Skip to main content

Full text of "Men and volts; the story of General electric"

See other formats


From the collection of the 

i f 
7 n 

z <r m 

o Prejinger h 
v iJibrary 




San Francisco, California 


Men and Volts 



CxAe G/tory oj 
general (Ol 


John Winthrop Hammond 


(J. oO. oLtf)f)tncoit (Somfaany 


Copyright, 1941, by 




Preface ix 

Prologue xi 


The Creative Period of the Arc Light and the 
Incandescent Light 

1. Planters of the Acorn 3 

2. A Light in the West 7 

3. Miniature Moons 11 

4. One of the Four 17 

5. At Menlo Park 20 


The Period of Commercial Introduction of 
Electric Lighting 

6. The Arc Light Takes the Field 27 

7. Transplanted Pioneering 33 

8. Manhattan Initiated 37 

9. The General Electric Succession 52 

10. Lighting Up with Incandescents 57 

11. Lighting Up with Arcs 67 



Electric Transportation, Motors, the Trans 
mission of Power 


12. Broomstick Cars 75 

13. From Shoes to Dynamos 87 

14. Onward with Volts 99 

15. Destiny Comes to Town 111 

16. The Motor Marches On 115 

17. Tribulation on Wheels 120 

The Period of Expansion and Consolidation 

18. The Dilemma of Patents 141 

19. Beside the Mohawk 149 

20. The Edison Consolidation 153 

21. "Experts" on the Job 162 

22. New Faces, New Companies 169 

23. A Famous Fight ISO 


The Formation of 

the General Electric Company 

24. General Electric Emerges 191 

25. The World's Fair 213 

26. Coffin Faces the Panic 220 

[PART six] 

The Development of Big Generating Units, 

Beginning of Hydroelectric Projects, Expansion 

of Systems through Transmission 

27. White Coal 229 

28. The Skilled Workman Appears 240 



29. The Steam Locomotive Challenged 254 

30. Campaigning for Candlepower 259 


The Development of Steam Turbines, Start of 

Industrial Research and of Commercial and 

Financial Expansion 

31. Courage and Grime 275 

32. Magic Comes to Manhattan 285 

33. A Venture Into Research 298 

34. Steinmetz and the Arc Light 303 

35. Brains and Brawn 314 


Spectacular Applications and Discoveries, Con 
tinued Expansion, Birth of Radio, Major 
Achievements Marking the End of 
the Pioneer Period 

36. The Romance of Work 325 

37. The Taming of Tungsten 333 

38. The Fire of Prosecution 340 

39. The Eternal "Why?" 345 

40. Conservatism Routed 364 

41. A Period Ends 372 
Epilogue The Second Generation 385 
Appendix 397 
Index 425 


JOHN WINTHROP HAMMOND spent more than three years in 
collecting the historical material for this volume. In 1922, 
when he started this work, many of the pioneers of the elec 
trical industry were still living, and he was able to get the 
story from their own lips. Hammond had the temperament 
of the true historian; he waded through records, he tracked 
traditional stories to their sources. And because he saw the 
epic of electricity in terms of men, the story he wrote was 
a human one. To him, this history was a labor of love. 

John Hammond died in 1934. The three hundred thousand 
word manuscript which he had prepared was later edited 
and condensed to approximately its present dimensions by 
Arthur Pound, well-known writer on industrial subjects. 
With a few minor changes, it is here presented. 

Because Hammond's manuscript closed with the year 1922, 
a brief Epilogue has been added, highlighting the major 
achievements of the succeeding years. In addition, there has 
been included as an Appendix the statement presented be 
fore the Temporary National Economic Committee by Owen 
D. Young, Honorary Chairman of the Board. 



This volume is published as a record of the amazing ac 
complishments of those pioneers of the electrical industry 
who, in the face of almost insurmountable difficulties, har 
nessed the gigantic force of electricity and put it to work to 
serve mankind. But the drama of the progress of electricity 
is a never-ending one; it goes forward today with an ever- 
changing cast. May this record of the courageous work of the 
pioneers serve as an inspiration to the men and women who 
carry on their work today! 


To PEER INTO THE FUTURE to be allowed for one precious 
moment to see the world as it will be sixty years from now 
what wouldn't any of us give to see the marvels yet un 
dreamed of that will mark the year 2000? And even could 
we, by some strange happenstance, see for a fleeting mo 
ment the wonders of that day, would we believe what we 
saw? Probably not, for the world of sixty years hence will be 
as different from today as today is different from the world 
of sixty years ago. 

The world of sixty years ago 1880 some can still re 
member it. No electric lights, except in the hands of a few 
ridiculed experimenters; no electric streetcars, except in the 
minds of a few "addled visionaries'*; no electricity at all, ex 
cept in a few isolated applications that had not yet pene 
trated the consciousness of the laity. Yet in the years to 
follow, the experimenters, the visionaries, the men of science 
were to create a world as unbelievable to the people of that 
day as the imaginings of a Jules Verne. And even they could 
not foresee the vast changes that their work would bring. 

We, who today enjoy the comforts, the conveniences, the 



high living standards that they created for us, see but the 
outer shell of a grand and intricately woven matrix, into the 
creation of which has been spent the genius of three genera 
tions. To most of us, the world of today is the world of all- 
time of yesterday, today, tomorrow. We cannot peer ahead 
we are intolerant of the struggles of the past. And yet, if 
we but realize it, to stand at our present vantage point and 
follow step by step the weird discoveries of the pioneers is 
as fascinating a journey as the mind of man can hope to take. 
And having taken it, the world of today will never again be 
to us the world of all-time it becomes rather a position 
achieved by the diligence of those who have gone before; a 
foundation upon which we too must build to that unknown 
destiny that holds the fate of man. 

With the past to inspire us, no goal is too difficult, no ob 
jective too high, no idea too visionary to command our ef 
forts. No matter how large or how small the part we play, 
each is essential in the scheme of things. Ours is the torch to 
carry forward, that the world of tomorrow may be an even 
better world in which to live. 

In this volume, the pioneer period of General Electric's 
history is related in detail the history of the last two decades 
all too briefly. But someday, when time has added the per 
spective so essential in preparing a historical narrative, the 
story of those two decades must be told, for it was in these 
years that many of the efforts of the pioneers were brought to 
fruition. And in this period, too, were sown the seeds that will 
mature to create the wonders of the world to come. 

PHILIP D. REED, Chairman 
General Electric Company 


The Creative Period of the Arc Light 
, and the Incandescent Light 

Planters of the Acorn 

twentieth-century eye a simple, unhurried existence. Horse- 
cars jangled leisurely through the principal streets of the 
larger cities. Multitudes of horses furnished their motive 
power, at the expense of immense stables and many grooms. 
The life of these horses was rigorous, and on the average 
they lived only about three years. In 1886, according to an 
estimate published at that time, the 500 horse railways 
operating in the United States utilized 120,000 horses for 
25,000 cars, or more than four horses to a car. 

In these conveyances business men of the seventies rode 
to store or office. In the wintertime passengers' feet were 
embedded in hay or straw, with which the floors were 
carpeted for the sake of warmth. 

As evening approached, lamplighters made their rounds 
through the city streets. Many were Civil War veterans, and 
the "lame lamplighter" became almost a proverbial figure. 
The lamps made small, yellow spots of illumination in the 
darkness, dotting the gloomy streets in double lines. They 
burned with a certain amount of flicker from air currents in 



the square glass chambers. A high wind was apt to extin 
guish them. In thrifty cities the lamps were turned out on 
nights of the full moon. 

This was the era of the reciprocating steam engine. 
Wherever power in large volume was required, there the 
reciprocating engine was to be found. The engines were 
bulky and towering a bewildering spectacle of intricate 
parts in noisy motion. 

Gas companies flourished, though they were targets of 
public criticism. The young clerk courted his sweetheart by 
gas-light. One of the perennial jokes of the day was built 
upon the romantic practice of turning down the gas in the 
front room while Sally entertained her beau. In most houses, 
however, it was an almost daily chore to fill the kerosene 
lamps, trim wicks and keep the glass chimneys washed. 

New York in 1876 was a city of ground-scrapers. Few 
buildings rose higher than five stories, and in downtown 
Manhattan there were quaint brick structures with a Dutch 
air about them. Fulton Street, going down to the old Fulton 
Ferry, was the suburban outlet, where twice a day crowds 
surged in the great American "rush-hour." In the nation's 
manufacturing plants lines of shafting turned hour after 
hour and a maze of belts whirred ceaselessly; in the back 
country, mountain streams tumbled impetuously toward the 
sea, unharnessed, unmolested. 

Eighteen seventy-six was America's centennial year, and 
it's Centennial Exposition was destined to open a new gate 
way, on which might well have been carved a proverb after 
the manner of Solomon: "It is the glory of God to conceal 
a thing; but the honor of man to search it out." Through that 
gateway lay the Electrical Era. 

The prophets of that age smiled in condescension when 
its pioneers dared to assert that the future belonged to the 


mysterious force with which Ben Franklin, more than a 
century before, had played. People might listen; but soon 
they shrugged skeptical shoulders and turned to the other 
wonders of the Centennial Exposition, marking the end of a 
century of existence for the United States of America. 

The Exposition, like all such displays, was chiefly a record 
of things done. It was concerned with history, not with 
prophecy. Nonetheless, there was a real prophet present, 
not of flesh and blood, but a thing of iron and copper; not 
a man with ringing voice, but a man-made machine, whis 
pering a new language. That prophet was the electric 

There were two types of dynamos represented at the 
Exposition; the Gramme and the Wallace, one Belgian and 
one American. Small and unobtrusive, each supplied cur 
rent to a single arc lamp. One of them, connected to a second 
dynamo, drove a small pump. 

"What curiosities!" spectators remarked, held for a mo 
ment by the dazzle of bluish light. Then they passed on, to 
lavish their admiration upon the huge Corliss steam engine. 
This gigantic machine overshadowed everything in Ma 
chinery Hall. Thirty feet high, the world's greatest prime 
mover, and wonder of the exposition, it was capable of 
producing one thousand horsepower of mechanical energy! 

The mighty engine was daily applauded, the little electric 
dynamos quickly forgotten. Purring quietly in their corners, 
they gave no intimation that they would one day compel the 
steam giant to bow its head in defeat. Even before the Ex 
position closed, the principle of electricity yoked for service 
had established itself. Inventive minds kept prodding it into 
more tangible reality. These were the successors of Davy, 
of Faraday, of Henry. And in the work of four of them is to 
be found the acorn that became the oak. Each of these four 


was a pioneer; each struggled courageously to introduce a 
startling innovation. They were men of imagination and 
faith, whom doubt and criticism did not daunt. 

Charles Francis Brush in the year of the Exposition was 
fashioning an electric dynamo which created illumination 
through the agency of arc lamps. Elihu Thomson and James 
John Wood were soon to fashion dynamos of their own, each 
of distinct design. The fourth of these pioneers was Thomas 
Alva Edison. 

None of these four men had an inkling that his activities 
would be of value to the others. Practical dreamers though 
they were, each with his own theories, his own workshop, 
and his own financial supporters, there is no intimation that 
any one of them foresaw the tidal currents which in hardly 
more than fifteen years were to sweep them into a single 
powerful organization. 

The gates to a new era began to swing open under the 
pressure of new ideas. 

A Light in the West 

OLD "UNCLE" BALDWIN, so the story goes, had come to Cleve 
land in 1877 to see the sights. One of the most amazing 
things he observed was a light of peculiar brilliance blazing 
at the door of a little shop. 

"That's a wonderful light, wonderful!" the old fellow 
murmured to a young man who stood in the doorway. 

"That, sir," the young man told him, "is the light of the 
future electricity. Some day you'll see it light the entire 

Charles F. Brush, who at the age of twenty-eight had 
successfully brought out his electric dynamo and arc lamp, 
was the young man in the little shop. 

When Brush graduated from the University of Michigan, 
he was thoroughly familiar with the principles of electricity 
as then understood and with the several dynamos which 
had appeared. The most important of these were the "ring 
armature" dynamo, devised by Pacinotti, an Italian, and 
later introduced anew by Gramme, a Belgian, whose name 
it bore; and the "drum armature" type, invented by Alteneck 
of the German Siemens-Halske Company, and generally 
designated as the Siemens dynamo. 


Brush believed that he could improve upon them both. 
Five years after graduation he had worked out the design of 
his dynamo. He could have prepared working drawings at 
a moment's notice, but he doubted if there would be a 
market in America for such a machine. 

Geo. W. Stockly was at this time Vice President and 
Manager of the Cleveland Telegraph Supply Company, 
manufacturer of telegraph instruments, electric bells, and 
fire alarm systems, where Brush worked. Once Brush 
chanced to remark to Stockly that he could build a more 
efficient dynamo. Stockly caught at the idea, offered Brush 
the shop facilities of the Telegraph Supply Company, and 
agreed that if the dynamo proved successful, his company 
would undertake its manufacture. This was the encourage 
ment Brush needed. 

He went to his country home near Wickliffe and set to 
work. When toward the end of the summer of 1876 he drove 
to the door of Stockly 's office in a buggy, there was on the 
seat beside him a machine that looked no larger than a 
model, hand-built throughout except for iron castings fur 
nished by the Telegraph Supply Company. 

The machine was taken into the company's shop and con 
nected to an arc lamp. The shaft of the armature was belted 
to the main shaft of the shop. It was a tense moment when 
the engine started. Brush with his dynamo stood at the limit 
of electrical knowledge. Was he to pass a boundary? The 
lamp, shining steadily, signalled his advance in dispelling 
some of the darkness that lay beyond. 

Brush had fulfilled his promise to Stockly, but he still 
needed an arc lamp to go with his dynamo. In that period 
there were not many lamps to be found and none of those 
tested gave the results that Brush desired. Finally he under 
took to design one of his own. 


In Brush's arc lamp automatic action permitted the proper 
regulation of the arc without the touch of human hands. The 
mechanism was ingenious but simple. Working entirely by 
magnetic control, it kept the carbons always a certain 
distance apart, the arc always at a given length. 

Two years later Brush completed his memorable invention 
of a series arc lamp with regulating shunt coil. This inven 
tion, which enabled him to introduce arc lighting from 
central stations as a commercial venture, marked the birth of 
that industry. Neat and compact in appearance, the new arc 
lamp included a short-circuiting device, the function of 
which was to carry the current around any disabled lamp 
to the others in the series, thus keeping the circuit unbroken. 

Ambition growing with success, Brush turned again to his 
dynamo, a one-light machine, with an armature only nine 
inches in diameter. Having achieved a series lamp, it was 
inevitable that Brush should seek to build a larger dynamo, 
his goal being a sixteen-light machine, which would give 
him the basis of a practical electric lighting system. 

Then began a period of designing and experimenting 
during which the Telegraph Supply Company furnished 
financial backing in such amount that the directors of the 
company became apprehensive. A majority of them pro 
tested to Manager Stockly that Brush was working in a 
field without a future. They insisted that it would be better 
for the company to confine itself to its "legitimate business." 
Fortunately for future generations, Stockly quieted the 
fears of his directors for a year or so. 

In the winter of 1876-77, the company entered into a 
formal agreement with Brush whereby they received the 
sole right to manufacture and sell the Brush system under 
any patents the latter might obtain. Patents were issued to 
Brush the following April for his arc dynamo, and in the 


fall for his arc lamp. Other patents followed, including one 
for a dynamo to operate at constant voltage, used first for 
electroplating, later for incandescent lighting. As yet the 
world of commerce, industry, and trade knew little of these 
inventions. In 1877, however, two of Brush's small-size 
dynamos and two of his lamps came suddenly into prom 
inence among several contemporary machines that were 
tested by the Franklin Institute of Philadelphia. The com 
mittee which studied these machines was composed of two 
men, Professors Edwin J. Houston and Elihu Thomson. 

Thomas A. Edison 

Elihu Thomson 

Charles F. Brush James /. Wood 


These photographs were taken about 1880, shortly after Edison invented the 
incandescent lamp. 



A dynamo built by Thomson in 1878. 

Miniature Moons 

AT THE AGE OF 11 Elihu Thomson had fashioned his first 
electrical device, an old-fashioned frictiorial machine, its 
revolving element consisting of a wine bottle with a hole 
knocked in the bottom. With it the lad charged a small 
Leyden jar and proceeded to demonstrate by inflicting a 
shock upon his incredulous father. Parental amusement 
spurred young Elihu to build a larger Leyden battery of five 
jars. When the elder Thomson touched this reservoir of 
electricity, he received a shock that nearly threw him off 
his feet. After that Elihu's experiments were viewed with 
respect from a distance. 

At the close of a brilliant high school career, Thomson 
was familiar with most of the branches of mechanics, and 
his incessantly active mind had approached the limits of 
contemporary electrical knowledge. Professional advance 
ment kept pace with his scientific development. At eighteen 
he was made Assistant Professor of Chemistry at the Boys' 
Central High School of Philadelphia, a school of excep 
tionally high standing, and five years later, in 1876, he 
succeeded to the full professorship of chemistry and me 



Before he was twenty-four, Thomson was invited by the 
Franklin Institute to deliver a course of five public lectures 
on electricity. Though young for such an honor, he seized 
the opportunity eagerly. Contemporary textbooks described 
electricity as being of two forms, static and dynamic. 
Through research Thomson had discovered error in the 
definition. It was the purpose of his lectures not only to 
prove the error but to establish his own theory in place of 
the old. 

For five meetings he packed the auditorium and held an 
audience spellbound. During one lecture he exhibited a 
small direct-current dynamo of his own design and construc 
tion, which created sufficient current to light an arc lamp on 
the platform. An astonished audience viewed the unfamiliar 
phenomenon in utter silence. By sheer accident in the course 
of another experiment he discovered the principle of electri 
cal resistance welding. When the series was concluded, 
Elihu Thomson had become a person of standing in the 
scientific world, for he had shown that electricity, instead of 
being two forces, was inherently the same throughout na 

Shortly after Thomson's lectures the Franklin Institute 
undertook the first measurement of the output of an electri 
cal dynamo. Desiring to purchase a dynamo, the Institute 
conducted scientific tests of several types of machines then 
available. Several months were consumed in testing a 
Gramme dynamo, two Brush dynamos, and two Wallace 
dynamos by Professors Houston and Thomson, colleagues 
at the Boys' Central High. 

It was no easy task which these men had undertaken, for 
electrical measuring facilities in those days were crude. But 
Thomson and Houston did their work so well that their 
results were studied by scientists for years to come. The 


Gramme dynamo made the best record an efficiency of 38 
per cent! The two Brush dynamos had respective efficiencies 
of 31 and 27 per cent. But it was found that the Brush 
machines yielded the most powerful current and hence 
produced the best light. For this reason, the two professors 
recommended that the Institute purchase a Brush dynamo. 
In this disinterested manner did Thomson first come into 
contact with Charles F. Brush and his work. 

Then occurred an event which greatly stimulated 
Thomson's thinking the first Brush arc lamps were in 
stalled in Philadelphia at the store of John Wanamaker. 
Five Brush dynamos were put to work, each supplying cur 
rent to four arc lamps in the windows of the store. These 
lights became one of the wonders of 1878. People gathered 
in throngs on the sidewalk to examine them. For weeks they 
were talked about and a contemporary writer called them 
"miniature moons on carbon points, held captive in glass 

Inspection of these developments convinced Thomson 
that the field of electrical activity had an unlimited future. 
Within a year he had justified his expectations by the per 
formance of one of the most remarkable electrical experi 
ments of the period. In the lecture hall of the Franklin Insti 
tute, before a small group of spectators, Thomson exhibited 
a revolutionary dynamo and two arc lamps which he and 
Professor Houston had made. While friends were exclaiming 
over this exhibition and the genius of the man who had pre 
pared it, Thomson was drawn by a chance encounter into 
the field of electrical manufacture. 

Enthusiasm for photography frequently brought Thomson 
to the studio of a commercial photographer in Philadelphia, 
Thomas H. McCollin. One day McCollin invited his cousin, 
George S. Garrett, up from Garrettsford, Pennsylvania, to 


witness a demonstration of Thomson's dynamo, induction 
coils, and vibrating arc lamp. 

During the demonstration Thomson remarked, "I can 
build a better machine than this; one that will run any num 
ber of lights you want/' Garrett was enthusiastic. "Let's 
build a four-lighter," he suggested. "Ill stand the expense." 
It was virtually an offer to Thomson to enter the business 
of building an arc-lighting system for market. He called 
upon his associate, Prof. Houston. Almost overnight they 
prepared drawings, engaged a pattern-maker, and arranged 
with the Harrison Machine Shop to do the machine work. 
Thomson himself put the electrical windings on the new 
dynamo. The inventor himself did not realize the possibilities 
of this dynamo until he watched it in operation. 

In the oven room of the Fuller Bakery in Philadelphia, 
Garrett was permitted to install the Thomson-Houston dy 
namo and its four arc lamps for all-night illumination of the 
bakeshop's operations. It was the beginning of summer, and 
the oven room registered a temperature of 
140 degrees Fahrenheit. The heat was so ter 
rific that Houston could stand it for only a 
short time. He worried about his colleague, 
who stayed in the oven room until late every 
night. But nothing could deter Thomson, 
who laughed at Houston's protests, rein 
forced himself with a pitcher of ice water 
and continued his vigil. 

The hot spell passed. Garrett, pleased with 
the results of the Fuller Bakery tests, asked 
for a machine that would operate more arc 
THE ARC LAMP lamps. Thomson set to work immediately. 
?5X 7 t> S r ,?o B J^ Changing the circuit connections of the dy- 

CHARLEsr. 111 i i i 

BRUSH namo enabled it to supply eight arc lamps 



instead of four. The machine was successfully tested just as 
it came from the shop; and several Brush arc lamps were 
borrowed to fill out the circuit, for Thomson had no supply 
of his own. 

Now came discovery which was to reward long hours of 
experiment. Thomson switched one of the lamps out of 
the circuit while the dynamo was running. He knew what 
would happen. The moment one lamp was cut out, the bal 
ance of the circuit was upset. Thomson stepped to the dy 
namo and changed the position of the metal brushes that 
were picking current off the whirring commutator until he 
had restored the current entering the circuit. He found that 
if he adjusted the brushes properly each time, he could turn 
off all of the lamps, one after the other, and the current 
would stay constant. Instantly he saw 
the possibility of automatic control, 
and set about designing a regulator. 
Its superiority lay in the fact that in 
dividual lamps could be switched off 
or on at will without upsetting the cir 
cuit. Other systems either introduced 
a large section of iron wire to set up 
a substitute resistance, or more com 
monly, maintained a reserve bank of 
lamps at the power station, one of 
which had to be switched in by an at 
tendant every time a lamp in the 
working circuit was switched out. 

There was a fire one night in John 
H. Gardner's brewery, where the sec- 

_ __ fit T 

ond installation or the Thomson-Hous- 

ton system was located. Several lamps THOMS AND 

were disabled, yet the circuit was un- HOUSTON 




impaired. The automatic regulator kept the other lamps 
operating normally and lighted the rescue of all the horses 
from the burning stable. One of the firemen, after seeing the 
arc lamps drenched with water, exclaimed, "What the dick 
ens kind of a light is that? You pour water on her and she 
won't go out!" 

Garrett and Thomson had survived an emergency, and 
upon the success of their system they now began to build a 
thriving business. For the moment, however, let us leave 
them and examine contemporary developments. 

One of the Four 

THE SHOP FOREMAN of the Brady Manufacturing Company 
of Brooklyn was James J. Wood, an intelligent and observant 
young man of twenty-one. He was good-natured and oblig 
ing, and so thoroughly likable that he was known to every 
one as "Jimmy." Some of his subordinates in the shop never 
knew his last name. Yet young Jimmy Wood was destined to 
take his place with Brush, Thomson, and Edison as one of 
the four outstanding pioneers in applying electricity to com 
mon use. 

Jimmy Wood had determined to build a dynamo that 
would reduce heat losses, and that would weigh well below 
the average of dynamos then in use, which was around six 
hundred pounds. He had no money, but his employer, James 
Brady, promised to manufacture the proposed dynamo if 
Wood would allow him a half -interest in the patent. To this 
the latter agreed. When the dynamo was finished, in May 
1879, it was, like Brush's, hardly larger than a model. It 
weighed 87 pounds; its armature was only eight inches in 
diameter and two and one-half feet in length. A trifle more 
than one horsepower would operate it. 



Officials of the Fuller Electrical Company were dumb 
founded by the achievement, and not until a reputable elec 
trical engineer tested it would they credit what this young 
man had done. 

Then came action. They paid Wood and Brady ten thou 
sand dollars for the patent rights. They renamed their organ 
ization the Fuller- Wood Company, with Wood's dynamo as 

Built by Jimmy Wood 

their chief stock-in-trade. Soon it was being sold, installed, 
and operated in a wide-spreading territory. 

Level-headed even after this sudden turn of fortune 
Jimmy Wood proceeded to improve his already efficient ma 
chine. He knew that his original dynamo had weaknesses 
which could be avoided. Accordingly his commercial dy 
namo, developed immediately, was built for higher voltage 
and for series-circuit work. Any tendency of the machine to 


The laboratory is the long building with the porch. In the foreground 
is the office building; the building at an angle is the "glass house" 
where experimental lamp bulbs were blown; at right angles to the 
laboratory in the rear is the machine shop; the experimental railway 
can be seen at the right. 


Edison's Menlo Park Laboratory, showing the lamp used in the first 
public demonstration on pole in foreground. 

OCTOBER 19, 1879 

Artist's reconstruction of the scene in the Menlo Park laboratory when 
the first successful lamp was turned on. 



emit flashes, or to "spark" at the point where the brushes 
were in contact with the commutator was avoided. Wood's 
dynamo showed no sparking; it ran with little noise; it was 
utterly simple. When ready for the market it weighed 120 

Like Thomson, Wood devised an automatic regulator, 
which included the use of two brushes under each magnetic 
pole. These brushes operated not only to in 
sure proper change in voltage, as more or 
fewer lamps were placed in the circuit, but 
also to prevent destructive sparking. 

By this time arc lamps were installed in 
doors as well as out, as promoters pushed the 
idea of interior arc illumination. Salesmen 
went after the business of factories, hotels, 
and theaters, but they were careful not to 
say too much about voltage. It was a weak 
ness of these systems that for interior instal 
lation a voltage dangerous to human safety 
had to be sent into customers' premises. 

About this time a phrase was occasionally 
heard: "The subdivision of the electric light/* 

What did it mean? Could the arc lamp be JIMMY WOOD'S 

, j. .j , ^ u i ARC LAMP 

subdivided r Could anyone cut up a lamp 

of two thousand candlepower into twenty lamps, each of one 
hundred candlepower? 

While a score of inventors had such a small unit in their 
thoughts, the true apostle of incandescence had sequestered 
himself at Menlo Park. He was still to be heard from, still 
an unknown. But the hour was at hand when his genius 
should quicken a world to gratitude. 


At Menlo Park 

AWAY FROM THE BUSTLE OF CITIES, picturesque in its seclu 
sion, the hamlet of Menlo Park was, until 1876, one of the 
byways of life. Then suddenly into it came a man who 
seemed little less than a magician and who raised it to fame 
among the cities of the nation. For a few teeming years it 
started into intense activity, recurring excitement, luminous 
nights. After that, quiet. The tide of activity, as suddenly as 
it flowed, was at ebb, and Menlo Park relapsed again into 
tranquillity, retaining only the halo which one man had 
given it to wear forever. 

The first intimation the village had of the newcomer was 
when a gabled dwelling began to rise near the railroad 
station. Inhabitants learned that the owner's name was Edi 
son; and after a while they began to see him an energetic 
young man with searching eyes and a ready smile. Not far 
from his home, he erected a laboratory, two stories high and 
a hundred feet long, built of white clapboards, with a porch 
across the twenty-five-foot front. Other buildings rose about 
the same time a machine shop, power house, and a library- 

Into this small center workers began to come. The hum of 



enterprise arose. Expectancy was in the air. Only shortly 
after his arrival the infectious enthusiasm of a young man 
had caught up, not alone associates, but the town where 
they lived. Menlo Park was a conquest of Edison's, and 
henceforth it was to acknowledge his residence with grati 

But the scientist, however great his zeal, cannot apply his 
ideas without practical support. Grosvenor P. Lowrey, a 
New York lawyer of high standing, was Edison's most loyal 
admirer among men of influence. Jovial and good-hearted, a 
typical old-school barrister with Horace Greeley whiskers 
under his chin, he had closely followed Edison's work with 
the phonograph and telephone transmitter, acting as legal 
adviser and acquiring a tremendous faith in the ability of the 
young inventor. When, around 1878, public interest began to 
lose itself in wonder over the arc light, Mr. Lowrey was one 
of several to think of electric lighting as an opportunity for 
Edison. Immediately he set about securing the necessary 

Through the interest of backers came the incorporation 
in October, 1878, of the Edison Electric Light Company. 
Edison was now committed to experimenting with electric 
lighting and he was provided with ample funds upon which 
he began to draw. He increased equipment, hired more 
men, and put all his characteristic intensity, his great tech 
nical imagination, into the effort of producing a small-unit 
electric light. 

Hundreds of experiments took place for fourteen months, 
and more than $40,000 was expended. Gradually a service 
able lamp filament of carbonized substances along with an 
all-glass, hermetically-sealed bulb were developed. Lastly, 
the means of obtaining and keeping a vacuum in that bulb 
were worked out. 


Toward the end of October, 1879, Edison carbonized a 
cotton thread. He placed it, bent in horseshoe form, inside 
one of his sealed glass bulbs. On the evening of October 19, 
this crude experimental lamp, standing upright on a table, 
was connected to an electric circuit. A knot of Edison work 
ers had gathered to see what would happen. The central 
figure was the wizard himself. With him were Francis R. 
Upton, his mathematician; Charles Batchelor, his model- 
maker; John Kruesi, his machine-shop expert; Ludwig 
Boehm, his glass-blower; and Francis Jehl and Martin Force, 
of the laboratory staff. 

Current was switched on. The lamp, responding instantly, 
glowed with a soft light. Quickly they measured the re 
sistance. It was 275 ohms ample for their purpose and 
overwhelmingly greater than the four or five ohms of pre 
vious incandescent lamps. Then all sat down to watch the 
slender horseshoe of light, half expecting it to vanish. But 
hour after hour it continued to glow until the night was 

No one had eaten, no one had thought of sleep. The grey 
of a second dawn found them at their vigil. Upton alone 
excused himself, feeling that he should not altogether desert 
his home, but he hurried back soon after daylight. 

About one o'clock on the second afternoon, October 21, 
more than forty hours after it first received the current, the 
filament burned out. 

The spell was broken. The men leaped up with cries of 
jubilation. Edison was quiet in the hour of his tremendous 
success. As the little lamp glowed, he had envisioned "great 
cities lighted from central stations," and his mind was alive 
with plans. But all he said, when the glow finally vanished, 
was, "That's fine that's fine! I think we've got itl If it can 
burn forty hours, I can make it last a hundred." 

Edison built these dynamos in 1883. 


This lamp, with a filament of carbonized 

bristol board, was exhibited by Edison on 

New Year's Eve, 1879. 

OF 1886 



The successful demonstration of a solitary lamp left the 
spectators awe-struck. Little did they realize, however, that 
what they had seen was to expand electrical usage literally 
to the confines of civilization; for the incandescent lamp 
could go where the arc lamp could not into office buildings 
and into homes. It was a lamp that could compete with gas! 
The London technical journal, Engineering, remarked that 
"if he (Edison) has indeed arrived at the solution claimed 
. . . gas has found a dangerous rival." No wonder gas stocks 
were disturbed! 

Edison had experimented with a dynamo throughout 
1879. The genius of Menlo Park had analyzed the scientific 
theories regarding dynamos and had studied the report of 
the Thomson-Houston tests at the Franklin Institute. The 
highest efficiency of any dynamo then examined was 38 per 
cent, and he determined that his machine must not lose 
over 60 per cent of the power put into it. 

The Edison dynamo had large magnets indeed, designed 
to create a powerful magnetic field. They represented much 
experimental work. Their dimensions were odd, for the mag 
nets rose three and a half feet into the air and were joined 
at the top by an iron crosspiece which completed the mag 
netic circuit and made the contrivance resemble an immense 
Roman numeral, II. Some of Edison's associates dubbed the 
dynamos "long-waisted Mary Anns," though officially they 
were termed the Edison bi-polar. The armature was modeled 
after the efficient drum type. But it had the advantage of 
being laminated; that is, it was constructed of sheets, or 
discs, of iron, insulated from each other although mounted 
on a single shaft. This innovation resulted from innumerable 
experiments by Edison's mathematician, Upton. When the 
ungainly machine was tested, the result surprised even Edi 
son and Upton. It was ninety per cent efficient! 


Notwithstanding this remarkable performance, the other 
inventors gave no indication of relinquishing the field to 
any "incandescent lighter/' even though sponsored by the 
miracle man of Menlo Park. 


The Period 

of Commercial Introduction 
of Electric Lighting 

The Arc Light Takes the Field 

AN ARC LAMP, hanging from the balcony of Dr. Long- 
worth's residence in Cincinnati, was attracting curious 
crowds through the early days of 1878. Current was supplied 
from a Brush dynamo in the basement, but to the mystified 
watchers the lamp was an entity in itself. 

One evening when Brush was in Cincinnati to examine his 
invention, he walked through the crowd before Dr. Long- 
worth's home. A man with a knowing air was explaining to 
his neighbors how the light was obtained. 

"See that little box at the top?" he said, pointing to the 
metal case containing the control mechanism. "That's a can 
full of oil. The oil flows down through those side-pipes to 
the burner of the lamp and feeds the flame. That's all there 
is to itl . . . And what are you smiling at?" he asked the 
stranger who had joined his listeners. 

"I didn't realize the lamp was as simple as that," answered 

In Cleveland, Brush's home town, there was no such skep 
ticism to surmount. On the streets and in Monumental Park 
the first large-scale installation of arc lighting, as an exhibi- 



tion rather than a commercial development, was accom 

It introduced a method of early outdoor arc lighting soon 
to become characteristic the elevation of lamps upon tall 
masts, or towers. This practice caused people to think of 
the arc light as a substitute for the sun and, with an immense 
naivety replacing their skepticism, they thought that a few 
such lights, hf ted high above all surroundings, would turn 
night into day. 

This first Cleveland installation was a modest one. Twelve 
lamps, mounted on eighteen-foot posts, were supplied with 
current from a Brush dynamo in the Telegraph Supply Com 
pany's shop. They were first lighted on April 29, 1879. 

The Cleveland Plain Dealer, describing the scene the day 
after the lights were first turned on, said: 

"Thousands of people gathered . . . and as the light shot 
around and through the Park a shout was raised. Presently 
the Grays Band struck up in the pavilion, and soon after 
ward a section of artillery on the lake shore began firing a 
salute in honor of the occasion. 

"The light varied some in intensity, when shining its 
brightest being so dazzling as to be painful to the eyes. In 
color it is of a purplish hue, not unlike moonlight, and by 
contrast making the gas lights in the store windows look a 
reddish yellow. 

"The Telegraph Supply Company's establishment . . . 
was thrown open to as many people as could be accom 
modated at a time to go through the works and inspect the 
machinery which sends light over the wire to blaze out 
between the carbon points in the lamps." 

Two months after Cleveland had so proudly applauded 
the achievement of one of her citizens, the Brush system 
found its way from the Middle West to the Pacific coast. 


In San Francisco, on June 30, 1879, was incorporated the 
California Electric Light Company. George H. Roe was the 
organizer, and William Kerr sold the electrical equipment, 
holding Brush territorial license for California, Oregon, 
Washington, and Nevada. 

This pioneer company proposed to sell electrical illumina 
tion to cash customers, something never before attempted. 
The few commercial installations of any inventor's apparatus 
then in existence had been solely for private use. Individ 
uals, merchants, hotels, and theaters had purchased plants 
and arc lamps; they had had their premises wired; and the 
lights, supplied from their own private generating plants, 
were utilized for private benefit on their own property. 
These were designated as "isolated lighting plants." 

The San Francisco Company was the first in this country, 
if not in the world, to enter the business of producing and 
selling electric service to the public. Its generating plant was 
in the rear of what is now the Pacific Building, and modest 
indeed was the equipment first installed. It consisted of two 
Brush dynamos, one supplying six lamps and the other six 
teen. In December two dynamos of the larger size were 
added. Customers were not lacking, though the rate was 
high. A flat rate of $10 per week per lamp was charged, as 
metering of the current was quite unknown. As the system 
was improved, rates were reduced, until eight years later it 
was $3.00 per week for current furnished up to 9:30 o'clock 
in the evening (11:00 on Saturdays), $4.00 for current up 
to midnight, and $6.00 for all-night service. No current was 
furnished on Sundays and holidays. The city and county of 
San Francisco were eventually among the customers of the 
plant, setting up a street lighting system of 21 masts, 50 feet 
in height, each having four 4,000-candlepower lamps. 

Brush saw that the early carbons had high electrical re- 


sistance and burned out rapidly, thus becoming expensive. 
They were made from gas-retort carbon and contained from 
three to five per cent of ash, which was fatal to steadiness 
in the light, causing a flickering. 

Both Thomson and Brush worked to improve carbons as 
soon as they had developed their systems of arc lighting. 
When George S. Garrett expressed doubt concerning a satis 
factory supply, Thomson accepted the challenge: "If we can 
not buy carbons in the market, we can certainly make them." 
In 1879, with the assistance of the photographer, Thomas H. 
McCollin, he procured a supply of hard carbon as raw 
material, powdered it, mixed it to secure cohesiveness, and 
moulded it into sticks, which were afterward baked. 

Brush attacked the problem in a totally different manner. 
He utilized the by-products of mineral oil distilleries to pro 
duce his "still-coke" carbon, which contained only three 
hundredths of one per cent of ash. After the carbon sticks 
were moulded and baked, he electroplated them with cop 
per to reduce electrical resistance and to retard combustion. 
This was the origin of his "copper-coated" carbons, on which 
he obtained a patent in 1877. 

Having won their way in the West, Brush lights were in 
troduced in the East, where they were installed at Wana- 
maker's Philadelphia store and exhibited at the fair of the 
Massachusetts Charitable Association in Boston. 

Niagara Falls was illuminated for the first time on July 4, 
1879, by a sixteen-light Brush dynamo and arc lamps. The 
dynamo, driven by a waterwheel, was a pioneer hydro 
electric plant. A full complement of lamps was always op 
erated; there was no need to switch off some and leave the 
rest burning. 

In the year 1880 the Brush fortunes were given a tremen- 


dous impetus by an actual sale of a system to a little com 
munity in northern Indiana. 

The governing authorities of Wabash, Indiana, had found 
that electric lighting would not only cost some $800 a year 
less than gas lighting, but would yield a greater volume of 
illumination. Accordingly they contracted for a Brush instal 
lation with four lamps of 3000 candlepower each, mounted 
on crossarms atop the dome of the Court House 200 feet 
above the ground. This was the first municipally owned 
electric lighting plant, and Wabash the first town wholly 
lighted by electricity, for it was planned to illuminate the 
city from a single point. 

Both press and public followed the progress of the experi 
ment with intense interest. Early on the day when the cir 
cuit was to be turned on, awed and wondering folk com 
menced to pour into Wabash from the surrounding country. 
Newspaper correspondents traveled from cities as far distant 
as Chicago and New York. By 8:00 o'clock on the moonless 
evening of March 31, 1880, more than 10,000 persons were 
crowded about the Court House. 

Let newspaper accounts tell the tale. 

The Fort Wayne Daily Sentinel: "Promptly as the court 
house clock struck eight, the thousands of eyes that were 
turned toward the inky darkness over the courthouse saw a 
shower of sparks emitted from a point above them, small, 
steady spots of light, growing more brilliant until within a 
few seconds after the first sparks were seen, it was absolutely 
dazzling. A loud shout went up from the crowd, the band 
began to play. . . ." 

A Chicago Tribune reporter went, he says, "up into the 
dome, right under the light, where (he) beheld a scene of 
magnificent splendor. For a mile around, the houses and 


yards were distinctly visible, while the far-away river flowed 
like a band of molten silver." 

An eye-witness pictures the emotions of his fellow specta 
tors: "People stood overwhelmed with awe, as if in the 
presence of the supernatural. The strange, weird light, ex 
ceeded in power only by the sun, rendered the square as 
light as midday. Men fell on their knees, groans were uttered 
at the sight and many were dumb with amazement. We 
contemplated the new wonder in science as lightning 
brought down from the heavens." 

The Wabash Plain Dealer was pardonably exuberant: 
"Yesterday morning the city of Wabash woke up and found 
itself famous. It is today the best advertised town in the 
United States. From Maine to California telegrams of the 
Associated Press flashed the intelligence that the problem of 
lighting the streets of an entire city solely by electricity was 
solved. ... In the street in front of the residence of Levi 
Linn, who is, we learn, still denouncing the electric light as 
a fraud, our reporter, standing three squares distant from 
the light, was able to read with ease the advertisement of 
Ebblinghouse and Austin." 

Thus was Wabash given its baptism of light. For the 
moment Brush's competitors were out-distanced and for 
gotten. But one of them at least, Elihu Thomson, already had 
his second wind and was pressing strongly forward. 

Transplanted Pioneering 

PROFESSOR THOMSON'S PRESTIGE had already travelled into 
New England when New Britain, Connecticut, sent a group 
of men, headed by Frederick H. Churchill, to make a proposi 
tion to this promising inventor. They wanted to organize an 
electrical company on the strength of his patents. It was an 
interesting offer and Thomson was not long in giving a favor 
able answer. 

News of his impending departure spread through the 
Boys' Central High School as commencement day drew 
near. The honor man in the class of 1880 was Edwin Wilbur 
Rice, Jr., who had been Professor Thomson's most brilliant 
pupil. A few days after graduating the young man cast his 
lot with the professor. Before autumn the two men, opti 
mistic and courageous, departed for New Britain. Neither 
suspected that they were to spend the rest of their lives in 
a common enterprise, were to "team up" together through 
many viscissitudes, were to help shape a new industrial 

A company was chartered with William Parker, president; 
Mr. Churchill, Treasurer and General Manager; and Pro- 



fessor Thomson, "Electrician" a term commonly used to 
designate the technical man, or inventor, whose patents pro 
vided the foundation for commercial organization. Profes 
sors Thomson and Houston executed a formal contract with 
the new American Electric Company. Briefly, it bound the 
company to manufacture "with all reasonable diligence" the 
inventions covered by the Thomson-Houston patents. These 
articles, enumerated in the contract, were to be sold and put 
into public use "by diligent and continuous attention . . . 
in all reasonable and advisable ways." In case the company 
failed to live up to stipulations, the patents were to revert 
to the patentees; stock of the company, given to the two 
professors in exchange for the patents, was to be surrendered 
by them. 

So Thomson, with the spirit of the scientific explorer tin 
gling in his brain, set up his humble working quarters in the 
basement of an old factory; and he and his young assistant 
set briskly to work to improve the Thomson-Houston arc 
light system. 

Both men personally wound the wire on the armatures of 
their first dynamos. When they were not helping in the shop 
work factory hands were few and inexperienced at first 
they were ceaselessly experimenting. They were always at 
work at seven in the morning; they seldom stopped before 
six at night. Frequently they worked in the evening, with a 
few assistants as devoted as themselves. On one such occa 
sion they improvised a luncheon of raw oysters, on which 
somebody suggested using paraffin in lieu of butter. Thom 
son, however, took his without garnishing, saying that he 
objected to "the candlepower flavor." 

His residence at that time was in New Britain's principal 
hotel. Rice had moved to a locally celebrated establishment, 
Mrs. Moore's boarding house, which became a counterpart 


to Mrs. Jordan's famous boarding house in Menlo Park. 
There Rice shared a room with a young bookkeeper and cost 
clerk, George E. Emmons, who had been hired by Churchill 
to establish a bookkeeping system for the new American 
Electric Company. Rice woke up once in the middle of the 
night and found his room-mate hard at work on the com 
pany's accounts. When he inquired what was going on, Em 
mons replied that he was trying to run down a shortage in 
his trial balance. 

"How much is the shortage?" asked Rice. 

"Three cents." 

"Good Heavens!" cried Rice. "Come to bed and 111 give 
you the three cents in the morning." 

Whereupon the young cost clerk laid down his pencil and 
gave his sleepy room-mate an academic lecture on the art 
of bookkeeping and the tremendous importance of a few 
missing cents in the trial balance. 

The first machines in these days ran only about ten or 
twelve lights. Machines of larger capacity were demanded, 
and Professor Thomson set to work to design one to operate 
twenty arc lamps. Thomson was constantly looking ahead, 
and he foresaw that, as arc light systems grew in size with 
correspondingly higher voltages, lightning protection would 
become a problem. Discussing the question with Rice, he 
pointed out that an electrical discharge caused by lightning 
could be led harmlessly to earth, but that a heavy arc, pro 
duced by the current of the circuit, would attempt to follow. 
This arc would have to be overcome. 

Thomson took a large steel magnet and walked over to 
several arc lamps which were shining in the test room. He 
placed the magnet in such a position that the arc was be 
tween the two magnetic poles. Immediately the arc vanished 
and the lamp went out. A magnet had blown out the arc. 


As long as the magnet remained in proximity it was impos 
sible to produce a fresh arc. From this experiment came 
Thomson's application of the magnetic blowout principle. 
He used it first in an efficient type of lightning arrester and 
later utilized it for electric switches. 

Manhattan Initiated 

A SINGLE MAN, IN 1880, precipitated a scientific contro 
versy, caused dismay to gas companies, and threw the stock 
market into turmoil. Shares in the Edison Electric Light 
Company rocketed. Stock that had a par value of $100 a 
share was not uncommonly sold at $500. In fact, the day 
came when three shares sold for $2000 each, were resold 
in a few minutes for $3700, and later the same day for 
$5000 apiece. While speculative activity was at its height, 
stock at times rose $100 a share within an hour. Yet this 
was but a surface indication of a general undercurrent of 
excitement. People everywhere were confident that a revolu 
tion in artificial illumination was at hand. And so it was 
largely through the resistless energy of Thomas Edison. 

Foremost in Edison's mind was the necessity of producing 
incandescent lamps in quantity. Even while experimenting, 
he sought to begin manufacture for he knew that he had a 
product which could be commercially handled. 

The word "filament" was Edison's contribution to the no 
menclature of the future industry. His immediate purpose 
was to find the best possible material for this filament. Ex- 



periment followed experiment; day-and-night activity at 
Menlo Park was incessant. 

"Somewhere in God Almighty's workshop," Edison said, 
"there is a dense, woody growth with fibres almost geometri 
cally parallel and with practically no pith, from which excel 
lent filaments can be cut." 

For nearly eight years Edison explorers searched the 
world for that "dense, woody growth" especially in the form 
of some species of bamboo. Bales and boxes containing plant 
specimens were shipped to Menlo Park from China, Japan, 
Ceylon, Cuba, and South America. No less than six thousand 
vegetable growths were tested at the laboratory, not to men 
tion scores of non- vegetable substances. 

Most of the specimens were of bamboo, the virtues of 
which Edison discovered by experimenting with fibre from 
a palm-leaf fan lying on one of the laboratory tables. In 
those days everything that met the eye was potential fila 
ment material. Edison placed the fan under a microscope, 
then had the bamboo cut into strips, carbonized, and 
mounted in test lamps. They proved to make better filaments 
than any he had yet developed. 

In the lamps for a second public exhibition, which Edison 
was planning, bamboo filaments were used exclusively. A 
complete system of incandescent lighting was to be tested 
on a large scale and its economy compared with that of gas. 
As an underground system it was to extend along the main 
streets of Menlo Park so as to provide light for most of the 

For the exhibition Edison's men dug trenches along the 
roads of Menlo Park. In them copper wire, without insula 
tion and encased only in wooden moulding, was laid. It was 
believed that a pressure of 100 volts would not require in 
sulation, but the first attempt to transmit current in this 

- . . 7.'i.< >v V-' 
^Nt ' -~ 


Arc lights, mounted on high towers, were expected to illuminate whole 
communities from one central spot in early attempts at street lighting. 


The factory at New Britain, Connecticut, where Thomson started his 
commercial career. 


fashion failed. Leakage of electricity blew out the whole cir 

Experiments in insulation were immediately begun. Coal 
tar, a composition of powdered slate, and wrappings of mus 
lin failed successively as non-conductors. When Wilson S. 
Howell of the Edison staff, after thorough study of formulas, 
offered to "cook" an insulating compound, he was installed 
with kettles and raw material in the chemical laboratory. 

More stench than success attended his first efforts. Dr. 
Otto A. Moses, chemist, after acute discomfort, gave up try 
ing to work. Howell not only smoked him out of his own 
laboratory, but earned the disgust of adjacent parts of the 
plant for his obnoxious "cooking." At length, however, the 
martyr brought forth a successful compound of refined Trin 
idad asphaltum, mixed with oxidized linseed oil, paraffin, 
and beeswax. 

Boys from neighboring farms were then rounded up, the 
conductors were placed on saw-horses and the boys wound 
on strips of muslin soaked in the compound. Each boy strad 
dled a conductor, walking along as he proceeded to wind. 
The tape was put on in three layers. 

It was Election Day, 1880, when Edison was told that the 
circuit was ready. A presidential campaign was just closing. 
Interested in the fortunes of James A. Garfield, Edison re 
solved to link the first trial of his system with the result of 
the election. 

"If Garfield is elected," he ordered, "light the circuit." 

For hours a telegraph key at the plant clicked off poll 

Edison in his residence waited as dusk changed into dark 
ness. Suddenly Menlo Park blazed with light as current was 
shot into many bamboo filaments. A political and an elec 
trical triumph were commemorated together. 


A delegation from the New York Board of Aldermen, 
which made the pilgrimage to Menlo Park late in December, 
1880, officially appraised the new inventions. The New York 
Herald, reporting the demonstration, observed that the al 
dermen wanted to form an opinion of the "resident wizard's 

"Stretching away on either side," the account went, "and 
intercepting the Park at intervals, ran long lines of light. 
There were illuminated spaces about all these gleaming 
points and the prospect was very beautiful as the visitors 
looked out upon it. 

"But at a sign from the Wizard, all changed. A workman's 
finger pressed the key and in an instant Menlo Park was in 
darkness. A ripple of applause involuntarily ran through the 
onlookers, but before it had subsided the finger was applied 
again, and the landscape was illuminated in a twinkling." 

The Newark Register recorded, on February 9, 1881, that 
"Menlo Park, the little hamlet where the great electro- 
scientist, Edison, holds communion with the hidden things 
of this world, was visited last evening by three hundred of 
Newark's best citizens." 

From the country-side farming folk poured into Menlo 
Park after nightfall to see the lights. One old fellow who 
drove in twenty miles was heard to say: "It's a pretty fair 
sight, but danged if I see how ye git the red-hot hairpin in 
the bottle." 

Multitudes that went to Menlo Park from New York night 
after night were eventually to see the "Edison lights" on 
their own streets. Edison and his counsellors had already 
planned the introduction of the incandescent lamp in Man 
hattan. On December 20, 1880 the Edison Electric Illumi 
nating Company of New York was chartered to operate on 
lower Manhattan, in an area termed by Edison the "First 


District" roughly a square mile included between Spruce 
and Ferry Streets and Peck Slip on the north, Nassau and 
Wall Streets, and the East River. 

Plans for installation were complete in essential detail, 
from a standardized lamp socket to the indispensable 
current-meter. It was one of few instances in electrical his 
tory of the invention by one man of an article for public use 
together with a perfected system for bringing it into accept 
ance. It was a Herculean task to plant this acorn of the 
electric light and power industry a far greater undertaking 
than the impatient public realized. 

Thus it was that there came first to Manhattan lights of 
the older type. In December, 1880, Brush lamps were in 
stalled along Broadway for three-quarters of a mile, the first 
electrical illumination of the famous street. The operating 
organization was the Brush Electric Light and Power Com 
pany of New York. This company also contracted to illumi 
nate Union and Madison Squares by means of masts 160 
feet high. 

The system was a spectacular success, and the New York 
Evening Post of December 17, 1880, tells of prospective cus 
tomers other than the city of New York: the Steinway Ware 
Rooms, the Park Theater, the Brunswick and Sturtevant Ho 
tels, and Koster and Bial's music hall. 

While sophisticated New York showed less excitement 
upon first seeing Brush lamps than did Wabash, Indiana, 
they were nevertheless well received. Nearly two years were 
to pass before the wizard of Menlo Park, with his bamboo 
filament and his ninety per cent efficient dynamo, would 
make good his promise to supplant the gas jet. 

During the winter of 1880-81, hundreds of persons wrote 
to ask the Edison Electric Light Company how soon incan 
descent lamps would be available. Hundreds of others so- 


licit ed agencies for the purchase of territorial rights. Had 
the company sold territory that winter, it could have en 
riched its treasury fabulously. But no one knew what prob 
lems would be encountered in putting the proposition on a 
commercial footing, and Edison himself had not obtained 
sufficient experience in the practical operation of his system; 
so the company refrained from engaging agents, who could 
not possibly have given satisfactory service to customers. 

A single agreement was made that winter with the Edison 
Electric Illuminating Company of New York regarding the 
famous "First District" where the Edison Electric Light 
Company was to make the installation. Edison would per 
sonally direct the work, conduct the necessary tests, start 
the central station machinery, and satisfy himself that the 
system was functioning properly. 

When Edison "personally directed the work," he was on 
the spot every day. He neglected sleep, and spent hours after 
the workmen had gone home going over the ground or 
solving minor problems. 

Meanwhile activities of a different character were devel 
oping downtown, where Edison had acquired a factory on 
Goerck Street for the establishment of the Edison Machine 
Works to manufacture the Edison dynamo. Miles of piping, 
or "tubing" were required to contain the underground con 
ductors; likewise shafting and pulleys, for connecting dy 
namos to the steam engines that drove them, lamp sockets, 
switches, meters, fuses, cutouts, brackets, junction-boxes, in 
struments, and electroliers. Most of these were Edison's in 
ventions, created in perfecting the system. As no existing 
manufacturers produced them, he set up plants of his own. 
"There was nothing that we could buy," he said, "or that 
anybody else could make for us." 

So, simultaneously with the Edison Machine Works, the 


An artist for Harper's Weekly pictures the laying of the Edison "tubes' 
or cables in the Citu's streets. 

Edison's first generating plant. 


Edison Tube Company was started on Washington Street, 
and the Edison Shafting Company on Goerck Street. Aux 
iliary appliances were produced by Bergmann and Com 
pany on Wooster Street, where Sigmund Bergmann, a for 
mer Edison employee, was proprietor. The manufacture of 
lamps by a force of forty-five men had been going on for 
six months at Harrison, New Jersey, at the rate of 700 a day. 
In the summer of 1882, among other new "hands," it gave 
employment to a youth named George F. Morrison, whose 
job was to unwrap and smooth tissue paper from lamp bulbs 
as they came from the glass works at Corning, New York, 
so they could be used again in packing finished lamps. His 
pay, if he worked sixty hours, was a dollar a week. This boy, 
undismayed by hard work, diligent and intelligent, was one 
of the future managers of the plant. 

As a "man of business" Edison showed himself coura 
geous as well as sagacious. He did not hesitate to manufac 
ture a lamp himself when the Edison Electric Light Com 
pany was reluctant to do so. He recognized that the com 
pany's policy was to hold patents, issue licenses, and promote 
commercial efforts, but not to manufacture. Accordingly on 
March 8, 1881, he executed a contract to sell the company 
all the lamps it needed for forty cents apiece, delivered in 
quantity, ready for commercial use, and with a manufac 
turer's guarantee as to qualify, the contract to run through 
the life of his patent. Each lamp cost Edison $1.10 to manu 

In the first year not quite 30,000 lamps were sold to the 
Edison Electric Light Company. Each year thereafter sales 
increased, and manufacturing costs lessened. In 1881 Edison 
produced lamps at seventy cents apiece; in 1882 at fifty 
cents; in 1883 at thirty-seven cents, realizing then a profit of 
three cents per lamp. The volume of sales that year wiped 


out deficits of the previous three years. In 1884, the cost 
dropped to twenty-two cents, and lamps were sold by the 
hundred thousand. The Edison Lamp Company was at last 
a profitable enterprise. 

All this while, there was going on at the Edison Machine 
Works a relentless search for a monster generating machine. 
The idea came to Edison as he studied the ten eight- 
horsepower dynamos which supplied current at the Menlo 
Park exhibition of 1880. A slow-speed engine drove all dy 
namos through a complicated system of belting and shafting. 
The power loss indicated considerable inefficiency. Hence 
Edison conceived the plan of substituting for the ten dy 
namos a single large-capacity unit with a direct-coupled, 
high-speed steam engine. 

No outsider saw the test, in the engine room at Menlo 
Park, of the experimental unit which preceded the first prac 
tical Jumbo. 

The steam engine used in that test was planned for a 
high speed of 600 rpm, and was to operate at a steam pres 
sure of 120 pounds per square inch. Heretofore the average 
speed of stationary engines was rarely more than 60 rpm 
and the average steam pressure ranged from 60 to 80 

It was a winter's night when the test took place. Charles 
T. Porter, who had designed and built the engine, opened 
the throttle a little at a time, his eye constantly on the gov 
ernor. It did not rise until the engine was making a racket 
that sounded "like an immense drop-forge foundry with ten 
thousand hammers in operation." The link motion was so 
rapid that it 'looked like a triangular-shaped bit of haze 
or fog." 

Edison related this experience with appreciative gusto: 


'We set the machine up in the old shop that stood on top 
of one of these New Jersey shale hills. We opened her up, 
and when she got to about 300 revolutions, the whole hill 
shook under her . . . 

"After a good deal of trouble we ran her up to 700 revolu 
tions. Every time the connecting-rod went up, she tried to 
lift the whole hill up with her." 

After this experience the engines finally built for driving 
the Jumbo dynamos were not run above 350 rpm. 

The first practical Jumbo dynamo was built at the Edison 
Machine Works in New York in 1881. Its size alone made 
visitors pause and stare. But they were chiefly puzzled by 
the seemingly endless operation of winding. They saw work 
men carry insulated copper wire around and around the long 
cylindrical cores, from one end to the other, with patient 
repetition; then start on a second layer and a third and a 
fourth, until six had been completed. The wire was wound 
around the core almost two thousand times, and there were 
six magnet cores. 

One further innovation was devised by Edison to meet the 
difficulty of a drop in voltage in the multiple circuit. While 
lamps nearest the dynamo received a full current of 110 
volts, the next lamps beyond received only 109 volts, the 
next 108, and so on. In the simplified system at Menlo Park 
such a defect was not serious, but when Edison began his 
layout for Manhattan, he perceived that, to correct a voltage 
drop, conductors would have to be thick at the generator 
end, and taper off gradually to a small diameter at the other 
end. To supply a large area by this means would have meant 
a prohibitive copper bill. Edison's tireless critics remarked 
that there was "not enough copper in the world" to permit 
of such a plan. 


Defeat seemed close at hand. But its very proximity pro 
vided the spur which led to the invention of the feeder 
system of current distribution. 

Instead of starting the parallel circuit direct from the 
dynamo, Edison interposed feeders, which ran between dy 
namo and central points in the parallel circuit, solely to sup 
ply current to the latter. This system was employed with 
complete success in the First District. 

A group of Jumbos was built for the District, which in 
the late summer of 1882 became a scene of activity prepara 
tory to actual operation. Edison was a tireless director, ex 
amining the work in the trenches where his tubing for the 
conductors was being laid, with insulating compound poured 
into every tube and junction-box. 

"I used to sleep on piles of pipes in the station," he said, 
"and I saw every box poured and every connection made on 
the whole job." A total of eighteen miles of street mains was 
put in. 

By August, 900 buildings had been wired and more than 
14,000 incandescent lamps placed in their sockets. No one 
knew, except Edison, when the station was to start gen 
erating. He told only his associates that it was to be Monday 
afternoon, September 4, 1882, at three o'clock. He wanted 
no publicity or excitement, for he was apprehensive of the 

He gave no appearance of this when he entered the dy 
namo room at Pearl Street about nine o'clock on Monday 
morning, attired in frock coat and tall derby neither of 
which went through the day unscathed. 

The day's work started prosaically. Through the forenoon 
Edison and his chief engineer gave the equipment a final 
inspection. Early in the afternoon their seclusion was inter- 


rupted by a tread of many feet, and a murmur of voices; 
newspaper reporters had learned, in the manner peculiar to 
their craft, that on this day the Edison light was at last to 
shine. They met no welcome, however. Edison, famous for 
his friendly smile, leaped upon them, and pushed the fore 
most bodily back to the door, exclaiming, "Get out! Get 
out I" He posted a man at the street door with orders to 
admit no one except the personal associates invited. 

Edison's motives were natural enough, for the gas com 
panies, in his own words, "were our bitter enemies in those 
days, keenly watching every move and ready to pounce 
upon us at the slightest failure. Big issues were at stake. 
Success meant a world-wide adoption of our central-station 
plan and its principles. Failure meant loss of money and 
prestige, and the setting back of our enterprise. . . . 

"All our apparatus, devices, and parts were home-devised 
and home-made. Our men were comparatively new, and of 
course without the slightest central station experience. What 
might happen when a big current was turned into the con 
ductors under the streets of New York, no one could say." 

A few weeks before, the system had been tested with 
curious results. A young man burst into the dynamo-room, 
crying excitedly, "Your electricity has got into the pavement 
up in Fulton Street and all the horses are dancing." 

When an Edison crew dashed to Fulton Street, they found 
the report was not exaggerated. The leak carried enough 
voltage to give horses setting hoofs upon a certain spot in 
the street a galvanic electric shock. Several days were re 
quired to locate and repair the faulty conductors. Small won 
der that Edison was so inhospitable to newspaper representa 
tives on the morning of September 4! 

There were scarcely twenty persons at the Pearl Street 


station when the great moment arrived. When Edison 
gave the signal, the connection was established and current 
flowed. The Edison system was working at last! 

Fifty-nine customers were served with current on that 
memorable day. Among them were the banking house of 
Drexel, Morgan and Company; the Herald and the Times 
offices; J. T. Pratt and Company; the Park Bank; and Sweet's 
Restaurant on Fulton Street. This was an eating place of 
great fame, frequented by Booth and Jefferson, and a stage 
coach terminal for uptown and Harlem. On the night of 
September 4, 1882 it was alive with excitement as were 
the other places where the wizard's lamps shone. 

"The giant dynamos," said the Times in recording the 
event "were started up at three in the afternoon, and accord 
ing to Mr. Edison they will go on forever, unless stopped by 
an earthquake." 

The system worked perfectly, except for a setback when 
a second engine and dynamo were started. The two engines 
seesawed, or "hunted," with a terrific racket. As Edison put 
it: "Of all the circuses since Adam was born, we had the 
worst!" He and Clarke spent two nights and a day in con 
tinuous session, trying to devise a remedy. They did not sur 
render until Edison had contrived a temporary solution, 
which permitted him to employ more than one generating 
unit at the same time. 

DESPITE THE SUCCESS of the incandescents, inventors and 
exploitors of the arc lamp systems were not to be downed. 
The Boston Journal of Commerce said in January 1882: 
"They (the incandescent lamp companies) claim to furnish 
light as cheaply as gas. But this is a subject that will require 
to be proved, as they also admit that they only furnish seven 
lights to a horsepower ... It costs less than a horsepower 


to furnish a single Brush light of 2000 candlepower . . . 
On this basis, neither the Brush Company nor the gas com 
panies have anything to fear from Edison." 

The American Electric people of New Britain also be 
lieved they could meet the incandescent lamp in eye-to-eye 
competition, with their more efficient arc lamp and their self- 
regulating dynamo. In fact, the Thomson-Houston system 
participated in such a tilt and came off the victorl 

In 1882 the P. and F. Corbin Company of New Britain 
went into the matter of electric lighting. They were ap 
proached by representatives of the Weston Electric Com 
pany, the Edison Company for Isolated Lighting, and the 
American Electric Company. The Corbin Company finally 
determined to give each system a week's trial. 

First to be tested was that of the Weston Company, in 
troducing an arc lamp developed by Edward Weston. One 
of the Corbin firm inquired: 

"Now that you have the lights going, can you turn any of 
them off?" 

"Not one at a time," was the reply. "They all turn off to 

The following week an Edison system was installed, with 
a bi-polar dynamo and a circuit of sixty incandescent lamps. 
Edison's system depended upon a voltage which had to be 
regulated constantly. In isolated plants, there was a pilot 
lamp on the headboard of the dynamo which allowed the 
dynamo tender to watch voltage variations. When necessary, 
voltage was adjusted, through a resistance box, which 
caused the voltage to increase or decrease at the will of the 

"Does a man have to be there all the time?" inquired one 
of the Corbin firm. 

"Yes," was the reply, "he has to regulate the voltage and 


adjust the commutator brushes as the load varies." The firm 
discovered also how fragile early Edison lamps were. 

When the Thomson-Houston system was tried Thomson 
and Rice personally installed a twelve-light arc dynamo, 
with its lamps. After starting the dynamo and assuring them 
selves that everything worked properly, they left the system 
to run itself. The Corbin Company found another advantage 
in the fact that it was possible to shut off some of the lights 
and leave others shining. 

At the end of the week the firm sent word to the American 
Electric Company: "Your check will be ready for you any 
time you call." The Edison system was not yet able to outdo 
its rivals. 

Meanwhile Thomson's laboratory continued to be the 
scene of incessant activity. In those days smooth, well- 
joined copper wire for winding armatures and field magnets 
was unknown. Lengths of wire were hard-soldered together 
with brass. Then the joints were roughly filed and cotton in 
sulation wrapped around the wire. Frequently slivers stuck 
through the insulation, or the joints were so irregular as to 
puncture the insulation and to leak current. 

Thomson demanded better joints of the wire manufac 
turer. "Don't solder them," he said. "Weld them and then 
draw the wire, instead of filing it." 

"But copper cannot be welded," exclaimed the wire manu 

"Yes it can," said Thomson, "if you'll use my method." 

Thomson explained the method of electric resistance 
welding which he had come upon unexpectedly during one 
of his lectures at the Franklin Institute four years before. 

While Thomson continued to tap new veins of knowledge, 
he realized that he was not receiving from the company the 
cooperation he had a right to expect. He was convinced that 



This cartoon, which was published in a popular magazine of the time, 

illustrates the ridicule which was directed at Edison while he was 

working on his incandescent lamp. 


The interior of a Thomson-Houston power plant in Boston, 1884. 


it was not marketing his inventions with "diligent and con 
tinuous attention," and "in all reasonable and advisable 
ways" as his contract demanded. He therefore tendered his 
stock in the American Electric Company, and that of Profes 
sor Houston, demanding the return of their patents on the 
ground that the company had not lived up to its agreement. 
It had no sales office outside of New Britain and lacked ade 
quate facilities for manufacture or experiment. Thomson's 
future was, for the moment, uncertain. 


The General Electric Succession 

ABOUT 1873 A FIRM under the name of C. A. Coffin and 
Company appeared in Lynn, Massachusetts, famous for its 
boots and shoes. Partners were Charles A. Coffin, a young 
man of twenty-nine, and Micajah P. Clough. They pros 
pered, for Coffin, though young, had a reputation for selling 
goods and designing styles. 

One morning he announced to Clough that he had joined 
a financial syndicate to buy up an electrical manufacturing 

"I'm putting some money into it, Clough," he said. "We 
can go in together or not, as you like." 

Though Clough did not join the syndicate, this step was 
the turning point of his partner's career. 

Until 1882 electric lighting had been little heard of in the 
shoe city. Yet a rumor was sufficiently enticing to interest 
the Lynn Grand Army Post, which was considering the il 
lumination of its new building. Silas A. Barton of the Post 
was chosen to investigate the possibilities of electric light. 

Having learned of an arc light exhibition by the American 
Electric Illuminating Company, Barton went to Boston, tak- 



ing with him a prominent leather manufacturer, Henry A. 
Pevear. Their questions regarding a dynamo system and its 
manufacturers met with evasive replies, probably because 
the Boston company was more interested in selling stock 
than in installing equipment. 

The visitors made their way surreptitiously to the base 
ment of the exhibition building. There they found a six-light 
dynamo. The nameplate, which the exhibiting company had 
forgotten to remove, read: "Manufactured by American 
Electric Company, New Britain, Conn." 

A day or so later Barton and Pevear got off a train at New 
Britain, engaged a station hackman to take them to the plant 
of the American Electric Company, and there met a young 
man by the name of Rice. The "electrician" he said, was out 
of town. Thomson was visiting several cities in an effort to 
interest capital in the electrical manufacturing business. 
While he searched for a rescuer, the rescuer came to him. 

As Barton and Pevear went through the New Britain fac 
tory they were impressed with the practical success of the 
arc light. They were impressed also by Rice's personality and 
technical knowledge. Their Yankee business minds sensed a 
beckoning opportunity. 

The Lynn men made tentative arrangements to secure a 
dynamo and lamps. Then they went home, called together 
some business friends and organized, on April 26, 1882, an 
arc-light central station company which they called the 
Lynn Electric Lighting Company. Electrical generating 
equipment was soon installed. When Thomson visited the 
plant, Barton was able to verify his understanding that the 
American Electric Company was in the market. 

Barton interviewed half a dozen shoe manufacturers in 
Lynn, and a syndicate was organized to buy control of the 
American Electric Company. Pevear joined the group on 


condition that Charles A. Coffin should also join and should 
agree to help actively in managing the unfamiliar enterprise. 
The syndicate made sure of the cooperation of Professors 
Thomson and Houston, holders of the patents. 

This time there was no quibbling over the contract with 


The first building occupied by this pioneer company in Lynn, 

the two professors. Mutual faith and cooperation persisted 
unbroken throughout succeeding years. The new company 
decided to reorganize at a meeting at New Britain on Feb 
ruary 12, 1883, under the name of Thomson-Houston Elec 
tric Company. 

Removal from New Britain to Lynn was not accomplished 
before the latter part of 1883, when a building was erected 
on Western Avenue consisting of three stories and a base 
ment ample, it was supposed, to accommodate the plant 
for a considerable period. Pevear thought of using some of 
the extra space for drying skins for his leather factory. But 


orders flowed in too rapidly for anything more to be heard 
about drying skins on the premises. 

Largely through Coffin's leadership, the young Thomson- 
Houston Electric Company began to obtain business in in 
creasing bulk. Dynamos and arc lamps rolled away to the 
freight yards in a constant stream. Shop hands increased 
from seventy-five to a hundred and fifty. On January 1, 1883, 
there were five central station plants whose Thomson- 
Houston equipment supplied current to 365 arc lamps. One 
year later thirty-one stations supplied 2400 arc lamps. 

At this time the company sold equipment to private in 
dividuals and concerns. It was not until 1884, under a policy 
suggested by Coffin, that the company began to establish 
central generating plants. Like Edison, Coffin foresaw that 
in this field lay the future of his company. 

Electrical science had its great inventors and technicians, 
but none of them were leaders of commerce. They were 
neither trained nor fitted to raise capital, or to finance the 
marketing of their inventions, particularly on a scale es 
sential in a country of tremendous size. Both Brush and Edi 
son for example worked out in detail their central station 
systems. Yet each of these men was glad to relinquish the 
commercial development of his ideas to others. 

And of the latter Coffin was the greatest. He it was who 
looked decades into the future and guided his policies to 
ward ultimate ends. His influence was undeniably the prin 
cipal factor in shaping a great development which began 
with the Thomson-Houston Company. Unsuspected by any 
one, a line of succession was being set up which consisted 
of two branches; and the roots of the second branch were 
just beginning to appear in the first of the Edison companies. 

Because of what it led to and because it was an unbroken 
chain of development, we have called it the General Elec- 


trie Succession. The term might have been applied when the 
Brush Electric Company or the Edison Electric Light Com 
pany were conceived. Both of these in due time were drawn 
into the succession. But the Thomson-Houston Company 
was the first in the succession in which policies and methods 
that were later to be perpetuated took form. 

And the Thomson-Houston Company meant Charles A. 
Coffin. For the former "shoemaker" had become a master 
mind in guiding America's progress in the age of electricity. 


Lighting Up with Incandescents 

IN THE SUMMER OF 1883 the Brush Electric Company be 
came interested in the American rights of an incandescent 
lamp invented by Joseph W. Swan, an Englishman. With 
the Swan lamp, Brush visualized an illumination system 
which would include both arc and incandescent lighting. He 
had completed a storage battery of cast-lead plates, which 
were pasted over with lead oxides and treated in an elec 
trolytic bath. Experimenting with the battery, he found that 
one horsepower, applied to a dynamo, would charge the bat 
tery sufficiently for it to operate ten Swan lamps for an hour. 

A number of these batteries Brush connected to an arc- 
light system, operated by a dynamo of large capacity. Dur 
ing the daytime he charged the batteries and at night op 
erated incandescent lamps from them. The lamps did not 
interfere with the arc-light system, as the batteries were 
equipped with an automatic "manipulator" which cut them 
into the circuit when they were to be recharged or cut them 
out as charging was completed. 

This worked so well, notwithstanding occasional trouble 
with the automatic manipulator, as to increase substantially 



the business of the Brush Electric Company for several 

Scarcely more than a year passed before another incandes 
cent lamp, backed by the Thomson-Houston Company, 
entered the incandescent field, as Thomson resolved to pro 
mote both arc and incandescent lamps and let the public de 
cide the issue. Where Brush used a storage battery to operate 
his arc and incandescent lamp circuit, Thomson used a "dis 
tributor box." This he designed for hooking up (anywhere 
in the arc-light circuit) a group of eight incandescent lamps, 
each requiring the same voltage but only one-eighth the cur 
rent of an arc lamp. The entire group absorbed the same 
current as the arc lamps, 9.6 amperes, thus preserving the 
constant-current characteristic of the system. 

Independent control of each lamp was secured through 
the distributor box, which contained eight resistance coils, 
one for each lamp. The coils, which were loops of wire of the 
same resistance as the lamp filaments, were disconnected 
while the incandescent lamps were in use; when one of the 
lamps went off, a resistance coil was switched into the cir 
cuit by a magnet to pass current in place of the lamp. 

For a lamp to operate in this system, the company secured 
a license from the Sawyer-Man Electric Company to manu 
facture and sell their incandescent lamp. A year or two later 
Thomson-Houston came out with its own dynamo and began 
to install complete incandescent systems, with both series 
and multiple distribution of the current. 

In adopting the Sawyer-Man lamp, the Thomson- 
Houston interests accepted the Sawyer-Man contention 
that they had invented a workable incandescent lamp before 
Edison had done so. Their lamp had a distinctive type of 
base, although it utilized a filament of carbonized bamboo 


A woman visitor inspects the Brush central station in Philadelphia, 




and embodied the three features which Edison had patented 
filament, vacuum, and glass container. 

When the Edison company refrained from bringing ac 
tion, various explanations were advanced to account for 
their passive attitude toward competitors. It is likely, how 
ever, that both Edison and his backers were too absorbed in 
the project of establishing their system to undertake patent 
litigation. They preferred to win against competition by 
commerical rather than by legal success. 

