:
TT-'iO-T
THE TOWER BRIDGE.
A LECTURE.
BY
JOHN WOLFE BARRY,
M.INST.C.E.
XonDon :
BOOT, SON AND CARPENTER,
24, OLD BAILEY, B.C.
1894.
ERRATA.
Page 43, line 11, for 12 inches" read " 27 inches;
Page 45, line 3, for "their" read "these;"
Page 51, line 16, for "down" read "into."
PREFACE.
Lecture was composed in 1893 for an audience
consisting in part of persons well acquainted with
engineering matters, but more largely of those who took an
interest in the Tower Bridge as members of the general
public. It is not to be regarded as being such a professional
description of the Bridge as would be suitable to a body of
experts ; but it may be of interest to a considerable number
of persons who have seen the Bridge in its various stages of
progress, or who wish to form a general conception of its
design and of the various considerations which led up to
its inception and determined its mode of execution.
J. W. B.
THE TOWER BRIDGE.
A LECTURE.
BY
JOHN WOLFE BARRY, M.lNST.C.E.
Introductory.
T)EFORE entering upon a description of the structure of
the Tower Bridge, it may be interesting if I touch
first upon the history of the principal bridges of London, and
then upon the necessity of means being afforded for crossing
the river below London Bridge, with a glance at the various
suggestions which have been made to meet the want.
A bridge appears to have existed on the site, or a few
yards lower down the river than the site, of the present
London Bridge, from remote antiquity. We read of Canute
attacking and being repulsed from London Bridge, and some
writers have held that a bridge at this spot dates from the
Roman occupation of London, which was then known to the
world as the town Augusta, and must have been of con-
siderable importance. Stow says that in the year 994 the
Danes were repulsed in an attack which they made on
London, because they took no heed of the bridge ; but
probably there is some misconception here, for it is recorded
that only in the previous year Anlaf, the Dane, sailed up
2 A LECTURE ON
the Thames as far as Staines with ninety-three ships and
ravaged the country. Discarding the tradition of a Roman
bridge, it seems clear that the first London Bridge of which
any record exists was erected between the years 993 and
1016, when Canute attacked London a second time, and made
a canal on the south side of the river so as to bring his ships
past the bridge. The first bridge was probably of wood, and
FIG 1. OLD PUTNEY BRIDGE.
somewhat of the form which many of us remember in the
cases of old Putney and old Battersea bridges, the piers being
formed of piles driven into the bed of the river, and the
openings spanned by timber beams. (The view (Fig. 1) is
taken from a photograph of old Putney Bridge, and iUustrates
that mode of construction.) This wooden bridge had great
vicissitudes. It was washed away in a flood in 1 091, was rebuilt
m 1097, and burnt in 1136. It was again rebuilt, but was
THE TOWER BRIDGE. 3
in so bad a state in 1163 that a new bridge was resolved on,
and this time in more durable materials.
A certain monk, Peter Curate of St. Mary, Colechurch
being known for his skill in bridge building, was employed to
erect the new bridge, which was begun in 1176 and finished in
1209. Peter of Colechurch did not live to see the completion
of his great work, but died in 1202.
The bridge though of stone, gave cause for anxiety early
in its history, for the structure was in bad condition in 1280,
and five of its arches were borne down by the ice and floods
after the great frost of 1282. When the design of the
bridge, with its 20 small arches and thick piers, is considered,
it will be seen what an obstruction it must have caused
to the flow of the water, and how serious the wash must
have been through it. In fact, we know that there was,
even without any exceptional flood, a fall of five feet on the
surface of the water through even the widest of the arches.
Apart from the danger to the structure from such a rush of
water, the impediment to river traffic must have been very
great, and shooting London Bridge in a boat must have been a
hazardous enterprise at most times of the tide.
The piers were built on a forest of piles, the tops of which
were about 9 feet above the bed of the river ; and either
originally, or as the necessity for protection to the piles was
recognised, starlings or projections all round the piers were
added. The starlings consisted of pilework, carried 3 feet above
the tops of the main piles of the bridge, the intervening
space being filled in with stone or chalk.
4 A LECTURE ON
The bridge at first was like modern bridges in being
unencumbered with houses, except that St. Thomas's Chapel
was built on its east side on the ninth pier from the north end.
The chap-dl was 65 feet long, 20 feet wide, and 14 feet high, and
it must have projected beyond the roadway on to the pier.
There was a drawbridge near the southern end, partly for
defence, but probably also for the passage of ships. To this
extent, it was prophetical of the Tower Bridge.
FIG. 2. OLD LONDON BRIDGE.
In 1426 a tower was built on the north side of the draw-
bridge to resist an enemy, and in 1471 a commencement was
made in building houses on the bridge. These were added to
from time to time, and amongst them was a famous structure
called Nonsuch House, which was a very beautiful building in
the architecture of its period, and was greatly admired. I
suppose that the houses were chiefly of wood ; for it is recorded
THE TOWER BRIDGE. 5
that, in 1632, forty-two houses were burnt, which were subse-
quently rebuilt ; and, again, in the fire of London in 1666, almost
all the houses were burnt, great lamentations being made over
their destruction. One would have thought that the bridge wa,s
rather improved by the catastrophe : but Londoners did not
think so in those days, for it appears to have been an important
business street, and the sites were let out to tenants at large
rents. In ten years' time the houses were all rebuilt, and the
bridge appeared as in the view (Fig. 2) which dates from 1725.
At that time the bridge was described as 9 1 5 feet and 1 inch
long. Observe the nicety of one inch in such a measurement.
Its height above the water was 43 feet 7 inches. The whole width
was 73 feet, of which only 20 feet was devoted to the roadway,
and 53 feet to the buildings on both sides of it. The approaches
to the bridge must have been very steep and inconvenient.
Such was old London Bridge, and so it existed till 1758.
It might, I think, be described as inconvenient for road traffic,
dangerous for river traffic, and more like an indifferent weir
than a bridge over a navigable river. Indeed, its functions
as a weir were utilised in 1582 ; for water wheels were erected
in several of the arches at the north end, by which water was
pumped for the supply of London.
Whatever we may think now of the ancient bridge,
people in the 18th century were proud enough of it, and
occasionally burst into poetry about it. Here is a translation,
made in 1725, of a Latin ode of the same date in its honour :
" When Neptune from his billows London spyed,
Brought proudly thither by a high Spring-tide,
(5 A LECTURE ON
As through a floating wood he steered along,
And dancing castles clustered in a throng ;
When he beheld a mighty bridge give law
Unto his surges and their fury awe ;
When such a shelf of cataracts did roar,
As if the Thames with Nile had changed her shore ;
When he such massy walls, such towers, did eye,
Such posts, such irons, upon his back to lye ;
When such vast arches he observed, that might
Nineteen Rialtos make for depth and height :
When the Cerulean God these things surveyed,
He shook his trident, and astonished said
Let the whole earth now all the wonders count,
This bridge of wonders is the paramount ! "
There is a little poetical license in more ways than one in
the ode ; for the Rial to has a span of 90 feet as compared with
the narrow openings of old London Bridge, varying from 10 to
30 feet in width and obstructed by the starlings round the
piers. You will also see that the local poet was even proud of
the cataracts, which endangered the lives of travellers in boats
when they shot the bridge with the current, and made it impos-
sible for any one to pass the bridge against the stream.
The houses on London Bridge or sufficient of them to
aiford a wide roadway and two good 'footpaths were ordered
to be finally pulled down in 1746.
Various designs were made from time to time for improving
London Bridge, and especially in respect of the waterway.
The views (Fig. 3) show two proposals, the upper one by
Labeleye, the other by Sir Christopher Wren. Neither design
was executed ; but that of Wren is a good example of
THE TOWER BRIDGE.
the grasp by that distinguished architect of the problem to be
dealt with, and would, if it had been carried out, have been an
immense improvement of the navigation.
In 1758-9 an arch 70 feet wide was made (by removing the
centre pier of the bridge) for the accommodation of the river
traffic : but before long it was felt that the old bridge had
outlived its day, and it was replaced by the present beautiful
FIG 3. LABELEYE'S & WREN'S DESIGNS FOR IMPROVING OLD LONDON BRIDGE.
bridge, by Rennie, which was begun in 1824 and opened in
1831, at a cost, including the approaches, of about 1,500,000.
