Abstract The flatifish Otharichthys sordidus (Pacific Sanddab) is a bottom dweller dependent on camouflage and quick escape response to protect it from danger. In an attempt to determine the nature of the substrates role in the quick liftoff of the sanddab from the ocean bottom, an experimental fish lifted off alternately from a porous and a solid substrate. The sanddab displayed similar displacement, velocity, and acceleration vectors in both x¬ and y-directions when lifting off from either substrate. The similarity of movement on both substrates indicates that a ground effect, or force caused by hydraulic flows redirected by the nearby solid substrate, is of littie conséquence to the sanddab as it lifts off from its substrate. I suggest three possible mechanisms, each independent of substrate, that may contribute to the liftoff of the sanddab. Introduction The sanddab Stharichthys sordidus is a bottom dwelling flatfish common in the waters of Monterey Bay. Like other flatfish, the distance between its pectoral fins is quite short, while it is quite wide between its dorsal and anal fins. Since it has evolved to living on its right side, it thus has à depressiform shape. Its right eye moves during development to the left side of its body so both can be used in observing its surroundings. It is known to eat squid and bottom-dwelling crustaceans (B. Hopkins, personal communication) The dorsal and anal fins of the Pacific sanddab reach nearly from its nose to its caudal fin, their rays just long enough to touch the bottom around the margin of the fish. Its left pectoral fin is large and used for steering and propulsion, while its right is reduced. This fin seems to serve the fish chiefly as as prop, supporting its anterior end off the substrate. It is well camouflaged in its favored sandy bottom habitat, featuring a highly developed system of chromatophores and iridophores on its upper surface for improving its ability to blend with different substrates (A. Deck, personal communication). In addition, sanddabs throw sand on their bodies to further blend with their substrate. The posture of the sanddab has a marked effect on its mode of locomotion. In contrast to fishes that swim freely in the water column, a sanddab commonsy rests for long periods of time on the bottom, moving only as danger or food approaches. Its reduced right pectoral fin has lost its powerful thrusting ability, and its dorsal and anal fins may retard lifting off the substrate by creating a partial seal with the bottom. It also might pay for its highly developed camouflage techniques: the namesake sand it throws on its left side to blend with its substrate may retard its escape when threatened. In contrast, the sanddab may gain locomotory advantages from living on the bottom. Ueland (1984) studied how the sanddab is able to maintain its position in swiftly moving water, avoiding locomotion altogether, by positioning itself in certain postures on the substrate. Interaction between seawater moved by the swimming sanddab and the solid substrate, called ground effect, may allow it to move more efficiently than without an interaction between the water and the substrate. Blake (1979) showed that the mandarin fish (Synchropus picturatus is able to hover more efficiently near a solid substrate due to the ground effect than it could without the substrate. This experiment sought to describe and quantify the effect of the substrate on the movement of the sanddab through comparing the liftoff of a sanddab from a solid substrate, which would allow a ground effect to occur, and a mesh substrate, whose porosity would prevent a ground effect from aiding in liftoff. Materials and Methods All experiments took place using a single sanddab (19 em long, 101.5 g) in a 0.80 by 0.40 m tank of still sea water, changed daily. The fish was caught in the waters of Monterey Bay in the vicinity of Hopkins Marine Station, Pacific Grove, California. The fish ate squid several times a week during the experimental period. A 9 volt electric prod, discharged at the extreme posterior end of the caudal fin of the sanddab, caused it to lift off from the substrate on which it was lying. A tripod-mounted Sony Handycam video camcorder, filming at 30 frames per second and 1/250 sec per frame, recorded a liftoff sequence on each of two different substrates. In each case, the depth of water above the substrate was 0.17 m. No sand was available for the sanddab to bury itself Recording continued until the camera captured the fish lifting off from both substrates in a similar fashion. The solid substrate was a Plexiglas sheet serving as the bottom of the tank. The porous substrate was a sheet of 1 1/2 mesh galvanized chickenwire resting 0.08 m off the bottom of the tank. The sanddab appeared to be equally comfortable on each substrate, although the mesh did tend to fray the dorsal, anal, and caudal fins of the fish slightly over time Playback of the two video sequences on a large-screen television provided large, clear images easily traceable onto acetate film. A digitizing pad, recording at centimeter intervals the position of the curved back of the fish, converted the shape of the fish into Cartesian