Divje babe I-novo paleolitsko najdišče in skupinsko grobišče jamskega medveda. Poskus tafonomske analize na podlagi vzorcev iz dveh sedimentnih in arheoloških kompleksov

The 1980-1986 excavations in the newly-discovered, as yet unpublished palaeolithic cave site Divje babe I in the Idrijca valley (Cerkno, Idrija District, Slovenia), yielded a vast number of remains of cave bear (Ursus spelaeus Rosenmiiller et Heinroth 1794). A portion of the remains has for the first time been studied systematically on the basis of standard samples from a sampling area of a specific cubic content (Figs. 1; 3). The cubic content of all sediments covered by the sampling is 20 m3. This volume, which comprises no more than 1/25 of all the sediments investigated, has yielded the remains of the cave bear as shown in Tables 2-3. The remains have been divided into seven arbitrary stratigraphic units of approximately the same thickness, with units representing individual bone and teeth samples from the homogeneous sampling area of a standard surface of 10 m2. The techniques applied in the analyses of samples have been made conformable to the technique of excavation and the degree of its exactitude, as well as to the nature of the osteoodontological material. Instead of subjecting the sediments to a systematic screening, all of the surviving finds have been taken in without exception. Any sort of selection of the material has deliberately been avoided. All the remains have been divided on the basis of clear-cut morphological differences into two groups, that of the juvenile and that of the adult specimens, with a further distinction in teeth between the left and the right. If necessary, parts of the skeletons and the slightly varying volumes of the sediments which were used for sampling have been loaded. All the samples derive from two sedimentary units of strata (Fig. 4) which, in turn, consist of two Moustčrian cultural levels of different quantities (Fig. 3). Sediments, which are diagenetically more or less changed, are composed of dolomitic gravels and fine sand. The upper unit represents the »phosphatic layer« (8) containing an average of 15.8 % of phosphates, while the lower unit consists of layers (10-14) containing some traces of phosphate incrustation made indistinct through some secondary mechanical agency, and averaging 13.2 % of phosphates. The line of demarcation between the two units is clear-cut, whereas those demarcating individual layers tend to be blurred. Typologically, diagenetically, and material-wise, there is no difference between the archaeological finds from the D- and E-level, or in fact from any other level. They all belong to the final stage of the Mousterian complex, and have been given an absolute date of over 38,000 B. P. (Zagreb 1981, Z-1033). In over 99%, the fauna is dominated by the cave bear. The finds of other animal species from the sampling area are shown in Tabic 5. All units have also been submitted to the palynological and anthracotomical analyses in several vertical series. However, only the results produced by the investigation of the units of the sampling area are given here. The pollen diagram (Fig. 5) indicates higher pollen values of herbs by comparison with those of the tree vegetation. This kind of pollen assemblage is typical of steppe vegetation, and indeed this had been one of the kind. It is only the question of how it is possible that a higher percentage of the pollen of entomophilous plants than of that of anemophilous plants had found its way to the cave, which gives occasion to doubts. Certainly not by force of wind. Obviously it had been brought here by some animals - either by solitary bees and other insects, or by the cave bears that must have been nibbling at the herbs on the grass-grown elevated plain over the cave, and their pollen is preserved in coprolites. However, the results of the pollen analyses admit of certain ecological conjectures about the cold climate vegetation (which is demonstrated by predominant coniferous trees) in the form of light forests associated with steppe vegetation (Compositae, Umbelliferae, Caryophyllaceae, and Gramineae). Nevertheless, the sporadic occurrence of the pollen and charcoal of mesophiles would seem to point to occasional, though slight rises of temperature. Since the radiocarbon dating of »over 38,000 years« establishes this segment of the profile as belonging to the middle Wiirm period, these results are found to be in perfect accordance with the results of the pollen analyses from other parts of Slovenia which have in a number of occasions established the heliophilous vegetation of conifers associated with a considerable prevalence of herbs, and always with the sporadic occurrence of the pollen of the mesophilous deciduous trees, mostly lime, oak, and elm. The common characteristic of the sediments and their archaeological, faunistic, and floristic contents is a high degree of homogeneity, which would seem to suggest a rapid and uninterrupted process of sedimentation. What we have here is probably a fairly perfect profile from a relatively short cold climate interval of time dating from the middle section of the Wurm glaciation, between the Br<6rup and Hengelo interstadials. All the remains of the cave bear have been analysed in terms of skeletal element and tooth representation, minimum number of individuals, determination of age, determination of sex, and fragmentation and specific damage on bones. The analysis of the parts of the skeleton has revealed a high degree of random dispersion and a strong intermixture of all remains. Both are accountable for by bioturbation. The samples display not one single anatomical arrangement in a group, and only few chaotic groupings of different long bones belonging to different individuals (the latter exclusively in E-level). The higher or lower degrees of disparity that have been established in terms of the skeletal element representation between individual samples are for the most part quantitative. Such disparities are found to be all the more conspicuous between sedimentary units and archaeological