The Padparadscha Handbook

There is a particular silence that happens in our Sydney workroom when a padparadscha comes out of the paper. Blue sapphires get an appreciative nod. Rubies get a low whistle. A padparadscha gets a pause, and then, almost always, an argument.

The argument is never about whether the stone is beautiful. It is about what colour it is. One person sees apricot. Another sees the inside of a shell. Someone tilts it a few degrees under the south light and announces that it has gone pink, and someone else takes it to the window and insists it is plainly orange, and both of them are looking at the same stone within the same minute. I have watched experienced people talk themselves in circles over a single three-carat oval, and I have done it myself more times than I would like to admit in print.

That is the whole handbook, really, compressed into one scene. Every other gem variety we sell can be named by anyone with a loupe and a little training. Padparadscha cannot. It is a colour so genuinely contested that an international committee of gem laboratories sat down and wrote a specification for it, complete with a lighting standard, a viewing angle and a list of seven things that disqualify a stone from the name. When professionals need a written protocol to agree with each other, a private buyer with a phone photograph and a seller's enthusiasm has no chance at all — unless somebody explains the rules first.

Why this stone deserves its own volume

We already publish a sapphire handbook. Padparadscha is a sapphire, so in principle it should be a chapter, not a book. It is a book because everything that makes ordinary sapphire buying manageable breaks down here.

With blue sapphire, the name is not in dispute; only the quality is. With padparadscha, the name itself is the single largest component of the price. Call a stone pink sapphire and it sits in one market. Call the same stone padparadscha and it sits in a materially higher one. That is an enormous amount of value resting on a judgement call — and it is a judgement call that two reputable laboratories, examining the same stone with the same care, can reach differently without either being wrong.

Then there is the newer problem, and the reason we rebuilt this handbook rather than reprinting an older one. Some sapphires in this colour range are not simply coloured; they are charged. Their orange component sits in colour centres that fill under ultraviolet light and drain in ordinary daylight over days or weeks. A stone can be a padparadscha in the shop and a pink sapphire by the time it has spent a fortnight on your hand at the beach. The laboratories have known this since the late 2010s and have been tightening their rules ever since, and as of August 2025 there is finally a single harmonised test for it. Almost no consumer guide on the internet has caught up. Chapter IV is the one I would read first if you are shopping this month.

What this book will and will not do

This handbook will teach you to read a padparadscha report line by line, to understand why one laboratory's opinion differs from another's, to tell an ordinary heated stone from a beryllium-diffused imposter and to know why your local jeweller's testing equipment cannot make that separation for you. It will give you a dated, sourced price ladder rather than the vague gesturing most guides offer, and it will show you the auction results that anchor the top of the market — including one figure that two respectable sources report differently, which I have flagged rather than tidied away.

It will not tell you what to buy, and it will not tell you that a gemstone is an investment. We are gem merchants, not licensed financial advisers, and anyone who blurs those two roles is selling you something other than a stone. Where I quote prices and auction results in this book, they are context for a purchase you are making because you want the object — nothing in these pages is investment advice.

I should also say plainly where we stand. Our family has worked the Ceylon gem fields across three generations since the 1960s, so we have a natural affection for the position that padparadscha is a Ceylon word for a Ceylon stone. Chapter III explains why that position is emotionally satisfying, commercially real and gemmologically unsupported, and why we tell buyers the truth about it anyway. If a handbook cannot concede the awkward point about its own tradition, it is a brochure.

Padparadscha is, to my eye, the loveliest thing corundum does. It is also the variety where the gap between an informed buyer and a hopeful one costs the most money. Read the eight chapters, take the checklist in Chapter VII to whoever is selling to you, and enjoy the argument about what colour it is — that part never stops, and it is not supposed to.

— Anna, FGAA
Ministry of Gems, Sydney

From lotus colour to laboratory specification

The word arrives in English by an unusually scenic route. It descends through Sinhalese from the Sanskrit padmaraga — padma, lotus, and raga, colour — and it reached Western gemmology through a German transliteration, which is why an English speaker looking at the spelling for the first time has no idea where to put the stress. Lotus Gemology, whose essay on the ownership of the word is the best single account of its history, traces the whole path; so does the historical survey on Ruby-Sapphire.com. It means, quite simply, the colour of the lotus blossom.

Which lotus, and which part of the blossom, is where the trouble starts. The historic Sanskrit sources are not describing a pale stone at all. The Garuda Purana, compiled somewhere between roughly 400 and 1000 AD, gives padmaraga a crimson colour, and a medieval lapidary of around 1200 to 1300 AD likens it to fire and to molten gold. Those are not pastel words. Anyone who tells you the ancient meaning of padparadscha was a delicate salmon pink is reading the modern definition backwards into the old texts.

Gemmologically, the colour is chromium's doing. Gem-A describes padparadscha as the rarest form of sapphire corundum, sitting on the boundary between pink and orange, coloured by traces of chromium — the same element that at higher concentration makes ruby, and quite distinct from the iron-and-titanium partnership that makes blue sapphire. In the stones that behave themselves, the pink comes from chromium absorption and the orange-yellow from a separate contribution that Chapter IV will get to in detail, because it is the part that can walk away.

