(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "The evolution of the steam locomotive (1803 to 1898)"

I 



cs S 

C S 

w q 

55 M 





THE 



EVOLUTION 



OF THE 



STEAM LOCOMOTIVE. 

(1803 TO 1898.) 
BY 

G. A. SEKON 

'Editor of the J** Railway Magazine" and " Railn-ay Year Book? 
Author of "A History of the Great Western Railway >" 



Confccm : 

THE RAILWAY PUBLISHING CO., LTD., 
79 TO 83, TEMPLE CHAMBERS, TEMPLE AVENUE, E.G. 

1899. 



PREFACE. 



In connection with the marvellous growth of our railway system 
there is nothing of so paramount importance and interest as the 
evolution of the locomotive steam engine. 

At the present time it is most important to place on record 
the actual facts, seeing that attempts have been made to disprove 
the correctness of the known and accepted details relative to several 
interesting, we might almost write historical, locomotives. 

Jn this work most diligent endeavours have been made to 
chronicle only such statements as are actually correct, without 
reference to personal opinions. 

In a broad sense, and taken as a whole, the old works on locomo- 
tive history may be accepted as substantially correct. 

From these, therefore, and from authentic documents provided 
by the various railways, locomotive builders, and designers, together 
with the result of much original research, has the earlier portion of 
this account of the evolution of the locomotive steam engine been 
constructed. The various particulars of modern locomotive practice 
have been kindly supplied by the locomotive superintendents of the 
different British railways, so that no question can arise as to the 
strict accuracy of this portion of the work. 

Nearly forty years ago it was authoritatively stated: "That kind 
of knowledge of the locomotive engine which answers the purpose of 
a well-informed man has already become so popular that it almost 
amounts to ignorance to be without it. Locomotive mechanism, is very 
simple in its elementary nature, and the mind is naturally disposed 
to receive and retain any adequate explanation of striking phenomena, 
whether mechanical or otherwise ; and hence it is that there aro 
thousands of persons who, although in no way concerned in the 
construction or working of railway engines, are nevertheless com- 
petent to give a fair general explanation of their structure and mode 
of working." 

If such were true at that time it is abundantly evident that 
it is more so at the threshold of the 20th century, consider- 
ing the growth of inquiry into, and appreciation of, scientific 
and mechanical knowledge by an ever widening and increas- 
ing circle of general readers, which has been one of the marked 
igns of intellectual development during recent years. Under 

A2 

498 



iv. PREFACE 

such circumstances it is not surprising that the locomotive and its 
history have received a large share of public attention. Whilst 
railway officers, with the intelligence for which they are justly dis- 
tinguished, have always evinced a proper desire to be acquainted 
with the evolution of the " steam horse," the spread of education 
has increased and quickened a desire for knowledge concerning the 
locomotive amongst all classes in a remarkable manner. Many of 
the numerous illustrations that embellish the book have been specially 
collected for the purpose, and several will be quite new to the majority 
of readers. Special pains have been taken to admit only such illus- 
trations the authenticity of which was known to the author, and 
for the same reason many otherwise interesting pictures, upon tne 
accuracy of which suspicion rested, were excluded from the collection. 

Despite these exclusions, we believe that no other book on loco- 
motive history in the English language is so fully illustrated. 

As it is proposed to deal with the railway locomotive only, 
it is not necessary to make more than a passing reference 
to the more or less crude proposals of Sir Isaac Newton, the Mar- 
quess of Worcester, Savery, Dr. Robinson, Leupold, and other 
writers and scientists, who hinted at the possibility of steam loco- 
motion. Nor does the writer propose to discuss the alleged use 
of railways and steam locomotives in Germany at a date prior to 
their general introduction into England. The claims of. Cugnot, 
Symington, Evans, Murdoch, and others as builders or designers of 
actual or model steam road locomotives will also be passed without 
discussion. 

We take this opportunity of expressing our sincere thanks to 
the locomotive superintendents of British railways, who have all 
been so willing to assist the author, not only in supplying accurate 
data concerning the locomotives of their own design, but also fo* so 
kindly revising the portions of the volume that relate to the loco- 
motive history of the particular railway with which each one of 
these gentlemen is connected. 

In conclusion, we leave the " Evolution of the Steam Locomotive " 
to the kindly consideration of our readers, hoping that from a 
perusal of it they may derive both information and pleasure. 

G. A. SEKON. 
December, 1898. 



CONTENTS. 



PAGE 

PREFACE iii. 

LIST OF ILLUSTRATIONS vi. 

CHAPTER 1 1 

IT 10 

III.- 28 

IV 40 

V 56 

VI 66 

VII. ..'. 82 

VIII 103 

IX 130 

X 156 

XI. 185 

XII 205 

XIII , 231 

XIV 260 

XV 294 

INDEX . 321 



LIST OF ILLUSTRATIONS. 



PAGK 

"90>" the latest type of Great Northern Railway express engine Frontispiece 

The First Railway Locomotive of which authentic particulars are known 3 

Locomotive built by Murray for Blenkinsopp's Railway 6 

Brunton's " Mechanical Traveller " Locomotive 

Hackworth's " Wylam Dilly," generally known as Hedley's "Puffing Billy" 11 

Hackworth's or Hedley's Second Design, used on the Wylam Rwy. in 1815 13 

Stephenson's Initial Driving Gear for Locomotives 15 

Stephenson and Dodd's Patent Engine, built in 1815 16 

Stephenson's Improved Engine, as altered, fitted with Steel Springs ... 17 

"Locomotion," the First QEngine to Run on a Public Railway 20 

The First Successful Locomotive, Hackworth's "Royal George" 23 

Hackworth's Blast Pipe in the "Royal George" 24 

Waste Steam Pipe in Stephenson's "Rocket" 25 

The "Novelty," entered by Braithwaite and Ericsson for the Rainhill Prize 29 

Hackworth's " Sanspareil," one of the Competitors at Rainhill 32 

Stephenson's " Rocket," the Winner of the Rainhill Prize of 500 35 

Winan's " Cycloped " Horse Locomotive 38 

Bury's Original "Liverpool," the First Engine with Inside Cvlinders, etc. 41 

The "Invicta," Canterbury and Whitstable Railway, 1830 ... 45 

The " Northumbrian," the Engine that Opened the Liverpool and Man- 
chester Rwy. 46 

Hackworth's "Globe" for the Stockton and Darlington Railway 43 

Stephenson's "Planet," Liverpool and Manchester Railway ... 49 

" Wilberforce," a Stockton and Darlington Railway Locomotive 53 

Galloway's "Caledonian," built for the Liverpool Manchester Rwy. in 1832 54 

Roberts's "Experiment," with Verticle Cylinders, Bell Cranks, etc. ... 57 

Hawthorn's " Comet," First Engine of the Newcastle & Carlisle Rwy., 1835 59 

"Sunbeam," built by Hawthorn for the Stockton and Darlington Railway 64 
The " Grasshopper," with 10ft. driving wheels, built by Mather, Dixon & 

Co., for the G.W. Rwy. ... 73 

The " Hurricane," with 10ft. driving wheels, a Broad-Guage Engine, built 

on Harrison's System 76 

The " Thunderer," a geared-up Broad-Guage Engine, built on Harrison's 

Plan ... ... 78 

Bury's Standard Passenger .Engine for the London and Birmingham Railway 33 

" Garnet," one of the First Engines of the London and Southampton Rwy. 85 

"Harpy," one of Gooch's "Firefly" Class of Broad- Gauge Engines 90 

Interior of Paddington Engine House, showing the Broad-Guage Locomo- 
tives of 1840 92 

" Jason," one of Gooch's First Type of Goods Engines for the G.W. Rwy. 93 

Paton & Millar's Tank Engine, for working on* the Cowlairs Incline, Glasgow 98 

Stephenson's "Long Boiler" Goods Engine, Eastern Counties Railway ... 104 

Gray's Prototype of the " Jenny Lind," No. 49, London & Brighton Rwy. 10* 

" Hero," a Great Western Railway Six-Coupled Broad- Gauge Goods Engine 106 

The "Great Western " Broad-Gauge Engine as originally Constructed ... 107 

The Original " Great Western," as Rebuilt with Two Pairs of Leading Wheels 1C9 

The "Namir," the First Engine built on Crampton's Principle 112 

Crampton's " London," First Engine with a Name, L. & N.W. Rwy. ... 113 

fP Great Britain," one of Gooch's Famous 8ft. "Singles," G.W. Rwy. ... 114 

"No. 61," London and Brighton Railway 115 

The "Jenny Lind," a Famous Locomotive, built by Wilson and Co. ... 119 
Trevithick's "Cornwall," with 8ft. 6in. Driving Wheels, and Boiler below 

the Driving Axle 120 



LIST OF ILLUSTRATIONS 

PAGE 

Trevithick's " Cornwall," as npw Running between Liverpool and Manchester 121 

" Old Copper Nob," No. 3, Furness Rwy., Oldest Locomotive now at work 1^3 

The "Albion," a Locomotive built on the "Cambrian" System 127 

The "Fairneld," Adams's Combinad Broad-Gauge Engine and Train ... 132 

The " Enfield," Combined Engine kind Train for the Eastern Counties Hallway 134 

1 lied Star," a 7ft. Single Broad-Gauge Saddle Tank Engine 136 

"No. 148," L. & N.W. Rwy.; Example of Stephenson's "Long Boiler" 

Engines ... ... ... ... ... ... ... ... ... ... 137 

Adams's "Light" Locomotive for the Londonderry and Enniskillen Railway 139 

England & Co.'s "Little England," Locomotive Exhibition, London, 1851 142 

Crampton's "Liverpool," London and Ncrth Western Railway 145 

Timothy Hackworth's " Sanspareil, No. 2" 149 

Caledonian Railway Engine, "No. 15" 153 

"Mac's Mangle," No. 227, London and North Western Railway 154 

" President," one of McConnell's " Bloomers," as originally built ... ... 155 

One of McConnell's "Bloomers," as Rebuilt by Ramsbottom 155 

The "Folkestone," a Locomotive on Crampton's System, built for the 

S.E.R., 1851 159 

One of J. V. Gooch's "Single" Tank Engines, Eastern Counties Railway ... 161 

" Ely," a Taff Vale Railway Engine, built in 1851 163 

McConnell's "300," London and North Western Railway 165 

Pasey's Compressed Air Locomotive, Tried on the E.G. Rwy., 1852 ... 170 
The First Type of Great Northern Railway Passenger Engine, one of the 

"Little Sharps" 171 

Sturrock's Masterpiece, the Famous {Great Northern Railway, "215" ... 172 

Pearson's 9ft. "Single" Tank Engine, Bristol and Exeter Railway 174 

One of Pearson's 9ft. " Single " Tanks, taken over by the Great Western 

Railway 176 

A Bristol and Exeter Railway Tank Engine, as Rebuilt (with Tender) 

by the G.W.R .' 178 

"Ovid," a South Devon Railway Saddle Tank Engine, with Leading Bogie 180 

"Plato," a Six-Coupled Saddle Tank Banking Engine, South Devon Railway 181 

The First Type of Narrow-Gauge Passenger Engines, Great Western Rwy. 182 
" Robin Hood," a Broad-Gauge Express Engine, with Coupled Wheels 7ft. 

in diameter , 183 

North British Railway Inspection Engine, No. 879 184 

The "Dane," L. and S.W.R., fitted with Beattie's Patent Apparatus for 

Burning Coal 187 

Cudworth's Sloping Fire Grate, for Burning Coal, as fitted to S.E.R. 

Locomotives 189 

" Nunthorpe," a Stockton and .Darlington Railway Passenger Engine, 1856 193 

Beattie's Four-Coupled Tank Engine, London & South Western Rwy., "1857 19; 
Sinclair's Outside Cylinder, Four-Coupled Goods Engine, Eastern Counties 

Railway (Rebuilt) 196 

Six-Coupled Mineral Engine, Taff Vale Railway, built 1860 202 

"Brougham," No. 160, Stockton and Darlington Railway 206 

Conner's 8ft. 2in. "Single" Engine, Caledonian Railway (Rebuilt) 208 

"Albion," Cambrian Railways, 1863 210 

A Great Northern Railway Engine, fitted with Sturrock's Patent Steam 

Tender 218 

Sinclair's Design of Tank Engine for the Eastern Counties Railway 219 

Beattie's Standard Goods Engine, London and South Western Railway, 1866 226 

Beattie's Goods Engine, London and South Western Railway (Rebuilt) ... 227 

Adams's Passenger Tank Engine, N.L. Rwy., as Rebuilt by Mr. Pryce ... 228 

Pryce's Six-Coupled Tank Goods Engine, North London Railway 229 

Locomotive and Travelling Crane, North London Railway ... ... ... 230 

"Python," a 7ft. lin. Coupled Express Engine, L. and S.W. Rwy. ... 232 

8ft. lin. "Single" Express Engine, Great Northern Railway 237 

"John Ramsbottom," one of Webbs "Precedent" Class, L. & N.W. Rwy. 238 

"Firefly," a London and South Western Outside Cylinder Tank Engine . . 239 
"Kensington," a Four-Coupled Passenger Engine, London, Brighton and 

South Coast Railway 240 

" Teutonic," a London and North Western jXailway " Compound " Loco- 
motive on Webb's System ... ... ... ... ... ... ... 244 

" Queen Empress," one of AY ebb's Compound Locomotives, L. & N.W. Rwy. 2JS 



LIST OF IL UST RATIONS 

PAGE 

"Black Prince," L. & N.W. Railway, a Four-Coupled Four-Cylinder Com- 
pound Engine ... 248 

Johnson's 7ft. 9in. "Single" Engine, Midland Railway 251 

"George A. Wallis," an Engine of the "Gladstone" Class, L., B. and 

S.C. Railway 252 

"1463," North Eastern Railway, one of the " Tennant " Locomotives ... 253 

Holmes' s Type of Express Engines for the North British Railway ... 254 
7ft. "Single" Engine, Great Eastern Railway, fitted with Holden's Liquid 

Fuel Apparatus 25b 

",No. 10," the Latest Type of Great Eastern Railway Express Engine, 

Fired with Liquid Fuel 258 

" Goldsmid," one of the new London, Brighton and South Coast Railway 

Express Passenger Engines 261 

"Inspector," London, Brighton and South Coast Railway .. 262 

"No. 192," a Standard Express Passenger Locomotive, L.C. & D.Rwy. ... 263 

Standard Express Passenger Engine, Cambrian Railways 264 

Standard Passenger Tank Engine, Cambrian Railways 265 

" No. 240," the S.E. Railway Engine that obtained the Gold Medal, Paris 

Exhibition, 1889 267 

Standard Goods Engine, South Eastern Railway 268 

Standard Passenger Tank Locomotive, South Eastern Railway 269 

Latest Type of Express Passenger Engine, South Eastern Railway 271 

Adams's Standard Express Engine, London and South Western Railway ... 273 
A " Windcutter " Locomotive, "No. 136," L. and S.W. Railway, fitted 

with Convex Smoke-Box Door 274 

Drummond's Four-Cylinder Engine, London and South Western Railway 275 

Four-Coupled Passenger Engine with Leading Bogie, North British Railway 277 

Holmes's Latest Type of Express Engine, North British Railway 279 

Four- Wheels-Coupled Saddle Tank Engine, London & North Western Rwy. 281 

Standard Express Passenger Locomotive, Lancashire and Yorkshire Railway 282 

Standard Eight- Wheel Passenger Tank Engine, Lancashire & Yorkshire Rwy. 233 

Oil-Fired Saddle Tank Shunting Engine, Lancashire rind Yorkshire Railway 284 

" Dunalastair," Caledonian Railway ... ... ... ... ... ... ... 285 

One of Mclntosh's "Dunalastair 2nd" Caledonian Express Locomotives ... 237 

Six- Wheels-Coupled Condensing Engine, Caledonian Railway ... ... ... 288 

" Carbrook," one of Drummond's Express Engines for the Caledonian Railway 289 

Mclntosh's 5ft. 9in. Condensing Tank Engine, Caledonian Railway ... ... 290 

"No. 143," Taff Vale Railway Tank Loccmotive, for working on incline ... 292 

A favourite Locomotive of the Isle of Wight Central Railway 293 

7ft. Sin. "Single" Convertible Engine, Great Western Railway 295 

"Empress of India," Standard G.W. 7ft. Sin. "Single " Express Locomotive 295 

" Gooch," a Four-Coupled Express Engine, Great Western Railway 297 

"Pendennis Castle," one of the Great Western "Hill Climbers" 298 

"Single" Express Engine, Six- Wheel Type, Great Western Railway ... .. 300 

6ft. 6in. Four-Coupled Passenger Locomotive, Great Western Railway ... 300 

6ft. Four- Coupled Passenger Engine, Great Western Railway ... 301 

" Barrington," New Type of Four-Coupled Engine, Great Western Railway 301 

Four-Coupled- in-Front Passenger Tank Engine, Great Western Railway ... 302 
"No, 1312," one of Mr. Ivafbt's (1073) Smaller Class of Four-Coupled 

Bogie Engines, Great Northern Railway 304 

The Latest Type of 6ft. 6in. Coupled Engine, Great Northern Railway .. 3C5 
Latest Type of G.N.R. Exrvoia Locomotive; 7ft. 6in. "Single," with Inside 

Cylinders, etc. 308 

"No. 100," one of the " T " Class Four-Coupled Passenger Engines, Great 

North of Scotland Kail way ill 

Pettigrew's New Goods Engine for the Furness Railway 315 

Six-Wheels-Coupled Bogie Engine, with Outside Cylinders, Highland Railway 316 

Liquid Fuel Engine, Belfast and Northern. Counties Railway ... ... .. 317 

"Jubilee," Four-Wheels-Coupled Compound Locomotive, Belfast and 

Northern Conties Railway 518 

"No. 73," Standard Passenger Engine, Great Northern Railway (Ireland)... 318 

Four-Coupled Bogie Express Engine, Great Southern and Western Railway 319 

'' Peake," a Locomotive of the Cork and Muskerry Light Railway .. ... 319 



EVOLUTION 



OF THE 



STEAM LOCOMOTIVE 



CHAPTER I. 

Trevituick's triumph; his first steam locomotives Mistaken for the dovil The 
Coalbrookdale engine >A successful railway journey at Merthyr Tydvil 
Description of the engine " Catch-me-who-can '' The locomotive in 
London Blenkinsopp's rack locomotive Chapman's engine Did Chapman 
build an eight-wheel locomotive? Brimton's ''steam horse' 1 Its tragic end. 

Richard Trevithick, the Cornish mine captain and engineer, 
belongs the honour of producing the first locomotive true, his 
original essay was a road locomotive. As long ago as 1796 he 
constructed a model locomotive which ran round a room ; and 
on Christmas Eve, 1801, he made the initial trip with his first steam 
locomotive through the streets of Camborne. This machine carried 
several passengers at a speed in excess of the usual walking pace of a 
man. Trevithick was joined in the enterprise by his 

cousin Vivian, who provided the money to build the steam 
engines, and to patent them, their first patent being dated 
24th March, 1802. It is described as "for improving the 
construction of steam engines, and the application thereof for 
drawing carriages on rails a>nd turnpike roads and other purposes." It 
was claimed that their engine would produce " a more equable rotarjr 



2 EVOLUTION OF THE STEAM LOCOMOTIVE 

motion on the several parts of the revolution of any axis which 
is moved by the steam engine, by causing the piston rods of tv, r o 
cylinder** to work on the said axis by means of cranks, at a quarter 
turn asunder." 

Among other improvements claimed in the specification, mention 
should be made of the return-tube boiler, bellows to urge the fire r 
and a second safety valve, not under the control of the driver. 

A steam carriage with these improvements was constructed, and 
Vivian and Trevithick commenced a journey on it from Camborne 
to Plymouth, from which port it was shipped to London. On tbo 
road to Plymouth a closed toll-bar was met, and the steam carriage 
stopped for the gate to be opened. " What have us got to pay here 1 " 
demanded Vivian. The affrighted toll-keeper, shaking in every limb, 
and his teeth chattering, essayed to answer, and at last said, 
" Ni>. na na na." " What have us got to pay, I say 1 " demanded 
Vivian. ''Na noth nothing to pay, my de dear Mr. Devil; do 
drive on as fast as you can. Nothing to pay." 

It must be remembered that to Cornishmen of a century ago 
the devil was a very real personage ; and, seeing the horseless 
carriage proceeding with a fiery accompaniment, the poor toll-keeper 
thought he had at last seen his Satanic majesty. He also appears 
to have remembered that it is well "to be civil to everyone, the 
devil included; there is no knowing when you may require his good 
wishes." Hence the toll-keeper's reason for calling Vivian "my 
dear Mr. Devil." 

As? early as August, 1802, R. Trevithick (according to his life, 
as written by his son, F. Trevithick) appears to have constructed a 
railway locomotive at Coalbrookdale. This engine had a boiler of 
cast-iron l^in. thick, with an interior return wrought-iron tube. 
The length of the boiler was 6ft., and the diameter 4ft. The cylinder 
working this engine was 7in. in diameter, the stroke being 3ft. The 
next railway locomotive was that constructed for the Pen-y-darren 
Tramroad near Myrthyr Tydvil. Of this particular locomotive (Fig. 1) 
H ir possible to obtain authentic particulars, although much that is 
legendary already clusters around this historic locomotive. For 
instance, we read that the locomotive in question had a brick chimney, 
and that it was demolished by colliding with an overhanging branch 
of a tree. Then the amount of the bet between Mr, Homfray, the 
owner of the tramroad, and his friend, as to whether the locomotive 
would successfully perform a journey from Pen-y-darren to the navi- 



EVOLUTION U1>' THE STEAM LOCOMOTIVE 



.3 



gallon at Plymouth, is a variable quantity. The amount staked hi3 
been stated to be 500 a side, and also 1,000 a side. 

It is evident that some days prior to February 10th, 1804, the 
engine successfully performed the journey, and that overhanging trees 
and rocks considerably impeded the travelling, several stoppages having 
to be made whilst these obstacles were removed. Mr. Homfray, how- 
ever, won the bet. On February 21st another trip was made by the 
locomotive. On this occasion the load consisted of 5 wagons, 10 tons 




FIG. 1. THE FIRST RAILWAY LOCOMOTIVE OF WHICH AUTHENTIC 
PARTICULARS ARE KKOWN 



of bar iron, and 70 passengers, the weight of the engine, with 
water and fuel, being 5 tons; the journey of nine miles being per- 
formed in 4 hours 5 minutes, including several stoppages ; the 
average speed when travelling being five miles an hour. On the 
return journey the engine hauled the empty wagons up an incline 
of 1 in 18 at the rate of five miles an hour. Several of the tram- 
plates, which weighed only 281b. per yard, were broken on the 
downward trip. Early in March the engine conveyed a load of 
25 tons from the iron-works to the navigation. 

B J 



4 EVOLUTION OF THE STEAM LOCOMOTIVE 

It will be observed that this engine from the first decided the 
practicability of conveying loads by means of smooth wheels on 
smooth rails, simply by adhesion. Yet, strange to say, for 
several years after, it was the firmly-fixed belief of succeeding locomo- 
tive constructors that it was impossible to obtain sufficient adhesion 
between a smooth surface and a smooth rail to successfully work a 
locomotive. The result was the invention of many curious method** 
to overcome this apparent difficulty, which, as a fact, never existed, 
save in the minds of the designers of the early locomotives. These 
men do not seem to have been fully acquainted with the results 
of Trevithick's experiments on the Pen-y-darren tramroad in 1804. 

A description of this locomotive prototype is of interest. The 
boiler was cylindrical, with a flat end. The fire-door and chimney 
were both at the same end, an extended heating surface being 
obtained by means of the return tube j above the fire-door was 
the single horizontal cylinder, the diameter of which was 8Jin. ; 
a considerable portion of the cylinder was immersed in the boiler, 
the exposed portion being surrounded by a steam jacket. The 
stroke was 4ft. Gin. ! The piston-rod worked on a motion frame 
extending in front of the engine. At the other end of the boiler 
was a fly-wheel some 9ft. Gin. in diameter, the motion being con- 
veyed to it by connecting rods from the cross-head; a cog-wheel 
on the fly-wheel axle conveyed the motion by means of an inter- 
mediate wheel to the four driving-wheels, which are stated to 
have been 4ft. Gin. in diameter. The exhaust steam appears to 
have been turned into the chimney, not for the purpose of a blast, 
but only as an easy method of getting rid of the vapour. It will 
be remembered that Trevithick, in his patent specification, specially 
mentioned bellows for urging the fire, and was, therefore, not 
acquainted with the nature of the exhaust steam blast. It is important 
to bear this in mind, as the reader will find in a later chapter. Thh 
engine is stated to have blown up through not being provided with a 
safety valve, though Trevithick specially ordered one to be fixed 
to the boiler, but his instructions do not appear to have been 
carried out. 

Trevithick made another locomotive, called " Catch-me-who-can." 
This ran on an ellipse-shaped railway specially laid down for it at 
Euston Square, London, and was visited by many people during the 
few days it wap on view. Another locomotive was constructed from 
the drawingb of Trevithick's Coalbrookdale locomotive of 1802, to 



EVOLUTION OF THE STEAM LOCOMOTIVE 5 

the orders of Mr. Blackett, the owner of Wylam Collieries. This 
engine weighing 4^ tons, had a single cylinder Tin. diameter, 
3ft. stroke, and, of course, a fly-wheel. For some reason or another 
this engine does not appear to have been used on the Wylam tram- 
road, but was used in a Newcastle foundry to blow a cupola. Mr. 
Armstrong, a former Locomotive* Superintendent of the Great Western 
Railway, was acquainted with this engine of Trevit hick's at the time 
it was so employed at Newcastle. 

Having given an outline of Trevithick's invention of the tram- 
road locomotive, and the other locomotive engines designed by him, 
we will deal with the locomotive built for J. Blenkinsopp (Fig. 2), of the 
Middleton Colliery, near Leeds, who, on April 10th, 1811, obtained 
a patent for a self-propelling steam engine, worked by means of a 
cog-wheel, engaging in a rack laid side by side with one of the rails 
forming the tramway. 

The erroneous idea that the locomotive of itself had not sufficient 
adhesion between the smooth wheel and the surface of the rail to 
propel itself and draw a load was strongly entertained by Blenkin- 
sopp, hence his patent rack and pinion system. Blenkinsopp having 
this opinion, which he published by means of his patent specification, 
caused succeeding inventors to fall into the same error regarding 
the adhesive properties of the locomotive, and consequently con- 
siderably retarded the development of the railway engine. 

Although this engine is generally known as Blenkinsopp's, it 
was constructed by Matthew Murray, the Leeds engineer. The 
boiler was cylindrical, with slightly convex ends, a single flue ran 
through it, which was in front turned upwards, and SD formed tlie 
chimney; the fire-grate was at the other end of the flue, as in the 
modern locomotive. 

This engine was provided with two cylinders, and was, in this 
respect, an improvement on Trevithick's single-cylinder engines. Tho 
cylinders were 8in. in diameter, and placed vertically, the major 
portion of them being placed within the boiler. The stroke was 
20in., and the motion was conveyed by means of cross-heads, working 
connecting-rods; these came down to two cranks on either side 
below the boiler. The cranks worked two shafts fixed across the 
frames, on which were toothed wheels, both working into a centre 
toothed wheel, which was provided with large teeth, these engaged 
on the rack rail previously described. The cranks were set at right 
angles, so that one piston was exerting power when the other 



6 EVOLUTION OF THE STEAM LOCOMOTIVE 

was at its dead centre, and vice versa. The engine was supported 
on the rails 'by four wheels 3ft, Gin. in diameter. The two cylinders 
were connected by a pipe which conveyed the exhaust steam and 
discharged it into the atmosphere through a vertical tube. The 
engine weighed 5 tons, burned 751b. of coal per hour, and evaporated 




FIG. 2. LOCOMOTIVE BUILT BY MURRAY FOR BLENKINSOPP'S RAILWAY 

50 gallons of water in the same time. This locomotive could haul 
94 tons on the level at 3J miles an hour, or 15 tons up an incline 
of 1 in 15 ; its maximum speed was 10 miles an hour.. The engine 
cost 400 to construct, and worked from August, 1812, for a period 
of about 20 years, and in 1816 the Grand Duke Nicholas, afterwards 
Emperor of Kussia, inspected the machine. The tramway on which it 
worked was about 3| miles long. 

In September, 1813, Murray supplied two of Blenkinsopp's engines 
to the Kenton Colliery. 

On December 30th, 1812, a patent was granted to William and 
Edward Chapman for a method of locomotion. A chain was stretched 
along the railway and fastened at each end; connected to the 



EVOLUTION OF THE STEAM LOCOMOTIVE 7 

locomotive by spur gear was a barrel, around which the chain 
was passed. When the barrel rotated, the chain was wound over 
it, and since the chain was secured at either end, the engine 
was of necessity propelled. An engine on this principle was tried 
on the Heaton Colliery Tramroad, near Newcastle-on-Tyne. The 
machine was supported on wheels travelling on the rails. The boiler 
was of Trevithick's design, and fanners were used to excite the 
combustion of the fuel. The weight of Chapman's engine was 6 tons. 
After e, few trials the scheme was abandoned, as it was found im- 
practicable to successfully work such a system. Every eight or ten 
yards the chain was secured by means of vertical forks, which held 
it when disengaged from the drum of the locomotive. 

By this method the pressure of one engine on the chain was 
limited to the fork on either side of the drum instead of being spread 
over the whole length of the chain, and it would, therefore, hove 
been possible for several engines to have used the chain at one and 
the same time. 

According to Luke Herbert and Lieut. Lecount, Chapman also 
built an 8-wheel locomotive for the Larnbton Colliery. This engine, 
it was stated, had vertical cylinders, and the motion was conveyed 
by means of spur wheels. It weighed 6 tons loaded, and drew 
18 loaded wagons, of a gross weight of 54 tons, from the colliery 
to the shipping place on the Wear; with the above load it attained 
a speed of four miles an hour up an incline of 1 in 115. The 
dimensions and capabilities accredited to this engine appear suspici- 
ously similar to those related of the first Wylam locomotive. 

On May 22nd, 1813, Mr. W. Brunton, of the Butterfly Ironworks, 
obtained a patent for a novel method of steam locomotion. This 
locomotive inventor was also suffering from the common belief that it 
was -impossible to obtain sufficient adhesion between a smooth rail 
and smooth wheels, despite the successes that had already been 
obtained in this direction by Trevithick. He therefore built an 
engine supported on four flanged carrying wheels, but propelled 
from behind by means of two legs. Indeed, another inventor con- 
sidered the idea of steam legs so natural that he constructed a steam 
road coach that was to be propelled by four legs, one pair partaking 
of the character and motion of the forelegs of a horse, and the other 
pair being fashioned on the model of the hind legs of the same 
quadruped. 

In. Bruuton's leg-propelled steam locomotive (Fig. 3) we find 



8 EVOLUTION OF THE STEAM LOCOMOTIVE 

that the boiler was cylindrical, with a single horizontal tube passing 
through it, and turned up in front in a vertical position, thus 
forming the chimney. The motion was obtained from a single 
horizontal cylinder, fixed near the top of the Boiler, the piston rod 
projecting behind; the end of the piston rod was attached to a 
jointed rod, the bottom portion of which formed one of the legs. 
The upper portion of this rod was attached to a framework fixed 
above the boiler of the engine, which formed a fulcrum, and then 
by an ingenious arrangement of levers, an alternate motion was 
given to the second leg. Each leg had a foot formed of two prongs 
at the bottom; these stuck in the ground, and prevented the legs 




FIG. 3. BRUXTON'S "MECHANICAL TRAVELLER" LOCOMOTIVE 

from slipping. Upon steam being applied, the piston in the ordinary 
way would have travelled to the end of the cylinder, but the leg, 
having a firm hold of the ground, presented a greater resistance 
to the steam than did the weight of the engine, so the steam acting 
on the surface that presented the lesser resistance, caused the 
cylinder to recede, and with it the engine, to which it was, of 
course, firmly attached. By means of the reciprocating levers, 
a horizontal rod travelled on the top of the boiler and 



EVOLUTIOX OF THE STEAM LOCOMOTIVE 9 

over a cog-wheel; then 011 the other side of this cog-wheel 
was another horizontal rod, which, actuated by the cog-wheel, 
travelled in a contrary direction, and being attached to the other leg 
of the engine, as the machine receded from the first leg, it drew 
the second leg close up to the back of the engine. The 
second leg was now ready to propel the engine, which it did upon the 
steam being applied to the other side of the piston, and the process 
was alternated with each admission of steam to the front or back 
of the piston. 

Whilst the legs were returning towards the engine the feet were 
raised by means of straps or ropes fastened to the legs and passing 
over friction-wheels, movable in one direction only by a ratchet 
and cafch, and worked by the motion of the engine. 

Brunton called his locomotive a "mechanical traveller," and 
stated that the boiler was of wrought-iron, 5ft. Gin. long and 3ft. dia- 
meter, weighing 2J tons, stroke of piston, 2ft., and at 2J miles 
per hour, with a steam pressure of 451b. per square inch, was equal 
in power to nearly six horses. This locomotive curiosity blew up 
at Newbottle in 1816, and about a dozen people were thereby either 
killed or seriously injured. 



CHAPTER II. 

Who is entitled to the honour of constructing the Wylam locomotives? The 
claims of Hackworth, Hedley und Foster " Puffing Billy "Rebuilt as an 
eight-wheel <ngine Stewart's locomotive Sharp practice causes Stewart to 
abandon locomotive building George Stephenson as a locomotive builder 
His hazy views as to his first engine " Blucher " The German General 
proves a failure Steo'lieiison and Dodd's engine Stephenson's third engine, 
with (so-called) steam springs Competent critics condemn Stephenson's 
engines The " Koyal William " The "Locomotion " Hackworth, General 
Manager of the Stockton end Darlington Railway Horse haulage cheaper 
than Stephenson's locomotives Hackworth to the rescue The " Royal 
George," the first successful locomotive The "exhaust" stoam blast- 
Rival olaim-mts raid its invention Locomotive versus stationary engine- - 
" Twin Sisters " " Lancashire Witch " " Agenoria " The " Maniac," 
" a Forth Street production." 

WE have now arrived at a point in the evolution of the steam 
locomotive where the claims of several men are in competition. The 
facts as to the experiments and construction of the engines at Wylam 
are not disputed. The question at issue is as to whom the honour 
of the success should be given. Christopher Blackett, of the Wylam 
Colliery, as previously stated, ordered a locomotive of Trevithick, 
but never used it. He, however, determined to make a trial of 
steam haulage on his plate way, and in 1811 some kind of experi- 
ments were made, having in view the above-mentioned object. At 
this time Timothy Hackworth was foreman of the smiths (he would 
now be called an engineer), and William Hedley was coal-viewer at 
Wylam. The friends of both Hackworth and Hedley claim for their 
respective heroes the honour of these early essays in locomotive 
construction. But it is probable the honours should be shared by 
both, as well as by Jonathan Foster, who also assisted in the experi- 
ments and construction of the Wylam locomotives. 

Hedley was colliery-viewer at Wylam, and therefore, most likely, 
Hackworth was, to an extent, under his orders, and probably had to 
defer to, and act under, the instructions of Hedley. 

But Hackworth's position as foreman-smith did not preclude him 
from making suggestions and introducing improvements of his own 
into the locomotives under construction. 

It is stated that Hedley was jealous because Hackworth obtained 



EVOLUTION OF THE STEAM LOCOMOTIVE 



11 



the praise for building the Wylam locomotives (or " Timothy's Dillies,'' 
as they were locally called), and to force Hackworth to leave Wylam, 
Hedley required him to do some repairs to the machinery on Sundays. 
Now, Timothy was a fervent Wesleyan, and spent his Sundays in 
local preaching, so he naturally refused to violate his conscience by 
working on that day. Consequently Hackworth sought employment 
elsewhere. 




FlQ. 4. HACKWOKTH'S "WYLAM DILLY," GENERALLY KNOWN AS 
HEDLEY'S "PUFFING BILLY" 

On the other hand, it was a sore point with Hackworth that George 
Stephenson. spent his Sundays at Wylam taking sketches and par- 
ticulars of the locomotives at that time at work on the Wylam Rail- 
way, the result of which observations was apparent in the locomotive 
built by Stephenson at Killingworth in 1814. 

The Wylam experimentalists in October, 1812, constructed a four- 
wheel vehicle driven by manual power working cranks connected 
with spur wheels. The carriage was loaded until sufficient weight 
had been placed upon it to cause the wheels to turn round without 



12 EVOLUTION OF THE STEAM LOCOMOTIVE 

The experiment, however, satisfied Mr. Blackett that locomotive 
engines with smooth wheels could be employed in drawing loads on 
his tramroad; and the construction of an engine was immediately 
proceeded with. This was completed and put to work early in 1813. 
It had a . cast-iron boiler, . and a single internal flue; the solitary 
cylinder was Gin. in diameter, and a fly-wheel was employed after 
the model of Trevithick's engine. The steam pressure was 501b. 
This four-wheel engine drew six coal trucks at five miles an hour, and, 
therefore, did the work of three horses not a very powerful example 
of a steam locomotive, it will be observed. This engine being some- 
what of a failure, it was decided to build another, and one- 
with a wrought-iron boiler and a return tube was constructed. In 
his engine (Fig. 4) it will be noticed the fire-box and chimney were both 
at the same end of the 'boiler. Two vertical cylinders were fixed over 
the trailing wheels of "Puffing Billy" (for it is this historical loco- 
motive, no'W preserved in the South Kensington Museum, that is 
now being described). The piston rods were connected to beams of 
the ".Grasshopper" pattern, being both centred at the funnel end *.i 
the engine. The driving rods were connected with these beams 
at about their centres, and passed down to spur wheels, which, by 
means of toothed wheels on either side, communicated the motion, 
to the four carrying wheels. The spent steam was conveyed from 
the cylinders to the chimney by means of two horizontal pipes laid 
along the top of the boiler. It was soon discovered that the cast- 
iron tram-plates, which were only of four square inch section, were 
unable to bear the weight of " Puffing Billy," and another change was 
decided upon. 

The engine was therefore placed on two four-wheel trucks (Fig. 5), so 
that the weight was distributed on eight instead of four wheels, the same 
method of spur gearing was employed, and the whole of the wheels 
were actuated by means of intermediate cog-wheels. To prevent, as 
far as possible, the noise caused by the escaping steam, a vertical 
cylinder was fixed on the top of the boiler between the cylinders and 
the funnel. Into this chamber the spent steam was discharged, and 
from it the same was allowed to escape gradually into the chimney. 
In addition to the improvement of a return tube, with its extended 
heating surface, with which this class of engine was provided, the 
funnel was only 12in. in diameter, as compared with 22in. diameter 
as used by Stephenson in his early engines. As already stated, the 



EVOLUTION OF THE STEAM LOCOMOTIVE 



13 



Wylarn locomotives were locally called "Timothy's Billies," after 
Timothy Hackworth, to TV hose inventive genius they were popularly 
ascribed. In 1830, the cast-iron plates on the road from Wylam to 
Leamington were removed, and the course was relaid with edge rails, 
so that the necessity for eight-wheel engines was at an end. 
" Timothy's Dillies " were then reconverted to four-wheel locomotives, 
and continued at work on the line till about 1862. 

Not many locomotive writers are acquainted with the fact tha" 




FIG. 5 HACKWORTH'S Oil HEDLEY'S SECOND DESIGN, AS USED ON 
THE WYLAM RAILWAY IN 1815 

in 1814 William Stewart, of Newport, Mon., constructed a locomotive 
for the Park End Colliery Company, which was tried on the Lydney 
Railway, and found to work in a satisfactory manner. The Park End 
Colliery Company were paying about 3,000 a year to contractors 
for horse haulage of their coal to the Forest of Dean Canal, and 
Stewart undertook to do the same by locomotive power for half that 
sum. The Company accepted his teraisi, and he set about the 



14 EVOLUTION OF THE STEAM LOCOMOTIVE 

construction of his engine. Whilst this was progressing the contrac- 
tors who provided the horses were told at each monthly settlement that 
tho Company were going to use a locomotive to, haul the coal, as 
horse-power was too expensive. By means of these threats the con- 
tractors were induced each month to accept a less price than pre- 
viously for "leading" the coal over the tramroad. U~on the specified 
date Stewart's locomotive was duly delivered on the line, and ac- 
cepted by the Park End Colliery Company for doing the work 
required; but the engineer was informed that the horse-power con- 
tractors were then only receiving 2,000 a year for the work, and 
that as Stewart had agreed to provide locomotive power at one-half 
of the sum paid for horses, he would only receive 1,000 a year. 

Stewart was so highly indignant at this piece of sharp practice 
that he refused to have anythin^ftirtner to do with the Park End 
Colliery Company, and at once removed his locomotive off their 
tramroad, and took it back to Newport. 

The earliest attempts of George Stephenson in connec- 
tion with the evolution of the steam locomotive now deserve 
attention. Stephenson himself is not very clear about his 
first engine, for, speaking at Newcastle at the opening of 
the Newcastle and Darlington Railway in 181-1, lie said that 
thirty-two years ago he constructed his first engine. "We called the 
engine 'My Lord/ after Lord Ravensworth, who provided the money 
for its construction." Both these statements are erroneous, for 
Stephenson did not build his first engine till 1814, and thirty-two 
years before 1844 would have been 1812. Then the engine could not 
have been called "My Lord," after Lord Ravensworth, for he title 
did not exist in 1814, the gentleman alluded to being only Sir 
Thomas Liddell till the coronation of King George IV. in 1821, when 
he was created Lord Ravens\vorth. 

The "Blucher," as this engine was in fact usually called, was first 
tried on the Killing-worth Railway on July 25th, 1814 ; she had a 
wrought-iron boiler, 8ft. long and 2ft. lOin. diameter, with a single 
flue 20in. diameter, turned up in front to form a chimney. The 
power was applied by means of two vertical cylinders located partly 
within the boiler, and projecting from its top, close together, and 
near the middle. The cylinders were Sin. diameter, the stroke 2ft. 
The motion was conveyed to the wheels by means of cross-heads and 
connecting-rods working on small spur wheels (Fig. 6), which engaged 
the four carrying wheels by means of cogged wheels fitted on the axles of 



EVOLUTION OF THE STEAM LOCOMOTIVE 15 

the flanged rail-wheels ; these were only 3ft. in diameter, and were 3ft. 
apart. The spur wheels engaged another cogged wheel, placed b 
tween them, for the purpose of keeping the cranks at right angles, 
No springs were provided for the engine, which was mounted on a 
wooden frame, but the water barrel was fixed to one end of a lever, 
and also weighted; the other end of this lever was fixed to the 
frame of the engine. This arrangement did duty for springs ! 




FIG. 6. STEPHENSOX'S INITIAL DRIVING GEAE FOR LOCOMOTIVES 

The best work done by " Blucher " was the hauling of loaded coal- 
wagons, weighing 30 tons, up an incline of 1 in 450, at about four 
miles an hour. This first effort of Stephenson had no original points 
about it ; the method of working was copied from the Wylam engines, 
whilst Trevithick's practice was followed with regard to the position 
of the cylinders i.e., their location, partly within the boiler. The 
average speed did not exceed three miles an hour, and after twelve 
months' working the machine was found to be more expensive thai? 
the horses it was designed to replace at a less cost. The absence of 
springs was specially manifested, for by this time the engine was so 
much, shaken and injured by the vibration that the Killingworth 
Colliery owners were called upon a second time to find the money to 
enable Stephenson to construct another locomotive. 

The second engine (Fig. 7) constructed by George Stephenson was 
built under the patent granted to Dodd and Stephenson on 28th Febru- 
ary, 1815. In this engine vertical cylinders, partly encased in the boiler, 
were again employed ; but their position was altered, one being placed 
at each extremity of the boiler over the wheels, the intermediate spur 
wheels formerly used for keeping the cranks at right angles were 



16 EVOLUTION OF THE STEAM LOCOMOTIVE 

abandoned, and the axles were cranked. A connecting-rod was fitted 
on these cranks, thus coupling the two axles. To give greater 
adhesion, the wheels of the tender were connected with those of the 
engine by means of an endless chain passing over cogs on the one 
pair of engine wheels, and over the adjoining pair of tender wheels ; 
by these methods six pairs of wheels were coupled. The mechanics 
engaged were not, however, capable of forging proper crank axles, 
and these had to be abandoned, and an endless chain coupling em- 
ployed for the engine wheels, similar to the one connecting the tender 
and engine, as previously described. 




FIG. 7. STEPHENSON AND DODD'S PATENT ENGINE, BUILT IN 1815 

This engine had no springs, and, to avoid excessive friction arising 
from the bad state of the tramroad, Stephenson employed " ball and 
socket" joints between the ends of the cross-heads and the con- 
necting-rods. In this way the necessary parallelism between the 
ends of the cross-heads and the axles was maintained. The spent 
steam in the engine was turned into the chimney, as in Trevithick's 
Pen-y-darren locomotive. This locomotive commenced to work on 
6th March, 1815. 

George Stephenson constructed a third engine (Fig. 8), under a 
patent granted to Lock and Stephenson on 30th September, 1816 ; this 
patent covered several matters, the most important in connection with 
the engine; being malleable iron wheels, instead of cast-iron, and what 
lias been described as " steam springs." The patentees called them 



EVOLUTION OF THE STEAM LOCOMOTIVE 



17 



'" floating pistons" ; of this description Colburn says emphatically 
" they are not," and the same authority continues, " and they (Lock 
and Stephenson) added, evidently without understanding the true 
action of the pistons, which were different in principle from the 
action of springs, that inasmuch as they acted upon an elastic fluid, 
they produced the desired effect, with much more accuracy than could 




FIG. 8. STEPHEXSOX'S IMPROVED ENGINE, AS ALTERED, FITTED \\ITH 
STEL SPRINGS (INVENTED BY NICHOLAS WOOD) 

be obtained by employing the finest springs of steel to suspend the 
engine. The whole arrangement was, on the contraiy, defective in 
principle and objectionable on the score of leakage, wear, etc. ; and, 
as a matter of course, was ultimately abandoned." 

In the drawings attached to the patent specification this engine 
is shown with six wheels, and the chain coupling is employed. Lecount 
says : " The six wheels were continued in use as long as the steam 
springs were applied, and when steel springs were adopted they were 
again reduced to four." So much praise has been given to Stephen- 
son for th,e "great improvements" he is supposed to have introduced 



18 EVOLUTION OF THE STEAM LOCOMOTIVE 

into the construction of the locomotive, that it will not be uninter sit- 
ing if we here reproduce the extremely pertinent remarks of Galloway, 
the well-known authority on the steam engine, which go far to prove 
that it was only the great success obtained by George Stephenson 
"from the construction of the Liverpool and Manchester and other 
railways, that caused historians and biographers to either magnify his 
locomotive successes, or to gloss over the evident faults in the design 
and construction of his engines. In his "History of tae Ste:,m 
Engine," published in 1827, Galloway says : These locomotive engines 
have been long in use at Killingworth Colliery, near Newcastle, and at 
Hilton Colliery on the Wear, so that their advantages and defects 
havo been sufficiently submitted to the test of experiment; and it 
appears that, notwithstanding the great exertions on the part of the 
inventor, Mr. Stephenson, to bring them into use on the different 
railroads, now either constructing or in agitation, it has been the 
opinion of several able engineers that they do not possess those 
advantages which the inventor had anticipated ; indeed, there can- 
not be a better proof of the doubt entertained regarding their utility 
than the fact that it has been determined that no locomotive engine 
shall be used on the projected railroad between Newcastle and Car- 
lisle, since, had their advantages been very apparent, the persons 
living immediately on the spot in which they are used, namely, 
Newcastle, would be acquainted therewith. 

"The principal objections seem to be the difficulty of sur- 
mounting even the slightest ascent, for it has been found that a rise 
of only one-eighth of an inch in a yard, or of eighteen feet in a mile, 
retards the speed of one of these engines in a very great degree ; 
so much so, indeed, that it has been considered necessary, in some 
parts where used, to aid their ascent with their load, by fixed 
engines, which drag them forward by means of ropes coiling round a 
drum. The spring steam cylinders below the boiler were found \ery 
defective, for in the ascending stroke of tne working piston they 
were forced inwards by the connecting-rod pulling at the wheel and 
turning it round, and in the descending stroke the same pistons were 
forced as much outwards. This motion or play rendered it necessary 
to increase the length of the working cylinder as much as there was 
play in the lower ones, to avoid the danger of breaking or seriously 
injuring the top and bottom of the former by the striking of the 
piston when it was forced too much up or down." 



EVOLUTIOX OF THE STEAM LOCOMOTIVE 19 

Stephenson must have felt himself to be a personage of somo 
importance when he received an order from the Duke of Portland for 
a steam locomotive. The engine, which had six wheels, was duly 
built and delivered in 1817, when it was put to work on the tramroad 
connecting the Duke's Kilmarnock Collieries with the harbour at 
Troon; but, after a short trial, its use was abandoned, as the weight 
of the engine frequently broke the cast-iron tram-plates. It has 
been stated that "this engine afterwards worked on the Gloucester 
and Cheltenham Tramroad until 1839, when the Birmingham and 
Gloucester Railway bought the line, and took up the cast-iron tram- 
plates." 

There is no doubt 'that a six-wheel engine with vertical cylinders 
partly encased in the top of the boiler, and called the " Royal William," 
was actually at work on this line the fact having been commemorated 
by the striking of a bronze medal ; but there is nothing to show that 
the " Royal William " and the engine built for the Kilmarnock and 
Troon Tramroad were one and the same locomotive ; whilst it is 
certain that the Gloucester and Cheltenham Tramroad was not 
purchased by the Birmingham and Gloucester Railway, but jointly 
by the Cheltenham and Great Western Union Railway and the 
Birmingham and Gloucester Railway, the price paid being 35,000. 

It would appear from a letter written by George Stephenson, and 
dated Killihgworth Colliery, 28th June, 1821, that he had but little 
idea to what a great degree the development of the steam loco- 
motive would be carried. The letter, which was addressed to Robert 
Stevenson, the celebrated Edinburgh engineer, proceeded as follows : 
" I have lately started a new locomotive engine with some improve- 
ments on the others which you saw. It has far surpassed my expecta- 
tions. I am confident that a railway on which my engine can work 
is far superior to a canal. On a long and favourable railway I would 
start my engine to travel 60 miles a day, with from 40 to 60 tons 
of goods." Taking Stephenson's "day" to mean twelve working 
hours, his idea of maximum speed did not exceed five miles an hour 
at that time. Before this in December, 1824 Charles MacLareu 
had published in the Scotsman his opinion that by the use of the 
steam locomotive lt we shall be carried at the rate of 400 miles a day," 
or an average speed of 33 1-3 miles an hour. 

Yet such is the irony of fate, that MacLaren, the true prophet, is 
forgotten, and George Stephenson is everywhere extolled. 

The Hetton (Coal) Railway was opened on November 18th, 1822, 



20 



EVOLUTION OF THE STEAM LOCOMOTIVE 



and five of Stephenson's "improved Killing-worth. " locomotives were 
placed upon the level portions. These engines were capable of hauling 
a train of about 64 tons, the maximum speed being four miles an 
hour. 




The Stockton and Darlington Railway, the first public railway, 
was cpaned on September 27th, 1825. The "Locomotion" (Fisr. 9) waj 
the first engine on the line. It was constructed at the Foitli Street 
Works of R. Stephenson and Co., at Newcastle-on-Tyne. At thi* 



EVOLUTION OF THE STEAM LOCOMOTIVE 21 

early peiiod these now celebrated Forth Street Works were little bettar 
than a collection of smiths' forges. 

Timothy Hackworth had been manager of these works, and he 
had a good deal to do with the construction of " Locomotion." His 
improvement of the coupling-rods in place of the endless chain pre- 
viously used for the purpose by Stephenson is worthy of passing 
notice. George Stephenson expressed a very strong desire that Hack- 
worth should remain in charge of the Forth Street Works, and 
went so far as to offer him one-half of his (Stephenson's) share in 
the business if he would remain. Hackworth agreed to do so if his 
name were added to that of the firm and he were publicly recognised 
as a partner; but this proposition was not accepted by Stephenson. 

Hackworth then took premises in Newcastle, and intended to 
commence business as an engine-builder on his own account, he 
having already received several orders from the collieries, etc., where 
his skill was well known and appreciated. George Stephenson, 
having heard of Hackworth's plans for carrying on a rival engine 
factory at Newcastle, saw Hackworth, and persuaded him to relin- 
quish the proposition and accept the office of general manager and 
engineer to the Stockton and Darlington Railway. 

Hackworth commenced these duties in June, 1825, and removed to 
Darlington. The " Locomotion" had four coupled wheels, 4ft. in 
diameter; two vertical cylinders, lOin. in diameter, placed partly 
within the boiler ; the stroke was 24in. ; steam pressure, 251b. pet 
square inch ; weight in working order, 6 J tons. The tender was 
of wood, with a coal capacity of three-quarter ton, and a sheet-iron 
tank holding 240 gallons; weight loaded, 2J tons. The tender 
was supported on four wheels, each of 30in. diameter. This engine 
worked on the Stockton and Darlington Railway till 1850. In 
September, 1835, "Locomotion" engaged in a race with the mail 
coach for a distance of four miles, and only beat the horses by 
one hundred yards! She was used to open the Middlesbrough and 
Redcar Railway on June 4th, 1846, being under the charge of Messrs. 
Plews and Hopkins on this occasion, when she hauled one carriage 
and two trucks, and took thirty-five minutes to cover the eight 
miles. From 1850 to 1857 she was used as a pumping engine by 
Pease at his West Collieries', South Durham, after which she was 
mounted on a pedestal at North Road Station, Darlington. This 
engine was in steam upon the Darlington line during the celebration 
of the Stockton and Darlington Railway jubilee in September, 1875. 



22 EVOLUTION OF THE STEAM LOCOMOTIVE 

She has been exhibited as follows: 1876, at Philadelphia; 1881, 
Stephenson Centenary; 1886, Liverpool; and 1889, Paris. In April, 
1892, she was removed from North Road to Bank Top, Darlington. 

The Forth Street Works in 1826 supplied three more engines to 
the Stockton and Darlington Railway, named " Hope," " Black Dia- 
mond," and " Diligence." These locomotives possessed many faults ; 
indeed, they were frequently stopped by a strong wind, and tho 
horse-drawn trains behind the locomotive-propelled ones were delayed 
because the engines could not proceed. " Jemmy " Stephenson 
(brother to George) was the principal engine-driver, and he was 
known far and near as most prolific in the use of oaths of a far 
from Parliamentary style. 

"Jemmy" would be cursing his engine and the horsemen cursing 
" Jemmy " for the delay ; and, indeed, the usual result was a general 
skirmish. We have already stated that Hackworth was a deeply 
religious man, and these scenes of lawlessness made a deep impression 
on his mind, so that he sought for some means to improve the 
locomotives, the radical fault of which was the shortness of steam 
Hackworth knowing that if things progressed smoothly "Jemmy" 
would have fewer occasions to display his oratorical gift. After 
eighteen months' working of the Stockton and Darlington Railway 
it was found that locomotive haulage was much more expensive than 
horse power; indeed, for every pound spent on horse power about 
three pounds were paid for locomotive power for doing an exactly 
similar amount of work. 

The 100 stock of the Stockton and Darlington Railway quickly 
fell to 50, and the shareholders began to get alarmed. 

There w T ere two opposite interests at stake that of the general 
body of shareholders and that of the locomotive builders (Messrs. Pease 
%id Richardson), who were also large shareholders in, and directors 
of the Stockton and Darlington Railway, as well as partners in the 
firm of R. Stephenson and Co. The question as to retaining the 
use of locomotive engines was fully discussed at a meeting of the 
principal proprietors, and Hackworth, as manager and engineer of 
the railway, was asked to give his opinion on the point. He replied : 
" Gentlemen, if you will allow me to make you an engine in my own 
way, I will engage that it shall answer your purpose." To have 
Defused him permission would have shown clearly to the other 
proprietors that Pease and Richardson did not care for the principles 
of steam locomotion, but that it was the locomotives constructed at 



EVOLUTION OF THE STEAM LOCOMOTIVE 



25 



the Forth Street Works they wished to retain. Therefore, after some 
discussion, it was agreed that " as a last experiment Timothy shall be 
allowed tD carry out his plan." 

Hackworth's opportunity had now arrived, ^ and the result wan 
the production of the first really successful locomotive steam engine. 

But although the shareholders "as a last experiment" gave Hack- 
worth leave to build a locomotive on his own plan, they do not appear 
to have had much belief in the success of the venture, for the hollar 
of an old locomotive was given him to use in the construction of the 
ne\\' engine. 




FIG. 10 THE FIRST SUCCESSFUL LOCOMOTIVE, IlACKWORTIi'S "HOYAL 
GEORGE," STOCKTON AND DARLINGTON RAILWAY, 1827 

The engine was originally a four-wheel engine, provided 
with four cylinders, two to each pair of wheels, anil it is 
stated to have been the first built in which a single pair of wheels 
was worked by two pistons actuating cranks placed at right angles 
to each other. She was built by Wilson, of Newcastle, and ^vas the 
fifth engine supplied to the Stockton and Darlington Railway. 

This boiler was a plain cylinder, 4ft. 4in. in diameter, and 13ft. 
'ong. A wrought-iron tube of ~* shape provided the heating 
surface, the vapour from the furnace travelling from the fire-grate un 



24 EVOLUTION OF THE STEAJl LOCOMOTIVE 

and down the tube to the chimney, which was at the same enti of 
the boiler as the grate ; indeed, the chimney was an elongation of 
the tube continued through the end of the boiler and turned up 
vertically. 

This return tube gave the new engine twice the heating surface 
of the ordinary engines, which were only provided with one straight 
tube. The locomotive was called the " Royal George " (Fig. 10), and 
was supported on six coupled wheels, each of 4ft. diameter. 

The cylinders were placed in a vertical position over the pair of 
wheels farthest from the chimney. They were 11 in. diameter, the 
stroke being 20in. Four of the wheels were provided with springs, 
but the pair connected to the pistons were not so fitted, the position 
of the cylinders rendering it impossible for springs to be used. 
The other improvements to be noted in the construction of the 
" Royal George " are : 

(1.) Springs instead of weights for the safety-valves. 
(2.) The short-stroked pumps. 

(3.) Self-lubricating bearings fitted with oil reservoirs. 
(4.) Tftie cylinders placed central with the crank journals and 
the centre of its orbit. 

(5.) The first example of six coupled wheels. 
(6.) The first really spring-mounted locomotive, the springs per- 
forming the double functions of "bearing springs" and "balance beams." 
(7) A portion of the exhaust steam used as a jet 
beneath the fire-grate and part also for heating the 
feed water ; and last and most important so impor- 
tant, indeed, that it has been described as the " life- 
breath of the high-pressure locomotive" the Steam 
Blast. (Fig. 11.) 

Trevithick, Nicholson, Stephenson, Gurney, and 
others have been credited with the production of this 
valuable contrivance, but an inquiry into the facts 
conclusively proves that before Hackworth built the 
FIG. 11- "Royal George'' the real nature of the exhaust steam 

BL\ST^PIPN blast was not understood b . v an 7 of t- n se who have 
THE "KOYAu since been credited with the invention. 

Doubtless several locomotive experimentalists, after 
various endeavours to get rid of the spent steam, at last turned the 
escape pipe into the chimney, as the meat practical way of discharging 
the exhaust steam. Trevithick did so, and George Stephenson and 





EVOLUTION OF Till- 8&EAM LOCOMOTIVE 25 

others simply followed Trevithick' s example, but knew nothing of the 
value of the exhaust steam as a means of increasing the heating powers 
of the locomotive. 

The claims of both Stephenson and Trevithick appear to be 
founded on the use of the words " steam blast " by N. Wood in his 
" Treatise," when describing the exhaust steam arrangement. This 
he probably did, not understanding the true nature of the blast, or 
contracted orifice, as invented by Hackworth. 

It is abundantly evident that Trevithick was 
absolutely ignorant of the effect of the blast on 
the fire, for in his patent ('No. 2,599) no 'mention 
is made of it, although the specification is mos*: 
minutely drawn ; indeed, thirteen years latar 
Trevithick actually patented ''fanners, etc., for 
creating an artificial draught in the chimney." 
Nicholson, in his patent (No. 2,990) dated Novem- 
ber 22nd, 1806, also says, "The steam must be 
high pressure ; the steam draught cannot be pro- 
duced by exhaust steam. This clearly shows 
Fm 12 he was not aware of the exhaust steam blast ; 

WASTE STEAM PIPE indeed, he expressly states that exhaust steam 
IN STEPHENSON'S cannot be used. With regard to George Stephen- 
"KOCKET" son, the fact that as late as July 25th, 1828, he 

wrote to Timothy Hackworth, " We have 
tried the new locomotive engine ('Lancashire Witch') at 
Bolton; we have also tried the blast to it for burning coke, and I 
believe it will answer. There are two bellows, worked by eccentrics 
underneath the tender" It will, therefore, be observed that Stephen- 
son's "blast" was produced by bellows. This letter was written ten 
months after Hackworth had successfully used the steam blast in 
the "Royal George." 

It will be shown later that it was only at the Rainhill trials, in 
October, 1829, that Stephenson learned Hackworth's secret of the 
blast pipe. Although Gurney, in 1822, used a coned pipe, he expressly 
states that the steam must be continuously ejected at a high velocity 
from a high-pressure boiler, which distinctly shows he did not us^ 
exhaust steam as Hackworth did. 

Walker and Rastrick were the engineers engaged by the directors 
of the Liverpool and Manchester Railway to report on the advantages 
to be gained from the adoption of stationary or locomotive engines on 



26 EVOLUTION OF THE STE AM LOCOMOTIVE 

the Liverpool and Manchester Railway. They decided in favour of 
the former, but they stated in their report, " Hackworth's engine 
(' Royal George ') is undoubtedly the most powerful that has yet been 
made, as the amount of tons conveyed by it, compared with the 
other engines, proves." The first year's work of the " Royal George " 
consisted of conveying 22,442 tons of goods 20 miles, at a cost of 
-only 466, whilst the same amount of work performed by horses cost 
998, showing a saving by the use of the "Royal George" of 532 
in one year. The "Royal George" was numbered, 1J, in the books 
oi the S. and D. R, 

This was the first time that a locomotive engine had worked for 
a whole year at a cheaper rate than horses. Upon this result being 
known to the Stockton and Darlington Railway directors, one of them 
exclaimed, "All we want is plenty of Timothy's locomotives." The 
" Royal George " worked night and day upon the Stockton and Dar- 
lington Railway until December, 1840, when she was sold to the 
Wingate Grange Colliery for 125 more than her original cost. 

R. Stephenson and Co. in 1828 supplied a six-Avheel coupled 
engine, " Experiment," to the Stockton and Darlington Railway. This 
locomotive had inclined outside cylinders, 9in. diameter, with a stroke 
of 24in. ; the wheels were 4ft. diameter. This engine did not give 
nearly so satisfactory results as Hackworth's "Royal George." 

Reference must here be made to Stevenson s locomotive, "Twin 
"Sisters," used in the construction of the Lancashire and Manchester 
Railway. She had two fire-boxes and boilers, and two. chimneys; 
she was supported on six coupled wheels of 4ft. diameter ; the cylinders 
were outside in an inclined position. The " Lancashire Witch " (pre- 
viously mentioned) was built by Stephenson and Co. in 1828 and sold 
to the Bolton and Leigh Railway. She was supported on four coupled 
wheels, 4ft. diameter; the cylinders were outside, 9 in. diameter, fixed 
"in an inclined position, projecting over the top and at the rear of 
the boiler. The engine is only mentioned for the purpose of noticing 
the fact that the fire was urged by means of bellows, worked by 
eccentrics fixed on the leading axle of the four-wheeled tender, which 
was specially built with outside frames for the purpose of allowing 
sufficient room to locate the bellows, etc. Yet some people have 
assurance enough to state that at the time Stephenson built this 
engine, and provided it with bellows for the purpose of urging the 
fire, he was fully acquainted with the nature and advantages of the 
steam 'blast ! ! 



EVOLUTION OF THE STEAM LOCOMOTIVE 27 

In the South Kensington Museum there is preserved the 
" Agenoria," a locomotive built for the Shutt End Railway by Fost ir, 
Rastrick and Co. in 1829, the engine being put to work on June Sn-I 
in that year. It is a four-wheel engine, with vertical cylinders, 7Jin. 
diameter, placed at the fire-box end; the stroke is 3ft., and the 
motion is token from two beams fixed over the top of the boiler, 
which is 10ft. long and 4ft. diameter. The slide valve eccentrics 
are loose upon the axle, and to enable the engine to work both ways 
a clutch is provided, as also is hand gear to the valves, to enable 
the axle to make a half turn, and so bring either the forward or 
backward clutch into action. The chimney was of abnormal height. 
The "Agenoria" worked for some thirty years. 

In 1829 R. Stephenson and Co. supplied an engine named " Rocket," 
No. 7, to the Stockton and Darlington Railway, similar in general 
design to "Experiment," No. 6 (already referred to). This engine 
was delivered at the time Hackworth was attending the Rainhill 
locomotive contest, and a director of the Stockton and Darlington 
Railway wrote to Hackworth, describing the shortcomings of this 
engine as follows : " The new one last sent was at work scarcely a 
week before it was completely condemned and not fit to be used in 
its present state. The hand gear and valves have no control in 
working it. When standing without the wagons at Tully's a few 
dayj ago it started by itself when the steam was shut off, and all 
that Jem Stephenson could do he could not stop it ; it ran down the 
branch with such speed that old Jem was crying out for help, everyone 
expecting to see them dashed to atoms ; the depots being quite clear 
of wagons, this would have been the case had not the teamers and 
others thrown blocks in the way and fortunately threw it off. A 
similar occurrence took place on the following day in going down 
to Stockton. As soon as the wagons were unhooked at the top of the 
run, away goes 'Maniac,' defying all the power and skill of her 
jockey, old Jem ; nor could it be stopped until it arrived near the 
staiths. Had a coach been on the road coming up, its passengers 
would have been in a most dangerous position. The force-pump is 
nearly useless, having had, every day it was at work, to fill the 
boiler with pails at each of the watering-places. No fewer than three 
times the lead plug has melted out. This 'Maniac' was a Forth 
Street production, and at last was obliged to be towed up to the 
'hospital' by a real 'Timothy' in front, on six wheels, and actually 
had twenty-four wagons in the rear as guard. It is now at head- 
quarters ait Shildon." 

Such was the opinion expressed by a director of the Stockton 
and Darlington concerning a Stephenson locomotive ! 



CHAPTER III. 

The Liverpool and .Manchester Railway Locomotive Competition The conditions 
of tha contest The competitors- The " Novelty " The " Sanspareil " The 
secret of the steam-blast stolen Mr. Hick's history of the " Sanspareil " 
The "Rocket" Colburn's comparison of the "Rocket" and "Sanspareil" 
Booth's tubular boiler fitted to the " Rocket " The prize divided History 
of the "Rocket" The "Perseverance" wihtdrawn from competition 
The " Cycloped " horse-propelled locomotive Winan's manumotive vehicles 
for the Liverpool and -Manchester Railway The directors purchase a dozen. 

ALTHOUGH Walker and Rastrick had reported to the directors of 
the Liverpool and Manchester Railway in favour of stationary engines, 
there were some of them who were enlightened enough to be desirous 
of giving steam locomotives a fair trial. The Stephensons being 
locomotive engine builders, naturally were not behindhand in fully 
and frequently describing the superiority of locomotive traction. 
Finally, at the suggestion of Mr. Harrison, the directors offered a prize 
of 500, to be awarded to the locomotive that at the trial appeared 
to be the best machine competing. The following is a copy of the 
notice detailing the conditions of the competition: 

" Railway Office, Liverpool, 25th April, 1829. 

" Stipulations and Conditions on which the Directors of the Liver- 
pool and Manchester Railway offer a premium of 500 for the most 
improved Locomotive Engine: 

"1st. The said engine must effectually consume its own smoke, 
according to the provisions of the Railway Act, 7, George IV. 

'" 2nd. The engine, if it weighs six tons, must be capable of draw- 
ing after it, day by day, on a well-constructed railway, on a level 
plane, a train of carriages of the gross weight of twenty tons, including 
the tender and water tank, at a rate of ten miles per hour, with a 
pressure of steam on the boiler not exceeding fifty pounds per square 
inch. 

"3rd. There must be two safety-valves, one of which must 
be completely out of the control of the engineman, and neither of 
which must be fastened down while the engine is working. 

" 4th. The engine and boiler must be supported on springs, and 
rest on six wheels, and the height from the ground to the top of the 
chimney must not exceed fifteen feet. 



EVOLUTION OF THE STEAM LOCOMOTIVE 



29 



" ,")th. The weight of the machine, with its complement of water in 
the boiler, must at most not exceed six tons; and a machine of less* 
weight will be preferred if it draw after it a proportionate weight; 
and, if the weight of the engine, etc., does not exceed five tons, then 
the gross weight to be drawn need not exceed fifteen tons, and in that 
proportion for machines of still smaller weight ; provided that the 
engine, etc., shall still be on six wheels, unless the weight (as above) 
be reduced to four tons and a half or under, in which case the boiler, 
etc., may be placed on four wheels. And the Company shall be at 
liberty to put the boiler, fire-tube, cylinders, etc., to a test of pressure 
of water not exceeding 150 pounds per square inch, without being 
answerable for any damage the machine may receive in' consequence. 




FIG. 13. THE "NOVELTY," ENTERED BY BRAITHWAITE AND ERICSSON 
FOR THE RAINHILL PRIZE 

. "6th. There must be a mercurial gauge affixed, .to the machine 
-with index rod showing the steam pressure above forty-five pounds 
per square inch. 

" 7th. The engine to be delivered complete for trial at the Liver- 
pool end of the railway not later than the 1st of October next. 

" 8th. The price of the engine which may be accepted not to 
exceed 550, delivered on the railway, and any engine not approved 
to be taken back by owner. 

"N.B. The Railway Company will provide the engine tender 
"with a supply of water and fuel for the experiment. The distance 
"within the rails is four feet eight inches and a half." 



30 EVOLUTION OF THE STEAM LOCOMOTIVE 

At this period there were but few men who understood even the 
outlines of locomotive construction, and unfortunately all of these 
did not take part in the competition. The fifth condition, limiting 
the weight of the loaded locomotive to six tons, probably deterred 
some makers from competing. Others had commenced constructing 
locomotives for the competition, but were unable to finish them by 
the date mentioned in the conditions. 

The actual entries were as follows: 1. Braithwaite and Ericsson's 
"Novelty"; 2. Timothy Hackworth's " Sanspareil" ; 3. R, Stephen- 
son's "Rocket"; 4. BurstalPs "Perseverance"; and 5. Brandreth's 
" Cycloped." 

The "Novelty" (Fig. 13) was far and awav the favourita engine at 
Rainhill, its neat appearance and smartness attracting universal atten- 
tion. It was a " tank " engine, and probably the first locomotive con- 
structed to carry its supply of water and co-al on the engine, being thus 
complete without a tender. This raised a difficulty in apportioning the 
load, as in the conditions it was arranged that the tender was to be 
counted as part of the load hauled. The machine with water r.nd 
coal weighed 3 tons 17 cwt. 14 Ib. ; the allowance made for the 
tender and fuel reduced the theoretical weight of the "Novelty,''' as 
an engine only, to 2 tons 13 cwt. 2 qr. 3J Ib. ; the gross weight 
hauled, including the locomotive, being 10 tons 14 cwt. 14 Ib. 

The "Novelty" was first tried upon October 10th, 1829 she 
had not previously been upon a railway and it was found necessary 
to make some alterations to her wheels. Timothy Hackworth, 
although he had an engine running in competition with the " Novelty," 
generously offered to repair the defect, and he personally took out 
the broken portion, welded it, and replaced it in position with his, 
own hands. 

The trials were conducted upon a level portion of line at Rainhill,. 
on a course only one and a half miles in length, and at either end 
an additional eighth of a mile was allowed for the purpose of getting 
up the speed and stopping after the run of a mile and a half. The engines, 
had to make forty runs over the course, or a distance of sixty miles 
in all, which was computed to be equal to a return journey between 
Liverpool and Manchester. 

After running two trips of one and a half miles each, the pipe 
from the pump to the boiler burst, in consequence of the cock between 
the boiler and pump having, by accident, been closed. The " Novelty " 
and train covered the first trip in five minutes thirty-six seconds, and 



EVOfA'TWX OF THE STEAM LOCOMOTIVE 31 

the return in six minutes forty seconds; being at the rate of 16.07 
and 13 J miles an hour respectively. After being repaired, the engine, 
with its train, imde an unofficial trip, and developed a speed cf 
21 1-6 miles an hour. Without a load the "Novelty" attained a 
speed of nearly thirty miles an hour. 

The " Novelty" was again tried on October 14th, but upon its third 
trip part of the boiler gave way, and it was decided to withdraw the 
locomotive from competition. 

The boiler of the "Novelty" was partly vertical and partly hori- 
zontal; the latter portion was about 12ft. long and 15in. in diameter. 
In the former was the fire-box, surrounded by water, coke being- 
supplied through what at first might be mistaken for the funnel of 
a steam fire-engine. This was, however, kept air-tight, the fuel 
being introduced by means of a descending hopper. The area of 
the fire-grate was 1.8 sq. ft., the fire-box heating surface 9J sq. ft.,, 
and the heating surface of the tubes, 33 sq. ft. 

The air entered below the fire-bars by a pipe traversing the length 
of the engine, and connected with bellows fixed above the frame at the 
other extreme of the engine. The bellows were worked by the engine, 
so that the " Novelty " was provided with a forced draught. The 
heated air was forced through a tube, which made three journeys 
through the horizontal portion of the boiler, and was consequently 
36ft. in length. It was 4in. in diameter at the grate end, and Sin. at 
the other extreme, where it was turned up as a chimney. The 
cylinders were located over the pair of wheels at the bellows end of 
the machine. They were fixed vertically, the diameter being 6in., 
and length of stroke 12 in. The piston rods worked through the 
top covers, and by means of cross-heads, side-rods, and bell-cranks 
the motion was conveyed to the crank axle beneath the vertical portion 
of the boiler, although, as previously mentioned, the cylinders were 
over the other pair of wheels. The wheels were 4ft. 2.1in. in diameter, 
and chains were provided for coupling the wheels together ; but these 
were not used at Rainhill. 

The water was carried in a tank located between the axles below 
the frame. The construction of the "Novelty" was only decided upon on 
August 1st, 1829, but so expeditiously was the work carried out 
that she was constructed in London and delivered in Liverpool a 
lengthy journey at that time by September 29th, 1829. Her dis- 
tinguishing colours at Rainhill were copper and blue. 

After the conclusion of the Rainhill Competition several alterations- 



32 EVOLUTION OF THE STEAM LOCOMOTIVE 

were made in the design of this engine, the position of the cylinders 
being altered from vertical to horizontal by Watson and Daglish, 
and in 1833 she was working on the St. Helens and Runcorn Gnp 
Railway. 

Although, through an accident, the "Novelty" had to be with- 
drawn from competing for the prize at Rainhill, the directors of the 



ww 




FIG. 14.-HACKWORTH'S "SANSPAREIL," ONE OF THE COMPETITORS 
AT RAINHILL 

Liverpool and Manchester Railway were so well satisfied with her 
performances that they gave Braithwaite and Ericsson an order for 
some locomotives of the same design. A description of these will bo 
found in Chapter IV. 

The engine next in order was Timothy Hackworth's " Sanspareil," 
(Fig. 14), now preserved in the South Kensington Museum. 



EVOLUTION OF THE STEAM LOCOMOTIVE 33 

The engine-shops at Shildon were not in a position to construct the 
whole of this locomotive; consequently Hackworth was forced to 
obtain the boiler and cylinders from other makers. The former 
was constructed at Bedlington Ironworks, and was of cylindrical form, 
6ft. long, 4ft. 2in. diameter, with one end flat and the other hemi- 
spherical. The heating surface was provided by means of a double 
return tube, the fire-grate and chimney being b(K-n at the same end. 

The area of the fire-grate was 10 sq. ft., the heating surface of 
same 15.7 sq. ft., the remaining heating surface 74J sq. ft. 

The -cylinders were constructed by R. Stephenson and Co., and six 
had to be made before two perfect ones were obtained, the sixth one, 
indeed, only being fitted at Liverpool when the contest was in progress. 

It has been stated that these cylinders were purposely constructed 
in a faulty manner to prevent the " Sanspareil " beating the " Rocket." 
This may or may not be true, but it is very evident that, save for 
Stephenson's imperfect workmanship in this respect, the " Sanspareil " 
would have won the 500 prize. When the " Sanspareil " was com- 
peting for the prize, one of the cylinders supplied by Stephenson and 
Co. burst, and it was found that the metal was only one-sixteenth of an 
inch thick! A nice state of things certainly! The cylinders were 
Tin. diameter, the stroke being 18in. The engine was carried on four 
wheels, 4ft. Gin. diameter. Total weight of engine, 4 tons 15 cwt. 2 qr. 

She was, of course, fitted with Hackworth's exhaust steam blast. 

During some preliminary trips at Rainhill, Stephenson was greatly 
surprised to see how well the " Sanspareil" ran, and he noticed sho 
always had a good supply of steam, so he got upon the engine and 
had a ride on her. During this trip he said to Hackworth, " Timothy, 
what makes the sparks fly out of the chimney?" Hackworth touched 
the exhaust pipe near the cylinders, and answered, "It is the end of 
this little fellow that does the business." 

After Stephenson got off the engine, John Thompson, the driver 
(he was Hackworth's foreman at Shildon), said to Hackworth, " Why 
did you tell him how you did it, sir? He will be trying to fit up the 
' Rocket ' in the same way." Hackworth said he did not think so, 
but Thompson determined to watch the "Sanspareil" all night. He 
therefore locked himself in the shed containing the engine that night, 
but towards daybreak sleep overcame him, and when he awoke he 
s-iw two men getting out of the window of the shed, and he found 
the chimney door of the "Sanspareil" open, and some materials 
inside the chimney. The secret of the exhaust steam blast was 



34 EVOLUJ.'ION OF THE STEAM LOCOMOTIVE 

stolen ! The next evening the " Rocket " again appeared ; this time 
she was fitted with a similar contrivance. The above is Hackworth's 
foreman's version of the theft, but the " Practical Mechanic's Journal " 
for June, 1850, gives the tale as told by the man who committed 
the theft. 

When in repair, the " Sanspareil " ran faster, took a heavier load, 
and consumed less coke than the "Rocket," and whilst the latter 
was remodelled within twelve months of the Rainhill contest, the 
former worked with practically no alteration until 1844. In 1864 
she was presented to the South Kensington Museum by Mr. John 
Hick, M.P., Bolton. 

The following is an extract from her history, as supplied by Mr 
Hick to the Museum authorities : 

"After the Rainhill trial the engine was purchased by the Liver- 
pool and Manchester Railway Company, and used by them for various 
purposes. In 1831, the engine was purchased by Mr. John Har- 
greaves, of Bolton, and was employed by him in the conveyance of 
passengers and general traffic on the Bolton and Leigh Railway for 
several years. In 1837, Mr. Hargreaves had the engine thoroughly 
repaired, and put on a pair of new cylinders of larger dimensions 
than the old ones, so as to increase the power. The original wood- 
spoked wheels were also removed at this time, and replaced with 
cast-iron hollow-spoked wheels. 

