THE GRAND JUNCTION CANAL
A HIGHWAY
LAID WITH
WATER.
PART IV. ― THE COMING OF THE RAILWAYS
THE RAILWAY ONSLAUGHT
“With the advance of railways the waterways lost the monopoly of
inland goods traffic which they had previously enjoyed, and their
position began to deteriorate. As railway competition
developed, many were reduced to a precarious position. Efforts
by Parliament to improve the competitive position of the canal
companies by enabling them to vary their tolls, to become carriers
of goods and to make working arrangements proved unsuccessful, and
ultimately, in many cases at the instance of the canal companies
themselves, about one third of the total mileage then existing
passed to the control of the railways. This occasioned much
disquiet, and during the second half of the nineteenth century
several public enquiries took place, but without any tangible
result.” [1]
Canals and Waterways,
The British Transport Commission (1955)
CANALS – THEIR WEAKNESSES
By the 1840s, the era of steam-hauled public railways had begun in
earnest and with it the inexorable decline of the canal network,
with significant parts falling into neglect or being abandoned.
It is worth considering the factors that led to this outcome.
In several ways the rot was already established within our canal
network by the beginning of the railway era. It stemmed from
the parochial outlook of the canal owners and from a lack of
standardisation in canal construction, both being defects that the
new railway companies were not to exhibit in any material way, as
events quickly proved:
“Canals in their day reached a far greater pitch of prosperity
than the railways have ever attained to, but they suffered fatally,
and do so now, from the want of any serious movement towards their
becoming a united system of communication. Each navigation was
constructed purely as a local concern, and the gauge of locks and
depth of water was generally decided by local circumstances or the
fancy of the constructors without any regard to uniformity.
The same ideas of exclusiveness seem to have been perpetuated in the
system of canal management; there is no Canal Clearing House, and
with few exceptions every boat owner has to deal separately with the
management of every navigation over which he trades.”
Bradshaw’s Canals and Navigable Rivers,
Henry Rodolph de Salis (1904)
Some of these weaknesses might have been addressed with foresight
and determination, but others are inherent in the canal concept.
Other than the comparatively small amount required to raise steam,
the new railways did not depend on a sufficiency of water to support
their operations; in this respect they were easier to construct and
maintain than a canal, and more versatile in the locations that they
could serve. A crew of three could operate a freight train
hauling a load of 500 tons or more; the same workforce manning a
pair of narrow boats could move around 50 tons. Quicker
transit time was another advantage in the railways’ favour.
Even in their earliest days, steam trains ran at a far higher speed
than their waterborne counterparts.
Although the railways generally had transit time on their side, this
was not
always so. Where a journey was short and the supplier
and consumer both had canal wharfs ― often the case on the
Birmingham canal network ― the speed advantage could lie with the
canal, and canals in this category quickly became targets for their
voracious new competitors. Efficient management could also
make a considerable difference to transit times, and here it is
interesting to note a passage from the Rusholme report, the writer
referring to government control of canals during WWI . . . .
“The Committee [2] sought to organise
traffic on these canals with a view to relieving the congestion on
the railways, and the result of their efforts was a considerable
increase in the amount of traffic carried and an improvement in the
speed of conveyance, until it compared on the whole not unfavourably
with that of heavy traffic on the railways.”
Canals and Waterways, The
British Transport Commission (1955)
. . . . and this despite a backlog of repair and maintenance work
that had built up during the war years, which suggests that transit
times might have been improved with more efficient management and
better co-ordination between canal companies, requirements that were
generally lacking. But little could be done to manage the
weather. Steam-hauled railways rarely suffered from delays due
to drought and were less affected by freezing conditions, but each
could bring canal traffic to a standstill, sometimes for long
periods despite the construction of reservoirs and bore holes, and
the operation of ice-breaking.
But lack of speed alone did not bring about the decline of canal
carrying. The failure of canal companies to develop
inter-working ― let alone to amalgamate ― was a significant
drawback, and one in which government might have intervened by
forcing mergers similar to those that they eventually imposed on the
railway companies in 1921 (the ‘grouping’). Various inquiries
and commissions on canals and transport over the years came to much
the same conclusion as that arrived at as recently as 1928:
“. . . . we are of the opinion that certain canals still possess
considerable value as a means of transport, and that, properly
rationalized and developed, they can be made to render much useful
service to the community in the future. We are satisfied that
a process of amalgamation is a necessary preliminary to any
developmental programme.”
The Royal Commission on Transport,
Sir Arthur Griffith Boscawen (1928)
Together with investment, company amalgamations might, over time,
have ironed out another significant difficulty, the lack of
standardization.
