Teleost fishes typically possess a homocercal caudal fin with a symmetrical morphology in which the dorsal and ventral lobes

a symmetrical morphology in which the dorsal and ventral lobes of the tail fin are the same size and project posteriorly beyond the axis of the vertebral column. In contrast, most elasmobranch fishes and some primitive actinopterygian fishes possess a heterocercal caudal fin with an asymmetrical morphology in which the ventral lobe of the fin is smaller than the dorsal lobe and the vertebral column extends into the dorsal lobe of the fin. For over a century, it has been assumed that these morphological differences generate functional differences because the two types of tail should deflect water differently during steady swimming. Early German morphologists (Schulze, 1894; Ahlborn, 1896) proposed that the asymmetrical morphology of the heterocercal tail creates asymmetrical water movement during the tail beat. According to the classic theory of heterocercal tail function, during the tail beat, the large, stiff upper lobe leads the smaller, flexible lower lobe, creating dorso–ventrally asymmetrical movements as the tail beats from side to side. The predicted consequence of this movement is that water is deflected ventrally as well as posteriorly during the tail beat, which creates an upward force, or lift, on the tail (Fig. 1). Support for this theory accumulated when experiments conducted using models (Grove and Newell, 1936; Affleck, 1950) and truncated elasmobranch caudal fins (Alexander, 1965) demonstrated that a heterocercal tail oscillated laterally produces upward lift under certain experimental conditions. More recently, Ferry and Lauder (1996) conducted the first comprehensive study of tail kinematics during steady swimming in a living elasmobranch (a leopard shark, Triakis semifasciata). Their study provides both three-dimensional kinematic data and flow visualization data that support the classic theory of heterocercal tail function. Therefore, this model appears to be accurate for some shark species, although it may not apply to all fishes that possess heterocercal tails (Lauder, 1999). 2433 The Journal of Experimental Biology 202, 2433–2447 (1999) Printed in Great Britain © The Company of Biologists Limited 1999 JEB2234