Archaeology of the Peak Forest Tramway

In this report the state of the studies on the Peak Forest Tramway is examined in the light of industrial archaeological research. Industrial archaeology is the youngest of the many branches of archaeology but to many, archaeology is still the study of antiquities of the prehistoric and Roman eras. Industrial archaeology is not just the study of the Industrial Revolution. The study of a Roman rutway is just as much industrial archaeology as the study of an 18^th century tramway.

Much work of this nature has been accomplished by the study of documentary evidence such as company drawings, minute books and letter books or by talking to elderly workmen. While this has been an invaluable aid in establishing maxims which have been found acceptable, the application of classical archaeological techniques sometimes supports them and sometimes opposes them.

While the quantity of data is increasing, concepts on the subject are constantly changing and it could be considered too early to write this report, but a start had to be made sometime. The restoration work at Bugsworth Basin has increased the data and the questions which have been asked have made a substantial enlargement of our understanding. Many of the questions asked do, however, remain to be satisfactorily answered. The new data also queries the many interpretive writings on the subject and the treatment of this report has been dictated by primary evidence rather than by conjecture. Where sufficient evidence was available a synthesis has been established to embrace major problem areas.

Following abandonment of the tramway in the 1920's it was dismantled and sold for scrap in the full knowledge that an important monument of early railway history was being destroyed. Efforts to make records and preserve a representative collection of equipment were made at the time but techniques were not as advanced as they are today and economic circumstances probably aggravated the situation, much as they do now.

Inevitably the scrap merchant did not remove everything and since the 1920's there has been an accumulation of quantities of artifacts, mainly into private hands but some into museums. The main collection is in the National Railway Museum at York but Buxton Museum and the North Western Museum of Science and Industry at Manchester also have some material. While museum material is always accessible the material in private hands is not, unless the owners kindly come forward and make it available for study.

The restoration of Bugsworth Basin therefore presented the final opportunity of increasing our understanding of one of the most important tramways in the country. Consequently volunteers working at Bugsworth were asked to look out for artifacts, large and small, without paying any attention to their state of preservation. They were also requested to look out for the remains of demolished buildings, wharfs and similar structures. The response to these requests was magnificent and armed with the lively mental picture of what the Basin looked like in its heyday, everyone soon developed sharp eyes. The local inhabitants of Bugsworth should not be forgotten either, for they too have generously made artifacts available for study.

The Blacksmith's Shop at the Top o' th' Plane, Chapel-en-le-Frith.
Harry Fletcher is holding the tramway horse and Joe Marchington is
standing in the entrance on the far left of the photograph.
This photograph is the only evidence to show that curved cast-iron rails
were actually used on the Peak Forest Tramway. It appears that these
rails were only laid in workshop areas and sidings where empty wagons,
or an occasional loaded one, would be used. Photo IWPS Archive

General Construction of a Tramway and the Roles Played by the Engineers Benjamin Outram, John Smeaton and John Curr

The principal features of a tramway were L-section (flanged) rails and stone sleeper blocks. The method of securing rails to the blocks was to drive a spike through the end of the rails into a hole in the block which contained a wooden plug. Wagons were fitted with flangeless wheels and the flanged rails provided guidance.

It has now been established that the flanged wheel, as used on modern railways, was a common feature of the medieval mining industry, particularly on the Continent. On the other hand, there is only one reference to the use of flanged rail before 1776. This was for a tramway built by John Smeaton in 1756 to carry stone for the construction of the Eddystone Lighthouse. Smeaton published an account of this in 1791, three years before the Peak Forest Canal Company was incorporated⁽¹⁾. It is thus a fallacy that the flanged rail was developed first and that there was a transition period during which the flange was moved from rail to wheel. In 1776 John Curr, a remarkable engineer from Sheffield, built a tramway in colliery workings at Sheffield. He then wrote a book about it which was published in 1797, about two years after work on the Peak Forest Tramway had commenced⁽²⁾.

Benjamin Outram was the consulting engineer to the Peak Forest Canal and Tramway and which of the other two engineers had the greater influence on him may never be conclusively proved. However, an examination of Outram's Grand Aqueduct⁽³⁾ at Marple suggests that Smeaton probably had the greater influence. Smeaton designed stone arches with pierced spandrels in order to reduce the weight of structures without any loss in strength. Outram's Grand Aqueduct also has pierced spandrels.

