In 2024, MAYS GEMS ran a blind test: fifteen members of the public, blindfolded, feeling six bangles against their cheek to test the oldest jade folklore there is — that real jade feels cold. One person picked the genuine piece correctly. That's a 6.7% success rate, worse than chance, testing the single most repeated jade authentication tip in circulation. Most participants had no prior knowledge of jade at all, which makes the result less about expertise and more about the test itself: the belief that you can feel the difference doesn't survive contact with a blindfold. It's why every piece in our jade collection ships with a laboratory report rather than my assurance.
I'm Hong, founder of BMjade. Nine years buying at the Hpakant rough auctions and finishing in our Kunming workshop, and I still send stones out for testing — including stones I was confident about and was wrong about.
Home tests are genuinely useful. They just don't do what most guides claim. They eliminate obvious imitations quickly and cheaply: plastic, glass, some dyed material. What they cannot do — at all, ever, by anyone — is confirm that a piece is natural untreated jadeite, because the treatment that matters most sits invisibly inside the stone.
This is the universal guide. It covers what "real" actually means (three separate questions, and most guides collapse them into one), which tests genuinely work, which are folklore, and what the evidence actually says about each. For tests specific to your piece — a bangle behaves completely differently from a set ring — the four format guides are linked at the end.
One number to hold onto while you read: the difference between natural Type A jadeite and treated material of similar appearance is five to over a hundred times in price. That's the size of the question every test below is trying to answer.
The short answer
Screen at home, verify at a laboratory. Home tests rule out plastic, glass and some dyed material in a few minutes. Only instrumental testing can confirm natural Type A status, because polymer impregnation is invisible to every sense you have.
| Step | What it does | What it costs |
|---|---|---|
| 1. Light and loupe | Rules out glass bubbles, catches dye in cracks | Free |
| 2. Weigh it | Rules out quartzite, serpentine, glass, plastic | A kitchen scale |
| 3. Laboratory report | Settles species and treatment | About $20–35 |
| 4. Verify the report | Catches forged certificates | 90 seconds |
Steps 1 and 2 tell you what a piece isn't. Only step 3 tells you what it is. Skipping step 4 makes step 3 worthless.
"Real" is three questions, not one
Most confusion in this subject comes from collapsing three separate questions into the word real.
1. Is it jade at all? Meaning jadeite or nephrite, rather than quartzite, serpentine, glass, resin or aventurine. This one home tests handle reasonably well.
2. If it's jadeite, has it been treated? This is the Type A / B / C question, and it's the expensive one.
3. Is it any good? Color, texture and translucency. A certificate doesn't answer this at all, and neither does any test. How jade quality maps to price.
The treatment grades
| Grade | What was done | What it means |
|---|---|---|
| Type A | Nothing beyond cutting, carving, polishing and a thin surface wax | Natural. Holds value |
| Type B | Acid-bleached to strip staining, then impregnated with polymer resin | Structure permanently weakened. Resin yellows and crazes over years |
| Type C | Dyed | Color isn't the stone's. Fades, pools in cracks |
| B+C | Both | Both problems |
A thin beeswax polish is accepted practice on Type A and does not disqualify a piece. The distinction between surface wax and deep resin impregnation is a laboratory judgment, and how the instruments make it is a separate article.
This detection method has been standard since the early 1990s, when GIA researchers published the foundational identification work on bleached and polymer-impregnated jadeite — the paper that first gave laboratories a reliable way to catch what your hands can't.
The thing to internalize: Type B is real jadeite. It isn't a fake stone. It's a real stone that has been chemically damaged and refilled, which is exactly why your hands can't detect it — in every way your senses can measure, it is jade.
The tests that actually work
Four, in order of how much they tell you per minute spent.
1. Weigh it
The single most informative home test, and the one almost nobody does.
Weigh the piece dry. Weigh it again suspended in water. Divide the dry weight by the difference between the two. That's specific gravity, and it separates species and catches most substitutes.
| Material | Specific gravity |
|---|---|
| Jadeite | 3.30 – 3.38 |
| Nephrite | 2.90 – 3.03 |
| Quartzite (incl. dyed "Malaysia jade") | ~2.65 |
| Serpentine ("new jade") | ~2.55 |
| Glass | ~2.5 |
| Plastic / resin | 0.9 – 1.4 |
A kitchen scale accurate to 0.01g and a cup of water. Ten minutes. It rules out four of the five common substitutes outright.
What it can't do: anything about treatment. Type B jadeite has essentially normal density.
2. Backlight it
A strong flashlight held against the piece, moved slowly.