The modern, narrow, pastel-leaning definition is largely traceable to one article: Robert Crowningshield's Padparadscha: What's in a name?, published in Gems & Gemology, volume 19 number 1, in 1983. Crowningshield restricted the term to light-to-medium tones of pinkish orange to orange-pink, and explicitly pushed out both the dark brownish orange East African material and the deep orangey-red stones that some dealers had been happily calling padparadscha. That article is why the word means something narrower today than it did in Sanskrit, and it is also why a certain kind of dealer has resented laboratory definitions ever since.

Four decades later the definition is a formal document. The Laboratory Manual Harmonisation Committee — the LMHC — publishes Information Sheet #4, and the current text, version 11, dates from May 2025. It defines padparadscha sapphire as a corundum from any geographical origin whose overall colour appearance is “pinkish orange to orangey pink with pastel tones and low to medium saturation”, judged in standard daylight between D50 at 5000 K and D65 at 6500 K, preferably under a light source with a Colour Rendering Index of 95 or higher.

Read that twice, because there are three separate ideas hiding in one sentence. The first is a hue window: pink and orange, in either order, and nothing else. The second is a saturation and tone ceiling: pastel, low to medium. The third — and this is the one buyers skip — is a lighting standard. A colour claim made under a warm shop spotlight or in a phone photograph with the white balance doing its best is not a colour claim under the LMHC definition at all. Daylight, or a properly specified daylight-equivalent lamp, or nothing.

The seven ways to lose the name

Information Sheet #4 also lists seven circumstances in which the name padparadscha shall not be applied. This list is the most useful page in modern padparadscha gemmology and I have never seen a consumer guide reproduce it, so here it is in plain English.

One: any colour modifier other than pink or orange is present — brown, purple, yellow, grey — including when that modifier is produced by light scattering off tiny inclusions rather than by chemistry. A milky stone that reads slightly brownish because of its own turbidity is out, even though nothing was added to it. Two: major uneven colour distribution is visible to the unaided eye with the stone table up, tilted up to thirty degrees either way. Colour zoning that you can see without magnification disqualifies the name. Three: yellow or orange epigenetic material sitting in fissures — that is, foreign material that arrived after the crystal formed — is contributing to the overall colour. Four: the stone has been treated in any of the ways covered by LMHC Information Sheets #2 and #3, which is to say lattice diffusion or fissure filling. Five: it has been irradiated. Six: its colour is unstable, and shifts out of the padparadscha range during a colour stability test. Seven: it has been dyed, coated, painted, varnished or sputtered.

What is not on the list matters just as much. Conventional heat treatment does not disqualify a stone from the name. The LMHC is explicit that padparadscha applies to both unheated and conventionally heated corundum. Heat changes the price, sometimes dramatically, but it does not change the noun.

Seven laboratories bound, three important ones not

The LMHC wording binds its seven member laboratories: Central Gem Laboratory in Japan, CISGEM in Italy, the DSEF German Gem Lab, GIA in the United States, GIT in Thailand, and Gübelin Gem Lab and SSEF in Switzerland. Those seven have agreed to say the same thing.

Three laboratories whose reports you will meet constantly in this market are not LMHC members and are not bound by that harmonised wording: GRS, AGL and Lotus Gemology. None of them is a lesser laboratory for it — Lotus in particular has done more original padparadscha research than most members — but it is the structural reason two good reports can disagree. If you are holding a GRS report and a friend is holding a GIA report and the variety names differ, you have not necessarily found a fraud. You may simply have found two houses applying two rulebooks.

The pastel war

And the rulebook is contested. In 2012 Richard Hughes surveyed nine experienced padparadscha dealers in Sri Lanka and Hong Kong about how they define the term. Not one of them used the word pastel. Two of them used words like deep and vivid. One summarised market taste geographically, observing that American buyers tend to want a sunset colour while Japanese buyers want the lotus. There was, in short, no consensus at all on where the boundaries sit — among the very people who trade the stones for a living.

So we have a formal definition that says pastel, a body of historic literature that says crimson and molten gold, and a working trade that says neither reliably. Our house position is to report all three honestly rather than pretend the matter is settled. When we describe a stone, we tell you what the laboratory called it, under which rulebook, and what we think you will see when you hold it in daylight. If those three things ever disagree, we say so — that is the whole job.

The awkward point we have to concede first

Let me get the uncomfortable part out of the way. The LMHC definition says, in as many words, that padparadscha is a variety of corundum from any geographical origin. Not Sri Lanka. Any. No major laboratory — not the seven LMHC members, not GRS, not AGL, not Lotus Gemology — restricts the designation geographically. If you are told that a stone cannot be a padparadscha because it came from Madagascar, you are being told a trade opinion dressed as a gemmological fact.

That trade opinion is widely held and sincerely felt, and I understand it better than most. The Sinhalese language named the colour. The reference stones against which every other deposit's material is judged came out of the Ceylon fields. Our own family has been in and around those fields across three generations since the 1960s, and there is a part of me that finds the origin-neutral definition slightly bloodless. Richard Hughes put the sympathetic case as well as anyone: he opposes restricting the term to Sri Lanka, but argues that Sri Lankan voices deserve real weight in defining their own word, and offers the analogy of excluding France from the definition of champagne. Even he stops short of a geographic restriction. So do we.

