TELEGRAPHY 
 
 FORTESCUE 
 
GIFT OF 
 E.P.Lewis 
 
 Physics der>t 
 
The Cambridge Manuals of Science and 
 Literature 
 
 WIRELESS TELEGRAPHY 
 
CAMBRIDGE UNIVERSITY PRESS 
 
 FETTEK LANE, B.C. 
 C. F. CLAY, MANAGER 
 
 : 100, PRINCES STREET 
 H. K. LEWIS, 186, GOWER STREET, W.C. 
 WILLIAM WESLEY & SON, 28, ESSEX STREET, STRAND 
 Berlin : A. ASHER AND CO. 
 fUtp>i s : F. A. BROCKHAUS 
 $eto gorfc: G. P. PUTNAM'S SONS 
 iSombau anti Calcutta: MACMILLAN AND CO., LTD. 
 
 All ri 
 
 reserved 
 
WIRELESS 
 TELEGRAPHY 
 
 BY 
 
 C. L. FORTESCUE, M.A. 
 
 Professor of Physics, Royal Naval 
 College, Greenwich 
 
 Cambridge : 
 at the University Press 
 
 New York : 
 G. P. P'ltnain's "Sons 
 
F7 
 
 PHYSICS DEPT. 
 
 PHYSICS DEPT 
 
 Cambridge: 
 
 FEINTED BY JOHN CLAY, M.A. 
 AT THE UNIVERSITY PRESS 
 
 With the exception of the coat of arms 
 at the foot, the design on the title page is a 
 reproduction of one used by the earliest known 
 Cambridge printer, John Siberch, 1521 
 
PREFACE 
 
 TN this book the author has had in his mind's eye 
 J- the reader who, possessing a general scientific 
 knowledge, is anxious to know something, not only 
 of the accomplishments of wireless, but also of the 
 means by which they are attained. The subject is 
 necessarily a highly technical one, and the first four 
 chapters are devoted solely to explanations of the 
 electrical phenomena involved. Which explanations 
 it may be added are really little more than state- 
 ments of facts illustrated where possible by mechanical 
 analogy. The fifth and sixth chapters deal with the 
 application of these principles to wireless apparatus, 
 and from the seventh chapter onwards the book is 
 devoted to a general survey of the uses to which 
 wireless is nowadays put. No forecast of the future 
 has been attempted as at the time of writing the 
 developement is so rapid that it is beyond anyone to 
 foresee the goal to which it is leading. 
 
 A short bibliography has been added including 
 most of the better known books on the subject. 
 
 6C5380 
 
vi PREFACE 
 
 The author's thanks are due to Marconi's Wire- 
 less Telegraph Co. Ltd. and Messrs Siemens Bros, 
 and Co. Ltd. for the use of photographs of Marconi 
 and Telefunken apparatus ; and to Mr H. W. Gregson 
 for reading through a large proportion of the proofs 
 and offering many valuable suggestions. 
 
 C. L. F. 
 
 LONDON. 
 March 1913. 
 
CONTENTS 
 
 CHAP. PAGE 
 
 I. Introduction . 1 
 
 II. Oscillatory currents, resonance and tuning . 17 
 
 III. Electromagnetic waves 29 
 
 IV. Absorption and atmospherics ... 40 
 V. The transmitting instruments ... 50 
 
 VI. The receiving instruments . . . . 66 
 
 VII. The uses of wireless telegraphy on board ship 77 
 VIII. The shore stations 95 
 
 IX. The use of wireless telegraphy between fixed 
 
 stations over land and sea ... 99 
 
 X. The uses of wireless telegraphy for naval and 
 
 military purposes 116 
 
 XI. Wireless telegraphy on airships and aero- 
 
 planes 120 
 
 XII. Wireless telephony 124 
 
 XIII. History 132 
 
 Bibliography .140 
 
 Index 141 
 
LIST OF ILLUSTRATIONS 
 
 FIG. PAGE 
 
 1. Diagram of Alternating Current ... 6 
 
 2. Diagram of Oscillatory Current ... 8 
 
 3. Magnetic Field near a coil carrying an Electric 
 
 Current 9 
 
 4. Mutual Induction between two coils . . 11 
 
 5. Hydraulic model of a Condenser . . . 15 
 
 6. Time diagram of Train of Oscillations . . 21 
 
 7. Kadiative Circuit 31 
 
 8 a, b, c, d, e. Electromagnetic Waves from an 
 
 Aerial 36 
 
 9. Diurnal variation of strength of signals at 
 
 Clifden 42 
 
 10. Diagram of connections for Transmitting by 
 
 "Plain Aerial" 52 
 
 11. Diagram of Coupled Transmitting Circuit . 54 
 
 12. Marconi Kotary Spark Gap .... 59 
 
 13. Marconi's Magnetic Detector .... 68 
 
 14. Diagram of Receiving Instruments ... 71 
 
 15. Arrangement of instruments in the transmitting 
 
 compartment of a Marconi ship ... 87 
 
 16. The operating room in a Marconi ship . . 89 
 
 17. Telefunken Transmitting Instruments . . 91 
 
 18. Telefunken Receiving Instruments ... 93 
 
 19. The Nauen Wireless Station, near Berlin . 110 
 
 20. Arrangement of Wireless Apparatus fitted to 
 
 Aeroplanes 121 
 
CHAPTER I., 
 
 INTRODUCTION *''-- ' ' "- - 
 
 THE general principles on which a message is 
 sent from one place to another by means of Wire- 
 less Telegraphy are in many ways analogous to the 
 principles involved in sending a sound signal from 
 one point to another. In the latter case the trans- 
 mitting instrument, consisting of a bell or syren for 
 instance, is made to set the air surrounding it into a 
 state of rapid vibration. These vibrations spread 
 themselves in all directions in the form of sound 
 waves in the air, and impinging on the receiving 
 instrument, usually the human ear, give rise to effects 
 which enable the sounds to be recognised. 
 
 In a Wireless Telegraphy Installation a similar 
 process is taking place during the sending of a 
 message. Corresponding to the mechanical vibra- 
 tions of the bell or syren are the electrical oscillations 
 in the aerial at the transmitting station, whose func- 
 tion it is to set the aether surrounding it into a 
 state of violent electrical vibration. These electrical 
 vibrations in the form of electromagnetic waves in 
 
 F. W. T. 
 
2 WIRELESS TELEGRAPHY [CH. 
 
 the aether are transmitted in all directions and 
 impinging on the receiving aerial, can be recognised 
 by means of the receiving instruments. 
 
 The apparatus, required at the transmitting station 
 
 **:tb;set up : tjiese : vibrations, and at the receiving 
 
 % .statipns. to. recognise them, is, however, rather more 
 
 *-\: :c<fi^fic&tf$.ti$Vivt^t required to give rise to and 
 
 recognise the sound waves, when signalling by that 
 
 means ; and consequently before describing these 
 
 instruments it is proposed to give a brief account 
 
 of the electrical phenomena upon which their action 
 
 depends. 
 
 The most familiar idea of electricity is the Current 
 of Electricity, generally regarded as the flow of a 
 kind of intangible fluid whose behaviour is similar 
 to the flow of a stream of water along a pipe. 
 Electricity, looked upon in this way, will flow freely 
 through some materials, the most important of which 
 is copper, which are known as "Conductors" of elec- 
 tricity; but only with extreme difficulty through 
 others which are called "Non-conductors" or "Insu- 
 lators." Well-known materials of the latter kind are 
 air, glass, mica, ebonite, etc. A long copper wire 
 forms a path for electricity to flow along just as the 
 inside of a water-pipe forms a space along which a 
 stream of water can flow. The quantity of electricity 
 passing a point in the wire in, say, a second is 
 analogous to the quantity of water passing a point 
 
i] INTRODUCTION 3 
 
 in the pipe in a second. When a large quantity of 
 electricity passes in a second the current is spoken of 
 as being a large one, and vice versa. If the wire 
 along which such a current of electricity is flowing 
 is broken the current will cease because at the break 
 a non-conducting barrier has been introduced. This 
 is equivalent to putting a valve into a water pipe 
 and closing it, or to blocking up the pipe with some 
 impermeable material. 
 
 To force water along a horizontal pipe there must 
 be a difference of pressure between the two ends, 
 which must be maintained if the stream is to con- 
 tinue. Similarly in the electrical case, there must be 
 a difference of electrical pressure to cause electricity 
 to flow from one end of a wire to the other, which 
 must be maintained if the current is to continue. It 
 may be pointed out that this electrical pressure is 
 not of the same mechanical nature as the pressure in 
 the water-pipe. It cannot be measured in pounds 
 per square inch or tons per square foot ; but it is 
 nevertheless closely analogous to it. Electricity cannot 
 flow without a difference of electrical pressure, but a 
 difference of pressure can exist without a current if 
 there is no conducting path for the electricity to flow 
 along. Electrical pressure is often called Electro- 
 motive Force. 
 
 If water is supplied from a pressure main to a 
 water motor through a small pipe a large proportion 
 
 12 
 
4 WIRELESS TELEGRAPHY [OH. 
 
 of the pressure is lost and only the remainder is 
 available for working the motor. This is bad 
 engineering as the water is wasting its energy in 
 the pipe in overcoming friction instead of doing 
 useful work in the motor. To prevent this wastage 
 a larger pipe must be used. A similar thing occurs 
 in the electrical case. If a current of electricity is 
 sent along a small wire it wastes its energy, and to 
 prevent this wastage a thick wire must be used. 
 The thin wire is spoken of as having a "high resist- 
 ance" and the thick one as having a "low resistance." 
 For wireless purposes it is highly undesirable for the 
 current to waste its energy in this way, and so all 
 circuits should be of low resistance, i.e. of large sized 
 copper wire, and where the currents are very large, 
 large copper tubes must be used. Copper is used 
 because size for size the wastage is less with copper 
 than with any other practicable material. 
 
 Electricity can under some circumstances be forced 
 through a non-conducting break in a circuit. This 
 occurs with a high electrical pressure and is of the 
 nature of a disruptive spark bursting through the 
 insulating medium. After the initial spark has formed 
 the path through the insulator becomes momentarily a 
 fairly good conductor of comparatively low resistance 
 on account of the volatilisation of the materials. In 
 many of the transmitting instruments used for wireless 
 telegraphy a spark of this kind is continually taking 
 
i] INTRODUCTION 5 
 
 place between two conductors separated by a layer 
 of air. This point at which the sparks are formed 
 is generally known as the " spark gap." With air as 
 the insulating medium which is burst through, the 
 non-conducting state quickly restores itself when the 
 current ceases and the same high electrical pressure 
 will be necessary to produce another spark. 
 
 A stream of water may flow in many ways. For 
 instance it may flow steadily in one direction at a 
 uniform rate. If a tap, connected to a constant 
 pressure main, is left turned on this steady flow will 
 take place. Again it may flow backwards and for- 
 wards, first in one direction and then in the other as 
 is approximately the case at the mouth of a tidal 
 harbour. Corresponding to these two ways there are 
 two ways in which electricity can flow. It can flow 
 steadily in one direction, in which case it is known 
 as a Direct or Continuous Current. Or again it can 
 flow alternately in one direction and then in the 
 other, when it is known as an Alternating Current. 
 The variations of an alternating current repeat them- 
 selves regularly after certain intervals. Starting 
 from an instant at which the current is zero, as at 0, 
 fig. 1, the changes are as follows : at first the current 
 increases to a maximum value at A and then dies to 
 zero again at B. During this time the electricity has 
 been flowing in one direction only, but beyond B the 
 flow is in the opposite direction, the current being a 
 
WIRELESS TELEGRAPHY 
 
 [CH. 
 
 maximum at C and falling to zero again at />. This 
 series of changes is what is termed a "complete 
 cycle." The two halves of the cycle occupy the same 
 time and at corresponding instants the currents are 
 equal but of course in opposite directions. The 
 "frequency" of an alternating current is the number 
 of cycles passed through per second. 
 
 As well as these two there is a third important 
 way in which a current of electricity may vary, which 
 
 Currcnf 
 
 Current 
 
 Q t.( pmplptf ( VC(P^-' 
 
 Fig. 1. Diagram of Alternating Current 
 
 is confined almost entirely to wireless work. It may 
 I flow in groups of oscillations to and fro, the amplitude 
 of the oscillations of each group gradually decreasing 
 and the intervals between the beginning of each 
 group and the next being generally large enough for 
 the oscillations to die down entirely in the meantime. 
 Such currents are known as Oscillatory Currents. As 
 an analogy, suppose a U-tube having arms about a 
 yard in length to be filled with water to within about 
 
i] INTRODUCTION 7 
 
 a foot of the top. Under normal conditions the level 
 of the water would be the same in each arm, but it 
 could be made different by, say, blowing down one of 
 them. Let the two levels be made different in this 
 or some other way and let the constraint be suddenly 
 removed. The levels immediately tend to readjust 
 themselves and if the bore of the tube is not too 
 small there will be several oscillations up and down 
 before the water finally comes to rest. Suppose the 
 disturbances of the level to be repeated at regular 
 intervals, generally greater than those required for 
 the water to settle down, and the resulting motion of 
 the water is analogous to the oscillatory currents. 
 Each group of oscillations is called a " Train" of 
 oscillations and may consist of one or two only or 
 of a large number of them. The term "cycle" is used 
 with oscillatory currents to denote two consecutive 
 surges to and fro. These are almost similar to a 
 cycle of an alternating current, the only difference 
 being that with the oscillatory current the amplitudes 
 of consecutive half cycles are steadily diminishing 
 instead of remaining constant as with the alternating 
 currents. Fig. 2 is a diagram of an oscillatory cur- 
 rent. The term "frequency" is used with oscillatory 
 currents in practically the same sense as with alter- 
 nating currents. It is the number of cycles which 
 would take place in a second if the train continued 
 uninterruptedly for that time. In practice the trains 
 
8 
 
 WIRELESS TELEGRAPHY 
 
 CH. 
 
 do not last as long as this and there is usually a 
 period of quiescence between them. The actual 
 number of cycles passed through in a second is 
 therefore less than the frequency. For suppose that 
 in a given case there are 100 trains per second each 
 lasting for one-thousandth of a second and consisting 
 of 50 cycles. The actual number of cycles passed 
 through in a second will then be 100 x 50 = 5000. 
 But if one train had lasted uninterruptedly for a whole 
 
 Fig. 2. Diagram of Oscillatory Current 
 
 second the number of cycles passed through would 
 have been 50 x 1000 = 50,000. This latter, and not 
 the former, is the frequency of the oscillatory current 
 in question. It is in fact the rate at which the cycles 
 are being passed through during a train, expressed as 
 so many per second. 
 
 The frequencies met with in wireless work are for 
 alternating currents 25 to 1000 cycles per second and 
 for oscillating ones 50,000 to 3,000,000. 
 
 Each of these three different types of current 
 
i] INTRODUCTION 9 
 
 involves three corresponding electrical pressures or 
 electromotive forces. A direct current will be pro- 
 duced by a steady unidirectional electrical pressure ; 
 for an alternating one an alternating electromotive 
 force of the same frequency will be required ; and for 
 an oscillating one an oscillating pressure is necessary. 
 
 i . ' 
 
 Current UP ' 
 
 DOWN 
 Fig. 3. Magnetic Field near a coil carrying an Electric Current 
 
 When a current of electricity flows along a wire 
 it produces magnetic effects in the surrounding air. 
 Suppose for instance that the wire is wound up into 
 the form of a coil as shewn in fig. 3, and that an 
 electric current is sent round it. Then, anywhere in 
 
10 WIRELESS TELEGRAPHY [OH. 
 
 the neighbourhood of the coil a freely suspended 
 small magnet will be found to be strongly affected. 
 Instead of pointing N. and S. as it usually does it will 
 be found to set itself in directions indicated by the 
 arrows and lines in the figure. This behaviour of 
 the small magnet shews the presence of what is 
 called a "Magnetic Field" around the coil. This 
 magnetic field depends upon the current for its 
 existence and varies with it. 
 
 A current of electricity flowing round a coil has 
 further the important property of being able, under 
 suitable conditions, to give rise to currents in other 
 coils near it to which it is not in any way connected. 
 For instance, consider two coils A and B, fig. 4, 
 A being connected to a battery or other source of 
 current and B being quite separate from A. Then 
 whenever the current in A is made to change an 
 electromotive force comes into existence in the coil 
 B tending to make a current flow round it. If the 
 two ends of B are joined by a conducting wire 
 currents will actually flow round the coils ; but if 
 the ends are not joined the electromotive force will 
 be there and there will be no current. This electro- 
 motive force in B is called an "Induced Electromotive 
 Force," and the current it will give rise to if the 
 ends of the coil are joined together is called an 
 " Induced Current." The induced electromotive force 
 in B is brought about by the agency of the magnetic 
 
INTRODUCTION 
 
 11 
 
 field which invariably accompanies the electric cur- 
 rent, the existence of which is indicated diagram- 
 matically by the dotted lines in the figure. It is 
 really the change of this field round about the coil B 
 that gives rise to these effects. Suppose now an 
 alternating current is flowing round the coil A. The 
 
 A 
 
 \ \ 
 
 Fig. 4. Mutual Induction between two coils 
 
 current and its accompanying magnetic field will be 
 in a continual state of change with the result that 
 a varying electromotive force is being continually 
 induced in B. This electromotive force and the 
 current it can produce are in this case similar alter- 
 nating ones of the same frequency as the current 
 in A, and may be large or small depending on the 
 
12 WIRELESS TELEGRAPHY [OH. 
 
 relative position, size, etc., of the two coils. In the 
 same way if there is an oscillatory current in the 
 coil A there will be induced an oscillatory electro- 
 motive force in B of the same frequency and dying 
 away in the same manner. This property of two 
 circuits is spoken of as their "Mutual Induction." 
 They are also often spoken of as being " Inductively 
 Coupled." 
 
 Water flowing along a pipe has the property of 
 Inertia in consequence of which an effort is required 
 to start it flowing, and when once started gives it a 
 tendency to go on flowing on its own account. The 
 sudden closing of a valve at the end of a long pipe 
 along which water is flowing at a good speed has 
 been known to burst the pipe near the valve on 
 account of the pressure set up in bringing the water 
 to rest. A current of electricity flowing along a wire 
 has an analogous property, especially if the wire is 
 wound up into the form of a coil. To start the 
 current flowing quickly, a large electrical effort (an 
 electromotive force) is required. Once this current 
 has been started it will have a tendency to go on 
 flowing, so much so in some cases that it will burst 
 its way through a non-conducting break introduced 
 into the circuit. This is due to what is known as the 
 "Self Induction" of the circuit and it gives to the 
 current a property of electrical inertia analogous 
 to the inertia of water flowing along a pipe. A 
 
i] INTRODUCTION 13 
 
 circuit in which the effect is very pronounced is 
 said to be "highly inductive," and one in which 
 the effect is negligibly small is said to be "non- 
 inductive." Thus in a highly inductive circuit a 
 large electromotive force will only comparatively 
 gradually set a current going, whereas in a nearly 
 non-inductive circuit a small electromotive force 
 quickly runs up the current to its full value. A coil 
 made up so as to be especially used on account of 
 this property is called an " Inductance." Often also 
 the self induction of a circuit is spoken of as its 
 " Inductance." 
 
 The analogy of water flowing along a pipe is of 
 assistance in forming a mental picture of the corre- 
 sponding electrical phenomena, but there is one very 
 important difference between them which must not 
 be overlooked, and that is that the changes in the 
 electrical case are all very much more rapid than 
 could be the case with the water. A stream of water 
 could be set up in a few seconds, or if a considerable 
 force is exerted, in a fraction of a second. The 
 electrical case corresponding to this is one in which 
 a current is started up in a few millionths of a 
 second only. Similarly the oscillations of water in 
 a U-tube probably require a second or more to com- 
 plete a cycle, but the analogous electrical oscillation 
 completes a cycle in a few millionths of a second 
 or less. 
 
14 WIRELESS TELEGRAPHY [CH. 
 
 A Condenser is an instrument in which electricity 
 may be stored, not in the form of chemical energy as 
 in a cell or accumulator, but in the form of actual 
 charges of electricity. 
 
 It consists of two conducting bodies separated 
 from one another by any insulating material known 
 as the "Dielectric." Its property of being able to 
 store up electricity is known as its " Capacity." Many 
 condensers consist of a large number of thin plates of 
 metallic foil separated by thin layers of paraffined 
 paper, glass, air or mica. These condensers are 
 known as " parallel plate " condensers. The elevated 
 wires constituting the aerials used for wireless pur- 
 poses together with the surrounding surface of the 
 earth also form a condenser, the dielectric in this 
 case being the intervening air. 
 
 When a condenser has charges of electricity stored 
 in it it is said to be " charged." One conductor may 
 be looked upon as having an " excess of electricity " 
 or positive charge, and the other a " deficiency " or 
 negative charge. 
 
 All condensers have this property, namely, that as 
 the amount of electricity stored in them increases the 
 electrical pressure between the two conductors also 
 increases exactly proportionally. If the conducting 
 plates of a condenser are small and some distance 
 apart a large difference of pressure will be required 
 to store up a small charge. Such a condenser is said 
 
I] 
 
 INTRODUCTION 
 
 15 
 
 to have a small capacity. If the plates are large and 
 near together the condenser will have a large capacity. 
 The amount of electricity which can be stored up in 
 any condenser is limited by the maximum electrical 
 pressure which can exist between the two conducting 
 
 Fig. 5. Hydraulic model of a Condenser 
 
 bodies before a spark jumps across between them, re- 
 sembling on a small scale a flash of lightning. This 
 spark consists of an actual flow of electricity, or current 
 of electricity through the dielectric. As a result of this 
 flow the positive and negative charges run together 
 and neutralise one another, the condenser becoming 
 
16 WIRELESS TELEGRAPHY [OH. 
 
 thus "discharged." A condenser can be discharged 
 without damage to the dielectric by simply connecting 
 the two conductors together by a wire. The wire 
 then has a transient current of electricity in it during 
 the time that the charges are running together. 
 These "Discharge Currents," as is explained later, are 
 of the greatest value for wireless telegraphy purposes. 
 The behaviour of a condenser is well illustrated 
 by the hydraulic arrangement shewn in fig. 5. It 
 consists of a large cylindrical vessel A connected up 
 at each end to a pump P. Sliding in the vessel is a 
 watertight piston B controlled by the springs (7(7. 
 Under normal conditions this piston divides A into 
 two equal parts. If now the pump is set working 
 water will be drawn out of one side and pumped into 
 the other with the result that the pressure on one 
 side increases and on the other side decreases, the 
 piston being pushed over to the right or to the left. 
 The excess of water on one side of the piston corre- 
 sponds to the excess of electricity or positive charge 
 on one of the conductors in the electrical condenser ; 
 the deficiency on the other side corresponds to the 
 deficiency of electricity or negative charge on the 
 other conductor. The excess of water on one side 
 over that on the other is proportional to the move- 
 ment of the piston against the springs, which in turn 
 is proportional to the difference of pressure on the 
 two sides of the piston. 
 
n] OSCILLATORY CURRENTS, ETC. 17 
 CHAPTER II 
 
 OSCILLATORY CURRENTS, RESONANCE AND TUNING 
 
 THE importance of condensers for wireless 
 telegraphy lies in the fact that it is principally by 
 their aid that the necessary very high frequency 
 oscillating currents are produced. When a condenser 
 is charged up to a high voltage and is then allowed 
 to discharge through an inductive circuit, the result 
 is a single train of oscillatory currents. The action 
 is as follows : as soon as the inductive circuit is 
 connected up to the two charged plates of the con- 
 denser the excess of electricity on one starts running 
 out to the plate on which there is a deficiency, so 
 producing a current of electricity in the wire. The 
 electrical pressure between the plates will quickly 
 speed up this current to a high value in spite of the 
 self induction of the circuit acting like the inertia of 
 a flywheel and opposing the increase of the rate of 
 the flow. A time, however, soon comes when the 
 excess of electricity will have completely run out 
 of one plate of the condenser and it will be for an 
 instant uncharged. At this instant there will be no 
 pressure tending to keep the current flowing and it 
 consequently tends to stop, but owing to the inertia 
 effect of the self induction it will go on flowing with 
 
 F. W.'T. 2 
 
18 WIRELESS TELEGRAPHY [CH. 
 
 the result that an excess of electricity collects on the 
 plate which originally had a deficiency. The elec- 
 tricity is, as it were, gaining impetus as it flows out 
 from one plate which carries it into the other one 
 in spite of the fact that as it collects there a rising 
 pressure is produced tending to make it flow out 
 again. This rising pressure will eventually stop the 
 flow and start it back in the reverse direction ; the 
 whole process being repeated several times, back- 
 wards and forwards. 
 
