Species · Carbonates

Magnesite

Magnesite is MgCO3, hardness 3.5 to 4.5 and density 3.00 — denser than the turquoise it is dyed to imitate, which is why the heft test gives the wrong answer.

Formula MgCO3Hardness 3.5 to 4.5Density 3.00Cleavage Perfect on {10-11}Habit Crystals uncommon, to 30 cm; usually massiveLustre Vitreous; streak whiteIMA status Approved, 1962 s.p.; Italy

Magnesite is magnesium carbonate, MgCO3, hardness 3.5 to 4.5 with a measured density of 3.00 and perfect rhombohedral cleavage. Dyed magnesite is the commonest turquoise imitation a dealer meets, and at 3.00 against a turquoise specimen's 2.86 it is the denser of the two — so the standard advice that real turquoise feels heavier gives exactly the wrong answer.

In short

  • Magnesite is 4.9% denser than turquoise, and that is below what a hand can feel. Twenty grams of turquoise occupies 6.99 cm3; the same volume of magnesite weighs 20.98 g. The heft test is not merely useless here — it points the wrong way.
  • Hardness is the test that works. Magnesite is 3.5 to 4.5 against turquoise's 5 to 6. A steel point near 5.5 marks magnesite and struggles on turquoise, and it gives the same right answer for howlite, variscite and chrysocolla.
  • Crystals are uncommon; magnesite is normally earthy, chalky, porcelaneous, fibrous or compact massive. That porcelaneous massive habit is exactly what makes it a convincing turquoise substitute once dyed — it takes dye evenly and carves cleanly.
  • It may fluoresce pale green to pale blue and is triboluminescent, meaning it can emit light when crushed or scratched. Neither response is reliable enough to identify it, and the Handbook hedges both with may.
  • In the carbonate density ladder magnesite sits third of six: calcite 2.7102, dolomite 2.86, magnesite 3.00, rhodochrosite 3.70, siderite 3.96, cerussite 6.55. Density separates that group far more cleanly than colour does.
The carbonate density ladder, and how each responds to cold dilute acid
SpeciesFormulaHardnessDensityCold dilute HCl
CalciteCaCO332.7102Effervesces briskly on a solid surface
DolomiteCaMg(CO3)23.5 to 42.86Slow or nil on a solid surface; brisk on powder
MagnesiteMgCO33.5 to 4.53.00Little or none cold; needs powder or warmth
RhodochrositeMnCO33.5 to 43.70Slow cold, brisk on powder
SideriteFeCO33.75 to 4.253.96Slow cold, brisk on powder; solution turns yellow
CerussitePbCO33 to 3.56.55Effervesces; density alone is decisive

The arithmetic that breaks the heft test

The most repeated piece of advice about buying turquoise is that the real stone feels heavy for its size. Run the numbers on the commonest substitute and the advice inverts.

The Handbook of Mineralogy gives turquoise a measured density of 2.86 and magnesite 3.00. Take a 20 g turquoise cabochon: its volume is 20 / 2.86 = 6.99 cm3. A magnesite piece of that same volume weighs 6.99 × 3.00 = 20.98 g.

So the substitute is 4.9% heavier than the stone it is imitating. Two effects follow, and both are bad. First, the difference is in the wrong direction. Second, 4.9% is far below the threshold at which a hand can compare two objects of different shapes — which means a buyer applying the rule will usually feel nothing at all, and when they do feel something they will reject the genuine piece.

The full six-substitute comparison, including the one case where neither hardness nor density works, is on how to tell real turquoise from fake. The measurement method itself is on how to measure specific gravity at home.

Why magnesite barely reacts to cold acid, and dolomite barely does either

Every collector learns that carbonates fizz in dilute hydrochloric acid. The useful version of that rule is that calcite fizzes, and the rest of the group is a spectrum.

Calcite effervesces briskly on a solid surface. Dolomite is slow to nil on a solid surface and brisk once powdered. Magnesite is the least reactive of the common carbonates cold, typically needing to be powdered or warmed before it responds at all. Siderite reacts slowly and leaves a yellow solution from the iron.

