216 
PHYSICS: C. BARUS 
It is difficult to account for this effect of frequency, so peculiarly marked in 
the instance given, where the observations were good. If different harmonics 
are in action, the overtones would have to respond in case of the wires under 
less tension, and this seems unlikely. The behavior of very tense steel wires, 
moveover, was the reverse of this. It is probable that the phenomena of §6 
contain the clue. 
6. Adjustable telephone. — The question as to the most advantageous position 
of the telephone is thus variously important. Consequently the telephone 
ihy figure 1, was mounted on a stout flat steel spring controlled by a micrometer 
screw. By actuating this, the face near g' could be approached as near g' 
as permissible, or withdrawn to a remoter position, with precision. The ex- 
periments made at length showed that within a wide range of tensions and of 
telephone positions, a particular degree of approach of the telephone corre- 
sponded to each particular stress of wire. Unless these paired positions are 
selected, the bifilar system does not respond. Tense wires require a nearer 
telephone; less tense wires a more remote telephone (even 0.5 cm.), within wide 
limits. Therefore the condition of resonance may be reached either by ad- 
justing the telephone on the micrometer screw for a given tension of wire; 
or by changing the tension of the wires for a fixed telephone. Between the 
admissible limits of tension, the sensitiveness passes through a flat maximum. 
With a distance of not more than 50 cm. between slit and ocular, a judicious 
disposition of parts eventually gave me 40 ocular scale parts per virtual mil- 
liampere, viz., 
n 
i X 103 
5X 102 
amperes 
cm. 
15 
0.67 
27 
20 
0.67 
27 
so that here the above effect of frequency, is no longer apparent. 
7. Coil tester. — An interesting application of the above apparatus, where a 
definite frequency is usually assigned, is its possible use for measuring the 
magnetic fields of different coils. For this purpose I wound a long thin pri- 
mary of an induction coil, which when inserted into the coil to be tested, should, 
from the measurement of the current induced in the secondary in question and 
in the absence of other mutual inductions, give the constants of the secondary. 
As many of the coils to be tested were of small internal diameter, the primary 
was wound on a long iron tube, fine wire being necessary. The data are for 
the 
Iron tube Helix 
Diameter outside 0.635 cm. 
Length 55 cm. 
Walls .0.08 cm. thick 
Diameter outside 0.7 cm. 
Wire 0.034 cm. in diameter 
nilh 21 turns per linear cm. 