" One pair of these are under the engine at the present time. The 
engine continued regularly at work in conveying coals, general goods, 
and passengers until 1844, when, being found much too small and 
short of power for the rapidly increasing traffic, Mr. Hargreaves took 
her to his colliery at Coppull, near Chorley, Lancashire, where the 
engine was fixed near a coal-pit. One axle and one pair of wheels 
were removed, and upon the other toothed gear was fitted, in order 
to give motion to winding and pumping apparatus, and the engine 
commenced its work as a regular fixed colliery engine, pumping and 
winding in the most satisfactory manner until the end of the 
year 1863 ; having raised many thousand tons of coal and many million 
gallons of water, and even at the time above named was in fair 
working order, and only removed because the coal in the pit was 
exhausted. 

" I hope the old engine will now find a permanent resting place ii? 
the Kensington Museum, where her end will be peace, if not pieces. 
Mr. Hargreaves has kindly given me the old engine, in consequence 



EVOLUTION OF THE STEAM LOCOMOTIVE 35 

of my having told him of my intention with regard to her. And 
having restored her as far as possible by collecting and putting 
together the available materials, I have pleasure in presenting this 
interesting relic to the Museum." 




Fi&. 15. STEPHENSON'S "ROCKET," THE WINNER OF THE RAINHILL 
PRIZE OF 500. 

The "Rocket" (Fig. 15), was entered in the name of 
Robert Stephenson, and was constructed at the Forth Street 
Works, Newcastle-on-Tyne, in 1829. Her distinguishing colours 
were yellow and black, with a white chimney. She was 
the first engine to be tried at Rainhill. Her weight was 4 tons 
5 cwt. ; load (including tender), 1 2f tons ; total, 1 7 tons. 
During the first twenty trips she attained a maximum speed of 
24.43 miles an hour, the average being 13.42 miles an hour; during 
the second twenty trips an average speed of 14.2 miles an hour was 
the result, with a maximum speed of 24 miles an hour. These short 

D 2 



36 EVOLUTION OF THE STEAM LOCOMOTIVE 

trips of one and a half mile^ each just suited the design of the 
" Rocket," as the steam raised before starting on each trip was 
sufficient to work her the one and a half miles; had the trips been 
longer, she not then being furnished with a -proper blast, 
but with that illustrated on Page 25 would probaL. A y have 
stopped for want of steam. Of this Z. Colburn signifi- 
cantly states : " The ' Rocket,' on the first day of her trial, 
derived but little benefit from the discharge of the exhaust steam 
up the chimney ; and, indeed, made steam nearly as freely when stand- 
ing as when running." Without a load, or tender even, she attained 
a speed of 29J miles an hour. The authority just quoted says : 
"The real power of the ' Sanspareil ' is to be estimated by its rate 
of evaporation, which was one- third greater than that of the 'Rocket/ 
and thus the ' Sanspareil,' after allowing for its greater weight, was 
the most powerful engine brought forward for trial. . . As far as 
it had gone, the mean rate of speed (of the ' Sanspareil ') was greater 
than that of the 'Rocket' up to the same stage of the experiment." 

The boiler of the " Rocket " was cylindrical, with flat ends, 6ft. 
long, 3ft. 4in. diameter; the fire-box was 3ft. long, 2ft. broad, and 
about 3ft. deep ; between the box and the outer casing was a space 
of 3in. filled with water. The cylinders were placed at an angle of 
45 degrees at the fire-box end, the connecting-rod being attached to 
a pin on the leading wheels, which were 4ft. 8jin. diameter, that of 
the cylinders being Sin. ; the stroke was 16 Jin. 

The "Rocket" had a great advantage over other engines because 
she was supplied with a tubular boiler, containing 25 tubes of 3in. 
diameter. The idea of the tubular boiler did not originate with 
the Stephensons. Mr. Booth, the Secretary of the Liverpool and Man- 
chester Railway, suggested their use in the " Rocket " ; but before this 
the tubular locomotive boiler had been patented by a Frenchman 
(M. Sequin), on February 22nd, 1828. Mr. Booth, however, states 
that he was unaware of the French patent, and, so far as he was 
concerned, the tubular boiler was an original discovery. The use of 
these tubes increased the evaporating power of the boiler three-fold, 
and at the same time reduced the consumption of coke 40 per cent. ; 
yet the "Rocket," with this great advantage, was not equal to the 
"Sanspareil," until the former was fitted with Hackworth's blast. 
When this had been done, the "Rocket" was capable of hauling 
20 tons (engine included) up an incline of 1 in 96, at 16 miles an 
hour, -for a distance of one and a half miles. The prize of 500 was 



EVOLUTION OF THE STEAM LOCOMOTIVE "6( 

divided between Robert Steplienson, the constructor of the " Rocket," 
and Mr. Booth, the suggester of the tubular boiler, which enabled 
that locomotive to be entitled to the prize. Tubular boilers had been 
successfully used in steam road coaches as early as 1821. 

After running a year or so, the " Rocket " was re-built, the cylin- 
ders being placed in a slightly inclined position over the trailing 
wheels, but still working the leading wheels; a smoke-box was added, 
and other improvements introduced. 

The "Rocket" was bought in the year 1837, from the Liverpool 
tind Manchester Railway, by Mr. J. Thompson, of Kirkhouse, the lessee 
of the Earl of Carlisle's coal and lime works. 

Here the engine was worked for five or six years on the Midgeholine 
line a local line belonging to Mr. Thompson for forwarding his 
coals from the pits towards Carlisle. 

Soon after the engine was placed on this line the great contest 
for East Cumberland took place, when Sir J. Graham was superseded 
by Mayor Aglionby, and she was used for conveying the Alston 
express with the state of the poll from Midgeholme to Kirkhouse. 
Upon that occasion the " Rocket " was driven by Mr. Mark Thompson, 
and accomplished her share of the work, a distance of upwards of four 
miles, in 4J minutes, thus reaching a speed nearly equal to 60 miles 
an hour. On the introduction of heavier and more powerful engines, 
the ''Rocket" was "laid up in ordinary" in the yard at Kirkhouse. 
This historic steam locomotive is now preserved in the South Kensing- 
ton Museum. It must not be forgotten, however, that the " Rocket " 
has been rebuilt, and its design considerably altered, since the Rainhill 
competition of 1829. 

The last of the steam locomotives entered for trial at Rainhill 
remains to be described. The " Perseverance" was constructed by Mr. 
Burstall, of Edinburgh. He was already known as a maker of steam 
road coaches. Unfortunately for the success, or rather want of 
success, of the "Perseverance," Mr. Burstall designed his railway 
locomotive on much the same lines as his steam coaches. 

The "Perseverance" had the misfortune to have some damage 
done to its wheels, etc., when being unloaded at Rainhill off the 
wagon on which it had been conveyed from Liverpool. A pre- 
liminary trial was made, and Mr. Burstall, finding the engine was 
unable to attain a higher s~eed than about six miles an hour, with- 
drew his locomotive from competition. 

The boiler was horizontal, and the water was admitted to shallow 



DEVOLUTION OF THE STEAM LOCOMOTIVE 

trays placed over the fire, and in this way was immediately converted 
into steam. The cylinders were vertical, and worked horizontal 
beams placed above them ; the wheels were worked by cranks fixed on 
the beams about half-way between the cylinders and the centre pivots 
of the beams. The second pair of wheels was driven by means of 
an axle with bevel wheels at each end, which conveyed the motion 
from the one axle to the other. 

This engine was distinguished by having the wheels painted red. 

Although not " steam " locomotives, we think it right to give a 
few details of the " Cycloped," (Fig. 16), and also of Winans' manumo- 
tife carriage, both of which were exhibited at Rainhill. The former was 




FIG. 16. WINANS' "CYOLOPED" HOUSE LOCOMOTIVE 

forked by a horse or horses fastened on a frame supported by four 
wheels ; the horses walked at a speed of one and a quarter miles an 
hour, on an endless platform formed of planks of wood. The horses 
being firmly attached to the frame could not go forward when they 
essayed to walk, and the consequence of their using their legs was tli3 
revolving of the floor, which worked round drums geared to the driving 
wheels. This motion caused the vehicle to move forward on the rails 
at a speed of about three miles an hour, with a, load of fifty passengers. 
Had the horses moved at a quicker rate, the speed of the " Cycloped " 
would have been increased in a proportionate ratio. 

Winans' carriage was worked by two men, who turned a windlass, 
which actuated the wheels. It accommodated six passengers, and it 



EVOLUTION OF THE STEAM LOCOMOTIVE 69 

was facetiously proposed that those passengers who worked at the 
windlasses should be conveyed by such vehicles at reduced rates. 
Although we now smile at the simplicity of such vehicles ever having 
been suggested for working on a railway, the Directors of the Liver- 
pool and Manchester Railway were considerably taken with the idea of 
Winans' man-propelled carriages, and they engaged two well-known 
engineers to report on their adaptability for passenger traffic on the 
railway. As might be expected, the experts reported against the 
proposed use of Winans' machines ; but, despite this adverse report, 
the Directors of the Liverpool and Manchester Railway actually 
bought twelve of these " manumotive " carriages of Winans. The 
purchase was made prior to March, 1830, and as we do not read of 
their being used after the railway was opened in September, 1830, 
we may conclude that during the six months that elapsed between 
tho purchase and the opening of the line the Directors had come to 
che same conclusion regarding the machines as did the engineers 
who reported against their use on the railway. 



CHAPTER IV. 

An important improvement in the locomotive Bury's original " Liverpool," 
the first inside cylinder < ngine Bury's own account of his invention- CKher 
authorities agree with Bury Extract, supplied by the Secretary of tlie 
L. & N.W.Rly., from the minute books of (he Liverpool and Manchester Rly. 
An r-nrly authentic list of Bury's locomotives Description of Bury's " Liver- 
pool " Last hoard of on the Bolton and Kenyon Railway The '' Invicta " 

for the first Kentish railway Still preserved by tie S.E.R First official trip 

on tits Liverpool -\t\t\ Manchester Railway Formal opening of the L. & Al.R. 
The locomotives that took part in the ceremony Ihe " William t. * 
Fourth '' and " Queen Adelaide " for the L. & M.Rly. Hackworth's 
" G-lobe " for the Stockton and Darlington Railway The romance of her 
construction, life, and end Stephensn/i's " Planet " S me of her feats on 
the L. & M.Rly. Meavier locomotives for the L. & M.Rly. Dodd's engine 
for the Monklind and Kirkintilloch Rly. Historical locomotive sold by 
auction for ?Q guineas Bury's "Liver" for the L. & M.Rly.- Mi, re Hack- 
worth "iron horses'' for the Stockton and Darlington Rly. Despite thei- 
peculiarities, they prove most successful The "Caledonian." 

We have now to deal with one of the most important improve- 
ments in the locomotive viz., that introduced by Mr. Edward Bury, 
of tfce Clarence Foundry, Liverpool, in the design of his celebrated 
"Liverpool," (Fig. 17). Of late years many extraordinary statements 
concerning various types and designs of locomotives have been made, 
and the "romancing" relative to the original "Liverpool" is perhaps 
the most conspicuous, whilst at the same time its uaer incorrectness i ; 
easily proved. 

One of these statements is that " the first engine built by Bury 
at Clarence Foundry was an outside cylinder engine, the Tread- 
nought/ which was completed March 30th, 1830, but proved a failure. 
However, he lost no time, but, with the assistance of his foreman, 
Mr. Kennedy, got out working drawings for a new engine, to be 
named the 'Liverpool.' This engine, No. 2 in the locomotive order 
book, and class A in the description book, was commenced early in 
January, 1831 ; it was completed in March of that year, and in May, 
1831, it was put to work on the Petersburg Railroad of America. It 
had four coupled wheels of 4ft. Gin. diameter." 

Now, as to the facts, Bury's books were sold by auction by his 
creditors on August 15th and I6th, 1851 ; and, even if the books 
are now in existence (which is extremely unlikely), it is obviously 
impossible for them to contain the particulars quoted above, for the 
very simple and conclusive reason that the facts relative to the 
original " Liverpool " are quite different to the statement just quoted. 



EVOLUTION OF THE STEAM LOCOMOTIVE 41 

There are three improvements with which Bury is justly credited 
in the locomotive now under review viz., the adoption of (1) horizontal 
inside cylinders below the smoke-box, (2) cranked driving axle, nnd. 
(3) coupled driving wheels of the (then) great diameter of six feet. 

In describing this historical " Liverpool" locomotive we cannot 
do better than quote Bury, the maker and designer of it, and 
Kennedy, his foreman, who constructed it. The former, at a meeting 
of the Institute of Civil Engineers, held on March 17th, 1840, read a 
paper on the locomotive, and, speaking of the inside cylinder engine, 
said : " This form of engine was adopted by the author as early as 
1829^, when he constructed the 'Liverpool/ which was the original 
model engine, with horizontal cylinders and cranked axles. It was set 
to work on the Liverpool and Manchester Railway in July, 1830." 



Tf 




FIG. 17. BURY'S ORIGINAL " LIVERPOOL," THE FIRST ENGINE 

\VITH INSIDE CYLINDERS AND CRANKED DRIVING AXLE. COUPLED 

WHEELS, 6ft. DIAMETER. 

About 1843 there was considerable discussion amongst engine 
builders and locomotive engineers as to the relative safety of inside 
and outside cylinder engines, and also regarding the superiority of 
the four-wheel or six-wheel locomotive. Bury and Co. thereupon 
issued a circular giving a history of the locomotive practice of their 
firm, and the various advantages claimed for their locomotive designs. 

From this circular we extract the following remarks, as bearing 
upon the point now under discussion : " It was the good fortune of 
the conductor of this foundry to originate the construction of four- 
w!iesl engines, with inside framing, crank axles, and cylinders placed 



42 EVOLUTION OF THE STEAM LOCOMOTIVE 

in the smoke-box. . . . The first engine on this principle was 
manufactured in this foundry in 1829, prior to the opening of the 
Liverpool and Manchester Railway." Such are Mr. Bury's statements 
concerning the original "Liverpool." 

We will now see what his partner, Mr. James Kennedy, the then 
President of the Institution of Mechanical Engineers, had to &ay 
regarding the "Liverpool." 

At a meeting of the Institute of Civil Engineers, held on Novem- 
ber llth, 1856, a communication was read from Mr. Kennedy, in 
which he stated that "the late George Stephenson had told both 
Bury and Kennedy, after having seen the e Liverpool ' engine on the 
Liverpool and Manchester Railway, that his son, the present Robert 
Stephenson, had taken a fancy to the plan of the 'Liverpool,' and in- 
tended to make immediately a small engine on the same principle." 
This he afterwards did, Stephenson's "Planet" being the s:iid engine 
* on (he same principle." Kennedy went on to state that " the 
letter-book of the firm (Bury and Co.) for the year 1830 contained 
the whole of the correspondence on the subject between the Directors 
of the Liverpool and Manchester Railway and Bury." 

The reader can readily judge as to which statement is likely to be 
correct those of such well-known men as Bury and Kennedy, which 
are concise, straightforward statements of known and accepted facts, 
or the recently published remarks concerning the "books, etc." 

Fortunately, students of locomotive history are not even obliged 
io decide either one way or the other on the statements pro and con 
already quoted concerning the original "Liverpool," but are able to 
gain independent and conclusive evidence on this important point in 
locomotive history. For the purpose of finally clearing up the point, 
the writer communicated with the Secretary of the London and North 
Western Railway, asking him to examine the Directors' Minute Book^ 
of the Liverpool and Manchester Railway for the year 1830, to see if 
these authentic documents contained any reference to Bury's " Liver- 
pool." Mr. Houghton most generously had the search we required 
made, and the result was as might have been expected. But let the 
letter tell its own tale. 

"London and North Western Railway, 

"Secretary's Office, Euston Station, N.W. 

"June, 3rd, 1896. 
"JDear Sir, With further reference to your request for information 



FVOLVTION OF THE STEAM LOCOMOTIVE 



43 



relative to Bury's locomotives, I have had the Minute Books of the 
Liverpool and Manchester Railway searched for the years 1829-30. 

"Towards the end of 1830 the Board sanctioned the [further] trial 
of the 'Liverpool,' and it was consequently allowed to work on the 
railway in competition with one of Mr. Stephenson's engines. The 
engineer was dissatisfied with the size of the wheels, which were 
6ft. instead of his maximum 5ft.; and there was a long con- 
troversy as to the respective merits of circular and square fire-boxes, 
which was ultimately referred to arbitration, when the square boxes 
recommended by Mr. Stephenson were given the preference. Tours 
truly, "(Signed) T. HOUGHTON." 

The above letter conclusively settles the points in dispute viz., 
that the " Liverpool " was tried on the Liverpool and Manchester Rail- 
way in, 1830, and that the diameter of the wheels was 6ft. 

We have thus pricked the specious bubble that stated the "Liver- 
pool" was duly commenced to be built in 1831, and that the diameter 
of the wheels was but 4ft. 6in. ! 

Readers may wonder why such obviously inaccurate statements 
should be published. One can only conjecture. Many lists 
of early locomotives have during the past few years been 
published. These should, however, be accepted with wiie very 

greatest caution. The following table of dimensions of Bury's 
early engines appeared as long ago as September 18th, 1857, in 

.the Engineer. As this was nearly forty years before "locomotive 
lists " had any marketable value, there can be no reason to call in 

question its accuracy : 



ol 

dS" 

fc 


Diameter 
rl 
Cylinder. 


. 0) 

fl 

KlW 


Diameter 
of 
Wheels. 


jj 

SS 


Diameter 
of Tubes. 


*J 

15 


, Diameter 


g 

a 




-al. 

IB 

32 


Diameter 
of 
Fire-box. 


Area of 
Tubes. 


o 

i! 

HOJ 


No. 


Ins. 


Ins. 


Ft. ins 


No. 


Ins. 


No. 


In 


8. 


Ft. Ins. 


Ft. Ins. 


Inches. 


Sq. Feet 


2 


12 


18 


6 


79 


1* 


52 


1 




7 11} 


8 9 


19 


450-25 


3 4 


11 


16 


5 


73 


f 


2-1 


1 




7 1 


3 


leg 


303-58 


5 


8 


16 


4 6 


40 


If 


13 


1 




6 5 


2 7J 


(6) 120(7) 


160-28 


6 & 7 


12 


18 


5 6 


76 


2 


13 


1 




8 6 


3 7 


18$ 181/16 


390 


8 & 


9 


16 





40 


If 


13 


1 




6 11 


2 101 


12| 


162 


10 


9 


16 


4 


43 


2 


9 


1 




7 1 


3 


14* 


189-1 


11 


8 


16 


6 


40 


If 


13 


1 




6 74 


2 9J 


12| 


158-2 


12 & 13 


10 


16 


6 


51 


2 


9 


1 




7 2$ 


3 04 


Ii 


222-74 


14 


8 


16 


6 


43 


If 


9 


1 




6 8 


2 10t 


12$ . 


155 


15. 16, &17 


12 


18 


4 6 


58 


N 


2 


2 




8 2 


3 7 


19J 191/19 


31840 


18 & 19 


10 


16 


4 6 


76 


2 


~ 


If 




7 2} 


3 2 


151 


251 



At the present time there exists a market for early locomotive 
details ; as with other marketable commodities, given a demand, a 
supply (of some kind) will be forthcoming. 



44 EVOLUTION OF THE STEAM LOCOMOTIVE 

We have a copy of the report prepared by the arbitrators, appointed 
by the Directors of the Liverpool and Manchester Railway, to inquire 
into the question of the round or square fire-box, as mentioned 
in Mr. Houghton's letter. The report was made by John Farey and 
Joshua Field, two celebrated engineers of that period, and was in 
favour of the square fire-box. 

It will now be of interest to give a description of Bury's original 
" Liverpool," which was designed and her construction commenced in 

1829. She contained many unusual features. Instead of a tubular 
boiler a number of convoluted flues were used. The fire was urged 
by bellows fixed under the tender ; the driver stood at one end of the 
engine in front of the smoke-box, and the fireman at the other end, 
behind the fire-box ; the cylinders were horizontal, placed, inside the 
frames beneath the smoke-box; their diameter was 12in., the stroke 
being 18in. ; the four wheels were 6ft. in diameter, and were coupled, 
and the driving axle was, of course, cranked. 

The "Liverpool," in this her original state, was used as a ballast 
engine in the construction of the Liverpool and Manchester Railway, 
but not being very successful, was withdrawn. After some altera- 
tions, she was again put to work on July 22nd, 1830. Then the 
crank axle appears to have broken, and she was again removed for 
repairs, and again put to work on the Liverpool and Manchester 
Railway on Octobor 26th, 1830. After the report previously men- 
tioned, the Directors refused to purchase the "Liverpool," and Bury 
removed her to the Bolton and Kenyon Railway. Here she attained 
a speed of' 58 miles an hour with twelve loaded wagons. On this 
line one of her wheels broke, and the driver was killed. As a result 
of this accident, she was then rebuilt and sold to Hargreaves, the con- 
tractor, for locomotive power on the Bolton and Kenyon Railway, and 
continued to work on that line for some years. 

The Canterbury and Whitstable Railway was opened on May 3rd, 

1830, and was the first locomotive line in the South of England. The 
original engine, the " Invicta " (Fig. 18), is still preserved by the South 
Eastern Railway at Ashford, but it is a mere chance that this engine did 
not disappear nearly sixty years ago. The Canterbury and Wliistable 
Railway Company, after a short time, let the working of the line to 
contractors, and they preferred to work it by horse-power, and wo 
find that in October, 1839, the contractors were advertising the 
" Invicta" for bale, describing her as of "12 horse-power, 18in. stroke, 
cylinders 24in. long, 9^-in. diameter, wheels 4ft. in diameter." Fortu- 



EVOLUTION OF THE STEAM LOCOMOTIVE 



45 



nately for students of early locomotives, there was no demand for the 
engine anywhere in the neighbourhood of Whitstable, there then being 
no other locomotive line nearer than Greenwich, on which she could 
have been used ; so no buyer was forthcoming, and the " Invicta" was 
thereupon laid up. The dimensions of the " Invicta," as supplied 
to us by Mr. J. Stirling, are as follows: Cylinders, 10|in. diameter, 
fixed in inclined position over leading wheels, and working the trailing 
wheels; stroke, 18in. ; four-coupled wheels, 4ft. diameter; wheel 
base, 5ft.; boiler, 10ft. Sin. long, 3ft. 4in. diameter, containing a 




FIG. 18. THE "INVICTA," CANTERBURY AND WHITSTABLE RWY., 1830 

single flue 20in. diameter ; distance from top of boiler to rails, 6ft. : 
from top of chimney to rails, lift. lin. ; chimney, 15in. diameter; 
total length over all, 13ft. Gin. At the bottom of the chimney is o 
kind of smoke-box, measuring about 2ft. 4in. high, 1ft. Sin. long, aud 
2ft. 4in. wide. The South Eastern Railway exhibited the "Invicta," 
at the jubilee of the Stockton and Darlington Railway in 1875, and 
at the Newcastle Stephenson Centenary in 1881. The "Invicta," when 
originally built, is said to have had a tubular boiler. 

The Directors of the Liverpool and Manchester Railway in 1829 
ordered of Stephenson and Co. seven engines of somewhat similar 
design to the " Rocket." The Directors made their first trip by rail- 
way from Liverpool to Manchester and back on Monday, June 14th, 
1830. The train was drawn by the "Arrow," and consisted of two 
carriages and seven wagons ; the total weight, including the engine, 
was 39 tons, the journey to Manchester being made in two hours one 




46 EVOLUTION OF THE STEA V .LOCOMOTIVE 

minute, whilst the return trip to Liverpool only took one and a half 
hours, a speed of 27 miles being attained for some distance. 

The Liverpool and Manchester Railway was formally opened on Sep- 
tember 15th, 1830, when the "Northumbrian" (Fig. 19), driven by 
George Stephenson, hauled the train consisting of the Duke of Welling- 
ton's carriage, the band, etc., on one line, whibt the " Phoenix." ' North 
Star," Rocket," "Dart," " Comet," '" Arrow," and " Meteor," each 

hauled a train upon the other 
line. Starting from Liverpool, 
the eight trains proceeded to- 
wards Manchester. At Parkside 
Mr. Huskisson was run over by 
the " Rocket," and he was placed 
on the "Northumbrian" and 
conveyed to Eccles in 25 minutes, 
or at the rate of 36 miles an hour. 
The Duke of Wellington's 
carriage was now left without an 
engine, and a curious sight was 
witnessed ; a long chain was 
obtained, and the trains which 
had been up to this point hauled 

by the " Phoenix " and " North Star," consisting of ten carriages, were 
joined together. The chain was then fixed to the Duke of Welling- 
ton's train on the other line, and so the rest of the journey was per- 
formed by the two engines and ten carriages on one line hauling 
another train upon a parallel set of rails. It may be of interest to 
observe that the carriage built for the Duke of Wellington was provided 
with eight wheels, so it will be noticed that eight-wheeled passenger 
stock is not at all a modern introduction, but, on the contrary, has 
been in use ever since the opening of the first railway built for the con- 
veyance of passengers. The vehicle in question was 32ft. long and 8ft. 
wide. 

The two engines ordered by the Directors of the Liverpool and 
Manchester Railway of Braithwaite and Ericsson after the style of 
the " Novelty," were named " William the Fourth," (by special per- 
mission of that monarch) and " Queen Adelaide." They were delivered 
to the Liverpool and Manchester Railway immediately the railway 
was opened, and on September 22nd, 1830, the "William the Fourth" 
ran off the rails on the Sankey Viaduct. A very considerable number 



FIG. 19. THE "NORTHUMBRIAN," THE 

ENGINE THAT OPENED THE LIVERPOOL 

VN'D MANCHESTER RAILWAY 



EVOLUTION OF THE STEAM LOCOMOTI V /; 47 

of trials were made with these locomotives on the Liverpool and 
Manchester Railway; but, as was the case with Bury's "Liverpool/' 
Stephenson. strongly objected to any 'other maker's engines being used 
on the line, and he was, therefore, always ready to find 
out some fault in the engines not of his construction ten- 
dered to the company. Braithwaite and Ericsson claimed 
four great advantages for their class of engines viz., (1) the 
total absence of all smoke; (2) the dispensing with a chimney; (3) 
a saving of 120 per cent, in the cost of the fuel, and of 30 per cent. 
in the space required to store it ; (4) a saving of 400 per cent. i> 
the space occupied by the boilers. 

Several improvements were introduced into the " William the 
Fourth," and "Queen Adelaide," so that they differed somewhat from 
the " Novelty." They were provided with four-wheeled tenders, which 
were placed in front of the engines. The four wheels of the engines 
were 5ft. in diameter, the wheel base being 6ft. 9in. The horizontal 
portion of the boiler was 8ft. long, the vertical portion, containing 
the fire, etc., being 6ft. 6in. high and 4ft. diameter. The cylinders 
were vertical, but worked dowmvards ; they were located one on each 
side of the vertical portion of the boiler, and a little to the rear of 
the leading wheels, to which the motion was conveyed by means of 
bell-cranks and connecting-rods the latter joined the axle within the 
wheels, so that the driving axle was cranked. 

The next engine that requires our attention is the celebrated 
ki Globe " (Fig. 20), designed for the Stockton and Darlington Railway 
by Timothy Hackworth, and built iby R. Stephenson and Co. The 
lt Globo " was built for passenger traffic ; she was provided with a steam 
dome, and -was the first locomotive built with this advantageous appen- 
dage for obtaining dry steam. The valve motion was reversible by a 
single lever. The heating surface was provided for by means of a 
single fire-tube, whilst behind the fire-bridge, and extending to the 
chimney, were seven small radiating tubes crossing the main flue. 

This idea of Hackworth's was afterwards introduced by Galloway 
in his stationary engine boilers, and patented by him. The engine 
"Globe" had a cranked axle and inside cylinders. 

Hackworth explained the construction of the "Globe" to the 
Directors of the- Stockton and Darlington Railway, and he was 
instructed to go to Newcastle and arrange for the building of the 
"Globe" by Stephenson and Co. He saw the officials at the Forth 



48 



EVOLUTION OF THE STEAM LOCOMOTIVE 



Street Works on March 3rd, 1830, and after the examination of the 
plans there, it is stated that one of the officials objected to the crank 
axle, saying " it would certainly inyolve a loss of power, as the efficient 
length of lever could only be calculated from the inside of the journal 
to the axle's centre." It is well known that Geo. Stephenson had 
previously seen Bury's " Liverpool," and said of it, " My son has taken 
a fancy to the plan of the 'Liverpool' engine, and intends to make 




FIG. 20. HACKWORTH'S " GLOBE " FOR THE STOCKTON AND DARLINGTON 
RAILWAY. THE FIRST LOCOMOTIVE WITH A STEAM DOME 

immediately a small engine on the same principle.'* Hackworth's 
reply to the objection to the crank axle was " that he held Stephen- 
son responsible only for supplying good workmanship, and not for 
any failure of the design, should such occur." 

On March 3rd, 1830, Hackworth, in company with Harris Dickin- 
son, one of R. Stephenson and Co.'s foremen, drove over to Bedling- 
ton Iron Works to order the boiler plates required for the construction 
of the " Globe." 

Hackworth remained at Newcastle till March 6th, and being satis- 
fied that the construction of the " Globe " would be immediately pro- 
ceeded with, he returned to Darlington, having obtained a promise 
of quick delivery. The boiler plates were delivered at the Forth 
Street Works, April 14th, 1830. 



EVOLUTION OF THE STEAM LOCOMOTIVE 49 

The completion of the engine was, however, delayed until after 
R. Stephenson and Co. had delivered the "Planet," inside cylinder 
locomotive, to the Liverpool and Manchester Railway. The " Globe " 
opened the Stockton and Middlesbrough Branch of the Stockton ai.'.d 
Darlington Railway on December 27th, 1830- Her speed frequently 
exceed 50 miles an hour with passenger trains. 

In consequence of a deficiency of water, she blew up in 1839. The 
engine was provided with a copper globe for the purpose of obtaining 
dry steam hence her name li Globe." She had four wheels, each of 
3ft. diameter. 







21. STEPHENSON'S 



PLANET," LIVERPOOL AND MANCHESTER 
RAILWAY 



Stephenson soon put into practice the borrowed idea of inside 
cylinder locomotives, to his own advantage, and on October 4th, 1830, 
ho delivered the first engine of his construction containing inside cylin- 
ders, placed in the smoke-box, as suggested to R. Stephenson by Trevi- 
thick. This locomotive was named the " Planet,"' and was constructed 
for the Liverpool and Manchester Railway. The cylinders 
were llin. diameter, stroke 16in. The boiler was 6ft. 6in. 
long, 3ft. diameter, and contained 129 tubes. She weighed 
eight tons ; the driving wheels were 5ft. diameter, and 
were placed just in front of the fire-box. The leading wheels were 
3ft. diameter, and projected beyond the front of the smoke-box. 



50 EVOLUTION OF THE STEAM LOCOMOTIVE 

The frames were outside the wheels, and were of oak lined with 
iron plates. As the " Planet " embodied several improvements not 
before used in the engines constructed by Stephenson for the Liverpool 
and Manchester Railway, it is natural that the locomotive should be 
able to perform better service than the earlier ones. On Novem- 
ber 23rd, 1830, she conveyed a train from Manchester to Liverpool 
in one hour, including a stop of two minutes for water. 

On December 4th, 1830, the " Planet " (Fig. 21) hauled a mixed train, 
weighing 76 tons without the engine and convoy (tender) from Liver- 
pool to Manchester in two hours thirty-nine minutes' running time. 

Stephenson continued to supply various locomotives to the Liver- 
pool and Manchester Railway with different minor improvements ; thus 
the "Mercury," built in December, 1830, had the outside frame placed 
above the driving axle, an improvement on the " Planet," which had 
the frames below the driving axle. But all these early engines of 
Stephenson were of a very unsatisfactory character. Pambour, writing 
in 1834, says of them: ''When an engine requires any repair, unless 
it be for some trifling accident, it is taken to pieces and a new one is 
constructed, which receives the same name as the first, and in the 
construction of which are made to serve all such parts of the old 
engine as are still capable of being used with advantage. The conse- 
quence of this is that a reconstructed or repaired engine is literally 
a new one. The repairs amount thus to considerable sums, but they 
include also the renewal of the engines." 

The directors of the Liverpool and Manchester Railway soon found 
tho method of working their heavy trains with four or five locomo- 
tives was far from economical, and Stephenson was required to supply 
more powerful engines for the merchandise traffic. He, therefore, 
built the " Samson " and " Goliath." These were only four-wheel 
engines, but all the wheels were made of one size and coupled together. 
The former was delivered in January, 1831, and on February 25th 
she conveyed a train weighing 164 tons (without reckoning the weight 
of engine or tender) from Liverpool to Manchester in two and a half 
hours. The dimensions of the engines were : Cylinders 1 4in. diameter, 
stroke 16in., wheels 4ft. 6in. diameter, heating surface 457.10 sq. ft. 

In 1831, the Directors of the Monkland and Kirkintilloch Railway 
decided to work their line by locomotives, and instructed Mr. Dodd, 
their engineer, to design engines for the purpose. He, however, 
merely adopted the plan used in the construction of the "Locomotion" 
(Stockton and Darlington Railway), with the cylinders placed partly 



EVOLUTION OF THE STEAM LOCOMOTIVE 51 

within the boiler over the wheels, working by means of cross-heads and 
connecting-rods. He also adopted the tubular boiler, which was, of 
course, wanting in the "Locomotion." The engines were con- 
structed by Murdoch and Aitken, of Glasgow, and were the first loco- 
motives built in that city. The first was put to work on May 10th, 
1831, and the second on September 10th, 1831. The boilers of these 
t\vu locomotives were lagged with wood, and metallic packing was 
for the first time employed in connection with the pistons. The 
cylinders were 10 Jin. diameter, stroke 24in., steam pressure 501b. 
The locomotives were supported on four coupled wheels, the coupling- 
rods having ball-and-socket joints at each end. A speed of six miks 
an hour was attained with Dodd's engines, and, although of rough 
design, they were much more economical in fuel and repairs than the 
engines supplied about the same time by Stephenson to the neigh- 
bouring Glasgow and Garnkirk Railway. These latter two engines 
were named the "St. Rollox" and "George Stephenson." Their 
dimensions were as follows : 

Diamete Wheels. Weight in 

of Stroke. working 

cylinders. Driving. Leading. order. 

St. Bollox llin. 14in. 4ft. 6in. se^in. 6 tons. 

George Stephenson ... llin. 16in. 4ft. 6in. 4ft. 6in. 8 tons. 

The gauge of this line was only 4ft. Gin. The "St. Rollox" cost 
tho G. and G. Railway about 750 ; that company sold it to the 
Paisley and Renfrew Railway for 350, and the latter, in December, 
1848, when the gauge of their line was altered, disposed of 
the locomotive by auction for 13. It had wooden wheals. At 
the same auction the other two locomotives of the Paisley and Renfrew 
Railway were also sold, and realised only 20 guineas each, although 
about ten years previously the Paisley and Renfrew Railway had paid 
Murdoch, Aitken, and Co. 1,100 for each of them. They were six- 
wheel tank engines. The Scotch engines we have just been describing, 
all burnt coal in place of coke, and as they caused a good deal of smoke 
they were much objected to on that account. 

We have previously stated that upon the advice of two engineers 
the Directors of the Liverpool and Manchester Railway had refrained 
from purchasing more locomotives from E. Bury, but other people 
soon saw the good points of his engines, and in 1832 the Liverpool 
and Manchester Railway considered it policy to purchase another 
locomotive from the Clarence Foundry. This engine was called the 

E 2 



til EVOLUTION OF THE STEAM LOCOMOTIVE 

"Liver." She had cylinders 11 in. diameter, 16in. stroke, and driving 
wheels 5ft. diameter. The " Liver " worked very successfully, and in 
1836 her fire-box was altered to burn coal, but this experiment turned 
out somewhat of a failure. 

Towards the end of 1831, and during 1832, the increasing traffic on 
the Stockton and Darlington Railway made a considerable increase in 
the number of locomotives necessary. Hackworth designed two new 
classes of engines to work the trains. One type was known as the 
" Majestic " class, and six engines of this description were soon at 
work. 

The " Majestic " locomotives had each six coupled wheels. The 
heating surface was obtained from a tube 9ft. long, 2ft. Gin. diameter, 
one end of which communicated with the fire-grate; the other was 
divided, from the boiler by a partition plate, inserted in which were 
104 copper tubes 4ft. long, and reaching to the smoke-box, ic should 
be observed that the boiler was 13ft. long. The cylinders were fixed 
in a vertical position in front of the smoke-box, the connecting-rods 
working on a straight shaft or axle parallel with the wheel axles : this 
driving shaft was coupled by outside rods to the six wheels. The slid 3 
valves had "lap," and were worked by two eccentrics, which aho worked 
the force pumps. The engine was reversed by means of a single lever. 
This class of engines included : 

" Majestic," built by Hackworth. 
"Coronation," built by Hawthorn. 
. " William the Fourth," built by Hackworth. 

" Northumbrian," built by Hackworth. 
ff Director," built by Stephenson. 
" Lord Brougham," built by Hackworth. 

All of them were built from Hackworth's designs, the leading 
dimensions being : Cylinders, 14Jin. diameter, stroke, 16in. ; boiler, 
13ft. long, 3ft. lOin. diameter; weight of engine empty, 10 J tons; 
full, 11^ tons. The other class of engines designed by Hackworth at 
this time included : 

"Darlington," built by Hawthorn. 
" Shildon," built by Hackworth. 
"Earl Grey," built by Hawthorn. 
"Lord Durham," built by Stephens 
, " Adelaide," built by Stephenson. 

" Wilberforce," built by Hawthorn. 



EVOLUTION OF THE STEAM LOCOMOTIVE 53 

u Wilberforce," an illustration of which is given (Fig. 22), was built 
by Hawthorn, of Newcastle, and commenced to work in 1832 ; it had 
six coupled wheels 4ft. in diameter; the cylinders were 14fin., with 
16in. stroke. Like many of the locomotives of that period, the " Wil- 
berforce," as will be observed, had two tenders, one at each end of trie 
engine. On the tender at the front end, which only carried coals (the 
fire-door being at the chimney end of the engines), the fireman stood ; 
whilst the other tender, at the footplate end, carried water in a barrel, 
and also the tool boxes. The engine wheels were made of two separate 
castings or rings, and the axles were all straight, the crank shaft being 
carried in separate bearings beneath the footplate. There were no 
tail lamps in those early days ; to make up for this deficiency a cresset 
containing burning coal was used. In some cases, when it was neces- 
sary to indicate the destination of the engine, or the section to which 
it belonged, as many as three of these cressets of glowing coals were 
employed on the same locomotive. 




FIG. 22.-" WILBERFORCE," A STOCKTOX & D ARLINGTON RAILWAY LOCOMOTIVE 

On certain favourable gradients the " Wilberforce " was capable of 
taking 36 loaded chaldron wagons, equal to about 171 tons, and its 
coal consumption is given as 681b. per mile. During the year ending 
June, 1839, this engine ran 16,688 miles, conveyed 635,522 tons over 
one mile, and cost 318 10s. 8d., or 4.5d. per mile run, for repairs. 
The wages of the driver and fireman during the same period amounted 
to 353 12s. 8d. 

The engines of this class, in their time, performed a greater amount 
of work than any others then existing. As late as 1846 one of the 
principal officials of the Stockton and Darlington Railway said of 
them : " Take them, weight for weight, they surpass any engine on 
the line." 



EVOLUTION OF THE STEAM LOCOMOTIVE 



The cylinders were 14 Jin. diameter, 16in. stroke; the valve gear- 
ing, wheels, etc., were similar to the " Majestic " class, but the 
cylinders were fixed on a framing extending 6ft. beyond the boiler over 
the driving shaft, which was coupled to the six wheels, each of 4ft. 
diameter. 

The heating surface of the engines was on a different system, to 
" return multitubular ' fire-tube " being employed. This comprised 
a principal tube 8ft. long and 28in. diameter at the fire-grate end, and 
24in. at the other. Here was fixed a D-shaped box ; from this, 
89 copper tubes conveyed the heated air back through the boiler to 
the semi-circular box fixed at the fire-grate end ; the chimney came 
out of this smoke-box extension. These flues proved most economical, 
many lasting as long as six years, and, when necessary, duplicate ones 
could be fixed, and the engine again at work in three days. The boiler 
was 10ft. long and 4ft. 4in. diameter, weight of engine 10 J tons empty, 
11 J tons loaded. 

The "Magnet;," built by Hackworth, at Shildon, in 1832, was 
an improvement on the above. The cylinders were 15in. 
diameter, 16in. stroke. The fire-tube at the furnace end was 
2ft. diameter, and was divided in the middle by a 4in. nre-b v iclj 
partition. The number of return tubes was 110. These were 7ft. Gin. 
long. Hackworth was at this time hauling all the trains on the 
Stockton and Darlington Railway by contract, at the rate of 2-5d. per 
^ ton of goods per mile ; 

\ T afterwards reduced to 

a still lower price. He 
paid the Stockton and 
Darlington Railway in- 
terest at 5 per cent, on 
the cost of locomotives 
employed on the line, 
which were the pro- 
perty of the Stockton 
and Darlington Rail- 

Pia. 23. GALLOWAY'S "CALEDONIAN," BUILT FOR THE ^ 

LIVERPOOL AND MANCHESTER RAILWAY IX 1832 ? Company, but 

leased to him. 

An engine named " Caledonian " (Fig. 23) was supplied to the Liver- 
pool and Manchester Railway in 1832, by Galloway, Borman and Co. 
She had inside frames, four coupled wheels oft. diameter, and n 
domed fire-box. The curious point about the locomotive was 




EVOLUTION OF THE STEAM LOCOMOTIVE 55 

tho location of tKe cylinders, which were placed on the fraino 
in front of the smoke-box, and were fixed vertically, with the piston- 
rods working through the upper cover, connecting-rods working down- 
wards to the leading wheels, the axle of which was below the frames, 
in front of the smoke-box. 

As might be expected, the " Caledonian " was far from being an 
easy-running locomotive, and, after several times running off the 
rails, she was rebuilt with inside cylinders and a crank axle. 



CHAPTER V. 

A Stephenson "bogie" engine for America-- -The genesis of a world-f: imous loco- 
motive film Its initi;u effort in locomotive construction, the "Experiment" 
*Ter cylinder valves Two early Scotch locomotives- -Stephenson favours 6- wheel 
Jngines, and constructs the " Patentea " Forrester's " Swif tsure " Opening 
of the Newcastle ;\nd Carlisle Rwy. The " Comet " R. S:ephenson's early 
"ultimatum," the ''Harvey Combe" Hackworth to the front with a locomo- 
tive novelty The first locomotive in Russia The "Goliath" The " Tyne " 

and her steam organ Other early Newcastle and Carlisle Rwy. engines An 

engine driver's reminiscences No eight hours day then The "Michael Long- 
ndg-3" Opening of the l*i'and Junction Rwy. Its first, locomotives. 

R. Stephenson and Co., in 1833, constructed a locomotive for the 
Saratoga and Schenectady Rail Road of America, which deserves 
mention from the fact that it had a leading bogie, rendered 
necessary because of the sharp curves on the Saratoga and 
Schenectady Rail Road. R. Stephenson named this locomo- 
tive the "Bt)gie," because the low wagons used, on the quarries at 
Newcastle were locally called "bogies," and it was from these vehicles 
that he developed the idea. of providing a small truck to carry the 
leading end of the locomotive in question. Ever since 1833 the 
swivelling* truck used for supporting locomotives and other railway 
rolling stock has, in England, been designated the "bogie." 

Richard Roberts, of the firm of Sharp, Roberts and Co. (the 
predecessors of Sharp, Stewart and Co., Limited), in the year 1833, 
turned, his attention to locomotive construction. His initial effort 
was of a somewhat novel kind. Four locomotives of his first design 
were constructed, one "Experiment" for the Liverpool and Man- 
chester Railway, and the others for the Dublin and Kingstown Railway. 
The cylinders, which were llin. diameter, were placed in a vertical posi- 
tion on the frames, just at the point were the boiler entered. the 
smoke-box. By means of cross-heads and side-links the motion was 
conveyed to a bell crank, and so transmitted by a connecting-rod to 
the driving wheels. There was, of course, a similar arrangement of 
cylinder, crank, etc., on both sides of the engine. The stroke was 
16in. The driving wheels, 5ft. in diameter, were placed in front of 
the fire-box, and had insude bearings; the leading wheels were 
located below the vertical cylinders, and had outside bearings. The 



EVOLUTION OF THE STEAM LOCOMOTIVE 



57 



pump was placed in a horizontal position above the frame over the 
driving wheels, and was worked by a rod actuated by the vertical 
member of the bell-crank. 

The ''Experiment" (Fig. 24) was unsuccessful, and was rebuilt, when 
:\ third pair of wheels was added, and the position of the cylinders r 
bell-crank, etc., altered. The valves were also of a novel kind, 
patented by Mr. Roberts in 1832. Colburn thus describes them : 
" The valve, of wrought-iron, was formed of two concentric tubes or 
pipes, the larger pipe having holes perforated to admit steam from 
the steam-pipe into the annular space. This annular space was closed 
steam-tight at each end of the valve, and steam could only escape 
from it alternately to each end of the cylinder through the slots. The 
exhaust steam passed from one end of the cylinder directly into the 
waste pipe, and from the other end it traversed the interior of the 



>( 







FIG. 24. ROBERTS'S " EXPERIMENT," WITH VERTICAL CYLINDERS, 
BELL-CRANK, CONNECTING-ROD, AND CYLINDER VALVES 

pipe of the cylindrical valve. These valves did not work well, as they 
did not expand equally with their cast-iron casings when heated by 
steam. For this reason the cylinder valves were soon abandoned. 
It should be mentioned that, in Mr. Roberts' first engines, the valve 
for each cylinder was worked with a motion derived from the opposite 
side of the engine. No eccentrics were employed, the requisite motion 
being taken from a pin ne.ar the fulcrum of each bell-crank, and trans- 
mitted thence through suitable gearing to the valve attached to the 
cylinder on the opposite side of the engine." 

The engines used on the Dundee and Newtyle Railway, con- 
structed in 1833, partook somewhat of the character of Roberta's 
" Experiment," inasmuch that right-angled cranks and vertical cylin- 
ders were employed, the diameter of the latter being llin., and 



58 EVOLUTION OF THE STEAM LOCOMOTIVE 

stroke 18in. These engines were named "Earl of Airlie" and "Lord 
Wharncliffe," and were constructed by J. and C. C'armichael, of Dun- 
dee. Both these engines were delivered at the end of September, 1833. 
The "single" driving wheels were placed in the leading position, 
the axle being just behind the smoke-box. The cylinders were placed 
on the side frames, about midway between the two ends. 

The piston-rods worked upwards, and the motion was conveyed by 
means of rods from the piston cross-heads. These connecting-rods 
passed down outside the pistons, and were connected to one end of the 
bell-cranks, which were fixed beyond the cylinders, with the pivots 
over the centre of the second pair of wheels. From the lower ends of 
the bell-cranks the driving-rods were pivoted, the other ends being 
connected to the outside cranks of the driving wheels. The fire-box 
end of the engines was supported on a four-wheel truck or bogie. 
These engines weighed 9J tons each, and cost 700 each. An 
ordinary four-wheel wagon, fitted with a water-butt, was used for a 
tender. 

An engine of similar design was ordered from Stirling and Co., 
of the East Foundry, Dundee, and delivered on March 3rd, 1834. 

Mr. A. Sturrock, the first manager of Swindon Works, and after- 
wards locomotive superintendent of the Great Northern Railway, 
helped to construct this engine, which was named " Trotter." Mr. 
Sturrock was at the time an apprentice at the East Foundry. 

The gauge of the Dundee and Newtyle Railway was only 4ft. 6in., 
but when the line was taken over by the Dundee and Perth Railway 
the gauge was altered to the normal gauge of Great Britain. The 
original engine, " Earl of Airlie," after some alteration, of course, 
could not run on the railway, but for some years after the change the 
" Earl of Airlie " was employed as a stationary pumping engine. 

Stephensons four-wheel passenger engines with a short wheel 
base were found to be very unsteady at the very moderate speeds 
then attained, and he, therefore, added a pair of trailing wheels, 
thus constructing a six-wheel " single " passenger engine. Stephen- 
son considered that the moderate wheel base of these small engines 
with six wheels would, on the easy curves of the Liverpool and 
Manchester Railway, offer considerable resistance, so he took out a 
patent, in which Jie provided that the middle or driving pair of 
wheels should be without flanges (or flanches, as they were then 
called). He claimed that by this modification the six-wheel passon- 
ger engine would pass round curves with much less strain and 



EVOLUTION OF THE STEAM LOCOMOTIVE 



59 



greater safety. The first engine so constructed by Stephenson he 
designated the "Patentee," and she was delivered to the Liverpool 
and Manchester Railway in January, 1834. She had outside frames, 
inside cylinders, 18in. stroke, 12in. diameter; the driving wheels 
were 5ft. diameter. 

George Forrester and Co., Vauxhall Foundry, Liverpool, 
in 1834 constructed a six-wheel engine named " Swiftsure." 
This locomotive possessed many novel features. It had outside 
horizontal cylinders; the frames were also outside, thus making the 
cylinders a considerable distance apart. The connecting-rods were 
keyed on cranks, at some distance outside the frames, whilst the 
Tact that the driving wheels were not counterbalanced caused the 
engines of this class to be most unsteady at even moderate speeds, 

gfn n nti( 




FIG. 25. HAWTHORN'S " COMET," THE FIRST ENGINE OF THE NEWCASTLE 
AND CARLISLE RAILWAY, 1835 

and they were soon known by the sobriquet of " Boxers." Colburn 
says: "A few pounds of iron properly disposed in the rims of iho 
driving wheels would have redeemed the reputation of these engines." 
The arrangement of cylinders and frames allowed the leading wheels 
to be placed well forward, the total length of the frames of the 
" Swiftsure " being 1 7ft. The driving wheels were 5ft. diameter, 
and the cylinders llin. ; the stroke was 18in. 

In the " Boxer' 7 Forrester employed his patent valve gear, witli 
vertical gab ends and four eccentrics. 

A portion of the Newcastle and Carlisle Railway was opened 
March 9th, 1835, and R, and W. Hawthorn constructed the first 



60 EVOLUTION OF THE STMAM LOCOMOTIVE 

engines for that railway. No. 1 was the " Comet," (Fig. 25), a four- 
wheel (coupled) locomotive; the cylinders (12in. diameter, 16in. stroke x > 
were placed below the smoke-box, the connecting-rods passing 
under the leading axle. The wheels were 4ft. diameter. 
Hawthorn's valve _gear was used in the engines of this 
class, which was actuated by four fixed eccentrics. The 
" Comet " .continued to work on the Newcastle and Carlisle 
Railway for a number of years, and was afterwards used as a 
stationary engine for driving the steam saws at the Forth Bank 
Engine Works, Newcastle. She was so engaged up to and subse- 
quently to 1863. 

About 1836 short-stroke locomotives came into favour, 
and Tayleur and Co. built ten for the Liverpool and Man- 
chester Railway. Although the cylinders were 14in. diameter, the 
stroke was only 12in. We need scarcely add the experiment wr.b 
not successful, although some of the original broad-gauge engines 
were built with short strokes. These will, however, be dealt with 
fully later on. 

In 1836 R. Stephenson and Co. constructed the "Harvey Combe" 
locomotive. She was a ballast engine, and was engaged in the con- 
struction of the London and Birmingham Railway. R. Stephenson 
had a minute description of this engine written by W. P. Marshall, 
and the work in question is stated to be " the most perspicuous and 
the illustrations of the most elaborate kind of any work describing a 
locomotive." 

The fact that at once strikes the intelligent reader as peculiar 
is that, although the " Harvey Combe " was designed " for conveying 
the earth excavated in the construction of a line of railway," as 
Marshall "perspicuously" puts it (but which we should shortly 
describe as a "ballast" engine), she is a "single" engine! and, 
therefore, is not much like a modern six-coupled ballast engine. 
She cost 1,4:00, and was of 50 horse-power. 

The principal dimensions of the " Harvey Combe " were : Cylin- 
ders, 12in. by 18in. ; driving wheels, 5ft., and leading and trailing 
3ft. 6in. diameter ; 102 tubes, If in. internal diameter ; total heat- 
ing surface, 480ft.; weight, empty, 10 tons; with fuel and water, 
11 tons 18 cwt. No flanges were provided to the driving wheels. 
Although the " Harvey Combe " was built for, and had rough usage 
as, a ballast engine, yet, when at the end of 1837 Nicholas Wood 
was making experiments for the purposes of his report to the 



EVOLUTION OF THE STEAM LOCOMOTIVE 61 

Great Western Railway as to the broad-gauge, the "Harvey Combe" 
was the principal narrow-gauge engine with which he experimented. 
With a gross load (including engine, etc.) of 81 tons, she attained a 
speed of 25 to 53 miles an hour, and consumed 0.471b. of coke per 
tor. per mile. With a gross load of 50 tons the speed reached was only 
32.88 miles an hour, with the above coal consumption. 

In 1836, Hackworth built a locomotive of novel construction 
viz., with double-acting, ram or trunk engines, by means of which 
piston-rods were dispensed with, the connecting rods being pivoted 
directly on to the piston and oscillated within the trunk. 

This was the first locomotive engine ever seen in Russia. She 
commenced work on the Zarskoe-Selo Railway on November 18th, 
ISoo, a religious service being held and the t locomotive consecrated 
before the first train was run. Of this engine the Russian Emperor 
remarked ir. English, "It is the finest I ever saw." An old officer of the 
Stockton and Darlington Railway, informs the writer that a locomotive 
or. the double-acting trunk principle was also built by Hackworth 
for that line, and so far as his memory serves him, he believes it was 
the " Arrow " passenger engine. The " Arrow " had leading and trail- 
ing wheels 3ft. Gin. diameter; driving wheels, 5ft. Gin. diameter; 135 
lubes in the boiler of If in. diameter ; cylinders, 20in. in diameter, and 
with p, stroke of only 9in. ! 

We have already mentioned the first engine (the " Comet ") sup- 
plied to the Newcastle and Carlisle Railway, but several of the other 
early locomotives used on - that line were powerful ones, and their 
design in advance of the generality of locomotives then in use. Thu<?, 
the "Goliath," one of the. first engines supplied to the line by Haw- 
thorn, in March, 1837, hauled a train consisting of 63 wagons of 
coal, weighing 267 tons, 12 miles in^less than 40 minutes. 

The "Goliath" had six-coupled wheels 4ft. diameter, cylinders 
14in. diameter, 18in. stroke. Total heating surface 550.91 sq. ft. 
Weight, empty, 11| tons; in working order, 13 tons. The "Atlas/ 
built by R. Stephenson and Co. in 1836, drew a train of 100 wagons, 
loaded with coal, coke, and lime, and weighing 450 tons, lOf miles 
in 45 minutes, but this was on a falling gradient, varying from 1 in 
215 to 1 in 106. This locomotive was also six-coupled, the wheels 
being 4ft. diameter; cylinders, 14in. by 18in. stroke; heating surface, 
553.77 sq. ft. , weighing 10 tons 6 cwt. empty, and 11 tons 6f cwt. in 
working trim. Another small locomotive on the Newcastle and 
Carlisle Railway, named " Tyne," built by Hawthorn, is worthy 



62 EVOLUTION OF THE STEAM LOCOMOTIVE 

oi notice, for the reason that the first steam organ was fitted to the 
engine. This was the invention of the Rev. James Birket, of Ovingham. 
It was fixed on the top of the fire-box, and was thus described : 
''The organ consists of eight pipes, tuned to compass an octave, but 
without any intervening tones or semi-tones. This is the first attempt 
to adapt a musical instrument to the steam engine capable of producing 
a tune, and though not so perfect as to admit of all the pleasing 
variety and combination of sound capable of being produced by the 
instrument to which we have compared it, there is no doubt but 
very considerable improvements will be made in this steam musical 
instrument by the inventor, who is a skilful musician as well as an 
ingenious mechanic." 

The "Tyne" had cylinders 13 Jin. by 16in. stroke, and four wheels, 
4ft. Gin. diameter; she weighed only 9J tons. After working for many 
years, a pair of trailing wheels 3ft. Gin. diameter was added, thus 
making her a six-wheel engine, with the leading and driving wheels 
coupled. She continued to work on the Newcastle and Carlisle 
Hallway till the end of 1857, when she was sold, but even at that 
time the "Tyne" was in good working order. Three other old 
locomotives were sold at the time viz. , " Eden," " Meteor," and 
" Lightning." 

The "Eden" was built by R. Stephenson and Co. in 1836, and had 
four-coupled wheels of 4ft. Gin. diameter, and a third pair 3ft. Gin. 
diameter; cylinders, 14in. by 15in. stroke, afterwards increased to 
IGin. stroke. Weight, empty, 10 tons, 6 cwt. 

The "Meteor" was built by Bury and Co., of Liverpool, and had 
only four wheels of 4ft. diameter; cylinders, 12in. diameter. The 
stroke at first was 15in., but afterwards was made IGin. Steam 
pressure, 551b. She was provided with hand gear, the slide valves 
working into the front of the steam chest by means of weight bars 
located between the front buffer beam and the smoke-box end. The 
piston connecting-rods, of course, actuated the rear axle, but the 
eccentric sheaves were upon the leading axle, so that if the crank pins 
upon which the side rods worked went a bit loose, the side rods had 
to be disconnected, and the valves worked by the gear handles. This 
was rather hard work for the driver and fireman, who, upon such occa- 
sions, took it in turns to thus work the valve gear. This Bury loco- 
motive opened the line from Blaydon to Newcastle on Sunday, October 
31st, 1839. The man who was fireman on this engine at that time 



EVOLUTION OF THE STEAM LOCOMOTIVE 63 

thus relates his experiences: ''The 'Meteor' engine was sent to 
Redheugh Station to work the passenger trains between that station 
and Blaydon, also coal trains and other things, with this tiny engino 
oi about eleven tons all told. We formed the connection at Blay- 
don with all trains to and from the west. For this new 
arrangement of running I was to be called out of bed by a 
watchman close after two o'clock each morning, to gather up my 
fire bars, put them into the box, and get a fire as best I could a& 
usual, and have steam ready by 5 a.m. to take our first train from 
Gateshead to Blaydon at 5.20 a.m. I had also to clean most of the 
little engine, the driver doing part. I had to clean up the shed, take 
all ashes out, coke the tender, etc. To turn the engine the tender 
had to be taken off, and pushed on one side to get past it, and 
reunited as often as we made a short trip. There is nothing like it 
in the divorce court. For this work my pay was 2s. 8d. per day r 
commencing at 5 a.m., when my driver made his appearance, little 
overtime being allowed, and we did well to finish by 8.45 p.m. I 
worked about 18 J hours daily, with one exception, weekly, and on 
this particular time we had our boiler to clean out, and had to fill 
by hand buckets this after our train work was finished. Water 
being a little scarce in the shed, it was frequently necessary to haul 
out of the river Tyne and carry to the shed, and pour into the boiler 
by the safety valve or man-hole by the driver, the fireman having the 
honour of carrying it from the river quay. 

"This work took so much labour and time that our only rest on 
that particular night and morning was upon the soft side of a plank 
while the steam was rising in the engine boiler, to leave for Blaydon 
at 5.20 a.m. with our usual first train. Then we were again at work 
until 8.45 p.m. There was not a guard for our passenger train, so 
I had the closing of the carriage doors, etc., to attend to, to fill up my 
spare time, and to keep myself awake. We had to load coals during 
part of the day from Wylam, etc., to Dunston, so that there was not 
much fear of falling asleep. I was coupler and guard for this work. 
When not otherwise engaged I had my cleaning to attend to, and 
tubes to keep clean daily, so I was really never committed for going 
to sleep during working hours. I was at this work over the winter 
almost the whole of 1839-40, when early one morning I had a fall from 
the boiler top in the shed, and came down the wrong end first. J 
injured one shoulder very much, which laid me off work one month. 



64 EVOLUTION OF THE STEAM LOCOMOTIVE 

I kept at work all the day after falling, but only one arm was of any 
use to me, and I was compelled to give up. 

"A bone-setter in North Shields had to do the needful for me, as 
they have often had to do for others before and afterwards." 

The " Lightning" was an engine with dimensions similar to those 
of the " Eden," previously described. 

Longridge and Co., of Bedlington, supplied the Stanhope and 
Tyne Kailway in 1837 with a very powerful locomotive named the 
"Michael Longridge." She had six coupled wheels, 4ft. diameter; 
cylinders, 14in. diameter; and a stroke of 18in. 

The Grand Junction Railway was opened in July, 1837, and 
R. Stephenson and Co. (together with other builders), supplied the 
original locomotives. Stephenson's engines at this time had become 
a little more dependable, for we find it Chronicled that three of them 
which had run uninterruptedly since they were first employed had, 
.between July 8th and September 30th, 1837, accomplished the follow- 
ing distances viz : the "Wildfire," 11,865 miles; " Shark," 10,018 
miles; and " Scorpion," 11,137 miles; and, moreover, they were then 




FIG. 26. "SUNBEAM," BUILT BY HAWTHORN FOR THE STOCKTON AND 
DARLINGTON RAILWAY 

still running in perfect working condition. They were six-wheel loco- 
motives, with leading and trailing wheels 3ft. Gin. diameter, driving, 
5ft. diameter; cylinders, 12 Jin. by 18in. stroke; weighing in work- 
ing order 9 tons 12 cwt. 

In 1837, No. 43, of the Stockton and Darlington Railway, 
the "Sunbeam" (Fig. 26) was turned out by Hawthorn. It was 
a "single" engine, having driving wheels 5ft. in diameter, 
and cylinders 12in. in diameter, with 18in. stroke. The 
"Sunbeam" worked well for 19 years, and in 1863 was re- 
ported as being " still in good working order, but too small for the 



EVOLUTION OF THE STEAM LOCOMOTIVE 



65 



present heavy traffic. The boiler of the " Sunbeam " was 8ft. long 
by 3ft. 2in. in diameter, and contained 104 copper tubes. The "Dart/' 
No. 4J, was built by Hackworth in 1840, at Shildon, and was a four- 
wheeled, engine, the wheels being 4ft. Gin. in diameter. The boiler, 
containing 122 tubes, was'.8ft..2in. long and 3ft Sin. in diameter. The 
fire-box was 4ft. high, 3ft. lOin. long, and 3ft. wide. The boiler 
pressure was lOOlb., and the heating surface of the engine 602 square 
feet; the cylinders were 14in. in diameter, and the stroke 16in. 
The extreme length of the engine and tender was 35ft. 3in., and the 
regular speed attained is said to have been thirty miles an hear. 



CHAPTER VI. 

A.n importune epoch, in locomotive history. The first brcad-gauge engines. 
Absurd incorrect statement* regarding these locomotives. The facts concerning 
same; extracts from directois' report. Brunei and the eiigina bui ders. 
The delivery of the first engines to the Great Western Baihva/. -Further 
extract from the directors' repoil Daniel Gooch appears on the scene- -Trial 
of the broad-gauge engines. Table of the original Great Western tngine-. The 
" Vulcan " " ^Eolus " " Bacchus " " Venus " " Apollo " " Mars '' 
and " Ajax," 10ft. -wheel engines. The builders' account of one of these giants. 
"Ajax," a sister engine. 10ft. disc wheels. Dr. Lardner. The "boat" 
engines. T. B. Crainplon and the "Ajax." The "Ariel." "Atlas." 
" Hurricane," a locomotive monstrosity with 10ft. driving wheels. The " Thun- 
derer," a geared engine on Harrison's system Gooch' s opinion of these two 

curious locomotives. The Haigh Foundry geared engines, described by an 
eye-witness. Table showing results of trials with the original broad-gauge 
enginos. The last of "Lion," "Planet," and "Apollo." 

WE have now come to an important era in the evolution of the 
steam locomotive viz., the first appearance in the arena of broad, 
or 7ft., gauge locomotives. Readers are probably aware that very 
much has been written on the subject of the early Great Western 
Railway locomotives during the past few years, and a surprising 
lack of knowledge of the subject has been exhibited by people taking 
part in discussions that have arisen. The facts are clearly estab- 
lished, so that it would be waste of time to recapitulate the many 
inaccurate statements that have been made relative to the original 
broad-gauge locomotives. Thus we read that "the first portion of 
the Great Western Railway was opened in 1837," also that "Mr. 
Brunei designed the 'Hurricane.'" These statements are, of course, 
utterly at variance with the facts, but they prepare one for yet more 
extraordinary statements on the same subject, such as "the directors 
of the Great Western Railway having appointed Mr. (afterwards Sir 
Daniel) Gooch as locomotive superintendent, the duty devolved upon 
him to design and provide the necessary engines. Mr. Gooch, having 
inspected all the locomotives on other railways, considered that 
5ft. Gin. wheels were far too small; he therefore designed the 
engines for the Great Western with driving wheels of 6ft., 7ft., and 
8ft. diameter, and placed orders for their construction with the 
leading builders of that time." 



EVOLUTION OF THE STEAM LOCOMOTIVE 67 

To commence with, therefore, it will be as well to give the exact 
particulars as to the ordering and delivery of the original broad-gauge 
locomotives, for the opening of the first portion of the Great Western 
Railway. 

The facts as given in the directors' reports to the shareholders, 
stated at the meetings of the shareholders, or mentioned in the 
various reports of Brunei, Wood, and Hawkshaw, are as follows : 
The first locomotive engines were ordered prior to August, 1836. The 
directors in their report of that date thus mentioned them: "Diffi- 
culties and objections were at first supposed by some persons to 
exist in the construction of engines for this increased width of rails, 
but the directors have pleasure in stating that several of the most 
experienced locomotive engine manufacturers in the North have 
undertaken to construct these, and several are now contracted for, 
adapted to the peculiar track and dimensions of this railway, cal- 
culated for a minimum velocity of thirty miles an hour." 

Instead of the builders having personal interviews to obtain 
orders for engines, as has been recently stated, it appears from 
Brunei's report of August, 1838, that he "left the form of construc- 
tion and the proportions entirely to the manufacturers, stipulating 
merely that they should submit detailed drawings to me for my 
approval. This was the substance of my circular, which, with your 
sanction, was sent to several of the most experienced manufacturers. 
Most of these manufacturers, of their own accord, and without pre- 
vious communication with me, adopted the large wheels as a 
necessary consequence of the speed required. As it has been sup- 
posed that the manufacturers may have been compelled or induced 
by me to adopt certain modes of construction, or certain dimensions 
in other parts, by a specification a practice which has been adopted 
on some lines and that these restrictions may have embarrassed 
them, I should wish to take this opportunity to state distinctly that 
such is not the case." 

Then, as to the delivery of the engines, from the directors' report 
it is clear that on August 12th, 1838, eleven locomotives were actually 
on the line. According to a statement drawn up by Mr. C. A. 
Saunders, the superintendent of the Great Western Railway, for the 
purposes of Mr. N. Wood's report, the following engines were then 
in use on the railway : " North Star," "^Eolus," "Venus," "Nep- 
tune," "Apollo," "Premier," and "Lion." This leaves four erginesto 
be accounted for. Sir Daniel Gooch states that the six engines built 



F 2 



68 EVOLUTION OF THE STEAM LOCOMOTIVE 

by the Vulcan Foundry Company could be depended upon. We caa, 
therefore, take it for granted that the " Vulcan " and " Bacchus " 
were two of the four, whilst the geared " Thunderer " was delivered 
before April 26th, 1838, and the "Ariel" before June 1st, 1838. 

The directors stated that the railway company had only accepted 
eight of these engines, and the three others required alterations 
before the engineer would accept them. 

This report continues with the following significant paragraph: 
" The directors are under the necessity of declining to receive two 
engines made for them, in consequence of a material variation in the 
plan of them since it was submitted to and approved by their 
engineer." These two engines may be the "Ajax" and her sister 
10ft. wheel engine, the "Mars," constructed by Mather, 
Dixon and Co., or the two geared engines built by the Haigh Foundry 
Company ; although it is probable that the two latter engines had 
not been delivered at this date. Besides the eleven engines already 
on the line, and the two refused by the engineer, the directors 
stated that nineteen others were then in course of construction, 
making a total of thirty engines. Of the seven engines mentioned 
as being in use on the line, according to Mr. Brunei, only four were* 
really used for the passenger service, the fifth being kept with 
steam up to take the place of one of the other four in case of a 
breakdown, and the other two were used for conveying ballast, etc., 
for the construction of the line. According to Hawkshaw's report, 
dated October 4th, 1838, fourteen engines had at that time been 
delivered to the Great Western Railway, and seven more were 
approaching completion, the nine remaining to complete the thirty 
not having then been put in hand. Mr. Daniel Gooch commenced 
his duties as locomotive superintendent of the Great Western Rail- 
way on August 18th, 1837. At this period the following engines 
had been ordered for the Great Western Railway : Six from the 
Vulcan Foundry, where Gooch had served under Stephenson; four 
from Mather, Dixon and Co., Liverpool; two from Hawthorn and 
Co., Newcastle : two from the Haigh Foundry Company, and, curiously, 
two from R. Stephenson and Co. 

Mr. Gooch states in his " diaries " that these two engines were 
constructed for a Russian railway with a 6ft. gauge, and that he 
himself prepared the working drawings from which they were 
constructed. There, however, appears to be some doubt as 



EVOLUTION OF TffE STEAM LOCOMOTIVE 69 

to whether it was a Russian or American railway for which the 
two locomotives in question were originally built. When ready for 
delivery the purchase money was not forthcoming, so the careful 
firm of R. Stephenspn and Co. did not part with the "North Star" 
and her sister engine. They afterwards widened the frames, fitted 
longer axles to the two locomotives, and then sold them to the Great 
Western Railway as 7ft. gauge engines. 

The " Vulcan," built by the Vulcan Foundry Company, was the 
first engine delivered to the Great Western Railway. One of Mather 
r.nd Dixon's 10ft. wheel engines arrived a few days after, 
having been sent ' by sea from Liverpool to Bristol in 
December, 1837, and forwarded by canal from Bristol to 
West Dray ton. A preliminary trial of these two engines was made 
on Wednesday, January 18th, 1838, and the following extract 
details the working of the two locomotives on this occasion: "A 
full trial was made during the whole of Wednesday in running the 
engines on two or three miles of the line near West Drayton, between 
London and Maidenhead. The object of the trial was to prove the 
rails, and most satisfactory was the result, both as to the increased 
width of gauge "and the use of continuous bearers of kyanised wood 
confined by piles, on which plan the line is constructed. An 
engine with 8ft. drawing wheels, made by Messrs. Tayleur and 
Co., Warrington, weight 23 tons, with the tender, water, coke, 
etc., and another engine made by Messrs. Mather, Dixon and Co., 
weight about 19 tons, with the tender, etc., ran the whole day with- 
out producing the slightest vibration either in the rails or the wood 
under them. The rails are, in fact, so beautifully firm, smooth, and 
true, that the engines glided over them more like a shuttle through a 
loom or an arrow out of a bow than like the effect on any previous 
railway. There is literally no noise no apparent effort nor can 
there ever be discovered any difference between the centre and the 
joint in the rails. A maximum speed was not attempted, as on so 
short a piece the momentum would be no sooner attained than it 
would require to be lowered, in preparation for stopping the engine. 
A speed of forty-five to fifty miles an hour was attained, -and when 
the engines are run, as they will be, either next or the following 
week, on an eight or ten-mile length, there is no doubt they will as 
easily run at a very much greater speed." 

Tho following table gives pr.rticulars of the original locomotives 



70 



EVOLUTION OF THE STEAM LOCOMOTIVE 



as supplied to the Great Western Railway. These engines were 
ordered by Brunei before Sir D. Gooch was appointed Locomotive 
Superintendent ; the first duty of the latter was to inspect these 
locomotives, then in course of construction, and he was not at all 
pleased with their dimensions : 



Builder. 


Name of Engine. 


Diameter 
of Driving 
Wheels. 


Diameter 
of 
Cylinder. 


Stroke. 


Grate 
area. 


Heating Surface. 


Tubes 


Fire- 
box. 


Total 


Vulcan Foundry Co. 

Mather, Dixon & Co. 

B. Stephenson & Co. 
Sharp,Roberts,<fcCo. 
Haigh Foundry Co. 
Hawthorn & Co 


Vulcan 


feet. 

8 
8 
8 

8 
8 
8 
10H 
10 
7 
7 

i 

6 
6 in. 
6 4 
6 4 
6 
10 


inches. 

14 
14 
14 
12 
12 
1-2 
14 
14 
14 
14 
16 
16 
14 
14 

16 
16 


inches. 

16 
16 
16 
16 
16 
16 
) 10 or 
j 11 
14 
14 
16 
16 
18 
18 

20 
20 


feet. 
9-58 

10-22 
10-22 

13-0 
13-0 

17-12 
17-1* 


sq. ft. 

534 
530 
530 
458 
458 
45S 
417 
417 
326 
326 
654 
651 
427 
427 

515 
515 


sq. ft. 

35-0 
57-15 
57-15 
52-35 
52 -35 
52-35 
57-3 
57-3 
51-71 
51-71 
70-10 
70-10 
51-17 
51-17 

108-26 
108-26 


sq, ft 

5 89 A 
587 A 
587 A 
510 B 
510 B 
510 B 
474 G 
474 C 
377 D 
377 D 
724 E 
705 
478 
478 
F 
F 
623 G 
623 G 


JEolus 
Bacchus 
Venus 
Neptune .. .. 
Apollo 
Mars 
Ajax 
Premier 
Ariel 


North Star .. 
Morning Star 
Lion 
Atlas 


Viper ...) 
Snake ...j" J 
Thunderer 
llurricane 



A. These engines had the driving axles above the frames. 

B. Gooch, N. Wood, Whishaw,(J. A Saunders, Z. Colburn, and other reliable authorities all state 

that these three engines had cylinders 12 inches in diameter.* 

C. J. Locke in his evidence before the Gauge Commissioners in 1846 stated that the stroke of 

these engines was only 10 or 11 inches. 

D. It will be noted that these two engines, also built by Mather, Dixon and Co., had very short 

strokes. 
B. Stroke was afterwards increased to 18 inches. 

F. These engines were geared, so that the driving wheels were equal to 12 feet diameter. 

G. The engines and boilers were on separate carriages. The "Thunderer" was geared up 3 to 1. 
H. Although the "Mars"\vas built with 10ft. driving -wheels, it is probable that the size was 

reduced after her trial trips on the G.W.E. After running 10,000 miles the G.W.R. Co. 
sold the "Mars." 

J. It is not certain that the " Viper" and ' l Snake" were the geared engines built by the Haigh 
Foundry Co., but they are generally accepted as such. 

The "Vulcan," it would seem, was a conspicuous failure. The 
Great Western Railway officials did not consider her good enough to 
be used in the experiments made during the autumn of 1838 for the 
purpose of Nicholas Wood's report to the" Great Western Railway in 
connection with the gauge controversy. Whishaw only gives an 
account of one trip to West Drayton and back with the l< Vulcan." 
This was made on the 12th of August, 1839, when, with a load of 
18 tons, she attained a speed of 50 miles an hour on a falling gradient, 

* There is some question as to this, as at one time the dumeter of the cylinders was 14 inches. 






EVOLUTION OF THE STEAM LOCOMOTIVE 71 

the average speed for the trip of 13 miles being 28.32 miles an hour. 
On the return trip, with a load of only 14 J tons, the average speed 
was only 21 miles an hour. The "Vulcan" was afterwards converted 
into a tank engine, and worked the traffic on a branch line for a few 
years. 

The " ^Eolus " appears to have been a somewhat better engine than 
her sister (although, by the way, Sir D. Gooch states that, excluding 
the " North Star," the engines from the Vulcan Foundry were the only 
once he coudld depend upon). N. Wood, in his tables, states that 
".Eolus" was capable of hauling 32 tons at fifty miles an hour, with 
a consumption of 0.761b. of coke per ton per mile, the water evapor- 
ated in an hour being 115-3 cubic feet. The greatest load drawn by 
"^Eolus" during N. Wood's experiments was 104 tons, the speed 
attained being 23 miles an hour, and the consumption of coke .301b. 
per ton per mile. Whishaw details four experiments with this 
engine, the most successful being on November 6th, 1838, when with 
a load of about 20 tons she attained an average speed of 31.39 miles 
an hour; the maximum on this occasion being 48 miles an hour. 
Whishaw's remarks concerning another journey are worth repeating. 
It was on July ,21st, 1838, when "^Eolus" took a train consisting of 
three first-class carriages, two open and one closed second-class 
carriages, and two stage coaches on trucks, or a load of 96,1641b., or 
about 43 tons, and essayed a trip to Maidenhead ; but " after about two 
and a half miles the train was suddenly stopped, and remained in 
statu qwo for 21 f minutes. In the meantime, 'zEolus' moved slowly 
away to recover her strength, and having sufficiently exercised herself, 
returned after a lapse of 21 f minutes to lead the train forward"; 
but the engine did not appear to have quite recovered her strength by 
this exercise ( !) for she had to stop at Slough, where she took water. 
This journey took 150 minutes to complete; but, deducting the 
34 minutes spent in four stoppages, the average travelling rate was 
11.71 miles per hour. 

On January llth, 1840, the "^Eolus" is stated to have made a 
remarkable trip. At this time certain Chartists were being tried 
at Monmouth, and the Dispatch, a Democratic Sunday paper, pub- 
lished detailed reports of the trial. Special messengers were 
despatched by road from Monmouth to Maidenhead, where an engine 
(the " JEolus ") was engaged to carry the messengers to London. She 
is said to have covered the first ten miles in seven minutes, or at the 
rate of 85 miles an hour. Here the preceding train was overtaken, 



72 EVOLUTION OF THE STEAM LOCOMOTIVE 

and the whole journey of 31 miles was completed in about 
minutes. 

Whishaw records a trip with "Bacchus" on December 13th, 1839, 
when, with a train of two second and one first-class (four-wheel) 
carriages, she covered 13 miles at an average speed of 29 miles an 
hour, the highest speed attained on the trip being 44.11 miles an 
hour. On January 9th, 1840, Whishaw made a trip to West Drayton 
and back with the "Bacchus." On the down journey, with a load 
of three coaches, 50 miles an hour was attained. On the up trip a 
similar maximum speed was attained three times, twice for a distance 
of a quarter-mile, and once for a half-mile. 

With "Venus" Whishaw records one experiment with a load of 
25J tons, made up of one open second-class, one first-class carriage, 
and two stage coaches on trucks. The average speed was 21 miles 
an hour, the highest being 48 miles an hour. The "Venus" was 
not much used during the first four months following the first open- 
ing of the Great Western Railway, her total mileage during that 
period being only 240 miles. Mr. Gooch found this engine was so 
extremely unsteady that he did not make use of her, save when no 
other engine was conveniently available hence her small mileage. 
The " Venus " was afterwards rebuilt as a tank engine, and her 
driving wheels reduced to 6ft. in diameter. When so rebuilt she 
worked the Tiverton branch traffic for some years. 

The "Apollo" drew the first up-train on the Great Western Rail- 
way, leaving Maidenhead for Paddington at 8 a.m. on June 4th r 
1838; whilst the next day, when leaving Maidenhead with the after- 
noon train of 13 carriages, she broke down, in consequence of a tube 
bursting, the train being delayed for some hours, and great excite- 
ment being caused in London consequent upon the exaggerated reports 
>f the mishap. 

