Feeding rates in the chaetognath Sagitta elegans:: effects of prey size, prey swimming behaviour and small-scale turbulence

The gut contents of Sagitta elegans were sampled twice daily (noon and midnight) during 9 days in October at an anchor station in the northern North Sea. Observations of the ambient prey field and of turbulent dissipation rates were collected simultaneously. The average number of prey per chaetognath was among the highest ever recorded, 0.57 +/- 0.10. Total gut content was independent of ambient prey concentration, suggesting that feeding rate was saturated. Clearance rates were estimated from gut contents and ambient prey concentrations and a literature-based estimate of digestion time. The clearance rate to prey size showed a dome-shaped relationship. The maximum clearance rates, about 100-300 ml h(-1), were observed for prey sizes corresponding to 6-10% of Sagitta length. Clearance rates varied not only with prey size, but also with prey type. For example, copepod males were cleared at rates up to an order of magnitude higher than similarly sized females, probably owing to differences in swimming behaviour. Sagitta elegans is an ambush predator that perceives its prey by hydromechanical signals. Faster swimming prey generates stronger signals and is, hence, perceived at longer distances. We develop a simple prey encounter rate model by describing the swimming prey as a 'force dipole' and assuming that a critical signal strength is required to elicit an attack. By fitting the model to the observations, a critical signal strength of 10(-2) cm s(-1) is estimated; this is very similar to estimates for copepods that also perceive prey by mechanoreceptory setae. Gut contents were independent of turbulent dissipation rate. Because feeding rates were saturated, we did not expect to see positive effects of turbulence. However, the strong wind-generated turbulent dissipation rates observed during the study (10(-3)-10(-1) cm(2) s(-3) in the upper mixed layer) could lead to negative effects by interfering with prey perception. At a dissipation rate of 10(-2) cm(2) s(-3) a 10-mm long S. elegans would experience fluid signals of order 0.3 cm s(-1) due to turbulence, 30 times stronger than the signal strength from the prey. It is, therefore, suggested that S. elegans is able to separate prey signals from turbulence signals due to their different spatial characteristics.