Tribute to brain titlepage.indd

Series Editors Kathy Schick and Nicholas Toth THE OLDOWAN: Case Studies into the Earliest Stone Age Nicholas Toth and Kathy Schick, editors Taphonomic Studies in Honor of C.K. (Bob) Brain Travis Rayne Pickering, Kathy Schick, and Nicholas Toth, editors New Approaches to the Archaeology of Human Origins Kathy Schick, and Nicholas Toth, editors Paleoneurological Studies in Honor of Ralph L. Holloway Douglas Broadfield, Michael Yuan, Kathy Schick and Nicholas Toth, editors S T O N E A G E I N S T I T U T E P U B L I C A T I O N S E R I E S Series Editors Kathy Schick and Nicholas Toth breathing life
into fossils:
Travis Rayne Pickering
Kathy Schick
Nicholas Toth
Stone Age Institute Press · www.stoneageinstitute.org 1392 W. Dittemore Road · Gosport, IN 47433 Front cover, clockwise from top left. Top left: Artist’s reconstruction of the depositional context of Swartkrans Cave, South Africa, with a leopard consuming a hominid carcass in a tree outside the cave: bones would subsequently wash into the cave and be incorporated in the breccia deposits. 1985 Jay H. Matternes. Top right: The Swartkrans cave deposits in South Africa, where excavations have yielded many hominids and other animal fossils. 1985 David L. Brill. Bottom right: Reconstruction of a hominid being carried by a leopard. 1985 Jay H. Matternes. Bottom left: Photograph of a leopard mandible and the skull cap of a hominid from Swartkrans, with the leopard’s canines juxtaposed with puncture marks likely produced by a leopard carrying its hominid prey. 1985 David L. Brill. Center: Photo of Bob Brain holding a cast of a spotted hyena skull signed by all of the taphonomy conference participants. 2004 Kathy Schick, Stone Age Institute. Bottom left: 2004 Kathy Schick, Stone Age Institute. Bottom right: 2005 Greg Murphy. Published by the Stone Age Institute.
Copyright 2007, Stone Age Institute Press.
All rights reserved under International and Pan-American Copyright Conventions. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, without permission in writing from the publisher.
CHAPTER 12
TAPHONOMY OF IMMATURE
HOMINID SKULLS AND THE TAUNG,
MOJOKERTO, AND HERTO SPECIMENS
GAIL E. KROVITZ AND PAT SHIPMAN
ABSTRACT
INTRODUCTION
This study explored quantitative and qualitative Although the tremendous importance of immature methods for deducing the taphonomic history of imma- crania for documenting the growth, development, and ture hominid crania, in the hopes of developing models evolution of various hominid species is widely recog- for use in diagnosing the timing and pattern of break- nized, their taphonomy has not been systematically age in specimens of unknown history. First, two sets of addressed. That immature crania are rarer in the fossil cranial inventory data (of modern human archaeological record than adult crania is generally attributed to their samples) were used to develop a model of the taphonom- greater vulnerability to damage because immature su- ic vulnerability of the different parts of the immature tures are unfused and immature bones are thinner than hominid cranium. We found that the nasal bones, vomer, adult ones (see discussion in Saunders, 2000). Here, we basilar and lateral elements of the occipital, and zygo- report a fi rst step toward deducing the taphonomic his- matic arches were highly vulnerable to breakage or loss, tory of immature hominid crania, which may prove use- while the orbital rim of the frontal bone was rarely miss- ful in determining the timing, causes, and implications of ing. In fact, the frontal bone plays an important role as a keystone that keeps crania articulated. Second, we re- The primary aim of our project was to establish the viewed the pertinent medical and forensic data on cranial expected pattern of breakage and destruction to imma- damage, and discussed three temporal stages of cranial ture hominid crania that have been subjected to minimal breakage: stage 1 (wet or fresh bone breakage), stage 2 taphonomic disturbance. To this end, we carried out both (dry bone or postmortem breakage), and stage 3 (post- quantitative and qualitative studies. We relied upon two fossilization breakage). Patterns of breakage and disar- sets of cranial inventory data. One of us (G.K.) conduct- ticulation in 20 immature fossil hominid crania were also ed an inventory of the immature cranial remains of 272 included in this discussion. Several new fracture types recent humans in six cemetery populations from four were observed in the fossil hominids, including temporal broad geographic regions. Here, we refer to these crania line, perpendicular, and metopic or para-metopic frac- as the Krovitz sample. For each cranium, she recorded tures, and a mosaic fracture pattern. Finally, the mod- the presence or absence of anatomical landmarks as an els discussed above were used to deduce the breakage indicator of the loss or breakage of cranial elements. histories of three immature fossil specimens that were From these inventory data we developed a model of the exposed to different taphonomic infl uences: Taung 1 taphonomic vulnerability of the different parts of the im- (Australopithecus africanus), Mojokerto (or Perning 1, mature hominid cranium. We also analyzed bone inven- Homo erectus), and Herto BOU-VP-16/5 (Homo sapiens tory data on a sample of 81 modern human crania from a cemetery population in England that was excavated and studied by the Sedgeford Historical and Archaeo-logical Project (SHARP); unpublished data were kindly 208  Breathing Life into Fossils: Taphonomic Studies in Honor of C.K. (Bob) Brain
provided to us by Patricia Reid of SHARP. The SHARP QUANTITATIVE STUDIES
sample had been scored by the original researchers for the presence or absence of eight major cranial bones; age Materials and methods
at death and sex had also been assigned. Only three of 82 individuals in the SHARP sample were immature, so To determine how immature crania were preserved this sample must be taken as representative of the cranial and damaged in specimens that were exposed to rela- taphonomy of adults. Description and comparison of the tively few taphonomic forces, we used two sources of results of the inventory studies constitute Part I of this The Krovitz sample consists of a landmark inven- In Part II, we fi rst review the pertinent medical and tory of 272 immature crania of recent humans from ar- forensic data on cranial damage and then describe and chaeological samples conducted by one of us (Krovitz, discuss the results of qualitative and quantitative obser- 2000). Although the inventory was not designed for a ta- vations conducted on photographs, casts, and occasion- phonomic study, the results are useful here. The samples ally upon originals of 20 immature fossil crania of Homo represent cemetery populations from England (Christ erectus, Australopithecus africanus, Neandertals, and an- Church Spitalfi elds, 18th and 19th centuries; Adams and atomically modern Homo sapiens (including H. sapiens Reeve, 1987; Molleson and Cox, 1993), Medieval Den- idaltu). Following forensic practice, we recognize three mark (A.D.1000-1500 ), Nubia (A.D. 0-1500; Vagn temporal stages of cranial or bone breakage: stage 1, wet Nielsen, 1970), Edo Period Japan (A. D.1603-1867; or fresh bone breakage, which incorporates the stages Mizoguchi, 1997), St. Lawrence Island Yupik Eskimo forensic scientists recognize as antemortem breakage, (A.D. 1800; Collins, 1937; Utermohle, 1984; Heathcote, which shows signs of healing on the broken edges, and 1986), and Indian Knoll (2500-2000 B.C.; Snow, 1948). perimortem breakage, which does not; stage 2, dry bone For each cranium, G.K. recorded the presence or absence or postmortem breakage; and stage 3, post-fossilization of a set of 39 anatomical landmarks (Figure 1, Table 1). (fossilized bone) damage. These three stages grade into This provides fi ne-grained data on the location of break- one another along a temporal continuum and the place- age or complete loss of cranial elements. ment of a specimen along this continuum has a marked Specimens were only included if they were undis- impact on its response to potentially damaging agents. torted, non-pathological, and had at least one anatomical From our observations of modern and fossil crania, we region (face or neurocranium) that was largely articu- summarize the types, frequency, and morphology of lated. Since disarticulated cranial bones, no matter how breaks seen in crania in each of these stages for potential complete, were not included in the Krovitz sample, this use in diagnosing the timing of breakage in specimens of inventory provides a conservative estimate of breakage for these samples. Recent human samples where com- Finally, in Part III, we use the quantitative and quali- pleteness of the crania was a primary criterion for collec- tative results from the previous two sections to deduce tion were excluded from the inventory.
the breakage histories of three immature fossil specimens Individuals in the Krovitz sample were divided into that were exposed to different taphonomic infl uences. four developmental age groups based on tooth formation These are: Taung 1, Australopithecus africanus, which and eruption sequences: 0 – 3.0 years (Age Group 1), 3.1 was dropped into a cave, probably by a leopard, while it – 6.0 years (Age Group 2), 6.1 – 9.0 years (Age Group was both fresh and fl eshed (McKee, 2001; McKee, 2004, 3), and 9.1 – 13.5 years (Age Group 4). Tooth formation personal communication to P.S.); Mojokerto, Homo was the primary method for dental age estimation (us- erectus, which was deposited in fl uvial sediments (Huff- ing data from Thoma and Goldman, 1960; Moorrees et man, 2001; Huffman and Zaim, 2003) and subjected to al., 1963; Smith, 1991), although tooth eruption was also breakage and plastic deformation (Anton, 2003, personal used when necessary (see discussion in Krovitz, 2000). communication to P.S.) at an unknown time; and Herto These developmental age groups roughly coincide with BOU-VP-16/5, Homo sapiens idaltu, which was modi- the following developmental criteria (after Minugh-Pur- fi ed and curated by hominids after the death of the indi- vis, 1988): 1) infancy (birth to completion of deciduous vidual (Clark et al., 2003; White et al., 2003) and under- tooth eruption and development), 2) early childhood (pe- went primarily post-fossilization damage (White, 2004, riod between deciduous tooth development and perma- personal communication to P.S.). We use the taphonomic nent tooth eruption), 3) mid-childhood (eruption of the vulnerability model developed from the inventory data fi rst permanent teeth), and 4) late childhood (completion and the patterns of breakage and taphonomic destruction of permanent tooth eruption and development, except for observed on casts and high resolution photographs of these specimens to deduce the specifi c taphonomic his- The SHARP data, which were made available to us but were not collected by us, consist of bone invento-ries of 82 individuals represented by articulated remains from the Anglo Saxon cemetery at Sedgeford buried be-tween 662 and 881 A.D. (Stillwell, 2002; Sedgeford His-torical and Archaeological Research Project or SHARP, Krovitz and Shipman  209
Figure 1. Landmarks used in this study. See Table 1 for landmark descriptions. Table 1. Description of landmarks used in this analysis; see Figure 1 for location of landmarks. All landmarks described with L/R were collected from either the right or left side, depending on which side was better preserved in that individual. 22. BRG (Bregma, coronal-sagittal 32. L/R CAR (Carotid canal, pos- 15. L/R PTN (Pterion, intersection 25. L-O (1/2 way between lambda 18. L/R IZA (Inferior temporal-zy- 28. L/R MXT (Maxillary tuberos- ity = junction maxilla and palatine in midline) 29. L/R PAL (Junction on palatine 39. L/R DPM (Behind DM2/P4 on 10. L/R ORB (Top of orbit, 1/2 way 20. L/R AST (Asterion = parietal- 30. VSJ (Vomer sphenoid junction, 210  Breathing Life into Fossils: Taphonomic Studies in Honor of C.K. (Bob) Brain
unpublished data; Reid, 2004, personal communication to P.S.). For each individual, the presence or absence of We could not derive a single predicted sequence of the frontal, maxilla, palatine, zygomatic, sphenoid, pa- disarticulation and damage from these landmark inven- rietal, temporal, and occipital was scored separately for tory data because the variability among specimens was the right and left sides; the values for both sides were too great. Instead, we identifi ed clusters of landmarks averaged for our purposes. If present, each bone was that exhibited high taphonomic vulnerability, interme- scored in completeness categories consisting of: < 25% diate taphonomic vulnerability, and low taphonomic complete; 25-50% complete; 51-74% complete; and 75- vulnerability (Figures 2 and 3; Table 2). The high and low vulnerability clusters together comprise 16 of the 39 Unlike the Krovitz sample (which consisted entirely landmarks (41%). The remaining 23 landmarks (59%) of immature individuals), 79 out of 82 (95%) individu- als in the SHARP sample were adult. Cranial fragments We argue that: (1) the primary factor determining and isolated bones that could not be associated with a the taphonomic vulnerability of a landmark is its struc- particular burial were excluded from the database pro- tural resistance to breakage and destruction, which is a vided to us; however, the SHARP sample did include function of the density of the skeletal element and of its disarticulated cranial bones that could be associated with placement within or projection from the cranium as a a burial (for example, one individual’s cranial remains whole; (2) the breakage of immature hominid crania is consist solely of an occipital bone). This is an important intimately related to the placement and physical nature difference from the Krovitz sample that only considered of sutures; (3) although crania of younger individuals relatively articulated crania and no disarticulated bones. were generally less complete, the taphonomic vulner- Although healed lesions were observed on fi ve individu- ability grouping of most cranial landmarks does not als in the SHARP sample (Stillwell, 2002), no crania change dramatically between the ages of 0-13 years.