coordinates. These coordinates provided values for the radius of curvature at each point along the back of the fish using the following formula from Thomas (1953). (14(dy/dx)2)3/2) rho - — - radius of curvature atx d2y/dx? By hanging each of six frozen dead sanddabs from each of three points along the perimeter of their bodies, the intersections of the three vertical lines showed where the approximate center of gravity of the fish was located. The three points of intersection were always within several millimeters of each other. The average ratio of distance from the nose to the center of gravity divided by total length of the fish served to locate the center of gravity of the experimental fish without needing to measure it directly. This center of gravity provided a convenient reference point for measuring the displacement, velocity, and acceleration of the sanddab as it lifted off. Results The typical center of gravity for a sanddab is 388 of the way from its nose to its tail (Figure 1). This point is the reference point for all further calculations of displacement, velocity, and acceleration. The sanddab moved its body in almost identical fashion when lifting off from each substrate (Figure 2). The speed at which it undulated its body was virtually identical between substrates (Figure 3). The shape of the wave, characterized by the radius of curvature at any given point along the wave, appears consistent between liftoffs from each substrate (Figure 4). Thus, direct comparison of displacements, velocities, and accelerations between liftoffs appears to be valid. In each liftoff, the fich held its body off the substrate, balancing as a tripod on the posterior part of the dorsal and anal fins and the right pectoral fin. Once lateral movement had begun, the displacement in the X direction (parallel to the substrate) was slightly greater at any given time for the experiment on the solid substrate than for the experiment on the porous substrate. In contrast, the displacement of the center of gravity in the Y direction ( e perpendicular to the substrate) was indistinguishable between substrates even after the fish began moving rapidly upward (Figure 5). In each case, the center of gravity of the fish moved about 7 em in each direction, for a total of about 10 cm along the diagonal, before beginning to glide about O. 167 s after beginning liftoff. Taken as the first derivative of the displacement versus time graph, the velocity versus time graph also shows little difference between the solid and porous substrates (Figure 6). Similarly, the acceleration versus time graph differs littie between substrates (Figure 7). Discussion It seems reasonable to hypothesize that the bottom habitat of the sanddab might give it a kinematic advantage over more freely swimming fishes. The ground effect provided by the solid ocean floor might allow the undulatory motions of the sanddab during liftoff to move it more quickly than similar motions could when at a distance from a solid substrate. To gain an equal acceleration, fish living in the water column would have to move more water at a higher velocity than flatfish due to the extra dissipation of energy in open water that would not occur against a solid substrate. Since the sanddab appeared to be consistent in both the rate of wave propagation along its body and the type of wave generated when lifting off trom each substrate, it is appropriate to asum attempting' to execute the same liftoff from each type of substrate. Thus any difference in displacement, velocity, or acceleration noted is precumably due to the difference in substrate. Surprisingly, all the data collected point to the conclusion that the fish is equally effective in lifting off from both the porous and solid substrates and thus gains littie through any ground effect it may expérience. The porosity of the substrate thus appears to have little effect on the movement of the sanddab. The sanddab thus appears to be lifting itself off the substrate using a force unaffected by the choice of substrate and great enough to overshadow any ground effect that the fish may also be experiencing. There seem to remain three possibilities, or combinations of possibilities, for the chief method by which the sanddab lifts itself off the substrate. One possibility is that a backwards, horizontal thrust by the sanddab's tail against the surrounding water is mainly responsible for its movement. This force would push the fish horizontally through the water. Depending on the angle of attack-the angle that the bottom of the fish makes with the horizontal substrate-the fish would gain differing amounts of vertical lift from its horizontal thrust. The displacement, velocity, and acceleration of the sanddab would thus not depend directly on the substrate. A ground effect caused by the solid substrate, would still affect lift but the experimental data indicate that such a contribution is negligible Alternatively, the main thrust may be to the rear but not be hydraulic at all. It may be that the dorsal and anal fin rays, whose tips lie along the bottom like two bulsdozer treads, move peristaltically to the rear to force the fish forward. The reduced right pectoral