In 1818 the Baroness Hyde de Neuville painted a water color of Montpelier’s western facade. The watercolor included a fence, evidence of which has been found archaeologically, depicted with black pickets, or pales as they were called in the 18th and early-19th century, black rails and white posts. Evidence that Madison’s fence remained painted a dark color throughout his retirement period (1817 to 1836) is found in John Gadsby Chapman’s circa 1832 sketch. The sketch also clearly shows dark colored pickets and rails and provides a more detailed image of the fences individual panels
Questions relating to how common dark colored pickets were in the early-19th century and what type of coatings could have been used to create the dark colors were developed after looking at the watercolor and the sketch. After reviewing period sources and current scholarship, it has been concluded that fences, including post-and-rail and pale fences, were commonly tarred, whitewashed or painted with linseed oil in the 18th– and early-19th centuries. It also appears that using linseed oil to paint fences became much more popular in the second-quarter of the 19th century as industrialization reduced the cost of linseed-oil based paints.
References to pigmenting all three of these finish types were also found, with a deep “invisible” green being the most common color, although white, black and gray were also noted. As opposed to an unpigmented tar coating, which seems to have been used primarily as a preservative, the pigmented coatings were recommended either to help the fence blend into the landscape or as an attempt to imitate a more expensive material. Based on the fact that other elements of Montpelier’s early-19th century picturesque landscape were reflective of the English landscape school, it appears that the first explanation best explains the dark colored pales found at Montpelier.
Evidence of fences with a comparable pattern of white posts and black panels, similar to the de Nueville water color and Chapman sketch, was also undertaken. Two period authors were found that mention such a pattern. The first one recommends white posts and dark panels because darkly colored pickets were “invisible” at night and so the lighter colored posts were needed to ensure that carriages and riders did not crash through the gate in the dark. The second description for a polychrome fence comes from a specification for a new baluster pale fence to enclose the New Haven burying ground. The new fence was intended to resemble an aged or oiled bronze fence with stone posts.
Finishes for Fences
After a search of late-18th and early- to mid-19th century English and American documentary sources, along with a report on Colonial Virginia fences by Vanessa Patrick of the Colonial Williamsburg Architectural Research Department, five main finish treatments for period fences were documented. The five options were: to simply do nothing and let the fence weather and decay naturally, to apply one to three coats of unpigmented tar to the fence, to apply a pigmented tar to the fence, to whitewash the fence and, finally, to paint with a pigmented linseed-oil paint. All five appear to have been used during the early-19th century, although painting a fence with linseed oil based paint appears to have been much more common towards the end of the study period.
Leaving a fence unpainted was the simplest, and cheapest, option available for treating fences in the early-19th century. However, it also allowed for the quickest failure of the fence’s wooden elements due to exposure to water and sunlight. Therefore, it appears that that unpainted fences were most often found away from the main house or associated with rural, lower class dwellings. Two surviving period Virginia pales illustrated in Patrick’s report on fences were unpainted.
Tarring a fence appears to have been a common option both in Virginia and in England. The term “tar” was applied to both wood tar and coal tar in the late-18th and 19th centuries. One author, writing about Baltimore in 1833, describes the two tar types as follows: “From the pit coal used to produce the gas a quantity of tar is obtained which is sold at $3 per barrel, a lower rate than the common tar extracted from pine wood.” He also states that the pit coal tar is “peculiarly adapted to other purposes such as painting roofs of houses, paling, &c.” Tar produced from wood was also known as pyroligneous acid, which specifically means a “substance produced by the effect of heat on wood, especially by destructive distillation.”
Patrick states that fences were frequently tarred in Virginia, although she does not illustrate any surviving examples from the period that are noted as being tarred.However, a single coat of tar would have been absorbed into the wood and, without additional analysis, might have been mistaken for an unpainted surface. The ability of tar to soak into the wood was also commented on in the period and one author states that a single coating of tar would leave no “body” because it “gets completely into the wood.” English and American landscape and agricultural writers from the late-18th and early-19th centuries also recommend using tar. Almost exclusively, they recommend tar because it is, in the words of one author, the best material to “preserve all outside wood-work” including “ornamental paling, and indeed all good outside work.” Tar coatings could also be pigmented a range of colors using chemically stable pigments such as charcoal and different shades of ocher.