It will be noticed that in the table of the original Great Western 
Railway locomotives we have given the diameter of the cylinders of 
"Venus," "Neptune," and "Apollo" as 12in., and we have also 
given the names of several men (whose probity is unimpeachable) as 
our authorities on the point. Nor is that all the weight of evidence 
in favour of 12in. being the original diameter of the cylinders. N. 
Wood, in his report to the Great Western Railway directors, specially 
refers to the point, thus :".... The performance of engines, 
such as 'Venus/ 'Neptune,' and 'Apollo/ with 12in. cylinders." This 
is in addition to the statement contained in Wood's Table, No. 3, 



EVOLUTION OF THE STEAM LOOOMOTIVI-: 



73 



where also he gives the dimensions as 12in. It is now, however, 
stated that the cylinders of these engines were 14in. in diameter. 

We now have to deal with the two locomotives with 10ft. driving 
wheels, constructed by Mather, Dixon and Co. for the Great Western 
Railway. 

Fortunately, one of the people who assisted in the construction of 
these engines is still living, and in the Engineer for January 3rd,. 
1896, he gave a detailed account of the building of the locomotive, 
and also a drawing of the " Grasshopper " (a nickname for the " Ajax " 
or "Mars"), which is here reproduced: 




^U^fBp- 

^ 



~mir 

ls/ 



FIG. 27. THE "GRASSHOPPER," ONE OF THE TWO BROAD-GAUGE ENGINES 
("AJAX" AND "MARS"), WITH 10FT. DRIVING WHEELS, DISC 
PATTERN, BUILT FOR THE GREAT WESTERN RAILWAY COMPANY 
BY MATHER, DIXON AND CO. 



The gentleman in question has favoured the writer with the 
following particulars concerning this engine: "The engine was 
designed by John Grantham, draughtsman at Mather^Dixon, and Co.,. 
North Foundry, Liverpool. The outside view resembled a steamer, 
the driving-wheel splashers like a paddle-box, and the handrail plates,, 
brought to the buffer planks, shaped like the stem of a vessel, and 
intended to take the wind pressure off the front end of the engine. 
The great diameter of the driving wheel shows that Brunei had 
something to say about it perhaps ordered it to be made twice the 
size of any other then made. The staff employed in the works then 
were : John Grantham, principal of drawing office, afterwards 
partner; Robert Hughes, manager of the marine department, after- 
wards of the Royal Arsenal, Woolwich, and inspector of steamships ; 
Mr. Banks, locomotive foreman, well known at Derby on the Midland 



74 EVOLUTION OF THE STEAM LOCOMOTIVE 

Railway ; Mr. Buddicomb, first locomotive superintendent of the 
Grand Junction Railway, and of the locomotive works at Rouen, 
France; Josiah Kirtley, first locomotive superintendent of the Mid- 
land Counties; George Harrison, first locomotive superintendent 
Scottish Central, and manager at Brassey's, Birkenhead; Mr. Potts, 
afterwards of the firm of Jones and Potts, Newton-in-the-Willows, 
locomotive builders, where the first solid locomotive wheel was made 
by the wheelsmith Frost. 

"All the above-named were apprentices and journeymen with me 
in my time. 

"William Tait, of the firm of Tait and Mirlees, Scotland Street, 
Glasgow, was the erector of the 10ft. wheel locomotive; I worked 
as mate with him on the .same engine. Tait was manager of Neilson's 
Hyde Park Locomotive Works, Glasgow, in 1845, and his mate 
John Wilson was manager from 1864 to 1884 under Mr. James Reid, 
sole owner of Neilson's Works. James Smith Scarf welded the 10ft. 
tyres. The crank axles were forged at the Mersey Forge, when Mr. 
Norris was manager, and turned by Charles Ackers. Ned Bursing 
turned the rims and tyres on a large lathe, driven By the gearing of 
the boring mill. I remember, having worked on the same lathe, that 
they had to cut a curved piece out of the shop wall for clearance/' 

The "Ajax" and "Mars" (Fig. 27), the 10ft. wheel engines supplied 
by Mather, Dixon, and Co. , had the driving wheels of peculiar construc- 
tion. Instead of the usual spokes, the circumference and the centres 
were connected by means of iron plates, bolted together in segments, 
And slightly convex in form. 

These disc wheels were constructed under a patent granted to 
Mr. B. Hicks, of Bolton, in October, 1834. The primary object of 
Mr. Hicks's pattht was not, however, the disc wheels, but a three- 
cylinder engine, with the cylinders placed vertically above the crank 
axle. Steam was only to be admitted at the top of the piston, so that 
the force of the steam was always pressing downwards ; by this method 
Mr. Hicks expected to considerably augment the adhesive properties 
of the engine. We cannot discover that an engine with three such 
cylinders was ever constructed, although the disc wheels were used in 
the " Mars," " Ajax," and other locomotives. 

As will be seen from the illustration of the " Grasshopper," these 
two 10ft. wheel engines had a projecting front, and the splashers 
covering the wheels above the frames were made to represent paddle- 



EVOLUTION OF THE STEAM LOCOMOTIVE 75 

boxes of a steamboat. For these reasons, Dr. Lardner says, they 
were generally known as the "boat engines," and he goes on to 
remark that they were found incapable of working the passenger trains 
(probably in consequence of the time lost in starting and stopping 
the monsters), and were used to haul the ballast trains during the 
construction of the Great Western Railway. Mr. Brunei gave the 
following evidence relative to these 10ft. wheel engines before the 
Gauge Commissioners in 1845: "Three engines were made for 10ft. 
The idea did not originate with me, but it was proposed by certain 
manufacturers, and although I expressed some fear of the feasibility 
of constructing 10ft. wheels, I thought it worth the trial. 
They were made, and it so happened that the three 
engines to which they were applied totally failed in other respects, 

and the whole engine was cast aside The engines to 

which I refer were a pair made in Liverpool by a maker there, who 
wa^ also making other engines for us. I take the whole responsi- 
bility, of course, of having allowed the 10ft. wheel to be made; but 
the engines, from other circumstances, were not successful, and the 
construction of the wheels was one which we should certainly never 
again adopt. It was an entire plate, and that with such a diameter 
is heavy, and offers such an enormous surface to the side wind that 
it certainly would not do to adopt it. In the other engine (' Hurri- 
cane '), which was tried with a 10ft. wheel, the wheel worked very 
well, but accidental circumstances threw the engine out of use; the 
-wheels got broken by an accident which would have broken any 
wheels, and no further attempt was made to use it." 

Mr. T. R. Crampton, the designer and patentee of the famous 
Crampton engines, gives the following particulars of the "Ajax"; 
"Area of fire-grate, 10.22ft.; total heating surface, 474.0ft.; diameter 
of driving wheels, 10ft.; diameter of cylinders, 14in. ; length of 
stroke, 20in. ; surface in fire-box, 57. 3ft. ; cubic contents of both 
cylinders, 7.09ft.; proportion of capacities to the wheel, 1:1.41." 

The " Ariel " appears to have come into collision with the " Hurri- 
cane" at Bull's Bridge (Hayes) on November 6th, 1838, whilst the 
" Lion " broke down near the same spot at five o'clock on July 30th, 
and was unfortunate enough to run over and kill a man at Baling on 
November 6th, 1838. 

About midnight on March 3rd, 1839, the "Atlas" was hauling a 
ballast train of 25 wagons towards Paddington, and instead of stopping 



76 



EVOLUTION OF THE STEAM LOCOMOTIVE 



at the usual place, the train continued on into the engine-houss, 
colliding with the "North Star," and doing considerable damage to 
that renowned locomotive; then, proceeding on its victorious career, 
it next charged the wall of the engine-house, and, finally, came to n 

stop. Upon inquiring into the cause of 
the accident it was discovered that both 
the driver and stoker were asleep on the 
engine, and that the train had been 
running for some miles with no one in 
charge. Although there were fifty men 
on the wagons, none of them were 
seriously injured. 

Great excitement was caused in 
London on the evening of October 
26th, 1838, by the ' report that Mr. 
Field (a partner in the firm .,of 
Maudslay and Field, the well-known 
engineers) had been run over and killed 
by the " Hurricane " (Fig. 28), but tliv* 
was not quite correct. The true facts 
were as follows : Dr. Lardner and his 
assistant, a youth of 19, named Field, 
were making experiments at Acton on 
the deflection of the rails, for the pur- 
pose of Wood's report to the directors, 
and were using the up line. The 
" Hurricane " was the engine employed, 
and this engine came down from, Pad- 
dington on the up line for their use. 
Young Field was stooping down to 
measure the amount of deflection as the 
engine passed, and just at the moment 
overbalanced himself in front of the 
" Hurricane," and, although it was only 
travelling at the rate of five miles an 
hour, it could not be pulled up in the 
short space, and he was, unfortunately, 
run over and killed. 

In December, 1836, T. E. Harrison 
L patented an arrangement for carrying 




EVOLUTION OF THE STEAM LOCOMOTIVE 77 

tho boiler of the locomotive on one carriage and the machinery on 
another, the idea being that when repairs were necessary to the boiler 
portion it could be disconnected from the machinery, and another 
boiler carriage substituted, and vice versa. Considering the amount 
of repairs necessary to locomotives at this early period of their evolu- 
tion, great ' economy was expected from the adoption of che 
Arrangement: 

The "Thunderer" (Fig. 29) was constructed in 1837 by Hawthorn's 
of Newcastle. The boiler portion of the machine was carried on six 
wheels, and viewed from its exterior, it appeared to be similar to an 
ordinary locomotive. In front, at the chimney end, was the machinery 
carriage, carried on four-coupled wheels of 6ft. diameter. The gearing 
being 3 to 1, therefore, one revolution of the prime driving wheels 
caused the travelling wheels to turn three times, thus making them 
equal to driving wheels 18ft. in diameter. 

The cylinders were horizontal, and the connecting-rods were 
attached to a double-cranked axle, on which was the cogged wheel; 
this worKed a pinion on the axle of the driving wheels. The axle of 
the driving wheels had a motion up and down, to allow for imperfec- 
tions in the road; and the cogged wheel and pinion were kept at 
the requisite distance in gear by the supports of the cranked axle 
being fixed over and connected with those of the driving wheels, and 
thus moving in conjunction with them. Two eccentrics on the cogged 
vrheel axle worked the slides with the usual levers and hand-gear, 
and the exhaust steam from the cylinders was discharged into tho 
chimney. 

The two carriages were connected by a bar, and the steam pipes 
had a ball-and-socket joint for lateral motion, with a metallic ring 
packing ; they also were composed cf two parts which slid one within 
the other, allowing by this means a motion in the direction of their 
length. The tank was under the boiler, and the engine-wheels were 
coupled, in order to have the whole weight for the purpose of obtain- 
ing adKesion. To keep the teeth at the right pitch, and prevent back- 
lash on reversing the motion, the pinion was in two parts, one of 
which was movable round the axle, and by means of keys these might 
be set so as to place the two halves of the teeth a little out of the 
' right line, and thus tighten 4heir action. 

The diameter of the boiler was 44in., that of the 135 tubes, If in. 
(internal) ; the tubes were 8ft. Tin. long. The fire-box was provided 
with a mid-feather. 



EVOLUTION OF THE STEAM LOCOMOTIVE 79 

On Friday, January 18th, 1839, the "Thunderer" drew a heavy 
ordinary train from Maidenhead to Paddington in 32 minutes, includ- 
ing the time occupied in stopping at Slough. 

The "Hurricane" was of the same general design and dimensions- 
as the " Thunderer," with, of course, the wide difference as to the 
mode of working. The machinery vehicle of the " Hurricane " was 
supported on six wheels, the leading and trailing being 4ft. Gin. dia- 
meter, whilst the driving wheels were 10ft. in diameter, the piston- 
rod connections working direct on the crank axle. The axle-boxes 
were above the frames, as was also the case with the two 10ft. wheel 
locomotives previously described. 

In a so-called locomotive history what purports to be an illustra- 
tion of the " Hurricane " is given ; ' the wheels are there shown with 
direct radiating spokes. The spokes of both these curious locomotives 
were, however, of the V description, as shown in the illustrations (Figs. 
28 and 29), and in Colburn's " Locomotive Engineering." 

At the end of September, 1839, when the 31 miles of the line 
was open to Twyford, the driver of the " Hurricane," having obtained 
a promise from the directors that they would provide for his wife 
and family if an accident happened to him, undertook to drive the 
" Hurricane " to Twyford at the speed of 100 miles an hour ; and r 
allowing three miles for getting up speed and stopping, it is stated 
that he successfully covered 28 miles at the rate of 100 miles an hour. 

In 1846, Grissell and Peto, the well-known railway contractors, 
undertook the task of removing the mammoth bronze equestrian 
statue of the Duke of Wellington from Mr. Wyatt's studio in the 
Harrow Koad, near the Great Western Railway locomotive shops, to 
Hyde Park. The car weighed 20 tons, and was borne by four wheels 
10ft. in diameter, lent by the Great Western Railway, one pair being 
open-spoked wheels from under the " Hurricane," the other pair being 
constructed of disc sheet iron, and were from under the "Mars" or 
" Ajax." Both pairs are clearly illustrated in the Illustrated London 
News for October 10th, 1846. 

Of the original Great Western locomotives there now only remain 
to be described the two geared engines supplied to the Great Western 
Railway by the Haigh Foundry Company. Unfortunately, little is 
known of these. Sir D. Gooch thus writes of them (after describing 
the spur and pinion gearing of the " Thunderer ") : "The same plan 
of- gearing was used in the two engines built by the Haigh Foundry ; 



80 EVOLUTION OF THE STEAM LOCOMOTIVE 

their wheels were 6ft.* diameter, and the gearing 2 to 1, but the 
cylinders were small. I felt very uneasy about the working of these 
machines, feeling sure they would have enough to do to drive 
themselves along the road." In the face of this emphatic and distinct 
statement of Sir D. Gooch respecting the two geared engines built by 
the Haigh Foundry Company, it has been stated that Sir D. Gooch 
was referring to the Haigh Foundry valve gear! Fancy reading "the 
same (spur and pinion) plan of gearing was used in the two engines 
built by the Haigh Foundry . . . the gearing being 2 to 1," 
and then being told that it was the Haigh valve gear that wag 
meant ! 

In addition to Sir D. Gooch's statement, we are fortunate to have 
the evidence of an independent person. This eye-witness, who saw 
one of the Haigh geared-up engines at Paddington in August, 1838, 
gives a very interesting and lucid account of this engine and its trial 
trips. He writes : " I have just returned from witnessing the per- 
formance of an engine on the Great Western Railway, built by the 
Haigh Company, upon somewhat of a new principle, which combines 
what the writer deems to be essential to the perfectibility of the 
locomotive engine namely, slower motion of piston with increased 
speed of engine. The experiment was completely successful, and, 
although Mr. Harrison has abandoned his plan, the principle of 
giving increased speed by the application of tooth and pinion gear 
is fully established by this experiment. 

"The engine started from Paddington with five carriages to 
Maidenhead, and returned with five carriages and two wagons loaded 
with iron, and frequently travelled at the rate of 40 miles an hour. 

"The engine then took the five o'clock train with passengers to 
Maidenhead, and performed the journey at the rate of 36 miles an 
hour with from 120 to 150 passengers." 

It will be noticed in the above statement that Harrison had 
already discontinued the 3 to 1 gearing of the " Thunderer." Sir 
D. Gooch says that he had to rebuild one-half of the original engines 
to make them of any service. It is more than probable that the 
two Haigh geared engines were thus rebuilt. Indeed, the fact that 
the books of the Great Western Railway show that the " Snake " ard 
''Viper" had driving wheels 6ft. 4in. in diameter is evidence tha* 

* The records at Swindon Locomotive Works show that the "Snake" ana - 
"Viper'' had wheels 6ft. 4in. in diameter. 



EVOLUTION OF THE STEAM LOCOMOTIVE 



such was the case, as the geared engines when delivered had wheels 
6ft. in diameter, and allowing that the small spur wheels were in a 
certain position, it would only be necessary to remove the spur 
wheels, slightly alter the length of the connecting-rods, and place 
wheels of 6ft. 4in. diameter on the crank axle to make ordinary 
locomotives of the engines in question. 

It is also possible that the discs of the "Ajax" wheels were cut 
down to 8ft., and new tyres provided, which would account for the 
fact that in 1842 Whishaw gives the diameter of " Ajax's " driving 
wheels as only 8ft. 

The following interesting table gives the result of the working of 
some of the original Great Western Railway locomotives: 



LOAD. 


BATE OF 
TRAVELLING. 


COKE CONSUMED. 




6 




1 


a 
| 


5 | 


H 


la 


' 


fe 
y - o S 


Names of 




II 


00 


.2 


IP"* 


(9 




a 




Engines. 


O 


u 

H 


5 


5 


1 * 






& 


f 


S. 

r 




Tots. 


Tons. 


Tons. 


Miles. 


Min. 


Miles 
per 
hour. 


In Ibs. 


J 


Goods. 


Gross 
load. 


North Star 


40-5 


28-5 


69 


8,848 


884 


25-45 


420,784 


47-5 


1-17 


63 


JEolus 


40-5 


28-4 


68-9 


7,292 


237 


23-81 


353,360 


48-4 


1-19 


71 


Venus 


40-5 


26-5 


67 


240 


300 


22-5 


12,656 


52-7 


1-3 


78 


Neptune 


40-5 


26-5 


67 


4,728 


949 


23-83 


188,384 


39-8 


95 


59 


Apollo 


40-5 


26-5 


67 


4,392 


942 


23-81 


193,080 


43-9 


1-08 


65 


Premier 


40 k 5 


25 


65-5 


3,024 


99 


22-73 


159,936 


52-8 


1-3 


87 


Lion 


40-5 


24 


64-5 


3,973 


96 


23-43 


226,576 57 


1-4 


89 



In consequence of the deficiency in the heating surface of many of 
the original broad-gauge engines, they had but a short career; among 
the first discarded were the "Ajax," "Planet," "Lion," "Apollo," 
" Hurricane," and " Thunderer." 

Although their lives as locomotives were ended, they were made 
to perform the functions of stationary engines; thus, during repairs 
to the beam engine in the fitting shops at Swindon Works in 1816 
or 1847, the "Lion" and "Planet" supplied the motive power to 
actuate the machinery, while the "Apollo" supplied steam to work 
the first Nasmyth's steam-hammer erected at Swindon. 



CHAPTER VII. 

Opening cf the London and Birmingham Eailway " Wallace," with feed water 
heating apparatus. Dr. Church's tank engine, "Eclipse." Balanced loco- 
motives. Smoke-consuming locomotives. Opening of the London and South- 
ampton Railway. "Soho," a locomotive without eccentrics A double 

flanged wheel engine Hancock's attempts to supply railway locomotives. 

American engii.es for England. Particulars of the engines and thsir workin?. 
Gooch commences to design engines for the Great Western Railway. His 
patent steeled tyres Grny introduces expansive working. Trial of his valve 
gear. The " long boiler " fallacy Stephenson's design for the York and 
North Mid'aii-1 Ka'Kvay. Rennies build a powerful loccmotive. Inventor 
of the link notion: Howe, Williams, or Stephenson? America claims the 

credit for the improvement. Beyer's single-plate frams engines Early Crewe 

engines. Robertson fits a steam brake to a locomotive. Engines for working 

the Cowlairs incline. Bodmer's reciprocating or "compensating" engines 

Tried on the Sheffield and Manchester, South Eastern, and london and Brigh- 
ton Rail ways.- -They prove failures. McConnelTu " Great Britain." Dew- 
ranee's coal -burning " Condor." 

EDWARD BURY, the celebrated locomotive engineer, of Liverpool, 
contracted to supply the London and Birmingham Railway with loco- 
motives. The first portion of the line was opened on June 20th, 1837 y 
and four-wheel Bury engines of his well-known types hauled the trains. 
Fig. 30 shows one of his standard passenger engines for the London and 
Birmingham Railway. 

In 1838 Kimmond, Hut-ton, and Steele, of Dundee, built a loco- 
motive, named " Wallace," for the Dundee and Arbroath Railway, at 
a cost of 1,012, including the tender. This engine had inside frames 
and inclined horizontal outside cylinders, 13in. diameter, 18in. stroke; 
the driving wheels were 5ft. Gin. diameter, the leading and trailing 
being 3ft. Gin. diameter ; the valve chests were on top of the cylinders. 
The exhaust steam was turned into the tender for the purpose of 
heating the feed-water. The "Wallace" was described as being r 
"without exception, one of the most splendid and beautifully finished 
pieces of mechanism ; indeed, all present who had seen the ' Scorpion,' 
' Spitfire,' and other celebrated English engines, gave the prefer- 
ence to the 'Wallace.'" The gauge of the Dundee and Arbroath 
Railway was 5ft. Gin. 

Dr. Church, a celebrated scientific experimentalist of Birmingham, 
constructed a four-wheel tank engine in 1838, named the "Eclipse." 



EVOLUTION OF THE STEAM LOCOMOTIVE 



83 




G '2 



84 EVOLUTION OF THE STEAM LOCOMOTIVE 

This locomotive was used in the construction of the London and 
Birmingham Railway. The cylinders were placed outside in a 
horizontal position, and were 11 Jin. diameter, the stroke being 24in. 
The leading or driving wheels were 6ft. 2Jin. diameter, and are said 
to have been the largest used up to that time on the narrow-gauge, 
being 2 Jin. larger than the 6ft. wheels of the original " Liverpool." 
The trailing wheels were 3ft. diameter. The water tanks 
were placed beneath the boiler, and when loaded the driv- 
ing wheels sustained a weight of 9 tons, and the trailing 
5 tons. The "Eclipse" hauled a load of 100 tons, and when running 
"light" attained a speed of 60 miles an hour. It will be observed 
that for the size of the driving wheels, weight of engine, design, and 
speed, the "Eclipse" was a considerable advance on the narrow- 
gauge practice then obtaining. The "Eclipse," after being rebuilt, 
was at work at Swansea in 1861. 

In 1838, two important improvements were introduced in loco- 
motive construction viz., the balancing of the reciprocating parts 
of the engine, and the partially successful use of coal in place of coke 
as fuel. Heaton, an engineer of Birmingham, introduced the 
balancing of locomotive wheels. This was in August, 1838, when he 
made a model engine on the suggestion of a director of the London and 
Birmingham Railway. The " Brockhall," one of the engines of the 
Company, was repaired at the Vulcan Works, Birmingham, early in 
1839, and ,was then fitted with Heaton's improvement. Sharp, 
Roberts and Co. had, in the previous December, supplied an engine 
to the London and Southampton Railway fitted with balancing weights 
just within the wheel rim ; while Heaton's weights took the form of an 
extension of the crank-throws on the opposite side of the axle, a 
method still employed in modern engines. The first locomotive that 
ever burned coal in a satisfactory manner, without the smoke causing 
a nuisance, was the "Prince George," a six-wheel engine belonging to 
the Grand Junction Railway. In 1838 it was fitted with Chanter's 
patent furnace,, the fire-bars of which sloped from the fire-box door 
to the tube-plate at an angle of 45 degrees ; over the fire-bars was a 
deflector. The motion of the engine caused all the fuel to fall 
to the lower end. Early in 1839 another six- wheel engine belonging 
to the Grand Junction Railway, the " Duke of Sussex," with cylinders 
loin, by 18in., was fitted with a Chanter furnace. This time the 
fire-bars did not slope so much, and on a trip from Crewe to Liver- 



EVOLUTION OF THE STEAM LOCOMOTIVE 



8b 



pool the engines covered several consecutive miles at the speed of 
60 miles an hour, the officials of the company at the same time 
declaring that the engine emitted no more smoke than the engines 
burning coke. 

The first portion of the London and Southampton Railway (now 
the London and South Western) was opened on May 12th, 1838, from 
London to Woking. The original locomotives were, with four excep- 
tions, six-wheel "single" engines, with driving wheels 5ft 6in. dia- 
meter. Fig. 31, "Garnet," is an illustration of one of these locomo- 




FIG 31. "GARNET," ONE OF THE FIRST ENGINES SUPPLIED TO THE 
LONDON AND SOUTHAMPTON RAILWAY 

tives; the cylinders were 13in. diameter, and the stroke 18m. The 
leading and trailing wheels were 3ft. 6in. diameter. The " Garnet '' 
weighed 13 tons empty. 

In 1839, Peel, Williams, and Peel, of Soho Works, Ancoats, sent 
the first locomotive constructed by them to the Liverpool and Man- 
chester Railway. This engine was named "Soho," and took a train 
of 25 loaded wagons, weighing 133 tons 18. cwt. 2 qrs., from Liverpool 
.to Manchester; whilst for a fortnight before this she was running 
with the ordinary passenger trains, and "no failure had taken place, 
and the trains having usually been brought in before their time." 
The improvement introduced into this engine consisted of a new 



86 EVOLUTION OF THE STEAM LOCOMOTIVE 

method of working the valves. The " Soho " had no eccentrics, but 
in place of them were two spur wheels, staked on to the crank axle, 
driving two other wheels of equal diameter placed immediately over 
them, so as to preserve the distance between the centres constantly 
the same, and unaffected by the motion of the engine on its springs. 
The wheels last mentioned were attached to a short axle, carrying 
at each end a small crank arm, which drove a connecting-rod attached 
to the valve spindle. 

Fenton, Murray, and Jackson, of Leeds, in 1839, supplied a six- 
wheel engine named " Agilis " to the Sheffield and Rotherham Rail- 
way. We have only very meagre details relating to this locomotive, 
but she is said to have had flanges an each side of the wheels, and 
also "that if either one or all the eccentrics which move the valves 
were broken, disarranged, lost off, or taken away, she is still under 
the control of the engineer, who can safely conduct her along the 
railway nearly as well as if those parts had remained entire." No 
explanation is given as to " how it was done " ! 

In 1840, Walter Hancock, of Stratford, Essex, who was well 
known as a steam road-coach builder, constructed a locomotive on 
somewhat the same system as his steam coaches. This engine was 
tried on the Eastern Counties Railway. The boiler was of peculiar 
design, containing a number of separate chambers, each enclosing 
several tubes. Each chamber or set of tubes connected with two 
general reservoirs, one at the bottom for the supply of water, the top 
one being a reservoir for the steam. The connection from each chamber 
to the water, steam pipes ? and reservoirs had self-acting valves, so 
that should an accident happen to any one chamber the self-acting 
valves were closed by the pressure of the steam above, or the water 
beneath, so that the remainder of the boiler retained its efficiency, 
the only result of the accident being a reduction of the heating sur- 
face. An accident of this kind was not so serious as a burst tube, 
as the damaged portion was automatically thrown out of use. Another 
advantage of this locomotive was the great heating surface contained 
in a comparatively small space ; a further improvement was a recipro- 
cating set of fire-bars. The cylinders were vertical, and actuated an 
independent crank shaft ; the progressive motion was conveyed to the 
wheel axle by means of endless chains working over pulleys fixed on 
the driving wheel axles, the diameter of the pulleys baing graduated, 



EVOLUTION OF THE STEAM LOCOMOTIVE 87 

so that the engine could be geared up or down, as either speed or 
power was required. 

As the machinery did not directly drive the wheels, it was possible 
to put that portion out of gear when it became necessary to work 
the feed pumps, etc. This 'was a considerable improvement on the 
usual locomotive, which upon such occasions either had to make a few 
trips for the purpose of supplying the boiler with water, or else 
perform over a "race." 

In 1839, Norris, the locomotive builder of Philadelphia, U.S.A., 
made an offer -to the directors of the Birmingham and Gloucester 
Railway to provide engines for working the severe gradient known as 
the Lickey Incline, 2 miles 3.35 chains in length. The agreement 
stipulated that the " locomotive engines were to be of a higher power, 
greater durability, and less weight than could be obtained in' this 
country. They were to be subjected to 15 trials within 30 days, and 
prove their capability by drawing up a gradient of 1 in 330 a load 
of 100 tons gross weight, at a speed of 20 miles an hour, and up a 
gradient of 1 in 180 a load of 100 tons at the speed of 14 miles an 
hour." If the American locomotives fulfilled these conditions the 
Birmingham and Gloucester Railway were under a contract to accept 
ten of the engines, at a price not exceeding 1,600 each, including 
the 20 per cent, import duty. Captain Moorsom, the engineer of the 
railway, stated that the " engines had not strictly complied with the 
stipulated conditions, yet he considered them good, serviceable 
engines." It will be observed that no guarantee was given as to 
what work these engines would accomplish on the Lickey Incline. 

The first three engines to arrive were the " England," " Colum- 
bia," and " Atlantic," and, according to the arrangements between the 
builder and the Birmingham and Gloucester Railway, they underwent 
a series of trials on the Grand Junction Railway before the directors 
of the Birmingham and Gloucester Railway accepted the engines. 
These trials took place during April and May, 1839, between Bir- 
mingham and Liverpool, a double journey of 156 miles being fre- 
quently made in one day. "The requisite load could not always be 
obtained, and it then became necessary to add empty wagons to the 
train to make up the right weight. The trains on some of the 
occasions exceeded 220 yards or 1-8 mile in length. With a steam- 
working pressure of 621b. per square inch, the results tabulated wers 



88 EVOLUTION OF THE STEAM LOCOMOTIVE 

as follows: On a rising gradient of 1 in 330, with a load ranging 
between 100 and 120 tons, the speed ranged from 13 4-5 miles to 
22J miles an hour; on an incline of 1 in 177, with a load of 100 tons, 
the variation in speed ranged between 9 4-5 miles and 13 4-5 miles 
an hour. Twenty-one trial trips were made, and in only five were 
the stipulated performances carried out, in five others doubt existed as 
to the work performed, but in eleven the engines failed to do the 
required amount of work. 

These experiments showed a curious result with regard to the 
fuel consumed. The aggregate rise of the gradients from Liverpool to 
Birmingham is about 620ft.; that from Birmingham to Liverpool is 
about 380ft. (exclusive in both cases of the Liverpool and Manchester 
Railway); the difference, therefore, up to Birmingham is about 240ft. 

In seven journeys of 596 miles up to Birmingham, the engine 
conveyed 682 tons gross, evaporated 12,705 gallons of water, and con- 
sumed 177 sacks of coke (1J cwt. each). In seven journeys of 596 miles 
down from Birmingham, the same engine conveyed 629 tons gross, 
evaporated 12,379 gallons of water, and consumed 177 sacks of coke. 
It would thus appear that the consumption of fuel was the same in 
both cases, and the only difference was the evaporation of 326 gallons 
of water more in the journey up than in the journey down, conveying 
nearly the same load both ways. The construction of these engines 
was very simple, and the work plain. The boiler was horizontal,- and 
contained 78 copper tubes 2in. diameter and 8ft. long, with an iron 
fire-box. The cylinders, 10 Jin. diameter, were inclined slightly down- 
wards, and so placed that the piston-rods worked outside the wheels, 
thus avoiding the necessity of cranked axles. 

The framing of these American engines was supported by six 
wheels; the two driving wheels of 4ft. diameter were placed close 
before the fire-box; the other four wheels, of 30in. diameter, were 
attached to a truck, which carried the front end of the boiler, and 
was connected with the frame by a centre-pin, on which it turned 
freely, allowing the truck to accommodate itself to the exterior rail 
of the curve, and, with the assistance of the cone of the wheels, to 
pass round with very little stress upon the rails. 

Tons cwt. 
The weight of the engine with the boiler and firebox full was... ... 9 11 

That of the tender with 21 cwt. of coke and 520 gallons of water was... 6 4i 

Total weight 15 15 

These engines, when empty, weighed only eight tons each. 



EVOLUTION OF THE STEAM LOCOMOTIVE 89 

Another of the American bogie engines supplied to the Birming- 
ham and Gloucester Railway was named the "Philadelphia." She 
was a more powerful locomotive than the three mentioned above, and 
Captain Moorsom, the engineer of the railway, in a letter dated from 
Worcester on June 22nd, 1840, gives an interesting account of her 
trial on the Lickey Bank. " Seventy-six chains in the incline of 1 in 
37 J were made ready with a single way, and three chains nearly level 
were laid temporarily to rest upon before starting. The road was 
quite new, and consequently not firm or well gauged, and the works 
going on close at hand occasionally covered the rails with dirt. The 
wagons used were of a large class, like those on the Manchester and 
Leeds line, and weighed when empty rather more than 2 tons, and 
at first worked very stiffly. They were loaded with 4 tons, and generally 
weighed, including persons upon them, about 6f tons. The 'Phila- 
delphia' weighed (as she worked) about 12 tons 3 cwt., and her 
tender weighed nearly 7 tons, being in all 19 tons. She had 12 Jin. 
cylinders, 20in, stroke, 4ft. driving wheel not coupled. The weight 
on her driving wheels was 6 1-3 tons (as weighed at Liverpool) 
without water. 

" The usual load she took was eight wagons, engine, and tender, 
with persons, equal to 74 tons gross weight, in ten minutes, or nearly 
6 miles per hour, the last quarter of a mile being at the rate of 
9f miles per hour. Seven wagons, etc., equal to 67J tons gross 
weight, in about 9 minutes, or 6 miles per hour mean speed. Six 
wagons, etc., equal to 61 tons gross weight, in sometimes 5J and 
sometimes 6J minutes, say in 6 minutes average, or 9 
miles per hour mean speed, the last quarter of a mile usually 
giving a speed of nearly 11 miles an hour. Five wagons, equal frc 
about 53 tons gross, were usually taken at a speed of 13 miles per 
hour for the last half-mile up. The foregoing results occurred 
generally during fine weather, but sometimes the rails were partially 
wet, and this occasioned a difference of speed in the ascent of half a 
minute to a minute and a half. One day when showery the men walked 
over the rails with marl on their boots, rendering the way very 
greasy and slippery, also the lower part of the plane had been formed 
only a few hours, and was very soft and badly gauged. 

Under these circumstances the 'Philadelphia' tooK five wagons, 
self, and tender, being a gross weight, including persons, of about 
53 tons, up at a mean rate of rather more than 5 miles per hour, 



90 



EVOLUTION OF THE STKAM LOCOMOTIVE 



and the last quarter of a mile was passed at the rate of 8 miles per 
hour. Two wagons were then taken off, and the ' Philadelphia ' 
took the remaining three wagons, self, and tender, being a gross 
weight, including persons, of 40 tons, up at a mean rate of 12 miles 
nearly per hour, her maximum speed being nearly 16 miles per hour." 

Sir D. Gooch was not at all satisfied with the original broad-gauge 
locomotives, and in 1839 he obtained the sanction of the directors of 
the Great Western Railway to design two classes of locomotives for 
the railway. These engines were known as the " Firefly " class and 
the "Fury" class, the former having 7ft. driving wheels, cylinders 
15in. diameter, 18in. stroke, and 700ft. of heating surface; the latter 
had 6ft. driving wheels, cylinders 14in. diameter and 18in. stroke, 
and 608ft. of heating surface. 

One hundred and forty-two locomotives of the "Fury" and 
" Firefly" design were constructed. Sir D. Gooch states that the best 




FIG. 32. "HARPY," ONE OF GOOCH'S "FIREFLY 
BROAD-GAUGE ENGINES 



CLASS OF 



were built by Fenton, Murray and Jackson, of Leeds. The sixty-two 
of the " Firefly" class were- built as follows: Twenty, by Fenton, 
Murray and Jackson, Leeds ; sixteen, by Nasmyth, Gaskell and Co., 
Manchester; ten, by Sharp, Roberts and Co., Manchester; six, by 
Jones, Turner and Evans, Newton; six, by Longridge and Co., Bed- 



EVOLUTION OF THE STEAM LOCOMOTIVE 91 

lington ; two, by Slaughter and Co., Bristol ; and two, by G. and J. 
Rerniie, London. 

It will be observed that most of these were built in the North 
of England, and it is a significant fact that these broad-gauge loco- 
motives were conveyed on mrrow-gauge trucks for some hundreds 
of miles to the Great Western Railway, thus showing that it would 
have been quite possible to widen the existing narrow-gauge railways, 
by simply decreasing the space between the two roads, comparatively 
at a small expense. 

All tnese engines were built from the specifications and drawings 
supplied by the Great Western Railway to the makers, and thin iron 
templates were also supplied of those parts which were to be inter- 
changeable. Fig. 32 illustrates the "Firefly" type. 

The " Firefly," built by Jones and Co., Viaduct Foundry, Newton, 
was the first of these engines delivered. On March 28th, 1840, she 
made an experimental trip from Paddington to Reading, with a load 
of two carriages, containing 40 passengers, and a carriage truck; 
she performed the journey in 46 minutes 25 seconds from start to 
stop. A spring of one of the tender wheels broke on the journey, 
necessitating careful running. On the return trip, between the 
26th and 24th mile posts, a speed of 56 miles an hour was reached, 
and the average speed from Twyford to Paddington was over 50 miles 
an hour. On the occasion of the Queen's accouchement in August, 
1844, the news was brought to London by a special messenger 
travelling on one of these engines. The journey from Slough to 
Paddington, 18J miles, was accomplished in 15 minutes 10 seconds, 
or at the rate of 75 miles an hour. The illustration (Fig. 33) 
shows the interior of the old Paddington engine shed, and 
amongst the locomotives to be seen are the " Ganymede " and " Etna." 
All the engines had domed fire-boxes, and outside frames, the 
principal dimensions, in addition to those already given, being : 
Leading and trailing wheels, 4ft. diameter; boiler barrel, 8ft. 6m. 
long, 4ft. diameter; 131 tubes, 2in. diameter, 9ft. long; weight, in 
working order on leading 4| tons, driving 11 tons 13 cwt., trailing 7 
tons 16 cwt. ; total, 24 tons 4 cwt. 

On November 20th, 1840, Daniel Gooch obtained a patent for 
steeled tyres, and the locomotives of the "Fury" and "Firefly" 
classes were fitted with these patent tyres. Although the tyres only 
contained one-fifth part of shear steel, yet the use of Gooch's tyres 



92 



EVOLUTION OF THE STEAM LOCOMOTIVE 



did not become general, as 56 years ago steel was an expensive com- 
modity, and consequently railway rolling stock generally was not 

fitted with steel tyres; indeed, 
the Great Western Railway 
went no further than using 
the improvement for their 
locomotive and tender wheels. 
Many locomotives fitted with 
these patent tyres ran nearly 
300,000 miles before new 
tyres were required. 

These first essays of Daniel 
Gooch as a locomotive de- 
signer at once placed him at 
the very head of locomotive 
engineers, and Gooch himself, 
usually so modest, says of 
these locomotives, "I may 
with confidence, after these 
engines have been working 
for 28 years, say that no 
W better engines for their 
weight have since been con- 
structed, either by myself or 
others. They have done, ana 
H continue to do, admirable 
5 duty." This candid eulogium 
5 of these engines by their de- 
fe signer certainly did not go 
beyond the truth in describing 
2 their good points. Gooch's 
H first design of broad-gauge 
goods locomotives had six 
coupled wheels 5ft. in 
diameter, inside cylinders 
16in. diameter, and a stroke 
of 24in. The fire-box was of the domed pattern. Fig. 34 ("Jason") 
represents one of these engines. 

John Gray, who was in 1840 locomotive superintendent of the 




EVOLUTION OF THE STEAM LOCOMOTIVE 93 

Hull and Selby Railway, introduced a striking improvement into the 
construction of locomotives at that time. (Gray had, on July 26th, 
1838, taken out a patent for his valve gear; and whilst on the 
subject of valve gears, it will be of interest to note that Dodds and 
Owen patented their wedge motion on September 16th> 1839.) In 
Gray's improvements in the Hull and Selby engines he adopted 




Fr& 34." JASON," ONE OF GOOCH'S FIRST TYPE OF GOODS ENGINES 
FOR THE GREAT WESTERN RAILWAY 



inside bearings for the driving wheels, an extended base for the 
springs, and, of course, his patent valve motion and expansive 
working. Shepherd and Todd, of the Railway Foundry, Leeds, con- 
structed the engines in question. The driving wheels were 6ft. dia- 
meter, cylinders 12in. diameter by 24in. stroke, fire-box 2ft, by 
3ft. 6in. (inside), and 94 2in. tubes, 9ft. 6in. long. Two of these 
locomotives, " Star " and " Vesta/' were tried in competition with other 
engines on Tuesday, November 10th, 1840. Sixteen trips were made 
by the " Star " and " Vesta," the average loads being 55.4 tons, or 
1,718 tons over one mile; coke consumed, 4651b., or 0-27llb. per 
ton per mile; water evaporated, 2,874113., or 1.621b. per ton per mile. 
Two other classes of locomotives were tried in competition with 
Gray's patent viz., the usual kind of engines then in use, and the 
same with the addition of Gray's expansion gear 



94 EVOLUTION OF THE STEAM LOCOMOTIVE 

The result of the trials is shown in the following table : 





X 


11 


,2 

I 


1 . 


fj 


c 


1 


Class 


o g j5 


fi 




" 


Ka 


S 


E= z 






^ CC -,: 












of 


c-o.S 


3*1 





d^ 




II 


C.S 







* & - 


S.c 


S's 


s s 


c ~ 


& 


Engine 




'5 


* 









5 5 




^ o 




M 


"o * 


"c* ^ 


^ 


ce 




o 


ss, 


a 


O 


^"3 




^ 


Patent 
Altered 
Unaltered 


1649-4 
1649-4 
1649-4 


446-98 
686-15 
1007-78 


14-41 
22-13 
32 k 59 


0-271 
0-416 
0-611 


2672 
4601-6 
6432-8 


86-19 
148-43 

207-5 


1-62 
2-90 
3-97 



The financial annual result of the three classes of engines for coke 
and boilers, with such a traffic as that of the Hull and Selby line, 
was about : 

4,500 for the unaltered engines. 

3,250 for the altered engines. 

2,000 for the patent engines. 

We have now reached the era of another development of the 
locomotive viz. the introduction of "long boiler" engines; but 
although the idea was well " boomed," it never was thought much of 
by competent locomotive engineers; indeed, many severely con- 
demned the plan. 

In 1841 Kobert Stephenson patented a new form of valve gear, 
with a top and bottom gab fixed to the valve spindle, and 
the ends of the eccentric rods kept apart by a straight link. Here, 
again, Stephenson introduced nothing new, his gear being but a 
clumsy adaptation of Roberts's valve gear. An engine of this 
description (generally known as Step'henson's patent "long boiler" 
engine) was tried on the York and North Midland Railway in January, 
1842, the dimensions being : 



Diameter of Cylinder 

Length of Stroke 

Diameter of Driving Wheels 

Diameter of small wheels ... 

There are 150 tubes, giving a heating surface of 

Copper Fire Box, with a heating surface of 



Length of Boiler, including fire and smoke boxes 
Weight of the Engine in working order 



14 inches 
20 
5$ feet 

765 



Total hea ing surface 795 feet 



17 

15 tons 



During a journey of 90 miles, a speed of 48 miles an hour was 
attained, but the train then consisted of only five carriages of light 
weight. 

The consumption of fuel during the above experiment was 19.21b. 



EVOLUTION OF THE STEAM LOCOMOTIVE 95 

per mile, with a load of eight coaches over half the distance (45 miles) 
and five coaches over the remaining half. 

This consumption included the whole of the fuel used in lighting 
the fire and raising the steam. 

R. Stephenson introduced tubes of wrought-iron instead of 
brass or copper, in order that the increased heating surface might be 
obtained without a corresponding augmentation in the price of the 
engine. This he did not adopt without making several experiments. 

During the last twelve months he had several boilers working 
under his own eye with iron tubes, for the special purpose of deter- 
mining how far he could recommend them for general adoption. The 
result was all that he could desire; and owing to this he introduced 
them with great confidence. The valve gear is thus eulogised : " In 
ordinary engines the mechanism for working the slide valves was very 
liable to derangement and considerable wear and tear. 

" This part of the engine he so far simplified that it required only a 
simple connection between the eccentrics and slide valves, thus doing 
away with a considerable number of moving parts. 

"This was attained by placing the slide valves vertically on the 
sides of the cylinders, instead of on the top as heretofore, so that the 
direction of the sliding motion of the valves and the central line of 
the valve-rods intersected the central line of the main axle at the 
point where the eccentrics were placed. In this case the eccentric- 
rods were connected immediately to the prolongation of the valve- 
rods, without the usual intermediate levers and weigh bars; the slide 
valves of both cylinders were placed in one steam chest, between the 
cylinders." Another improvement was in the working of the feed 
pumps; it consisted in connecting the pump-rods to the eccentrics 
used for reversing the engine. By this arrangement the velocity of 
the moving part of the pump was greatly diminished, by which 
was secured greater regularity of action. 

Messrs. G. and J. Rennie, of Holland Street,^ Blackfriars, S.E., in 
1841, constructed a locomotive named the "Lambro" for the Milan 
and Monzo Railway. The " Lambro " was built from the design of Mr. 
Albano, the engineer to the railway; the cylinders were 13in. diameter, 
18in. stroke, driving wheels 5ft. Gin. diameter, steam pressure 501b., 
weight 22 tons. Her average coke consumption with trains weighing 
143 tons at 36 miles an hour, was only 221b. per mile. The loco- 
motive engineer of the railway reported that " no engine he had 



96 EVOLUTION OF THE STEA M LOCOMOTIVE 

seen at all approached the locomotive engine ' Lambro ' in any respect 
whatever, in the economy of fuel, in her immense dragging power, 
and in the excellency and solidity of her framing and working gear." 

The particular evolution now about to be described occupies a 
foremost position in locomotive history. Like many other useful 
inventions, the link motion has been proclaimed as the production of 
different people. 

Its popular title, the "Stephenson" link motion, is a well-known 
misnomer ; indeed, Stephenson never appears to have put forward a 
claim in which he figured as the inventor of the curved link motion , 
perhaps, at first, he did not fully appreciate its value. 

The germ of the idea belongs to Williams, of Newcastle, who, in 
1842, designed a form of straight link coupling the two eccentrics 
together. Of course, such an arrangement was utterly impossible in 
practice, as the crank, in revolving, would soon place the two eccen- 
trics in such a position that the link would be destroyed. The 
curved link, placed half-way between the valves and eccentrics, was 
soon evolved from Williams' crude idea, and up to 1846 it was most 
generally called Williams' motion. In an article describing expansion 
valves, in the Practical Mechanics' Magazine for April, 1846, it is 
&o described; but in the May number of the magazine a letter 
appears from William Howe, a fitter employed by R. Stephenson and 
Co., Newcastle. In this communication Howe states that Williams 
proposed the straight link, previously mentioned, but that Howe saw 
its utter impracticability, and evolved the curved link. Williams 
made no reply to this communication ; although he may not have seen 
Howe's letter claiming the invention. Be this as it may, Howe was 
thereafter given the credit for the curved link. It is, however, 
significant that he never patented it, and it is probable that at first 
neither he nor Stephenson saw its value as a means of effectually 
working the valves expansively, or one or the other would have 
protected the invention, seeing that Stephenson had then quite 
recently patented the top and bottom gab-gear. Then, again, 
Howe's supposed claim may have been a reason for not protecting it. 

In the invention of the link motion, this country does not appear 
to have been forestalled by the Celestial Empire, as (it is asserted) is 
the case with so many useful discoveries. But the glory does not rest 
with us, for it has been shrewdly " guessed " that the idea originated 
with one of our American cousins, W. T. James, of New York, who, as 



EVOLUTION OF THE STEAM LOCOMOTIVE 97 

early as 1832, constructed the "James" locomotive, which was pro- 
vided with link motion. The invention at this period does not 
appear to have been considered of any value, for its use was not 
perpetuated in later locomotives in America until after it had been 
re-discovered by the Williams-Howe experiments of 1842-3. 

In 1843, Mr. C. Beyer, then employed with Messrs. Sharp Brothers 
and Co., but afterwards of the well-known firm of locomotive builders, 
Messrs. Beyer, Peacock and Co., Manchester, introduced the single 
iron plate for locomotive frames. 

Trevithick's son directed his attention to the evolution of the 
steam locomotive, and while chief engineer of the Grand Junction 
Railway, the now world-famous Crewe Works were erected, beiijg 
opened in 1843. Mr. A. Allan became manager at Crewe, and under his 
superintendence a new class of engines was constructed, the novel 
points being the coupling of the driving and trailing wheels Allai 7 
having, in 1863, publicly claimed this innovation as wholly and solely 
due to him. 

The engines in question are usually described as "the old Crewo 
goods class," and had outside cylinders, 15in. by 14in. The coupled 
wheels were 5ft. diameter, and were placed one pair before and the 
other behind the fire-box; these wheels had inside bearings, and the 
small leading pair had outside bearings. The steam pressure was 
1201b. These useful engines weighed 19 \ tons, and were used for 
goods traffic for many years. Mr. Ramsbottom afterwards rebuilt 
several of them as tank engines, and some, as such, are still in use 
on the London and North Western Railway. Alexander Allan, who 
died as recently as 1891, was noted for his invention of a straight 
link motion in 1855. 

The need of a powerful brake has always been one of the greatest 
necessities of locomotive engineers. For a long time they all agreed 
that it was not advisable to brake the driving wheels of locomotives ; 
but Peter Robertson, the locomotive superintendent of the Glasgow 
and Ayr Railway, was of a different opinion, and in April, 1843, he 
fitted a locomotive on that railway with his patent steam brake. 
The apparatus consisted of a flexible metal band, of a semi-circular 
shape, surrounding the upper half of the driving wheel. One end of 
the band fastened to a hinge, and the- other was fixed to a piston- 
rod. When "off," the piston-rod held the band away from the tyre 
of the driving wheel, but when steam was applied behind the piston. 



98 



EVOLUTION OF THE STEAM LOCOMOTIVE 



the band was tightly pressed against the tyre. Such was the simple, 
but effective, application of Robertson's steam brake. A familiar 
example of its action can be seen in the hand brakes still fitted to 
cranes. 

The Cowlairs incline at Glasgow is the bete noir of the North 
British Railway, and is situate just outside the Glasgow terminus of 
what was originally the Edinburgh and Glasgow Railway. When 
first opened this incline was (as is, indeed, at present the case) worked 
by stationary engines; but towards the end of 1843 Mr. Paton, the 
locomotive superintendent, and Mr. Millar, the engineer of the Edin- 
burgh and Glasgow Railway, designed and built a powerful loco- 
motive for working this two-mile incline of 1 in 42. The engine was 




FIG. 34A. PATON AND MILLAR'S TANK ENGINE FOR WORKING ON THE 
COWLAIRS INCLINE, GLASGOW 

put to work in January, 1844, and during that year the cost of 
working the incline was, with the locomotive, one-third of the amount 
expended during the previous twelve months on the stationary engine. 
Upon reference to the illustration (Fig. 34A) of this remarkable loco- 
motive, the first detail that attracts notice is the immense steam dome. 
The engine was supported on six coupled wheels of 4ft. 3 Jin. diameter. 
The cylinders were " outside," fixed in an inclined position about 
half-way up the smoke-box, their diameter being 15|in. The stroke 
was 2 Sin. These dimensions, it will be noticed, were considerably 
in advance of the general practice obtaining 55 years ago. The 



EVOLUTION OF THE STEAM LOCOMOTIVE 



99 



valve chests were above the cylinders, and the eccentrics were fixed 
on the driving axle, within the frames; the springs were underhung, 
and all the wheels were counterbalanced. Two lever safety valves were 
provided. The heating surface of the fire-box was 60 sq. ft., that of 
the tubes 748ft. The other principal dimensions of this engine 
were: Fire-box, 4ft, long by 4ft. Gin. deep; smoke-box, 2ft. 6in. 
long by 4ft. 4in. deep; 136 tubes, 2in. diameter, and 10ft, 6in. long. 
This engine, it should be observed, was of the "tank" class, 200 
gallons of water being stored in a tank below the smoke-box, that 
amount being sufficient for two trips. The water was supplied from a 
stand-pipe, and not from the usual columns. 

The driving wheels were furnished with brakes, the levers of 
which were worked by a screw, the handle of the latter being placed 
within reach of the engineer. 

The trailing pair of wheels had a steam brake, something like those 
applied to the engines of the Ayr line by Mr. Robertson. Sand-boxes 
were placed in front on each side of the water tank for dropping sand 
on the rails, which was done by the stoker on the footplate, by a 
handle and rod from valves or stoppers in the boxes. The most effectual 
remedy against slipping was to keep the rails clean, which was done 
by means of two jets from the boiler in going down the incline plane. 
When very dirty two other jets of cold water were used, a small air 
vessel and one of the feed pumps being used for that purpose. 

The total weight of the engine was 26 \ tons; the rate of speed 
with 12 carriages of the gross weight of 54 tons was 15 miles per 
hour ; the rate of speed with 20 trucks of goods of a gross weight of 
104 tons was 9 miles per hour, up the Cowlairs incline. 



WORK OF EXGIXE FOR THE MONTH OF NOVEMBER. 1844. 
TOTAL WORK DONE ON INCLINE PLANE. . 



i 


X 


i 


1*3 


00 





h 
















i 


2 


1 


OH 


5 





Is 


No. 


Tons. 


No. 


Tons. 


No. 


Tons, 


Tons. 


Total. 


Hrs 




335 


1,675 


694 


3,817 


298 


1,937 


7,429 


121 


120 


7 


344 


1,720 


673 


3,701$ 


268 


1 742 


7,163$ 


107 


118 


14 


373 


1,875 


58 


3619 


248 


1,612 


7,106 


104 


118 


21 


376 


1,900 


640 


5,525$ 


254 


1660 


7,076j 


103 


118 


28 


1,430 


7,170 


2,665 


14663 


1,068 


6,951 


28,775 


435 


474 






if 2 



100 EVOLUTION OF THE STEAM LOCOMOTIVE 

The table on page 99 gives the results of one month's working of a 
second locomotive of similar design, the cylinders, however, being 
16 Jin. diameter, and additional heating surface being provided by 
means of a water' space dividing the fire-box. The second engine 
was put to work towards the end of 1844. 

These engines were named " Hercules " and " Sampson," and were 
built at Cowlairs, whilst two others of the same general design, and 
named " Millar " and " Hawthorne," were constructed at Newcastle. 

Mr. A. E. Lockyer states that these engines " had not run any 
length of time, however, before the foreman platelayer complained of 
the engines destroying the rails, which, it must be remembered, were 
only 581b. per yard, with the sleepers 3ft. apart." In consequence of 
this report the incline was relaid, the distance between the sleeper* 
being reduced to 2ft. between the centres. This did not much mend 
matters, and to crown all, the Forth and Clyde Canal 
began , to leak, in consequence, no doubt, of the vibration 
induced by the constant passage of the heavy locomotives. A 
strategic movement to the rear then became necessary, and an 
eminent engineer (Mr. McNaught) was appointed by the directors t<s 
strengthen the land engine, and put it in proper working order, so as 
to reintroduce the haulage system for working the incline. 

A Newcastle firm (R. S. Newall and Co., the original inventors 
and patentees of untwisted iron rope) supplied the railway company 
with one of their wire ropes. The land engine was finished by 
March 4th, 1847, and on trial under the new conditions the haulage 
system proved highly satisfactory, so much so that the four loco- 
motives were removed altogether. 

The Manchester and Sheffield line was, in 1845, supplied with four 
powerful goods locomotives, built on Bodmer's patent prin- 
ciple. The cylinders were 18in. diameter, stroke 24in. ; the six 
coupled wheels were 4ft. Gin. diameter; but the weight of these 
engines was only 24 tons each. They are, however, stated to have 
been equal to hauling a gross load of over 1,000 tons. Bodmer's 
locomotives deserve recognition in the evolution of the steam loco- 
motive, because of their curious construction, and also because other 
locomotive histories do not mention these peculiar engines. 

The engines are described as "compensating," the whole strain 
being confined to the pistons, piston-rods, connecting-rods, and cranks. 
There were two pistons in each cylinder, one being connected with 



EVOLUTION OF THE STEAM LOCOMOTIVE 101 

one crank and the other with the opposite crank of an axle with 
double cranks on each side, so that the driving axle was fitted with 
four cranks. 

The steam was admitted alternately between the two pistons at the 
time the pistons met in the middle of the cylinders, also between tiie 
ends or tops of the cylinders and the pistons when the latter arrived 
at the other end of the stroke. 

Bodmer claimed that by this arrangement the engine was perfectly 
balanced, and no oscillation or pitching of the engine resulted, no 
matter what speed was attained. Another engine of this description 
was supplied to the Sheffield and Manchester Railway, constructed by 
Sharp Bros, and Co. The cylinders were 14in. diameter, strode 
20in. (two strokes of lOin. each in both cylinders), driving wheels 
5fc. diameter, steam pressure 901b. per square inch. During Novem- 
ber, 1844, the average coke consumption of this engine amounted to 
only 21.921b. per mile. 

A larger and more powerful engine on the same principle v:as 
supplied to the Joint Locomotive Committee of the Soutn Eastern 
and London and Brighton Railways, and when the 
Committee was dissolved the engine was taken over by the South 
Eastern Railway in 1845, and was numbered 123. The cylinders 
were 16in. diameter and 30in. stroke, or rather, two pistons each 
working a stroke of 15in. Heating surface was : box 73 sq. ft. ; 
tubes 769 sq. ft. \ steam pressure 951b. ; weight 18 tons; coke Con- 
sumption 151b. per mile. The driving wheels were 5ft. Gin. diameter. 
Shortly after the South Eastern Railway took over this engine it 
broke down, and one of the men in charge was killed. 

Bodmer also supplied the London and Brighton Railway with 
one of these patent reciprocating engines. This was in 
December, 1845 ; and she ran the first 5 p.m. express from London 
to Brighton. The locomotive in question was No. 7, and had single 
driving wheels, 6ft. diameter. The cylinders were 15in. diameter, 
and the 20in. of stroke was, of course, covered by two pistons in each 
cylinder working lOin. The fire-box was of the well-known "Bury" 
type. No. 7 was rebuilt in January, 1850, when Bodmer's recipro- 
cating pistons were taken out, and ordinary ones put in. In later 
years No. 7 was named "Seaford." 

Bodmer designed another engine on this plan, with outside 
cylinders 22in. diameter and 24in. stroke i.e., two pistons of 12in. 



102 EVOLUTION OF THE STEAM LOCOMOTIVE 

stroke each. The driving wheels were 7ft. diameter. The boiler 
pressure of this extraordinary engine was lOOlb. and the coke 
consumption was estimated at lOlb. per mile, with trains of 12 coaches. 
This engine was fitted with cylindrical slides and expansion valves, 
under a patent obtained by Bodmer. 

In 1845, J. E. McConnell, then locomotive superintendent of the 
Birmingham and Gloucester Railway, determined to construct a more 
powerful engine for working the Lickey Incline than the American 
engines previously described. The " Great Britain " was the result 
of his essay. She was a six-wheel coupled saddle-tank locomotive. 
The wheels were 3ft. lOin. diameter, and the cylinders 18in. by 2 Gin. 
stroke. This powerful "iron-horse" easily hauled trains weighing 150 
tons up the Lickey Bank. McConnell also rebuilt one of the American 
engines, as a saddle-tank locomotive, for working the Tewkesbury 
branch of the Birmingham and Gloucester Railway. This curious 
specimen of a saddle-tank engine had outside cylinders 10 Jin. dia- 
meter, 20in. stroke, single driving wheels 4ft. diameter, and a leading 
bogie. 

Mr. Dewrance, of the Liverpool and Manchester Railway, about 
this time turned his attention to the experiments which were, ever 
and anon, being made towards the long-wished-for goal of a perfect 
coal-burning locomotive. In the "Condor" he tried the effect of 
two fire-boxes. The fuel was inserted in the usual manner into the 
exterior fire-box; the second, or combustion chamber being designed 
to consume the gaseous matter that escaped from the first furnace. 

During the period of special attention to the working of the 
" Condor " this system of coal burning appears to have been of a 
fairly successful character. The idea of a combustion chamber as a 
solution of the vexed question of a successful smoke-consuming loco- 
motive was afterwards tried by other locomotive designers. The 
division between the two fire-boxes of the " Condor " consisted of a 
transverse water-space, fitted with short tubes. Air was admitted to 
the combustion-chamber by means of a pipe, with a head perforated 
with small holes. 



CHAPTER VIII 

Stephenson's "long boiler" 1 goods engines for the Eastern Counties Railway- 
Gray's prototype of the "Jenny Lind " Hackworth builds twelve of the class 
for the Brighton Uailway Stophenson and Howe's three-cylinder locomotive 
not a success The "Great A," another Stephcnson absurdity The competitive 
trials betweau broad and narrow-gauge locomotives Gooch to tho rescue! 
The " Premier," the first engine constructed at Swindon The " Great Wes- 
tern " the forerunner of the standard express engine of to-day Trial trip cf 
this "mammoth"' A notable run of the "Great Western" The "Great 
Western" altered to an eight-wueel engine Galloway's incline-climbing loco- 
motive tried on the Great Western Beyer's "Atlas" for the Manchester and 
Sheffield Railway The Eastern Union "Essex" draws 149 loaded goods 
wagons Step iienson's " White Horse of Kent " Crampton, as a locomotive 
designer, the " Namur " constructed Gooch's "Iron Duke" and "Lord of 
the Isles " make the broad-gauge still more popular The " Jenny Lind," a 
"storm in a tea cup" Trial of the <( Jenny Lind" and "Jenny Sharp" 
Trevithiek's "Cornwall,"' a locomotive monstrosity Exhibited at the 1851 
Exhibition Rebuilt in her present form, and still running McConnell's 
"counter-balancing" experiments The "most powerful narrow-gauge engine 
ever built "" No. 185" of the Y.N. and B.R. The oldest locomotive now 
running, " Old Coppernob," of the Furness Railway " Lablache," another 
locomotive freak "Cambrian' locomotives, and the peculiarities of their 
construction The "Albion,"' of 1848- -Half a century later, the writer un- 
earths the working drawings of this engine and her sisters. 

DURING 1845 R. Stephenson and Co. built seven of their "long- 
boiler " engines, with outside cylinders, for working the goods traffic 
of the Eastern Counties Railway. Fig. 35 is an illustration of one of 
these ungainly specimens of locomotive construction. The boiler 
barrel was no less than 13ft. Gin. in length, all the axles were beneath 
tho barrel, the leading wheels were 3ft. diameter, and the driving and 
trailing (coupled) wheels t>ft. 9 Jin. diameter. The cylinders were 16in. 
diameter, the stroke being 21 in. In working order, these locomotives 
weighed 23 tons 12 cwt. After looking at the illustration, it is 
scarcely necessary o add -.that these .engines were very unsteady when 
travelling, the oscillation being excessive. 

In the arrangement of inside and outside bearings to the various 
wheels of the patent engines, designed by John Gray for the Hull and 
Selby Railway (previously described), we make acquaintance with the 
embryo design, afterwards perfected, and known the whole world ever 
as the "Jenny Lind" class. 



104 



EVOLUTION OF THE STEAM LOCOMOTIVE 



In 1846 Gray had become locomotive superintendent of the 
Brighton Railway, and he prepared another design of express engines 
for that line, in which the type now known as " Jenny Lind " was 




FIG. 35. STEPHENSON'S "LONG-BOILER" GOODS ENGINE, EASTERN 
COUNTIES RAILWAY 

further developed. J. Hackworth and Co. obtained the contract for 
the supply of twelve of these locomotives, and in November, 1846, 
they delivered the first pair, numbered 53 and 54. Fig. 
36 represents No. 49, one of these engines. The leading and 
trailing wheels were 3ft. 6in. diameter, the drivers being 6ft. dia- 
meter. Cylinders 15in. by 24in. stroke. Heating surface: tubes, 
700 sq. ft. , fire-box, 79 sq. ft. Inside bearings were provided to the 
driving, and outside to the leading and trailing wheels ; the engines 




FIG. 36. GRAY'S PROTOTYPE OF THE "JENNY LIND," No. 49, 
LONDON AND BRIGHTON RAILWAY 

were fitted with Gray's "horse-leg motion," and several of the dozen 
had two square-seated steam domes, one located on the centre of the 
boiler barrel, the other over the fire-box. Each dome was provided 



EVOLUTION OF THE STEAM LOCOMOTIVE 105 

with a steam safety-valve. The steam pressure was lOOlb. per square 
inch. 

These engines were found to be good at hauling heavy loads (us 
computed 50 years ago) at speeds up to and slightly exceeding 40 
miles an hour. 

In 1846 Stephenson and Howe obtained a patent for a three- 
cylindered engine. Z. Colburn, in his "Locomotive Engineering," 
exposes the fallacy of the idea that the action of the steam admitted 
alternately to cylinders whose centres are far apart, sets up a dan- 
gerous sinuous motion. The object of Stephenson and Howe's three- 
cylinder engine was to overcome this winding motion. Colburn states 
that a "few pounds of counterweight would have served a better 
purpose than the extra cylinder and working parts." Two engines 
appear to have been built on this plan before the true cause of the 
rocking motion and the real way of overcoming it, were fully grasper) 
by the patentees. The outside cylinders were only 10 Jin. diameter 
and 22in. stroke; whilst the centre or inside cylinder was 16in. 
diameter, but the stroke in this case was restricted to 18in. It is 
needless to add that these three-cylinder locomotives were not 
successful. 

Passing reference must be made to the celebrated gauge experi- 
ments which took place during the last days of December, 1845, and 
resulted so greatly in favour of the broad-gauge, despite the fact that 
the Great Western Railway had no new engines prepared for the com- 
petition, but used those regularly in work on the broad-gauge railways. 

The narrow-gauge experiments were made on the Great North of 
England Railway, a special engine being built for the purpose by R. 
Stephenson and Co., and called "A." The "A" was a six-wheel 
long-boiler engine, with outside cylinders and 6ft. 6in. driving wheels. 
Hot water for supplying the boiler was used on the narrow-gauge 
in place of cold on the broad-gauge. The latter started 
from a state of rest, but the narrow-gauge approacl ed 
the starting-point at as great a velocity as possible; yet, notwith- 
standing these sharp practices of the narrow-gauge officials, they were 
completely beaten in the experiments. 

The Swindon Works commenced to build locomotives early in 
1.846 ; and, as its name implies, the " Premier " was the first engine 
constructed at these now world-famous locomotive shopb. 

She was a six-coupled goods engine, with wheels 5ft. diameter. 



106 



EVOLUTION OF THE STEAM LOCOMOTIVE 



Numerous engines of this type, with, slight modifications, vrere 
built, at Swindon ; "Hero" (Fig. 37) is a good example of the G.W. 
standard goods engine at the time. 

The narrow-gauge engineers having made frantic efforts to pro- 
duce locomotives as powerful as those in use on the Great Western' 




323 


1[~ "~TT " "I " j 




II IhTf 
ferS 




FIG. 37." HERO," A GREAT WESTERN RAILWAY SIX-COUPLED 
BROAD-GAUGE GOODS ENGINE 



Railway, the directors of the latter company decided to have a larger 
and more powerful engine constructed, and Mr. Gooch received orders 
to construct a colossal locomotive, and to have it in work before the 
commencement of the Parliamentary Session of 1846. From the 
time the decision was arrived at, until the " Great Western " was at 
work, only 13 weeks elapsed, during which short period the design 
of the engine had to be decided upon, the drawings made, the patterns 
prepared, and the whole of the complex machinery made and put 
together ; yet those three months were sufficient to produce this most 
famous- locomotive. 

As originally constructed, the " Great Western " (Fig. 38) was a six- 
wheel engine, the dimensions being: Cylinders, 18in. diameter and 
2 4in. stroke ; driving wheels, 8ft- diameter ; leading and trailing wheels, 
4ft. Gin. diameter; 278 tubes, 9ft. long, 2in. diameter; fire-box (out- 
side), 5ft. 6in. by 6ft., inside 4ft. lOin. by 5ft. 4in-, with partition 
through the centre; heating surface, tubes 1,591 sq. ft.; fire-box, 
160 sq. ft.; grate area, 20ft.; height, from level of rail to top of 



EVOLUTION OF THE STEAM LOCOMOTIVE 



107 



boiler, 9ft. Gin. ; the chimney was oft. 2 in. high ; length of engine, 
24ft. ; weight (empty), 36 tons. In this engine Gooch retained the 
Gothic fire-box, as supplied to the engines he had previously designed. 
By the way, a picture, purporting to be an illustration of this engine, 
was given in a book on locomotive history, with the flush 




FIG. 38. THE 



GllEAT WESTERN" BROAD-GAUGE ENGINE AS 
ORIGINALLY CONSTRUCTED 



top fire-box and four leading wheels! The "Great Western" con- 
tinued to work trains on the Great Western Railway until the end of 
1870, having run a total distance of 370,687 miles during the 23| 
years she was in work. 

On Saturday, June 13th, 1846, the "mammoth" locomotive (as 
the " Great Western " was usually called) made a sensational trip from 
London to Bristol and back, and, but for the failure of one of the six- 
feed pumps, necessitating slower running, even better results would 
ha,ve been attained. But, despite the accident, the result of the trip 
came like a " bolt from the blue " upon the narrow-gauge engineers. 

The train weighed 100 tons, and consisted of ten first-class 
carriages, seven of which were ballasted with iron, the other three 
being occupied by the directors and those interested in the experi- 
ment. The train started from Paddington at 11 hours 47 minutes 
52 seconds; at Didcot a stop of 5J minutes was made; Swindon was 
reached in 78 minutes. After staying there 4 minutes 27 seconds, 
the journey was continued to Bristol, the whole distance of 118^ 
miles being covered in 2 hours 12 minutes, or at the rate of 54 miles 
an hour, or, excluding the 9| minutes spent in the two stoppages, at 
about 59 miles an hour for the complete journey, including the slowing 
down and getting up speed again on three occasions. The maximum 
speed was obtained between the 82nd and 92nd mile-posts (from the 
80th to the 85th mile there is a falling gradient of 8ft. per mile, 
and from the 85Jth to about the 86Jth mile there is a falling gradient 



108 



EVOLUTION OF THE STtiAM LOCOMOTIVE 



of about 1 in 100, and a fall of 8ft. per mile then reaches to about 
the 90Jth mile-post; a rising gradient of 8ft. per mile then succeeds 
and extends beyond the 92nd mile-post), performing the ten miles 
in 9 minutes and 8 seconds, or at an average speed of nearly 66 
miles an hour. The 87th and 88th miles, on a falling gradient of 
8ft. per mile, were run over at a rate of 69 miles per hour. 

One Monday early in June, 1846, the "Great Western" was 
attached to the 9.45 a.m. express Paddington to Exeter, the crack 
train of that time, which, indeed, continued to be the fastest ordinary 
passenger train until the establishment of the "Flying Dutchman" 
many years later. When it was advertised that this train would 
perform the journey between London and Exeter in 4J hours, people 
said it was impossible; what, then, must have been thought of the 
run performed by the " Great Western " and chronicled below 1 The 
193 J miles from Paddington to Exeter were covered in 214 minutes 
(3 hours 34 minutes) running time, being an average rate of 55 J miles- 
per hour. The actual running time on the journey was as follows : 



From Paddington to Didcot ... 
Didcot to Swindon 
S wind on to Bath 
Bath to Bristol 
Bristol to Taunton 
Taunton to Exeter 



53 miles 

24 

29f 

11J 



111 

30f 
193| 



55 minutes 



14 
45 
37 

214 



The return journey was performed in less time, and could have 
been accomplished with ease at a rate exceeding 60 miles an hour. 
The actual running time, exclusive of stoppages, was as follows: 



From Exeter to Tauntou 
Taunton to Bristol 
Bristol to Bath 
Bath to Swindon 
Swindon to Didcot 
Didcot to Paddington 



30f miles 
443 



34 minutes 

43 

14 

34 

26 

56 



Miles, 193| 



Minutes, 208 



After the engine had been running a short time, Gooch found the 
weight on the leading axle too much to be safely carried by one axle, 
and he fitted another pair of leading wheels to the " Great Western '' 
(Fig. 3 9), making her an eight-wheeled engine, having a group of four 
wheels in front of the driving wheels. It must be remembered that these 
four wheels were not affixed to a bogie frame. So well satisfied were 
the directors of the Great Western Railway with the " Great Western " 
that 29 more engines of almost similar design (except the domed 



EVOLL'TION OF THE STEAM LOCOMOTIVE 



109 



lire-box) were constructed during the next eight years, and these 
engines, with a few of the same design, built at a more recent period, 
worked the famous broad-gauge expresses between London and New- 
ton Abbot until the abolition of the broad-gauge in May, 1892. 

In March, 1847, the Great Western Eailway laid down a length 
of line at Maidenhead for the purpose of testing Elijah Galloway's 




FIG. &. iJiE ORIGINAL "GREAT WESTERN" AS REBUILT WITH TWO 
PAIRS OF LEADING WHEELS 

system of locomotive propulsion with horizontal driving wheels. 
The horizontal wheels gripped a centre rail, and the engine not being 
dependent upon the weight placed upon the driving wheels for adhe- 
sion, was enabled to ascend inclines that were impossible for ordinary 
locomotives; whilst the fact that the two horizontal driving wheels 
were pressing one on either side of the centre rail enabled tne engine 
to safely pass round curves of extremely short radii, such as would be 
impossible with ordinary locomotives. The line put down at Maiden- 
head was on an incline of 1 in 19, but a model engine and train suc- 
cessfully ascended an incline of 1 in 6. Mr. D. Gooch gave the follow- 
ing account of the experiments: 

" Engineer's Office, Paddington, 

"March 25th, 1847. 

"The following is the result of the experiment I made with 
Mr. Galloway's locomotive engine, in which the driving wheels are 
placed horizontally, and act against the sides of a centre rail : 

Weight of engine 23 tons. 

Weight of load H* 



83| tons. 

" This weight was taken at a slow speed up an incline of 1 in 1 9, 
with a pressure on the boiler of 601b. on the inch, and calculating the 
power of the engine and actual duty performed, we have as follows : 
With steam at 601b. in the boiler, the average effective pressure on 
the pistons, after deducting back pressure, will be about 501b. on 



110 EVOLUTION OF THE STEAM LOCOMOTIVE 

the inch, then the area of the two cylinders 308 X 50 = 15, 4001 D., 
and double stroke of piston equals 32in., and circumference of driving 
wheel 11 Gin. 

"Therefore, as llGin. : 15,400 : : 32: 4,248 tractive power on the 
rim of the wheel, 

And gravity per ton, 1 in 19 = 118 lb, 
Friction ditto ... 7 lb. 

125 X 33-5 tons = 4,187'5 lb. 
resistance overcome. 

therefore, 4,248 - 4,187 = Glib., the total loss from the friction of 
the working parts of the engine, which I think, is as small a loss as 
can be hoped for in any class of engines, and from the facility of 
applying screws to increase the weight on the driving wheels to an/ 
required amount, there is no difficulty from slipping. 

" (Signed) DANIEL GOOCH." 

The " Atlas," constructed for the Manchester and Sheffield Railway, 
deserves notice. She was built by Sharp Bros, and Co., from the 
designs of Mr. Beyer, their then chief engineer, but afterwards head 
of the well-known firm of locomotive builders, Beyer, Peacock and 
Co., of Manchester. 

The "Atlas" commenced work in May, 1846, and during the suc- 
ceeding 17 months she travelled 40,222 miles, with a coke consump- 
tion of 36.531b. per mile, although engaged in hauling heavy goods 
trains. The engine had inside cylinders, 18in. diameter, 24in. stroke; 
the whole of the framing and bearings were inside the wheels; the 
boiler was 13ft. Gin. long and 3ft. Gin. diameter, and contained 175 
brass tubes of If in. external diameter; the wheels were cast-iron, 
4ft. Gin. diameter ; a copper fire-box was provided, its inside measure- 
ments being 3ft. Sin. long, 3ft. 3jin. wide, and 3ft. 4jin. from the 
fire-bars to the top. The water space around the fire-box was 3in., 
and a mid-feather, . 4in. wide, divided the fire-box. 

The cylinders were secured to each other by internal flanges, which 
formed the bottom of the smoke-box, and also the chief cross-stay 
between the frames. The valves were in one chest, located below the 
cylinders, and inclined towards each other. The weight of the 
valves was carried by spindles working through stuffing-boxes. The 
regulator was provided with two perforated discs, so that the steam 
was admitted very gradually, the volume increasing as the two sets of 
perforations came opposite each other. 

The weight of the " Atlas " was 24 txDns, and five other engines of 



EVOLUTION OF THE STEAM LOCOMOTIVE 111 

exactly similar designs were supplied to the Manchester and Sheffield 
Railway. 

Another engine of the same description was supplied to the Man- 
chester and Birmingham Railway, and on October 3rd, 1836, "No. 30" 
hauled a train of 101 wagons, weighing 597 tons, from Longsight to 
Crewe, a distance of 29 miles, at the average speed of 13.7 miles an 
hour. 

The mention of a powerful engine and a record train on one rail- 
way naturally suggests a better one on another line, so we have the 
"Essex" going "one better" than " No. 30." 

This time we have a load of 149 loaded wagons (probably equal 
to 390 tons), and forming a train nearly half a mile long. The 
"Essex" is also stated to have hauled a train of 192 empty trucks. 
The engine in question was built for the Eastern Union Railway by 
Stothart, Slaughter, and Co., Bristol, in 1847, and had wheels 4ft. 9 in. 
diameter, cylinders loin, by 24iii. stroke, weight 22 tons. 

In 1846, Stephenson and Co. supplied the South Eastern Railway 
with an engine called the " White Horse of Kent" (the " Whits Elephant 
of Newcastle " would have been a far more descriptive name). This 
engine probably exhibited the " long boiler " folly in a more marked 
manner than any other engine of that notorious class. She waa 
21ft. lOin. long, with a wheel base of only 10ft. 3 Jin. ! She had 
cylinders loin, by 22in. stroke, 5ft. 6in. driving wheels, and weighed 
18| tons. Gooch says this engine was so unsteady that it was 
necessary to be tied on to make experiments on the smoke-box temper- 
ature, and that the tubes were so long that one end of the engine was 
actually condensing the steam generated at the other end! 

At this time Mr. T. 11. Crampton turned his attention to loco- 
motive construction, and patented a design of locomotive. He claimed 
for his design the following advantages viz., a reduction of the rocK- 
ing and vibrating motion, obtained by lowering the centre of gravity, 
and by locating the greater portion of the weight between the sup- 
ports ; an increased heating surface ; and a superiority of arrangement 
of the working parts, the whole of which were placed immediately 
under the eye of the driver. 

The first engine constructed on this principle was the " Namur " 
(Fig. 40), built under Crampton's patent by Tulk and Ley, of tne 
Lowcra Works, Whitehaven, for the Namur and Liege Railway. 

The illustration shows that the chief peculiarity of the " Namur " 
waa the position of the driving wheels, the axle of which was behind 



112 



EVOLUTION OF THE STEAM LOCOMOTIVE 



the fire-box, so that the axle extended across the foot-plate. One 
spring, formed of plates, also extended across the back of the fire-box, 
parallel with and above the driving axle, and acting upon it at the 
bearings. 

The chimney was 6ft. Gin. high ; the smoke-box was very narrow, 
being no wider than the diameter of the chimney ; all the wheels had 
inside bearings; the cylinders were outside, and horizontal; the 
valve-chests were on the outer side of the cylinders, so that the eccen- 
trics were at the extreme ends of the axles, beyond the wheels, and 
quite exposed. 




