Contrast Control in Iron-Based Printing

Almost every published contrast control for these processes is a chemical addition, and almost none of them is where the contrast actually comes from. Mike Ware, who has spent a career measuring iron processes, puts the answer in one sentence at the end of the section where he catalogues the additions: “For pictorial purposes, it is strongly recommended that the negatives should be made correctly in the first place.”

That is not a counsel of perfection. It is the consequence of a fact you already know: these processes have a fixed exposure scale. There is no contrast filter, no grade, and — for the Van Dyke — no development stage to lengthen or shorten. What a printer has instead is one large control and four small ones, and this page ranks them in the order of how much they move the curve rather than in the order the literature usually presents them.

The controls, ranked

The five controls, ranked by how much they move the curve

1. 1The negative — the only control large enough to make an unprintable negative printable. Published requirements: about 1.8 to 2.4
2. 2The developer, in a kallitype only — documented for image colour; its effect on exposure scale is not published and is what the lab measures
3. 3The sensitiser — acid content raises argyrotype contrast by dissolving highlight silver; the iron-to-silver ratio has no published figure
4. 4Multiple coating — documented for maximum density in the shadows, not for contrast
5. 5A chemical restrainer — the largest chemical effect in the platinum literature, and in this family it means chromium(VI). Not used here

Drawn to rank, not to a measured scale — no source publishes the five effects on one axis, and putting numbers on the bars would be inventing them. The ordering is the course's, argued from the sources named in each row.

The rest of this page works down that list, and then spends its second half on the one the course will not hand over — because the argument for refusing it is a piece of teaching in its own right, and because a reader who does not understand what the dichromate was doing cannot judge the substitutes.

The negative, which is not a control at all but a specification

These processes have a fixed exposure scale, which is the whole reason the ranking above comes out as it does. Ware’s rule for the family, from the cyanotype monograph, inverts the habit of Parts XVIII and XIX: the contrast of the negative should always be adapted to the process, not the other way round.

The published requirements agree well enough to plan against, and they are all long:

Process	Required negative density range	Source’s own words
Van Dyke Brown	up to 1.85	“capable of an extremely long tonal range”
Kallitype	up to 1.85, or about log 1.8	King reaches it by developing film about 50 per cent longer than normal for a grade 2 paper
Argyrotype	2 to 2.4 in the ultraviolet	Ware, “as for my other siderotype processes”
New Cyanotype	at least 1.8, by overdeveloping 70 to 80 per cent	and up to about 2.6 with citric acid added
Classic cyanotype	about 0.9	three stops — the odd one out, and the one you met first

Two things follow. One negative made for the long scale serves the Van Dyke, the kallitype, the argyrotype and — from Part XXV — the platinum-palladium prints, which is exactly why Part XXI asked you to choose a standard negative and make it long. And a negative made for grade 2 enlarging paper, with a density range around 1.05, is not a soft negative for these processes: it is an unprintable one, and no chemical addition on this page will rescue it.

The three routes to a long negative are Part XXI’s and are named rather than re-taught: develop the film substantially longer than normal, use a staining developer whose stain adds ultraviolet density the eye does not see, or output a digital negative with a correction curve. Photographers’ Formulary’s New Kallitype sheet gives the same three in the same order and puts a figure on the third — a contrast boost of “10, 20, or even 30 percent” in software.

!Ultraviolet density is not visual density, and this is where that bites

Every figure in the table above is a density range, and not all of them are measured in the same light. Ware states his argyrotype requirement explicitly as “in the UV”; the kit sheets do not say.

A negative’s ultraviolet density and its visual density are different numbers, and the difference is large for a stained negative and for an ink-jet negative, because the ink’s absorption in the near ultraviolet has nothing to do with its appearance. Ware makes the point in his general alternative-printing notes: silver-gelatin negatives have a neutral density that does not vary much with wavelength, and ink-jet negatives do not, so an ordinary visible-light densitometer says nothing about how one behaves under a UV lamp.

The consequence for this page: a negative you measured at 1.8 visually may be anything under the lamp, and the only measurement that settles it is a step tablet exposed through the process. That is the last section of this page, and it is why it is the last section.

The paper, its sizing and its humidity

The second row of the ranking is where the design of this part expected humidity to sit, and the evidence does not put it there for the processes this part teaches. The section is worth having anyway, because the paper genuinely does move the curve — by a different route.

