Silviculture practices interact with multiple sources of variability to influence regeneration trends in northern hardwood forests. There is uncertainty whether low-intensity selection harvesting techniques will result in desirable tree regeneration. Our research is part of a long-term study that tests the hypothesis that a silvicultural approach called "structural complexity enhancement" (SCE) can promote accelerated development of late-successional forest structure and functions. Our objective is to understand the regeneration dynamics following three uneven-aged forestry treatments modified to increase postharvest structural retention: single-tree selection, group selection, and SCE. In terms of regeneration densities and composition, how do light availability, competition, seedbad, and herbivory interact with overstory treatment effects? To explore these relationships, manipulations and controls were replicated across 2-hectare treatment units at two sites in Vermont, USA. Forest inventory data were collected pre-harvest and 13 years post-harvest. We used linear mixed effects models with repeated measures to evaluate the effects of treatment on seedling and sapling abundances and diversity (Shannon-Weiner H'). Multivariate analyses evaluated the relative predictive strength of treatment versus alternative sources of ecological variability. Thirteen-years post-harvest, the harvested treatments were all successful in recruiting a sapling class with a significantly higher mean than the control. However, in all of the treatments prolific beech regeneration dominated the understory in patches. Seedling densities exhibited pulses of recruitment and mortality with a significant positive treatment effect on all harvested treatments in the first four years post-harvest. Seedling diversity was maintained, while sapling diversity was negatively influenced by herbivory (deer and moose browse) and leaf litter substrate. Multivariate analyses suggest that while treatment had a dominant effect, other factors were strongly influential in driving regeneration responses. Results indicate variants of uneven-aged systems that retain or enhance stand structural complexity, including old-growth characteristics, generally regenerate at adequate and desirable densities depending on site conditions.