Why Uvarovite Usually Appears as Small Green Crystals—and What the Faceted Exception Changes

A good uvarovite specimen can look less like a conventional gemstone than a piece of rock dusted with green light. Hundreds of crystal faces catch the eye, yet there may be no single crystal that could sensibly become a faceted stone. The surface is not a lesser version of the material. It is often the geological form in which the material grew and survived.

Identity comes before colour

Uvarovite is a mineral species in the garnet group. Its ideal endmember formula is Ca₃Cr³⁺₂(SiO₄)₃. In the garnet structure, calcium occupies the X, or dodecahedral, site; trivalent chromium occupies the Y, or octahedral, site; and silicon occupies the tetrahedral Z site. This site-based identity is more useful than a label such as “chrome-green garnet”, because several garnets can be green.

An ideal formula defines the compositional pole used to name and classify a mineral. Natural garnets need not match that pole perfectly. Aluminium can substitute for chromium towards grossular composition, while ferric iron can introduce an andradite component. The analysed Outokumpu specimen in the RRUFF database, for example, contains substantial aluminium at the Y site even though chromium remains dominant.

Colour cannot identify uvarovite on its own. Green grossular, tsavorite and demantoid can overlap it visually. So can a mixed garnet whose chemistry falls somewhere between named endmembers. Crystal shape, dark matrix and intense green are useful observations, but none supplies a composition.

Why the surface is typical

The word “surface” is a clue. In the Moa-Baracoa chromitite of Cuba, researchers described uvarovitic garnet in millimetre-scale veins and along altered chromite. The crystals formed from hydrothermal fluid in open fractures; chromite was also corroded and replaced along grain boundaries and pull-apart cracks. The textures are consistent with local fluid–rock interaction along the vein walls.

That study is especially useful because it preserves two boundaries. First, the material was compositionally zoned across the uvarovite–grossular series, from about 17 to 63 mol% uvarovite. Not every green grain in the suite was uvarovite-dominant. Second, it was one deposit. Its textures show how a drusy surface can form, not a law governing every uvarovite occurrence.

The broader mineral record points to related settings. The Handbook of Mineralogy places uvarovite in the hydrothermal alteration of chromite-bearing serpentinite, and also in metamorphosed limestone and skarn where dolomite and chromite react. These are chemically selective environments: calcium, chromium and silica have to meet while garnet can crystallise. A reasonable geological inference is that, when this happens at a fracture wall, chromite boundary or narrow reaction front, the preserved result can be an attached field of crystals rather than a few detached gems.

What the evidence does not show

No authoritative source has shown that chromium itself inhibits uvarovite growth. A locality study discusses fluid composition, transport near the fluid–solid interface and surface kinetics as controls on garnet chemistry and form. Those factors may help explain dense, small crystals in a particular vein. They do not justify a universal sentence that chromium “stops” the crystals growing.

The most defensible explanation is therefore spatial. Many documented crystals occupy confined, localised mineralising surfaces. Their small size is real; the proposed single-element growth rule is not established.

What the rare faceted exceptions change

Uvarovite can grow larger than its familiar crystal crust suggests. The Handbook records crystals to 4.5 cm and calls those from Outokumpu exceptional. A primary study of that Finnish locality likewise reports individual crystals as large as 45 mm. External size, however, says nothing by itself about transparency, fractures, darkness or the clean volume available to a cutter. A 4.5 cm crystal is not automatically 4.5 cm of facet-grade rough.

The best-documented gemological exception in the sources reviewed here is more specific. A Fall 1993 Gems & Gemology abstract of work by U. Henn and H. Bank describes a 0.95 ct dark-green faceted stone from Outokumpu. It was singly refractive, had a specific gravity of 3.76 and an index above the limit of a standard refractometer. Electron-microprobe data yielded calculated endmember proportions of 59.80% uvarovite, 38.24% grossular and 1.96% andradite.

That composition makes the exception more revealing, not less. It is a uvarovite-rich mixed garnet, consistent with the solid solution seen in natural material. The specimen proves that suitably transparent, uvarovite-dominant material can be faceted. One documented stone cannot establish regular availability, typical size or a market category of pure faceted uvarovite.

Drusy, cluster and matrix describe presentation

“Drusy” or “druzy” means a coating of small crystals on a rock surface. “Cluster” is a looser description for crystals growing together. Neither is a mineral species or a special uvarovite variety. A listing that says “uvarovite drusy” still needs the mineral identification to be supported separately.

Presentation changes how the object should be judged. Look for even coverage if that is important to the design, but also inspect bare patches, crushed points, cracks through the matrix and areas that appear repaired. In jewellery, notice whether a rim shields the crystal surface and whether prongs or edges can catch clothing. A bright green carpet may be visually continuous while its physical support is not.

Four green garnet labels that are not interchangeable

A faceted green stone removes one visual clue rather than solving the identity. It no longer shows the typical uvarovite-on-matrix presentation, and its polished appearance may resemble other green garnets more closely. This is where a seller’s species name needs evidence, especially when rarity is part of the description.

