The partial skull of a lion from Natodomeri, northwest Kenya is described. The Natodomeri sites are correlated with Member I of the Kibish Formation, dated to between 195 ka and ca. 205 ka. The skull is remarkable for its very great size, equivalent to the largest cave lions (Panthera spelaea [Goldfuss, 1810]) of Pleistocene Eurasia and much larger than any previously known lion from Africa, living or fossil. We hypothesize that this individual represents a previously unknown population or subspecies of lion present in the late Middle and Late Pleistocene of eastern Africa rather than being an indication of climate-driven size increase in lions of that time. This raises questions regarding the extent of our understanding of the pattern and causes of lion evolution in the Late Pleistocene.

A small collection of carnivoran fossils from the South Turkwel hominid site is described. The fauna is composed of Megantereon ekidoit new species, Homotherium sp., Crocuta cf. dietrichi, cf. Pachycrocuta sp., Canis new species A., cf. Civettictis sp., Viverridae or Herpestidae indet., and Lutrinae indet. The record of Megantereon and Canis, as well as Pachycrocuta and Civettictis, if these genera are identified correctly, represents the earliest occurrences of their respective taxa in Africa. These specimens suggest a relatively rapid reorganization of the carnivore guild some time around 3.5 Ma, followed by a longer period of transition to a fauna more comparable in composition to the modern one.

In both jaw geometry and molar morphology, eutherian carnivores (order Carnivora) as a whole display greater diversity (plasticity in evolution from the primitive type) than marsupial carnivores (order Dasyurida). This is related to the difference in tooth replacement between the two taxa. In Carnivora, the permanent carnassial is preceded by a deciduous carnassial; the permanent tooth can erupt in its (geometrically) permanent position, and the post-carnassial molars are free to evolve for specialized functions or be reduced. In Dasyurida, there is relative molar progression, each erupting molar in turn functioning as a carnassial, and subsequently being pushed forwards in the jaw by the next erupting molar. Thus, all molars have carnassiform morphology, and none are free to develop for other functions. The greater plasticity of Carnivora has led to their adaptive zone being broader (as a group they are relatively more eurytopic than Dasyurida), which in turn has led to greater taxonomic diversity within Carnivora than Dasyurida. The resulting pattern from a macroevolutionary point of view is that, even in the absence of direct competition, Carnivora have had greater evolutionary “success” than Dasyurida.

The borophagine canids were bone-cracking scavengers in the Miocene-Pleistocene of North America. In this they parallel the Recent hyenas. This paper analyzes the borophagine adaptation in relation to that of hyaenids, using Osteoborus cyonoides as an example. The emphasis during canid evolution on the posterior molars, which is a derived condition, created a constraint on the adaptation of borophagines. This constraint meant that the borophagines used P4/4 as bone-cracking teeth, whereas hyaenids use P3/3. The latter adaptation has the advantage of separating the bone-cracking teeth from the meat-cutting portion of the dentition, thereby allowing a dual purpose dentition in hyaenids. In borophagines, no such dual purpose was possible, and it is suggested that they were closer to obligate bone-cracking scavengers than Recent hyaenids. Other than the evolution of a specialized bone-cracking tooth, the borophagines adapted to bone cracking by evolving a vaulted and strengthened skull for the dissipation of the strong forces generated during bone cracking. In this they again parallel the hyaenids. Evolution within borophagines involved an elaboration of patterns already set at the group's inception, creating an evolutionary trend which was mediated by the constraint on the bone-cracking morphology. This trend may be due to selection or sorting, or may, under certain assumptions, be stochastic. Other evolutionary trends may also be epiphenomena of constraints that lock morphological evolution.

African wild dogs (Lycaon pictus) occupy an ecological niche characterized by hypercarnivory and cursorial hunting. Previous interpretations drawn from a limited, mostly Eurasian fossil record suggest that the evolutionary shift to cursorial hunting preceded the emergence of hypercarnivory in the Lycaon lineage. Here we describe 1.9—1.0 ma fossils from two South African sites representing a putative ancestor of the wild dog. the holotype is a nearly complete maxilla from Coopers Cave, and another specimen tentatively assigned to the new taxon, from Gladysvale, is the most nearly complete mammalian skeleton ever described from the Sterkfontein Valley, Gauteng, South Africa. the canid represented by these fossils is larger and more robust than are any of the other fossil or extant sub-Saharan canids. Unlike other purported L. pictus ancestors, it has distinct accessory cusps on its premolars and anterior accessory cuspids on its lower premolars—a trait unique to Lycaon among living canids. However, another hallmark autapomorphy of L. pictus, the tetradactyl manus, is not found in the new species; the Gladysvale skeleton includes a large first metacarpal. Thus, the anatomy of this new early member of the Lycaon branch suggests that, contrary to previous hypotheses, dietary specialization appears to have preceded cursorial hunting in the evolution of the Lycaon lineage. We assign these specimens to the taxon Lycaon sekowei n. sp.

Palmqvist (2002) has criticized our attribution of the mandibular ramus of Megantereon from South Turkwel, Turkana Basin, Kenya, to a new species, M. ekidoit (Werdelin and Lewis, 2000), suggesting instead that it “unequivocally” belongs to the African species M. whitei. We believe that this criticism stems from a misunderstanding of our statements and from an erroneous view of variability within Megantereon. In this reply we shall address Palmqvist's criticisms, showing that M. ekidoit is not synonymous with M. whitei, and also that the specific attribution of various Megantereon specimens is not as clear cut as Palmqvist (2002; Martinez Navarro and Palmqvist, 1995, 1996) believes.

The evolutionary history of the lion Panthera leo began in Pliocene east Africa, as open habitats expanded towards the end of the Cenozoic. During the middle–late Pleistocene, lions spread to most parts of Eurasia, North America, and may have eventually reached as far south as Peru. Lions probably evolved group-living behaviour before they expanded out of Africa, and this trait is likely to have prevailed in subsequent populations. The first lions were presumed to have been maneless, and maneless forms seem to have persisted in Europe, and possibly the New World, until around 10 000 years ago. The maned form may have appeared c. 320 000–190 000 years ago, and may have had a selective advantage that enabled it to expand to replace the range of earlier maneless forms throughout Africa and western Eurasia by historic times: ‘latest wave hypothesis’.