AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 79:43-49(1989) Carnivore Alteration of Human Bone From a Late Prehistoric Site in Illinois GEORGE R. MILNER AND VIRGINIA G. SMITH Department of Anthropology, The Pennsylvania State Uniuersity, University Park, Pennsylvania 16802 (G.R. M.); Department of Anthropology, University of Kentucky, Lexington, Kentucky 40506 (KG.S.) KEY WORDS Taphonomy, Human bone, Oneota ABSTRACT The nature and pattern of destruction of human bone by scavenging animals, probably canids, is described for 30 Oneota skeletons from a west-central Illinois cemetery dated at ca. A.D. 1300. The most frequently damaged parts of the skeleton include bones covered by relatively little soft tissue, such as the major joints of the limbs, and those in the facial, abdominal, and gluteal regions. Over the past several decades, researchers in several disciplines have directed increased attention toward the taphonomic processes affecting bones in archaeological and paleontological assemblages (Behrensmeyer, 1976; Gifford, 1981; Johnson, 1985; Shipman, 1981; Shipman and Rose, 1983). Important aspects of this research have included studies of the disintegration and dispersal of mammalian skeletons, including the nature of damage attributable to carnivores and scavengers (Binford, 1981; Bonnichsen and Will, 1980; Brain, 1981;Bunn, 1981;Haynes, 1980,1982, 1983, 1985; Hill, 1976, 1980; Klippel et al., 1987; Miller, 1969; Potts and Shipman, 1981; Shipman, 1981; Shipman and Rose, 1983; Sutcliffe, 1970),the differential susceptibility of various skeletal elements to destruction (Binford, 1981; Binford and Bertram, 1977; Brain, 1981; Behrensmeyer and DechantBoaz, 1980; Haynes, 1982, 1985; Klippel et al., 1987), and patterns of carcass disarticulation (Binford, 1981; Brain, 1981; Haynes, 1982, 1985; Hill, 1976, 1979a,b, 1980; Hill and Behrensmeyer, 1984; Klippel et al., 1987; Shipman and Phillips, 1976; Toots, 1965). There is little published information, however, that specifically covers damage to human remains caused by carnivores and scavengers (see comments in Maples, 1986; Morse, 1983; Sutcliffe, 1970; Zimmerman et al., 1981). This information is of importance to the interpretation of early hominid finds (Brain, 1970, 1981; Tobias, 1974), some archaeological human skeletal series (Milner and Smith, n.d.; Owsley et al., 1977;Zimmer- 0 1989 ALAN R. LISS, INC. man et al., 19811, and modern forensic cases (Krogman and Iscan, 1986; Maples, 1986; Morse, 1983). The scavenger-induced alteration of human bone described here is based on an examination of 30 partial and complete skeletons from a late prehistoric cemetery in the American Midwest. The kinds of bone damage and the sites of involvement on the elements available for study are identified. Since the initial collection and subsequent burial of these remains were not conducted under controlled conditions, it is not possible to identify unambiguously the animals responsible for the damage or to assess the rate of bone destruction, the differential loss of entire skeletal elements, or the typical disarticulation sequence. The sample acquires comparative significance, however, because it is large, the surviving skeletal elements are exceptionally well preserved, and the remains are drawn from one locality featuring a single suite of possible scavengers. THE SKELETAL SAMPLE The study specimens are part of a collection of 264 skeletons from a carefully and completely excavated Oneota burial mound morris Farms #36) in west-central Illinois dated at ca. A.D. 1300. Forty-three skeletons (16% of the total) exhibited one or more of the following characteristics: damage resulting from arrow wounds or blows from weapReceived January 25, 1988; revision accepted June 24, 1988. 