The gradients leading to Rennie's bridge from the south are
unfortunately steep, being 1 in 27 from the Borough and 1 in
21 from Tooley Street, and are very distressing to the horses
of heavily laden wagons and omnibuses.
It is remarkable how long London existed and flourished
with only one bridge across the Thames ; for it was not till 1729
that any other bridge was built. This was Putney Bridge,
A LECTURE ON
already alluded to, which could, indeed, from its distance from
the town, scarcely be called a bridge to serve the needs of
London. Such, however, was the jealousy of the Corporation
of London that they espied in the project of building a bridge
at Putney an attack on the interests of London, petitioned
Parliament urgently against it, and prophesied that, if it were
authorised, the trade of London would wane and decay.
Putney Bridge cost the modest sum of 23,750, and lasted
till 1885, when the handsome stone bridge now existing was
built out of the rates, costing, with its approaches, about
600,000. For some years before the bridge was pulled down
it had a central opening, which did not exist in the original
structure, but was made by removing some of the old piers and
spanning the opening by iron girders, supported on iron cylinders.
;. 4. OLD WESTMINSTER BRIDGE.
Old Westminster Bridge (Fig. 4) followed soon after Putney
THE TOWER BRIDGE. 9
Bridge in order of date, an Act of Parliament having been
granted for it in 1735, and the bridge having been opened in
1750. It was 44 feet wide and 1,223 feet long. The centre arch
was 76 feet wide, and the whole structure was greatly admired.
The cost of it appears to have been about 390,000, which
sum perhaps includes the cost of the approaches. The engineer
was Charles Labeleye. It fell into decay from failure of the
foundations, which rested on timber caissons, and was replaced
by the present bridge about 1861. The view is taken from an
old print, and gives some idea of the old bridge in its palmy
days, before settlements and decay had set in.
Old Blackfriars Bridge (Fig. 5) was authorised in 1750, and
opened in 1769. It was not unlike Westminster Bridge and was
FIG. 5. OLD BLACKFRIARS BRIDGE.
built by Robert Mylne. As in the case of Westminster Bridge,
its foundation failed, and from the same causes. It was replaced
10 A LECTURE ON
in 1869 by the present fine bridge, constructed of iron and
stone, from the designs of Joseph Cubitt.
It is usual to hear, in these days of iron and steel
construction, that the old stone bridges had an element of
permanence in them greater than can be hoped for in the case of
those constructed of iron or steel. But it is a curious com-
mentary on this view that within our memory Westminster
and Blackfriars bridges had to be pulled down within 100
years of their construction, that the piers of Waterloo Bridge
have required to be strengthened, and that some of the founda-
tions of new London Bridge have given rise to anxiety.
The main reason of these failures or partial failures is, no
doubt, to be found in the increased scour of the river, due to
the removal of old London Bridge, which, as has been said,
acted very much as a weir, and prevented the free flow of the
tide. But it is doubtful whether this was the only cause.
Apart from the question of scour, there can be no doubt
that stone bridges bring a heavy weight on the foundations,
which therefore require great care. They also involve an
arched form of construction. The Eastern proverb says truly
that the arch never sleeps, and thus defective foundations in
an arched bridge involve serious results.
Another bridge of the 18th century was old Battersea
Bridge, which was similar to old Putney Bridge.
Vauxhall Bridge was opened in 1816, at a cost of 300,000.
It is a poor example of bridge building, and is unworthy of the
age which produced Southwark Bridge. It is already found
inadequate, and is to be replaced by a better structure.
THE TOWER BRIDGE.
1 1
Waterloo Bridge was designed by Rennie, and was opened
in 1817 at a cost of 800,000. It is a very fine work, but
is narrower than foresight should have predicted.
Southwark Bridge (Fig. 6), designed by the same engineer
as New London and Waterloo Bridges, was, at the time of
its erection, and still remains, an excellent example of iron
and stone construction. It was opened in 1819 at a cost of
FIG. 6. SOUTHWARK BRIDGE
800,000, but it also is narrow and the approaches are unduly
steep. In the view of Southwark Bridge old London Bridge
appears with the central arch as widened in 1759.
Proposals for Crossing the Thames beloiv London Bridge.
We see, from the above imperfect history of bridge building
in London during this period, that, though London Bridge alone
12 A LECTURE ON
served our forefathers for upwards of seven centuries, no less
than eight bridges were built in a single century, from 1730 to
1830. To this number five more bridges, exclusive of railway
bridges, were added in the next half century.
All this activity must have resulted from a great develop-
ment of South London, and no doubt aided immensely in that
development. But London Bridge still remained the bridge
lowest down the river, and but little was heard of any other
means of crossing the river below this spot till 1824, when Sir
Marc Isambard Brunei conceived the project of the Thames
Tunnel between Wapping and Rotherhithe.
The difficulties of that work are known to everybody.
There were frequent irruptions of water into the works, and it
was not till 1843 that the Tunnel was opened for foot traffic
only, the original idea of accommodating vehicular traffic having
been abandoned, owing to the expense which had already been
incurred, and which had amounted to about 468,000. The
Thames Tunnel was purchased about twenty-five years ago by
the East London Railway Company, whose trains now run
through it.
South London, probably, did not begin to extend much
below London Bridge till after Lambeth and Southwark had
first been developed. I came upon an old print lately (repro-
duced in Fig. 7) which indicates what Bermondsey was like only
90 years ago. It shows the ruins of the old and famous Abbey
of Bermondsey, which stood near the present Tooley Street, and
you will see that fields and hedges at that time extended in all
directions where now an enormous population lives and labours.
THE TOWER BRIDGE. 13
This is merely an instance of what is true of all South
London for miles down the river.
It has been stated, I believe correctly, that 39 per cent, of
the population of London now live east of London Bridge, and,
indeed, a town as large as Manchester and Birmingham put
together has rapidly grown up south of the Thames below
London Bridge, as is evidenced by any map of London of the
present day, which compares strikingly with the print of the
7. BERMOXDSEY ABDEY.
ruins of Bermondsey Abbey. The question of further com-
munication thus became more and more the subject of discussion,
and various attempts (apart from the Thames Tunnel already
referred to) were made to deal with the subject, commencing
with the Tower Subway.
The Tower Subway, which consists of an iron tube about
14 A LECTURE ON
7 feet in diameter, extending from Great Tower Hill (on the west
side of the Tower of London) to near Pickle Herring Stairs
on the Surrey side of the river, was constructed by a joint-
stock company, and was opened in 1871, Mr. Peter Barlow
being the engineer. It was originally intended to work the
traffic through the subway by means of a car drawn by
ropes, but, this plan having been found inconvenient, the
subway has been since used merely as a footway. In spite of
its small size and the inconvenience of its being approached by
long nights of stairs, nearly a million foot passengers have
passed through it per annum, at a charge of ^d. each.
FIG. 8. METROPOLITAN BOARD OF WORKS PROPOSED TOWEK BRIDGE.
In 1879 the Metropolitan Board of Works applied to
Parliament for powers to construct a high level, or rather medium
level bridge, on the site of the present Tower Bridge. This
bridge, designed by the late Sir Joseph Bazalgette, was
proposed to consist of a single arch of 850 feet span, supporting
a horizontal roadway at a height of 65 feet above high water,
and is shown in Fig. 8. It was intended to be approached on
the north side by a straight raised approach, rising from
opposite the Mint, and on the south side by a long spiral
approach, each with a gradient of 1 in 40.
THE TOWER BRIDGE. 15
The estimated cost of the bridge was 600,000 for works,
800,000 for property, total 1,400,000, and this expenditure
was to have been met out of the rates.
The proposed bridge was strongly opposed by shipowners
and wharfingers as being too low, and the long gradients (no
less than 5,700 feet in length of 1 in 40) being also thought very
objectionable, the scheme was rejected after a long enquiry.
In the Session of 1883 a proposal was made by a private
company for a large subway for vehicular traffic, from Great Tower
Hill to the Surrey side of the river. Access to this subway would
have been afforded exclusively by large and numerous hydraulic
lifts. The Bill was opposed by the Metropolitan Board of
Works, and was rejected. The subway would not have been
toll free.
