Experimental Induction of Localized Reproduction in a Marine Bryozoan

The control of reproduction and growth rate within colonies of marine invertebrates is often conditional and can be very localized. We demonstrate experimentally large and localized shifts in the timing and pattern of reproduction within colonies of a temperate bryozoan (Membranipora membranacea) in response to simulated damage by predators and crowding by conspecifics. In these protandrously hermaphrodite colonies, zooids on the damaged side of a colony reproduced sooner than in unmanipulated regions of the same colony. To examine the influence of the pattern of edge damage on localized reproduction, we damaged the perimeter of circular colonies in two patterns: (1) a continuous half of the edge was trimmed (1/2-Damage) and (2) the edge was trimmed in four alternating one-eighth sections (4/8-Damage). The 1/2-damage treatment triggered localized reproduction, and the more localized four-eighths-damage did not. These experiments demonstrate that the configuration rather than the total amount of edge damage affects the localization of reproduction. In parallel experiments, conspecifics were allowed to crowd half the perimeter of experimental colonies. This treatment also resulted in localized and accelerated reproduction near the contact zone adjacent to a conspecific. Not only do patterns of reproduction change in crowded or damaged colonies, but obstructed colonies also compensate for reduced growth at an obstructed edge by extending the adjacent unobstructed perimeter edge at a greater rate. One model to explain the sort of local cues governing the observed shifts in reproduction and growth rate is a source-sink model. A similar mechanism is proposed to underly growth and reproductive allocation in plants. We suggest that the balance between growth and onset of reproduction in zooids is determined by the rate of translocate moving through each zooid. The rate of translocate movement through zooids is, in turn, affected by the strength and proximity of sinks for that translocate, such as the growing edge of the colony. We propose a simple source-sink model of carbon flow to explain our experimental results. This model would account for the induction of localized reproduction in 1/2 damaged colonies and the lack of localization in 4/8 damaged colonies.