The formation of epicormic sprouts on the boles of trees is a phenomenon that has, until recently, been poorly understood. Renewed interest in the topic in the last two decades has led to significant advances in our knowledge of the subject, especially in regard to bud anatomy, morphology and ontogeny. There exists, however, no comprehensive synthesis of results from different disciplines across genera and geographical areas; this review seeks to fill that void and provide a comprehensive framework capable of guiding future research. A tree's potential for producing epicormic branches is dependent on the number of buds that are produced on a growing shoot, the development of those buds and associated meristems over time and the factors that promote sprout formation or bud death. Based on the descriptions of a limited number of researched species, we were able to describe four different developmental strategies for epicormics based on characteristics of meristem development. Control over epicormic bud dormancy is complex, but it is clear that the traditional view of auxin-mediated dormancy release is incomplete. Genetic control over epicormic development is yet to be empirically proven. Future research should focus on clarifying these physiological and genetic controls of epicormic bud development as well as developing more robust methods for tracking epicormics in ecological and silvicultural studies.

Using inventory data from a long-term silviculture experiment in east-central Maine, spatial models were developed to analyze 28 years (1974–2002) of stand structural dynamics. Differences in spatial pattern, species mingling, height differentiation, and relative stand complexity index (rSCI) were compared among five treatments: commercial clear-cutting, fixed diameter-limit, 5 year single-tree selection, three-stage shelterwood (both with and without precommercial thinning), and unharvested natural areas. Regardless of treatment, regeneration events (whether induced by natural breakup of the overstory or by harvesting) increased aggregation in spatial pattern and reduced species mingling, more so in the commercial clearcut and fixed diameter-limit treatments where hardwood densities were highest. Regular spatial patterns were rare. Height differentiation values for individual trees and stand-level mean rSCI were generally highest in untreated natural areas and 5 year selection treatments, intermediate in commercial clearcut and fixed diameter-limit treatments, and lowest in three-stage shelterwood treatments. After a brief adjustment period, precommercial thinning in a shelterwood treated stand generally increased species mingling, height differentiation, and rSCI. Two untreated natural areas exhibited divergent pathways of structural development. Dynamics in uneven-aged selection treatments more closely resembled that of the untreated natural areas than did the shelterwood, commercial clearcut, or fixed diameter-limit treatments.

Securing desirable regeneration is essential to sustainable forest management, yet failures are common. Detailed seedling measurements from a forest inventory across 24 northern US states were examined for plausible regeneration outcomes following overstory removal. The examination included two fundamental regeneration objectives: 1) stand replacement- securing future forest and 2) species maintenance- securing upper canopy species. Almost half the plots lacked adequate seedlings to regenerate a stand after canopy removal and over half risked compositional shifts. Based on those advance reproduction demographics, regeneration difficulties could occur on two-thirds of the plots examined. The remaining one-third were regeneration-ready. However, compared to historical norms, increased small-tree mortality rates reduces that proportion. Not all forest types rely on advance reproduction and results varied among the forest types examined. Some variability was associated with browsing intensity, as areas of high deer browsing had a lower proportion of regeneration-ready plots.