672 Prof. Fleming and Mr. Dyke on the Production of 



other, which can be equivalent to two radiotelegraphic 

 stations separated by several hundred miles. 



In order to supply the necessary conditions for quantitative 

 work, one o£ these closed circuits must be made the seat of 

 perfectly constant oscillations, damped or undamped, by 

 preference damped oscillations. The appliances which have 

 now been in use in the radiotelegraphic research Laboratory 

 at University College, London, for many years past for this 

 purpose are as follows : — 



The source of electromotive force may be an induction-coil 

 or transformer. If a transformer, it must then be operated 

 by a current from some supply circuit or from an alternator. 

 If an induction-coil, it is preferably operated by large 

 secondary cells, the primary current being interrupted by 

 some good form of mercury coal-gas brake. We have used 

 for this purpose with great advantage the Beclere mercury 

 coal-gas brake, in which a jet of mercury is raised by a rotary 

 pump and squirted against the copper plate in an atmosphere 

 of coal-gas. The mercury brakes that are generally sold 

 for use with paraffin oil as the insulating medium for the 

 mercury, are very messy in use and give great trouble by 

 necessitating constant purification of the mercury and are 

 by no means constant in action. Any mercury turbine 

 brake, however, can be converted into a coal-gas brake 

 by making the vessel gas-tight and employing, instead of 

 paraffin oil, an atmosphere of coal-gas supplied from a small 

 rubber bag under slight pressure. We have also used with 

 even greater advantage for some time a mercury turbine 

 brake by Schall, thus converted into a coal-gas brake which 

 will work for hours at a time for many months without the 

 slightest attention. The next element in the oscillatory 

 circuit is the spark-gap. To obtain perfectly constant 

 results, it is necessary to cause a jet of air to impinge upon 

 the spark-gap to destroy the arcing which otherwise would 

 take place, and, as shown in another paper, the result of this 

 air-blast is to remove the causes of the irregularity in the 

 discharge current (see the following paper on " The Effect of 

 an Air-Blast upon the Spark-Discharge of a Condenser charged 

 by an Induction-Coil or Transformer," by J. A. Fleming and 

 H. W. Richardson). 



It is also desirable to enclose the spark-balls in a cast-iron 

 chamber for silencing purposes. 



The form of spark-gap therefore used by the authors consists 

 of a cast-iron chamber S (see fig. I) closed by a lid with a glazed 

 peephole in it, the distance of the spark-balls being adjustable 

 by a screw, and a glass jet J being arranged so as to cause a 

 steady jet of air from a small Lennox blower under a pressure 