References have also been found to fences being coated with whitewash. While not a particularly high-status coating, whitewash appears to have been used to reflect a sense of domestic order and cleanliness. Two of the period sources found provide references to whitewashing fences, including one which states “a range of black broken paling, which undergoes a periodical whitewashing every now and then” and another that writes “less than a bushel of lime whitewashed the paling” Interestingly none of the authors writing on landscape design or agricultural reform found so far advise using whitewash on fence, suggesting that it was reserved for low-class or urban locations and not as effective a preservative as tar or linseed oil.
Linseed oil also appears to have been a common coating to preserve and decorate pale fences. In 1815 Samuel Park says that fences “might be finished with white lead and oil” and an 1823 passage states that it was “customary” to coat the portion of fence post that was to be below ground “with a strong coat of coarse oil paint.” While references to pails painted with linseed oil are fairly common, they almost always date to post-1825. The increase in the recommendations in the use of oil-based paints for exterior work after 1824 appears to be tied to the increased availability, and lower price, of linseed oil. The landscape gardener John C. Loudon, in his 1825 Encyclopedia supports this idea by stating. “There are now coarse oil paints sold of all colours so cheap as to enable persons erecting palings or other works of wood to paint them at a small expense.” Another reference has been found in an 1847 literary magazine that describes a return visit to a house after many years stating that the “rough fence of brown bars” had been “replaced by a white paling.” The mention of “brown bars” suggests that the house originally had tarred post-and-rail fence that was later replaced with a painted white-pale fence.
While many references were found that recorded white painted fences, dark green was the most common color recommended by landscape gardeners and agricultural reformers for painting pale fences during the period. Writers of the period state that the “invisible” green color was applied to help disguise the fence and give the surrounding landscape an unbroken appearance. Rising out of the English landscape movement, this same desire to unify the countryside would also inspire the development of haha fences. Period sources that recommend painting fences green almost always suggest the color for its ability to partially disguise the fence within the larger landscape. The earliest reference found for this use was in 1785 when William Marshal writes that “besides the sunk fence, another sort of unseen barrier may be made, though by no means equal to that especially if near the eye. This is constructed of paling painted of the invisible green.” Similarly, in 1811 William Mason writes that green paint on fences, “if used with judgment, will be found fully answerable to the most sanguine expectations. The limits, as it were, retire from the view, and use and beauty, which seemed to have suffered a momentary divorce, are now indistinguishably united again.” Mason goes on to say that the main objective for painting fences green was to make sure the barrier gave off “no tawdry glare” that would disrupt the scenery. Finally, in 1825 Loudon writes, “Painted green, or even with the paint called blue anticorrosion (ground glass and oil chiefly), or coated over with the pyrolignous liquor from the gas works, such fences are not obtrusive and less liable to suggest ideas of limitation, confinement, restraint, etc.” While Loudon does appear to be confused on the nature of pyrolignous tar, his words relate that fences were still being painted green throughout the first quarter of the 19th century, documenting the use of green paint on fences for an approximately 30 year period. In her report, Patrick also illustrates an undated period picket from Maryland with a green top and white painted body. Even though the white body of the pale would make it stand out in the landscape instead of receding, this example does show that green paint was applied to pales in the United States.
In addition to the effect of a green painted fence on the landscape, fences also appear to have been painted green for more practical reasons. Specifically, when talking about a green paint made from pigmented tar, another author writing on gates and fencing in 1801 states, “I have lately used the ‘invisible Green Coal Tar Paint,’ prepared at the ‘British Colour Manufactory,’ for ornamental fences, which I think is a good colour, as well as a cheap and effectual preservative of the wood and iron.” Linseed oil paint, while not as effective a preservative as tar, would have also helped to mitigate the effects of water and sunlight on the wooden fences.
Finally, in her report Patrick mentions that fences in Virginia were painted Spanish brown. No evidence of this pigment has been found in the period writings and no sources or illustrations are found in Ms. Patrick’s report. However, one reference to painting the fence enclosing the Pohick church red was found and, since Spanish brown was almost universally used in houses and more finished buildings in the early- and mid-1800s, its use on fences and gates would be logical.
A search through period journals and books was also made for period examples of white posts and black pickets similar to those depicted in the de Nueville watercolor. Two documentary sources were found that describe a polychromed fence; one which used different color finishes imitating higher value materials and a second that used white posts as a functional measure to make a dark green fence visible to carriages drivers and riders at night.
The first reference comes from a report assembled by an 1839 committee charged with improving the New Haven Connecticut burying ground. After deciding against a stone fence because of the cost and a closed board fences due to maintenance issues, they decided on what they called an “open baluster fence.” The fence consisted of 2-1/2 inch square balusters mounted into railings that were supported by obelisk shaped cedar posts. However the committee still had pretensions for a higher status fence and recommended painting the balusters and railings the “black of bronze” and coating the posts with a white colored sanded finish to simulate stone. While the color scheme of this fence is similar to what the Baroness is depicting, the fact that they are specifying balusters suggests that flat pales would have made less convincing stand-ins for brass balusters.