The first District was in full operation with five thousand 
lamps burning by May 1, 1883. Licenses for Edison com 
panies in other places had been issued. Edison had worked 
out a plan for installing his system in thinly-settled com 
munities by means of overhead wires strung on poles, call 
ing this idea the "village plant system." And he had con 
ceived brilliant improvement upon the parallel circuit. 

One morning he came to the Machine Works on Goerck 
Street with a rough diagram. "I want you to lay out an ex 
perimental line in the laboratory and connect some lamps 
to it like this, and put two dynamos in series on the circuit. 
Then test it out and give me a report of what happens." The 
sketch presented the rudiments of the three-wire system, 
one of the greatest single factors in the success of illumina 
tion. It was to place electrical service within the limit of 
every purse. Edison had developed it in an effort to save 
copper in the conductors of his system. The saving that it 
finally effected amounted to as much as 62 per cent. Elec 
trical systems of today could hardly operate without it. 

Four months after Edison found that the three-wire plan 
would work, William S. Andrews and Frank J. Sprague, who 
had come from London to join the Edison staff, left New 
York for Sunbury, Pennsylvania. A license for a local central- 


station had been issued to the town of Sunbury by the Edi 
son Electric Light Company. Andrews and Sprague were 
to install the lighting system for operation throughout the 
entire town by the evening of July 4. 

They arrived at Sunbury with forty-eight hours to work 
in and proceeded to hook up two "L" type bi-polar dynamos 
in the local generating station, then to wire the town for five 
hundred 16-candlepower lamps. From the station to the 
main circuits, feeders which embodied the three-wire plan 
were carried on poles instead of underground. Andrews and 
Sprague might have slept the second night had not Sprague 
in his enthusiasm given the dynamos a trial run. He burned 
out the engine bearings and with his partner spent most of 
the night scraping babbitt metal. 

Edison arrived at noon on the Fourth to inspect the plant. 
Half an hour before sunset he ordered the dynamos started. 
As the glass bulbs glowed, excited shouts arose in the town. 
At the City Hotel a noisy crowd gathered. There was an in 
cessant fusillade of firecrackers, and as evening drew on 
fireworks such as the little town could rarely afford were set 

Townsfolk asked Edison and his men all manner of ques 
tions, "What makes the hairpin red hot?" "How do you light 
it from a distance?" "Can you blow it out?" Almost dis 
tracted, Andrews complained that they were asking ques 
tions he couldn't answer. "This is a new business," Edison 
told him, "we must invent answers to go with our appli 

Andrews recalled the advice on a day later in the summer, 
when a severe thunderstorm broke. In the midst of it, a boy 
ran into the central station, yelling that the City Hotel was 
afire. Andrews hurried downtown, found a crowd standing 
about the hotel in the rain, but he could see no smoke. In- 


side a dozen people told him excitedly that they had seen 
sparks and flashes leaping from the electric light wires to 
the metal gas fixtures around which the wires were twisted. 

The wires were insulated only by a cotton wrapping 
painted with white lead. Static electricity, built up by the 
storm, had leaked through the insulation and shot along the 
metal fixtures. 

Although Andrews calmed the excited crowd the hotel 
proprietor sent for him next day and demanded the removal 
of the wires. 

"You may not realize it," said Andrews, "but your hotel 
was struck by lightning yesterday. If it hadn't been for us 
you'd be proprietor this morning of nothing but a heap of 
ashes. Those sparks were the lightning being shunted into 
the ground on our wires." 

"Well!" exclaimed the astonished proprietor, "if that's the 
case, well let the wires stay, of course." 

Andrews left Sunbury, not long after, to install another 
Edison three-wire system in Shamokin. And in the fall of 
1883 Sprague became identified with a much more preten 
tious installation. This was the Edison central station at 
Brockton, Massachusetts, where the three- wire system was 
for the first time put in by means of underground con 

The system, supplying about 150 lamps, was installed de 
spite considerable apprehension on the part of the public. 
Officials of the local gas company were particularly dubious. 
One of them remarked that if the lodge rooms were ever 
lighted by electricity he would stop going to the sessions. 
Timid citizens wrote to the newspapers, expressing the fear 
that life and property would no longer be safe. 

Within a year the number of lamps in circuit exceeded 
two thousand. In 1885 it was reported to Edison that only 


two customers had gone back to gas one other had gone 
so far back as to resume the burning of kerosene! 

Sprague was the electrician of this station, and served as 
Edison's representative. He was also acting as Edison's 
mathematician at that time, calculating the size and cost of 
conductors for projected central stations. Blueprints of pro 
posed systems were regularly sent to him at Brockton. 

This was one of Edison's concessions to the mathemati 
cians. His own methods were not mathematical. But he rea 
lized that Upton and Sprague could substitute precise, time- 
saving calculations for cumbersome, protracted experiments. 
The method in the matter of conductors had been first to 
construct in miniature an electric-light system for individual 
towns. By experimenting with the miniature system, dimen 
sions of the conductors were fixed upon. This method took 
time and money, and Edison was willing to allow mathe 
matics to simplify it. 

Sprague's calculations showed that Edison's original plan 
for main conductors and feeders could be improved. The 
mains were intended to be tapering in diameter; Sprague 
made them of uniform thickness throughout. The resistance 
in the feeders was intended to be alike in all cases; Sprague 
made the resistance low for heavy loads of current and high 
for normal loads. This proved to be efficient, and the cover 
ing patent was presented by Sprague to the Edison Electric 
Light Company. Sprague had joined Upton and Thomson 
in being among the first to show what the college-trained 
man could do in electrical engineering. 

Sprague now seized the opportunity to experiment with 
an electrical device he had long dreamed of. Ruthlessly he 
converted the little office of the Brockton station into a tem 
porary electrical workshop. The office had a luxurious tap 
estry carpet and lounge. Sprague soon had them littered 


with armature cores, magnet wire, jack knives, asphaltum 
varnish cans, japan and shellac bottles, paper insulation, and 
magnetic testing bells. Most of this bric-a-brac was stored 
under the lounge during the hours of mental activity de 
voted to the central station blueprints at which times the 
blueprints were spread on the tapestry carpet. 
* Not the most favorable conditions, one would say, for con 
triving anything so intricate as an electric motor. Yet the 
motor this young enthusiast finally put together was to lead 
the way for electricity across the boundaries of illumination 
and into the vast field on the other side of the fence the 
field of electric power. 

It has seemed a long time from Edison's promises in 1878 
regarding incandescent lighting to their fulfillment in 1882. 
But four years was a brief period in which to devise and in 
troduce on a commercial basis so far-reaching a system. It 
was his original purpose to introduce his light solely by 
means of the central-station method. Only after tremendous 
pressure had he acquiesced in the formation of the Edison 
Company for Isolated Lighting. Pursuing its purpose ag 
gressively, this company engaged lighting salesmen and sent 
them at an early date into conservative New England. Iso 
lated Edison plants came to be numerous in that region 
within a year. And the business activity which resulted 
furnished training and experience to one of the later apostles 
of electric power, Sidney B. Paine. 

As a youth of nineteen Paine entered a textile mill at Fall 
River, Massachusetts. After four years of learning this busi 
ness he became in 1881 a sales agent of the Edison Electric 
Light Company, and later was established in Boston as as 
sistant manager of the New England department of the Edi 
son Company for Isolated Lighting. 

During the next few years Paine went up and down New 


England, interesting textile mills in Edison lighting. He 
could meet the mill men on their own ground; he knew their 
business and how electric lights could best serve them. 
Among his first installations was the Hampden Mill of 
Holyoke, where 120 lights were put in circuit. Mill after mill 
became a customer. So energetic were Paine's efforts and so 
universal his success that in ten years not a textile factory of 
any size in New England lacked its isolated Edison plant. 

A great stir was always caused by the lights. The Lowell 
Morning Times proudly announced as its "big news" on Feb 
ruary 21, 1882, that the prominent Merrimack Mills were 
being equipped with 262 Edison lamps, one for each loom in 
the plant. 

Paine was constantly meeting incredulity and bedrock 
conservatism. One mill owner would not hear of lighting his 
mill by any such "new-fangled method." 

"All right," said Paine, "I won't say any more about it. 
But I have an hour till train time and I'd like to tell you 
how the electric light works." 

"Go ahead," said the mill man, "provided you don't try to 
tell me why I should light my mill with it." 

As Paine talked, the skeptic listened more and more in 
tently. At last he said in an apologetic tone, "How much 
would it cost to light my mill by this system?" 

And Paine walked out of the mill with another signed con 
tract in his pocket. 

But men were developing Edison central stations, just as 
Paine was cultivating the isolated lighting business. Ar 
thur R. Bush, trained both in engineering and commercial 
practices, aided many generating stations to take the first 
step in supplying current for the "red-hot hairpins" in their 
air-tight bottles. 

Bush had begun his electrical career on the vessels of the 


Old Colony Steamship Company, which operated the two 
sound steamers, Pilgrim and Providence, between New York 
and Fall River. He helped to wire the Providence, for which 
2000 lamps were required. Every stateroom had incandes 
cent lights, so that at night the vessels were a blaze of glory 
and soon were famed up and down the coast. 

At Rutland, Massachusetts, Bush helped to establish a 
small generating station. Realizing that the station attendant 
had little knowledge of things electrical, Bush stayed in the 
neighborhood for several days after the installation. Early 
one evening the attendant asked him to fill in a supper re 
lief. The attendant had scarcely gone when Bush discovered 
that the clutch mechanism connecting the jack-shaft to the 
dynamo driving pulley was out of order and the clutch was 
slipping. If the dynamo stopped, all the electric lights in 
town would go out. 

Bush knew that it would take two men to repair the 
mechanism without stopping the engine and he was alone. 
It was a case of leaping into the gap as best he could. He 
grasped the clutch lever with both hands and held on. He 
was able to keep the clutch in place by main force. The 
minutes passed; the strain increased. His muscles ached 
sorely, but nobody came. Desperately he stuck to the job. 
The tardy attendant finally arrived. And Bush, faint with 
exhaustion, discovered that he had held the clutch for an 
hour and a half. 

Until the late 'eighties there were more Edison isolated 
plants than Edison central-station plants. In 1883 there were 
154 isolated Edison plants in the country, employing nearly 
30,000 lamps. 

The Detroit Post in January 1883 published the following 
summary of the lighting situation in New York: 

"A year ago there were a dozen companies ready to prom- 


ise any city incandescent lights that is, ready to light in 
teriors with small shining bulbs like gas jets but at this 
moment Edison's company is the only one actively in the 
field. It not only occupies the field, it comes near to filling it. 

"The Times is lighted on the circuit of the first district. I 
went into the counting room and editorial room last night to 
inquire about it. They spoke warmly in its praise. 'The best 
artificial light I ever wrote by,' said the manager." 

The tide was running strong and there was little doubt 
as to its direction. 

** m 

Frank J. Sprague 

Charles J. Van Depoele 

E. M. Bentley W. K. Knight 



Lighting Up with Arcs 

PAYING SCANT ATTENTION to the prophet of incandescence, 
the pioneer sales agent of the arc light resourceful, rough- 
and-ready, versatile as a Jack-of-all-trades was going 
everywhere and accomplishing a great deal. He flourished 
from 1882 until the early nineties. And he was more than a 
salesman; he was a missionary and a promoter. 

He went into communities where electric light was known 
only by hearsay and where even intelligent folk had their 
superstitions about it. He won over men of means; interested 
them in financing an electric light company; helped them 
find property sites; persuaded the town government to grant 
an operating franchise, which he frequently drew up him 
self; assisted in the organization of the company; sold it the 
necessary equipment, personally drawing up the contract 
in proper legal form; and considered his task completed only 
upon the arrival of the construction man who took charge of 
the actual installation. 

The training that electrical manufacturers gave their field 
representatives was of the utmost importance in the conduct 
of their business. Young men educated at college in elec- 



tricity were hard to find. Technical schools had not yet be 
gun to offer special electrical courses. In physics courses 
then offered, emphasis was on the principle of static elec 

So electrical companies themselves had to teach the men 
to sell and install their equipment. The Thomson-Houston 
company lost no time in establishing a regular course of in 
struction. Men who completed this course were known as 
"experts" and were sent out either as salesmen or construc 
tion men. 

They took their work with profound earnestness, but they 
got plenty of fun out of it as well. They called themselves 
"expects" because they expected to do bigger things and re 
ceive bigger pay. The first "experts" were a rather nonde 
script lot. They were recruits from the "highways and 
hedges," and comprised machinists, steam fitters, carpenters, 
school teachers, and anybody else with a turn for the me 
chanical who was willing to pitch in. "We took agents and 
peddlers, prize fighters and preachers," as Coffin put it. 

Their first duties were menial. They took turns keeping 
the floor of the testing room clean. They fired the ovens 
where dynamo windings were dried, and carried coal the 
length of the shop. They operated machine tools, or labored 
at the workbenches. Eventually they wound up in the test 
ing room, where they had to measure and adjust arc-lengths 
in the lamps, paint dynamo rods, fit belts to the dynamos, 
and sweep up shop at the end of day. 

A good deal of fun came when they were taught to climb 
poles. As construction men, they did all the work of linemen 
today. Sometimes they became nervous when they reached 
the top of the pole and wrapped their arms around it des 
perately. This brought their bodies perpendicular, loosen 
ing the grip of their spurs. Down they would slide with 


hands and arms full of splinters. But there was no respite; it 
was "up and at 'em" again. 

Sometimes arc-lamp salesmen penetrated into frontier 
regions. The veteran expert of the Brush Company, 
Thomas E. Adams, later the joint inventor of the Brush- 
Adams lamp, spent months on the upper Mississippi, install 
ing and adjusting arc headlights on lake and river steamers 
engaged in logging operations day and night. The Brush 
arc headlight, with suitable reflector, was so popular that 
more than half the steamers were thus equipped, and a sales 
agent was kept busy throughout the logging season. 

On one of these trips Adams put up for the night at Lan 
sing, Iowa, where the lodging house was by no means luxu 
rious. In the dead of night there came a thundering rap on 
his door. The voice of the landlord bellowed, "Say, mister, 
are you an electrician?" 

Adams emerged in his night clothes to meet a couple of 
men from a river steamer whose lighting outfit had gone 
wrong. In response to their entreaties he got into his clothes, 
walked with them half a mile along a railroad track, and 
rowed in their skiff six miles down the river. 

He found that a novice who installed the headlight had 
adjusted it in such a manner that the carbons would not 
separate when current was applied. It took Adams about 
two minutes to set things right. 

The following season Adams went into Canada to install a 
Brush arc headlight for Gilmore and Company, a large con 
cern that was cutting timber in the wilderness. He took the 
apparatus into the woods by boat and pack. It was installed 
on a company steamer, and successfully tested. 

When Adams prepared to return, the Gilmore people ob 
jected. "If you go back," they said, "well have trouble with 
the light right away." Adams assured them that everything 


would be all right he couldn't take up a permanent resi 
dence in their midst. 

He roughed it for several weeks, getting back to the 
border and finally reaching Buffalo. There he made for the 
first barber shop he could find. It was filled with men who 
turned and stared at him. He looked like a wild man, with 
tangled hair and a thick beard, and his face, neck, and 
wrists were swollen from bites of the Canadian black fly. 

Months later the Brush offices received a letter from the 
Gilmore Company which said curtly, "Your man came up 
here, put your light on our boat and then abandoned it. We 
must confess, however, that it has given no trouble." 

Colorful legends have also been handed down of the 
breezy methods of R. T. McDonald, general manager of the 
Fort Wayne Electric Light Company. The chance which 
created this electrical plant in Fort Wayne has been re 
counted by John Kise, one of its earliest employees. 

Kise, who in 1881 was employed in the warehouse of 
Evans, McDonald & Company, overall manufacturers, made 
a week-end excursion to Cleveland in July. There he saw 
the dazzling light of Brush arc lamps, and was so attracted 
by them that he visited the works of the Brush Electric Com 
pany to learn more about electric lighting. 

When, some time later, Kise met two strangers in the din 
ing room of the old Aveline Hotel, in Fort Wayne, he could 
talk of nothing but electricity. He was astonished to learn 
that his table companions were themselves electrical inven 

"We are in Fort Wayne," they said, "trying to arrange a 
demonstration for an arc lamp we've designed, but no one 
seems interested. You can't sell patent rights without a dem 

"Have you tried my employer, R. T. McDonald?" asked 

, &***> Spa! 



This Bentley-Knight car got its power by contact with an electric con 
duit between the tracks. 



Frank J. Sprague and his demonstration electric car. It was here that 
Jay Gould tried to jump off when a safety fuse blew out. 


Kise. "He'll help you exhibit the lamp. We've a small steam 
engine that could run your dynamo. You come out to the 
factory tomorrow and I'll introduce you to Mr. McDonald." 

So James Jenney and his son Charles, who had invented 
an arc lamp in their home town of Ann Arbor, Michigan, 
were introduced to Ronald T. McDonald. The moment Mc 
Donald understood the nature of Jenney's invention, he saw 
its future possibilities. Not only did he offer the warehouse 
for an exhibition of the light, but with characteristic vigor 
he also rounded up his townsmen, invited them to the dem 
onstration, and then stood by to give Jenney the opportunity 
of proving his invention. 

McDonald soon interested other Fort Wayne men in the 
new lamp. After a brief interval of negotiations and con 
ferences, a new concern came into existence, on November 1, 
1881. It was known as the Fort Wayne "Jenney" Electric 
Light Company though legally Jenney's name was not in 
cluded in the title. McDonald went after business with un 
bounded energy, and was usually successful despite his er 
ratic methods. 

McDonald is said to have traveled sometimes in a private 
railroad coach. He would roll into town with a flourish, hire 
a local band, and start a parade to draw the crowd to a public 
hall. There McDonald addressed them on the wonders of 
electric light and the importance of a local company for the 
community. Before he left he usually had a company started, 
and tucked away in his pocket was an order for electrical 

Frequently his salesmanship outran his company's ability 
to meet contracts. But he had a remarkable construction 
man, W. H. Driftmeyer, and from all accounts Driftmeyer 
saved McDonald's bacon more than once. 

During some jobs Driftmeyer slept on the premises, work- 


ing every minute of his waking hours. On one occasion a 
pulley burst and struck him on the head. He couldn't afford 
to quit work, so a private hospital was rigged up for him in 
the power plant and he directed the installation from a cot, 
his head wound in bandages. 

While he was completing an installation in New London, 
he received a hurry order to go out to Paris, Illinois, where 
McDonald was in difficulties. Two dynamos had been in 
stalled with the understanding that they were to be self- 
regulating. When the company had found that they were 
not, payment amounting to $28,000 was held up. Could 
Driftmeyer build a regulator and get it on in time? He 
rushed home on a Friday night, worked until the following 
Tuesday, hardly stopping to sleep, and made a regulator to 
fit the dynamos. It worked. McDonald received the delayed 
payment and entertained the whole construction crew at a 
celebration dinner. 

In New York and the nearby territory, during this period, 
the Fuller- Wood Company and the compact, low-weighted 
dynamo of "Jmmry" Wood continued to find numerous cus 
tomers. Wood himself did not hesitate to canvass for pros 

By the middle eighties each of these companies was mar 
keting a dynamo of respectable capacity. The Brush dynamo 
operated forty arc lamps; the Thomson-Houston, twenty- 
eight; the Wood, twenty-five; the Jenny, thirty. The Edison 
bipolar incandescent dynamos at that time were made 
in four sizes: the "E" lighting fifteen lamps of sixteen- 
candlepower; the "Z" 60 lamps; the "IT 150; the "K" 250. 

Now as arc-light stations multiplied and incandescent sta 
tions began to appear, there came upon the scene a new 
type of experiment the electric streetcar. 


Electric Transportation, Motors, 
the Transmission of Power 


Broomstick Cars 

OLIVER WENDELL HOLMES, at the age of eighty, wrote a set 
of rollicking verses called, The Broomstick Train; or The 
Return of the Witches. He fancied that the curious pole on 
the roof of every electric streetcar, reaching up to the 
slender "cobweb" of wire above, was a witch's broomstick 
put to prosaic use. 

Thus in 1889 did a poet herald the innovation in methods 
of transportation. He discerned, as many did not, the spirit 
of conquest which it typified. 

"Look here!" he exclaimed in Over the Teacups, "There 
are crowds of people whirled through our streets on these 
new-fashioned cars, with their witch-broomsticks overhead 
if they don't come from Salem they ought to! and not 
more than one in a dozen of these fish-eyed bipeds thinks or 
cares a nickel's worth about the miracle which is wrought 
for their convenience . . . 

"We ought to go down on our knees when one of these 
mighty caravans, car after car, spins by us, under the mystic 
impulse which seems to know not whether its train is loaded 
or empty." 



To be introduced to the reading public by Oliver Wendell 
Holmes was a bit of fortune. Such lines in the poem as "On 
the rattling rail, by the broomstick train," must have made 
many persons feel more kindly toward the strange electric 
car. But for years after an indubitable triumph over techni 
cal difficulties the electric car struggled against popular 
prejudice. People who felt at home with the "red-hot hair 
pin" could not get accustomed to a vehicle moving through 
the streets with nothing to pull it. 

Men had been attempting to move locomotives or single 
cars by electricity for two generations. Thomas Davenport 
of Brandon, Vermont, in 1835 "ventured his little all" in 
striving to construct an electric motor, operating a minia 
ture railway with it; Robert Davison of Aberdeen, Scotland, 
drove a five-ton electric locomotive in 1839 between Edin 
burgh and Glasgow. Dr. Werner Siemens in 1879 operated a 
sizeable electric railway in Berlin, the first to be driven by 
current from a dynamo instead of storage batteries. 

And Thomas A. Edison built and operated two experi 
mental electric railways at Menlo Park in 1880 and in 1882. 
He employed reversed dynamos for his driving motors, as 
Siemens had done, conducting current from dynamos to 
motors and back again through the rails and wheels of the 

The first of these railways was a crude and ludicrous 
affair. The locomotive was a four-wheel species of flatcar 
with board seats and a medley of apparatus, from which two 
long brake-handles protruded. Motive power came from a 
"long-waisted Mary Ann" utilized as a motor and laid upon 
her side. 

With this queer locomotive Edison "rode to glory," or at 
least to newspaper fame, at a speed of twenty miles an hour. 


He hauled a train of two cars, one of which had a canvas 
awning and was fondly dubbed "Pullman." 

The second experimental railroad had two locomotives 
and was far more sophisticated. The track was two and a 

Edison's electric railway at Menlo Park, in 1880. 

half miles long and the locomotives were capable of forty 
miles an hour. 

Nothing tangible came of Edison's adventure into the 
realm of electric railroading, nor of the joint three-ton loco 
motive which he produced with Stephen D. Field in 1883 to 
appear under the name of "Judge" at the Chicago Exposition 
for Railway Appliances. The locomotive was supplied with 
current from a third rail and had a Weston dynamo for a 
motor. In two weeks the "Judge" had run 446 miles. But 
though it led to nothing practical, Edison's experimental 
electric railway set people thinking for the first time about 


electric traction upon the elevated roads of Manhattan. The 
New York Herald took up the idea and used the Edison rail 
way to create considerable agitation. But it was twenty years 
before any action was taken. 

The electric locomotive was never thought of as adaptable 
for street railways. And the "broomstick car" was finally 
evolved through long and arduous effort by a memorable 
group of American pioneers, among them Frank J. Sprague, 
Leo Daft, Charles J. Van Depoele, Edward M. Bentley, 
Walter H. Knight, and J. C. Henry. 

Sprague had worked out a motor in his luxurious office in 
Brockton. His development of the trolley pole was later to 
bring him into a dispute with Van Depoele. To follow the 
contribution of Bentley and Knight, however, carries us 
back to Cleveland and the Brush Electric Company, which 
furnished shop facilities for these two men who set out to 
master one of the most obdurate problems in the electrical 

"If you want to pull a car, you must put something in 
front to pull it," said a serious street-railway official in 1884. 
That was the way people talked when electrical men an 
nounced their plan of moving a car by means of a motor in 
stalled inside. 

Frank J. Sprague put the problem of the electric street 
car a little differently. He summed it up as follows: 

"On the car must be mounted one or more motors, weigh 
ing anywhere from 1000 to 3000 pounds, whose speed must 
be reduced ten or twelve times, while perfect freedom of 
axle movement and resilience of car springs must be pre 
served, as, with this additional weight and system of gear 
ing the car is driven over irregular tracks, around sharp 
curves and up heavy grades. It must, moreover, be handled 
by men ignorant of electrical technique; must be reliable, 


not apt to get out of order; and must have a reasonable de 
preciation, although exposed to the worst abuses of any 
class of machinery." 

The word "abuse" sums up the contrast between the 
streetcar motor and its brother, the stationary motor. The 
latter is started in the morning and runs all day at the same 
speed. But the streetcar motor is started and stopped hun 
dreds of times a day and is made to run at all speeds up to 
its capacity. It is constantly in the hands of strangers. 

It was a nerve-straining enterprise upon which Bentley 
and Knight embarked when they ran their first electric car 
on Garden Street, Cleveland, July 26, 1884. The Associated 
Press announced it as the "first electric railroad for public 
use in America." 

Unquestionably electrical apparatus was ready at that 
moment to venture into the domain of transportation. For 
several years the dynamo had been earning its salt, though 
its construction had become increasingly intricate. 

Electrical men had known for ten years what would hap 
pen if a dynamo, supplied with power by a steam engine, 
was connected to another dynamo in the same curcuit so 
that current generated in the first machine would flow to the 
second. They knew that the mysterious interaction between 
magnets and armature would cause the armature in the 
second dynamo to rotate with enough force to drive a 
machine shaft or car axle. Thus the mechanical power which 
originated in the steam engine would reappear in the second 
of the two dynamos called now a motor. This is the idea 
of dynamo reversibility, revealed as far back as the days of 
Pacinotti and Gramme. 

Means had been found, long before this, to magnetize 
field magnets in a direct-current dynamo by a connection of 
the armature coils with the field-windings. Variations in this 


practice resulted in the series-wound, shunt-wound, and 
compound-wound dynamos. 

Bentley and Knight understood the principle of the motor, 
but to apply it to propelling cars was not easy. The prob 
lem was three-sided. There was the design of the motor, the 
manner in which it was to be installed in the car, and the 
method of transmitting power from the motor armature to 
car axles. 

Bentley and Knight used as their source of power a thirty- 
horsepower, 500-volt constant-current dynamo of the Brush 
Electric Company, located at the car barns on Euclid Ave 
nue, Cleveland. Another Brush dynamo, series-wound and 
of smaller capacity, was used as a motor. It was suspended 
underneath the car body and connected to front and rear 
axles by belts of coiled wire. 

The attitude of the railway company toward the new 
fangled mode of locomotion seems to have been one of 
chilly tolerance. The electric car was assigned to the worst 
piece of track on the system and left to bob along as well as 
it could. 

Prejudice against overhead wires was so widespread that 
the inventors enclosed them in an underground wooden 
conduit which ran in the middle of the track. It had a con 
tinuous slot through which a contact device called a "plow" 
reached down and carried current for transmission from the 
wires to the motor. 

The backers of the East Cleveland Railway were not 
exempt from the general feeling of awe which the strange, 
horseless vehicle aroused. 

Said the Cleveland Herald when cars commenced to make 
regular runs: "It was amusing to watch the passengers who 
boarded the car. Some took the invention as a matter of 


course, while others, especially the ladies, evinced great 
curiosity ... An unexpected drawback is the fact that half 
the horses that pass the car are frightened by it. There is 
nothing unusual in the appearance of the car; it makes no 
noises besides the whirr of the motor . . . But even old 
plugs were frightened; and one passenger opined that the 
horses, jealous of a loss of business, had combined to express 
their disapproval of the invention." 

This was all very fine. But within three days the electric 
car encountered difficulties. The motor was supported at one 
end by a spring connection, and the springs started to break, 
with loud reports, at the rate of one an hour. When the wire 
belts began to give way, the car had to go to the repair shop. 

A new method of connection between motor and axles 
was tried. Friction wheels made of strawboard with a sys 
tem of link belts were employed. But the first time the car 
endeavored to move a heavy load the friction wheels slipped 
and developed flat spots. Finally the inventors adopted spur 
gearing, built up of paper to lessen the noise. This worked 
fairly well. But when two and then three cars were run in 
parallel operation electrically, with a fixed or constant cur 
rent, the dynamo balked. It was being overloaded to almost 
double its capacity. 

Despite these problems Bentley and Knight struggled to 
keep one or two cars running. They did so through the rest 
of the year and well into 1885, changing their methods to 
use constant voltage and varying current. The service 
proved sufficiently reliable during winter storms to give a 
hint of the superiority latent in electric traction. It was cer 
tain, however, that the conduit method of supplying cur 
rent was impractical, for it involved heavy expense. 

Bentley and Knight did not make a practical success of 


the electric street car. But they with Van Depoele and others 
started a ripple of expectation which lasted until the electric 
car seriously took up its task. 

Charles J. Van Depoele was early apprenticed to a cabinet 
maker in his native town of Lichtervelde, Belgium. Artistic 
in his tastes and skillful with his hands, he became a wood- 
carver of considerable reputation, until the mechanical and 
scientific side of his nature began to gain the upper hand. 
He hob-nobbed in the railway shops of East Flanders, where 
his father was a master mechanic. There he found an atmos 
phere to his liking, and there, despite his father's disap 
proval, he served another apprenticeship, this time to study 
elementary electricity. 

In 1869, at the age of twenty-three, Van Depoele came to 
the United States and established in Chicago a prosperous 
wood-carving business. Unexcelled at this trade, he was soon 
making more money than he needed and the surplus went 
promptly into a series of experiments with electric arc lights. 

In less than ten years he had an arc-light system of his 
own; had installed lamps in Chicago; and had lighted the 
front of an opera house in Detroit, where the strange glow 
of the first lights caused a nervous citizen to send in a fire 

Van Depoele was also experimenting in a modest way 
with motors, operating them on his series arc circuit. On one 
of these motors he placed a current-collecting brush made 
from a block of carbon, which served the purpose as well as 
the old copper brush. Without knowing it, he had made one 
of the most valuable discoveries of his life which was one 
day to play a dramatic role on streetcar development. 

By 1880 this erstwhile wood carver had made friends in 
financial circles, and his own manufacturing concern, the 
Van Depoele Electric Manufacturing Company of Chicago, 


was in the forefront of those trying to move the streetcar by 
electricity. Most notable of his early demonstrations was the 
mile-long railway which he operated in 1884 from the street 
car terminal in Toronto to the Toronto Annual Exhibition. 
The cars were made up in a train and drawn by a motorcar. 

The method conceived by Van Depoele for supplying 
current to the motor was an innovation. On the roof of the 
motorcar he mounted a pivoted beam with a contact wheel 
on its upper end. This wheel was placed against the under 
side of an overhead wire. At the opposite end of the beam 
was a powerful spring which pressed the beam, and hence 
the contact wheel, continuously against the wire. Connected 
to the wheel was a flexible cable which provided a path for 
the current from wire to motor. This was nothing more than 
the simple trolley pole, now universal. It was the most practi 
cal and workable method that had been tried. 

Upon applying for a patent a year or so later, Van Depoele 
found himself in conflict in the United States Patent Office 
with Frank J. Sprague, who laid claim to the invention of 
the under-running trolley as far back as 1882. He was at that 
time an officer in the United States Navy, on a year's leave 
of absence. While riding on the Metropolitan District Rail 
way in London, Sprague had pondered the notion of run 
ning the road by electricity, with a conductor supported on 
insulators between the tracks, and automatically kept in 
tension. But at the Kensington Station he saw such a formi 
dable complication of switches connecting the four tracks 
that the idea came to him of substituting an overhead con 
ductor and an upward-pressing contact mounted upon the 
roof of the car. 

The patent office refused to recognize Sprague's inventive 
work previous to his return to the United States in 1883, de 
spite the fact that part of the time he was on board an 


American vessel and hence on American territory. The rul 
ing gave priority to Van Depoele, to whom the patent was 

In November 1884 the Sprague Electric Railway and 
Motor Company was incorporated under an agreement 
which committed Sprague to experiment with motors both 
for traction and stationary work. It represented his major 
ambition and had caused his resignation from Edison's 
organization the previous April. 

Under the agreement, Edward H. Johnson, later president 
of the Edison Electric Light Company, assumed the financ 
ing of whatever work was undertaken. This was probably one 
of the most unusual concerns, as regards its finances, in the 
whole range of electrical activity. For months it existed 
mostly on paper. 

The company did little to attract public notice until the 
winter of 1885-86, when Sprague gave a demonstration be 
fore officials of the Manhattan Elevated Railway on a short 
stretch of track in New York. Beneath an ordinary flatcar 
mounted on an elevated-railway track, he placed two of his 
motors. The experiment was not without results despite the 
fact that, in working the controller too abruptly, the safety- 
fuse blew out with a startling flash. Jay Gould was one of 
those who tried to jump from the car, and never afterward 
could he be persuaded to take the slightest interest in elec 
tric traction. 

The results came when Sprague was approached a few 
days later by C. E. Chinnock, superintendent of the Pearl 
Street Edison Station, with an offer to purchase for $30,000 
one-sixth of Sprague's interest in the company. At the mo 
ment Sprague did not have enough ready cash to pay his 
boarding-house bill, yet he declined the offer. 


Sprague was away on a vacation when he received a 
telegram from Johnson announcing that he had promised to 
show Cyrus W. Field, chief owner of the Manhattan Ele 
vated Railway, an electric car in operation on his road the 
following Tuesday only four days distant. 

Sprague rushed to New York and set feverishly to work. 
Monday night found him with one assistant completing the 
electrical connections of the motors by candlelight. He had 
no opportunity to make tests and had no idea whatever 
whether the motor would work. 

The situation was tense. But the motors responded with 
out a fault. For two hours every possible feat was performed 
with the car, including regenerative braking, which seems 
to have been tried that day for the first time in America. 
Geared motors were mounted by the "wheelbarrow" method, 
later universal; interpole winding was used, as well as rheo- 
static control and other innovations. 

At the end of the day Sprague received another call from 
Chinnock, who offered $25,000 for one-twelfth of Sprague's 
interest, nearly double his first offer. And this time Sprague 

He was rather mystified about the proposition, however, 
for Chinnock was not a man of means. Later he learned that 
the Superintendent was acting as agent for an officer of the 
Edison Electric Light Company, who had been advised to 
buy into the Sprague Company by a spiritualistic medium! 

In less than three weeks another twelfth interest was sold 
for $26,500 to other parties, who doubtless bought on more 
materialistic advice. When officials suddenly lost all interest 
in the project, the Sprague Company had to abandon its 
attempts to electrify the elevated railway. It turned conse 
quently to the street-railway field, where half a dozen elec- 


trical inventors had small installations operating with just 
enough success to encourage the sanguine, just enough fail 
ure to arouse scoffers. 

With people beginning to talk of electricity as the emanci 
pator of the horse, it was high time for an emancipator to 
appear. Horses frequently collapsed in the traces from the 
heat. They could be worked only two or three hours a day. 
The streetcar stables were immense, and property surround 
ing them invariably depreciated in value. 

As William Wharton, Jr., put it: "Electric cars never get 
sick with epizooty or other diseases, and during strikes or 
other periods of enforced idleness do not require to be fed." 

But opposition to electric transit, continuing up to 1888 
and even later, revealed human motives at their worst. The 
backers of cable roads did everything they could to impede 
the entrance of electric cars. Superstitions of the people 
were played upon; the danger of electric wires was exag 
gerated; readers were gravely informed that the heavy cur 
rents passing over the wires would kill the shade trees; or 
that the presence of so much electricity in the air would 
cause sickness. 

Meanwhile electric light companies were coming more 
and more to be an important factor in the transit situation. 
The day was near when the Edison Electric Light Company 
and the Thomson-Houston Electric Company were to place 
themselves solidly behind electric traction. 




Construction gang of the Thomson-Houston Company engaged in 
erecting poles and stringing wires. 



From Shoes to Dynamos 

THERE WAS NO SPOT IN Lynn, Massachusetts, more indus 
trious than the factory of the Thomson-Houston Electric 
Company in 1885. But it was a mysterious business. Passers- 
by loitered as their ears caught the unfamiliar drone of dy 
namos. Through basement windows they beheld with star 
tled eyes the weird green flashes of commutator brushes and 
the steady blue gleam of arc lamps. 

By 1885 a broad, smoothly functioning organization was 
shaping. Executive minds were developing above the com 
mon level and gravitating into their proper spheres. Rice 
was made general superintendent, his career diverging at 
this point from his former close association with Professor 
Thomson. He held a combined engineering and plant- 
production office. Under him was an able assistant engineer, 
Albert L. Rohrer, and under them both was a fast-growing 
and efficient departmental organization. George E. Emmons, 
meanwhile, had become factory auditor from which he ad 
vanced by logical steps into the realm of factory manage 

The company was expanding from insignificance to first 



importance. Routine was in flux; minor departments were 
taking shape. There was one draftsman, W. O. Wakefield, 
and the small supply of blueprint paper he required was 

Everywhere there was teamwork. Rohrer was in the shop 
at seven in the morning, laying out work with the foreman 
and starting the production of new products as they were 
worked out in Thomson's laboratory, a definite procedure 
for manufacture having been decided upon in consultation 
with Rice. Thomson's was the true analytical mind. He per 
ceived difficulties before they were reached, and altered 
his plans accordingly. It was the antithesis of the "cut-and- 
try" practice, which demanded innumerable experiments, 
and was a conserver of time and effort. 

Sometimes he and Rice received helpful suggestions un 
expectedly from Charles A. Coffin, the man of finance and 
commerce. They would look at each other in surprise and 
confess that he had brought up a technical idea which de 
manded consideration. 

From intensive activity and the rising hum of industry 
the nucleus of a great corporation was emerging. Already 
the company contained rich human talent. It had a great 
scientist-inventor in Thomson, a keen engineer in Rice, a 
capable factory manager in Emmons, and a genius of finance 
and salesmanship in Coffin. 

One thing alone clouded its horizon the restriction of 
competitive patents. This was a problem in which the whole 
future of electrical development was inextricably bound up. 

Finance rather than patents absorbed Charles A. Coffin 
in the first two years of Thomson-Houston activity. It was a 
double-headed problem with a manufacturing side, and a 
public-service or operating side. The company did not sell 
direct to the public but to local electric lighting companies 


whose welfare meant the success or failure of the Thomson- 
Houston Company. 

Local companies must receive financial stimulus. Much 
depended on them; yet they were small and inexperienced, 
and their problems were considerable. The investment re 
quired for launching local companies often was prohibitive 
to the small capitalist. As an associate of Coffin's said in later 

"The business was new and presented problems which 
were substantially without precedent . . . People generally 
did not at all appreciate the need or the value of electric 
ity ... There were few who had the courage and the neces 
sary capital to buy and install apparatus. Customers did not 
exist; they had to be created." 

These matters filled Coffin's days with anxiety and his 
evenings with study. But the policy which he finally inau 
gurated was simple. 

He offered to sell electrical equipment for part payment 
in cash and the balance in securities of the local company. 
That was all, but it was enough to assist many local com 
panies in getting a start. It was a policy which resulted in 
many new Thomson-Houston customers. 

Twenty-eight local companies were equipped with 
Thomson-Houston apparatus during 1884, and forty-seven 
during 1885. Three thousand arc lamps were sold in 1884, 
and seven thousand in 1885. 

Meanwhile Coffin was working out a commercial organi 
zation. The young company needed salesmen men with 
youth, initiative, and brains. He found such men now and 
then among young leather salesmen who came into his office 
at C. A. Coffin and Company. Whenever one of these men 
made a favorable impression, Coffin talked electricity to him 
and offered a chance with Thomson-Houston. Between 1883 


and 1885 he drew an appreciable group of salesmen from 
shoes to dynamos, and the Thomson-Houston yearly class 
of "experts" swelled in proportion. 

Coffin realized about this time that he must give up either 
the manufacture of shoes or the manufacture of electrical 
equipment. He could not do justice to both. Should he stake 
everything upon the innovation of electrical development? 

In the midst of his doubts there came an unexpected prop 
osition. The American Electric and Illuminating Company 
of Boston made overtures to Coffin for the purchase of the 
Thomson-Houston Company for $300,000. 

Coffin's decision was quickly made. He informed them 
that he could not consider the offer and told his shoe partner 
that he was giving up shoe manufacturing to devote himself 
to electricity. From that decision he never turned back. 

About the beginning of 1885 Coffin asked one of his 
former shoe customers, J. R. McKee, if he would take over 
the managership of one of the Thomson-Houston commer 
cial departments. 

"But I don't even know what electricity is!" answered 

"That's just our trouble," Coffin replied with a laugh. 
"We've too many men who know what electricity is, or 
think they do. What we want now is somebody to care for 
the commercial side. Perhaps the less you know about what 
electricity is, the better!" 

Coffin felt that Thomson-Houston needed in the com 
mercial field men with reliable merchandising instinct, who 
were free from the conservatism which technical electricians 
would be apt to display in the selling field. Coffin was exer 
cising his talent for picking out individuals of ability. He 
had known McKee for years, since the time when as a young 
man McKee had gone farther west than any other salesman 




84 5700 

The Thomson-Houston float in the Li/nn parade. 



on the road. He believed that McKee had the energy and 
enthusiasm to take the electrical message wherever he could 
get a hearing. 

So McKee took charge of some of the supplemental activi 
ties carried on under Thomson-Houston auspices. He be 
came president of the Thomson- Van Depoele Electric Min 
ing Company, a subsidiary organized to sell an electric drill 
invented by Van Depoele, and an electric coal-cutting ma 
chine. He also headed the Thomson-Houston Motor Com 
pany, which began to compete with other electrical manu 
facturers in offering stationary motors in sizes ranging from 
1/2 to 110 horsepower. 

The Thomson- Van Depoele Company also embarked 
upon a campaign for equipping mines. A line of electrically 
driven hoists and pumps and a type of electric mine loco 
motive made their appearance. In 1889 the first installation 
of a mine railway was made at a colliery near Scranton. A 
forty-horsepower electric locomotive operated on a track 
having a gauge of thirty-six inches. 

Thousands of dollars were spent in endeavoring to de 
velop mining mechanisms, particularly electric drills and 
coal cutters. Young men just out of college, some from aristo 
cratic Boston families, were hired to operate coal-cutting 
machines in the mines. But eventually the line of electric 
coal-cutting apparatus had to be abandoned. The machines 
were not commercially feasible. 

The Thomson-Houston International Company was or 
ganized in 1884. Men had been sent abroad from the first 
and had found opportunities for trade as good as at home. 
The South American business became extensive. An arc- 
light station at Hammerfest, Norway, was long the most 
northerly electrical installation in the world. Products went 
to Egypt, Hawaii, Peru, Russia, Spain, and elsewhere. 


Thomson brought out his transformer for electric resist 
ance welding in 1885, thereby creating a new division of 
the electrical art. It was introduced commercially by the 
Thomson Welding Company, organized in 1888 for that 

In 1885 Thomson designed a direct-current dynamo for 
incandescent lamps. Thus the Thomson-Houston Company 
was prepared to compete with the Edison Electric Light 
Company in the field of incandescent lighting. Desperate 
rivalry and astounding expansion were close at hand in the 
commercial development of this little unit of light, which 
Edison could not keep in his own hands without a struggle. 

When the first glow-spots of incandescence shone out 
around the First District in lower Manhattan, the lamp was 
transformed from an experiment to a standard utility. Its 
welfare and its future became the concern of the factory as 
well as of the laboratory, and the saga of its commercial 
development had begun. 

Skeptics flourished, nonetheless, on through the middle 
eighties. Conservative Henry Morton, President of Stevens 
Institute of Technology, was not yet won over. In December 
1884 he dourly remarked: "Ten thousand mechanical en 
gineers are wanted to every ten electricians; and it would be 
a mistake for a very great number of young men to deter 
mine to devote themselves to electrical enterprises." In that 
very year 125,000 incandescent lamps had been sold, an 
increase of 55,000 over the previous yearl 

It was impossible to make lamps more than just so fast, 
however great the demand. Manufacturing processes at 
Edison's factory at Harrison, New Jersey, were mostly hand 
work. A hundred workmen produced only three hundred 
lamps a day. Expert glassblowers were necessary to blow 
the bulbs and other glass parts. The bamboo filament went 


through eight operations, all performed by hand, before it 
was ready for carbonizing. And it took five hours to exhaust 
the air from each bulb to produce the vacuum. 

Men sat for weary hours at workbenches handling bam 
boo. The cane, received in long strips, was slit with a knife 
over and over until the smallest possible diameter was ob 
tained. One piece of cane yielded five filament-strips, each 
of which was planed on one side before it was baked. Next 
it was finished in straight form, then bent into horseshoe 

The same laborious operations were in vogue at the 
Thomson-Houston Electric Company, where the Sawyer- 
Man lamps were manufactured. But here the filament 
went through an extra process of treating, or "flashing," the 
invention of William E. Sawyer. The filament was placed 
in a bulb containing rarefied hydrocarbon gas, such as gaso 
line vapor, and was heated by passing a current of electric 
ity through it. Heat decomposed the vapor and deposited 
a coating of graphite upon the filament. 

The filament produced by flashing was a better con 
ductor of electricity, and was practically uniform through 
out its length. It could be treated to the correct resistance to 
obtain proper voltage and efficiency, and treating increased 
its life. The incandescent lamps of the Thomson-Houston 
company were now better than Edison's, which gave them 
an advantage on the market. There were doubts among 
their own men, however, whether in using a bamboo fila 
ment instead of the original thread they might infringe on 
Edison patents. But, as Thomson put it, "We started out to 
make the best incandescent lamp we knew how." 

The Thomson-Houston Company was planning an output 
of 2000 lamps a week. Already the Edison Lamp Works 
was producing almost this many, and Edison was seeking to 


simplify his manufacturing processes so as to get costs below 
the forty cents per lamp for which he had contracted to sell 
them to the Edison Electric Light Company. He had in his 
employ an engineer, John W. Howell, whose ingenuity con 
tributed one of the most valuable manufacturing advances 
of the period. Howell reduced the time of exhausting air in 
the lamp bulbs from five hours to a bare thirty minutes. 

Toward the end of 1883 Edison, while watching pumps 
working on his lamps, noticed a bluish glow inside the glass 
globes. This glow kept up until a high vacuum was obtained, 
then disappeared. 

The glow had no effect upon the lamp, which gave its 
incandescent light as brightly as ever, but seemed merely 
incidental to the process of exhausting the air. Edison sus 
pected that it came from a current inside the loop of the fila 
ment, because the glow appeared only when electricity was 
sent through the filament, heating it so that the gases would 
be driven out. 

To see if he was correct, he inserted a wire inside the loop 
between the two "legs," connected it to one of the contacts 
outside the bulb through which current flowed to the fila 
ment, and placed a galvanometer between the terminal and 
the wire. The needle of the galvanometer showed a small 
current flowing between the legs of the filament. But it 
flowed only when the wire was connected to the positive 
side of the circuit. 

Edison had answered the first question about the bluish 
glow when he took out a patent based upon the "Edison 
Effect" in October 1884; however, he did not suspect how 
much more lay back of it. Electrical science was not suffi 
ciently advanced to explain the glow. Not until thirty years 
later did other minds make the "Edison Effect" the basis of 


Edison's inventive ingenuity, spurred on by a forty-cent 
contract, made several cost-reducing changes in his lamp. 
The lead-in wires were of platinum, because that metal could 
be sealed into the glass neck of the bulb in an airtight joint. 
But the cost of platinum was too high for Edison to reduce 
his costs below the forty-cent limit. 

He overcame the difficulty by placing the platinum-copper 
weld in the glass seal, benefiting the lead-in wires, which 
thus could be made appreciably shorter. Changes in the 
structure of the base aided still further the reduction in 
platinum, until finally each lamp contained only one-eighth 
of the amount used in 1881. 

Electric light companies were now springing up on every 
hand, scattering through communities, as lavishly as citizens 
would allow, the ''bottled sunlight" of incandescent or the 
"blue moons" of the arc. But a generating station able to 
operate day and night was then unknown and apparently 
unthought-of. Central stations started their dynamos shortly 
before sunset, ran on a reduced scale after midnight, and 
shut down entirely at dawn. For the most part they were 
places of night work only, their function being to supply 

Edison wrote to W. S. Andrews, under date of January 2, 
1884, on the general subject of central station operation: 

"I think the wages for men at the station should be as 
follows: Engineer, $65 per month, daylight to daylight 
running, starting, of course one-half hour before required in 
the evening. The meter man should be called manager, 
salary, $50 a month. Duties, taking meters, collecting, keep 
ing books and running electrical part from one-half hour 
before dusk until seventy-five per cent of the load goes 

"You should have these men on probation and subject to 


passing an examination by me. That will wake them up. Also 
I think it very essential that for several days steam should be 
raised in the morning, and if there is a spare engine at the 
station they should be practiced. We have found that the 
Brockton men were not sufficiently practical, hence if any 
thing went wrong they lost their heads." 

This last referred to an incident the previous autumn at 
Brockton, where the engine was run by a locomotive engi 
neer who was supposed to be fearless with machinery. The 
electric wiring in the station was somewhat crude, as was 
often the case in those days. The large conductors from the 
dynamos to the switchboard were laid in grooves cut in the 
wooden flooring and carried to a wooden partition, where 
they were connected to outgoing feeders. One quiet evening 
the insulation on these conductors gave way. There was a 
sudden flash and flames shot along the floor. The "fearless" 
engineer jumped out of the window, while the fireman hid 
behind the boiler. 

About this time there came a change in the policy of the 
Edison Machine Works as to the finished appearance of its 
dynamos. Seeking to keep down cost, little effort was made 
toward ornamentation. The castings were rough and indif 
ferently painted. One day there came a complaint from 
Baltimore something the matter with the Edison dynamos. 

An inspector went down. He found that the Baltimore 
dynamos were all right except for an accumulation of dirt 
due to pure neglect. And the neglect was due to the fact that 
the station owned several dynamos of another manufacturer, 
handsomely painted, neat and trim. They were the pride of 
the station attendant, who lavished all his time upon them 
at the expense of the Edison dynamos. The inspector's re 
port caused the officials back home to sit up and take notice. 

As central stations multiplied, so did inexperienced opera- 


tors. And instruments by which voltage and amperage could 
be measured were not yet to be had. At Pearl Street the 
voltage of the lamps was regulated by a homely, though 
ingenious, device, which was an arrangement of magnets by 
which a red lamp was lighted if the voltage was too high and 
a blue one if it was too low. New stations at first did not have 
even this crude voltmeter; yet upon the regulation of elec 
trical pressure the life of the lamps depended. 

It was small wonder that some stations boasted of good 
records for their lamps, while others had such poor records 
that the managers thought there was something wrong with 
the lamps. Complaints of this sort were heard at meetings 
of the Association of Edison Illuminating Companies, or 
ganized in 1884. 

A central station manager got up and declared the lamp 
was "rotten," that it could not last two hundred hours. 
Whereupon Wilson S. Howell, manager of the unusually 
successful station at New Brunswick, N. J., reported that 
very careful records had been kept of the life of their lamps, 
and the average had been 3000 hours. Whereupon the 
original speaker exclaimed: 

"Gentlemen, it is easy to account for this difference the 
poor quality of my lamps and the excellence of his. The man 
who runs the New Brunswick station has a brother in the 
lamp works!" 

Not long after this Edison selected the New Brunswick 
station for the first commercial test of a high grade lamp 
having a hydrocarbon bamboo filament. It was one of the 
first steps in increasing incandescent efficiency. 

To the public, however, efficiency seemed to appeal less 
than convenience. An old German shopkeeper who had arc 
lamps in his place tried to explain their failings to an elec 
trician of the arc-lamp company. The lamps had behaved 


strangely a few evenings before, and he threw up both 
hands and brought them down slowly, pantomiming their 

"I don't like dot ding," he solemnly remarked. "Why not 
you have dose schmall condensed lights what comes in 

The spectacular was not overlooked in exploiting "dose 
schmall condensed lights." Blase Manhattan was startled 
on the evening of October 13, 1884, to see a singular proces 
sion coming down aristocratic Fifth Avenue. 

Several hundred marching men moved in the form of a 
hollow square, each wearing a helmet surmounted by one 
of the little glow-bulbs. Inside the hollow square moved a 
portable steam engine and Edison dynamo; and the marchers 
carried a long rope containing a conductor from which cur 
rent was drawn through flexible wires passing inside their 
sleeves to the lamps on their hats. A marshal led the column 
on a war charger, bearing a baton tipped with a light. 

This was also the year of the "Edison darky," a negro at 
tendant at Edison exposition booths, whose clothing was 
wired to connect with an incandescent lamp on top of his 
helmet. The heels of his shoes were equipped with copper 
plates that had sharp points which pierced the carpet and 
made contact with hidden conductors. Many a timid lady 
was startled upon seeing a sudden glow of light appear on 
top of the darky's head just as he handed her an Edison 


The two buildings of the McQueen Locomotive Company, on the Mo 
hawk flats near Schenectady, to which Edison moved his machine 



The first Bentley-Knight "tram cars" upset the peace of Brooklyn's 



Onward with Volts 

TEN YEARS HAD NOW PASSED since Charles F. Brush built 
his first dynamo and Elihu Thomson started to design a one- 
light arc machine with which to astonish his audiences at 
the Franklin Institute. 

It had been a brilliant decade. At its beginning electric 
lighting was largely in the realm of the magical. Hard- 
headed business men smiled at the notion that it was practi 
cal, or that it could serve a utilitarian purpose. American 
industry, in the column marked "electrical," was an absolute 

But within a single decade what a transformation! The 
hum of dynamos was in the air. Factories, generating sta 
tions, lights, and horseless cars were proving what the blue 
spark could do. 

The growing array of essential minor apparatus was keep 
ing pace with the development of the basic processes and 
machines. Meters and switches had appeared. Edison's 
chemical meter in 1882 took its share of the day's work, 
measuring the current that each customer used. 

This meter was remarkable for its simplicity and operated 



entirely by electrolytic action. A fraction of the electric cur 
rent which entered the customer's premises was shunted 
through a pair of cells, each containing two zinc plates im 
mersed in a solution of zinc sulphate. The cells, connected 
in series, served as checks upon each other. Through the 
effect of the electric current zinc in proportion to the volume 
of the current was removed from the positive plate and 
deposited on the negative plate. From weighing the plates 
before and after use, the amount of current consumed could 
be calculated. 

The weighing was done with the utmost precision. Cen 
tral stations had a regular meter crew, which made the 
rounds with a horse and wagon. New meter plates were put 
in to replace old ones, which were tagged with the customer's 
name and carried back to the station for weighing. 

Experience proved this meter perfectly reliable. Mistakes, 
when they did occur, were usually caused by human fallibil 
ity, as was the case at Sunbury, where the meter in a large 
clothing store indicated at one time that $200-worth of cur 
rent had been consumed in a month. The meter-man was 
aghast. He lost sleep for two nights. Then suddenly the ex 
planation came to him. In order to fit the plates into the 
meter-box he had been obliged to clip off an inch or two of 
copper wire, forgetting that he had previously weighed the 
plates, wire and all. The loss in weight represented by two 
inches of wire had registered about $150 to the customer's 

Another humble device was the switch. Yet of what practi 
cal value was it to know how to produce electric current, 
efficiently and in volume, if the current could not be con 
trolled? How could an electric circuit operate daily unless it 
could be closed and opened at will? 

Before the end of the decade 1876-86, the switches, while 



still serving each its own dynamo, were grouped together on 
a single frame. In large stations these switchboards were 
formidable affairs, the multiplicity of plugs suggesting a 
gigantic game board. The early switchboards worked on the 
plug principle and were constructed of wood. If a poor 


contact occurred between plug and terminal, the frame was 
liable to take fire. 

With the incandescent lamp, the switch entered a new 
phase. Edison's system reversed the relation between am 
perage and voltage. Arc-lamp systems had a current which 
remained fixed or constant; but they had a changing voltage, 
increasing according to the number of lamps in circuit and 
going as high as 2000 volts. Edison employed a constant 
voltage, no higher than 110; but the operating circuits 
handled over 700 amperes of current. The switch now con- 


trolled an electric current beyond all previous experience. 
Edison developed perhaps the first type of knife-blade 
switch, certainly the first of large capacity. He was the first 
to grapple with the erratic arc in an electric switch. The 
voltaic arc in the arc lamp is useful, desirable and en 
couraged by the construction of the lamp; but the voltaic 
arc at almost any other point in an electrical system is 
harmful. Seven hundred amperes of current, suddenly inter 
rupted by a switch, was likely to cause such an arc, resulting 
in damage to the apparatus. Edison prevented this by ar 
ranging the knife-blade contacts of his switch so that the 
current passed through them in series, one contact after 

IN THE FIRST TEN YEARS of electrical growth, the forging of 
policies was hard put to catch up with the brilliancy of in 
vention. In various executive offices, at desk or council table, 
America's economic future was taking form. In the plain 
office of Edwin W. Rice, Jr., general superintendent of the 
Thomson-Houston Electric Company, were originated some 
historic policies, determined by him through sheer necessity. 
There were few industrial executives of that period who 
were so young in years, so discerning, or so prolific as Rice. 

It was a time of crowding emergencies and unremitting 
labor. Upon the young superintendent fell the task of ana 
lyzing a multitude of perplexing situations. He watched the 
extraordinary growth of the plant and studied the question 
of factory efficiency. And he developed methods so strategic 
and fundamental that they became steppingstones toward 
a day of huge production, methods that worked a revolution 
in industrial supervision. 

He divided plant operation along functional lines. A new 
staff of supervisory officials was created all of them familiar 


in present industry, but each one then of a fairly unknown 

There was a production manager, responsible for getting 
the manufactured product out on time; a mechanical super 
intendent in charge of mechanical work, new building con 
struction and factory discipline; a cost manager, with juris 
diction over cost accounting, bookkeeping, and pay-roll; a 
purchasing agent, directing the purchase of all raw materials, 
supplies, and equipment; and a technical superintendent, 
supervising engineering, research, and drafting. All reported 
directly to the factory manager. 

Then the young general superintendent introduced a 
second innovation in factory management, a new method of 
determining manufacturing costs in which scientific prin 
ciples superseded guesswork. The company set about ac 
curately calculating its overhead taxes, insurance, fuel, 
maintenance, depreciation and the variation of overhead 
as applied to different products of manufacture. To the 
actual cost of direct labor and materials were added the 
computed percentages of overhead. 

The first ten years made large and flourishing enterprises 
out of the Thomson-Houston and Edison companies and 
they also brought prosperity to the Brush Electric Company. 
Out of Cleveland had come in the latter part of the decade 
news of the first experiments with an electric industrial 

The promoters of this endeavor were Alfred and Eugene 
Cowles, two brothers interested in the production of alumi 
num by electric smelting. They contracted with Charles F. 
Brush for a dynamo to be used in their experiments, specify 
ing a constant-potential, direct-current unit, capable of pro 
ducing 300 amperes at 60 volts. 

Current was conducted to carbons mounted in a horizontal 


position in the interior of the furnace, which was a clay 
retort, and the experiments, begun on October 2, 1884, re 
sulted in the production on October 9 of the first alloy of 
aluminum obtained in the United States by this method. It 
is also recorded that on October 11 synthetic rubies and 
sapphires were produced in this furnace. 

The Cowle brothers now ordered a larger unit, which was 
built for them in 1886. This dynamo was one of the early 
"monsters." It was of the characteristic Brush open-coil type, 
and had a capacity of 3200 amperes at 80 volts, or 256 kilo 
watts. It was larger than Edison's Jumbo and next to the 
largest dynamo in the world. 

The Brush Company toward the end of 1887 embarked 
upon a new field to keep abreast of the times. Following the 
lead of Thomson-Houston, it entered the alternating-current 
field. A visitor to Cleveland would have been impressed by 
the Brush factory, one of the most pretentious of its day. It 
was a beehive of activity, for there were local Brush com 
panies in numerous cities, all of them licensees of the parent 

"Six years ago," wrote Charles Lever in 1885, "a few Brush 
arc lamps strung up along Broadway comprised nearly all 
the electric lighting New York could boast of ... Now 
there are nearly 3000 arc lamps in operation every night in 
New York and Brooklyn . . . These lights are supplied on 
the rental system . . . The usual charge is forty to forty-five 
cents per lamp per night . . . Cleveland, the home of 
Brush, is the largest manufacturing city for electric light ap 
paratus in the world." 

The Fort Wayne Electric Light Company had branched 
into the field of incandescents in 1885. R. T. McDonald, head 
of the concern, had induced a new electrical man, M. M. 
Slattery, to join his company and work up a system of in- 


candescence, using the alternating current. Within a short 
time Slattery's system an alternating-current dynamo, a 
transformer, and auxiliary apparatus was on the market. 
Customers were expected to purchase their lamps of other 
manufacturers thus did the Fort Wayne company avoid 
the danger of patent infringement. 

Wood, in the meantime, had pioneered in arc lighting by 
installing an outfit, despite competition, for lighting the 
Statue of Liberty on Bedloe's Island. Representatives of 
several arc-light systems interviewed General Stone prior to 
Wood's call. Wood found the General nourishing a notion to 
throw electric light upon the clouds so that reflected rays 
would illuminate the statue. 

"General," said Wood, "such a scheme as that would make 
you the laughing stock of the country. What you need for 
the job is direct lighting." 

The General sprang up with an outburst of emphatic lan 
guage. "By gad, sir," he cried, "you're the first one who has 
told me the truth. Those other rascals simply agreed with 
me, with all sorts of flatttery about the fine results I would 
get with my plan." 

Wood got the order. And there was no reason for laughter 
when the current was turned on and electric rays cast their 
brilliance upon the symbolic figure of Miss Liberty. 

THE SECOND DECADE of electrical development was inaugu 
rated by an event in the Berkshires which influenced the 
years that followed. This was the establishment on March 6, 
1886 of a local lighting system in Great Barrington, Massa 
chusetts, which was totally different from any preceding it. 
First it made use of alternating current and was the first 
such system to be commercially installed. Second, it used 
transformers which were of a particularly efficient type. The 


system was the work of a young man of twenty-eight, an 
accomplished electrical technician, William Stanley. 

Stanley was the first to develop commercially a practical 
transformer, although not the first to conceive it. He utilized 
the only efficient principle of design and the only efficient 
system of circuit connection. 

As far back as 1879, Thomson's dynamo fulfilled the ele 
mentary requirements of an alternating-current generator, 
and he had devised two induction coils in a working circuit 
that were perfect counterparts of the modern transformer. 
But he did not utilize alternating current, because of the 
greater simplicity of direct current. In 1885 he foresaw that 
the day of alternating current was near; and he set up an 
experimental system at the Thomson-Houston plant, sending 
current from Factory A to Factory B, and making use of 
transformers for so doing. This of course was not a com 
mercial installation. 

Stanley, meanwhile, was working in the same direction 
independently of Thomson, and what he did had in the be 
ginning no relation to the General Electric Succession. It 
was an outgrowth of his association with George Westing- 
house. The latter had founded, in 1884, the Westinghouse 
Electric and Manufacturing Company, which has become 
one of the large electrical manufacturing units of the present 

Westinghouse saw the possibilities of alternating current, 
and the need of such a device as the transformer to give the 
current utmost utility. He bought the American rights to the 
patents of a Frenchman, Gaulard, and an Englishman, 
Gibbs, who had worked out transformers. Then he com 
missioned Stanley to experiment with the transformers, 
whose principal drawback was in their series connection so 


Derailed cars were frequent enough occurrences on the Richmond road 
to insnire this nicture hu n. ma^nzine artist. 


that they were dependent upon each other in operation and 
could not work independently. 

Stanley's notable contributions were two. First, he con 
nected the transformers in parallel, giving independent 
operation, such as Edison obtained by connecting incandes 
cent lamps in parallel; and second, he made each trans 
former automatically regulate itself, thus permitting inde 
pendent control of the devices supplied by the transformer. 

Thomson in 1879 had already accomplished the same 
things. He, too, had connected induction coils in parallel and 
had designed those coils to be self -regulating. 

Stanley obtained a patent covering the method of design 
ing his transformer, rather than the method of connection in 
circuit. Thomson made application for a patent in November 
1885, but it was stubbornly contested by several opponents, 
one of whom was the Gaulard-Gibbs combination. It was 
nearly twenty years before Thomson's patent was granted, 
only to be declared invalid in its first court test. 

Thomson, Stanley, and Westinghouse all perceived that the 
transformer was the key to a new door in electrical progress. 
Up to this time men generated electric current with great 
success. They distributed current after it was generated and 
utilized it in ways that were expanding every year. But they 
could not transmit the low-voltage current employed for 
incandescent lighting farther than three miles. To send it 
over greater distances required heavy currents, hence large 
conductors and the incurring of a prohibitive expense for 
copper wire. 

The arc-light systems had an advantage due to the 
higher voltages used. Brush had used his forty-light dynamo 
to operate thirty lights at full brilliance ten miles away. This 
restriction of transmission was the serious limiting factor of 


Edison's system. An Edison generating station could not 
economically supply a territory greater than sixteen square 
miles. Transmission became feasible only when the trans 
former was made practical. 

In Stanley's historic installation at Great Barrington, cur 
rent was sent from his laboratory in an old rubber mill on 
the outskirts of town into the town proper, a distance of four 
thousand feet, the current being transformed from 500 volts 
to 3000 volts for transmission, and then reduced back to 500 
volts for use. 

Thomson did not place an alternating-current dynamo 
upon the market until the spring of 1887. His policy in one 
regard was distinct from that of Westinghouse and Stanley. 
He believed that high-voltage systems, which would ob 
viously grow up as alternating current came into extensive 
use, required protective devices to safeguard customers. In 
1885 he devised such protective equipment and incorporated 
it into his first commercial transformers. Until he had done 
this, he would not allow the Thomson-Houston Company 
to exploit the alternating current commercially. His feeling 
on this point was inflexible and won the endorsement of 
both Coffin and Rice. Thomson told his colleagues that he 
would not permit an alternating-current system to enter his 
own house without protection and therefore he would not 
sanction its use in the houses of others. 

Westinghouse and Stanley disagreed with Thomson. They 
recognized the latent danger in a high-voltage line, but they 
held that it was exaggerated. So they went ahead, while the 
Thomson-Houston Company refrained from entering the 
field though at no time did it condemn or criticize its 

Not so the Edison Electric Light Company, which assailed 
alternating current and kept up the attack against com- 


petitors that ventured to advocate it. At first the Company 
trained its guns upon the Westinghouse people. When the 
Thomson-Houston Company put a transformer in the field 
with a Thomson protective device it was also included in 
the attacks. Sales agents had no trouble convincing cus 
tomers that its transformer was now perfectly safe. But the 
Edison Company remained adamant toward the alternating 
current, regardless of whether protective devices were in use 
or not. 

In a printed pronouncement, "A Warning from the Edison 
Electric Light Company," it sought to show that the alternat 
ing current was a menace to life and limb. And this attitude 
persisted despite the obvious electrical advantages of uti 
lizing that kind of current. 

The clash of opinions was intensified by the decision of 
the New York State prison authorities to adopt electrocution 
for capital punishment and to employ alternating current 
for the purpose. That was ammunition for the direct-current 
artilleries! What could be more convincing, they demanded, 
than the official selection of this current as the most effica 
cious means of executing men? 

So there grew up this curious conflict known as the "War 
of the Currents." The Westinghouse group frequently struck 
back at the Edison camp, and these two became the princi 
pal antagonists. Coffin never permitted Thomson-Houston 
to participate in the skirmishing. He insisted that his men 
keep quiet. 

As a matter of fact there was never quite the protective 
advantage in the direct current that its protagonists sup 
posed. At equal voltages, alternating current, as Thomson 
said, is less deadly than direct. The danger in alternating 
current is that the primary coil in the step-down trans 
formers, from which circuits enter houses and offices, may 


accidentally come in contact with the secondary coil, al 
lowing the high voltage of the former to enter the secondary 
circuit. Should that occur, any person touching the second 
ary was likely to receive a serious shock. Under prevalent 
methods of manufacture and installation Thomson believed 
that the two coils might readily short circuit. 

To take care of such a contingency, he invented three 
methods of protection, each of them effective. The one 
finally adopted, because of maximum reliability and sim 
plicity, was to ground the secondary wires of the transformer 
with a metallic conductor. Once adopted, the Thomson- 
Houston Company did not hesitate to exploit the alternating 
current. A complete commercial system was planned at once, 
using Thomson alternators and transformers. The latter 
raised the current pressure for transmission to a thousand 
volts, which was for some years the standard Thomson- 
Houston transmission voltage. 


Destiny Comes to Town 

IN THE BRIEF STRETCH of her existence America has grown 
with unbelievable swiftness. It has been such a lusty growth 
that the present has almost lost sight of the past from which 
it developed. The "good old days" have disappeared with 
the inrush of machinery, inventions, innovations in a word, 
with the touch of progress. Transformations in individual 
communities have come about almost within a generation. 
And electricity has taken a prominent part in working these 
social changes. 

In the eastern reaches of the Mohawk Valley of New York 
State, along the course of that low-banked, quiet-moving 
stream, stands the city of Schenectady. In bygone times 
it was a peaceful, prosperous town of Dutch settlement 
and Dutch atmosphere. Washington Irving might have said 
of it, as he did of Tarrytown: "It is in such little retired 
Dutch valleys, found here and there embosomed in the great 
state of New York, that population, manners and customs 
remain fixed . . . They are like those little nooks of still 
water which border a rapid stream, where we may see the 

straw and bubble riding quietly at anchor, or slowly revolv- 



ing in their mimic harbor, undisturbed by the rush of the 
passing torrent." 

Schenectady was much like that in 1886. It had no large 
pretensions, no particular desire to alter its simple mode of 
life. The worthy citizens who traversed the grass-fringed 
cobblestones of elm-shaded State Street were content with 
their Dutch traditions and their unassuming place in the 
affairs of the nineteenth century. 

Small local industries supported the fourteen thousand 
inhabitants. Broomcorn in spreading abundance waved in 
the breezes up and down the river banks, for the town 
manufactured most of America's brooms and brushes. The 
one large-scale factory was the plant of the Schenectady 
Locomotive Works, colloquially known as the "big shop." 

No citizen of Schenectady in 1886 supposed that the com 
munity would be noted industrially other than as a place 
where locomotives were built. No more did the president of 
the Schenectady Locomotive Works, Charles G. Ellis, or his 
plant superintendent, Walter McQueen. These men had 
been associated for years. They had guided the growth of 
the works; they were fast friends. But they had a falling-out, 
and their quarrel, by an odd chance, swayed the whole 
future of their community. McQueen, with the assistance of 
local capital, organized the McQueen Locomotive Works 
and began the erection of two factory buildings. 

He secured as a site a broad expanse of level ground west 
of the town and along the river. Once it was called the "big 
flat," a famous Indian hunting ground. Now it offered many 
advantages for the industrious white man with his machines, 
his mechanical skill, his manufacturing genius. 

McQueen began to build his factory on this spot in 1885. 
He had two structures completed, except for the roofs, when 
his principal financial backers died, and the project was 



halted. The unfinished buildings stood in plain sight from 
the railroad as the spring of 1886 came in. 

In New York, meanwhile, another event occurred which 
bore upon Schenectady 's future. Edison dispatched agents 
in various directions to scout for suitable sites for a new 


The factory on Goerck Street, New York, shortly before it was 
moved to Schenectady. 

factory. One went to New Jersey, one to Pennsylvania, and 
a third to up-state New York. 

So it was that Destiny called to the Mohawk Valley one 
day in the spring of 1886 a man who felt an immediate inter 
est in such a spectacle as met his eye from the car-window of 
a train puffing across the "big flat," past the two lonely factory 
buildings on the outskirts of Schenectady. He was Harry M. 
Livor, who reported promptly to Edison. Edison went up to 
inspect the site in person and the location made a favorable 
impression. An appraisal of the property had placed its value 
at $45,000. Edison felt it was too high, and offered $37,500, 


declaring he would not pay a cent more. The owners were 
equally adamant. 

It would have been an impasse had not civic pride burned 
in the breasts of some of the townsfolk. These men wanted 
to see Edison's plant located in their community. They held 
a meeting, discussed ways and means, and determined at 
length to appeal to the merchants of the town to make up 
the difference between Edison's offer and the owners' price. 
The merchants of Schenectady were canvassed, and urged 
to consider what it would mean for the town to have the 
plant of the famous Edison. The necessary amount was 
raised, so that the owners, as well as Edison, were satisfied. 

The stage was set for Schenectady to emerge from her 
seclusion into industrial prominence among American cities. 

The shops immediately took up the task of turning out the 
dynamos needed by the numerous Edison generating sta 
tions. Within a few months, other buildings sprang up along 
side the original shops which McQueen had reared. 



The organizer of the Fort Wayne 
Electric Company, and the Com 
pany's factory in 1883. 


The Motor Marches On 

lighting, utilizing only the direct current, Brush, Thomson, 
and Wood, the great arc-light pioneers, ready to exploit the 
alternating current as well as the direct these four con 
stituted the great pioneers in the nation's electrical illuminat 
ing field. All of them had given some thought to electric 
power. But none of them, even as late as the middle eighties, 
had expressly designed an electric motor as such, nor made a 
business of developing electric motors for stationary work. 

Edison, when questioned by a journalist in 1884 as to the 
"transmission of electrical force," said: "That problem has 
been pretty well worked out. A young man named Sprague 
. . . has worked the matter up in a very remarkable way. 
His is the only true motor." 

Sprague had vigorously pursued a well-defined theory of 
electric power since his days as a member of Edison's com 
pany, when, in his office at Brockton, he had found time to 
work out his motor. After forming the Sprague Electric 
Railway and Motor Company he perfected the design of a 
constant-speed, non-sparking motor to be used on direct- 



current circuits, as well as a motor adapted particularly for 
street-railway work. 

Specimens of these types were displayed at the Philadel 
phia Electric Exposition in September, 1884. The largest was 
of six-horsepower capacity; the smallest, rated at one and a 
half horsepower, drove a weaving loom. These motors were 
so designed that they would run at a constant, uniform 
speed despite variations in load, with a fixed, non-sparking 
position for the commutator brushes. 

In the spring of 1885 endorsement of Sprague motors 
came from the Edison Electric Light Company, which sent 
out a circular to its local companies advising them that "A 
practical motor has been a want seriously felt in our sys 
tem . . . The Sprague motor is believed to meet ... all 
the exigencies of the case, and the Edison Electric Light 
Company feels it can safely recommend it to its licensees as 
the only practical and economic motor existing today. Our 
own company has no interest whatever in the Sprague 
company, and does not derive any benefit . . . through 
this agreement." 

The Sprague Company soon had sales agents in Philadel 
phia, New England, and elsewhere. A dozen motors were 
installed in textile factories at Lawrence, Massachusetts. 
The man who was perhaps most active in the field was 
George F. Steele, who served for several years as superin 
tendent of the New England Electric Company, eventually 
becoming president. Steele was an able salesman with a 
good product to sell. He made a reputation for the Sprague 
motors and for himself by introducing it into the Boston 
garment trade. 