Outram, who was born at Alfreton near Matlock in 1764, became an adherent of tramways, or plateways as they were sometimes known, and due to his enthusiasm they came into widespread use. With William Jessop, John Wright and Francis Beresford he founded the famous Butterley Ironworks in 1790, which traded under the name of Outram & Company. When the Peak Forest Canal Act was passed in 1794, Outram was appointed as consulting engineer and it was natural for him to recommend a tramway to connect the limestone quarries at Dove Holes to the Peak Forest Canal. The canal company agreed and work on the tramway commenced in circa 1795, the rails being provided by Outram & Company⁽⁴⁾.

Stone Sleeper Blocks: From 8 to 12 thick and circular, square or rectangular in shape and weighing between 150 and 200 lbs. The base to be flat with a small portion of the upper surface level to provide a bed for the end of the rails.

Holes and Plugs: A hole 1˝ in diameter and 6 deep to be drilled in the centre of the stone block to receive an octagonal oak plug 5 long.

Spikes: The plug to receive an iron spike or large nail with a flat point and long head adapted to fit the countersunk notch (slot) in the end of the two rails and so secure them to the block.

Standard Rails (L-section): To be of cast iron, one yard long with a flange on the inner edge 2˝ high at the ends and 3˝ high in the centre. The base of the rail to be not less than 4 broad, and the thickness proportional to the work they were intended for.
(The key identification feature is that the rail flange was hump-backed or arched).

Level Crossing Rails: The base to be double the thickness with low flanges.
(This means that a channel or 'U' section was used for level crossing rails).

Points: Rails of particular forms are necessary where roads (lines) branch out from or intersect each other.
(Unlike Curr, Outram was quite brief about this significant feature).

Gauge: The distance between the flanges should be 4' - 2". This being fixed by an iron gauge.
(This is the most controversial aspect of Outram's recommendations. The word between means the space separating two places and it is not necessarily a synonym for inside. It would have been easier and more logical to use a gauge inside the rail flanges than outside them and even today the standard gauge of 4' - 8˝" is measured inside the rails).

Curves: Great care is required to make them perfectly easy and that the inner part of the curve should be fixed a little lower and that the rails should be set a little under gauge.
(Outram was also quite brief about the construction of curves, whereas Curr went into greater detail. The overall length on the inner part of a curve is less than on the outer part. If rails 1-yard long were used in curves then this would have resulted in stone blocks no longer being opposite to each other; they would be out of phase. Our present state of knowledge shows that Outram only used straight rails to traverse curves whereas Curr had them specially cast to suit the radius. Technically, this made Curr's design more advanced than Outram's).

Around Bugsworth and along the tramway the stone blocks conform reasonably well to Outram's original recommendations, even though many of them were probably replaced in 1835. They are from 8 to 12 thick and are irregularly shaped. The upper and lower surfaces are fairly flat and parallel. Subsequent modifications to the tramway were also constructed with similar sleepers. Interestingly, evidence for the use of conventional wooden sleepers, which physically tie adjacent rails together, has been discovered, particularly on the south side of the Lower Basin at Bugsworth.

Track detail 1797 to 1832/37 conforming to Outram's recommendations. Note the groove worn into the surface of the stone block by the rail. This wear, coupled with the wear on the spikes and rail breakage, caused serious problems in the operation of the early tramway

Track detail for some parts of the tramway from the early 1830s. This design was probably influenced by the principle of fixing edge rails on the Liverpool and Manchester Railway which opened on 16 September 1830

The holes are about 1˝ in diameter and 6 deep but sometimes they are lobed rather than round. This feature can easily be simulated today by using a drill which is sharpened off centre. The type of drill used was a jumper or cruciform-shaped drill of a design which can still be obtained today. In use it had to be struck with a heavy hammer while turning it slowly by hand.

Concerning the plugs, little material has been found and it was in such a poor state of preservation that lengths could not be determined nor was impossible to tell whether or not they were originally octagonal in section. The type of wood actually used has yet to be determined.