fin might also give some small shove against the bottom. In this case, as in the above possibility, horizontal movement would produce vertical lift because of the angle of attack the fish presents to the water. The third possibility is that the fish gains a vertical normal force from the substrate, but the nature of the substrate does not matter. The bottom may act as a solid springboard to allow the fish to gain a vertical displacement that would not be possible without a substrate to push against. Freely swimming fish that accelerate using an undulatory motion like that of the sanddab during liftoff initially slip laterally from their original axis of orientation, then cross their original axis to a point displaced slightly laterally and forward from their original position. (Figure 8) Since for a sanddab the normal force of the bottom prevents the initial lateral slippage it can end up with a greater net lateral displacement than freely swimming fish. This might give it an advantage in striking prey or escaping from a predator. Images from a camera filming at more than 30 frames per second would help resolve the above three possibilities into a model for the sanddab during liftoff. A final, less obvious advantage the sanddab gains from its bottom habitat is its ability to optimize itself as a 'transient swimmer. This type of fish tends to remain stationary, conserving energy and swimming only occasionally in short bursts of acceleration. Of advantage to this type of fish is a large muscle to body mass ratio, a deep silhouette to maximize the amount of water moved with a single stroke of the body, and a body flezible enough to create large amplitude waves (Webb, 1984 and Daniel, 1988). Because of its flattened posture and bottom habitat, the sanddab is able to maximize all three variables. Fish enjoying a transiently swimming lifestyle high in the turbulent water column, by contrast, cannot afford such a deep silhouette lest it act as a sail to drive them through the water away from their desired position. Conclusion Evolving to live flat on the bottom of the ocean has resulted in the sanddab making a number of adaptations. Its camouflage disguises it from predator and prey, while the shape it assumes allows it to stay in place despite water moving swiftly over its body. While it seems to receive little, if any advantage from a ground effect during liftoff from the substrate, further work is required to determine the chief method used by the sanddab for lifting off from the ocean bottom. Acknowledgements Thanks to Mark Denny of Hopkins Marine Station for providing laboratory space and useful advice during the course of this study. Additional thanks to Charles Goldman of UC-Davis for his encouragement and support. Bibliography Blake, R. W., 1979. The energetics of hovering in the mandarin fish (Synchiropus pieturatus J. Erp. Biol 82: 25-33. Blake, R. W., Median and Paired Fin Propulsion, in Webb, P. W., and Weihs, D., ed. Fiah Remechanis pp. 221-224. Praeger, New York, NY, 1983. Daniel, T. L., 1986. Forward Flapping Flight from Flexible Fins. Can / 222 66: 630-638. Thomas, George B., Jr. Clculus and Analytie Geometry p. 406. Addison- Wesley, Cambridge, MA, 1953. Ueland, F., 1984. Rheotaxis in dtharichthys sordidus Ciass papers, Stanford University Biology department“ 175H. Webb, P. W., Form and Function in Fish Swimming, S. Am, July 1984, p. 72. Figures 1: Location of Center of Gravity of the Sanddab. 2: Liftoff Sequence of the Sanddab. Each frame 1/30 5. 3: Speed of Wave Propagation along the Sanddab during liftoff. 4: Radius of Curvature of Wave along the Sanddab during liftoff. 5. Displacement of the Center of Gravity of the Sanddab during liftoff. 6: Velocity of the Center of Gravity of the Sanddab during liftoff. 7: Acceleration of the Center of Gravity of the Sanddab during liftoff. 8: Undulatory motion of a sanddab during liftoff from a solid substrate (top) and typical undulatory motion of a transiently swimming fish (bottom) (from Webb, 1974) Distance from Nose to Center of Gravitu 0.4 0.3 92 0 U05 007 009 011 013 0.15 0.17 0.19 0.21 0.23 0.25 0.27 0.23 Length (m) osee 2 0.18 0.16 20.14 50.12 80.10 -008 5006 2004 52002 000L E Mesh Substrate • Solid Substrate Wave Propagation time (s) 0.2 § 04 0.2 0.0 3-0.2 -0.4 0.4 02 0.0 3-0.2 -0.- — Takeoff from Hesh Substrate p — 0335 0675 — 100 5 1335 123 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Em from nose Takeoff from Solid Substrate F 0335 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 067 5 Em from nose 100 5 1335 X-Displacement 003 E Mesh Substrate 006 • Solid Substrate 0.04 0.02 000 -0.02 0.1 time (3) Y-Displacement 0.08 E Mesh Substrate 006 • Solid Substrate 0.04 002 -0.02 L U.0 0.1 time (3) 0.2 X-Yelocity 1.0 o8 06 0.4 + 0.2 —0— Solid Substrate (x) —— Mesh Substrate (x) 0.0 -02 033-067 067-100 133-167 200-233 233-267 Time (3) Y-Yelocity 10 08 06 0.4 02 —9— Solid Substrate (9) —— Mesh Substrate (u) 00 -0.2 0-033 033-067 067-.100 133-167 200-233 233-267 Time (3) 20 10 -10 * -20 U.O 30 20 U.O X-Accelerations —#— Mesh Substrate (x) —— Solid Substrate (x) time (s) Y-Acceleration — Mesh Substrate (u) — Solid Substrate (u) time (s) 0.2 0.2 —7 — JO8JeAE 12 . 4 5