44 G.R. MILNER AND V.G. SMITH TABLE 1. Carniuoredamaged bones Bone Cranial Mandibles Clavicles Scapulae Humeri Radii Ulnae Carpals Metacarpals Hand phalanges Manubria Sterna Xiphoid processes Ribs Cervical vertebrae Thoracic vertebrae Lumbar vertebrae Sacra Coccyges Innominates Femora Patellae Tibiae Fibulae Calcanea Other tarsals Metatarsals Foot phalanges N Present N Damaged % Damaged 12 10 32 37 37 34 37 125 89 167 16 13 4 372 117 252 110 22 3 0 9 14 16 13 16 0 9 3 2 2 1 33 5 46 40 24 41 32 27 130 31 12 9 0 26 24 3 24 9 7 9 66 10 5 loo 2 25.0 0.0 28.1 37.8 43.2 38.2 43.2 0.0 10.1 1.8 12.5 15.4 25.0 8.9 4.3 12.3 10.9 40.9 0.0 56.5 60.0 12.5 58.5 28.1 25.9 6.9 2.0 15.2 ‘The facial bones were reasonably intact in seven of the nine crania that did not exhibit carnivore-related damage. ons, including Celts; cut marks from scalping or decapitation; cut marks from partial postcranial dismemberment; and small, conical tooth impressions accompanied by extensive destruction of cancellous bone and a splintering of the cortex attributable to scavenging animals. These forms of bone destruction were distinctly different from antemortem osteolytic lesions (Ortner and Putschar, 1985; Steinbock, 1976) and the fractures, warping, surface erosion, and rodent gnawing that can modify the appearance of bone long after death (Behrensmeyer, 1978; Bonnichsen and Will, 1980; Johnson, 1985; Miller, 1975; Shipman, 1981; Shipman and Rose, 1983). Differentiation of defects was facilitated by the state of bone preservation where intact bone surfaces permitted the identification of fine anatomical detail (the skeletal series is described fully in Milner and Smith [n.d.]). Forty-one of the 43 affected skeletons were from individuals who were over the age of 15 years at the time of death. Thirty of these adults displayed a postmortem modification of bone attributable to scavenging animals. These individuals were killed in intergroup conflict, and occasionally they were partially dismembered, as indicated by multiple cut marks on selected skeletal elements. Their remains were subsequently recovered and interred in the burial mound. All bones in the skeletal series were examined for evidence of carnivore-induced alteration identified by criteria that closely followed descriptions of damage by previous researchers (Binford, 1981; Bonnichsen and Will, 1980; Brain, 1981; Haynes, 1980, 1982, 1983, 1985; Hill, 1976, 1980; Miller, 1969; Potts and Shipman, 1981; Shipman, 1981; Shipman and Rose, 1983; Sutcliffe, 1970). Only unambiguous examples of scavengerrelated bone destruction are included in Tables 1 and 2, and this damage conformed to the characteristics described below. Postmortem defects of undetermined origin, such as localized areas of bone erosion or fragmentation, were not counted as having been gnawed by scavengers. Identification of the likely cause of bone destruction proved to be consistently difficult only for broken ribs; consequently, only relatively intact specimens were included in the counts presented in the tables. Damage attributable to rodent gnawing, which occurred occasionally, was readily distinguishable from carnivore-related bone destruction, and it was not included in the tables. DAMAGED BONES IN THE STUDY SAMPLE Bone damage ranges from discrete or confluent puncture marks where cortical bone had been penetrated to complete destruction of significant portions of flat and irregularshaped bones as well a s the ends of long bones (Figs. 1-4). Punctures are circular, oval, or, less frequently, elongated depressions bordered by adherent fragments of cortex inclined toward the defect’s center. Many of the discrete circular to oval defects appear to have been made by canines. A number of the circular to oval, shallow puncture marks that penetrated thin cortical bone probably approximate the dimensions of the tips of the teeth, presumably canines, responsible for damaging the bones. Small fragments of cortex around the margins of these defects are usually oriented approximately perpendicular to the bone surface. These small fragments of adherent cortex had been pressed into the interior of the bone and deflected laterally, in many instances producing well-defined walls. These particular defects measure