Whitehouses Pumping Station

The Wendover Arm of the Grand Union Canal was built in 1799 to link springs in the Chiltern Hills in the town of Wendover to the Tring summit pound of the Grand Union Canal, just over 6.7 miles to the east. The water from the Wendover springs is critical to the operation of the canal as it drops south towards the Thames valley in London, and north towards the Ouse valley in what is now Milton Keynes.

The Wendover Arm follows the 120m / 390ft contour of the Chiltern north-facing escarpment, and at Wilstone around 2 miles before the junction with the main line at Bulbourne, a reservoir was built on the Aylesbury Vale plane 20m below. The reservoir took excess water from the Wendover Arm via a weir, shaft and adit at the Whitehouses pumping station, which was built in 1802 to house a Newcomen atmospheric beam engine to pump water back up to the Arm from the reservoir to supplement water flowing from the springs at times of low rain fall and high demand.

By 1815, three additional reservoirs had been built at Marsworth, with a larger more efficient Boulton and Watt beam engine installed at the new Tringford pumping station, somewhat closer to Bulbourne. A second beam engine was installed at Tringford in 1838, at which point the original engine at Whitehouses was decommissioned, and the pumping station converted to a cottage.

See this OS Map of the Marsworth reservoir complex, and satellite view:

Detail from map at Tringford Pumping Station

Unfortunately due to the chalk geology at times of high rainfall, many small springs would puncture holes in the clay lining of the canal. Then at times of low rainfall, the Arm suffered catastrophic leakage through the damaged lining and into the chalk substrate. Repairs to the clay lining were carried out as early as 1802, and in 1857/8 the worst section was relined in asphalt. However, these repairs were unsuccessful and in the very dry summer of 1897 more water was leaking from the Arm than was being supplied to the summit pound by the springs. A stop lock was installed at Tringford pumping station to isolate the leaking section, but by 1904 the Arm from Tringford onwards was abandoned to navigation. The far 3.5-mile section from Wendover to Drayton Beauchamp was repuddled and the water level lowered to reduce leakage, and a pipeline was installed under the worst 1.5-mile section from Drayton Beauchamp to Little Tring, terminating at Tringford pumping station, with excess diverted to the Marsworth reservoirs. At times of high flow, further water could be diverted from the pipeline to the lower Wilstone reservoir via the Whitehouses pump shaft.

The cottage at Whitehouses was maintained and occupied right through to the Second World War when it is believed refugees from Europe were housed there. The cottage was finally demolished in 1945 as the site had no running water, sanitation or road access.

The Wendover Arm Trust was formed in 1989 with the aim of restoring the Arm to navigation. The 100-year-old pipeline is now badly leaking, with all the water diverted to Wilstone reservoir, and the pumps at Little Tring require constant maintenance and use a huge amount of electricity to keep the summit level in water. The eventual aim of rewatering the arm would allow the spring water to once again flow freely to the summit pound, massively improving the water supply to the Grand Union canal – and instantly saving the Canal and River Trust a large proportion of the cost and operational risk of pumping at Little Tring.

Through the 1990s and early 2000s, the Trust’s restoration work was concentrated on scrub clearance along the dry section of the canal, and on the rebuilding of Little Tring bridge and reopening the first 100 yards or so of the dry section, including providing a winding hole to allow full-length boats to navigate to the end of the arm. Kescrg were involved in this phase of the restoration and built a relationship with the trust while helping with the construction of the concrete retaining walls that lined this section of the canal.

Around 2007 the trust volunteers then moved onto their long term aim of relining the dry section of the Arm with modern bentonite clay impregnated carpet matting. Kescrg again were involved from time to time with this work, helping with one of the early sections of lining, the footbridge installations, and with building some of the mooring layby sections, where a concrete retaining wall allows boats to moor close-in to the bank of the eventual channel.

Shuttering for Mooring Layby, 2008. Photo Bobby Silverwood


During our May dig weekend in 2010, we were asked to investigate an old retaining wall hidden in undergrowth, which was believed to be related to the Whitehouses pumping station – though no details of the layout or nature of the site were known at that time.

The front wall had three arches, but it was not known where these led. we were tasked with finding the ends of the arches, and to start to establish the layout of the pumping station. By the end of the weekend, we had located the other end of the culverts, which appeared to lead to some kind of tank structure.


By 2011, the Wendover Arm team were still a couple of years away from getting their relining work from Drayton Beauchamp to the Whitehouses site – but by the time they reached Whitehouses, the outfall wall had to be restored so it could be incorporated into the completed relining works. The Trust volunteers were working at capacity on the relining, so they asked us to take on the excavation, investigation and restoration of the pumping station structures; so we put together a programme of dig weekends and summer camps to work on the Whitehouses site.

We returned to Whitehouses for our 2011 May weekend to discover as much as we could about the site, so plans for its restoration could be made. The first job was to clear the site of undergrowth – being a previously cultivated garden, it was home to swathes of very vigorous nettles. We then proceeded to continue with the excavation behind the wall we had uncovered in 2010. As we chased the walls around, we found that this appeared to be a square structure, with a brick floor. Once the extent of the structure had been carefully determined, the bulk of the material was removed by excavator, which started to reveal what was believed to be the outfall tank for the pumping station.

