Species · Micas

Lepidolite

Lepidolite is the pink lithium mica, hardness 2.5 to 4 and density 2.8 to 2.9. It is not a valid species: the name covers polylithionite and trilithionite.

Series formula K(Li,Al)3(Si,Al)4O10(F,OH)2Hardness 2.5 to 4Density 2.8 to 2.9 measured, 2.693 calculatedCleavage Perfect on {001}; flexible and elasticHabit Pseudohexagonal plates to 20 cm; scaly aggregatesColour and lustre Pink, purple, rose-red, violet-grey, white; pearly to vitreous lustreAssociates Spodumene, elbaite, amblygonite, columbite, cassiterite, topaz, berylIMA status Not a species. The Handbook's own sheet calls it a group name

Lepidolite is the pink to purple lithium mica of granite pegmatites, hardness 2.5 to 4, measured density 2.8 to 2.9, with perfect {001} cleavage and flexible elastic plates to 20 cm. It is not a valid mineral species. The September 2026 IMA list has no entry for it, and the Handbook of Mineralogy's own sheet states plainly that lepidolite is a group name.

In short

  • Two independent references agree that lepidolite is not a species. The September 2026 IMA-CNMNC Master List contains no entry for the name at all, and the Handbook of Mineralogy sheet, under Polymorphism and Series, ends with the words a group name. The species behind it are polylithionite, KLi2AlSi4O10F2, and trilithionite.
  • Only three sheets in this site's 158-sheet reference set span a full 1.5 Mohs points, and all three are retired names. Lepidolite 2.5 to 4, tetrahedrite 3 to 4.5, tennantite 3 to 4.5. Of the 102 sheets that print a hardness range, no other exceeds one point. A wide hardness bracket is a symptom of a name covering more than one mineral.
  • The measured and calculated densities disagree by about 4%. The Handbook gives 2.8 to 2.9 measured against 2.693 calculated. Polylithionite, one of the species inside the name, measures 2.58 to 2.82. The spread is what a solid solution looks like from the outside.
  • The pink is lithium and manganese, and it is the reason the mineral is mined. Lepidolite is a lithium ore, found by metasomatic replacement of muscovite or biotite in granite pegmatites, with spodumene, elbaite, amblygonite, columbite, cassiterite, topaz and beryl.
  • It is not radioactive and it is not dangerous to hold. Lithium micas contain no uranium or thorium as an essential component, and the specimen on your shelf is an inert silicate. Dust from grinding or sawing is a different question, and one for a workshop rather than a collection.
What the label lepidolite actually covers
NameFormulaIMA statusHardnessDensity
LepidoliteK(Li,Al)3(Si,Al)4O10(F,OH)2Not a species; a series or group name2.5 to 42.8 to 2.9
PolylithioniteKLi2AlSi4O10F2Approved, 19982 to 32.58 to 2.82
TrilithioniteKLi1.5Al1.5(Si3Al)O10F2Redefined, 1998, type locality Sweden
MuscoviteKAl2(AlSi3O10)(OH)2Valid species2.52.77 to 2.88
BiotiteK(Mg,Fe)3(AlSi3O10)(OH)2Not a species; a series name2.5 to 32.7 to 3.3

The reference literature says it twice, in two different ways

This is worth doing carefully, because it is the kind of claim that gets repeated without being checked.

First check: the IMA list. The September 2026 IMA-CNMNC Master List is the official register of valid mineral species. Searching it for lepidolite returns nothing. Polylithionite is there, approved in 1998, with the formula KLi2AlSi4O10F2. Trilithionite is there, marked Rd for redefined, also 1998, with a Swedish type locality.

Second check: the Handbook's own sheet. Its Polymorphism and Series line reads 1M, 2M2; 3A polytypes common; 2M1; 3M2 rare; a group name. The primary descriptive reference for the mineral says, in its own fourth clause, that the thing it is describing is a group.

Two sources, no conflict, and the page is filed under lepidolite anyway. That is deliberate. The name is on every dealer's label, every old collection card and every search a collector types, and renaming the page would make it unfindable by the people holding the specimens. This site does the same for apophyllite, heulandite, stilbite and tetrahedrite. Five of the 73 species pages here carry a name that is not a valid IMA species, and each one says so.

