Variability in life history traits and structural diversity of commercially exploited fishes in response to stress can impact their population dynamics and sustainability. Using data from a fishery dependent sampling program from 1978 to 2011, we evaluated temporal variability of size and growth of adult Atlantic herring (Clupea harengus) in the Gulf of Maine. We then developed and tested the hypotheses on the links of such temporal changes to population density and environmental factors and found decreases in size and growth potential. Generalized additive models found that density dependence was the main driver of such changes over sea surface temperature and salinity. Our results highlight the importance of density dependent processes in regulating growth and population size structure for Atlantic herring in the Gulf of Maine. 1 | INTRODUCTION The structural diversity of life history traits (e.g., size and age) within and among fish populations is an important consideration that may impact the population dynamics and productivity of managed fish stocks (Hilborn & Walters, 1992; Ricker, 1975). Large variations in size and age diversity have been demonstrated among populations, as well as within populations among years, and can be related to various biotic and abiotic factors (e.g., population density, prey availability, temperature and salinity) (Blaxter, 1992; Chen & Mello, 1999; Hare et al., 2016; Yan et al., 2012). Additionally, truncated size and age distributions can indicate a stressed stock due to overfishing (Brunel & Piet, 2012). As such, an understanding of the complex relationship between fish populations and their environment is important for investigating the potential impact of environmental changes on fish population dynamics (Fischer et al., 2013). An inverse relationship between spawning stock biomass (SSB) and growth, as well as a decline in asymptotic length per unit biomass density, has been described (Lorenzen & Enberg, 2002; Walters & Post, 1993). The parameters of the Von Bertalanffy growth function (VBGF) maximum attainable length (L ∞), maximum attainable weight (W ∞) and the Brody growth rate (K) are influenced by density-dependent growth.