History of the Caisson Lock On the Somersetshire Coal Canal

Please note: This is a very exciting topic. This caisson was "one of the wonders of the waterway world". The exact location and structure are still a mystery, but new evidence is constantly coming to light.

If you wish to read pages on its possible locations then please visit possible locations site

For visitors who have navigated directly to this site, may I suggest that you look first at the various diagrams and potted history of the site first? Please go to: History of the SCC

(Diagram on right is from Clew )

The original idea was to cut a 3/4 mile tunnel at Combe Hay to link the 2 sections of the canal. The decision to place the canal at a higher level, so as to avoid a long tunnel, was probably also what made it possible to concentrate the whole drop at Rowley Bottom (at Combe Hay) and that, in turn, made caissons a viable proposition.

Robert Weldon (b:?1754 to d:1810) from Sunderland had a demo. half-sized model of the caisson lock on the Ketley Canal (near Oaken Gates) in working order in Sept. 1792. (The patent is believed to date from June 1792).This obviously impressed the SCC committee. However it seems likely that, being a native of Lichfield, he came across the writings of Charles Darwins' famous Grandfather - Erasmus - another native of this town. Erasmus published, in 1777, details of a wooden box, capable of receiving a boat which could be raised or lowered between the upper and lower levels of canals (Not in a water-filled well though).

The expression "Caissoon" as seen in a comprehensive dictionary means "a large water tight chest", and the expression was first used in engineering circles as early as 1704.The Caisson became a subject of interest to researchers in the Telford area when we read the details of a letter by John Ward written from Leeds to Thomas Gilbert of Cotton Hall,.Staffordshire which reads " I saw Mr Thomas Gilbert at Cotton on Tuesday night. He was very earnest in recommending it to our committee to send to John Rennie (Civil Engineer:1761-1821) into Shropshire to see a

"cassoon"(sic) which is calculated to raise and sink boats from one level to another without waste of water.Weldon the patentee and the cassoon is at Wrockwardine, near Lilleshall in Shropshire. If it would save a tunnel and a steam engine at Crofton or Wolfhall I should be very glad For I do not know but a steam engine (which will be necessary if a tunnel is not adopted) Will be a greater nuisance than a tunnel and deep cutting on account of the smoke It will send up ".".

The caisson was described as follows by the Bath Herald for 27 June 1795, as follows:

"This ingenious contrivance consists of a wood cistern, having two square apertures with a slide door to each, at the respective levels of the upper and lower Canals; In this reservoir (being always full of water) is immersed a Caisson, or hollow vessel; having a door water-tight at each end, for the purpose of receiving and enclosing a boat; this Caisson is ballasted, to be specifically of the same gravity as water, consequently will descend or ascend in its surrounding medium (with or without a load) with the greatest facility. Each end of the Caisson, when brought in contact with the square apertures of the reservoir, is also water- tight; this is accomplished by an inverted valve, discharging the water from the space between the slide-door of the reservoir and the door of the Caisson, the pressure instantly then fixes the Caisson to the apertures, therefore the slide- door may be opened with much ease, and of course that of the Caisson. The boat may then be received for ascension, the doors re-shut, and the great difficulty of releasing the Caisson from the aperture (against which it was so power- fully pressed) obviated, by opening another valve to refill the space just mentioned; it consequently follows, the pressure on the Caisson will be as before, equal on all its sides - by discharging a small quantity of water from the Caisson by means of a cock into the lower Canal, it becomes lighter, and instantly ascends to the upper aperture; the same process with the valves completes the design which is truly philosophical" .

From "The Daily Telegraph"


From "The Daily Telegraph"


Adrian of the SCC has now completed an excellent series of diagrams to show how it works.


See Adrian's Caisson "how it works"


In Jan.1796 adverts appeared in the local (Bath) press for masons to build a caisson lock at Combe Hay.

Dimensions were: Chamber: 66 ft deep;width 20ft in middle, tapering to 11ft 6in at each end. Length 88ft; Lift 46ft. Brick lined(Clew). However Torrens states that it was built with 'freestone' (Local inferior oolitic limestone; this is probably correct).

Long-boat turn-around-time: about 7 mins.


1st trial Feb. 1798. partly broke ... a failure.

2nd trial June 1798 worked.

3rd trial April 1799 worked.

4th trial 12 days later a spectacular and successful demo. in front of HRH The Prince of Wales!

5th trial 10 days later, over 60 ladies and gentlemen took "rides" in the caisson.

6th trial May 1799. Bulging brick/stone-work caused the wooden caisson to stick.

