Wet-plate collodion negative

☠Level D: studied, never performed

Wet-plate collodion is nitrocellulose dissolved in ether and alcohol, poured by hand onto a plate; its fixing was frequently cyanide; and cadmium salts were standard in the salting mixture. The Level D policy names it under the dangerous solvents and again under cadmium and cyanide, and the register classifies this row D.

No procedure appears here — no quantity, no sequence, no conditions. Where a step is named it is named because the chemistry is unintelligible without it.

This is the process that made photography ordinary, and it did so on a licensing accident as much as on chemistry: it was the first good process nobody had to pay to use. It displaced both the daguerreotype and the calotype within a decade.

The chemistry

The halide is made inside the film. A collodion plate is not coated with silver bromide; it is coated with a solution of soluble halides in nitrocellulose, alcohol and ether, and the silver halide forms within the film when the plate is dipped in the silver bath.

Because the crystal grows in place, everything about it — its size, its distribution, how the nitrocellulose holds it — depends on what the halide was dissolved in and what came with it. That is why the choice of cation mattered so much to collodion workers, when in a simple precipitation it would have washed away as a spectator ion. Towler’s reasoning, given for the iodide, transfers: a collodion iodised with an alkaline iodide is unstable and not permanent in its sensitiveness, whereas one iodised with a cadmium salt is more glutinous and much more stable, taking longer to ripen and keeping its sensitiveness far longer. Photographers combined the two, cadmium in excess, and then added a bromide — in Barreswil and Davanne’s 1851 proportion of four parts iodide to one of bromide — because silver iodide alone is sensitive only to the ultraviolet, violet and blue.

⚗Why the plate must stay wet, and why it is a vapour pressure that says so

Diethyl ether and ethanol are a pair and neither is a diluent. Archer records that gun-cotton is not soluble in pure ether free from alcohol — a fact that caused real confusion at the time, because the early accounts said ether and the early experimenters using dry ether could not reproduce them.

The two also do different physical jobs. Ether is the volatile one, boiling at 34 °C: it leaves fast, and its leaving is what turns a poured liquid into a film. Hardwich’s chapter heading says the rest — the use of alcohol to retard evaporation. Alcohol boils at 78 °C, and setting one solvent against the other makes the mixture adjustable: a plate that sets too fast is unworkable, and one that sets too slowly runs.

Then the AIC’s sentence that defines the entire workflow: once the alcohol and ether have dried, the collodion becomes impervious to the processing solution. That is why the plate had to be sensitised, exposed and developed in one continuous run, and why photographers carried darkrooms into fields — the AIC puts the equipment and materials needed to make one negative away from the studio at around 250 kilograms. The constraint that shaped a whole industry is a vapour pressure.

Development is physical. A plate taken from the silver bath is still wet with it, so it carries free silver ions in the collodion as well as the silver halide. Iron(II) sulfate reduces that free silver, and the metal deposits where light has already made a nucleus — one electron from iron(II), one silver atom, a one-to-one stoichiometry. Because the silver comes out of solution rather than out of the grain, the image is plated onto the exposed specks, which is why it looks grey and metallic, why it lies on the surface, and why the plate must not be allowed to dry or be rinsed before the developer is poured.

The acid is the throttle, not the iron. Towler is explicit: in warm weather or after a long exposure more acid is needed “to keep the reducing agent in check”, and for instantaneous work the iron may be used with no acid at all. He adds the observation that decides the whole balance — diminishing the iron and increasing the acid “are correlative expressions”.

Historical workflow

Clean the glass; flow the collodion; sensitise in the silver bath; expose; develop with iron; fix; wash; varnish. All of it in one continuous run, in the dark, near the camera.

Fixing had two answers and the choice showed. Hardwich states that potassium cyanide is a most energetic agent in dissolving the insoluble silver salts, far more so than hyposulphite of soda, and that the double salt it forms is not decomposed by dilution — so a cyanide-fixed plate needs less washing. He did not recommend it for negatives, for a reason that reads very differently now than it did then: negative images dissolve in a fixing agent more easily than collodion positives do, so without much care the cyanide lowers the intensity and whitens the surface. He tells the reader to prefer sulfocyanides or hyposulphite as safer in the hands of a beginner — meaning safer for the negative.

Drying it out was the obvious improvement and it was tried. Towler’s 1864 manual gives the tannin process: after the plate is collodionised, sensitised and thoroughly washed, it is flowed with a preservative solution of tannin, drained and dried, and “when perfectly dry, the plates will keep in the dark for a long time”. His strength is 15 grains to the ounce, about 3.4 per cent, with the range itself a control: the greater the quantity of tannin, the greater the density of the shades. He also records one prohibition with a mechanical reason — such a plate is fixed in hyposulphite of soda “but not of the cyanide, because it is apt to loosen the film”.

