Cloverfields as of August 2020: An Extraordinary Find In The Eighteenth-Century Roof
/An Extraordinary Find!
In this newsletter, we learn about how the eighteenth-century roofers working at Cloverfields “swept the valley” and “combed the ridge” of the shingles they installed—and about how sometimes they did not, and instead, they used very thin shingles. These slender shingles and the way they were manipulated and installed constitute, according to historian Willie Graham, an “extraordinary find.”
In the video above, Graham starts by explaining to us how the roofing technique called “sweeping the valley” works:
So, this roof has valleys where the dormers meet up with their various roof slopes. And traditionally, they didn’t flash those valleys; what they would do is called “sweeping the valley,” and the shingles would fan as they went around the valleys, and they would cut the shingles in almost pie shapes to fit that sweeping. They put a little cant strip underneath to help ease the shingles around the corners, and it makes for a beautiful roof. It was done in the period I think, less for its looks and more as a way to avoid having to put flashing in.
There were a couple of roofs in the eighteenth century that had leaded valleys, but they were very rare. This was really the common way of doing it; it was to sweep them. To recreate that sweeping today takes a lot of practice, experimentation, and we had to do that here to get them to look just right. I think the roofer did a really fabulous job working out that detail with us.
We agree! The roofer that worked with Lynbrook of Annapolis did a wonderful job sweeping the valley. Graham then explains what “combing the ridge” means:
You’d like to think that each row would have the same radius in order to create that fanning, but in reality, that radius changes slightly from one row to the next in order to make it work out.
And then the real complicated thing is that, as you get to the top, the shingles are coming around to form the ridge at the top of the dormers. The shingles on one side need to extend past the shingles on the other side; it’s called ”combing the ridge.” Instead of putting a cap on it like you often see on shingle roofs today, they simply extend the shingles on one side a couple of inches beyond the shingles on the opposite roof. But all that’s got to come together as you are still sweeping into it, and it’s got to flatten out in the main roof. That is a very tricky thing to do.
It is tricky to do, and it may result in water damage. Because of this potential damage the Cloverfields’ restoration team decided not to imitate this particular eighteenth-century technique. As Graham explains:
We did cheat and put flashing underneath the shingles. We flashed actually between each course as we were doing the sweeping. You see none of this, but just in case something leaks, the flash between each course and the valley underneath is flashed all as a safety measure. But otherwise, we really wanted to limit how much flashing there is on the roof because you want those shingles to breathe from underneath.
Graham then turns our attention to the section of the roof that in the eighteenth century connected the stair tower in the rear of the house with the roof of the main house. It is here where we see “an extraordinary find”.
The extraordinary find here is where the roof of what had been over the stair tower, this kind of rear wing to the main house, how that roof was treated as it fitted into the back of the main roof of the house.
And so we know there is a clapboard base that was put on everything. The main roof gets a clapboard base, the wing gets a clapboard base, and where the two meet, they put clapboards that they bent to fit the valleys. Almost like the sweeping of the valleys with their shingles, they’ve done them with horizontal boards that are bent around. And to think about it, they’re bending these things that are going to be tough to bend anyway; they’re set on, you know, lapping overtop of each other just like weatherboards. They are doing all of that, and they are trying to make it watertight.
And my guess when we were looking at this roof from underneath was that the shingles followed that curve, and it swept the valley. But in fact, we discovered that did not happen, that there are no nail holes in the valley for the shingles to go through, so those valleys were exposed, and they simply tucked the ends of those clapboards that are being bent around the corner underneath the clapboards that are coming from the main roof.
In fact, as these clapboards, as I said, are 4 or 5 feet long, and at the ends of them, they feathered each along the length, they feathered the ends of them together to create a seal. One nail goes through two boards that are kind of shaved down and fitted together.
This roofing technique is one of the most unusual architectural features of the historic house:
I think in some ways this is one of the most unusual features that we have seen in this house. It’s been full of all these great surprises; it’s the part of the building that just never survives, and it is also particularly early, and you combine those two things to be able to see how craftsmen in the period resolved this.
Graham then observes how thin the original boards were:
To create those curved boards, the first thing they do is they split the boards very thinly. The thickest board I could find was just shy of 3/8 of an inch thick, so they’re very thin boards going around there, which helped them bend them, but you’d think it wouldn’t give them much life but yet they worked for at least 50 years, so you can’t argue with it; I mean that is something that was successful.
So they fit them and bend them there, and the shingles on the main roof came up and stopped short, 18 inches or something, of that valley. At least on the main roof, we don’t know if the shingles continued on over top of the stair tower roof, just because nothing survives back there, but presumably, they continue once you got beyond the valley. They were doing everything they could to make that area watertight without adding any kind of flashing or tar, or any other kind of product up there to facilitate it.
Although unusual, the roofing technique was not unique to Cloverfields. Graham points out how we can see something similar going on at “Melwood,” a 1714 Southern Maryland house:
You might think that this was an idiosyncratic case where one person figured it out, did it this way, the next guy is doing it another way. However, since we discovered it here, there is a house in Southern Maryland, a place called “Melwood” that dated from 1714, and it’s missing its rear wing just like we are here. When you go up on the roof, you can see the tell-tale signs where it had clapboards that were bent in the valley like this one, and here we are today an- hour-and-a-half drive away, and somebody solving that problem in a very similar fashion. So I think it must have been understood by carpenters in the period that this is a way to treat competent roofs at the turn of the 18th century.
Graham appreciates the originality of the roofing technique:
I think in some ways this is one of the most unusual features that we have seen in this house. It’s been full of all these great surprises; it’s the part of the building that just never survives, and it is also particularly early, and you combine those two things to be able to see how craftsmen in the period resolved this.
A roofing technique used at the Cloverfields house in Maryland’s Eastern Shore in the year 1704 was used ten years later in Southern Maryland, at Mellwood. Once again, then, Cloverfields helps us to write the architectural history of the Chesapeake Bay region; this time Cloverfields adds a section about how eighteenth-century roofers experimented with different techniques to waterproof the houses they worked on.
By: Devin S. Kimmel, of Kimmel Studio Architects
For: Cloverfields Preservation Foundation