Light-Hardened Colloids: Bitumen, Dichromate and the Pigment Processes

Open a well-made book of photographs published between 1870 and 1900 — a portfolio of actresses, a survey of a city, a volume of art reproductions — and a great many of the pictures in it will not be photographs. Some books of that period do carry real prints, tipped in by hand. The rest are ink, or pigmented gelatin, pressed onto paper by a machine, and every one of those machines was aimed by a photographic exposure onto a layer of gelatin, gum or albumen carrying a chromium(VI) salt. Wall’s 1912 dictionary measured the reach of it in a single index entry: under Bichromate Methods he lists gum, carbon, photogravure, collotype, oil and bromoil, and most photographic block and plate making besides.

That is the scale of the family this page is about, and it is the reason a course could not leave it out and call itself complete. It is also a family in which not one process shares a single mechanism with anything else in this course. There is no latent image. There is no development that amplifies a small change into a large one. There is no fixer, and none is needed. What light does here is change how soluble a layer is, and the picture is whatever survives the wash.

This page explains that chemistry properly, tells the history that the chemistry produced, describes the objects well enough that you could pick one out of a box, and gives no procedure for any of it — no quantity, no bath, no sequence, no time, no temperature, and no substitution that would make one possible.

Where this page sits, and what it does not repeat

Seven entries in the process atlas already carry these processes one by one, and the chemical encyclopaedia carries the salt. Part I owns the Niépce story and the invention narrative that goes with it. The chromium policy owns the rule, and the Level D policy owns the reason a page like this exists at all.

What is left for a lesson is the part none of them can do: the argument that these are one family, the mechanism worked through at the resolution the evidence actually supports, the engineering problem that shaped fifty years of invention, and the exclusion stated as a decision rather than as a reflex.

The idea that is not silver

Every silver-halide process in this course runs on the same four-step logic: a photon frees an electron, a handful of silver atoms accumulate at a sensitivity site, a developer finds those few atoms and reduces the whole crystal, and a fixer removes what was never developed. Delete any one of those and the process stops working.

The dichromated colloid deletes three of them at once, and the fourth is not recognisable.

No latent image. There is nothing invisible to be found later. The chemical change light makes is the whole of the effect, and it is a change in the layer, not in a dispersed crystal.

No amplification. In a silver halide crystal a few atoms of photolytic silver make a whole crystal developable, so the developer supplies almost all the chemical work and the light supplies only the instruction. Here there is no such lever. Ware’s estimate for Niépce’s bitumen — under 10⁻⁶ ISO, which works out at a camera exposure of roughly five days — is the price of that in one number. Dichromated gelatin is very much faster than that — Cassell’s 1911 comparison puts sensitised carbon tissue at around half the exposure a printing-out silver paper needs to reach a printable depth, which places the family respectably on the printing-out scale — but it is still a printing-out material with no gain, and it was a contact-printing material throughout its working life rather than a camera one.

No fixer. A silver print has to be fixed because the halide that was not developed is still light-sensitive and still in the sheet. In a dichromated colloid the residue is simply soluble, so water carries it away and there is nothing left that light can act on. Mungo Ponton said so in 1839, in the plainest words the subject ever got: to fix it, all that is required is careful immersion in water.

And development is a solvent. Warm water for gelatin, cold water for gum, a mixture of lavender oil and a petroleum distillate for bitumen. Nothing is reduced, nothing is amplified, nothing is converted. A quantity of material is removed, and the picture is the remainder.

⚗The consequence that makes the whole industry possible: tone is a thickness

In a silver print, density is how much metal is sitting in a layer of fixed thickness. In a dichromated colloid, density is how much colloid is left, and the colloid carries its pigment with it. The image is therefore a relief: physically thicker where more light fell.

That single fact is why this family, and not silver, produced the printed page.

A relief can be cast from — press it into soft metal and you have a mould whose depth varies with tone, which is the Woodburytype. A relief can be etched through — where the hardened layer is thin the acid reaches the metal sooner and bites deeper, which is photogravure. A relief of hardened gelatin holds water differently from soft gelatin, and greasy ink goes where the water is not, which is collotype. And a relief is a resist: it protects what is under it, which is what Niépce was actually after and what the semiconductor industry still does.

None of those is available to a silver image, because a silver image is a stain rather than a shape.

Bitumen: the first member, and one honest gap

Bitumen of Judea belongs to this family, and the belonging is not a loose analogy — it is the same operational fact. Light makes the coating insoluble in the solvent that dissolves the coating light did not reach, and a solvent then develops it. The Part I lesson tells the story, the heliography atlas entry records the object, and the coating formulary entry holds what Niépce published. Three things are worth adding here that none of them is the right place for.

The spectral behaviour is not the dichromate’s, and the difference is diagnostic. The 1911 Britannica survey gives the action of the spectrum on a dichromate as confined to the most refrangible end, beginning in the ultraviolet and reaching no further than the blue-violet — and, crucially, that a dichromate solution absorbs only those rays which are effective in altering it. Asphaltum behaves differently: the same survey reports its action continuing into and below the red, with the blue rays most effective. Two materials in the same family, with different absorption and therefore different useful light. That correlation between absorption and chemical action is a rule of photochemistry worth carrying out of this page: a substance is altered by the light it absorbs, and by no other.

The mechanism is genuinely unresolved, and the course says so rather than repeating a phrase. The nineteenth century said the bitumen was “hardened”; the twentieth sometimes said “polymerised”; a modern reader will meet “photo-crosslinked” and “oxidatively insolubilised” as well. Those are not synonyms — they name different chemistry, with different roles for atmospheric oxygen and different predictions about what an aged coating contains — and this course has not read a peer-reviewed mechanism paper on bitumen photochemistry. It therefore states the phenomenology, which is certain and reproducible, and does not state a mechanism, which is not.

The reason is not laziness on anyone’s part. Bitumen is a natural mixture, not a compound; Ware, who is a chemist, calls its molecular structure so complex and variable as to constitute a chemist’s worst nightmare. A mixture does not have a mechanism, and a claim that it does is the sort of tidy sentence that survives in textbooks precisely because nobody can check it.

☠Why the bitumen route is excluded, and why the reason is not the bitumen

The cold solid has no notified GHS classification at all. That is an absence of classification, not a finding, and the nearest classified material — petroleum asphalt on ILO-WHO card 0612 — records that fumes of the substance are possibly carcinogenic to humans and that a harmful concentration of airborne particles is reached quickly when it is dispersed.

But the hazards that actually govern belong to the other three materials the process requires, and they are different in kind from anything else in this part. Development is a warm organic solvent worked in quantity. The historical purpose — a printing plate — ends in nitric acid. And the industrial descendants of the process, in the British patent abridgments of 1877 to 1883, dissolve the resin in benzene, which EH40 sets at 1 ppm with both the Carc and Sk notations: the tightest solvent limit anywhere in this course.

