Room 101 (The Pink Parlor), Wall D, Photo by John Butler
On Friday, Scott finished removing the trim from the Pink Parlor, pictured above. I had eagerly anticipated the joinery surrounding the fireplace, given our recent work on another fireplace surround. The displaced surround, turned upside-down, is below:
Pink Parlor Surround Turned Upside Down
When I think about the era in which this house was built, in a relatively new country, with newly earned independence, it can feel very foreign to me. I struggle to understand the mindset of these post-colonial carpenters. But when I see the joinery detail below, and a fireplace surround constructed nearly the same way I’d construct it today, I feel much closer and more connected with our region’s history. I realize that we are still a young country, and in the context of the rest of the world, this is a pretty young house. The importance of preservation is emphasized not simply because the house is “old,” but because so much hard work went into constructing it.
Mid Rail and Stile Joinery
Oh, and the stiles were buried deep, just like in the Blue Room.
Buried Stile
For more photos of last week’s progress, click on the slideshow, below:
On Tuesday, Dan was removing southern yellow pine flooring in a room we’ve dubbed “the Pink Parlor.” As an earlier layer of flooring was uncovered, he detected beneath the scrim of sand and dust a pattern in the mottled finish. A little washing revealed a fine stenciling.
Stenciling Detail
Scott had also been working on dismantling the Pink Parlor, and discovered some craftsmanship that has made me re-think a hasty declaration about the best way to field panels
Fielded Wall Panel
The crisp beveled edges of these raised panels are beautiful. It’s hard to believe that this was the back face, appreciated by the carpenter alone. Large panels need to have a certain amount of thickness for stability, but their edges must be narrow in order to fit into a groove that runs along the inner edges of the stiles and rails. At O’Kane, we have discovered a variety of techniques for relieving that edge, each one more impressive than the last.
Wall Panel Detail
When we were first assessing the room, and removing the plaster, many of us thought that the wall beneath the chair rail was finished in plaster, because the wall section was so wide, and smooth. In our day and age, it is hard to believe that a 24 inch, 17 foot wall section could be composed of a single, clear board, but in this building’s day and age, that was entirely possible.
Scott removing nails from wainscot panel
The wainscot that Scott removed was 24 1/2 inches wide, tightly grained, and completely free of knots and defects. Lain upon the horses, in the strafing light, the scallops left by the joiner’s plane were obvious.
Wainscot Panels
This is a limited selection of what we’ve revealed in the past week. For more, see the slideshow below:
Before it was dismantled, the fireplace in O’Kane’s Blue Parlor got a lot of attention. It is a simple-looking surround, with a single large panel above and an applied mantle, but it’s a good representation of the vernacular style from its era. Aside from a little bit of backband added in a Greek Revival-era renovation, the surround was intact, and allowed a visitor to feel transported in time. While I found the piece pleasing aesthetically, I didn’t fully appreciate the workmanship until it was dismantled, revealing another chapter in the story of this building.
After Scott removed the adjacent paneling, and had cut or pulled the wrought nails attaching it to the wooden lintels, we realized that removing the surround wouldn’t be so easy. The stiles on either side of the panel and fireplace opening extended past the first layer of brick, but we could never have guessed just how deeply. After removing a piece of subfloor and digging into crumbled clay mortar, we found that the stiles extended below the surface of the subfloor by 8 inches. Eight Inches!
We haven’t found anything like this elsewhere in the building, and, based on the adjacent wall paneling, which went no deeper than the first layer of flooring, there is no reason to think that the original floor was eight inches lower. My theory is this: while the frame was being fit, joiners were cutting this and the other frame & panel walls (joinery shots, below). As soon as the frame was erected and sheathed, joiners installed this surround first so that this hearth could warm and feed the carpenters as they finished the rest of the house.
After digging out the stiles, we carefully laid the surround onto a specialized piece of preservation equipment called a Trash Can, and then we discovered something AWESOME.
A Pulvinated Panel! I have a thing for pulvinated, or “breasted” panels (would an analyst draw some connection between my interest and being a woman in a male-dominated field?) I have loved them ever since I first encountered them at Hancock Shaker Village on a NBSS class trip. At Hancock, the technique is seen on the front of the panel, and elsewhere, it seems to refer largely to friezes. I just think it is The Number One Most Elegant Way to field a panel, and ought to be used more often, and visibly. It is appealing to me how present the crafts-person is in this method of shaping a panel. The curve is shaped by his eye and hand, rather than a combination square. To me, the process is nearer to the construction of a chair than that of the austere wall panel.