 There are many mechanical analogies to this 
 oscillation of the current when a condenser dis- 
 charges through an inductive circuit. Take for 
 instance the water analogy suggested in fig. 5. Sup- 
 pose there to be an excess of water on one side of 
 the piston and the pump to be replaced by a large 
 valve which can be suddenly thrown fully open. 
 Immediately this is done the compressed springs will 
 force the water round from the side on which there 
 is an excess to the side on which there is a deficiency. 
 The speed at which the water flows round will go 
 on increasing so long as the springs are exerting 
 any force. The piston, however, soon reaches its 
 middle position where the springs exert no force. 
 But the stream of water would not stop at this 
 point. Its inertia would carry it on into the side of 
 the cylinder where originally there was a deficiency 
 but where there will now be an excess. As the result 
 
n] OSCILLATORY CURRENTS, ETC. 19 
 
 of this excess the piston is displaced beyond its middle 
 point and immediately the springs come into action 
 and oppose the further flow of the water, eventually 
 bringing it to rest. The springs, being compressed, 
 then force out the water in the opposite direction to 
 that of the first stream, the speeding up and over- 
 running being repeated several times. 
 
 This oscillation of the water backwards and 
 forwards will go on until the frictional resistance of 
 the pipes brings it to rest, and in point of fact it 
 would not be possible to make an apparatus of this 
 kind which would oscillate to and fro many times. 
 A similar thing occurs in the electrical case. Energy 
 will be wasted owing to the resistance of the wires 
 and other similar causes in consequence of which 
 the oscillations get less and less, finally dying out 
 altogether. Unlike the water analogy, however, the 
 electrical oscillations may go on, in some cases, for 
 several hundred cycles before dying out. 
 
 This dying away of the oscillations is what is 
 termed "Damping"; and the oscillation is called a 
 " Damped Oscillation." If it dies away very quickly 
 it is " strongly " damped, but if it dies away very 
 slowly it is said to be "weakly" or "slightly" damped. 
 In some cases it is possible by special means to 
 produce an oscillation of this kind which does not die 
 away. Such an oscillation is called an " undamped " 
 or "continuous" oscillation, and is often very desirable 
 
 22 
 
20 WIRELESS TELEGRAPHY [OH. n 
 
 and much sought after. It is really an alternating 
 current of very high frequency. 
 
 These oscillations are illustrated by the curves in 
 fig. 6 which shew the variation of the pressure 
 between the two plates of a condenser and the 
 current flowing into and out of them, during part 
 of an oscillatory discharge. They are represented by 
 the full and broken line respectively, the scales of 
 current and pressure being given. 
 
 Oscillations of this kind which are not too strongly 
 damped have the property that the time for one 
 complete oscillation is the same whether the oscilla- 
 tion has a large or small amplitude. This will be 
 noticed to be the case by referring to fig. 6. The 
 frequency of the oscillations is therefore constant 
 whatever their magnitude. 
 
 This frequency is known as the "Natural Fre- 
 quency " of the circuit. It is high or low depending 
 upon the magnitudes of the capacity and the in- 
 ductance. If both are small the natural frequency is 
 high and vice versa. That this is the case would 
 be expected from considerations of the hydraulic 
 analogy. If the piston has strong controlling springs 
 a small excess of water on one side of the piston 
 over that on the other will give rise to large forces 
 tending to equalise the amounts. This corresponds 
 to a condenser of small capacity in which a small 
 charge causes large electrical pressures. If the 
 
o_O 
 
 
 
 o 
 
 
 O 
 
 *C f 
 
 
 O 
 
 D 
 
 O 
 
 
 
 o 
 
 O 
 O 
 
 o 
 
 < ^ 
 
 trj 
 
 (M 
 
 > CJ 
 
 ir> 
 
 N_ 
 
 v_^. 
 
 
 
 1 
 
 1 
 
 1 
 
22 WIRELESS TELEGRAPHY [CH. 
 
 valve in the connecting pipe is opened the strong 
 springs will quickly force out the excess of water 
 which rushes quickly into the other side and is 
 equally quickly forced out again. Also the shorter 
 the connecting pipes the smaller will be the amount 
 of water the inertia of Avhich will have to be over- 
 come, the more quickly will the strong springs 
 set the water moving and the higher will be the 
 frequency of the oscillations. This corresponds to a 
 discharge circuit of small self induction. 
 
 From this analogy it will be easily understood 
 that small capacity and small self induction mean 
 high frequency, and large capacity and large self 
 induction low frequency. The frequency as a matter 
 of fact is inversely proportional to the square root of 
 the product (capacity of the condenser) x (self induc- 
 tion of the circuit), and so long as this product of 
 these two remains the same the frequency does not 
 in any way depend upon their individual values. 
 
 A circuit consisting of a condenser and an in- 
 ductance with low resistance in which an oscillatory 
 current can thus be set up is generally spoken of as 
 an " Oscillatory Circuit." 
 
 If an alternating or oscillating electromotive force 
 be applied to a circuit consisting of a condenser and 
 an inductance it will produce the greatest effect if its - 
 frequency is the same as the natural frequency of 
 the circuit. 
 
ii] OSCILLATORY CURRENTS, ETC. 23 
 
 Referring again to fig. 5, suppose that the pump 
 can be made to give small impulses to the water 
 firstly to the left, then to the right, and so on. These 
 small impulses if they are properly timed will produce 
 a large oscillation of the water. The timing of the 
 impulses must exactly correspond to the natural 
 frequency with which the water tends to flow 
 backwards and forwards on its own account. If 
 the timing is incorrect large effects are not pro- 
 duced, because no advantage is being taken of the 
 inertia of the water and the forces exerted by the 
 springs. 
 
 Many other examples of this same effect could be 
 ({noted. For instance, a company of soldiers marching 
 over a bridge receive the order to " break step," that 
 is to cease marching regularly all in step together. 
 The reason for this is that the regular impulse of the 
 whole company marching in step might coincide with 
 the natural frequency with which the bridge would 
 spring up and down on its own account if once 
 started. Accurate timing of the applied impulses 
 would then produce very serious oscillations in the 
 bridge and possibly the destruction of it. 
 
 The regular movement of a few sailors from one 
 side to the other of a large ship can be made to set 
 up a heavy rolling. They must move from side to 
 side exactly as required by the natural frequency 
 with which the ship rolls on her own account. The 
 
24 WIRELESS TELEGRAPHY [OH. 
 
 moving of vastly greater loads will produce negligible 
 effects if the timing is incorrect. 
 
 These mechanical illustrations are very closely 
 analogous to the electrical case. If the condenser is 
 charged and allowed to discharge through the in- 
 ductance, oscillations of current take place of a 
 definite frequency. This corresponds to the natural 
 frequency of the oscillations of the bridge or of the 
 rolling of the ship. The applied electromotive force 
 corresponds to the regularly applied impulses to the 
 bridge or the ship and like them produces the 
 greatest effect when properly timed. 
 
 In the mechanical case the applied impulses will 
 have to be reversed in direction after intervals which 
 may be measured in seconds or tenths of seconds. 
 In the electrical case the frequencies are very high 
 and the impulses will have to be reversed after 
 intervals measured in millionths of seconds. 
 
 The principle of the timing of small impulses so 
 as to produce large effects is known as the principle 
 of " Resonance." If the frequency of an applied 
 electromotive force is the same as that of the circuit 
 to which it is applied the two are sometimes spoken 
 of as being " in resonance." 
 
 Two oscillatory circuits are often so arranged 
 that there is mutual induction between them. In 
 that case when an oscillatory current is set up in 
 one of them a similar oscillatory electromotive force 
 
n] OSCILLATORY CURRENTS, ETC. 25 
 
 will be induced in the other. If the two circuits 
 have different natural frequencies this induced oscil- 
 latory electromotive force will not generally produce 
 a large oscillatory current. When however the fre- 
 quencies of the two circuits are the same the induced 
 electromotive force in the second circuit will be of 
 the right frequency to set up a large oscillatory 
 current there, and if the trains of oscillations last 
 long enough the oscillatory current will reach a high 
 value even though the electromotive force is itself 
 quite small. It should be remembered that in 
 speaking of a current being gradually set up in this 
 way each separate train of oscillations has to be 
 considered separately. During the intervals between 
 the trains in the first circuit no electromotive force is 
 being induced in the second and there is no current 
 there. 
 
 As is pointed out in the next chapter, two oscilla- 
 tory circuits placed a long way apart may be made to 
 influence one another by the aid of electromagnetic 
 waves. The impulses produced in one circuit by the 
 oscillating currents in the other are then extremely 
 small. Unless the timing of these impulses is exact, 
 the impulses cannot be expected to produce any 
 appreciable result. Properly adjusted, however, quite 
 an appreciable oscillatory current can be set up in 
 the distant circuit. 
 
 The adjusting of two circuits to the same natural 
 
26 WIRELESS TELEGRAPHY [OH. 
 
 frequency is known as "Tuning." The adjustment of 
 any circuit is made by varying .the self induction or 
 the capacity. In some cases, where a wide range 
 of adjustment has to be provided for, both can be 
 varied. 
 
 This tuning of two aerials to the same frequency 
 is of the utmost importance and is the fact upon 
 which the practical utility of wireless telegraphy 
 depends. All over the world there now exist hun- 
 dreds of wireless stations, some of which are very 
 powerful. Any one of these stations when "sending" 
 will radiate its electromagnetic waves in all direc- 
 tions. As they advance they will strike other aerials 
 and give rise to oscillating electromotive forces in 
 them. The effects of these electromotive forces are, 
 however, generally negligible unless the receiving 
 aerial is tuned to the frequency of the sending aerial. 
 This tuning can be effected by a simple adjustment 
 and it is therefore possible to arrange the receiving 
 aerial to be responsive to the waves from a particu- 
 lar sending aerial or not at will. Thus, if several 
 stations are sending at the same time and using waves 
 of different frequencies any receiving aerial can be 
 adjusted to receive from any one of them and neglect 
 the others as desired. The practical value of this is 
 obvious because otherwise it would not be possible 
 for more than one pair of stations to be working in 
 the same locality at the same time. 
 
n] OSCILLATORY CURRENTS, ETC. 27 
 
 It sometimes happens, however, that two aerials 
 whose frequencies are nearly the same are sending 
 at the same time. Each receiving station will then 
 receive its own signals strongly and the others weakly. 
 The state of affairs is then rather like two people 
 talking at once, one loudly and the other softly. A 
 certain amount of confusion is inevitable. The same 
 trouble may arise in another way. If one station is 
 endeavouring to receive from another one at a dis- 
 tance at the same time that a third station near at 
 hand is sending, then the powerful impulses of the 
 adjacent station may set up appreciable oscillations 
 in spite of the frequencies being widely different, and 
 the signals from the distant station may be quite 
 swamped. Confusion of this kind is what is termed 
 "Interference." There are other forms of interfer- 
 ence especially noticeable when thunderstorms are 
 near, which are caused by the atmosphere discharging 
 itself to earth through the aerial. If these discharges 
 take place when a message is being received they 
 may obliterate whole letters or whole words. These 
 discharges are what are termed "Atmospherics" and 
 are the bugbear of wireless operators, more par- 
 ticularly in the Tropics. 
 
 An aerial which is arranged to respond very 
 freely to its own frequency but only very slightly to 
 impulses differing but little from this natural fre- 
 quency is spoken of as being very "selective." A 
 
28 WIRELESS TELEGRAPHY [CH. 
 
 high degree of "selectivity" is clearly a most valu- 
 able adjunct. A very selective aerial has however 
 the slight disadvantage that its tuning with the 
 sending aerial must be very accurate, otherwise 
 signals intended for it will be passed over. 
 
 The nature of the oscillations in the sending 
 aerials will have a large effect on the selectivity. 
 If the oscillations are strongly damped the impulses 
 in the receiving aerials will be few in number, and, in 
 spite of accurate tuning, will not generally have time 
 to set up large currents. If, however, the arrange- 
 ments are such that these few impulses can set up 
 large currents, then the conditions will also be such 
 that impulses of other frequencies will do the same, 
 and selectivity will be lost. For suppose the damped 
 train of waves to consist practically of one large im- 
 pulse only. An aerial which can receive this signal 
 will be in a condition in which this single blow sets 
 up the necessary oscillations. But a blow under such 
 circumstances will set up practically the same oscilla- 
 tions whether the sending aerial is tuned to the 
 receiving one or not, and if all sending aerials were to 
 send out these damped waves it would be impossible 
 for more than one pair of stations to work at a time 
 in any given locality. With very slightly damped 
 oscillations in the sending aerials, on the other hand, 
 there will be in the receiving aerials long series of 
 impulses which with correct tuning will gradually set 
 
n] OSCILLATORY CURRENTS, ETC. 29 
 
 up the necessary large currents. Impulses of the same 
 magnitude but of different frequency will not cause 
 these currents because they are wrongly timed. Thus 
 with the sustained oscillations a receiving aerial may 
 be adjusted to pick out messages sent at a particular 
 frequency and to neglect all others. It is largely for 
 this reason that as nearly as possible undamped 
 oscillations are desirable in all transmitting aerials. 
 
 CHAPTER III 
 
 ELECTROMAGNETIC WAVES 
 
 WHEN a condenser of the parallel plate type is 
 charged up, the dielectric between the conducting 
 plates is in a condition different from that when 
 the condenser is uncharged, owing to the electrical 
 pressure then existing between the two plates. 
 This electrical pressure gives rise to what is termed 
 an " Electrical Strain " in the dielectric, or more 
 probably in the all pervading aether contained in 
 the space occupied by the dielectric. With the 
 parallel plate type this strain is confined to the 
 part of the dielectric immediately between the two 
 plates, just as when a piece of thin india-rubber is 
 pressed between two flat surfaces the compression 
 is distributed over the part of the rubber between 
 
30 WIRELESS TELEGRAPHY [CH. 
 
 the surfaces only. If this condenser is connected 
 up with an inductance so as to form an oscillatory 
 circuit and if an oscillating current is set up, then 
 the condenser will be rapidly charged and discharged 
 first in one direction and then in the other. Accom- 
 panying these rapid charges and discharges there 
 will be rapid changes in the state of the dielectric 
 between the conducting plates. The electrical strain 
 will be alternately in one direction and then in the 
 other, but it will be confined to that part of the 
 dielectric immediately between the plates and will 
 not tend to spread itself out beyond these limits, 
 except to a very small extent. 
 
 During the time that the oscillatory current is 
 flowing an oscillatory magnetic field will be produced 
 around the inductive part of the circuit. The effects 
 of this magnetic field will be mostly confined to the 
 region of the inductance though it will be noticeable 
 much further away than is the case with the electrical 
 strain of the dielectric in the condenser. 
 
 A circuit such as this, consisting of a parallel plate 
 type of condenser and an inductance, in which the 
 effects of the electrical strain, and to a large extent 
 the magnetic field, are quite local, is termed a 
 " non-radiative " circuit. 
 
 It will be noted that in this oscillatory circuit 
 the greatest magnetic field is produced when the 
 current has its maximum value, which occurs at the 
 
Ill] 
 
 ELECTROMAGNETIC WAVES 
 
 31 
 
 instant at which there is no charge in the condenser 
 (see fig. 6), no pressure between the conducting plates, 
 and therefore when the electrical strain in the di- 
 electric is zero. The maximum electrical strain on the 
 other hand occurs when the charge is a maximum, 
 and the current and with it the magnetic field, zero. 
 During the oscillations there is thus a continual 
 
 Magnetic Flax 
 
 Electrostatic 
 Strain 
 
 Fig. 7. Radiative Circuit 
 
 changing from the maximum of electrical strain to 
 the maximum of magnetic field and back again. The 
 electrical strain is alternately in one direction and 
 then in the other; the same with the magnetic field. 
 Many circuits in which oscillating currents can be 
 set up are not thus " non-radiative." Two insulated 
 plates A and /? (see fig. 7) held horizontally some few 
 
32 WIRELESS TELEGRAPHY [CH. 
 
 feet apart will constitute a condenser and can be 
 charged up. If they are then connected together by 
 means of an inductive coil C they will discharge, and 
 an oscillatory current will be set up in the inductance. 
 In this case when the plates are charged the electrical 
 strain is no longer confined to a definite part of the 
 air, but will be distributed somewhat as indicated in 
 the figure, but of course in all planes round the plates. 
 The effects will be noticeable at some distance from 
 the plates and theoretically would extend indefinitely 
 in all directions. Now when the condenser discharges 
 itself through the wire the current will set up a 
 magnetic field near the coil which again will not be 
 confined only to the immediate neighbourhood of the 
 wire. 
 
 Such a circuit as this in which both the electrical 
 strain and the magnetic field spread out into the 
 surrounding space is termed a " radiative " circuit, 
 or an electrical "radiator." 
 
 Another and more important form of radiative 
 oscillatory circuit is one in which the conducting 
 plates consist of a large number of elevated wires 
 \ all joined together, for one, and the surrounding 
 surface of the earth for the other, the intervening 
 air being the dielectric. The wires and the earth 
 are joined by a conductor which may be straight or 
 may have an inductively wound coil in series with it. 
 This arrangement constitutes the usual type of aerial 
 
in] ELECTROMAGNETIC WAVES 33 
 
 which is an essential part of all wireless telegraphy 
 installations. 
 
 With radiative circuits of this kind in which 
 oscillating currents can be set up, the electric and 
 magnetic effects are transmitted to considerable 
 distances by the electromagnetic waves which the 
 oscillating currents give rise to. 
 
 It will be well to consider a mechanical case of 
 wave transmission to illustrate these electromagnetic 
 waves. A very simple case in which wave motions 
 can be observed is that of a rope tightly stretched 
 between two supports. If a part of the rope say a 
 few feet from one end is gradually depressed, careful 
 observation will shew that there has only been an 
 appreciable movement of those parts of the rope 
 adjacent to the part depressed. A few feet from 
 that part the rope will apparently be exactly as 
 before. But now if the part of the rope instead of 
 being gradually depressed is suddenly pushed clown 
 the same distance by the application of considerable 
 force the results will be quite different. The whole 
 rope will be seen to quiver, a little wave of depression 
 can, with care, be seen to spread in both directions 
 from the part which has been lowered. After a short 
 time the rope will settle down to the same position as 
 when the depression was gradual, but before doing so 
 all parts of it will, on account of the little wave which 
 has run along it, have been moved temporarily through 
 
 F. W. T. 3 
 
34 WIRELESS TELEGRAPHY [OH. 
 
 a much greater distance than the final steady displace- 
 ment, Avhich as previously stated cannot be observed 
 for more than a few feet on either side of the point 
 depressed. If the part of the rope is given a series of 
 say three or four movements up and down in quick 
 succession instead of a single depression only, then a 
 little group of waves can be observed to start from 
 the point of disturbance and move outwards from it. 
 If the end supports are rigid each little group of the 
 waves will be reflected back to the starting-point. 
 As they pass this point the parts of the rope will 
 move up and down just as they were made to do in 
 the first instance but to a smaller extent. After the 
 group of waves has passed, these parts come to rest 
 and will remain so until a group of waves returns 
 again. Eventually the rope will come to rest owing 
 to the oscillations gradually frittering away their 
 energy in friction, stirring up the air, etc. 
 
 The propagation of the electromagnetic waves is 
 somewhat analogous to the propagation of the little 
 train of ripples along the rope. The gradual charg- 
 ing up of a condenser of a radiative circuit produces 
 very local eifects only. If the charging up is rapid a 
 zone of electrical strain is set up near the conductors 
 which immediately spreads itself out in all directions 
 with the definite, but very great, velocity of light, 
 namely 186,000 miles per second. This wave of strain 
 produces far greater effects at each point as it reaches 
 
in] ELECTROMAGNETIC WAVES 35 
 
 it, than the permanent effect produced when the 
 condenser is charged. This is analogous to the way 
 in which the ripple running along the rope causes 
 temporary depressions of each portion of the rope 
 as it reaches it, which are much greater than the 
 permanent ones. (It must be remembered of course 
 that the electric waves spread themselves in all planes 
 and are not confined to one path as in the case of 
 the rope.) 
 
 The distribution of the permanent strain in the 
 air surrounding a charged aerial is somewhat as 
 indicated by the fig. 8 a. With an oscillating current 
 in the aerial the elevated wires will be alternately 
 charged in opposite ways positively and negatively. 
 With each charge a wave of electric strain in one direc- 
 tion or the other is sent out from the wires. This is 
 analogous to the ripples produced in the rope when 
 one part is quickly moved up and down several times. 
 
 The production of these waves is shewn in a 
 diagrammatic way in figs. 8 a to 8 e, where one cycle 
 of the oscillating current has been taken. Between 
 each positive and negative charging of the elevated 
 wires there will be large rushes of current of very 
 short duration. Each rush of current sets up a 
 magnetic field surrounding the vertical wire in the 
 form of a thick horizontal ring. These rings spread 
 out from the aerial in the same way as the zones of 
 electric strain and in between them. 
 