So the bitumen page is Level D for a reason that has nothing to do with acute toxicity and everything to do with the assessment the course would have to write first, for solvents and an acid it does not otherwise handle. Naming the right reason matters. A reader told that bitumen is dangerous learns nothing; a reader told that the solvent load and the etch are the problem can assess the next resist process they meet.

1832, 1839, 1852, 1855: four claims that get collapsed into one

The sentence “Ponton discovered that dichromates are light-sensitive” appears in a great many books, and it is wrong in at least two directions at once. The photoglyphic engraving entry sets out the priority dispute and the timeline records the corrections; what a lesson can add is why each step was a different kind of discovery, because that is the transferable part.

1798 — Vauquelin found that chromic acid forms a carmine-red silver salt that darkens in light. Eder records it and then sets it aside, and he is right to: that is a silver compound behaving like a silver compound. It belongs to Part IV’s chemistry, not to this one.

1832 — Suckow found that chromate salts mixed with organic substances are light-sensitive with no silver present at all. Eder: he must be recognised as the first discoverer of this light reaction. This is the discovery of the family, and it is the one that gets forgotten, because it produced no picture.

1839 — Ponton made the first photographic application of it, on paper, and reported that fixing required only immersion in water. That is a discovery of a use, not of a phenomenon. Eder adds two things about it that a careful reader should keep. Ponton’s own explanation of the chemistry was, in Eder’s words, quite incorrect. And Ponton missed the important half: he did not mix his dichromate with a colloid.

1840 — Becquerel repeated the work and identified what had actually been happening. Ponton’s paper was not a bare dichromate coating: the Getty’s account is that the cellulose of the sheet and the small amount of gelatin present as internal size were the only organic material available for the photochemical reduction, and Becquerel showed that the size mattered. He then went further and reported that paper deliberately sized with iodide of starch was more sensitive than unsized paper. That is the moment the organic component stops being a contaminant and becomes an ingredient.

1852 — Talbot patented the mixture: dichromate with gelatin, glue, gum or starch, and the finding that light renders it insoluble. Eder gives the English patent as 29 October 1852, with the full account published as Gravure photographique sur l’acier in the Comptes rendus the following year. Talbot’s own description of the developed plate is worth reading closely, because it describes something subtler than “insoluble”: after washing, the light image appeared somewhat prominent, since the water washed away the chromium salt from the parts affected by light and swelled the glue coating somewhat. Differential swelling, not simple dissolution — and swelling is the property collotype would later be built on.

1855 — Poitevin turned it into two industries in a single patent specification. Eder’s reading of the December 1855 English patent finds both halves in it: dampen the exposed colloid and roll it up with greasy ink, which only adheres to the parts exposed to light — that is collotype — and, in the same document, the remark that coloured prints could be had by mixing a pigment into the colloid and washing away the portions light had not changed, which is carbon printing and, by extension, gum. One specification, two families, and both of them still in use.

⁇Three cautions about the names, and one about the spelling

Priority here is genuinely tangled, and the tangle is informative rather than embarrassing. Pretsch’s English patent for a gravure process using the swelling property is dated 9 November 1854, ahead of Poitevin’s English filing; Poitevin’s French licence is dated 6 August 1855; Sutton and Pouncy reached a pigment process independently in England in 1858; and Cassell’s 1911 encyclopaedia credits Pouncy, whose patent it dates 10 April 1858, with being the first to actually produce carbon prints rather than to describe them. Being first to see, first to file and first to make are three different achievements, and the sources disagree about which one they are ranking.

The Getty atlas spells Talbot’s process “photoglyptic engraving”; this course and most of the literature spell it photoglyphic. The course does not correct the Getty and does not treat the variant as evidence of anything; it is recorded here so that a reader searching the atlas can find the passage.

And Ponton’s date is reported as both “May 1839” and simply “1839” — the Britannica gives the month, Eder gives the year and the venue as the Royal Society of Scottish Artists. Where two Tier 1 sources give different precision for the same event, the course prints the less precise one and notes the other, rather than choosing.

The chemistry, stated to the resolution the evidence supports

Everything above depends on one photochemical reaction, and the honest account of it is shorter than most books make it.

Light does not act on the dichromate by itself. The Britannica states the constraint without hedging: the change is only possible in the presence of organic matter of some kind, such as gelatin or albumen. That is why Ponton’s plain paper worked at all — the sizing was the organic matter — and why Becquerel’s starch-sized sheet worked better.

What light drives is a reduction of chromium. The Getty calls it the photochemical reduction of the dichromate salt: chromium(VI) takes electrons from the colloid around it and falls to chromium(III). The reduction half-reaction in acid is standard and tabulated:

Cr₂O₇²⁻ + 14 H₃O⁺ + 6 e⁻ → 2 Cr³⁺ + 21 H₂O

Dichromate reduced to chromium(III) in acid, standard potential +1.232 V — the half of this photochemistry the course can write down

Six electrons per dichromate ion, and fourteen hydronium ions to carry the oxygen away as water, which is why the acidity of a coating is not a detail.

The other half is missing, and the course leaves it missing. Those six electrons come from the colloid — from gelatin, from gum arabic, from the paper’s own sizing — and no source this course has read says which groups are oxidised, into what, or in what proportion. So there is no second equation on this page. Writing one would mean inventing chemistry, which is the one thing this course will not do, and a plausible-looking equation is far more dangerous here than a stated gap, because a reader cannot audit it.

!What the course asserts about the cross-link, what it does not, and what would settle it

Asserted, on the Getty’s account and on the identical statement in the potassium dichromate entry: chromium(VI) is photochemically reduced in the presence of the colloid; the reduced chromium binds the colloid; a bound network no longer swells and dissolves in water. That is enough to explain every observable in this page.

Not asserted: which groups on the polymer are bridged, in what ratio, and whether the bridging species is chromium(III) itself or a shorter-lived chromium(V) or chromium(IV) intermediate formed on the way down from chromium(VI). Those are real, answerable questions with a real literature behind them, and this course has not read that literature. Nor has it found an account of the mechanism by which chromium(VI) is carcinogenic, and it does not publish one.

What would settle it, so a reader knows what a good answer looks like: a spectroscopic identification of the chromium species actually present in an exposed layer, dated and quantified; and an identification of the bridged residues on the protein or the polysaccharide, from a polymer or photographic chemistry reference rather than from a process manual. Until the course has read those, it prints the sentence it can support and marks the boundary.

Why this matters more than it looks. An exposed carbon print carries chromium for the rest of its life, and a conservator reasoning about that object needs to know what species it is. Elemental analysis cannot tell them, and neither can this page.