One more piece of honesty. We looked for a formal published position from Sri Lanka's National Gem and Jewellery Authority restricting the word to Sri Lankan stones and could not locate one. As far as we can establish, the Ceylon-only position is a trade and dealer stance, not a regulatory one. If that changes, we will update this chapter.

Where else the colour comes from

Padparadscha-type material is produced in several places, and knowing them is part of buying sensibly rather than superstitiously.

Sri Lanka remains the reference, and the deposits are old enough to have supplied the stones that formed the definition in the first place. Madagascar arrived in force with Ilakaka from the late 1990s, and then with Bemainty near Ambatondrazaka from around 2016 to 2017 — a deposit that will reappear in the next chapter for reasons you will not enjoy. Tanzania contributes Tunduru material and the Winza find of 2007. Vietnam produces padparadscha-type corundum at Quy Chau.

There are tendencies by source, and Gem-A summarises them usefully: Madagascan stones lean pinker, Tanzanian stones lean browner, and Sri Lankan stones remain the benchmark. Tendencies are not rules. I have handled Madagascan stones that would embarrass a Ceylon stone of the same price, and Ceylon stones that had no business asking for a premium. Origin is a probability statement about a parcel; the stone in your hand is a fact.

The name is origin-neutral. The price is not.

Here is how the market actually resolves the contradiction, and it is remarkably tidy: the name follows the colour, and the money follows the origin.

The Natural Sapphire Company's buying guide puts Madagascan padparadschas at roughly twenty per cent below comparable Sri Lankan stones, and that figure matches what we see when we buy. Twenty per cent is not a rounding error, but neither is it the chasm the romance implies. Auction practice shows the same mechanics at work. Among Sotheby's padparadscha lots across 2024 and 2025 you can find stones catalogued with an AGL report reading Malagasy origin with indications of low temperature heating, and others with AGL or Gübelin reports reading Ceylon origin with no indications of heating — and the second description is the one positioned as the premium tier. Note that in those descriptions origin and treatment are travelling together. Much of what buyers experience as an origin premium is actually a compound premium: Ceylon and unheated, which is a scarcer combination than either alone.

What we would actually tell you to do

If two stones are genuinely comparable in colour, clarity, cut and treatment status, and one is Ceylon and one is Madagascan, the Ceylon stone will cost more and will be easier to resell. That is a real, observable market fact and you are entitled to weigh it. If you want the Ceylon stone because your family is Sri Lankan, or because the word and the place belong together in your head, that is a perfectly good reason to spend the money and I am not going to argue you out of it.

What I would argue you out of is paying a Ceylon premium on a Ceylon claim. Origin determination is a laboratory discipline based on inclusion suites and trace-element chemistry, and it is an opinion — a well-founded one from a good laboratory, but an opinion, and one that laboratories occasionally decline to offer when the evidence is ambiguous. If origin is load-bearing in your decision, it needs to be on a report from a laboratory that does origin work seriously, not in a seller's sentence. A stone described as Ceylon with no report saying so is a stone described as Ceylon.

And if you find that the twenty per cent matters more to you than the postcode, buy the Madagascan stone with a clear conscience. It is the same species, the same chemistry, the same name under every rulebook that counts. You can see how we describe origin on the stones we hold in the Ceylon sapphire collection, and every stone we sell carries its origin position in writing rather than by implication.

Two colours, and only one of them is loyal

A padparadscha's colour is a mixture, and the two halves of the mixture behave completely differently.

The pink half comes from chromium — trivalent chromium substituting into the corundum lattice, absorbing light in a way that has not changed since the crystal grew and will not change while you own it. That half is entirely stable. The orange-yellow half, in a significant proportion of stones, comes from something else: colour centres, which are essentially trapped-charge defects in the crystal structure. Work published in the Journal of Gemmology, volume 36 number 4, in 2018 set out the mechanism clearly. Those centres charge under ultraviolet light and discharge under ordinary illumination, and the effect is reversible. The technical words are reversible photochromism and, when a laboratory prints it in a comment, tenebrescence.

The practical consequence is what should worry a buyer. A stone can arrive at a laboratory or a shop having spent time in sunlight or under a UV lamp, looking gloriously orange-pink, and then slowly drain toward pink over days or weeks of normal wear. The same 2018 study documents a case that makes the point vividly: a 9.1 carat stone submitted as a padparadscha shifted from orange-pink to a vivid pink; its orange charged straight back within minutes under long-wave ultraviolet, then faded again over days to weeks in ordinary daylight. It was, in effect, a rechargeable padparadscha.

The same study looked at forty-eight metamorphic samples from Madagascar and Sri Lanka. The stones from Ambatondrazaka — the Bemainty area that came on stream around 2016 to 2017 — showed the highest frequency of the worst category: distinct shifts that take the stone right out of the padparadscha range. That deposit's timing is not a coincidence in the story of why the rules tightened when they did.

It is tempting to file all this under laboratory curiosity. Do not. At the International Gemmological Conference in 2023, Aaron Palke, Shane McClure and Nathan Renfro reported that roughly sixty per cent of the pink sapphires they tested, across various origins, showed some colour change after about four hours of long-wave ultraviolet exposure. Stones from Sri Lanka, Burma and Vietnam were the most UV-reactive; Ilakaka material was largely unchanged. One stone in their work needed heating to 325 °C to reach its ground state — 200 °C did nothing at all.