were grossly pathological. The SHARP bone inventory provided coarser-grained data on the preservation of var- High taphonomic vulnerability
ious cranial elements in a predominantly adult sample of Landmarks with high taphonomic vulnerability are illustrated in Figure 3 and listed in Table 2a. Several We selected deliberately buried specimens because landmarks in the facial region are missing with remark- this greatly simplifi ed the potential range of taphonomic ably high frequency in the Krovitz sample. These most histories exhibited by the samples. Further, most mod- vulnerable landmarks are missing in almost 70% of in- ern humans are buried in some fashion, so most size- dividuals in the youngest age group and are absent in able samples of modern human crania are derived from 40-47% of the specimens across all age groups. All of cemetery populations. Since interment occurred shortly the most vulnerable facial landmarks are associated with after death, the possibility of lengthy surface weathering, the nasal bones (NAL), the zygomatic arch (SZA, IZA), signifi cant waterborne transport, or substantial carnivore and the vomer (VSJ). Each of these bones has sutures damage was eliminated. What we observed on these with other bones that cover small linear distances and specimens represents a generalized or baseline pattern which therefore probably break or separate more easily of destruction, damage, and preservation of immature than do more extensive sutures. The nasals and the vo- crania undergoing deposition rapidly after death. There- mer are thin and fragile bones prone to damage, and both fore, the condition of crania in our samples refl ects the have an edge projecting into open space. In contrast, taphonomic vulnerability of various parts of the cranium the zygomatic is not a particularly fragile bone but the based primarily on their mechanical resistance to break- zygomatic arch projects from the generally ovoid shape age and on the structural integrity of various sutures be- of the cranium, which makes it vulnerable to breakage. tween bones. The ways in which specimens of unknown The anterior portion of the zygomatic arch is also a thin taphonomic history, such as the fossil crania considered strut of bone that is very susceptible to crushing. In ca- in Part III, deviate from this baseline pattern should pro- daver experiments, McElhaney and colleagues found vide clues to their exposure to other destructive agents.
that the zygomatic arch will break under as little as 130 psi (McElhaney et al., 1976; Mackey, 1984), whereas the Results and discussion
pressure required to fracture the cranial vault is much From the landmark inventory, we calculated the greater: 450 to 750 psi (Cox et al., 1987). On dry crania, percentage of specimens in which each landmark was the zygomatic arch encloses empty space and requires absent in each age group, and in all age groups averaged. The presence or absence of many landmarks was high- Out of the 272 individuals inventoried, 163 (60%) ly correlated with that of other nearby landmarks, with had a face judged to be in good condition, while pres- three obvious clusters of covarying landmarks (face, ervation of the face was judged to be fair or poor in the neurocranium, and basicranium). These groupings were remaining specimens. The entire face was missing in 25 undoubtedly caused by the close spatial relationships of specimens (9% of the sample), suggesting that loss of the landmarks within each anatomical region and the the entire face, usually from nasion downwards, is only general similarity in terms of robustness and/or geom- moderately common in archaeological remains. This Krovitz and Shipman  211
Figure 2. Percentage of landmarks missing for each age group, ordered into low, intermediate and high taphonomic vulnerability. Landmark numbers as in Table 1. Figure 3. Landmarks with high and low taphonomic vulnerability (see Table 2). 212  Breathing Life into Fossils: Taphonomic Studies in Honor of C.K. (Bob) Brain
Table 2. Taphonomic vulnerability of landmarks, separated into High (a), Intermediate (b) and Low (c) taphonomic vulnerability. Landmark abbreviations and numbers as in Table 1. 2b. Intermediate Taphonomic Vulnerability Krovitz and Shipman  213
pattern of damage is known in the forensic literature into the intermediate vulnerability group; just over 20% as a LeFort III fracture (described below, after Rogers, of the specimens (regardless of age group) were missing 1992) and occurs in about 10% of patients seeking medi- at least one of the landmarks on the squamous temporal cal attention for cranial fractures (Richardson, 2000). A LeFort III fracture leaves the cranium in two pieces, the face and the skullcap (neurocranium plus basicranium). Low taphonomic vulnerability
The thinness and vulnerability of the facial bones to ta- Landmarks with low taphonomic vulnerability are phonomic forces means that once these two portions of illustrated in Figure 3 and listed in Table 2c. These land- the cranium have separated, the skullcap is much more marks were present in nearly all of the specimens in the likely to survive and be recovered than the face. Addi- samples examined here regardless of the age at death of tionally, the sphenoid is extremely likely to be broken or the individual. All of the landmarks with the lowest ta- phonomic vulnerability are facial and all are located on Three landmarks in the basicranial region are also the sturdy orbital rim of the frontal bone. They are na- consistently missing in high frequencies (25-42% across sion (NAS), orbitale (ORB), the top of the nasomaxillary all age groups): the anterior part of the jugular process suture at the frontal bone (NMT), the frontomaxillary (JUG); the anterior point of the foramen magnum on the suture at the orbit (FMO), and the frontozygomatic junc- midline or basion (BAS); and the posterior border of the tion at the orbit (FZJ). It is apparent that the robustness occipital condyle with the foramen magnum (CFM). All and structural strength of the orbital rim of the frontal are clustered spatially and are intimately related to the bone has a substantial impact on the frequency of preser- basilar and lateral parts of the occipital bone. The sutures vation of this region of the cranium.
between these ossifi cation centers and the squamous oc- We also believe that the frontal bone acts as a physi- cipital are short and vulnerable to separation; the squa- cal keystone in holding the cranium together and is criti- mous fuses to the basilar portion at about fi ve years, and cal in determining how an immature cranium will break. the lateral and basilar parts fuse in the sixth year (Byers, The frontal has sutural connections with the parietals, 2002). The basilar suture is vulnerable to separation until temporals, sphenoid, maxillae, zygomatics, nasals, and it fuses to the sphenoid between 18-21 years.
ethmoid. The frontal bone plays a key role in hafting Three landmarks on the vault are also missing in the face onto the neurocranium and in reinforcing and moderately high frequencies (25-30% across all age strengthening sutural connections (such as the sagittal groups). These are: lambda (LAM) at the junction of suture) within the neurocranium. If the coronal suture the lambdoid and sagittal sutures; a point (L-O) halfway opens and the face and frontal separate as a unit from the between lambda and opisthion on the midline; and opis- rest of the cranium, then the face has an improved chance thion (OPI), the posterior midline point of the foramen of survival, although the rest of the cranium will almost magnum. All are associated with the squamous occipital. certainly disarticulate. However, if the face breaks off One of the most likely types of taphonomic damage to below nasion in a LeFort II or III fracture (discussed occur to an immature cranium is the loss of the occipital below), then the neurocranium has a better chance of bone due to separation of the lambdoid suture.
survival, but the facial bones will almost certainly dis- The high frequency with which both basilar and squamous occipital landmarks are missing shows that In general, once a cranial bone is isolated its indi- loss of part or all of the occipital must be considered vidual chance of survival is lessened. However, isolated one of the most common types of cranial damage among cranial elements vary in their likelihood of survival due immature individuals, as is loss of the vomer, nasals, to their structural or mechanical properties. Sturdier and zygomatic arches. Separation at the coronal suture bones (or bone parts, such as the petrous temporal, suit- seemed common in the Krovitz sample but was not spe- ably named for its rocklike properties) almost always survive in higher frequencies and with less damage than delicate bones (such as the nasals or vomer), thin bones Intermediate taphonomic vulnerability
with a complex shape (such as the sphenoid), or bones This grouping includes most of the neurocranial with projecting processes (such as the zygomatic).
landmarks and a mixture of facial and basicranial land- The Krovitz sample considered only fairly well marks (Table 2b). Because these landmarks exhibit the articulated crania; thus most individuals preserved the greatest variability, both within and between age groups, frontal bone, which explains the apparent low tapho- we suggest that the preservation of and damage to these nomic vulnerability of the landmarks on the frontal landmarks tends to refl ect particular differences in indi- bone. Individuals with a missing or badly broken fron- vidual growth and taphonomic history.
