The Online Gem Buyer's Handbook

Chapter I

Foreword: The Parcel Nobody Ordered

Anna, our FGAA-qualified gemmologist, on the parcel scams, paper mills and countdown clocks of the online gem trade — and why the internet is still, handled correctly, the best market a gem buyer has ever had.

The box with no postage label

In August 2023, people across the United Kingdom and northern Europe began finding small parcels in their letterboxes that they had never ordered. Inside grey and pink boxes, tucked into pink ‘Princess’ gift bags, were ‘diamond solitaire’ rings — each one accompanied by a laminated identification certificate. There was no postage label, no receipt, no sender information of any kind. The certificates announced themselves as the work of the ‘Fellowship of Gemological Association of Great Britain’, and carried the logos of IGI, ILAC, Platinum Guild International and GIA — four organisations that had precisely nothing to do with any of it.

Two details gave the game away, and I want you to notice how small they are. First, the real Gemmological Association of Great Britain — Gem-A — spells gemmological the British way, with two m’s; the certificate used the American spelling. Second, and more fundamentally: Gem-A does not grade gems and does not issue gem reports of any kind, so a ‘Gem-A certificate’ is fake by definition, the way a Qantas train ticket would be. A police report was filed on 22 August 2023, National Jeweler covered Gem-A’s public warning the following day, and the Italian Gemological Review was still reporting fresh parcels bearing false Gem-A credentials in January 2024. Recipients were eventually traced to purchases made through an e-commerce site, which is the clue to what the whole strange exercise actually was.

Because nobody mails out free rings for fun. This is a ‘brushing’ scam: a seller ships unordered goods so it can post glowing ‘verified purchase’ reviews in the recipients’ names. GIA had issued its own consumer alert on 2 March 2023 after fielding more than fifty calls in three months — from California, Texas and across the United States — about low-quality jewellery that arrived uninvited, some of it accompanied by counterfeit GIA diamond grading reports. Some of those counterfeit reports identified moissanite as diamond, in writing, under a forged version of the most trusted letterhead in gemmology. Sit with that for a moment: the certificate — the thing buyers are told to demand as their protection — was the payload.

Why listen to a seller about scams

A fair question, so let me answer it plainly. I am the gemmologist at Ministry of Gems, a Sydney house built on three generations in the Ceylon gem fields since the 1960s. We sell gemstones online ourselves. Every stone that leaves us does so with its treatment status stated in plain words, because our view — and you will meet it again and again in this book — is that treatments are legitimate when they are disclosed and priced accordingly, and that the sin is never the treatment; the sin is the silence. That doctrine gives us an odd qualification for this subject: we know exactly which corners can be cut in an online gem listing, because we pay for the privilege of not cutting them. To our knowledge, no other guide to online gem fraud has been written by a seller who discloses; the existing ones are written by bloggers and scam-watchers, and they are useful, but they have never had to photograph an unheated sapphire honestly at 4 pm in winter.

The state of play in Australia

This is not a niche worry. According to the ACCC’s National Anti-Scam Centre, shopping scams were the most-reported scam type involving financial loss in the first half of 2025, with more than 6,300 loss reports to Scamwatch, and total reported scam losses trending up 24% year-on-year to A$174.8 million for that half. Zoom out further and the ACCC’s combined figures — Scamwatch, ReportCyber, AFCX, IDCARE and ASIC together — put Australian scam losses at A$2.03 billion in 2024 and A$2.18 billion in 2025. Gems are a small slice of that, but they are a slice with a special vulnerability: almost nobody can check the goods by eye, the price range for ‘a red stone’ spans five orders of magnitude, and the trade’s own paperwork can be counterfeited for the cost of a laminator.

How this handbook works

Here is the route. Chapter II gives you the price floors — the plain arithmetic that exposes most fakes before you read a single word of the listing. Chapters III and IV cover the fake-certificate industry and the proper way to verify a real report. Chapter V is a bestiary of documented marketplace scams, with dates, so you can recognise the shapes rather than memorise the names. Chapter VI is a phrasebook of true words used to mislead. Chapter VII deals with photographs and origin stories, and Chapter VIII with payment — the armour you choose before you ever click buy. Along the way I will point you to our free Gem Lexicon when a term needs more room, and to our Treatment Register, where we publish exactly how each treatment works and what it does to price.

One more thing before we start. This book is sceptical, but it is not a warning against buying gems online. I buy stones over the internet; the house buys stones over the internet; done properly, the online market gives an Australian buyer access to goods our grandparents would have needed a Colombo broker and a fortnight at sea to see. The internet is the best gem market ever built and the cheapest fraud machine ever built, and they share a front door. The whole craft is knowing which one you are walking into — and after eight chapters, you will.

— Anna, FGAA
Ministry of Gems, Sydney

Chapter II

Arithmetic Before Romance: The Price Floors

Before you learn a single red flag, learn the price floors — the plain arithmetic that makes most online gem fraud collapse before the certificate even comes out of the envelope.