levels relative to the skeletal element representation of juvenile and adult individuals respectively. On the whole, the E-level exhibits a superior skeletal element preservation of both age profiles, which in turn suggests the exclusion of man as a predominant taphonomic agent prevailing upon the skeletal element representation (Figs. 6; 7). The teeth display a much smaller degree of dispersion and taphonomic loss by comparison with that in bones. For the purpose of a standard, similar pattern in pairs of isolated teeth (Fig. 11) and metapodials (Fig. 8) have been taken. By way of cumulative frequences of isolated teeth arranged according to the arbitrary stratigraphic units, surpluses of lower molars have, been established (Fig. 9), which is explainable in terms of the surplusage of mandibles, or the deficiency of cranial parts of the skulls. Additionally, more mandibles than maxillae have been recovered from the E-level (Fig. 10). The figures shown have been obtained through adding together individual adult and juvenile teeth, and then singling out maximum amounts for either upper or lower teeth which in turn represent the minimum possible numbers of all maxillae and mandibles respectively. The frequency of occurrence in individual isolated teeth in samples is, in spite of some interruptions, in general agreement with the sequence of the eruption of teeth (Fig. 13). (Couturier M. A. J., 1954, 145; Ehrenberg K., 1931, 675 ff.; Ehrenberg K., 1964.) The minimum number of individuals has been approximated on the basis of four pairs of upper (I3-M2), and five pairs of lower teeth (I3-M3) (Fig. 12). It has been found that this number adheres to certain regular recurrences relative to a higher degree of durability in certain teeth. Eventually, these teeth always act as the determining factor of the minimum number of individuals. All that is needed is a sample of ample size. In this case, it is the lower Ml and, M2 which have been singled out previously by help of cumulative frequences (Fig. 9). All available methods used for the determination of age profiles (Schmid E., 1972, 75; Watson J.P.N., 1978; Sergeant D.E., 1967; Klein R.G., Cruz-Uribe K., 1984, 41 ff.; Geiger G. e t al., 1977; Morris P. A., 1972) are practicable only under the condition that the bones and teeth are distinguishable according to their parent individuals. The reason is that there is a stage in ontogenesis in which some of the osteoodontological elements have already reached the stage of adulthood morphologically, whereas others, morphologically speaking, are still in their stage of juvenility. Thus, as a result of disintegration of skeletal and dental associations °f certain individuals which persist in diverse juvenile stages of their ontogenesis (which is the case in most of the sites with mass accumulations of cave bear bones), a highly heterogeneous sample is produced. This fact has unfortunately been completely disregarded up till now (cf. Biichler H., 1957; Kurt6n B., 1958), although K. Ehrenberg did pointed it out in passing (id., 1935, 105). In view of the morphological ambiguity at a certain juvenile stage of development, "igh mortality rate in present-day populations of bears at this very stage, and the fact that the frequences of most of the individual permanent teeth comply reasonably with the natural sequence of their growth, the following procedure, based on the method of minimum number of individuals, has been adopted for the purpose of determining the number of juvenile and adult individuals in the present samples. To begin with, left-left and right-right pairs of juvenile and adult teeth were selected for the purpose of investigation. To achieve by this was the avoidance of mixing several individuals in the instance of one and the same tooth, and the by-passing of the morphological ambiguity between several individuals. Left-right teeth clearly definable beyond any doubt were arranged according to the sequence of their growth, which l s Ml, M2, 13 for the upper, and Ml, M2, 13, M3 for the lower set. P4 was intentionally left out as it has no effect on the final result owing to the rare cases of its occurrence. On account of the bad state of their preservation, few cases of them being found, and low degree of their definability, canines, which are the last to grow and are therefore determinative for the number of adult individuals, were not taken into consideration either. In the case of a tooth which is the last to grow, the pair displaying a larger number of adult teeth was chosen invariably. The adult teeth of the final pair (in terms of eruption) represent the minimum number of adult individuals. In cases of all other teeth, pairs displaying the larger number of juvenile teeth were chosen in all instances. Out of these, the maximum number of juvenile teeth was adopted as the minimum number of juvenile individuals. By way of such procedure the results shown in T. 8 have been obtained. The combination of both maximum numbers was then selected as the minimum number of both the adult and juvenile individuals. In view of a high probability that all pairs of teeth selected in this way (3 upper, and 4 lower) do not derive from the same individuals, the numbers of both the adult and juvenile individuals may well be increased. However, the possibilities of the minimum number of adult individuals to increase will be decreased by an increase in the minimum number of juvenile individuals, and vice versa. A similar result has been obtained through a comparison of left-left and right-right juvenile-adult pairs of 4 upper and 5 lower teeth (Fig. 14). It has been established that there are very few individuals from the early stages of diphyodontia (all permanent teeth still have hollow roots; there are a number of deciduous teeth still present, mostly canines), or from an age under 8 months according to K. Ehrenberg (id., 1964, 217). This fact admits of an explanation in that the taphonomic losses were larger, or the mortality rate was lower, due to the presence of mother bears during the cubs' first months in the winter and spring seasons. It may well be that during a following winter an increasing number of cubs were compelled to hibernate on their own for various reasons, which presumably resulted in a higher mortality rate during the first years of their lives. Incidentally, this is roughly what is discernible from the present samples. Furthermore, there is a conspicuous difference between the D- and E-levels in terms of their age structures; namely, D-level display a relatively higher number of adult individuals. This fact becomes even more interesting if compared with those concerning the ratios between the sexes. Essays at sex determination of cave bear fossil remains have, in most instances, been performed on teeth, mostly canines (Koby F.