Canals were generally built to serve the needs of local industries,
which fostered a parochial attitude to their construction and
management. Their dimensions were influenced by the type of
trade they were designed to serve, by local conventions, [3]
by the amount of capital available for their construction, by water
supply and other topographical limitations:
“The locks and waterways of canals are altogether wanting in
uniformity. One of the great difficulties of canal traffic in
Great Britain, in England and Wales especially, has been that
scarcely two canals have a continuous gauge. We find even upon
one canal two or three different gauges of locks. In some
instances the canals forming a continuous line are approximately of
the same size, but the gauge of the locks is entirely different.
Either their locks are considerably shorter, or are considerably
narrower, or the water-way is considerably shallower. There is
nothing like a common uniform gauge of canal throughout the whole
distance.”
E. J. Lloyd, Engineer to the Warwick canals, giving
evidence to the
Select Committee on Canals (1883)
Even the dimensions of the ubiquitous ‘narrow boat’ varied:
“’Narrow boats’ or ‘monkey’ boats are by far the most numerous
class of vessel engaged in inland navigation. They are from
70ft to 72ft long by 6ft 9ins to 7ft 2ins beam, and draw from 8ins
to 11ins of water when empty, loading afterwards to about 1ins to 1
ton.” [4]
Bradshaw’s Canals & Navigable Rivers,
H. R. de Salis (1904)
Because government took no interest, statutory standards designed to
facilitate inter-working between different canal systems were not
imposed, and waterways continued to vary in the maximum size and
loading of craft they could accommodate. The Grand Junction
Canal main line is an example; it can accommodate craft 72ft long by
14ft 3ins beam and 3ft 8ins draught, [5] but the
locks on the adjoining systems at Gayton and Norton restrict craft
to 7ft beam, although they in turn lead to broad waterways. In
the 1930s, the Grand Union Canal Company widened the line from
Braunston to Birmingham via Napton Junction, which includes a short
section of the adjoining Oxford Canal, but with that exception the
Oxford Canal is narrow:
“As almost all the through routes between important centres at
the present time contain links to narrow canal, the effect of these
diversities of gauge is to confine any long distance through traffic
to narrow boats. Nothing but a narrow boat can navigate
between London and Northampton, London and Leicester, London and
Nottingham or Manchester, and nothing but a narrow boat can get into
or out of Birmingham from anywhere. If we attempt to take our
narrow boat from London to Leeds we shall fail altogether, as we
shall be stopped at either Wigan, Sowerby Bridge, or Cooper Bridge,
by the locks on the Leeds and Liverpool Canal, or Calder and Hebble
Navigation, which, although twice the width required by the narrow
boat, are 10 feet too short.”
H. R. de Salis (Vice Chairman
Fellow, Morton & Clayton) giving evidence to
the Royal Commission (1906).
Narrow locks also slowed traffic, due to pairs of narrow boats
having to lock through individually rather than abreast. Even
within the otherwise broad Grand Junction Canal, the Blisworth and
Braunston tunnels were too narrow to allow wide boats to pass, while
river barges (14ft beam) could only be accepted without restrictions
on its more heavily worked southern section below Berkhamsted. [6]
Coupled with the physical limitations on through working, craft
traversing different companies’ systems faced a chaotic pricing
structure that subjected them to varying charges at toll offices
along their route. And where a craft briefly entered another
canal company’s system in order to reach its destination,
disproportionately high ‘compensation tolls’ [7]
were often imposed. An example was the Oxford Canal, which was
authorised to impose a compensation toll of 2s 9d a ton on coal and
4s 4d on all other items at the junction of the Warwick & Napton
Canal, and in this way received a quarter of a million pounds over
twenty years. [8] All this meant that
‘through charges’ could not be determined accurately in advance;
craft were gauged and the captains paid in cash at each toll office
they passed on their journey, or if working for a carrying company
with an account, then the company was billed. Chaos would have
ensued had the same charging system operated on the railways.
