The Origins of the Narrow Boat
For many years, Chester was the main port for the north west of England. However, by the start of the eighteenth century, the estuary of the River Dee was becoming increasingly silted. A second problem for the port was that, geographically, Chester's communications with the rising industries of Lancashire were poor. Thus the opening, in 1715, of Liverpool's first wet dock, designed by Thomas Steers who was also to be the engineer for several local navigations, began a shift of the centre of trade from Chester to Liverpool.
The canalisation of the Dee below Chester in 1749 came too late to stop this change. Steers had constructed the Mersey & Irwell, and Douglas Navigations by then, and the Weaver Navigation had also opened, creating the foundation for an inland waterway system linking the new industrial areas of the north west with Liverpool. The port's trade had also increased as a result of the Spanish and French wars, when shipping using the English Channel was a target for privateers. The rise in Liverpool's importance is reflected in the increased number of ships belonging to the port, the number rising from 84 in 1709 to 226 by 1760. At the same time, the number of boatbuilders (counted by voting eligibility) increased from 47 in 1734 to 86 in 1761 indicating a tremendous expansion in the demand for boatbuilding in the north west.
For trade to the colonies, for coastal shipping and for carriage on the new inland waterways, the need for boats was making great demands on the boatbuilding industry of the north west. The Irish Sea was becoming more important as a trade route, and boats were built in Northern Ireland not just for this trade, but also for the Newry Canal and Lagan Navigation, opened in 1742 and 1763 respectively. Inland waterway and coastal trading boats built in Ulster show similarities to those built in the north west of England suggesting well-established links with the boatbuilding industry of the Merseyside area.
In Lancashire, an increased use of barges on river navigations must have been one of the results of the doubling of cotton imports between 1730 and 1751, from 1.5 to 3 million lbs annually, as cotton textile manufacturing began its rapid rise to become England's most important industry. The majority of this cotton would probably have been moved from the port of Liverpool to the Manchester area along the Mersey & Irwell Navigation. Not just cotton was carried, with coal also rising in importance as supplies of wood fuel diminished and coal burning technology improved. Salt was also being carried in increasing tonnages, both for the new Lancashire chemical industry and for use as a preservative by the Irish dairy and beef industry. Grain, limestone and iron products were also being carried routinely by coastal and river barges. It was against this background of increasing waterborne trade that the Duke of Bridgewater was to build his canal and coal mining empire.
In 1760, Worsley was just a small village with few skilled inhabitants. Because of this, and the difficulty in obtaining trained workmen in the area generally due to the rapid growth of industry, the Duke had to look elsewhere for skilled workers for both his mines and canal. Stone masons came from Cumbria, while miners moved to Worsley from Staffordshire and Northumberland. When the Duke required boatbuilders, there was a second problem; the poor supplies of timber to be found locally. The lack of forests in the south Lancashire area was itself the reason for the demand for coal and hence the canal. Wood for boatbuilding needed to be of high quality and was difficult to obtain. The Day Books of John Brockbank, a Lancaster shipbuilder at the end of the eighteenth century, show that he travelled all over the country to obtain supplies. In March 1790 he went to London, returning by way of Pershore, Ironbridge and Frodsham, with other journeys particularly to Wales and the area around the rivers Dee and Severn where timber could be procured more easily than in Lancashire.
Small wonder then that the Duke set up a boatbuilding yard at Bangor-on-Dee, with the wood for the boats coming from his own estate at Ellesmere. The yard built the sailing flats which the Duke used on the Mersey estuary, the Mersey & Irwell Navigation and on his own canal when it was joined to the Irwell. The flats were delivered by being sailed down the River Dee and round the Wirral to the Mersey. There is a possibility that the canal and river were linked at Castlefield in Manchester when the canal was first opened. Definite details have so far proved impossible to obtain, but the existence of the Bangor-on-Dee boatyard does suggest that a link was built. However, smaller boats were also needed for the construction of the canal, for use in the mines at Worsley and to transport the coal from there into Manchester. These would not have been suitable for a sea passage from the Bangor boatyard because of their small size and so a separate yard was set up by the Duke at Worsley.
