When the roof shingles were removed, we hoped to finally solve the last big architectural mystery of Comstock House - but what we found was completely unexpected, and has led to other questions we still can't answer for certain.
The puzzle concerned the rain gutter system in the front of the house. We knew that there were downspouts inside all three of the columns, which were (presumably) attached to the back of the original gutters, and the aesthetic goal probably was to avoid big ol' metal downspouts from spoiling the look of the otherwise all wood-and-window house. Nice, but drawbacks include the risk that heavy rains could overwhelm such an enclosed system and cause leaks inside the house (the architectural history of this topic was discussed in an earlier essay). However that system was designed, we believed it apparently wasn't very successful. Between the earliest known photo (probably 1905, a few months after construction) and the 1908/1909 postcard, a rainstop board appeared above the actual gutter, likely intended to slow the torrent of rain as it shoots down our steep-pitched roof. Even that didn't perform very well; during the heavy rains of 2009-2010, every downpour turned the front of the house into a waterfall.
Until the shingles were ripped off, it was unknown what architect Brainerd Jones had engineered, or whether he had left that problem to contractors Williamson & McKenzie. Aside from the gutter profile shown above, there was nothing specified in the blueprints or in the written specifications. Thus we were completely surprised to find ourselves staring at not one, but two pipes emerging from each downspout. It was the oddest thing that any of us had ever seen.
Each pipe was 1½ inches and had bent-up lead flashing, which suggested that both pipes once emerged from the roof. The lower one was clearly once attached to the gutter; but what purpose served the upper pipe?
One theory was that the roof originally had oversized dutch gutters (also called box gutters, yankee gutters or integral gutters). These were sometimes used in Victorian-era America and were part of the house structure, built between the end of the roof and the flat board fascia that is visible from the ground (good diagram here, if trouble visualizing this). In this theory, our pipes acted as dual drop outlets (or maybe, a drop outlet with an emergency overflow drop outlet). Logical, yes, likely, no. The monster-sized fascia covering this design would have to be at least 8" high, and nothing like that is seen on the early house portraits or in the blueprints.
Theory #2: When the original gutter system failed, they improvised a second gutter above it to help intercept the cascade, and what we thought was a rainstop was actually a trough gutter (AKA stop gutter or Yankee Barn gutter). It's an interesting possibility, but couldn't be true; the two pipes were made and installed when the house was being constructed.
So to paraphrase Sherlock, once you've eliminated the impossible, whatever remains - however unusual - must be the truth, and this is it: Each downspout had its own private air vent.
 |  |  | |
| |||
Technically these are "back vents," and Brainerd Jones might have suggested that the contractor build these as part of his terse directions found on page 21 of the specs: "Vents from other fixtures, back vent where practical..." But the two big questions are A) was this design unique? and B) did it even work?
Researching this proved a great challenge; understanding turn-of-the-century plumbing is sometimes like trying to solve a crossword puzzle written in Ye Olde English. Terminology wasn't always the same as today (and sometimes descriptors meant the complete opposite), and there were a jumble of terms to describe the same thing. Quiz: Around 1900, what did they call the pipes that carried rain from the gutters? Possibilities include "downspouts," "conductors," and "leaders," sometimes mixed with a prefix of "roof" or "rain" or "water". When I discovered hyphens were sometimes used ("rain-water down-spout") and using that combination returned still more unique results in a Google book search, my brain popped a tiny aneurysm.
Victorians had the notion that a downspout could have multiple purposes. They could double as air vents for the sewer system, or rainwater could be diverted into regular waste pipes in the house, with the hope that a big storm would flush away grease and other clogs. Of course, if the downspout itself was clogged with leaves or snow, the entire venting system failed; sinks didn't drain and toilets didn't flush. That practice ended and you'll find many building codes from c. 1890 onwards adamant that "Rain water leaders must not be used as soil, waste, or vent-pipes." The emerging uniform building code also included two other points of interest to our mystery. There was a rule for downspouts concealed inside buildings, specifying that the downwpout be made of sturdy material and absolutely watertight with the gutter - a no-brainer, but it shows inner-wall downspouts were considered an acceptable building practice. And at least from 1915 on, the Uniform building code required vents for each plumbing fixture, but specifically stated no vents were necessary on downspouts. That might have been included for completeness' sake, or it might suggest that some builders were venting downspouts, and the practice had come to be considered superfluous.
But after all that research, I could find only a single mention of an actual vented downspout. It appeared in a 1889 journal published by MIT, then reprinted the same year in the widely-read trade magazine, "American Architect and Architecture" (Volume 26). It described an ancedote told by New York's Chief Inspector of Plumbing, where a basement was flooded because the downspouts couldn't keep up with high water volume from a heavy rain. The problem was solved by adding an air vent to the downspout. Now the idea makes sense; vents allowed water to shoot down those pipes like a firehose.
So finally we're at the big question: Did Brainerd Jones' system work?
Engineering reference books in that era provided tables showing every calculation a builder would probably need, including the proper diameter of gutter and downspouts calculated for the size of the roof. The rule at this time was one square inch of downspout for every 250 square feet of roof area. With each watershed of the Comstock House roof covering roughly 2,000 square feet, the house needed eight square inches of downspout per side. The vented system here came up short. Each 1½ inch diameter pipe provided about 1.75 square inches, so all three downspouts together offered a combined 5.3 inches.
The other part of the calculation was the size of the gutter itself, and here either/or the architect or builder stumbled. The charts specified a gutter of eight inches wide; the redwood gutters used on Comstock House were the standard five-inch width, which we verified after finding a discarded cut in a crawl space, seen at right.But the undersized gutter wasn't the worst problem; we discovered that the middle downspout was never connected. The pipe was still sealed with a lead plug that would have been used during a plumber's smoke test for leaks. Considering the incredible speed of the construction of Comstock House, it's likely that the construction foreman didn't notice that the roofers had shingled over the plumber's mistake. The result was that the front downspouts could not have handled a watershed for a roof less than half this size.
