XXXVI. Action of Radium on the Electric Spark. By R. S. Willows, M.A., D.Sc, and J. Peck, B.Sc [Communicated by the Physical Society: read January 27, 1905.] It is well known that radioactive substances, when placed near a spark, in general cause the discharge to pass more readily by the formation of ions in the electric field. The action on a long spark is, however, different, and appears, as far as our knowledge goes at present, somewhat irregular. In some cases the discharge passes more readily, in others with greater difficulty [citation redacted]. The present paper describes experiments which were undertaken with the object of studying the origin of this influence. The spark was produced between two brass spheres of unequal diameters, one 27 mm. the other 48 mm., by means of a Wimshurst machine carrying three pairs of plates of 2 feet diameter. The radium bromide, 5 mgm., was contained in a small capsule closed with mica. The strength of the [header] 379 sample was not known, but in most of the experiments it was not of the highest activity. The action was altogether different according to the direction of the discharge. Using a spark-gap longer than 2 cms., and making the larger knob positive, the radiations had practically no influence. With the smaller sphere positive the radium in most cases extinguished the spark, the visible portion of the discharge being confined to a glow at the surface of both electrodes. With the machine used a torrent of sparks could be kept up with a gap of 3 cms. ; in these circumstances the spark was least sensitive to the radium. As the knobs were gradually pulled apart, a good deal of brush discharge became mixed with the spark, and the effect was more easily observed. With still longer discharges there appeared only a positive brush, beyond this a dark space, and on the surface of the cathode a faint glow. In this condition the discharge is very susceptible to influence, it may be extinguished instantaneously by the radium held at a distance of 1 metre. The following series of observations and the accompanying curve (fig. 1) plotted from the numbers, show how the sensibility increases with the spark-length. The two spheres were set apart so that their nearest points were distant from each other the lengths given in the first column, the machine was put in action and the radium gradually approached along a line at right-angles to the discharge. The numbers in column 2 give the distances at which the discharge was just extinguished; in each case the numbers are the mean of several observations. Distance apart of knobs. Smaller sphere positive. Distance of radium from spark-gap when the discharge is just put out. [formula redacted] [The discharge in this case is mixed brush and spark. Most of the brush is stopped at 10 cms., the spark also at cms.] 380 [header] The curve is practically a straight line up to a spark-length of 12 cms., after which the sensibility rapidly increases. It is at this point also that the dark space first appears. [figure redacted] The appearance of the discharge with spark-lengths 4, 7, and 13 cms. is shown in the sketches A, B, &c. respectively of fig. 2. When the anode consisted of a piece of No. 22 wire, with which the discharge passed nearly entirely in brush form, the radium had no, visible effect. Both this and the fact that [header] 381 the radiation has no influence when the discharge passes from the larger to the smaller sphere, are consistent with the theory that a minimum discharge potential is necessary for the [figure redacted] effect to be produced. In each of these cases, the mean potential gradient is less than it is when the spark is in its sensitive condition. Although the visible discharge is extinguished, it does not follow that the current actually passing is diminished ; the conductivity might be so largely increased that a non-luminous discharge takes place. It was found possible to measure this current, the numbers, although variable, being sufficiently accurate for the purpose of the experiment. To do this, the positive side of the spark-gap was connected directly to the machine, the other side to one terminal of a dead-beat D'Arsonval galvanometer, the other terminal of which was earthed. The negative pole of the machine was also earthed. The current thus passed from the anode across to the cathode, and through the Galvanometer to earth. It was found that the extinction of the spark corresponded with a decrease in the current. Further, by encasing the radioactive salt in a thick lead case, so that a narrow beam 382 [header] only of rays issued, it was possible to investigate the effect at different points of the discharge. The lead case was closed by a thick lid of lead, working on a hinge, which could be opened or closed by the observer at a distance. The discharge had the form shown in C, fig. 2. The two sets of numbers given below will show the results and the order of accuracy obtained. Per cent. Decrease of current Near positive knob. Middle of discharge. Near negative knob. [table redacted] The numbers are the mean of a large number of separate observations. It is seen therefore that the positive end is the most sensitive under these conditions. With shorter discharges, the currents were too irregular to admit of conclusions being drawn. A reference to 0, fig. 2, shows that the discharge near the anode is concentrated into a single branch very narrow in section. It hence seems probable