nitrogen was added, and the nitrogen line was then seeii. 
This gives 1’2 per cent. 
To 118 C.C. hydrogen, dried by having concentrated sul- 
phuric acid in the eudiometer on the top of the mercury, 
1 c c. nitrogen was added, and the nitrogen line was not then 
visible: on addition of '2 c.c. more nitrogen the nitrogen 
line was not visible, but on addition of *2 c.c. more it came 
into view. 
This gives 1’2 per cent. Mean of the four experiments is 
115. 
Therefore the amount of nitrogen that must be present in 
hydrogen to be detected in the spectroscope is IT per cent. 
Again, 4*5 c.c. nitrogen were added to 143 c.c. hydrogen, 
and the nitrogen line wave length 5004 was easily seen; on 
adding 1 c.c. more nitrogen the yellow nitrogen lines wave 
lengths 5681 and 5666 came out, and another green line 
wave length 5164 makes its appearance on the addition of 
2*5 c.c. more. 
The result is not affected in the least, whether air or pure 
nitrogen is added to the hydrogen. 
(6) Hydrogen in nitrogen. 
To 112 c.c. pure nitrogen (made by the method already 
given) 67 c.c. hydrogen were added, and the green and red 
hydrogen bands were distinctly visible. 
To 144 c.c. nitrogen 1*6 c.c. hydrogen were added and the 
green band became visible, not as a band but as a very broad, 
thick line, quite sharp at the edges, and on the addition of 
more hydrogen it widened out into a band. The red hydro- 
gen band was also visible at this time as a broad line. 
To 140 c.c. nitrogen *35 hydrogen was added; the red 
hydrogen band was easily seen, but the green band only 
flashed in occasionally as a broad line. 
On addition of *3 c.c. more hydrogen it was still a broad 
line, and broadened out into a band on adding more hydro- 
gen. 