The second reference to a polychrome fence comes much closer to what is seen in the 1818 watercolor. In his 1801 version of an essay on hanging and fastening gates and wickets, Thomas Parker recommends painting “ornamental fences” a dark green color called “invisible Green Coal Tar Paint”. However, he also cautions that the gate posts for “invisible green” fences installed along roads should be painted white to prevent them from being damaged by horses and carriages running into them at night. While Parker is writing almost exclusively about gates and wickets, the same logic would apply to fence panels as well. It is also interesting that Parker refers to using invisible green paint on ornamental fences, indicating that even elaborate fences in the period were painted green to recede into the landscape when seen from a distance.
Composition of Period Fence Finishes
As mentioned above there were two types of tar in the late-18th and early-19th centuries; wood (pyroligneous) tar made from destructively distilling pine and coal tar that was produced as a by-product of gasifying coal. Both are still used today to preserve rail road ties and telephone/power poles, although both materials are more often referred to as creosote. The color most commonly associated with tar coatings is brown, with additional coats giving a darker, almost black, color. An early 19th-century source recommends the following application method for tar:
“It should be gently heated in an iron pot, and put on with a brush. When first put on, it gets so completely into the wood that many who have tried it have supposed that it would be of little use. It soaks in, and seems to leave no body, as the artists express it: but if the work thus treated be examined after it has been exposed for some days to the sun, its surface and the texture of the wood will evidently be much altered; for it will now be found so impresious and hard, that it will be very difficult to make any impression upon it.
If a second, and especially if a third coat of this tar be put upon the wood, it will then bear out , as the painters call it, sufficiently well’ and I have reason to believe that it will preserve all outside wood-work much more effectually then it can be by any other expedient that has hitherto been contrived for this very important purpose for ornamental paling, and indeed for all good outside work, a first, or perhaps a second coat of this tar might be used with great advantage; and when these shall be complete dry, the work might be finished with white lead and oil, as usual.”
The last sentence of the quote is also suggestive of how these coatings would have been used and indicates that tar was also used as a primer for exterior woodwork that was to be finished with linseed oil paint.
Finally, it also appears that pigments were added to tar in order give it a less “objectionable” color. In his 1845 “Encyclopedia of Domestic Economy”, the American author Thomas Webster gives several recipes for pigmenting tar. The first recipe is for a “tar varnish” and includes mixing coal dust or powered charcoal to give the cured coating an opaque black color or Spanish brown to make a brownish coating. The second calls for “beating twelve pounds of resin in a mortar and adding to it three pounds of sulphur and twelve pints of whale oil. This mixture must then be melted over a fire and stirred well while it is melting. Ochre of any required colour ground in oil may be put to it. This composition must be laid on hot and when the first coat is dry, which will be in two or three days, a second coat may be given and a third if necessary.” Finally, Webster also suggests that to make a “very cheap and durable green paint for iron rails and coarse woodwork” that yellow ocher could be added to “gas tar.” This may also be very similar to a green tar paint referred to as “Lord Dundonald’s coal varnish.” Webster’s reference to using tar paint only for “course woodwork”, also suggests that by 1845 tar was no longer a popular choice for ornamental work. It is also interesting to note that a green paint made from tar pigmented with ocher would produce a much more stable color then a green paint made with verdigris, which would turn almost black in five to ten years.
The simplest recipe for white wash is to mix quick or slacked lime with water to the desired consistency. However, whitewash recipes can also be very complex and many suggest additives for increasing its durability. Common additives include alum, salt, sugar, blood, rice, milk, linseed oil and hide glue. An 1837 reference to a compounded white-wash recipe that was specifically intended for paling is included below: “…take three parts of this air -slaked lime, two of wood-ashes, and one of fine sand; pass them through a fine sieve, and add as much linseed oil to the composition as will bring it to a proper consistence for working with a painter’s brush.” While a test batch of the wash has not been made, the recipe would result in a grayish coating that, because of the inclusion of sand, might have been used to approximate White Haven sandstone.