 32 
 
t ; (' 
 
 ~ i i 1 
 
 -^ \\ *v 
 
 <- r 
 
 snH 
 
 *w 
 
 4- 
 
 ' 4r 1 
 
 a t> 
 ui ^ 
 W 
 
 be 
 
 s 
 
CH. in] ELECTROMAGNETIC WAVES 37 
 
 The complete waves will therefore be made up 
 of alternate zones of vertical electric strain and 
 horizontal magnetic field. Each zone expands from 
 the aerial at the rate of 186.000 miles per second 
 and there is consequently no catching up or being 
 left behind. 
 
 A " Wave-length " is the distance from a crest 
 to a crest in the case of the wave in the rope. The 
 same term is used for the electromagnetic waves and 
 it means the distance measured in the direction of 
 movement from one state of maximum positive strain 
 to the next one. The wave-length is consequently the 
 distance a wave can move forward during one com- 
 plete cycle of the oscillating current producing it. 
 Thus with an oscillating current of frequency 100,000 
 cycles per second an interval of nju^nroth of a second 
 will elapse between the throwing oif of consecutive 
 states of positive electric strain. During this 
 interval the first state will have moved forward 
 186,000 x j^Vi^o miles. The wave-length correspond- 
 ing to this frequency is therefore 1*86 miles. Waves 
 of lengths from a few hundred feet to four or five 
 miles are used in practice. 
 
 An aerial as used for wireless purposes constitutes 
 a condenser 1 pfjlfitiiiite magnitude. When it is charged 
 and allowed to discharge, the frequency of the oscilla- 
 tions will therefore depend upon the self induction of 
 the discharge circuit. If this consists of a straight 
 
38 WIRELESS TELEGRAPHY [OH. 
 
 wire connecting the elevated ones to the surface 
 of the earth it will have the smallest possible self 
 induction and the frequency of the discharge will be 
 as high as can be obtained. If inductances are con- 
 nected in series with the wire from the aerial to earth 
 the frequency of the oscillations can be reduced to any 
 required extent. But since the wave-length depends 
 upon the time of one oscillation it follows that by 
 adding inductance in series between the aerial and 
 earth the length of the wave radiated can be increased 
 to any extent. 
 
 Electromagnetic waves of all lengths move with 
 the velocity of light, and as a matter of fact they are 
 almost certainly of precisely the same nature as the 
 wave motion whereby light is conveyed from one 
 point to the other, the only difference being one of 
 wave-length. The luminous waves have frequencies 
 of from four to eight hundred billions of complete 
 cycles per second and wave-lengths measured in 
 millionths of a foot, whereas the frequency of the 
 electromagnetic waves is from 50,000 to a few million 
 only. 
 
 Owing to the very high velocity of propagation of 
 the electromagnetic waves, the time taken for a signal 
 to cover a distance of as much as, say, 4000 miles is 
 very small. The transmission may in fact be looked 
 upon as instantaneous for practical purposes (other 
 than considerations of wave length), just as visual 
 
in] ELECTROMAGNETIC WAVES 39 
 
 signalling by means of a flashing lamp is usually 
 regarded as instantaneous. For instance, the time 
 required for a signal to cross the Atlantic is about 
 g^th of a second. An observer at the transmitting 
 station can observe the actual spark causing a signal 
 but he would not be conscious of its existence any 
 sooner than an operator at the receiving station 2400 
 miles away. 
 
 The electromagnetic waves carry with them a 
 certain amount of energy drawn in the first case 
 from the oscillating current in the aerial. In con- 
 sequence of this a radiative circuit is more damped, 
 i.e. its oscillations die away more quickly, than a 
 non-radiative circuit where this dissipation is not 
 taking place. It is also important to note that the 
 rate of radiating energy is much greater for short 
 waves than for long ones, the energy liberated per 
 cycle varying, in fact, inversely as the cube of the 
 wave-length. With an aerial the rate at which the 
 energy is radiated depends also to a great extent 
 upon whether the aerial is used with added induct- 
 ance in series or not. For a given wave-length an 
 aerial with straight connecting wire gives the 
 maximum rate of radiation, the rate falling off 
 rapidly with a smaller aerial and added inductance. 
 
40 WIRELESS TELEGRAPHY [CH. 
 
 CHAPTER IV 
 
 ABSORPTION AND ATMOSPHERICS 
 
 MAXWELL and Hertz in their mathematical theory 
 of electromagnetic waves assume a perfect non-con- 
 ducting, homogeneous aether ; such an aether as is 
 supposed to occupy interstellar space. The nearest 
 approach to the practical case to which these theories 
 can be applied is to assume the earth to be a flat, 
 perfectly conducting surface and that the surrounding 
 atmosphere does not differ appreciably from the ideal 
 aether. The theory then shews that the intensity of 
 the electromagnetic waves varies inversely as the 
 distance from the transmitting aerial. Thus with 
 oscillations in the transmitting aerial of a given magni- 
 tude the electromotive force in the receiving aerial 
 will fall off in proportion to the distance between 
 them. The current will fall off in the same way 
 and consequently the energy in the receiving aerial 
 (depending, as it does, on both the current and the 
 electromotive force) will fall off as the square of 
 the distance between the aerials. But in practice 
 the conditions assumed are not fulfilled, and to 
 determine the extent to which the inverse square 
 law was applicable experiments were carried out by 
 Duddell and Taylor in 1904 between a land station 
 
iv] ABSORPTION AND ATMOSPHERICS 41 
 
 and a small steamer in the Irish Sea. Up to distances 
 of about 60 miles the law was found to be approxi- 
 mately correct, but even at that distance the falling 
 off of the signals appeared to be more rapid than was 
 given by the theory. This result has been confirmed 
 since by experiments carried out by the United 
 States Navy Department over distances up to 1000 
 miles, which shew that the falling off from the 
 theoretical value gets greater as the distances 
 increase. 
 
 Practical wireless experience has also brought 
 to light numerous other facts not explained by the 
 theory. In the early days when only comparatively 
 short waves were in use, it was discovered that signals 
 could be made with more certainty and over longer 
 distances by night than by day. Later it was found 
 that by day the short waves had a much longer range 
 over water than over land but by night the range 
 was about the same, even though the land was 
 mountainous or consisted of a sandy desert. Long 
 waves on the other hand have very nearly the same 
 range over land as over water whether by day or 
 night. Mr Marconi, who has probably had more 
 experience of these phenomena, at any rate over long 
 ranges, than anyone else, gave some most interesting 
 facts concerning his long distance transatlantic ex- 
 periments in a lecture at the Royal Institution on 
 June 2nd, 1911. Between Clifden and Glace Bay two 
 
42 
 
 WIRELESS TELEGRAPHY 
 
 [CH. 
 
 waves had been used, one of 7000 metres (23,000 ft.) 
 and the other 5000 metres (16,400 ft.). The variation 
 of the strength of the signals on a typical day is 
 shewn in fig. 9. During the period of daylight at 
 both stations the signals are approximately uniform 
 in strength, the long wave being the better one of 
 the two. In the interval between sunset at the two 
 stations signals first of all fall ofl* and then rise again 
 
 Scale of o 
 
 Signal Daytime over , 
 
 Strength Who le AUantic % 
 
 Very 
 Strong 
 
 Strong 
 
 Medium 
 Strong 
 
 Medium 
 Weak 
 
 Zero 
 
 7000 Metres 
 
 Night over 
 Whole Atlantic. 
 
 During which signals are 
 very variable in strength 
 
 varying from vary weak to I 
 cry strong. 
 
 Storms have a decided 
 t and moonlight 
 
 iy have an influence 
 
 a 
 
 Noon 
 
 a 9 to n 11 i a 
 Greenwich Time 
 
 * 5 6 
 
 10 n \i 
 
 Noon 
 
 Fig. 9. Diurnal variation of strength of signals at Clifden 
 
 to a maximum, the short wave usually being the 
 strongest. During the night they are uncertain, 
 sometimes strong and sometimes weak. At the in- 
 tervals between sunrise at the two stations the same 
 phenomena are observed as in the intervals between 
 sunset. Mr Marconi also mentions another remark- 
 able fact, namely, that it is easier to signal in a 
 northerly and southerly direction than in an easterly 
 and westerly one. It had frequently occurred that 
 
iv] ABSORPTION AND ATMOSPHERICS 43 
 
 ships fitted with the usual installations had been able 
 to communicate with the English shore stations from 
 the Mediterranean, distances much greater than could 
 have been accomplished in an easterly and westerly 
 direction with the same power. At Buenos Aires, 
 Mr Marconi had been able to receive signals from 
 Clifden and Glace Bay quite clearly by night, over 
 a distance of 6000 miles. They could not, however, 
 be detected by day and yet during this same time 
 signals between Clifden and Glace Bay themselves 
 were equally strong by day or night. The behaviour 
 in a northerly and southerly direction thus appears 
 to be quite different from the behaviour in the 
 easterly and westerly direction. 
 
 "Freak" distances have also been numerously 
 recorded when stations have been able to com- 
 municate temporarily over much greater distances 
 than usual. It sometimes happens that an operator 
 can sit at his instruments and hear very distant 
 stations, the signals waxing and waning in a manner 
 similar to that in which the sound from a distant 
 church bell is occasionally noticed to vary on a 
 stormy day. 
 
 The reasons for this very irregular behaviour are, 
 as previously stated, at present not thoroughly under- 
 stood. The falling off of the signals more rapidly 
 than is indicated by the theory is due to two possible 
 causes ; either the waves waste their energy en route 
 
44 WIRELESS TELEGRAPHY [CH. 
 
 or they are deflected from the receiving aerial on 
 account of the curvature of the earth or the electrical 
 properties of the atmosphere. 
 
 Energy may be wasted in two ways ; owing either 
 to the air being slightly conductive or to the surface 
 of the earth not being a perfect conductor. In the 
 upper strata of the atmosphere a point is reached 
 at which it possibly becomes a partial conductor on 
 account of its rarefaction and by the action of the 
 sunlight. In these layers the waves would be accom- 
 panied by conduction currents which would be flowing 
 along paths of high resistance and therefore wasting 
 energy. These layers would become therefore to some 
 extent opaque to the electromagnetic waves just as 
 fog is partially opaque to light waves. This semi- 
 conductive state is thought to be produced by the 
 ultra violet light in the sunlight which brings about 
 a change known as " ionisation." This would explain 
 the fact that short waves have a smaller range by 
 day, when the ionisation is going on most rapidly, 
 than they have by night, but it does not explain the 
 fact that the long waves have practically the same 
 range by day as by night or the fact that the range 
 of the short waves is the same over water as over 
 land at night but differs in the daytime. As the 
 bases of the waves spread out from an aerial, currents 
 of electricity are set up in the surface of the earth. 
 If the surface has a high electrical resistance as would 
 
iv] ABSORPTION AND ATMOSPHERICS 4f> 
 
 be the case over a sandy desert, these currents lead 
 to a similar waste of energy which would not take 
 place over the sea as the water is a good conductor. 
 This very probably has some connection with the 
 longer ranges which can be obtained over water, 
 especially with the short waves. 
 
 The alternative theory is that the waves do not 
 spread out over the spherical surface of the earth in 
 the same way that they would if it were flat. The 
 light waves which are of the same nature, but of 
 much shorter wave length, are certainly propagated 
 only in what are for practical purposes straight lines. 
 From analogy it would not therefore be natural to 
 expect the electromagnetic ones to follow curved 
 paths as they undoubtedly do in spreading out from, 
 say, Clifden to Buenos Aires, a distance of one 
 quarter of the earth's circumference. The problem 
 has been attacked by mathematicians with at present 
 uncertain results, but the trend of opinion seems to 
 be that to some extent waves could thus spread 
 themselves round the surface of the earth, not in 
 the form of space waves, as described on page 35, 
 but in the form of waves on or near the surface of 
 the earth. An alternative possibility is that though 
 in a homogeneous atmosphere the waves can pro- 
 ceed in practically straight lines only, they are- 
 reflected and refracted in passing from one layer of 
 the atmosphere to another, thus causing the bending 
 
4fi WIRELESS TELEGRAPHY [OH. 
 
 round the earth which is found to exist in practice. 
 The electrical properties of the air will undoubtedly 
 vary slightly with its physical conditions and the 
 extent of the ionisation. These variations will cause 
 slight variations of the velocity with which the waves 
 are propagated through it, which would cause a 
 change of the direction in which the waves are 
 moving. It is a well-known fact that owing to the 
 bending of the rays of light as they pass through the 
 layers of air the whole of the sun's disc is sometimes 
 visible after it has really sunk below the horizon. 
 There are reasons for supposing that the depth to 
 which the ionisation of the atmosphere proceeds 
 varies with the time of day, being a maximum at any 
 one place at noon and a minimum at midnight. Also 
 it is probable that the extreme outer layers are 
 permanently ionised to a considerable extent, making 
 them more effective in bending the waves than the 
 middle or lower layers. Further, it can be shewn 
 theoretically that the long waves are more bent by 
 the ionised layers than the short ones. Assuming 
 the atmosphere to have these properties, it may be 
 that at night the efficient outer layers refract both 
 long and short waves back to the earth, even into 
 the valleys of a hilly country. But by day it may be 
 that only the long waves are bent back to the earth 
 by the less ionised middle layers ; the short waves 
 being only bent sufficiently to keep them away from 
 
iv] ABSORPTION AND ATMOSPHERICS 47 
 
 the outer layers and not sufficiently to bring them 
 back to the earth again before they have frittered 
 away their energy. A supposition of this kind will 
 explain the different behaviour of the short and long 
 waves by day and by night. It will also more or less 
 fully explain the variation of the strength of the 
 signals between Glace Bay and Clifden. This theory 
 may seem a little far-fetched in some instances, but 
 none of the hypotheses involved are wholly devoid 
 of experimental foundation and it is the only theory 
 at present in existence which can in any way be 
 made to explain all the facts. The suggestion also 
 offers an explanation of the "freak" distances, the 
 idea being that a large mass of air may act as an 
 enormous electromagnetic lens or mirror and focus the 
 waves on to the particular station when very strong 
 signals may be received from stations which under 
 ordinary circumstances would not be heard at all. 
 There seems so far no way of deciding which, if any, 
 of these views is the correct one, and until a great 
 many more experimental data are forthcoming the 
 uncertainty is likely to continue. 
 
 Another cause of uncertainty in the reception of 
 wireless signals, coming into an entirely different 
 category from absorption or refraction, is the so- 
 called atmospheric interference. This trouble is not 
 in any way connected with the inefficiency of the 
 atmosphere as a medium for the transmission of 
 
48 WIRELESS TELEGRAPHY [OH. 
 
 electromagnetic waves but is due to the powerful 
 disturbances liberated from flashes of lightning and 
 other natural electrical discharges. In the case of 
 a large flash of lightning an enormous amount of 
 energy is temporarily available, and short but very 
 violent trains of electromagnetic waves are liberated 
 which, on striking an aerial, are often sufficient to set 
 it oscillating in spite of its being tuned to a wave 
 length very different from that of the waves them- 
 selves. The result is that a loud signal is received 
 sounding like a single " rap " or like a series of two 
 or more "raps" following one another. When very 
 bad these sounds continue almost without interrup- 
 tion and render the reception of proper signals very 
 difficult. Being due to thunderstorms and similar 
 phenomena they naturally vary in intensity with the 
 time of year. In the temperate zones they are at 
 their worst in summer, but are never entirely absent. 
 The waves are so powerful and have such long ranges 
 that it is probable that many of those observed in this 
 country in winter have their origin in the Tropics. 
 In the Tropics the interference is worse than in 
 other zones on account of the prevalence of electrical 
 storms there, especially at certain periods of the 
 year. During the worst storms the aerial has to be 
 connected straight to earth to prevent the instru- 
 ments from being damaged by the heavy discharges. 
 The same thing has to be done in the temperate 
 
iv] ABSORPTION AND ATMOSPHERICS 49 
 
 zones, too, if a storm passes directly over the 
 station. 
 
 A vast amount of ingenuity has been expended in 
 trying to overcome the interference produced by these 
 stray waves. The use of a high pitched musical note 
 is one of the most effective means of reducing it. The 
 atmospherics, fortunately, are quite irregular and 
 signals in a clear piercing note can easily be read 
 through them if they are not too strong. The strength 
 of all but the very worst can be reduced by the use 
 of very selective receiving circuits, unless the wave 
 length of the atmospheric happens to be the same as 
 that of the wave being received. The very powerful 
 ones, however, cannot be got rid of in this way and 
 are particularly harmful in that they often render the 
 operator's ear insensitive for several seconds, with 
 the result that whole letters or even words may be 
 missed. To prevent this occurring the Marconi 
 Company have devised a very ingenious arrangement 
 of two rectifying detectors acting in the same circuit 
 in opposite directions. One of these is in a sensitive 
 condition but the other is not. For ordinary signals 
 the insensitive one has practically no effect and they 
 are received on the sensitive one in the ordinary way. 
 Very strong signals, however, will bring both of them 
 into action in opposite directions, with the result that 
 the sounds though loud are not deafening. Other 
 arrangements having ultimately the same results are 
 
 F. W. T. 4 
 
50 WIRELESS TELEGRAPHY [CH. 
 
 also employed. During the discussion of a paper 
 on this subject before the Institution of Electrical 
 Engineers Mr Duddell suggested the use of more 
 power as the remedy for these troubles. As the 
 available power of some of the recently projected 
 stations is approaching 10,000 horse-power it would 
 appear that his advice was being acted upon. 
 
 CHAPTER V 
 
 THE TRANSMITTING INSTRUMENTS 
 
 THE International Morse Code of "dot" and 
 "dash" signals is almost universally employed for 
 wireless telegraphy just as for telegraphy with wires. 
 It consists of a number of signals of long and short 
 duration, a " long " being equal to two " shorts." By 
 various combinations of these " longs '' and " shorts " 
 or " dots '' and " dashes " all the letters of the 
 alphabet are represented. To an expert operator the 
 Morse code is as easy to " read " either from photo- 
 graphic records or by means of sounds in a telephone 
 receiver as ordinary print or ordinary conversation. 
 On account of each word being spelt out with letters 
 of from one to four separate signals, the transmission 
 is necessarily slower than speaking or writing. With 
 hand signalling the speed of "sending" rarely exceeds 
 
v] THE TRANSMITTING INSTRUMENTS 51 
 
 20 to 25 words per minute (a word being taken as 
 consisting of, on an average, five letters). Some ten 
 times this speed may be obtained by the use of 
 automatic sending by means of apparatus similar to 
 that used for rapid transmitting with the ordinary 
 telegraphy. 
 
 The apparatus used for all systems of wireless 
 telegraphy has certain parts in common. In all cases 
 some form of aerial is essential, which is used to 
 radiate the electric waves when a station is "sending" 
 and to absorb the waves radiating from some other 
 station when it is "receiving." Generally the same 
 aerial is used for both these purposes but in some 
 recent large stations separate aerials are used, one 
 for transmitting the other for receiving. 
 
 When a given station is "sending," instruments 
 are necessary to generate in the aerial the wave- 
 producing oscillating currents and having produced 
 them to control their duration, so as to produce 
 "dots" and "dashes" as required. It is largely in 
 the means of producing these currents that the 
 systems diifer from one another. Considerable 
 differences also occur in the arrangement of the 
 receiving instruments. 
 
 In the earliest wireless telegraphy stations the 
 apparatus was of the simplest, and consisted of a 
 small aerial carried on a mast and connected up to 
 one of a pair of spark balls at its lower end (see 
 
 42 
 
52 
 
 WIRELESS TELEGRAPHY 
 
 [CH. 
 
 fig. 10). The other spark ball was connected to a 
 wire running down into the ground. An induction 
 coil was used to charge up the aerial until a spark 
 jumped across between the two spark balls. When 
 
 Plain 
 
 13 
 
 Aerial 
 
 Spark 
 Balls 
 
 Earth. 
 
 Fig. 10. Diagram of connections for Transmitting by "Plain Aerial" 
 
 this took place the aerial discharged itself and an 
 oscillating current flowed backwards and forwards 
 from the elevated wire to the earth through the 
 bridge of heated vapour between the spark balls 
 which was formed by the first spark. After a few 
 
v] THE TRANSMITTING INSTRUMENTS 53 
 
 hundred-thousandths of a second these oscillations 
 ceased, the air resuming its normal insulating con- 
 dition and the aerial then being ready to receive its 
 next charge from the induction coil. After receiving 
 this charge the process repeated itself, and every 
 time that the aerial discharged a train of electric 
 waves was sent out from it Each train striking a 
 suitably tuned receiving aerial produced a small effect 
 which was made known in some way or other to the 
 operator on the look out for the message. 
 
 If left to itself the induction coil would have gone 
 on charging the transmitting aerial at the fairly rapid 
 rate of some five or ten times per second and at 
 the receiving station there would have been a con- 
 tinuous series of small effects produced. To transmit 
 signals from one station to another it was necessary 
 to interrupt this series so as to divide it up into the 
 necessary " longs " and " shorts " as required for the 
 Morse code. This was done by interrupting the 
 supply of current to the induction coil by means of a 
 hand-operated switch or " key." In this way the dis- 
 charges could be broken up as required, a "long" 
 lasting for something of the order of one quarter of a 
 second during which the aerial discharged itself two 
 or three times and a "short" for one half of this 
 time, the interval between the signals constituting 
 one letter being of the same length as a " short." At 
 the receiving aerial the small effects would be broken 
 
54 
 
 WIRELESS TELEGRAPHY 
 
 [CH. 
 
 up into exactly corresponding " longs " and " shorts " 
 and the message could be " read." 
 
 This apparatus was very simple but had many 
 very serious disadvantages. Its range was limited 
 and it was impossible to render the system selective. 
 This was due to the strong damping, caused largely by 
 the fact that a very large proportion of the energy 
 
 Choking Coil. 
 
 Fig. 11. Diagram of Coupled Transmitting Circuit 
 
 Earth 
 
 stored up in the aerial just before the spark took place 
 was wasted in the spark gap, and partly, also, by the 
 useful radiation of energy into the surrounding space. 
 The oscillation consisted of one large rush of current 
 followed by two or three very rapidly diminishing 
 ones, and then ceased. 
 
 On account of these disadvantages the system was 
 very quickly replaced by the " coupled " system. In 
 
v] THE TRANSMITTING INSTRUMENTS 55 
 
 this system there are two quite separate oscillatory 
 circuits so arranged that they have the same natural 
 frequency and so that they react upon one another 
 inductively. One of these circuits (see fig. 11) consists 
 of the aerial and the earth connected together by 
 an inductance. The other circuit, often called the 
 primary circuit, consists of parallel plate condensers 
 joined up to an inductance and a pair of spark balls. 
 The inductance in this circuit is so placed that it is 
 inductively coupled with the inductance connecting 
 the aerial to earth. This arrangement thus consists 
 of a non-radiative primary circuit in which oscilla- 
 tions can be set up and which acts as a reservoir from 
 which the radiative aerial circuit can take its supply 
 of energy. 
 