Scores of small clothing factories and tailor shops in 
Boston used batteries of sewing machines operated labori 
ously by hand. Larger establishments were equipped with 


machine-tables, which had a shaft connecting all the sewing 
machines, so that they could be run by power. This device 
gave Steele his opportunity, for the Boston Edison Company 
had just opened up, making central-station power available. 
It needed only one or two Sprague motors, belted to the 
shafts of these machine-tables, to sell the idea to the trade. 


They eliminated the necessity of a steam engine on the 
premises and lessened the expense for power. Steele came 
into all the business he could handle; and as he visited the 
motors regularly, making any adjustments needed in their 
operation, he made staunch friends of his customers. The 
clothing men would take no motors except the Sprague. 

In December, 1886, the first 220-volt Sprague motor was 
installed in a building at Purchase and Pearl Streets, Boston, 
for the purpose of running a freight elevator. The motor was 
a fifteen-horsepower unit, connected by about three thou 
sand feet of copper wire to the three-wire system of the 
Boston Edison Company. 


The length of the line was a new factor at the time, and 
the size of copper wire had been incorrectly calculated for 
maintaining the voltage under load. Steele realized this 
the moment he looked at the wires, but it was too late to 
make changes, as the elevator was needed to move a new 
wheel concern to the sixth floor. A load of big wheels was 
immediately rolled into the elevator. When the attempt was 
made to start it, the voltage of the motor dropped to 170 
and the elevator could not be stirred. All the time fresh 
consignments of wheels kept arriving until the sidewalk was 
blocked with them. 

Steele, in desperation, changed the connections on the 
motor so that it would operate with the reduced voltage. But 
in doing so he had to subject the motor fields to such intense 
heat that the winding insulation began to boil and flow out 
through the motor frame. Continuous operation with over 
heated windings was impossible. Steele stood duty ten hours 
a day, in an unheated building, to start and stop the motor 
for every trip made by the elevator. Between trips he had to 
jump up and down to keep from freezing. 

But the wheels were moved in without further difficulty. 
Everyone was delighted. Steele accepted the compliments of 
the building owners with a frostbitten smile and was wise 
enough to keep his comments to himself. Within a few days 
he had another circuit run into the building so that the 
motor, supplied with proper voltage, could operate contin 
uously under normal conditions. 

At the end of 1886 the Sprague Company had 190 sta 
tionary motors installed and in use, and 80 more under con 
struction. In many cities in the East and Middle West, they 
served more than a hundred trades and industries. They 
drove boot and shoe machinery in Detroit and Boston; coffee 
mills in Elgin, Illinois, and Lancaster, Pennsylvania; emery 


wheels in Des Moines and Chicago; lathes in Chicago, 
Boston, and New York; and printing presses, ventilators, ice 
cream freezers, and various other mechanisms. The Chicago 
fire department; the Gold and Stock Telegraph Company of 
New York, where Edison started his career as mechanical 
repair man; Drexel, Morgan and Company; and the New 
York Stock Exchange installed Sprague motors. 

No meters had yet been installed on motor circuits, so that 
in most cases a flat-rate charge was made. In Boston, for 
example, a five-horsepower motor cost about $33 a month. 

The metering of motors began toward the end of 1886. 
Steele installed a metered motor for Bigelow and Kennard, 
jewelers, in October of that year, to operate a buffing lathe 
for polishing silverware; and this motor was destined to 
continue in operation for thirty years, staunchly exemplify 
ing by is own performance the tireless, muscle-saving power 
of the electrical handyman. 

But the difficulties the stationary motor had encountered 
were small compared with those at that moment besetting 
the traction motor. 


Tribulation on Wheels 

BENTLEY AND KNIGHT IN 1887 had abandoned Cleveland as 
a scene of operations and now had a contract with the street 
railway company at Woonsocket, Rhode Island. By agree 
ment, the Thomson-Houston company furnished the gen 
erators and motors. After consultation, a rugged fifteen- 
horsepower machine was adapted. It was planned for 400 
volts, and its speed was twice reduced, by gears and pinions, 
before the power passed into the car-axles; that is, it was a 
"double-reduction" motor. 

Rohrer, one of their engineers, was apprehensive lest a 
complicated rig would be required to shift the brushes of the 
motor when the direction of operation was reversed. During 
the tests, Knight acted as motorman on the front platform, 
while Rohrer stretched himself at full length on the floor of 
the car, trap door open, to observe the behavior of the 
motor. For some hours they ran the car back and forth, 
Rohrer shifting the brushes on the motor each time by hand. 
Finally he discovered a fixed position for the brushes where 
the electrical neutral point was identical for either direction, 
and he realized with considerable relief that a brush-shifting 



rig would not be needed. A laminated copper brush and a 
fixed yoke were accordingly manufactured and attached to 
the motor. 

In October, 1887, the Woonsocket road went into com 
mercial operation with one car. On the first day this car ran 
steadily for thirteen hours, carrying eleven hundred persons, 
"curiosity loads," as they were termed. The road had an ele 
vated conductor. 

The Bentley-Knight Electric Railway Company now ob 
tained a contract for equipping the road of the Observatory 
Hill Passenger Railway Company of Allegheny City, Penn 
sylvania. This was one of the most difficult propositions they 
could have accepted. The total ascent on Observatory Hill 
was a rise of 295 feet in a stretch of 4900 feet, an average 
grade of six per cent, with a maximum grade of twelve per 
cent located on a sharp curve. On the entire road there was 
scarcely a fifty-foot stretch where a car could stand without 
the aid of brakes and there were thirty-four curves. This was 
a tremendous undertaking for a car still in the experimental 

The line was four miles in length. It was decided that one- 
quarter of the distance should be of conduit construction 
and the remainder of the overhead system. The cost per mile 
of the conduit was $23,000; of the overhead wire, $4000. 
These figures indicate the reason why the overhead wire, 
except in densely populated centers, survived the conduit. 

So severe was the grade on Observatory Hill that the cars 
were given additional mechanical assistance in making the 
ascent. This consisted of a steel rack laid parallel to one 
rail, the teeth of which engaged a gear wheel mounted on 
the car axle. 

The Thomson-Houston men again went into conference 
over the type of motor to serve the road. They determined 


upon a voltage of 500, and placed two motors instead of one 
on each car. Generating current were four dynamos, built for 
250 volts each and operated as two units, two machines in 
series constituting a unit. 

The road opened in January 1888, and service was main 
tained for nineteen hours a day. The records show that on 
one occasion a single car carried forty passengers up the 
long grade at eight miles an hour. 

That seemed to indicate success. Yet a few months later 
the Thomson-Houston company received an urgent call for 
help. There was trouble with the commutators on the 
motors, and six cars were laid up for repairs. A few days 
later Rohrer, who had been sent out to adjust the difficulty, 
had his first contact with what proved the most vexatious 
difficulty in the history of the streetcar motor. 

"I well remember my sensations," he said, "when I stood 
at the end of the road looking up the street on which the 
underground conduit was laid, and saw two streaks of cop 
per dust on the paving. ... I donned my overalls, got busy 
with file and sandpaper, and after working the rest of that 
day, all night, and all the next day, I had four of the cars 
in service. After getting some sleep, I soon had the other 
two ready. Fortunately, I had carried with me a supply of 
new copper brushes, which I put on the motors. The old 
brushes had become cutting tools, the edges having melted 
from the severe sparking, and the commutators were a sight." 

The severity of streetcar service, repeated starts and 
stops, reversal of direction, all going on for nineteen hours 
a day, had been accompanied by flashing and sparking 
whenever these variations occurred. Every flash melted cop 
per from the brushes until their edges were so sharp that 
they ground copper off the commutators. 

It was said that there was a small fortune in copper de- 


A view in the wire-insulating department at Schenectady, about 1888. 


posited along the tracks of the Allegheny City street railway, 
and for a while express shipments of brushes left Lynn 
almost nightly. 

The day of triumph for the electric street railway seemed 
far in the future. Yet with all its present uncertainty, an 
enterprise far more audacious than this venture at Allegheny 
City was being planned. And the man to undertake it was 
Frank J. Sprague. 

Out of a clear sky his company was offered the contract 
to build an electric road at Richmond, Virginia. Sprague 
himself remarked years later, "We had little to show, but 
faith was strong and the contract was taken under terms, 
price, and guarantee easily placing it in the Icnave or fool' 
class, especially in view of the unprepared state of the Com 
pany to undertake work of such magnitude." Probably no 
one familiar with the circumstances would have dreamed 
that the enterprise, foolhardy though it seemed, meant the 
turn of street-railway fortune. 

Certain it is that no particular attention was drawn to 
Sprague's departure for the Virginian capital with his care 
fully designed motor, his boundless faith and little else. 

The astonishment of the darkies in Richmond gave 
Sprague his first taste of renown. His car eliminated animal 
power on the city lines. He was the "emancipator of the 
mule." An old negro, watching with bulging eyes the mule- 
less vehicle steadily climbing a long hill, cried out: 

"Fo' Gawd, what am de white folks a-gwine do nex'? 
Fust dey freed de darkey, an* now dey freed de mulel" 

Certain it is that nothing so "impossible," so beset by 
difficulty, discouragement, and tribulation was ever turned 
into a virtual success. Yet it is universally conceded that the 
birth of electric traction dates from the solution of the tech 
nical problems presented by Sprague's road at Richmond. 


It was Sprague's first attempt to set up a street railway 
system of any size. It was to be an entirely new road new 
construction throughout. It was to be city-wide, twelve miles 
in length, and the contract specified that forty cars were to 
be equipped and thirty were to be operated simultaneously. 
No electrical railway in America had yet operated more than 
a dozen cars at once. 

Electrification, under the contract, was to be completed 
in ninety days from the time work started, and the price to 
be paid was $110,000. The contract did not state that Rich 
mond has several hills, making the steep grades many and 
the curves sharp. Yet Sprague signed that contract without 
either sight or knowledge of Richmond or the route of the 
road, and knowing that not a foot of track had yet been laid. 

The construction syndicate did a poor job of roadbed 
construction. The track was laid with light-weight rails of 
inferior quality; the curves were built on a radius of thirty- 
three feet, some only twenty-four feet, and were not bound 
in any way; the track was not ballasted from one end to the 
other. All this added heavily to Sprague's task. And already 
he had enough to weigh him down. He needed advice, as 
sistants, time, and experiments. 

Sprague knew that he must have trained, reliable men at 
the scene of operations; as he put it, he needed "a man who 
has nerve and grit and coolness, one who, if he gets on the 
front of a car going seventy-five miles an hour and there is 
danger ahead, will stay there." So Oscar T. Crosby was per 
suaded to resign his commission in the Army engineer corps 
and join the new enterprise. Not long afterward Ensign S. 
Dana Greene offered his services. It was soon arranged that 
Greene should handle affairs at Richmond, while Crosby 
took charge of manufacturing details in New York. 

Experiments were begun with the utmost perseverance. 


Trolley poles were tried out, including an under-running 
wheel; motor-control received unending attention; motor- 
suspension beneath the car, brush construction, current 
supply, powerhouse equipment, absorbed everyone's 
thoughts. Cars were run up and down an experimental track 
in the yard of the Sprague company works in New York. 
.During one of the first attempts, before a reliable motor 
controller had been devised, the car got out of control and 
smashed through the side of the office. The only thing that 
stopped its wild career was the office safe, in which by an 
odd chance there lay the one asset of value possessed by 
the company the Richmond contract. 

Within a few weeks the motor was under control and the 
system began to take tangible shape. Sprague had heard that 
Van Depoele sent out his cars with a man riding on the 
roof to keep the trolley wheel from jumping the wires, and 
he was determined to avoid a similar arrangement. 

But overwork and loss of sleep lowered his physical vital 
ity, and without warning he came down with typhoid fever. 
It was the very moment when the organization needed to 
drive ahead at full speed. Yet for nine weeks during that 
crucial period the helmsman's hand was absent, while the 
complication of problems grew and difficulties multiplied 
with every day that passed. The entire burden of piloting 
the Sprague ship of fortune during those nine weeks rested 
upon the shoulders of the two young assistants, one trained 
in the Army and one in the Navy. 

When Sprague, convalescent from his sickness, went to 
Richmond late in 1887, the tracks had been laid and the 
overhead wire system was in place. He was now able to go 
over the route, and he watched with dire apprehension as the 
car approached a ten per cent grade. As he himself recalls 
that moment: 


"My heart fell within me, and I said, It is utterly impos 
sible for any car to climb that hill/ I felt that there were two 
things that would probably happen. First, the car would not 
ascend the grade, no matter how powerful the machines, for 
lack of track adhesion; and second, the machines were not 
powerful enough, even if the wheels would cling to the 

'We had built very light machines (motors) with one 
reduction of gearing, to which the street-railway practice is 
now gravitating, and the torsional effort of these machines, 
while great, was not sufficient for the duty now demanded 
of them. An eight per cent grade would strain them, a ten 
per cent grade would be fatal. . . . 

"Should we operate the particularly heavy grades with a 
cable, to be run by electric motors, in sunken pits under 
neath the track, and depend upon motors for the regular 
duty on the rest of the road? This seemed feasible if the cars 
could not by their own adhesion mount a ten per cent grade. 
On the other hand, if a car did have sufficient adhesion to 
mount this grade, it was plain that there must be a change 
in the machines. We must double the reduction of gearing, 
and this was a serious problem to face. . . . 

"We had on board a number of our employees and Gen 
eral Manager George Burt, who was in Richmond represent 
ing the syndicate's interests. ... A short distance from the 
end of the line we stopped in the middle of a sharp curve. 
Burt thought we could not get out of it. I said we would if 
I wrecked the machine, and out of it the car came. No 
more enthusiastic man, I think, was on the car that night 
than Burt himself, after seeing that exhibition of what a 
motor could do when pushed to an abnormal strain. . . . 

"Arriving at the foot of the hill we stopped and I said, 
'Burt, we won't make it!' He said, 'You will; if you can get 


out of a curve like that, you can go up the side of a wall/ 
I offered to bet him five dollars, but needless to say I was 
in hopes I would lose. If we succeeded in climbing the 
hill, I knew what would happen to the machines. But it was 
vital to know whether a self-propelled car could be made to 
go up that grade." 

It did go up the grade. Reaching the top, the highest point 
on the line, it came to a stop amid the cheers of an enthusi 
astic throng. But the motors were hotter than the outside of 
a furnace. 

Sprague was in no hurry to proceed. He hoped the motors 
would cool down after a brief delay. But the crowd was 
impatient, and he again applied the current. As the car 
stirred, Sprague felt a peculiar bucking movement and re 
alized that it was seriously disabled. One of the armatures 
was crossed, and the odor of burning insulation filled the air. 

Sprague called out loudly enough for all to hear, "Greene, 
there is some trouble with the circuits. You'd better go back 
to the carbarn and get some instruments so that we can 
locate it." 

Greene realized the situation at once. He took his time 
about getting the instruments. Sprague turned the lights off 
in the car, and the crowd, tired of watching a dark, motion 
less car, gradually dispersed. Greene finally arrived with 
four powerful mules, which corrected the trouble by hauling 
the car back to the shed. 

The long hill had been surmounted. But the problems left 
in the wake of this achievement seemed as insurmountable 
as the hill itself. A change in the gearing between motors 
and car axle was imperative. But time was practically up 
and the construction syndicate was clamoring for operation 
to start. 

Sprague went to Brown & Sharpe of Providence, and told 


the foreman, "We have forty cars and are under contract to 
run. If we fail, we're likely to go under. I have everything 
at stake. The electric railway is largely at stake. The road has 
to succeed. Would you put some of your men and material 
at my disposal until I recover the position we have lost?" 

The foreman said he would. In a few weeks the only 
practical mechanism for converting the single-gear reduction 
equipment into double reduction equipment was completed. 
Sprague had conceived the method himself. 

There were countless other problems to be solved. Be 
tween forty and fifty schemes of trolley construction were 
attempted before one of Sprague's draftsmen, Eugene Pom- 
mer, thought out the plan of the swivel trunnion mounted 
on a tripod. This supported a trolley pole with a retrieving 
line attached to it. It proved to be the happy solution and 
was installed on all the cars. 

The brushes on the motor commutators were as aggra 
vating to Sprague as they had been to Bentley and Knight 
at Allegheny City. Every shape and design of metal brush 
was tried. Flat brushes wore through, doubled over, and 
stuck to the commutators; solid and laminated brushes 
caught in the commutator bars or split and straddled half a 
dozen bars, causing crossed armatures and burnt-out fields. 
Tilting brushes would not work. Brushes of copper, bronze, 
or brass in various shapes, set on end and pressed down by 
springs, served for a while but finally gave out. At last a 
number of bars of brass about three-eighths of an inch square 
were installed and lasted tolerably well. But as the cars 
moved through the streets, a shower of brass particles fell 
at the rate of nine dollars worth a day! 

In October, 1887, experimental operation had been at 
tempted, and on February 2, 1888, in a drizzling rain, the 
road was opened for public service. People swarmed upon 


the cars. Service was satisfactory until cars began to stop 
and could not be budged. The new gears had developed a 
trick of locking. One of Sprague's workmen thought the 
trouble was lack of lubrication. And so it proved. 

During a sleet storm workmen had to get up on the roofs 
of the cars and knock the ice off the overhead wire by 
clugging it with a broom. 

Then the metal brushes began to ruin the commutators, 
and in turn the armatures. Interruptions became continuous. 
Motor fields had to be rewound and rewound. It was a cease 
less nightmare. But the road kept running, regardless of 
mental anguish and financial drain. Dismay stalked through 
the offices of the Sprague company in New York as reports of 
these troubles came in every mail. 

Somehow the cars continued to run. Mechanics sweated 
and electricians swore, and Sprague himself spent his time 
commuting between Richmond and New York. The public, 
unaware of the strain upon tempers behind the scenes, ex 
claimed over the electric cars whose fame became wide 

Sprague gradually increased the number of cars in service 
from ten to twenty, then thirty, and finally forty. "We felt 
that we owned the whole street, and the city as well," he 
said the first time thirty cars were run May 4. 

By the end of April nearly 6000 passengers were riding 
on the cars every day. There were delays and interruptions, 
but the public was tolerant and patient. The motormen were 
referred to as "motoneers." Sometimes a car ran off the 
poorly-laid rails and passengers and pedestrians alike would 
hoist it back on the track. 

Sprague's practice in motor installation on these cars in 
general adhered to the plans adopted for his Elevated Rail 
way experiment of the preceding year. He was the first man 


to sleeve the motor on the car axle at two separate points, 
and to support the free end of the motor from the car-body. 
This arrangement maintained parallelism of the gears under 
all conditions of car movement. 

In distributing current Sprague won success on the prin 
ciple of the parallel circuit. There was a working conductor, 
a main conductor, and a feeder. The first was suspended over 
the center of the car track; the two latter ran above the 
center of the street curb, parallel to the road. Current from 
the dynamos travelled at 450 volts over the feeder, which 
supplied the main conductor at four widely separated points. 
The main conductor in turn supplied the working conductor 
by connecting wires placed at intervals of five hundred feet. 
Each car was driven by two motors, originally of seven and 
a half horsepower each. 

Trouble with the metal brushes continued. A man was on 
duty all day in front of the carbarn to meet each car as 
it came in. The motors would be cut out of circuit in turn, 
and while the car was run backward and forward, the work 
man would smooth off the mutilated commutators with a 
file. This trouble with the brushes was far more serious than 
people realized. Not to solve it was to endanger the prospect 
of ultimate success. 

Relief for the Richmond road, and for many others as 
well, finally appeared in the fall of 1888. The nightmare was 
laid forever when Charles J. Van Depoele, then in the serv 
ice of the Thomson-Houston Electric Company, made one of 
the most brilliant suggestions in the history of electric street 

Early in January, 1888, Charles A. Coffin told Professor 
Thomson that he believed the Thomson-Houston company 
should undertake electric-railway development. Coffin's de 
cision to enter this new field came before the success of 

Coil winder and helper in the Schenectady plant. 


Taking fathers lunch to the Schenectady Works each noon used to be a 
small boy's chore. 

The searchlight tower erected on the summit in 1892. 


the Richmond road was consummated. He foresaw at an 
early period the great possibilities of electric traction in 
urban development. People could live in residential suburbs 
if quick transportation was at their disposal. Every city and 
town the country over could use electric streetcars. 

Coffin's faith was rising at the moment when that of others 
was declining. One magnate who had grown faint-hearted 
was Aaron K. Stiles, manager and chief financier of the Van 
Depoele Electric Manufacturing Company. The company 
had been doing an excellent business. Van Depoele himself 
was as optimistic and persevering as ever. But Stiles found 
the financing of the company far more difficult than he had 

The previous autumn Van Depoele had told the American 
Street Railway Association that the electric railway was 
"ready for the market." He had enumerated eight electric 
railways installed by his company, all of them running at 
that time, and averaging seven or eight cars in operation. 
The most notable railway was at Montgomery, Alabama, a 
road thirteen miles long, with twelve cars running simulta 

But Stiles had not sufficient resources upon which to draw, 
and he knew he would have to look elsewhere for financial 
backing. He had already approached Sprague in the matter 
at the very time when Sprague was swamped with activities 
at Richmond. But Sprague, too absorbed in his own prob 
lems to undertake new ones, rejected the offer. 

So it happened that Coffin purchased the Van Depoele 
Company and brought it into the Thomson-Houston com 
pany, which acquired the rights to Van Depoele's patents, 
and added Van Depoele himself to its staff of engineers. 

Terms of sale were largely on a royalty basis. Stiles was 
to receive a royalty of $50 for every streetcar equipped 


under Van Depoele's company patents; and Van Depoele 
received a royalty of $5 per car in addition to his salary. 
Before a year had passed Stiles was satisfied that electric 
streetcars would never amount to much, and asked for a 
cash settlement of his interests in lieu of further royalties. 
Years after his interest was bought, enough cars had been 
equipped to make him a wealthy man had he retained the 
old royalty basis. 

In the month of March, 1888, the month of the great bliz 
zard, two men were added to the Thomson-Houston organi 
zation. One was William J. Clark; the other was Eugene Grif 
fin. Griffin had made an extensive trip through the United 
States, gathering data on the electric-light business, and inci 
dentally on the electric street railway. 

Now began a new and busy era in Thomson-Houston 
affairs. The company was committed to the great enterprise 
of electric traction. Yet the disquieting situation centering 
around the metal brush was unsolved. Until that spectre 
was laid, the development of electric traction could not 

Van Depoele had his own ideas. He had been placing the 
motor on the front platform of the car. This arrangement 
required a sprocket chain to convey the driving power from 
the motor to the car axle which Thomson and Rice at once 
pronounced a clumsy method. Van Depoele's reason for so 
placing the motor was that the motorman could shift the 
brushes by hand whenever serious sparking occurred, and 
especially when the direction of rotation was reversed. 

But the plan was abandoned in the system which the 
company now sought to design. Hoping to dispose of the 
metal brush dilemma in some other way, Thomson and Rice 
put the motor back under the car and equipped it with the 
best type of metal brushes they could make. 


Then they set about electrifying their first streetcar line, 
a modest road a mile and a half long, at Crescent Beach, a 
coast pleasure resort between Boston and Lynn. 

On the evening of July 3, 1888, a party of five men regis 
tered at Young's Hotel in Boston. Three of them were di 
rectors of the newly organized electric railway company of 
Des Moines, Iowa. The other two were Thomson-Houston 
sales agents, Theodore P. Bailey and William J. Clark. 

The clerk on duty handed Clark a note which he and 
Bailey read with dismay. They managed to conceal their 
feelings from their three prospective customers through din 
ner, and made an appointment with them for the next 

Alone together, Bailey and Clark exchanged looks of con 
sternation. Their guests had come east expressly to see the 
first Thomson-Houston road at Crescent Beach, where they 
expected to be taken the next day. But the note announced 
briefly that the motors on both cars at Crescent Beach had 
burned out and the cars were not yet back in service. 

No need to ask the cause of the trouble. The copper 
brushes were again working havoc, and the cars might not 
be running for two or three days. What should they do with 
their prospects? 

The next day was a holiday and Clark and Bailey sug 
gested a sight-seeing trip for their guests before starting for 
Crescent Beach. First they visited Nantasket Beach, which 
proved so delightful that they missed their return boat and 
had to wait two hours for the next one. 

At lunch the service was painfully slow, although the 
westerners were fortunately unaware of the covert signals 
passing between Clark and the waiter. After lunch the visi 
tors were persuaded to go out to Bunker Hill, which they 
might not have another opportunity of seeing. On their re- 


turn the carriage broke down, and the driver devoted him 
self endlessly to searching for a lost bolt in the road, while 
the close of day drew on. 

When they finally got back to Boston, it was too late to 
visit Crescent Beach, and the unsuspecting Des Moines di 
rectors agreed to postpone the trip until morning. Returning 
to their hotel, Clark and Bailey were greeted by another 
note. It said, simply: 

"Motors replaced. Crescent Beach road running beauti 
fully." It had been delivered early that morning! 

They lost no time in getting to Crescent Beach the fol 
lowing day. The cars behaved perfectly on every trip. The 
Des Moines men were greatly impressed, and signed a con 
tract with Thomson-Houston in due course. 

It may be surmised that the success obtained at Crescent 
Beach was hardly more than superficial. But it was no time 
for giving up, and before long the company signed a con 
tract for equipping the Eckington and Soldiers' Home Rail 
way at Washington, D. C. 

No sooner were preparations under way on this work than 
the United States Senate reported a resolution revoking the 
permit previously granted to the railway company to erect 
overhead wires in the city of Washington. The railway 
company, however, was granted a public hearing before the 
resolution was reported. 

Documentary evidence, amassed in considerable bulk, 
was laid before the committee. Several street-railway offi 
cials sent letters asserting in positive terms that overhead 
wires were not dangerous and that 500 volts, such as these 
systems employed, and not yet caused a fatal accident. 

The argument of the Thomson-Houston counsel con 
cluded with the following excellent logic: 

"No doubt there is danger in electric wires, but the danger 

A portrait taken shortly after his arrival in America. 


to people generally in the city is extremely small. The danger 
in crossing the streets is thousands of times as great. The 
danger on the railroad train, on the steamer, in eating your 
lunch or drinking soda-water, is far greater; you do these 
things without a care for the consequences. The other day 
I read of two or three fatalities caused by falling bricks 
blown from a chimney, but the paper said nothing about a 
proclamation by the mayor ordering all chimneys taken in." 

The committee of the Senate was not proof against such a 
strong defense, and the resolution was not passed. 

On October 17 the road was opened and created favorable 
comment because of its neat and ornamental appearance. 
At the same time the hearing before the Senate committee 
and particularly the endorsement of the overhead wire sys 
tem had given electric street railways excellent publicity. 

By the fall of 1888 the Thomson-Houston company had 
equipped several electric railways. The work was done with 
the utmost care, and under Coffin's policy the company gave 
service along with the sale of the products. Hence there were 
continual reports on the behavior of the motors, continual 
supervision by the experts over the commutator brushes. 

An alarming situation was created both by unexpectedly 
heavy operating loads placed upon the street-railway motors 
and by the neglect which the motors had to undergo. Spark 
ing at the commutators became so troublesome and the effi 
cient operation of roads so difficult that something had to 
be done before the company was financially ruined. 

Thomson immediately called a conference. Everyone 
agreed there was desperate need for a remedy. But no one 
appeared to know what could be done. Finally Van Depoele 
spoke in his quick way: 

"Why don't you try carbon for a brush?" 

Rice exclaimed, "Carbon has a resistance a thousand times 


greater than that of copper. It would be impossible to carry 
the current from the commutator." 

Van Depoele was unperturbed. "I tried carbon brushes on 
a motor in Chicago once, and they worked pretty well." 

"What did you try them on?" persisted Rice. 

"On a small motor operating on a series arc circuit." 

"Well, they might work for a current of ten amperes and a 
small motor for one-quarter or one-half horsepower, but 
they wouldn't be likely to work on the fifty or sixty amperes 
current which our railway motors use. But we can try them." 

The trial was made. Van Depoele and Rohrer fitted the 
extemporized carbon brushes to one of the standard F-30 
motors and gave it a test. Rohrer was delegated to inspect 
the motor every half hour. Before long he reported with con 
siderable excitement that the commutator showed not the 
slightest sign of sparking and was taking on a smooth, glazed 

A severer test was immediately conducted, with a fairly 
heavy load. Still no sparking, not even when the direction 
of rotation was reversed. 

The good news was soon all over the plant. One of the 
most disturbing problems ever encountered had been ban 
ished. The spectre of the metallic brush was laidl 

Rice communicated with the carbon manufacturers and 
before many days passed, carbon brushes had been placed 
on a Thomson-Houston railway motor in Woonsocket. In 
due time a report was made that the motor had run 4000 
miles without the renewal of either brush or commutator, 
an incredible distance in those days. The report added that 
the brush showed hardly any wear and that the commutator 
had taken on a splendid polish. 

Those were days of rejoicing in the Thomson-Houston 
organization. Congratulations were bestowed upon the ami- 


able, quick-spoken young Belgian, who had remembered at 
the crucial moment about his block of carbon. 

Only a few weeks slipped by before the news of the 
carbon brush went among other electric railway inventors. 
Sprague heard of it almost at once and seized upon it for 
his widely heralded, but sadly harassed, road at Richmond. 
The carbon brush saved the day for that famous under 

Thomson said of the brush, "It is the most important 
invention ever made in the electric-railway field." It seems 
almost incredible that on this unobtrusive little block of 
carbon the whole future of the streetcar rested. But the 
only alternative, had no satisfactory brush been discovered, 
would have been some form of alternating-current streetcar 
far more complicated and far less efficient than the direct- 
current apparatus. 

It was probably fortunate that the serious nature of the 
brush problem had not been obvious to the laity. For there 
came to Richmond in the early summer of 1888 a man who 
was to set the streetcar humming at a faster pace. This was 
Henry M. Whitney, president of the West End Street Rail 
way of Boston who was considering either electric or cable 
propulsion for his road. 

Sprague arranged a special demonstration for him in Rich 
mond. There had been some criticism as to the ability of 
electric power to handle traffic when the cars were closely 
bunched. To prove that it could be done, Sprague gathered 
twenty-two cars at one end of the line, far removed from the 
powerhouse. Operations had concluded for the night, and 
Whitney was aroused from sleep to witness the demonstra 

At the wave of a lantern, twenty-one motormen all started 
their cars at once. The line pressure, which had been raised 


to 500 volts, immediately dropped to 200 volts, dimming all 
the car lights. Then it gradually rose again, and the cars, 
which had started slowly, went rapidly forward. There was 
no more talk of cable power in Boston. Sprague, however, 
did not get the contract, but that story will be taken up later. 

Meanwhile, a new contribution to the electric street- 
railway had been made, an invention of Professor Thom 
son's. It embodied the magnetic blowout principle of his 
pioneer lightning arrester, used also for correcting other 
disturbances than lightning. In the design of an efficient con 
trol device for streetcar motors the idea proved of the 
utmost value. The controller is simply a form of switch, 
and as with all switches, it is necessary to prevent the cur 
rent from forming a destructive arc when it is interrupted. 
Thomson's arrangement was easily adapted to Thomson- 
Houston railway work, and was embodied therein from the 
outset. It consisted of two metallic horns surrounded by a 
magnet, the latter "blowing out" the arc by forcing it toward 
the tips of the horns where it broke harmlessly. 

It was one more step toward reliable streetcar apparatus 
which would hasten the final departure of the plodding 
horse from the car tracks of American cities. 


The Period of Expansion 
and Consolidation 

The Dilemma of Patents 

As THE YEAR 1889 came in, the Thomson-Houston Com 
pany had little reason to feel dissatisfied with its progress. 
The volume of business per year had grown from $300,000 
in 1883 to $3,500,000 in 1888. 

The only fly in the ointment was the question of patents. 
During the hours that mean to most men leisure and relax 
ation, Coffin pondered over this problem. Suppose some of 
his customers should desire an underground conduit instead 
of an overhead trolley? He could not install a conduit system 
without the risk of infringing the Bentley-Knight patents. 

Regarding the best type of motor and the best method of 
motor installation he was again on uncertain ground. The 
practice of the Bentley-Knight company had been followed 
in a general way in the Thomson-Houston product. There 
were many who felt that this motor summarized the best 
qualities of its predecessors. It was a fifteen-horsepower unit 
weighing almost a ton and having a speed of a thousand 
revolutions a minute. It delivered its power by double- 
reduction spur gearing, the armature revolving about nine 
times as rapidly as the car axle. The magnetic strength of the 



fields was particularly good in proportion to the total weight. 
Two motors were used per car, one to drive each axle. 

Successful as it was, no one supposed that this represented 
the ultimate in streetcar motors; and rumors that there were 
others which embodied radically new ideas hardly lessened 
Coffin's perplexity over the prospect of patent difficulties. 

The principal advocate of the single reduction motor at 
this time was no longer Sprague, who had had to abandon 
the type at Richmond, but Professor Sidney Howe Short, 
whose inventions formed the nucleus of the new Short Elec 
tric Railway Company of Cleveland. His experiments with 
single-reduction motors were awakening excitement in the 
offices of other electrical manufacturers. Before his retire 
ment, he was to produce the first motor that operated with 
out gears of any sort, having its armature direct-connected to 
the car axle. 

All of Short's experimental work was performed at the 
plant of the Brush Electric Company, which also manufac 
tured his motors for the market. In effect, Short was the suc 
cessor of Bentley and Knight. He made use of Brush dynamos 
for his generators and motors as they did. In so doing he took 
full advantage of the easy regulation afforded by a series- 
wound motor. This regulation could be accomplished simply 
by properly shifting the commutator brushes whenever a 
change in speed was desired. In the Sprague motor the regu 
lation was obtained by means of a commutator field, and in 
the Thomson-Houston, a resistance rheostat was employed. 

The Brush company itself never entered the street-railway 
field. Its specialty continued to be arc lighting, and as such 
it was a powerful rival of Thomson-Houston. Brush had 
patent control over the double-carbon arc lamp, and it was 
believed that his storage-battery patent was so broad as to 
exclude any competitor in that field. 


Another competitor was the Fort Wayne Electric Light 
Company, which was active in the middle west and south, 
installing Jenney and Schaffer direct-current arc dynamos 
and Slattery alternating-current incandescent dynamos. Mc 
Donald was a personal friend of Coffin's, and toward the 
middle of 1888, perceiving the need of additional working 
capital, he entered into negotiations with Coffin. As a result 
the Thomson-Houston company purchased a majority of the 
stock of the Fort Wayne company, obtaining a controlling 
interest. Coffin did not seek to acquire any patents by the 
transaction, as he felt that his company could hold its own 
in the open markets without infringing upon the Fort Wayne 
patents. What he had in mind was the desirability of con 
ciliation and confidence among competing concerns. Coffin 
appreciated McDonald's unusual executive ability, and knew 
him for a bluff individual, erratic and daring, but just and 
shrewd. McDonald's methods were not Coffin's, but the two 
men were probably the most successful merchandisers in the 
electrical industry in that period. 

Almost simultaneously, the Schuyler Electric Company 
entered the market. This concern had started a competing 
arc-light campaign in New Britain, in the days when Thom 
son and Rice were working there, under the guidance of a 
commercial promoter named Spencer D. Schuyler. Its prin 
cipal inventors were D. A. Schuyler and A. G. Waterhouse, 
who had put out a sixteen-light arc dynamo and a lamp. In 
1887 they moved to Middletown, Connecticut, and carried 
on a fairly prosperous business. In 1888 they were installing 
a 45-light machine in New England under the nose of the 
Thomson-Houston people. 

But the company lacked Coffin's genius for securing new 
capital and could not expand as its operations required. In a 
financial quandary, the Schuyler officials were apprised that 


the Thomson-Houston company needed additional plant 
facilities. When Coffin made an acceptable offer, the chief 
construction man, J. A. Dalzell, as well as Schuyler and 
Waterhouse, joined the Thomson-Houston forces. 

THROUGH THE 'EIGHTIES, the Thomson-Houston experts con 
tinued to meet with adventure at home and abroad. Their 
day's work was seasoned with more than the average spice 
of excitement. 

One of them, William B. Potter, who had technical train 
ing to his credit, and who went through the instruction 
school with unwavering zest, was sent out to adjust com 
plaints from the central stations of local companies. 

He was delayed in a small town in North Carolina one 
night when there was no train out until the following morn 
ing. While chatting with an attendant at the generating sta 
tion, the door suddenly opened and a determined-looking 
gentleman stalked in. He was the president of the local gas 
company with a few threatening remarks to address to the 
arc-light people. 

"I've heard how your plant is going to put our gas com 
pany out of business," he said, "and I've come around to see 
that you don't do it. I'm going to shoot the daylights out of 
your machine." 

He w r hipped out a revolver and advanced upon the dy 
namo. Potter acted quickly. 

"Here, have a care!" he cried. "You put a bullet in that 
dynamo and you'll cause a flash of electricity that will kill 
us all. Don't go near with metal in your hands." 

As he spoke, the dynamo emitted a bright, blue flash. The 
gun was drawn back in panic. The dynamo flashed again, 
and the gas president waited no longer. He dashed for the 
door and disappeared into the night. The two electrical men 


wiped a cold perspiration from their brows. And the dynamo 
placidly continued its intermittent sparking. 

Shortly after this Potter was despatched to San Antonio, 
Texas, to install the Alamo Electric Railway, with instruc 
tions that it must be in operation in forty days, generating 
station and all. The road was about five miles in length, and 
there was one street within the city where right-of-way per 
mits had not been granted. Potter decided he would have to 
lay track on that street between Saturday and Monday and 
get the permits afterward. 

About two A.M. on Sunday, while operations were at their 
height, he was arrested. The chief of police was particularly 
hostile because he had been awakened from sleep. But Potter 
was able to put his arguments so convincingly that the chief 
sent him back to work. The right of way was legally secured 
a day later. Operation was begun on the thirty-eighth day. 
Potter himself broke in the green motormen, stressing the 
importance of deft work with the hand brakes in making 
emergency stops. 

Meanwhile, at Lynn, Professor Thomson was doing sig 
nificant work with alternating current. His experiments had 
been suggested back in 1884, when he built an electro 
magnet, large for those days, having a core six inches in 
diameter and twenty inches in length. He had noticed that 
after it was magnetized, if he dropped a disc of sheet copper 
about one-sixteenth of an inch thick upon one of the mag 
netic poles, the disc would not fall with a quick contact but 
would settle gently, at variance with the laws of gravity. It 
was evidently slowed up as it encountered the magnetic 

Before long he found that when alternating current, in 
stead of direct current, was passed through the coils of the 
magnet, the copper disc was alternately attracted and re- 


pelled but repulsion was stronger than attraction. He ex 
plained the cause as "electro-inductive repulsion." 

It was not until 1887 that he worked out a practical device 
based upon this discovery. Then he built a motor having a 
laminated field; that is, a field constructed of many discs of 
metal pressed together, and a laminated iron armature, with 
three coils, three commutator segments, and brushes. He 
short-circuited the armature coils and passed an alternating 
current through the field coils. The armature of the little 
machine thereupon set up a revolution and he obtained con 
siderable power. It was one of the first induction motors ever 
built, and certainly the first repulsion motor. 

But motors, both traction and stationary, had not yet come 
into their own. There were scores of Sprague motors already 
spinning in countless New England industries. But Thom 
son's repulsion motor was to become the father of a long 
line of machines which were to outstrip them in efficiency 
and number. 

Thomson described his experiments with the repulsion 
motor in a paper before the American Institute of Electrical 
Engineers in May, 1887. It aroused scientific excitement, and 
started the keenest among contemporary inventors working 
with alternating current. As a result, refinements of practice 
in alternating-current design were accelerated throughout 
the whole electrical world. 

Every effort was made to introduce the safeguarded 
alternating-current system. Old customers were encouraged 
to substitute it for the direct-current system. By the end of 
1887, the shipping department at Lynn had seen twenty-one 
alternators go out to central station customers, and more than 
four hundred transformers. These represented a capacity of 
more than 10,000 incandescent lamps of sixteen candle- 


power. In 1889, over 300 alternators and more than 9000 
transformers left the works. 

In one period of twenty-four months, no less than 300 
local companies were established by Thomson-Houston 
efforts; and year after year the number of new companies 
regularly averaged between 100 and 150. On January 1, 
1887, there were 171 companies in existence, operating 21,- 
000 arc lamps; two years later there were 419 companies and 
51,000 lamps. 

Coffin kept himself posted on everything. It is hardly con 
ceivable that one man could follow so intimately and so 
unfailingly the intricacies of a many-sided business, growing 
by such tremendous bounds. At the same time he was inter 
viewing some of the biggest men in Boston to offer them 
stock in the company. He believed whole-heartedly in the 
electrical business. It was a subject which he presented with 
all his innate friendliness of spirit, and with the utmost ear 
nestness of purpose. 

He encountered far more than the usual hesitancy in such 
matters. He heard all the customary inquiries as to the earn 
ings and volume of business of his company, plus expres 
sions of skepticism as to the future of electricity. But he was 
able to win over prospective stockholders nine times out of 
ten. He secured the participation of such influential men as 
F. L. Ames, former Governor of Massachusetts; Henry L. 
Higginson, of Lee, Higginson and Company; T. Jefferson 
Coolidge; S. Endicott Peabody; and George P. Gardner. 
These were among the most conspicuous men in Boston and 
they became a phalanx of strength to Thomson-Houston. 

No man more completely held the trust and admiration 
of his co-workers than did Coffin. No man exercised his 
leadership with greater simplicity, greater humility, greater 


regard for others. He dominated, it is true; but he never 
domineered. His dominance sprang directly from the confi 
dence which others placed in him; those outside the com 
pany, who watched his masterly direction, and those within 
the company, who served under him and found that service 
a memorable experience. 

Beside the Mohawk 

EDISON WAS PROSPERING ALSO. So satisfactory had the placid 
Schenectady river site proved that the two original Mc 
Queen buildings became the nucleus around which a thriv 
ing plant took form. 

The Edison Tube Company and the Edison Shafting Com 
pany had followed the Machine Works to its new quarters 
on the Mohawk. A species of amalgamation was accom 
plished on July 30, 1886, when both companies were formally 
merged, with the consent of their stockholders, into the 
Edison Machine Works. All auxiliary equipment required 
for a lighting system, from dynamo to lamp, was manu 
factured in these three plants. 

Two men of pronounced mechanical talents had followed 
Edison from Menlo Park. One was Charles Batchelor, for 
mer Shop Superintendent at Goerck Street, and the other 
John Kruesi, the skillful Swiss who built the first model of 
Edison's phonograph and heard it "talk back" to its inventor. 
He had been in charge of the tube works and now began to 
oversee several departments, with the title of Assistant Gen 
eral Manager. 



Under Kruesi was that well-remembered figure, William 
B. Turner, familiar to the working force as "Pop." He was a 
rough and ready individual, a type which predominated in 
the Edison personnel. His manner was genial and he ad 
dressed his men as "lad"; he was seldom seen with his coat 
on, and never was known to appear at the plant in a white 

Turner was an expert millwright, and his services were 
invaluable in setting up the machinery that was brought up 
from New York. Kruesi, although rough after his own fash 
ion, came to be known as the most gentlemanly among the 
immediate bosses. He, like "Pop," was a hard worker and 
drove his men. He gave his directions in a bluff bass roar, 
but kept his language free from profanity, which was some 
thing unusual. His heart was kindly the interests of the 
men became a personal concern, and he revealed a fatherly 
manner when dealing with a subordinate who he thought 
was not making the most of himself. 

The Edison bosses soon became personages of standing 
in the town. Indeed, the entire plant rapidly took on the 
status of a local institution. There was always smoke coming 
out of its chimneys. The ring of the hammer and whir of the 
saw were heard as new shops rose. The community discov 
ered that it had acquired an energetic, progressive enterprise 
where the old McQueen buildings once stood in forlorn deso 
lation, and felt in it considerable civic pride. 

With the spring of 1887 the plant began to spread, and 
this enlargement continued with recurrent pauses for nearly 
forty years. The original of Building 11 (destroyed by fire in 
1892 ) was reared as a new home for the underground tube 
works and for the manufacture of wire and cable, under the 
general foremanship of Christian Rach. This building stood 


A portrait of the first president of the General Electric Company, taken 
at about the time he took office. 


The Schenectady plant the year before Edison and Thomson-Houston 
joined forces to form the General Electric Company. In the back 
ground can be seen the Mohawk River and the Erie Canal. 


In eight years the Thomson-Houston company had expanded to this 
size from the single building on Western Avenue, seen at the ex 
treme left. 


on the other side of what was later to become the mile-long 
main street of the works. 

Two years later Thomas Cromerford Martin, writing for 
the Electrical World of August 25, 1888, sketched graph 
ically his experiences during "A Day with Edison at Sche- 
nectady." Had he known that the handful of shops he then 
described would one day become a huge industrial center, 
he could hardly have written with more enthusiasm. 

"We are looking," he said, "at one of the greatest exem 
plifications of the power of American inventive genius, and 
at an establishment where, from beginning to end, a new 
art is illustrated by new processes." 

The works by that time embraced twenty-six separate 
buildings and employed from 750 to 800 hands. "The depart 
ments," wrote Martin, "include the building of Edison dy 
namos, the construction of Sprague electric motors . . . ev 
ery branch of the insulated wire business . . . Sims-Edison 
torpedo work, the Edison processes for dealing with refrac 
tory ores, and a general business in shafting, pulleys, and 
hangers, and in millwright and foundry work." 

Martin was greatly moved by the "noble machine shop." 
"In such a place as this," he says, "the prosaic and the 
marvellous jostle each other. Here are six thousand feet of 
shafting and some fifty thousand feet of belting, driving 
nearly four hundred separate mechanisms in the production 
of apparatus whose birth was yesterday." 

Even in those days the schedule of work-hours in the 
electric generating stations was beginning to undergo a 
revolution. The days of night-duty only were beginning to 
pass. Says Martin: "The telegraph office, requiring current 
for its circuits; the telephone exchange, with its magnetos to 
be run; the medical establishments, with patients to be cured 


by electricity; the printing offices; the ice-cream saloons; the 
buildings with elevators; the wood- working factories; the 
chemical works with bad ventilation . . . each of these and 
hundreds of other places need current all day long for direct 
use or to drive motors, and they are all becoming customers 
of the central stations." 

Nor had the day of the isolated plant come to an end. 
Mills, iron works, machine shops, publishing offices, and asy 
lums, as well as theaters, clubs, banks, office buildings, steam 
ships, and ferry-boats, in considerable numbers had installed 
Edison isolating lighting plants. 

In 1886 the Edison Electric Light Company took over all 
the business which had been previously handled by the 
Edison Company for Isolated Lighting. After four years it 
was merged back into the parent company. The Edison 
Electric Light Company still had its headquarters at 65 
Fifth Avenue in New York City. It issued licenses, under 
Edison's electric-light patents, to the teeming family of Edi 
son local companies. 

Two other Edison organizations did not move to Sche- 
nectady. These were the lamp works at Harrison, New 
Jersey, and Bergmann and Company. The latter was op 
erated by Sigmund Bergmann, of the Menlo Park group, who 
combined to a high degree both mechanical and executive 
talents. The plant, at Seventeenth Street and Avenue B, New 
York, manufactured practically everything Edison needed 
in the line of auxiliary equipment, such as meters, switches, 
lamp-sockets, cut-outs, fuses, and electroliers. 

The Edison Consolidation 

PROSPERITY IN THE FORM of new business had also poured 
her bounties upon the Sprague Electric Railway and Motor 
Company during the years of 1888 and 1889. The long night 
mare at Richmond was past, and the prestige of the final tri 
umph there brought other street-railway contracts flooding 
in. Moreover, the company's business in stationary motors 
was far more important than its competitors realized. 

The company had for an emergency man Dave Mason, 
known as the "wet nurse" because he had hurried out at short 
notice to save many an infant electric road from the fatalities 
of babyhood. He had installed and watched over the first 
long-distance circuit on which Sprague motors worked, one 
of the earliest instances of transmission over any appreciable 
stretch of territory. 

It would seem like child's play to the present electrical 
engineer, for this line was but eighteen miles in length. But 
for 1887 it was a considerable undertaking. It was located 
in a mountainous California wilderness on the Feather River 
where it encircles Big Bend Mountain. A hydroelectric 
generating station supplied the circuit, in which were con- 



nected fourteen Sprague motors scattered along the course 
of the stream. Their work was to drive pumps and hoists and 
run dinky cars in the operations of placer mining, for the Big 
Bend had gold within its rugged bosom. 

Officials of the Sprague Company had formed no con 
ception of the amazing welcome which would be accorded 
the electric motor. It was instantly recognized as a worker 
whose driving arm would never tire and whose disposition 
fitted well with that of the human operative. 

William L. R. Emmet, who like Sprague was trained in 
the navy, and Horace F. Parshall, a young man skilled in 
engineering mathematics, were valuable additions to the 
Sprague personnel. Parshall took charge of the establishing 
in Cleveland of the Euclid Avenue electric street railway. 
Sprague was astonished one morning to find a letter from 
Parshall condemning the No. 5 motor in such language as 
only a much-irritated man would employ. 

"That motor," he wrote, "is so extremely rotten that unless 
a No. 6 type is immediately developed, 111 quit the job." 

Sprague knew its weaknesses well; and shortly thereafter 
the No. 6 motor appeared. It was larger in capacity fifteen 
horsepower and embodied the double-reduction principle. 
Its greater physical size was accommodated by the practice, 
begun by car builders, of producing independent car trucks. 

Emmet, while promptly recognizing the good qualities in 
S prague's apparatus, did not consider that it was always well 
built, either in the Sprague or the Edison shops. "In most 
cases," wrote Emmet at a later period, "the motors gave very 
expensive trouble and had to be settled for on terms more or 
less disadvantageous to the manufacturer. In spite of these 
troubles, there was a great demand for equipment which 
would run." 

Because of poor workmanship in the shops, Sprague found 




it imperative to have Emmet rewind many of the No. 6 
motors after they had been shipped. He was equipping sixty 
cars for a new road at Allegheny City. Under instructions 
from New York, Emmet rebuilt the trolleys and the control 
lers, altered the system of lubrication, and enclosed the gears 
in gear cases. The road ran under Emmet's direction for 
nearly a year with little trouble. Maintenance costs and elec 
trical efficiency on the road were good. 

This experience brought out Emmet's faculty for finding 
the weaknesses in apparatus and correcting them. It was a 
faculty which was one day to solve one of the great engineer 
ing problems in the realm of apparatus design. 

Yet with all its success, the internal condition of the 
Sprague company during 1888 was far from stable. Possessed 
of a splendid future, it lacked proper financing. Its president, 
Edward H. Johnson, was burdened with two presidencies 
that of the Edison Electric Light Company being the second. 
Sprague spoke his mind freely upon the question of Johnson's 
relation to both companies. 

"Unless this company were to sacrifice everything for 
which it has worked so long, and which after such hard work 
it has won," he said, "I see no alternative, certainly for the 
present, save that you should give up one position or the 
other. The two are antagonistic, and held as they now are 
will, I fear, end not only disastrously to this company but to 
your career as business manager." The crisis was not resolved 
until early in the spring of 1889, when the status of the 
Edison company itself was changed. 

Edison, as a man of inventive genius, is perhaps unique in 
entering successfully the field of electrical manufacture and 
the field of public-utility operation. But it was as inventor 
that he was greatest, and from the other activities he with 
drew little by little. 


Edison dropped out of active relationship with the Edison 
Electric Illuminating Company. The separation was con 
summated suddenly in the spring of 1889 by a merger of 
individual Edison manufacturing enterprises into a corpo 
rate organization, the Edison General Electric Company. 

Four companies were consolidated by this move the 
Edison Electric Light Company, the Edison Machine Works, 
the Edison Lamp Company, and Bergmann & Company. In 
addition, the new concern bought out the Canadian Edison 
Manufacturing Company, and the Edison United Manufac 
turing Company, which was the selling outlet for the three 
American manufacturing units. 

The Edison Electric Light Company had throughout its 
life been comparatively inactive in exploiting electric power. 
Few company officials, from Edison down, could give much 
consideration to motors while the lighting business continued 
to be so brisk. In issuing licenses to local Edison companies, 
the Edison Electric Light Company found business at flood 
tide. It had its own sales agents, of the promoter type, super 
vised by W. Preston Hicks and similar to those of the 
Thomson-Houston Company. And it had a policy similar to 
Coffin's of accepting stock from local companies in payment 
of royalties due the parent company under Edison patents. 
By the autumn of 1889 the parent company had an interest 
in more than seventy Edison licensee companies over the 
United States. The company's month-to-month revenue was 
composed of dividends and interest on stocks, and a royalty 
for every incandescent lamp. In the fiscal year ending in 1889 
the total receipts from these two sources were $160,500. 

Nearly all the large Edison central-station companies had 
been established by 1889 Boston, Chicago, Detroit, Brook 
lyn, Philadelphia. It had been sufficiently absorbing to main 
tain relations with these ventures, to meet expanding needs 


for equipment, to supply lamps, and to cope with the progress 
of sales agents in securing new contracts. But an attempt to 
consider electric motors while electric lighting was so rapidly 
expanding would have precipitated serious problems. The 
addition of a new department could only come about through 
a general enlargement of facilities and this the new merger 

The Sprague company needed healthier financing; and the 
Edison company was now in a position to enter the field of 
electric power. What resulted was the bodily absorption of 
the Sprague company into the Edison General Electric. 

Joseph P. Ord, a capable executive and a master of financial 
procedure became comptroller of the new organization. 
Young Whitestone, the bookkeeper of the Sprague concern, 
was hard at work over his books when the red whiskers and 
strapping figure of Ord appeared in the doorway, totally 
unannounced. No official announcement had been made to 
the Sprague personnel concerning the change in officers. Ord 
strode over to the counter and demanded to see Whitestone's 
general ledger. The lad gave one look at the towering appari 
tion and shoved the ledger onto his stool and sat upon it. The 
bookkeeper's fears were soon put at rest. Indeed, Whitestone 
and Ord became fast friends, despite that first encounter. 

Thus came into corporate life the Edison General Electric 
Company. From the shoulders of Edison it lifted the multitu 
dinous details essential to the manufacture of his inventions. 
Edison, in alliance with the new corporation, gravitated with 
it away from the utility operating field. Since the manufac 
ture of his inventions was now to be handled by an organiza 
tion fortified by large financial resources, he began to turn 
from the manufacturing field as well, back to that mysterious 
realm which no true inventor can forsake. 

As for Sprague, his affiliation with the new company was 


brief. While the change of control was maturing, he con 
tinued as vice president of the Sprague company, but in 
July he resigned his office and started out again by himself. 

When President Whitney and his directors decided to pro 
ceed with the electrification of the West End Road of Boston, 
they turned to the man who had accomplished the Richmond 
miracle, and in the autumn of 1888 awarded to Sprague's 
company a contract for a trial installation on the Brookline 
division of thirteen miles of track to be equipped for the 
operation of twenty cars. The Sprague overhead wire was 
built from the Brookline terminus to the Boston city line, 
making use of tapering steel poles. Within the city limits 
overhead construction was not allowed, so the Bentley- 
Knight underground conduit was laid between the two 
tracks instead of in the center of each track. The beginning 
of actual operation on the line was delayed by the construc 
tion of the conduit, regarded by Sprague as a necessary evil. 

The first official run occurred on January 2, 1889, amid 
snowstorms which caused trouble with the conduit. Sprague 
equipped a snow-sweeper with one of the motors used in his 
demonstrations of 1886 upon the elevated railway in New 
York, and with this the way was cleared for an impressive 
demonstration. The press registered enthusiasm; and the 
people through official bodies and organizations hastened to 
taste the thrills of riding on this grown-up brother of the 
''broomstick cars." 

But Whitney was not so easily satisfied. As a business man 
he relished competitive trial, and it was to the Thomson- 
Houston company that he assigned the Cambridge division of 
the West End system. The road was equipped with overhead 
wires, utilizing a cross-arm method of suspension from iron 
poles. Its cars were supplied with fifteen-horsepower motors, 


larger in capacity than Sprague's No. 5, more rugged, more 
powerful, and giving a better electrical performance. 

Operation of this division began on February 16, 1889, 
from which day Coffin and his associates argued the merits of 
their system before Whitney with all the vigor and diplomacy 
at their command. They knew the prestige it would mean for 
their company should they win the final contract. 

Three months slipped by. Whitney was a keen trader. He 
wanted reliable electrical equipment and required guarantees 
that would protect him from unforeseen reverses. The 
Sprague company would not make a proposition which 
would guarantee to assume the risks. 

It was Charles A. Coffin who produced what Whitney re 
quired. It consisted of supporting the work of his electrical 
engineers with financial guarantees. As a result on June 27, 
1889 Whitney placed the contract for the West End Road in 
the hands of Thomson-Houston. 

A new shop Factory C had to be built to fill the huge 
order. Manufacturing schedules were adjusted and the work 
ing force increased. Rice recommended a program of expan 
sion, which Coffin immediately endorsed. 

Already intensive redesigning of the dynamo for the 
severe conditions of electric railway service had set in. The 
slotted armature, as finally designed, was freely ventilated by 
air ducts and was of the four-pole arrangement, finally rated 
at 270 horsepower. Rohrer followed the shop work and con 
ducted the tests. It required two days to assemble the first 
machine for trial, using wooden blocks and hydraulic jacks 
in lieu of cranes, which had not yet come into extensive use. 

The big dynamos created considerable stir among the 
entire force. Yet still larger ones, rated at 360 horsepower 
and having six magnetic poles, were brought out for use on 


the West End Road. By 1892 the powerhouse of this road 
was the talk of street-railway men far and wide. There were 
over a hundred generators weighing some thirty-five tons 
apiece. They were driven by belting which aggregated a 
length of 1800 feet over a third of a mile and was fifty- 
four inches in width. 

By 1891 the line was running 130 electric cars at an 
average speed of eight miles per hour two miles an hour 
better than the best horse-cars could do. Traffic on the road 
had increased enormously. During snowstorms, more than 
the normal number of cars could be run, as well as electric 
plows and sweepers. After the first snowstorm, when the 
system was stalled for an hour, the elements were unable 
to derange its operation. 

Nor was this the only success that electric traction won on 
the fateful West End Road. A clause in the original contract 
stipulated that the Thomson-Houston Company should main 
tain in good working condition and keep in proper repair the 
entire overhead wire system on the West End Road, as well 
as the motor equipment on the cars, for a period of five years. 
Compensation for this service was to be paid by the West 
End Company according to a graduated scale based on the 
revenue per car mile. Back of this provision can be discerned 
the doubts which Whitney still entertained regarding the 
practical value of electrical equipment. Opposed to this was 
the complete confidence of the Thomson-Houston Company. 

In less than two years of operation Whitney was completely 
convinced. Cost of maintaining the overhead lines and car 
equipment was so much less than what he paid Thomson- 
Houston for such maintenance that the latter were realizing 
a substantial profit. Whitney thereupon exercised an option 
allowing him to take over the work, which saved his company 
several thousands of dollars yearly. 


The skeptical years were definitely at an end. Electricity, 
as a practical agent of power, had come to stay. From then on 
there were few to doubt that the pigmy which in 1876 had 
whispered such unintelligible secrets of the future would 
become in time a giant. Not a master, to rule the peoples of 
the earth, but a giant slave, to serve with unlimited strength 
and unending efficiency those who would command him. 


"Experts" on the Job 

IN THE MIDST OF SUCCESS Charles A. Coffin was troubled by 
visions of the future. He would never be satisfied until the 
day he offered his customers complete electric service. Only 
thus, as he saw it, could the electrical era dawn in America. 
The alternative was a feverish, interminable period of conflict 
between patent holders, during which the major competitors 
in the field exhausted their energies trying to avoid each 
other's patents or fighting alleged infringements. 

That was the depressing picture of the future which 
troubled his mind, even while the tap of the mason's trowel 
proclaimed the rearing of new factory buildings. 

Coffin already had a glimpse of what the future would be 
like. The Thomson-Houston Company was trying to discover 
a double-carbon arc lamp that would avoid the Brush patent. 
After hours of conference by engineers and attorneys, the 
lamp was declared legally sound. But the Brush Company 
promptly served notice of an infringement. A legal combat 
loomed, and its menace caused Coffin to think again of 

To his colleagues it seemed a daring proposal. The Brush 



OF 1893 

Exhibits and installations by the Company at Chicago in 1893, 


Company was as strongly entrenched as Thomson-Houston, 
possessed the prestige of a firstcomer in the field, and boasted 
of successful installations throughout the land. Coffin never 
theless dispatched a personal representative to Cleveland 
with an offer to purchase Brush's stock, which constituted a 
controlling interest in the company. 

The emissary arrived in Cleveland in October, 1889, em 
powered to offer $40 per share. Brush had no interest in this 
figure and held out for $75 per share. When he remarked 
that he did not suppose the Thomson-Houston Company 
would pay his price, the agent wagered a new overcoat that 
the offer would be accepted. The wager was promptly 

Next day Brush received a telegram stating that he owed 
Coffin's representative a new overcoat. The price of $75 per 
share for forty-thousand-odd shares of stock had been ac 
cepted and control of the Brush Electric Company passed 
into Thomson-Houston hands for a sum in excess of 

So rapid and so quiet had been the negotiations that the 
electrical world was taken by surprise. The first reports of 
the transaction were scarcely credited. 

No change in the conduct of the Brush Company's opera 
tions resulted from the transfer of control for many years 

By purchasing control of the Brush concern, Thomson- 
Houston strengthened its position as proprietor of one of the 
soundest electrical manufacturing organizations in the United 
States. It was doing business on a scale equal to that of the 
Edison General Electric Company. 

A year later there came a new alignment among the 
Thomson-Houston subsidiaries. The Fort Wayne Company 
believed that the Jenney arc-light apparatus could be im- 


proved. To do this job, they selected a young man named 
Wood. At that time Wood was affiliated with the Thomson- 
Houston Company, conducting a modest manufacturing 
enterprise in Brooklyn which turned out three or four dyna 
mos a week. McDonald effected a transaction by which the 
Fort Wayne Company purchased from Thomson-Houston a 
part interest in Wood's arc-light system with the understand 
ing that the designer become a member of the Fort Wayne 

Wood, accompanied by a hundred employees, arrived in 
Fort Wayne on December 3, 1890. It was the beginning of 
the most productive period in his career. McDonald en 
couraged Wood with characteristic ardor in every inventive 
idea which offered practical possibilities. Before long Mc 
Donald's tactics had established Fort Wayne as the home 
of the Wood system of arc lighting. 

Hopeful young men kept arriving at the now celebrated 
plant of the Thomson-Houston Company, seeking to become 
"experts." By the beginning of the 'nineties, college graduates 
were noticeable among them. But they were novices when it 
came to practical electrical work, and there was nothing 
"greener" around the shops than a new expert. 

Most of the young men needed all the ambition and grit 
they could muster to carry them through weeks of soul-trying 
work at menial tasks. But these weeks separated the wheat 
from the chaff. The men who survived the testing were 
ready to advance in the company. 

The expert course had developed into a separate depart 
ment. In early years the General Manager of the company, 
Silas A. Barton, hired the beginners. Later J. B. Cahoon was 
chief of experts. 

How green the new arrivals were is revealed in the 


reminiscences of A. K. Baylor, who has described his feelings 
when he first entered Factory C. 

"On either hand crackled and hummed various mighty 
dynamos. Belts! I had never seen so many horizontal, verti 
cal, and slanting. Those driving the generators were wide 
and heavy, flapping as they ran, and shooting little spurts, 
here and there, to an occasional guard rail . . . 

"By the time we had gone half-way down the line 
becoming accustomed to the racket and observing that no 
one was being struck dead I gained courage and became 
quite elated . . . Then I came to earth, almost literally, with 
the proverbial dull thud, as I was put to work wiping up 
grease on the floor under the frame of a railway motor. The 
task was uninspiring, to say the least; and it seemed so useless. 
What mattered a little grease, more or less, on an area already 
covered and impregnated with it?" 

His working companion that day was a chap whose con 
ception of electricity was limited to the familiar explanation 
that current passing through a wire is analogous to water 
flowing through a water pipe. He had taken this literally, 
believing that the cable running across the floor from the 
starting rheostat to the dynamo actually encased a tangible 
fluid. Planting both feet firmly on the cable he exclaimed to 
Baylor, who was about to start a motor, "There! Now let's 
see you make her go!" 

The experts spent a month among shop motors. Upon them 
rested the responsibility of starting the motors before the 
whistle sounded, morning and afternoon. "Woe betide the 
expert who was late on that job!" writes Baylor. "It was a 
well-nigh capital offense. Suspension, following goodness 
knows what ordeal in the inner office, was the minimum 
sentence. , 


"I had quarters almost across the street, and I well recall 
springing out of bed at the five-minutes-to-the-hour whistle, 
to race through the check-house with shoe laces flying, collar 
unfastened, coat on my arm, and buttoning my waistcoat as 
I ran breathless and breakfastless." 

A vanished expedient, the check-house with its wall-board 
on which hung the brass check of each employee, was a fore 
runner of the time clock and the timecard. As each worker 
arrived, he unhooked his check and dropped it into a box. 
The instant the starting whistle blew the box was locked 
and note was taken of the checks remaining on the board. 

Relations between the young experts and the company 
officials were warm. Baylor speaks with appreciation of "the 
privilege of visiting the home of Mr. Coffin under a standing 
invitation to the boys of the course to 'come around' Sunday 
afternoons. Done up in our best, and after painful efforts to 
make fingernails presentable, we often took advantage of 
this opportunity to be served with tea by Mrs. Coffin and 
looked over and 'interviewed' by the great 'C.A.' himself. 

"In that way, and by a continuation of the personal touch, 
augmented by E. W. Rice, Jr., Eugene Griffin, J. R. Lovejoy, 
and other officials, were sown seeds that have blossomed into 
a world-wide harvest of loyalty that few men or institutions 
have ever inspired." 

There were occasional nerve-racking occurrences during 
the testing of new machines. One happened on the day that 
an alternator with bearings of the self -oiling type was first 
tested. The engineers had seen a direct-current dynamo so 
equipped by the Fort Wayne Company, and were captivated 
with the device. They induced Rohrer to permit the bearings 
for their own new alternator a seventy-kilowatt "giant" 
to be redesigned. When the day for the test arrived, hope 


View in Building 12 at Schenectady, about 1890. 

Arc lamps being assembled at the Lynn Works, about 1892. 

Machine shop at the Lynn Works, about 1895. 


and anxiety alternated like an electric current in the breasts 
of the engineers. 

As the belt was slipped upon the pulley of the dynamo, the 
armature began to revolve. Then there came a groan, a 
shriek, smoke and the belt went sailing through the air. 
Dependable, self -oiling bearings indeed! They were almost 

Someone asked the test man who had set up the machine, 
"How much oil did you put in those bearings?" He looked up 
with a blank expression. "Oil? I didn't put any in. I was told 
they were self -oiling." 

It took two days for a skilled machinist to work the melted 
babbitt off the steel shaft. And it was two weeks before the 
machine could be tested again, this time well oiled and un 
questionably successful. 

Every dynamo of greater size completed in the shops 
became a nine-days' wonder. When a new manufacturing 
tool or shop practice was added to the day's routine, work 
men marveled and visitors came to see. The early electric 
cranes were held in awe at Lynn as well as at Schenectady. 
The first of these began operations in November, 1888, in 
Factory K, and had a capacity of twenty tons. Many a week 
passed before workmen refrained from staring when the 
crane lifted a load. 

When a five-ton electric crane was put into use at Schenec 
tady in 1887, old-timers recall how excursion trains on the 
New York Central stopped by the Edison works to give 
passengers a glimpse of the great beast at work. 

While these giant industrial "muscles" were lending fasci 
nation to the humming electrical factories, Thomson was hard 
at work in the laboratory giving scientific counsel, and trying 
out new devices. For several years he had been perfecting 


his electric meter. Unlike Edison's chemical meter, it meas 
ured not only current consumed over a specific time but also 
voltage, thus establishing a true and equitable means of 
rendering bills for electric light and power. The meter won 
fame as the Thomson recording wattmeter. 

So urgently had it been needed that within two years a 
meter department and special shop facilities had to be set up 
to handle the business. Its ruggedness and simplicity left 
little to be desired. 

There are some who believe that this meter was one of the 
determining influences in electrical development. It won high 
honors at the Paris Electrical Exposition of 1890, dividing the 
prize with a clock-type meter which was more costly to install 
and required regular winding. A factory, established shortly 
afterward in France to manufacture the Thomson meter, be 
came the nucleus of the French Thomson-Houston company. 


New Faces, New Companies 

IN 1889 A YOUNG MAN left his native country and crossed the 
ocean to America. His name was Steinmetz. Steinmetz came 
to an America that was conscious of the new pulse of in 
dustrial life and the throb of mighty economic and material 

As a young immigrant he entered New York harbor in 
steerage accompanied by a fellow-student, Oscar Asmussen. 
Together they stood on deck, gazing at the busy harbor. They 
saw a multitude of buildings, symbol of a crowded metropolis, 
of national vigor, of surging life. 

The lad's misshapen body and diminutive figure were in 
odd contrast to his friend's sturdier build. A seeming pygmy, 
he gazed at vigorous America, preparing to enter the conflict 
for sustenance and position that was fought daily on those 
shores. In the air was a newly awakened force, a power that 
men had aroused from lethargy which seemed to confront 
them with passive, inscrutable curiosity as if wondering if 
there were any who would truly master it. On the deck of 
the steamer, a dwarf in stature, insignificant to his f ellowmen, 
stood one of those masters. 



He was not conscious of his destiny and had no intimation 
that he was to solve some of the riddles of the electrical era. 
He had no thought, even, of an electrical career. 

Since men are human, and hence unaware of the forces of 
destiny, it was not strange that the immigration officers hesi 
tated to admit Carl Steinmetz upon that June day in 1889. 
Steinmetz, besides his shrunken body, had a severe cold 
which had swollen one side of his face. The immigration offi 
cers determined that he should be deported. 

Then Asmussen took up the cause of his friend. He de 
clared his money was Steinmetz's money, that his home 
should be his friend's. In brief, he accepted full responsibility 
for Steinmetz's future. 

Thus friendship won for Steinmetz an entrance to his 
future country. He found himself in the heart of the metropo 
lis and in the hands of destiny. Events not entirely of his 
own shaping guided him into his niche in the new world. 
Of the places where he applied for employment, one gave 
him encouragement. In the office of Eickemeyer and Oster- 
held in Yonkers he had a talk with the head of the firm, 
Rudolf Eickemeyer. There he went to work on the tenth of 
June as an electrical draftsman, at twelve dollars a week. 

Eickemeyer was a manufacturer and inventor, well sea 
soned by experience and, like Steinmetz, a native of Germany. 
From him Steinmetz caught the inspiration for putting his 
great technical ability to work upon electrical problems and 
through him was provided the most natural outlet for the 
young man's mathematical talents. 

So for a year or two Steinmetz, hidden away in an obscure 
corner, worked for a modest manufacturer in a humble, 
unspectacular post. His life was quiet, contented, assiduous. 
His surroundings were pleasant and wholly untroubled. 

Only a few weeks after Steinmetz crossed the Atlantic, 


another traveller landed in New York and hurried up to 
Boston and Lynn. This was Rudolf Langhans, a young 
German scientist whom Rice had discovered during a trip 
abroad, and who was believed to have found a way of pro 
ducing a new filament for incandescent lamps. 

He was installed in the Thomson-Houston laboratory, pro 
vided with assistants, and given an interpreter. What 
Langhans proposed to do was to combine pure silicon and 
carbon in the filament. But pure silicon was difficult to obtain 
a fact of which Langhans appeared to be ignorant. The 
Thomson-Houston chemists discovered that what he was 
using was not pure silicon, but oxide of silicon. 

In the end, Langhans regretfully admitted that his idea 
was not practicable. He left Lynn, and the attempt to pro 
duce a silicon-carbon filament came to an end. Though the 
Thomson-Houston Company lost $75,000 on the effort, ex 
perimental work did not lag. "A better incandescent lamp," 
was the cry; and it echoed from Lynn to Harrison, where the 
Edison Lamp Works was wrestling with the same idea. 

Looking at the electric light bulb of today, seemingly a 
device easy to construct, even with its hairlike filaments, one 
finds little indication of the infinite pains and delicate han 
dling that produced it. So intricate is it that in the early days 
at the Edison Lamp Works, when a curious soul asked why 
the various steps could not be done by machines, he was 
laughed at. The work was far too delicate! 

Glass blowing was an old and conservative art, and neces 
sarily slow. Not only the bulb but the glass stem, in which 
the lead-in wires were mounted, had to be blown. Operators 
went through months of training, and even then the per 
centage of spoilage was considerable. 

But one change after another affected the shape and size 
of stems until Edison was able to develop a machine which 


held the bulb in a horizontal position and, after the filament 
had been mounted, rotated it in a flame while the stem was 
sealed in. At Lynn, Branin thought of flaring the stem at one 
end, so that it could be sealed to the bulb on the flange prin 
ciple a process which permitted a more perfect joint, ob 
viated strains in the glass, and caused a saving in glass of half 
a cent a lamp. 

Hand work persisted in such operations as blowing off 
the lower end of the bulb to a uniform length, and in making 
an opening in the bulb top through which a tube was inserted 
for pumping out air. A succession of efforts to produce a 
satisfactory base culminated in Bergmann's base, which is in 
use today. The idea for it came to Bergmann while he con 
templated the screw-cap on a kerosene can. 

But the Edison, Thomson-Houston, and Brush companies 
were not the only manufacturers producing incandescent 
lamps. In various sections of the country, particularly in the 
middle west, small aggressive companies were doing a flour 
ishing business. Lamps manufactured by these concerns were 
similar to Edison's. The filamentary illuminant in a sealed 
glass globe from which most of the air had been exhausted 
was always present. Hence all of these companies, as well 
as Thomson-Houston and Brush, stood in the precarious po 
sition of possible infringers upon Edison's lamp. 

They had taken the risk deliberately, because the policy 
of the Edison Company gave them a feeling of security. Years 
had come and gone since Edison's lamp appeared, and no 
steps had been taken to uphold by legal action that valuable 
patent. But the issue was soon to be fought out at the bar. 
Preliminary proceedings had been begun late in 1886, al 
though so quietly and after so many legal delays that the 
sense of security persisted among competitors. They kept on 
manufacturing and selling. 


Within the stately Edison Building, which stood at 16 and 
18 Broad Street, New York, the busy organism of the Edison 
General Electric Company found its brain center. It was an 
immense corporation for its day, and the forerunner of one 
still more immense. 

Said the Electrical Engineer in a special supplement en 
titled "Edisonia," published with its issue of August 12, 1891: 

"One cannot visit the Edison Building without being im 
pressed with the sense of irresistible power that a large, well- 
knit body in swift, onward motion always arouses. The halls 
and elevators are thronged, and at every floor one encounters 
the same bustle and hum of activity. . . . The industry of 
which these are the headquarters has over 6000 names on 
its payrolls, and has a monthly income and outgo of a million 
of dollars. . . . The company has between 4000 and 5000 
customers on its books." 

