ME. W. CEOOKES ON EEPULSION RESULTING EEOM RADIATION. 
529 
turned mass of magnesium, weighing 42 grains, suspended by a very fine platinum 
wire, the distance between the point of suspension and the centre of gravity of 
the magnesium bob being 39-139 inches, so that it forms a seconds’ pendulum ; f is a 
spiral made of platinum plate, fastened to two stout copper wires which pass through 
the thick plate of glass b, and thence pass to a contact-key and a battery. The plate b 
is cemented (83) to the end of the tube a b, which is ground flat, g is an arm fused 
into the upright tube for the purpose of connecting it to the glass spiral of the pump ; 
it is contracted at h for convenience of sealing off. The fine platinum wire is fastened 
at its upper end to a thick wire which is sealed into the glass, and passes through to 
the outside for electrical purposes (120). The distance between the pendulum bob 
and the spiral is 7 millims. To ignite the spiral the current from two Grove’s cells 
was used ; this brought it to a bright red heat in air, and to a white heat in vacuum. 
Three feet from the pendulum a telescope was firmly clamped to the bench ; 
it was furnished with a micrometer-eyepiece, with movable spider-threads and gra- 
duated circle. The edge of the magnesium bob was brought into the same focus as 
the traversing cross wire. Observations were taken in the following manner : — The 
observer at the telescope brought the cross wire to zero, and then adjusted it to coincide 
with the edge of the pendulum bob. An assistant, guided by a seconds’ watch, 
pressed the contact-key down for one second, then broke contact for a second, next made 
contact for the third second, and so on, alternately making and breaking contact for 
either 10, 20, or 40 seconds, counting the seconds aloud. At each second the swing of 
the pendulum increased ; and the milled head of the micrometer was kept turning so as 
to let the cross wire keep up to the furthest point to which the pendulum vibrated. 
At the end of the experiment the position of the cross wire was taken and its distance 
from zero recorded. 
100. Experiments were first tried in air of normal density. The pump was then set 
to work, and observations were taken at different heights of the gauge. The difference 
between the height of the gauge and that of the barometer gave the tension of air in 
the apparatus in millimetres of mercury ; this is recorded in the first column of the 
following Tables. The second column gives the greatest amplitude of the half oscilla- 
tion of the pendulum in millimetres — the sign plus signifying attraction, and minus 
repulsion. 
Near the centre of Table I., in the second column, are five observations to which I 
have affixed no sign. When trying the experiments I thought that either I had mis- 
taken the direction of impulse, or my assistant had commenced to count the make-and- 
break seconds wrongly, as the movement seemed to be repulsion. Never having had 
repulsion at such a pressure before, I was not prepared for it ; and fearing there might 
be an error, left the sign queried. Another series of observations were taken to re- 
examine this point ; they are given in Table II. 
It is worthy of notice in these Tables that the attraction by the incandescent 
spiral is only moderate in air of ordinary density. The attraction diminishes to a 
MDCCCLXXV. 4 A 