FIG. 40. THE "NAMUR," THE FIRST ENGINE BUILT ON CRAMPTON'S 
PRINCIPLE 

The boiler barrel was surmounted by an immense fluted dome, 
which was fitted with two lever safety valves, whilst a third one, of 
the spring pattern, was provided on the fire-box casing. 

The following are the principal dimensions of the " Namur " : 
Diameter of driving-wheels, 7ft. ; diameter of leading and middle 
wheels, 3ft. 9in. ; total wheel base, 13ft.; cylinders, 16in. diameter, 
20in. stroke; number of tubes, 182 length lift., external diameter 
2in. ; fire-box, 4ft. 3in. long, 3ft. 5in. wide; area of fire-tube, 14ft. 6 in. ; 
heating surface : fire-box 62ft., tubes 927ft., total 989ft. 



EVOLUTION OF THE STEAM LOCOMOTIVE 



113 



The engine was completed early in February, 1847, and previous 
to its exportation, it was tried for several weeks on the London and 
North Western Railway, running over 2,300 miles. All classes of 
traffic were hauled by the engine, and she gave general satisfaction. 
A speed of 75 miles an hour was attained between Willesden and 
Harrow, when running "light." On another occasion, 50 miles r.n 
hour was attained on a trip from Camden Town to Wolverton with a 
coke train, weighing 50 tons, between Tring and Wolverton. 

The " Namur " weighed 22 tons, of which 7^ tons were on the 
leading wheels, 4 tons on the centre wheels, and 10J tons on the 
driving wheels- 

The L. and N.W.R. were so satisfied with the " Namur " that Tulk 
and Ley were instructed to build a Crampton engine for that railway ; 
and the " London " (Fig. 41) was produced in 1848 in response to this 




FIG. 



41. CRAMPTON'S "LONDON," THE FIRST ENGINE WITH A NAME 
ON THE SOUTHERN DIVISION OF THE L. & N.W.R. 



order. She was the first engine on the southern division of the L, and 
N.W.R. to have a name. The driving wheels were 8ft. diameter, the 
cylinders 18in. diameter and 20in. stroke. The boiler was oval in 
shape, its vertical diameter being 4ft. 8in., and its horizontal diameter 
3ft. lOin. The heating surface was 1,350 sq. ft. The fire-box ex- 
tended below the driving axle. 

In April, 1847, Mr. D. Gooch's famous broad-guage 
express engine, "Iron Duke," commenced to run. Fig. 42 repre- 
sentn an engine of this class. She was the first of a set of twenty- 
nine locomotives of almost similar construction, designed to work the 



114 



EVOLUTION OF THE ST8AM LOCOMOTIVE 




EVOLUTION OF THE STEAM LOCOMOTIVE 



115 



Great Western express trains. The " Iron Duke " was an improve- 
ment on the celebrated " Great Western," previously described ; the 
most noticeable difference was the absence of the domed fire-box in 
the " Iron Duke.'' The total mileage of this engine, up to October, 
1871, when it was withdrawn from service, amounted to 607,412 miles. 
The best-known engine of the class is "Lord of the Isles," built at 
Swindon in 1850, and exhibited at the International Exhibition, Lon- 
don, 1851 ; she commenced to run July, 1852, and continued inactive 
service on the Great Western Railway for 29 years, during which time 
789,300 miles were covered by the "Lord of the Isles." This famous 




FIG. 43. "No. 61," LONDON AND BRIGHTON RAILWAY 

broad-gauge locomotive is still preserved by the Great Western Railway. 
The next point in 'the evolution of the locomotive that deserves 
attention is the famous class of engines known as the "Jenny Lind" 
design. 

Much has been written concerning these, engines during recent 
years, and many uncorroborated and absurd statements have been 
made ; but it was most clearly demonstrated that to Mr. David Joy 
was due the chief honour of designing the successful class of loco- 
motive known far and near as "Jenny Linds." Such a design was 
elaborated from the adoption of the best features of the several 
descriptions of locomotives then in use. 

The first of the type of engine afterwards known as the 
" Jenny Lind " class was constructed for the London and Brighton 
Railway by E. B. Wilson and Co., Railway Foundry, Lseds, 
and was commenced building in November, 1846, and completed in 

i 2 



116 EVOLUTION OF THE STEAM LOCOMOTIVE 

May, 1847. The principal features of the engines may be summarised 
as follows: Steam pressure 1201b. per square inch, inside bearings 
to driving and outside bearings to the leading and trailing whaels, 
outside frames, outside pumps located between the driving and trailing 
wheels, and worked by cranks fixed on the outside of the driving axles. 
The engine had a raised fire-box; the dome was fluted and had a 
square seating; the safety valve was enclosed within a fluted column, 
and fixed on the fire-box. 

Polished mahogany lagging was used for both the boiler and fire- 
box, the same being secured by bright brass hoops. The tops of the 
safety valves and dome were bright copper. The first trip of the 
" Jenny Lind" was from Leeds to Wakefield and back. Ten engines 
of this class were supplied to the London and Brighton Railway, and 
were numbered 61 (Fig. 43) to 70. The principal dimensions were : 
Driving wheels 6ft. diameter; leading and trailing wheels, 4ft. dia- 
meter; cylinders (inside), 15in. diameter, 20in. stroke; boiler, lift, 
long, 3ft. Sin- diameter; 124 tubes, 2in. diameter. A water space of 
3in. was left between the inner and outer shells of the fire-box. Heat- 
ing surface, tubes 700 sq. ft., fire-box 80 sq. ft. 

It is significant to note that in the original description of the 
"Jenny Lind," published in 1848, we are informed that "in establish- 
ing this class of engine Messrs. Wilson have studied less the introduc- 
tion of dangerous novelties than the judicious combination of isolated 
examples of well-tried conveniences." This statement exactly agrees 
with those recently made by Mr. Joy. 

The great success of the " Jenny Lind " type caused Sharp Bros, and 
Co. to introduce a rival class of engines nicknamed " Jenny Sharps." 

The engines were provided with a mid-feather in the fire-box 
for the purpose of augmenting the heating surface. The principal 
dimensions of the "Jenny Sharps" were as follow: Steam pressure, 
801b. ; cylinders, 16in. diameter, 20in. stroke; driving wheels, 5ft. 6in. 
diameter ; heating surface, tubes (of which there were 161, each 10ft. 
long and 2in. diameter) 847 sq. ft., fire-box, 72 sq. ft. ; total, 919 sq. ft. 
Mi'. Kirtley, the locomotive superintendent of the Midland Railway, 
arranged a trial between the rival "Jennies," and the event came off 
on May 4th, 5th, and 6th, 1848. 

Sharp's engines were Nos. 60 and 61, and Wilson's Nos. 26 and 27. 
The first trip was with a load of 64 tons, made up of nine carriages 
and two brake-vans, weighted with iron chairs to 64 tons. 



EVOLUTION OF THE STEAM LOCOMOT1 V /; 117 

Sharp's No. 60 took the first train, the weight being, engine 
21 tons 9 cwt., tender 12 tons 11 cwt., load 64 tons; total, 98 tons, or, 
including officials, etc., about 100 tons. 

The journey was from Derby to Masborough, 40J miles, the lire 
rising for the first 20 miles at about 1 in 330, and falling for the 
remainder of the distance at about the same rate. The weather was 
fine, the metals dry, and there was no wind. 

William Huskinson drove the train, which l^ft Derby at 
3h. 39min. 5isec. p.m., and arrived at Masborough at 4.28 p.m. 
Among the passengers were Messrs. Kirtley, locomotive superintend- 
ent ; Marlow, assistant locomotive superintendent ; Harland, carriage 
superintendent; E. B. Wilson and Fenton, of the firm of E. B. Wilson 
and Co. ; and T. R. Crampton. 

The first 18 miles up the bank of 1 in 330 were covered in 
25 minutes 12J seconds, being at an average speed of nearly 43 miles 
an hour. Before starting, the water in the tender had been heated 
to nearly boiling point ; 16 cwt. of coke were consumed, or 44.8ib- par 
mile ; 10,2901b. of water were evaporated, equal to 5.71b. of water 

to lib. of coke. 

/ 

Wilson's engine, No. 27, was next tried. She weighed 24 tons 
1 cwt., and her tender, loaded, 15 tons 13 cwt., the total load with train 
thus being 103 tons 14 cwt. William Carter drove the train, which left 
Derby at 7h. lOmin. 20sec., and arrived at Masborough at 7h. 56min. 
42sec., the speed averaging 52 miles an hour. The first 18 miles w en- 
negotiated in 22 minutes 44f seconds, or at nearly 47 miles an hoar. 
Only 13 cwt. of coke was used, equalling 36.41b. per mile. 

The following table shows the working of the two engines up the 
bank to the seyenteenth mile-post : 







'Jenny 
Sharp." 


" Jenny 
Liad." 






" Jenny 
Sharp." 




" Jenny 
Linrt." 


Mile 
Post. 




Miles per 
hour. 


Miles per 
hour. 


Mile 
Post. 




Miles per 
hour. 




Miles per 
hour 


1 




21-6 


21.9 


10 




46-8 


.. 


520 


2 




39-6 


44-5 


11 




45-9 




51-4 


3 




42-0 


51-0 


12 




45-9 




5S-3 


4 




42-5 


51-4 


13 




45-6 




52-7 


5 




45-4 


51-4 


14 




46-6 




51-8 


6 


. 


46-8 


51-2 


15 









51-8 


7 




44-5 


48-9 


16 




" 48-0 




51-4 


8 




46-2 


50-0 


17 




47-0 




51.8 


- 9 


. 


47-0 


52-5 













Trials were then made with trains of 17 coaches, weighted to 
99 tons 16 cwt. Twenty passengers were carried, including Captain 
Symmons, the Government Inspector. The gross load was 101 tons. 



118 



EVOLUTION OF THE STEAM LOCOMOTIVE 



William Mould drove the Sharp engine, and William Barrow the 
Wilson 'engine (No. 26). 

The coke consumption was Sharp's, 16 cwt., or 44.81b. per mile; 
Wilson's, 12 cwt., or 33.61b. per mile. 

Water evaporated Sharp's, 10,8401b., equal to 27.11b. per mile, 
or 61b. of water by lib. of coke; Wilson's, 10,1161b., equal to 25.291b. 
per mile, or 7.51b. of water by lib. of coke. 

The first 18 miles up the bank were covered in 26 minutes 
19 seconds by the "Jenny Lind," and in 27 minutes 55 seconds by 
the "Jenny Sharp." 

The tables show the speeds at which the posts were passed : 





Sharp's 




Wilson's 




Mile 


Engine. 
Miles per 




Engine. 
Miles per 


Mile 


Pot. 


hour. 




hour. 


Post. 


1 


15-0 




18-3 


10 


2 


36-5 




40-9 


11 


3 


48-0 




45'6 


12 


4 


42-4 




46-8 


13 


>j 


43-9 




46-8 


14 


C, 


43-9. . 




46-2 


15 


7 


41-9 




434 


16 


8 


42-4 




43 4 


17 


9 


43-9 




44-5 


18 



Sharp's 
Engine. 


^ 


Vilson's 
Engine. 


Miles per 


1 


liles per 


hour. 




hour. 


43-9 




44-5 


44-5 




44-5 


43-9 




45-0 


43-4 




45-0 


43-4 




44-5 







45-0 


42-9 




43-9 


429 




42-4 


41-9 


.. 


41-4 



Beyond the thirtieth mile-post Wilson's engine, which had been 
considerably in advance, according to the time taken, began to lose 
ground, in consequence of the driver allowing the fire to get low, and 
upon arrival at Masborough he had scarcely sufficient steam to shunt 
tho train. 

Mr. Kirtley considered the trial unsatisfactory for this reason, and 
a second one was arranged for the next day, but with no more satis- 
factory result, as upon this occasion, after travelling a mile, a- joint 
cover of one of the cylinders worked loose, consequently a great deal 
Oj! steam escaped during the remaining 39 miles of the trip. We 
have given the real facts in connection with the original " Jenny 
Linds " at some length, for the purpose of placing on permanent record 
the details of these capital locomotives, and so prevent our readers 
and students of locomotive history generally from being misled by the 
absurdly inaccurate romances that have, for some obscure purpose, 
been recently circulated concerning' the " Jenny Lind." (Fig. 44.) 

The original design of the locomotive now to be described is s-.o 
singular that we are reminded of the extravagant examples of locomo- 
tive construction appertaining to 1830, or thereabouts, rather than io 
the year now under review. Yet, strange as it may appear, the "Corn- 



EVOLUTION OF THE STEAM LOCOMOTIVE 



119 



wall" (Fig. 45) is still running express trains, although it must be con- 
fessed it has undergone a complete metamorphosis since it was built 
at Crewe in 1847. The engine in question was designed by Mr. F. 
Trevithick, son of the famous " father of the locomotive," and was 
intended to be a narrow-gauge improvement on Gooch's famous 
" Great Western," as Trevithick wished to build a locomotive that 
would be able to attain a higher rate of speed than the renowned broad- 
gauge engine. To do this, he considered an increase of the diameter 
of the driving wheels, a sine qua non. He therefore constructed 
the " Cornwall " with driving wheels 8ft. Gin. in diameter. His next 



^/>^3C3 




- -ir~ rar 



I 



I,', G . 44. THE "JENNY LIND," A FAMOUS LOCOMOTIVE BUILT BY 
WILSON AND CO., LEEDS, IN 18'6 



proposition was that as 8ft. was then considered the limit of 
for driving wheels on the broad-gauge, with the boiler above the 
driving axle, it was necessary to place the boiler below the driving 
axle with wheels 8ft. Gin. diameter on the narrow gauge. And, there- 
fore, Trevithick constructed the " Cornwall," with underhung boilers, 
i.e., beneath the driving axle. The cylinders were outside, 17-Jin. dia- 
meter, with a stroke of 24in. The heating surf ace was 1,046 sq. ft. Ths 
locomotive was carried on eight wheels a group of four leading wheels, 
the driving, and a single pair of trailing wheels. Weight of engine in 
working order, 27 tons. The "Cornwall" was very successful 'v 



120 EVOLUTION OF THE STEAM LOCOMOTIVE 

attaining high rates of speed, and, indeed, far exceeded TVevithick's 
expectations in this respect. 

It has been stated that she attained a spaed equal to 117 
miles an hour down the Madeley Bank. Such a statement must be 
accepted with reserve not that the bond fides of the engineer who 
made it are doubted, but rather because of the difficulty of obtain- 
ing correctly the exact speed of engines when travelling at a great 
rate, even when proper instruments are employed. We krow that 




FlG 45. TREVITHICK'S " CORNWALL." WITH SFT. 6lN. DRIVING WHEELS, 
AND BOILER BELOW THE DRIVING AXLE 

with an ordinary watch cornet results are almost impossible, and an 
error of a second or two when calculating a quarter of a mile vill 
make a very great difference when arriving at the approximate rates 
in miles per^ hour. However, be this as it may, it is generally 
acknowledged that the " Cornwall " attained speeds that may fairly 
be called phenomenally high. 

On November 9th, 1847, the "Cornwall" was hauling v -a goods 
train from Liverpool, and upon rounding the curve near Winsford 
Station, ran into a coal train, the result being the death of the driver 
of the " Cornwall," the engine being thrown across both lines, whilst 
rhe tender and trucks were projected over the engine, and did not 
come to a standstill for- several yards. 

The " Cornwall " was one of the features of the first International 
Exhibition (held in Hyde Park, London, in 1851). In 1862 Mr. 



EVOLUTION OF THE STEAM LOCOMOTIVE 



121 




122 EVOLUTION OF THE STEAM LOCOMOTIVE 

J. Kamsbottom rebuilt the "Cornwall," and placed her new boiler over 
the driving wheels. She was numbered "173," and still .works the 
three-quarter-of-an-hour express trains between Liverpool and Man- 
chester. She completed her jubilee of active service last year, and 
is still running. The present number of the "Cornwall" is "3020," 
and she is now only a six-wheeled engine. 

McConnell made an experiment in counterbalancing a locomotive 
on the London and North Western Railway in 1848. The engine ; n 
question was the "Snake," No. 175, built by Jones and Potts on 
Stephenson's long-boiler principle. McConnell's plan was to provide a 
connecting-rod attached to a block working between slide bars, on the 
opposite side of the driving axle to that on which the piston, etc., 
were located. By this method he considered that, providing his extra 
rod-block, etc., weighed the same as the pistons and other recipro- 
cating parts, he had attained a perfect method of counterbalancing. 
The result was a rude disillusion of the idea, and a complete wreckage 
of b^th the theory and the " Snake," the engine breaking down on its 
first trip, after being fitted with this reciprocating counterbalance. 
The only result of such an addition to the " Snake " was an increase 
in the weight of the engine and an augmentation of the friction and 
axle strains. 

In the spring of 1848 McConnell built an engine which he expected 
" to prove the most powerful narrow-gauge engine ever yet built." 

It had outside cylinders 18in. diameter, and 7ft. Gin. between 
centres. The driving wheels were 6ft. diameter, leading and trailing 
3ft, lOin. The boiler was 4ft. 3in. external diameter, 12ft, 7in. long, 
and contained 190 tubes of 2in. diameter. Height of top of boiler 
from rail level, 7ft. &n. 

The fire-box was 5ft. 9Jin. wide, by 5ft. 5in. long, and of the same 
height. The wheel base was as follows : Leading to driving, 6ft, Sin. ; 
driving to trailing, 10ft. 6in. 

Another combination design in locomotive practice is to be found 
in engine "No. 185," delivered to the York, Newcastle, and Berwick 
Railway on October 3rd, 1848, by R, Stephenson and Co. 

This engine had inside cylinders, but outside valve gearing and 
eccentrics". The cylinders were 16in. diameter, with 20in. stroke. The 
boiler was 3ft. lOin. diameter and lift, long; there were 174 tubes, 
If in. outside diameter, lift. 5in. long, the heating surface being: 
tubes, 964 sq. ft, ; fire-box, 82 sq. ft. The driving wheels were 6ft. 6in. 



EVOLUTIOX OF THE STEAM LOCOMOTIVE 123 

diameter, the leading and trailing being 3ft. 9in. diameter. Inside 
bearings were provided for the driving wheels and outside bearings 
for the leading and trailing wheels. Inside and outside iron-plate 
frames, lin. thick and Sin. deep, were provided. This engine 
weighed 22 tons in working order, and consumed 181b. of coke per mile 
with express trains of four carriages. The peculiar feature of 
" No. 1 85 " was the vertical valves, worked by eccentrics outside the 
driving wheels; the pumps were also worked off the same eccentrics, 
and were consequently outside, as in the " Jenny Lind " design. The 
exhaust ports were below the cylinders, the pipes from which united 
at the blast orifice. 




F.o. 47." OLD COPPER-NOB," No. 3, FURNESS RAILWAY, THE OLDEST 
LOCOMOTIVE NOW AT WORK 

Locomotives that attain their "jubilee" of active service are 
indeed very few and far between, and it redounds much to the honour 
of the late firm of Bury, Curtis and Kennedy, of the Clarence Foundry, 
Liverpool, that locomotives constructed by them in the year 1846 are 
still engaged in hauling trains on an English railway. 

This firm of builders ceased to exist 46 years ago, but engines 
Nos. 3 (Fig. 47) and 4 of the Furness Railway are continuing monu- 
ments of the good material and sound workmanship of Bury, Curtis and 
Kennedy. The locomotives in question are mounted on four wheels 



124 EVOLUTION OF THE STEAM LOCOMOTIVE 

(coupled) of 4ft. 9in. diameter, the cylinders are 14in. diameter, and 
stroke 24in., the valves being between the cylinders. The wheel 
base is 7ft. Gin. The boiler is lift. 2in. long, with a mean diameter 
of 3ft. 8in., and contains 136 tubes of 2in. diameter, the total heating 
surface being 940 sq. ft. Steam pressure, llOlb. The tenders are 
carried on four wheels of 3ft. diameter, the wheel base being 6ft. 9in. 
The tank holds 1,000 gallons of water, and the coal space is 100 cubic 
feet. The engines weigh 20 tons each, and the tenders 13 tons each. 

The prominent " Bury " features bar framing and round back 
fire-boxes with dome tops are, of course, en evidence. 

The chimneys appear abnormally high when viewed side by sida 
with modern engines ; whilst the pair of Salter safety valves with 
long horizontal arms, the one reaching from the centre to the back of 
the fire-box, and its fellow continuing to the front, are also noticeable 
objects. These engines are usually employed in shunting goods trains 
in the Barrow Docks and goods yards, and are locally called the 
" old copper nobs." 

Two further peculiarities of these Bury engines are worth recording 
viz., the splashers, which are extended in a curious way over the rear 
of the wheels, and reach within a few inches of the rails, and the round 
" old copper nobs." 

The period under review was a time of considerable competition 
between the rival gauges, and this competition naturally led to the 
projection of various extraordinary designs in locomotive construction, 
such designs being the results of the efforts made by the narrow- 
gauge engineers to equal the splendid broad-gauge locomotives then 
recently introduced. 

During the first weeks of 1848 E. Wilson and Co., of the Railway 
Foundry, Leeds, turned out a remarkable specimen of locomotive con- 
struction ; the engine in question was named " Lablache " (after a 
celebrated singer). This locomotive had two inside cylinders 16in. 
diameter, 20in. stroke, and was supported on four wheels each 7ft. 
diameter; the wheel base was 16ft. 

It is necessary to describe the mode of working introduced into the 
"Lablache." Between the two pairs of wheels was a straight bar, or 
shaft, extending under the boiler, parallel with the axles, and pro- 
jecting on each side beyond the frames. Between the frames two 
levers were attached to this shaft, and the other extremities of these 
levers were attached to the pistons by the usual piston-rod and con- 



EVOLUTION OF THE STEAM LOCOMOTIVE 125 

necting-rods. Now comes the difference in working ; the driving axle, 
it will be observed, was not cranked, but provided with arms. The 
axle did not revolve, but simply vibrated backwards and forwards. 
Outside the frames were double-ended levers, one end being coupled 
to a crank on the leading wheel, and the opposite end connected in a 
similar manner to the trailing wheels. The wheels on both sides of the 
locomotive were connected in the same way that a rotary motion is 
communicated to a lathe by a treadle. When first constructed india- 
rubber springs were provided for this engine's bearings. 

Another engine of a similar design was built, but much lighter. 
It ran upon the York, Newcastle, and Berwick line for some years. 
We may say that no other engine on this system was ever built. 
With a train of three carriages, an average speed of 75 miles an hour 
is said to have been maintained between Rugby and Leicester. This 
was, however, due to the high pressure of the steam. Upon another 
occasion 80 miles an hour was attained ; and the engine hauled 
a train of 53 loaded wagons, weighing 430 tons, at an average speed 
of 30 miles an hour. After some little time, the fire-box of tlie 
" Lablache " was destroyed, and she was then returned to the Railway 
Foundry, and altered into a four-coupled engine of the usual type, and 
sold to a railway contractor. 

Another locomotive of peculiar design now deserves notice. At a 
first glance it might be supposed that the " Albion " was propelled on 
the same principle as the "Lablache" previously described. Such is 
not, however, the case, the machinery being of an entirely different 
character. We have been fortunate enough to secure the original 
working drawings of the " Albion " and the three other engines con- 
structed on the same method, designated the " Cambrian " system. A 
patent for this method of working steam engines was obtained in 
1841 by Mr. John Jones, of Bristol, and applied to stationary engines. 

Broadly speaking, the modus operandi is as follows : A central 
shaft is provided, extending under the boiler of the locomotive and 
projecting beyond the frames on both sides. Between the frames the 
shaft passes through a segment al cylinder, within which and fitted to 
the shaft was a species of disc piston, made to vibrate throughout the 
length of the hollow segment of the cylinder. It will, therefore, fee 
observed that the motion was obtained from a vibrating disc engine, 
the blades of which were fixed on the driving shaft ; the difference 
between Wilson's locomotive and the ones we are now describing being 



126 EVOLUTION OF THE STEAM LOCOMOTIVE 

that the former was actuated by two horizontal engines working a 
rocking shaft by connecting-rods, whilst the latter were driven by a 
disc engine, fixed directly upon the rocking shaft. The arrangement 
for connecting the driving wheels with the shaft was very similar 
in both classes of engines. 

The premier "Cambrian" locomotive was named "Albion" (Fij. 
48), and was built in 1848 by Messrs. Thwaites Bros-, of the Vulcan 
Foundry, Bradford. She was a six-wheel engine, the leading and 
middle pairs of wheels both receiving motion by means of the con- 
necting-rods from the outside levers attached to the driving shaft. 
Tho top of the fire-box was considerably above the level of the top of 
the boiler barrel. Upon this raised fire-box was fitted a steam dome 
with a square seating, above the dome was an enclosed Salter pattern 
safety valve. 

The principal dimensions of the "Albion" were: Leading anil 
driving wheels, 5ft. Gin. diameter, and trailing, 3ft. 9in. diametsr : 
wheel base leading to driving, 9ft. Gin. ; driving to trailing, 5ft. Sin. ; 
boiler, 12ft, long, containing 149 tubes; throw of cranks, 20in. 

it should be observed that the "Albion" was fitted with the "link"' 
motion. 

The patentee claimed the following advantages for locomotives 
built on the "Cambrian" system viz., perfect balance of working 
parts, thus entirely doing away with the centre pressure and strain ; 
the complete avoidance of all dangerous oscillation; the ends of the 
oscillating levers, in passing through the greater part of a circle, 
gained increased power at the extremities of the stroke, and so com- 
pensated for the loss of power in the cranks as they approached the 
dead centres. 

This is explained by observing that as the lever approaches the 
extremities of the stroke the actual length diminishes, and becomes 
from 18in. to l7lin., IGin., 15Jin., 14in., and 13Jin. at the centres, 
so that the power of the lever increases in proportion to its diminution 
in length. 

The wear and tear of the machinery was less than in an ordinary 
locomotive, there being fewer working parts, whilst the centre of 
gravity was considerably lowered. 

The above advantages summarised amounted to the advantages 01 
the long-stroke crank without a long-stroke cylinder, and consequently 
the absence of a high-piston velocity. 



EVOLUTION 0V THE STEAM LOCOMOTIVE J27 




128 EVOLUTION OF THE STEAM LOCOMOTIVE 

The "Albion" made its initial trip in June, 1848, the length of 
line selected being from Bradford to Skipton, "on the Leeds and Brad- 
ford .Railway; the distance was about 18 miles. The speed 
attained and the low fuel consumption are stated to have more than 
satisfied the builders and others concerned. The " Albion" was 
afterwards tried on the Midland Railway between Derby and Birming- 
ham, and the result of these trials showed that the coke consumption 
was 51b. per mile less than with the ordinary locomotives, although 
the trains hauled were of greater weight than usual. We have been 
unable to obtain further details of the working of this interesting loco- 
motive. The patentee appears to have sent details of the duties per- 
formed by the "Albion" to the Institution of Mechanical Engineers 
in 1849; but these were not printed in the "Proceedings," nor is 
the Secretary of the Institution now able to find any trace of the papers 
in question among the archives of the Institution. Messrs. Thwaites 
Bros., the builders, inform us that about 30 years ago the engine in 
question was working at Penistone, near Sheffield, and that she v. as 
afterwards taken over by the Manchester, Sheffield and Lincolnshire 
Railway. 

Unfortunately, the locomotive department of that railway does 
not appear to have preserved any particulars relating to the "Cambrian" 
locomotive after it came into the possession of the Manchester, Shef- 
field and Lincolnshire Railway. 

Th'3 other three engines with " Cambrian " machinery were tank 
locomotives. Two of these were propelled in a similar manner to tne 
" Albion," the segmental cylinder being below the frames, and located 
between the driving and leading wheels, both pairs of which were 
5ft. Sin. diameter, the trailing wheels being 3ft. 9in. diameter. One 
of these two tank engines had a raised fire-box, similar to that of the 
" Albion " ; but the other' had a " Gothic " fire-box, with the wood 
lagging exposed to view. The other features of the former were a boibr 
12ft. long, and a steam dome on the fire-box, fitted with two Salter 
safety valves, placed side by side. This engine had the " link '' motion. 
Three water-tanks were provided, one beneath the foot-plate, the 
second below the frames between the leading and driving wheels, and 
th0 third extended from the front of the leading axle under the 
smoke-box, and terminated at the buffer beam. The wheel base was, 
.L. to D. 9ft. 6in., D. to T. 5ft. Sin, The engine had inside frames and 
bearings. 



EVOLUTION OF THE STEAM LOCOMOTIVE 129 

The locomotive with the " Gothic " fire-box was fitted with a gab 
reversing gear, worked off the leading axle; the throw of the cranks 
was 19in. The boiler was 12ft. long and 3ft. Sin. diameter, and con 
tained 121 tubes. Two water-tanks were provided one beneath the 
foot-plate, the other below the frames between the leading and driving 
wheels. The wheel base of this engine was, L. to D. lift., D. to T. 
oft. Sin. 

The third Cambrian tank engine of which we possess the drawings 
was a six-wheel locomotive, with single driving wheels 5ft. Gin. dia- 
meter, the leading and trailing wheels being 3ft. 9in. diameter. The 
wheel base was 15ft. 5in., equally divided. 

This engine also had a " Gothic " fire-box, and was provided with 
a sledge brake, which acted on the rails between the driving and 
trailing wheels. The reversing gear was of the fork pattern. The 
water tanks were fixed one below the foot-plate, the other beneath 
the frames, between the driving and trailing wheels. The 
boiler was lift. 2in. long, and contained 126 tubes- The machinery 
in this engine was arranged in an entirely different manner, 
the segmented cylinder being below the smoke-box. The 
driving shaft passed through the cylinder, and projected beyond 
the frames on either side of the engine, and vibrated in an arc, as did 
that of the " Albion " ; but instead of a lever being attached to each 
end of the cranks, the latter only extended in one direction, so that 
at one end the crank was fixed on the driving shaft, while to its other 
extremity was pivoted a connecting-rod, 4ft. long, the other end of 
which was pivoted on a vertical arm, the upper end of this arm being 
attached to the frame by a horizontal bolt, on which it hung. It is 
very difficult to explain the method of propulsion without a drawing, 
but it will be understood that the connecting-rod frona the driving 
shaft to the hanging-rod only vibrated. Another crank, 6ft. long, was 
also attached to the bottom end of the vertical swinging-rod ; thf 
ol her end of this crank was connected with the driving wheel by means 
( f the usual outside pin. It will, therefore, be seen that by means of 
the hanging-rod the vibrating motion was transformed into a rotary 
one. The feed-pumps were worked off the vertical rod, the motion of 
which was similar to that of a pendulum, with the connecting-rods 
fastened to its bottom end. The drawings of these 'four remarkable 
locomotives are on a large scale, and are well executed j parts of them 
being coloured, they are also mostly in a good state of preservation. 



CHAPTER IX. 

The era of "light" and combination locomotives Samuel's "Lilliputian" and 
"Little \Vonder '' The broad -gauge " Fairfield," constructed by Bridge* 
Adams Samuel's "' Enfield" Original broad-gauge "singles" converted into 
tank engines The rise of "tank" engines, "saddle," and "well" Adams' 
"light" engines on Irish railways The Norfolk Railway adopts them Eng- 
land's "Little England" exhibited ut the 1851 Exhibition- Supplied to the 
Edinburgh and Glasgow, the Liverpool and Stockton, Dundee and Perth, 
and Blackwall Railways Hawthorne's " Plews " for the Y.N. and B.R. 
Crampton's monster "Liverpool" Taylor's design for a locomotive-- Pear- 
son's prototype of the " Fairlie " engine Ritchie's non-oscillating engine 
Timothy Hackworth again to the front His celebrated " Sanspareil, No. 2 " 
His challenge to Robert Stephenson unaccepted- -Bury 's " Wrekin " Cale- 
donian Railway locomotive, No. 15 "Mac's Mangle" on the L. and N.W.R. 

MANY curious contrivances were introduced into the construction 
of the locomotive abouiT the period now under review. Among these 
early proposals for the improvement of locomotion, few are more 
interesting than the combined locomotive and carriage introduced 
some fifty years ago by Mr. W. Bridges Adams. 

Mr. Adams had a wide experience of every section of railway con- 
struction. Indeed, in the preface to one of his books, in writing of his 
experience, he says that he had "years of practical utility in planning 
the construction of nearly all machines that run on roads and rails also 
from navvy's barrow up to a locomotive engine." 

Nor are Mr. Adams's contributions to railway literature inconsider- 
able, for, besides writing several books between 1838 and 1862, he was 
at one time editor of a periodical, and also wrote voluminously under 
the pseudonym of " Junius Redivivus." 

Having thus briefly mentioned Mr. W. B. Adams as being entitled 
to a far more important position in the evolution of our locomotives 
than is usually accorded him, we will now proceed to discuss the 
subject of combined locomotives and railway carriages, of which 
Mr. Adams was the chief advocate. The first machine of the kind, 
however, appears to have been constructed by Mr. Samuels, of the 
Eastern Counties Railway, for the purpose of quickly and economically 
conveying the officials of the railway over the system. 

This engine was apparently called both the "Lilliputian" 
and the "Little Wonder." It was constructed in 18-i7, and made its 



EVOLUTION OF THE STEAM LOCOMOTIVE 131 

first trip to Cambridge on Saturday, October 23rd, leaving London at 
10.30 a.m., and reaching the University town at 2.45 p.m. Stops 
were made at three iSttermediate stations for water, etc., which occu- 
pied about half an hour, so that the 57 J miles were covered in about 
105 minutes' running-time. 

The total length of the "Little Wonder" was 12ft. 6in., in which 
space was included the boiler, machinery, water-tank, and seats for 
seven passengers. The frame was hung below the axles, and carried 
on four wheels 3ft. 4in. diameter. 

The floor was 9in. above rail level. The machinery consisted of 
two cylinders, 3 Jin. diameter, and placed one on each side of the 
vertical boiler ; the driving axle was cranked. The stroke was Gin. 
The boiler was cylindrical in shape, 19in. diameter and 4ft. 3in. high; 
it contained 35 tubes, 3ft. Sin. long and Ijin. diameter; the tube 
heating surface being 38 sq. ft. The fire-box was circular in shape, 
16in. diameter and 14in high, its heating surface being 5J sq. ft. 

The link motion, feed pumps, etc., were provided. The water-tank 
held 40 gallons, and was placed under the seats. The usual speed of 
the "Little Wonder" with a full load was 30 miles an hour; and as 
high a rate as 44 miles an hour was often attained. The coke con- 
sumption was only 2 Jib. per mile. The weight of the whole vehicle, 
including fuel and water, was only 25 J cwt. 

Samuels' initial effort with light locomotives having been so suc- 
cessful, "it occurred to him that branch traffic could be much more 
cheaply worked by means of a combined engine and carriage, instead 
of the usual locomotive and train of carriages. 

Mr. Adams also had for some time been in favour of a combination 
of the kind, and Mr. Gregory, the engineer of the Bristol and Exeter 
Railway, was also in favour of the system being tried on the short 
branches of that railway, the passenger carriages on one at least of 
which were at that time drawn by horses. Acting upon the advice o; 
Mr. Gregory, the directors of the Bristol and Exeter Railway ordered 
Mr. Adams to construct a vehicle and engine for working the traffic 
on the Tiverton branch. The machine was completed in December, 
1848, and a satisfactory trial of it was made upon the broad-gauge 
metals of the West London Railway. This combination, which was 
constructed by Mr. Adams at Fairfield Works, Bow, E., was called 
the " Fairfield " (Fig. 49), and was brought into use on the Tiverton 

branch on December 23rd, 1848. 

K 2 



132 EVOLUTION OF THE STEAM LOCOMOTIVE 

Its length was 39ft., and the boiler was placed in a vertical posi- 
tion. The driving wheels were 4ft. Gin. diameter, and were originally 
made of solid wrought iron. The middle and trailing wheels, 3ft. Gin. 
diameter, were of wood, and loose on their axles as well as their 
journals, the middle wheels having a lateral transverse of Gin. 




FIG. 49. THE "FAIRFIELD," ADAMS'S COMBINED BROAD-GAUGE ENGINE 
AND TRAIN, FOR THE BRISTOL AND EXETER) RAILWAY 

The boiler was vertical, 3ft. in diameter and 6ft. high, and con- 
tained 150 tubes; the fire-box was 2ft. high and 2ft. Gin. in diameter. 
The cylinder was Sin. diameter, with 12in. stroke. The connecting- 
rods worked on a separate crank shaft, which communicated with the 
driving wheels by side rods, the axle of the driving wheels being 
straight, with crank pins on the outside. 

The boiler was placed behind the driving axle, the tank, capable 
of holding 200 gallons of water, being in front of it ; and the coke- 
box was attached to the front part of the carriage behind the 
driver. The working pressure was lOOlb. 

The bottom of the framing was within 9 in. of the rails, so that 
by keeping the centre of gravity low greater safety might be ensured 
at high speed, and freedom from oscillation obtained. 

The first-class carriage was in the form of a saloon, and accom- 
modated sixteen passengers ; whilst the second-class compartment 
seated thirty-two. The entire weight of the machine was about 10 
tons, and when occupied with forty-eight passengers it amounted to 
about 12 J tons. 

On the experimental trip, on December 8th, 1848, the " Fairfield'" 
left Paddington Station at 10.30 a.m. for Swindon, 77 miles down 
the line, with a party of gentlemen connected with various railways. 
Mr. Gooch officiated as driver on both the up and down journeys. 

Though the rails were greasy from the prevailing rain, in addition 



EVOLUTION OF THE STEAM LOCOMOTIVE lib 

ro a head wind and, what was worse, a leak in the boiler the 
machine soon attained considerable speed, and for a portion of tho 
\viiy reached the rate of 49 miles an hour. On arriving at Swindon 
the fire was extinguished, the leak partially repaired, and, after a 
reasonable sojourn, the party returned to town. The run back was 
exceedingly satisfactory, the speed of 49 miles being maintained for 
a considerable part of the way, the passage from Slough to Paddiagtoo 
being performed in 30 minutes. 

As previously stated, the crank-shaft was unprovided with wheels, 
the motion being conveyed to the driving wheels by means of craiks 
fixed on the outsides of the driving axle, and connected to similar 
cranks on the driving wheels by means of cannetting-rods. 

This method has erroneously been called " Crampton' s system," but 
it should be noticed that Adams used it for several years previous to 
Crampton adopting the plan in question. These combined engines 
and carriages were, in fact, built under a patent obtained by Mr. 
Adams in 1846, .and, therefore, some time before Crampton adopted 
the- inside cylinder and intermediate driving shaft. 

It was found in practice that the vertical boiler of the " Fairfield " 
was not a success, so after some nine months' trial it was replaced 
by a horizontal tubular boiler. Then, after further experience, several 
drawbacks to the efficient working of branch line traffic by means of 
the combined- engine and carriage were evident. So the engine was 
disconnected from the carriage and given an extra pair of wheels, arid 
became, in fact, a miniature four-wheeled tank locomotive, a style of 
engine Adams afterwards became noted for building. 

Mr. Samuel having obtained the sanction of the directors of the 
Eastern Counties Railway, Mr. Adams constructed a locomotive car- 
riage for the Enfield branch traffic. The "Enfield" (Fig. 50), in 
appearance resembled a four-wheel tank engine and a four-wheel car- 
riage, built together on a continuous frame, instead of being connected 
by couplings and buffers. 

The whole framing, with the exception of the two buffer bars, was 
01 wrought-iron, and was 8ft. Gin. in width, bound together by deep 
cross-bars. 

The engine was of the outside cylinder class. The cylinders were 
Tin. in diameter, with a 12in. stroke. They were simply bolted down 
to the surface of a stout wrought-iron plate, in the middle of which 
the boiler was placed. 



134 



EVOLUTION OF THE STEAM LOCOMOTIVE 



The driving wheels were 5ft. in diameter, and, as well as the front 
pair of wheels of the carriage, were without flanges, those of the 
leading engine wheels and the hind pair of the carriage being sufficient 
to retain the engine on the rails, whilst greater freedom was thus 
obtained for passing around curves. The boiler was constructed in 
the usual manner, and was 5ft. in length by 2ft. Gin. in diameter, and 
.had 115 IJin. tubes 5ft. Sin. long, giving 230ft. of tube-heating sur- 
face. The dimensions of the fire-box were 2ft. 10 Jin. by 2ft. Gin., 
being an area of 25 sq. ft., making the total heating surface 255 sq. ft. 
The water was carried below the floor of the carriage in wrought-iron 
tubes 12in. in diameter and 12ft. long. 

The coke was carried in a chest placed behind the foot-plate of the 
engine and immediately in front of the carriage head. The side 
frames were ingeniously trussed by diagonal bars of iron, and were 
thus rendered of great strength without adding much weight to the 
machine. 




FIG. 50. THE "ENFIELD," COMBINED ENGINE AND TRAIN FOR THE 
EASTERN COUNTIES EAILWAY 

The leading engine wheels, together with the running wheels of 
the carriage, were 3ft. in diameter. The carriage was divided into four 
compartments, the two middle ones being for first-class and the two 
external ones for second-class passengers. The guard's seat was on 
the top of the carriage head. A vertical shaft with a hand-wheel on 
its upper end passed down the side of the head, and was connected 
beneath the framing with two transverse rocking shafts, carrying the 
brake blocks, placed one on each side of the driving wheels, thus 
giving the guard a ready means of control over the speed of the 
engine. 

To bring up the buffers to the line of those of ordinary carriages, 
separate timber beams were passed across each end of the carriage, 
the front one being supported by neat wrought-iron brackets, rising 



EVOLUTION OF THE STEAM LOCOMOTIVE 135 

from the framing. The total weight of the whole was not more than 
10 tons, including its supply of coke and water, and accommodation 
was afforded for 42 passengers, to convey which, at 40 miles per hour, 
the calculated consumption of coke was 71b. per mile. 

Mr. Samuel stated that the accommodation provided by the com- 
bined engine and carriage was not sufficient for the traffic, so two 
additional carriages (one with a guard's compartment) were added, 
the train thus having accommodation for 150 passengers. The 
"Enfield" worked this train regularly at 37 miles an hour speed. 

From January 29th to September 9th, 1849, the train travelled 
14,021 miles, and was in steam 15 hours daily, but only five of which 
were spent in running. The total time in steam during the above 
period was 2,162 hours, the total coke consumed being 1,437 cwt., o.t 
which 743 cwt. was consumed in running, 408 in standing, and 286 in 
raising the steain. The average coke consumption per mile was 
11.481b., but a considerable portion of this was spent in standing, the 
actual consumption for running being only about 61b. per mile. 

In addition to the passenger traffic, the "Enfield" hauled all the 
goods and coal traffic on the branch, which, during the period under 
review, amounted to 169 tons of goods and 1,241 tons of coal. On 
June 14th, 1849, the "Enfield" took the 10 a.m. train from Shore- 
ditch to Ely, 72 miles, the train consisting of throe passenger carriages 
and two horse-boxes'; but the "Enfield" arrived eight minutes before 
time, and the coke consumed only amounted to 8f Ib. per mile for the 
trip, including that used in raising steam. 

When tried between Norwich and London, the " Enfield " performed 
the journey of 126 miles in 3 hours 35 minutes, including stoppages. 
An ordinary train had, at that time, never made the journey so quickly. 

Although the "Enfield" appeared to use so little fuel, the broad- 
gauge " Fairfield " does not seem to have been an economical machine. 
A special trial was made between Gooch's famous 8ft. single "Great 
Britain" and "Fairfield," between Exeter and Bristol. A loaded 
wagon weighing 10 tons was drawn by the "Fairfield," making a total 
weight of 26 J tons, of which the engine portion can be reckoned at 
9 tons and 17 tons for the weight of the train. The distance is 
76 miles, and the time allowed for the 8 a.m. train, including ten stops, 
was 2 hours 35 minutes; but the "Fairfield" took 3 hours 17 minutes 
to cover the distance, and consumed 131b. of coke per mile, only 6.3!b> 
of water being evaporated for each pound of coke. 



136 EVOLUTION OF THE STEAM LOCOMOTIVE 
The duty performed by the two locomotives is thus tabulated: 

Consumption 

Load in Coke per of Coke per 

tons. mile. ton per mile. 

11 Great Britain" 100 ... 26 Ib. ... 0-26 Ib. 

"Fairfleld" 17 ... 13 Ib. ... 0-76 Ib. 

But, in comparison with the old " Venus," the " Fairfield " conies 
out no better. 

The "Venus," it will be remembered, was one of the original 
broa'd-gauge engines built for the Great Western Railway by the 
Vulcan Foundry Company, with 8ft. driving wheels. This engine had 
her driving wheels reduced to 6ft. diameter, and a small water-tank 
fitted on the foot-plate in place of a tender, thus being converted into 
<i six-wheel " single " tank engine. The " Venus " only used 14:lb. of 
coke per mile in working the Tiverton branch; while the " Fairfield" 
consumed 191b. of coke per mile on the same work. The evaporating 
powers of the "Venus" had been greatly improved since N. Wood' a 
experiments in 1838, as at that time she consumed 52.71b. of coke per 
mile run. 




FIG. 51." BED STAR," A 7*T. SINGLE BROAD-GAUGE SADDLE TANK 
ENGINE. CYLINDERS, 16in. BY 18in. 

In addition to " Venus," several other of the early broad-gauge 
locomotives were reconstructed as tank engines. Fig. 51 ("Red 
Star") is a good example of the peculiar tank locomotives on the 
G.W.R. 60 years ago. 

In addition to the "Fairfield" and "Enfield," combined engines 
and carriages were constructed by Mr. Adams for several other rail- 
ways. One for the Cork and Bandon Railway had cylinders 9in. dia- 
meter, and accommodation for 131 passengers. This engine was con- 



EVOLUTION OF THE STEAM LOCOMOTIVE 



137 



structed in such a mariner as to enable it to run independently of the 
carriage. Another engine and carriage was built for a Scotch railway, 
and was guaranteed to work at 40 miles an hour. But the advantage 
of having the engine separate from the carriage was so great ihut 
Mr. Adams soon ceased to build the combination vehicles, and instead 
constructed his celebrated "light" locomotives; these, and the some- 
what similar "Little England" engines, built by England and Co., 
were at one time very popular. 

Fig. 52, representing " No. 148," one of the first batch of outside 
cylinder engines on the Southern Division of the L. and N.W.R., shows 
also a good example of Stephenson's " long boiler " locomotive. " 148 '' 
was built by Jones and Potts, of Newton-le-Willows in 184Y. The 
cylinders were 15in. diameter, the stroke being 24in. The driving 




FIG. 52. "No. 148," LONDON AND NORTH WESTERN RAILWAY; AN 
EXAMPLE OF STEPHENSON'S "LONG BOILER" ENGINES 

wheels were without flanges, and were 6ft. Gin. in diameter. Tlie 
le{?din</ wheels were 4ft. diameter. This engine was destroyed in a 
collision at Oxford on January 3rd, 1855, in which accident seven 
people lost their lives. 

At this period a fashion for "tank" engines had become prevalent, 
and most of the locomotive builders produced designs, each having 
characteristic features. Thus Sharp Brothers and Company's 
"tank" engines had outside cylinders, with the tank between the 
frames and below the boiler, whilst the coal was carried in a bunker 
affixed to an extension of the foot-plate. Somewhat similar " single 
tank. " engines were made by the same firm for the Manchester and 
Birmingham Railway (London and North Western Railway). The 
two engines in question were Nos. 33 and 34, and were used in 



138 EVOLUTION OF THE STEAM LOCOMOTIVE 

vorking the traffic between Manchester and Macclesfield, the daily 
duty of each averaging 114J miles. These engines commenced work 
ir* May, 1847. They weighed 21 tons in working order; the driving 
wheels were 5ft. Gin. diameter, and the leading and trailing 3ft. Gin. 
Two water-tanks were provided, one between the leading and driving 
wheels, the other under the coal bunker, at the rear of the trailing 
wheels. The two tanks contained 480 gallons of water. A woo len 
float attached to a vertical rod was fitted to show the amount of water 
in the tanks ! The bunker contained half a ton of coals. These engines 
were fitted with sand-boxes; but these were placed in front of the 
leading wheels only, although the locomotives were specially constructed 
for running either bunker or chimney in front. However, the intro- 
duction of the sand-box was a step in the right direction ; yet Tredgold 
only mentions the innovation in an apologetic manner. He says 
(after describing the working of the apparatus) that " it is very seldom 
required on the Macclesfield line, owing to the ballast between the rails 
being mostly sand; but when the rails are moist it is necessary in 
starting a heavy train to open the sand-cock." Tredgold then pro- 
ceeds to give a detailed explanation of " how it is done." 

In September, 1849, Walter Neilson,, of Glasgow, obtained a patent 
for his design of tank engine. 

The tank was of the now well-known " saddle " kind, and covered 
the whole boiler, barrel, and smoke-box; the bottom of the saddle 
tank rested on the frames on either side of the boiler, so that the 
tank was semi-circular in shape, instead of being but an arc, as is the 
practice with modern " saddle tanks." Neilson was, however, suffici* 
ently ingenious not to limit the design of his saddle tank, for we find 
that "the tank may be supported from the boiler, instead of the 
framing, if necessary, and its length may be made shorter than that 
of the boiler, if required." The boiler was fed with water drawn 
from the smoke-box end of the tank, to obtain the advantage of the 
escaping heat. The coal bunkers were placed at the sides of the fire-box, 
and extended some distance towards the back buffer beam, but a 
bunker was not provided at the end, so as to allow " of ready access 
to the couplings of the wagons behind." The engine in question had 
inside frames, underhung springs, outside cylinders, single driving 
wheels, unprovided with flanges, and small leading and trailing wheels. 
A short cylindrical dome was placed over the fire-box, and on this wer<j 
fixed two "Salter" pattern safety valves, covered by a brass casing. 



EVOLUTION OF THE STEAM LOCOMOTIVE 



139 



" Light locomotives " was the popular name of tank engines when 
the general use of such engines was being urged as a method of reduo 
ing the working expenses of unremunerative railways. We have 
previously alluded to Mr. W. Bridges Adams and his combined engines 
and carriages. This gentleman and Mr. England were the principal 
advocates of the "light" locomotive, and both attained some success 
in connection therewith. 

The engines in question would now be considered absurdly light, 
but nearly fifty years ago far different ideas of "light" and "heavy,", 
as applied to locomotive engines, obtained. 

The practice of Adams and England regarding " light " locomotives 
differed considerably. The former was a firm advocate of four wheels 




FIG. 53. ADAMS'S "LIGHT" LOCOMOTIVE FOR THE LONDONDERRY AND 
ENMSKILLEN EAILWAY 

and a long wheel base. England, on the other hand, preferred his 
light locomotives to be supported by six wheels. In 1847, Adams built 
a light locomotive (Fig. 53) for the Londonderry and Enniskillen Rail- 
way (Ireland), with outside cylinders 9in. in diameter, the stroke being 
15in. The driving wheels were 5ft. in diameter, and located in front of 
the fire-box; the other pair of wheels were 3ft. diameter, and were 
placed beneath the smokenbox. The fire-box was 2ft. 9in. long, the 
boiler 2ft. 3in. diameter and 10ft. Sin. long; height of top of boiler 
from rails, 5ft. Sin. The connecting-rods were 5ft. 3in. long ; the steam 
pressure was 1201b. The water-tank was placed beneath the boiler, 
and reached to within a few inches of the surface of the rails. Mr. 



140 EVOLUTION OF THE STEAM LOCOMOTIVE 

Adams built a similar engine for the St. Helen's Railway. In Novem- 
ber, 1849, a broad-gauge light locomotive was built at Mr. Adams's 
Fairfield Works, for service on the Holyhead breakwater. The engine 
in question was from designs prepared by Mr. Thos. Gray, resident 
engineer of C. and J. Rigby, the contractors for the breakwater. This 
engine had cylinders Sin. diameter, the stroke being 18in. 

In July, 1849, Adams supplied two of his light engines to the 
Cork and Bandon Railway. These differed from those already 
described, as the driving wheels were the leading ones, the smaller 
pair of wheels being at the rear. The Irish names of the engines 
signified " Ilunning Fire " and " Whirlwind." 

In August, 1853, the engineer of the Cork and Bandon Railway 
reported that " the cost of repairs to the engines was very small, more 
particularly on the light engines, which have worked all the fast 
passenger trains in a satisfactory manner, and with the same con- 
sumption of coke as heretofore viz., about lOlb. per mile. These 
engines were put upon the line in July, 1849, since which period they 
have been daily working the passenger traffic- The principal item 
of cost in their repairs during the four years has been a new crank 
axle to each of the two light engines, as also a new set of tyres on chc 
driving wheels. The light special trains conveyed by these engines 
generally occupy about 26 minutes between the two termini of Cork and 
Bandon." These two light locomotives continued to work traffic over 
the Cork and Bandon Railway for several years. 

On May 1st, 1851, Mr. Peto, the chairman of the Norfolk Railway, 
provided four, light engines with 12in. cylinders, and weighing 10 tons 
each, to work the branch traffic of that railway under the following 
circumstances. 

The Norfolk Railway was worked by the Eastern Counties, and 
the branch or local trains of the former were supposed to meet the 
main line trains of the latter line at the junctions. 

But the Eastern Counties trains had a habit of being behind- 
hand, putting in an appearance at the junctions any time between 
thirty minutes and an hour after the times given in the time-tables. 
As a result, the traffic on the Norfolk branch lines was thoroughly dis- 
organised ; indeed, so little could it be depended upon that local 
passengers almost completely neglected the line. Then the Eastern 
Counties Railway worked the Norfolk branches with the main line 
engines, and charged the Norfolk Railway the average expense per 
mile incurred in working with these engines. 



EVOLUTION OF THE STEAM LOCOMOTIVE 141 

Such a method did not meet with the approval of the chairman 
<>i the Norfolk Railway, so Mr. Peto obtained the sanction of the 
K;isi"rn Counties Railway to allow the Norfolk Company to work the 
local branch traffic itself, and independent of the arrival and departure 
oi the main line trains. Mr. Peto's new system met with instantaneous 
and complete success, a great saving being effected. Thus the coke 
consumption of Adams's light engines, introduced by Mr. Peto, only 
averaged 1011). per train mile; but the Norfolk Railway had been 
paying the Eastern Counties Railway at the rate of 271b. per mile, 
that being the average coke consumption of the Eastern Counties 
Railway main line engines. A large and remunerative local passenger 
traffic was built up by reason of the improved local services. 

The advantages claimed by Mr. Adams for his light engines were as 
follows : Less dead weight, less friction, and less crushing and 
deflecting of the rails. 

We will now proceed to give some account of England's 
light locomotives, popularly called " Little Englanders " ; but 
this cognomen then had a very different meaning, as applied to loo 
motives, than the words have at the present time in their application 
to certain individuals. England constructed his premier light engine 
in 1849, and the "Little England" (Fig. 54) was exhibited at the Exhi- 
bition of 1851. The chief dimensions were : Driving wheels, 4ft. Gin. 
diameter, located in front of the fire-box ; leading and trailing wheels, 
3ft. diameter; inside cylinders, placed between the leading and driv- 
ing wheels, and not under the smoke-box ; the frames were outside. 
The fire-box was of the Bury type, with safety valves, similar to those 
previously described as on the Bury engine still at work on the Furness 
Railway. A dome was placed en the boiler barrel ever the cylinders, 
so that the steam-pipes proceeded in a curved vertical line from the 
dome to the cylinders. The dome was on a square seating. An 
auxiliary pipe for the escape of the steam was provided at the back 
of the chimney, but was only about one-half as high as the chimney. 
At the rear of the foot-plate was a well-tank, holding water sufficient 
for a 50-mile trip. A prize medal was awarded to this engine at Iho 
Exhibition. 

England and Co. in August, 1850, sent one of their light engines 
to the Edinburgh and Glasgow Railway on the following conditions : 
A guarantee that the engine should work the express trains between 
Edinburgh and Glasgow, consisting of seven carriages, and keep good 



142 EVOLUTION OF THE STEAM LOCOMOTIVE 

time as per time-bill, while the fuel consumption was not to exceed 
lOlb. of coke per mile. If the light engine performed these condi- 
tions to the satisfaction of the railway company's engineer, the Edin- 
burgh and Glasgow Railway was to purchase the locomotive for 1,200. 
But if the work done and the quantity of fuel consumed were not 
as guaranteed, England and Co. were to remove the engine and pay 
all expenses of the trial. 

This " Little England " was tried in competition with the " Sirius," 
the coke consumption of the former being 81b. 3oz. per mile against 
291b. loz. of the "Sirius," both performing exactly the same work. 
The "Little England" so frequently ran in before her time that the 
driver had to be ordered to take longer time on the trips for fear of 
an accident happening in consequence of the train arriving before it 
was expected. The speed of this light engine frequently exceeded 
60 miles an hour, and during the heavy winds and gales of January, 
1851, the "Little England" was the only locomotive on the line that 




FIG. 54. ENGLAND & CO.'S "LITTLE ENGLAND'- LOCOMOTIVE, 
EXHIBITED AT THE PEEMIEB INTERNATIONAL EXHIBITION, LONDON. 1851 

kept time. With a train of five carriages the coke consumption only 
amounted to 6Jlb. per mile. On the Campsie Junction line, the 
"Little England" hauled a train of seven carriages and a brake-van, 
all of which were overloaded with passengers, over the several gradient* 
of Nebrand, at 30 miles an hour. Although the train stopped at a 
station on the incline, the light engine successfully started from the 
station and continued the ascent. An ordinary engine was sent to 



EVOLUTION OF THE STEAM LOCOMOTIVE 



143 



assist the train at the rear, in case the " Little England " proved un- 
equal to the task, but it is said that the bank engine was unable to 
keep up with the train! 

The following table shows the result of the trial of the "Little 
England" on the Edinburgh and Glasgow Railway: 



1 


i- 


00 






"E 


S' 


** 


s 


Coke Consumed. 




f 








?H 




g ^3 


|| 


ca 




m 


1 


i s 


o 
B 
P 


"While running. 
Per Mile. 


Including lighting up and 
standing 4 hours be- 
tween each trip, ft, [ _.j 










Ibs. oz. 


Ibs. oz. 


7 








8 3 


9 *v, 7 


7 
5 
4 


) 95 
f miles. 


6 


90 j 
min. 1 


7 4 
6 5 


9 7 
8 5 



On September 7th, 1850, another "Little Englander" commenced 
service on the Liverpool and Stockport Railway, under guarantee to 
haul a train of seven carriages up an incline of 1 in 100, stopping and 
starting upon it, at a speed of 25 miles an hour, and consuming not 
more than lOlb. of coke per mile; on the level the speed was to fce 
45 miles an hour. This engine frequently drew ten carriages under 
the conditions laid down for only sev.en. In June, 1849, a "Littlo 
Englander" had been supplied to the Dundee and Perth Railway for 
working the mail train 'of four carriages. This the engine did suc- 
cessfully for a considerable time. 

After the abolition of rope traction on the Blackwall Railway 
"Little Englanders" were used for the passenger trains. 

England and Co. guaranteed these light engines to haul trains of 
six carriages at a speed of 40 miles an hour on gradients of 1 in 100, 
at a coke consumption of only lOlb. per mile. These engines cost 
1,200 each, and the builders were willing to back them for 1,000 
guineas a-side, with a load in proportion to the weight of any other 
engine, or the amount of fuel consumed. We do not think anyone 
ever cared to accept this challenge. 

In March, 1848, a patent was granted to McConochie and Claucle, 
of Liverpool, for various improvements in the locomotive. The 
cylinders were inside, behind the leading wheels, the valve gearing 



144 F, VOLUTION OF THE STEAM LOCOMOTIVE 

being outside the frame and worked by eccentrics on the naves of the 
driving wheels. It will be remembered that the valve gearing of 
Stephenson's "No. 185" was on this plan. The pumps were worked 
off the driving wheels, as in the " Jenny Linds." A double-beat 
safety valve was provided. 

To enable a low-pitched boiler to be employed, the axle was 
cranked at the extreme ends, so that at each extremity of the axle 
only one return crank-arm was provided, the wheel itself forming the 
second one, and a pin connecting the wheel and axle-crank formed 
the shaft upon which the connecting-rod worked. 

To increase the weight upon the driving axle, a toggle joint was 
placed between the bearing of the trailing axle and the springs ; a rod 
connected the knuckle of the toggle joint with the piston of a small 
steam cylinder. 

When the driver wished to obtain additional adhesion for the driv- 
ing wheels, he admitted steam to this auxiliary cylinder, which drove 
the toggle joint into an upright position, thereby removing the weight 
from the trailing wheels and placing it upon the driving wheels. 
Several other novel proposals were included in the specification in 
question. 

In 1848, Hawthorne, of Newcastle, built an engine named " Plews," 
No. 180, of the York, Newcastle, and Berwick Railway (makers' 
number of engine, 711). The locomotive had a copper fire-box. The 
boiler was 10ft. Sin. long, of oval shape, and consequently had to be 
stayed with four plates ; 229 brass tubes of Ifin. external diameter were 
provided ; two lever safety valves were fixed on a raised fire-box and 
enclosed in a brass casing; the steam pressure was 1201b. A very 
large cast-iron dome placed on the centre ring of the boiler was a 
characteristic of the "Plews." 

The cylinders were placed between the outside and inside frames, 
diameter 16in. and stroke 20in. ; whilst the slide valves were outside 
the cylinders, being worked by four eccentrics, on the outside of the 
wheels, but within the outside frames. The driving wheels were 7ft. 
diameter, the leading and trailing being 4ft. diameter ; the whole of 
the bearings were outside. 

When at rest, the steam was turned into the tender for the pur- 
pose of heating the feed-water. The tender was carried on six wheels 
of 3ft. Gin. diameter, and was capable of holding 1,400 gallons of 
water. 



EVOLUTION OF THE STEAM LOCOMOTIVE 145 

Brake blocks were provided for both sides of the six wheels, and 
an ingenious arrangement of tooth wheels and rack applied the whole of 
the blocks by means of a few turns of the brake handle. 

Crampton's engine, " Liverpool " (Fig. 55), has been described as the 
" ultimatum for the narrow-gauge." Why, we are at a loss to under- 
stand ; many other narrow-gauge engines have been constructed of 
greater power, and certainly of more compact and pleasing design. 
The " ultimatum of locomotive ugliness " would have been a, correct 
title for thQ_" Liverpool." 

The engine in question was built by Bury, Curtis, and Kennedy, for 
the London and North Western Railway, in 1848. The one good point 
about the engine was the immense heating surface, which amounted 
to 2,290 sq. ft. When our locomotive superintendents make up their 
minds to construct express locomotives with such an amount of heating 
surface, we shall hear no more of " double engine running," and our 




FIG. 55. CRAMPTON'S "LIVERPOOL," L. AND N.W.R. . 

express trains may be expected to average a speed of over 50 miles an 
hour from start to finish (including stops) on all trips. 

The general arrangement of the "Liverpool" was similar to the 
engines on Crampton's system already described viz., the driving 
wheels at the back of the fire-box and outside cylinders fixed about 
the centre of the frames. This engine had three pairs of carrying 
wheels under the boiler, in addition to the driving wheels. The 
cylinders were outside, fixed upon transverse bearers, formed of iron 
plates IJin. thick, curved to the shape of the boiler and passing below 
it. The cylinders were 18in. diameter, the stroke being 24in. Metallic 
packing, consisting of two concentric rings of cast-iron, each with a 
wedge and circular steel spring, was used for the purpose of making the 
pistons steam-tight. The valves were above the cylinders, and were 

L 



146 EVOLUTION OF THE STEAM LOCOMOTIVE 

inclined, the eccentrics being of large size and outside the driving 
wheels. The regulator was located in a steam-box on the top of the 
boiler barrel ; the steam reached the valves by means of curved vertical 
copper pipes outside the boiler, whilst the exhaust was conveyed to the 
smoke-box by similar horizontal "outside" pipes. The two exhaust 
pipes united within the smoke-box beneath the bottom of the chimney, 
the blast orifice being 5jin. diameter. 

The leading wheels were 4ft. Sin. diameter, the two intermediate 
pairs 4ft., and the driving wheels 8ft. in diameter. The area of the 
fire-grate was 21.58ft. The tubes were of brass, 12ft. Gin. long; 
292 were 2 3- 16th in diameter, the remaining eight being If in. dia- 
meter. The heating surface was: Tubes, 2,136.117 sq. ft.; fire- 
box, 154.434 sq. ft. The pumps were horizontal, fixed on the frames 
over the leading wheels; they were worked by extension piston-rods, 
worked through the covers of the cylinders. 

The engine weighed (loaded) 35 tons, of which weight 12 tons 
were OQ the driving axle. The tender weighed 21 tons. With a light 
load the "Liverpool" attained a speed of nearly 80 miles an hour, 
whilst on one occasion she hauled the train* conveying Franconi's 
troupe and horses, consisting of 40 vehicles, from Rugby to Eustan 
under the schedule time. Three engines had been engaged to haul 
the same train from Liverpool to Rugby, when time was lost. The 
power of the "Liverpool" would, therefore, appear to have exceeded 
that of three of the usual London and North Western Railway loco- 
motives oi that date. 

Adams's idea of a straight driving" shaft connected by means of 
outside rods with the driving wheels soon attracted attention, and in 
1849 Crampton incorporated the principle in his patent locomotive 
specification of that year. But it was some two years later before 
any engines were built under this particular patent of Crampton's. 
These locomotives will be described in due sequence. 

We will now give a few details of some engines that would have 
been most interesting had we knowledge that they were ever built. 
We possess drawings of the engines in question, but lack authentic 
details of their performances, so we will mention the principal features 
of the designs, as given in the patent specifications. George Taylor, 
of Holbeck, Leeds, obtained his patent on June 3rd, 1847. The 
drawing shows the boiler to be hung below the wheels, of which there 
are only four; these were to be 15ft. diameter, and in addition the 



EVOLUTION OF THE STEAM LOCOMOTIVE 147 

wheels were geared up 2 to 1, so that one revolution of the cogged 
driving wheel would have propelled the engine six times the distance 
of a driving wheel of 5ft. diameter. The cylinders were inside tb? 
frames, over the boiler, and, of course, at the rear of the smoke-box ; 
the connecting-rods were attached to cranks on either side of a central 
cog-wheel, which engaged with a cog-wheel of half its diameter, fixed 
on the centre of the rear axle. The motion being conveyed to the 
centre of the axle, instead of alternately on each side, as is usual, 
practically abolished the oscillating motion so apparent in two-cylinder 
engines. An examination of the drawing of this locomotive design 
of George Taylor shows with what ease and slight alteration it was 
possible for the two geared engines supplied to the 'Great Western 
Railway by the Haigh Foundry to have been altered to ordinary direct 
action engines. 

Large wheels were also to be used for the tender, the axles passing 
through the water-tank, so that the centre of gravity was lowered. 

James Pearson, the locomotive superintendent of the Bristol and 
Exeter Railway, obtained a patent on October 7th, 1847, for a double 
locomotive. Fairlie's " Little Wonder " narrow-gauge engines were 
probably suggested by Pearson's design of 1847; whilst the latter's 
famous broad-gauge double-bogie tanks were decidedly evolved from 
his earlier form of locomotive. 

The boiler was to have the fire-box in the centre, the latter being 
divided into two parts, connected below the furnace doors ; the driving 
ax'e wars across this central foot-plate, to allow of very large wheels and 
a low centre of gravity. Each boiler (there being practically two, one 
each side of the central double fire-box) was carried on a four-wheel 
bogie, so that the locomotive was carried on ten wheels, as in the later 
design. The bogie frames were connected by tension-rods, passing 
outside the fire-box. India-rubber springs were employed, their use 
being to allow each bogie to adjust itself to any inequality of the 
road, and to bring the bogies back to the straight position on an 
even road. The coke was to be stowed in bunkers over the boilers, 
and the water could be either in tanks between the tops of the boilers 
and the coke bunkers, or a separate tender could be provided. The 
steam domes were on the fire-box, and were to be of abnormal height, 
and. connected over the head of the foot-plate, thus forming the roof of 
the cab. An exhaust fan was fixed in the smoke-box to draw ilie 
heated air through the tubes and discharge it up the chimney, or it 



148 EVOLUTION OF THE STEAM LOCOMOTIVE 

could be used again as a hot blast for the furnace, and a chimney and a 
smoke-box were provided for each boiler. The fans were to be 
driven by pulleys off one of the axles, and it was claimed that,' as 
the exhaust steam was not required for the purpose of creating a blast, 
extra large exhaust pipes could be used, and the cylinders thereby 
relieved of "back pressure." The cylinders were outside, and rhe 
valves were beyond the cylinders. These were fixed between the 
wheels of one of the bogies.- The general design of this engine, as 
shown in the drawings, was very ingenious, and is certainly the most 
symmetrical "double-ended" type of engine we have seen illustrated. 
Pearson for some reason did not construct an 'engine after this stylo, 
but produced the well-known 9ft. "single" (double-bogie) tanks 
instead. 

The third patent now to be described had also for its leading fea- 
ture extra large driving wheels. The specification is that of Charles 
Ritchie, of Aberdeen, the patent being granted to him on March 2nd, 
1848. The principal feature was the providing of two piston-rods to 
each piston, one on each side. Four driving wheels were proposed, 
one pair placed in front of the- smoke-box and one pair behind the 
fire-box. The cylinders were outside, and were, of course, fixed at an 
equal distance between the two pairs of driving wheels. One pair of 
carrying wheels was to be used, placed below the cylinders. It was 
claimed that this arrangement of pistons and connecting-rods exactly 
balanced the reciprocating parts of the machinery, and therefore 
abolished oscillation. Another improvement related to the slide- 
valves, the starting, stopping, and reversing of the engine, together 
with the expansive working of the steam, the whole to be controlled 
by a wheel on the foot-plate, connected by cogs with the link of the 
valve-gear. 

Other improvements were compensating safety-valves, an "nnti- 
primer," and an improved feed-water apparatus. The last is described 
as follows : " Upon steam being admitted from the boiler into the 
cylinder, through the steam-port, the piston will be acted upon, and 
the ram be withdrawn ; the water will then raise the valve and enter 
the barrel, to occupy the space previously occupied by the ram. By 
this time the piston will have acted upon a lever, so as to cause the 
slide-valve to uncover one port and cover the other, thereby allowing 
the steam on the other side of the piston to escape through the 
exhaust pipe. 






EVOLUTION OF THE STEAM LOCOMOTIVE 



149 



' - The piston will now be impelled in a contrary direction, and the 
ram entering the barrel will cause the one valve to be closed and the 
other to be opened by pressure of the water therein, which, as the 
mm advances, will be forced into the boiler." 

Another part of the specification related to an " anti-fluctuator." 
.V partition-plate was to be fixed between the tube-plate and the fire- 
box, and the water was to be let into the boiler at the fire-box end, 
and would only reach that portion of the boiler beyond the fire-box 
by flowing over the top of the partition-plate. By this means, the 
fire-box would always be covered with water. It will be seen that the 
sj vification contained several useful propositions, which, however, do 
not appear to have been put into practice. 




FIG. 56.-TIMOTHY HACKWOBTH'S "SANSPAREIL No. 2" 

We have previously, upon more occasions than one, shown i\i*> 
important position occupied in the evolution of the steam locomotive 
by the engines built or designed by Timothy Hackworth. We now 
have to give an account of his last locomotive, the " Sanspareil No. 2.*' 

A. comparison of the drawings of this engine (copies of which are 
in our possession) with Hackworth's earlier efforts of 20 years before, 
clearly discloses the remarkable strides made in the improvement of 
the locomotive during that period, and also most clearly shows that in 
1849 Hackworth was still in the very van of locomotive construction, 
even as he had been in the days of his " Royal George." 



150 EVOLUTION OF THE STEAM LOCOMOTIVE 

The "Sanspareil No. 2" (Fig. 56) was constructed by Timothy Hack- 
worth at his Soho Engine Works at Shlldon. The patent was obtained 
in the name of his son, the late John Wesley Hackworth. We are 
indebted to the executor of the will of Timothy Hackworth for many 
of the following details concerning the engine now under review. 

The locomotive was of the six-wheel "single" type, with outside 
bearings to the L. and T. wheels, and inside bearings of the driving 
wheels. The cylinders were inside. A cylindrical steam dome was 
placed on the boiler barrel close to' the smoke-box. The fire-box was 
of the raised pattern, and on it was an encased Salter safety-valve. 
Cylindrical sand-boxes were fixed on the frame-plates in front of the 
driving wheels. The principal dimensions of the engine were: 
Driving wheels, 6ft. 6in. diameter; leading and trailing, 4ft. dia- 
meter; cylinders, 15in. diameter, 22in. stroke. Weight in working 
order : L., 8 tons 6 cwt. ; D., 11 tons 4 cwt. ; T., 4 tons 5 cwt. Total, 
23 tons 15 cwt. 

It would be well if we mentioned the principal novelties in con- 
struction viz. : Welded longitudinal seams in boiler-barrel ; the 
boiler was connected to the smoke-box and fire-box by means of welded 
angle-irons, instead of the usual riveted angle-irons; the lagging of 
the boiler was also covered with sheet-iron, as is now general, instead 
of the wood being left to view, as was at that time the usual practice. 

A baffle-plate was fitted at the smoke-box end of tubes, as well as 
at the fire-box, end. 

The pistons and rods were made of wrought-iron in one forging. 

The valves were constructed under Hackworth's patent, and were 
designed to allow a portion of the steam required to perform the 
return stroke to be in the cylinder before the forward stroke was com- 
pleted, and thus to form a steam cushion between the piston and 
cylinder covers. Such working was said to economise 25 to 30 per 
cent, of fuel. 

The engine conveyed 200 tons 45 miles in 95 minutes, consuming 
21 cwt. of coke, and evaporating 1,806 gallons of water. She also 
drew a train of six carriages over the same distance without a stop, in 
63 minutes, with an expenditure of 13 cwt. of coke and 1,155 gallons 
of water. 

Upon the' completion of this engine, J. W. Hackworth sent the 
following challenge to Robert Stephenson : 

"Sir, It is now about 20 years since the competition for *he 



EVOLUTION 0*' THE STEAM LOCOMOTIVE 131 

premium of locomotive superiority was played off at Rainhill, on the 
Liverpool and Manchester Railway. Your father and mine were the 
principal competitors. Since that period you have generally been 
looked to by the public as standing first in the construction of loco- 
motive engines. Understanding that you are now running on the 
York, Newcastle and Berwick Railway a locomotive engine 
which is said to be the best production that ever issued from. Forth 
Street Works, I come forward to tell you publicly that I am prepared 
to contest with you, and prove to whom the superiority in the con- 
struction of locomotive engines now belongs. 

"At the present crisis, when any reduction in the expense of work- 
ing the locomotive engine would justly be hailed as a boon to railway 
companies, this experiment will no doubt be regarded with deep interest 
as tending to their mutual advantage. I fully believe that the 
York, Newcastle and Berwick Railway Company will willingly afford 
every facility towards the carrying out of this experiment. 

" Relying upon your honour as a gentleman, I hold this open for a 
fortnight after the date of publication. 

"I am, Sir, yours, etc., 

JOHN W. HACKWORTH." 

We do not think Robert Stephenson accepted the challenge ; at all 
events, no records of such a competition have ever been made public, 
and had it taken place the victor would have doubtless well published 
the result. 

The " Sanspareil " frequently attained a speed of 75 miles an hour 
on favourable portions of the line. She was sold to the North Eastern 
Railway by the executors shortly after the death of Timothy Hack- 
worth, something like 3,000 being obtained for the engine, which 
continued to work upon the North Eastern Railway until recent years, 
having, of course, been rebuilt during the long time it was in active 
service. 

We have now to describe another specimen of the locomotives con- 
structed by the celebrated firm of Bury, Curtis and Kennedy. This 
locomotive was one of the last engines built by the firm before it* 
final dissolution. The "Wrekin" was a six-wheel engine with inside 
bar frames and inside cylinders, and was constructed for the Birming- 
ham and Shrewsbury Railway in 1849. 

The special points noticeable in the construction of the engine in 
question are the width of the framing, which was arranged horizontally 



152 EVOLUTION OF THE STEAM LOCOMOTIVE 

instead of vertically, and only two bearings to each axle. The axle- 
boxes of the leading wheels were bolted to the frames, those of tho 
other wheels being 1 welded to the frames, and the cylinders were also 
directly affixed to the framing. An advantage claimed by the builders, 
as resulting from the method of construction employed, was time the 
weight being placed entirely within the wheels, such weight had a 
tendency to press down the axle between the bearings, and so counter- 
act the constant tendency arising from the flanges of the wheels^ when 
pressing against the edge of the rails, especially in passing round 
curves. 

The- cylinders were 15in. diameter, the stroke being 20in The 
driving wheels were 5ft. Tin. diameter, the leading 4ft. lin. and the 
trailing 3ft. Tin. 

The boiler contained 1T2 brass tubes, lift. Gin. long and 2 Jin. 
external diameter. The heating surface was: Tubes, 1,059 sq. ft.; 
fire-box, 80 sq. ft,; total, 1,139 sq. ft. Grate area, 15 sq. ft. 

No steam dome was provided, the main steam-pipe being of iron, 
with a longitudinal opening 3-1 6th inches wide along the top; thi.-i 
-pip 3 extended to the smoke-box, at which end of it the regulator valve 
was placed ; the actuating-rod passing through the main steam-pipe 
from end to end. Two encased Salter safety valves were fixed on the 
fire-box. The wheel base of the " Wrekin " was : leading to driving, 
8ft. lin. ; 'driving to trailing, 6ft. llin. 

In 1849 the Vulcan Foundry Company supplied the Caledonian 
Railway with an engine known as " No. 15." In general appearance 
the locomotive was very similar to Allan's " Velocipede " engine on 
the London and North Western Railway. 

" No. 15 " (Fig. 5T) was a six-wheel engine, with inclined outside 
cylinders, 15in. diameter and 20in. stroke. The driving wheels were 
6ft. diameter, leading and trailing wheels 3ft. 6in. diameter. The boiler 
barrel was 9ft. 9in. long and 3ft. 6|in. diameter, containing 158 brass 
tubes of If in. external diameter. Wheel base, L. to D., 6ft. ; 
D. to' T., 6ft. 6Jin. The chimney was 6ft. 6in. high ; on the centre 
of the boiler was a man-hole, surmounted by a column safety-valve of 
Salter's pattern, the blowing-off steam pressure oeing OOlb. The 
steam dome was of brass, placed on the. raised fire-box, and sur- 
mounted with a second Salter's safety-valve. The driving and leading 
wheels were provided with underhung springs, but the trailing wheels 



EVOLUTION OP THE STEAM LOCOMOTIVE 153 

had the springs over the axle-boxes. These latter springs were of 
elliptic shape, and were provided with a screw device fixed on the foot- 
plate, by means of which the weight was taken off the trailing wheels 
and thrown upon the driving wheels. 

In addition to the semi-circular brass name-plates (i.e., Caledonian 
Railway) affixed to the splashers of the driving wheels, brass number- 
plates of diamond shape (12in. long by Gin. diameter) were fixed 
on the buffer beams of "No. 15." The tender was supported on four 
wheels, 3ft. Gin. diameter, and held 800 gallons of water. 

