178 
this groove the magnetometer, with its attached 
magnet, slides ; so that, by placing the needle at 
different distances from the centre of the coil, 
the instrument may be used for measuring cur- 
rents differing very greatly in strength. 
The adjustment of the instrument consists 
in placing the magnetometer upon the table, 
without the field-magnet, and adjusting the 
whole so that the index is at zero of its scale; 
after which the magnet is put in its place, and, 
if necessary, one end is moved by means of 
the screw until the index again points to zero. 
The interpretation of the reading of the in- 
strument is a very simple operation. A scale 
along the edge of the V-groove shows the posi- 
tion of the magnetometer. It is necessary to 
know the strength of field at the needle, which 
is, of course, that due to the magnet, plus the 
horizontal component of the earth’s magnetism ; 
and then, if this strength of field in c.G.s. units 
be multiplied by the reading of the needle, and 
divided by the scale-reading of the magne- 
tometer, the result will be the current in am- 
péres. : 
This galvanometer was permanently mounted 
on one of the stone piers, and its connections 
were so arranged that it could be quickly 
thrown in or out of the circuit without inter- 
fering with the continuity of the same. ‘The 
direction of the current through the galvanom- 
eter could be instantly reversed ; and through- 
out the observations reversals were regularly 
made, so as to eliminate error arising from dis- 
placement of the zero. ‘The zero-point was 
adjusted, however, at the beginning of every 
series of observations, and examined at fre- 
quent intervals during the same. 
A differential galvanometer was also used in 
measuring current strength. This instrument 
was kindly loaned to the jury by Professor 
Brackett of Princeton, by whom it was devised 
and constructed. A description of it will be 
found in the American journal of science, vol. 
xxi. p. 395. It consists essentially of two 
heavy rings, one within the other, through 
which the current goes in opposite directions. 
The needle in the centre has a silk-fibre sus- 
pension. The radii of the rings being given, 
the constant of the instrument can be readily 
calculated. It was especially constructed for the 
measurement of strong currents, ten ampéres. 
giving a deflection of fourteen to fifteen de- 
grees when the value of H is a little more than 
.2. When the current was steady, it behaved 
admirably ; but with a fluctuating current it 
became extremely difficult to get trustworthy 
readings, owing to the constant vibration of the 
needle. The graduated circle was so small that 
SCIENCE. 
‘and the Brackett 9.77. 
ae 
, a 
[Vov. III., No. 54. 
the estimation of fractions of a degree was quite 
uncertain. Even with steady currents, more 
time was required for reading the Brackett than — 
the Thomson, owing to the length of time 
needed for the needle to come to rest after a 
reversal of current. For these reasons its use — 
was not continued throughout the test. Itwas — 
observed continuously during the tests of the 
Thomson-Houston dynamo, and during a part 
of the tests of the Weston dynamo, when, owing 
to fluctuation in the current, its use was neces- 
sarily discontinued. It served a useful purpose, 
however, as a check upon the indications of 
the Thomson instrument, the close agreement 
of the two justifying confidence in the indica- 
tions of the latter. The most carefully made 
series of comparisons was that of Sept. 20. 
During the afternoon of that day, eight simul- 
taneous readings of the two instruments were 
made; the current, which was remarkably 
steady, being furnished by the Thomson-Hous- 
ton dynamo. The means gave 9.92 amperes 
as indicated by the Thomson, and 9.93 am- 
péres for the Brackett. The regular tests were 
begun on Sept. 25, and in the mean time 
several additional conducting-wires had been 
brought into the testing-room. The indica- 
tions of the Brackett galvanometer were, after 
this, constantly somewhat less than those of 
the Thomson, which was doubtless due to the 
alteration of the field by the presence of the 
currents. The difference was quite constant, 
and amounted to about two per cent. Thus, 
on Sept. 25 the Thomson gave 9.97 ampéres, 
On the 26th the Thom- 
son indicated 10.0, and the Brackett 9.80, and 
on the 27th, with the Weston are dynamo, 
showed 18.6 ampéres, and the Brackett 18.3. 
That this discrepancy could be accounted for 
by the effect of the current upon the strength 
of field was established by vibrating a needle 
under two conditions, — with and without the 
currents. Herein is shown the advantage of 
a strong, permanent magnetic field, such as 
exists in the Thomson instrument. An altera- — 
tion of the field, which might considerably in- — 
fluence the results from the Brackett, would — 
hardly be perceptible with the Thomson. M 
During the tests of the arc-light machines 
the whole current was taken through the gal- 
vanometers. With the incandescent systems, — 
however, in which the current was sometimes 
as high as 170 ampéres, this was impossible; — 
as tle coils and connections would have been 
greatly heated. The current might possibly 
have been safely divided between four or five 
instruments ; but, these not being at hand, it 
became necessary to make useof a shunt. 