The dark reaction, and why keeping properties are part of the process

A sensitised dichromated colloid does not sit still. The insolubilisation, once started, continues in the dark, and every period manual treats this as a working constraint rather than a curiosity.

Cassell’s 1911 encyclopaedia gives it a name — the continuing action of light — and three properties. It occurs only in the carbon and kindred processes. It can complete a partially exposed print stored in a perfectly dark place. And it is very slow and uncertain, depending on the presence of dampness in the air, so that it runs much faster in wet weather than in dry and can be prevented altogether by keeping the material in an absolutely dry receptacle such as a calcium tube — a desiccator, in modern words.

A working carbon printer in the modern revival describes the same thing from the other end. Sandy King calls it the dark effect, a slow insolubilisation of the gelatin of sensitised but unexposed tissue, whose practical consequences are a gradual gain in speed and a loss of contrast — which is exactly what a partial, unmodulated hardening of the whole layer would produce.

⚗Read those two accounts carefully: neither one is a rate

This is a good place to practise the distinction the whole part is about.

What is established: the effect exists; it is progressive; it is faster in humid air; it costs contrast; and both a 1911 manual and a modern practitioner describe it in compatible terms, which is strong agreement across a century.

What is not established, anywhere the course has read: a number. Not a half-life, not a percentage of speed gained per day, not a humidity threshold. “Very slow and uncertain” is a practitioner’s summary, not a measurement, and the course will not convert it into one.

So the statement this page makes is qualitative and says that it is: sensitised dichromated material has no useful shelf life, and the length of that non-shelf is not a figure the course can give you.

The mechanism the effect implies is worth stating even so. If dampness accelerates it, then water is a participant — which fits a chemistry in which the colloid must be mobile enough for chromium to bridge it, and which is also why a dried, desiccated layer is stable enough to sell.

The commercial consequence shows up in the way the material was sold. Cassell records carbon tissue being manufactured in two forms, sensitive and insensitive — the second with the dichromate left out, to be sensitised by the user shortly before use. That is the same commercial logic as an alternative-process kit sold as separate bottles, and for the same reason: the mixed, dried, sensitised article is the one with the clock on it. Set that beside a silver gelatin dry plate, which keeps for a year in a box, and you have measured the practical distance between the two families.

The choice of salt was a lever on the same problem. Cassell reports that ammonium dichromate has a stronger sensitising power and is more soluble than the potassium salt, gives richer pictures in carbon printing, and was said to be more than twice as sensitive to light — a claim the encyclopaedia attributes rather than asserts, and this page does the same.

One exposure, six second halves

The exposure is identical across the whole family: a negative in contact with a dichromated colloid, sunlight or ultraviolet, no visible image, and the print timed against an actinometer or a strip of printing-out paper because there is nothing to inspect. Everything that distinguishes the processes happens after it.

What the nineteenth century did with a hardened relief

1. 1Wash it, keep it, and the pigment in it is the picture — carbon printingWarm water removes the unhardened pigmented gelatin. The survivor is a continuous-tone image in whatever pigment was milled into the tissue, and no silver is anywhere in it. Poitevin 1855; made practical by Swan in 1864
2. 2Build it in coats with a brush — gum bichromateOne coating of gum, watercolour and dichromate carries a short scale, so the print is made in several coats in register, and the printer intervenes during development. The one member of the family in which the hand is part of the method
3. 3Cast from it under pressure — the WoodburytypePress the dried relief into soft lead and you have an intaglio mould whose depth tracks tone. Fill it with pigmented gelatin ink and print. The only fully continuous-tone photomechanical process ever made practical, on the Getty’s assessment
4. 4Ink it wet and print flat — collotypeHardened gelatin takes up less water than soft gelatin, and greasy ink goes where the water is not. Poitevin described it in the same 1855 patent; Tessié du Motay and Maréchal got halftones from it in 1865; Albert of Munich made the plates last
5. 5Etch through it — photogravureThe hardened layer is a resist of graded thickness, so the acid reaches the metal sooner where the layer is thin and bites deeper there. Klič combined a carbon resist with an aquatint grain on copper in 1879 and the art trade was his
6. 6Use it to protect metal, and throw the image away — the photoresistNiépce’s original purpose, the half-tone printing block, and, a century later, every integrated circuit ever made. The picture is not the product; the etched plate is

Six industries from one exposure. The branch point is not the chemistry, which is the same in every row, but what the relief is used for once the wash has finished.

Wall’s list of bichromate methods, quoted at the top of this page, is that diagram written out as an index entry in 1912. It is not a niche. For roughly half a century this family was how photographs reached the public.

The half-tone problem, and three answers to it

The single most instructive thing in this whole family is a failure, and the sequence of engineering answers to it. It is worth working through properly, because it is a mechanism problem rather than a recipe problem, and because it explains the otherwise baffling structure of carbon printing.

The failure. Poitevin’s and Pouncy’s early carbon prints came out harsh and lost their middle tones. Abbé Laborde diagnosed it in 1858, in a note to the French Photographic Society, and the diagnosis is exact enough to quote: in the sensitive film, however thin it may be, two distinct surfaces must be recognised, an outer and an inner which is in contact with the paper; the action of light commences on the outer surface; in the washing, therefore, the half-tones lose their hold on the paper and are washed away.

Why that happens is pure optics plus pure chemistry. The colloid layer absorbs the light that hardens it, so hardening is graded downward from the illuminated face. Under the darkest parts of the negative, only a skin at the top hardens. Cassell puts the geometry plainly: the film is so thick that even the strongest shadow does not penetrate right through, so a layer of soluble gelatin of varying thickness always underlies the insoluble image. Now put that in warm water. The water attacks from the edges and from below, dissolves the soluble underlayer, and the hardened skin above it — which is the half-tone — floats off intact. Cassell: any attempt to develop the film on its original paper would result in its floating right off as soon as the soluble gelatin commenced to dissolve.

The whole of carbon printing’s peculiar architecture exists to solve that one problem, and it was solved three different ways in the six years after Laborde named it.

Why carbon prints lost their half-tones, and the two shapes of the fix

1. 1Expose through the front — hardening is graded downward from the lit face, so every hardened zone hangs from the top surface and only the deepest shadow reaches the paper
2. 2Develop from the front — the failure — water dissolves the soluble gelatin underneath the hardened skin, and the half-tones float off whole. Laborde named the cause in 1858
3. 3Burnett, 1858 — expose through the back — anchors the hardened zone to the paper. The cost is a much longer exposure and the paper’s own texture printed into the image
4. 4Swan, 1864 — transfer, then develop from the back — squeegee the exposed layer face down onto a new support and strip the original paper, so the water meets the soluble side first. The cost is a lateral reversal and a second transfer to undo it

Drawn to show the relationships, not to scale: the tissue is a thin coating, not a slab, and the hardened zones grade continuously rather than in three steps. No quantity, time or temperature appears here or anywhere on this page.

