regarded as the sole African palaeochoerid and the sole African bunodont anthracotheriid, respectively, are unambiguously included within the Hippopotamidae. They are the earliest known hippopotamids and set the family fossil record back to the early Miocene (approximately 21 Ma). The analysis reveals that hippopotamids displayed an unsuspected taxonomic and body size diversity and remained restricted to Africa during most of their history, until the latest Miocene. Our results also confirm the deep nesting of Hippopotamidae within the paraphyletic Anthracotheriidae; this finding allows us to reconstruct the sequence of dental innovations that links advanced selenodont anthracotheriids to hippopotamids, previously a source of major disagreements on hippopotamid origins. The analysis demonstrates a close relationship between Eocene choeropotamids and anthracotheriids, a relationship that potentially fills the evolutionary gap between earliest hippopotamids and cetaceans implied by molecular analyses.
Abundant remains of a small bovid have been unearthed at the basal Middle Miocene locality of Arrisdrift, Orange River Valley, Namibia. Among the more than 10,000 vertebrate fossils excavated at the site only two horn cores have been found. The bovid from Arrisdrift is one of the oldest known species with horn cores. It is here classified as Namacerus gariepensis nov. gen. nov. sp. Namacerus differs from Eotragus artenensis by its smaller size, greater hypsodonty, and other morphological characters of the dentition. From a phylogenetic point of view the Arrisdrift bovid may well belong to the stem group from which the rest of the bovid lineages were derived. The hypothesis of an African origin for the family Bovidae is strongly supported by the new data. During the Early and basal Middle Miocene Africa possessed a diversity of bovid genera and it was the scene of a major radiation within the family, as indicated by the presence of diverse cranial morphotypes which are primitive with respect to those known from Eurasia.
Size distributions, based on molar length, are presented for various groups of living and fossil primates. I examine (1) the extent to which particular taxonomic groups with distinctive morphological and behavioral attributes are characterized by distinctive size ranges or distributions and (2) the way in which size ranges and distributions are affected by the presence or absence of other primates within the same geographical area and time. Distinctive behavioral and ecological “adaptive zones” are characterized by distinctive size ranges and behavioral and morphological parallelism or convergence among living and fossil primate taxa usually results in similar size distributions.
The phylogenetic relationships and adaptations of Kenyapithecus have been of special interest since Leakey (1962, p. 696) first described the genus as possessing a number of characters exhibiting “a marked tendency in the direction of the Hominidae.” Expectations regarding the hominid affinities of Kenyapithecus influenced the reconstruction of many functionally and phylogenetically significant aspects of its anatomy in the absence of tangible fossil evidence. The type species Kenyapithecus wickeri is represented by four jaw fragments, 11 isolated teeth, and a distal humerus collected from a single site, Fort Ternan (Pickford, 1985). As of 1985, the hypodigm of the referred species Kenyapithecus africanus consisted of the type maxilla together with 46 isolated teeth, four incomplete postcranial pieces, and a poorly preserved mandible collected from the Maboko Formation and an isolated lower molar from Nyakach (Pickford, 1985). Features for which little or no fossil evidence existed, but which Kenyapithecus was said to share with hominids, include: small lower incisors relative to cheek tooth size, reduced incisor procumbency, arcuate dental arcade, short rostrum, and a humerus that is longer than the femur (Simons and Pilbeam, 1972). The supposed reduction in upper canine and lower incisor size, and facial abbreviation, in combination with thickenameled molars in Kenyapithecus was interpreted as being related to an australopithecine-like emphasis on molar grinding resulting from the consumption of hard objects (Andrews and Walker, 1976), with incisors being “relatively unimportant in food preparation” (Simons and Pilbeam, 1978, pp. 149, 152).