By comparison, from the Liverpool & Manchester Railway onwards, the
4ft 8½ins gauge applied to most of our mainline railway network, [9]
thereby permitting traffic to move from one company’s system to
another. Railway company mergers also took place resulting in
longer routes under single managements. [10]
Where passengers and freight passed through the domains of different
companies to reach their destination, arrangements were negotiated
for through working and end-to-end charging ― set up in 1842, the
Railways Clearing House allocated the revenue received in proportion
to the resources that each had provided. [11]
Overall, railway companies had a much wider vision of their business
and how it might be developed than did their blinkered waterway
competitors of their own assets. Boyle summed the position up
succinctly:
“The canal, on the contrary, is not calculated to deceive
anybody. Beside the railway it appears the embodiment of
quiet, plodding, undisguised sluggishness; and if this is not
actually the case, it must be admitted the erroneous impression is
not altogether without foundation. There is certainly nothing
fast about a canal. It seldom changes or improves; and when it
does so, it is only by imperceptible degrees. It is apparently
the same yesterday, today, and for ever; ― perpetuating its
primitive arrangements to all time and under all circumstances.
Enlightened enterprise appears to have shunned it as an ungenial
sphere for all its operations, and the world in its march of
improvements seems to have left it behind hopelessly on the road.”
Hope for the Canals,
Thomas Boyle (1848)
In Britain, the need for large-scale modernisation of the canal
system was an issue in which the state took no part, for transport
was seen as a matter for individual companies. Some of our
early canals ― especially those built before the French Wars drove
up construction costs ― paid large dividends rather than investing
their revenue in straightening, widening and deepening their
waterways, and in providing bank protection when steam-power became
feasible ― some systems would not at first accept powered craft.
In France, Belgium and Germany, where the introduction of railways
also had a damaging impact on inland waterways, the state
intervened. The outcome was that waterways were modernised and
steam haulage introduced, while protective legislation regulated
canal and railway tariffs. Not until nationalisation in 1947
did the British government take a material interest in inland
waterways, but by then it was too late. The development of
road transport and of the motorway network made canal modernisation
along northern European lines wholly uneconomic.
――――♦――――
THE ONSLAUGHT BEGINS
The success of the Bridgewater Canal proved the viability of canal
transport, and within a few years of its opening an embryonic
national canal network came into being with the construction of
canals such as the Trent & Mersey and the Oxford. The early
1790s saw the period known as ‘canal mania’, when huge sums were
invested in new canal schemes, often with little serious
consideration of their viability. Partly as a result, the
canal network grew to over 4,000 miles, becoming both cause and
effect of the rapid industrialisation of the Midlands and the North
of England. The period between the 1770s and the 1830s is
often referred to as the ‘Golden Age’ of British canals, an age in
which canal companies faced no serious competition and the
complacency of the monopoly operator became established.
If any single date marks the beginning of the decline of our canal
network, the 15th September 1830 must be a serious contender, for on
that day the world’s first mechanically-operated inter-city public
transport link, the Liverpool & Manchester Railway, commenced
business:
“Although the advantages of a railway for the conveyance of
passengers and goods are very superior to any other mode, even when
horses are employed as the moving power; yet, these advantages are
vastly increased by the substitution of Locomotive Steam Engines . .
. . The fare between these towns [Liverpool and Manchester]
for inside passengers by the railway is five shillings; before the
opening of the railway it was twelve and fourteen shillings by the
ordinary coaches. The distance between the towns is run on the
railway in two hours: by the turnpike road it occupies four and a
half or five hours. . . . Goods shipped at Liverpool by water
conveyance to Manchester, are thirty-six hours on the passage; but
by the railway they are delivered in about five hours . . . . And in
winter, the canal part is frequently frozen for many weeks, so as to
obstruct all passage; and goods have to be transferred from the
canal to land carriage, suffering loss of time, increase in expense,
and risk of damage in the transfer.”
Observations on Railways: particularly on the
proposed London & Birmingham Railway, Anon
(1831).
Although the march of technology would have delivered the
steam-hauled public railway, the Liverpool & Manchester Railway grew
out of a commercial need, to combat the high charges and poor
service provided by the Bridgewater Canal and the Mersey & Irwell
Navigation in transporting goods between Liverpool and Manchester
(it was said that goods could take longer passing between the two
cities than in crossing the Atlantic). The irony was that the
Bridgewater Canal had been built seventy years earlier to address
the very same problem, the expensive and poor service then provided
by the Mersey & Irwell Navigation.
The waterways did what they could to oppose the Liverpool &
Manchester Railway, even to the extent of threatening physical
violence to the railway surveyors. The first Bill laid before
Parliament failed, but at the second attempt the railway company
succeeded [12] in obtaining the necessary private
Act, and construction commenced. Faced with a fait accompli,
the waterways attempted to retain their trade by reducing tolls and
improving services; the Irwell Navigation straightened sections of
its line and improved its water supply; the Bridegwater Canal
Company built new docks and warehouses, improved access to the River
Mersey at Runcorn and linked their docks to the Weaver Navigation.
