Seed limitation restricts population growth in shaded populations of a perennial woodland orchid.

Seed production and seedling recruitment are thought to be of minor importance in determining population dynamics and long-term viability in long-lived perennial plants. Seed addition experiments, on the other hand, have amply shown that supplemental addition of seeds almost always, irrespective of longevity, results in increased seedling recruitment. Any change in the environment that affects fruit and seed production can thus be expected to affect seedling recruitment, but the extent to which increased fruit and seed production affect overall population dynamics remains relatively unknown. In this paper, we present demographic data of six populations of the long-lived woodland orchid Orchis purpurea that were monitored for seven consecutive years (2002-2008) occurring in two contrasting light environments. We use a nested life table response experiment (LTRE) at the vital rate level to disentangle the relative contributions of each of six annual transitions, six sites, and two light environments on the population dynamics of this species and to determine vital rate variations that contributed most to variation in population growth rate. Population growth rates (lamda) were significantly higher in the light environment than in the shaded environment (average lamda = 0.9930 and 1.0492 in the shaded and light environment, respectively). The LTRE analysis showed that variation in fecundity and, to a lesser extent, variation in growth made the largest total contributions to variation in lamda, whereas the contributions of variation in survival were almost negligible. Fruit production was two times larger and the net reproductive rate (R0) was approximately six times higher in the light environment than in shaded areas, suggesting that variables related to reproduction are the key drivers of population dynamics of this long-lived orchid species in different light environments. Our results indicate that light is an important factor affecting population dynamics of Orchis purpurea and illustrate that, even in long-lived species, flower and seed production can have important effects on the population dynamics.