• alpine plants;
  • Campanula thyrsoides;
  • elasticity analysis;
  • European Alps;
  • evolutionary stable strategy;
  • flowering threshold;
  • life-history evolution;
  • semelparity;
  • size-structured populations;
  • stable size distribution


  • 1
    The evolution of flowering strategies (when and at what size to flower) in monocarpic perennials is determined by balancing current reproduction with expected future reproduction, and these are largely determined by size-specific patterns of growth and survival. However, because of the difficulty in following long-lived individuals throughout their lives, this theory has largely been tested using short-lived species (< 5 years).
  • 2
    Here, we tested this theory using the long-lived monocarpic perennial Campanula thyrsoides which can live up to 16 years. We used a novel approach that combined permanent plot and herb chronology data from a 3-year field study to parameterize and validate integral projection models (IPMs).
  • 3
    Similar to other monocarpic species, the rosette leaves of C. thyrsoides wither over winter and so size cannot be measured in the year of flowering. We therefore extended the existing IPM framework to incorporate an additional time delay that arises because flowering demography must be predicted from rosette size in the year before flowering.
  • 4
    We found that all main demographic functions (growth, survival probability, flowering probability and fecundity) were strongly size-dependent and there was a pronounced threshold size of flowering. There was good agreement between the predicted distribution of flowering ages obtained from the IPMs and that estimated in the field. Mostly, there was good agreement between the IPM predictions and the direct quantitative field measurements regarding the demographic parameters λ, R0 and T. We therefore conclude that the model captures the main demographic features of the field populations.
  • 5
    Elasticity analysis indicated that changes in the survival and growth function had the largest effect (c. 80%) on λ and this was considerably larger than in short-lived monocarps. We found only weak selection pressure operating on the observed flowering strategy which was close to the predicted evolutionary stable strategy.
  • 6
    Synthesis. The extended IPM accurately described the demography of a long-lived monocarpic perennial using data collected over a relatively short period. We could show that the evolution of flowering strategies in short- and long-lived monocarps seem to follow the same general rules but with a longevity-related emphasis on survival over fecundity.