The floor under every stone

Every gem species has a number underneath it, and almost nobody selling fakes wants you to know it. That number is the cost of producing the cheapest thing that could plausibly be passed off as the stone in question. For ruby, the arithmetic is brutal and clarifying. Flame-fusion synthetic ruby — the Verneuil process, chemically genuine corundum grown in a furnace — costs roughly US$1–5 per carat, with the raw material often under fifty US cents a carat, according to production-cost breakdowns published by Synthetic-Ruby.com and GemSelect. Flux-grown synthetic ruby, the premium tier of man-made material, runs about US$100–300 per carat. And lead-glass composite ‘rubies’ — low-grade corundum whose fractures are filled with lead glass, sometimes in proportions where the glass is a substantial fraction of the stone — sell for a few dollars per carat and have circulated in volume since roughly 2004, as the Skyjems encyclopedia documents.

Now hold those numbers against the top of the market. Fine unheated Burmese ruby trades at roughly US$30,000–100,000 and more per carat, per Estate Diamond Jewelry’s market summaries, and The Natural Ruby Company notes that fine unheated ruby over two carats from any origin commonly exceeds US$35,000 per carat. At the very summit, the 25.59 ct Sunrise Ruby — unheated, Burmese, pigeon’s blood — sold at Sotheby’s Geneva in May 2015 for CHF 28.25 million, around US$30 million, better than a million US dollars per carat; the 8.62 ct Graff Ruby had made US$8.6 million at Sotheby’s Geneva the previous November. Between a furnace product at a dollar a carat and an auction record at a million, the word ‘ruby’ alone tells you almost nothing. The price tells you nearly everything.

The ninety-dollar ruby, solved

So take the classic listing: a ‘genuine natural ruby ring’ for $90. Run the floor test. Ninety dollars cannot buy a natural ruby of the size photographed, because natural material of even mediocre gem quality costs more than that at wholesale before anyone cuts it, mounts it or ships it. It can, however, comfortably buy a flame-fusion synthetic at $5 a carat, or a lead-glass composite at less — with enough margin left over to pay for the gold-plated setting, the velvet box and the advertising that found you. The listing is not mispriced; it is misdescribed. The stone in the photo may even show convincing ‘natural’ inclusions, and Chapter V will show you how those are manufactured too. The point of the floor is that you did not need to see the stone, read the reviews or examine the certificate. The arithmetic disposed of the listing before romance ever got a vote.

The same test works across the counter of the coloured-stone world. Beryllium-diffused sapphire — heavily treated material whose colour is driven in from outside, which we cover properly in Chapter VI — is worth less than US$1,000 per carat, with commercial goods as low as about US$15 per carat, per Skyjems and Crescent Gems, even when it visually mimics a natural padparadscha that can fetch around US$50,000 per carat; a conventionally heated sapphire of similar appearance might be around US$1,500 per carat. Ruby-in-zoisite, an opaque decorative rock of ruby crystals in green matrix, trades at roughly US$250 per kilogram for rough — yet Gemology Resources documents specimens being fraudulently ‘appraised’ at millions, even hundreds of millions, as an ‘investment’. When a $40 cabochon arrives with a document valuing it at $14 million, the document is not optimistic. It is a prop.

Why size multiplies price

The second piece of arithmetic that protects you is the size ladder. Gem prices are quoted per carat, and the per-carat rate itself climbs with size, because large clean rough is exponentially rarer than small clean rough. The Sapphire Bank’s published price data for fine unheated Ceylon blue sapphire shows roughly US$4,000 per carat at one carat rising to roughly US$10,000 per carat at three carats for the same quality — and a fine two-carat Ceylon stone typically costs four to six times its one-carat sibling, not twice. The Natural Sapphire Company’s pricing charts show the same curve. Top unheated Ceylon stones above three carats run US$5,000–25,000 per carat and beyond. This is why a listing offering a ‘10 ct natural sapphire’ for the price of a decent one-carat stone is not a bargain hiding in plain sight; it is the ladder snapped in half. If you would like to see the curve with your own eyes, our Ceylon sapphire collection publishes carat weights beside prices precisely so the ladder is visible, and The Price of Colour in this library walks through it species by species.

A floor is not a valuation

Two cautions before we move on. First, floors work in one direction only. A price below the floor proves a listing false; a price above the floor proves nothing at all, because fraud is perfectly happy to charge full freight — the Bangkok operators in Chapter V charge thousands for near-worthless stones. Second, none of this chapter — none of this book — is investment advice. I quote auction results and market ranges, with dates and sources, because they calibrate your sense of what is plausible, not because I am suggesting any stone will appreciate. Nobody knows that, and as you will see in Chapter V, ‘guaranteed resale profit’ is itself one of the oldest scam signatures in the trade. Buy a gem because it is beautiful, honestly described and fairly priced today. Our Gem Market Report — 2026 Edition exists to keep those calibrations current; use it as a rangefinder, never as a prospectus.