What a buyer can observe, and what remains unknown

Without special equipment, a buyer can make useful material observations:

• colour under stated lighting, including whether the green is even or patchy;
• approximate crystal size, coverage and the presence of exposed sharp points;
• chips, missing areas, matrix cracks, loose sections and visible repairs;
• how a setting protects—or exposes—the crystal field;
• whether the seller discloses matrix, coatings, adhesives, repairs or treatment.

Those observations cannot determine the garnet species, endmember proportions, locality, matrix identity, treatment history, assembly method or structural integrity. They also cannot show that a larger-looking crystal is transparent through enough volume to facet.

Documentation occupies a middle ground. A seller may state a locality or species, and a report may record tests, but the wording should be tied to the exact object. On a matrix holding many tiny crystals, ask which crystal or area was tested. A result from one point should not silently become a claim about every green grain on the surface.

How identification is built

Standard gemmological observations can narrow the field on a loose or faceted stone: optic reaction, refractive index, specific gravity and absorption spectrum all contribute. The Outokumpu exception required several of these observations plus microprobe chemistry. A single number or a green colour was not enough.

For very small crystals on matrix, the practical route may be targeted. Raman spectroscopy can identify a minute exposed garnet phase by its vibrational spectrum. X-ray diffraction can resolve crystalline phase or structure, while electron-microprobe or another suitable compositional analysis is needed to quantify chromium, aluminium and iron and calculate endmember proportions. GIA cautions that mineral species depends on both composition and structure; estimating garnet composition from Raman peak positions alone is not accurate enough for practical use.

The USGS record for an intensely green National Museum of Natural History specimen shows why restraint matters. X-ray diffraction found garnet with other phases, and chemical work indicated a significant uvarovite component. Even then, the record states that the identity and proportions of the other garnet components were not known. Bright green was evidence to investigate, not a final name.

Care and wear depend on the whole object

The Handbook gives uvarovite a Mohs hardness of 6.5–7 and describes it as brittle. Hardness concerns resistance to scratching; it does not protect tiny crystal points from impact, the matrix from cracking or the interface between the two from failing. A cluster also has many edges that can snag.

GIA describes drusy as delicate and better suited to pendants, earrings and brooches. Rings and bracelets can work when the design protects the surface and the wearer accepts more care. This is a setting judgement, not a promise that one jewellery category is safe in every construction.

For a known, untreated garnet in a water-safe mounting, GIA recommends warm soapy water and advises against steam. A uvarovite cluster is a composite object, so the matrix, fractures, setting and any adhesive or repair set the stricter limit. If those are unknown, avoid soaking and vigorous brushing. Do not use an ultrasonic cleaner unless a qualified professional has examined the entire piece and confirmed that its construction is suitable. Use minimal pressure, prevent snagging while drying, and stop wearing the piece if crystals or matrix feel loose.

Questions worth asking before buying

• Is “uvarovite” based on visual opinion, supplier documentation or independent testing?
• If tested, which crystal or area was examined, and by what method?
• Is the locality documented or merely associated with the material in general?
• What is the matrix, and are there disclosed fillers, coatings, adhesives or repairs?
• For jewellery, how is the crystal surface protected and what cleaning method does the maker approve?

These questions do not demand laboratory work for every modest specimen. They keep observation, seller documentation and independent testing in their proper places. The rarer and more specific the claim—especially a faceted uvarovite claim—the more precise the evidence should be.

References

• Grew, E. S. et al. (2013), “Nomenclature of the garnet supergroup”, American Mineralogist 98, 785–811. DOI; institutional PDF.
• Handbook of Mineralogy, Uvarovite, version 1.2.
• Proenza, J., Solé, J. & Melgarejo, J. C. (1999), “Uvarovite in podiform chromitite: The Moa-Baracoa ophiolitic massif, Cuba”, The Canadian Mineralogist 37, 679–690. Full paper.
• von Knorring, O., Condliffe, E. & Tong, Y. L. (1986), “Some mineralogical and geochemical aspects of chromium-bearing skarn minerals from Outokumpu, Finland”, Bulletin of the Geological Society of Finland 58. DOI; full paper.
• GIA, Garnet Description.
• GIA, Fall 1993 Gems & Gemology Gemological Abstracts, “Faceted dark-green uvarovite from Outokumpu, Finland”, abstracting U. Henn and H. Bank. PDF, p. 222.
• USGS Spectroscopy Laboratory, Uvarovite NMNH106661 analysed specimen record, Open-File Report 03-395.
• RRUFF Project, Uvarovite R061041, Outokumpu, XRD- and chemistry-confirmed sample sourced from the Royal Ontario Museum.
• Handbook of Mineralogy, Grossular and Andradite.
• GIA, “What’s the difference between the two green garnets: tsavorite and demantoid?”
• GIA, “Raman Spectroscopy and X-Ray Diffraction: Phase Identification of Gem Minerals and Other Species”, 2024.
• GIA, Garnet Care and Cleaning Guide.
• GIA, “Dreamy Drusy: Minute Crystals Shimmer in Dramatic Designs”. Its general drusy definition and setting guidance are used here; its obsolete single-locality statement for uvarovite is not.
• Smithsonian Institution, National Museum of Natural History, Uvarovite specimen NMNH 93602-01.