Decide which question you are asking
“Real” carries three meanings in the gem trade and they have wildly different consequences. An imitation is a different material pretending to be the stone: glass, plastic, or a cheaper mineral dyed to suit. A synthetic has the same chemistry and crystal structure as the natural stone and was grown in a factory rather than the ground. A treated stone is natural and has been improved — heated, filled, dyed, irradiated.
Most disappointment comes from the third category rather than the first. Nobody pays a fortune for red glass, but people pay natural prices for glass-filled rubies and dyed quartz every week, because those stones are, in a narrow sense, real.
The expensive mistake is almost never glass. It is a genuine stone whose treatment nobody mentioned.
What a loupe shows in ten seconds
Glass is the easiest thing in gemmology to catch, and a ten-power loupe does it. Look for round or elongated gas bubbles, for swirl or flow lines like the inside of a marble, and for facet edges that are soft and rounded rather than crisp where two planes meet. Moulded imitations often show a faint seam and facets that do not quite line up.
Then look at what is inside. Natural stones are geological objects and usually contain something: crystals, needles, healed fractures, colour zoning in straight bands following the crystal's growth. A completely empty interior in a large coloured stone is a reason to keep looking, not a reason to relax.
- Bubbles and swirls — glass, almost always.
- Rounded facet junctions and worn edges on a “new” stone — glass or a soft imitation.
- Straight colour banding and mineral inclusions — consistent with natural growth.
- Curved striae, as though the colour had been stirred — flame-fusion synthetic corundum.
The physical checks worth doing
Weight in the hand is informative once you have handled a few stones: most gem minerals are denser than glass of the same size, so a large stone that feels light is suspect. Temperature is the other free test — glass and plastic warm quickly against the skin, while most gem materials feel cool for noticeably longer. Neither is proof, and both are quick.
Look at wear, too. Facet edges on an old stone should be crisp if the material is hard; a soft imitation set in a decades-old ring will have abraded edges and a dull polish where a corundum or a spinel would not. And where a stone is doubly refractive and strongly so — zircon, tourmaline, calcite — a loupe can show a doubling of the opposite facet edges that a glass imitation cannot produce.
Tests not to do
Do not scratch it. Hardness testing has a place in a laboratory with reference points and a discreet corner of a specimen; on a set stone it damages the thing you are trying to value, and a scratch tells you far less than the loupe already did. The same goes for heating a stone, tapping it, or dropping it into anything.
One exception, for one material. Amber floats in heavily salted water where most plastics sink, which is a genuine, non-destructive test and the standard first check on a bead necklace. Even there, do it with a loose bead rather than a strung strand, and rinse and dry it afterwards.
Synthetics are real stones, at another price
Flame-fusion ruby and sapphire have been produced commercially since the beginning of the twentieth century, which means a great many antique rings contain them entirely innocently. Chemically they are corundum; a laboratory identifies them by growth structure — curved striae rather than the straight, angular growth of a natural crystal — and by their inclusions, or the absence of the ones nature leaves behind.
The same applies to synthetic emerald, spinel and quartz. None of it is fraud when it is disclosed, and none of it is worth natural money. If a colour is vivid, the clarity is high and the price is comfortable, synthetic is the most likely explanation, and it is a perfectly good thing to own knowingly.
Treatments, in order of how much they cost you
Heating ruby and sapphire is ancient, permanent and so nearly universal that untreated stones command a premium precisely because they are unusual. That is the benign end. Glass or lead-glass filling of fractured ruby is the other: it turns cheap, heavily fractured material into something that looks like a fine stone, it is damaged by ordinary jewellery repair and ultrasonic cleaning, and it is worth a small fraction of the untreated equivalent.
In between sit dyed material — agate, quartz, jade, howlite sold as turquoise — and irradiated or heated colour changes. Almost all commercial citrine is heated amethyst. “Goldstone” is glass with copper crystals in it. A great deal of “turquoise” is reconstituted powder in resin. None of these are frauds by name, and all of them are frauds at natural prices.
Heated sapphire is a normal sapphire. A glass-filled ruby is a repair with a colour, and the two prices should not resemble each other.
What a photograph can and cannot do
A camera reads colour, transparency, lustre and shape, and from those an identifier can often propose what a stone is likely to be and what its kind is worth — which is a useful place to start when a box of inherited jewellery arrives with no paperwork. Our Rock, Crystal & Mineral Identifier is built for that first pass, and we would rather you knew its limits than discovered them.
What no photograph can do is measure. Refractive index, specific gravity, hardness, spectrum and fluorescence are the properties that actually separate one red stone from another, and every one of them requires the stone in a hand and an instrument against it. A picture narrows the field; it does not close it.
When to pay for a laboratory
Three situations justify the fee: you are insuring the piece, you are selling it as something specific, or the answer changes the number by more than the report costs. For coloured stones the report you want states species, whether it is natural or synthetic, and what treatments were detected — and for ruby, sapphire and emerald, origin, because geography carries a premium in those three.
Ask for the treatment line explicitly. A report that says “natural ruby” and nothing else has told you the least interesting of the three things you needed to know.