A page for each of the minerals people actually pick up, with the three field tests that settle most of them and an honest account of what a photograph cannot reach.
16 minerals34 modelsFree · no account
Mineral identification has a reputation for being difficult, and it is not — it is just that almost everybody starts with the wrong property. Colour is the first thing anyone notices and the least reliable diagnostic in mineralogy. Quartz occurs in at least seven colours under seven different names. Fluorite occurs in all of them. Meanwhile two stones that look nothing alike can be the same species, and two that look identical can differ in value by a thousandfold.
What actually settles a mineral is hardness, streak and density, and all three are free. A copper coin, a steel knife, a piece of glass, the unglazed back of a tile and your own hand will identify most of what you find, and no photograph can measure any of them. That is not a criticism of the instrument the house makes — it is the reason the instrument's page says the same thing.
So each page here is built the same way: the measured properties in a table, then the checks ordered by how much each one settles, then what the mineral gets confused with and what is sold in its place. Where a material has a real fraud problem — turquoise, amber, malachite, jade, lapis, hematite — that section is long, because it needs to be. Where nothing pretends to be it, there is no such section, because there would be nothing honest to put in it.
The order is the whole method. Hardness, streak and heft between them identify most of what anyone finds, and every one of them is free.
A fingernail is about 2.5, a copper coin 3.5, a steel knife or a glass plate 5.5, a steel file 6.5. Work up: find the softest thing that scratches your specimen and the hardest thing it scratches. That brackets the hardness, and hardness alone eliminates most candidates. Wipe the powder away before deciding — soft minerals leave dust that looks exactly like a scratch.
Drag the specimen hard across the unglazed back of a tile and read the colour of the powder, not the colour of the stone. This is how hematite gives itself away — red-brown streak whatever the surface looks like — and how pyrite is separated from gold in one stroke. Minerals harder than about 6.5 will not streak at all, which is itself information.
Pick it up before you look at it. Pyrite, hematite and garnet feel wrong for their size; opal and amber feel too light. Then find a broken edge: cleavage means flat mirror faces meeting at consistent angles, and no cleavage means curved conchoidal shells. Quartz never cleaves. Fluorite, calcite and feldspar always do.
Carbonates effervesce in weak acid and silicates do not, which separates calcite, malachite and marble from quartz, agate and jasper in ten seconds. Run that, and only then let yourself think about colour — as a way of choosing between varieties of a species you have already identified, never as a way of choosing the species.
One page each, in the mineral's own terms — its code system, its weighted checks, its models and what they do on the secondary market.
Lovenhall StoneThe tests tell you what it is. The instrument reads a photograph and offers a name to start from — then hands you back to the tile and the coin, which are what settle it.