Conclusion: The original rainwater system for Comstock House never worked. The gutters were too small, they needed an additional downspout, and one of the downspouts they had was permanently plugged. The rainstop was installed within a few years, but that didn't solve the problem either. Finally the entire system was abandoned; the redwood gutters were ripped out and replaced with a standard "ogee" five-inch sheet metal gutter, with three downspouts destroying Brainerd Jones' look. And despite all that, the gutters still couldn't handle the volume of water in even a moderately strong rain.
Hopefully our new system, which is over-engineered according to the period specs, will prove more robust. We'll see.
In the months leading up to the roofing project, most attention was over something no one will ever notice. Hopefully.
Besides the failing shingle roof that had us playing attic shuffleboard with buckets during every rain, there was clearly serious problems with the bargeboard - namely, that most of it was either gone or rotted.
Bargeboard is the trim along the edge of the roof; when it's horizontal, it's called a fascia board and is the plank of wood behind the rain gutter. But when it's on a diagonal eave or gable, the same piece is called a bargeboard, rake board or vergeboard, and its purpose is to cover structural details, like the exposed end of the rafter. In the Carpenter Gothic style, the bargeboards were often elaborate mouldings, such as lacy "gingerbread." Here are the before and after pictures of the bargeboard on the west face of Comstock House (CLICK to enlarge):
 |  |
Our bargeboard is an odd, custom shape that none of our experts had seen before, and isn't found in the E. L. Roberts millwork catalog that was published in 1903, just before the house was constructed. The exterior of the board is rather plain, so it wasn't intended to be decorative; architect and preservation expert Mark Parry thought that Comstock House architect Brainerd Jones used the bargeboard to hide the edge of the shingles to prevent a ragged profile. But the side of the bargeboard facing the shingles has a deep notch with an unclear purpose, the mystery made even more difficult to solve because all of the surviving pieces were in very rough shape. On some boards, it looked like the notch was intended as a channel to carry away rain that might otherwise wick up under the shingles. Or was the original purpose of the notch to create an interlocking system between the bargeboard and shingles to seal the roof? Whatever the intent of the original system is now unknown; the wood is so badly eroded by more than a century of weather that precise interpretation is possible. Jones didn't write about it in his specifications to the contractor, and those who reroofed the house in later years made no effort to figure out the purpose, either roughly nailing the old bargeboard on the rafter ends as trim or replacing it with ordinary 1x6 planks.
It was contractor David Jessen who (as usual) figured it out. The flat surfaces on the back were meant to be attached to 30° furring strips, long lost. If the bargeboard is nailed to that and the whole assembly is attached to the end of the rafters, it creates a bed for the outside edge of the shingles. The roof gains a clean visual line as seen from the ground, and there is no exposed bottom to the shingle for possible rain wicking.
 |  |  | |
| |||
To mill these custom pieces (and others), we provided moulding profiles to Redwood Lumber Company of Healdsburg, who can actually supply wood from old-growth redwood trees that were felled in the same era as the construction of Comstock House. (Unsolicited endorsement: Highest quality work at great prices.) The outline above left shows one of the best surviving profiles of the original bargeboard, with the believed original profile sketched in.
 |  | |
| ||
As Jim Scotchler began shingling the back of the house, the "things" became a consideration. The things had to come off the wall, then the things had to go back in place once the shingling was done. And what in the world are the things for, anyway?
The "things" are 5-foot lengths of 2½ inch cast iron pipe, as seen at right (CLICK on this and any other image to enlarge). The downspout (currently galvanized steel, but we're replacing them with copper) is attached to the conductor head. At the bottom, the cast iron flares away from the foundation; in a regular downspout system, this is a separate curved piece called a "shoe."
The "things," as it turns out, are Victorian "downspout boots" and you can still buy modern versions of them, even made in cast iron. One catalog says their purpose is "protecting the light metal downspout from damage of traffic, vandalism, and other abuses," and that makes sense, particularly on a two story building like Comstock House; a serious bump against the downspout and over 15' of metal pipe could be ruined, and maybe even the gutters ripped loose. Architect and preservation expert Mark Parry pointed out that these may have been a familiar sight in rural Sonoma County in that era, as they're just the right height to prevent a horse or cow with an itchy backside from doing damage. If removed, the boots also provide a way to clean out the downspout from underneath - not an insignificant benefit, considering last year we had to hire a plumber to remove walnuts packed into a plumbing vent by a very industrious squirrel.
As the roofing project neared, a decision had to be made: Should we keep the boots or extend the downspouts all the way to the ground? The vote was overwhelmingly against the boots, and I was often inclined to agree - without the boots we could have 3" downspouts, which would be less likely to back up the gutters during a heavy rain. But then I'd ponder my central tenet: You don't get to choose which historic details to save and which to remove in a restoration (a friend calls this my "Nazi preservationist" tendency) and I'd also recall the plumber's long face as he delivered the bad news that "you've got walnuts," and back I'd be, finding new love for the ugly downspout boots.
The answer should come from architect Brainerd Jones, but his voice is not clear on the subject. The notes to the contractor state only that the "conductors" should be 2-inch galvanized iron, "well secured to building." No mention of boots, or whether the pipe should be one contiguous piece. (His note only concerns the gutter system in the back; originally the front downspouts were concealed in porch columns, which will be the topic of a separate article.) We do know from a Comstock family photo that the boots were in place in 1919 and an additional downspout had been added to the southwest corner; it's possible that the boots were also a post facto fix.
A little survey of the use of downspouts in this period in architectural history yields fascinating results (and that's probably the only time you'll ever encounter "fascinating" and "downspouts" in the same sentence). Victorian architects apparently didn't care much for downspouts. Never do you see them in blueprints or drawings, and you can examine photographs in books such as "Artistic Country-Seats" (1886) and nary see one for a dozen pages or more. What happened at these houses during a rainstorm is a mystery; hopefully there was a well-engineered and concealed gutter system, because otherwise the water would sheet off the roof, likely flooding the basement.
 |  |
Attitudes began to change in the early 20th century. The famed partnership of Greene & Greene viewed downspouts and the straps that attached them as decorative architectural elements, incorporating them in drawings as early as 1903. In the 1904 Reeve House at Long Beach shown above left, the gutters and downspouts are painted a light color to contrast with the shingles, and the "elbow" between the gutter and downspout appears almost structural. (A recent view shows these features are no longer present). Their 1906 Bolton house in Pasadena, shown above right, used a pair of downspouts to symmetrically frame the entrance.
