480 Llectrical Effects produced by Explosion of Hydrogen. 
be reduced to as little as one fiftieth of this value. This 
result is got by dividing the largest value of ( obtained, 
11:4, given by the first experiment with the low-resistance 
galvanometer, by 356, the number of c.c. in the cylinder, 
It appears that an increase of electric force produces little 
or no effect, and that therefore, presumably, all the ions are 
collected when the outer cylinder is at 82 volts. This con- 
clusion was arrived at from the experiments of Table I. 
There is not sufficient evidence to show whether the 
presence of air produces an effect, but the experiments given 
in Table III., in which the quantity of air was varied, seem 
to point to the presence of air causing some increase in the 
ionisation. But, if this is so, it is far from proportional to 
the amount of air present. The matter is further compli- 
cated by the diminution in the completeness of the explosion 
when the amount of air is considerable. This is shown in 
Table ITI. 
The chief interest of these results lies, perhaps, in their 
bearing on the question whether the production of molecules 
of water-vapour from atoms of hydrogen and oxygen is 
accompanied by the formation of a comparable number of 
ions. The number of molecules of water-vapour formed is 
the number of molecules of hydrogen which enter into com- 
bination. Hence the number formed per c.c. for each milli- 
2 Nia 
3° 760 Tay 
where N is the number of molecules per c.c. at 760 mm. 
pressure and 15°C. The number of coulombs produced 
nee 10" , : 
per ¢.c. 1s ee , where vc.c. is the volume of the cylinder 
3000 9 
Uv 
metre fall of pressure of the mixed gases is 
and g has the same meaning as in the tables, or 
electrostatic units. If then e is the electrostatic charge on 
an ion, the number of pairs of ions produced bears to the 
number of molecules formed the ratio 
30009 x 1140 _».8 x 10-4 Z 
ve N Vv 
since Ne=1:22x 10" (J. S. Townsend, Phil. Trans. 1899). 
Taking the largest value of g given by the experiments, 
-144, and v=356 c.c., the ratio becomes approximately 
l-1x10-*. It would appear then that about ten million 
molecules of water-vapour are formed for every pair of ions 
produced, and that, too, when nearly 10 per cent. of the 
mixture is air. On the other hand, it is quite possible that 
a more suitable apparatus would produce greater electrical 
a t 
Pgh’ 
