CHAPTER 9 JACOVEC CAVERN CARNIVORES AND OTHER FAUNA 9.a. Taxonomy Carnivores The Order Carnivora is represented by five families- Viverridae, Herpestidae, Canidae, Felidae, and Hyaenidae. The Viverridae is represented by Genetta genetta. The Herpestidae is represented by Cynictis penicillata and a taxonomically indeterminate Herpestes. The Canidae is represented by Canis mesomelas, Vulpes chama,, and a taxonomically indeterminate canid. The Felidae is represented by Homotherium latidens, Felis caracal, Panthera leo, Panthera pardus and taxonomically indeterminate felids. The Hyaenidae is represented by Chasmaporthetes nitidula, Chasmaporthetes silberbergi, Chasmaporthetes sp., and a taxonomically indeterminate hyaenid. Measurements are provided in Table 9.1. Systematic palaeontology Family: VIVERRIDAE Subfamily: Viverrinae Genetta genetta (Linnaeus, 1758) Small-spotted or common genet BP/3/22769, left hemimandible (Plate. 9.1) 147
The specimen is well preserved. The hemimandible has the M 1, P 4, P 3 and P 1 in place. The tip of the metaconid on the M 1 bears a fresh break, while the mesial portion of the P 3 is missing. The freshness of the breaks indicates that they occurred either during excavation or perhaps during preparation. The ascending ramus, condyle and angle are also broken. The break through the ascending ramus has sediments embedded indicating this to be an old break, which most probably happened during the pre-depositional period. The hemimandible compares very well in dimensions and overall appearance to G. genetta. Family: HERPESTIDAE Subfamily: Herpestinae Cynictis penicillata (G. Cuiver, 1829) Yellow mongoose BP/3/33614, left hemimandible (Plate. 9.2) The specimen is a near complete left hemimandible, missing only the coronoid process. The P 1, P 2, P 3, and P 4 are present. The posterior border of the canine alveolus displays fresh breaks on the buccal side. The tip of the prominent process on the angle of the body is slightly broken though not a fresh break. The specimen compares well in all aspects to C. penicillata. Family: CANIDAE Subfamily: Caninae Vulpes chama (A. Smith. 1833) Cape fox 148
BP/3/22770, left hemimandible (Plate. 9.3) The specimen is in fair condition, and retains the M 1 and part of P 3. The mesial border of the P 3 has been freshly broken from the cervical region. The hemimandible displays two carnivore tooth marks on the basal border and at near right angle to the distal end of the present M 1 and perpendicular to the position of M 2. The ascending ramus, condyle and angle are missing, and there are fresh breaks along the alveoli of P 4, M 1, M 2 and M 3. The specimen is undoubtedly a V. chama in dimensions and overall appearance. Family: CANIDAE Subfamily: Caninae Canis mesomelas (Schreber, 1775) Black-backed jackal BP/3/32169, left P 2 (Plate. 9.4) An isolated, complete and well preserved left P 2. The specimen compares closely with C. mesomelas in dimensions and overall appearance. Family: FELIDAE Subfamily: Machairodontinae Homotherium latidens (Fabrini 1890) Extinct feline BP/3/22478, right I 3 (Plate. 9.5) 149
The specimen is equivalent in size to a modern adult lion upper third incisor. However, the major difference is the presence of a robust accessory cusp on the mesial side. This is a characteristic feature of Homotherium latidens. The root is freshly broken at the cervical junction and the enamel inferior to the accessory cusp on the lingual side has flaked exposing the dentine. Family: FELIDAE Subfamily: Felinae Felis caracal (Schreber, 1776) Lynx caracal BP/3/31881, left P 4 (Plate. 9.6) The specimen is in good condition, and almost complete, missing only the anterior lateral root. The specimen is too small to be leopard and compares well with caracal. Family: FELIDAE Subfamily: Pantherinae Panthera leo (Linnaeus, 1758) Lion PB/3/23269, left P 4 (Plate. 9.7) The anterior roots have broken off at the cervical junction, while the posterior root has broken approximately 9.17mm away from the closest cervical junction. The paracone and metacone are moderately worn on the lingual surface exposing the 150
dentine, while the protocone is missing. The tooth compares well in overall appearance with modern P. leo. Panthera leo (Linnaeus, 1758) Lion BP/3/23743, right M 1 (Plate. 9.8) The posterior root has broken off at the cervical junction, while the tip of the metaconid has also broken slightly. The metaconid is also chipped slightly on the lingual surface. The tooth compares closely with lion. Panthera leo (Linnaeus, 1758) Lion BP/3/23151, right P 4 (Plate. 9.9) The roots have broken at the cervical junction. The lingual surface on both the metacone and the paracone are slightly worn. This specimen is smaller than BP/3/23269, in both length and robusticity. Though they derive from different sides of the body, size differences (Table 9.1) indicate they derive from different individuals. Family: FELIDAE Subfamily: Pantherinae Panthera pardus (Linnaeus, 1758) Leopard BP/3/31708, right P 4 (Plate. 9.10) 151