 Arrangements are made whereby the parallel plate 
 condenser of the primary circuit can be charged up 
 until a spark jumps across between the spark balls. 
 The condenser then discharges itself through the 
 primary inductance and the spark gap, setting up the 
 required oscillatory current. This oscillatory current 
 in the inductance induces an oscillatory electromotive 
 force in the aerial circuit which in turn gives rise to 
 an oscillatory current there, with the result that 
 electric waves are radiated. With this arrangement 
 a very much longer series of oscillations could be 
 produced in the aerial for one charging-up of the 
 primary condenser. The reaction between the two 
 
50 WIRELESS TELEGRAPHY [OH. 
 
 circuits is rather complicated, but the final result is 
 that the aerial radiates a complex wave made up of 
 two simple waves of different wave-length. With the 
 instruments as used in practice the reaction between 
 the two circuits is not a very powerful one and under 
 these circumstances the two component waves have 
 very nearly equal wave-lengths. 
 
 Though not ideal this system was a great advance 
 on the old system of " plain aerial " as it was called. 
 Ranges were greatly increased and there was an 
 enormous gain of selectivity owing to the much 
 longer duration of the oscillatory currents in the 
 aerial. 
 
 The primary condenser is usually charged by 
 means of an alternating electrical pressure, a spark 
 taking place generally once for every complete cycle. 
 This alternating pressure is usually obtained from an 
 alternator, which is a machine designed to produce 
 alternating currents, driven by a steam or oil engine 
 or by an electric motor. The alternating current is 
 generated at a comparatively low pressure, lower 
 than would be suitable for charging the primary 
 condensers, and is increased to that required by a 
 static transformer. This is an instrument having no 
 moving parts and in which it is easier to secure the 
 very high insulation necessary for the high pressure. 
 An inductively wound coil is also included in the 
 charging circuit so as to give it a natural frequency 
 
v] THE TRANSMITTING INSTRUMENTS f>7 
 
 equal to that of the alternator. A switch placed in 
 the charging circuit is generally used for controlling 
 the time during which the discharges are taking 
 place, producing thereby the required "longs" and 
 "shorts." The number of sparks per second when 
 working depends upon the frequency of the electrical 
 pressure. Frequencies of from 25 up to 800 and 
 1000 cycles per second are employed. High fre- 
 quency sparking is a great advantage where aural 
 reception is employed. The reason for this is that 
 each separate spark liberates a single train of waves 
 which, acting upon the receiving aerial and its de- 
 tector, produces a single movement of the diaphragm 
 of a telephone receiver. If these sparks follow one 
 another regularly and with great rapidity, at the rate 
 of say 1000 per second, the movements of the tele- 
 phone diaphragm will be correspondingly rapid and 
 the sound produced will be a high pitched musical 
 note. These high pitched notes can be heard when 
 a lower pitched note would probably be inaudible or 
 swamped by other noises, particularly the sounds 
 produced in the telephone by atmospheric discharges 
 through the aerial. This is just the same as if two 
 persons were speaking at the same time, one spas- 
 modically in a low bass voice and the other regularly 
 in a high pitched piercing treble. A third person 
 can with a very little concentration follow the latter 
 without being in any way confused by the former. 
 
58 WIRELESS TELEGRAPHY [OH. 
 
 There is also the further advantage that the telephone 
 receiver is more sensitive to high frequency currents 
 than to low ones. These high musical notes are 
 therefore very useful Avhere serious interference is 
 experienced. In most modern land stations and in 
 many of the larger ships' installations, rates of spark- 
 ing of from 200 to 1000 sparks per second are 
 employed. A much greater amount of energy per 
 second is then, too, sent out from the aerial and 
 the range is increased. A corresponding increase 
 of power is of course required at the transmitting 
 station. 
 
 The disadvantages under which this system labours 
 are twofold. Firstly, during practically the whole 
 time that oscillating currents are flowing in the aerial 
 there are also oscillating currents in the primary 
 circuit wasting energy at the spark gap. Relatively, 
 the energy wasted in this spark is much less than 
 that wasted in the spark when using the "plain 
 aerial" system, but nevertheless the result of this 
 wastage is that only a part of the energy supplied is 
 actually radiated in the form of a still somewhat 
 damped wave train and the spark balls become 
 so heated that irregular sparking takes place, so 
 spoiling the clearness of the note in the telephone 
 at the receiving station. This is particularly the case 
 with the rapid sparking previously mentioned, so 
 much so in fact that all kinds of complicated devices 
 
v] THE TRANSMITTING INSTRUMENTS 59 
 
 in the form of powerful blowers, rapidly rotating 
 spark gaps, etc., are being employed, with varying 
 success, to overcome the difficulty. 
 
 Of these devices perhaps the most successful is 
 one now used by the Marconi Co., which consists 
 
 Primary D . 
 
 primary 
 Inductance r 
 
 Condenser 
 
 Fig. 12. Marconi Kotary Spark Gap 
 
 of a wheel, on the periphery of which are a 
 number of projecting studs, fixed on to the shaft of 
 the alternator producing the alternating pressure, as 
 shewn diagrammatically in fig. 12. This wheel with 
 its projecting studs rotates between a pair of fixed 
 
00 WIRELESS TELEGRAPHY [OH. 
 
 studs which are just clear of it. The double gap 
 from one fixed stud to the wheel and thence to the 
 other fixed stud here constitutes the gap in the con- 
 denser discharge circuit. The position of the fixed 
 and moving studs is so arranged that they come 
 opposite to one another just at the instants at which 
 the spark is required, that is at the instants at which 
 the condenser is fully charged. When the studs are 
 not opposite to one another the gap in the discharge 
 circuit is so large that no spark can possibly take 
 place. In this way sparks can be obtained with 
 perfect regularity and a clear musical note produced. 
 With this device there is no trouble with overheating 
 of the spark balls on account of the fanning of the 
 wheel. 
 
 With the large Marconi installations an additional 
 effect is produced with this rotary spark gap. The 
 peripheral speed of the wheel (which has to be very 
 high, in the neighbourhood of 36,000 feet per minute) 
 and the number of studs are so arranged that the studs 
 are opposite to one another for only a very short time. 
 Then by the correct adjustment of the coupling of the 
 two circuits the oscillations in the primary are only 
 allowed to continue for just long enough to set up 
 the maximum oscillations in the aerial, the increasing 
 distance between the studs making it impossible for 
 them to go on longer. The oscillations in the aerial 
 then continue whilst the primary condensers are 
 
v] THE TRANSMITTING INSTRUMENTS 61 
 
 charging up for the next spark. This final result is 
 almost exactly the same as that obtained by the 
 "quenched spark" system to be explained later. 
 
 The second disadvantage of the coupled system is 
 that the waves radiated from the aerial are complex 
 ones made up of two waves of slightly different 
 frequencies, one above and one below the frequency 
 to which each of the circuits are separately tuned. 
 At the receiving station the best results are obtained 
 with the receiving aerial tuned to one or other of 
 these two waves. Signals can, however, be received 
 with the aerial timed to any intermediate frequency. 
 The result of this is that a wider range of frequency is 
 involved for each "tune" and the whole of the energy 
 radiated is not concentrated in the particular wave 
 for which the receiving aerial is tuned. For the 
 conditions under which this system is used in practice 
 the two waves have very nearly the same frequency 
 so as to overcome this difficulty as much as possible. 
 
 Both these difficulties have been largely overcome 
 by the use of one or other of the applications of 
 Prof. Max Wien's "charging by impact" method, 
 notably in the "Telefunken" system and in the nearly 
 identical system of Lepel. 
 
 In these systems there are again two coupled 
 circuits as in the system just described. The funda- 
 mental difference lies in the fact that in Wien's 
 method means are employed to "quench" the spark 
 
62 WIRELESS TELEGRAPHY [CH. 
 
 in the primary circuit so preventing the current from 
 flowing there for more than two or three cycles. If 
 this is done in a suitable way the whole of the energy 
 of the primary circuit is transferred to the aerial 
 circuit almost in the form of an electrical blow and 
 by the time the primary current has ceased large 
 oscillations will have been set up in the aerial. This 
 circuit then goes on oscillating by itself, the oscilla- 
 tions dying down quite gradually, there being no 
 damping owing to currents flowing across the spark 
 gap. The ultimate result is that a larger propor- 
 tion of the energy is actually radiated, and this 
 radiation, be it noted, is in the form of a single wave 
 of the natural frequency of the aerial circuit. 
 
 The overheating of the spark gap, which with the 
 ordinary spark gap leads to irregular sparking, is not 
 so likely to arise with this method because of the 
 very short time during which the current is flowing 
 there. Also for the same reason the charging up of 
 the primary condenser can start whilst the aerial is 
 still radiating. The system is therefore particularly 
 well adapted to high spark frequencies, and the use 
 of a high musical note which can be "read" through 
 atmospheric interference without difficulty. The 
 oscillations in the aerial circuit are well sustained, 
 making the system very selective and susceptible to 
 fine tuning, 
 
 The charging of the primary condenser and the 
 
v] THE TRANSMITTING INSTRUMENTS 63 
 
 interruption of the sparking for signalling purposes 
 are brought about in exactly the same way as before. 
 
 The same principle can be employed with a direct 
 current for charging the condenser with but slight 
 modifications. ,This was as a matter of fact the 
 method originally used by Lepel. 
 
 The quenching of the spark, which is the distinctive 
 feature of the system, is obtained by breaking the 
 spark gap up into a number of small gaps of about 
 one-hundredth of an inch each between metallic discs 
 of about three inches in diameter. The best adjust- 
 ment of the circuits and the practical management of 
 the gaps presented some difficulty at first but this 
 has now been entirely overcome. 
 
 Besides the two coupled circuit methods just 
 described and generally known as the Marconi and 
 the Telefunken system respectively, there are systems 
 using undamped high frequency currents generated 
 by special types of alternating current dynamos, or 
 as in the Poulsen system by means of an electric arc 
 between suitable electrodes in a non-oxidising atmo- 
 sphere. In America Alexanderson has designed 
 alternators capable of giving quite large outputs, of 
 the order of twenty to thirty horse-power, at fre- 
 quencies of 100,000 cycles per second and upwards. 
 Goldschmidt in Germany working on entirely different 
 principles has achieved the same final results. The 
 Telefunken Co. have also recently introduced a method 
 
64 WIRELESS TELEGRAPHY [CH. 
 
 of generating these currents for use in their high 
 power station at Nauen. 
 
 The method of using these alternators is to connect 
 them directly into the aerial circuit, which is tuned 
 to have a natural frequency equal, to that of the 
 alternator. By means of a suitable switch the dura- 
 tion of these high frequency currents is controlled so 
 as to produce the necessary "longs" and "shorts" for 
 telegraphic signalling. 
 
 In the Poulsen arc system an electric arc, similar 
 to that in the ordinary electric arc lamp used for 
 illuminating purposes, is maintained between copper 
 and carbon electrodes surrounded by an atmosphere 
 of methylated spirits or other hydrocarbon vapour. 
 When subjected to a powerful magnetic field and 
 suitably adjusted, this arc has the property of 
 causing continuous oscillatory currents to flow in any 
 oscillatory circuit connected across its terminals. 
 These oscillations are of a frequency equal to the 
 natural frequency of the oscillatory circuit, but appear 
 to be slightly less regular than the oscillations 
 produced by a high frequency alternator. If the 
 oscillatory circuit is a suitably tuned aerial, continuous 
 oscillations will be induced in it just as (or almost 
 just as) when a high frequency alternator is used. 
 
 In these systems absolutely undamped wave trains 
 are sent out from the aerial and the maximum of 
 selectivity is obtained. At a receiving station, working 
 
v] THE TRANSMITTING INSTRUMENTS 65 
 
 with a transmitting station sending out these un- 
 damped waves, special methods have to be adopted 
 to receive the signals, which will be explained later. 
 Intercommunication between one of these stations 
 and an ordinary spark station is not therefore possible 
 under ordinary circumstances although they will 
 interfere with one another freely. 
 
 A further advantage is that with a continuous 
 wave system the aerial is actively radiating during 
 the whole of a " dot " or " dash." To illustrate this 
 point with figures ; take a spark system using an 
 oscillatory frequency of 100,000 cycles per second, 
 each train consisting of about 50 effective cycles, and 
 let the spark frequency be 500 per second. The 
 intervals between the sparks will then be -gfath of 
 a second. Of this ^yth of a second, the aerial will 
 only be actively radiating for 50 x y^oVw = wooth f 
 a second. For the remaining ^ 3 ^ths it will be in- 
 effective. Had a continuous wave system been in 
 use the aerial would have been radiating for the 
 whole 5ooth of a second and for the same range an 
 aerial of only about ^ of the size would have been 
 necessary. As the aerial represents about one half 
 of the total cost of a modern station, this is a strong 
 point in favour of the continuous wave systems. There 
 are unfortunately practical disadvantages. The alter- 
 nators are expensive, require very careful attention, 
 and have yet to prove their durability. With the 
 
 F. W. T. 5 
 
66 WIRELESS TELEGRAPHY [OH. 
 
 Poulsen arc difficulty is experienced in keeping it 
 ready to oscillate when required and for as long as 
 is required ; moreover it is inefficient, only a small 
 percentage of the total energy supplied being 
 radiated. 
 
 CHAPTER VI 
 
 THE RECEIVING INSTRUMENTS 
 
 THE detectors used for receiving signals are very 
 sensitive devices enabling the small currents in a 
 receiving aerial to be observed. They are used indi- 
 rectly, being made to operate other instruments by 
 means of which the presence of the current is made 
 known either as a sound signal or by the deflection 
 of a beam of light. The former method is the one 
 now most generally in use, the sounds being produced 
 by a pair of ordinary telephone receivers. By their 
 means every series of oscillating currents is made to 
 produce a corresponding series of sounds which enable 
 the message to be interpreted. If the detector is 
 arranged to deflect a beam of light, the message can 
 be recorded photographically, which has the advan- 
 tage of reducing the possibility of mistakes but is not 
 nearly so sensitive as aural reception with telephones, 
 and messages are liable to be rendered quite unread- 
 able by strong atmospheric disturbances. 
 
vi] THE RECEIVING INSTRUMENTS 67 
 
 It may be pointed out that telephone receivers 
 alone cannot be used because in the first place the 
 inductance of the windings would allow only a very 
 minute oscillating current to flow through them, 
 secondly the diaphragm could not move appreciably 
 at the frequency of the oscillating currents generally 
 used, and lastly even if it could move at this rate it 
 would give out a note beyond the range of audibility, 
 for few ears can detect sound waves of a frequency 
 above 25,000 cycles per second. 
 
 In the old days the metallic filings coherer was 
 the detector universally used, but at its best it 
 was not very sensitive compared with the modern 
 detectors and suffered from the further disadvantage 
 of being easily put out of action by vibration. It 
 was quickly replaced by the far more sensitive and 
 reliable magnetic, thermal and rectifying detectors 
 now in use. 
 
 Marconi's Magnetic Detector is one of these and 
 consists of a band of soft iron kept moving beneath 
 the poles of a pair of horse-shoe magnets (see fig. 13). 
 Just beneath these magnets the wire is made to pass 
 through two small coils, the outer of which is con- 
 nected to the telephone receivers and the inner one 
 to the aerial so that the small oscillatory currents 
 will pass through it. Every train of oscillatory 
 currents then causes a "click" in the telephone. 
 These "clicks" combine to form a distinctive sound 
 
 52 
 
68 
 
 WIRELESS TELEGRAPHY 
 
 [CH. 
 
 which is interrupted in the same way as the sequence 
 of the trains of oscillations. Thus the "dots" and 
 "dashes" are reproduced in the telephones exactly as 
 they are sent out from the transmitting station. 
 
 The winding of the coil through which the oscilla- 
 tory current flows may be of quite low electrical 
 resistance and consequently the instrument can be 
 
 Farth 
 
 Te/ephones 
 
 Fig. 13. Marconi's Magnetic Detector 
 
 connected directly in the aerial circuit without in- 
 troducing too much damping. This, however, is 
 undesirable when a very selective circuit is required, 
 as a single circuit, even of very low resistance, does 
 not allow interference to be "tuned out." A better 
 method, which is the usual procedure of the Marconi 
 Co. with this detector, is to connect it to the aerial 
 through intermediate resonant circuits, inductively 
 
vi] THE RECEIVING INSTRUMENTS 69 
 
 coupled with one another. These circuits are very 
 selective because they all have to be set oscillating 
 in turn by impulses of the frequency to which they 
 are all adjusted. Waves of other frequencies may 
 set the aerial oscillating to some extent and the 
 intermediate circuits very slightly, but unless the 
 interference is very powerful the final detector circuit 
 will be unaffected. 
 
 This detector is very stable and reliable, and the 
 only attention it wants is regularly to wind up the 
 clockwork which keeps the iron band moving. Much 
 of the success of the Marconi ships' installations must 
 be put down to its excellent qualities. 
 
 Fessenden's thermal detector consists of a very 
 short length of exceedingly fine platinum wire sealed 
 into an exhausted glass bulb, rather like the bulb of 
 a tiny electric lamp. The oscillatory current is passed 
 through this wire, heats it, and consequently causes 
 an increase of its electrical resistance. This change 
 of resistance is made use of indirectly to operate 
 a telephone receiver. 
 
 A fundamental difference between the magnetic 
 detector and this thermal detector may be pointed 
 out. The heating of the wire in the latter depends 
 on both the magnitude of the current and the time 
 for which it is flowing. The same result can be ob- 
 tained by a large current flowing for a short time or 
 by a small current flowing for a long time. With the 
 
70 WIRELESS TELEGRAPHY [OH. 
 
 magnetic detector, on the other hand, all the indica- 
 tions point to the fact that it is upon the maximum 
 value of the current that its action depends. A long 
 continued small current will not affect it to the same 
 extent as a large current flowing for a short time. 
 The thermal detector belongs to a class known as 
 integrating detectors, i.e. those which will add together 
 a series of small effects. With the use of more and 
 more sustained trains of oscillations, to secure greater 
 selectivity, detectors of the type of the magnetic 
 detector tend to fall into disuse, detectors of the 
 integrating type taking their place provided they 
 can be made equally stable and reliable. 
 
 The modern rectifying detectors of course belong 
 to the integrating class. They depend for their action 
 upon their property of being in some way or another 
 able to store up a little electricity from each train of 
 the oscillatory currents in the aerial. This charge 
 they give out again as a pulsating unidirectional 
 current to telephone receivers, one pulsation corre- 
 sponding to every train of oscillations. One method 
 of connecting them up to the receiving aerial is shewn 
 in fig. 14. The aerial is connected to earth through 
 the inductance LI. A second inductance L% and a 
 condenser K* form a second oscillatory circuit tuned 
 to the same natural frequency as the aerial and so 
 placed that the two circuits are inductively coupled 
 together. The detector D is connected across the 
 
vi] THE RECEIVING INSTRUMENTS 71 
 
 terminals of the condenser KI in series with another 
 condenser K^. The telephone receivers T through 
 which the pulsating currents are to be passed are 
 connected across the terminals of this condenser K 2 , 
 and for the best results with most detectors a battery 
 B is connected in series with them. 
 
 The action of this apparatus is as follows. When 
 
 A 
 
 L, 
 
 <z> 
 
 I 
 
 Fig. 14. Diagram of Receiving Instruments 
 
 a train of waves of the right frequency strikes the 
 aerial, oscillatory currents are set up in L^ which 
 induce oscillatory electromotive forces in L 2 . As 
 the two circuits are tuned to the same frequency, 
 these oscillatory electromotive forces will set up large 
 currents and there will be large oscillatory differences 
 of pressure between the terminals of the condenser 
 
72 WIRELESS TELEGRAPHY [OH. 
 
 K^ These of course tend to make currents flow 
 through the detector into or out of the condenser 
 K 2 . The detector is so adjusted, however, that these 
 currents can flow one way but not the other, with the 
 result that the condenser K z becomes charged. For 
 suppose that the current can flow downwards through 
 D and not upwards. Then when the upper terminal 
 of K l is positive a charge passes through the detector 
 and collects on the upper part of K 2 . When the 
 pressure is reversed the tendency is for this charge 
 to run out again. This the detector prevents. Thus 
 every time that the upper terminal of KI is positive 
 the charge in K z will be added to, and by the 
 end of the train quite an appreciable charge will 
 have been deposited in K z . As it is collecting, 
 this charge will have started flowing comparatively 
 slowly through the telephone receivers. It would 
 appear at first sight that it would run out in this way 
 as fast as it collected. It does not do this on account 
 of the inductance of the receiver windings. For every 
 train of oscillations, therefore, the condenser K 2 has 
 a charge given to it in a series of rapid jerks which 
 it gives out again to the receivers as a single 
 pulsation of current in one direction. Each current 
 impulse causes a "click" in the receivers and if 
 continued rhythmically gives rise to a clear musical 
 note. The function of the battery in series with the 
 telephones is to maintain a steady current through 
 
vi] THE RECEIVING INSTRUMENTS 73 
 
 the detector and the telephone by way of the induct- 
 ance L 2 . The detectors requiring this battery are 
 found to be better rectifiers when they already have 
 a certain steady current flowing through them as well 
 as the oscillations. The adjustment of this current 
 is generally rather critical. The oscillatory currents 
 and the impulsive telephone currents are super- 
 imposed on this steady current, and the ultimate 
 result, viz. a regular variation of the current in the 
 receivers, is the same. Owing to the aerial and 
 intermediate circuits being tuned together this ar- 
 rangement of receiving instruments is fairly selective. 
 Only powerful interference of a different frequency 
 from that of the signal being received will set up 
 sufficiently large currents to prevent the message 
 being read. 
 
 Many of these rectifying detectors consist of 
 crystals or crystalline materials which have this 
 property of unidirectional conductivity when placed 
 in contact with one another. In some cases, car- 
 borundum for instance, the property appears to lie 
 in the crystal itself ; in others it seems to be at the 
 point of contact. This latter is the case with such 
 combinations as galena and graphite or a copper 
 point on iron pyrites. Some of them are more sensi- 
 tive with the battery in series with the telephones 
 but others work better without it. There seems to be 
 no guiding principle in this and it is a matter of trial 
 
74 WIRELESS TELEGRAPHY [OH. 
 
 to find whether a battery is an advantage or not. 
 The disadvantage of these detectors is that most of 
 them require very careful adjustment and are easily 
 rendered insensitive by vibration or heavy atmo- 
 spheric discharges. 
 
 Another detector coming into this class is the 
 Fleming oscillation valve, which depends for its 
 action on the fact that if a piece of platinum or 
 other metal is sealed into the globe of an ordinary 
 incandescent electric lamp, then, even although this 
 metal makes no electrical contact whatever with the 
 glowing filament, a current can be made to flow from 
 it to the filament with a comparatively small voltage 
 but not in the opposite direction. The Marconi Co. 
 use this detector with both carbon and tungsten 
 filament lamps, and in America a detector called 
 the Audion is used by de Forest which, if not iden- 
 tically similar, is at any rate on the same lines. Both 
 types are fairly sensitive when correctly adjusted and 
 are not influenced by vibration or heavy discharges. 
 They both require a battery in the telephone circuit, 
 the pressure from which must be carefully adjusted. 
 The current flowing through the filament must also 
 be regulated, there being usually two or three sensi- 
 tive points. 
 