several millimeters in diameter (N = 67; length: X = 3.8 mm, s = 45 CARNIVORE DAMAGED HUMAN BONE TABLE 2. Carnivore-damaged anatomical features of selected bones Bone Clavicles Medial end Lateral end Scapulae Acromion process Coracoid process Axillary border Inferior angle Vertebral border Superior angle Superior border Humeri Head Greater tubercle Medial epicondyle Lateral epicondyle Distal joint surface Radii Proximal end Distal end Ulnae Proximal end Distal end Innominates Anterior superior spine Posterior superior spine Iliac crest Ischial tuberosity Ischiopubic ramus Pubic body Superior pubic ramus Femora Head Greater trochanter Lesser trochanter Distal end Tibiae Proximal end Distal end Fibulae Proximal end Distal end N Present N Damaged % Damaged 32 32 4 7 12.5 21.9 37 35 34 32 26 26 23 10 2 27.0 5.7 2.9 34.4 7.7 7.7 8.7 36 36 37 37 37 1 11 2 2 2 5 8 6 3 13.9 11.1 21.6 16.2 8.1 34 34 7 10 20.6 29.4 37 37 13 9 35.1 24.3 45 44 45 43 43 41 44 15 7 12 17 9 7 5 33.3 15.9 26.7 39.5 20.9 17.1 11.4 9 4 40 40 40 40 3 23 22.5 27.5 7.5 57.5 41 41 19 16 46.3 39.0 30 32 7 7 23.3 21.9 0.9 mm, range = 2.3-6.7 mm; width: X = 2.9 mm, s = 0.7 mm, range = 1.7-4.7 mm). Penetration of the cortex, however, usually was accompanied by the formation of irregular cortical fragments with jagged edges that were inclined at various angles toward the point of impact. These defects varied in size, but were typically several millimeters larger than the well-defined punctures through thin cortical bone. Where the cortical bone is thick, the presence of small, shallow pits or linear troughs indicate where individual teeth had crushed but failed to penetrate the cortex. Occasionally a tooth apparently skidded across a n area of extremely thin cortical bone, including that of many long bone epiphyses. These excursions produced linear, broad, shallow, 11 more-or-less U-shaped bands of crushed cortex pressed into the spongy bone interior. A coalescence of defects resulted in a rough surface of broken bone composed of numerous spicules arising from the irregular edge of intact cancellous bone as well as from dislodged, but still partially adherent, trabeculae and cortical bone splinters. Further gnawing resulted in a considerable loss of bony substance, particularly the removal of cancellous bone. Sometimes a thin cortical shell is all that remains of the end of a long bone. The edges of extensively damaged areas where teeth had penetrated the cortex are frequently scalloped. Internally, the surface of damaged cancellous bone often exhibits shallow, conical depressions or linear, U-shaped troughs. Multiple linear excava- 46 G.R. MILNER AND V.G. SMITH Fig. 1. The medial epicondyle of the humerus is damaged. and a large puncture mark borders the coronoid fossa. tions through cancellous bone are often uniformly aligned and extend across much of the damaged area. Cortical surfaces adjacent to extensively damaged ends of large diameter long bones occasionally exhibit short, shallow, broad grooves where teeth had apparently ground into and skidded across the shaft’s surface. Some grooves are oriented transversely, whereas others extend longitudinally or obliquely along the shaft for several millimeters before terminating at the edge of a n extensively damaged area where both the cortex and underlying spongy bone had been destroyed. This point of intersection is sometimes marked by a semicircular indentation, i.e., a tooth mark, a t the margin of the damaged area. In the relatively few instances where most, or all, of a long bone’s end had been destroyed, the removal of small fragments of compact bone produced a jagged edge on the surviving portion. Extensive splintering of diaphyses is restricted to relatively few bones, usually radii and ulnae with fragmented distal shafts. When this occurred, the corresponding hand bones were not found with the burial. This longitudinal and oblique splintering is quite distinctive and bears no resemblance to the kinds of fragmentation, such as transverse