Why? The Science of Athletics

288 WHY?-THE SCIENCE OF ATHLETICS . behind this point must be "push". Again, supposing the weapon be held at a and thrown, then, as this point is in front of c, it naturally follows that the force applied must be "pull". Pull in throwing is far less likely to produce error than push, as will be seen from Fig. 105. Let ab represent a javelin travelling in the direction of the arrow heads and c its common centre of gravity. Now assume again that the weapon is grasped at b and thrown. It would naturally oe very hard to hold the javelin in ("( FIG. 105 this position and maintal.n the true direction of flight, i.e . along ab ; therefore, assuming the error is a' b', a force applied at b and with a d.irection parallel to ab would produce an upward tendency to Totafe round c, as shown by the arrow u (or vice versa, should the error be the other side of ab), and it will be seen easily that the nearer the javelin is held to the centre (c) the less likely is the error to occur. Or should the weapon be held a{ a point near a, the tendency will be to drop it in the direction d. Therefore, to maintain the javelin in the line of its true direction of flight, it is essential that it be thrown with the hand-hold established at its exact centre of gravity, c, as shown in Figs 106, 107 and 108, Plate 23. Another consideration is in relation to the circum– stance that it is necessary for the thrower to transmit his