Changes in land use have been shown to have profound effects on forest plant community structure and diversity. Dispersal limitation has been invoked as a major factor hampering colonization of forest plant species, while seed-sowing experiments and performance observations have provided some evidence for recruitment limitation determining forest plant distribution in post-agricultural forests. However, most of these studies were relatively short-term, and very few studies have investigated long-term growth rates of populations occurring in recent and ancient forests. In this study, matrix models using demographic data collected for four consecutive years were used to study the effect of forest age on population dynamics of the temperate forest herb Primula elatior. A life table response experiment (LTRE) and elasticity analysis were used to analyze the effect of forest age on population growth rate (lambda) and to decompose the effect of forest age on lambda into contributions from each matrix element. Population growth increased logarithmically with increasing forest age. Bootstrap analyses showed that populations located in very recent forests (< 50-years-old) had growth rates that were significantly < 1, whereas populations located in forests > 150-years-old had growth rates that were significantly > 1. Summed elasticities for individual growth significantly decreased with increasing forest age, whereas summed elasticities for survival and fertility significantly increased with increasing forest age. The LTRE analysis showed that the increase in lambda with increasing forest age was mainly due to increased seedling and juvenile growth and increased juvenile and adult survival. Our results indicate that past agricultural land use has long-lasting effects on the demography of forest herbs and may provide an additional mechanistic explanation for the poor colonization capacity of many forest herbs in post-agricultural forests.