Some wrought-iron spikes conforming to Outram's original recommendations have survived but it is unlikely that any of them are contemporary. They have rectangular heads to fit the countersunk slots in the rails. They are variable in length and section and this indicates that they were forged without the use of shaped dies.

A greater quantity of round spikes have survived and these belong to the period 1865-70 when much of the main line was relaid using steel rail. They too are variable in length and section, which again indicates forging without shaped dies. The apparent absence of more advanced techniques for their manufacture in the second half of the 19^th century is surprising. Judging by their grained appearance these spikes were also made of wrought iron but it would be useful if a metallurgical examination could be made to settle this question. Other round spikes with larger heads are associated with a saddle having two fixing holes.

Standard cast-iron rails (L-section), formerly known as gang rails or plates

The word standard could be considered misleading and one is reminded of the statement made by Joseph Whitworth in 1880, "That candle butts and candlesticks came in so many sizes that they often did not match." And so it was with standard rails.

Unquestionably, the most important finds were several fragments of rail which, after cleaning and measurement, were found to have hump-backed flanges. These conform to Outram's recommendations and belong to the earliest period of the tramway. Another feature of the design was that the ends were cast thicker to enable them to seat better on the stone blocks. These have been designated as foot-mounted rails and they were mounted directly onto the sleeper without a saddle. It was next necessary to conclusively prove that they were examples of the earliest type of rail to be used on the tramway. If the theory was correct then somewhere along the tramway there should be stone blocks with an impression in them, the width of which corresponded to that of the rail, but most important of all, there should be a striation across the hole corresponding to the rail joint. Brian Lamb kindly walked along the tramway to look for this evidence and eventually found it in the vicinity of Chapel Milton. Subsequently stone blocks conforming to this were also found at Bugsworth.

As built, the track was single line throughout but in 1803 it was doubled except for the sections through Stodhart tunnel and the bridge carrying the A624 over the tramway at Chapel-en-le-Frith. Although it cannot be proved, it is almost certain that the second line was identical to the first because there would have been insufficient experience at that time to show the major defects in the Outram design. These defects will be discussed later.

From this period onwards the design of standard rails, as found, becomes confusing. From the beginning Outram and the canal company recognised that economic circumstances dictated the size of the rail section. Rails for sidings and little used branches could be made with a cheap light section, whereas rails for the main line had to be made of a more expensive heavy section. Wooden patterns, used to make sand moulds, quickly wore out and had to be replaced. A new pattern was always slightly different than its predecessor. A rail section subjected to heavy use would wear differently than the same section subjected to light use and similarly a rail section located on a curve would wear differently than the same section on a straight portion of track. Finally, it is likely that the rails were eventually purchased from several suppliers. It has not yet been possible to find a way out of this maze but the main details of standard rails are summarised as follows:

Fragments of two types have been found and these only differ in the design of the foot.

One fragment, presented by villagers, has been examined and though it is foot mounted the flange is of uniform height. It is not known whether this is of Outram or post-Outram design.

Note: The next two groups of rail are post-Outram, have uniform height flanges and are not foot mounted. They were designed to overcome the serious breakage problem experienced with the Outram rail.

Examples have been found and the exhibits in the National Railway Museum and Buxton Museum also belong to this group. These rails are generally well worn and are dimensionally inconsistent. They probably date from 1822-35. One rail, though damaged, was found to be less than 1-yard long and could have been cut shorter to suit a set of points.

Several examples have been found but there are no exhibits of this type in museums. The rails in this group are only slightly worn and are dimensionally consistent. Considering the slight wear, they are probably of late date and could even be contemporary with the change over to steel rail on much of the main line.

Assembly of hump-back, foot mounted rails to Outram's recommendations. Viewed from outside the track.
Photo P J Whitehead

Various types of rail shown assembled. Viewed from outside the track. From left to right: level crossing rail, low-flanged heavy section rail, transition rail and steel rail.
Photo P J Whitehead

Cast iron has a low tensile strength and it was found that even a heavy section cast-iron rail did not completely solve the breakage problem.

