Finding out what finish is on your furniture
A cotton swab, three solvents and an inconspicuous corner will usually name the finish, or name the group it belongs to. Nearly everything you may safely do to a surface follows from that answer.
- Start here: the essentials
- 1. Why the finish decides what you may do
- 2. Before you open anything
- 3. Where to test
- 4. Wax comes off before anything else
- 5. The sequence
- 6. What a negative result means
- 7. See it: three pieces tested
- 8. Practice: the hidden-corner test
- 9. Your piece
- 10. Check your work
- 11. Knowledge check and scenario
Start here Essentials
At minimum this chapter asks you to put three solvents on one hidden spot, in order, and write down what each one did. The reasoning behind the order and the finish-by-finish table are Go deeper on a first pass.
- Pick a place nobody will ever see, and read the safety note first. An inside face, an underside, the back of a leg (section 3, and the solvent note in section 2).
- Wipe it with mineral spirits on a swab and see what comes off. Wax comes off first, and until it does you are testing the wax (section 4).
- Try denatured alcohol, then lacquer thinner, in that order. One fresh swab each, one small spot each (section 5).
- Write down what happened, including nothing happening. Nothing happening is a result (your condition report).
Minimum time: about 20 minutes, plus drying. Done when: your report names a finish or the group it belongs to, says where you tested, and says whether the finish softened.
1. Why the finish decides what you may do Understand
The finish is the layer between you and the wood, and almost every reversible thing in this manual is done to it rather than to the wood. So knowing which one it is settles a long list of questions at once: whether water may touch the surface, what a white ring probably is, whether wax will stick or sit in a smear, what a cleaning solvent will do, whether the piece can be handled by you or belongs to somebody with training.
Finishes fall into a few families, and they behave differently because they cure differently. Physical principle Shellac and lacquer are evaporative: the resin is dissolved in a solvent, the solvent leaves, and the film that remains will redissolve in that same solvent for as long as it exists. That is why they can be tested, and it is also why they are vulnerable to the liquids people put on tables. Varnish and the cured coatings that came after it are reactive: they change chemically as they cure, and the cured film does not redissolve in the solvent it came in. That is why they resist the test, and why a damaged one is so much harder to mend invisibly. Oil and wax finishes are different again: rather than forming a film on top, they sit in and just above the surface.
The consequences run through the rest of the manual:
| If the finish is | Cleaning | Waxing | What to watch for |
|---|---|---|---|
| Shellac | Keep water off it; the published routine uses mineral spirits on a cloth | Suitable, on a sound stable film | Alcohol in any form, including in drinks and in cleaning products |
| Lacquer | The published cleaning routine applies | Suitable, on a sound stable film | Crazing as the film ages; treat a crazed surface gently |
| Varnish or a newer cured finish | The most tolerant of the published routine | Named by conservation guidance as the best maintenance for clear-varnished furniture | Little, which is why it is also the hardest to identify |
| Oil or wax | Sparingly; mineral spirits keeps lifting wax for as long as you keep wiping | Wax over wax is ordinary; wax over bare wood is not recommended | Confusing a wax layer with the finish underneath it |
| Paint | Conservation guidance stops at dusting unless you are trained | Not recommended on painted surfaces | Paint that may predate 1978: see What can hurt you in old furniture |
| Unfinished | No water: it raises the grain and drives soil into the pores | Not on bare wood | Anything liquid, including the test solvents |
The full version of that table, with what each finish tolerates and the stop conditions, is Identifying the finish, which is this chapter's reference page.
2. Before you open anything
You are about to open three solvents. None of this is exotic, and all of it has a label that should be read.
The National Park Service's own instruction to staff cleaning wood furniture is to follow all safety instructions on the solvent containers, to obtain and read the safety data sheets for every solvent used, and to ensure ample ventilation throughout the cleaning and the drying process. Drying is included deliberately: the vapor does not stop when you put the lid back on.
Mineral spirits, sold as paint thinner and known industrially as Stoddard solvent, is a combustible liquid. NIOSH's pocket guide entry for it gives a flash point of 102 to 110 °F and classes it as a Class II combustible liquid, lists eyes, skin, respiratory system, central nervous system and kidneys as target organs, and says to prevent skin and eye contact. Denatured alcohol and lacquer thinner are more volatile still. Work away from flames, pilot lights and heaters, keep the containers closed between uses, and do not do this in a closed room.
Rags and swabs used with these solvents, and with any oil finish, go flat and outdoors. The reason is in Cleaning furniture, waxing it, and reviving a tired finish, and it is the one instruction in this manual about a fire.