Three practical notes on using the floors well. First, currency: the trade quotes in US dollars, and every figure in this chapter is USD — convert before you compare, because an AUD price read as USD flatters a listing by a third. Second, watch for the per-carat two-step: a listing shouting ‘$500 per carat’ on a claimed two-carat stone is asking $1,000 for something the floor test must still judge as a whole — and sellers switch between per-carat and total pricing precisely where the switch confuses. Third, mounted stones muddy every number: a setting hides weight, forgives windows and conceals chips, which is why serious per-carat arithmetic belongs to loose stones, and why our own jewellery listings state the centre stone’s carat weight separately rather than folding it into a ‘total gem weight’ that no floor can be tested against.

Chapter III

The Certificate Industry That Isn’t

There are real laboratories, and there is a printing industry dressed up as one — laminated cards in unmarked parcels, invented acronyms, and verification websites run by the counterfeiters themselves.

All you need is a printing press

The insurance analysts at JCRS, reviewing a fake laboratory called ‘KGCL’ whose acronym returned no search results and whose paperwork listed no contact details, put the economics of the fake-certificate trade in a single line: to run a fake lab, all you need is a printing press. That is the entire business model. There is no building, no spectrometer, no gemmologist — just cardstock, a logo drawn to look institutional, and sometimes a website. The output is designed to survive exactly one test: a nervous buyer glancing at it and feeling reassured.

The naming strategies are worth learning because they repeat. Strategy one: sound-alike acronyms. The ‘GRA’ — rendered in full as the ‘Gales Gemological Research Association’ — appears on mass-produced moissanite certificates by the thousand and on counterfeit paperwork flagged in GIA’s 2023 brushing-scam alert, and no officially recognised laboratory of that name exists; the American Gem Society’s coverage notes the name appears chosen to shadow GIA. Strategy two: one-letter surgery on a real name. A fraudulent valuation document surfaced under ‘Instituto Gemmologico Italiano’ — one vowel away from the genuine Istituto Gemmologico Italiano in Milan, as Gemology Resources reported. Strategy three: borrow real logos wholesale, as the fake ‘Fellowship of Gemological Association of Great Britain’ certificate did with IGI, ILAC, Platinum Guild International and GIA in the mail scam from Chapter I. In every case the trick is adjacency: stand close enough to a real institution that its credibility falls on you like shade.

The certificate that appraises itself

Here is the single most reliable tell on a fake document, and I want it in muscle memory: a real laboratory report does not tell you what the stone is worth. GIA states outright, in its own FAQ, that its documents are reports rather than certificates and that no appraisal value is stated on them. The reason is structural — a laboratory’s authority comes from measuring what a stone is, not opining on what someone should pay — and every serious lab holds the same line. So when a document styled as a lab certificate arrives with ‘Appraised value: US$18,500’ printed in a gold box, it has confessed. Fake labs print values because the value is the product: it powers the inflated-appraisal retail scheme Scambusters describes, where a stone is ‘independently valued’ far above its asking price so the listing feels like theft in your favour. Remember the ruby-in-zoisite specimens from Chapter II, appraised at millions against a US$250-per-kilogram market. A printed value on gem paperwork is not a bonus. It is the watermark of the industry that isn’t.

The verification site run by the counterfeiter

‘But I checked it online, and it verified.’ This is the heartbreaking sentence, because verification theatre is now part of the product. The definitive case study comes from John Burrows of Albion Fire and Ice, who in August 2021 made a deliberate test purchase: a $1.29 ‘watermelon tourmaline’ that turned out to be two pieces of glued glass. It arrived with a professional plastic certificate card from ‘AGI Labs’ — and, crucially, the certificate verified perfectly on AGI Labs’ own online database, because the counterfeiter operates the database. A working lookup page proves only that someone can run a website. JCRS documented the same move at higher stakes: a fraudulent report steering buyers to ‘AIGSThailand.co’, a spoof one letter removed from the genuine Asian Institute of Gemological Sciences at AIGSThailand.com. The QR code printed on a fake certificate resolves, obediently, to the fake lab’s own confirmation page. Never verify a document through anything printed on the document — no QR code, no URL, no phone number. Type the laboratory’s known domain yourself. We will do exactly that, step by step, in the next chapter.

What real paper looks like

Real laboratory documents fight back. Printed GIA reports carry security features including microprinting and a hologram, and GIA has moved some formats — the Diamond Dossier among them — digital-only, partly for fraud resistance. Real labs write in a deliberately harmonised dialect: the Laboratory Manual Harmonisation Committee, whose members include GIA, Gübelin, SSEF and GIT, standardises the wording used for treatment and origin disclosure, which is why genuine reports from different labs sound almost eerily alike. A ‘certificate’ declaiming that a stone is a ‘100% natural earth-mined precious ruby’ is speaking a language no LMHC lab uses; grandiosity is a dialect fakes speak among themselves. And real reports are honest about their own limits — under ISO/IEC 17025, the standard to which major gem labs are accredited, opinions and interpretations must be marked as such. A document that is certain about everything, effusive about value and decorated like a share certificate is none of the things it claims. For the full anatomy of a legitimate report, field by field, this library’s Certification Handbook does that job properly, and I will not repeat it here.

