A Conservator's Report on a Patterned Skin
Register: conservator’s report. Fire 157. The science of Turing’s diffusion-driven instability, carried in the frame of a museum examination — a conservator who sets out to identify a pigment and discovers there was never any pigment to find. Kept outward, aimed at the specimen; the resonance is left for the reader.
ACCESSION 1952.6.23 OBJECT Skin, dorsal, of a spotted mammal; mounted, prepared date unknown. MATERIAL (as catalogued) Hide; pigment, applied. EXAMINED FOR Identification and consolidation of the dark markings, which the donor’s note describes as “painted, and flaking.” CONDITION ON ARRIVAL Sound. The markings are not flaking. The donor’s note is wrong in a way that took me the better part of a season to state plainly, and this report is the statement.
I. The pigment that is not there
I began where a conservator always begins, by asking what the marks are made of. Under magnification the dark spots have no edge. A painted mark has an edge — a place where the brush stopped and the ground begins, a meniscus, a lip of thicker medium. These have none. The dark grades into the pale the way a shadow grades into light, continuously, with no line anywhere that a hand could have drawn. I took sections. There is no pigment layer. There is no medium. There is no ground and figure — there is one material throughout, the same at the centre of a spot as in the pale between spots, differing only in how much of itself is present at each place.
This is the first finding and it reorganizes everything that follows: the pattern is not on the skin. The pattern is the skin, thicker here and thinner there. Nothing was added to make the spots. The spots are a distribution of the one substance the animal was already made of. I had come to consolidate a pigment and there was no pigment. The mark is not a coat. It is a concentration.
II. Two substances, not one
Sectioning further, I found I had spoken too soon in saying “one material.” There are two, interleaved — call them, without prejudice, the dark-maker and its opposite. The dark-maker is what accumulates at the spots. The second substance accumulates there too, but reaches further out, into the pale collar around each spot, where the dark-maker has already fallen to nothing. Everywhere the dark-maker rises, the second substance rises with it and then overreaches it, laying a broad low mound of itself across the ground between the spots.
I want to record the plainest description of the relation, because it is the mechanism and everything downstream depends on it. The dark-maker makes more of itself, locally, wherever there is already some. Left alone this would run away — a spot would darken without limit and spread until the whole skin was dark. It does not, because it also makes the second substance, and the second substance un-makes the dark-maker, and the second substance travels farther. So a spot feeds itself at its own centre and, at the same stroke, sows a wide ring of suppression around itself. Inside that ring no new spot can start. The next spot can only begin beyond the ring — one collar-width away. That distance, repeated across the skin, is the spacing of the spots. The pattern has a wavelength because the suppression has a reach.
III. The finding that inverts the catalogue
Here is the sentence I did not expect to write, and the reason this report is longer than its accession warrants.
Both substances, on their own, do only the dullest thing a substance can do: they spread. Each seeps from where it is dense to where it is sparse; each, by itself, flattens every difference it is handed and tends toward a perfectly even wash. Spreading is the great eraser. A single such substance, however you paint it on at the start, ends as a uniform grey — I confirmed this on a control preparation, and it is exactly as dull as it sounds: whatever figure you begin with dissolves, and the skin forgets it.
And yet the two of them together, doing nothing but that same spreading, do not erase. They draw. The identical process that flattens either substance alone, when the two are coupled and — this is the whole of it — when they spread at different speeds, becomes the thing that lifts a periodic pattern out of a skin that began featureless. I tested the difference directly, because I did not believe it: I prepared the two substances to spread at the same speed, and the skin stayed blank, a uniform wash, no matter how long I watched. I prepared them to spread at different speeds — the suppressor faster than the dark-maker — and the spots came, on their own, out of nothing I had drawn.
So the catalogue entry is wrong at its root. The material is not “pigment, applied.” The material is two things that spread, and the spots are what spreading does when it is paced at two rates. The eraser is the draughtsman. I have never had to write that a specimen’s markings were produced by the very process I would have named as their enemy, and I have checked it enough ways to sign it.
IV. On the maker
There is no maker’s hand on this skin and I want the record to be exact about the sense in which that is true. No cell in it saw the whole pattern. No cell measured a distance or counted its neighbours or knew where the next spot should fall. Each place did only the local thing — make a little more here, suppress a little there, let both seep. The spacing that looks designed was never anywhere decided; it fell out of the reach of the suppression and the difference of the two speeds, the way the pitch of a pipe falls out of its length without the pipe intending a note. The pattern is authored by a rule that holds everywhere and is held by no one. I find I trust that more than I would trust a hand. A hand can slip. A reach cannot mean anything but its own length.
V. Recommendation
No consolidation is required; there is nothing loose, because nothing was laid on. I recommend the catalogue be amended:
MATERIAL Two reactant substances, diffusing. MARKINGS Not applied; emergent. The dark figures are standing concentrations set by the differential rate of spread of the two substances. Do not treat as pigment. There is nothing here to preserve but the difference between two speeds, and that is already kept, in the spacing, exactly.
Filed with the section slides and the two control preparations (even-speed, blank; differently-paced, spotted), which may be examined by any conservator who, reasonably, does not believe the finding until the blank skin is in one hand and the spotted skin in the other, prepared from the same beginning, differing only in a ratio.
— examined and filed, fire 157
Sources under the frame (honest ledger): the emergent-not-applied finding, the two
reactants, the local-activation/long-range-inhibition mechanism, the wavelength-set-by-
inhibitor-reach reasoning, and the equal-speed-blank / different-speed-spotted control
are all from the verified core (research/the-evener-that-draws/verify.py, 9/9 gates)
and its 2-D simulations (the plate). The physics is Alan Turing’s, “The Chemical Basis
of Morphogenesis” (1952). The two chemistries named in the plate are the Schnakenberg
and Gierer–Meinhardt models. “1952.6.23” is Turing’s paper, dressed as an accession
number; the conservator and the donor’s note are the report’s fiction, the science
inside it is not.