In 1844, the Bridgewater bought the Mersey & Irwell Navigation and
eventually reached a trading agreement with its new competitor, the
London & North Western Railway.
In these respects the Bridgewater Canal is unusual. Faced with
the railway onslaught, the company responded with vigour and
continued to trade profitably until acquired by the Manchester Ship
Canal Company in 1885. But compared with many other canal
companies the Bridgewater had the advantage of a strong business
foundation. At one end of its line lay Manchester, a major
manufacturing city, and at the other end lay the Runcorn docks and a
river link to a major seaport, Liverpool. In addition, there
was a healthy coal trade that lasted into the 1970s. When
faced with railway competition, canal companies in less fortunate
circumstances followed the Bridgewater example by reducing their
charges, generally by between 30% and 50% of their former level, [13]
but although this slowed their commercial decline it did not halt
it. Tonnage often increased in response to the cheaper toll
rates, but total revenue fell, resulting in reductions in manpower
levels and waterway maintenance, and in the sale of company assets.
The new railways quickly acquired business where transit time was
important, such as with perishables, packages and passengers.
Passenger traffic was never a staple source of revenue for canals,
although in the north-west of England and in Scotland it did make a
contribution on some systems. [14] Most of
the railways’ passenger traffic was either entirely new, or was
taken from the stagecoach operators on routes where they competed
(such as Liverpool to Manchester), the stagecoach operators going
out of business (or becoming feeders) almost as soon as a competing
railway commenced business. Having established a profitable
line in passenger traffic, the railways then began to capture
freight where another significant, although less obvious advantage
of speed, lies in swifter logistics. This allows manufacturers
and retailers to reduce their stocks of raw materials and finished
goods, thereby saving on financing charges and/or releasing some of
their capital for other purposes.
――――♦――――
SOME CASE STUDIES
Faced with a losing battle, some canals ― mostly the less successful
― soon sold out to the competing railways, sometimes even offering
themselves for sale. By purchase, amalgamation or lease, [15]
the railway companies gradually acquired a substantial interest in
the canal network to the extent that by 1865, about one third of
canal mileage in England and Scotland was under some form of railway
control. [16] On occasions a canal fitted
the railway’s business, acting as a feeder or giving freight access
to areas of the country served by a competing railway company, but
generally a railway company’s acquisition permitted it to remove
waterway competition by a combination of increased tolls and
declining maintenance:
“The cases in which railway companies have a more or less strong
or less strong positive interest in pushing the trade on the canals
belonging to them are exceptions to the rule.”
Report of the Royal Commission on Canals,
pub. 1909.
One company stifled by railway ownership was the Kennet & Avon
Canal, which links Reading on the Thames with Bristol. This
waterway comprises sections of the rivers Kennet and Avon, which
between Newbury and Bath are linked by a 57-mile ‘broad’ canal.
Engineered by John Rennie Snr. and opened throughout in 1810, the
Kennet & Avon was successful in its early years ― mainly carrying
coal and stone ― although its dividends never exceeded a modest 3¾%
(1840) due mainly to its heavy construction costs. The opening
of the Great Western Railway in 1841, followed by the Berks & Hants
Railway in 1847, diverted most of the Kennet and Avon’s trade to
rail despite the canal company having lowered its tolls. When
taken over by the Great Western Railway in 1852 for a fraction of
its construction cost, its tolls were increased and its trade,
unsurprisingly, declined still further. By 1868, the canal’s
annual tonnage had fallen from 360,610 in 1848 to 210,567, and the
canal ceased to make any profit after 1876. [17]
Following WWI, when more trade was lost to road transport, the Great
Western Railway Company attempted to close the canal but were
prevented from doing so:
“The local authorities in Wiltshire, Berkshire, and Somerset,
through whose area the Kennet and Avon Canal runs, are still a
little anxious as to the fate that awaits this old waterway.
The Great Western railway, which originally proposed to apply for
powers to abandon the canal altogether, has now, in consequence of
widespread opposition, promised to modify its application so that,
though the canal will be closed to traffic, the company will remain
liable to carry out its present obligations ‘except those which have
been necessary for traffic purposes.’
In that qualifying phrase a great many fears which the railway
company has sought to answer have been revived. The one
specific promise given is that the company will maintain the pumping
station at Crofton and so ensure a flow of water into the canal.
The local bodies are asking if the locks and their gates will be
kept in proper repair; whether the channel will be allowed to become
choked with weeds and silted up; and whether ultimately there may be
stretches of virtually stagnant water in and near the towns once
served by this waterway that joins the eastern and western seas. . .