 The electrolytic detector may probably also be 
 included in the class of rectifying detectors. It has 
 been made in various forms, usually consisting of a 
 
vi] THE RECEIVING INSTRUMENTS 75 
 
 fine platinum point immersed in a weak acid solution 
 in a lead container. If a battery of suitable electro- 
 motive force is connected up with one terminal to the 
 platinum point and the other to the container, a 
 transient current flows when the circuit is first closed, 
 which quickly dies away owing to the platinum point 
 in some way insulating itself from the solution. If 
 now an oscillatory pressure is applied between the 
 point and the container as well as the steady pressure 
 of the battery, a rectifying action takes place the 
 exact nature of which is uncertain and has led to 
 long discussion in scientific circles. This detector is 
 used with the same arrangement of receiving instru- 
 ments as the other rectifying detectors, and whatever 
 its action may be it is remarkably sensitive when the 
 battery pressure is adjusted to the critical value, 
 being able to detect currents which would be quite 
 unnoticed on the magnetic detector. Like the crystal- 
 line detectors it is liable to be rendered insensitive 
 by heavy atmospheric discharges. 
 
 The receiving arrangements for the undamped 
 waves sent out by transmitting stations using high fre- 
 quency alternators or oscillating arcs are necessarily 
 different from those just described. If for instance 
 an ordinary rectifying detector were used, the result 
 would simply be an increase of the current through 
 the telephone during each signal and a "click" at the 
 beginning and end of it only. It would obviously be 
 
76 WIRELESS TELEGRAPHY [OH. 
 
 very difficult to pick out the "dots" and " dashes'" 
 under these circumstances. One of the best methods 
 of receiving these undamped waves is that due to 
 Poulsen. The aerial is connected to earth through 
 an inductance, with which a second very efficient 
 oscillatory circuit carefully tuned to the same natural 
 frequency is inductively coupled. By means of a 
 little instrument known as a "ticker " the condenser 
 of this circuit is disconnected from the circuit and 
 connected to a pair of telephone receivers at regular 
 intervals. Whilst it is connected to the oscillatory 
 circuit larger and larger oscillatory currents accumu- 
 late in it, gradually reaching quite high values on 
 account of the very slight damping of the circuit. At 
 whatever instant the condenser happens to be dis- 
 connected from this circuit it will be charged to a 
 greater or less extent. It might happen on very rare 
 occasions to be disconnected when it was not charged 
 at all, but the tendency is for the break to take place 
 at the instant of zero current, i.e. at the instant when 
 the charge in the condenser is a maximum. When 
 the condenser is connected to the telephone receivers 
 this charge runs out and causes a "click." These 
 "clicks" will take place every time the condenser 
 connections are changed over if the waves are still 
 coining in. Thus if the waves are interrupted to 
 indicate "dots" and "dashes" the sounds given by 
 the telephone will be similarly broken up. It will be 
 
vi] THE RECEIVING INSTRUMENTS 77 
 
 noticed that the note heard in the telephone is in this 
 case dependent not on the transmitting station but 
 on the speed at which the ticker is working. This 
 speed is generally so arranged that the ticker changes 
 the connections of the condenser at the rate of about 
 one thousand times per second. Reception by this 
 method is very sensitive and selective but is un- 
 fortunately not as immune from atmospheric inter- 
 ference as would be expected. 
 
 CHAPTER VII 
 
 THE USES OF WIRELESS TELEGRAPHY 
 ON BOARD SHIP 
 
 WIRELESS telegraphy serves two distinct purposes 
 on board ship, firstly for safety and use in emergency, 
 and secondly for general communication between ship 
 and ship and between ship and shore. Taking the 
 former of these, it is estimated that not less than 
 5000 persons have been rescued from serious pre- 
 dicaments through its agency, the majority of whom 
 would in all probability have otherwise lost their 
 lives. Numerous cases could be quoted. The col- 
 lision between the s.s. " Florida " and s.s. " Republic," 
 when the operator of the latter ship achieved world- 
 wide notoriety for the manner in which he stuck to 
 
78 WIRELESS TELEGRAPHY [CH. 
 
 his work under the most adverse circumstances, is a 
 case in point. More recently the P. and 0. liner 
 "Delhi" went ashore on a practically desert coast 
 of Northern Africa. By the aid of wireless, assistance, 
 which could not otherwise have been obtained, was 
 forthcoming in a few hours. Or again, what the 
 sufferings of the survivors of the "Titanic" disaster 
 would have been but for their timely rescue by 
 the "Carpathia," it is impossible to say. The in- 
 habitants of the distant island of St Kilda will also 
 appreciate the value of wireless, as it was by its aid 
 that a cruiser was despatched to them with food 
 supplies in the spring of 1912, within a few hours 
 of their famished condition being known. 
 
 The utility of wireless in this connection is not 
 only after a disaster has taken place. It now serves 
 as an invaluable means of circulating information 
 with respect to dangers to navigation in the form 
 of ice, derelicts and bad weather. The tracking of 
 typhoons in the Indian Ocean and China Sea has 
 now been reduced to a fine art, and all ships in these 
 seas receive regular warnings arid are able to adjust 
 their courses so as to evade the danger. Further, 
 experiments are now being carried out with several 
 types of so-called "Wireless Compasses," which en- 
 able a ship to find the direction from which a signal 
 is being received. Thus on a dangerous coast, 
 if a lighthouse or lightship is fitted with wireless 
 
vn] USES ON BOARD SHIP 79 
 
 apparatus, it becomes "visible" to passing ships fitted 
 with these compasses even in the densest fogs or the 
 worst snowstorms. The French Government are 
 fitting up several stations for this purpose only, 
 designed to give out regular intermittent signals so 
 that they may be identified in the well-known way 
 in which lighthouses are recognised at night, by the 
 regular interruption of their beams of light. 
 
 To a smaller extent, the sending of "time signals" 
 to ships far out from land is of value, because unless 
 a navigator knows the Greenwich Mean Time accu- 
 rately it is difficult for him to fix his position on 
 the ocean with any degree of certainty. The " time 
 signals" are therefore intended to enable him to 
 check the accuracy of his chronometers. The signals 
 are sent out at prearranged times by the powerful 
 shore stations, e.g. the German station at Norddeich 
 gives the signal at 12 midday, and the Eiffel Tower at 
 10 a.m. and 12 midnight; and by prearranged signals, 
 one definite part of which is the hour to be indicated. 
 The method used by Norddeich is to give some three 
 or four minutes' warning so that operators may be 
 ready, and then to give four series of five " dots," the 
 last of which is 12 o'clock. This method allows very 
 accurate checks to be taken, owing to the rhythmical 
 manner in which the series of " dots " are sent out. 
 
 Wireless is thus a great addition to the safeguards 
 of navigation as well as being invaluable in case of 
 
80 WIRELESS TELEGRAPHY [CH. 
 
 accident, and it is for both these reasons that many 
 countries, e.g. Great Britain, The United States of 
 America, Spain, Italy, etc., are making or have 
 already made laws to the effect that ships beyond 
 a certain size and going beyond a certain distance 
 shall be equipped with reasonably good apparatus for 
 the purpose. At the 1912 International Conference 
 it was unanimously decided to recommend all Govern- 
 ments to make it compulsory for all ships, cargo or 
 passenger, to be fitted with wireless instruments and 
 to have on board at least one fully qualified operator. 
 The fact that in Lloyds' Register of Shipping a special 
 section has for some years been devoted to those ships 
 fitted with wireless, shews the importance attached 
 to it by ship-owners and underwriters. 
 
 The arrangement for distress signals to take pre- 
 cedence over all others under all circumstances has 
 been in existence from the first. It was one of the 
 earliest provisions of the Marconi organisation, the 
 distress signal then being "CQD." At the 1906 In- 
 ternational Conference this was changed to "SOS," 
 but the Marconi operators still used the "CQD" 
 signal among themselves, and it is interesting to 
 notice that the first distress signal sent out from 
 the "Titanic" was "CQD." After distress signals urgent 
 Government messages, English or otherwise, take 
 second place in the order of precedence, being fol- 
 lowed, before any private or press messages whatever, 
 
vn] USES ON BOARD SHIP 81 
 
 by messages relating to navigation such as warnings 
 and weather reports. 
 
 For general communication on board ship, wire- 
 less is used for much the same purposes as the 
 ordinary telegraph on land. On all the frequented 
 routes, a ship fitted with wireless apparatus is never 
 isolated from the world. If it is not in direct com- 
 munication with a shore station itself, it will be within 
 range of some other ship which can take any desired 
 messages and "relay" them on to a shore station. In 
 this way it is now possible to start from, say London, 
 and travel completely round the world without ever 
 being out of telegraphic communication with the 
 starting point. Across the Pacific from Japan to 
 Vancouver and between New Zealand and San 
 Francisco are perhaps the few regular routes on 
 which uninterrupted communication cannot be de- 
 pended upon during both day and night. Even these 
 possible chances of being isolated from the world 
 will shortly be removed. On the North Atlantic 
 routes, every liner has the necessary apparatus and 
 can be communicated with from either Poldhu or 
 Cape Cod for the whole passage. This has made it 
 possible to publish small daily newspapers, "The 
 Atlantic Daily News" and the "Cunard Bulletin," 
 on board these ships, giving all the most important 
 news of the day, market fluctuations and reports, 
 stock exchange quotations, and so forth. The busy 
 
 F. W. T. 6 
 
82 WIRELESS TELEGRAPHY [OH. 
 
 business man or financier is thus no longer out of 
 touch with the world for five or six days every time 
 he crosses the Atlantic. Before the introduction of 
 wireless, these crossings were looked upon as mental 
 rests with the mind free from all worry and trouble. 
 Those days are now past. 
 
 At most post and telegraph offices in Europe and 
 America, messages are now received for transmission 
 to ships within range of the ordinary shore stations 
 at rates very little exceeding those of an inland tele- 
 gram. In Great Britain and America messages are 
 accepted for any ship on the Transatlantic service, 
 to be sent at extra cost via Poldhu or Cape Cod if 
 the ship is out of range of the ordinary shore stations. 
 On other frequented routes the arrangements made 
 are on similar lines, but necessarily not so complete. 
 In time, however, it is probable that there will be the 
 same continuous communication for a ship in any part 
 of the world, whether it is on a regular route or not. 
 
 The value of this continuous communication is 
 not to the passengers only. To the shipping com- 
 panies alone, wireless more than justifies its existence. 
 The exact time of arrival of a ship is now known 
 beforehand, together with a full knowledge of the 
 docking accommodation needed, the stores required 
 for the next voyage, and any information with respect 
 to repairs which may be necessary. By the time the 
 ship arrives everything is in readiness, and the time 
 
vn] USES ON BOARD SHIP 83 
 
 saved is generally nearly a day, and often more, per 
 voyage. The record Christmas trip of the "Lusitania" 
 in 1911 from Liverpool to New York and back, was 
 only possible under these conditions. 
 
 To the submarine cable companies, wireless has 
 been in one way a veritable godsend. All their 
 repair ships are now fitted, and instead of having 
 to come to a port after every repair, they now get 
 their instructions by wireless and can proceed direct 
 to wherever they are next required, often saving 
 voyages of several hundred miles. 
 
 In the fishing fleets in the North Sea and in 
 American waters, the larger boats are now being 
 fitted. The value of communication between ship 
 and shore, whereby the supply of fish reaching the 
 markets is known beforehand, can easily be under- 
 stood. 
 
 Wireless has not infrequently been the means 
 whereby escaping criminals have had their flight 
 summarily cut short. The sensational discovery and 
 arrest of Dr Crippen and Miss le Neve on board the 
 s.s. " Montrose " is a case in point which will be well 
 remembered. 
 
 At the present time, with the exception of some 
 30 or 40 vessels, all the wireless ships are under the 
 control of one or other of the Marconi companies, 
 and working under the Marconi organisation with 
 the shore stations of the different countries. These 
 
 62 
 
84 WIRELESS TELEGRAPHY [CH. 
 
 companies train and to a large extent control their 
 own operators, and have introduced a regular service 
 similar to that of the Cable Companies' services. The 
 operators have definite qualifications imposed upon 
 them by the International Convention and by the 
 Governments of the countries to which the ships 
 belong, but beyond this their control is in the hands 
 of the Marconi companies subject to the commanding 
 officer of each ship whilst actually at sea. The 
 operators sign on at the beginning of each voyage, 
 and are given the relative rank of a junior officer. 
 At the beginning of each voyage, sailing charts are 
 prepared from which the operator can tell every day 
 and every hour what ships should be within range. 
 The list of these is generally posted on the ship's 
 notice boards at intervals for the information of the 
 passengers. Every ship as soon as it is within range 
 of another, promptly calls it up and sends off to it 
 any telegrams which may have been accumulating 
 for it during the previous few hours. 
 
 In the larger and more important liners two 
 operators are carried, and a continuous watch is 
 kept day and night. During a voyage across the 
 Atlantic these operators have their hands fairly full, 
 sending and receiving telegrams to and from shore 
 stations and other ships. Some three to four hours 
 of the night are devoted solely to the reception of 
 press news from Poldhu or Cape Cod for insertion 
 
vii] USES ON BOARD SHIP 85 
 
 in the following morning's newspaper. Ship to ship 
 communication ceases during this interval, except 
 for distress signals or very urgent messages. 
 
 In the smaller ships only one operator is carried, 
 with the result that the watch is not continuous. 
 This is a disadvantage as distress signals may be 
 missed. The 1912 Conference has recommended that 
 in ships of this class, the operator or other competent 
 person capable of reading Morse should listen for 
 the first ten minutes of every hour, and that distress 
 signals should always be made at those times as well 
 as at others. By this means a ship near another one 
 in distress would be certain of picking up the distress 
 signals within an hour. 
 
 Although such a large proportion of the ships 
 have their wireless under the control of the Marconi 
 companies, they are by no means all of them fitted 
 with Marconi apparatus. All the systems used can, 
 however, intercommunicate freely so that no diffi- 
 culties are experienced. The arrangement of the 
 apparatus in its main features is very much the 
 same in all cases. It of course depends upon the size 
 of the ship, as the requirements of an Atlantic liner 
 are very different from those of a trawler. In the 
 one case skilled operators are available, and in order 
 to get long ranges and rapid signalling more com- 
 plicated apparatus can be employed than in the other, 
 where the operator has other duties and only just 
 
86 WIRELESS TELEGRAPHY [CH. vn 
 
 sufficient wireless training to be able to send and 
 receive messages in Morse. 
 
 The power of the machines supplied for trans- 
 mitting purposes varies from about one-half horse- 
 power up to 10, giving effective ranges of from 50 
 up to 400 or 500 miles when working with the medium 
 power shore stations. The aerials consist usually of 
 from two to six horizontal wires arranged lengthways 
 of the ship and suspended from the highest possible 
 points of the masts. One or two connecting wires 
 are brought down from these elevated ones to the 
 wireless cabin, which is generally situated on one of 
 the upper decks not far from the bridge, with which 
 easy communication is provided. These wireless cabins 
 on the larger ships are divided into three parts, namely, 
 sleeping quarters for the operators, a space for the 
 transmitting instruments, and the operating room 
 with the receiving instruments. Figs. 15 and 16 
 shew the transmitting compartment and the operating 
 room of a ship fitted by the Marconi Company. The 
 lower part of the transmitting compartment contains 
 a "rotary converter," which is a machine for con- 
 verting the usual direct current from the ship's 
 electrical supply, to an alternating one of fairly high 
 frequency. This alternating current is taken to a 
 transformer placed on the shelf above, which raises 
 the voltage to that required for charging the con- 
 densers placed on the same shelf alongside it. These 
 
Fig. 15. Arrangement of instruments in the transmitting 
 compartment of a Marconi ship 
 
88 WIRELESS TELEGRAPHY [OH. vn 
 
 condensers are made up of glass sheets with layers of 
 metallic foil between them, the whole being immersed 
 in insulating oil. The discharge circuit of the con- 
 densers is through the primary inductance, contained 
 in the box fixed on the right-hand side high up on 
 the back of the compartment, and the spark gap 
 placed on the shelf above the rotary converter and 
 marked "2," fig. 15. In the latest type of trans- 
 mitting instruments, a rotary spark gap on the 
 shaft of the rotary converter is used. The secondary 
 inductance is split up into two parts, one of which is 
 in a box in front of the primary inductance with 
 which it is inductively coupled, and the other is in 
 a second box attached to the back of the compart- 
 ment on the left. The wires from the aerial are 
 connected to the latter, one end of the former being 
 connected to the earth connection which is attached 
 to the steel frame of the ship itself. These two 
 circuits thus constitute the two circuits of the 
 ordinary coupled system. Both primary and second- 
 ary inductance can be varied to enable several tunes 
 to be used. The usual tunes worked on are the 
 commercial 300 and GOO metre waves, but the pro- 
 vision is there for others if they are required. In 
 the operating room will be observed, on the right, 
 a switchboard and apparatus for the control of the 
 rotary converter in the transmitting compartment. 
 On the kneehole table, to the right, is an ordinary 
 
90 WIRELESS TELEGRAPHY [CH. vn 
 
 induction coil which can be used for transmitting 
 distress signals in emergency if the supply of elec- 
 tricity from the ship's dynamo fails. To the left of 
 this coil is the " multiple tuner," a device for rendering 
 the receiving circuits very selective. Above it is the 
 magnetic detector and to the left of it Fleming valve 
 detectors, either of which may be used at will. The 
 use of these two types of detector is a luxury provided 
 on the larger ships only, the smaller ones having the 
 magnetic detector alone. A point of interest in the 
 practical use of the multiple tuner is a small "change- 
 over " switch by means of which it may be put into 
 or cut out of the receiving circuits. It is cut out 
 when the operator is keeping a "general lookout." 
 His receiving circuit is not then very selective, and 
 any signals over quite a wide range of wave-length 
 may be picked up. This is an advantage because all 
 ships nominally on the same tune may not have ex- 
 actly the same adjustments, so that a very selective 
 receiving circuit adjusted to one ship might miss 
 signals from another. With the non-selective re- 
 ceiving circuit this is almost impossible. When the 
 operator hears his ship called, he then puts in his 
 tuner and adjusts it so as to get the best results for 
 the particular message. 
 
 Fig. 17 shews one type of transmitting appa- 
 ratus fitted by the Telefunken Company, except for 
 the motor-alternator which is placed in a separate 
 
Fig. 17. Telefunken Transmitting Instruments 
 
92 WIRELESS TELEGRAPHY [CH. vn 
 
 compartment. This latter consists of an alternator giv- 
 ing alternating currents of a frequency of 500 cycles per 
 second driven by an electric motor. The current from 
 this alternator is taken to a transformer shewn under 
 the table on the left-hand side. Here the voltage is 
 increased to that suitable for charging the condensers, 
 which can be seen in the form of four vertical tubes 
 at the back of the table. These condensers consist 
 essentially of glass tubes coated inside and out with 
 thin metallic foil. The inside layers are all joined 
 together and form one plate of the condenser, the 
 outside layers the other, the intervening glass serving 
 as the dielectric. The discharge circuit of these con- 
 densers consists of the Telefunken multiple spark gap. 
 the cooling vanes of which can be seen over the back 
 of the table on the right, and the primary inductance 
 under the table, not visible in the figure. This in- 
 ductance is adjusted by pushing a plug into suitable 
 holes on the table. The aerial is connected up to the 
 inductances on the left of the table on the white 
 porcelain insulators, thence to the primary inductance 
 and so to the earth connection, a direct or conduc- 
 tively coupled circuit thus being used. Two oscillatory 
 circuits thus " conductively " coupled have exactly 
 the same properties as the two " inductively " coupled 
 circuits of the Marconi apparatus. Adjustments of 
 the inductances in the aerial and primary circuits 
 are provided for tuning purposes. 
 
Fig. 18. Telefunken Receiving Instruments 
 
94 WIRELESS TELEGRAPHY [OH. 
 
 Fig. 18 shews the receiving apparatus. By means 
 of the switch at the back of the instrument, the aerial 
 is changed over from transmitting to receiving. At 
 the top of the instrument are two inductively coupled 
 coils. One of these is connected to the aerial and 
 to earth, the other to moving vane air condensers 
 seen in front of the apparatus. These two circuits 
 are adjusted for fine tuning by turning the handle 
 of the air condenser. Coarse tuning is obtained 
 by replacing one or other of the coils by the spare 
 ones shewn at the right-hand side. The detectors 
 are arranged to slip into holes which can be seen 
 just behind the condensers, the electrical contact 
 being automatically made by the two projecting 
 clips. The detectors are thus easily changed if they 
 become insensitive. The telephones have moveable 
 plug terminals which fit into sockets at the side 
 of the detectors. The adjustment of this receiving 
 circuit is very simple, and yet at the same time it 
 is very fairly selective. Electrolytic and crystalline 
 detectors of a very permanent and sensitive type 
 are used. In some ships fitted with this apparatus 
 tuned microphones are used, tuned to the spark 
 frequency of the transmitting station, to amplify the 
 sound produced in the telephones. This, however, 
 is at the expense of considerable complication of the 
 adjustments which are easily thrown out by vibration. 
 Also, as would hardly be expected, atmospherics and 
 
vn] USES ON" BOARD SHIP 95 
 
 other interference appear to be magnified to the same 
 extent as the signals which are being received. 
 
 CHAPTER VIII 
 
 THE SHORE STATIONS 
 
 BY this term is generally meant those stations 
 situated on or near the coasts for the particular 
 purpose of communicating with passing ships, and 
 which are open for intercommunication with all ships 
 on one or other of the commercial wave-lengths. 
 
 In Great Britain there is a complete set of these 
 stations round the coast which are under the direct 
 control of the General Post Office. In Germany, 
 Italy, France, and most other countries, a similar 
 control is exercised, but in America there are shore 
 stations belonging to private firms as well as to the 
 Government, both of which are open for commercial 
 use. A few countries, e.g. Germany, Italy and 
 Holland, have individual stations of much longer 
 range than the majority of the English stations. 
 The geographical fact of the British Isles lying right 
 on the more important of the great ocean highways 
 renders a large number of stations necessary, but 
 owing to the existence of the powerful Marconi 
 stations it has not so far been necessary for the 
 
96 WIRELESS TELEGRAPHY [OH. 
 
 Post Office to establish long range stations of their 
 own. For other countries, however, if they wish to 
 communicate with their ships directly as soon as 
 they are within range of land they will generally 
 have to do so over long ranges. The majority of 
 the shore stations are connected up by land lines 
 to the telegraphic systems of their own countries. 
 A few of them, situated on isolated islands are really 
 cable stations on which the wireless installation acts 
 as a feeder to the cable. A good example of a 
 station of this kind is the one in the Cocos Islands 
 in the Pacific. The Eastern Extension Australia and 
 China Telegraph Companies' cables pass this island, 
 and the wireless instruments belong to the company 
 and are worked by their staff. 
 