Burnett, 1858 — turn the sheet over. Expose through the paper, so that the hardening starts at the face that is stuck to the support. His own reasoning, in the paper he read to the Photographic Society of London, is that it is only by printing this way that the unacted-upon portions can afterwards be removed by washing. The Getty calls the idea theoretically sound, and it is; it failed in practice for two reasons that are both about the paper. Light has to get through the sheet, so exposures became very long, and the structure of the sheet printed itself into the image.

Fargier, 1860 — varnish the front and strip the back. Coat the exposed layer with collodion varnish, take the original paper off in warm water, develop from behind, and attach a new support. It worked and gave full half-tones. It was also, in the Getty’s word, complicated, and Poitevin successfully opposed the patent on the grounds that he had detached films the same way in his own powder-carbon process.

Swan, 1864 — transfer first, then develop. Attach the exposed tissue face down to a support, strip the original paper in warm water, and wash from what is now the top. The most-hardened face is protected by the new support and the most-soluble part meets the water first, so the whole tonal scale develops. Swan’s other change was a plasticiser: adding a little sugar to the tissue gave a less brittle layer when it dried. Neither idea was new on its own — the Getty says as much — but the combination is what made the process an industry, and the carbon tissue that resulted was sold ready-coated, so that most photographers working the process bought their materials rather than making them.

Everything after that is materials engineering. Johnson, in 1869, found that briefly soaked tissue adheres to a waterproof surface by its own swelling with no adhesive at all, and in 1870 swapped soap for sugar as the plasticiser. Sawyer, in 1874, patented a flexible temporary support — paper, gelatin made insoluble with chrome alum, then shellac and borax — because on a rigid metal plate the thick dark areas contracted away from the thin light ones as they dried.

⚗Two structural facts that fall straight out of the geometry

Single transfer reverses the picture; double transfer un-reverses it. If the image is developed through the face that was against the negative, then that face ends up against the new support, and the print reads left-for-right. The trade lived with it for landscapes, used a second transfer where orientation carried meaning — the Getty names letters, a wedding-ring hand and familiar scenery — and avoided the whole question where a reversed negative was available.

The tissue has to be coated thicker than the deepest shadow it will ever have to render. Cassell states the rule directly, and the reason is the same geometry: a thin layer of soluble gelatin must remain between the insoluble shadow and the paper, or the shadow anchors itself to the support and the transfer will not lift. So the maximum density of a carbon print is decided at the coating machine, before any negative exists — which is not a sentence that can be written about any silver material.

Gum: the same chemistry with the printer’s hand in it

Gum bichromate is the member of the family that went the other way. Where carbon printing was industrialised, standardised and sold ready-made, gum stayed a hand process and became the pictorialists’ instrument precisely because of the properties that made it commercially useless.

The chemistry is identical: gum arabic in place of gelatin, watercolour pigment in place of milled carbon, a dichromate as the sensitiser, and light hardening the colloid in proportion to exposure. Development is a soak in cold water — and it is here that gum diverges from everything else in this course, because the printer intervenes during it. Cassell’s account is candid about this: the worker may lave, spray, sponge or brush, and by such means retains only such of the pigmented gum as is required for the rendering of his idea of the subject.

One coat carries a short tonal scale, so a gum print is usually built in several coats, each exposed and developed in turn, which raises the registration problem that Cassell also names honestly: even with a device to replace the sheet in position, the paper itself expands and contracts between printings. And the encyclopaedia’s closing sentence on the process is the most useful thing in the entry — that the very existence of these variable elements precludes the possibility, even if it were desirable, of laying down any hard and fast rules for working it.

That sentence is worth carrying out of this page for a reason that has nothing to do with gum. It is a period source saying, in 1911, that its own subject cannot be reduced to a recipe. Most of what this course teaches can be, and the fact that one process resists it tells you something real about where the control lives.

☠What a gum print leaves behind, and why the clearing bath does not fix it

Period practice finished a gum print with an alum bath to remove the yellow bichromate stain — the unreduced dichromate still sitting in the paper. That is a real and effective step, and it is why old gum prints are not yellow.

It is not, however, a decontamination. The Getty’s X-ray fluorescence work on carbon prints is unambiguous on the general point: washing and clearing procedures cannot remove all the chromium, and what remains is embedded in the image layer at a concentration roughly proportional to the tonality of the photograph. The dichromate that was reduced by light is not soluble dichromate any more; it is bound into the colloid, which is exactly what made the picture.

So a dichromated colloid print is a permanent record of its own chemistry, and a bath that removes the stain is removing the part that did not react.

Printing without a screen: the Woodburytype, collotype and photogravure

Three of the second halves in the diagram above solved the same commercial problem — how do you get a photograph onto a printing press? — and each solved it by exploiting a different property of the same hardened layer.

The Woodburytype used the relief as a mould. A thick dichromated gelatin slab, cast on glass and dried, was exposed under a negative, washed out in hot water to leave a positive relief thickest where most light fell, dried hard, and then pressed into a smooth lead plate under a hydraulic press — the Getty gives 35 MPa — to make an intaglio matrix whose depth tracked the tone. The matrix was filled with hot pigmented gelatin ink, a shellac-varnished receiving paper was pressed onto it, and after a few minutes the print was lifted off and trimmed. Because the ink layer is genuinely thicker in the shadows, the result is the only fully continuous-tone photomechanical process ever made practical in the Getty’s assessment: no screen, no dots, no threshold. It lost anyway, because it could not be adapted to rotary printing, and collotype and the letterpress halftone took the work by the end of the century.

Collotype used the swelling difference. This is the property Talbot noticed in 1852 and Poitevin patented in 1855: dampen an exposed dichromated colloid and the hardened parts take up less water than the soft parts, so greasy ink adheres where the light acted. Poitevin’s priority for this half is contested as well — the Britannica records that Paul Oreloth’s patent of 1854 already states that his designs were inked with printing ink before transfer to stone or zinc — and it credits the first successful half-tones from an inked gelatin film to Tessié du Motay and Maréchal of Metz in 1865. There is no relief to speak of and no screen at all — the tone is carried by a fine reticulation in the gelatin surface. The practical obstacle was durability, and Albert of Munich’s solution is one of the neatest pieces of reasoning in this whole family: rather than adding gum-resins, alum or tannin to harden the film, which coagulated the mass before it could be coated, he used light itself as the hardener, exposing the face of the layer that lay against the support. The plate had to become transparent for that to be possible, so he abandoned metal for glass, and the edition length went from fewer than a hundred impressions to over a thousand.