That gradient is itself diagnostic, and it is more useful than the yes-or-no version. A white rhombohedral carbonate that will not fizz cold but fizzes as powder is magnesite or dolomite, and the density separates those two: 3.00 against 2.86.

An acid test is destructive and it is not reversible. It takes a scratch of powder, and on a display face that scratch is permanent. The general position on destructive testing is set out on how to identify a mineral specimen.

Magnesite as a specimen rather than a substitute

Magnesite crystals reach 30 cm, with large {10-11} or {01-12} rhombohedra modified by {10-10} and {11-20}, or tabular on {0001}. That is a serious crystal, and the localities that produce them are few.

The Handbook names Brumado in Bahia, Brazil for fine large crystals; Sunk near Trieben in Styria and Radenthein in Carinthia, Austria, also for large crystals; Baldissero Canavese in Piedmont; Magnesia in Greece, which supplies the name of the element and therefore of the species; Snarum in Norway; the Mt Brussilof mine near Radium in British Columbia; Gabbs in Nye County, Nevada; and immense deposits in Liaoning Province, China.

The occurrence is what you would expect from the company it keeps: a primary mineral in igneous or sedimentary rocks, formed by metamorphism or alteration of serpentine and peridotite, uncommon in marine evaporites and hydrothermal veins, rare in carbonatites. Its listed associates are talc, antigorite, chlorite, dolomite and calcite — which is why the talc page and this one describe the same rocks from opposite ends.

The IMA-CNMNC Master List records magnesite as approved by a 1962 special procedure with Italy as the country of first description. We hold no stock and quote no prices; the route for a Brumado rhombohedron or a Styrian crystal is the wanted list, and the rest of this section is indexed under mineral species.

What the dyed material looks like when you know what to look for

Dyed magnesite sold as turquoise is usually beads, cabochons or carved pieces rather than anything on matrix, and that alone is a strong signal — a mineral specimen with no matrix and a polished surface has had its evidence removed.

Three checks, in the order they are worth doing. Hardness first: a steel point marks magnesite at 3.5 to 4.5 and struggles on turquoise at 5 to 6. Then the veining: magnesite's natural fracture network takes dye preferentially, so the dark web on a dyed piece is often too even and too continuous, and a drilled bead may show pale untinted material in the bore. Then the edges: a chipped edge on dyed material commonly shows white beneath.

None of this is a claim about any particular piece, and none of it is a price judgement — there is no catalogue on this site and no specimen here is for sale. The broader treatment-disclosure argument, including what to ask a seller, is on how to tell if a crystal is dyed.

Questions

Is dyed magnesite the same as howlite?
No, though both are dyed blue and sold as turquoise. Howlite is Ca2B5SiO9(OH)5, hardness 3.5, density 2.53 to 2.59; magnesite is MgCO3, hardness 3.5 to 4.5, density 3.00, per the Handbook of Mineralogy. Both are softer than turquoise, so the hardness test catches both. Density distinguishes the two substitutes from each other.
Does magnesite fizz in acid?
Barely, cold. It is the least reactive of the common carbonates in cold dilute hydrochloric acid and generally needs to be powdered or warmed to respond, where calcite effervesces briskly on a solid surface. That difference in reaction rate is a more useful test than a simple yes or no.
How do I tell magnesite from dolomite?
By density: 3.00 against 2.86, a difference of 4.9% and measurable on a 0.01 g scale with a few grams of clean material. Hardness overlaps, both are rhombohedral with perfect {10-11} cleavage, both are commonly white, and both are sluggish in cold acid. See how to measure specific gravity at home.
Does magnesite fluoresce?
The Handbook records that it may exhibit pale green to pale blue fluorescence and phosphorescence under ultraviolet, and that it is triboluminescent. The hedge is the point: a magnesite that does not fluoresce is still magnesite, so the response can support an identification but never settle one. See fluorescent minerals explained.
Is magnesite a British mineral?
Not notably. The IMA-CNMNC Master List gives Italy as its country of first description, and the Handbook's locality list runs through Austria, Brazil, Greece, Norway, Canada, the United States and China without a British entry among the well-studied occurrences. Eleven of the species this site covers do have United Kingdom type localities; they are listed on the pharmacosiderite page.