In 1884 the Metropolitan Board of Works promoted their
Thames Crossings Bill, by which they sought to construct a
subway for vehicular and foot traffic under the Thames at
Nightingale Lane, about a mile east of the Tower. This subway
would have been approached by gradients 5,800 feet long of 1 in
40, or somewhat steeper. In the same Session an independent
company promoted a " Duplex " bridge at the Tower, on the
site of the present Tower Bridge. This bridge, as its name
implies, was a double one at the central portion, and its action
resembled that of a lock. The down river bridge being opened
would allow a vessel to pass up into the space between the two
bridges, and when this was done the down river bridge would be
closed and the up river bridge opened to allow the vessel to
proceed. Of course for vessels descending the river the operation
16 A LECTURE ON
would be reversed. The " Duplex" bridge would have been a
great obstruction in the river, and it would have been difficult
to manage vessels passing through it with a tide running as
it does in the Thames.
The two Bills were referred to a hybrid committee of the
House of Commons, who reported that neither Bill ought to be
proceeded with, and that a low level opening bridge at the east
side of the Tower was the best way of meeting the various
opposing views, and they invited the Corporation of London to
undertake the work.
Prior to 1884, there had been several proposals for low
level bridges at or near the site of the present Tower Bridge
one for example by Colonel Hay wood for a fixed bridge without
an opening span, others by Sir Douglas Galton and Sir John
Hawkshaw for an opening bridge of the ordinary type, with
horizontal girders revolving on a central pier. There were also
proposals by Sir G. Bruce for a bridge which would roll
transversely across the river on isolated piers, and various
modifications of Sir Joseph Bazalgette's proposal. Early among
these various suggestions was a proposal made by Sir Horace
Jones (the City Architect) for a bascule bridge.
I may mention, in passing, that the French word " bascule "
means a see-saw, and, as applied to a bridge, means a balanced
bridge, which can move up and down. The view (Fig. 9) shows
one of the old bascule bridges across a Dutch canal. It will
be seen that there are two beams balanced on two upright
posts, the inner ends of the beams being attached by chains
to a hinged platform across the canal, and the outer ends having
THE TOWER BRIDGE. 17
counter-balance weights on them. As the outer or landward
ends of the beams are lowered, the inner ends are raised and
pull up the hinged platform and thus open the bridge for the
passage of craft. Numerous examples of similar bridges exist
in this and other countries.
The Bridge House Estates Committee of the Corporation
of London, who, as representing that body, had given for many
Fie. 9. BASCULE BRIDGE ACROSS A DCTCH CANAL.
years much attention to the subject, appointed in 1884
a deputation to visit the Continent, Newcastle-upon-Tyne,
and other places where there were opening bridges of various
descriptions, and finally they came to the conclusion to recommend
the Corporation to accept the invitation of the hybrid committee
of 1884 and promote a Bill for the erection at the Tower site
of a " bascule " bridge as the best means of meeting the case.
18
A LECTURE
After the original sketches made by Sir Horace Jones,
^hich are reproduced in the two views (Figs. 10 and 11),
FIG. 10. SIR H. JONES'S DESIGN* OF A B.\scn,E BKIDGK (.shut against i-
FIG. 11. SIR H. JONES'S DESIGN OF A BASCULE BRIDGE (open for ships.)
it was seen that any arched form of construction across
the central opening would be very objectionable; as the
masts of ships would be in danger of striking the arch
unless they were kept exactly in the centre of the span.
THE TOWER BRIDGE. 19
Accordingly, various other sketches were made, and this was
the condition of affairs when in 1884 I became connected
with the undertaking as engineer, Sir Horace Jones being
at the same time appointed architect to the work, and we
decided that any girders over the central span, when open
for the passage of ships, must be horizontal and not arched.
The result of these considerations was an amended sketch
FIG. 12. DESIGN SUBMITTED TO PARLIAMENT.
shown in Fig. 12. In 1885 the Corporation, in accordance
with these ideas, promoted the Act of Parliament, of which the
present structure, though greatly altered from the original
sketch, is the result.
The problem to be solved is one of no small difficulty, for it
is necessary to reconcile the requirements of the land traffic with
the very important interests of the trade of the Upper Pool.
This part of the river is always crowded with craft of various
kinds, and this fact made the "bascule" system unusually
desirable. Any opening bridge revolving horizontally would
have occupied so large an area of the river as to be
undesirable from many important points of view, whereas
a bridge revolving in a vertical plane, not only occupies the
c
20 A LECTURE ON
minimum of space in the river, but also at an early stage of the
process of opening affords a clear passage for ships in the central
part of the waterway, increasing in width rapidly as the operation
of opening is continued.
The mode in which the traffic of the Pool is conducted pre-
scribed the general arrangement of the spans of the bridge.
Sea-going vessels of all kinds are moored head and stern in two
parallel lines in the Upper Pool, on each side of the centre line
of the river, leaving a central channel from 200 to 250 feet wide
free for the passage of vessels up and down the river, and this
space is frequently contracted, by barges and small craft lying
alongside the larger vessels, to a width of from 160 to 180 feet.
The spaces in the river occupied by the large vessels on
each side of the free central channel are called tiers, and as
vessels lie in the tiers two or sometimes three abreast, with
barges alongside them, it will be seen that if the piers of a
bridge were made alignable with the tiers there would be no
obstruction to navigation and little to the flow of water by two
piers of a width not greater than that of the tiers. On each
landward side of the tiers channels are preserved for the passage
of vessels to and from the wharves, and it was of course necessary
that these side channels should not be obstructed by any pier of
the bridge. Thus the mode, in which the river traffic has for
many years adjusted itself, made it evident that a bridge with a
clear central opening of from 160 to 200 feet, and two side
openings of about 280 or 300 feet, would meet all requirements,
and that there could be no objection to piers wide enough to
accommodate a counter-balance, seeing that the width of two
THE TOWER BRIDGE. 21
vessels lying in the tiers would be more than the width necessary
for such an extension of the moving girders into the piers, as
would provide for a sufficient counterpoise.
The proposals of the Corporation, however, encountered
a strong opposition in Parliament from the wharfingers, who
carried on business with sea-going vessels above the site of the
bridge, and the Bill was also opposed by the Board of the
Thames Conservancy. The Bill was referred to strong Com-
mittees of both Houses of Parliament, and eventually was
passed very much in the condition in which it was brought in,
but with several stringent clauses for the protection of the
interests of the river traffic.
With these few words on the principles that governed the
main features of the Tower Bridge, we will proceed to consider
the details of the structure generally.
General Description of the Bridge.
The Act of Parliament defined the leading dimensions of
the Tower Bridge to be as follows :
(1) A central opening span of 200 feet clear width,
with a height of 135 feet above Trinity high water when
open, and a height of 29 feet when closed against vessels
with high masts. (It may be mentioned in passing that
the height of the centre arch of London Bridge is 29|
feet above Trinity high water.)
(2) The size of the piers to be 185 feet in length
and 70 feet in width.
22 A LECTURE ON
(3) The length of each of the two side spans to be
270 feet in the clear.
The Act also defined the utmost permissible size of the
temporary stagings in the river.
The Conservators of the Thames, who very properly con-
sidered chiefly the importance of the river traffic, procured the
The doffed //nes she**"
re/sresonr oufs'de ///7 o
above toed of fttver .
(first fty' of Works)
.. ,__ t-le*- Hf*t*r_W*j[_
(Second stag* of Works)
#?/, 2. 3 . S.6.78 /J J4. /S /6. /9. Sf.
22 $ 23 *re Square Cussons
f/.' 9, /O. // /2, /7. IS. 2O. j Z4- JLT*
Tr/asiffuJa-r
FIG. 13. PLAN TO SHOW LIMIT OF TEMPORARY WORKS.
insertion in the Act of Parliament of a clause obliging the
Corporation to maintain at all times during the construction of the
bridge, a clear waterway of 160 feet in width, and this necessity
occasioned much delay in the construction of the permanent piers,
THE TOWEtl BfclDGE. 23
as the opening defined was too wide to permit of both piers being
constructed simultaneously. The plan (Fig. 13) shows the limits
of the temporary works as defined by Parliament. The outer lines
round each pier are the limits of the necessary stagings, and it
will be seen that, in order to give a navigable width at all times
of 160 feet, there could be only one staging at a time of the
full width required for building the piers.
The formal ceremony of commencing the works of the
Tower Bridge was performed by the Prince of Wales on 21st
June, 1886.