Chapter IV

Verification: Checking a Report Properly

A report number that checks out online is a good start and nothing more — the proper verification takes ten minutes, a typed web address, and a set of scales.

Type the address yourself

The first rule was earned in the last chapter, so let us make it operational. When a stone arrives with a report, open a fresh browser tab and type the laboratory’s domain from your own knowledge or from an independent search — never from the QR code, link or phone number printed on the paper. For GIA, that means gia.edu and its Report Check service, which verifies every report the institute has issued since 1 January 2000, free, by cross-referencing the paper against GIA’s live database. Other serious labs run equivalents, some with a second factor: Lotus Gemology in Bangkok, a ruby-and-sapphire specialty lab, requires the report number plus a PIN printed on the report card — a small, elegant defence, because a counterfeiter who has photographed a genuine report’s public face still cannot verify against it. If a lab’s name yields no independent search results at all, no address, no staff, no history — the KGCL problem — you already have your answer, and it cost you nothing.

The number is not the stone

Now the harder lesson, and the one that separates careful buyers from lucky ones: a report number that verifies proves the report exists. It does not prove the stone in your hand is the stone in the database. The documented fraud here is genuinely clever. In 2021, Indian police uncovered an operation that laser-inscribed lab-grown and treated diamonds with the GIA report numbers of nearly matching natural diamonds — so the stone ‘verified’ online against someone else’s legitimate paperwork. The bust, covered in JCRS’s insurance newsletters, seized genuine reports, fake reports and the laser-inscribing machine itself. GIA responded by offering same-day inscription-match verification, in which the lab confirms that the inscription on your stone was actually applied by GIA to that stone.

Your countermeasure is beautifully low-tech: cross-check the physical stone against the physical claims. A gem report states the carat weight to two decimal places and the measurements to hundredths of a millimetre. Carat scales accurate enough for a sanity check cost less than a dinner out; a digital calliper costs less than the scales. If the report says 2.04 ct and 8.12 × 6.35 × 4.10 mm, and your loose stone weighs 2.31 ct at 8.9 millimetres, the paperwork belongs to a different stone, whatever the database says. This is also, quietly, the best argument for buying loose stones rather than mounted ones where serious money is involved: a mounted stone cannot be weighed, which is exactly why fake-lab paperwork so often grades stones in their settings. When in doubt, an independent gemmologist can confirm identity and match a stone to its report — that is precisely the work I do in-house at Ministry of Gems, and any buyer can commission it from a qualified professional in their own city.

Which laboratories count

Buyers understandably crave a short list of ‘safe’ labs, and the internet obliges with dogma. A widely shared JewellersReviews piece from December 2025 argues that only three laboratories in the world can be trusted — GIA, Gübelin and SSEF. I understand the impulse, and those three are indeed beyond reproach; but as a buying rule it is too tight, and it would wrongly condemn a great deal of honestly papered goods. Auction practice is the tell: Sotheby’s and Christie’s routinely sell top rubies and sapphires on GRS, SSEF and Gübelin paper, and trade comparisons such as Signed Vintage Jewelry’s laboratory guides also accept AGL, Lotus Gemology, GIT, HRD and IGI depending on the category of stone. The real test of a laboratory is behavioural, not tribal: does it use the harmonised LMHC-style disclosure wording; can its reports be verified online through a domain you typed yourself; and does it refuse to print valuations? A lab that passes those three tests belongs on your list. A lab that fails any of them does not, whatever its acronym.

What a report is — and is not

Finally, calibrate what the document actually gives you, because even genuine reports are widely misread. A report states the stone’s identity — natural corundum, say, rather than synthetic or glass — its measurable properties, and its treatment status as far as testing can detect. That is the factual core, and it is gold. But a report is not a certificate of value: GIA is explicit that no appraisal figure appears on its documents and that a report is not a guarantee. And parts of a report are formally opinion: under ISO/IEC 17025, the accreditation standard for major gem labs, opinions and interpretations are required to be marked as such — a requirement Jack Ogden’s analysis in Gem-A’s Journal of Gemmology notes few labs fully honour — and geographic origin is the leading example, as Chapter VII will show. So read a report the way you would read a surveyor’s report on a house: trust the measurements, weigh the opinions, and remember that it says nothing about whether the price is fair. That last judgement belongs to you, armed with Chapter II. For the field-by-field anatomy of legitimate reports, our Certification Handbook covers what each line means and how the major labs differ.

Chapter V

The Marketplace Bestiary

Live-stream countdown clocks, industrial grinding balls sold as meteorite diamonds, parcels nobody ordered — a field guide to marketplace gem fraud, with dates and case files.