Virility was heightened by semimystery. Not yet was the 
veil wholly drawn so that the untutored public could com 
prehend what lay back of its electric lights and motors. Edi- 
sonia, however, described some of the broad features of the 
process. "The mere handling of the bamboo, as it arrives in 
dainty bundles from Japan, tied up in flowery wrappers, and 
passing from stage to stage, becoming smaller by degrees 
until at last it is ready for the carbonizing retorts, involves the 
employment of many sensitive mechanisms and many agile 

Even the tissue-paper squad, smoothing out the used paper 
for repacking, had its glamor. In that lowly environment 
lurked the gold nugget of opportunity. Young George Mor 
rison, now a foreman, had started out in the squad ten years 
before. He was bound to keep on rising, for men still noticed 
that whatever he did, he did well. 

From process to process the visitor continued his rounds, 


coming at last to the packing room, "where in huge racks like 
egg boxes, or in big barrels, the lamps await labeling for the 
nearest city or for the far ends of the earth." It was a wide 
practice to ship the lamps in barrels. Says another writer: 
"A barrel was the piece de resistance in a day when all goods 
were merchandised loosely over the counter, stuffed with 
excelsior, bound with iron hoops, padded with packing bus 
tles." The barrels were made of the stoutest materials for 
shipping their delicate cargoes, and "the agent who received a 
package of lamps had to be a strong and mighty man and a 
good feller with a broom." 

Barrel shipments were massive in the aggregate, and the 
capacity of the "huge factory" in 1891 was 25,000 lamps every 
working day. 

"Edisonia" reports of the Schenectady works that they 
"grow with the rapidity of a western town. . . . These huge 
cathedral shops, swarming, every one of them, with hundreds 
of busy artisans and filled to overflowing with machines 
and tools and labor-saving appliances, epitomize the progress 
to date both in mechanics and in the arts of dynamoed motor- 
construction. If you would find the monument of American 
skill and genius in such fields, look around you." Colorfully 
the writer describes the production of the Edison tubes, in a 
building "where huge caldrons simmer and huge piles of 
black iron pipe rise to the roof, for all the world like exag 
gerated lead pencils," with an electric drill "pounding away 
in a ferocious manner at the heart of a block of granite," 
elbow-to-elbow with "demure young ladies splitting mica." 

After the consolidation of 1889, the shops of Bergmann 
and Company were transferred to Schenectady, and a new 
group of buildings was set up. Sockets for lamps, switch 
boards, and wiring devices were now made at Schenectady. 
Punch presses hammered and thumped from morning to 


night in the new section; and the first porcelain shop was 
there established. 

Edison and Villard were frequent visitors at the plant. 
Their appearance created a general stir. The cry would go 
up, "Here comes Edison!" and there would be a lively cran 
ing of necks. They were an odd pair, Edison and Villard. The 
financier, his moustache close-cut, was always well-tailored, 
while the inventor was never mindful of his clothes, usually 
wearing a shiny Prince Albert coat, and in summer an old 
straw hat. 


kingdom am a-comin' in, an' de year ob jubilee!" Whatever 
salvation electricity has brought, its "year ob jubilee" began 
in 1890. The great formative period of electrical development 
was at its height. 

The nation was passing from one era to another. A new 
mode of life was emerging from the mists of uncertainty, was 
adjusting, adapting, solidifying. Industry and commerce, tak 
ing shape around a new enterprise, developed with it at 
tremendous speed. 

In the rush and stir of this formative period there arose 
much confusion in technical and commercial activity and 
competition among sales agencies. The Edison consolidation 
set out to minimize this as rapidly as possible by adopting a 
policy of centralization. Says Edison General Electric's an 
nual report for 1890: 

"It was found practically impossible to exercise over so 
many distinct organizations the close supervision necessary 
to secure rigid accountability and to conduct the business on 
an economical basis. It was found further that the prevailing 
system of allowing commissions on business obtained through 


local agencies was very expensive, without bringing adequate 

"Your board therefore decided ... to bring about a uni 
fication of all the allied manufacturing organizations and 
selling agencies. ... It further determined that the entire 
business of manufacturing, selling and installing should be 
conducted in the General company's own name, through its 
own employees. 

"The Edison Electric Light Company was excepted from 
this arrangement, and it remains an active, independent body 
in all respects." 

Vice-President Insull and Comptroller Ord addressed 
themselves to the task of a complete reorganization. Out of 
their work came principles of business administration and 
methods of financial accounting which persist today. 

Their salient achievement was to set up a district organi 
zation. The United States was divided into seven large areas 
with a district office in each, and an eighth in Canada. In 
New England the district office was in Boston; in the Eastern 
district, in New York; Central district, Chicago; Southern, 
New Orleans; Rocky Mountain, Denver; Pacific Northwest, 
Portland; Pacific Coast, San Francisco; Canadian, Toronto. 

These were the units which contacted customers. Many 
departments of the company maintained branches in each 
district; and each district had its manager. Not the least of 
their responsibilities was the negotiation of modes of pay 
ment for products purchased by customers. In the company's 
annual report for the following year appears this significant 
statement: "The General Company could have done a much 
larger business if it had been willing to accept securities in 
payment for orders; but ... a strict rule was adopted of 
declining all such and doing business exclusively on a cash 
or short credit basis." 


This represented a departure from Coffin's policy of ac 
cepting securities of local electric light companies. Securities 
already in the possession of Edison Electric Light repre 
sented more than a million dollars, and gave the parent com 
pany a financial interest in more than seventy operating 
companies. Feeling that its own fortunes were linked too 
intimately with those of the local companies, the parent com 
pany arranged for a specialist to take charge of the financial 
operations of licensees, appointing Henry W. Darling, presi 
dent of the Canadian Bank of Commerce. 

There was much to be done, for the operating utilities like 
the manufacturing companies were expanding lustily. The 
local company in New York erected in 1890 a generating sta 
tion with a maximum capacity of 200,000 lights, the largest 
in the country. The report points with pride to the aggregate 
capacity of all the Edison generating stations in the nation 
1,300,000 incandescent lamps. 

All of these operated with direct current, and alternating 
current was left severely alone. The line of cleavage on this 
subject between the Edison, and the Thomson-Houston and 
Westinghouse policies was never more rigidly drawn. From 
1889 to 1891 Edison was in frequent controversy with his 
technical antagonists. Even when his company purchased 
the American rights to a transformer system invented in 
Europe, his inflexible position caused the idea of putting the 
system into practice to be abandoned. 

Yet alternating current was the touchstone of the future. 
Signs proving this were increasing. It was in this same year 
of 1890 that William Stanley, who had remodelled the Gaul- 
ard and Gibbs system for the Westinghouse company, re 
marked to his young assistant, Cummings C. Chesney, "Let's 
go off by ourselves and establish a plant for the manufacture 
of transformers." 


Stanley had completed his work for Westinghouse, and he 
and Chesney went to Pittsfield, Massachusetts, where they 
succeeded in interesting several capitalists, and on Novem 
ber 1 the Stanley Laboratory Company and the Stanley 
Electric Manufacturing Company were organized. A small 
plant was acquired where a force of sixteen men went to 
work. Within three months the first transformers were de 
signed, built, and shipped. 

A year later John F. Kelly joined forces with Stanley and 
Chesney. From that moment the three constituted a team, 
whose fame penetrated in ten years to every corner of the 
electrical realm. Together they originated a complete line 
of alternating-current equipment, announced and advertised 
as the "SKC" system Stanley, Kelly, Chesney. They were 
an aggressive triumvirate, and their work greatly accelerated 
the development of the alternating current. 

There is no evidence that the Edison company publicly 
opposed the Stanley company during the "war of the cur 
rents." When the Stanley company appeared, the greatest in 
dividual issue in the Edison camp was the future of the 
incandescent lamp, and particularly the validity of its patent. 

In 1888 Edison had improved the efficiency of the lamp by 
coating the carbon filament with asphalt. As a result the lamp 
consumed only 3.1 watts per candlepower, compared with 
7 watts per candlepower in 1880. This efficiency held for 
fifteen years, even after the "squirted filament" appeared. 
Efficiencies of 3.5 and 4 watts per candlepower were also 
supplied in 1890, according to the voltage variation in the 
circuits of different lighting companies. 

Already there had been mutterings of controversy. Feel 
ing which centered around the question of patents bordered 
on bitterness. Johnson had issued his pamphlet "A Warning 
from the Edison Electric Light Company." A counter-blast 


came from competitors, who published anonymously a par 
ody of Johnson's style bearing the title: "E N Trium 
phant! An inspirational rhapsody in seven cantos. Translated 
from the Johnsonese by W. W., poet lariat." 

As the 'eighties came to an end, conflict over the Edison 
lamp patent No. 223,898 burst in the fury of a legal storm. 
The future of electricity hung upon the issue; aspirations, 
labors, fortunes, were at stake. 


A Famous Fight 

IT WAS THE AUTUMN OF 1889 when a group of lawyers met 
before a court examiner in New York and began taking testi 
mony in the momentous patent suit instituted to determine 
who was legally entitled to call the incandescent lamp his 
own invention. 

The opponents were the Edison Electric Light Company, 
complainant, and the United States Electric Lighting Com 
pany, for alleged infringement of Patent No. 223,898, granted 
to Edison in January 1880 and embracing his claims for an 
electric lamp consisting of a high-resistance carbon filament 
in a sealed glass container which formed a nearly perfect 
vacuum. The complainant held that the Maxim lamps, pro 
duced by the United States Electric Lighting Company, in 
fringed upon his patent. 

A spark of controversy that had lain dormant since 1882, 
almost hidden from the heedless public, ignored by the elec 
trical interests, had suddenly flamed out. 

When the Edison company raised the cry of infringement, 
other inventors affirmed that his invention had been antici 
pated, that his patent was worthless. Many testified on the 



witness stand that others, Hiram S. Maxim in particular, had 
previously done what Edison laid claim to have done. 

Although no evidence was presented in court until Octo 
ber 15, 1889, the Edison company did not delay so long in 
bringing action against infringements as would appear. The 
suit was formally begun in 1886, about the time the Edison 
company realized that other companies could market com 
peting lamps successfully and profitably. Not until then did 
it cease temporizing and procrastinating. In June 1886 the 
Edison Electric Light Company filed its amended bill of 
complaint, to which the United States Electric Lighting Com 
pany answered by submitting its amended plea. Preliminary 
skirmishes occupied the three-year interval. 

The trial, when it was finally instituted, occupied many 
months; the record swelled into many printed volumes of 
testimony; technical questions were explained, analyzed, in 
terpreted, and legally vivisected; the patent, the grand bone 
of contention, was systematically picked to pieces, its word 
ing attacked, its validity flouted. It bore the full brunt of 
battle, and both sides introduced an endless succession of 
exhibits, physical and documentary. 

Richard N. Dyer was the generalissimo who handled the 
legal artillery for the complainant. His right-hand man was 
Clarence A. Seward, a keen lawyer. And as advisor to both 
was Grosvenor P. Lowrey, Edison's veteran legal friend. 

On the opposing side was Samuel A. Duncan, shrewd, 
quick, and thorough, usually addressed as General, chief of 
counsel for the United States Electric Lighting Company; 
with his able assistant, Edmund Wetmore. 

The defendant's position hinged upon two main points. 
First, it held that the Edison patent was invalid because the 
description of the invention would not enable a person 
"skilled in the art" to make incandescent lamps by following 


its directions. Second, it asserted that even if lamps could be 
so made, Edison's work had been anticipated by other in 
ventors who had produced the same sort of electric lamp be 
fore the date of the patent. 

The defendant introduced the names of five inventors in 
the effort to prove that Edison was not the first man to bring 
out a successful incandescent lamp: William E. Sawyer, Al- 
bon Man, Hiram S. Maxim, Edward Weston, and Moses G. 
Farmer. And there was an array of witnesses to testify as 
experts that an incandescent lamp could not be made by fol 
lowing the directions given in the patent. 

The specifications of the patent disclosed how to prepare 
filaments of a very small diameter out of lamp black and coal 
tar. To prove that these directions were clear to anyone 
skilled in the art, the prosecution called upon John W. 
Howell, electrician of the Edison Lamp Company, to con 
struct lamps by the method described. 

Howell made twenty-seven such lamps, and then went on 
the witness stand and told how he made them. He kneaded 
the lamp black and coal tar into the consistency of putty and 
rolled it out upon a plate of ground glass to the hairlike or 
filament diameter of from six- to seventeen-thousandths of 
an inch. Short pieces were cut off, wound upon a wooden 
mandrel into a spiral form, carbonized, and sealed into vacu 
ous bulbs all precisely as described in the patent. 

When they were tested the lamps burned for six hundred 
hours. The record of the tests, the twenty-seven lamps, and 
the set of tools which Howell employed were introduced as 
exhibits. Howell even offered to construct the lamps in the 
presence of the Court, but this was not deemed necessary. 

But the chief battle arose when the complainant called 
Charles L. Clarke, acting chief engineer of the Edison Elec 
tric Light Company up to 1884. He was relied upon to refute 




the assertion that Edison's invention was anticipated by 
others. Clarke took the witness stand on July 21, 1890, and 
held it almost continually for three months. General Duncan 
led the charge in an extremely clever cross-examination. He 
asked the witness nearly six hundred questions, seeking again 
and again to expose a flaw, a contradiction, an incompetency 
in his testimony. 

The light-giving element of the lamp what Edison termed 
the filament was the storm center of the controversy. Sev 
eral inventors contemporary with Edison had employed car 
bon sticks or "rods," which differed only in diameter from 
Edison's "thread," or filament. The defense contended that 
both types were in effect the same thing, and that the carbon 
rods therefore constituted a legal anticipation of Edison's 

Day after day and week after week the clash of wits con 
tinued. But Clarke's testimony could not be broken down, 
and he showed such familiarity with a multitude of published 
articles and books that General Duncan found him a full 
match for his skillful fencing. 

At length Duncan asked the witness if the arc lamps were 
not, after all, much more efficient than the incandescent 
lamps, and if they were not so regarded by the "prior art" 
the inventions brought out prior to the date of the patent in 
suit. It was evidently intended as a key question, upon which 
a sequence of other questions was to depend. But Clarke 
was not wholly unprepared. The moment the questions was 
put, Dyer, sitting across the table, turned to Duncan and re 
marked: "Here is where you are going to get it in the neck!" 

Then Clarke gave his reply. He held that the arc lamp was 
only slightly more efficient than the incandescent and that 
the prior art so regarded it, citing many authorities on the 
point. He pointed out the evident conviction among investi- 


gators prior to Edison that this margin of efficiency was not 
so great as to deter them from striving to solve the riddle of a 
practical incandescent lamp. 

The answer checkmated General Duncan. It upset his line 
of attack and left him taken aback. For several moments he 
sat silent, stroking his beard. 

Dyer at length remarked, "Well, General, why so silent?" 
To which Duncan replied, "I was just thinking how much 
I'd like to kick a certain young man." 

Finally the time came when the evidence was completed 
and the case rested in the hands of the Court. The Court's 
opinion, written by Judge Wallace, was handed down on 
July 14, 1891. It was a complete victory for Edison. 

The defendant appealed, and on October 4, 1892, the 
United States Circuit Court of Appeals handed down an opin 
ion affirming the decision of the lower court. The patent was 

Judge Wallace looked upon the attenuated carbon burner 
of high resistance and small radiating surface as Edison's 
signal achievement. It was this, in the opinion of the court, 
that turned the incandescent lamp from a failure into a 

"It was a remarkable discovery," said Judge Wallace, in 
the opinion handed down, "that an attenuated thread of car 
bon would possess all the long-sought qualities of a practical 
burner. . . . The extreme fragility of such a structure was 
calculated to discourage experimentation with it. ... The 
futility of hoping to maintain a burner in vacuo with any 
permanency had discouraged prior inventors, and Mr. Edi 
son is entitled to the credit of obviating the mechanical diffi 
culties which disheartened them. . . . 

"By doing these things, he made a lamp which was practi- 


cally operative and successful, the embryo of the best lamps 
now in use." 

The higher court, affirming this decision, gave its opinion 
in part as follows: 

"Edison's invention was practically made when he ascer 
tained the theretofore unknown fact that carbon would stand 
high temperature, even when very attenuated, if operated 
in a high vacuum, without the phenomenon of disintegra 
tion. ... It was an invention, in view of the teaching of the 
art as to the disintegration of carbon under the action of an 
electric current, to still select that substance as a suitable 
material from which to construct a burner much more at 
tenuated than had ever been used before." 

The Court was explicit in maintaining that a carbon rod 
is not the same as a carbon thread one of the classic issues 
in the suit. Said Judge Lacombe of the Higher Court: 

"The evidence fails to satisfy us that the prior art furnished 
any burners less than twice this size. In contradiction to these 
earlier burners, Edison calls his burner a filament" 

These judicial statements make it apparent that the re 
duction of the carbon burner from the diameter of a pencil, 
or small rod, to that of a thread constituted the germ of a 
great invention. Judge Lacombe, in deciding that the Maxim 
lamps infringed Edison's filament, declared that the carbon 
burners in the Maxim lamps were of threadlike diameter and 
that they "indisputably lie wholly on one side of the dividing 
line between rods and filaments." 

This opinion is identical with that of the English court 
which passed on the suit of the Edison and Swan United 
Electric Light Company brought against Woodhouse and 
Rawson, in which the Edison English patent was sustained 
both by the lower and the higher court. 


Four courts had agreed that the prior inventors stopped 
short of success when they failed to cross the significant "di 
viding line between rods and filaments." When but a few 
steps from the goal they halted. Edison alone continued to 
the goal. 

Legal triumph meant immediate commercial triumph. 
Companies desiring to manufacture incandescent lamps were 
obliged to pay a license fee to the Edison Electric Light Com 
pany. But the patent, which had been so dearly upheld, had 
almost run the course of its life; it would expire in 1894. 
Hence, counsel for the Edison Company moved as rapidly 
as possible in bringing injunctions against infringers. And 
the latter, feeling the pinch of sudden stringency, became 
extremely resentful. The Edison Company was roundly con 
demned for waiting so long before bringing suit, allowing the 
infringing companies to invest large sums of money in their 
respective enterprises. 

But their complaint was well disposed of by the Circuit 
Court of Appeals in these words: 

"Every one of the manufacturing corporations, the com 
petitors of the Edison companies, commenced their opera 
tions with a knowledge of the existence of the patent in suit. 
They were controlled by business men of intelligence and 
experience. Their promoters and managers may have be 
lieved, and probably did, that the patent could not be suc 
cessfully maintained. But they entered upon the business 
with an understanding of its risks and of the consequences 
which would befall them as infringers if the patent should be 

Many of the small lamp manufacturers that had come into 
existence through the middle and late 'eighties were among 
those enjoined the Sunbeam, the Columbia, the Buckeye, 


as well as the Sawyer-Man Electric Company and its succes 
sor, the Consolidated Electric Lighting Company. 

It might be supposed that the future now looked bright 
for the Edison General Electric Company, as smoke of the 
long battle in the Federal Courts. cleared away. 

But the fight left traces of unmistakable bitterness and 
feeling of apprehension, a sense of insecurity. Electrical 
men were forced to take the patent question very much to 
heart. Were they safe in running the risk of infringing other 
patents held by rival companies? Were they safe even in 
supposing their own patents valid, in cases where those pat 
ents had not yet been tested in court? 

The Edison General Electric Company itself shared in this 
reaction. The suit had been a costly procedure. If other 
patents should be found to require adjudication, the financial 
drain might become serious, not to mention the effect upon 
the public whose own economic well-being was likewise at 
stake. The prospect, even in the face of so decided a victory, 
was not bright. 

Yet the competing men and their companies were not 
naturally belligerent. All the Edison Company sought, all 
that any of them sought, was the chance to pursue their busi 
ness activities undisturbed; to make and sell articles that they 
knew the public wanted without fear of legal complications 
either for themselves or their customers. Such was the simple 
assurance they desired; and that assurance was the last thing 
which the future appeared to offer. 

But there was a way out in consolidation. The idea was 
germinating in the minds of officials of the Edison General 
Electric Company even before Judge Wallace rendered his 
opinion in 1891. 

The foreman who completed the building of the 1903 turbine on time. 

Left to right, Peter Mulvey, Charles L. Clarke, and Ben Helm. 


The Formation 

of the 
General Electric Company 


General Electric Emerges 

TOWARD THE END OF FEBRUARY, 1891, A. L. Rohrer, assistant 
to E. W. Rice, Jr., at the Thomson-Houston factory, received 
a telephone call from Charles A. Coffin at his office in Boston. 

"Watch for the arrival at the factory of Mr. Lovejoy, ac 
companied by Mr. Henry Villard, President of the Edison 
General Electric Company," said Coffin. "Mr. Villard is to be 
shown through our plant and is to be introduced to Professor 
Thomson and Mr. Rice. Please see that his identity does not 
become generally known." 

Rohrer hung up the receiver with a tingle of excitement. 
A visit, with Coffin's sanction, from the head of this great 
competitor seemed to indicate but one thing. 

Mr. Villard was taken around the factory and shown the 
utmost courtesy. At the conclusion of his tour he departed 
with an exchange of friendly felicitations. Rice and Rohrer 
commented on the significance of all this in a guarded man 
ner, which was, nonetheless, privately jubilant. It seemed 
as if some of their weightiest problems were to be dissipated. 

Coffin was no longer the only one to realize the seriousness 
of patent deadlocks. It was evident to everyone that the 



burden of heavy license fees disheartened investors who con 
sidered buying shares in electrical companies. Natural ex 
pansion was threatened by the difficulty of enlisting new 
capital. There were irritating delays in meeting public de 
mand for electric service, requests for reliable and efficient 
electrical systems that could not be met the indefinite post 
ponement of America's complete electrical development. 

The Edison patent controlling the incandescent lamp had 
two years to run, and a two-year handicap was something 
the Thomson-Houston Company might have difficulty in 
recovering from. And there were other Edison patents the 
feeder patent, the three-wire patent without which direct- 
current service would be unsatisfactory. Thomson-Houston, 
on the other hand, held valuable transformer patents, which 
unlocked the whole field of the alternating current. 

The conviction was taking shape that the incandescent 
lamp and the alternating-current transformer system be 
longed together, just as did the overhead trolley and the 
magnetic blowout. The two were complements of each 
other. Yet they had been kept apart because the patents were 
held by rival concerns. It was found that no plant could be 
constructed and no system installed by either company with 
any hope of rendering efficient service to the public without 
infringing the rights of the other company. In many of the 
larger cities two rival electric lighting systems existed. One 
local company exploited the Edison low-tension direct- 
current method of incandescent illumination; the other oper 
ated the Thomson-Houston high-tension alternating-current 
arc-lighting circuits and series incandescent circuits. 

With affairs in this state, minds in the opposing camps 
were in a receptive mood regarding consolidation. Overtures 
came from the Edison Company before the patent suit was 
settled. They were made through a third party, Hamilton 

Manager of the Schenectady Works for many years. 

The Correspondence Department at the Schenectadij Works. 


McKay Twombley, who was chosen because of his familiarity 
with the Thomson-Houston Company. 

After the visit of Henry Villard and J. R. Lovejoy to the 
plant nothing was heard about consolidation by the operating 
officials at Lynn. Months came and went; spring passed, 
summer, autumn, and winter succeeded each other. The 
Thomson-Houston people concluded that the project had 

But conversations were going on at Twombley 's residence 
in Boston, attended by Twombley, Coffin, and Frederick P. 
Fish, counsel for the Thomson-Houston Company. Diffi 
culties had arisen over the question of terms. Coffin did not 
consider that the initial proposition submitted by Twombley 
gave adequate recognition to the record of his company, and 
he asked for a modification. 

It was nearly a year before he satisfied himself that even 
the revised proposition was suitable for recommendation to 
his directors. At length, in February 1892, he notified Twom 
bley that he was ready to accept the new terms. 

So one March morning Rice entered the factory offices and 
sought out Rohrer and Knight. They were to go at once to 
Boston to meet Mr. Coffin at his office. There they found 
Henry Villard and John Kruesi of the Edison General Elec 
tric Company. It was then announced that a consolidation 
had been agreed to, and that an exchange of visits between 
officials was to take place preliminary to the signing of the 
papers and the public announcement of the new company. 

Formalities were completed on April 15, 1892. On that day 
was incorporated the enterprise that combined the wisdom, 
traditions, and vitality of two vigorous rivals under the title 
of General Electric Company. 

Coffin had insisted upon an equality of treatment between 
the merging companies and the recognition of certain ideals 


in undertaking the new organization. He proposed that the 
new company should consider first the public which it served, 
and only secondly its own success. He proposed that the 
enterprise be organized to perform better the task of develop 
ing the cornerstone of modern industrial and economic life. 
Having that for its purpose, the bread-and-butter interest of 
the new company would never suffer. No other policy, he 
affirmed, could be regarded as "good business." 

Conferences, initiated in the late winter of 1891-92, were 
concluded nearly a year later. When terms of consolidation 
were finally drawn, they were laid before the respective 
boards of directors for ratification. The Thomson-Houston 
directors met in Boston and the Edison directors in New 
York. Few doubted that the two boards would act favorably, 
yet their action was momentous. 

So passed into one corporation two great undertakings. In 
resources and in achievements, the merging companies were 
almost equal. The Edison company was capitalized for $15,- 
000,000; Thomson-Houston for $10,400,000. The Edison 
Company reported a gross business of $10,940,000 for the 
preceding year; Thomson-Houston showed a gross of $10,- 
304,500. There were 6000 employees on the Edison rolls, 
4000 on the Thomson-Houston. The Edison Company had 
two manufacturing plants fourteen acres, forty buildings, 
and 400,000 square feet of floor space. The Thomson-Houston 
Company had one plant covering eight acres, with eleven 
buildings and 340,000 square feet of floor space. Both had 
between 3000 and 4000 customers each. The Edison Com 
pany had approximately 375 central-station companies, and 
more than 2300 isolated lighting installations. Against this 
showing Thomson-Houston reported 870 central-station com 
panies, but very few isolated plants. The Edison Company 


had equipped 180 street railways and 2230 cars; the 
Thomson-Houston 204 roads and 2760 cars. 

To most of the employees of both companies, the consoli 
dation came as a surprise, despite rumors that had been 
abroad during the preceding year. It was announced to the 
employees by circular letters posted on June 1, 1892, the date 
on which the General Electric Company began to operate in 
its own name. 

Before this Thomson-Houston men began to appear at the 
Edison works in Schenectady, and a general plan of reorgani 
zation was introduced. Thomson-Houston ideas and policies 
were adapted to existing conditions wherever they seemed 
likely to produce improvement. Otherwise the policies of the 
Edison Company continued, with considerable readjustment 
of personnel. 

Among the notable group which had created General Elec 
tric, Charles A. Coffin stood out by force of personality. Yet 
his new associates were themselves persons of large affairs 
and of high administrative caliber. 

Distinguished names appeared on the new Board of Di 
rectors. Industrialist rubbed elbows with banker, and both 
shared places with technical and business experts. The eleven 
directors were: F. L. Ames, Boston financier and business 
pioneer; Charles A. Coffin; T. Jefferson Coolidge, Boston 
banker; C. H. Coster, of Drexel, Morgan and Company; 
Thomas A. Edison; Eugene Griffin, Thomson-Houston offi 
cial; Frank S. Hastings, Edison official; Henry L. Higginson, 
of Lee, Higginson and Company; D. O. Mills, New York 
banker; J. P. Morgan; and Hamilton McKay Twombley. 

The representation of two such notable banking firms as 
Lee, Higginson and Company of Boston, and Drexel, Mor 
gan and Company of New York, later to become J. P. Morgan 


and Company, gave uncommon financial stamina to the 
board, which was recognized at the time as probably the 
strongest brought together for an American industrial con 

Yet among these exceptional men, there stood out that 
unassuming and affable New Englander who won every 
body's friendship and who thought least of all of his own ad 
vancement. To one who met him for the first time, he might 
not have appeared to be cast for an industrial chief of high 
position. But his associates knew him for what he was. With 
out hesitation they selected him unanimously as President 
of General Electric. 

It was a conspicuous position. And its prominence in 
creased with the genuine leadership of the Quaker financier, 
who twenty years before had been a salesman of shoes. 

As its Chairman the Board elected H. M. Twombley. It is 
striking to note that the Board itself included only one of 
the inventive geniuses whose achievements had disclosed 
the possibilities of electrical manufacture to which the com 
pany was now dedicated. Edison was the sole representative 
of this group. 

His position on the Board, as the champion of direct cur 
rent, may seem to have been a difficult one. Henceforth 
alternating current was to be both advocated and utilized. 
The new company would recommend for a particular in 
stallation whichever of the two types was preeminently fitted 
for the work to be done. But while Edison did not endorse 
this policy, it is certain that he concerned himself less and 
less at this period with manufacturing activities. 

It was not Edison, but Professor Thomson, who became 
the sole member of early electrical inventors to maintain 
active affiliation with General Electric. Edison retired to 
continue his independent researches. Thomson continued at 


work in his laboratory in Lynn. He declined the proffered 
honor of membership on the board of directors, declaring 
his dominant interest was to pursue laboratory work un 
burdened by other demands. 

The others of those technical pioneers in the General 
Electric Succession had gone their several ways. Charles F. 
Brush had retired, with the sale of his company, and was 
now concerned with independent laboratory operations. 
Frank J. Sprague had branched off into the development of 
electric elevators, and had organized with Edward H. John 
son the Sprague Elevator, and the Sprague Interior Conduit 
and Insulation companies. 

Charles J. Van Depoele died in the very year of this con 
solidation at the early age of forty-six. Edward M. Bentley 
had returned to the practice of law. Walter H. Knight, how 
ever, came into the consolidated company as one of the 
Thomson-Houston engineering staff, continuing in electric 
streetcar development until the middle 'nineties. 

When it was at length completed, the organization of Gen 
eral Electric showed a predominance of Thomson-Houston 
men in the corporate offices. Eugene Griffin (Thomson- 
Houston) was First Vice President. Second Vice President, 
Samuel Insull ( Edison ) , soon succeeded by J. P. Ord ( Edi 
son); General Counsel, Frederick P. Fish (Thomson- 
Houston); Secretary, E. I. Garfield (Thomson-Houston); 
Treasurer, Benjamin F. Peach, Jr. (Thomson-Houston). 

Men of tried ability, drawn from both companies, suc 
ceeded to the management of the technical and manufactur 
ing interests. E. W. Rice, Jr., (Thomson-Houston) was made 
Technical Director; John Kruesi (Edison) continued as 
Manager of the Schenectady Works; and Francis R. Upton 
(Edison) was placed in charge of the Edison Lamp Works 
at Harrison. 


In commercial organization, General Electric appropriated 
the best in the methods of both its predecessors. It estab 
lished four departments, as follows: 

Lighting Department, under S. Dana Greene (Edison); 
Railway Department, O. T. Crosby (Edison); Power De 
partment, J. R. McKee (Thomson-Houston); Supply De 
partment, J. R. Lovejoy (Thomson-Houston). 

The scientific factory-costs system of the Thomson- 
Houston Company and its functional system of factory or 
ganization were ingrafted throughout the shops of the con 
solidated company introduced gradually, accompanied by 
education of the men in the new methods and the training 
of shop foremen. 

The nationwide district office plan of the Edison company 
was made the foundation of General Electric's activities in 
the selling field. 

The personnel throughout was so strong that the prestige 
of the company was quickly established. Coffin took great 
pride in the human side of his organization. He wanted fit 
ness and loyalty; but he wanted the loyalty to be spontane 
ous. He wanted General Electric to give its workers both 
opportunity and commendation. 

The watchdog of economies in all branches of the com 
pany was the gruff but painstaking Joseph P. Ord, who 
served as Comptroller for two years before becoming a Vice 
President. His business genius was preventing waste of any 

Shortly after consolidation, a name was placed upon the 
rolls which was to assure the future of the new company as 
the exponent of the intricate alternating current. In January 
1892, Thomson and Rice had listened at a meeting of the 
American Institute of Electrical Engineers to a mathematical 
exposition which they would not soon forget. The speaker 

Millwright of the Schenectady Works for years. 


Foreman of the Edison machine shop a\ 
Menlo Park and later manager of the 
Edison Machine Works at Schenectady 


eneral Superintendent of the Edison 
Machine Works at Schenectady. 



was an odd-looking young fellow, dwarfed in stature, intel 
ligent in manner, who spoke with a foreign accent but in a 
manner that carried conviction. 

The speaker was Steinmetz, the penniless immigrant of 
1889, now, in 1892, revealing new mathematical laws to the 
acknowledged heads of American electrical engineering! 

Steinmetz had distinguished himself in the Eickemeyer 
factory by making known the mathematics of the law of 
hysteresis the law governing losses in the magnetic circuit 
of an electric motor. It was a brilliant discovery. It threw 
light into a dark corner of electrical engineering, and simpli 
fied one aspect of the art of designing efficient electric power- 
consuming apparatus. 

As it happened, Eickemeyer's work at Yonkers was under 
scrutiny at that time in the Thomson-Houston offices. His 
"iron-clad" armature and methods of armature winding had 
attracted the attention of Thomson and Rice. They discussed 
the proposition with Coffin of acquiring Eickemeyer's pat 
ents. Together with F. P. Fish, Thomson-Houston Counsel, 
they went to Yonkers to inspect the plant. On their return 
they agreed that by all odds Eickemeyer's most pronounced 
asset was the young Steinmetz. 

That summer negotiations were opened which resulted in 
the purchase by the General Electric Company of Eicke 
meyer's plant, business, and patents. 

Rice has recorded his astonishment on his first meeting 
with Steinmetz, describing his "small, frail body, surmounted 
by a large head with long hair hanging to the shoulders, 
clothed in an old cardigan jacket, cigar in mouth, sitting 
cross-legged on a laboratory worktable." This apparition be 
gins to speak, and immediately his strange appearance is 
forgotten. "I instantly felt the strange power of his piercing, 
but kindly, eyes," recalls Rice, "and as he continued, his en- 


thusiasm, his earnestness, his clear conception and marvel 
lous grasp of engineering problems convinced me that we 
had indeed made a great find. I was delighted when, with 
out a moment's hesitation, he accepted my suggestion that 
he come with us." 

Within a few months Steinmetz left Yonkers to become a 
General Electric engineer, located in the Calculating Depart 
ment at the Lynn plant, at the head of which was H. F. 

Consolidation brought no lag in engineering or sales ac 
tivities. Nearly all the engineers and sales agents of both 
companies continued under the consolidated company. Now 
they worked together instead of in opposition. Some aided 
in establishing consolidation in the field. None had a larger 
task of this sort than young Dr. Thomas Addison. 

Addison had left the medical profession to devote himself 
to volts and amperes. He was a salesman in the Thomson- 
Houston Chicago office, then headed by B. E. Sunny, who 
sent him in March 1890 to the Pacific Coast, where electricity 
was one day to go through some of its most dramatic mo 
ments. Addison opened a Thomson-Houston office in San 
Francisco the following May, and in 1892 was authorized to 
effect a consolidation in Pacific Coast territory. He took over 
a five-story building and gradually brought under his juris 
diction the commercial and construction activities which 
agents of the Edison company had been pursuing throughout 
the Pacific Coast region. 

Among these agents was Sidney Z. Mitchell, who had 
taken to the great northwest a tremendous capacity for fight 
ing against odds. Mitchell went into Seattle, a bold prophet 
of the electrical day. He wanted $10,000 for a hydroelectric 
plant, to be operated by a proposed local lighting company. 


But the people to whom he talked did not perceive the 
dawn as clearly as did Mitchell. 

Defying fate, he finally secured $10,000, with a towering 
forfeit imposed by local politicians should his electric plant 
be not running by a specified time. But Mitchell mastered 
the situation, completed the plant on time, and saw it become 
the first hydroelectric generating station on the Pacific Coast. 
Then he went up and down that country, throughout the 
great inland empire, promoting, organizing, building, and 
operating. He was an economic apostle to one town after 
another, bringing the Edison lamp to brighten the dark 
shores that bordered the Pacific. 

Meanwhile in the East, Hewlett of the New York office of 
Thomson-Houston was preparing to install new lights on 
the Statue of Liberty to celebrate, in October, 1892, the quad- 
ricentennial of Columbus' voyage. Thomson-Houston incan 
descent lights were used extensively around New York at 
that time. Hewlett had put them on several lightships, a line 
of ferry boats, in Tiffany's store, and had installed them for 
Paine's fireworks at Coney Island. 

A complaint came in one day from one of the ferryboats 
that the installation was giving trouble. Hewlett went on 
board and became immediately suspicious when his exami 
nation failed to reveal anything wrong. He made a conspicu 
ous mark on the brush holder of the dynamo, so that he 
could tell instantly if the brushes had been shifted. He 
showed the engineer what he had done, then got the purser 
to put up a cot in an empty coal bunker and stretched out 
for a nap. 

At two in the morning, after the boat had stopped running 
for the night, he went into the machinery room and looked 
at the dynamo. It was just as he had left it, and, there was 


never any more trouble with the lights. His suspicions were 
well founded the engineer, from professional jealousy, had 
tampered with the brushes on the dynamo. It was a case of 
regulating not the machine but the man. 

People living in western Massachusetts noticed during 
that summer of 1892 a long shaft of light sweeping the night 
sky from somewhere in the White Mountains. The news 
quickly got about among the rural communities: an electric 
searchlight surmounted the summit of Mt. Washington. It 
was the largest unit of the sort then known. The lens was 
thirty-six inches in diameter. The volume of illumination was 
equivalent to 20,000 candles, and this was magnified to 100,- 
000 candles by the use of a Mangin lens. The light had been 
manufactured by the Thomson-Houston plant at Middle- 
town, Connecticut. 

To get the searchlight in place was a task to daunt the 
most determined engineer. A tower fifty-five feet high was to 
be built, and it was recorded that gales of over a hundred 
miles an hour swept that rugged peak. The Superintendent 
of the Mount Washington Inclined Railway wagered the 
construction man a silk hat for each one of his thirteen work 
men that the tower would not last a year. He lost the bet, but 
never paid it. 

Watchful, wondering eyes curious eyes in most cases, 
unfriendly eyes in some observed the formation of General 
Electric. It was a "combine," exclaimed the critic, a "trust." 
It brought into existence a corporation which was gigantic 
for a period when people were alarmed about trusts. 

Coffin heard extremists attack General Electric as a mo 
nopoly with menacing possibilities, questioning every item 
in the annual financial statements over a period of several 

Perils conjured up by suspicious folk never eventuated, 


nor did the criticisms directed at Coffin and the other offi 
cials have any logical justification. The attacks displayed a 
signal misconception of the possibilities inherent in the elec 
trical business, producing a disquieting effect upon General 
Electric's finances in the great panic of 1893. 

A weekly magazine called Electricity made savage ac 
cusations against the company's bookkeeping methods and 
impugned the personal integrity of Coffin himself. Its most 
frequently repeated charge was that of overcapitalization. In 
its issue of September 7, 1892, it declared: 

"Last week we expressed our belief that the Trust was 
foredoomed to failure ... To talk of earning interest on 
this immense capitalization by honest business methods is 

Only a total inability to gauge the immediate future could 
have led to such an assertion. 

Far different were the comments of technical men, who 
have always considered that the merging of Thomson- 
Houston and Edison General Electric was a move of great 
benefit to American industry. The Schenectady Electrical 
Handbook, published by the American Institute of Electrical 
Engineers in 1904, says: 

"Never in the industrial world did organization effect a 
more magical change in releasing pent-up energy. Guided by 
master hands, electrical arts leaped into industrial pre 
eminence; the volume of manufacture of appliances, progress 
of invention, public confidence in electricity, and its general 
utilization, all took long strides forward." 

Whatever the fortunes of other electrical manufacturing 
organizations, whatever their virtues and achievements, Gen 
eral Electric set for itself certain ideals which, for forty years, 
have kept it from becoming a "soulless corporation." It had 
an idealist at its head during its first crucial years. It was 


to a large extent his vision which gave General Electric the 
momentum that produced its progress. 

From the day of amalgamation, economies were effected. 
The company found itself with two factories in which in 
candescent lamps were produced, but it was apparent that 
one factory would permit of a considerable gain in manu 
facturing efficiency. Which of the two should be discon 

The Thomson-Houston bamboo filaments were treated by 
the Sawyer process, and the patent covering this process 
had expired about the time of the consolidation. Whereupon 
John W. Howell, electrician at the Edison Lamp Works, 
visited the plant at Lynn to study the treating process. Back 
in Harrison he shut himself up on the top floor of the factory, 
started experimenting, and before long discovered a way to 
simplify the method. 

Under the original Sawyer process, a stream of gasoline 
vapor was drawn through a highly heated bulb containing 
the filament. Heat decomposed the vapor and coated the 
surface of the filament with graphite. When Howell per 
ceived that the flow of vapor was too great, causing an un 
equal cooling of the filament, he diminished the flow a little 
at a time, and with every reduction found that he obtained 
a better filament. 

While he was experimenting, Rice came out to the factory 
and announced the decision to maintain a single lamp works. 
"The factory which will continue lampmaking will be the 
one at which the best grade of lamps is produced," he said 
to Howell. "To determine which one that is, we are going to 
make a test. Your factory will make one hundred lamps, send 
fifty to Lynn for test and retain the other fifty. Lynn will do 
the same." 

Howell's discovery was in the nick of time. A month later 


Rice again visited Harrison and said to Howell, "The Lynn 
men admit that your lamps are better than theirs, so we will 
shut down the Lynn factory and the men will come down 
and work for you." 

Howell, meanwhile, kept on with his filament-treating 
experiments, until he eliminated the flow mechanism, mak 
ing the apparatus practically automatic. This was the most 
satisfactory of the filament-treated lamps. 

The first annual report of the General Electric Company 
remarks of the progress of the lamp, that: "The prices at 
which apparatus and lamps are now furnished by the Gen 
eral Electric Company for the use of such stations (licen 
sees) are, in many instances, lower than those which pre 
vailed prior to the sustaining of the lamp patent, while the 
quality and efficiency of the apparatus are largely increased. 
Thus the licensees are enabled to cheapen their production, 
and their prosperity is greatly enhanced." 

There are also to be found in this first report other signs 
of development: first, generators of larger capacities; second, 
an innovation in the manner of operating the generators. 

"On February 1, 1892," says the report, "the largest 
power-generator manufactured was of 275 horsepower. Ma 
chines of 2000 horsepower are now (February 1, 1893) be 
ing manufactured by your company. . . . The largest light 
ing generators in use on February 1, 1892, were capable of 
supplying only 2000 incandescent lights each. There are 
now being constructed generators of the direct-coupled type 
with a capacity of 12,000 incandescent lights from each 

The first use of the direct-coupled, or direct-connected, 
method was in Edison's Jumbo dynamos. But not until the 
present period were the direct-connected units started on 
their commercial career. Their birth meant the gradual pass- 


ing of the previous type, the belt-driven generator, except 
in isolated individual plants. The change resulted both in 
improved efficiency of operation and economy of space. The 
forest of flapping leather straps which met the eye in the 
larger central stations had always produced friction losses. 
These were now avoided by connecting the steam engine 
and the electric dynamo on the same shaft. 

All this had its effect upon the incandescent-lighting situ 
ation. More than 2,000,000 lamps were sold in 1892, and the 
annual sale increased steadily. No one but General Electric 
could legally market the Edison Lamp from 1892 to 1894. 

JUST BEFORE THE CONSOLIDATION an engineer in the Thomson- 
Houston Railway Department, W. B. Potter, had a conversa 
tion with W. H. Knight, the engineering head of the depart 
ment, on the subject of the magnetic blowout. There had 
been considerable groping about for a better method of con 
trolling the speed of streetcar motors during the starting of 
the car, when it was necessary to accelerate the speed 
gradually in order to prevent the sudden jerks which would 
otherwise occur. 

Technical differences had developed in the procedures of 
the Thomson-Houston and Edison companies. The streetcar 
motors of the former, which were connected in parallel, were 
controlled by a rheostat. This device inserted resistance into 
the circuit, and as the speed of the car increased, the resist 
ance was switched out a step at a time. Edison streetcar 
motors were controlled by changes made during acceleration 
in the wiring combinations of the motor's internal field 
circuit commutating the field, as it was termed. 

The Thomson-Houston men felt that a master stroke 
would be achieved if some way could be found of operating 
the motors in series during acceleration and then changing 


to parallel connection while the motors were running and 
after full speed had been attained. The difficulty was the 
excessive arc in the switch at the point where the circuit was 
broken, an arc too large for the magnetic blowout, as then 
utilized, to handle. 

Finally Potter proposed that the magnetic blowout should 
be altered so that instead of a small, intense magnetic field, 
one of large area should be sought which would draw the 
arc out to an increased length, making it more susceptible to 
interruption. Days of study followed, during which Potter 
drew a rough sketch of motors connected by the series- 
parallel method to accompany the new form of magnetic 
blowout. The moment Knight saw the sketch he exclaimed, 
"That will work let's try it." 

They proceeded to experiment. The control apparatus 
proved the most baffling factor of all, for mechanical rather 
than electrical reasons. Finally, Potter worked out a control 
ler, designated as the "J" typ e > which was an arrangement 
of plunger contacts, operated in a vertical plane through 
bevel gearing from either end of the car. It was encased in 
a long flat box and placed beneath the car body, with a con 
trol handle on each of the outside platforms. When the 
handle was moved, one set of plunger discs was raised until 
contact was made with a similar set above. The alignment 
had to be perfect, however, and careless handling crippled 
the apparatus. 

But it worked well enough on early installations to demon 
strate the superiority of series-parallel control for streetcar 
motors. The Hartford Street Railway conducted a compara 
tive trial of this method with the commutated-field method 
and the resistance method of the Edison and Westinghouse 
companies respectively. The test, concluded in April, 1892, 
on the day of the consolidation, showed a saving in electrical 


energy of twenty per cent by the series-parallel method. 
Representatives of the competing companies were consider 
ably chagrined. Potter remarked that evening to the Edison 
Company's man: 

"This test gives me a unique satisfaction. Today we have 
beaten you as competitors; tomorrow we welcome you as 

But there was trouble ahead for the "J" controller. Com 
plaints began to come in about the burning out of plunger 
contacts. The Ninth Avenue Elevated Railway in Brooklyn 
had ordered a large equipment of the controllers, but only a 
few had been installed, and they kept a repair force con 
stantly busy. 

Eugene Griffin, head of the Thomson-Houston Railway 
Department, receiving an appeal from the road, asked Potter 
if he could get the cars running again by July Fourth. Potter 
said he could. But a brief inspection of the cars in Brooklyn 
was sufficient to satisfy him that the "J" controller was a fail 
ure. He instructed the shop superintendent to take off the 
"J" controllers and replace them with some of the old resist 
ance rheostats still in stock. So the cars were ready to operate 
long before the Fourth. 

Potter went immediately to the foreman of the model room 
and said to him, "I see a crash coming on the "J" controller 
and I want to forestall it. I don't know how many men I 
shall need, but I'd like to have you put men on the work as 
fast as I can send you sketches." 

Rapidly he evolved a new form of controller. With his 
draftsmen and the model-room force working like beavers, a 
model was completed within ten days, the final trick lasting 
nearly all night. At three in the morning Potter and two shop 
men carried the model down to the test room for its first 
trial. It worked to perfection. 


Knight was skeptical when he heard the news, and told 
Potter he wanted to see the new controller after it had been 
turned on and off a thousand times with the motors running 
at full load. Potter forthwith assigned a man to carry out the 
test. He spent the best part of a day turning the controller 
on and off, on and off. He counted up to a thousand times, 
and still no injury was done. 

This sturdy mechanism was cataloged as the "K" con 
troller. It was of upright cylindrical design, permitting a si 
multaneous movement of the handle and the contacts. The 
latter met in the horizontal plane instead of the vertical. 
What made it particularly simple was the plan to mount it 
on the motorman's platform, thus eliminating over-delicate 
mechanisms which had been the undoing of its predecessor. 
The "K" controller inaugurated a new era of efficiency in 
electric transportation. 

MEANWHILE DOWN IN DIXIE LAND, a region heretofore noted 
more for agriculture than for industrial ventures, events un 
folded which were to bring stationary motors into a white 
glare of prominence. 

Until 1891 installed motors had been direct-current units, 
not above twenty horsepower in capacity. Thomson's experi 
ments with his first repulsion motor still remained a pioneer 
study of motors using alternating current. But in 1891 engi 
neers of Thomson-Houston began work on an experimental 
three-phase induction motor under the direction of H. G. 
Reist. Most of the later work on this motor was done by 
W. J. Foster and Dr. Louis Bell, the latter resigning the 
editorship of the Electrical World in 1892 to Join the staff 
of General Electric. Together they worked out the new motor 
and its generator, conducting a long series of tests. 

Two of the units were immediately sold. At the same time 


a direct-current generator at Baltic, Connecticut, and a simi 
lar machine, used as a motor in the Ponemah Textile Mills 
four miles away in Taftsville, were proving unsatisfactory. 
The cost of the heavy copper conductors required for the 
transmission line was excessive, and the motors did not oper 
ate well for constant-speed textile machinery. The new 
alternating-current generators and motors proved a god 
send, and two of each were substituted, late in 1893, at 
Baltic and Taftsville. A third installation by the Thomson- 
Houston plant of General Electric occurred in the same 
year at Redlands, California, later to be described. 

But it was in the far south, at Columbia, South Carolina, 
that there occurred the first historic installation in a textile 
mill. : ..-V 

From the beginning the Columbia mills scouted the idea 
of electricity. C. K. Oliver, agent for the mills, and Stephen 
Greene, his engineer, faced a dilemma. They wanted to em 
ploy waterwheels driven by the rapids of the Congaree 
River, 800 feet from the mills; but a canal for conveying 
river boats around the rapids interfered with the plan, since 
it ran between the river and the mills. No feasible plan of 
mechanical drive could be worked out. Under pressure of 
dire necessity Greene turned at last to electricity. 

He was besieged by salesmen from three electrical con 
cerns. One was Sidney B. Paine, who had put Edison Elec 
tric lights in more textile mills than any of his rivals. Now he 
devoted weeks to urging General Electric motors upon 
Greene. The cost of an electrical installation seemed pro 
hibitive, and Greene held out until the last shred of hope in 
some form of mechanical drive had vanished. The three 
electrical salesmen all submitted propositions, and Paine's 
bid was the highest $75,800. Greene dismissed Paine's 
proposal from the outset, for it covered alternating current, 


while his competitors advocated direct current at a much 
lower cost. 

"But direct current is not suitable for driving your machin 
ery," Paine told him. Then he compared the two types of 
electric drive, explaining that machines and lights for the 
mills would both be supplied from the same compound- 
wound direct-current generators. When the lights were 
turned on the voltage of the generators, and hence of the 
feeders supplying the motors, would be raised automatically, 
increasing the speed of the motors. Yet to run efficiently, 
the textile machines required a constant, never-changing 

While the issue hung fire, the World's Fair opened at 
Chicago. One of Paine's competitors invited Greene to at 
tend the exposition as his guest. Paine promptly said, "Very 
good; now will you return as my guest?" 

On the return trip Greene told Paine he had made up his 
mind in favor of direct current. But somewhere between 
Chicago and Columbia, Paine's arguments took effect. 
Greene was completely won over. 

Now came a struggle to fulfill the contract, which speci 
fied fourteen alternating-current motors of sixty-five horse 
power each. Thus far the largest alternating-current motors 
built by General Electric were for slightly more than ten 
horsepower. It was a tremendous jump. 

The contract, following the new routine, went before the 
sales committee and was approved. Then it went before the 
executive committee and was there disapproved. It was too 
radical too venturesome and hazardous. The committee, 
however, went to the head technical official, E. W. Rice, Jr., 
for advice. Rice promptly told them that the thing was per 
fectly feasible, and the day was saved. 

The work of designing the units fell upon Parshall, Bell, 


Steinmetz, and the Calculating Department. Parshall was 
willing to guarantee that the motors would work. The con 
tract, however, required them to be installed on the ceiling 
of the mill, to operate in an inverted position, and Bell de 
murred at this. "This is the last application of an inverted 
motor well ever make," he objected. To which Paine replied, 
"Then it's the last motor well ever put into a cotton mill, 
because there is no room on the floors of these mills for any 
more machinery." 

It was far from the last motor General Electric ever put 
into a cotton mill. For it was destined to open the eyes of 
mill owners all over the south and to precipitate an economic 
step that brought industrial wealth to whole communities. 

The generating station began to operate in April 1894, 
with two 500-kilowatt three-phase generators, water-driven. 
The transmission line was only 800 feet long. But in the size 
of the alternating-current motors, their inverted position, 
the use of both ends of the motor shafts for driving belting, 
and above all the innovation of subdividing the mill into 
sections each with its own driving motor, the installation 
was destined to influence the future equipment of all textile 

The World's Fair 

SUCCEEDING GENERATIONS astound their predecessors. Sons 
amaze fathers and are themselves dumbfounded by their 
sons. It was a brief generation that elapsed between the 
Centennial Exposition at Philadelphia in 1876 and the 
World's Fair at Chicago in 1893. And the startling contrast 
between the two was due largely to the magic of electricity. 

Spectators at the Centennial marvelled at the interminable 
lines of shafting which distributed power; at the World's 
Fair they saw scarcely a single shaft instead, they heard 
the purr of motors. At the Centennial they paid little atten 
tion to the light of flaming gas burners; at the Fair, the 
illumination was itself a dazzling spectacle. 

General Electric, though only a year old, was able to 
present at the Fair an almost perfect epitome of the elec 
trical art as it then existed. The great Electrical Building was 
entirely illuminated by the Edison lamp, and in its center 
arose a "majestic luminous column," thickly studded with 
thousands of incandescent lamps. Edison lights also illumi 
nated the Manufacturers and Liberal Arts Building, the 
"largest room in the world," and General Electric search 
lights played upon the illuminated fountains. 



Even more famous was the Intramural Railway, a complete 
electric elevated railroad encircling the exposition grounds, 
and installed by General Electric. Here was shown the new 
series-parallel controller in operation; in the power plant 
was an immense direct-current generator of 1500 kilowatts 
capacity (2000 horsepower), gloriously hailed as "the largest 
generator in operation in the world;" and in this same power 
house were the beginnings of modern switchboard develop 
ment. Parshall and the calculating department had designed 
the big generator; Hewlett and Knight were prominent in 
devising the switchboard. 

A "colossus of industry," a "tamed Titan of traction," this 
generator weighed ninety tons, while the flywheel of the 
engine weighed eighty-five tons and was twenty-four feet 
in diameter. The generator was of the multi-polar type with 
twelve magnetic poles, and the field frame was sixteen feet 
in diameter. It was a direct-connected unit, driven by a 
cross-compound engine, and had a horsepower capacity 
equal to 250 of the original Edison bi-polar machines or al 
most 30 of the Edison jumbos. 

Parshall said later, recalling the moments of anxiety while 
the unit was under construction, "The machines were so 
large and so important, I had the thought in my mind that 
if I did not get the anticipated result, my job would not be 
worth having." Many times he rose in the dead of night, 
turned on the light and went over his notes to be sure he 
had made no mistake in calculating the electrical constants. 
These constants were so vividly fixed in his mind that thirty- 
five years later he could enumerate them! 

Another transportation thrill was provided by the motor- 
driven sidewalk, 4300 feet in length, equipped by General 
Electric. Its driving power comprised twenty-four railway 

- " 



The electric locomotive demonstrates its power by outpulling two 
steam locomotives. 


A General Electric locomotive pulling a train through the Baltimore 

tunnel in 1896. 


Oil switches in the Fiske Street station of the Commonwealth Edison 



motors, each of fifteen horsepower, placed at intervals be 
neath the platform trucks. The latter, 340 in number, were 
coupled together into one long train, continuously moving in 
one direction. The sidewalk carried 6000 seated passengers 
at one time, at the rate of six miles an hour. 

On the water, General Electric had its motors in the fleet 
of fifty electric gondolas that plied the lakes and lagoons. 
The motors were of two sizes, supplied from Brush storage 
batteries. In Machinery Hall, General Electric exhibited its 
electric locomotive, its illumination of the battleship 'Illinois," 
some of professor Thomson's early experimental apparatus, 
and a pictorial story of Edison's evolution of the incandescent 

A service and repair shop were required to keep dynamos, 
motors, and lights at top efficiency. Few saw this General 
Electric shop hidden behind the scenes. Yet it harbored 
during that summer a future leader of General Electric. 

He was a student at Massachusetts Institute of Technology 
just through his sophomore year. No place in the world 
looked so alluring to him that summer as did Chicago. He 
wanted to work there, and he had no fancy notions as to 
what that work should be. And so a new name was added to 
the payroll: "G. Swope, helper, per day, $1." 

General Electric's principal competitors in that year were 
the Stanley Electric Manufacturing Company, the Siemens 
and Halske Electric Company of America, the Brush Elec 
tric Company, and the Fort Wayne Electric Company. 

The latter two were under the control of General Electric 
and classified as subsidiaries; the operation of their plants 
was directed by Rice. Yet both companies were free to com 
pete with General Electric in whatever line of products they 
were particularly strong. Brush and Fort Wayne salesmen 


continually pitted themselves against those of General Elec 
tric. Coffin and his associates saw an advantage in competi 
tion between different General Electric plants, since it 
disclosed the plant which had the lowest manufacturing 
costs for a particular product. Manufacture could then be 
concentrated at the low-cost plant and low costs meant 
low prices, more customers, and ultimately greater prosperity 
for the company as a whole. 

There was still much activity by the Brush Company in its 
own name. Arc lamps were gleaming by the hundred thou 
sand in streets, in parks and outdoor areas, and in large halls. 
Arc-light generators were brought out in increased capacities, 
with four poles instead of two and running to a maximum of 
7000 volts. To construct such a generator had been dismissed 
as totally impossible in 1877. 

There were still local Brush lighting companies, many of 
which had added incandescent lighting to their lines. Even 
in far-away Manila a Brush agent, W. S. Culver, was install 
ing a station equipped to supply 10,000 incandescent lamps 
and 300 arc lamps. 

No less active was the Fort Wayne Company. McDonald 
had remarked to James J. Wood that he considered the 
Slattery alternating-current system out of date and proposed 
that Wood design a new one. Wood immediately set to work. 
He was able and unusually diligent as a mechanic and de 
signer. His apparatus was noted for its simplicity of construc 
tion, his models had the appearance of finished machines, and 
he usually brought out a product not costly to manufacture, 
yet possessing great efficiency. 

Wood's alternator proved all that McDonald could wish 
for. Its maximum internal temperature was considerably 
lower than that of the Slattery machine. It rose to this 


temperature quickly, however, while the temperature of the 
S lattery machine rose more slowly. The Wood machine 
weighed about two-thirds less and ran at a slower speed, 
although its output was the same. 

The first central station engineer to encounter the Wood 
alternator remarked, "These little machines will never do the 
work the big ones did. And they get so hot you could fry eggs 
on 'em!" 

Wood hastened to the station and succeeded in convincing 
the engineer that the heat losses in his machine were less 
than in Slattery's. The egg-frying test was not resorted to. 

The Stanley Company at Pittsfield, Massachusetts, was 
devoting itself exclusively to alternating-current problems. 
They were first and foremost transmission engineers. Stanley 
was the trail-breaker, never defeated, never down-hearted, 
alive with ideas. Kelly, on the other hand, was phlegmatic, 
contemplative. He had unquestioned inventive talent and a 
prodigious memory, so that whenever Stanley or Chesney 
thought of something new they laid the idea before Kelly, 
who could usually tell if anything of the sort had ever been 
done before. Chesney was the detail man of the group, who 
kept the routine going the production engineer who got in 
ventions through the drafting room and shop and out upon 
the market. 

When in 1892 a delegate to the National Electric Light 
Association asserted that transformers of greater capacity 
than five kilowatts would never be practicable, the "SKC" 
men smiled, for at that moment they were building trans 
formers of twenty and thirty kilowatts capacity for com 
mercial use. 

In constructing these units, they established an efficient 
practice in the insulation of transformer coils, first by treated 


cotton cloth and later by an impregnating process, a com 
pound being forced into the coils until wires and compound 
together formed a solid mass. 

Stanley was satisfied in 1892 that he had built a transformer 
of high efficiency. His data were corroborated when the 
department of electrical engineering at Cornell University 
tested a Stanley transformer of seventeen kilowatts capacity 
and found it to be 96.9 per cent efficient. 

Transmission lines at that time usually used single-phase 
current of 133 cycles. No satisfactory motor for such circuits 
seemed to be forthcoming, and the Stanley company began 
to agitate for polyphase circuits of lower frequencies. It 
boldly designed a polyphase alternator having a stationary 
armature and a revolving field, but for either 66 or 133 cycles, 
and for any voltage up to and including 12,000. This was 
first used for long-distance transmission in 1893 at Housa- 
tonic, Massachusetts. Current was carried to Great Barring- 
ton, seven miles away. The transmission line impressive in 
length for those days was mounted on poles and insulated 
with porcelain insulators and iron pins, both imported from 
England. This was one of the first instances of polyphase 
transmission, vying with those at Taftsville and Redlands. 

Meanwhile the Siemens and Halske Company of America 
had been developing business in the Middle West and creep 
ing outward from Chicago over a constantly widening radius. 
Much electrical apparatus had been imported from Germany 
and placed on the American market. Early in 1893, however, 
the company began manufacturing at Chicago its "inpole 
dynamo," in which the field and field poles were inside the 
armature. The capacities of these machines ranged from 
about 50 to 1500 kilowatts; they were direct-connected to 
the steam engines which drove them. 

At first the personnel was almost entirely German. Later a 

Charles G. Curtis 

William Le Roy Emmet 


The first Curtis steam turbine, built in 1901 for experimental purposes. 


few Americans were taken on, and effort made to introduce 
American methods and policies. 

These, with the Westinghouse Company, were General 
Electric's competitors. All of them met on a competitive basis 
at the World's Fair in Chicago in 1893. 


Coffin Faces the Panic 

UNDER THE SOMBER CLOUD of the financial panic of 1893 
General Electric's future suddenly darkened. Throughout 
the greater part of the World's Fair, depression was deep 

The actual money stringency lasted from early in April 
until the middle of September, but it was the aftermath which 
proved hardest to bear. It was then that the pinch of hard 
times was felt by the majority of people. 

Buffeted by this economic storm, General Electric needed 
a stout heart and a steady hand at the wheel. It needed espe 
cially a calm, intrepid mind to chart a course of safety, let 
the winds howl as they might. 

The corporation had had but a year in which to gather 
strength to face the trial. Coffin, though still in the leader 
ship, was no longer dominant as he had been in the Thomson- 
Houston organization. There were new members on the 
Board of Directors, critical New York financiers, who were 
not always disposed to concede his soundness of judgment. 

One of their first proposals was the setting up of an execu 
tive committee to pass upon Coffin's plans and policies, and 



even to override him, should it seem desirable so to do. 
Under this committee were departmental committees, which 
supervised the conduct of affairs in their respective spheres 
of activity the manufacturing committee, the sales com 
mittee, the engineering committee. 

Coffin was still adjusting himself to the unfamiliar curbs 
upon his administrative powers when the panic came. Busi 
ness fell off abruptly during May and June. 

General Electric had to pay unheard-of rates of interest 
for money with which to meet its pay roll week after week. 
Few knew the burden with which Coffin struggled. Yet there 
was a problem with a more menacing outlook for the Quaker 
business man. Its roots went beyond the panic; but because 
of the panic, the life of General Electric was threatened. 

On July 31, 1893 the financial reports submitted to Coffin 
showed that the total indebtedness of the company was 
$10,028,000. This consisted of $4,446,000 in notes payable, 
$1,579,000 in accounts payable, $609,000 in dividends, and 
$3,394,000 in paper discounted under endorsement. Cash on 
hand amounted to $1,294,000, making the net indebtedness 

Direct indebtedness the company's own notes and ac 
counts payable was troublesome enough at such a time. 
But the situation was aggravated by indirect indebtedness, 
the notes of other companies, chiefly customers, which bore 
General Electric's endorsement and had been discounted 
under that endorsement for ready cash. 

These notes were eloquent of the way in which electricity 
had taken hold. Enthusiasm in starting local light and power 
companies had often outrun capacity to finance the enter 
prise. Local companies required equipment and frequently 
purchased from General Electric apparatus for which they 
could not pay in cash. Such purchases they financed by 


obtaining bank loans, then offering these loans in part pay 

General Electric, believing in the future of the electrical 
business, required only that men of integrity and ability man 
age the business, and on this condition accepted and endorsed 
their notes. This was the Coffin policy. It was his way of 
giving local companies their start and had advanced business 
in rapid strides. Yet it became a boomerang when the panic 

General Electric's endorsement made the corporation liable 
for repayment of the loans should the original borrowers ( the 
local companies) prove unable to pay them off at maturity. 
Because of the panic, the local companies proved unable to 
pay the notes, and banks began to clamor for repayment of 
their loans. 

That was the prospect General Electric faced as the sum 
mer of 1893 wore on. Having encouraged a vast economic 
movement during its incipient stages, the company found 
itself travelling a thorny path. And the end of the path, unless 
an immediate way out could be found, was bankruptcy. 

Particularly critical were the New York bankers regarding 
Coffin's policy which had been so vital in the development 
of both Thomson-Houston and General Electric. It was no 
secret that New York banking circles believed General Elec 
tric could not survive. It is even said that Twombley, Chair 
man of the Board, was openly twitted regarding his connec 
tion with the "defunct General Electric," and that this had 
something to do with his eventual resignation from the 

But Coffin did not falter when criticism of his methods 
was outspoken and sharp. He did not for a moment repudiate 
his actions or deny his responsibility for the existing situa 
tion. He stood unflinchingly and faced the issue, even though 


he is reported to have said, years later, "There were months 
that seemed like scalding centuries." 

He had one carefully considered plan, one hope of averting 
disaster. Without delay he laid it before his associates on the 
board for their scrutiny. 

The proposal was to liquidate, even at substantial loss, the 
stocks and bonds of local companies which had accumulated 
as an outcome of the same business policy and which repre 
sented millions in ultimate market value. These securities 
had come into Thomson-Houston hands in part payment for 
the purchases made by customers, and Coffin had always suc 
ceeded in converting them into ready funds through the 
United Electric Securities Company. Some of them had 
come from the old Edison company, which received them 
from its customers in payment of license fees. 

The total amount of securities was reported in the balance 
sheet of January 31, 1893, at a book value of $9,173,251, 
par value of $16,220,391. Some day these stocks and bonds 
would be of immense value. But in the great panic they could 
scarcely be sold in the open market. The most attractive of 
them would hardly have brought twenty-five cents on the 

Coffin listed the best of these stocks and bonds, a list show 
ing an approximate par value of $12,200,000, and proposed 
to offer them to the stockholders of the company on the basis 
of thirty-three and a third cents on the dollar. The stock 
holders had the opportunity to purchase the securities as 
soon as the panic passed; meanwhile they were to be de 
posited with trustees. Bankers interested in the company were 
to be asked to underwrite the eventual sale, which meant an 
immediate advance by the bankers of about $4,000,000 in 

It was a great sacrifice to make, but desperate times require 


desperate measures. Nor was the proposition coldly received. 
Lee, Higginson and Company stood behind Coffin from the 
outset and promptly agreed to the plan. 

But what would the New York bankers say? Morgan, who 
held the determining vote, was at his summer home in 
Maine. The plan was forwarded to him for scrutiny. His 
reply was anxiously awaited. For Coffin, it was one of those 
months in his life that "seemed like scalding centuries." 

Jesse R. Lovejoy, manager of the supply department, who 
knew in a measure what Coffin was passing through and 
what he hoped to accomplish, inquired one day, "Is every 
thing going through all right, Mr. Coffin?" To which Coffin 
replied, "Everything is either going through or falling 
through by tomorrow morning." 

The crisis was at hand. The silence from the Maine coast 
had grown ominous. Coffin made plans to catch the next 
train for New England to urge the plan in person. Action of 
any sort was better than the tension of waiting. 

Opening the door of his office, he nearly collided with a 
telegraph messenger. Morgan had accepted the plan without 
reservation, and another "scalding century" had come at last 
to an end. 

Thenceforth future endorsements of customers' notes were 
based on a more conservative policy. "Today," says the 
second annual report (with what a feeling of relief hidden 
between the lines! ) "no paper is under discount except such 
as it is believed will be paid by the makers; consequently the 
indirect obligations of the company are nominal rather than 

Other evidences of caution are to be found in this report 
of January 1894: 

"While the liquidation of the debt has been going on, the 
company has also readjusted its basis for sales, either to cash 


or to short credits to desirable customers. . . . It is believed 
that your company has lost little legitimate business in 
consequence of its curtailment of credit to customers. It 
intends to confine its business to this basis." 

The panic had burst upon General Electric before it be 
came a smooth-running personnel operating at maximum 
efficiency. A problem still unsettled in the spring of 1893 was 
the status of the district offices, many of which were working 
with such a degree of independence as to be almost separate 
concerns. However, the necessary reorganization was, if any 
thing, hastened by the severe lessons of the panic. It brought 
to the district offices a realization of the value of centralized 
control, of having back of them in times of stress the seasoned 
board of General Electric. 

It was only a matter of months before Coffin, with charac 
teristic tact and firmness, transformed these independent 
district offices into branches of the general office, cemented 
into the general organization and responsible to central offi 
cials. It was an arrangement that permitted efficiency and 
economy and definitely contributed to the slowly returning 
tide of prosperity. 

Meanwhile, at Fort Wayne the panic had brought develop 
ments which concerned General Electric. The Fort Wayne 
company suffered severely, and at the beginning of 1894 had 
financial obligations which it appeared unable to meet. Mc 
Donald exerted every effort to obtain the required funds, but 
with only partial success. 

Of the capital stock of $4,000,000, General Electric, the 
largest stockholder, held $1,137,000. Coffin and his associates 
advised McDonald that they favored a reorganization which 
would allow the plant to be operated on a sound basis. 
McDonald hurried east for a conference with Coffin, and on 
his return to Fort Wayne he petitioned the Superior Court 


for the appointment of receivers and the liquidation of the 
company's affairs. This petition was granted. Almost simul 
taneously, McDonald organized the Fort Wayne Electric 
Corporation, with capital stock of $1,500,000, which forth 
with contracted with the receivers to continue the business 
of the company. 

By this time the middle of 1894 business was slowly 
improving. In General Electric shops the six-day week had 
returned. The home office was transferred upstate from New 

Orders from every part of the country began to come in. 
Much of the increased business originated with the local 
companies, which were thus building up a strong and durable 
new prosperity for General Electric. To them the Coffin 
helping-hand policy, despite criticism and misgiving, had 
imparted a renewed vitality. And that policy, once the great 
panic passed into history, began to pay dividends. 


The first large Curtis steam-turbine generator being installed in the 
Fiske Street station in Chicago. 


The Fiske Street turbine soon after it had been brought back to Sche- 

nectady and mounted in Works Avenue. Left to right: E. W. Rice, Jr., 

W. L. R. Emmet, Thomas A. Edison, George F. Morrison, Charles P. 

Steinmetz, and H. F. T, Erben. 


The Development 

of Big Generating Units, 

Beginning of Hydroelectric Projects, 

Expansion of Systems 

through Transmission 

White Coal 

THE TALK AMONG ENGINEERS in 1893 and 1894 was all of 
alternating current and transmission the two developments 
that represented a new technical era. The foundations of 
General Electric's work were laid in the early experiments of 
Professor Thomson, but the influence that was to bring Gen 
eral Electric into leadership in the field was the brilliant mind 
of Steinmetz. The moment he joined the staff the mathematics 
of the law of hysteresis and of the alternating current became 
available at first hand to General Electric engineers. Into their 
midst came the master of these intricate methods to explain 
his equations in person. 

One of the trials of electrical engineering was the task of 
applying mathematics to the phenomenon of alternations, 
cycles, and phases. It was being done by slow, complicated 
methods. But the young mathematical genius had discovered 
a method so simple that the formidable difficulties of the 
alternating current were forever eliminated. It meant a sav 
ing of years, perhaps of generations, in the electrification of 
American life. 

But the Steinmetz formula was not easy to comprehend. 



Even experienced engineers had to be instructed, and they 
became students of this odd young fellow with the hunched 
shoulders, the jet-black hair, the keen eyes. 

This period of classroom engineering, begun in the old 
Calculating Department at Lynn, continued and expanded 
when the department moved to Schenectady in 1894. At the 
time of that transfer Parshall went abroad, and into his shoes 
as Head Engineer stepped Steinmetz. 

This happened at the very time when the period of power 
and light together was beginning to succeed the period of 
light. Lines already established to furnish light by night 
offered a new source of revenue if they could also furnish 
power by day. The growth of the systems was inevitable. 
But if they used direct current, the only way they could 
expand without an exorbitant investment in copper wire 
augmented by losses in the conductors was to establish a 
great number of generating stations. 

Alternating current solved the perplexity. With the in 
valuable transformer the current could be sent economically 
for long distances; lights and motors could be operated from 
the same circuit; and economical transmission meant flexible 
systems, capable of immense expansion. 

In Yonkers, a stone's throw from the Eickemeyer and 
Osterheld factory, where Steinmetz began his career, the 
Bradley Electric Power Company in 1893 was built around 
an invention which meant almost as much in the develop 
ment of the alternating current as did the transformer. 
Charles S. Bradley was one of Edison's pioneers. He had 
later developed a mechanism which permitted electrical 
energy, transmitted as alternating current, to be converted 
into direct current. Trolley lines, which invariably employed 
direct current, and the direct-current lighting lines already 
in existence, could thus become part of the unified system. 


Sprague (in white suit) demonstrates his multiple-unit control on the 
test track in Schenectady. 

or tt its 

M it IS. U u .. 
M H II SS t; .i, 
l ti tl II I J . 
Si H U n u 

> U t$ Irt ii U 



The transformation in Manhattan brought about by the electrification 

of the Grand Central Terminal. These two pictures were taken from 

almost the same spot, the upper one in 1906, the lower, down modern 

Park Avenue, in 1929. 


In his laboratory at Yonkers Bradley worked out a rotary 
converter, or synchronous converter, having a revolving field 
and a stationary armature. It was so designed that it would 
receive alternating current at one end and change it by a 
commutator into direct current, to be drawn off on a working 
circuit at the opposite end. 

After the converter, the next step was the substituting 
of polyphase for single-phase systems. Polyphase systems 
proved far better adapted to light-and-power work. General 
Electric in its polyphase service preferred the three-phase, as 
used in the first induction motors at Concord, Taftsville, and 
Columbia. But the first commercial polyphase system built 
by General Electric was at Redlands, California, and it was 
the first such system in the world. 

The problems presented at Redlands would seem amusing 
today but at that time they were gigantic. Redlands was the 
nucleus of what finally became one of the largest central- 
station service systems in the west that of the Southern 
California Edison Company. 

