Inertia as a ‘safe harbour’: do fish larvae increase length growth to escape viscous drag?

Introduction Fish larvae differ from adults in their interaction with the physical environment. They hatch at a size small enough to experience water as a mainly viscous medium. Their external morphology and development are adapted accordingly (Osse and Drost, 1989). The boundary layer around the gill filaments would be too thick to allow efficient gas exchange. Only as the larvae grow bigger do the gills become functional in breathing (Osse, 1989). Viscosity also affects swimming. The flow regime around an organism depends on its size and swimming speed and can be characterized by the dimensionless Reynolds number (the ratio of inertial to viscous forces). Fish larvae hatch in the viscous flow regime and reach the inertial regime around first feeding. Experimental evidence (Vlymen, 1974) and theoretical considerations (Weihs, 1980) indicate a change of swimming style as the fish larva passes from a viscous to an inertial flow regime. The fish larva responds to the changing regime, not only by changing its swimming style, but also by adjusting its growth pattern. Fish larvae delay the gut development to keep their body flexible enough to achieve the high body wave amplitudes suitable for swimming in a more viscous regime (Osse, 1990). In spite of the obvious adjustments to the viscous flow regime, small fish larvae still experience mortality close to that of eggs (Pepin, 1991). The mortality drops drastically as the larvae grow and swimming performance improves (Bailey and Batty, 1984; Pepin, 1991). This hints at the possibility that fish larvae, instead of adapting optimally to the viscous flow regime, invest into outgrowing viscosity as quickly as possible ('safe harbour' hypothesis of Shine, 1978). To escape the viscous flow regime, a fish larva can increase its swimming speed or its size. Webb and Weihs (1986) suggest that fish larvae escape the viscous flow regime by increasing their body length. Fuiman (1983) and Osse (1990) found some indication to support this idea in the fish larvae's allometric growth patterns. This review tries to test the hypothesis that fish larvae increase their growth in length to escape the viscous flow regime. It does so by looking at the weight-length relationship of larvae from 23 species. Fish larvae should increase their growth in length until they reach the inertial flow regime at some critical length. The allometfic exponent of the weight-length relationship below this critical length should be lower