Greene & Greene continued to emphasize downspouts - the 1908 Gamble House had no fewer than four on the front face, including one next to the front door - and Stickley also began to work them into his plans around that time. But the most over-the-top use of a downspout has to be Maybeck's design for the 1909 Goslinsky House in San Francisco, shown at right. Not only was the downspout prominent, but the unique twisted copper design and flower header made it the focus point.The contrarian in this period was Frank Lloyd Wright, who famously disliked downspouts because of their strong vertical lines, which in part explains why his buildings also famously leak. His 1904 design for Unity Temple in Oak Park, Illinois had downspouts hidden inside walls, and for that reason the National Trust has named the church as one of the most endangered historic places in America: "The system was undersized and essentially inaccessible, and to this day water continually overflows the drains and permeates the concrete roof slabs. Heavy rains in September 2008 caused a large chunk of plaster and concrete to fall from the sanctuary ceiling." Wright also had a 1907 house plan with a downspout hidden inside a chimney flue that was in the middle of the building. I'm sure that likewise worked out just swell.
The bones of Comstock House may be made of old-growth redwood, but the soul of the grand old place is its light. The daylong sunshine that fills the south and east bedroom windows; the warm green pastels that wash through the stained glass an hour before sunset; the twilight orange glow from soft electric lights in evenings.
The home at night could seem gloomy because the rooms never have the bright illumination favored in modern houses. It's one of those tradeoffs for living here, just as the hydronic radiator system denies you the quick gratification of a blast of warmth from the furnace on a chilly morning. Not having the option to flick a switch and turn midnight into noon is no great loss, in my mind; having deep shadows in some places makes the house feel quieter overall, even if a video or music is loudly playing in another room. Or, to reprise a theme often mentioned here, an old house is comfortable only if you understand and embrace the systems inherent.
LEFT: The entrance hall gas-electric chandelier, with three electric lights and three gas jets
Evening lighting would have been slightly different in 1905. Although we're using lightbulbs of approximately the same luminance as in that period (much more about that below), the Oates family also could light the gas on some chandeliers and wall sconces. The bright flame from the gaslight - which was coal gas, not natural gas - would have cast hard chiaroscuro shadows, which in conjunction with the gentler, dimmer electric bulbs, would have produced a harmonized "blending of the many lights," as described in a newspaper item about a 1906 party in the house.
Having fixtures that were both gas and electric was more of a necessity than luxury when Comstock House was designed in 1904, and probably every home in town had mantle lamps or some other means to augment the electric service. The reason is simple: The power company couldn't keep the lights on. The "juice" for Santa Rosa came from the Colgate hydroelectric power station in the Sierra foothills, and the lines often failed - sometimes for days on end - when the weather was less than perfect. An alternative source of illumination was also practical because electricity was very expensive. Adjusted for inflation, electricity cost 25 times more than it does today.
Although we have table and floor lamps around the house to make particular living areas brighter, we've sought to maintain the original brightness and color of light from wall and ceiling fixtures. In this period, incandescent bulbs were typically lower wattage than common today, and the filaments were usually thicker. The combination meant that the filament ran much cooler, thus emitting an orange glow instead of today's brighter blue-white.
It's possible to buy reproductions of old style bulbs (here's one source), but at nearly $18 each, lighting the whole house with them would be prohibitively expensive. Another product on the market, marketed as a means of extending the lifetime of incandescent lamps, is the “Button." These are adhesively attached to the base of modern light bulbs to reduce the wattage, allowing the filament to run cooler. The Button has a diode inside, which only allows normally alternating current (AC) to flow one way, literally cutting the power available to the lamp in half. The recommended way to use the Button would be to install a higher wattage bulb than is needed; since the lamp would then run cooler, it will last longer, saving the consumer money by replacing the bulbs less often. For our purposes, we use lower wattage modern bulbs which the diode cuts down further, taking the output into the range that was commonly used in the early 20th Century.
If the Button was the easy solution, accurately defining the problem was more of a trick. Namely, how bright were incandescent lightbulbs in 1905, really? As far as we can tell, no one in modern times has waded into this research, so the analysis below is original; corrections and other interpretations are most welcome.
Judging from multiple sources from that period including the trade journal "Illuminating Engineer," the typical lightbulb for home use was sixteen candle power, at around three watts per candle. This would mean that average bulb was a little less than 50 watts. Thus you can screw a modern 40-50W bulb into a socket and have that "Ragtime Era" look, right?
Wrong. The old bulbs were considerably less efficient than a 50W incandescent bulb today. Modern bulbs use tungsten filaments, which didn't come into common use until just before the World War I years. In 1905, lamps still had carbon filaments, which radiated more heat than light. The carbon filaments also burned much cooler than tungsten: 1300 vs. 3410 degrees centigrade. To compare the efficiency of old and new filaments, refer to the graph on this page from the 1912 "Cyclopedia Of Architecture, Carpentry, And Building" which shows the logarithmic efficiency curve for carbon filament bulbs. If plotted on the same graph, the much hotter tungsten filament would be shown to produce about EIGHT times the amount of light for the same wattage.
Another way to look at the problem is to compare modern and old lumens, but there's the little snag that early electricians compared watts per candle, not lumens per watt. Fortunately, a 1907 paper in Illuminating Engineer states that the standard bulb had the efficiency of 3.3 lumens per watt. By that calculation, the modern incandescent is about FIVE times brighter.
Whether the old lights were precisely 12 or 20 percent as bright as today's bulb doesn't really matter; once in use, the bulbs also rapidly lost efficiency as the inside of the glass blacked with soot. And then there was the issue of unreliable voltage; the more juice the brighter the light, and electrical transmission standards were notoriously unreliable. All that you can say with assurance is that home electric lighting was much, much dimmer than it is today.