The anterior root has broken at the cervical junction, while the posterior root is has broken at the tip. The anterior accessory cusp is broken at the tip, while the entire posterior accessory cusp is missing. The specimen is too small and gracile to be a lion but bigger than a cheetah, and compares well with a leopard. Family: HYAENIDAE Chasmaporthetes silberbergi (Broom 1948) Extinct long-legged hunting hyaena BP/3/22740, left P 4 and BP/3/22745, left P 3 (Plate. 9.11) The left P 4 is relatively complete except for minimal wear at the roots. A major characteristic of this tooth that distinguishes it from the equivalent tooth in other hyaenid species is its bucco-lingually slender paracone compared to other modern hyaenids, in addition to the presence of robust mesial and distal accessory cusps. The roots of the P 3 have broken, but the crown is complete and in good condition. Whereas the tooth is slender bucco-lingually compared to those of other modern hyaenid species, it possesses both mesial and distal accessory cusps, while in other modern hyaenids, only the distal accessory cusp is present. The distal accessory cusp of P 3 in Chasmaporthetes silberbergi s is more pronounced than the mesial accessory cusp. Comparison of the new specimen and those of already identified materials of extinct hyaenids housed at the Bernard Price Institute of Palaeontology. was conducted in order to rule out the other extinct hyaenid species. The specimen has a more primitive premolar development which distinguishes it from 152
Chasmaporthetes nitidula, which has a cheetah-like development of the anterior and posterior cusps of the premolars (Turner 1987). Chasmaporthetes nitidula (Ewer, 1955) Extinct long-legged hunting hyaena BP/3/22742, left P 4 (Plate. 9.12) Only the anterior half of the tooth is preserved. However the preserved section is in good condition and necessary morphological characteristics to identify it to species are well preserved. The presence of robust mesial accessory cusp distinguishes it from the modern hyaenid species. The distinction of this species from other Chasmaporthetes species derives from the extra robustness of this mesial accessory cusp in Chasmaporthetes nitidula when compared with those of other Chasmaporthetes species. The major differences in the lower premolars of C. silberbergi and C. nitidula are: those in C. nitidula are more oval in occlusal outline and have a fleur-de lis profile on anterior and posterior edges of the main cusp, while those of C. silberbergi are rectangular in occlusal outline and tend to be straight on the anterior and posterior edges of the main cusp (Turner 1987). Chasmaporthetes sp. Extinct long-legged hyaena species BP/3/22741, left P 2 (Plate. 9.13); The specimen is in good condition, except for the missing lingual portion of the paracone. The distinction of this specimen with the equivalent from modern 153
hyaenids is with the presence of both mesial and distal accessory cusps. In modern hyaenids, only the distal accessory cusp is present. The other distinction is the size of the distal accessory cusp, which is more robust in Chasmaporthetes than in modern hyaenas. Carnivore Skeletal Part Representation Large and small carnivores are represented in the Jacovec Cavern infill. The majority of the species were identified to generic and/or species level on the basis of cranial elements. Among the extinct carnivores, the Chasmaporthetes comprise the majority of the specimens (Appendix II). Postcranial deriving from the Chasmaporthetes is distinct from the rest of hyaena in its elongation. Chasmaporthetes has been identified to be an actively hunting rather than a scavenging and occasionally hunting hyaena. Only one juvenile hyaena was identified on the basis of a distal humerus. In addition to the felids identified to species on the basis of cranial materials, two other mature medium felids (leopard/cheetah size), one medium juvenile felid and one juvenile large felid are represented. In total, a hundred and forty-seven cranial and postcranial specimens derived from the felids. Among the small carnivores represented is Canis mesomelas, identified from a left P 2, Genetta genetta, identified from a left mandible, Felis caracal, identified from a left P 4, Vulpes chama and Cynictis penicillata each identified from a left mandible. A species indeterminate Herpestinae was identified from a right M 1, and a juvenile small carnivore was identified on the basis of a left calcaneum (Appendix II). 154