Volcanic cycles of doming and eruption of the Miocene Kisingiri volcano produced three sedimentary cycles recorded in the volcaniclastic strata of Rusinga and Mfangano Islands, Lake Victoria, Kenya. Each of the three cycles began with the deposition of cobble and boulder conglomerates shed from the volcanically domed Precambrian basement, followed by deposition of pyroclastic and volcaniclastic strata, representing nephelinite-carbonatite eruption of the Kisingiri volcano. Volcanogenic strata produced by the first two cycles (lower two-thirds of the Rusinga Group) are predominantly fine-grained tuffs and medium-grained volcaniclastic deposits, indicating alluvial deposition from a low-relief volcanic edifice. The third cycle is dominated by boulder debris flows, lava flows, and minor tuffaceous beds (upper part of the Rusinga Group and overlying Kisingiri Group). This last cycle records the formation of the high-relief Kisingiri stratocone, much of which is preserved in the dissected flanks of the volcano. The first two cycles are recorded in distal apron deposits but are not well preserved in the core of the volcano. Second-order sedimentary cycles, consisting of fining-upward sequences (5-10 m thick) of granular tuffaceous sandstones and conglomerates that fine into siltstones, stacked floodplain paleosols, and airfall tuff beds, dominate the strata of the first two cycles. These fining-upward sequences represent alluvial aggradation that accompanied and followed eruptive episodes that were much shorter in duration than the main cycles of doming and eruption.
A widespread hypothesis in anthropological science is that at some stage in the evolution of humans there must have been an adaptive shift from an arboreal forest existence to a terrestrial savannah one. The reasoning behind this idea seems to be that most extant apes are forest-dwelling, leading to the conclusion that this is probably the primitive (plesiomorphic) hominoid adaptation. Humans, in contrast, are adapted to savannah settings, and this represents a highly derived (apomorphic) condition among hominoids.
Previous chapters in this book have focused on the interaction between functional and phylogenetic analysis in interpreting fossil hominoid taxa or anatomical regions. The point has been made repeatedly that functional and phylogenetic analysis inform one another. Both functional analysis and phylogenetic analysis in turn are informed by paleoenvironmental data. Reconstructions of positional behavior and diet in fossil hominoids based on their functional anatomy are tested or at least reinforced by reconstructions of paleoenvironments. Paleoenvironmental reconstructions in turn are based in part on the behavioral implications (ecomorphology) of fossil taxa, including hominoids. Paleoenvironmental patterns during hominoid evolution also represent responses to environmental conditions, and as such can be treated as characters relevant to phylogenetic analysis (e.g., Andrews, 1982). In this chapter we summarize the paleoenvironmental data from most of the localities that have produced the fossils examined in this book. The functional anatomy and behavioral implications of the hominoid fossils are also summarized in the context of now-current paleoenvironmental reconstructions.
The species Proconsul africanus from the early Miocene of East Africa has long been recognized from Koru (the type locality), Songhor, Rusinga Island and Mfangano Island. Comparison of the type series from Koru with specimens from Rusinga and Mfangano indicates that this material does not belong to a single species. The material from Koru and from Songhor belongs to a single species that takes the name P. africanus and which is not represented at Rusinga or Mfangano. The Rusinga and Mfangano material exhibits levels of metric variation in the teeth and postcranial skeleton that are incompatible with its recognition as one species. The larger species at Rusinga and Mfangano takes the name Proconsul nyanzae. The smaller species does not have a name available for it and is described here as a new species, Proconsul heseloni.
This paper presents a review of the evolutionary relationships of the early catarrhine primates. The first stage of the analysis involves the reconstruction of the inferred ancestral morphotypes of the major groups of extant anthropoids. The introduction of the fossil taxa into the phylogenetic scheme represents the second and final stage of the analysis. The results of this cladistic analysis suggests that: (1) the parapithecids are a specialized group of basal anthropoids, (2) Oligopithecus savagei may represent the earliest recognizable catarrhine, (3) Propliopithecus (= Aegvptopithecus) and Pliopithecus apparently represent the successive sister taxa to the modern catarrhines, (4) Dendropithecus and Proconsul are best regarded as basal catarrhines of modern aspect, and (5) Victoriapithecus is a primitive cercopithecoid monkey which represents the siter taxon of the extant Old World monkeys.