During June, 1849, " No. 15 " made a number of trial trips between 
Glasgow and Carlisle, with seven, eight, and nine coaches of an 
average weight of five tons each, the weight of the engine and tender 




FlO. 57. CALEDONIAN RAILWAY ENGINE, "No. 15'' 

being 28 tons. On the trips to Glasgow the Beattock Summit had, of 
courso to be climbed. This consists of 10 miles of stiff gradients, 
varying between 1 in 75, 80, and 88. The run of 13 J miles from 
Beattock to Elvanfoot, consisting of the 10 miles just described and of 
3J down at 1 in 100, was negotiated by "No. 15" in 33 minutes, 
with a train of six coaches; with seven coaches the time was 41 
minutes, and with a pilot and eleven coaches, 30 minutes, or at* the 
rate of 27 miles an hour. These were considered exceptionally good 
specimens of hill-climbing performances 48 years back, but are, of 
course, entirely out of comparison with modern Caledonian records 
over the same line with much heavier trains. 

McConnell, the locomotive superintendent of Wolverton, turned out 
several remarkable locomotives for the London and North West-am 
Railway, and No. 227, or, as she was generally called, "Mac's Mangle," 
(Fig. 58), was one of these peculiar specimens of McConnell's design. The 
cylinders were of large size, being 18in. diameter, with a 24in. stroke; 



154 



EVOLUTION OF THE STEAM LOCOMOTIVE 



they were outside, as were also the axle bearings a very uncommon 
combination. No. 227 was a six-wheel "single" engine, the driving 
wheels being 6ft. Gin. diameter, and the leading and trailing wheels 
4ft. diameter. The fire-box was of the raised pattern, and a Salter 




safety valve (encased) was fixed on it. A huge steam dome was pro- 
vided, located, originally, close to the smoke-box end of the boiler 
barrel, but afterwards (in 1850) placed near the fire-box end, over the 
driving wheels. The boiler-heating surface of " Mac's Mangle " was 
1,383 sq. ft. JNo. 227 enjoyed but a short locomotive career, being 
built in April, 1849, and "scrapped" in May, 1863. It is stated that 
in consequence of the extreme width of this engine, caused by outside 



EVOLUTIOX OF Til STEAM LOCOMOTIVE 



155 



cylinders being employed in conjunction with outside axle-boxes, it 
became necessary to set back the platforms at some of the stations, so 
that the engine could clear these erections without coming to grief. 




FIG. 59." PRESIDENT," ONE OF McCONNELI/S "BLOOMERS,' 
AS ORIGINALLY BUILT 



L. & N.W.R. 



l:i 1850 McConnell designed a very powerful class of passenger 
engines for the L. and N. W. R. These are generally called the 
" Bloomers." "President" (Fig. 59) illustrates this favourite class 
of L. and N. W. R. locomotive, when built. The cylinders were 
inside, 16in. diameter, with a stroke of 22in. The driving wheels were 
7ft. in diameter. The heating surface was 1,152 sq. ft. These engines 
weighed 28| tons. (Fig. 60) is from a photo of a u Bloomer" as 
rebuilt by Ramsbottom. 




FIG. 60. ONE OF McCONXELI/S "BLOOMERS" AS REBUILT BY RAMSBOTTOM 



CHAPTER X. 

The locomotive exhibits of 1851 The " Hawthorn " Wilson's two-boiler engine, 
the "Duplex" Fairbai-:n's tank engine The S.E.R. "Folkestone" on 
Crarupton's system Sharp's "single" engines for the S.E.R,. J. V. Gooch's 
designs for the Eastern Counties Railway The "Ely," Taff Vale Railway 
Beattie s "Hercules" A much- vaunted locomotive, McConnell's "300" 
L. & N. W.Tv London and Birmingham in two hours The chief features of 
"300" Competitive trials with other engines Coil v. coke An earlier 
"record" bciler Dodd's " Y^a'bel " The first compound locomotive 
Another Beattie design Pisay's compressed air railway engine Its trial 
trips on the Eastern Counties Railway The original Great Northern engines 
Sturrock's masterpiece, "Xo. 215,' G.N.R. Pearson's famous Sft. "single" 
double bogies, Bristol and Exeter Railway Rebuilt with Sft. drivers, and 
n lender added by the G.W.R. More old Furness Railway engines Neilson's 
outside cylinder locomotives -A powerful goods engine on the Maryport and 
Carlisle Railway Go cola's 7ft.-c.apled broad-gauge locomotives His first 
narrow-gauge engines. 

THE premier International Exhibition, which, as all the world well 
knows, was held in Hyde Park, London, 1851, brought together quite 
a respectable collection of railway appliances. The British exhibitors 
showed the following locomotives : 

London and North Western Railway's "Cornwall " and "Liverpool." 

Great Western Railway 'y "Lord of the Isles." 

Hawthorne's express, " Hawthorn." 

Adams's combined engine and carriage, " Ariel's Girdle," built by 

Wilson and Co., Leeds. 

England's light locomotive, built by Fairbairn. 
Fairbairn's tank engine. 
South Eastern Railway's ''Folkestone." 
E. B. Wilson and Co.'s double boiler tank engine. 

Several of these have been described, in an earlier chapter, whilst 
details of other types (such as the li Lord of the Isles " type) have also 
been given, so that it is not necessary to describe such designs again. 
We have, however, to give particulars of Hawthorne's express, Fair- 
-bairn's tank, the "Folkestone," and Wilson's "double boiler" tank 
engine. The dimensions of the first are : cylinders, 16in. diameter, 



EVOLUTION OF THE STEAM LOCOMOTIVE 157 

22in. stroke; driving wheels, 6ft. Gin. ; leading and trailing wheels, 
3ft. 9in. diameter; heating surface of fire-box, including water bridge, 
110 sq. ft.; tubes, 865.4 sq. ft. The tubes were of brass, of 2in. 
external diameter, and 158 in number. 

The " Hawthorn " had inside cylinders and double sandwich frames, 
a raised fire-box, with an enclosed safety-valve, no dome, but a per- 
forated steam-pipe for the collection of the steam was provided. The 
engine was designed for running at 80 miles an hour ; the special 
features of the engine being double compensating beams for distri- 
buting the weight uniformly on all the wheels, equilibrium slide valves, 
*aid an improved expansion link suspended from the slide-valve rods. 
Instead of fitting a spring to each wheel,, two only were placed on each 
side of the engine between the wheels. These springs were inverted, 
and. sustained by central straps attached to the framing. Their ends 
were connected by short links to the wrought-iron dou Die-compen- 
sating beams placed longitudinally on each side of the engine, inside 
and beneath the framing. 

The two inner contiguous ends of these beams were linked by a 
transverse pin to an eye at the bottom of the axle-box of the driving 
axle, whilst the opposite ends of the beams were respectively linked 
in a similar manner to eyes on the top of the leading and trailing axle- 
boxes. The action of these beams was obvious. By them a direct 
and simultaneous connection was given to all the axle-bearings, and 
consequently a uniform pressure was always maintained on all the 
wheels, irrespective of irregularities on the permanent-way. The 
slide valves were placed on vertical faces in a single steam-chest, 
located between the two cylinders. One slide valve had a plate car,t 
on its back, and the other had an open box cast on its back to receive 
a piston, which had it's upper end parallel with the valve face. This 
piston was fitted steam-tight in the box, and its planed top bore against 
the face of the plate in working. By this arrangement the slides were 
relieved from half of the steam pressure ; and to assist a free exhaust, 
a port was made in the back plate of one of the slides, so providing an 
additional exit for the spent steam by means of the piston and tho 
exhaust ports of the opposite valve. 

The expansion link was placed in such a position as to allow the 
bottom of the boiler to be quite near the axle. The link, instead of 
being fixed to the ends of the eccentric-rods, so as to rise and fall with 
them when the reversing lever was moved, was suspended from its 



158 EVOLUTION OF THE STEAM LOCOMOTIVE 

centre, by an eye, from the end of the slide-valve spindle. This 
removed the weight of the link, etc., from off the reversing gear. The 
eccentric-rods were jointed to the opposite ends of the link slide-block, 
to secure steadiness and durability of the parts. It was claimed that 
this method of a fixed link-centre as fitted to the " Hawthorn " ensured 
a more correct action of the valves. 

Wilson and Co., of the Railway Foundry, Leeds, exhibited a curious 
tank engine at the Exhibition of 1851, called the "Duplex," in con- 
sequence of it being provided with two boilers. The idea of the 
designer was to obtain sufficient steam from an engine of light weight 
to haul a heavy train. The original drawings of this engine are still 
in the possession of Mr. David Joy, who designed it ; and at first it waa 
proposed to build the "Duplex" with three cylinders and six-coupled 
wheels, but afterwards fresh drawings were prepared, and it was from 
these latter ones that the engine was built. The two boilers were 
placed side by side, and these each measured 10ft. Gin. long by 
1ft. 9in. diameter, and together contained 136 tubes of If in. diameter, 
the heating surface of which was 694 sq. ft., that of the fire-box 
being 61 sq. ft., making a total of 755 sq. ft. The cylinders were 
outside, their diameter being 12 Jin., and the stroke 18in. The lead- 
ing wheels were 3ft. 6in. diameter ; the driving and trailing (coupled) 
5ft. diameter. Some other dimensions were : Total length, 24ft. 3in. ; 
breadth, 5ft. Sin. ; height from rail to top of chimney, 13ft. 6in. ;, 
weight, empty, only 16 tons, with fuel and water 19 tons 17 cwt. The 
capacity of tank was 520 gallons, sufficient for a journey of 25 miles ; 
coke bunker, 42 cubic feet), equal to 26 bushels, or 15 cwt. 
The " Duplex " was sold to a Dutch railway after the Exhibition, 
and its further career is, therefore, unknown to those interested 
ir. it. 

Fairbairn's tank locomotive was of the "well" type, supported on 
six wheels, the driving pair being 5ft. diameter, and the L. and T. each 
3ft. 6in. diameter. The cylinders were inside, measuring lOin. by 
15in. stroke. The boiler was 8ft. long by 3ft. diameter, and contained 
88 brass tubes of 2in. diameter. The heating surface amounted to 
480 sq. ft. The internal fire-box was of copper, and measured 2ft. 5in. 
long, 3ft. wide, and 3ft. Sin. deep. The tank behind and under the 
foot-plate held 400 gallons of water. The coke consumption of this 
little engine was only lOlb. per mile with trains of six carriages, the 
weight in working order only 13 tons; and it may interest our readers 



EVOLUTION OF THE STEAM LOCOMOTIVE 159 



to know that this diminutive locomotive was described as "a fair 

specimen of the heavier class of tank engine " 

The engine calling 
for the greatest atten- 
tion at the Exhibition 
of 1851 was the 
"Folkestone" (Fig. 61), 
exhibited by the South 
Eastern Railway. This 
was an engine built by 
B. Stephenson and Co., 
under one of Cramp- 
ton's patents, but the 
principal feature in its 
design was an inter- 
mediate driving axle, 
connected by means of 
outside cranks, and 
coupling-rods to the 
driving wheels, which 
were (under Cramp- 
ton's patent) behind 
the fire-box, the axle 
extending across the 
foot-plate. It will be 
well, perhaps, if we at 
this point reiterate the 
fact that the method 
of working locomotives 
by means of an inter- 
mediate crank-shaft 
was not introduced by 
Crampton, it having 
been used some years 
previously by W. B. 
Adams, not to mention 
some of the early 
Stockton and Darling- 
arrangement was em- 




ton Bailway engines, where the same 



ployed, but with vertical cylinders. Readers -rill, therefore, 



160 EVOLUTION OF THE STEAM LOCOMOTIVE 

see it is incorrect to describe locomotives with this system of machinery 
as " Crampton's patent," although it is quite possible for a " Cramp- 
ton patent" locomotive to be provided with an intermediate driving- 
shaft, as was the case with the " Folkestone." 

Eight engines of this type were built by Stephenson and Co. for 
the South Eastern Railway,, and were numbered 136 to 143, the first 
of which was named "Folkestone." These engines were supported by 
six wheels, a group of four being arranged close together at the smoke- 
box end. Their diameter was 3ft. Gin. The driving wheels were 6ft. 
in diameter, the wheel base 16ft. These engines weighed 26J tons 
each, of which only 10 tons were on the driving wheels, the remainder 
01 the weight being supported by the four leading wheels. The 
cylinders were inside, 15in. diameter, and the stroke 22in. The fire- 
box top was flush with the boiler barrel, the straight lines of which 
were unrelieved by a dome, but an encased safety valve was fixed near 
the back of the fire-box top. The boiler contained 184 tubes, of 2in. 
diameter and lift, in length. 

The "Folkestone" ran its trial trip on Monday, March 31st, 1851, 
when Mr. McGregor, the chairman of the South Eastern Railway, Mr. 
R. Stephenson, the builder of the engine, Mr. Barlow, the South 
Eastern engineer, and Mr. C'udworth, the South Eastern locomotive 
superintendent, were present. From London Bridge to Redhill no 
great speed could be attained, as a Brighton train was in front; but 
beyond the latter station, and with a train of nine carriages, the 19 \ 
miles to Tonbridge were covered in 19 J minutes, a maximum speed 
of 75 miles an hour being attained. After a short stop, the journey 
to Ashford was resumed, and that town was reached in 20 J minutes 
after leaving Tonbridge. The times and distances were as follow : 
Redhill to Tonbridge, 19 miles 47 chains, start to stop in 19 J minutes ; 
Tonbridge to Ashford, 26 miles 45 chains, start to stop in 20J 
minutes, or at the rate of 78 miles an hour; the whole 46 miles 12 
chains being covered in 40 minutes, running time, or, including the 
stop at Tonbridge, in 43 minutes. It must be remembered that the 
line between Redhill and Ashford is, perhaps, the most level and 
straight in England for so long a distance. 

These eight engines did not prove very successful in general work- 
ing, and they were afterwards rebuilt as four-coupled engines, an. 
ordinary cranked axle with wheels being provided in place of the 
intermediate driving shaft. 



EVOLUTION OF THE STEAM LOCOMOTIVE 161 

It will not be out of place if we here mention eight "single" 
engines built by Sharp Bros, in 1851 for the South Eastern Railway, 
and numbered 144 to 151. The general dimensions were similar to 
the Cramptons, except that the wheel base was only 15ft., and that the 
heating surface was 1,150 sq. ft. The admission of the steam to the 
cylinders was controlled by a hand lever, with catch and notches, 
similar to and placed by the side of the ordinary reversing lever. Six 
eccentrics were on the driving axle, two of them working the pumps. 
The framing and springs of these engines were afterwards perpetuated 
by Cudworth in his later and better known types of South Eastern 
locomotives. 




FIG. 62. ONE OF J. V. GOOCH'S "SINGLE" TANK ENGINES, 
EASTERN COUNTIES EAILWAY 

The locomotives of the despised "Eastern Counties," that were 
designed about 1850 by Mr. J. V. Gooch, will now be concisely 
described. They were of three kinds viz., " single " tanks, " single " 
express, and four-wheels-coupled tender engines. Of the tanks, three 
sizes were constructed, chiefly at the "Hudson Town" (or Stratford 
Works). The largest of these were provided with outside cylinders, 
14in. diameter and 22in. stroke, the boiler being 10ft. Gin. long, erd 
containing 164 tubes of 1 3-1 6th in. diameter. The leading and trail- 
ing wheels had outside bearings, the driving wheels being provided 
with inside bearings only. A steam dome was placed over the raised 
fire-box, and a screw-lever safety valve on the boiler barrel. The 
water was stored in two tanks, fixed between the frames, one below 
the boiler and the other beneath the foot-plate. These tank-engines 
were known as the " 250 " class, and some of our readers may recollect 

M 



162 EVOLUTION OF THE STEAM LOCOMOTIVE 

that when Peto, Brassey and Belts leased the London, Tilbury and 
Southend Railway, engines of this design were used to work the traffic 
on that railway. We understand it is now 20 years since the last of 
them (No. 08) reached the final bourne of worn-out locomotives 
the "scrap heap." 

The dimensions of the smallest class of these tanks (Fig. 62) were : 
Cylinders, 12in. diameter, 22in. stroke; boiler, 10ft. long and 3ft. 2in. 
diameter, 127 tubes of 1 7-8th in. diameter; the total heating surface 
was 709 sq. ft.; grate area, 9.7 sq. ft. The driving wheels were 
6ft. 6in. diameter, and the L. and T. 3ft. 8in. The total weight of 
these engines was 23 tons 19 cwt., of which 9 tons 14 cwt. was en 
the driving axle. The wheel base was : L. to D., 6ft. Sin. ; D. io 
T., 5ft. 9in. 

J. V. Gooch's four-coupled, or " Butterflies," had leading wheels 
3ft. 8in. diameter, and driving and trailing (coupled) 5ft, 6in. Wheel 
base, L. to D., 6ft. Sin. ; D. to T., 7ft. 9in. The cylinders were 15in. 
diameter, the- stroke being 24in. The boilers of this class, and also of 
the singles, next to be described, were of the same dimensions as 
those of the " 250 " class of tanks. 

The " single " expresses were provided with 6ft. 6in. driving 
wheels, and cylinders 15in. diameter and a 22in. stroke; in this class 
also the leading and trailing wheels were 3ft. Sin. diameter. The 
wheel base wa.s 14ft., the driving wheels being 6ft. 9in. from the 
leading and 7ft. Sin. from the trailing wheels. Ten engines of this 
design were constructed, some at Stratford, and others at the then 
recently opened Canada Works of Brassey and Co. at Birkenhead. 
Their official numbers were from 274 to 283. 

The " Ely " (Fig. 63) represents the type of 6-wheel passenger engine 
in use on the Taff Vale Railway at this period. She was built in 1851 
by Messrs. Kitson and Company, from Taff Vale designs. She had 
13in. cylinders, with 20in. stroke, and four wheels coupled, of 5ft. 3in. 
diameter. She carried a pressure of lOOlbs., she had a four-wheel 
tender, carrying 900 gallons of water, and as the gross weight of the 
tender was about 11 tons in working order, the gross weight of the 
engine and tender would be 33 tons. The "Ely" could not take a 
train of three carriages, weighing only 21 tons, up the Abercynon 
bank of 1 in 40 without the assistance of a " bank " engine. 

In 1851 Mr. Beattie, the locomotive superintendent at Nine Elms, 
built for the London and South Western Railway the four-wheels- 



EVOLUTION OF THE STEAM LOCOMOTIVE 163 

coupled engine, " Hercules," No. 48. The frames of this engine were 
of the " lattice " type, examples of which can be still seen on seme of the 
older Great Northern Railway tanks. 

The diameter of wheels was: L., 3ft. Gin.; D. and T., 5ft. 6in. ; 
tender, 3ft. Gin. ; wheel base, L. to D., 7ft. lin. ; D. to T., 6ft. Gin. ; T. 
to leading tender, 7ft. 3 Jin. ; the tender wheel base being 10ft. 3in. 
equally divided. 




FIG. 63. "ELY," A TAFF VALE EAILWAY ENGINE, BUILT IN 1851 



The weight was distributed as follows: Engine, L. axle, 8 tons 
17 cwt. ; D., 9 tons 17 cwt. ; T., 9 tons 16 cwt,; tender, L., 4 tons 
19 cwt.; M., 5 tons 19 cwt.; T., 7 tons 10 cwt. The cylinders were 
15in. by 22in. ; tractive force on rail, 7,5001b. ; 1,800 gallons of water 
could be carried in the tender. tank. The "Hercules" had a flush top 
boiler, and a raised fire-box surmounted by a large inverted, urn- 
shaped dome. This design of locomtive was a favourite one on the 
London and South Western Railway for many years, but the last 
engine of the kind has now been scrapped. 

Having favoured the London and South "Western Railway, to 
equalise matters, we cannot do better than give a description of a 
locomotive belonging to its cousin-german, the London and North 
Western Railway. The latter was indeed the more famous, being no 
other than McConnelFs notorious " No. 300," (Fig. 64) which, being 
introduced with a vast amount of publicity, became a nine days' 
wonder, then sank into quiescent mediocrity, and after a brief loco- 
motive career, was seen no more a rather different fate, be it ob- 
served, to that of the London and South Western Railway's " Hercules." 

It has been stated that only one drawing of this engine exists. 

M2 



164 EVOLUTION OF THE STEAM LOCOMOTIVE 

This is incorrect; the writer possesses a complete set of drawings 
relating to " No. 300," together with the whole of the specifications 
from which the engine was constructed. To reproduce this specifica- 
tion in detail would give too technical a character* to this narrative, 
and would try the patience of even the most ardent locomotive 
enthusiast. 

The directors of the London and North Western Railway in 1851 
expressed their determination to run their express trains from London 
to Birmingham in two hours, and gave instructions to McConnell, the 
locomotive superintendent at Wolverton, to* design the necessary loco- 
motives. The salient features of the design were : Inside cylinders, 
18in. by 14in. ; six wheels, with inside and outside frames; driving 
wheels, 7ft. Gin. diameter; leading, 4ft. 6in. ; and trailing, 4ft. 
diameter. 

The boiler was lift. 9in. long and 4ft. 3jin. external diameter. 
The tubes were of brass, 303 in number, only 7ft. in length, and l|in. 
outside diameter. The : crank axle bearings were outside, 
7in. deep and lOin. in length, the inside ones being Vin. 
and 4 Jin. respectively. The leading and training axles were 
hollow, the metal being IJin. thick, and the hollow centre 4Jin. dia- 
meter, thus making the total diameter of the straight axles 7Jin. 
The slide valves had an outside lap of 1 Jin. The principal innovations 
were : Coleman's patent india-rubber springs, fitted below the driving 
axle and above the leading and trailing axles, and also to the buffers. 
McConnelFs patent dished wrought-iron pistons, forged in one piece 
with the piston-rod, and encased with continuous undulating flat metal 
packing. The steam-pipe was of flat section, and passed through a 
superheating chest in the smoke-box ; the steam was thus dried during 
its journey from the dome to the cylinders. The great feature of the 
design was the arrangement of the fire-box, with a mid-feather, a com- 
bustion chamber, hollow stays for a free supply of air to the fire-box, 
and the cutting away of the bottom of the fire-box to obtain clearance 
for the cranks and yet retain a low centre of gravity with large driving 
wheels. Assertion to the contrary notwithstanding, it should be 
observed that so much did McConnell insist upon a low centre of 
gravity that he specially mentioned it in his patent specification of 
February 28th, 1852. 

A more particular description of the fire-box, etc., is requisite. It 
extended into the cylindrical portion of the boiler a distance of 4ft. 9in., 



EVOLUTION OF THE STEAM LOCOMOTIVE 



165 



so that the boiler tubes were only 7ft. long. The whole length of 
the fire-box was 10ft. Gin. ; depth at front-plate 6ft. 5in., at door- 
plate 6ft. lOin. ; length on fire-bars 5ft. lOJin., thus leaving 4ft. 7jiu. 
for the portion over the axle and the combustion chamber. At its 
narrow part (directly at the top of the recess above the driving axle) 
the fire-box was only 2ft. 3in. in height ; height at tube-plate 3ft. 
(beyond the cut away portion) ; width at tube-plate 3ft. 9in. It will 
be noticed that Webb's "Greater Britain" class of locomotives is 
designed with the long fire-box and combustion chamber ; but as Mr. 
Webb, unlike McConnell, does not object to the high-pitched boiler, 
the former does not recess the boiler barrel for the purpose of obtain- 
ing a low centre. Webb also divides his tubes' into two sets by having 
the combustion chamber between them. McConnelFs combustion 
chamber was a continuation of the fire-box. We must now describe 
the general appearance of this engine. 




FIG. 64. McCOXNELL'S 



LONDON AND NORTH WESTERN RAILWAY 



The cylinders were inclined upwards from the front, and the 
valve-chests were above them, below the smoke-box. Two Salter safety 
valves were provided, encased within a sheet-brass covering of 
Stirling's Great Northern pattern. The steam pressure was 1501bs. 
The dome 1 was also of brass, with a hemispherical top surmounting the 
cylindrical lower part. The steam regulator was at the mouth of the 
steam-pipe, which was placed at the top of the dome (inside, of course). 

The heating surface was : Tubes, 980 sq. ft. ; fire-box, 260 sq. ft. 
Wheel base, 16ft. lOin. Sufficient steam could be raised in 45 minutes 



166 EVOLUTION OF THE STE A M LOCOMOTIVE 

after lighting the fire to move the engine. Two of these engines were- 
built about the same time one (No. 300) by Fairbairn and Co., Man- 
chester, the other by E. B. Wilson and Co., Leeds. The orders were 
given early in July, 1852, and the engines delivered the second week 
in November, Wilson and Co. having occupied but eight weeks in the 
construction of the one given to them. 

Both engines were delivered at Wolverton on the same day, and on 
Thursday, November llth, 1852, Wilson's engine was tried for the 
first time, when on her first journey to Euston she attained a speed of 
60 miles an hour. 

It was soon found that "No. 300" and her sister engine were 
unablo to cover the 111 miles Euston to Birmingham in two hours, 
as was confidently .predicted, and the failure to do so was perhaps: 
justly attributed to the inferior condition of the permanent-way. 
On March 8th, 1853, "No. 300" hauled a train of o4 
carriages, weighing 170 tons, from Birmingham to London in three 
hours eight minutes, including five stoppages. A similar train drawn 
by the " Heron " and " Prince of Wales " took ten minutes longer to 
perform the same journey. These two engines had cylinders loin, by 
20in., and 6ft. driving wheels. The results of this trial are thus 
tabulated : 





Cuke. 


Cote 
per mile. 


Average 
s peed 
per hour. 


Maximum 
speed 
per hour. 


No. 300 
" Heron " & "Prince of Wales" 


4,529 lb. 
4,851 lb. 


40-8 lb. 
43-7 lb. 


36-4 miles 
34-5 miles 


.14 
48 



Upon the result of this run it was claimed that McConnell's patent 
engines were considerably superior to two of the ordinary London arid 
North Western Railway locomotives, and one of Stephenson's "long 
boiler" abortions was altered by McConnell, being fitted up with his 
patent combustion chamber, short tubes, and the other innovations, 
as mentioned in our description of " No. 300." 

The "long boiler" originally had 1,013 sq. ft, of tube-heating sur- 
face; when altered, the length of the tubes was reduced to 41ft., and 
some additional ones were fixed diagonally across the combustion 
chamber. By this alteration the tube-heating surface was reduced to 
547 sq. ft., and the engine is stated to have drawn 170 tons at 60 
miles an hour, and to have attained a speed of 70 miles an hour wich 



EVOLUTION OF THE STEAM LOCOMOTIVE 



161 



light trains. From the working of this locomotive the following table 
(by which a reduction of 23 per cent, in the amount of fuel consumed 
was claimed for the altered engine) was prepared: 





Miles 
run. 


Average 
load. 


Coke 
consumed. 


Coke 
per mile. 


Coke 
per ton 
per mile. 


Original 


29,442 


115 tons 


1,715,952 Ib. 


58-28 Ib. 


5"4 Ib. 


Altered 


12,060 


144 tons 


519,120 Ib. 


43-04 Ib. 


2981b. 



But D. K. Clark's paper on " Locomotive Boilers," read before rhe 
Institution of Civil Engineers, soon placed a very different complexion 
upon the result of the trials between the ordinary and patent engines, 
resulting in the " air tubes " to the combustion chamber being speedily 
abandoned. The attention of the directors of the London and North 
Western Railway was called to the failure of these engines, with fhe 
result that they ordered Messrs. Marshall and Wood to report on the 
two classes of engines viz., the ordinary London and North Western 
type and McConnelPs patent locomotives. This report was ready hi 
August, 1853, but for some reason its publication was suppressed at 
the time, but the directors countermanded the construction of other 
engines already ordered on McConnelPs patent principle. 

In the summer of 1854 Marshall and Wood conducted another set 
of experiments for the directors of the London and North Western 
Railway, with the object of determining the relative value of coke and 
coal as fuel for the locomotives. 

The engines chosen were McConnell's patent " No. 303 " and the 
" Bloomer/' No. 293. Double trips were run between Rugby and 
London daily for six consecutive days, coal being burnt on three days 
and coke on the three alternate days. The trains chosen were the 
12.55 p.m. up and 5.4.5 p.m. down. 

It was found that lib. of coal evaporated 5.831b. of water, and lib. 
of coke 8.651b. of water; but the monetary saving was 6s. 9d. per ton 
in favour of coal. 

McConnell's patent engines were again condemned. Marshall and 
Wood's report concluded as follows : " Although we consider the experi- 
ments we made with No. 303 engine satisfactory in point of smoke 
burning, we cannot resist the belief that the consumption of coal is in 
excess of what it ought to be, and that there is room for considerable 



168 EVOLUTION OF THE STEAM LOCOMOTIVE 

improvement in this respect, by means which shall tend to utilise 
the heat which is at present wasted." 

The whole report is of great interest to the technical reader ; it is, 
however, too long to reproduce in extenso, 

It is abundantly evident that there is no great pecuniary gain from 
locomotive designing, or we should be treated to great law-suits 
regarding the validity of the patents, such as have recently been the 
case with pneumatic tyres and incandescent gas-burners. We have 
already, upon more occasions than one, pointed out that certain 
patented locomotive designs had previously been anticipated, although 
the later patentees were probably unaware of the fact. We find this 
to have been the case with McConnell's " recessed " boiler locomotives 
just described, for on December 2nd, 1846, W. Stubbs and J. J. Grylls, 
of Llanelly, enrolled a design of locomotive. The specification in 
question not only mentioned the recessing of the boiler for the purpose 
of allowing the use of a large driving wheel and yet retaining a low 
centre of gravity, but it even anticipated McConnell's combustion 
chamber between the fire-box and tubes. An adaptation of Bodmer's 
double piston motion was also specified by Stubbs and Grylls. The 
two cylinders were placed below the boiler, four wheels being con- 
nected by means of side-rods with the cross-heads of the two cylinders 
in such, a manner that from each cylinder two wheels were driven, by 
means of a cross head, and each cross-head, by means of two con- 
necting-rods, rotating the wheels. Another claim under this patent 
related to driving a locomotive by eccentrics fitted with antifriction 
rollers as a substitute for the ordinary cranks. 

Although in the "Evolution of the Steam Locomotive" it is only 
intended to describe locomotives for British railways, it may not be out 
of place to mention an engine for a foreign railway, for two reasons 
first, because it was built by an English firm in England, and, secondly, 
because it was tried on an English railway before exportation. The 
"Ysabel" was constructed in 1853 by Dodds and Sons, of Kotherham, 
for the " Railway of Isabella II. from Santander to Abar del Rey," and 
was tried on the Lickey incline of 1 in 37 for two miles, under the 
direction of Mr. Stalvies, the locomotive superintendent at Broomsgrove. 
The "Ysabel" had four-coupled wheels 4ft, Gin. diameter; cylinders, 
14Jin. by 20in. stroke; 137 tubes, IJin. diameter, and lift. Sin. in 
length, and was fitted with Dodds' patent wedge expansive motion, 
v/hich required only two eccentrics. For the purpose of easy trans- 



EVOLUTION OF THE STEAM LOCOMOTIVE 169 

portation, the " Ysabel " was so constructed that when disconnected no 
single portion weighed more than six tons ; in addition to the fittings 
necessary to secure the boiler, the only connections between it and 
the frames, machinery, etc., were the steam-pipe and the two feed- 
pump connections. When tried upon the Lickey bank this locomotive 
hauled six trucks weighing 45 tons 12f cwt. up the two miles one 
furlong in 12 minutes 12 seconds, and with a train weighing 29 tons 
4J cwt. the incline was negotiated in seven minutes five seconds. 

The compound locomotive is not quite so modern an invention as 
is popularly supposed, for, putting aside the suggestion emanating in 
1850 from John Nicholson, an Eastern Counties Railway engine-driver, 
whose plan of continuous expansion is generally accepted as the 
foundation of the compound system, we find that in 1853 a Mr. 
Edwards, of Birmingham, patented a " duplex " or in other words a 
compound engine, the steam, after working in a high-pressure cylinder, 
being used over again in a low-pressure one. The cylinders were so 
placed that the dead centre in one occurred when the other piston 
was at its maximum power. 

In 1853 Beattie constructed for the London and South Western 
Railway at Nine Elms Works, the "Duke," No. 123, a six-wheel 
" single" express engine; driving wheels, 6ft. Gin. diameter; 
L. and T. 3ft. Gin. diameter; cylinders, IGin. by 21in. 
stroke. The weight was arranged in an extraordinary 
manner, 10 tons 9 'cwt. being on the leading axle, only 9 tons 9 cwt. 
on the driving axle, and 5 tons 11 cwt. on the trailing axle. Ine wheel 
base was, L. to D., 6ft, 8 Jin. ; D. to T., 7ft. Gin. The "Duke" had a 
raised fire-box, surmounted by a large dome similar to that of the 
''Hercules," whilst another dome was located on the centre of the 
boiler barrel. The shape of this centre dome resembled a soup-tureen 
turned upside down. 

At this point we take the opportunity to briefly describe a railway 
locomotive which, although not propelled by steam, deserves to be 
mentioned as an initial attempt at railway haulage by means of com-, 
pressed air. 

The engine in question was constructed by Arthur Pasey, and was 
tried on the Eastern Counties Railway in July, 1852. This machine 
was, in point of size and power, nothing more than a model, the 
dimensions being: Cylinders, 2?, in. diameter, 9in. stroke; driving 
wheels, 4ft. diameter; weight, IJ-tons; air capacity of reservoirs, 
39 cubic ft. 



170 



EVOLUTION OF THE STEAM LOCOMOTIVE 



By reference to the illustration (Fig. 65) it \rill be 
seen that this curious little locomotive had the six wheels 
of 4ft. diameter within the frames, and the horizontal cylinders out- 
side the frames, and actuating the centre pair of wheels- 
Above the "frames was placed a cylindrical air reservoir, with egg 
shaped ends. This extended from the buffer beam at one end of the 
vehicle to the leading axle, a distance of about 1 2ft. The remainder of 
the space, about 4ft., was occupied by the pressure-reducing and other 
apparatus, and afforded a place of vantage for those in charge of the 
machine. The reservoir was constructed to withstand a pressure cf 
2001b., but the engine was only pressed to 1651b., and this at the time 




FlO. 65 PASEY'S COMPRESSED AIR LOCOMOTIVE, TRIED ON THE 
EASTERN COUNTIES RAILWAY IN 1852 

of the trial at Stratford was reduced to 201b. working pressure. With 
a load of eight people, the engine ran the four miles, Stratford to Lea 
Bridge and back, in 30 minutes. The incident of the trial so aroused 
the curiosity of the men engaged at the Stratford Works, that they all 
left their employment for the purpose of witnessing the trial of so great 
an innovation as Pasey's compressed air locomotive. For this reason 
no further trials could be held at Stratford, but on July 2nd a second 
trip was made at Cambridge, and on this occasion, with six passengers, 
the following results were recorded : Starting from the 60th mile-post 
near the Waterbeach Junction-, with a, working pressure of 151b. per 
sq. in., the first mile was covered in five minutes. By increasing the 
pressure on the pistons, the second mile was' covered in four minutes; 
the pressure was then reduced to 18-851b., and 3J additional miles 
were covered in ten minutes. The designer of this little machine gives 



EVOLUTION OF THE STEAM LOCOMOTIVE 



171 



eight reasons by which he apparently succeeds at all events to his 1 
own satisfaction in proving the great superiority of compressed air 
traction over that of steam. Unfortunately for Mr. Pasey's theory, 
steam is still triumphant, and compressed air dead or nearly so 
for tractive purposes. 

The opening of the Great Northern Railway next claims our atten- 
tion. The first locomotives were supplied by contract, an order for 
50 passenger engines having been given to Sharp Bros, and Co. These- 



i 




FIG. 66. THE FIRST TYPE OF GREAT NORTHERN RAILWAY PASSENGER 
ENGINE, ONE OF THE "LITTLE SHARPS" 

were six-wheel single engines (Fig. 66), the driving wheels being 5ft. 6in. 
diameter. The cylinders were 15in. by 20in. stroke. Weight of engine,, 
loaded, 18 tons 8 cwt. These engines were called "Little Sharps/' 
and (Fig. 66) is an illustration of one of them 

We will now describe the famous " No. 215 " (Fig. 67) of the Great 
Northern Railway, designed towards the end of 1852 by Mr. Archi- 
bald Sturrock, constructed by Hawthorn and Co., Newcastle, and de- 
livered to the Great Northern Railway on August 6th, -1853. 

Fortunately, Mr. Sturrock has supplied the writer with complete 
and authentic details, together with a drawing, of this engine, so that 
readers may rely upon the information being strictly accurate, 
although it should be noted that it does not correspond in several 
particulars with other statements concerning "No. 215" that have 
been published. 

It is a matter of railway history that in 1852 the "Gladstone" 
award settled the great rivalry existing at that period between the 
London and North Western and Great Northern Railways. The 



172 



EVOLUTION OF THE STEAM LOCOMOTIVE 



'competition had been carried on in a manner still in favour in American 
railroad warfare viz., the cutting of rates and fares; but Mr. Glad- 
stone having decided this point, the Great Northern Railway intro- 
duced the method of rivalry now universally recognised as English 

^railway competition that is, trial of speed. Mr. Sturrock, with the 
experience gained under the daring broad-gauge leaders, was, ot 

course, conversant with what a locomotive could do, and his published 
reasons for the construction of "No. 215" are as follow: 

"This engine was constructed to prove to the directors of the 




FIG. ft7. -STURROCK'S MASTERPIECE, THE FAMOUS G.N.R. "215" 

'Great Northern Railway that it was quite practicable to* reach Edin- 
burgh from King's Cross in eight hours, by only stopping at Grantham, 
York, Newcastle, and Berwick. This service was not carried out, 
because there was no demand by travellers for, nor competition 
amongst, the railways to give the public such accommodation." 

Although delayed for 35 years, the demand for such a service 
arose in 1888, and Mr. Sturrock then had the satisfaction of seeing 
runs such as he had built ''No. 215" to perform become daily 
accomplished facts. It should be noted that when "No. 215" was 
originally built, she was fitted with a leading bogie, such an arrange- 



EVOLUTION OF THE STEAM LOCOMOTIVE 173 

ment being a principal feature of Mr. Sturrock's original design for- 
the engine. The bogie and trailing wheels were 4ft. Sin. diametery 
the driving wheels being 7ft. Gin. diameter ; the cylinders were 
inside, and had a diameter of I7in., with a stroke of 24in. The- 
heating surface was large, this being another of the strong points in 
Mr. Sturrock's design. Tubes, 1,564 sq. ft.; fire-box, 155.2 sq. ft.; 
total heating surface, 1,718.2 sq. ft. The weight was, empty, 32 tons: 
11 cwt. 2 qr. ; in working order, 37 tons 9 cwt. 2 qr. Wheel baae r 
21ft. 8Jin. Water capacity of tender, 2,505 gallons. The frames 
and axle bearings were outside; the latter were curved above the- 
driving axle, as in the broad-gauge " Lord of the Isles " type. 

The boiler and raised fire-box were also after the same pattern. 
The engine had no dome, but an encased safety valve on the fire-box 
a further evidence of attention to* the Swindon practice. Com-, 
pensation beams connected the two pairs of bogie wheels, ami the 
underhung springs of the driving wheels were also connected with 
the trailing axle springs by means of compensation levers. " No.. 
215" frequently ran at 75 miles an hour. She appears to have been* 
broken up about 29 years back, for in 1870 Mr. Stirling built an. 
engine, "No. 92," in which he used the 7ft. Gin. driving wheels of 
Mr. Sturrock's famous "215." Engine No. 92, is still at work, so 
that the driving wheels must be 45 years old. A comparison of Mr. 
Sturrock's "215" with McConnelPs "300" will show the immense 
superiority of the former, especially with regard to the amount of. 
heating surface, the pitch of the boiler, and the bogie in place of the 
rigid wheel base. 

In the last chapter, Mr. Pearson's initial patent for a locomotive: 
was described, and a description of his famous double-bogie tank 
engines, with 9ft. "single" driving wheels, is given below. The design. 
(Fig. 68), which was brought out in 1853, was a modification of 
the patent specification already alluded to. The engines were con- 
structed by Rothwell and Co., Union Foundry, Bolton-le-Moors, and. 
were famous for the low average cost for repairs and fuel consump- 
tion per mile run; indeed, a feature of most of the broad-gauge loco- 
motives was the low average cost of maintenance and working. The 
ends of the frames were supported on a four-wheel bogie, 
the wheels of which were 4ft. in diameter, and the driving 
wheels 9ft. diameter; these latter had no flanges. The- 
cylinders (the ends of which projected beyond the front of the- 



174 



EVOLUTION OF THE STEAM LOCOMOTIVE 




EVOLUTION OF THE STEAM LOCOMOTIVE 175 

smoke-box) were 16 Jin. diameter and 24in. stroke; the driving axle 
was above the frame. The boiler was 10ft. 9in. long and 4ft. Jin. 
diameter; it contained 180 brass tubes of one and thirteen-sixteenths 
inch external diameter. The steam pressure was 1301b. No dome 
was provided, and the Salter safety valves were located on the top 
of the fire-box and enclosed by a brass casing. The weight of the 
engine, in working order, was 42 tons. The water was stored in 
three tanks, one beneath the boiler, another below the 
fire-box, and the usual well tank, behind the foot-plate. 
The two suspended tanks were connected by means of a 
stuffing-box jointed pipe, which was continued to the bottom of the 
wheel-tank, so that the water in the three tanks was thus able to pass 
from one tank to any other .one. The feed-pumps were worked from 
the piston-rod cross-head, and the feed-pipes passed along behind the 
splashers to the boiler. To steady the suspended tanks, link-rods were 
passed between the two. There were also "bogie safety links" con- 
necting the bogie frames with the main frame at each end, and 
similar links connected the suspended tanks with the other ends of 
the' bogie frames. 

These links were each fitted with india-rubber disc buffers, to allow 
of the necessary elastic working. The parts were thus so strongly 
linked together, that should a bogie centre-pin break, or should the 
bogie movement fail in any way, the wheels would still remain in 
their right position. The whole of the springs were of the india- 
rubber disc kind. Those of the driving axle presented some re- 
markable peculiarities. 

They were double, an elastic connection J3eing formed between 
the boiler and the axle-boxes by large plate brackets projecting from 
the boiler barrel, and carrying centre studs for a short double-armed 
lever; each end of this lever had a separate spring-box attached to 
it by a long link. 

The inner spring-box worked down behind the disc plate of the 
driving-wheel splashers, whilst the outer one worked parallel to it, 
outside the driving wheel. 

The springs for the other wheels were all beneath their axles, and 
were very compact and neat in appearance. The brake action was 
confined to the after bogie, all four wheels being used for the fric- 
tional effect, the sliding bars carrying the brake blocks being actuated 
in reverse directions by a screw spindle, which carried a winch 
to be worked by tne driver. 



176 EVOLUTION OF THE STKAM LOCOMOTIVE 




EVOLUTION OF THE STEAM LOCOMOTIVE 177 

The regulator valve was a slide, worked in a simple and certainly 
a convenient manner by a short lever, set on a pillar stud on the 
front of the fire-box, and passing through a slot in the end of the 
slide spindle. This was a far more effective plan of working the 
valve than the ordinary rotatory handle. 

These engines were remarkable for their steady running at high 
speeds, 80 miles an hour and over being a daily performance of the 
engines on certain portions of the main line between Exeter and 
Bristol. 

One reason for the freedom from excessive oscillation for which 
these engines were famous was attributable to the 9ft. driving wheels, 
and the slow piston velocity arising therefrom ; thus with 6ft. wheels 
at a speed of 60 miles an hour, the pistons have to make no less than 
280 double strokes per minute without making allowance for " slip." 
With the 9ft. driving wheels the double piston strokes per minute at 
60 miles an hour fall to 186, and consequently with so consider- 
able a reduction in the movements of the reciprocating and rotating 
machinery of the locomotive, it is only reasonable to expect and 
obtain a much more steady movement of the machine. 

In the matter of coal consumption the engines were no less suc- 
cessful. Writing in August, 1856, Mr. Pearson reported: "Engine 
No. 40 has run 81,790 miles since her delivery in October, 1853, and 
has consumed 794 tons 17 cwt. 2 qr. of coke, or 21.761b. per mile; 
the repairs as yet have been very trifling, consisting chiefly of re- 
turning the tyres. This enghie has been working passenger trains 
on the main line almost the whole of the time since she was delivered. 
Our mileage is rather heavy, each engine averaging 750 miles per 
week." 

After 1876, when the Bristol and Exeter Railway was amalga- 
mated with the Great Western Railway and the former company's loco- 
motive stock became the property of the latter, 4 of the 8 original 
9ft. tank engines then in existence were rebuilt, and their character 
and design entirely remodelled. The diameter of the driving wheels 
waa reduced to 8ft., and tyres fitted to them, a pair of trailing wheels 
were provided in place of the rear bogie, and a separate tender was 
added, the tanks being done away with. The B. and E.R. numbers of 
these engines were 39 to 46. The G-W.R. numbered the four taken 
over 2001 to 2004. The latter was hauling the "Flying Dutchman" 
when the Long Ashton accident happened on July 27th, 1876. It was 

8 



178 



EVOLUTION OF THE STEAM LOCOMOTIVE. 




EVOLUTION OF THE STEAM LOCOMOTIVE 179 

in consequence of this disaster that the engines were rebuilt with 8ft. 
wheels. In concluding this sketch of Pearson's famous broad-gauge 
double tanks, we may state that until recent years, when phenomenally 
high locomotive speeds have been recorded, these engines held the 
"blue ribbon" in that respect with an authenticated speed of 81 miles 
an hour. Figures 69 and 70 represent them as rebuilt. 

The Furness Railway Company is certainly notorious for the 
manner in which it preserves its locomotives; not only has it the 
two old Bury engines (already described) yet in active service, but 
there are still at work on the same Company's iron roads other 
.engines manufactured as long ago as 1854. These locomotives are 
first cousins to Bury's four wheel (coupled) goods engines ; they were 
built by Fairbairn, of Manchester, and have cylinders 15in. diameter, 
with a stroke of 24in. Of course, they are technically inside cylin- 
der i.e., of the Bury "inside" type, with the cylinders within the 
frames, but below the smoke-box, instead of within it. The cylinders 
are, in fact, but a few inches above rail level; they incline upwards, 
and the connecting-rods pass beneath the leading axle and actuate 
the trailing axle ; the four wheels are 4ft. 9in. diameter, and are 
coupled by means of round section side-rods; the wheel base ;s 
7ft. 9in. ; the frames are of the inside bar pattern; the fire-box is 
round, with circular top, and surmounted by a double Salter safety 
valve. The boiler is lift. 2in. long and of 3ft. llin. mean diameter; 
it contains 148 tubes, 2in. diameter. The total heating surface is 
940 sq. ft. ; steam pressure, 1201b. ; weight of engine in working order, 
22 J tons. There is no dome on the boiler; but some modern attach- 
ments have been fixed on the upper portion of the round fire-box, the 
.steam pressure gauge being very noticeable. The tender is sup. 
ported on four wheels of 3ft. diameter, the wheel base being 8ft., 
capacity of tank 1,000 gallons, and coal space 100ft. ; weight in work- 
ing order, 14 J tons. The tender has outside frames, and the brake 
actuates blocks to both sides of the four wheels. These engines are 
used for working goods and mineral traffic over the Furness Railway. 
The particular engine we have been describing is "No. "9." 

"Ovid" (Fig. 71) represents a type of bogie saddle-tank engines, 
with four-coupled wheels, designed by D. Gooch for working the pas- 
senger trains on the steep inclines of the South Devon Railway. The 
cylinders were I7in. diameter, with a stroke of 24in. The coupled 
wheels were 5 ft. in diameter. Weight, in working order, 38 tons, 

N2 



180 EVOLUTION OF THE STEAM LOCOMOTIVE 

Steam pressure, 120 Ibs. per square inch. "Ovid" was built by Haw- 
thorn in 1854. 

" Plato " (Fig. 72) was one of the six-coupled banking engines, da- 
signed by Gooch for the South Devon Railway. She was built at 
Swindon-in 1854. The steam pressure, cylinders, stroke, ana weight 
were the same as in the " Ovid " class. The wheels were 5ft in dia- 
meter. The tanks contained 740 gallons of water. The rectangular 
projection in front of the smoke-box is the sand-box ! 

Neilson and Co., of the Hyde Park Works, Glasgow, pro- 
duced in 1855 a type of outside cylinder goods engine. Readers will 
remember that at that period goods locomotives were not necessarily 




FIG. 71. "OVID," A SOUTH DEVON RAILWAY SADDLE TANK ENGINE, 
WITH LEADING BOGIE 

of the six or eight wheels coupled description ; they more generally had 
but the leading and driving wheels coupled. This type of engine, it will 
be remembered, is now usually described as "four-coupled in front" 
or a "mixed traffic" engine. The locomotive in. question was built 
for the Edinburgh and Glasgow Railway, and was numbered "353" in 
Neilson and C'o/s books. 

The boiler was of considerable length, and appeared longer from 
the fact that the fire-box top was not raised, so that a long, un- 
broken line of boiler top met the eye, relieved at the extremity of 
the fire-box end by being surmounted by an immense steam dome, en 
the top of which was fixed an enclosed Salter safety valve. The 
horizontal outside cylinders were below the foot-plate side frames, 
located as usual at the smoke-box end. Their diameter was 16in. 



EVOLUTION OF THE STEAM LOCOMOTIVE 181 

and stroke 22in. The coupled wheels were 5ft. and the trailing 
wheels 3ft. Gin. diameter. 

The frames were u inside," and the driving and leading wheels 
were provided with inside bearings only, but by a curious practice cf 
bolting on to the main frames at about the mid41e of the fire-box 
an elongated portion, which curved outwards, the trailing wheels were 
provided with outside bearings. The rams actuating the boiler feed- 
pumps were simply extensions of the piston-rods, the pumps being 




FIG. 72. "PLATO," A SIX-COUPLED SADDLE TANK BANKING ENGINE, 
SOUTH DEVON RAILWAY 

fixed between the leading and driving wheels. The engine was pro- 
vided with a steam-pressure gauge, fixed on a vertical pillar over 
the top of the fire-box indeed, in much the same position the steam 
gauge still occupies, save that "No. 353" had no cab or weather- 
board, and it therefore appeared singular to see the gauge in the 
place indicated. 

Rotatory valves for locomotives are almost annual "inventions," and 
as old friends as the " biggest gooseberry " and " sea serpent," which 
appear regularly year by year. Under such circumstances, we may be 
excused for giving an account of Locking and Cook's patent rotatory 
valve, fixed to the York and North Midland Railway engine, " No. 48," 
on January 26th, 1854, and taken out in May of the same year, the 
locomotive in the interim having run 10,000 miles. " No. 48 " was 
used on the Hull and Bridlington branch and although she was an 
old engine, having been built for the Hull and Selby Railway in 1840, 
yet with the rotatory valve, good old " 48 " is stated to have con- 
sumed 20 per cent, less coke than a modern engine doing the same 



182 EVOLUTION OF THE STEAM LOCOMOTIVE 

work on the same branch; we also read that when the valve was 
removed no perceptible wearing; was to be noticed. We are not, 
however, aware that il No. 48 " or any other of the York and 
North Midland Railway locomotives were afterwards fitted with 
Locking and Cook's patent rotatory valves. 

Mr. G. Tosh, locomotive superintendent of the Maryport and Car- 
lisle Railway, designed in 1854 a powerful goods engine to work the 
heavy mineral traffic over the railway. This engine had six coupled 
wheels, 4ft. Tin. diameter; cylinders, 16fin. by 22in. stroke; heating 
surface tubes, 1,181ft.; fire-box, 84ft.; total, 1,265 sq. ft.; steam 
pressure, 1201b. ; weight, 26 tons 12 cwt. ; cost, 2,175. She hauled 
a train of 100 loaded wagons, weighing 445 tons, for a distance of 
28 miles in 1} hours. The line is of a very undulating character, 
including an ascent nine miles long, one mile of which is 1 in 192. The 
wagons were borrowed from, the Newcastle and Carlisle Railway, and 
the 100 only weighed 172 tons, or an average of less than If tons 
each. 

The dead weight of mineral wagons has largely increased sines 
1854, although it is to be feared their carrying capacity has not 
increased in the same proportion. 




FIG. 73. THE FIRST TYPE OF NARROW-GAUGE PASSENGER ENGINES 
ON THE GREAT WESTERN RAILWAY 

About this time, the growth of narrow gauge lines in the districts 
served by the G.W.R., together with the amalgamations and alliances 
of narrow gauge railways with the G-W.R,, made it necessary for the 
tetter railway to provide narrow-gauge engines. Fig. 73 represent 
one of the first narrow-gauge Great Western locomotives. It will be 



EVOLUTION OF THE STEAM LOCOMOTIVE 



183 



seen that Daniel Gooch introduced all his well-known features into 
these engines. These locomotives were built by Beyer, Peacock, and 
Co. The " single " driving wheels were 6ft. Gin. diameter, the cylin- 




ders being 15 Jin. diameter, and the stroke 22in. Compensation levers 
connected the leading and driving springs. 

Tn 1855 Sir D. Gooch designed a class of coupled express boad 
gauge engines for the Great Western Railway. These engines had a 



134 



EVOLUTION OF THE STEAM LOCOMOTIVE 



group of four leading wheels, like the " Lord of the Isles " class. The 
driving and trailing wheels were coupled, and were 7ft. in diameter. 
At that time, no coupled wheels of so large a diameter had been con- 
structed. The cylinders were 17in. diameter, with a 24in. stroke- R. 
Stephenson and Co. built the engines, of which there were 10. They 
were a most successful class of engine, and ran about 500,000 miles 
each before being " scrapped." " Robin Hood " (Fig. 74) was one of 
these engines. By reference to the illustration, it will be seen that 
the tender was fitted with the sentinel b.ox for the " travelling porter " 
that formerly accompanied the G.W. broad-gauge expresses. 

Fig. 75 represents the inspection or cab engine of the N.B.R., it is 
numbered 879, and was originally built by Messrs. Neilson and Co., in 
1850, for the Edinburgh and Glasgow Railway. She is now used for 
inspection purposes. The cylinders are lOin. diameter by 15in. 
stroke. Other dimensions are : Wheels, leading and trailing, 3ft. 
diameter; driving, 5ft. diameter; wheel base, 15ft. Sin. ; centre of 
leading to centre of driving, 10ft. Sin. ; centre of driving to centre of 
trailing, 5ft. Tubes, No. 88, Ifin. diameter outside. Heating sur- 
face : Tubes, 324 sq. ft. ; fire-box, 35 sq. ft. ; total, 359 sq, ft. Fire- 
grate, 5 sq. ft. Weight, in working order, 22 tons Icwt- 3qrs. Tank 
capacity, 426 gallons. 




F;o. 75. NORTH BRITISH RAILWAY INSPECTION ENGINE, No. 879 



CHAPTER XI. 

Improvements in coal-burning locomotives Beattie'a system Trials of the 
"Canute" Yorston'a plan Cmlworth's successful efforts Yarrow's apparatus 
P. K. Clark's system tried on tho North London and other railways Wilson's 
plan fitted to engines working the O. W. & W.R. Lee and Jacques' experiments 
Frod-sharu's device tried on the E.C.R. Douglas' system The various plans 
reviewed " Nunthorpe,'' a S. & D.R. engine Double engine on the Turin 
and Genoa Railway Cromptoii's engines on the E.K.R. French locomotives 
on the F C.R. Gifford's inveuticn of the injector First fitted to the "Pro- 
blem" Ramsbottom's vatwr "pick-up ' ypparatus Brunei's powerful B.G. 
tanks for the Vale of Neath Railway Incorporation of the Metropolitan 
Railway Trial of Fowler's "hot brick" engine Its end Fletcher's saddle 
tanks "75," T.V.R. Second-hand locomotives on the L. & S.W.R. The 
" Meteor "Early L.C. .fc D.R. engines. 

We have now reached an era in the "evolution of the steain 
locomotive" which, in its after development, amounted to a complete 
revolution in the character of the fuel used for locomotive purposes. 
The year 1855 found the locomotive, or rather those responsible for 
its working, on the threshold of successful experiments, which re- 
sulted in the complete substitution of the "black diamonds" in their 
natural state for locomotive fuel in preference to the use of coal after 
it had undergone the process of carbonification necessary to form coke. 
It must not be forgotten that steam-users never had a preference 
for coke, but they were compelled to use it, because the more volatile 
coal produced so much smoke in the process of combustion that 
legislative action (which compels locomotive engines to be so con- 
structed as " to consume their own smoke ") practically prevented the 
use of coal until science discovered a method of consuming the smoke. 
There had been various attempts to reach this desirable state, 
and we have from time to time in this series of articles described 
certain of these efforts; but none of them up to the date under re- 
view had been sufficiently successful to warrant the adoption of any 
one of the methods proposed as a complete smoke-consumer. 

The successful efforts made by Beattie, of the London and South 
Western Railway, to solve the problem of smoke consumption in the 
locomotive so as to admit of coal being used as fuel stand out promi- 
nently. The salient points of his smoke-consuming locomotive com- 
prised an enlarged fire-box, a combustion chamber,, the transverse 



186 EVOLUTION OF THE STEAM LOCOMOTIVE 

division of the fire-box by means of an inclined water bridge, and 
the fire-box arched with fire-bricks. A perforated fire-door for the 
admission of air to the fire-box was another of the features of Beattie's 
system, as were also the use of the ashpan dampers and the employ- 
ment of an auxiliary steam jet in the chimney for use when the engine 
was at rest and the ordinary exhaust blast consequently not available. 
With the addition of a feed-water heating apparatus Beattie reduced 
the fuel consumption to from .12 to .171b. per ton mile. 

The dimensions of the London and South Western Railway loco- 
motive "Canute^ (an engine filled with Beattie's coal burning appara- 
tus) were : -Cylinders (outside), 15in. diameter, 21in. stroke; driving 
wheels, 6ft. Gin. diameter. The fire-box was 4ft. llin. long, 3ft. 6in. 
wide, oft. lin. deep at the back, and 4ft. lin. in front. The combustion 
chamber had a flat roof, was 4ft. 2in. long, and 3ft. Gin. diameter. 
The tubes were 6ft. long, IJin. diameter, and 373 in number. Total 
area of fire-grate, 16 sq. ft. 

The heating surface of the ''Canute" was as follows: Box, 
107 sq. ft. ; combustion chamber, 37 sq. ft. ; tubes, 625 sq. ft. ; 
total, 769 sq. ft ; in addition to which red-hot bricks presented a 
surface of 80 sq. ft., not, however, for heating the water, but for the 
purpose of burning the smoke. Four series of trials were made with 
the "Canute" engine No. 135, and these are detailed in " Locomotive 
Engineering/' The experiments are described as " 1st, the engine 
in its usual order, with coal, bricks, and hot feed-water ; 2nd, with 
coal, bricks, and cold water; 3rd, with coke, bricks, and hot feed- 
wuter; 4th, with coal and hot feed-water, but without the bricks." 
Three different kinds of coals were used for the experiments. The 
following is a brief summary of the experiments: 1st, a regular 
express train, of 10J coaches, weighing 66 tons, or with the engine 
and tender, 99 tons. Average speed, exclusive of stoppages, 34 miles 
an hour; consumption of coal, 151b. per train mile; water evaporated, 
9.351b. per Ib. of coal consumed; average temperature of heated feed- 
water, 187 degrees. 2nd trial, a weighted train of 28 coaches, 
weighing with engine and tender 236 tons. Average speed, exclusive 
of stoppages, 301 miles an hour; coal consumed, 28|lb. per mile, 
8.87lb. of water evaporated by each pound of coal; temperature of 
feed-water, 212 degrees. 3rd experiment with an express train, but 
without the fire-bricks in the fire-box, showed that a saving of 12 pei 



EVOLUTION OF THE STEAM LOCOMOTIVE 



187 



cent, was due to the use of the fire-bricks, and with coke instead, of 
coal as fuel, the saving was 24 per cent, in favour of coal ; whilst the- 
use of the feed-water heating apparatus showed a saving of 30 per 
cent, of fuel. Beattie's apparatus is illustrated by Fig. 76, the 
" Dane,'.' being a similar locomotive to " Canute." 




FIG. 76. THE "DANE," L. & S.W.R., FITTED WITH BEATTIE'S PATENT 
APPAKATUS FOE BURNING COAL 



As the feed-water heating apparatus was an important innovation 
in locomotive practice, it will be of interest if we append a descrip- 
tion of the same. In outward appearance, the most noticeable portion 
of the apparatus was the condenser, a cylindrical appendage placed in 
a vertical position on the top of the smoke-box and in front of the- 
chimney. From a casual glance, the condenser much resembled the 
steampipe of a steamship which is usually to be observed outside the-, 
smoke-stack. From the bottom of the condenser, outside the 
engine, a pipe conveyed the heated water and steam back to the 
tender. The method of working was for the exhaust steam to be 
discharged from the blast pipe into the condenser, which, as previ- 
ously explained, was on. the top of the smoke-box, and consequently 
right over the blast orifice. Here the exhaust steam was mixed with a 
jet of cold water, which was pumped into a condenser. The result 
of such meeting was the condensing of the steam and heating of the 
water, which flowed by gravitation through the pipe previously 
described. The supply pump for the boiler was worked off this pipe, 
and both the heated water and that from the tender were together 



188 EVOLUTION OF THE STEAM LOCOMOTIVE. 

pumped into the boiler. If the boiler were not being fed, the heated 
water from the condenser, instead of passing into the boiler, flowed 
through the pipe into the tender, and thus raised the temperature of 
the whole of the water in that vessel. 

It should be mentioned that before entering the boiler the tem- 
perature of the feed-water was further increased by passing through 
a special heating apparatus, fixed in the smoke-box. This smoke- 
box chamber was heated by the exhaust steam, which passed through it 
after leaving the blast pipe, and before entering the external con- 
denser placed above it. By these methods the temperature of the 
feed-water was raised above the boiling point before entering the 
boiler. 

The engines of this design gave satisfaction, both as regards smoke- 
consuming and feed-water heating, and to Beattie, therefore, is due 
much of the honour of successfully overcoming the defects that pre- 
viously existed in so-called " smoke-consuming " locomotives. The 
" Canute " can, therefore, be considered amongst the earliest of the 
locomotives burning coal in such a manner as to consume the smoke. 
It should be mentioned that in later engines built under Beattie's 
patent the external condenser fixed on the top of the smoke-box in 
front of the funnel was not used, a modified form of interior appa- 
ratus being substituted. 

It must not be supposed that at this period Beattie was alone in 
the field of experiment relating to " smoke-consuming " locomotives. 
Several other engineers were engaged in the same useful research, 
amongst whom we mention Yorston, Cudworth, Yarrow, D. K. Clark, 
Wilson, Lee and Jacques (jointly), Sinclair, and Douglas. Yor&tons 
plan was patented by Sharp, Stewart, and Co. in 1855. The fire-box 
was divided into two 1 parts by a transverse mid-feather, which was 
perforated by a series of tubes, to allow the coal gases to escape and 
air to enter. The coal was fed into the portion of the fire-box next 
the tubes, the front part being reserved for coke; separate fire-doors 
were used for introducing the coke and coal into the fire-box. The 
air entering through the perforations in the fire-box, at the tube-plate 
end, was expected to force the smoke, etc., from the coal fire over 
the incandescent coke, where the combustion of the coal would be com- 
pleted. The system, however, appears to have been better in theory 
than practice, as no particular steps were taken to push the inven- 
tion in question. 



EVOLUTION OF THE STEAM LOCOMOTIVE 



189 



With Cudworth's system the opposite course was adopted, and 
resulted in his engines taking a foremost position among those burn- 
ing coal as fuel. 

Mr. Cudworth, the locomotive superintendent of the South Eastern 
Railway appears to have made his first experiments with engine No. 
142, which during July, 1857, was tried as a coke-burning locomotive ; 
but during October and November of the same year experiments were 
made with this engine, fitted with Cudworth's patent grate, etc. 

The principal dimensions of Cudworth's standard passenger engines 
were as follows: Cylinders, 16in. by 24in. stroke; driving wheels,. 
6ft. diameter; wheel base, 15ft.; heating surface, 965ft.; grate area, 
21 sq. ft. Total weight in working order 30 J tons, of which the 
leading axle supported 9 tons 9 cwt., driving lOf cwt., and 




FIG. 77. CUDWORTH'S SLOPING FIRE GRATE, FOR BURNING COAL, AS 
FITTED TO SOUTH EASTERN RAILWAY LOCOMOTIVES 

trailing 10 tons 6 cwt. The tender was carried on six 
wheels, and weighed in working order 20 \ tons. These engines had 
inside cylinders and "back-coupled" driving wheels, and for many 
years comprised the principal type of South Eastern passenger 
locomotives. Several of them are still running, but rebuilt, their former 
distinguishing features viz., the large brass dome on the centre of the 
boiler barrel, the raised fire-box, with a brass encased Salter safety 



190 EVOLUTION OF THE STEAM LOCOMOTIVE 

valve, the sloping fire-grate, and the peculiar chimney--all having 
been -removed during the present . locomotive regime. 

The chief feature in Cudworth's system was the long, sloping fire- 
box, which was 7ft. Gin. in length, the grate being 7ft. long, illustrated 
by Fig. 77. The fire-box was divided into two parts by a longitudinal 
mid-feather, thus forming two furnaces, with separate doors ; the two 
furnaces united at the lower end in front of the tube-plate. 'j.he 
coal was introduced alternately into each furnace, being placed just 
within, the doors ; the sloping grate and the motion of the engine 
caused the fuel to gradually slide down the grate towards tne tube- 
plate, and by the time the fuel had reached the lower end of the grate, 
the smoke had become separated from the carbon of the coal, and was 
consumed by the incandescent mass of fire at the lower end of the 
grate, as it passed over the same on its way to the tubes. . 

Cudworth employed neither combustion chambers nor air-bricks in 
his system; but air was admitted to the fire-box by means of a 
damper fixed in the front of the lower end of the grate. A steam-jet 
was fixed in the chimney to create a sufficient draught when the 
engine was still. Cudworth's " smoke-consuming " locomotives were 
as economical in coal as Seattle's, whilst the former's system was 
much more simple. 

On March 18th, 1857, Thomas Yarrow, of Arbroath, was granted 
a patent for his smoke-consuming apparatus for locomotives, which 
was, used on the Scottish North Eastern Railway. Th-e leading char- 
acteristic of the design was a flat arch of fire-bricks constructed inside 
an ordinary fire-box. The lower end of the arch commenced below the 
bottom row of tubes, and the arch was continued upwards in a slant- 
ing direction till within 8 or 10 inches of the roof of the fire-box. 
Upon the top of this arch were fixed a number of tubes, through 
which the vapours passed before reaching the ordinary boiler tube**. 
Hot air was supplied to the fire by means of pipes with trumpet- 
shapsd mouths placed in front of the ashpan. The fire-bars were fixed 
on a transverse rocking-shaft fitted with several short arms, upon 
which the ends of the fire-bars rested. To. prevent the formation of 
clinkers, an occasional rock was given to the fire-bars by the fireman, 
a sector being provided for the purpose. Yarrow's system required 
the coal to be placed at the extreme front of the fire-box, so that the 
smoke was forced by the brick arch to return towards the fire-door 
; before it could get over the arch and enter the tubes, and in the 



EVOLUTION OF THE STEAM LOCOMOTIVE 191 

passage the denser portion of the smoke was burnt. The patent also 
included the use of a steam-jet in the chimney for use when the 
engine was not working, and the heating of the feed-water by means 
of the exhaust steam. 

Late in 1857 D. K. Clark devised his system of smoke-consuming 
furnaces : the air was forced through tubes into the fire-box by the 
action of minute jets of steam, which acted much in the same way 
as the blast pipe in the smoke-box. The air-tubes were Ijin. dia- 
meter, with the steam-jet orifice contracted to one-sixteenth inch 
diameter. 

The first locomotive fitted with D. K. Clark's system was one of 
the North London Railway's tanks. This was in January, 1858, but 
only one side of the fire-box was fitted; four air-tubes were em- 
ployed, and with a small fire the prevention of smoke was complete. 
In April of the same year one of the passenger engines on the 
Eastern Counties Railway was fitted with Clark's apparatus. Four 
air-tube's were fitted to one side of the fire-box, and three to the 
other side. In the following January a South Eastern Railway 
passenger locomotive was fitted with two rows of seven tubes each, 
through the front and back of the fire-box. In March, 1859, a Great 
North of Scotland Railway engine was fitted with tubes on Clark's 
system, with such satisfactory results that the whole loco- 
motive stock of that railway was speedily fitted with the apparatus. 
No complete investigation appears to have been made as to the work 
performed by the jets of steam as employed by Clark. It is generally 
supposed that the steam had a merely mechanical effect viz., that 
cf drawing The air into the fire-box. It has also been suggested that 
the steam produced a chemi,cal combination which facilitated the 
combustion of the volatile gases, besides precipitating the unconsumed 
carbonaceous matter on the fire. The result of the adoption of the 
system on the Great North of Scotland Railway's locomotives was 
mich that the coal consumption fell to under .21b. of coal per ton 
mile. A trial was also made of Clark's system on the London 
Brighton, and South Coast Railway, one of the old passenger engines 
being fitted with air-tubes and steam-jets to the front of the fire-box, 
with good results. 

In 1858 Mr. Edward Wilson, who supplied the Oxford, Worcester, 
and Wolverhampton Railway with locomotive power by contract, fitted 
his system to -several of the engines on that line. Mr. David Joy, 



192 EVOLUTION OF THE STEAM LOCOMOTIVE 

the inventor of the celebrated Joy. valve-gear, was at that time loco- 
motive superintendent of the Oxford, "Worcester, and Wolverhampton 
Railway, and he possesses records of many runs of the engines so 
fitted, and the comparisons between the fitted and unfitted engines 
show an immense saving of fuel by the former; indeed, the coal con- 
sumption was remarkably low considering the severe nature of the 
line between Oxford and Worcester. Some short time ago Mr. Joy 
showed the writer the tabulated results of these trials, and, if 
memory serves .correctly, the coal consumption averaged about 2 lib. 
per train mile. Wilson's system consisted in fixing several tubes from 
the bottom of the fire-box underneath the whole length of the boiler 
and smoke^box, so that the mouths of the air-tubes projected in front 
of the engine, and the resistance of the train when travelling forced the 
air through the tubes into the fire-box. By his method Wilson ob- 
tained a forced draught without the expenditure of the steam, which 
was necessary in Clark's system. 

Lee and Jacques' system was introduced on the East Lancashire 
Railway in July, 1858. It consisted of a narrow fire-brick arch, and 
a deflector fixed at the top of the underhung fire-door. The deflec:or 
projected in a downward sloping direction into the fire-box. A valve 
for controlling the supply of air to the fire-box was fitted to the fire- 
door, and this valve was worked by means of a sector. The air 
entered the fire-box through the valve, and the deflector caused the 
air to be projected downwards on to the fuel, whilst the brick arch 
prevented the immediate escape of the gases, and kept them within 
the fire-box sufficiently long for the smoke to be consumed. 

In December, 1858, Mr. Sinclair, the locomotive superintendent of 
the Eastern Counties Railway, commenced to fit some locomotives 
with the deflecting plate, etc., on a plan introduced by a Mr. Frod- 
sham. The fire-door was underhung, and the baffle-plate was fixed 
above it, to direct the air down on to the fuel; whilst instead of a 
brick arch, two steam-jets were used, one on each side of the door. 
These also helped to force the air on to the burning fuel and to drive 
the liberated, but unconsumed, smoke back into the fire, when it 
was consumed. 

Mr. Douglas's plan was adopted by the Birkenhead Railway. He 
combined the use of an inclined fire-grate of large area,, and a baffle- 
plate. In January, 1858, when first introduced, the deflector was fixed 
to the inner side of the fire-door, but in June of the same year an 



EVOLUTION OF THE STEAM LOCOMOTIVE 193 

underhung fire-door and movable baffle-plate were employed. These 
afterwards gave place to a plain inverted scoop, to project the air 
right on to the fire. 

After reading the description of the various plans adopted for the 
consumption of the smoke, readers will at once observe that each 
and every designer had the same object in view viz., to supply a 
sufficient volume of air to the fire, and mix the air with the uncon- 
sumed gases given off by the burning coal, and then to prevent the 
immediate escape of this gaseous mixture from the fire-box. Being 
retained within the heated fire-box, the temperature of the vapour 
was raised sufficiently, so that the vapour readily burnt when forced 




FIG. 78. "NUNTHORPE," A STOCKTON AND DARLINGTON RAILWAY 
PASSENGER ENGINE, BUILT IN 1856 

by the steam deflector, or brick arch (according to the system adopted), 
back on to the incandescent fuel. As stated, the object of all the 
inventors was the same, but the methods adopted were different, and 
these latter (though some systems had advantages that others lacked) 
were successful in each case; but from the whole could be chosen 
some that certainly were more noteworthy, both as regards simplicity 
of application and design, and others that were more successful in 
attaining the object in view viz., a consumption of the smoke given 
off by the coal. In these four years 1855-59 however, the problem 
of consuming the coal smoke was successfully accomplished, 
and the era of the coal-burning locomotive definitely inaugurated. 



194 EVOLUTION OF THE STEAM LOCOMOTIVE 

Fig. 78 is an illustration of the "Nunthorpe," No. 117 of the 
Stockton and Darlington Railway. This engine shows a distinct advance 
in locomotive construction ; indeed, it is possible at the present time 
to see on some lines engines somewhat similar in appearance still at 
work. She was built by Gilkes, Wilson and Co., in 1856, and was in- 
tended for passenger traffic. Four of the six wheels were coupled, these 
being 5ft. in diameter. The cylinders were inside, 16in. in diameter, and 
with 19in. stroke. The tender was on six wheels, and the tank capa- 
city was 1,200 gallons. The cost of the engine was 2,550. It will 
be observed that the weather-board of the " Nunthorpe " afforded very 




FlG. 79. BEATTIE'S 4-COUPLED TANK ENGINE, L. & S.W.R., 1857 

little protection to the driver and fireman, but its inclusion in the 
design of the engine was a step in the right direction. 

In 1857 Beattie designed a handy class of passenger tank engines 
for the L. and S.W.R. Three were built at first, and named " Nelson/' 
" Howe," and "Hood." They had four coupled wheels, 5ft. diameter, 
and a small pair of leading wheels. The cylindars, which were out- 
side, were 15in. diameter, the stroke being 20in. These engines are 
illustrated by Fig. 79. They were good locomotives, and "Hood" 
and "Howe" continued in work till 1885. 

Fairlie is usually given the credit of introducing double loco- 
motives with a centre foot-plate. By reference to Chapter IX., it wLl 
be seen that the design was patented by Pearson, of the Bristol and 
Exeter Railway, as long ago as 1847, and in 1855 a double 
engine, built by R. Stephenson and Co., was at work on the Giovi 
incline of the Turin and Genoa Railway. The incline in question 
commences 7f miles after leaving Genoa, and is six miles long, the 
average gradient being 1 in 36. The double locomotive was of the 
tank type. The wheels were 3ft. 6in. diameter, the cylinders 14m. 



EVOLUTION OF THE STEAM LOCOMOTIVE 195 

diameter, and the stroke 22in. The machine actually appears to have 
been two engines placed fire-box to fire-box, and connected by means 
of a foot-plate between the two fire-boxes. The combination, with 
fuel and water, weighed 50 tons. In fine weather a load of 100 tons 
was hauled up the Giovi bank at 15 miles an hour; in bad weather 
the load was reduced to 70 tons. 

The first portion of the East Kent Railway from Chatham to 
Faversham was opened in January, 1858, the original locomotives 
being designed by Crampton, who was one of the contractors for the 
construction of the line. The engines in question were "tanks," and 
weighed 32 tons each at that period considered an excessive weight 
for an engine. They were also unsteady and generally unsatisfactory, 
frequently running off the metals. 

Mr. Robert Sinclair was appointed locomotive superintendent of 
the Eastern Counties Railway in 1858, and his first design of engines 
was a class for working the goods traffic, of which only six were con- 
structed, Rothwell and Co. being the builders. The engines had a 
pair of leading wheels, 3ft. Tin. diameter, and two pairs of coupled 
wheels, 5ft. diameter; the cylinders were 18in. diameter, the stroke 
being 22in. 

During the following years another class of goods engines (Fig. 80) 
were built by various firms from Mr. Sinclair's improved design. 
Indeed, as will be seen later on, some were even constructed 
by the French firm of Schneider and Co. These Lad 
outside cylinders, and inside frames to all wheels. The 
coupled wheels (D. and T.) were 6ft. Sin. diameter, and the leading 
3ft. 9in. diameter. The boiler was 10ft. 9in. long by 4ft. 2in, dia- 
meter, and contained 203 tubes, of If in. diameter; heating space, 
1,122 sq. ft.; weight, 35 J tons. Twenty-one of these engines, built 
by Neilson and Co., had Beattie's patent fire-box, which was sur- 
mounted by a large dome. These were numbered 307 to 327. When 
Mr. W. Adams was appointed locomotive superintendent of the Great 
Eastern Railway, he rebuilt several of these engines with a leading 
bogie in place of the pair of wheels. 

In November, 1858, a design of locomotive engine was patented, 
four pairs of coupled wheels being employed, all of which were 
located under the boiler barrel. The two leading pairs of wheels 
had outside axle-boxes, and the two trailing pairs inside 
axle-boxes, the latter having a lateral motion. The cylinders 

o 2 



196 



EVOLUTION OF THE STEAM LOCOMOTIVE 




EVOLUTION OF THE STEAM LOCOMOTIVE 197 

were inside, under the smoke-box, but the method proposed 
for working the locomotive was of a curious type, being somewhat 
after the fashion employed in ancient steamboats, the pistons work- 
ing out towards the front buffer beams, but connected to the leading 
wheels by outside cranks working off the cross-heads. 

A design for four-wheel tank engines was patented by S. D. 
Davison, in February, 1859, the leading feature being plate-iron 
frames formed into tanks for holding a supply of water. 

Attention must now be given to an invention that has proved of 
enormous value to the locomotive engineer, but which from its sim- 
plicity of action, yet apparent impossibility, -was not at first deemed 
worthy of practical use. On July 23rd, 1858, a patent was granted 
to H. J. Giffard, a Frenchman, for his injector, or boiler feeder, which 
in a short period almost completely superseded feed pumps, with 
their attendant friction, uncertainty of action, and excessive 
outlay for maintenance and repair. But above these minor 
disadvantages of the feed pumps, the injector removed 
from the minds of locomotive engineers that great source 
of danger, a short supply of water in the boilers, as well as the 
additional expense and inconvenience of "exercising" the locomotives 
solely for the purpose of filling the boiler, or, where such a method was 
inconvenient, of working the engine over a "race" for the same pur- 
pose. The theory of the injector did not originate with Gift'ard, for 
as long ago as 1806 Nicholson mentioned it as applicable for forcing 
water, whilst other philosophers have suggested its utility; indeed, 
the principle was used in connection with vacuum sugar boiling 
pans 20 years before Giffard's patent. The story of Giffard's acci- 
dental discovery of the action of steam and water in supplying a 
steam boiler with additional water reads almost like an extravagant 
romance, but many other great inventions and scientific discoveries had 
beginnings that appeared quite as improbable. The action of the 
injector, although curious, is well known, and therefore needs no 
description here. It is stated that Rainsbottom's " Problem," built 
at Crewe in November, 1859, was the first locomotive fitted with 
Giffard's " injector." This engine was the prototype of the world- 
famous "Lady of the Lake" class. Her dimensions were, outside 
cylinders, 16in. by 24m. ; single driving wheels, 7ft. 7 Jin. diameter; 
weight in working order, 27 tons. These engines have inside frames 
and bearings to all the six wheels. 



