220 Passage of Argon through thin Films of Indiarubber. 



The first, or rising, part corresponds to a spark in opening 

 the circuit ; the second, or descending, part of the curve 



ki 



corresponds to the law X= — == without spark, the two 



parts meeting in a point that corresponds to a maximum for 

 the spark-length in the secondary circuit, and an optimum 

 for the capacity in regard of the efficiency of the secondary 

 coil. The two parts of the curve would probably form a 

 cusp in the common point if a pair of pieces of graphite were 

 inserted in the primary circuit on each side of the interruptor, 

 the capacity being diminished gradually from its greatest 

 value during the observations. 

 Stockholm, Sweden. 



XV. On the Passage of Argon through thin Films of India- 

 rubber. By Lord Rayleigh, F.R.S* 



SOON after the discovery of Argon it was thought desira- 

 ble to compare the percolation of the gas through india- 

 rubber with that of nitrogen, and Sir W. Roberts-Austen 

 kindly gave me some advice upon the subject. The proposal 

 was simply to allow atmospheric air to percolate through the 

 rubber film into a vacuum, after the manner of Graham, and 

 then to determine the proportion of argon. It will be re- 

 membered that Graham found that the percentage of oxygen 

 was raised in this manner from the 21 of the atmosphere to 

 about 40. At the time the experiment fell through, but 

 during the last year I have carried it out with the assistance 

 of Mr. Gordon/ 



The rubber balloon was first charged with dry boxwood 

 sawdust. This rather troublesome operation was facilitated 

 by so mounting the balloon that with the aid of an air-pump 

 the external pressure could be reduced. When sufficiently 

 distended the balloon was connected with a large Toepler 

 pump, into the vacuous head of which the diffused gases 

 could collect. At intervals they were drawn off' in the usual 



The diffusion was not conducted under ideal conditions. 

 In order to make the most of the time, the apparatus was left 

 at work during the night, so that by the morning the internal 

 pressure had risen to perhaps three inches of mercury. The 

 proportion of oxygen in the gas collected was determined from 

 time to time. It varied from 34 per cent, when the vacuum 

 was bad to about 39 per cent, when the vacuum was good. 

 * Com inimical el by the Author. 