Sizing changes the image, and there is a sourced mechanism. Photographers’ Formulary state the observation on both kallitype sheets: “Prints made with arrowroot starch-sized-paper will have a brown color while those sized with gelatin will have a blue tone.” Ware supplies the mechanism from the platinum work, and it is about particle size: gelatin, like a surfactant, may interact by “‘protecting’ nanoparticle metals, favouring formation of smaller particles and therefore warmer colours”, and it “seriously inhibits the precipitation of platinum”. Whether that changes contrast as well as colour is not published, and the honest reading is that sizing is a colour and maximum-density variable with an unmeasured effect on scale.

Absorbency changes how much sensitiser is in play. A heavily sized sheet holds the sensitiser at the surface; an absorbent one takes it in. Photographers’ Formulary tie that to the clearing test rather than to contrast — a print taking more than ten minutes to clear is on paper too absorbent to use unsized — and King’s four-to-five-minute criterion is the same test with a stricter limit.

⁇Humidity: a real contrast control, for processes that are not the ones in this part

Where the evidence is. Ware’s Platinomicon tabulates exposure scale against relative humidity for the print-out platinum and palladium sensitisers, and the effect is substantial: the scale lengthens with humidity in every column, so damp paper prints softer, and the optimum working band is 50 to 80 per cent. His argyrotype notes make humidity a deliberate colour control, and include glycerol as a humectant for that purpose. His own siderotype record sheet — a blank one-page form he hands out at workshops — has fields for relative humidity at coating, three drying rows including one headed Humidify, and a processing column headed Steam. That is a practitioner building humidity into the record because it is one of his variables.

Why it does not transfer to this part. Every one of those is a printing-out process, in which the reduction happens inside the coating and the water in the fibres decides how far the ions can travel and how large the metal particles can grow. A kallitype makes its image in a tray with as much water as it wants. No kallitype source read for this course treats humidity as anything but a factor in drying time, and the course will not transfer a requirement between processes without evidence.

Where it might transfer, and this is a real open question. The Van Dyke is a printing-out process, its image is nanoparticle silver, and Ware’s mechanism for humidity and particle size is stated for silver as well as for the platinum metals. No Van Dyke source read for this course mentions humidity at all — but the argument that it should matter is the same argument that works for the argyrotype, which is the same chemistry with a different silver salt. That is a hypothesis with a cheap test, and it is one of the experiments at the foot of this page.

What the course does: record relative humidity at coating, drying and exposure on every page in this part; control it on none of them; and say which of those two things you are doing.

The developer, in a kallitype

The third row, and the one where a printer with a bath in front of them has something to change.

What the sources establish is colour, and they establish it cleanly. Photographers’ Formulary’s rule covers all three of their developers with one sentence in each direction: more Rochelle salt gives more sepia, more borax gives more blackness. Wall’s 1912 dictionary records four period developers giving black, sepia, warm maroon and purple. Bostick & Sullivan say only that “different developers will change the color tone of the print”.

What they do not establish is contrast. Not one of the four kallitype sources read for this course attributes an exposure-scale difference to the choice among sodium citrate, Rochelle salt and borax. Every time contrast appears in a kallitype developer discussion, the thing doing the work is a dichromate addition. That gap is real and this page will not paper over it, which is why the kallitype lab asks for a step wedge in all three prints: it is a measurement nobody has published.

Two things about the developer genuinely are contrast controls, and both are about its state rather than its identity.

Temperature is written into two of the three recipes. The black-tone and brown-tone developers are specified warm, “around 38 °C/100 °F or higher”; the sepia one is specified at room temperature for twice the time. A bath that has drifted ten degrees over three prints is a different developer, and whether the difference shows as colour, as speed or as scale is exactly the kind of thing a step wedge answers and an argument does not.

And the accumulated iron load is a control you did not choose. King’s warning is about clearing — unreplenished developer accumulates iron(II) and makes the print progressively harder to clear — and the consequence for contrast follows from it: a print whose masked border will not clear has a stain in its highlights, and a stain in the highlights is a loss of contrast measured from the paper base. An exhausted developer flattens prints from the light end, which looks like a contrast problem and is really a clearing problem. The distinction matters because the remedy is different.

The sensitiser

The fourth row is thin, and it is worth being precise about how thin.

