202 Mr. G. Marconi [June 13, 



quently the subject of litigation between the companies concerned. 

 Prof. Fleming, who has witnessed the working of a great number of 

 syntonic wireless telegraph stations, was sufficiently impressed by 

 what he saw to make the following statement in his Cantor lectures 

 on " Electrical Oscillations and Electric Waves," delivered before the 

 Society of Arts in December 1900 : — " The objections as to inter- 

 ference of stations which imperfectly informed persons are in the 

 habit of raising with regard to Mr. Marconi's system of wireless 

 telegraphy, as a matter of fact no longer exist." 



I shall now say a few words on the subject of the detector of the 

 electric waves, called sometimes " the electric eye," which consists of 

 that essential part of the receiving apparatus especially aifected by 

 the electrical oscillations. In all wireless telegraph apparatus used up 

 to quite a recent date, a detector, now called a coherer, has been 

 employed. This detector is based on discoveries and observations 

 made by S. A. Varley, Prof. Hughes, Colzecchi, Onesti, and especially 

 Prof. Branly. Prof. 0. J. Lodge has made large use of this apparatus, 

 which he first named " coherer," in the very numerous experiments 

 and studies he has carried out on the effects produced by Hertzian 

 waves. The form of coherer I have found most trustworthy and 

 reliable for long-distance work consists of a small glass tube about 

 4 cm. long, into which two metal pole-pieces are tightly fitted. They 

 are separated from each other by a small gap, which is partly filled 

 with a mixture of nickel and silver filings. Provided such a coherer 

 is properly constructed, and the tapper and relay in good adjustment, 

 it proves to be quite reliable when within range of the transmitting 

 station. Experiments with syntonic systems have, however, shown 

 that certain kinds of coherers can be far more advantageously employed 

 than others. One apparently all-important condition is that the resist- 

 ance of the coherer in its sensitive state, or after being tapped, should 

 appear to be infinite when measured with an E.M.F. of about 1 volt. 

 If the tapping does not entirely do away with the conductivity of the 

 filings very poor results are obtained, which can be explained as follows. 

 According to the systems I have described, electrical syntony between 

 the transmitter and receiver is dependent on the proper electrical 

 resonance of the various circuits of the transformers used in the 

 receivers. The condenser and secondary of the transformer must not 

 be partially short-circuited by the coherer, otherwise the oscillations 

 cannot mount up or sum up their eifect, as is essential in order to 

 produce the difference of potential at the ends of the coherer neces- 

 sary for breaking down its resistance ; but the electrical oscillations 

 will leak across the conductive coherer without causing it to record 

 any signal. Of course, tlie condenser is short-circuited when the 

 filings cohere under the influence of the received oscillations ; but in 

 this case the signal is already recorded, and the tapper at once restores 

 the coherer to its non-conducting condition, and in this way restores 

 its sensitiveness. 



By using coherers containing very fine filings the necessary con- 



