Understanding the history of anthropogenic vegetation is often difficult due to the lack of tangible historical evidence. In this study, we examined chronological changes of farmland demarcation trees planted on alluvial plains along the Hijikawa River in southwestern Japan based on species distribution patterns, folk nomenclature, and multiple usage of the trees.The species composition of demarcation trees was investigated at 47 sites in 13 villages. We performed hierarchical clustering using Bray-Curtis measures to detect groups of similar tree composition and permutational multivariate analysis of variance to test whether differences in species composition correspond to village units. To better understand the traditional knowledge of demarcation trees, we conducted interviews with 53 farmers, most of whom were over 60 years old.Clustering resulted in six tree composition groups. The group characterized by the most frequently planted species, Chaenomeles speciosa, dominated around lower reach villages. The group characterized by Euonymus japonicus dominated around middle reach villages, and that characterized by Salix pierotii was mainly located around upper reach villages. Chaenomeles speciosa was always identified with the standard Japanese name boke or similar names. Euonymus japonicus and several other species were also called boke by many farmers. Several elderly farmers stated that C. speciosa was pervasive in upper and middle reach villages in their youth, suggesting the prototypical use of C. speciosa in the study area. In addition, some minor species were likely to have been left after commercial crop production or subsistence use between the late nineteenth and mid-twentieth centuries, including Morus sp. and Celtis sinensis for sericulture, Salix koriyanagi for fiber production, and Gardenia jasminoides for food coloration. The name kōshin bana recorded for E. japonicus suggests that the species' use originated from the folk faiths Kōshin-shinkō and/or Shōmen-Kongō.The composition of demarcation trees in the region has not been stable over time, but instead changed to reflect the local livelihood, industry, and faiths. Despite the lack of tangible historical evidence, the spatial distribution patterns, folk nomenclature, and traditional knowledge of plants can provide clues to trace the chronological background of ecotopes in anthropogenic landscapes.