330 Profs. J. J. Thomson and R. Threlfall. [May 6, 



the volume of the nitrogen takes place, which reaches a maximum, 

 after which the passage of sparks of the same kind produces no 

 permanent effect upon the volume. 



2. That for nitrogen at a pressure of 8 mm. of mercury, which is 

 the pressure at which we have usually worked, the permanent diminu- 

 tion in volume is from 8 to 12 per cent, of the original volume, while 

 at a pressure of 16 mm. of mercury the diminution is not more than 

 from 2 to 3 per cent. ; thus, though there are twice as many molecules 

 in the tube the effect is not so big. 



3. The diminution in volume takes a considerable time to reach its 

 maximum value ; in our experiments, where the discharge-tubes are 

 1 cm. in diameter and 25 cm. long, and the sparks were produced by 

 an induction coil giving a spark about 4 inches long in air, it took 

 about eight hours' sparking to produce the maximum diminution. 



4. That this diminution takes place equally well whether platinum 

 or aluminium electrodes are used. 



5. That the ratio of the maximum diminution to the original 

 volume is independent of the volume of the discharge- tube and of the 

 extent of its surface. 



6. That if the tube be maintained at a temperature of over 100° C. 

 for several hours, the gas regains its original volume. 



We attribute this diminution in the volume of the gas to the forma- 

 tion of an allotropic modification of nitrogen.* 



We now proceed to give a detailed description of the experiments 

 and of the various parts of the apparatus. 



Discharge-tube and Gauge. 



We had a good deal of trouble in getting this part of the apparatus 

 satisfactory; we found that discharge- tubes of the ordinary kind 

 were very liable to leak round the electrodes after a series of sparks 

 had been passed through them. The form of tube we finally adopted 

 is represented in fig. 1. AB is a glass tube about 25 cm. in length 

 and 1 cm. in diameter, into which the \J -piece E, F, Gr, fused up at Gr, 

 is fused, sulphuric acid or mercury is placed at the bend of this \J -tube, 

 and serves as a gauge to measure alterations in the pressure of the 

 gas in AB. The end B dips into a vessel containing mercury, the 

 level of which is higher than that of the part of the tube through which 

 the electrode passes ; a piece of glass tubing is placed over the top of 

 the discharge- tube, and the interval between the tubes caulked with 

 glass wool ; the cup thus formed is filled with mercury, which reaches 

 above the entrance of the electrode into the tube. The electrodes are 



* Since this paper was sent to the Koyal Society we have seen a book by Mr. 

 Stillingfleet Johnson, entitled "Elementary Nitrogen," in which the same conclusion 

 is come to from purely chemical reasons. 



