This article contributed, in part, to Adam Porter's article on the tunnel in Canal Boat January 2014.

A REVISED STORY OF THE CONSTRUCTION
OF BLISWORTH TUNNEL

Tony Marsh    February 2007

This article is a re-write of one placed on this website in October 2005.  The aim of the
re-write is to improve clarity - there is no change to the information presented.

Claim:  This account justifies the statement that two, not one, lines for the tunnel were tried before the
successful line was adopted.  It accounts for all of the known evidence for subsidence on Blisworth Hill,
including the strangely off-line vent near to Buttermilk hall farm, and provides an explanation for the
remains of what appears to be a water-course in a field known as The Park.  It also prompts a suggestion
for why a part of the canal bank is anomalously steep and why there remains only a narrow course for the
hill railway near the same place.

 

  Introduction      Plans for the tunnel through Blisworth Hill were deposited in 1793 and the northern part of the canal had been dug as far south as Blisworth by 1796. The deposited plans show the configuration of the canal (ie. its bends along the 'Parliamentary Line') outside the tunnel in accord, more-or-less, with the layout we now have.  Attempts to scale from these plans are hardly worthwhile.  The early attempt to pierce the hill with initial headings (narrow 'pilot' diggings) was reported to be dogged by excess water and some brickwork collapse. Work continued for many months on draining the hill, especially near the centre and south end. It was then reported that the 'works were out of line' which we take to mean that the line of vertical shafts (the bases of which would be joined by headings) were not in a sufficiently good straight line. It was then resolved to start again with a new alignment; that is, dig fresh vertical shafts and the new line was said to be 150 yards to the west of the aborted attempt at the south and a short distance to the east at the north. In the new alignment there is some reference to the line consisting of two tunnel sections joined by a short curved 'open-air' section near the north end. This complexity is later dropped and the project proceeds to completion in 1805 along the line we have today. This summary may be filled out with considerable detail by reading David Blagrove's book "Two Centuries of Service" available at the Museum at Stoke Bruerne. He has based his account primarily on the project minutes of the Grand Junction Canal Company.

A little more detail can be gleaned from the evidence on the ground.  A recent survey has been completed and the information, some of it entirely new, is presented in the figure to the left. The figure is based on an old OS map and the section on view is a strip of land running from the NNW to the SSE. In other words, to save space in the browser window, the map has been rotated somewhat in a clockwise direction. This is evident from the appearance of some reference points in red eg. E and D which can be seen to the left.  So, let's go through the evidence . . .

Fist the features labelled in blue; the Greek pi symbol, π (the browser may/may not successfully show this symbol) refers to a bend in the canal where the western bank is much steeper than anywhere else in the whole project.  I return to this in a discussion near the end.  The culvert entrance and exit labelled 'S' highlights the underground path provided for Fisher Brook by the engineers so that the water discharges safely into the canal.  North and south portals of the finished tunnel are labelled 'P'.

In establishing the headings for the final tunnel, 22 vertical shafts were dug down to a datum level having carefully aligned them all along a straight line which appears to have Stoke Bruerne church tower as its southern end point, being a 'target' visible quite a long way over Blisworth Hill.  The headings were dug on two faces at the bases of the shafts in order to establish a continuous heading.  When the tunnel was repaired in 1982, surveyors found all these shafts and checked that they were safely capped with brickwork.  They needed to know their precise positions prior to any repair work.  The marks labelled in red relate mostly to 'off-line' shafts, or rather, the evidence of them.  They were created whilst engineers c. 1797 pursued earlier, unsuccessful, tunnel alignments and are described below.  Some of these locations indicate land subsidence and the area near them can be assumed potentially dangerous to the public.

E.  A one foot depression over an area some 6 to 8 feet across.  This is centred on a small plateau of land near a field fence - see photograph which is taken from a low vantage point to the east in order to illustrate the 'plateau'.  The feature is located in a field known as The Park or Park Slopes - coloured lilac.  Note the counter argument regarding this feature given in the series of photographs.

D.  A low hump of spoil on which a electric distribution post happens to be placed.  In 1976 a large hole appeared at the centre of the hump and the pole keeled over.  In this photo the pole is shown sawn off while repairs were done.  In the distance is the Stoneworks farm house.  The feature is located at the north end of a field known as Fisher Close or Fishweir (also coloured lilac).  There was once a deep depression in Fishweir (partly filled by the traverse of a horse railway - orange line, see below) which is believed to have accommodated a large medieval fish pool.  The horse railway is also known as the hill railway, tramway, plateway, etc.

C.  A feature near Blisworth Hill farm which was discounted and is therefore not discussed.

 However the reason it was discounted is based on the farmer's statement that some ground subsidence that I picked up on at this location actually related to a swimming pool.  Another farmer, an acquaintance to the first farmer, 5 years later has told me that the idea of a swimming pool just there is nonsense.  So maybe location C is really associated with the proposed first tunnel.  I will not be able to confirm this without offending the first farmer ... oh well!