tal bone probably did not survive to be included in the Because the squamous temporal suture is bev- Krovitz sample. Because of its general robustness and its eled and not interdigitated, we expected that this suture many sutural attachments to other bones, the frontal has would be more likely to open and fall apart than other an unusually large effect on the taphonomic survival of neurocranial sutures. Contrary to our expectations, the landmarks along this suture (AST, PTN, and SPH) fell 214  Breathing Life into Fossils: Taphonomic Studies in Honor of C.K. (Bob) Brain
Application of taphonomic
landmarks change taphonomic vulnerability categories vulnerability categories
with increased age: the juncture of the palatine suture with the edges or curve of the palate (PAL); the top of The categories described above represent three lev- the zygomatic-maxillary suture at the inferior orbital rim els of taphonomic vulnerability. In general terms, know- (ZYS); the foramen ovale (FOV); and the incisive fora- ing which cranial parts are missing from a specimen can be used with the taphonomic vulnerability categories to The taphonomic vulnerability of PAL is intermedi- judge the intensity of the taphonomic damage to which a ate in the youngest individuals and steadily diminishes as age increases, until PAL eventually ranks among the Specimens showing only a few fractures or absences landmarks showing the lowest vulnerability in individu- in the high vulnerability category have probably been als aged 9-13 years. Between the ages of 0-13 years, the subjected to minimal taphonomic destruction. Those palate lengthens and strengthens considerably and, be- showing breaks in high and intermediate categories (but tween ages of 5-12 years, the fi rst and second permanent not necessarily the loss of all landmarks in those catego- molars erupt. We hypothesize that the presence of these ries) have been subjected to moderate taphonomic de- molars may partially shield the palatine suture from de- struction. Those crania showing some fractures to or loss struction in older individuals. A similar drop in vulner- of some landmarks in all three groupings have been sub- ability is shown in ZYS, which is in the intermediate jected to extensive taphonomic destruction. The catego- vulnerability group in the youngest group of individuals ries of taphonomic vulnerability are nested, so that spec- and drops to the low vulnerability group in individuals imens subjected to any level of taphonomic destruction older than 3 years. This change may be related to the have also, by defi nition, been subjected to the less severe increasing strength and buttressing in the face as adult- types of damage as well. Especially indicative of intense hood is approached. We can offer no hypotheses for the taphonomic destruction is the breakage and partial loss change in taphonomic vulnerability of FOV and ICF in of the orbital margin on the frontal bone or separation the different age groups other than individual variability of the frontal bone from other parts of the cranium. As argued above, once the frontal is absent, the cranium as a whole is less likely to be preserved in the archaeological Comparison of breakage in the
or fossil record. However, it is important to note that the Krovitz and SHARP samples
response of a cranium to taphonomic agents is highly variable according to the particular circumstances of The results of the Krovitz landmark inventory were burial. Further, these categories do not necessarily rep- compared with those of the SHARP bone inventory to resent three widely separated points in the time (relative see if these patterns of survival were consistent between to death) at which damage occurred. For example, ex- immature and adult crania. To calculate frequency of tensive damage can and does occur early in a specimen’s damage/absence in the SHARP sample, we added the taphonomic history while the bone is fresh and the cra- numbers of bones that were entirely missing to those in nium is fully fl eshed; conversely, minimal damage can which a signifi cant portion (25% or more) of the bone was missing (Table 3). As there are only three immature If the age at death of an immature cranium can be crania in the sample (Reid, 2004, pers. comm. to P.S.), determined with some precision, then a more refi ned we did not subdivide the sample into age groups.
deduction can be made using both age at death and ta- Generally, the frequency and location of cranial phonomic vulnerability data. Within the Krovitz sample, damage is similar between the two samples. The Krovitz most landmarks show decreasing vulnerability as age data showed that the category of high taphonomic vul- increases (Table 2), although landmarks rarely shift nerability included facial landmarks associated with the from one vulnerability category to another. Only four maxilla, nasals, zygomatic arches, and vomer. Although Table 3. Completeness of bones in the SHARP sample (N=82; 79 adults and 3 immature crania). Completeness data represent an average of lefts and rights for each bone. Bone abbreviations as follows: Front = frontal, Pariet = parietal, Occipit = occipital, Temp = temporal, Sphen = sphenoid, Zygo = zygomatic, Max = maxilla, and Pal = palatine. Krovitz and Shipman  215
data on the nasals and vomer were not available for the sample did not. If a large number of the sphenoids and SHARP sample, facial bones (zygomatics and maxillae) temporals contained in the SHARP sample were from were damaged or absent in a high number of individuals disarticulated crania then they would likely be less well (63% and 53% respectively). Thus, in both the immature preserved than those from the more articulated crania in sample inventoried by Krovitz and in the largely adult SHARP sample, aspects of the facial bones were among In summary, cranial breakage and survival in the im- the most highly vulnerable to taphonomic destruction. mature sample studied by Krovitz and the adult SHARP Landmarks in the category of lowest taphonomic vulner- sample show a generally similar pattern. There are im- ability, based on the Krovitz sample, were related to the portant exceptions pertaining to the survival of the oc- sturdy superior margin of the orbit and the crucial role cipital, the sphenoid, and the temporal bones.
of structural keystone that the frontal bone plays within the cranium. This is consistent with the observation that QUALITATIVE BREAKAGE PATTERNS IN
only 33% of individuals in the SHARP sample had seri- ously damaged or absent frontals, making the frontal the MMATURE FOSSIL CRANIA
second least vulnerable bone after the occipital.
Qualitative observations on patterns of fracture lo- The two samples differed strikingly in the vulner- cation and morphology provided additional tools with ability of the occipital bone. Landmarks in this region which to deduce the approximate timing of damage were among the most highly vulnerable in immature in an immature cranium. We examined some original individuals in the Krovitz sample. In the largely adult specimens but more usually photographs and casts of 20 SHARP sample, the occipital bone was the least vulner- fossilized immature hominid crania that were relatively able bone, being missing or seriously damaged in only complete and thus might be comparable to the Krovitz 32% of individuals. We hypothesize that this difference is sample (Appendix I). These data were compared with related to the adult nature of the SHARP sample. Despite similar observations on crania in the forensic and medi- considerable variability in the timing of the closure of cal literature (primarily representing damage to living or cranial sutures in adults (Todd and Lyon, 1924; Todd and recently dead individuals) and with archaeological spec- Lyon, 1925a, b, c; McKern and Stewart, 1957; Meindl imens from the inventory sample (representing primarily and Lovejoy, 1985), the lambdoid sutures in adults of the SHARP sample would have fused to some degree, and For each fossil specimen we observed, we noted the sutures between the basilar and lateral occipital elements general frequency of damage, the location of fractures, would be completely fused. Stronger bony attachments and the attributes (length, course, texture, and type) of between the occipital and other cranial bones would fractures and of the fractured surfaces. We used stan- make the occipital bone itself less likely to separate from dardized terminology from the forensic literature, where the rest of the cranium and less vulnerable to damage.
possible, to describe fracture type and morphology and Another marked difference in the survival of cranial to deduce the time of fracture relative to death. Appendix bones in these two samples involves the sphenoid and I summarizes our observations for each of the 20 imma- the temporal bones. In the Krovitz sample, landmarks as- sociated with the sphenoid and temporal bones fall into the intermediate category. In the SHARP sample, these Chronology of taphonomic damage
bones are highly vulnerable, showing serious damage or We divide the taphonomic history of an immature destruction in 62% and 55% of the individuals respec- cranium into three phases at which breakage can occur. tively. Thus the sphenoid is the second most vulnerable bone and the temporal is the third most vulnerable bone Stage 1) Wet or fresh bone breakage, incorporating in the SHARP sample. We hypothesize that the beveled antemortem and perimortem breakage, generally takes nature of the squamous temporal suture is a more impor- place while the bone is partially or wholly fl eshed. An- tant source of vulnerability in adults because the other temortem breakage is recognized by the fact that healing endocranial sutures are partially or wholly fused. In began before death; under controlled conditions, grossly contrast, in immature individuals, the squamous tempo- detectible healing may be evident in as little as one to ral suture is not distinctly more vulnerable to separation two weeks after the time of injury (Sauer, 1998; Gal- than the other cranial sutures. Similarly, the sphenoid is loway, 1999: 15, citing Murphy et al., 1990 and Rogers, relatively more vulnerable in adults than in immature in- 1992). Identifying perimortem breakage (occurring at dividuals. The sphenoid is a very thin and fragile bone the time of death) can be less straightforward than an- in adults compared to the other vault bones, which have temortem damage, but still incorporates fresh bone frac- thickened and acquired greater robustness with age; in ture patterns (Sauer, 1998). Although antemortem and immature crania, the sphenoid is not so markedly dif- perimortem fractures can be differentiated by evidence ferent in robustness from the other cranial bones. This of healing, they both occur on wet, fresh bone and both difference could also be due to sampling differences be- have similar fracture characteristics; therefore, they are tween the Krovitz and SHARP samples, as the SHARP sample included disarticulated bones and the Krovitz 216  Breathing Life into Fossils: Taphonomic Studies in Honor of C.K. (Bob) Brain
Fresh or living bone is a composite tissue comprised tissue left and behaves more like a brittle, inorganic ma- of fl exible protein (mostly collagen) and brittle hydroxy- terial, such as stone or ceramic, than like fresh bone. We apatite. In both antemortem and perimortem breakage, do not know of anyone who has measured of mechanical the bony tissue and the sutures between bones con- strength of fossilized bone; strength would presumably tain intact collagen and other organic components that vary with the degree of mineralization. As a crude ap- give bone its elasticity, meaning that it is able to bend proximation, we might expect fossilized bone to have the or deform under load before failure (breaking) occurs. very low tensile strength, very low modulus of elasticity, The fl exible collagen and membranes surrounding un- and high compressive strength of stony substances such fused sutures stop cracks from propagating through the bony tissue by deforming and dissipating force (Currey, Another issue in breakage is microstructure. Living bone is riddled with osteons and other spaces that house The mechanical properties of one square inch sam- bone cells; these “holes” can and do serve as the sites of ples of fresh tissue from long bones has been measured. fracture origin because they are inherently weak places In tension, such a sample of whole fresh bone fractures or in the tissue. Similarly, foramina act as stress concentra- fails at only about two-thirds the pressure that is required tors where fractures often initiate (Currey, 1984). In fos- to fracture bone under compression (Gordon, 1968: 42- silized bone, these “holes” are more or less completely 44). This is why, in fresh bone subjected to blows, frac- fi lled with mineral. The increased strength caused by the tures are initiated not at the point of impact where the absence of holes seems to be more than offset by the bone is compressed but in the surrounding bone which is placed under tension (Rogers, 1992). Whole, fresh bone Both the diminished fl exibility and the geometry or has a low modulus of elasticity, which means it has a morphology of the bone(s) assumes greater importance tendency to bend without breaking (Gordon, 1968: 42- in determining breakage in stages 2 and 3 than in stage 1. 44). Younger individuals with bones that are more car- It is also important to appreciate that breakage of stage 1 tilaginous and less mineralized will have an even lower crania is more likely to be the result of accident or human violence than breakage in older crania. Fractures caused Stage 2) Dry bone breakage occurs postmortem af- by violence are more likely to be directed at the face than ter the bone has lost much of its organic content, and fractures caused in other ways during stages 2 or 3.
is usually defl eshed, although dried or mummifi ed fl esh can be found on stage 2 crania. Dry bone breakage may Review of the forensic and medical
occur during a lengthy time span ranging from shortly terminology: fracture types
after death to centuries later, depending on specifi c pres- Six basic types of cranial fractures, differentiated on ervation conditions; the organic component of bone will the basis of location and morphology, are frequently dis- also vary according to the time since death and preserva- cussed in the medical and forensic literature (e.g., Gallo- tion conditions. The archaeological sample we invento- way, 1999; Byers, 2000; Richardson, 2000). Three com- ried had been subjected primarily or exclusively to stage bined fracture patterns are frequently identifi ed in the face (i.e., LeFort fractures), and disarticulation of cranial Immature crania in stage 2 are substantially more sutures is also noted. These types of cranial damage are vulnerable to separation along unfused sutures than stage reviewed here with particular focus on immature crania.
