107 
Therefore the amount of oxygen that must be present in 
nitrogen in order to be detected in the spectroscope is 4*5 
per cent. 
The addition of air to the oxygen gives the same result as 
adding pure nitrogen to the oxygen, so that it is immaterial 
which is used ; neither does the dryness of the gas seem to 
affect the results. 
The following experiments were made at diminished pres- 
sures to see how the above results are affected at different 
pressures. 
Thus, to 95 c.c. hydrogen, at 11 inches mercury pressure, 
1*5 c.c. air was added, and the nitrogen line was just visible. 
Now 95 c.c. hydrogen at 11 inches pressure = 35 c.c. at 30 
inches pressure, and this gives 3*5 per cent nitrogen. 
Again, to 90 c.c. at 10 inches pressure were added 
•5 c.c. air, and no nitrogen lines were visible, 
•3 c.c. air more, „ „ 
•4 c.c. air more, „ „ 
*2 c.c. air more, and nitrogen line became visible. 
Therefore 1*4 c.c, air in 30 c.c. hydrogen = 3*7 percent 
nitrogen. 
Thus a diminution of pressure from 29J inches to lOJ 
inches mercury alters the amount of nitrogen that must be 
present in hydrogen in order to be detected in the spectro- 
scope from IT per cent to 3*6 per cent. 
96 c.c. hydrogen at 2 inches pressure showed the three 
hydrogen hands as wide lines, sharp at the edges, the red 
being the broadest On the addition of *25 c.c. air the nitro- 
gen line wave length 5004 was seen at the edges of the field 
only, and not extending across. 
On adding further small quantities of air (up to 3*8 c.c.) 
more nitrogen lines came into view, till at last they were all 
in view, but the}’ were only visible at the edges of the field, 
and would not extend across the field. The hydrogen lines 
became broader on each addition of nitrogen, and ultimately 
