Genotypic tradeoffs along environmental gradients help maintain diversity in functional traits in the wild and limit the range of suitable environments for each genotype in tree breeding programmes. Little is still known of the capacity of abiotic and biotic soil variation to generate marked shifts in genotypic performance ranks. We examined the potential of belowground soil factors to bring about genotype × environment interactions in a tree species by combining data from extant field trials and a new seedling-based progeny experiment. We first analysed genotypic growth patterns of Norway spruce (Picea abies) at two field trial locations, a native forest site and a former agricultural field, that exhibited biotic and abiotic soil variation. While we found significant genotype × location interactions in growth, we also observed positive between-location genotypic correlations, indicating similar genotypic rank orders across divergent soil types. Contributing to the genotype × environment interactions, differences in age at the time of growth measurements may explain why genetic variances nevertheless differed between the locations. The subsequent progeny experiment with soil and seeds collected at the trial locations enabled controlled treatments in a growth chamber that tested the capacity of between-location soil variation to induce genotype × environment interactions in seedling traits. The progeny experiment revealed that the soil treatment had major effects on averages in all 14 shoot and root functional traits, with four groups of correlated traits (e.g., estimates of shoot and root system size) identified by a principal component analysis. Seed collection location affected only few traits, and the more southern agricultural field trial yielded slightly larger seedlings with delayed phenology. Yet, despite significant genetic variation, no seedling trait manifested genotype × soil treatment interactions, which may be due to the soil treatments not mirroring spatial heterogeneity of the soils at the trial locations, or to our limited subsample of ten genotypes in the progeny experiment. Taken together, our results on adult trees and seedlings indicate that overall tree growth is impacted by variation in belowground environmental factors, but further research with more comprehensive sampling is needed to determine whether they have potential to generate location-specific patterns of genotypic performance in economically valuable tree species.