Documents: Special Interest: Gardening In England:

Victorian Greenhouse Restoration at

Winsford Walled Garden
by Michael Gilmore
August 31, 2008

Introduction The first time I went inside a Victorian glasshouse was at Kew during a boy scout trip to London. But even the great Palm House was a ‘no contest’ for a boy who had just discovered girls and rock and roll. In later years I was to visit many stately homes and their gardens around the UK, and at each and every one the glasshouse interiors were off-limits to public visitors, - and they still are.

Then in February 1999, I accompanied my mother on a visit to a derelict and forgotten Victorian walled garden. Aileen had always been a keen gardener and asked only two questions during that visit. Are the walls in good shape? And can we do anything with the greenhouses? My response was positive on both counts. The walls were in perfect condition, not a brick or coping tile missing. At the time I wasn’t quite so definite about the greenhouses. But I recognised the wood and my bradawl failed to make any impression upon the rotten-looking base plate timbers. So I said “yes”. Aileen then advised the owner we would purchase the property, even though our home wasn’t even on the market to pay for it.

This was how I became interested in Victorian greenhouses and how I started gardening. It’s been a near-vertical learning curve ever since and I love it!

Site Description

Unlike almost every other garden I have visited, the greenhouses at Winsford Walled Garden are not located inside the walls. Even though the walled garden once contained a 200ft long Peach House, the walled garden was primarily designed and built as a summer flower garden around 1882. The adjacent vegetable garden comprised an open square encircled by a protective ring of 11 glasshouses, each designed for a specific purpose. Three boiler houses were located in the corners.

The subject of this article is the restoration of the stove or hothouse wing of a Victorian Conservatory Glasshouse built of teak. The complete structure originally extended to almost 90ft in length and is located along the southern perimeter of what was once the vegetable garden. The stove wing itself measures 28 ft long by 19 ft wide rising to 12ft at the ridge. The opposite wing, located furthest from the boiler house was the temperate wing containing an entirely different internal arrangement.

Between the two was an octagonal fern house which was removed by US Forces who used the gardens and the surrounding estate before the Normandy landings.

We arrived at Winsford at the end of June 1999, and for two months we used machetes to clear the overgrowth. During the same period we invited representatives from various National and Regional Conservation groups to inspect the greenhouses and advise us what we should do.

It was obvious to everyone who saw the stove house and the multitude of roof props inside that the following winter would be its last unless something was done to preserve it. Unfortunately, no suggestions were offered and, by the time two representatives from The Lost Gardens of Heligan arrived we were becoming apprehensive as to whether or not anything could be done to save any of the greenhouses.

The Heligan representatives provided us with the motivation we needed. By suggesting the stove house could only be restored with professional help and that such help would take two years to complete. Alternatively, they would take it off our hands. . . . We bade them farewell and the next day began clearing out all the accumulated debris of sixty years.

Restoration work begins

The stove house has a central partition, within each identical half are two brick-built growing beds on either side of an ornate cast iron central bench.

Our first task was to uncover what was there. To identify what could be retained and remove what was irretrievable. Since I had never done any restoration before, deciding which was which was not always straightforward. Questionable items were set aside until it was discovered where they went, how they worked and why they were used. When I had answers to all three I could convince myself I was making the correct decision. While I removed the soil from the growing beds, Aileen removed the rubbish in the central duct beneath the benches. The removal of the soil would prevent the site turning into a muddy quagmire when the timbers received a high-pressure wash down. Any obviously loose panes were removed from the roof for safety reasons.

Initially, the greenhouse received two washes. The first wash removed the moss, dirt and slime from the wood. The second cleaned up after the first. Within moments of the first pressure wash we knew we had struck ‘gold’. Where there was slime before, now the wood miraculously became a golden, honey brown. It was exhilarating to watch and filled us with the belief “we can do this”.

All the glass was removed after the wash down. This task became gradually more dangerous as the glass removal proceeded because the whole structure became increasingly more flexible. We were removing the buildings’ rigidity. Glass panes could easily slip between the roof glazing bars. After the glass was safely put aside, all the loose and rotten roof timbers were removed. Temporary diagonal bracing was added to improve stability.

Galvanised nails hadn’t even been invented when the glasshouse was built. The original steel nails at the end of each glazing bar in the roof had rusted, expanded, then blown off the surrounding wood. Yet just inches away from the ends of each glazing bar, the Burmese Teak was in perfect condition. The best of the glazing bars were set aside and later reused to build new glazed lids to 60ft of the original 300 ft of garden frames.

