Abstract
In this study, we try to combine traditional archaeozoological biometry, based on outer bone measurements, with stable isotope analyses of bone collagen. Right from the start of archaeozoological research in Central and Western Europe, the important size variability in Roman domestic cattle has puzzled scholars. According to an established view, these differences in bone size are attributed either to the simultaneous presence of different types or even breeds or to the result of crossbreeding of smaller, native, and larger Roman cattle. Likewise, the episodic import of large-sized animals has been considered. First, we selected thirty proximal phalanges of cattle from three sites including five archaeological contexts from eastern Austria (Roman provinces of Noricum and Pannonia). The bone sample comprised the whole hitherto observed metric variability in Roman provincial cattle, and we tried to include minimal and maximal specimens. The results from stable isotope analyses (δ(15)N, δ(13)C, δ(18)O, δ(2)H) carried out on thirty proximal phalanges indicated that isotope signals were rather site-specific and, generally, not related to bone size. Therefore, we conclude that at least in the area investigated, small and large cattle types were raised and herded in the same areas and not spatially separated. There are, however, uncertain indicators of intra-site differences in isotope signals related to bone size, which should be checked on much larger sample sets.