.”
The Times, 19th January,
1927
By 1950 most of the Kennet & Avon was disused and impassable, an
outcome shared by much of the railway controlled canal system.
A further attempt at closure in the 1950s having failed, sufficient
of the Canal survived into the preservation era, since when, with
unstinting volunteer effort and Lottery grant support, it has been
transformed into what the 1968 Transport Act classifies as a
‘cruising waterway’. [18]
Another, and in some ways more damaging strategy adopted by railway
companies to stifle canal trade, was to gain control of a short
section of a much longer through route. The toll for that
section could then be increased to the highest permissible level,
thereby imposing a disproportionately high increase to the overall
through charge. An example was the through toll on iron goods
from South Staffordshire to London, of 4s 6d per ton. The sum
charged over the 10-mile Birmingham section, controlled by the
London & North Western Railway Company, amounted to 1s 6d, while the
independent canals received 3s for the remaining 146 miles.
Had the Birmingham section being charged at the same rate as the
remainder, the through toll would have been reduced by a quarter. [19]
However, it must be said that the railway companies were merely
adopting a strategy already operated by some canal companies in the
same circumstances:
“One great impediment has been found to exist, in the present
disjointed state of the Canal interests and the varying systems
under which they carry on their operations. Some of the
existing Companies, possessing lines of Canal which form central
links in a great chain, take advantage of their peculiar position,
and establish a rate of charges so high as to secure to themselves a
large return for their capital, even upon a small amount of traffic.
This practice, while it obliges the other Companies, not so
advantageously situated, to reduce their rates to such an extent
that they are unable to conduct their business with a profit, at the
same time prevents such a reduction in the general charges of the
line as would enable the several Companies, as a body, to maintain a
fair competition with the Railways.”
Second Report from the Select Committee on
railways and Canals Amalgamation (1846)
Following WWI., the canals were faced with increasing road
competition and difficulties in recruiting boat crews (many who had
joined the forces or found better-paid employment in factories
engaged on munitions work, never returned) while the railways were
by then being subsidized by the State:
“The main difficulty with regard to the extension of the use of
canals lies in the freight charges. The railways, subsidized
by the State, have been able to keep their charges down, while on
canals increased expenses have forced freight charges up. In
1913 the canal charges were lower than those on the railways; to-day
they are considerably higher.”
The Times, 10th October,
1919
But aggressive railway competition was not the sole reason that
canal companies went to the wall. Other causes of failure lay
in their original business case being unsound (e.g. under-estimated
construction costs, over-estimated traffic receipts) [20]
or in the local industry(s) on which they relied disappearing under
changed circumstances, such as coal mines becoming worked out,
canal-side iron works closing in favour of large steel mills, or
grain mills moving to the ports to be more economically placed to
mill cheap imported grain. The Basingstoke Canal (also known
as the London and Hampshire Canal) in part fell victim to London
increasingly receiving its agricultural produce from the midlands
and north of England via the Grand Junction Canal, and later the
railways, and importing cheaper foodstuffs through the London docks.
The Basingstoke Canal linked that town with the Thames at Weybridge,
via the River Wey Navigation [21] ― several plans
to continue it to the South Coast never materialised.
Engineered by William Jessop and built as a broad canal, it was
opened throughout in 1794. Unlike the canals in the midlands
and the north of England, which served mainly industrial needs, the
Basingstoke Canal was intended to stimulate agricultural development
in central Hampshire by transporting timber, flour, malt and
agricultural produce to London, with return cargoes of coal, manure
and groceries. Construction costs were much higher than
anticipated added to which the traffic projections proved
over-optimistic By 1796 the Company was on the brink of
bankruptcy. To stave off failure bonds were sold, but
servicing the interest absorbed the profits resulting in the
Company’s shareholders failing to receive a dividend at any time in
the Canal’s history. By the 1820s, improved roads in the area
led road transport presenting serious competition, with a faster
door-to-door service at comparable charges.
Ironically, trade revived briefly during the 1830s with the
construction of the South Western Railway, the canal being used to
transport railway construction materials; but once the railway had
been opened, canal trade slumped. Price-cutting by both
companies continued throughout the 1840s, the railway having the
advantage of being able to cross-subsidise reduced freight rates
from its passenger receipts (by then a common practice):
“Another consideration which must not be overlooked is, that
although it has been stated that with proper management Canals might
maintain a successful competition with railways in the carriage of
heavy goods, still such competition has hitherto been carried on
under great disadvantage, owing to the large profits made by Railway
Companies on passenger traffic, which enables them to submit in some
instances even to a loss on the carriage of merchandize with a view
to withdraw traffic from Canals.”