Another £1 000 was more than the SCC could justify as the work had cost £4 582 so far. (The original estimate was for £ 1 200; Outram later estimated that it would cost £16 000, although whether this was for one or the complete set of caissons is not clear). Outram recommended that the caisson should be replaced by an inclined plane. As already stated, the probable reason for this was that the whole of the cistern (well) had been cut in the Fullers Earth Measures - a rock formation that swells like cat litter when wet. William Smith from Tuckingmill (on the canal near Midford) the "father of English Geology", surely must have known of this problem, especially as he had visited other areas of England such as the Sapperton Tunnel in 1798 and 1794 which had exactly the same problems and for exactly the same reasons. It may be that this was the reason for him eventually being dismissed from the SCC on the 5th of June 1799 (especially as he had sided, according to a Mrs Eyres - (quoted by Torrens), with the anti-caisson lobby). He should have spoken up sooner! He had also 'taken advantage' of the SCC in building his house on the banks of the SCC at Tuckingmill where, ironically, there later was a Fullers Earth Mill. (The house that stands here has a plaque to Smith but this is the wrong house! (It is the next house))

Whilst there, Smith planned to open up a Bath Stone quarry higher up the hill side in Combe Down and bring the stone down to Tuckingmill (for loading onto the Canal) by a tramroad/inclined plane. For more detail see Combe Down Tramway

Other candidates for Weldon's failure include the poor Tarras cement-mortar that was used. However this being partly based on the Bath Blue Lias limestones which had an excellent reputation, not only in use in canals but mines as well, would appear to contradict this notion.


Jane Austen and her father visited the site in 1801, so obviously it had not yet been filled-in.


The Rev. Warner also visited the site in 1801 ..."but the bason (sic) of which alone remains".


There were plans for another 2-3 such caissons. Certainly the Rev. Warner talks of the building of another caisson lock being suspended, so obviously at least 2 were (partly) built, one (perhaps) marked by a former flag pole, the other 180 feet away & (perhaps) marked by a chestnut tree atop an artificial mound. These 2 markers are important clues it is thought in the location of the positions of the 2 caissons. (See later).

"...(the caisson) may then be easily raised or lowered at pleasure by means of racks and pinions, or chains and pulleys ... from one level to the other...." (lines 12-14 from the patent). This is at variance with the account from Clew, where in the passage describing label 'A', more emphasis appears to be placed on increasing the mass of the caisson by admitting water to the caisson, in order to lower it.


The last 3 lines of the patent do however leave room for the above method as a subsidiary method of (fine) control, ... "...and the machine is further regulated by means of a pump, an air pipe, and other apparatus, to be applied as occasion may require". (My emphasis). I have not encountered details before of such a "pump". (See later).

However 6 years previous to this (a long time methinks?) the same Canal Company had... "... launched a competition for the 'best means of raising and lowering heavy weights from one navigation to another" ".

The contest was won by Henry Williams and James Loudon, and by 1792 tub-boats were being hauled on rails up and down the steep slope, by teams of men and horses. This was the famous "Hay Inclined Plane". Boats entered and left the water, carried on wheeled cradles. By the year end a steam engine at the top of the incline had been installed, allowing full operation of the incline by a four-man team. It is 305m (333 yd) long and has a vertical rise of over 61m (66 yd) equivalent to twenty-seven canal locks. A pair of five-ton tub-boats could be passed in four minutes whereas it would have taken at least three hours to raise a vessel through conventional locks. The inclined plane worked for just over a century, its last recorded operation was in 1894, and its formal closure in 1907. "


It may thus seem strange that the proprietors of the SCC came to favour Weldon's relatively unproven invention, and, even after the failure of Weldon's caisson, approved another version of the inclined plane that obviously was not half as successful as the Hay version! In these next paragraphs it should be noted that IE3 has problems reading the French! Netscape/Safari/ Firefox appearsmuch better!

A Frenchman called Dutens also wrote extensively on the caissons etc in England. This source may well have been overlooked by previous historians. (Mike Clarke reports that the pages of the book had not been cut!). Dutens quoted extensively from both Reess (sic) and Chapman. (Full references are in Clew (ibid)). However it is obvious from his writings that he "visited the canal near Bath" .

In a passage he appears to have written himself, and it has to be admitted that it is somewhat ambiguous, Dutens might be suggesting that the caisson had two doors (at each end): ..."les deux portes du caisson et du puits étaient ouvertes pour laisser passer le bateau". (The 2 doors of the caisson and pit (or well) were opened to let the boat through).


However it could be read that the caisson and exit doorway on the side of the pit each had one door - making the pair. Probably this is what he meant...