198 EVOLUTION OF THE STKAM LOCOMOTIVE 

An invention of Mr. Ramsbottom in connection with the improve- 
ment of the working of the locomotive deserves attention at this 
point. We refer to his self-filling tender apparatus, as introduced 
in 1860 on the London and North Western Railway system, and 
afterwards partially on the Lancashire and Yorkshire Railway, but 
which until the last year or so ha,s not been used on other lines. The 
speed competition of recent years, and the expiration of the patent, 
has now caused the Great Western, Great- Eastern, and North Eastern 
to adopt the water pick-up apparatus. One advantage of the system 
is, of course, the considerable reduction in the dead weight a not 
unimportant factor in express train running. The superiority of 
Ramsbottom's system is easily seen by comparing the small light 
tenders in use on the London and North Western Railway with the 
gigantic ones adopted by the Great Northern, Midland, and other 
lines running long distances without stopping, but which systems 
are unsupplied with the water trough and the necessary pick-up 
apparatus. The first pair of water troughs appear to have been put 
down near Conway, on the North Wales section of the London and 
North Western Railway. They were of cast-iron, 441 yards long, 
18in. wide, and Tin. deep, the water being 5in. deep. At each end of 
the main trough was an additional length of 16 yards, rising 1 in 100. 
It was towards the end of 1860 that the first trial of the trough 
system was made. Here, again, as in the case of the " injector," the 
arrangement requisite to produce the effect is so simple that at first 
blush the effect appears to be the result of some marvellous secret 
power rather than the operation of a simple natural law, the effect of 
the travelling scoop upon the water being exactly the same as if the 
water were forced against a stationary scoop at a velocity equal to that 
at which the train is travelling. The lowest speed at which the appara- 
tus works properly is something about 22 miles an hour. This speed, 
however, brings it within the scope of fast goods trains, whilst express 
trains can scoop up the water when travelling at 50 miles an hour, and 
can pick up about 1,500 gallons in the length of the trough quarter 
of a mile. The speed of the train would not appear to have much 
effect upon the water picked up in passing over a trough, as although 
with a slower train less water would be raised per second, yet the 
extra length of time spent in travelling over the trough would com- 
pensate for the smaller amount of water raised per second. The water 
supply-pipe is fixed inside the tender ; it is slightly curved throughout 



EVOLUTION OF THE STEAM LOCOMOTIVE 199 

its entire length, and is expanded towards its upper end to about ten 
times the area of the bottom, in order to reduce the speed or force 
of the incoming stream, which is directed downwards by the bent end 
or delivering mouth at the top of the pipe. To the lower end of Ihis 
pipe is fitted a movable dip-pipe, which is curved forward in the 
direction of the motion of the tender, so as to act as a species of 
scoop. This dip-pipe is rendered movable and adjustable in various 
ways, with a view to its being drawn up clear of any impediments, 
such as ballast heaps lying on the way, and also to regulate the 
depth of immersion in the water of the feed-water trough, the dip- 
pipe being capable of sliding up inside the feed-pipe by a convenient 
arrangement of rods and levers. 

In order that the dip-pipe may enter and leave the feed-trough 
freely at each end, the rail surface at that part of the line is lowered 
a few inches, a descending gradient at one end of the trough serving 
to allow the dip-pipe to descend gradually into the trough, whilst a 
rising gradient at the opposite end enables it to rise out of the trough 
again, the intervening length of line between the two gradients being 
level. To meet emergencies, Mr. Ramsbottom provided a small ice- 
plough, to be used occasionally during severe frost for the purpose of 
breaking up and removing any ice which might form in the trough. 
This plough consisted of a small carriage mounted on four wheels, 
and provided with an angular-inclined perforated top, which worked 
its way under the ice on being pushed along the bottom of the trough, 
and effectually broke it up and discharged it over each side. 

A very powerful class of broad-gauge saddle tank locomotives was 
designed by Brunei for working the heavy coal traffic over the severe 
gradients of the Vale of Neath Railway. These engines were sup- 
ported by six coupled wheels of 4ft. 9in. diameter, the cylinders being 
18in. diameter, and the stroke 24in. The heating surface was 1,417.6 
sq. ft.; the water capacity of tanks was 1,500 gallons. The engines, 
which were fitted with Dubs' wedge motion, were built by the Vulcan 
Foundry Company, and weighed 50 tons in working order. A 
noteworthy performance of one of. these locomotives consisted in 
hauling a train of 25 loaded broad-gauge trucks, each weighing 15 
tons, the gross weight, including the engine, amounting to 425 tons. 
This train travelled up a bank of 1 in 90 for a distance of 4J miles. 
Such a load on the gradient mentioned is equal to one of 1,275 tons on 
the level, and in a general way we do not find engines hauling trains of 



200 EVOLUTION OF THE STEAM LOCOMOTIVE 

the latter weight upon our most level lines. The Vale of Neath perform- 
ance must, therefore, be regarded as an exceptional locomotive feat. 
These engines were numbered 13, 14, and 15, and not being provided 
with compensating beams between the wheels, it is stated that one axle 
frequently carried 20 tons of the total weight. During 1860 these 
thre'e locomotives were, under the advice of Mr. Harrison, rebuilt as 
tender engines, to reduce the weight on the wheels, the excessive 
amount oi : which had been very destructive to the permanent-way. 
The cost of the alterations to the engines and the addition of the 
tenders was 700 each engine. About the same time some of the 
other Vale of Neath six-wheels-coupled engines were converted into 
four-wheels-coupled bogie locomotives. 

The locomotive now to be described had but a very shadowy exist- 
ence; it was- rather a tentative essay to produce a steam locomotive 
without the aid of a fire. The idea when proposed by Sir John 
Fowler was not new, for more or less successful essays had already 
been made on a small scale s with engines, the steam for propelling 
which was generated in the same manner as in Fowler's locomotive. 

In 1853 a railway was incorporated as the North Metropolitan; 
the next year a new Act was obtained, and the title changed to the 
Metropolitan. This authorised the construction of a railway from the 
Great Western Railway at Paddington to the General Post Office; 
powers were afterwards obtained to allow the City terminus to be 
in Farringdon Street instead of at the Post Office. The Great 
Western Railway subscribed <! 75,000 of the capital, and for the 
convenience of that Company's through traffic the Metropolitan was 
laid out on . the mixed-gauge, and when it was first opened it was 
worked on the broad-gauge only, by the Great Western Railway a 
most sensible arrangement, and one which ought never to have been 
relinquished, seeing how well adapted the wider vehicles were for 
conveying the imme'nse crowds that travel by every train on this line. 
The Act of Incorporation specially provided that the line was to 
be worked without annoyance from steam or fire. At first it was 
proposed to convert the water into steam by means of red-hot bricks 
placed around the boiler, and Mr. (afterwards Sir) John Fowler designed 
such a locomotive, which was built by a Newcastle firm, and tried on 
the Metropolitan Railway between Bishop's Road and Edgware Road 
Stations before the line was opened. The first trial took place on 
Thursday, November 28th, 1861. The following is an account of the 



EVOLUTION OF THE STEAM LOCOMOTIVE 201 

trip : " The engine was of considerable size, and it was stated that, 
it could run on the railway from the Great Western at Paddington to 
Finsbury Pavement without allowing the escape of steam from the- 
engine or smoke from the fire. A few open trucks were provided 
with .seats, and when the gentlemen were seated, the new engine 
propelled them under the covered way of the Metropolitan Railway 
to the first station at the eastern side of the Edgware Road, and 
back again to the Great Western Station, the steam and smoke ^being 
shut off. The tunnel, or covered way, was perfectly fresh and free 
from vapour or smoke. On the signal being given to work the engine 
in the ordinary way, a cloud of smoke, dust, and steam soon covered 
the train, and continued until it emerged from the tunnel into the 
open air. The experiment was perfectly successful, but it was under- 
stood that engines so constructed would be rather more expensive 
to work than those running in the ordinary way." To work the 
Metropolitan Railway on this system would have required the erection 
of immense boilers at both ends of the line to heat the water for 
the locomotive, and also furnaces for making the bricks red-hot,, 
whilst the charging of the locomotive boilers with hot water and the 
fire-boxes with hot bricks would have occupied some considerable time- 
at the end of each trip. 

It is, of course, well known that the experiment was very far 
from being "perfectly successful." Indeed, "failure" would be n 
much better definition of the hot-brick engine, since the proposed 
method of working was not carried out. We understand the engine was 
sold to Mr. Isaac Watt Boulton, the well-known purchaser of second- 
hand locomotives, and for somo time remained in his " railway 
museum" before being finally scrapped. The Metropolitan Railway 
had, consequently, upon the failure of the hot-brick engine, to fall 
back upon the Great Western Railway for working the underground 
line, until Sir John Fowler's later design of engines, constructed by 
Beyer, Peacock, and Co., were ready to work the traffic. 

In 1862 Fletcher, Jennings, and Co., of Whitehaveii, designed a. 
handy type of saddle tank engine for shunting purposes, etc. The 
engine ran on four wheels, 3ft. 4in. diameter, the wheel base being 
6ft. The cylinders were lOin. diameter, with 20in. stroke. Allan's 
straight link motion was employed, and was worked off the leading 
axle (it will be understood that the four wheels were coupled). This 
method of actuating the valves was not conducive to good working,. 



202 



EVOLUTION OF THj. STEA M LOCOMOTIVE 



as, of course, if the coupling-rods worked slack the valve gear motion 
became disorganised. 

Fig. 81 is a photograph of engine No. 75, of the Taff Vale Railway, 
built at the Company's Cardiff Works in 1860. The six-coupled wheels 
were 4ft. Sin diameter, the cylinders were 16in. diameter, and the 
stroke was 24in. No. 75 weighed 32 tons in working order ; the steam 
pressure was 130 Ibs. per sq. in. She was employed in the heavy 
mineral traffic of the Tail Vale Railway, and from her design well calcu- 
lated to work over the heavy gradient of that system. 




FIG. 81. SIX-COUPLED MINERAL ENGINE, TAFF VALE EWY., BUILT 1860 

In 1862 the L. and S.W. Railway purchased some second-hand 
engines from a contractor. They were built by Manning, Wardle, and 
Co., Leeds, and comprised six-wheels-coupled saddle tank engines. The 
wheels were 3ft. diameter; cylinders, 12in. by 18in. stroke; wheel 
base, 10ft. Sin. ; length over buffers, 21ft. 6in. ; weight, empty, U tons 
8 cwt., loaded, 16 tons 4= cwt. The fire-box was surmounted by a safety 
valve enclosed within a high fluted pillar. The steam pressure was 
1201b. One of these engines is leased to the Lee-on-the-Solent (Light) 
Railway, and may be seen working the traffic on this little line, 
which, by the way, spends over twopence to earn each penny of its 
gross income. 

Before leaving the London and South Western Railway and its 
goods locomotives, it is as well to record the dimensions of the 



EVOLUTION OF THE STEAM LOCOMOTIVE 203 

"Meteor," No. 57, constructed at Nine Elms in 1863 from, the designs 
of Mr. Beattie. The cylinders were 16Jin. diameter, 22in. stroke; 
the leading wheels were 3ft. 3in., and the coupled (D. and T.) wheels 
-5ft. diameter; the wheel base was 14ft., of which 8ft. 2 Jin. was be- 
tween the coupled wheels. The leading wheels were under the boiler, 
.and the front buffer beam was about 6ft. in advance of the centre of 
this axle. An immense dome was fixed on the raised fire-box; the 
safety valve was within an inverted urn-shaped case on the boiler 
barrel. The weather-board had slight side-wings, and was curved 
upwards at the top, and so formed an incipient cab. The fire-box 
.sloped from the tube-plate towards the foot-plate. The total weight, 
in working order, was 32 tons 18 cwt., of which 11 tons 9 cwt. was 
on the leading, 11 J tons on the driving, and 9 tons 18 cwt. on the 
trailing axle. The tender was supported on six wheels, 3ft. 9 fin. 
diameter, and had a tank capacity of 1,950 gallons. 

By a marvellous addition of a big head and a bigger tail (to say 
nothing of various legs), the diminutive body of the East Kent Rail- 
way had, in August, 1859, blossomed into the London, Chatham and 
Dover Railway; and for this railway 24 locomotives were supplied 
by various firms from Crampton's designs. They were numbered 
3 to 26. The design was peculiar a leading bogie having wheels 
3ft. 6in. diameter, and a base of 4ft., and four-coupled wheels 5ft. 6in. 
diameter. The cylinders were outside, and had a stroke of 22in., the 
diameter being 16in. As in the "London" and other Crampton 
engines, the cylinders were placed about mid-way between the smoke 
and fire-boxes, whilst the connecting-rods actuated the rear pair of 
coupled wheels, so that in describing the position of the wheels of 
these engines we should have to enumerate them as " leading bogie," 
" centre," and " driving." A compensation lever connected the centre 
and driving wheels. Gooch's valve gear was used. Like other engines 
of Crampton's design, this class was a failure, and within three or four 
years they were rebuilt as six-wheel engines, with inside cylinders and 
outside frames; some of them, as reconstructed without a bogie, are 
still in active service on the London, Chatham and Dover Railway. 

Before the grave faults inherent in the previously described clasa 
of engines had been fully appreciated, the London, Chatham, and 
Dover Railway had arranged for a second batch of engines from 
another of Crampton's designs. These consisted of five engines con- 
structed by R. Stephenson and Co. in 1862. The locomotives in 



204 EVOLUTION OF THE STEAM LOCOMOTIVE 

question were worked on the principle patented by W. Bridges Adams, 
and previously described in an earlier chapter viz., an intermediate 
driving shaft, coupled by outside rods to the driving wheels, situated 
behind the fire-box. The cylinders were 16in. diameter by 22in. 
stroke, and within the frames. The driving wheels were 6ft. 6^in. 
diameter, and bogie wheels 4ft. OJin. diameter. Cudworth's sloping 
fire-box, fitted with a longitudinal mid-feather, was employed. 
The heating surface amounted to 1,200 sq. ft., made up of 
130 sq. ft. fire-box and 1,070 sq. ft. tubes, which were 
2 in. diameter, 10ft. lOin. long, and 189 in number. The grate area 
was 26 sq. ft. 

The engines in question were named, etc., as follows: 

Company's No. Name. Builder's No. 



Echo" ... ... l?81 

Coquette" .. ... 1382 

Flirt" ... ... 1383 

Flora" .. ... 1384 

Sylph " ... 1385 



As remarked in describing the previous class, Crampton'u engines 
were in this case also found to be unsuitable, so that the London, 
Chatham and Dover Railway rebuilt the five engines, when the 
intermediate driving shaft was provided with a pair of wheels, and the 
engines became "four-coupled bogies." Tlie diameter of the cylin- 
ders was increased to I7in. ; the Cudworth fire-box was dispensed with, 
and the heating surface reduced, the present dimensions being fire- 
box, 100 sq. ft.; tubes, 987 sq. ft.; grate area, 16J sq. ft.; weight in 
working order: on bogie, 14 tons 12 cwt. ; driving wheels, 14 tons 
12 cwt. ; and on trailing wheels., 10 tons; total, 38 tons 16 cwt. 



CHAPTER XII. 

'Brougham," Stockton and Darlington Railway L. & N.W.R. engines at tho 
1862 Exhibition -Sinclair's "Single" engines for the G.E.R. French loco- 
motives on the G.E.R. L. & S.W.R. tank engines, afterward converted to 
tender engines Conner's 8ft. 2in. " Single " engine on the Caledonian Rail- 
way The liliputian "Tiny," the Crowe Works locomotive " Dignity and 
Impudence '' Bridges Adams's radial axle tank engines His spring tyrea 
Account of the St. Helens Railway locomotive with these innovations Broad- 
gauge engines for the Metropolitan Railway Rupture betwe?n the Great 
Western and Metropolitan Sturrock to the rescue G-.N. tender engines on 
t!,e Metropolitan Delivery of the Underground Company's own engines 
Great Northern "condensing" locomotives The Bissell bogie truck well ad- 
vertised End of the "hot brick" engine Sturrock's steam-tender engines 
on the G.N.R. Sinclair's tank engine with Biwell trucks Fell's system of 
locomotive traction Tried on the Cromford and High Peak line Adopted 
on the Mount Cenis Railway Spooner's locomotives for the Fesliniog Rail- 
way Fairlie's double bogie engines Tho "Welsh Pony" and " Litt'e 
Wonder ' --Fairlie's combined trains and engines Cudwortlvs trailing bogie 
North London engines, a model fcr tank locomotive construe- tors- Pryce's 
designs for the North London Railway. 

Fig. 82 illustrates the "Brougham," No. 160, of the Stockton and 
Darlington Railway. This engine was designed for hauling passenger 
trains. She was a bogie engine, as will be -noticed by reference to the 
illustration, and had four-coupled wheels 6ft. in diameter. The cylin- 
ders, placed outside, were 16in. in diameter, with a stroke of 24in. 
The tender was on six wheels, and the tank was capable of carrying 
1,400 gallons. No. 160 was constructed in 1860, not a very long time 
prior to the amalgamation with the North Eastern Railway Company, 
by R. Stephenson and Co., of Newcastle, at a cost of 2,500. 

The London and North Western Railway exhibited at the London 
International Exhibition of 1862 a locomotive constructed at Wol- 
verton from the designs of Mr. McConnell; the engine was built the 
previous year, was numbered 373, and named "Caithness." The 
cylinders were 18in. by 24in. ; driving wheels, 7ft. 7in. diameter; 
L. and T., 4ft. 7Jin. ; steam pressure, 1501b. ; wheel base, 18ft.; 
heating surface (14 tubes IJin. diameter, 9ft. 4in. long), 980.319 sq. 
ft. ; fire-box, 242.339 sq. ft. ; weight in working order (engine and 
tender) 59 tons 14 cwt. A combustion chamber 2ft. Sin. long was 
provided. Two other engines of this design were built, No 372 



206 



EVOLUTION OF THE STEAM LOCOMOTIVE 



<: Delamere" and No. 272 "Maberley." Apparently these engines 
were not very successful, as we do not find accounts of their later 
performances. 

In 1862 Fairbairn and Co. constructed for the Great Eastern 
Railway a class of " single " engines designed by Mr. R. Sinclair. 
These locomotives had outside cylinders, 16ft. by 24in. ; .driving 
wheels, 7ft. 3in., and leading and trailing wheels, 3ft. 9in. dia- 
meter; heating, surface, tubes (203, If in. diameter), 957.6 sq. ft.; 




FIG. 82. "BROUGHAM," No. 160, STOCKTON AND DARLINGTON RAILWAY 

fire-box, 94.9 sq.. ft.; grate area, 15.27 sq. ft.; weight, 32 tons, of 
which 13 tons 13 cwt. 1 qr. was on the driving axle. Gooch's link 
motion was employed. 

The design in question was of rather attractive appearance, the open 
splasher being an attractive feature, as was also the cab somewhat of 
an innovation 35 years ago. Mr. S. W. Johnson succeeded Mr. Sinclair 
at the end of 1865 as Great Eastern Railway locomotive superinten- 
dent, and under the regime of the former some of these engines were 
rebuilt with a leading bogie, and the diameter of the cylinders was 
increased to 18in. Another form of cab was introduced, the Salter 
safety valve on the dome was removed, and one of Ramsbottom design 
placed on the flush top fire-box, which had superseded the raised 
pattern as employed in this class of engine by Mr. Sinclair. One of 
the engines of this class (No. 0295) was in active service as recently as 
July, 1894. In connection with this class of engine a special circum- 



EVOLVTloX OF THE t TEA M LOCOMOTIVE 207 

stance needs mention viz., that 16 of these locomotives were made 
not " in Germany," but in the country of her foe ; the French engineer- 
ing firm with the German name of Schneider, in 1865, contracting 
to supply the 16 locomotives at a less price than any English maker. 
This event was certainly a curiosity in the economic history of this 
country's trade. We import many articles; let us hope, however, 
that foreign locomotives will not again be seen on English railways. 
There is some consolation to be found in the statement that all the 
British locomotive builders were so full of orders at the time that 
they practically refused to accept orders for the engines in question by 
tendering for them at outside prices, so that consequently the order 
had to be given to a foreign firm. 

In 1863 Beyer, Peacock and Co. commenced to construct a class 
of tank engines for the London and South Western Railway from 
the designs of Mr. J. Beattie. The locomotives in question had out- 
side cylinders 16 Jin. by 20in. stroke; four coupled wheels, 5ft. Tin. 
diameter; and a pair of leading wheels, 3ft. 7fin. diameter. The 
boiler contained 186 tubes, If in. diameter. The heating surface was 
made up of tubes 715.17 sq. ft., and fire-box 80 sq. ft. The grate 
area was 14.2 sq. ft. 

A lock-up safety valve was placed on the front ring of the boiler 
barrel, and two of Salter's pattern on the immense dome which sur- 
mounted the raised fire-box. The steam pressure was 1301b. The 
engine weighed in working order 29 tons 17 cwt., of which 10 J ton& 
was on the driving axle. We have already stated that the engines 
were built as tanks, but Mr. W. Adams, who had succeeded Mr. J. 
Beattie as locomotive superintendent of the London and South 
Western Railway, added tenders to some of their engines in 1883. It is 
p, common practice to rebuild tender engines as "tanks," but the 
opposite practice is somewhat of a novelty. The tenders were sup- 
ported on six wheels, 3ft. 9 fin. diameter, and weighed 20J tons in 
working order, the water capacity being 1,950 gallons. 

An engine that attracted considerable attention at the 1862 
Exhibition was one built by Neilson and Co. from the designs of Mr. B. 
Conner, locomotive superintendent of the Caledonian Railway (Fig. 83). 
The engine in question had outside cylinders, 17 Jin. diameter, with a 
stroke of 24in. ; driving 8ft. 2in. in diameter, with inside bearings and 
underhung springs. The trailing and leading wheels had outside 
bearings. The engine had 1,172 sq. ft. of heating surface: the- 



208 EVOLUTION OF THE STEAM LOCOMOTIVE 

grate area was 13.9 sq. ft.; wheel base, 15ft. Sin. ; weight, empty, 
27J- tons; in working order, 30 tons 13 cwt., of which 14 tons 11 cwt. 
was on the driving axle. 

Colburn describes the locomotive as a " fine, well-constructed 
engine, standing gracefully on its wheels, large, yet compact, and 
qualified to run at any speed with ease and steadiness," Nor can 
this description be in any measure contradicted. For, until Stirling 
built his famous 8ft. lin. "singles" for the Great Northern Railway, 
Conner's 8ft. 2 in. Caledonian engines were far and away the most 




FIG. 83. CONNER'S SFT. 2lN. "SINGLE" ENGINE, CALEDONIAN RAILWAY 
(REBUILT) 

graceful locomotives ever placed on the 4ft. 8Jin. gauge. In general 
design, the engine was a modification of the old Crewe pattern engine. 
The dome was, however, of rather a peculiar shape : it was placed on 
the top of the raised fire-box. The driving axle was of cast steel, 
and the tyres of Krupp steel. The large number of spokes in the 
driving wheels was noticeable, being at only lOin. centres at the rim 
of the wheels. The slide-valves, were provided with l^in. lap. A 
great improvement was the provision of a cab, and that of not dis- 
proportionate dimensions, considering the " year of grace " in which 
the engine was constructed. Trains of nine carriages were hauled at an 
average speed of 40 miles an' hour, with a coal consumption of 2|lb. 
per mile ; 1 4 loaded carriages were frequently taken up the terrible 
Beattock bank, 10 miles in length, at 30 miles an hour. 

The late Khedive of Egypt was so taken with the appearance of 
this engine when it was at the Exhibition that he immediately 
ordered one for his own railway. He was searching for a locomotive 
to convey him at 70 miles an hour, and Conner's 8ft. 2in. single 



EVOLUTION OF THE STEAM LOCOMOTIVE 209 

appeared to be the one most likely to fulfil his requirements. Nor do 
we hear that he was in any way disappointed with his purchase. 

It is interesting to know that the Caledonian Railway has still a 
specimen of this notable design unscrapped may it ever remain so. 
To prevent our appetite becoming vitiated with a, galaxy of Brob- 
dingnagian locomotives, we will descend to the other end of the scale, 
and detail the Liliputian " Tiny," as used in the Crewe locomotive 
works. The railway is of 18in. gauge, and was opened in May, 1862, 
for a length of three-eighths of a mile. In its course the engine 
traverses curves of 15ft. radius each, no difficulty being found in 
going round these curves with loads of 12 to 15 tons,' or in taking 
7ft. Gin. wheel fcrgings or tyres on edge by means of trucks specially 
adapted for the purpose. This engine has four wheels coupled; 
inside cylinders, 4 Jin. diameter, and Gin. stroke; the wheels are 15in. 
in diameter, on a base of 3ft. The total heating surface is about 
42 sq. ft. A No. 2 Giffard's injector supplied the boiler with water ; 
this precious liquid is stored in a saddle tank, with a capacity cf 
28 gallons. "Tiny," when "right and tight and ready for action," 
weighs only 2| tons. 

The duties of the Liliputian engines consist in hauling materials 
to and from different parts of the works, and as the 18in. rails are 
in most places laid parallel with the standard gauge line's, '"Tiny" 
is also called upon to fly shunt the trucks, etc., when necessary. 

An engine of this type, the "Nipper," forms with the giant 
" Cornwall " that well-known photographic picture the railway 
"Dignity and Impudence." 

Fig. 84 represents Sharp, Stewart, and Co.'s standard design of 
passenger engine of this period. The " Albion " was delivered to the 
Cambrian Railway in May, 1863. She was an inside cylinder engine, 
with a pair of leading wheels, and an enclosed Salter safety valve. 
Altogether, the " Albion " is a fair example of locomotive practice 38 
years ago. 

We have on previous occasions referred to the improvements in 
locomotive construction introduced by Mr. W. Bridges Adams, and 
we now have again to record a successful employment of his design. 
In the first week of November, 1863, Mr. James Cross, locomotive 
engineer of the St. Helens Railway, completed a tank locomotive, sup- 
ported on eight wheels, the leading and trailing pairs of which were 
fitted with the radial axle boxes patented by Mr. W. B. Adams; 

p 



210 



EVOLUTION OF THE STEAM LOCOMOTIVE 



whilst the four coupled wheels were fitted with spring tyres, which 
were another invention of the same engineer. 

The St. Helens Railway was famous or, from an, engineer's point 
of view, we should say, perhaps, infamous for the severe gradients, 
sharp curves, and numerous points, crossings, and junctions. The 
inclines were as steep as 1 in 35, 1 in 70, and 1 in 85, whilst the 
curves were constructed with radii of 300ft, and 500ft., and reverse or 




Fio. 84. " ALBION," CAMBRIAN RAILWAYS, 1863 

S curves were also more frequent than pleasant. The St. Helens 
Railway was only 30 miles long, but within two miles of the St. Helens 
Station no less than 12 miles of sidings were located. We do not 
mean to suggest that the whole line of railway was so thickly covered 
with siding connections, but such were distributed over the remaining 
mileage of the railway in too plentiful profusion. Here, then, was 
a length of railway containing the three great hindrances to smooth 
and quick running, but the locomotive about to be described was so 
constructed as to successfully overcome these impediments. 

This engine had inside cylinders, 15in. diameter and 20in. stroke. 
The coupled wheels were 5ft. lin. in diameter, the rigid wheel base 
being 8ft., but as these wheels had spring tyres, each pair of wheels 
was practically as free to traverse the curves as uncoupled wheels. 
Other dimensions were: Heating surface, 687 sq. ft.; grate area, 



EVOLUTION OF THE STEAM LOCOMOTIVE 211 

16.25 sq. ft.; total wheel base, 22ft.; weight in working order, on 
leading wheels, 7 tons 15 cwt. ; on driving, 111 tons; on rear coupled, 
11 J tons; on trailing, 10 tons, including 4J tons water and 1J tons 
coal. Total weight, 401 tons. 

The boiler contained 121 tubes, 10ft. llin. long, and IJin. dia- 
meter; steam pressure, 1401b. ; water capacity of tank, 950 gallons. 
The fire-grate was 5ft. long, and sloped from the door to the tube- 
plate. The springs of the coupled wheels were connected by means o 
a compensation lever. The dome was placed on the raised fire-box 
and fitted with a screw-down safety valve; a second valve of the 
same pattern was fixed on the boiler barrel. A roomy and well- 
enclosed cab, fitted with side windows, thoroughly protected the 
enginemen. 

Adams's radial axle-boxes are, of course, still in use on the Great 
Northern Railway, London, Chatham, and Dover Railway, and other 
lines, so that a detailed account here is not necessary, the salient 
feature being that they are made with a radius, having its centre in 
the centre of the adjoining axle, the axle-box guide-boxes being 
curved to fit. In the engine we are now describing the radius of the 
boxes was 7ft., and the lateral play of the boxes was 4jin. on each 
side. The spring-pins were not fixed on the top of the boxes, but 
were each fitted with a small roller to allow the boxes to freely 
traverse. The axle-boxes weighed 3J cwt. each. 

It will be understood that when an engine fitted with these boxes 
enters a right-hand curve the flanges of the leading wheels draw the 
boxes to the right, so that the engine itself remains a tangent to the 
curve, whilst, since the axle-boxes are themselves curved, the effect 
is that the right-hand side axles are brought nearer the rigid wheels, 
and consequently the radial wheels on the opposite side of the engine 
further from the fixed wheels, the whole effect of the radial axle-boxes 
being that the trailing and leading axles actually become radii of 
the curves being traversed, although the flanges continue parallel to 
the rails. 

Adams's spring tyres require a more precise description, and 
before we describe them, readers may perhaps be reminded that Adams 
had strong views on the subject of railway rolling-stock wheels. 
He enters rather fully into the matter in his book, " Roads and Rails," 
especially in the chapter dealing with "the mechanical causes of 

p 2 



212 EVOLUTION OF THE STEAM LOCOMOTIVE 

accidents." In this, Adams maintains that the usual forms of wheels 
are in reality rollers, and not wheels. 

Tho spring tyres had been tried on the North London 
Railway, Eastern Counties, and on another locomotive on the St. 
Helens Railway, before the engine now under review was constructed. 
Upon the coupled wheels of the new locomotive for the latter railway, 
double spring hoops were employed, the single form having been 
used in the three previously mentioned engines. The plan adopted 
was as follows : 

"The tyres chosen were constructed with a deep rib in front; this 
was bored out, internally, to a depth of Jin., and to a conical section, 
and, of course, parallel to the tread. A flat edge, fin. wide, was 
thus left on either side. 

" The springs, formed of tempered hoop steel, were placed on the 
inner surface of the tyres. Corresponding curves were turned across 
the outer circumference of the wheels. The wheels were forced into 
the cones containing the springs, and retained by three I'm. 
bolts, and a flat ring in the groove at the back of the tyre, 
the effort of the spring tyres being to allow of a slight lateral 
motion in running round curves and also to give a better grip of the 
rails, as the tyres, by reason of the weight upon them being trans- 
mitted through the tyre springs, slightly flattened upon the rails, and 
so presented a larger surface for adhesion between the tyres and rails." 

The following interesting account of the working of the radial 
axle and spring tyre locomotive on the St. Helens Railway is ex- 
tracted from a paper by Mr. J. Cross, the designer of the locomotive, 
and read before the Institution of Civil Engineers. Mr. Cross stated 
that "the engine was completed in the first week of November, 1863, 
and has since been running very regularly, taking its turn of duty 
with passenger trains or coal trains, or as a shunting engine; and 
about the numerous works connected by sharp curves with the St. 
Helens line. The motion round curves is free from all jerking, and 
on straight lines the speed is more than 60 miles an hour; either 
end of the engine being first, without any train behind to< give steadi- 
ness; and the motion is so smooth that it has only been by taking 
the actual time that the engineers have convinced themselves of the 
fact of the speed exceeding 40 miles an hour. It was built to traverse 
curves of 200ft. radius. This it does with the greatest facility, and 



EVOLUTION OF THE STEAM LOCOMOTIVE 213 

it has regularly worked the passenger trains round a curve of 1,000ft. 
radius, going directly off the straight line by a pair of facing points 
at a speed of more than 30 miles an hour, and it has gone round 
curves of 132ft. radius. It has also run a train of 12 passenger 
carriages, weighted up to 100 tons, exclushe of its own weight, at 
60 miles an hour on the level. From the advantages it possesses over 
the ordinary mixed engines for weighting the trailing coupled wheel, 
it, without difficulty, on a wet, slippery day, started, and took this 
load up a gradient of 1 in 70, drawing seven of the carriages with a 
load weighing 72 tons 5 cwt., up a gradient of 1 in 36, round a 
curve of 440ft. radius; and coal trains of 250 tons are worked over 
long gradients of 1 in 200 with the greatest ease. 

"It is evident, then, that engines on this principle, affording 
facilities for the use of high power in hilly countries, are peculiarly 
adapted for Metropolitan lines, where sharp curves are a necessity 
(being equally safe whichever end is foremost), and are also well 
suited for light lines in India and the Colonies. It may likewise be 
remarked that carriages and wagons on this principle would carry 
heavier freights, with a saving in the proportion of dead weight, while 
their friction round curves would be less than at present." 

The improvements adopted in the construction of this locomotive 
for the St. Helens Railway were so successful that, as usual, other 
claimants, who appropriated the radial axle-boxes as their invention, 
were soon contending with Adam's and Cross as to who was entitled to 
the honour of introducing the improvement. 

The first portion of the Metropolitan Railway was opened on 
January 18th, 1863, and the line was then worked on the broad-gauge 
by the Great Western Railway for a percentage of the receipts. The 
Great Western Railway provided the stations, staff, locomotives, and 
rolling stock. 

Mr. D. Gooch, in 1862, designed a special class of tank engines 
for working the Metropolitan Railway. They were six-wheel engines, 
the driving and trailing wheels being 6ft. diameter and coupled. The 
cylinders were outside. A special form of fire-box and baffle-plate 
was employed, and tanks were provided beneath the boiler barrel, 
into which the exhaust steam was discharged by means of a revers- 
ing valve fitted to the bottom of the blast pipe. When in the open 
air, the waste steam escaped up the chimney in the usual manner. 



214 EVOLUTION OF THE STEA M LOCOMOTIVE 

The first of these engines were named : Bee, Hornet, Locust, 
Gnat, Wasp, Mosquito, Bug, Khan, Kaiser, Mogul, Shah, and Czar. 
Later ones were named after flowers and Great Western Railway 
officers. 

A dispute arose between the two companies at the beginning of 
August, 1865, and immediately developed into a complete rupture. The 
smaller quasi vassal railway, through the energy displayed by its chief 
officers, successfully overcame the apparently insurmountable obsta- 
cles that beset it, and consequently the Metropolitan Railway asserted 
its complete independence of the Great Western Railway, and has 
since maintained it. 

It was indeed a nine days' wonder that the Metropolitan Railway 
was called upon to perform, for it had to obtain from somewhere 
locomotives and carriages to work the underground line, commencing 
on the morning of August 10th, 1863. 

Mr. Sturrock, the locomotive superintendent of the Great Northern 
Railway, had at this time under construction a class of condensing- 
tank engines that he had designed to work the Great Northern 
Railway traffic over the Metropolitan Railway. The directors of the 
Metropolitan Railway in this emergency applied to Mr. Sturrock for 
assistance, and by working day and night he managed to fit up some 
Great Northern tender engines with a temporary condensing apparatus. 

The difficulty was to provide some kind of condensing apparatus on 
the Great Northern tender engines, i being necessary to use flexible 
connecting pipes between the engine and tender strong enough to 
withstand the steam pressure, but Mr. Sturrock was successful enough 
to contrive the necessary flexible pipes by which the exhaust steam 
was conveyed from the engine to the water tank of the tender, but 
these pipes very frequently burst, and all concerned were far from 
sorry when the proper engines were delivered. 

An order for eighteen had already been placed with a well-known 
Manchester firm of locomotive builders by the Metropolitan Railway, 
Beyer, Peacock, and Co. building them from the designs of the late 
Mr. (afterwards Sir) John Fowler. 

The type is well known to London readers, the engines having 
side tanks, a leading bogie, the wheels of which were 3ft. diameter, 
with a base of 4ft. The driving and trailing wheels (coupled) were 
5ft. 9in. diameter, their base being 8ft. lOin. ; the total wheel bas 
being 20ft. Pin., or to centre of bogie, 18ft. 9in. The cylinders were 



EVOLUTION OF THE STEAM LOCOMOTIVE 215 

outside, slightly inclined from the horizontal, I7in. diameter, and 
24in. stroke. The grate area was 19 sq. ft. The fire-boxes had 
sloping grates, which were Gin. deeper at the front than the back. The 
boiler barrel was 4ft. in diameter, and 10ft. 3in. long; it contained 
166 tubes, 2in. diameter, the total heating surface being 1,014 sq. ft. 
The working pressure was nominally 1301b. per sq. in., but when 
working through the tunnels, condensing the steam, and with the 
dampers closed, a very much lower pressure resulted. The frames 
were inside, the dome (fitted with a Salter valve) was on the boiler 
barrel, close to the smoke-box, a sand-box being also fixed on the 
boiler barrel at the back of the dome. 

The bogie truck was built of plate frames, and was on the Bissell 
system, turning on a centre-pin fixed to the engine frame, at a radial 
distance of 6ft. 8in. from the centre of the truck. "Locomotive 
Engineering" says that "this radial length ensures a nearly correct 
radiality of the bogie to curves of all radii, the proper length of the 
radius to ensure exact radiality of the centre of the bogie for all curves 
being 7ft. 2in., or 6in. more than the actual length a difference 
which is, perhaps, of no great importance in practice." 

For the purpose of effectually condensing the exhaust steam the 
side tanks were only filled with water to within 6in. of the top, and the 
steam was discharged upon the surface of the water, from a 7in. pipe 
on each side one to each tank. Into the mouth of these 7in. pipes a 
4in. pipe was projected a short distance, and the other end of the 
4in. pipe was below the surface of the water, so that a portion of the 
steam was discharged right into the water in the tanks, and agitated 
the water sufficiently to prevent the surface of the water from becoming 
too hot, as would have been the case if the same portion of tha 
water had always been presented to the waste steam. The tanks 
held 1,000 gallons, and at the end of a journey the water had become 
too warm to properly condense the exhaust, and it therefore became 
necessary to quickly empty the tanks and to take in a fresh supply of 
cold water. 

To expeditiously perform the former operation, each tank was 
provided with a pipe 7in. in diameter; this led to a cast-iron valve- 
box being placed below the foot-plate. By means of a screw, worked from 
the foot-plate, a lOin. valve was operated, and the water in the tanks 
could be discharged into the pits below the engine in the course of 
some 60 seconds. 



216 



EVOLUTION OF THE STEAM LOCOMOTIVE. 



The following list gives the names and builders' numbers of the 
first locomotives constructed for the Metropolitan Railway : 



Engine No. 


Name. 


Builder's No. 


Engine No. 


Name. 


Builder's No. 


1 


Jupiter. 


412 


10 


Cerberus. 


421 


2 


Mars. 


413 


11 


Lutona. 


422 


3 


Juno. 


414 


12 


Cyclops. 


423 


4 


Mercury. 


415 


13 


Daphne. 


424 


5 


Appollo. 


416 


14 


Dido. 


425 


6 


Medusa. 


417 


15 


Aurora. 


426 


7 


Orion. 


418 


16 


Achilles. 


427 


8 


Pluto. 


419 


17 


Ixion. 


428 


9 


Minerva. 


420 


18 


Hercules. 


429 



These engines were fitted with a very small coal bunker, only 
18in. wide. Weight of engine in working order: on bogie, 11 tons 
3 cwt. ; driving, 15 tons 9J cwt. ; and trailing, 15 tons 10 cwt. 
Total weight, 42 tons 3 cwt. 

Mr. Sturrock's engines for working the Great Northern trains over 
the Metropolitan Railway were numbered 241 to 250, their leading 
dimensions being: Cylinders (inside), 16 Jin. diameter, 22in. stroke; 
leading and driving wheels (coupled), 5ft. Gin. ; trailing wheels, 4ft. 
diameter; wheel base, L. to D., 7ft. 6in. ; D. to T., lift. 9in. ; total, 
1 9ft. Sin. Weight, empty, 32 tons 4 cwt. 1 qr. ; in working order, 
39 tons 12 cwt, 2 qrs. 

These Great Northern Railway locomotives were fitted with 
Adams's radial axle-boxes to the trailing wheels, and commenced 
working at the end of October, 1865. 

The patentee of the Bissell bogie truck did not intend to hide the 
light of his invention under a bushel, for he advertised the improve- 
ment in a truly American style. The following advertisement was to 
be found in the columns of the sober railway newspapers soon after 
the Metropolitan locomotives were at work: 
" Important to Railway Directors, Engineers, and the Travelling 

public. 

"No more accidents from engines running off the line (see Queen's 
letter to Railway Directors copied in the railway papers 
January 28th, 1866). 

"The Bissell bogie, or safety truck, for locomotive engines, so much 
prized on American and foreign railroads for the great safety 
and economy it affords on curved roadways, after years of 
probationary trial in England, has at length been adopted by 



EVOLUTION OF THE STEAM LOCOMOTIVE 217 

John Fowler, Esq., C.E., F.G.S., upon all the^ new engines, 
eighteen in number, now working on the Metropolitan Railway, 
and by Robert Sinclair, Esq., C.E., upon twenty new eight- 
wheeled engines on the Great Eastern Railway, whicn may be- 
seen daily. The royalty for the use- of the Bissell Patents has- 
been reduced to 10 per engine, so that every engine requiring 
a bogie underframe should be provided with the Bissell safety 

truck. Apply to ." 

Whilst on the subject of railway advertisements we take the oppor- 
tunity to record the obituary announcement of the tentative " hot 
brick " engine, previously referred to, designed to work on the Metro- 
politan Railway. It appeared in the railway newspapers during the 
early months of 1865, and was to the following effect : "Metropolitan 
Railway. One locomotive engine for sale, either entire or in parts. 
For particulars apply to the Locomotive Superintendent, Bishop's-*- 
Road, Paddington." 

Reference must here be made to Mr. Sturrock's system of steam 
tenders, as adopted by him to work the heavy coal and goods trains, 
on the Great Northern Railway. In addition to the usual engine, the 
pistons of a pair of cylinders, 12in. diameter, with a stroke of 17iu. 
actuated the centre axle of the tender, and the six tender wheels were 
coupled by outside rods. The tender wheels were 4ft. Gin. diameter. 
The steam tenders weighed about 35 tons, with water and coal, and of 
this weight over 13 tons was on the driving wheels. After use in the 
tender cylinders, the exhaust steam was condensed in the tender tank. 
Forty-six of these steam tenders were constructed, and some are still 
running, but as simple tenders, the propelling apparatus having been 
done away with many years ago. Fig. 85 represents a Great Northern 
engine fitted with one of Sturrock's patent steam tenders- 

Mr. Robert Sinclair, whilst locomotive superintendent of the Great 
Eastern Railway, only designed one type of tank engine, and Neilson 
and Co. constructed the first of this class in 1864. Twenty of the class 
were built, being originally intended to work the Enfield Town Branch, 
but in later years-these engines were used on the North Woolwich line. 
The engines (Fig. 86) were supported by eight wheels, the leading and 
trailing being 3ft. Tin. diameter, and the driving and back coupled 
5ft. 6in. diameter. The cylinders were outside, 15in. diameter, and 
22in. stroke. The leading and trailing wheels were fitted with the 
Bissell truck, referred to in the advertisement just quoted. So that 



218 EVOLUTION OF THE STEAM LOCOMOTIVE 




EVOLUTION OF THE STE 1 M LOCOMOTIVE 19 

although the whole wheel base was 17ft. 4in., the rigid base that of 
the coupled wheels was only 6ft. The boiler was 13ft. Gin. long, and 
the water was carried in the tanks beneath the boiler and between the 
frames. An enclosed cab with front and rear spectacle plates was 
provided. 

This improvement so delighted the Great Eastern Railway drivers 
that they presented a testimonial to Mr. R. Sinclair in May, 1864, in 
which they described him as the "inventor" of the weather-board or 
" cab," as fitted to locomotives. The tank engines in question weighed 




FiO. 86. SINCLAIR'S DESIGN OF TAKE: ENGINE FOR THE EASTERN 
COUNTIES RAILWAY 

38 tons 6 cwt. 3 qrs., of which weight 20 tons 5 cwt. 2 qrs. was on the 
coupled wheels. 

In January, 1863, Mr. J. B. Fell patented a locomotive designed 
for working over extremely steep gradients. At that time there was a 
break 47 miles long in the continuity of the iron road communica- 
tion between France and Italy by the Mount Cenis route. This break 
has in later years been abolished by the construction and working of 
the famous Mount Cenis tunnel. Brassey and Co. in 1863 proposed 
that during the construction of the tunnel a temporary mountain rail- 
way worked on Fell's system should be built over the mountain. An 
experimental locomotive was, therefore, constructed at the Canada 
Works, Birkenhead. This engine weighed 14J tons loaded. The 
boiler was 2ft. 9in. diameter, and 7ft. 9J in. long, and contained 100 
tubes of IJin. external diameter. The heating surface was 420 sq. 
ft., and the grate area 6J sq. ft. The engine had two sets of machinery 
one for working the vertical wheels, acting on the ordinary carrying 
rails, and the other actuated the special horizontal clutch wheels. 



220 EVOLUTION OF THE STEA M LOCOMOTIVE 

which were pressed against the centre rail. The outside cylinders 
which worked the four-coupled vertical wheels, of 2ft. 3in. diameter, 
were 11 fin. diameter, the stroke being 18in. The horizontal coupled 
wheels were 16in. diameter, with a base of 19in. ; these were driven, by 
inside cylinders llin. diameter and lOin. stroke. A pressure of 12 
tons, actuated by means of a screw apparatus, could be applied to the 
horizontal wheels. 

By permission of the London and North Western Railway, an ex- 
perimental railway, 800 yards long, was laid down upon the Whalley 
Bridge Incline of the Cromford and High Peak Railway. 

The gauge was 3ft. 7f in., and there were 180 yards of straight line 
on a gradient of 1 in 13.5, and 150 yards of curves, with radii of 
2J and 3J chains, on a gradient of 1 in 12. The third rail upon this 
line, to be clipped between the horizontal driving wheels of the engine, 
was laid" on its side, 7 Jin. above the other rails. 

In the course of a series of experiments carried on from Septem- 
ber, 1863, to February, 1864, the engine, working up to a pressure of 
1201b. to the square inch, never failed, with a maximum load of 30 
tons, to take a load of 24 tons up the above inclines and round the 
curves. The outer cylinders working on the four vertical wheels 
could only draw up, besides the weight of the engine, a loaded wagon 
weighing seven tons; while the inside cylinders, acting upon the 
horizontal wheels, which pressed with 12 tons against the middle 
rail, enabled the engine to take up 24 tons on the same day and under 
the same conditions. The inside cylinders alone were able to carry 
up the engine itself, round the curves, and exhibited the power of 
taking up altogether 17 tons. 

The results of the experiments on the High Peak Railway were 
considered so satisfactory that the line up Mount Cenis was com- 
menced without delay. The engine was not properly adapted for 
working the mountain traffic, in consequence of the crowded and 
complicated nature of the machinery, and also because the feed-oil 
dropped on to the horizontal wheels and lessened the bite on the 
centre rail. The weight on the horizontal wheels was increased to 
16 tons., and an additional pair of guide wheels acting on the centre 
rail was provided at the trailing end of the engine, after the High 
Peak experiments. 

The Board of Trade was at that time so far interested in railway 
matters as to send out Captain Tyler, one of its inspectors, to report 



EVOLUTION OF THE STEAM LOCOMOTIVE 221 

on the Mount Cenis Railway. We extract from his report the follow- 
ing account of the working of this engine on the mountain railway : 

" In the course of two days I took six trips with this engine up and 
down the experimental line, carrying each time a load of 16 tons, in 
three wagons, including the weight of the wagons, and it performed 
in the ascent 1,800 metres in SJmin., with a loss of 141b. of steam ond 
of 5 l-3in. of water in the gauge glass, at steam pressure, varying 
between 92 and 1251b. to the square inch in the boiler, a,s the average 
of all those experiments. 

" The speed attained was in every case greater than that which it 
is proposed to run with the same load with the express trains ; and the 
average speed, as above given, was at the rate of 13 1-3 kilometres (or 
8 1-3 English miles) per hour, instead of 12 kilometres (or 7J English 
miles) per hour, which is the highest running speed allowed in the 
programme given to the French Government for this part of the line. 

"The weather was fine and calm, and the bearing rails were in 
firsi>rate order ; but the middle rail, as well as the horizontal wheels, 
were oily, and, therefore, in a condition very unfavourable for good 
adhesion." 

A second engine was built on Fell's system specially for working 
over the steep Mount Cenis Railway, and in its construction several 
improvements, suggested by the shortcomings of the first engine, were 
introduced. 

The second engine wa,s built partly of steel, and weighed 13 tons 
empty, and 16 tons 17 cwt. fully loaded, afterwards increased to 17 
tons 2 cwt. The boiler was 8ft. 4 Jin. long, and 3ft. 2in. in diameter, 
and contained 158 tubes of IJin. external diameter. Fire-box and 
tubes contained altogether 600 superficial feet of heating surface, and 
there were 10ft. of fire-grate area. There were only two cylinders, 
with a diameter of 15in. and stroke of 16in., which worked both tlw 
four-coupled horizontal and four-coupled vertical wheels, which were 
all 27in. in diameter. The wheel base of the vertical wheels was 
6ft. lOin., and that of the horizontal wheels, 2ft. 4in. The maximum 
pressure in the boiler was 1201b., and the effective pressure on the 
piston was 751b. to the square inch. 

Besides possessing a greater amount of boiler power, this engine 
travelled more steadily than No. 1, its machinery was more easily 
attended to, and the pressure upon its horizontal wheels could be 
regulated by the engine-driver at pleasure from the foo1>plate. This 



^ EVOLUTION OF THE STEAM LOCOMOTIVE 

pressure was applied through an iron rod connected by means 
of right and left-handed screws, with a beam on each side of the 
middle rail, and these beams acted upon volute springs which pressed 
the horizontal wheels against that rail. 

The pressure employed during the experiments was 2J tons on 
each horizontal wheel, or 10 tons altogether; but the pressure actually 
provided for, and which when necessary was employed, was 6 tons 
upon each, or 24 tons upon the four horizontal wheels. 

The vertical wheels were worked indirectly by piston-rods from 
the front, and the horizontal wheels directly by piston-rods from the 
back of the cylinders. 

Having already given Captain Tyler's account of his experiments 
with the first engine, we cannot do better than reproduce his state- 
ment concerning the second of the Fell engines, built for the Mount 
Cenis Railway. 

Captain Tyler stated that with the new engine he "was able to 
take up 1,800 metres of the experimental line with the same load as 
before, of 16 tons in three wagons, in 6J minutes, or at a speed of 
171 kilometres per hour, as against 12 kilometres per hour which it 
is proposed to run with the express trains. The steam pressure in the 
boiler fell from 1121b. to 102 Jib., and Sin. of water were lost in the 
gauge-glass, the feed having been turned on during the latter period 
only of this experiment. 

" The engine exerted in this instance, omitting the extra resistance 
from curves, about 177 horse-power; or, adding 10 per cent, for the 
resistance from curves, 195 horse-power, or more than 12 horse-power 
to each ton of its own weight, and nearly 60 horse-power in excess 
of what was required to take the same load up the same gradient and 
curves at 12 kilometres per hour, as proposed in the programme. I 
observed on the following day that 401b. of steam-pressure in the 
boiler, or one-third of the maximum pressure employed, was sufficient 
to move the engine alone up a gradient of 1 in 12 J; and the friction 
of carriages or wagons being proportionately much less than that of 
an engine, the same engine ought, a fortiori, to be able to move a 
gross load of three times its own weight, or 48 tons, at its greatest 
working pressure, up the same gradient." 

Having now given some details of locomotives constructed for 
working on a foreign steep grade railway, it will not be out of place 



TTTOLUTION OF THE STEA M LOCOMOTIVE 223 

to describe the special form<* of engines designed for the Welsh 
narrow-gauge line, usually called the Festiniog Railway. The line 
kas been open for a great number of years, but up to June, 1863, 
had only been used for conveying slates from the quarries to the 
shipping port. Horses were employed to haul the empty trucks up 
to the quarries, the loaded wagons running down to Portmadoc by 
gravity. . 

The average gradient for 12 J miles was 1 in 92, the steepest 1 in 
60. The radii of the curves ranged between two and four chains. 
Unlike the Mount Cenis line just reviewed, the Festiniog Railway 
waa worked with locomotives depending solely on the adhesion of the 
carrying wheels, no central rail being provided. The gauge was 
1ft. 11J in. 

The engines were designed by Mr. C. E. Spooner, the engineer of 
the railway. At first two were constructed, England and Co. being 
the builders. These miniature iron horses (one was more correctly 
called the "Welsh Pony") hstd two pairs of coupled wheels, with a 
wheel base of 5ft. The cylinders, which were outside the framing, 
were 8 Jin. in diameter, with a length of stroke of 12in., and they we*e 
only 6in. above the rails. 

The maximum working pressure of the steam was 2001b. to the 
square inch. Water was carried in tanks surrounding the. boilers, 
and coal in small four-wheel tenders. 

The heaviest of these engines weighed 7J tons in working order, 
and they cost 900 each. They could take up, at 10 miles an hour, 
about 50 tons, including the weight of the carriages and trucks, but 
exclusive of that of the engine and tender. They 'actually conveyed 
daily on the up journey an average of 50 tons of goods and 100 
passengers, besides parcels. Two hundred and sixty tons of slates 
were taken down to Portmadoc daily. The engines were well 
adapted for convenience in starting and in working at slow speeds, but 
their short wheel base and the weight overhanging the trailing wheels 
gave them more or less of a jumping motion when running. 

Safety guards, similar in form to snow ploughs, were afterwards 
added in front of the engines, behind the tenders, and under the 
platforms of the break-vans, in consequence of their being so near 
to the rails. 

After a few years' experience of these four-wheel loco- 
motives, the directors of the Festiniog Railway determined to experi- 



224 EVOLUTION OF THE STEAM LOCOMOTIVE 

ment with an engine constructed on Fairlie's double-bogie system, 
and the "Little Wonder" was constructed. In February, 1870, 
several trials were made with this engine, when a train of 72 wagoni, 
of a total length of 648ft., and of a gross weight, including the engine, 
of 206 tons 2 qrs.,, was drawn up an incline of 1 in 85 at a speed of 
five miles an hour, the steam pressure being 2001b. per square inch. 
The " Welsh Pony's " best performance in these trials upon the same 
gradient, but with a pressure of 1501b.,* consisted in drawing 26 
wagons, the gross load of which, with engine, amounted to 73 tons 
16 cwt. Tabulated, the result of these trials were as follows : 

Total Gravity. Fractional 

resistance. Ibs. resistance. 

Ibs. per ton per Ibs. 

gross. ton. per ton. 

"Little Wonder" 40 ... 26-3 ... 13-7 

" Welsh Tuny" wuh 1501 bs steam 51-4 ... 26-3 ... 25'1 

Do. 1301bs steam 44-5 ... 26-3 ... 18'2 

The general arrangements of the "Little Wonder" may be 
described as follows. The boiler was double, having two fire-boxes 
united back to back with two distinct barrels and sets of flue-tubes, 
and consequently a chimney at each end. A bogie was placed under 
each barrel, and each bogie had two pairs of wheels coupled together, 
worked independently by a pair of steam-cylinders to each bogie. 
Thus a total wheel base of 19ft. lin. in length was covered by the 
bogies; each bogie had a 5ft.-wheel base, and the distance between 
the centres of the bogies was 14ft. lin. The four cylinders were 
8 3-1 6th in. in diameter, and had a stroke of 13in. ; the wheels were 
2ft. 4in. in diameter. The combined grate area was 11 sq. ft., and the 
heating surface 730 sq. ft. Fairlie's system of double engines soon 
came into repute for working steep gradients, and many very powerful 
engines were and are still constructed on his system for use on foreign 
railways. Fairlie, in conjunction with Samuels,, adapted his system to 
# species of combined locomotive and carriage, and,, in 1869, one was 
constructed for working on the London, Chatham, and Dover Railway 
between Swanley Junction and Sevenoaks. Seven passenger com- 
ipartments were provided in this vehicle, accommodation comprising 
seats for 16 first-class and 50 second-class passengers; its total length 
was 43ft., and weight, empty, 13 J tons. The leading end was sup- 
ported by the engine bogie, and the trailing end by an ordinary 
bogie truck. Curves of only 50ft. radius were easily passed over by 
the combination vehicle. 



EVOLUTION OF THE STEAM LOCOMOTIVE 225 

Leaving Fairlie and his combinations, both of locomotives and 
carriages, and also of double locomotives, we now glance at a class of 
tank engines designed by Cudworth for working the trains between 
Cannon Street and Charing Cross upon the opening of the former 
terminus in 1866. These engines were seven in number, and were con- 
structed at the Canada Works. They were of the " coupled in front" 
pattern, with a trailing bogie. The cylinders were inside, 15in. dia- 
meter and 20in. stroke. The coupled wheels were 5ft. Gin. diameter. 
Outside frames were employed, and also compensation beams both to 
the coupled and bogie wheels. The coal bunker, with water-tank under, 
was of exceptional length. It was always a puzzle to the writer as 
to how a stout driver could manage to squeeze through the narrow 
entrances to the foot-plate, especially as these apertures were situate 
at the side of the fire-box ; but evidently the " trick was done " by follow- 
ing the axiom, " Where there's a will there's a way," and doubtless 
the drivers, if asked, would have replied, " It's very easy if you only 
know the way." These South Eastern Railway locomotives were 
numbered 235 to 241. 

Mr. Wm. Cowan, locomotive superintendent of the Great North 
of Scotland Railway, designed a class of engine, which Neilson arid 
Co. constructed. The design was stated to be that of a " goods'' 
locomotive, but upon examination we find the engines in question to 
be no other than the popular four-coupled behind, with a leading 
bogie and outside cylinders. The latter were arranged in a hori- 
zontal position immediately below the frames. The coupled wheels 
were 5ft. 6Jin. diameter, with underhung springs connected by 
means of an equalising lever-beam. The bogie wheels were 3ft. in 
diameter, with a base of 6ft. Inside bearings were supplied to the 
bogie axles. The boiler barrel measured 10ft. 10'Jin. between the tube- 
plates, its external diameter was 4ft. lin., and it contained 206 tubes 
of If in. diameter. The engine was fitted with D. K. Clarke's system 
of smoke consuming apparatus, previously described. The fire-box 
was of the raised pattern, and the steam dome was placed on it. The 
engine weighed 39 tons 13 cwt, and the tender 27 tons, in working 
order. 

In general appearance this "goods" engine resembled in a 
remarkable degree the London and South Western Railway express 
passenger engines as built by Mr. Adams. The tender was carried 
on six wheels. 

Q 



226 EVOLUTION OF TRu STEA M LOCOMOTIVE 

Fig*. 85 represents Beattie's standard design of goods engine for the 
L. and S.W.R. in 1866, the wheels were 5ft. 1 in. diameter, the cylinders 
being inside,, and having a diameter of I7in., the siroke 24in. Beyer, 
Peacock and Co. were the builders. Fig. 86 represents an engine of 
this class as rebuilt some years later at Nine Elms Works. 

In 1868. Mr. W. Adams placed upon the North London Railway 
the first locomotive constructed from a design which has, in its 
broad features and general outline, ever since been a model of sim- 
plicity, attractiveness, and utility, showing, as the design does, what 
engines constructed to work important local traffic should be like. 

In its original form there were some points that need alteration, 
as they certainly spcilt the general symmetrical effect of an other, vise 



r 




FIG. 85. BEATTIE'S STANDAKD GOODS ENGINE, L. & S.W.E., 1866 

artistic appearance. We may as well allude to these defects at once, 
and then proceed to detail the locomotive. 

The first of such blots on the design was the placing of a cylin- 
drical sand-box on the top of the boiler barrel, between the chimney 
fircl the dome. To show that such a position for this useful appendage 
was not necessary, we mention that only the driving wheels were sup- 
plied with sand from this unsightly excrescence, the supply of sand 
for the trailing wheels (for use when running bunker in front) being 
placed in an unobtrusive position. If the latter sand-boxes could thus 
be located, why was it necessary to place that for the leading wheels 
in so conspicuous a position? This example of awkward location of 
so useful an adjunct is further emphasised when we remember that 
these engines run just as frequently bunker first as chimney first. 



EVOLUTION OF THE STEA M LOCOMOTIVE 



227 



Further, in consequence of the position of this sand-box, the rod for 
working the sand valves was carried along the top of the boiler 
barrel, several inches above its surface, thus still more detracting 
from the symmetry of the design. The other feature \ve wish to 
allude to, is the shape of the dome cover, the whole of which was of 
ft needlessly ugly contour. Then, again, in later years an enclosed 
nib was added, the back and front of which, being of sheet-iron, ex- 
tending to the extreme of the coal bunker, and with no return sides, 
lias given a rather toy-like appearance to these otherwise fine loco- 
motives. We are glad to be able to mention that when these 
engines were rebuilt, the objectionable sand-box was removed, and a 
more pleasing form of steam dome provided, but this improvement was 




FIG. 86. BEATTIE'S GOODS ENGINE, L. & S.W.H., REBUILT 

in a great measure negatived by the black enamelled iron which is now 
used for the cover in place of the bright brass formerly employed for 
the purpose. 

Having thus mentioned the defects in appearance, rather than 
utility, of the North London Railway passenger tanks (Fig. 87), we 
can proceed to do justice to this really fine class of engines designed 
by Mr. Adams. 

The outside cylinders were I7in. diameter, and the stroke was 
24in. Tho driving and trailing wheels (coupled) were 5ft. 3i.n. 
diameter, the bogie wheels being 2ft. 9in. diameter. The heating 
surface was 1,015 sq. feet. The boiler was 4ft. lin. dia- 
meter, and contained 200 tubes of Ijin. diameter. A 
good feature in the design was the high steam pressure employed 
viz., 1601b. per sq. in. and there can be no doubt that much of 

Q 2 



228 



EVOLUTION OF THE STEA M LOCOMOTIVE 



the success of this class of engine can be traced to the use of so high 
a pressure of steam at a time 29 years ago when other lines 
were using a much lower pressure. Indeed, to-day it is only necessary 




to watch a North London and any of several other railway 
companies' trains starting side by side, and it will be observed that 
the North London generally gets away first; these engines are, iii 
fact, capital at starting, and soon attain a high rate of speed. 



EVOLUTION OF THE STEA M LOCOMOTIVE 
The weight was as follows : 



229 



On bogie wheels 
On driving wheel ? . . . 
On trailing wheels... 



Empty, 
tons cwt. 
15 14 
11 11 
11 7 

Total 38 12 



Loaded, 
tons cwt. 
14 14 
14 6 
14 12J 

43 12~ 




It will be observed that, when empty, the bogie axles supported 
19 J cwt. more of the gross weight than when the engine was in 
working order. 



230 EVOLUTION OF THE STEAM LOCOMOTIVE 

The wheel base of the bogie was 5ft. Sin. The coupled wheels 
have underhung springs connected by a compensation beam. India- 
rubber springs are used in connection with the ' hanging of the 
springs, and also to guide the bogie, etc., and it was found that such 
springs answered the use to which they were put in a most admirable 
manner. 

In. all the new engines that have lately been built, and when re- 
building old engines of this type, the cylinders have been increased 
to 17 Jin. diameter, and other things considerably modified in detail. 

Mr. Pryce has also built 24 powerful six-wheel tank engines (Fig. 
88) for dealing with the N.L.R. goods traffic. These engines are very 




FIG. 89. LOCOMOTIVE AND TRAVELLING CRANE, N.L.R. 

efficient. They have outside cylinders I7in. diameter, 24in. stroke, 
and 4ft. 4in. coupled 'wheels. Boiler pressure, IGOlb. per sq. in. 
Weight in working order, 45 tons 9 cwt., all available for adhesion. 
The total wheel base is lift. 4in. ; consequently, they take curves 
easily. 

The coal consumption of these engines was very satisfactory. 
The trains of the North London Railway consist of twelve vehicles., 
weighing, empty, 90 tons 14 cwt., and loaded 112 tons 6 cwt., but 
the coal consumption, with very frequent stoppages, only averaged 
30.281b. per mile. 

Fig. 89 represents the combined saddle-tank locomotive and crane 
belonging to the North London Railway, as .recently rebuilt by 
Mr. Pryce. 



CHAPTER XIII. 

Beattie's express engines Kendall 'a three -cylinder engine for the Blythe and 
Tyne Railway Heavy engines for the Metropolitan and St. John's Wood 
Railway Sold to the Taff Yale Railway "The most powerful locomotive in 
the world " for sale" Jinks's Babies "The " Areo-steam " locomotive on the 
Lancashire and Yorkshire Railway- Tank engines on the Furness Railway- 
Patrick Stirling's world famous " 8ft. singles " for the G.W.R. Webb's " Prece 
dents" forihoL. and N.W.K. The "John Ramsbjttom" " The Firefly," r.n 
engine that has "played many parts " J. Stirling's 7ft. coupled engines on the 
G. and S.W.R. Stirling's reversing apparatus Watkin's express engines for 
ohe S.E.R. Stroudloy's " Grosvenor," L.B. and S.C.R. The era of "com- 
pounds" W. F. Webb's first compound locomotive Boweu-Cooke's views on 
the subject The 'Experiment'"' 7ft. lin. compounds "Queen Empress" 
" Black Prince " Wordsell compounds -Midland coupled expresses Stroud- 
ley's "Gladstone" class- -Tlio ''General Maiiagers" on the North Extern 
N.B.R. locomotive, "No. 592 " Holmes' s "633" class Great Eastern 7ft. 
coupled Holden's liquid fuel locomotives Serve tubes in locomotives Caere's 
7ft. bin. " Singles." 

1 

Fig. 90 represents the "Python," one of J. Beattie's four-coupled 
express engines, constructed for the L. and S.W.R. The cylinders were 
outside, I7in. diameter by 22in. stroke. The coupled wheels were 7ft. 
lin. diameter, and the leading wheels 4ft. diameter. The heating surface 
was 1,102 sq. ft. Weight of engine in working order, 35 tons 11 cwt. 
For some years this class of engine was the favourite express engine 
on the L. and S.W.R. 

Locomotive engineers have always one great difficulty to provide 
f or viz., the extra power required to start locomotives, especially on 
steep inclines, and as such grades are particularly en evidence on 
the mineral lines, it is not surprising to find Mr. W. Kendall, of 
Percy Main, Northumberland, patenting a locomotive designed to 
overcome the defects just indicated. The patent is dated October 
26th, 1867. The engine was of the three cylinder type, with one 
inside and two outside cylinders. When running on a level road 
only the inside cylinder was used, but for starting or ascending 
inclines the power of all three was brought into use, the whole 
arrangement of the power being actuated by the reversing gear 
apparatus. By a peculiar adaptation of the lap of the valves, a 
small quantity of steam was admitted to the valves of the outside 



232 EVOLUTION OF THE STEAM LOCOMOTIVE 

cylinders when these cylinders were not working, for the purpose of 
lubrication. The engine in question was built at the Percy Main 
Works of the Blythe and Tyne Railway. She was of the " four-coupled 
behind " type, with a single pair of leading wheels. The inside 
cylinder was connected in the usual manner to the cranked axle of 
the centre wheels, the outside cylinders actuating the trailing pairs 
of wheels. Without diagrams it is rather difficult to explain the 
method employed to prevent the pistons, etc., of the outside cylinders 
from reciprocating, but shortly it may be stated that the connecting- 
rod was divided into two pieces, and at the joint each end fitted into 
an enclosed link. When disconnected, that portion of the rod 
coupled to the wheels which was in the link merely travelled up and 
down the link, whilst the part connected with the piston, etc., was 




FIG. 90. "PYTHON," A ?PT. llN. COUPLED EXPRESS ENGINE, 
LONDON AND SOUTH WESTERN RAILWAY 

at rest. By means of a screw gear this latter portion of the con- 
necting-rod was lowered in the link, and engaged with the other part 
of the rod, which was coupled to the wheels, and so the outside 
cylinders were brought into action. If required, the outside cylin- 
ders could be used independently of the one inside cylinder, so that 
the engine could be a one, two, or three cylinder locomotive. Sepa- 
rate regulators were provided for the inside and outside cylinders, but 
the handles were coupled together, so that, if required, one move- 
ment actuated the admission of steam to all the cylinders. To pre- 
vent too strong a blast, the driver could, by the operation of a ball 
valve, discharge the exhaust steam from the outside cylinders into 
the atmosphere by means of a pipe in front of the engine. On the 



ION OF THE STEAM LOCOMOTIVE ' 233 

other hand, the whole of the exhaust from the three cylinders could 
be discharged up the chimney in the usual manner if preferred. 

Upon April 13th, 1868, the Metropolitan and St. John's Wood 
Railway was opened for traffic. The line branches from the Metro- 
politan Railway at Baker Street, and was worked by the Metropolitan 
Company. The gradients on the short line are very severe, and it 
was not considered advisable to attempt to work the railway by the 
usual type of engine employed on the underground line ; so Mi- 
Burnett, the then locomotive superintendent of the Metropolitan 
Railway, designed a special class of engine for the St. John's Wood 
Railway. These were constructed by the Worcester Engine Com- 
pany, and were numbered 34 to 38. They were provided with six 
coupled wheels of 4ft. diameter, with outside bearings; the cylinders 
were 20in. diameter, with a 24in. stroke ; they were placed within the 
frames at 2ft. 2 in. centres. The wheel base of these powerful loco- 
motives wa f j divided as follows: L. to D., 6ft. lOin. ; D. to T., 
7ft. 2in. The boiler was lift, long, and 4ft. 3in. diameter, r.nd 
contained 176 tubes of 2in. diameter. 

The fire-boxes were .exceptionally large, the measurements being : 
Length, outside 7ft. lin., inside 6ft. 6in. ; width, outside 4ft., inside 
3ft. 6in. The depth was 5ft. 5in. in front, sloping to 3ft. llin. at 
back. The steam pressure was 1401b. ; heating surface, 1,165 GO. 
ft. ; grate area, 22J sq. ft. The water capacity of the tanks was 
1,000 gallons. 

These mammoth engines weighed 46 tons in working order, and 
it was soon discovered tliat they were far too powerful for working 
the light traffic over the St John's Wood line, the ordinary type of 
Metropolitan locomotives being quite capable of successfully work- 
ing the trains over these inclines. So, in 1873, when the Taff Vale 
Railway was in urgent need of some powerful engines for hauling the 
heavy coal trains over the Penarth Dock lines, the Metropolitan Rail- 
way succeeded in disposing of these five engines to the South Wales 
Company, and they can still be seen employed on work more adapted 
to their construction than was that of hauling ligjit" passenger trains 
on the St. John's Wood Railway. 

It is evident that both the patentee and builders of the " double 
bogie" locomotives had a very exalted opinion of the capabilities of 
these peculiar engines. In December, 1870, G. England and Co. 



234 EVOLUTION OF THE STEA M LOCOMOTIVE 

were- advertising for sale by private tender to the best bidder " the 
most powerful locomotive at present known upon any railway in the 
United Kingdom, irrespective of gauge." 

This "most powerful" locomotive was constructed for the 4ft. 
8 Jin. gauge on Fairlie's double bogie system. She had four cylin- 
ders, 15in. diameter and 22in. stroke, eight wheels, all drivers of 
4ft. 6in. diameter, and with steel tyres. 

Amongst other useful features claimed for this "most powerful" 
locomotive, we read that she "would take a load up an incline at 
a speed exceeding that of any other engine at present known, and 
would round the sharpest curves with ease." 

"Jinks's Babies" consisted of a batch of ten engines constructed 
towards the end of 1871, and early in 1872. They had outside 
cylinders, l7in. diameter and 30in. stroke, with a leading bogie and 
four coupled wheels of 7ft. diameter; they had, perhaps, as good a 
right to the title "most powerful" as the Fairlie engine just men- 
tioned. 

Be this as it may, however, " Jinks's Babies " were not successful. 
They were built at the Stockton and Darlington Locomotive Works, 
at Darlington, and originally numbered 238 to 240, etc., and upon 
the consolidation of the North Eastern Railway were renumbered 
1238 to 1240, etc. They were rebuilt by Mr. Fletcher as six-wheel 
engines, the bogie giving place to a single pair of leading wheels, and 
the stroke of the pistons was reduced from 30in. to 26in. Even after this 
metamorphosis, " Jinks's Babies " could not be truthfully described as 
successful locomotives. Amongst other peculiarities the circular 
valves should be enumerated. The steam pressure was 1401b. per 
sq. in. 

In 1871 the Lancashire and Yorkshire Railway fitted up an 
engine with an apparatus said to have been invented by Mr. Richard 
Eaton, but called "Warsop's Areo-Steam system," by means of 
which a continuous supply of heated air was forced into the 
bottom, of the boiler, so causing the water to be continually agitated, 
and. thereby preventing incrustation of the metal, as well as more 
quickly generating steam, and last but far from least economising 
the fuel. The engine experimented upon was a six-coupled goods, 
No. 369, with cylinders 15in. by 24in., 5ft. wheels, and working at a 
pressure of 1301b. per sq. in. An air pump, single acting, Gin. diar 
meter by 2ft. stroke, with piston and metallic rings, driven from ono 



EVOLUTION OF THE STEAM LOCOMOTIVE 



235 



of the main cross-heads, was secured to the framework of the engine 
in the place originally occupied by the feed pump. The compressed 
air passed along a pipe IJin. in diameter, 6ft. long, to a coil of l^in. 
lap-welded iron pipe, within the smoke^box, 61ft. in length, so 
arranged as to avoid contact with the blast pipe or the- ashes 
deposited in the smoke-box by the action of the blast. After tra- 
versing the coil, the expanded air became heated to a temperature 
nearly as high as that of the waste gases, and thus ranging between 
500 degrees and 800 degrees, or 850 degrees Fahr., lifted the self- 
acting valve, and entered the perforated distributing pipe within 
the boiler, and was constantly passing in jets through the water to ihe 
steam space, whence the combined powers of steam, and air proceeded 
to the cylinders to carry out their duty. A very simple apparatus was 
used when desirable to stop compression, by keeping the inlet valve 
open when steam was shut off; otherwise an undue proportion of air 
would enter the boiler, and impede the feed-water injectors. 

At the same time, occasions arose where a judicious use of the 
air injection was made with great advantage, even with steam shut 
off. It is stated that "on March 21st, 1872, there wa^ 
a heavy fall of snow, and the driver of No. 369 had to 
make the most of his resources. In coming down Rain- 
ford bank he had but lOOlb. of steam at Balcarres siding, 
with steam shut off. He allowed the air pump to continue work, 
and in 400 yards his gauge rose to 140, when he opened his regulator 
again to mount the incline with his heavy load, and so successfully 
gained the summit." The annexed table shows the working of engine 
No. 369, with and without the apparatus, and also of an exactly 
similar engine, No. 38, employed on the, same length of line, and 
hauling the same trains. No. 38 was not fitted with the apparatus. 



Engine. 


Miles run. 


Coal 
consumed. 


Average IDS. 
per mile. 


369 (without apparatus) ... 
369 (with apparatus) 
88 (without apparatus) 


21,948 
27,934 
28,053 


Tons. cwt. 
403 6 
472 10 
550 10 


42-92 

87-89 - 
43-95 



Although the above glowing statement is made about this inven- 
tion, which was fitted to no less than six engines, and tried for a 
period of about five years, it was not found to be commercially success- 



236 EVOLUTION OF THE STEA M LOCOMOTIVE 

ful, the power consumed in working the pump, and the cost of repairs 
running away with the economy supposed to have been gained in the 
original experiment. 

About this time the Furness Railway introduced a powerful 
design of six-coupled tank engines. The cylinders were : Inside, 
18in. diameter and 24in. stroke; heating surface tubes, 1,048 sq. ft; 
fire-box, 96 sq. ft.; grate area, 15 sq. ft. The frames were "inside." 
The side-tanks were capable of containing 1,000 gallons of water. 

Weight in working order : L., 13 tons 13 cwt. ; D., 16 tons 6 cwt. ; 
T., 14 tons 15 cwt. ; total, 44 tons 14 cwt. On the level this class of 
engine hauled 372 tons at 20 miles an hour, and up an incline of 1 in 
80 a load of 367 tons was drawn at llf miles an hour. The steam 
pressure was 1451b., and the coal consumption 40.161b. per mile. 

The name of Patrick Stirling, the late locomotive superintendent of 
the Great Northern Railway, will long be remembered and held in 
high honour amongst those of his confreres, consequent upon his suc- 
cessful design of 1870, in which year he built the first of his now 
world-famous 8ft. lin. singles, a type of locomotive which immedi- 
ately leaped into public favour, which for elegance and simplicity 
of design it is not saying too .much in stating that no modern engine 
has surpassed or is likely to surpass. These engines soon showed the 
travelling public that really express speed could be safely indulged in 
for continuous runs of great length without fear of accident or failure. 
Indeed, modern express speed can date its foundation from the intro- 
duction of these engines. The Great Northern Railway undoubtedly 
owes its popularity and fame as the " express " route to the successful 
running of Patrick Stirling's 8ft, lin. outside cylinder " single " 
engines. 

The following may be accepted as a correct description of the 
earlier type of this locomotive design. Later engines of the same 
class have, in common with the development of locomotive design, 
increased in weight, grate, tube, and cylinder area, and steam pres- 
sure ; but the general outline to-day, as seen in Fig. 91, 
is -the same as that of 27 years ago, and we do not 
think this compliment can be paid to the design of any 
other locomotive built at the present time. The cylinders were 
18in. diameter, with a length of stroke of 28in. The small ends of 
the connecting-rods were furnished with solid bushes of gun metal. 
The inner and the outer fire-boxes were connected together by stays, 



EVOLUTION OF THE STEA M LOCOMOTIVE 237 

screwed into each of the plates, without the intervention of iron 
girder bars. By this arrangement, which had been in use for some 
time in Belgium, the large amount of deposit generally existing upo?i 
girder-boxes was prevented, the facility for cleansing was much 
greater, and the liability of the tube holes in the copper-plate to 
become oval had been got rid of. 

The heating surface in this engine was, in the tubes, 1,043, and 
in the fire-box 122 sq. ft. The fire-grate had an area of 17.6 sq. ft. 




FIG 91. 8FT. llN. "SINGLE" EXPRESS ENGINE, GREAT NORTHERN 
RAILWAY 

When the engine was in working order, the weights upon the 
driving, trailing, and bogie wheels were 15, 8, and 15 tons respec- 
tively. The distance from the centre of the trailing wheels to tha 
centre of the bogie pin was 19ft. Sin. These engines were capable of 
drawing a weight of 356 tons on a level at a speed of 45 miles an 
hour, with a working pressure of 1401b. to the sq. in. The con- 
sumption of coal, with trains averaging sixteen carriages of 10 tons 
weight each, had been 271b. per mile, including getting up steam and 
piloting. The cost of maintaining and renewing passenger engines 
on the Great Northern Railway was in 1873 estimated to amount to 
2Jd. per mile. 