The paper describes the results of geological mapping and palaeontological collecting in the Gumba peninsula, at the south-western end of Rusinga Island, Lake Victoria, Kenya.
A description of the rocks is given, with discussion of the conditions under which they were formed. The sediments seem to represent deposits of shallow, impersistent lakes and flood-plains in a semi-arid region which was subjected to continued volcanicity, often violently explosive. The stratigraphical sequence at Gumba closely resembles that described by R. M. Shackleton at Kiahera in western Rusinga.
Structurally the Gumba peninsula conforms to the simple plan established elsewhere in Rusinga. No evidence of violent post-Miocene compressive activity emerges, although minor fractures occur, which may represent local squeeze, dependent on major faulting of normal type.
An account is given of early Miocene vertebrates collected at Gumba, and of recent additions to the Rusinga fauna.
The paper describes the succession in the Miocene beds of localities near the Kavirondo Gulf of Lake Victoria, with particular reference to Rusinga island. Fossiliferous lacustrine beds form the lowest part of a sequence laid down during a phase of yolcanic eruption, and there is an upward passage from clays through tuffs to agglomerates. Plateau lavas were subsequently extruded over the Miocene beds. Conclusive evidence establishes that the beds are Lower Miocene, a fact of importance in relation to the dating of the rich fauna and of the volcanic activity, and in proving the date of deformation of the main Central. African peneplain to be pre-Miocene. The subsequent history is traced, and it is shown that the Miocene beds were successively subjected to dyke intrusion, strong faulting, valley erosion and (in the Lower Pleistocene) to raised beach formation. The disturbance of the Miocene beds includes reversed faulting and is interpreted as due to horizontal compression; it is suggested that it is related to a late Miocene phase of Rift Valley formation.
Early Miocene (c. 18 Ma) hominoid sites (Proconsul) from Rusinga Island, Lake Victoria occur in palaeosols and volcaniclastic strata deposited in a semi-arid, seasonal climate on the flanks of the active, low-relief edifice of the Kisingiri volcano. Palaeosols interbedded with sequences of primary pyroclastic deposits have stable carbon isotope values indicative of C3 vegetation from semi-arid environments. Isotopic values of palaeosol organic matter and pedogenic carbonate from the Rusinga Group have average δ13C of -23.8 ± 0.8 and -7.7 ± 1.1‰, respectively, considerably heavier than average C3 vegetation. These isotopic values were most likely caused by reduced photosynthetic fractionation under water-stressed conditions. Prolonged dry seasons and semi-arid precipitation levels for Rusinga Group fossil soils are also supported by the profile depth of calcareous horizons and vertic clay structures. The principal fossil-bearing units of the Rusinga Group (Kiahera and Hiwegi formations) are dominated by sandy strata which are interbedded with palaeosols and have features indicative of hydromagmatic pyroclastic deposition. Pyroclastic surge features include very low-angle cross-bedding, low amplitude and long-wavelength dune bedforms, moderate to poorly sorted layers of tuffaceous and pebbly sandstones, and common, isolated cobbles and boulder clasts of Precambrian basement rocks with associated impact sags. These features are interpreted as the deposits of pyroclastic surge dune bedforms and ballistically implaced volcanic ejecta, both produced by powerful hydromagmatic explosions from the Kisingiri volcano. According to this interpretation, the initial stages of this carbonatite-nephelinite volcano had repeated episodes of hydromagmatic eruptions which built a large (15-20 km radius), low-relief tuff ring or maar volcano. The palaeosol sequences interbedded with the primary pyroclastic deposits represent periods of volcanic quiescence lasting hundreds to thousands of years. Thus, Proconsul and associated fauna and flora inhabited a frequently disturbed landscape that experienced repeated catastrophic volcanic deposition in an overall semi-arid seasonal climate.
