Whether you pronounce it like “tapenade” or “lemonade,” it has a certain ring to it, doesn’t it?
Way back when, the house had a balustrade around the edge of the second floor roof:
We believe the roof balustrade was original to the 1812 house (it is faintly visible in William Wall’s 1819 watercolor ). And, characteristically, Davis chose to retain and extend it when he designed the 1849 addition. A portion survives on the west face of his Italianate tower:
But a roof balustrade is essentially a fence at the bottom of a hill and so rain, snow, and ice tends to build up behind it and work their destructive ways.
As we worked on repairing the eaves, portions of the original balustrade support posts started to be uncovered:
The posts had been cut down and covered by sheathing boards during a re-roofing campaign (probably in the first half of the twentieth century).
We carefully recorded the layout of the support posts for future reconstruction of the balustrade.
Well not really… but close. Yesterday, we found what appears to be a well-preserved lead-lined wood toilet tank right underneath the roof boards tucked in next to the SE chimney. Kind of cool. This example probably dates to the second half of the nineteenth century based on the design.
An unsolved question is where the rest of the toilet was located. The space directly below is very narrow (the chimney stack to the right of the photo flairs out in the room below). The top flush tank could have serviced a bowl installed in the back bedroom behind the chimney or the pipes could have run straight down to the first floor where a bathroom still exists. If the bowl was, in fact, two floors away from the tank originally, it would have made for a rather delayed flush response time and a very powerful flush as roughly fifteen gallons of water came down from on high.
Carpenters have always left their mark (figuratively and literally) on their work. In the early nineteenth century, when Samuel Plumb’s house was originally built, they made a lot of marks.
Under the so-called “scribe rule” which had been used for hundreds of years, every joint was marked with a system of Roman numerals gouged into the wood. Each mortise was scribed (see this post for a description) to exactly fit one tenon. So every joint between a post and a beam or a stud and a plate was unique. To keep track of which part went with which, each side of the joint was numbered.
After 1830, a new system, called the “square rule” entered into general use which “squared up” the area around the mortise so that standard parts could be used. This was a heck of a lot quicker and marked the beginning of the transition toward dimensional lumber and all-nailed construction that accelerated after the Civil War. Isolated examples of square rule framing as early as 1816 have been found in New England as discussed in this link from Historic Barns of Connecticut.
But Samuel Plumb’s carpenters, working in 1812, were still “old school” and marked every joint. Since the joint only went together in one way, the carpenters needed to make sure not only that the numbering matched up but also that the parts were facing the right way. That’s probably what the arrow in the top photo is showing: which side of the plate should be facing out when they fit the studs and joists into it.
The photo below of the roof deck support structure shows another example of this marking system in use. It’s tough to figure out exactly what the “VIII” is marking in this location. It could be number of the ceiling joist next to it (which was pinned to the plate underneath the rafter) or perhaps the number of a mortise (out of view) cut into the plate below to receive a tenon for a support post. Anyway, it’s a “VIII” looking for another “VIII” as its mate.
Work began in earnest on Monday on the re-roofing project. A four-foot-wide strip of metal roofing was cut back and the sheathing boards removed all along the original North eave of the house. To create a strong and even bearing surface for the new roof, blocking was added between each rafter, and then new tongue and groove sheathing boards were nailed vertically across the entire eave. The new work was then temporarily covered with tar paper until the new EPDM roofing is installed.
On Wednesday afternoon, Jesse began carefully prying up the metal roofing along the North wall bay. His intent was to evaluate the condition of the first floor top plate and examine the original Federal period (circa 1812) gutter encapsulated underneath the sheathing boards. And this time, it was all there:
Looking left to right, you can see (1) the later crown molding added after the gutter was enclosed (2) a 1″ wide nailer used to attach the crown molding (3) the gutter itself, with a hand carved 3″ channel to conduct rainwater (4) later blocking (running across the gutter) used to support the sheathing boards nailed on top when the gutter was enclosed (5) the original bevelled sheathing boards which guided water down to the gutter.
Here’s another view, with all the later work removed:
Jesse said this is the best preserved original gutter he has seen in a long time. Also visible in the photo (right of the gutter) is the top plate timber supporting the wall below, which, thankfully, survives in good condition on top of most of the bay (one less thing to repair).
The gutter channel was carved out a single piece of wood (about 8″ wide) with the bottom carved into the profile found scribed into the shingles of the addition. The gutter was hung approximately level and the pitch was controlled primarily by the depth of the channel (about 1.5″ on on end and 3/4″ on the other over an 8′ run). This gutter was clearly not intended to handle the volume of storm water we have experienced recently.
Another very interesting thing revealed on Wednesday was evidence that this gutter was painted (rather than tarred or lined with lead) at some point in its history. There are remnants of a light yellow ochre paint (approximately the color of buttermilk or pancake batter) which roughly matches samples taken on the shingles and an earlier/primer coat of dark brown paint splashed on the sides of the gutter:
More analysis is needed to determine exactly what we have here. Was the paintwork original to the house? Was it just slop from the shingles or an intentional protective coating for the gutter? Or does it date from the Davis renovations of 1839 or 1849?
The Phase I Restoration project specs call for installing new half-round (5″ on the third floor pavilion and tower roofs, and 6″ on the second floor roof) galvanized steel gutters and downspouts to channel water away from the house. A portion of the original Federal built-in gutter system is also scheduled to be repaired during the project.
These new gutters will replace rusted out gutters of similar form (dating probably to the early twentieth century) that have failed in many places, adding to our water-related woes.
An interesting question to consider is what was A.J. Davis’s original solution to water run-off at the Dr. Oliver Bronson House. A roof probe on the North side of the house suggested that the original Federal built-in gutters were simply covered over when the second floor roof eaves were extended in 1839. But did Davis specify the installation of some type of early hung gutter system in their place, or did he view his design for “bolding projecting bracketed eaves” as sufficient to keeping water away from the house?
In Cottage Residences (1844), Davis’s friend and collaborator Andrew Jackson Downing praises extended bracketed eaves as offering functional as well as aesthetic benefits. In describing Design V, A Cottage in the Bracketed Mode, he states that “the protection afforded by the projection of the roof, will give complete security and dryness to the walls.”
Assuming that Davis shared this view (which is likely given Davis’s and Downings’ close association in this period) then no additional rainwater conductor system would have been needed. From a technological perspective, an attached metal gutter system for a residence, rather than a commercial structure, in 1839 would be a rarity (Although attached metal gutter systems are documented as early as the 1820s, they didn’t become widely used for houses until the late nineteenth century).
Since the house has been re-roofed multiple times and the eaves partially reconstructed in some locations, it is difficult to know for certain whether a Davis-era hung gutter system ever existed. But, perhaps, additional roof probes will tell us more. In the mean time, below are a couple more photos of the gutter hangers being installed:
This week, we had a new toy arrive from our friends at United Rentals: a 60 ft JLG boom lift with four-wheel drive. This replaces the prior lift (fitted with a bucket rather than a work platform) that we were using for the chimney dismantling operation. The new man lift will be used to install half round hung gutters on the second and third floors of the house. Toys are good…