Second Report from the Select Committee on
railways and Canals Amalgamation (1846)
Another short boost to trade occurred in 1854, when the Canal was
used to transport materials for the construction of Aldershot Camp,
after which business again declined. In June 1866, the
proprietors resolved to go into liquidation. By this time,
other canals in the south ― notably the Wey and Arun, the Thames and
Severn, and the Wilts and Berks ― had already expired, but the
Basingstoke was not abandoned. There followed a succession of
owners, one attempting to sell in 1904:
“The canal was purchased in 1896 by its present proprietors, who
have expended a considerable amount of money on its restoration and
improvement, but apparently without success, as at the present time
(1904) the navigation is again for sale. Of late the towing
path has become much overgrown in places, thereby causing horse
towing to be very difficult.”
Bradshaw’s Canals & Navigable Rivers,
H. R. de Salis (1904)
Despite the auctioneer’s encouraging sales pitch [Appendix]
there were no takers. WWI brought a further brief revival of
trade, with Government stores and munitions being carried to
Aldershot Camp with return cargoes of manure and timber. In
1923, the Canal was bought by Alexander Harmsworth, who ran a
successful business until his death in 1947, but only on the section
below Woking, the upper reaches having by then become a picturesque
backwater:
“Long stretches of this waterway are among the quietest and
least-frequented places in the country. Rich vegetation
abounds on its banks. In places the trees meet above it,
forming in summer long tunnels in which the light is tinged a
greenish shade, giving to whoever may pass through them, in a boat
hardly propelled through the weed-thick water, a ghostly and baleful
air. Here and there a broken bridge crosses the water, here
and there an empty boat-house stands mournfully on the bank”.
The Times, 28th August,
1929
The Canal was not taken into the nationalisation programme in 1947,
and by the 1960s it was derelict. But as with the Kennet &
Avon, sufficient of it survived for restoration. Today, after
years of effort by members of the Surrey & Hampshire Canal Society
in partnership with local authorities, 32 miles of the Basingstoke
Canal are again open to navigation, albeit restricted at times due
to water shortage.
Some waterways did manage to hold their own against the railway
onslaught, but these tended to be the river navigations, which
required comparatively few locks, had a dependable water supply and
could more readily be widened and deepened. An example is the
Aire & Calder Navigation in West Yorkshire, which runs for 34 miles
from Leeds to Goole, with a 7½ mile branch from Wakefield (where it
connects with the Calder & Hebble Navigation) to Castleford.
During its life the Navigation has been widened and deepened, its
locks have been lengthened, and canal cuts have been constructed to
straighten sections of the route.
During the 1840s, the Aire & Calder’s trade was reduced by about a
third following the arrival of the railways, but the Navigation was
fortunate in having two innovative engineers, Thomas Bartholomew
and, following his death in 1852, his equally talented son, William.
Thomas was an early advocate of steam propulsion and by the time of
his death two-thirds of the traffic on the Navigation was hauled by
steam tug. Besides carrying out many improvements to the
waterway, William Bartholomew developed a sectional barge system.
Known as ‘Tom Puddings’, each comprised six separate compartments
with a bow and stern section, which, under the control of four men,
could move 800 tons of coal. On arrival at Goole the contents
of each compartment (‘tub’) could then be emptied mechanically into
a collier by means of dockside hydraulic hoists. By 1913,
there were 18 tugs, 1,010 compartments and 1,560,006 tons of coal
was being carried annually. [22] The
compartments were still carrying around half a million tons of coal
until the late 1960s, long after most British canals had ceased to
be used for commercial traffic, but the gradual demise of the coal
industry led to compartment traffic ceasing in 1986.

A
‘Tom
Puddin’
― Bartholomew’s sectional boat.
In addition to investing in the Navigation’s development, its
success can also be attributed to a firm business base in the shape
of the large Yorkshire coalfields that it serviced and on the
company’s development of Goole Docks, from which coal and other
goods were exported in quantity and into which pit props for the
mining industry were landed:
“. . . . the collector of customs at Goole, said that there were
large warehouses there belonging to the Aire and Calder Navigation
Company. He produced a mass of reports made to the Board of
Customs, from which it appeared that cotton-twist, fustians, woollen
goods, coals, and iron were exported thence.”
The Times, 15th July,
1845
The decline of the coal industry and of heavy manufacturing
generally saw commercial traffic on the Navigation dwindle, but not
disappear. As recently as 2007 the Navigation is reported to have
carried 300,000 tons, mostly of petroleum and building aggregates.