Photogravure used the relief as an etch resist. Klič’s 1879 process transferred a carbon tissue image onto a copper plate that had first been given an aquatint grain, developed it in warm water, and etched it in iron(III) chloride baths of varying strength. Where the resist is thin the mordant reaches the copper sooner and the cell bites deeper, so it holds more ink; where the resist is thick the cell stays shallow. The aquatint grain is what makes it printable at all: it breaks the surface into cells that can hold ink and support the doctor blade, which is a screen in function without being a screen in appearance. Klič went on to put the same logic on a cylinder, with a crossline screen replacing the grain, and invented rotogravure.

Eder spends three pages on how the man’s name should be spelled, and the digression is worth the space for what it teaches about records. The parish register at Arnau has Klitsch; the inventor himself signed his drawings, gravures and government contracts Klič after his father adopted the Czech form; and, because English colleagues could not pronounce it, he called himself Klietsch in England and that is what went onto his tombstone. Four sources, four spellings, one man, and only one of them is what the documents say. It is the same problem as a period formula attributed to three different people, and the same discipline answers it: say which document you are reading.

🕰The dichromate's second career, and the one nobody notices

Cassell’s entry for carbon tissue makes an observation that is easy to read past: in process work, it says, the photogravure process is solely worked with a carbon resist developed on the copper plate.

The carbon tissue that photographers bought to make permanent pigment prints and the carbon tissue that printing houses bought to make gravure cylinders were the same article. So the material that produced a few thousand exhibition prints also produced, for decades, most of the illustrated magazines, the catalogues, the art reproductions and the newspapers’ picture sections in the industrialised world.

The pigment process is remembered as a fine-art curiosity and was in fact an industrial commodity, and the ratio between those two uses is many orders of magnitude.

The objects: what they look like, and how they are told apart

These processes made things that are now in collections, in dealers’ boxes and in family houses. Being able to recognise one is the practical payoff of the chemistry above, because almost every identifying feature is a direct consequence of the mechanism.

Separating a pigment print from a silver print, and then from its own cousins

in order of how little equipment it needs

1. Raking light, unaided eye: is there relief?Both carbon prints and Woodburytypes carry a slight surface relief, most pronounced at a sharp boundary between light and dark, and the Getty’s viewing instruction for carbon is to look at nearly 180 degrees to the surface. Dark areas stand higher, and look glossier, because there is more gelatin there
2. The failure modes that are absentNo silver mirroring and no image fading, in either process, where the pigment is carbon or another stable inorganic. A print made with organic dyes can fade, and the Getty says so — the permanence belongs to the pigment, not to the process
3. The mount, and what the maker claimed on itPrinted or embossed permanence claims are an identification feature in their own right: "permanent photograph", "procédé au charbon", and — on a Ghent carte de visite in the Graphics Atlas — "charbon inaltérable", a photographer boasting that his prints will not do what everyone knew silver prints did
4. Above 25×: pigment particlesRandomly distributed pigment microparticles and larger clusters that manufacture could not eliminate, visible in the lighter areas of both carbon and Woodburytype. Brighter red or pink particles of a red-lake pigment appear in the warm-toned versions, added to imitate a gold-toned albumen print
5. Size, trim and edges: carbon or Woodburytype?Woodburytypes are trimmed flush on all four sides, because the gelatin ink smeared the margins, and the shear leaves a microscopic edge deformation. The Getty’s rule of thumb is that anything larger than 11 by 14 inches is a carbon print or a Stannotype rather than a Woodburytype proper, and that above 8 by 10 a Woodburytype is unlikely. Air-bubble white spots occur in Woodburytypes
6. X-ray fluorescence: no silver, and how much chromiumThe decisive analytical signature is the total absence of silver together with the presence of chromium. Carbon prints carry roughly five times the chromium of Woodburytypes, so it is the relative amount in the Dmax and not the mere presence that separates them — and the mounting board contributes most of the inorganic signal, so it has to be measured separately
7. Infrared spectroscopy: which binder, and which varnishGelatin shows Amide I at 1626 and Amide II at 1533 reciprocal centimetres, with the 1450 to 1300 region separating gelatin from albumen. A shellac layer in the Dmin points to a Woodburytype. A collodion-varnished carbon print can be mistaken for a collodion photograph, and the Amide II peak of the gelatin underneath is what gives it away

Every row is the chemistry showing itself in the object: relief because tone is thickness, no mirroring because there is no silver, chromium because the sensitiser is bound into the image, and gelatin because the image substance is the binder.

!Three honest limits on that analysis, which a reader should be able to state

X-ray fluorescence identifies the element, not the oxidation state. “Chromium is present at concentrations proportional to tonality” is a finding about chromium. It is not a finding about chromium(VI), and nothing on this page should be read as one. What species survives in a century-old print is a question the course cannot answer, and it has said so above.

Most of the inorganic signal may not be the picture. The Getty’s own analysis of a Woodburytype found that every element detected in the print was also present in the mounting board, and concluded that most of the signal originates there. That is why the atlas insists on measuring board, substrate, Dmin and Dmax separately, and it is a general lesson about analysing anything that is glued to something else.

A recommendation in the literature is not evidence that anybody followed it. The technical literature on the Woodburytype recommends small amounts of mercury(II) chloride to extend the shelf life of the gelatin ink. After examining several dozen prints, the Getty’s authors report that they have not been able to identify such a treatment. Period instruction and period practice are two different bodies of evidence, and this part of the course exists partly to keep them apart.

If you own one. The advice is the same as for every object in this part and is mostly a list of things not to do. Cool, dry, dark and flat; no water, no solvent, no adhesive tape; handle by the mount and not the surface. The gelatin layer of a carbon print or a Woodburytype lifts and cracks at the edges and in the thick shadow areas, and the Getty records cracking and lifting inside the image area as well in objects now more than a century old. A print that is already lifting is a conservator’s problem, not a home one. The chromium bound in the image layer is not a reason to avoid handling a framed or mounted print, and it is a reason not to sand, scrape, solvent-clean or otherwise disturb one.

Why the family mattered, and to whom

Three arguments carried it, and they are unusual in that all three were mostly correct.

Permanence, at the moment when it was the industry’s crisis. The generation that watched albumen prints yellow and fade had a genuine commercial emergency on its hands, and here was a process whose image substance was a pigment. The claim went onto the mounts: permanent photograph, procédé au charbon, and the Ghent photographer’s charbon inaltérable. The Graphics Atlas note on that object reads it exactly right — the tendency of silver-based photographs to fade was well known, and the photographer is boasting. He was largely entitled to. Carbon and Woodburytype prints made with stable inorganic pigments do not fade and do not mirror; what they can do is crack, lift, and suffer biologically at high temperature and humidity, because the binder is still gelatin.

