Chalcedony, agate, and jasper

A banded, translucent, waxy-looking stone that a steel point will not mark is almost certainly a member of the chalcedony family. So is a mottled opaque red-and-green one, and so is most of the uniform black material sold as onyx. This is a large family with a great many trade names, and all of its members are the same substance as a quartz crystal — arranged differently.

That difference in arrangement is the whole of what makes them behave the way they do.

One composition, two fabrics

Chalcedony is silica: the same composition as quartz. What separates it is grain size and habit. Where a quartz crystal is a single continuous lattice, chalcedony is a dense mass of submicroscopic fibres, too fine to resolve without a microscope. It is the microcrystalline end of the crystalline-to-amorphous range, and the consequences run through every property that matters in a tool.

Hardness is around 6.5 to 7 — at or just below coarse quartz, the slight reduction attributable to the fabric rather than the composition. Steel will not mark it. It will mark glass.

Specific gravity runs a shade under quartz’s 2.65, in the region of 2.58 to 2.64, because the spaces between fibres hold a small amount of water and void.

Lustre is waxy to dully vitreous rather than glassy — the family’s most recognisable qualitative feature, and the reason the material reads as “stone” where a quartz crystal reads as “gem”.

Toughness is the important one. Chalcedony has no cleavage, and a crack propagating through it has to cross an enormous number of fibre boundaries, each of which absorbs a little energy and can deflect the crack. This makes it appreciably tougher than a single quartz crystal of identical hardness — the same aggregate mechanism that makes nephrite tough, in a different mineral. A rose quartz tool and an agate tool are equally hard and not equally survivable.

The family, by variety

Every name below is the same mineral with a different colour cause, texture or banding pattern.

Agate — chalcedony with visible banding, usually concentric or parallel, formed by successive deposition on the walls of a cavity. The banding is a growth structure, not a fracture.

Onyx — in gemmology, parallel-banded chalcedony, classically black and white. In the trade, “black onyx” is a solid black material, and solid black is rare in nature; the standard commercial route to it is dyeing a pale grey agate.

Carnelian — translucent orange to red, coloured by iron oxide, and very commonly heat-treated to develop or deepen the colour. Sard is the browner end of the same material.

Chrysoprase — apple-green chalcedony, its colour generally attributed to nickel. It is the one member of the family with a genuine claim on jade’s appearance, and it carries the trade name “Australian jade” as a result.

Bloodstone — dark green chalcedony with red spots of iron oxide.

Jasper — the opaque end: chalcedony so loaded with fine-grained impurities that no light passes. Red, brown, yellow, green, banded and brecciated. The boundary between jasper and chalcedony is textural and gradational, not a compositional line, which is why the two names are applied inconsistently.

Why this family is dyed more than any other

Chalcedony’s fibrous fabric leaves micropores between the fibres, connected well enough that liquids travel through them. The material is therefore unusually receptive to colour, and dyeing agate is one of the oldest and most established practices in the stone trade — long predating the current market, widely disclosed at the wholesale level, and entirely legal.

The mechanism produces its own tell. Because the porosity varies from band to band, the dye does too: a dyed agate typically shows colour concentrated in the more porous layers and weak in the tight ones, giving a banded intensity that follows the growth structure rather than an even body colour. In blue, green and purple agate — colours that are uncommon to absent in natural chalcedony at high saturation — this is close to diagnostic.

Sugar-and-acid treatment is the other classical method for black: a sugar solution is drawn into the pores and then carbonised with acid, leaving carbon in the stone rather than a dye. The result is stable and permanent, and it is still a treatment. These are cousins of the processes described for imitation and enhanced stone generally.

What you can establish

WHAT YOU CAN TELL — a hard, waxy, banded stone

  · Steel will not mark it; it marks glass
      → hardness near 7; consistent
  · Waxy rather than glassy lustre → chalcedony
      family rather than a quartz crystal
  · Concentric or parallel banding → agate;
      a growth structure, not a crack
  · Colour strong in some bands, weak in others
      → dyed. Strong indicator
  · Saturated blue, green or purple → very likely dyed
  · Heated carnelian vs natural → NOT DETERMINABLE
      at home, and often difficult in a laboratory

Where it gets confused with jade

Chalcedony is harder than nephrite, tough for its hardness, waxy in lustre and available in green. That combination puts it closer to jade’s overall impression than most substitutes manage, and green chalcedony has served as a jade substitute for a very long time.

Two observations separate them. Density: nephrite runs 2.90 to 3.03 and jadeite 3.30 to 3.38, both clearly above chalcedony’s 2.6 or so, so a chalcedony piece feels light next to either. And structure: jade’s fabric is felted or granular and shows no concentric banding, while agate’s banding is its signature. A banded green stone is not jade; an unbanded one needs more than looking.

The trade names follow the usual pattern — “Australian jade” for chrysoprase, “green onyx” for dyed green agate — a qualifier letting a gem name do the selling. Against these, incidentally, sits the other onyx: the soft honey-banded carbonate that shares the word and is four Mohs steps softer.

The limit

You can place a stone in this family with reasonable confidence from hardness, lustre and banding, and you can often catch the dye. That is most of what a non-specialist needs.

What you cannot do is name the variety with authority, because the variety names are trade categories resting on appearance rather than measurable species boundaries — nobody’s instrument reports “jasper”. You cannot tell heated carnelian from unheated, which is difficult in a laboratory too and routinely undisclosed because the treatment is regarded as normal. And on an evenly coloured piece in a plausible natural hue, you cannot establish that no dye is present; the absence of banded intensity is the absence of one tell, not proof of anything.