C. K 
 
 ; Libris 
 OGDEN 
 
 SIR DAVID SALOMONS BART 
 
 BROOMHILL , 
 TUNBRIDGE WELLS . 

 
 s 
 
 o 
 w 
 lj 
 
 W 
 
 Q 
 5
 
 LAMP? 
 
 ELE6TRI6 LMTM. 
 
 51 Bourse of ^?oxtr ^ecfures on {Sfecfrtc ^fCtttntnatton 
 "giot;>aC ^nsftfttiton of greaf Britain. 
 
 } > <- .- 
 
 I. A. FLEMING, ALA., D.Sc, F.R.S, 
 
 PROFESSOR OF ELECTRICAL ENGINEERING IN UNIVERSITY COLLEGE, 
 LONDON, 
 
 MEMBER OF THE INSTITUTION OF ELECTRICAL ENGINEERS. 
 
 The Illustrations, Designs, and Text of these Lectures are entered at Stationers' 
 Hall, and are, therefore, copyright. 
 
 [: KLKCTKICIAN" PRINTING & PUBLISHING COMPANY, 
 LIMITED, 
 
 SALISBURY COURT, FLEET STREET.
 
 PREFACE. 
 
 HE following pages are a trans- 
 cript, with some additions, of 
 a course of Four Lectures on 
 "Electric Illumination," deli- 
 vered by the Author to after- 
 noon audiences at the Royal 
 Institution, London. ' They are, 
 however, in no sense presented 
 as an adequate treatment of a 
 subject which has long since 
 outgrown the limits of a lecture, 
 or even of a course of lectures. 
 
 The original aim was merely to offer to a general 
 audience such non-technical explanations of the physical 
 effects and problems concerned in the modern applica- 
 tions of electricity for illuminating purposes as might 
 serve to further an intelligent interest in the subject, 
 and perhaps pave the way for a more serious study of 
 it. The growth of electric lighting is assisted by the 
 diffusion of all accurate (even if simple) information, on 
 
 2000148
 
 ii. PREFACE. 
 
 the mode of production and employment of electric 
 currents for illuminating and other purposes. 
 
 Many of those who were present at the Lectures 
 expressed a desire to be again able to refer to the 
 descriptions and illustrations then given, and at their 
 request, and in the hope that they may be of use to 
 others, the lecture notes have been revised and pub- 
 lished. The aim throughout has been rather to deal 
 with principles than with details, and to give such 
 general and guiding explanations of terms and processes 
 as are essential for a grasp of the outlines of the sub- 
 ject. For this purpose simple drawings have been 
 given of as many of the experiments and apparatus 
 used as possible. For permission to reproduce a set 
 of photographs of views of the interior of rooms 
 artistically illuminated with incandescent lamps, and 
 which are to be found in the second Lecture, the author 
 is indebted to Mr. A. F. Davies, under whose guidance 
 these particular instances of electric lighting were 
 carried out. 
 
 J. A. F. 
 
 University College, London, 
 October, 1894.
 
 CONTENTS. 
 
 (For Index to Contents, see page 225,) 
 
 LECTURE I. 
 
 PAGES 
 
 ELECTBIC MEASUREMENTS ... ... ... ... 1 50 
 
 14 liberations. 
 
 LECTURE II. 
 
 ELECTRIC GLOW LAMPS ... ... ... ... 51 126 
 
 40 Illustrations. 
 
 LECTURE III. 
 
 ELECTRIC ARC LAMPS 
 
 ELECTRIC DISTRIBUTION 
 
 11 Illustrations. 
 
 LECTURE IV. 
 
 28 Illustrations. 
 
 127170 
 
 ... 171222
 
 ERRATA. 
 
 The reader is requested to make the follow in<j corrections: 
 
 Page 15, line 10 from top 
 
 instead of "movement of fixed plates," 
 
 read "movement of suspended plates." 
 
 Page 69, line 4 from bottom 
 
 instead of " If a small piece of wire," 
 
 read " If a small piece of iron wire." 
 
 Page 206, line 8 from bottom 
 
 instead of " 90lb. on the square inch," 
 read " 60lb. on the square inch." 
 
 Page 207, bottom line 
 
 instead of "one of these larger machines," 
 read "one of these smaller machines."
 
 LECTURE I. 
 
 ELECTRIC LIGHTING in Great Britain. A Glance Backward over Fourteen 
 Years. Present Condition of Electric Lighting An Electric Current. 
 Its Chief Properties. Heating Power. Electrical Resistance. 
 Xanies of Electrical Units. Chemical Power of an Electric Current. 
 Hydraulic Analogies. Electric Pressure. Fall of Electric Pressure 
 down a Conductor. Ohm's Law. Joule's Law. Units of Work and 
 Power. The Watt as a Unit of Power. Incandescence of a Platinum 
 Wire. Spectroscopic Examination of a Heated Wire. Visible and 
 Invisible Radiation. Luminous Efficiency. Radiation from Bodies at 
 Various Temperatures. Efficiency of Various Sources of Light. The 
 Glow Lamp and Arc Lamp as Illuminants. Colours and Wave- 
 Lengths of Rays of Light. Similar and Dissimilar Sources of Light. 
 Colour-distinguishing Power. Causes of Colour. Comparison of 
 Brightness and Colour. Principles of Photometry. Limitations due 
 to the Eye.- Luminosity and Candle-power. Standards of Light. 
 Standards of Illumination. The Candle-foot. Comparison of Sun- 
 light and Moonlight. Comparison of Lights. Ritchie's Wedge. 
 Rumfordand Bunsen Photometers. Comparison of Lights of Different 
 Colours. Spectro-photometers. Results of Investigations. 
 
 HOEVER undertakes to describe the re- 
 markable progress during the last two 
 decades of the art of electric illumination 
 must certainly direct attention to three 
 important dates which, in England at any 
 rate, formed turning points in the history 
 of this application of scientific knowledge. 
 ^ In the year 1879 the Government of our 
 
 country had its attention directed for the first time to electric 
 lighting as a possible subject of legislation, and referred the 
 whole matter to a Select Committee of the House of Commons, 
 of which Lord Playfair was appointed Chairman. This Com- 
 mittee met, and took voluminous evidence from numerous and
 
 2 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 various experts, but the broad conclusion reached in the report 
 which it finally presented was generally the expression of 
 opinion that there was no reasonable scientific ground at that 
 date for supposing that domestic electric lighting had obtained 
 a sufficient footing to entitle it to be described as a practical 
 success, and that, therefore, there seemed no useful result to be 
 attained by interfering at once, by special legislation, with 
 electric lighting. Passing over a gap of three years, we find 
 in the year 1882 the whole aspect of affairs entirely changed 
 by the completed invention of the electric glow lamp. At 
 the beginning of that year the first Crystal Palace Electrical 
 Exhibition enabled the public to properly appreciate the 
 extent to which the then perfected electric incandescence 
 lamp had revolutionised artificial lighting, and also the charm 
 and beauty of that illuminant. Schemes were rapidly put 
 forward for the supply of electric current from public 
 generating stations for the purpose of electric lighting, 
 and the enthusiasm which was thus awakened in the public 
 mind led inventors and promoters to prognosticate a very 
 speedy and entire revolution in the art of illumination. As a 
 result, immediate legislative action was taken to control 
 the public supply of electric current. In 1882 a Bill was 
 introduced into the House of Commons entitled "An Act 
 for Facilitating Electric Lighting." We need not pause to 
 enumerate the various causes which interfered with the 
 immediate fulfilment of the sanguine expectations which were 
 formed in 1882 as to the immediate future of electric lighting. 
 Practical difficulties presented themselves the moment that 
 many of the immature schemes which were then launched 
 were attempted to be put into practice. The dynamo machine 
 in some of its forms had hardly emerged from the condition 
 of being a large laboratory or workshop instrument, and 
 mechanical engineers bad not yet brought to bear upon it 
 that knowledge which subsequently enabled them to convert it 
 into a most efficient and trustworthy machine for generating
 
 ELECTEIC MEASUEEMENT. 3 
 
 electric current for public electric supply. Countless details 
 remained to be perfected in the many arrangements for dis- 
 tributing, using, and measuring electric current so produced. 
 From the commercial point of view much information had 
 to be slowly accumulated before even approximately correct 
 opinions could be formed as to the revenue to be gained from 
 the sale of electric energy for domestic purposes, and there 
 was at that date but little information available for enabling an 
 accurate forecast to be made concerning the probable average 
 annual consumption of electrical energy by incandescent lamps 
 when used instead of gas jets in different classes of buildings for 
 illuminating purposes. There is no doubt, however, that the 
 Act of 1882, though much abused at the time, performed the 
 important function of preventing the survival of the unfit. 
 
 Between 1882 and 1888 exceedingly important improvements 
 (some of which it will be necessary later on to examine) were 
 made, and in 1888 the time seemed ripe for a fresh forward 
 movement. This was effected by the pressure brought upon 
 the Legislature to repeal one of the clauses in the Act of 1882, 
 by which revision much more favourable conditions were 
 created for inviting the support of capital ; and as soon as the 
 Electric Lighting Amendment Act of 1888 was an accomplished 
 fact, a very important inquiry was held by the Board of Trade 
 in May, 1889, in the Westminster Town Hall, London, under 
 the chairmanship of Major Marindin. At this inquiry, which 
 lasted for eighteen days, the whole subject of electric 
 lighting by public supply, especially with reference to the 
 needs of London, was carefully debated by many of the 
 leading scientific and legal experts, and, as a result, the 
 Metropolis was divided up into certain areas of electric 
 supply, and conditions were laid down under which the 
 distribution of current might be undertaken either by Public 
 Companies or by the Local Authorities. Up to the date 
 of that inquiry the total amount of electric lighting in London 
 
 B 2
 
 4 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 or in the Provinces had been, comparatively speaking, very 
 small. From and after that date it has advanced by leaps and 
 bounds. The progress made in the use of the incandescent 
 lamp as a means of artificial illumination is shown by the 
 following figures, and graphically indicated in Fig. 1. 
 
 At the end of 1890 there were probably in use about 
 180,000 incandescent lamps in London, but at the end of 1892 
 some 500,000 were being employed, and at the end of 1893 
 
 1.000,000 8-c.p. Lamps Connected 
 
 700,000 
 
 600,000 
 
 500,000 
 
 / 
 
 
 400,000- / 
 
 
 
 300,000 _ri<V1 
 
 / 
 
 / 
 
 
 200,000 
 
 ^ 
 
 
 / 
 
 
 100,000 ^f" x *"^ 
 
 
 <&'' 
 
 
 
 fcf 
 
 
 9?-' 
 
 
 
 1880 1890 1891 1892 1893 
 
 FIG. 1. Diagram showing the Progress of Electric Lighting in London 
 and the Provinces in Five Years. The altitude of the vertical black lines 
 represents to scale the growth, from year to year, in the total number of 
 lamps supplied. 
 
 700,000. In the provinces the grand totals at the end 
 of 1892 and 1893 were probably 147,000 and 425,000. 
 Hence, while in 1888 the total number of incandescent lamps 
 in use in the United Kingdom probably hardly exceeded 
 100,000, even if so many, we find that at the end of 1893 
 there was a grand total of rather more than 1,125 000 electric 
 glow lamps in use in the United Kingdom. There are at the
 
 ELECTRIC MEASUREMENT. 5 
 
 present time (1894) some thirteen or fourteen Companies and 
 Corporations supplying electric current for lighting purposes 
 in London, and these have a total length of more than 
 250 miles of copper mains laid down under the streets. In the 
 Provinces there are between 60 and 70 towns in which a 
 similar public electric supply is given, either by the Local 
 Authority or by a Public Company, and these "undertakers," 
 as they are called in the Act, have a total length of probably 
 more than 400 miles of mains in use at present. Some idea 
 may be formed of the extent to which electric lighting has 
 proceeded in the United Kingdom in the five years which 
 have elapsed since the Board of Trade inquiry was held at 
 the Westminster Town Hall in 1889, by picturing to ourselves 
 an underground electric main or pair of conductors of copper 
 stretching across Great Britain from Penzance to Perth, with 
 an 8-candle-power incandescent lamp placed every yard along 
 this distance. This would roughly represent the total usage 
 of electric lamps and length of electric mains at the present 
 date (1894). 
 
 The 700,000 incandescent electric lamps in London have 
 probably already begun to make their presence felt by the Gas 
 Companies. If these 700,000 glow lamps were replaced by gas 
 jets of equal illuminating power, their annual consumption 
 of gas might at least amount to about 1,000 million cubic feet. 
 The average rate of increase of the metropolitan gas consump- 
 tion between 1886 and 1891 was 947 million cubic feet, the 
 increase for 1891 alone being 1,313 millions. In 1892, how- 
 ever, the year when the London Electric Lighting Companies 
 got fairly to work, this increase fell to 35| millions ; and in 
 1893 the consumption was less by 1,007 million cubic feet. 
 The year 1893 was unusually bright and sunny in the summer 
 and free from fog in winter in London. Hence this cause 
 amongst others may have operated to arrest the growth of gas, 
 but it is clear that a stage has now been reached in which the
 
 6 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 older illuminant will begin to feel the competition of the 
 younger. An industry which has progressed with such rapid 
 strides, and which in the short space of half a decade has made 
 electric lighting in large towns one of the necessary luxuries 
 of life, is not likely to be arrested in its present stage ; and as 
 it has, in its various aspects, not only much scientific and 
 artistic, but also considerable practical interest for every 
 householder, your attention in this and the three succeeding 
 lectures will be directed to the elucidation of facts which 
 ought to be known by every user of the electric light, and the 
 knowledge of which will enable him to understand something 
 of the principles which underlie the art of electric illumination, 
 and to comprehend as well the aid which it can render, when 
 properly applied, to beautify and please. 
 
 It will be necessary to open the whole of our discussion by 
 some simple illustrations of the meaning of fundamental terms. 
 Every science as well as every art has its necessary technical 
 terms, and even if these words at first sound strangely, they are 
 not therefore necessarily difficult to understand. We are all 
 familiar with the fact that electric illumination depends upon 
 the utilisation of something which we call the electric current. 
 Little by little scientific research may open up a pathway 
 towards a fuller understanding of the true nature of an electric 
 current, but at the present moment all that we are able to 
 say is that we know of what it can do, how it is produced, 
 and the manner in which it can properly be measured. Two 
 principal facts connected with it are, that when a conductor, 
 such as a metallic wire or a carbon filament, or any other 
 material which is capable of being employed as a conductor, is 
 traversed by an electric current, heat is generated in the con- 
 ductor, and the space round the conductor becomes capable 
 of influencing a magnetic needle. These facts can be simply 
 illustrated by passing an electric current through an iron wire 
 thus (Fig. 2). You will notice that as the current is gradually
 
 ELECTRIC MEASUREMENT. 7 
 
 increased the iron wire is brought up from a condition in 
 which it is only slightly warm to one in which it becomes 
 visibly red hot in the dark, and finally brilliantly incandescent. 
 At the same time if I explore the region round about the wire 
 with a suspended compass needle, we find that at every point 
 in the neighbourhood of the wire the magnetic needle places 
 itself, or tries to place itself, in a position perpendicular to the 
 wire. This fact, of capital importance, was discovered by 
 
 FIG. 2. An iron wire W is rendered incandescent by an electric current 
 sent through it from a battery B. The magnetic needle N S held near it 
 sets itself across the wire. 
 
 H. C. Oersted in 1820, and in the Latin memoir in which he 
 describes this epoch-making discovery he employs the following 
 striking phrase to express the behaviour of a magnetic needle 
 to the wire conveying the current. He says, " The electric 
 conflict performs circles round the wire ; " and that which he 
 called the electric conflict round the wire, we now in more 
 modern language call the magnetic field embracing the con- 
 ductor. We shall return in a later lecture to this last fact.
 
 8 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 Meanwhile I wish at present to fasten your attention on the 
 heating qualities of an electric current, and the laws of that 
 heat production and radiation. The same electric current 
 produces heat at very different rates in different conductors, 
 and the quality of a body in virtue of which the electric current 
 produces heat in passing through it is called its electric 
 resistance. If the same electric current is passed through con- 
 ducting wires of similar dimensions, but of different materials, 
 it produces in them different quantities of heat in the same 
 time. Before you (Fig. 3) is a chain composed of spirals 
 of iron and copper wire. These wires are each of the same 
 length and of the same diameter. Sending through this com- 
 
 FIG. 3. A chain W composed of alternate spirals of copper C and iron I 
 is traversed by an electric current sent through it by a battery B. The 
 iron links become red hot, the copper links only slightly warm. 
 
 pound chain an electric current, we notice that the iron wire 
 links are very soon brought up to a bright red heat, whilst the 
 copper links, though slightly warm, are not visibly hot. We 
 have, therefore, before us an illustration of the fact that a 
 current heats the conductor, but that each conductor has a 
 specific quality called its electrical resistance, in virtue of which 
 the same strength of current produces heat in it at a rate 
 depending on the nature of the material. Other things being 
 equal, the bodies which are most heated are said to have 
 the highest resistance.
 
 ELEGTRIG MEASUREMENT. 9 
 
 It is now necessary to notice the units in which these two 
 quantities, namely, electric current and electric resistance, are 
 measured. For the sake of distinction, units of electric 
 quantities are named after distinguished men. We follow a 
 similar custom in some respects in common life, as when we 
 speak of a " Gladstone " bag or a " Hansom " cab, and 
 abbreviate these terms into a gladstone and a hansom. 
 Primarily the distinctive words here used are the names of 
 persons, but by application and abbreviation they become the 
 names of things. An electric current is measured in terms of 
 a unit current which is called an ampere, and electrical 
 resistance is measured in terms of a unit which is called an 
 ohm, these being respectively named after two great investi- 
 
 FIG. 4. Glass lantern trough containing two lead plates n p, and a 
 solution of sugar of lead. When a current from a battery B is sent through 
 the cell it deposits the lead in tufts on the negative plate n. 
 
 gators, Andre Marie Ampere and Georg Simon Ohm. In 
 order to understand the mode in which an electric current 
 can thus be denned, we must direct attention to another 
 property of electric currents, namely, their power of decom- 
 posing solutions of metallic salts. You are all familiar with 
 the substance which is called sugar of lead, or, in chemical 
 language, acetate of lead. Placing in a small glass trough 
 a solution of acetate of lead and two lead plates (see Fig. 4), I 
 place the cell in the electric lantern and project the image 
 upon the screen. If an electric current is passed through 
 the solution from one lead plate to the other it decomposes
 
 10 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 the solution of acetate of lead, extricating from the solution 
 molecules of lead and depositing them on one of the lead 
 plates, and you see the tufted crystals of lead being built up 
 in frond-like form on the negative pole in the cell. We might 
 employ, in preference to a solution of acetate of lead, a solution 
 of nitrate of silver, which is the basis of most marking inks, 
 and the same effect would be seen. It was definitely proved 
 by Faraday that we might define the strength of an electric 
 current by the amount of metal which it extricates from the 
 solution of a metallic salt in one second, minute, or hour. The 
 Board of Trade Committee on Electrical Standards have now 
 given a definition of what is to be understood by an electric 
 current of one ampere in the following terms : An electric 
 current of one ampere is a current which will in one hour 
 extricate from a solution of nitrate of silver 4-025 grammes 
 of silver.* Otherwise we might put it in this manner : A 
 current of electricity is said to have a strength of one ampere 
 if, when passed through a solution of nitrate of silver, it 
 decomposes it and deposits on the negative plate one ounce of 
 silver in very nearly seven hours. We are acquainted in the 
 laboratory with currents of electricity so small that they 
 would take 100,000 years of continuous action to deposit 
 one ounce of silver, and we are familiar in electric lighting 
 practice with currents great enough to deposit one hundred- 
 weight of silver in thirty minutes. The simple experiment 
 just shown is the basis of the whole art of electro-plating. 
 Hence, when we speak later of a current of one ampere, or ten 
 amperes you will be able to realise in thought precisely what 
 such a current is able to achieve in chemical decomposition. 
 It may be convenient at this stage to bring to your notice the 
 fact that an 8 candle-power incandescent lamp working at 100 
 volts usually takes a current of about one-third of an ampere, 
 a current which would deposit by electro -plating action one 
 ounce of silver in about twenty- one hours. 
 
 * 28'3495 grammes 1 ounce avoirdupois.
 
 ELECT ETC MEASUREMENT. 
 
 11 
 
 We pass next to consider another important matter, viz., 
 that of electric pressure or potential ; and we shall be helped 
 in grasping this idea by considering the corresponding concep- 
 tion in the case of the flow of fluids. When a fluid such as 
 water flows along a pipe it does so in virtue of the fact that 
 there is a difference of pressure between different points in 
 the pipe, and the water flows in the pipe from the place where 
 the pressure is greatest to the place where the pressure is 
 least. On the table before you is a horizontal pipe (Fig. 5) 
 which is connected with a cistern of water, and which delivers 
 
 FIG. 5. Horizontal pipe P, having six vertical gauge tubes attached to 
 it. The pipe P is in connection with a water cistern C, and when the tap 
 T is shut the water stands up at the same level, A B, in all the tubes. 
 
 water to another receptacle at a lower level. In that pipe are 
 placed a number of vertical glass tubes to enable us to measure 
 the pressure in the pipe at any instant. The pressure at the 
 foot of each gauge glass is exactly measured by the head or 
 elevation of the water in the vertical gauge glass, and at the 
 present moment, when the outlet from the horizontal pipe is 
 closed, you will notice that the water in all the gauge glasses 
 stands up to the same height as the water in the cistern. In 
 other words, the pressure in the pipe is everywhere the same.
 
 12 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 Opening the outlet tap we allow the water to flow out from 
 the pipe, and you will then observe that the water sinks (see 
 Fig. 6) in each gauge glass, and, so far from being now uniform 
 in height, there is seen to be a regular fall in pressure 
 along the pipe, the gauge glass nearest the cistern showing 
 the greatest pressure, the next one less, the next one less 
 still, and so on, the pressure in the horizontal pipe gradually 
 diminishing as we proceed along towards the tap by which 
 the water is flowing out. This fall in pressure along the pipe 
 takes place in every gas and water pipe, and is called the 
 
 FIG. 6. Horizontal pipe P, through which water is flowing from a 
 cistern C to a reservoir R. When the tap T is open the water stands at 
 gradually decreasing heights in the pressure tubes. The dotted line A B 
 shows the hydraulic gradient. 
 
 hydraulic gradient in the pipe. The flow of water takes 
 place in virtue of this gradient of pressure. It will be next 
 necessary to explain to you that there is an exactly similar 
 phenomenon in the case of an electric current in a wire, and 
 that there is a quantity which we may call the electric pres- 
 sure, which diminishes in amount as we proceed along the 
 wire when the current is flowing in it. In order to under- 
 stand the manner in which this electric pressure can be
 
 ELECTRIC MEASUREMENT. 
 
 1., 
 
 measured, a few preliminary experiments will be essential. 
 Every body which is charged with electricity has, in virtue of 
 that charge, a certain electrical potential, or pressure, as it is 
 called, and electricity always tends to flow from places of 
 
 FIG. 7. A small Wim?hurst electrical 
 paper A and B attached to its terminals, 
 the paper strips are drawn together. 
 
 nachine, having two strips of 
 When the machine is worked 
 
 higher to lower potential, just as water or other fluids tend 
 to flow from places of greater to less pressure. When two 
 bodies are at different electric pressures, or potentials, it is 
 found that there is an attraction or stress existing between
 
 14 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 them, and a tendency for them to move, if possible, nearer 
 together. If I attach to the terminals of a small electrical 
 machine two paper strips, and then charge those paper 
 strips to different electric pressures, \ve find the strips 
 
 FIG. 8. Lord Kelvin's Multicellular Electrostatic V T oltmeter. 
 
 The fixed plates or cells are marked A in the sectional drawing. The 
 
 movable plates n are attached to a suspended axis S. 
 
 are drawn together (see Fig. 7). The difference of pressure 
 between these two bodies can be exactly measured by the 
 mechanical force with which they attract one another, or by 
 the force required to keep them apart by a certain distance. 
 This fact is taken advantage of to construct many instruments
 
 ELECTRIC MEASUREMENT. 
 
 15 
 
 which are called electric pressure-measuring instruments, or 
 voltmeters. One of the most valuable of these is the electro- 
 static voltmeter, invented by Lord Kelvin. It consists (see 
 Fig. 8) of a series of fixed plates, which are called cells, and 
 suspended between these are a number of movable plates, all 
 attached to a common axis, this axis being suspended by a 
 very fine wire. The suspended plates are so arranged that, 
 when they are at a different electric pressure or potential from 
 the fixed plates, they are attracted in between them, and the 
 movement of the fixed plates is resisted by the torsional force 
 
 FIG. 9. A battery C, which in the actual experiment consists of 50 cells, 
 sends a current through a wire W W. By means of a pair of contact 
 pieces A B, the terminals of an electrostatic voltmeter V are connected to 
 various points on the wire W W, and the fall in electric pressure along it is 
 thus measured. 
 
 of the suspending wire. The extent to which they have so 
 moved can be measured by an indicating needle fastened to the 
 movable plates. If the fixed and movable plates in this instru- 
 ment are brought to different electric pressures, or different 
 electric potentials, they will be attracted towards one another, 
 and we then have an instrument which can be converted by 
 proper graduation into an electric pressure-measuring instru- 
 ment. Furnished with such an appliance we can now explore 
 the change in pressure down a wire through which the 
 electric current is flowing. Through this manganese-steel 
 wire we are now passing a current of electricity. One ter- 
 minal of the voltmeter, namely, that connected to the fixed
 
 16 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 plates, is kept connected to one end of that manganese-steel 
 wire, whilst the wire connected to the movable plates of the 
 voltmeter can be slid along the manganese-steel wire to 
 different points (see Fig. 9). You will see from the indica- 
 tions of the voltmeter that as I slide the movable contact 
 along the conductor conveying the current the indications of 
 the voltmeter increase, and it is possible thus to show that 
 there is a difference of electric pressure between two points 
 on the wire conveying the current, precisely as there is a 
 difference of fluid pressure between two points in a pipe along 
 which water or gas is flowing. The water flows in the pipe 
 from places where water pressure is greatest to places where 
 water pressure is least. The electric current, whatever it 
 may really be, likewise flows from places where the electric 
 pressure is greatest to places where it is least. A unit of 
 electric pressure has been selected which has been called a 
 volt, after Volta, who, in 1801, gave us the first galvanic 
 battery. It may be remarked in passing that the pressures at 
 which it is most usual to work incandescent lamps are either 
 50 or 100 volts between the terminals of the lamp, and that 
 the pressure between the terminals of a single galvanic cell, 
 such as is used for working electric bells, is from 1 to 1 J volts. 
 
 We are then able to connect these two ideas of electric cur- 
 rent and electric pressure when measured in the units above 
 denned, and to give a definition of the unit in which electric 
 resistance is measured. If we assume that a wire is taken 
 through which a unifonn current of one ampere is passed 
 always in one direction, and two points on this wire are 
 found such that the difference of electric pressure be- 
 tween those points is one volt, then such wire is said to 
 have a resistance of one ohm between those chosen 
 points. The resistance of any conductor can be measured 
 by comparing it by certain methods with that of a conductor 
 whose resistance is one ohm, and the electrical resistance of
 
 ELECTRIC MEASUREMENT. 17 
 
 any wire or conductor can thereby be expressed in units, each 
 of which is called one ohm. The great service which Dr. G. S. 
 Ohm rendered in 1827 to electrical science was that he gave 
 the first clear definition of the manner in which electric cur- 
 rent, electric pressure, and electric resistance are related to 
 one another. This is now embodied in a statement which is 
 called Ohm's law, and which is stated as follows : The 
 strength of the electric current which flows in any conductor 
 when that current has a uniform flow in one direction being 
 measured in amperes, and the difference of the electric pres- 
 sures between the ends of that conductor being given in volts, 
 the electrical resistance of that conductor in ohms is obtained 
 by dividing the last number by the first. Ohm's law is not 
 only a definition of the mode of measuring electrical resistance, 
 but is also a statement of a physical law. It has been experi- 
 mentally proved that the resistance of a conductor, as 
 measured above, does not depend upon the value of the cur- 
 rent, and is the same for large currents as for small ones if 
 a correction is applied for the change of temperature of the 
 conductor which is produced by the current. The resistance 
 of a conductor is generally affected considerably by change 
 of temperature ; for some bodies, such as pure metals, it is 
 increased, and for other bodies, such as carbon and certain 
 metallic alloys, it is decreased. 
 
 Apart, however, from changes of temperature, it is an 
 experimental fact that the strength of the current produced in 
 any given conductor is exactly proportional to the fall in 
 electric pressure down the conductor, and the numerical ratio of 
 the numbers representing the fall of pressure in volts and the 
 strength of the current in amperes is the value of the electrical 
 resistance of that conductor in ohms. In order that you may 
 see the application of this in electric lighting, let us consider 
 the above in connection with an incandescent electric lamp. 
 Such a glow lamp, as at present constructed, consists, as
 
 18 ELECTRIC LAMPJ AXD SLEOTRW 
 
 we shall see in the next lecture, of a fine carbon thread 
 or filament, which, as usually made, is traversed by a 
 current of about two-thirds of an ampere, when that con- 
 ductor is one which is suitable for a 16-candle-power lamp 
 worked at the usual pressure of 100 volts. The electric 
 supply companies bring into our houses two wires, between 
 which they are constantly engaged in keeping an electric pres- 
 sure difference of 100 volts or 110 volts, or thereabouts. If, 
 therefore, the terminals of a lamp are connected to these two 
 supply wires, the ends of the carbon filament are exposed to 
 an electric pressure of 100 volts. The electric resistance of 
 that carbon, when incandescent, is therefore expressed in ohms 
 by dividing the number expressing the pressure difference in 
 volts by the number defining the current in amperes ; hence 
 it is the quotient of 100 by two-thirds, or 150 ohms. 
 
 Another fundamental law in connection with the flow of an 
 electric current in conductors was enunciated by Mr. Joule in 
 1841, and is called Joule's law. It is thus stated: If a cur- 
 rent flows through an electric conductor, the heat produced in 
 that conductor per second is proportional to the product of 
 the square of the current strength as measured in amperes 
 and the resistance of the conductor measured in ohms. Joule 
 deduced this law from elaborately careful experiments made 
 on the quantity of heat produced in a certain wire when tra- 
 versed by an electric current, that wire being immersed in 
 water. His experimental procedure was as follows : He 
 immersed a wire, formed into a spiral, in a vessel of water so 
 protected as not to be able to lose heat from the outside. He 
 then passed measured currents of electricity through the 
 spiral, and observed with delicate thermometers the rise of 
 temperature of the water in a stated time. The whole of the 
 energy which is thus being spent in the wire is converted into 
 heat, and that heat is employed in raising the temperature of 
 the water. If by suitable means we prevent the loss of heat
 
 ELECTRIC MEASUREMENT. 19 
 
 from the containing vessel, or otherwise take it into account, 
 and if we try this experiment with currents of two different 
 strengths, say of one ampere and two amperes, it will be 
 found that, if the resistance of the conductor remains the same, 
 the heat generated in a given time by the current of two 
 amperes will be four times as great as the heat generated in 
 the same time by a current of one ampere, and in like manner 
 a current of three amperes would generate nine times as much 
 heat as a current of one ampere, always provided that the 
 resistance of the wire is not sensibly changed when the cur- 
 rent is altered. A little consideration of the law of Joule 
 and the law of Ohm, when taken together, will show you 
 that, since the total amount of heat produced per second by a 
 current of a given magnitude is proportional to the products 
 of the numbers representing the resistance of the circuit 
 in ohms, and the square of the strength of the current 
 measured in amperes flowing through it ; and, since the 
 product of the value of the resistance of the circuit in 
 ohms and the current strength in amperes is numerically 
 equal to the difference of pressure between the two ends 
 of the conductor, it follows that the total rate at which 
 energy is being expended in any conductor to produce heat 
 when a current of electricity is flowing through it is measured 
 by the numerical product of the strength of that current 
 in amperes, and the pressure difference between the terminals 
 of that conductor measured in volts. If we apply this rule 
 to the case of an electric lamp we find that, in order to 
 measure the total rate at which energy is being transformed 
 into heat and light in an incandescent electric lamp, we 
 have to measure, in the first place, the current passing 
 through it in amperes and the pressure difference between 
 the terminals of the lamp in volts, and the product then gives 
 us, in certain units, which are called u-atts, the rate at which 
 energy is being dissipated or converted into heat in the carbon 
 filament. Thus, for example, if a lamp which takes two- 
 
 c 2
 
 20 ELECTRIC LAMPS AXD ELECTRIC LIGHTING. 
 
 thirds of an ampere is placed upon a circuit having a 
 pressure difference of 100 volts, the product of 100 and two- 
 thirds being- 66, the lamp would be taking 66 watts, and 
 this is the measure of the rate at which energy is being 
 supplied to the lamp, and converted by it into light and heat. 
 
 It is important that you should possess a very clear con- 
 ception of the exact meaning attached to the n-att as a unit of 
 power. If a weight of one pound is lifted one foot high, the 
 amount of exertion or work required to raise this weight is, 
 in engineering language, called one foot pound of work. If 
 550 pounds, or nearly one-quarter of a ton, are so lifted one 
 foot high, the work done is called 550 foot-pounds. Imagine 
 this last exertion made in one second, and repeated every 
 second ; the rate at which icork is being done, or exertion made, 
 will be equal to that which is called one horse- power. An 
 ordinary man might for some time keep on lifting 55 pounds, 
 say the weight of a good-sized full portmanteau, one foot 
 high per second, and he would then be doing one-tenth 
 of a horse-power. The work done in lifting one pound 
 about nine inches high, or more nearly 0-7373 of a foot, 
 has been selected as a unit of work, and is called one 
 joule, and, if this work is repeated every second, this rate 
 of doing icork, or making exertion, is called one watt. It 
 is necessary to notice carefully that a watt is a unit rate of 
 doing work and not a unit of work. If the pound weight is 
 lifted against gravity nine inches high slowly or quickly the 
 work done is the same, viz., one joule. If it is lifted nine 
 inches high in one second this rate of doing work is called one 
 watt. If it is lifted nine inches high in half a second, the rate 
 of doing work is doubled, and is then tn-o u-atts. Hence 
 it will be seen that one watt is a unit of power, which is 
 t ne TTB^h P ar k f a horse-power. The reason for choosing 
 such a small unit is that it gives convenient numbers in which 
 to measure the rate at which work is done in such ener^v
 
 ELECTRIC MEASUREMENT. 21 
 
 transforming agents as incandescent lamps. When a larger 
 unit is necessary, it is usual to employ the kilowatt, which is 
 equal to 1,000 watts, as a unit of power, and it is obvious 
 that a kilowatt is equal to about one and one-third of a horse- 
 power. Accordingly, such a lamp as we have above assumed 
 dissipates energy at a rate which would require one horse-power 
 to maintain eleven or twelve lamps, and the rate at which 
 physical energy has to be supplied to a 16-candle-power glow 
 lamp to keep it at full candle-power is almost as great as the 
 maximum rate at which a strong man can do work. 
 
 Having prepared the way for further explanations by these 
 elementary definitions, we must now proceed to examine more 
 carefully what happens when an electric current is sent 
 through a conductor, and the temperature of that conductor is 
 allowed to rise. Coming back to our initial experiment with 
 the iron wire heated by an electric current, we must note that 
 there are three ways in which the energy supplied to that wire 
 is being dissipated. In the first place, it is dissipated by con- 
 tact with the cold air around it. The air molecules are con- 
 tinually beating against the hot wire, coming up to it cool, 
 taking energy from it to heat themselves, and going away 
 warm, and this process of carrying off of heat by the air mole- 
 cules is called convection. In the next place, the wire loses 
 heat by conduction out at the ends ; the warm wire is held 
 in cool metallic supports, which are so large and such good 
 conductors that they are not sensibly heated by the current. 
 Hence, part of the heat of the wire is carried away by them, 
 and if you examine the red-hot wire you will find that it is 
 cooler at the ends than it is in the middle, being not quite so 
 brilliantly incandescent close up to the clamps as it is in the 
 centre. Then, furthermore, the wire is losing energy by a 
 process called radiation. It is imparting energy to the ether, 
 and sending out waves into this ether which represent an 
 energy transformation, some of these waves being of such a
 
 22 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 character that they can affect the retina of the eye, and con- 
 stitute what we call light ; but by far the larger quantity of 
 that wave energy is not capable of affecting the retina of the 
 eye, and is called the non-luminous radiation. The luminous 
 radiation, as we shall see later, in this case is only about 
 one or two per cent, of the total radiation. Let us follow, then, 
 the changes that take place as any conductor, say a wire or 
 thin carbon rod, is gradually heated to incandescence. Prof. 
 Draper, as far back as 1847, carried out such a series 
 of experiments with a platinum wire, which he heated up 
 gradually by an electric current, measuring at each stage the 
 total amount of energy which was thrown out from the wire 
 in the form of heat and the amount of energy thrown out 
 from the wire in the form of luminous radiation or light. 
 Before you is a table of Draper's results. His figures are in 
 certain arbitrary units. 
 
 Table of Draper's Experimented Results (in ISJfi) on the Incan- 
 descence of Platinum Wire. 
 
 Temp'rature 
 of the Wire in 
 Centigrade 
 degrees. 
 
 The Heat 1 The Light 
 given out by radiated bv 
 the Wire. the Wire. 
 
 Remarks. 
 
 525C 
 
 527C 
 
 practically nil 
 '87 units 
 
 Just visible in the dark. 
 
 590C 
 
 1-10 
 
 
 
 
 Spectrum visible to line E. 
 
 653C 
 
 1-50 
 
 
 
 
 F. 
 
 718C 
 
 180 
 
 
 
 
 "F-G 
 
 782C 
 
 2-80 
 
 
 
 
 " ',', " G. 
 
 910C 
 
 3-70 
 
 
 
 
 Full spectrum visible. 
 
 1038 C C 
 
 6-80 
 
 0-39ui 
 
 its 
 
 
 1100C 
 
 8-60 
 
 0-62 
 
 
 
 
 1L66C 
 1230C 
 
 10-0 
 
 12-5 
 
 173 
 
 292 
 
 
 Spectrum to H. 
 
 1293G 
 
 155 
 
 4-40 
 
 
 
 
 1367C 
 
 7-24 
 
 
 
 
 142FC 
 
 12-34 
 
 
 
 
 The above amounts of heat and light are given in arbitrary units. The 
 letters E, F, G, refer to the fixed lines in the solar spectrum.
 
 ELECTRIC MEASUREMENT. 23 
 
 Let us, however, follow the whole progress of the pheno- 
 mena experimentally hy the employment of an incandescent 
 lamp. Before me on the table is a carbon glow lamp, having 
 a straight filament of carbon as the conductor to be rendered 
 incandescent. By means of a lens we can project an image of 
 the glowing carbon upon a screen, and by means of a prism 
 we can expand that linear optical image into a prismatic spec- 
 trum, or rainbow strip, and notice the gradual changes which 
 occur as the carbon is heated up from its lowest temperature to 
 the highest incandescence it will safely bear. By passing a care- 
 fully graduated current through the lamp, we find out that at 
 a certain point the carbon just begins to be visible in the dark. 
 It has sometimes been inferred, as a deduction from Draper's 
 experiments, and was, indeed, stated by him, that all bodies 
 begin to be visibly red hot in the dark at the same tempera- 
 ture, namely, at about 525C. This, however, is certainly 
 not the fact. It has been shown by Weber, Bottomley, and 
 others, as the result of careful experiments on carbon fila- 
 ments and platinum wires, that bodies with a black, sooty 
 surface have to be brought up to a higher temperature than 
 bodies with a bright metallic surface before they begin to be 
 visible in complete darkness. The crucial experiment has 
 been made by J. T. Bottomley of taking two platinum wires, 
 one of which is made to have a dull, sooty surface by coating 
 it with the finest possible coating of lampblack, the other 
 being highly polished. If these are placed in closed glass 
 tubes from which the air is exhausted, and electric currents 
 passed through the wires so as to heat them, it will be 
 found, on carefully increasing the current and examining the 
 wires in complete darkness, that the wire which has a bright 
 metallic surface becomes visible in the dark at a lower 
 temperature than the wire which has a dull, sooty surface. 
 In Mr. Bottomley's experiments the temperature of the 
 two wires was obtained from their electrical resistance ; 
 this last having been carefully measured at various tempera-
 
 24 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 tares previously. This and other experiments show that 
 it is not strictly true that all bodies become luminous in 
 the dark at the same temperature. But approximately we 
 may say that most bodies, such as metals and carbons, when 
 heated to 600 Centigrade, begin to give out radiation which 
 is capable of affecting the eye as dull red light. Returning 
 to our lamp, let us gradually increase the current through 
 the carbon conductor of this Bernstein lamp, whilst at the 
 same time we project the image of the straight carbor upon 
 the screen, and examine it with a prism. As the current is 
 gradually increased the carbon gives off radiation which is 
 first entirely non- luminous, and which, though not capable of 
 affecting our eyes, can be detected by a very delicate thermo- 
 meter or other suitable instrumental means. As the tem- 
 perature of the wire is increased to a higher point, raciation 
 makes its appearance which is capable of affecting the retina 
 of the eye. It is sometimes stated that the first colour which 
 makes its appearance is red, but careful observation shows that 
 the light which impresses the eye first, and which most 
 easily stimulates the retina, is a f/reyish-f/rem, which occupies 
 in the spectrum the position of maximum brightness. If the 
 temperature of the conductor is still further increased, we see 
 that the prism shows us that red and yellow rays have made 
 their appearance also, and a short spectrum is projected upon 
 the screen in which the red, yellow, and green rays are clearly 
 visible. Increasing still more the temperature of the conductor 
 by passing a stronger current through it, we notice that the 
 spectrum lengthens, greenish-blue, blue, and violet rays 
 successively making their appearance, and are added to the 
 others, whilst the previously existing green and red rays, and 
 especially the green and yellow, are much strengthened. The 
 spectrum, therefore, exhibits a growth, which growth consists 
 in the addition of more and more refrangible rays, or rays 
 towards the violet end of the spectrum, whilst at the same 
 time a gradual increase in the intensity or luminosity of all
 
 ELECTRIC MEASUREMENT. 25 
 
 the rays takes place, so that, when finally we have the com- 
 plete spectrum exhibited on the screen, the conductor is found 
 on examination to be in a state of brilliant incandescence, 
 throwing out white light. We can, therefore, by the assist- 
 ance of the prism, watch the gradual progress in the emission 
 of radiation from the carbon conductor, which is capable of 
 affecting the eyes; but experiments can be made which, at the 
 same time, show that non-luminous or invisible radiation is 
 being sent out from the conductor, and that this, at all stages 
 of the increase in luminosity of the visible spectrum, is also 
 increased in a certain proportion. Broadly speaking, therefore, 
 as we increase the temperature of any body, it first begins by 
 emitting rays which are not capable of affecting the eye, but 
 finally, as the temperature is carried up to a higher and 
 higher point, it emits successively rays which are capable of 
 affecting the optic nerve; and in the end, when a temperature 
 of 1,600 or 1,700C. is reached, we have a total radiation from 
 the conductor, which affects the eye as white light. The 
 annexed table shows approximately the character of the light 
 emitted from bodies when raised to different temperatures : 
 Radiation froin Bodies at different Temperatures. 
 
 'Centigrade 
 
 Bodies begin to be just visible in the dark at about . . . 500 
 
 Dull red heat at 700 
 
 Dull cherry-red heat at 800 
 
 Full cherry-red at 900 
 
 Melting point of silver at ... ... ... ... 945 
 
 Clear red heat at 1,000 
 
 Melting point of gold at 1,045 
 
 White cast-iron melts at... ... ... ... ... 1,040 
 
 Orange-red heat at ... ... ... ... ... 1,100 C 
 
 Bright orange heat at 
 White heat at 
 Steel melts at 
 Bright white heat at 
 Dazzling white heat at 
 
 1,200 
 1,300 
 1,300 
 1,400 
 1,500 
 
 Palladium melts at ! 1,500 
 
 Wrought-iron melts at , 1^(500 
 
 Platinum melts at \ 1,775 
 
 Iridium melts at ... i 1 950
 
 26 ELECTEIC LAMPS AND ELECTRIC LIGHTING. 
 
 Returning again to our incandescent wire, it must be noted 
 that, of the total radiation sent "out by such a hot body, only 
 a small fraction of that radiant energy is capable of affecting 
 the eye, and making its impression upon us as light. By far 
 the larger proportion of radiation from most incandescent 
 bodies is non-luminous radiation, and makes no impression at 
 all upon the organ of sight. The proportion of luminous to 
 total radiation from various sources has been measured by 
 different observers, with the following results : 
 
 Proportion of Luminous to Non-Luminous Radiation in 
 Various Sources of Light. 
 
 Source. 
 
 Luminous 
 Radiation. 
 
 Non-Luminous 
 Radiation. 
 
 Red-hotwire ... 
 
 practically nil. 
 
 100 
 
 per cent. 
 
 Hydrogen flame 
 
 ditto. 
 
 100 
 
 
 Oil flame 
 
 3 per cent. 97 
 
 
 Gas flame 
 
 4 , 96 
 
 t 
 
 White-hotwire 
 
 4-5 95-4 
 
 
 Electric glow lamp ... 3 to 7 , 95 
 
 
 Arc lamp 5 to 15 , 90 
 
 , 
 
 Sunlight 
 
 34 , 66 
 
 
 
 The above table shows us, then, the percentage of luminous 
 to non-luminous radiation in these various sources of light. 
 The luminous efficiency of any illuminant is defined as the 
 fraction, expressed as a percentage, which the luminous 
 radiation is of the total radiation. We may also express, in 
 the unit previously defined, called the watt, the rate at which 
 energy is being expended in any illuminant to produce an 
 illuminating power equal to that of one candle, and coupling 
 this with the known luminous efficiency of the illuminant, 
 we obtain two numbers which precisely express the energy 
 consumption of the agent, and that which may be called its 
 efficiency as a translating device for converting energy from 
 one form, whether electrical or chemical, into another form 
 namely, eye-affecting wave motion in the ether. Below are
 
 ELECTRIC. ME A X UREMENT 
 
 27 
 
 collected together these numbers for various 'sources of 
 light: 
 
 Efficiency of Various Sources of Light. 
 
 Source of Light. 
 
 Total Consumption 
 of Energy in Watts 
 required to produce 
 a light of one Candle. 
 
 Ratio of Luminous to 
 total Radiation, or 
 Luminous Efficiency. 
 
 Candle ' ... 
 
 86 watts 
 
 2 to 3 per cent. 
 
 Oil lamp 
 
 57 
 
 3 
 
 Petroleum lamp 
 
 42-8 
 
 3 
 
 Argand gas lamp 
 
 68 "8 
 
 4 
 
 Electric glow lamp ... 
 
 34 
 
 3 to 7 
 
 Electric arc 
 
 0-8 
 
 5 to 15 
 
 Magnesium wire 
 
 
 15 
 
 Electric discharge in 
 
 
 
 rarefied gases 
 
 
 33 
 
 One great problem awaiting solution in the future is the 
 discovery of a source of artificial illumination which is, 
 relatively speaking, much more efficient than those which are 
 at present available ; that is to say, of a method which will 
 convert a greater proportion of the energy which is transformed 
 into radiation strictly limited to that which can affect the 
 eye ; whether that energy be electrical energy in the form 
 of an electric current, or whether it be chemical potential 
 energy associated with materials to be combined. It appears 
 probable that the solution of this problem rests in the develop- 
 ment of phosphorescence into a practical method for yielding 
 light. By the employment of apparatus of extraordinary 
 delicacy, Prof. S. P. Langley and Mr. Very, in America, have 
 succeeded in determining the luminous efficiency of the light 
 emitted by the Cuban fire-fly. They find that in this natural 
 light the whole energy of radiation is comprised within the 
 limits of the visible spectrum, and, what is more important, 
 chiefly within the limits of the green and greenish-yellow rays, 
 which are especially important for the purposes of vision. These 
 experiments show that the light emitted by this insect is,
 
 28 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 indeed, light without heat, and that its luminous efficiency 
 is not far below 100 per cent. In our artificial production of 
 light we have much lee-way to make up before we can rival 
 the efficiency of the glow-worm and fire-fly. 
 
 Before, however, we can discuss this matter further, it will 
 be necessary to turn our attention a little more at length to 
 the subject of photometry, or the comparison of different 
 sources of illumination in regard to their light-giving quali- 
 ties. This is a part of our subject which, we may say at 
 once, is in a much less satisfactory condition, as regards exact 
 measurement, than the other practical electrical measurements 
 to which reference has been made. The various light-giving 
 bodies which we know and use, such as the sun, candle, gas 
 lamp, incandescent electric lamp, electric arc lamp, &c., are 
 bodies which are at very different temperatures, and they emit 
 light, therefore, of very different composition. Every ray of 
 light which has a pure and simple ray is characterised by two 
 qualities first, its wave length, which determines the colour 
 impression it makes on the organ of vision ; and, secondly, the 
 amplitude of that wave motion, which determines its luminous 
 intensity. 
 
 It may fairly be assumed that my audience is familiar 
 with the fact that all optical research has indicated the fact 
 that what we call light is a disturbance of some kind taking 
 place in a universal medium called the ether. We do not 
 know precisely the nature of that medium or that disturbance, 
 but crucial experiment shows that, when a ray of light is 
 passing through space, some change is being repeated very 
 rapidly at every point in its path, and that this disturbance 
 travels out from any point with a velocity which is, approxi- 
 mately, 186,000 miles a second ; that is, it moves nearly 
 one foot in the thousand-millionth part of a second. This 
 disturbance is a wave motion that is to say, at points in
 
 ELECTRIC MEASUREMENT. 29 
 
 space separated by certain intervals similar motions or actions 
 are taking place in a periodic manner at the same time, and 
 the distance between two points in space at which similar 
 actions or movements are taking place is called a wave length. 
 We are familiar with such wave motion in the case of sound 
 and the surface waves of water. By optical processes the 
 wave lengths of light can be measured, and they have the 
 values given in the table below. The unit of length in 
 which these wave lengths are expressed is the one-hundred- 
 millionth part of a centimetre. 
 
 Tables of Colours and Wave Lengths of Light. 
 
 Colour of the Light. 
 
 Wave Length. 
 
 Red 
 
 Orange-red 
 
 6,800 
 6,550 
 
 Orange 
 Yellow 
 Yellow-green 
 Pure green ... 
 Green-blue 
 Cyan-blue ... 
 Pure blue . . 
 
 5,950 
 5,680 
 5,370 
 5,200 
 4,910 
 4,700 
 4,580 
 
 Violet-blue ... 
 Violet 
 
 4,410 
 4,330 
 
 We may call to mind the fact, that in music the length of 
 the air wave producing any note is exactly half the length of 
 the wave corresponding to the note an octave below. 
 
 It is evident from the above figures that the eye is sensitive 
 to about an octave of colour. It may assist the imagination if 
 it is here mentioned that the wave length of green or greenish- 
 blue light is about l/50,000th part of an inch. This is not 
 by any means an inconceivably small length. Gold leaf may 
 be beaten out so fine that the thickness of a single sheet 
 is only 1/300, OOOfch part of an inch, or something like one- 
 sixth part of the length of a wave length of green light.
 
 30 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 In pure white light we may have rays of all these wave 
 lengths present, together with any or all the intermediate 
 ones whose wave lengths have not heen given. Each different 
 source of light emits in general a great group of rays of 
 many different wave lengths, and each particular ray may 
 be present in varying intensity or brightness. The proportion 
 in which the different rays exist in any source, and the relative 
 intensity which these respective rays have, is not the same 
 for different sources of light. Hence the radiation from any 
 illuminant is in general a very complex thing, and the effect 
 which is produced when it falls on our eyes is a resultant 
 or joint effect due to all the several individual rays that exist 
 in that light. We have in the prism a means of analysing 
 any compound ray by separating out the individual light 
 rays which compose it in a fan-like form, so that we can 
 detect which rays are present and which are absent, and 
 measure also their relative brightness or luminosity. We did 
 this just now when we expanded the linear image of our in- 
 candescent carbon filament into a broad many-coloured band 
 called a spectrum. Two lights are said to be similai' when 
 their spectra can be made to have equal brightness in all 
 their corresponding parts. That is to say, if the yellow ray 
 in one spectrum is made equal in brightness to the yellow in 
 the other, then the green, blue and red in the two spectra are 
 also of equal brightness. A very little investigation shows us, 
 however, that different sources of light are not similar. I 
 throw upon the screen the spectrum of the light given by the 
 carbon of an incandescent lamp, and also throw upon the 
 same screen another spectrum just above the first, which is 
 formed from the light of an electric arc lamp. These spectra 
 are formed by exactly similar prisms placed symmetrically 
 with regard to the screen, and you will notice that not only 
 are similarly coloured rays in the two spectra not equally 
 bright, but you will note the especial richness of the spectrum 
 of the arc light in violet rays. We can so adjust these two
 
 ELECTRIC MEASUREMENT. 31 
 
 spectra that they become exactly equal in brightness or 
 luminosity for any particular ray, say the yellow that is to 
 say, we can so weaken the light of the arc lamp that 
 its spectrum in the yellow is exactly equal in brightness to 
 the spectrum of the incandescent lamp in the yellow. When 
 this is done, we find that the arc lamp spectrum is very 
 much brighter in the violet, green, and blue than that of 
 the incandescent lamp spectrum, but that it is not so 
 bright in the red. It is clear therefore, that these two 
 lights are dissimilar in quality, and before we can speak 
 of the comparison of two lights, we must understand dis- 
 tinctly what we are going tc compare. 
 
 Broadly speaking, the two great purposes for which 
 we require artificial light are to discriminate form and 
 to discriminate colour difference. With regard to the 
 discrimination of form, it is not necessary that the 
 light which is employed should possess rays of more 
 than one colour or wave length. Such a light is 
 called mono-chromatic. If I illuminate this room with 
 light which is wholly yellow or wholly red, by burning 
 metallic sodium in the flame of a spirit lamp, or, in 
 a similar manner, by burning some nitrate of strontium, 
 which is the basis of red fire, we find that we are 
 able to discriminate the form of surrounding objects, 
 but that their normal colour differences have vanished. I 
 need not spend much time in reminding you that what 
 we call the colour of different objects is only a physical 
 difference in their surface or substance which enables them 
 to select certain rays out of a compound bundle of rays 
 falling upon them for reflection, and to absorb the rest. 
 Thus, a ripe cherry is red because, out of the whole col- 
 lection of rays of light falling upon it, it sends back to 
 our eyes by preference a large proportion of the red, yellow, 
 and green rays, but absorbs some violet and blue ; and the
 
 32 ELECTRIC LAMPS AND ELEQTRIC LIGHTING. 
 
 leaves of the cherry tree are green for the reason that 
 they absorb a large proportion of rays other than green, 
 but reflect more copiously these last. This selective re- 
 flection is, however, due to the absorption or diminution 
 in intensity of certain rays in the compound light falling 
 on the body, and which, passing throu/jli the surface of the 
 reflecting body, is internally reflected, and then is returned 
 back through the surface, having in the double transit 
 suffered a deprivation or weakening of certain constituent rays. 
 Hence, the colour of a body is dependent upon the character 
 of the radiation which falls upon it, and if we illuminate a 
 group of coloured bodies, first by light which is purely red, 
 and then by light which is purely green, we find that in 
 these two cases the apparent colour of the bodies is totally 
 different. 
 
 These facts we are able to illustrate before you in a 
 very simple manner by throwing some red light, produced 
 by passing the light from the electric lantern through a pure 
 ruby-red glass, upon a group of coloured ribbons or pieces 
 of coloured paper. We note that the red paper still looks 
 red in the red light, but that the green and blue paper or 
 ribbons look almost perfectly black. On the other hand, 
 when plunged into the green light, the red paper or ribbons 
 now present an almost perfectly black appearance, whilst 
 the green and greenish blue retain their normal colour. 
 In order to understand in what sense sources of light can 
 be compared together, it is necessary to keep in view the 
 fact that every ray of light has three qualities : firstly, what 
 we call its colour ; secondly, its luminosity or brightness ; 
 and thirdly, its purity, or the degree to which that ray is 
 mixed with white or other light. 
 
 When white light, the standard of comparison being day- 
 light, falls upon a coloured surface, be it leaf or flower, some
 
 ELECTRIC MEASUREMENT. 33 
 
 part of the light is reflected back unaltered in quality. Some 
 portion passes through the surface, and is reflected back by 
 internal reflection, and in performing the double journey 
 through the surface layer it has certain of its constituent rays 
 weakened or destroyed. The apparent hue or colour of the 
 object depends upon the nature of this selective absorption. 
 The brightness or luminosity of the surface depends upon the 
 intensity of the light which falls upon it, and upon the 
 amount by which that light which is reflected from it, whether 
 'superficially or internally, is weakened. The eye is sensitive 
 to both these qualities of colour and brightness, and can 
 pronounce judgment upon either of them, or on both. More- 
 over, we appreciate the extent to which the altered light is 
 mixed with the unaltered light. In the case of coloured surfaces 
 which we call pale or light in tint, such as a light pink, or 
 light or Cambridge blue, there is a considerable admixture of 
 white light when these surfaces are seen by normal daylight. 
 Hence, when white light falls upon a surface, some of it is 
 reflected unaltered, and some part is returned to the eye after 
 having suffered a weakening or destruction of some, or all, of 
 the rays present in it. A surface we call white reflects a large 
 fraction of the light which falls upon it unaltered in quality ; 
 a surface we call black reflects a very small fraction, mostly 
 unaltered, of the light which falls upon it ; and a surface 
 we call coloured sends back to our eyes a fraction of the light 
 which falls upon it, but the quality of the light is altered by 
 the diminution in brightness of some, or all, of the various 
 rays present in it. 
 
 Suppose that there are two white surfaces, on each of 
 which pure red light from different sources is being thrown. 
 They may differ in brightness or luminosity, but there is 
 generally no difficulty in deciding which of these two surfaces 
 is the brighter. If, however, we have two surfaces, one of 
 which is reflecting pure blue light and the other pure red
 
 34 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 light, then we have two differences to appreciate, namely, a 
 real difference in colour due to the difference in the wave 
 lengths of the lights reflected by these surfaces, and a differ- 
 ence in brightness or luminosity, which may, or may not, 
 exist, due to the different intensities of the lights reflected 
 from the two surfaces. A certain training of the eye is 
 necessary in order to distinguish a difference in the luminosity 
 of two surfaces, altogether apart from the question whether 
 they have, or have not, a difference of tint. An inexperienced 
 eye cannot do this with the exactitude of a trained eye. It is 
 comparatively easy in extreme cases. If, for instance, I place 
 before you, in a white light, a piece of dark blue paper and a 
 piece of light yellow paper, you have no difficulty, quite apart 
 from the colour difference, in deciding at once that the blue 
 is darker than the yellow, or less bright ; and if, on the other 
 hand, I present to you similar pieces of a light Cambridge blue 
 and a very dark red, you have also no difficulty in deciding as 
 to the relative brightness of those surfaces. The artistic eye 
 is especially trained in this sort of discrimination of bright- 
 ness or luminosity, as distinguished from colour. An artist, 
 for instance, does not admit that any objects are correctly 
 depicted in which the proper differences of luminosity, or 
 brightness, or of light and shade, as he would say, are not 
 properly expressed or suggested. Very often it is quite out 
 of the power of the artist to give the different parts of the 
 surface of his paper or canvas the same actual relative 
 luminosity or brightness as exists in the case of the objects he 
 is depicting. He cannot, for instance, depict on his canvas 
 dark green leaves when seen against a bright background of 
 white cloud or snow in a manner which shall give them the 
 proper relative brightness or luminosity as seen in Nature, 
 especially when that paper or canvas has to be viewed in the 
 interior of a room. And a great part of the art of painting 
 consists in suggesting differences of luminosity which cannot 
 be exactly obtained. But when an artist makes a sketch with
 
 ELECTRIC MEASUREMENT. 35 
 
 monochrome, or crayon, or pencil, he endeavours in some 
 sense to so alter the surface of the paper in its various parts 
 that, without regard to colour differences, these patches of the 
 paper imitate more or less nearly the relative luminosity or 
 brightness of the various parts of the surfaces of the objects 
 to be depicted. 
 
 When, therefore, objects having what we call colour are 
 viewed by the aid of two different illurninants, each sending 
 out light of different qualities, as far as regards tlie colour 
 (Ustinifitishiny powers of those two lights there is no sense in 
 which they can be compared. As regards the power of dis- 
 tinguishing colours, we cannot compare an electric arc lamp 
 with a candle, because no number of candles, however great, 
 are, or can be, the equivalent of any arc lamp in their power 
 of revealing or producing colour differences. We can, how- 
 ever, within certain limits, compare lights in regard to their 
 different powers of producing on a white surface equal 
 luminosity or brightness, whether these lights be monochro- 
 matic that is to say, emit rays of only one wave length 
 or whether they are emitting a compound light composed of 
 rays of many wave lengths. Taking any surface, such as 
 white paper, which is capable of reflecting rays of all wave 
 lengths, at least as far as those which affect the eye are con- 
 cerned, \ve may illuminate part of this paper from one source 
 of light and part from another source, and we may adjust the 
 intensity of these two sources of light in such a way that a 
 discriminating eye can assert that the two parts of the white 
 surface are of equal brightness, whether they are apparently 
 of the same colour or not. There are, however, certain great 
 difficulties in doing this which must not be ignored when 
 comparing together the illumination produced by two lights 
 of different character. 
 
 It is necessary to bear in mind that in making these 
 photometric comparisons, our eye is the only instrument 
 
 D2
 
 30 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 which we can employ, and we are limited in our opera- 
 tions by the physiological properties of that organ. The 
 eye is wonderfully susceptible of education, but there are 
 certain inherent properties of it which no education can 
 overcome or displace. It is often taken for granted that 
 the physiological effect namely, the amount of visual 
 sensation produced by a given light is proportional to the 
 luminous or intrinsic brilliancy of that light, but this is not 
 really the case. If two white surfaces are illuminated respec- 
 tively by pure red and pure blue light, and the lights are ad- 
 justed so that the surfaces are of apparently equal brilliancy, 
 then, if both the sources of light are doubled or trebled in 
 intensity, the relative illuminations on those surfaces will no 
 longer be equal. In other words, the visual sensation does 
 not increase proportionately to the luminous intensity of the 
 source of light. The visual impression produced by violet 
 or blue light increases more slowly than that of red light 
 when the absolute or intrinsic intensities of the two lights 
 are steadily increased by equal steps. As long, however, as 
 we are dealing with lights of fairly similar character, we have 
 a practical basis for comparison in the experimental fact that 
 the illumination produced on a white surface, due to any such 
 source of light, say a standard lamp, placed at any distance 
 from the surface, can be exactly imitated by placing four such 
 equal sources of light at double the distance from the surface ; 
 provided that in both cases the rays of light fall in a similar 
 manner on the surface. 
 
 It is owing to the above-mentioned different physiological 
 action of different coloured lights (a phenomenon discovered 
 by Purkinje) that high authorities such as Helmholtz 
 ("Physiological Optics," p. 420) have stated that any com- 
 parison of lights of different colours is impossible. One fact is 
 perfectly certain, and that is, that there is no way in which we 
 can by any simple number, such as the candle-power, express
 
 ELECTRIC MEASUREMENT. 37 
 
 the relative colour-distinguishing powers of different lights, 
 and any attempt to do so is pure nonsense. Taking daylight 
 from a bright northern sky as our -standard of normal light, we 
 cannot express the degree in which the light from an electric 
 arc or glow lamp can reveal colour differences as seen by such 
 normal daylight by stating the candle-power of those lights, 
 and hence all such expressions as that an electric arc lamp 
 has so many candle-power are insufficient and inaccurate 
 methods of assigning a visual value to the light. We can, 
 however, compare within certain limits the relative brightness 
 of white surfaces when exposed to the two lights which are 
 to be compared. For many purposes for which we require 
 light, such as reading or writing, this gives us a measure of 
 the value of the light for visual purposes. At present, the only 
 scientific basis for photometry is to be found in the approxi- 
 mately correct fact that brightnesses can be added together, and 
 that the illumination or brightness produced 011 a white surface 
 by two separate lights is the sum of the brightness produced 
 by the individual illuminations. The limitations set upon 
 photometry by the physiological properties of the human eye 
 seem, in some aspects, to be insurmountable ; and, since 
 seeing means discriminating colour and brightness difference, 
 no substitute for the eye which is not an eye can be found. 
 Our chief resource is the power we have of training the eye 
 to increased visual sensibility, but a person with a naturally 
 " bad eye " for colour or luminosity difference will never make 
 a good photometrist. 
 
 In order to effect such a comparison of lights we must start, 
 first, with a standard illuminant, and, second, with a standard 
 of illumination. Unfortunately, the legal standard illuminant 
 is agreed on all hands to be an exceedingly unsatisfactory 
 standard. This standard is, by the Metropolis Gas Act of 
 1860, defined to be a standard sperm candle Jin. in diameter, 
 burning 120 grains in the hour, and it is called a standard or
 
 38 ELECTEIC LAMPS AND ELECTEIC LIGHTING. 
 
 parliamentary candle. We will return in a moment to con- 
 sider the various other and much better standards of illumi- 
 nation which have been suggested, but, for the present, let 
 us assume that by the candle we mean a normal standard 
 candle, or, at any rate, the mean of a large number of such 
 standard candles in illuminating power. The illumination 
 which such a candle produces on a white surface, say a sheet 
 of paper, held at a distance of one foot from it, is called one 
 candle-foot, and the candle-foot is the unit of illumination, 
 just as the candle is the unit of illuminating power. The 
 principle on which practical photometry is based is the experi- 
 mental fact that the brightness of the surface of white paper 
 produced by two candles at a distance of one foot from the 
 surface is twice that produced by one candle at a distance of 
 one foot, and, moreover, that the illumination of such a sur- 
 face produced by four candles at a distance of two feet is 
 equal to the illumination produced by one candle at a distance 
 of one foot. Similarly, nine candles at three feet and six- 
 teen candles at four feet distance from the white surface 
 produce the same illumination as does one candle at a distance 
 of one foot. The candle-foot is found to be an exceedingly 
 convenient illumination for the purposes of reading. Most 
 persons can hardly read comfortably if the illumination on 
 their book or paper is less than one candle-foot. The illumina- 
 tion in a well-lighted room may vary from two candle-feet on 
 the tables to half a candle -foot on the walls and a quarter of 
 a candle-foot on the floor, while a brilliant illumination, such 
 as that required on a theatre stage, is from three to four 
 candle-feet. The illumination in a street due to the street 
 gas lamps may be from one-third to one-fourth of a candle- 
 foot ; in a picture gallery, from one to three candle-feet. 
 The illumination due to full, high moonlight in London is only 
 7 y,h to y^th of a candle-foot, depending on the clearness of 
 the sky and altitude of the moon. These fractional figures 
 must be understood as follows : An illumination of a quarter
 
 ELECTEIC MEASUREMENT. C9 
 
 of a candle-foot means illumination which would be produced 
 by a light of one candle held at two feet distance from the 
 white surface, and an illumination of T ^th of a candle-foot is 
 the illumination produced by a candle held at a distance of 
 ten feet from the white surface. The illumination of full sun- 
 light may be from 7,000 to 10,000 candle-feet when the sun 
 is high in a bright clear climate. The mean daylight in the 
 interior of a well-lighted room may be from ten to forty candle- 
 feet. Attempts have been made at various times to estimate 
 the relative brightness of a white surface when held in full 
 sunlight and when held in full moonlight. Various numbers 
 have been obtained by different observers. Bouguer, in 1725, 
 and Bond, in America, in 1851, made estimates of the relative 
 brightness of sunlight and moonlight. Unless the altitudes 
 of the sun and moon and atmospheric conditions are stated, 
 these relative values do not mean very much. Eoughly 
 speaking, full sunlight produces an illumination from 400,000 
 to 700,000 times as great as that of full moonlight. It is 
 difficult to read anything but fairly large type in full moon- 
 light* 
 
 * Bouguer dispersed the rays of the sun falling on a small hole by a 
 concave lens and compared the diffused sunlight with the light of a candle. 
 Bond formed the image of the sun on a sphere of silvered glass, and com- 
 pared it when so weakened by reflection with a standard light. Bouguer 
 found that the intensity of full sunlight was 300,000 times greater than 
 that of full moonlight, and Bond found the ratio to be 470,980 to 1. 
 More recently Prof. Young, in America, has made an estimate of the sun's 
 mean candle-power. After correcting for atmospheric absorption, he finds 
 the candle-power of the sun to be 1,575 x 10 24 . Prof. L. Weber, of 
 Breslau, finds that the red rays in sunlight are 1,110 x 10 24 times more 
 intense than the corresponding rays in caudle light, and the green rays 
 in sunlight 2,294 x 10' 24 times as intense as green candle-light rays. Since 
 the sun's mean distance from the earth is 48 x 10 10 feet, the sun's illumina- 
 
 tion at the earth is A 575 x l ~^ = 7,000 candle feet. Average moonlight 
 
 a candle-foot, and hence full sunlight illumination, on a white 
 surface, is to full moonlight in the ratio of 490,000 to 1. (10 24 stands for 
 a billion times a billion.)
 
 40 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 We turn, then, to consider some of the methods of photo- 
 metric comparison which depend on the principle that one 
 source of light is said to have an illuminating power equal 
 to that of another when the brightness of a white surface held 
 at the same distance from both these sources is the same. 
 Taking, for instance, the normal candle as a standard, any 
 other source of light is said to have four candle-power if it 
 produces the same apparent brightness on a white surface 
 held two feet from it as does a candle if held one foot from 
 the same surface ; such comparison being made, wholly with 
 regard to luminosity or brightness of the white surface, and 
 no attention being given to the difference in colour in any 
 
 W B 
 
 FIG. 10. Comparison of Illuminating Power of Glow Lamp and Candle 
 by means of Ritchie's Wedge. 
 
 white surface when viewed by the two sources of light. The 
 simplest method of effecting this comparison is by the use of 
 Eitchie's wedge. A block of wood is cut in the shape of a 
 wedge with a rather obtuse angle, and has its two adjacent 
 surfaces covered with fine white writing paper, the separating 
 edge being made as sharp as possible (see Fig. 10). The two 
 sources of light which are to be compared say a candle and 
 an incandescent lamp are placed together in a darkened 
 room. The wedge is held between them, with its edge 
 vertical, and on looking at the vertical edge we see that one 
 side of the wedge is illuminated by the candle and the other 
 surface by the electric glow lamp. \Ye can then move the
 
 ELECTRIC MEASUREMENT. 41 
 
 wedge to and fro until we find a position such that the two 
 sides of the wedge respectively illuminated by the two agents 
 appear to be of the same brightness. We are assisted in 
 getting rid of any difficulty due to difference of colour in the 
 light by adopting the principle suggested by Capt. Abney, 
 viz., by oscillating the wedge or swinging it to and fro. 
 When we have very nearly found the position of balance, if 
 we move the wedge to and fro in slowly diminishing arcs, we 
 shall alternately increase and diminish the relative brightness 
 of the two surfaces, but we shall not affect any difference in their 
 relative colour tints ; hence the attention of the eye is by this 
 device fastened upon that which is varying, namely, the relative 
 luminosity or brightness of the two surfaces, and distracted 
 from that to which we wish to give no attention, namely, 
 their relative colour difference. Having found the position in 
 which the two inclined surfaces of the wedge appear equally 
 bright when each respectively is illuminated by one of the two 
 sources, we measure the distance from the candle to the 
 wedge and from the lamp to the wedge. The illumination pro- 
 duced by the candle is then obtained by dividing unity, since 
 the candle is the unit of illuminating power, by the square of 
 its distance in feet from the wedge ; and, likewise, the illu- 
 mination produced by the electric glow lamp is equal to its 
 unknown candle-power divided by the square of its distance 
 in feet from the wedge. Since these illuminations are equal, 
 it follows that the candle-power of the glow lamp must be 
 numerically given by dividing the square of the distance of 
 the glow lamp from the wedge by the square of the distance 
 of the candle from the wedge. Thus, if the balance has been 
 found when the glow lamp is four feet from the wedge and 
 the candle is one foot from the wedge, the candle power of the 
 glow lamp is four times four divided by one times one, or 16. 
 
 Instead of employing a wedge, some observers make use of 
 a method due to Rumford, in which the two illuminants are
 
 42 ELECTRIC LAMPS AND ELECTRIC LIGHTING, 
 
 made to cast the shadow of a stick upon a white screen. 
 If a glow lamp and a candle are placed a short distance 
 apart (see Fig. 11), and a pencil or other rod of wood is 
 held an inch or two from a white card, which is placed 
 so as to be equally illuminated hy the two sources of light, 
 it will be found that there are two shadows on the card. 
 One is a shadow due to the lamp, and the other is a shadow 
 due to the candle. The shadow due to the candle is a 
 space into which candle-light does not shine, and upon 
 which the light from the lamp shines, whilst the shadow 
 due to the lamp is a portion of the surface which is 
 illuminated by candle-light alone. By suitably moving the 
 
 
 FIG. 11. Comparison of Illuminating Power of Glow Lamp and Candle 
 by means of Rumford Shadow Method. 
 
 candle and the lamp we find we can obtain positions in 
 which the two shadows are of equal depth, and we then 
 obtain the candle-power of the incandescent lamp by dividing 
 the square of its distance in feet from the screen by 
 the square of the distance in feet of the candle from the 
 screen. All that we are doing in this case is to compare 
 together the illumination produced upon a white surface 
 which is partly illuminated from one source and partly from 
 another, and the advantage which the rod gives us is that 
 we can make these two shadows or illuminated surfaces just 
 touch one another, and so be assisted in making a comparison 
 between their relative brightnesses. If the same experiment is
 
 ELECTRIC MEASUREMENT. 43 
 
 tried with moonlight and a candle, the observer will very soon 
 realise the difficulties that attend photometric measurement 
 when the lights are of different qualities. On any bright 
 moonlight night hold a white piece of card at the window, 
 and place a candle so as to illuminate the card at about the 
 same angle. Hold a pencil in front of the card, and it will 
 be seen that there are two shadows, one of which is a bright 
 blue and the other a bright yellow. The yellow shadow is, 
 apparently, the shadow thrown by the moon, and the blue 
 shadow is, apparently, the shadow thrown by the candle. In 
 reality, that which we take for the shadow due to the moon is 
 a space illuminated by candle-light alone, and the shadow we 
 take for the shadow produced by the candle is a space on the 
 card illuminated by moonlight alone.* These two surfaces 
 differ not only in colour, but in brightness, because the light 
 falling upon them is very different in quality. By moving 
 the candle we can find a position, generally from seven to ten 
 feet from the card, in which the blue and yellow shadows, 
 though different in colour, have apparently the same depth. 
 But the difficulty of determining when this relative luminosity 
 is the same will give an observer who tries the experiment 
 for the first time an insight into the difficulties of photometric 
 measurement. 
 
 A third and more frequently-used method for the com- 
 parison of sources of illumination is the grease-spot photo- 
 meter of Bun sen. If a piece of thin paper has a spot of 
 grease placed upon it, and if we hold up the paper between 
 our eye and the light, we find the grease- spot more trans- 
 parent than the rest of the paper, and it looks lighter. But 
 if we hold the paper so that the light falls on it, we see that 
 the grease-spot is apparently darker than the rest of the paper. 
 
 * It was this and other similar phenomena which drew Goethe into 
 some confusion of thought on the subject of colour, and led him in his 
 Farlenlchrc to dispute Newton's theory of colour.
 
 44 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 If the disc of paper with a grease-spot upon it (see Fig. 12) is 
 held up between two lights, such as a gas flame and a candle, 
 and looked at, first from one side and then from the other, it 
 will be found possible, by moving the disc, to find a position 
 for the piece of paper such that the grease-spot can hardly be 
 seen from either side. When this is the case the two sur- 
 faces of the paper are equally illuminated by the two sources 
 of light, and if one of these sources of light is a standard 
 candle, then, as above explained, the candle-power of the 
 other source of light is obtained by dividing the square of the 
 distance in feet of the source of light from the screen by the 
 square of the distance in feet of the candle from the screen. 
 
 -: 
 
 FIG. 12. Comparison of Illuminating Power of Glow Lamp and Candle 
 by means of Bunsen Grease-Spot Method. 
 
 In practice various devices are used, such as a pair of inclined 
 mirrors at the back of the disc, and arrangements for rever- 
 sing the disc, to enable the observer to rapidly make a fair 
 comparison between the two surfaces of the disc, and to 
 ascertain if, when equally viewed from both sides, the grease- 
 spot is invisible. The grease-spot method is one which is 
 largely used in the determination of the candle-power of 
 illuminating gas. The Gas Companies are bound to supply 
 gas of a certain candle-power when burnt from a particular 
 standard burner, and official comparisons have to be made in 
 order to determine if they comply with their obligations. If
 
 ELECTRIC MEASUREMENT. 45 
 
 the grease-spot or shadow methods are employed for the com- 
 parison of a caudle with a glow lamp, we do not find any 
 serious difficulties in determining the relative brightness of 
 the white surfaces illuminated by the two sources of light, 
 because the qualities of these two lights are not very diffe- 
 rent that is to say, if we formed a spectrum from each light 
 we should find that the different rays in these spectra, which 
 are of the same wave length, can all be made to be equal in 
 luminosity ; or, in other words, if the luminosity of the red 
 ray in the glow lamp is equal to the luminosity of the red 
 ray of the same wave length in the candle, then the green, 
 blue and violet rays of corresponding wave lengths are 
 also equal as regards luminosity. If the experiment is tried 
 with two lights of very different qualities, such as 
 an electric glow lamp and an electric arc lamp, con- 
 siderable difficulty will be found by the unpractised observer 
 in determining when two adjacent surfaces illuminated by 
 these two sources of light are of equal brightness. In order 
 to diminish this difficulty, observers have sometimes employed 
 coloured glasses to select particular rays from the arc lamp 
 and compare them with similar rays in the candle-light, 
 and numerous determinations exist of the caudle-power of 
 different arc lamps for red and green rays. But this really 
 serves no useful purpose. The practical value of a light is not 
 the relative intensities of certain rays when compared with 
 rays of the same wave length in the light of a candle, but the 
 total brightness which that light is capable of producing on a 
 white surface when compared with the total brightness which 
 the unit illumiuant is capable of producing on the same sur- 
 face. 
 
 We may further illustrate by an experiment the diffi- 
 culty of comparing together the brightness of two surfaces of 
 different colour in the following manner : An electric glow 
 lamp with a red glass bulb and an electric glow lamp with a
 
 4G ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 green glass bulb (see Fig. 13) are so placed as to equally 
 illuminate a white surface. A rod or other opaque 
 body is then placed so as to cast two shadows, one 
 from each lamp. You observe that the two shadows 
 are respectively red and green ; the red lamp appears to 
 cast a green shadow and the green lamp appears to cast 
 a red shadow, and you will find considerably more difficulty 
 in comparing together these two lights by this Runiford 
 method of photometry than would be the case if the two 
 lights were both of them white or both of them red. By 
 adopting the method of oscillating one of the lights so as to 
 
 FIG. 13. Comparison of Illuminating Power of Two Lights of Different 
 Colours by Shadow Method. The two glow lamps have glass bulbs of 
 different colours. 
 
 alternately increase and diminish the brightness of one of the 
 shadows, the difficulty of making this luminosity comparison 
 is partly overcome. 
 
 With regard to the standards of illuminating power, 
 it has been previously stated that the British standard 
 candle is an unsatisfactory standard. The French adopted 
 a standard oil lamp, called a carcel standard lamp, burn- 
 ing 42 grammes of colza oil per hour with a wick of a 
 certain size. This carcel standard gives a light equal to
 
 ELECTRIC MEASUREMENT. 47 
 
 about 9i British candles. A standard which has been adopted 
 to some extent in Germany, and to a less extent in England, 
 is the Hefner-Alteneck arnyl-acetate lamp. This is a little 
 spirit lamp burning pure pear oil from a wick of certain 
 size, and yielding a flame 40 millimetres high, with an 
 illuminating power of rather less than one British candle. 
 But the objection to its use as a standard is that it gives a 
 somewhat reddish light. Another standard largely used in 
 England is the Methven gas standard. In this standard, 
 ordinary coal gas, enriched with benzol or pentane, two 
 hydro -carbon liquids, is burnt at a particular form of 
 argand burner. In front of the flame is placed a metal plate, 
 Laving a slit in it of such a size that it only permits light 
 to pass from the centre of the flame, and this slit is ad- 
 justed so that the light is equal to two standard candles. 
 Variations of quality and pressure in the gas are said not to 
 affect the intensity of the light which is emitted through the 
 slit. In practice this is not found to be exactly the case. By 
 far the most satisfactory standard which has yet been pro- 
 posed is the pentane air-gas standard of Mr. Vernon Harcourt. 
 In this standard a mixture of volatile hydro-carbon (called 
 pentane) and air is burnt at a jet of a certain kind in such a 
 manner as to yield a flame, produced by the combustion of a 
 definite chemical compound, such flame having a certain 
 height and dimensions. By taking suitable precautions, such 
 a pentane gas standard can be made to yield excellent results 
 as a standard of light. It has been proposed by M. Violle, a 
 distinguished French chemist, that an absolute standard of 
 light should be obtained by taking the light emitted from one 
 square centimetre of molten platinum. Such a standard, as 
 a practical standard of reference, is not very convenient or 
 suitable for use as a general primary standard of light, but 
 it may be found valuable as an ultimate reference. For some 
 purposes the only satisfactory method of comparing together 
 two sources of light, and examining their qualities in regard
 
 48 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 to the relative proportions and brightness of the different rays 
 existing in those lights, is by employing the spectro-photo- 
 meter. This instrument consists of an arrangement whereby 
 the same prism is made to analyse the light from two sources 
 of light, and to expand the rays sent out from these two 
 sources of light into two rainbow bands or spectra in such a 
 manner that these spectra are placed one above the other, the 
 corresponding colours being vertically over one another. It 
 is then possible to weaken one of these sources of light by 
 interposing either a disc with an aperture in it of variable 
 size, revolving at a rapid rate, or by employing a pair of nicol 
 prisms, by means of which the light from one source can be 
 weakened to any extent. The brighter of these sources of 
 light is weakened until one particular ray, say the yellow ray, 
 in that light is exactly equal in brightness to the correspond- 
 ing yellow ray in the spectrum of the other source of light. 
 In other words, the spectra are made to match each other 
 exactly at this particular spot. It will then be found that 
 if the two sources of light are of different qualities, say a glow 
 lamp and an arc lamp, the two spectra will not match each 
 other as regards brightness in any other places. 
 
 Prof. Nichols, of Cornell University, U.S.A., has carried 
 out a series of observations with an apparatus of this kind. 
 Taking an Edison 16-candle-power incandescent lamp as his 
 standard, he calls the brightness of the spectrum of this light 
 everywhere unity, and he compares with it the spectrum of 
 any other light, say an arc lamp, of which the spectrum has 
 been weakened so as to make it of identical brightness for one 
 particular yellow ray with the corresponding ray in the 
 spectrum of the glow lamp. It is then found that the 
 spectrum of the arc lamp is less bright in the red rays, but 
 much more bright in the green, and especiaUy much more 
 bright in the violet. In the spectrum of the arc lamp, 
 especially when that spectrum is taken in such a way as to 
 utilise a large portion of the light proceeding from the carbon
 
 ELECTRIC MEASUREMENT. 49 
 
 vapour between the poles, a very bright violet band is found at 
 the extreme end of the spectrum. These observations have 
 been expressed by Prof. Nichols in a series of curves (see 
 Fig. 14). The horizontal line at the base of the diagram 
 represents the spectrums ; the letters A, B, D, E, &c., indicate 
 
 FIG. 1C 
 
 the position of the characteristic black lines or missing rays 
 in the solar spectrum, which are taken as points of reference. 
 The intensity of the spectrum of the glow lamp is taken 
 everywhere as unity, and is represented by the altitude of the 
 horizontal line marked yloic lamp. The other curves then
 
 50 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 represent, by their ordinates or heights at various points, the 
 relative brightness of the respective rays in the different 
 portions of the spectra of the light taken from the sun, the 
 arc lamp, and magnesium wire, compared with the ray of 
 the same wave length in the light from the electric glow lamp. 
 It will be seen how enormously brighter the sunlight and the 
 arc light spectrum are in the neighbourhood of the violet rays 
 when compared with the corresponding ray of the glow lamp, 
 but that their light, relatively speaking, is less bright in the 
 red than the corresponding light of the glow lamp, provided 
 that the spectra of all these sources of light have been 
 equalised so as to make them of identical brightness in the 
 neighbourhood of the D line of the solar spectrum that 
 is, of the yellow rays.
 
 LECTURE II. 
 
 THE Evolution of the Incandescent Lamp. The Nature of the Problem. 
 Carbon the only Solution. Allotropic Forms of Carbon. The Modern 
 Glow Lamp. Processes for the Manufacture of the Filament. Edison- 
 Swan Lamps. The Expansion of Carbon when Heated. Various 
 Forms of Glow Lamps. Focus Lamps. High Candle-power Lamps. 
 Velocity of Molecules of Gases. Kinetic Theory of Gases. Processes 
 for the Production of High Vacua. Necessity for a Vacuum. Mean 
 free path of Gaseous Molecules. Voltage, Current and Candle-power 
 of Lamps. Watts per Candle-power. Characteristic Curves of 
 Lamps. Life of Glow Lamps. Molecular Shadows. Blackening of 
 Glow Lamps. Self-recording Voltmeters. Necessity for Constant 
 Pressure of Supply. Changes Produced in Lamps by Age. Smashing 
 Point. Efficiency of Glow Lamps. Statistics of Age. Variation of Candle- 
 powerwith Varying Voltage. Cost of Incandescent Lighting. Useful Life of 
 Lamps. Importance of Careful "Wiring." 
 Average Energy Consumption of Lamps 
 in various Places. Load Factors. Methods 
 of Glow-lamp Illumination for Production 
 of Best Effects. Artistic Electric Lighting. 
 Molecular Physics of the Glow-Lamp. 
 
 HE illustrations and explanations 
 in the previous Lecture will have 
 prepared the way for a study of the 
 electric incandescent lamp as a source 
 of illumination. We shall not occupy 
 time by discussing the stages by 
 which the glow lamp has attained 
 its present perfection. Neither is 
 it necessary to dwell upon questions 
 of scientific or legal priority which are not most advantageously 
 discussed from the lecture table. Suffice it to say that in its 
 modern form the electric incandescent lamp always has been, 
 and always will be, inseparably associated with the names of 
 
 K2
 
 52 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 Mr. Thomas Alva Edison and Mr. Joseph Wilson Swan, and 
 in a lesser, though by no means negligible degree, with 
 those of Mr. Lane-Fox and Messrs. Sawyer and Man. 
 These inventors were preceded by many who attempted 
 solutions of the problem, and they have been followed by 
 others who have added countless details of invention to the 
 methods and means by which progress has been made from 
 early ideas to a perfection which is even now not yet final. 
 These successful workers, however, started from a standpoint 
 similar to that which initiated the ideas and failures of early 
 pioneers in this region of discovery and invention. The know- 
 ledge of the simple fact that a metallic wire or other conductor 
 could be heated by an electric current, and that such a heated 
 body when raised to a proper temperature gave out rays of 
 light, was the starting point for all the attempts made by early 
 and by later inventors to discover a means of producing a 
 conductor which would fulfil the following conditions : (1) 
 It must be capable of being heated to a high temperature 
 without fusion or sensible volatilisation ; (2) it must be a 
 material of high specific resistance ; and (3) it must be 
 capable of being shaped into such a form as to be conveniently 
 and economically utilised as a means of producing small units 
 of light. 
 
 The fundamental fact on which the modern glow lamp is 
 based is that carbon can be heated in a highly perfect vacuum 
 or in rarefied gases to a temperature near, or higher than, 
 the melting point of platinum without suffering very rapid 
 change, and that in certain forms it fulfils the three con- 
 ditions named above, as necessary for the incandescing 
 material in a glow lamp. Carbon is capable of existing in 
 three modifications, which are chemically termed aUotropic 
 forms. These are : charcoal as ordinarily obtained by the 
 carbonisation of some organic substance, such as wood, paper, 
 thread, silk, and other similar substances, which consist
 
 ELECTEIC GLOW LAMPS. 53 
 
 principally of carbon united with other elements, which 
 can be driven off by heat. Next, carbon exists in the form 
 known as rp-aphite, and we have it in this form in plumbago 
 as used in lead pencils, and in the hard carbon deposit in gas 
 retorts. And, lastly, carbon occurs in a transparent or opaque 
 crystalline form known as diamond. All forms of carbon are 
 by a sufficiently high temperature converted into graphite. 
 Of these three forms of carbon, diamond is practically a non- 
 conductor of electricity ; ordinary charcoal or carbon, such as 
 that produced by carbonising wood, cotton, or other organic 
 materials, has an electrical conductivity which varies greatly 
 with the circumstances under which it is produced; but, ap- 
 proximately speaking, it may be said that the hard and 
 dense form of carbon which is obtained from gas retorts as a 
 product of the distillation of coal, and which is very largely 
 composed of graphitic carbon, has an electrical resistance 
 which is, for the same volume, from one to three thousand 
 times that of copper. The temperature of fusion and tem- 
 perature of volatilisation of carbon is also very high. If car- 
 bon is heated to a temperature near to red heat in ordinary 
 air, or in an atmosphere containing oxygen, an immediate 
 combustion of the carbon takes place, produced by the union 
 of the carbon with the oxygen of the surrounding atmosphere 
 and the resulting production of carbonic acid gas. Hence, in 
 order that permanence may be obtained, it is essential that the 
 heating of the carbon should take place either in a vacuous 
 space or else in an atmosphere which does not contain free 
 oxygen. The practical invention of the incandescent lamp, 
 therefore, turned upon the discovery of the proper method for 
 producing carbon in the form of a thin wire, thread, or 
 filament, as it is called, and supporting this carbon filament 
 in a glass bulb quite free from all oxidising gases or air. 
 
 The final outcome of failure and research was ultimately 
 to show that in order to obtain light by the incandescence of
 
 54 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 a conductor, such incandescence being produced electrically, 
 the following conditions must be met : Firstly, the substance 
 to be rendered incandescent must be carbon. All attempts to 
 employ metallic wires, such as platinum and platinum-iridium 
 alloys and they have been many have proved unsuccessful. 
 Up to the present time no really practical substitute has been 
 found for carbon, although many have been described and 
 claimed. Secondly, the carbon must be enclosed in a bulb 
 which is exhausted of its air as perfectly as possible. It has 
 been frequently asserted by different inventors that carbon is 
 equally or more permanent when contained in glass bulbs 
 filled with rarefied gases, such as nitrogen or hydrogen, 
 chlorine, bromine, or hydrocarbon gases ; but it has not yet 
 been definitely shown that these methods are as efficient as 
 the employment of a very high vacuum. Thirdly, the enclos- 
 ing bulb must be entirely of glass. Fourthly, the electric 
 current must be got into and out of the conductor by means 
 of platinum wires sealed through the glass. Without stop- 
 ping to detail the attempts which have been made, or which 
 are still being made to modify these conditions, or to con- 
 struct lamps to give light by incandescence which do not 
 fulfil them, suffice it to say that the modern electric glow 
 lamps in general use all comprise the above four elements : 
 carbon, vacuum, platinum, glass. Let us consider each of 
 them in turn. 
 
 The carbon conductor or filament is made by carbonising 
 some material which must be capable of being reduced to 
 nearly pure carbon after it has been given the necessary 
 filamentary form. This carbonisation is conducted by heat- 
 ing a suitable material reduced to a thread-like form to a very 
 high temperature in a closed crucible. If a piece of paper, 
 linen, cotton or thread, which consists essentially of cellulose 
 (the chemical constitution of which is carbon united to oxygen 
 and hydrogen), is raised to a high temperature, say a white
 
 ELECTRIC GLOW LAMPS. 55 
 
 heat, in a closed crucible made of very infusible material, 
 such as compressed blacklead or plumbago, the hydrogen 
 and oxygen are driven out from their combination with 
 the carbon, and a residue is obtained which is more or 
 less pure carbon. The temperature at which this carbonisa- 
 tion is conducted has a great effect upon the kind of carbon 
 produced. Carbonisation at a very high temperature, near the 
 melting point of steel, when conducted out of contact with air, 
 produces a highly dense and elastic form of carbon. Some- 
 times the material employed for this purpose is cotton thread 
 of a pure kind which has been parchmentised by being treated 
 in a bath of sulphuric acid, commonly known as oil of vitriol, 
 and water. This bath destroys the fibrous structure of the 
 thread, and produces a material somewhat resembling catgut. 
 Parchmentised paper is not infrequently used as a material for 
 the covers of jam-pots. 
 
 The dilute sulphuric acid with which the thread is treated 
 has the power of producing a change in the cellulose, which 
 results in the formation of the above tough material. The 
 parchmentised cotton thread is cut into the necessary lengths, 
 wound on a frame made of carbon rods, and buried in a 
 crucible which is packed full of powdered plumbago. The 
 closed crucible is then exposed to a vivid white heat in a 
 furnace, and after a sufficient exposure to this temperature 
 the crucible is opened and the carbonised thread is taken put. 
 It is then found that the threads have been converted into a 
 material which is almost wholly pure carbon of a dense variety 
 and highly elastic. In place of using parchmentised thread, 
 as first suggested by Mr. J. W. Swan, other means have been 
 adopted for preparing the cellulose in the form of a fine 
 structureless thread. Pure cotton wool, which is nearly pure 
 cellulose, is capable of being dissolved by several chemical 
 materials, such as chloride of zinc, and the gummy mass so 
 produced can be pressed out into a thread. This thread is
 
 56 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 then treated as above described, and converted into carbonised 
 thread or carbon filaments of the necessary form. When so 
 prepared the carbon thread is an elastic material, having an 
 electrical resistance which is approximately two thousand 
 five hundred times that of a copper wire of the same 
 length and thickness. An immense variety of materials have 
 been used and suggested for producing the carbon filament. 
 At one time Edison used largely very thin slips of Japanese 
 bamboo cut into the requisite shape. We will project upon the 
 screen the image of an Edison lamp, of which the carbon fila- 
 ment is formed of carbonised bamboo and has been broken on 
 
 FIG. 15. Modern Electric Glow Lamp (Edison-Swan), with Looped 
 Carbon Filament. 
 
 one side, and set it in vibration ; you will see by the rapid 
 movement of the carbon loop when the lamp is shaken 
 how exceedingly elastic is the delicate carbon filament. 
 Spiral springs can be prepared in this manner from carbonised 
 thread which have a considerable degree of elasticity and 
 tenacity. 
 
 The carbon thread, having been thus prepared, has next 
 to be mounted in a glass bulb, and it must be so held that
 
 ELECTRIC GLOW LAMPS. 57 
 
 when heated the expansion which it experiences can be 
 permitted to take place without endangering its rupture 
 or severance from the wires to which its ends are fastened. 
 Early experimentalists found an insuperable difficulty in 
 connecting their carbons to the leading-in wires, and in 
 constructing any practical lamp by the use of straight 
 carbon rods. In most modern lamps the carbon conductor 
 is given a horse-shoe or double-loop form, as shown in Fig. 15, 
 in order that it may have perfect liberty to expand when 
 heated without becoming detached from, or from putting any 
 strain upon, the supporting wires by which it is held. It is 
 very easy to show that carbon, like most other solid bodies, 
 expands when heated. I have here an Edison-Swan incandes- 
 cent lamp with a long straight carbon filament (see Fig. 16). 
 One end of the carbon filament is attached to a platinum wire 
 
 FIG. 16. Electric Glow Lamp with Straight Carbon Filament for showing 
 the Expansion of the Carbon. 
 
 sealed through the glass at the top, and the other end is 
 attached to a platinum wire sealed through the other end of 
 the glass, but a little spiral spring of steel is inserted between 
 the leading-in wire and the carbon. Projecting an image of 
 this upon the screen, and asking you to fix your attention 
 upon the end of the carbon attached to the spiral spring, it 
 will be seen that when the electric current is sent through the 
 carbon filament, making it incandescent, the spring contracts, 
 thus showing the expansion of the carbon thread. If the 
 carbon is rigid and is not given a loop or horse-shoe form, 
 then it is necessary to make provision for this expansion of 
 the carbon by attaching it to flexible supports or springs, as 
 was done in the Bernstein lamp (see Fig. 17). The carbon
 
 58 ELEGTEIC LAMPS AND ELECTEIC LIGHTING. 
 
 filament, before inclusion in the bulb, is frequently subjected 
 to a process which is called treating. If the oarbon filament 
 is rendered incandescent in an atmosphere of hydro -carbon 
 gas if, for instance, it is rendered incandescent when it is 
 immersed in coal gas or benzol vapour the heated carbon 
 decomposes the hydro-carbon gas, and a deposit of carbon of 
 a very dense form is made upon the filament, the carbon 
 
 Fid. 17. Bernstein Lamp, with Straight Carbon C, and Springs S to 
 permit Expansion. 
 
 which is so deposited being of a graphitic form, and having, 
 under the proper conditions of deposit, a brilliant steel-like 
 lustre. This process is not, however, applied to all carbon 
 filaments. This deposited carbon has a lower electrical resist- 
 ance for the same volume, viz., about one-sixth or seventh,
 
 ELECTRIC GLOW LAMPS. 59 
 
 when compared with the carbonised cellulose of which the 
 carbonised parchmentised thread is composed. In some cases 
 the original carbonisation of the material is performed in 
 an electric furnace at a very high temperature and under 
 pressure ; under these conditions the organic material which 
 is carbonised is converted into an exceedingly dense variety 
 of carbon, called adamantine carbon, the surface of which has 
 a smooth and polished appearance and a steel-like lustre with- 
 out having been subjected to the above-described process of 
 treating. However prepared, the carbon filament has then to 
 be attached to the leading-in wires, which are sealed through the 
 glass. These, as above stated, are made of platinum, and no 
 really successful substitute has yet been found for this. The 
 reason for this selection is as follows : In order that the 
 wires by which the current is conveyed to the carbon filament 
 may be sealed into the glass airtight, and not shrink away 
 from the glass when cold, it is necessary that the material of 
 which the wires are composed shall have practically the same 
 expansion as the glass, because otherwise the wires would 
 crack out when the glass bulb was heated and cooled, 
 and the air from outside would leak into the vacuum. It 
 happens, by a fortunate coincidence, that the co -efficient 
 of expansion of platinum is exactly the same as that of some 
 varieties of glass, and platinum wires can therefore be sealed 
 through the glass whilst hot, and will remain firmly attached 
 to the glass when cold. The carbon filament is fastened to 
 the ends of the platinum wires by means of a carbon cement, 
 or a deposit of carbon made over the junction, and the carbon 
 filament with its attached platinum wires is then sealed into 
 a glass bulb. We are thus able to pass the necessary electric 
 current through the filament, and yet preserve the carbon in a 
 highly perfect vacuum. 
 
 In Fig. 18 are shown two lamps with carbon conductors 
 of different forms. In some cases the carbon filament has
 
 60 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 a simple horse-shoe shape as originally adopted by Edison ; 
 whilst in other cases it is given a double twist, as in the 
 Edison-Swan lamps, one or more curls being used. For 
 some purposes it is necessary to coil the filament into a tight 
 compact coil, in order that, when the lamp is used in the focus 
 of a mirror, the light may be all concentrated in one place, and 
 
 A 
 
 FIG. 18. Glow Lamps with Zigzag and Straight Carbons. 
 
 be gathered up by the mirror or lens. Such lamps are called 
 focus lamps, and are employed in optical lanterns, in ships' 
 side-lights, and wherever it is necessary to collect all the rays 
 from the filament by optical means. These lamps are very 
 useful in optical lanterns, and, now that so many houses are
 
 ELECTEIG GLOW LAMPS. 
 
 01 
 
 supplied with electric current, afford an easy means of show- 
 ing lantern views in a drawing-room.* In cases in which 
 high candle-power is required, it is obtained by putting two or 
 more filaments in the same lamp bulb. In Fig. 19 is shown 
 a representation of a multiple-filament lamp of 2,000 candle- 
 
 FIG. 19. 2,000 candle-power Glow Lamp with Multiple Filaments. 
 
 * Photographic amateurs who practise the art of making lantern slides 
 from negatives, want an easy and ready mode of exhibiting them. If 
 electric current is laid on to the house, the simplest method of doing this 
 is to fit an ordinary optical lantern with a 100-volt focus lamp of 50 
 candle-power. This can be fitted on the tray which usually carries the 
 oxyhydrogeu jet. A screen of drawing paper six feet square makes an 
 admirable surface for projection, and the views can be shown in a drawing- 
 room without danger or smell.
 
 62 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 power as constructed by the Edison-Swan Company. In these 
 lamps a number of carbon conductors of the horse-shoe shape 
 are arranged between two rings, so that the electric current 
 can pass from one ring to the other through four, or five, or 
 six, or more carbon conductors, each of which has one leg 
 attached to one ring and the other leg to the other. Lamps 
 of this high candle-power are made up to 3,000 candle- 
 power or more. In such cases the glass bulbs become globes 
 of glass nearly a foot in diameter, and have to be made of 
 considerable thickness to withstand the enormous pressure of 
 the air on the outside. The large lamp now shown to you in 
 action has a globe having a surface of 280 square inches, and 
 the air pressure on the exterior amounts to one and three- 
 quarter tons ! One limitation to the safe size of such lamps 
 is this great external air pressure 
 
 The final stage of manufacture of the lamp is the exhaus- 
 tion of the lamp bulb. It was mentioned above that this 
 was rendered essential, in the first place, by the fact that, if 
 heated in the air to a high temperature, the carbon filament 
 would soon be destroyed by combustion. Taking a non- 
 exhausted lamp, I pass an electric current through the 
 filament. You see that the lamp burns for a few seconds 
 and then goes out. The oxygen of the air in the bulb has 
 combined with the carbon of the filament in one or more 
 places and destroyed it by producing combustion of the 
 material. But if we take a filament of carbon not enclosed 
 in a bulb and plunge it into a glass vessel full of a non- 
 oxidising atmosphere, such as carbonic acid gas or coal gas, 
 and heat it electrically, you will see that the carbon, though 
 rendered incandescent, is not then rapidly destroyed. The 
 carbon filament is enclosed in a vacuum, not merely to pre- 
 vent its combustion, but there is another important reason 
 for exhausting the bulb of all gases or air. Even if we suppose 
 the bulb filled with a non-oxidising atmosphere of nitrogen or
 
 ELECTRIC GLOW LAMPS. G3 
 
 hydrogen, such a lamp would yet not be so perfect as a vacuum 
 lamp, and for the following reason : Modern physical 
 research has provided arguments well nigh irresistible to prove 
 that gases consist of molecules or little particles which are 
 in rapid motion. This kinetic theory of the structure of gases, 
 as it is termed, is supported by an immense body of physical 
 evidence, and no less an authority than Lord Kelvin has called 
 it one of the surest articles of the scientific creed. Certain 
 lines of investigation have shown how to determine the average 
 velocity of these gas particles. It has been shown, for 
 instance, that in hydrogen gas at ordinary pressure and at the 
 temperature of melting ice the molecules are moving with an 
 average velocity of about 6,000 feet per second, or 69 miles 
 per minute, and in oxygen with an average velocity of 18 miles 
 per minute. These gas molecules in their rapid flight collide 
 against one another in moving to and fro, and the average 
 space over which the molecule flies before it has a collision 
 against some other molecule is called its mean free path. 
 
 In ten cubic inches of air or oxygen at the ordinary 
 pressure, and at the temperature of melting ice, there are, 
 in all probability, a number of gas molecules represented 
 numerically by a billion times a billion, or by unity followed 
 by twenty-four cyphers (10 24 ). The mean free path of the 
 oxygen gas molecule is rather more than two-millionths of an 
 inch, and the average number of collisions during the one 
 second in which it darts over a distance in all of 1,500ft. is 
 7,600 millions. We are here dealing with minute portions 
 of time and space, in which the unit is not a second and 
 an inch, but a millionth part of each of these. The 
 millionth of an inch is about the same fraction of an inch 
 that one foot is of 200 miles ; and the millionth part of a 
 second is the same fraction of a second that one second is of 
 about twelve days. All the air we breathe and feel consists, 
 therefore, of these flying and colliding oxygen and nitrogen
 
 64 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 molecules. Each particle or molecule is always having its 
 direction changed by its collisions against neighbours, just 
 as a billiard ball has when it "cannons" against another. 
 It is the constant bombardment of these molecules against 
 the walls of a containing vessel which creates the elastic 
 pressure of a gas, and when any solid body is surrounded 
 by a gas, an enormous number of collisions take place 
 against that surface in every second of time. 
 
 If the surface is heated, the molecules which come up 
 against it take heat from it in their myriads of contacts per 
 second, and this removal of heat from the body by the gas 
 molecules is called convection. If the carbon filament is 
 enclosed in a glass bulb not exhausted of its air, a certain 
 amount of energy is thus removed from the filament and 
 given up to the glass, being conveyed by these molecular 
 vehicles. The energy so abstracted from the filament is not 
 in any way conveyed to the eye as light. Hence, for all 
 illuminating purposes, it is lost, and a lamp so constructed 
 with a non-vacuous bulb cannot in some respects be as 
 efficient a device as one with a high vacuum, because of this 
 additional loss of energy from the filament. 
 
 The process adopted for removing the air from the bulb is 
 generally the employment of some form of mercury pump in 
 combination with a mechanical air-pump. Time will not per- 
 mit me to describe the many forms of this device which have 
 been suggested. Briefly speaking, the principle upon which 
 many of these mercury pumps act is as follows : A vertical 
 glass tube F (see Pig. 20) has a side tube S sealed into it, to 
 which is attached the lamp to be exhausted. Drops of mercury 
 are allowed to fall down the vertical tube, and as they fall down 
 they push the air in the form of bubbles before them. The 
 air is, therefore, gradually drawn out from the lamp bulb L, 
 and the completion of the process of exhaustion is recognised
 
 ELECTRIC GLOW LAMPS. 
 
 65 
 
 when the globules of mercury fall down the vertical tube in 
 close contact with one another without carrying any air 
 between them. Under these circumstances, if the bottom 
 end of the vertical fall tube is placed in a vessel of mercury V, 
 the mercury will gradually rise in the fall tube to the height 
 
 FIG. 20. Diagram of arrangements of a Mercury Pump for Exhausting 
 Lamps. 
 
 M Mercury Reservoir ; F, Fall Tube ; L, Lamp being Exhausted ; 
 S, Side Tube ; V Vessel of Mercury. 
 
 of the mercury barometer that is, to about 30in. and then 
 each succeeding drop of mercury which falls down on to the 
 top of the column, does so with a sharp click, thus showing
 
 66 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 that the lamp bulb has been exhausted of its air to a very 
 high degree. It is not difficult to abstract from such a glass 
 lamp bulb all but one-millionth of the air originally contained 
 in it. A highly exhausted space of this kind is commonly 
 called a vacuum ; nevertheless, when we realise that every 
 ten cubic inches of air at ordinary pressure contains a billion 
 times a billion molecules of air, it is easily seen that the 
 reduction of this enormous number, which may be contained 
 in a lamp bulb of about three inches in diameter before 
 exhaustion, to one-millionth of their original value, still 
 leaves included in the bulb a very respectable number of 
 molecules, viz., a million times a billion. One important 
 change that is, however, made in the physical state of 
 the gas when so reduced to a very low pressure is that the 
 mean free path of the molecules is enormously increased. By 
 this reduction in pressure, the mean free path of the molecules 
 in space, on an average, traversed before collision, is increased 
 just in proportion as the pressure is diminished. 
 
 By the reduction of the pressure to one-millionth of its 
 normal value by the abstraction of 999,999 millionths of 
 the air, the mean free path of the air molecules is increased 
 a million-fold in other words, t3 something not far from 
 two inches. In the interior, therefore, of a large glow 
 lamp, which may have a volume, say, of ten cubic inches, 
 there is good reason to believe that, when the bulb is 
 exhausted to the highest point usually obtainable in practice, 
 there still remain an immense number of molecules of air, 
 but that the space is, comparatively speaking, so sparsely 
 populated with molecules that each molecule in flying to and 
 fro, may move, on an average, over a distance of an inch or two 
 without collision against a neighbour. The rarefaction of 
 the air must, therefore, decrease immensely the rate at which 
 energy is taken from the filament by convection, because it 
 decreases the number of molecular bombardments against the
 
 ELECTRIC GLOW LAMPS. 
 
 67 
 
 filament, and this is one object of producing the exhaustion. 
 When the proper vacuum has been obtained in the lamp, the 
 glass inlet tube is melted, and the lamp is sealed off from the 
 pump. The lamp is finished by adding to it a brass collar 
 with two sole-plates of brass, which are fastened on to the 
 ends of the platinum wires which protrude through the glass. 
 In this way an electric current can be sent through the 
 filament, and yet the highly perfect vacuum in the glass 
 bulb be indefinitely preserved. 
 
 Lamps so made may take numerous forms. In Fig. 21 
 is shown a form of lamp called a candle-lamp, employed 
 
 FIG. 21. Electric Candle Lamp, of which the Glass Bulb is shaped like a 
 Candle Flame, 
 
 for decorative and other purposes. Lamps may be made 
 so small that they can be employed for theatrical, decorative, 
 microscopic, dental, and surgical purposes, in which case they 
 are called micro-lamps ; they may be made so large, and 
 have so many carbon conductors in them, that they can yield 
 a candle-power of two to three thousand, and the glass bulbs 
 may, in both cases, be given any requisite form and colour. 
 
 Having thus briefly described the process of the construction 
 of a glow lamp, we have now to study a little more closely 
 
 F2
 
 68 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 the laws which govern its operation. From what has been 
 already stated in the First Lecture, it will be seen that, in 
 order to possess a full knowledge of the behaviour of the 
 lamp as an energy-translating device, it is necessary to 
 know four things about it. First, we must know the 
 current, in amperes, which is taken by the lamp when a 
 certain electric pressure is produced between the ends of the 
 carbon filament. Generally speaking, lamps are denomi- 
 nated by the pressure at which they must be worked to 
 give satisfactory results as to duration. This pressure 
 measured in volts is called the marked volts or maker's rolts, 
 and lamps may be distinguished as 100-volt lamps, 50-volt 
 lamps, 10-volt lamps, &c. The current taken by a lamp 
 can be measured by many kinds of instruments, which are 
 called ampere-meters or ammeters. The pressure between 
 the terminals of the lamp can be measured by suit- 
 able voltmeters. Approximately speaking, a lamp intended 
 to work at 100 volts and to give 16 candles light has a 
 carbon filament about five inches long and about one hundredth 
 of an inch in diameter, and when worked at a pressure of 
 100 volts it takes from f to f- of an ampere. From explana- 
 tions previously given it will be remembered that if these 
 two numbers namely, the number representing the volts 
 and the number representing the amperes are multiplied 
 together, we have a number representing the electrical power, 
 estimated in watts, which is being taken up in the lamp, and 
 such a 16 -candle-power 100-volt lamp generally takes from 
 50 to 60 watts. We have already described the methods 
 by which the candle-power can be measured. If we divide 
 the number representing the electrical power in watts taken 
 by the lamp by the number representing the candle-power 
 of the lamp, the quotient gives us the number representing 
 the watts per candle-power.' In most modern lamps this last 
 number varies from 2J to 8$, when the lamp is being used 
 at the marked volts. There are, therefore, four quantities in
 
 ELECTEIC GLOW LAMPS. 
 
 69 
 
 connection with every lamp which it is important to know. 
 First, the marked volts or voltage of the lamps ; second, 
 the ampere current taken by the lamp at the marked volts ; 
 third, the candle-power of the lamp; and, fourth, the watts 
 per canille-pou'er. Very briefly we may describe the process 
 by which these four quantities are determined. 
 
 We have already described the instrument called an electro- 
 static voltmeter, which is used for measuring electric pressure. 
 
 FIG. 22. Simple Ampere-Meter for Measuring Lamp Currents. 
 
 One form of instrument which may be used for measuring the 
 strength of the current taken by the lamp, and which is called 
 an ammeter or an ampere-meter, is made as follows : A coil 
 of wire C (see Fig. 22) is wound on a bobbin, and through this 
 coil of wire the current to the lamp is passed. This current 
 creates a magnetic field round the wire, as described in our 
 First Lecture. If a small piece of wire I is suspended on the end 
 of a delicately balanced lever, so that the iron is held freely just 
 at the entrance of the bobbin, the passage of an electric 
 current through the bobbin will attract the iron into the coil.
 
 70 ELECTRIC LAMPS AND ELECTE1C LIGHTING. 
 
 This movement of the iron is resisted hy a weight, W, properly 
 placed on the balance arm, or by the -weight of an indicating 
 needle. Such an instrument may be graduated to show on 
 a divided scale the strength of the current in amperes which 
 is passing through the coil. 
 
 It is important that, in making thus a determination of 
 the constants of a lamp viz., the current passing through 
 the lamp, the pressure difference at the terminals of the lamp, 
 and the candle-power of the lamp by comparison with a 
 suitable standard of light these quantities should all be mea- 
 sured at the same instant. By varying the current through the 
 lamp, as we can do by using pressures of different values, we 
 can obtain a series of observations by which the candle-power, 
 current, and volts on the terminals of the lamp are all measured 
 simultaneously for different values of the pressure on the lamp 
 terminals, and these values can be set out in a series of curves 
 which are called the characteristic curves of the lamp. In 
 Fig. 23 is shown a curve illustrating the manner in which the 
 current of a 16-c.p. lamp varies as the current changes. It 
 will be seen that a very small increase in the current is 
 accompanied by a large increase in the light of the lamp, and 
 it can be shown that the candle-power varies very nearly as 
 the sixth power of the current. That is to say, if a lamp 
 gives 1 candle light when a current of one ampere is passing 
 through it, then it would give 2x2x2x2x2x2 = 64 candle- 
 power if two amperes were passed through it. For a reason 
 to be explained later, however, no lamp would bear such a 
 relative increase in current. But the above-mentioned law 
 holds good proportionately for all glow lamps, viz., that the 
 candle-power varies approximately as the fifth or sixth power 
 of the current. In Fig. 24 is shown another curve, illustrating 
 the manner in which the candle-power varies with the volts, 
 and in Figs. 25 and 26 curves showing the manner in which 
 the candle-power varies with the watts and with the watts per
 
 ELECTRIC GLOW LAMPS. 
 
 VI 
 
 7 
 
 35 40 45 50 55 60 66 -7Q 
 
 Amperes. 
 
 FIG 23. Curve showing Variation of Candle-power with Current in an 
 Electric Glow Lamp. 
 
 80 90 
 
 Volts. 
 
 100 110 
 
 Fio. 24. Curve showing Variation of Candle-power with Voltage in 
 an Electric Glow Lamp.
 
 72 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 candle-power. From the curve in Fig. 25 it will be seen that 
 the candle-power of the lamp varies very nearly as the cube 
 of the total power in watts supplied to it. In other words 
 if we supply the lamp with twice the electrical power, the 
 candle-power is increased eight times. If we supply it with 
 three times the electrical power, it is increased twenty-seven 
 times. These curves will show, amongst other things, the 
 great importance of preserving the pressure at the terminals 
 of a lamp quite constant when that lamp is used as an 
 
 25 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 / 
 
 
 C 16 
 
 
 
 
 
 [/ 
 
 
 
 
 
 
 / 
 
 
 s. 
 
 
 
 
 / 
 
 
 
 f 
 
 
 
 / 
 
 / 
 
 
 
 
 
 
 ^ 
 
 ~jr 
 
 
 
 
 C 30 40 50 60 70 8 
 
 Watts. 
 
 FIG. 25. Curve showing Variation of Candle-power with Wattage 
 in a Glow Lamp. 
 
 illuminating agent. Any variation of the pressure, however 
 small, produces a very serious variation in the candle-power 
 of the lamp. It is convenient to bear in mind the following 
 figures for the 16-c.p. lamp as generally made. Such a lamp 
 takes, when worked at a pressure of XOO volts, 0-6 or f of an 
 ampere, absorbs 60 watts of power, and therefore in 16| hours 
 consumes an amount of energy which is 16f times 60 = 1,000
 
 ELECTRIC GLOW LAMPS. 
 
 73 
 
 watt-hours. This amount of energy is called one Board of 
 Trade Unit, and is generally sold by the Electric Supply 
 Companies in England at prices varying from 3d. to 8d. 
 
 As soon as we begin to use incandescent lamps in practice, 
 even although the pressure at the terminals may be kept 
 exceedingly constant, we find that lamps are like human 
 beings they have a certain life or duration, and moreover 
 most of them undergo a process of deterioration in light-giving 
 
 6 10 15 20 26 30 35 
 
 Watts per Candle 
 
 Fin. 26. Curve showing the Variation of Candle-power with Watts per 
 Candle-power in an Electric Glow Lamp. 
 
 power. After a certain time the filament of the lamp is 
 destroyed, and the lamp ceases to work. The duration of any 
 lamp when worked under certain conditions as to pressure 
 and supply is called its life, and from a large number of 
 observations of similarly-constructed lamps when worked in 
 a particular manner we can deduce the averaye life of a lamp. 
 We cannot predict from observations of one or two lamps
 
 74 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 what the average duration will be, any more than we can tell 
 from observations on one or two human lives what is the 
 average duration of human life in any place. The life of 
 any particular individual lamp may be only a few hours, or 
 it may be many thousands of hours. We shall presently 
 point out that this average duration is only one factor in 
 determining the real value of the lamp as a translating 
 device. Apart from accidental circumstances, the factors 
 that determine the life of the lamp are the temperature at 
 which the carbon filament is kept, and the nature of the 
 surrounding gas. Assuming, however, that an incandescent 
 lamp is supplied with electric current at a constant electric 
 pressure : it is always found that two marked changes occur 
 in the lamp in the process of time ; first, the lamp begins 
 to blacken by a deposit of carbon which is made upon the 
 interior of the bulb ; and, second, the carbon filament 
 undergoes a change by which its electrical resistance becomes 
 increased and its surface altered, so that, whatever its initial 
 condition, its surface finally assumes a much darker and more 
 sooty appearance than when first manufactured. 
 
 From these physical changes it follows that the candle-power 
 of the lamp is diminished, first by an obstruction of light by 
 the black deposit of carbon on the inner surface of the glass, 
 and, second, owing to the increase in the resistance of the 
 filament, the lamp taking less current and therefore furnishing 
 less light. The deposit of carbon on the interior of the bulb 
 is produced partly, or perhaps generally, by a process which 
 is an ordinary evaporation of the carbon, but it may be pro- 
 duced partly by volatilisation of condensed hydrocarbons from 
 the interior of the conductor. In addition, however, to this, 
 there is a process of projection of carbon molecules from the 
 filament in straight lines which is not of a kind generally 
 included in the term volatilisation. Many blackened carbon 
 lamps show lines of no deposit on the glass, which
 
 ELECTRIC GLOW LAMPS. 75 
 
 have been called molecular shadows. It is not unusual to 
 find lamps the interior surface of the glass of which has 
 become very black by a deposit of carbon. On examining the 
 lamp it will be found that on one side of the glass, lying in 
 the plane of the loop, there is a line of clear glass on which 
 no deposit has been made. This is very often shown well in 
 lamps in which the filament has been worn away at a point 
 about halfway down one leg of the loop (see Fig. 27) ; and 
 on examining such a lamp it will be found that there is a 
 well-marked line of no deposit upon the glass, just in the 
 
 FIG. 27. Glow Lamp with the glass bulb blackened by deposit of carbon, 
 molecular scattering having taken place from the point marked a on the 
 filament, and a shadow or line of no deposit thus produced at 6 on the glass 
 
 plane of the loop, and on the side farthest removed from the 
 point of rupture. Blackened lamps of this kind, therefore, 
 indicate clearly that carbon molecules have been shot off 
 in straight lines from the place where the carbon has subse- 
 quently been ruptured. If in a carbon filament there happens 
 to be a weak spot or place of high resistance, at that point 
 the current will generate heat at a greater rate than at other
 
 76 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 places, and the temperature of that spot will rise above the 
 temperature of the rest of the filament. Such points of greater 
 temperature can often be seen on carefully examining an old 
 or bad filament. From these high-temperature spots, carbon 
 molecules must be projected off by an action which may be 
 partly electrical, and are deposited upon the glass on all sides. 
 But it is clear that the unbroken leg would shield part of the 
 glass from this carbon molecule bombardment, and a shadow 
 of one-half of the loop would therefore be formed upon the 
 glass. 
 
 This shielding action may be illustrated by a simple experi- 
 ment. Here is a f| -shaped rod (Fig. 28). This shall represent 
 
 Fifl. 28. "Spray Shadow" of a rod thrown on Cardboard Screen to 
 illustrate formation of molecular shadow in Glow Lamps. 
 
 the carbon conductor in the lamp, and this sheet of cardboard 
 placed behind it the side of the glass bulb. I have affixed a 
 little spray-producer to one side of the loop, and from that 
 point blow out a spray of inky water. Consider the ink spray 
 to represent the carbon atoms shot off from the overheated 
 spot. We see that the cardboard is bespattered on all points 
 except along one line where it is sheltered by the opposite 
 side of the loop. We have thus produced a " spray-shadow " 
 on the board. The existence of these molecular shadows 

 
 ELECTRIC GLOW LAMPS. 77 
 
 in incandescent lamps leads us, therefore, to recognise 
 that the carbon atoms must be shot off in straight lines, or 
 else, obviously, no such sharp shadow could thus be formed. 
 It has been asserted by some writers that the whole of the 
 black coating on the interior of the carbon glow lamps is 
 produced solely by a process of evaporation of carbon ; but 
 the frequent existence of these shadows on blackened lamps 
 show that under some conditions the projection of carbon 
 from the filament is not merely a general and irregular 
 volatilisation, but a copious projection of carbon molecules, 
 which move out in perfectly straight lines from one part of 
 the filament.* 
 
 It is clear, therefore, from all the foregoing considerations 
 connected with the processes which are going on in the 
 interior of the lamp and the resulting ageing of the filament, 
 and from the changes that take place in the candle-power with 
 change of current, that the following conditions must be 
 observed in order to secure the best results in the employment 
 of incandescent lamps for illuminating purposes. In the 
 first place, the electric pressure which is supplied to the 
 consumer must be exceedingly constant. Where a supply 
 is drawn from the mains of a public electric lighting company 
 the consumer will, usually, have no control over the pressure, 
 at any rate in the case of continuous currents. The public sup - 
 ply companies of Great Britain are bound by the Electric Light- 
 
 * The author possesses a large collection of lamps showing these 
 molecular shadows. In the case of the old Edison lamp, the bamboo 
 filament was attached to the platinum leading-in wires by a deposit of 
 copper made over the clamp. Under these conditions, if one clamp 
 became loose, or made a bad contact, a scattering of copper molecules took 
 place from that clamp, producing a green transparent deposit of copper 
 over the interior of the bulb, which deposit generally showed a well- 
 marked shadow of one-half of the carbon loop upon the glass. Shadows 
 have also been produced by the writer by the volatilisation of alumiuium 
 plates in lamps, the volatilisation having been created by passing a sudden 
 strong current through an aluminium wire or plate included in the lamp 
 bulb. This aluminium deposit U of a fine blue colour.
 
 78 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 ing Acts to furnish current at a certain constant pressure, which 
 is generally 100 or 110 volts, and conditions are laid down by 
 the British Board of Trade that the pressure shall not vary 
 more than four volts above or four volts below this specified 
 pressure. These are very wide limits of variation, and very 
 few of the supply companies avail themselves of them. 
 The consumer who takes current from a public supply 
 
 FIG. 29. Holden's Self-recording Hot- Wire Voltmeter. 
 
 company should always take pains to ascertain for himself 
 whether the pressure is constant. He can do this by 
 the employment of a voltmeter, such as Lord Kelvin's electro- 
 static voltmeter, described in the previous Lecture, Fig. 8, or 
 he can employ a self-registering voltmeter, of which there are 
 many in use, which will record upon a drum covered with
 
 ELECTEIG GLOW LAMPS. 79 
 
 paper a curve indicating the variation of pressure during the 
 day and night. Two of these recording voltmeters which 
 are now much in use are those devised by Prof. G. Mengarini 
 and Captain Holden. In the latter instrument (see Fig. 29), 
 a very fine wire is connected across between the two points 
 between which pressure has to be determined. A small cur- 
 rent flows through this wire and heats it, causing it to expand. 
 The expansion of this wire is then made to move a lever, 
 which, by means of a writing pen, records upon a drum driven 
 by clockwork a curve, and any variation in the pressure or 
 voltage is shown by the form of the line so drawn. The 
 other instrument, due to Prof. Mengarini, is of a different 
 type. In this apparatus a current taken from the two points 
 between which the pressure has to be measured is passed 
 through two coils of wire one a fixed coil, and the other a 
 coil suspended by two steel wires. The fixed and movable 
 coils are so arranged (see Fig. 30) that when a current passes 
 through the two it causes the movable coil to twist round, 
 or tend to twist round, into a new position. This action 
 of the current is resisted by the suspended wires. A writing 
 pen is attached to the movable coil and draws a curve upon 
 a revolving drum, as above described, and this curve records 
 any variation in the voltage. The chart so drawn can then be 
 examined at the end of every day, and a record is thus kept of 
 the variation of pressure during the twenty-four hours. If the 
 consumer finds that the light of his incandescent lamps varies 
 a great deal, or if the life of the lamps is very irregular, it is 
 well to have such a curve of pressure automatically drawn 
 by a self-recording voltmeter for three or four days, and in 
 this way to detect the limits within which the electric pres- 
 sure varies. Consumers are very apt to lay the blame imme- 
 diately upon the lamps for any variation in light or brevity 
 in life, but as the electric pressure is partner with the lamp 
 in producing the result, it is evident that the sins of the 
 supply companies in giving a variable pressure ought not to
 
 80 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 be laid upon the lamps. On the other hand, if the recording 
 voltmeter shows a great uniformity in electric pressure, and 
 if still the life of the lamps is not satisfactory, then either the 
 lamps are in fault, or else the consumer is not using lamps of 
 the proper voltage for his circuits. Before ordering lamps, it 
 is well to ascertain also in this way what is the actual pressure 
 
 FIG. 30.--Mengarini's Self-recording Voltmeter.* 
 
 on the circuits, and then to order lamps of the proper marked 
 volts for the particular pressure. Incandescent lamps 
 which have been made and marked by the maker to work at 
 100 volts are not intended to be worked at 102 or 104, and, if 
 they are, abbreviation in the average life must certainly 
 * Reprinted by permission from the Society of Arts Journal. 

 
 ELECTRIC GLOW LAMPS. 
 
 81 
 
 occur, or at any rate a more rapid blackening of the globe and 
 diminution of the candle-power than would be the case if the 
 consumer employed lamps of the proper voltage. If the supply 
 is very irregular in pressure, it is never possible to obtain 
 the same good results in the duration of the lamp and 
 uniformity in brilliancy which can be obtained if the supply 
 is maintained constant in pressure, at least within the limits 
 of one or two volts. 
 
 Let us at this point note that, apart from the question of 
 duration or cost, the physical value of a glow lamp is 
 
 FIG. 31. Glow Lamp Burning under Water. 
 
 measured by its utility as an energy- transforming device. 
 An electric glow lamp is a machine for transforming energy 
 from one form, viz., the energy of an electric current, into 
 another form, viz., eye-affecting radiation. As explained in 
 Lecture I., it transforms the electric energy into radiant energy 
 or energy of ether waves, only a fraction of which can impress 
 the eye as light. This fraction varies from 3 to 7 per cent. 
 This numerical expression for the " luminous efficiency " can 
 be determined by suitable experiments in which the relative
 
 82 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 amounts of luminous and dark heat sent out by a glow lamp 
 are separately measured. If a glow lamp L (see Fig. 31) is 
 immersed in water in a transparent vessel V the water absorbs 
 or stops nearly all the dark or non-luminous radiation, but 
 permits a large fraction of the luminous radiation or light to 
 pass out. By photometering the lamp both when in and 
 when out of the water the fraction of the luminous radiation 
 which is absorbed can be determined. By measuring the rise 
 of temperature of the water which is created by the absorption 
 of the radiation, both when the luminous radiation is allowed 
 to pass out and when it is wholly stopped by blackening the 
 lamp, we can determine the ratio between the amount of the 
 radiation which affects the eye as light and that which cannot 
 affect it at all. A well-devised series of experiments of this 
 kind have been carried out by Mr. E. Merritt, at Cornell 
 University, U.S.A. The result has been to show that, on 
 an average, about 5 per cent, of the energy supplied to the 
 glow lamp as ordinarily used is utilized to make light, and 
 the other 95 per cent, is non-effective as far as the eye is 
 concerned. Hence, if a glow lamp is supplied with 60 watts 
 and gives 16 candles light, only 3 watts are really transformed 
 into light, and the mechanical value of the light so produced 
 is rather less than a quarter of a watt per candle. 
 
 This luminous efficiency varies with the temperature of the 
 filament. At a low red heat it is not more than 1 per cent., 
 but increases to about 5, 6, or 7 per cent, at the normal 
 working temperature. Thus, for an Edison 16-c.p. lamp 
 Mr. Merritt found the following figures when the lamp was 
 worked at various pressures so as to give different candle- 
 power. 
 
 It will be seen that the luminous efficiency increases as 
 the "watts per candle-power" decreases that is, as the 
 temperature of the filament increases. 

 
 ELECTRIC GLOW LAMPS. 
 
 83 
 
 The "luminous efficiency" must be carefully distinguished 
 from that which is sometimes called "the efficiency" of the 
 lamp, and which is the reciprocal of the watts per candle or 
 the number of candles light yielded per horse-power or per 
 watt expended in the filament. 
 
 Luminous Efficiency of a 16-c.p. Edison Glow Lamp. 
 
 Working 
 
 Candle- 
 
 Power 
 absorbed 
 
 Watts per 
 candle- 
 
 Luminous 
 radiation 
 
 Luminous 
 efficiency as a 
 
 
 pow 
 
 in watts. 
 
 power. 
 
 in watts. 
 
 percentage. 
 
 74-2 
 
 0-9 
 
 34-6 
 
 38-0 
 
 0-18 
 
 0'5 per cent. 
 
 91-6 
 
 48 
 
 56 -2 
 
 12-0 
 
 0-68 
 
 1-2 
 
 97-3 
 
 7-3 
 
 64-6 
 
 9-0 
 
 1-13 
 
 17 
 
 100-3 
 
 8-9 
 
 69-3 
 
 7-8 
 
 1 62 
 
 23 
 
 107-6 
 
 14-6 
 
 81-6 
 
 5-6 
 
 2-97 
 
 3-6 
 
 109-3 
 
 16-3 
 
 84-4 
 
 5-2 
 
 457 
 
 5-4 
 
 124-1 
 
 38-2 
 
 115-4 
 
 3-0 
 
 7-46 
 
 6-5 
 
 It will be convenient for me at this stage to enter a 
 little more fully into the changes that go on in an incandes- 
 cent electric lamp during its life. We have pointed out 
 above the causes which are at work to change the surface and 
 ultimately to destroy the carbon filament. These causes com- 
 mence to act from the very moment when the lamp begins to 
 be used. If a carbon filament electric glow lamp is placed 
 upon a circuit of perfectly constant pressure, and if the con- 
 stants of the lamp namely, the candle-power, the current, the 
 resistance of the filament, the power taken up in watts, and 
 the watts per candle-power are measured at intervals, say of 
 every 100 hours, it will be found that the changes which go 
 on in the lamp are somewhat as follows : During the life of 
 the lamp there is a progressive rise in resistance, a progres- 
 sive diminution in candle-power, and a progressive increase 
 in watts per candle-power. An illustration of these changes 
 is diagrammatically shown for certain forms of ordinary 
 vacuum lamps in the curves given in Fig. 32. These curves 
 
 G 2
 
 84 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 are deduced,* not from observations on one lamp alone, but 
 from the average of a large number. If, instead of working 
 the lamp at its marked volts or normal pressure, we raise the 
 voltage, say, 10 per cent, above the normal, changes of the 
 same character take place, only they are very much ac- 
 celerated, and the result for a collection of lamps worked 
 at 10 per cent, above the normal pressure is given in Fig. 33. 
 
 14 
 
 v 
 o 
 
 i 
 
 I 
 
 10 
 
 I 
 
 -o 8 
 
 c 
 
 9 
 1 6 
 
 a 
 
 r 
 
 
 2 
 
 O 
 
 \ 
 
 
 
 
 
 
 ^ 
 
 J 
 
 270 
 260 
 250 I 
 
 
 
 
 
 \ 
 
 s> 
 
 
 
 
 / 
 
 
 
 
 N^ 
 
 ^er 
 
 
 ^y 
 
 / 
 
 
 
 
 / 
 
 
 x^ 
 
 . 
 
 ^ 
 
 ^ 
 
 " 
 
 
 / 
 
 / 
 
 <&i? 
 
 ^ 
 
 
 
 *~ . 
 
 
 
 5 
 
 230 | 
 
 220 
 
 210 
 
 // 
 
 
 
 /" 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 IOO 200 300 4OO, 50O 6OO 700 8OO 
 Life m Hours 
 
 FIG. 32. Curves showing the increasing resistance and diminishing 
 candle-power of a Glow Lamp during its life. 
 
 On the other hand, if we desire to keep the candle-power 
 always constant, we have to continually raise the pressure, 
 and at the same time we are increasing the watts per candle 
 and the resistance of the lamp. These progressive changes 
 are illustrated in the two curves in Figs. 34 and 35, in the 
 
 * These curves are borrowed from a very instructive Paper by Prof. 
 Nichols, of Cornell University, U.S.A., on "The Artificial Light of the 
 Future."
 
 ELECTRIC GLOW LAMPS. 
 
 86 
 
 first of which a 16-c.p. lamp was worked at continually- 
 increasing pressures sufficient to keep the candle-power 
 always constant, and in the second of which the candle-power 
 was raised to four times its normal value and kept constant. 
 
 The value of a lamp to a user is not to be measured merely 
 by the watts per candle-power it takes at the commencement 
 
 3-00 
 
 t'2-50 
 
 I 
 
 i. 2 00 
 
 65 p 
 
 6 
 
 60 - 
 
 6 
 
 Life 
 
 8 10 
 
 Hours. 
 
 FIG. 33. -Curves showing the rapid decrease in candle-power and in- 
 crease in resistance in a Glow Lamp worked at 10 per cent, above its 
 normal pressure. 
 
 of its life, but also, to a large degree, by the extent to which 
 this consumption of energy and production of light remains 
 constant during the life of the lamp. This is best expressed 
 by the percentage of the variation of the candle-power during 
 the life of the lamp when taken at equal intervals of time. 
 A 16-c.p. glow lamp, taking say GO watts at the commence- 
 ment of its life, but which after two or three hundred hours
 
 86 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 in use is giving no more than 8 candles light, although taking 
 the same power, is costing the consumer twice as much per 
 candle at the end of this time as at the beginning. Some 
 writers have, therefore, suggested that at the end of a certain 
 period, when the candle-power has suffered a definite 
 diminution in value, say, of more than 30 per cent., the 
 lamp should be "smashed," and a new lamp put in its 
 place. This, however, is not always advisable or necessary. 
 
 240 
 
 FIG. 34. Curves showing the increase in Voltage required to keep the 
 candle-power of a Glow Lamp constant. 
 
 If the lamp in the position in which it is being used is giving 
 light enough for the purpose for which it is required, then, 
 although its candle-power may be diminished, if the total 
 power taken by the lamp is diminished also, the total sum 
 which is being paid by the consumer for the energy supplied 
 to that lamp is on the whole diminished, and therefore the 
 lamp is costing the consumer less on the iclwle to maintain
 
 ELECTRIC GLOW LAMPS. 87 
 
 in action than a new lamp would do if it replaced the old one, 
 although at the same time the cost of the lamp per candle- 
 light may be considerably increased. Generally speaking, 
 users of electric glow lamps have an idea that the lamp which 
 lasts the longest is necessarily the best. From what has 
 been stated above, it will be seen that it is not merely long life, 
 but constancy of candle-power, combined with high luminous 
 efficiency and low cost, which are really the factors deter- 
 
 I 
 
 A 
 O 
 
 
 
 
 
 
 
 / t 
 
 
 
 
 
 
 
 
 
 
 
 yL 
 
 
 
 TD 
 
 
 
 
 
 * e* 
 
 2SZ 
 
 
 
 
 
 o ' 85 
 
 & 
 
 
 
 ^ 
 
 ^* 
 
 
 */ 
 
 / 
 
 
 
 
 
 
 / 
 
 
 
 A 
 
 // 
 
 
 
 
 
 
 
 / 
 
 
 
 *?/ 
 
 & 
 
 
 
 
 
 J 
 
 
 / 
 
 
 
 *// 
 
 
 
 
 
 
 
 
 
 
 2 
 
 / 
 
 
 
 
 
 
 1-60 
 
 114 
 
 Y 
 
 ~ 
 
 
 
 
 
 
 
 
 Life in Minutes. 
 
 FIG. 35. Curves showing the increasing Voltage necessary to keep the 
 candle-power of a Glow Lamp constant and equal to four times its normal 
 amount. 
 
 mining the value of the lamp to the consumer. The duration 
 of the lamp is intimately connected, apart from accidents, 
 with the temperature at which the carbon filament is main- 
 tained, and this is determined by the watts per candle-power 
 being expended in the lamp. Makers of glow lamps some- 
 times make statements which lead the uninitiated to believe
 
 88 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 that there is a certain "efficiency" or peculiarly low value 
 of the watts per candle-power belonging especially to a 
 certain make of lamp. It should be, however, clearly under- 
 stood that any glow lamp may be worked at values of the 
 watts per candle-power varying over very wide limits, but 
 experience has shown that carbon filaments, as at present 
 made, cannot be worked at temperatures higher than 
 those corresponding to about 2 to 2| watts per candle- 
 power without very rapidly being destroyed. 
 
 A number of tests have been made by Messrs. Siemens and 
 Halske on lamps of various types, and in which groups of 
 these lamps were worked at different watts per candle-power 
 from 2 to 4 and upwards. The lamps were measured at 
 intervals of 100 hours, and the diminution in the observed 
 candle-power expressed as a percentage of the original candle- 
 power was laid down in the form of curves. The results 
 of these tests seemed to show that, on the whole, lamps 
 worked at from 3 to 3| watts per candle-power had a greater 
 average constancy in candle-power than those worked at a 
 lower number of watts per candle-power. That is to say, if we 
 take a 16-c.p. glow lamp and work its filament at such a 
 temperature that it is taking on the whole 32 watts, or 
 2 watts per candle-power, it may take less power to begin 
 with, but it will not preserve nearly the same constancy of 
 candle-power as if it had been worked at such a tempera- 
 ture that it absorbed from 48 to 56 watts at the com- 
 mencement of its career ; in other words, absorbed 3 to 3| 
 watts per candle-power. No hard and fast rule, however, 
 can be laid down about these matters. No deductions can 
 be made which have any great value from tests made 
 upon a very small number of lamps. As I have previously 
 stated, the lives of electric lamps are in this respect like 
 human lives, and just as the insurance companies have to 
 draw their deductions and to make their calculations for
 
 ELECTRIC GLOW LAMPS. 89 
 
 life insurance on the results of figures obtained from a 
 very large number of human lives, so inferences made 
 from the results of statistical tests carried out on incan- 
 descent lamps are only of use if they are made on a very large 
 number of lamps. Broadly speaking, however, the duration 
 of an incandescent lamp is dependent upon the electric 
 pressure at which the lamp is worked, and if a lamp is marked 
 to be used on a 100-volt circuit, then if it is used on a circuit 
 at a pressure of 105 volts, its life would be greatly abbreviated, 
 probably to one quarter or one fifth of the time which it would 
 last if worked at the normal pressure. On the other hand, 
 if it is worked at a pressure of 95 volts, its duration, accidents 
 apart, will be very greatly increased. It is useful, therefore, 
 to have, under these circumstances, a general knowledge of 
 what would be the variation in candle-power if the same lamp 
 is employed at different voltages. The tables on the next page 
 give the candle-power yielded by two particular electric glow 
 lamps when worked at varying pressures, the first table being 
 for a 100-volt 16-c.p. lamp, and the second table being for a 
 100-volt 8-c.p. lamp. A very cursory examination of these tables 
 will show that the light given by the lamp varies enormously 
 with the pressure, and that a very small diminution of pressure 
 suffices to deprive the user of a very large fraction of the 
 candle-power for which the lamp is constructed. 
 
 It may not be inappropriate at this point to call the atten- 
 tion of users of incandescent lamps to the fact that price is 
 not the only factor to be considered in connection with the 
 choice of a glow lamp. Consideration should also be given 
 to the rate at which the candle-power decays when constant 
 pressure is kept on the terminals of the lamp, as well as to 
 the initial efficiency, and the useful life of the lamp. The 
 useful life of the lamp, so far as most consumers are con- 
 cerned, must be considered to be limited to that period of the 
 total lamp life in which its candle-power is not diminished by
 
 90 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 Tables showing the Candle-power of certain nominal 8 and 16 
 Candle-power Incandescent Lamps when worked at various 
 pressures in Volts. 
 
 A 16 Candle-power Lamp. 
 
 An 8 Candle-power Lamp. 
 
 Worked at 
 
 Yielded a candle- 
 power of 
 
 ***** "Ery* 
 
 105 volts. 
 
 22-8 c.p. 
 
 102 volts. 7'9 c 
 
 P- 
 
 100 
 
 
 16-7 
 
 
 100 
 
 69 
 
 
 95 
 
 
 122 
 
 
 98 
 
 61 
 
 
 90 
 
 
 8-7 
 
 
 96 
 
 535 
 
 
 85 
 
 
 5-9 
 
 
 94 
 
 4-7 
 
 
 80 
 
 
 4-0 
 
 
 92 
 
 4-2 
 
 
 75 
 
 
 2-5 
 
 
 90 
 
 3-65 
 
 
 70 
 
 
 1-5 
 
 
 88 
 
 3-2 
 
 
 65 
 
 
 0-8 
 
 
 86 
 
 2-8 
 
 
 60 
 
 
 0-6 
 
 
 84 
 
 2-35 
 
 
 
 
 82 
 
 2-2 
 
 
 
 
 80 
 
 1-7 
 
 
 
 
 78 
 
 1-4 
 
 
 
 
 76 
 
 1-15 
 
 
 
 
 74 
 
 0:9 
 
 
 more than 30 to 40 per cent, of its original candle-power, 
 although this is not by any means a definite limit to the 
 utility of the lamp. The user has to determine what is the 
 total cost to him, during that useful life, of the energy 
 supplied to the lamp, and what is the total candle-power 
 hours given by the lamp. Suppose, for example, that the 
 lamp is a 100-volt 16-c.p. lamp, and that it takes originally 
 50 watts to bring it to this candle-power, when on a 
 100-volt circuit. Then let us suppose, in the course of 
 600 hours' burning, that it is reduced to 8 c.p., and that 
 it then takes 40 watts at the same pressure. If the rate 
 of decay is tolerably uniform, the total energy taken by 
 the lamp in watt-hours is the product of 45 = | (50 + 40), 
 and 600, or 27,000 watt-hours ; which is 27 British Board of 
 Trade units (frequently written B.T.U.). The average 
 candle-power hours given by the lamp is (16 + 8) = 12,
 
 ELECTRIC GLOW LAMPS. 91 
 
 which multiplied by 600 gives 7,200 candle-power hours. 
 If the Board of Trade unit of electric energy costs 6d., 
 then the total cost of the energy taken up is 13s. 6d., 
 for which the consumer gets 7,200 candle-hours of light at a 
 mean candle-power of 12. If the lamp originally cost Is. 6d., 
 the total cost of the 7,200 candle-hours, including lamp, is 
 15s., or the consumer gets 40 candle-hours for a penny, given at 
 a mean candle-power of 12. In other words, he is paying 
 0-3 of a penny per hour for his lamp and light. A penny 
 includes, therefore, the cost of lamp and power for about 3| 
 hours. Suppose, however, the user is offered a lamp at the 
 price of Is., but that extended experience shows that a 
 16-c.p. lamp of this kind falls to 8 candle-power in 300 hours' 
 use. It is clear that this last lamp, if it takes the same power 
 at each candle-power as the other, will transform in this time 
 45x300 = 13,500 watt-hours, or 13 Board of Trade units, 
 and the total cost of lamp at Is. and power at 6d. per 
 unit to the user will be 7s. 9d. For this he gets 12 x 300 
 = 3,600 candle-hours, or 39 candle-hours for a penny. 
 Hence, the cheaper lamp has not really much benefited him. 
 
 Therefore, the question which is so commonly asked, viz., 
 Is electric light cheaper or dearer than gas? cannot be an- 
 swered by a simple "Yes " or " No." The factors which decide 
 the relative cost, so far as the electric glow lamp is concerned, 
 are as follows : 1st. The price paid per Board of Trade unit 
 of electric energy. 2nd. The price of the lamp. 3rd. The 
 average useful life of the lamp. This last will depend not 
 only upon the lamp itself, but also upon the constancy of the 
 electric pressure of the supply. 4th. The manner in which 
 the "wiring" of the house or building has been done ; a factor 
 which greatly determines the degree to which the consumer 
 can effect economy by not using more lamps than is actually 
 necessary. 5th. The intelligence shown in placing the lamps 
 so as to make each do the utmost illuminating duty, and the
 
 92 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 care with which the control is effected by switching off 
 lamps not actually needed. We may add also a 6th. The 
 employment of a by-no-means negligible precaution in 
 securing the best results the keeping of the outside of 
 the lamps clean and free from dust and dirt by having them 
 well washed at intervals. A dusty lamp gives much less 
 light than a clean one. 
 
 To assist the reader in making certain essential com- 
 parisons, a table is given below which shows the total 
 cost in power and lamp renewals for working for 200 
 hours a 16-c.p. 100-volt glow lamp taking on an average 
 50 watts. The costs are given for various prices of the 
 Board of Trade unit from 3d. to 8d., and for lamps costing 
 Is. and Is. 6d., and for useful lives of from 200 to 1,000 hours. 
 Let it be borne in mind that the " useful life " of a lamp is 
 not a sharply marked period, but is the approximate average 
 time during which its candle-power has not deteriorated by 
 more than say 40 per cent, of its original value. 
 
 Table shouting the cost of 200 hours' use of a 100-volt 16-c.p. Gloiv 
 Lamp, including the proportionate cost of Lamp Renewals and 
 cost of Power. The Lamp is assumed to take on an average 
 50 watts, and to cost either Is. or Is. 6d. 
 
 Cost of Energy. 
 Board of Trade Unit. 
 
 3d. ! 4d. 5d. 
 
 6d. 
 
 7d. 
 
 8d. 
 
 Price of Lamp. 
 
 I/- 1/61/- 1/61/- 1/6 
 
 I/- l/6|l/- 1/6J1/- 1/6 
 
 Useful Life of Lamp 
 in hours. 
 
 Cost of 200 hours' use of Lamp. 
 
 200 hours. 
 
 3/6 4/- 
 
 4/4 4/10 
 
 5/2 5/8 
 
 6/- 6/66/107/4 
 
 7/8 8/2 
 
 400 
 
 3/- 3/3 
 
 3/10 4/1 
 
 4/8 4/11 
 
 5/6 5/9 
 
 6/4 6/7 
 
 7/2 7/5 
 
 600 
 
 2/10 3/- 
 
 3/8 3/10 
 
 4/6 4/8 
 
 5/4 5/6J6/2 6/4 
 
 71- 7/4 
 
 800 
 
 2/9 2/10 
 
 3/7 3/8 
 
 4/5 4/6 
 
 5/3 5/4 
 
 6/1 6/2 
 
 6/11 71- 
 
 1,000 
 
 2/8 2/9 
 
 3/6 3/7 
 
 4/4 4/5 
 
 5/2 5/3 /- 6/1 
 
 6/10 6/11
 
 ELECTRIC GLOW LAMPS. 93 
 
 The foregoing table must be read as follows : Suppose 
 the user gets current at 6d. per unit, and buys lamps at Is. 6d., 
 and let the lamp he buys be assumed to have an average useful 
 life of 800 hours. On looking along the column in line with 
 800 and under the heading 6d. and Is. 6d., will be seen 5s. 4d. 
 This is the cost of the 16-c.p. lamp for 200 hours of burning, 
 including proportion of lamp cost and power. Hence, in this 
 case the 800 hours' burning cost 21s. 4d. on the whole, and 
 for this the user gets an average candle-power of about 13 
 candles for 800 hours, or 13 x 800 = 10,400 candle-hours. It 
 is obvious that no hard and fast comparison with the cost of 
 equivalent candle-hours given by gas light is possible, unless 
 a careful statement is made of the manner in which the two 
 lights are used. One great advantage which electric illumina- 
 tion possesses, from a domestic and economical point of view, 
 is the ease with which it can be turned off and on. A switch 
 placed near the door enables a person leaving a room to 
 turn out all the electric lights without calling for the smallest 
 trouble in relighting them, which cannot be done in the case 
 of gas. In effecting a comparison of cost, we have to take 
 into consideration the time of wastage and non-useful com- 
 bustion in the consumption of gas ; and this, apart altogether 
 from the immense superiority of the illuminant which fouls 
 no air and destroys no decorations. Actual extended ex- 
 perience shows* that, with care and good management, 
 electric energy at 8d. per unit can be made to give as much 
 illumination by means of incandescent lamps as gas at 3s. 2d. 
 per thousand cubic feet used with ordinary gas burners. At 
 this rate, one Board of Trade unit of electric energy can be 
 made to do the work of about 200 cubic feet of gas. 
 
 A very important factor which enters into the question 
 of the cost of incandescent lighting is the care with which 
 the arrangements of the electric light " wiring " have 
 
 * See The Electrician, March 16, 1894, p. 561.
 
 94 ELECTRIC LAMPS AND ELECTEIC LIGHTING. 
 
 been carried out. The " wiring " may be done in such a 
 way that the user can only turn on and off large groups of 
 lamps at once. As far as possible each lamp should be on 
 a separate switch, and be capable of being used and put out 
 of use alone. Such independent wiring costs more per light 
 than group lighting, but it enables the user to create the 
 necessary illumination with a minimum of expenditure of 
 electric energy. 
 
 This is the moment to offer a word of caution against all 
 and every form of " cheap " electric lighting work. Too much 
 care cannot possibly be bestowed upon the quality of the 
 materials and character of the work in laying the electric service 
 lines and mains in buildings. This " wiring " has now become 
 a special trade, and in the hands of competent firms is carried 
 out with the utmost care and with all the numerous pre- 
 cautions which long experience has shown to be necessary 
 for safety and efficiency. No amount of inspection after the 
 work is finished and no amount of electric testing is really an 
 efficient substitute for a watchful and experienced eye during 
 the progress of the work. All important electric wiring work 
 ought to be thus watched on behalf of the owners by a clerk 
 of the works trained in this kind of supervision. It is not 
 sufficient for a consulting engineer to draft the specification 
 of the work ; it must be suitably supervised during its pro- 
 gress. It is advisable that every architect should thus be 
 competent to draft his own specification for electric house- 
 wiring, and that every clerk of the works should be competent 
 to supervise and inspect the electric wiring work during its 
 entire progress. 
 
 The actual average annual consumption of electric energy by 
 incandescent lamps placed in buildings of various classes differs 
 very greatly. Taking the case of supply by meter, it is found 
 that for a large class of private houses the average number of
 
 ELECTRIC GLOW LAMPS. 95 
 
 Board of Trade units taken per annum by an 8-c.p. lamp will 
 not exceed 12 to 20. Indeed 20 is generally an outside limit. 
 Lamps in clubs, hotels and restaurants show a very much 
 larger average than 20, and may reach 30 to 40 B.T.U. per 
 8-c.p. lamp per annum, or occasionally even more. It may 
 be mentioned that a 16-c.p. lamp would, under the same condi- 
 tions, take twice, and a 32-c.p. three times the above amount. 
 Since a 85-watt 8-c.p. lamp takes one B.T.U. of electric 
 energy in 30 hours of continuous use, it follows that 
 from 360 to 600 hours is the average total time during 
 which a lamp in a private house is in use in the year. 
 That is, from one hour to one hour and a half is the 
 average daily time during which each lamp is alight in 
 residential buildings. During the other 23 or 22| hours 
 of the day, however, the electric supply station has to 
 keep in readiness, machinery or other apparatus capable of 
 giving current for that lamp should the user require it. The 
 had factor of a supply station is the ratio of the actual 
 amount of Board of Trade units sold to the amount which 
 could be sold if the demand were constant and equal to the 
 maximum demand. It will be seen from the above figures 
 that the load factor of an electric supply station, supplying a 
 purely residential district, does not exceed 8 or 10 per 
 cent. The load factors of various classes of buildings are 
 very different. The load factor of private houses is seen 
 from the above figures to be even below 8 per cent. The 
 most profitable consumers, therefore, from the point of view 
 of the suppliers of electric current, are not those who have 
 most lamps, but those who use the lamps they have for the 
 longest number of hours. 
 
 We will now turn to consider the manner in which 
 incandescent lamps can be employed in order to produce 
 the best illuminating effect. The immense advantage 
 which electric incandescent lighting possesses is, that it
 
 96 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 lends itself so readily to decorative purposes. In order that 
 it may be employed in this sense to the best advantage, 
 certain general guiding principles have to be held in view 
 and followed. Unfortunately, this is not always done, and the 
 result is that the effectiveness of the light, from the artistic 
 point of view, is greatly diminished. 
 
 The first principle which should be allowed to guide us in 
 the production of an artistic effect by electric lighting is that, 
 while a proper and sufficient illumination is thrown upon 
 the surfaces to be illuminated, the source of light must be 
 itself, as far as possible, concealed. The illuminating power 
 of a light for the purpose of vision does not depend merely 
 upon the candle power of the lamp, but it depends upon the 
 amount of light which is received by the eye from the surfaces 
 from which it is reflected. The pupil of the eye is capable of 
 being varied over wide limits, probably from about four square 
 millimetres, or the 150th part of a square inch, to 70 square 
 millimetres or the tenth part of a square inch. This ex- 
 pansion or contraction of the pupil of the eye is effected by 
 muscles over which the will has no control, and is termed 
 in physiology a " reflex action." If the eye has been kept in 
 darkness for some time, then the pupil becomes expanded to 
 its greatest extent. Under these conditions, if the eye is 
 opened, and especially if it is exposed to a bright light, this light 
 falling on the retina causes an immediate muscular con- 
 traction of the pupil to be effected. It is very easy to see this 
 adjusting process, going on in the eye by shutting the eyes for 
 a few minutes and then opening them in a bright light whilst 
 a hand mirror is held before the face so as to examine the 
 pupils of the eyes. The pupils will then be seen to be rapidly 
 contracting in area immediately the eyes are opened. Hence, 
 if the eye is turned upon a brilliant line of light, the contraction 
 of the pupil which immediately sets in imposes a limitation 
 upon the amount of light which can enter the eye ; and if the 

 
 ELECTRIC GLOW LAMPS. 97 
 
 eye is turned immediately towards a dull, badly illuminated 
 surface, the pupil is not at once suddenly expanded again. 
 
 One part, at any rate, of the injurious effect produced by 
 trying to read or write in the twilight is due to the strain 
 produced in the eyes by the muscular effort thus demanded in 
 the eye. When, therefore, we enter a room in which there 
 are a number of bright lights, such, for instance, as a room in 
 which incandescent lamps are being employed, the filaments 
 of which can be seen, the moment that the eye is turned upon 
 these bright lines of light, muscular contraction of the pupil sets 
 in, and on turning the eye away again to other less illuminated 
 surfaces, the retina of the eye does not receive sufficient light 
 to observe details. In common language, the eye is ' ' dazzled. ' ' 
 There is a certain fascination about brilliant points or lines of 
 light which captivates and attracts the eye even against the 
 will. 
 
 It will thus be seen that the presence of brilliant lines 
 or points of light in a room has a tendency to keep the pupil 
 of the eye in a state of partial contraction, which unfits it for 
 obtaining the best visual effect when turned upon surfaces less 
 illuminated. Some eyes are especially sensitive in this respect. 
 If, however, the lamp globes are made of frosted glass, or in 
 other ways protected, so that the image of the incandescent 
 filament cannot be directly thrown upon the retina, the actual 
 visual effect may be increased in spite of the fact that such 
 frosted globes or screens may cut off from 30 to 50 per cent, of 
 the light. This is a common experience with everybody who 
 tries to read or write in the neighbourhood of a brilliant in- 
 candescent lamp. By covering the globe with tissue paper, or 
 with a ground glass or porcelain shade, although a diminution 
 to a large extent is caused in the total light actually dis- 
 tributed from the lamp, yet, nevertheless, we are able to see 
 better and more comfortably by it. The process, however,
 
 98 ELECTRIC LAMPS AND ELECTEIC LIGHTING. 
 
 which is commonly adopted of cutting off a large proportion 
 of the light by a semi-opaque globe is a wasteful one, because 
 there can be no object in making light merely to absorb it. 
 
 The proper method is to so place the incandescent lamp 
 that no light from the filament can directly enter the eye, 
 but so that all the light sent out from the filament shall be 
 reflected to the eye from the objects to be seen, such as 
 the walls of the room and the various objects in the room, 
 
 FIG. 36. View of the Interior of a Dining Room, Illuminated with 
 Shaded Electric Candles. 
 
 and that the light shall reach the eye only after reflection 
 from these surfaces. Bare or uncovered lamps, and especially 
 if suspended half way down a room or on a level with, or 
 a little above the eye, are a crude and exceedingly disagreeable 
 method of illumination. 
 
 In all cases where incandescent lighting is carried out by those 
 who understand the proper methods of using it, so as to pro- 

 
 ELECTRIC GLOW LAMPS. 99 
 
 duce the best artistic effects, the guiding principle is followed 
 of so placing the lamps that the whole of the light emitted 
 from them only reaches the eye after reflection from sur- 
 rounding objects. This result can be attained in a variety 
 of ways. The lamps may be placed in shades so as to 
 throw the light upwards upon the ceiling of the room or 
 upon the walls, from which surfaces it is ultimately reflected 
 down on to the objects in the room. Wherever highly 
 decorated and light ceilings exist in a room to be lit, this is 
 a very effective method of distributing the light. In Figs. 36, 
 37, 38, 39, and 40, are shown photographs of interiors so 
 illuminated. It is always possible to arrange glow lamps 
 protecting them by metallic or other ornamental shields, so 
 so as to obtain this desired effect. A natural shell, properly 
 supported, forms a very effective screen and reflecting surface. 
 The white pearly interior of the shell acts as a reflecting 
 surface, and if the shell is very slightly translucent, then 
 a pleasing effect is produced by the small and diffused light 
 which passes through. 
 
 The second important principle of decorative electric 
 lighting is to distribute the light properly and to prevent 
 the concentration of light in large masses, thereby avoid- 
 ing the production of harsh shadows. Everyone is familiar 
 with the exceedingly disagreeable effect which is produced 
 by sharp and strong shadows. A soft gradation of 
 light and shade is essential in the production of an 
 artistic effect in a room. The inartistic productions of 
 many amateur photographers, especially in the region of 
 portraiture, are much more due to the fact that they have 
 not the means at their disposal for producing the right 
 effects of light and shade on their subjects, than to any 
 defects in their photographic manipulation. Any concentra- 
 tion of the light in large masses, in public or private rooms, 
 
 H2
 
 100 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 FIG. 37. View of the Interior of a Drawing Room Lighted Electrically 
 with Screened Lamps. 
 
 FIG. 38. View of a Ball Room Illuminated with Shaded Electric 
 Caudles in Chandeliers.
 
 ELECTRIC GLOW LAMPS. 101 
 
 FIG. 39. View of a Reception Room Illuminated with Inverted 
 Incandescent Lamps throwing Light upwards. 
 
 FJG. 40. View of a Drawing Room Illuminated with Shaded 
 Electric Candles.
 
 102 ELECTRIC LAMPS AND ELECTEIC LIGHTING. 
 
 and especially when thrown downwards, invariably produces 
 disagreeable shadows upon the face; and, therefore, in places 
 like ball-rooms or drawing-rooms no such process of employ- 
 ing downwardly- suspended incandescent lamps should ever be 
 permitted if it is desired to obtain the best results. All artists 
 know the immense importance of obtaining proper lighting 
 by diffused light, and the northern aspect which they invariably 
 select for their studios is determined by this consideration. 
 It is essential that the light should be so distributed in small 
 units, in order that there shall be no sharp shadows cast 
 upon any object. A simple test for the effectiveness of the 
 distribution of light in any room may be obtained as follows : 
 Take a white card or a sheet of white writing paper and hold 
 it horizontally about the level of the eyes, then hold a pencil 
 or other small rod vertically on the card, and note if any 
 marked shadow of the pencil is thrown in any direction : if it 
 is, then the light is insufficiently diffused, and the lamps 
 ought to be re-arranged. The best effects as to distribution 
 of light are generally obtained by the employment of 5 candle- 
 power lamps. These glow lamps are constructed with bulbs 
 of semi-opaque glass having the shape of a candle flame. 
 (see Fig. 21). These candle lamps are carried on the ex- 
 tremity of an artificial porcelain candle, so as to resemble in 
 some degree an ordinary wax candle alight. These lamps are 
 then arranged in groups on chandeliers or brackets, being pro- 
 tected in many cases from direct visibility by plain or orna- 
 mental shades. In Figs. 36 and 41 are shown photographic 
 illustrations of rooms which are lit by such candle lamps, and 
 in which the light from these is thrown up against the walls 
 of the room, the candle lamps not being themselves directly 
 visible. 
 
 A third great guiding principle of artistic lighting is to pro- 
 portion the light to the nature of the surfaces from which it 
 is reflected. We require light in our living rooms to see the 

 
 ELECTRIC GLOW LAMPS. 
 
 103 
 
 FIG. 41. Drawing Room Illuminated with Electric Lamps, so shaded 
 that no direct view of the lamp filament can be obtained. 
 
 FIG. 42. Interior of Freemason's Hall Illuminated by Shaded Incandescent 
 Lamps on Frieze and Ceiling.
 
 104 ELECTRIC LAMPS AND ELECTEIC LIGHTING. 
 
 pictures, decorations, objects, and, above all, to see one another; 
 and we can only see these objects by the light which they 
 reflect back to the eye. The aim should, therefore, be to have 
 sufficient but not excessive incident light. It should be borne 
 in mind that the different surfaces with which the walls of 
 rooms are covered have very different reflective powers. Dull 
 walls absorb eighty per cent, of the light falling on them, and 
 reflect only about twenty per cent, of the light ; this is the case, 
 for instance, with walls covered w T ith dark oak panelling or dulJ- 
 coloured paint or paper. Ordinary light tints of paper or 
 paint may reflect from forty to sixty per cent. ; clean white 
 surfaces, such as white painted and varnished wood or plaster, 
 may reflect as much as eighty per cent., and mirrors from 
 eighty to ninety per cent. 
 
 The following figures have been given by Dr. Sumpner 
 for the reflecting power of various surfaces : 
 
 Table of Eeflecting Pmcers of Various Surfaces. 
 
 White blotting paper 82 per cent. 
 Ordinary foolscap ... 70 ,, 
 Newspapers 50-70 
 Yellow wall paper ... 40 
 Blue paper 25 ,. 
 Dark brown paper... 13 
 Deep chocolate paper 4 
 
 Plane deal (clean) 40-50 per cent. 
 Plane deal (dirty) ... 20 
 Yellow painted wall 
 (dirty) 20 
 Yellow painted wall 
 (clean) 40 
 Black cloth 1-2 
 Black velvet '4 
 
 Everyone is familiar with the way in which the laying 
 of a white table-cloth brightens up a dark dining room, or 
 in which the substitution of white for dark curtains lightens 
 up a room. The reason for this is obvious from the above 
 table. Hence, rooms with dark oil paintings, dark woodwork, 
 hangings, curtains, or portieres require relatively much more 
 illumination per square foot than rooms with very light 
 decorations, water-colour paintings, mirrors, &c. Experience 
 has shown that in a room of the latter description an illumi-
 
 ELECTRIC GLOW LAMPS. 105 
 
 nation is required which is equal to an expenditure in the 
 lamp of about f or 1 watt per square foot of floor surface, on 
 the assumption that the lamps are placed, on an average, 
 about 8 or 9 feet above the floor. A room of the former 
 description, such as a picture gallery with dark oak walls, 
 may require from 2 to 3 watts per square foot of floor surface. 
 These figures, however, cannot be taken too absolutely. 
 Experience is the only guide as to the amount of light to be 
 placed in a room. As a broad general rule, 100 square feet 
 of floor surface will be barely illuminated by one 16-c.p. 
 lamp placed about 8 feet above the floor. It will be 
 well illuminated by two such lamps placed 8 feet above the 
 floor, and brilliantly illuminated by four such lamps. In 
 other words, one 16-c.p. lamp per 100 square feet of 
 floor surface is hardly sufficient, one 16-c.p. lamp per 50 
 square feet of floor surface is fairly good, and one for every 
 25 square feet of floor surface is brilliant. These figures 
 are given on the assumption that the lamp is placed about 
 8 feet above the floor, and that the floor surface is fairly 
 reflective. The lighting of pictures is a very special study, and 
 can only be properly undertaken by those who have given 
 considerable attention to the method of placing and using 
 incandescent lamps. The mere illumination of a room by a 
 number of lamps hung head downwards by cord pendants 
 from the ceiling, a method which is not unusually adopted, is 
 the most crude and imperfect method of applying the electric 
 light, and ought never to be permitted in any case where the 
 smallest decorative effect is really desired. 
 
 The brief limits of a lecture will not permit us to enlarge 
 at greater length on this part of the subject, but it opens up 
 a wide field for the exercise of the inventive and aesthetic 
 faculties. 
 
 Before leaving the subject of incandescent lamps it will 
 not be without interest to refer briefly to the investigations
 
 106 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 which have been made into a curious effect, found to exist 
 in lamps which have a third or idle metallic wire sealed 
 through the glass. Our starting-point for this purpose is a 
 discovery made by Mr. Edison in 1884, and which received 
 careful examination at the hands of Mr. Preece in the follow- 
 ing year,* and by the author more recently. Here is the 
 initial experiment. A glow-lamp having the usual horseshoe- 
 shaped carbon (Fig. 43) has a metal plate held on a platinum 
 wire sealed through the glass bulb. This plate is so fixed that 
 it stands up between the two sides of the carbon arch without 
 touching either of them. We shall illuminate the lamp by a 
 
 FIG. 43. Glow Lamp having insulated metal middle plate M sealed into 
 the bulb to exhibit the " Edison effect." 
 
 continuous current of electricity, and for brevity's sake speak 
 of that half of the loop of carbon on the side by which the 
 current enters it as the positive leg, and the other half of the 
 
 _ __ 
 
 * Mr. Preece's interesting Paper on this subject is published in the 
 Proceedings of the Royal Society for 1885, p. 219. See also The Electrician, 
 April 4, 1885, p. 436, "On a Peculiar Behaviour of Glow-lamps when 
 Raised to High Incandescence." The following statements of experiments 
 are chiefly taken from a Friday evening discourse given by the author at 
 the Royal Institution, February 14, 1890, on " Problems in the Physics of 
 an Electric Lamp," and from a Paper on "Electric Discharge between 
 Electrodes at different Temperatures in Air and in High Vacua." (Proceed- 
 ings of the Royal Society, 1890. )
 
 ELECTRIC GLOW LAMPS. 
 
 107 
 
 loop as the negative leg. The diagram in Fig. 45 shows the 
 position of the plate with respect to the carbon loop. There 
 is a distance of half-an-inch, or in some cases many inches, 
 between either leg of the carbon and this middle plate. 
 Setting the lamp in action, I connect a sensitive galvanometer 
 between the middle plate and the negative terminal of the 
 lamp, and you see that there is no current passing through 
 the instrument. If, however, I connect the terminals of my 
 galvanometer to the middle plate, and to the positive electrode 
 of the lamp, we find a current of some milliamperes is passing 
 through it. The diagrams in Fig. 45 show the mode of con- 
 
 FIG. 44. Sensitive Galvanometer connected between the middle plate 
 and positive electrode of a Glow Lamp, showing current flowing through it 
 when the lamp is in action (" Edison effect "). 
 
 nection of the galvanometer in the two cases. This effect, 
 which is often spoken of as the "Edison effect," clearly 
 indicates that an insulated plate so placed in the vacuum of 
 a lamp in action is brought down to the same potential or 
 electrical state as the negative electrode of the carbon loop. 
 On examining the direction of the current through the galva- 
 nometer we find that it is equivalent to a flow of negative 
 electricity taking place through it from the middle plate to the
 
 108 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 positive electrode of the lamp. A consideration of this fact 
 shows us that there must he some way hy which negative 
 electricity gets across the vacuous space from the negative leg 
 of the carbon to the metal plate, whilst at the same time a 
 negative charge cannot pass from ihe metal plate across to 
 the positive leg. 
 
 Before passing away from this initial experiment, I 
 should like to call your attention to a curious effect at the 
 moment when the lamp is extinguished. Connecting the 
 galvanometer, as at first, between the middle plate and the 
 negative electrode of the lamp, we notice that, though made 
 
 (No current.) (A current.) 
 
 F IG . 45. Mode of connection of galvanometer G to middle plate M and 
 carbon horseshoe-shaped conductor C in the experiment of the " Edison 
 effect." 
 
 highly sensitive, the galvanometer indicates no current flowing 
 through it whilst the lamp is in action. Switching off the 
 current from the lamp produces, as you see, a violent kick or 
 deflection of the galvanometer, indicating a sudden rush of 
 current through it. 
 
 In endeavouring to ascertain further facts about this effect 
 one of the experiments which early suggested itself was 
 directed to determine the relative effects of different portions of 
 the carbon conductor. -Here is a lamp (see Fig 46) in which 

 
 ELECTEIC GLOW LAMPS. 109 
 
 one leg of the carbon horseshoe has been enclosed in a glass 
 tube the size of a quill, which shuts in one half of the carbon. 
 The bulb contains, as before, an insulated middle plate. If we 
 pass the actuating current through this lamp in such a direc- 
 tion that the covered or sheathed leg is the positive leg, we 
 find the effect existing as before. A galvanometer connected 
 between the plate and positive terminal of the lamp yields a 
 strong current, whilst if connected between the negative 
 terminal and the middle plate there is no current at all. Let 
 us, however, reverse the current through the lamp so that the 
 
 FIG. 46. Glow Lamp having negative leg of carbon enclosed in glass tube 
 T, the " Edison effect" being thereby annulled or greatly diminished. 
 
 shielded or enclosed leg is now the negative one, and the galva- 
 nometer is able to detect no current, whether connected in one 
 way or in the other. We establish, therefore, the conclusion 
 that it is the negative leg of the carbon loop which is the active 
 agent in the production of this " Edison effect," and that, if 
 it is enclosed in a tube of either glass or metal, no current 
 is found flowing in a galvanometer connected between the 
 positive terminal of the lamp and this middle collecting plate
 
 110 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 Another experiment which confirms this view is as follows : 
 The lamp (Fig. 47) has a middle plate, which is provided 
 with a little mica flap or shutter on one side of it. When the 
 lamp is held upright the mica shield falls over and covers one 
 side of the plate, but when it is held in a horizontal position 
 the mica shield falls away from the front of the plate and 
 exposes it. Using this lamp as before, we find that, when the 
 positive leg of the carbon loop is opposite to the shielded face 
 of the plate, we get the " Edison effect " as before in any posi- 
 tion of the lamp. Eeversing the lamp current, and making 
 
 FIG. 47. Glow Lamp having mica shield S interposable between middle 
 plate M and negative leg of carbon, thereby diminishing the ' ' Edison 
 effect." 
 
 that same leg the negative one, we find that, when the lamp is 
 so held, the metal plate is shielded by the interposition of the 
 mica, and the galvanometer current is very much less than 
 when the shield is shaken on one side and the plate exposed 
 fully to the negative leg. 
 
 At this stage it will perhaps be most convenient to outline 
 briefly the beginnings of a theory which may be proposed to 
 reconcile these facts, and leave you to judge how far the
 
 ELECTRIC GLOW LAMPS. Ill 
 
 subsequent experiments seem to confirm this hypothesis. 
 Very briefly, this theory is as follows : From all parts of the 
 incandescent carbon loop, but chiefly from the negative leg, 
 carbon molecules are being projected which carry with them, 
 or are charged with, negative electricity. In a few moments 
 a suggestion will be made to you which may point to a 
 possible hypothesis on the manner in which the molecules 
 acquire this negative charge. Supposing this, however, to be 
 the case, and that the bulb is filled with these negatively- 
 charged molecules, what would be the result of introducing 
 into their midst a conductor such as this middle metal plate 
 which is charged positively ? Obviously, they would all be 
 attracted to it and discharge against it. Suppose the positive 
 charge of this conductor to be continually renewed, and the 
 negatively- charged molecules continually supplied which 
 conditions can be obtained by connecting the middle plate to 
 the positive electrode of the lamp the obvious result will 
 be to produce a current of electricity flowing through the wire 
 or galvanometer by means of which this middle plate is 
 connected to the positive electrode of the lamp. If, however, 
 the middle plate is connected to the negative electrode of the 
 lamp, the negatively -charged molecules can give up no charge 
 to it, and produce no current in the interpolated galvanometer. 
 We see that, on this assumption, the effect must necessarily be 
 diminished by any arrangement which prevents these nega- 
 tively-charged molecules from being shot off the negative leg 
 or from striking against the middle plate. 
 
 Another obvious corollary from this theory is that the 
 " Edison effect " should be annihilated if the metal collect- 
 ing plate is placed at a distance from the negative leg much 
 greater than the mean free path of the molecules. 
 
 Here are some experiments which confirm this deduction. 
 In this bulb (Fig. 48) the metal collecting plate, which is to be
 
 112 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 connected through the galvanometer with the positive terminal 
 of the lamp, is placed at the end of a long tube opening out 
 of and forming part of the bulb. We find the " Edison 
 effect " is entirely absent, and that the galvanometer current 
 is zero. We have, as it were, placed our target at such a 
 distance that the lowest range molecular bullets cannot hit 
 it, or at least that but very few of them do so. Here, again, is 
 a lamp in which the plate is placed at the extremity of a tube 
 opening out of the bulb, but bent at right angles (Fig. 49). 
 We find in this case, as first discovered by Mr. W. H. Preece, 
 that there is no " Edison effect." Our molecular marksman 
 
 FIG. 48. Collecting plate placed at end of tube, 18in. in length, opening 
 out of bulb. 
 
 cannot shoot round a corner. None of the negatively-charged 
 molecules can reach the plate, although that plate is placed 
 at a distance not greater than would suffice to produce the 
 effect if the bend were straightened out. Following out our 
 hypothesis into its consequences would lead us to conclude 
 that the material of which the plate is made is without 
 influence on the result, and this is found to be the case. 
 
 We should expect also to find that the larger we make our 
 plate, and the nearer we bring it to the negative leg of the 
 carbon, the greater will be the current produced in a circuit
 
 ELECTRIC GLOW LAMPS. 
 
 113 
 
 connecting this plate to the positive terminal of the lamp. I 
 have before me a lamp with a large plate placed very near the 
 negative leg of the carbon of a lamp, and we find that we 
 can collect enough current from these molecular charges to 
 work a telegraph relay and ring an electric bell. The current 
 which is now working this relay is made up of the charges 
 collected by the plate from the negatively- charged carbon 
 molecules, which are projected against it from the negative leg, 
 across the highly perfect vacuum. I have tried experiments 
 with lamps in which the collecting plate is placed in all kinds 
 
 FIG. 49. Collecting plate placed at end of elbow tube opening out of bulb. 
 
 of positions, but the results may all be summed up by saying 
 that the greatest effects are produced when the collecting plate 
 is as near as possible to the base of the negative end of the 
 loops, and, as far as possible, encloses, without touching, the 
 carbon conductor. It is not necessary to make more than 
 a passing reference to the fact that the magnitude of the 
 current flowing through the galvanometer when connected 
 between the middle plate and the positive terminal of the lamp 
 often " jumps " from a low to a high value, or vice versa, 
 in a remarkable manner, and that this sudden change in the 
 current can be produced by bringing strong magnets near the 
 outside of the bulb.
 
 114 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 Let us now follow out into some other consequences the 
 hypothesis that the interior of the bulb of a glow lamp, when 
 in action, is populated by flying crowds of carbon atoms all 
 carrying a negative charge of electricity. Suppose we connect 
 our middle collecting plate with some external reservoir of 
 electric energy, such as a Ley den jar, or with a condenser 
 equivalent in capacity to many hundreds of Leyden jars, 
 and let the side of the condenser which is charged positively 
 be first placed in connection through a galvanometer with the 
 middle plate (see Fig. 50), whilst the negative side is placed 
 
 FIG. 50. Charged condenser C discharged by middle plate J/, when the 
 positively-charged side of condenser is in connection with the plate and 
 the other side to earth e. 
 
 in connection with the earth. Here is a condenser of two 
 microfarads capacity so charged and connected. Note what 
 happens when I complete the circuit, and illuminate the lamp 
 by passing the current through its filament. The condenser 
 is at once discharged. If, however, we repeat the same ex- 
 periment, with the sole difference that the negatively-charged 
 side of the condenser is in connection with the middle plate, 
 then there is no discharge. 

 
 ELECTRIC GLOW LAMPS. 115 
 
 These very interesting experimental results may also 
 be regarded from another point of view. In order that the 
 condenser may be discharged, as in the first case, it is essential 
 that the negatively-charged side of the condenser shall be in 
 connection with some part of the circuit of the incandescent 
 carbon loop. This experiment with the condenser discharged 
 by the lamp may be, then, looked upon as an arrangement in 
 which the plates of a charged condenser are connected respec- 
 tively to an incandescent carbon loop and to a cool metal 
 plate, both being enclosed in a highly-vacuous space ; and it 
 appears that the discharge takes place when the incandescent 
 conductor is the negative electrode of this arrangement, but 
 not when the cooler metal plate is the negative electrode of 
 the charged condenser. The negative charge of the condenser 
 can be carried across the vacuous space from the hot carbon 
 to the colder metal plate, but not in the reverse direction. 
 
 This experimental result led me to examine the condition 
 of the vacuous space between the middle metal-plate and the 
 negative leg of the carbon loop in the case of the lamp 
 employed in our first experiment. Let us return for a 
 moment to that lamp. I join the galvanometer between the 
 middle plate and the negative terminal of the lamp, and find, 
 as before, no indication of a current. The metal plate and 
 the negative terminal of the lamp are at the same electrical 
 potential. In the circuit of the galvanometer we will insert a 
 single galvanic cell, having an electromotive force of rather 
 over one volt. In the first place let that cell be so inserted 
 that its negative pole is in connection with the middle plate, 
 and its positive pole in connection through the galvanometer 
 with the negative terminal of the lamp (see Fig. 51). Regard- 
 ing the circuit of that cell alone, we find that it consists of 
 the cell itself, the galvanometer wire, and that half-inch of 
 highly vacuous space between the hot carbon conductor and 
 the middle plate. In that circuit the cell cannot send any 
 
 i2
 
 116 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 sensible current at all, as it is at the present moment con- 
 nected up. But if we reverse the direction of the cell, so that 
 its positive pole is in connection with the middle plate, the 
 galvanometer at once gives indications of a very sensible 
 current. This highly-vacuous space, lying between the 
 middle metal plate on the one hand and the incandescent 
 carbon on the other, possesses a kind of unilateral conduc- 
 tivity, in that it will allow the current from a single galvanic 
 cell to pass one way but not the other. It is a very old and 
 familiar fact, that in order to send a current from a battery 
 through a highly-rarefied gas by means of metal electrodes 
 
 FIG. 51. Current from Clark cell Ck being sent across vacuous space 
 between negative leg of carbon and middle plate M. Positive pole of cell 
 in connection with plate M through galvanometer G. 
 
 the electromotive force of the battery must exceed a certain 
 value. Here, however, we have indication that if the negative 
 electrode by which that current seeks to enter the vacuous 
 space is made incandescent the current will pass at a very 
 much lower electromotive force than if the electrode is not 
 so heated. 
 
 .A little consideration of the foregoing experiments led to the 
 conclusion that, in the original experiment, as devised by Mr. 

 
 ELECTRIC GLOW LAMPS. 
 
 117 
 
 Edison, if we could by any means render the middle plate very 
 hot, we should get a current flowing through a galvanometer 
 when it is connected between the middle plate and the negative 
 electrode of the carbon. This experiment can be tried in the 
 manner now to be shown. Here is a bulb (Fig. 52) having in it 
 two carbon loops : one of these is of ordinary size, and will be 
 rendered incandescent by the current from the mains. The 
 other loop is very small, and will be heated by a well-insulated 
 secondary battery. This smaller incandescent loop shall be 
 employed just as if it were a middle metal plate. It is, in 
 fact, simply an incandescent middle conductor. On repeating 
 
 FIG. 52. Experiment showing that when the " middle plate " is a 
 carbon loop rendered incandescent by insulated battery B, a current of 
 negative electricity flows from M to the positive leg of main carbon C 
 across the vacuum. 
 
 the typical experiment with this arrangement, we find that 
 the galvanometer indicates a current when connected between 
 the middle loop and either the positive or the negative 
 terminal of the main carbon. I have little doubt but that, 
 if we could render the platinum plate in our first-used lamp 
 incandescent by concentrating on it from outside a powerful 
 beam of radiant heat, we should get the same result.
 
 
 118 ELECTRIC LAMPS AND ELECTEIC LIGHTING. 
 
 A similar set of results can be arrived at by experiments 
 with a bulb constructed like an ordinary vacuum tube, and 
 having small carbon loops at each end instead of the usual 
 platinum or aluminium wires. Such a tube is now before 
 you (see Fig. 53), and will not allow the current from a few 
 cells of a secondary battery to pass through it when the 
 carbon loops are cold. If, however, by means of well- 
 insulated secondary batteries, we render both of the carbon 
 loop electrodes highly incandescent, a single cell of a battery 
 is sufficient to pass a very considerable current across that 
 vacuous space, provided the resistance of the rest of the 
 circuit is not large. We may embrace the foregoing facts 
 by saying that if the electrodes, but especially the negative 
 electrode, which form the means of ingress and egress of a 
 
 G 
 
 Fio. 53. Vacuum tube having carbon loop electrodes, c c, at each end, 
 rendered incandescent by insulated batteries, S l B,,, showing current from 
 Clark cell, Cl~, passing through the high vacuum when the electrodes are 
 incandescent. 
 
 current into a vacuous space, are capable of being rendered 
 highly incandescent, and if at that high temperature they 
 are made to differ in electrical potential by the application 
 of a very small electromotive force, we may, under these cir- 
 cumstances, get a very sensible current through the rarefied 
 gas. If the electrodes are cold, a very much higher electro- 
 motive force will be necessary to get the discharge or 
 current through the space. These facts have been made 
 the subject of elaborate investigation by Hittorf and Gold- 
 stein, and more recently by Elster and Geitel. It is to
 
 ELECTRIC GLOW LAMPS. 119 
 
 Hittorf, I believe, that we are indebted for the discovery of 
 the fact that by heating the negative electrode we greatly 
 reduce the apparent resistance of a vacuum. 
 
 Permit me now to pave the way by some other experiments 
 for a little more detailed outline of the manner in which I 
 shall venture to suggest these negative molecular charges are 
 bestowed. This is really the important matter to examine. In 
 seeking for some probable explanation of the manner in which 
 these wandering molecules of carbon in the glow-lamp bulb 
 obtain their negative charges, I fall back for assistance upon 
 some facts discovered by the late Prof. Guthrie. He showed 
 some years ago new experiments on the relative powers of in- 
 candescent bodies for retaining positive and negative charges. 
 One of the facts he brought forward* was that a bright red- 
 hot iron ball, well insulated, could be charged negatively, but 
 could not retain for an instant a positive charge. He showed 
 this fact in a way which it is very easy to repeat as a lecture 
 experiment. Here is a gold-leaf electroscope, to which we will 
 impart a positive charge of electricity, and project the image 
 of its divergent leaves on the screen. A poker, the tip of which 
 has been made brightly red hot, is placed so that its incandes- 
 cent end is about an inch from the knob of the electroscope. 
 No discharge takes place. Discharging the electroscope with 
 my finger, I give it a small charge of negative electricity, and 
 replace the poker in the same position. The gold leaves in- 
 stantly collapse. Bear in mind that the extremity of the poker 
 when brought in contiguity to the knob of the charged electro- 
 scope becomes charged by induction with a charge of the 
 opposite sign to that of the charge of the electroscope, and you 
 will at once see that this experiment confirms Prof. Guthrie's 
 statement, for the negatively-charged electroscope induces a 
 positive charge on the incandescent iron, and this charge 
 
 * " On a New Relation between Electricity and Heat," Phil. Mag., 
 Vol. XLY., 1873, p. 308.
 
 120 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 cannot be retained. If the induced charge on the poker is a 
 negative charge it is retained, and hence the positively- 
 charged electroscope is not discharged, but the negatively- 
 charged electroscope at once loses its charge. 
 
 Pass in imagination from iron balls to carbon molecules. It 
 may be asked whether it is a legitimate assumption to suppose 
 the same fact to hold good for them, and if a hot carbon 
 molecule or small carbon mass just detached from an incan- 
 descent surface behaves in the same way, and has a greater 
 grip for negative than for positive charge ? If this can possi- 
 bly be assumed, we can complete our hypothesis as follows : 
 Consider a carbon molecule or small congerie of molecules 
 
 FIG. 54. Rough diagram illustrating a theory of the manner in which pro- 
 jected carbon molecules may acquire a negative charge. 
 
 just set free by the high temperature from the negative leg of 
 the incandescent carbon horseshoe. This small carbon mass 
 finds itself in the electrostatic field between the branches of 
 the incandescent carbon conductor (Fig. 54). It is acted upon 
 inductively, and, if it behaves like the hot iron ball in Prof. 
 Guthrie's experiment, it loses its positive charge. The mole- 
 cule then being charged negatively is repelled along the lines 
 of electric force against the positive leg. The forces moving 
 it are electric forces, and the repetition of this action would 
 cause a torrent of negatively- charged molecules to pour across 
 from the negative to the positive side of the carbon horseshoe.
 
 ELECTRIC GLOW LAMPS. 121 
 
 If we place a metal plate in their path which is in conducting 
 connection with the positive electrode of the lamp carbon, the 
 negatively- charged molecules will discharge themselves against 
 it. A plate so placed may catch more or less of the stream 
 of charged molecules which pour across between the heels of 
 the carbon loop. There are many extraordinary facts, which 
 as yet I have been able only imperfectly to explore, which 
 relate to the sudden changes in the direction of the principal 
 stream of these charged molecules, and to their guidance 
 under the influence of magnetic forces. The above rough 
 sketch of a theory must be taken for no more than it is worth, 
 viz., as a working hypothesis to suggest further experiments. 
 
 It has been noticed that such a middle plate placed between 
 the legs of the carbon horseshoe becomes black soonest on 
 the side facing the negative leg of the carbon. It has also 
 been noticed that, in the case of lamps with treated filaments, 
 the surface of the deposit assumes a dead lamp-black 
 appearance soonest on the negative leg. These observations 
 tend to support the contention that there is an ejection, or 
 more rapid ejection, of carbon from the negative leg of the 
 carbon horseshoe. Much has yet to be done before the full 
 meaning of this " Edison effect " is unravelled ; but, as far as 
 it has yet been examined, the following summary of facts 
 includes most of those as yet observed. 
 
 If a platinum wire is sealed through the glass bulb of an 
 ordinary carbon filament lamp, and carries at its extremity a 
 metal plate so placed as to stand up between the legs of the 
 carbon horseshoe without touching either of them, then, when 
 the lamp is actuated by a continuous current, it is found 
 that : 
 
 (1.) This insulated metal plate is brought down instantly to 
 the potential of the base of the negative leg of the carbon, and 
 no sensible potential difference exists between the insulated
 
 122 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 metal plate and the negative electrode of the lamps, whether 
 the test be made by a galvanometer, by an electrostatic volt- 
 meter, or by a condenser. 
 
 (2.) The potential difference of the plate and the positive 
 /electrode of the lamp is exactly the same as the working 
 potential difference of the lamp electrodes, provided this is 
 measured electrostatically, i.e., by a condenser, or by an elec- 
 trostatic voltmeter taking no current ; but, if measured by a 
 galvanometer, the potential difference of the plate and the 
 positive electrode of the lamp is something less than that of 
 the working lamp electrodes. 
 
 (3.) This absolute equality of potential between the negative 
 electrode of the lamp and the insulated plate only exists when 
 the carbon filament is in a state of vivid incandescence, and 
 when the insulated plate is not more than an inch or so from 
 the base of the negative leg. When the lamp is at inter- 
 mediate stages of incandescence, or the plate is considerably 
 removed from the base of the negative leg, then the plate is 
 not brought down quite to the same potential as the negative 
 electrode. 
 
 (4.) A galvanometer connected between the insulated plate 
 and the positive electrode of the lamp shows a current increas- 
 ing from zero to four or five milliamperes, as the carbon 
 is raised to its state of commercial incandescence. There is 
 not any current greater than 0-0001 of a milliampere* between 
 the plate and the negative electrode when the lamp has a good 
 vacuum. 
 
 (5.) If the lamp has a bad vacuum this inequality is 
 destroyed, and a sensitive galvanometer shows a current 
 flowing through it when connected between the middle plate 
 and either the positive or negative electrode. 
 
 (6.) When the lamp is actuated by an alternating current, a 
 continuous current is found flowing through a galvanometer 
 * A milliampere is the a - Voth of an ampere. 

 
 ELECTRIC GLOW LAMPS. 123 
 
 connected between the insulated plate and either terminal of 
 the lamp. The direction of the current through the galva- 
 nometer is such as to show that negative electricity is flowing 
 from the plate through the galvanometer to the lamp terminal. 
 This is also the case in (4) ; but, if the lamp has a bad vacuum, 
 then negative electricity flows from the plate through the gal- 
 vanometer to the positive terminal of the lamp, and negative 
 electricity flows to the plate through the galvanometer from 
 the negative terminal of the lamp. 
 
 (7.) The same effects exist, on a reduced scale, when the 
 incandescent conductor is a platinum wire instead of a carbon 
 filament. The platinum wire has to be brought up very near 
 to its point of fusion in order to detect the effect, but it is 
 found that a current flows between the positive electrode of 
 a platinum wire lamp and a platinum plate placed in the 
 vacuum near to the negative end of that wire. 
 -- (8.) The material of which the plate is made is without 
 influence on the result. Platinum, aluminium, and carbon 
 have been indifferently employed. 
 
 (9.) The active agent in producing this effect is the negative 
 leg of the carbon. If the negative leg of the carbon is covered 
 up by enclosing it in a glass tube, this procedure entirely, or 
 nearly entirely, prevents the production of a current in a 
 galvanometer connected between the middle plate and the 
 positive terminal of the lamp. 
 
 (10.) It is a matter of indifference whether a glass or metal 
 tube is employed to cover up the negative leg of the carbon ; 
 in any case this shielding destroys the effect. 
 
 (11.) If, instead of shielding the negative leg of the carbon, 
 a mica screen is interposed between the negative leg and the 
 side of the middle plate which faces it, then the current 
 produced in a galvanometer connected between the positive 
 terminal of the lamp and the middle plate is much reduced. 
 Under the same circumstances hardly any effect is produced
 
 124 ELECTRIC LAMPS AND ELECTEIC LIGHTING. 
 
 when the mica screen is interposed on that side of the metal 
 plate which faces the positive leg of the carbon. 
 
 (12.) The position of the metal plate has a great influence 
 on the magnitude of the current traversing a galvanometer 
 connected between the metal plate and the positive terminal 
 of the lamp. The current is greatest when the insulated 
 metal plate is as near as possible to the base of the negative 
 leg of the carbon, and greatest of all when it is formed into 
 a cylinder which embraces, without touching, the base of the 
 negative leg. The current becomes very small when the 
 insulated metal plate is removed to 4 or 5 inches from the 
 negative leg, and becomes practically zero when the metal 
 plate is at the end of a tube forming part of the bulb, which 
 tube has a bend at right angles in it. Copious experiments 
 have been made with metal plates in all kinds of positions. 
 
 (13.) The galvanometer current is greatly influenced by the 
 surface of the metal plate: being greatly reduced when the 
 surface of the plate is made small, or when the plate is set 
 edgeways to the negative leg, so as to present a very small 
 apparent surface when seen from the negative leg. In a lamp 
 having the usual commercial vacuum, the effect is extremely 
 small when the insulated metal plate is placed at a distance 
 of 18in. from the negative leg, but even then it is just sensible 
 to a very sensitive galvanometer. 
 
 (14.) If a charged condenser has one plate connected to the 
 insulated metal plate, and the other plate connected to any 
 point of the circuit of the incandescent filament, this con- 
 denser is instantly discharged if the positively-charged side of 
 the condenser is connected to the insulated plate and the 
 negative side to the hot filament. If, however, the negative 
 leg of the carbon horseshoe is shielded by a glass tube, this 
 discharging power is much reduced, or altogether removed. 
 
 (15.) If the middle plate consists of a separate carbon loop, 
 which can itself be made incandescent by a separate insulated 

 
 ELECTRIC GLOW LAMPS. 125 
 
 battery, then, when this middle carbon is rendered incandes- 
 cent, and is employed as the metal plate in the above experi- 
 ment, the condenser is discharged when the negatively-charged 
 side of it is connected to the hot middle carbon, the positively- 
 charged side of it being in connection with the principal 
 carbon horseshoe. 
 
 (16.) If this last form of lamp is employed as in (4), the 
 subsidiary carbon loop being used as a middle plate, and a 
 galvanometer being connected between it and either the 
 positive or negative main terminal of the lamp, then, when 
 the subsidiary carbon loop is cold, we get a current through 
 the galvanometer only when it is in connection with the 
 positive main terminal of the lamp ; but, when the subsidiary 
 carbon is made incandescent by a separate insulated battery, 
 we get a current through the galvanometer when it is con-, 
 nected either to the positive or to the negative terminal of 
 the lamp. In the first case the current through the galvano- 
 meter is a negative current flowing from the middle carbon 
 to the positive main terminal, and in the second case it is 
 a negative current from the negative main terminal to the 
 middle subsidiary hot carbon. 
 
 (17.) If a lamp having a metal middle plate held between 
 the legs of the carbon loop has a galvanometer connected 
 between the negative main terminal of the lamp and this 
 middle plate, we find that, when the carbon is incandescent, 
 there is no sensible current flowing through the galvano- 
 meter. The vacuous space between the middle plate and 
 the hot negative leg of the carbon possesses, however, a 
 curious unilateral conductivity. If a single galvanic cell is 
 inserted in series with the galvanometer, we find that this 
 cell can send a current deflecting the galvanometer when 
 its negative pole is in connection with the negative main 
 terminal of the lamp, but if its positive pole is in connec- 
 tion with the negative terminal of the lamp, then no current
 
 126 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 flows. The cell is thus able to force a current through the 
 vacuous space when the direction of the cell is such as to 
 cause negative electricity to flow across the vacuous space 
 from the hot carbon to the cooler rnetal plate, but not in 
 the reverse direction. 
 
 (18.) If a vacuum tube is constructed, having horseshoe 
 carbon filaments sealed into it at each end, and which can 
 each be made separately incandescent by an insulated battery, 
 we find that such a vacuum tube, though requiring an electro- 
 motive force of many thousands of volts to force a current 
 through it when the carbon loops are used as electrodes 
 and are cold, will yet pass the current from a single gal- 
 vanic cell when the carbon loop which forms the negative 
 electrode is rendered incandescent. It is thus found that 
 a high vacuum terminated electrically by unequally -heated 
 carbon electrodes possesses a unilateral conductivity, and 
 that electric discharge takes place freely through it under 
 an electromotive force of a few volts when the negative 
 electrode is made highly incandescent.
 
 LECTURE III. 
 
 THE Forms of Electric Discharge : Brush, Glow, Spark, Arc. Vacuum 
 Tubes. Sparking Distance. The Electric Arc. The Optical Pro- 
 jection of the Arc. The Arc a Flexible Conductor. The High 
 Temperature of the Arc. Non-Arcing Metals Lightning Protectors. 
 The Distribution of Light from the Arc. Continuous and Alter- 
 nating Current Arcs. Voltage Required to Produce an Arc. The 
 Physical Actions in the Arc. The Changes in the Carbons. The 
 Distribution of a Potential in the Arc. The Unilateral Conductivity 
 of the Arc. The Temperature of the Crater. Comparison with Solar 
 Temperature. The "Watts per Candle" of the Sun. Intrinsic 
 Brightness and Dissipative Power of Heated Surfaces. Comparison 
 of Glow Lamp. Arc Lamp, and Sun, in respect of Brightness and 
 Radiation. Arc Lamp Mechanism. Arc Lamp Carbons. The Hissing 
 of Arc Lamps. The Application of 
 Arc Lamps. Inverted Arcs. Series 
 and Parallel Arc Lighting. 
 
 E must now forsake the study of 
 the incandescent electric lamp, 
 and turn our attention to the 
 older form of electric illumina- 
 tion, namely, the electric arc 
 lamp. There are three principal 
 ways in which electric discharge 
 seems to be capable of taking place across a space filled with 
 air or gas. These are: firstly, by the glow discharge; secondly, 
 by an electric spark, or disruptive discharge ; and thirdly, by 
 the electric arc. When two conductors which are at diffe- 
 rent electric pressures are brought within a certain distance 
 of one another, and when the electric pressure difference
 
 128 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 between these conductors is gradually raised, one of the 
 above three modes of electric discharge is always established, 
 and takes place in a manner depending upon the gaseous 
 pressure of the air or gas in which the conductors are 
 immersed, upon the electric pressure difference between the con- 
 ductors (caUed the electrodes), and upon the strength of the 
 current which the source of the electric pressure can pro- 
 duce. If the two conductors are, say, two brass balls which 
 are connected with the terminals of an ordinary electrical 
 machine, and capable of making a considerable difference of 
 electric pressure between them, but only capable of ^gene- 
 rating a very feeble electric current, the following effects 
 are observed when the balls are brought near one another : 
 In the air, at ordinary pressures, the conductors, if viewed 
 in the dark, will be found to be covered with a violet glow 
 of light, and this is more visible if their ends are small 
 if, for instance, they are the rounded ends of two wires 
 attached to the electrical machine. If one of the con- 
 ductors has a large surface if, for instance, it is a brass 
 plate and the other conductor is not a very good conductor 
 as, for example, a wooden ball or knob then in the dark 
 we find a form of glow discharge taking place between the 
 ball and the plate, which is called a brush discharge. It 
 has a violet, broom-shaped glow of light, and Sir Charles 
 Wheatstone, on examining it by reflection in a rapidly 
 revolving mirror, showed that it consisted of a series of 
 very rapid electric discharges between the air particles. 
 It is generally accompanied by a slight hissing sound. 
 
 This form of brush discharge occurs in nature under some 
 conditions, and is known by the name of St. Elmo's Fire. 
 In the high Alps, when in the neighbourhood of a thunder- 
 storm, travellers have often observed these whizzing brushes 
 of purple light proceeding from their axes and alpenstocks, 
 and from sharp-pointed rocks; and a similar phenomenon
 
 ELECTRIC ABC LAMPS. 129 
 
 has been seen proceeding from the masts and yardarms 
 of ships under certain electrical conditions of the atmo- 
 sphere. When the knob or ball is positively electrified the 
 electric brush is larger and more brilliant than when the 
 ball is negative. The finest brushes are formed in nitrogen 
 gas. On the other hand, if the conductor is very small, 
 and n^ade of metal, then the glow that is seen on the end 
 of this conductor in the dark is not intermittent, but is 
 associated with a current of air proceeding from the con- 
 ductor. A blunt metal point attached to an electrical machine 
 can thus be made to blow out a candle flame. 
 
 The form and nature of the electric discharge in rarefied 
 gases has been studied by many observers. The phenomena 
 are exceedingly complicated, but generally speaking are 
 somewhat as follows : Into the extremities of a closed glass 
 tube or bulb are sealed platinum wires, which can be con- 
 nected to an electrical machine or battery or any source 
 of electric current. If the tube is exhausted of its air, or 
 filled with highly rarefied gas of any kind, and an electric 
 current is passed through it, the following facts can be noticed : 
 A glow of light surrounds each end of the platinum wire, 
 and, if the air pressure is gradually diminished until 
 the air is rarefied to about -ro-oth part of its ordinary 
 pressure, the end of the negative electrode or conductor is 
 seen to be surrounded by a bluish violet light separated 
 from the electrode by a dark space which increases in width 
 as the rarefaction of the air is increased. The positive 
 conductor, or electrode, is also surrounded with a glow of 
 light, frequently of a different colour, and the space in 
 between may be filled up with a glow of light which may 
 or may not be cut up by a series of dark spaces, and is then 
 said to be stratified. This phenomena of stratified discharge 
 can be shown to you very beautifully by connecting the 
 terminals of an induction coil to two platinum wires, which
 
 130 ELECTRIC LAMPS AND ELECTEIC LIGHTING. 
 
 are sealed into the ends of a long 
 glass tube filled with rarefied car- 
 bonic acid (see Fig. 55). This tube 
 is one of a large collection which 
 belonged to the late Mr. Warren 
 de la Eue, and was presented to 
 this Institution. You will see that 
 the space in between the platinum 
 wires is filled with a light-coloured 
 violet glow, which is stratified or 
 cut up into a number of saucer- 
 shaped layers of light. During 
 this form of discharge the nega- 
 tive electrode, or wire by which 
 the current leaves the tube, is found 
 to become hot, and becomes, in 
 time, raised to a red heat ; whilst, at 
 the same time, it is more or less dis- 
 integrated, depositing a film of metal 
 upon the glass in its neighbourhood. 
 
 This glow discharge in gases is 
 a very complex and yet attractive 
 subject, and has engaged the atten- 
 tion of numerous able physicists, 
 such as Faraday, Pliicker, De la 
 Eive, Hittorf, Spottiswoode, Moul- 
 ton, Crookes, and, more lately, 
 J. J. Thomson and many others. 
 It is found that this phenomenon 
 of discharge, both in rarefied gases 
 and in gases at ordinary pressure, 
 can be produced not only by the cur- 
 rent of electricity supplied by an 
 ordinary statical electrical machine
 
 ELECTRIC ABC LAMPS. 131 
 
 or by an induction coil, but also by the current given by a bat- 
 tery or series of galvanic cells. When the electric discharge takes 
 place in air at ordinary pressures, and when the supply of cur- 
 rent is not very rapid, and especially if one of the conductors is 
 a poor conductor of considerable surface, the discharge takes 
 the form of a series of electric sparks. The form of this spark 
 and the distances over which it can be produced are very much 
 determined by the nature of the surfaces. Bounded metallic 
 knobs give bright snapping sparks, whereas metallic points 
 give thin whizzy sparks. Besides the form of surface, the 
 interposed gases affect the result. Faraday found that 
 different gases have different restraining powers upon the 
 spark. At the same distance of conductors, hydrogen gas 
 permits a discharge through it more easily that is, at a lower 
 electric pressure than air, whereas some other gases, such as 
 hydrochloric acid gas, have peculiar restraining powers. An 
 electric spark is in reality a brief current of electricity taking 
 place between the two conductors, during which time the air 
 or other gas between them is rendered incandescent and 
 therefore luminous, and portions of the material of the con- 
 ductors between which the discharge is taking place are 
 carried across from one side to the other. For each particular 
 pressure there is a certain sparking distance, which, however, 
 is partly determined by the form of the conductors between 
 which the sparks jump. By the term " sparking distance," 
 corresponding to any electric pressure, is meant the distance 
 over which the spark will spring when the difference of 
 electric pressure between the conductors is gradually raised to 
 that value. If we bring two brass balls, connected with the 
 terminals of an electrical machine, within a certain distance of 
 one another, and then darken the room, you will see these 
 two balls surrounded with a glow of light. On bringing them 
 within a certain lesser distance of one another a snapping 
 spark passes, which is repeated at intervals as the machine is 
 worked.
 
 132 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 The first accurate experiments on the sparking distance 
 corresponding to different electric pressures were made by 
 Lord Kelvin. He showed that between slightly curved metal 
 plates it required a pressure of about 5,000 volts to jump 
 across the eighth part of a centimetre, or the twentieth 
 part of an inch. Mr. De la Kue found that between flat 
 plates the difference of pressure required to produce a spark 
 was 8,000 volts when the surfaces were a fifth part of a 
 centimetre, or yf^ths of an inch, apart; and 11,300 volts 
 when the surfaces were one-third of a centimetre, or ^ths 
 of an inch, apart. Similar experiments have been made by 
 numerous other physicists with like results. 
 
 The following table of results for sparking distances under 
 certain conditions has been calculated from the observations 
 of MM. Bichat and Blondlot. If metal balls ^ths of an 
 inch in diameter (one centimetre) are separated by a distance 
 beginning at ^jth of an inch, and increasing by equal 
 amounts viz., by one millimetre at a time the electric 
 pressure difference in volts which will just make a spark 
 jump across at the various distances will be as follows (the 
 distances are given in millimetres ; one millimetre is nearly 
 ^th of an inch) : 
 
 Distances of 
 the balls in 
 millimetres. 
 
 Sparking 
 pressure in 
 volts. 
 
 Distances of 
 the balls in 
 millimetres. 
 
 Sparking 
 pressure iu 
 volts. 
 
 1 
 
 4,765 
 
 12 
 
 27.024 
 
 2 
 
 8,140 
 
 13 27.'765 
 
 3 
 
 11,307 
 
 14 
 
 28,359 
 
 4 
 
 14,119 
 
 15 
 
 28,949 
 
 5 
 
 16,664 
 
 16 
 
 29,363 
 
 6 
 
 19,210 
 
 17 
 
 29,837 
 
 7 
 8 
 
 21,823 
 22,792 
 
 18 
 19 
 
 30,133 
 30,547 
 
 9 
 
 24,153 
 
 20 
 
 30.932 
 
 10 
 
 25,071 
 
 21 31.198 
 
 11 
 
 26,255 
 
 22 
 
 31,494
 
 ELECTEIC ARC LAMPS. 133 
 
 Hence we see that, between such metallic balls, an electric 
 pressure of 30,000 or 40,000 volts is required to make a spark 
 jump over a distance of about one inch. It has, however, 
 been shown that these sparking distances are greatly affected 
 by the state of polish of the surfaces and by their form, and 
 that sparking is promoted by light of particular kinds falling 
 on the balls. The light from other electric sparks, as also that 
 from burning magnesium wire, has a strong effect in breaking 
 down the insulation of the air and promoting the passage of a 
 spark between polished metal balls across a distance which it 
 would otherwise not pass. Although an electric spark of a 
 few inches in length represents an electric pressure of many 
 thousands of volts, it represents very little current or quantity. 
 It is somewhat surprising to see a torrent of noisy sparks 
 passing between the discharger of an electrical machine, and 
 yet to know that this represents in quantity, probably, not a 
 thousandth of an ampere of electric current, and that it could 
 hardly effect the chemical decomposition of one drop of water 
 or solution of a metallic salt, although its pressure or potential 
 may be hundreds or thousands of volts. An electric pressure 
 of 100 volts, such as is used for incandescent lamps, will 
 hardly produce a spark over any visible distance ; certainly the 
 distance would be less than T ^th part of an inch. 
 
 These various forms of discharge the spark, the brush, 
 and the glow discharge have all been the subject of 
 elaborate investigations. Faraday particularly studied them 
 in the twelfth and thirteenth series of his " Experimental 
 Eesearches in Electricity." Whatever be the means by 
 which the difference of electric pressure is produced, it is 
 found that, if there exists a sufficient resistance in the 
 circuit limiting the rate at which the supply of electricity 
 can take place, then no other form than the spark, brush, 
 or glow discharge is possible. If, however, we employ 
 a primary or secondary battery that is, a collection of
 
 134 ELECTRIC LAMPS AND ELECTEIC LIGHTING. 
 
 galvanic cells for producing the difl'erence of pressure 
 between the two conductors, such an arrangement differs 
 from an ordinary electrical machine only in the fact that 
 it can supply a stronger or more powerful electric current. 
 The galvanic battery, when compared with an electrical 
 machine such as the Wimshurst machine, must be thought 
 of as a very large pump compared with a very small one. 
 Both these pumps can create a difference of level, pumping 
 up water from one level to a cistern or reservoir at a higher 
 level, but the pump of larger capacity differs from the other 
 in that it can supply faster and can keep up the difference 
 of level in spite of out-flow of water. Accordingly, if a 
 large number of galvanic cells are joined together, and the 
 ends of this battery are connected to two brass balls, it is 
 found that, if the battery is one which has a high internal 
 resistance, it will produce a series of small sparks, which 
 jump over continually from one ball to the other, provided 
 the difference of pressure is great enough and the balls are 
 brought near enough together. "When the battery is one 
 of low internal resistance, then the spark discharge cannot 
 be maintained ; but if the conductors are brought near 
 enough together, and the pressure sufficiently raised, the 
 spark or glow discharge passes spontaneously and immedi- 
 ately into a third form of discharge, which is caUed the 
 electric arc. 
 
 It is possible that this arc discharge was first observed 
 by Curtet in 1802, one year after the invention of the gal- 
 vanic battery by Volta ; but the first careful study of the 
 phenomena of the electric arc in air and in vacuum was 
 made by Sir Humphry Davy at the beginning of this 
 century. At that time Sir Humphry Davy was engaged 
 in the remarkable series of electro - chemical discoveries 
 which resulted in the production of metallic potassium and 
 sodium from caustic potash and soda for the first time. 

 
 ELECTRIC ARC LAMPS. 135 
 
 Finding the necessity for a larger battery than he possessed, 
 he laid a request before the managers of the Eoyal Institu- 
 tion in 1808, asking them to provide the means of con- 
 structing a battery of 2,000 cells with which to continue these 
 researches.* Shortly afterwards this battery was provided, 
 and with it Davy not only continued his electro- chemical 
 discoveries, but studied the phenomena of the electric arc. 
 Connecting the ends of this large battery of 2,000 pairs 
 of plates to two pieces of hard charcoal, which had been 
 heated and then plunged into quicksilver to make them 
 better conductors, Davy observed for the first time on a 
 large scale the phenomena of the electric arc. In his 
 " Elements of Chemical Philosophy " (Vol. IV.) he thus 
 describes the first production of this very large 2,000-volt 
 electric arc : 
 
 " The most powerful combination (battery) that exists in 
 which number of alternations (plates) is combined with 
 extent of surface is that constructed by the subscriptions 
 of a few zealous cultivators and patrons of science in the 
 laboratory of the Royal Institution. It consists of two 
 hundred instruments connected together in regular order, 
 each composed of ten double plates arranged in cells of 
 porcelain and containing in each plate 32 sq. in. ; so that 
 the whole number of double plates is 2,000 and the whole 
 surface 128,000 sq. in. This battery, when the cells are 
 filled with 60 parts of water mixed with one part of nitric 
 
 * Sir Humphry Davy laid a request before the managers of the Royal 
 Institution on July 11, 1808, that they would set on foot a subscription 
 for the purchase of a large galvanic battery. The result of this suggestion 
 was that a galvanic battery of 2,000 pairs of copper and zinc plates was 
 set up in the Royal Institution, and one of the earliest experiments per- 
 formed with it was the production of the electric arc between carbon poles, 
 on a large scale. It is probable, however, that Davy had produced the 
 light on a small scale some six years before, and, according to Quetelet 
 Curtet, observed the arc between carbon points in 1802. Sec Dr. Paris' 
 " Life of Sir H. Davy."
 
 136 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 acid and one of sulphuric acid, afforded a series of brilliant 
 and impressive effects. "When pieces of charcoal about an 
 inch long and one-sixth of an inch in diameter were brought 
 near each other (within the thirtieth or fortieth part of an 
 inch) a bright spark was produced and more than half the 
 volume of 4he charcoal became ignited to whiteness, and 
 by withdrawing the points from each other a constant 
 discharge took place through the heated air in a space 
 equal at least to 4in., producing a most brilliant ascend- 
 ing arch of light, broad and conical in form in the middle. 
 Fragments of diamond and points of charcoal and plumbago 
 rapidly disappeared and seemed to evaporate hi it, but there 
 was no evidence of their having previously undergone 
 fusion." 
 
 We may repeat his experiments on a smaller scale. Taking 
 two pencils of hard carbon formed of gas retort coke, these 
 are connected to the two ends of a battery furnishing a 
 pressure of about 50 volts. If these two carbons are brought 
 within a short distance of one another, say y^th of an 
 inch, about the thickness of a thin visiting card, it will be 
 found that this electric pressure of 50 volts is not sufficient 
 to create a spark capable of jumping over this distance. The 
 very thin layer of air existing between the two carbon ter- 
 minals is sufficient to insulate entirely this pressure. If the 
 two carbons are touched together at the point of contact 
 they immediately become red hot, and then, on separating 
 them again a slight distance, the electric discharge continues 
 to take place between them, and this is called the electric arc. 
 Instead of bringing the carbons in contact we can break 
 down the insulation of the air by passing a small electric 
 spark from one carbon to the other, or even at some distance 
 away. This can be done by taking a discharge from a smaU 
 electrical machine, and when the experiment is performed 
 before you you will notice that the electric arc discharge 

 
 ELECTRIC ARC LAMPS. 137 
 
 follows immediately upon the passage of a spark which is 
 too small to be visible. 
 
 A very convenient arrangement for producing and examin- 
 ing an electric arc is one in which one of the carbon rods is 
 fixed and the other is moved by a rack and pinion, so as to be 
 brought in contact with the first, and then moved away as 
 required when the arc is produced between them. This 
 arrangement, one form of which, as devised by Mr. Davenport,* 
 is shown in Fig. 56, is needed hi an experimental study of 
 the arc. 
 
 FIG. 56. The Davenport Arc Lamp. 
 
 In order to view more comfortably the whole phenomena 
 that are taking place in the interior of this dazzling source 
 of light, we will project an image of the electric arc by 
 means of a lens upon the screen, and examine the whole 
 of the effects taking place in it. The arc discharge is most 
 easily obtained and maintained between conductors which 
 are good but not very good conductors, and which are only 
 
 * This convenient form of hand-regulated arc lamp was described in 
 The Electrician, Vol. XXXIL, p. 393.
 
 138 ELECTRIC LAMPS AND ELECTEIC LIGHTING. 
 
 volatilised at a high temperature. As we shall see presently, 
 it cannot be maintained at all between certain metals when 
 close together. No substance has been found superior for 
 the purpose of producing an electric arc to a dense variety of 
 carbon, produced either from the graphitic deposit of carbon 
 found in the interior of gas coke ovens, or by the production 
 of a hard variety of carbon from gas coke, lamp-black, or 
 sugar. Twenty-five years ago the carbon rods which were 
 used for this purpose were generally produced by sawing 
 lumps of very hard retort carbon into small rods a quarter 
 of an inch square, but of late years an immense manufacture 
 has sprung up in the manufacture of arc-light carbon rods. 
 In the manufacture of these rods some form of dense and 
 very pure carbon is made into a paste with syrup or coal tar, 
 and from this rods are pressed out which, after being dried 
 and baked at a high temperature, furnish carbon rods of 
 various sizes which are used for the production of the electric 
 arc. These carbon rods are generally sold in lengths, varying 
 in size from a quarter of an inch to an inch in diameter, and 
 from nine to eighteen inches in length. Taking two of these 
 rods, and producing between them an electric arc in a closed 
 lantern, we have now projected upon the screen an enlarged 
 image of the electr-ic arc. 
 
 Four things can be noticed in this electric arc as soon as it 
 has been formed by bringing the carbons into contact (see 
 Fig. 57). Looking at the optical image of the arc projected 
 upon the screen, we notice that both the ends of the two 
 carbons are brilliantly incandescent, but that, if the arc is 
 formed by a current of electricity flowing always in one 
 direction, called a continuous current arc, then the positive 
 carbon, or the one attached to the positive pole of the battery 
 or dynamo, and marked + in the illustration, is more 
 brilliantly incandescent than the other. This positive pole 
 soon has its end hollowed out into a cup-shaped depression,
 
 ELECTRIC ARC LAMPS. 
 
 139 
 
 which is called a crater, and this positive or crater carbon, 
 after being in use for a few minutes, will be found to 
 have its extremity converted into graphitic carbon or plum- 
 bago, which will mark upon paper like a pencil, in a 
 way that it would not do before being so used. At the 
 
 FIG. 57. The Electric Arc. 
 
 same time the negative carbon appears to get more pointed, 
 and becomes surrounded a little below its tip with small 
 globules, which are, in all probability, condensed carbon 
 vapour. If the arc is maintained for some time, it will be
 
 140 ELECTRIC LAMPS AND ELECTE1C LIGHTING. 
 
 found that both these carbons are being worn away ; but the 
 positive or crater carbon wears away about twice as fast as 
 the other when the arc is formed in air. In the space 
 between the two carbons we notice a band of brilliant violet 
 light, which is surrounded by a less luminous aureole of a 
 golden colour. In a carefully-regulated, steady electric arc it 
 will be seen that the violet -coloured portion seems to spring 
 from the white-hot crater, and that a well-marked dark space 
 separates this purple part from the golden-coloured, wing- 
 shaped flames which appear to spring from the negative 
 carbon. This intermediate central portion is called the true 
 arc, and hence we have four separate portions of the arc 
 to consider: namely, the crater, the violet-coloured core of 
 the arc, the aureole, and the negative carbon. On looking 
 closely at an arc so projected upon the screen, it will be 
 seen that little bits of carbon are continually becoming 
 detached from one pole, and they immediately fly over to 
 the opposite carbon. It can be shown by several experiments 
 that there is probably a double transport of material going on 
 in the arc. The material of the positive pole is being carried 
 across to the negative side, and the material of the negative 
 pole is being carried across to the positive side. 
 
 At this stage it will be interesting to analyse the light of 
 the arc by means of the prism. Placing a vertical slit before 
 the lens which is projecting the image of the arc upon the 
 screen, we cut off all but a linear band of light, the upper 
 part of which proceeds from.the crater carbon, the middle part 
 from the violet core of the arc, and the bottom part from the 
 negative carbon. The prism being then placed in front of the 
 lens, we expand this linear image of the slit into a spectrum 
 which is divided longitudinally into three parts. The upper 
 portion, you will observe, is a brilliant continuous spectrum of 
 light, in which all the prismatic rays are present. This spec- 
 trum proceeds from the light emitted from the crater of the arc.
 
 ELEGTEIG AEG LAMPS. 141 
 
 The middle portion of the spectrum is much less bright 
 in the orange, red, yellow and green, but, as you will 
 see, is distinguished by two remarkably brilliant violet bands, 
 which are characteristic bands in the spectrum of carbon 
 vapour. The bottom of the spectrum, formed from the 
 negative carbon, is a continuous spectrum similar to the 
 upper one, only less brilliant. On lengthening out the arc 
 by drawing the carbons apart, we find that the two upper 
 spectra move away from one another, the middle spectrum 
 increasing in width. At a certain point the arc breaks down, 
 the middle spectrum with its two violet bands disappearing 
 instantly ; but for a short period of time the upper and 
 under spectra linger as the carbons cool, because the carbons 
 continue to glow faintly ; and as these carbons cool you 
 observe the two spectra shrinking rapidly from the violet 
 end, until at last only faint traces of the red and green 
 remain, and these finally vanish. But you will see that 
 the lower spectrum, which proceeds from. the light of the 
 negative carbon, vanishes first. This only indicates that 
 the crater carbon has a much higher temperature than the 
 negative carbon, and is in accordance with all that we 
 know about the arc. 
 
 If we examine an electric arc of this kind, we notice that, 
 as the carbons are gradually drawn apart and the arc length- 
 ened, the aureole of golden vapour increases and forms a sort 
 of large lambent flame, which disappears when the arc is 
 lengthened beyond a certain point. After such a continuous 
 current arc has been extinguished, we can always tell which 
 has been the positive carbon, not only from the crater hol- 
 lowed in it, but also from the fact that it remains red hot for 
 a longer time than the negative carbon. The exceedingly high 
 temperature of the positive carbon is shown by the fact, above 
 mentioned, that after being used for some time it is converted 
 into graphite at the extremity and will mark paper.
 
 142 ELECTEIC LAMPS AND ELECTRIC LIGHTING. 
 
 The electrical power taken up in the continuous current 
 electric arc is measured in an exactly similar way to the power 
 taken up and dissipated in the filament of an incandescent lamp 
 namely, by measuring the current in amperes flowing through 
 the arc, and the difference of electric pressure in volts between 
 the carbons. Multiplying these numbers together, we obtain 
 the value of the power, measured in watts, being absorbed in 
 the arc. By far the best way to denominate electric arcs 
 is by the power in watts absorbed. We may thus speak 
 of a 300- watt arc, or a 500-watt arc, or of a 1,000-watt arc. 
 The length of the arc can be measured most conveniently 
 by projecting the image of the carbons upon the screen. 
 If a sheet of polished metal is then held behind the arc, a 
 faint diffused light will be thrown upon the screen, which 
 will enable us not only to see the arc, but to see also the 
 image of the two carbon rods. By means of a pair of com- 
 passes or a foot rule we can measure the length of the 
 magnified image of the arc that is to say, the distance 
 between the image of the tips of the carbon rods. We can 
 also measure the width of the magnified image of either of 
 the carbon rods. Let us suppose that the rods are each of 
 them half an inch in diameter, and that their image upon 
 the screen is eight inches in diameter ; then the arc has 
 been magnified 16 times. If, then, we find that the image 
 of the arc is, say, two inches in length when measured upon 
 the screen, it shows us that the real length of the arc is 
 T Vth part of this, namely, one-eighth of an inch, because 
 the length of the arc is magnified in the same ratio as 
 the width of the carbons. This optical method of measur- 
 ing the length of the arc affords a very easy means of 
 accurately ascertaining the length of the arc without danger 
 or difficulty. Electric arcs may, roughly, be distinguished as 
 long arcs and short arcs, according to the distance between 
 the carbon poles, although the dividing line between the 
 two is not in any way accurately determined.
 
 ELECTRIC ARC LAMPS. 143 
 
 Before passing on to consider the physical measurements 
 connected with the electric arc, let us notice that the violet 
 core of the arc, which consists of a torrent of incandescent 
 carbon vapour, acts like a perfectly flexible conductor. If I 
 bring a magnet near to such an electric arc, you will notice 
 that the magnet apparently impels the arc sideways, and if it 
 is brought near enough it will actually blow it out. The 
 reason for this is because the arc, as a conductor traversed by a 
 current, exerts an electro-magnetic force upon the magnet, 
 and in like manner the magnet exerts an electro-magnetic 
 force upon the arc. The arc always tends to move across the 
 lines of force of the magnet, and in this way may actually 
 be bent into a bow shape, or be drawn out into a kind of blow- 
 pipe flame. 
 
 Brief references must now be made to the generation of 
 light and heat in the arc. Experiment seems to show that in 
 the continuous current arc that is, the arc produced by a 
 current of electricity always flowing continuously in the 
 same direction by far the larger proportion, perhaps 85 
 per cent., of the light thrown out comes from the highly 
 incandescent crater carbon, about 15 per cent, comes from 
 the negative carbon, and very little indeed, certainly not 
 more than 5 per cent., from the true electric arc. The great 
 source of light, therefore, in the electric arc is the intensely 
 luminous crater formed at the end of the positive carbon, which 
 is brought up to an exceedingly high temperature. Various 
 estimates have been made of the temperature of this electric 
 crater, but the physical difficulties connected with these 
 experiments are very great. Eossetti in 1879 stated, as the 
 result of some of his experiments, that the positive carbon 
 had a temperature of 3,200C., and the negative carbon 
 2,500C. More recently, Violle has given a temperature 
 of 3,500 D C. as the temperature of the crater carbon, and 
 other observers have given even higher results than these.
 
 144 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 This temperature is sufficient not only to melt but actually 
 to boil away even the most refractory metals. We can easily 
 show the high temperature of the crater by boiling in it 
 a piece of copper or silver. If an electric arc is formed with 
 the crater downwards, and a small piece of copper is placed in 
 the crater, and then the image of the arc is projected upon 
 the screen, you will see the little piece of copper boiling 
 violently and sending out a torrent of copper vapour, which, 
 in its incandescent condition, emits a brilliant green light. In 
 the same way, silver, platinum, and even the refractory metal 
 iridium, can be melted and boiled into vapour in the electric arc. 
 It has been already mentioned that there is a transport of 
 material in the arc, material passing over from the positive to 
 the negative pole, and vice versd. It is only natural, therefore, 
 to expect that, since the electric arc is a phenomenon of electric 
 discharge taking place in a highly-conducting vapour, the 
 longest arcs will be capable of being maintained between the 
 most volatile metals, and experience shows this to be the case 
 generally. It has been found that the difference of tempera- 
 ture between the two poles is greater in proportion as they 
 are worse conductors and more easy of disaggregation ; also 
 that, to produce an arc of a given length requires a greater 
 current in proportion as the infusibility of the electrode is 
 greater. 
 
 It is, however, a very striking fact, as discovered by 
 Wurts, that there are four metals namely, zinc, cadmium, 
 magnesium and mercury which are called non-arcing metals, 
 and between which it is difficult or impossible to maintain 
 the electric arc, if the surfaces of these metals are very near 
 together. This fact is taken advantage of in the production 
 of a very ingenious lightning protector. 
 
 When overhead circuits, consisting of copper wires carried 
 on insulators on posts, are employed for the distribution of
 
 ELECTRIC ARC LAUl'it. 145 
 
 electric currents, such overhead lines are very much exposed 
 to lightning stroke. If they should be struck by lightning, 
 the lightning discharge will generally pass back into the 
 generating station, and do damage to the dynamo machines. 
 In order to prevent this, a lightning protector of the 
 following kind has been designed, especially for use with over- 
 head alternating-current circuits, the object of which is to 
 allow the lightning striking the overhead circuits to pass to 
 earth, but yet at the same time to prevent the current from 
 the dynamo machine following it. On a slab of marble or 
 porcelain are placed an odd number of small zinc cylinders, 
 one inch in diameter and about three inches long. These 
 cylinders are placed close to one another, being separated 
 by a distance only of -g^th of an inch. Each cylinder is 
 insulated. The middle cylinder is connected by a wire with 
 a plate sunk in the earth. The two outside cylinders are 
 connected to the two lines which form the overhead circuit. 
 If the lightning strikes the line on either side, the discharge 
 jumps over from cylinder to cylinder until it reaches the 
 .middle cylinder, and then it goes to earth by the earth-plate. 
 In so doing it will attempt to start an electric arc between the 
 cylinders ; but, as a permanent arc cannot be maintained 
 between zinc surfaces, this arc is almost instantaneously ex- 
 tinguished, and the dynamo machine, therefore, is unable to 
 produce an electric arc across between the two mains. In 
 this way a safe passage for the lightning to earth is provided. 
 
 It will, in the next place, be necessary to direct attention 
 to the distribution of light from an electric arc. On 
 examining an electric arc formed by a continuous current, 
 the positive or crater carbon being uppermost, a very cursory 
 examination shows us that the light sent out from such 
 an arc is not uniform in all directions. Very little light is 
 sent upwards. By far the larger portion of the light is sent 
 downwards at an angle of from 30 deg. to 40 deg. below
 
 146 ELECTRIC LAMPS AXD ELECTRIC LIGHTING. 
 
 the horizon. In the light of explanations given above, it is 
 very easy to see the reason for this. It is really due to the 
 fact that the greater portion of the light is sent out from the 
 bottom surface of the top carbon ; hence, when viewed in a 
 horizontal position, but very little of this incandescent crater 
 can be seen. On elevating the arc, so as to look up underneath 
 it, the amount of apparent surface of the crater which is seen 
 increases up to a certain position as we elevate the arc. But 
 if the arc is elevated beyond a certain amount above the eye, 
 then the bottom or negative carbon begins to stand in the way of 
 the light sent out from the crater, and to diminish the apparent 
 
 FIG. 58. Model showing the Relati%-e Areas of Crater seen at different 
 angles of view. 
 
 area. By taking two cylinders of pasteboard, N and P, held 
 by threads a little distance from one another (nee Fig. 58) and 
 fixing on the underneath end of the upper cylinder a disc of 
 red paper to represent the incandescent crater, we may easily 
 convince ourselves, by looking at this model arc in different 
 directions, that there is a certain position in which the eye can 
 be placed in which it will see the largest apparent area of the 
 crater. 
 
 There is, therefore, a position in which the maximum 
 amount of light will be thrown out by a real arc represented
 
 ELECTRIC ARC LAMPS. 147 
 
 by this model. Accordingly, if the intensity of the light 
 given by an electric arc is measured in different directions, 
 say for every ten degrees of inclination above and below 
 the horizontal, we find different photometric values for 
 these rays in these directions. We can represent this 
 different photometric intensity in different directions by a 
 curve (see Fig. 59), which is called the photometric curve of 
 the arc. The magnitude of the radii of this curve, taken in 
 different directions, represent the intensity of the light pro- 
 
 .--60 
 
 FIG. 59. Curve Showing the Distribution of Light from a Continuous 
 Current Arc. The figures printed on the radial lines represent the candle- 
 power in different directions. 
 
 ceeding from the arc in these directions. The exact position 
 of the maximum intensity will depend upon the length of the 
 arc, the size of the crater, and the relative widths of the nega- 
 tive and positive carbons. Accordingly, in street arc lighting 
 advantage is taken of this fact. If the electric arcs are 
 put up on high poles, or street supports, the positive or 
 crater carbon being upermost, they will throw down their 
 
 L 2
 
 148 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 maximum amount of light in a direction inclined at 40 or 
 50 degrees to the horizon, and illuminate a circular area 
 around them with a maximum intensity. Just underneath 
 the arc there will be more or less shadow, and in a direction 
 above the horizon very little light will be thrown out. 
 
 In order to obtain the best results and the most uniform 
 distribution of light, Mr. Crompton has found it best to use a 
 very small negative carbon, in order that the negative carbon 
 may stand as little as possible in the way of the light sent out 
 from the crater in a downward direction, and that the arc may, 
 therefore, illuminate the largest possible area underneath it. 
 This unequal distribution of the light from the arc can, to a 
 great extent, be modified by the use of a somewhat peculiar 
 glass globe ; but we shall point out in an instant a method by 
 which superior results are gained for internal lighting. 
 
 Not only can we form an electric arc by means of a continuous 
 current of electricity that is to say, one always flowing in the 
 same direction but we can employ an alternating current of 
 electricity. In such an alternating current there is a very 
 rapid change in the direction of the electric current, the 
 current flowing first in one direction for a short period, then 
 being arrested, and then flowing in the other direction for an 
 equally short period. An alternating current of electricity 
 may therefore be compared to the ebb and flow of water in a 
 tidal river, in which the water flows in one direction for a 
 certain number of hours and is then reversed and flows in 
 the opposite direction for about an equal period. 
 
 In what are called alternating current systems of electric 
 lighting distribution it is very usual to employ an alter- 
 nating current having a frequency of from 40 to 100 periods 
 per second that is to say, the electric current changes its 
 direction from 80 to 200 times in one second. A continuous
 
 ELECTRIC ARC LAMPS. 149 
 
 current of electricity may be likened, on the other hand, to a 
 non-tidal river in which the water flows always in one direction. 
 We can form an electric arc not only with a continuous but with 
 an alternating current of electricity, and the reason for this is 
 that there is a certain persistence in the electric arc. It is 
 found that, when the arc is made with a continuous current of 
 electricity, the current can be interrupted for a short fraction 
 of a second without putting out the arc. Hence the current 
 of electricity can be reversed in direction without extinguish - 
 
 FIG. 60. Curve Showing the Distribution of Light from an Alternating 
 Current Arc. The figures marked on the the radial lines indicate candle- 
 power in these directions. 
 
 ing the arc. If, however, it is reversed very rapidly, then the 
 result will be that, instead of having one positive and one nega- 
 tive pole or carbon, each carbon will be alternately positive 
 and negative, and, instead of having only one crater, we shall 
 have both carbons cratered out and sending out light. Hence 
 an alternating current arc has a very different distribution of 
 light from a continuous current arc. In an alternating current
 
 150 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 arc the light is not only sent downwards, but it is sent upwards 
 as well, the distribution of light being found to be about 
 equal in directions above and below the horizon. The curve 
 representing the distribution of the light of an alternating arc 
 lamp is shown in Fig. 60, from which it will be seen that the 
 alternating current arc sends its maximum illumination in 
 directions both upwards and downwards equally inclined to 
 the horizon. Objections are sometimes taken to the use 
 of an alternating current arc on the ground that it makes 
 a disagreeable noise or humming. This, however, is much 
 less evident in some arcs than in others, and, by the employ- 
 ment of suitable carbons and a proper frequency of current, 
 can very nearly be abolished. It appears to be proved, how- 
 ever, that an alternating current arc does not yield the same 
 total illumination for a given expenditure of energy in it as 
 does a continuous current arc. 
 
 Returning, then, for the moment, to the continuous current 
 arc, we must note some more physical facts connected with 
 the formation of an electric arc between various terminals. 
 When the arc is formed between carbon poles it is very easy 
 to show that the true electric arc cannot be formed for much 
 less than 35 volts ; that is to say, it is not possible to pro- 
 duce a separation of the two carbons from one another, pro- 
 ducing a true electric arc as opposed to a mere incandescence 
 of the tips of the carbons, unless the pressure difference 
 between the carbons amounts to about the above value. 
 When the electric arc is formed with a greater pressure diff- 
 erence, it is found that there is a certain constant pressure 
 not far from 35 or 39 volts, according to most experimentalists, 
 which does not, as it were, take any part in the production 
 of the current through the arc. If electric arcs of various 
 lengths are formed, and if at the same time we measure 
 the current flowing through the arc and the difference of 
 pressure between the carbons, it has been found by various 

 
 ELECTRIC ARC LAMPS. 151 
 
 observers that the result can always be expressed in the follow- 
 ing way : Let the difference of pressure between the carbons, 
 measured in volts, be called the working volts of the lamp ; 
 then the length of the arc is found to be proportional to the 
 working volts diminished by a certain constant amount, and 
 to be inversely proportional to the current flowing through 
 the arc. The question then arises, What is this constant 
 pressure which is first to be produced before any true electric 
 arc can be formed ? 
 
 It has been maintained by some that the electric arc 
 acts as if it contained a source of electric pressure which 
 operates against the external electro -motive force producing 
 an arc. There are various reasons, however, why this 
 position is untenable, and there are also arguments supporting 
 the view that this constant pressure, which must be first 
 applied before any true arc can be formed, represents one 
 factor in the measurement of the physical work that must 
 be done in order to volatilise the carbon in the crater, 
 and to maintain that production of carbon vapour. The 
 view now generally held is that the temperature in the 
 interior of the crater of the electric arc is the temperature 
 of boiling carbon, or of the material of the poles between 
 which the electric arc is formed. This theory receives 
 support, amongst other results, from the experiments of 
 Victor von Lang. In his experiments he formed electric 
 arcs between poles of various metals, and measured this 
 constant pressure difference below which no arc of measurable 
 length can be maintained. His results are seen in the table 
 below, and it will be at once noticed that the order in which 
 the metals stand as regards the magnitude of this so-called 
 back electromotive force is the order of their infusibility or 
 volatility. Carbon, which is the most infusible and non-volatile, 
 heads the list, and the most fusible and volatile metals, 
 such as cadmium and silver, are found at the bottom. If, then,
 
 152 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 Table nf Electric Pressures or Voltages required to begin an Electric 
 Arc betioeen poles of the materials named. (Victor von Lang.) 
 
 Material used for poles between 
 which the Arc is formed. 
 
 Initial voltage for production of 
 an Arc between these poles. 
 
 
 35 volts. 
 
 
 27 
 
 Iron 
 Nickel 
 
 25 
 26 
 
 
 23 
 
 Silver 
 
 15 
 
 Cadmium 
 
 10 
 
 the temperature of the crater is the temperature of the 
 boiling metal or material between which the electric arc is 
 formed, we should expect such a crater to have a constant 
 temperature and to emit light of a perfectly constant composi- 
 tion. Very good reasons have accordingly been shown by 
 different physicists for believing that, in the carbon arc, the 
 temperature of the crater of the positive carbon has a perfectly 
 constant value, which is that of the boiling point of carbon. 
 
 From the above table of results it will be seen that the most 
 infusible material (carbon) requires the highest voltage to begin 
 a true electric arc, and generally that the order of the initial 
 voltage is the order of infusibility of the materials. It has 
 been shown that the lower the electric conductivity of the 
 material of which the poles are formed, the greater is the 
 difference of temperature between the positive and negative 
 poles when an arc is produced between them. The following, 
 therefore, are the reasons why carbon, in its hard graphitic 
 form, is so peculiarly suitable for use as the poles or electrodes 
 in forming an electric arc : 
 
 1. Tt is a rather poor conductor of electricity (compared 
 with metals). Hence we obtain a great difference of tern- 

 
 ELECTRIC ARC LAMPS. 153 
 
 perature between the poles ; in other words, we gain the 
 advantage of having a highly- heated crater on one pole. 
 
 2. It is easily disintegrated. Hence we obtain a longer 
 arc for a given electric pressure difference between the poles. 
 
 3. It is very infusible, and its oxide is a gas ; and therefore 
 does not form a deposit of oxide in and around the arc. 
 
 4. It is an inferior conductor for heat. Hence the high 
 temperature is localized at the ends of the poles. 
 
 Just as no material has yet been found which is superior to 
 carbon for the incandescing conductor of glow lamps, so no 
 substance has been yet discovered superior to carbon for 
 use in the production of an electric arc. 
 
 Hence we must conclude that the processes at work in the 
 production of an electric arc are somewhat as follows : In the 
 crater, carbon is being boiled into vapour at a temperature 
 probably about 8,500 and 4,000 centigrade. On looking 
 carefully at a magnified image of the crater projected by a 
 lens on to white paper, we can see an apparent seething and 
 effervescence of the surface of the intensely white hot floor of 
 the crater. This torrent of carbon vapour passing towards 
 the negative pole forms the violet core of the arc, the violet 
 colour being that of the vividly-incandescent carbon vapour, 
 possibly superheated to a higher temperature than the boiling 
 point of carbon by the passage of the electric discharge through 
 it. At the cooler negative pole some of the carbon vapour is 
 condensed ; but a portion of the torrent of carbon molecules 
 carried across is deflected back and returns to the positive pole, 
 causing the golden aureole or flame, and creating thus a double 
 carbon current in the arc. The negative carbon gives evidence 
 after use of having been worn away by a kind of sand-blast 
 action, and is so worn away into a series of terraces. The 
 negative carbon also undergoes a curious series of changes, by 
 which a point or knob of carbon is gradually formed on it.
 
 154 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 This knob or mushroom end finally breaks off, and is thrown 
 out of the arc. In Fig. 61 are shown a series of diagrams illus- 
 trating the gradual changes the negative carbon undergoes. 
 
 It has been shown by the author that, if a third carbon pole 
 is dipped into the continuous current arc, so that its tip is in 
 the core of the arc but not touching either positive or negative 
 carbon, or if the electric arc is projected sideways by a magnet 
 on to the third carbon, then a great difference of electric 
 pressure exists between the positive or crater carbon and this 
 
 Stage 1. 
 
 Stage 2. 
 
 Stage 3. 
 
 FIG. 61. Diagrams showing the Changes in Form of the End of the 
 Negative Carbon. 
 
 third pole, sufficiently great to illuminate a small incandescent 
 lamp or to ring an electric bell ; but that no sensible difference 
 of electric pressure exists between the negative carbon and the 
 third pole. 
 
 If a continuous current electric arc is formed in the usual 
 way, and if a third insulated carbon, held at right angles to 
 the other two, is placed so that its tip just dips into the arc 
 (see Fig. 62), we can show a similar series of experiments to 
 those described in Lecture II. under the head of the 
 "Edison effect." The arc is rather more under control if 

 
 ELECTRIC ARC LAMPS. 155 
 
 we cause it to be projected against the third carbon by means 
 of a magnet. We have now formed on the screen an image of 
 the carbon poles and the arc between them in the usual way. 
 Placing a magnet at the back of the arc, the flame of the arc 
 is deflected laterally and is blown against a third insulated 
 carbon held in it. There are three insulated wires attached 
 respectively to the positive and to the negative carbons of the 
 arc and to the third or insulated carbon, the end of which dips 
 into the flame of the arc projected sideways by the magnet. On 
 starting the arc this third carbon is instantly brought down to 
 the same electrical potential as the negative carbon of the arc, 
 and connecting an amperemeter in between the negative carbon 
 
 FIG. 62. Electric Arc projected by magnet against a third carbon, and 
 showing strong electric current flowing through a galvanometer, O, con- 
 nected between the positive and third carbon. 
 
 and the third or insulated carbon, we get, as you see, no indica- 
 tion of a current. If, however, we change the connections 
 and insert the circuit of the galvanometer between the positive 
 carbon of the arc and the middle carbon, we find evidence, 
 by the violent impulse given to the galvanometer, that there is 
 a strong current flowing through it. The direction of this 
 current is equivalent to a flow of negative electricity from the 
 middle carbon through the galvanometer to the positive carbon 
 of the arc. We have here, then, the " Edison effect " repeated
 
 156 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 with the electric arc. So strong is the current flowing in a 
 circuit connecting the middle carbon with the positive carbon 
 that we can, as you see, ring an electric bell and light a small 
 incandescent lamp when these electric-current detectors are 
 placed in connection with the positive and middle carbons. 
 
 We also find that the flame-like projection of the arc 
 between the negative carbon possesses a unilateral conductivity. 
 Joining this small secondary battery of fifteen cells in series 
 with the galvanometer, and connecting the two between the 
 middle carbon and the negative carbon of the arc, just as 
 
 FIG. 63. Galvanometer G and battery B inserted in series between 
 negative carbon of electric arc and a third carbon to show unilateral con- 
 ductivity of the arc between the negative and third carbons. 
 
 in the analogous experiment with the incandescent lamp, we 
 find we can send negative electricity along the flame of the arc' 
 one way but not the other. The secondary battery causes the 
 galvanometer to indicate a current flowing through it when 
 its negative pole is in connection with the negative carbon of 
 the arc (see Fig. 63), but not when its positive pole is in con- 
 nection with the negative carbon. On examining the third 
 or middle carbon after it has been employed in this way for
 
 ELECTRIC ARC LAMPS. 157 
 
 some time, we find that its extremity is cratered out and con- 
 verted into graphite, just as if it had been employed as the 
 positive carbon in forming an electric arc. 
 
 Much interesting work yet remains to be done in studying 
 the physics of the electric arc, and we cannot say yet that we 
 fully understand the mechanism which underlies this familiar 
 electric phenomenon. 
 
 We must not pass on from this part of the subject with- 
 out making reference to experiments which have been made 
 to determine the intrinsic brightness of the crater of the 
 electric arc. It has been estimated by Mr. A. P. Trotter that 
 each square millimetre of the incandescent crater sends out 
 light equal in brightness to 170 candles. Other observers, 
 such as M. Blondel, have determined a similar value, viz., 
 160 candles per square millimetre. That is to say, an arc 
 light crater, having an area, say, of 30 square millimetres, 
 would send out light which would produce an illumination or 
 brightness on a white surface placed parallel to it equal to 
 about 5,000 candles placed at the same distance. The intrinsic 
 brilliancy of the arc crater exceeds that of any other source of 
 light with which we are acquainted, except the sun. Certain 
 materials exist for making a comparison between solar tempera- 
 ture and brilliancy, and that of the arc light crater. It is well 
 known that the sun's edge (limb) is much less bright than the 
 centre (40 per cent, less) ; but, as a first rough approximation, 
 we may take it as equally bright, and assume that the sun 
 radiates light as if it were an incandescent circular disc of 
 852,900 miles in diameter. The actual area of its apparent 
 surface is then 1J billion billion square millimetres (1-48'10 24 
 sq. mm.). Prof. Young, in America, has made an estimate of 
 the illuminating power of the sun, and his conclusion is that 
 it is equal to 1,575 billion billion candles (1575. 10 24 c.p.), 
 correction being made for terrestrial atmospheric absorption.
 
 158 ELECTRIC LAMPti AND ELECTRIC LIGHTING. 
 
 Hence the mean intrinsic brilliancy of the sun's apparent surface 
 is about 1,000 candles per square millimetre. 
 
 The rate at which energy is being sent out from the sun's 
 surface has been estimated from data given by Herschell 
 and Pouillet ; and more recently Forbes and Langley have 
 corrected these data, so that from them Lord Kelvin 
 estimates the solar radiation as equal to 133,000 horse-power 
 per square metre. This corresponds very nearly to 100 
 watts per square millimetre. Accordingly every square 
 millimetre of solar surface is sending out energy as radiation 
 at a rate equal to 100 watts, and producing a candle- 
 power of 1,000 candles. The sun is, therefore, working at 
 an " efficiency," as a glow-lamp maker would say, of th of a 
 watt per candle. If we take the mean value of the estimates 
 that have been made for the intrinsic brilliancy of the crater 
 of the electric arc as 160 candles per square millimetre, and 
 the rate of energy radiation as 30 watts per square millimetre, 
 which is probably not far from the truth,* we see that the 
 solar surface radiation is 100 watts, and the solar intrinsic 
 briUiancy 1,000 candles per square millimetre; whilst the 
 electric arc light crater has a radiation of about 30 watts 
 and a brilliancy of 160 candles per square millimetre. The 
 brilliancy of the sun is, therefore, according to these figures, 
 six times that of the electric arc crater, and its rate of sur- 
 face radiation about three times. Compare this again with 
 the incandescent carbon of the glow lamp. A 16-c.p. glow 
 lamp with circular filament has a carbon filament five 
 inches long and T ^th of an inch in diameter. The area of 
 
 A series of experiments made in the author's laboratory seemed to 
 show that in various sized arcs, using from 300 to 1,000 watts, the crater 
 area was always of such size that in every case about 27 to 30 watts was 
 issipated per square millimetre of crater surface. This is on the assump- 
 tion that the whole work done in the are is in the first place expended in 
 the crater, and that the rest of the heating of the carbon is due to secondary 
 effects.
 
 ELECTRIC ARC LAMPS. 
 
 159 
 
 its apparent surface is, then, about 30 square millimetres, 
 its total surface is very nearly 100 square millimetres, and 
 it will radiate energy at a rate equal to about 50 watts when 
 giving a light of 16 candle power. Hence, per square milli- 
 metre of apparent surface, its intrinsic brilliancy is about 
 half a candle, and its rate of radiation half a watt. Compar- 
 ing the radiation power and intrinsic brilliancy of the three 
 illuminants -sun, electric arc crater, and glow lamp filament 
 per square millimetre of surface, we have a rough comparison 
 as follows : 
 
 -J 
 
 Rate of radiation of energy 
 per square millimetre 
 of surface. 
 
 Intrinsic brilliancy or 
 candle-power per square 
 millimetre. 
 
 Sun 
 
 100 watts. 
 
 1,000 candle-power. 
 
 Electric Arc 
 Crater 
 Glow Lamp 
 Filament... 
 
 30 
 
 i 
 
 160 
 * 
 
 The law connecting radiation with temperature at these 
 high temperatures is not yet definitely known, but the above 
 figures warrant the conclusion that the solar temperature, 
 though much higher, is not enormously higher than that of 
 the crater of the electric arc. The temperature of the sun 
 has sometimes been assumed to be millions of degrees Centi- 
 grade. All such guesses must be exceedingly wide of the mark. 
 The solar temperature, at least at the surface, is probably 
 much nearer 6,000 or 7,000C. 
 
 It is necessary to note the proper manner of comparing 
 various illuminating agents in respect of their specific 
 energy radiation and light-producing powers. If we consider 
 a straight cylindrical carbon filament which is, say, 5in. 
 long and y^th of an inch in diameter, this filament will 
 have a total surface of nearly T V^ths of a square inch, 
 but its apparent or projected surface will only have an
 
 160 ELECTRIC LAMPS AND ELECTEIC LIGHTING. 
 
 area of J ff th o a S( l uare inch * Let this filament > when 
 incandescent, give a candle-power of 16 candles measured 
 in a direction at right angles to itself, and absorb a 
 total power of 48 watts. If we take as our unit of area 
 j-oVffth of a square inch, called an inch-mil, it is clear 
 that the, energy waste is at the rate of 48 watts for 160 
 units of surface, or 1 watt per 3| inch-mils. The candle- 
 power given by the circular section filament is, however, 
 just the same as would be given out by a flat strip 
 of carbon, kept at the same temperature, the length of 
 which was the same and the width of which was equal 
 to the diameter of the round filament. Hence the brightness 
 of the surface, or candle power given out perpendicularly 
 by each unit of surface of the round filament, is obtained 
 by dividing the whole candle-power viz., 16 by the area 
 of the projected surface of the circular filament, viz., J^th 
 of a square inch. If, therefore, the unit of surface is the 
 j^th of an inch, the brightness of the surface is iths 
 of a candle per inch-mil, the energy waste is at the rate 
 of igths of a watt per inch-mil of surface. 
 
 In the usual way of reckoning what is called the " efficiency " 
 of the glow lamp it is usual to divide the whole power 
 taken up in the filament (in this case 48 watts) by the 
 resultant or observed candle-power in one direction (in this 
 case 16 c.-p.), and to obtain a quotient (in this case 3) of 
 watts per candle-power. If we are, however, defining the 
 rate at which energy is dissipated per square unit of area 
 of the incandescent surface, and the brightness or normal 
 candle-power per square unit of the surface, we see that 
 the ratio of the numbers representing brightness and power 
 radiated per unit of area is not the same as the ratio of 
 the numbers representing the observed candle power and the 
 total power dissipated when we are dealing with glow lamp 
 filaments. If we call the total power in watts radiated
 
 ELECTRIC AEG LAMPS. 161 
 
 per square unit of area from the incandescent surface the 
 dissipating power of that surface, we see that the ratio of 
 dissipating power to brightness is about one watt per candle 
 for a carbon filament when worked under the conditions of 
 temperature at which its "efficiency" would generally be 
 called three watts per candle-power. Hence it happens that 
 in the above table, comparing solar, arc, and glow lamp 
 radiation, the brightness of the carbon filament is given 
 as half a candle per millimetre, and the dissipating power 
 as half a watt per millimetre. It is necessary to distinguish 
 between the ratio of dissipating power to brightness and 
 total radiation of energy to observed candle-power. We 
 shall call the former the watts per normal candle-power 
 and the latter the watts per observed candle-power. 
 
 We cannot assume that the law which experiment shows 
 connects together the candle-power and the rate of energy 
 waste in watts in a glow-lamp filament is followed at 
 temperatures much higher than that at which the incan- 
 descent filament is usually worked. The temperature of the 
 carbon filament when being used at about 8 watts per 
 candle-power is probably not far from 1,500C. or 1,700C. 
 Up to that point the candle-power appears to increase 
 nearly as the cube of the power in watts taken up. We 
 cannot, however, heat the carbon to a temperature higher 
 than that which corresponds to one watt per observed 
 candle-power without rapidly volatilising it in vacuo. In 
 the electric arc crater the temperature appears to correspond 
 to about one- fifth of a watt per normal candle-power; and 
 this temperature M. Violle has stated to be about 3,500C., 
 or about double that of the incandescent carbon of the glow 
 lamp at normal temperature. These estimates of arc crater 
 temperature, however, probably err on the side of being too low, 
 and the true crater temperature may be even higher than the 
 above value. Stefan, in 1879, suggested that the temperature
 
 162 ELECTEIC LAMPS AND ELECTRIC LIGHTING. 
 
 of a heated black body and its rate of radiation of energy were 
 connected as follows : The fourth power of the absolute tem- 
 perature of the body varies as the rate of radiation of energy 
 from a unit of the surface of the body. In other words, if the 
 rate of radiation of energy is increased 16 times the tempera- 
 ture of the surface would be doubled, if 81 times trebled, 
 and so on. It has been asserted by several experiment- 
 alists that this law is inapplicable to high temperatures, 
 but if we assume for the moment that it can in any degree 
 apply to the range of temperature extending from the glow lamp 
 to the solar surface temperature, then, since the dissipating 
 powers of the surfaces of the glow lamp filament, arc crater, 
 and sun, are, from the table above, seen to be nearly in the 
 ratio of 1 : 50 : 200, the absolute temperatures should be 
 in the ratio of 1 : 2f ; 3f roughly, and this would indicate 
 the solar surface temperature as being not much greater than 
 7,000'C. 
 
 We must now briefly consider the practical side of electric 
 arc lighting. We shall make no attempt to enter into details as 
 to the countless forms of arc lamp mechanism, but in very 
 general terms, explain the nature of that mechanism. From 
 what has been said above, it will be seen that, in order to 
 maintain an electric arc between carbon points, some form of 
 apparatus has to be devised which will automatically perform 
 the three following operations : First, bring the carbon rods 
 or points together when no current is passed between them ; 
 second, as soon as a current passes, separate them to a 
 determined distance, or "strike the arc," as it is termed ; and, 
 third, will bring the carbons together slightly if the current 
 decreases, or separate them slightly if it increases. This last 
 action is called "feeding the arc." In a good arc lamp 
 mechanism the "feed" is very smooth and uniform, and is 
 marked by an entire absence of all jerky movements. All 
 early attempts at the invention of arc lamp mechanism com-
 
 ELECTRIC AEG LAMPS. 
 
 163 
 
 prised the use of clockwork, and this involved many objec- 
 tions. Modern arc lamp mechanism almost entirely depends 
 upon the employment of electromagnets, which act as the 
 source of power to move the carbons. Without attempting any 
 extended description of even a portion of the devices employed, 
 a very general idea may be obtained by considering the simplest 
 form of shunt and series coil lamps. One of the carbon rods, 
 say the upper one, is fixed to an iron rod or core I (see Fig. 64), 
 which has one end inserted just inside the hollow of a bobbin 
 wound over with thick insulated wire, Se, and the other end in- 
 serted just inside a bobbin, Sh, wound over with fine wire. The 
 
 FIG. 64. Diagram of Shunt and Series Coil Arc Lamp Mechanism. 
 M, M, are the two Electric Supply Mains ; Se, the Series Coil ; Sh, the 
 Shunt Coil ; I, the Iron Core, and C, the Carbons. 
 
 former bobbin is inserted in the line, bringing the current 
 to the arc, and is called the series coil. The fine wire coil has 
 its wire joined across between the two carbons, and is called 
 the shunt coil. The arrangement will be understood by 
 reference to Fig. 64. 
 
 When the carbons are placed in contact, and a current is 
 passed through them, this current passes through the series 
 coil on its way to the carbons. It therefore energises that 
 coil, and causes it to attract or pull in the iron core. It will be
 
 164 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 understood by reference to the figure that this action immediately 
 separates the carbons a little way. At the same time the 
 shunt coil is traversed by a current, the action of which is 
 to attract its half of the iron core, and therefore to exert 
 an influence drawing the carbons together. The carbons 
 therefore become separated by a small amount, thus striking 
 the arc, and the amount by which they become so separated is 
 determined by the relative magnetic power of the two coils. If 
 the current becomes weakened by the carbons burning away, 
 then the action of the shunt coil preponderates, and draws the 
 carbons together again, thus shortening the arc and increasing 
 the current. When the current increases in strength the 
 series coil exerts the most powerful action of the two, and 
 separates the carbons again by a small amount. It will be 
 seen that the iron core is kept floating between two balanced 
 attractions, one of which tends to draw the carbons attached to 
 it together, the other to separate them. By this simple means, 
 modified in practice by various details of construction, an arc 
 lamp can be constructed which is self-feeding and which sets 
 itself in action by the simple process of switching it on to the 
 circuit. 
 
 There are an immense variety of arc lamps depending 
 upon this principle, in all of which various means are 
 taken to make the feed of the lamp properly gradual. If the 
 feed of the lamp is not sufficiently regular it shows itself 
 by rapid changes in the appearance of the arc. As the arc 
 lengthens the violet light of the true arc preponderates, and 
 the lamp burns with a much more bluish colour. If the carbons 
 then come together suddenly the lamp is very liable to hiss 
 and to burn for some time with a more reddish light until 
 the carbons are separated. In lamps with imperfect feeding 
 mechanism these changes give rise to irregularities in the 
 light which are very annoying. In a good arc lamp the feeding 
 mechanism should work with such perfect regularity that the
 
 ELECTEIC ARC LAMPS. 165 
 
 movement of the carbons can hardly be discerned with a 
 magnifying glass. The best manner of examining the feed of 
 an arc lamp is to project, by means of a lens, an image of the 
 arc lamp upon the screen, and then to examine this optical 
 image. The motion produced by the feeding mechanism 
 is then magnified in the same ratio as the dimensions of the 
 arc, and any irregularity of motion is very easily detected. 
 
 One word ought to be said with regard to the quality of 
 carbons used in arc lighting. A vast amount of ingenuity 
 and capital has been expended in investigations intended 
 to discover the best processes for the manufacture of arc 
 light carbons. These carbons are now made as above 
 mentioned by forming into a paste some very pure form of 
 carbon by means of a fluid, such as tar or syrup, capable 
 of being carbonised. This paste is squeezed out into rods by 
 powerful hydraulic presses, and these rods are then dried and 
 baked. Carbons are classified as cored or non-cored carbons. 
 In the cored carbons the centre of the carbon rod is perforated 
 by a longitudinal hole, thus forming a carbon tube, and this 
 carbon tube is filled up with a less dense form of carbon. 
 These cored carbons are generally used for the positive carbon 
 of the arc, and the object of making a softer central portion to 
 the carbon is to determine the position of the crater and 
 compel it to keep a central position. 
 
 As already explained, in operating continuous current arc 
 lamps the positive or crater carbon is generally placed upper- 
 most. Under these circumstances there is a cone of shadow 
 directly under the lamp which is due to the negative carbon. 
 Mr. Crompton has found it to be a great advantage to use 
 very much smaller negative carbons than is usually done, and 
 in his arc lamp he uses the smallest negative carbon which 
 can be employed practically. On the other hand, a too 
 small negative carbon is liable to become heated over a greater
 
 166 ELECTRIC LAMPS AND ELECTEIC LIGHTING. 
 
 portion of its length, and therefore a practical limit is placed 
 to the use of very small negative carbons. It appears probable, 
 from the experiments which have been made, that the 
 hissing of an arc lamp at any rate, of a continuous current 
 arc lamp is caused by a too great current density ; that is to 
 say, by employing a current too large in comparison with the 
 cross-section of the carbons. Generally speaking, in most of 
 the ordinary arc lamps used for outdoor illuminating purposes, 
 the positive or crater carbon is about 14 or 15 millimetres in 
 diameter that is to say, f ths of an inch and the negative 
 carbon about 10 or 11 millimetres, or f-ths of an inch. 
 Under these circumstances the current density in the carbon 
 is somewhere about 50 amperes per square inch. Under 
 some circumstances an electric arc burning quietly sets to 
 work suddenly to hiss loudly, and at that moment the current 
 through the arc also suddenly increases and the difference of 
 pressure between the carbons decreases. M. Blondel thus 
 explains the cause of this effect. At the moment when 
 the hissing begins the voltage falls and the arc loses its 
 transparency, and is transformed into an incandescent mist 
 which hides the crater. When the hissing ceases the arc 
 suddenly becomes violet and transparent again, and the surface 
 of the crater appears to be covered with black specks, which 
 gradually disappear. Whilst the hissing lasts, the brightness 
 of the crater is diminished, and the violet colour of the arc 
 becomes replaced by a very characteristic blue-green tint 
 which seems to point to a lowering of the temperature. M. 
 Blondel thinks that in this case the uniform flow of carbon 
 vapour is replaced by an intermittent disruptive discharge in 
 which the carbon is torn away as it were in pieces, and not 
 evaporated. Hissing occurs when the current density exceeds 
 about an ampere per square millimetre of crater surface, and 
 is more likely to occur with soft carbons than with hard. 
 
 In the employment of arc lamps for illuminating purposes, 
 owing to the fact that the light is radiated from an exceed-
 
 ELECTRIC ARC LAMPS. 
 
 167 
 
 ingly small area (in a 300- watt lamp, about Jyth part of a 
 square inch), the light sent out by the lamp throws very 
 sharply denned shadows if the lamp is used without a shade. 
 Hence it is usual to enclose the lamp in a large semi-opaque 
 globe. These globes cut off from 40 to 60 per cent, of the 
 light in various directions ; but, by presenting a larger light- 
 giving area they cast less well-marked shadows and are less 
 dazzling to the eye. On the other hand, this wasteful absorp- 
 tion of the light causes the lamp to be less efficient as an 
 illuminating agent. In order to meet this difficulty the 
 device is frequently adopted of employing inverted arc lamps. 
 In this arrangement the arc lamp is hung from the 
 
 FIG. 65. Inverted Arc Lamp and Shade for Workshop and Factory 
 Lighting. The dotted line shows the place of the Conical Reflector. 
 
 ceiling or roof of the building to be illuminated with the 
 negative carbons uppermost (see Fig. 65), and the positive or 
 crater carbon is placed at the bottom of the pair. The lamp 
 is surrounded with a conical shade, the purpose of which is to 
 throw the light up, and in addition sometimes a white umbrella- 
 like shade is suspended over the lamp, from which the light is 
 again reflected downwards. A room illuminated by these 
 inverted arc lamps is therefore flooded with light, none of the
 
 168 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 rays of which come directly from the arc, but only after reflec- 
 tion from the hood or the ceiling. A very pleasing, diffused 
 light is thus produced, and inverted arc lighting has been tried, 
 in many cases with great success in workshops, dye works and 
 factories. You are able to judge of the effect of it yourselves 
 if I illuminate this theatre for a few minutes by means of a 
 couple of inverted arc lamps which have been lent to me by 
 Messrs. Parsons and Co. If a workshop is illuminated by arc 
 lamps in the ordinary manner, even although they are pro- 
 tected by semi-opaque globes, the light casts sharp shadows 
 around and under the tools, and the workmen are therefore 
 sometimes unable to get the light necessary for their work ; but 
 the inverted arc lighting does away with this difficulty, and in 
 places so illuminated there are practically no shadows at all. 
 
 We cannot do more than make a brief allusion to the 
 manner in which arc lamps are employed in practical electric 
 lighting. One method is by means of what are called series 
 arc lamps. In this arrangement the arc lamps to the number 
 of 30 or 50 are placed on one circuit, so that the same electric 
 current, say one of ten amperes, flows through each lamp in 
 turn. As each lamp requires to have about 50 volts on 
 the terminals in order to make it work properly, it will be 
 seen that the arrangement necessitates the use of a dynamo 
 machine giving a very high electric pressure from 1,500 
 to 3,000 volts. Series arc lighting is employed to a con- 
 siderable extent in street lighting, for the reason that the 
 conductors conveying the current can be, comparatively speak- 
 ing, small and inexpensive. Arc lamps may, however, be 
 arranged in parallel instead of in series. In this case two 
 or more lamps are placed together across the mains between 
 which a constant electric pressure is being maintained. 
 Suppose, for instance, in any district an electric lighting 
 supply company has the mains laid down for giving a supply 
 at 100 volts for incandescent lighting, then two arc lamps can
 
 ELECTRIC ARC LAMPS. 169 
 
 be placed across these mains in series, a small resistance being 
 added in order to control the flow of current through the 
 lamps. At the same time any number of these pairs can be 
 arrranged across between the mains like the rungs of a ladder. 
 In some cases five to ten arc lamps are worked in series across 
 mains having a difference of potential between them of from 
 250 to 500 volts, a number of these series of five or ten being 
 strung across in parallel between two supply mains. An ob- 
 jection which has always been urged against series arc lighting 
 is that it necessitates the employment of high pressure currents, 
 which are, of course, relatively much more dangerous than 
 low pressure currents. In series arc lighting each lamp has, 
 moreover, to be provided with an apparatus called an automatic 
 cut-out, so that if the lamp becomes extinguished by the 
 carbons being separated too far or burning out, the cut-out 
 closes the circuit, and does not permit the current which is 
 supplying the other lamps in the series to be interrupted. 
 In the case of alternating-current arc lamps, they may either 
 ^ be run in series on a high tension alternating-current circuit, 
 'as is done in the electric lighting of Eome, or each lamp may 
 be provided with a small transformer, so that current can be 
 taken from high tension mains and reduced down to a 
 pressure at which it is convenient to work the arc lamps. 
 We shall in the next Lecture proceed to explain more in detail 
 the nature of this pressure-reducing device.

 
 LECTURE IV. 
 
 THE Generation and Distribution of Electric Current. The Magnetic 
 Action of an Electric Current. The Magnetic Field of a Spiral 
 Current. The Induction of Electric Currents. The Peculiar 
 Magnetic Property of Iron. Iron and Air Magnetic Circuits. The 
 typical cases of an Iron Circuit with and without Air Gaps. The 
 prototype forms of Dynamo and Transformer. The Transformation 
 of Electrical Energy. Hydraulic Illustrations. The Mechanical 
 Analogue of a Transformer. The Mode of Construction of an 
 Alternate Current Transformer. The Fundamental Principle of all 
 Dynamo Electric Machines. Alternating and Direct Current Dynamos. 
 Alternating Current Systems of Electrical Distribution. Descrip- 
 tion of the Electric Lighting Station in Rome. The Tivoli-Rome 
 Electric Transmission. Views of the Tivoli Station. Continuous 
 Current Systems. The Three-Wire System. Description of St. 
 Pancras Vestry Electric Lighting Station.- 
 Liverpool, Glasgow, and Brussels Electric 
 Lighting Stations. Direct-driven Dynamos. 
 Alternating and Continuous Current 
 Systems Contrasted. The Centenary of the 
 Birth of the Electric Current. 
 
 N tliis fourth and last Lecture it will 
 be necessary to direct our atten- 
 tion very briefly to the subject of 
 the generation and distribution of 
 electric current for the purposes 
 of electric illumination. To en- 
 deavour to cover anything but a 
 small portion of this immense 
 subject in the space of one lecture 
 
 would be to confuse and bewilder rather than to inform. 
 
 It will, therefore, be necessary to limit our discussion to
 
 172 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 certain narrow lines, and not to attempt, in any way, to 
 give even an outline of the whole of the methods which are 
 at the present moment employed for this purpose. Avoiding 
 technicalities, however, the endeavour will be made to enable 
 the reader to grasp certain broad general principles which 
 underlie the construction of all the appliances used for the 
 generation and distribution of electric current for industrial 
 purposes, and then we shall proceed to describe generally 
 the arrangements employed in the electric lighting of a 
 modern city. We must return, in the first place, to facts 
 touched upon in the First Lecture, and examine more in 
 detail the relation of magnetism and electric currents. In so 
 doing it may be an advantage, for the sake of brevity and 
 clearness, to depart a little from the usual phraseology. We 
 must also trace our way in the light of a series of pre- 
 liminary experiments. 
 
 It has already been pointed out that a wire, conveying what 
 we call an electric current, exerts a magnetic influence, and 
 at the same time becomes heated. In fact, the phrase, "an 
 electric current," is merely a comprehensive expression used 
 to denote all the effects of heat and magnetism that are known 
 to exist in and around a wire which we speak of as being 
 traversed by an electric current. If we take a conducting wire 
 say a copper wire and pass through it a strong electric 
 current, we find that such a wire, when dipped into iron 
 filings, takes them up, the filings clinging to the wire forming 
 a bunch around it. This fact was discovered about 1820 by 
 the French experimentalist Arago. We can more carefully 
 and deliberately explore the nature of this magnetic action 
 of a wire conveying a current in the following way : Passing 
 the copper wire through a hole in a glass plate, or card, so 
 that the wire stands vertically to the plate, we then sprinkle 
 the glass plate uniformly with steel filings, and, by causing a 
 powerful current to pass through the copper wire, and by gently
 
 ELECTRIC DISTRIBUTION. 
 
 173 
 
 tapping the glass plate, the filings arrange themselves in 
 a series of concentric circular lines round the wire (see 
 Fig. 66). The glass plate being placed in this vertical 
 lantern, the image of these circular lines of filings is projected 
 upon the screen. Taking a very small magnetised compass 
 needle, and holding it in any position near the glass plate, 
 it at once becomes evident that the little magnetised needle 
 always places itself in the same direction as the circular 
 lines mapped out by the iron filings. This fact may be 
 
 FIG. 66. Curves formed by sprinkling Iron Filings on a Card C, per- 
 forated by a Wire conveying an Electric Current, thus showing the lines of 
 Circular Magnetism round the Conductor A B. 
 
 viewed in the following light : The circular magnetisa- 
 tion of the space round the wire is a manifestation of 
 a part of the properties of the electric current, and may 
 properly be called the magnetic part of the current, or, 
 more shortly, may be called the magnetic current enveloping 
 the conductor. In fact, the magnetic current is a part of 
 that which we call the electric current. The direction of flow 
 of this magnetic current round a conductor may be ascertained 
 at any place by the use of iron filings as above described.
 
 174 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 A very little examination of the distribution of the iron filings 
 shows us that they He closer together at points near to the 
 wire than at points far away. 
 
 It is important to examine another instance of the same 
 kind. Here is a large bobbin of wire, the wire being covered 
 over with cotton and wound on a paper tube. Into the 
 interior of this bobbin is placed a slip of glass on which steel 
 filings have been sprinkled. On passing an electric current 
 through the wire and tapping the glass, we find that the steel 
 filings arrange themselves inside the bobbin in a series of 
 nearly parallel straight h'nes (see Fig. 67), and that, in this 
 particular case, it is, therefore, evident that the electric current, 
 
 FIG. 67. Curves formed by Iron Filings sprinkled on a Card, through 
 holes in which a spiral wire is laced, showing the h'nes of Magnetic Current 
 or Induction linked with a Spiral ElectricCurrent. 
 
 flowing through the wire in a series of nearly circular turns, 
 is accompanied by a flow of magnetism, or a magnetic current, 
 in a direction along the axis of the bobbin. By exploring 
 the space outside the bobbin with a magnetic compass needle 
 it is very easy to show that the current of magnetism which 
 flows in one direction in the ulterior of the bobbin flows 
 back along the outside of the bobbin in an opposite direction, 
 and so completes what is called a magnetic circuit. What- 
 ever particular form we give to the conductor conveying 
 the electnc current, we always find that there is a magnetic 
 current flowing in the space round it in a magnetic circuit 
 which is linked with the conducting circuit. This co-linking 

 
 ELECTRIC DISTRIBUTION. 175 
 
 of electric conducting circuits and magnetic circuits is the 
 fundamental fact of electro-magnetism. 
 
 So far the cases we have been examining have been such 
 that the magnetic current flows in a circuit which is wholly 
 occupied by air as a medium. If, instead of permitting the 
 magnetic flow to take place in air, it takes place in iron wholly 
 or partly, then it is found that the same electric current 
 flowing in the wire would produce a much more powerful 
 magnetic current in the iron circuit than it does in the 
 air circuit. 
 
 It is necessary to examine this point with some care, and to 
 do this we must define a little more accurately the mode of 
 measuring the magnetic quantities concerned. If a wire 
 covered with silk or cotton is wound any number of times 
 round a bobbin of any kind, and if a current of a certain 
 strength, measured in amperes, is allowed to flow through 
 that wire, the product of the number of turns of the wire 
 and the strength of the current measured in amperes which 
 is flowing round the bobbin wire is an important quantity, 
 which is called the ampere-turns of that bobbin. The total 
 number of ampere-turns on the bobbin, divided by the length 
 of the bobbin, gives us the ampere-turns per unit of length 
 of the coil. We have already seen that in the interior of 
 such a bobbin of insulated wire when traversed by an electric 
 current, there is a magnetic current or magnetic induction, 
 as it is called the direction of this field being in the direction 
 of the axis of the bobbin. Before we can show how this 
 induction is measured it will be necessary to direct attention 
 to the fundamental discovery in connection with this subject. 
 
 Imagine that in the interior of the bobbin employed a 
 moment or two ago in our experiments another bobbin of wire is 
 inserted, so as to occupy a portion of the space in the interior. 
 Let this second bobbin be called, for the sake of distinction,
 
 176 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 the secondary conducting circuit, whilst the first is called the 
 primary circuit. Faraday discovered that, as long as the 
 electric current flowing in the primary circuit remains con- 
 stant or unaltered, no effect will be produced by it upon the 
 secondary bobbin. The aperture, or central hollow, of this 
 secondary bobbin remains traversed by part or all of the mag- 
 netic induction, or, as we have called it, the magnetic current 
 of the first bobbin, but otherwise it is not affected. If the 
 electric current traversing the primary bobbin is either stopped 
 or reversed in direction, or is altered in strength, the immediate 
 result is to produce a current of electricity circulating in the 
 secondary bobbin ; which current is, however, only a very brief 
 or transitory current, and does not continue for any length of 
 time. It is, as it were, a sort of wave of electricity passing 
 through the secondary circuit. By appropriate instrumental 
 means we can measure the quantity of electricity which is in 
 this way set in motion. We can also measure, in the unit 
 called an ohm, as explained in the First Lecture, the electrical 
 resistance of this secondary circuit. Faraday showed that the 
 strength of the magnetic induction which is traversing the 
 secondary bobbin could be exactly measured by the product 
 of the resistance of this secondary circuit and the whole of 
 the quantity of electricity set flowing in it when the primary 
 current was suddenly arrested. Hence the use of a secondary 
 circuit of this kind affords us the means of exploring the 
 magnetic condition of the space in the interior of such a 
 primary bobbin. 
 
 In Fig. 68 is shown a pair of spiral conducting circuits, 
 P and S, which consist of spirals of wire wound through 
 suitable holes in a card. If a current of electricity is passed 
 through the circuit P, and if iron filings are sprinkled on the 
 card, we find that the lines of magnetic induction, due to the 
 primary circuit, are marked out. Some of these lines will be 
 seen to pass through, or be linked with, the secondary circuit S. 

 
 ELECTRIC DISTRIBUTION. 
 
 177 
 
 It may be well to state at this point that the quantity of 
 electricity which flows past any point on a conducting electric 
 circuit in one second, when it is being traversed by a current 
 of one ampere, is called one coulomb. We are able to measure 
 quantity of electricity by means of an instrument called a 
 ballistic galvanometer, and, if we are provided with such an 
 appliance, it is a comparatively simple matter to determine 
 the whole quantity of electricity which flows through a circuit 
 when any change takes place in the magnetic current or 
 induction linked with it. 
 
 i^Xx^-A 
 
 rvv$C x -Vi\ 
 
 [^'^l^<\ 
 
 n p 
 
 4- - 
 
 FIG. 68. Curves delineated by Iron Filings on a Card, showing the line 
 of Magnetism of a Spiral Primary Circuit, P, passing through and linked 
 with a Secondary Circuit, S. 
 
 Eeturning, then, to the case of our primary and secondary 
 bobbin : let us, for the sake of simplicity, assume that the 
 secondary bobbin is wound closely on the outside of the primary 
 bobbin, and that the primary bobbin has a certain given number 
 of turns, and is traversed by a current of a certain strength 
 measured in amperes. When a primary current of, say, ten 
 amperes is passing through the primary bobbin, we have a 
 magnetic current of a certain strength, or a magnetic induc- 
 tion, flowing hi the direction of the axis of that primary 
 bobbin, part or all of which magnetic induction or current 
 perforates or passes through the secondary circuit. If the 
 secondary circuit is connected to a ballistic galvanometer,
 
 178 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 and the primary current is suddenly stopped, the galvanometer 
 will give an indication that a brief wave of electricity has 
 passed through the secondary circuit, and this is called a secon- 
 dary current. The product of the quantity of electricity in this 
 secondary current and the resistance of the secondary circuit 
 is, under these circumstances, a measure of the amount of the 
 total magnetic induction, or the total magnetic current due to 
 the primary current in the primary bobbin which is flowing 
 through or linked with the secondary circuit. In the above 
 case the magnetic circuit consists of air that is to say, the 
 magnetic current flows in a circuit which is merely the air 
 space in and around the primary bobbin. 
 
 Supposing that we insert in the interior of the primary 
 bobbin a thick iron rod wholly filling up the interior space, 
 and either straight or bent round so as to form a closed ring. 
 Experiment shows that under these circumstances the state 
 of affairs is greatly altered. On passing through the primary 
 bobbin the same current of, say, ten amperes, and then 
 stopping that current suddenly, we should find that the electric 
 impulse produced in the secondary circuit is immensely in- 
 creased, and that the quantity of electricity circulated in the 
 secondary circuit would then be a thousand or more times 
 greater multiplied probably 50 times if the iron bar were 
 a short straight rod, and multiplied 2,000 times if the iron 
 was bent round in the form of a ring. This fact shows us 
 that a given number of ampere-turns in the primary bobbin 
 produces an electric impulse, or electromotive force, in the 
 secondary circuit which is dependent upon the nature of the 
 material filling the space in the interior of, and outside of, 
 the primary bobbin. 
 
 These facts can all be summed up in the following state- 
 ments : If there be two circuits of insulated wire, which are 
 called respectively the primary and secondary circuits, and
 
 ELECTRIC DISTRIBUTION. 179 
 
 which are placed near to one another or wound over one 
 another, then the flow of a continuous current of electricity 
 through the primary circuit produces no efl'ect whatever upon 
 the secondary circuit, so long as that primary current remains 
 unaltered in strength. If the primary current is changed in 
 strength, either being increased or diminished, reversed or 
 reduced to nothing, this change in the number of ampere- 
 turns in the primary bobbin gives rise to an electromotive 
 force, or force setting electricity in motion, acting in the 
 secondary circuit, and this electromotive force, or electric 
 impulse, depends essentially upon the rate of change of the 
 current strength, or the rapidity of the change which is made 
 in the strength of the primary current, and, therefore, in the 
 rate of change of strength of the magnetic current or magnetic 
 induction which is produced by the primary current, and which 
 passes through the secondary circuit. 
 
 Following a conception of Faraday's, it is usual to speak 
 of the direction of the magnetic current which surrounds the 
 electric current as the direction of the magnetic lines of force 
 due to the primary current, and these lines of force can, as we 
 have already explained, have their directions made manifest 
 by the employment of steel or iron filings. .The peculiar 
 property which iron possesses, when used as a magnetic 
 circuit, is that it permits the production through it of a given 
 magnetic current by the employment of a far less number 
 of ampere-turns per unit of length than does any non- 
 magnetic material. 
 
 We may examine the behaviour of iron in this respect most 
 easily by returning to the fundamental experiment by which 
 it was discovered. Let us take a wooden ring having a 
 circular section, and wind over it insulated copper wire to 
 form a primary circuit, and over that again another insulated 
 secondary circuit. Through the primary circuit let an electric 
 
 N2
 
 180 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 current be passed, producing, therefore, a certain magnetising 
 force, measured by the number of ampere-turns per unit of 
 length of that bobbin. From what has been above explained 
 it will be seen that this primary current gives rise to a 
 magnetic induction or magnetic current which passes round 
 the ring, and some or all of which traverses the secondary 
 circuit. If the primary current is stopped, the arrest of 
 the magnetic current flowing through the secondary circuit 
 creates in it an electromotive force which causes a brief 
 flow of electricity to take place through that circuit, if it is 
 complete, and that electromotive force at any instant is 
 
 FIG. 69. Iron Ring I, having wound on it two Circuits, a Primary, P, 
 connected to a battery, B, through a switch or key, K, and a Secondary 
 Circuit, S, connected to a galvanometer, G. When the Primary Current 
 is stopped or started it creates an Induced Current in S. 
 
 measured by the rate at which the magnetic field or magnetic 
 current passing through the secondary circuit is being changed. 
 Supposing, however, that these circuits, instead of being wound 
 on a wooden ring, are wound on an iron ring of exactly the 
 same size (see Fig. 69), and that we pass through the primary 
 circuit a current of the same magnitude as before. The arrest 
 or stoppage of this current would now be found to produce a 
 vastly greater electromotive force in the secondary circuit, and 
 the explanation of this fact is that the primary ampere-turns 
 are now able to produce a much greater flow of magnetic 
 current through the secondary circuit, and that, therefore, the 
 arrest of the primary current creates a greater rate of change 
 in the magnetic current flowing through the secondary circuit. 
 The nature of the material filling the space, either partly or
 
 ELECTRIC DISTRIBUTION. 181 
 
 wholly, in and around the primary and secondary circuits, 
 has a very important influence on the inductive effect of the 
 primary circuit upon the secondary. 
 
 It is sometimes customary to speak of the primary circuit 
 as exerting a mayneto-mottie fora on the magnetic circuit- 
 In the above simple instance of the iron ring, we see that 
 we have two electric circuits and one magnetic circuit, 
 which are linked together like the links of a chain, and 
 that between the three circuits there is a fixed relation of 
 action which is as follows : The electromotive force acting 
 in the primary circuit produces in the primary circuit an 
 electric current. This electric current, flowing a certain 
 number of times round the primary circuit, produces a 
 magneto-motive force of a given magnitude, which creates a 
 magnetic current, or, as it is called, a magnetic induction in 
 the magnetic circuit. The magnetic circuit, in turn, exercises 
 an influence upon the secondary circuit, but only when the 
 magnetic induction or magnetic current is being changed. In 
 the secondary circuit an electromotive force is set up when- 
 ever the magnetic current is being altered in magnitude, and 
 the electromotive force acting to produce an electric current in 
 the secondary circuit is measured by the rate at which the 
 magnetic current or magnetic induction throughout is being 
 changed. 
 
 We have, therefore, the following operations related to one 
 another in these three circuits, the primary circuit, the 
 secondary circuit, and the magnetic circuit. The primary 
 electromotive force acting in the primary circuit produces 
 a primary current ; the primary current acting on the 
 magnetic circuit produces a magneto-motive force ; the 
 magneto-motive force produces a magnetic current or mag- 
 netic induction in the magnetic circuit ; the magnetic in- 
 duction, if/ten clianyiny in amount, produces a secondary
 
 182 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 electromotive force in the secondary circuit, and therefore 
 a secondary current. In order to distinguish that quality 
 of iron and air in virtue of which a given magneto- motive 
 force produces a much greater magnetic current or magnetic 
 induction in the iron than it does in the air, we say that 
 the iron possesses a less magnetic resistance than the air. 
 The magnetic resistance of iron may be as much as from one 
 to two thousand times less than that of air. 
 
 Supposing, in the next place, we make a cut across each 
 side of our iron ring (as seen in Fig. 70), separating the 
 iron ring into two portions by a narrow air-gap across each 
 
 FIG. 70. Iron ring I, as in Fig. 69, but having a pair of air-gaps made by 
 cutting the ring in two at the points A A. 
 
 side. It would then be found that a given electric current 
 flowing in the primary circuit would, when arrested, produce a 
 much smaller secondary electromotive force in the secondary 
 circuit than it does when the iron ring is uncut. The reason 
 for this is because the air-gap on both sides has increased 
 the magnetic resistance of the magnetic circuit ; and, accord- 
 ingly, although the magnetic current or induction is able 
 to traverse the air-gaps, the magneto-motive force of the 
 primary circuit is now not able to produce so much magnetic 
 current or magnetic induction in the magnetic circuit ; and 
 hence, when the primary current is arrested, the change in 
 the magnetic induction or magnetic current flowing through 
 the secondary circuit is not so great as before. 

 
 ELECTRIC DISTRIBUTION. 
 
 183 
 
 The two cases we have just examined namely, the iron 
 ring complete, with a primary and a secondary circuit wound 
 over it, and an iron ring with two air-gaps in it that is to 
 say, two half- iron rings, on one of which is wound a primary 
 circuit, and on the other a secondary circuit constitute two 
 typical models which will enable us to explain the action 
 of appliances which are called dynamo- electric machines and 
 transformers. 
 
 FIG. 71. Iron Ring magnetised in lines round the ring, but producing no 
 external magnetic field or influence. 
 
 Attention should, at this point, be directed to four 
 diagrams, which are reproductions from photographs of 
 experiments easily made. 
 
 In Fig. 71 is shown an iron ring, circular in form. Let a 
 spiral of insulated wire be wound on this ring and an electric 
 current be passed through it. This current will magnetise 
 the ring along lines shown as dotted lines. If iron filings 
 are sprinkled over a card held over the ring they will not
 
 184 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 arrange themselves in any particular form in the space 
 outside the ring, thus showing that the magnetism is wholly 
 confined to the ring. Next, let a narrow cut be made in the 
 ring (we Fig. 72). At once we find, by applying the iron 
 filings test, that there is now a development of magnetism 
 in the space outside the ring, and that there are lines of 
 magnetic induction or current passing across from one side 
 of the air-gap to the other, producing a powerful magnetic 
 
 FIG. 72. Magnetised Iron Ring with cut or air-gap. The ring is 
 magnetised along the dotted lines. The external magnetic field or induc- 
 tion is delineated by iron filings. 
 
 field in the gap space. By making wider gaps, as shown in 
 Figs. 73 and 74, the varying forms of the external fields can 
 be detected by the use of the sprinkled filings. 
 
 We have already seen that, in order to produce an induced 
 electromotive force, and therefore an electric current, in the 
 secondary circuit, it is necessary to produce a change in the 
 magnetic current or magnetic induction passing through that
 
 ELECTRIC DISTRIBUTION. 
 
 
 185 
 
 FIG. 73. Magnetised Iron Ring, as in Fig. 72, but with wider air-gap. 
 
 
 Fin, 74. Magnetised Iron Ring, as in Fig. 72, but witli still wider air-gap.
 
 186 ELECTEIC LAMPS AND ELECTPIC LIGHTING. 
 
 circuit. This change may be produced in two ways either by 
 stopping the primary current or by reversing its direction. 
 In the second case the amount of change in the induction is 
 doubled, because the reversing of the primary current amounts 
 to first stopping it, and then starting it again in the opposite 
 direction. Take the first case, viz., the complete iron ring, 
 and imagine the direction of the primary current continually 
 being changed or alternated. This would produce an alternat- 
 ing flow of magnetism round the ring, the direction of the 
 magnetisation changing its direction with every change in the 
 direction of the primary current. This changing magnetism 
 passing through the secondary circuit would set up in the 
 secondary circuit an alternating secondary electromotive force, 
 and therefore, if the secondary circuit is completed, an alter- 
 nating secondary current. It will be at once evident, however, 
 that there are two ways in which we may set up this changing 
 magnetism in that half of the ring on which the secondary 
 circuit is wound. We may either employ the undivided ring, 
 and keep continually changing the direction of the primary 
 current in the primary circuit, or we may maintain the primary 
 magnetising current constant in one half of the ring and turn 
 round that half of the ring on which the secondary circuit is 
 wound, so as to keep on reversing the direction of the magnetic 
 current through the secondary circuit. A little consideration 
 of these two modes of producing the changing induction through 
 the secondary circuit will show that they really amount to the 
 same thing. 
 
 In the light of the above remarks it will be found very easy 
 to understand the general action of the alternate current 
 transformer and dynamo. It has been explained in the previous 
 Lectures that the power conveyed by a continuous electric 
 current is measured by the product of the current strength 
 measured in amperes and the difference of pressure or poten- 
 tial between the ends of the circuit measured in volts. The
 
 ELECTRIC DISTRIBUTION. 187 
 
 same amount of power, say one horse-power, can, therefore, 
 be conveyed, either in the form of a large current having a 
 small fall in pressure or in the form of a small current having 
 a large fall in pressure. We are familiar enough with this 
 effect in the case of water. A waterfall may consist of a very 
 large body of water falling down a very moderate height, such 
 as the well-known falls of the Ehine at Neuhausen, in Switzer- 
 land ; or it may present itself in the form of a much less 
 quantity of water falling from a much greater height, like the 
 falls of the Anio, at Tivoli, in Italy. In either case the power 
 which the water would be capable of exerting, if utilised by 
 means of turbines or waterwheels, would depend, not on the 
 height of the fall nor on the quantity of the water flowing down, 
 taken alone, but on the product of the numbers representing 
 respectively the height and the quantity. The height of the 
 fall determines the velocity or pressure of the water at the 
 base of the fall. It is impossible to get any work out of 
 water unless we are able to let it fall from one level to a 
 , lower one. Hence it is that stores of water at a height above 
 'the level of the sea represent available sources of energy. 
 
 The power which originally lifted the water up before it 
 can fall must have been the evaporative power of the sun's 
 heat. We commonly speak of using water-power; but, as 
 a matter of fact, we are really employing in these cases 
 sun-power. The falls of Niagara represent the accumu- 
 lated water drainage of half a continent falling over a 
 precipice of 150 feet in height on its way to the sea. This 
 water, however, was originally evaporated from lakes, rivers, 
 or the sea, before it could be precipitated as rain over 
 Canada and North America. The power which lifted these 
 water molecules from the surface of the Atlantic or Pacific 
 Oceans into a position in which we are able to make them 
 give back part of their energy of position was the radiation 
 proceeding from the sun. The dynamos now being placed in
 
 188 ELECTEIC LAMPS AND ELECTRIC LIGHTING. 
 
 position to tap off some of the available power of the Falls of 
 Niagara will in reality be driven by sun-heat, radiated to the 
 earth at some previous time. 
 
 These hydraulic facts have an exact parallel in electrical 
 science. A current of electricity can only do work by falling 
 down from one electric pressure or potential to a lower one, 
 and the rate at which it can do work is measured, not by 
 the current strength or by the fall in potential alone, but by 
 the product of the numbers representing these two quantities. 
 A current of electricity of 10 amperes falling in pressure by 
 1,000 volts, represents the same power of doing work as a 
 current of 1,000 amperes falling 10 volts in pressure. There 
 is, however, a considerable difference between the two agents. 
 The small current of high pressure is very different in many 
 respects from the large current of low pressure. In the first 
 place, the high-pressure current is more dangerous to handle 
 or deal with, and requires much better insulation than the 
 low-pressure current. On the other hand, when a current of 
 electricity flows through a conductor a part at least of its 
 energy is frittered away into heat by reason of the necessary 
 resistance of the conductor. The amount so dissipated is, by 
 Joule's law, measured by the product of the square of the 
 current strength and the resistance of the conductor measured 
 in ohms. Hence, if flowing in conductors of the same size in 
 cross-section and length, a current of 1,000 amperes would 
 generate 10,000 times more heat per second than a current of 
 10 amperes. Suppose, then, that we wish to convey a power 
 of 10,000 watts to a distance, we can do it either by using a 
 current of 10 amperes flowing out and back along electric 
 mains between which 1,000 volts pressure is maintained, or by 
 using a current of 100 amperes flowing in mains between which 
 100 volts pressure difference is preserved. If, however, we 
 wish to dissipate or waste only the same fraction of our 10,000 
 watts power, say 10 per cent., in conveying it, we shall have,
 
 ELECTRIC DISTRIBUTION. 189 
 
 in the case of low pressure current, to employ electric mains 
 of 100 times less resistance in the second case than in the 
 first. In other words, we must provide a conducting channel 
 for the large current of 100 amperes which shall be 100 times 
 the cross-section, and therefore 100 times the conductivity, of 
 that which it would he necessary to lay down for conveying 
 the small current of 10 amperes, with the same proportionate 
 loss of energy. It will, therefore, be evident that, as far as 
 mere cost of copper is concerned, it is cheaper to convey 
 electric energy in the form of small electric currents at high 
 pressure than as a large electric current at lower pressure. 
 On the other hand, the high pressure conductors cost more 
 to insulate, so the advantage is not simply proportionate to 
 the economy in copper. 
 
 The point to which I wish now to direct your attention is 
 that we have in the tranfformer an appliance which enables 
 us to convert small currents of high pressure into large 
 currents of low pressure. In so doing we do not create any 
 energy in fact, we waste some of it ; but the transformer is 
 a device which enables us to change the form of electrical 
 power just as a pulley or other simple machine enables us to 
 change the form of mechanical power. By the use of a pulley 
 and tackle a man exerting a small force through a great 
 distance can raise a very heavy weight through a smaller 
 distance. 
 
 The alternate-current transformer is an appliance which 
 operates on electric power just as a lever or pulley does on 
 mechanical power. It consists of an iron ring or circuit of 
 some form or other which is wound over with two wires, a 
 thick wire and a thin one. These circuits are carefully 
 insulated one from the other. The thin wire generally makes 
 some ten or twenty times more turns round the iron ring 
 than the thick one. If a small alternating current of elec-
 
 190 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 tricity from a high pressure source is passed through the thin 
 wire, it creates an alternating magnetism in the ring of iron, 
 magnetising it first one way and then the other in a direc- 
 tion round the axis of the ring. This changing magnetism 
 creates, as before explained, another alternating electric current 
 in the thick-wire circuit, called a secondary current, and this 
 current can be made to have a lower pressure and greater 
 strength than the primary current by suitably proportioning 
 the number of windings of the two circuits. Hence we can 
 transform a small current produced by a pressure of 2,000 
 
 FIG. 75. One form of an Alternate-Current Transformer in process of 
 manufacture. 
 
 volts into a current nearly 20 times as strong, but having a 
 pressure only of 100 volts. In this transformation there* is 
 a certain loss of power, or dissipation of energy, but in good 
 transformers this will not exceed from 1 to 4 per cent, of the 
 amount of power, transformed. We can, then, by means of 

 
 ELECTR1 C DISTEIB UTION. 
 
 191 
 
 the transformer, carry electric power at high pressure and 
 transform it down to a low pressure at the place where it is 
 to be used. In transformer systems of electric supply this 
 
 is the method adopted, as opposed to the other or direct- 
 current supply, in which continuous currents flowing uniformly
 
 192 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 in one direction are used, and which distribute current at 
 a lower pressure for use without transformation. 
 
 The practical construction of the transformer is carried out 
 as follows : A number of thin iron bands, strips, or plates are 
 so arranged as to form an iron ring, or its equivalent. This 
 ring is then wound over with two circuits of insulated wire, 
 one a fine wire, called the primary circuit, and one a thick 
 wire, called the secondary. In Fig. 75 is shown such a trans- 
 former in process of manufacture. 
 
 The transformer, when completed, is generally enclosed in 
 an iron case, and is provided with proper terminals for the 
 two circuits (see Fig. 76). Transformers made as above are 
 generally constructed for various transforming powers, such 
 as 1 horse-power (H.P.), 10 H.P., &c. This means that the 
 transformer, say of 10 H.P., can transform the energy of a 
 current of about 4 amperes at a pressure of 2,000 volts into 
 energy in the form of a current of about 74 amperes at a 
 pressure of 100 volts. The sizes of transformers are often 
 given in kilowatts, and then we have to remember that one 
 kilowatt is equal to about 1 H.P. Hence a 30-kilowatt trans- 
 former is a 40-H.p. transformer. It will, therefore, be seen 
 that a transformer is only a development of the simple iron 
 ring wound over with two circuits, as described at page 180, 
 and illustrated in Fig. 69. The object of building up the 
 iron core, or ring of iron strips or plates, is to check, as far 
 as possible, the production of electric currents in the mass of 
 the iron core, which, if unprevented, would be a source of 
 waste of power. 
 
 "We are also able to present a similar simple explanation of 
 the mode of action of the dynamo machine. Referring again 
 to Fig. 70, we see that we can reverse or change the direc- 
 tion of the magnetic induction or current flowing through
 
 ELECTRIC DItiTKIB UTION. 
 
 193 
 
 the secondary circuit by making two cuts or air-gaps in the 
 iron ring, and then rotating that half of the ring on which 
 the secondary circuit is bound. This can, perhaps, be better 
 understood by reference to another diagram (see Fig. 77). In 
 this figure the shaded horse-shoe shaped part marked M is 
 supposed to represent an iron core, which is magnetised by an 
 electric current passed through a wire wound round that core, 
 but which wire, for the sake of simplicity, is not shown on the 
 diagram. This magnetic circuit is completed by placing 
 another iron mass, represented in the figure by the shaded 
 rectangle in between the poles or ends of the horse-shoe 
 magnet. We thus obtain an iron circuit having two cuts 
 or air-gaps in it, corresponding with the split ring shown in 
 Fig. 70. 
 
 FIG. 77. Skeleton Diagram, showing the principle of the Dynamo Machine 
 
 Suppose, now, a magnetic induction or magnetic current 
 to exist in one direction round this air-iron circuit, and 
 to be produced by a primary current flowing in a primary 
 circuit of insulated copper wire which is wound on the horse- 
 shoe shaped iron core. This magnetic induction or current 
 will pass across the air-gaps, producing there a powerful mag- 
 netic field. Next, let us suppose that the iron block, or, as it 
 is called, the armature core, is wound over with a circuit of 
 wire indicated by the rectangle A A in Fig. 77. Let this core 
 and armature wire be capable of rotating round an axis, and 
 suppose that, by a couple of contact rings B B, we can always 
 keep the ends of the armature circuit A A in conducting
 
 194 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 connection with an external circuit C- If the rectangular 
 core and its wire winding is made to revolve round its axis, it 
 is easy to see that the direction of the magnetic current or 
 induction through this secondary or armature circuit will be 
 continually reversed, and hence that an induced secondary 
 current will be created in it, which will also be continually 
 reversed that is, it will be an alternating current. 
 
 A dynamo, therefore, consists essentially of an iron core,, 
 or field mat /net, as it is termed, which is wound over with a 
 primary circuit or field wire circuit, and this circuit, when 
 traversed by an electric current, creates a powerful magnetic 
 field in the space between the poles or ends of the field 
 magnet. In the interpolar space is placed an iron cylinder 
 or ring, called the armature core, and this serves to complete 
 the magnetic circuit. On this armature core is wound another 
 coil or coils of wire called the armature windings, which 
 correspond to the secondary circuit of the transformer. The 
 armature core and its windings are rotated by being fixed 
 on a spindle to which is usually attached a pulley. In 
 Fig. 78 is shown a representation of an Edison-Hopkinson 
 dynamo, in which the field magnets and armature are easily 
 distinguished. In one large class of dynamos, called alternate- 
 current machines, or alternators, the secondary current which 
 is sent out from the secondary or armature circuit is, as 
 observed above, an alternating current. 
 
 For many purposes, however, it is essential to produce a 
 continuous current, or a current always in one direction. 
 To obtain this, an additional organ has to be furnished 
 to the dynamo. This is called a commutator, the function 
 of which is to convert the alternating current produced 
 in the armature coils into a continuous current in the 
 external circuit. A little attentive consideration will, how- 
 ever, show that a dynamo and transformer are essentially
 
 ELECTRIC DISTRIBUTION. 
 
 195 
 
 the same appliances in nature, only that in the latter the 
 necessary reversal of the flow of magnetism through the 
 secondary circuit is obtained by alternating or reversing the 
 
 FIG. 78. Edison- Hopkinson Dynamo. 
 
 direction of the primary current, whereas in the case of the- 
 dynamo the necessary reversal is obtained by rotating or 
 turning round that part of the iron magnetic circuit on which 
 
 o2
 
 196 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 the secondary circuit is wound. This rotation of the arma- 
 ture of the dynamo may, in practice, be effected either by 
 putting a pulley on the shaft and driving it by a belt or rope, 
 or by coupling the dynamo shaft directly to the shaft of a 
 high-speed steam engine. 
 
 One of the methods of direct driving, as this is called, most 
 widely adopted in England is to employ a high-speed Willans 
 engine, the shaft of which is continuous with that of the 
 
 FIG. 79. View of an Alternator direct driven by 
 Willans Engine. 
 
 jing coupled to a 
 
 dynamo. In Fig. 79 is shown an illustration of such a 
 combination. It is obvious that this arrangement is exceed- 
 ingly economical of space, and hence has been widely adopted 
 in the arrangement of electric generating stations, in which 
 space is of importance. 
 
 It is impossible, as observed at the beginning of this 
 Lecture, to do more than give here the briefest outline of the
 
 ELECTRIC DISTRIBUTION. 197 
 
 different methods employed for distributing current for illu- 
 minating purposes ; but the non-technical reader will probably 
 be able to grasp most easily the nature of the various 
 systems employed by my selecting one or two typical 
 instances, and describing these somewhat in detail. At the 
 present moment there are two principal systems of electric 
 lighting, both of which have their advantages : these are 
 called respectively the alternating current supply system 
 and the continuous current supply system. In the first of 
 these a generating station is established, which is provided 
 with dynamo - electric machines called alternators, which 
 are capable of producing an electric current alternating 
 in direction a large number of times in a second. In 
 the case of alternating currents the current flows in one 
 direction in the circuit for a short fraction of a second, is 
 then reversed, and then flows in the opposite direction for a 
 short fraction of a second, and repeats the same cycle of 
 operations continuously. The number of limes per second 
 which the current cycle is repeated namely, a flow in one 
 direction succeeded by a flow in the opposite direction, 
 like the ebb and flow of the tide is called the frequency 
 of the electric current, and the frequencies most usually 
 employed are either 40, 83, 100 or about 125. In England 
 the practice generally is to use a frequency of 80 or 100, 
 on the continent of Europe the lower frequency of 40, and 
 in America the higher of 120. This alternating current is 
 generated by machines called alternators at a high pressure, 
 which is generally either 1,000, 2,000 or 2,400 volts, and 
 in some cases 5,000 or 10,000. This high-pressure alter- 
 nating current is then led out from the station by highly 
 insulated conductors, which are called the primary mains, 
 and these primary mains are generally made in what is 
 called the concentric form. A stranded copper cable is covered 
 over with insulating material, and then a circular strand of 
 wire is plaited over the insulating material, and this again
 
 198 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 is protected by a further layer of insulation, and finally by a 
 covering of steel wires, called the armour. In Fig. 80 is 
 shown a section of this steel-armoured concentric cable. 
 
 This armoured cable is laid down in the streets under 
 the pavements, and is then usually conducted to certain 
 places which are called transformer sub-centres. These are 
 small rooms, either excavated out underground or built 
 above ground, and to them the high-pressure primary 
 
 FIG. 80. Cross-section of a Concentric Lead-covered Steel- Armoured 
 Cable, for High-Pressure Transmission. The central white dots represent 
 the inner conductor, and the outer circle of white dots the several wires 
 composing the outer conductor. The shaded portion shows the outer 
 covering of lead. 
 
 current is brought by these " primary " cables. In these 
 transformer sub-centres are placed a number of trans- 
 formers, which, as already explained, are connected with 
 the primary mains, so that all their primary circuits are 
 in parallel across the mains. The secondary circuits of 
 the transformers are then connected to another set of 
 underground mains, which are called the secondary dis- 
 tributing mains, and these distributing mains are laid down 
 in the streets which are to be furnished with light. It is
 
 ELECTRIC DISTRIBUTION. 
 
 199 
 
 most usual to employ a transformation ratio of 20 to 1, 
 so that the pressure of 2,000 volts is reduced to 100 volts 
 at the other side of the transformer. 
 
 In many cases the secondary distributing system is laid down 
 on the three-wire system (see Fig. 81). In the transformer 
 sub-centre are placed a number of transformers, T T, which 
 have primary and secondary switches, by which their primary 
 circuits can be connected to the primary mains MM, and their 
 secondary circuits to the secondary mains ABC. During the 
 
 FIG. 81. System of Distribution by Alternating Currents and Three- 
 wire Secondary Circuits. A is the Alternating Current Dynamo, M M 
 the Primary Mains, T T the Transformers, and A B and C the Three- 
 wire Secondary Circuits with Lamps across them. 
 
 hours of small demand (or "light load," as it is technically 
 called) only one of these transformers, which is called the 
 master transformer, is kept connected to the system; but 
 during the hours of heavy demand (or " full load ") two or 
 more transformers are added on, so as to share the load. In 
 some cases of very scattered lighting, instead of collecting 
 together the transformers in sub-centres, one or more trans- 
 formers are placed in each house or building which is to 
 be supplied with current. 
 
 In such a transformer system the sources of losses of energy 
 are as follow : There is a certain constant loss of energy in
 
 200 -ELECTRIC LAMPS ASD ELECTRIC LIGHTING. 
 
 every transformer which is kept connected to the primary 
 mains, owing to the fact that the incessant reversal of the 
 magnetisation of the iron core of the transformer necessitates 
 an expenditure of energy. This is called the " open circuit 
 loss" of the transformer, and in thoroughly efficient trans- 
 formers does not exceed from one to two per cent, of the 
 whole of the nominal output of the transformer. Thus, 
 for instance, if the transformer is one which is capable 
 of transforming 10 horse-power of electric energy from the 
 form in which it exists as a current of about four amperes 
 flowing under a pressure of 2,000 volts into electric energy 
 which exists in the form of 75 amperes at a pressure of 
 100 volts, such a transformer can be made to dissipate or 
 waste only about one-seventh of a horse-power in the con- 
 tinual magnetisation of its iron core. When the transformer 
 is doing its full work, or is loaded up to any extent, in 
 addition to this open circuit loss which is also called the 
 iron loss in the transformer, because it takes place in the iron 
 core there is an additional loss, which is called the " copper 
 loss," and this is due to the heating effect produced by the 
 currents in the primary and secondary copper coils of the 
 transformer. In well-designed transformers the copper loss 
 at full load is about equal to the iron loss at no load, and 
 it will therefore be easily seen that the result is to give the 
 transformer an " efficiency," as it is called, of about 97 per 
 cent, at full load that is to say, of the whole power supplied 
 to the transformer in the form of high -pressure current 97 
 per cent, is given out by the transformer in the form of low- 
 pressure current. It is possible to construct a transformer so 
 as to have as much as 90 per cent, efficiency at ^th of full 
 load. The most important point to consider, however, is, not 
 the efficiency of the transformer at full load, or its efficiency at 
 light loads, but to consider the so-called all-day efficiency of the 
 transformer that is to say, if the transformer is employed for 
 24 hours, doing such lighting as may be demanded of it, the
 
 ELECTRIC DISTRIBUTION. 201 
 
 important quality which it should possess is that of having a 
 high all-day efficiency ; in other words, the ratio between the 
 total number of units of electrical energy which are taken out 
 of the transformer must bear a large proportion to the number 
 put into it. 
 
 As already explained in the Lecture on " Electric Glow 
 Lamps," the ordinary demands for lighting in residential or 
 public buildings are of so varied a nature that the full current 
 for the whole of the lamps placed in them is only required for a 
 very small portion of the 24 hours. For instance, in an ordinary 
 private residence, during the early part of the day there may 
 be a small demand for lighting in the lower part of the house ; 
 during the greater part of the day very little ; in the afternoon 
 some more lamps will be turned on ; in the evening (in the 
 winter between 4 and 10 or 11) there will be a variable but 
 much larger demand; and during the night very little. If on 
 a piece of paper a line is taken and divided into twenty-four 
 parts, and a perpendicular drawn at each of those points 
 representing by its altitude the number of amperes of current 
 taken by the house at that moment, a curve joining the 
 tops of all these lines gives us the load (Uaijram of the 
 house, and the area of this load diagram gives the whole 
 quantity of current taken by the house. If, instead of 
 erecting perpendiculars proportional to the current, we erect 
 perpendiculars which are proportional to the power measured 
 in watts being taken up in the house at that time, we get 
 the watt-hour diagram, and the whole area of this diagram 
 measures the total amount of energy in Board of Trade 
 units taken up in the house during the twenty-four hours. 
 The ratio between this amount and the full amount which 
 would be taken if all the lamps were kept burning for the 
 whole period is called the load factor of the house, and the 
 load factor of the house is on an average not more than 10 
 per cent.
 
 502 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 Accordingly, if a house or a series of houses are supplied 
 from one transformer centre, the actual demand at any 
 one moment for electric power may be very small ; and the 
 important point with regard to transformers is that they 
 should possess a very high efficiency at low loads, in order 
 that, on the whole, the number of units of electrical energy 
 that are wasted in the transformers in magnetising the iron 
 ores may not be a very large proportion of the total energy 
 which is supplied to the houses. Transformers can now be 
 easily made to have an 80 per cent, all-day efficiency, and, 
 under these circumstances, the total amount of energy supplied 
 to the customers may be, and often is, as much as 80 per cent, 
 of that which is sent out from the station. In inferior 
 systems of alternating current supply the efficiency of distri- 
 bution may fall very far short of 80 per cent., not amounting 
 to even more than 50 per cent. 
 
 Having placed these general principles before our minds, 
 we can now proceed to consider the main features of an 
 existing alternating current supply system, which has been 
 worked out with great care and forethought, namely, that 
 adopted for the supply of electric energy for the lighting of 
 the city of Rome. The electric lighting of Kome is conducted 
 from two electric lighting stations, one of which has been 
 established in the gas works at Rome, at the foot of the 
 Palatine Hill, and the other at Tivoli, eighteen miles from 
 Rome. The first station was put down in the year 1889 with 
 the object of supplying current for incandescent and arc 
 lighting within the city. This station occupies the site of 
 the old Circus Maximus, and at the present moment it 
 supplies an equivalent of nearly 40,000 10-c.p. incandescent 
 lamps. The station itself is a substantial brick building in 
 close contiguity to the gasworks. It contains six Ganz alter- 
 nating current dynamos four of GOO horse-power and two 
 of 250. Each alternator consists of a series of armature
 
 ELECTRIC DISTRIBUTION. 
 
 203 
 
 coils which are contained on the inside of an iron ring frame. 
 A central wheel, which constitutes the flywheel of the engine 
 
 FIG. 82. 600 horse-power Alternator in the Cerchi Elecrric Station at 
 Home. 
 
 FIG. 83. 600 horse-power Alternator in the Cerchi Electric Station at 
 Rome. 
 
 driving the dynamo, carries on its external surface a sent* of 
 insulated coils, which are called the field-magnet coils. The
 
 204 
 
 ELECTRIC LAMPS ANI> ELECTRIC LIGHTING. 
 
 engines are coupled direct to this flywheel, the cylinders of 
 the engines being one on either side. In Figs. 82 and 83 are 
 shown views of these 600 H.P. alternators with the direct- 
 coupled engines. Smaller alternators of 250 H.P. are used 
 during the daytime, or at a time when the demand for light- 
 ing is not great. These alternators generate an alternating 
 current at a frequency of 40 and a pressure of 2,000 volts. 
 
 FIG. 84. View of the Hydraulic Tower at Tivoli. 
 
 They send their current into three primary cables or feeders, 
 which are concentric steel-armoured conductors, and these 
 feeders deliver current to certain transformer centres and to 
 transformers which are placed in the various buildings to be 
 lighted. The total length of mains laid down is about 25 
 kilometres, or 16 miles. At the end of 1892 there was a total
 
 ELECTRIC DISTRIBUTION. 
 
 205 
 
 number of lamps connected to the system equal to 34,46 J 
 10-c.p. lamps. In the transformer centres and in the houses 
 are placed transformers for reducing the pressure from 2,000 
 volts to 100 volts, two sizes of transformers being mostly 
 used, one capable of transforming power ejual to 10,000 
 
 FIG. 85. View of Power Station at Tivoli. 
 
 watts and the other equal to transforming power of 5,000 
 watts. From these transformers secondary cables are laid, 
 distributing secondary current at a pressure of 100 volts 
 to various buildings.
 
 206 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 The capacity of this station being fully exhausted, it was 
 determined to furnish an additional supply by means of a 
 second station, and to take advantage for this purpose of the 
 large available water supply of Tivoli. The village of Tivoli, 
 which occupies the site of the ancient Tibur, stands on a spur 
 of the Sabine Hills. The beauties of Tivoli and its surrounding 
 country were celebrated in undying verse by Horace, whose 
 Sabine farm was not very far distant. The chief natural 
 attraction is the fine cascade formed by the River Anio, one 
 fall of which is 340 feet in height. Here also in classic times 
 stood Hadrian's villa, of which the grounds once covered an 
 area of several square miles, and contained an unrivalled 
 collection of works of art. Upon this romantic spot the 
 modern engineer has laid his hands, and has compelled a large 
 portion of the power running to waste in these waterfalls to be 
 directed to the purpose of electrically lighting modern Rome. 
 With this object, a hydraulic canal was first constructed, 
 leading from one of the upper reaches of the River Anio to 
 the top of a tall tower (see Fig. 84). This tower contains 
 an iron hydraulic fall tube six feet in diameter. At a distance 
 of 150 feet below the top of this tower, and about half-way 
 down the side of the hill a power house (see Fig. 85) has 
 been constructed, into which the termination of this hydraulic 
 fall tube is brought. The water from the upper level conveyed 
 by the hydraulic canal falls down this hydraulic main, and the 
 tube has a capacity for delivering 100 cubic feet of water per 
 second at a pressure equal to that due to a height of 150 feet, 
 which is about equal to 90 Ib. on the square inch. The end 
 of this hydraulic tube terminates in three lateral branches, 
 each three feet in diameter, and closed with a hydraulic valve 
 capable of resisting the enormous pressure of the water above. 
 Each branch of the main also sends off three subsidiary 
 branches, which are led to three Girard turbines or hydraulic 
 motors. These machines are practically water engines, in 
 which the flow of water is made to cause the revolution
 
 ELECTRIC DIS TRIE UTION. 
 
 207 
 
 of a wheel with curved blades which is enclosed in an iron 
 case. In one room of the power house there are nine of 
 these turbines, six of 350 H.P. and three of 50 H.P. The 
 water that passes through these turbines is delivered into a 
 tailrace, and then returns back again into the Tivoli Falls 
 at a lower level. Coupled to each of those turbines is a Ganz 
 alternating- current dynamo. Six of these dynamos are of 
 350 H.P. each, and can furnish an alternating current of 45 
 
 FIG. 86. Turbine and Excitor Dynamo in the Tivoli Station. 
 Rome System. 
 
 amperes at a pressure of 5,000 volts ; three of the dynamos 
 are direct-current machines of 50 H.P., and can furnish a 
 continuous current of 150 amperes at a pressure of 125 
 volts. These smaller dynamos are employed for furnish- 
 ing the current required to magnetise the field magnets of 
 the larger machines. A view of one of these larger machines
 
 208 ELECTXIC LAMPS AXD ELECTRIC LIGHTING, 
 
 and its associated turbine is shown in Fig. 86. These six 
 large alternators send their current to a switchboard which 
 is 60 feet long and 15 feet high. From the power house 
 proceed four stranded copper cables, each consisting of 19 
 copper wires, the over-all diameter of each cable being about 
 one inch. Each cable weighs 980 kilogrammes per kilo- 
 metre, or nearly two tons per mile. These cables convey 
 the current across the Campagna from Tivoli to Borne, a 
 distance of about 18 miles, and in all four lines there is a 
 total weight of about 120 tons of copper. These four cables 
 
 FIG. 87. The Half-way House. Tivoli-Rome Electric Lighting System. 
 
 are supported by porcelain insulators, which are carried upon 
 crossbars on the iron posts, placed at intervals throughout 
 the whole length of the line. These porcelain insulators are 
 similar to those used for carrying telegraphic lines, with the 
 addition that they contain an arrangement for keeping the 
 edges of the insulator moistened with a highly insulating 
 oil. These oil insulators have been designed specially with 
 the object of insulating the line for the 5,000 volts pressure.
 
 ELECTRIC DISTRIBUTION. 
 
 209 
 
 Half-way between Tivoli and Rome there is a half-way 
 house (see Fig. 87), in which dwells a custodian whose duty 
 it is to inspect the whole of the line. The power station 
 at Tivoli provides in all a total of 2,000 H.P. in the form 
 of an electric current at a pressure of 5,000 volts. The 
 copper transmission cables are made of hard -drawn copper 
 wire, and of a cross-section of 100 square millimetres. 
 The poles carrying the cables are placed 85 metres apart, 
 
 FIG. 88. View of the Switchboard Arrangements in the Porta Pia Trans- 
 former House. Tivoli-Rome System. 
 
 in almost a straight line across the Campagna, and 
 there are 707 poles altogether between Tivoli and Rome. 
 The resistance of each cable is about 38 ohms for the 
 eighteen miles. Each cable can convey 100 amperes, and 
 the four cables that is, the two pairs can therefore con- 
 vey 200 amperes from Tivoli to Rome, which is equal to 
 the output of five of the large alternators. In the passage 
 from Tivoli to Rome the current undergoes a fall in pressure
 
 210 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 due to the resistance of the line, which is about 800 ohms in 
 the eighteen miles when working at full load. The cables, 
 therefore, di ssipate about 16 per cent, of the whole of the power 
 capable of being transmitted at full load. The alternators 
 have been so constructed that they can all work in parallel, 
 sending their current into one or both pairs of cables as may- 
 be desired. On arriving at Home the current enters a 
 transformer house (see Fig. 88) placed just outside the Porta 
 Pia of Rome. In this transformer house are placed a 
 series of transformers, sixteen in number, each capable 
 of transforming 30,000 watts from 5,000 to 2,000 volts. This 
 transformer house is a substantially-built stone building 
 two stories in height. The current, reduced from 5,000 to 
 2,000 volts, is then transmitted by underground cables through 
 the streets of Eome to certain transformer centres, where it is 
 again reduced in pressure to 100 volts. In addition to this, a 
 portion of the current, at a pressure of 2,000 volts, is utilised 
 for working a series of alternating current arc lamps. There 
 are six circuits of these arc lamps, each circuit being arranged 
 for 48 lamps. Each series of lamps takes a current of 14 
 amperes, and there is a special automatic pressure-regulator 
 attached to the transformer supplying these arc light circuits, 
 so that the current is kept perfectly constant, whatever may be 
 the number of lamps upon the circuit. In addition to the arc 
 light circuits, there are five or six incandescent lighting circuits 
 consisting of Siemens concentric steel- armoured cables laid 
 underground. The current supplied from Tivoli can be 
 arranged to assist the Cerchi station, so that the two stations, 
 18 miles apart, one in the gasworks and the other at Tivoli, 
 assist one another in furnishing current for incandescent 
 and arc lighting in Rome as may be required. This Tivoli 
 station is one of the finest examples of an alternating-current 
 station operated by water-power, and it has been in perfectly 
 successful operation since the year 1890. In order to protect 
 the overhead line, going over 18 miles along the Campagna,
 
 ELECTRIC DISTRIBUTION. 211 
 
 from damage from lightning there are lightning protectors 
 of the kind described on page 145, placed at Tivoli, in the 
 transformer house at Rome, and in the half-way house shown 
 in Fig. 87. These protectors have proved sufficient to guard 
 the dynamos and transformers from danger by lightning 
 stroke. 
 
 The general arrangement of a low-pressure continuous cur- 
 rent station on the three-wire system may be described as 
 follows : In the station are placed a series of dynamos, each 
 dynamo in most English practice being direct-coupled to a 
 high-pressure compound steam engine. In Fig. 79 is shown 
 
 FIG. 89. Diagram showing the arrangement of Dynamos D D, and 
 Lamps L on the " Three-wire " System of Distribution. 
 
 such a steam dynamo, in which the armature shaft is coupled 
 direct to the crank shaft of a Willans engine. This combina- 
 tion of engine and dynamo on the same bedplate is obviously 
 very economical in floor space, and has therefore naturally 
 become very popular in places where space is valuable. 
 A series of these engine-dynamo sets is generally arranged 
 in a central station, each set being either similar in size 
 and appearance, or else in graduated sizes. In Fig. 91 
 is shown the interior of the St. Pancras electric lighting 
 station, and in Fig. 92 the interior of the central station at 
 Glasgow. In each of these the steam dynamos are arranged 
 to generate current generally at either 110 volts or 220 volts, 
 delivering this current to a switchboard. If the current is
 
 212 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 supplied direct to the circuits, then the dynamos are joined in 
 pairs between the three bars of the switchboard, as shown in 
 Figs. 89 and 90. From the ends of these " omnibus bars, ' 
 as they are called, are laid out a series of feeding mains or 
 feeders, which feeders terminate in the distributing mains laid 
 down hi the streets of the district to be supplied. The circuits 
 of the various houses are connected up to these triple distribut- 
 ing mams in such a manner that as nearly as possible one-half 
 of the lamps are joined in between one pair of mains and the 
 
 FIG. 90. Dynamos in a Three-wire Low-Pressure Station. 
 
 other half between the other. In laying out such a district 
 great discretion and knowledge have to be brought to bear in 
 selecting points, which are called the feeder centres, so that at 
 those points the pressure may be kept as constant as possible. 
 Each dynamo in the central station may be regarded as a 
 pump, pumping its current into the positive main omnibus 
 bar, from which it flows out by the feeders to the distributing 
 circuit and returns back again to the opposite main. During 
 the hours of heavy demand for current there is, of course, a
 
 ELECTRIC DISTRIBUTION. 213 
 
 large current flowing out by these feeders, and a corres- 
 pondingly large fall of pressure down the feeder. Hence the 
 pressure at the terminals of the dynamos has to be kept up in 
 order to allow for this fall. It is usual to maintain the pres- 
 sure at certain feeding points constant at 100 or 110 volts, 
 whatever may be the nature of the supply, and to do this the 
 pressure at the dynamo terminals has to be varied from 110 to 
 150 volts or thereabouts, as the demand varies at different 
 hours of the day or night. It will thus be seen that in 
 the feeders in a low-pressure station the loss of energy is 
 greatest at the time of greatest load, because they are then 
 traversed by the largest currents, and, therefore, the total 
 supply of energy from the dynamos has to be equal to the 
 amount required by the lamps in use plus that required to 
 supply the corresponding loss in the feeders. A low-pressure 
 continuous current station has, consequently, its least efficiency 
 of distribution at the time of fullest load. Exactly the opposite 
 is the case with the alternating current supply. In this case 
 *, the efficiency of distribution is a maximum at the time of 
 largest demand. 
 
 We may take as a good example of a low-pressure station 
 the St. Pancras Vestry Electric Lighting Station, which 
 has been carried out by the local authority of the parish 
 of St. Pancras, in the north-west of London, for the supply 
 of electrical energy for lighting and motive power purposes 
 in that district. In this part of London, which is densely 
 populated, it was considered advisable to adopt the continuous 
 current low-pressure system, and a typical station of this kind 
 was, therefore, designed for the district by Prof. Henry 
 Kobinson. The following is a description of the works as 
 given by him : The station buildings consist of an engine- 
 house 106ft. by 26ft., a boiler-house 53ft. by 35ft., a coal 
 store 43ft. by lift., a battery room 40ft. by 14ft. 6in., as 
 well as a testing-room, office, stores, and an underground
 
 214 ELECTRIC LAMPti AND ELECTRIC LIGHTING. 
 
 tank for condensing water capable of containing 170,000 
 gallons of water, and a chimney shaft 5ft. square inside 
 and 90ft. high. The dynamo room contains eleven 
 engines and dynamos, which are erected on a concrete 
 foundation, surrounded by sand to prevent vibration being 
 communicated to the walls. The floor is carried indepen- 
 dently of the engine foundations by cantilevers from the 
 walls. The engines employed are of the Willans compound 
 
 FIG. 91. View of the Interior of St. Pancras Electric Lighting Station, 
 London, showing the Dynamos. 
 
 central valve type, and the dynamos are of the 6-pole Kapp 
 type made by Messrs. Johnson and Phillips. A view of the 
 interior of this station, showing the long perspective of 
 engines and dynamos, is given in Fig. 91. Each Willans 
 engine is coupled direct to its corresponding dynamo. These 
 dynamos furnish a continuous current, nine of them giving 
 680 amperes at a pressure varying from 112 to 180 volts, and 
 three of them giving 145 volts, with a small current for 
 charging secondary batteries at a distant sub-station. The
 
 ELECTRIC DISTRIBUTION. 215 
 
 remaining two dynamos are wound for an output of 90 
 amperes at 540 to 575 volts. These are used for working 
 the street arc lamps as well as for charging in series four 
 sets of storage batteries at the central station which are 
 capable of discharging at a rate of 60 to 75 amperes. The 
 boiler plant consists of five Babcock-Wilcox boilers, each 
 capable of evaporating over 5,000lb. of water per hour, the 
 working pressure being 170lb. on the square inch. Along 
 the top of these runs a steam main into which all the boilers 
 deliver their steam and from which the engines separately 
 take their steam. The steam, having done its work in the 
 engines, is then either delivered into the atmosphere or passes 
 into a jet condenser, which draws its water from the bottom 
 of the underground tank. From the top of this tank the hot 
 water is pumped by independent pumps to a cooling appa- 
 ratus, which is capable of dealing with a minimum of 10,000 
 gallons of water per hour. The cooling arrangement consists 
 of a large surface of corrugated sheet iron attached to a 
 framing placed round the chimney. The water, after flowing 
 along the corrugated surface, is collected underneath and 
 returned to the bottom of the tank. There is also an air 
 pump in the station for delivering dry air into the street main 
 culverts at the rate of 5,000 cubic feet of air per hour. 
 
 The street lamps are Brockie-Pell arc lamps, placed on tall 
 iron posts placed in the centre of the road where admissible. 
 They are fixed at distances varying from 160ft. to 245ft. 
 apart, the height of the lamp varying from 22ft. to 25ft. 
 above the pavement. These lamps are worked 11 in series 
 near the central station and 10 in series at a distance, and 
 each lamp takes a current of 10 amperes, which is sup- 
 plied by the two 500- volt dynamo machines. The main 
 switchboard for the incandescent lighting is arranged on the 
 "three-wire" system. Each dynamo is provided with a 
 double-pole switch and copper coupling strips to connect the
 
 210 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 machines to the third or middle wire and to the different bars 
 on the feeder board on either side of the circuit. There is an 
 amperemeter on each dynamo to indicate the current being 
 given out. On the switchboard there are four positive 
 omnibus bars and four negative bars, and at full load one 
 machine can be run on to each bar and be worked at any 
 pressure which may be necessary to serve the feeders which 
 are switched on to it. Thus the dynamos may be worked at 
 different pressures to suit the demand in the districts which 
 they serve. 
 
 There are seven feeder mains going out of the station, in 
 addition to a direct supply to the distributing mains. The 
 mains are laid throughout the district on the three-wire 
 system, the principal mains being sufficient to supply 25,000 
 incandescent lamps of 16-candle power in use simultaneously. 
 These conductors are laid down under the streets, and are 
 composed of copper strips 1 Jin. wide and iin. thick, supported 
 on edge in glazed porcelain insulators, which are carried on 
 small cast-iron brackets built into the walls of the culverts. 
 Some of the mains are cables laid in cast-iron pipes, and some 
 of them are armoured cables. In addition to this, the current 
 is supplied from the central station to a distant secondary 
 battery station 1,140 yards away, which contains two batteries 
 of 58 E.P.S. cells each. By varying the number of dynamos 
 in use at the station and the number of feeders in connec- 
 tion with the omnibus bars, the engineers in charge of the 
 station have it in their power to regulate the electric pressure 
 between the distributing mains laid down in the streets, and 
 the supply is so regulated as to keep the pressure between the 
 middle main or wire and each of the outer wires to 110 volts. 
 
 A very similar station has been erected for the supply of 
 electric current in Glasgow by the Glasgow Corporation. A 
 view of part of the interior of this station is shown in Fig. 92.
 
 ELECTRIC DISTRIBUTION. 
 
 217 
 
 The station buildings are substantially built, standing on a 
 layer of concrete two feet thick laid throughout the entire block. 
 These buildings consist of an engine and dynamo room, boiler 
 room, workshop, stores, and offices. The boiler room con- 
 tains six steel boilers of the marine type, measuring 12ft. by 
 10ft. in diameter, and working at a pressure of 1601b. Each 
 contains two furnaces of the corrugated type, with an internal 
 diameter of 3ft. These boilers deliver steam through a ring 
 
 FIG. 92. View of Part of the Glasgow Corporation Electric Lighting 
 Station. 
 
 main into the engine room. The engine room contains a series 
 of Willans engines coupled to dynamos (see Fig. 92). The 
 engines are Willans central valve compound engines, and the 
 nine dynamos are two-pole shunt-wound machines with 
 drum armatures. The boilers are fed with gas coke obtained 
 from the Corporation gas works. Over the boiler house is 
 placed an accumulator room containing 114 secondary 
 battery cells. Each cell contains 61 lead plates, and has a 
 capacity for storing up and discharging 1,000 ampere-hours
 
 218 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 of electrical quantity. The current from the dynamos, 
 which is generated at from 210 to 230 volts, is led to 
 a distributing board in the station, and this supply has 
 a three-wire system of distribution laid down in the 
 streets. The three copper conductors consist of copper 
 strip, which is laid on porcelain insulators contained in iron 
 culverts. The current is brought to the two outer wires of 
 the three-wire strip by conductors, which, as in the St. 
 Pancras station before described, are called " feeders," and 
 which consist of only two mains, a positive and a negative 
 main. The lamps are all joined in parallel between the 
 middle main of the distributing mains and one of the outer 
 ones, the total of about 46,000 eight candle-power lamps being, 
 as far as possible, connected so that one-half of the lamps are 
 ioined in between the middle main and one of the others, 
 called the "positive," and the other half of the lamps being 
 cpnnected in between the middle main and the other conductor, 
 called the " negative." The middle wire, therefore, serves to 
 carry the balance of current from one part of the distributing 
 system to the other at those times when the number of lamps 
 joined in between the two sides of the three-wire system is not 
 exactly 
 
 These two stations may be taken as fairly typical examples 
 of low-pressure stations as established in Great Britain, 
 although many other equally good, such as the Kensington 
 and Knightsbridge and Westminster stations in London, 
 might be described. Such brief space as we are able to 
 give to this portion of the subject in no way enables us 
 to exhaust the description of the whole of the methods for a 
 direct current low-pressure supply ; but the general tendency, 
 at any rate in England, is to adopt the direct-driven type 
 of dynamo, on account of the considerable economy which 
 is thereby effected in space. In some low-pressure stations, 
 where space is not of much importance, it is found convenient
 
 ELECTRIC DISTRIBUTION. 
 
 219 
 
 to drive the dynamos by means of belts from the flywheels 
 of engines. 
 
 In Fig. 98 is shown a portion of the inteiior of a Con- 
 tinental electric lighting station that of the City of Brussels.
 
 220 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 This station is constructed to contain plant of 3,000 indi- 
 cated horse-power, the unit adopted being a 500 horse-power 
 engine driving two 250 horse-power dynamos. Provision is 
 made for placing six such units, one being for reserve. The 
 present boiler plant consists of three Babcock-Wilcox water- 
 tube boilers capable of evaporating 7,500lb. of water per 
 hour. The engine room contains engines, at present two in 
 number, of the horizontal compound condensing type. The 
 low-pressure cylinders are 40 inches in diameter, and the 
 high-pressure cylinders 26 inches in diameter, the length of 
 the stroke being 4 ft. The speed is capable of variation 
 between 62 and 75 revolutions per minute. Each engine is 
 fitted with two grooved flywheels 20 feet in diameter, and 
 weighing 20 tons. These drive the dynamos attached to each 
 engine by means of ropes. The dynamos shown in the 
 illustration are of the four-pole type, shunt-wound, with 
 drum armatures, and with an output capacity of 145 kilowatts. 
 These dynamos supply a three-wire distributing system, the 
 cables being laid down underground in iron pipes. 
 
 In addition to the generating plant a storage plant of 
 accumulators is provided, in two batteries of 70 cells 
 each. Each battery is capable of giving a normal discharge 
 of 350 amperes for ten hours. The underground cables 
 consist of copper-stranded cables insulated with india- 
 rubber, which are drawn into iron pipes. This work was 
 designed and carried out by the engineers of the Silver- 
 town Company. 
 
 The above descriptions of alternating current and con- 
 tinuous current generating stations will be sufficient to give 
 the student a general view of the methods which are at 
 present (1894) employed for the generation of electric current 
 for illuminating purposes. It would lead us into matters 
 too highly technical to attempt to discuss fully when, and
 
 ELECTRIC DISTRIBUTION. 221 
 
 under what conditions, each system finds its best appli- 
 cation. Suffice it to say that no one system can be described 
 as the best system ; both the low-pressure direct current and 
 the high-pressure alternating current systems have certain 
 peculiar advantages and disadvantages which have to be 
 considered in the designs for a system of electric distri- 
 bution. The engineer who is called upon at the present 
 time to design and carry out a system of electric supply 
 for public purposes has to consider a large number of facts 
 which determine the choice that shall be made of the method 
 of supply. Broadly speaking, the low-pressure continuous 
 current system finds its best field in areas in which the 
 density of the lighting is considerable. 
 
 Taking such cases as the central areas in large towns, it 
 will be found that at the present time the demand for lighting 
 will vary from one 8-c.p. lamp to three 8-c.p. lamps per yard 
 of distributing main or of house frontage. In the outer, 
 portions of large towns and in smaller country ones the demand 
 for light is much more scattered, and in these cases the only 
 possible system of supply which meets the conditions is the 
 alternating current system. It is possible to some extent to 
 combine together the advantages of the two systems. The 
 generating plant in the station can be made to furnish an 
 alternating current at a low pressure. This can be used to 
 supply the district within f of a mile or a mile round the 
 station with current furnished directly from the generating 
 machines. Transformers can then be used to raise this 
 pressure and transmit the current to the distant portions of 
 the area of supply, and tlie pressure can then be reduced, 
 either in sub-centres or in the houses, by means of 
 transformers, to the normal pressure of 100 volts. 
 
 For a fuller discussion of the questions of electrical distribu- 
 tion we must refer those who desire to know more of these
 
 222 ELECTRIC LAMPS AND ELECTRIC LIGHTING. 
 
 matters to the many special treatises and text books in which 
 the problems are handled in a thoroughly technical manner. 
 The aim in these Lectures has been merely so to place the broad 
 outlines of the subject before the general reader that the way 
 may be prepared for more advanced instruction. No one at 
 the present time can afford to be entirely ignorant of the 
 principles which underlie the industrial uses of electricity. By 
 the end of the nineteenth century all large towns in the world 
 will be underlayed by a network of copper conductors dis- 
 tributing from house to house the electric current as an 
 essential requisite of modern life. When in 1901 we celebrate 
 the centenary of the birth of the galvanic battery, by which 
 Volta in 1801 gave us the first practical means of producing 
 this wonder-working agent, we shall witness the opening of 
 an era in which the utilisation of electric current for the 
 purposes of domestic life will have, to us, as yet uncorn- 
 prehended consequences in serving to enhance the comfort 
 and convenience of mankind.
 
 INDEX TO CONTENTS.
 
 INDEX. 
 
 PAGE 
 
 Absorption of Light 31 
 
 Alternate Current Transformer 189 
 
 Ampere, The 9 
 
 The Definition of the 10 
 
 Ampere-meter 68 
 
 Ampere turns 175 
 
 Amplitude 28 
 
 Annulment of Edison Effect 109 
 
 Arc, Crater Temperature of 157 
 
 Arc Discharge 127, 134 
 
 Arc, Electric 139 
 
 between Various 
 
 Materials 152 
 
 'ArVLamp, Hand-Kegulated 137 
 
 - Inverted 167 
 
 Mechanism 162 
 
 Armoured Cable 198 
 
 Artistic Electric Lighting 98 
 
 Bamboo Filament 56 
 
 Bernstein Glow Lamp 58 
 
 Blackening of Glow Lamps 74 
 
 Board of Trade Unit 90 
 
 Board of Trade Inquiry at West- 
 minster 3 
 
 Bottomley's Experiments on Radia- 
 tion 23 
 
 Brightness 34 
 
 Brush Discharge 127 
 
 Brussels, Electric Lighting Station at 219 
 
 Bunsen Photometer 43 
 
 Cable, Armoured 198 
 
 Candle-Foot .. 38 
 
 Lamp 
 
 Power 
 
 Standard 
 
 Carbon, Advantage for Electric Arcs 
 
 Boiling Point of 
 
 - Deposit of in Glow Lamp ... 
 
 Filament 
 
 67 
 
 69 
 
 37 
 
 152 
 
 153 
 
 74 
 
 56 
 
 Molecule, Negative Charge of 111 
 53 
 
 55 
 
 Various Forms of 
 
 Cellulose... 
 
 Cellulose Solubility of in Zinc- 
 Chloride 
 
 Characteristic Curves of Glow Lamps 
 
 Chemical Power of an Electric 
 Current 
 
 Colour 
 
 Causes of 
 
 First Visible 
 
 Production of, by Absorp- 
 
 tion ... 
 
 Colour-Distinguishing Powers of 
 Lights 
 
 Comparison of Lights by Spectro- 
 Photometer 
 
 Comparison of Lights of Different 
 Colours 
 
 Comparison of Sunlight and Moon- 
 light 
 
 Comparison of Surfaces in respect of 
 Luminosity 
 
 Convection 
 
 Cost of Electric Lighting 
 
 of Incandescent Lighting 
 
 Coulomb, The Unit called the 
 
 Crater of Electric Arc 
 
 Curve connecting Candle-Power and 
 Current for Glow Lamp 
 
 Current, Magnetic Field of 
 
 Davenport Arc Lamp 
 
 Davy, Sir Humphry 
 
 Experiments on Arc 
 
 Discharge, Forms of Electric 
 
 Disruptive Discharge 
 
 Draper's Experiments on Radiation.. 
 
 Dynamo, Edison-Hopkinson 
 
 Principle of 
 
 Edison, T. A. 
 
 Effect in Glow Lamps 
 
 Edison-Swan Glow Lamp 
 
 Efficiency of Light of a Fire-Fly ... 
 of Various Sources of Light 
 
 Electric Arc 
 
 36 
 39 
 
 33 
 21 
 91 
 
 92 
 177 
 139 
 
 71 
 7 
 
 137 
 
 134 
 
 135 
 
 127 
 
 127 
 
 21 
 
 195 
 
 186 
 
 52 
 
 106 
 
 56 
 
 27 
 
 27 
 
 139
 
 226 
 
 INDEX. 
 
 Electric Arc, Alternating 149 
 
 Continuous 147 
 
 Crater Temperature of 143 
 
 Distribution of Light 
 
 from 147 
 
 Edison Effect in i5 
 
 Long and Short 140 
 
 Mechanism 162 
 
 Spectrum of 140 
 
 Unilateral Conduc- 
 tivity of 
 
 Electric Current, Magnetic Field of... 
 Currents, Generation of 
 
 Magnetic Power of 173 
 
 Electric Discharge 127, 129 
 
 Electiic Distribution, Test of Uni- 
 formity of 102 
 
 Electric Light and Gas 5, 91 
 
 Electric Lighting Act 
 
 Amendment Act ... 
 
 Committee of House 
 
 of Commons 
 
 Costof 
 
 in Great Britain ... 
 
 in London .... 
 
 Progress of 
 
 Station. Brussels ... 
 Station, Glasgow... 
 
 Station, Rome 
 
 Station, St. Pancras 
 Station, Tivoli 
 
 Electric Potential ... 
 
 Pressure 
 
 Resistance 
 
 Power, Transmission of 
 
 Units 
 
 Wiring 
 
 Electrical Exhibition, Crystal Palace 
 
 Electrolysis 
 
 Electrolytic Cell for Lantern 
 
 Electrostatic Voltmeter 
 
 Energy, Consumption of, by Glow- 
 Lamps 
 
 Evolution of Incandescent Lamp ... 
 
 Fall of Pressure, Electric 
 
 down Wire. 
 
 in Water Tube 
 
 Field, Magnetic 
 
 Filament of Glow Lamp 
 
 Bamboo 
 
 Carbon 
 
 Focus Lamp. Edison-Swan 
 
 Foot-pound, Definition of the 
 
 Frequency 
 
 Fundamental Terms, Meauing of 
 Gas and Electric Light.. 
 
 2 
 3 
 
 1 
 
 92 
 1 
 5 
 4 
 
 219 
 
 216 
 
 2C3 
 
 214 
 
 205 
 
 11 
 
 11 
 
 16 
 
 188 
 
 9 
 
 93 
 2 
 
 14 
 
 95 
 51 
 12 
 15 
 11 
 
 194 
 55 
 56 
 56 
 60 
 20 
 
 197 
 6 
 5 
 
 Gas Molecules, Mean Free Path of ... 
 Number of in cubic inch 
 
 Velocity of 
 
 Gases, Kinetic Theory of 
 
 Glasgow Electric Lighting Station... 
 
 Glow Discharge 
 
 Glow Lamp, Artistic uses of 
 
 Bernstein 
 
 Brightness of 
 
 Candle Form 
 
 Edison-Swan 
 
 Efficiency of 
 
 Elements of 
 
 Exhaustion of 
 
 Filament, Brightness of. 
 
 High Candle-Power ... 
 
 Life changes of 
 
 Life of 
 
 Luminous Efficiency of 
 
 Resistance of during life 
 
 Smashing Point 
 
 Useful Life of 
 
 Variation of Light of 
 
 with Pressure 
 
 Variation of Resistance 
 
 of during life 
 
 Blackening of 
 
 Characteristic Curves of 
 
 Graphite 
 
 Grease Spot Photometer 
 
 Guthrie, Prof 
 
 Half-way House, Tivoli, Rome 
 
 Head of Water 
 
 Heat and Electricity, Relation of 
 
 Heating Effect of Electric Current .. 
 
 of a Conductor by Current... 
 
 of Different Conductors by 
 
 Currents 
 
 Hefner- Alteneck Amyl-Acetate Lamp 
 High -Pressure and Low - Pressure 
 
 Distribution 
 
 High- Pressure Transmission 
 
 Horse-power, Definition of 
 
 Hydraulic Gradient 
 
 Illuminating Power, Unit of 
 
 Illumination, Unit of 
 
 Incandescent Lamp, Evolution of ... 
 
 Invention of ... 
 
 Incandescent Lighting, Artistic 
 
 Problem of. .. 
 
 Interior Views of Rooms Lit witli 
 
 Incandescent Lamps 100, 101 & 103 
 
 Inverted Arc Lamp 167 
 
 Iron Ring, Divided, with two Circuits 182 
 with Primary and Secon- 
 dary Circuit 180
 
 INDEX. 
 
 227 
 
 PAGE 
 
 Joule, Mr ................................. 18 
 
 Definition of the Unit called 
 
 the ............................. 20 
 
 Joule's Law ............................. 18 
 
 Kelvin Voltmeter ....................... 14 
 
 Kilowatt, Definition of the ............ 21 
 
 Lane-Fox, Mr. .......................... 52 
 
 Lightning Protector .................... 145 
 
 Light. Mono-Chromatic ............... 31 
 
 -- Reflection of .................... 31 
 
 -- Velocity of ....................... 28 
 
 -- Wave Lengths of Various 
 
 Colours ....................... 29 
 
 LinesofForce ......................... 179 
 
 Load Factor ........ ..................... 95 
 
 Luminosity ................................. 32 
 
 - - -- of Surface ................. 33 
 
 Luminous Efficiency .................... 26 
 
 of Glow Lamp ... 81 
 
 Luminous Radiation ................... 21 
 
 Magnetic Circuit ..... .................. 174 
 
 -- Circuits with Air Gaps... 184. 185 
 
 Effect of Current 
 Fieldof Conductor 
 Induction ...... 
 
 Magnetisation of Iron Circuits ....... 
 
 Magnetomotive Force .................. 
 
 Marked Volts ............................. 
 
 Mean Free Path of Molecules ......... 
 
 Mengarini Voltmeter .................... 
 
 Mercury Pump, Form of ...... ... ..... 
 
 Methven Gas Standard .................. 
 
 Micro- Glow Lamp ...................... 
 
 Molecular Electric Charge .......... 
 
 -- Physics of Glow Lamp ...... 
 
 Shadow in Glow Lamp ... 
 
 7 
 
 7 
 
 181 
 185 
 181 
 
 68 
 63 
 79 
 65 
 47 
 67 
 120 
 121 
 76 
 
 Moonlight, Intensity of .............. 39 
 
 Multicellular Voltmeter ............... 14 
 
 Multiple Filament Glow Lamp ...... 61 
 
 Oersted, Prof. H. C ................... 7 
 
 Oersted's Discovery ..................... 7 
 
 Ohm, Dr. G. S ............................ 17 
 
 - The ................................ 9 
 
 Ohm's Law ........................... ...... 17 
 
 Parallel Arc Lamps ................... 169 
 
 Parchmentised Thread ................ 55 
 
 Pentane Standard, Vernon Harcourt. 47 
 
 Photometer, Bunsen .................... 43 
 
 -- Grease Spot ... ........... 44 
 
 -- Ritchie's .................. 40 
 
 -- Rumford ................. 41 
 
 Photometric Comparisons .............. 35 
 
 Photometry, Basis of ..................... 57 
 
 --- of Glow Lamp ............ 41 
 
 --- of Various - Coloured 
 
 Lights ................... 42 
 
 PAGE 
 
 Platinum Standard 47 
 
 Positive and Negative Charge, Dif- 
 ference between 119 
 
 Potential, Electric 11 
 
 Pressure, Electric 11 
 
 Primary Circuit 177 
 
 Pump, Mercury 65 
 
 Purkinje's Phenomenon 36 
 
 Radiation at Different Temperatures 25 
 Luminous and Non- 
 Luminous 26 
 
 Reflecting Power of Various Surfaces 104 
 
 Resistance, Electric 16 
 
 Retrospect of Electric Lighting over 
 
 Fourteen Years 1 
 
 Ritchie's Wedge 40 
 
 Rome, Electric Lighting at 202 
 
 Electric Lighting Station 203 
 
 Rumford Photometer 41 
 
 Sawyer and Man 52 
 
 Secondary Circuit 177 
 
 Series Arc Lamps 168 
 
 Smashing Point 86 
 
 Spark Discharge 127 
 
 Sparking Distance 132 
 
 Spectra, Similar 30 
 
 Spectro-Photometer 48 
 
 Spectrum of a Source of Light 30 
 
 Spiral Electric Current, Field of 174 
 
 St. Pancras Electric Lighting Station 214 
 
 Standard, Candle 37 
 
 Stefan's Law 161 
 
 Summary of Facts concerning the 
 
 Edison Effect on Glow Lamps 121 
 
 Sun, Brightness of 157 
 
 Efficiency of 158 
 
 Temperature of.. ... 159 
 
 Sunlight, Intensity of 39 
 
 Surface Emissivity of Wires 23 
 
 Surfaces, Reflecting Powers of 104 
 
 Swan, J. W 52 
 
 Table of Wave-Lengths of Light 29 
 
 Temperatures, Fundamental 25 
 
 Scale of 25 
 
 Terms, Fundamental, Meaning of ... 6 
 Three-wire Electric Lighting Station, 
 
 Arrangement of Dynamos in ...211, 212 
 
 Three- wire System of Distribution... 211 
 
 Tivoli, Electric Lighting Station at. . . 204 
 
 Tivoli-Rome Transmission 208 
 
 Transformer House 209 
 
 Transformer, Construction of 192 
 
 Principle of 186 
 
 Sub-centres 198 
 
 Systems 199 
 
 Turbines at Tivoli 207
 
 228 
 
 INDEX. 
 
 United Kingdom, Electric Lighting 
 
 in 4 
 
 Units, Electric 9 
 
 Vacuum, Conductivity of 118 
 
 Tube 130 
 
 Variation of Candle - Power with 
 
 Voltage 71 
 
 Variation of Candle - Power with 
 
 with Wattage 71 
 
 Various Systems of Electric Distri- 
 bution 221 
 
 Vernon Harcourt Pentane Standard. 47 
 
 Violle Standard 47 
 
 Volt, The 16 
 
 Volta 16 
 
 Voltmeter 68 
 
 Holden 78 
 
 PAGE 
 
 Voltmeter Kelvin Multicellular 14 
 
 Mengarini 79 
 
 Self-recording 78 
 
 Self-recording (Mengarini) 80 
 
 Water Power 187 
 
 Water Pressure and Electric Pres- 
 sure 11 
 
 Watt, Definition of the 20 
 
 Watts per Candle-power 69 
 
 Wave Length 28 
 
 Wave Lengths of Light 29 
 
 Willans Engine " 196 
 
 Wimshurst Electric Machine 13 
 
 Wiring, Electric 93 
 
 Inspection of 94 
 
 Work 20 
 
 AVurt's Lightning Protector 145 
 
 COURr, FUEET STREET,
 
 ELECTRI 
 
 LAMPS 
 
 AND 
 
 ELECTS :s 
 LIGHTING. 
 
 11 -LEG PRICIAN