Sixty per cent reframes the whole question. Colour instability is not an edge case to be checked on suspicious stones; some degree of it is closer to the default in this material, and the meaningful question is not whether a stone shifts but whether it shifts far enough to leave the padparadscha range. Palke and his colleagues also made the argument that follows logically: the state after standard fade testing is the stone's actual colour, and that is what grading should describe.

The harmonised test, in detail no consumer guide we know of publishes

The laboratories agreed. The LMHC made a colour stability test mandatory for any corundum of padparadscha-like colour, and in August 2025 it published Information Sheet #16, version 1, which harmonises the protocol. Since no consumer guide I can find explains what actually happens to your stone, here it is.

Step one: grade the initial colour under standardised D50 to D65 light against a colour reference chart — Munsell, Color-Codex or similar. Step two, the deactivation: place the stone five to ten centimetres from a strong LED or halogen source, with a suggested rating of 100 watts and very little or no ultraviolet content, for a minimum of three hours. Step three: regrade under the same standardised light. There is an optional activation cycle as well — the stone goes directly onto a strong ultraviolet lamp, long-wave or short-wave, for at least ten minutes, and is graded again. Only the full deactivation-plus-activation cycle reveals the complete range of a reversible shift. Laboratories are advised to record an absorption spectrum at each grading step.

The reporting language is standardised too: no colour shift; minor colour shift, with an example such as orange to pinkish orange; or distinct colour shift, such as orange to pink. Reversible cases may be described in the comments as reversible photochromism or tenebrescence.

Then comes the consequence. Under LMHC rules, a stone whose colour shifts out of the padparadscha range during stability testing must be called pink sapphire or fancy sapphire — not padparadscha — optionally with a comment about the instability. A stone that shifts but stays within the range keeps the name, with an optional tenebrescence note. That exclusion was adopted at the 29th LMHC meeting in Lucerne on 14 and 15 November 2018, and Lore Kiefert, chief gemmologist at Gübelin, gave the plain-language reason at the time: if you buy a padparadscha and days later it is a pink sapphire, you have probably overpaid, so the report needs to say something about stability.

Where the laboratories still differ

They test. They do not all conclude the same way, and this is the single most useful thing to understand before comparing two reports.

GIA exposes sapphires of padparadscha colour to intense incandescent illumination for a few hours and reports the colour the stone shows after the test. That rule holds in both directions, which produced one of the more surprising findings of recent years: in December 2022, in a laboratory note by Aaron Palke, Shane McClure and Nathan Renfro, GIA reported the reverse phenomenon — some pink and pale yellow sapphires gain orange or yellow during the incandescent stability test, with pink stones shifting toward padparadscha, then fading back to their ground state in darkness and recharging under ultraviolet or sunlight. GIA still reports the post-test colour in every case, for consistency.

SSEF runs a comparable fade test: about three hours of intense halogen light on a reflective surface, colour graded before and after using Munsell charts or spectroscopy, with a ten-minute ultraviolet re-activation test in a dark box. Stones showing marked shifts are labelled fancy sapphire rather than padparadscha, with an explanatory letter available on request.

GRS takes a different road. Rather than demoting the stone, it classifies stability into published types. CST Type 1 is no fading observed. Type 2a and 2b describe stones whose orange or yellow component fades — 2a keeps a padparadscha colour in both the charged and discharged states, 2b does not. Type 3a and 3b describe stones whose orange or yellow component increases, with the same a-and-b logic. Type 4 is an irreversible colour change. GRS built an automated lightbox facility that charges, discharges and recharges stones while documenting every state.

That 2a-versus-2b distinction is price-critical, and it is printed on certificates that buyers are holding right now without knowing what it means. A 2a stone is a padparadscha whichever way you catch it. A 2b stone is a padparadscha only when it is charged. If you take one line of this chapter to your next viewing, take that one.

What a buyer should actually do

Ask, in writing, whether the stone has been colour-stability tested, by whom, and what the result was. If the report carries a GRS CST type, ask which one and what it means for the discharged state. If it is an LMHC-member report and the stone is still called padparadscha, the stability test has already been done and passed — that is precisely what the name now guarantees under those rules, which is a genuine improvement on the market of ten years ago.

And if you are buying a stone with a known minor shift at a price that reflects it, that is a perfectly rational purchase. Tenebrescence disclosed and priced is a characteristic. Tenebrescence undisclosed is somebody selling you a photograph.

Treatments are legitimate when they are disclosed and priced accordingly. A heated padparadscha is a real padparadscha, sold honestly at a heated price, and there is nothing wrong with owning one. What is wrong — the only thing that is actually wrong in this whole area — is silence. A treated stone sold as untreated is not a gemmological problem; it is a conduct problem. Keep that distinction in mind while you read the rest of this chapter, because the technical material can make treatment sound like a moral failing, and it is not. The failing is non-disclosure.