To manufacture rails of wrought iron would have been an obvious solution to the problem, for this material has a far higher tensile strength. Joseph Hall developed his wet puddling process in circa 1830, which greatly reduced the price of wrought iron but the canal company do not appear to have considered its use. Economic factors and the scale of production of wrought iron may have made it so impractical that the subject was never considered.

Following the introduction of the Bessemer converter process in 1856, steel became available on a commercial basis and about a decade later it was being used extensively for conventional railway rails. In 1863 the Peak Forest Canal Company passed into the full control of the Manchester, Sheffield and Lincolnshire Railway Company. This company had extensive workshops in Gorton, Manchester and they had the facility to roll flanged steel rail. The decision was made and sometime between 1865 and 1870 the changeover to steel rail was made.

These rails were 9 and 12 long with two countersunk fixing holes in the flange at each end. The existence of the 9 rail has only been proved from photographs but there is both photographic and physical evidence for the 12 rail.

The method of fixing to the stone blocks was similar to that adopted for the earlier cast-iron rails. The saddle (see next section) was still used but at every new rail joint the old saddle was removed and replaced by one of special design. Abutting rail flanges were connected using a steel fishplate 18 long.

The fishplate contained four bolt holes for fixing purposes. The countersunk-headed bolt incorporated a nib which fitted into a groove in the rail fixing hole and so prevented the bolt from turning when the square nut was tightened.

The dimensions of the screw thread and the square nut have been carefully measured and it was found that they conformed to the original Whitworth standard for a thread size. This must be viewed with the knowledge that it was in 1841 that Joseph Whitworth read his distinguished paper "On a Uniform System of Screw Threads".

It would be impossible to traverse sharp curves using 9 or 12 foot-long rails and photographic evidence indicates that only the straighter sections of the tramway were converted. Typical locations where this was done are the incline in Chapel-en-le-Frith and on the approach to Bugsworth. The 242 foot-long mobile tippler pier situated on the north side of the Lower Basin (the New Drop) is also known to have been fitted with steel rails.

At certain places it was necessary to connect a steel rail to a cast-iron rail. There were four methods of accomplishing this transition.

Fishplate joint between successive lengths of steel rail viewed from inside the track. Only the second nut and bolt from the right are original. Washers were not used. Note the incised number '6' in the saddle.
Photo P J Whitehead

The same joint viewed from outside the track. Note the nib incorporated in the head of the bolt which prevented the bolt turning when the nut was being tightened.
Photo P J Whitehead

The first three decades of the 19^th century showed major defects in the Outram design. The constant movement of wagons over the rails wore deep grooves in them and this caused two problems. It will be remembered that the early rails were foot mounted and this was done by making the ends thicker. The grooves made the rails much weaker in the centre and rail breakage became a serious problem. The wagons also wore the spikes securing the rails to the stone blocks, with the result that the rails became loose.

The solution to the first problem was to design a new heavy section cast-iron rail with a flange of uniform height. The solution to the second problem was the introduction of the saddle. This was made of cast iron and was approximately 8˝ x 7˝ with a central fixing hole and two flanges to locate the rail. The method of fixing was the same as before, using a rectangular-headed spike to fit the countersunk slots in the rails, the spike passing through the central hole in the saddle. The oak plug in the stone block was made longer so that it extended into the hole in the saddle.

The proceedings of the Peak Forest Canal Company for the 19 and 20^th June 1833 state that, ".... all the Plates and Pedestals to be examined at Bugsworth prior to being sent to the Limestone Quarry. The blocks where the Pedestal rests upon to be paralleled to the underside of the Pedestal not to be dished out in the manner that some have that were pointed out, but quite flat on the surface⁽⁶⁾ ....". For the 4^th June 1835 the proceedings state, ".... that the Committee have continued to improve the working state of the railway and in the last year 4,010 yards have been taken up and laid with new rails and Blocks and Pedestals upon a new principle and 3,158 yards have also been repaired ....".

Five types of cast-iron saddle with a central fixing hole have been identified at Bugsworth. Four of these date from or after 1833-35, while the fifth dates from 1865-70. This particular saddle is readily identifiable in that it has low flanges and was invariably cast with an incised number '6' underlined to prevent any confusion with '9'. This special saddle was used at the joint between lengths of steel rail and was so designed that the inner flange did not catch the fishplate connecting the rails. As the rails were connected with a fishplate, the special saddle was fixed directly to the stone block by a headless spike which was driven into the oak plug until it was flush with the surface of the saddle.