Finally (!) our conclusion: the average 50W lightbulb in 1905 would be the equivalent to 7-12W today. We vary between both. For the newel post (pictured here) we are using the Button to cut the wattage from a modern 11W bulb by half, using a lower wattage at that location because it's functionally a night light. For wall sconces and chandeliers, we have 25W modern bulbs with the Button. In either case, the overall effect is the same welcoming orange glow that you should have seen when walking through the door on a fine evening in 1905.
Santa Rosa weather usually gives no reason for complaint, so the recent cold snap, where thermometers struggled to climb much beyond freezing even in afternoons, came as an unwelcome surprise. When it's this frigid, Comstock House can seem like a big old barn of a place that sucks in the chill - or, it can be reasonably warm, even comfortable. It all depends on how well you understand the house's systems, and how willing you are to work with those systems instead of fighting them.
Most everything about this house was designed for the comfort of the Oates family in 1905, and the first step to living happily in this old house is embracing their standards of comfort. Not every corner of every room can be brightly lit in the evenings as many people today desire (more about that above), and some bedrooms will be warmer than others because they have larger radiators.
And you can also forget the gratification of instant warmth on a chilly morning by jacking up the thermostat; the hydronic radiators can take more than an hour to make the slightest difference if they started cold. The Oates family likely either ordered a servant to fix the fire in the cast-iron boiler several hours before they arose on a winter's morning, or left coals burning overnight so the radiators never cooled. The former owner of this house chose to use a timer that kicked the boiler alive before dawn (she was apparently not bothered by the hammering that results as hot water begins flowing through stone-cold iron pipes, and unworried that the shock of doing same every day might crack the century-old plumbing). We instead are setting the thermostat low overnight, but never turn it off completely - the equivalent of letting the coals die down.
But when the weather turns this extreme, no radiator management is enough to keep our house comfortable; somehow the main source of the seeping cold must be also blocked off, and here that means closing down all of the servant's portion of the house, including the kitchen.
Architect Brainerd Jones may have tended well to the comforts of the Oates family, but his disregard for the people who worked for them was deplorable. In the blueprints there was only a single toilet for the help, and it was accessed via a nook on the kitchen porch. The only radiator in that entire section of the house was specified for the upstairs servant's bedroom, and it's the tiniest radiator in the whole place. Mercifully, the contractor varied from the plans by adding another radiator to the "sitting room" (which we believe was more likely the cook's bedroom) and installing a toilet in the second upstairs bathroom.
The lack of radiators aside, Jones' design is also much of the reason why this northwest corner of the house turns arctic. The sitting room juts away from the house, exposed on three sides; the kitchen and hallway wrap around the back porch, which has no basement underneath it or second story above it to help insulate (no second floor above the sitting room, either).
In the Oates' day, only if the poor servant(s) kept the stove red hot all day could these rooms be usable, but the family probably rarely experienced their employees' discomfort. Every path into the kitchen/sitting room has a pair of doors that can be closed to buffer smells and cold from the main portion of the house. The most obvious pair are the swinging doors on either side of the china pantry, which allows it to function as a kind of airlock.
Today, the design allows us to zone off the old servant's portion of the house, which means that there is about 7,000 cubic feet less for the radiators to heat. While the results aren't perfect, it makes a noticeable difference. On a recent night where temps plunged to 22 degrees and only slightly edged past the freezing point by noon, it was a tolerably-comfortable 65 in the main part of the house, while in the zoned-off section it was 42 degrees.
Obl. history sidebar: The original post about our heating system neglected to mention that most homes built in that era didn't have radiators and hot-water heating, but instead used a gravity furnace. Here warm air from a coal-fired furnace positioned in the center of the basement drifted upwards through ducts (wrapped in asbestos, natch) to registers in the floors or baseboards. Besides being ridiculously inefficient, the furnaces were labor-intensive because they needed to be kept as hot as possible, requiring frequent trips down to the dungeon to "fix the fire."
it wasn't until after WWI that the first furnace blowers became available to force the air upwards using a propeller-like fan, and about the same time oil burners began to replace coal fires. The switchover to heating oil was encouraged because of war-related coal shortages that led to a "coalless Monday" order that lasted into 1918. Unlike coal - which was needed to fuel transatlantic ships - there was no rationing of oil whatsoever.
(BELOW: Cartoon from the November 3, 1907 Press Democrat. CLICK to enlarge)
The first in a series of essays on the architectural background of Comstock House is now available. Behind the Design, Part I lays the groundwork by covering the Colonial revival fad and the emergence of the Shingle and Queen Anne styles in the late 19th century; the next entry will cover how these elements further evolved in the 20th century San Francisco Bay Area.
Demolishing the front steps meant removing the handrail as well, with no tears shed; it was a plain wooden pole added in the 1990s, and we know from historic photographs that no earlier rail existed. But when we rebuild the steps, should we install another handrail, albeit something nicer? There was also pressure that decisions needed to be quick coming; if a support post was to be added, a footing had to be ready when the new concrete pad was poured just a few days later.
A conversation about the handrail was underway between contractor David Jessen, architect Mark Parry, and myself. We agreed that we didn't want to reuse the old "mop stick" handrail or replace it with a new but just-as-bad equivalent. Possible custom wrought iron rails were most often discussed; a thin black rod would be the less obtrusive option, for certain. But should it run along one side? Both sides? Down the middle? Have some curly decorative flair at the ends? Candice and I were simultaneously kicking around ideas for something fancier with both a top and bottom rail, either art nouveau grillwork or Roycroft-like Arts & Crafts ballusters between. Either might be period appropriate, but the debate itself was drifting away from the true question: what would architect Brainerd Jones have done?
Since Jones didn't specify a handrail, we could only guess what he might have done on these front steps, if it had been included. But in the rear of the house were two possible examples: the once-identical handrails for the kitchen porch stairs and the back stoop, both made of only wood and with Jones' hallmark "Union Jack X." Wouldn't he have used this same design in the front steps? I went further, and argued that such a rail would have been down the middle, rather than blocking the side wall/platforms, and cobbled together a crude Photoshop mockup of what this might look like.
Candice endorsed the concept, but our contractor objected strongly. If we were really restoring the house we shouldn't be introducing major new elements, such as a 4-inch-wide wooden railing. Plus, he insisted, the handrail would never have been placed in the center. "I'll bet I can find five examples of rails in the middle of steps," I pushed back. "Show me one," he challenged.