The town in 1895, a community of 4500 inhabitants whose 
chief occupation was growing oranges, was lighted by oil, 
except for a few private plants. Manufacturing of any sort 
was at a premium, for coal cost eleven dollars a ton. Alert 
citizens felt that electric power would insure for their town a 
bright future. In 1892 they engaged as consulting expert an 
electrical engineer, Almarian W. Decker. Decker investigated 
the field and reported favorably on a plan to locate a hydro 
electric generating station on Mill Creek, a branch of the 
Santa Ana River. He prepared specifications and corre 
sponded with four electrical manufacturing companies, in 
cluding General Electric, to secure bids for the equipment. 

Then the hesitancy and caution that were characteristic of 
the period came to the surface. It was essentially a trans- 


mission installation, and transmission was a new departure. 
The manufacturers expressed doubts. One was unwilling to 
try three-phase machines, which Decker had specified. An 
other insisted upon direct current. 

Even General Electric did not bid on generators to operate 
at 5000 volts, as Decker originally demanded. It offered to 
install two three-phase generators rated at 250 kilowatts and 
2400 volts. But it won the contract on the basis of its three- 
phase plan. 

So to that picturesque region under the snow-crowned 
Sierras went two of the first polyphase generators that Gen 
eral Electric built. 

They were trucked up the creek from Redlands by teams 
of horses and installed under the supervision of Fred Barber, 
General Electric's versatile sales agent and field engineer in 
that section. The construction work was in charge of young 
Albert C. Jewett, later to make a reputation for getting things 
done against staggering odds. The generators were driven by 
one of the first of Lester A. Pelton's impulse waterwheels. 

While the station was building, the Redlands Electric 
Light and Power Company was hustling for business. The 
Union Ice Company became its first large customer, buying 
electric power for its plant at Mentone, not far distant. So 
cheap was the cost of power that the ice company could ship 
7000 tons of ice per year to Los Angeles, at two dollars a 
ton railroad freight, and still sell ice in Los Angeles fifty cents 
a ton cheaper than anyone there could manufacture it. 
Electricity was becoming an economic revolutionist! 

On November 7, 1893 the Mill Creek Power House No. 1 
began to operate. Dr. Louis Bell undertook to synchronize the 
two motors, as specified in Decker's contract, with an acoustic 
synchronizer. Two diaphragms of sheet iron, each connected 
to a phase of the two generators, were encased in a brass 


cylinder with an opening in its center. Bell expected the 
diaphragms to vibrate in harmony when the generators were 
in unison. The men at the power house dubbed the contriv 
ance the "growler," a term which proved to be far too mild. 

As the generators hummed, the vibrating diaphragms set 
up a din. A scrambled medley of sound issued. Bell with 
tingling ears backed rapidly away from the listening hole in 
the cylinder. Neither he nor anyone else could learn by 
acoustics when the generators were in unison. Eventually 
the problem of synchronization was adjusted by other means. 

Thus the birth of transmission work for General Electric 
occurred more than three thousand miles away from Sche- 
nectady. But it did not outgrow infancy for some years, even 
in California, for three limiting factors delayed progress: the 
size and insulation of transmission cables and their protection 
from lightning. 

MEANWHILE, OBSERVE NIAGARA! Majestic, sublime and use 

For generations Niagara appealed only to esthetic instincts; 
then in almost an instant its practical aspects were realized. 
Poets, won by the beauty of the spectacle, were replaced by 
economists and engineers who saw it as a potential empire 
builder. 'Their aspirations to harness its energies formed a 
strange contrast to the early superstitions that haunted the 

Indians worshipped Niagara as an embodiment of the 
Great Spirit, offering human sacrifices to the thundering, 
mist-veiled cataract. Every year the tribe which dwelt beside 
the Onigara (Niagara) placed its fairest maiden in a birch 
canoe loaded with flowers and fruit, and pushed the little 
craft into the boiling current. 

From the riverbank the redskins watched unmoved as the 


maiden swept through the upper rapids and shot over the 
curving crest of the falls. She was a sacrifice to Onigara, a 
hostage to security and prosperity. 

Of a different breed was the white man of modern times, 
whose thought lay along another pathway. His ambitions 
were not to make war across Niagara's rushing flood, nor to 
walk tightropes, as did some of his ancestors, above its thun 
dering sweep, but to apply its giant strength to machines that 
would increase mankind's material prosperity a far differ 
ent sacrifice. 

He came not to worship Niagara, but to tame it. Niagara 
inspired him to exploits as thrilling as those of the adventurers 
of early days and awakened in him as much imagination as 
it did in the poet. 

In 1845 canals and mechanical waterwheels were first pro 
posed for utilizing this concentration of natural force. Not 
until the early 'eighties was an installation of any size made. 
By that time from 5000 to 10,000 horsepower was harnessed 
through canals and waterwheels to the machinery of a group 
of mills. 

At that time electrical men were awakening to the possibili 
ties of the giant working in their whirring dynamos. They 
were fascinated by the scheme of making Niagara turn dy 
namos, of causing Nature to pay the power bills, of capturing 
some of the "white coal" that wastefully plunged itself "from 
a rock taller than the tallest pines." 

Although they knew there was no prospect of utilizing all 
the power of Niagara, they calculated what the aggregate 
might be. Estimates ranged from four to seven million horse 
power, though Steinmetz, years later, placed it nearer nine 

Many ways for harnessing this power were studied; thou 
sands of dollars were spent. Compressed air had its propo- 


nents, as had hydraulic pipes, and a system of manila or wire 
ropes. It was even proposed to stretch a steel shaft the length 
of New York State to enable factories within reach to attach 
belts and gear wheels and secure as much power as they 

Not until 1892 were bids asked from electrical concerns. 
A power-house of three generators was specified, each of 
5,000 horsepower. It was the first large waterpower installa 
tion to be proposed and promised to add a new chapter to 
the glamor that was Niagara's. The best brains of General 
Electric went to work on the problem. S. Dana Greene han 
dled the commercial aspects. Rice, Parshall, and Steinmetz 
addressed themselves to the engineering problems. In De 
cember 1892 General Electric sent in its preliminary proposi 
tion, and submitted a final bid early in 1893. It was an 
exhaustive array of data, contained in sixty typewritten sheets 
accompanied by twelve exhibits, blueprints, study sketches, 
and photographs. It was said that of the five bids none was 
more thorough than General Electric's. 

On May 12 the Cataract Company notified the bidders 
that it had rejected the three-phase idea in favor of two- 
phase, and had decided to have its own engineers prepare 
plans and specifications for the generators. Rice, Parshall, 
and Steinmetz inspected these plans, and then prepared a 
new proposition based thereon. But they found it necessary 
from professional conviction to depart from certain tech 
nical features laid down by the Cataract technicians. 

All bids were now in, and the days of waiting began. On 
October 27, 1893, President Edward D. Adams of the Cata 
ract Company notified General Electric that the contract 
had been awarded to a competitor. 

It was one of the fortunes of war, but for General Electric 
it proved a spur to keener endeavor. Such stamina as this 


was no mean asset. For there were other laurels to be won at 
Niagara. Again General Electric leaders went to work. And 
this time they were successful. 

The proposition submitted on July 23, 1894, was accepted. 
It was a transmission job, to send power from Niagara to 
Buffalo over a line twenty-six miles long, at a potential or 
pressure of 10,000 volts. In the Niagara Falls power station 
electricity was generated at 2000 volts. This current was to 
be passed into General Electric transformers which were to 
subject it to a double process. They would raise the voltage 
from 2000 to 10,000, and at the same time change it from 
two- to three-phase. 

Into a substation at Buffalo current at 10,000 volts passed, 
after its transmission over a bare copper cable supported on 
several thousand wooden poles with porcelain insulators of 
a special type. In the substation it was stepped down by 
companion transformers to the original voltage of 2000. 

Some of the current at that voltage went direct to syn 
chronous motors, driving dynamos in an electric-light station; 
some of it supplied motors in a city pumping station at 110 
volts; some went to a general lighting system at 110 volts; 
and some to a set of rotary converters into which it entered 
as alternating current at 370 volts, emerging as direct cur 
rent at 500 volts to operate the Buffalo street-railway system. 

Here was an example of the flexibility, versatility, and 
practical usefulness of an electric service system. It was 
Niagara at work a score of miles distant from the thunder of 
the great falls. It meant a total of 30,000 horsepower in trans 
former capacity. 

Around the new powerhouse of the Cataract Construction 
Company began to rise factories on land leased to them by 
the power company to use the power it sold. The first two 
were the Pittsburgh Reduction Company, manufacturers of 


aluminum, and the Carborundum Company, manufacturers 
of carborundum. Both ordered transformers, rotary con 
verters, and special voltage regulators from General Electric. 

From 1894 on, the field work of General Electric at Niagara 
was in charge of W. L. R. Emmet, the former field engineer 
of the old Sprague Company, now engineer of General Elec- 
tric's Lighting Department, and a student of the alternating 

Nor was this the end of General Electric's work at Niagara. 
Shops at Schenectady were to contribute other and larger 
halters for the taming of the cataract, the harnessing of this 
greatest source of "white coal" in America. 

So THEY'VE GOT Mr. Edison's electric lights in the city now? 
Got 'em in all the big places, have they? New York, Boston, 
Chicago? Well, that's fine, but guess well keep on with gas 
light for a while." 

So spoke many a sturdy citizen to the wife of his bosom 
during the hard times that reigned after the panic. Electric 
lights were too expensive, thought the average man, for 
those who had to figure carefully over ways and means. A 
pretty penny to put into wires, not to mention the little bulbs 
with the red-hot hairpins! 

It was the alternating current that was soon to send the 
little hairpin-container into just such modest homes to replace 
the open gas burner that had already been replaced in the 
big houses. Hand in hand with the return to more prosperous 
days went the transition from direct to alternating current, 
reaching further and further into the homes of the common 

This was in 1894-95. The Pearl Street Station now served 
5000 customers. Electrical lines were crawling northward 
toward the Harlem River and into the Bronx. Service circuits, 


however, did not extend beyond Sixty-third Street. People 
living farther north began to agitate for light and power, so 
the New York Edison Company leased property at Fifth 
Avenue and Seventy-second Street for a substation. It was 
the task of the Stanley Company, contractors, to lay out an 
alternating-current underground transmission line at 2400 
volts between these two stations. The current at this sub 
station went into a motor-generator set, the alternating- 
current motor operating a direct-current generator. The di 
rect current was distributed upon the customers' service 

This method was later superseded, almost everywhere, by 
substituting a Bradley rotary converter for the motor- 
generator equipment, thus realizing Bradley 's prediction that 
the converter could be made the "heart of a system." The first 
use of this converter was made by General Electric in Chi 
cago in 1897. 

In nearly all its installations for central station companies 
General Electric now put in alternating current. Motors and 
lights came into use upon the same circuit and expansion to 
the lively growth of demand was made possible. Communi 
ties could be served by distant waterfalls or from generating 
stations in other towns. And all because men like Elihu 
Thomson and William Stanley had developed an ingenious 
agent called the transformer. 

Professor Thomson obtained a patent in 1887 on his dis 
covery that a pure mineral oil not only insulated, or confined, 
the voltage but also carried off the internal heat of the ma 
chine. It was not easy, however, to induce customers to em 
ploy oil-immersed transformers. They objected that the oil 
added to the cost, and constituted a fire risk should it leak 
through the case. General Electric engineers sometime in 
1893 discovered through a series of tests that the oil- 


immersed transformer could work at much greater capacities 
than transformers insulated otherwise. Thus the oil, in ad 
dition to insulating and cooling, was proved to permit 
greater operating loads; and since transformers have no 
moving parts, they lend themselves perfectly to oil insulation. 
This made possible the eventual discontinuance of air-blast 
transformers, and constituted a significant step in trans 
former development. 


The Skilled Workman Appears 

BACK IN THE SPRING of 1891, "Pop" Turner, the horny- 
handed superintendent of the old Edison Machine Works, 
saw a young chap coming between the shops with a handful 
of tools and called him into his office. 

"Lad," he said, "I want you to take charge of 10 tomorrow. 
Take off your overalls." 

The mechanic was a foreman in charge of experimental 
work in Building 10. Sometimes he knocked elbows with 
Edison when work brought the great man to the Schenectady 
shops. He could keep up with the "wizard" when it came to 
long hours and little sleep, but he was nonetheless astounded 
at Turner's words. To "take charge of 10" was a supervisory 
position of a high order, such as only master mechanics 
could aspire to hold. General foremen, then as now, did not 
wear overalls nor handle tools with their men. They were 
shop executives of intelligence and pronounced mechanical 
and technical skill. 

William Madigan obediently doffed his overalls and set 
to work. One could see that the smell of the shop, the scent 
of lubricated metal, and acrid machine grease carried a 



tang to his nostrils. "Pop" Turner knew, if "Bill" didn't, that 
he was ripe for running the principal shop of the plant. 

It was imposing to the dynamo builders of that day, Build 
ing 10. It purred and hummed from morning to night and 
was cluttered with electrical machines in all stages of manu 
facture. Madigan, who loved to see a piece of metal ma 
chined true to a hair, delved into the task of turning out fine 
work and turning it out on time. He watched fondly over the 
old Edison bi-polars and the stolid Sprague motors, and no 
less fondly over the later Titans. 

The shop had only one crane, a small affair, but a wonder 
to those who compared it with the ungainly yard derricks. 
There was a planing machine which would take a piece of 
work twenty feet long, and a ten-foot boring mill. Both were 
considered monsters of their kind. 

Bill McCool, one of the millwrights under Turner's direc 
tion, handled the placing of all machinery which went into 
the plant. McCool once needed an extra man for a few days, 
and John Miller of the carpenter's gang was detailed to lend 
a hand. Miller never again wore a carpenter's apron. He 
stepped into McCool's place when McCool went elsewhere, 
and was to move forward, little by little, into the post of 
Chief Millwright of the works. 

Miller petted every new machine that came in; he and his 
men eased them off the freight cars, coaxed them through 
doors, and swung them by derrick and crane into place. 
Sometimes they took them in piecemeal, put the intricate 
parts together, and set them running, with never a hitch or a 
miss. As much as Madigan loved to make them do good work, 
so Miller loved the machines for themselves. Soon he was 
speaking of them with a touch of pride as "my machines." 

He had been through some experiences with them, he 
and McCool. Once the Mohawk River, transformed into a 


flood by melted ice and spring rains, poured over its low 
banks and spread across the "big flat." The Edison works 
was soon under water three feet deep, which rose to the 
level of shop benches and office desks. 

Everybody went to work mopping up. Miller put in 120 
hours in one week. He and some of McCool's gang made a 
raft at the gate of the plant and poled their way among the 
shops, rigging tackle with which to lift machinery out of the 
water, to dry it and clean off the mud. Kruesi, the Works 
Manager, waded about in hip boots, his coattails dragging 
on the surface of the flood. 

Some of the office force rowed in a boat to what is now 
Building 6. H. L. Baltozer went in through a window, made 
steppingstones of bookkeepers' stools, and climbed over 
them to the safe to get out the books. 

A year later a fire broke out in Building 11, where under 
ground tubing was produced. Six hours later Rach, the Gen 
eral Foreman, Kruesi and Turner stood surveying the hot 
ruins while a crowd of smoke-grimed workmen went on 
playing streams of water into the shell that remained. 

It was at the height of a busy season for underground tub 
ing. Said Rach to the two bosses, "If you'll give me all the 
men I need, I'll be running again within a week." 

"A week!" cried Turner, "Chris, you're crazy." 

But Kruesi gave him the men and Rach went to work while 
the ruins were still ablaze. He laid out the work without a 
moment's hesitation, giving each man a particular job and 
making sure he understood it. Then they started to clear out 
the debris. 

Men went into the ruins with ropes which they attached 
to the ends of smoking beams and ragged corners of hot 
metal. Hoses were played on the beams as they were jerked 


The men worked in shifts with no cessation, day or night. 
As soon as the charred shell was cleared out, rebuilding 
began. The works lumber mill, in Building 4, hummed 
twenty-four hours a day to keep pace with construction. 
Masons, working under canvas covering because of a siege 
of rain, rebuilt the brick walls and the ovens used for baking 
insulation. Rach slept in snatches until Thursday night, when 
he collapsed. But the following Monday morning, almost a 
week to the hour, he closed a switch that gave the new shop 
its working power and production at full tide went on. 
There were orders to fill and promises to keep, new designs 
to be placed in production. 

A few months later the pall of the panic set in. But the 
skeleton organization of foremen the Madigans, the Millers, 
and the Rachs carried on. With fresh zest they entered the 
new burst of activity that began with the closing years of 
the nineteenth century. 

In the "New York Shops" other developments were com 
ing to a boil. Equipment for Niagara and Buffalo, and shortly 
after railway apparatus for New York, raised the inevitable 
problem of insulation. Not a mile of line, not a switch could 
work at high voltages without proper insulation. How Wil 
liam Cermak, the veteran porcelain maker of the old days of 
Bergmann and Company, wrestled with the problem. He 
had made good electrical porcelain in small pieces, for sock 
ets and small switches, employing what was known as the 
dry process. But insulators that would withstand 10,000 volts 
were another matter. 

Experiment after experiment was made. Scores of insula 
tors were designed and tested, only to go to pieces under less 
than 10,000 volts. Loads of broken porcelain were hauled 
from the scene of the struggle. It was a nerve-racking period. 
At length Cermak and the transmission engineers produced 


the well-known "petticoated" insulator, a succession of out 
spreading ridges, like the old-fashioned hoop skirt done in 
flounces which stood every test up to and beyond 10,000 

These insulators for the moment kept the spectre at bay. 
But it was yet to be completely vanquished. Cermak and the 
factory heads moved heaven and earth to do it. Cermak well 
understood that it was his task, that if he could not measure 
up to it the works would get someone who could. But the fac 
tory heads had not the slightest idea where such a man was to 
be found. Yet, unknown to all of them, he was in the Schenec- 
tady works at that time in the person of Edward M. Hewlett, 
whose great contribution will be discussed in later pages. 

Many a service record of which the possessor is proud was 
begun by young mechanics shaping themselves into artisans 
in those transition days. It was monotonous, winding arma 
tures and field magnets, yet it was work for skilled hands to 
do. If the winding was done carelessly, there would not be 
room on the shaft for the bearings and the commutator. 
There were nice calculations to be made to insure proper 
clearance between the fields, maintaining the air-gap, and 
to insure proper connections and the adequate number of 
turns for the machine's capacity. An Edison motor of ten 
kilowatts required over sixty pounds of wire for the windings, 
and took three men twelve hours to wind. 

About this time there appeared at the Schenectady works 
a man short in stature with black, well-trimmed beard and 
keen, pleasant eyes. He was seen going and coming in the 
office of John Kruesi, whose assistant he had been made. This 
was the same George Emmons who, in the days of the Ameri 
can Electric Company, had astonished his roommate, E. W. 
Rice, Jr., by sitting up half the night to locate an error of 
three cents in his trial balance. Rice had sent for him a few 


years later when the Thomson-Houston Company wanted a 
man to keep track of its costs. He was then appointed Factory 
Auditor, and, under General Electric, Plant Manager at 

Kruesi had laid a solid foundation for the growth of the 
plant at Schenectady. Emmons was made his assistant, and 
through the following year they gauged together the needs 
of the thriving factory and planned its future. Then Kruesi 
was appointed consulting engineer, and Emmons succeeded 
him as Works Manager. 

To every man his profession. That of George E. Emmons 
was originally the study of costs. But from this scrutiny in 
business, and especially in industrial business, Emmons 
worked out policies directed toward developing economy. 
This took him into the domain of management, where his 
executive qualities came to the fore. 

When he came to Schenectady, Emmons applied the 
Thomson-Houston cost system and functional plan of factory 
organization to their utmost. As Works Manager, he had un 
restricted authority to build slowly, soundly toward his ob 
jective an efficient factory, conducted on scientific princi 
ples of management and cost. 

Before long it was apparent among the shop force how 
well Emmons understood men, and how completely he 
represented the cooperative conception of industry. All, 
himself included, were working together for industrial 

Before long he had an extensive acquaintance among the 
men. He had a habit of greeting everyone he passed, whether 
he knew their names or not. He had a way of swiftly looking 
a man over as if to place that man in his mind; yet his glance 
seemed to carry the assurance that they were brothers in a 
common task. After awhile a quaint phrase grew up among 


the men of the plant. They alluded to Emmons as the "little 
man." It was a token of respect. Sometimes there were differ 
ences of opinion or definite disagreements between Emmons 
and the workers; but that sense of personal relationship 
between them was never lost. They called him the "little 
man" even when problems were knottiest and opinions most 
divergent. They still felt confident of fair treatment at his 
hands, and they knew that his first concern was the efficiency 
of the plant. 

That efficiency did not come easily. It meant the sub 
stitution of new methods. It meant putting factory work on a 
planned schedule, and creating a system of reports by means 
of which the factory heads could tell the daily status of a 
specific piece of work; such innovations as assigning each 
man a permanent place in the shop a practice which saved 
much confusion and valuable time. 

Good business came back to General Electric. New shops 
sprang up. Machines grew to giants, and with them grew the 
problems to be faced. Emmons and his men became the 
veritable giant killers of their time, for every problem had 
its gigantic aspect. 

But the problem that never stayed killed was that of keep 
ing manufacturing facilities abreast of commercial demands. 
The rush of orders was ceaseless. For a while neither the 
Schenectady nor the Lynn plant could handle them. 

So it happened that despite Emmons' efficiency, the Sche 
nectady works toward the middle of 1895 found itself ten 
weeks behind in deliveries, and customers became querulous. 
Every effort was made to relieve the congestion. Several lines 
of products were transferred bodily to Lynn. Langdon Gib 
son, taking up the task of Production Manager, worked out 
improvements in the shipping routine which permitted faster 



The barn on Liberty Street, Schenectady, where General Electrics Re 
search Laboratory was founded. 

The founder of the General Electric Research Laboratory. 


shipments to be made. And Emmons was on the firing line 
from early morning until after dark. 

The situation was closely watched, too, by the indefati 
gable Coffin, by Rice, now Vice President in charge of En 
gineering and Manufacturing, and by Eugene Griffin, Vice 
President in charge of Commercial Operations. Griffin's de 
partment was directly responsible for the inrush of new busi 
ness. Of the salesmen at work under him, there was none 
more zealous than the department chief. Griffin believed in 
the electrical revolution, heart and soul. He was so full of his 
subject that to meet him was to become immersed in an 
atmosphere of confidence. Few prospective customers could 
resist the contagion. Local leaders considering the organi 
zation of an electric light and power company could enter 
tain few doubts of the future after Griffin had enveloped 
them with his enthusiasm. The result was more orders, busier 
shops, a swifter pace. 

Griffin went to Coffin one day, considerably perturbed. 
"We're in a pretty fix!" he said. "The Lynn factory cannot 
turn out more than forty motors a week and I have orders 
for a hundred." 

Coffin chuckled. "I wouldn't mourn over that, although 
I appreciate your predicament and 111 see what we can do. 
I'd feel a lot more distressed if the factory were able to turn 
out a hundred motors and you had orders for only forty." 

J. R. McKee, Manager of the Power and Mining Depart 
ment, had told the sales committee recently that the possi 
bility of electric motors for operating large mills was at 
tracting more attention every day. Frequently he had im 
pressive summaries to present of industries which had turned 
to electric power, and he issued a weekly bulletin describing 
the installations. 


A second textile mill in the South, at Pelzer, South Caro 
lina, was about to undergo electrification. Even yet, the own 
ers required courage to sign a contract for electric motors 
rather than for a form of mechanical rope drive. In Charles 
ton, where most of the capital stock for the Pelzer Mills was 
subscribed, the news that the electric drive was going in 
caused the stock to drop $25 a share. The mill hands pro 
tested flatly, "The mill never will run with those little wires 
to pull it." On the day the mill started operation, a kind 
friend approached Captain Smythe, who was responsible for 
the contract, and offered his condolences on the failure of 
the electrical transmission system. "I've watched those wires 
all day," he said, "and they haven't moved yet." 

Twenty-five General Electric motors drove the spindles 
in that mill in 1895, and now one textile mill after another 
sent in orders. In one week of that year, McKee's bulletin 
reports that motors were sold for use in mines, shoe factories, 
yarn mills, tanneries, powder mills, watch factories, and even 
for blowing church organs. 

Meanwhile, the annual sales of the incandescent lamp had 
reached the six million mark, despite the fact that the Edison 
patent expired in the previous November. The directors, in 
their annual report for that year, outlined their policy as 

"Your company will chiefly rely upon the high quality of 
the lamp manufactured by it and its facilities for manu 
facturing at a low cost to maintain its commanding position 
in the lamp business, irrespective of patent control. Lamp 
prices have been greatly reduced. While the volume of this 
important part of your business will without doubt be largely 
increased in the future, the profit thereon will be less." 

This amounted to an announcement that General Electric 
intended to hold its own in competition for business by sheer 


efficiency of methods. Competition was quick to show itself. 
The independent lamp manufacturers who had been forced 
to discontinue their infringing lamps now swarmed into the 

Local light and power companies equipped by General 
Electric in 1894 aggregated nearly 1,500. They still had finan 
cial problems, and the parent company still accepted, to 
some extent, negotiable paper in lieu of full cash payments, 
especially in the form of bonds of the local companies. This 
revived the need of an agency for handling the bonds as 
they accumulated, and so there was organized in 1894 the 
Electrical Securities Corporation. It bought from General 
Electric the bonds of local companies which were not 
readily marketable and issued against them its own col 
lateral trust bonds. 

Almost completely had the memory of the fearsome days 
of 'ninety-three faded from the minds of men. The manager 
of the Metropolitan Telephone Company in New York re 
ported that more telephone messages were handled daily for 
General Electric than for any two other concerns in the city. 

About this time Coffin proposed to Griffin a trip to the 
Pacific Coast. "But think of the expense," protested Griffin. 

"I think it would be worth while," said Coffin, thought 
fully. "I should like to show you what we are doing on the 
coast, these transmission enterprises, the outlook in that 

And so a cross-continent trip, which causes so little con 
cern to business today, was carefully weighed and pondered 
by the leaders of a young industrial concern. Indeed, there 
were events of vast significance taking place on the coast, 
among the foothills of the lofty ranges where a giant's energy 
awaited the harness of the pioneer. 

Into the mountain hinterland of California went this new 


explorer, the electrical engineer, making his way into regions 
where human habitation was scant or nonexistent. Miners 
had already discovered gold in those heaved-up mountains. 
But the new explorer sought treasure of a different sort, 
though scarcely less valuable than the metal for which the 
forty-niners struggled. It was treasure as old in the ranges 
as gold, yet more difficult to extract white coal. 

This treasure, too, had its boom, when the mining camp of 
other days gave place to the bustling construction camp of 
hydraulic engineers. Invasion of the mountain solitudes 
went on without cessation. The initial project at Redlands 
was scarcely in operation when the city of Sacramento sent 
engineers to the falls of the American River, twenty-five 
miles distant, to prospect for a waterpower site. In the power 
house erected by the Folsom Falls, General Electric installed 
four generators, each of 750 kilowatts capacity, producing 
three-phase alternating current, to which the company had 
now committed itself. Over the intervening miles a trans 
mission line was built to Sacramento and energy was first 
delivered in that city on July 14, 1895, at 11,000 volts. Elec 
tric streetcars, a city-wide system of arc lights and incan 
descent lights, and a variety of motors for stationary power 
drew energy from the falls at Folsom. 

The Sacramento News proclaimed jubilantly: "The Folsom 
power is here! It came early yesterday morning, and a hun 
dred guns awakened the town, proclaiming the glad news. 
. . . The Sacramento plant is the most extensive yet put in 
place anywhere. . . . No event ever occurred in Sacramento 
that has been heralded to her greater advantage than the 
splendid industrial feat of yesterday." 

Thereupon one after another of those western white coal 
centers yielded up its energy. Competition among the elec 
trical companies was keen. General Electric, in the following 


year, built an 11,000-volt line from the Kern River fourteen 
miles to Bakersfield; another extending fourteen miles from 
the Big Cottonwood River to Salt Lake City to work at 10,- 
000 volts; and a third from the Ogden River thirty-six miles 
to Salt Lake City, at 15,000 volts. In 1897 a transmission 
project of eighty-one miles and 33,000 volts was undertaken, 
the longest commercial electric power transmission system 
and the highest voltage yet attempted. It stretched from the 
Santa Ana Canyon to Los Angeles and Pasadena. 

Triumph indeed. Yet in the midst of it the Journal of 
Electricity, under date of May 1897, noted that all was not 
yet under full control. "Not to that trustworthy servant, the 
transformer, may the present limitations of electric trans 
mission be ascribed,'' it said, "the barrier is embodied in 
the insulator alone. In addition to being a practically perfect 
nonconductor, the insulator should be wind, rain, snow, sleet, 
dust, and insect-proof. . . . Truly may it be said that un 
limited reward awaits the inventor of a perfect high-tension 

Nevertheless, General Electric struggled along with the 
insulators that were at hand and transmission and voltages 
in California crept slowly upward. The fight for the business 
grew intense. And those bold specialists, the SKC trio, shone 
in one of the daring projects of the white coal region. 

DEVELOPMENTS AT SCHENECTADY now shift our scene from 
the far west to the city of Washington, where smart naval 
uniforms color the streets and the corridors of the federal 
buildings. In the office of the Bureau of Ordnance, United 
States Navy, early in 1894, Captain W. T. Sampson, head of 
the bureau was talking to Lieutenant Bradley A. Fiske and 
an associate. Sampson suddenly inquired, "Do you think you 
could turn gun turrets by electricity?" 


Lieutenant Fiske, already a successful inventor, had his 
doubts. But the problem interested him so much that he 
accepted the assignment, established headquarters in New 
York, and began to consult with various electrical manufac 
turers. He spent some time with J. W. Kellogg, Commercial 
Manager of Marine Work for General Electric, and other 
members of the staff. These men showed him a scheme of 
motor application and control known as the Ward Leonard 
system and built by General Electric under license from its 
originator, E. Ward Leonard. Fiske decided to try the system 
and experiments were begun in adapting it to gun turrets 
on the cruiser Brooklyn, then under construction. 

After two years Lieutenant Fiske finally worked out a plan 
that made the whole scheme practical. He took out a cover 
ing patent and sold it to General Electric. 

Then came the task of convincing the ultra-conservatives 
that electricity for turning turrets was superior to steam. 
The naval constructors who were officially sent to Sche- 
nectady to examine the apparatus sent in an unfavorable re 
port. Lieutenant Fiske suggested that two of the Brooklyn's 
turrets be equipped with steam and two with electric ma 
chinery, so that the systems could be compared under iden 
tical conditions. 

The Bureau of Construction and the Bureau of Steam En 
gineering opposed the proposal. General Electric then 
offered to stand the entire cost of the installation should the 
tests prove a failure. The question reached the desk of Sec 
retary of the Navy Hilary A. Herbert, who, although quite 
impartial, was unversed in the technical phases involved. 
Accordingly General Electric sent two of its executives, 
S. Dana Greene and J. R. McKee, to take up the fight. Dana 
Greene is said virtually to have lived in Washington for 
months. McKee appealed in person to Secretary Herbert to 


permit the test to be made "so that this administration would 
go into history as the one identified with inaugurating this 
forward step." 

Finally the Secretary was won over. "On his own orders," 
writes McKee, "and against the recommendations of the 
Navy Department's bureaus" he ordered the four turrets 

The test was conducted toward the close of 1896, in the 
presence of a board of officers. In Fiske's own words, "No 
triumph could have been more complete. The forces of ultra- 
conservatism were utterly routed, and a most important step 
in the forward progress of the Navy thereby permitted. The 
. . . system . . . was adopted by the Navy and was one of 
the important reasons for the improvement in gunnery which 
afterward resulted." 

The board of officers reported of the test that: "The electric- 
controlled turrets could be turned . . . and brought to rest 
with the object previously selected between the cross hairs 
of the sighting telescope with great facility, the turret having 
a smooth and regular motion. While it was possible to arrive 
at the same result with the steam-controlled turrets, it was 
only done with considerable difficulty." 

These were fifty-horsepower motors, and McKee calls their 
adoption "the real beginning of the use of electrical apparatus 
on naval vessels," although General Electric had, a year or 
two previously, put seven-horsepower motors on a cruiser 
and three battleships for operating ammunition hoists. Gen 
eral Electric also installed the first electric motors for deck 
winches on two battleships, as well as on two grain-carrying 
vessels. But the turret equipments were functioning in ample 
time to take their share of the glory that came to the Brooklyn, 
as one of the fleet under command of former Captain 
Sampson, in the battle of Santiago less than two years later. 


The Steam Locomotive Challenged 

A NEW DAY HAD COME for the traction motor. Its years of 
trial and tribulation had vanished. Within five years of the 
struggle at Richmond it had fulfilled aspirations that would 
have dazed the staunchest railway men of a decade earlier. 
The traction motor had leaped from the streetcar tracks of 
the teeming cities to the railroad tracks of trunk-line systems, 
and was challenging even the province of the steam loco 

Before the consolidation of 1892, the Baltimore & Ohio 
Railroad was confronted by a problem which steam locomo 
tion only aggravated. In Baltimore the road was building a 
tunnel more than a mile long with an appreciable grade. A 
considerable saving in running time of trains was effected 
between New York and Washington, but the smoke and coal 
gas of the locomotives in the confined space were unendur 

In their dilemma the railroad turned to the Thomson- 
Houston Electric Company, asking if an electric locomotive 
could be built for service in the tunnel. Knight and Potter 
promptly replied in the affirmative. The railroad began 



negotiations for three such locomotives, which were eventu 
ally contracted for by the new General Electric Company. 

The engineers who had bargained to move heavy railroad 
trains by electricity were not entirely actuated by over- 
confidence. They had tried their hands at the electric locomo 
tive of thirty tons which was exhibited at the World's Fair, 
and had also designed the Cayadutta, which weighed thirty- 
five tons and became a "yarder," or light-service locomotive. 
But the big fellows for the Baltimore tunnel weighed ninety- 
six tons, and their gearless motors, four to each locomotive, 
were rated at 360 horsepower. They could make fifty miles 
an hour by themselves, and could haul thirty freight cars at 
fifteen miles an hour or ten passengers at thirty miles 
including in each case the steam locomotive, which was left 
coupled to the trains. 

These big fellows operated with never a hitch or a failure. 
Actual service began on August 4, 1895, and one "monster" 
made a name for itself by hauling forty-four loaded freight 
cars and three steam engines up the grade of the tunnel at a 
steady speed of twelve miles an hour. "Not a sputter, spark, 
or slip of wheel indicated the tremendous energy which was 
developed by the locomotive," said one of McKee's bulletins. 

Old railroad men, raised in an atmosphere of steam, 
scarcely realized what this meant. It was the formal introduc 
tion of a new locomotive power. 

The division superintendent of the Baltimore and Ohio 
could not believe the electric locomotive had a drawbar pull 
equal to that of a steam locomotive. Potter suggested that he 
apply any sort of test he desired, whereupon the superin 
tendent rode on one of the electric locomotives to see what it 
could do. 

The rule had always been followed of accelerating gradu 
ally the speed of the train in leaving Camden station, and the 


motorman was proceeding in that manner when the super 
intendent asked him why he didn't "pull her wide open." 
The motorman demurred. The superintendent told him to do 
as he was told, and the controller was turned clear around to 
full speed. A shout went up from the train crew. The sudden 
surge of power had pulled the end out of a boxcar that was 
loaded with oats. The track was knee-deep in loose grain. By 
the time the mess was cleaned up the division superintendent 
was satisfied as to the pulling power of the electric locomo 

All this time the traction motor for streetcar operation was 
by no means standing still. Toward the end of 1893 appeared 
the G-E 800 motor, which incorporated most of the merits 
and avoided most of the errors of the early motors which 
General Electric had inherited. 

One significant feature of this motor was the appearance 
of the Eickemeyer method of winding. This consisted of 
separate coils of copper bars, all alike, bent so that they could 
be inserted in the slots of the armature core until the surface 
of the core was a solid mass. This allowed for the removal 
of any individual coil, in case it became defective, so that a 
new one could replace it. The problem of insulation was also 
simplified. The process of manufacture was both accelerated 
and standardized because the coils could be made and the 
insulating tape wound upon them by machine. Farewell at 
last to the laborious hand-winding. 

Motors at this period became more workmanlike in appear 
ance. But they were still intricate. The G-E 800, for merely 
insulating purposes, contained such a variety of substances 
as Irish linen, oiled pressboard, oiled cotton, mica, asbestos, 
hickory, Japan pressboard, canvas, and tape. 

It was the G-E 800 that was placed upon the cars of the 
Metropolitan Street Railway Company in New York when 


that road finally adopted the underground conduit for its 
Lenox Avenue line in upper Manhattan. Determined to dis 
cover what the best method of motive power might be, the 
directors of the road offered a prize of $50,000 for a practical 
and economical mode of traction. They were immediately 
swamped with suggestions. Three thousand applicants sub 
mitted ideas. 

One inventor would have placed small windmills on the 
roofs of the cars; another would have harnessed the tide at 
Sandy Hook; a third wanted to hitch each car to a balloon! 
The prize was never awarded. The directors, almost pros 
trated by the multitude and variety of the proposals, decided 
to save their money, which doubtless went in 1894 toward 
paying the expense of the conduit electric system put in by 
General Electric. The five-mile line, subject to construction 
conditions peculiar to a great metropolis, cost more than 
$100,000 per mile to build. 

In Chicago, the demonstration at the World's Fair had won 
over the Metropolitan West Side Elevated Railway to the 
idea of electric instead of steam operation. General Electric 
also equipped the Lake Street Elevated, and for this worked 
out a new style of motor rated at 160 horsepower. It had a 
box frame devised by E. D. Priest to solve the difficulty of 
fitting motors of increasing size into the small space beneath 
the cars. 

In the same year, 1895, the New York, New Haven and 
Hartford Railroad wanted electric equipment for a branch 
line running to Nantasket Beach. The G-E 2000 motor, a 
heavy-duty unit, went in, and traffic increased 300 per cent. 

Fast-flying years! Old equipment disappearing as new 
came in. Potter, succeeding Knight in 1895 as engineer of 
General Electric's railway department, wanted to locate one 
of the old Van Depoele cars, with the motor on the platform 


and a chain running to the axles, to have it preserved as a 
relic. Not one of the old cars was to be found anywhere! 
They had disappeared completely, even from the carbarns 
and the scrap yards. 

"Somebody once asked, Where do the pins go?' What I'd 
like to know is, Where do the trolley cars go?' " said Potter as 
he abandoned the search. 


Campaigning for Candlepower 

IN THE OLD DAYS electric lighting stations required a great 
array of dynamos, for each dynamo supplied only about 
fifty lamps. In the larger cities, the arc light generating sta 
tions needed a battalion of dynamos. The Chicago Arc Light 
and Power Company in 1893 had fifty-six dynamos, serving 
2242 arc lamps, or about forty lamps apiece. 

One of the rotating giants which the nineties ushered in 
could equal in performance many of the old-time midgets. 
The battalion gave way now to a corporal's guard, or even a 
lone soldier. It was a time of replacement; equipments be 
came obsolete rapidly; space was conserved. 

The local companies were rapidly combining. Edison and 
Thomson-Houston licensees were merging in each local com 
munity f or two electric lighting companies in one town had 
come to be as undesirable as two telephone companies. But 
there was a delicate problem involved in these mergers: the 
question as to which company in each town should preserve 
its identity. It was natural that Charles A. Coffin wanted the 
Thomson-Houston licensees to dominate the local consolida 
tions. But he was too astute to adopt such a policy. He appre- 



elated that it was the incandescent lamp Edison's lamp 
which guaranteed the electrical future. And how potent was 
the name of Edison! From motives of justice, of sentiment, 
and of commercial value, Coffin felt that it should be perpet 
uated, so he championed the name of his former competitor. 
Soon the entire country was dotted with utilities bearing the 
name of Edison. 

So the time came when one electrical company served a 
single town, generating only alternating current. Yet for 
street lighting purposes, arc lamps on series circuits were 
essential. To change the original street-lighting systems to 
multiple circuits, corresponding with the incandescent-lamp 
circuits, meant a large investment in labor and in copper 
wire. Special circuits would also be needed if the street lights 
were to be turned on and off from the central station. Other 
wise each lamp would have to be turned on and off indi 
vidually, and the "limping lamplighters" would be recalled 
to the streets. Independent operation, so vital to the incan 
descent system, in the case of arc lamps was an actual 

Something new was needed, a regulating device to con 
trol current in the circuit after it had left the generator, some 
thing functionally like the transformer and the rotary 
converter. And it was Professor Thomson's constant-current 
transformer which met the situation. Here was another 
contrivance for making electricity its own regulator with 
greater than human precision. 

Existing series circuits, on wires that had been put in years 
ago, could now be connected to the secondary coils of the 
constant-current transformers. The series circuits were kept at 
constant current, though street lights were part of a city- 
wide electric system of multiple circuits at constant voltage. 


And the arc-light circuits could be enlarged whenever 
growth demanded it. 

The arc lamp itself was moving toward greater efficiency, 
as was the incandescent. Until 1893 the arc, although pro 
tected from the weather by a globe, had been exposed to the 
air, and the carbon electrodes had been consumed so rapidly 
that almost daily replacements were necessary. L. B. Marks, 
an illuminating engineer of New York, devised an enclosed 
arc lamp and sold his patent rights to General Electric in 
that year. The electrodes lasted for days without trimming, 
and the life of the lamp was ten times prolonged. 

The volume of illumination, however, was not so great, and 
some of the larger cities continued using the old-style open- 
arc lamps. The new style lamps, however, were designed for 
operation upon every existing type of circuit. Yearly General 
Electric put out 100,000 arc lamps, and more than half of 
them were of the enclosed type. It came eventually to be the 
most widely used of all types, and was used for illumination 
even in the larger stores. 

While the arc lamp and the alternating current were get 
ting acquainted with each other, the incandescent lamp con 
tinued to thrive. Where the arc lamp was called for by the 
thousands, the incandescent was turned out by the millions. 
The hope of the Edison Lamp Works was for machines that 
would replace the painstaking work of glass blowers. 

Edison had already made a beginning, and others were at 
work on new experiments. If bulbs could only be blown in 
moulds! Once or twice it was tried. It could be done but 
the stem still had to be sealed in by hand, and bulbs made in 
moulds were not much of a gain if they must immediately go 
through the uncertainties of hand operation. Then about 
1895 two inventors of the Buckeye Incandescent Lamp 


Company of Cleveland, Spiller and Massey, brought out a 
sealing-in machine, and Howell, at Harrison, began im 
proving it. This machine sealed the glass stem and the glass 
bulb together in an air-tight joint. Edison had worked out a 
similar machine, which was unsatisfactory because the bulbs 
were rotated in a horizontal position and repeatedly lost their 
symmetry; the Spiller and Massey machine rotated the bulb 
in a vertical position, insuring perfect symmetry. Almost at 
the same time there came an improvement in the tubulating 
machine, which pierced the top of the bulb and attached a 
glass tube through which the air could be pumped from the 
globe. A high vacuum could now be obtained more easily 
and more quickly than before. Hence bulbs blown in a 
mould, instead of free-blown, were again attempted, this time 
with success beyond all previous experience. 

The Edison Lamp Works could now supply yearly the 
million lamps that had become the demand of the American 
public. General Electric was producing almost half the num 
ber itself while independent lamp companies were turning 
out the remainder. 

In the midst of this promotion of electric incandescence, 
there appeared one man, D. McFarlan Moore, who took his 
stand as the proponent of artificial illumination by chemical 
gases. These gases became radiant when placed in a vacuum 
and made to conduct electric currents. Moore had been with 
Edison and later with General Electric until 1894, when he 
obtained the financial backing to organize a company of his 
own to experiment in the field of gaseous conduction. He re 
marked to Edison that he hoped to produce an imitation of 

"What's the matter with my light?" Edison inquired. 

"Too small, too red, and too hot," replied Moore. 

Eventually Moore produced a lamp, tubular in form like 

"Dad' VinaTs crew 

takes time out for 


One of the motors 




the mercury vapor lamp of Peter C. Hewitt, which imitated 
daylight so closely that it found immediate use for color- 
matching. But it was far from acceptable for universal light 
ing, in the home or elsewhere, because of the high voltage 
required to operate it. It was the incandescent lamp, small, 
red, and hot as it was, which multiplied its radiance in 
homes and business buildings the country over. 

The early lamps had lasted as long as 3000 hours, giving 
neither brilliant nor efficient illumination. After a few hun 
dred hours, candlepower dropped. In 1893 the Edison Lamp 
lasted for 1200 hours, but it burned at 80 per cent or more of 
its rated candlepower for scarcely more than one-fifth of its 
life. Yet the lamp consumed electric current during the 
inefficient four-fifths of its life for which the customer paid. 
He was plainly not receiving full value for his money. Illumi 
nation cost him more per candlepower than if the lamp 
lasted for a shorter period, retaining more of its efficiency 
during the entire time. 

By 1895 General Electric had produced lamps designed to 
give better light for a shorter period. With a total life of 700 
hours they gave at least 80 per cent of their rated power for 
more than half their life. As the lamps of later years con 
tinued to improve, the consumer profited by getting more 
illumination for the same amount of money. 

Perhaps the most important contribution to lamp manu 
facture at that time was the use of red phosphorus vapor in 
producing the vacuum. It was a process discovered by an 
Italian engineer, Arturo Malignani. The news that lamps 
were being exhausted without the use of mercury caused a 
stir at Harrison. The factory heads sent for John W. Howell. 
"We want this invention," they told him. "Go over there 
and buy it for us. Catch the first boat." 

Howell was about to be married, and the invitations were 


ready to be mailed. There was a hasty scratching out of dates 
and writing in of earlier ones. It was a rapidly married couple 
that caught an early boat for Italy. They were still out of 
breatn when they found the inventor, Malignani, in a tiny 
Alpine village. Howell bought the American rights to the in 
vention, stopped in England on his way home to order a 
special pump required by the new process, and returned to 
Harrison triumphant. That was the end of the mercury- 
exhaust process, and the sale of the discarded quicksilver 
paid practically the entire cost of altering manufacturing 
facilities for the phosphorus exhaust. 

This was the most important change since the treated fila 
ment. It reduced the time required to secure a vacuum from 
half an hour to less than a minute; it created a far higher 
vacuum; and of great importance, it removed the danger of 
mercury poisoning. 

Howell could not see a new apparatus without looking it 
over for possible improvements, so that before long he hit 
upon a weak point in the Malignani process. The lamp was 
connected to the exhaust pump by a glass tube, which was 
painted on the inside with the phosphorus. After the pumping 
was completed the phosphorus was vaporized, and to insure 
the vapor's passage into the bulb and not into the pump, the 
connecting tube had to be closed. This was originally done by 
a glass-blowing process, but Howell inserted a section of 
rubber tubing which could be completely closed by a pinch- 
cock pressed upon the rubber before the phosphorus was 

This eliminated one glass-blowing step, and permitted 
more exact control of the vaporizing. Finally, Howell and 
William R. Burrows devised an apparatus for performing the 
complete exhaust process upon ten lamps at a time. 

In 1894 the famed bamboo filament was replaced by the 


squirted filament. Under this procedure cotton was dissolved 
in zinc chloride, and the cellulose mass which resulted was 
squirted through a small die into alcohol or water to harden 
it. Then it was washed, dried, shaped, packed into crucibles, 
and carbonized. This type of carbon filament was more uni 
form and more homogenous in structure than the carbonized 
bamboo. It required less labor and less material to produce. 
It made longer filaments possible, and a more uniform distri 
bution of the light was obtained by making the filament more 
oval and steadying the middle loop with an anchor. 

This was the efficient lamp of the middle and late nineties 
the lamp which would produce economical illumination. 
It was a product calculated to build good will. At the time it 
appeared, the Edison central station companies, comprising 
the Association of Edison Illuminating Companies, led the 
way in urging the policy of renewing customers' lamps as 
soon as they grew dim. 

Consideration of customer needs was one of the ideals 
upon which General Electric built. It was a policy that 
Thomson, Rice, and Coffin had woven into the fabric of the 
Thomson-Houston Company and then transferred to General 
Electric. Because of this policy, a field of service, as yet un 
suspected, was brought into existence. 

THERE WAS AN OLD-TIME "EXPERT" of the Thomson-Houston 
Company named Walter D'Arcy Ryan, who was convinced 
that electric lighting was being applied by careless methods. 
No scientific study was made of lighting projects, and lights 
were put in by guesswork. Inevitably, thought Ryan, that 
sort of thing is costly. And to whom? To the users of lights 
the customers of General Electric. 

And so he conceived the idea of putting electrical illumina 
tion upon a scientific basis. Careful studies of a customer's 


premises would form the basis of recommendations for scien 
tific illumination. This would be done without consideration 
for the business which General Electric would derive from 
the installation. 

It was a new proposition. Ryan went to E. W. Rice, Jr., 
Technical Superintendent of the Company. And there his 
idea met with a friendly reception, for Rice authorized the 
necessary appropriation, though largely as an experiment. 

Thenceforward, customer needs received special study and 
expert recommendation. The objective in every installation 
was the correct volume of light from the most efficient equip 
ment. As the practice extended, the laboratory grew. Custom 
ers realized a saving in money. And General Electric was 
proving again its principle of service. 

In 1902 the laboratory, now officially designated the 
Illuminating Engineering Laboratory, was transferred to 
Schenectady, and Ryan received the title of Illuminating 
Engineer. From this laboratory, in 1903, came a concentric 
light diffuser, a type of reflector which produced evenly 
distributed rays of light throughout the illuminated area. 
New instruments were brought out for demonstrating what 
the scientific analysis and measurement of lighting could be 
made to accomplish the spectrophotometer, the luximeter, 
the luminometer, and later a line of devices for studying the 
effects of light on colored materials. 

car man, was reaching fulfillment; electricity had grown from 
youth to manhood and was fast becoming the "giant of the 
future." Day and night without a lull, energy in such volume 
as the great pioneers had never dreamed of issued from a 
multitude of rotating monsters that whirred and hummed, 
making and distributing the electric current that meant light, 


power, and heat over a spreading network of cables and wires. 

But energy uncontrolled is not only useless but perilous. 
It must be mastered after it is brought into existence. And 
this mastery involved the problem of the electric switch. The 
future of electric power depended upon the invention of 
switches capable of controlling a titanic energy. If greater 
switches could not be developed, the trend toward generating 
stations of greater capacity would be halted. 

In the midst of their early work with switchboards, Gen 
eral Electric engineers perceived this problem drawing nearer 
and taking shape. 

In the electric switch two contact points are drawn apart, 
and an arc is produced in the gap, growing longer and longer 
as the gap lengthens, until it breaks and the circuit is inter 
rupted. The procedure lasts but a fraction of a second. In 
some of the tests that Rice and Hewlett conducted, the arc 
leaped a dozen feet before it broke. They agreed that no one 
would want to handle a switch that created such formidable 
arcs. But there was one remedy possible mineral oil. If the 
arc were produced beneath the surface, the oil would 
smother it. 

The oil switch succeeded. It became the "H" type, famous 
for its perfect work and its high capacity. In the generating 
station at Ninety-sixth Street, New York, these switches 
proved a perfect complement to the big generators. Within a 
few years the Manhattan Elevated, transformed from steam to 
electric operation, installed these switches in its power house, 
as did the Interboro road and later the New York Central. 
This was the only type of switch that could handle properly 
the volume of energy now speeding forth upon the circuits 
of electric service lines. It was to go down on the books, not as 
a switch, however, but as a circuit breaker. 

Meanwhile a race of midgets was growing up to join the 


circuit breakers in their task of controlling the energy of the 
rotating giants. For potent as they are, the relays are indeed a 
race of midgets. 

The relay is a collection of metallic parts without life or 
consciousness, yet it is eternally on guard; devoid of human 
faculties, it is yet a sentinel more dependable than humans. 

Edward M. Hewlett, the premier designer of switches and 
switchboards, fathered the relay. He both fathered and 
nurtured it; and all the while urgent men were bringing new 
switchboard problems to his office, and the five major de 
partments with professional insistence dinned their needs in 
his ears and consumed the bustling energies of his little staff. 

Hewlett experimented with dashpots, with pendulums, 
with clockwork. Finally he turned to air pressure for retard 
ing the movement of the solenoid core. He undertook to 
introduce an air cushion by means of a little sack of leather. 
But he found that a special grade of leather was required, a 
leather that was both dense and flexible. 

He visited leather factories to discover the differences in 
grades and to study samples. He finally settled upon a type 
of kangaroo leather, and with this he originated the bellows 
relay. In this the plunger rod, or solenoid core, in attempting 
to rise under the magnetic influence of the solenoid, pushed 
against the bellows; it could not complete its movement and 
close the circuit contacts until the air had been pushed out of 
the bellows, a matter of several seconds. By that time, if the 
short circuit had been slight, the trouble would have passed; 
the excessive current, magnetizing the solenoid, would have 
ceased, and the plunger would have dropped back to its 
original position. 

Meanwhile, on the Pacific coast, the electrical engineer was 
tussling with the great mountain ranges, planning a hydro 
electric project among the rugged Sierras. The distance of 


transmission would be greater, and hence the voltage higher, 
than any previous system. They would have to transmit at no 
less than 40,000 volts. The organization of the Standard Elec 
tric Company was begun, and through the self-confidence 
and technical originality of John F. Kelly, the Stanley Elec 
tric Manufacturing Company finally won the contract. 

To those who said it couldn't be done, Kelly remarked that 
it depended upon the size of the transmission cables. Two 
generating stations were immediately begun, one at Colgate 
on the North Yuka River 142 miles from Oakland, the other 
at Electra (Blue Lakes) 149 miles from Oakland in the 
opposite direction. Kelly went into the mountains to the site 
of the Colgate powerhouse, lying far north in the Sacramento 
Valley. He rode along a precarious mountain trail that 
climbed and wound through defiles and skirted sheer cliffs 
the road over which all equipment and every item of supplies 
was laboriously carted to make that power-house a reality. 
It was thirty-five miles from the nearest settlement, but an 
eight hour trip. 

On the Yuka River, winding through a great gorge in the 
wooded mountainside, Kelly and his men put in their cum 
brous electrical machines. They stretched transmission lines 
upon poles that were perched on the shoulders of the moun 
tains, and fixed their insulators upon crossarms that projected 
into dizzy space. 

The insulators were a problem in themselves. They had 
been made at first either of glass or of porcelain, but later only 
of porcelain. Their function was to keep the high-tension cur 
rent from leaping off the transmission lines and through the 
supporting poles. Mechanical requirements demanded that 
the insulators should be small, strong, and of simple con 
struction; electrical requirements demanded that they be 
extremely large, complicated, and built of porcelain, a sub- 


stance particularly fragile. They had of course to be non 
conducting, and their surfaces had to be kept dry and clean. 
Yet they were perched upon a succession of poles deep in the 
recesses of the hills, out of sight and care of men. 

Nonetheless the line from Colgate crept toward the distant 
city; and to meet it there crept up from the south a twin line 
from Blue Lakes. At last they met at Oakland, 152 miles from 
Colgate and 147 miles from Blue Lakes. The former line was 
that of the Bay Counties Electric Company, the latter was 
built for the Standard Electric Company. 

Blue Lakes and Stockton, fifty miles apart, were connected 
in 1897 and operation begun, three years before that span of 
the system crept on to Oakland. The Bay Counties Company 
in 1899 began delivering electricity at 40,000 volts as far as 
Sacramento, seventy-six miles from the powerhouse, thereby 
reinforcing Sacramento's service from the Folsom plant 
which General Electric had equipped a few years before. 

General Electric engineers were constructing an eighty- 
mile line to supply Los Angeles at 33,000 volts, and they too 
experienced difficulties with the paradoxical demands of the 
porcelain insulators. But all activities in their company were 
overshadowed by events in the East which had suddenly 
come to a climax. It was a situation again involving patents, 
the immense volume of which was now underlying the 
whole electric service systems of America. 

EDISON ONCE SAID that a patent is an excuse for a fight. Yet 
the United States Patent Office has been one of the sources 
from which has sprung the greatness of America. When the 
Japanese government sent a special commissioner to Wash 
ington in 1899 to study the American patent law, he was 
asked why the Japanese desired a patent system. He replied 
that Japan had been striving to emulate the acknowledged 


great nations of the world, among them the United States. 
Then he added: 

"We said: What is it that makes the United States such a 
great nation?' We investigated and found that it was patents; 
so we will have patents." 

Back of this naive explanation there is much truth. The 
assurance of definite legal protection as a reward for con 
tributing something new and useful to society has encouraged 
innumerable inventors. Yet, stimulating progress as it un 
questionably does, the patent system has also come close at 
times to threatening it another paradox. 

In the twenty years since 1876, there had been issued 
thousands of patents covering the adaptation of electrical 
energy to the needs of man. Keen minds thought of the same 
ideas, made the same discoveries. Many controversies were 
fought out in the patent office; others went through the 

Companies could not offer perfect electrical systems unless 
they held patents covering all their equipment. As rival com 
panies frequently each held patents vital to competitors' 
systems, neither could offer the perfect system with legal 
right and safety. True, both could "muddle through" until 
the patents expired. But letters patent in the United States 
are granted for a term of seventeen years, and it was a dismal 
prospect to think of suspending progress for such a period. 

New patents came out faster than old ones expired. Litiga 
tion swelled, and executives mopped harassed brows. It was 
a situation which had tormented Charles A. Coffin as Presi 
dent of Thomson-Houston and which precipitated the 
mergers finally resulting in the General Electric Company. 
But in 1896 the ramifications were wider, and threatened 
with a more sinister aspect the public weal. For electrical 
systems were large, prosperous, and indispensable, with mil- 


lions of dollars invested in them and millions of people de 
pendent upon them for a necessity of modern life. 

Inextricably enmeshed in the patent deadlock were the 
two largest electrical manufacturers in the United States 
General Electric and the Westinghouse Electric & Manufac 
turing Company. Competitors for years in every type of 
electrical enterprise, each held a multitude of patents, many 
of them deadlocking in their effect upon the rival manufac 
turer, many of them infringed, either deliberately or uncon 
sciously, by the other company. Westinghouse held the 
patents of Sawyer-Man, Maxim, Weston, Tesla, Stanley, and 
many others; General Electric held those of Thomson, Brush, 
Edison, Sprague, Van De Poele, Bradley, and others. And 
the customers of both were doomed to share the resulting 
enormous burden. 

Coffin did not propose consolidation now. In the advanced 
and more intricate form of the industry, consolidation did 
not seem practicable. 

Instead, a working agreement was proposed, providing, in 
essence, that each of the two companies should license the 
other to manufacture under its patents. The agreement was 
signed on March 31, 1896, when a Board of Patent Control, 
representing both companies, was set up with headquarters 
in New York, a central clearinghouse for handling licensing 

This agreement insured freedom of action in the legal re 
lations of the signatories for a period of fifteen years. Instead 
of lessening commercial competition, it stimulated it, for 
both companies were now free to manufacture and sell in 
competition with each other all essential equipment required 
by their customers. The merit of each company's product 
governed the amount of trade and the range of prices which 
it could obtain and neither of them controlled the market. 


The Development of Steam Turbines, 

Start of Industrial Research and 

of Commercial and 

Financial Expansion 


Courage and Grime 

IN HIS QUIET EXECUTIVE OFFICE in New York the Quaker presi 
dent of General Electric was reviewing the year's events. It 
was nearing the close of 1896, and Coffin was preparing his 
annual report to the stockholders. 

A memorandum lay before him: in all more than 9,000 
generating machines sold their total capacity representing 
400,000 horsepower. That would contribute a pretty volume 
of illumination to the continent. Six thousand railway motors 
marketed, and 8,000 horsepower in stationary motors. In 
what new way was electricity next to work? Coffin thought 
of what his lieutenant in all things technical, E. W. Rice, Jr., 
had told him. There was a man trying to interest the Com 
pany in a new form of steam drive inner wheels with curved 
blades, to be whirled at high speeds by the force of steam. 
It was a new form of turbine, distinctly different from the 
Parsons turbine already in use in England. 

The new year was still young when a tall young man, 
quick in movement, entered Rice's office in Schenectady. He 
announced himself as Charles G. Curtis, of New York, and in 
his travelling bag he brought the plans and the description 



of his steam turbine, for which he had unsuccessfully sought 
a market in the city. 

The men to whom Curtis talked were open-minded. They 
were interested when he told them he proposed to build a 
new kind of wheel with a succession of concave steel blades, 
technically known as buckets, around its circumference; and 
to revolve this wheel by the force of steam striking against 
the blades. 

Rice had the imagination to believe that it could be done. 
He analyzed, appraised, examined drawings. And at last he 
told Curtis that General Electric would make an agreement 
with him. He was to have the best facilities of the Schenec- 
tady Works to carry on the experiments necessary to make 
the turbine practical, in return for which General Electric 
was to be allowed to purchase the patent rights in case the 
turbine proved a commercial success. 

In the factory yard a pit was excavated, in which was 
placed a small, crude model of a turbine. The revolving parts 
were metal discs to the rims of which were riveted the 
buckets. There was a boiler, and from it the steam passed into 
a steam chest. It was then shot against the blades of the 
wheels from nozzles. 

Gingerly the shop crew worked over this modest affair, 
which was to revolve at such high speed that the more fearful 
believed it would fly apart. Curtis was assisted in the work 
chiefly by John Kruesi, now Consulting Mechanical Engi 
neer. He had been assigned by Rice to report upon the prog 
ress of the experiments, making his own recommendations. 

There were trying days when the revolving discs gave 
trouble. Their dimensions had to be exact to a hair, and this 
called for shop precision of a high order. Curtis made test 
after test, with the wheels dangerously near the casing of the 


Problems sprang up one after another. When after two 
years of experimenting, Curtis had given several demonstra 
tions, Kruesi sent in his final report to Rice, and gave an ad 
verse verdict! He recommended that the tests be abandoned. 

Engineering heads conferred in considerable consterna 
tion. The tests, aside from the period of time consumed, had 
meant a large financial expenditure. Rice turned in his di 
lemma to a man who knew nothing of the turbine experi 
ments, who was disinterested and reliable in judgment. Wil 
liam LeRoy Emmet, Engineer of the Lighting Department, 
had distinguished himself by his handling of the problems 
of alternating current. Rice asked Emmet to investigate 
Curtis' turbine and study the results of the tests. 

Emmet conceded that the tests did not show a performance 
comparable with the ability of reciprocating engines, but he 
saw something in the turbine that made him think that it 
required further experimenting to bring out unsuspected 
possibilities. If turbines could be made to work efficiently, 
he reasoned, their mechanical simplicity and economy of 
manufacture would prove of vast importance. Emmet rec 
ommended that two machines be built for commercial usage, 
one of 500 kilowatts capacity and the other of 1500. 

Rice called him in for consultation. "Would you take 
charge of the design and manufacture of the trial turbines?" 
he asked. "If you would, 111 authorize their production." 

Emmet, having had no experience with turbine principles, 
said that he would. Within twenty-four hours he was down 
in the shops, immersed in the tremendous question of the 
future of the new turbine. 

What Emmet saw in Curtis' steam turbine was the prin 
ciple, in embryo, of the great steam turbines of today; wheels 
revolving upon a common shaft and jets of steam blowing 
with terrific force against their blades. There were the noz- 


zles from which the steam, starting from the steam chest 
under high pressure, was expanded to a lower pressure, to 
acquire velocity before it was applied to the blades of the 
moving wheels. There were guides for the steam, channels 
which received the jets and passed them on so that they 
struck the next wheel from exactly the same direction as their 
impact upon the preceding wheel. 

It is a tempestuous journey that the steam experiences 
inside that immense iron jacket, squat and black, which en 
closes the vitals of the modern steam turbine. Superheated 
to a high temperature and acted upon by tremendous pres 
sure, it crowds its way into the steam chest, surging irre 
sistibly toward any available outlet. Through valves into the 
nozzles it flings itself, is immediately expanded to a lower 
pressure, and hurls its energy at tremendous speed in some 
turbines more than sixteen miles a minute against the first 
revolving wheel, escaping instantly between the blades into 
the first set of stationary buckets. There it changes swiftly 
to its original angle of impact, strikes the next revolving 
wheel, and shoots into the succeeding stationary buckets. At 
the end of the first stage it is caught in a second set of nozzles, 
which expand it to still lower pressure, giving it renewed 
velocity for the increased diameters of the next wheel. In 
the far end of the turbine, at reduced pressure, it is condensed 
in a vacuum chamber. 

Emmet felt that Curtis' first turbines should be redesigned 
in order to perform efficient work. He investigated and ex 
perimented. Presently he made changes in detail. He studied 
especially the arrangement of the buckets and the proportion 
of bucket rows to the number of stages. Here was the man 
who had redesigned and reconstructed some of the early 
Sprague railway motors until they gave such service as simi- 


Despite their size, huge turbines must be built and assembled with 
unbelievably close accuracy. 


This motor, built by workmen at the Schenectady plant, was at the 
time of its construction the largest electric motor in the world. It de 
velops 22,500 horsepower. 


Building 60 at Schenectady. The size can be judged from the figures of 
workmen in middle foreground. 


lar motors on other roads could never produce. He had found 
again an opportunity to exert his peculiar genius. 

In 1900 two turbines had been jointly designed by Curtis 
and Emmet. The smaller turbine had a capacity of 500 kilo 
watts and a speed of 1800 revolutions per minute. Curtis 
designed the buckets and the nozzles; Emmet designed the 
general mechanical construction and the arrangement of 
parts; Emmet, in conjunction with H. G. Reist, designed the 
generator which the turbine was to drive. 

Shop process was slow and laborious, as the shop men for 
the most part knew nothing about turbines. The only ex 
perienced men were a crew of thirty-six mechanics, all of 
whom had been wrestling with Curtis' first small units for 
three years or more, and now formed the nucleus of a turbine- 
shop organization. 

Within a year the first Curtis-Emmet turbine was complete 
and ready for testing under service conditions. 

It came up to every expectation. A few months later the 
1500-kilowatt unit was well advanced. Emmet and Curtis 
next designed a vertical 5000 KW turbine, and offered it to 
Samuel Insull who was now president of the Commonwealth 
Edison Company of Chicago. He was building a new generat 
ing station on Fiske Street in Chicago. 

Rice had his hands full to win the Board of Directors to an 
undertaking so beset with difficulties. But he did it in the 
end. And all the while Emmet was busy with engineering 
data, which he finally took to Chicago to lay before Insull. 
Assured by Emmet that the turbine was possible, Insull 
staked his business reputation and designed his new station 
for it. 

That was in 1901. The Fiske Street Station was to begin 
operating in the fall of 1903. Emmet, confident and optimis- 


tic, hastened back to Schenectady, and the executive offices 
hummed with the news he brought. 

Bill Madigan was directing the activities of Building 10. 
Again and again Emmet came to him with turbine parts to 
be machined. And Madigan, watching the turbines develop, 
discovered points about them that appealed to him as a con 
noisseur of machines. 

The year 1903 dawned, and with it came anxiety and ap 
prehension for General Electric. Chicago began to ask when 
the first factory test of the big turbine would be made In- 
sull and his engineers wanted to be present. A tentative 
promise was given. It had to be postponed, then postponed 

A whisper went around, especially among the makers of 
reciprocating engines, that General Electric was having trou 
ble with the turbine. Perhaps their engineers had been a 
little too cocksure. 

Then Coffin, who understood the selling power of good 
will, wrote to Rice that the turbine must be built on time! 
Rice called in Emmons, the "little man" who was giving the 
Schenectady works an esprit de corps to be proud of. 

A few hours later Emmons had assembled all the factory 
superintendents in the plant. He put to them one question, 
"Who, among our general foremen, is one hundred per cent 
qualified to complete the building of that turbine on time?" 

A ballot was taken. Only one name on the list had one 
hundred per cent marked after it by every man in the meet 
ing. The name was Madigan's. 

The next day it was early in February, 1903 Emmons 
sent for Bill Madigan, and when the friendly mannered fore 
man appeared, Emmons instructed his secretary to admit no 
visitors. Madigan felt the tenseness in the air before a word 
was said. 


"Bill," said Emmons, "do you want to make a change?" 

"Which way?" said Madigan promptly. "Up or down?" 

"Up, if you can swing the job. The story is this, we're in 
hot water over that Chicago turbine. We've promised to have 
steam in her by March, and she's nowhere near ready. The 
company's reputation depends on keeping that promise. 
We've picked you to swing it. Think it over and give me 
your answer." 

"I don't need to think it over," Madigan replied. "I've been 
watching the turbine job, and I'd like to tackle it." 

The following Monday the new foremen of Building 20, 
where the big turbine lay almost finished but unassembled, 
began laying out his work. Early as it was, Emmons was there 
before him. All the resources of the plant were placed behind 
this pleasant-spoken, quick-moving master mechanic. 

The parts of the turbine were scattered all over the shop, 
and scarcely another man in the plant knew how to assemble 
them into a finished machine. Two wheels were to be ma 
chined over again because of flaws. Everyone who looked at 
those wheels shook his head and said, "Six weeks at least." 
Madigan said nothing, but as he studied the job his fighting 
spirit rose to the challenge. 

Like a general he marshalled his men, organizing their 
different operations. He kept the work going without pause. 
Repeatedly he stepped in and worked with his own hands. 
Everything he did was planned ahead. 

Emmons came down to the turbine shop day after day. 
"When will you steam her, Bill?" he asked. 

"March 7th," was the reply. 

"You're too optimistic," said Emmons. Those to whom he 
repeated this promise said, "Madigan's crazy!" 

Nevertheless word was sent to Chicago that the turbine 
would receive its factory test on March 7, and Insull made 


his plans accordingly. By that time the turbine builders were 
working at all hours in Shop 20. 

March 1 arrived. Rice came down and asked when the 
turbine would be steamed. Madigan, grimy with dirt and 
grease, replied, "On March 5!" His eyes were red from lack 
of sleep, but his head was as clear as ever, his smile as reas 
suring. The big turbine was almost assembled, a mammoth 
towering four times as high as a man. A few more days of 
intensive, unflagging work and on March 4, three days ahead 
of his first promise, Madigan told the test men to let in steam. 

The turbine performed excellently for President Insull at 
the official test of March 7th. The good name of General 
Electric had been saved! 

The turbine-generator which Samuel Insull ordered from 
General Electric required only one-tenth of the space and 
weighed one-eighth as much as the reciprocating engines it 
replaced. But Emmet's problems were not yet settled. The 
unit in the Chicago powerhouse had still to be installed. For 
months things kept going wrong. Construction men had to 
stay there and watch the turbine as it ran. They fairly lived 
in the generating station. Whenever the turbine ran well, 
General Electric stock seemed to rise; when it ran poorly, the 
stock sagged. 

But at the end of three months no more unfavorable re 
ports were received. The turbine ran steadily day and night, 
month after month. And Emmet busied himself with new 
experiments and new designs, for a clamor was going up 
from central station executives for turbines of their own. 
None were ever demanded that Madigan couldn't build, 
though he remarked years later that he never enjoyed any 
thing so much as that first "scrap with the turbines." 

Other turbines were already on the market but not yet in 
as large sizes. The DeLaval, and the Parsons, in the hands of 


Westinghouse, were from the first great competitors of the 
Curtis turbine. 

As the General Electric shops hummed in the effort to 
supply the turbine demand, its engineers improved both 
turbine and generator year after year, eventually reverting 
to the horizontal type, until the machines grew in efficiency 
beyond the rosiest dream. But their biggest selling point, in 
a period of rapid expansion, was economy in space. The 
capacity of plants could be expanded without adding a 
square inch to the buildings already in existence. More kilo 
watts without laying a brick! And the cost of a turbine- 
generator unit was one-third that for a reciprocating engine 
and generator. 

Curtis' relations with General Electric continued to be 
active, although the purchase of his patents had been con 
cluded. Under the contract finally entered into he received 
$1,500,000 for his patent rights in the invention. The contract 
permitted him to develop on his own initiative turbines for 
marine applications in all cases where the mode of propulsion 
was other than electrical. But it left General Electric free to 
put the turbines on ships if they were electrically run. The 
possibility of propelling a ship by electricity was something 
that no one as yet took seriously. But Rice believed in it, 
and he saw to it that the clause went into the contract. Most 
electrical scientists, in 1900, dismissed the idea with the re 
mark that to drive an ocean liner by electricity, the generat 
ing plant would have to be so large that it would sink the 
ship! Rice was looking fifteen years ahead in inserting that 
clause, which was one day to play a leading role in the des 
tiny of the Company. 

For the time being General Electric was busy supplying 
the light and power companies with turbines. In 1909 the 
original unit at Chicago and a duplicate unit subsequently 


installed gave place to two larger sets. These were rated at 
12,000 kilowatts, more than twice the capacity of the old 
units. Yet there was so little difference in their dimensions 
that the new machines were placed upon the original founda 
tions and supplied with steam by the original boilers. 

Six swift years had passed. None of those who had borne 
a hand in the turbulent production of the first Chicago unit 
was likely to forget the experience. Hence it was proposed 
that this first large turbine should be taken back to Schenec- 
tady, the place of its birth, and set up as a perpetual monu 
ment to an industrial victory. There the old pioneer stands 
today, on the main thoroughfare of the General Electric 
Works, just outside an enormous building where its own 
titanic successors, now far more powerful, are carefully as 
sembled and shipped. Madigan himself was there until his 

So was launched the turbine era. Meanwhile, other events 
were at work shaping the career of the turbine's guardian 

Ronald McDonald, the eccentric genius of the Fort Wayne 
Electric Corporation, died in 1898, whereupon the following 
year General Electric purchased it. 

Four years after this, in 1903, General Electric purchased 
the Stanley Electric Manufacturing Company at Pittsfield. 
The Stanley Company had brought out its inductor alterna 
tor, and its transformers had reached a point of great effi 
ciency. General Electric could not offer to its customers 
complete electrical systems without this progressive concern. 
In 1900 the Siemens and Halske Company also sold its busi 
ness to General Electric. 


Magic Comes to Manhattan 

THOSE WERE EXCITING DAYS for General Electric's commer 
cial men. Particularly exciting, perhaps, for the selling ex 
perts whom Coffin had taken into his organization from other 
lines, such men as John R. McKee, the former boot and shoe 
wholesaler of Indianapolis, William J. Clark, the erstwhile 
coal dealer and postmaster of Birmingham, Connecticut, 
and many another. Both these men had vindicated Coffin's 
judgment by rising to high departmental posts; McKee be 
came manager of the Power and Mining Department, and 
Clark of the Railway Department. 

McKee always valued his breaking-in days with the old 
Thomson-Houston Company, grubbing in the coal mines 
with young fellows fresh from college, trying to make prac 
tical the Van De Poele electric mine-digger. Now he was 
pushing the electrical idea for coal and metal mines for all 
he was worth, and motors for almost every use under the 

"We are biting into the mining business," he said to Coffin 
in the summer of 1895, and produced the latest issue of his 
sales bulletin to verify the statement. The superintendent of 



the Berwind- White Coal Mining Company, of Osceola Mills, 
Pennsylvania, announced in that publication that he had a 
mine "turning out over 11,000 tons of bituminous coal per 
month, and not a mule about the place." 

