The Ink Is Held by Cells That Keep Dying
A tattoo is permanent — but not because anything in your skin is. The cells that hold the ink die and are replaced, over and over, for the rest of your life. Permanence is not a thing that sits still; it is a thing that keeps being handed on. In 2018, a lab killed every pigment-holding cell in a tattooed patch of mouse skin at once — and the tattoo stayed exactly as it was. Below, you can run that experiment yourself.
Ask why a tattoo lasts and the usual answer is: the ink gets locked inside your skin cells and stays there. It's a reasonable guess. It's also wrong in the one way that makes the real story worth telling — because the cells holding the ink don't stay. They keep dying. The pigment stays anyway. Understanding how is the whole point.
First: where the ink has to go
A tattoo needle doesn't paint the surface — it punctures through it. Your skin is two working layers, and only one of them can keep a mark.
The outer layer rebuilds itself continuously: new cells form at its base and migrate up, turning over roughly every 36–40 days. Anything caught in it is carried to the surface and shed. Ink left here would vanish within weeks — which is exactly why a scratch-deep "tattoo" fades and a real one doesn't.
Below the epidermis lies the dermis: dense connective tissue that does not desquamate. This is where the needle deposits the ink, and where the pigment settles for good. The stability of the tattoo begins with the stability of this layer — but, as it turns out, not with the stability of the cells in it.
The dermis — operate it
Here is a patch of tattooed dermis. Each teal dot is a clump of pigment; the violet halo around it is the dermal macrophage currently holding it — an immune cell whose day job is swallowing debris. Watch: macrophages die (the halo fades, the pigment sits bare for a moment), and then a new macrophage arrives and re-swallows the same pigment, in place. The carrier cells are being replaced constantly. The picture doesn't move.
running · recapture ON · carrier cells replaced 0× over
The animation is sped up so you watch years of carrier-cell turnover pass in seconds — the dynamics are illustrative. The numbers (half-life, retention) are computed from the model formula at your slider settings, and they match the verifier exactly.
Try the ablation. Press Kill every carrier cell now. Every macrophage dies at once — the halos all vanish and the pigment is left bare. Then new macrophages move in and re-swallow it, and the tattoo is unchanged. This is the actual experiment Baranska and colleagues ran in 2018: using a genetic switch (a diphtheria-toxin receptor driven by the CD64 macrophage gene), they wiped out the pigment-holding macrophages in tattooed mouse skin. The tattoo "kept the appearance it had just before" — the released pigment simply waited, extracellular, until new macrophages recaptured it.
Now turn recapture OFF and ablate again. With no cell to catch the released pigment, each death is a small permanent loss, and the tattoo dissolves. The difference between a permanent tattoo and a fading smudge is entirely this: is there always another cell to hand the pigment to?
The cycle, in three steps
Pigment particles injected into the dermis are too large to drift away. Dermal macrophages — professional swallowers of debris — engulf them and hold them in place.
Macrophages don't live forever. When a pigment-laden one dies, it releases its cargo. The particles are too big to be cleared, so they stay put, extracellular, at the tattoo site.
A neighbouring or newly arrived macrophage engulfs the same particles, in the same place. Persistence depends on the renewal of macrophages, not the longevity of any one of them.
The same story, as one equation
If macrophages hand the pigment on every τ years on average, and a fraction c of it leaks away to the lymph at each handoff instead of being recaptured, then the pigment remaining after time t is a clean exponential:
half-life t½ = τ · ln2 / c — set by cell turnover and leak, not by permanence
This one formula holds both halves of the truth at once. Because c is small, R stays high for a human lifetime — the tattoo is permanent. But because c isn't zero, R still slowly decays — the tattoo fades. Same handoff dynamics; the timescale is the lifespan of a macrophage, and nothing in the skin needs to last forever for the mark to.
Why they fade and blur anyway
Permanent doesn't mean unchanging. Three real processes are the leak term c made concrete:
- Drainage to lymph nodes. Measured in humans: the same pigment species show up in tattooed skin and in nearby lymph nodes. Large clumps stay in the skin; smaller (nano-scale) particles travel off-site. The ink is being slowly, genuinely carried away (Schreiver 2017). That transport is documented; a health harm from it is not established — don't read more into it than the data says.
- UV breakdown. Sunlight chemically degrades pigment over years — which is why an unprotected tattoo dulls faster. Well-supported in principle; a precise in-vivo fading rate is not something we can put a clean number on.
- Dermal remodelling. The collagen matrix is slowly reworked over decades, and pigment spreads a little each time — the reason old lines soften and blur rather than staying razor sharp.
Why removal is so hard — and what the laser really does
A removal laser does not vaporise the ink. It delivers ultrashort, high-energy pulses that the pigment absorbs and that shatter the particles — a photoacoustic shockwave breaking big clumps into fragments (picosecond pulses break them smaller than older nanosecond ones). The laser only breaks the ink up. What actually removes it is your own immune system: fragments small enough to carry are hauled off by macrophages and the lymphatics. That's why it takes many sessions weeks apart — each pass fragments a portion, and clearance is slow.
And the recapture cycle is exactly why removal is "strenuous," in the paper's word: pigment freed by one pass is immediately re-swallowed by neighbouring macrophages before it can be cleared. The authors' proposed fix follows straight from the mechanism — laser the tattoo while transiently ablating its macrophages, so the fragments have no cell to be recaptured by. That's their hypothesis from the mouse work, not established clinical practice.
What the popular answer gets wrong
The check
The retention law and every number below are recomputed — analytically and by a seeded Monte-Carlo agent simulation — in research/why-tattoos-permanent/verify.mjs (10/10 pass). The on-page instrument runs the same model.
| Retention law | R(t) = exp(−c·t/τ) |
| Illustrative parameters (τ handoff time, c leak/handoff) | τ = 1 yr · c = 2% |
| Model half-life t½ = τ·ln2/c | 34.7 yr |
| Retention after 40 yr | 44.9% |
| Ablation: pigment kept when 100% of carriers die at once | 98% |
| Spatial: pigment still on its original site after 25 full turnovers (local vs non-local recapture) | 100% vs 3% |
| Epidermal turnover (cited, not derived) | ~36–40 days |
What the model proves: that permanence and slow fade both fall out of one handoff process; that a single mass ablation removes only the leak fraction and the pattern persists; and that local recapture is what keeps the lines sharp while a non-local one scatters them. What the model is not: a measurement. The mechanism it models is the mouse science (Baranska 2018); τ and c are illustrative constants chosen to land the half-life in the observed multi-decade range, not values measured in human skin. Absolute half-lives are model outputs, not clinical figures.