The countdown clock

Begin with the newest beast, because it is the one most likely to find you. Live-stream gem selling — a charismatic host, a tray of stones, a ticking ‘live exclusive’ discount — is a documented scam vector, not a hypothetical one. On 26 March 2023, the Singapore Police Force issued a formal advisory after at least ten victims lost at least S$95,000 buying jade through TikTok live streams in barely three months, from January of that year. The documented tactics: time-bound discounts available ‘only while the stream is live’, counterfeit or substituted items, and in some cases nothing shipped at all. Notice what the countdown clock actually does — it is a machine for deleting your verification time. Every chapter of this book requires minutes or days of checking; the live format is engineered so you have seconds. That is not an accident of the medium. It is the point of the medium. No honest seller of a four-figure stone needs your decision before the stream ends; our own stones sit patiently in the collection for weeks while buyers think, and that patience costs us nothing but interest.

Impossible stones and industrial leftovers

Some frauds sell a story so exotic that the price seems almost reasonable. The current champion: ceramic mill balls — industrial grinding media, roughly 81% aluminium oxide, worth pennies per ball in bulk — being sold online as ‘lonsdaleite’ or ‘hexagonal meteorite diamonds’ at prices from US$20 to US$195,000, per Gemology Resources and Washington University in St. Louis’s meteorite identification pages. Real lonsdaleite crystals are smaller than a micrometre, require X-ray diffraction to identify, and have never been found as visible crystals — let alone polished spheres with seam lines. The tell here is a general one: when the story requires the object to be something science has never once produced at that size, the story is the product.

Lower down the price ladder, the manufacturing is subtler. Those cheap eBay ‘rubies’ with convincing internal ‘feathers’ — the wispy inclusions buyers are taught to read as proof of natural origin — are frequently quench-crackled flame-fusion synthetics: furnace-grown corundum heated and plunged into cold fluid so it fissures like the real thing, or quench-crackled dyed quartz doing the same dance a caste lower, as InVogueJewelry documented. And bulk parcels of ‘emeralds, rubies and sapphires’ from some marketplace wholesalers have been identified as dyed sillimanite — a cheap mineral that drinks dye readily. The lesson stings a little: the ‘look for inclusions’ folk test has been priced in by the counterfeiters. Inclusions can be manufactured. Only identity testing — refractive index, spectra, microscopy in trained hands — actually separates species, which is why Chapters III and IV keep sending you back to real laboratories.

The oldest trick in Bangkok

The ancestor of every online gem scam predates the internet entirely. The Bangkok ‘gem scam’ — a friendly stranger, a tuk-tuk, a ‘government-sponsored, one-day-only’ gem sale, and the promise that the sapphires you buy today will resell at home for a 100% profit, sometimes pitched at two to five times — has operated for more than thirty years, with average losses of US$500–5,000 per victim, per the long-running documentation at Into-Asia and Sea Insider. The stones are real enough, usually; the prices are fantasy; and the accompanying ‘certificates’ come from unrecognised institutes with no industry standing. I raise it not for travel advice but because its DNA is in every modern variant: manufactured urgency (one day only), borrowed authority (government-sponsored), and — the deepest tell of all — the resale-profit promise. Hold that last one as an absolute: any seller who pitches a gemstone by what you will make selling it is running the Bangkok script, whatever the platform. And let me head off the lazy conclusion, because a 2021 UK guide fell into it: the answer is not ‘avoid Asian sellers’. Some of the finest coloured-stone laboratories on earth — AIGS, GIT, Lotus Gemology — are in Bangkok. Fraud is a behaviour, not a postcode.

Reviews, cadmium and the behavioural audit

Two final specimens, then the field method. First: the review count is not evidence. Shill reviews are now a federal offence in the United States — the FTC’s Rule on Consumer Reviews and Testimonials, announced 14 August 2024 and effective 21 October 2024, bans creating, buying or selling fake reviews, explicitly including AI-generated ones, with civil penalties up to roughly US$52,000 per violation. Regulators do not build rules like that against imaginary problems; the brushing parcels of Chapter I were review-manufacturing machines. Second, the hazard nobody prices: a Sunday Times investigation reported by JCK found jewellery sold via Amazon and TikTok Shop — including items advertised as sterling silver — containing cadmium, a carcinogen, far above legal limits. Marketplace risk is not only misdescription; sometimes it is metallurgy.

So can you buy safely on eBay, Etsy or Instagram? Yes — honest dealers genuinely trade there, and the International Gem Society’s position that careful online buying is fine is one I share. The separator is behavioural, and it makes a five-point audit you can run on any seller in ten minutes: full treatment disclosure in the listing itself; paper from a respected laboratory, or an explicit offer to submit the stone to a lab of your choosing; daylight video on request; a real, stated return policy; and traceable payment rails. Run the audit before you fall in love with the stone. Sellers who pass all five are rarer than they should be — and worth keeping.

