«Item type text; Dissertation-Reproduction (electronic) Authors Munro, Natalie Dawn Publisher The University of Arizona. Rights Copyright © is held ...»
assemblages in low frequencies and they all have some potential to naturally inhabit or die in caves, particularly those that are well lit such as Hayonim Cave and Hilazon Tachtit. These species are also on the smaller end of the potential game size continuum, increasing the possibility that they were consumed by small to medium-sized diumal raptors or bam owls. Overall, the evidence for the collector of these four species is ambiguous, and not enough data are available to demonstrate or disprove that humans ate them. For this reason, as well as their low frequencies, they are omitted from further analyses. Because their abundance is low, particularly in the case of hedgehogs and squirrels, their removal from further consideration has no significant impact on the total representation of species at the sites.
Summary of Large and Small Game Assemblage Formation Histories There is no question that the Hayonim Cave and Hilazon Tachtit assemblages were collected almost entirely by humans. Ungulates, mammalian carnivores, tortoises, hares, partridges, and Falconiformes all bear strong evidence of human modification.
The only problematic species are a few relatively rare small animals, including the two lizards Agatna stellio and Ophisaunis apodus and the squirrel and hedgehog. These rare animals are excluded from further analysis.
After disposal, bone is subject to structural and chemical breakdown through natural and cultural processes. Chemical decomposition occurs mainly through
the complete leaching of bone at the other. Mechanical decomposition is caused by physical processes that alter the macrostructural properties of bone (see Lyman 1994).
Mechanical processes of decomposition include abrasion, water transport, and gnawing by rodents and carnivores; and to some extent also weathering. Here, the term in situ decomposition is limited to breakdown by chemical processes and non-human mechanical agents. Human effects will be examined separately in the next chapter.
Bone Decomposition and Surface Damage Each bone in an archaeological assemblage may record clues to its taphonomic history through damage sustained from attritional processes. Here, damage likely to have been caused by non-human mechanical and chemical forces after deposition is considered at the individual specimen and assemblage level.
Root Etching After human activity ceases at a site, plants and animals quickly recolonize disturbed deposits and frequently leave traces of their presence on bones. In less than a year between excavation seasons at Hayonim Cave and Hilazon Tachtit, the caves were rapidly revegetated by a blanket of weeds and grasses, as well as large plants such as young fig trees. The rootlets of growing plants often come into contact with buried bones. As they grow, roots secrete chemicals with the strength to dissolve bone mineral, and leave squiggled etchings on bone surfaces (Lyman 1994). Root etching may vary significantly in intensity from light imprints to the complete destruction of bone, depending on the size and species of the plant and the amount of time roots are in direct
after discard. In this study, root etching was recorded as a presence/absence variable, with an additional note on its severity (light, moderate, heavy).
Weathering Weathering is caused by the prolonged exposure of bone or other materials to some combination of air, sunlight, moisture, and changing temperatures. Weathering causes the breakdown of bone in predictable ways as the fibrous collagen component of its structure decomposes (Behrensmeyer 1978; Gifford 1981; Miller 1975). Mild mineral recrystallization may also occur as weathering progresses (Stiner et al. 1995).
Behrensmeyer (1978) divides the weathering process into six stages, defined by straightforward changes in the condition of the bone's outer surface. In the initial stages, bones become bleached and begin to crack in lines parallel to their main axis or grain.
Next, cracks expand and separate, and the exterior layer begins to exfoliate. Cracks eventually widen, their edges become rounded, and fi-agments of bone break away fi-om the surface. Severely weathered bone can be reduced to dust in situ and may completely disappear. Weathering can also blur and distort diagnostic features on a bone's surface such that the element and species are no longer recognizable. Burial beneath the ground surface generally arrests the weathering process. A bone's weathering stage may therefore provide a rough, relative gauge of how long a bone was exposed on the surface before it was buried (Lyman and Fox 1989). Here, weathering was recorded on a scale from 0 to 5 following Behrensmeyer (1978).