The Government authorities gave every facility for the
execution of the works, and, to enable the north approach to the
bridge to be made without interfering with very important
wharf property, allowed a small part of the Tower Ditch to be
occupied by a portion of the works. If this concession had not
been made, the cost of the land for the undertaking would have
been almost prohibitory. It was stipulated in return that the
design of the bridge should be made to accord with the archi-
tecture of the Tower, and at one time it was intended that the
new works should be made suitable for the mounting of guns and
for military occupation. The latter idea was afterwards to a
great extent discarded.
The piers of the Tower Bridge are essentially different from
the piers of an ordinary bridge, inasmuch as they have to contain
the counterpoise and machinery of the opening span, as well as
to support the towers which carry the suspension chains of the
fixed spans and the overhead girders above the opening span.
They are thus very complex structures, as will be seen by the
24
A LECTUBE
illustrations (Figs. 14, 15 and 16.) Theirfonn in plan (Fig. 15)
may be described as a square of 70 feet elongated by cutwaters
__ _. _
LONGITUDINAL SECTION OF PIER.
FIG. 14.
QUARTER PLAN AT A. 8.
THE TOWER BRIDGE.
25
at each end, bringing the total length to 185 feet 4 inches.
Their total depth from the roadway level to the London clay, on
which they rest, is 102 feet.
TRANSVKRSK SECTION Of PIER
x-
FHJ. 10.
We will first consider the form of the piers up to the level
of the roadway, which is 32 feet above Trinity high water.
Each contains (l) a large cavity to receive the landward end and
counterbalance weight of one leaf of the opening span ; (2) two
large chambers for the hydraulic accumulators ; (3) two chambers
for the machinery which actuates the opening span ; and (4)
two long tunnels, one for receiving the main pivot shaft on
which the leaf of the opening span revolves, and the other for
the pinion shaft by which power is transmitted to the opening
span from the machinery.
A diagram (Fig. 28, page 39,) will explain the method of
26 A LECTURE ON
actuating the opening span. The old bascule bridges of Holland
(Fig. 9) had their counterbalance above the roadway level,
mounted on posts at the abutments, and attached to the bridge
by chains or ropes. The dimensions of the Tower Bridge forbade
such an arrangement of an overhead counterweight, and the
counterbalance is there applied, as shown in Fig. 28, directly to
a prolongation of the girders of the opening span. These girders
turn on the main pivot, behind which a space, or cavity,
has been provided, to permit of the movement up and down of
the landward ends of the girders and the counterweight.
This space, which is called the bascule chamber or opening, is
in the form of a quadrant, and its dimensions are 50 feet from
north to south, 44 feet from east to west ; it is 50 feet in height
next the central or opening span, diminishing to nothing next
the landward or fixed span of the bridge. The two machinery
chambers are each 35 feet by 30 feet, and 10 feet high, and the
two chambers for accommodating the accumulators are each 30
feet by 20 feet 4 inches, and are 50 feet in height, extending
from below the floor of the machinery chamber to within
26 feet of the bottom of the foundations.
Before describing the mode in which the substructure of
the pier was constructed, it will be best shortly to describe
the general arrangement of the remainder of the fixed portion
of the bridge.
The mode adopted for spanning the landward openings is
by suspension chains, which in this case are stiffened. The
chains are anchored in the ground at each end of the bridge,
and united by horizontal ties across the central opening at a
THK TOWER BRIDGE. 27
high level (Fig. 30, page 49). These ties are carried by two
narrow bridges 10 feet in width, which are available as foot
bridges when the bascule span is open for the passage of vessels.
The foot bridges are 140 feet above Trinity high water, and, as
their supports stand back 1 5 feet from the face of the piers, their
clear span is 230 feet. Access is given to them by hydraulic
lifts and by commodious staircases in the towers.
Above the landings, at the tops of the stairs and on which
the foot passengers land from the lifts, come the roofs of the
towers, the tops of which are 162 feet above the roadway level,
or 264 feet from the bottom of the foundations.
Having now given the leading dimensions of the structure,
I will proceed to describe (l) the mode in which the piers were
constructed up to the roadway level ; (2) the details of the
opening span and machinery ; (3) the details of the fixed super-
structure, namely, the towers, the suspension chains, and the
overhead footways ; (4) the mode of erection of the super-
structure.
Tlie Mode of Constructing the Substructure of the Piers.
Iron caissons strutted with strong timbers were used in
excavating the bed of the river and building the foundations
of the piers. During these operations the external pressure of
the water and earth surrounding the caissons was very great,
as there is a depth of 32 feet of water at high tide at this part
of the river, and the caissons had to be carried about 21 feet into
the bed of the river to secure a good foundation. The caissons
employed were boxes of wrought iron, without either top or
28
A LECTUHE ON
bottom, and with the bottom edges made sharp and strong (Fig.
18,) so as to easily penetrate the ground.
There are twelve caissons for each pier, as will be
THE TOWER BRIDGE.
understood from Fig. 13. Those about the central parts
of the pier are 28 feet square in plan, and those near the
cutwaters are triangular in plan, the dimensions being 35
feet by 33 feet 8 inches. Fig. 17 shows a square caisson and a
triangular caisson in plan, with their timbering and other details.
I will describe the mode of sinking one
caisson, and the description will apply more
or less to all, though, of course, the circum-
stances attending the different caissons
required some differences of treatment.
The temporary staging for the pier
having been made with piles, the next
operation was to erect the caissons upon
the staging.
ROLLED STEEL CUTTING FOGE.
Fio. 18.
HALF CROSS SECTION OP PIER. ' HALF CROSS SECTION OF PIER,
SHOWING CAISSON BEFORl LOWERING SHOWING CAISSON WHEN SUNK
TO THE BED OF THE RIVER. 6 BEING FILLED WITH CONCRETE.
FIG. 19.
The bottom part of the caisson, having to be sunk deep into
the bed of the river, could not be removed on the completion of
the pier, and was thus named the permanent caisson. The
purpose of the upper part of the caisson was merely to exclude
30 A LECTURE ON
the water during the process of building the pier, and it could
be removed when the brickwork and masonry were finished.
This part was thus called the temporary caisson.
The permanent caisson was 19
feet in height, divided horizontally
into two portions. It was erected
on timber supports, which were
slightly above low water mark (Fig.
19), where it was rivetted together
and firmly strutted inside with strong
timbers, 14 inches square. It was then
lifted slightly by four powerful screws
attached to four rods, from which was
slung the weight of the caisson and the
timbering in it. The timber supports
were removed, and the caisson was lowered by the screws on to
the bed of the river, which had previously been levelled by divers.
After the permanent caisson reached the ground various
lengths of temporary caisson were added to it till the top of the
temporary caisson came above the level of high water. The
junction between the permanent and temporary caissons was
made with india rubber, as shown in Fig 20.
Divers working inside the caisson, excavated first the
gravel and then the upper part of the clay forming the bed of
the river, and as they dug away the soil, which was hauled up
by a crane and taken away in barges, the caisson gradually sank,
until at length its bottom edge penetrated some 5 to 10 feet into
the solid London clay.
PETAI L OF
INDIA-RUBBER JOINT.
FlG. 20.
THE TOWER BRIDGE. 31
London clay is a firm, water-tight stratification, and, when
the above-mentioned depth was reached, it was safe to pump out
the water, which up to this time remained in the caisson, rising
and falling with the tide through sluices in the sides. The
water having been pumped out, navvies proceeded to the bottom
of the caisson and dug out the clay in the dry.
Additional lengths of temporary caisson were added as the
caisson sank, so that at last each caisson was a box of iron, 57
feet high and of the dimensions above stated, in which the
preparation of the foundations could be commenced. A detailed
view of one of the completed caissons is given in Fig. 21.
I may mention in passing how important it is, in order to
ensure success in sinking caissons or cylinders, that they should
be controlled from above and be prevented from sinking unevenly.
It is very easy to prevent a caisson from going wrong (like many
animate subjects as well as inanimate) by timely control, but it
is a very different thing to put the matter right when a wrong
course has been pronouncedly taken.
London clay being peculiarly hard and uniform in texture,
advantage was taken of this circumstance to increase the area of
the foundations by digging out sideways or undercutting
below the edge of the caisson, as shown at the bottom of Fig. 21.
