IN a recent paper (this vol., p. 130) an account was given of e'xperiments made to check the observation by Thornton that the ignition of a mixture of methane and air by the impulsive electrical discharge I' proceeds per saltum as the pressure is changed " (Colliery Guardian, 1916, 112, 504). Lack of the requisite apparatus prevented a t that time the prosecution of the' expelriments a t pressures above atmospheric. Such experiments have1 since been made; and i t seems desirable to lay the

more »

... ts before the Society, not only because they difler in two essential respects from Thornton's, but because a possible explanation of one of the1 differences is suggested. A diagram representing Thornton's results is shown, in dotted line in the figure. The " igniting-currents " recorded are those currents passing through the primary circuit of the particular induction coil used which, when broken, gavel secondary discharges just capable of igniting the mixture, which contained 9.5 per cent. of methane. It will be seen that not only are there well defined "steps," but that when the pressure of the mixture is increased above 2000 mm. the direction of the stepped curve is reversed. This resmarkable result, indicating decreased '( ignitibility " of a given mixture as its pressure is increased, would, if corre8ct, be as important practically as theoretically. I have been unable to confirm Thornton's observations, as will be seen from the curve shown in full line in the figure, representing the, results of my experiments. An attempt was made to use one of the explosion vessels that had been employed f o r the work a t pressures less than atmospheric so as t o make the results directly comparable, but it was found that the spark-gap (1 mm.) was too great. For when the pressure of the mixture was but little greater than atmospheric it was found that the least discharge that could cross the gap caused ighition of the mixtxre, and there was, of course, no means of ascertaining whether, if a discharge of less intensity oould have crossed, it also could have caused ignition. A new explosion vessel with a sparkgap of between 0.25 and 0.5 mm. was therefore employed, the electrodes being of platinum. This reduction in the spark-gap enabled a secondary discharge of less intensity to pass, whilst a t the same time, owing t o the decreased length of the sparks, a greater The electrodes were of nickel.