Heat: ordinary, ancient, and harder to detect than you think

Conventional heat treatment does not cost a stone the padparadscha name. Both the LMHC and CIBJO permit untreated and conventionally heated stones to be called padparadscha; CIBJO's Gemstone Book, in its edition of 22 December 2022, sets out the same boundary, excluding beryllium-diffused, irradiated, coated and dyed material from the name.

What heat does cost you is money, and the gap is discussed properly in Chapter VI. What is worth knowing here is that the detection problem is far more delicate at the low end than most buyers assume. Low-temperature heating, roughly 800 to 1200 °C, is common for Madagascan pink and padparadscha material, and SSEF describes it as one of the hardest calls in gemmology. Their detection relies on a combination of microscopy, FTIR spectroscopy looking at hydroxyl-related absorption peaks, and Raman spectroscopy — three techniques cross-checked, not one. Krzemnicki and colleagues, writing in Minerals in 2023, added a further published criterion: the dehydration of diaspore and goethite inclusions during low-temperature heating can separate heated from unheated corundum when microscopy and FTIR are inconclusive.

Two things follow. First, no jeweller looking through a loupe can tell you a padparadscha is unheated, and anyone who does so with confidence is telling you about their confidence rather than about the stone. Second, this is exactly why laboratory language is hedged. A report reading no indications of heating is a properly cautious formulation, not a weasel one.

One more piece of vocabulary, because it circulates and confuses people. At its November 2018 meeting the LMHC clarified that the sapphire treatment marketed as HPHT is in fact standard heat treatment, with pressure incidental, requiring no special notation beyond ordinary heat disclosure — although laboratories noted it can create or expand fissures. If someone offers you an HPHT padparadscha as though it were a distinct and superior category, it is heated.

Beryllium: the treatment that invented the fake padparadscha

Now the serious one, and it is worth telling as history because the history explains the safeguards.

In October 2001 a burner in Chanthaburi, Thailand worked out how to convert off-colour Songea corundum from Tanzania into orange and orange-pink material. Large numbers of padparadschas duly appeared in the Thai market in late 2001, and in December of that year Ken Scarratt, then running the AGTA laboratory, and Bill Larson of Pala International obtained samples in Bangkok. The method was lattice diffusion of beryllium, and the full account was published by Emmett and colleagues in Gems & Gemology, volume 39 number 2, in the summer of 2003. Lotus Gemology's essay on the subject remains the most readable narrative of what happened next.

What happened next was that the padparadscha market briefly came apart. GIA, Gübelin and SSEF identified and disclosed the treatment to the industry across 2001 and 2002 following the AGTA alerts, and previously certified orange-pink stones had to be re-examined. If you ever wonder why this variety is surrounded by more procedural caution than any other fancy sapphire, that episode is the answer.

Beryllium diffusion runs at extreme temperatures, up to around 1800 °C, and beryllium diffuses through corundum so rapidly that the induced colour can penetrate the entire stone. That defeats the classic immersion check for a colour rim in a great many cases — the traditional trick simply does not apply when the diffusion has gone all the way through.

Worse, beryllium is too light an element for XRF to detect. Not difficult: impossible, as a matter of physics rather than of budget. Identification requires mass-spectrometry-class methods — LA-ICP-MS, which GIA notes routinely detects beryllium down to around 0.1 parts per million atomic, or SIMS, or LIBS. Those instruments live in major laboratories, not in shops.

This is the single most important practical paragraph in the handbook. When a seller tells you a stone was tested with a machine, ask which machine. A handheld or benchtop XRF is a genuinely useful instrument for many purposes and cannot rule out beryllium diffusion in a padparadscha. For a stone of any significance in this variety, a report from a top-tier laboratory is not a nicety; it is the only reliable protection that exists.

Synthetics, and one very determined fake

Synthetic sapphire in padparadscha colours is not a modern threat — it has existed since 1908, via the Verneuil flame-fusion process. A century-old hazard is still a hazard, and it is one a competent gemmologist can usually resolve quickly through inclusion study.

The instructive case is more elaborate. In November 2015 a 14 carat stone described as a Sri Lankan padparadscha was submitted to Lotus Gemology in Bangkok; the results were published in June 2016 by E. Billie Hughes, C. Chankhantha, A. Burkhardt, W. Manorotkul and R. W. Hughes. The stone was a Verneuil synthetic pink sapphire that had been deliberately fractured, infilled with copper-bearing lead glass and given orange-stained fissures to counterfeit padparadscha colour. Part of what gave it away was that its short-wave and long-wave ultraviolet fluorescence were suspiciously equal. Somebody went to a great deal of trouble. They were caught because the stone was submitted to a laboratory — which is the entire moral of the story.

What the law expects, in two jurisdictions

In the United States, the FTC Jewelry Guides at 16 CFR Part 23, revised in 2018, require disclosure of treatments that are not permanent, that require special care, or that significantly affect value. Diffusion treatment triggers that obligation, and the accepted descriptors are diffusion-treated sapphire or beryllium-treated sapphire.

Closer to home, the ACCC's jewellery industry guidance, launched in September 2005, states that describing a treated gemstone as natural without disclosing the treatment risks breaching the misleading-and-deceptive-conduct provisions now found in section 18 of the Australian Consumer Law. The regulator's stated position is that prudent practice is disclosure even where a treatment is hard to detect. Read that sentence again if you are selling as well as buying: difficulty of detection is not a defence.

