186 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1907. 
Time only permits me to give you a brief sketch of some interest- 
ing experiments which have been carried out lately by the German 
Wireless Telegraph Company between Berlin and their large station 
at Nauen, 20 miles distant. At the transmitting station they employ 
12 electric ares in series, each of which is composed of a carbon nega- 
tive and a water-cooled capper positive electrode. These arcs take 4 
amperes at 440 volts. (See fig. 14.) In parallel with this series of 
arcs is joined a condenser and inductance, to which is inductively but 
loosely coupled an antenna from which undamped electric waves, 800 
meters in wave length, are radiated, having a frequency, therefore, of 
400,000. The oscillations set up in this antenna can be more or_less 
enfeebled by shunting them to earth through a microphone trans- 
mitter, the resistance of which is varied by the act of speaking against 
it. Hence, although the wave length of the emitted electric waves is 
not altered, their intensity is modulated in accordance with the wave 
Sending Antenna 
Receiving Antenna 
Wic. 15.—Wireless telephony by electric waves. 
form of the sounds impressed on the transmitter diaphragm. At the 
receiving station there is a receiving antenna tuned to the wave 
length used, having a quantitative electrolytic detector in connection 
with a telephone coupled inductively to the antenna circuit. Hence 
the vibrations of the transmitter diaphragm vary the intensity of the 
radiated electric waves but not their wave length. These waves travel 
through space, fall on the receiving antenna and affect the resistance 
of the electrolytic detector in proportion to their intensity. Hence 
the receiving telephone repeats the sounds or articulations made 
against the transmitting microphone and reproduces speech. The 
German experimentalists say that a satisfactory wireless transmission 
of speech can be made in this manner, 20 kilometers or 12 miles over 
water with antenne 25 meters or about 80 feet high. 
Ruhmer has recently described in the Elektrotechnische Zeitschrift 
some similar experiments made with a 220-volt Poulsen are. In this 
