Staurolite
Staurolite is Fe2+2Al9Si4O23(OH), hardness 7 to 7.5, density 3.74 to 3.83. Its 90° crosses are the less common twin; Roberts in 1934 found fakes cut from talc.
Staurolite is an iron aluminium silicate, Fe2+2Al9Si4O23(OH) on the September 2026 IMA list: monoclinic but pseudo-orthorhombic, hardness 7 to 7.5, density 3.74 to 3.83. Its name is Greek for a cross. The Handbook records 60° twins as common and 90° cruciform twins as less common, and Roberts documented talc-schist fake crosses in 1934.
In short
- Two crosses, and the common one is not the famous one. The Handbook staurolite sheet records twins at 60° on {231} as common and 90° cruciform twins on {031} as less common. Roberts, Virginia staurolite as gems (American Mineralogist 19, 549, 1934) found the same in Henry and Patrick Counties, Virginia, yet the crosses offered as gems there were almost all of the right-angle pattern, which, he wrote, leads the collector to suspicion.
- The Roberts check. Roberts found artificial crosses cut from a grey talcose schist of hardness about 2½ to 3, filed to shape and soaked in linseed or other oil to darken them; some had holes bored to imitate the cavities of weathered crystals. Staurolite is 7 to 7.5. Talc itself reads 12.25 to 12.93 g in water for a 20.00 g piece, staurolite 14.65 to 14.78 g: 1.72 g apart.
- Pits are not proof. Roberts says natural crystals develop small cavities on weathering, which at times identifies them, but that holes were bored into the artificial ones to make them look genuine. The International Gem Society staurolite article also records imitations pockmarked to simulate weathering. A pitted surface is consistent with staurolite; it does not prove it.
- Formula, worked. On standard atomic weights the IMA formula is 28.21% SiO2, 53.86% Al2O3, 16.87% FeO and 1.06% H2O. The sheet’s Black Mountain, New Hampshire analysis reads 27.82%, 54.91%, 12.39% and 2.26%: less iron, with lithium, magnesium and zinc in its place, and 2.14 times the ideal water. The IMA list has two more staurolite-root rows, magnesiostaurolite and zincostaurolite.
- Where. The Distribution field calls staurolite widespread and names Pizzo Forno and Alpe Piona, Ticino; Mt. Greiner, Zillertal; the Keivy massif, Kola; large twins in Finistère and Morbihan, Brittany; Fanzeres, Portugal; Franconia, New Hampshire; Chesterfield, Massachusetts; Blue Ridge, Fannin Co., Georgia; Picures (as the sheet prints it), Taos Co., New Mexico; and four places in Minas Gerais, Brazil.
| Material | Source | Hardness | Density | 20.00 g piece in water | Crystal system, optics | What separates it from staurolite |
|---|---|---|---|---|---|---|
| Staurolite | Handbook sheet | 7 to 7.5 | 3.74 to 3.83 | 14.65 to 14.78 g | Monoclinic, pseudo-orthorhombic; biaxial (+) | — |
| Zincian staurolite | IGS | Not given separately | 3.79 | 14.72 g | Indices 1.721 to 1.731 (IGS) | Nothing by weight: inside the staurolite range |
| Talcose schist, cut as a fake cross | Roberts 1934 | About 2½ to 3 | Not given | Not worked | A rock, not a crystal | Hardness 4 or more below staurolite; filed, oiled, sometimes bored |
| Talc, the mineral itself | Handbook sheet | 1 | 2.58 to 2.83 | 12.25 to 12.93 g | Triclinic or monoclinic | 1.72 g lighter in water at the closest; hardness 1 |
| Almandine (an associate) | Handbook sheet | 7 to 7.5 | 4.318 | 15.37 g | Cubic; isotropic, n = 1.830 | 0.59 g heavier in water; no cross twins; deep red to brownish red |
What is staurolite? An iron aluminium silicate of the schists
Staurolite is the iron-dominant member of a small group of aluminium silicates, Fe2+2Al9Si4O23(OH) on the IMA list. The September 2026 IMA-CNMNC Master List gives it status G, grandfathered, with 1792 as the year and the country unknown. A script over the list finds three rows whose name ends in staurolite: staurolite; magnesiostaurolite, Mg(Mg,Li)3(Al,Mg)18Si8O44(OH)4, approved as 1992-035, Italy; and zincostaurolite, Zn2Al9Si4O23(OH), approved as 1992-036, Switzerland. The Handbook writes the formula (Fe2+,Mg)2Al9(Si,Al)4O20(O,OH)4.
What the sheet records. Monoclinic, pseudo-orthorhombic; prismatic crystals showing {110}, {010}, {001} and {101}, typically rough, to 12 cm. Cleavage distinct on {010}; fracture subconchoidal; brittle; hardness 7 to 7.5; measured density 3.74 to 3.83. Dark reddish brown to blackish brown, yellowish brown, rarely blue; streak white to greyish; subvitreous to resinous lustre. It is common in pelitic schists, gneisses and argillaceous rocks regionally metamorphosed to amphibolite grade, and also occurs as a detrital mineral, with almandine, sillimanite, kyanite, tourmaline, muscovite, quartz and chloritoid.