For consumer purchases in Australia, the consumer guarantees under the Australian Consumer Law also apply — goods must match their description, among other things — and a stone sold as unheated that is not unheated does not match its description. Those guarantees attach to consumer purchases; they are a floor set by law, not a favour granted by a seller.

Our own practice is simpler than the law requires and easier to check: every stone we sell carries its treatment status in writing, in plain words, on the Digital Gemstone Passport that travels with it, and the vocabulary we use is defined publicly in our treatment register so you can hold us to it.

Everything below is context for a purchase, not a forecast. We are gemmologists and merchants; we are not licensed financial advisers, and nothing in this chapter — or anywhere in this handbook — is investment advice. Gem prices are reported by dealers and auction houses with their own interests, they move, and a headline result achieved by an exceptional stone in a good room on a good night tells you almost nothing about what your stone would fetch on a Tuesday. Buy a padparadscha because you want to own it. Read the numbers so you are not overcharged, not so you can predict anything.

The retail ladder

Gemdaia's padparadscha pricing guide, current to 2025, puts fine retail material at roughly US$2,000 to US$30,000 and more per carat, depending on size, colour balance, clarity and treatment. Certified stones over two carats with even colour sit in the upper part of that band, around US$10,000 to US$30,000-plus per carat.

Pala International's long-standing buying guide reaches a similar top: fine padparadschas run up to about US$30,000 per carat at the top tier. Pala also supplies the sentence I would have every buyer memorise — that any fine untreated padparadscha of quality above two carats is a rare stone. Not expensive. Rare. Those are different claims, and the second one is the reason for the first.

Above the retail band sits the trade quote. JewelleryNet and The Jewellery Editor have both cited figures above US$50,000 per carat for top non-heated stones of good quality. Those are quotes for exceptional material, not shop prices, and you should treat any local offer at that level as requiring extraordinary documentation.

What the no-heat premium is actually worth

This is where I have to disagree, politely, with a lot of retail copy.

The documented, conservative premium for an unheated padparadscha over a comparable heated one runs from about 50 per cent — Pala's figure — to roughly double. Gemdaia's 2025 guide gives a concrete worked example at the three-carat mark: a heated stone up to about US$10,000 per carat against about US$20,000 per carat for a comparable untreated stone. So: fifty per cent to two times, well documented.

You will also see claims that unheated stones are worth three to five times their heated equivalents. Those figures circulate, they are dealer-sourced and retail-tier, and we could not verify them independently. I am not saying they never happen — an extraordinary stone at the top of the market can do extraordinary things — but if someone quotes you a five-times premium as though it were a market rule, they are quoting marketing. Fifty per cent to double is the number to plan around. You can see the shape of the premium in the stones we hold in the unheated sapphire collection, where the no-heat position is stated on every stone rather than implied.

Origin stacks on top, at roughly twenty per cent for Ceylon over comparable Madagascan material, as Chapter III set out. And the two premiums compound: Ceylon and unheated together is a scarcer proposition than either alone, which is why auction catalogues put that combination at the front.

The auction record, with the discrepancy left in

Auction results are the only truly public prices in this market, so they are worth knowing even though they describe stones most of us will never handle.

The highest total price for a padparadscha is a 28.04 carat Sri Lankan stone sold at Christie's Hong Kong on 28 November 2017. Lotus Gemology's auction-records table gives the result as US$2,489,354, or US$88,779 per carat. Several trade reports, including National Jeweler, give US$2,464,470 for the same sale. Both are honest; the gap is a Hong Kong dollar conversion difference. I am leaving it in rather than picking the prettier number, because a handbook that silently launders a discrepancy has taught you nothing about how to read the ones you will meet later.

The highest price per carat belongs to a smaller stone: 8.01 carats, Sri Lankan, sold at Sotheby's Hong Kong on 3 October 2017 for US$780,800 — US$97,478 per carat. Note the order of those two records. The most expensive padparadscha by total was not the most expensive by carat. Weight and value are related in this variety, but they are not the same axis.

Then there is the du Pont Padparadscha, which is my favourite story in the file. A 24.58 carat natural and untreated stone, bought by a member of the du Pont family holidaying in Ceylon in 1937, sold at Christie's New York Magnificent Jewels on 8 December 2020 for US$930,000 — about US$37,835 per carat, against an estimate of US$500,000 to US$700,000. It was admired, and priced, for a saturation richer than the modern pastel definition strictly encourages, which is a quiet argument in favour of the dealers Hughes surveyed.

Two more, for calibration in the other direction. A 30.77 carat padparadscha ring at Bonhams Fine Jewellery on 19 September 2017 made US$199,500. That is about US$6,500 per carat — a very large stone at a per-carat price a fine three-carat comfortably beats, and it is the cleanest available proof that weight alone does not set price in this variety. And a 12.94 carat heart-shaped padparadscha at Sotheby's New York Magnificent Jewels on 19 April 2016 made around US$250,000.

The older benchmark, for anyone reading a guide that still quotes it: a 20.84 carat padparadscha at Christie's in June 2005 made US$374,400, or US$18,000 per carat. That result stood as the reference sale for roughly a decade, and a surprising number of consumer guides still cite it as if it were current. It is twenty-one years old.