Since the Comstock House design is based on the shingle style country house of the 1880s, I looked first at houses designed by McKim, Mead & White, the most famous architectural firm for this school. Almost immediately, I found an example of a center handrail: The 1882 Isaac Bell House in Newport. But that was a simple metal pipe, and it wasn't free-standing - at the porch landing it terminated on a support column (not visible in this photograph). Only in the most technical sense was this a match, and search I might, no other examples could be found. Down-the-middle handrails rarely/never existed in the era, even expanding the search to late Victorians in general. Lesson learned: if you're lucky enough to have a contractor like David Jessen with decades of experience working on Victorian-era homes, pay heed.
But what did the old technical manuals say on the topic? Like shingle preservation and other obscura discussed here, the study of Victorian porch handrails lacked any easy Google answers. Turn-of-the-century carpentry books contained much on building stairs and handrails, but the focus was exclusively on construction of interior steps and 
fine joinery (gotta have a handrail punch to turn those handrail screws) and such complex craftsmanship as the geometry of making winding stairs. Rarely were exterior stairways ever mentioned, and never was found a discussion of center handrails. A passing reference to a center railing was found in the classic 1897 Practical Building Construction (which remained in print for more than thirty years), which implied using a center rail as a median divider when there was heavy foot traffic in both directions.
The period photographs also show that side handrails on entrance stairs began evolving -- and sometimes disappearing -- during the 1880s. Homes built in the Victorian Gothic Revival style usually had a first floor high off the ground, which required many steps; houses built during the Colonial Revival craze after the Centennial year were lower to the ground, with rarely more than a handful of steps needed. Victorian Gothic front steps were usually narrow and straight, almost like a chute; steps found on houses after that period tended to be much wider, and when there was a long distance to the front door, there were landings or switchback designs that broke the climb into shorter segments.
(Sidebar: After looking at hundreds of 19th century front steps, I've come to believe that Americans in the Victorian era mostly saw handrails as an aid to pulling themselves upward, not as a safety measure to prevent a tumble down. Examples abound of railing designs that would be be easy for climbing, but possibly dangerous for descending -- the rail is too short, begins at the third or fourth step from the top, is interrupted by a large newel post, and so on. Some academic has studied this to death, I'm sure.)
Above are details from a selection of examples from "Artistic Country-Seats" (1886, reprinted by Dover with new commentary in 1982) that show what happened to American front steps after the High Victorian era. In some cases, the sides to the stairs completely -- and dangerously -- disappear, but more often the sides evolved to frame the steps, integrating them into house design. 
In the middle example, the sides continue the spindle woodwork found on the porch balustrades. On the right, the steps are encased with a low wall that continues the shingling on the house, and is topped with a cap that could be used as a planter box or some sort of platform.
In the end, we were asking an irrelevant question: Brainerd Jones didn't neglect to build a handrail on the Comstock House front steps -- he did include it in his design, via a shorter and a taller sidewall that was completely appropriate to the 1880s style, giving visitors a bit of support while climbing the seven stairs. Or the two caps on each side could simply be used as a nice place to sit, which is exactly what the Comstock family did, posing for a snapshot during a 1937 reunion.
As the reshingling of Comstock House began, we found a few old shingles with a faint maker's mark stamped on the back:
ABERDEEN LBR. & SHINGLE CO. 6/2 EXTRA * A * ABERDEEN, WASH. |
This discovery resolves a question from this blog's very first post re: the "Star A Star" shingles that architect Brainerd Jones demanded in his specifications. Sure enough, here were old cedar shingles with exactly that marking. But they weren't white cedar, as Jones indicated -- which raises the possibility that the contractor sneaked in a lesser-quality product.Although Jones specified Eastern white cedar, Western red cedar shingles were used instead. It was likely a cost and availability issue; white cedar mills were far away in the upper Great Lakes area and New England. White cedar shingles are remarkably light individually, but the weight adds up quickly when they are applied; expect about 1 sq ft. of coverage equal to 1 lb. of shingles. Now multiply that by the size of this house, and imagine all those tons of shingles being shipped thousands of miles. By rail. In 1904. That may have been too much of a strain on Mr. Oates' deep pockets, despite his boast that "nothing that money and taste can provide will be omitted in making it a comfortable and attractive home." By contrast, the red cedar mills in Washington state were regularly sailing boatloads of wood into the San Francisco ports, according to shipping notices in newspapers of the day. Western red cedar was far cheaper and readily available. Plus all "* A *" shingles were of comparable quality, right?
The 1924 "Lumber Inspection Rules" manual shows both Western red and Eastern white shingles had star-A-star grades. White cedar had three basic classifications: "Extra star-A-star" at the best, followed by "Standard star-A-star" and finally, "Sound Butts" grade. But the manual says that red cedar had no fewer than seven first-class grades -- and the kind used on Comstock House were the lowest. The best-quality red cedar shingle were called Royals, followed by Perfections, Eurekas, Perfects, Extra Clears, Extra Star-A-Stars 5/2, and finally Extra Star-A-Stars 6/2. (The x/2 refers to thickness at the bottom, or butt end, of the shingle: 6/2 meant that each shingle was 1/3 of an inch at the thickest, and thus six of them made a pile 2" tall.)
Note that these rules are from a reference book twenty years later; it's possible that when Comstock House was built, the red-westerns had only three grades, like the white-easterns. Or maybe it was presumed that star-A-star really meant, "sidewall grade," regardless of type of wood. It's really a quibbling point, except for speculating whether Oates was gypped; even if these are bottom-first-class grade shingles, they have weathered remarkably well -- cupping and warping aside, the things are still (mostly) in place and (mostly) intact over a century later.
A more interesting question to explore: How do we know these are indeed the original shingles, and not from repairs made decades later?