The Structural Problem

The reasons for the demise of the glasshouse were apparent from the start. One previous owner had, at some point, decided they didn’t want to grow anything beneath the roof glazing and removed the supporting steel and wire framework. Unfortunately, the steel was also the bracing for the roof.

Later, the roof glazing must have deteriorated to such an extent as to compel one owner to replace it. However, instead of cutting the glass to fit the 12 inch gap between each glazing bar, the owner had accelerated his repairs by not cutting the 2ft wide replacement glass sheets. Consequently, every alternate rafter was removed. From which moment, the glazed roof was doomed.

The wider uncut replacement glass added about 5% to the total weight of the roof. Only now there was no bracing and half the structural glazing bars had been removed. There was only thing the roof could do. It slumped. The ridge dropped slightly and the roof weight pushed out the walls. This caused maximum stress at the corners and they all ‘burst.’ We arrived shortly after.


To this day, I have never seen a greenhouse with a higher degree of ornamentation and ingenuity than the Foster & Pearson greenhouses at Winsford Walled Garden. Some of the original metalwork had been due to impact damage. But we were fortunate in that we were able to unearth and recycle replacement items from the garden. Some cast iron components used in ventilating the glasshouses had rusted solid. Where possible, these were removed from the roof and a local farrier heated the pieces in a portable furnace until their eventual release.

With all the glass gone and most of the roof missing it was a much safer place to work. At this point, Aileen and my wife Tanya rejoined my efforts, and for over a week the three of us cleaned all the cast ironwork back to the bare metal. We worked from daybreak to sunset, spurred on by the task at hand, and by the threat of the approaching winter upon our stock of tender plants that would surely perish at the first frost.

Then, while the ladies began painting the metalwork I repaired the brickwork to both the corners of the greenhouse and to the growing beds. Later rejoining the ladies to complete the painting of the metalwork in the roof ridge. Each corner of the teak base plate was secured with a nut and bolt running diagonally across the corner. But only the head of the bolt was visible on the one side. Where was the nut? By removing a small teak cover underneath, the nut was revealed.

Another example of the manufacturers’ attention to detail were the zinc inserts set at an angle into the top of each door frame (there are four doors). These prevent condensation dripping beneath the frame and remaining as a pool of water until the door is opened.

A word about Teak and the restoration By the time we were ready to rebuild the roof I had discovered a few things about Teak wood.

Foster & Pearson catalogues show 40ft high piles of Burmese teak that were left to season outside for at least two years. But the high price of teak garden furniture today ensured that we would not even entertain the idea of replacing the roof in teak. Instead, we chose a local softwood that could be treated under vacuum pressure, a process that was guaranteed for at least twenty years. Thus far, after seven years, with no sign of rot I have every confidence that it will last even longer. Furthermore, the treated softwood has now weathered to a colour very similar to the adjacent teak.

Mature and weathered teak is greyish in colour, and looks rather like old concrete. Certainly not the warm golden honey colour we are all used to seeing on sale. That look is due to the use of teak oil. Teak, the natural slow-growing teak that takes hundreds of years to mature, which was used by Foster & Pearson, owes its fantastic water resistance to its cellular structure. As such, the application of teak oil is insignificant. When old teak is planed back to the bare wood the warm golden brown colour reappears, but it will disappear after a few weeks outside. Teak oil is used to help maintain the natural colour of the wood.

The first time I glued softwood to teak, I used a water resistant polyvinyl acetate (PVA) and clamped both pieces together overnight. The next morning I released the clamps and the wood slid apart and fell to the floor! A superior wood glue, ‘Cascamite’ was then used, in exactly the same way, with exactly the same result. I then called a boat yard, one that uses teak timber on the decks, and they gave me the name of the two-part adhesive they used, consisting of a powder and a syrup that was like melted chocolate. An eye-boggling cheque and a week later I re-clamped both pieces together and they were bonded solid. The new glue set like glass.

Ingenious Heating Design Running the length of the greenhouse is a central duct that originally contained the 4 inch diameter primary heating pipe work, which exited the stove house and ran through to the other sections of the Conservatory Greenhouse.

Beneath both central benches lying across the central duct are two galvanised steel water tanks of 150 imp galls capacity each, which were, and still are, used to irrigate and humidify the greenhouse. The tanks straddled the duct in order to benefit from the rising heat off the primary pipe work. In this way, heated plants were never chilled by cold water applied to their roots or foliage. The heated duct also contained the irrigation water supply pipe. As a result, the irrigation pipe never froze and always brought warm water to the greenhouse.

