Evolutionary trajectory of phenological escape in a flowering plant: Mechanistic insights from bidirectional avoidance of butterfly egg-laying pressure.

Phenological escape, whereby species alter the timing of life-history events to avoid seasonal antagonists, is usually analyzed either as a potential evolutionary outcome given current selection coefficients, or as a realized outcome in response to known enemies. We here gain mechanistic insights into the evolutionary trajectory of phenological escape in the brassicaceous herb Cardamine pratensis, by comparing the flowering schedules of two sympatric ecotypes in different stages of a disruptive response to egg-laying pressure imposed by the pierid butterfly Anthocharis cardamines, whose larvae are pre-dispersal seed predators (reducing realized fecundity by ~70%). When the focal point of highest intensity selection (peak egg-laying) occurs early in the flowering schedule, selection for late flowering dependent on reduced egg-laying combined with selection for early flowering dependent on reduced predator survival results in a symmetrical bimodal flowering curve; when the focal point occurs late, an asymmetrical flowering curve results with a large early flowering mode due to selection for reduced egg-laying augmented by selection for infested plants to outrun larval development and dehisce prior to seed-pod consumption. Unequal selection pressures on high and low fecundity ramets, due to asynchronous flowering and morphologically targeted (size-dependent) egg-laying, constrain phenological escape, with bimodal flowering evolving primarily in response to disruptive selection on high fecundity phenotypes. These results emphasize the importance of analyzing variation in selection coefficients among morphological phenotypes over the entire flowering schedule to predict how populations will evolve in response to altered phenologies resulting from climate change.

Multiple temperature effects on phenology and body size in wild butterflies predict a complex response to climate change.

Temperature-induced alterations in phenology and body size are the cumulative outcome of sequential effects impacting development and are universal responses to climate change. Most studies have so far focused on phenological responses to warming in multiple taxa across space and time, or the ontogenetic effects of temperature in the laboratory. I here complement this work by investigating shifts in phenology and body size (wing length) attributable to temperature changes operating over the entire lifespan of the univoltine orange-tip butterfly Anthocharis cardamines in a single wild population over 14 generations. Phenology was affected by temperatures during three discrete periods in the year prior to emergence, corresponding to late larval/early pupal life, the onset of the chilling period required to break pupal diapause, and postdiapause pupal development prior to eclosion. Higher temperatures during late larval/early pupal life and postdiapause pupal development advanced the subsequent emergence of the butterflies, whereas higher temperatures at the onset of the chilling period retarded it. The synchronization of the butterflies' emergence schedule increased when pupae were exposed to milder midwinter temperatures. Wing length increased with warmer temperatures at distinct points in the early and midpupal periods; such direct effects of temperature on body size could complement season length effects in explaining the reversal of the temperature-size rule in univoltine insects. The periods during which temperature affected the phenology of the butterfly only partially overlapped those affecting the first flowering date of its host plants lady's smock (Cardamine pratensis) and garlic mustard (Alliaria petiolata). Observed thermal effects on flowering time, emergence timing, and emergence synchronization indicate that phenological convergence as well as phenological mismatching could affect host-plant availability and diet breadth; thermal effects on body size imply that important population-level processes could be impacted through correlated changes in fecundity and dispersal rate. In general, the combined effects of phenological and ontogenetic responses to temperature changes across the whole lifespan will likely be important in modeling the demographic responses of interacting species to climate change.

Male emergence schedule and dispersal behaviour are modified by mate availability in heterogeneous landscapes: evidence from the orange-tip butterfly.

Protandry (prior emergence of males) in insect populations is usually considered to be the result of natural selection acting directly on eclosion timing. When females are monandrous (mate once), males in high density populations benefit from early emergence in the intense scramble competition for mates. In low density populations, however, scramble competition is reduced or absent, and theoretical models predict that protandry will be less favoured. This raises the question of how males behave in heterogeneous landscapes characterized by high density core populations in a low density continuum. We hypothesized that disadvantaged late emerging males in a core population would disperse to the continuum to find mates. We tested this idea using the protandrous, monandrous, pierid butterfly Anthocharis cardamines (the orange-tip) in a core population in Cheshire, northwest England. Over a six-year period, predicted male fitness (the number of matings a male can expect during his residence time, determined by the daily ratio of virgin females to competing males) consistently declined to <1 in late season. This decline affected a large proportion (∼44%) of males in the population and was strongly associated with decreased male recapture-rates, which we attribute to dispersal to the surrounding continuum. In contrast, reanalysis of mark-release-recapture data from an isolated population in Durham, northeast England, showed that in the absence of a continuum very few males (∼3%) emerged when fitness declined to <1 in late season. Hence the existence of a low density continuum may lead to the evolution of plastic dispersal behaviour in high density core populations, maintaining late emerging males which would otherwise be eliminated by selection. This has important theoretical consequences, since a truncated male emergence curve is a key prediction in game theoretic models of emergence timing which has so far received limited support. Our results have implications for conservation, since plastic dispersal behaviour in response to imperfect emergence timing in core (source) populations could help to maintain sink populations in heterogeneous landscapes which would otherwise be driven to extinction by low mate encounter-rates (Allee effects).