B.  A particularly tall brick chimney mounted over a shaft that is in use over the existing tunnel.  There is good reason to suppose it was first dug for one of the unsuccessful alignments.

B1.  A particularly short (6 feet) chimney over an off-line shaft that is connected to the existing tunnel by a slantwise heading.  This vent can be inspected from the Stoke Road in the direction of the pink coloured arrow.  Associating this vent with an earlier tunnel alignment provides a reason for it being built, yet still useful as a vent, 40 feet off the line of the finished tunnel.  Its diminutive stature is a puzzle.

G5.  This is a feature just south of B1 as indicated in a diagram prepared by George Freeston in May 1985.  He doesn't say what G5 is on the diagram but it is a safe bet that it marks another ground collapse.  G5 is exactly on the green line - see diagram as a footnote to this article.

A.  A small area at the edge of a field that has been fenced off by the farmer.  It had sunk in the past.  Interestingly, this feature was re-discovered in the recent survey by noting that T. W. Millner, regional chief engineer GJC Company, found a sunken feature in 1922 and gave, in a letter (GF Collection 1996 - Canals), its 'co-ordinates' as 1564 feet along the tunnel (from the south portal) and off-line to the east, perpendicularly, by 200 feet.  It is probable that no one, until now, has made use of the data in Millner's letter.  Transferring this data to a map placed us exactly at the feature - photo taken from the fenced path.

A1.  A sunken feature that has dropped 10 feet in the recent past and after refilling stands at 3 feet deep at present.  It was pointed out to me by the farmer.

Z.  The same farmer drew my attention to two zones of rather broad subsidence which have become established in the arable field. (no photographs)

First Portal.  The same farmer agrees with my supposition that the first portal at the south end was located here. (no photograph)

Grouping the Features.     All the features listed above, except for one - D, are on a good straight line.  It is therefore supposed that they were created to enable a continuous heading to be established from the first south portal to one at the north end that is tentatively labelled 'Greek pi', being in line with the bridge at the mill.

The 'good straight line' can be appreciated scientifically, in that there are a total of seven features which support each other with a high precision, excluding the location of the first portal and the feature marked E.  That seven such features might occur as a joint coincidence is very improbable indeed - in other words, the alignment is there for a supposed good reason. This is the chief revelation of this article.  That a less clear alignment simultaneously seems to exist for the turn in the canal at the north end, point E and the first portal provides the 'good reason' and prompts us to write the revisions to history - see below.

Let us assign the name First Portal Green to this line - FPGreen for short.  The feature at D is presumably associated with another set of headings (also using the initial south portal?) which we name FPBlue.  There is little to define it other than the feature at D but the farmer has pointed out that the stream seems to disappear underground now at the point, between A and A1, where it crosses FPBlue (the stream is marked blue in the map).  The point at which the stream disappears should be regarded as another potentially dangerous area.  Knowing that the hillside was pierced by numerous drainage headings, especially at the south end, we cannot attach much significance to this point.

In any case, FPBlue does fit the description for the alignment revision mentioned in the introduction and therefore, on the face of it, FPBlue might be supposed to be the second unsuccessful alignment.  However, the fact that FPGreen and the final tunnel both pass through (or under) the vent at B, being a particularly tall chimney, is very probably significant.  It bears on our view as to the order in which the two alignments were attempted - in fact contradicting the supposition above, ie. FPGreen followed FPBlue.

Writing the History ?

The status of the perceived straight line given in the sections above raises the proposition that FPGreen was really attempted. In the paragraphs labelled "Try this Idea" there is my interpretation. Note there is a piece of evidence from the management minutes was unearthed by George Freeston:

Minute of 17th May 1796: (paraphrased) that the agreed line through the hill be taken by "crossing the line of the present ruins near the north end and to enter the ground a little to the east and come out about 130 yards to the west of the present line, being the best and most eligible line that can be adopted"   So, a variance was proposed...  Let's call this FPWhite
Minute of 7th June 1796: "that the tunnel be executed on the original line" and, in the margin written Original line to be pursued.

This sounds like FPWhite was really FPBlue...

Try this Idea   For reasons of wishing to use the two completed halves of the canal, while the tunnel was being built, the engineers established a toll road in around 1795 to join Blisworth with Stoke Bruerne.  The road, which is now called the Stoke Road and is clearly shown running the length of this map, was by the requirements of the Parliamentary Act to run as close to the line of the tunnel as feasible and this is the case for FPBlue.  The toll point on this road has been indicated at one of the two orange letter Ts on the map.   A horse railway was established in 1800 because the horse and carts used on the road were an inadequate means of transport.  The horse railway was used until the tunnel project was finished in 1805 - its path is shown approximately by the orange line in the map.  It too was supposed to follow the line of the tunnel but, from an inspection of the topography, it is clear that it could not do so whilst achieving an even grade, being an important consideration for horse-drawn vehicles.  Actually, for some of the way, it followed FPBlue but lay directly over FPGreen near to Blisworth Hill Farm where it runs behind the farmhouse.  This could be coincidental as the farmhouse was established at some time before 1806 (Grafton rental records - farmer Samuel Wilson).  