 The duties of these stations lie almost entirely 
 with the transmission of messages from ship to shore 
 and vice versa, which by the 1906 Convention it is 
 their duty to regulate when the traffic is heavy. As 
 well as this primary function they also act as efficient 
 coast-guards, in that the distress signals of any ship 
 going ashore would be certainly picked up by them 
 and assistance sent. 
 
 In Great Britain the shore stations are Crook- 
 haven in the extreme S.W. corner of Ireland, the 
 first station with, which the Atlantic liners are in 
 touch and consequently a very busy one requiring 
 a staff of six operators working day and night in 
 
vin] THE SHORE STATIONS 97 
 
 pairs ; Rosslare, a station used by shipping passing 
 to Liverpool ; Seaforth, near Liverpool, for the use 
 of steamers entering that port ; The Lizard, another 
 busy station, being the first to come into touch with 
 the ships coming up the English Channel ; Bolt Head, 
 the first station taken over by the Post Office, and 
 Niton in the Isle of Wight near St Catherine's Point, 
 both of which are used in connection with the up- 
 channel traffic; the North Foreland for the use of 
 ships entering or leaving the Port of London and 
 passing through the Straits of Dover ; Caistor in 
 Norfolk for use over the southern part of the North 
 Sea ; Cullercoats, near Newcastle, for service with 
 ships on the N.E. coast and on the North Sea; and 
 lastly Malin Head in the north of Ireland for service 
 with ships from the Clyde and Liverpool passing to 
 the northward of Ireland. 
 
 As well as these Post Office stations there are a 
 certain number of more or less private ones used on 
 particular services only. For instance, the London, 
 Brighton and South Coast Railway have such stations 
 at Newhaven and Dieppe for use with their cross- 
 channel steamers between those ports, and the Great 
 Western Railway have a station near Fishguard for 
 use with their Irish boats. 
 
 On the Continent, Germany has a complete set of 
 stations along both the Baltic and North Sea coasts. 
 At Norddeich on the extreme western point of 
 
 F. W. T. 7 
 
98 WIRELESS TELEGRAPHY [OH. 
 
 Friesland there is a powerful station capable of com- 
 municating with ships well out into the Atlantic. In 
 Holland is another powerful station used for the 
 same purpose, at Scheveningen, near The Hague. 
 France maintains stations along her northern coast 
 as well as in the Mediterranean. In Spain a very 
 complete network of powerful stations has recently 
 been opened. There is a centrally situated station 
 at Aruanjuez near Madrid with which the shore 
 stations proper at Vigo, Cadiz, Tenerifie, Las Palmas, 
 Soller (in the Balearic Islands) and Barcelona, main- 
 tain regular communication. Portugal is following 
 suit in the near future with stations connecting up 
 Lisbon with the Azores, Madeira and St Vincent. 
 Ships on the South American and South African 
 routes will thus have a number of stations with which 
 to communicate on both outward and homeward 
 journeys. Italy of course has a very complete 
 system round the whole of her coastline, and there 
 are several stations in the Eastern Mediterranean. 
 In Canada and America the chain is complete round 
 both eastern and western coasts and on the Great 
 Lakes. Many of the principal islands of the West 
 Indies have stations, and there are a few scattered 
 round both the Atlantic and Pacific coasts of South 
 America. In the Pacific there is an important station 
 in Honolulu, and three in the Fiji Islands used partly 
 for communication with passing ships and partly for 
 
viii] THE SHORE STATIONS 99 
 
 ordinary telegraphic purposes between themselves. 
 Australia and New Zealand have rather lagged behind 
 for various reasons, but are now making up for lost 
 time. Round the coasts of Asia stations are sparsely 
 distributed from Petropavlovsk in Eastern Siberia to 
 Aden in the West. Japan has a few stations, and 
 there are several in the Dutch East Indies. In 
 Africa a number of stations have been erected in 
 the Italian East African colonies ; there is one at 
 Durban, another at Cape Town, and there are a few 
 on the western coast. 
 
 CHAPTER IX 
 
 THE USE OF WIRELESS TELEGRAPHY BETWEEN 
 FIXED STATIONS OVER LAND AND SEA 
 
 FOR the purposes of signalling between fixed and 
 permanent stations, whether over land or sea or 
 both, wireless comes immediately into competition 
 with the older telegraphy by submarine cable and 
 land lines. 
 
 In comparing the advantages of the two systems 
 in any given case four principal points have to be 
 dealt with, namely, Reliability, Rapidity of Trans- 
 mission, Cost, and Secrecy. 
 
 Reliability has been placed first and is without 
 
 72 
 
100 WIRELESS TELEGRAPHY [CH. 
 
 doubt the most important adjunct of any telegraph 
 service. At present the older system has probably 
 a slight advantage for both short and long distances. 
 Wireless signals may be delayed or wrongly trans- 
 mitted for three principal reasons, namely interference 
 or atmospherics, personal errors in sending or receiving, 
 and failure in the instruments. The two last possi- 
 bilities are common to both the old and new systems 
 equally, for the personal element is the same for both, 
 and against the greater possibility of failure of a 
 wireless station must be placed the chance of damage 
 to the cable or land line. Against the difficulty of 
 atmospherics and interference must be placed the 
 possibility of magnetic storms which entirely dis- 
 organise the ordinary telegraphy, but have no effect 
 on wireless. There is a case on record in which 
 Mr Marconi failed to get a message through by cable 
 owing to one of these storms, and had to fall back 
 upon a partially finished wireless station, with 
 successful results. Magnetic storms of this magni- 
 tude are, however, much less frequent than atmo- 
 spherics, and for this reason cable or land wire 
 telegraphy must be said to have the advantage. 
 Methods of eliminating the effects of all but the very 
 worst of these wireless pests are however proceeding 
 so rapidly that there is every hope of the new system 
 becoming as reliable as the old one in the course of 
 a few years. 
 
ix] USE BETWEEN FIXED STATIONS 101 
 
 With regard to Rapidity of Transmission wireless 
 is at present by no means as far advanced as the 
 ordinary telegraphy, which has had something like 
 50 years' start and has reached H Jrigh state of 
 perfection, whereas rapid transmission *6y wireless 
 is only just developing. With cable telegraphy it is 
 possible to arrange the instruments Iso'iKat' messages 
 can be sent both ways between two stations at the 
 same time, a method known as "duplex working." 
 The same results can be obtained with wireless. 
 
 The question of Cost is intimately connected with 
 the question of the speed of working. Obviously the 
 more work that can be got out of a system in a given 
 time the cheaper will be the service. Given equal 
 rates of working the relative cost of the two systems 
 must depend entirely on the conditions. Land wire 
 or cable will have to be used in some cases, quite in- 
 dependent of cost, e.g. for telegraphic communication 
 between London and Birmingham or Manchester. 
 Leaving out cases of this kind and dealing with those 
 in which both systems are possible, the choice between 
 the two will ultimately depend upon the cost of con- 
 struction and maintenance of a cable or land wire as 
 compared to the cost of erection and maintenance 
 of the wireless stations. Generally speaking the 
 constructional cost is less for the wireless stations. 
 With regard to maintenance the advantage will lie 
 with wireless in tropical and uncivilised countries, 
 
102 WIRELESS TELEGRAPHY .[OH. 
 
 where continual clearing and patrolling is necessary 
 for the land lines or in places where the shores are 
 rocky and the cost of maintenance of cables is high. 
 It must of course be remembered that wireless 
 signalling tfver long distances requires a good deal 
 of power 5 and the 'running costs of the power station 
 must-' be. considered in comparison with the main- 
 tenance costs of land lines or cables. The wireless 
 station has one advantage, namely that if there is a 
 breakdown it is on the spot with a staff and appliances 
 for repair immediately at hand, whereas with a land 
 line or submarine cable if there is a failure it has 
 first to be located and then repaired, and a long time 
 may elapse before communication is re-established. 
 
 Finally in comparing the two systems there is the 
 question of Secrecy to be considered, one to which 
 a good deal of importance has been attached. It is 
 an undoubted fact that wireless under all conditions 
 is not as secret as cable or land-line telegraphy. But 
 this objection is probably far less serious than appears 
 at first sight. It is presumably in connection with 
 commercial work that the difficulty is supposed to 
 arise as the wave-lengths are fixed and universally 
 used. Consequently if, say, a ship is sending some 
 message to a distant shore station, all surrounding 
 ships and perhaps several other shore stations will be 
 able to read the message. But it must be remembered 
 that it is only the operators at these stations who read 
 
ix] USE BETWEEN FIXED STATIONS 103 
 
 it, and they are under the same obligations as to the 
 secrecy of any message they overhear as they, or the 
 cable operators, are to any message which they 
 actually send or receive. Private stations may of 
 course also receive the message, but they are all 
 known and licensed, and in Great Britain, at any 
 rate, one of the conditions of granting the licence 
 is that no use shall be made of any messages which 
 may be picked up. It is unlikely that any leakage 
 could go on for any length of time without being 
 detected and promptly put a stop to. In the United 
 States, where up to 1912 little or no control of private 
 stations was exercised, the difficulty is (or was) 
 greater, but from recent events it would appear that 
 the trouble there was more in the direction of false 
 messages being sent out than in legitimate ones being 
 read and the information thus gained improperly used. 
 As a last resort some system of code can always be 
 made use of, as well for wireless as for other tele- 
 graphy. This proceeding is of course only applicable 
 to short messages. Press news could hardly be 
 transmitted in that way owing to the labour of 
 coding and decoding. In many cases, for example 
 Marconi's transatlantic service, this kind of news is 
 rendered sufficiently secret by the fact that it is not 
 worth anyone's while to put up a sufficiently large 
 aerial to "tap" the message privately. The news 
 can of course be picked up on a small aerial near to 
 
104 WIRELESS TELEGRAPHY [OH. 
 
 the station from which it is being sent, but it will 
 usually be of no value there. Nine-tenths of the 
 press news from Poldhu could be picked up on a 
 small aerial anywhere in the South of England, but 
 it would be of no value to anyone after they had 
 got it. 
 
 Summing up, it may be said that for some time 
 to come there is a wide and useful field for both 
 systems in which they will actually assist one another, 
 but that in the future competition between wireless 
 and submarine cable telegraphy is inevitable over 
 both long and short distances. Similar competition 
 will also in all probability arise between wireless and 
 long land lines. 
 
 At present Marconi's transatlantic service is more 
 advanced than any other case of telegraphy between 
 fixed land stations by wireless. A brief history of 
 the development of this service, together with some 
 indication of the difficulties which had to be over- 
 come before the now regular communication could 
 be maintained is given in Chap. XIII. Both at Clifden 
 and at Glace Bay the stations are in wild and isolated 
 positions overlooking the Atlantic. A large amount 
 of power is required, and consequently a large and 
 self-contained power installation is necessary at each 
 station. The wireless part of the apparatus is similar 
 on both sides. Separate sending and receiving aerials 
 are used to radiate long waves so as to enable the 
 
ix] USE BETWEEN FIXED STATIONS 105 
 
 large distances to be covered even under the worst 
 conditions. To enable a large amount of power to 
 be used and for these long waves to be efficiently 
 radiated enormous aerials are necessary. At Clifden 
 the transmitting aerial is about 170 feet in height 
 and extends from the power house for over three- 
 quarters of a mile in a direction straight away from 
 Glace Bay. The earth connections are taken to a 
 lake in a direction directly towards Glace Bay. 
 With this arrangement the signals are transmitted 
 powerfully in a direction towards Glace Bay, but 
 comparatively weakly in other directions. The 
 receiving aerial is of the same shape, but still longer, 
 extending for over a mile horizontally. Another 
 peculiar feature of these stations is that high voltage 
 direct current is used for transmitting purposes, 
 obtained from several specially constructed dynamos 
 all connected in series and working in conjunction 
 with a battery of about 6000 accumulator cells. The 
 condensers are also rather unusual, consisting of 
 large zinc sheets suspended from heavy insulators. 
 Alternate sheets are joined together forming the 
 plates of a large parallel-plate condenser having air 
 for its dielectric. This is rather a cumbrous arrange- 
 ment, but has the advantage that if by accident it is 
 over-charged and sparks across between the plates, 
 the dielectric is self-healing. With a glass-plate con- 
 denser such an occurrence necessitates the removal 
 
106 WIRELESS TELEGRAPHY [OH. 
 
 of the damaged plate which may take some time. 
 A rotary spark gap is used so as to ensure regular 
 sparking. The high voltage direct current supply 
 is connected straight up to the condensers, charging 
 them up in the intervals between the sparks ; and to 
 prevent a large current rushing from the dynamos 
 and battery when the spark takes place, inductively 
 wound coils are also connected in the circuit from 
 the dynamos. Thus every time the projecting studs 
 of the wheel come opposite the fixed ones a spark 
 takes place and the condensers discharge with 
 powerful oscillations. Inductively coupled with the 
 primary inductance is a secondary one connected to 
 the aerial and earth with an additional inductance 
 in series for tuning purposes. The arrangement is 
 therefore just the same as the usual coupled trans- 
 mitting circuit with the high voltage alternating 
 supply replaced by the direct. The "dots" and 
 "dashes" are made by means of a magnetically 
 operated switch in the circuit to the dynamos. This 
 switch required very careful design. Electrical 
 engineers will appreciate the difficulty of breaking 
 from two to three hundred horse-power at a pressure 
 of nearly 20,000 volts at the rate of about 200 times 
 a minute. This duty the switch has to perform day 
 in and day out with unfailing regularity. For receiv- 
 ing purposes Mr Marconi says that he uses the 
 Fleming oscillation valve with complete success, no 
 
ix] USE BETWEEN FIXED STATIONS 107 
 
 doubt with some highly selective form of receiving 
 circuit so as to prevent interference from other 
 signals as well as from atmospherics. 
 
 A fairly regular public service has been main- 
 tained since 1908, one of the difficulties experienced 
 being, curiously, with the connections to the land 
 telegraph systems, especially on the Canadian side, 
 where they are exposed to bad weather conditions. 
 Arrangements with the Postal Authorities are now 
 complete, and telegrams can be handed in at any 
 Post Office for transmission from England to America 
 and vice versa. The Transatlantic traffic however 
 is so large that the Cable Companies have not at 
 present felt this competition to any appreciable 
 extent. 
 
 With one exception the Clifden and Glace Bay 
 stations are the most powerful in existence. This 
 exception is a very large and imposing station erected 
 by the Marconi Co. in Italy at Coltano in the Gulf 
 of Genoa. Details of this station have not been 
 published, but it is understood to be designed for 
 a power of 1000 horse-power, and is to be used for 
 communication with the Italian East African colonies, 
 with Italian ships on the South American and South 
 African routes, and possibly with the South American 
 stations themselves. 
 
 Other well-known high-power stations are the 
 Eiffel Tower station in Paris, the large experimental 
 
108 WIRELESS TELEGRAPHY [CH. 
 
 station at Nauen near Berlin, belonging to the Tele- 
 funken Co., and the American stations at Arlington 
 near Washington arid at Brant Rock, on the eastern 
 coast, some 20 miles south of Boston. The Eiffel Tower 
 station is under the control of the French Government 
 War Departments and is used largely for experi- 
 mental purposes. Few details of the instruments in 
 use have been published. The aerial is in the form 
 of an inverted fan suspended from near the top of 
 the tower and spreading out over the surrounding 
 gardens at an angle of about 45 degrees. It is the 
 highest aerial in existence at present, and some very 
 long ranges have been obtained with the use of 
 comparatively small powers. The transmitting and 
 receiving instruments are placed in an underground 
 compartment partly for secrecy and partly so as not 
 to disfigure the gardens more than is necessary. The 
 latest installation is reported to be of 225 horse-power, 
 using a spark frequency of 1000 per second. 
 
 Nauen is principally interesting on account of its 
 aerial. This is of what is termed the "umbrella" 
 type, and is suspended from one large mast. The 
 original mast (see fig. 19) consisted of two parts with 
 a ball and socket joint at the bottom and at the 
 junction between the two and was 650 feet in height. 
 The lower part was of heavy steel lattice work con- 
 struction of triangular cross-section, the sides of 
 which were 13 feet in width. The upper part was 
 
ix] USE BETWEEN FIXED STATIONS 109 
 
 of similar construction but much lighter, and both 
 parts were supported by strong steel guys anchored 
 to heavy masses of concrete. This very bold con- 
 struction failed on account of one of these guys 
 breaking in a heavy gale with the result that the 
 whole mast and aerial collapsed. The present mast 
 is of similar construction but about 900 feet in 
 height, making it very nearly equal in height to the 
 Eiffel Tower. The aerial wires are spread out in all 
 directions from the top of the mast to a number of 
 small masts placed on the circumference of a circle 
 of about 500 yards in diameter, some of which can 
 be seen in the background. 
 
 The large American stations were erected partly 
 for experimental purposes and partly for communi- 
 cation with the naval ship and shore stations. They 
 are both situated on high land and have large 
 umbrella-type aerials suspended from masts some 
 600 feet in height. Power for transmitting purposes 
 is obtained from alternators giving a high spark 
 frequency and using rotary spark gaps on the alter- 
 nator shaft. It is proposed to use very high frequency 
 alternators and undamped waves. 
 
 The part played by wireless in the development 
 of South America, principally in the Amazon region, 
 is one of its most interesting applications to the duties 
 of ordinary inland telegraphy. Up to 1906 the only 
 regular communication over the whole of this vast 
 

 19. The Nauen Wirch-HH Station, near I'.nliri 
 
rn. ix] USE BETWEEN FIXED STATIONS 111 
 
 area consisted of a fable from Para to Manaos and a 
 very leisurely postal service carried by the steamers 
 up and down the river. This region is particularly 
 one for wireless as opposed to land line telegraphy. 
 For a land line wide clearances would have to have 
 been made and maintained through the forest, the 
 poles would have needed protection against both 
 insects and animals and constant patrolling and in- 
 spection would have been necessary. The cable it is 
 true could have been extended, but this would have 
 been expensive, and interruptions of the existing cable 
 were frequent. Both Marconi and Telefunken stations 
 are in use and there is now a complete line of com- 
 munication from Para, at the mouth of the Amazon, 
 to Lima on the Pacific coast of Peru. Stations have 
 also been erected to the south of this line in Peru, 
 Bra/il and Bolivia, forming a system of about ;>0 
 stations in all connecting up all the most important 
 settlements. The crossing of the Andes between Lima 
 and Kquitos was a new departure in wireless engineer 
 iug successfully carried out by the Telefunken Co. 
 lu the Congo a similar development has taken place. 
 Here the difficulties in the way of land wires are 
 even greater than in South America. No telegraph 
 poles have yet been designed capable of standing up 
 against a herd of wild elephants, 
 
 Lastly there is the great Imperial Wireless Scheme, 
 not yet certainly an accomplished fact because it has 
 
112 WIRELESS TELEGRAPHY [OH. 
 
 yet to receive the sanction of the House of Commons. 
 This scheme was entered upon after full discussion 
 at the Imperial Conference in 1911, partly for strate- 
 gical and partly for commercial reasons. It is felt 
 that a limited number of wireless stations can be 
 more effectively protected than many thousand miles 
 of cable which in case of international difficulties 
 could be so easily cut. The proposed arrangements 
 between the Postmaster-General, acting for both 
 Home and Colonial Governments, and the Marconi 
 Co. as to the cost and maintenance have been made 
 known, but technical details have only been published 
 in part. Six stations are suggested, situated in 
 England, Egypt, the East African Protectorate, the 
 Union of South Africa, India and near Singapore. 
 In addition to these another powerful station at 
 Fort Darwin, Australia, is proposed, which is to be 
 erected by the Australian Government but which is 
 to form a part of the whole scheme. These stations 
 will each have to be able to communicate under all 
 conditions with the stations on either side in the chain, 
 and the one in East Africa with three others, viz. 
 South Africa, India and Egypt. The stations are to 
 work duplex at 20 words per minute or with a simplex 
 automatic system at 50 words per minute. The Post- 
 master-General will pay the Company 60,000 per 
 station in two sums, 40,000 when the stations are 
 erected and the remaining 20,000 after they have 
 
ix] USE BETWEEN FIXED STATIONS 113 
 
 been satisfactorily worked by the Company for six 
 months. For 28 years after the completion of the 
 scheme the Postmaster-General will pay the Company 
 10 per cent, of the gross receipts by way of a royalty 
 in return for which they have the free use of all the 
 Marconi patents as well as the technical advice of the 
 Company. The wave-lengths to be used are from 
 17,000 to 50,000 feet. Care is to be taken to prevent 
 interference between the stations included in the 
 scheme as well as existing large power stations near 
 the route. Large aerials of course will have to be used 
 and as much advantage of directional effect will be 
 taken as possible. Power will be provided by alter- 
 nators driven by steam turbines varying from 1300 to 
 2500 horse-power. The Marconi type disc discharger 
 directly coupled to the alternator shaft is to be used 
 and a different note will be given to each station to 
 assist in the elimination of interference. The trans- 
 mitting condensers will be of the parallel-plate type 
 immersed in oil and the usual coupled transmitting 
 circuits will be employed. At the terminal stations 
 the steam generating plant and the turbo-alternators 
 are to be duplicated, one set of transmitting instru- 
 ments only being proposed. At the intermediate 
 stations a complete set of power generators and 
 transmitting instruments is to be provided for each 
 station with which communication will have to be 
 maintained, together with one set of spare boilers 
 
 F. W. T. 8 
 
114 WIRELESS TELEGRAPHY [OH- 
 
 and turbo-alternators. Thus in the East African 
 station there will be three complete sets of trans- 
 mitting instruments, including primary condensers, 
 primary and secondary inductances, and three aerials. 
 This multiplication of the apparatus is necessary in 
 order that communication may be maintained with 
 all three surrounding stations at the same time, a 
 state of affairs that may easily arise at times of heavy 
 traffic. The 2500 horse-power turbo-alternators, of 
 which there will be four sets, will be installed in 
 this station and when working simultaneously in all 
 three directions 7500 horse-power will thus be in use. 
 The Clifden and Glace Bay stations pale into insig- 
 nificance compared to this. It should be noticed, 
 however, that the power of all stations is expected 
 to be in excess of that actually required. In conse- 
 quence of the enormous amount of power to be 
 handled the signalling switches for breaking up the 
 signals into "dots" and "dashes" are in triplicate 
 throughout, with means of changing over from one 
 to the other very quickly in case of failure. Sig- 
 nalling at 50 words per minute, the circuit to the 
 primary condensers will have to be made and broken 
 at the rate of about eight times a second, and the power 
 to be broken may, at this switch, be about 2000 
 horse-power. It would not be surprising if these 
 switches presented greater difficulties both in design 
 and operation than any other part of the scheme. 
 
ix] USE BETWEEN FIXED STATIONS 115 
 
 The receiving stations are to be situated not less than 
 ten miles from the transmitting stations, this distance 
 being necessary for duplex working. No details of 
 the proposed receiving instruments are published 
 at present. 
 