The image

Sharper than a calotype and cheaper than a daguerreotype, and printable many times. That is the whole commercial argument, and it is why the second half of the nineteenth century looks the way it does.

Grey and metallic, and lying on the surface, because the development is physical rather than chemical. The distinction is Part IV’s and it recurs across the course: a chemically developed image is built out of the crystal, a physically developed one is plated onto it.

Blue-sensitive, which is why skies are blank and why bromide was added to give the collodion, in Towler’s phrase, a greater capacity for colours.

And the fixer leaves a signature. The Image Permanence Institute records the difference as a process variation with a visible result: fixing in hypo gives a darker, brownish grey image colour, fixing in cyanide a lighter, milky tan. The AIC’s ambrotype entry still describes the developed plate as fixed in hypo or potassium cyanide, because both are in the historical record and both are found in surviving objects.

Permanence

As a negative, it did the job the calotype could not. A glass plate has no fibres to scatter the printing light, so a collodion negative prints sharper than a paper one, and it prints as many times as anybody wants.

The binder has its own clock, and the evidence is from the material’s later career. Nitrocellulose combined with a plasticiser became celluloid, and celluloid became the film base that, in IPI’s phrase, caused many fires and innumerable losses of archival records; Kodak last manufactured nitrate film in 1951. Whether a thin collodion layer on glass behaves like a nitrate film base over a century is a question this course has no source for, and it does not assert one in either direction.

Varnish is part of the object. The AIC records an aqueous pre-coat of gum arabic, dextrin, albumen or gelatin applied before varnishing a collodion negative, so that the alcohol in the varnish does not dissolve the collodion binder — which is a good illustration of how much of this process’s practice is about protecting the binder from its own solvents.

Hazards

Diethyl ether is the reason the process is where it is. It is extremely flammable, volatile at room temperature, and poured in quantity in a small dark space. The Level D policy names the dangerous solvents as one of the families it covers, and this is the case it was written for.

Nitrocellulose is a family rather than a substance, and the safety card draws the line as a threshold: the formulation with more than 12.6 per cent nitrogen is used exclusively for the manufacture of explosives. Hardwich lists four varieties in order of substitution, the most substituted being true gun-cotton and insoluble in ether and alcohol; the less substituted ones are what photography used, which is why the photographic grade is called collodion wool. The course states the threshold and gives no route to either side of it.

Potassium cyanide, where the period fixing is followed, carries an aggregated GHS classification as fatally toxic by every route, and liberates hydrogen cyanide on contact with acid — in a room that has an acid stop bath in it.

Cadmium bromide in the salting mixture is named in the Level D policy on its own account.

Silver nitrate at bath strength, in a standing dipping tank rather than a brushed film.

Five families of hazard in one operation is unusual even among the Level D processes, and it is why Part XXVI gives this process a lesson of its own.

Why we study it and do not reproduce it

Because it is why the nineteenth century looks the way it does. Between about 1851 and the arrival of the dry plate, almost every photograph that reached an audience passed through this material.

Because the binder’s permeability, not its optical quality, is what decides whether a plate can be developed at all — which is a lesson about binders that Part V’s gelatin work depends on and that no other process states so plainly.

Because it made photography cheap by being unpatented, which is a fact about law rather than chemistry and is worth setting beside Talbot’s patents and Daguerre’s English one.

And because two of its children are in this atlas as separate rows. The ambrotype and the tintype are the same film read against a dark ground rather than printed from, and understanding that is the cheapest possible demonstration that a negative and a positive can be the same object seen two ways.

Where the course carries it, and it carries no steps. Part XXVI owns it, in Wet-Plate Collodion: The Process That Made the Nineteenth Century Look Like That, which gives the chemistry and the history at Level D and no procedure at all. That is the part’s rule rather than this process’s misfortune: its overview sets out what Level D means and why every page in it has no steps.

The lesson carries the arguments this entry compresses. A solution, not an emulsion is the distinction that governs everything downstream, and why the plate had to stay wet follows from it — which is this entry’s point about permeability deciding whether a plate can be developed at all, argued rather than asserted. The developer plates rather than reduces, which is a different mechanism from anything in Part VIII. Three objects out of one chemistry is where the negative, the ambrotype and the tintype are separated properly. And the varnish is not a finish; it is the reason the object still exists — a conservation fact this entry does not otherwise carry.

The hazards are counted rather than gestured at, which is the section that justifies the classification: ether, the cadmium salts, and the cyanide fixing that its own lesson takes up in full — including why any acid in the room turns that salt into a gas. The lesson also describes what present-day practice actually looks like, and where to go if you intend to pursue it, which is what the Level D policy requires of a page that gives no steps.

Assignment: Assessing a Historical Process for Hazard is where a reader does the classification themselves, on three period procedures, with a deliverable that may not contain a procedure — including for the one judged least hazardous.