JoLty— On the Conservation of Mass. dl 
the scale at either side of zero may be left out of account as not actually influential 
in restricting the sensitiveness of the apparatus. 
To investigate the more serious question of the effects of air-friction, a fine 
needle was magnetised, and then broken in halves. One half was suspended 
within a tall glazed box by a fibre of silk 54 ems. in length. The needle was 
balanced horizontally, and, of course, pointed in the magnet meridian. he other 
half of the needle was now brought to within 3 cms. of the suspended needle, 
attracting it out of the vertical. A micrometer microscope served to measure the 
displacement of the fibre just above the needle. It was found to be 0:09 ems. 
The weight of the suspended needle was 0:0230 grms. The force at 3 cms., 
therefore, is 0-039 dynes. One of the magnets is now attached to the extremity of 
the beam, and at right angles to the plane of the beam, and is thus nearly 
perpendicular to the magnetic meridian. The other magnet is mounted at the 
end of a wooden rod, and introduced through the wall of the case containing the 
beam. ‘To avoid any air-disturbance attending this operation, the rod is arranged 
to move within a glass tube entering the case from outside, and closed at its inner 
end. Thus the magnet can be presented, at any desired distance, to the magnet 
carried on the beam. It is now sought to determine the rate of change of 
velocity of the spot of light when the magnet is presented at various distances. 
Many observations were made. These were examined by plotting the rate of 
motion before and after the introduction of the magnet from without. I repro- 
duce further on (see p. 38) some of these observations as typical of the 
results. 'The mode of observation is the same here as was adopted in the actual 
experiments on possible inertia change—that is, the position of the spot of light 
on the divided scale is read every 15 seconds. ‘The scale reads 0 at its central 
point, and, as already stated, is divided to each side of 0 into divisions three-fifths 
of a millimetre in length. The scale-readings are plotted vertically, the time- 
intervals horizontally, the units in the latter case being 15 seconds. ‘Thus the 
intersection of the curve with each ordinate is an experimentally determined point. 
It will be seen that for several consecutive intervals of 15 seconds, the beam 
hag fluctuated irregularly, sufferimg minute displacements. When now the magnet 
is introduced, a steady motion is imparted to the beam (which is fully loaded), 
attaining at the moment of the second observation after application of the force— 
that is, in 30 seconds—a velocity which is almost steady, and amounts, in the 
mean, to the rate of 15 scale-divisions per 15 seconds, or one scale-division per 
second. The magnet being presented at an initial distance of 3 cms., we may state 
this result as indicating that a force of 0:0392 dynes acting for 30 seconds imparts 
a velocity of one scale-division to the spot of light, or 0:055 mms. per second to the 
masses suspended on the beam. Now this total mass being here just 160 grammes, 
it is easy to calculate that a force of 0:0288 dynes acting for this interval of time 
TRANS. ROY. DUB. SO0., N.S., VOL, VII., PART Il. G 