Where did the boatbuilders for these yards come from? Those at Bangor-on-Dee could well have come from the declining industry at Chester, but where did those employed at Worsley originate? The Mugg family is one mentioned in early canal records as being involved with boatbuilding at Worsley, and Frank Mullineux, Worsley's local historian, recalled talking to descendants of the Mugg family who still visited 'home' in the Welshpool area. So the first Muggs to be employed at Worsley could have originally built boats on the River Severn before moving north. Another boatbuilder employed by the Duke was John Lloyd, whose name also suggests Welsh origins. The River Severn would have been a good area for the Duke to seek skilled boatbuilders, as it was to be another 80 years before it was improved for navigation. In the middle of the eighteenth century boats navigated the upper river only at times of high water. Trade was steady but did not expand to any great extent until the opening of the Staffordshire & Worcestershire Canal in 1772. Consequently the demand for boats in the north west was far greater than on the Severn, and boatbuilders may well have considered the opportunities for regular work much better in Lancashire. Further evidence that they moved north is a drawing of the method used to construct embankments on the Bridgewater. This shows boats similar to those using the Upper Severn around the turn of the century.
If boatbuilders on the Bridgewater Canal did come originally from the River Severn, their techniques for building narrow boats would probably have been similar to those they used for Severn barges. Do such techniques vary across the country, and if so what deductions can we draw from this? A brief survey of surviving wooden barges and documentary evidence from around the country provides some answers to these questions.
The traditional barge of the north west was the Flat. Surviving wooden flats are the Mossdale at the Boat Museum, the Staintondale next to the Weaver at Frodsham (now destroyed) and the Oakdale in Liverpool (now at Lytham), while it is possible to see sunken remnants at Sutton Level Lock on the Weaver or at Spike Island in Widnes where the St. Helens Canal starts. Among those at Sutton Level Lock are the remains of the Daresbury, Gowanburn and Eustace Carey.
The Mersey flat is a carvel built boat, that is the planks are placed edge to edge and the gap between filled with oakum. It also has a flat bottom to allow it to take ground safely at low waters in tidal areas, essential for loading and unloading where there is no dock. It is round chined, which means that, in cross section, the hull curves round from the bottom to the side. To accommodate these criteria all the planking is parallel and runs from stem to stern. However the most important features to notice, in this comparison of national boatbuilding techniques, are the bows and sterns, in particular the stem and stern posts. These are of composite construction, the stem or stern post having a piece of wood called the apron, about three times as wide, being bolted behind it. The planking is spiked to this apron with the stem and stern post protruding beyond the surface of the planking. This protects the ends of the planks, which butt up to it, from damage in locks and docks. The bow and stern were strengthened by two hooks, large curved pieces of wood fitted just under the deck and joined by a triangular knee just behind the apron. These features are common to all barges found in the north west of England, including those of the Leeds & Liverpool Canal and northern Irish canals.
Boats on the Leeds & Liverpool varied from one end of the canal to the other. Those in Lancashire were built in a similar way to flats, with many boats having a square stern. Around the Mersey estuary this square stern formed an important part of the boat_s carrying capacity, the transom being fully immersed when the boat was loaded enabling it to carry up to five tons more cargo. Elsewhere transoms are much less pronounced and were used more to increase the size of the stern deck and the stern cabin. They would undoubtedly have given more stern buoyancy when sailing in tidal waters. The square transom was only introduced into Yorkshire around the start of the twentieth century, and it seems likely that the technology came via the Leeds & Liverpool Canal. The methods for constructing the full bows of Yorkshire boats reach Lancashire in the same way, with the framing of the bow and stern of the Oakdale, the last wooden barge to be built on the Mersey, showing a definite Yorkshire influence.