breaks, found on bones broken long after burial. Cortical bone is also splintered where portions of flat bones, especially the ilia, and irregularly shaped elements, such a s the vertebral trans- Fig. 2. The caudal end of this sacrum is destroyed. Note the scalloped inferior border and several puncture marks. verse and spinous processes, are extensively damaged. The number of bones present for examination and those displaying unambiguous evidence of scavenger-related damage are listed in Table 1. Only certain portions of these bones are consistently damaged (Table 2). DISCUSSION The sample provides indirect information on animal feeding patterns, since the remains presumably originated as surface assemblages where scavenger-induced damage was truncated at different stages by recovery and subsequent burial. The distribution of affected areas on the available skeletal elements, however, clearly indicates that certain portions of bones are more likely to be damaged by gnawing than other parts of the skeleton. Frequently damaged areas include those in the vicinity of the abdomen, especially the anterior superior iliac spine and iliac crest, and the gluteal region, including the ischial tuberosities, the greater trochanter and head of the femur, and the caudal end of the sacrum. Bones covered by little soft tissue are also frequently damaged, including the bones of the face and those comprising the major joints of the upper and lower limbs. The calcanea clearly illustrate differential destruction attributable to their anatomical position and structure. Some specimens exhibited only minor damage in the exposed distal portion of the bone, whereas others featured extensive damage CARNIVORE DAMAGED HUMAN BONE 47 Fig. 3. The cancellous bone of this tibia has been scooped out, undercutting the more resistant cortical bone, and several shallow, transverse excavations through the spongy bone interior are visible. Fig. 4. This distal femur displays several discrete puncture marks and an extensively destroyed medial surface bordered by a scalloped margin. resulting in the loss of all but more-or-less intact anterior and superior articular surfaces. Significant underenumeration of damaged ribs, the bones of the hands and feet, and coccygeal elements is likely. Many ribs are fragmented, especially distally, but only specimens with unambiguous scavenger-related alterations were counted as damaged elements. Definite gnawing damage, including conical defects or small, semicircular fractures along bone margins, occurred most often on the distal ends of the ribs. Presumably, hands and feet were frequently gnawed as well. The nature of damage on surviving elements, however, suggests that gnawing often resulted in the complete destruction or dispersal of these bones by scavengers. The likelihood of recovery for burial of appreciable numbers of damaged hand and foot bones, therefore, would have been diminished. Certainly these particular bones are underrepresented in this skeletal sample. Finally, coccygeal elements often would have been destroyed by the gnawing that resulted in damage to the sacrum, since sacral destruction was most pronounced at the caudal end of the bone. Identification of the animals responsible for the damage rests on the nature of destruction and the fauna conceivably present in late prehistoric, west-central Illinois. Over- all, the appearance of the damaged areas and the sites of involvement conform closely to descriptions of nonhuman mammalian bones subjected to gnawing by canids, especially wolves and dogs (Binford, 1981; Brain, 1981; Haynes, 1980, 1982, 1985). The extent of destruction is consistent with that reported for bones found a t canid kill sites in contrast with the more highly fragmented elements, which often exhibit eroded surfaces, recovered from wolf dens and human camps frequented by dogs (Binford, 1981; Haynes, 1982; Klippel et al., 1987). Dogs, coyotes, and wolves were present in late prehistoric Illinois (Purdue and Styles, 19861, and