1 bones since the connective tissue holding the sutures (1) Linear fractures are elongated, single breaks that together is degraded or decayed in stage 2. Therefore, the go through the outer table of the bone, the diploe, and the modulus of elasticity of the bone tissue is compromised inner table of the cranial vault bones. Linear fractures as is the crack-stopping ability of the fl exible compo- comprise 70-80% of observed fractures in the forensic nents of bone. Thus breakage will occur in stage 2 bones context (Gurdjian, 1975; Rogers, 1992), which usually at lower loads than in stage 1 bones (Lyman, 1994; Gal- involves stage 1 breakage. Linear fractures are often the loway, 1999). The loss of elastic tissue from the bone result of impact with objects having a large mass, such as not only lowers the force needed to produce failure in heavy weapons or automobiles. In forensic cases, linear bone but also alters the morphology of the resulting frac- fractures occur less commonly in children than in adults ture. As elastic tissue degrades, the fracture surface be- due to the greater elasticity of immature bone (Duncan, comes progressively fl atter, more planar, and less likely 1993) but are known to occur in cases of child abuse, es- to splinter or bevel. The course of vault fractures is more pecially in children under the age of three years (Naim- likely to be curvilinear and longer in fresher crania and Ur-Rahman et al., 1994). In fresh crania, linear fractures occur as a result of forces between 450 to 750 psi (Cox Stage 3) Post-fossilization breakage occurs after the et al., 1987) although there is considerable individual bone has been mineralized, but fossilization is not an all- or-nothing event. Bones at a single site may range from (2) Diastatic fractures are linear fractures that fol- fully mineralized to a condition close to that of dry bones low the course of sutures, fused or unfused, in stage 1 with only minimal geochemical changes. What typifi es crania. In antemortem or perimortem circumstances, dia- this stage is that the bone has no signifi cant elastic or soft static fractures cause traumatic interruptions of sutures, Krovitz and Shipman  217
sometimes leading to the springing outward of the vault (Tillier, 1999), but this does not appear to be a ping-pong bone on one side of the fracture and the evulsion of brain fracture and the individual is older than those who typi- tissue through the crack. This springing out is a result of cally incur such fractures. Close-up photographs in Til- the release of the inherent tension of the intact cranial lier (1999) support her suggestion that some healing had vault by a fracture. This phenomenon is very unlikely to occurred at the time of death, proving that this fracture occur in dry or fossilized crania because elasticity of the bones is so diminished that fracture is more likely to oc- (4) Stellate fractures are a set of linear fractures cur than a rebounding outward of part of the cranium. radiating in a star-shaped pattern from a single point In the forensic context, diastatic fractures consti- where impact occurred. Gurdjian (1975) found that stel- tute about 5% of all fractures and occur most commonly late fractures are typical of heavy loads of relatively low in the coronal and lambdoid sutures (Galloway, 1999). velocity on stage 1 crania and are somewhat more com- Blount (1955, 1977) observed that true diastatic frac- mon on upper parietals than elsewhere. Where stellate tures are rare in children in stage 1. He postulated that fractures are centered on a depressed fracture, they are linear fractures made on fresh or wet bone very rarely functionally identical to radiating fractures. cross a suture because the area of the suture has differ- Although radiating fractures do not usually occur ent mechanical properties from the bone surrounding it. later than stage 1, we observed stellate fractures with- The greater fl exibility of connective tissue in and near out depressed fractures on crania in stages 2 and 3. We sutures in immature crania acts to stop cracks by dis- inferred that these were caused by the slow crushing or fl attening of curved bones or parts of bones probably (3) Depressed fractures, together with comminuted under sedimentary load. Examples of stellate fractures and stellate fractures (discussed below), comprise 15% centered at inion but not involving a depressed fracture, of fractures in forensic contexts (Gurdjian, 1975). De- which probably occurred in stage 2 or 3, can be seen on pressed fractures involve deformation of the cranial vault the immature fossil crania from Engis, Pech de l’Azé, or in response to impact, usually of high velocity by a blunt object of small or moderate diameter. The bone at the (5) Communited fractures of the vault involve large point of impact is pushed inward while the area immedi- numbers of small fragments, usually produced by low ately surrounding the impact is bent outward, placing the velocity/heavy impact force. Crushing incidents are one bone under tension and initiating fractures (Gurdjian et common cause of comminuted fractures of fresh (stage al., 1953; Gurdjian, 1975; Rogers, 1992). The fragments 1) crania. We suggest that comminuted fractures may of bone pushed inward by the impact in stage 1 remain also result from sedimentary pressures acting on stage 2 attached to the cranium; Byers (2002) refers to this phe- or 3 (dry or fossilized) crania. The fossil crania Dederi- nomenon as hinging and regards it as diagnostic of stage yeh 1 and 2, Qafzeh 11, and KNM-WT 15000, among 1 breakage from depressed fractures. We have observed others, exemplify an overall comminution of the cranial a specimen that received a depressed fracture in stage 1 vault (which we call a mosaic fracture pattern, see be- with the fragments still in place some 200 years later, low). We discuss below ways in which stage 1 and stage well after the cranium had reached stage 2 (G. Milner, 2 comminuted fractures may be distinguished. personal communication to authors, 2003). Depending (6) The tripod or zygomatic-maxillary fracture on the strength of the blow, a depressed fracture may is one of the most common cranial fractures observed be surrounded by a number of linear breaks that radi- in medical and forensic circumstances (Rogers, 1992; ate outward from the depressed area known as radiating Richardson, 2000). Frequently caused by a blow to the fractures. A depressed fracture surrounded by radiating malar eminence, the tripod fracture separates the zygo- fractures is one typical result of blunt force trauma un- matic bone from the rest of the cranium by breaks in the der stage 1 conditions. Radiating fractures are unlikely zygomatic arch, at or near the zygomatic-maxillary su- to occur in dry stage 2 bone (Byers, 2002: 270) or in ture, and at or near the zygomatic-frontal suture (Rogers, Depressed fractures are 3.5 times more common in In addition to the six types of cranial fracture de- stage 1 children than in stage 1 adults (Zimmerman and scribed above, forensic experts distinguish three com- Bilaniuk, 1981). Even though an immature cranium is bined fracture patterns involving the face called LeFort more fl exible than an adult’s, the absolute thinness of the fractures (see Figure 4) (Galloway, 1999; Byers, 2002). cranial vault bones makes immature crania more prone These fractures may occur in combination as well as to fracture. Sometimes depressed fractures occurring in separately. A LeFort I fracture is an approximately hori- young individuals do not break through both inner and zontal break above the alveolar processes of the maxillae outer bony tables of the vault but may simply dimple the and below the nasal aperture. The typical cause of a Le- surface; this stage 1 phenomenon is known as a ping- Fort I fracture is a blow to the lower face from the front pong fracture because similar depressions occur on or side. A LeFort II fracture isolates the midface from the ping-pong balls. Among the immature fossil crania we vault, with breakage passing through the maxilla, the in- examined through photographs, Qafzeh 11 shows a clear fraorbital foramen, and nasion; these fractures typically depressed fracture on the frontal just above the left orbit result from a blow to the midface at midline. A LeFort III 218  Breathing Life into Fossils: Taphonomic Studies in Honor of C.K. (Bob) Brain
Figure 4: Illustration of LeFort fractures (after Byers, 2002, and Galloway, 1999). fracture passes through nasion, through the bony orbits sutural membranes themselves. Although diastatic frac- below the superciliary ridge, back through the sphenoid tures are breaks that follow the course of a suture, they and zygomatic arch. This fracture separates the cranium differ from sutural separations in that the application of into a skullcap and a facial portion and occurs in forensic force is necessary to produce a diastatic fracture; while circumstances when a blow is struck to the upper central sutural separations occur naturally as a by-product of de- compositional changes in an immature cranium. Sutural In the Krovitz sample, 9% of the crania showed separations occur in stage 2 rather than stage 1 unless breaks that appeared to be LeFort III fractures, i.e., the there is some processing of the fresh cranium (such as specimens were neurocrania without faces. Eight per- cooking or chemical treatment) that destroys or degrades cent of the SHARP sample showed LeFort III fractures, the organic component of the bone. Sutural separations as judged by the simultaneous absence of right and left yield cranial fragments that may be entirely unbroken maxillae, zygomatics, and palatines. However, because but which are separated from the rest of the cranium. The we did not have data on isolated cranial bones (those not vast majority of crania from individuals who die under associated with skeletons) in the SHARP sample, the the age of one year (and many older immature individu- actual frequency of LeFort III fractures may have been als as well) will come apart along sutural lines once the collagen and membrane at the sutures has broken down Breakage that appears as a LeFort III fracture seems (i.e., once stage 2 is reached). However, since the exposed to be relatively common among fossil hominids. Numer- sutural edges involve many small, protruding points that ous adult crania of Homo erectus show this pattern as are vulnerable to further breakage, not all cranial bones do immature crania including Engis 2, Subalyuk 2, Mo- isolated by sutural separation will be complete. Close jokerto 1, and Skhul 1. We hypothesize that during burial inspection of the sutural edge should reveal whether the and taphonomic destruction, crania with LeFort I and II breakage is a single diastatic fracture or a series of frac- fractures (or other facial damage) are so weakened that tures subsequent to a sutural separation. they tend to deteriorate further and present as LeFort III The Taung 1 cranium shows a separation of the fractures by the time of excavation and study. It is un- coronal suture, with subsequent minor breakage of the common to fi nd fossil crania with LeFort I or II types of frontal. Gibraltar 2, the Devil’s Tower Neandertal, also experienced a separation of the coronal, sagittal, and Finally, sutural separations are another type of cra- temporal sutures; subsequent to that separation, there nial “damage” that is commonly observed in immature has been some breakage of the exposed sutural edges. crania. A sutural separation involves the falling apart of The Amud 7 specimen preserves the occipital bone of a an unfused or open suture due to the loss or degrada- young Neandertal which was isolated by sutural separa- tion of the collagen component of bone tissue and of the tion; the rest of the cranium was not recovered.
