Back in black: recolonization of surface habitats by cave isopods challenges the evolutionary dead-end paradigm

Life-history specialization is hypothesized to constrain evolutionary reversibility, potentially leading to evolutionary dead ends. Subterranean organisms exemplify it through extreme adaptations, such as depigmentation. However, ecotones, like springs (interfaces between caves and surface freshwaters), may offer opportunities for specialized lineages to retain phenotypic plasticity and evolutionary potential. After the discovery of spring populations showing pigmentation in freshwater cave-dwelling isopods (Monolistra pavani), we combined population genomics, long-term field surveys, spectroscopy and demographic modelling to determine the taxonomic identity, persistence, demographic history, and genetic basis of pigmentation. Phylogeny confirmed that spring populations belong to M. pavani but are distinct from nearby cave populations. Demography supported a divergence between cave and spring populations 43,000 generations ago, indicating long-term isolation rather than accidental occurrence. Spring populations exhibited higher genetic diversity and effective population sizes than cave populations and maintained pigmented, partially pigmented, and depigmented individuals over multiple years. Spectroscopy detected melanin-consistent absorbance profiles, while genomic analyses identified an association between pigmentation and loci showing similarity to pigmentation-related genes, although no strong genome-wide signatures were detected and functional validation is still required. These findings suggest that endogenous melanogenic pathways remain partially functional in this cave lineage and may be reactivated through genetic and environmentally induced mechanisms following surface recolonization. Caves may have acted as glacial refugia, whereas spring habitats became available after ice retreat. We provide the first genetic evidence of stable surface populations in cave-dwelling isopods and show that specialization does not preclude evolutionary flexibility or the potential to exploit novel ecological opportunities.