Natural jadeite transmits unevenly — clouds, fibrous drifts, and structure that shifts as the light moves. Glass shows round bubbles under a loupe. Dead, perfectly even transmission across an entire piece means something standardized it, and nature doesn't standardise stones.
What it can't do: distinguish Type A from Type B. Both transmit.
3. Loupe the surface
A 10x loupe on the surface reflection, angled to the light.
You're looking for two things. Cui xing (翠性) — the frost-fleck glints of jadeite's crystal grains catching light. Present is consistent with natural. And the etch web — spiderweb lines or orange-peel dimpling that acid bleaching leaves and polish can't fully bury.
One critical rule: treatment is never local. Find etch marks anywhere on a piece and the whole piece is Type B.
4. Look at the dye
Dye concentrates. It pools in surface cracks, gathers in drill holes on beads, and follows fissures rather than sitting evenly in the body of the stone. A 10x loupe on any crack or drill hole is the fastest dye check there is.
What it can't do: catch polymer impregnation, which is colorless.
The tests that don't work
Every one of these appears in high-ranking guides. Here's what each actually measures.
| Test | What it claims | What it actually is |
|---|---|---|
| The cold test | Real jade feels cold | Thermal conductivity. Glass, quartzite, serpentine and Type B jadeite all feel cold. Proves nothing |
| Water drop / float test | Real jade sinks fast | Density, measured badly. Use a scale, not a glass of water |
| Scratch test | Real jade resists steel | Partly true and it damages your piece. Nephrite at 6–6.5 can mark. Quartzite at 7 won't. Backwards results are common |
| Burning hair / lighter test | Real jade doesn't scorch | Thermal damage to the wax finish and, on treated material, the resin. It can wreck a genuine piece |
| Hanging a hair and burning it | Folklore | Nothing at all |
| "It feels heavy" | Density by hand | Useless without a reference piece to compare against, and even then unreliable |
The sound test: it depends entirely on the format
This one gets a section because a flat verdict is wrong in both directions.
On a bangle, tapping works. A jade bangle is one continuous ring of stone with no setting, no glue and no metal. It can resonate freely, and the difference between the clear sustained note of a sound bangle and the dull thud of a cracked or heavily treated one is genuinely audible. It's one of the few traditional tests that survives scrutiny, and it survives only here. The bangle method in full.
On rings, earrings and set pendants, it's useless. A small stone clamped in metal cannot resonate. Whatever you hear is the setting.
If a guide tells you the sound test works, or that it doesn't, without asking what shape your piece is, that guide hasn't thought about it.
UV light: useful, but only in one direction
The UV test is widely described as a reliable home check for Type B. It isn't reliable, but it isn't worthless either, and the precise shape of its usefulness is worth getting right.
Use a 365nm long-wave UV flashlight — the kind sold for currency checking. Not 395nm, which is the common cheap option and gives poorer results.
The polymer resin in Type B jadeite commonly fluoresces a chalky blue-white under long-wave UV. Natural Type A jadeite is usually inert or very weak. Sweep the light across the stone, and on a set piece pay particular attention to the setting edge, where excess adhesive also glows.
The asymmetry is the whole point:
A glowing piece is a red flag. A dark piece is not an all-clear.
Some treatments barely fluoresce. Some natural stones react faintly. Traditional beeswax finishing on a genuine Type A piece can produce a response. Laboratories use UV fluorescence as a screening step that flags a stone for instrumental testing — never as a conclusion.
Treat it as supporting evidence, in one direction only.
The $880,000 lesson
A woman in Shanghai paid the equivalent of about $880,000 for a piece that turned out to be plastic worth roughly $15.
The instinct is to assume she was careless. She wasn't unusually careless — she did what most buyers do, which is trust the seller and the appearance.
Put that alongside the blind test at the top of this article. Fifteen people who handle jade, six bangles, one correct answer. The failure rate isn't a failure of attention. It's a failure of method. Hands are the wrong instrument for this question, and no amount of experience upgrades them.
That's the case for laboratory testing stated as plainly as I can state it, and I'd rather you took it from a study and a court case than from a jade seller.
The certificate isn't automatically the answer
One more trap, and it's the one that's grown fastest.
Forged jade certificates cost a few dollars to produce, and the good ones carry convincing reproductions of CMA and CNAS accreditation marks. Some are more elaborately printed than genuine reports — gold foil, holograms, impressive seals.
So a certificate you haven't verified online is worth approximately nothing.
The verification is short: type the laboratory's address into your browser yourself, enter the report number and anti-counterfeit code, and compare what comes back — material, treatment status, weight, dimensions, photograph — against the physical piece.
Never scan the QR code printed on a certificate you haven't verified. A forged certificate can carry a QR code pointing at a page the forger controls, and it will confirm the document beautifully.