During the 1960s, the navigation underwent further modernisation, in
which the locks from Goole to Leeds were upgraded and enlarged to
accommodate vessels conforming to the 700-tonne Euro-barge standard.
The Navigation also provides important leisure cruising links to,
among other waterways, the Leeds & Liverpool Canal and, via the
Calder & Hebble Navigation, to the Huddersfield Broad and Narrow
canals and the Rochdale Canal (the latter two being restored to
use).
But overall, the Aire & Calder’s success was untypical of British
inland waterways, which went from a position in the 1840s in which
they carried more freight than the railways, to one at the end of
the nineteenth century when their share of the market had slumped to
about a tenth; or put another way, overall growth in canal carrying
during the period remained almost static whilst the corresponding
rail freight business burgeoned.
――――♦――――
THE LONDON & BIRMINGHAM RAILWAY
“In 1825 the tract of country lying between London and Birmingham
was surveyed by Messrs John and Edward Grantham, for the purposes of
ascertaining the practicability of forming a railroad between the
metropolis and the town of Birmingham. Sir J. Rennie,
likewise, surveyed a tract of country for the same purpose carrying
the route of line along the valley of the Thames to Oxford and
thence to Birmingham. This, however, was abandoned on account
of the tunnelling necessary at Oxford, and the periodical floods to
which the Thames is subjected. Various portions of country
were also surveyed and diversions made in all directions until the
most eligible, and that which presented least, difficulties, was
evident. In 1828, the first public intimation appeared of the
formation of ‘A Company for making a Railway from London to
Birmingham,’ with the name of Mr Stephenson, of the Liverpool and
Manchester Railway, subjoined as its engineer.”
The London and Birmingham Railway Guide:
Joseph W. Wyld (1838)
The seeds of the London & Birmingham Railway were sewn as early as
1825. Sir John Rennie proposed a London to Birmingham railway
passing through Qianton and Banbury, but nothing came of the scheme
or of a later proposal by Francis Giles for a line passing through
the Watford Gap, Rugby and Coventry. However, both schemes
caused considerable alarm among canal companies, turnpike trusts,
stagecoach operators and the proprietors of coaching inns and
stables. Each group could be depended on to oppose with vigour
any railway bill that eventually came ― they did not have long to
wait.
In 1829, two schemes were published in which lines were proposed
along the routes put forward by Rennie and by Giles. Rather
than compete with each other ― in addition to numerous opposing
commercial and landed interests ― the schemes’ proprietors sensibly
combined forces, employing George and Robert Stephenson to review
their plans and recommend which to adopt. The Stephensons
having chosen the eastern route, a detailed survey was then
undertaken by Robert Stephenson and Thomas Gooch, [23]
which formed the basis of a Bill submitted to Parliament in 1832.
This Bill passed the Commons but was rejected by the Lords, the
dissenting landowners being in the majority of 59 to 53. The
original route was then altered to avoid the estates of most of the
opposition, while the remaining dissenters along the line were made
financial offers they found difficult to refuse. The outcome
was that on the second attempt the London & Birmingham Railway Bill
passed both houses and received the Royal Assent on 6th May, 1833.
Construction then proceeded swiftly. Boxmoor was reached in
July, 1837; Tring in January 1838; Denbigh Hall in April; and the
line was opened throughout in September of that year. The
112½-mile railway had taken five years to complete; the 131½ miles
of the Grand Junction Canal (including its pre-1805 branches) had
taken twelve.
The route chosen by Robert Stephenson keeps close company with the
Canal for many miles. From Cassiobury Park, Railway and Canal
ascend the valleys of the Gade and the Bulbourne before crossing the
Tring Gap in deep cuttings. There then follows a long descent
to Milton Keynes where Canal and Railway briefly part company, the
Canal weaving a sinuous path around the modern town while the
Railway takes a more direct route to Wolverton. Both then
cross the valley of the Great Ouse, the Canal on a substantial earth
embankment, the Railway on a brick viaduct. At Roade, rather
than following Barnes’s example of tunnelling the ridge, Stephenson
took the Railway through in a deep cutting. At Buckby, Railway
and Canal finally diverge, the Canal heading west towards its nearby
rendezvous with the Oxford Canal, the Railway departing in a
north-westerly direction towards the Watford Gap and the notorious
Kilsby Tunnel. Here Stephenson met his Nemesis in the form of
the quicksand deposits and severe flooding that had been encountered
by the canal tunnelers at Blisworth ― in neither case did the trial
borings reveal the problem.