Chapter VI

The Phrasebook: True Words That Mislead

The most effective deceptions online are built entirely from true words — a phrasebook for ‘created’, ‘genuine’, ‘composite’ and the other sentences engineered to be misread.

The words the law watches

The most dangerous listings contain no lies at all. They are assembled from technically true words arranged so that you do the misreading yourself — and the reason we know exactly which words, is that regulators have had to legislate for them. The US FTC’s Jewelry Guides, revised in 2018, require that a man-made diamond be described as ‘laboratory-grown’, ‘laboratory-created’, the manufacturer’s name followed by ‘-created’, ‘synthetic’, or an equally clear equivalent — placed immediately before the word ‘diamond’, not buried in a description tab three clicks away. The same Guides disallow ‘real’, ‘genuine’, ‘natural’ and ‘precious’ for any manufactured product, and caution against ‘cultured’ for lab-grown diamonds unless it carries qualifying language. So when a listing reads ‘created sapphire’, that single word is the entire confession — it means the stone came from a factory, and the seller is betting you will read it as ‘created by nature’. In Australia the same logic runs through the general prohibition on misleading and deceptive conduct; the FTC vocabulary is still worth learning because it is the de facto dialect of the international online trade. For the honest version of this entire subject — how lab-grown stones are made, what they cost, and when they are a perfectly good choice — this library’s Natural vs Lab-Grown Handbook gives it full and fair treatment.

‘Composite ruby’: a confession in two words

Now the phrase that does the most damage in the coloured-stone world. A lead-glass composite ruby, as Chapter II established, is low-grade corundum whose fractures have been filled with lead glass — and the filling is nearly invisible because the glass’s refractive index of 1.76–1.77 sits almost exactly on corundum’s 1.76–1.78, per the Skyjems encyclopedia. The FTC’s position is blunt: calling such a stone an unqualified ‘ruby’ is deceptive; the acceptable terms are ‘composite’, ‘hybrid’ or ‘manufactured’, and only when effectively qualified to convey that the product lacks the named stone’s properties and needs special care. This is not theoretical enforcement muscle. Macy’s — the American department-store chain — was sued in 2014 over lead-glass-filled rubies sold without disclosure, the highest-profile consumer case establishing that a composite sold as ‘ruby’ is actionable misrepresentation.

And the special care is real, which is why the disclosure is mandatory rather than polite. The lead-glass filler sits at Mohs 5–6 against corundum’s 9, and it is damaged by household acids as mild as lemon juice, by ultrasonic cleaners, by steam, and by a jeweller’s torch — the glass etches, dissolves or discolours. Picture the honest failure case: a buyer takes her ‘ruby’ ring for a routine resize, the bench jeweller applies heat as he would to any corundum, and the stone comes back looking like a cracked windscreen. Nobody in that story did anything wrong except the person who left a word out of a listing years earlier. When you see ‘composite ruby’ disclosed plainly and priced at composite prices, you are looking at an honest seller of an inexpensive product. When you see the same stone as ‘genuine ruby, special price’, you are looking at the Macy’s case waiting to happen.

The scandal that built the rules

If you want to know why modern disclosure rules are so uncompromising, the answer is a scandal with a date. Around 2000, treaters in Thailand discovered that diffusing beryllium into sapphire at high temperature could transform cheap material into glowing padparadscha-like oranges and pinks — and the output was sold into the trade as conventionally heated sapphire at natural-colour prices. The world’s laboratories took roughly twelve to eighteen months to work out what they were looking at, as Ganoksin’s account records, and in that window an enormous tonnage of misdescribed stone changed hands. The episode did lasting institutional damage — and lasting institutional good, because it drove today’s mandatory treatment-disclosure norms. The modern rulebook is CIBJO’s Gemstone Blue Book, and its standard is worth quoting in spirit: sellers must disclose all treatments even if the purchaser does not ask, and where treatment status cannot be confirmed or economically tested, the suspected treatment must be disclosed rather than passed over in silence. Nothing concealed, nothing minimised. That is the standard I hold our own listings to, and it is the standard you are entitled to hold over any seller on earth.

What honest language sounds like

Calibrate your ear with the base rates, because they make honest language easy to recognise. An estimated 90–95% of all commercial sapphires are heat-treated before they reach a buyer, per Ceylons Munich and The Natural Sapphire Company — heat is permanent, stable, and accepted by every major gemmological body. Nearly all emeralds — commonly cited around 99% — are clarity-enhanced, most often with oil, graded by GIA as F1 minor, F2 moderate or F3 significant, per the International Gem Society. Treatment is therefore the normal condition of the market, and honest sellers say so without flinching: ‘heated sapphire, no other enhancements’, ‘minor oil (F1), lab report attached’. Dishonest language works the two ends against the middle — either weaponising the base rate (‘all sapphires are heated, so unheated premiums are a con’) or trading on silence (‘100% natural’, meaning natural species, saying nothing about treatment). And an undocumented ‘unheated’ claim is its own red flag: precisely because 90–95% of stones are heated, unheated commands a serious premium, and a serious premium claimed without laboratory paper is a cheque written on an empty account. Our Treatment Register publishes, for every treatment we will and will not sell, what it is, how it is detected and what it does to price — and every stone in our collection states its treatment status in the listing, not in the fine print. That is not virtue; it is just what the words are for.