The contemporary type of engines on the ' West Coast" 



238 



EVOLUTION OF THE STEAM LOCOMOTIVE 



route was the celebrated "Precedent" class, illustrated by "Joliii 
Ramsbottom" (Fig. 92). 

These London and North Western Railway locomotives were con- 
structed at the Crewe Works, from the designs of Mr. F. W. Webb, 




locomotive superintendent, the first of them being constructed in De- 
cember, 1874. The engines, as our readers well know, have four- 
coupled wheels, 6ft. Gin. diameter, and a leading pair of wheels 3ft. 
9in. diameter. The principal dimensions originally were they may 



EVOLUTION OF THE STEAM LOCOMnri \ '/; 



239 



vary <, little in some details in certain engines inside cylinders, 17in. 
diameter, with a stroke of 24in. Heating surafce, 980 sq. ft. in tubes, 
and 103.5 sq. ft. in fire-box; grate area, 17.1 sq. ft.; weight in work- 
ing order, L., 10 tons 5 cwt. ; D., 11 tons 10 cwt. ; T., 11 tons; total, 
">2 tons 15 cwt. Steam pressure, 1201b. ; wheel base, 15ft. Sin. 

The most famous engine of this class is the " Charles Dickens," 
No. 955, built at Crewe in 1882; the "Inimitable" is shedded at 
Manchester, and the daily journey to and from Euston consists uf 
366|- miles; the train's worked by this engine are the 8.30 a.m. up, 
and the 4 p.m. down. As long ago as September 21st, 1891, the 
" Charles Dickens " had obtained the premier position in 
enjrine mileage. On that dav '"'she" if the shades of 



FIG. 93 




CTSIDE CYLINDER TANK ENGINE 



" Boz " will allow the bull completed her millionth mile, 
consisting of 2,651 trips between Manchester and London, 
in addition to 92 other journeys. During this period of 9 yeai*s 
219 days the engine had burned 12,515 tons of coal. Up to the 
end of February, 1893, the total mileage of "Charles Dickens" 
amounted, to 1,138,557, and up to the present time it has exceeded 
the enormous total of 1,600,000 miles! 

In April, 1874, Mr. Webb introduced another type of locomotive for 
the London and North Western Railway: the " Precursor,'' No. 2145, 
gives its title to the design in question. 

The cylinders were I7in. by 24in. stroke. The leading wheels were 
3ft. 6in. diameter, whilst the driving and trailing wheels (coupled) 
were 5ft, 6in. diameter. The tubes contributed 980 sq. ft., and the 



240 



ti SOLUTION OF THE STEA M LOCOMOTIVE 



fire-box 94.6 sq. ft., of the heating surface. The weight in working 
order was 31 tons 8 cwt. 

" Firefly " (Fig. 93) is one of the numerous six-wheel outside cylinder 
tank engines built from the designs of J. Beattie by Beyer, Peacock and 
Co. for the L. and S.W.R. between 1863. and 1875. The cylinders 
were 15 Jin. by 20in. stroke, the leading wheels 3ft. 7fin. diameter, 
and the coupled wheels 5ft. 7in. diameter. The heating surface was 
795.17 sq. ft.; the weight, in working order, 34 tons 12 cwt. A 
number of these engines had the cylinder diameter increased to 16 Jin, 
and a tender added by W. Adams in 1883. "Firefly" was built in 
1871. 




FIG. 94,," KENSINGTON," A 4-COUPLED PASSENGER ENGINE, L.B. k S.CE. 

"Kensington" (Fig. 94), a L., B. and S.C. locomotive, was, in 
December, 1872, rebuilt by Mr. Stroudley in the form illustrated. The 
cylinders were I7in. by 24in. stroke; coupled wheels, 6ft. 6in. dia- 
meter; leading wheels, 4ft. 3in. diameter. In 1872, "Kensington" 
was domeless, that appendage being added later. This engine was ori- 
ginally a single engine, built by R. Stephenson and Co. in 1864. 
Altogether, this engine, like many individuals, has " played many parts-" 

We have now reached a period in locomotive history when the 
engines to be described are of comparatively modern construction, a 
very large proportion of them being still in work on the various 
lines of railway, and readers interested in such matters are probably 
acquainted with the particulars of the locomotives. Under such cir- 
cumstances, a, detailed and particular account of each design would 
be rather wearisome, therefore the general features of modern engine:* 



EVOLUTION OF THE KTEA M LOCOMOTIVE 241 

will l)e less fully described. At the same time any uncommon points 
in their design or construction will be mentioned. 

The standard type of express passenger engines now used on the 
South Eastern Railway has developed from a class introduced by Mr. 
J. Stirling, when locomotive superintendent of the Glasgow and 
South Western Railway. 

In 1873 he constructed at the Kilmarnock Works an engine with 
a leading bogie and four coupled wheels of 7ft. diameter. The cylin- 
ders were inside, 18in. diameter and 26in. stroke. In this design, as 
in the later type on the South Eastern Railway, the boiler was un- 
provided with a dome, but in the latter the duplex safety valve is 
placed about the centre of the boiler barrel, whilst on the Glasgow 
and South Western Railway engines it surmounts a flush-top fire-box. 

Mr. Stirling's reversing apparatus is a very useful contrivance; it 
enables the driver to reverse his engine without the expenditure of 
any muscular power. At first the new reversing gear was frequently 
mistaken for the Westinghouse air-brake pumps. It consists of two 
vertical cylinders placed tandem fashion at the side of the boiler barrel. 
One piston-rod passes through both cylinders, and the pistons are 
attached to it ; this rod is connected with the reversing apparatus. t 
One cylinder contains steam, the other oil. The duty of the latter is 
to prevent the movement of the piston or rod. It will be understood 
that, since the cylinder is quite full of oil, it is impossible for the 
piston and connections to move unless the oil can pass from one sid j 
of the piston to the other. 

This is accomplished by a handle, which also actuates the \alve 
of the steam reversing cylinder so that when the steam is admitted 
into one cylinder to move the piston, the oil is at the same time 
permitted to flow through a valve to the other piston, and the revers- 
ing apparatus is worked. 

The oil keeps the piston in any desired position. As soon as the 
oil cannot pass from one side of the piston face to the other, the 
gear is firmly locked. 

Mr. A. M. Watkin became locomotive superintendent of the South 
Eastern Railway in 1876, and he introduced a very pretty desigii of 
express passenger engines. Twenty engines of the type were con- 
structed : Nos. 259 to 268 by Sharp Stewart, and Co., and Nos. 260 
to 278 by the Avonside Engine Company. Inside frames were pro- 
vided ; the leading wheels were 4ft. and the four-coupled wheels 



242 EVOLUTION OF THE STEA M LOCOMOTIVE 

6ft, Gin. diameter. The cylinders were I7in. diameter and 2 tin. 
stroke. The weight in working order was 34 J tons; the total heating 
surface, 1,103J sq. ft. The splashers to the coupled wheels were of 
open-work design. The chimney was of the rimless South Eastern 
pattern ; a dome was provided on the centre of the boiler barrel, and 
a duplex safety-valve on the fire-box top. A cab very much resembling 
the standard London and North Western Railway pattern was fitted 
to the engines. 

Several of these engines, as rebuilt by Mr. Stirling, remain in 
work at the present time; they are principally employed on the Mid- 
Kent services. 

In 1874 Mr. Stroudley, the then locomotive superintendent of 
the London, Brighton and South Coast Railway, built the " Gros- 
venor" with 6ft. 9in. single driving wheels, inside cylinders. 1 Tin. 
by 24in., and a total heating surface of 1,132 sq. ft. The "Stroud- 
ley" speed indicator was fitted to this engine. On August 13th, 
1875, the "Grosvenor" conveyed a train from Victoria to Ports- 
mouth (87 miles) without a stop. This was the first occasion on which 
such a trip had been performed; the time taken was 110 minutes. 

No other engine exactly similar to the "Grosvenor" was con- 
structed; but in 1877 the " Abergavenny " with. 6ft. Gin. single 
drivers and cylinders 16in. by 22in. was built, and in 1880 the first 
of the " G " class of singles was turned out at the Brighton Work.*. 
These also have single driving wheels 6ft. 6in. in diameter, but the 
cylinders are l7in. diameter, the stroke being 24in. The weight on 
the driving wheel is 13 tons. 

An interesting era in the evolution of the steam locomotive is at 
this point arrived at viz., the first really practical trial of compound 
engines, or the use of steam twice over for the purpose of propelling 
a locomotive. 

To Mr. Webb, the chief mechanical engineer of the London and 
North Western Railway, is due the honour of introducing the com- 
pound system on an extended scale in railway practice. Although 
21 years have now passed since the premier attempt of giving the 
system a fair trial on an English railway was made, it does not seem 
to have gained much favour with English locomotive engineers. In- 
deed, at the present time, excepting a few minor trials elsewhere, tha 
London and North Western Railway is the only company that con- 
structs and uses compound locomotives. 



EVOLUTION OF THE STEAM LOCOMOTIVE 243 

Mr. Webb employs the three-cylinder type of engine, which is an 
adaptation of the system introduced by M. Mallet on the Bayonne 
and Biarritz Railway. Three engines were built from Mr. Mallet's 
design by Schneider and Co., Creusot, and were brought into use in 
July. 1876. In these locomotives Mallet employed two outside 
cylinders, one being 15 Jin. and the other 9^in. diameter. 

Mr. Webb uses three cylinders : an inside cylinder for the 1 p. 
steam and two outside cylinders for the high-pressure steam. But at 
first one of Trevithick's old " single " engines was fitted up on Mallet's 
two-cylinder plan. This was in 1878. The engine worked success- 
fully for five years on the Ashley and Nuneaton branch of the London 
and North Western Railway, and thereupon Mr. Webb decided to con- 
struct compound engines on his three-cylinder system. 

The first of such engines was the "Experiment." Her outside 
h.p. cylinders were lljin. diameter, the. inside l.p. being 26in. dia- 
meter. Joy's celebrated valve gear was employed to regulate the 
admission of steam to the cylinders. 

" Webb " compounds have two pairs of driving wheels, but these 
are uncoupled, so that practically the engines are " singles." Whether 
the four driving wheels work well together,, or whether, on the other 
hand, there exists a considerable amount of either slip or skidding is 
another matter. The trailing pair of wheels is driven from the h.p. 
cylinders, and the middle pair from the inside or l.p. cylinder. 

Mr. Bo\ven-Cooke,~an authority on London and North Western 
\\ a ihvay locomotive practice, s\ims-up the advantages of the "Webb" 
compound system under the five following heads : 

1. Greater power. 

2. Economy in the consumption of fuel. 

3. The whole of the available power of the steam used. 

4. A niore even distribution of the strains upon the working parts, 
and larger bearing surfaces for the axles. 

5. The same freedom of running as with a single engine, with 
the same adhesion to the rails as a coupled engine. 

The 6ft. wheel type of London and North Western Railway 
compound was introduced in 1884. The outside cylinders are 14in, 
and the inside 30in. diameter, stroke 24in. Joy's gear is used for 
all the valves; the valves to the outside cylinders are below, and the 
valve. of the l.p. cylinder is above the cylinder. The boiler steam- 
pressure is I751b. per square in., but it is reduced to 801b. when enter- 

R2 



244 



EVOLUTION OF THE STEAM LOCOMOTIVE 



ing the low-pressure cylinder. The weight of the engine in working 
order is 42 tons 10 cwt. Heating surface: Tubes, 1,242 sq. ft.; fire- 
box, 159.1 sq. ft.; total, 1,401.5 sq. ft. Grate area, 20.5 sq. ft. 




An engine built to this design the "Marchioness of Stafford'' 
was exhibited at the London Inventions Exhibition of 1885, and 
gained the gold me.dal. 

In 1890 the first of the "Teutonic" (Fig. 95) class of 7ft. lin. com- 
pounds was constructed at Crewe Works. The leading wheels of this 



EVOLUTION OF THE STEAM LOCOMOTIVE 



245 



type are 4ft. Ijin. diameter. Total weight in working order, 45 tons 
10 cwt. In these engines Mr. Webb's loose eccentric motion is used 
for the low-pressure inside cylinder, but Joy's gear is retained for the 
h.p. outside cylinders. 




Another type of compound is the "Greater Britain." During 
1897 the "Greater Britain" and oth,er engines, of the class were 
coloured red, white, and blue, and employed to haul the Royal train 



246 EVOLUTION OF THE STEAM LOCOMOTIVE 

when travelling over the London and North Western Railway system. 
They were then nicknamed the ''Diamond Jubilees." 

The special feature of this class is the length of the boiler, which 
is divided into two portions by means of a central combustion cham- 
ber. The heating surface is: fire-box, 120.6 sq. ft.; combustion 
chamber, 39.1 sq. ft.; front set of tubes, 875 sq. ft.; back set of 
tubes, 506.2 sq. ft. ; total, 1540.9 sq. ft. The two pairs of driving 
wheels are located in front of the fire-box, and in addition there are 
a pair of leading and a pair of trailing wheels. 

An engine of this class the " Queen Empress " (Fig. 96) was ex- 
hibited at the World's Fair held at Chicago in 1893. Her leading 
dimensions are: Two high-pressure cylinders, 15in. diameter by 24:n. 
stroke; one low-pressure cylinder, 30in. diameter by 24in. stroke; 
wheels driving, 7ft. lin. diameter (four in number) ; leading, 
4ft. l^in. diameter; trailing, 4ft. IJin. diameter. Weight on each 
pair of driving wheels, 16 tons. Total weight of engine in working 
order, 52 tons 15 cwt. Total wheel base, 23ft. Sin. Centre to centre 
of driving wheels, 8ft. 3 in. 

The most recent type of compound goods locomotives constructed by 
Mr. Webb has eight-coupled wheels, three pairs of which are located 
under the fire-box, the trailing pair being close to the back of the 
fire-box. The outside cylinders are below the top of the frame-plate, 
and incline towards the rear, This type of engine was designed by 
Mr. F. W. Webb, chief mechanical engineer of the London and North 
Western Railway, principally for working the heavy mineral traffic 
over that Company's South Wales district, the first engine being 
built in 1893. The wheels (all coupled) are 4ft. 5 Jin. in diameter, 
with tyres 3in. thick. The distance between the centres of eaoh pair 
is 5ft. 9in., the total wheel base being 17ft, Sin. All the cylinders 
drive on to one a*de the second from the front of the engine; the 
two high-pressure cylinders are connected to crank pins in the wheels 
set at right angles to each other, the low-pressure cylinder being 
connected to a centre crank-pin set at an angle of 135 degree with the 
high-pressure cranks; the high-pressure cylinders are 15in. diameter 
by 24in. stroke, and the low-pressure cylinders are 30in. diameter by 
24in. stroke. All the cylinders are bolted together and in line, the 
low-pressure being placed immediately under the smoke-box and the 
High-pressure cylinders on each side outside the frames, the steam 
chests being within the frames. 



EVOLUTION OF THE STEAM LOCOMOTIVE 247 

The engine weighs, in working order, 53 tons 18 cwt. 
The empty weights are as follows : 

Tons Cwts. 

Leading wheels ... ... ... ... 12 10 

Driving wheels ... ... ... ... 14 8 

Intermediate wheels ... ... ... .... 12 14 

Trailing wheels ... ... ... ... 9 15 

Total (Empty) ... 7 

The latest type of a passenger compound locomotive built by Mr 
"Webb is the 7ft. four wheels coupled engiire " Black Prince" (Fie:. 97), 
which was built at the Crewe Works in July, 1897. 

The engine has> two high-pressure and two low-pressure cylinders, 
all being in line, and driving on to one axle, the high and low pressure 
cranks being directly opposite each other. 

One of the features of this engine is the method adbpted for 
working the valves, two sets of gear only being used for working the 
four valves. , 

Joy's valve motion is used for the low pressure valves, the valve 
spindles being prolonged through the front of the steam chest, and 
on the end of the spindle a crosshead is fixed which engages with a 
lever of the first order, carried on a pivot firmly secured to the engine 
frame. The other end of this lever engages with a crosshead fixed 
on the end of the high pressure valve spin^fe, and by this means the 
requisite motion is given to the high pressure valve. 

The leading end of the engine is carried on a double radial truck, 
the centre of w r hich is fitted with Mr. Webb's radial axle box and central 
controlling spring. This arrangement permits of lin. side play, and 
gives greater freedom to the truck when passing round curves than is 
possible in the ordinary type of bogie with a rigid centre pin. 

One important object aimed at in the construction of tnis engine 
has been to get all the bearing surfaces throughout as large as possible. 
Each of the four journals in the radial truck is Gjin. diameter and 
lOin. long. The driving axle, in addition- to the two ordinary bear- 
ings, which are each 7Mn. diameter and 9in. long, has a central bear- 
ing between the two cranks., 7in. diameter and 5Jin. long. In the 
trailing axle the journals are 7Jin. diameter, by 13Jin. long. 

This engine made its first trip on August 2nd, 189T, and since 
then has been principally engaged in working the " up " dining saloon 
express, which leaves Crewe at 5.2 p.m., running through to Willesden 
without P, stop a distance of 152J miles and returning the same 



243 



EVOLUTION OF THE STEAM LOCOMOTIVE 



night with the Scotch sleeping saloon express, which leaves Euston at 
11.50 p.m., running through to Crewe without a stop, 158 miles. 




The total distance run by this engine up to June 30th, 1898, was 
52,034- miles. 

The high-pressure cylinders are loin, diameter by 24in. stroke, 
and the low-pressure cylinders are 20|in. diameter by 24in. stroke. 



EVOLUTIOX OF THE STKAM LOCOMOTIVE 249 

Tho heating surface is: Tubes, 1,241.3 sq. ft,; firebox, 159.1 
sq. ft. ; total 1,409-1 sq. ft. Grate area, 20.5 sq. ft. 

A concise survey of other compound locomotives will be of interest 
at this juncture. 

Mr. Wordsell, the then locomotive superintendent of the Great 
Eastern Railway, in 1882 built a compound engine, with two inside 
cylinders, the h.p. 18in. and the l.p. 26in. diameter; the stroke was 
24in. ; steam pressure, 1601b. per sq. in. The coupled wheels were 
7ft. diameter. The engine was fitted with a leading bogie, the wheels 
of which were 3ft. lin. diameter; with her tender she weighed 77 
tons in working order;, her number was 230. A similar engine, No. 
702, with Joy's valve gear, was built in 1885. 

Mr. Wordsell also built a two-cylinder, six-coupled goods engine- 
for the Great Eastern Railway, on the compound principle. This was 
fitted with the ordinary link motion. 

Mr. Wordsell, upon his appointment as locomotive superintendent 
of the North Eastern Railway, introduced compound engines on that 
line. These were provided with two cylinders, both inside, with 
the valves on top. 

Tlia h.p. cylinder is 18in. and the l.p. 26in. in diameter, the 
stroke being 24in. Mr. Worthington thus describes the North Eastern 
Railway standard compound goods engine: "In outside appearance 
this engine is neat, simple, and substantial. It weighs 40 tons 7 cwt., 
and has six coupled oft. IJin. in diameter. 

'* The cylinders are placed, as in the passenger compound engines, 
beneath the slide valves and inside the frames. 

''The chief features of this goods engine to be observed are the 
starting and intercepting valves, which enable the engine-driver to 
start the engine by admitting sufficient high-pressure steam to the 
large cylinder without interfering with the small cylinder, in case the 
latter is not in a position to start the train alone. 

" The two valves are operated by steam controlled by one handle. 
If the engine does not start when the regulator is opened, which v/ill 
occur when the high-pressure valve covers both steam ports, the 
driver pulls the additional small handle, which closes the passage irom 
the receiver to the low-pressure cylinder, and also admits a small 
amount of steam to the low-pressure steam-chest, so that the tv>o 
cylinders together develop additional starting power. 

"After one or two strokes of the engine the exhaust steam from the 
high-pressure cylinder automatically forces the two valves back to 



250 EVOLUTION OF THE STEAM LOCOMOTIVE 

their normal position, and the engine proceeds working com- 
pound." 

The North Eastern Railway has other compound engines con- 
structed on the Wordsell and Van Borries system, a 6ft. 8Jin. four- 
coupled, with a leading bogie of locomotive, being turned out in 1886. 
Engines of this type have a heating surface of 1,323.3 sq. ft., a grate 
area of 17.33 sq. ft., and a working pressure of 1751b. per sq. in. 

Another North Eastern Railway type of compound has 7ft. 6in. 
single driving wheels and a leading bogie. The h.p. cylinder is 20in., 
and the l.p. 26in. diameter, compared with 18in. and 26in. in the four- 
coupled class, the stroke being the same in each design viz., 24in. 

The first of the 7ft. 6in. compound class of locomotives was con- 
structed at the Gateshead Works in 1890. The engines of this design 
appear capable of doing very heavy work with a low coal consumption, 
the average being 281b. per mile, which, considering the heavy traffic 
and speed maintained, is low, being, in fact, 21b. per mile below that of 
any other class of engine engaged on the same traffic. 

With a train of 18 coaches, weighing 310 tons (including 87 tons, 
the weight of the engine and tender), a speed of SB miles an hour 
was attained on a level portion of the road, the horse-power indicated 
being 1,068. These engines have a commodious cab, and the tenders 
carrying 3,900 gallons of water, thvs making it possible for the run of 
125 miles, from Newcastle to Edinburgh, to be performed without a 
stop. There is also a class of six-coupled tank engines, with a trailing 
radial axle. The stroke is 2-iin., and the diameter of cylinders h.p. 
18in., and l.p. 26in. Compound engines have also been tried on the 
Glasgow and South Western Railway and on the London and South 
Western Railway. 

The advantages of express locomotives being fitted with leading 
bogies were speedily recognised by most of the locomotive super- 
intendents. Mr. S. W. Johnson, the Midland chief, introduced a 
design of such engines in 1876. The steam pressure of the early 
engines of this class was 14:01b., but in later years this was increased 
to 1601b., whilst in the recent engines the pressure is still further 
augmented. 

The same progress is to be noticed in the diameter of the cylinders 
of the Midland engines, the diameter having increased from 17 Jin. in 
1876 to 19Jin. at the present time. The size of the coupled wheels 
has also increased from 6ft. 6in. to 7ft. 9in. The length of stroke has 
been the same in all engines of this design viz., 26in. 



EVOLUTION OF THE STEAM LOCOMOTIVE 



251 




252 EVOLUTION OF THE STEAM LOCOMOTIVE 

The new Midland single express engines are illustrated by Fig. 98. 
These locomotives have inside cylinders 19jin. diameter, with a stroke 
of 26in. The driving wheels are 7ft. 9in. in diameter. By standing 
on a railway station platform alongside one of these engines, one gets 
a good idea of their immense proportions, the abnormally high pitch 
of the boiler being especially noticeable. 

Mr. Stroudley, in his " Gladstone " class of engines for the London, 
Brighton and South Coast Railway, adopted an entirely opposite 
practice. His engines had the leading and driving wheels coupled, 
and a pair of smaller trailing wheels. The coupled wheels are 6ft. 
6in. diameter, and the trailing 4ft. Gin. diameter. The cylinders are 
inside, and measure ISJin. diameter, the stroke being 26in. 

The reversing apparatus is actuated by means of compressed air, 
supplied by the Westinghouse brake pump; whilst part of the exha-ist 
steam is projected against the flanges of the leading wheels, and upon 
condensation upon the flanges forms a lubricant to the flange surface, 
when pressing against the inner sides of the rails. Fig. 99 is from 
a photograph of "George A. Wallis," an engine of the "Gladstone" 
class. 




FIG. 99." GEOEGE A. WALLIS/' AN ENGINE OF THE " GLADSTONE " 
CLASS, L.B. & S.C.R, 

The "Ten.nant" (Fig. 100) class of express engines, on the North 
Eastern Railway, deserves mention, being the design of a general 
manager during the North-Eastern locomotive interregnum of 1885. 

The engines have four-coupled wheels, 7ft. diameter, and a lead- 
ing pair of small wheels, cylinders being 18in. diameter, and 24in. 
stroke. 

The cab is somewhat similar to the Stirling pattern on the Great 
Northern Railway. 



EVOLUTION OF THE STEAM LOCOMOTIVE 253 

The North British Railway engine, No. 592, was exhibited at the 
Edinburgh Exhibition of 1886, and Mr. Holmes, her designer, was 
awarded the gold medal. 

The driving and trailing wheels are coupled, their diameter being 
7t't. Tlit fore part of the machine is supported on a four-wheeled 
bogie. The symmetrical appearance of this and other North British 
liailway locomotives is spoilt by having the safety valve located above 
the dome casing. The cylinders are 18in. diameter, and 26in. stroke. 
No. 602, another engine of this design, is notorious as being the lirst 
engine to cross the Forth Bridge, when formally opened by the Prince 
of Wales on March 4th, 1890, the Marchioness of Tweeddale driving 
the engine upon the occasion. 

Mr. Holmes, in 1890, introduced another very similar design of 
North British Railway engines, but with coupled wheels only 6ft. 6in. 
diameter. These are known as the "633" class, illustrated by Fig. 101. 

Turning to the Great r.-isiern Railway, we have to chronicle some 
types of locomotives designed by Mr. Holden. The express passenger 
engines have a pair of small leading w r heels and four coupled wheels of 
7ft. diameter, with cylinders 18in. by 24in. The valves are below the 
cylinders, which, by the way, are both cast in one piece. 




FIG. 100. "1463," N.E.R., ONE OF THE " TENXANT " LOCOMOTIVES 

In connection with this design of locomotive, the triumph of skilled 
mechanism, combined with the application of scientific research, 
deserves record, seeing that a troublesome waste product has been 
turned into a valuable calorific agent. We refer to the introduction 
of liquid fuel for locomotive purposes, as carried out under Mr. 
Holden's patent. 



254 



EVOLUTION OF TRK STEAM LOCOMOTIVE 



Now, sanitary authorities have large powers, and they are very 
fond of abusing these powers, and pushing matters to extreme issues 
although at times, as we know from personal experience, they some- 




times exceed their statutory powers, and find themselves in a tight 
place, from which they can only retreat by payment of compensation 
and heavy law costs. 



EVOLUTION OF THE STEAM LOCOMOT1VK 255 

In connection with the pollution of streams the authorities have 
vi-ry wide powers, and when they found the waters of the never clear 
or limpid Channelsea and Lea were further polluted by some oily, 
iridescent matter, with a pungent odour, the sanitary inspectors were 
soon ferreting out the offenders. The waste products from the Great 
Eastern Railway oil-gasworks at Stratford were found to be responsible 
i'or the nuisance, and the service of a notice requiring immediate 
abatement of the same was the result of the discovery. 

Mr. Holden, remembering the good old proverb, " Necessity is the 
mother of invention," soon commenced to experiment with the matter 
which the sanitary authorities refused to allow to be emptied into 
the already, impure waterways under their jurisdiction. The result 
>f a, series of trials on, first a six-coupled tank engine, and then on a 
single express, was a four-coupled express engine on the G-E.R,, No 
760, named " Petrolea," 

This locomotive was constructed in 1886, and in general appear- 
ance ia| similar to the four-coupled express engine just described. The 
heating arrangements are, however, supplemented by the liquid fuel 
burning apparatus, which may be briefly described as follows : The 
oil fuel is carried on this engine in a rectangular tank of 500 gallons 
capacity, but in later examples occupies two cylindrical reservoirs, 
which contain 650 gallons, placed on the top of the tender w r ater-tanks, 
<>no en each side. 

The -"liquid fuel is supplied to these reservoirs through man-holes 
at the footplate end. The feed pipes from these tanks unite on the 
tender .footplate at the centre, and from this junction the oil is con- 
veyed by the flexible hose pipe to the engine, Where the supply is again 
divided to feed the two burners situated on the fire-box front just 
under t'he footboard. 

Both the liquid in the tanks and the injected air are heated before 
use, the former by means of steam coils in the tanks, and the latter 
by coiled pipes in the smoke-box. The heated liquid fuel and air 
are injected into the fire-box, through tw r o nozzles in the form of fine 
spray, steam being injected at the same moment through an outer 
ring of the same nozzles. The steam divides the mixture of air and 
liquid into such fine particles that it immediately ignites when in con- 
tact with the incandescent coal and chalk fire already provided in the 
fire-box. The fire-box is fitted with a brick arch deflector. 

The whole of the apparatus is controlled by a four-way cock fitted 



EVOLUTION OF THE STEAM LOCOMOTIVE 



on the fire-box case, near the position usually occupied by the regu- 
lator. The positions of the cock in question are: (1) steam to warm 
coils in liquid fuel tank; (2) steam to ring-blowers on injectors; (3) 




steam to centre jets of ejectors ; and (4) steam to clear out the liquid 
fuel pipes and ejectors. The success of "Petrolea" was so apparent 
.and unquestionable that Mr. Holden's patent system of burning liquid 



<>i< y///<; STEAM LOCOMOTIVE 257 

fuel was immediately fitted to other Great Eastern Railway loco- 
motives, with the result that at the present time a number are fitted 
with his patent apparatus. 

The following Great Eastern Railway locomotives have been fitted 
to burn liquid fuel: 

9 four-wheel coupled express engines. 

6 single express engines. (Fig. 102) 

1 six-wheel coupled goods engine. 

1 six-wheel coupled tank, and 
20 four-wheel coupled bogie tanks. 

and the 10 engines of the new class of "single" bogie expresses. 
(Fig. 103.) 

The application of the "Serve" corrugated tube has also been 
introduced on the Great Eastern Railway in connection with the 
liquid fuel. The goods engine and also two of the express passenger 
engines have the "Serve" tubes. The experiment of burning liquid 
fuel has been very successful, only 161b. of oil having been consumed 
per mile run, against an average of 351b. of coal per mile, with coal- 
fired engines. 

Some very handsome Bogie Single Express Locomotives have 
recently been built at the Stratford Works of the G.E.R. Company 
to the designs of Mr. James Holden. They have been specially con- 
structed for 'running the fast Cromer traffic. The boiler has a tele- 
scopic barrel lift, long, in two plates, and is 4ft. Sin. diameter outside 
the smaller ring. It contains 227 tubes If in. external diameter, and 
the height of; its centre line above the rail level is 7f. 9in. 

The fire-box is 7ft. long, and 4ft. Jin. wide outside, and has a grate 
area of 21. 3J sq. ft., and is fired with oil fuel. The total heating 
surface is 1.2-92.7 sq. ft , the tubes giving 1,178.5 sq. ft., and the fire- 
box 114.2 sq. ft. The working pressure is 160 Ibs. per sq. in. 

Tho driving wheels are 7ft., the bogie wheels are 3ft. 9in., and tho 
trailing wheels 4ft. diameter. The total wheel base is 22ft. 9in., the 
bogie wheels centres being 6ft. 6in. apart, from centre of bogie pin to 
centre of driving wheel is 10ft. 6in., and from centre of driving wheel 
to centre of trailing? i 9ft. The total length of engine and tender, over 
buffers, is 53ft. 3in. " 

The cylinders are 18in. diameter by 26in, stroke, the distance be- 
tween centres being 24in. The slide valves are arranged underneath, 

8 



258 



EVOLUTION OF THE STEAM LOCOMOTIVE 



and are fitted with a small valve, which allows any water that may 
collect in the slide valve to drain off. 




Steam sanding apparatus is fitted at front and back of the driving 
wheels. 



EVOLUTION OF THE STEAM LOCOMOTIVE 259 

Macallan's variable blast pipe is used, the diameter of the pipe 
being 5Jin., and of the cap 4Jin. 

This variable pipe is being adopted on all the Company's engines. 
The pipe has a hinged top, operated from the footplate. When the 
hinged top is on the pipe, the area is that of a suitable ordinary pipe, 
and when the top is moved off the area is about 30 per cent, larger. 

It is found that a large proportion of the work of the engines can 
be done with the larger exhaust outlet, the result being a reduced 
back pressure in the cylinders and also a reduced vacuum in the 
smoke-box, and less disturbance of the fire and consequent saving of 
fuel. 

The tender is capable of carrying 2,790 gallons of water, 715 
gallons of oil fuel, and l tons of coal. It is provided with a water 
scoop for replenishing the tank whilst running. 

The weights of the engine and tender in working order are : 
engine, 48J tons; tender, 36 tona; total, 84J tons. 

The oil-firing arrangements embody a number of ingenious details, 
among them the supply of hot air for combustion from a series of 
cast-iron heaters placed around the inside of the smoke-box, the air 
being drawn from the front through the heaters to the burners for the 
exhausting action of the steam jets used for injecting the oil fuel. The 
latter is warmed before leaving the tender in a cylindrical heating 
chamber, through which the exhaust steam from the air-brake pump 
circulates. 

The regulation of the oil supply is effected by a neatly designed 
gear attached to the cover and hood of the ordinary fire-door, and 
finally the burners or injectors are so constructed that should one 
require cleaning, inspection, or renewal, the internal cones can all be 
removed from the casing by simply unscrewing one large nut. Thesa 
engines have polished copper chimney tops, and are painted and lined 
in the standard G.E.R. style, and fitted with the Westinghouse auto- 
matic brake. 



8 2 



CHAPTER XIV. 

Modern L.B. and S.C.R. locomotive? Foxnr-coupled in front pf^eenger tank Six- 
coupled tank with radial trailing wheels Goods engines " Bessemer," four- 
coupled bogie express " Inspector " Standard L.C. and D.R. paessngtr 
engines Goods locomotives Three classes of tanks Ca.mbdun locomotives, 
passenger, goods, and tank S.E. engines A "Prize Medal" locomotive- 
Stirling's goods and tank engine His latest type of express engines Adams's 
locomotives on the L. & S.W.R. Mixed traffic engines Passenger and six- 
coupled tanks -Dru-nmond's " Wmdcutter " moke-box His four-cylinder ex- 
press engine North British passenger locomotives Engines for the West 
Highland Railway Holme's goods and tank engines His latest express type 
of engi le Classification of N.B.R. locomotive? N.B.R. inspection or cab 
engine L. and Y. locomotives Aspinall's water " pick-up " apparatus 
Severe gradients on the L. and Y. system 7ft. Sin. coupled expresses 
"A" class of goods engines Standard tank engine? L. and V. oil-burn : ng 
tank locomotives Caledonian Railway engines Drummond's famous 
" Dunalastairs " Excelled by his " Dunalastairs 2" Six-coupled " t,ondexig- 
ing" tender engines " Carbrooke " class Dimensions of 44 types of 
Caledonian locomotives 

The modern engines on the London, Brighton, and South Coast 
Railway are designed by Mr. Billinton, and comprise 

The four-coupled in front tank with a trailing bogie, of which 
"Havant," No. 363, is an example. This engine was built at 
Brighton Works, 1897. Inside cylinders, 18in, by 2Sin. ; diameter 
Coupled wheels, 5ft. Gin. diameter. Heating surface, 1,189 sq. ft. 
Steam pressure, IGOlb. Weight in working order, 47 tons. 

" Watersfield," No. 457, built at Brighton in 1895, is a specimen 
of the six-coupled goods tank engines, with radial trailing wheels- 
This class have inside cylinders^ 18in. by 2 Gin. ; heating surface, 
1,200 sq. ft.; diameter of wheels, 4ft. Gin.; steam pressure IGOlb. ; 
weight in working order, 51 tons. 

No. 449 represents the six-coupled goods tender engines, built by 
Vulcan Foundry Co. in 1894, from Mr. Billinton's designs. Inside 
cylinders, 18in. by 26in. ; wheels, 5ft. diameter ; heating surface, 
1,212,, sq. ft.; steam pressure, IGOlb. ; weight in working order: 
engine, 38 tons ; tender, 25 tons. 

"Bessemer," No. 213, is one of the new type of four-coupled 
express passenger engines, with leading bogie, and was built at 
ton Works, 1897. Inside cylinders, 18in. by 26in. ; diameter 
of coupled wheels, 6ft 9in. ; heating surface, 1,342 sq. ft.; working 



E VOL I TlOX or Til /; ,s TEA M LOCOMOTIVE 



261 



pressure, 1701b. ; weight in working order: engine, 44 tons 14 cwt. ; 
tender, 25 tons. Fig. 104 is from a photograph of " Goldsmith," an 
engine of this class. 

Before closing this short description of the London, Brighton, c.ixl 
South Coast Railway locomotives, attention must be called to the 
combined engine and carriage named "Inspector," No. 481 (Fig. 105). 
This engine was constructed in 1869 by Sharp, Stewart, and Co., as 
an ordinary four-coupled passenger tank, and rebuilt in its present 
form some 11 years or so ago. 

The cylinders are inside, lOjin. diameter, 16in. stroke; coupler- 
wheels, 4ft. diameter; weight in working order, about 20 tons; 
steam pressure, 1201b. In addition to the coupled wheels there are 
also a pair of leading and a pair of trailing wheels. There is r.o 
steam dome, and the side tanks are as long as the boiler barrel, being 




FIG. 104. "GOLDSMITH," ONE OF THE NEW L.B. & S.C.R. EXPRESS 
PASSENGER ENGINES 

extended on each side to the smoke-box. The inspection car is fixed 
on to the back of the coal bunker, its floor is some distance below the 
level of the engine frames, and the car is entered from a platform, at 
the end, which is in turn entered from the outside by steps on either 
side, as in a tram-car. The back of the platform is quite open, 
whilst the partition dividing the platform, from the enclosed portion of 
the car is glazed, so that anyone sitting with his back to the coal 
bunker can see the permanent-way, etc., over which "Inspector" has 
just passed without leaving his seat if necessary. There is a speaking 
tube, to enable those m the saloon to communicate with the driver. 



262 EVOLUTION OF THE STEAM LOCOMOTIVE * 

A special form of indicator board, not used for any other train, is 
carried by "Inspector" viz., a white board with black horizontal 
stripes. 

The modern locomotives of the London, Chatham and Dover Rail- 
way are built from designs prepared by Mr. William Kirtley, the Com- 
pany's locomotive superintendent. The main line passenger engines 
(Fig. 106) are of the M3 class, and have the following dimensions : 
Cylinders, 18in. diameter, 26in. stroke; 
Couple^ wheels, 6ft. 6in. in chamber ; 
Bogie 3ft. 6in. 

Heating surface: tubes, 1,000.2 sq. ft.; fire-box, 110 sq. ft. 
Grate area, 17 sq. ft.; working pressure, 1501b. 
Weight, in working order, 42 tons 9 cwt., of which the driving 

and trailing coupled wheels support 28 tons 18 cwt. 
The standard tender, for both goods and passenger engines, is 




FIG. 105. "INSPECTOR," LONDON, BRIGHTON, AND SOUTH COAST BY. 

carried on six wheels, and, loaded, weighs 34 tons; accommodation is 
provided for 4J tons of coal, and 2,600 gallons of water. 

The standard goods engines have six coupled wheels, 5ft. in 
diameter. 

Cylinders, 18in. by 26in; 

Heating surface: tubes, 1,000.4 sq. ft.; fire-box, 102 sq. ft.; 

working pressure, 1501b. per sq. in. ; 

Weight, in working order : leading 13 tons 2 cwt. ; driving, 
15 tons 4J cwt,; trailing, 10 tons 19 ct.; total, 39 tons 
6 cwt. 



EVOLVTI<>.\ nV THE STEAM LOCOMOTIVE 



263 



These engines are known as " Class B2." 

The tank engines consist of three classes. 

The dimensions of those for working the main line and suburban 
services are as follows : 

Inside cylinders, with an incline of 1 in 10, I7in. diameter; 24in. 
stroke. Wheels, four coupled in front, 5ft. Gin. diameter. A trailing 
bogie with 3ft. wheels. Heating surface: tubes, .971.7 sq. ft.; fire- 
box, 99-3 sq. ft.; grate area, 16 J sq. ft. Tank capacity, 1,1(0 gallon? 
of water, 2 tons of coal. Weight, in working order, 49 tons 15 rwt. 
Steam pressure, 1501b. per sq. in. 

These engines are officially described as Class R. 




FIG. 106. "No. 192," ONE OF THE STANDARD EXPRESS PASSENGER LOCOMOTIVES, 
LONDON, CHATHAM AND DOVER RAILWAY 

The A class of bogie tank engines were specially designed for 
working through tunnels. The inside cylinders are I7|in. diameter, 
and 26in. stroke. The coupled (leading and driving) wheels are oft. 
Gin. diameter, the wheels of the trailing bogie being 3ft. in diameter. 
The heating surface is made up as follows : Tubes, 995 sq. ft. ; fire- 
box, 100 sq. ft.; grate area, 16 J sq. ft.; working pressure, 1501b. ; 
water capacity of tanks. 970 gallons ; fuel space, 80 cubic ft. ; weight, 
in working order, 51 tons. 



264 EVOLUTION OF THE STEAM LOCOMOTIVE 

All of these engines are fitted with steam condensing apparatus 
to allow of working over the Metropolitan Railway between Snow Hill 
and King's Cross and Snow Hill and Moorgate Street. 

Class T comprises the goods or shunting tanks. These have six 
coupled wheels of 4ft. Gin. diameter, with a wheel bore of 15ft. The 
cylinders are inside, with a I7in. diameter and 24in. stroke. The 
heating surface is as follows : Tubes, 799.3 sq. ft. ; fire-box, 88.7 




Fl(J. 107. STANDARD EXPRESS PASSENGER ENGINE, CAMBRIAN RAILWAYS 

sq. ft.; grate area, 15 sq. ft.; steam pressure, 1501b. ; tank capacity, 
830 gallons ; coal bunker, 48 cubic ft. ; weight, in working order, 
40f tons, 

In 1889 these shunting engines were fitted with the Westinghouse 
Automatic Brake, which is the continuous brake adopted by the 
London, Chatham and Dover Railway. 

In general outline the modern locomotives on the Cambrian Rail- 
ways are similar to those of the London, Chatham, and Dover Railway. 

The express passenger engines (Fig. 10 7) 'on the Cambrian Railway 
have a leading bogie, with wheels 3ft. Gin. diameter, and four coupled 
wheels of 6ft. diameter. The Inside cylinders are inclined 1 in 15, 
and. are 18in. diameter, and 24in. stroke. The heating surface is : 
Tubes, 1,057 sq. ft. ; fire-box, 99 J sq. ft. ; grate area, 17 sq. ft. There 
are 230 tubes, 10ft. 5- 16th in. long, and If in. diameter. The wheel 
base is : Centre to centre of bogie, 5ft. Gin. ; leading to trailing, 7ft. ; 
centre of bogie to driving, 9ft. 3 Jin. ; and driving to trailing, 8ft. 3in. 
Boiler pressure, 161b. per sq. inch. These engines have underhung 
springs to the driving and trailing wheels, are fitted with a steam 
sanding -apparatus, the vacuum brake, screw reversing gear, and 
other improvements. They were built by Sharp, Stewart, and Co., 



EVOLUTIOX OF Till': STKAM LO<'< MOTIVE 



265 



Atlas Works, Glasgow, the particular one we have described having 
been turned out in 1893. 

The modern goods engines numbered 73 to 77 werebuilfcby Neilson 
and Co., (Jlasgow, in 1894, the maker's numbers being 4,691 to 4,695. 
The six-coupled driving wheels are 5ft. l^m. ; diameter ; the wheel 
base being : leading to driving, 7ft. 5in. ; driving to trailing, 7f fc. 
1 Oin. The springs to all the wheels are underhung, ' the driving 
wheel springs being of Tinimis' patent design. Steam sanding 
ipparatus is provided in front of the leading wheels. The cylinders 
iro inside, and are inclined 1 in 10. The boiler barrel is 10ft. 3in. 
long, and contains 204 tubes of Ijin. diameter; the heating surface 
la-ing: tubes, 986.2 sq. ft; fire-box, 98.3 sq. ft.; fire-grata area, 
16 J sq. ft.; working pressure, 1601b. per sq. in. 

The tenders have six wheels. 3ft. lOin. diameter, with a wheel 
base of 12ft., equally divided. Water capacity, 2,500 gallons; coal 
space, 200 cubic ft. 




FIG. 108. STANDARD PASSENGER TANK ENGINE, CAMBRIAN RAILWAYS 

The above dimensions are those of the Cambrian Railways modern 
standard tender, and apply both to the passenger and goods engines. 

The bogie passenger tank engines (Fig. 108) have inside cylinders, 
I7in. diameter, 24in. stroke, inclined 1 in 9- The coupled wheels (leading 
and driving) are 5ft. 3in. diameter, the bogie wheels being 3ft. l|in. 
diameter. The boiler barrel is 10ft. 2fin. long, and contains 134 
tubes of 2in. diameter, and 38 tubes of Ijin. diameter 
Boiler (pressure, 1601b. per sq^ inch. Heating surface : 



266 EVOLUTION OF THE STEAM LOCOMOTIVE 

Tubes, 920.1 sq. ft; fire-box, 90 sq. ft,; grate area, 13.3 sq.ft. 
The tanks contain 1,200 gallons of water, and the bunkers 2 tons of 
coal. The total wheel base is 20ft. lin., the coupling side-rods being 
7ft- 8in. long. Weight, in working order, 45 tons 9 cwt. 3 qrs. 

These engines were built by Nasmyth, Wilson, and Co., Ltd., 
Bridgewater Foundry, near Manchester. 

Mr. James Stirling, the present locomotive superintendent of the 
South Eastern Railway, soon aft-er his appointment, took steps to 
thoroughly renovate and classify the various types of locomotives on 
the system. 

He has now succeeded in doing so; indeed, save for a few 6ft. 
D. and T. coupled, of Cudworth's design, now rebuilt without a dome, 
and the six-wheel four-coupled express engines built during the short 
Watkin locomotive regime, and now rebuilt by Mr. Stirling, nearly 
every engine on the South Eastern Railway is from Mr. Stirling's own 



It should be mentioned that Mr. James Stirling, like his brother, 
the late Patrick Stirling, of Great Northern Railway fame, does not 
believe in a steam dome. Another feature of resemblance in their 
designs is discovered in the style of cab. Patrick favoured a brass 
encased safety-valve, located on the top of the fire-box; whilst James 
chooses the boiler barrel for the position of that useful feature in a 
locomotive, which he, however, constructs after the Ramsbottom type. 

The modern South Eastern Railway engines all have inside cylin- 
ders, and Mr. Stirling's excellent reversing gear previously described. 
They may be divided into the following classes : 

Four-wheels coupled bogie express engine of two sets of 
dimensions. 

Four-wheels coupled bogie passenger engine. 
Four-wheels coupled bogie tank engine. 
Six-wheels coupled, goods engine. 
Six-wheels coupled shunting tank engine. 

Tho standard express class of engines was introduced about 15 
years ago, and the locomotives were then painted black, but fortunately 
for their appearance, Mr. Stirling has recently reverted to the South 
Eastern Railway colour obtaining before his appointment as loco- 
motive superintendent, and the newer engines are now painted a pleas- 



EVOLUTION OF THE STEAM LOCOMOTIVE 267 

ing tint of olive green. " No. 240 " (Fig. 109), an engine of this class, 
was exhibited at the Paris Exhibition of 1889, and obtained the Gold 
Medal. 

The leading dimensions are: Cylinders, 19in. diameter, 26in. 
stroke (incline 1 in 30); leading bogie wheels, 3ft. 9in. diameter; 
wheel base of bogie, oft. 4in. ; driving and trailing wheels (coupled), 
7ft. diameter; wheel base of coupled wheels, 8ft. Gin. The driving 
wheels have Timmis's springs; the trailing wheels underhung 
laminated springs. 

The tender is carried on six wheels of 4ft. diameter, with a wheel 
base of 12ft., equally divided. The tender tank holds 2,650 gallons of 
water, and the coal capacity is ' 4 tons. 




FIG. 109. "No. 240," THE SOUTH EASTERN RAILWAY ENGINE THAT 
OBTAINED THE GOLD MEDAL AT THE PARIS EXHIBITION, 1889 

Weight, in working order: on bogie, 13 tons 12 cwt. ; driving 
wheels, 15 tons 18 cwt.; on trailing wheels, 13 tons; tender. L., 
10 tons 6 cwt.; centre, 10 tons 1 cwt.; T., 10 tons 3 cwt.; total 
weight of engine and tender, 73 tons. 

From the above description it will be seen that these locomotives 
are finely proportioned and should be capable of doing excellent ser- 
vice. They are good for hauling heavy loads, and the "direct" line 
via Sevenoaks has some severe gradients, which these engines nego- 
tiate in fine style. 

Another point in their favour is the coal consumption, the average 
being low, although the fuel is of inferior quality. 



268 



EVOLUTION OF THE XTEAM LOCOMOTIVE 



The speed, however, of the fast trains is disappointing. Probably 
it is not right to blame the engines for this, but rather the timing 
of the trains. 

Whilst other railways are accelerating their services, the South 
Eastern Railway retrogrades in the matter of speed. 

Yet there is not a finer length of line in the kingdom for showing 
what an engine can do than that between Redhill and Folkestone, or 
leaving the main line at Ashford and on to Ramsgate. For many 
miles these tracks are practically straight and level; but no advantage 
is taken of the circumstances so far as speed is concerned ; hence 
travellers are apt to blame the locomotives. These probably have 
never had a chance to show what speed they are capable of. 




FIG. 110.-STANDARD GOODS ENGINE, SOUTH EASTSRX RAILWAY 

Mr. Stirling's other class of bogie tender engines is very similar in 
appearance to the one just described, but of smaller dimensions. The 
engines now to be described were first constructed some years before 
the 7ft. coupled expresses ; indeed, soon after Mr. Stirling was 
appointed locomotive superintendent. They are principally used for 
working the passenger trains on the North Kent line (London to 
Maidstone). 

Cylinders, 18in. by 26in. (incline 1 in 15); bogie wheels, 3 ft. Sin. 



EVOLCTIOX OF THE STL'-iM LOCOMOTIVE 



269 



diameter; wheel base, 5ft. 4in. ; driving and trailing wheels 
(coupled), 6ft. (Hin. diameter (wheel base, 8ft. 2in.); springs and 
similar to the 7ft. class; tender wheels, 3ft. Sin. diameter; 
base, 12ft., equally divided; water capacity, 2,000 gallons; 
coal, 3 tons. Weights in working order: on bogie, 12 tons 12 cwt. ; 
driving axle, 14 tons 2 cwt.; trailing, 11 tons 5 cwt.; tender, L., 
8 tons 12 cwt.; centre, 8 tons 2 cwt.; T., 9 tons; total weight 
(tngine and tender), 63 tons 13 cwt. 

The tender goods engines (Fig. 110) have six wheels (coupled) of 5ft. 
2in. diameter; cylinders, 18in. by 26in. (incline 1 in 9) ; wheel base, L. 
to D., 7ft, 4in. ; D. to T., 8ft. 2in. The tenders are of similar 
dimensions to the 6ft. passenger engines, with 100 gallons additional 
water capacity. Weights in working order: engine, L., 12 tons 
2 cwt.; D., 15 tons 3 cwt,; T., 11 tons. Tender, L., 9 tons 5 cwt.; 
C, 9 tons 1 cwt.; T., 9 tons 17 cwt.; total (engine and tender), 
64 tons 18 cwt. 

The four wheels coupled bogie tanks (Fig. Ill) have the leading 




FIG. 111. STANDARD PASSENGER TANK LOCOMOTIVE, S.E. MWY. 

and driving wheels coupled ; these are 5ft. 6in. diameter ; cylinders, 
18in. by 26in. (incline 1 in 9); trailing bogie, with wheels, 3ft. 9in. dia- 
meter ; side v.ater tanks, capacity, 1,050 gallons; coal bunker capacity, 
30 cwt,; wheel base, L. to D., 7ft, 5in. ; D. to bogie centre, lift, 
llin. ; bogie wheel base, 5ft. 4in. Weight in working order: L., 
13 tons 17 cwt.; D., 16 tons; bogie, 18 tons 16 cwt,; total, 
48 tons 13 cwt. 



270 EVOLUTION OF THE STEAM LOCOMOTIVE 

The above is a capital type of passenger tank engine, of which 
the South Eastern Railway possess a large and increasing number. 
They are mostly constructed by Glasgow firms, whilst the tender 
engines are built at Ashford Works. 

There is a similar type of bogie tanks, fitted with condensing 
apparatus, and used for working the through South Eastern trains 
over the Metropolitan Railway to the Great Northern Railway. Some 
of these engines were also used for hauling the South Eastern trains 
through the Thames Tunnel, when the through service between Croydon 
(Addiscombe Road) and Liverpool Street was in operation. For this 
purpose they were fitted with a short funnel, to enable them to clear 
the Thames Tunnel. 

The illustration (Fig. 112) shows Mr. Stirling's latest type of express 
engine for the South Eastern Railway, the first of which commenced 
to work at the end of July, 1898. Several differences of detail com- 
pared with Mr. Stirling's previous South Eastern Express engines are 
introduced. The more noticeable are the large bright brass stand 
upon which the safety-valves are mounted, the improvement in the 
shape of the cab on the engine, whilst the sides of the tender are 
painted in two panels, with the Company's coat of arms between (Mr. 
Stirling, it will be observed, has not slavishly copied other practice in 
lettering the tenders S.E.R.) ; the springs are below the frames, and 
steps at the back are provided on either side of the tender. 

The diameter of the wheels and cylinders, the stroke, and wheel- 
base remain the same. The tender rs a trifle longer, making the total 
length over buffers 52ft. 8in., instead of 52ft. 4in. The working 
pressure is now I701b. per sq. in., there being 215 tubes of If in. ex- 
ternal diameter, 10ft. 4 Jin. long. The other differences in the dimen- 
sions are tabulated below : 

"440" Class, illustrated by FlG. 112. "240" Class, illustrated by FIG. 109. 

Bail level to centre of Boiler . . . . 7ft. lOin 7ft. Sin. 

Total Heating Surface .. .. .. 1,100 sq. ft .. 1 020i sq. s t 

To top of Chimney is 13ft. 4in. in both classes,the new Engines having Funnels 2in. shorter. 
Weight loaded 

ENGINE. ENGINE. 

Bogie. D. T. Bogie. D. T. 

15 tons. 16 tons 8 cwt. 14 tons 13 cwt. 13 tons 12 cwt. 15 tons cwt. 13 tons. 

TENDER. TENDER. 

L. C. T. L. C. T. 

10 tons 15 cwt. 10 tons 18 cwt. 12 tons 9 cwt 10 tons 6 cwt. 10 tons 1 cwt. 10 tnns 3 cwt. 

TOTAL 80 tons 3 cwt. TOTAL 73 tons. 

Water Capacity of Tender .. .. 3,000 galls .. .. 2,650 galls. 

Goal ,, ,, .... 3 tons ... 4 tons. 

With the increased weight, boiler pressure, and heating surface of 
these engines, coupled with a compromise towards a steam dome, such 
fine locomotives ought to be quite equal to hauling the heavy trains 



EVOLUTION OF THE STEA M LOCOMOTIVE 



271 




272 



EVOLUTION OF THE STEAM LOCOMOTIVE 



run by the South Eastern Railway at high speeds. Mr. Stirling is to 
be congratulated upon the appearance of the machines. 

The standard engines now in work on the London and South 
Western Railway were constructed from the designs of Mr. Adams, 
the late locomotive superintendent, who resigned about three years 
ago. Mr. D. Drummond, who succeeded Mr. Adams, has built sever.il 
new types of engines, viz., large bogie tank engines, six-wheels-coupled 
goods engines, four-wheels-coupled bogie express engines, as well as 
.?, "four-cylinder" engine, which latter is decidedly a new departure 
in London and South Western Railway practice. The most important 
of Mr. Adams' designs can be classified thus : 

Four-coupled bogie express engine and tender. 

Four-wheels-coupled in front, mixed engine and tender. 

Six-wheels-coupled goods engine and tender. 

Four-wheels-coupled bogie tank engine; and 

Six-wheels-coupled bogie shunting tank engine. 

There are two classes of four-wheels-coupled bogie passenger 
engines, both of the same design, but of different dimensions. 

The appended table Avill show the variations in the two classes : 










1 


1 ! '1 


o 









1 


G 
O 

O 


* 








-* d -s 


^ ^ 


^ 






H 


E 


% 35 


6 

ft 

"C 

5 


Cylmders. 


Boiler Pres 


f 


fo , 0-3 
o 1 fe^ 

^ =i^ 
60 i 3 

1 1 5 


Di:imeter o 
Coupled 


Heating Su 
Tubes. 


Fire Bos. 


Grate Area 


Number of 


Tractive F 
Hails. 


Water Cap 
of Tend 


IB. ia 


Ibs. 


ft. 


ft. in. ft. in. 


fr. in. 


Sjuarefect. 




Ibs. 


gjA*. 


: 


















26 17x2G 
43 19 x2<! 


160 ' 11 
175 ! 11 


60133 
6 10 j 3 7 


6 7 
6 7 


1121 
1193.7 


110 
112-12 


17 
19-75 


216 
230 


10,079 
13,0<>9 


3,000 
3,300 


i. ; 




i 

















These engines (Fig. 113) have outside cylinders, underhung springs 
to the coupled wheels, the springs being connected by means of a com- 
pensation beam; a dome on the boiler barrel, and a Ramsbottom 
safety valve on the fire-box. A notable feature in the design is the 
distance the frames project in front of the smoke-box. The style 
of cab is also very neat. 

A great feature in Mr. Adams's later engines was his patent 
vortex blast pipe, the introduction of which very considerably re- 
duced the coal consumption of the locomotives fitted with the 
invention. 



OF THE STEAM LOCOMOTIVE 



273 



The mixed traffic engines have inside cylinders, 18in. diameter, 
2 Gin. stroke, leading and driving wheels (coupled) 6ft. diameter, and 
trailing wheels 4ft. diameter, underhung springs and compensation 
beams to the coupled wheels; steam pressure, 1601b. The heating 
surface and grate area are similar to the " 26 " design. Tractive force 
on rails, ll,7001b. Tender capacity the same as "43" design. 

The six-coupled goods engines have inside cylinders, I7in. dia- 
meter, 26in. stroke; wheels, 5ft. lin. diameter; steam pressure, 
1401b. ; underhung springs; boiler barrel, 10ft. Gin. long, 4ft. 4in. 
diameter; fire-box, oft. lOin. long, 5ft. high. The smoke-box front 




FIG. 113. -ADAMS'S STANDARD EXPRESS ENGINE, LONDON AND SOUTH 
WESTERN RAILWAY 

inclines, so that the box is wider at the base than at the top, as is 
the case with the London and North Western goods engines. There 
are 218 tubes of If in. external diameter; the heating surface being: 
tubes, 1,079 sq. ft.; fire-box, 108 sq. ft.; grate area, 17.8 sq. it. 
Tractive force on rails, 10,442Jb. ; water capacity of tender, 2,500 
gallons. 

The suburban and other short distance passenger traffic is per- 
formed by tank engines, having the leading and driving wheels 
coupled, and a trailing bogie. The cylinders are inside, 18in. dia- 
meter, 26in. stroke; coupled wheels, 5ft. 7in. diameter; bogie wheels, 

T 



274 



EVOLUTION OF THE STEAM LOCOMOTIVE 



3ft. diameter; heating surface and grate area the same as in the 
" 26 " class and the mixed traffic engines already described. Steam 
pressure, 1601b. per sq. in.; fuel space of bunkers, 80 cubic ft.; 
water capacity of tanks, 1,200 gallons. Tractive force on rails, 
12,5731b. 

The six-wheels-coupled shunting tanks are altogether of smaller 
dimensions, the cylinders being 17 Jin. diameter, and having a 24in. 
stroke wheels, 4ft. lOin. diameter; boiler barrel, 9ft. 5in. long and 
4ft. 2in. diameter, containing 201 tubes of llin. external diameter. 
The heating surface is : tubes, 897.76 sq. ft. ; fire-box, 89.75 sq. 
ft. ; the fire-box is 5ft. long and 4ft. 9in. high, the grate area being 
13.83 sq. ft.; the steam pressure, 1601b. Tractive force on rails, 




Photo by] 



[P. Moore 



FIG. 114. A "WINDCUTTER" LOCOMOTIVE ENGINE, "No. 136," L. 
FITTED WITH A CONVEX SMOKE-BOX DOOR 



S.W.R , 



12,6721b. ; fuel capacity of bunker, 77J cubic ft.; capacity of water 
tanks, 1,000 gallons. 

The London and South Western Railway at one time had an 
extraordinarily large number of different designs of locomotives, and 
at the present time the number of designs in use probably exceeds that 
on any other British railway, despite the fact that the older clashes 
are being rapidly " scrapped," although some of the very ancient 






EVOLUTION OF THE STEAM LOCOMOTIVE 275 




276 EVOLUTION OF THE STEAM LOCOMOTIVE 

types have in recent years been rebuilt with new boilers. The older 
engines of Battle design mostly have names, but this practice, unfor- 
tunately, has been disregarded by recent London and South Western 
Railway locomotive superintendents, save in the case of one tank 
engine, named " Alexandra," under special circumstances, and even 
this name has lately been removed. 

Since Mr. Drummond has become chief at Nine Elms, two at least 
of his innovations deserve notice. One is an experiment with a 
windcutter smoke box door (Fig. 114), constructed in the belief that 
the wind resistance is thereby decreased. In addition to the tender 
engine 136 being so fitted, this form of convex smoke-box door is fitted 
to a L. and S.W.ll. tank engine, and also to some of the tender goods 
engines. 

Another type of engine, designed by Mr. Drummond, that has 
attracted considerable attention, is the four-cylinder engine (Fig. 115), 
built at Nine Elms at the end of 1897. This engine is supported on 
four driving wheels (uncoupled) of 6ft. 7in. diameter, and a leading 
bogie. Joy's valve-gear is used for the outside cylinders ; all the 
cylinders are 15in. diameter; the stroke is 26in. A very large heat- 
ing surface, including the water tubes in the fire-box, amounting to 
1,700 sq. ft., is provided. The steam pressure is I751b. p^r sq. in. 
The tender is carried on two four-wheel bogies, and carries 4,300 
gallons of water. The motion is reversed by means of a steam 
apparatus. A portion of the exhaust steam is discharged at the 
back of the tender. 

The locomotive works of the North British Railway are situate 
at Cowlairs, Glasgow, and Mr. M. Holmes is the present locomotive 
superintendent. 

Originally the North British Railway works were located at St. 
Margaret's, near Edinburgh, but when the Edinburgh and Glasgow 
Railway was amalgamated with the North British, in 1865, the Cow- 
lairs works of the former were chosen as the locomotive headquarters 
of the Company. 

Considerable power is required to work the trains over the North 
British system, as not only are the trains heavy, but many are run 
at a good speed, whilst steep gradients are not unknown. 

II; is not, therefore, surprising that "single" engines should be 
absent from the locomotive stock. 



EVOLUTION OF THU STEAM LOCOMOTIVE 



277 



The passenger engines are mostly of the four-coupled leading 
bogie type. (Fig. 116). 

The principal passenger engines have the coupled wheels of 
6ft. 6in. and 7ft. diameter, both with cylinders 18in. by 26in. The 
steam pressure is 1401b. usually, but some of the engines are credited 
with an additional lOlb. per sq. in. 

The other dimensions are : 



Heating Surface Tubes 

Fire-box 

Grate Area 

Weight in Working order 



7 pp. WHEELS. 

1,007 sq. ft. 

119 ,, 

22 

45 tons 6 c\vt. 



6 FT. 6 IN. WHEELS. 

1,148 sq. ft. 

118 

20 

46 tons 10 cwt. 



The driving wheels of both sizes have a weight of 15 tons 12 cwt. 
upon them. The tenders weigh 32 tons, and hold 5 tons of coal and 
2,500 gallons of water. 

Engines of these classes work the East Coast expresses between 




F;o. 116. FOUR-COUPLED PASSENGER ENGINE WITH LEADING BOGIE, N.B.R. 

Edinburgh and Berwick, 57 miles 42 chains. The booked time is 
72 minutes, but the runs are frequently performed under the hour; 
indeed, a train has been timed from start to stop in 57 minutes 21 
seconds, on the journey up from Edinburgh to Berwick. 

For working the West Highland Railway Mr. Holmes designed a 
class of four-coupled bogie engines of exceptional power. The coupled 
wheels are only 5ft. 7in. in diameter; cylinders, 18in. diameter, 2 4in. 
stroke; heating surface tubes, 1,130.41 sq. ft. ; fire-box, 104.72 sq. ft.; 



278 EVOLUTION OF THE STEAM LOCOMOTIVE 

grato area, 17 sq. ft.; steam pressure, 1501b. per sq. in.; weight 
of engine in working order, 43 tons 6 cwt., of which 14| tons rest 
on the driving axle. 

The tender is similar to that previously described. 
Goods engines are very numerous on the North British Railway, 
the most modern ones being known as the " 18in. standard" type. 
These have six coupled wheels of 5ft. diameter; cylinders, 18in. by 
26in. .stroke; heating surface tubes, 1,139.96 sq. ft.; fire-box, 107. 7i 
sq. ft.; grate area, 17 sq. ft.; weight in working order, 40 tons 13 
cwt., of which 15 tons 8 cwt. are supported by the driving wheels. 
The tender is of the usual type. Other goods engines have cylinders 
I7in. diameter, with 26in. stroke. 

The short distance passenger traffic is worked by four classes of 
tank engines, one type of which is very similar to the London, 
Brighton, and South Coast "terriers," though of larger dimensions. 
These have cylinders 15in. by 22in., coupled wheels, 4ft. Gin. dia- 
meter, tanks to hold 600 gallons of water, and weigh 33J tons in 
working order. Another class of bogie tank has coupled wheels 5ft, 
in diameter, a leading bogie with solid wheels 2ft. 6in. diameter, 
cylinders 16in. by 22in. stroke. These engines originally condensed 
the exhaust steam, but the usual practice is now followed, and the 
exhaust is used as a blast for increasing the draught. 

The two other classes of tank engines have the following 
dimensions : 

494 Class : Cylinders, I7in. by 26in. ; diameter of driving wheels, 
6ft. ; water capacity, 950 gallons ; coal, 30 cwt. ; weight, 
47 tons 4 cwt. 

586 Class : Cylinders, 1 7in. by 24in. ; diameter of driving wheels, 
5ft. 9in. ; water capacity, 1,281 gallons; coal, 50 cwt.; weight, 
50 tons 7 cwt. 

There is a handy little type of saddle tanks, known as " shunting 
pugs." These run on four (coupled) wheels of 3ft. Sin. diameter; 
they have outside cylinders, 14in. diameter and 20in. stroke. The 
wheel base is 7ft.; weight in working order, 28 tons 15 cwt.; water 
capacity of saddle tank, 720 gallons. 

Mr. Holmes' latest type of express engines fortheN.B.R. (Fig. 117) 
has a working pressure of I751b. per sq. in. The principal dimen 
sions being : Cylinders, 18Jin. diameter by 26in. stroke. Wheels : 
Bogie, 3fti. 6in. diameter; driving and trailing, 6ft. 6in. diameter; 



EVOLUTION OF THE STEAM LOCOMOTIVE 



279 



wheel base, 22ft. lin. ; centre of bogie to centre of driving wheels, 9ft 
lOin. ; centre of driving to centre of trailing wheels, 9ft. Tubes No. 
254, Ifin. diameter outside. Heating surface: Tubes, 1,224 sq. ft.; 
fire-box, 126 sq. ft. ; total, 1,350 sq. ft. Fire-grate, 20 sq. ft. Weight 
of engine in working order, 47 tons. Weight of tender in working 
order, 38 tons. Tank capacity, 3,500 gallons. 

The North British Railway locomotive stock comprises about 30b 
engines, but many of these are in the A or duplicate list, and are not, 
therefore, included in the statutory returns. 




FIG. 117. HOLMES' S LATEST TYPE OF EXPRESS ENGINE, NORTH BRITISH RAILWAY 



The North British Railway tender locomotives are classified under 
seven headings four goods and three passenger. 

By a recent return the number of engines under each, head was: 



GOODS. 

18in. cylinder, 6 wheels coupled main line 

1st class, 6 wheels coupled ... 

2nd claaa, 6 wheels coupled ... ... ... ... ... 

(Of which 1 (No. 17a) is on the duplicate list.) 
3rd class, 6 wheels coupled ... 

(Of which 2 (18a and 250a) are on the duplicate list.) 

PASSENGER. 

1st class, 4 wheels coupled ... 
2nd class, 4 wheels coupled ... 



144 

267 



7:, 



121 

22 

(Of which 5 (268a, 269a, 394a, 395a, and 404a) are on the duplicate list.) 
3rd class, 4 wheels coupled ... ... ... ... ... 29 

(One (247a) is on the duplicate list.) 



280 EVOLlfTIOX OF THE ,v TEAM LOCOMOTIVE 

The locomotive works of the Lancashire and Yorkshire Railway 
are situate at Horwich, in the vicinity of Bolton, and are the newest 
of the immense assemblages of workshops and factories designated by 
the various railways as their "works," which have been erected by 
the principal railway companies. It is not, therefore, surprising to 
find that the Horwich works are quite equal to all, and exceed many 
other, of the railway establishments in the matter of modern machine 
tools, and in the general completeness of the undertaking. 

Mr. J. A. F. Aspinall is chief mechanical engineer of the Lanca- 
shire and Yorkshire Railway, and under his supervision the locomo- 
tive stock of the railway has been raised to a degree of excellence 
seldom equalled and never exceeded. 

This position has been attained because Mr. Aspinall has always 
shown a determination to introduce the best features of all kinds into 
his locomotive designs. The Joy valve gear is very extensively em- 
ployed in the construction of Lancashire and Yorkshire locomotives, 
and has always given excellent results on that line. 

For many years past the Lancashire and Yorkshire Railway has 
adopted the Ramsbottom system of water tanks, while the pick-up 
apparatus is actuated by a vacuum arrangement patented by 
Mr. Aspinall. The water troughs are situate at nine different places 

on the system viz : 

Better No. 

Horbury Junction : East end of Horbury Junction Station :-: 

Hoscar Moss : Between Hoscar Moss and Burscough Bridge .. ... 7 

Kirk by : Between Kirkby and Fa zakerly "> 

Lea Boad : Between Lea Boad and Salwick G 

Rufford : Between Bufford and Burscough North Junction 8 

Smithy Bridge: West end of Smithy Bridge Station 1 

Sowerby Bridge: WfSt end of Sowerby Bridge Tunnel 2 

Walkden : Between Moorside and Wardley and Walkden 4 

Whittey Bridge: West end of Whittey Bridge Station ... 9 

Very severe gradients are to be found on the Lancashire and 
Yorkshire Railway, many stretches of 1 in 50, between which rate of 
inclination and that of 1 in 100 very many banks exist, some oi 
which are of considerable length; whilst from Baxenden to Accring- 
ton the line falls 1 in 40 for two miles at a stretch, and at the 
same rate for 1J miles, from Padiham Junction to Padiham Station, 
and also for If miles at 1 in 40 from Hoddlesden Junction co 
Hoddlesden. From Britannia to Bacup the gradients are as follows: 

Fall 286 yards, 1 in 61. 

Fall 550 yards, 1 in 35. 

Fall 154 yards, 1 in 70. 

Fall 1,056 yards, 1 in 34. 



EVOLUTION OF THE STEAM LOCOMOTIVE 281 

But this bank is even eclipsed in severity by the Oldham incline' 
of 1 in 27 for three-quarters of a mile. All these stiff banks are 
worked by locomotive engines without the help of stationary engines. 

Every train which leaves Victoria Station, Manchester, in an east- 
ward direction, has to start off by ascending a serious incline of 1 in 
77, followed by another of 1 in 65, round a sharp S curve, on its way 
to Newton Heath, or else to ascend gradients towards Miles Platting of 
1 in 59 and 1 in 49. 

The locomotive stock consists of 1,333 engines. Of this number,. 
590 are of the standard types described below as being of the three 
leading types designed by Mr. Aspinall The balance is made up- 




FIG. 118. FOrR-WHEELS-COUPLED, SADDLE TANK ENGINE, LONDON 
AND NORTH WESTERN RAILWAY 

mainly of engines of older forms, which are gradually being replaced 
with engines of the standard type, though a large number of thes<? 
engines have been altered so as to require a boiler of one type only. 

The locomotive sheds of the Lancashire and Yorkshire Railway 
are situate and numbered as below : 

Newton Heath, No. 1 ; Low Moor, No. 2 ; Sowerby Bridge, No. 3 ; 
Leeds, No. 4 ; Mirfield, No. 5 ; Wakeneld, No. 6 ; Normanton, No. 7 - r 
Barnsley, No. 8; Knottingley, No. 9; Goole, No. 10; Doncaster r 
No. 11; , No. 12; Agecroft, No. 13; Bolton, No. 14; 

Horwich, No. 15; Wigan, No. 16; Southport, No. 17; Sandhills,- 
No. 18; Aintree Sidings, No. 19; Bury, No. 20; Bacup, No. 21; 
Accrington, No. 22; Burnley, No. 23; Skipton, No. 24; Lower 



282 



EVOLUTION OF THE STEAM LOCOMOTIVE 



Darwen, No. 25; Hellifield, No. 26; Lostock Hall, No. 27; Chorley, 
No. 28; Ormskirk, No. 29; Fleetwood, No. 30; Blackpool (Talbot 
Road), No. 31 ; and Blackpool (Central), No. 32. 




3 3 

ii 

Is 



We will now proceed to describe some of the types of Lancashire 
and Yorkshire locomotives. 

The "H," or Standard class of four-wheels coupled passenger 
engines is illustrated by engine No. 1,093 (Fig. 118A). The cylinders 



EVOLUTION OF THE STEAM LOCOMOTIVE 283 

are inside, and the axles also have the bearings inside. The principal 
dimensions are: 

Cylinders, 18in. diameter by 26in. stroke. 

Bogie wheels, 3ft. Ojin. diameter 

Coupled wheels (driving and trailing), 7ft. Sin. diameter 

Wheel base, 21ft. 6Jin. 

Centre of bogie to centre of driving wheel, 10ft. 2|in. 

Centre of driving to centre of trailing wheel, 8ft. 7in. 



Weight loaded (bogie), li 

(driving), 



13 tons 16 cwt. 



(trailing), 14 tons 10 cwt. 
Total, 44 tons 16 cwt. 
Boiler, 4ft. 2in. diameter, 10ft. 7|in. long. 
Firebox, 6 ft. long, 4ft. lin. wide, 5ft, lOin. high. 
Number of tubes, 220. 
Tubes (outside diameter), If in. 
Heating surface, tubes, l,108-73sq.-ft. 
firebox, 107-681sq.-ft. 
Total, l,216-41sq.-ft. 
Firegrate area, 18-75sq.-ft. 
Pressure of steam per sq. inch, 1601bs. 
Weight of 6-wheel tender, loaded, 26 ions 2 cwt. 2 qrs. 
Capacity of water tank of tender, 1,800 gallons. 
Fuel capacity of tender, 3 tons. 

The "A." or standard class goods engines have cylinders, boilers, 
heating surface, steam pressure, etc., the same as the "H" class of 
passenger engines just described;, whilst a similar pattern of tender 
is employed, the six-coupled wheels are 5ft. lin. diameter; the wheel 
base is : L. to D., 7ft, 9in. ; D. to T.,. 8ft. 7in. ; total, 16ft. 4in. 




FlO. 119. STANDARD 8-WHEEL PASSENGER TANK ENGINE, L. & T.R. 

Weight in working order: L., 13 tons 16 cwt. 2 qr. ; D., 15 tons; 
T., 13 tons 6 cwt. 2 qr. ; total, 42 tons 3 cwt. 

Tank engines are employed to work the trains between Man- 
chester and Blackburn, a distance of 24J miles, of which 13 miles 
are on rising gradients, and six on falling gradients, most of them 
being steeper than 1 in 100. The most serious gradients affecting the 



284 



EVOLUTION OF THE STEAM LOCOMOTIVE 



working of this line are those fromBoltonup towards Entwistle, where, 
for a mile and a quarter, the gradient is 1 in 72, and for the following 
44 miles is 1 in 74 ; a more serious incline than the celebrated one 
over Shap Fell. These tank engines are fitted! with the water 
pick-up apparatus which can be used when running either chimney 
or bunker in front. The trains each consist of thirteen coaches, which 
including the engine weigh about 250 tons. 

The engines (Fig. 119) have eight wheels viz., a pair of leading 
radial, two pairs of coupled, and a pair of trailing radial. The cylin- 
ders are inside, and have 26in. stroke, the diameter being 18in. The 
diameters of the wheels are : 

Radial, 3ft. 7f in. 

Coupled (driving and trai.ing\ 5ft. Sin. 

Wheel base, 24ft. 4in., divided as follows : Front radial wheel to centre 

of driving, 7ft. lOJin. : driving to rear coupled, 8ft. Tin. ; rear coupled 

to trailing radial, 7ft. 10in. 
Weight loaded (leading radial wheel;, 13 tons 10 cwt. 

,, ,, (driving), 16 tons 12 cwt. 

,, ,, (rear coupled), 15 tons 2 cwt. 

,, ,, (trailing radial), 10 tons 15 cwt. 
Totil, 55 tons 19 cwt, 




FIG. 120. OIL-FIRED SADDLE TANK SHUNTING ENGINE, LANCASHIRE 
AND YORKSHIRE RAILWAY 

The boiler, fire-box, etc., dimensions are the same as the " H n 
class. The tanks of these locomotives hold 1,340 gallons of water, and 
the bunkers two tons of coal. 



EVOLUTION OF THE STEAM LOCOMOTIVE 



285 



The above three classes form the leading types of locomotives of 
Mr. Aspinall's designing. Fir. 120 illustrates a four-wheel-coupled 
saddle tank locomotive designed by Mr. Aspinall, and fired with oil, on 
Holden's system. It is used for shunting at Liverpool. 




At the present time the locomotives of the Caledonian Railway 
hold first place in the popular mind for speed and hauling capacity. 
This result has been attained through the remarkable performances of 
the engines of the " Dunalastair" class, constructed at St. Rollox 



286 EVOLUTION OF THE STEAM LOCOMOTIVE 

Works from the designs of Mr. J. F. Mclntosh, the present locomotive 
superintendent of the Caledonian Kailway. 

These engines (Fig. 121) have been frequently described, but it is as 
well to recapitulate the leading dimensions. The cylinders are inside, 
18 Jin. diameter and 2 Gin. stroke. The engine is supported by a 
leading bogie, and by four-coupled wheels of 6ft. Gin. diameter. The 
bogie wheel base is 6ft. Gin. ; centre of bogie to driving wheel, 9ft. 
llin. ; D. to T., 9ft.; total length over buffers (engine and tender), 
53ft. 9fin. The weight in working order is : Engine bogie, 15 tons 
14 cwt. 3 qr. ; D., 16 tons; T., 15 tons 5 cwt. ; tender L., 12 tons 
13 cwt.; M., 13 tons 4 cwt.; and T., 13 tons 4 cwt. 2 qr. ; total, 
86 tons 1 cwt. 1 qr. 

The tractive force is 14,4001b. Water capacity of tender is 3,570 
gallons. The working pressure is 1601b. The leading feature of the 
engine consists of the large heating surf ace viz., tubes, 1,284.45 sq. ft., 
and fire-box, 118.78 sq. ft. To obtain this result the boiler has been 
"high pitched," giving the engine a rather squat appearance, and 
causing the driving wheels to appear to be of smaller diameter than 
is actually the case. 

An extended cab is provided for the protection of the driver and 
fireman. The splendid work performed by these machines has fre- 
quently been chronicled, the principal feature being the daily run 
from Carlisle to Stirling, 118 miles, in 123 minutes, without a stop; 
this trip includes the tremendous pull up the Beattock Bank, with a 
rise of 650ft. in ten miles. Yet Sir James Thompson, the general 
manager of the Caledonian Railway, isaid of this class of engine, 
" But, effective as it is, we are already improving upon it, and it will 
undoubtedly be superseded by our next type of engines." 