Pulvinated Panel at Hancock Shaker Village
Given that the back of the panel we found was mostly rough, and totally invisible, the gently curved backside was not really where this crafts-person showed his stuff. That was in the triple stub tenon we found in the wide bottom rail, and the double tenon up top. The joinery involved is partly what leads us to believe that the surround may have been made ahead of time, off-site.
To see more photos of our process click on the slideshow, below
The Hill fireplace is nearing completion. From the outset, this project has been among our most rewarding. We designed the panelled wall using HABS drawings from a house built by the father of the builder of this house. Knowing that the design is grounded in historical precedent lends the project a sense of purpose greater than that of simply filling in the space around a fireplace.
Fireplace Return Panel
The paneled wall is composed of three frames, joined by mortise and tenon. One, that surrounds the fireplace opening itself, contains a large panel approximately 3′ x 6′. Its 8′ stiles extend past the panel to the floor, and the joint between the flat panel and the stile is covered by a bolection molding developed from the John Cram HABS drawings. The same bolection molding will cover the joint between the lower half of the stile and the masonry.
Adjacent the right side of the large panel is an open frame that will enclose a bookshelf, and perpendicular to the left edge is a third frame with a panel in its lower half, and a bookshelf above.
Pilaster Layout Close Up
The front wall plane will be adorned with three pilasters, designed from the Cram drawings, which handily obscure the joint between the two front frames, and their edges. The pilasters are not only decorative, as their obfuscation allows the front frames an expansion joint, that will move with the weather.
When I first approached this job, I was sure that the many mortise and tenon joints would pose the greatest challenge. And while they were a lot of fun, they came together with relative ease, and the glue ups were the most interesting aspect of the first phase of the process. Years ago, I used to wonder whether there was a word for glue-up phobias, because I clearly had it. The idea of the glue dripping and drying as I scrambled to fit the swelled joinery once filled me with dread, but over time I’ve borrowed tips here and there to develop a stress-free system that results in remarkably flat panels.
The larger of the two panels was 6′ x 3′. Historically, a plane of this size would have been broken up by a stile or rail to accomodate panels composed of single boards, some of which were over 2′ in width. However, the client has a specific painting that she wants to hang over the fireplace, and we thought it an appropriate compromise to create the panel by gluing together four boards. We’ve seen wide glued up panels in houses of a similar age, such as in the O’Kane House, in Durham, NH.
Half Glue Up
The biggest challenge when gluing together a wide panel is keep the whole assemblage flat. My contemporaries would recommending gluing together narrower pieces to keep it stable, but we wanted to use wider boards to better approximate the panel that may have originally been; so we used four, 9 1/2″ wide, eastern white pine boards. Tom and I started by jointing a face (flattening the face on a jointer), and then running the board through the thickness planer to create a parallel face, and then jointing an edge, and using our new SawStop to create a parallel edge. I then laid the boards out with their best faces up, arranging and rearranging to achieve the tightest joints and flattest initial surface. I then drew marriage marks across each of the joints, and used the marks to reference the biscuit jointer. A biscuit jointer adds to the prep time for glue ups, but I think that it has been the most important changes in my path towards glue-up enlightenment. After the glue is applied, no longer do the boards twist and slide past one another like Jamiroquai and his couch in the Virtual Insanity video. As the pipe clamp is tightened, the biscuits help to hold unruly boards to their better behaved brothers. They allow you more time to arrange your pipe clamps and cauls.
Panel, half scraped, see its shine?
I glued the panels in two pairs, and then glued the pairs together. I then used a cabinet scraper to remove some of the larger globs that had squeezed from the joint, and followed it with a random orbital sander. I tend to be snobby about sandpaper, thinking it a lesser method of smoothing and flattening, but for taking down the glue between joints (if it has hardened too long) I have found nothing better. Sandpaper abrades the surface of the wood, always leaving a slightly fuzzy surface, while planes and cabinet scrapers cut the fibers, leaving a smoother, shiny surface. In most cases, a plane is also faster. But I couldn’t risk the tear out that might occur if the flat-sawn boards had adjacent grains that run in the opposite direction. I was still able to achieve a traditional finish by following the sander with a cabinet scraper and extremely sharp block plane.
