Trait analysis in a population of the Greater Butterfly-orchid observed through a 16-year period

Abstract

A large English population of the temperate tuberous Greater Butterfly-orchid,Platanthera chlorantha, was monitored through a 16-year period. Each June the number of flowering plants was counted and 60 flowering plants were measuredin situfor four morphological traits, selected for both ease of measurement and their contrasting contributions to the life history of the species. Trait data were tested annually in pairwise combinations for individual plants, before mean values throughout the study period were regressed and cross-correlated against each other and against local data for four meteorological parameters. Labellar spur length proved to be more constrained than either flower number or stem height, and rarely yielded statistically significant correlations with other traits, whereas the three remaining traits reliably showed modest but significant correlations. Mean values and coefficients of variation differed only modestly among years and showed few of any meaningful trends. Spring rainfall and insolation had no detectable effect on traits of plants flowering that June; instead, they impacted on trait expression during the following year, presumably as a result of differential resourcing of replacement tubers formed during the previous year. High spring rainfall in year t–1 increased leaf area and stem height in year t, whereas the widely fluctuating number of flowering plants was highest in years immediately following those characterised by relatively dry and/or sunny springs. The “decision” to flower is taken during the previous summer, though it may be modified through winter/spring abortion of above-ground organs. The proportion of the population electing to flower is the only measured parameter that impacts significantly on annual reproductive output, emphasising the under-rated difficulty of evolving through directional selection. Any attempt to predict the behaviour of plant species in response to climate change must integrate information on demography with that on life history, habitat preference and intimate symbioses.