( 410 ) 
the telephone. And with 15 A the self-induction is also greater thai 
with the other four A-coils, so in this respect also 15 A will be the 
least suitable for pure articulation. 
With 15 B of course the double curves do not appear at all and the 
self-induction is much less than with 15 A. So we can expect 15 B 
to articulate better than 15 A. 
In order to investigate in how far difference of articulation was 
perceptible with the different coils, I compared 3 A with 15 B, likewise 
6A with 15B and finally also 15 A with 15B. Of course the 
battery was chosen in such a way that the intensity of the sound 
remained the same with the two coils under comparison. So for 
instance in using 15 B (see table on pag. 406) 3 times more elements 
had to be taken than for 6 A. It was however evident, that the 
microphone got too much current with 3 Leclanché-cells and 15 B, 
as, even though no sound was made in its vicinity, it began to 
vibrate and to make a noise in the telephone. I therefore made 
use for these experiments of a thermo-electric battery of GULCHER 
consisting in all of 66 couples. In comparing 6 A with 15 B I made 
the former coil act with 7 couples, the other with 2! couples; in 
comparing 15 A with 15 B [I used respectively 15 and 21 couples, 
ete. Now an article of a newspaper was read before the microphone 
and by pressing down or releasing 4 Morse-keys the 2 coils under 
comparison were exchanged. It appeared that not the slightest 
difference in articulation was perceptible, either when comparing 
3 A with 15 B, 6A with 15B, or 15 A with 15B. Women’s voices 
generally sounding clearer out of our small telephones (with small, 
thin diaphragms) than men’s voices, the experiments were also made 
with these, but with the same negative result. The telephone spoke 
equally clear in all the different cases. 
To investigate whether in a musical sound change of quality 
would be perceptible when the coils were exchanged, the experi- 
ment was made with a musical box and also with the tuning-fork 
Fa 3, but without any result. To avoid the influence of the proper 
tones of the telephone-diaphragm, the telephone was substituted by 
a condensor with a permanent charge of about 32 volts, but the 
result was the same. 
As in many cases, also here practice has found the right way. 
The induction coils with iron and 4 or 6 primary layers of wire in. 