Though continuous growth prevents their incisors from dulling, they must be honed to a practical length to prevent dysfunction and injury. Most rodents keep their teeth in check by gnawing on hard substances and prefer bone if available, especially dense, compact fragments (Brain 1980; Lyman 1994). Gnawing produces deep double grooves on bone surfaces, the impressions of pairs of upper and lower incisors working in opposition.
Rodents rarely completely consume a bone, but they can obscure diagnostic features or cause structural instability that may encourage decomposition by other processes. The frequency of gnawing reflects the extent of rodent activity at a site either immediately or long after human occupation. It can be a useful indication of the level of disturbance caused by rodent burrowing. When present on bones in the Natufian assemblages, the intensity and location of gnaw marks were also noted. Intensity was recorded as very light, light, moderate, heavy or breaching the bone's cavity.
Results on Surface Damage The results on the frequency of damage from root etching, weathering and rodent gnawing in the Natufian faunas are presented in Table 4.6. Root etching was present on 5.4% of ungulate bones from Hayonim Cave and 5.3% from Hilazon Tachtit. Root etching is heavy in less than one twentieth of these cases, and severe enough to erode holes in only two instances at Hayonim Cave. In no case was the identifiability of recovered bone compromised by root etching. The impact of weathering is even less severe. Though 3.9% of the Hayonim and 1.9% of the Hilazon ungulate assemblage displayed evidence of weathering reached only stage I, light cracking, in more than three
to stage 3, and one to stage 4. Rodent gnawing is even less common, affecting 1.4% of the ungulate assemblage; gnawing damage on only one bone from each assemblage was considered heavy. In all macroscopic categories bone damage is very light. The low intensity of in situ macroscopic damage caused by non-human agents confirms observations made previously that the assemblage is in a very good state of preservation.
Though only ungulate frequencies are reported here (accounting for greater surface areas on bone), the proportions of such damage on carnivore and small game bones are similar.
Table 4.6: Frequencies of damage caused by natural mechanical processes of in situ decomposition on ungulate and carnivore remains from the Namfian layer of Hayonim Cave (HAYC) and Hilazon Tachtit (HLZT).
In Situ Attrition and Mineral Density of Skeletal Tissues Many methods for isolating the influence of in situ attrition caused by non-human forces of decomposition compare relative differences in the mineral density of skeletal elements. The mineral density of bone has been shown to correlate reasonably well with skeletal survivorship under a variety of conditions (Binford and Bartram 1977; Brain
depositional taphonomic processes including carnivore ravaging (Brain 1969; Haynes 1980; Marean et al. 1992), chemical decomposition (Lyman and Fox 1989), trampling (Nicholson 1992), and water transport (Boaz and Behrensmeyer 1976; Voorhies 1969), may be mediated in part by skeletal density. Dense bones have lower porosity, and lower surface area to volume ratios, and are therefore argued to be less vulnerable to attack from sources of attrition (Lyman 1994). Teeth are most resistant of all because they are highly mineralized. Skeletal tissues incorporate a diverse range of mineral densities ranging from dense teeth to compact bone, cancellous bone and paper-thin cranial sinuses and turbinals. By examining the relative representation of bone portions with known densities, it is possible to gauge the influence of post-depositional attritional processes (Lyman 1984; 1994: 235-258).
Tooth- and Cranial Bone-Based MNI Comparisons Stiner (1994: 99-103) introduced a test of in situ attrition that capitalizes on differences in the mineral density of teeth and cranial bones. As the densest component of the vertebrate skeleton, teeth are more resistant to decomposition and fragmentation than any bone, no matter how dense. By comparing tissues of different densities in the same anatomical unit (the skull), Stiner's test provides one relative measure of in situ attrition that should be independent of transport decisions. The test assumes that crania and possibly mandibles are deposited in the archaeological record with the teeth intact, so the MNI represented by both teeth and bone was potentially equal at the time of deposition. The intensity of attrition is measured by comparing the MNI of a species
of ideal preservation, the excavated assemblage is expected to mimic the deposited assemblage, and the ratio of bone- to tooth-based MNIs should be one. In cases where in situ bone loss has occurred, the ratio of bone- to tooth-based MNIs will drop below one, depending on the severity of decomposition. Values close to one indicate relatively good preservation. As conditions worsen the slope approaches zero.
