68 
PHYSICS: BARUS AND BARUS 
around its axis a with the speed of N turns per second and if there are n 
segments, A, the corresponding mean current passing through the galvanom- 
eter, with inclusion of the value of A above, on reducing to practical units 
will be 
/ = NDLHv/3.6 X 10 20 amperes (3) 
Let N = 100; D = 100 cm.; L = 100 cm.; H = 0.4 dynes per unit pole; 
v = 50 cm./sec. or about 1 knot. Then 
r 10 2 X 10 2 X 10 2 X.4 X 50 cvy1 „ l4 
1 = 3 6 X 10 2D = amperes, nearly. 
i.e., the current traversing the galvanometer per knot of axial motion through 
the earth's vertical field would scarcely be perceptible by an extremely sen- 
sitive galvanometer. 
In our trial of the apparatus we used a disc pattern like figure 3, putting 
both a galvanometer and a telephone in series, at G. On charging auxiliary 
armatures like B, B" in figure 1, with a small Wimshurst machine, a loud rattle 
was immediately heard in the telephone and the galvanometer showed an 
average current of 2 X 10~ 6 amperes at about 10 (or less) rotations per 
second. This slowly dropped owing to leakage of charge from the armatures; 
but at least 10~ 6 amperes persisted indefinitely. Discharging the armatures 
was followed with immediate permanent silence in the telephone. If we es- 
timate the difference of potentials of the armatures as 10 4 yolts, and the earth's 
field H (as above) to supply 5 X 10 -5 volts, the latter should produce 5 X 10 -5 
(2 X 10~ 6 /10 4 ) = 10 -14 amperes, which is of the order of values computed. 
Charging the armatures with 25 volts from a storage battery produced no 
result, as was to be expected. 
Intensification. — The drum was now to be modified as suggested in figure 2, 
though the experiments below were carried out with the disc form, figure 3, 
which is more easily constructed. All contacts must be synchronous and 
momentary. 
In the disc type of machine (fig. 3) (as here constructed about 2 feet in 
diameter), A, B, C, D (in front) are the segments, rotating around the axis 
m, and E, F the armatures behind the diagram. The metallic posts or com- 
mutators a, b, c, d stick out normally from the disc both in front and behind 
it. The vertical brushes d and b touching diametrically opposite posts in 
front successively, include the galvanometer, etc., G, K being an insulating 
holder. The horizontal brushes, g and h, in metallic connection with E 
and F, respectively, touch the other two posts successively behind the disc 
and store the charges of A and C on E and F.- H shows the earth's magnetic 
field and v the motion through it. 
If we suppose that any charge on the armature E, for instance, induces the 
same but opposite charge on the segment B, the charges successively pass- 
ing through the galvanometer for each quarter turn may be scheduled in 
