70 
PHYSICS: BARUS AND BARUS 
troscope and electrometer were also tested; but these instruments are un- 
suitable as on open circuit the potential may suddenly rise to sparking values 
and ruin them. 
Great care was taken to secure all insulators firmly and to allow only metal 
appurtenances (brushes) to touch each other. 
The sounds (taps) were usually strong on starting the intensifier. As the 
speed increases to about 5 or 10 rotations per second a maximum of cur- 
rent was reached. This may be an average current of =±= 10 -5 amperes or more, 
as specified. The current frequently has an approximately regular period of 
reversal; as for instance (L, low current, here about — 2.5 X 10 -6 amperes; 
H, high current, + 5 X 10 -6 ampere, both fluctuating). 
Successive currents L H L H L H L H L H L H, etc. 
Period of each (seconds) 25 30 30 30 30 35 45 35 40 35 45 35 etc. 
The current is zero when the armatures are metallically connected. This 
shows that stray magnetic lines from the motor are without direct effect, a 
conclusion which was further tested by supplying stationary looped magnetic 
lines from a strong horseshoe magnet. Left to itself the machine picks up 
potentials either from within itself, incidentally, or from the room or sur- 
sounding walls in cold weather; and these are much higher than any which 
can be produced with a moving magnetic field. The machine, thoroughly 
flame-cleaned on both sides, with armatures in contact, soon* charges itself, 
the maximum potentials being usually approached by successive alternations 
of rapidly increasing amplitude. Finally sparks appear at the brushes. If a 
bunsen flame is near, its depolarizing effect is appreciable at about a foot; 
the machine ceases to function at about six inches. Only under favorable 
conditions were we able to build up potentials from the induction of a small 
horseshoe magnet. It is obvious therefore that before putting such an ap- 
paratus to the final test it will have to be constructed with precision and ex- 
periments made in an environment free from stray potentials, with elimina- 
tion of voltaic potentials, etc., in the parts, as everything of this nature is 
so quickly assimilated and intensified. As to the means of rating the value 
of the inducing potential, the current obtained on disconnecting the armatures 
for a definite small interval of time suggests itself. 
The efficiency is enormously increased by diminishing the distance between 
the armatures and the rotating segments (condenser) though in the above 
improvised apparatus it was not possible to approach closer than an average 
half inch. 