 The contract for this scheme came in for very 
 adverse criticism on its publication, and a Select 
 Committee of the House of. Commons was appointed 
 to consider it. At the time of writing only an interim 
 report has been made by the Committee, who are 
 awaiting the results of an enquiry by an independent 
 Scientific Committee into the relative advantages of 
 the different systems. 
 
 Several other ambitious schemes on similar lines 
 are also mooted. For instance, the Marconi Company 
 are said to have started with the erection of a 
 chain of large stations connecting up England with 
 N. America, California, the Sandwich Islands, and 
 Japan. The French Government has been reported 
 to have in hand a scheme connecting up Europe with 
 S. Africa, S. America, and the Pacific. Germany, 
 with the aid of the new Nauen aerial, proposes to 
 maintain communication with N. America and her 
 colony of Togo in central Africa. If all these schemes 
 materialise the tuning will have to be very accurate 
 and very selective circuits will be necessary if serious 
 interference is to be prevented. 
 
 82 
 
116 WIRELESS TELEGRAPHY [CH. 
 
 CHAPTER X 
 
 THE USES OF WIRELESS TELEGRAPHY FOR NAVAL 
 AND MILITARY PURPOSES 
 
 EFFICIENT means of communication are one of 
 the most important essentials of modern warfare and 
 it is not surprising that both naval and military 
 authorities have carefully watched the development 
 of wireless from the very first. Before its introduc- 
 tion a fleet at sea was dependent for its information 
 upon high speed cruisers. Signals could not be 
 transmitted beyond the limits of vision either by day 
 or night, and in fog communication between ships 
 was possible by sound signals only. All this isolation 
 has now gone, with far-reaching results. An army in 
 the field has for years been able to signal over long 
 distances by means of temporary land wires. These 
 wires, however, were easily cut and consequently 
 required constant protection in an unfriendly country. 
 Wireless is not subject to this disadvantage as 
 both scientific training and the necessary apparatus 
 would be required to interrupt the service. Thus for 
 Military uses wireless has been merely an advance 
 on the old system but for Naval purposes it has been 
 a revolution. 
 
 It should be remembered that wireless signals can 
 be received by an Enemy as well as a Friend, so that 
 
x] FOR NAVAL AND MILITARY PURPOSES 117 
 
 war-time may perhaps be the quietest of times from 
 a wireless point of view. A destroyer prowling by 
 night near an enemy's fleet would be as likely to 
 use her wireless as switch on her searchlights. Her 
 presence would be given away as much by one as 
 by the other. Unless used with the greatest care 
 therefore wireless may be a source of danger, and it 
 is possible that the results of the whole of a series of 
 operations may depend upon the skill with which it 
 is employed. On account of this the great American 
 strategist, Admiral Mahan, has gone so far as to pre- 
 dict that in the near future it will be safer in some 
 circumstances to fall back upon the old system of 
 sending information by fast cruisers than to use 
 wireless. 
 
 Wireless has been used in actual warfare on 
 several important occasions. At the time of the 
 Boxer trouble in China temporary stations were fitted 
 up at the Taku forts and were found invaluable for 
 communication between them and the ships outside. 
 In the Russo-Japanese war it played a prominent 
 part, all the principal ships on both sides being fitted, 
 and the Russian army having at least three portable 
 sets in use. The Japanese both outside Port Arthur 
 and before the battle of Tsushima were able to keep 
 their principal ships at a safe distance and yet remain 
 in constant touch with their cruisers arid destroyers 
 who were watching the Russians. The latter were as 
 
118 WIRELESS TELEGRAPHY [OH. 
 
 a matter of fact on some occasions warned of the 
 intentions of the Japanese by overhearing and de- 
 ciphering their signals. 
 
 It is curious that there is no record of the Russian 
 ships in Port Arthur having made any use of their 
 wireless to maintain communication with the outside 
 world. Presumably the distances were too great for 
 the early type of apparatus with which they were 
 fitted. Could regular communication have been kept 
 up even during the first six months of the siege it is 
 possible that the ultimate end of the war might have 
 been quite different from that which was actually the 
 case. The Russian general, Stbssel, would have been 
 compelled to hand over supreme control of the defence 
 to the Fort Commandant, General Smirnoff, and other 
 inefficient officers would have been superseded, with 
 the probable result that at the very least, the capture 
 of the fortress and the destruction of the fleet shelter- 
 ing in the harbour would have been delayed for 
 several months. This would have greatly added to 
 the difficulties of the Japanese as it might have 
 allowed the Russian Baltic Fleet to effect a junction 
 with the Port Arthur ships, in which case the Japanese 
 fleet would have been inferior in numbers and the 
 results of a battle would have been doubtful. Had 
 the Japanese been defeated at sea they could not have 
 continued the campaign on land and victory would 
 have rested with Russia. 
 
x] FOR NAVAL AND MILITARY PURPOSES 119 
 
 During this war a small steamer, the "Hainum," was 
 fitted with de Forest apparatus and used with a 
 similar shore station near Wei-hai-wei for the purpose 
 of collecting first hand information for the Times 
 newspaper. Some annoyance was caused to both 
 belligerents and the steamer was probably very lucky 
 in escaping from being sunk or at any rate from 
 being seized and having her wireless gear removed 
 by one or other of them. 
 
 The Italians took a number of Marconi portable 
 sets to Tripoli, but the extent to which they were 
 used is rather uncertain. The large station at Derna, 
 which had previously been used for communication 
 with Turkey, was destroyed early in the war but was 
 refitted by the Italians for their own use between 
 their ships and their army. 
 
 Little is known at present of the extent to which 
 wireless has been used in the Balkan War. Adrianople 
 has certainly been in touch with the Turkish head- 
 quarters during practically the whole of the siege, 
 but beyond that practically no mention of its use 
 has been made. That a number of portable sets 
 of instruments have been available on both sides is 
 known, and if ever the veil of secrecy is lifted it is 
 more than probable that wireless will be found to 
 have played no unimportant part. 
 
120 WIRELESS TELEGRAPHY [OH. 
 
 CHAPTER XI 
 
 WIRELESS TELEGRAPHY ON AIRSHIPS AND 
 AEROPLANES 
 
 THE fitting of wireless apparatus to airships and 
 aeroplanes has generally been in connection with their 
 military duties. The advantage of wireless com- 
 munication to -an airship or aeroplane when engaged 
 in scouting operations is obvious. It will become to 
 them in the near future what it now is to a warship, 
 but at present this application, especially in the case 
 of aeroplanes, is in quite an experimental state. The 
 great difficulty lies in the fact that no earth con- 
 nection is possible. To radiate or absorb any 
 electromagnetic energy it is therefore necessary to 
 arrange for two sets of conductors to form a radiative 
 circuit similar to that used by Hertz. With airships 
 the difficulty is overcome by using the body of the 
 ship as one conductor and a long trailing wire for 
 the other, thus converting them into a huge vertical 
 Hertz oscillator. This arrangement would radiate 
 waves which could be picked up on the usual type 
 of earthed aerial if the two were tuned to the same 
 frequency. The same device has been tried with 
 aeroplanes but with little success. On one occasion 
 a Farman biplane was fitted with two parallel trailing 
 
xi] ON AIRSHIPS AND AEROPLANES 121 
 
 wires which together made up the radiative circuit. 
 These wires were naturally found to be a nuisance on 
 the aeroplane to say nothing of the danger of their 
 becoming involved with the propeller, and the latest 
 arrangement is to attach two sets of wires to the 
 
 7afline o/ Aeroplane, 
 
 u Oaf line o/ Aeropla 
 
 P 
 
 Fig. 20. Arrangement of Wireless Apparatus fitted to Aeroplanes 
 
 planes themselves as shewn in the diagram. The 
 capacity of one of these sets with respect to the 
 other will obviously be very small on an aeroplane 
 whose greatest dimension does not much exceed 
 20 feet, and consequently waves of only about 300 
 
122 WIRELESS TELEGRAPHY [OH. 
 
 feet in length can be efficiently used. The trans- 
 mitting instruments must obviously be as light as 
 they can possibly be made. On airships comparatively 
 powerful sets can be employed, absorbing some two or 
 three horse-power which can be obtained from the 
 main engines, but on aeroplanes small batteries have 
 to be used, the greatest output from which does not 
 exceed about one-tenth of a horse-power. So far as 
 details have been published it appears that one form 
 or other of the usual coupled transmitting circuits has 
 been used. The arrangements shewn in the diagram 
 are those fitted to an aeroplane used in trials carried 
 out on Salisbury Plain in connection with the Annual 
 Manreuvres of 1910. Signals from such an arrange- 
 ment could not be exchanged with an ordinary earthed 
 aerial because the waves given out are in the wrong 
 plane. Some form of horizontal radiator and absorber 
 was necessary and took the form of two similar sets 
 of wires arranged fan wise in a horizontal plane, the 
 apices of the two fans being near one another, with 
 the receiving and transmitting instruments between 
 them. 
 
 In comparison with the wireless apparatus in use 
 on board ship and in the land stations these arrange- 
 ments appear crude, and it is not surprising that no 
 very long ranges have been attained. Receiving 
 signals appears to be more difficult than sending 
 them, and there are at present only one or two 
 
XT] ON AIRSHIPS AND AEROPLANES 123 
 
 records of them having been received at all on 
 aeroplanes. In the case of the Wellman airship on 
 the other hand the receiving was better than the 
 sending, probably owing to the small amount of 
 power used for the latter purpose. 
 
 The German military airships are all stated to 
 have been fitted, and have been at various times 
 reported to have maintained communication over 
 long distances. The Clement-Bayard airship which 
 came over from Paris to London in October 1910 
 was equipped but the extent to which the apparatus 
 was used is uncertain. The ill-fated Wellman airship 
 " America " had a small Marconi set of instruments 
 and a Marconi operator on board which were found 
 useful, though curiously when the airship had to be 
 abandoned no reply could be got to the distress 
 signals, and the attention of a passing steamer was 
 called during the night by hand signalling with a 
 lamp. The steamer presumably carried only one 
 operator who was not then looking out for signals. 
 The large French air cruiser "Adjudant Vince'nnot" 
 was able in September 1911 to maintain easy com- 
 munication with both the Eiffel Tower station and 
 with one at the fortress of Verdun during the whole 
 of a 13-hour flight. 
 
 Turning to the aeroplanes, during the experiments 
 on Salisbury Plain distances up to three miles were 
 obtained, and better results have since been reported 
 
124 WIRELESS TELEGRAPHY [OH. 
 
 with similar apparatus. In France the Farman 
 biplane with the two trailing wires was able to 
 transmit signals to the enormous Eiffel Tower aerial 
 over distances up to 40 miles. In May 1911 a German 
 aeroplane was reported to have been in touch with 
 the large station at Nauen during the whole of a 
 flight from Berlin to Hamburg. Very recently after 
 a long flight from Eastchurch to Portsmouth, Com- 
 mander Samson is said to have reported his safe 
 arrival to officers at the former place by means 
 of the wireless apparatus fitted to his hydro- 
 aeroplane. 
 
 CHAPTER XII 
 
 WIRELESS TELEPHONY 
 
 ALMOST from the first introduction of the tele- 
 phone, efforts have been made to get rid of the 
 connecting wires. A long series of experiments were 
 made in Germany by Ruhmer with a " speaking arc " 
 and a selenium cell, but the apparatus was cumber- 
 some, inefficient, and of no practical use. The 
 "conductive-inductive" method of telegraphy be- 
 tween two parallel wires was used for telephony with 
 better results. A microphone transmitter was con- 
 nected in series with the sending wire, and when 
 
xn] WIRELESS TELEPHONY 125 
 
 spoken into, the sounds were reproduced in the 
 telephone receivers in the receiving wire. This 
 method was used by Preece between the Skerries 
 and Cemlyn and also between the island of Rathlin 
 and the mainland with such success that the former 
 of these two installations is still in use. The range 
 of this method is, however, limited, even when the 
 parallel wires can be made several miles in length. 
 
 The only system of wireless telephony really pre- 
 senting any hope of being of use over long distances, 
 is the one using the same electromagnetic waves as 
 wireless telegraphy. Fessenden and de Forest in 
 America have devoted a great deal of attention to 
 this system since 1899. The former was in 1908 
 successful in transmitting speech quite clearly from 
 Brant Rock to New York, a distance of about 400 
 miles. In the same year, de Forest apparatus was 
 fitted to several of the United States warships making 
 their famous round-the-world tour. The installations 
 do not seem to have been of much practical use 
 although no difficulties were experienced in communi- 
 cating over distances up to 50 miles. In Germany 
 the Poulsen system was adapted to telephonic trans- 
 mission with some success and ranges of 250 miles 
 were obtained. A number of systems, de Forest, 
 Poulsen and others, were tried at the Eiffel Tower with 
 the result that speech was clearly transmitted on one 
 occasion over a distance of 155 miles. During 1909 
 
126 WIRELESS TELEGRAPHY [OH. 
 
 the French cruiser "Conde" was fitted, and distances 
 up to 100 miles reached. The British Admiralty 
 were also at this time reported to have been carrying 
 out trials with the de Forest apparatus, but with 
 only moderate success. Latterly it appears that 
 rather less interest has been taken in the problem, 
 at any rate fewer "inventors of the wireless tele- 
 phone" have appeared. 
 
 To understand the action of the Wireless Tele- 
 phone it will be necessary to touch very briefly on 
 the question of the vibrations of the air set up when 
 a person is speaking. The act of speaking consists in 
 expelling air from the throat and mouth in such a 
 way that certain particular vibrations are set up 
 corresponding to the particular sounds it is desired 
 to produce. These vibrations are of small amplitude 
 and high frequency and are of course quite inde- 
 pendent of the bodily movement of the mass of the 
 air. Some of these vibrations are quite regular ones 
 of definite frequency, as for instance those which go 
 to make the sound 55 in "Coo." Others such as e in 
 "Me" or a in "Ma" are a little more complex, being 
 made up of combinations of two or three simple 
 vibrations of different frequencies. The consonant 
 sounds, on the other hand, are much less regular and 
 do not consist of vibrations which repeat themselves 
 periodically. The problem of the transmitter with 
 all telephony is to make use of these air vibrations to 
 
xn] WIRELESS TELEPHONY 127 
 
 produce changes in an electric current which shall be 
 exact reproductions of the air vibrations. If this can 
 be done the currents can be used to reproduce the 
 sounds at a distance in a telephone receiver. For the 
 ordinary telephony with wires the air vibrations are 
 thrown up against a thin metallic diaphragm by means 
 of the trumpet mouthpiece, causing small movements 
 in it exactly similar to the movements of the air. 
 Behind this diaphragm and between it and a fixed 
 plate is a small quantity of fine granulated carbon. 
 A battery is connected up with one terminal to the 
 diaphragm and one to the fixed plate. When the 
 diaphragm is set vibrating the contact resistance of 
 the granules varies in a manner nearly proportional 
 to the magnitude of the diaphragm's motion. This 
 variation of the resistance causes a variation of the 
 current from the battery, which can be made use of 
 to cause corresponding movements in the receiver 
 diaphragm. The latter diaphragm in turn sets up 
 vibrations in the air surrounding it. If the whole 
 motions have been faithfully reproduced throughout, 
 then these vibrations given out will be of the same 
 nature as those taken in, and to an ear placed near 
 the receiver the sounds will appear the same as those 
 produced at the transmitter end. 
 
 With wireless telephony practically the same pro- 
 cess has to be gone through. In the transmitting 
 aerial continuous oscillations have to be maintained 
 
128 WIRELESS TELEGRAPHY [OH. 
 
 whose magnitude can be varied in exact reproduction 
 of the vibrations constituting the sounds which have 
 to be transmitted. Variable contact devices similar 
 to those used for ordinary telephony are as a matter 
 of fact employed. The receiving aerial, which is tuned 
 electrically to the same frequency as the transmitting 
 aerial, thus receives a continuous train of waves of 
 varying magnitude. With a suitable detector con- 
 nected up with telephone receivers in exactly the 
 same way as for wireless telegraphy, these variations 
 of the magnitude of the waves can be converted into 
 variations of the magnitude of the current in the 
 telephone receivers. Suppose for example the sound 
 56 is being transmitted. Before the sound begins the 
 transmitting aerial is giving out, and the receiving 
 aerial is taking in, a continuous wave of invarying 
 amplitude. At the receiving end this is accompanied 
 by a steady current through the telephone receivers 
 and no sounds are produced. But after the speech 
 has started there will be regular rhythmical variations 
 of the intensity of the waves given out and received. 
 With increase of the intensity the detector in the 
 receiving aerial rectifies more of the oscillations, and 
 with decrease of the intensity, less. The result there- 
 fore is that the current in the telephone receivers 
 increases and decreases in the same way as the 
 intensity of the waves, and the sounds produced in 
 these receivers will be the same as those actuating 
 
xn] WIRELESS TELEPHONY 129 
 
 the transmitter at the transmitting station. For the 
 consonant sounds the same process goes on except 
 that the variations are irregular. 
 
 It may be noted that the electrical wave frequency 
 has no direct connection with the frequency of the 
 air vibrations. Electrical wave frequencies of from 
 50,000 to 1,000,000 are used, whereas the frequency 
 of the air waves varies from about 200 to 5000. It is 
 obvious that the frequency of the electrical oscilla- 
 tions must be much greater than that of the air 
 vibrations, because for each variation of the latter a 
 series of electrical oscillations is required from which 
 current for the telephone receivers can be rectified. 
 If this was not the case each variation of the air 
 movements would not be reproduced and the speech 
 would not be clear. 
 
 Continuous oscillations in the transmitting aerial 
 are not absolutely essential for wireless telephony. 
 Spark systems may be used as long as the spark 
 frequency is not less than 10,000 per second. As each 
 spark is accompanied by a train of oscillations there 
 will be currents in the aerial practically the whole of 
 the time and the behaviour is very much the same as 
 with continuous oscillations. 
 
 For satisfactory telephonic communication there 
 are three principal requirements; firstly the source 
 of continuous, or practically continuous, oscillations ; 
 secondly a means of varying the intensity of these 
 
 F. W. T. 9 
 
130 WIRELESS TELEGRAPHY [CH. 
 
 oscillations in the transmitting aerial corresponding 
 to the air vibrations ; and thirdly a detector at the 
 receiving end whose rectifying action is proportional 
 to the intensity of the incoming waves. The first and 
 third of these requirements present no serious diffi- 
 culties. The continuous oscillations of the required 
 frequency are produced by high frequency alternators 
 of the Alexanderson type as used by Fessenden in 
 his Brant Rock-New York experiments or by the 
 Poulsen or de Forest oscillating arcs. For the 
 detectors, the electrolytic and several other integrat- 
 ing detectors have the necessary properties. The 
 second requirement, however, is the difficulty which 
 has yet to be overcome. For telegraphic signalling 
 over ranges up to, say, 500 miles some two or three 
 horse-power is required in the aerial. The varia- 
 tions of this power are from the full power during the 
 signals to zero in the intervals between the signals. 
 To telephone over the same distance with the same 
 aerial nearly as large a variation of power would be 
 necessary. But these variations of power have to 
 take place, not at the rate of three or four times per 
 second, but at rates of from 200 to 5000 times per 
 second, and the only source of energy that is available 
 to effect this variation is that which can be obtained 
 from the vibrations set up by the voice. Looked at 
 from this point of view the results which have actually 
 been obtained are almost unbelievable. Microphones 
 
xn] WIRELESS TELEPHONY 131 
 
 similar to those used for ordinary telephony are 
 generally employed, but with modifications to enable 
 them to deal effectually with the large currents and 
 pressures which are necessary in the transmitting 
 aerials. Water cooling devices have been tried with 
 these microphones to make their action more regular, 
 as it was found that overheating was one of the 
 reasons for the carbon granules setting together and 
 being unaffected by the vibrations of the diaphragm. 
 In some cases several microphones have been used in 
 series or in parallel with improved results. Various 
 devices, too, have been tried intended to relieve the 
 microphone of its very arduous duties by arranging 
 it merely to throw the transmitting aerial out of tune 
 instead of directly varying the magnitude of the 
 current in it. As a matter of fact the work the 
 microphone has to do is not so much reduced as 
 appears to be the case, and the improvement is not 
 very great. The difficulty, in fact, seems at present 
 to be insuperable, and is the barrier in the way of 
 the extensive use of the wireless telephone. 
 
 The utility of wireless telephony if these difficulties 
 could be overcome lies in the fact that it would be 
 more rapid in its action and would not require a 
 practised operator to read the signals. Anyone 
 understanding the language in use would be able to 
 communicate by its means, and arrangements could 
 be made for two or three people to carry on a 
 
 12 
 
132 WIRELESS TELEGRAPHY [CH. 
 
 conversation together. In any case its application 
 would of course be limited to ships and possibly to 
 a very few long distance installations. Its use in 
 a busy town with several thousand subscribers is 
 inconceivable. 
 
 CHAPTER XIII 
 
 HISTORY 
 
 THE history of Wireless Telegraphy, or to give it 
 its official title, Radio-telegraphy, must be regarded 
 as dating from the publication of Maxwell's famous 
 paper, " A Dynamical Theory of the Electromagnetic 
 Field," read before the Royal Society on December 
 8th, 1864. It is true that signalling between stations 
 by electrical means without the use of connecting- 
 wires had been carried out earlier than this, but the 
 methods employed differ from those of modern wire- 
 less and, although they continued to be used until as 
 late as 1897, their practical application is very limited. 
 In his paper Maxwell proved theoretically the exist- 
 ence of electromagnetic waves, but the practical 
 confirmation of the theory was not forthcoming until 
 1878 when Hertz published the results of his experi- 
 ments with oscillating currents. Following this many 
 eminent physicists took up the subject and the 
 
xni] HISTORY 133 
 
 possibility of using the waves for signalling purposes 
 occurred to some of them. It is interesting to note 
 that in some of Hertz's experiments tuned circuits 
 were used, and that in 1894 Lodge had demonstrated 
 before a large audience the possibility of signalling 
 from one room to another through several walls. 
 
 It remained however to Marconi to produce the 
 first set of apparatus specifically constructed for sig- 
 nalling from point to point by the agency of electro- 
 magnetic waves. He brought his apparatus, which 
 was of the simple type now known as "plain aerial," 
 to England early in 1896 and took out his first patent 
 in June of that year. Marconi's great advance lay in 
 the use of the earth connection which is one of the 
 most important points covered by this first patent. 
 Many demonstrations were given in the presence of 
 representatives of the various Government Depart- 
 ments and other prominent people, and as more 
 experience was gained greater ranges were obtained. 
 By 1899 signalling across the English Channel had 
 been accomplished and the East Goodwin Lightship 
 had been placed in communication with the South 
 Foreland Lighthouse. This is perhaps the first 
 practical application of wireless. A few ships and 
 shore stations had been equipped and the difficulties 
 of interference immediately presented themselves. 
 To overcome them, Lodge, in 1898, patented his 
 method of using more sustained oscillations by means 
 
134 WIRELESS TELEGRAPHY [OH. 
 
 of which more advantage could be taken of the tuning 
 of the transmitting and receiving stations. For the 
 same purpose Marconi in 1900 introduced the coupled 
 transmitting circuit with fixed spark balls, the famous 
 "four sevens" patent being taken out in that year. 
 Others besides Marconi and Lodge were working on 
 the subject, notably Braun and Slaby in Germany, 
 Fessenden and de Forest in America, each contribut- 
 ing his share to the general advance and in many 
 cases taking out patents for similar devices at almost 
 the same time. These many patents led later to long- 
 patent actions in which, with the exception of the 
 Lodge patent, Marconi has been consistently suc- 
 cessful, from which it must be assumed that he was, 
 on the whole, ahead of his competitors. 
 