The Aberdeen Lumber & Shingle Company was founded in 1899 and was among a dozen or-so mills around Chehalis County (renamed Gray's Harbor County) Washington that cranked out an enormous volume of wood products. The area was also a key battleground in efforts to unionize the timber industry; the IWW organized a strike of loggers and sawmill operators there in 1912, but were violently opposed by a "Citizen's Committee" (read: company goons) who clubbed the strikers and forced 150 of them into boxcars for deportation out of the county. Their plan was foiled, however, when train crews refused to roll the locomotive. This is one of those tales that reads like a Steinbeck novel, with each player revealing his true character as the crisis unfolds; the mayor of the town of Hoquiam stood with the train crews refusing to deport the strikers, but Aberdeen's mayor tried to deputize city workers into strike-breakers, with most of them quitting their jobs rather than obey the order. The Wobblies won that strike, but vigilante attacks on labor organizers continued for more than a decade.
Aberdeen Lumber & Shingle (AL&S, to keep it short) was one of the largest mills in these years of strife. The industry peaked in 1925, when more than a hundred mills in Gray's Harbor County cut about 1.3 billion board feet (see: "Tempest in the Timber" for a good read on the overall history). From then on, it was a spiraling collapse; AL&S owner Cliff M. Weatherwax made the news in 1928, when he brokered a deal to merge 75% of the lumber mills in the area.
The last direct mention I can find of AL&S is from the Centralia (Washington) Daily Chronicle, March 5, 1930: "Lumber Industry is Humming Again" noted that AL&S and other lumber operations that "have been closed down for some time" are reopened. The mill employed 200 men, the article says. But a 1933 item on sawmills in Aberdeen reopening after a strike lists the status of nine mills but doesn't mention AL&S, so perhaps it had already merged with another operation. The reference desk at the Aberdeen public library says AL&S was listed in the Polk business directories for the last time in 1937.
Conclusion: The AL&S shingles must be original to the 1904/1905 construction of Comstock House. Although some shingles were later painted by the Comstocks, it's extremely unlikely that Nellie Comstock would have reshingled the the house when it was only 30 years old, at most.
(Extra historical footnotes and obl. Ripley's-Believe-it-or-Not tie-in: Cliff Weatherwax may have owned AL&S at least through 1928, but he was no longer managing the mill; by then, he and the missus were society swells living in San Mateo, where he was a director of the polo club. Cliff was murdered at age 60 during a 1939 robbery that was apparently never solved. Leaving a New York City nightclub he hailed a cab and told the driver to take him to Brooklyn; en route, the cabbie testified that a passer-by yelled to him that his passenger had fallen out of the car. He told police that he found Weatherwax sitting on the curb with a head injury. The frightened taxi driver said he fled, but returned later and found Weatherwax gone. Police told the Times that Weatherwax had been beaten and robbed of $1,000 and a gold watch. The family also left a big thumbprint up in Gray's Harbor County; grunge rocker Kurt Cobain went to Aberdeen's Weatherwax High School, named for Cliff's father.)
After weeks of research on cedar shingle treatments and testing four products (see: "Preserving the Aging Shingle"), I had enough information to write a book on the topic (well, a lengthy blog post, anyway) but still felt that there was no obvious best choice. A TWP formulation that appeared to last about a decade in this Mediterranean-like climate seemed to be the better of the bunch, but it was still a far cry from the forever-lasting creosote stain that was available when Comstock House was built.
But as I was about to make a decision, Kelly-Moore sales rep Greg Fitch asked why I hadn't looked into TWP's 200 series, which is made specifically for shakes and shingles. Mistakenly, I assumed that the product was either discontinued because no stores carried it, or it wasn't legal in California -- after all, the stuff was 93% solids, far higher than anything else on the market. A call to the distributor revealed that not only was it available in the state, but that they sold quite a bit of it in Marin County. From him I also learned of the Marin Wood Restoration and Painting Company, which offers an extremely helpful web page on how they blend different TWP formulas (don't miss their photo gallery). The drawbacks are that this treatment takes longer to dry -- no problem here, since I'm hand-dipping each shingle -- and requires extensive stirring; a tar-like layer settles in the bottom of a can after only a few hours. The shingles also dry to a considerably lighter color.
Thus armed with new info, I ordered three cans of TWP and began experimenting afresh.
My new test panel had 7 samples, shown below. As with the first set, these shingles are all Maibec grade A white cedar. These shingles were dipped in different recipes containing TWP 203 (gold), TWP 200 (clear) and TWP 515 (cedar, which was tested by itself in the first panel). The clear formula offers no UV protection, but is used to lighten other formulas while still providing the non-colorfast wood preservation benefits of the other TWP 20x product line.
The shorthand recipe is shown in parentheses; (1-2-3) means 1 part gold, 2 parts clear, and 3 parts cedar formula, for ex. Because the TWP 20x products lighten in color, the test panel below was photographed ten weeks after dipping, except for shingle 2A, which is a freshly dipped shingle with the same recipe as shingle 2.
 | 1: Just TWP 203 (1-0-0) |
| 2: Equal parts gold and clear (1-1-0) |
| 2a: Equal parts gold and clear. but newly applied (1-1-0) |
| 3: (2-1-0) |
| 4: (1-0-1) |
| 5: (1-0-3) |
| 6: Equal parts of all (1-1-1) |
| 7: (1-1-2) |
Curiously, the only significant color change appears when mixing the clear formula with cedar, as shown in samples 6 and 7. Our final choice was recipe two, an equal mix of TWP 200 and TWP 203. So far, it looks pretty good; we'll see.
Reshingling the house was a project we hoped could wait for a year or two, but porch repairs left the southeast corner with exposed new plywood sheathing that needed to be covered before the winter rains. There was no question that we would use unpainted white eastern cedar shingles, as architect Brainerd Jones described in his plans. But without paint, how should the new shingles be protected? In Jones' day, shingles and shakes were often treated with creosote stain, which is now banned. In evaluating the wood preservatives available today, durability was a paramount concern; using a product that needed to be reapplied every couple of years or so would mean that a house this size would be too-often surrounded with scaffolding. Another factor to consider was that white cedar weathers to slate gray fairly quickly, and Candice wanted to keep the live-wood color as long as possible.
