Adapting insect caste concepts to the demands of bio-ontologies.

AbstractTerminological and conceptual ambiguity surrounding core entities such as colonies and castes continues to hinder comparative analysis, data integration, and ontological interoperability in social insect research. While extensive work has addressed the descriptive and functional aspects of social organization, less attention has been paid to the ontological status of these entities and the consequences this has for formal representation in bio-ontologies. In this study, I provide a realist ontological analysis of colonies and castes grounded in distinctions between natural units, natural kinds, and realizable entities. I argue that insect colonies constitute bona fide biological individuals (natural units) demarcated through relations of causal unity and historical continuity, and therefore suitable for treatment as unified causal systems within ontology frameworks. By contrast, I show that castes fail to satisfy the criteria for either natural units or natural kinds, due to their lack of independent causal cohesion, developmental plasticity, temporal variability, and context sensitivity. Treating castes as autonomous entities or essentialized kinds introduces systematic instability and misrepresentation in bio-ontological models. To address these issues, I propose modeling castes as realizable entities, specifically as roles instantiated in individual organ-isms and realized in biological processes under colony-regulated developmental and ecological conditions. This approach accommodates morphological canalization, phenotypic plasticity, temporal polyethism, and cross-taxon variability within a unified framework, while remaining consistent with the Basic Formal Ontology (BFO). My analysis clarifies longstanding conceptual ambiguities in the representation of social insect organization and provides a flexible, biologically faithful foundation for ontology design, supporting robust data interoperability and future empirical refinement.