Growing competitive or tolerant? : significance of apical dominance in the overcompensating herb <em>Gentianella campestris</em>

Abstract

<div lang="en" class="abs"><p class="abs_header">Abstract</p><p>As a compensatory response to herbivory, plants may branch vigorously when the growth of dormant meristems is triggered by shoot damage. Undamaged plants, on the other hand, often restrain branching, and this limitation on growth can be considered a cost of tolerance to herbivory. Restrained branching is caused by apical dominance and may, alternatively, be associated with fitness benefits in competitive environments that favor fast vertical growth. To test these hypotheses regarding selection for restrained branching, we compared the performance of two subspecies of the biennial grassland herb <em>Gentianella campestris</em>; the tall, apically dominant ssp. <em>campestris</em> and the short, multi‐stemmed ssp. <em>islandica</em>, which shows reduced apical dominance. For both subspecies, we manipulated the height of surrounding vegetation (competition) and damage intensity in grasslands of differing productivity (high, medium, low), and examined population growth rates using matrix population models combined with life table response experiments. In the absence of damage, ssp. <em>campestris</em> exhibited a higher population growth rate than ssp. <em>islandica</em> in the tallest vegetation, however with the growth rate still being below one. In the medium and low productivity environments where the vegetation was shorter, the population growth rate of ssp. <em>islandica</em> was considerably higher than that of ssp. <em>campestris</em> as long as no more than about 50% of the plants were damaged. When plants were damaged, the apically dominant ssp. <em>campestris</em> showed a positive population growth rate (λ > 1) and often overcompensatory seed production in all productivity levels, while ssp. <em>islandica</em> showed no compensation and therefore the population was predicted to decline (λ < 1). We conclude that restrained branching in <em>Gentianella</em> cannot be selected for by competition alone, but that episodes of apical damage are required to maintain the trait. Furthermore, because of the costs of restrained branching, apical dominance should be selected against in grasslands where competition and disturbance are low.</p></div>