Until railway competition changed the picture, the Grand Junction
Canal Company had taken full advantage of its monopoly status as a
transporter of goods:
“As a special instance of a canal taking advantage of its
position to raise its rates, we may mention the Grand Junction
Canal, which extends from Paddington to Braunston where it joins the
Oxford Canal. The Grand Junction Canal was an important link
between London and the great mining and manufacturing sections of
Warwickshire, Cheshire, Staffordshire, etc. It was a monopoly
without competitor; its exactions, excessive rates, discriminatory
rates, and its supercilious conduct caused loud and general
complaints even as late as 1836.”
Remarks on the tonnage rates and drawbacks of the
Grand Junction Canal, Mercator (1836)
No sooner had the new railway company commenced operations, than it
too began to exploit the monopolistic position that it had won from
the stagecoach operators:
“THE LONDON AND BIRMINGHAM
RAILWAY: No system appears to us to
require more vigilant scrutiny, nor the conduct of any body of
persons to be more jealously watched, than railway companies and
railway directors. The moment that all means of opposition
ceases an attempt is made to impose on the public the most
exorbitant and unjustifiable charges for goods and passengers.
We last week expressed our opinion on the subject of the increase of
price for the carriage of parcels from 1d to 1½d per pound, and we
believe we may now congratulate the public on its abandonment, and
the immediate reduction to the old prices. But there are other
ways that these great monopolies are made to press hardly on public
convenience; among others, we beg this week to direct the attention
of our readers to the sudden increase in the price of conveyance
that has taken place to all the intermediate stages since the
opening of the railway throughout to London, and the consequent
removal of the stage-coaches from that line of road.”
The Times, 29th
September, 1838
The London & Birmingham Railway’s tussle with the Grand Junction
Canal Company over their respective shares of the freight traffic
between London and the Midlands was slow to start, for the Canal was
not affected by the Railway’s prime interest at that time ―
passengers. But as the railway moved into freight carrying, a
price-cutting war began that was to last off and on until
nationalisation eventually brought railways and canals under a
single owner, the State.
――――♦――――
APPENDIX
(The Times, 28th October, 1904)
THE BASINGSTOKE CANAL.—By
order of the High Court of Chancery, the Woking, Aldershot, and
Basingstoke Canal was offered for sale by auction at the Mart
yesterday, by Mr. B. I’Anson Breach, of the firm of Messrs.
Farebrothers, Ellis, Egerton, Breach, and Co.
In giving particulars of the property the auctioneer remarked that
the occasion presented a unique opportunity for speculators.
The canal, which was constructed under two Acts of Parliament passed
in the reign on George III., was built at a cost of between £150,000
and £200,000, and it was opened in 1794. It started in the
town of Basingstoke, and, passing through some of the most
picturesque and beautiful residential neighbourhoods of Hampshire
and Surrey, it ended at the junction of the river Wey, by means of
which a direct line of navigation was opened to London, a distance
of nearly 70 miles. The canal was about 37 miles long, and
there were 29 locks. The traffic at the outset was very
remunerative, but when the South-Western railway was constructed it
seemed gradually to “swamp” the Canal. There were 13 wharves
conveniently placed at important military and trading centres; and
along the canal were a number of lockkeepers’ dwellings, warehouses,
&c., principally let. The property embraced a total land and
water area of about 366 acres. There were low rentals now
coming in amounting to £500 a year, and the revenue — the general
trading of the company — was £5,000. He saw no reason why the
Canal should not be carried on to Southampton; and a splendid
opportunity was afforded to the War Office of connecting their
waterways with the camp at Aldershot. Replying to questions,
the auctioneer stated that the lack of water in the canal about a
month ago near Woking arose from the fact that the receiver had been
putting in a new lock gate.
Under the existing Act it was clearly defined that the Canal was to
be used for the purposes of a canal; but he knew no difficulty which
would prevent a purchaser from obtaining a short Act of Parliament
to enable him to convert it into a motor track if he wished to do
so. Adjoining property owners, under section 101 of the Act,
had the right of pre-emption on terms which would have been settled
by Commissioners. He then invited bids for the property.
To his suggestion that they should begin at £50,000 there was no
response, upon which he invited offers of £40,000 and £30,000 with
the same result. He was, he remarked, in the hands of intended
buyers; but he was acting under sealed instructions, and could
therefore give no hint as to prices. Subsequently he invited
bids of £25,000 and £20,000, adding that he could not go below the
last-mentioned amount. No offer, however, was made for the
property, which was accordingly withdrawn. |