Minimum Number of Individuals A total of twenty-two carnivore individuals were recovered from the Jacovec Cavern assemblage. This accounts for 32% of the total number of individuals recovered from the Jacovec Cavern. The carnivores range from large extinct carnivores such as Homotherium latidens to small carnivores such as Cynictis penicillata. Among the carnivores, the felids account for 41% of carnivore individuals, with three large size adults, one large size juvenile, three medium size adults, one medium size juvenile and one small size juvenile; the hyaenas account for 36% of the total carnivore individuals, with seven adults and a juvenile: while the remaining 23% (five individuals) is represented by small carnivores, the size of bateared fox and other smaller sizes. Other Fauna Other fauna represented in this infill include a single tortoise individual, identified on the basis of forty-four specimens deriving from the carapace and postcrania. One equid is represented by an upper third molar. Two rodents, a Pedetes capensis identified from two femora and upper right incisor, and another unidentifiable to species level, identified from a left lower incisor, is represented in addition to the hundreds of thousands of other rodent and microfauna which are not included in this study. A Potamochoerus porcus was identified from a left hemimandible. Two lagomorphs are represented in this infill, a Lepus capensis, identified from two tibiae and a femur, and an indeterminate lagomorph identified from five postcranial elements. The hyraxes are represented by four adults, identified from nine specimens, with a majority deriving from the cranium; and two juveniles 155
identified from eleven specimens, the majority of which derive from the cranium (Appendix II). Systematic palaeontology Family: EQUIDAE Equidae indet. An indeterminate sub-adult equid BP/3/22648, right M 3 (Plate. 9.14) The specimen is of a sub-adult, as the roots are not yet in formation. In equids, the crown continues to grow and the roots appear only at the last stages of tooth formation (Hillson 1995). The crown is broken at various sections exposing the infundibulums. The breaks on the crown at the buccal surface are old while those on the lingual surface are fresh. Family: SUIDAE Potamochoerus porcus (Linnaeus, 1785) Bush pig BP/3/22871, left hemimandible (Plate. 9.15) The specimen is a portion of hemimandible with P 2 -P 4 in place. The specimen has been crushed medially causing loss of bone surrounding P 3 and P 2 on the lingual surface. All the premolars present are heavily worn, indicating the individual was very mature. 156
Family: PEDETIDAE Pedetes capensis (Forster, 1778) Spring hare BP/3/22572, right upper incisor; BP/3/22801, right femur; BP/3/22502, right femur (Plate. 9.16) The right upper incisor is preserved on the buccal side and a small portion of the mesial surface. The breakage that led to loss of the whole of lingual surface, portion of the mesial and the entire distal portion is fresh, indicating it occurred during or after excavation. Both femora derive from juvenile individuals, as indicated by the unfused surface for the femur head and great trochanter. The distal shaft and distal epiphyses of the femora are also missing. The preserved portion of the shaft for specimen BP/3/22502 is 50% complete, missing the posterior section, while it is 100% complete for the other specimen. The breakage on specimen BP/3/22801 is fresh while that in the other specimen is old. 9.b. Taphonomy Bone Surface Modification In total, eighty-one bone elements preserve signs of surface modifications. Mammalian (non-rodent) feeding damage There are fifty-nine unidentifiable specimens that preserve signs of mammalian feeding damage. Eighteen of these are digested specimens, while the rest display other types of carnivore mastication damage. One small felid metapodial displays tooth pits and tooth scores; a size class II mammal caudal vertebrae displays 157
gnaw marks; while eight specimens belonging to size class I mammal display carnivore modifications in various forms ranging from tooth pit, tooth scores to gnaw marks. A Vulpes chama hemimandible displays carnivore tooth marks on the basal border. Rodent Gnawing Only two specimens were recognized as having been modified by rodent gnawing. One of the specimens is a metapodial shaft of a small mammal, while the other is an unidentifiable long bone shaft fragment. Insect Damage Nine specimens preserve indications of insect damage. Eight of these are unidentifiable long bone shaft fragments less than 4 cm in length, while one carpal/tarsal deriving from a size class III mammal displays insect boring. Assemblage formation There are various taphonomic processes that lead to accumulation of bones. Some of these processes are easily discernable from the characteristic signatures they imprint on bones. For example, suggestions of carnivore involvement are recognized by the presence of carnivore related modification such as tooth marks and/or digestion related corrosion. Researchers thus investigate for both conspicuous and inconspicuous signatures that may shed light on the taphonomic history of an assemblage. The depositional environment of an assemblage may contribute in preservation or deletion of previous taphonomic signatures. In volcanic areas, preservation is usually better than in other depositional environments in dolomitic caves, the preservation of taphonomic signatures is more or less a gamble as these zones are characterized by cyclical processes of deposition, collapse and erosion. 158