This test is applied to all mammalian species from Hayonim Cave and Hilazon Tachtit that naturally possess both teeth and cranial bones. Gazelles provide the largest sample sizes. The only other ungulates with adequate representation are fallow deer and wild boar at Hayonim Cave, though samples for both of these groups are much smaller than for gazelles. Also considered are fox, wild cat and hare representing smaller mammal categories.
The tooth- versus bone-based MNIs for all taxa in the sample are listed in Table
4.7. In no instance does the MNI based on bone exceed that calculated from teeth, in keeping with the fact that teeth are more resistant to post-depositional processes than bone. In most cases, however, the tooth-based counts only slightly exceed bone-based counts. Figure 4.7 is a scatterplot of the tooth and bone-based MNIs listed in Table 4.7.
The slope of the line demonstrates that for most species the ratios of tooth- versus bonebased MNIs are close to 1, and that the results are consistent across taxa with the exception of hare. The MNI represented by hare teeth (27) from Hayonim is greater than for cranial bones (18). This may be caused by human processing of hare skulls for their
bones may be particularly fragile. Overall, the results of this test indicate that the quality of preservation of compact bone at both sites is very good.
Table 4.7: Tooth- and bone-based MNIs of ungulate, carnivore, and small mammal cranial bone and teeth from the Natufian layer at Hayonim Cave (HAYC) and Hilazon Tachtit (HLZT).
mandibles. The best represented portions (e.g., the petrous) are also the most dense (sensu Lyman 1984, 1985, 1994) and tend to be most resistant to decomposition.
Compact bone features also occur in many other elements of the skeleton, thus there is a good chance that MNE estimates for the full skeleton will be accurate representations of the number of elements that were originally deposited (Stiner n.d.). The high quality of bone preservation at Hayonim Cave and Hilazon Tachtit is not surprising given the freshness of specimens on visual inspection.
Density-Mediated Attrition and the Postcranial Skeleton Like the preceding tests comparing tooth- and bone-based MNIs, postcranial tests examine the relative representation of bone portions within an element, or between elements according to variation in mineral density. Here bone density is defined as grams per cubic meter following Lyman (1984, 1994), who measures bone density using a nondestructive technique called photon densitometry (or photon absorptiometry). This technique calculates density by directing a beam of photons through a predetermined location on a bone (scan site) and measuring its strength as it emerges from the bone (Lyman 1984, 1994). Lyman has published density standards for scan sites on sheep, antelope, and marmots, derived from the average bone density of a few individuals from each species (Lyman 1984; Lyman et al. 1992). Since then, other researchers have applied his methods to other mammals, providing standards for species including bison, deer (Kreutzer 1992), and hares (Pavao and Stahl 1999). Some of these standards are used here.
missing from an assemblage, and to what extent bone loss may be explained by structural density (sensu Lyman 1984, 1994). Bone density varies significantly within the mammalian postcranial skeleton and even within individual elements (Kreutzer 1992;
Lyman 1984; Lyman et al. 1992; Pavao and Stahl 1999). Though most elements contain at least some dense compact bones portions, the bones of the vertebral column are dominated by cancellous bone and thus have low overall density. Thick compact bones are more resistant to many attritional processes. The mineral density of most appendicular bones, including the long bones can be much greater, but still varies considerably within elements. Some articular ends are composed of spongy, cancellous bone, which is less dense than the thick cortical bone of long bone shafts and certain other articular ends such as the distal tibia and humerus. To compare the survivorship of portions of variable densities within the same element, the MNE for gazelle long bones is
calculated independently for the shaft and the distal and the proximal ends (Stiner 1994: