Thefunctionofthefloatingweightistoautomaticallykeepthestylusincloseengagementwiththerecord，thusinsuringaccuracyofreproduction。Theweightpressesthestylustoitswork，butbecauseofitsmassitcannotrespondtotheextremelyrapidvibrationsofthestylus。Theyarethereforecommunicatedtothediaphragm。
SomeofEdison\'smostremarkableinventionsarerevealedinanumberofinterestingpatentsrelatingtotheduplicationofphonographrecords。Itwouldbeobviouslyimpossible，fromacommercialstandpoint，toobtainamusicalrecordfromahigh-classartistandsellsuchanoriginaltothepublic，asitscostmightbefromonehundredtoseveralthousanddollars。Consequently，itisnecessarytoprovidesomewaybywhichduplicatesmaybemadecheaplyenoughtopermittheirpurchasebythepublicatareasonableprice。
Themakingofaperfectoriginalmusicalorotherrecordisamatterofnosmalldifficulty，asitrequiresspecialtechnicalknowledgeandskillgatheredfrommanyyearsofactualexperience；butintheexactcopying，orduplication，ofsucharecord，withitsmanymillionsofmicroscopicwavesandsub-waves，thedifficultiesareenormouslyincreased。Theduplicatesmustbemicroscopicallyidenticalwiththeoriginal，theymustbefreefromfalsevibrationsorotherdefects，althoughbothoriginalandduplicatesareofsucheasilydefacablematerialaswax；andtheprocessmustbecheapandcommercialnotascientificlaboratorypossibility。
Formakingduplicatesitwasobviouslynecessarytofirstsecureamoldcarryingtherecordinnegativeorreversedform。Fromthiscouldbemolded，orcast，positivecopieswhichwouldbeidenticalwiththeoriginal。Whiletheartofelectroplatingwouldnaturallysuggestitselfasthemeansofmakingsuchamold，anapparentlyinsurmountableobstacleappearedontheverythreshold。Wax，beinganon-
conductor，cannotbeelectroplatedunlessaconductingsurfacebefirstapplied。Thecoatingsordinarilyusedinelectro-
depositionwereentirelyoutofthequestiononaccountofcoarseness，thedeepestwavesoftherecordbeinglessthanone-thousandthofaninchindepth，andmanyofthemprobablytentoonehundredtimesasshallow。Edisonfinallydecidedtoapplyapreliminarymetalliccoatingofinfinitesimalthinness，andaccomplishedthisobjectbyaremarkableprocessknownasthevacuousdeposit。Withthisheap-
pliedtotheoriginalrecordafilmofgoldprobablynothickerthanonethree-hundred-thousandthofaninch，orseveralhundredtimeslessthanthedepthofanaveragewave。
Threehundredsuchlayersplacedoneontopoftheotherwouldmakeasheetnothickerthantissue-paper。
Theprocessconsistsinplacinginavacuumtwoleaves，orelectrodes，ofgold，andbetweenthemtheoriginalrecord。
Aconstantdischargeofelectricityofhightensionbetweentheelectrodesiseffectedbymeansofaninduction-coil。Themetalisvaporizedbythisdischarge，andiscarriedbyitdirectlytowardanddepositedupontheoriginalrecord，thusformingtheminutefilmofgoldabovementioned。Therecordisconstantlyrotateduntilitsentiresurfaceiscoated。
AsectionaldiagramoftheapparatusFig。6。willaidtoaclearerunderstandingofthisingeniousprocess。
Afterthegoldfilmisformedinthemannerdescribedabove，aheavybackingofbasermetaliselectroplateduponit，thusformingasubstantialmold，fromwhichtheoriginalrecordisextractedbybreakageorshrinkage。
Duplicaterecordsinanyquantitymaynowbemadefromthismoldbysurroundingitwithacold-waterjacketanddippingitinamoltenwax-likematerial。Thiscongealsontherecordsurfacejustasmeltedbutterwouldcollectonacoldknife，andwhenthemoldisremovedthesurpluswaxfallsout，leavingaheavydepositofthematerialwhichformstheduplicaterecord。NumerousingeniousinventionshavebeenmadebyEdisonprovidingforavarietyofrapidandeconomicalmethodsofduplication，includingmethodsofshrinkinganewlymadecopytofacilitateitsquickremovalfromthemold；methodsofreaming，offormingribsontheinterior，andformanyotherimportantandessentialdetails，whichlimitsofspacewillnotpermitofelaboration。Thosementionedabovearebutfairexamplesofthepersistentandeffectiveworkhehasdonetobringthephonographtoitspresentstateofperfection。
InperusingChapterXoftheforegoingnarrative，thereaderundoubtedlynotedEdison\'sclearapprehensionofthepracticalusesofthephonograph，asevidencedbyhispropheticutterancesinthearticlewrittenbyhimfortheNorthAmericanReviewinJune，1878。Inviewofthecrudityoftheinstrumentatthattime，itmustbeacknowl-
edgedthatEdison\'sforesight，asvindicatedbylatereventswasmostremarkable。Nolessremarkablewashisintenselypracticalgraspofmechanicalpossibilitiesoffuturetypesofthemachine，forwefindinoneofhisearlyEnglishpatentsNo。1644of1878thediskformofphonographwhich，sometentofifteenyearslater，wassupposedtobeanewdevelopmentintheart。ThisdiskformwasalsocoveredbyEdison\'sapplicationforaUnitedStatespatent，filedin1879。
ThisapplicationmetwithsomemerelyminortechnicalobjectionsinthePatentOffice，andseemstohavepassedintothe“abandoned“classforwantofprosecution，probablybecauseofbeingoverlookedinthetremendouspressurearisingfromhisdevelopmentofhiselectric-lightingsystem。
IX
THEINCANDESCENTLAMP
ALTHOUGHEdison\'scontributionstohumancomfortandprogressareextensiveinnumberandextraordinarilyvastandcomprehensiveinscopeandvariety，theuniversalverdictoftheworldpointstohisincandescentlampandsystemofdistributionofelectricalcurrentasthecentralandcrowningachievementsofhislifeuptothistime。Thisviewwouldseementirelyjustifiablewhenweconsiderthewonderfulchangesintheconditionsofmodernlifethathavebeenbroughtaboutbythewide-spreademploymentoftheseinventions，andthegiganticindustriesthathavegrownupandbeennourishedbytheirworld-wideapplication。Thathewasinthisinstanceatruepioneerandcreatorisevidentasweconsiderthesubject，fortheUnitedStatesPatentNo。223，898，issuedtoEdisononJanuary27，1880，foranincandescentlamp，wasofsuchfundamentalcharacterthatitopenedupanentirelynewandtremendouslyimportantart——theartofincandescentelectriclighting。Thisstatementcannotbesuccessfullycontroverted，forithasbeenabundantlyverifiedaftermanyyearsofcostlylitigation。
Iffurtherproofweredesired，itisonlynecessarytopointtothefactthat，afterthirtyyearsofmoststrenuousandpracticalapplicationintheartbythekeenestintellectsoftheworld，everyincandescentlampthathaseversincebeenmade，includingthoseofmoderndays，isstilldependentupontheemploymentoftheessentialsdisclosedintheabove-namedpatent——namely，afilamentofhighresistanceenclosedinasealedglassglobeexhaustedofair，withconductingwirespassingthroughtheglass。
Anincandescentlampissuchasimple-appearingarticle——
merelyafilamentsealedintoaglassglobe——thatitsintrinsicrelationtotheartofelectriclightingisfarfrombeingap-
parentatsight。TothelayminditwouldseemthatthismusthavebeenTHEobviousdevicetomakeinordertoobtainelectriclightbyincandescenceofcarbonorothermaterial。
ButthereaderhasalreadylearnedfromtheprecedingnarrativethatpriortoitsinventionbyEdisonsuchadevicewasNOTobvious，eventothemosthighlytrainedexpertsoftheworldatthatperiod；indeed，itwassofarfrombeingobviousthat，forsometimeafterhehadcompletedpracticallampsandwasactuallylightingthemuptwenty-fourhoursaday，suchadeviceandsucharesultweredeclaredbythesesameexpertstobeanutterimpossibility。ForashortwhiletheworldoutsideofMenloParkheldEdison\'sclaimsinderision。Hislampwaspronouncedafake，amyth，possiblyamomentarysuccessmagnifiedtothedignityofapermanentdevicebyanoverenthusiasticinventor。
Suchcriticism，however，didnotdisturbEdison。HeKNEWthathehadreachedthegoal。Longago，byacloseprocessofreasoning，hehadclearlyseenthattheonlyroadtoitwasthroughthepathhehadtravelled，andwhichwasnowembodiedinthephilosophyofhisincandescentlamp——
namely，afilament，orcarbon，ofhighresistanceandsmallradiatingsurface，sealedintoaglassglobeexhaustedofairtoahighdegreeofvacuum。Inoriginallycommittinghimselftothislineofinvestigationhewaswellawarethathewasgoinginadirectiondiametricallyoppositetothatfollowedbypreviousinvestigators。Theireffortshadbeenconfinedtolow-resistanceburnersoflargeradiatingsurfacefortheirlamps，butherealizedtheutterfutilityofsuchdevices。
Thetremendousproblemsofheatandtheprohibitivequantitiesofcopperthatwouldberequiredforconductorsforsuchlampswouldbeabsolutelyoutofthequestionincommercialpractice。
Hewasconvincedfromthefirstthatthetruesolutionoftheproblemlayinalampwhichshouldhaveasitsilluminatingbodyastripofmaterialwhichwouldoffersucharesistancetotheflowofelectriccurrentthatitcouldberaisedtoahightemperature——incandescence——andbeofsuchsmallcross-sectionthatitwouldradiatebutlittleheat。Atthesametimesuchalampmustrequirearelativelysmallamountofcurrent，inorderthatcomparativelysmallconductorscouldbeused，anditsburnermustbecapableofwithstand-
ingthenecessarilyhightemperatureswithoutdisintegration。
ItisinterestingtonotethattheseconceptionswereinEdison\'smindatanearlyperiodofhisinvestigations，whenthebestexpertopinionwasthatthesubdivisionoftheelectriccurrentwasanignisfatuus。Hencewequotethefollowingnoteshemade，November15，1878，inoneofthelaboratorynote-books：
“Agivenstraightwirehaving1ohmresistanceandcertainlengthisbroughttoagivendegreeoftemperaturebygivenbattery。Ifthesamewirebecoiledinsuchamannerthatbutone-quarterofitssurfaceradiates，itstemperaturewillbeincreasedfourtimeswiththesamebattery，or，one-
quarterofthisbatterywillbringittothetemperatureofstraightwire。Orthesamegivenbatterywillbringawirewhosetotalresistanceis4ohmstothesametemperatureasstraightwire。
“Thiswasactuallydeterminedbytrial。
“Theamountofheatlostbyabodyisinproportiontotheradiatingsurfaceofthatbody。Ifonesquareinchofplatinabeheatedto100degreesitwillfallto，say，zeroinonesecond，whereas，ifitwasat200degreesitwouldrequiretwoseconds。
“Hence，inthecaseofincandescentconductors，iftheradiatingsurfacebetwelveinchesandthetemperatureoneachinchbe100，or1200forall，ifitissocoiledorarrangedthatthereisbutone-quarter，orthreeinches，ofradiatingsurface，thenthetemperatureoneachinchwillbe400。Ifreducedtothree-quartersofaninchitwillhaveonthatthree-
quartersofaninch1600degreesFahr。，notwithstandingtheoriginaltotalamountwasbut1200，becausetheradiationhasbeenreducedtothree-quarters，or75units；hence，theeffectofthelesseningoftheradiationistoraisethetemperatureofeachremaininginchnotradiatingto125degrees。Iftheradiatingsurfaceshouldbereducedtothree-thirty-secondsofaninch，thetemperaturewouldreach6400degreesFahr。Tocarryoutthislawtothebestadvantageinregardtoplatina，etc。，thenwithagivenlengthofwiretoquadrupletheheatwemustlessentheradiatingsurfacetoone-quarter，andtodothisinaspiral，three-quartersmustbewithinthespiralandone-quarteroutsideforradiating；hence，asquarewireorothermeans，suchasaspiralwithinaspiral，mustbeused。TheseresultsaccountfortheenormoustemperatureoftheElectricArcwithonehorse-power；as，forinstance，ifonehorse-powerwillheattwelveinchesofwireto1000degreesFahr。，andthisisconcentratedtohaveone-quarteroftheradiatingsurface，itwouldreachatemperatureof4000degreesorsufficienttomeltit；
but，supposingitinfusible，thefurtherconcentrationtoone-
eighthitssurface，itwouldreachatemperatureof16，000degrees，andtoone-thirty-seconditssurface，whichwouldbeabouttheradiatingsurfaceoftheElectricArc，itwouldreach64，000degreesFahr。Ofcourse，whenLightisradiatedingreatquantitiesnotquitethesetemperatureswouldbereached。
“Anothercuriouslawisthis：Itwillrequireagreaterinitialbatterytobringanironwireofthesamesizeandresistancetoagiventemperaturethanitwillaplatinawireinproportiontotheirspecificheats，andinthecaseofCarbon，apieceofCarbonthreeincheslongandone-eighthdiameter，witharesistanceof1ohm，willrequireagreaterbatterypowertobringittoagiventemperaturethanacylinderofthinplatinafoilofthesamelength，diameter，andresistance，becausethespecificheatofCarbonismanytimesgreater；
besides，ifIamnotmistaken，theradiationofaroughenedbodyforheatisgreaterthanapolishedonelikeplatina。“
Proceedinglogicallyupontheselinesofthoughtandfollowingthemoutthroughmanyramifications，wehaveseenhowheatlengthmadeafilamentofcarbonofhighresistanceandsmallradiatingsurface，andthroughaconcurrentinvestigationofthephenomenaofhighvacuaandoccludedgaseswasabletoproduceatrueincandescentlamp。Notonlywasitalampasamerearticle——adevicetogivelight——
butitwasalsoanintegralpartofhisgreatandcompletesystemoflighting，toeverypartofwhichitboreafixedanddefiniteratio，andinrelationtowhichitwasthekeystonethatheldthestructurefirmlyinplace。
TheworkofEdisononincandescentlampsdidnotstopatthisfundamentalinvention，butextendedthroughmorethaneighteenyearsofamostintenseportionofhisbusylife。Duringthatperiodhewasgrantedonehundredandforty-nineotherpatentsonthelampanditsmanufacture。
Althoughverymanyoftheseinventionswereoftheutmostimportanceandvalue，wecannotattempttoofferadetailedexpositionoftheminthisnecessarilybriefarticle，butmustreferthereader，ifinterested，tothepatentsthemselves，afulllistbeinggivenattheendofthisAppendix。
Theoutlinesketchwillindicatetheprincipalpatentscoveringthebasicfeaturesofthelamp。
ThelitigationontheEdisonlamppatentswasoneofthemostdeterminedandstubbornlyfoughtcontestsinthehistoryofmodernjurisprudence。Vastinterestswereatstake。Allofthetechnical，expert，andprofessionalskillandknowledgethatmoneycouldprocureorexperiencedevisewereavailedofinthebitterfightsthatragedinthecourtsformanyyears。AndalthoughtheEdisoninterestshadspentfromfirsttolastnearly$2，000，000，andhadonlyaboutthreeyearsleftinthelifeofthefundamentalpatent，Edisonwasthoroughlysustainedastoprioritybythedecisionsinthevarioussuits。Weshallofferafewbriefextractsfromsomeofthesedecisions。
InasuitagainsttheUnitedStatesElectricLightingCompany，UnitedStatesCircuitCourtfortheSouthernDistrictofNewYork，July14，1891，JudgeWallacesaid，inhisopinion：
“Thefutilityofhopingtomaintainaburnerinvacuowithanypermanencyhaddiscouragedpriorinventors，andMr。Edisonisentitledtothecreditofobviatingthemechanicaldifficultieswhichdisheartenedthem……Hewasthefirsttomakeacarbonofmaterials，andbyaprocesswhichwasespeciallydesignedtoimparthighspecificresistancetoit；thefirsttomakeacarboninthespecialformforthespecialpurposeofimpartingtoithightotalresistance；
andthefirsttocombinesuchaburnerwiththenecessaryadjunctsoflampconstructiontopreventitsdisintegrationandgiveitsufficientlylonglife。Bydoingthesethingshemadealampwhichwaspracticallyoperativeandsuccessful，theembryoofthebestlampsnowincommercialuse，andbutforwhichthesubdivisionoftheelectriclightbyincandescencewouldstillbenothingbuttheignisfatuuswhichitwasproclaimedtobein1879bysomeofthereamedexpertswhoarenowwitnessestobelittlehisachievementandshowthatitdidnotrisetothedignityofaninvention……Itisimpossibletoresisttheconclusionthattheinventionoftheslenderthreadofcarbonasasubstitutefortheburnerspreviouslyemployedopenedthepathtothepracticalsubdivisionoftheelectriclight。“
AnappealwastakenintheabovesuittotheUnitedStatesCircuitCourtofAppeals，andonOctober4，1892，thedecreeofthelowercourtwasaffirmed。ThejudgesLacombeandShipman，inalongopinionreviewedthefactsandtheart，andsaid，interalia：“Edison\'sinventionwaspracticallymadewhenheascertainedthetheretoforeunknownfactthatcarbonwouldstandhightemperature，evenwhenveryat-
tenuated，ifoperatedinahighvacuum，withoutthephenomenonofdisintegration。Thisfactheutilizedbythemeanswhichhehasdescribed，alamphavingafilamentarycarbonburnerinanearlyperfectvacuum。“
InasuitagainsttheBostonIncandescentLampCompanyetal。，intheUnitedStatesCircuitCourtfortheDistrictofMassachusetts，decidedinfavorofEdisononJune11，1894，JudgeColt，inhisopinion，said，amongotherthings：
“Edisonmadeanimportantinvention；heproducedthefirstpracticalincandescentelectriclamp；thepatentisapioneerinthesenseofthepatentlaw；itmaybesaidthathisinventioncreatedtheartofincandescentelectriclighting。“
Opinionsofothercourts，similarintenortotheforegoing，mightbecited，butitwouldbemerelyinthenatureofreiteration。TheabovearesufficienttoillustratethedirectclearnessofjudicialdecisiononEdison\'spositionasthefounderoftheartofelectriclightingbyincandescence。
EDISON\'SDYNAMOWORK
ATthepresentwriting，when，afterthephenomenallyrapidelectricaldevelopmentofthirtyyears，wefindonthemarketagreatvarietyofmodernformsofefficientcurrentgeneratorsadvertisedunderthenamesofdifferentinventorsnone，however，bearingthenameofEdison，ayoungelectricalengineerofthepresentgenerationmightwellinquirewhetherthegreatinventorhadevercontributedanythingtotheartbeyondamereTYPEofmachineformerlymadeandbearinghisname，butnotnowmarketedexceptsecondhand。
Foradequateinformationhemightsearchinvainthebooksusuallyregardedasauthoritiesonthesubjectofdynamo-electricmachinery，forwithslightexceptionstherehasbeenasingularunanimityintheomissionofwriterstogiveEdisoncreditforhisgreatandbasiccontributionstoheavy-currenttechnics，althoughtheyhavebeenuniversallyacknowledgedbyscientificandpracticalmentohavelaidthefoundationfortheefficiencyof，andtobeembodiedinallmoderngeneratorsofcurrent。
ItmightnaturallybeexpectedthattheessentialfactsofEdison\'sworkwouldappearonthefaceofhisnumerouspatentsondynamo-electricmachinery，butsuchisnotnecessarilythecase，unlesstheyarecarefullystudiedinthelightofthestateoftheartasitexistedatthetime。Whilesomeofthesepatentsespeciallytheearlieronescoverspecificdevicesembodyingfundamentalprinciplesthatnotonlysurvivetothepresentday，butactuallylieatthefoundationoftheartasitnowexists，thereisnorevelationthereinofEdison\'sprecedingstudiesofmagnets，whichextendedovermanyyears，norofhislatersystematicinvestigationsanddeductions。
Dynamo-electricmachinesofaprimitivekindhadbeeninventedandwereinusetoaverylimitedextentforarclightingandelectroplatingforsomeyearspriortothesummerof1819，whenEdison，withanembryoniclightingSYSTEM
inmind，castaboutforatypeofmachinetechnicallyandcommerciallysuitableforthesuccessfulcarryingoutofhisplans。Hefoundabsolutelynone。Onthecontrary，allofthefewtypesthenobtainablewereuneconomical，indeedwasteful，inregardtoefficiency。Theart，ifindeedtherecanbesaidtohavebeenanartatthattime，wasinchaoticconfusion，andonlybecauseofEdison\'smanyyears\'studyofthemagnetwasheenabledtoconcludethatinsufficiencyinquantityofironinthemagnetsofsuchmachines，togetherwithpoorsurfacecontacts，renderedthecostofmagnetizationabnormallyhigh。Theheatingofsolidarmatures，theonlykindthenknown，andpoorinsulationinthecommutators，alsogaverisetoseriouslosses。Butperhapsthemostseriousdrawbacklayinthehigh-resistancearmature，baseduponthehighestscientificdictumofthetimethatinordertoobtainthemaximumamountofworkfromamachine，theinternalresistanceofthearmaturemustequaltheresistanceoftheexteriorcircuit，althoughtheapplicationofthisprincipleentailedtheuselessexpenditureofatleast50percent。oftheappliedenergy。
Itseemsalmostincrediblethatonlyalittleoverthirtyyearsagothesumofscientificknowledgeinregardtodynamo-
electricmachineswassomeagrethattheexpertsoftheperiodshouldsettleuponsuchadictumasthis，butsuchwasthefact，aswillpresentlyappear。Mechanicalgeneratorsofelectricitywerecomparativelynewatthattime；
theirtheoryandpracticewereveryimperfectlyunderstood；
indeed，itisquitewithintheboundsoftruthtosaythatthecorrectprincipleswerebefoggedbyreasonofthelackofpracticalknowledgeoftheiractualuse。Electriciansandscientistsoftheperiodhadbeenaccustomedformanyyearspasttolooktothechemicalbatteryasthesourcefromwhichtoobtainelectricalenergy；andinthepracticalapplicationofsuchenergytotelegraphyandkindreduses，muchthoughtandingenuityhadbeenexpendedinstudyingcombinationsofconnectingsuchcellssoastogetthebestresults。Inthetext-booksoftheperioditwasstatedasasettledprinciplethat，inordertoobtainthemaximumworkoutofasetofbatteries，theinternalresistancemustapproximatelyequaltheresistanceoftheexteriorcircuit。
Thisprincipleanditsapplicationinpracticewerequitecorrectasregardschemicalbatteries，butnotasregardsdynamomachines。Bothweregeneratorsofelectricalcurrent，butsodifferentinconstructionandoperation，thatrulesapplicabletothepracticaluseoftheonedidnotapplywithpropercommercialefficiencytotheother。Attheperiodunderconsideration，whichmaybesaidtohavebeenjustbeforedawnofthedayofelectriclight，thephilosophyofthedynamowasseenonlyinmysterious，hazyoutlines——
justemergingfromthedarknessofdepartingnight。Perhapsitisnotsurprising，then，thatthedynamowaslooselyregardedbyelectriciansasthepracticalequivalentofachemicalbattery；thatmanyofthecharacteristicsofperformanceofthechemicalcellwerealsoattributedtoit，andthatifthemaximumworkcouldbegottenoutofasetofbatterieswhentheinternalandexternalresistanceswereequalandthiswascommerciallythebestthingtodo，somustitbealsowithadynamo。
ItwasbynomiraclethatEdisonwasfarandawayaheadofhistimewhenheundertooktoimprovethedynamo。HewaspossessedofabsoluteKNOWLEDGEfarbeyondthatofhiscontemporaries。Thisheadacquiredbythehardestkindofworkandincessantexperimentwithmagnetsofallkindsduringseveralyearspreceding，particularlyinconnectionwithhisstudyofautomatictelegraphy。Hisknowledgeofmagnetswastremendous。Hehadstudiedandexperimentedwithelectromagnetsinenormousvariety，andknewtheirpeculiaritiesinchargeanddischarge，lag，self-
induction，staticeffects，condensereffects，andthevariousotherphenomenaconnectedtherewith。Hehadalsomadecollateralstudiesofiron，steel，andcopper，insulation，winding，etc。Hence，byreasonofthisextensiveworkandknowledge，Edisonwasnaturallyinapositiontorealizetheuttercommercialimpossibilityofthethenbestdynamomachineinexistence，whichhadanefficiencyofonlyabout40percent。，andwasconstructedonthe“cut-and-try“principle。
Hewasalsonaturallyinapositiontoassumethetaskhesetouttoaccomplish，ofundertakingtoplanand-buildanimprovedtypeofmachinethatshouldbecommercialinhav-
inganefficiencyofatleast90percent。Trulyaprodigiousundertakinginthosedarkdays，whenfromthestandpointofEdison\'slargeexperiencethemostpracticalandcorrectelectricaltreatisewascontainedintheEncyclopaediaBritannica，andinaGermanpublicationwhichMr。UptonhadbroughtwithhimafterhehadfinishedhisstudieswiththeillustriousHelmholtz。ItwasatthisperiodthatMr。UptoncommencedhisassociationwithEdison，bringingtothegreatworktheverylatestscientificviewsandtheassistanceofthehighermathematics，towhichhehaddevotedhisattentionforseveralyearspreviously。
AssomeaccountofEdison\'sinvestigationsinthisconnectionhasalreadybeengiveninChapterXIIofthenarrative，weshallnotenlargeuponthemhere，butquotefromAnHistoricalReview，byCharlesL。Clarke，LaboratoryAssistantatMenloPark，1880-81；ChiefEngineeroftheEdisonElectricLightCompany，1881-84：
“InJune，1879，waspublishedtheaccountoftheEdisondynamo-electricmachinethatsurvivedintheart。Thismachinewentintoextensivecommercialuse，andwasnotableforitsverymassiveandpowerfulfield-magnetsandarmatureofextremelylowresistanceascomparedwiththecombinedexternalresistanceofthesupply-mainsandlamps。
Bymeansofthelargemassesofironinthefield-magnets，andcloselyfittedjointsbetweentheseveralpartsthereof，themagneticresistancereluctanceoftheironpartsofthemagneticcircuitwasreducedtoaminimum，andtherequiredmagnetizationeffectedwiththemaximumeconomy。
AtthesametimeMr。Edisonannouncedthecommercialnecessityofhavingthearmatureofthedynamooflowresistance，ascomparedwiththeexternalresistance，inorderthatalargepercentageoftheelectricalenergydevelopedshouldbeutilizedinthelamps，andonlyasmallpercentagelostinthearmature，albeitthisprocedurereducedthetotalgeneratingcapacityofthemachine。Healsoproposedtomaketheresistanceofthesupply-mainssmall，ascomparedwiththecombinedresistanceofthelampsinmultiplearc，inordertostillfurtherincreasethepercentageofenergyutilizedinthelamps。Andlikewisetothisendthecombinedresistanceofthegeneratorarmaturesinmultiplearcwaskeptrelativelysmallbyadjustingthenumberofgeneratorsoperatinginmultipleatanytimetothenumberoflampstheninuse。Thefield-magnetcircuitsofthedynamoswereconnectedinmultiplewithaseparateenergizingsource；
andthefield-current；andstrengthoffield，wereregulatedtomaintaintherequiredamountofelectromotiveforceuponthesupply-mainsunderallconditionsofloadfromthemaximumtotheminimumnumberoflampsinuse，andtokeeptheelectromotiveforceofallmachinesalike。“
AmongtheearliestofEdison\'sdynamoexperimentswerethoserelatingtothecoreofthearmature。Herealizedatoncethattheheatgeneratedinasolidcorewasaprolificsourceofloss。Heexperimentedwithbundlesofironwiresvariouslyinsulated，alsowithsheet-ironrolledcylindricallyandcoveredwithironwirewoundconcentrically。Theseexperimentsandmanyothersweretriedinagreatvarietyofways，until，astheresultofallthiswork，Edisonarrivedattheprinciplewhichhasremainedinthearttothisday。
Hesplituptheironcoreofthearmatureintothinlaminations，separatedbypaper，thuspracticallysuppressingFoucaultcurrentsthereinandresultingheatingeffect。Itwasinhismachinealsothatmicawasusedforthefirsttimeasaninsulatingmediuminacommutator。[27]
[27]Thecommercialmanufactureofbuilt-upsheetsofmicaforelectricalpurposeswasfirstestablishedattheEdisonMachineWorks，GoerckStreet，NewYork，in1881。
Elementaryastheseprincipleswillappeartothemodernstudentorengineer，theyweredenouncedasnothingshortofabsurdityatthetimeoftheirpromulgation——especiallysowithregardtoEdison\'sproposaltoupsetthethensettleddictumthatthearmatureresistanceshouldbeequaltotheexternalresistance。Hispropositionwasderidedinthetechnicalpressoftheperiod，bothathomeandabroad。Aspublicopinioncanbebestillustratedbyactualquotation，weshallpresentacharacteristicinstance。
IntheScientificAmericanofOctober18，1879，thereappearedanillustratedarticlebyMr。UptononEdison\'sdynamomachine，inwhichEdison\'sviewsandclaimsweresetforth。Asubsequentissuecontainedasomewhatacri-
moniousletterofcriticismbyawell-knownmakerofdynamomachines。Attheriskofbeinglengthy，wemustquotenearlyallthisletter：“Icanscarcelyconceiveitaspossiblethatthearticleontheabovesubject“Edison\'sElectricGenerator“inlastweek\'sScientificAmericancouldhavebeenwrittenfromstatementsderivedfromMr。Edisonhimself，inasmuchassomanyoftheadvantagesclaimedforthemachinedescribedandstatementsoftheresultsobtainedaresomanifestlyabsurdastoindicateonthepartofbothwriterandprompterapositivewantofknowledgeoftheelectriccircuitandtheprinciplesgoverningtheconstructionandoperationofelectricmachines。
“Itisnotmyintentiontocriticisethedesignorconstructionofthemachinenotbecausetheyarenotopentocriticism，asIamnowandhavebeenformanyyearsengagedinthemanufactureofelectricmachines，butrathertocallattentiontotheimpossibilityofobtainingthedescribedresultswithoutdestroyingthedoctrineoftheconservationandcorrelationofforces……
“Itisstatedthat`theinternalresistanceofthearmature\'
ofthismachine`isonly1/2ohm。\'Onthisfactandthedisproportionbetweenthisresistanceandthatoftheexternalcircuit，thetheoryoftheallegedefficiencyofthemachineisstatedtobebased，forweareinformedthat，`whilethisgeneratoringeneralprincipleisthesameasinthebestwell-knownforms，stillthereisanall-importantdifference，whichisthatitwillconvertanddeliverforusefulworknearlydoublethenumberoffoot-poundsthatanyothermachinewillunderlikeconditions。\'“ThewriterofthiscriticalletterthenproceedstoquoteMr。Upton\'sstatementofthisefficiency：“`Nowtheenergyconvertedisdistributedoverthewholeresistance，henceiftheresistanceofthemachineberepresentedby1andtheexteriorcircuitby9，thenofthetotalenergyconvertednine-tenthswillbeuseful，asitisoutsideofthemachine，andone-tenthislostintheresistanceofthemachine。\'“
Afterthisthecriticgoesontosay：
“HowanyoneacquaintedwiththelawsoftheelectriccircuitcanmakesuchstatementsiswhatIcannotunderstand。
Thestatementlastquotedismathematicallyabsurd。
ItimplieseitherthatthemachineisCAPABLEOFINCREASING
ITSOWNELECTROMOTIVEFORCENINETIMESWITHOUTANINCREASED
EXPENDITUREOFPOWER，orthatexternalresistanceisNOTresistancetothecurrentinducedintheEdisonmachine。
“DoesMr。Edison，oranyoneforhim，meantosaythatr/nenableshimtoobtainnE，andthatCISNOT=E/r/nR？
IfsoMr。EdisonhasdiscoveredsomethingMOREthanperpetualmotion，andMr。Keelyhadbetterretirefromthefield。
“FurtheronthewriterMr。Uptongivesusanotherexampleofthismodeofreasoningwhen，emboldenedandsatisfiedwiththeabsurdtheoryaboveexposed，heendeavorstoprovethecauseoftheinefficiencyoftheSiemensandothermachines。Couldn\'tthewriterofthearticleseethatsinceC=E/rRthatbyR/norbymakingR=r，themachinewould，accordingtohistheory，havereturnedmoreusefulcurrenttothecircuitthancouldbeduetothepoweremployedandintheratioindicated，sothattherewouldactuallybeacreationofforce！……
“InconclusionallowmetosaythatifMrEdisonthinkshehasaccomplishedsomuchbytheREDUCTIONOFTHEINTERNAL
RESISTANCEofhismachine，thathehasmuchmoretodointhisdirectionbeforehismachinewillequalINTHISRESPECT
othersalreadyinthemarket。“
AnotherparticipantinthecontroversyonEdison\'sgeneratorwasascientificgentleman，whoinalongarticlepublishedintheScientificAmerican，inNovember，1879，gravelyundertooktoinstructEdisonintheABCofelectricalprinciples，andthenproceededtodemonstratemathematicallytheIMPOSSIBILITYofdoingWHATEDISONHADACTUALLYDONE。Thiscriticconcludeswithagentlerebuketotheinventorforill-
timedjesting，andasuggestiontofurnishAUTHENTICinformation！
Inthelightoffacts，astheywereandare，thisarticleissofullofhumorthatweshallindulgeinafewquotationsItcommencesinABCfashionasfollows：“Electricmachinesconvertmechanicalintoelectricalenergy……Theratioofyieldtoconsumptionistheexpressionoftheefficiencyofthemachine……Howmanyfoot-poundsofelec-
tricitycanbegotoutof100foot-poundsofmechanicalenergy？Certainlynotmorethan100：certainlyless……
Thefactsandlawsofphysics，withtheassistanceofmathematicallogic，neverfailtofurnishpreciousanswerstosuchquestions。“
Thewould-becriticthengoesontotabulatetestsofcertainotherdynamomachinesbyacommitteeoftheFranklinInstitutein1879，theresultsofwhichshowedthatthesemachinesreturnedabout50percent。oftheappliedmechanicalenergy，ingenuouslyremarking：“Whyisitthatwhenwehaveproducedtheelectricity，halfofitmustslipaway？Somepersonswillbecontentiftheyaretoldsimplythatitisawaywhichelectricityhasofbehaving。ButthereisasatisfactoryrationalexplanationwhichIbelievecanbemadeplaintopersonsofordinaryintelligence。Itoughttobeknowntoallthosewhoaremakingorusingmachines。
Iamgrievedtoobservethatmanypersonswhotalkandwritegliblyaboutelectricitydonotunderstandit；someevenignoreordenythefacttobeexplained。“
HerefollowsHISexplanation，afterwhichhegoesontosay：“Atthispointplausiblycomesinasuggestionthattheinternalpartofthecircuitbemadeverysmallandtheexternalpartverylarge。Whynotsaymaketheinternalpart1andtheexternal9，thussavingnine-tenthsandlosingonlyone-tenth？Unfortunately，thesuggestionisnotpractical；
afallacyisconcealedinit。“
Hethengoesontoprovehiscasemathematically，tohisownsatisfaction，followingitsadlybycondolingwithandawarningtoEdison：“ButaboutEdison\'selectricgenerator！……Noonecapableofmakingtheimprovementsinthetelegraphandtelephone，forwhichweareindebtedtoMr。
Edison，couldbeotherthananaccomplishedelectrician。
Hisreputationasascientist，indeed，issmirchedbythenewspaperexaggerations，andnodoubthewillbemorecarefulinfuture。Butthereisadangernearerhome，indeed，amonghisownfriendsandinhisveryhousehold。
“……Thewriterofpage242“theoriginalarticle“isprobablyafriendofMr。Edison，butpossibly，alas！awickedpartner。Whydoeshesaysuchthingsasthese？`Mr。Edisonclaimsthatherealizes90percent。ofthepowerappliedtothismachineinexternalwork。\'……Perhapsthewriterisahumorist，andhadinhismindColonelSellers，etc。，whichhecouldnotkeepoutofaseriousdiscussion；butsuchjestsarenotgood。
“Mr。Edisonhasbuiltaveryinterestingmachine，andhehastheopportunityofmakingavaluablecontributiontotheelectricalartsbyfurnishingauthenticaccountsofitscapabilities。“
Theforegoingextractsareunavoidablylengthy，but，viewedinthelightoffacts，servetoillustratemostclearlythatEdison\'sconceptionsandworkwerefarandawayaheadofthecomprehensionofhiscontemporariesintheart，andthathisachievementsinthelineofefficientdynamodesignandconstructionwereindeedtrulyfundamentalandrevolutionaryincharacter。Muchmoreofsimilarnaturetotheabovecouldbequotedfromotherarticlespublishedelsewhere，buttheforegoingwillserveasinstancesgenerallyrepresentingall。InthecontroversywhichappearedinthecolumnsoftheScientificAmerican，Mr。Upton，Edison\'smathematician，tookupthequestiononhisside，andansweredthecriticsbyfurtherelucidationsoftheprinciplesonwhichEdisonhadfoundedsuchremarkableandradicalimprovementsintheart。ThetypeofEdison\'sfirstdynamo-
electricmachine，thedescriptionofwhichgaverisetotheabovecontroversy，isshowninFig。1。
AnyaccountofEdison\'sworkonthedynamowouldbeincompletediditomittorelatehisconceptionandconstructionofthegreatdirect-connectedsteam-drivengeneratorthatwastheprototypeofthecolossalunitswhichareusedthroughouttheworldto-day。
InthedemonstratingplantinstalledandoperatedbyhimatMenloParkin1880tendynamosofeighthorse-powereachweredrivenbyaslow-speedenginethroughacomplicatedsystemofcounter-shafting，and，toquotefromMr。
Clarke\'sHistoricalReview，“itwasfoundthataconsiderablepercentageofthepoweroftheenginewasnecessarilywastedinfrictionbythismethodofdriving，andtopreventthiswasteandthusincreasetheeconomyofhissystem，Mr。Edisonconceivedtheideaofsubstitutingasinglelargedynamofortheseveralsmalldynamos，anddirectlycouplingitwiththedrivingengine，andatthesametimepreservetherequisitehigharmaturespeedbyusinganengineofthehigh-
speedtype。Healsoexpectedtorealizestillfurthergainsineconomyfromtheuseofalargedynamoinplaceofseveralsmallmachinesbyamorethancorrespondinglylowerarmatureresistance，lessenergyformagnetizingthefield，andforotherminorreasons。Tothesameend，heintendedtosupplysteamtotheengineunderamuchhigherboilerpressurethanwascustomaryinstationary-enginedrivingatthattime。“
Theconstructionofthefirstoneoftheselargemachineswascommencedlateintheyear1880。Earlyin1881itwascompletedandtested，butsomeradicaldefectsinarmatureconstructionweredeveloped，anditwasalsodemonstratedthatarateofenginespeedtoohighforcontinuouslysafeandeconomicaloperationhadbeenchosen。Themachinewaslaidaside。Anaccurateillustrationofthismachine，asitstoodintheengine-roomatMenloPark，isgiveninVanNostrand\'sEngineeringMagazine，Vol。XXV，oppositepage439，andabriefdescriptionisgivenonpage450。
Withtheexperiencethusgained，Edisonbegan，inthespringof1881，attheEdisonMachineWorks，GoerckStreet，NewYorkCity，theconstructionofthefirstsuccessfulmachineofthistype。Thiswasthegreatmachineknownas“JumboNo。1，“whichisreferredtointhenarrativeashavingbeenexhibitedattheParisInternationalElectricalExposition，whereitwasregardedasthewonderoftheelectricalworld。Anintimationofsomeofthetremendousdifficultiesencounteredintheconstructionofthismachinehasalreadybeengiveninprecedingpages，henceweshallnotnowenlargeonthesubject，excepttonoteinpassingthattheterriblydestructiveeffectsofthesparkofself-inductionandthearcingfollowingitwerefirstmanifestedinthispowerfulmachine，butwerefinallyovercomebyEdisonafterastrenuousapplicationofhispowerstothesolutionoftheproblem。
Itmaybeofinterest，however，tomentionsomeofitsdimensionsandelectricalcharacteristics，quotingagainfromMr。Clarke：“Thefield-magnethadeightsolidcylindricalcores，8inchesindiameterand57incheslong，uponeachofwhichwaswoundanexciting-coilof3。2ohmsresistance，consistingof2184turnsofNo。10B。W。G。insulatedcopperwire，disposedinsixlayers。Thelaminatedironcoreofthearmature，formedofthinirondisks，was333/4incheslong，andhadaninternaldiameterof121/2inches，andanexternaldiameterof267/16inches。Itwasmountedona6-inchshaft。
Thefield-poleswere333/4incheslong，and271/2inchesinsidediameterThearmaturewindingconsistedof146copperbarsonthefaceofthecore，connectedintoaclosed-coilwindingbymeansof73copperdisksateachendofthecore。
Thecross-sectionalareaofeachbarwas0。2squareinchtheiraveragelengthwas42。7inches，andthecopperend-
diskswere0。065inchthick。Thecommutatorhad73sec-
tions。Thearmatureresistancewas0。0092ohm，[28]ofwhich0。0055ohmwasinthearmaturebarsand0。0037ohmintheend-disks。“AnillustrationofthenextlatesttypeofthismachineispresentedinFig。2。
[28]HadEdisoninUpton\'sScientificAmericanarticlein1879proposedsuchanexceedinglylowarmatureresistanceforthisimmensegeneratoralthoughitsratiowasproportionatetotheoriginalmachine，hiscriticsmightprobablyhavebeensufficientlyindignantastobeunabletoexpressthemselvescoherently。
ThestudentmayfinditinterestingtolookupEdison\'sUnitedStatesPatentsNos。242，898，263，133，263，146，and246，647，bearingupontheconstructionofthe“Jumbo“；
alsoillustratedarticlesinthetechnicaljournalsofthetime，amongwhichmaybementioned：ScientificAmerican，Vol。
XLV，page367；Engineering，London，Vol。XXXII，pages409and419，TheTelegraphicJournalandElectricalReview，London，Vol。IX，pages431-433，436-446；LaNature，Paris，9thyear，PartII，pages408-409；ZeitschriftfurAngewandteElektricitaatslehre，MunichandLeipsic，Vol。IV，pages4-14；
andDredge\'sElectricIllumination，1882，Vol。I，page261。
Thefurtherdevelopmentofthesegreatmachineslateron，andtheirextensivepracticaluse，arewellknownandneednofurthercomment，exceptinpassingitmaybenotedthatsubsequentmachineshadeachacapacityof1200lampsof16candle-power，andthatthearmatureresistancewasstillfurtherreducedto0。0039ohm。
Edison\'sclearinsightintothefuture，asillustratedbyhispersistentadvocacyoflargedirect-connectedgeneratingunits，isabundantlyvindicatedbypresent-daypractice。
HisJumbomachines，of175horse-power，soenormousfortheirtime，haveservedasprototypes，andhavebeensucceededbygeneratorswhichhaveconstantlygrowninsizeandcapacityuntilatthistime1910itisnotuncommontoemploysuchgeneratingunitsofacapacityof14，000kilowatts，orabout18，666horse-power。
WehavenotenteredintospecificdescriptionsofthemanyotherformsofdynamomachinesinventedbyEdison，suchasthemultipolar，thediskdynamo，andthearmaturewithtwowindings，forsub-stationdistribution；indeed，itisnotpossiblewithinourlimitedspacetopresentevenabriefdigestofEdison\'sgreatandcomprehensiveworkonthedynamo-electricmachine，asembodiedinhisextensiveex-
perimentsandinoveronehundredpatentsgrantedtohim。
Wehave，therefore，confinedourselvestotheindicationofafewsalientandbasicfeatures，leavingittotheinterestedstudenttoexaminethepatentsandthetechnicalliteratureofthelongperiodoftimeoverwhichEdison\'slaborswereextended。
Althoughhehasnotgivenanyattentiontothesubjectofgeneratorsformanyyears，aninterestinginstanceofhisincisivemethodofovercomingminordifficultiesoccurredwhilethepresentvolumeswereunderpreparation1909。
Carbonforcommutatorbrusheshasbeensupersededbygraphiteinsomecases，thelattermaterialbeingfoundmuchmoreadvantageous，electrically。Troubledeveloped，however，forthereasonthatwhilecarbonwashardandwouldwearawaythemicainsulationsimultaneouslywiththecopper，graphite，beingsofter，wouldwearawayonlythecopper，leavingridgesofmicaandthuscausingsparkingthroughunequalcontact。AtthispointEdisonwasaskedtodiagnosethetroubleandprovidearemedy。Hesuggestedthecuttingoutofthemicapiecesalmosttothebottom，leavingthecommutatorbarsseparatedbyair-spaces。
Thisschemewasobjectedtoonthegroundthatparticlesofgraphitewouldfilltheseair-spacesandcauseashort-
circuit。Hisanswerwasthattheair-spacesconstitutedthevalueofhisplan，astheparticlesofgraphitefallingintothemwouldbethrownoutbytheactionofcentrifugalforceasthecommutatorrevolved。Andthusitoccurredasamatteroffact，andthetroublewasremedied。Thisideawassubsequentlyadoptedbyagreatmanufacturerofgenerators。
XI
THEEDISONFEEDERSYSTEM
TOquotefromthepreambleofthespecificationsofUnitedStatesPatentNo。264，642，issuedtoThomasA。EdisonSeptember19，1882：“Thisinventionrelatestoamethodofequalizingthetensionor`pressure\'ofthecurrentthroughanentiresystemofelectriclightingorothertranslationofelectricforce，preventingwhatisordinarilyknownasa`drop\'inthoseportionsofthesystemthemoreremotefromthecentralstation……“
TheproblemwhichwassolvedbytheEdisonfeedersystemwasthatrelatingtotheequaldistributionofcurrentonalargescaleoverextendedareas，inorderthataconstantanduniformelectricalpressurecouldbemaintainedineverypartofthedistributionareawithoutprohibitoryexpenditureforcopperformainsandconductors。
Thisproblemhadatwofoldaspect，althougheachsidewasinseparablyboundupintheother。Ontheonehanditwasobviouslynecessaryinalightingsystemthateachlampshouldbeofstandardcandle-power，andcapableofinterchangeableuseonanypartofthesystem，givingthesamedegreeofilluminationateverypoint，whetherneartoorremotefromthesourceofelectricalenergy。Ontheotherhand，thismustbeaccomplishedbymeansofasystemofconductorssodevisedandarrangedthatwhiletheywouldinsuretheequalpressurethusdemanded，theirmassandconsequentcostwouldnotexceedtheboundsofpracticalandcommerciallyeconomicalinvestment。
Thegreatimportanceofthisinventioncanbebetterunderstoodandappreciatedbyabriefglanceatthestateoftheartin1878-79，whenEdisonwasconductingthefinalseriesofinvestigationswhichculminatedinhisinventionoftheincandescentlampandSYSTEMoflighting。Atthistime，andforsomeyearspreviously，thescientificworldhadbeenworkingonthe“subdivisionoftheelectriclight，“asitwasthentermed。Someleadingauthoritiespronounceditabsolutelyimpossibleofachievementonanyextendedscale，whileaveryfewothers，ofmoreoptimisticmind，couldseenogleamoflightthroughthedarkness，butconfidentlyhopedforfuturedevelopmentsbysuchworkersasEdison。
Theearlierinvestigators，includingthoseuptotheperiodabovenamed，thoughtoftheproblemasinvolvingthesubdivisionofaFIXEDUNITofcurrent，which，beingsufficienttocauseilluminationbyonelargelamp，mightbedividedintoanumberofsmallunitswhoseaggregatelightwouldequalthecandle-powerofthislargelamp。Itwasfound，however，intheirexperimentsthatthecontraryeffectwasproduced，forwitheveryadditionallampintroducedinthecircuitthetotalcandle-powerdecreasedinsteadofincreasing。
IftheywereplacedinseriesthelightvariedinverselyastheSQUAREofthenumberoflampsincircuit；whileiftheywereinsertedinmultiplearc，thelightdiminishedastheCUBEofthenumberincircuit。[29]TheideaofmaintainingaconstantpotentialandofPROPORTIONINGTHECURRENTtothenumberoflampsincircuitdidnotoccurtomostoftheseearlyinvestigatorsasafeasiblemethodofovercomingthesupposeddifficulty。
[29]M。Fontaine，inhisbookonElectricLighting1877，showedthatwiththecurrentofabatterycomposedofsixteenelements，onelampgaveanilluminationequalto54burners；whereastwosimilarlamps，ifintroducedinparallelormultiplearc，gavethelightofonly61/2burnersinall；
threelampsofonly2burnersinall；fourlampsofonly3/4ofoneburner，andfivelampsof1/4ofaburner。
Itwouldalsoseemthatalthoughthegeneralmethodofplacingexperimentallampsinmultiplearcwasknownatthisperiod，theideaof“drop“ofelectricalpressurewasimperfectlyunderstood，if，indeed，realizedatall，asamostimportantitemtobeconsideredinattemptingthesolutionoftheproblem。Asamatteroffact，theinvestigatorsprecedingEdisondonotseemtohaveconceivedtheideaofa“system“atall；henceitisnotsurprisingtofindthemfarastrayfromthecorrecttheoryofsubdivisionoftheelectriccurrent。Itmayeasilybebelievedthattheterm“subdivision“
wasamisleadingonetotheseearlyexperimenters。
ForaveryshorttimeEdisonalsowasthusmisled，butassoonasheperceivedthattheproblemwasoneinvolvingtheMULTIPLICATIONOFCURRENTUNITS，hisbroadconceptionofa“system“wasborn。
Generallyspeaking，allconductorsofelectricityoffermoreorlessresistancetothepassageofcurrentthroughthemandinthetechnicalterminologyofelectricalsciencetheword“drop“whenusedinreferencetoasystemofdistribution
isusedtoindicateafallorlossofinitialelectricalpressurearisingfromtheresistanceofferedbythecopperconductorsleadingfromthesourceofenergytothelamps。
Theresultofthisresistanceistoconvertortranslateaportionoftheelectricalenergyintoanotherform——namely，heat，whichintheconductorsisUSELESSandwastefulandtosomeextentinevitableinpractice，butistobeavoidedandremediedasfaraspossible。
Itistruethatinanelectric-lightingsystemthereisalsoafallorlossofelectricalpressurewhichoccursinovercomingthemuchgreaterresistanceofthefilamentinanincandescentlamp。Inthiscasethereisalsoatranslationoftheenergy，buthereitaccomplishesaUSEFULpurpose，astheenergyisconvertedintotheformoflightthroughtheincandescenceofthefilament。Suchaconversioniscalled“work“asdistinguishedfrom“drop，“althoughafallofinitialelectricalpressureisinvolvedineachcase。
Thepercentageof“drop“variesaccordingtothequantityofcopperusedinconductors，bothastocross-sectionandlength。Thesmallerthecross-sectionalarea，thegreaterthepercentageofdrop。Thepracticaleffectofthisdropwouldbealossofilluminationinthelampsaswegofartherawayfromthesourceofenergy。ThismaybeillustratedbyasimplediagraminwhichGisagenerator，orsourceofenergy，furnishingcurrentatapotentialorelectricalpressureof110volts；1and2aremainconductors，fromwhich110-voltlamps，L，aretakeninderivedcircuits。ItwillbeunderstoodthatthecircuitsrepresentedinFig。1aretheoreticallysupposedtoextendoveralargearea。Themainconductorsaresufficientlylargeincross-sectiontoofferbutlittleresistanceinthosepartswhicharecomparativelynearthegenerator，butasthecurrenttraversestheirextendedlengththereisagradualincreaseofresistancetoovercome，andconsequentlythedropincreases，asshownbythefigures。
Theresultofthedropinsuchacasewouldbethatwhilethetwolamps，orgroups，nearestthegeneratorwouldbeburningattheirproperdegreeofillumination，thosebeyondwouldgivelowerandlowercandle-power，successively，untilthelastlamp，orgroup，wouldbegivingonlyabouttwo-thirdsthelightofthefirsttwo。Inotherwords，averyslightdropinvoltagemeansadisproportionatelygreatlossinillumination。
Hence，byusingaprimitivesystemofdistribution，suchasthatshownbyFig。1，theinitialvoltagewouldhavetobesohigh，inordertoobtainthepropercandle-powerattheendofthecircuit，thatthelampsnearestthegeneratorwouldbedangerouslyoverheated。Itmightbesuggestedasasolutionofthisproblemthatlampsofdifferentvoltagescouldbeused。But，asweareconsideringsystemsofextendeddistributionemployingvastnumbersoflampsasinNewYorkCity，wheremillionsareinuse，itwillbeseenthatsuchamethodwouldleadtoinextricableconfusion，andthereforebeabsolutelyoutofthequestion。Inasmuchasthepercentageofdropdecreasesinproportiontotheincreasedcross-sectionoftheconductors，theonlyfeasibleplanwouldseemtobetoincreasetheirsizetosuchdimensionsastoeliminatethedropaltogether，beginningwithconductorsoflargecross-sectionandtaperingoffasnecessary。
Thiswould，indeed，obviatethetrouble，but，ontheotherhand，wouldgiverisetoamuchmoreseriousdifficulty——
namely，theenormousoutlayforcopper；anoutlaysogreatastobeabsolutelyprohibitoryinconsideringtheelectriclightingoflargedistricts，asnowpracticed。
Anotherdiagramwillprobablymakethismoreclear。
Thereferencefiguresareusedasbefore，exceptthatthehorizontallinesextendingfromsquaremarkedGrepresentthemainconductors。Aseachlamprequiresandtakesitsownproportionofthetotalcurrentgenerated，itisobviousthatthesizeoftheconductorstocarrythecurrentforanumberoflampsmustbeaslargeasthesumofALLtheseparateconductorswhichwouldberequiredtocarrythenecessaryamountofcurrenttoeachlampseparately。
Hence，inaprimitivemultiple-arcsystem，itwasfoundthatthesystemmusthaveconductorsofasizeequaltotheaggregateoftheindividualconductorsnecessaryforeverylamp。Suchconductorsmighteitherbeseparate，asshownaboveFig。2，orbebunchedtogether，ormadeintoasolidtaperingconductor，asshowninthefollowingfigure：
Theenormousmassofcopperneededinsuchasystemcanbebetterappreciatedbyaconcreteexample。SomeyearsagoMr。W。J。Jenksmadeacomparativecalculationwhichshowedthatsuchasystemofconductorsknownasthe“Tree“system，tosupply8640lampsinaterritoryextendingoversosmallanareaasninecityblocks，wouldrequire803，250poundsofcopper，whichatthethenpriceof25centsperpoundwouldcost$200，812。50！
Such，inbrief，wasthestateoftheart，generallyspeaking，attheperiodabovenamed1878-79。Asearlyintheartasthelatterendoftheyear1878，Edisonhaddevelopedhisideassufficientlytodeterminethattheproblemofelectricilluminationbysmallunitscouldbesolvedbyusingincandescentlampsofhighresistanceandsmallradiatingsurface，andbydistributingcurrentsofconstantpotentialtheretoinmultiplearcbymeansofaramificationofconductors，startingfromacentralsourceandbranchingtherefromineverydirection。ThiswasanequivalentofthemethodillustratedinFig。3，knownasthe“Tree“system，andwas，infact，thesystemusedbyEdisoninthefirstandfamousexhibitionofhiselectriclightatMenloParkaroundtheChristmasperiodof1879。Herealized，however，thattheenormousinvestmentforcopperwouldmilitateagainstthecommercialadoptionofelectriclightingonanextendedscale。Hisnextinventivestepcoveredthedivisionofalargecitydistrictintoanumberofsmallsub-stationssupplyingcurrentthroughaninterconnectednetworkofconductors，thusreducingexpenditureforcoppertosomeextent，becauseeachdistributionunitwassmallandlimitedthedrop。
Hisnextdevelopmentwastheradicaladvancementofthestateofthearttothefeedersystem，coveredbythepatentnowunderdiscussion。Thisinventionsweptawaythetreeandothersystems，andatoneboundbroughtintobeingthepossibilityofeffectivelydistributinglargecurrentsoverextendedareaswithacommerciallyreasonableinvestmentforcopper。
Thefundamentalprinciplesofthisinventionwere，first，toseverentirelyanydirectconnectionofthemainconductorswiththesourceofenergy；and，second，tofeedcurrentataconstantpotentialtocentralpointsinsuchmainconductorsbymeansofotherconductors，called“feeders，“
whichweretobeconnecteddirectlywiththesourceofenergyatthecentralstation。Thisideawillbemademoreclearbyreferencetothefollowingsimplediagram，inwhichthesamelettersareusedasbefore，withadditions：
Infurtherelucidationofthediagram，itmaybeconsideredthatthemainsarelaidinthestreetalongacityblock，moreorlessdistantfromthestation，whilethefeedersareconnectedatoneendwiththesourceofenergyatthestation，theirotherextremitiesbeingconnectedtothemainsatcentralpointsofdistribution。Ofcourse，thissystemwasintendedtobeappliedineverypartofadistricttobesuppliedwithcurrent，separatesetsoffeedersrunningoutfromthestationtothevariouscentres。Thedistributionmainsweretobeofsufficientlylargesizethatbetweentheirmostextremepointsthelosswouldnotbemorethan3volts。
Suchaslightdifferencewouldnotmakeanappreciablevariationinthecandle-powerofthelamps。
Bytheapplicationoftheseprinciples，theinevitablebutuselessloss，or“drop，“requiredbyeconomymightbeincurred，butwasLOCALIZEDINTHEFEEDERS，whereitwouldnotaffecttheuniformityofilluminationofthelampsinanyofthecircuits，whetherneartoorremotefromthestation，becauseanyvariationsoflossinthefeederswouldnotgiverisetosimilarfluctuationsinanylampcircuit。Thefeedersmightbeoperatedatanydesiredpercentageoflossthatwouldrealizeeconomyincopper，solongastheydeliveredcurrenttothemainconductorsatthepotentialrepresentedbytheaveragevoltageofthelamps。
Thusthefeederscouldbemadecomparativelysmallincross-section。Itwillbeatonceappreciatedthat，inasmuchasthemainsrequiredtobelaidONLYalongtheblockstobelighted，andwerenotrequiredtoberunallthewaytothecentralstationwhichmightbehalfamileormoreaway，thesavingofcopperbyEdison\'sfeedersystemwasenormous。
Indeed，thecomparativecalculationofMr。Jenks，abovereferredto，showsthattooperatethesamenumberoflightsinthesameextendedareaofterritory，thefeedersystemwouldrequireonly128，739poundsofcopper，which，atthethenpriceof25centsperpound，wouldcostonly$39，185，orASAVINGof$168，627。50forcopperinthisverysmalldistrictofonlynineblocks。
Anadditionalillustration，appealingtotheeye，ispresentedinthefollowingsketch，inwhichthecomparativemassesofcopperofthetreeandfeedersystemsforcarryingthesamecurrentareshownsidebyside：
XII
THETHREE-WIRESYSTEM
THISinventioniscoveredbyUnitedStatesPatentNo。
274，290，issuedtoEdisononMarch20，1883。Theobjectoftheinventionwastoprovideforincreasedeconomyinthequantityofcopperemployedforthemainconductorsinelectriclightandpowerinstallationsofconsiderableextentatthesametimepreservingseparateandindependentcontrolofeachlamp，motor，orothertranslatingdevice，uponanyoneofthevariousdistributioncircuits。
ImmediatelypriortothisinventionthehigheststateoftheartofelectricaldistributionwasrepresentedbyEdison\'sfeedersystem，whichhasalreadybeendescribedasastraightparallelormultiple-arcsystemwhereineconomyofcopperwasobtainedbyusingseparatesetsofconductors——minusload——feedingcurrentatstandardpotentialorelectricalpressureintothemainsatcentresofdistribution。
Itshouldbeborneinmindthattheincandescentlampwhichwasacceptedatthetimeasastandardandhassoremainedtothepresentdaywasalampof110voltsorthereabouts。Inusingtheword“standard，“therefore，itisintendedthatthesameshallapplytolampsofaboutthatvoltage，aswellastoelectricalcircuitsoftheapproximatepotentialtooperatethem。
Brieflystated，theprincipleinvolvedinthethree-wiresystemistoprovidemaincircuitsofdoublethestandardpotential，soastooperatestandardlamps，orothertranslatingdevices，inmultipleseriesoftwotoeachseries；andforthepurposeofsecuringindependent，individualcontrolofeachunit，todivideeachmaincircuitintoanydesirednumberofderivedcircuitsofstandardpotentialproperlybalancedbymeansofacentralcompensatingconductorwhichwouldbenormallyneutral，butdesignedtocarryanyminorexcessofcurrentthatmightflowbyreasonofanytemporaryunbalancingofeithersideofthemaincircuit。
Referencetothefollowingdiagramswillelucidatethisprinciplemoreclearlythanwordsalonecando。Forthepurposeofincreasedluciditywewillfirstshowaplainmultiple-seriessystem。
InthisdiagramGandGrepresenttwogenerators，eachproducingcurrentatapotentialof110volts。Byconnect-
ingtheminseriesthispotentialisdoubled，thusprovidingamaincircuitPandNof220volts。ThefiguresmarkedLrepresenteightlampsof110voltseach，inmultipleseriesoftwo，infourderivedcircuits。Thearrowsindicatetheflowofcurrent。Bythismethodeachpairoflampstakes，together，onlythesamequantityorvolumeofcurrentrequiredbyasinglelampinasimplemultiple-arcsystem；
and，asthecross-sectionofaconductordependsuponthequantityofcurrentcarried，suchanarrangementastheabovewouldallowtheuseofconductorsofonlyone-fourththecross-sectionthatwouldbeotherwiserequired。Fromthestandpointofeconomyofinvestmentsuchanarrangementwouldbehighlydesirable，butconsideredcommerciallyitisimpracticablebecausetheprincipleofindependentcontrolofeachunitwouldbelost，astheturningoutofalampinanyserieswouldmeantheextinguishmentofitscompanionalso。Byreferringtothediagramitwillbeseenthateachseriesoftwoformsonecontinuouspathbetweenthemainconductors，andifthispathbebrokenatanyonepointcurrentwillimmediatelyceasetoflowinthatparticularseries。
Edison，byhisinventionofthethree-wiresystem，over-
camethisdifficultyentirely，andatthesametimeconservedapproximately，thesavingofcopper，aswillbeapparentfromthefollowingillustrationofthatsystem，initssimplestform。
Thereferencefiguresaresimilartothoseintheprecedingdiagram，andallconditionsarealsoalikeexceptthatacentralcompensating，orbalancing，conductor，PN，ishereintroduced。Thisistechnicallytermedthe“neutral“wire，andinthedischargeofitsfunctionsliesthesolutionoftheproblemofeconomicaldistribution。Theoretically，athree-
wireinstallationisevenlybalancedbywiringforanequalnumberoflampsonbothsides。Ifalltheselampswerealwayslighted，burned，andextinguishedsimultaneouslythecentralconductorwould，infact，remainneutral，astherewouldbenocurrentpassingthroughit，exceptfromlamptolamp。Inpractice，however，nosuchperfectconditionscanobtain，hencethenecessityoftheprovisionforbalancinginordertomaintaintheprincipleofindependentcontrolofeachunit。
ItwillbeapparentthatthearrangementshowninFig。2
comprisespracticallytwocircuitscombinedinonesystem，inwhichthecentralconductor，PN，incaseofemergency，servesintwocapacities——namely，asnegativetogeneratorGoraspositivetogeneratorG，althoughnormallyneutral。
Therearetwosidestothesystem，thepositivesidebeingrepresentedbytheconductorsPandPN，andthenegativesidebytheconductorsPNandN。Eachside，ifconsideredseparately，hasapotentialofabout110volts，yetthepotentialofthetwooutsideconductors，PandN，is220volts。
Thelampsare110volts。
Inpracticalusetheoperationofthesystemisasfollows：
IfallthelampswerelightedthecurrentwouldflowalongPandthrougheachpairoflampstoN，andsobacktothesourceofenergy。Inthiscasethebalanceispreservedandthecentralwireremainsneutral，asnoreturncurrentflowsthroughittothesourceofenergy。Butletussupposethatonelamponthepositivesideisextinguished。Noneoftheotherlampsisaffectedthereby，butthesystemisimmediatelythrownoutofbalance，andonthepositivesidethereisanexcessofcurrenttothisextentwhichflowsalongorthroughthecentralconductorandreturnstothegenerator，thecentralconductorthusbecomingthenegativeofthatsideofthesystemforthetimebeing。Ifthelampextinguishedhadbeenoneofthoseonthenegativesideofthesystemresultsofasimilarnaturewouldobtain，exceptthatthecentralconductorwouldforthetimebeingbecomethepositiveofthatside，andtheexcessofcurrentwouldflowthroughthenegative，N，backtothesourceofenergy。Thusitwillbeseenthatathree-wiresystem，consideredasawhole，iselasticinthatitmayoperateasonewheninbalanceandastwowhenunbalanced，butineithereventgivingindependentcontrolofeachunit。
Forsimplicityofillustrationalimitednumberofcircuits，showninFig。2，hasbeenemployed。Inpractice，however，wheregreatnumbersoflampsareinuseas，forinstance，inNewYorkCity，whereabout7，000，000lampsareoperatedfromvariouscentralstations，thereisconstantlyoccurringmoreorlesschangeinthebalanceofmanycircuitsextendingoverconsiderabledistances，butofcoursethereisanetresultwhichisalwaysononesideofthesystemortheotherforthetimebeing，andthisismetbyproperadjustmentattheappropriategeneratorinthestation。
Inordertomaketheexplanationcomplete，thereispresentedanotherdiagramshowingathree-wiresystemunbalanced：
Thereferencefiguresareusedasbefore，butinthiscasetheverticallinesrepresentbranchestakenfromthemainconductorsintobuildingsorotherspacestobelighted，andtheloopsbetweenthesebranchwiresrepresentlampsinoperation。Itwillbeseenfromthissketchthattherearetenlampsonthepositivesideandtwelveonthenegativeside。Hence，thenetresultisanexcessofcurrentequaltothatrequiredbytwolampsflowingthroughthecentralorcompensatingconductor，whichisnowactingaspositivetogeneratorGThearrowsshowtheassumeddirectionofflowofcurrentthroughoutthesystem，andthesmallfiguresatthearrow-headsthevolumeofthatcurrentexpressedinthenumberoflampswhichitsupplies。
Thecommercialvalueofthisinventionmaybeappreciatedfromthefactthatbytheapplicationofitsprinciplesthereiseffectedasavingof621/2percent。oftheamountofcopperoverthatwhichwouldberequiredforconductorsinanypreviouslydevisedtwo-wiresystemcarryingthesameload。Thisarisesfromthefactthatbythedoublingofpotentialthetwooutsidemainsarereducedtoone-quarterthecross-sectionotherwisenecessary。Asavingof75percent。wouldthusbeassured，buttheadditionofathird，orcompensating，conductorofthesamecross-sectionasoneoftheoutsidemainsreducesthetotalsavingto621/2percent。
Thethree-wiresystemisinuniversalusethroughouttheworldatthepresentday。