These processes lead to loss of portions of bone, loss of bone surface, and in some instances loss of entire bones. These caves are also characterized by accumulation of travertine, which in some instances is embedded on bone surfaces, thus masking any taphonomic signatures that may have been present. Such a scenario is well seen in Jacovec Cavern where the orange breccia was first deposited, then partially eroded before the brown breccia was deposited. Eventually, there was collapse which resulted in the mixing of the two breccias. It is with this history in mind that the reconstruction of the dynamics involved in the assemblage formation within this cavern is attempted. Though the collapse in Jacovec Cavern resulted to breakage of bones, which may have resulted in the loss of signs of bone modification, there are crucial pointers preserved that can shed light into formation of the assemblage. First of these is the carnivore and rodent modified bones. The presence of tooth marks indicates that carnivores and rodents had access to the bones but does not necessary implicate them in direct involvement in the contribution to the accumulation of the assemblage. This is true when one considers that other unrelated processes such as fluvial activity, may incorporate surface accumulations into the caves. However, when presence of tooth marks is expressed with other lines of data such as the presence of digested bones, it makes a strong case in arguing for direct involvement of the carnivores. Digested bones may result from regurgitations and/or faeces. Observations of modern hyaenas indicate that they void close to their dens where they often regurgitate (e.g., Bearder 1977; Kruuk 1972; Sutcliffe 1970). This leads to their regurgitations and faeces becoming incorporated in den assemblages by fluvial activity and other natural processes (Pickering 2002). The recovery of digested bones indicates that the cavern may have been used as a denning site or carnivores frequented and rested on the 159
grounds above. The extent of carnivore involvement in the accumulation of the assemblage is indirectly investigated in the light of skeletal part representation and presence or absence of some taxa. Hyaenas use caves and burrows as denning and breeding sites, where they sometimes accumulate bones from carcasses brought in for their cubs or for further processing. Juvenile hyaenas spend most of their time in the dens and if they venture out, they do not wander away from the entrance where they run a risk of being killed by other carnivores or by conspecifics (e.g. East et al. 1989; Kruuk 1972; Mills 1990). In case of natural death or where the cubs are killed by conspecifics, their carcasses are not normally transported or consumed (e.g., Kruuk 1972; Mills 1990). Thus complete or near complete remains are expected in such dens, if postdepositional processes have not deleted signs of their presence. A single juvenile hyaenid postcranial element is not substantial enough to support the argument in favour of denning. Cranial remains especially complete tooth rows, isolated teeth or tooth elements have higher chances of survival in cases of intense postdepositional destruction. It is thus possible that the juvenile-hyaena humerus element probably washed in from the grounds above. Another aspect worth investigation is the representation of other carnivores in the assemblage especially that of small carnivores. Researchers have documented numerous instances of interspecific killings among mammalian carnivores (e.g. Rosenzweig 1966; Polis et al. 1989; Palomares & Caro 1999). The killings may be symmetrical (both species kill each other) or asymmetrical (one species kills the other), and in some interactions adults of one kill young but not adults of the other (Palomares & Caro 1999:492). Though not in exclusivity, smaller carnivores are more prone to be killed than big carnivores (Palomares & Caro 1999). However, their 160