it is likely that some or all of these animals were responsible for most of the scavenger-induced damage in this skeletal collection. Other scavenging animals may have contributed to the postmortem destruction of bone as well. The pattern of bone destruction is similar to the damage reported by Zimmerman et al. (1981) for a late prehistoric, South Dakota skeletal series consisting of individuals who had been killed and whose remains were subsequently gnawed by scavengers, probably canids. In addition, modern human skeletons exhibiting similar damage, also attributed to dogs, have been recovered in forensic fieldwork in Kentucky (personal communication, David J. Wolf; one of the authors [V.G.S.] has examined many of these specimens). 48 G.R. MILNER A ND V.G. SMITH dnvai Gorge. Nature 291574-577. Gifford DP (1981) Taphonomy and paleoecology: A critiFragmentation by scavenging animals of cal review of archaeology’s sister disciplines. In MB Schiffer (ed.): Advances in Archaeological Method and human bones from a n archaeological site in Theory, Vol. 4. New York: Academic Press, pp. 365the American Midwest follows a consistent 438. pattern where certain parts of the skeleton G (1980) Evidence of carnivore gnawing on Pleisare damaged before other elements. The Haynes tocene and recent mammalian bones. Paleobiology areas most frequently affected include the 6:341-351. ends of long bones, where there is generally Haynes G (1982) Utilization and skeletal disturbances of North American prey carcasses. Arctic 35266-281. little covering soft tissue, and the bones in the vicinity of the facial, abdominal, and glu- Haynes G (1983) Frequencies of spiral and green-bone fractures on ungulate limb bones in modern surface teal regions. Overall, the tooth impressions assemblages. Am. Antiquity 48:102-114. and the fragmentation patterns closely match Haynes G (1985) On watering holes, mineral licks, death, the kinds of bone destruction produced by and predation. In JI Mead and DJ Meltzer (eds.): Environments and Extinctions: Man in Late Glacial North canids, including dogs, coyotes, and wolves, America. Orono: Center for the Study of Early Man, all of which were present in late prehistoric University of Maine, pp. 53-71. Illinois. Hill A P (1976) On carnivore and weathering damage to bone. Curr. Anthropol. f7t335-336. ACKNOWLEDGMENTS Hill AP (1979a) Butchery and natural disarticulation: An investigatory technique. Am. Antiquity 44:739The excavation of the Norris Farms #36 744. site was conducted by Illinois State Museum AP (197910) Disarticulation and scattering of mamarchaeologists and supported by the Illinois Hill mal skeletons. Paleobiology 5261-274. Department of Transportation. This study of Hill AP (1980) Early postmortem damage to the remains the skeletons was facilitated by the generous of some contemporary East African mammals. In AK Behrensmeyer and AP Hill (eds.): Fossils in the Makloan of the collection to the senior author by ing: Vertebrate Taphonomy and Paleoecology. Chithe Illinois State Museum. We thank David cago: University of Chicago Press, pp. 131-152. J. Wolf (Forensic Anthropologist, Medical Ex- Hill AP, and Behrensmeyer AK (1984) Disarticulation aminer Program, Justice Cabinet, Commonpatterns of some modern East African mammals. Paleobiology 1Ot366-376. wealth of Kentucky) for his comments regarding modern forensic cases. Robert B. Johnson E (1985) Current developments in bone technology. In MB Schiffer (ed.): Advances in Archaeological Eckhardt and anonymous reviewers proMethod and Theory, Vol. 8. New York: Academic Press, vided useful comments on earlier versions of pp. 157-235. this paper. Klippel WE, Snyder LM, and Parmalee PW (1987) Taphonomy and archaeologically recovered mammal bone from southeast Missouri. J. Ethnobiol. 7:155-169. LITERATURE CITED Krogman WM, and Iscan MY (1986) The Human Skeleton in Forensic Medicine, 2nd ed. Springfield Charles Behrensmeyer AK (1976) Taphonomy and paleoecology in the hominid fossil record. Yrbk. Phys. Anthropol. C. Thomas. 