What you need: mineral spirits, denatured alcohol and lacquer thinner, in small quantities; cotton swabs; a few clean white cotton rags; and somewhere ventilated to work. White rags matter, because the color that comes off onto them is part of the result.
3. Where to test
The test removes a very small amount of finish where it works. So the place you choose should be a place where that does not matter and where nobody will look.
Good places, in rough order:
- The inside face of a rail or apron, where it faces into the piece.
- The underside of a top, an inch or two in from the edge.
- The back of a leg, low down.
- The inside of a door, near a hinge.
- The back of a case, where the back carries a finish.
- Under a drawer runner.
Poor places: anywhere on a top, any front surface, anywhere near an edge that catches light, and anywhere the finish is already damaged, because a damaged spot tells you about the damage rather than about the finish.
One qualification. The whole method assumes the spot you test is representative of the piece. On a piece that has been repaired, partly refinished or assembled from parts, it may not be. If the finish on the inside of a door behaves differently from the finish on a drawer front, that is worth knowing and worth writing down, and it belongs with the conflicting clues in How old your furniture is.
Work small. A spot the size of a pea is enough to see a reaction. Use a fresh swab for each solvent, because a swab that has been in mineral spirits will carry it into the alcohol and confuse the result.
4. Wax comes off before anything else
The first wipe is mineral spirits, and it is doing a different job from the two that follow. Mineral spirits dissolves wax. So the first wipe tells you whether there is a wax layer on top of the finish, and it has to come first, because until the wax is off you are testing the wax rather than the finish.
What to watch for: a dulling or clouding where you wiped, a waxy smear on the swab, and a change in the sheen of that spot. If you get any of those, there is wax on the surface. Keep wiping that small area with fresh swabs until nothing more lifts, then let it dry before going on.
Convention Flexner, whose published sequence this follows, makes the same point about wax in a second article: wax cannot be neutralized, which is what some product labels claim, and has to be washed off with a petroleum-distillate solvent. Mineral spirits is one. So the first wipe is both a test and the beginning of the cleaning in Cleaning furniture, waxing it, and reviving a tired finish.
Two results from the first wipe matter beyond the wax question.
Dirt comes off, and the finish does not change. Good: the finish is behaving, and you have learned that mineral spirits is a safe cleaning agent for this surface, which is the National Park Service's test for exactly that.
The finish itself softens, goes tacky, or color comes off onto the swab beyond ordinary dirt. This is the stop condition, and it is the most important sentence in this chapter.
The Park Service's instruction here is unambiguous and this manual carries it unchanged: if the mineral spirits softens the finish, do not attempt to clean the piece, and consult a furniture conservator for professional assistance. It is not an instruction to carry on more gently, or to try a weaker solvent, or to work faster. It is a stop.
A finish that softens under the mildest of these three solvents is telling you it is fragile, degraded, or something other than what you expected. What happens next is a conversation with someone who can look at it, which is Who to hire for furniture work. Write the result on your report and leave the surface alone.
5. The sequence
With the wax question settled and the spot dry, the test proper is two more solvents, in this order.
| Step | Solvent | If it reacts | If nothing happens |
|---|---|---|---|
| 1 | Mineral spirits (also naphtha or turpentine) | Wax on the surface. Keep going until no more lifts, then dry and continue | No wax. Continue |
| 2 | Denatured alcohol | Shellac. The surface goes sticky or the finish dissolves | Not shellac. Continue |
| 3 | Lacquer thinner | Lacquer, if the finish dissolves or goes sticky. A finish that goes soft and tacky rather than dissolving may be water-based | Varnish or one of the newer solvent-resistant finishes: see the next section |
How to run each step: a fresh swab, lightly wetted rather than dripping, held on the spot for a few seconds, then lifted and looked at. Look at both the surface and the swab. A reaction shows as the surface going soft or sticky, as the finish visibly dissolving, or as finish color coming off on the swab.
Give each step a moment. A reaction is often not instant, and wiping vigorously immediately will take off dirt and tell you nothing. Let the spot dry between solvents.
Why the order is what it is
The sequence runs from the mildest solvent to the strongest, and from the outermost layer inward, and those two happen to coincide.
Mineral spirits is the mildest of the three and dissolves wax but not the film finishes, so it strips the top layer without touching what is underneath. Denatured alcohol is shellac's own solvent and does not dissolve lacquer or cured varnish. Lacquer thinner is the strongest, a blend that will dissolve lacquer and will also attack shellac, so running it first would give you a reaction that could be either and tell you nothing.
The practical rule generalizes beyond this test, and it is the one conservation sources keep repeating: start with the weakest thing that might work. The Park Service's waxing leaflet gives the same preference in a different context, advising weaker organic solvents such as turpentine and mineral spirits and advising against waxes carrying the stronger solvents xylene and toluene, which it says can damage some finishes and are more of a health hazard.