Where the ladder actually lives

Headline records make poor anchors, so here is one from the other end. Sotheby's sold an SSEF-certified Madagascan unheated padparadscha ring for around US$33,500 — a real stone, a real report, a real auction, at a price a serious private buyer might genuinely pay. The distance between that figure and the 2017 records is the whole market, and most of us shop in the lower half of it.

For species-level context: the ceiling for sapphire as a whole is not padparadscha's. The Regent Kashmir, a 35.09 carat blue sapphire, made US$272,374 per carat at Christie's Hong Kong on 27 May 2025. Padparadscha's US$97,478 per carat sits well below top Kashmir blue — and above nearly every other fancy sapphire colour.

Finally, the rarity anchor that explains the whole ladder: the largest known fine padparadscha is a 100.18 carat oval from Sri Lanka, held by the American Museum of Natural History in New York. One stone, in a museum, at the top of a species. That is the supply picture, and it is why a clean, even, unheated two-carat stone with a good report is a genuinely difficult thing to find rather than merely a costly one.

Which laboratory, and under which rulebook

Start at the letterhead, not at the picture. The first thing a padparadscha report tells you is which rulebook is in force.

If the report comes from one of the seven LMHC members — CGL Japan, CISGEM, DSEF, GIA, GIT, Gübelin or SSEF — then the word padparadscha on that report carries the whole Information Sheet #4 apparatus behind it: the hue window, the pastel ceiling, the daylight standard, the seven disqualifiers, and a mandatory colour stability test the stone has passed. That is a lot of meaning packed into one word, and it is why those reports command the confidence they do.

If the report comes from GRS, AGL or Lotus Gemology, the word carries that laboratory's own standard, which may be stricter, looser or simply differently organised. GRS, for instance, uses padparadscha directly in the colour field of its reports and does not tie the term to origin. It is worth knowing that GRS trademarks certain colour terms — GRS-type pigeon blood for ruby and GRS-type royal blue for sapphire — but padparadscha is not one of them.

The buyer's conclusion from all this is not cynicism. It is specificity. Ask which laboratory, and understand that a difference between two good reports is usually a difference between two rulebooks rather than evidence that one of them is careless. Lotus Gemology has written about how genuinely fine the pink-sapphire-to-padparadscha boundary is; expert laboratories acknowledge openly that it is a judgement call. Two reputable reports can disagree on the variety name without either being wrong.

The five lines that matter

Once you know whose rules apply, read in this order.

The variety or identification line. Does it say padparadscha, or pink sapphire, or fancy sapphire? Under LMHC rules, fancy sapphire or pink sapphire on a stone of obviously padparadscha-ish colour is a meaningful signal — it may mean the stone failed the stability test, or carries a disqualifying modifier, or has eye-visible zoning. It is not necessarily a bad stone. It is a differently priced stone.

The treatment or enhancement line. The two formulations you want to be able to read are no indications of heating and indications of heating, sometimes qualified as low temperature. Remember from Chapter V that low-temperature detection is genuinely hard, so the hedged phrasing is a mark of care. If the line describes lattice diffusion, that stone cannot be called padparadscha under LMHC or CIBJO rules at all — and if the paperwork calls it one anyway, stop.

The origin line, if there is one. Origin is a laboratory opinion, and a good laboratory will sometimes decline to give one. If you are paying a Ceylon premium, this line is the only place that premium is documented.

The colour stability comment. This is the newest line and the one most people skip. Under the harmonised wording of Information Sheet #16 you may see no colour shift, minor colour shift or distinct colour shift, and comments referring to reversible photochromism or tenebrescence. On a GRS report you may instead see a CST type code — and as Chapter IV explained, the difference between 2a and 2b is the difference between a stone that reads padparadscha in every state and one that only reads padparadscha when charged.

The date. Reports are opinions given on a date, with the knowledge and instruments of that date. Two moments in this variety's history make that concrete. The beryllium episode of 2001 and 2002 forced the re-examination of stones that already held certificates. And the LMHC only adopted the unstable-colour exclusion at its 29th meeting in Lucerne on 14 and 15 November 2018, with the harmonised test protocol arriving in August 2025 — so a report issued before those dates was not applying the stability rules that a report issued today applies. An old report is not worthless. It is simply older than some of the questions.

The four questions, in writing

Take these to any seller, anywhere, including us. Ask for the answers by email so that they exist afterwards.

One: Is this stone treated, and if so, how? A trustworthy answer is one plain sentence. Vagueness, a change of subject, or a lecture about how all gemstones are treated is the answer.

Two: Which laboratory issued the report, what date does it bear, and may I see the full report rather than a summary card? Summary cards and website badges are not reports.

Three: Has the stone been colour-stability tested, and what was the result? If the seller does not know what you are asking, you have learned something useful about the seller.

Four: If origin is part of the price, where is it documented?

A fifth is worth adding for any significant stone: if this were sent to a top-tier laboratory tomorrow and the report differed, what would happen? The answer to that question is a returns policy, and it should be in writing too. For consumer purchases in Australia the consumer guarantees under the Australian Consumer Law apply regardless — goods must match their description — but a clear written policy saves everyone the argument.

