applied across the space between target and electrode. On 

 impact of the bullet on the target the electrometer indicated 

 the reception of a considerable charge, and this independently 

 of the direction of the field, showing that ions of both signs 

 were present in the air between target and electrode. 



In the earlier experiments a Winchester rifle was 

 employed, but in order to avoid effects due to exploded gases 

 accompanying the bullet the Winchester was discarded in 

 favour of an air-rifle, with which no such parasitic effects 

 could be detected. 



The rifle was mounted w T ith its muzzle at a distance of 

 about 6 ft. from the target ; it threw a leaden bullet weigh- 

 ing '96 gram, with a velocity of about 700 ft. per second. 



The results of preliminary qualitative experiments were 

 stated in a paper read by one of us 0) before the Australasian 

 Association for the Advancement of Science at Sydney, 

 January, 1911. 



II. a. Methods and Results of Experiments with 

 Air-rifle. 



The method of investigation employed in the first in- 

 stance, i.e., when the ions are produced by the impact of a 

 bullet on a target, will be understood best by reference to 

 the accompanying diagram (fig. 1). The bullet fired from 

 the air-gun perforates a thin sheet of paper, LN, and im- 

 pinges upon the target, P, the nature of which can be varied 

 at will. An annular metal electrode, AB, is supported in 

 front of the target at a distance from it of about 1"5 cm. and 

 insulated as shown with sulphur and guard-ring. Target 

 and electrode are enclosed in hollow metal box, the lid, CD, 

 of which is removable. This box with the target can be 

 raised to any desired potential by means of a battery of 

 500 small accumulators and the charge driven on to the 

 electrode measured by a Dolezalek electrometer of suitably 

 arranged sensibility. 



In making measurements of the total ionization the 

 electrometer reading (mean of two or three successive 

 swings) was taken 15 seconds after impact of the bullet, 

 in which time practically all ions of a given sign produced 

 in the chamber were found to be collected. A number of 

 measurements (usually five or six) were made in each 

 experiment and the arithmetic mean of these taken. The 

 amount of variation in the effect is indicated by the mean 

 error of these measurements. Where necessary, correction 

 was made for the natural leak of the instrument. The order 



(l) Notes on Ionization by Impact, by Kerr Grant. M.Sc. 