The caisson having been controlled from the first by the
suspending rods to which allusion has been made, its descent
any further than was desired was easily arrested by the rods,
when the bottom of the caisson was 20 feet below the bed of
the river. The clay was then excavated 7 feet deeper than the
bottom of the caisson, and outwards beyond the cutting edge
32
A LECTURE ON
28 FEET CAISSONS .
HALF INSIDE ELEVATION. HALF OUTSIDE ELEVATION.
1 1
FIG. 21. COMPLETE CAISSON WITH TIMBERING AND SUSPENSION RODS.
THE TOWER BRIDGE.
33
for a distance of 5 feet on three of the four sides of the caisson.
In this way not only was the area of the foundations of the pier
enlarged, but as the sideways excavations adjoined similar
excavations from the next caissons
(Fig. 22), the whole foundation was
made continuous. The whole of the
permanent caissons with the spaces
between them were then completely
filled with concrete, upon which the
BETWEEN W TWO ADjo", C N U G TT cA ( lssoNi. brickwork And masonry were com-
FIG. 22. , . . ,
menced in the temporary caisson, and
carried up to 4 feet above Trinity high water, as shown
in Fig. 23.
Fiu. 23. CROSS SECTION OF PIER, SHOWING OUTSIDE WALL COMPLETED.
It was not desirable to build isolated portions of the brick-
work and masonry, even if they were joined together afterwards.
Accordingly the temporary caissons were so designed as to admit
of their sides being removed (Fig. 15) and of the whole area
enclosed by their front and back plates being thrown together to
34
A LECTURE OX
permit of continuous building. For this purpose the corners of
the caissons were united by timber piles, which were driven
in a groove on each caisson (Fig. 24), and when these
had been driven and made water-tight (as to which no difficulty
Xolca Cor- 1 "
ZK-ct afxirV
' uu/ IftujtJiJt cf Gtlascfn'.
LI.3-X 3fi
DETAIL OF VERTICAL dOlNTS
AND PILE GROOVES
FIG. 24. ENLARGED PLAN OF THE ANGLE OF A CAISSON.
occurred), the sides of the temporary caissons were removed.
In this way the outside portions of the piers were built, and
eventually formed a continuous ring of a strong masonry wall,
THE TOWER BRIDGE.
35
water-tight and able to resist the external pressure of the
water. (Fig. 25). The foundations of the central portion of
the pier, enclosed by the outside walls, were then excavated
arid the pier completed.
N CAISSONS BEFORE DRIVING THE P.LES,^
BETWHN THC CAISSONS /
FIG. 25. HALF PLAN OK PIER AT DIFFERENT PERIODS OF CONSTRUCTION.
The abutments of the bridge were built within ordinary
coffer dams, and, though formidable in size and depth, presented
no new features of construction such as have been explained
with regard to the piers.
When the piers and abutments had reached a height of
4 feet above high water, the first contract was finished, and new
contracts for the superstructure were let.
36 A LECTURE ON
The work of the foundations was troublesome and tedious,
owing to the isolation of the piers, and still more to the
great amount of river traffic, rendering the berthing of barges
difficult. The substructure thus occupied a considerably longer
time than was anticipated.
FIG. 26. VIEW OF THE FOUNDATIONS IN PROGRESS.
The view Fig. 2G gives an idea of the appearance of the
works during the construction of the piers.
The Opening Span.
The stipulated dimensions of the opening span have been
already given, as providing, when the bridge is open for ships, a
clear waterway of 200 feet in width, with a clear height
throughout the 200 feet of 135 feet (which has been increased in
THE TOWER BRIDGE. 37
construction to 140 feet) from Trinity high water mark. I may
mention in passing that I think these dimensions constitute the
largest opening span in the world. The next largest opening
is, I believe, at the Newcastle bridge, where there are two
separate spans of 100 feet each.
The opening span of the Tower Bridge consists, as already
explained, of two leaves, each turning on a horizontal pivot of
solid forged steel. This pivot, which rests on live rollers, is
1 foot 9 inches in diameter, and weighs 25 tons. Each leaf is
formed with four main longitudinal girders, 13 feet 6 inches apart
from centre to centre, which together provide for a clear width of
bridge between the parapets of 50 feet. This width will be
divided into 32 feet for a roadway (which is sufficient for four
lines of road traffic) and for two footpaths each 9 feet wide. The
spaces between the longitudinal girders are filled with cross
girders and roadway plates, on which will be laid the wood
pavement of the roadway and footpaths.
The total length of the moving girders resting on each pier is
162 feet 3 inches (Fig. 27). They turn on the main pivot, the
position of which is 12 feet 9 inches back from the face of the
pier. Thus from the centre of the main pivot riverwards to the
end of the girder is 112 feet 9 inches, and to the end of the
girders landwards is 49 feet 6 inches. The centre of gravity of
the part over the river is 48 feet from the centre of the main
pivot, and the estimated weight of this portion of the semi -span
is 424 tons. A counterbalance box is attached to the landward
end of the girders, and is intended to be filled with 422 tons of
iron and lead. The centre of gravity of this portion of the
gg A LECTURE ON
girders and the counterbalance weights will be 32 feet 9 inches
from the centre of the main pivot, and the total weight land-
ward of the main pivot will be 621 tons. Thus the total
weight of each leaf of the opening span, resting on the main
pivot when the bridge is being moved, 'or on the blocks
when the movement is completed, will be the sum of the two
weights given above, or 1,045 tons. The pier has, therefore,
to carry on its riverward face a heavy load, which is distributed
on the masonry by a longitudinal and several cross girders.*
>.
i -
r---fe-49 6 *-- - ur H2-9
LUX*
FIG. 27. DIAGRAM OF WEIGHTS OF BASCULE.
The two outside girders carry at their counterbalanced ends
a quadrant (Fig. 28), on which are fixed teeth, and these teeth
engage with similar teeth on revolving pinions, which are
actuated by the hydraulic machinery. It will be seen that, when
the pinions are turned round, the whole of each leaf of the
* The various weights above given are matters of estimate. They may be slightly
modified in execution.
THE TOWER BRIDGE.
39
opening span is made to rise or fall at will. It may be
mentioned that there are two distinct pinions to each quadrant,
making four pinions to each leaf, and that any one of the four
pinions is strong enough to actuate and control the whole leaf.
|_ 100.0
FIG. 28. DIAGRAM OF BASCULE TO SHOW QUADRANT OR SECTOR.
It will be observed that the landward ends of the
main girders are bent downwards, and extend backwards into
the bascule chamber in the pier, so as to carry the counterbalance
40 A LECTURE ON
box ; but it is, of course, necessary that the road and footpaths
should extend continuously up to the opening span. This is
effected by eight fixed longitudinal roadway girders, which carry
the roadway over the bascule chamber, and between which the
longitudinal moving girders can rise and fall. Where the
moving roadway over the central span adjoins the fixed roadway
over the bascule chamber, there must be, of course, a cross
slit or joint. This space is covered by a hinged flap, which
rises and falls with the movement of the moving girders.
Such are the main features of the opening span, and
before leaving the subject, I will give a short account of the
actuating machinery in its main features. It will be, of course,
impossible to describe it within the time now available in any
minute detail.
Some of my hearers may not be acquainted with the
leading principles of hydraulic machinery in general, and so I
may be excused if to others my remarks appear somewhat
elementary. The principle of the application of water to
hydraulic engines is shortly as follows. First of all, a pump is
required, which is powerful enough to pump water under great
pressure in the present instance under a pressure of about
850lbs. on every square inch. This pressure is nothing unusual,
but its magnitude will be appreciated when we remember that in
the boiler of a locomotive engine the steam pressure is usually
not more than about one-fifth of the above amount.
The water pressure, being so high per square inch, enables
the pistons and pipes of hydraulic machinery to be comparatively
small, and, as the water can be conveyed anywhere in pipes of
THE TOWER BRIDGE. 41
suitable strength, the power produced by the pump can be
applied at any desired place. In the present case the pumps
are on the Surrey side of the bridge, and are actuated by two
steam pumping engines of 360 horse power. The high-pressure
water is brought thence by pipes along the Surrey fixed span,
up the Surrey tower, across the high level foot bridge, and down
the Middlesex tower. Return pipes convey the water when it
has exerted its pressure at the hydraulic engines, which are
fixed below the footway on each pier back again to the pumps
at the steam engines, where it is again subjected to pressure
and made ready for work again.