Preserving the natural look of the wood meant narrowing the search to semi-transparent stains with UV blockers to prevent fading. To block the ultraviolet range of sunlight, all of these products include trans oxides. More formally known as "transparent" iron oxide pigments, these are extremely fine metal powders that will absorb UV light when suspended in a solvent or water (trans oxides are most commonly used in automotive paint). Since truly transparent metal only exists in the Star Trek universe, no wood stain with UV protection is actually "clear;" all of these products have at least some tint of color because of the trans oxide powders, which are available in wood-friendly pigments of yellow, red, green, brown and black.
Except for the common trans oxide ingredients, there are wide variations in the formulas. Much is concealed as proprietary in the MSDS and technical data sheets, but common ingredients in many of these products include mineral spirits and xylene and/or another form of benzene. Sometimes the recipe includes surprises: a few have ethylene glycol (antifreeze), and some have a fungicide banned in some countries.
A common way to evaluate these products is to compare the percentage of VOCs vs solids in the formula. California and many states in the Northeastern U.S. have limits on the total amount of Volatile Organic Compounds that can be included in paints and stains, and all of these products are compliant. Asking knowledgeable professionals (roofers, senior paint store salespeople, shingle and wood stain company reps) about this issue, I found confusion abounds. Workers in the trade assume high VOC numbers are signs of a better product, and more than once it was suggested that better stuff is available out of state -- a claim that approached urban legend status because no one could name what product and what state had the goodies. Contradictory, I was also told by some that low VOC numbers are a measure of quality -- that the chemicals are rip-off fillers that evaporate quickly. In truth, I don't think any generalizations can be made that low/high VOCs are a way to judge quality, but percentages (usually in grams per liter, or "g/liter") are given with some products below.
To compare the different products, I set up an experiment: four shingles (Maibec grade A Natucket) were each treated with a product that might be suitable for reshingling. On July 4, 2008, the shingles were nailed to a board, which in turn was mounted on the southern face of the house. To accelerate aging, I spray the shingles with chlorinated tap water 10-12 times a day. The experiment will have limited value to this project because our decision will have to be made in a few weeks; I intend to continue the test until 2010 or so, however, and post updates here.
Pictured below are the four treated shingles; in the lower part of the image is a closeup showing the wood after it was hit with a spritz of water.
| 1 Flood's CWF-UV Shown here in their "clear" finish, Flood's UV formula is not being considered for this project because it's not recommended for use on new wood. The company's website states that new wood should be instead saturated with their water-soluble "New Wood Defender" for a year, but oops -- that turns out to be a discontinued product. Their telephone customer support rep says they now suggest that new wood just be left untreated until it's dry enough to absorb a drop of water in twenty seconds. This product has an unusual oil/water formula with about 60% VOCs, most of it paraffin/naphta. Flood is the only company here that offers a warranty on UV damage (4 year). |
| 2 Weather-Bos™ Shown here is three coats of their Roof Boss "Formula 5." With its water-based formula, this is the greenest product on the market. Because it's so safe, the company doesn't have to reveal much about the formula (it says only that the product contains "natural oils and resins"), but their stats are far better than the competition: "up to" 72 percent solids, and a tiny amount of volatiles -- just 120 g/liter. Weather-Bos™ offers no warranty (except for defective products) but does offer an interesting rant on the meaninglessness of "guarantees." |
| 3 Cabot Clear Solution Shown here in their 9200 series "natural" formula. Cabot's has an excellent reputation, and white cedar shingle distributors Maibec and WCS offer shingles pre-treated with the product. A linseed oil-based formula that the company suggests be re-applied every 2-4 years, it has 250 g/liter VOCs and is 64 percent solids. Warranty: replacement cans or refund if defective. |
| 4 TWP (Total Wood Protectant) Shown here is the formula 515 "Light Cedar." Its heavy paraffin oil content results in far more water bead than any other product, as seen in the photo detail above. This particular recipe is 60 percent solids and 350 g/liter VOCs. TWP also contains Folpet, a broad-spectrum fungicide that is highly toxic to fish and invertebrates. But kudos to the company for alerting consumers to this risk; Cabot Clear also contains Folpet, but doesn't mention its toxicity. |
| Sikkens Cetol SRD A popular Canadian finish that contains over 60 percent volatiles (mostly mineral spirits) and about 38 percent solids. The U.S. shingle mill WCS offers wood pre-treated with Sikkens. |
| Penofin Based on the exotic ingredient of Brazilian Rosewood Oil, Penofin also includes a familiar mix of benzines and paraffin/naphta. Performance is about the same as other products; reapplication every 2-4 years is recommended, and as often as every 9 months on horizontal surfaces. |
| Olympic Semi-Transparent Premium Acrylic Stain This latex coating is a possible contender because a leading Candadian white shingle mill, Waska, offers a 15-year basic warranty (fading NOT included) when two coats of the product is applied at the factory. |
Of all these products, Weather-Bos™ is by far the most obscure; I've yet to meet anyone else in California who knows about it. Candice had used it at her home in Portland, and I later tried it on a redwood deck and cedar fence at our Sebastopol house. Weather-Bos™ certainly has the best stat sheet, winning both as the most environmentally safe and also having the highest levels of solids. So why doesn't everyone use their product? Partly because of the application issues: They recommend a minimum of three coats, but for optimum performance you should apply as much of the stuff as the wood will absorb. For maintenance, users are advised to "power wash it once every year or so" and apply another light coat. They also suggest washing your precious wood at a pressure setting of an incredibly high 3,000 PSI (don't stand in front of THAT fire hose).This is a product that I'd like to give an unqualified endorsement, but can't. That earlier cedar fence was treated with over three coats (I recall it was at least five, maybe six), and was beautiful for a couple of years. But now five years later, there's little color left. I'm certain that a power wash followed by applying an oxalic acid cleaner and another coat would do wonders, but I can't do that level of maintenance every three years on a house of this size. I would probably try it again on another fence or similar project, however.
A June, 2000 Consumer Reports survey on lightly-tinted deck treatments ranked TWP, Sikkens, and Penofin as the top 1, 2, and 3 (respectively) products. Sold locally by Kelly-Moore Paints, TWP was also anecdotally recommended more than any other product, and I couldn't even find anyone complaining about the stuff in online discussion forums, which is a remarkable tribute in itself. The best testimonial, however, comes from the product in its worst shape.