Chapter VII

The Photograph and the Origin Story

Cameras are not eyes and origin is an opinion — how to discount the photograph, ask for the right video, and weigh an origin story at its true evidentiary value.

Cameras are not eyes

Even when the stone is genuine, the treatment disclosed and the paper real, one gap remains between the listing and your hand: the photograph. Professional gem photography is a craft of controlled lighting, polarisation, contrast enhancement and macro magnification, and its practitioners are candid about the result — Cecile Raley Designs, a dealer who writes unusually honestly about her own trade, puts it plainly: there is a major discrepancy between how a stone is photographed for sale and how it looks in your hand. Note that this is the honest end of the business. At the other end, the photo-editing guides themselves draw the line for us: Retouching Zone’s jewellery-editing guidance warns that pushing vibrance and saturation beyond minimal levels distorts a stone’s natural colours and introduces hues not typically found in gemstones. When an editing manual for sellers has to say ‘stop doing this’, you may safely conclude it is being done at scale. The jade trade shows the ceiling of the problem: as the Cuijuan jadeite buying guides document, filters and saturation adjustments routinely make stones look greener, brighter and more translucent than they will ever look in daily wear.

So discount the still image — every still image, including ours. A photograph is an argument, not a record. What we do about it at Ministry of Gems is shoot in daylight and describe the stone’s behaviour in prose as well; what you should do about it, anywhere you shop, is refuse to buy a coloured stone on stills alone.

Two cheap checks close the gap from your side of the screen. Reverse-image-search the listing photos: stolen images — a fine dealer’s photograph fronting a stranger’s listing — are among the commonest marketplace tells, and the search takes thirty seconds. And remember that your own screen is a second distortion layer stacked on the seller’s first: the same photograph reads differently on a phone at full brightness at midnight and on a laptop by a window at noon. If a purchase is serious, look at the images on more than one device, in daylight, before you decide what colour you think you have been shown.

The video to ask for

The corrective is specific footage, and you are entitled to request it. Ask for three things, per the practical standard the jadeite guides articulate for all coloured stones: video in natural daylight, hand-held movement footage so you can watch the colour shift as the stone tilts, and a close-up rotation clip. Static images are not enough — colour in a cut gemstone is a moving phenomenon, and video is much harder to flatter than a frame. A seller with an honest stone finds this request routine; a seller with an edited photograph finds it existential. Watch for the giveaways in what arrives: footage shot only under shop halogens (which warm every stone), only at one angle, or with the stone never leaving a white light-tent. And when your stone arrives, judge it the way a gemmologist does: in soft, indirect daylight from a south-facing window — that is the gentle, even light in our southern hemisphere; my northern colleagues use a north-facing one — at arm’s length, over several days and several skies. A stone you love in Tuesday’s overcast and Saturday’s sun is a stone you will love for decades. That standard, incidentally, is exactly the one to bring to our own opals and Australian sapphires — stones whose play and bi-colour behaviour no still frame has ever captured honestly.

Origin is an opinion

Now the subtler seduction: the origin story. ‘Ceylon sapphire.’ ‘Burmese ruby.’ The names carry real price power — and buyers deserve to know what stands behind them. Here is the trade’s own answer, from Jack Ogden’s analysis in Gem-A’s Journal of Gemmology: geographic origin on a laboratory report is an opinion, not a measurement. Laboratories reach origin conclusions by consensus of gemmologists — educated guesses, in Ogden’s candid phrase — because stones from different deposits can be genuinely indistinguishable; certain Madagascar sapphires resemble Kashmir stones closely enough to trouble the best instruments in the world. The documented case that should recalibrate you permanently: a 30.08 ct sapphire at auction carried three laboratory reports giving two different origins — Sri Lanka and Myanmar. Nobody in that story was corrupt. Origin determination is simply hard, and major labs regularly disagree.

The honest institutions wear that difficulty openly: SSEF and Gübelin state a geographic origin only when confident, and when the data is inconclusive the report reads ‘origin undetermined’. Hold that phrase close, because it is a trust signal in its purest form — an honest lab admits uncertainty; a fake one never does. Which yields the buying rule: an origin claim without a top-laboratory report is a story, not a fact; and even with the report, it is a professional opinion whose price premium you should weigh accordingly. We sell Ceylon sapphires, plainly labelled as such, on precisely that footing — the origin opinion comes from the laboratory, named on the paper, never from our romance about our own history. For how provenance is actually established, documented and honestly sold — a different and richer thing than a single word on a report — this library’s Provenance Handbook takes the subject on in full.