Buxton Museum has an example of yet another cast-iron saddle. This has a fixing hole in the centre of a lug on either side of the flanges as well as a central hole. Additionally, each flange has two rectangular holes cast into it, the purpose of which is completely unknown. Similar saddles have been discovered at Bugsworth but these lack the central hole and the rectangular holes in the flanges. The method of securing rails to this type of saddle was by means of steel or wrought-iron wedges. Dating evidence is of even more importance. Does this saddle represent the "new principle" referred to in the proceedings for 4^th June 1835? Were single-hole saddles a later design? They would certainly have been cheaper to manufacture and assemble.

Another saddle consisted of a simple channel section with a central fixing hole. It was of fabricated construction made from wrought iron/steel. Only one has been discovered and this was in association with a low flanged heavy section rail. The wharfs at Bugsworth were examined and an impression which fitted this saddle was found in a coping stone on the south side of the Lower Basin Arm.

A small artifact used in connection with saddles was a spacing plate. This was placed below the rail joint and it had a central hole to enable the spike to pass through it. These plates were used where necessary to facilitate assembly of the rails by ensuring that they seated properly in the saddles. Subsequently it was discovered that these plates were also used in cases where the rail was fastened directly onto stone block without the use of a saddle.

Gritstone sleeper block with linear rail indentation and fixing spike hole. Photo P J Whitehead	Iron fixing spike. Photo P J Whitehead
Cast-iron saddle. Photo P J Whitehead	Cast-iron saddle in situ on the Central Peninsula at Bugsworth Basin. Photo P J Whitehead

Only one broken example of a level crossing rail has been found and this conforms to Outram's recommendations. The base is double the thickness of the centre portion of a standard hump-back rail and it consists of a channel or 'U' section with low flanges of uniform height.

The method of fixing was the same as for L-section rails , using spikes, but it is not known whether special saddles were ever used with them. Photographic evidence shows that they were sunk into the road until the flanges were level with or slightly below the surface of the road. Following the transfer of the canal and tramway to the Sheffield, Ashton-under-Lyne & Manchester Railway in 1846, edge rails lying on their side were also used as level crossing rails.

One example has been found at Bugsworth, complete with switch⁽⁷⁾, and there are two others in the National Railway Museum. All are of the hump-back type and are foot mounted.

As there were no point levers, the mode of choosing the required track was to kick the two switches to point in the direction needed.

One small fragment has been found at Bugsworth. The underside of this was cast with an incised number '4' which could date it to the period 1865/70. The angle of cross is just over 18 and from measurements of this fragment and an examination of a photograph of a cross-over rail it has been possible to synthesise the design. There should be an example of a cross-over rail in the National Railway Museum but this was removed many years ago. There is also photographic evidence of another type of cross-over rail with an angle of cross of about 39⁽⁸⁾.

A few double spacing plates have been found and these were used in connection with point saddles.

The set of point rails in the National Railway Museum is not a complete set removed whole from the track bed. Rather the rails were collected at random from a pile of scrap as the track was being dismantled.

It has not been possible to determine the actual gauge of the track to compare it with the recommended gauge of 4 - 2. Opposite stone blocks were not fastened to each other in any way and over the years these have moved relative to each other. The measurement of distances between the holes of opposite stone blocks in situ would consequently be of little value. Additionally, engineering limits and fits were not used and clearances were always made very generous in order to ensure that the wagons would run no matter how inaccurate the gauge.

For similar reasons, it has not been possible to do any work on the construction of tramway curves. Available evidence indicates that standard 1 yard-long rails were used to traverse both the inner and outer parts of the curve. As mentioned previously, this would result in opposite stone blocks becoming out of phase with each other.

Photographic evidence shows that in some locations curved cast-iron rails were used when the radius was relatively tight. One such location was the track leading into the blacksmith's shop at the Top o' th' Plane, Chapel-en-le-Frith.

The most laborious part in the restoration of iron and steel artifacts is the removal of rust. In the absence of shot blasting equipment this has to be done using a hammer and chisel followed by wire brushing either by hand or using an electric drill.

