Obsidian, and natural glass

Obsidian is glass that a volcano made. It has no crystal structure, it fractures conchoidally into sharp curved surfaces, and it can contain gas bubbles — all of which are the standard indicators of a manufactured glass imitation. On a black stone tool, the usual glass tells therefore establish very little, and this is the one material where the ordinary examination routine breaks down.

How it forms, and why it is rare in old rocks

Obsidian is a volcanic glass, formed when silica-rich lava cools too fast for its atoms to organise into crystals. The composition is broadly that of rhyolite — high in silica — and the material is amorphous: no lattice, no cleavage planes, no grain boundaries.

Hardness is around 5 to 5.5, below both jades and well below quartz. Specific gravity is approximately 2.3 to 2.6, which is low for a stone and close to manufactured glass, for the good reason that they are chemically similar substances.

Glass is thermodynamically unstable over long periods. Obsidian slowly devitrifies — crystallises from within — and also takes up water, so old obsidian turns into something else. This is why obsidian is geologically young wherever it occurs, and why there is no ancient obsidian: the material has a lifespan.

Its varieties come from what is suspended in it rather than from its own structure. Snowflake obsidian carries white radiating clusters of crystals that grew in the glass. Mahogany obsidian is streaked reddish-brown by iron oxide. Sheen and rainbow obsidian show a metallic or iridescent play from oriented microscopic inclusions — structural colour, not pigment. All of these are obsidian with something in it.

The brittleness is the point

Amorphous material has no internal boundaries to stop a crack, so once a fracture starts it runs. That is why obsidian was one of the most valuable tool stones in the pre-metal world: controlled fracture produces an edge finer than any metal blade, because the break follows the material rather than a grain.

The same property makes a thin flat object of obsidian fragile. It chips readily at edges, and the chips are curved, glassy and very sharp — sharper than a broken piece of any crystalline stone. A damaged obsidian tool has genuinely keen edges on the damage, which is a straightforward mechanical consequence of conchoidal fracture in a homogeneous brittle solid, and worth knowing before handling a broken one.

Thermal shock affects it as it affects manufactured glass: a brittle amorphous solid with low thermal conductivity is close to the worst case for a rapid temperature change, and why that happens is the same mechanism in both materials.

What breaks down here

The standard glass indicators — bubbles, conchoidal fracture, vitreous lustre, no visible crystals — are indicators of glass, and obsidian is glass. Every one of them is expected in genuine obsidian and none distinguishes it from a moulded black glass imitation. This is a real gap in the home identification repertoire and it should be stated rather than papered over.

A few things do still discriminate:

Mould seams and flow lines. A moulded object was formed in a mould and the parting line is sometimes detectable at an edge, usually polished but occasionally not. Obsidian is cut and ground from a block, so it has no seam. Manufactured glass may show flow structure oriented to the shape of the object; obsidian’s internal banding relates to lava flow and bears no relation to the tool’s outline.

Inclusions with mineral character. Snowflake clusters, iron oxide streaks and oriented reflective platelets are natural growth or alteration features. A manufactured imitation of these tends to look painted or suspended rather than grown.

Hardness, weakly. Obsidian and soda-lime glass are close in hardness, so this discriminates poorly.

What you can establish

WHAT YOU CAN TELL — a black glassy tool

  · Conchoidal chips, vitreous lustre, no crystals
      → glass of some kind. Does not distinguish
        obsidian from manufactured glass
  · Mould seam at an edge → manufactured
  · Internal banding unrelated to the outline
      → natural flow structure, suggests obsidian
  · Snowflake clusters or iron-oxide streaking
      → obsidian varieties
  · Steel point marks it → softer than 6; consistent
      with either obsidian or glass
  · Obsidian versus black manufactured glass
      → OFTEN NOT DETERMINABLE at home. Needs
        refractive index or spectroscopy

Other black stones, and one confusing name

The dark tool market contains several unrelated materials that get discussed as though they were interchangeable.

“Black jade” is a real thing and usually means very dark nephrite, which is a legitimate material and not glass at all. It is opaque, waxy rather than vitreous, and much tougher — a nephrite piece and an obsidian piece of the same shape are in different mechanical universes.

“Black onyx” in the trade is normally dyed chalcedony, a cryptocrystalline quartz. Genuinely banded black-and-white onyx exists; uniform black onyx is generally a dyed product, and this has been standard practice for a very long time.

Dark carbonate rock sold as bian stone is fine-grained, opaque, soft and non-glassy, and separating it from obsidian by look is straightforward: obsidian is glassy and carbonate rock is not.

Basalt, slate and various dense dark rocks are also carved into tools and are structurally aggregates, so they show grainy rather than conchoidal fracture.

The general point is that “black stone” names a colour, and colour is the weakest identifying feature a material has.

The limit

For obsidian, the honest position is unusually stark. The material is natural glass, so the tests that catch glass cannot catch a glass imitation of it. What you can determine is structural class — that the object is amorphous rather than crystalline — and that is genuinely useful because it separates the obsidian-or-glass bracket from nephrite, serpentine and carbonate rock.

Deciding which of obsidian and manufactured glass you are holding usually needs a refractive index reading, which puts it in a gemmologist’s hands. The consolation is that the two materials behave similarly enough in an object of this kind that the distinction is mostly about origin rather than function.