There is, however, one important feature of hydraulic
machinery to be explained, and that is the accumulator. The
object of the accumulator is, as its name suggests, to accumu-
late power, which is effected as follows. We have seen
that the origin of the power is the steam engine actuating
a pump or pumps, and, if the machinery were always working
and always requiring the same amount of power, an accumulator
would be of comparatively little use ; for the pumps could pump
the water direct to the hydraulic engines, to be used there in a
continuous effort. But such is not the requirement. The
hydraulic machinery at the Tower Bridge, and in almost every
other application of hydraulic power, is not called upon for a
continuous effort, but, as it were, for spasmodic efforts lasting
over a certain number of minutes, and the accumulators are
employed to store up the power provided by the pump, in order to
give it out at a greater speed than the pump, though working
continuously, would provide.
42
A LECTURE ON
In accordance with these principles, the accumulator (Fig.
29) is a large cylinder, in which fits a long plunger, on the top
of which are placed weights, which bring a pressure of 850lbs.
on every square inch of the area of the cylinder. The steam
pumps pump water into the accumulator, with that pressure, and
STEAM EJVG/fJS
Fio. 29. DIAGRAM OF APPLICATION OF ^HYDRAULIC POWER.
as it is a principle of hydraulics that any pressure applied to
water in a closed vessel is communicated to the whole of the
water in that vessel, the pumps, though much smaller than the
large plunger of the accumulator, raise it with its superincum-
bent weight, but proportionately slowly.
The pipes to work the hydraulic engines are in communi-
cation with the cylinder of the accumulator, and this, being of
great internal capacity, can supply high pressure water with
great rapidity to work the hydraulic engines for what I have
called their spasmodic efforts. In the meantime, the pumps are
working away to re-supply the accumulator.
At the Tower Bridge there are six accumulators, viz., two
near the pumps on the Surrey side of the river and two on each
THE TOWER BRIDGE. 43
pier. Their capacity is ample for the most liberal demands of
the hydraulic engines.
It will not be possible to describe the hydraulic engines
themselves in any detail. It will be sufficient to say that there
are two sets of engines on each pier, and that their leading
principle is that of reciprocating cylinder engines. Rams in
cylinders are worked to and fro by the high pressure water, and
by cranks communicate rotatory motion, through a series of geared
wheels, to the toothed pinions, to which allusion has already been
made, and which actuate the quadrants of the openmg span.
Each engine has three rams with a stroke of tt inches ;
those on the smaller engines are 7^ inches in diameter, and
those on the larger engines 8|- inches in diameter.
When the Tower Bridge was being discussed in Parliament,
the disaster to the Tay Bridge was fresh in the minds of many,
and some alarm was expressed lest the machinery might not be
strong enough to control the opening span in heavy winds.
The Board of Trade had reported with regard to the Tay
Bridge, that provisions should be made in all future structures
for a wind pressure of 56lbs. per square foot, and, though this is,
I think, an excessive estimate, even for the wind in such an
exposed place as the Firth of Tay, and is much more excessive
in the comparatively protected position of the Tower Bridge, it
was considered right not only to provide for the extreme
pressure of 56lbs. per square foot, but also to provide the
machinery of this strength in duplicate on each pier. Thus we
have machinery equal to twice the requirements of the Board of
Trade, An ordinarily strong wind, however, will riot give a
44 A LECTURE ON
pressure exceeding about 17lbs. per square foot, and therefore
the hydraulic engines are arranged in pairs, one engine of each
pair exerting a power equal to a 17lbs. wind, and the other
equal to a 39lbs. wind, the two together being equal to the
extravagant pressure of a wind of 56lbs. On each pier the
duplicate pair of engines follow up the work of the first pair
and provide against breakdowns.
I should mention that when the two leaves of the opening
span are brought together, there will be long wedge-shaped
bolts, actuated by hydraulic machinery, fixed on one leaf and
shooting into the other leaf, to complete the union of the two.
All the machinery of the opening span will be worked from
cabins on the piers, in which there will be levers like those in a
railway signal box, so interlocked one with the other that all
the proper movements must follow in the arranged order.
The time required for the actual movement of the opening
span from a position of rest horizontally to a position of rest
vertically is estimated at about 1% minutes. To this must be
added the time necessary for stopping the road traffic and clearing
the bridge, and withdrawing the bolts. This may take, perhaps,
some 1 J minutes more, and we then have to add the time for
the passage of a ship and the lowering of the bridge. The time
of Ij- minutes for opening or shutting the bridge gives a
mean circumferential speed at the extremity of each leaf of 2 feet
per second, which is a moderate speed for an opening bridge.
Signals will be provided by semaphores by day and signal
lamps by night, to show ships whether the bridge is open or
shut. By night when the bridge is open for ships, four green
THE TOWER BRIDGE. 45
lights will be shown in both directions, and when it is shut
against ships four red lights will be similarly exhibited, and
then- lights will be interlocked with the machinery, so that
wrong signals cannot be shown. By day similar intimation will
be afforded by semaphore arms on the same posts as those which
carry the signal lamps. During foggy weather, a gong will be
used in specified ways.
One other part of the machinery remains to be mentioned.
This is that of the passenger lifts between the roadway level and
the high level foot bridge.
There are two lifts in each tower, consisting of a cage, 13
feet by 6 feet, and 9 feet high, raised and lowered by an ordinary
hydraulic ram with chain gearing, and capable of lifting 20 to 25
passengers in about lj minutes, including the delays of opening
and shutting the doors. As the lift will have to descend carrying
a cargo of passengers before it can take a second load of ascend-
ing passengers, we may assume three minutes from one start to
the next ; or, as there are two lifts on each tower, 1^ minutes.
In addition to the lifts, there are ample flights of stairs
in the towers.
The Fixed Superstructure.
The fixed parts of the superstructure of the Tower Bridge
consist, as has been said, of two shore spans, each of 270 feet,
and of a central high level span of 230 feet. The fixed bridge
is of the suspension form of construction, and the chains are
carried on lofty towers on each pier and on lower towers on
each abutment.
I will first describe the towers. When an opening bridge
46 A LECTURE ON
was first proposed there was some outcry by aesthetical
people on the score of its ruining the picturesqueness of the
Tower of London by hideous girder erections, and it seemed to
be the universal wish that this bridge should be in harmony
architecturally with the Tower.
You will have seen the various architectural ideas evolved
out of these considerations, and observed that it was originally
intended that the towers should be of brickwork in a feudal
style of architecture, and the bridge somewhat like the draw-
bridge of a Crusader's castle. Subsequently, Sir Horace Jones
suggested a combination of brick and stone, and towers of a
form similar to that of the view Fig. 12 on page 19.
The ideas were in this condition when I was appointed
engineer to the scheme, with Sir Horace Jones as architect, and
the Corporation went to Parliament for powers to make the
bridge. Sir Horace Jones unfortunately died in 1887, when
the foundations had not made much progress, and up to
that time none of the architectural designs had proceeded further
than such sketches and studies as were barely sufficient to
enable an approximate estimate to be made of the cost. Since
the death of my coadjutor I have preserved the general
architectural features of the Parliamentary sketch designs,
but it will be seen that the structure as erected differs
largely therefrom, both in treatment and material.
The width, and consequently the weight, of the bridge was
increased by the requirements of Parliament, and the span of
the central opening was enlarged from 160 feet, as originally
intended, to 200 feet. At the same time, the provision of
THE TOWER BRIDGE. 47
lifts and stairs, to accommodate foot passengers when the
bridge was open, was felt to be a necessity.
In this way it became apparent that it would not be
possible to support the weight of the bridge on towers wholly
of masonry, as in the first designs, unless they were made of
great size and unnecessary weight. It was, consequently,
necessary that the main supports should be of iron or steel,
which could, however, be surrounded by masonry, so as to
retain the architectural character of the whole structure.
It was clear that in any event a large part of the steelwork
of the towers must be enclosed in some material, for the
moving quadrants project upwards some 40 feet from the
level of the roadway, while the stairs and lifts also required
protection from the weather. It thus became a question of
surrounding the towers either with cast iron panelling or with
stone, and eventually a granite facing with Portland stone
dressings was adopted.