Halfway across the tank floor, there was a step and what appeared to be a weir crest, with an adit exiting the corner of the tank linking up to the original pump shaft. It is likely that the structure was rearranged after the pump was removed in 1838 to provide a static level-control overflow weir back down the shaft to the reservoir.

We left the site with the tank three-quarters clear, so the Canal and River Trust heritage officer could survey the site and agree on the next steps with the Trust.


In 2012 we again returned in May, to prepare the site for the initial restoration of the front wall and tank structure over the course of a WRG Summer camp, which we would be leading.

Having again cleared site of undergrowth, the excavator was used to grub out a large number of small trees and saplings to make a large neat compound for the summer camp and steps were cut into the canal bank on either side to create a safe access to the site for the camp.

Work continued to clear the rest of the outfall tank and prepare the brickwork for repointing and reconstruction. An interesting date stone was discovered for 1865. This does not directly correlate with a known date for the site, as the pumping engine was removed around 1838, though it may indicate the tank and associated structures were rebuilt or remodelled around this time.

A collapsed and infilled vaulted cellar was found running parallel to the tank, and the stairwell was excavated to investigate the structure. Parallel again to this first cellar, a tiled floor was uncovered, and next to this a hole in the ground was found to lead to a second vaulted cellar, which though also partially filled with rubble appeared to be in relatively good structural condition. It was conjectured that the first cellar lay under the original steam engine, possibly as the coal store or housing the boiler, and the tiled floor and second cellar were the footprint for the pumping station accommodation.

This diagram, part of the project plan for the summer camp, shows the layout of the pumping station site.

The camp was a good mix of experienced and first-time volunteers from across the country and made great progress removing blown brickwork from within the tank, and rebuilding the arches and front wall of the outflow structure.


Returning again for our May weekend in 2013, we carried out more preparation for our second summer camp. This included starting to lay out the path which would lead from the footpath and footbridge over the canal to the pumping station site, as once the restoration is complete the area will be opened up for public access as a nature trail and site of historic interest on the canal. We also spent half a day helping the local team with laying protective blocks on the bentomat lining – these prevent the clay from swelling when the canal is rewatered and also protect the matting from damage if boats come too close to the side. This was very heavy and hard work, and rests were well earned while waiting for the next dumper delivery.

In July we then returned for our second WRG Summer camp. The nature trail pathway was extended, laid with a weed barrier membrane then finished with scalpings. The work reinstating brickwork in the outfall tank continued, and the front wall was rebuilt, including rebuilding the arches and the paddle recess. This paddle set into the wall opened a sluice directly through to the pump shaft. This was presumably used either cracked open to control the water level in the summit pound, or to drain the Arm for maintenance if required. The paddle gear had been relocated to the middle of the canal bed when the arm was abandoned – repurposed to control the flow in the pipeline and divert the water away from the Tringford pump and down the old pump shaft to Wilstone reservoir as required.

Having brought the front wall up to coping level on the summer camp, we returned for an August weekend to set the coping stones with the help of the hydraulic grab excavator attachment.

The key stone to set in place was over the paddle recess. This stone has an inset in the top for the paddle gear and a hole for the paddle shaft which had to be lined up precisely with the slot in the brickwork. Once this stone was in place the one to the right could be set, and then the rest of the wall one by one. The joints between the stones were then pointed up. By the end of the weekend, the front wall was complete and looking very smart. More progress had been made inside the tank, where a design for a new weir crest had been agreed with CRT to eventually allow them to set the level of the pound with overflow to Wilstone reservoir via the pump shaft.


With the front wall complete, the goal of getting the structure to the point where the lining could marry up to the outfall was achieved. The trust had had some setbacks with the progress of their lining due to both weather conditions and ground conditions encountered over the preceding 3 years. However, getting the structure to the point where it was ready for the Trust was a great achievement.

There was still work to do inside the outfall tank to get the structure ready for rewatering. CRT had provided a design for a weir crest across the centre of the tank, with accommodation for weir boards to be added or removed to control the eventual water level. Once this was complete, CRT would take over the project as they had plans for further improvements to the water control features, and would finish off the tank walls and install a grill over the top to allow the public to view and interpret the water control features. So we returned for our April and May weekend digs to complete the rebuild of the weir and invert in the tank.

Contractor Works

With the weir pillars in place, our work on the structure was complete. At this point, CRT took over the site, and in late 2015 contractors came in to reconfigure the water control features, whereby the wall sluice and overflow weir both empty into a new chamber, and this chamber in turn empties into the pump shaft. This allows the flow to be controlled and gauged with equipment to modern standards.


We made one final visit to Whitehouses in May 2016, to further investigate the second cellar as a prelude to a survey to be carried out by the local archaeological society. A stairwell to the cellar had been discovered at the very back of the site, and our task was to clear the cellar and make the structure safe for the survey, so a final plan could be made with CRT to either preserve or fill in the cellar with inert material. As can be seen from the first photo, the hole in the ground couldn’t safely be left how it was.

It was a great honour to be entrusted with the Whitehouses project and to do the work on behalf of CRT and the Wendover Arm Trust. In particular, this wouldn’t have been possible without the support of Ray Orth and Roger Leishman from the Wendover Arm Trust. We continue to support and watch the progress on the Wendover Arm with great interest, and will return as project and work opportunities arise.