Let's propose here that the progress in the project was made along the lines given here and then give a speculative account of the story of the project:

FPBlue and Toll Road  > [Revision 1]  > FPGreen and Horse Railway  >  [Revision 2]  >  Final Success

It is necessary to assume that the action to attempt FPGreen was never minuted by the management and that it was either quietly agreed with the Duke of Grafton or it was carried out illegally.  Of course we shall never know.  No measure of surprise need be felt as sharp practice and creative book-keeping were endemic in society circa 1800.  We must be thankful that the real heroes of the project were the engineers.  Within the limits of what was known then of tunnelling expertise and what we know now of the geology, the achievement is to be appreciated by all as simply monumental.

So, what happened in the two Revisions?  

Revision 1   The earliest work on the tunnel c. 1794 was carried out by Messrs. Jones and Biggs.  These contractors were soon laid off as the chief engineer, Barnes, took over the work in 1795.  Biggs had pulled out and Jones was found to be too slow and have presided over poor brickwork.  Although 'in charge', Barnes was presumably absent dealing with other parts of the project.  By the end of 1795, work was nearly at a standstill.  [see Blagrove]  Mention is made of excesses of water without indicating whether this was at the north end or the south end, or both.  Villagers these days know that Fisher Brook is extremely variable in volume, being fit for toddlers to paddle in (in high summer) and of sufficient volume to supply 50 cubic yards a minute of water at other times (after a few days of very wet weather).  The brook is the result of rainfall collection over an area of 3 square miles and is capable of filling a major works in hours.  It is possible that, in trying to establish the north portal for FPBlue, the rages of Fisher Brook had taken them by surprise - in any event the engineers must have become terrified by it.  Indeed, by choosing the 'cranked' course of the canal in its approach upto the intended tunnel entrance, the engineers had picked the approximate course of the brook and had probably invited inundation.

The rather radical revision to adopt the line FPGreen would bypass all the problems of Fisher Brook which could then be easily ducted to pour into the canal from the east, not far from the bridge near the mill.  Two other advantages were to arise from the plan.  One was that a long excavation into the hillside would not be needed, thus saving labour.  The other, as far as the villagers and probably the Duke of Grafton were concerned, was that a place known as 'The Green' (South Gutter Green on the 1729 map) would not be sacrificed to that long excavation.

To even consider the direct line, FPGreen, was to go against the terms of the Parliamentary Act and the detailed plans therein.  How could a band of engineers, led by Barnes, backed by managers in London, do that?  Evidently there was an arrogance, partook also, probably, by the Duke of Grafton that reasoned, "Look, this is a better plan, actually, we'll go this way and codge-up the terms later - Parliament will see our point of view!  By the way, let not our ideas enter the records..."  Does this sit well with what we know of the 1780 - 1820 times?  I leave others to comment.

Revision 2   By the evidence of a number of shafts, the FPGreen line was worked on for quite a time.  No doubt a north portal would have been created at a point near that labelled with a Greek pi.  The line was eventually abandoned. It could be that, in working the tunnel from the south portal, the engineers had experienced basically their first real encounter with the problems of underground water and they were unprepared.  It could be that their work collapsed.  There are two depressions in the fields to the north of the south portal, marked by red dots at Z in the map, which certainly line up with FPGreen.  A collapse, of course, could have occurred at any time since.

If there was a drastic problem with FPGreen, the engineers would be forced to again rethink.  They had an option to return to approximately the Parliamentary Line if the problem of Fisher Brook could be overcome.  Comparing FPGreen with the successful line, one cannot help noticing that the approaches to the portals at both ends are distinctly different.  For FPGreen the tunnel would begin with very little ground overhead, at both ends.  For the final line, a long excavation into the hillside (as originally planned, it seems) was adopted.  Whether the FPGreen portals with their brickwork were giving trouble we shall never know.  In abandoning the portals the one at Greek pi would have been dismantled.  It is at this time the engineers were mostly engaged on 'draining the hill' and many months would go by as they prepared for the final line.  The horse railway would have been completed and, to a degree, the pressure would have been relieved.