Linseed-oil paint for architectural uses is a versatile material that was colored with a relatively (by today’s standards) small number of pigments in the late-18th and early-19th century. Many sources exist that detail how linseed-oil paints were made and applied, but two of the references found were explicitly recommended for fences. These include one reference to straight “white-lead and oil” and a recipe for an oil based “invisible green” colored paint. The author of the oil-based paint recipe, William Mason, unfortunately only lists the pigments needed, stating that the actual color “must always depend upon the circumstances of the scenery in which the paint is made use.” The pigments he mentions are “white lead, oker [ocher], blue black and a proportionally small quantity of verdigrease [verdigris].”
Conclusion and Draft Specifications
Based on the existing documentary sources it appears that it was common to have tarred or whitewashed fences in the late-18th and early-19th centuries in England and the United States. While linseed-oil based paints also were used to finish fences in the period, they appear to have become more common later in the 19th century due to an increase in the availability of cheap, industrially produced linseed oil. Both tar and linseed oil paints were also mixed with pigments to make a variety of colors, including white, reds, blacks, browns and greens. Green fences were especially being recommended for landscape gardeners attempting to emulate the rural English landscape tradition. Because other elements of the circa 1818 landscape at Montpelier also were influenced by this tradition, and because Madison had access to period books on English gardens, it appears very likely that Madison‘s fence pales and rails were painted the deep “invisible” green color during de Neuville’s visit.
Whether Madison used a green paint that used tar or linseed oil for the binder is not known. However, evidence that the fence was probably painted with a linseed-oil based paint can be surmised from the available documentary evidence. Specifically, an account from John H. B. Latrobe’s 1832 visit describes the fence in front of the Montpelier mansion as being “very handsome”. Latrobe, whose other descriptions of elements at Montpelier are not always as complimentary, was observing what he felt was a fairly high status fence that would have been finished with an equally high status coating. Since tar coatings were often seen as more utilitarian in the period, it can then be inferred that Latrobe was looking at an oil-based paint. A linseed-oil based paint that was pigmented green with verdigris would also darken naturally into an almost black color, which might help to explain why de Nueville would have depicted the pales as black in her watercolor.
The exact color of the posts can also be debated. In both the de Neuville and Chapman images the posts appear to be painted white. While white was a common color for fences in general, it also appears to have been common to paint posts an off-white, light gray or buff color to imitate stone, with sand occasionally added to the paint to complete the effect. Because no physical evidence for the post finish survives, the best choice of color would be white to match the description of white posts that were recommended to demarcate a darkly painted fence at night. The choice of white would also be the more conservative choice from a restoration perspective. A sanded, stone colored post would imply that Madison attempted to imitate a brass-panel, stone-post fence, which is an assumption that cannot be documented at this time.
Gardiner Hallock – Montpelier Department of Architectural Research – 11/9/07
1 Charles Varle. A Complete View of Baltimore: With a Statistical Sketch. Samuel Young, Baltimore. (1833) 61 .
2 The Columbia Electronic Encyclopedia, Sixth Edition Copyright © 2003, Columbia University Press. Licensed from Columbia University Press. All rights reserved.
3Vanessa Patrick. “Partitioning the Landscape: The Fence in Eighteenth-Century Virginia.” Colonial Williamsburg Department of Architectural Research (unpublished). 1983
4 Chemical Essays and Encyclopedia Britannica; Or A Dictionary of Arts, Sciences, and Literature; Enlarged and Improved. Archibald, Constable and Company Edinburgh. 1823. p. 275
6 The new monthly magazine and Literary Journal London (January to June, 1825), 495. Sarah Joseph Buell Hale. American Ladies’ Magazine. J.D. Row, Boston (1834) 59
7 Chemical Essays and Encyclopedia Britannica; or a Dictionary of Arts, Sciences, and Literature; Englarged and Improved. Archibald, Constable, and Company: Edinburgh, 1823. p. 275
8 The vistor or Monthly Inspector. Religious Tract Society (1847) 143.
9. William Marshal. On Planting and Rural Ornament: A Practical Treatise. W. Bulmer, Loundon (1785) 258-259. 258
10. Chemical Essays and Encyclopedia Britannica, 275.
11. Allen Brown’s correspondence, Montpelier Department of Archaeology
12. Chemical Essays and Encyclopedia Britannica, 275.
13. Thomas Webster. Encyclopaedia of Domestic Economy. Harper & Brothers, New York (1845) 78
14. Chemical Essays and Encyclopedia Britannica, 517
15. Maria Eliza Rundell. The new family receipt book. John Murray, London (1837) 84-85
16. Chemical Essays and Encyclopedia Britannica, 275. William Mason The Works of William Mason. T. Cadell and W. Davies, London 1811 349
17. Mason, 349.