Krovitz and Shipman  219
In summary, six types of fracture, three patterns of Burial per se may protect the cranium from further facial fracture, and sutural separations are distinguished damage by many taphonomic agents. Although crania in the forensic and medical literature. Many but not all buried in stage 1 may undergo non-traumatic sutural of these were recognized in the archaeological or fossil separations, especially on very young (<1 year old) in- dividuals, it is important to remember that these sutural separations do not normally occur at the time of burial Infl uence of chronological stage
but after burial, during stage 2, when the organic com- on fracture morphology
ponents of the cranial bone have decayed. Stage 1 crania do not generally show sutural separations unless there Stage 1
is some processing of the fresh cranium that destroys or degrades the organic component of the bone.
To assess stage 1 fractures, we turned again to the In summary, stage 1 fractures are usually few in medical and forensic literature. Richardson (2000), a number per specimen, except in cases of warfare, vio- radiologist, reports the frequency of different fractures lent attack, or accidents. Stage 1 depressed fractures fre- for a hospital population. Less than 10% of the injuries quently show a rounded outline that refl ects the shape in his sample were incurred by children. Most probably, of the object responsible for the damage; depressed his sample was biased toward adults and toward young fractures often show hinged pieces or retention of the males, since in many medical reports young males are pushed-in fragments. Linear fractures occur and are found to sustain a higher frequency of facial fractures elongated and curving in course; where they intersect than other age and sex categories (Barker et al., 2003). sutures, the course may change abruptly to follow the In Richardson’s sample, automobile accidents and as- plane of weakness represented by the suture, becoming saults were the two most common causes of the facial a diastatic fracture. Stage 1 breaks are often beveled and the fracture surface itself will show irregularities caused Richardson found that the most common type of by microscopic variations in the amount of elastic tissue midfacial fracture is a tripod fracture of the zygomatic in the bone. The fracture itself is sharp-edged and ap- and maxilla (40%). LeFort I fractures comprise 15%; Le- pears crisp and cleanly defi ned. Sutural separations are Fort II, III, simple zygomatic arch fractures, and commi- nuted fractures each comprise another 10%; and alveolar fractures, often associated with fractured teeth, make up Stage 2
the last 5%. Richardson points out that 60-70% of all Breakage occurring after the bony tissue has dried facial fractures (the tripod fractures plus LeFort II and and its organic component has decayed differs from that in stage 1 because the material properties of the bone Any of the cranial fractures described above may oc- tissue have changed (as discussed above). Generally, the cur on stage 1 crania, with the exception of sutural sep- edges of fragments fractured during stage 2 are fl atter and arations (see discussion on sutural separations above). more planar than those resulting from stage 1 fractures, Since most stage 1 crania do not sustain trauma prior to indicating that the bone is responding more uniformly to death, Richardson’s sample is not expected to parallel pressure since it is no longer elastic. Stage 2 fractures of the incidence of facial fractures in a general cemetery the cranial vault may or may not be beveled. population. Prior to the invention of mechanized forms Synthesizing our original observations of immature of transport, an individual cranium probably experienced archaeological and fossil crania with those reported in few breaks during stage 1 except in cases of attack, war- the literature, we observed meaningful differences in the fare, or a major accident. Even then, the point or points general frequency of the different fracture types on cra- of impact may be evident. The effects and sequence of a nia in different temporal stages. Long, linear fractures series of blows or impacts can be deduced by a skilled with a curving course across the cranial vault are rare examiner in forensic or archaeological cases (e.g., Sauer, in crania that were dry (stage 2 or 3) when broken, as are true diastatic fractures. In contrast, sutural separa- The morphology of fractures is helpful in deduc- tions commonly occur in immature crania during stage ing the chronological stage of the cranium at the time of 2, judging from the Krovitz sample. Fully or partially breakage. Linear fractures in stage 1 crania are usually separated sutures, especially interdigitated ones, are very long and curved rather than straight; their course relates vulnerable to further breakage because of the irregular- more to the direction and magnitude of force applied ity of the exposed edge. One common consequence of than to the geometry of the cranium per se. The outline this post-separation breakage is the loss of an angular of a depressed fracture is also often curved into a round- piece of vault bone at bregma, as in the specimen from ed or oval shape, refl ecting the shape of the object which Le Figuier or Qafzeh 11. Another is the loss of small impacted the cranium. Concentric rings of fracture may fragments along the course of separated coronal, sagittal, encircle a depressed fracture. Stage 1 fractures of crania or lambdoid sutures, such as in the Mojokerto specimen. are often beveled, with the direction of the bevel indicat- The lack of a linear course in such sutural fragments dis- ing the direction of movement of the force causing the tinguishes them from true diastatic fractures. 220  Breathing Life into Fossils: Taphonomic Studies in Honor of C.K. (Bob) Brain
Dry crania (stage 2) show fewer discrete depressed involved in the orbital rim. The frequency of orbital frac- fractures than wet ones (stage 1). Oval or curving holes tures in this sample can be assessed only from data on where pieces are missing—holes which might represent the completeness of the frontal and maxilla. Twenty-two depressed fractures that have lost their small fragments— percent of the SHARP individuals show breakage to are uncommon in stage 2 breakage. An exemplar might some part of the frontal bone and an additional 11% are be the rounded hole on the right side of the Roc de Mar- missing their frontal bones. Twenty-seven percent of the sal cranium. However, if this was originally a depressed SHARP individuals show maxillary breakage, with an fracture, there is no hinging nor are the depressed frag- additional 26% percent missing the maxillae entirely. ments still attached. Holes in stage 2 are much more Pure LeFort I fractures, which comprised 15% of commonly geometric or angular in outline. Comminuted Richardson’s sample, were not observed in the Krovitz fractures in stage 2 crania may separate into fragments. sample and LeFort II fractures appeared to be rare, sug- Their recovery and recognition depends largely on exca- gesting that both types probably progressed to LeFort III vation and curation techniques. Cases where large num- fractures by the time they were observed in archaeologi- bers of fragments have been reassembled into partial fos- cal or fossil samples. In other words, specimens missing sil crania include Dederiyeh 1 and 2, KNM-WT 15000, the landmarks near the maxillary teeth were invariably also missing much more of the face. LeFort III fractures A comparison of the frequency of various fracture occurred in 9% of the Krovitz specimens versus 10% in types in stage 1 and stage 2 crania is instructive. It is Richardson’s sample. Our inventory data show that the important to remember, however, that breaks occurring nasal sutures and the zygomatic arches are especially in stage 1 crania are likely to weaken the specimen and likely to open or break in dry immature crania. Loss of make it more vulnerable to further breakage. This means both of these regions would encourage the separation of that the initial fractures may be obscured by later dam- the face and vault in a LeFort III pattern. The number of age, and if stage 1 damage is signifi cant, the specimens LeFort I and II fractures could not be readily estimated may not survive to be examined as archaeological (stage from the SHARP data. LeFort III fractures, as judged by the simultaneous absence of maxillae, palatines, and zy- The frequency of tripod and zygomatic arch fractures gomatics, apparently occurred in 8.5% (N=7 out of 82) (together 50%) in stage 1 crania reported by Richardson of the SHARP sample. Among the immature fossilized matches closely with the high frequency of fractures to crania, there was only one LeFort I fracture (Herto BOU- the zygomatic arch, 40-47% across all age groups in the VP-16/5). LeFort II or III fractures occurred in 8 of 20 Krovitz stage 2 crania (Table 2a). The zygomatic frac- (40%) of the specimens. Because the original specimens tures observed in the Krovitz sample were not sutural were not examined in most cases and the sample size separations but fractures of the zygomatic arch. In the is small, we do not know if this difference between the SHARP sample, 27% of the individuals showed zygo- archaeological and fossil samples is meaningful. matic breakage and another 37% were missing the zygo- In the Krovitz sample, common sutural separations in stage 2 crania occur at the basilar suture (37-64% of In contrast, the high frequency of orbital fractures the specimens according to age) and, less frequently, (60-70%) cited by Richardson is not paralleled by the along the lambdoid suture. Separation along the basilar frequency of orbital fractures in stage 2 crania. Richard- suture contributes to further breakage of the basicrani- son does not specify where the orbits are fractured, sim- um; separation along the basilar and lambdoid sutures ply that fractures involving the orbit are very common results in the isolation of the occipital from the rest of the in patients seeking medical attention for cranial trauma. cranium. Separation at the coronal suture occurred, sepa- In the Krovitz sample, four landmarks (ORB, FZJ, ZYS, rating the face and frontal bone from the rest of the cra- and FMO) refl ect damage to the superior, lateral, in- nium in 38 specimens (14% of the Krovitz sample); note ferior, and medial parts of the orbital rim respectively. that in these individuals the rest of the cranium did not Three of these landmarks (ORB, FZJ, and FMO) are survive intact once the face and frontal bones separated. so rarely missing in the Krovitz sample that they are in Other observed (though not quantifi ed) sutural separa- the low taphonomic vulnerability category. The fourth tions include the zygomatic-temporal suture (contribut- landmark, ZYS, falls into the intermediate taphonomic ing to zygomatic arch breakage), zygomatic-maxillary vulnerability category, being missing in 18% of the 272 suture (contributing to loss of the maxilla or zygomatic), specimens in the landmark sample. Clearly, then, either and springing out of the beveled parietal-temporal suture orbital fractures are substantially less common in stage 2 (contributing to loss of the temporal bone, especially if than in stage 1 immature crania or, because of the impor- the occipital bone is also missing).
tant structural role the frontal bone plays in protecting We observed two new forms of linear fracture regu- crania from breakage (as discussed above), crania with larly in stage 2 crania although we did not quantify their frontal or orbital fractures that occurred during stage 1 occurrence. We call the fi rst of these a temporal line did not survive to be inventoried in stage 2.