The Horse Railway:   It is a fact that the path of the railway generally shows room for a southbound and a northbound rail placed side by side.  Furthermore, sleeper stones have been found in 4 lines at places on Blisworth Hill.  If one walks south, starting at the bridge near the mill, there are to be seen in the first field two levels, side by side, for the railways.  As the path skirts the first bend in the canal, the accommodation curiously narrows to that sufficient for only one railway and the ground appears to have slumped somewhat.  Furthermore, the bank of the canal is much steeper at the bend.  There is a minute referring to a discussion about the safe angle to adopt when cutting the long excavations.   Barnes, it seems, wanted the angle to be 20 or 25 degrees as this is the angle everywhere else.  But someone asked that maybe the angle, generally could be steepened to reduce the amount of spoil to be removed.  The minute records the suggestion was turned down.  So, given the strong preference for a gentle slope at ~20 degrees, why is there a steeper bank just where the accommodation for the railway had narrowed?

One answer is this:  A line was adopted for the pair of railways curved over the ground just south of the portal at Greek pi and continued over the hill - in other words it was innocently placed very close to the edge of the eventual canal works but thought safe at the time it was placed.  When it became clear that a new line for the tunnel was needed, the portal would be need to be removed and then the course of the canal excavation would be close to the railways.  Barnes must have relented his grip on excavation angles and perhaps left a bit of the portal brickwork in place, for a while, so as to support the railway nearest to the canal, this being satisfactory and not requiring further excavation on the other side of the railways (the west side) to make further accommodation.  He knew that stability for the railway was not a long-term requirement.

Another partial answer is this:  A 1947 aerial photograph (NRO) of the area shows that the footpath was located some 3 or 4 yards further south in 1947 compared to now.  (It is an intention to overlay Google and RAF maps in due course)  The bank seems to have slipped and the slippage could have created a steeper descent to the canal.  Now the slippage seems to be quite noticeable at the turn in the canal and we need to ask why - maybe the previous excavation of a portal!

An attempt to preserve the portal at Greek pi (but perhaps rotating it slightly counter-clockwise) for the sake of the railways would perhaps have generated the idea to have two tunnel sections (as mentioned in Blagrove's book).  There would perhaps have been planned a short section of tunnel to replace a substantial part of the deep excavation, followed by an open curved section leading to the north portal as we know it.  The problems regarding Fisher Brook would be solved if the brook could then be ducted over the short section.  What happened, presumably, was that a new way to deal with the brook was found and the complexity of the extra section could be dropped and the risk to the railways deemed minimal.

Fisher Brook:  In the now out of print book "Blisworth", available on this website, a theory is presented for the diversion of the brook.  Starting from Fisher Close, a uniformly graded line could be found which runs north and parallel to the railways.  When arriving at the field known as The Park, it appears that the line diverts abruptly to the west following the present field boundary (there is a hollow in this hedge that resembles a water-course) and from there runs north again to join the line of a spring which discharges into the canal roughly where the portal Greek pi was located.  Knowing that this line for the brook was temporary, Barnes could have bolstered it with stakes supporting shiplap boards to ensure it did not spill into the new works.  Problems for the engineers would have been that the diverted brook must cross the railway near Greek pi and removal of spoil from the deep excavation towards the west would have been inconvenienced by both the railway and the brook - perhaps all spoil was moved up the Toll Road side instead.

At this point we recall that FPGreen and the final line seem to share the particularly tall vent shaft at B.  In the organisation of the new and final line, engineers noted this, as a saving in labour, along with the fact that Stoke Bruerne church tower would offer the perfect anchor point for the line.  All the way south from B both the church and B itself would be visible.  For a considerable distance to the north of B it would be visible and the fact that it was built very tall is suggestive in itself that it was used as a primary surveying point.  The new line puts the north portal into a position 'just a few yards to the east of the old works'.  Those old works mentioned in the minutes were not part of FPGreen but were a reference to a long abandoned FPBlue.  Perhaps the reader at this point feels there is far too much to be swallowed as a historical account.  That's reasonable - the fact is, there were two unsuccessful lines under consideration and the way in which they can be folded into history, presented here, is not intended to be definitive.

If, in the future, a lump of blue-brick brickwork should emerge from the slumping bank near Greek pi then that would make a big difference.  I take a look and have a poke every year!  My imagination has gone into a cross-sectional diagram for this area - click on the Greek pi area to see this, the yellow cross-section depicts the brickwork of the supposed portal.

Comment:  In David Blagrove's book, published in Stoke Bruerne, there is no mention of a plan to bring the tunnel out in line with the bridge near the mill.  In the National Waterways Museum at Stoke Bruerne there is a model of Blisworth Hill created in 2000 by Brian Collings, who was manager of the museum at that time.  Across the model run two tapes of differing colours, one shows the line of the tunnel and the other, representing the unsuccessful attempt is in close proximity to FPGreen . . .    I therefore claim nothing in terms of originality!

----------------------------------------------------------------------------------


Note "G5" on this map - probably another ground collapse in support of this general account.