Defining a mineral

Mineral definitions

It is remarkably difficult to define a mineral. Like other well-intended definitions, it is often derailed by exceptions, becoming ever more abundant with increased complexity of the definition. The standard definition usually goes along lines like 'a mineral is a chemically pure, solid substance with a uniform and coherent internal organisation, formed by natural processes.'

Natural

The 'natural' is ever so often replaced by the term 'mineralogical' giving a wonderful circular and inherently nonsensical definition or 'geological' which is less obvious but no more informative. Regrettably, very few if any minerals fit the definition!

Pure

How should we understand 'chemically pure'? Practically every mineral contains traces of alien elements, and some are even known to form series of mixed minerals - this includes well-known groups like feldspars and amphiboles. And obviously, if we replace 'chemically pure' with 'varying within fixed limits' we might as well forget about it.

Solid

Solid? Well, we like to think of Mercury as a mineral ... OK, that's the exception then. But an otherwise abundant mineral like Ice [yes, it is a mineral - pure, solid, internal organisation] does become mobile and slippery under pressure. That's how glaciers move. Many other minerals have similar properties, we just don't have the chance to observe it as readily. Some have a more eccentric lack of solidity, where you can shift the crystal lattice, even if the mineral is neither soft, liquid or even warm. A single crystal of Antimonite can be bent, shifting the crystal lattice, but it is a one-time only process; it can not be bent back [please don't try this on a good specimen!]. You may also press a corner of a Calcite cleavage rhombohedron against a firm base, and flip the corner - technically you induce a twin. This only works on very small pieces, but does demonstrate that even a 'solid' mineral is not solid in the sense it wont move.

Internal organisation

Specimen of Almandine from New South Wales,
Australia

'Uniform and coherent internal organisation' basically means it is crystalline - all the atoms are arranged in a repetitive pattern. That works fine ... up to a point. We already mentioned Mercury, that is liquid and thus has no internal repetitive pattern. The same applies to Opal and all other amorphous minerals (there are a few). With good analytical methods, you may detect coherent internal organisations (crystal lattices), that are not uniform. Every now and then, crystals are zoned and two chemically different substances grow on top of each other, sharing a crystal lattice, that changes slightly in at least one dimension. This is very common in f.ex. amphiboles, that often differ chemically from the core through several layers towards the surface of the crystal.

Lost?

So we didn't get too far on that definition part, did we? And you know what? It doesn't bother me at all! It doesn't bother me, because it is a pseudo-problem. I can define 'Calcite' [you can look that one up yourself] - we know it's chemistry, crystallography, lattice constants, etc., and even know it can incorporate foreign elements like magnesium and iron without appreciably affecting the properties. That is, we have a firm definition of Calcite, and we have a good understanding of how real objects in nature may deviate from that; the definition of 'Calcite' allows us to distinguish between objects that are 'Calcite' and those that are 'not Calcite' and hence it is a good definition. We have similar definitions of 'Gold', 'Azurite', 'Quartz', 'Kristiansenite', etc. That is, whenever we put a name on things, we can go back and look up what that name means.
The problem arises, when we wish to define a group of all these good definitions, excluding equally good definitions of other objects (sugar, steel, grandpa, ...), and that definition has to be non-circular and simple ... and be ready to include definitions of future minerals? Finding the common denominator of over 3,500 definitions, simultaneously predicting future scientific understanding is not a reasonable expectation. Regrettably, I can only offer two possible definitions of the concept of 'mineral': I) I know one, when I see one (rather presumptuous); II) a substance recognised as such by the International Mineralogical Association (no hard feelings, but still 'old boys network'). But do note, that even if I can not define 'mineral', I can explain 'mineral' and I can at any time define any mineral.

Bibliography

Blackburn, William H. & William H. Dennen. 1997. Encyclopedia of mineral names. Canadian Mineralogist, special publication 1.

Gaines, Richard W., H. Catherine W. Skinner, Eugene E. Foord, Brian Mason, Abraham Rosenzweig & Vandall T. King. 1997. DanaÕs new mineralogy: the system of mineralogy of James Dwight Dana and Edward Salisbury Dana, 8th ed.

Ramdohr, Paul & Hugo Strunz. 1980. KlockmannÕs Lehrbuch der Mineralogie, 16th ed.

This page is written and maintaned by Claus Hedegaard