Editions, and the printed page. The Autotype Company, founded on Swan’s patents in 1868, is the shape of the industry in one firm: carbon tissue in dozens of catalogued colours; a photomechanical department that added the Woodburytype in 1880; a Keramic department firing carbon prints into enamel; an Artistic Finish department putting them on opal glass and canvas at sizes up to eight feet; compass dials printed on mica during the Second World War; and an eventual absorption into a chemicals group in 2005. The Woodburytype’s own list of publications — Galerie Contemporaine, Thomson’s Street Life of London, The Theatre, Woodbury’s Treasure Spots of the World — is a list of the kind of book that made photography a public medium rather than a private one.

Control, which is the argument that kept gum alive. For the pictorialists, everything Cassell listed as an obstacle — the variability, the hand development, the multiple coats — was the attraction. A gum print is visibly made rather than taken, and it is the process in which a photographer’s decisions during development are as consequential as their decisions during exposure. That is why gum survived the collapse of every commercial dichromate process and is still printed today.

🕰How the family ended, and what the ending was not

It was not a chemical defeat. Nothing better at hardening a colloid arrived, and the processes still work.

Carbon printing was displaced commercially by materials that were easier, and then by ink-jet pigment printing, which reaches the same argument — a stable pigment image with no silver in it — by a route with no chromium in it at all. Commercial tissue manufacture collapsed to almost nothing: Hanfstaengl in Germany continued small quantities until about 1990, and by early 2007 a working practitioner could name a single United States supplier of monochrome tissue.

What did change was knowledge. The exposure limits and the classifications that now govern chromium(VI) arrived long after the processes did, and they arrived because of what was found out about lungs and skin. The retreat is a safety retreat wearing commercial clothes, and the next section is why.

The hazard that stops us

The chromium policy states the rule and carries the classification, the exposure limit and the argument for separating chromium(VI) from chromium(III); this page cites it rather than restating it. Three things belong here instead, because they are specific to this family rather than to the element.

The period name for it, and what the period got wrong

Cassell’s 1911 encyclopaedia has an entry headed Bichromate Disease, and it is one of the most instructive paragraphs in the corpus behind this part. It describes a skin disease affecting some workers who use potassium bichromate extensively, taking the form of small ulcers or an irritating rash. It gives the preventive as rubber gloves or finger-stalls. And for very severe cases it advises rubbing in a nitrate of mercury ointment.

Read what that paragraph knows and what it does not.

It knows the effect is real, occupational and dose-related, and it knows the control. Rubber gloves in 1911 is the right answer, and Wall’s dictionary of the previous year says the same thing about soaking the hands as in the carbon process, with the aggravation where there are cuts.

It gets the mechanism backwards in a way that matters. The entry says the disease occurs only when the skin is particularly sensitive — that is, it treats susceptibility as a pre-existing personal quirk. The modern classification says the opposite: potassium dichromate carries H317, may cause an allergic skin reaction, in 98.6 per cent of 491 notified reports, and H334, may cause allergy or asthma symptoms if inhaled, in 99.8 per cent. Sensitisation is something the exposure does to you, not something you brought with you, and once acquired it does not go away.

It has no idea about the rest of it. There is no cancer in that entry, no mutagenicity, no reproductive toxicity, no respiratory sensitisation, and no exposure limit — because none of those concepts, in the form we now use them, existed yet. The notified classification runs to twelve statements including H340, H350 and H360, each of the first two in 99.8 per cent of reports. Ammonium dichromate’s aggregated record has the same shape.

Those percentages were read on 6 September 2026, and the date is part of the claim. A notified classification is a live aggregate of what companies have told a regulator, not a fixed property of a substance: the report count moves, notifications are added and withdrawn, and a figure quoted without a date cannot be checked against the record it came from. The GHS and CLP reference explains how to read one.

And its treatment is a second hazard. A mercury ointment rubbed into ulcerated skin is the nineteenth century’s pharmacopoeia meeting the nineteenth century’s dermatology, and it is the clearest possible illustration of why a period source is evidence of what was believed and never of what is true.

⚠The material is still sold to photographers, and its own sheet says what it is

This is not a hazard that belongs to the past tense. A current supplier safety data sheet for ammonium dichromate, purified, revision G1 dated 20 August 2015, lists the recommended uses of the product and gives the first of them as “In photography”.

The same sheet, in section 2, classifies the material Carcinogenicity Category 1A — the category for a known human carcinogen — together with Germ cell mutagenicity 1B, Reproductive toxicity 1B, Respiratory sensitization 1, Skin sensitization 1, skin corrosion and serious eye damage both Category 1, acute toxicity Category 3 by mouth and Category 2 by inhalation, and specific target organ toxicity Category 1 on repeated exposure.

Those two facts sit on the same document: the product is sold for photography, and the product is a Category 1A carcinogen. Both are true, and a course that showed you only one of them would be misleading you either way round.

The waste, which does not end when the print is finished

Two separate streams come out of this chemistry and only one of them is obvious.

The bath and the wash water. The Environment Agency’s waste classification guidance defines a heavy metal, for the purposes of the List of Waste, as any compound of a listed set that includes chromium(VI) explicitly, so far as those compounds are classified as hazardous — which dichromates are. Its assessment route provides mandatory group entries for chromium compounds, to be used where no substance-specific entry exists. A dichromate sensitising bath, the wash water from a carbon development, and the alum clearing bath that took the stain out are all chromium-bearing, and none of them is a domestic drain’s business. GOV.UK’s household guidance routes hazardous waste from a home to the local authority’s collection service, and that service differs between authorities. This is a reading of guidance for England and Wales, it is dated, and your local regulations govern. The disposal caveat sets out what the course can and cannot say here.

And the print. This is the stream nobody counts. The Getty’s X-ray fluorescence work establishes that the chromium in a carbon print is photochemically produced, embedded in the image layer, and not removable by the washing and clearing the process uses. The picture is a chromium-containing object, permanently, in the same way that it is a gelatin-containing object. That is not a reason to fear a print in a frame, and it is a reason to understand that “the chemistry is finished” and “the chromium has gone” are different statements.

What the course could not source, stated so that nobody mistakes silence for safety

The mirrored occupational-hazard guide card for chromium(VI) has no extractable text. It is one of the known holes in this course’s research corpus, and the chromium policy records it. No exposure limit is quoted on this page from that card, and the figures the course does use — chromium(VI) compounds at 0.01 mg/m³ over eight hours, with the Carc and Sen notations and a biological monitoring guidance value — come from HSE’s EH40 and from the aggregated notified classifications, each attributed on the pages that carry them.

No mechanism of chromium(VI) carcinogenicity appears anywhere in this course, for the same reason no cross-link mechanism appears above: it has not been read in a source that meets the evidence standard. What is published is the classification, the limit and the notations.

EH40’s own caution is the one to end on: absence from a list does not indicate that a substance is without risk. Neither does absence from a course’s corpus.