It seems odd that a paint store would have a shabby look, but that's the case of the Kelly-Moore store in Novato; its rough cedar siding was last treated with TWP nearly eleven years ago, in the autumn of 1997. The photo at right shows the south face, which has endured the most weathering. The original color is intact wherever the wood grain is raised; the rest has aged to gray. The color hasn't turned black, which is an often-heard fear about these newer formulations, and the stain hasn't flaked off in patches. Most impressive of all, if the image is desaturated of color, (see inset) the overall brightness remains nearly uniform to its original levels. Yeah, viewed closely and on such long boards, the finish does appear mottled -- but on shingles with 4 ½ inch exposure, I believe the effect would be a graceful transition from the natural cedar to classic weathered gray. Or should we try to forever preserve the color of new wood?An update appears in the post above.
What if you could take a delicate thing like a house shingle and make it last forever, never losing its color? Well, maybe not "forever," but certainly for decades beyond its expected life? Such a technology existed once, and played a nearly-forgotten role in the pageant of American architecture.
Creosote, or more specifically, coal tar creosote, to distinguish it from the nasty and dangerous stuff that builds up inside chimneys from wood fires, was the most popular wood preservative for a hundred years. A 1874 patent mentions in passing a still-older "creosote process" for pressure treating wood, but the first specific recipe for preserving building materials using creosote is found in a 1880 patent for a mix of creosote, turpentine, and paraffin. Several patents were granted in the following years claiming a new or improved formula or process for wood preservation, but the breakthrough was the 1884 patent by Samuel Cabot Jr. for a formula specifically designed for coloring and protecting the "ornamental shingles now so much used by architects." Cabot approached the problem from a different angle than the other inventors, who only wanted to drive preservatives deeper into the wood; his main objective was to find a way to preserve color. The key element in his formula was to use distilled creosote, and twice distilled at that.
Cabot's invention came near the end of the golden age of Shingle Style architecture, so it's unlikely that pioneers such as H.H. Richardson or McKim, Meade, & White ever used his "creosote stain" on their later Gilded Age masterpieces. But as it turned out, Cabot's timing was perfect; the American Queen Anne Style was just about to sweep the nation in the latter part of the 1880s. An interesting "chicken or the egg" question to ponder is whether Queen Anne's popularity was due in part to Cabot's vivid new stains, or whether Cabot owed his success to popular demand for vivid colors on all those new Queen Annes.
By the turn of the century, creosote stain was well established. We know Frank Lloyd Wright used the stuff because his son once fell into a barrel of the gunk. Craftsman architects Greene & Greene certainly relied on it; architect and preservation expert Mark Parry recalls his mentor being told by Henry Greene to specifically use creosote stain for restoration work on their 1908 Gamble house masterpiece ("one carefully applied coat of Cabot’s Creosote Stain #7, allowed to dry two weeks and followed up by another, and it'll last another eighty years"). Was creosote stain used on the shingles for Comstock House? Alas, on this the contract specs don't specify either way. Brainerd Jones indicated only on pg. 8 that both the roof and siding were to be covered using white cedar shingles, but Parry believes Jones must have assumed that the contractor would've used some sort of preservative.
There certainly were other options available than Cabot’s, but many were dolorous; a 1916 reference, "The Chemistry and Technology of Paints" noted that some stains were made by dumping colored linseed oil into crude creosote, sometimes with carbolic acid or kerosene. A 1912 analysis found some contained no creosote at all (their example formula of a "fairly high-grade" stain was nearly one-third creosote oil). A 1930 analysis even found that some were making a homebrew stain with used auto crankcase oil.
Creosote stain earned its reputation for preventing tinted wood from fading. As a 1886 Cabot ad boasted truthfully, "Owing to the strange PRESERVATIVE POWER of the Creosote, wood treated with this Stain cannot decay but simply wears away from the force of the weather." The definitive 1930 USDA report, "The Preservative Treatment and Staining of Shingles" (PDF) performed a 40-month experiment with pine shingles, which found they kept their color well -- although they tended to ooooooze creosote in hot weather.
As for fire protection, creosote stain wasn't great, and several references mentioned that abestine (powdered asbestos) was often added as a fire-retardant. A set of 1916 experiments found creosote stain offered little more protection than paint. The hands-down winner in that test, by the way, was "Penetim" treated wood, which took over nine minutes to ignite -- more than three times longer than any other treatment. Alas, nothing today can be found about this formula, or even the Penetim Manufacturing Co. of New York. (If you know anything, please contact me.)
Creosote stain continued to be widely used until about 25 years ago (see chapter ten in this very informative 1974 Defense Dept. manual on all kinds of roofing). But because of concerns that creosote was a possible human carcinogen, the EPA proposed restrictions in 1984; creosote oil compound hasn't been available for sale in the U.S. since 1988. The danger isn't deemed high enough for an asbestos-like cleanup -- railroad ties, utility poles, and other creosote-impregnated wood is still reused and sold. Creosote is also banned in Europe, Canada, and probably elsewhere.
As an alternative to creosote, the 1930 USDA writer thought that pentachlorophenol (PCP) had "cleaner colors" along with none of the oozing problem. PCP, also known as Penta, is still an option today for pressure-treated wood. But alas, there's never been a formula using anything but creosote that will make shingles last. And last. And. Last.
Creosote may also have gotten a bad rap as a human health risk. The Wikipedia entry reports that a 2005 mortality study of creosote workers found no evidence supporting an increased risk of cancer (no endorsement of accuracy from me until I can read the actual data).
The EPA rulings on creosote is up for evaluation this year and public comments are welcome -- at the master index to docket # EPA-HQ-OPP-2003-0248-0048, there's plenty to read. Among the questions posed by the EPA: "Are there unique uses and benefits of creosote?" Read the guideline for public comments (PDF) and respond by June 16, 2008. Comments can be submitted via regulations.gov; enter the docket number and follow the directions given.
Obl. research note: The gentleman at the end of "Monty Python's The Meaning of Life' who has an unfortunate eating accident after consuming a final ''wafer-thin mint'' was named "Mr. Creosote." You'll NEVER guess what bobs up often during creosote web searching, no matter how much the search criteria is refined.