The finishes and what each tolerates
| Finish | How it cures | Redissolves in | Typically damaged by |
|---|---|---|---|
| Wax | Nothing: the solvent evaporates and wax remains | Mineral spirits, naphtha, turpentine | Heat, abrasion, and anything that dissolves it |
| Shellac | Solvent evaporates; the film stays soluble | Denatured alcohol; attacked by lacquer thinner | Alcohol, water, heat |
| Nitrocellulose lacquer | Solvent evaporates; the film stays soluble | Lacquer thinner | Solvents, and crazing as it ages |
| Oil-based varnish and polyurethane | Cures chemically into a film that does not redissolve | Nothing in this test | Abrasion, light, and eventual embrittlement |
| Catalyzed, polyester and UV-cured finishes | Cure chemically, often very hard | Nothing in this test | Impact and abrasion rather than solvents |
| Water-based finish | Coalesces as water leaves | Goes soft and tacky in lacquer thinner, toluene, xylene | Solvents, and some cleaning products |
| Oil finish | Oil cures in and just below the surface | Not a film to dissolve | Drying out in appearance, and wear through to bare wood |
Dating from the answer
The answer carries a date range, and How old your furniture is uses it as the one dating clue in this manual with a published source behind it. The short version: shellac widely available from the 1820s and dominant for about a century; lacquer from the 1920s; catalyzed finishes from the 1960s; water-based from the 1980s and 1990s. All of it subject to the assumption that the finish is original.
6. What a negative result means
Suppose nothing happens. No reaction to any of the three. This is a common outcome and it is worth being precise about what it means, because it is where people overclaim.
It does not mean "varnish." Flexner's own wording is that the finish is varnish or one of the newer more-solvent-resistant finishes, and that is a group, not a name. The group includes oil-based varnish, polyurethane, catalyzed conversion finishes, polyester and UV-cured coatings. The earliest and the latest of those are separated by most of a century, and they behave differently from each other in wear and in repair.
A negative result is a group, not an answer, and the honest entry on your report is the group. Writing "varnish" when what you established was "none of these three solvents touched it" converts a real observation into a guess, and the next person, perhaps a professional you hire, will work from what you wrote. For almost every purpose in this manual the group is sufficient: it tells you the surface is solvent-resistant, which is what the cleaning and waxing decisions turn on. Where it is not sufficient, a conservator can narrow it further with tests this manual does not teach.
The second thing a negative result might mean is more interesting: it might mean the piece has been refinished. A modern cured coating on a piece whose construction looks much older is a strong hint that somebody stripped it, and that hint should go on the report beside the conflicting clues in How old your furniture is.
7. See it: three pieces tested See it
Dana's walnut chest. Tested on the underside of the top rail, where the rail faces into the case.
What happened: mineral spirits lifted a little gray dirt and no wax smear. Denatured alcohol made the spot sticky within a few seconds, and a faint brown came off on the swab.
Known: the finish on that rail is shellac.
Assumed: that the rest of the piece carries the same finish, which Dana did not test, and that the shellac is original rather than a later recoat.
What it changes: under those assumptions, water stays off this piece. That rules out the soap-and-water step of the published cleaning routine, and it makes the white ring on the top a likely alcohol or water mark in a shellac film rather than a stain in the wood, which raking light in Surfaces, veneer and hardware can test further.
On the report: "Finish: shellac. Tested underside of top rail with mineral spirits (no wax, light soil), then denatured alcohol (softened within seconds, faint color on swab). Did not soften under mineral spirits. Other surfaces untested."
Marcus's four oak chairs, bought for $60 at an estate sale. All four have a sticky, gummy surface.
What happened: on the inside face of a rear seat rail, mineral spirits lifted a brown, greasy film onto the swab, and kept lifting it onto a second and third swab. Under it the surface was dull but firm. Alcohol did nothing. Lacquer thinner did nothing.
Known: there was a thick accumulation on the surface that mineral spirits removes, and the finish under it did not react to the two stronger solvents.
Assumed, and it is Marcus's assumption: that the gummy layer is years of aerosol polish rather than a finish breaking down. The mineral spirits wipe is what tests this, and the fact that the surface under it was firm rather than soft supports the assumption without proving it.
Still open: whether the polish contained silicone, which matters for anyone who later tries to put a new finish on these chairs. Cleaning furniture, waxing it, and reviving a tired finish explains why that question outlives the chairs' current owner. The Park Service's own leaflet on silicone speaks of it being known or suspected rather than observed, so suspected is what goes on the report.
