In bioarchaeology, the biometry of archaeozoological remains is an important component of studies on domestic species and size has been used for multiple purposes from identifying domestication, to track environmental changes or evolution of husbandry practices. The establishment of the Roman Empire has been accompanied by social, political and economic transformations that also reflect in farming practices and animal husbandry. In southern France, biometric variation has already been partially perceived during Roman times, particularly for cattle, but lack chronological accuracy and statistical validation. This paper presents a diachronic analysis of linear measurements of post-cranial bones belonging to the domestic triad (sheep, goat, cattle and suids) in western Gallia Narbonensis (France), between the Roman conquest and Late Antiquity, i.e. from 200 BCE to 400 AD. Biometric data from 64 archaeological sites, excavated and studied over more than 30 years of preventive archaeology, were analysed using a Log Size Index (LSI) approach using time as a continuous variable. The analysis of 5533 measurements first analysed per bone and variable, then separating length, breadth and depth dimensions, revealed different trends, highly influenced by the number of measurements, reflecting allometric differences but also cases of asynchronous evolution. However, these allometries within species are small when compared to interspecies differences. Overall, the size of the four taxa increased from the 2nd century BC to the 2nd century AD. Cattle and pig sizes then decreased from the 2nd century AD and only from the 4th century onwards for goats, while sheep size tends to increase during the 3rd-4th centuries. If the Roman conquest influences the size of the domestic animals, this does not affect the four species in the same way. This potentially reflects differentiated agropastoral strategies for each of the species in the western part of Gallia Narbonensis during the Roman period. This study, which provides a diachronic and cross-species study framework, should be seen as a first step for a more in-depth understanding of micro-regional and socio-economic variation in domestic species morphologies.