Raising Panel
After the panel was smoothed, I relieved the backside, creating a raised panel. Because the client wanted a flat panel on which to hang her painting, the work will never be seen, but like a lot of traditional forms, the fielding serves a purpose. It relieves the edge so that it will fit in a dado (groove) that runs in the edge of the stiles and rails, but allows the center of the panel to remain at a full 3/4 inch thickness, which increases its stability.
I transported the frames and panels to Hampton Falls, where they were assembled, and pinned. Tom and I spent the last week installing the frames, and adding the pilasters and trim. Please peruse the installation photos below, and check back for final shots of the completed work.
On Friday, we peeled plaster from the walls of the Blue Parlor, in the O’Kane Farmhouse. Scott was Bill and I, Ted, as we traveled in our proverbial telephone booth through layers of plaster, lath, wallpaper and time.
There were clues to some of what we might find. Surrounding the door openings were wooden strips, wedge-shaped in profile, that served as plaster grounds. They were a little over an inch thick, and were nailed four inches away from the opening itself, creating a border around the door opening that looked like a recessed casing. The application of these plaster grounds became popular in the mid-18th century, and allowed the plasterer to create a flat wall plane within the borders of the ground (Much of this initial dating information comes from James Garvin’s A Building History of Northern New England, pp 65-71). In most cases, the chair rail, baseboard, and door and window casings were applied directly to the frame, and served as the plaster ground. In Shaker buildings, for example, the casing is nearly flush with the plane of plaster. But in this wall, the plaster plane was an inch proud of the recessed door border, due to the applied ground, and there was a vertical, beaded joint between the side sections of the border and the top. It looked like a larger section of beaded paneled wall was peeking through.
As we carefully peeled the plaster from the lath, we paid attention to the composition of the plaster. Older plasters have a higher concentration of goat hair, and regularly one will uncover a multi-colored tuft that was was never fully mixed in. Older plasters were applied in three coats, a base coat, a straightening, or “brown” coat, and a skim coat. The base coat is thickly applied and creates the keys that lock the plaster onto the lath. The base coat squeezes through the slits between the pieces of lath, and droops behind. A skilled plasterer will use the right amount of pressure to create an even pattern of keys, enough pressure to create a key big enough to hold, but not so much that the plaster breaks off and splooges into the wall cavity. After the base coat has dried to a leather hard consistency, the brown coat is applied, and the plasterer drags a long straight-edged board, or screed, over the surface, flattening the wall plane between the grounds. The brown coat is usually where you see the most goat hair. The skim coat is the last, thinly applied coat, devoid of hair and leaving that hard, cured, eggshell finish. On the first two sections of wall, we found sawn lath behind the plaster, hung with machine-cut nails. This dates the added plaster surface to sometime after the mid-19th century, as we suspected (Garvin, p. 67).
One Wall, Many Coverings
Behind the lath we found beautiful, psychedelic wallpaper. The profile of beaded paneling telegraphed through, and punched vertical lines in the wallpaper at each panel’s joint. On the wall, one could see three different periods all at once. The horizontal shadow lines left behind by the lath, the lively geometric pattern of the wallpaper, and the vertical beads poking their noses through the surface.
Scott, Salvage Detective
So I liked that, but the most exciting discoveries were yet to come. So far, all we could determine in terms of dates was that the plaster was applied before the advent of wire nails during the late-1800’s, leaving no real indication of the date of the beaded paneling. Farther along the north wall, to the west, was a section that appeared to have once been partitioned off into a different room (according to a long joint in the floorboards). When Scott began dismantling this section, the wall cavity was different. There was a void behind the lath, and in its depths he could see the horizontal, bevelled profile characteristic of feather-edged paneling, and shiny, chrome yellow paint. The paneled wall he uncovered was was hung horizontally, and it had a feather-edged profile, where the edge of the board is beveled to a thin tongue that slips into a groove on the adjacent board. Conversely, beaded tongue and groove has a bead with an edge perpendicular to a quarter inch tongue. The joints in such boards are typically tighter. Both styles were used and re-used during the first half of the 18th century, but the feather-edged stuff is reminiscent of an earlier era.
