Intelligence and Miscellaneous Articles, 



155 



and the spark, coiideused by iiiue jars, springs between two small 

 knobs of pLatinum. The sides have no effect, because of their dis- 

 tance. Only the principal lines are quoted; the plate contains TO. 

 The intensity varies from 1 to 6. 



*A = 43^-8 



432-5 3. 



431-4 4. 



429-9 2. 



428-7 2. 



427-4 4, 



423-8 6, 



422-0 4. 



420-7 3. 



419-1 3. 



417-5 4, 



416-2 4 ( 



415-2 ....'.. 4. 



414-3 4. 



413-1 4. 



spread. 



\ = 



=411-8 . . . 



. . . 5. 







411-0 ... 



...3. 







410-2 . . . 



...4. 





* 



409-5 . . . 



...4. 







407-6 ... 



...4. 



spread. 





406-9 . . . 



. .. 4, 



spread. 



X 



406-5 . . . 



. . . «J. 





X 



406-2 . . . 



...3. 







405-7 . . . 



. . . 2. 





X 



404-9 . . . 



... 5. 



spread. 



* 



404-2 . . . 



...5. 



(2 in the 





403-5 . . . 



. .. 4. 



[middle). 



X 



402-6 . . . 



...3. 





* 



399-6 ... 



. .. G. 







398-0 . . . 



... 2. 



double. 



The lines indicated by an asterisk have been already noted as 

 belonoing to nitrogen ; the others have not before been observed. 

 Among these, those which are indicated by the letter x are not 

 produced in air \vith knobs of tin ; the rest are common to both 

 spectra, and consequently belong to nitrogen or hydrogen. Most 

 of the X lines are more intense at the extremities than in the middle ; 

 the spectrum obtained with knobs of tin showed that the lines of 

 the electrodes present this peculiarity. This effect is rendered 

 most apparent by placing the spark parallel to the slit of the spec- 

 troscope. 



In the chief experiment which remains to be described the spark 

 is perpendicular to the slit. We have then on the photograph 

 some short lines, and long ones which extend on both sides beyond 

 the median band of the spectrum. After operating on nitrogen at 

 the ordinary pressure, I compressed the gas in the cylinder to 

 eight atmospheres, and took a photograph without making any 

 change in the apparatus. The comparison of the two proofs ob- 

 tained at 1 and at 8 atmospheres fully confirms my proposition. 

 At 1 atmosphere the nitrogen-lines extend on both sides of the 

 median region impressed by a feeble continuous light ; throughout 

 their length they are admirably sharp, since more than thirty of 

 them could be measured in the space of 3 millimetres. 



At 8 atmospheres the median region is quite continuous, and 

 the lines of the gas can hardly be distinguished there ; but these 

 extend in both directions as in the preceding spectrum, and pre- 

 serve their sharpness. This is precisely the effect which would be 

 produced by a trail of fire formed by solid particles of dust and en- 

 compassed by an aureole of luminous gas. 



I am pursuing these researches, as far as the means at my dis- 