The Eureka Colliery never had a mule from the day it 
opened. A rough track, a generator, and a two-ton, ground- 
hugging electric locomotive did the pulling. The locomotive 
was like a flat box on wheels. It was small indeed for a loco 
motive ten feet long and five feet wide, and so low that 
the overhead conductor was less than thirty-six inches above 
the rail. It was dubbed the "turtle-back" because of its slop 
ing, rounded top, and it had the pulling power of many 

A few weeks later McKee was describing a power plant at 
Scott Haven from which electrical energy was supplied to 
several mines. "A central station in mining work," he re 
marked to his colleagues of the sales committee. But McKee 
had little time for waving the feathers in his cap. Sales op 
portunities poured in upon his staff. 

In those stirring, strenuous days many an epic was related 
at the meetings of the sales committee. There was the mis 
sionary motor at Middletown, Ohio, which was given only 
"tolerance room" in a bicycle factory by the skeptical own 
ers; but it was soon being "tolerated" to such an extent that 
the owners ordered eight others like it. 

There was many a tale of being marooned overnight in 
the sort of hotels that make men dream of home. W. J. Han- 
ley, who was ranging the region south of the Great Lakes 
with an energy that was to turn him one day into a district 
manager, had one such experience when a window broke 
and the snow of a blizzard whirled in over his bed all night. 
Later in the same winter in a temperature fifteen below zero 
he had sold an electrical outfit to a snowbound miller. He 


made a dash to catch a freight train in an open cutter with 
part of his pajamas wrapped around his head to keep his ears 
from freezing. He caught the freight, and clambered over 
the rocking cars in pitch darkness to ride on the locomotive, 
which carried him to his station. 

McKee's salesmen of the Power and Mining Department 
and those of S. Dana Greene, of the Lighting Department, 
thought themselves thoroughly conversant with modern 
electrical machines. Then Coffin announced the patent agree 
ment of 1896 with Westinghouse, which permitted General 
Electric to handle the Tesla type of polyphase induction 
motor. Again the commercial personnel studied charts and 
digested test reports between sales, until they were able to 
present the virtues of this new type of motor as whole 
heartedly as salesmen should. 

The motor sales went well over twelve million dollars in 
1895, and reached seventeen million dollars in 1899. A yearly 
increase in total orders of twenty per cent was not uncom 
mon. The Supply Department, under Jesse R. Lovejoy, re 
ported an annual increase running from 20 to 24 per cent, 
and incandescent lamp sales were swelling yearly by leaps 
varying from 23 to 25 per cent. Total orders in 1900 stood 
above twenty-six million dollars. 

At Lynn, where doubts had once been expressed that the 
business would occupy an entire three-story building, the 
buildings were now numbered by the score. A new plant 
had arisen upon the flats along the Saugus River. Begun in 
1892, halted by the panic of 1893, resumed in the middle 
nineties, the future River Works was outstripping even the 
original factory on Western Avenue, where the Thomson- 
Houston Company had struggled in the optimistic eighties. 

The inroads made upon the conservative railroad business 
were slower in taking shape, despite the electric locomotives 


of the Baltimore and Ohio. Then in 1898 came an oppor 
tunity. Clark and his men were soon beaming over the rail 
way outlook as were McKee's power salesmen over the sales 
of motors. They had a customer in the Hoboken Shore Road, 
which operated a two-mile freight terminal line from Wee- 
hawken to Hoboken, where the docks of the North German 
Lloyd Steamship Company were located. To handle this 
haulage a stout two-truck locomotive of 540 horsepower 
capacity had been built, with an expected speed of eight 
miles an hour under load. 

This installation, though not large, had great missionary 
value. It was close to New York, and when its trial trip took 
place railroad officials from every road that served the me 
tropolis were watching closely as it pulled its load of nearly 
300 tons. 

From the center of Manhattan Island there once arose in 
pre-electrical days a pall of sooty smoke that rolled up in 
cessantly. Its lower depths were shot through with white, 
fanlike billows of wind-blown steam. At night lurid shafts 
of firelight pierced the ascending vapors; and infernos of 
flame could be seen down in the darkness of the pit. 

Fifteen blocks through the heart of the metropolis and 
thirty or forty acres of high-priced real estate were occupied 
by this blanketed area. It seemed one of the inevitable bur 
dens of modern metropolitan life, a grimly exacted penalty. 

To this spot there came every twenty-four hours, loudly 
puffing and panting, seven hundred steam locomotives and 
30,000 persons riding in the trains they hauled. The marvel 
lous steam toilers could be seen moving up and down that 
vortex with easy effort and murky breath, hundreds of lusty 
locomotives, each one adding its hissing voice to the deafen 
ing chorus, as if exulting in the superhuman power with 
which human brains had endowed them. 


Such was the passenger yard of the Grand Central Ter 
minal at Forty-second Street as the nineteenth century 
closed. Stretching northward was its one outlet into the 
north, the Park Avenue tunnel, where passengers and crews 
endured the nuisances and dangers of smoke in the narrow 

These conditions had existed for years, and on the Man 
hattan Elevated Railway snorting little steam locomotives 
were pulling the crowded "L" trains. Each of them consumed 
its quota of bituminous coal and gave back in return not only 
useful hauling power but low-trailing clouds of smoke. For 
years attempts had been made to replace the little steam 
locomotives with electric motors. Public campaigns favoring 
the change extended back into the eighties. But in twenty 
years no day had dawned when the smoke had not filled the 
lungs of patrons. 

If the passenger yard of the Grand Central Terminal was a 
smoking pit, the Manhattan Elevated was a smoke-belching 
serpent twining the length of the island metropolis. Thus it 
remained when Sprague, last of the hopeful demonstrators, 
withdrew his electric motors in defeat. After the merger of 
1889 which added the Sprague Electric Railway and Motor 
Company to the Edison General Electric, Sprague struck 
out again for himself with an idea that had germinated in his 
mind some years before. The Sprague stationary motors were 
winning friends in Massachusetts through the agency of 
George F. Steele and the old New England Electric Com 
pany. One day Steele heard of a new type of motor which 
had given a startling demonstration of speed in lifting an 
elevator. He talked with its designer, a young Boston me 
chanic, Charles E. Pratt, and notified Sprague, who became 
interested at once in Pratt's accomplishment. Sprague and 
Pratt finally decided to join forces, and early in the nineties 


the Sprague Electric Elevator Company was organized. 

Sprague understood perfectly what electricity could do 
in the operation of elevators, from the days when his station 
ary motor had run scores of belt-driven elevators. But, in 
vading a domain already in the possession of the hydraulic 
elevator companies, he met with intense opposition. Working 
steadily against it, he obtained his first promising contract 
in 1892. The Postal Telegraph Building in New York pur 
chased six Sprague-Pratt passenger elevators. But in so doing 
they demanded a cast-iron guarantee that in case of failure 
of the electrical method the Sprague Company would install 
any hydraulic system the customer might select. 

These hydraulic systems never had a chance. Sprague ran 
four of the electric elevators at a speed of 325 feet per minute 
with a "live load" of 2500 pounds, and the other two at a 
speed of 400 feet per minute with a load of 1800 pounds. So 
came into existence the modern high-speed electric elevator, 
driven by directly connected motors, with a control system 
expressly adapted to such work. From the status of pioneer, 
Sprague rose to that of leader in the field. T. Comerford 
Martin spoke of him years later as "the father of electric 
transportation, horizontal and vertical." 

Sprague continued to experiment with the motors in the 
Postal Telegraph Building. He connected the individual 
control circuits to a master control switch so as to operate all 
the elevators simultaneously. They responded, but their 
movement was far from synchronous. For some time he pon 
dered the experiment. There was something about it that 
haunted him. Then came the swift inspiration that trains of 
cars could be operated by the inter-control, master-switch 
plan which he had tried with the elevators. 

It was the golden key to a hundred problems of traction. 
Sprague visualized the application clearly. He was able to 


design that system almost at a single sitting. It became the 
multiple-unit method of train control a method that proved 
to be the final step in the practicability of electric trains on 
elevated roads, in subways, and even on electrified main 
railroad lines. No other schemes of train operation ap 
proached it in simplicity, flexibility, and effectiveness. 

In its perfected form, multiple-unit control permitted as 
sembling into trains, with control from any point in the train, 
any number of cars each with its own motors and controllers. 
The electrical units were connected with each other and 
with master controllers by a "train line" which produced 
unison in the motive power provided by each car. The num 
ber and sequence of the cars was immaterial; they could be 
operated in any combination. 

What feats were not made possible quick control, rapid 
acceleration, fast train schedule, short intervals between 
trains, and easy adaptation of train lengths to fluctuations of 
traffic! It was the solution of operating problems in congested 
city areas, with their recurring rush hours and expanding 

Sprague thought at once of the Manhattan Elevated Rail 
way. A year later, in 1896, and again in 1897, he sent letters 
to the officials of the road describing his multiple-unit idea 
and offering to demonstrate it with their trains at his own 
financial risk. When nothing resulted from these proposals, 
Sprague went back to his elevators. 

Two years went by. Then out of a clear sky Leslie Carter, 
President of the South Side Elevated Railway of Chicago, 
known as the "alley L," wrote to Sprague asking if he would 
serve as consulting engineer in the electrification of that 
road. -ri s 

Sprague was in the midst of an ambitious elevator project; 
and he was on crutches as well. During his convalescence 


after a bad fall, John McKay, the largest stockholder of his 
company, had suggested that Sprague go to London to secure 
a tempting contract which involved half a million dollars 
a proposed elevator installation for the Central London Tube 

It was the largest installation yet attempted. There were 
to be forty-nine elevator cars. Sprague assembled data and 
prepared drawings, planning to sail the last of March. It was 
then that he heard from Leslie Carter, and the engineers of 
the Chicago road called upon him with the plans for its 

This was the opportunity Sprague had been waiting for. 
He explained his idea of multiple-unit control to the engi 
neers and to William J. Clark, manager of General Electric's 
Railway Department, who accompanied them. 

Sprague delayed his sailing for a month. His bid for the 
electrification contract he submitted purely as an individual, 
with no company back of him, and in competition with Gen 
eral Electric, Westinghouse, and the Walker Company. 
Sprague not only stood his ground with these organized con 
cerns, but entered into a lively contention to prove to Carter 
that an eighteen-mile schedule, as proposed by one of his 
competitors, was unsound engineering, and that his own rec 
ommendation of fifteen miles an hour should be adopted. He 
and Carter spent several hours one day in April talking by 
telegraph over an open wire, each standing at the elbow of 
an operator, totally oblivious of the swelling telegraph tolls. 

The day before Sprague was to sail for Europe he received 
word that he had been awarded the contract, one of the 
conditions being that he put up a $100,000 bond as a guaran 
tee of performance. Sprague wired that he would furnish the 
bond upon his return from London. 

So he departed on April 29, 1897, facing a tremendous 


task to be carried through in a conservative country, and 
leaving behind him a situation that bristled with problems. 

In London Sprague felt as if he were trying to straddle the 
Atlantic, with one foot in England and the other in the 
United States. He must work fast with his London prospect, 
yet he had arrived in the midst of the Whitsuntide holidays. 
He must protect his interest in the execution of the Chicago 
contract, yet all he could do was to cable instructions and 
rely upon the judgment of his assistant, McKay, who was 
to sign the contract under power of attorney. 

It did not ease his mind that one of the stipulations in this 
contract was the guarantee of a demonstration of multiple- 
unit control by July 15th. It was then the first of May. 
Multiple-unit control existed only on paper, and he had no 
idea how long he would be detained in London. 

One encouraging thing happened during those first two 
weeks with the arrival of William J. Clark, sent by General 
Electric to get the sub-contract for the 240 electric motors to 
be installed on the South Side Elevated. After Clark got his 
contract, Sprague obtained through him the use of General 
Electric's experimental railway track at Schenectady. 
Sprague had assumed the obligation of providing the track, 
the railway furnishing the cars. 

Meanwhile things were dragging in London. Sprague had 
a full-size controller equipment and a typical elevator car 
safety mechanism set up in the basement of the Cecil Hotel 
for demonstrations, and had submitted his bid for the con 
tract. But he could not get the Central London Tube Rail 
way to come to a decision. He guaranteed that the cost of 
operating the elevators would not exceed one pound per 
thousand single trips of 67 feet rise, regardless of the size of 
the cars. But still they hesitated. 

Time was passing and he had to get back to the United 


States. In a last effort to clinch the matter, he offered to pro 
ceed with the contract staking all his prospects in the London 
elevator project unreservedly upon the performance of trial 
elevators in one shaft, Sir Benjamin Baker, the railway's en 
gineer, to be the judge. Upon this basis he was awarded the 

Sprague sailed for New York on June 16th, landing three 
weeks before the date on which he was required to demon 
strate multiple-unit control. He found that some beginning 
had been made, guided by what meager instructions he had 
been able to send by letter and cable. But a colossal task 
remained, and no one but a man with driving willpower, un 
sparing of himself, could hope to swing it. 

Sprague did not spare himself; and he did not hesitate to 
spend money. He worked day and night, and on July 15 he 
was in Schenectady with multiple-unit control installed upon 
two cars of the South Side Elevated Railway. He operated 
those two cars himself the following day, and saw his idea 
admirably vindicated. 

Within a year after the tests at Schenectady, the South 
Side road was fully equipped and multiple-unit control was 
in service on its first installation. Almost at the same time, 
Edward H. Johnson sought another business alliance with 
his former associate, and early in 1899 the Sprague Electric 
Elevator Company merged with the Interior Conduit and 
Insulation Company, which Johnson had organized, to form 
the Sprague Electric Company. 

Fresh from his success on the Chicago line, Sprague turned 
again to the New York situation. He had long advocated 
before the Rapid Transit Commission an electric under 
ground transportation system for the city. Now he renewed 
his campaign. He also wrote repeatedly to the officials of 
the Manhattan Elevated, and published a number of news- 


Dr. Whitney, Dr. Coolidge, and Dr. Langmuir, of the General Electric 
Research Laboratory. 

The General Electric plant at Nela Park, Cleveland, Ohio. 

Airplane view of the Schenectady Works 


paper interviews describing the virtues of multiple-unit con 
trol. It seemed to be effort wasted. There was no interest 
from 1897 to 1900. 

In another quarter, however, the importance of Sprague's 
system was more fully appreciated. Electric train control was 
an issue with the General Electric engineers. They were 
working on various ideas, and believed they had found the 
most reliable method in a contactor form of the series-parallel 
control system, which embodied the work of Elihu Thomson, 
William B. Potter, and Frank E. Case. After seeing Sprague's 
tests of 1897, they considered the application of their con 
tactor type of apparatus to Sprague's multiple-unit system. 

When the Manhattan Elevated finally began planning 
electrification in 1900, negotiations had been begun by Coffin 
for the acquiring of Sprague's multiple-unit patent. These 
were completed in 1903, and General Electric acquired the 
Sprague Company; the contract for the electrification of the 
Manhattan Elevated had not yet been closed. The question 
of what the road should pay for the equipment resulted in a 
deadlock, and the commercial men could do no more. They 
perceived that a master hand was needed, a final, diplomatic, 
understanding word. When the situation was presented to 
Coffin, he went alone to the office of the Manhattan Elevated 
Railway to speak the word that no one else could speak. He 
brought back with him one of the largest contracts in months, 
representing more than a million dollars and involving the 
initial installation of nearly 1700 motors. 

Electric service on the road began on December 20, 1900, 
and the first ticket for passage on an electric train in New 
York was sold at 11 o'clock that morning to a woman at the 
92nd Street Station. 

Meanwhile the situation at the Grand Central Terminal 
was approaching a climax. As trains fought their way through 


the smoke of the tunnel, the firemen in the locomotives were 
obliged to lie on the floor of the cabs and peer beneath the 
swirling vapors to read the signals and guide the engineers. 
Sentiment was so strongly awakened that the city authorities 
adopted an ordinance permanently banning the operation 
of trains into the city by steam power. Thereupon the New 
York Central appointed an electrical commission to under 
take a study of electric power. And the commission discov 
ered that electrification would mean a saving of hundreds 
of thousands of dollars every year! 

A project was launched for making over the great terminal 
station at a cost of some forty million dollars. The station was 
to accommodate 200,000 passengers a day instead of 30,000; 
the great passenger yard would handle 3000 trains a day 
instead of 700. 

And what was to become of the smoking pit, with its dark 
vapor-cloud? It was to be concealed from sight beneath the 
city. Above it, forty acres of real estate would be created in 
the heart of the city that was hungry for more space. 

By 1908 the program was under way. General Electric 
locomotives hauled trains through the northward tunnel in 
clear, clean air, and for thirty miles to Croton, and twenty- 
five to White Plains. Originally there were thirty of these 
clean, quiet, but no less powerful locomotives. Designed by 
Asa F. Batchelder, the motor armatures were mounted gear- 
less upon the axles and the field magnets embodied in the 
mechanical structure of the locomotive. This line, built in 
the Schenectady shops, was introduced by historic old 
"6000." Each unit weighed ninety-four tons complete and 
had four driving motors. They were the first gearless loco 
motives of General Electric make since those designed for 
the Baltimore and Ohio Tunnel, but were more sturdy and 
simple in form. Pulling with 2800 horsepower, they were as 


powerful as the steam monsters they replaced. By means of 
the multiple-unit system two of them could be coupled to 
gether as a single unit of double power, and as the Electrical 
World put it, could "dally lightly with the biggest trains the 
Central system has ever known." 

They drew their power from steam turbine-generators in 
a distant powerhouse. Eight turbine-generators, 60,000 horse 
power all told, supplied the energy. And it was there, where 
they were not a menace, that the smoky vapors now rolled 

The Interborough Rapid Transit Company soon followed 
the Manhattan Elevated in determining upon electrification. 
General Electric designed a new type of motor expressly for 
use on this road and sent two trial units to appear in the 
competitive tests. After they had been delivered, it was dis 
covered that a certain change in the internal design of the 
motor would greatly increase its efficiency. To conceal from 
the competitors what was in the wind, one of the two motors 
already installed on an Interborough car was changed over 
night. When the comparative trials took place, this motor 
out-performed any of the others, and General Electric was 
awarded half of the entire motor order in addition to the 
whole order for control equipment. 


A Venture Into Research 

THERE HAD BEEN A PROBLEM in the back of Rice's mind since 
the expiration of the Edison incandescent lamp patent in 
1894 how to increase the lamp's efficiency. One day toward 
the fall of 1900 the door of Rice's office opened to Charles P. 
Steinmetz, now becoming famous as a mathematical engi 
neer, and with him, Albert G. Davis, manager of the Com 
pany's Patent Department. 

"Mr. Steinmetz and I," said Davis, "believe it would be 
an excellent idea to create a laboratory where scientific 
investigations might go forward on the incandescent lamp 
and other problems. We should like to recommend such a 
step for your serious consideration. 

"To improve electric lighting is our foremost thought," 
he continued, while Steinmetz, leaning in characteristic pose 
on a corner of Rice's desk, nodded assent. "I personally feel, 
and Mr. Steinmetz endorses my views, that the electric light 
has a future more brilliant than its past. We should not like 
to assert that the carbon filament lamp is the best lamp we 
can have. There may be a better type of electrode for arc 
lamps than the carbon electrode. These things cannot be 
determined properly without research." 



Research! It was what Rice himself had been thinking of, 
Yet hardly another executive in the country in those days 
was considering such a step. 

The three entered into an analytical discussion of the ques 
tion. They all agreed that the proposed laboratory at Sche- 
nectady should be entirely separate from the factory and the 
sales branches, a place where all talk of sales should be out 
of order. 

Dr. Willis R. Whitney, of the Massachusetts Institute of 
Technology, was finally engaged for the post. In order to 
start working at once, Whitney agreed to take advantage of 
a laboratory which Steinmetz had established for his own 
use in an old barn that stood in the rear of the rooming house 
occupied by Ernst J. Berg, Eskil Berg, and himself on Lib 
erty Street. There in 1901 Whitney came for three days a 
week to work on preliminary problems, usually with assist 
ance from Steinmetz. The entire staff consisted of Whitney 
and a young assistant, J. T. H. Dempster. But that little 
barn was the cradle of General Electric's Research Labora 

COFFIN REALIZED AS FEW OTHERS DID the role of the incandes 
cent lamp as the herald of the electrical idea, the first con 
tact which many thousands of people had with the electrical 
way of doing things. If it made a good impression, electric 
service in general would benefit by the resultant good will. 
So Coffin's policy of helping the independent companies to 
improve their product, either with or without amalgamation, 
was pre-eminently sound. 

Among the independents there was one man who held 
the key to the problem, Franklin S. Terry, proprietor of the 
Sunbeam Incandescent Lamp Company of Chicago. This 
unpretentious company was established just about the time 


of the great lawsuit over Edison's lamp. Terry had found 
himself restricted by the infringement actions, but after the 
patent expired in 1894 he had slowly built his business anew, 
watching with wary eye the progress of his big competitor. 
Shortly before 1900 Terry began talking consolidation with 
some of the other independents. One evening in 1901 he 
sat next to B. G. Tremaine at a jobbers' dinner in Chicago. 

"I tell you, Tremaine, we've simply got to combine!" he 
exclaimed. "If we don't we'll never get anywhere and Gen 
eral Electric will get everywhere." 

"What have they been doing to you now?" said Tremaine. 

"Taking business away from us," snapped Terry. "We had 
a big prospect in sight and submitted samples of our lamps. 
So did the Edison Lamp Works. Then they proposed to the 
prospect that he test their lamps and ours. Even provided 
the equipment for the testing. Of course they won. Why? 
The old story a better lamp. We can't stand that sort of 

Slowly, as they talked, the virtues of consolidation as 
sumed clearer outlines to both of them and a plan suggested 
itself a plan for merging the independents as actual com 
petitors of General Electric and at the same time obtaining 
from General Electric adequate financing to assure success 
in operation. It was a bold plan. 

But General Electric was prepared to accept the proposi 
tion. They knew that Terry and Tremaine's interest would 
stimulate their activities, and they were confident that these 
men could be relied upon to do a good piece of work. It was 
not the first time that General Electric had acquired control 
of an enterprise and vested full executive management in 
the men already operating it. It was men always men that 
Coffin sought, the driving force behind all enterprise, the 
dynamic source of all progress and all success. 


Let Terry and Tremaine have full play! If they developed 
new ideas, excellent. There was no doubt that General Elec 
tric would share in these, just as the National Association 
would be allowed to share in new ideas emanating from the 
Lamp Works at Harrison. 

And so, before 1901 drew to a close, the National Electric 
Lamp Association was created, without a General Electric 
man in it. By 1904 the National Electric Lamp Association 
was in full operation. In all essential respects it was an active 
competitor of the Edison Lamp Works of General Electric. 
In the Cleveland factory of the Association on 45th Street 
had been organized such service departments as cost, sales, 
and advertising analysis, credit systems, production methods, 
and engineering and scientific research. Each of the affiliated 
companies was termed a "lamp division." They were fully 
autonomous and so nearly independent that, although au 
thority centered in Cleveland, it was seldom apparent. Expert 
aid, such as they could never have afforded by themselves, 
was quickly available when needed. 

As time went on, it was repeatedly demonstrated that 
leadership in the manufacture of incandescent lamps was 
not to be wrested from the veterans at Harrison. Notable 
improvements still came, for the most part, from the Edison 

Amazing and brilliant work was being accomplished, be 
tween 1900 and 1910 by William R. Burrows, once a 
Thomson-Houston apprentice, and John W. Howell, the 
veteran Edison lamp engineer. Burrows had devised a 
machine which accomplished the double weld process auto 
matically and made use of a smaller size of platinum wire than 
an operator could hold in his fingers. The first reduction in 
the costly amount of platinum used in the lamp came with the 
invention of the double weld process. 


Platinum was used originally because it is the only metal 
which expands and contracts at the same rate as glass, so 
that during the cooling of the stem seal, containing platinum 
wires embedded in the glass, the airtight joint was not im 
paired. Around 1912 research under Howell's direction de 
veloped dumet wire, which consists of a core of nickel-iron 
alloy within a thin shell of copper. Dumet wire shrinks at the 
same rate as glass, and by its use platinum is avoided. 

So platinum had gone the way of the bamboo filament. But 
no one was satisfied yet with what had been done to this tiny 
globe of glass. 

Burrows set to work to devise an automatic filament- 
cementing machine and an automatic exhaust machine in 
which the air was pumped out of the bulbs and the exhaust 
tube sealed off from the bulb at the rate of several thousand 
lamps a day lamps far superior to any that had been pro 
duced by hand methods of exhausting. A similar machine 
was developed about the same time at Cleveland. 

Already bulbs were blown in moulds instead of by hand. 
But now came the remarkable bulb-blowing machine, known 
as the Westlake machine. It was devised by August Kadow 
of the Libbey Glass Company, and purchased by General 
Electric in 1917. Nothing like the automatic glass blower had 
been seen before. It did what generations of hand blowers 
had never believed possible it produced symmetrical, 
moulded bulbs of glass automatically. 

Lower factory costs meant lower selling prices. The Na 
tional Electric Lamp Association, because it was part of 
General Electric, was able to share in these cost-reducing 
innovations. The effect of the trend in "co-operative compe 
tition'* between the two establishments made itself apparent 
in the greater number of lamps annually purchased. 


Steinmetz and the Arc Light 

STEINMETZ HAD THE CREED of a searcher. He was never tired 
of laboratory work. That was where happiness lay for Stein 
metz. He lost himself for long hours, for half a night, in his 
private laboratory in Schenectady, and what he did there he 
never alluded to as "work." 

When he joined with Davis in recommending the experi 
mental laboratory which Whitney was beginning, Steinmetz 
endorsed Davis' views of the need of applied electrical 
science. Already there was the carbon-filament lamp (incan 
descent), the carbon-electrode lamp (arc), and the carbon 
brush of the direct-current motor and generator. Electric 
light and electric power had really passed through what 
might be called the "carbon era." 

Was carbon the ultimate medium for all these products? 
Steinmetz would not be satisfied to say that it was. And so he 
bent his scrutiny upon the arc lamp. The arc lamp had not 
changed in essential character since the evening in 1808 
when Sir Humphry Davy of the Royal Institution produced 
the first electric arc, with the aid of a battery of 2000 voltaic 
cells. Two electrodes of carbon were used on that night, and 



had been ever since. Changes were all concerned with the 
manner of producing the current that passed through the 
electrodes, the mode of connecting the lamps to the circuit, 
and the method of regulating the length of the arc by mag 
netic mechanism. These were the contributions of Brush, 
Thomson, Wood, and the rest. Davy had contributed the 
light source; the others had contributed the various means 
of utilizing that light source for practical illumination. 

At night, absorbed in the enthralling search, Steinmetz 
was to be found in his laboratory. By day he pursued his 
mathematical engineering and his consultation work at his 
office at the General Electric plant. People passing the new 
Steinmetz dwelling on Wendell Avenue, unfinished in 1900, 
could see shining from the windows of a long, low structure 
in the rear electric lights of an intense bluish- white radiance. 
More and more frequently as the weeks slipped into the year 
1901 were those lights seen. They were an illuminated gauge 
of the progress the black-bearded little man was making in 
his search for something that would supplant Sir Humphry 
Davy's arc-lamp electrode. 

Toward the end of 1900 all his experiments centered upon 
a type of electrode composed of magnetite an oxide of iron 
mixed with a proportion of titanium. The resulting light 
was far more brilliant than that produced by the carbon 
electrodes. But only long investigation could determine the 
quantities of magnetite and titanium that would work best 
together. Both substances had drawbacks. They must be 
combined in the one ratio in a thousand which would suc 
ceed in allowing their advantages to outweigh their objec 
tionable points. 

To assist him Steinmetz called upon the new General 
Electric laboratory which now, after its first few months of 
existence, was located in a small wooden building at the 


General Electric Works, designated then as Building 10. 
There and at the Steinmetz laboratory on Wendell Avenue 
the study of the magnetite-titanium electrode went on, until 
its brilliant future was at length assured. 

The new lamp had peculiarities of its own. The arc instead 
of springing forth pointed at both ends and bulging in the 
middle, as did the arc of the carbon electrodes, became tri 
angular, spreading in both directions from the tip of the 
upper electrode and coming to a point at the tip of the lower. 
In the second place, the illumination originated in the arc 
itself, whereas in the carbon lamp the illumination was caused 
by the incandescent tip of the lower electrode, heated white- 
hot by the arc. 

These were advantages that caused Steinmetz and Whitney 
to watch their laboratory lamps with keenest attention, and 
to consider it only a minor episode when they discovered 
that after the current had once been turned off they could 
not reproduce the arcl 

They saw at once that when the current ceased to flow, 
the upper electrode dropped against the lower, as in the case 
of the carbon lamp, and owing to the melted magnetite both 
stuck together, so that when the current was again turned on 
the control mechanism could not pull them apart. Current 
flowed through the two electrodes without encountering the 
break in the circuit on which the arc depends. Magnetite has 
a lower melting point than carbon, so that this difficulty was 
not encountered with the carbon lamp. 

Steinmetz soon had another disciple in the person of 
C. A. B. Halvorson, one of the most active engineers at the 
River Works at Lynn, where arc lamps and their auxiliary 
equipment were manufactured. So active a disciple had he 
become, as the problem of the magnetite arc lamp came to a 
focus, that it was soon apparent that he might do for 


Steinmetz's discovery what Wood and Brush and Thomson 
did for Sir Humphry Davy's put it into practical working 

Halvorson succeeded in designing a control mechanism 
which would keep the two electrodes apart when the current 
was off, and would automatically move them into the correct 
relation to form the arc when the current came on again. 
Then the melted magnetite caused more trouble. Globules 
of molten iron, instantly solidifying, dropped from the upper 
electrode and smashed through the glass globe of the lamp. 

Again and again it happened. Globe after globe was 
broken, and at the end of each day the janitor was swamped 
with the job of sweeping up glass fragments. Hundreds of 
globes were broken before it was reluctantly admitted that 
magnetite could not be used. 

After more experimenting, Halvorson tried pure copper for 
the upper electrode, retaining the magnetite-titanium com 
bination for the lower. Here was the great solution. The 
difficulties of melted magnetite were eliminated, and the 
brilliance of the arc retained, despite the fact that only one 
electrode was active. The copper was permanent to an un 
usual degree, requiring renewal only about once in three 
years. This electrode, however, had to be encased in a steel 
shell to prevent undue oxidation. 

A fresh difficulty now arose from fumes and vapors given 
off by the magnetite under the heat of the arc. The vapor, 
thick and brownish, settled on the globe, obscuring the arc. 
A number of expedients were tried, but Halvorson finally 
placed inside the lamp a smoke outlet or chimney constructed 
of wrought-iron pipe. This drew the vapor into the outer air, 
and a small cap with lateral openings allowed it to escape 
without being blown back into the chimney by the wind. 

At length the new lamp was ready for quantity manu- 


facture, and ready to face manufacturing problems, now that 
problems in design had been met and defeated. The magnet 
ite electrode consisted of a tube into which the magnetite 
and titanium, in powder form, were forced under pressure. 
The two kinds of powder had to be uniformly distributed 
throughout the tube, and it was impossible for anyone to do 
this with such speed as to allow quantity production. 

Again Halvorson attacked the problem, while the work of 
preparing the new lamp for the market was suspended. He 
was completely baffled, until one day, walking along the 
shore, he noticed how closely the sand was packed by the 
action of the waves. A solution came to him in a flash. He 
got back to his office as rapidly as possible. There he worked 
out his scheme for subjecting the electrode tubes to a constant 
gentle bumping as the pressure was applied. The uniform 
distribution of the powder was assured, and the whole 
process could be performed by a machine. 

Early in 1903 most of the new lamp's advocates had given 
it up in despair. Someone suggested that a batch of fifty of 
the lamps should be manufactured and presented to Dr. 
Steinmetz, and that the undertaking should then be con 
sidered closed. The fifty lamps were sent over to Schenectady, 
and Halvorson went along to install them. About half of the 
lamps were put up on poles around Steinmetz* new house 
and along the adjacent streets. Steinmetz was delighted. He 
had no inkling of the skepticism at Lynn or of the virtual 
rejection of his arc lamp in that quarter. He believed in the 
lamp, never doubting that its day of practical utility was 
close at hand. As he and Halvorson stood in the picturesque 
grounds surrounding Steinmetz's home and watched those 
lamps flash their blue-white brilliance, they knew that this 
was the arc lamp of the future. It still had its imperfections 
but it would not fail! 


Yet this ceremony, planned as the funeral of a tremendous 
project, proved to be a celebration of its success. Halvorson 
was able to obtain operating data from the installation that 
indicated readjustments and refinements in the lamps. The 
Lynn men were bowled over by the news that the magnetite 
lamp was a practical success and was to go on the market 

Out of an odd, almost Bohemian environment came the 
magnetite electrode. No workshop in America, perhaps in 
all the world, had quite the bizarre atmosphere of Steinmetz's 
laboratory during those years when the new century was 
coming in, and with it the gleaming, blue-white lamps. 

In that environment the study of all the problems of com 
mercial introduction and operation of those lamps went 
steadily forward. Research and science were mingled with 
the informal housekeeping of a bachelor menage. There 
were ordinarily two persons in the establishment, Steinmetz 
and his general assistant, J. LeRoy Hayden. They lived and 
worked in the laboratory while the new house was being 
built, for Steinmetz refused to leave his beloved paradise. 

Hayden had come up from the General Electric works to 
take charge of the power plant adjoining the laboratory, 
where a Brush arc-light dynamo generated direct current for 
the twenty-five magnetite lamps of the outdoor exhibit. 
Steinmetz offered him quarters in the laboratory to save the 
long trip back to his room in downtown Schenectady. Hayden 
was only too glad to accept, little supposing that it was the 
beginning of a working combination that was to last twenty 
years, or that he was to become Steinmetz's most intimate 
friend. He moved his personal effects into the laboratory, 
sleeping in the room next to Steinmetz's on the second floor, 
and eating the generous meals that Steinmetz cooked on a 
little gas stove. 


It was a necessary evil, the business of eating, but Stein, 
metz took the burden upon himself, and he cooked not as a 
chef but as a chemist. He was partial to all shades of yellow. 
Eggs were consequently a staple of diet, varied, when he 
could get his mind off his favorite color, by beefsteak and 
boiled potatoes. 

The two ate unceremoniously, did their chores on a co 
operative basis, and hurried to the test room and the glowing 
bulb, while day vanished and the hours of the night marched 
on. Neither thought of sleep until late and yielded to it only 
with regret. 

Halvorson and others of the engineering group frequently 
visited Steinmetz who offered them the hospitality of his 
laboratory. They sat in faded, threadbare parlor chairs or on 
a horsehair lounge that stood amid a maze of laboratory 
apparatus, as talk drifted to the magnetite lamp. 

"But it works only on direct-current circuits," one salesman 
complained in dismay, for there were not many direct-current 
circuits in existence. 

"That only means," said Steinmetz, in his characteristic 
way, "that we must have some way of changing current from 
alternating to direct. In our laboratory Dr. Weintraub is 
experimenting now with a tube in which an electric current 
is changed from alternating to direct by means of mercury 

Dr. Ezechiel Weintraub, who had discovered that mercury 
vapor will permit the passage of an electric current in one 
direction only, was a physicist in the General Electric Re 
search Laboratory. He made the discovery in a mercury-arc 
lamp of fair illuminating properties but cursed by a ghastly 
greenish tint. 

Peter Cooper Hewitt, son of a former mayor of New York, 
was following the same line of experiments in Newark. He, 


too, had invented a mercury-arc lamp of brilliant, greenish 
light, which he was attempting to market in long tubes. He, 
too, brought out a mercury-vapor rectifier. Inevitably the 
claims of the two inventors clashed in the United States 
Patent Office. To clear up the situation, General Electric 
made an offer to purchase Hewitt's concern. Accepting the 
offer, the Cooper Hewitt Electric Company thereby joined 
General Electric, and the latter marketed Hewitt's tubular 
mercury lamp, which developed a special field of usefulness 
in photographic studios. 

In the meantime Steinmetz had adapted Weintraub's 
mercury-arc rectifier for commercial work in connection with 
the magnetite lamp. After several changes in the design of 
the tube, he substituted this convenient device for the bulky 
and costly rotating machines which had hitherto performed 
the converting function. 

In Jackson, Michigan, magnetite lamps first went into 
commercial use in 1904. The interest of electric light men all 
over the country was focused upon that city. The new lights 
were given the trade name of "luminous arc lamps," and 
their merits were so apparent that scores of orders material 

Then Boston demanded a street light of even greater bril 
liance than the luminous arc lamp as yet supplied. General 
Electric engineers set to work to raise the capacity of the 
lamp from 300 to 500 watts, and to increase the proportion of 
titanium in the electrode. It was no simple task. Day and 
night for almost three months Halvorson and his assistants 
worked with an isolated lamp in the outskirts of Boston. 
Then they installed twenty-five lamps on Huntington Avenue 
with a success which amazed the Boston engineers. Four 
thousand of these lamps eventually supplied the entire area 
of Boston proper. 

President of General Electric from 1913 to 1922. 

Dr. E. F. W. Alexanderson and his high-frequency alternator. 


In 1902-03 General Electric built searchlights with lenses 
thirty-six, forty-eight, and sixty inches in diameter. Most of 
these were ordered by the Army and the Navy, which were 
engaged in extensive maneuvers to test the coast defences on 
Long Island Sound and Casco Bay. The naval craft were 
attempting, theoretically, to get past the coastal fortifications 
by night. Searchlights were set to work to hunt down the 
"enemy." The first 60-inch light was placed in Fort Wright, 
on Long Island Sound, and a battery of 36-inch lights was set 
out in other forts around New London. Unwaveringly the 
powerful beams "protected" New York from the attack. Not a 
vessel escaped detection. 

One 80-inch monster was built by General Electric in 1904 
and sent to the St. Louis Exposition the only one of its size 
ever built. The handling of this great projector was a con 
struction job in itself. A stationary steam engine, rigged to a 
block and tackle, was required to swing it up the side of a 
building to its resting place. It was not the sort of thing that 
anyone would expect to get lost yet that is what happened. 
General Electric never knew what became of it after the 
exposition ended. A rumor drifted in some years later that 
the big light was sold to the Russian government when the 
Russo-Japanese war opened, and that it joined in the defence 
of Port Arthur on the coast of Siberia. 

These multifarious developments were of necessity the 
work of a highly specialized personnel. Technical problems 
were most frequently mastered by technical men, but not 
always. Now and then an idea emanated from the shops. 
Such was the case when, in preparing for a railway motor 
test, Peter J. Mulvey, foreman of the test, contributed a no 
table improvement to the carbon brush. 

A dozen years had passed since the carbon brush was 
applied to the infant street-railway motor, in which time it 


had stood the burden well. But the increasing capacities of 
traction motors proved greater than the brush in its original 
form could handle. In the electrification of the Manhattan 
Elevated in 1901, the motors were rated at 153 horsepower, 
and the carbon brushes sparked alarmingly when current 
was applied. The engineers tried to lubricate the brushes by 
dipping them in oil, but the sparking continued. The problem 
became a thorn in the side of "Pete" Mulvey, then spending 
his tenth year in the plant. 

The boys who worked for Mulvey on the railway motor 
test college graduates fresh from the campus were great 
admirers of his sagacity as foreman, his knowledge of electric 
motors, and his open-handed manner. They found him not 
only fair in his dealings, but a jolly fellow in a crowd and one 
who appreciated a joke. They were delighted when the news 
swept their department late in 1901 that Pete had solved the 
problem of lubricating the carbon brush that Van De Poele 
had invented a few years earlier. 

Various accounts tell how he turned the trick. The version 
most generally related, even by Mulvey himself, attained in 
time the dignity of tradition. The story went that Pete had 
become so annoyed one morning over the obstinacy of the 
carbon brush that he flung it the whole length of the shop 
and saw it vanish in a bucket of oil. Repenting of his im 
pulsive conduct he rescued the brush, took it home that night, 
dried it out by baking it in the kitchen stove, and discovered 
the next day that it worked with complete freedom from 

What Mulvey actually did was a trifle different, though he 
achieved the same results. Experimenting on his own account, 
he immersed brushes in oil, then subjected them to repeated 
bakings, with an interval of time between each baking. That 
was the secret of the lubricated brush. The lubrication was 


so thoroughly baked in that the brush would transmit a 
heavy current without danger to the commutator from spark 
ing. Later the Research Laboratory devised a self-lubricating 
brush, and reduced the costs of manufacture. 

An entire shop at the Schenectady Works was now devoted 
to carbon products, including brushes, and 2,000,000 pieces 
were manufactured there annually. The period of tremendous 
expansion in the shops was then at its height. And John 
Miller, the weather-beaten millwright who so dearly loved 
his machines, was encountering tight situations in equipping 
the spacious new shop buildings. A big building, the present 
Building 16, was completed in 1899 and into it went a huge 
sixty-foot boring mill. To handle this mass of metal, a steel 
casing twelve feet in diameter had first to be sunk thirty-five 
feet in the ground to contain the operating mechanism and 
the spindle shaft on which the bed of the boring mill was to 


Brains and Brawn 

THE RESEARCH LABORATORY was thriving in its new home in 
Schenectady, having graduated from the one-story shed-like 
structure to Building 19. In its deep cellar Whitney and his 
few assistants had carefully collected their stock of chemicals 
and mechanical appliances in the spring of 1901. That was a 
high-water year for the impetuous Mohawk, which over 
flowed its banks and poured into the cellar of Building 19, 
setting afloat all the supplies of the laboratory. The place was 
converted into a lake clinking with bottles and jars afloat. 

Before another spring the laboratory had been moved, and 
in 1904 was lodged in Building 6. Whitney's staff now con 
sisted of three chemists drawn from the Massachusetts Insti 
tute of Technology. Various enterprises were afoot and more 
men were being required. Whitney's own work resulted, be 
fore the end of 1904, in the development of a new type of 
electric resistance furnace. 

In subjecting various articles to the heat of the furnace, 
Whitney was amazed at the effect which the heat had upon 
carbon filaments. Analyzing and testing the filaments inside 
incandescent lamps after this high-temperature treatment, 



he realized that the very nature of the carbon was changed. 
It now behaved in a way characteristic of metallic sub 

These metallized carbon filaments were a new step in the 
line of Edison's original objective, which was to utilize a 
high-resistance filament. When the new lamp was finally 
ready for commercial usage, it consumed only 2.5 watts per 
candlepower and yielded nearly 300 candles per horsepower. 
It represented an increase over its predecessor of twenty-five 
per cent in candlepower for the same volume of energy. 

This new type of lamp, known by the trade name of "Gem," 
was the final achievement of the long carbon period in 
incandescent lighting. It was recognized as the greatest single 
improvement of the entire span of development and was in 
itself a justification of all the efforts expended in the research 

Two years of study and experiment were required before 
the manufacture of the new filaments was possible com 
mercially. They were more expensive to produce than the 
old-style filament. But in 1907 the sales of "Gem" lamps 
approached four million, and in 1909 they leaped to ten and 
a half million, outdistancing for the first time the unmetal- 
lized-filament lamp. 

Yet the carbon period in incandescent illumination was 
fated to pass. Even as Whitney was making ready to turn his 
lamp over to the factory, a newcomer was being heralded 
from abroad. 

In Germany, under the encouragement of the Siemens and 
Halske Company, Dr. Werner von Bolton had been experi 
menting with the hard, brittle metal tantalum, which he had 
purified until he was able to render it ductile, or pliable, so 
that it could be drawn out into a flexible wire. Incandescent 
lamps with filaments made of tantalum wire were of even 


greater efficiency than those with the metallized filament. 

There was a fresh stir in electric light circles. General 
Electric obtained a license to manufacture the lamp in 
America, and it was commercially introduced before the close 
of 1906. The tantalum lamp consumed only two watts per 
candlepower and yielded 370 candles per horsepower, which 
was about twenty per cent better than the metallized carbon 

But the tantalum lamp had a great drawback; it was con 
fined almost entirely to direct-current circuits. When con 
nected to alternating-current circuits, its life was curtailed 
because of the crystallizing of the tantalum. Nevertheless, 
sales rose as high as twenty-seven million a year by 1913. 

Invention is like lightning. It strikes none can tell where 
or when. Nor does it always require the man of broad tech 
nical schooling to work its will. While Whitney in his labora 
tory was firmly fixing the principle of research in General 
Electric's program, at the opposite end of the works brawny 
John Miller, the millwright, was turning things upside down 
to contribute an unheard-of idea to the development of 
mechanical processes. 

Direct-current motors as they came into extensive use gave 
rise to many months of searching for noiseless gears and 
pinions. The strain upon gear teeth, meshing together with 
great force one moment, stopping entirely the next, caused 
them to wear out rapidly and unevenly. Most of the gear teeth 
in the thousands of motors that were driving machinery 
underwent a constant banging action. In the days of belts 
and shafting, such shocks were absorbed by the belting, but 
those days had passed. And the noise of the gears was 
nerve-racking. * 

Miller spent some time experimenting with many kinds 
of gears. He tried bronze, but it was too soft to last. He tried 


hard steel, but it made such an uproar that it was out of the 
question. He tried rawhide, but it did not stand the strain. 
The same was true of leather. He tried copper, laminated 
iron, wood, compressed paper; all in vain. 

"The gear we want is a gear that will wear smooth," he 
told himself, watching the punch and shear. "These teeth 
are wearing unevenly, some places giving out faster than 
others. What we want is a gear that wears evenly, all over 
its surface, so that it continues to mesh perfectly even while 

A few days later a thought occurred to him while he was 
gazing at a shiny spot on his coat sleeve. 

"Worn smooth!" he exclaimed. 

With sudden inspiration he secured some buffing wheels 
from one of the shops, wheels made of cloth sewed together 
in layers and used for polishing the finished parts of ma 
chines. He took these to his own shop, pressed them together 
between clamps, cut out gear teeth and mounted the cloth 
gear upon the hack saw. Then he started the saw going and 
watched it closely. 

Two months slipped by. The cloth gear seemed to be 
standing the test. It was silent in operation. And it was 
wearing smooth and shiny like Miller's coat sleeve. It was 
getting hard as well, so that one could scarcely tell it from 
metal. Convinced that he was on the right track, Miller took 
the gear to John Riddell, the mechanical superintendent of 
the plant. 

"There's a gear made out of cloth," he announced with 

"Out of cloth?" exclaimed Riddell. "I'm afraid you're 
going crazy!" 

Miller took him down to the shop and let him watch the 
construction of another cloth gear. Riddell became half 


convinced. He notified the Patent Department, but he took 
precautions to make light of the story and even drew a 
humorous sketch representing General Electric scouts forag 
ing among neighboring clothes lines for raw material for the 
"lingerie pinion." 

But the cloth pinion was no joke; the Patent Department, 
the engineers, and even Riddell himself, were soon taking it 
seriously. Miller soon had a cloth gear and a cloth pinion on 
the punch and shear, where they worked both silently and 
successfully for three years. The best previous gear on that 
machine had given out in a week. 

So it was that General Electric manufactured and mar 
keted cloth gears. The first step in preparing the original 
type involved placing cotton cloth under enormous pressure, 
and from this it was termed Fabroil. A later adaptation, 
known as Textolite, made its own place in the world in the 
timing mechanism of automobile engines. 

Again hard times came "a-knockin' at de door," and Ameri 
can business and industry were forced to stop and listen. It 
was the financial stringency of 1907. But this time General 
Electric did not suffer. Though its sales were curtailed, its 
supply of available cash was abundant. It was better off than 
scores of contemporaries, large and small. 

What gave it such a financial advantage at a time of crisis 
was the financial genius of a man who had an extraordinary 
capacity for appraising the trend of business developments. 

During the first six months of 1907, when orders received 
by General Electric increased twenty-two per cent over those 
of the corresponding period of the preceding year, General 
Electric issued bonds to a total of nearly thirteen million dol 
lars. The interest rate was five per cent. It was the first ex 
tensive financing the company had undertaken in more than 
ten years. Again and again the annual report had, contained 


such paragraphs as these: "During the past year the company 
has not borrowed money or incurred obligations." General 
Electric had lived within its income. 

But as the second half of 1907 set in, there appeared the 
stringency which Coffin had sensed. From August 1, 1907 to 
January 31, 1908, General Electric's orders declined twenty- 
three per cent over the corresponding period of the preced 
ing year. Business was dull, work slackened, markets wavered 
or went to pieces. Money became scarce and rates rose alarm 
ingly. Business houses on every side were paying ten or 
twelve per cent for what cash they could borrow and the 
supply of cash was limited. 

Through it all General Electric sailed on an even keel. 
Although its factories were compelled to slow up, as the tide 
of orders ebbed to a slender stream, its executives were un 
disturbed about meeting payrolls or financing current obliga 
tions. Indeed, Coffin was able to help some of his large 
public-utility customers, loaning them thousands of dollars 
from the funds accumulated by his issue of bonds. And that 
stored up in turn a fund of future good will which gave 
General Electric salesmen a golden advantage year after year 
when the skies had cleared. 

LATE IN 1907 a construction foreman left the Chicago office 
of General Electric and jolted over rough, sandy roads to the 
wild southern shore of Lake Michigan, where a steel com 
pany was fashioning a new community. The foreman was O. 
B. Vinal, experienced as a mechanic, capable in handling 
men, brusque yet kindly in manner. He could mix with any 
crowd, live under any conditions. He had the pioneer spirit. 
The town smacked of the early gold-rush days in California 
or the oil-boom period in western Pennsylvania. A single 
street of swirling sand, churned ankle-deep by traffic; a strag- 


gly line of shacks that were lodging houses; brawny, sun- 
browned fellows of the construction gangs and erecting 
crews, who formed nine-tenths of the population; a constant 
rattle of explosive language as man and beast toiled daily 
through the sand dunes or shivered in the gales; and sand 
fleas everywhere, burrowing and biting, day and night that 
was Gary, Indiana, when Vinal and his men first arrived in 
November 1907. 

The great steel mill was under construction while the big 
motors were going in. The salamander stoves, scattered about 
in those drafty interiors, flung their warmth hardly beyond an 
arm's length against the cutting winds. 

The men of the General Electric crew bunked two in a 
room in the one substantial hotel, where all united in a cease 
less warfare against the fleas! 

"Dad" Vinal was boss of the cuisine, general manager of 
the hotel, fatherly preceptor to his sixty huskies, and general 
construction foreman on the motor-assembly work. By one 
means or another he kept up morale through months of grind 
in a backwoods environment. 

Slowly the huge motors were machined to precision meas 
urements and their ponderous parts fitted together. They 
were bulky driving units, those motors, towering nearly thirty 
feet above the mill floor. And they had never been com 
pletely assembled before, even in the factory. 

The great flywheels were shipped in sections which the 
construction crew had to machine, mount, and bolt in place; 
the face of the rotors had to be pared down, so that the air 
gap would be two-tenths of an inch across; and four hundred 
tons of sheer weight for each complete unit had to be lifted 
into place. All this with the knowledge that the motors were 
to be tested, for the first time, while steel men from all the 
districts looked on. 


Time pressed, and the sand fleas stuck as relentlessly to 
their task as did the men to theirs. Questioning letters came 
from Schenectady, and Vinal wrote back reassurances, 
though himself not altogether reassured. It was an exhaust 
ing life but no one was idle, and excitement added its spice. 

What the construction men did not know though the en 
gineers at Schenectady did was the skepticism among steel- 
mill operators over the mammoth motors. There had been 
but a few instances of steel-mill electric drives, all on a com 
paratively small scale, and there were none in which the steel 
mill was entirely and continuously run by electric motors. 
Hence the predictions up to the day of operation that the 
motors would never deliver the required power. Back in 
Schenectady the young designing engineer, Howard Max 
well, held no doubts, for he had designed these motors to de 
liver as much as 20,000 horsepower in case of need. 

On the day of the test, January 17, 1909, Vinal looked a 
good deal calmer than he felt. Steel men by the score had 
gathered, and competitors of General Electric, not to men 
tion General Electric's own engineers, led by E. W. Rice, 
Jr. Vinal moved the handle of the pole-changing switch 
which regulated the speed of the steel mill, a speed that was 
double that of any previous mill. Hewlett had designed that 
mechanism which controlled current at 4000 volts, and had 
heard the critics predict its failure. 

There was a precipitate backward movement among the 
crowd as the motors started a tremendous droning song. The 
steel mill was in operation! Vinal and his crew withdrew and 
left the big motors to themselves. 


Spectacular Applications and 

Discoveries, Continued Expansion, 

Birth of Radio, Major Achievements 

Marking the End of the 

Pioneer Period 

The Romance of Work 

IN THE EARLY YEARS of the new century, an epochal struggle 
was developing. It dealt with transmission. How could cur 
rent be transmitted without undue loss? Unless this question 
were solved, transmission distances would be forever limited. 

The thing that held them back was nothing other than the 
insulators on the poles. Queer-looking tyrants they were, imi 
tations of "a lady with an umbrella and a variety of petti 
coats." Some of them by 1902 stood as high as a table, were 
as plump as a barrel, and weighed nearly sixty pounds. They 
were constructed of fragile porcelain, difficult to transport, 
to erect, and to replace without breakage. Yet these delicate, 
unwieldy objects, weighty as they were, had to be mounted 
upright upon the crosspieces of every transmission pole, held 
in place by long vertical pins, and there support the swinging 
cables attached to their topmost points in spans varying from 
400 to 500 feet between poles. To go beyond 70,000 volts, 
they would have to be too enormous either to manufacture 
or to handle. 

What tales the linemen could relate of getting these insu 
lators up intact. How difficult the task of replacing one that 



had been smashed. They made wonderful targets for thought 
less youngsters and sharp-shooting cowboys. And replace 
ments had to be made immediately to avoid interruption in 
current, no matter how the gales might howl. Insulators 
were then selling at $3.50 apiece, and systems of size re 
quired from 5000 to 20,000 of them. 

Harold W. Buck, electrical engineer of the Niagara Falls 
Power Company, who had started with General Electric, 
was reflecting in 1905, as was every other transmission engi 
neer, upon the things that could be done if electricity could 
be transmitted at more than 70,000 volts. At that moment he 
had a project in mind that hinged upon the possibility of 
transmitting at 100,000 volts. 

He had asked the switch engineers of General Electric if 
they could construct oil switches for controlling energy at 
such a voltage, and they had quickly answered "Yes." But 
no one could guarantee him a practical insulator; his plans 
were halted there. 

Buck pictured his dilemma to Hewlett, General Electric's 
veteran switchboard engineer. It was an appeal for help. And 
Hewlett was not the man to dismiss a challenge. Together 
they discussed problems and sketched ideas. 

Buck suggested insulators made of hickory sticks and im 
pregnated rope, inserted in the transmission lines between 
the dead ends of the cable. Hewlett suggested distributing 
the insulation over a series of units. Then he asked why 
cables could not be attached to insulators that were sus 
pended downward instead of mounted upright, and why 
there could not be a unit type that could be made up into 
strings as desired. 

Hewlett returned to Schenectady and Buck to Niagara, 
and both began to experiment. Out of this there came the 
suspension type of insulator, Hewlett's conception; and its 



The development of control equipment for the Canal was one of the 
most important achievements of the early part of this century. 


A transcontinental passenger train on the Chicago, Milwaukee 6- St. 
Paul being pulled bij an electric locomotive. 


complement, the strain insulator, which was Buck's. They ob 
tained a joint patent in 1907. 

Broad, flat discs of corrugated porcelain were suspended 
from each crossarm in strings one beneath the other, like 
Chinese gongs, the bottommost disc supporting the trans 
mission cable; this was the suspension type. A severing of 
the cable, the insertion of a section of nonconducting ma 
terial uniting the severed ends yet itself attached to the cross- 
arm of the pole, and a swinging loop of electrical conductor 
hung from one dead end of the cable to the other to furnish 
a path for the high-voltage current; this was the strain insu 

Between them they carried the day. The nonconducting 
function of the insulator, distributed between those several 
discs, could be increased as the line voltage was raised by 
adding more discs. The strain insulator, which was mounted 
only on certain of the poles or towers, took up at regular 
intervals the mechanical strains which were imposed upon 
the transmission cables. 

Steel towers were replacing the less substantial wooden 
poles. Since they were more expensive, they were placed 
farther apart, and the spans swinging between them were 
lengthened to seven, eight, and nine hundred feet. The 
cables themselves grew in less than twenty years from the 
thickness of one's little finger to the size of a shovel handle. 
But the two new insulators sustained those weights with both 
mechanical and electrical success. 

One of Hewlett's engineer friends asked him as time went 
on, "J ust where is the limit in all this?" To which Hewlett re 
plied, "I don't know, it's not in our department." And so it 
appeared. From 70,000 to 110,000 volts in one leap meant ex 
panding two and a half times the area which an electrical 
system could serve. 


FROM THE PEN of Chester T. Crowell came a description of 
a visit to Schenectady. In telling of his impression of the 
great turbine shop, he said: 

"An enormous door rolled back on little wheels . . . My 
first impression was that we confronted miles and miles of 
roofed space. Sunlight, filtering through windows placed 
fifty or sixty feet above ground, showed that far away in the 
distance there was a rear wall. One saw it through a haze 
not of smoke or steam, but just distance . . . 

"The mechanism of a steam turbine is as meticulously ac 
curate as that of a fine watch, but the delicate measurements 
are applied to pieces of metal so large that a dozen men work 
ing on them look like ants. These are proud men artists. 
Watch their sensitive fingers caress that steel eggshell. Watch 
them measure distances as they signal to the ponderous 
crane. What pride shines in those faces as ten tons or more 
of machinery silently meets the floor, every corner in place. 

"These are not horny-handed laborers . . . their features 
are keen and sensitive; brawn is not a premium here , . . 
They are working on what will eventually be one of the 
world's great power plants; they are working under the direc 
tion of engineers and scientists whose acquaintance is an 
honor . . . 

"I watched it (the central shaft) turning slowly round and 
round, while a man wearing glasses appeared to be polishing 
it ... Coming nearer I observed that this man was working 
with several queer implements. 

" Tm trimming this to measurement/ he explained. A rib 
bon of steel thinner than paper was curling up from one of 
his implements. 

" 'What do they allow you on this?' 

" 'One-thousandth under; nothing over/ was the reply. 


"At the point where he was working the shaft was about 
three feet in diameter. I was told it weighed fifty-five tons. 
He was trimming it to the precise measurement named in 
the specification and the margin allowed him for human 
error was one-thousandth of an inch! 

"That man remains indelibly in my memory as a figure of 
romance. He and his sharp eyes and steady nerves, calmly 
facing that margin of one-thousandth of an inch! . . . 

"That man is an artist. Of the men I saw working in that 
cathedral of manufacture, none impressed me as a shy, mod 
est creature. Make no mistake, they have a keen sense of the 
romance and adventure of work." 

Going from the busy shops to a quiet little workshop in 
one of the laboratory buildings, we find picturesque old 
Julien Tournier, who had joined Edison's little group as an 
instrument worker in 1882. He had become an interesting 
"old-timer" at the Schenectady plant; a slight, white-haired, 
white-bearded figure, with a touch of French vivacity in his 
speech, he was always surrounded by an orderly confusion 
of tools and materials. 

When Gerrit A. Beneker, the painter of men, visited Schen 
ectady searching for types of industrial workers, he painted 
Tournier, and called his portrait "The Inventor." And when, 
a few years later, old Julien Tournier died, General Electric's 
current advertising space in a well-known technical journal 
consisted of a reproduction of Beneker's portrait of Tournier, 
and under it these words: 

"In the life program of Julien Charles Tournier, public 
recognition played no part. 

"He began as an instrument maker in Edison's laboratory. 
His forty-five years of service to electricity were completed 
in the same work ... He invented and contributed improve 
ments to switches, sockets, fuse-plugs, and attachment-plugs. 


He might have retired, had he so chosen, and lived in com 
fort; but his love of electricity was his life, and he was con 

"We publish his picture as a tribute to him, and because he 
typified the thousands of men and women who have dedi 
cated their lives to electrical development. The world will 
never know their names. They have no craving to be known. 
But their devotion is quite beyond the interest of men in 
ordinary business. Within the ranks of the General Electric 
Company are many such men. Their spirit is the best as 
surance that electricity will year by year find more and better 
ways to serve." 

IN 1907, TWO FIREBOATS launched upon the Chicago River 
by the city of Chicago, the Graeme Stewart and the Joseph 
Medill, were the first vessels in America to be propelled ef 
fectively by electrical energy. General Electric had not for 
gotten its contract with Charles G. Curtis which permitted 
it to use steam turbines with electric generators for driving 
the propeller of a ship. The Manitowoc Dry Dock Company, 
builders of the two fireboats, awarded to General Electric 
the contract for the electrical equipment of the craft. A. A. 
Babcock, the builders' engineer, worked with General Elec- 
tric's marine engineers, particularly W. L. R. Emmet, John 
H. Clark, and Maxwell W. Day, over the engineering eco 
nomics and operating efficiencies of so grouping the ma 
chinery that one 660-horsepower Curtis steam turbine drove 
both the electric generator and the water pumps. 

In this same year the alternating-current transformer was 
helped over a hurdle. The transformer in its smaller sizes, 
which distributed electrical energy for local neighborhoods, 
had developed disconcerting electrical losses, whose only 
apparent remedy involved too great an expense to make 


their use practicable. When the prospect looked particularly 
dark, there came a ray of light from England. Sir Robert 
Hadfield, scientist, had discovered an alloy of steel and 
silicon, and General Electric, under a patent license, at once 
undertook to produce it in quantity. W. S. Moody, Chief 
Transformer Engineer, experimented until at last he suc 
ceeded in rolling the new steel in large tonnages. The di 
lemma of the small transformer was solved. 

It was also in 1907 that Rice and Ord, vice presidents, rec 
ommended to the directors the purchase of seven hundred 
acres of land at Erie, Pennsylvania, whereon to build a new 
plant when times were brighter. Confident that times would 
be brighter, they sought to shift manufacturing toward the 
geographical center of population. The plan was amply justi 
fied; that land became the site of the Erie Works of today, 
the center of its electric transportation activities. 

SHORTLY BEFORE 1900 a glass tube, 186 feet in length, had 
been placed one day in the foyer at Madison Square Garden. 
As night drew on this tube glowed with a brilliant white 
radiance. The illumination extended to all parts of the foyer. 
It was the first of Moore's practical gaseous lamps. Its effec 
tiveness attracted great attention, and the test established 
the fact that it was more economical of electric current than 
the incandescent light. Within a few years a number of radi 
ant tubes had appeared in Manhattan. 

Such events were not to be ignored by General Electric, 
manufacturers of the lamp with the carbon filament, which 
Moore had once told Edison was "too small, too red, and too 
hot." Was that lamp to be supplanted after twenty-five years 
by a luminous gas? 

It was possible, despite the fact that the Moore tube was 
cumbersome. It had to be installed by expert workmen, ex- 


hausted of its air upon the premises, and its replacement 
when it got out of order was a troublesome procedure. 
Hence, although Coffin and Rice negotiated for Moore's 
patent rights, they also implored Whitney to add an able 
man to his staff to study the elements which had promise of 
yielding better filaments. Whitney immediately recom 
mended Dr. William D. Coolidge, of the faculty of the Mas 
sachusetts Institute of Technology, who had been doing 
original work in physical chemistry. 

So Coolidge, in Boston, received a surprising offer from 
Whitney. But it was an offer that proved satisfactory, and 
early in 1905 Coolidge became a General Electric researcher. 

Just as these preparations were being made, scientific 
men were stirred by news of a great discovery in Europe. 
An incandescent electric lamp had been developed which 
contained a filament made of the most intractable of sub 
stances, the metal known as tungsten. 


The Taming of Tungsten 

TUNGSTEN WAS INTRACTABLE, more brittle than dry bone, 
more fragile than an eggshell. Yet in 1904 men fashioned it 
into slender, hairlike filaments for incandescent lamps. 

Two laboratory workers in a far land brought forth these 
filaments of tungsten. They had no funds with which to fi 
nance their work, but they summoned to their task patience 
and sustained, intelligent endeavor. 

In 1902 Dr. Alexander Just, a laboratory assistant in the 
Technical High School in Vienna, began experimenting in 
his spare time to develop a new form of filament for incandes 
cent lamps. He did not think at first of tungsten but tried a 
number of other substances. He associated with himself a col 
league, Franz Hanaman, and they studied tungsten between 
them, experimenting for two years. 

Both were men of small financial means. They paid their 
laboratory expenses from their humble salaries. Dr. Just's 
income was $55 a month. And they had set themselves a 
mountainous task. 

Tungsten existed only as a gray-black powder of fine hard 
particles, or as a rough, half -fused mass. Except as an alloy 



in tungsten steel, it had never had any commercial value. In 
itself it had never been shaped into any useful form. Yet the 
two impoverished Austrians believed they could put this un 
workable, brittle, and hitherto useless metal into incandes 
cent lamps as thin, looped filaments. 

Realizing that they could not work tungsten as ordinary 
metals are worked, Just and Hanaman tried various chemi 
cal treating methods, usually at high temperatures. At length 
they hit upon two ways of producing filaments of pure 
tungsten. The most satisfactory process of the two was that 
in which the tungsten was sintered together. The tungsten 
powder was first mixed with a solution of sugar and gum 
arabic. The thick paste resulting was squirted under high 
pressure through a diamond die. As it issued from the die, 
it was caught upon pieces of cardboard in loops. It was then 
treated so that the sugar and gum arabic were turned into 
carbon, which was removed by a further treating process, 
leaving pure tungsten. 

In the final treating, which required a very high tempera 
ture, the tungsten, upon being freed from the carbon, under 
went a curious evolution. The tiny particles became sintered, 
or drawn together, in a form resembling dovetailing; and so 
tungsten filaments of coherent mass, but of extreme fra 
gility, were made. 

At that moment was born the tungsten era in incandescent 
illumination, though to all intents and purposes it seemed 
doomed to strangulation at birth. For Just and Hanaman 
lacked the funds to commercialize their invention. They 
even lacked the funds with which to obtain a patent. In 
1904 they managed to borrow $60, which enabled them to 
apply for British and French patents. They applied for an 
American patent the following year. But they lacked the 
further funds needed to place their invention on the market. 


By that time two other European inventors, Dr. Werner 
von Bolton and Dr. Hanz Kuzel, in Germany, had made 
counter claims. 

In the spring of 1906 Coffin and Rice dispatched two 
scouts to investigate, Dr. Whitney and John W. Howell. 
They went first to Berlin, and there by chance picked up the 

They were walking along the street one evening when 
they observed some unusually brilliant electric lamps shin 
ing in a store window. The shopkeeper was dazed by the 
rapidity of their questions. "Ach so, from the Auergeselschaft 
came those lamps." The Americans hastened to the plant and 
there heard the story of von Bolton's tungsten lamps. 

Tests proved the lamps far better than any Whitney and 
Howell had seen. They made a bid at once for the American 
rights. There the plant officials told them of KuzeFs work, 
and also that of Just and Hanaman. Whitney and Howell 
went straightway to Augsburg, Budapest, and Vienna to see 
the other inventors and in each case to open patent negotia 
tions. They recognized the possibility of a patent conflict. 
There were three inventors. But who had priority? It was 
impossible at that moment to determine. 

General Electric paid $400,000 to Just and Hanaman alone, 
and it expended a total of $1,500,000 before it acquired the 
rights to all three patents, thus assuring the tungsten-lamp 
market in America, regardless of the patent outcome. 

The tungsten-filament lamp was commercially introduced 
by General Electric in 1907. Millions of the lamps were 
sold yearly despite the fragile filament, because they were 
one hundred per cent more efficient than the tantalum lamps. 
The tungsten filament reduced the consumption of electric 
energy from two watts to one per candlepower. The tantalum 
lamp had yielded 370 candles per horsepower; the tungsten 


lamp yielded 746. It was the greatest forward step made in 
incandescent electric lamps. 

A long leap, indeed, from Edison's first carbon lamps, con 
suming seven or eight watts per candlepower and yielding 
but 100 candles per horsepower. Yet something else as 
wonderful was about to happen. Coolidge, at Schenectady, 
had proposed that tungsten could be rendered ductile! 

It was a bold adventure, along a trail many had ignored 
altogether. For tungsten, of all elements, seemed destined to 
remain untamed. It's brittleness was traditional in the scien 
tific world. But in the face of tradition, Coolidge set to 
work. To decisive assertions that it could never be done, 
Coolidge said simply, "Let's try!" 

Experiments carried him through six years, demanding 
more patience than would seem humanly possible. They im 
posed a drudgery that would kill the spirit of anyone lacking 
the true inventive temper. And it cost Coolidge a degree of 
mental concentration that lined his face for life. 

In that expanding workshop where Whitney, the direct 
ing genius, made the rounds of the laboratory rooms each 
day with the exuberant query, "Good morning! Having good 
fun today?" Coolidge began wrestling with that most in 
tractable of metals. Preoccupied from morning to night, from 
month to month, he worked grimly to make it ductile and 
pliable as wire. 

Seldom was there such an unyielding subject. It permitted 
of being bent when heated to a very high temperature, but 
was as brittle as ever when it cooled. Yet Coolidge dis 
covered that at certain temperatures he could hammer the 
metal and elongate it, something no one had ever before ac 
complished. It was enough to save him from discouragement, 
and it started him working in a new direction. 

Coolidge now hovered incessantly around heat-treating 


bottles and electric resistance furnaces, perspiring through 
hot summer days in the dusty laboratory workroom. He 
pressed tungsten in heated rolling mills used by jewellers. 
He pressed it between hot slabs of tungsten steel. He tried 
drawing it through heated dies whose diameter was no 
greater than a thousandth of an inch. With larger dies the 
tungsten rod invariably broke. 

And then late in the fall of 1908, Coolidge first held ductile 
tungsten in his hands! For the pressed tungsten filament, 
drawn through one heated die after another, each slightly 
smaller than the last, and itself heated and reheated, had 
lost its brittleness. It could be bent without breaking, even 
when cold! 

Verifying his experiments, Coolidge found that tungsten 
behaved in a manner that was the exact opposite of all other 
metals. The longer it was worked at high temperatures the 
more ductile it became; whereas other metals under this proc 
ess lose their ductility and turn brittle. Peering through his 
microscope he found that ductile tungsten was fibrous in 
structure, while non-ductile tungsten was crystalline which 
is also the reverse of all other metals. 

But Coolidge's work was by no means finished. It was one 
thing to transform tungsten laboriously in the laboratory; 
it was quite another to work with large quantities of the 
metal on a commercial basis in the factory. It looked as if 
commercial production of ductile tungsten was not to be at 
tained. For two years more Coolidge kept at it, his hopes 
almost smothered. 

He could not produce the right kind of tungsten slug for 
the hot-drawing process. For commercial use, a larger slug 
was needed than the tiny one of pressed tungsten used in 
his experiment. He pressed the tungsten powder in a mold. 
When the pressed slugs were finally obtained, they were so 


fragile that they could not be lifted, and a heat-treating proc 
ess was developed to give them mechanical strength. 

Then the slugs would shrink under the heat, break in two 
or pull away from the clamps. 

They were so brittle, both hot and cold, that nothing could 
be done with them. 

Coolidge, still unbeaten, turned to swaging machines. At 
last he was able to hot-swage the slug as it came, sintered 
into a coherent mass, from the treating bottle, to a diameter 
of an eighth of an inch. The hot-swaging process won the 
day. The slug could now be worked into smaller and smaller 
diameters. Shortly before it reached the die-drawing di 
ameter of 25 one-thousandths of an inch, the tungsten be 
came ductile. It had become a slender gossamer wire, with 
all the properties of wire and only one-sixth the diameter 
of a human hair! 

Small and marvellously strong. The more ductile it be 
came the stronger it grew. When its diameter was measur 
able by thousandths of an inch, it attained a tensile strength 
of 590,000 pounds to the square inch. 

So Coolidge converted an obstinate metal, more brittle 
than eggshells, into a pliable wire smaller than a hair and 
stronger than any other substance known to man. It was a 
transformation as extraordinary as if one were to take flour 
and by some magic turn it into a wire stouter than a steel 

Ductile tungsten, in recognition of the man who tamed it, 
is often referred to as Coolidge metal or Coolidge wire. It 
meant physical stability for the tungsten lamp. And it meant 
a revolution, almost an upheaval, at the Edison Lamp Works. 
Only four years before, the manufacture of the sintered fila 
ment had begun, and now, in 1911, every bit of equipment 
for producing sintered filaments was scrapped. 


Scores of new machines were disposed of in the junk mar 
ket without being unpacked. Half a million dollars was sacri 
ficed in replacing the old equipment with new. Another half 
million dollars was sacrificed in the stock of unsold sintered- 
filament lamps, for which there was no longer a market. 

The public benefited, in reduced lamp prices, as well as in 
the greater efficiency of the lamp. In 1914 the volume of 
electric illumination utilized by the American people repre 
sented a saving of $200,000,000 a year over what its cost 
would have been with the metallized carbon-filament lamp. 
And the cost of light as a commodity was dropping while the 
costs of other commodities were going up. 

The Fire of Prosecution 

IN 1911 GENERAL ELECTRIC was brought into court. It was 
the chief of thirty-four defendants in a suit instituted on 
March 3, 1911, by the United States government, through 
Attorney General Wickersham, "against what is alleged to 
be one of the most powerful and complete monopolies in the 

The arraignment of the "electrical trust," and a recital of 
the attendant allegations, was part of the news of the day. 
The formal charge was a violation of the Sherman anti-trust 
law in the marketing of incandescent lamps. And there was a 
long recital of practices which the government believed to 
be in restraint of trade. 

The attitude of the government in submitting these charges 
can perhaps be best understood by a brief examination of 
its views in the matter. The interest of General Electric in 
the National Electric Lamp Association was disclosed, with 
the intimation that competition between the different com 
panies of the Association, both with each other and with 
General Electric, was artificial rather than genuine. The 
government in its bill of complaint pointed to the practice 



of requiring wholesale houses and dealers to maintain the 
price of lamps as fixed by General Electric, and to the patent 
agreement with the Westinghouse Company, which expired 
that year. 

At first General Electric's counsel planned to make formal 
answer to the complaint. Indeed, such an answer was drawn, 
printed, and filed. It admitted the facts embodied in the prac 
tices cited, but denied that those practices constituted a viola 
tion of the law and particularly denied any intent to commit 
such violation. It pointed out that the National Association 
with its service units was fully able to compete with the 
Edison Lamp Works in research and experiments in all 
branches of lamp manufacturing, and that the stifling of 
competition was what they wished to avoid by financing this 
Association, whose management was entirely in the hands of 
the minority stockholders. 

The bill of complaint stated that General Electric was pro 
longing its patent monopoly of incandescent lamps after the 
patents had expired. This seems to have referred especially 
to the carbon-filament lamp. General Electric replied that al 
though the patents on the lamp had run out, those governing 
the methods of its manufacture were still in force. Every ma 
chine used in the process of manufacture was patented; 
most of them were essential to volume manufacture, and the 
processes could still be infringed upon. 

The government's petition went on to charge that General 
Electric had bought patents and patent rights until it had a 
monopoly of every succeeding type of incandescent lamp 
that had ever appeared. This, it contended, prevented the 
different types the carbon, the metallized, the tantalum and 
the tungsten from competing with each other. The defence 
replied that the different types of lamps could not in any way 
be considered competitive, since each constituted an im- 


provement and was designed to replace the preceding one. 

