The kinematics of phototaxis in larvae of the ascidian Aplidium constellatum

Although phototaxis has an important influence on the vertical distribution and settlement of marine invertebrate larvae, few studies have explored the mechanisms of taxis in larvae at the organismal level. We examined how phototaxis changes over ontogeny in larvae of the ascidian Aplidium constellatum and experimentally tested hypotheses about the kinematics of oriented swimming. By video recording their swimming movements at regular intervals over their ontogeny, we found that larvae switched from positive to negative phototaxis. We tested hypotheses about the kinematics of phototaxis by recording the three-dimensionalmovement of larvae in response to a change in the direction of illumination and by tracking the tail motion of tethered larvae in response to sinusoidal changes in light intensity. Larvae swimming with negative phototaxis changed their rate of rotation about their antero-posterior and dorso-ventral axes in response to a change in the direction of illumination. These changes in the rates of rotation caused the axis of the helical trajectory to orient away from the light source. Tethered larvae oscillated their tails at a constant tail beat frequency and with a slow periodicity that was correlated with the stimulus frequency. These findings suggest that ascidian larvae orient by changing tail motion in proportion to perceived changes in light intensity. This method of orientation predicts that larvae achieve the switch from positive to negative phototaxis by changing the delay of their kinematic response to changes in perceived light intensity. Introduction A broad diversity of marine invertebrate larvae are capable of oriented swimming toward or away from vector cues such as gravity and light (for reviews see Thorson 1964; Young and Chia 1987; Forward 1988; Young 1995). Such tactic swimming influences the vertical distribution of larvae (e.g. Crisp and Ghobashy 1971; Forward 1984; Forward 1985; Young 1986; Olson and McPherson 1987; Stoner 1992; Shanks 1995b) and their selection of a microhabitat for settlement (e.g. Thorson 1964; van Duyl et al. 1981; Durante 1991; Svane and Dolmer 1995). Despite widespread interest in understanding how changes in larval behavior affect dispersal (e.g. Scheltema 1986; Davis and Butler 1989; Stoner 1990; Shanks 1995a), few studies have considered the kinematic mechanisms of tactic behavior at the organismal level. An exception is Mast (1921), which investigated phototaxis in the ascidian Aplidium constellatum and proposed a model for the kinematic response to a light stimulus that facilitates orientation. In the present study, we quantified the kinematics of phototaxis over ontogeny in A. constellatum and tested Mast’s (1921) model by experimentation on individual larvae. Like other species of compound ascidian, larvae of A. constellatum switch from positive to negative phototaxis during a brief dispersal phase. Upon hatching, these 2 mm long larvae initially swim away from their parent colony by moving toward light (Grave and Woodbridge 1924). Shortly after hatching, larvae switch to negative phototaxis, which is how they swim for the rest of the 10 to 100 min larval stage. Qualitative observations suggest that ascidian larval swimming becomes slower, more intermittent, and less directed with age (e.g. Trididemnum solidum, Bak et al. 1981; Podoclavella moluccensis, Davis and Butler 1989; Chelyosoma productum, Young and Braithwaite 1980). We tested whether larvae of A. constellatum exhibit these changes as they age by measuring swimming trajectories of individual larvae at hatching and just prior to settlement. Marine Biology (2003) 142: 173–184 DOI 10.1007/s00227-002-0929-z Communicated by P.W. Sammarco, Chauvin M.J. McHenry (&) Æ J.A. Strother Department of Integrative Biology, University of California, Berkeley, CA 94720, USA Present address: M.J. McHenry The Museum of Comparative Zoology, Harvard University, Cambridge, MA 02138, USA, e-mail: [email protected] According to Mast (1921), a larva is capable of phototaxis because it moves its tail laterally with a largeamplitude tail beat in response to a particular rate of change in perceived light intensity (Fig. 1). Between these tail flicks, body rotation occurs about the anteroposterior axis (rotation known as roll) and the larva follows a straight trajectory. Because the larva possesses a single ocellus that is directed toward one side of the body, roll exposes the ocellus to temporal oscillations in light intensity. These oscillations are low in amplitude when the body is aligned with a light source, but large when the body moves at an angle with respect to the source. In response to a large-amplitude change in intensity, the larva flicks its tail and changes the direction of swimming. During tail flicks, the body rotates strictly around its dorso-ventral axis (rotation known as yaw). Mast (1921) suggested that positive phototaxis is achieved by this mechanism when the tail flicks toward the abocular side of the body in response to a decrease in light intensity and toward the ocular side in response to an increase. Later in ontogeny, this response is reversed to achieve negative phototaxis. We evaluated Mast’s (1921) model in A. constellatum by testing three hypotheses. (1) By video recording the movement of larvae in three dimensions, we tested whether larvae rotate strictly by roll and thereby swim straight when not turning. (2) By rapidly changing the direction of illumination while larvae are swimming, we tested whether their bodies rotate primarily by yaw to cause a sharp change in direction during turning maneuvers. (3) By tethering larvae and stimulating their ocellus with sinusoidal changes in light intensity, we tested whether tail flicking occurs in response to changes in light intensity. Fig. 1A–E The mechanism of phototaxis proposed by Mast (1921). A In an experimental situation, this negatively phototactic larva begins by swimming up and away from a source of illumination that is beneath it (the ‘‘under light’’). At a, the under light turns off and the side light turns on. In response, the larva flicks its tail toward the right and ocular side at b, which directs swimming toward an oblique angle to the illumination. After one half of a body rotation, the tail flicks at c, toward the right, which is then the abocular side of the body. This directs swimming nearly in the opposite direction of light. B During periods of straight swimming, the larva rotates around the antero-posterior axis of the body at the roll rate, xroll, and during turns, the body rotates around the dorso-ventral axis at the yaw rate, xyaw. CWith the change in light direction at a, the perceived light intensity oscillates because of the rotation of the body. The amplitude of these oscillations decreases, after c, as the body is directed away from the side light. D The yaw rate remains at a value of zero during periods of straight swimming but changes during the turns stimulated at b and c. At b, the body turns to its left and therefore has a positive value for yaw rate. At c, the body turns to its right, which requires a negative yaw rate. E The roll rate remains at a steady value during straight swimming but decreases to zero during the turns stimulated at b and c 174