fracture. This fracture differs from an elongated, curving This conclusion is supported by the SHARP sample linear fracture by its anatomical placement. A true linear data which do not show high levels of breakage on bones fracture occurring on a stage 1 cranium curves across the Krovitz and Shipman  221
vault of the cranium with no characteristic placement. A ditionally, the cause might be due to the change in cur- temporal line fracture is found on one or both sides of a vature which becomes more acute at or near the midline cranium or skullcap that has been subjected to a compres- of the frontal bone, thus rendering this region especially sive force after burial and drying. Dry crania can be con- vulnerable to breakage. The Le Figuier and Qafzeh 10 sidered ovoids that may be hollow, incompletely fi lled specimens show metopic or parametopic breaks, which with sediments, or fi lled with unconsolidated sediments. are distinct from patent metopic sutures such as in Pech Slow compression, such as the weight of accumulating sediments overlying the cranium, will tend to fl atten the During our reading of descriptions of fossil crania, cranial vault from side to side. Such compression will we noted a disturbing tendency for any fracture along produce a linear fracture that follows the approximate the midline of the frontal bone to be described as a pat- course of the temporal line on one or both sides of the ent metopic suture, even when the edge of the break was specimen. It is not the location of this muscle marking planar and not interdigitated. Similarly, missing frag- but the more acute curvature of the cranial vault in this ments at bregma were sometimes labeled as patent ante- region that renders it especially vulnerable to fracture. rior fontanelles without anatomical evidence. Great cau- This acuteness of curvature is especially evident in pos- tion is needed in concluding that missing bone indicates terior view, where modern human crania show parietal a natural anatomical consequence of immaturity rather bosses or the typical “en maison” shape. Temporal line fractures typically pass from the superior margin of the A stellate fracture of the squamous occipital, cen- orbit (or from the coronal suture) through the parietal and tering on inion, occurred in a number of fossil crania al- stop when the fracture encounters the lambdoid suture, though the bone tissue is thicker at inion. This pattern of as in the La Quina H18 and Teshik Tash specimens.
breakage is common on stage 2 specimens. The cause of The second new type of fracture we observed in such breaks would seem to be the geometry of the oc- stage 2 crania is a perpendicular fracture in our termi- cipital, which is effectively a very blunt cone the point of nology. These fractures run perpendicular to the sagittal which is at inion. Virtually any diffuse pressure on such suture inferiorly from that suture until they encounter the a structure will tend to fl atten the cone, producing a stel- temporal suture or a temporal line fracture. Single speci- late fracture at inion, as in Engis 2 and Amud 7. mens often show two or more perpendicular fractures, We may make some broad generalizations in com- which are the natural result of diffuse pressure applied to paring stage 1 and stage 2 fractures. Breakage in stage 2 the ovoid cranium. Perpendicular fractures are common typically involves the orbit much less often than in stage in archaeological specimens (G. Milner, 2004, personal 1, and in stage 2 the lateral or inferior orbital margins are communication to G.K. and P.S.). Longitudinal bending more often damaged than the superior margin. Temporal stress tends to fl atten the curvature toward the front and line, perpendicular, and metopic/parametopic fractures back of the cranium, causing the perpendicular fractures, are typical stage 2 breaks. Sedimentary pressure during as in the Teshik Tash, Engis 2, Grotte des Enfants 6, and stage 2 may cause a widespread mosaic fracture pattern comprised of numerous geometric fragments lacking a The intersection of temporal line and perpendicular discrete area of impact, as distinct from a circumscribed fractures effectively breaks the parietal into large, rough- area with a comminuted or depressed fracture, which is ly rectangular or trapezoidal fragments. We hypothesize more typical of stage 1. The surfaces of stage 2 fractures that continued sedimentary pressure (or some other type are usually planar with blunt edges lacking beveling or of diffuse compressive load) will break these rectangular hinging. The course of stage 2 fractures is not curved or fragments further until most or all of the cranial vault rounded. Immature crania damaged during stage 2 sus- surface is broken into triangular or irregularly geometric tain more breaks per specimen and frequently show non- fragments in what we call a mosaic fracture pattern. This traumatic sutural separations. Overall, stage 2 breaks pattern differs from a comminuted fracture in that the rarely show a discrete point of impact and instead result mosaic fracture pattern covers a large area of the cranial vault and has no clear outline or point of impact. The mosaic fracture pattern can be observed in KNM-WT Stage 3
15000, Mojokerto, Herto BOU-VP-16/5, Dederiyeh 1 Post-fossilization breakage occurs after the bone has and 2, Subalyuk 2, and Qafzeh 10 and 11. been mineralized. Though rarely of concern in forensics, Another new type of fracture we saw in stage 2 cra- post-fossilization breakage is important to paleoanthro- nia is the metopic or parametopic fracture. Younger in- pologists as it may yield clues to the circumstances under dividuals, under the age of about four years, may show which a hominid died and became buried in sediments. a vertical fracture of the frontal bone either along the In this stage, the cranium acts very much like a ceramic metopic suture or parallel to it. Although Cobain et al. vessel of similar shape. The primary infl uences on break- (2002) report that the metopic suture is fused in most age of a stage 3 cranium seem to be the geometry of the individuals by the time of birth, the site of the former cranium or parts of the cranium in question, as in stage suture may be weaker than the adjacent bone over the 2, and the lack of bony elasticity. Bone density per se orbits, which is reinforced by the superciliary ridge. Ad- is a less important issue in post-fossilization breakage 222  Breathing Life into Fossils: Taphonomic Studies in Honor of C.K. (Bob) Brain
than in earlier breakage because stage 3 bone tissue is age than the surrounding rock, the endocast helps the cranium resist fl attening with the result that the vault Determining the timing of postmortem fractures bone shatters but the pieces are not destroyed. A partial relative to the time of death can be diffi cult; we are not endocast, as in Taung 1 or Skhul 1, obviously leaves the confi dent of our ability to distinguish between stage 2 portion of the cranium without the endocast extremely and stage 3 breakage in most cases. Stage 3 breakage vulnerable to fragmentation and loss.
may reveal internal surfaces of the fossil that are dif-ferent (often lighter) in color than those that have been THREE CASE STUDIES
exposed longer; thus fresh breaks on fossils are often readily recognizable. Some fossil specimens we exam- ined showed mosaic fracture patterns generally similar to those produced in stage 2 with many discrete, short, Taung 1, the type specimen of Australopithecus afri- and straight fractures yielding numerous geometric frag- canus, was collected during quarrying of a South African ments. These specimens generally lacked either a long limestone cave and was recognized by Raymond Dart as temporal fracture or perpendicular fracture, which we a previously unknown species of hominid (Dart, 1925). believe are more common in stage 2 crania. We hypothe- The specimen is that of a young individual about 3-4 size that crania subjected primarily to stage 3 damage are years old (Bromage, 1985); only the partial cranium and more likely to show an overall diffuse shattering rather mandible have been recovered. There are no other homi- than the creation of longer, more linear fractures along nid remains to date from the site. Subsequent studies of geometric planes of weakness. As a subjective impres- the remaining portions of the Taung deposits and similar sion, we believe that the average fragment size is some- caves nearby suggest a complex taphonomic history for what smaller in stage 3 mosaic fracture patterns than in the bones preserved in those caves. It is most probable stage 2 patterns, but it will be diffi cult to test this because that the Taung 1 skull was washed or dropped into a cave we cannot reliably separate stage 2 and 3 breakage. Frag- in the tufa by a leopard or other mammalian predator, ment edges, where visible, are generally planar in both while the skull was both fresh and fl eshed (Brain, 1981; McKee and Tobias, 1994; McKee, 2001; McKee, 2004, Another taphonomic factor that has a considerable personal communication to P.S.). An alternative interpre- infl uence upon the breakage of stage 3 crania is the tation made by Berger and Clarke (1995), that the Taung presence or absence of a consolidated natural endocast individual was preyed upon by a large avian raptor, is that was formed while the cranium or a portion of the less likely (McKee 2001). Whatever the precise cause of cranium was still intact. A cranium that is largely intact death, the skull became a sedimentary particle while it when it becomes a sedimentary particle in an open air or was largely or completely intact and the mandible was cave site is likely to become fi lled with sediment which will eventually harden into an endocast. At least some The Taung skull is one of those rare specimens with deliberately buried crania also develop natural endocasts a natural endocast, which preserves the impression of the (i.e., Skhul 1, see McCown and Keith 1939: 299-301). entire right side of the cranial vault and occiput although Because these sedimentary infi llings obscure the internal those parts of the cranium are now missing. The endocast surfaces of the fossilized bones, they are usually removed did not fi ll the skull completely and does not preserve the during preparation. The exceptions are cases where the left side of the cranium, which was not recovered. The sedimentary infi lling (endocast) preserves an impression face, frontal bone, and mandible of the individual were of bones which were not recovered or where the endo- intact and in anatomical position when found. Because cast separates naturally from the fossilized bone without the venous markings and the sulci and gyri of the inte- rior surface of the right side of the braincase are clearly We have few examples to generalize from in which preserved on the endocast, it is obvious that the skeletal endocasts have been recovered intact or where their ex- elements of this side of the cranium were also present tent and placement at the time of discovery is document- in the rock. The vault fragments from the right side and ed. However it is intuitively obvious that the presence occiput were either destroyed by the blast that exposed and nature of a consolidated natural endocast effectively the skull or were not collected by the workmen who re- transforms a fossilized cranium from a hollow ovoid, structurally similar to a ceramic vessel in terms of poten- Remarkably, the face and frontal bone show no tial for breakage, into a solid comprised of a dense center weathering and no fractures; very fragile regions of the (the endocast) and a relatively thin outer covering (the skull with high taphonomic vulnerability are preserved fossilized bone). We suggest that a consolidated, natural (such as the nasal bones and zygomatic arch). The coro- endocast will not necessarily prevent fragmentation of nal suture has separated neatly and there is only minimal a fossilized cranium as pressure is exerted but will act additional breakage on the frontal. The right zygomatic to keep fragments together and in or near their original arch is intact. The mandible was found in place, attached anatomical position. Although a fossilized cranium with to the maxilla by sediment (Dart, 1925). This is strong a complete endocast may be more vulnerable to break- evidence that the cranium and mandible were deposited Krovitz and Shipman  223
shortly after death while the bone was in stage 1 condi- tures did not separate. Most of the small fractures along tion and held together by soft tissue. Burial in alkaline the sutures have beveled edges with the inner table being sediments and the partial infi lling of the skull protected more extensive; they may represent bending and fractur- it from further damage until the specimen was exposed ing of the specimen in situ that caused small fragments to separate from the cranium. Alternatively, it is possible There is one area where a small fragment, probably that this damage occurred during excavation or prepa- of parietal, was pushed into the then-unconsolidated ration, procedures which are not well documented. The endocast, to which the fragment still adheres although cranium shows a possible temporal line fracture, several the surrounding bone is missing. This damage could not perpendicular fractures on the cranial vault and a stellate have occurred when the specimen was fresh or the frag- fracture at inion. A number of fragments of the cranial ment would have been pushed into the brain tissue and vault have beveled edges, suggesting that these fractures would not now adhere to the endocast (McKee, 2001). A occurred before all organic tissue and fl exibility of the pointed rock or other object probably caused this small bones was lost. Most of the occipital portion of the basi- fracture before the endocast was fully consolidated. cranium is missing although the (damaged) petrous por- We conclude that the Taung skull came into the ancient tufa cave during stage 1 when its fl esh was still Once buried, the Mojokerto skullcap fi lled with sedi- intact, as has been proposed before (e.g., Brain, 1981; ment which became a natural endocast. Venous markings McKee and Tobias, 1994). A natural endocast was are visible on the better preserved (and exposed) left side formed. The specimen was subjected to minimal tapho- of this endocast, showing that additional vault fragments nomic destruction thereafter except for sutural separa- were present in the rock. If these fragments survived tion during stage 2. Possible crushing or destruction of until the moment of discovery, they were unfortunately left side of cranium occurred during late stage 2 or early not collected. It is important to note that the fossil was stage 3, after drying of the cranium and before consoli- collected as an aid in geologic mapping and biostratigra- phy, not for paleontological studies (Duyfjes, 1936). An alternative interpretation is based upon the fact that at Mojokerto (Perning 1)
