hertz's experiments. 217 



tliroug'h wliicli the osciilatiiig- ciiriciit rushes backward and forward. 

 Once the air gap lias been broken tlirough, tlie character of the air gap 

 as au opponent of the passage ol" electricity is coniph^tely changed. 

 Before the air gap breaks down it requires a considerable initial dif- 

 ference of electric pressure to start a current. Once it has been broken 

 down, the electric current oscillates backward and forward across 

 the incandescent air gap until the whole difterence of electric pressure 

 has subsided, showing that the broken air-gap has become a conductoi' 

 in which e\ en the feeblest electric pressure is able to produce an elec- 

 tric current. If this were not so, Leyden jars would not be discharged 

 by a single s]»ark. 



All this is (pute in accordance with what we know of air that is — or 

 even has lately been — incandescent; such air conducts under the feeb- 

 lest electric force. All this is most essential to the success of our oscil- 

 lator. ( )nly for this valnable pro[)erty of air, that it gives way suddenly, 

 and thence forward offers but a feeble opposition to the rapidly alternat- 

 ing discharge, it would have been almost impossible to start these rapid 

 oscillations. If we wish to start a tuning fork vibrating we must give 

 it a sharp blow; it will not do to press its prongs together and then let 

 them go slowly; we must apply a force which is short-lived in compari- 

 son with the i)eriod of vibration of the fork. It is necessary then that 

 the air-ga]) must l)reak down in a time short compared witli the rate of 

 oscillation of tlu^ discharge; and when this is rcipiired to be at the rate 

 of 400,000,000 per second, it is evident liow very remarkably suddenly 

 the air-gap breaks down. From the exi)eriments themselves it seems as 

 if any even minute roughness, dust, etc., on the discharging surface 

 interfered with this rapidity of break-down; it seems as if the points 

 spluttered out electricity and gradually broke down the air-gap, for 

 the vibrations originated are very feeble unless the discharging- surfaces 

 are kept highly jjolished; gilt brass knobs act admirably if kept pol- 

 ished np every ten minutes or so. One of the greatest desiderata in 

 these experiments is some method of making sui'e that all the sparks 

 should iiave the same character and be all good ones. 



III. 



In the foregoing, the ])rinciples ui)()ii which a rapidly vibrating elec- 

 tric oscillator should be constructed have l)een considered, and how the 

 sudden break-down of the air-gai) enabled these rapid vibrations to be 

 started. It is probable that this break-down occurs in a time smaller 

 than the thousand-millionth of a second. How very rajtid the inter- 

 atonn'c motions nnist be! 



Consider now the princii)les on which an apparatus is to be con- 

 structed to receive the vibrations pi-oduced by this oscillator. We may 

 observe in the tirst place that as we are dealing with a succession of 

 impulses at equal intervals of tinn^ w(^ (;an utilize resonance to accuuui- 

 late the effect of a single impulse. Kesouance is used in an immense 