-Ed., 1950; Kurtčn B., 1955). We have reason to believe that canines are not the most felicitous choice for the determination of ratios between the sexes, if at all we should accord recognition to the hypothesis of sexual dimorphism in teeth. The weight is on sex rations, not on the determination of sex. By reason of their specific status relative to other teeth, the canines were easily shown to exhibit high taphonomic losses (deficits) in most samples (T. 9). For this reason adequate ratios between the sexes would not be obtainable on the basis of canines. Molar teeth, on account of lower taphonomic losses, seemed to fit the purpose more adequately. The measurements (with an accuracy of 0.1mm) of three morphologically least variable molars and the third upper incisor have shown, in consideration of individuals, that D-level displays predominantly higher values (left-skewed histograms), whereas E-level displays predominantly lower values (right-skewed histograms (Fig. 13). From the viewpoint of sexual dimorphism this would argue for a higher number of males in D-level, and a higher number of females in E-level, on the supposition that the measured qualities have retained their normal distribution, and that the basic characteristics of the population (range of variation) have remained unchanged. A higher number of females (both young and old) suggests a variation in the use of den. Under natural conditions, approximately the same number of cubs of both sexes is born, with approximately the same mortality rate in both cases. Consequently the same number of cubs of both sexes is to be expected in the present samples. The mortality rate of adult subjects, on the other hand, is dependent upon the displayed number of either males or females in the den, i. e. upon the use of the den. An increase in the number of either males or females may therefore suggest a highly probable change in the utilization purpose of the den. For it is a well-known fact from the ethology of recent bears species that adult females with their young and adult males are definitely mutually exclusive, in dens and elsewhere (Manvill A.M., 1986). In the course of the study of the fragmentation of the material, consideration was given to its heterogeneous and multilayer character. The fact is that the analysed damage took its origin in different stages of the transition of samples from biosphere to lithosphere by cause of the activity of diverse biotic and abiotic agencies. This finally resulted in as much as 85 % of the bones being more or less damaged. Main sources of the damage thus produced are: - mechanical and chemical weathering, - predatory destruction for the purpose of consuming bones and marrow for food, - human destruction for the purpose of extraction of marrow and other purposes, - inattention during excavations. All fragments have been arranged into five groups according to their sizes (T. 11), and into sharp-edged and rounded-off according to the outline of their fractures (T. 12). The number of fragments smaller than 5 cm is a highly conservative one, since ca. 64 % of such have been overlooked due to the field-work technique. The average weights of individual osteoodontological finds have pointed to a relatively higher amount of smaller fragments at E-level by comparison with that of D-level (Fig. 16). Special attention should be paid to the fragments with edges of their fractures rounded off. This rounding of edges was in most instances caused mechanically, and is independent of the size of fragments. Rounded edges of such fragments testify to fractures which were obviously occasioned on death assemblages, and are attributable almost to none other than biotic agencies. There is a marked difference in roundness between the bone fragments of D-level and those of E-level (Fig. 18), which may have resulted from certain changes in rates of sedimentation (shorter hiatuses?) and general conditions before and during sedimentation. There are also significant differences in the fragmentation of individual skeletal remains between the juvenile and adult individuals. Some of the damage, such as cranial and long marrow bones of the adult individuals' broken open, are perhaps attributable to the palaeolithic men considering the facts such as rounded-off edges of fractures, small dimensions of fragments, and the impossibility of assembling these fragments. Some bones of juvenile individuals, on the other hand, display the kind of damage which makes it certain beyond any doubt that it was caused by predators (in view of significant traces such as gnawing and tooth marks). Specific kinds of damage in bones, such as leaching, tooth marks, burns, and cutmarks, are presented in T. 13. An interesting feature is the complete absence of cutmarks such as Would be attributable to the activities of palaeolithic hunters. Predatory marks, on the other "and, are in perfect agreement with the frequence of the skeletal remains of the wolf. A part °t predatory damage on bones could be also attributable to the cave bear which may have crushed the bones but did not leave many traces on them. Unfortunately, the role of ancient men in the den of his presumed game remains unaccounted t°r at the present stage of investigation. For this purpose a more detailed treatment of the samples is being prepared that will proceed through the application of non-parametric statistical methods