A wide hardness bracket is evidence of a wide name

Here is a small result from this site's reference set that turned out to be worth more than it looked.

Of the 158 Handbook sheets in the set, 102 print hardness as a range rather than a single figure. Exactly three of those ranges span a full 1.5 Mohs points: lepidolite at 2.5 to 4, tetrahedrite at 3 to 4.5 and tennantite at 3 to 4.5. Everything else that varies, varies by a point or less — zoisite 6 to 7, uraninite 5 to 6, turquoise 5 to 6, polylithionite 2 to 3.

All three of the widest are names the IMA has since retired as species. That is not a coincidence and it is not a mystery. A hardness bracket is measured on real specimens, so a name covering a compositional range collects the hardness of everything inside it. The bracket is the variation showing through the label.

The practical version for a collector: when a reference gives you a hardness spanning more than a point, treat the name with suspicion before you treat the measurement with suspicion. It is a cheap screening test for names that are about to be split, and it works on old books as well as new ones. The method and its limits are on how to test mineral hardness properly, and the same story in a sulphosalt is on tetrahedrite.

Where it comes from, and what it grows with

The Handbook places lepidolite in granite pegmatites, derived by metasomatic replacement of muscovite or biotite, and also in some high-temperature quartz veins, greisens and granites. Its associates are the pegmatite lithium suite: spodumene, elbaite, amblygonite, columbite, cassiterite, topaz, beryl and other micas.

That association list is the reason lepidolite matters to a collector who is not collecting micas. A lilac scaly mass is a signpost. Rubellite and multicoloured elbaite, blue and pink beryl, and the phosphates all come out of the same pockets, and a pegmatite that has produced good lepidolite has produced them somewhere.

The named localities are the classic pegmatite fields: Rožná in the Czech Republic; Alabashka in the Urals; the Varuträsk pegmatite in Sweden; Mount Mica near Paris, Oxford County, Maine; the Pala and Mesa Grande districts of California; the Brown Derby pegmatite in Colorado; the Ingersoll mine in South Dakota; the Tanco mine at Bernic Lake, Manitoba; Maharitra on Mount Bity, Madagascar; and Bikita.

Britain is not on that list and that is the honest position: British lithium pegmatites are minor and lepidolite here is a mineralogical curiosity rather than a collecting destination. The rest of this section is indexed at the species index. We hold no catalogue and nothing is offered for sale; the wanted list is the only commercial route.

Questions

Should I stop calling it lepidolite?
No, but know what you are saying. The word is an accurate and useful description of a pink lithium mica and it will remain in trade use indefinitely. It is not a species determination. The correct label form is lepidolite (polylithionite-trilithionite series) with the locality, and our note on how to label and catalogue a mineral collection covers adding to an old card without overwriting it.
Which species is my lepidolite?
Almost certainly not answerable without analysis. Polylithionite and trilithionite differ in the lithium-to-aluminium ratio across a continuous series, and no hand test resolves that. The Handbook records polylithionite as possibly fluorescing lemon-yellow under shortwave ultraviolet, which is suggestive at best and not a determination.
Is lepidolite radioactive?
No. It is a lithium, potassium and aluminium fluorosilicate with no uranium or thorium in its formula. The trace potassium-40 in any potassium mineral is present but trivial. Our note on whether your mineral is radioactive sets out which species actually warrant a detector.
Why do so many mica names turn out not to be species?
Because the micas are dominated by continuous substitutions between iron and magnesium, lithium and aluminium, fluorine and hydroxyl, and the historic names were coined for appearance rather than composition. Biotite went the same way as lepidolite and is now a series name. The IMA-CNMNC nomenclature rules require a species to be defined by a dominant occupancy of a site, which the old appearance-based names simply do not do.
Will lepidolite survive being cleaned?
Gently, yes. It is hardness 2.5 to 4 with perfect basal cleavage and flexible elastic plates, so a brush and water are fine and anything mechanical is not. An ultrasonic bath will delaminate scaly aggregates. Never pick at the edges of a plate: mica splits at the touch.