Which laboratories to trust and how to read what they issue.
Which piece do you have?
Everything above applies to any jade. What comes next depends entirely on format, because each one hides and reveals different things.
| Your piece | Why it's different | Guide |
|---|---|---|
| Bangle or bracelet | No setting, no glue, no metal. The only jade that can be backlit around its entire circumference — and the only one where the sound test works | Bracelet authentication |
| Ring | Metal can hide seams, adhesive, treatment clues, or an entirely different material beneath a visible cap. Doublets are a real risk | Ring authentication |
| Earrings | The only jade that comes as a matched pair, and the only one where metal sits inside a piercing. Both facts are authentication tools | Earring authentication |
| Necklace or beads | Beads are read as a population, not individually. Drill holes are where dye gives itself away | Necklace authentication |
Where I'll be straight with you
We sell jade, so weigh this accordingly.
We certify everything, and it costs us very little. Bulk laboratory submission through market channels runs a few dollars per piece. Any seller telling you certification is impractical at your price point has decided not to do it, not been prevented from doing it.
I still get pieces wrong. Nine years in and I send stones out because the honest position is that nobody's hands settle this. A seller who tells you they can always tell is either inexperienced or selling you something.
And home tests are worth doing anyway. Not because they prove anything, but because a seller who becomes evasive when you say "I'm going to weigh it and check the report number" has told you what you needed to know before you ever reach for the scale.
Frequently asked questions
Can you tell if jade is real at home?
You can rule things out. Weight, backlighting and a 10x loupe reliably eliminate glass, plastic, quartzite and serpentine, and catch obvious dye. What no home test can do is confirm that jadeite is natural and untreated, because polymer impregnation sits inside the stone and passes touch, weight, sound and light. In a 2024 blind test, fifteen members of the public — blindfolded, testing jade's oldest folklore claim that it feels cold — picked the genuine bangle correctly 6.7% of the time.
Does real jade feel cold?
Yes, and so does glass, quartzite, serpentine and treated jadeite. The cold sensation is thermal conductivity, and it doesn't distinguish jade from the materials most often sold as jade. It's the most repeated and least useful test in circulation.
Does the water test work for jade?
Not as usually described. Dropping a piece in a glass and watching it sink measures density very badly. Density is genuinely useful, but measure it properly: weigh the piece dry, weigh it suspended in water, and divide the dry weight by the difference. Jadeite sits at 3.30–3.38, nephrite at 2.90–3.03, quartzite around 2.65.
Does the sound test work?
On a bangle, yes. A bangle is one continuous ring of stone with nothing damping it, and the difference between a clear sustained note and a dull thud is audible. On rings, earrings and set pendants it's useless, because a small stone clamped in metal can't resonate. Anyone giving a flat answer without asking your piece's shape hasn't thought it through.
Does Type B jade glow under UV light?
Often, but the test only works in one direction. Polymer resin commonly fluoresces chalky blue-white under 365nm long-wave UV, so a glowing piece is a red flag. But some treatments barely fluoresce, some natural stones react faintly, and traditional wax finishing can respond — so a dark piece is not an all-clear. Use it as supporting evidence, never as a verdict.
Should I scratch-test my jade?
No. It damages the piece, and the results mislead. Nephrite at 6–6.5 on the Mohs scale can take a mark from hardened steel, while quartzite at 7 won't — which means the test can point exactly the wrong way. Weigh it instead.
Is Type B jade fake jade?
It's real jadeite that has been acid-bleached and impregnated with polymer. So it's genuine jade material, chemically altered, and worth a small fraction of untreated material of similar appearance. It isn't an imitation, which is precisely why home tests can't catch it.
How do I know a jade certificate is genuine?
Verify it online yourself. Type the laboratory's address into your browser rather than scanning the QR code on the document, enter the report number and anti-counterfeit code, and compare the returned record against the physical piece. Forged certificates cost a few dollars and often carry convincing accreditation marks, so an unverified certificate provides no protection.
Conclusion
The uncomfortable finding at the top of this article is the most useful thing in it: fifteen blindfolded people, six bangles, one right answer, testing the exact claim — "real jade feels cold" — that shows up in nearly every other guide to this subject. Your hands are not a bad instrument for this. They're the wrong instrument, and so are mine.
So screen at home to eliminate the obvious, then let a laboratory answer the question that determines the value, and verify the report yourself. That sequence costs about thirty dollars and ninety seconds, and it's the whole of what protection in this market looks like.
If you have a piece and don't know where to start, send me a photograph and its weight at jadeworldchina@outlook.com and I'll tell you what to check first. — Hong