Also he states that the building of a second caisson was suspended and that the mechanism for the raising and lowering this caisson was much simpler (than in the first). "Quant à la seconde écluse à flotteur qui est restée sans exécution, bien que le mécanisme en dut paraître d'une application beaucoup plus facile." (As for the second floating lock this has remained without use, although the mechanism of it would appear to be a much simpler implementation).


As a footnote Dutens relates that a French Engineer (M. Sganzin) observed that the first caisson mechanism at Combe Hay was soon modified and was thus very similar to another planned caisson lock mechanism for the canal at Ruabon (near Wrexham in Wales). (My emphasis). ..."(il) fait observer qu'il fut modifé et corrigé peu de temps après son exécution. La modification qu'indique M. Sganzin, semble rapprocher beaucoup ce système de celui de l'écluse de Ruabon,..." There are three possible interpretations of these passages:


(1) the first lock used the rack and pinion gear to raise and lower the caisson whilst the second made more use of adjusting the mass of the caisson by water.

(2) Dutens is describing another, cradle-caisson lock, patented by Rowland & Pickering, in another part of the country.

(3) The second (Combe Hay) lock was different from the first and was much more like that of Rowland & Pickering's design. This is most unlikely. However see later!.


Just how Weldon's device was to be like that of the said Rowland & Pickerings' is not made clear. It is evident from Dutens that he believed that their design was superior to that of Weldon's, and would almost certainly have worked. I, too, believe it was superior. However, accepting the 'classic' reason for the demise of Weldon's lock - that of building it in Fullers Earth, - R&P's design would have foundered for the same reason!

"As to the second floating lock which was not constructed, even though its mechanism must have seemed much easier, I shall content myself with reporting from Reess (sic) (who states that) Messrs Rowland and Pickering took out a patent on the 18th of March 1794 for the invention of a barrel vault- or cradle- mobile lock, which, supported by props borne by a large plunger (diver), was able to be raised or lowered, so as to connect, according to need, with the level of the upper or lower reach whenever boats are ready to enter the airlock". (Translated from the French). (The boat entered an open-top "caisson" with doors at each end. The enormous mass was counterbalanced by a giant watertight box in water. There were struts that connected the two together (the "caisson" or cradle being directly atop of the watertight box ("plunger or diver")). The cradle, when it descended, did so in an air-filled shaft. There were clever devices (spirals of weights and/or weights in conical barrels) to counter the apparent loss in mass as the struts sank into the water.)


Other interesting snippets from the few pages I have are: William Smith (not unreasonably) visited the Combe Hay site. (Torrens dates this as 1798 and 9) Dutens states that Reess (sic) gives more detail on ..."which pumps are used by the boatmen who are shut away in the caisson aboard their boat in order to empty or to bring in the necessary volume of water in order to lighten or load the boat; and how the air pipes work so as to prevent accidents which would result from a lack of air in a situation where some circumstances would hold the caisson and the men below the water for too long." (translated form the French). (One wonders whether the caisson was thus served by a suitable flexible (leather?) pipe from its roof to the atmosphere on the higher level). No such mention is made of this sensible precaution in Weldon's patent.



On reflection, it seems a little odd for the boaman to be in the caisson at all!. It would have been potentially dangerous and scary; there was no need to be there, especially if he was working alone and needed to walk the horse to the other level.

Dutens observed the inclined plane.

Dutens states not just that the walls of the chamber bulged but that they collapsed, and that this had led to the total abandonment of the workings ..." even though they had got the diver/plunger to work!" (My emphasis). The plunger/diver was a term reserved for R&P's airtight box. Surely, if it had been Weldon's design, he would have used the word "caisson". Again this reinforces the possiblity that the first caisson lock of Weldon's was modified to accommodate refinements from Rowland and Pickering's design. Very interesting, but hard to believe, especially as, if fully implemented, it would have necessitated the well to be twice as deep! Perhaps only minor alterations were made to Weldon's first design and that this explains the forementioned differences in the accounts of how the effective mass of the caisson was changed to raise or lower it. (Pulleys/weights/ water).

If any more information comes to light I would be pleased to pass it on.

Adrian has also done an excellent History of Rowley Bottom and the Combe Hay Locks (with other links) .


Torrens. H.(1975) "The Somersetshire Coal Canal Caisson Lock". (BIAS)
I am indebted to Simon Jones for a copy of this document.
Clew. K.R. "The Somersetshire Coal Canal and Railways".2nd ed. Pub.: Head. 1982.

Dutens. "Notice sur les Elevators".
Sorry no more info. on this important and, perhaps, previously untapped, source. I have asked for more detail.
A. Tuddenham ( countless e'mails!)

I am indebted to Jackie Dixon for the translation from the French.

If anyone can assist with re-reading the French to ensure the translation is accurate please contact me.