A lacustrine, volcanically influenced fan-delta is recognized in the Kulu Formation of the Miocene Rusinga Group in western Kenya. In the Nyamsingula Gully area on Rusinga Island in Lake Victoria, mapped lithologic units include the classic topset, foreset (mega-foreset) and bottomset structure of a Gilbert-type delta. A distinctive mega-breccia unit of colluvial origin interfingers laterally with the delta deposits and consists of pyroclastic and sedimentary rock fragments derived from underlying and laterally adjacent strata. The breccia formed from the bevelling of a fault escarpment, from which the fan-delta also originated. Away from the fan-delta axis, slumping of fault-scarp colluvium into the lake produced matrix-supported debris flows that are interbedded with lacustrine shales. In the topset beds of the delta, clasts of a distinctive melilitite-nephelinite tuff of airfall origin can be identified and traced down a palaeoslope through laterally adjacent units from the original bedded tuff, into the mega-breccia unit, and finally into debris-flow deposits in the foreset and bottomset beds. During resedimentation the tuff behaved in a brittle fashion due to quick cementation as a result of its water soluble, ultra-alkaline composition. The Kulu Formation is a distinctive facies that interfingers with the lowermost Hiwegi Formation. Both formations are part of the Rusinga Group. The bulk of the Rusinga Group, excluding the Kulu Formation, represents braidplains and alluvial aprons that flanked the Miocene volcano of Kisingiri. The braidplain and alluvial apron deposits have primary dips away from the volcano and were derived from material erupted and eroded from the volcano. The Kulu Formation, in contrast, was deposited in two small, elongate lacustrine basins separated by a small horst (approximately 5 km long and 2-3 km wide), from which the deposits were largely derived. The fan-delta systems prograded symmetrically away from the horst margins, into the adjoining lacustrine basins. The long dimensions of the horst and lake basins roughly parallel the local slope of the Kisingiri volcano, and therefore deposition in the Kulu Formation was perpendicular to deposition of the other Rusinga Group formations in the vicinity.
New total-fusion K/Ar ages indicate that all of the fossiliferous formations that make up the lower part of the Early Miocene Kisingiri sequence in W Kenya at Rusinga Island, Mfwangano Island, and Karungu were deposited during an interval of <0.5Ma at approx 17.8Ma ago. This contrasts markedly with K/Ar ages previously published from these detrital-tuffaceous formations, which suggested that they were deposited over an interval of as much as 7Ma between 23-16Ma, overlapping the age-ranges of all other E African Early Miocene sites including Koru, Songhor, Napak, Bukwa, Loperot, Muruarot and Buluk. In addition, the analytical problems revealed by the new Kisingiri results cast doubt on biotite ages which provide dating for the most important sites. Thus, the strong differences between the Kisingiri fauna and those of Koru, Sonhor and Napak, long held to be due to ecology because of the apparent overlap in ages, may actually be due to a difference in time. If this view of the geochronology is correct, it may now be possible to identify adaptive trends and evolutionary succession in the E African Early Miocene faunas. -Authors
The effects of volcanicity often mimic those of aridity and can lead to paleoenvironmental misinterpretations. The occurrence of volcanically induced barrenness, xeric conditions, and extreme geochemical alkalinity or salinity in the context of a regionally more humid climate is dubbed here ``mock aridity.'' Biotic recovery at Mount St. Helens (Washington) and Oldoinyo Lengai (Kenya) points to potential long-term effects of volcanicity on the overall ecosystem. Contraindicating sedimentary rocks and fossils from Kenya Miocene rocks and contraindicating sites in U.S. Pacific Northwest Miocene rocks both suggest interpretive problems due to mock aridity. This calls for a reevaluation of volcanogenic sites derived from supposed climax ecosystems in the light of mock aridity.