The dichromate-free routes, at the strength of the evidence

The obvious question after all of that is whether the pigment processes can be run without chromium(VI). The honest answer has three parts and only the first is encouraging.

Substitutes exist and are named in the literature. Mike Ware, in an article about something else entirely — gas evolution during iron-process exposures — refers in passing to the Chibatype process for hardening pigmented colloids, which uses an ammonium ferric citrate sensitiser, and to the use of diazidostilbenes to photoharden pigmented colloids, where the photolysis evolves nitrogen. He notes, correctly, that dichromated colloids evolve no gas at all and so are not subject to the problem he is describing.

That is a mention, not a formulation. It establishes that the routes exist and that a chemist of Ware’s standing regards them as real. It does not give a composition, a working method, a hazard classification, a waste route, or a source of supply, and this course has read nothing that does.

So here is what would have to be true before this course taught one. A published, reproducible formulation from a source that meets the research standard. A hazard record for the sensitiser — a notified classification or a manufacturer’s safety data sheet, dated — rather than an absence of one, because an unclassified substance is an uncharacterised substance and not a mild one. A waste route that a domestic reader can actually use. And, ideally, an independent account of somebody working the process, so that the formulation is known to survive contact with a real coating rod.

☠What the course will not do in the meantime

It will not take a dichromated formula, swap in a different crosslinker, and publish the result as carbon printing. The chromium policy already rules on this and the reason is worth having in front of you: a substituted formulation is a new, untested process wearing a historical name, and presenting it otherwise would breach the rule against silently altering historical formulations and the rule against publishing anything the course has not sourced, in the same sentence.

Where the course has no dichromate-free route it can source, the process is study-only, and the page says the omission is a decision rather than an absence of evidence.

Level D here, Level C in the rubric, and the difference stated plainly

The safety classification is explicit about where chromium(VI) sits, and the wording is worth quoting exactly rather than paraphrasing. Among the criteria for Level C — specialist or supervised laboratory, the rubric lists:

Chromium(VI) compounds, formaldehyde and similar substances in any quantity.

And Level C’s own summary describes what that level assumes: facilities a home darkroom does not have — a fume cupboard, controlled waste streams, or supervision by someone trained in the specific hazard.

So by the rubric, supervised carbon printing is Level C work, and this page presents it at Level D. That is a decision of this course, not a rule of the rubric, and it is made on one ground. This course is written for a reader working at home. It cannot assume a fume cupboard, it cannot assume a licensed waste contractor, and it cannot assume a second competent person in the room. Level C exists here as a real classification with real pages behind it, and where the course can specify the controls it does. For chromium(VI) it cannot specify them to a domestic reader, so it gives no procedure to anyone and says so. That is what Level D means: not that nobody may do this, but that a page cannot assume the conditions under which anybody should.

!Where a reader who wants a pigment print should actually go

Carbon and gum printing are alive, and they are practised well. Not by people who ignored the hazard, but by people who have the extraction, the gloves, the waste route and the habit of using them. That is the population this course cannot assume you belong to, and the Level D policy is explicit that where institutions and practitioners work with a process under controlled conditions, the right move is to point at their published accounts rather than paraphrase them into something that reads like instructions.

Three routes are worth naming. A supervised workshop run by a practitioner who works the process regularly. A printmaking or photography department with a fume cupboard and a chemical waste contract, where these processes are often already taught. And, if and when the course can verify one, a dichromate-free route to the same result, on the terms set out above.

And one route the course does teach now. If what you want is a print whose image substance is not silver and whose permanence does not depend on silver chemistry, Part XXI’s cyanotype and Part XXV’s palladium are both in this course, with procedures, because their controls are ones a reader can assemble. Neither of them is a pigment process and neither gives you a relief. That is an honest trade rather than a substitution, and it is offered as one.

Recap

1. This is the other family of photochemistry, and it deletes three things silver depends on. No latent image, no development amplification, no fixer. Light changes the solubility of a colloid, a solvent removes what light did not act on, and the picture is the remainder — which also means that every photon has to do its own work, and the whole family is a printing-out chemistry with no gain.
2. Tone is a thickness, and that is why this family, not silver, produced the printed page. A relief can be cast from, etched through, inked wet or used as a resist. The Woodburytype, photogravure, collotype and the half-tone block are four different exploitations of one hardened layer.
3. The mechanism is chromium(VI) photoreduced in the presence of the colloid, and the course states it to there and stops. The half-reaction is tabulated and printed here; the other half is not, because no source read says what the colloid becomes. Which groups are bridged, and whether chromium(III) or a shorter-lived intermediate does the bridging, is a named gap rather than a simplification.
4. The sensitised material has no shelf, and that is a property of the chemistry rather than a nuisance. The dark reaction runs on, faster in damp air, costing contrast as it goes — described compatibly by a 1911 encyclopaedia and a modern practitioner, and quantified by neither.
5. The half-tone failure and its three fixes are the best mechanism lesson in the part. Hardening is graded downward from the lit face, so the half-tones sit on a soluble underlayer and float off in the wash. Burnett exposed through the back and paid in time and texture; Fargier varnished and stripped; Swan transferred first and paid in a lateral reversal. The architecture of carbon printing is that geometry made physical.
6. The objects announce themselves once you know the chemistry. Relief in raking light and gloss in the shadows because tone is thickness; no mirroring and no fading because there is no silver; pigment clusters under a microscope; and, under X-ray fluorescence, no silver at all together with chromium proportional to tone — about five times more of it in a carbon print than in a Woodburytype, which is what separates two processes that otherwise look identical.
7. The exclusion is about controls, and the page says which decision is whose. The rubric puts chromium(VI) at Level C in any quantity; this course presents it at Level D because it is written for a home darkroom and cannot assume the fume cupboard, the waste route or the second person that Level C means. And a period source that names bichromate disease, blames the worker’s skin for it, and prescribes a mercury ointment is the reason the course reads every historical hazard statement as evidence of belief rather than of fact.

Check your understanding

Question 1. A reader who has learned silver chemistry asks why a carbon print does not need a fixer. What is the correct reason?

• The warm-water development destroys the remaining dichromate, so there is nothing left to fix
• The alum clearing bath is the fixer under another name
• Nothing light-sensitive remains in the sheet in an insoluble form: the unhardened colloid and the unreduced dichromate are simply soluble, so water removes them, whereas an undeveloped silver halide is insoluble and stays light-sensitive until a fixer complexes it away
• The image is a pigment, and pigments are not light-sensitive, so no fixing step is possible

Show the answer and why

Answer: Nothing light-sensitive remains in the sheet in an insoluble form: the unhardened colloid and the unreduced dichromate are simply soluble, so water removes them, whereas an undeveloped silver halide is insoluble and stays light-sensitive until a fixer complexes it away

The purpose of a fixer is to remove something that will not wash out on its own and that goes on responding to light if it stays. In a dichromated colloid there is no such residue: what light did not act on is soluble, and Ponton said so in 1839 — to fix it, all that is required is careful immersion in water. Option one has the direction of the chemistry wrong, since the water removes the dichromate rather than destroying it; option two confuses a stain-clearing bath with a fixing one; and option four is a non-sequitur, because the pigment was never the sensitive component in the first place.