19:36-50. Maples WR (1986) Trauma analysis by the forensic anBehrensmeyer AK (1978) Taphonomic and ecologic inforthropologist. In K J Reichs (ed.): Forensic Osteology: Advances in the Identification of Human Remains. mation from bone weathering. Paleobiology 4t150-162. Springfield: Charles C. Thomas, pp. 218-228. Behrensmeyer AK, and Dechant-Boaz DE (1980) The recent bones of Amboseli Park, Kenya, in relation to Miller GJ (1969) A study of cuts, grooves, and other marks on recent and fossil bone: I. Animal tooth marks. East African paleoecology. In AK Behrensmeyer and AP Hill (eds.): Fossils in the Making: Vertebrate TaTebiwa 12t20-26. phonomy and Paleoecology. Chicago: University of Miller GJ (1975) A study of cuts, grooves, and other Chicago Press, pp. 72-92. marks on recent and fossil bone: 11. Weathering cracks, fractures, splinters, and other similar natural phenomBinford LR (1981) Bones: Ancient Men and Modern ena. In E Swanson (ed.): Lithic Technology: Making Myths. New York: Academic Press. and Using Stone Tools. The Hague: Mouton PublishBinford LR, and Bertram JB (1977) Bone frequenciesers, pp. 211-226. and attritional processes. In LR Binford (ed.): For Theory Building in Archaeology. New York: Academic Milner GR, and Smith VG (n.d.1: Oneota human skeletal Press, pp, 77-153. remains. (Unpublished manuscript ) Bonnichsen R, and Will RT (1980) Cultural modification Morse D (1983) The skeletal pathology of trauma. In D Morse, J Duncan, and J Stoutamire (eds.): Handbook of bone: The experimental approach in faunal analysis. of Forensic Archaeology and Anthropology. TallahasIn BM Gilbert (ed.): Mammalian Osteology. Laramie: see: Florida State University Foundation, pp. 145-185. Privately published, pp. 7-26. Brain CK (1970) New finds a t the Swartkrans australo- Ortner DJ, and Putschar WGJ (1985) Identification of pathological conditions in human skeletal remains. pithecine site. Nature 225:1112-1119. Smithsonian Contributions to Knowledge 28. WashBrain CK (1981) The Hunters or the Hunted? Chicago: ington: Smithsonian Institution Press. University of Chicago Press. Bunn HT (1981) Archaeological evidence for meat-eating Owsley DW, Berryman HE, and Bass WM (1977) Demographic and osteological evidence for warfare at the by Plio-Pleistocene hominids from Koobi Fora and 01CONCLUSION CARNIVORE DAMAGED HUMAN BONE Larson site, South Dakota. Plains Anthropol. Memoir 13.119-131. Potts R, and Shipman P (1981)Cutmarks made by stone tools on bones from Olduvai Gorge, Tanzania. Nature 291t577-580. Purdue JR,and Styles BW (1986) Dynamics of mammalian distribution in the Holocene of Illinois. Reports of Investigations 41. Springfield Illinois State Museum. Shipman P (1981) Life History of a Fossil. Cambridge: Harvard University Press. Shipman P, and Phillips JE (1976) On scavenging by hominids and other carnivores. Curr. Anthropol. 17:170-172. Shipman P, and Rose J (1983) Early hominid hunting, butchering, and carcass-processing behaviors: Ap- 49 proaches to the fossil record. J. Anthropol. Archaeol. 2.57-98. Steinbock RT (1976) Paleopathological Diagnosis and Interpretation. Springfield: Charles C. Thomas. Sutcliffe AJ (1970) Spotted hyaena: Crusher, gnawer, digester and collector of bones. Nature 227t1110-1113. Tobias PV (1974)Aspects of pathology and death among early hominids. The Leech 44t119-124. Toots H (1965) Sequence of disarticulation in mammalian skeletons. Contrib. Geol. 4:37-39. Zimmerman LJ, Emerson TE, Willey P, Swegle M, Gregg JB, Gregg P, White E, Smith C, Haberman T, and Bumsted MP (1981) The Crow Creek site (39BFll) massacre. DACW45-78-C-0018.US Army Corps of Engineers, Omaha District. NOTE ADDED IN PROOF The forensic work of Haglund and coworkers (19881, which was published after the acceptance of this article, closely parallels the results presented for the archaeological sample (Haglund WD, Reay DT, and Swindler DR  Tooth mark artifacts and survival of bones in animal scavenged human skeletons. J. Forensic Sci. 33:985-997).