A shadow box is not a picture frame with extra depth. It is a small sealed room. Once the glass goes on, whatever is inside — moisture, a stray fibre, the beginnings of a beetle — is sealed in with the work for as long as the piece exists. Everything I do before that moment is either helping the piece survive or quietly setting a fuse.
People ask me fairly often how long one of these takes. The honest answer is that the composition takes an afternoon and the construction takes three days, most of which is waiting: waiting for a specimen to relax, waiting for it to dry on the setting board, waiting for adhesive to cure before anything goes near glass. What follows is every layer between the wall and the wings, from the back forward, and the reasoning behind each one.
Why depth changes everything
A flat frame presses its contents against glazing. That is fine for a print. It is fatal for a three-dimensional specimen: a butterfly compressed against glass loses scales at every contact point, and those scales are the colour. They do not grow back and they cannot be replaced. Wing scales are modified hairs, each one socketed into the wing membrane, and they shed under the lightest abrasion — a fingerprint on a forewing is permanent.
So the whole design problem is clearance. The specimen needs air around it on every side, and the box has to hold that air without the contents shifting when the piece is carried, posted or knocked by a passing elbow. My working minimum is 12 mm of clearance between the highest point of the specimen and the inside face of the glazing. For a swallowtail with a strongly arched thorax that can mean 20 mm or more.
Rule of thumb
Measure the deepest specimen at its highest point, add 12 mm, then add the thickness of your mount and backing. That total is the minimum internal rebate depth. If the frame you like is shallower, you need a different frame, not a flatter specimen.
The stack, back to front
Every box I build is the same seven layers, in the same order. The materials change; the sequence does not.
| Layer | Material | Depth | Job |
|---|---|---|---|
| 1 · Dust seal | Kraft or Tyvek tape | — | Closes the back against airborne spores and insect eggs |
| 2 · Backing board | 4 mm acid-free foam board | 4 mm | Rigid substrate that will take a pin without splitting |
| 3 · Ground | Aged paper, board or painted panel | 0.5 mm | The visual field the specimen sits against |
| 4 · Specimen & flora | Pinned insect, dried botanicals | 8–18 mm | The work |
| 5 · Air gap | Nothing at all | 12 mm min | Keeps scales off glass; lets the piece read as dimensional |
| 6 · Spacer | Acid-free barrier strip | varies | Holds the glazing off the contents and hides the rebate |
| 7 · Glazing | UV-filtering acrylic or glass | 2–3 mm | Blocks dust, pests and most ultraviolet |
1 · The backing board
Acid-free, always. Ordinary cardboard and MDF release acidic vapour as they age — the same slow chemistry that turns the pages of a paperback brown at the edges. In a sealed box those vapours have nowhere to go. Over a decade they will yellow a pale ground and embrittle anything organic sitting on it. Foam board with acid-free facings costs perhaps two dollars more per piece. There is no argument for the cheaper option.
Thickness matters for a practical reason: the pin has to bite. A 4 mm board holds a size 3 entomological pin firmly enough that the specimen will not rotate. Anything thinner and the pin pivots the first time the box is tipped.
2 · The ground
This is the layer where the piece stops being a specimen and starts being art, and it is the only layer I let myself be undisciplined about. I use tea-stained paper, torn book pages, gold leaf, india ink washes, dyed silk. The one constraint is that whatever goes in must be fully dry and fully cured before it is sealed. Oil paint is out — it off-gasses for months. Acrylic needs a week. Ink needs a day. Anything still evaporating inside a sealed box will condense on the inside of the glazing and stay there.
3 · Pinning
Stainless steel entomological pins, sizes 2 to 5 depending on the thorax. Not sewing pins — those are nickel-plated brass or plain steel and they corrode, staining the specimen with a green or orange bloom at the entry point within a few years. The verdigris ring around an insect in an antique case is almost always a cheap pin, not age.
The pin goes through the right side of the thorax by convention, vertical, and stops with the specimen sitting about a third of the way down the shaft. That leaves a handle above and enough below to seat firmly in the board.
Relaxing and setting: the slow part
A dried specimen is brittle to the point of absurdity. Attempt to open the wings of an insect that has been dry for a month and you will hear a sound like a dead leaf and be left holding two halves of a butterfly. Before anything can be posed it has to be relaxed — rehydrated until the joints move again.