Question 2. Why must a carbon tissue be coated thicker than the deepest shadow it will ever be asked to render, and what does that imply about where the maximum density of the finished print is decided?

• To give the squeegee something to grip; the maximum density is decided by the exposure
• Because hardening is graded downward from the lit face, a layer of soluble gelatin must survive between the hardened shadow and the paper or the image cannot be transferred; so the maximum density is decided at the coating machine, before the negative exists
• Because thick coatings dry faster, and the maximum density is decided by the dichromate concentration
• To allow for shrinkage during drying; the maximum density is decided during development

Show the answer and why

Answer: Because hardening is graded downward from the lit face, a layer of soluble gelatin must survive between the hardened shadow and the paper or the image cannot be transferred; so the maximum density is decided at the coating machine, before the negative exists

Cassell states both halves. The film must be appreciably thicker than the depth of the strongest shadow so that a thin layer of soluble gelatine remains between the insoluble shadow and the paper — and if that layer is missing, the shadow is anchored to the original support and the transfer fails. The consequence is genuinely strange to anyone trained on silver: a carbon tissue has its Dmax built into it as a coating thickness, and no negative or exposure can exceed it. Dichromate concentration is a contrast variable, not a Dmax variable, and it is not printed anywhere in this course.

Question 3. X-ray fluorescence of a nineteenth-century pigment print finds no silver and a chromium signal that is strongest in the darkest areas. What has been established, and what has not?

• That the print is a carbon print rather than a Woodburytype, and that the chromium present is chromium(VI)
• That the object is a dichromated-colloid process rather than a silver one, and that chromium is present in proportion to tone — but not which oxidation state it is in, and not, without measuring the mounting board and comparing relative amounts, whether it is a carbon print or a Woodburytype
• That the print is unsafe to handle and should be sealed
• That the chromium was left by inadequate washing, which a rewash would remove

Show the answer and why

Answer: That the object is a dichromated-colloid process rather than a silver one, and that chromium is present in proportion to tone — but not which oxidation state it is in, and not, without measuring the mounting board and comparing relative amounts, whether it is a carbon print or a Woodburytype

Three separate limits have to be respected at once. The technique reports elements, not oxidation states, so a chromium peak is not a chromium(VI) finding. The Getty establishes that both carbon prints and Woodburytypes contain chromium and that the discriminator is the relative amount — roughly five times more in carbon — with the mounting board measured separately because most of the inorganic signal usually comes from it. And the chromium is not a washing failure: it is photochemically produced and bound into the image layer, which is precisely what made the picture, and the Getty says the process’s own washing and clearing cannot remove it.

Question 4. Cassell’s 1911 encyclopaedia says that bichromate disease "occurs only when the skin is particularly sensitive". Where does that sentence go wrong, and why does the error matter more than the missing cancer warning?

• It does not go wrong; modern classification agrees that only susceptible individuals are affected
• It treats susceptibility as something the worker brought with them, when sensitisation is something the exposure produces — so the sentence implicitly tells an unaffected worker that they are not at risk, which is the reader most likely to relax the control
• It is wrong only about the word "skin", since the real hazard is inhalation
• It understates the dose required, which is the only substantive error in the entry

Show the answer and why

Answer: It treats susceptibility as something the worker brought with them, when sensitisation is something the exposure produces — so the sentence implicitly tells an unaffected worker that they are not at risk, which is the reader most likely to relax the control

The missing carcinogenicity is an absence of knowledge that nobody in 1911 could have supplied. The sensitivity sentence is different: it is an active misdirection about causation, and it points the wrong way for exactly the reader who is still fine. Potassium dichromate carries H317 in 98.6 per cent of 491 notified reports and H334 in 99.8 per cent, and sensitisation, once acquired, is permanent. Note also that Cassell gets the control right — rubber gloves — while getting the mechanism wrong, which is a common shape in period sources and a reason to read the instruction and the explanation separately.

Question 5. This page prints one reaction equation and deliberately omits its counterpart. What is the omitted half, and why is the omission not a simplification?

• The oxidation of the colloid that supplies the six electrons; it is omitted because no source the course has read states what the colloid becomes, and writing a plausible equation would be inventing chemistry a reader could not audit
• The chromate-to-dichromate equilibrium; it is omitted because it is too elementary for this level
• The reduction of chromium(III) to chromium metal, omitted for safety reasons
• The hydrolysis of gelatin in warm water, omitted because it belongs to Part V

Show the answer and why

Answer: The oxidation of the colloid that supplies the six electrons; it is omitted because no source the course has read states what the colloid becomes, and writing a plausible equation would be inventing chemistry a reader could not audit

The tabulated half-reaction says where six electrons go. It does not say where they come from, and they come from the gelatin, gum or sizing around the chromium. No source in this course’s corpus states which groups are oxidised, into what, or in what proportion, so no second equation is written. This is the same discipline the page applies to the cross-link itself and to bitumen’s photochemistry: state the boundary of the evidence rather than smoothing over it, because an equation is exactly the kind of claim a reader will trust without checking.

Question 6. On what basis does this course place carbon printing at Level D when its own rubric places chromium(VI) at Level C?

• The rubric is out of date and Level C no longer exists
• Level D is assigned to any process a manufacturer stopped supplying
• The rubric’s Level C entry names chromium(VI) compounds in any quantity and assumes a fume cupboard, controlled waste and a trained supervisor; this course is written for a home darkroom and can assume none of those, so it gives no procedure to anyone — a decision of the course, stated as one, rather than a rule of the rubric
• Carbon printing is more hazardous than the other Level C processes, so it is promoted a level

Show the answer and why

Answer: The rubric’s Level C entry names chromium(VI) compounds in any quantity and assumes a fume cupboard, controlled waste and a trained supervisor; this course is written for a home darkroom and can assume none of those, so it gives no procedure to anyone — a decision of the course, stated as one, rather than a rule of the rubric

The rubric wording is quoted exactly on the page: "Chromium(VI) compounds, formaldehyde and similar substances in any quantity", under Level C, whose summary is facilities a home darkroom does not have. Nothing in the rubric puts carbon printing at Level D. What puts it there is the audience this course is written for, and saying so plainly is more useful than pretending the classification fell out of a table — because a reader who has a fume cupboard and a waste contract can now see exactly which assumption does not apply to them.

Check my answersClear