Worked: the ideal against an analysis. On standard atomic weights Fe2Al9Si4O23(OH) has a formula mass of 851.85: 28.21% SiO2, 53.86% Al2O3, 16.87% FeO, 1.06% H2O. The sheet’s analysis, an average of 23 microprobe analyses from Black Mountain, Grafton Co., New Hampshire, reads SiO2 27.82, Al2O3 54.91, FeO 12.39, with Li2O 0.36, MgO 0.62 and ZnO 0.19 sharing the iron site, and H2O 2.26 determined on a separate sample: 2.14 times the ideal water. The IGS adds a zinc-bearing staurolite of specific gravity 3.79 and a cobalt-bearing blue one called lusakite, from Lusaka, Zambia; a script finds no lusakite row on the IMA list.
Why does staurolite form crosses? The 60° and 90° twins
Staurolite crosses are twins, and the Handbook records the 60° twin as common and the 90° cruciform twin as less common. The sheet’s twinning line reads: commonly as 60° twins with {231} as twin plane, may be cyclic; less commonly as 90° cruciform twins with {031} as twin plane. The name, from the Greek for a cross, describes the common twins. The general idea of a twin law is on mineral twinning explained.
What a collector in the field found. Roberts, collecting in Henry and Patrick Counties, Virginia, in 1922 and 1923, reports that simple untwinned crystals were fewer than twinned ones, that the majority were twinned at approximately 60°, that quite a few made a cross of almost 90°, and that very few were triplets combining both. He calls the 60° form the Cross of St. Andrew and the 90° form the Roman Cross. The International Gem Society staurolite article says crystals with both 60° and 90° twinning are extremely rare, and that the crosses are popularly called fairy crosses or fairy stones.
A note on indices. Roberts names the twin planes in a different crystallographic setting from the Handbook’s, so his indices are not copied here; the Handbook’s {231} and {031} were read from the rendered sheet.
Are fairy crosses real? The Roberts check, a documented failure
Some are, and some were carved: Roberts (1934) documented fake staurolite crosses cut from talcose schist. He reports that in Virginia, well over a quarter of a century before he wrote, people began wearing staurolite crystals as watch charms and pendants, and that of the crosses offered, some were genuine and some artificial. Practically all were of the right-angle pattern, which in Henry and Patrick Counties is rare compared with the others; the numbers offered, against that scarcity, are what raised his suspicion.
How the fakes were made. The artificial crosses were cut from a grey talcose schist of hardness about 2½ to 3 with thin band saws, finished with flat files and soaked in linseed or other oil to give a dark brown. It is quite possible, he adds, that many of the plain Roman crosses were genuine, filed and smoothed from natural twins. Natural crystals develop small cavities as they weather, and holes were bored into the artificial ones to imitate them. The IGS cites a 1963 GIA note on a similar talcose imitation dipped in paraffin.
The check, from the numbers. Staurolite is hardness 7 to 7.5; Roberts’s schist about 2½ to 3; talc itself, on its sheet, 1. In water a 20.00 g staurolite reads 14.65 to 14.78 g and pure talc 12.25 to 12.93 g. Roberts gives no density for the schist, which is a rock, so its reading is not worked here. A cross that is very smooth, perfectly right-angled and soft fits the documented fake; one that is rough, heavy and hard fits staurolite. The methods are on how to measure specific gravity at home and how to test mineral hardness properly; the general fakes page is how to tell if a crystal is fake.
Staurolite or garnet? Two brown crystals of the same schists
Almandine is staurolite’s first-named associate, and it is cubic, isotropic and denser. The Handbook almandine sheet gives hardness 7 to 7.5, the same as staurolite, but density 4.318: 15.37 g in water for a 20.00 g piece, 0.59 g heavier than staurolite at its heaviest. Almandine is deep red, brownish red or red-violet, and typically forms well-formed dodecahedra or trapezohedra, not prisms or crosses; see almandine.
The blue one in the same rocks. Kyanite also appears on staurolite’s association line; its species page is kyanite. Staurolite’s sheet records it as rarely blue, so colour alone does not settle the pair; the cross twins and the prism shape do.
Where staurolite is found
The Distribution field calls staurolite widespread, with fine crystals from, for example: Pizzo Forno and Alpe Piona, Ticino, Switzerland; Mt. Greiner, Zillertal, Tirol, Austria; the Keivy massif, Kola Peninsula, Russia; large twins in the old provinces of Finistère and Morbihan, Brittany, France; Fanzeres, Portugal; Franconia, Grafton Co., New Hampshire; Chesterfield, Hampshire Co., Massachusetts; near Blue Ridge, Fannin Co., Georgia; near Picures (as printed), Taos Co., New Mexico; and, in Minas Gerais, Brazil, Rubellita, Ardenella, Agua Quente at Rio Pardo de Minas, and Fazenda Graniais (see Minas Gerais).
Virginia, on Roberts’s authority. The sheet does not name Virginia. Roberts describes a staurolite belt in the Virginia Piedmont from near Lynchburg south-westward into North Carolina, about 60 miles long, with the better collecting in Henry and Patrick Counties in the Wissahickon schist, and records collections he made at Ducktown, Tennessee in 1916 and in Fannin County, Georgia in 1917. Nothing is said here about access to any site today.
The other silicates of the schists are indexed at the species index. We hold no catalogue and nothing here is offered for sale; the wanted list is the only commercial route.