^Esthetically speaking, stone seems better than cast iron,
which would equally hide the constructive features, and
practically speaking I think it also better, for I know of no mode
so satisfactory for preserving iron or steel from corrosion as
embedding it in brickwork, concrete or masonry. Careful pro-
vision has been made in all parts for expansion and contraction
of the two materials, and, though we have had great extremes
of heat and cold since the masonry has been built, no effects
resulting from any difference of temperature have been observed.
The Fig. 30, which is a half elevation of the steelwork of
the bridge, shows the general arrangement of the towers,
48 A LECTURE ON
which have been since enclosed with masonry and brickwork,
and may be described as being steel skeletons clothed with
stone.
I am afraid some purists will say that the lamp of truth
has been sadly neglected in all this, and that the old architects
would not have sanctioned such an arrangement as a complex
structure of steel surrounded by stone.
One reason may be that the old architects did not know
much about iron or steel. Perhaps, if they had been acquainted
with their capabilities, they would have been as ready to
employ them as they were to back up stone-faced walls with
brick, and to hide the constructive features of their buildings as
Sir Christopher Wren did when he used a brick cone to support
the internal and external domes of St. Paul's.
However all this may be, " needs must when Parliament
drives," and, if the appearance of the Tower Bridge is approved,
I hope that we may forget that the towers have skeletons as much
concealed as that of the human body, of which we do not think
when we contemplate examples of manly or feminine beauty.
I return, however, to the details of construction. The
skeleton of each tower consists of four wrought steel pillars,
octagonal in plan, built up of rivetted plates. The pillars
start from wide spreading bases, and extend upwards to the
suspension chains, which they support. They are united by
horizontal girders and many diagonal bracings, to which it is not
necessary here to refer in detail. The chains are carried on the
abutments by similar but lower pillars. All these and other
particulars appear in Fig. 30.
THE TOWER BRIDGE.
49
50 A LECTURE ON
Between the pillars are spaces for the public stairs and the
passenger lifts, and for the quadrants of the opening span when
in their upward position. When all these necessary things are
accommodated, it will be seen that there is very little room left
in the towers for the first 40 feet of their height, that is to say,
up to the level of the archway over the road. The horizontal
girders in the towers above the archway carry various floors to
provide for landings from the public stairs and possibly for
rooms for the police and staff.
On the tops of the octagonal pillars rest a series of rollers,
which will allow the chains so to move as to accommodate
themselves to changes of temperature and to unequal distribu-
tion of the road traffic.
The arrangements of the rollers are peculiar. The amount
of space available for their reception on the tops of the steel
octagonal pillars is limited, while the weight which they have
to support is very large, being estimated when the bridge was
fully loaded at 1,000 tons. It was thus necessary that the
weight should be equally distributed over the whole series of
rollers, and that there should be no possibility of a concentration
of weight on any one or two rollers. This is effected by an
arrangement by which the weight brought on the top of the
system is first carried by two blocks, thus ensuring equal division
between them, then the weight so divided is sub-divided
between two plates, and the weight on each plate is again sub-
divided between two rollers.
The main chains, which are 60 feet 6 inches apart from
centre to centre, extend from the rollers on the piers to other
THE TOWER BRIDGE. 51
rollers on each abutment, and support the platform of the bridge
by suspension rods, extending from the bottom of the chains to
the cross girders of the platform. These arrangements appear
in the view, and do not call for any further detailed description.
It may be asked why are these structures, which look like
girders, called chains? They are, in fact, chains, stiffened to
prevent deflection, and the object of the form is to distribute the
local loads due to passing traffic, which, in the case of an ordinary
suspension chain, distort the chain, continually depressing each
part as the load passes, and consequently distorting the
platform of the bridge. By making the chain, as it were, double,
and bracing it with iron triangulations, these local deflections of
the chain are avoided.
The ends of the chains on the abutments and on the towers
are united by large pins to the ties. The ties on the abutments
are carried (4S5S? the ground below the approaches, and are there
united to anchorage girders, which rest against very heavy
blocks of concrete, and are abundantly adequate to resist the pull
of the chains. The ties between the towers are for the purpose
of uniting the ends of the two chains, and by their means the
stress on the chains is conveyed from anchorage to anchorage.
The platform of the bridge is formed of cross girders,
which extend transversely from side to side, their ends being
immediately under the chains, to which they are suspended by
solid steel rods. Between the cross girders are short longitudinal
girders, and on these rest corrugated steel plates, which carry
the paving and footways. The total width between the parapets
on the fixed spans of the bridge and on the approaches is 60 feet,
52
A LECTUBE ON
Fu:. 31. PROGRESS OF TOWER ON PIER ix RIVER.
THE TOWER BRIDGE. 53
which is divided into 36 feet for the vehicular traffic and into
two pathways each 1 2 feet wide. I may mention in passing that
London Bridge is 54 feet wide between the parapets.
The total weight of steel and iron in the Tower Bridge will
amount to nearly 12,000 tons.
The Erection of the Superstructure.
The erection of the superstructure of the bridge has been
effected without any considerable difficulty. Temporary stagings
of pile work were erected in the river forming piers, upon which
wrought iron horizontal girders were erected which supported
timber beams and planking thus forming continuous tem-
porary bridges communicating from the shores to each pier.
Less difficulty .was experienced than was expected in
driving the pile work either round the permanent piers or to
form the supports of the temporary bridge, and but little
damage was ever caused to it during the time of the erection
of the bridge by collisions with it of river craft, which from
time to time occurred. Access being thus given from the
shores, the next thing to be done was to erect the pillars,
girders and bracing, forming the steel framework of the
abutment and river towers. All the steel work had been put
together at the works of Messrs. Sir W. Arrol & Co., at
Glasgow, and was brought thence by sea in small pieces to be
rivetted together at the bridge. The view (Fig 31) shows this
work in progress at one of the river towers.
The next work taken in hand was the erection of the high
54
A LECTURE ON*
level footway bridges between the towers. These bridges when
finished are cantilevers for a distance from eacli tower of 55 feet,
and are girders for the remaining space of 120 feet between the
ends of the cantilevers. They were erected, however, by
FIG. 32. PROGRESS OF HIGH LEVEL_BRIDGES.
temporary expedients, as cantilevers, piece by piece from the tops
of the towers and without scaffolding from below, for the whole of
the semi-span of 1 1 5 feet till they met over the centre of the river.
The progress of the girders from week to week as they advanced
towards each other, was watched with much interest by the
public from London Bridge. A view (Fig. 32) shows the girders
a short time before they met.
THE TOWER BRIDGE. 55
The chains were erected in their position by means of
scaffolds and trestles resting on the temporary bridge. The
succeeding views (Figs. 33, 34 and 35) show this work in
progress. Cranes on stages, travelling along the temporary
FIG. 33. PROGRESS OF ERECTION OF MAIN CHAINS. SHORT SEGMENT.]
wooden bridge, served to place the various parts of the chain on
the trestles, where they were rivetted together in their
permanent positions. In the meantime the land ties from the
anchorages had been brought up to the tops of the abutment
towers, and the long horizontal ties, which are carried by the high
level bridges, had been erected and rivetted in their places. The
holes for the connecting pins at the ends of the ties and at the
56
A LECTURE ON
junctions of the short and long chains were then bored finally
and the pins inserted, thus forming a through connection from the
i'lo. 34. FEOGRESS OF J^KECTION 01- MAIN LUA
anchorage on the north side of the river to the corresponding
anchorage on the southern side.
THE TOWER BRIDGE. 57
Another part of the work was the erection of the fixed
roadway girders and the moving girders on the river towers.
The fixed girders were first erected on the temporary bridge,
and moved from thence into their position over the bascule
chamber. The portions of the moving girders which would
eventually be landward of the main pivot and extend about 10 feet
riverwards therefrom were similarly erected on the temporary
FIG. 35. PROGRESS OF ERECTION OF MAIX CHAINS. GENERAL VIEW.
bridge, launched forward, and placed in a vertical position. The
main pivot could now be threaded through the moving girders
and the bearings on it adjusted. A sufficient length of the moving
girders to reach about 40 feet over the central span of the
bridge was then erected vertically, with the accompanying cross
girders and bracing, after which the quadrants to carry the
58
A LECTURE ON
teeth by which the moving span was to be actuated, were erected
in their places. The teeth of the quadrants were bolted on
to them, and, to ensure an accurate fit, with the pinions, which
engage with the teeth on the quadrants, the moving girders
were revolved by temporary means, so that the teeth on the
quadrants could be fitted carefully to those of the pinions. The
view (Fig. 36) shows this work in progress. The regulations of the
FIG. 36. MOVING GIRDERS IN PROGRESS.