On the report: "Finish: no reaction to denatured alcohol or lacquer thinner, so varnish or a newer solvent-resistant finish. Heavy greasy surface layer removed by mineral spirits over three swabs; silicone suspected, not established."
Priya's sideboard. Mid-century, the finish gone gray and dry.
What happened: mineral spirits darkened the spot noticeably while it was wet and it went pale again as it dried. No film dissolved under any of the three solvents, and no distinct layer came off on a swab.
Known: there is no solvent-soluble film finish here.
Assumed: that this is an oil finish rather than a very hard cured coating. The evidence for it is the darkening-and-returning behavior, which is a surface absorbing and releasing solvent rather than a film sitting on top, together with the era and the dry gray look.
Still open: whether what remains is oil, or oil that has largely worn away leaving bare wood. That changes what the cleaning chapter permits, because bare wood and an oil finish take different treatment, and pressing the question further is work for someone who can handle the piece.
The two published sources behind this chapter:
- Using solvents to date furniture (Bob Flexner, Popular Woodworking, April 2016): the solvent sequence, in the author's own words, with the period dates and his own statement of the original-finish assumption.
- Cleaning wood furniture (National Park Service, Conserve O Gram 7/1, July 1993): the mineral spirits swab test, the stop condition in the Limits box above, and the solvent safety caution. Written for park staff working on collection objects, which is worth remembering when reading its tone.
- NIOSH Pocket Guide: Stoddard solvent: the flash point, combustible class, target organs and skin-contact guidance for mineral spirits, as cited in the solvent box above.
8. Practice: the hidden-corner test Practice
Minimum: one spot, three solvents, three lines written down. Time: about 20 minutes plus drying between steps. Materials: mineral spirits, denatured alcohol, lacquer thinner, cotton swabs, clean white cotton rags, ventilation, and a place to put the used swabs flat and outdoors.
- Read the labels and open a window. Both of them, and keep the room ventilated while the spot dries as well as while you work.
- Choose and photograph your test spot before you touch it, so you have a before image and so you can find it again.
- Mineral spirits first. Watch for wax, for dirt, and for the stop condition. If the finish softens, stop the whole exercise and write it down.
- Let it dry, then denatured alcohol. Fresh swab, a few seconds, then look at the surface and at the swab.
- Let it dry, then lacquer thinner. Fresh swab again. Note whether a reaction is dissolving or going tacky.
- Write three lines on the report, one per solvent, plus where you tested. Then put the swabs and rags flat on concrete outdoors.
The second-spot exercise. Run the same three steps on a second hidden spot on a different part of the piece: a drawer front rather than a rail, or the inside of a door rather than a leg. The useful outcome is one of two things. Either the two spots agree, which raises your confidence that the whole piece carries one finish, or they disagree, which has just told you that part of the piece was refinished or replaced. That second finding is worth far more than the first and is invisible to anyone who tests once.
9. Your piece Your piece
10. Check your work Check your work
Criteria
- Does your report say where on the piece you tested, specifically enough that you could find the spot again?
- Does it record what each solvent did, including the ones that did nothing?
- If nothing reacted, does the report say "varnish or a newer solvent-resistant finish" rather than "varnish"?
- Does it state separately whether the finish softened under mineral spirits, which is the stop condition and not the same question as which finish it is?
- If there was wax, is that recorded as a separate layer rather than as the finish?
Worked example
Marcus's entry names the spot, records all three solvents including the two that did nothing, gives the group rather than a name, and records the greasy layer separately from the finish under it, with the silicone question flagged as suspected rather than established. Checked against the criteria it passes on all five, and the fifth is the one that matters most for the chairs' future: the layer and the finish are two different things and his report keeps them apart.
Common mistakes
- Testing on a visible surface. The test takes finish off where it works. Hidden means hidden.
- Testing on damage. A worn or cracked area behaves differently from sound finish, so the result describes the damage.
- Skipping the mineral spirits step. If there is wax, the alcohol step tests the wax and the result is meaningless.
- Reusing a swab. Carries one solvent into the next.
- Writing "varnish" on a negative result. Write the group.
- Carrying on after the finish softened. That is the one result in this chapter that ends the session.
- Dropping the swabs in a wastebasket indoors. Flat, outdoors, on concrete.
What is still unresolved
This test does not narrow the solvent-resistant group, does not tell you whether the finish is original, and does not tell you how thick or how sound it is. It also does not tell you anything about the wood underneath, which is What your piece of furniture is actually made of. What to do with the answer is Cleaning furniture, waxing it, and reviving a tired finish, which is why that chapter asks you to read this one first.