 The coupled circuit system with improvements is 
 the system in almost universal use to-day. The 
 tendency of the changes has been in the direction 
 of increased efficiency, higher spark frequency and 
 longer wave-lengths. The spark gap has probably 
 been subject to the greatest alterations. The fixed 
 spark balls being found unsatisfactory for high spark 
 frequency, Marconi gradually adopted a rotary gap. 
 At first this was driven by a separate small motor, 
 but during 1910, and after, it was fitted to an extension 
 of the alternator shaft. A similar arrangement had 
 been successfully used by Fessenden in 1906 in a 
 station in Scotland used for signalling across the 
 
xm] HISTORY 135 
 
 Atlantic. In the same year, 1906, Wien, in Germany, 
 pointed out the advantages of "charging by impact." 
 This was taken up by the Gesellschaft fur Drahtlose 
 Telegraphic, the manufacturers of the "Teleftmken" 
 apparatus, who use it in their now well-known "singing 
 spark" instruments introduced in 1909. 
 
 The practical use of continuous oscillations also 
 dates from about 1906 when Poulsen developed his 
 oscillating arc apparatus into a more or less reliable 
 system. Some three or four years later means of 
 generating these continuous oscillations directly from 
 special types of alternators were devised by Alexander- 
 son in America, and Goldschmidt in Germany. Still 
 later, in 1912, a third method was introduced by the 
 Telefunken Company. 
 
 The development of the receiving instruments has 
 been similarly progressive. The metallic filings coherer 
 of 1896 proved unreliable and was quickly superseded. 
 In America and Germany thermal and electrolytic 
 detectors were used. Marconi's magnetic detector 
 was introduced in 1901 and immediately proved 
 itself superior for general use to any other detecting 
 instrument then known. The modern rectifying 
 detectors and their highly selective circuits came in 
 by degrees from about 1906 onwards. 
 
 Marconi's early efforts at signalling across the 
 Atlantic are not without interest. The first attempt 
 was made in December, 1901. The Poldhu station, 
 
136 WIRELESS TELEGRAPHY [CH. 
 
 the largest then in existence and having installed in 
 it machinery of some 30 to 40 horse-power, was just 
 completed. A temporary aerial attached to a kite 
 was sent up at St John's, Newfoundland, and the pre- 
 arranged signals from the Cornwall station were 
 picked up on three consecutive days. Having thus 
 proved the possibility of signalling over this distance, 
 large stations were designed for Cape Cod and Cape 
 Breton, and an increase in the size of the aerial 
 was arranged for at Poldhu. This work was com- 
 pleted in 1902 and signalling in both directions was 
 carried out. For the regular communication required 
 for commercial purposes larger aerials and more 
 power were soon found necessary. The large Clifden 
 (Ireland) and Glace Bay (Nova Scotia) stations were 
 taken in hand in 1906, and after many preliminary 
 experiments a fairly regular communication was 
 established on commercial lines in 1908. 
 
 Soon after Marconi's first demonstrations in 
 England a company, "The Wireless Telegraph and 
 Signal Co., Ltd." was formed to work his patents and 
 manufacture his apparatus. This company, the name 
 of which was changed in 1900 to "Marconi's Wireless 
 Telegraph Co., Ltd.," occupies a very important place 
 in the history of wireless telegraphy. A related com- 
 pany, "The International Marine Communication Co., 
 Ltd." soon followed, which took over the working of 
 the apparatus installed in the ship and shore stations, 
 
xm] HISTORY 137 
 
 the organization of the signalling and the control and 
 training of the operators. Subsidiary manufacturing 
 companies in which the original company is financi- 
 ally interested, have also been formed in America, 
 Germany, France and many other countries. Owing 
 to the decisions arrived at in recent patent actions, 
 here and abroad, these companies have for all prac- 
 tical purposes a world-wide monopoly. Their only 
 important competitors at present are the Telefunken 
 and Poulsen Companies, and possibly, in the near 
 future, the Compagnie Universelle de Telegraphic et 
 de Telephonic sans Fil, who own the Goldschmidt 
 patents. 
 
 Early in its development the necessity for some 
 international agreement with respect to wireless 
 signalling became obvious. In 1903 a conference 
 was held at Berlin without definite results. A second 
 was held in 1906 and a long Convention was signed 
 by the delegates. The principal provisions of this 
 Convention were : 
 
 1. The fixing upon 300 and 600 metres as the 
 standard wave-lengths for commercial purposes ; a 
 range of from 600 to 1600 metres being reserved for 
 naval and military uses. 
 
 2. An agreement that all stations should inter- 
 communicate freely, independently of the type of 
 apparatus with which they were fitted. 
 
 3. A number of service regulations relating to 
 
138 WIRELESS TELEGRAPHY [OH. 
 
 methods of calling up and answering, distress signals, 
 control of tariffs, etc., etc. 
 
 This Convention of course required ratification 
 from the various Governments before becoming law 
 in their countries. Many Governments ratified at 
 once. The Marconi Co. looked upon the intercom- 
 munication clause as an attempt to rob them of the 
 advantages which they were gaining from their private 
 organization, their operators at that time not being- 
 allowed to communicate with stations fitted with 
 other than Marconi apparatus. They therefore pro- 
 tested against ratification of the Convention both in 
 England and America. In England a Parliamentary 
 Committee was appointed to go into the question 
 fully. After hearing a vast amount of expert evidence 
 the Committee decided by a majority of one in favour 
 of ratifying, which was consequently carried out. 
 The United States, Italy and Japan did not ratify 
 immediately, but by the time of the holding of the 
 third Conference in London in 1912 they had all 
 come into line. At this third Conference many details 
 were arranged for the improvement of the service, 
 but the most important matter dealt with was the 
 question of the use of wireless for saving life at sea. 
 This question was put before the Conference by the 
 British delegates at the request of the Government 
 as a direct consequence of the "Titanic" disaster. 
 The result of the discussion was a recommendation 
 
xni] HISTORY 139 
 
 that all ships above a certain size should be 
 equipped with wireless apparatus and carry qualified 
 operators. 
 
 At the same time that international regulations 
 were being evolved, most countries found it necessary 
 to exercise some control over the use of wireless in 
 their own territories. In Great Britain the Wireless 
 Telegraphy Act was passed in 1904 and is still in 
 force. Its most important enactment is that no 
 station of any kind may be used except under licence 
 from the Postmaster- General. In this way the in- 
 discriminate use of powerful transmitting apparatus 
 by amateurs is largely prevented, whereas in America, 
 where no such act was in force, the result was chaos. 
 
BIBLIOGRAPHY 
 
 A Handbook of Wireless Telegraphy. Erskine Murray. (London. 
 
 Lockwood and Sons.) 
 A History of Wireless Telegraphy. Fahie. (Edinburgh. Blackwood 
 
 and Sons.) 
 A Manual of Wireless Telegraphy. Collins. (New York. Wiley 
 
 and Sons.) 
 An Elementary Manual of Radio-Telegraphy and Radio-Telephony. 
 
 Fleming. (London. Longmans and Co.) 
 Die Elektrische Wellentelegraphie. Arendt. (Brunswick. F. Vieweg 
 
 und Sohn.) 
 Electric Waves. Hertz. English translation by D. E. Jones. 
 
 (London. Macmillan and Co.) 
 
 Electromagnetische Schwingungen und Drahtlose Telegraphie. Zen- 
 neck. (Stuttgart. Enke and Co.) 
 
 La Telegraphie sans Fils. Poincare. (Paris. Naud.) 
 La Telegraphic sans Fils et les Ondes Electriques. Boulanger et 
 
 Ferric. (Paris. Berger-Levrault.) 
 
 La Telegraphie sans Fils. Van Dam. (Paris. Ch. Beranger.) 
 Leitfaden der Drahtlosen Telegraphie. Zenneck. (Stuttgart. Enke 
 
 and Co.) 
 
 Les Oscillations Electriques. Poincare. (Paris. Carre.) 
 Les Oscillations Electriques : Principes de la Telegraphie sans Fil. 
 
 Tissot. (Paris. Doin et Fils.) 
 Principles of Wireless Telegraphy. Pierce. (New York. M c Graw- 
 
 Hill Book Co.) 
 
 Radio-Telegraphy. Monckton. (London. Constable and Co.) 
 Signalling across Space without Wires. Lodge. (London. The 
 
 Electrician Publishing Co.) 
 The Principles of Electric Wave Telegraphy and Telephony. Fleming. 
 
 (London. Longmans and Co.) 
 Wireless Telegraphy and Wireless Telephony. Kennelly. (New 
 
 York. Moffat Yard and Co.) 
 
INDEX 
 
 Aerial, 14, 32 
 
 capacity of, 37 
 
 on board ship, 86 
 
 uses of, 52 
 Airships and Aeroplanes, wireless 
 
 in, 120 
 
 Alternating currents, for trans- 
 mitting, 56 
 
 Alternators, high frequency, 63 
 Atmospheric absorption, 44 
 Atmospherics, 27, 48 
 
 elimination of, 47 
 
 Cape Cod station, 81, 84, 136 
 Circuit 
 
 non-radiative, 30 
 
 oscillatory, 22 
 
 radiative, 32 
 
 receiving, 70 
 Clifden station, 104, 136 
 Coherer, 67, 135 
 Coltano station, 107 
 Compass, wireless, 78 
 Condenser, 14 
 
 capacity of, 14 
 
 parallel plate type, 14 
 Conductors, 2 
 Continuous oscillation, 19, 135 
 
 for transmitting, 63 
 
 for receiving, 75 
 
 Coupled system 
 
 advantages of, 56 
 
 disadvantages of, 58 
 
 for transmitting, 54, 134 
 Coupling 
 
 inductive, 12 
 
 conductive, 92 
 Current, 2 
 
 alternating, 5 
 
 direct, 5 
 
 induced, 10 
 
 magnetic effects of, 9 
 
 oscillatory, 6 
 Cycle, 6, 7 
 
 Damping, 19 
 
 of spark gap, 54, 58, 62 
 Detectors 
 
 crystalline, 73 
 
 electrolytic, 74 
 
 Fleming valve, 74 
 
 function of, 66 
 
 magnetic, 67, 135 
 
 rectifying, 70, 135 
 Dielectric, 14 
 Distress signals, 80, 85 
 
 Eiffel Tower station, 79, 107 
 Electrical pressure, 3 
 Electric strain, 29 
 
142 
 
 INDEX 
 
 Electromotive force, 3 
 induced, 10 
 
 Frequency 
 
 of alternating current, 6 
 of oscillatory current, 7 
 natural, of oscillatory circuit, 20 
 
 Glace Bay station, 104, 136 
 Hertz, 40, 132 
 
 Imperial Wireless Scheme, 111 
 
 Inductance, 13 
 
 Induction 
 
 mutual, 12 
 
 self, 12 
 
 Interference, 27, 47 
 International Convention, 137 
 
 Lodge, 133, 134 
 
 Marconi 
 
 apparatus on board ship, 86 
 Companies, 136 
 Telegraphic Service, 82 
 
 Marconi's 
 
 early demonstrations, 133 
 transatlantic experiences, 42 
 transatlantic service, 104, 135 
 
 Maxwell, 40, 132 
 
 Morse code, 50 
 
 Musical note, 49, 57 
 
 Nauen station, 108, 115 
 Non-conductors, 2 
 
 Oscillatory current, 6 
 production of, 17 
 
 Plain aerial system, 51 
 disadvantages of, 54 
 
 Poldhu station, 81, 84, 136 
 Poulsen arc system, 63, 135 
 Press news, 81, 84, 103 
 
 Quenched spark system, 61, 135 
 
 Kesistance, 4 
 Resonance, 22 
 
 Eusso-Japanese War, wireless in, 
 117 
 
 Selectivity, 27 
 
 effect of damping of trans- 
 mitting circuit on, 28 
 of transmitting systems, 54, 
 
 62, 64 
 
 Ship stations, 77 
 Shore stations, 95 
 Sound signals, analogy to wire- 
 less signals, 1 
 South America, wireless develope- 
 
 ments in, 109 
 Spark gap, 5 
 multiple, 63, 92 
 rotary, 59, 88 
 
 use in transmitting circuit, 52, 
 55 
 
 Telefunken system, 61 
 
 apparatus on board ship, 90 
 Time signals, 79 
 Tuning, 25 
 
 Warnings, wireless, to ships. 
 
 78 
 Water analogy 
 
 of condenser, 16, 18, 23 
 
 of current, 2, 5 
 
 of oscillatory current, 18, 23 
 Wave-length, 37 
 
 of waves radiated from an 
 aerial, 38 
 
INDEX 143 
 
 Waves, electromagnetic, 33 Wireless Telegraphy Act, 139 
 reflection and refraction of, 45 Wireless Telegraphy versus Tele- 
 variation of intensity with graphy with wires, 99 
 
 distance, 40 Wireless Telephony, 124 
 
 Wieii, 61, 135 developement of, 124 
 
 Wireless, compulsory on board difficulties of, 130 
 
 ship, 80 principles of, 126 
 
PRINTED BY JOHN CLAY, M.A. 
 AT THE UNIVERSITY PRESS 
 
THE 
 
 CAMBRIDGE MANUALS 
 OF SCIENCE AND LITERATURE 
 
 Published by the Cambridge University Press 
 GENERAL EDITORS 
 
 P. GILES, LittD. 
 
 Master of Emmanuel College 
 
 and 
 A. C SEWARD, M.A., F.R.S. 
 
 Professor of Botany in the University of Cambridge 
 
 70 VOLUMES NOW READY 
 HISTORY AND ARCHAEOLOGY 
 
 Ancient Assyria. By Rev. C. H. W. Johns, Litt.D. 
 
 Ancient Babylonia. By Rev. C. H. W. Johns, Litt.D. 
 
 A History of Civilization in Palestine. By Prof. R. A. S. 
 
 Macalister, M.A., F.S.A. 
 
 China and the Manchus. By Prof. H. A. Giles, LL.D. 
 The Civilization of Ancient Mexico. By Lewis Spence. 
 The Vikings. By Prof. Allen Mawer, M.A. 
 New Zealand. By the Hon. Sir Robert Stout, K.C.M.G., LL.D., 
 
 and J. Logan Stout, LL.B. (N.Z.). 
 The Ground Plan of the English Parish Church. By A. 
 
 Hamilton Thompson, M.A., F.S.A. 
 The Historical Growth of the English Parish Church. By A. 
 
 Hamilton Thompson, M.A., F.S.A. 
 
 English Monasteries. By A. H. Thompson, M.A., F.S.A. 
 Brasses. By J. S. M. Ward, B.A., F.R.Hist.S. 
 Ancient Stained and Painted Glass. By F. S. Eden. 
 
 ECONOMICS 
 
 Co-partnership in Industry. By C. R. Fay, M.A. 
 Cash and Credit. By D. A. Barker. 
 The Theory of Money. By D. A. Barker. 
 
LITERARY HISTORY 
 
 The Early Religious Poetry of the Hebrews. By the Rev. 
 
 E. G. King, D.D. 
 The Early Religious Poetry of Persia. By the Rev. Prof. J. 
 
 Hope Moulton, D.D., D.Theol. (Berlin). 
 The History of the English Bible. By John Brown, D.D. 
 English Dialects from the Eighth Century to the Present Day. 
 
 By W. W. Skeat, Litt.D., D.C.L., F.B.A. 
 King Arthur in History and Legend. By Prof. W. Lewis 
 
 Jones, M.A. 
 
 The Icelandic Sagas. By W. A. Craigie, LL.D. 
 Greek Tragedy. By J. T. Sheppard, M.A. 
 The Ballad in Literature. By T. F. Henderson. 
 Goethe and the Twentieth Century. By Prof. J. G. Robertson, 
 
 M.A., Ph.D. 
 
 The Troubadours. By the Rev. H. J. Chaytor, M.A. 
 Mysticism in English Literature. By Miss C. F. E. Spurgeon. 
 
 PHILOSOPHY AND RELIGION 
 
 The Idea of God in Early Religions. By Dr F. B. Jevons. 
 
 Comparative Religion. By Dr F. B. Jevons. 
 
 Plato : . Moral and Political Ideals. By Mrs A. M. Adam. 
 
 The Moral Life and Moral Worth. By Prof. Sorley, Litt.D. 
 
 The English Puritans. By John Brown, D.D. 
 
 An Historical Account of the Rise and Development of Presby- 
 
 terianism in Scotland. By the Rt Hon. the Lord Balfour 
 
 of Burleigh, K.T., G.C.M.G. 
 Methodism. By Rev. H. B. Workman, D.Lit. 
 
 EDUCATION 
 
 Life in the Medieval University. By R. S. Rait, M.A. 
 
 LAW 
 
 The Administration of Justice in Criminal Matters (in England 
 and Wales). By G. Glover Alexander, M.A., LL.M. 
 
 BIOLOGY 
 
 The Coming of Evolution. By Prof. J. W. Judd, C.B., F.R.S. 
 Heredity in the Light of Recent Research. By L. Doncaster, 
 
 M.A. 
 
 Primitive Animals. By Geoffrey Smith, M.A. 
 The Individual in the Animal Kingdom. By J. S. Huxley, B.A, 
 Life in the Sea. By James Johnstone, B.Sc. 
 The Migration of Birds. By T. A. Coward. 
 
BIOLOGY (continued) 
 
 Spiders. By C. Warburton, M.A. 
 
 Bees and Wasps. By O. H. Latter, M.A. 
 
 House Flies. By C. G. Hewitt, D.Sc. 
 
 Earthworms and their Allies. By F. E. Beddard, F.R.S. 
 
 The Wanderings of Animals. By H. F. Gadow, F.R.S. 
 
 ANTHROPOLOGY 
 
 The Wanderings of Peoples. By Dr A. C. Haddon, F.R.S. 
 Prehistoric Man. By Dr W. L. H. Duckworth. 
 
 GEOLOGY 
 
 Rocks and their Origins. By Prof. Grenville A. J. Cole. 
 The Work of Rain and Rivers. By T. G. Bonney, Sc.D. 
 The Natural History of Coal. By Dr E. A. Newell Arber. 
 The Natural History of Clay. By Alfred B. Searle. 
 The Origin of Earthquakes. By C. Davison, Sc.D., F.G.S. 
 Submerged Forests. By Clement Reid, F.R.S. 
 
 BOTANY 
 
 Plant-Animals : a Study in Symbiosis. By Prof. F. W. Keeble. 
 Plant-Life on Land. By Prof. F. O. Bower, Sc.D., F.R.S. 
 Links with the Past in the Plant- World. By Prof. A. C. Seward. 
 
 PHYSICS 
 
 The Earth. By Prof. J. H. Poynting, F.R.S. 
 The Atmosphere. By A. J. Berry, M.A. 
 Beyond the Atom. By John Cox, M.A. 
 The Physical Basis of Music. By A. Wood, M.A. 
 
 PSYCHOLOGY 
 
 An Introduction to Experimental Psychology. By Dr C. S. 
 
 Myers. 
 The Psychology of Insanity. By Bernard Hart, M.D. 
 
 INDUSTRIAL AND MECHANICAL SCIENCE 
 
 The Modern Locomotive. By C. Edgar Allen, A.M.I.Mech.E. 
 
 The Modern Warship. By E. L. Attwood. 
 
 Aerial Locomotion. By E. H. Harper, M.A., and Allan E. 
 
 Ferguson, B.Sc. 
 
 Electricity in Locomotion. By A. G. Whyte, B.Sc. 
 Wireless Telegraphy. By Prof. C. L. Fortescue, M.A. 
 The Story of a Loaf of Bread. By Prof. T. B. Wood, M.A. 
 Brewing. By A. Chaston Chapman, F.I.C. 
 
SOME VOLUMES IN PREPARATION 
 HISTORY AND ARCHAEOLOGY 
 
 The Aryans. By Prof. M. Winternitz. 
 
 Ancient India. By Prof. E. J. Rapson, M.A. 
 
 The Peoples of India. By J. D. Anderson, M.A. 
 
 The Balkan Peoples. By J. D. Bourchier. 
 
 Canada of the preseni day. By C. G. Hewitt, D.Sc. 
 
 The Evolution of Japan. By Prof. J. H. Longford. 
 
 The West Indies. By Sir Daniel Morris, K.C.M.G. 
 
 The Royal Navy. By John Leyland. 
 
 Gypsies. By John Sampson. 
 
 A Grammar of Heraldry. By W. H. St John Hope, Litt.D. 
 
 Celtic Art. By Joseph Anderson, LL.D. 
 
 ECONOMICS 
 
 Women's Work. By Miss Constance Smith. 
 
 LITERARY HISTORY 
 
 Early Indian Poetry. By A. A. Macdonell. 
 
 The Book. By H. G. Aldis, M.A. 
 
 Pantomime. By D. L. Murray. 
 
 Folk Song and Dance. By Miss Neal and F. Kidson. 
 
 PHYSICS 
 
 The Natural Sources of Energy. By Prof. A. H. Gibson, D.Sc. 
 
 The Sun. By Prof. R. A. Sampson. 
 
 Rontgen Rays. By Prof. W. H. Bragg, F.R.S. 
 
 BIOLOGY 
 
 The Life-story of Insects. By Prof. G. H. Carpenter. 
 The Flea. By H. Russell. 
 Pearls. By Prof. W. J. Dakin. 
 
 GEOLOGY 
 
 Soil Fertility. By E. J. Russell, D.Sc. 
 Coast Erosion. By Prof. T. J. Jehu. 
 
 INDUSTRIAL AND MECHANICAL SCIENCE 
 
 Coal Mining. By T. C. Cantrill. 
 Leather. By Prof. H. R. Procter. 
 
 Cambridge University Press 
 
 C. F. Clay, Manager 
 
 London : Fetter Lane, E.C. 
 
 Edinburgh: 100, Princes Street 
 
7 DAY USE 
 
 RETURN TO DESK FROM WHICH BORROWED 
 
 PHYSICS LIBRARY 
 
 This publication is due on the LAST DATE 
 stamped below. 
 
 RB 17-60m-12,'57 
 (703slO)4188 
 
 General Library 
 
 University of California 
 
 Berkeley 
 
K6"74I 
 T7 
 
 4<s-p~fr 
 
 UNIVERSITY OF CALIFORNIA LIBRARY