Act of Parliament rendered it necessary to confine operations in
the first instance to a length of 40 feet of the moving girders, so
as not to reduce the free passage for vessels through the central
opening during the process of adjustment to less than 160 feet.
When the adjustment was completed the erection of the
remaining portions of the moving girders was taken in hand
and the work completed to a height of about 100 feet above the
THE TOWER BRIDGE.
59
roadway where the ends of the girders are when in their
vertical position.
The succeeding view (Fig. 37) shows the general appaarance
which the bridge will present when opened for the passage of
sea-going vessels. It will be observed that the tops of the
towers and the ends of the moving girders are incomplete, and
FIG. 37. VIEW OF THE BRIDGE IN THE AUTCMX OF 1893.
that part of the temporary staging in the river, on which the
steelwork of the side spans was erected, is still standing.*
I am glad to say that the loss of human life during the
construction of the bridge has, considering the magnitude and
nature of the works, been small. In all, six men have met with
fatal accidents, and at least one of these was the result of
sudden illness, or of a fit.
A more recent view of the Bridge, taken early in 1894, i? given as a frontispiece.
GO A LECTURE ON
The Approaches.
The northern approach to the bridge is constructed partly
on a viaduct of brick piers and arches, faced with stone, and
partly by means of retaining walls, and extends from the northern
abutment to Tower Hill, opposite the Royal Mint, a distance of
330 yards. The ground on which these works stand was
acquired from Government, and was the glacis and part of the
eastern ditch of the Tower. An entrance to the Tower property
from the east is afforded by a wide archway beneath the
approach at its southern end. Some of the arches of the viaduct
adjoining this entrance are used for a guard room, and others
for stores for the fortress and for the bridge. Most of the
northern approach is level, and there is a gradient of 1 in 60
on the remainder, extending across the north shore span to the
northern pier of the bridge. The southern approach stands
wholly on property acquired from private owners, and extends
from the southern abutment to Tooley Street, a distance of 280
yards. It is partly constructed on a viaduct near the bridge, and
under the arches of the viaduct are placed the engine and
boiler houses with the coal stores for the hydraulic pumping
engines. The rest of the southern approach is upheld by
retaining walls, built go as to form cellarage for houses to be
built on each side of the approach. The gradient of the
southern approach is, throughout its length, 1 in 40, arid this
gradient extends across the south shore span to the southern
pier of the bridge. From the piers the inclines are continued to
the centre of the river by gradients of 1 in 75, on both leaves of
the opening span.
THE TOWER BRIDGE. 61
Communication betiveen existing thoroughfares and the
Tower Bridge.
The approach to the bridge on the north side of the river
connects directly with the southern end of the Minories,
where that important street, which runs north and south, joins
wide thoroughfares extending eastward and westward, but the
access from the southern bank of the river is not at present so
satisfactory. The approach constructed by the Corporation
extends from the bridge to Tooley Street, which is an important
street running east and west, but there is no good thoroughfare
in a southerly direction from Tooley Street. The London
County Council undertook to make a wide southern approach,
extending from Tooley Street to the Old Kent Road and
Bricklayers' Arms Station ; but up to the present time nothing-
has been done, except to bring into Parliament bills which have
been abortive, from the fact that the Council had included
in them the principle of what is called betterment. It is much
to be hoped that this greatly needed southern approach will be
made expeditiously, as it cannot be doubted that, though the
heavy traffic of Tooley Street will be served by the present
arrangement, much of the utility of the Tower Bridge will
remain unrealised, until a direct north and south thoroughfare
for traffic is opened up. The Corporation have performed their
part of the enterprise, and it is to be regretted that the London
County Council are wholly in arrear with their share of the
undertaking.
62 .A LECTURE ON
Conclusion.
The accommodation of the interests of the road and river
traffic at the site of the Tower Bridge has presented many
difficulties, hut I venture to hope that the bridge which is now
being erected to a great extent solves the problem. Being a low
level bridge the total rise of the road traffic is not great, and the
gradients of the approaches are short and easy. The river traffic
has ample width and height for passing through the bridge,
and the machinery for opening the bridge for the passage of
ships will be rapid in its action. Lastly, there are arrangements
for the continual accommodation of foot traffic.
The seagoing ships which pass the site of the Tower Bridge,
and for which the central span would have to be opened, number
on the average, about 17 daily. They pass by chiefly at or near
the time of high water, and it may well be arranged that several
may pass one behind the other. The number of seagoing ships
proceeding above the site of the bridge does not show any
tendency to growth, but, on the contrary, the volume of such
traffic will rather, I think, gravitate to the docks down stream
as time goes on.
I am afraid that some disappointment will occasionally be felt
when vehicular traffic is stopped by the opening of the bridge,
but it may be hoped that no serious delays will occur either to
seagoing ships or to vehicular traffic, as the periods during
which the opening span will be raised, though sufficient for the
accommodation of the river traffic, will not be of frequent
occurrence or of long duration. The Tower Bridge will,
THE TOWER BRIDGE. 63
it is thought, fairly meet all the difficulties of the case, but if
the road traffic becomes of greater importance, and the
sea-going river traffic grows less, I suppose the fate of the bridge
will be to become a fixed bridge. How soon this may happen no
one can tell. It is able to fulfil its duties either as an opening
or as a fixed bridge.
The cost of the bridge, with its approaches and including the
cost of the property purchased, will be about a million sterling,
and the whole of the expense will be defrayed out of the funds
carefully husbanded and administered by the Bridge House
Estates Committee. Londoners will thus be presented, without
the charge of one penny on the rates, with a free bridge. The
expense of working the bridge, which will be very considerable
from the quantity of machinery comprised within it, will also
be paid by the Corporation.
As this is the first public authorised notice of the Tower
Bridge, I am very glad to take the opportunity of saying how
much the works generally, and myself in particular, are indebted
to many gentlemen who have assisted in the undertaking. First
and most important of all, my acknowledgments are due to my
partner, Mr. H. M. Brunei, who has supervised the whole of the
complicated calculations and details of the structure, and has
taken a very active share in the carrying out of the work from
first to last. Afterwards follow the resident engineer, Mr.
Cruttwell, who has been in control of the works from their
commencement ; Mr. Fyson, who has had the duty of the prepa-
ration of most of the detailed working drawings and calculations
of engineering matters, and Mr. Stevenson, who has acted as my
64 THE TOWER BRIDGE.
architectural assistant. In connection with this subject, I
cannot but express my great regret that the work was so soon
after its commencement deprived of the architectural knowledge
and experience of Sir Horace Jones, and that he has not lived
to see the mode in which his conception of a large bascule
bridge across the Thames has been realised. In another branch of
duty I have to express my thanks to the various contractors
who have been employed to Mr. Jackson and Mr. Webster,
who made the foundations and approaches ; to Sir William
Arrol to whom the erection of the steelwork of the super-
structure is entrusted ; to Messrs. Perry, who are carrying out
the masonry ; lastly, and in a very important degree, to the
firm of Sir W. G. Armstrong, Mitchell and Co., to whom is
entrusted the hydraulic machinery, which, I believe, is without
rival in size and power.
The time of construction, some 7 years to the present time,
has seemed long, but it may be some comfort to those who are
impatient, to remember that old London Bridge was 33 years
in building, old Westminster Bridge llf years, and new London
Bridge 7J years, and I think my hearers will have seen that
the Tower Bridge is no ordinary bridge, and in no ordinary
position. The structure and its machinery are full of the most
elaborate and complicated work of all kinds.
In drawing this description of the works to a conclusion, I
may be allowed to express a hope that the Tower Bridge, when
finished, will be considered to be not unworthy of the
Corporation of the greatest city of ancient or modern times.
BOOT, SON AND CARPENTER, PRINTERS, 24, OLD BAILEY, E.C,
TG
83
THE LIBRARY
UNIVERSITY OF CALIFORNIA
Santa Barbara
THIS BOOK IS DUE ON THE LAST DATE
STAMPED BELOW.
fl
RECDYRL
^ SWTHERN REGI^L LIBR^Y FAQLITy
A 000 622 631 o