remains might be more infrequent in an assemblage since their bodies are consumed almost entirely. In their research, Palomares and Caro (1999) recorded that the herpestids, viverrids, and procyonids were the least likely killers of other carnivores, and smaller carnivores that group are able to kill victims 12 times their weight but solitary species have a limit above which a victim cannot be killed (Palomares & Caro 1999: 492). Other researchers have also documented cases where smaller carnivores have killed bigger ones and in some instances consumed the carcasses (Palomares & Caro 1999: Table A1). Modern spotted and brown hyaenas have been documented to consume a variety of items (e.g., Kruuk 1972; Bearder 1977; Mills 1990; Skinner et al. 1992). Of note was the high abundance of small carnivores in brown hyaena diet (Mills 1990). Pickering (2002) documented as many as 6 out of 12 species of animals consumed by brown hyenas were small carnivores. Various researchers have suggested why small carnivores are prone to predation by hyaenas. Brain (1981) suggested that since the bodies of small carnivores are easily ravaged, hyaenas prey upon them to serve as provision for food for the cubs. Mills (1990) and Mills and Mills (1997) suggested that small carnivores are easily transported over long distances by brown hyaenas and the overlap in food acquisition strategies between the hyaenas and these small carnivores makes the latter more vulnerable and prone to becoming prey (Owen & Owen 1978; Klein et al., 1991). Eight of the twenty-one carnivore individuals identified from the Jacovec Cavern are small carnivores. These include potential hyaena prey such as Canis mesomelas, Vulpes chama, Genetta genetta, and Cynicitis penicillata. On a study of potential predation of carnivore species by other carnivores, Caro & Stoner (2003) discovered that small size, overlap in geographic range and habitat with many of the large carnivores contributes to small canids and small 161
mongooses being the most vulnerable to predation by large carnivores. It is thus possible that the small carnivores within the Jacovec Cavern assemblage were victims of larger carnivores. The low representation of postcranial elements in addition to presence of carnivore tooth marks on the anterior border of a Vulpes chama hemimandible supports this view. The larger carnivores represented in the Jacovec Cavern may equally have been victims of other carnivores or they may have died naturally in the grounds within the vicinity of the cave and their skeletal remains were incorporated in the cave through other processes such as slope wash. The presence and representation of hyraxes by a majority of cranial elements may indicate possible predation. Ravaging of carcasses belonging to hyraxes by eagles and carnivores such as leopards and hyaenas results in a refuse assemblage dominated by cranial elements (Brain 1981). However, absence of carnivore related modification on these specimens does not strengthen such an argument. Considering the complexities involved in accumulation of bones into cave, other processes such as slopewash of material within the caves vicinity cannot be ignored. The involvement of fluvial action in Jacovec Cavern has been proved through the analysis of breccia sediments. Geological analysis of nine rock types sampled from within the breccias indicated that they were fluvially transported from variable distances but not further away from the cave (Kibii 2000). This fluvial action may have most probably brought within it other materials from the vicinity which may have included bones. The absence of porcupine and porcupine gnawed bones is crucial in inferring the nature of entrance to the cave. Porcupines are quite agile and can move about in most caverns except vertical shafts. They, in most cases, exploit burrows that have been abandoned by carnivores and other animals. It is in these caves/burrows that they accumulate a range of items especially bones which they gnaw in an effort to trim 162
down and sharpen their ever growing incisors (e.g., Brain 1981). Thus the absence of the porcupines and/or porcupine gnawed bones indicates that the entrance was not easily negotiable. This in extension provides support against the cave having acted as a den. Pickering (2002) outlined some of the criteria that should be used in differentiating faunal assemblages accumulated by hyaenas and hominids. Among the criteria is the abundance of carnivores in relation to other faunal species in hyaenaaccumulated assemblages. In the Jacovec Cavern assemblage, the 32% carnivore representation may provide additional support, in addition to signs of carnivore modification, to argue for carnivore having significant contribution to the accumulation of the assemblage, not within the cavern but within the vicinity of the cavern in the grounds above. However, since the deposits within the Jacovec Cavern have undergone collapse and erosion, it is difficult to determine how much of the original assemblage is represented by what has been excavated. Thus some of the biased skeletal body part representation may equally be as a result of postdepositional processes. Nevertheless, this doesn t compromise the making of general observations and conclusions. 163