Abstract:
The quaternary glaciation of Europe brought dramatic climate change that strongly affected both the environment and its inhabitants. The primary aim of this dissertation is to investigate biological changes that were influenced by these climatic fluctuations, specifically correlations between body size of the variable species of caballine horse (Equus ferus) and environmental change during the Pleistocene. In order to analyse the morphological changes in horse most effectively, it is necessary to build a complete paleoenvironmental framework, which is possible through evaluation of a combination of data from archaeobiological, palaeontological, geological, sedimentological, and stratigraphical contexts.
This study is concerned with three aspects of the size variability in caballine horses: 1) the temporal variability of body size from the Middle to Late Pleistocene in central Europe, 2) the geographical variation of Equus ferus during the Late Palaeolithic from western to central Europe, and 3) the problem of taxonomy for the Genus Equus. In order to best characterize 'body size' of equids, the the osteometric method of "Variability-Size-Index" (VSI) (Uerpmann 1982) is used. VSI is a special method to compare diverse skeletal elements in biometrical evaluations with the effect of enlarging the number of comparative bone fragments. Size indices are defined as measures for the deviation of a particular dimension from the respective dimension in a 'standard', which can either be a single individual (Logarithmic-Size-Index LSI, Uerpmann 1990) or a standard population. The VSI method requires the selection of a 'standard population' for quaternary horses, this standard population comes from the Middle Pleistocene (01s 13 or 15) site of Mosbach near Wiesbaden, Germany. The VSI distribution of the Mosbach horses approaches a normal distribution, rellecting a high level of biological homogeneity.
Isotope analyses on a series of Middle and Late Pleistocene horse bone from the study area conducted by Elisabeth Stephan (1999) provide a paleoenvironmental framework in which to evaluate changes in horse body size. Oxygen isotope ratios in animal bone tissue depend on the global oxygen cycle. The oxygen isotope ratios ('80/160i)n the drinking water of mammals change as a function of temperature, humidity, and other related effects. This ratio is changed by fractionation in the body in ways characteristic for each species and independent of environmental temperature until the oxygen isotopes are incorporated in the hydroxyapatite of the bones. Investigations of recent equids and other mammals show a linear relationship between the oxygen isotope ratios of drinking water and bone phosphate (Longinelli 1984; Luz & Kolodny 1985). Because of this linear relationship and the stability of phosphate, the 6180 value of animal bones reflect the isotopic composition of the drinking water at the time of the formation of the hydroxyapatite. Therefore, it is possible to obtain specific information about palaeoenvironments and climate on the basis of the oxygen isotope ratios of bone phosphate (Ayliffe & Chivas 1990; Luz et al. 1990). The data from isotope analyses allow us to predict morphological changes in body size of horses during the Pleistocene: if low levels of oxygen isotope ratios indicate cold climate, the mean VSI value of the horses was low and therefore the animals were small in size. In contrast, if high levels of oxygen isotope ratios indicate warm climate, the mean VSI value was also high and horses were larger in size. These predictions are indeed reflected in the osteometric data analysed in this study, in that the smallest horses are found in the coldest sites and the largest horses from warm sites. This correlation is not only obvious in a single unit of time, but it is repeated throughout the glacial cycles.
Examples of small horses from the coldest periods of the glaciation come from Salzgitter-Lebenstedt, a site recently dated with the ESR method to the maximum cold stage of the RissISaale (01s 6) (Dirk Hoffmann, in preparation) and from the site of Wiesbaden-Igstadt, dated to the maximum cold stage of the WurrnIWeichsel (01s 4). Largebodied horses came from interglacial sites in the Cromerian and Eemian but the optimum in temperature is not the optimum in body size: horses are largest before or after a climatic optimum, in cases where the temperature is somewhat lower and vegetation changes from oak woodland to a more open form of grassland with steppic elements. This ecosystem provides horses with the best resources for growth, reproduction, and survival, even during winter when plant food is not always available in central Europe.
Temperature is only one component in the complex combination of elements that make up climate and is not the only factor that influenced the body size of horse. The oxygen isotope ratios in horse bone show that in some cases, the largest and smallest horses lived under similar conditions of temperature. On the other hand, comparison of results from different geographical regions shows that larger horses are found in central Europe and smaller individuals in western Europe, a second indication that temperature is not the only factor that influences body size in horses.
Osteometrics of horse bones from Gravettian deposits between Spain and Poland indicate an increase in body size from western to central Europe. The Spanish horses are smaller than those from France, which are then smaller than horses from those in Gravettian sites from Czech Republic and Poland. The largest horses were recovered from excavations at the Jaskinia Raj site in eastern Poland. This variation in body size is also discernable in reindeer (Rangifer tarandus, Weinstock 2000). Weinstock (2000: 101) discusses a correlation between body size and continentality: "...more continental areas may suffer from more rigorous winters, with higher mortality rates during this season - and thus less intraspecific competition during the growth season - than more oceanic areas. Continentality could be directly proportional to the 'width' of the winter bottleneck." In contrast to the situation in the Gravettian, when body size of caballine horses increases from the west to east, body size does not increase during the Magdalenian. The very cold and dry climate of the glacial maximum changed into better climatic conditions in oceanic areas earlier than continental regions, which is the reason that Magdalenian horses in Spain, living in warmer habitats with good food resources, are slightly larger than their relatives in Switzerland and Germany.
Lastly, this dissertation discusses the problem of nomenclature and taxonomy of the Genus Equus. According to Siewing (1980:811): "Individuals, who correspond in all important characteristics with each other and with their descendants, are one species." This definition of species incorporates populations that are geographically and temporally separated. In this sense, all species of Pleistocene caballine horses are the continuation of one species of wild horse, called Equus ferus Pallas 1775, however it might be sensible to assign subspecies for different chronological periods that accounts for the extensive variation in body size over time.