Yorkshire and Tyne keels and sloops were originally clinker built, with their planks overlapping and clenched together, until the end of the nineteenth century. Recently surviving examples had carvel construction. Stem and stern posts were of composite construction, though here there was a rebate in the post into which the ends of the planks fitted. There is a suggestion that the planks were not spiked into the apron at the bow as, because of the full lines of the bow, the planks became too thin where they met the stem. Elsewhere the spikes were driven right through the plank and frames and then bent over, a survival of clinker technique not found in Lancashire. A further difference was the layout of the frames at bow and stern; in Yorkshire they were almost in line with the keel and butted up to a transverse frame, while in Lancashire they radiated from a single point. The Yorkshire method allowed the full bows, typical of keels and sloops, which increased carrying capacity. A further difference was the use of three hooks, one fitted across the bow or stern, the other two linking this to the first transverse beam. This was the Yorkshire technique used on the Oakdale.
Further south the Norfolk wherry was also clinker built, though with the much finer lines necessary for them to sail on the comparatively shallow waters of the Broads. Thames sailing barges are completely different, the edge of each plank having a rebate which overlapped the adjacent plank when assembled, the joint being covered with elk hair and tar beforehand to make it water-tight. The bow and stern are round chined whilst the hold area is hard-chined, the flat bottom of the boat meeting the side at a right angle. Here the planks met edge to side, whilst at bow and stern they met edge to edge. This resulted in a strange stepped joint where the construction changed from hard to round chine.
The construction of narrow boats is not the same as any of these various types barge. However, things are different when we look at the barges of the south west of England. On the River Severn there were two types of barge. Firstly there was the clinker-built up-river barge carrying 20-40 tons. These barges were mainly found upstream of Worcester, the last ones ceasing work early this century. Unfortunately none survive. The second type was the 'trow' which carried 60-80 tons and was found on the lower reaches of the river below Worcester and on the tidal estuary. They were of carvel construction, were round chined and had the flat bottom of the typical coastal trader. But it is in the type of stempost that an important variation occurs. On Severn Trows, along with similar sized boats built in south west England and southern Ireland, the stempost was made from one piece of wood, with the ends of the planks forming the hull being spiked into a rebate in the side of the stempost to protect the plank-ends from damage. The bow shape which accompanied this method of construction was much less bluff than the bow which is found on the Mersey flat or the Humber keel. The strengthening for the bow was also less substantial then the northern boats, comprising simply several knees and cross beams as opposed to large 'hooks'.
Although Severn Trows have round chines, square chined boats were also found on the Severn and in the south west. Barges for the Stroudwater Canal were built using this technique, as were those associated with the Hackney (Teignmouth) Canal, and many dumb barges used on the Severn also used this method of construction. Planking for the bottoms of these boats runs transversely in stark contrast to barges elsewhere in the country. It is possible to find the remains of many trows and river barges at Purton near Sharpness while the trow Spry may be seen at Ironbridge. The preserved Tamar barge Shamrock, and Galway Bay Hookers, which used to carry cargo along the west coast of Ireland, also show similarities to trows.
How do these various types of construction compare with narrow boats? The starvationers or mine boats and the box boats built at Worsley to transport coal into Manchester are arguably the original narrow boats, though it was only after the decision to reduce the dimensions of the Grand Trunk Canal that the narrow boat became a distinct type. Early lithographs of narrow canals often show boats similar to those built at Worsley and used until the 1970s by the NCB. Other contemporary illustrations are more difficult to interpret. Some show boats with rounded bows, though without more detail of the construction it is difficult to know where this form came from. The design certainly seems to have disappeared fairly quickly, possibly because it was found unsuitable for shallow canals. The narrow boats around Manchester were constructed using the hard chine method with transverse bottom planks while the stem and stern posts, unlike other barges built in the north west, were made from one piece of wood in the same way as boats built on the Severn. In conjunction with the survey of similar boats nationally, this suggests that the boatbuilding techniques used at Worsley had their origins in the area around the river Severn.
One question which does arise concerns transverse bottom planking. This type of construction may have been introduced at the same time as canals as earlier flat-bottomed hard-chined boats had longitudinal planking similar to that still found on some wooden inland boats in Holland. The caulking of transverse seams make it less likely to be damaged by contact with objects on the bed of the canal than the more traditional longitudinal seams. Unfortunately little documentary evidence remains of eighteenth century boat building to confirm or deny such speculation. Like most skilled trades, boatbuilders handed down their knowledge verbally, and they were certainly conservative by nature and only took up new techniques slowly or if they were obviously of advantage.
