Fig. 1. (A) Sampling locations for flow speed. Flow speeds were averaged over 21 points (red crosses; for clarity only 11 points are shown) equally spaced on a radial transect with a diameter of gape, ranging from +50 to –50 deg. from the imaginary line projecting at a right angle to the mouth. The measure of mean fluid velocity at gape distance from the mouth is referred to throughout the paper as `flow speed'. Velocity profiles were calculated for points under the imaginary line (broken blue line) projecting at a right angle to the mouth. (B) Determination of time to peak gape (TTPG), linear speed of mouth opening (G/t) and of the acceleration of the flow in front of the fish's mouth, illustrated on kinematic and flow speed measurements from a typical strike of a 180 mm bluegill sunfish with TTPG=32 ms. For bluegill, the change in jaw angle is constant between strikes (jaw angle transforms from 10 to 180), and thus, the time it takes the fish to open its mouth (TTPG; measured from 20% to 95% peak gape; gray reference lines) is inversely proportional to the angular speed of mouth opening, /t. The linear speed of mouth opening was determined by regression of gape distance on time (dotted blue line) through at least two-thirds of the opening phase of the mouth (filled blue circles). The slope of that regression (400 mm s^–1) describes the mean rate of change in gape distance, G/t, during mouth opening. Fluid acceleration was determined similarly, as the mean rate of change in flow speed over the duration of increasing flow speed (dotted red regression line through closed red circles; 6.67 in this case). We retained strikes for further analysis only if R² for the above regression was higher than 0.9.