After filing its answer to the government's charges, Gen 
eral Electric determined that instead of going on trial and 
contesting the charges, endeavoring to justify its practices 
before the court, it would ask permission to withdraw its 
answer and indicate its willingness to submit to a decree. It 
had never had any intention of engaging in illegal practices 
and was therefore willing to have the government pass judg 
ment without pleading its case. It stipulated that it would 
not appeal from the Court's decision. The defendent affirmed 
its "desire and intention ... to comply with each and all 
the provisions of the Statutes of the United States . . . and 
not to operate under or make or carry on any such contracts 
or practices as are condemned by said Act of Congress as 
now construed by the Court." 

Judge John M. Killits, in the Circuit Court, northern dis 
trict of Ohio, permitted this procedure and issued his decree 
on October 12, 1911. In brief, he found that most of the 
practices described in the government's petition were acts 
in restraint of trade, as embraced within the meaning of the 
law, particularly the method of marketing the incandescent 

He ordered the dissolution of the National Electric Lamp 
Association, ordered General Electric to do business only in 
its own name, to refrain from fixing the prices at which in 
candescent lamps might be resold to the public, and to re 
frain from bringing pressure to bear in order to market types 
of lamps lacking any legitimate demand. 

The decree raised a new problem. How could General 
Electric legally continue to market its incandescent lamps to 
the general public without risking arbitrary price manipula 
tion by retail dealers? In the belief of General Electric only 
one basis existed for raising or lowering lamp prices. That 




basis was the rise or fall of costs in other words, a legitimate 
economic basis. Artificial price changes by local dealers, ac 
cording to this theory, would introduce complications result 
ing in an unfriendly reaction by the public toward the in 
candescent lamp. 

To meet this situation the officials of the Company worked 
out the present agency plan of selling lamps by which local 
dealers became agents under contract of General Electric. 
Stocks of lamps were not sold to them outright but taken on 
consignment, the company retaining ownership in the lamps 
while they were in the agents' hands, and relinquishing own 
ership only when the agent sold the lamps. Thus the sale by 
an agent to a consumer was a first sale rather than a resale, 
as formerly, and General Electric could therefore legally con 
trol the price of its product to the consumer. 

The company submitted this plan to the Attorney General 
at Washington, who declined to pass upon its legality. But 
feeling sure of its ground, General Electric put the plan into 
effect in 1912. 

This practice was not passed upon by a court of law until 
1926. The government again brought suit against both Gen 
eral Electric and Westinghouse, charging that the agency 
plan was only an evasion of the decree of 1911 and "intended 
to carry out the same evil result." The case went, on appeal, 
to the United States Supreme Court, which held that the 
agency system as practiced by General Electric was a per 
fectly legal method of selling a product, and that the thou 
sands of dealers holding contracts as lamp agents were in 
effect in the same status as employes of the company and 
entitled to make first sales. 

Meanwhile a radical readjustment in the National Electric 
Lamp Association was going on. Even before the Court's 
decree of 1911 was issued, General Electric had exercised 


its option of purchasing the remaining stock in the Associa 
tion, and became sole owner. 

About this time the Cleveland headquarters were moved 
to Nela Park, just outside of Cleveland, so pronounced in 
natural beauty, so impressive in man's handiwork, that it 
became known as one of the most remarkable institutions 
possessed by American industry. 

The name "Nela" was coined from the initials of the 
National Electric Lamp Association, and so became a memo 
rial to that organization which was dissolved by order of the 
Court, as were all its component companies. They were 
merged into General Electric; the factory at Cleveland be 
came the National Lamp Works of General Electric; the 
other individual factories became simply additional plants. 

Terry and Tremaine had dreamed of Nela Park long before 
they set to work, in 1910, to bring the dream to pass. Terry 
had said to his colleagues: "Let us create a center like the 
well-known institutions of the fine arts to stimulate morale 
and encourage individuality, and at the same time to make 
us better known before the public." The spot they finally 
selected was isolated amid the charm of forest and landscape. 
There, in 1910, they erected a group of buildings harmonious 
in architecture, marked by a dominant simplicity. Twelve 
buildings were arranged in the form of an extensive quad 
rangle, and have often been called "the university of light." 

So emerged Nela Park, self-contained and self -maintained, 
yet consciously linked with the successors of Thomson and 
of Edison. 


The Eternal "Why?" 

IN TAMING TUNGSTEN, Coolidge had revealed an "unknown 
land" where scores of mysterious roads stretched away to 
a beckoning horizon. Both he and Whitney felt their pulses 
quicken as they surveyed the prospect that lay before them. 

They needed new explorers. The call went out to the 
world of technical men, and from institutions far and near 
recruits flocked to the town beside the Mohawk. 

By 1912 the research personnel numbered nearly two hun 
dred. One of the new arrivals was young Irving Langmuir, 
who came in 1909 from Stevens Institute of Technology, 
once presided over by Dr. Morton, one of Edison's severest 
critics. The new recruit came with a fine academic record, a 
prolonged period of training and an inquisitive mind. He 
was one of the "why" men who are ever prying into the 
secrets of the world's mysterious phenomena. Such men 
never outgrow the habit formed in childhood of continually 
seeking answers to their questions. 

Langmuir's curiosity fitted exactly into Whitney's plan. 
There were a great many "whys" to be asked about the in 
candescent lamp. Why did the lamp bulbs sometimes show 



a blackening on their inside surfaces, which interfered with 
illumination? Why did current flow inside the bulb, between 
the legs of the filament, in what was known as the Edison 
Effect? Such questions were to Langmuir as scent is to the 

There were opinions and theories enough about the black 
ening of the bulb. It was commonly believed that minute 
traces of gases remained inside the bulb, and that these 
gases, striking the hot filament, caused it to disintegrate im 
perceptibly, and that the particles of tungsten from the dis 
integrating filament were deposited on the surface of the 
bulb, causing it to blacken. 

Langmuir was not content with theories. He established 
the fact that there were gases in the supposed vacuum, and 
then he determined just what those gases were. Their quan 
tities were so small that he had to invent special instruments 
of the utmost sensitiveness. 

He found that there were five gases in the bulb, water 
vapor, hydrocarbon vapors, carbon monoxide, carbon diox 
ide, and hydrogen; but that the only gas which acted on the 
filament in such a manner as to blacken the bulb was water 
vapor. The next step was to eliminate the criminal. He tried 
to do this by producing a vacuum higher than any previously 
known; from a millionth of an atmosphere he raised the 
vacuum to less than a billionth. And he resorted to endless 
precautions to keep out water vapor. 

But still the blackening went on. Why? He could think of 
only one other possibility the evaporation of the tungsten 
in the filament. 

A succession of experiments proved that the tungsten fila 
ment did lose weight; and that blackening came from its 
evaporation. The water vapor merely facilitated the process; 



Visitors on the balcony of the High Voltage Laboratory watch elec 
tricity put through its more spectacular tricks. 


the other gases, some more than others, retarded or even 
nullified it by reducing the rate of evaporation. 

There might be one gas, thought Langmuir, which would 
so reduce the rate of evaporation as to increase the efficiency 
of the lamp. With the first experiments a new situation arose. 
He succeeded in retarding the evaporation of the tungsten, 
but the dissipation of heat from the filament through gas 
conduction was found to be considerable. Some gases cooled 
the filament so greatly that the amount of electrical energy 
required to maintain the proper temperature of the filament 
rendered them commercially impractical. 

Langmuir now faced the problem of what combination of 
inert gas and tungsten filament would reduce the evapora 
tion of the tungsten and overcome the dissipation of heat 
through the gas. It meant a study of the laws governing heat 
losses through convection, a subject which had hardly been 
touched upon. 

This stage of the investigation yielded surprising informa 
tion. Langmuir discovered that filaments of relatively large 
diameter do not suffer as large a heat loss in proportion as 
do smaller filaments, a fact contrary to expectations. The 
explanation was curious. Langmuir found that the filament, 
when heated in any inert gas, became surrounded by a film 
of hot gas. This film had a thickness independent of the dia 
meter of the filament. If the filament was doubled in size, 
the film became only slightly thicker and the rate of heat 
dissipation did not become doubly greater. It was obvious 
that filaments of larger diameter would be more efficient in 
gases. This, unfortunately, meant a larger quantity of raw 
material in the filament. 

Langmuir continued his experiments until he had satisfied 
himself that argon was the most desirable inert gas for the 


lamp bulb. He also found that the gas should be introduced 
at atmospheric pressure, rather than in a high vacuum. But 
the lamps he made for testing contained filaments that were 
relatively large, and consumed from two to six thousand 

One step was needed to make this the light of lights. Was 
it impossible to use a filament of small diameter? Anyone 
else would have said yes. But Langmuir discovered that a 
coiled filament of small diameter gave all the advantages of 
one of larger diameter. 

Patiently he sought to discover how large a coil was de 
sirable, how far apart the turns should be, and what would 
be the effect on the gas film. The outcome was an incandes 
cent lamp which was twice as efficient as the ordinary 
vacuum tungsten lamp. 

What did this mean? A one hundred per cent advance at 
a single stride! The new lamp required one-half of one watt 
per candle, and gave 1492 candles per horsepower. It was 
fourteen times as efficient as the first carbon lamps. Above all 
it meant that incandescent lamps of large capacity would be 
able to compete with the arc lamp, even with the magnetite 
arc! It was a marvelous illustration of what can come to pass 
when a trained investigator with the gift of asking questions 
starts to answer that enormous little word "Why?" In 1913 
the first commercial gas-filled incandescent lamps made their 
appearance and became instantly popular. 

THE YEAR 1913 saw a number of changes in the executive 
organization of General Electric. Charles A. Coffin, the mas 
ter builder of General Electric's destiny, announced that he 
wished to relinquish his position as President and become 
Chairman of the Board. He had reached the age of sixty-nine. 
For twenty-one years he had been president of General Elec- 


trie. During those years he had combined in his own person 
the positions of President of the company and Chairman of 
the Board, so that his action in 1913 meant virtually a division 
of his responsibilities. It did not deprive the General Electric 
Company of his counsel, nor did it decrease his keen super 
vision over its affairs. 

E. W. Rice, Jr., was made the new General Electric Presi 
dent. It was just thirty years since Rice had gone to Lynn as 
Professor Thomson's assistant after earning his salt by clean 
ing castings and whitewashing walls at New Britain. Now 
he was the head of a great technical and industrial institu 

He entered upon his term of office under far more favora 
ble auspices than Coffin had done, twenty-one years before. 
It was no longer a time of stormy financial skies and of hostile 
criticism. No one was attacking General Electric's methods 
on the ground that there could not be enough business in 
electrical manufacturing to pay dividends on so much capital. 
Yet technical problems there were in ever-greater numbers, 
and these Rice was supremely qualified to handle. Great as 
had been the accomplishments of the past, still greater and 
more amazing things were wrapped in the silence of the 

Soon after the turn of the century, a tall, stalwart recruit, 
noticeably Scandinavian of feature, boarded a train bound 
for Schenectady. He had left Sweden a year or two earlier 
and his English was still halting. It was 1902, years before 
Coolidge or Langmuir had joined the research laboratory, or 
Hewlett had strung porcelain "dinner plates" together and 
called them insulators. 

This recruit eagerly watched the buildings on the river 
flats glide into view from the train window. In his eyes there 
was a look of curiosity and expectancy. But what he was 


searching for, and how successful he would be in finding it, 
time alone would reveal. 

For the next year or two no one heard of Ernst F. W. 
Alexanderson, who was bending diligently over a drafting 
board, then studying the construction of railway motors. His 
soul was there among the armatures and field coils, the 
watts and torques. The world was going to hear from him. 

The thing that first drew attention to his talents was the 
Alexanderson single-phase railway motor of 1904. Not long 
after, when he asked for a chance at another problem, he was 
assigned to the new field of radio. 

Radio in those days was just plain "wireless." Embryonic, 
immature, firing the enthusiasm of the few, its future was 
ignored by the many. But the young engineer from Sweden 
had kept abreast of radio work from those days in 1896 and 
1897 when Marconi proved to an astonished world that mes 
sages could be transmitted without wires. 

Since that time General Electric had worked on radio de 
velopment. It had designed and manufactured efficient spark 
sets working at the low frequency of five hundred cycles. 
This meant that the alternating current which set free those 
mysterious radio waves pulsated five hundred times per sec 
ond. The great need of the moment was for a generator to 
produce alternating current of high frequency. Professor 
Reginald A. Fessenden had attempted to build such a gen 
erator, and finally he took his problem to General Electric. 
The person to attack the proposition with determination was 
young Alexanderson. 

For two years he devoted himself to it. For two years he 
was a target for many a sly shot loaded with that overworked 
word "impossible." But like the rest of his breed, he was in 
vulnerable. At the end of two years, the critics were con 
verted. The first models of his high frequency alternator 


would generate two kilowatts of alternating current at 60,- 
000 cycles. Soon they reached 100,000 cycles. Inconceivable 
that anything in the material universe could pulsate so many 
times in a single second. 

An alternator invented in England had a frequency of 
120,000 cycles, but its output was only a fraction of a watt. 
Alexanderson produced an alternator of 200,000 watts. In 
it, precision was combined with the most delicate adjust 
ments. The alternator when ready for the market was a pon 
derous affair, including a huge rotating disc that weighed 
tons and an array of supplementary parts which maintained 
exact operating regulation. 

The immense rotating disc revolved in strong magnetic 
fields, and its periphery travelled 700 miles an hour; in four 
hours it could have rolled across the Atlantic. There were 
fearful moments in the laboratory when the first models 
were brought up to speed, and furtive glances toward the 
door. Yet for all its speed, that huge disc was not permitted 
to vary from its position by so much as the three-hundredth 
part of an inch. And the speed was kept constant, to the 
smallest fraction of a revolution. 

This first achievement of the young Swedish inventor 
suggested so many possibilities that he forsook other fields 
of electrical endeavor. In 1906 the first installations of his 
pioneer achievement were made; two units were sent to the 
laboratory of John Hays Hammond, Jr., at Gloucester, and 
one to the American Marconi Company at New Brunswick. 

No one could deny that Alexanderson had metamorphosed 
radio. The old spark set and arc generators could transmit 
only short distances; the new alternator created electro 
magnetic waves that leapt the oceans. The 200-kilowatt alter 
nator, completed in 1918, was to become the foundation of 
the first great trans-oceanic radio system. 


Alexanderson aspired to a complete radio transmitting 
system, just as Edison had once aspired to a complete electric 
lighting system. Soon he worked out a magnetic device for 
controlling the energy generated by the alternator under the 
impulse of the dots and dashes of a telegraph key or the 
sound of the human voice. He tested this magnetic modu 
lator, as he called it, for radio telephony, but the electrical 
impulse coming from the microphone was so weak that the 
modulator would not operate. Some means of amplifying 
the impulse must be found. 

Alexanderson turned to the Research Laboratory. Dr. 
Whitney once remarked that "research needs more aviators," 
that is, aviators in spirit. Langmuir, in the laboratory, had 
been "cultivating a flying spirit/' And in so doing he had 
produced a new type of vacuum tube. 

He had made this discovery in following the trail of the 
blackened bulbs. In solving the problem of the gases in the 
bulb he had invented an exhaust pump which produced a 
higher vacuum than ever before achieved. Formerly one 
molecule of air in a hundred thousand left in the bulb was 
thought to be a high vacuum. But Langmuir's mercury con 
densation pump left only one molecule in ten billion. His 
work with the vacuum led to another of those stimulating 
"Whys?" For Langmuir noted that with a certain vacuum 
the Edison Effect occurred, while above that range it could 
not be observed. 

Langmuir was the latest of a brilliant succession to investi 
gate the Edison Effect. Edison himself was the first to notice 
the strange blue glow within the bulb, and he proved it to 
be caused by the passage of current between the legs of the 
filament. That was thirty years earlier. Fourteen years later, 
in 1897, Sir Joseph Thomson, foremost scientist of his day, 
discovered the reason for that phenomenon when he dis- 


covered the electron, the unit of electricity. He studied par 
ticularly the electron in a vacuum, and proved that when the 
filament of an incandescent lamp becomes hot it throws out 
electrons, constituting negative electricity, and that these 
electrons are attracted by a wire or metallic plate connected 
to the positive side of the supply circuit. 

Langmuir went to work at once to design a vacuum tube 
based on his discovery. It could work at 250 volts, whereas 
the old audion tube was limited to 30 volts. It could handle 
watts and even kilowatts, while the previous tubes were con 
fined to small fractions of a watt. It could amplify the impulse 
of a microphone to any volume for radiation as electromag 
netic waves from an antenna; and thus it provided the ampli 
fier for which Alexanderson was looking. 

Edison's discovery of thirty years before resulted in a new 
art, a new industry. Through it were assured practical radio 
telephony and radiobroadcasting. As Dr. Whitney once said, 
"In the advance of civilization it is new knowledge which 
paves the way and the pavement is eternal." 

Through the ages electrons had pursued their course 
through the world without once being detected. Into that 
invisible electronic world, which science now believes is 
identical with the material world, Langmuir was the first to 
enter and make those "big little things" do his bidding. He 
harnessed them for as long as mankind endures. 

When two of Langmuir's fellows in General Electric be 
came aware of what he had done, they applied his electronic 
discovery to further practical ends. One of these was Alex 
anderson; the other was Coolidge, the tamer of tungsten. 

It was Langmuir's radiotron, the first of the high-vacuum, 
high-voltage tubes, which rendered Alexanderson's magnetic 
modulator efficient, and so permitted proper control of the 
energy produced by the high-frequency alternator. Alex- 


anderson could now offer an efficient, long-distance, high- 
power transmitting unit. And his work from 1913-1916 was 
closely watched by Marconi. 

Alexanderson sought to develop equipment for commercial 
companies. He studied the existing apparatus of commercial 
stations, particularly the adaptation of apparatus to his new 
unit, the alternator. The antenna, as Alexanderson himself 
once said, "corresponds to the hull of a rocking boat, with 
which we produce waves in the water." If the hull is heavy, 
greater energy is necessary to rock it because of its resistance. 
And Alexanderson found that the radio antennae of com 
mercial stations possessed unduly high resistance, so that 
they wasted power which might otherwise have produced 
radio waves. 

Alexanderson eliminated the waste by grounding the an 
tenna at equal intervals producing the effect of several an 
tennae connected in parallel, each tuned to the same wave 
length. Resistance was so reduced that 200 kilowatts of en 
ergy had the same effect as 1200 kilowatts upon the former 
type of antenna. This was his multiple-tuned antenna, dem 
onstrated at Schenectady in February, 1916. 

Both radio telegraphy and radio telephony were now prac 
tically possible. Alexanderson immediately designed a duplex 
radio telephone system, and on November 24, 1916, occurred 
one of the earliest demonstrations of two-way wireless te 
lephony, when men at the General Electric plant at Pitts- 
field talked to men at the Schenectady plant. 

The Marconi wireless station at New Brunswick, however, 
brought the Alexanderson system to fame. It was to become 
one of the most notable radio stations in America during the 
Great War. 

But before that time came, one of the most significant 
happenings among General Electric men was the application 


A three-unit General Electric turbine-generator set installed in a cen 
tral station. 




Edison's first lamp, carbon-filament lamp, tantalum-filament lamp, 
drawn-tungsten filament lamp with tip, gas-filled lamp with tip, vac 
uum lamp without tip, gas-filled lamp without tip, inside-frosted lamp. 


by Coolidge of Langmuir's electronic research to Roentgen- 
ray work. 

General Electric had manufactured X-ray equipment since 
Professor Wilhelm K. Roentgen, in 1895, discovered a new 
ray whose properties were so little known at the time as to 
warrant the title "X"-ray. Professor Thomson at Lynn had 
studied the phenomenon and designed the apparatus, which 
went by his name. They were the Thomson inductorium, a 
form of induction coil, and the Thomson Roentgen-ray trans 
former. The latter supplied alternating current to the tube, 
which contained three electrodes, two of aluminum and one 
of platinum. 

No radical change in these tubes took place for fifteen 
years. Then Coolidge, fresh from ductilizing tungsten and 
equipped with the new knowledge of electronic behavior, 
began to study tungsten as an X-ray electrode, or "target," 
in place of platinum. X-rays of greater power were produced, 
as high voltages were employed. 

Day after day Coolidge entered a lead-lined booth with a 
window of leaded glass through which he watched the ac 
tion of his tubes. Lead offers an impassable barrier to the 
X-ray and hence constitutes a safe screen for the research 
worker. Outside the room was a danger zone, filled with 
powerful electrical emanations from the tube which was 
under test. But the worker in his tower was able safely to 
take notes of every change in the conditions governing the 

Finally in 1913 Coolidge brought out a tube that operated 
at 100,000 volts or higher; a tube in which the X-rays were 
created by a bombardment of electrons from the tungsten 
electrode; a tube which could be kept in operation for hours 
and whose intense rays were capable of a high degree of 


YET RESEARCH WAS ONLY ONE of many departments of ac 
tivity. Even as Coolidge was fashioning a new tool in thera 
peutics, a group of men in Hewlett's office in the Switch 
board Engineering Department was talking over a control 
system for the mighty locks of the Panama Canal. 

Here was a task fit for electricity to make available to 
commerce this enormous ditch, itself one of the great engi 
neering feats of human history. It was a task involving the 
movement of gates weighing four hundred tons, the largest 
more than seven hundred; the filling and emptying of locks 
a thousand feet long, three hundred feet wide; and the con 
trol of operating mechanisms scattered over fifty miles, the 
largest single group extending for more than a mile. 

These problems were further complicated by the tropical 
climate. Metal parts must be made noncorroding, coils must 
be made moistureproof. Maximum reliability was impera 
tive, because the Panama Canal was far removed from manu 
facturing centers and weeks must pass before repair parts 

Edward Schildhauer, engineer of the Isthmian Canal 
Commission, had asked every electrical manufacturer in 
America to submit a bid. The rejoicing in General Electric 
offices over winning the contract, however, was brief; coats 
were taken off and sleeves rolled up, for intensive days of 
planning lay ahead. 

No nation in the world could have asked for a more per 
fect test of the electrical giant at work, for if electricity could 
operate the Panama Canal, surely it could do anything under 
the sun! 

It did operate the Canal and is still doing it. Five hun 
dred motors, rating from seven horsepower up to seventy, 


are on duty at the great locks, and more than five hundred 
others are at work at the dams, the spillways, and elsewhere 
along the canal a total of over a thousand electric motors, 
with a combined capacity of nearly 30,000 horsepower. 

And the motors were the easier half of the problem. The 
real task was the control system. It must be centralized at 
one spot for each of the locks; it must be an exact duplicate 
in miniature of the locks themselves, their gates, their valves, 
their fender chains. Everything that happened at those big 
locks must be repeated simultaneously on the control boards. 
The operators must have before them a complete record of 
the movement of those ponderous parts, a perfect gauge of 
the rise and fall of the water, accurate to one-tenth of one per 

Nothing short of infallibility was demanded. Errors by 
the operators, endangering lives and property on board ves 
sels passing through, must be made impossible. 

For previous canals there had been a large force of opera 
tors scattered along the canal-way, each charged with one 
control operation. Safety from error increased as the number 
of human beings increased. But control for each lock cen 
tered in one man's hands, and that man had to proceed with 
out error this was a staggering demand. 

Edward M. Hewlett, designer of switches and controls, 
went tc work over the requirements quietly. Upon Hewlett's 
assertion that he could make this system work, the control 
contract for the Panama Canal was awarded to General Elec 
tric. Now he had to live up to his promise. 

First Hewlett consulted with the engineer of a concern 
that made interlocking railroad switch devices, to obtain a 
bid on that part of the work. 

"You people don't know anything about interlocking de- 


vices 1" exclaimed the engineer. "We know the interlocking 
business; you don't. You're suggesting something that can't 
be done at the price you require." 

"Not at all," Hewlett replied with some spirit. "We were 
building interlocking apparatus when you were in short 
trousers, and what's more, we're in the business of doing 
things that can't be done!" 

There followed weeks of reading, designing, experiment 
ing. Within six months Hewlett had a working model of the 
indicating system and the interlocking system as well the 
system that was accepted and operated when the canal was 
opened on August 15, 1914. It has been working there ever 
since. A system of many parts, it was yet not fundamentally 
complicated. It was made up of vertical and horizontal bars, 
two and a quarter miles of them, whose movements were 
governed by mechanical stops. 

There was keen ingenuity in the way the movements of 
those bars were planned. And in the interlocking principle 
which ran through an intricate network of those lug-stopped 
rods, governing dozens of control operations, permitting 
only one sequence of action in the order of turning the con 
trol handles upon control boards as long as sixty-four feet. 
This was Hewlett's selsyn system. 

PARTICIPATING LUSTILY in the work of the world, rubbing 
shoulders daily with cosmopolitan interests, a buyer and a 
seller in many markets, General Electric like other groups of 
humans came in for its share of controversy. Situations were 
continually arising which could be cleared up only by the 
rulings of the law courts. 

On one such occasion, counsel for General Electric met its 
match. General Electric was in litigation with Stone and 
Webster of Boston, counsel for whom was Tyler and Young, 

President of General Electric from 1922 to 1940. 

Chairman of the Board of General Electric from 1922 to 1940. 


and the case was handled in court by the junior member of 
the firm, Owen D. Young. 

He was a comparative stranger in Boston, although his 
associates remembered vaguely that he had been born and 
raised a farmer's boy in northern New York. And they knew 
that Charles H. Tyler had given him a start upon his gradua 
tion from the law school of Boston University. Within ten 
years Tyler had admitted him to partnership. 

The General Electric lawyers did not rate Young as a 
formidable opponent. He sat slouched in his chair, "his long 
legs lost in the shadows of the table." But when he towered 
before the court on those long legs, his slouch gone, head 
erect, they came to recognize not only his physical domi 
nance, but his penetrating discernment and the clear-cut 
effectiveness of his arguments. When the trial was over, 
Owen D. Young was a name to be mentioned with respect 
among General Electric people. It reached the discriminating 
ears of Charles A. Coffin; and for a year or two Coffin watched 
him at a distance, noting the working of that keen mind, 
meditating upon the personality that had gotten the better of 
General Electric lawyers. 

Coffin thought again of this display of talent and character 
when, in April, 1912, they brought him the news of the tragic 
death of Hinsdill Parsons, the brilliant lawyer who had been 
Vice President and General Counsel of General Electric for 
eleven years. During this time Young had made himself 
well known in and around Boston. He had, in fact, ably 
handled private law business for at least one of the Directors 
of General Electric. 

So, one day, Coffin telephoned to Young and asked him to 
call. And he offered the young lawyer the position of General 
Counsel of the corporation. 

Young accepted. Some years later he confided to one of 


his colleagues part of what lay back of that acceptance. On 
Young's frequent trips from Boston to his boyhood home in 
New York State, his train passed close to the General Electric 
plant in Schenectady. Often, as he passed, he noticed the 
busy stir and activity. He sensed the feel of constructive 
effort, of building something useful. And sometimes, tired 
with the eternal bickering of the law courts and the litigation 
over affairs past and done with, he looked longingly at the 
Schenectady plant and wished that he could have a part in 
some such work, doing something truly constructive, helping 
to build something of lasting value. So, although Coffin's 
offer came as a surprise to Young, the idea was neither new 
nor unwelcome. 

So came to General Electric this "tall rangy man of 
Lincolnesque height," Coffin's eventual successor as Chair 
man of the Board. He had come to help it grow and to grow 
with it. 

Indeed, the growth of General Electric was so well- 
balanced and so healthy that it was continually able to 
measure up when new opportunities arose for yoking elec 
tricity. That was one reason why its salesmen and its engi 
neers had just convinced a group of electric power prospects 
in the northwest that electric motors could run their trunk- 
line railroad. 

The Chicago, Milwaukee and St. Paul Railroad extended 
far west of Milwaukee and St. Paul, climbed the great 
continental divide, penetrated the Rockies, wound upward 
among the Sierras, gathered its energies to cross the Cascade 
range, and halted only on the shores of the blue Pacific, at 
the northwest metropolis of Seattle. It passed through moun 
tains that were exhilarating to passengers, but trying upon 
the capacities of its steam locomotives, particularly in the 
northwest winters. 


Now electricity was to be given a chance. The inheritance 
of all the early traction pioneers was embodied in the electric 
motors that set out to conquer the Rockies. It was December, 
1915, when the first electric locomotives took up their task 
on the St. Paul road. Nothing like them had been seen in that 
region before. The forty-two freight locomotives, weighing 
280 tons each, were as massive as the stoutest steam hauler on 
the road. Each was able, through its twelve gearless motors, 
to haul a freight train weighing 2500 tons over steep grades 
at fifteen miles per hour. In 1920 electric passenger locomo 
tives that weighed 260 tons apiece and were able to haul 
twelve cars weighing a total of 960 tons at twenty-five miles 
an hour, joined in the work. On 400 miles of mountain 
trackage these giant electric locomotives hauled their load 
with ease. One of the grades on the route extended for 
twenty-one miles at a rise of two per cent, but the electric 
giants mastered it easily. 

The new locomotives worked on the principle of regenera 
tive braking, which put a new item of economy in the budget. 
On the mountain divisions conditions for this type of braking 
were ideal. The gearless motors, the shafts of whose armatures 
were the axles of the wheels, after mounting those long 
grades, automatically reversed their action on the "down" 
side. Their rotation was no longer produced by the electrical 
impulse but by gravity. Instead of converting the electrical 
energy of the supply line into mechanical energy, they con 
verted the mechanical energy created by the train's motion 
on the downgrade into electrical energy. And the process 
slowed up, or braked, the momentum of the heavy cars. 
Electricity produced by those generator motors constituting 
a saving of about twelve per cent in total power consumption, 
flowed back into the electrical transmission system, speeding 
over the distribution network as a contribution to the motive- 


power required by electric locomotives elsewhere on the 
line. A perfect operating economy electricity helping to 
pay its own bills. 

FROM THE BEGINNING automatic action had been important 
in electrical devices. Automatic action regulated arc lamps, 
kept current constant in dynamos, became the essence of the 
relay, the electric meter, the transformer. 

When Alex Dow, President of the Detroit Edison Com 
pany, took one of his problems to General Electric engineers, 
they were interested at once. Dow wanted to apply the 
automatic idea in such a way that he could extend his lines to 
the outskirts of Detroit, where new homes were springing 
up though population was sparse. 

The ordinary substation in such a district, with its pay 
roll, would be too costly in proportion to the possible busi 
ness. What Dow wanted was a station that could be con 
trolled from a distance, through automatic action. That was 
what the engineers designed for him. In less than a year's 
time this remotely controlled substation was at work in 
Detroit. It was situated on Rowena Street and controlled from 
the generating station on Elizabeth Street, a mile away. It 
distributed electrical energy up to 500 kilowatts; and except 
for an occasional inspector, no human set foot within its 

Two years later the completely automatic substation ap 
peared. An engineer in General Electric's Chicago office, 
Edward Taylor, designed the first of this type, which began 
to operate at Union, Illinois, in December 1914. It was magic 
both in its working and in its effect upon operating costs. 
Here were machines which began working when they were 
needed, and stopped working when they were not. 

Picture an electric car on that road running far out from 


the main generating station, farther than direct current can 
be transmitted economically. Presently it draws near the 
automatic substation. At a certain point its approach actuates 
relays. Instantly the synchronous converters, by which alter 
nating current is converted into direct, go into action. 

In swift sequences, handled entirely by automatic control, 
the converters are connected through transformers to a 
33,000-volt supply line, then to a 600-volt trolley line and the 
speeding car draws the motive power it needs. When its 
power is supplied from the next substation, the current in the 
trolley wire at the first substation drops below a certain 
point, causing the automatic control to shut down the con 
verters until the next car approaches. 

In less than ten years the total installed capacity of 
equipment of this type reached half a million kilowatts. On 
interurban lines especially, the cars take their power from 
several automatic stations in succession. The automatic gen 
erating station was to grow out of this at a later period. 

Almost as marvelous was the growth of the electric genera 
tor itself. In 1902 people spoke of generators that fed 7500 
kilowatts of electricity into transmission lines as "mammoth." 
In 1914, 20,000 kilowatts were generated by one General 
Electric unit. What do such capacities mean to practical folk? 
They mean that whereas in 1904 one generating machine 
could supply enough electric current to light 100,000 lamps, 
in 1914 it could supply 1,700,000! Only twenty-five years 
before, a dynamo supplying 150 incandescent lamps was 
thought to be wonderful. Capacities had increased eleven 
thousand times in a quarter century! 


Conservatism Routed 

I HAVE TONIGHT seen the greatest revelation of beauty that 
was ever seen on the earth," exclaimed Edwin Markham, the 
poet, one February evening in 1915. 

"I may say this, meaning it literally," he added, "and with 
full regard for all that is known of ancient art and architec 
ture, and all that the modern world has heretofore seen of 
glory and grandeur. I have seen beauty that will give the 
world new standards of art and a Joy in loveliness never 
before reached. This is what I have seen the courts and 
buildings of the Panama-Pacific Exposition illuminated at 

The illumination which so inspired the poet was largely the 
work of Walter D'Arcy Ryan, director of General Electric's 
Illuminating Engineering Laboratory. 

Ryan had tackled illuminating jobs of this magnitude be 
fore. In 1907 he had supervised the lighting of Niagara Falls 
a feat that required the installation of batteries of 
projectors giving a combined illumination equivalent to 
1,115,000,000 candles. It, too, had been a huge success. The 
throng that gathered on the opening night was so large that 



the suspension bridge swayed perceptibly and traffic had to 
be halted. The roar of the cataract came through the dark 
ness, the cataract that could not be seen. Then without warn 
ing the Falls leaped out of the night, a vast, shimmering mist 
of plunging water, gleaming in the concentrated light. The 
thousands of spectators stood in awe and silence. 

But Ryan encountered new difficulties when, in San 
Francisco, he presented his plans before the architects, de 
signers, and color artists who were involved in preparation 
for the Exposition. His proposals seemed so fantastic that 
there was scarcely a detail which was not opposed. For 
Ryan's theory was founded upon the idea that to run strings 
of incandescent lamps along the edges of buildings so that 
they would be outlined by dots of light, was crude and 
archaic. His plan was twofold: first he proposed to throw a 
flood of light upon the fagades of the buildings; and second, 
he proposed a scheme for depth, or "shadow" lighting. 

A terraced tower, four hundred feet or more from a square 
base with broad archways, dominated the wide array of 
picturesque Spanish mission buildings. Ryan's idea for light 
ing this tower was even more fantastic. He wanted something 
lively and prismatic. Immediately he thought of imitation 
cut-glass jewels played upon by powerful searchlights. He 
instituted a search which finally uncovered a type of jewel 
made in imitation of diamonds, rubies, sapphires, and emer 
alds; and the glass was so cut as to possess a high index of 

Many cunning hands had worked over those gems. Native 
workmen among the hills of Austria had kept the craft in 
their families for generations, treating the original uncut 
pieces by heat and glass-blowing methods of which they 
alone knew the secret. In 1913 there came from distant 
America an order for no less than 130,000 jewels! 


The Austrian craftsmen gasped, but in time the order was 
filled. Ryan called the jewels "Novagems." 

His laboratory workshop, crowded with color paintings of 
the buildings as they would appear when illuminated, color 
charts, and booths for demonstrating special illumination 
effects aroused the greatest curiosity. Some of the other 
specialists felt that Ryan was usurping the functions belong 
ing to others. Upon returning from one of his trips East, 
Ryan found that his critics had burst into open condemna 
tion of his plans. Someone had tried a few of his jewels upon 
the tower and found that they cast shadows. The architects' 
commission wanted to throw all the jewels bodily into San 
Francisco Bay. 

Ryan promptly demonstrated by means of photographs 
that, when properly mounted, the jewels would not cast 
shadows, while incandescent lamps, strung up and down the 
tower would. Then he was told that his idea of placing 
heraldic banners along the avenues to add color in the day 
time and to screen floodlight units at night would cheapen 
the exposition into a mere Coney Island spectacle because 
the banners would flap. Ryan replied that he intended to 
hang fifty-pound weighted tassels from each banner so that 
no ordinary breeze would disturb them. 

Another outcry went up when he disclosed his plan for a 
steam scintillator to be played upon by a great battery of 
electric searchlights. He had arranged with the Southern 
Pacific Railroad to have a steam locomotive, painted a rich 
cream color, backed upon a pier on the waterside front of the 
exposition. Shell pits were dug and forty-eight big search 
lights were placed in position, their total capacity equalling 
2,600,000,000 candles. Surely "this man Ryan" was running 
wild! The architects' commission started a movement to have 
the locomotive removed. 


But Ryan would not be stampeded. He stoutly defended 
his "fireless fireworks." The locomotive and the steam scintil- 
lator remained. 

But exposition officials were secretly as fearful as the 
architects and designers. They hoped that Ryan was right, 
while they quietly consulted a local illuminating expert and 
had an alternate plan worked out down to the last detail, 
ready to be rushed into the gap should the Ryan plan fail 
upon trial. This alternate plan was based upon strings of 
incandescent lamps. 

At last came the night for the trial exhibition, the night 
of February 15, 1915, three weeks before the date of the 
official opening. Upon one of the wide avenues the official 
party gathered. Dusk slowly began to fall. A silvery shaft 
leaped into the darkening sky. 

Then, in a burst of glory, a scene of romantic beauty un 
folded, as palaces and halls stood out richly down long 
thoroughfares of light. Radiance glowed from every window, 
as if the interiors were brightly lighted. Deep rose tints 
bathed the recesses in archways and columned porticoes, 
where forbidding shadows usually gathered. It was a city of 

In the center rose the iridescent Tower of Jewels. And 
every pool and lagoon caught and reflected the glimmering 
tower and the fairy palaces. A 20th Century Aladdin had 
rubbed his lamp. 

The international jury of awards of the Exposition did 
something never before dreamed of. It adjudged the illumina 
tion of the Exposition to be a "decorative art." 

TURNING TO A FAR DIFFERENT FIELD of the electrical arts, the 
vear 1915 also saw the battleship New Mexico launched. 
Back in 1909 Emmet had felt that the turbine-electric drive 


could be designed for sea craft. Units of 14,000 kilowatts 
were already a reality. He turned to the United States Navy, 
knowing that there he would find vessels large enough to 
receive maximum benefit from the turbine-electric drive, as 
well as financial resources sufficient to undertake so large an 

But the Navy was not interested in turbine-generators and 
electric motors for its vessels. Steam propulsion had been the 
traditional method for so many years, and too many naval 
men had been bred in the sacred belief that there could be no 
substitute. No one in the Navy Department at Washington 
took Emmet seriously. Three colliers were at that time being 
equipped, and preparations for installing other types of pro 
pulsion went forward without interruption. The Cyclops was 
launched, with a marine engine to propel her. A little later 
the Neptune followed, equipped with a geared turbine as her 
driving unit. Then Emmet, pleading fair play and an equal 
chance with his competitors, won a reluctant consent to the 
installation of a turbine-generator upon the Jupiter. 

Naval officials did not for a minute believe the Jupiter 
could be propelled efficiently by electric motors. They even 
had the vessel designed so that the mode of propulsion could 
be altered to the marine-engine type when the anticipated 
breakdown of the electric drive occurred. 

In this wise was the Jupiter built at the Mare Island Navy 
Yard at San Francisco, and put in commission in 1913. Her 
commanding officer was Commander C. S. Kernpff, and her 
chief engineer Lieutenant S. M. Robinson. Aside from Lieu 
tenant Robinson, only one man on board, the chief electri 
cian, had any conception of the nature of the electrical equip 
ment. The apparatus was handled by enlisted men who had 
to be instructed from the ground up in their work. 

But machinists' mates within a week's time learned how to 


operate the electrical apparatus, and Commander KernpfFs 
report was an enthusiastic endorsement of the new method of 
propelling naval craft. 

The report caused a stir in naval circles at Washington. 
The officers of the Jupiter were immediately replaced by 
others to check their sanity in expressing such enthusiasm. 
The new officers turned in an equally hearty endorsement. 

But naval officials at Washington were not easily satisfied 
and accordingly they detailed two officers to serve on the 
Jupiter as inspecting officers, whose sole duty was to observe 
and report on the electrical equipment. Their report was even 
more complimentary than its predecessors. 

With that, criticism temporarily subsided. Emmet, follow 
ing up the advantage he had gained, recommended that 
turbine-generators and electric motors be provided for the 
battleship New Mexico, now under construction. The pro 
posal immediately raised an outcry. It seemed that the new 
radicals would stop short of nothing. It was all very well to 
turn turrets, operate deck winches, revolve capstans, or 
hoist ammunition by electricity. And all very well to propel a 
vessel no larger than the Jupiter. But a battleship! It couldn't 
be done. It would require generators too large for the ship. 

Emmet, undaunted, drew up a definite proposal on behalf 
of General Electric and submitted it to the Navy Department, 
but without avail. Then he interviewed the Secretary of the 
Navy, Josephus C. Daniels, and persuaded him to allow 
estimates to be prepared showing the cost of equipping the 
New Mexico for electric drive as compared with the cost of 
installing steam-propulsion machinery. 

Emmet took his figures to the industrial engineer of the 
Brooklyn Navy Yard, who had been ordered to draw up the 
estimates. This was Admiral George E. Burd, who was fair- 
minded and had a passion for facts. Admiral Burd's estimate 


swayed the issue, for it stated that turbine-generators and 
electric motors on the New Mexico would cost a quarter of a 
million dollars less than either the geared turbine or the 
marine engine. 

Emmet appeared at a conference in Secretary Daniels' 
office and spoke at length, drawing on his ten years' experi 
ence in designing and installing Curtis steam turbines and 
generators. His figures had been tested and verified, and he 
pointed out considerations which others had lost sight of. In 
the end, Secretary Daniels gave him the contract. 

In 1915 the New Mexico was launched, a powerful floating 
fighting machine of 32,000 tons. She burned oil for fuel, and 
her dozen big tanks had a capacity of a million gallons. She 
required only nine boilers, while other battleships of her 
weight had twelve. Steam pressure on the boilers ran to 280 
pounds, which meant that turbine pressures had risen 105 
pounds since the Chicago turbine of 1903. 

As for electrical equipment, its aggregate weight was about 
600 tons. There were two main generators, each rated at 
15,500 horsepower, and they supplied four propulsion motors, 
each with a 7000-horsepower capacity. The generators were 
each driven by a Curtis steam turbine, so that only two tur 
bines were necessary, while the battleship Pennsylvania, with 
the geared-turbine drive, required ten. 

At ten knots the New Mexico consumed about twenty-five 
per cent less fuel than the best turbine-driven ship that pre 
ceded her. And she was an all-electric ship. Electric motors 
operated the steering gear, boat cranes, winches, capstan, 
refrigerating system, laundering machines, water pumps, oil 
pumps; they raised anchors, revolved the turrets, hoisted am 
munition, compressed air, turned the fans and blowers; they 
supplied the machine shop, carpenter shop, printing shop, 
and kitchen; and electricity heated the air, cooked toast, 


boiled coffee, supplied telephone systems, operated the bat 
tery of searchlights, illuminated the entire ship. It was a 
veritable electric home, in which 1100 men found the tasks 
of Me amazingly simplified. 

It was a triumph indeed. Yet electrical engineers admitted 
that the original prejudice of the Navy was no more than 
natural. E. W. Rice, Jr., President of General Electric, spoke 
later of the success to a group of Navy men. 

"That electrical gear," he said, "should by any possibility 
prove to be more generally efficient seems a paradox . . . 
It is due, we believe, to the fact that in the electric drive an 
open air space is substituted for the teeth of the mechanical 
gear. Although we remove the teeth, we are able to transmit 
tremendous power without any mechanical connection . . . 
The electrical method substitutes an indestructible, infinitely 
elastic air space for the teeth of the mechanical gear, and 
obviously the air space cannot be bent, or broken, requires 
no lubrication, and contains nothing to rub or to get out of 


A Period Ends 

CLOUDS OVER EUROPE brought sudden problems to the 
United States as April 1917 came in. In her youthful strength 
America was marching into the vortex of a titanic human 
struggle that flamed destructively across a continent. 

America was separated from the scene of conflict and 
from all her allies by three thousand miles of ocean. She must 
talk continually to those allies, to her military and naval 
leaders, though an ocean intervened and though the enemy 
attacked the underwater cables. 

There was but one way to do it radio. Every army and 
naval radio station in the country was instantly keyed up for 
service. Yet few of them were equipped to span the ocean. 
Transoceanic radio communication was still embryonic. But 
now it was imperative. 

In this dilemma one station alone stepped into the gap. 
The American Marconi Company, affiliated with the British 
Marconi Company, agreed to give a practical trial of the 
Alexanderson alternator at its large transmitting center 
at New Brunswick. Alexanderson had developed his 200- 
kilowatt alternator, and a complete unit of this size was 



rented to the Marconi Company, replacing the old 50- 
kilowatt unit at the Marconi station. In January 1918 this be 
came for the time being a United States naval radio trans 
mitting station, operated by naval personnel. 

War continued through the spring and summer of 1918. 
The New Brunswick station bridged the ocean daily in the 
transaction of government war business. But soon the opera 
tors at receiving stations both in America and Europe found 
themselves in a new predicament, created by the activities 
of the enemy. 

The large German transmitting station at Nauen was "jam 
ming the air," that is, it was sending out interfering waves 
for the purpose of choking incoming messages on the receiv 
ing antennae of the American and allied stations, thus pre 
venting communication. The scheme succeeded often enough 
to make things disconcerting. 

Radio engineers of the allied nations set to work to over 
come this interference. Alexanderson soon developed his 
barrage receiver, which was a unidirectional system so ef 
ficient that it could eliminate the signals of a high-powered 
sending station operating close by. One unit of this device 
was sealed in the diplomatic mail of the United States gov 
ernment and sent to France. Another was installed in the 
United States government receiving station at Bar Harbor, 

Meanwhile the station at New Brunswick had been bom 
barding Germany and her allies with a new sort of missile 
radio messages and bulletins relating to the war prepara 
tions which America was making. The constant arrival of 
American troops in France, the launching of ship after ship, 
the raising of billions for waging war all this was poured 
upon the air for assimilation by German stations. It was an 
offensive against morale, a psychological attack. 


Late in October, 1918, the last great German offensive had 
been stopped. The attackers of heroic Verdun had again 
been checked, and the American divisions were steadily 
hammering back German lines which had been thought im 
pregnable. The tide was slowly, surely turning. At that mo 
ment there leaped upon the air, to be picked up alike by 
Nauen and the allied stations, a set of call-letters from the 
American station at New Brunswick. 

"POZ . . . POZ . . . POZ ... de ... NFF" buzzed the 
receivers. POZ was the call for the German station at Nauen, 
and NFF was the station at New Brunswick. The two had 
not been on speaking terms for a long time. 

The Nauen operator replied a moment later: "Your signals 
are fine, old man." 

Whereupon the "old man" in New Brunswick dispatched 
in English the first of President Wilson's statements to the 
German people, carrying the ultimatum that the allies would 
conduct no negotiations for an armistice or for peace with 
the German government as then constituted. In other words, 
Kaiser Wilhelm would have to step down. 

"Thereafter," says the New York Post, "Washington was 
in constant communication with Berlin. President Wilson's 
memorable 'fourteen points' were broadcast to Germany 
from New Brunswick. Wireless was making history at a 
faster pace than all the engines of destruction. Innumerable 
electrical impulses, flashing back and forth across the At 
lantic, were settling the war. It might have taken a month 
longer to negotiate the armistice if it had not been for those 
radio exchanges." 

The war ended, but radio went on. As Alexanderson said, 
"It was the war which brought radio into its own. Before the 
war it had been interesting. With the advent of the war it be 
came vital." Its demonstrated usefulness was only beginning. 

About 11 years after his retirement as President of General Electric. 

The buildings that house the Research Laboratory, at Schenectady. 


For a few months after the armistice was signed the most 
significant events in American radio centered around the 
transmitting station at New Brunswick and its peculiarly ef 
fective radio generator, the Alexanderson alternator. 

Efforts to obtain customers for the high-frequency alter 
nator had been made by General Electric before America 
entered the war. The government was first approached, but 
was already committed to the Poulsen arc. At the close of 
the war General Electric recommenced its efforts to sell 
equipment to the Marconi Company. Formerly the terms had 
not been agreed upon. The Marconi Company insisted upon 
an agreement by which it would obtain a monopoly on the 
Alexanderson alternator. Under this agreement they placed 
an order amounting to $5,000,000. 

General Electric felt no hesitancy now in entering this ex 
clusive agreement, as with one exception no other customer 
had been found for the costly alternator. The Swedish gov 
ernment planned to erect a powerful station, and the Swedish 
minister at Washington, W. A. F. Ekengren, had begun 
negotiations with General Electric. 

But the Swedish government wanted only one alternator. 
The Marconi Company was a far larger customer, and in 
acquiescing to its proposal, General Electric was obliged to 
terminate its negotiations with Sweden. Minister Ekengren 
was seriously disturbed. He knew what an advantage that 
alternator would give to radio development in Sweden. He 
called upon Rear Admiral William Bullard, then Director of 
Communications for the Navy. Admiral Bullard became as 
disturbed as Minister Ekengren, but for a far different reason. 

Familiar with the merits of the alternator, he knew that 
the American Marconi Company, despite its name and loca 
tion, was dominated by the British Marconi Company. He 
foresaw what radio would become within a few years; and 


he wanted America, not a foreign country, to have control 
over the alternator which was at that time the most practical 
device for producing continuous radio waves at high power. 

This was the situation in March, 1919, when E. P. Edwards, 
later Manager of General Electric's Radio Department, re 
turned to Schenectady from New York with the necessary 
data to draw up the contract giving the Marconi Company 
exclusive purchasing rights to the Alexanderson alternator. 
As he began his work, he was called on long distance by 
Commander Hooper, in charge of the Radio Section of the 
. Naval Bureau of Engineering, who besought him to delay 
negotiations until a conference could be arranged. 

That conference, a momentous one in the history of Ameri 
can radio, took place on April 5, 1919. E. W. Rice, Jr., Presi 
dent of General Electric; Charles A. Coffin, Chairman of the 
Board; Owen D. Young, Vice President; Albert G. Davis, 
Vice President in Charge of Patents; and others, were met 
by Admiral Bullard and Commander Hooper. 

Admiral Bullard was spokesman for the Navy. "Gentle 
men," he said, "we are satisfied that the General Electric 
Company, through the Alexanderson alternator, has de 
veloped the most perfect system of radio communication in 
existence. We should like to make use of it ourselves, but we 
cannot secure the necessary appropriation. We know that 
you are negotiating with the Marconi interests, expecting to 
sell this apparatus exclusively to them. We are here to per 
suade you on the grounds of self-interest and patriotism not 
to do it!" 

Then he pointed out what such a step would mean, in 
giving to Great Britain the control of world radio com 
munication. In effect, his message was: "Save radio control 
for America!" 

Owen D. Young, as spokesman for General Electric, re- 


plied: "You have placed a definite problem before us. You 
tell us that we have a wonderful system, that the Navy itself 
cannot buy it, and that we should not sell it to the only com 
mercial company which is prepared to buy it. 

"We realize the patriotic note in your appeal; but what do 
you want us to do with this apparatus? Do you want us to 
put it to one side, where neither ourselves nor the public will 
derive any benefit from it?" 

"We did not come without an alternate plan," replied Ad 
miral Bullard. "I would recommend that your company 
should either exploit this system directly, or should organize 
a subsidiary company to do so." 

The appeal struck home. General Electric broke off nego 
tiations with the Marconi Company. 

Next the General Electric officials launched a new com 
pany for radio communication, rather than put their own 
company into a field foreign to its province. They purchased 
the block of stock in the American Marconi Company which 
was owned by British interests. Then the American Marconi 
Company was purchased by General Electric in behalf of the 
proposed new company. 

The new concern was established around December 1, 
1919, under the name of the Radio Corporation of America. 
It was to be the selling agent for General Electric's radio 

General Electric officials now conceived of a mobilization 
of all American radio and communications interests behind 
the new company, and undertook negotiations which led to 
the participation in the Radio Corporation of the American 
Telephone & Telegraph Company, the Western Electric 
Company, the Westinghouse Electric & Manufacturing Com 
pany, and the United Fruit Company. 

This procedure adjusted a patent dilemma which was as- 


suming a gravity reminiscent of the deadlocks of 1890 and 
1896. Now, in 1919, the conflict of patents held by compet 
ing companies, yet necessary for the operations of all, would 
have crippled radio communication in all its aspects. Such 
an outcome would have been calamitous, for radio was di 
recting the public mind to technical innovations in a manner 
without precedent. 

The marvels which electricity was producing were meet 
ing with far less prejudice and a far readier welcome than 
did those of three decades ago. People had become electri 
cally minded! 

As STEINMETZ ENTERED the high-studded laboratory room, 
with its wide windows and its maze of apparatus, conversa 
tion ceased. Heads turned in his direction, attention was 
riveted on that impressive figure, diminutive in stature, yet 
suggestive of a power beyond physical strength. 

"Good morning! Good morning!" he exclaimed. "How are 
you?" Then, to his adopted son, J. LeRoy Hayden, he added, 
"What's new?" 

It was the greeting with which he had opened the day at 
office and laboratory for nearly thirty years. But on this 
morning, Steinmetz knew better than anyone else what was 
"new." He was about to demonstrate his famous lightning 

For weeks he had been investigating lightning with this 
strange device. His purpose was to discover new knowledge 
which he could pass on to the designers of lightning ar 
resters, so that these watchdogs of transmission systems might 
be made more reliable. As he picturesquely termed it, they 
are the traffic officers; lightning is the criminal of these 

In the study which he had been making, the lightning 


IN 1925 



generator was but a means of an end. Yet it was spectacular 
in itself. 

There were many who wanted to see, on this winter's 
morning of 1922. And what they saw were rows of large 
glass panes, connected by wires to a group of glass tubes 
and by other wires to a circuit, or "discharge path," in which 
was placed a small tree limb. The glass plates were the con 
denser, which stored up electricity as does a thundercloud 
until the point was reached when it could hold no more and 
electricity was discharged in a quick flash that was the 
artificial lightning-bolt. The lightning "flashed," the thunder 
"roared," and the tree limb was split into fragments. It was a 
perfect imitation, in miniature. 

"We get a discharge," said Dr. Steinmetz, "of 10,000 
amperes at 120,000 volts; that is, over one million horse 
power, lasting for a hundred thousandth of a second. This 
gives us the explosive, shattering effect of real lightning." 

According to his calculation, natural lightning might rep 
resent as little as 3000 amperes but with a hundred million 
volts behind it, or five hundred million horsepower. 

This was the most spectacular of many deeds which Gen 
eral Electric men performed in these later years of the great 
electrical enterprise. General Electric was more than a manu 
facturing concern. It was an institution of science and of 
engineering, of vocational training and of character build 
ing, an institution of wide social and economic influence. 
Through its scientists, engineers, workmen, and salesmen it 
had become what it was. 

A few months before Steinmetz manufactured lightning, 
the General Electric laboratory at Pittsfield transmitted elec 
tric current at a million volts. This was achieved by Guiseppe 
Faccioli and Frank W. Peek, Jr., upon the spot where the 
famous SKC trio had pitted their confidence against the elec- 


trical conservatism of the early nineties. The plant manager 
at Pittsfield at that very moment was Cummings C. Chesney, 
last of the SKC men. 

As for the manufacture of the incandescent lamp, hardly a 
hand operation now remained. One by one machines had 
taken over these intricate, patience-wearing operations. 

In 1918 William R. Burrows, Howell's resourceful asso 
ciate, was still looking for more economical methods of manu 
facturing the lamp. He reflected that there were six or eight 
major processes, each located in a different department. 
Some machines produced more rapidly than others, so that 
surplus parts had to be stored between operations. Burrows 
and his staff soon reduced the number of basic machines to 
four, which were placed next to each other. The work was 
fed in sequence from one to the next. A balance of produc 
tion was .thereafter maintained, in which there were no 
surplus parts. Production costs dropped down, as did the 
price to the consumer. Within thirty months General Electric 
reduced its lamp prices five times, the total reduction 
amounting to thirty-seven per cent. 

A further economy was the tipless lamp. For forty years 
no one had been able to eliminate that little glass tip, which 
was a liability for it increased the breakage hazard. The tip 
was caused by the method of exhausting the air from the 
bulb, through a glass tube welded to the top of the bulb, 
which in being sealed off left a protuberance. In 1929 two 
foremen of the National Lamp Works of General Electric at 
Cleveland, Loris E. Mitchell and Arthur J. White, contrived 
to exhaust the lamp through the base of the bulb. This new 
method abolished one step of manufacture, and production 
cost and the market price dropped again. 


THE PIONEERS OF THE COMPANY, the men who had brought 
the organization through its earlier years, grew older. In 
evitably, younger shoulders assumed the heavier burdens 
of management. 

In January, 1919, General Electric segregated its export 
business, which up to that time had been handled by a for 
eign department. The foreign business had increased three 
fold during the preceding five years. It needed separate at 
tention attention of a kind different from that accorded 
domestic affairs. So it was decided to set up a new company, 
to be called the International General Electric Company. 

To head that new company, a man was chosen from out 
side the organization, a vice president of the Western Electric 
Company, who had built on a background of engineering 
training a reputation for merchandising ability and a keen 
grasp of the fundamentals of business. That man's name was 
Gerard Swope. Strictly speaking, he was returning to the 
fold. But probably few were aware that he was the same 
"G. Swope, helper, per day $1" who had been on the Com 
pany's payroll at the Chicago World's Fair of 1893, when 
he was a student at Massachusetts Tech and General Electric 
was only a "million a month" concern. 

For almost two years Swope traveled constantly in the 
foreign countries in which General Electric was doing busi 
ness. He reorganized the scattered personnel of the various 
foreign offices, he worked out standard and consistent con 
tracts with foreign business firms, he developed a smooth- 
running organization to handle the large amount of business 
which had all but swamped the previous staff. 

The duties of Owen D. Young, as General Counsel, had 
been at first largely in giving legal advice and handling 
strictly legal affairs. But gradually he was called on to be 


arbitrator in other matters. All sorts of questions were 
brought to him, and he showed an uncanny faculty for com 
ing to a quick decision, but a decision that the passage of 
time showed to be as wise as though it had been the result 
of long and laborious discussion. When matters reached an 
impasse, it became a habit to take them to Mr. Young. 

In the organizing of the Radio Corporation, too, Young 
had taken a leading part. As Chairman of the Board of that 
company, he had already, by 1922, shown his ability in guid 
ing the new venture through the hazards and pitfalls of its 
earliest years. 

In the spring of 1922, the industrial world learned of im 
portant executive changes in the General Electric organiza 
tion. The directors had elected Gerard Swope as President 
of General Electric, and Owen D. Young as Chairman of the 
Board. Edwin W. Rice, Jr., whom Swope succeeded as Presi 
dent, became Honorary Chairman of the Board. 

And so Coffin, the great leader, retired. He kept his desk 
in the suite of offices at 120 Broadway, New York, where the 
executive heart of the Company had long been centered. 
He came regularly to his office. Visitors there met with the 
same kindly reception, the same quiet humor that had 
smoothed the deliberations of so many executive and di 
rectors meetings. But his day-by-day interests were now more 
personal, confined largely to innumerable philanthropic ac 
tivities that had been one side of his life for a long span of 

In honor of the first President, the Charles A. Coffin 
Foundation awards were added to a long and growing list 
of activities in which General Electric had long been pio 
neering again leading the way that others would eventually 
follow. This was the field of employee relations, which Gen 
eral Electric was exploring by establishing plans contributing 


to employee security. A pension plan had been set up in 
1912; a mutual benefit association to provide benefits in case 
of sickness, death, or hospitalization had been established in 
1913; a profit-sharing system in 1916; a life-insurance plan in 
1920; a savings plan in 1922. A suggestion system, whereby 
employees were paid for suggestions helpful to the operation 
of the Company, had been established as far back as 1906, 
and through the intervening years one plan after another had 
been added to protect employees against the vicissitudes of 
life, and to provide opportunity and a measure of security. 

IN THE SPRING OF 1926 a young General Electric man, Wil 
liam W. Trench, called upon Charles A. Coffin at the latter's 
office in New York. It was one of the last interviews in Mr. 
Coffin's career of which there is a record. Mr. Trench de 
scribes it as follows: 

"That May morning found him full of the exuberance of 
youth. Just in from his home on Long Island, he was full of 
the beauty of the country had time to discuss two child 
poets whose works aroused his enthusiasm, to speak glow 
ingly of the notable men who had been attracted to Gen 
eral Electric by the fascination of the industry, possessing, 
as it does, no known bounds of service or accomplishment. 

"The conversation touched on the extraordinary contribu 
tions of young men to the advancement of the industry. Mr. 
Coffin remarked on the age at which Edison did his revolu 
tionary work at which Langmuir and Coolidge brought 
forth their first great discoveries. His thoughts turned to the 
possibilities of making greater use of youth in other fields of 
activity. He deplored the program of several young men he 
knew who were still pursuing their studies at twenty-eight, 
with the realm of actual life yet unexplored he looked with 
some hope at ways and means of aiding youths to acquire 


wisdom as well as knowledge during the early vital years of 
their lives. 

"As in the case of any discussion with Mr. Coffin, one left 
the room stimulated aroused. One felt a greater interest in 
the manifestations of nature in the beauty of cultural things 
in the possibilities of building and training youth for larger 
activities in the part America should play in world affairs." 

Mr. Coffin was then in his eighty-second year. Two months 
later he passed away. 


The Second Generation 

WITH THE DEATH OF Charles A. Coffin the first great era in 
General Electric history came to a close. During his admin 
istration he had launched a new industrial art and estab 
lished it firmly as an essential part of mankind's way of liv 
ing. Now, under the leadership of Owen D. Young and 
Gerard Swope, a new era in the Company 's history began. 
Before them lay the task of bringing a frontier organization 
to industrial maturity, of refining the rough-hewn policies 
that had served the pioneer period. There was plenty of pi 
oneering yet to be done, but it was pioneering of another 
sort. The struggle to overcome the skepticism that had con 
fronted the early efforts of the pioneers was largely won. 
Now came the task of moulding the giant force of electricity 
to human needs; of applying it to bring new comforts and 
conveniences, higher living standards for the nation; of fit 
ting a new industry to take its place beside the older indus 
tries to help build a greater America. 

It was to this task that Owen D. Young and Gerard Swope 
now applied themselves. They early worked out a division 
of responsibilities that was to continue as long as they held 



office. As Mr. Young described it, "One of us shall act as cap 
tain of the ship, the other as navigator." Mr. Young would 
concern himself with policy, Mr. Swope with production, re 
search, engineering, and sales. 

In the 18 years of their leadership, the teamwork of these 
two men produced many marvels, but none of more lasting 
significance than their common effort on behalf of labor. For 
years the Company had been in the forefront of American 
industry in providing fair compensation and good working 
conditions for employees, but these were to be supplemented 
by many employee benefit plans. Many of these plans had 
been inaugurated before Mr. Coffin retired, but under Mr. 
Young and Mr. Swope they were expanded and welded into 
a unified program for the employees' welfare. Since 1916 
there had been in effect supplementary compensation and 
profit-sharing plans for the employees, and Mr. Swope and 
Mr. Young applied themselves to the perfecting of these 
plans. Together they further developed the pension plan 
which had operated since 1912, and the mutual-benefit asso 
ciations which provided sickness, hospitalization, and death 
benefits. They established life insurance plans; a savings 
plan; the suggestion system, whereby employees reap the 
benefits from their suggestions for improving methods and 
working conditions; vacations with pay; a plan to adjust em 
ployee earnings automatically as variations occur in the cost 
of living; loan funds; a plan to assist employees in acquiring 
homes; and year by year additional methods were worked 
out to protect employees against the vicissitudes of life, and 
provide opportunity and a measure of security. Employee 
representation was instituted, unemployment insurance was 
put into operation, and educational and recreational facilities 
for employees developed. 

These efforts on behalf of labor, described as 'liberal" and 

Chairman of the Board of General Electric. 

President of General Electric. 


even as "radical" by many of their contemporaries, sprang 
from a new conception of their responsibilities as leaders of 
the General Electric Company. Up to the time the electrical 
industry was founded most businesses had been managed by 
their owners directly. As the years passed and large corpora 
tions came into being, there came about a separation of man 
agement and ownership now represented by stockholders 
but management still considered itself the agent of the 
owners, working in their interests. It was this conception of 
management that prevailed throughout the business world 
at the time Mr. Young and Mr. Swope succeeded to the 
Board Chairmanship and Presidency of General Electric. 
They were among the first to recognize a new conception of 
management's responsibilities a conception of manage 
ment, not as an agent of the owners, but as a trustee of all 
groups vitally interested in industry owners, employees, 
and the general public, including customers. It was their de 
termination to guard the interests of all three groups. 

This conception of their responsibilities had a profound 
influence on the General Electric Company. No serious labor 
disturbance occurred during their entire administration, even 
though that period saw the nationwide struggle which ac 
companied the union movement in America. 

Another field in which this, the second generation of Gen 
eral Electric administrators, was outstandingly successful 
was the field of public relations. Mr. Coffin had been of a 
retiring nature. He himself said, "A company's job is simply 
to make goods and sell them. The less said about personali 
ties, the better." As a result of this reticence, many consid 
ered General Electric an "interest," controlled by "Wall 
Street." Under the direction of Mr. Young and Mr. Swope, 
the public began to get better acquainted with General Elec 
tric. The introduction of home appliances, together with the 


appliance advertising, helped; the public relations program 
that they instituted was instrumental in bringing it about; 
and they themselves, rapidly growing to the stature of inter 
nationally known figures, became widely recognized as the 
leaders of General Electric. The emphasis which Mr. Young 
placed on this phase of the Company's activities is illustrated 
by his admonition to a group of employees: 

"Everyone," he said, 'lias had mornings when he hates to 
hear the telephone ring, or to see the office door open. I beg 
of you, gentlemen, when next you meet such a morning, take 
a stick of dynamite and blow up one of our plants. But do 
not take it out on a customer of the General Electric. We can 
replace the plant you have destroyed; we know its value; we 
have a reserve from which we can rebuild. But we cannot 
measure the goodwill you have destroyed, and we can never 
know if we have replaced it." 

Consideration for the public and public opinion directed 
several important policy changes during the early years of 
their administration. General Electric's ownership of the 
Electric Bond and Share Company, giving it an influential 
interest in a goodly number of public utilities, was a poten 
tial source of public criticism, and Mr. Young and Mr. Swope 
soon advocated taking the Company out of this business. 
"We can't carry water on both shoulders," said Mr. Swope; 
"it is wrong to our own customers." Even against the advice 
of Mr. Coffin they urged this step to their Board of Directors, 
and in December 1924 they disposed of General Electric's 
holdings in the Electric Bond and Share Company by dis 
tributing its stock to General Electric's stockholders. 

In another direction, too, this new concept of their man 
agement responsibilities began to be felt. On behalf of the 
general public additional emphasis was placed on making 
electricity more useful, and less costly so that everyone could 


afford its advantages. When they took control, lamps and 
fans were about the only General Electric products sold to 
the general public in fact, few electric appliances of any 
sort were on the market. General Electric products consisted 
principally of huge turbines and generators, motors, and 
other apparatus, sold largely to industry, railroads, public 

Between 1925 and 1930 General Electric devised and put 
into production many new electrical appliances, designed to 
help banish drudgery from the home. Electric refrigerators, 
radios, washing machines, vacuum cleaners, and many 
smaller appliances were placed on the market. Air-condition 
ing equipment was developed; electric ranges, water-heaters, 
electric ironers, and garbage-disposal units took their place 
beside the others. At first these devices were crude and 
costly, but as sales volume increased and improved methods 
of production were worked out, the selling prices were 
steadily reduced. At the same time, vast improvements were 
made in the products themselves they lasted longer and 
consumed less electricity, so that each year millions more 
families were able to afford them. It was not many years be 
fore the manufacture of electrical products for the home be 
gan to be almost as large a part of General Electric's business 
as the large apparatus which had previously accounted for 
most of the Company's sales. In creating these new comforts 
and conveniences for the home, contributing directly to the 
benefit of the public, General Electric also assisted in creat 
ing a whole new branch of the electrical industry, providing 
new employment for thousands of people in the factory, 
thousands of salesmen all over the country, and a host of 
workers in other industries providing the raw materials from 
which these appliances are made. 

Nor was General Electric idle in the field of industrial 


equipment. Under the leadership of Mr. Young and Mr. 
Swope, General Electric's scientists, engineers, and workmen 
were busy fitting electricity and electric apparatus to the 
needs of the world. They were revolutionizing methods of 
production throughout industry by creating electrical meth 
ods for doing existing jobs, and for making possible new op 
erations which could not be done before. In every factory 
and shop, electricity was being put to work to enable many 
industries to produce goods by the thousands at lower cost, 
so that millions of people could enjoy them. 

By constant improvement of the apparatus for generating, 
transmitting, and distributing electricity, General Electric 
was making it possible for the power and light companies to 
expand the areas they served, and helping them to reduce 
the cost of electricity year by year so that more people could 
enjoy its benefits. The steam turbine has been so improved 
that today twice as much electricity is produced for each 
pound of coal consumed as was produced in 1922. Hydrogen 
cooling was developed for large generators, further increas 
ing their efficiency. Transformers have been vastly improved, 
new steels developed for the cores which have greatly re 
duced power loss in the transformers, and a new method de 
veloped in which the cores of distribution transformers are 
wound, producing better transformers at less cost. Research 
in lightning has been carried forward and expanded, and 
many new protective devices developed to minimize light 
ning damage. 

Throughout the whole range of electric apparatus em 
ployed in making and distributing electricity, constant im 
provement has resulted in making electricity more readily 
available to everyone. In this field, General Electric is still 
seeking new and better ways. Research and development on 
the mercury turbine, direct-current transmission, and hun- 


dreds of other experiments give promise of the progress yet 
to come. 

To help industry provide in abundance the thousands of 
articles in common use, General Electric has developed hun 
dreds of new electrical devices. Arc welding has enabled 
industry to fabricate giant machines for the factory, as well 
as smaller products for the home. Bridges, buildings, ships, 
and hundreds of industrial products are made more easily, 
more quickly, and at less cost than would be possible with 
out arc welding. 

Electric heat, in the form of huge electric furnaces and 
small elements, is employed in melting, annealing, harden 
ing, and enameling metal products. Many industrial proc 
esses would be practically impossible without electric heat. 

New motors more compact, more flexible, more adapt 
able give vast amounts of mechanical power to help in 
dustry do its many tasks. Small motors fractional horse 
power have been applied to bookkeeping machines, elec 
tric appliances, air-conditioning equipment, and hundreds 
of other useful applications. Electric control has been ap 
plied to hundreds of machines, making them capable of do 
ing things that would be impossible by hand. In many fields, 
electric control governs the processes of converting raw ma 
terials to useful finished products. 

In mining operations and in the oil industry, electricity has 
made possible improved methods, and today electricity oper 
ates the drills, pumps oil through electrically welded pipe 
lines to electrically operated oil refineries, and operates the 
gasoline pumps at the service stations. In every industrial 
endeavor electricity has been applied to enable men to pro 
duce more with less effort. 

In the field of transportation, electricity has made vast im 
provement possible. General Electric played an important 


role in electrifying the Pennsylvania Railroad from New 
York to Washington and Harrisburg the world's largest 
main-line electrification project. High-speed electric locomo 
tives have won new passenger revenue for the railroads. 
Auxiliary equipment for air-conditioning trains has made 
travel more comfortable. General Electric has developed a 
new, more powerful, steam-electric locomotive, which carries 
its own steam turbine to produce the electric power it uses 
to drive its wheels. 

For urban transportation, General Electric has developed 
trolley coaches, streamlined street cars, and oil-electric buses. 
Faster, smoother, more comfortable transportation has been 
furnished hundreds of American cities. On the high seas, 
electricity has made ships faster, safer, more comfortable. 
Turbine-electric and geared drives have revolutionized 
methods of ship propulsion. And in dozens of other ways, 
electricity has contributed to passenger, freight, and naval 
vessels alike. 

In the air, General Electric superchargers have become al 
most standard equipment on commercial and military air 
planes. Airplane instruments, airport lighting, radio beacons, 
all owe their efficacy to electricity. 

In the constant war against disease, electricity has been 
placed at the aid of the physician and surgeon in countless 
ways. The X-ray, developed by Dr. W. D. Coolidge of the 
General Electric Research Laboratory, has proved invalu 
able in diagnosis, and in the treatment of disease. The 
electro-cardiagraph, for analyzing heart conditions; the arti 
ficial fever machine for treating several painful, crippling 
diseases; electro-surgical apparatus; diathermy; ultraviolet 
radiation; infrared radiation; ionic radiation; surgical ioniza- 
tion; and other remedial measures have been provided by 


Through research and engineering developments, the in 
candescent lamp has been vastly improved and reduced in 
cost. The "Science of Seeing" has been developed, aiding 
and safeguarding the sight of millions of people. The lamp 
has invaded the field of photography, and street and highway 
lighting have made our streets safer after dark. New light 
sources of ever greater efficiency have been developed 
sodium lights, fluorescent lights all are opening up new 
fields of usefulness. Sealed-beam headlamps for automobiles 
are making night driving safer and more comfortable. 

Another achievement of vast importance in the home has 
been the radio. Dr. Irving Langmuir's development of the 
high-power vacuum tube made radiobroadcasting possible, 
and General Electric's own radiobroadcasting stations have 
pioneered in making radio entertainment available to almost 
everyone. Today, General Electric has regular broadcasting 
stations at Schenectady, Denver, and Oakland; has two 
shortwave stations at Schenectady and one at San Francisco 
broadcasting to South America, Europe, and the Orient in 
several languages; has television stations at Schenectady and 
at Bridgeport; and has a frequency-modulated station at 

In still another field General Electric has been pioneering 
the new field of plastics, already important, and promising 
vast benefits for the future. As an outgrowth of the early non- 
metallic gears, the present list of plastic products ranges 
from heels for ladies shoes to radio cabinets, from pencils to 
parts for machines. In any forecast of the future, plastic 
products must be given a leading place. 

This then, was the picture of General Electric that Mr. 
Young and Mr. Swope could look back upon as 1939 drew to 
a close. Behind them lay a period in which General Electric 


had made electricity of service to mankind, contributing to 
greater comfort of living for millions of people. Behind them 
lay a period of post-war collapse, an unprecedented period 
of business boom, followed by a decade of devestating de 
pression. Yet through it all General Electric had made steady 
progress in applying electricity to the needs of civilization, 
creating "More Goods for More People at Less Cost." Mr. 
Young was just past 65, Mr. Swope was 66. They had long 
advocated the retirement of General Electric employees at 
65; now they faced the task of turning over to others the 
responsibilities they had discharged so well. 