The benches themselves were covered in thick slate which retained a tremendous amount of heat to warm the terracotta pots resting upon them. A slate bottom at the base of each growing bed hides the secondary pipe work that once heated the compost. A valve on each side of the greenhouse meant that one pair of growing beds could have a different soil temperature from their neighbour opposite. Coupled with the partition and individual venting this made for a very flexible growing environment. An important by-product of the heated beds was their slow release of stored heat to the surrounding air. This battery-like behaviour helped to smooth out any thermal inclines. In the event of boiler failure they could be the difference between the survival of a rare and expensive collection and its complete ruin.

A third heating circuit used to rise up through the floor beside the growing beds and ran along the front of each. Its convected heat was designed to remove any roof condensation and the resulting damage to foliage that dripping condensation can bring. The removal of roof condensation also led to an improvement in light transmittance, especially during the winter. The pipe work continued along the floor towards the nearest end wall before rising up the side and along the rear of each growing bed. Effectively encircling the bed with a warm air curtain that could be adjusted to suit the foliage growing in each bed. This curtain also prevented condensation forming on the glasshouse walls.

On cold winter days when the windows and ridge vents were closed and the risk of botrytis high, cold air was still permitted to enter the hot house through a series of vents along the side walls. This cold air passes over the third heating circuit pipe work running along the outer walls and the fresh air is preheated before it reaches the warm plants.

Replacing the ridge ventilator The greenhouse ridge had been covered in sheet steel and secured with 4” nails. Beneath this sacrilegious eyesore were found some fantastic examples of the Victorian engineers’ art in cast iron. The greenhouse ridge was not only beautiful, it could be raised vertically to exhaust hot air at the hottest point, along the full length of the greenhouse.

After the sheet steel was removed I was fearful of dismantling the whole ridge structure at once and then being unable to put it back on my own. The ventilators themselves consist of a number of cast pieces which have to be set at the correct distance apart along the ridge and I was not certain I could do it working alone.

So, the existing timbers were measured up and the wood ordered. Upon its arrival the ridge ventilators were completely rebuilt by removing one piece at a time and replacing it with the new. Where there was sheet steel now there was glass. The ornate windless and worm drive used to raise each ventilator had seized solid with rust. Unable to remove them to heat them, they each had to be gradually eased free with WD40, cleaning them and then applying two coats of smooth black Hammerite paint. It took about thirty-six man-hours before both mechanisms looked like new.

Replacing the timber roof

When condensed water falls from roof glazing and drops on valuable plants, the resulting droplets act like miniature magnifying glasses, that can often lead to unsightly burning on the foliage. Foster & Pearson designed a new rafter with an integral drip which carried the condensed moisture down to the eaves and outside. The cross section of these rafters is quite complex for any wood machinist to reproduce. Which meant ‘prohibitively expensive’ from our point of view. Thus the replacement rafters are a simple design based on the original overall dimensions but without the integral drip. This time the rafters were secured with galvanised nails.


Approximately 95% of the glass on the stove house was recycled from the site. But every piece had to be re-cut. It was filthy and at the time a bitterly cold wind blew from the north, the greenhouse (then) was very exposed. Aileen volunteered to wash the glass but only if it could fit in the kitchen sink where it was warm. Previously, we had been advised that ‘you can’t cut dirty glass’. An expensive oil-filled glass cutter was purchased.

The rafters were braced laterally. Yet despite this, a brief survey showed the width of every pane would be slightly different. A practical method was devised in which every vertical run was measured up and the panes cut accordingly. Several runs were cut at a time and then handed over to Aileen for cleaning. Once rinsed and dried, the panes were stacked beneath the rafters where they would be installed. Care was needed to ensure the order of the panes did not change.

Six hundred panes were needed for each restoration and there have only been eleven breakages. The glasscutter tool is nothing short of amazing. We began our restoration on the first day of September, 1999. The tender plants went undercover in the first completed partition on November 4, 1999. The first frost arrived two days later. The restoration was finally completed on November 15, 1999.

Restoration purists might argue against our use of simplified rafters and silicone sealer. But I would argue that such methods have enabled the preservation of a fantastic greenhouse that would otherwise be lost. Also, because it was teak, this greenhouse was never painted. It does not need painting. Yet modern glazing putty needs to be painted otherwise it breaks down under direct UV light. As a result of our 3-month effort, garden visitors today can wander inside the magnificent restored greenhouses and witness Victorian horticultural engineering at its very best for themselves. A polite request allows visitors to work the windows and vents themselves.

please visit this wonderful site at http//

  • New Eden
  • Kids Garden
  • Plant a Row Grow a Row