Zoisite

What is zoisite? One formula, two crystal systems

Zoisite is the orthorhombic form of Ca2Al3(Si2O7)(SiO4)O(OH); clinozoisite is the monoclinic form of the same formula. The Handbook of Mineralogy zoisite sheet gives point group 2/m 2/m 2/m, space group Pnma, prismatic crystals to 10 cm, typically deeply striated parallel to [010] and poorly terminated, and columnar to compact masses. Cleavage is perfect on {010} and imperfect on {100}; hardness is 6 to 7, measured density 3.15 to 3.36. The sheet records it as dimorphous with clinozoisite, and the clinozoisite sheet says the same in reverse.

What the IMA list records. A case-insensitive scan of every line of the September 2026 list finds 3 containing zoisite: clinozoisite, zoisite and zoisite-(Pb). Two of them, zoisite and clinozoisite, have exactly the same formula string. Zoisite is grandfathered, G, 1805, with Austria as the country of first description; clinozoisite carries a 2006 special-procedure entry, also Austria. Armbruster et al. 2006, Recommended nomenclature of epidote-group minerals (European Journal of Mineralogy 18, 551) records that the mineral was first named saualpite for the Saualpe in Carinthia, where it occurs in eclogites, and that Werner chose the name zoisite in 1805 to honour Siegmund Zois, from whom he obtained the holotype specimen from the Saualpe.

Group membership depends on the reference. The Handbook sheet lists zoisite in the epidote group. The 2006 report limits the group to monoclinic species and states that zoisite, as the orthorhombic polymorph of clinozoisite, is not considered a member; it adds that zoisite is the one orthorhombic species originally assigned to the group. The epidote–clinozoisite series is set out on epidote, and its density argument is not repeated here.

Can you tell zoisite from clinozoisite by weight? No: the shared window

A balance cannot separate zoisite from clinozoisite, because their density ranges share 3.21 to 3.36. Worked as 20 − 20/D on the measured densities, a 20.00 g piece reads 13.65 to 14.05 g in water as zoisite and 13.77 to 14.08 g as clinozoisite. The overlap, 13.77 to 14.05 g, covers 71% of zoisite’s density span and 88% of clinozoisite’s. Only a reading under 13.77 g points to zoisite, and only one over 14.05 g to clinozoisite; the method and its limits are on how to measure specific gravity at home.

The other hand tests fail too. Zoisite is hardness 6 to 7 and clinozoisite 6.5. Both are biaxial (+): zoisite’s 2V is 0° to 69°, clinozoisite’s 14° to 90°. Both are prismatic and striated parallel to [010], and each has one perfect cleavage, {010} in zoisite and {001} in clinozoisite. What separates them is the crystal system, which needs an X-ray pattern or an oriented optical study, not a scratch or a scale. The hardness method is on how to test mineral hardness properly; reading a broken surface is on cleavage or fracture.

As a rule. A pale prismatic calcium aluminium silicate labelled zoisite is a determination to the pair unless the label records how the system was established. Deeply coloured material is easier: pistachio-green with density over 3.38 points to epidote, and violet-blue gem crystals from Tanzania were identified as zoisite by X-ray work in 1967 and 1968, as Hurlbut records.

Tanzanite, thulite and anyolite: what the zoisite variety names mean

Tanzanite is blue to violet zoisite, thulite is pink zoisite, and anyolite is a green zoisite rock with ruby. GIA’s tanzanite description calls tanzanite the violet blue to blue violet variety of the mineral zoisite and says most gems are heat treated. Hurlbut 1969, Gem zoisite from Tanzania (American Mineralogist 54, 702) describes the Tanzanian crystals: on 12 crystals 40 forms were identified, 17 of them new; density 3.355; hardness between 6½ and 7; and red-violet, deep blue and yellow-green pleochroism, which led to their first being taken for cordierite or dumortierite.

The vanadium ledger, worked. The zoisite sheet’s analysis is Hurlbut’s, with the locality printed as Umba Valley: SiO2 39.55, Al2O3 33.39, V2O3 0.20, CaO 24.44 as corrected on the sheet. Normalised on 12.5 anhydrous oxygens it gives Ca 1.987, Al 2.986, Si 3.001 and V 0.012, which matches Hurlbut’s own table. The sheet notes that CaO was originally given as 22.44%; with 22.44 the printed oxides sum to 97.74, with 24.44 to 99.74, within 0.05 of Hurlbut’s printed total of 99.69, so the corrected figure is the one that fits. Ideal zoisite is 24.68% CaO, 33.66% Al2O3, 39.67% SiO2 and 1.98% H2O. Hurlbut attributes the colour to vanadium and the change to sapphire-blue on heating to V3+ becoming V4+.

Thulite and anyolite. The International Gem Society tanzanite (zoisite) article calls thulite pink zoisite, with unionite and rosaline as trade names, and describes ruby in zoisite, also known as anyolite, as a lapidary rock of chrome-rich green zoisite, black hornblende and opaque ruby. The 2006 report lists thulite and unionite among obsolete names or synonyms for zoisite cited by Dana in 1896, and says anyolite means green in the language of the Masai. The gem society’s thulite figures, specific gravity 3.09 and biaxial (−), sit outside the zoisite sheet’s 3.15 to 3.36 and biaxial (+), so a thulite label is a colour name to be confirmed, not a determination. The ruby in anyolite is corundum. Whether a blue stone should be called tanzanite or zoisite is answered on is zoisite the same as tanzanite.

Where zoisite is found: the Saualpe, Merelani and Shigar Valley

The Distribution field calls zoisite widespread and lists classic localities for fine crystals and decorative rocks: the Saualpe, Carinthia, and the Zillertal, Tirol, Austria; near Trondheim, Norway; Goshen and Williamsburg, Massachusetts; the Keystone quarry near Cornog, Pennsylvania; Milford Heights, Maryland; around Spruce Pine, North Carolina; Ducktown, Tennessee; the Tunk Creek area east of Riverside, Washington; Asbestos, Quebec, Canada; the Juarez district, Baja California, Mexico; exceptional gem material in the Merelani Hills, Lelatema Mountains, 40 km southeast of Arusha, Tanzania; and well crystallised material from Alchuri village, Shigar Valley, Baltistan, Pakistan.

Where it grows. The sheet places zoisite typically in medium-grade regionally metamorphosed crystalline schists formed from rocks relatively high in calcium, and in eclogites and blueschist-facies rocks, with garnet, albite, biotite, “hornblende”, quartz and calcite. That is the company of vesuvianite and epidote too, which is why pale green prisms from such rocks need the checks above before a name goes on the label.

Related pages. The orthorhombic blue look-alikes are cordierite and dumortierite; benitoite compares tanzanite with benitoite. The rest of the reference section is at the species index. We hold no catalogue and nothing here is offered for sale; the wanted list is the only commercial route.

Questions