When the reports disagree

It happens, and it is not the disaster it feels like. If you hold two reports with two variety names, work through the causes in order: different rulebooks (LMHC member versus not), different dates (before or after the stability rules), different colour states (charged versus discharged at the time of grading), and only then genuine disagreement on a boundary call.

If the difference is a rulebook or a date, you now know more about your stone than a single report would have told you. If it is a boundary call between two expert houses, then you own a stone that lives on the boundary — which is, after all, the definition of the variety. Price it accordingly and enjoy it. The vocabulary on any of these lines is defined in plain English in our gem lexicon, and if you want the general skill rather than the padparadscha-specific version, The Certification Handbook reads a laboratory report line by line.

The January 2018 effect

Princess Eugenie's engagement to Jack Brooksbank was announced on 22 January 2018, and the ring was an oval padparadscha sapphire in a diamond halo on yellow gold — a deliberate echo, as Gem-A and the BBC both noted at the time, of the 1986 Sarah Ferguson ruby engagement ring. Trade estimates put the centre stone at around five carats, with a wholesale value somewhere between US$35,000 and US$55,000 and retail estimates above US$140,000.

What happened next is the most instructive part. Gem traders reported a demand uptick for padparadscha, driven simply by new consumer awareness that the variety existed. Tom Heyman of Oscar Heyman described the general mechanism to National Jeweler: royal and celebrity coloured-gemstone rings have historically correlated with upticks in consumer demand for similar styles. The search data supports it — interest in sapphire engagement rings had already risen 115 per cent across 2017, spiked further after the January 2018 announcement, and Eugenie's ring still draws in the order of 1,600 searches a month.

And the trade's immediate worry, voiced in the discussion forums at the time, was exactly the one this handbook has spent seven chapters on: a popularity spike would tempt vendors to pass off brownish sapphires and beryllium-diffused material as padparadscha at inflated prices. Given what happened in 2001, that worry was well-earned. The remedy has not changed. Report, treatment line, stability line, date.

Is it a sensible engagement stone?

Yes, with two caveats that are worth stating honestly.

The first is practical and reassuring. Padparadscha is corundum, the same species as the blue sapphire in your grandmother's ring, and corundum is among the hardest materials used in jewellery. It is genuinely suited to daily wear in a way that many coloured stones are not. If you want a coloured centre stone that will still look like itself after a decade of washing-up, corundum is the right family to be shopping in.

The second is about light, and it is specific to this variety. Padparadscha is a pale, delicately balanced colour, and pale stones are more sensitive than saturated ones to what surrounds them. A yellow-gold setting pushes the stone warmer and orange-ward; white metal reads it cooler and pinker; a halo of colourless diamonds makes it look paler than it does loose against a neutral card. None of this is a defect, but it does mean you should insist on seeing the stone in the metal you intend to use, in daylight, before you commit. Ask to see it near a window rather than under a display lamp — a shop spotlight flatters this colour more than almost any other, and if you have read Chapter II you now know the laboratories specify daylight for exactly that reason.

If the stone has a documented minor colour shift, add one more habit: look at it after it has been indoors for a week, not only after an afternoon in the sun. You will simply be doing at home what the laboratory does with a halogen lamp.

Wearing, cleaning, and the one conversation to have with your jeweller

Day to day, a padparadscha needs less care than its price suggests. Warm water, a drop of mild detergent and a soft brush will handle almost everything. Take rings off for the gym, the garden and the beach, less because the stone will suffer than because settings do.

Ultrasonic and steam cleaning deserve a moment's thought rather than a blanket rule. A clean, unfractured stone in a sound setting is generally untroubled by either; a stone with reaching fissures is a different proposition, and if any fissure-filling has ever been done — which, remember, would have cost the stone the padparadscha name in the first place — the answer is simply no. When in doubt, hand cleaning costs nothing.

The conversation that genuinely matters is the one before any repair. Tell your jeweller, before the work starts, what the stone is and what its report says, and ask them to unset it rather than work with a torch near it. Sizing, retipping and re-shanking are all routine; heat applied to a set coloured stone is how routine work becomes an incident. This is not padparadscha-specific paranoia — it is the standard rule for coloured stones — but a stone whose value rests substantially on a no-heat report has more than the usual amount to lose.

The note you leave with it

Padparadscha is a variety whose meaning is documentary as much as visual. In fifty years, the person who inherits your ring will not be able to tell by eye whether the stone was heated, where it came from, or which rulebook the word on the box was using. They will only know what you leave with it.

So leave the full laboratory report, not the summary card. Leave the date. Leave a plain sentence in your own handwriting about what you were told when you bought it and by whom. If the stone has a stability comment, write down what it means, because in half a century the vocabulary will have moved on again. Every stone we sell leaves here with a Digital Gemstone Passport that holds exactly this material, and free lifetime re-authentication so that a future owner can bring it back and have the record confirmed rather than reconstructed.

Do that, and the stone stays a padparadscha in the full sense of the word — a lotus-coloured sapphire with a name, a history and a piece of paper that still means something. Do not, and in two generations it becomes a pretty orange-pink stone that somebody has to start investigating from scratch. Chapter I began with an argument about what colour a stone was. This is how you make sure your descendants get to have the enjoyable version of that argument instead of the expensive one.