least one credible report of the discovery of the cranium Mojokerto is an immature Homo erectus specimen mentions that there were fossil fragments lying on the (Anton, 1997), approximately 4-6 years old, that was surface, which prompted Andoyo to excavate there and discovered in Java in 1936 by Andoyo, an Indonesian discover the cranium (Duyfjes, 1936). Possibly the now- geological assistant (Duyfjes, 1936; von Koenigswald, missing fragments of the left side of the cranial vault 1936a, b). The specimen was deposited in fl uvial sedi- were the surface fragments seen by Andoyo and presum- ments (Huffman, 2001; Huffman and Zaim, 2003). To ably judged too small to be useful. If so, the discovery our knowledge, no one has attempted to reconstruct the occurred after these pieces became separated from the taphonomic history of the Mojokerto skull as a bony rest of the fossilized specimen but before weathering and specimen, although its taphonomic history as a geologi- erosion could destroy the impression of interior surface cal and sedimentary particle has been discussed (Huff- of the parietal and temporal fragments on the endocast. Several pieces of bone from the right side of the The Mojokerto cranium appears to have suffered a vault and from the occipital bone are pushed sharply LeFort III fracture; the face and much of the basicranium into the endocast, which is not complete in this area, and of the specimen was lost or destroyed. Transport of the there are sizeable areas where there is no preserved bone specimen after this fracture may have occurred but was at all but only endocast. The placement of bevels and probably not extensive, judging from the preservation of pushed-in fragments suggests that, prior to the complete fragile edges of the broken right parietal, the occipital, consolidation of the endocast, sedimentary pressure pro- and the frontal where it articulates with the ethmoid. Von duced numerous fractures and forced some of the result- Koenigswald (1936a) perhaps overstated the fragility of the parts that remain intact, writing: “It is in fact a mira- The Mojokerto cranium is subtly but markedly cle that such a fragile object has been so well preserved deformed (Anton, 2003, pers. comm. to P.S.); symme- under these circumstances.” Later he wrote (1937: 25): try could not be restored even if all of the pieces were “we are certain that it [the cranium] was found in situ, separated from the matrix endocast. The remaining por- because the bone is so thin that it would have been de- tion of the left temporal, bearing the zygomatic process, stroyed by any movement or rewashing.” has been moved in an anterior direction and rotated in The Mojokerto cranial vault is broken into many a clockwise direction from lateral view. It is possible angular fragments. A piece of the frontal is missing at that the plastic deformation and warping of the specimen midline, and a fracture which runs from the edge of the occurred in stage 1, while the bone was still somewhat missing section to bregma suggests that there was prob- elastic. However, we cannot judge with certainty when ably a metopic fracture. Small pieces of bone are missing in the taphonomic history of the specimen this plastic at bregma and at various points along the coronal, sagit- deformation and warping of the bony tissue occurred, tal, and occipital sutures. However, the main cranial su- since sedimentary pressures are capable of warping con- 224  Breathing Life into Fossils: Taphonomic Studies in Honor of C.K. (Bob) Brain
and portions of the temporal bones were not recovered. Fractures of the Mojokerto cranium typical of post- Only the squamous occipital is preserved; the basicranial mortem stage 2 include: the deduced metopic fracture, part of the occipital and the sphenoid are both missing. the perpendicular fractures, and the mosaic fracture pat- These features might be expected in either in dry-bone tern of the cranial vault. The number of angular frag- damage or in post-fossilization fractures (stages 2 or 3). ments are neither as numerous nor as small as those in, White (2004, personal communication to P.S.) con- for example, Herto BOU-VP-16/5 (see below). The Mo- cluded that the fractures were primarily or wholly post- jokerto specimen also shows various fragments of bone fossilization based on three observations. First, matrix- pushed into the endocast, a possible temporal line frac- fi lled cracks between pieces and ectocranial matrix that ture, and a rotation of the temporal, all of which must bridged adjacent pieces indicate that the cranium was have occurred after the bone had dried but before the embedded whole. Second, the fi lling of various voids (such as sinuses, diploe spaces, etc.) with matrix shows Many areas of the cranium in the high and interme- in many cases that anatomically adjacent fragments were diate taphonomic vulnerability groups have been broken in place when the matrix hardened. Finally, there were or are missing in this specimen: the entire face, the zy- plant rootcasts on the endocranial and ectocranial surfac- gomatic arches, and the basilar occipital. Much of the es but none on the fracture surfaces, showing that break- squamous occipital is preserved, as is most of the neuro- age occurred well after sedimentary burial and probably cranium. The superior margin of the right orbit is broken, even though this is an area of the cranium most likely The Herto BOU-VP-16/5 cranium shows defl eshing to be intact in archaeological specimens (see discussion cutmarks around the perimeter of the glenoid fossa and polishing of the broken edges of the occipital and tem- In summary, the Mojokerto cranium was probably poral bones. These alterations are taken as evidence of subjected to late stage 1–early stage 2 breakage. Damage postmortem treatment of the cranium by hominids, per- most likely occurred after sedimentary burial but before haps as part of a mortuary ritual (Clark et al., 2003: 751). consolidation of the endocast and while the cranial bones This damage was most probably infl icted during stages 1 were suffi ciently elastic to warp and deform as well as (the defl eshing) and 2 (the polishing of broken edges).
break with beveled edges. Without the natural endocast, Despite the extensive fragmentation of Herto BOU- it seems likely that many of the individual fragments VP-16/5, which bespeaks intense exposure to tapho- would have separated from one another along fracture nomic agents of destruction, large parts of the fragile lines. There is little or no evidence of separation along facial bones are preserved. Both nasals are present; the sutures. The face was broken off in a LeFort III pattern. left orbital rim is intact as is most of the left zygomatic The preservation of this cranium suggests exposure to arch; substantial parts of both maxillae are present. The survival of some (but not all) of the elements in the most taphonomically vulnerable category combined with ex- Herto BOU-VP-16/5
tensive fragmentation suggests that breakage occurred Herto BOU-VP-16/5, an immature cranium of Homo after fossilization had enhanced the structural strength sapiens idaltu, was discovered in 1997 in the Herto Bouri of elements that are fragile in stages 1 and 2. The pres- region of Ethiopia (Clark et al., 2003; White et al., 2003). ence of a LeFort I type fracture is very rare in imma- The specimen was recovered in over 180 pieces, which ture fossils. If the specimen had been subjected to more were found on the surface after eroding out of an indurat- extensive taphonomic destruction, the LeFort I fracture ed sandstone. In the view of the discoverers, the cranium would have probably progressed to a LeFort III fracture. was modifi ed and curated by hominids after the death If efforts to recover fragmentary pieces of the cranium of the individual (Clark et al., 2003; White et al., 2003). had been less intensive, the specimen might well appear The pieces of the Herto cranium are numerous, angular, and appear to be planar on the edges (White et al., 2003; We fi nd no evidence that would lead us to ques- White, 2003, personal communication to P.S.). The mid- tion the interpretation that the cranium was defl eshed dle part of the face is missing but the maxillary alveoli and curated (during stage 1), resulting in polishing of are preserved; this is the only example of a LeFort I frac- edges around the broken-out basicranium (probably dur- ture we observed among the fossil crania. Much of the ing stage 2). From the observations and data presented basicranium is damaged or missing with the exception above, we deduce that most of the mosaic fragmentation of the petrous temporals. Metopic or parametopic, tem- and fracturing of the vault and face of Herto BOU-VP- poral line, and perpendicular fractures are absent as are 16/5 occurred during stage 3, the post-fossilization pe- elongated linear and diastatic fractures. Fractures of the cranial vault are numerous, short, and straight, but do not follow the sutures. Thus the entire neurocranium is com- CONCLUSIONS
prised of angular fragments in a mosaic fracture pattern. While the coronal, sagittal, and lambdoid sutures have We have summarized and integrated quantitative and not separated, the temporal sutures apparently opened qualitative data from medical, forensic, archaeological, Krovitz and Shipman  225
and paleontological sources in an attempt to characterize Berryman, H.E., Symes, S.A. 1998. Recognizing gunshot and the taphonomic attributes of immature hominid crania. blunt cranial trauma through fracture interpretation. In: From these diverse observations, we have created a set of Reichs, K.J. (Ed), Forensic Osteology, Advances in the expectations that relate fracture patterns to taphonomic Identifi cation of Human Remains, 2nd Edition. Charles C. Thomas, Springfi eld (IL), pp. 333-352 vulnerability and that describe fracture morphology and placement in relation to the time of breakage relative to Billy, G. 1979. L’enfant Magdalenien de la Grotte due Figuier. the death of the individual. Data on the breakage and preservation of individuals from the Krovitz and Sedg- Blount, W.P. 1955. Fractures in Children. Williams and eford samples have been used to identify key differences in breakage between immature and adult crania, respec- Blount, W.P. 1977. Fractures in Children. Robert E. Krieger We tried to show how these expectations might be Brain, C.K. 1981. The Hunters or the Hunted? An Introduc- tion to African Cave Taphonomy. University of Chicago, used in practical terms by re-analyzing three immature fossil crania from Taung, Mojokerto, and Herto. We re- Bromage, T.G. 1985. Taung facial remodeling: A growth gard the work reported here as a fi rst approximation and and development study. In: Tobias, P.V. (Ed), Hominid still speculative. We encourage further research along Evolution: Past, Present and Future. Alan R. Liss, New these lines in order to produce more refi ned and useful diagnostic tools for the taphonomist, paleontologist, and Byers, S.N. 2002. Introduction to Forensic Anthropology. Al- Clark, J.D., Yonas Beyene, Y., WoldeGabriel, G., Hart, W.K., ACKNOWLEDGEMENTS
Renne, P.R., Gilbert, H., Defl eur, A., Suwa, G., Katoh, S., Ludwig, K.R., Boisserie, J.-R., Asfaw, B., White, We wish to gratefully acknowledge receiving infor- T.D. 2003. Stratigraphic, chronological and behavioural mation and assistance from Susan Anton, Berhane As- contexts of Pleistocene Homo sapiens from Middle faw, Frank Huffman, Patricia Reid, Alan Mann, Jeffrey McKee, George Milner, Erik Trinkaus, and Tim White Cobain, L.R., Nabipour, S., Richards, G. 2002. The role of on this paper. G.K. would like to thank those who gave sutures in frontofacial growth: evidence from the metopic her access to the skeletal collections in their care, or who suture. Paper delivered at the 80th General Session of helped her obtain or interpret the dental x-rays for the the International Association for Dental Research, the recent human samples: Hisao Baba, Pia Bennike, Jodie American association for Dental Research, and the Cali- Blodgett, Luca Bondioli, Jennifer Clark, Kevin Conley, fornia Association for Dental Research, San Diego, CA. Jean-Marie Cordy, Kate Hesseldenz, Louise Humphrey, Collins, H.B.J. 1937. Archaeology of St. Lawrence Island, David Hunt, Robert Kruszynski, Helen Liversidge, Alaska. Smithsonian Miscellaneous Collections 96(1). Niels Lynnerup, Roberto Macchiarelli, Giorgio Manzi, Smithsonian Institution Press, Washington, DC.
