I collected and delivered codes from the electrodes to a terminal box, held on the lumbar, and relayed them to an input terminal of the electromyograph by cable, the length of which was sufficient so as not to disturb pitching. I recorded the data on a personal computer connected to the output terminal at 1OV absolute. I connected a terminal from a video camera to the input terminal of a video timer,and the output terminal to a video recorder to record images of the pitcher from the side view simultaneously with the measurements on EMG. I could identify the position every 1/100 second on therecorded image. Each set of data was synchronized (Fig. 1). I repeated rehearsing experiments and removed noise as much as possible before the actual take.