The relaxing chamber is the least glamorous object in my studio: a sealed plastic box with damp paper towel in the base, a rack to keep the specimen off the water, and a few drops of an antifungal so the humidity does not grow anything of its own. The specimen goes in, papered, and the box is closed.
| Specimen | Time in chamber | Note |
|---|---|---|
| Small butterfly (wingspan < 50 mm) | 12–24 hours | Check early; over-relaxing invites mould |
| Medium butterfly (50–100 mm) | 24–48 hours | The usual case |
| Large swallowtail or birdwing | 48–72 hours | Thick thorax takes the longest |
| Beetle | 3–5 days | Chitin resists rehydration |
| Detached wing | 2–6 hours | Membrane only; watch it constantly |
Once relaxed, the specimen is pinned to a setting board — a grooved block with the channel taking the body and the flat shoulders taking the wings. The wings are drawn forward with a pin behind a strong vein (never through the wing itself, which leaves a hole) until the trailing edge of the forewing is at right angles to the body. That right angle is the museum convention and the reason pinned butterflies all look subtly the same. Strips of glassine, weighted, hold everything flat.
Then the piece sits. Small specimens dry in three or four days, large ones in two weeks. Take one off the board early and the wings will slowly drop over the following months — a slump you cannot undo without starting the whole process again.
Glazing: the decision people get wrong
The glazing choice is the single biggest determinant of how the piece looks in ten years, and almost nobody thinks about it at the point of purchase.
| Option | UV blocked | Weight | Verdict |
|---|---|---|---|
| Standard float glass | ~45% | Heavy | Cheap, clear, breaks into the specimen. Avoid for shipping. |
| Standard acrylic | ~60% | Light | Scratches easily, attracts dust by static. Acceptable. |
| UV-filtering acrylic | ~98% | Light | What I use. Best survivability per dollar. |
| Museum glass (anti-reflective, UV) | ~99% | Heavy | Gorgeous, invisible, four times the price. |
The reason to care: ultraviolet light degrades the pigment in wing scales the same way it fades a poster, but faster, because those pigments were never engineered to sit in a window. Reds and yellows go first. A red lacewing in unfiltered daylight will visibly shift within five years. Behind UV acrylic in the same spot it will still be red when your grandchildren inherit it.
Note that the structural colours — the metallic blues and greens of a Papilio, produced by the physical shape of the scale rather than by pigment — do not fade at all. They are colour made out of geometry. Nothing bleaches a shape.
Sealing, hanging and the things that go wrong
The back is closed with a continuous dust seal — kraft tape or Tyvek, edge to edge, no gaps. This is not decorative. A carpet beetle larva needs a gap of about half a millimetre to get in, and once inside a sealed box with a dead insect it has no reason ever to leave. Most infested antique cases I have opened had a lifted corner on the backing paper.
Hanging hardware goes on before the seal, because retrofitting means punching through a finished back. Two D-rings at one third down the sides, wire with no more than 25 mm of slack. For anything over about 2 kg, two hooks rather than one — a single point lets the piece rotate, and a rotating shadow box will eventually shed its flora into the bottom of the frame.
The four failures I see most
- Glue on the specimen. Someone adheres a butterfly directly to the backing to avoid pinning. The adhesive wicks into the wing membrane, darkens, and the piece is unrecoverable. Pins are removable; glue is a decision you make once.
- Sealing damp flora. Botanicals that feel dry can still hold internal moisture. Sealed in, that moisture equalises into the air of the box and condenses on the inside of the glazing on the first cold night. Then it grows mould.
- Non-entomological pins. Discussed above. Verdigris within five years.
- Hanging it in the sun because it looks lovely there. It does look lovely there. See the next paragraph.
That last one deserves its own article, and it has one — see Caring for Specimen Art for the numbers on light levels, humidity and pest inspection.
A shadow box built properly is a hundred-year object. Built carelessly it is a five-year object that looks identical on the day you buy it. Studio notes
A build checklist
If you make your own, this is the sequence I would hand you on a card.
- Confirm the frame's internal rebate is deep enough for specimen + 12 mm + mount.
- Cut acid-free foam backing to a snug — not tight — fit.
- Finish the ground layer. Let it cure fully. Days, not hours.
- Relax the specimen. Check every twelve hours.
- Set on the board with stainless entomological pins. Wait out the drying.
- Dry-lay the composition on the backing before committing a single pin.
- Pin the specimen. Attach flora with tiny points of reversible adhesive, never a bead.
- Fit the spacer. Clean the glazing on both faces under raking light.
- Assemble, inspect through the glass for fibres, then close.
- Fit hanging hardware. Seal the back edge to edge.
- Sign and date the back. Note the species and its source.
That last step matters more than it sounds. A specimen without provenance is just a dead insect. With a species name and a source it becomes a record — and in fifty years, when nobody remembers what was in the box, the writing on the back is the whole story.