Yuji Mizoguchi, Theya Molleson, Søren Nørby, Nancy Cox, T.C., Buchilz, D.J., Wolf, D.J. 1987. Blunt force trauma O’Malley, Rosine Orban, Doug Owsley, Ildiko Pap, from police impact weapons: some skeletal and neuro- Rick Potts, Edourard Poty, Mary Powell, Patrick Semal, psychological considerations. Journal of Police Science Ib Sewerin, Gabriella Spedini, Chris Stringer, and Erik Trinkaus. G.K.’s research was supported by grants from Currey, J. 1984. The Mechanical Adaptations of Bones. Princ- the L.S.B. Leakey Society, the National Science Foun- dation, and the Japanese Society for the Promotion of Dart, R.A. 1925. Australopithecus africanus: The man-ape of Science. P. S. especially wishes to thank Bob Brain for being an inspiration throughout her entire career.
Dean, C., Leakey, M.G., Reid, D., Schrenk, F., Schwartz, G.T., Stringer, C., Walker, A. 2001. Growth processes in teeth distinguish modern humans from Homo erectus and REFERENCES
earlier hominins. Nature 414, 628-631.
Adams, M., Reeve, J. 1987. Excavations at Christ Church, Dodo, Y., Kondo, O., Muhesen, S., Akazawa, T. 1998. Anat- Spitalfi elds 1984-6. Antiquity 61, 247-256.
omy of the Neandertal infant skeleton from Dederiyeh Cave, Syria. In: Akazawa, T., Aoki, K., Bar-Yosef, O. Anton, S. 1997. Developmental age and taxonomic affi nity of (Eds), Neandertals and Modern Humans in Western Asia. the Mojokerto child, Java, Indonesia. American Journal of Physical Anthropology 102, 497-514.
Duncan, C.C. 1993. Skull fractures in infancy and childhood. Akazawa, T., Muhesen, S., Dodo, Y, Kondo, O., Mizoguchi, Y. 1995. Neanderthal infant burial. Nature 377, 585-586.
Duyfjes, J. 1936. Zur Geologie und Stratigraphie des Kend- Barker, R., Hockey, R., Spinks, D., Miles, E. 2003. Facial enggebietes zwischen Trinil und Soerabaja (Java). De Injury. Injury Bulletin, Queensland Injury Surveillance Ingenieur in Nederlandsch-Indië, Mijnbouw & Geologie, Berger, L.R., Clarke, R.J. 1995. Eagle involvement in the ac- Ferembach, D. 1970. Le crane de l’enfant du Pech-de-l’Azé. cumulation of the Taung child fauna. Journal of Human Archives de l’Institut de Paleontologie Humaine 33, 226  Breathing Life into Fossils: Taphonomic Studies in Honor of C.K. (Bob) Brain
Fraipont, C. 1936. Les hommes fossils d’Engis. Archives de Meindl, R.S., Lovejoy, C.O. 1985. Ectocranial suture closure: l’Institut de Paleontologie Humaine 16, 1-52. A revised method for the determination of skeletal age Galloway, A. 1999. Fracture patterns and skeletal morphol- at death based on the lateral-anterior suture. American ogy: Introduction and the Skull. In: Galloway, A. (Ed), Journal of Physical Anthropology 68, 57-66.
Broken Bones: Anthropological Analysis of Blunt Force Minugh-Purvis, N. 1988. Patterns of craniofacial growth and Trauma. Charles C. Thomas, Springfi eld (IL), , pp. 63- fevelopment in Upper Pleistocene hominids. Ph.D. Dis- sertation, University of Pennsylvania.
Gordon, J.E. 1968. The New Science of Strong Materials Mizoguchi, Y. 1997. Human Bones from the Ikenohata-Shi- or Why You Don’t Fall Through the Floor. Penguin, chikencho Relics (in Japanese). Ikenohata-Shichikencho Iseki (Keianji-Ato). Taito-ku Ikenohata-Shichikencho Gurdjian, E.S., Webster, J.E., Lissner, H.R. 1953. Observation on prediction of fracture site: Head injury. Radiology 60, Molleson, T., Cox, M. 1993. The Spitalfi elds Project, Vol. 2: The middling sort. Council for British Archaeology Gurdjian, E.S. 1975. Impact Head Injury, Mechanistic, Clini- cal and Preventive Correlations. Charles C. Thomas, Moorrees, C.F.A., Fanning, E.A., Hunt, E.E. 1963. Age varia- tion of formation stages for ten permanent teeth. Journal Heathcote, G.M. 1986. Exploratory Human Craniometry of Recent Eskaleutian Regional Groups from the Western Murphy, S.P., Sledzik, P.S., Mann, R.W., Kelley, M.A. 1990. Arctic and Subarctic of North America. British Archaeo- Macroscopic bone remodeling following trauma: Re- logical Reports International Series 301, Oxford.
considering the term perimortem. Paper presented at the Huffman, O.F., Zaim, Y. 2003. Mojokerto Delta, East Java; annual meeting of the American Association of Forensic Paleoenvironment of Homo modjokertensis—fi rst re- sults. Journal of Mineral Technology 10, 1-9. Naim-Ur-Rahman, Jamjoon, Z., Jamjoon, A., Murshid, W.R. Huffman, O.F. 2001. Geologic context and age of the Perning/ 1994. Growing skull fractures: Classifi cation and man- Mojokerto Homo erectus, East Java. Journal of Human agement British Journal of Neurosurgery 8, 667-679. Pap, I., Tillier, A.-M., Arensburg, B., Chech, M. 1996. The Ishida, H., Kondo, O., Muhesen, S., Akazawa, T. 2000. A new Subalyuk Neandertal remains (Hungary): A re-exami- Neandertal child recovered at Dederiyeh Cave, Syria in nation. Annales Historico-Naturales Musei Nationalis 1997-1998. American Journal of Physical Anthropology Patté, E. 1957. L’Enfant Néandertalien du Peche de l’Azé. Krovitz, G.E. 2000. Three-dimensional comparisons of cra- niofacial morphology and growth patterns in Neandertals Rak, Y., Kimbel, W.H., Hovers, E. 1994. A Neandertal infant and modern humans. Ph.D. Dissertation, The Johns from Amud Cave, Israel. Journal of Human Evolution Lyman, R.L. 1994. Vertebrate Taphonomy. Cambridge Uni- Rak, Y., Kimbel, W.H., Hovers, E. 1996. On Neandertal auta- pomorphies discernible in Neandertal infants: A response Mackey, M. 1984. Injuries to the face and to skin. In: Aldman, to Creed-Miles et al. Journal of Human Evolution30, B., Chapon, A. (Eds), Biomechanics of Impact Trauma. Elsevier Science Publisher, Amsterdam, pp. 335-340.
Richardson, M. L. 2000. Approaches to Differential Diagnosis Madre-Dupouy, M. 1992. L’Enfant du Roc de Marsal: Etude in Musculoskeletal Imaging. <http://www.rad.washing- Analytique et Comparative. Editions CNRS, Paris.
Marks, M.K., Hudson, J.W., Elkins, S.K. 1999. Craniofacial Rogers, L. 1992. Radiology of Skeletal Trauma. Churchill fractures: Collaboration spells success. In: Galloway, A. (Ed), Broken Bones: Anthropological Analysis of Blunt Saunders, S. 2000. Subadult skeletons and growth-related Force Trauma. Charles C. Thomas, Springfi eld (IL), pp. studies. In: Katzenberg, M.A., Saunders, S. (Eds), Biological Anthropology of the Human Skeleton. Wiley- McCown, T.D., Keith, A. 1939. The Stone Age of Mount Carmel: Volume II. Clarendon University, Oxford.
Sauer, N. 1998. The timing of injuries and manner of death: McElhaney, J.E., Reynolds, V.L., Hilyard, J.F. 1976. Hand- Distinguishing among antemortem, perimortem and post- book of Human Tolerance. Japan Automobile Research mortem trauma. In: Reichs, K.J. (Ed), Forensic Osteol- ogy: Advances in the Identifi cation of Human Remains. Charles C. Thomas, Springfi eld (IL), pp. 321-332.
McKee, J.K. 2001. The Taung raptor hypothesis: Caveats and evidence. American Journal of Physical Anthropology Schwartz, J.H., Tattersall, I. 2002. The Human Fossil Record, Volume I: Terminology and Craniodental Morphology of Genus Homo (Europe). Wiley Liss, New York.
McKee, J.K., Tobias, P.V. 1994. Taung stratigraphy and tapho- nomy: Preliminary results based on the 1988-93 excava- Schwartz, J.H., Tattersall, I. 2003. The Human Fossil Record, tions. South African Journal of Science 90, 233-235.
Volume II: Craniodental Morphology of Genus Homo (Africa and Asia). Wiley Liss, New York.
McKern, T.W., Stewart, T.D. 1957. Skeletal Age Changes in Young American Males. Quartermaster Research and Development Center, Matick (MA).
Krovitz and Shipman  227
Smith, B.H. 1991. Standards of human tooth formation and Todd, T.W., Lyon, D. 1925b. Cranial suture closure, its prog- dental age assessment. In Kelley, M.A., Larsen, C.S.

Source: http://www.stoneageinstitute.org/pdfs/breathing-fossils-ch12-krovitz-shipman.pdf