Across the face of the older, yellow paneling, we saw the regular shadow lines of shelving, leading us to believe that this section of wall had been obscured by pantry storage and left alone when the rest of the room was upgraded. The wall plane of the beaded section is sufficiently proud of the feather-edged wall plane that the beaded paneling could be hiding more yellow feather-edged paneling–but this is only one of a number of possible scenarios. The beaded paneling could be contemporary with the feather-edged paneling; the feather-edged paneling might have been recycled from elsewhere, or simply used to delineate a different room in the house. In most homes, we’d never know the answer to these queries, because we wouldn’t dismantle the wall any more than was needed to make repairs, but the O’Kane house will be completely dismantled, and it is exciting to know that as we proceed, some questions will be answered, and even more created.
Three Walls in One
Adjacent the feather-edged paneling was the most exciting section yet. A section so exciting as to make the author flap her arms in an improvised, peacock-like dance. Behind the plaster Scott found accordian, or split board lath, hung with wrought nails. Accordian lath is hung using a wide, rough, knotty board. The first edge is nailed to the studs, and then the board is split along the grain and checks are stretched open and nailed, creating voids for the plaster to “key” into. This kind of lath supplanted the use of split, or riven lath around 1800. It was used until the mid-19th century, with the introduction of sawn lath.
Wrought Nail Detail
Wrought nails were used until the advent of machine-cut nails, invented in 1790. So we were finding a relatively newer lath style with an older nail technology allowing us to date the wall to sometime between 1790 and 1800 (Garvin, p 66). Dating a building without recorded documentation is a fuzzy practice. Often, the invention of a technology allows us to bracket a building’s date into “well, we know it wasn’t built before such-and-such,” and this is an unsatisfactory conclusion. Most technologies were in favor for at least fifty years. We use all these time brackets, and the popularity of certain styles throughout the house, to come up with an approximate date. It is unusual, and thrilling, to uncover a wall that so neatly falls between the advent of one technology and the extinction of another–and that was why I found myself flapping my arms, wildly.
Please peruse the photos below for more information about our process, and stay-tuned for more exciting discoveries.
Tomorrow we embark on the building of the Hill fireplace. The Hill house is a turn of the 19th century farmhouse, with all the attendant revisions and additions. The owner wants to restore her fireplace to reflect the time period it was built and the building trends in her region.
Through her own research, she believes the original building may have been constructed by John Cram, a prolific local builder. Fortunately, the John Cram farmstead, also in Hampton Falls, NH, was documented by the Historic American Building Survey, and was available through its database online. The Historic American buildings Survey was a program of the National Parks Service, that was developed in 1933 to provide work for architects, draftsmen and photographers left jobless by the Great Depression. Not only are the drawings and photos an incredible resource for our work, browsing through them leaves one with an incredible sense of patriotism. They document the best work of craftsmen throughout America, and are evidence with our abiding connection with history. Not only that, but one happens upon some pretty great photography.
John Cram Fireplace
So, the John Cram Farmstead. Out of the three documented fireplaces, the design above best fit the bare hearth that we are working with. We experimented in SketchUp with a number of pilaster designs. The others were faithful to classical elements and proportion, with a proper capital, and entablature. We kept coming back to this John Cram design, however, with its unorthodox use of crown incorporated into the capital. Though purists might reject it, the John Cram pilaster is simpler, and more appropriate to a vernacular building. This pilaster may not have been found in Greece, but it was found here, in Hampton Falls, NH. It makes sense, given the house’s location and style, but more importantly, because of HABS, we have good evidence that this was what was built in this area at that time.
Hill Molding Profiles, from Cram
The John Cram HABS drawings also gave us detailed molding profiles. Using a Williams and Hussey molder shaper, we will mill custom crown, bolection, and bed moldings based on the HABS details. The client wanted to make one significant change to the John Cram style, instead of the raised panels, she wanted flat panels with a bolection molding, which would accommodate a painting to be hung above the fireplace. In A Building History of Northern New England, by James Garvin (pg 137), we found a bolection molding that was identical to the astragal found at the top of the John Cram pilaster. The only difference is in the millling of a rabbet on the back of the bolection molding stock. Utilizing HABS, we were able to design a fireplace surround that not only fits the house’s age and vernacular style, but adjusts to the needs of its contemporary homeowner.
Tomorrow, we will lay out the framing around the hearth. Dave and Brian will prepare a plumb and level wall, while Tom and I will mill the stock, and cut and fit the joinery back at the shop. I can’t wait to get started on the frame and panels, but I’ll pause to post about our progress along the way.