Renewed field work on the Miocene deposits of Maboko Island Benefit & McCrossin 1989 has produced a weath of new fossil material with important implications for vertebrate evolution in the early Neogene of East Africa. The deposits are of particular importance due to the abundance of primate fossils recovered from them. This report presents new geologic data concerning the stratigraphy and age of the fossiliferous deposits on Maboko.Maboko Island lies in the Winam Gulf of Lake Victoria (Figure 1). The geologic setting and history of research on the island have been reviewed by Andrewset al. 1981. Fossiliferous deposits were first reported from the island in 1930s and a Miocene age attributed to them on faunal grounds. Some controversy arose over the dating of the Maboko deposits, sparked by suggestion of an unconformity in the sequence and the presence of faunal elements of clearly Pleistocene age within the Miocene assemblage. The fluvial character of the Maboko deposits does include erosional surfaces, but there is no indication that they are of significant duration. Pleistocene fossils are found scattered about the island as a surface lag deposited during a late Pleistocene highstand of Lake Victoria.Subsequent studies on Maboko Island recovered additional fossil material, and suggested an age of 15–16 ma for the fauna, based on faunal considerations and isotopic dating of presumed correlative units exposed on the mainland (Andrewset al., 1981). This report presents the first numerical calibration of the age of the Miocene sequence on Maboko Island.
In his many papers on the functional morphology of the primate hand John Napier paid particular attention to the thumb and its kinematics. He suggested that in most New World monkeys the trapezium-1st metacarpal joint (t-lmcj) is almost cylindrical in form, allowing flexion-extension, limited abduction-adduction, and no rotation. He referred to this joint type as a modified hinge, allowing pseudo-opposition of the thumb. By contrast, the t-lmcj joint is sellar in catrrrhines, allowing all three of these movements to be combined in the movement of opposition. On the evidence of the trapezium, Napier & Davis (1959) considered Proconsul africanus to have a t-lmcj of the modified hinge type. These views were challenged by Lewis (1977), who pointed out that a sellariform t-lmcj is a primitive feature in mammals, although it is modified from this shape in some New World monkeys. Lewis (1977) also believed that the trapezium of P. africanus does in fact bear a sellar articular surface. Beard et al. (1986) observations on recently collected trapezia of P.africanus confirm Lewis' interpretation.Currently there is a considerably larger sample of fossil material than that available to Napier & Davis. A consideration of the functional morphology and kinematics of the t-lmcj in extant anthropoids, Proconsul spp. and Afropithecus turkanensis, suggests that the sellariform nature of the t-lmcj in these taxa is emphasized enough to result in abduction-adduction and rotatory movements as extensive as those typical of extant large hominoids. This contrasts with the more limited movements found at the t-lmcj of other extant anthropoids. This similarity of the t-lmcj and its kinematics in large hominoids, and the Miocene species is most likely to be a shared derived character.
We propose a formal stratigraphic nomenclature for fossil-bearing strata exposed in the Lothidok Range of northern Kenya. About 1540 m of sedimentary and volcanic rocks are defined (in ascending order) as the Kalakol basalts (and Eragaleit beds), the Lothidok Formation (Moruorot, Kalodirr, Naserte, Lokipenata and Kalatum Members), the Loperi Basalts, and the Turkwel beds. These new units provide a stratigraphic context for fossils collected from the region since 1932, including the recently described primate taxaTurkanapithecus andAfropithecus, and also a framework for isotopic ages of the strata. Fossiliferous strata occur within the Eragaleit beds of the Kalakol basalts, the Kalodirr, Naserte and Kalatum Members of the Lothidok Fm., and the Turkwel beds of presently undifferentiated strata. On the basis of potassium-argon (K/Ar) dates on associated volcanic rocks, faunas of four ages are recognized: those from the Eragaleit beds (24–27·5 Ma; Late Oligocene), those from the Kalodirr and Naserte Members (17·8–16·8 Ma; latest Early Miocene), those from the Kalatum Member (13·8–12·2 Ma; Middle Miocene) and those from the Turkwel beds (<11·9 Ma).Turkanapithecus andAfropithecus derive from the lower part of the Lothidok Fm. (latest Early Miocene). Primate fossils have also been recovered from the Eragaleit beds (Late Oligocene), and the Kalatum Member (Middle Miocene). The assemblage of fossil higher primates from the Lothidok region appears to be distinct from those of western Kenya and Napak, and is perhaps more similar to that recovered from Buluk, northeast of Lake Turkana. Age determinations are also reported for a basalt that overlies fossiliferous strata at Loperot, situated about 90 km south of the Lothidok Range; faunas at Loperot predate 15 Ma.
