We investigated the influence of population density on juvenile body mass in two contrasting roe deer populations, in Sweden (Bogesund) and France (Chizé), in which density was monitored for ≥15 years. We investigated the effect of population density and climatic conditions on cohort performance. We predicted that: (1) body mass of growing fawns should be sensitive to environmental changes, showing marked between-year variation (i.e., cohort effects), (2) fawns in the less productive (weakly seasonal, weakly predictable summer weather) habitat of Chizé should show stronger density-dependent responses due to more severe food competition during summer than fawns in the more productive (markedly seasonal, moderately predictable summer weather) habitat of Bogesund, and (3) fawns at Bogesund should be heavier both in absolute terms and relative to their size than their conspecifics in Chizé due to a higher degree of fat accumulation in northern environments. In both study sites we found marked cohort variation and clear effects of density, with body mass varying by as much as 29% over years. While neither summer nor winter climate influenced fawn body mass at Bogesund, fawns tended to be lighter after summers with high temperatures at Chizé. In addition, fawns were heavier after acorn mast years experienced in utero at Bogesund. As expected, the strength of the density-dependent response of fawn body mass was greater at Chizé than at Bogesund. For a given density, male fawns were consistently heavier than females in both sites. Lastly, both sexes at Bogesund had higher absolute body mass and were larger for a given body size than in Chizé. Our results clearly demonstrate that absolute density is a poor predictor of roe deer performance and supports the view that habitat quality has an overwhelming importance for determining fawn body mass in roe deer populations