74 



one end, with the concave edge inward, and the plane side 

 undermost, the plane making an angle of about forty degrees 

 with the horizon ; throwing as if to strike the ground at the 

 distance of about thirty yards, and giving it, on leaving the 

 hand, a rapid rotatory as well as progressive motion. In- 

 stead of striking the ground at which it was aimed, its plane 

 becomes horizontal at the distance of twenty-five yards, and 

 so continues for about fifteen yards, when it commences 

 rising in the air, and moving towards the left ; its plane then 

 becomes inclined, and continues at an angle of from thirty to 

 forty degrees, whilst it describes apparently a segment of a 

 circle to the left. Having, at the distance of sixty or seventy 

 yards, attained an altitude of from forty to sixty feet, the pro- 

 jectile returns, descending to the point from which it was 

 projected, when its plane becoming once more horizontal, it 

 skims along within a few feet of the ground, and passes close 

 by the right hand of the person who threw it. On passing, 

 its plane becomes elevated once more, it rises a second time, 

 and performs another smaller curve, (fifteen or twenty yards 

 behind the projector,) in like manner as the first, with this 

 singular exception, that the second curve is described from 

 left to right, contrary to the course of its rotation and of the 

 first curve, which is invariably from right to left." 



In bringing the instrument under the notice of the Aca- 

 demy, Mr. MacCullagh wished to draw attention to the 

 theory of its motion. When a body of any form whatsoever 

 is projected in vacuo, we know that its centre of gravity 

 must describe a parabola in a vertical plane, while the body 

 spins about an axis passing through that centre. In the pre- 

 sent case, therefore, it is clear that the continued swerving from 

 the vertical plane must be ascribed to the action of the air. 

 But to compute accurately the mutual action of the air, and 

 of a body endued, at the same time, with a progressive and 

 a rotatory motion, is a problem far beyond the present powers 

 of science. The problem can only be solved approximately ; 