As Sir J. Thompson intimated, Mr. Mclntosh improved upon 
the above type, the result being the excellent " Dunalastair 2 " 
(Fig. 123). These fine engines also are employed to haul the heavy 
West Coast corridor trains between Carlisle and Glasgow, and Edin- 
burgh and the North. 

From Glasgow to Carlisle one of the engines hauls the 2.0 p.m. 
corridor train without a pilot throughout the journey, the weight of the 
train, excluding passengers, luggage, and tender of engine, is upwards 
of 350 tons. The dimensions are : wheels, 6ft. 6in., D. and T. coupled, 
with leading bogie; cylinders 19in. by 26in. Tender runs on two 
four-wheel bogies; water capacity, 4,125 gallons. The weights on. 



ETOirTION OF THE STEAM LOCOMOTIVE 



287 




288 



EVOLUTION OF THE STEAM LOCOMOTIVE 



wheels are as follows: engine bogie, 16 tons 6 cwt. ; drivino- wheels, 
16 tons 17 cwt.; trailing, 15 tons 17 cwt, total, 49 tons. Tender: 
front bogie, 22 tons 1 If cwt. ; hind b:gie, 22 tons 6 cwt. total, 
45' tons. Total weight of engine and tender in working order, 94 tons. 
'Total length over buffers (engine and tender), 57ft. 3|in. ; tractive 
force, 16,8401b. ; working pressure, I751b. Der sq. in. Heating sur- 
face: tubes, 1,381.22 sq. ft.; fire-box, 118.78 sq. ft. total, 1,500 sq. ft. 

Bogie wheel base, 6ft. 6in. ; centre of bogie to driving, 
10ft. llin. ; driving to trailing, 9ft.; distance between bogie 
centres of tender, lift. 3in. ; total tender wheel base, 16ft. 9in. 

Another new type of engine introduced by Mr. Mclntosh has 
5ftr. 6in. coupled wheels. It is a passenger-goods, or mixed traffic 
engine (Fig. 123), for working goods, mineral and heavy passenger and 



-Jl 




FIG. 123. SIX-WHEELS-COUPLED CONDENSING ENGINE, CALEDONIAN BAILWAY 

excursion trains through the Glasgow Central Underground Railway. 
Wheel base, L. to D., 7ft, 6in. ; D. to T., 8ft. 9in. ; cylinders, 18in. 
by 26in. ; six-wheeled tender ; water capacity, 2,800 gallons. 

Another good design of Caledonian Railway engines is the " Car- 
brook" (Fig. 124) class, constructed from Mr. D. Drummond's specific 
cation with a leading bogie, and four-coupled wheels of 6ft. Gin. 
diameter. The weight of these engines is : bogie, 14 tons 15 cwt. ; D., 
15 tons 4 cwt.; T., 15 tons; L., 10 tons 16 cwt, 2 qr. ; M., 14 tons 
6 cwt. 3 qr. Wheel base : bogie, 6ft. 6in. ; centre of bogie to D., 9ft. 
lOin. ; D. to T., 9ft. ; total length over buffers (engine and tender), 54ft. 
6in. Water capacity of tender, 3,560 gallons. The safety valve is 
located on top of the dome, an unsymmetrical practice which spoils the 
outline. There is also another type of Mr. Drummond's engines, with 



EVOLUTION OF THE STEAM LOCOMOTIVE 289 

cylinders, 18in. by 26in. stroke. Wheel base and water capacity as 
in the " Dunalastair " class ; but the weight and tractive force- are dis- 
similar. The former, on bogie, is 14 tons 13 cwt. 2 qr. ; D., 15 tons 
7 cwt. 3 qr. ; tender, L., 12 tons 13 cwt.; M., 13 tons 4 cwt. ; T., 13 
tons 4: cwt. 2 qr. ; total, 84 tons 6 cwt. 3 qr. The tractive force is 
12,9001b. 




FIG. 124. "CARBEOOK," ONE OF DRUMMOND'S EXPRESS ENGINES I 
FOR THE CALEDONIAN RAILWAY 

To give full details of all the 44 types of Caledonian Railway 
engines would be rather wearisome to the reader, so of the remaining 
classes, particulars only are appended : 

PASSENGER ENGINES WITH TENDERS. 

Diameter of Driving Wheels : 

5ft. 9in. four-coupled, with leading bogie. Cylinders, 18in. iy 

26in. stroke. 
7ft. single, with leading bogie and pair of trailing wheels. 

Cylinders, 18in. by 26in. stroke. 
7ft. four-coupled, with leading bogie. Cylinders, 18in. by 24m. 

stroke. (This is a rebuilt type of engine.) Tender only 

holds 1,880 gallons. 
7ft. four-coupled, with a small pair of leading wheels. Cylinders, 

I7in. diameter by 24in. stroke. No dome on boiler. 
6ft. Gin. four-coupled (D. and T.), with a small pair of leading 

wheels. Cylinders, 17in. by 24in. stroke. No dome to 

engine, and only four wheels to tender, with a water capar 

city of 1,428 gallons. 
6ft. D. and T. coupled, small leading whesls. No dome. 

Cylinders, I7in. diameter by 22in. stroke. Six-wheel tender. 

u 



290 EVOLUTION OF THE STEAM LOCOMOTIVE 

8ft. 2in. single, small leading and trailing wheels. No dome. 

Cylinders, 1 Tin. diameter by 24m. stroke. Six- wheel tender. 
6ft. D. and T. coupled, ^mall leading wheels, rebuilt by Drum- 

mond, with safety valve on dome. Cylinders, 18in. by 24in. 
5ft. D. and T. coupled, with leading bogie. Cylinders, 18in. 

by 24in. Four-wheel tender ; water capacity, 1,550 gallons. 
7ft. single. Cylinders, I7jin. by 22in. No dome. Four-wheel 

tender, 1,384 gallons. 




FIG 125. MCINTOSH'S 5FT. 9lN. CONDENSING TANK ENGINE, 
CALEDONIAN RAILWAY 

PASSENGER TANKS. 

.5ft. L. and D. coupled, trailing bogie; cylinders, 16in. by 22in. ; 

Drummond valve ; water capacity of tanks, 830 gallons. 
5ft. single (for use of officials) : cylinders, 9^in. diameter by 15in. 

stroke; well tank holds 520 gallons; bunker, 30 cwt. of 

coal ; wheel base : L. to D., 6ft. 6in. ; D. to T., 7ft. 6in. Weight : 

L., 7 tons 10 cwt. 3 qr. ; D., 11 tons 6 cwt. 2 qr. ; T., 7 tons 

16 cwt, 1 qr. ; tractive force, 2,4891b. 
5ft. D. and T. coupled; cylinders, l7Jin. by 22in. Water, 820 

gallons. 
5ft. 6in. L. and D. coupled; cylinders, 16in. by 20in. Water, 450 

gallons. 
4ft. 6in. L. and D. coupled with trailing bogie ; cylinders, 18in. by 

22in. Water capacity, 950 gallons. 



EVULI'TIOX OF THE STE < M LOCOMOTIVE 291 

3ft. Sin. L. and D. coupled, and pair of trailing wheels; cylinders, 

14in. by 20in. stroke. The saddle tank holds 800 gallons. 
5ft. Sin. radial L. and T. wheels, and 4 coupled wheels (eight wheels 

in all); cylinders, I7|in. by 22in. Water in side tanks, 1,200 

gallons. Coal in bunker, 3 tons. 

5ft. D. and T. coupled with leading bogie ; cylinders, 17in. by 24in. 
5ft. 9in. L. and D. coupled, with trailing bogie ; cylinders, ISin. by 

26in. This class is fitted with condensing apparatus. (Fig. 125.) 

GOODS ENGINES WITH TENDERS. 

6ft. D. and T. coupled, with pair of leading wheels. Cylinders, ISin. 

diameter, by 24in. stroke. Six-wheel tender ; water capacity, 

1,840 gallons. 
5ft. six-wheels coupled. Cylinders ISin. by 26in. 6-wheel tender; 

water capacity, 2,500 gallons. 

The following engines have no domes : 
5ft. 6-wheels coupled. Inside cylinders, I7in. by 24in. 6-wheel 

tender ; water capacity, 1,800 gallons. 
5ft. (mineral engine) L. and D. coupled, small trailing wheels, no dome. 

Cylinders, I7in. by 24in. 4-wheel tender; 1,542 gallons. 
5ft. 6-wheels-coupled mineral engine. Wheel base : L. to D., 5ft. 6in. ; 

D. to T., 5ft. 6in. ; all wheels under boiler barrel. Cylinders, 

ISin. by 24in. 6-wheel tender; water capacity, 1,840 gallons. 
5ft. 6-wheels coupled mineral engine. Cylinders I7in. by 24in. 

4-wheel tender; water capacity, 1,383 gallons. 
5ft. 6in. L. and D. coupled and small pair trailing wheels ; insido 

cylinder, 16in. by 20in. 4-wheel tender. 
4ft, Sin. L. and D. coupled, mineral engine. Cylinders, I7in. by 20in. 

4-wheel tender; water capacity, 1,000 gallons. 

(A similar class of engines has cylinders I7in. diameter by ISin. stroke.) 
5ft. D. and T. coupled, with pair of small leading wheels. Cylinders, 

I7in. by 24in. 4-wheel tender; 1,545 gallons. 
5ft. D. and T. coupled, with small leading wheels. Cylinders, 17in. by 

20in. 6-wheel tender; water capacity, 1,700 gallons. 

MINERAL TANK ENGINES. 

4ft. 6in. 6-wheels coupled, saddle tank, holding 1,000 gallons of water; 
safety valves on dome ; cylinders ISin. by 26in. 

U2 



292 EVOLUTION OF THE STEAM LOCOMOTIVE 

4ft. Gin. 6-wheels coupled, side tanks, with condensing apparatua ; 

cylinders, 18in. by 26in. 

4ft 6in. 6-wheels coupled; saddle tank; cylinders, 18in. by 26in. 
4ft. 6-wheels coupled; saddle tank, 1,000 gallons; cylinders, 18in. by 

22in. stroke. 
4ft. 6-wheels coupled; saddle tank, 940 gallons. Cylinders, I7in. 

by 20in, No dome. 
3ft. Sin. 4-wheels-coupled ; wheel base, 7ft.; saddle tank, 800 gallons. 

Cylinders, 14in. by 20in. 
There is a similar class of engine built by Neilsons, the difference 

being in the weight. That of the former is, on leading axle, 

13 tons 14 cwt. 1 qr. ; on driving axle, 13 tons 13 cwt. 1 qr. 
Weight of Neilson's class: L., 13 tons 10 cwt. 3 qr. ; D., 13 tons 

9 cwt. 1 qr. 
3ft. Sin. 6-wheels-coupled, saddle tank; water capacity, 900 gallons. 

Drummond's safety valves. Cylinders, 14in. by 20in. stroke. 
Lastly, a class of 4- wheel engines,' with coupled wheels, 3ft. 6in. 

diameter; side tanks hold 500 gallons. No" dome, cab, or weather 

board; wheel base, 6ft. 3in. Cylinders, 14in. diameter, 22in. 

stroke. 




FIG. 126. "No. 143," TAFF VALE RAILWAY INCLINE TANK LOCOMOTIVH 

Engine "No. 143" (Fig. 126) is one of three peculiar locomotives, 
specially constructed for working on the Pwllyrhebog Incline, of 1 in 13, 
on the Taff Vale Railway. The fire-box and roof slopes backwards, so 
that when the engine works bunker first up the incline, the water 
Is level over the top of the fire-box. She is fitted with, two draw- 



EVOLUTION OF THE STEAM LOCOMOTIVE 293 

bars for attaching a wire rope. This rope is coupled to a low draw- 
bar under the drag-plate, so as to keep the rope below the axles of 
the wagons, which follow the engine down the incline, or are pushed 
up before the engine. " 143" has cast-iron " Sleigh" brakes acting on 
the rails, in addition to the usual steam brakes on the wheels. The 
dome is placed on the fire-box, and the regulator is within it, so as 
to ensure dry steam when working on the incline. Wheels, 5ft. 3iu. 
diameter. Cylinders, 17 Jin. by 26in. Weight, 44 tons 15 cwts. 




A FAVOURITE LOCOMOTIVE OF THE ISLE OF WIGHT 
CENTRAL RAILWAY 



CHAPTER XV. 

Great Western " convertible " locomotives The value of names in locomotive prac- 
tice Water troughs on the G.W.R. Dean's 7ft. Bin. singles His "Arm- 
strong" class An extension smoke-box on the G.W.R.; the '* .Devonshire " 
class 7ft. " singles : '- " 2202 " and "3225," four-coupled G.W. engines The 
" Barrington " Great Western passenger tanks "Bull Dog" design "No. 
36," Gredt Western Railway A six- wheel eoupleu goods engine with a. 
leading bogie Ivatt's advent c.n the Great Northern, and his innovations 
"Domes" to the fore New goods and tank engine,? Rebuilt " Stirlings " 
Ivatt's inside cylinder four-coupled bogie engines His chef d'ceuvre 
"990" A ten- wheel tank on the G.N.R. " 266," the latest Great 
Northern engine Possibilities of the future Great North of Scot- 
land locomotives Manson's designs James Johnson's tank and 
tender engines Furness* engines, passenger and goods The 1896 "express"' 
design Pettigrew's new goods engines Highland Railway engines A 
Great Central Railway locomotive Some Irish Jocjuio lives Beliii^t raid 

Northern Counties Compounds The " Restrevor " class, G.N. (I.) Great 

Southern and Western standard passenger design A locomotive for an 
Ivish "light"' railway. 

The broad gauge having been finally abandoned on the U.W.R. in 
May, 1892, it became necessary to re-arrange the locomotive power. 
Previous to that date Mr. W. Dean, the G.W.R. Locomotive Superin- 
tendent, had constructed at Swindon several six-wheeled express loco- 
motives (Fig. 127), with " single" driving wheels, 7ft. Sin. in diameter, 
inside cylinders 20in. in diameter, and a stroke of 24in., and weighing 
44 tons 4 cwt., of which 13 tons 4 cwt. was on the leading axle. 

This class of engine was designed to work the West of England 
expresses between London and Newton Abbot, consequent upon the 
conversion of the gauge, and the locomotives were therefore built upon 
strictly narrow-gauge dimensions, but some few of them were worked 
on the West of England expresses whilst the gauge was yet broad, anr! 
for this purpose the wheels were fixed outside the framing. In this 
condition they had a very curious and ungainly appearance, intensified 
by the squat chimney, large dome, and bulged fire-box covering. 

After the alteration of the gauge had been effected, and the wheels 
of the engines of this class had been fixed in their normal position, 
their appearance was considerably improved, but there still remained 
about the locomotives a somewhat indescribable want of symmetry and 
unison of outline. However, it was decided to substitute a bogie for the 
pair of leading wheels, whilst the diameter of the cylinders was reduced 
to 19 irches. These alterations, coupled with other minor improve- 
ments, added to the admittedly good qualities of the engines as loco- 
motive machines, soon caused the class, thus improved, to gain a 



EVOLUTION OF THE STEAM LOCOMOTIVE 



295 



high place in the estimation of both experts and the railway public. 
The amount of bright brass about the engines and the names 
carried by them mostly those of famous broad-gauge engines, or 
popular broad and narrow-gauge, Great Western Railway officials 
have also added to the prestige of the design. Let cynics say what 
they will, one feels more interest for, say, the "Rover-' than he can 
ever expect to for plain " No. 999." 

The adoption of water troughs on the Great Western Railway, 
and the addition of the "-pick-up" apparatus to the tenders of these 
engines, enables the Great Western Railway to perform many daily 
runs for length and speed that, a few years back, would rightly have 
been considered quite phenomenal. Happily, we improve with giant 
strides in matters locomotive at the tail end of the 19th century. 




Photo] IF- Moore 

FIB. 127. 7FT. SIN. "SINGLE" CONVERTIBLE ENGINE, 
GREAT WESTERN RAILWAY 

With the adoption of the normal gauge over the whole of the 
{Jreat Western Railway system, engines of this class are now used on 
the expresses on all sections where the character of the gradients 
allows such engines to be run with proper economy. Under these 
circumstances, it is not surprising to learn that additional batches of 
engines of Mr. Dean's 7ft. Sin. " single " design (Fig. 128) are being 
added to the Great Western Railway locomotive stock at not infre- 
quent intervals. At the present time, there are 71 of these engines 
at work, and nine others under construction probably a larger number 
of one class of modern express locomotives than can be found elsewhere. 



EVOLUTION OF THE STE 'A W LOCOMOTIVE 




EVOLUTION OF THE STEAM LOCOMOTIVE S97 

The huge pipe for delivering the feed-water to the boilers of chese 
engines, formerly placed in a conspicuous position,, has been removed, 
an alteration that has added much to the beauty of outline of these 
fine-looking locomotives. 

Mr. Dean has constructed a class of four-coupled engines, with a 
leading bogie, known as the " Armstrong " class. In its salient 
features, the design is a modification of the 7ft. 8in. single class 
described above, but naturally several of the dimensions are dis- 
similar in the two classes. " Armstrong " is No. 7, " Gooch " (Fig. 
129), No. 8, "Charles Saunders," No. 14, and "Brunei," No. 16. 




FIG. 129." GOOCH," A 4-COUPLED EXPRESS ENGINE, 
GREAT WESTERN RAILWAY 

Immediately subsequent to the change of gauge in May, 1892, a 
class of tank engines, with wheels four-coupled in front and a trailiig 
bogie, was built for working the fast passenger traffic west of Newf-on 
Abbot. The bogies of these engines were fitted with Mansel wheels 
quit an exceptional practice in locomotive building. 

Mr. Dean has since designed another class of locomotive to work 
the fast train traffic over the severe gradients and curves so common 
to the Great Western Railway main line west of Newton Abbot. 

These engines are popularly called the "Devonshire" or "Pen- 
dennis Castle" class (Fig. 130), after the name given to the first engine 



298 



EVOLUTION OF THE STEAM LOCOMOTIVE 




EVOLUTION OF THE STEAM LOCOMOTIVE 299 

constructed on the plan. A prominent feature of the design is the 
" extension " smoke-box a feature copied from modern American 
practice. Before constructing the " Pendennis Castle," Mr. Dean had 
fitted another engine No. 426 with an extended smoke-box, and 
the result of the trials made with this locomotive satisfied the Great 
Western Railway locomotive superintendent as to the advantages of 
tho arrangement. 

The cylinders of this class are 18in. diameter, the stroke being 
26in. The coupled wheels (D. and T.) are 5ft. 7|in. diameter, that of 
the (leading) bogie being 3ft. 7Jin. The use of Mansel wheels has 
also been adopted both for the bogies and the tenders of the loco- 
motives of tfcis class. The frames are double, and are specially con- 
tracted at the smoke-box end to allow sufficient play to the 
bogie wheels. Both inside and outside bearings are provided for the 
driving axle. The boiler is of steel, the heating surface being : Tubes, 
1,285.58 sq. ft.; fire-box, 112.60 sq. ft,; steam, pressure, 1601b. ; 
grate area, 19 sq. ft.; weight of engine, 46 tons, of which 15 tons 
7 cwt. is on the driving axle, 17 tons on the bogie, and 13 tons 
3 cwt. on the trailing (coupled) axle. The tender holds 2,000 gallons 
of water, and weighs, loaded, 24 tons. Ten engines of this design 
were originally constructed at Swindon viz. : 

3252 Duke of Cornwall. 3255 Cornubi*. 3859 Lizard. 

3253 Pendennis Castle. 3256 Excalibur. 32<>0 Merlin. 

3254 Boscawen. 5257 Guinevere. 3261 Mount Edgcumbe. 

3258 King Arthur. 

These proved so satisfactory in performing the peculiar duties 
required from passenger engines on the West of England main line 
of the Great Western Railway that a second batch of twenty was put 
in hand. These commenced running in the early months of 1898. 
They are named and numbered as follow : 

3262 Pow.lcrh.m. 3272 Amyas. 3282 Ma istowe. 

3263 Sir Lancelot. 3273 Armurel. 3283 Mounts Bay. 
S2t St. Anthony. 3274 C rnishman. ii'st Newquay. 
>>>>.' St. German*. 3275 Chough. 3285 St. Erth. 

3266 St. Ives. 3276 D-Utmoor. 3286 St Just. 

3267 St. Michael. 3277 Earl of Devon. :!2*7 St, Agnes. 

3268 Tamar. 8278 Eddystonp. 3288 Tresco. 

3269 Tintagel. 3279 Exmoor. :;i's! Trefusi. 

3270 Trevithick 3280 Falmouth. 3290 Torbay. 

3271 Tre Pol and Pen. 3281 Fowey. 3291 Tregenna. 

Several of these engines have the tenders fitted with the water 
pick-up apparatus. 

The names, it will be observed, should specially please the patrons 



300 EVOLUTION OF THE STEAM LOCOMOTIVE 

of the Great Western Railway residing in Devon and Cornwall, and 
help to palliate the keen regret with which the abolition of the broad- 
gauge was felt in those counties. 

Among types of Great Western locomotives, one may be mentioned 




FIG. 131. SINGLE EXPRESS ENGINE, 6-WHEEL TYPE, GREAT WESTERN 
RAILWAY 

the 7ft. "singles" (Fig. 131), largely used for hauling the express 
trains on the Birmingham and Northern lines. The cylinders are 
18in. diameter, the stroke being 24in. Heating surface, 1,250.31 

square feet. 

Many of the passenger trains on the Gloucester and Weymouth 
sections are worked by the 6ft. Gin. four-coupled engines, illustrated 
by engine 2,202 (Fig. 132). The leading dimensions of this class are : 




FIG. 132. 6FT. 6IIC. 4-COUPLED PASSENGER LOCOMOTIVE, GREAT WESTERN 
RAILWAY 

Cylinders, I7in. diameter ; stroke, 24in. ; heating surface, 1,363.5 sq. ft. 
Weight of engine and tender, in working order, 59 tons 8 cwt. 

North of Wolverhampton, for working the West to North expresses, 



EVOLUTIOX OF THE STEAM LOCOMOTIVE 



301 



and for other fast trains in the North Western district of the 
G.W.R., the engines represented by 3,225 (Fig. 133) are largely used. 




FIG. 133. -6FT. 4-COUPLED PASSENGER ENGINE, GREAT WESTERN 
RAILWAY 

This class has cylinders 18in. by 24in. stroke; leading wheels 4ft. dia- 
meter, and coupled, driving, and trailing wheels, 6ft. diameter. The 
heating surface totals to 1,468.82 sq. ft. ; and the weight of engine 




FlO. 134. "BARRINGTON," NEW TYPE OF 4-COUPLED ENGINE, - 
GREAT WESTERN RAILWAY 

and tender, including the load of 4 tons of coal and 3,000 gallons of 
water, amounts to 74 J tons- 

' Barrington" (Fig. 134) is one of Mr. Dean's latest type of express 



302 EVOLUTION OF THE STEA M LOCOMOTIVE 

passenger engine. These powerful locomotives are somewhat of the 
"Devonshire" type, having an extended smoke-box, whilst the "rJel- 
paire " fire-box is also introduced. In the framing, it will be noticed, 
early G.W. practice is reverted to. The cylinders are 18in. by 26in. 
stroke. The bogie wheels are 4ft., and the coupled wheels 6ft. Sin. in 
diameter. The engine weighs 51 tons 13 cwt. ; the tender, with the 
same amount of water and coal as " 3,225 " class, 32 J tons. 

A good deal of the G.W. passenger trains are hauled by smart 
little six-wheel (four-coupled) tank engines, which are specially noted 
for getting away quickly, and immediately attained high speeds. 
"No. 576" (Fig. 135) represents a coupled-in-front engine of this de- 
scription, but the more generally-known Great Western Railway pas- 




FlG. 135. 4-COUPLED-IN-FRONT PASSENGER TANK ENGINE, G.W.R, 

senger tank engines have the driving and trailing wheels coupled ; 
these are 5ft. diameter, the cylinders being 16in. diameter by 24ft. 
stroke. 

Mr. Dean's latest creation for the Great Western Railway is named 
"Bull Dog," No. 3,312, and the design will be known as the "Bull 
Dog" class. Except that the bogie wheels have spokes, the wheels, 
framing, and motion are similar to the "Devonshire" class (Fig. J.30). 
The boiler is of gigantic proportions; the fire-box is of the Belpaire 
type, and projects over the top and sides of the boiler barrel. The 
smoke-box is extended, and steaming reversing gear is employed, 
whilst another improvement, Davies and Metcalfe's patent exhaust 
steam injector, is fitted to the engine- and is being extensively adopted 
on Great Western Railway locomotives. The name-plates are on the 
sides of the fire-box; the clack valves are below the boiler barrel, 



EVOLUTION OF THE STEAM LOCOMOTIVE 303 

behind the smoke-box. The cab of the "Bull Dog" extends to the 
edge of the footplate, with a door in the front on the fireman's side. 

Before closing these remarks on modern Great Western Railway loco- 
motives, some description of No. 36 is necessary. Here again we have 
an adaptation of American practice a six-wheels-coupled engine, with a 
leading bogie, and an extension smoke-box. The cylinders are inside, 
20in. diameter by 24in. stroke, with the steam chests below them. 
The driving wheels are 4ft. 6in. diameter, the bogie wheels only 2ft. 
8in. diameter. All the wheels have outside bearings, and the driving 
wheels have inside bearings in addition. The boiler con- 
tains 150 "Serve" tubes of 2 Jin. diameter. The total 
heating surface is 2,385 sq. ft. ; steam pressure, 1651b. ; grate area, 
35 sq. ft. The weight is as follows: On bogie, 12 tons 6 cwt.; 
leading coupled wheels, 15 tons \2 cwt.; driving wheels, 16 tons 
11 cwt.; and trailing wheels, 15 tons 1 cwt. Total weight of engine, 
59 J tons; of tender, 32 tons; together, 91 J tons. The tender is 
fitted with a water pick-up apparatus. This locomotive has been 
employed in hauling goods trains for many months past, and it is 
stated, to have hauled a train weighing 450 tons through the Severn 
Tunnel despite the severe gradients and length in ten minutes, 
although for such a load two goods engines of the usual Great Western 
design would be required, and they would take 18 minutes to perform 
the trip. 

Consequent upon the death of the late Mr. Patrick Stirling one 
of the best locomotive superintendents of his time the directors of 
the Great Northern Railway appointed Mr. H. A. Ivatt to the supreme 
command at Doncaster. Mr. Ivatt received his early training in the 
science of locomotive construction at Crewe, and left the Great Southern 
and Western Railway (Ireland), where he was locomotive superinten- 
dent, to succeed Mr. P. Stirling on the Great Northern ! ail way. 

Mr. Ivatt, having decided opinions of his own relative to locomotive 
design, soon set to work to introduce his ideas on the Great Northern 
system ; so that after many years more than two decades of dome- 
less locomotives, Doncaster awoke one morning to find a Stirling 
8ft. "single" fitted with a steam dome encased in a green-painted 
cover. It was certainly a great surprise the colour especially, for 
many had hoped to see bright brass but those interested survived the 
shock, and waited to see some engines of Mr. Ivatt's design on the 
Great Northern Railway. 



504 



EVOLUTION OF THE STEA If LOCOMOTIVE 



Several engines, with pronounced Ivatt features, were soon run- 
ning, but the main designs of all of them are cast after distinctly 
Stirling models, as they were already under construction at the time 
of Mr. Ivatt's appointment. 

In the 1070 class (four-coupled, six- wheeled engines) we find that 
the dome and cab, amongst external signs, are the work of the new chief 




FIG. 136. "No. 1312," ONE OF MR. 1VATTS (1073) SMALLER CLASS OF 
4-COUPLED BOGIE ENGINES FOR THE GREAT NORTHERN RAILWAY 

at Doncaster; whilst those of the 1073 design have his leading bogie, 
splasher over the coupled-wheels, dome, and cab. 

Coming to " No. 34," a rebuilt 8ft. " single," Mr. Ivatt is responsi- 
ble for the dome, cab, and safety-valve casing, whilst in the 1206, six- 
coupled saddle-tanks, we again find the dome and new pattern valve 
casing. 

Readers will notice that we have only referred to the apparent 
details that are attributed to Mr. Ivatt, but, by reference to the 
appended tables of dimensions, they will find that several alterations 
that do not so readily meet with notice have been made in other 
matters connected with the Great Northern locomotives. 



OFT. Gix. FOUR-WHEELS COUPLED ENGINE, No. 1070. 
CYLINDERS. 



Diameter 

Stroke * 



WHEELS. 

Driving ; 

Trailing 

Leading 

WHEEL CENTRES. 

From centre of trailing to centre of driving wheels 
From centre of driving to centre of leading wheels 
Total wheel base ... 



26in. 



6ft. 6in. diameter. 
6ft. 6in. diameter. 
4ft. Oin. diameter. 

8ft. Sin. 
9ft. 8in ] 
17ft. llin 



EVOLUTIOX OF THE STEA M LOCOMOTIVE 305 

BOILEE. 

Length of barrel 10ft. lin. 

Diameter of barrel 4ft 5in. 

Length of lirebox casing 5ft. 6in. 

HEATING SUEFACE. 

Tubes l,020'7sq.-ft. 

Firebox 103'lsq.-ft. 

Total l,123-8sq -ft. 

Crate area 17-8sq.-ft. 

Tubes 215 IJin. diameter cutside 

OFT. Oix. FOUR- WHEELS COUPLED BOGIE ENGINE, No. 1073. 

(Illustrated by FIG. 13G.) 

CYLINDEES. 

Diameter 17$in. 

Stroke : 26in. 

V HEELS. 

Driving 6ft. Gin. diameter. -, 

Trailing. 6ft. 6in. diameter. 

Bog e 3ft. 6in. diameter. 

WHEEL CENTEES. 

From centre of trailing to centre of driving wheels 8ft. 3in. 

From centre of driving to centre of bogie pin ... 9ft. !tin. 

Centres of bogie wheels 6ft. 3in, 

Total wheel base 21ft. Sin. 

BOILEE. 

Length of barrel 10ft. lin. 

Diameter of barrel 4ft. Sin. 

Length of firebox casing 5ft. 6in. 

HEATING SUEFACE. 

Tubes l,020-7sq.-ft. 

Firebox 103-lsq.-ft. 

Total l,123-8sq.-ft. 

Grate area 17 l 8sq.-ft. 

Tubes 215 Ifin. diameter outside. 




FIG. 137. THE LATEST TYPE OF 6FT. 6iN. COUPLED ENGINE, 
GREAT NORTHERN RAILWAY 

-Fig. 137 represents the larger and later type (just out) of the 
6ft. Gin. four-coupled engine, with a leading bogie, on the Great 
Northern Railway. In these engines the boiler diameter has been 
augmented by 3in., so that it bulges out over the splashers; the 
heating surface is increased to 1,250 sq. ft., while the fire-box is 
greatly enlarged, having 120 sq ft. This enlargement of the fire- 

x 



306 



EVOLUTION OF THE STEAM LOCOMOTIVE 



box has involved a lengthening- of the side rods and coupled-wheel 
base by 9in. The fire-grate area is 20. Sin. , instead of 17.8in., in the 
smaller engines. The chimney, which is much shorter owing to the 
height of the boiler, is built up in three pieces. 

SFT. SINGLE PASSENGER ENGINE No. 34. 



CYLINDERS. 



Diame er 

Stroke 



Driving 
Trailing 
Bogie 



WHEELS 



WHEEL CENTRES. 



From centre of trailing whrel to centre of driving 
From centre of driving wheel to centre of bo^ie pin 

Centres of bogie wheels 

Total wheel b.ise 

BOILER. 

Length of barrel 

Diameter of barrel ... ... ... 

Length of firebox casing next to barrel 

Length of firebox casing at bottom 



8ft. Oin. .liameter. 
4ft. Gin. diameter. 
3 ft. loin, diameter. 



Oft Oin. 
l()ft. 9in. 

6it. 6in. 
23ft. 3in. 



lift. Jin. 
4ft. 9in. 
6ft. Din. 
7fi. 2in.. 



HEATING SURFACE. 



9SOsq.-ft. 

; 114-2sq.-ft. 

1, 94--2sq -ft. 

23' Gsq-ft. 

... ISi-ljjin. diameter outside. 

4PT. GIN. SIX-WHEELS COUPLED SADDLE TANK ENGINE, No. 1206. 



Tubes 
Firebox 

Totil 
Grate ate i 
Tubes 



CYLINDERS. 



Diameter 

Stroke 



Driving 
Trailing 
Leading 



WHEELS. 



18in. 
26in. 



4ft. Gin. diameter. 
4ft Gin. diamster. 
4ft Gin. diameter-. 



WHEEL CENTRES. 

From centre of trailing to centre of driving wheels 8ft. 3in. 

From centre of driving to centre of leading wheels Tt't. :-Jin. 

Total wheel base i:.ft. Gin. 

BOILER. 





4ft 5in 




5ft Gin 


Tubes 


HEATING SURFACE. 
1 061-13sq -ft 


Firebox 
Total 


103-lsq.-ft. 
] 164-2-jisq -ft 


Grate area 
Tubes 


17-8sq.-ft 
215 l^in. diameter outriue. 



Mr. Ivatt's express passenger engine No. 990 (illustrated as a 
frontispiece to this volume) is quite a new departure in British loco- 
motive practice, having a leading bogie, four-coupled wheeh in front 



EVOLUTION OF THE STEAM LOCOMOTIVE 



307 



of the fire-box, and a pair of trailing wheels under the foot-plate. 
The dimensions are : 



OYLINDKRS. 



IHaim-ter 
Stroke 



Trailing- 
Coupled 



WHEELS. 



WHEEL CENTRES. 



Total wheel base 



BOILER. 



Length of barrel between tube plates 

Diameter of barrel 

Length outsTde firebox casing 



Tubes 
Firebox 



HEATING SURFACE. 



Total 

Grate Area 
Tubes 



19in. 
24in. 



3ft, 6in. 
6ft. 6in. 
3ft. 6in. 



From centre of trailing to centre of driving wheels 8ft. Oin. 

Centres of coupled wheels . 6l't. lOin. 

Centre of leading coupled to centre of trailing 

bogie wheel 5ft. Sin. 

Centres of bogie wheels 6ft. 3'n. 



26ft. I'm. 



13ft. Oin. 
4ft. Sin. 
8ft. Oin. 



l,302sq.-ft. 
140sq.-ft. 

l,442jq.-ft 
26.7sq.-ft. 

191 2in. external diameter. 



Mr. Ivatt has also designed a new class of 10- wheel tank engines 
for the G.N.R., the leading dimensions being 

CYLINDERS. 



Diameter 

Stroke 



Coupled 
Trailing 
Bogie 



WHEELS. 



26in. 



5ft. 6in. 
3ft. 6in. 
3ft. 6in. 



. WHEEL CENTRES. 

From centre of trailing: to centre of back coupled 6ft. Oin 

From centre of back coupled to centre of driving 8ft. 3'n. 

From centre of driving to centre of trailing bogie 6ft. 9in. 

Centres of bogie wheels . . . . . . . . 6ft. 3in. 

Total wheel base 27ft. 3in. 



BOILER. 



Length of barrel 
Diameter of barrel 
Length of firebox casing 



10ft. lin. 
4ft. 5in. 
5ft. 6in. 



HEATING SURFACE. 



Tubes . . 
Firebox 


l,020.7sq.-ft. 
103sq -ft 




Total 


11237sq-ft 




Grate Area 
Tubes 


17.8sq.-ft. 
215 l|jn external 








X 2 



308 



EVOLUTION OF THE STEAM LOCOMOTIVE 



There now remains to be described Mr. Ivatt's newest engine, a 
7ft. Bin. single-wheeler with leading bogie, a large boiler, lift. 4in. 




a l 

CO Q 
W 

^ 



o Q 

J5 51 



long and 4ft. 5in. diameter, giving 1,268 sq. ft. of heating surface, 
I751b. steam pressure, and a fire-box 7ft. long. The boiler centre 
stands 8ft. Sin. above the level of the rails. The cylinders are 18in. 



EVOLUTION OF THE STEAM LOCOMOTIVE 309 

diameter by 26in. stroke. The grate area is 23 sq ft. As this engine 
is only just out of the Doncaster shops none of her performances have 
as yet been recorded, but if she prove as good as she looks the Great 
Northern Railway will have a valuable addition to its already 
numerous '' single " locomotives. 

And now, perhaps, may be ventured an opinion on Mr. Ivatt's 
innovations in Great Northern locomotive practice. In the first place, 
from an aesthetic point, there can be no two opinions that a dome 
greatly improves the appearance of a locomotive, but one of bright 
brass is infinitely superior to one covered with green paint. "To 
win the eye is to win all," and plenty of bright brass about a loco- 
motive- is certainly an attraction ; a large amount of the popularity of 
the Great Western engines is due to the fine display of brass. The 
same reason that causes us to prefer a brass dome makes us sorry to 
see the Stirling brass casing of the safety valve give place to Mr. Ivatt's 
design. Green paint undoubtedly is a good thing, but then you can 
have "too much of a good thing." Again, a curved splasher for 
coupled wheels, following the outlines of both wheels, looks much 
neater than the design used with Class 1073. The bogie is decidedly 
an improvement; so is an extended cab, but graceful outlines might 
be used in connection with the latter. Mr. Ivatt has certainly intro- 
duced some decided improvements into the composition of the Great 
Northern Railway locomotives, but the tout ensemble might be more 
pleasing ; a few alterations in matters of detail would give observers 
a more appreciative opinion of modern Great Northern Railway 
engines. 

Now water-troughs are so much in fashion, it should not be diffi- 
cult to find suitable locations for them on the Great Northern system, 
and with a. double-bogie tank engine, with outside cylinders, a 9ft. 
or larger driving wheel, York ought to be reached in less than three 
hours from King's Cross, and without an intermediate stop. Will 
the 19th century see such an achievement? We hope so, but fear to 
prophesy; its sands are almost run. 

The Manson engines of the Great North of Scotland Railway de- 
serve notice. As long ago as 1878 and 1879 it was decided to place 
heavier and more powerful engines on that railway. The engines 
weighed 41 tons 5 cwt. each, and the tender 28 tons 5 cwt. in working 
order. The working pressure was 1501b. per square inch. 

In 1884 Mr. Manson, who succeeded Mr. Cowan, got sonie six- 



310 EVOLUTION OF THE STEAM. LOCOMOTIVE 

wheel coupled inside cylinder tank engines from Kitson and Co., of 
Leeds. The following are the principal dimensions, viz. : 

Cylinders 16in. by 24in. 

Coupled wheels 4ft. 6in. diameter. 

Wheel base 13ft. 8in. 

Tubes 140 Ifin. external diameter 

Heating surface Tubes 680s<i.-ft. 

Heating surface Firebox 66sq.-ft. 

Total 756sq.-it. 

Working steam pressure . . Ituib. per sq.-in. 

Weight in working order 37 tons 7 cwts. 

- ^> 

In the same year Messrs. Kitson' and Co. also supplied some four 
coupled passenger engines, with leading bogie and a six- wheeled tender. 

The cylinders are " inside," and the bogie is Kitson's swing link 
type, which this Company has used since 1884. These engines were 
delivered with a brick arch in the fire-box, but this was afterwards 
taken out, and air tubes put into the front and rear of the fire-box, so 
as to consume the smoke. The principal dimensions are : 

Cylinders 17iin. by 26in 

Coupled wheels 6ft. Oin. diameter. 

Bogie wheels - - 3ft. Oin. diameter. 

Tender wheels (6), 3ft. 9in diameter. 

Wheel base of engine - 20;t. Sin. 

Wheel base of tender lift. Oin. 

Total wheel base of engine and tender .. ', 40ft. 3|in. 

Tubes . . . . . 189 IJin. external diameter. 

Heating surface Tubes .. .. .. .. 946sq.-ft. 

Heating surface Firebox 90sq.-ft. 

Total l,036sq.-ft. 

Tank capacity 2,000 gallons. 

Working steam pressure : 1401b. per sq.-in. 

Weight in working order Engine . . . . 37 tons 2 cwts. 

,, Tender .. . .. 29 tons cwts. 

Total . . . . 66 tons 2 cwts. 

In 1888 Mr. Manson brought out his engine with inside cylinders, 
having the valves placed on the top, which were of the balanced type 
introduced by Mr. Cowan. The valves were driven by the ordinary 
Stephenson link motion working on a rocking shaft. In other respects 
the engine very much resembled those just described, except that the 
engine and tender were coupled by a central bar and one solid central 
rolling block in place of side spring buffers. 

The cylinders were 18in. by 26in. and the coupled wheels 6ft. 0in. diameter 

The engine weighed 41 tons 9 cwts. 

The tender weighed 29 tons cwts 

In working order. 
They were built by Messrs. Kitson and Co. 

In 1890 Mr. Manson increased the capacity of the tender to 3,000 
gallons, and in doing this introduced a bogie tender. The tender was 



EVOLUTION OF THE STEAM LOCOMOTIVE 



311 



carried on eight wheels 3ft. 9 Jin. diameter. The four trailing wheel 3 
were fixed, and the four leading carried a bogie similar to that on the 
engine. 




The wheel base of the tender was 16ft. 6in., and the weight in 
working order 38 tons. The engine for these tenders was the same as 
that- just described. These were built by Stephenson and Co. 



312 EVOLUTION OF THE STEAM LOCOMOTIVE 

In 1893 Mr. Janies Johnson, who succeeded Mr. Manson, designed 
some heavy bogie tank engines. They were four wheels coupled in 
front, with trailing four-wheeled bogie. The valves were of the ordi- 
nary type, placed between the cylinders, which were " inside." 

These engines were fitted with the brick arch, and since that time 
all the Company's engines have had the air tubes removed, and brick 
archea fitted. 

The following are the principal dimensions: 

Cylinders _. .. 17J,in. by 26in. 

Coupled wheels 5ft. Oin. diameter. 

Bogie wheels 3ft. O^in. diameter. 

Fixed wheel base .. 7ft. 6in. 

Bogie wheel base 5ft. 6in. 

Total wheel base 22ft. Oin. 

Tubes 220 l|in. external diameter. 

Heating surface Tubes .. .. .. .. l,093.5sq.-ft. 

Firebox 113.5sq.-ft. 



Total 1.207.0sq.-ft. 

Grate area . 18sq.-ft. 

Working steam pressure .. 1651b. per sq.-in. 

Tank capacity 1,200 gallons. 

Bunker capacity 2 tons coal. 

Weight in working order 53 tons 15 cwts. 

Built by Neilson. 

In the same year Mr. Johnson designed some inside cylinder pas- 
senger engines (Fig. 139), which had the same size of boiler as the 
bogie tank engines. 

They had a four-wheeled bogie in front, and four-coupled driving 
wheels. The tender was on six wheels, and carried the same amount 
of water as the bogie tenders previously described. Spring buffers are 
used between engine and tender. The principal dimensions are : 



Cylinders 

Coupled wheels 

Bogie wheels 

Tender wheels 

Wheel base of engine 

Wheel base of tender 

Total wheel base of engine and tender 

Working- pressure 

Weight in working order Engine . . 
Tender . . 



18in. by 26in. 

6ft. lin. diair.eter. 



3l't, 9in. diameter. 
6 4ft. lin. diameter. 
21ft. 9^in. 
13ft. Oin. 
43ft. 4in. 
.1f<5 h. per sq.-in. 
43 tons 18 cwts. 
35 tons cwts. 



" Total 78 tons 18 cwts. 

Built by Neilson and Co. 

This is the present standard type of passenger engines on the 
Great North of Scotland Railway. 

In addition to the engines of the Furness Railway, previously de- 
scribed, others deserve recognition, and it should be placed on record 
that the red-brown colour distinguishing the locomotives of this line 
has been the standard colour for a number of years. Some sixteen years 
or so back, the Midland Railway discarded green as the distinguishing 
colour fcr its engines, and adopted the red-brown shade of the Furness 



EVOLUTION OF THE STEAM LOCOMOTIVE 



313 



Railway. Some people have imagined that the Furness Railway loco- 
motives are painted in imitation of the Midland, but the facts show 
the opposite to be the case. 

In 1870 a type of four-wheels-coupled passenger engines. were intro- 
duced on the Furness Railway. The leading dimensions of these 
were : 



Diameter of cylinders 

Stroke 

Diameter of coupled wheels 
Diameter of leading wheels 



1ft. 4in. 

1ft. Sin. 

5ft. ?iin. 

3ft. Sin. 



CENTRE TO CENTKE OF WHEELS. 



Leading to driving 
Driving 1 to trailing 
Total wheel base .. 
Diameter of boiler (mean) 
Length of barrel 
Length of firebox (shell) 
Number of tubes 



6ft. 6in. 

7ft. 9in. 
14ft. Sin. 

3ft. llin. 
10ft. Oin. 

4ft. 4in. 

157 2in. external diameter 



Tubes 
Firebox 



Total 



HEATING SURFACE 



839.5sq.-ft. 
77.0sq.-ft. 

916.5sq.-ft. 



TENDER (Four Wheels). 



Grate area 

Diameter of wheels 

Wheel base 

Capacity of tank .. .. 

Capacity coal 

Total wheel base, engine and tender 



ll.5sq.-ft 

3ft. Sin. 

9ft. 6in. 
1,200 gallons. 
3 tons. 

120. 



WEIGHTS IN WORKING ORDER. 

T. o. 

~ 8 10 

~ 11 10 



Leading 
Driving 
Trailing 



Total 

Total weight of tender 

Working pressure in Ibs. per sq.-in. 




At this period the standard goods engines of the Furness Railway 
were six- wheels-coupled, of the following dimensions: 



Diameter of cylinders 

Stroke 

Diameter of coupled wheels 



1ft. 4in. 
At. Oin. 
4ft. 7in 



CENTRE TO CENTRE OF WHEELS. 



Leading to driving 
Driving to trailing 
Total wheel base 
Diameter of boiler (mean) 
Length of barrel 
Length of firebox (shell) 
Number of tubes 



6ft. 9in. 

8ft. Oin. 
14ft. 9in. 

3ft. llin. 
10ft 4in. 

4ft. lliin. 
156 2<n. external diameter. 



314 



EVOLUTION OF THE STEAM LOCOMOTIVE 



HEATING SUEFACE. 



Tubes 
Firebox 



Total 



TENDER (Four Wheels). 

Grate area 

Diameter of wheels 

Wheel base ,. 

Capacity of tank 

Capacity coal 

Total wheel base, engine and tender 



Leading 
Driving 
Trailing 



Total 

Total weight of tender 

Working pressure in Ibs. per sq.-in. 




.-ft. 



13.8sq.-ft. 

3ft. 8in. 

9ft. 6in. 
1,600 gallons. 
3 tons. 
32ft. 7in. 



WEIGHTS IN WORKING ORDER. 



T. 0. 

10 11 

11 10 
8 18 



30 19 
19 10 
120. 



The modern main line Furness Railway passenger engines have 
four wheels coupled of 6ft, diameter, with a leading bogie, the wheels 
of which are 3ft. Gin. diameter. The cylinders are inside 18in. dia- 
meter, with a 24in. stroke. The other dimensions are: 



CENTRE TO CENTRE OF WHEELS. 

Centre of bogie to centre of driving axle . . . . 9ft. 6iin. 

Centre of driving to trailing 8ft. 6in. 

Centres of bogie wheels 5ft. 9in. 

Total wheel base .. .-. 20ft. llin. 

Diameter of boiler (mean) 4ft. 3in. 

Length of barrel 10ft. 3in. 

Length of firebox (shell) 5ft. 9in. 

Number of tubes 230 l|in. external diameter. 

HEATING SURFACE. 

Tubes ... l,109.0sq.-ft. 

Firebox .. .. .. ., 99.5sq.-ft. 

Total l,208.5sq.ft. 

TENDER (6 wheels). 

Grate area 17sq.-ft. 

Diameter of barrel 3ft. lOin. 

Wheel base 12ft. Oin. 

Capacity of tank .. 2,500 gallons 

Capacity coal 4 tons. 

Total wheel base, engine and tender .. ... 42ft lin. 

WEIGHTS IN WORKING ORDET? 

T. 0. Q. 

Leading bogie 13 12 

Driving 14 10 

Trailing .. , 13 4 Q 

Total .. 4 1 I Q~ 

Total weight of tender 28 5 

Working pressure in iKs. per sq.-in 150. 



EVOLUTION OF THE bTEA M LOCOMOTIVE 



315 



This express class of passenger engines was introduced in 1896. 

When Mr. W. Pettigrew, M.Inst.C.E., who was, during the latter 
years of Mr. Adams's regime, practically the chief at Nine Elms Loco- 
motive Works, was appointed locomotive superintendent at Barrow, to 

n 




FIG. 140. PETTIGREW'S NEW GOODS ENGINE FOR THE FURNESS RAILWAY 

succeed Mr. Mason, he got out designs for a new and powerful class 
of goods engines, which are now being delivered to the Furness Rail- 
way. Fig. 140 represents one of these engines, the leading dimen- 
sions of which are : 



Diameter of cylinders . . 

Stroke 

Diameter of coupled wheels 
Wheel base of engine . . 
Diameter of boiler (inside) 
Length of barrel . .* 

Length of fire-box (outside) 
The boiler contains 



18in. 

26in. 

4ft. Sin. 

5ft. 6in. 

4ft. 4in. 
10ft. 6in. 

6ft. 9in. 
208 tubes, l|in. external diameter 



HEATING SURFACE. 



Tubes 
Firebox 



Total 

Grate surface 

Weight of engine in working order (about) 
Working pressure 

TENDER. 

Diameter of wheels 

Wheel box 

Weight in working order (about) .. 

Capacity of tanks 

Coal 

Total wheel base of engine and tender . . 
Total weight in working order (about) .. 



l,029sq.-ft. 
105sq.-ft. 

l,134sq.-ft. 

20.5sq.-ft. 

38i tons. 
1501bs. per sq.-in. 



3ft. lOin. 
12ft. 

28i tons. 
2,500 gallons. 
4 tons. 
37ft. llin. 
66| tons. 



Fig. 141 represents one of the Highland Railway's 10-wheel main 
line engines, with outside cylinders. The six-coupled wheels make 



316 EVOLUTION OF THE STEAM LOCOMOTIVE 

this design to be well adapted for the heavy traffic of the system, 
whilst the leading bogie gives sufficient facility for easily negotiating 
the curves of the Highland Railway. 

The first newest class of express engines, designed by Mr. P. Drum 
mond, is just delivered, and is very similar to those designed for 
the Highland Railway by Mr. D. Jones, the late locomotive super- 
intendent, except that the new class has inside cylinders, whilst those 
built two years ago had outside cylinders. The dimensions of No. 1, 
" Ben-y-Gloe," just delivered, are: cylinders, 18 Jin. by 26in. The 
coupled wheels 6ft. and the leading bogie wheels 3ft. Gin. diameter. 
Heating surface, 1,175 sq. ft. Steam pressure, I751b. per sq. in. 
Weight, in working order: engine, 46 tons; tender, 37J tons. 

Mr. H. Pollitt's design of locomotive for working the express traffic 
over the London extension of the Great Central Railway has four- 
coupled wheels 7ft. diameter; cylinders, 18jin. by 26in., with piston 
valves; a Belpaire fire-box, and steam-pressure I701bs. per sq. in. 
The tender holds 4,000 gallons of water and 5 tons of coal. 




FIG. 141. SIX-WHEELS-COUPLED BOGIE ENGINE, WITH OUTSIDE CYLINDERS, 
HIGHLAND RAILWAY 

Before closing this account of locomotive evolution, some few 
details of modern Irish locomotives will be of interest. 

Fig. 142 represents a four-coupled passenger engine of the Belfast 
and Northern Counties Railway. This engine is of the compound 
type, and is fired by petroleum on Holden's system. 

"Jubilee" (Fig. 143) is also a compound express passenger engine 
of the same railway. Both these engines were designed by Mr. B. 
Malcolm, the Company's locomotive superintendent. 



EVOLUTION OF THE STEAM LOCOMOTIVE 



317 



The modern passenger engines on the Great Northern Hallway 
(Ireland) arc of the four-coupled type, with a leading bogie, and are 




FIG. 142. LIQUID FUEL ENGINE, BELFAST AND NORTHERN COUNTIES RAILWAY 



Rostrevor " class. The leading dimensions are as 



known as the 
follows : 

CYLINDERS. 

Diameter of piston IS^ifl. 

Stroke of piston 24in. 

Centie to centre 2ft. 7m 

Steam ports 14iin. by l^in. 

Exhaust ports lli'.n. by 5im. 

Outside tap Tin. 

Lead .. l-8in. 

Maximum travel . . . . . . . . . 3}in. 

WHEELS 

Diaoieter of bogie wheels 3ft l^in 

Diameter of driving wheels . . . . . . . . 6ft. Via. 

Diameter of trailing wheels .. .. 6ft. 7in 

Bogie wheel base . . . . 5f c. 3in 

From bogie wheel centre to trailing 17ft. 9in, 

Total wheel ba~t .. .- 20ft. 4iin. 

HEATING SURFACE. 

In firebox _ 109sq -ft 

Tubes l,013sq.-ft. 

Total l,122i.-ft 

Grate area . . 18isq. fr. 

Working pressure per sq. in.) .. - 1601b 

WEIGHT. In working order. 

T. o. Q. 

Bogie 13 5 

Driving axle . . . 14 15 

Trailing 14 



Total 



42 



318 EVOLUTION OF THE STEAM LOCOMOTIVE 




FIG. 143." JUBILEE," 4-WHEELS COUPLED COMPOUND LOCOMOTIVE, 
BELFAST AND NORTHERN COUNTIES RAILWAY 

Fig. 144 represents one of the engines of the Great Northern 
(Ireland) Railway, as decorated to haul the Duke of York's train daring 
his recent visit to Ireland. 




144. " No. 73," STANDARD PASSENGER ENGINE, GREAT NORTHERN 
RAILWAY (IRELAND) 



EVOLU.TIOX OF THE STEAM LOCOMOTIVE 



319 




FIG. 145. FOUR-COUPLED BOGIE EXPRESS ENGINE, GREAT SOUTHERN 
AND WESTERN RAILWAY 

Fig. 145 is from a photograph of one of the standard passenger 
engines of the Great Southern and Western Railway. This engine 
was designed by Mr. R. Coey, the Company's locomotive superinten- 
dent. The coupled wheels are 6ft. Gin. diameter, the cylinders being 
18in. diameter, with a stroke of 24in. * 

Our last illustration (Fig. 146) is produced from a photograph of 
<: Peake," one of the " light " engines of the Cork and Muskerry Light 
Railway. Engines of this type are specially designed for working on 
" light " railways. 




FIG. 146. "PEAKE," A LOCOMOTIVE OF THE CORK AND MUSKERRY 
LIGHT RAILWAY 



INDEX. 



[N.B. The letters B.G. denote a Broad Gauge locomotive.] 



PAGE 



Adams, Bridges, combination en- 
gines and carriages 130, 133 

Adams, Bridges, radial axle-boxes 209 
Adams, Bridges, system of in- 
termediate driving shafts 133 

Adams, Bridges, spring tyres 211 

Adams, W., engines for the N.L. 

Ry. ,.., 226 

Adams, W., engines for the L. & 

S.W. Ry 272 

"^Eolus," B.G 71 

"Agenoria," 27 

" Agilis," with double flanged 

wheels 86 

"Ajax," B.G 73, 75 

" Albion '' on the " Cambrian 

system" 125 

Allan claims to have introduced 
"back-coupled" engines, 97; 

link motion 97 

American engines for the Birming- 
ham & Gloucester Railway 87 

"Apollo," B.G 72 

" Areo-steam "' engines 234 

"Ariel," B.G 75 

"Armstrong" class, G.W.R. ... 297 
Aspinall, J. A. F., locomotives 
for the Lancashire & York- 
shire Railway 280 

Aston, W., engines for the Cam- 
brian Railways 264 

"Atlas," B.G 75 

"Atlas," M. & S.R 110 



"Bacchus," B.G. 72 

Back-coupled engines by Allan ... 97 

Bilamed locomotives 84 

'Beattie's engines, 162, 169 (coal- 
burning), 185, 194, 203. 207, 

226, 231, 240 



PAGB 

Belfast & Northern Counties Rail- 
way engines t 316 

Beyer's single iron plate 

{frames 97 

Beyer's "Atlas," for the M. & 

S.R 110 

Billinton, R. J., engines for L.B. 

& S.C.R 260 

Birmingham & Gloucester Ry., 

American engines on 87 

Birmingham & Gloucester Ry., 

McConnell's engine for 102 

"Black Prince," L. & N.W.R. ... 247 
Blackett, Hedley, and Hackworth 

construct an engine 10 

Blenkinsopp's, J., engine 5 

"Blucher'' 14 

"Boat engines," B.G 73 

Bodmer's reciprocating engines... 100 

Bogie tenders 241, 277, 310 

Bogie engines (early) 56, 173 

Braithwaite & Ericsson's "No- 
velty" 50 

Braithwaite & Ericsson's "Wil- 
liam the IV." and "Queen 

Adelaide" 46 

Bristol & Exeter Ry. locomotives, 

B.G 173 

Broad gauge engines (see G.W. & 
Bristol & Exeter Railways) 

"Brougham," S. & D.R. 206 

Brunei, I. K., and broad gauge 
locomotives, 67, 75; Vale of 

Neath Ry 39 

(See also Great Western 
^ Railway engines) 
Brunton's " leg propelled " engine 7 

"Bull Dog," G.W.R 302 

Burnett's tanks for the M & " 

S.J.W.R ..233 

Bury, Edward, inventor of the in- 
side cylinder locomotive 40 

Bury, his first " Liverpool " 40 

Bury, Autheptic list of his first 
engines 43 



322 



EVOLUTION OF THE STEAM LOCOMOTIVE 



PAGE 

Bury, Contractor to the London 
& Birmingham Ry 82 

Bury, Engines on the Furness 
Ry. 123, 179 

Bury, Extract from the Minute- 
books of the L. & M.R. relat- 
ing to the "Liverpool" 42 

Bury, " Liver," for the L.' & 
M.R 52 

Bury, "Meteor," N. & C.R 62 



Cambrian locomotive system 125 

Caledonian Ry. : Efcgine "No. 
15," 152; 8ft. 2in. "single," 

207; modern 285 to 292 

"Caledonian," L. & M.R 54 

Canterbury & Whitstable Rail- 
way 44 

Cambrian Railways engines 209, 264 
"Canute," an early coal-burning 

engine 186 

Chapman's chain locomotive 6 

" Charles Dickens," L. & N.W.R. 239 
Clark's smoke consuming engines 191 
Coal-burning locomotives, 84 ; 
Chanter's system, 84; Dew- 
rance's, 102 ; London & North 
Western, 167; Beattie, 185; 
Yoraton, 188; Cudworth, 189; 
Yarrow, 190; Clark, 191; Wil- 
son. 191 ; Lee and Jacques, 
192: Sinclair, 192; Douglas 

or Frodsham 192 

Coey, R.. engines for the G.S. 

& W.R. 319 

Oork & Muskerry Light Railway 319 
Combined engines and carriages 

130, 136, 224 
"Comet," Newcastle & Carlisle 

Railway 60 

Compound Iiocomotdves ... 169, 242 
249, 316 

" -Compressed air locomotive 169 

"Cornwall'' 119 

Cork & Bandon Ry., Adams's 

light engines on 140 

"Caithness," L. & N.W.R 205 

Cowan, W., goods engine 225 

Cowlairs incline, 98 ; rope trac- 
tion on 100 

Crampton, T. R., locomotives, 
75; on the lOft.-wheel. B.G., 

112, 145, 159, 203 

' Crewe Works erected 97 

Cudworth, I. , coal-burning en- 
gines 189 



PAGE 



Cudworth, coal burning engines... 189 
" Cycloped " horse locomotive ... 38 
Cylinder valves, fitted to 
Roberts's " Experiment " ......... 57 



.Davis & Metcalf e's exhaust-steam 

_- injector 502 

Dean, "W., locomotives for the 

Great Western Ry 2G4 

"Devonshire" class, G.W.R. ... 297 

Disc wheels 74, 75 

Dodd's engines for the Monkland 

and Kirkintilloch Ry 50 

Douglas, coal burning engine 192 

Drummond, D., engines for the 

L, & S.W.R 276 

Drummond, D., engines for the 

Caledonian Ry. 288 

" Dunalastair, " Cal. Ry 285 

Dundee & Newtyle Ry. engines ... 57 
"Duplex," a two-boiler engine... 158 



Eastern Counties Ry., Hancock's 
locomotive for, 86 ; " Essex," 
111 ; compressed air engine, 
169 ; coal burning, 192 ; Sin- 
clair's engines 195 

(See also G.E.R.) 

" Eclipse," Dr. Church's tank en- 
gine &2 

'Eigftt - wheels. - coupled engines, 
early 195 

Eight - wheels - coupled engines, 
Webb's 246 

Eight- wheel rolling stock, the 
first 46 

" Enfield," combined engine and 
carriage 133 

England's " Little England " loco- 
motives 141 

"Essex," E.C.R Ill 

Exhaust steam blast (see Hack- 
worth) 

Exhaust steam injector (Davies & 
Metcalfe's patent) 302 

"Experiment" engine for the L. 
& M.R 56 

"Experiment," L. & N.W.R, ... 243 



INDEX. 



523 



F 



PAGE 



" Fairfield " combined engine and 

carriage, B.G 131, 153 

Fairlie's " double bogie " en- 
gines 224, 234 

Festiniog Railway, Fairlie's 

engines on . 223 

Fell's steep gradient engines 219 

Fletcher's 4-wheel tank engine ... 201 
" Folkestone," a Cramp ton engine 

for the S.E.R 159 

Four-cylinder engines, L. & N.W. 

Railway 248 

Four-cylinder engines, S.W.R 276 

Fowler, Sir J., "hot-brick" en- 
gine 200, 217 

French locomotive on the Eastern 

Counties Ry 195, 207 

Furness Ry. engines, 123, 179, 236, 312 
"FHiry" and "Firefly" classes, 
G.W.R., B.G. . 90 



Galloway's incline climbing 

experiments 109 

Gauge locomotive experiments ... 105 
Geared-up engines, B.G. 77, 79, 147 

Giffard's injector 197 

"Gladstone" class, L.B. & S.C. 

Railway 252 

Glasgow & South- Western Ry. 

locomotives 241 

" Globe," the first engine with a 

steam dome 47 

" Goliath/' Newcastle & Carlisle 

Railway 61 

Gooch, Daniel (see G.W.R.) 
Gooch, J. V., engines by ... 161, 162 
Grand Junction Ry., opening ... 64 
Grand Junction Ry. early locomo- 
tives 64 

"Grasshopper," B.G 73 

Gray's expansion gear 93 

"Great Britain/' M'Connell's 102 

"Greater Britain," L. & N.W.R. 245 
Great Central Ry. 4 , Pollitt's en- 
gines for 316 

Great Eastern Ry., locomotives 
(see also Eastern Counties 

Ry 206, 217, 249, 255 to '259 

Great Northern Railway en- 
gine, "215" ". 171 

Geat Northern Ry. engines, 171, 

216, 236, 303 to 309 
Great North of Scotland Ry. en- 
- gines 225, 309 to 311 



Great Northern (Ireland) Ry. en- 
gine 318 

Great Southern & Western Ry. ... 319 

"Great Western," B.G 106 

Great Western Ry. locomotives, 
the original, 66; first trial of, 
69 ; table of dimensions, 70 ; 
the 10ft. wheel engines, 73, 76; 
geared-up engines, 77, 79 ; table 
of mileage of original engines, 
81 ; Gooch's first engines, 90 ; 
first engine built at Swindon, 
105; "Great Western," 106; 
trial trips, 107, 108 ; Galloway's 
engine, 109; "Iron Duke," 
113 ; first narrow gauge 
engines, 182; "Robin Hood," 
184 ; "Metropolitan Ry. , engines 
for, 213 ; Dean's designs, 294 to 303 
"Grosvenor," L.B. & S.C.R 242 



Hackworth, Timothy, first en- 
gine 10 

Hackworth, Timothy, and the 
Stockton and Darlington Ry. 

locomotives 21 

Hackworth, Timothy, " Royal 

George'' j .' 24 

Hackworth, Timothy, " Saiis- 

pareil" 32 

Hackworth, Timothy, and the ex- 
haiist steam blast, 24 ; the 

secret stolen at Rainhill 33 

Hackworth, Timothy, "Globe" 

for the S. & D.R 47 

Hackworth, Timothy, "Majestic" 
and Wilberforce " classes " for 

the S. & D.R. ., 52, 53 

Hackworth, Timothy, trunk, or 

ram engine, 61 ; " Arrow " ... 61 
Hackworth, Timothy, builds 

"Jenny Linds " 1C4 

Hackworth, Timothy, " Sans- 
pareil 2," 149; challenge to R. 

Stephenson concerning 150 

Harrison's patent engines, B.G. ... 76 
" Harvey Combe " ballast engine 60 
Hancock's engine for the Eastern 

Counties Ry 86 

Haigh Foundry engines, B.G. ... 79 

"Hawthorn'' 157 

Hawthorne's engines 52, 59, 156 

Hedley (see Blackett) 10 

Highland Railway locomotives ... 316 



,524 



EVOLUTION OF THE STEAM LOCOMOTIVE 



Historical locomotives sold by 
auction 51 

Holden, J., liquid fuel locomo- 
tives 254, 316 

Holden, J., engines for the G.E.R. 253 

Holmes, M., engines for the 
N.B.R 253, 277 

" Hot-brick " locomotive, Fowler's, 
for Met. Ry 200, 217 

Howe and the "link" motion, 
96; 3-cylinder engine 105 

''Hundred miles an hour!" B.G. 79 

Hurricane 10ft. wheel engine, 
B.G 76, 79 



Injector, Gifford's invention of ...197 

"Iron Duke" BG. 113 

Inside cylinder locomotive, " Liver- 
pool," the first 40 

nside cylinder locomotive extract 
from the minute books of the L. 

& M.R. relating to same 42 

"Inspector," L.B. & S.C.R. ... 261 
International Exhibition, 1851, 

locomotives at 156 

"Invicta," Canterbury & Whit- 
stable Ry 44 

Ivatt, H. IA., engines for the 
G.KR. ., 303 to 309 



"Jason,"' B.G., Gooch's first 

goods engine 92 

James and the link motion 96 

"Jenny Lind " engines 104, 115 

"Jenny Sharps" 116 

"Jinks's Babies" 234 

Johnson, S. W., engines for the 
Midland Ry 250 



Kirtley, W., engines for the L.C. 

& D.R 262 

Kendall, W., 3-cylinder engine ... 231 
Kennedy's, James, testimony re- 
garding the first inside cylinder 
locomotive . 42 



"Lablache" 124 

"Lambro" 95 

Lancashire & Yorkshire Ry. en- 
gines 234, 280 

L.B. & S.O.R. locomotives, 240, 

242, 252, -260 
L.C. & D.R. locomotives ... 203, 262 



PA9B 

" Light locomotives," Samuels' 
130 ; Adams's, 139 ; England's 141 

Liquid fuel locomotives 253, 285 

"Little Wonder," Festiniog Ry. 224 

"Little England" 141 

" Little Wonder," Samuels' com- 
bined engine and carriage 130 

" Liverpool," Crampton's engine 

for the L. & N.W.R 145 

" Link " Motion, 96 ; Allan's P7 

"Liver," Bury's, for L. & M.R.... 52 
" Liverpool," the first engine with 
inside cyclinders and crank axles 40 

"Liverpool," description of 44 

Liverpool & Manchester Ry., 
earfv locomotives on, 45, 46, 50, 

52, 85 
Liverpool & Manchester Ry. , 

opening of .".. 46 

Liverpool & Manchester Ry. : 
8-wheel passenger carriage at 

opening 46 

Liverpool & Manchester Ry., 
Rainhill contest, 28; the com- 
petitors 30 

"Locomotion," S. & D.R 20 

London & Birmingham Ry. , open- 
ing, 82; Bury's engines for ... 82 
London, Brighton & S.C.R., Bod- 

mer's engine on . 101 

"Long boiler" engines, 94, 103, 

111, 122, 137 
London, Brighton & S.C. Ry., 

"Jenny Lindsl" 116 

"London," Crampton's engine for 

theL. &N.W.R 113 

"Lord of the Isles," B.G 115 

L. & S.W.R. locomotives, 162, 
169, 187, 194, 202, 207, 226, 

231, 240, 72 

L. & N.W.R. locomotives, 163, 
122, 153, 155, 205, 238, 239, 

243, 281 
(See also London & Birmingham Ry.) 

M 

"Magnet," S. & D.R 54 

" Majestic " class, S. & D.R 52 

Malcolm, B.. engines for the Bel- 
fast & Northern Counties Ry.... 316 
Manson, engines for the G.N. of 

S. Ry 309 

"Mars," B.G 73 

McCo-nnell's " Great Britain," 
102 ; counterbalancing experi- 
ments, 122 ; " most powerful 
KG. engine," 122; "Mac's 
Mangle," 153; "Bloomer's 
155; "300," 163; "Caithness," 205 



INDEX. 



325 



Mclntosh, J. F., locomotives for 

the Caledonian Ry 286 

Metropolitan Ry.,* hot brick 

engine for, 200; B.G. engines 

on, 213; first engines 214 

"Meteor," L. & S.W. R 203 

* Meteor," Bury's, for tb.e N. & 

C. Ry. 62 

Metallic piston packing, first 

used 51 

" Michael Longridge," for the S. 

fc D. Ry 64 

Midland Ry., trials of "Jenny 

Sharps'' and " Jenny Lands" on 

116 to 118 

Midland Ry., Johnson's engines 250 
Monkland & Kirkintilloch Ry., 

first engines on the 50 

Murray's, M., engine (see Blenkin- 

sopp] 5 



N 



" Namur," on Crampton's system 112 
Narrow gauge engines on the 

G.W.R., the first 182 

Neilson's type of goods engine... 180 
Newcastle & Carlisle Ry., open- 
ing of 59 

Newcastle & Carlisle Ry., "Go- 
liath' locomotive 61 

Newcastle & Carlisle Ry., 

"Atlas" locomotive 61 

Newcastle & Carlisle Ry., 

" Tyne " locomotive 61 

Newcastle & Carlisle Ry., 

"Eden" locomotive 62 

Newcastle & Carlisle Ry., 

"Meteor" locomotive 62 

Norfolk Ry., light engines on ... 140 
North British Ry. engines . . . 253, 276 
North Eastern Ry. locomotives, 

249, 252 
North London Ry. engines, 191, 

226, 230 

"No. 266," G.N.R 508 

"No. 990," G.N.R 3D6 

"Nunthorpe," S. & D.R 193 



"Old Ooppernob," Furness Ry, 
the oldest engine now at work T23 

Opening of the Canterbury and 
Whitstable Railway 44 

Opening of the Liverpool and Man- 
chester Ry 46 

Opening of the Stockton & Dar- 
lington Ry 47 



PAGB 

Opening of the Newcastle & Car- 
lisle Ry 69 

Opening oi the Grand Junction 

Railway 64 

Opening of the Great Western Ry. 72 
Opening of the London & Bir- 
mingham Ry 82 

Opening of the London & South- 
ampton Ry 85 

Opening of the East Kent Ry. ... 195 
Opening of the Metropolitan Ry., 

B.G 213 

Opening of the Metropolitan & St. 
John's Wood Ry 233 



Pambour, on the early L. & M. 

Ry. engines .......................... 50 

Casey's compressed air locomotive 169 
" Patentee," Stephenson's 6-wheel 

passenger engine for the L. & 

M.R .................................. 59 

Paton's Cowlairs Incline engine... 98 
Pearson's design for a double tank 

locomotive, 147; 9ft. singles... 175 
Pettigrew, W., engines for the 

Furness Ry., ........................ 315 

"Perseverance," at Rainhill ...... 37 

"Planet," L. & M. Ry ............. 49 

" Plews," Y.N & B.R ............ .... 144 

Pollitt, H., engines for the Great 

Central Ry ........................... 316 

" type, L. & N.W. R. 238 



Precedent 



"Precursor" type, L. &N.W.R. 239 
" G.E.R 



255 
105 



"Pretolea, 

"Premier," B.G ...................... 

"Problem," L. & N.W.R., first 

engine fitted with the injector... 197 
Pryce, H. J., engines for the 

N.L.R. .............................. 230 

"Puffing Billy" ...................... 12 

" Python," L. & S.W.R ............. 231 



" Queen-Empress," L. & N.W.R. 244 

R 

Rainhill locomotive contest, "Cy- 

cloped" at the 38 

Rainhill locomotive contest, 

Manumotive carriages at the ... 38 
Rainhill locomotive contest, 

"Perseverance" at the 1W 

Rainhill locomotive contest, 

"Rocket" at the 35 



326 



EVOLUTION OF THE STEAM LOCOMOTIVE 



locomotive contest : 

. "Sanspareil" at the 33 

Rainhill locomotive contest, 

"Novelty" at the 30 

Ramsbo'ttbm's' water pick-up ap- 
paratus 198 

"Red Star," B.G 136 

RennieV" Lambro " 95 

Ritchie's design for a locomotive 148 
Roberta's "Experiment/ L. & 

M.R:, with cylinder valves 56 

Robertson's steam brake 97 

: -" R6ckeV: at Rainhill, 35; later 

- history" .:. 37 

^Rocket," Colburn's opinion of 

her 36 

" Rocket," her tubular boiler in- 
vented by Booth 36 

" Rocket," awarded the Rainhill 

$ prize 36 

Russia, Hackworth's trunk en- 
gine, the first locomotive in 61 

" Royal George," first financially 

successful locomotive 23 

.'.' Royal William" ..: 19 



S 



Samuels' "Little Wonder'' 130 

" Sanspareil," at Rainhill, 33 ; 

later history 34 

"Sanspareil 2" 149 

Sheffield & Manchester Ry., 

Bodmer's engines on 100, 101 

Sheffield & Manchester Ry., " At- 

~, las" 110 

]Short-stroke engines 60, 61 

Sinclair's smoke consuming en- 
gines, 192; "singles," 206; 

tanks 217 

"Snake" , 122 

Steam organ, fitted to the " Tyne " 62 

Steam tenders 217 

Steam blast (see also Hackworth) 24 
Stephenson, Geo., & Hackworth, 

11, 21, 25, 33 

Stephenson's engines : first, 14 ; 
second, 15 ; third, 16 ; later 
types, 26, 27, 35, 43, 46, 49, 
50, 51, 56, 59, 60, 64; long 
boiler, 103; 3-cylinder, 105; 
. ,. "A," 105; "White Horse of 

Kent," 111, double engine 194 

Stephenson's, Robt., valve gear, 

94; link motion 96 

Stephenson's, Robt., challenge to, 

re "Sanspareil 2" 150 

.Stewart, W., early locomotives by 13 



PAGK 

Stirling, P., 8ft. lin. "singles" 
for the G.KR 236 

Stirling, J., design for the G. & 
S.W.R 241 

Stirling, J., reversing apparatus... 241 

Stirling, J., locomotives for the 
S.E.R 266 

Stockton & Darlington Ry., open- 
ing of 20 

Stockton and Darlington Railway, 
"Royal George'"' 23 

Stockton & Darlington Ry. 
locomotives, 22, 27, 47, 52, 53, 
54, 61, 64, 65, 193, 206 234 

Stroudley's engines for the L.B. 
& S.C.R 242, 252 

Sturrock, Archibald, apprenticed, 
58; "215," G.KR., 171; and 
Met. R. , 214 ; condensing en- 
gines, 216; steam tenders 217 

"Soho" 86 

South Eastern Ry., Bodmer's en- 
gines on 101 

South Eastern Ry., "White 
Horse of Kent," 111; "Folke- 
stone," 159 ; Sharp's engines, 
161; Cud worth's, 189, 225; 
Watkin's, 241; Stirling's 266 

South Devon Ry. locomotives, 
B.G, 179 

"Sunbeam,' S. & D.R 64 

" Swiftsure," Forrester's, engine 
for the L. & M.R , 59 



Tayleur r s short stroke locomotives 60 

Tank engines, early 137, 158 

Taff Valley Ry. engines, 163, 202, 

233, 292 
Ten feet driving wheel engines, 

B.G 69, 73, 76 

"Teutonic," L. & KW.R 244 

"Thunderer," B.G. 77 

Three-cylinder engines, Stephen- 
son & Howe's 105 

Three-cylinder engines, Kendall's 231 
"Tiny," Crewe Works, engine ... 209 

Tosh's goods engine 182 

Trevithick' s, F., "Cornwall" ... 119 
Trevithick, R., inventor of the 

steam locomotive 1 

Trevithick, R., his first railway 

engine 2 

"Tyne," K & C.R 62 



INDEX. 



327 



PAGB 



Vale of Neath Railway (see Brunei). 
Valve gears, " Experiment/' 57 ; 
"Soho," 86; Gray's, 93; 
Dodds Owen's, 93 ; Stephen- 
son's, 94; -link motion, 96; 
vertical, 123; rotatory, 181; 

Dubs' .". 199 

"Venus," B.G 72, 136 

"Vulcan," B.G. 69, 70 j 

W 

" Wallace," Dundee & Arbroath 
Railway 82 \ 

Water pick-up apparatus, Rams- 
bottom's 198 

"Welsh Pony," Festiniog Ry. ... 223 

Webb, F. W., engines for the 
L. & N.W.R., " Precedents," 
238; "Precursors," 239; com- 
pounds, 243 ; " Greater Britain," 
245; 8 wheel-coupled, 246; 
"Black Prince" .. ..248 



PAflB 

Whishaw on Great Western Ry. 
B.G. engines (see Chapter VI., 
pages 66 to 81) 

Williams and the "link" motion S6- 
Wilson's, Ed., system of smoke 

consuming 191 

" Wilberforce," S. & D.R 53 

"William the 4th," and "Queen 

Adelaide ' ' locomotives 46 

Winans' Manumotive carriages at 

Rainhill 38 

Worsdell's compounds 249 

' ' Windcutter, ' ' locomotive 274 

Wood, N., on Great Western 

Ry. B.G. engines (see chapter 

VI. , pages 66 to 81) 
"Wrekin, 151 



Yarrow's coal-burning engine 190 

Yorston's coal-burning engine ... 188 
"Ysabel" 168 



RETURN CIRCULATION DEPARTMENT 

TO ^ 202 Main Library 


LOAN PERIOD 1 
HOME USE 


2 


3 


4 


5 


6 



ALL BOOKS MAY BE RECALLED AFTER 7 DAYS 

1 -month loans may be renewed by calling 642-3405 

6-month loans may be recharged by bringing books to Circulation Desk 

Renewals and recharges may be made 4 days prior to due date 

DUE AS STAMPED BELOW~ 



1C LF 






OCT 1 198 


o 






RECEIVED b 








SFP ^ MO 






CIRCULATION /DP 


















LIBRARY USE ON 
Vllll 0/|Hh; 








CIRCTIT ATION D? 


PT. 














AUG08 1 


)B7 






AUTO. DISC. 






V JUL 1 5 1957 







UNIVERSITY OF CALIFORNIA, BERKELEY 
FORM NO. DD6, 60m, 3/80 BERKELEY, CA 94720 

s 



U.C. BERKELEY LIBRARIES