Chapter VIII

Payment Armour and the Verification Clock

Your payment method is the last line of defence and the first decision — how chargebacks, PayPal and marketplace guarantees actually work, and why the clock starts the moment the parcel lands.

Choose the rail before the ring

Everything in this book so far reduces your odds of buying badly. This chapter is about what happens if you buy badly anyway — and the strange truth is that most of your protection is decided before you click, by the payment rail you choose. The ACCC and Scamwatch guidance is unambiguous: pay by credit card or a secure service such as PayPal rather than by debit card or bank transfer, so that a chargeback remains possible; and treat any request for bank transfer, gift cards or cryptocurrency as a scam red flag in its own right. A seller who steers you away from reversible payment is telling you, in the politest available dialect, that they expect you to want the money back. If the worst happens, contact your bank immediately — recovery odds decay by the hour. And a note for home-ground purchases: when you buy as a consumer from an Australian business, the Australian Consumer Law’s consumer guarantees apply to the purchase — goods must match their description, among other things — which sits alongside, not instead of, everything below.

The four windows

Your armour comes in four pieces, each with its own clock, and the numbers are worth memorising. Credit-card chargebacks: across Visa, Mastercard, Amex and Discover, the window generally runs up to 120 days from the transaction or delivery date, and it covers ‘not as described’ and ‘defective merchandise’ disputes on online purchases, per the chargeback-industry references at Chargebacks911 and Chargeflow. PayPal: Purchase Protection applies to ‘Goods and Services’ payments only — never, ever ‘Friends & Family’, which carries no protection at all and is therefore the scammer’s favourite request — and covers Significantly Not as Described claims: a materially different item, misrepresented condition, undisclosed missing features. PayPal reimburses the purchase price plus original shipping on eligible claims, and disputes must be filed within 180 days of payment. One candour note, in the house style: some secondary guidance describes a shorter not-as-described sub-window measured from delivery, so file early rather than testing the boundary, and read the current PayPal user agreement for your purchase. eBay: the Money Back Guarantee lets you open an ‘item not as described’ case up to 30 days after actual or estimated delivery, eBay enforces refunds regardless of the seller’s own return policy, and the seller gets three business days to resolve before eBay steps in. Etsy: Purchase Protection cases must be opened within a 30-day window, covering items that arrive seven or more days after the maximum estimated delivery date or that differ significantly from the description or photos.

The ten-day rule

Now put the windows against the workflow, because this is the practical insight that no guide I have read connects. The marketplace windows are 30 days. Independent laboratory verification — shipping a stone to a lab, queueing, testing, reporting — can take longer than that, and even a local gemmologist’s appointment can drift a fortnight if you let it. Which means a buyer who does everything right but slowly can hold a worthless stone and a dead claim. So here is the rule I give friends, and it is the closest thing this book has to a single commandment: the verification clock starts the moment the parcel lands, and your target is days, not weeks. On arrival day: photograph the stone and all paperwork in daylight, beside the listing images. That week: weigh and measure it against the report as Chapter IV taught, verify the report through a domain you typed yourself, and book the stone in front of an independent gemmologist — within roughly ten days at the outside, faster if the seller is overseas. If anything fails, file inside the window with your documentation attached. A dispute filed on day 12 with photographs, a weight discrepancy and a gemmologist’s note is close to unanswerable. The identical dispute on day 45 is a sad story. Sydney readers are welcome to bring an online purchase to me for exactly this check — talk to our gemmologist — and no, it does not have to be our stone. Especially if it is not our stone.

What honest looks like

I want to end the way none of the scam guides do — not with fear, but with a worked example of the thing itself, because after seven chapters of forgeries you should see the genuine article in full light. Here is what an honest online listing for a fine stone looks like, element by element. Species and variety, stated plainly: natural corundum, blue sapphire. Treatment status, in the listing body, not a tab: unheated, with the laboratory’s phrasing — ‘no indications of heating’ — quoted exactly. The laboratory named, the report number printed, and the report verifiable through the lab’s own site by a URL you type yourself. Origin given as the lab’s opinion, attributed as such. Carat weight and millimetre measurements you can check against the stone on a $30 scale. Daylight video, offered before you ask. A price that sits inside the ranges Chapter II taught you — a four-to-five-figure number for a fine unheated Ceylon stone, because the honest version of that stone cannot be cheap. A stated return window. Traceable payment. And documentation that follows the stone for life — in our house, a Digital Gemstone Passport carrying the identity, treatment status and paper trail together, plus free lifetime re-authentication whenever you or your heirs want the file reopened. That is the standard in action across our unheated sapphire collection, and I commend it to you not as advertising but as calibration: once you have seen every element present in one listing, you will notice every listing where something is missing. The absences are the tell. Buy where nothing is missing — from us or from anyone — and the internet becomes what it always promised to be: the widest, best-lit gem counter ever built.