The terminology used by those who built narrow boats is similar to that used by other boatbuilders. The term 'knee', used by narrow boat builders for pieces of wood which strengthened the joint between the side and bottom of a boat, is in general use in boatbuilding for any piece of wood which strengthens an angled joint. 'Frame' in narrow boat construction is the term used for the individual pieces of wood onto which the planking is fixed around the bow. In much the same way, in boat building this term applies to the parts of the wooden skeleton of a boat onto which the planking is fixed. The method of hanging the rudder on Worsley-built boats also follows normal boat-building practice, a vertical iron pin through iron eyes holding the rudder to the boat. The method now used on dumb narrow boats, where the rudder hangs on cups fitted to the stern post, is probably a more recent development, designed to reduce damage to the rudder and stern post by allowing the rudder to rise when hitting obstructions on the canal bed.
Although narrow boats seem to have followed the traditions of the south west in their construction, their size is probably related to the north west. Prior to the Bridgewater canal, locks on the river navigations of the north west were approximately 68ft long with a breadth of 17ft; a size suiting inland Mersey flats (Those on the Douglas Navigation was probably somewhat smaller). These dimensions come from local registers of shipping, flats seeming to divide into two groups; the larger coastal vessels and those smaller ones which could be used on inland navigations.
The first boats built to work in the Worsley mines came in two main sizes, 47 by 4.5ft and 55 by 6ft, although there was a further smaller inspection type and actual sizes would vary. The only cargo carrying boat to survive until recently - once preserved at the former NCB museum at Lound Hall near Retford - measured 49ft 3ins by 6ft 4ins. Other preserved mine boats are much smaller and were of the maintenance type.
Larger boats built exclusively for canal use were constructed to fit the then standard lock length of 68ft and most narrow boats built for the Bridgewater Canal, particularly those from Worsley, were constructed to this length until boatbuilding finished at Worsley. When the Manchester, Bolton & Bury Canal was built, 68 feet was the length of lock used, the canal being designed to suit existing narrow boats and the wide boats from the Leeds and Liverpool Canal.
Why, then, when the Trent & Mersey Canal was built, were its locks 72ft long resulting in the development of the narrow boat? Financial restraints were the original reason, though the actual difference in costs between a wide and narrow waterway were not that great. The promoters of narrow canals were still unsure of the benefits of canal transport, though not all canal promoters thought so - almost twice as many canals were built wide as narrow. A further reason could be the difficulty of providing standardised measurement in the eighteenth century. Canals were the first civil engineering undertakings to require co-ordination in building standards over long distances, and to ensure that the original standard Bridgewater narrow boat of 68ft by 6ft 6in could navigate a canal, locks were made slightly larger to take account of any variation in measurements by local builders.
This could be why 7ft was used for the width and not the 6ft plus standard of the Worsley mine boats. A certain amount of excess width would also make it easier for boats to enter locks by allowing the water to be displaced. However, when the canal was finished, this precaution was found to be unnecessary and so boats for the Trent & Mersey were built using the maximum dimensions of the locks. It was important to do this on the shallow Midland canals in order to maximise the carrying capacity. However, in the north west, narrow boats remained at 68ft long and to increase the carrying capacity the Bridgewater 6-planker was developed, increasing the draught to increase the tonnage carried, only possible because the canals in the north west were built deeper to accommodate flats. Although it is difficult to be conclusive about the origins of the narrow boat, it seems likely that its construction design originated on the River Severn and that its dimensions are based on the early river navigations of the north west.
Finally, why were there such great variations in barge construction methods. One reason is that in Britain boat building is based on two methods: the Scandinavian clinker and the Mediterranean carvel traditions. Over the country, from north east to south west, there is a gradual change in the influence of these two traditions, though there is no definitive dividing line. The composite stem and stern post possibly originates from Nordic traditions where, on early boats, they were very much of a 'Y' cross-section, the back of the post having to be hollowed out to allow the plank fixings to be clenched or bent over. Carving such a shape from a single piece of wood created weakness, and the introduction of the apron may have improved matters. It certainly allowed for a much bluffer bow to be built which increased carrying capacity.
First published in 1997