A small correction before anything else. Last week this newsletter said full results from Sotheby's Hong Kong would be in "next Tuesday." That sale is September 17, a Thursday. This is next Tuesday, and the sale has not happened yet. The spinel result is still two days out. Apologies for the miscount, results next week.

In the meantime, something more useful than waiting: a pattern that has run quietly under every edition since June. Every focus stone this newsletter has covered carries a different relationship to treatment, and every one of those relationships has a real number attached. Spinel gets nothing done to it. Ruby usually gets heated. Tanzanite almost always does. Padparadscha's premium for staying natural is wider than any other stone in the index. Pull all eight together and a single table emerges that no single edition could show on its own.

Key Takeaways

→ Sotheby's Hong Kong, two days out. The September 17 High Jewelry sale, and the unheated Burmese spinel pair inside it, is still ahead of us. Full results and the multiple-over-estimate read land in next week's edition.

→ The untreated premium is now a market-wide data point, not a connoisseur's opinion. Trade sources tracking 2026 pricing put the general unheated premium for fine sapphire at 30 to 50 percent over heated material of comparable quality, with some houses now testing UV fluorescence as a standard pricing input for ruby and pink sapphire (CaratX; Jewelry Appraisal Denver).

→ The "No Oil" designation is doing to emerald what unheated does to sapphire. 2026 trade guides describe a stone certified with no clarity enhancement as now commanding roughly double the price of an otherwise identical stone with minor oil, a premium wide enough that treatment disclosure has become the primary lever on emerald value (Gems and More).

The index tracks 8 reference stones, weekly. Each row holds until a new dated transaction moves it. Sotheby's Hong Kong has not happened yet, so every row is exactly where it was last week.

Auction prices realized, not retail. Sources: Christie's, Sotheby's, Gemfields, Bonhams, GemGuide, Pala International.

Notable this week: read this table next to the focus below and every row tells a second story. The stones near the top, sapphire, alexandrite, Paraiba, all carry meaningful treatment premiums or treatment risk. The stones that are simply, structurally untreated, spinel and tsavorite, sit in the middle of the table despite offering the cleanest paperwork in the index. Price does not reward the absence of treatment on its own. It rewards rarity that happens to come with clean paperwork.

Why this table has never existed in one place before. Each stone in this index got its own deep dive because each has a genuinely different relationship to treatment. Put together, they form something more useful than any single edition: a map of where "untreated" is a real price lever, where it is simply the default, and where it barely matters at all.

Trade-estimate ranges compiled from CaratX, Gems and More, and this newsletter's own focus editions. These are benchmarks, not realized transaction data, and should be read as a band.

The pattern the table reveals. Treatment only becomes a price lever where it is common enough to create a real choice. Ruby is usually heated, so an unheated stone is genuinely scarce and commands a real multiple. Sapphire follows the same logic at a smaller scale. Emerald almost always carries some filler, so a credible no-oil report is rare enough to roughly double the price. But where a stone is naturally, structurally untreated across the board, spinel, tsavorite, and functionally alexandrite too, there is no separate premium to earn, because there was never a treated version to be scarce against. Untreated is simply what the stone is, and the market prices it on rarity and beauty instead.

Where this cuts against instinct. It is tempting to assume "never treated" is an automatic value advantage. The table says otherwise. Tsavorite and spinel are two of the cleanest stones in this entire index by disclosure standards, and neither commands anything like Kashmir sapphire or fine Paraiba pricing, because their value ceiling is set by demand and rarity, not by the absence of a treatment line on a lab report. Meanwhile tanzanite, almost universally heated, trades at the bottom of the table not despite the treatment but for entirely unrelated reasons, single-source geology and a market built mostly on one company's marketing rather than centuries of demand.

The one genuine exception. Padparadscha, covered here two weeks ago, breaks the pattern in the other direction. Because the color itself is chemistry-dependent and heat can nudge a stone toward the definition without fully creating it, an untreated stone that lands naturally inside the window is priced as an entirely different category, 2 to 3 times the heated price for comparable material. That is the widest treatment-based gap in this whole table, wider than ruby's, because color-window rarity and treatment rarity compound on top of each other in a way no other stone in the index replicates.

1. Ask what "untreated" is competing against. The premium exists only where treatment is the norm. An unheated ruby matters because most rubies are heated. An unfilled emerald matters because most emeralds are filled. A naturally untreated spinel matters less as a premium and more as a baseline, because there was never a treated majority to stand out from.

2. The report has to say two different things, and know which one matters. For ruby, sapphire and emerald, the treatment line is the whole story. For alexandrite, the natural-origin line matters more than any treatment disclosure, because synthetic material is the real risk. Read every lab report for the specific claim that actually moves that stone's price, not just for the presence of a certificate.

3. Rarity sets the ceiling. Treatment status sets the discount underneath it. No amount of clean paperwork turns tsavorite into Kashmir sapphire. But within any one stone's own market, the treatment line can still be the difference between the top and the middle of that stone's own price range. Both things are true at once, and conflating them is the most common pricing mistake in this market.

→ September 17, Sotheby's Hong Kong. Two days out. The unheated Burmese spinel pair is the direct test of everything in this edition's table: does the market price an untreated spinel with a real premium, or does it confirm spinel's baseline-untreated status and price purely on size and saturation.

→ GIA's expanded origin services. Now running across ruby, spinel, Paraiba, opal and more since January. As more stones get formal origin reports, watch whether origin premiums start compressing the gap between "documented" and "undocumented" untreated stones.

→ October's Gemfields Mixed-Quality ruby auction. Still the next like-for-like read on the ruby row, and a chance to see the heated-versus-unheated premium show up in real Mozambican material rather than trade estimates.

Pick any stone you own or are considering, and ask a more specific question than "is it treated." Ask what a treated version of that exact stone would look like, and how common it is. If treated is the overwhelming norm, untreated is worth real money. If untreated is already the norm, you are not buying a premium, you are buying the baseline, and the price should reflect rarity and beauty instead.

— Carat Brief

Carat Brief is editorial. Nothing here is investment, legal, or tax advice. Gemstones are illiquid and require certified verification before purchase. Work with credentialed labs: GIA, AGL, SSEF, Gübelin, Lotus Gemology.

Sources: CaratX; Gems and More; Jewelry Appraisal Denver; GIA; and this newsletter's prior focus editions on spinel, alexandrite, tanzanite, tsavorite and padparadscha. Premium ranges are trade estimates and vary by source, size and origin.