How to split rocks in the field
Of 109 mineral species this site covers, 57 cleave perfectly and only 15 do not. The specimen is usually weaker than the rock, and that decides where you hit.
Work the matrix, not the specimen. Across the 109 species sheets in this site's reference set, 57 record perfect cleavage and only 15 record none, so the crystal in the rock is usually the weakest thing in it. Split with a chisel along an existing plane, keep the hammer off the mineral, and take more rock than you think you need.
In short
- The specimen is more likely to split than the rock is. 57 of 109 species in this site's reference set record perfect cleavage. A blow transmitted through a specimen finds the cleavage plane inside the crystal before it finds anything in the host rock.
- Only 15 of the 109 have no usable cleavage — quartz, pyrite, crocoite, wulfenite, willemite, vesuvianite, scheelite, chalcopyrite and seven others. Those are the specimens you can be relatively rough near. Everything else, you cannot.
- 61 of the 103 species with a recorded hardness sit below 4, which is softer than the steel of the chisel you are holding. The tool marks the mineral, not the other way round, so the chisel edge never touches the crystal.
- Take the whole block out first and trim it at home. Field trimming is done standing up, in bad light, with cold hands and a heavy hammer. Every irreversible decision made in that state is made badly.
- Permission and safety come before technique. Nothing on this page is legal advice; whether you may collect at all is covered on is it legal to collect minerals in the UK, and dump safety on is it safe to collect on old mine dumps.
| Cleavage recorded | Count in the reference set | Examples | How close the chisel can work |
|---|---|---|---|
| None, poor, indistinct or not observed | 15 | Quartz, pyrite, crocoite, wulfenite, vesuvianite, scheelite | Closest. These fail by fracture, which is less predictable but not planar |
| Distinct, in traces, or interrupted | 28 | Arsenopyrite, marcasite, mimetite, pyromorphite, skutterudite | Moderate. A parting exists but takes real force to open |
| Good | 9 | Pharmacosiderite, cerussite, anglesite | Keep the blow in the matrix; work from 3 cm out |
| Perfect | 57 | Calcite, fluorite, baryte, galena, muscovite, talc, brucite, the zeolites | Furthest. Never strike a block containing one of these on a face that points at the specimen |
The survey behind the rule, and what it actually counted
Method. On 20 September 2026 the Handbook of Mineralogy sheet for every species this site covers was converted to text and the Cleavage field extracted by regular expression. That gives 109 sheets after removing duplicates, mirror copies and non-species files. Each was classified on the Handbook's own word.
Result: 57 perfect, 9 good, 15 with none or nothing usable, and 28 in between. More than half the species a collector is likely to be digging for have a plane inside them along which they will part cleanly under a fraction of the force needed to break the host rock.
The second count matters as much. Of the 103 sheets recording a hardness, the distribution by minimum value runs: 7 species below 2, 28 in the 2s, 26 in the 3s, 13 in the 4s, 16 in the 5s, 9 in the 6s, and 4 at 7 or above. Sixty-one of 103 are below hardness 4 — softer than a steel chisel edge, softer than a hammer face, and softer than most of the country rock around them.
Put those two numbers together and the technique writes itself. The thing you are trying to extract is usually both softer than your tools and more cleavable than the rock holding it. Everything below follows from that, and the identification side of the same evidence is on cleavage or fracture, and how to tell.
The tool order, and why the hammer is not first
A club hammer used alone delivers a shock wave through the whole block. A chisel concentrates the same energy into a line you choose. The hammer is the power source; the chisel is the steering. Using the hammer alone on a specimen block is the single commonest way collectors destroy what they came for.
1. Read the block before touching it. Find the bedding, the joint, the vein wall, the old fracture. Rock splits along planes it already has far more readily than along ones you are trying to create. If the block has no plane pointing the right way, take the whole block.
2. Set a cold chisel in the existing plane, well away from the specimen. Three to five centimetres of matrix between the chisel edge and the nearest crystal is a reasonable working minimum for a perfect-cleavage species, and you can work closer on the fifteen that do not cleave.
3. Tap, do not swing. Repeated light blows walk a split open along the plane. One heavy blow starts a fracture wherever the rock is weakest, which is often through the vug.
4. Use wedges for anything large. Two or three plug-and-feather sets in a line open a bed steadily and symmetrically. A wedge line is slower than a sledge and it is the only method that gives any control over where a big block parts.
5. Wrap before it goes in the bag, not when you get home. A specimen that survives extraction and then travels loose in a rucksack has not survived. The packing standard is on how to pack and ship mineral specimens.
The three mistakes that account for most losses
Splitting along the wrong plane. A vein-filling of calcite or baryte will part on its own cleavage rather than along the vein wall, and the split runs through the crystal lining rather than behind it. The cure is to attack from the country rock side, not from the vein side, so the crack arrives at the vug from behind.
Hammering the specimen instead of the matrix. This is usually not deliberate: a collector strikes a block held in one hand, and the shock travels through the whole piece. Put the block down on solid ground or on another rock. A block resting on soft ground absorbs the blow and returns nothing.
Trimming in the field to save carrying weight. Understandable and almost always wrong. The trim you make on a hillside cannot be undone, the specimen is dirty so you cannot see what you are cutting into, and a carried kilogram of waste rock costs an hour of effort where a bad trim costs the specimen. Take the block; trim it clean, dry and in good light.
A fourth, less common but worse: prying with the chisel as a lever. Cold chisels are hardened and brittle, they snap under side load, and the fragment goes somewhere. Eye protection before the first blow, every time. Broader site-safety questions are dealt with on is it safe to collect on old mine dumps, and the rest of this section is indexed at collecting guides.
What to do with the block when you get it home
The block that leaves the field is not the specimen. Trimming is a separate operation done on a bench, and the first decision is whether to trim at all — a piece with its matrix intact carries more information than a trimmed one, and the fashion for close trimming destroys context that cannot be put back.
When you do trim, the same cleavage logic applies with more control: nibblers or a trim saw rather than a hammer, cuts made from the back, and the specimen supported along its whole length so it is never bridging a gap. Cleaning comes after trimming, not before, and is covered on how to clean mineral specimens — with the important exception of pyrite, which has its own rules on how to clean pyrite safely.
Label it the same day. A self-collected specimen with a date, a grid reference and a note of the working is worth keeping; the same specimen without them is a rock. What goes on the label is set out on how to label and catalogue a collection, and what self-collected means to a later buyer is on what self-collected means on a label.
We do not sell specimens and there is no catalogue on this site. If you are after material from a working or a locality you cannot reach yourself, the route is the wanted list.