64 INDUCTION SHOCKS 
A strip of vulcanite, d (Fig. n), 18 mm. long, 8 mm. 
wide, and i mm. thick, flat on one side and on the 
other tapered toward the edges, is supported at the top of 
the block by a horizontal rod working freely in a collar, 
e (Fig. n), in such fashion as to press closely against 
the inner surface of the cavity, a, and when rotated 
about its axis of support to cover or uncover the open- 
ing, c. When the vulcanite strip is brought over the 
opening, it cuts the mercury connection between cavi- 
ties a and b, and therefore breaks any electric circuit 
which may include them. This method of breaking a 
circuit has many points in its favor. The break cannot 
be delayed through the tendency of mercury drops to 
cling together, for the severance of the mercury column is 
not the withdrawal of one mass of mercury from another, 
but is the forcible interposition of a nonconductor in the 
path.* Moreover, the vulcanite strip cuts off not only 
the liquid mercury, but if it fits tightly, as it should, cuts 
off as well any mercury vapor that may be formed. 
Thus the effect of volatilization of mercury is minimized. 
Since the point where the break occurs is beneath a con- 
siderable depth of mercury, air does not have access to it, 
and oxidation does not occur. I have found, as a matter 
of fact, that the same mercury may be used in one of 
these keys for months without any appreciable varia- 
tion in the effectiveness of the break. 
* A device employing the same principle was described by Lombard 
in 1902: Am. Jour, of Physiol., 1902, viii, p. xx. 