The artifact is then treated with a proprietary rust remover. If it is to be left in this condition it is next treated with a rust inhibiting fluid consisting of equal parts of raw linseed oil and turpentine (not turpentine substitute). If it is to be painted then, after application of rust remover, it is given two coats of zinc chromate primer paint.

It was found that wrought-iron artifacts did not respond very well to this treatment and after unsuccessful attempts using pickling solutions another method was adopted. This consists of heating the artifact to about 900C and immediately quenching it in water. After repeating this process once more it was dried at about 150C and any rust still adhering was removed using an old saw blade or a wire brush. The rust remover was then applied in the normal way.

Parts of flanges on rails and saddles are frequently missing and if the particular artifact justifies it, then it has to be repaired. This is done by cutting a piece of plywood roughly to shape and fastening it in place with epoxy resin adhesive. The plywood is then built up with a proprietary filler which is then sanded down to shape. Small holes and cracks are filled using cold casting iron.

If money were no object, then there are very few artifacts which could not be repaired to a fully working condition. This particularly applies to wagon wheels, most of which are broken when found. This could be done using the well known thermit process.

It has only been possible to examine tramway artifacts discovered or donated, together with the small amount of material in museums, as these were the only sources available. Many pieces of the jigsaw are still missing and if artifacts discovered prior to commencement of this work could be made available for study, it may be possible to obtain a more complete picture of the tramway for the benefit of posterity.

This section of the report must be considered as purely speculative. Various parts of the country are establishing heritage centres for particular industries but a heritage centre for the limestone extractive industry does not yet exist. The Peak Forest Canal and Tramway were built almost exclusively for the limestone extractive industry and, because of its historical importance, Bugsworth Basin is the natural choice for the site of such a centre.

In connection with this it would seem appropriate if the goods warehouse in the Lower Basin Arm and the lime sheds in the Upper Basin and Middle Basin Arm could all be rebuilt as heritage centres. Artifacts associated with the tramway and canal could then be displayed in them. Some track could be relaid in its original position inside the buildings and tramway wagons of different types could be reconstructed, using available artifacts and evidence, and displayed on it.

Only one wagon from the Peak Forest Tramway has survived. This is mineral wagon number 174 which is now in the National Railway Museum at York. They were used to carry limestone from the quarries near Dove Holes to Bugsworth under the action of gravity. It consisted of a substantial wooden chassis with a wrought-iron body held in place by two wooden wedges. The wheel axles were bolted onto axle trees and cast-iron wheels (about 20 inches diameter) were secured on the axles by a large washer and a lily-pin. A wagon could carry a load of about 2 to 2˝ tons. The ganger and nipper (apprentice) controlling a gang of wagons stood on the lily-pins and kept the speed at 4 to 6 mph by spragging the wheels to make them skid. A gang of wagons comprised up to 40 wagons, loosely couple together by chains, with a net weight of 80 to 100 tons.

I would like to acknowledge the contributions of all those friends who have helped by the discussion of ideas and content and by helping me check typescripts, proofs, etc, also in particular:

Brian Lamb, for his help in the investigation of the original Outram rail, for photographic evidence and for the information from the Proceedings of the Peak Forest Canal Company

Buxton Museum and the National Railway Museum for allowing their collections of material to be measured and photographed.

A Narrative of the Building and a Description of the Construction of the Eddystone Lighthouse with Stone, (1791).

John Curr, The Coal Viewer, and the Engine Builder's Practical Companion, (John Northall, 1797).

M J T Lewis, Early Wooden Railways, (Routledge & Keegan Paul Ltd, Paperback edition, 1974).

1. John Smeaton, A Narrative of the Building and a Description of the Construction of the Eddystone Lighthouse with Stone, (1791)

2. John Curr, The Coal Viewer, and the Engine Builder's Practical Companion, (John Northall,1797).

3. The Grand Aqueduct is superlative in its class and is, without doubt, Outram's masterpiece.

4. Minutes of the Peak Forest Canal Company
10 December 1794, '....4. Outram to supply Peak Forest Canal Co with Cast Metal Railways by Benjamin Outram & Co.... Clerks to affix Seal'.