that the most sensitive part is where the electric field is greatest. We found it impossible to measure this field at different points of the gap. [figure redacted] The apparatus shown in fig. 3 was used to investigate the effect of altering the pressure of the gas in which the discharge took place. The bell- jar A was 25 cms. long and 12 cms. in diameter. One end was closed with a brass plate, B, through which passed a rubber bung ; D was connected with a water-pump and gauge. The radium capsule, 0, was fastened to a piece of soft iron which allowed of it being shifted to any [header] 383 part of the jar by means of an external magnet. The spark-gap was about 6 cms. long. The results are as follows : — Pressure in ems. of mercury Effect of radium. [table redacted] Visible discharge (spark and brush) stopped. Discharge (all brush) stopped. Discharge (all brush) concentrated but not extinguished. The action of the radium is therefore less marked as the pressure is reduced. This may be due to the reduction of the voltage required to maintain the current. A very much weaker sample of the salt enclosed in a glass capsule showed quite an appreciable effect when the discharge was in a fairly sensitive condition. Thus the action cannot be due to the [alpha] rays, since these would be absorbed by the glass. It must be due to either the [beta] or [gamma] rays. Since the latter are supposed to be identical with Röntgen rays, the influence of these on the spark was tried, but they brought about no stoppage of the discharge. The ionizing powers of the Röntgen rays and of the rays from the radium on the air in the spark-gap were of altogether different orders. Thus the radium at 50 cms. readily extinguished the spark, a Röntgen-ray tube 25 cms. away had no effect : but if the gap was replaced by the knob of a charged electroscope, the latter was completely discharged in less than 1 sec. when the tube was working, while with the radium complete discharge took 5 min. to accomplish. The result obtained cannot therefore be ascribed to the [gamma] rays, nor does it depend directly on the number of ions produced by the ionizing agent, but rather is a special action of the [beta] rays. The negative result obtained with Röntgen rays also precludes the following explanation : — The ions in the discharge are produced in two ways, directly by the field and by collision of ions already present with neutral atoms if the field is sufficiently intense. If the ionizing agent, rays of any kind, produces sufficient ions, the voltage between the electrodes may be so reduced that the second method of formation is impossible, and the final result may be a decrease in the current which passes. The effect would then be more marked as the external agent produces more ions, but, as is seen, this is not the case. It remains only to show directly that the effects observed are due to the [beta] rays. The radium was enclosed in a lead 384 [header] capsule of the shape shown in fig. 4, and was placed near enough to the spark to extinguish it. [figure redacted] The capsule was further placed between the poles of a strong electromagnet, so orientated that the [beta] rays would be driven into the lead by its action. Immediately the magnetic field was created, the discharge passed again with undiminished intensity, while when the current in the magnet was stopped the visible discharge also stopped, showing clearly that it was the easily deflectable [beta] part of the rays that was concerned in the extinction. As these rays are of the same nature as Lenard rays, the action on the spark of the latter was also tried. With the aluminium window of the tube distant from the spark about 10 cms., an appreciable effect could he seen ; when the tube was in good working order and the spark in its most sensitive condition, extinction of the latter could be readily brought about. If the cathode rays inside the tube were deflected from the window by means of a magnet, so that no Lenard rays existed outside, no action on the spark could be seen. In fact the discharge from the Wimshurst could be started and stopped quite readily by simply approaching a magnet to the cathode of the Lenard tube. The ionizing action of the rays as tested by an electroscope 10 cms. away from the window, was more than that of the radium 50 cms. away, but less than that of the Röntgen rays when the tube emitting the latter was 25 cms. away. This experiment also shows that the failure of the Röntgen rays to put out the spark cannot be ascribed to their intermittence, as both Lenard and Röntgen tubes were worked by the same coil. It would also tend to prove that the effects are dependent on the high velocity of the [beta] rays, rather than directly on the number of ions they produce in the gap ; for the ionization produced by the Lenard rays is here greater than that produced by the radium, while it is well known that their velocity is much less than that of the [beta] rays that the radium emits. In conclusion we must express our gratitude to Mr. J. J. Vezey for the loan of some radium bromide, with which the preliminary experiments were performed, and also to the Governors and Principal of the Institute for providing us with the radium bromide necessary to continue those experiments. We are also indebted to Mr. F. G. Bratt for considerable help in the experimental part of the investigation. Sir John Cass Technical Institute, Jewry St., E.C. November 10, 1904.