Results for the seven balls tested are given in Figs. 2–4, and further details are given in Table I. The force wave forms are all of a similar general form, being an approximate half-sine wave form, but asymmetrical in time. For most of the balls, the maximum force is recorded at a time close to 0.5t, where t is the duration of the impact, indicating that the experimental compression and expansion phases are of approximately equal duration. However, the impulse during the compression is larger than the impulse during the expansion, with the result that the ball rebounds at a speed less than the incident speed. The plasticene ball did not bounce and remained permanently deformed after the collision. All of the hysteresis curves have a finite area, indicating that all collisions were inelastic. The golf and superballs have an approximately linear compression phase, with F}y, and a nonlinear expansion phase. The y displacement wave forms are more closely sinusoidal than the force wave forms, at least during the compression phase. In all cases it was found that the ball rebounds in a compressed state since y remains finite at the end of the impact. This was confirmed for the tennis and superballs by aligning the beam L2 , as shown in Fig. 1, so that it grazed the top of the ball when the ball was at rest on the piezo. The results for the tennis ball are shown in Fig. 3, where it can be seen that the beam is blocked just prior to the impact, it is unblocked during the impact, and remains unblocked for about 0.5 ms after the impact. The spatial resolution was not sufficient to observe this effect with the other balls, since the ball compression was too small. A totally unexpected result was obtained with the silly putty ball. Silly putty has the property that it stretches easily when stretched slowly, it breaks when stretched quickly and bounces when dropped on a surface. When the silly putty ball was dropped on the piezo, it was discovered, with some initial astonishment, that the piezo generated a negative output signal commencing about 30 ms before the ball made contact with the piezo, as shown in Fig. 4. No other ball had this effect, and the effect was observed only with a freshly prepared silly putty ball, created by stretching the putty and rolling it into a ball. The effect was traced to electrostatic charging of the ball to about 1 kV when it was stretched. The effect was simulated by charging a plastic rod and moving it toward or away from the piezo. The capacitance between the ball and the piezo was only a few pF, but this was sufficient to generate a voltage of about 0.2 V across the piezo. The ball was able to hold its charge for about 20 min, despite repeated handling and dropping of the ball.