142 



NA TURE 



Decembjr 2, 1909 



same time there is some penetration of the wave into the 

 earth, and consequent dissipation of energy. 



Dr. Zenneck has considered the case of electric waves 

 1000 feet in wave-length, and has represented the final 

 result by some interesting curves. He defines the effect 

 of the absorption of energy by the soil by stating the 

 distance in kilometres at which the wave amplitude would 

 be reduced by the effect of this absorption to 0-367=1/6 

 of its amplitude at the sending station, altogether apart 

 from the weakening due to the spreading of the waves 

 out in a hemisphere, which we may call the spherical or 

 space decrease. These curves are plotted to abscissae re- 

 presenting the specific resistance of the soil (Fig. 3). You 

 will see from this diagram that when a plane electric wave 

 having the above wave-length is propagated over sea 

 water, it would have to travel 10,000 kilometres before 

 its amplitude would be reduced in the assigned ratio, and 

 over fairly dry soil about 100 to 1000 kilometres ; but over 

 very dry soil, having a small dielectric constant, only 

 about I to 10 kilometres. Also vou will notice that the 

 curves rise up again for still higher resistivities. This, 

 of course, is as it should be. .Ml the practical cases lie 

 between two ideal extremes : the case of an infinitely 

 perfectly conducting earth, in which case the waves would 

 not penetrate into it at all, and the other case, an infinitelv 

 perfect non-conducting earth, in which the wave would 



i XI. 



large capacity, and the inductance is kept small. If the 

 capacity is measured in electrostatic units, and the in- 

 ductance in electromagnetic units, the ratio of capacity to 

 inductance may be something of the order of 5/1 or even 





2 ,- 



Fig. 3 — Cu 

 (300 metre 



[Jr. ,Wat«! D^">PSo;i. | Dry Soil. | 



Specific Resistance in Ohms per Metre Cube. 



es showing the Distance in which Electric Waves i 

 in length have r 

 ac«s. (Dr. Zen 



elling ( 



penetrate into it, but would sufifer no dissipation of energy. 

 This theory is quite in accordance with practical experi- 

 ence in radio-telegraphy. Every receiving apparatus 

 associated with an antenna of a certain height and kind 

 must be subjected to waves of a certain minimum ampli- 

 tude to give any appreciable signal. For all lower ampli- 

 tudes that particular receiving arrangement is perfectly 

 deaf. Now it is a matter of common experience that with 

 a given radio-telegraphic apparatus and antenna it is 

 possible to receive signals for greater distances over sea 

 water than over dry land, and that if the soil is very dry 

 the distance may be cut down very considerably indeed. 

 This is not due merely to the difficulty of making what 

 the telegraphists call a good earth at the sending station, 

 it is due to the absorption of the wave by the earth for 

 the whole distance which extends between the two stations. 

 Hence, also, it is a common experience that when particu- 

 larly dry weather is succeeded by wet weather the radio- 

 telegraphic communication between two stations on land 

 is considerably improved. 



The next point in connection with the antenna to be 

 noticed is the means adopted of setting up the oscilla- 

 tions in it. _ The universal custom at present is to excite 

 oscillations in a reservoir circuit consisting of a condenser 

 and an inductance by means of the spark or arc. If the 



Fig. 



20/1. In this case the condenser is charged by means 

 of an induction coil or transformer, and discharged across 

 a spark-gap, and this discharge consists of intermittent 

 trains of electric oscillations with a periodic time equal 

 to the free natural period of the oscillatory 

 circuit. These discharges are made to succeed 

 each other from 50 to 600 times a second by 

 using an induction coil with an appropriate inter- 

 rupter, or else an alternator and a transformer. 

 If the arc method of exciting the oscillations is 

 employed, then the ratio of capacity to inductance 

 must be much smaller, and the oscillations are 

 excited in this circuit by a continuous current arc 

 worked with a voltage from 200 to 400 volts or 

 more, the arc being traversed by a strong mag- 

 netic field, and generally being placed in a 

 chamber kept free from oxygen. The oscillations 

 set up in the condenser circuit are then persistent 

 or unbroken. The oscillations are excited in the 

 antenna by coupling it inductively or directly with 

 the condenser circuit (Fig. 4). If the former 

 method is employed, then an oscillation trans- 

 former is used consisting of two coils of wire, one 

 coil being inserted in the condenser circuit and 

 one in the antenna circuit, and according as these 

 coils are near or far apart they are said to be 

 ^^5 closely or loosely coupled. These two circuits 



have, then, each their own natural period of 

 electric vibration, like tuning-forks, and they have 

 to be adjusted to syntony. It is well known that 

 under these conditions oscillations set up in one 

 circuit immediately create oscillations of two 

 frequencies in both circuits. This action can 

 be easily illustrated by two pendulums, which are of 

 the same length and are hung side by side on a 

 loose string distinguished by red and blue bobs. If 

 one pendulum is set swinging it imparts little jerks 

 to the other and sets the latter in motion, but to do 



Red Pendulum 



Blue Pendulum 



this the first must part with its own energy, and 

 hence is gradually brought to rest. Then the operation 

 is repeated in the reverse direction. The motion of each 



-- — -, ". ". -. -. — I pendulum may then be represented by the ordinates of a 



spark method is used, then the condenser is one of relatively curve such as' those in Fig. 5. This kind of motion can, 

 NO. 2092, VOL. 82] 