At the Board of Directors meeting on November 17, 1939, 
Mr. Young read the statement that he had written on his 
scratch pad a week before, "We took up these offices to 
gether and we wish to lay them down together." January 1, 
1940. Listening to his words were two men who had entered 
the room a few minutes before as the Assistant to the Presi 
dent and Executive Vice President, respectively, but who 
were to leave a few minutes later as Philip D. Reed, Chair 
man of the Board of Directors, and Charles E. Wilson, Presi 

So ended the second period of General Electric history; 
so started the third. What lies ahead none can know, but if 
the history of the past can be taken as a guide, the future 
holds in store possibilities as vast and as promising as any 
that have gone before. 


Statement of 

to the 

Temporary National Economic Committee 
Regarding the Formation of Capital 

of the 
General Electric Company 

May 17, 1939 

THIS is AN EFFORT to draw an over-all picture of the material 
growth of the General Electric Company and to show the 
origin and development of its capital. The statement does 
not deal with the adventures and the accomplishments of 
the Company in the field of science, nor with its experience 
in welding together great numbers of human beings of widely 
diversified capacities and skills into an organized and effec 
tive service. 

Even in the material field, the attempt to reflect the story 
truly and briefly is a daring one. None will understand that 
better or make more charitable allowance than the group for 
which this memorandum is prepared. 

The corporate age of the General Electric Company is 47 
years, but as it sprang from the consolidation of the Edison 
and Thomson-Houston Companies, it is perhaps more cor 
rect to say that its present age is sixty years. 

At the close of its sixtieth year on December 31, 1938, we 
find its material resources are carried on its books at the net 
value of $322,739,000, classified as follows: 



Net working capital (current assets less current 
liabilities ) 

Affiliated companies and other investments, (in 
cluding International General Electric Co., 
and manufacturing, selling, real estate and 
other companies, $105,585,000) 

Plant and equipment 

Other assets 

Less: other liabilities and miscellaneous reserves 
Total capital investment 

This capital was accumulated as follows: 
From retained earnings, now carried as surplus and 

general reserve 
From issue of capital stock 

This capital stock was issued: 

For properties 

For cash 

In exchange for bonds of the Company 

As dividends 






$ 23,982,000 



$ 37,784,000 


Representing this stock there is now issued and outstand 
ing 28,845,927 shares, held by 208,580 stockholders. There 
are no bonds and no preferred stock outstanding. In a word, 
the two hundred thousand-odd stockholders own the Com 
pany, not an equity in its assets. 

This picture, which is one of affluence in the foreground, 
might well be misleading unless we can see in the back 
ground the rough and dangerous road on which the pioneers 
set out, with confidence and courage, more than a half cen 
tury ago. In that great caravan were scientists and inventors 
to originate, engineers to apply, manufacturers to make, com- 


mercial men to sell, people with savings to finance, entre 
preneurs and administrators to develop and manage, and a 
vast army of conscientious and skillful men and women do 
ing their respective jobs with pride in a new art and en 
thusiasm in the creation of a new industry. 

In looking for the beginning of the road it is necessary for 
us to recall what has been earlier stated, that the progenitors 
of the General Electric Company were the Edison General 
Electric Company and the Thomson-Houston Electric Com 
pany. Let us speak of the latter first, not because it is the 
more important of the two, but rather because, in typical 
American fashion, it arose out of the unknown. 

If we exclude the telegraph, it is fair to say that in 1878 
the electrical manufacturing industry was not only non 
existent, but the art as a practical one was unknown. The 
men who created, financed and managed the Thomson- 
Houston Company were likewise relatively unknown. At that 
time Edison was well enough known through his inventions 
so that his initial adventure into the electrical industry was 
financed by a group of prominent New York capitalists. Not 
so with the Thomson-Houston group. There were no prior 
inventions, there were no names of great capitalists, and there 
was very little money. 

Elihu Thomson and Edwin J. Houston were teachers in 
the Philadelphia Central High School. They built a dynamo, 
induction coils and an arc lamp. Thomson showed the con 
traption to his friend, Thomas H. McCollin, a commercial 
photographer in Philadelphia. McCollin asked his cousin, 
George A. Garrett, to see a demonstration. Thomson, then 
26 years old, said, "I can build a better machine than this, 
one that will run any number of lights you want." Garrett, 
with enthusiasm replied, "Let's build a four-lighter. Ill 
stand the expense." That was America speaking in 1879! 


The first installation was in an all-night bakery, the second 
in a brewery. When the brewery caught on fire one of the 
firemen who was holding the hose said, "What the dickens 
kind of a light is that? You pour water on her and she won't 
go out!" That was news in America sixty years ago. 

At the same time, Mr. Edison, having incorporated the 
Edison Company with the backing of J. P. Morgan and a 
capital of $300,000, was receiving considerable well-deserved 
publicity in the New York press. 

Thereupon, Frederick H. Churchill of New Britain, Con 
necticut, seeking a new industry for his town, invited Thom 
son and Houston to settle there. When they accepted, Church 
ill circulated a prospectus in New Britain which, after de 
scribing the new company as "a rare opportunity for the 
upbuilding of a successful business enterprise," ended with 
the following sentence: "Such an industry would have within 
it a capacity for expansion and growth such as is furnished 
by but few of the business opportunities that today present 
themselves to the business world." 

Churchill died in 1881 and the New Britain venture failed; 
not, however, before one of its installations had been made in 
an armory in Lynn, Massachusetts. A Lynn newspaper pro 
prietor, Silas A. Barton, became interested in the possibility 
of a second installation in that city and visited New Britain. 
Learning that the business was for sale, he suggested the 
idea of purchase to a group of Lynn shoe manufacturers, 
among whom was Charles A. Coffin, then under forty years 
of age. The upshot was that Coffin and his associates pur 
chased a controlling interest in the company, moved it to 
Lynn and changed its name to the Thomson-Houston Electric 

The members of the Lynn syndicate were successful shoe 
manufacturers. Initially they were primarily interested in 


bringing a new industry to their town. That business soon 
captured the imagination of Charles A. Coffin, and from that 
time until his retirement in 1922 he was its leader through 
bad times and good. 

It is necessary for us to remember that in 1882 the problem 
which faced both the Thomson-Houston Company and the 
Edison Company was not only the origination, development 
and manufacture of electrical machines. That was only half 
the task. Broadly speaking there were no buyers. New cus 
tomers had to be created to buy the new machines. This led 
to the creation of local light and power companies. The 
sponsors were usually made up of local capitalists and poli 
ticians, for it was necessary to have a franchise and some 

For the most part they were motivated by civic pride and 
the promotion of local development. The capital of the com 
panies was proverbially inadequate. This forced the manu 
facturing companies to accept in part payment for the new 
equipment the bonds and stocks of the new customer com 
panies. The Edison Company, through a somewhat different 
technique that of issuing exclusive licenses against secu 
rities of local operating companies ultimately faced the 
same necessity of accepting additional securities in part pay 
ment for its products. 

Leaving now the Thomson-Houston Company at Lynn, 
with Charles A. Coffin and his associates in charge, let us 
look at the development of the Edison Company. 

In 1879, while Thomson, the high school teacher, was 
tinkering with his arc light system, Edison, still under 35, 
had already established a modest position as an inventor. 
One of his earlier inventions, relating to the electric tele 
graph, had been the subject of patent litigation. Grosvenor 
P. Lowrey, a prominent New York lawyer, had defended 


Edison's interests and as a result became one of his most 
faithful and staunch supporters. Lowrey, having confidence 
in Edison's ability and knowing of his interest in developing 
an incandescent lamp, approached Mr. J. P. Morgan and 
other influential financiers of his acquaintance. The result 
was the organization of the Edison Electric Light Company. 
The capital stock amounted to $300,000, of which $250,000 
was issued in payment for equipment and the remaining $50,- 
000 was subscribed for, with 80 per cent paid in in cash. 
That, too, was America speaking in 1878. 

Edison's immediate problem was the discovery of some 
material for a filament that would stand incandescence for 
a considerable period without disintegration. 

On October 21, 1879, his experiments were crowned with 
success at Menlo Park, when his carbonized cotton filament 
burned continuously for forty hours. Truly a historic demon 

To successfully operate an incandescent lighting system, 
however, it was necessary for Edison, in addition to the lamp, 
to develop a completely new set of mechanisms to generate 
and transmit the electric current. This was the kind of stag 
gering task which only a great genius could master. It soon 
made Edison's facilities at Menlo Park inadequate. In 1880, 
therefore, he organized the Edison Machine Works in New 
York City to manufacture generators, and the Edison Lamp 
Works near Newark, N. J., to manufacture lamps. 

In 1886, Edison decided to move the Edison Machine 
Works out of the metropolitan area. He became interested in 
an unused plant at Schenectady, and made an offer which 
was refused by the owners. 

An impasse threatened. But civic pride burned bright in 
the breasts of some, at least, of Schenectady 's citizens. They 
grasped what it would mean to the town to have a plant of 


the famous Edison. A meeting was called, and eventually by 
local subscription a fund was raised which made up the dif 
ference between "bid and asked" prices. 

So it was that the Mohawk Valley became the center for 
manufacturing the generating equipment required by the 
growing number of Edison licensee operating companies. 

These were stirring times. The formative period of electric 
development was approaching its height. The Edison system, 
as an organization for manufacturing, licensing and selling 
electric light systems, was beginning to sprawl. Greater in 
tegration seemed desirable, and in 1889 a merger of the in 
dividual Edison enterprises was consummated through the 
formation of the Edison General Electric Company. 

At the same time that the Edison Machine Works was es 
tablished at Schenectady and the Thomson-Houston Com 
pany was getting under way at Lynn, another idea was 
germinating in the minds of several competent and creative 
engineers, notable among whom were Charles J. Van Depoele 
and Frank J. Sprague. In 1887 and 1888 these two, working 
independently, had demonstrated the practicability of using 
electricity to propel streetcars. 

This opened almost at once a vast new field for the electri 
cal manufacturers. Larger power units had to be made, longer 
transmission was necessary, and cars needed to be equipped 
with motors. The old horse railways had to be transformed, 
franchises modified or enlarged and additional capital pro 
vided. As in the lighting field, the problem of financing was 
difficult but imperative. It invited the manufacturing com 
panies to extend their limited credit further in order to en 
large their business. 

The rapidity of the development of the Edison and Thom 
son-Houston Companies is shown in the following statement 
of their condition in 1891: 


Edison General 

Electric Thomson-Houston 

Capitalization $15,000,000 $10,400,000 

Gross business 10,940,000 10,304,500 

Profits 2,098,000 2,700,000 

Number of employees 6,000 4,000 

Factory space, sq. feet 400,000 340,000 

Customers 3-4,000 3-4,000 

Central stations 375 870 

Isolated installations 2,300 very few 

Street railways equipped 180 204 

Street railway cars 2,230 2,760 

The Thomson-Houston Company, springing from the un 
known, is catching up with the great Edison Company and in 
points surpassing it. With two-thirds the capital and two- 
thirds the number of employees, Thomson-Houston is equal 
ling Edison in gross business and exceeding it in profits. It 
has a larger number of central lighting stations, more street 
railways and more streetcars equipped. The Edison Com 
pany, in its annual report for that year, said that it "could 
have done a much larger business if it had been willing to 
accept securities in payment for orders; but ... a strict rule 
was adopted of declining all such and doing business ex 
clusively on a cash or short credit basis." 

The Thomson-Houston Company had been rediscounting 
its customers' notes with its banks and pledging operating 
company securities as collateral for its loans. Finally the 
banks became hesitant, particularly in accepting securities 
as distinguished from customers' notes, and in 1890 Charles 
A. Coffin organized the United Electric Securities Company. 
The purpose was to transfer to it large blocks of securities of 
utility operating companies and to sell to the public its own 
debentures, and so obtain longer term money to bolster the 


waning treasury of Thomson-Houston. The Thomson-Hous 
ton Company kept the common stock of the United Electric 
Securities Company. 

The public, however, was hesitant too. It soon became ap 
parent that debentures could not be sold in sufficient quan 
tities to finance the rapidly increasing amount of utility 
securities which the manufacturing company was obliged to 
take. The Edison Company was drawing in its horns as a mat 
ter of business policy. The Thomson-Houston Company was 
doing the same thing from necessity. 

Perhaps at this point we should remind ourselves that there 
were several other companies operating at this time in the 
electrical manufacturing field; notable among them was the 
Westinghouse Company. They were all more or less in the 
same condition with the same problems and the same dif 

Under such circumstances, what could be more natural 
than to combine the Edison and Thomson-Houston Com 
panies, giving the consolidated concern the business leader 
ship of Charles A. Coffin, the inventive genius of Edison and 
Thomson, eliminating patent conflicts which were threaten 
ing both concerns, and bringing together the financial in 
terests in Boston which had supported Thomson-Houston 
with the great prestige of Mr. Morgan and his associates in 
New York? 

So it was that the General Electric Company was incor 
porated on April 15, 1892, and began active business on the 
first day of June in that year. Charles A. Coffin was made 
President, and the contest began for the market of electrical 
manufactures between the Westinghouse Company led by 
George Westinghouse, the brilliant and imaginative engi 
neer, and Charles A. Coffin, the daring and equally imagina 
tive man of business. 


The first board of directors of the General Electric Com 
pany reflected its distinguished and substantial backing; F. 
Lothrop Ames, Charles A. Coffin, T. Jefferson Coolidge, Jr., 
and Henry L. Higginson, all of Boston, Thomas A. Edison, 
Charles H. Coster, Frank S. Hastings, General Eugene Griffin, 
Darius O. Mills, H. McK. Twombley, and J. Pierpont Morgan 
of New York. 

The directors made their first report to stockholders on 
April 11, 1893, covering the eight months from June 1, 1892, 
to January 31, 1893. At that time it stated its capital invest 
ment to be $45,688,755, against which it had issued $10,000,- 
000 of bonds, $4,236,900 of preferred stock and $30,426,900 
of common stock with a surplus of $1,024,955. It had 3,272 
stockholders. There were then 1,277 central station lighting 
companies using Edison and Thomson-Houston apparatus, 
supplying 2,500,000 incandescent and 110,000 arc lamps. Of 
the lighting companies, the report says: 

"The growth of these companies has been phenomenal, and it 
is very satisfactory to note that those which have been established 
longest are making the most rapid increase in size of plant and 
volume of business. 

"During the past year there has been a very marked apprecia 
tion in the value of the securities of local companies, especially 
in the larger cities, testifying to the increased confidence of in 
vestors in such properties." 

It is interesting to note, at this time, the growth in two 
years of the electrification of street railways. On February 1, 
1891, there were 151 roads operating electrically or under 
contract; on the same date in 1893 there were 435. On 
February 1, 1891, there were 1,578 electric cars; on the same 
date in 1893 there were 8,386. On February 1, 1891, there 
were 1,252 miles of electric railway in operation; at the same 
date in 1893 there were 4,927 miles. 


In its first eight months of its existence, the Company did 
$11,728,000 of business, at a profit of $2,996,011 of which 
$1,971,056 was distributed in dividends and $1,024,955 was 
carried to surplus. 

The prestige of the Company enabled it to sell $454,300 
par value of the preferred stock of the United Electric Secu 
rities Company for a price in excess of $408,000. It was still 
necessary for the Company, however, to endorse and dis 
count customers' notes, because such notes unendorsed were 
not acceptable to the banks. On January 31, 1893, these 
amounted to $3,787,312.69. In addition it had outstanding 
$10,000,000 of five per cent bonds. 

In such a condition the new General Electric Company 
faced the panic of 1893. It will be interesting to see what hap 
pened during the next few years. 

The first shock of the depression was dramatically told in 
the second annual report of the Company issued under date 
of January 31, 1894: 

"It is needless to say that the past year has been a most trying 
one to all corporations. It has been especially so to companies 
like your own, dealing with local enterprises situated in all parts 
of the United States, and largely dependent on normal conditions 
for their success and development. During the summer of 1893, 
even old and strong customers were obliged to ask for leniency 
in paying their accounts and notes. Under these circumstances, 
your Company found itself with its own obligations to meet, but 
unable at that time to collect the money with which to meet them. 
The difficulties thus presented were carefully considered by your 
Board and were met by selling to a syndicate certain of the Com 
pany's assets consisting of claims against, and stocks and bonds 
of, local lighting and railway companies, the same being of a 
class of which your Company sold several million dollars in 1892 
and which your Directors, in their last report, said they intended 
to continue to sell from time to time as heretofore 'through the 


ordinary channels.' The channels through which your Company 
usually made such sales having become unavailable owing to the 
panic, your Directors adopted a plan used on several occasions in 
the earlier days of the Thomson-Houston Company, and made 
the sale of assets above described to a syndicate which paid over 
$4,000,000 in cash. Although the transaction involved a large 
shrinkage from book valuations, the sale was at a price high 
under the conditions then prevailing. Few of the securities sold 
were listed on any exchange or commonly dealt in, and it was 
not possible to effect a ready sale except in bulk to a syndicate. 
These assets were placed in a trust known as 'The Street Railway 
and Illuminating Properties/ After the financial stringency had 
subsided, the right to subscribe to them was offered to the Stock 
holders of your Company. 

'The depreciation in value of the assets thus sold applies 
equally to those still on hand. Holders of stocks and bonds of 
almost every kind find them quoted today much lower than a 
year ago, and this Company, as a holder of electrical stocks and 
bonds, is no exception to the rule. In fact, the shrinkage in values 
of electrical securities has been greater than in most others. The 
last year has been characterized by shrinkage in every direction, 
and your Company has suffered severely from it." 

The officers of the Company were making a titanic effort 
in their race with receivership. By January 31, 1894, the debt 
had been reduced by $6,750,000. The cash had dwindled 
from $3,871,000 on January 31, 1893, to $591,000 on January 
31, 1894. The directors stated: 

"While the liquidation of the debt has been going on, the 
Company has also readjusted its basis for sales, either to cash 
or to short credits to desirable customers. In view of the extreme 
depression and the uncertainty as to the early future, your Direc 
tors have not felt justified in any other course than that of ad 
hering strictly to sales on this basis. It is believed that your Com 
pany has lost little legitimate business in consequence of its cur- 


tailment of credit to customers. It intends to confine its business 
to this basis, and to accept smaller profits." 

The capital investment of the Company dropped in that 
year from $45,000,000 to $32,000,000 and its surplus from 
$1,000,000 in the black to $12,000,000 in the red. Such was 
the first impact of the panic on the capital structure of the 
new concern. 

It is interesting to note that the red figures in the surplus 
continued at substantially the same figure until 1898, when it 
was wiped out by a reduction of capital stock of $14,000,000. 
In January 1899, the surplus was in the black to the extent 
of $156,000, and it has remained in black continuously ever 
since. The directors further said in their report of January 31, 

"Your Directors do not believe that it will be possible for some 
time to come to do as large a business as was done by the Com 
pany prior to the panic, although a gradual improvement has 
been apparent during the last two months. The street railway 
business, which to a considerable extent was formerly done 
through syndicates and promoters, many of whom have become 
embarrassed, promises to be smaller than during the previous 
year. Arc lighting business is also reduced, largely because of the 
inability of local companies to secure capital with which to ex 
tend their business for the purpose of carrying out municipal 
contracts. The business of the Company, with respect to incan 
descent lighting, which is to a great degree performed by strong 
and conservatively managed local companies, is in a more healthy 
condition, and has not suffered so severely." 

In the third annual report, as of January 31, 1895, to cover 
additional losses the directors "arbitrarily charged $2,000,000 
to Profit and Loss Account." 

During that year, however, they were busy dealing master 
fully with their debt. Of it they say: 


"In the last report, your Directors expressed the opinion that 
they could, out of the then unliquidated assets, pay off the bal 
ance of your floating debt and also provide all necessary working 
capital. These expectations have been realized, and in addition 
thereto the Company has purchased and cancelled $1,250,000 of 
its own Debentures at an average cost of less than 89 per cent." 

In the fourth report as of January 31, 1896, the Directors 
make the following statement: 

"In their last report, your Directors referred at some length to 
the liquidation of old assets, and stated that the sum of $2,000,000 
had been charged to Profit and Loss for the purpose of providing 
for all shrinkages which could then be anticipated in the liquida 
tion of old matters. 

"Much has been accomplished in the year just closed in liqui 
dating old and slow assets, and the condition of the assets of like 
character which still remain on the books of the Company is such 
as to enable your officers to more definitely fix their proper values. 
Information regarding these matters will be found in the report 
of the Second Vice-President, to which particular attention is in 
vited. There have been charged against the $2,000,000 item above 
referred to the sums of $530,152.16, representing the shrinkages 
which have accrued from the liquidation so far as completed, 
leaving $1,469,847.84 still standing to provide for possible shrink 
ages in the future. It is the belief of your Directors that this 
amount is sufficient to cover all the purposes for which the above 
sum of $2,000,000 was originally set apart/' 

The purge continued, but signs of health were reappearing. 

"The business secured by your Company for the fiscal year 
just closed was less than ten per cent greater in value of sales 
than for the year previous. The actual increase in output of fac 
tories, based upon capacity of machines and number of articles 
produced, is more than thirty per cent greater than for the pre- 


vious year. While the selling prices as thus shown have been ma 
terially reduced, there has been a corresponding curtailment in 
manufacturing and other expenses and lowering in costs, largely 
due to improved designs and methods of manufacture/' 

The gross business was still running around $12,000,000. 
Even in 1897 the business had not increased. In that year, 
however, the directors referred again to the $2,000,000 that 
had been set aside in 1894 to provide for shrinkage in assets. 

"The business of your Company has suffered during the past 
year, in common with that of all manufacturing enterprises, from 
the disturbed financial and political conditions which have pre 
vailed during a considerable portion of the time. These conditions 
have curtailed the amount of capital ordinarily available for the 
establishment and extension of Power and Lighting Plants, and 
have enforced the practice of great economy on the part of its 
customers. As a result, the shrinkage in orders received by your 
Company was very marked, especially during the latter half of 
the year. This shrinkage is not shown by a material falling off in 
shipments, as given in the Profit and Loss Statement on page 17 
of this report, but the amount of work in progress and unfilled 
orders on hand is considerably less than a year ago. 

"With a return to normal commercial conditions, a correspond 
ing revival in the business of your Company may be expected. 
The volume of business secured by it for the first three months 
of the current year is slightly in excess of that for the same period 
in either of the three previous years. 

"On January 31, 1895, the sum of $2,000,000 was set aside, as 
shown in the Annual Report of that year, to provide for shrinkage 
in assets, the exact values of which it was then extremely difficult 
to fix. During the past year your Directors have been able to 
value these items with substantial accuracy, and the $2,000,000 
fund has been found sufficient and has been used to provide for 
the proper adjustment of all accounts and other assets for which 
it was created." 


In opening the sixth report as of January 31, 1898, the 
President said: 

"The past year witnessed a revival in business which increased 
rapidly in activity and volume during its latter months." 

He was there referring to orders taken rather than sales 
billed. The latter remained at substantially $12,000,000. The 
clouds were lifting the sum was beginning to shine. For 
five years the depression had continued. The struggle for 
survival was over at last. 

A reduction of forty per cent in the share capital of the 
Company was made, bringing the common stock from $30,- 
460,000 to $18,276,000 and the preferred stock from $4,252,- 
000 to $2,551,200. 

In 1898 the Company was clearing the way for dividends. 
The surplus went into the black, as heretofore stated, to 
$156,000. The business jumped to $15,500,000. In the report 
to stockholders for that year there appears the following 
significant statement: 

"The Company has no Note Payable, nor is there under dis 
count any paper bearing the Company's endorsement or guar 

"It has not borrowed any money, nor has the Company's credit 
been used during the year either by issuing notes, endorsing cus 
tomers' paper for discount or lending its name in any way; but by 
adhering to the policy of the previous four years and maintaining 
sales on a basis of cash or short credit to desirable customers, all 
purchases have been paid for in cash." 

By 1898 the volume of business had grown to more than 
$22,000,000, and the market was beginning to take securities 
which the Company was selling at prices substantially in ex 
cess of their book value after the drastic write-downs of the 


depression years. The Company was off to a new start and 
was to participate profitably in the truly amazing develop 
ments of the next thirty years. 

By the turn of the century, then, the capital investment of 
the Company as shown by the report of January 31, 1901, 
was $32,114,681 (in contrast with the $46,000,000 with which 
it had started eight years before) and was represented by 
$1,534,000 in bonds, $2,551,200 of preferred stock, $21,400,- 
300 of common stock, and $6,629,181 of surplus. The sales 
billed for that year were approximately $28,000,000. 

The two succeeding years were very profitable ones, es 
pecially due to the liquidation of securities above the then 
book values, and the surplus had risen to $15,000,000. All the 
preferred stock and about $1,000,000 of bonds were con 
verted into common. About $2,000,000 of new bonds were 
issued. The Company restored the stock reduction which 
had been made in 1898, which brought its capital stock to ap 
proximately $42,000,000. 

During the next four years it was able to find new capital 
with which to carry on its rapidly increasing business by is 
suing its common shares to stockholders for cash at par to 
the extent of approximately $20,000,000. 

So the Company faced the critical year of 1907 with a 
capital investment of $81,000,000 represented by $2,000,000 
of bonds, $64,000,000 of common stock, and $15,000,000 of 
surplus. Its volume of business had grown to approximately 

As a result of the panic of 1907 the directors said in their 
report of January 31, 1908: 

"Late in the year there was a sudden and severe shrinkage in 
the value of all merchandise and materials used by your Com 
pany, notably copper. All said materials, whether raw, manu 
factured, or in process, which were on hand January 31st, 1908, 


were inventoried at the prevailing lower prices. The book value 
of such inventories was thereby reduced by about $2,000,000." 

The stockholders having subscribed for only $1,500,000 of 
common stock in 1907, the Company issued $13,000,000 of 
5% ten year debenture bonds convertible into stock at par 
on and after June 1, 1911. The financial pressures had forced 
the use of convertible debentures to maintain an adequate 
capital structure, but fortunately by this time the resources 
and credit of the Company enabled it to withstand the shock 
of the short depression of 1907. 

During the year 1911, $12,000,000 of the bonds were con 
verted into common, bringing the common capital to $77,- 
000,000. In 1912 the Company declared a stock dividend of 
30% out of surplus for the purpose, as stated in the report, 
"of recouping the stockholders in part for dividends passed 
or reduced during the years 1893 to 1902." During that same 
year the Company authorized an issue of $60,000,000 of forty 
year debentures to be sold from time to time as required, 
and they sold, in 1912, $10,000,000 of such bonds bearing a 
rate of 5% per annum. 

So, at the end of 1912, the capital investment stood at 
$125,000,000, with bonds of $12,000,000, surplus of $12,- 
000,000 and common stock of $101,000,000. 

During the war years, 1917, 1918 and 1919, there were 
substantial temporary borrowings, increases of capital and 
surplus, the detail of which throws little light on the normal 
and progressive capital development of the Company. 

The year 1920 brought another crisis. The orders for that 
year were $318,000,000, more than eighty millions in excess 
of the previous year. Inventories and receivables had in 
creased, so when the sudden and severe decline in market 
prices came, the Company wrote off of inventory some $18,- 


000,000. During the year it made short term loans of $45,- 
000,000, but between January and April of the succeeding 
year, $42,000,000 of such loans had been repaid. The com 
mon capital, however, was increased in excess of $50,000,000 
during the two years 1920 and 1921 ($40,000,000 being for 
cash and property and the balance as stock dividends ) bring 
ing the total capital investment up to $298,000,000, consisting 
of $176,000,000 of common stock, $38,000,000 of bonds and 
$84,000,000 of surplus and general reserve. 

In 1922 a special stock was created having no preference 
over the common except a limited dividend rate of 6%. For 
the next several years this special stock was issued as stock 
dividends, and no further dividends in common stock were 
issued. The last dividend paid in special stock was in 1926, 
and the total amount issued was $43,000,000. 

In 1923 the Company called and paid off $15,000,000 of 
its bonds. In 1925 it called $15,000,000 more, and it had ac 
quired enough in the market in the meantime so as to re 
duce its funded debt to $2,000,000. In 1935 the Company 
called this remaining $2,000,000 of bonds and retired all of 
its $43,000,000 of special stock. 

The Company had in the year 1935 $180,000,000 of com 
mon stock, $136,000,000 of surplus and general reserve, mak 
ing a total capital investment of $316,000,000. 

From the above brief sketch of capital growth, I have only 
drawn attention to the major changes and the problems which 
faced the management. The resume of the source of capital 
was shown at the beginning of this statement. 

When one looks at the growth of the volume of sales of 
the General Electric Company, he finds it remained sta 
tionary at $12,000,000 for the first five years of its life, but at 
the turn of the century it reached more than $25,000,000. At 
1910 it was $70,000,000; at 1920 it was $275,000,000; and at 


1929 it reached its high point of $415,000,000. In 1933 it 
dropped to $136,000,000; in 1937 it recovered to $350,000,- 
000 and in 1938 it was $260,000,000. The enormous growth 
in the two decades after 1910 needs to be accounted for by 
anyone who wishes to understand the background of the 
capitalization of the electrical industry. 

I wish now to go back to the trying year of 1893, when the 
General Electric Company transferred reams of utility secu 
rities to the Street Railway and Illuminating Properties in 
order to raise $4,000,000 in cash, and thereby saved itself 
from bankruptcy. The Street Railway and Illuminating Pool, 
as it was then called, was strictly a liquidating concern. It had 
no intention of entering the utility business as such. It had 
no particular interest in or knowledge of management of 
such properties. In order to make its securities marketable, 
it was obliged in many instances, if not indeed in most, to 
reorganize the operating units through receivership by re 
ducing the prior lien bonds and preferred stocks. It thereby 
threw larger equities in the new common. The new bonds and 
preferred were sold to the public to the extent and at such 
prices as the market would take them. The new common 
shares remaining in the liquidating pool were, for the most 
part, not marketable through then established channels. 
Something new needed to happen and it did. 

For the ten years preceding the turn of the century, the 
glamour of the new electrical art had attracted many of the 
most brilliant young men of the time into the field of electri 
cal engineering. Most of them took a post-graduate course 
with the manufacturing companies known as "The Test." 
Thereafter, some of them remained in manufacturing, but 
many others went out to the lighting and railway companies. 
A few of them started in business on their own account, first 
as expert advisers to existing companies, and later as ex- 


perts in management. It was to this last group that liquida 
tors like the Illuminating Pool went for the purpose of creat 
ing, through expert management and engineering skill, values 
in common shares. The usual procedure was to turn over to 
such management groups a small block of the common 
shares at a rather nominal figure and install the engineering 
organization as manager. To spur the management, it was 
usually customary to grant options on additional blocks to 
be taken up in specified periods at increasing prices. 

So the young engineers came into positions of manage 
ment and later control of operating public utilities. They re 
placed the nominees of the old politician and the local capi 
talist who, with little knowledge, had endeavored to operate 
in the early days and whose influence was much diminished 
and in many cases entirely eliminated by the purge of the 
panic of 1893. 

So we find at the beginning of the twentieth century groups 
of highly trained young engineers, able and ambitious to de 
velop utilities, with an opportunity open to them not only to 
render great service to the community, but to acquire a com 
petence for themselves. At the same time in the manufactur 
ing organizations their contemporaries were coming into 
positions of influence. For the first time the manufacturers 
had competent and appreciative buyers, and conversely these 
young managers of utilities were stimulating and urging the 
electrical producers to make more efficient and more econom 
ical equipment. 

Here then were producers and users venturing into a new 
field unafraid of new things. The rate of technical advance 
was tremendous. Human brains alive and at work every 
where were enlarging the field of electrical service and ma 
terially reducing the costs. Values began to be reflected in 
these common stocks and then the public itself began to buy 


them, in limited quantities to be sure, but nevertheless a 
market was emerging. Soon the old liquidating concerns had 
finished their task, usually with great profit to the stock 
holders who had dared to invest their money in the dark days 
of the panic. 

But when the utility common shares became valuable the 
new managers did not wish to sell them, although they could 
do so at a profit, because they might lose control of the com 
panies which they had really made. As energetic Americans, 
however, they were ambitious to extend their management 
operations and achieve the added economies that such ex 
tension promised. 

To meet this situation the utility holding company was in 
vented, which enabled these ambitious young managers not 
so much to sell their original common shares at a profit as to 
obtain some funds against them through the issue of holding 
company securities, and thereby enable them to enlarge their 
operations in the communities which they served or to ac 
quire an interest in other properties and so extend their man 
agement program. Then, too, the holding company was a 
better medium for the investor. Instead of taking common 
shares in an individual operating company, he was able to 
invest in a diversified group of shares, and if he took pre 
ferred stock of the holding company he would gain some 
security through the equity margin represented by the com 
mon. , 

This, broadly, is the story of utility holding companies be 
fore the great war. If holding companies have, in these later 
days, fallen under criticism, we must ever remember the 
service which they performed in the rapid expansion of elec 
trical services to the public. As an effective instrument for 
providing capital it made possible the rapid development and 
use of the steam turbine, not only in great centers of popu- 


lation but in power plants serving large numbers of rural 
communities through networks of transmission systems. In 
no other way could the small communities have received such 
high character of service at such low cost. 

This mechanism was developing during the period from 
1900 to 1910, during which time the business of the General 
Electric Company trebled. But even so, the full effect was 
not felt until the next two decades when the business of the 
Company rose from $70,000,000 to more than $400,000,000. 
This tremendous increase was not experienced alone by the 
General Electric Company, but by the entire electrical manu 
facturing industry. Indeed, General Electric's percentage of 
the industry business has remained through the years at be 
tween twenty and twenty-five per cent. 

There is another aspect, however, of this co-operation be 
tween the engineers of the utilities and the engineers of the 
electrical manufacturers. As I have suggested before, it was 
difficult for the manufacturers to keep pace with the demands 
of their own customers for new things. To do so, the General 
Electric Company in 1901 established a research laboratory 
and called to it men of great vision and ability to explore the 
unknown. Those men had no specified jobs. To improve what 
we had was the work of the engineers. The research men 
were to move out into an unknown land, and they were sup 
ported on their adventure in the hope that they would bring 
back among their discovered treasures a few at least that 
would be of practical advantage to the electrical industry. 
One might well have criticized in those days a corporate man 
agement which gambled stockholders' money on such remote 
chances. Yet I suppose no venture has been more profitable 
to the General Electric Company, to investors and operators 
in utilities and to the public served by them than these re 
search laboratories. 


The drive for economy and efficiency in the production of 
electrical power so urgently demanded by the utility engi 
neers was supported by the work of the research laboratories 
and the engineers of the manufacturers. Indeed, the profits 
of the utility operating companies became so large as not only 
to attract investors, but to precipitate constantly calls from 
customers for reduced rates. In the beginning, municipalities 
endeavored to regulate rates, but the utility rapidly outgrew 
the municipality. Then State Commissions were charged with 
that responsibility, but the utilities in some cases overran 
State boundaries and so the Federal government found its 
justification for a limited entry into the field of utility regu 

I should like to say in passing that in the political debates 
incident to the entrance of the Federal government into the 
field of regulation, there has been criticism of the incompe 
tence not only of local regulating bodies, but frequently of 
State commissions. I should like to take the opportunity of 
saying here that by and large I think those criticisms are un 
fair and unwarranted. The fact is that the research labora 
tories and the engineers in the electrical industry made such 
rapid progress in efficiencies and economies that regulating 
bodies were unable to keep up with them. As a consequence, 
by the time the rather cumbersome machinery of regulation 
established a rate, the efficiencies introduced meanwhile fre 
quently kept the profits on the new rate as high as they were 
on the old ones. This fact led customers to criticize the regu 
lating boards. Such criticism to my mind was unjustifiable. 
Indeed, on the whole, I think the fact that the utility com 
panies were highly prosperous during the early days was of 
great public advantage. Their earnings were largely turned 
back into their properties and so supplemented the capital 
which they were able to get from the public and which in 


amount would, I think, have been inadequate to have pro 
vided for the rapid expansion of facilities which took place 
between the years 1910 and 1930. 

I have spoken of the part which the engineer managers of 
the electrical utility companies played in the development 
of the business, and consequently the capital formation of the 
General Electric Company. To complete the picture it is 
now necessary to speak of the major adventures of the Gen 
eral Electric Company itself in the utility field. 

I have already told how essential it was in the earlier days 
for the Company to accept public utility securities in part 
payment for its apparatus and how large blocks of such utili 
ties were sold in the panic of 1893 to save the Company's 
life. Attention has also been called that prior to the organiza 
tion of the General Electric Company, Charles A. Coffin had 
organized the United Electric Securities Company of Boston 
as a medium through which, by the issue of its debentures 
and preferred stock, he could find an indirect market for the 
securities of operating companies which he wished to take. 
The United Electric Securities Company was continued in 
existence as a subsidiary of the General Electric Company 
and was used from time to time for the purpose of liquidating 
in part the utility securities of the General Company. Its 
headquarters were in Boston and its securities were largely 
sold in that market. 

In 1904 the General Electric Company caused to be or 
ganized the Electrical Securities Corporation in New York 
for the purpose of doing a similar business in New York and 
thereby widening the opportunity of the General Electric 
Company to aid in the financing of utilities. The common 
stock of this company too was held by the General Electric 
Company. Many years later, to save overhead and simplify 
the organization, the United Electric Securities Company of 


Boston was liquidated by a transfer of its assets in the do 
mestic field to the Electrical Securities Corporation, and in 
the foreign field to the International General Electric Com 
pany, which is the subsidiary of the General Electric Com 
pany doing its foreign business, except in Canada. It is 
unnecessary to speak of the United Electric Securities Com 
pany further than to say that it served a useful purpose for 
more than forty years with satisfactory returns to its investors 
and the General Electric Company and that it had rendered 
a substantial service to the financing of utilities during pe 
riods when the market was not ready to supply capital de 

The Electrical Securities Corporation still exists, but it no 
longer issues its securities to the public, all of the outstanding 
ones having been called and retired. It had developed 
through many years a highly specialized management and it 
is today, practically speaking, only an incorporated Securities 
Department of the General Electric Company. It too, as a 
financing mechanism to protect utility credits, had served a 
useful purpose. 

These two companies were investment companies. They 
were not in any sense management companies although at 
infrequent times they were compelled as investors to partici 
pate as directors, and occasionally as managers. Such activi 
ties, however, were purely incidental. 

I shall now speak of another adventure of the General Elec 
tric Company into the utility field which did not originally 
contemplate, but which ultimately eventuated, in manage 
ment and in a "holding company." I speak of the Electric 
Bond and Share Company. It was natural that the two invest 
ment companies above referred to should take the securities 
of the larger public utility operating companies. It was, of 
course, the ambition of the engineer managers, of whom I 


have spoken, to operate initially in the larger communities. 
Even after the turn of the century, the utilities operating in 
the smaller communities were finding it difficult to get ade 
quate capital and particularly equity capital. Prior lien se 
curities, such as bonds and preferred stocks of the smaller 
companies could only be sold on a high interest basis and 
when adequate capital for common stock was not available, 
the senior securities often could not be sold at all. 

Under these circumstances, the General Electric Company, 
again under the leadership of Charles A. Coffin, conceived 
the idea in 1905 of organizing the Electric Bond and Share 
Company for the purpose of aiding the smaller utilities in 
raising junior capital and in selling their senior securities at a 
better price. The development of the utility business in serv 
ing many communities from centralized power plants through 
the use of long transmission lines practically forced the Elec 
tric Bond and Share Company, through the normal evolution 
of its business, into a large holding and management com 
pany. Meantime, other operating units, through holding com 
panies and otherwise, were extending their lines to serve not 
only small cities, but small villages and hamlets; indeed, they 
were reaching out for the farms. As these operating units be 
came larger and the general trend of the utility business was 
profitable, the Bond and Share Company was well able to 
handle financing and to provide its own capital requirements 
without help from the General Electric Company. Accord 
ingly, on December 30, 1924, the General Electric Company 
decided to divest itself of ownership of the Electric Bond and 
Share Company by distributing the common stock of that 
Company to its own stockholders as a dividend. Such stock 
at the time of its distribution was valued on the books of the 
General Electric Company at $25,000,000. 

In 1919, after the close of the great war, the General Elec- 


trie Company, at the suggestion of the President of the 
United States, undertook to secure the co-operation of all 
American concerns interested in wireless in the creation of 
a unified organization to develop and protect the communi 
cation interests of the United States throughout the world. 
The story of radio and its rapid development need not be re 
peated here. It is only referred to because the Radio Corpora 
tion of America was organized for the above purposes in 1919 
and as the result of a decree of the Federal Court, the com 
mon stock of the Radio Corporation which the General Elec 
tric Company held was distributed in 1933 to the stockholders 
of the General Electric Company as a dividend. The value 
of that stock as carried on the books of the General Electric 
Company was approximately $26,500,000 and that amount 
was deducted from surplus in that year. 

While speaking of dividends, perhaps I should say at this 
point that in addition to the Electric Bond and Share and the 
Radio Corporation dividends of $51,000,000, the General 
Electric Company has, during the forty-eight years of its life, 
paid cash dividends to its stockholders of $655,662,000 and 
an additional $47,000,000 to retire its special stock issued as 
stock dividends. During this same period the Company has 
sold approximately $7,000,000,000 worth of electrical prod 
ucts and it has paid to its employees in wages and salaries 
something over $2,700,000,000. 

Now we are back to the point where I started. The enter 
prise has been profitable, but it took the vision and daring of 
genius to create it. It took confidence, persistence and cour 
age to develop it and carry it through its earlier days. It re 
mains only for us at this time and for those who come after 
us to so administer this great concern that it may continue 
to render a service in the future comparable to the past. 


Adams, Thomas E. 

Brush-Adams lamp invented 69 

Addison, Dr. Thomas H. G-E 
representative on West Coast 200 

Alexanderson, Ernst F. W. 

alternator on trial at New Bruns 
wick, 372; duplex radio telephone 
system, 354; single-phase railway 
motor, 350; high-frequency alter 
nator, 351; magnetic modulator, 
352; multiple-tuned antenna, 354; 
two-way wireless telephony, 354 

Alteneck, inventor of Siemens dy 
namo 7 

American Electric Company 

organized, 33; bought by Lynn 
Syndicate, 54 

American Marconi Company 

Alexanderson alternator tested, 
372; purchased by G. E. Co., 377 

American Telephone & Telegraph 
part in organizing RCA 377 

Ames, F. L. 

financial supporter of Thomson- 
Houston Co., 147; Director of 
G. E. Co., 195, 406 

Andrews, William S. 

Edison lighting installation, Sun- 
bury, Pa., 59 

Arc lamps 

Brush type, 7, 8; displayed at Dr. 
Longworth's residence, 27; first 
installation, Wanamaker's store, 
13, 30; automatic regulator, 15, 
19; street, first installation, 27; 
Jenney type, 71; luminous, 310 

Asmussen, Oscar, friend of Stein- 
metz 169 

Association of Edison Illuminating 

organized, 97; renewing custom 
er's lamps, 265 


Babcock, A. A., turbine-electric ship 
propulsion 330 

Baltimore & Ohio Railroad, tunnel 
electrification 254 

Barton, Silas A. 

General Manager of Thomson- 
Houston Co., 164; organizer of 
Lynn Electric Lighting Co., 52 

Batchelor, Charles 

came to Schenectady, 149; Edi 
son's model maker, 22 




Bell, Dr. Louis 

installation at Mill Creek, Cal., 
232; three-phase induction motor 
tests, 209 

Bentley, Edward M., pioneer in 
electric transit 
first electric car 78 

Bentley-Knight Electric Railway Co. 
contract for equipment, Observa 
tory Hill Passenger Railway Co., 
121; patents, 141 

Bergmann and Company 

organized, 43; formed part of 
Edison G. E. Co., 152, 156 

Boehm, Ludwig, Edison's glass- 
blower 22 

Bolton, Werner von 

lamp filaments; tantalum wire, 
315; tungsten, 335 

Bradley, Charles S. inventor of rotary 
converter 230 

Brady Manufacturing Co. 

Wood's dynamo manufactured 17 

Brockton, Mass. 

three-wire lighting system in 
stalled 61 

Brush, Charles Francis 

arc headlight, 69; arc lamps, 7, 8; 
combined incandescent and arc 
lamp system, 57; copper-coated 
carbons, 30; double-carbon arc 
lamp, 142; dynamo designed, 6, 
8; lamps, Broadway installation, 
41; patent agreement with Tele 
graph Supply Co., 9; sixteen-light 
dynamo, Niagara Falls, first instal 
lation, 30; street lighting installa 
tion, 27; retired, 197 
U. of Michigan graduate 7 

Brush Electric Co. 

lamps and generators, 215216; 
manufactured Short motors, 142; 
stock, controlling interest sold to 
Thomson-Houston Co., 163 

Brush Electric Light and Power 
Broadway lighting 41 

Buck, Harold W. 

strain insulator invented 326 

Buckeye Incandescent Lamp Co., 
186; sealing-in machine, 261 

Burrows, William R. lamp improve 
ments 264, 301, 380 

Bush, Arthur R. 

electric lighting on steamers 64 

Cahoon, J. B. Chief of Experts, 
Thomson-Houston Co. 164 

California Electric Light Company 
incorporated, 29; selling electric 
service, first, 29 

Canadian Edison Manufacturing 
Company purchased 156 

Carbon brushes 

lubrication, 312; railway motor 
use, 137 

Carbon products manufactured 313 

Cataract Construction Company, 
Niagara Falls 235 

Centennial Exposition, machinery 
exhibits 5 

Central London Tube Railway 
Sprague elevator contract 292-294 

Cermak, William, porcelain maker 


Charles A. Coffin Foundation, 
Awards, established 382 

Chesney, Cummings C. 

started Pittsfield plant 177 

Churchill, Frederick H. 400 

Treasurer American Electric 
Co. 33 

Cincinnati, Ohio 

arc lamp displayed 27 

Circuit breakers, oil, early develop 
ment 267 

Clark, John H., electric ship pro 
pulsion study 330 

Clark, William J. 
joined Thomson-Houston Co., 132; 
Manager Railway Department, 
285; motor contract for South Side 
Elevated Railway of Chicago, 293 

Clarke, Charles L. Acting Chief 
Engineer, Edison Electric Light 
Co. 182 

Cleveland, arc lighting, first large 
scale installation display 27 



Coffin, Charles A. 

company consolidation plans ac 
cepted, 193; full time to Thomson- 
Houston Co., 90; first Director of 
G. E. Co., 195, 406; becomes 
President of G. E. Co., 196, 405; 
President of G. E. Co., resigns, 
348; Chairman of Board, G. E. 
Co., 348; Chairman of the Board, 
G. E. Co., retired, 382; died, 384 

Columbia, lamp manufacturer 186 

Columbia, South Carolina, textile 
mills electrical installation 210 

Condensation pump, mercury 352 

Consolidated Electric Lighting Com 
pany, lamp manufacturer 187 

Coolidge, T. Jefferson, Jr. 

financial supporter of Thomson- 
Houston Co., 147; first Director of 
G. E. Co., 195, 406 

Coolidge, William D. 

came to Schenectady, 332; ductile 
tungsten experiments, 338; x-ray 
tube invented, 355 

Cooper Hewitt Electric Co. 

merged with G. E. Co. 310 

Corbin, P. and F. Company 

electric light tests 49 

Coster, C. H. 

first Director of G. E. Co. 195, 406 

Cowles, Alfred and Eugene 

electric furnace experiments 103 

Crescent Beach 

street car line electrification 133 

Crosby, O. T., Railway Department, 
first 198 

Culver, W. S., Brush Co. salesman 


Curtis, Charles G. 

patent agreement with G. E., 276; 
steam turbine design and patents, 
275, 279 


Dalzell, J. A., joins Thomson- 
Houston Co. 144 

Darling, Henry W., Treasurer of 
G.E. Co. 177 

Davis, Albert G. 

Vice-President of G.E. Co. in 
charge of Patents 298, 376 

Day, Maxwell W., electric ship pro 
pulsion study 330 

Dempster, J. T. H. 
Assistant in Research Laboratory 


Direct-current dynamo, Thomson's 
invention 12 

Dow, Alex. President Detroit Edison 
Co., idea for automatic substa 
tion 362 

Dumet wire 302 

Duncan, Samuel A., lawyer 

181, 183 

Dyer, Richard N., patent attorney 



alternating current, Slattery type, 
105; "drum armature" type, 7; 
Edison's bi-polar, 23; exhibited, 
Centennial Exposition, 5; Gramme 
type, 5, 7; "ring armature" type 
designed, 7; Siemens-Halske type, 
7; Wallace type, 5 


East Cleveland Railway, first electric 
car 80 

Edison, Thomas Alva 

bi-polar dynamo, 23; current 
measuring chemical meter, 99; 
dynamo manufactured, Edison 
Machine Works, 42; electric rail 
ways, experimental, 76; incandes 
cent lamp, 22; knife-blade switch, 
first type, 102; lamp manufacture, 
171; patent controversy, 179; 
patent suit, 180; patent, English 
Court decision, 185; three-wire 
system, 59 
Personal Notes 

central station operation, wages, 
95; personality, 175; first Director 
of G.E. Co., 195, 406; retired 
from G. E. Co., 196 



Edison Company for Isolated Light 
ing 63 
merged with Edison Electric 
Light Co. 152 

Edison effect 94, 346 

Edison Electric Illuminating Com 
pany of New York 
chartered, "First District," 40; 
merged with Edison G. E. Co., 

Edison Electric Light Company 
contract to sell lamps, 43; financ 
ing of, 402; incorporation, 21; 
formed part of Edison G. E. Co., 
156; lamp patent suit, 180; merger 
of Edison Company for Isolated 
Lighting with, 152; sale of stock, 

Edison General Electric Co. 

business compared with Thomson- 
Houston, 1891, 404; directors act 
on consolidation terms, 194; hold 
ings at time of consolidation, 194 

Edison Laboratory, Menlo Park, first 
buildings erected 20 

Edison Lamp Company, organized, 
43, 402; formed part of Edison 
G. E. Co., 156 

Edison Machine Works 

formation of, 42, 402; built first 
Jumbo dynamo, 45; manufactured 
Edison dynamo, 42; moved to 
Schenectady, 114, 149, 402; 
formed part of Edison G. E. Co., 

Edison Shafting Company 

organized, 43; merged with Edi 
son Machine Works, 149 

Edison Tube Company 

organized, 43; merged with Edi 
son Machine Works, 149 

Edison United Manufacturing Com 
pany, purchased 156 

Edwards, E. P. Manager, Radio De 
partment, G. E. Co. 376 

Eickemeyer and Osterheld, first to 
employ Steinmetz 170 

Eickemeyer plant, purchased 199 

Electric Bond and Share Co. stock 
distributed to G. E. Co. stock 
holders 388, 423 

Electric furnaces 

arc, 103; resistance, 314 

Electric locomotives 255, 361 

pioneer inventors 78 

Electric motors 

Eickemeyer winding method, 256; 
operating ammunition hoists, 253; 
repulsion first built, 146; railway 
types, 142; Sprague early uses, 
115; Sprague experiments, 62; 
three-phase induction, 209 

Electric ship propulsion 368 

Electrical Securities Corporation, 
organized 249, 421 

Electricity, first commercial sale 

29, 31 

Electroplating, first use of dynamo 


Elevated electric railroad, Chicago 
World's Fair, 1893 214 


Sprague motor used, 117, 290; 
Sprague-Pratt type, first installa 
tion, 290 

Emmet, William LeRoy 

Curtis steam turbine tests, 277; 
electric ship propulsion, 330, 368; 
steam turbine mechanical con 
struction, 279; work at Niagara, 

Emmons, George 

executive ability, 245; personality, 
244; bookkeeper, American Elec 
tric Co., 35; Factory Auditor, 
Thomson-Houston Co., 245; au 
ditor at Lynn Works, 87; Plant 
Manager at Lynn Works, 245; 
Assistant Manager of Schenectady 
Works, 245; Manager of Schenec 
tady Works, 245 

English Court, decision on Edison 
lamp patent 185 

"Experts," (early engineers) 68, 164 


Fabroil, cloth gear 318 

Faccioli, G. transmission million volts 
of electricity 379 

Factory organization, Thomson- 
Houston Co. plan 198 



Farmer, Moses G., inventor incan 
descent lamp 182 

Fireboats, first electrically propelled 


Fish, Frederick P. 

Counsel for Thomson-Houston 
Company, 193; General Counsel 
of G. E. Co., 197 

Fiske, Lieutenant Bradley A. 251 

Fiske Street Station, Chicago, to be 
gin operation 279 

Folsom Falls, California, electric 
power plant 250 

Force, Martin, Edison's laboratory 
assistant 22 

Fort Wayne Electric Company 

Eurchased interest in Wood's arc- 
ght system, 164; reorganized, 
225; Slattery alternating-current 
system, 215-216 

Fort Wayne Electric Corporation 
organized, 226; purchased by 
G. E. Co., 284 

Fort Wayne Electric Light Com 
pany 71, 143 

Foster, W. J., three-phase induction 
motor tests 209 

Franklin Institute of Philadelphia, 
tested dynamos 10, 12, 23 

French Thomson-Houston Company 


Fuller Bakery, Philadelphia, Thom 
son-Houston dynamo installed and 
tested 14, 400 

Fuller-Wood Co., Wood's dynamo 



Gardner, George P., financial sup 
porter of Thomson-Houston Co. 


Garfield, E. I, first Secretary of G. E. 
Co. 197 

Garrett, George S., witnessed dem 
onstration, Thomson's inventions 


Gary, Indiana, steel mill, electric 
drive 320 

Gaulard and Gibbs system 107, 177 
"Gem," incandescent lamp devel 
oped 315 
General Electric Co. 

broadcasting stations, 393; Comp 
troller, first, 198; consolidation an 
nouncement to employees, 195; 
cooperation with labor, 387; De 
partment Heads, first, 198; Direc 
tors, first, 195, 406; District Offices 
established, 176, 198; Electric 
Bond and Share Co. stock dis 
tributed to stockholders, 388; En 
gineering Committee formed, 221; 
Erie Works, land purchased, 331; 
Executive Committee formed, 
220; executive officers, first, 197; 
exhibits, World's Fair Chicago, 
1893, 215; financial reports, 221, 
397-398, 406-415; frequency- 
modulation station, 393; Home 
Office transferred to Schenectady, 
226; incorporation date, 193, 405; 
Manufacturing Committee formed, 
221; Moore's gaseous lamp patents 
obtained, 332; Mutual Benefit As 
sociation, established, 383; new 
title beginning date, 195; Niagara 
power plant plans, 235; patent 
agreement with Curtis, 276; patent 
agreement with Westinghouse, 
Tesla type polyphase induction 
motor, 287; patents held, 272; 
Pension Plan established, 383; 
Pittsfield Works started, 177; 
profit-sharing system established, 
383; progress of industry since 
1922, 385-394; purchase Cooper 
Hewitt Electric Co., 310; Research 
Laboratory organized, 299, 314; 
River Works, Lynn, new plant, 
287; Sales Committee formed, 
221; Schenectady Works, history 
of buildings, 150, 242-243, 305, 
313-314; sales, 1892-1938, 415- 
416; stockholders and shares is 
sued, 398; tantalum lamp manu 
facture, 316; tungsten filament, 
commercially introduced, 335; 
tungsten filament patent pur 
chased, 335; Young, Owen D., 



General Electric Co. Continued 
statement on G-E financial his 
tory, 397-424 

Generators, expansion in use of 
205, 214, 363 

Gibson, Langdon, Production De 
partment Manager 246 

Gramme, early dynamo inventor 7 

Grand Central Terminal, electrifica 
tion 288 

Great Harrington, Mass., alternating- 
current lighting system 105 

Greene, S. Dana 

commercial engineer, 235; gun 
turrets electrically operated, 252; 
head of Lighting Department of 
G. E. Co., first, 198 

Greene, Stephen, textile mill en 
gineer 210 

Griffin, Eugene 

first Director of G. E. Co., 195, 
406; joined Thomson-Houston Co., 
132; personal interest in young 
experts, 166; Vice-President of 
G. E. Co. first, 197; Vice-Presi 
dent of G. E. Co. in charge of 
Commercial Operations, 247 

Gun turrets, electrically operated 252 


Hadfield, Sir Robert, silicon steel 
alloy discovered 331 

Halvorson, C. A. B., perfected arc 
lamp manufacture 305 

Hanaman, Franz, tungsten filament 
experiments, success of 333 

Harrison Lamp Works, lamps manu 
factured 43 

Hastings, Frank S., first Director of 
G. E. Co. 195, 406 

Hayden, J. LeRoy, Steinmetz's as 
sistant and friend 308 

Hewitt, Peter Cooper, invented 
mercury-arc lamp and mercury- 
vapor rectifier 310 

Hewlett, Edward M. 

light installations, New York and 

vicinity 201 


relays designed, 268; Panama 

Hewlett, Edward M. Continued 
Canal control system designed, 
357; steel mill motors control de 
signed, 321; switches and switch 
boards designed, 214, 268; sus 
pension type insulator, 326; de 
veloped Selsyn system, 358 . 

Hicks, W. Preston, Sales Manager, 
Edison Electric Light Co. 156 

Higginson, Henry L. 

first Director of G.E. Co., 195, 
406; financial supporter of 
Thomson-Houston Co., 147 

Hoboken Shore Road, two-truck 
locomotive built 288 

Horse cars, horses and cars estimated 
number 3 

Houston, Edwin J. 

Brush inventions studied, 10; 
patent contracts with American 
Electric Co., 34; Professor of Boys 
Central High School, 12, 399 

Howell, John W. 

electrician Edison Lamp Com 
pany, 182; lamp engineer, 94, 301; 
lamp experiments, 204, 252, 264 

Howell, Wilson S., insulating com 
pound made 39 

Hydroelectric plant 

first on Pacific Coast, 201; Mill 
Creek, 231; Niagara Falls, 235 


Illuminating Engineering Labora 
tory, transferred to Schenectady 


Incandescent lamps 

early experiments, 37; legal 
definition, 183-185; improve 
ments in, 178, 265; tungsten, 
first, 333; tungsten, ductile, 336 
first sale 43 

gas-filled 346, 348 

"Gem" 315 

invented 22 

early manufacture 43, 92 

patent suit 180-187 

sales, 1892 206 

Sawyer-Man type 58, 93 



Incandescant lamps Continued 
sealing-in machine (Massey and 
Spiller) 262 

Swan type 57 

tungsten manufacture begun 338 

Induction motors 

first repulsion type built, 146; 
three-phase experiments, 209 

Inductor alternator, developed by 
Stanley Electric Manufacturing 
Co. 284 


made and tested, 243; porcelain, 
269; strain type, 327; suspension 
type, 326 

Insull, Samuel 

President Commonwealth Edison 
Company of Chicago, 279; Vice- 
President of Edison G. E. Co., 
176; 2nd Vice-President of G. E. 
Co., 197 

Interborough Rapid Transit Co., 
electrification 297 

International General Electric Co., 
organized 381 

Intramural Railway, Chicago World's 
Fair, 1893 214 

Jackson, Mich., magnetite lamps, 
first used 310 

Jehl, Francis, Edison's laboratory 
assistant 22 

Jenney, James, invented arc lamp 71 

Jenney and Schaffer direct-current 
arc dynamos 143 

Johnson, Edward H. 

organized Sprague Elevator Co., 
197; President Edison Electric 
Light Co., 84; President, Sprague 
Co., and Edison Electric Light 
Co., 155 

Jumbo dynamo, first 45 

Just, Alexander, tungsten filament 
experiments, success of 333 


Kellogg, J. W. Commercial Manager 
of Marine Work 252 

Kelly, John F. joined with Stanley 
and Chesney 178 

Knight, Walter H. 

first electric car, 79; electric street 
car development, 197; switchboard 
designed, 214 

Kruesi, John 

Assistant General Manager Edison 
Machine Works, 149; came to 
Schenectady, 149; Consulting en 
gineer, 245, 276; Edison's ma 
chine-shop expert, 22; Manager 
Schenectady Works, 197, 242 

Kuzel, Hanz, tungsten filament ex 
periments 335 

Langhans, Rudolf, lamp filament ex 
periments 171 

Langmuir, Irving 

came to Schenectady, 345; lamp 
filament experiments, 346; mer 
cury condensation pump invented, 
352; radiotron invented, 353 

Libbey Glass Company, purchased 
by G. E. Co. 302 


arc lamps, Brush type, 7; New 
York, first installation, 46-48; 
World's Fair, Chicago, 1893, 213; 
Wabash, Indiana first town wholly 
lighted by electricity, 31 

Lightning, artificial, Steinmetz's ex 
periments 379 

Livor, Harry M. chose Schenectady 
for site of factory 113 

Locomotives, electric, 255, 361 

Lovejoy, J. R. 

personal interest in young experts, 
166; Supply Department, first, 
198, 287; visits Lynn factory with 
President Henry Villard, 193 

Lowrey, Grosvenor P. 

Edison's friend and promoter of 
Edison Electric Light Company, 
21, 402; Edison's lawyer, 181 

Lynn Electric Lighting Company, 
organized 53 




McCool, Bill, millwright 241 

McDonald, R. T. 

arranged arc lamp demonstration, 
70; negotiated sale of Fort Wayne 
Co., stock, 143; died, 284 

McKee, John R. 

gun turrets electrically operated, 
252; Manager, Commercial De 
partment Thomson-Houston Co., 
90; Manager, Power Department 
of G. E. Co., 198; Manager, Power 
and Mining Department of G. E. 
Co., 247, 285; President, Thom 
son-Houston Motor Co., 91; Presi 
dent, Thomson- Van Depoele Elec 
tric Mining Co., 91 

Madigan, William 

foreman, Schenectady Works, 240, 
243, 280, 281; planned building 
Curtis turbine, 281-282 

Magnetic blowouts, electric switches 
and street car motors 35, 206 

Magnetite lamps, first used 310 

Malignani, Arturo, vacuum process 
for lamps 263 

Man, Albon, inventor incandescent 
lamp 182 

Manhattan Elevated Railway, elec 
trification 288, 295 

Marks, L. B., enclosed arc lamp pat 
ent purchased 261 

Mason, Dave, emergency man, Spra- 
gue Co. 153 

Maxim, Hiram S. 

inventor incandescent lamp, 182; 
witness in lamp patent suit, 181 

Maxwell, Howard, designer electric 
drive for steel mills 321 

Menlo Park, first buildings erected 20 

Metal brushes, for railway motors 129 

Metropolitan Street Railway Co., 
underground conduit adopted 257 

Miller, John 

Chief Millwright, 241; cloth gear 
perfected, 316; foreman, 243 

Mills, D. O., Director of G. E. Co., 
first 195, 406 

Mine locomotives, electric 91, 286 

Mitchell, Loris E. design tipless lamp 


Mitchell, Sidney Z. hydroelectric de 
velopment, Seattle 201 

Moody, W. S., Transformer Engineer, 
silicon steel alloy, rolling process 


Moore, D. McFarlan, gas-discharge 
lamp development 262, 331 

Morgan, J. P., Director of G. E. Co., 
first 195, 406 

Morrison, George F. 

first work and pay, 43; foreman, 
Edison G. E. Co., 173 

Mulvey, Peter J. foreman carbon 
brush improvement, 311; personal 
ity, 312 


National Electric Lamp Association 
organized, 301; G-E interest in, 
340; buildings erected, 344; dis 
solved, 344 

"Nela" origin of name 344 

New Mexico, battleship, turbine elec 
tric drive, first 368 

Niagara Falls 

first illumination, 30; power plant, 

"Novagems" 366 

Observatory Hill Passenger Railway 
Co., road equipment 121 

Oil switch, early use 267 

Old Colony Steamship Company, in 
stall Edison lighting system 65 

Ord, Joseph P. 

Comptroller of Edison G. E. Co., 
157; Comptroller of G. E. Co., 
first, 198; 2nd Vice-President of 
G. E. Co., 197 

Overhead trolley wire system, U.S. 
Senate hearing 135 

Pacinatti, dynamo, "ring armature" 
type designed 7 



Paine, Sidney B. 

electric lighting in textile mills, 
63; electrical salesman, 210 

Panama Canal control system 356 

Panics, effect on G. E. Co. 

1893 221, 407 

1907 413 

Parker, William, President American 
Electric Co. 33 

Parshall, Horace F. 

Manager, Calculating Department 
of G. E. Co., 200; designed gen 
erator, 214; designed motor, 211; 
engineer, Niagara plant, plans, 
235; engineer, Sprague Co., 154 

Parsons, Hinsdill, Vice-President and 
General Counsel, G. E. Co. 359 


agreements, Brush and Telegraph 
Supply Co., 9; consolidations, 192; 
Patent Control Board, 272; lamp 
monopoly, 340 

Peabody, S. Endicott, financial sup 
porter of Thomson-Houston Co. 


Peach, Benjamin F., Jr. Treasurer of 
G. E. Co., first 197 

Peek, Frank W., Jr. transmission mil 
lion volts of electricity 380 

Pelton impulse water wheels 232 

Pevear, Henry A., organizer of Lynn 
Electric Lighting Co. 53 

Polyphase system, first in world 231 

Polyphase transmission 218 

Porter, Charles T., designed steam 
engine 44 

Potter, William B. 

engineer, Railway Department of 
G. E. Co., 257; selling experiences, 

Pratt, Charles E., designer elevator 
motor 289 

Protective equipment, for transform 
ers 108 

Radio Corporation of America 
organized, 376; distribution of 
stock held by G. E. Co., 424 

Railway motors, Brush, Sprague, 
Thomson-Houston compared 142 

Reed, Philip D. Chairman of the 
Board, G. E. Co., elected 394 

Regenerative braking 

electric trains, 361; first time used, 

Reist, H. G. 

fenerator design for steam tur- 
ine, 279; three-phase induction 
motor work, 209 

Resistance welding 

principle discovered, 12; Thom 
son's method, 50 

Rheostats, railway motor demonstra 
tion 85 

Rice, E. W., Jr. 

personal interest in young experts, 
166; steam turbine interest, 275; 
work on Niagara power plant 

Elans, 235; superintendent at 
ynn, 87; Thomson-Houston or 
ganization plan, 102; Technical 
Director of G. E. Co., first, 197; 
Vice-President in charge of En 
gineering and Manufacturing, 247; 
President G. E. Co., 349; Honor 
ary Chairman of the Board, G. E. 
Co., 382 

Rohrer, Albert H. 

assistant engineer at Lynn, 87; en 
tertains President Villard, Edison 
G. E. Co., 191 

Rotary converter Bradley type, first 
used 238 

Ryan, Walter D'Arcy 

Illuminating Engineer, 266; illu 
mination installations supervised, 
364; designed Tower of Jewels, 
Panama-Pacific Exposition, 367; 
scientific basis for illumination, 

Rach, Christian, foreman 150, 243 
Radio, trans-oceanic system, first 351 

San Francisco, headquarters Pacific 
Coast territory 200 



Sawyer-Man Electric Company, in 
candescent lamp manufacture 

58, 93, 182, 187 

Schaffer and Jenny, direct-current 
arc dynamos 143 


in 1886, 111; experimental rail 
way track, 293; home office G. E. 
Co., 226 

Schuyler, D. A. 

sixteen-light arc dynamo invented, 
143; joined Thomson-Houston Co., 


Mt. Washington, 202; St. Louis 
Exposition, 311; U.S. Army and 
U.S. Navy, 311 

Securities, G. E. Co., 1893 221-223 

Short, Sidney Howe, single-reduction 
motor experiments 142 

Siemens-Halske Co. 
Alteneck's drum-armature type 
dynamo invented, 7; purchased by 
G. E. Co., 284 

Slattery alternating-current system 


Smythe, Captain, plan electrification 
textile mills 248 

South Side Elevated Railway of Chi 

electrification, 292; multiple-unit 
train control, first installation, 294 

Southern California Edison Co. 231 

Sprague, Frank J. 

electric street railroad built, Rich 
mond, Va., 123; elevator develop 
ment, 117, 289; motor experi 
ments, 62, 84; multiple-unit 
system of train control, 291, 295; 
railway motor experiments, 84, 
115; streetcar motor designed, 78, 
115; three-wire lighting system in 
stalled, Sunbury, Pa., 60; trolley 
pole invention refused, 83; re 
signed from Sprague Co., 158 

Sprague Electric Co. organized 294 

Sprague Electric Elevator Co. 
organized, 197, 290; merged with 
Interior Conduit and Insulation 
Co., 294 

Sprague Electric Railway and Motor 

incorporated, 84; electric street 
railroad built, 123-125; electrifica 
tion of West End Road of Boston, 
158; growth of motor business, 
1888-1889, 153; merged with Edi 
son G. E. Co., 157, 289 

Sprague Interior Conduit and In 
sulation Company 197 

Stanley, William 

Pittsfield Works started, 177; trans 
former designed and built, 106, 

Stanley Electric Manufacturing Com 

organized, 178; purchased by G. E. 
Co., 284 

Statue of Liberty, lighting 105 

Steel mills, electric driven, Gary, 
Indiana 320 

Steele, George F. 

Sprague motor salesman, 116; 
Sprague elevator motor, interest 
in, 289 

Steinmetz, Charles Proteus 

arrived in America, 169; personal 
ity, 199, 308; appointed Head En 
gineer, 230; experimental labora 
tory, 304; explained hysteresis 
law, 229; magnetite-titanium elec 
trode study, 305; manufactured 
lightning, 379 

Stiles, Aaron K., Manager Van De- 
poele Electric Manufacturing Co. 


Stockly, George W., offered shop 
facilities to Brush 8 

Streetcar motors, series-parallel con 
trol 207 

Street lighting 

arc lamps, first demonstration, 27; 
Wabash, Indiana, first town wholly 
lighted by electricity, 31 

Street railroads 

Crescent Beach, electrification of, 
133; Observatory Hill, electric 
equipment, 121; Richmond, Va., 

Substations, automatic first in opera 
tion 362 



Sunbeam Incandescent Lamp Com 
pany of Chicago 186, 299 

Sunbury, Pa., central lighting sys 
tem 59 

Sunny, B. E., Manager Chicago Of 
fice of Thomson-Houston Co. 200 

Swan, Joseph W., incandescent lamp 
invention 57 

Swope, Gerard 

work at World's Fair Chicago, 
1893, 215; first President, Interna 
tional General Electric Co., 381; 
President G. E. Co., 382; Presi 
dent, G. E. Co., retired, 394 

Synchronous converter, invented 231 

Taylor, Edward, designed first auto 
matic substation 362 

Telegraph Supply Co., part in 
Brush's early invention 8-9 

Terry, Franklin S. 

organizer of National Lamp Works, 
300; proprietor Sunbeam Incan 
descent Lamp Co. of Chicago, 299 

Textile mills 

electrical machinery, first installa 
tion, 210; install Edison's lighting 
system, 64; Pelzer, So. Carolina 
electrification, 248 

Textolite 318 

Thomson, Elihu 

experiments, first, 11; Brush inven 
tions studied, 10; constant-current 
transformer, 260; direct-current 
dynamo invented, 12, 92; dynamo 
designed, 6, 106; inductorium in 
vented, 355; oil-immersed trans 
formers, patent, 238; protective 
equipment, transformers, 108; re 
cording watt-meter, 168; repulsion 
motor, first built, 146; resistance 
welding method, 50; Roentgen-ray 
transformer invented, 355; trans 
former designed, 107 
Personal Notes 

declined membership on Board of 
G. E. Co., 197; "Electrician" 
American Electric Co., 34; Profes- 

Thomson, Elihu Continued 

SOT Central High School of Phila 
delphia, 11, 399 

Thomson-Houston Electric Company 
organized, 54, 400; early days, 88, 
165; business, 1883-1888, 141; 
business compared with Edison 
G.E. Co., 1891, 404; consolida 
tion terms, 194; patent contract 
with American Electric Company, 
34; electric railway developments, 
130; electrification of West End 
Road of Boston, 159; incandescent 
lighting system planned, 58; elec 
tric locomotive for tunnel use, 254; 
purchased Ft. Wayne Co. stock, 
143; Van Depoele Co. patents ac 
quired, 131 

Thomson-Houston International Co. 
organized 91 

Thomson- Van Depoele Electric Min 
ing Co., electrical mining equip 
ment 91 

Thomson Welding Co., organized 92 

Tournier, Julien Charles, type of in 
dustrial worker 329 

Train control 

contactor form of series-parallel, 
295; multiple-unit method, Spra- 
gue's idea, 291 


design, early types, 106-107; de 
sign, Stanley, first built, 178; oil- 
immersed, began use of, 238 

Transmission lines 

California and vicinity, 250; Niag 
ara to Buffalo, 236 

Tremaine, B. G., Organizer of Na 
tional Lamp Works 300 

Trench, W. W. last interview with 
Mr. Coffin 383 

Trolley poles 

invention, 83; Sprague invention 
refused, 83 

Turner, William B., personality 150 

Twombley, Hamilton McKay 

Chairman of Board of G. E. Co., 
196; Director of G. E. Co., first, 
195, 406; consolidation plans, 
Thomson-Houston Electric Co. 
and Edison G. E. Co., 193 




Underground conductors, first used, 
three- wire lighting system 61 

United Electric Securities Co. 

223, 404, 421 

United Fruit Company, radio inter 
ests combined with Radio Corpo 
ration 377 

United States Electric Lighting Com 
pany, lamp patent suit 180 

Upton, Francis R. 

Edison's mathematician, 22; Man 
ager Edison Lamp Works at Harri 
son, 197 

Utility holding companies, value of 


Van Depoele, Charles J. 

carbon for brushes, suggested and 
accepted, 135; trolley pole inven 
tion, 82-83; died, 197 

Van Depoele Electric Manufacturing 
Company of Chicago 82, 131 

Villard, Henry, President Edison 
G. E. Co. 191 

Vinal, O. B., foreman, Gary installa 
tion 319 


Wabash, Indiana, first town wholly 

lighted by electricity 31 

Wages, central station operators, 

Edison's plan 95 

Wallace, Judge, opinion in lamp suit 

Wanamaker's store, Brush arc lamps, 

first installation 13 

Ward Leonard system 252 

Waterhouse, A. G. 

join Thomson-Houston Co., 144; 

sixteen-light arc dynamo invented, 


Weintraub, Ezechiel, mercury arc 
rectifier 309 

West End Road of Boston, electri 
fication 158 

Western Electric Co., part in or 
ganizing RCA 377 

Westinghouse Electric and Manufac 
turing Company 

organized, 106; part in organizing 
RCA, 377; patents held, 272; poly 
phase induction motor, Tesla type, 
patent agreement, 287; trans 
former design, 106 

Weston, Edward, inventor incandes 
cent lamp 182 

White, Arthur J., foreman, design 
tipless lamp 380 

Whitestone, Samuel, bookkeeper of 
Sprague Co. 157 

Whitney, Willis R. 

magnetite-titanium electrode study, 
305; organized Research Labora 
tory of G. E. Co., 299 

Wilson, Charles E., President G. E. 
Co. 394 

Wood, James J. 

arc-light system, 164; automatic 
regulator invented, 19; dynamo 
invented, 6, 17 

World's Fair, Chicago 1893 

General Electric exhibits, 215; il 
lumination, 213 

Young, Owen D. 

came to Schenectady, 360; Chair 
man of the Board, G. E. Co., 382; 
Chairman of the Board, G. E. Co., 
retired, 394; General Counsel, 
G. E. Co., 359, 381; Report to 
Temporary National Economic 
Committee, 397