Seven distinct types of faunal assemblage occur in East African Miocene deposits. Time and ecology both influence the faunas but the main feature of these seven assemblages, termed Faunal sets, is that they appear largely to be time successive. Faunal changes are considered to be due mainly to in situ evolution, to intercontinental immigration/emigration events or to local ecologically induced faunal translocations. The sum effect of these influences appears to have been expressed chronologically in Western Kenya so that a biostratigraphic succession can be erected. The preliminary framework presented here contrasts periods of which our knowledge is weak with those where we have adequate information. Correlations with deposits elsewhere in Kenya and Africa are suggested with the understanding that the faunal sets used for these purposes are both ecologically as well as chronologically controlled.
Collections of vertebrate fossils from the Miocene of Kenya began in 1911, the year of Dr Felix Oswald's expedition to Kenya. It was not until 1927 that more sites were found, when, in the space of five years, four more important sites were discovered at Koru, Rusinga, Maboko and Songhor and the first collections made. During the following 40 years, excavations at these sites and several additional ones in Kenya were directed principally by L.S.B. Leakey working initially from Cambridge and later from Nairobi. Miocene work in the Nyanza Valley has been continued more recently by his students and associates, John and Judy Van Couvering, Martin Pickford and Peter Andrews, with short visits made by a number of other workers, and it is this work that is reported in the other papers in this issue.
The results of excavations at Songhor in 1972–73 and at Koru in 1978–80 are described. An account is given of the geology of the localities and the attempt is made to correlate these and other exposures in the Tinderet region of western Kenya. The oldest beds were excavated at a new site near Muhoroni called Meswa Bridge in river channel deposits of the Muhoroni Agglomerates. Succeeding formations are named the Koru Formation, Legetet Formation, and Kapurtay Agglomerates, all fossiliferous. The Kapurtay Agglomerates in the Koru area outcrop mainly in the Chamtwara Valley, and these beds are correlated with the Songhor deposits on the basis of a sequence of Augite Nephelinite Agglomerates common to both. The vertebrate fossils are found in terrestrial deposits (except at Meswa Bridge in the Muhoroni Agglomerates) and are usually associated with palaeosol development. The faunas have highly diverse small mammals but more limited large mammals (again except at Meswa Bridge), suggesting the existence of selective factors operating either in bone accumulation or during deposition. There may have been a combination of selection by carnivores and size sorting during deposition.
Fossils of an unnamed large suiform have been recovered from two Early Miocene localities in East Africa. The material is distinct from other species of the suborder, including the anthracothere Brachyodus aequatorialis, which is of similar size. The upper molars of the new form are bunodont, quadricuspidate (with a tiny paraconule), and have no buccal styles (parastyle, mesostyle, metastyle) and the enamel is thin and lightly wrinkled to smooth, which contrasts strongly with upper molars of Brachyodus which are pentacuspidate, selenodont, have pervasively wrinkled enamel and well-developed parastyle, mesostyle and metastyle. A new genus and species is erected for this suiform, which is most likely an anthracothere.