122 Mr. Gr. V. MacLean on the 



touching the key in the oscillator circuit, sparking at the 

 oscillator began, the electric waves traversing the free space 

 act upon the coherer, and at. once there is or is not an 

 increased reading given by the milliamperemeter-needle. 

 By moving the carriage to different distances from the re- 

 flector, and repeating the above operations, different readings 

 are found. There is seen to be a regular increase and de- 

 crease in the milliamperemeter-needle readings as the coherer 

 is moved farther and farther from the reflector. At certain 

 places no change in the readings is observed, while at certain 

 other places there is a maximum reading. Thus by means of 

 the coherer we are enabled to locate the nodal and antinodal 

 points of the electric waves, and hence to determine the 

 wave-length. 



Many precautions had to be taken during the course of 

 the experiment. The coherer responds to any (secondary) 

 electric spark ; the milliamperemeter readings differ for 

 different kinds of sparks. Not only will the coherer respond 

 to the sparks from the oscillator, but it responds to an}' other 

 (secondary) spark which may take place in any other part 

 of the building, or on a distant trolley-wire. So sensitive 

 is the coherer that the slightest jarring of the room, such 

 as caused by the slamming of a door, or by the walking of 

 persons in the corridor, will be sufficient to very materially con- 

 fuse the readings. Satisfactory and trustworthy measurements 

 can only be made when such disturbing causes are not 

 present. Accordingly the night hours were selected in which 

 to carry on the observations. If the mercury in the mercury- 

 cup of the break becomes coated, the nature of the secondary 

 spark changes, and so the readings under such conditions 

 are not correct. If too strong a current be sent up from the 

 storage-battery into the Ruhmkorff coil, the nature of the 

 secondary sparks changes so much that the readings must be 

 disregarded. The greatest care had to be exercised to pre- 

 serve the equality of the oscillator sparks throughout the 

 whole of the measurements. 



It was further noticed, whenever the key of the oscillator 

 circuit was closed (the coherer-key being first closed), that 

 there was always a very distiuct increase in the milliampere- 

 meter readings, even before the sparking at the oscillator began. 

 However, as soon as the sparking at the oscillator commenced, 

 this increased reading was augmented more or less according 

 to the distance of the coherer from the reflector. This 

 peculiar motion of the needle before the sparking begins at 

 the oscillator seems to point to an electrostatic effect upon 

 the coherer- globules, due to electric or magnetic causes 