Acid raises contrast in the argyrotype, with a stated mechanism. Ware’s sensitiser carries a deliberate 20 per cent excess of sulfamic acid over what the silver oxide needs, bringing it to about pH 3.5. To make a more contrasty version, “dissolve an extra 1 g Sulphamic Acid in 100 cc sensitizer”, and by mixing the two “the contrast can be ‘fine-tuned’”. His mechanism is one clause long and it is the whole of the argument: the extra acid “tends to dissolve silver in the highlights.”

Read that carefully, because it is a different kind of control from a restrainer. A restrainer stops silver being made; this dissolves silver that was made. Both raise contrast by emptying the light values, and the aesthetic cost is the same — you are buying separation with highlight detail.

⁇Acid raises contrast in one process and lowers it in another

Ware publishes both of these, and a reader who meets them a week apart will think one is a misprint.

Argyrotype: extra sulfamic acid raises contrast, because it dissolves silver in the highlights.

New Cyanotype: “The contrast of the sensitizer can be lessened by adding citric acid, so that it can even accommodate a negative density range of 2.6 or so.” Acid lowers contrast.

They are not in conflict, because they are not the same mechanism. The argyrotype’s acid acts on the image substance, which is silver and is soluble in the wrong conditions. Prussian blue is not, and the New Cyanotype’s citric acid is acting on the iron chemistry, not on the pigment. Ware states the mechanism for the first and not for the second, so this page reports both effects and offers no mechanism for the cyanotype case.

The lesson to take away is a general one about this literature. “Add acid” is not a contrast instruction; it is an instruction whose direction depends on what the image is made of. A practice transferred from one siderotype to another on the strength of the word acid is a practice with no reasoning behind it.

The iron-to-silver ratio has no published figure. Photographers’ Formulary’s New Kallitype sheet is the only source that raises it: “Most research shows that the best print density will result from our recommendation of equal proportions. Some studies show that slightly better densities can be achieved with slightly more silver nitrate solution.” They name no proportion and cite no study, and the claim is about density rather than contrast. This course therefore publishes no ratio other than 1:1, and treats the variation as an experiment to be run rather than a control to be used.

And one practitioner’s report, marked as one. Wynn White found that adding more tartaric acid to a Van Dyke sensitiser “seemed to increase contrast slightly and move the image color to a more neutral gray but then graininess became a problem”, while more silver nitrate and more ferric ammonium citrate “didn’t have much effect”. That is consistent with Ware’s argyrotype mechanism — more acid, less silver in the highlights, more contrast, smaller particles — and it is one printer’s observation on his own papers, not a measurement.

Multiple coating

The fifth row, and the sources are clearer than you might expect about what it does and does not buy.

What is documented is maximum density. Wynn White: “I have tried single coating Vandyke prints but have found the dark areas of the print to be very weak. Double coating is a must.” Photographers’ Formulary’s New Kallitype sheet offers a second coat as a remedy for visible brush strokes or thin spots. Neither claims a change in contrast.

Why density and contrast are not the same thing here, and this is the general point the course keeps returning to: contrast is the slope between two densities, and adding more image substance in the shadows raises the top of the curve. If the highlights are unchanged, the scale lengthens and the slope rises; if the second coat also builds in the light values, it does not. Which happens depends on where the second coat’s silver ends up, and no source read for this course has measured it.

So the honest statement is: double coating is a documented remedy for weak shadows and an undocumented contrast control. Run it against a step wedge and you will know for your paper — which is more than the literature knows.

The restrainer principle, which is the mechanism behind every chemical contrast agent here

Now the fifth row of the ranking, and the reason it needs a whole section.

The principle, stated for the platinum case by Ware. Potassium chlorate “is a strong oxidising agent, and its effect is to reoxidise some of the iron(II) photoproduct, thereby making it unavailable for reducing platinum(II) and so truncating the exposure scale of the process, providing a more distinct tonal separation in the final image.”

6 Fe²⁺ + ClO₃⁻ + 6 H⁺ → 6 Fe³⁺ + Cl⁻ + 3 H₂O

A strong oxidising agent puts the photoproduct back where it started, before it can reduce any metal

Why re-oxidising some of the iron(II) raises contrast

1. 1In a weakly exposed area, light made a little iron(II)A highlight is a region where only a few photons arrived, so only a few iron(III) centres were reduced
2. 2The oxidising agent removes a fixed amount of iron(II) everywhereIt does not know which part of the sheet it is in. It takes the same quantity out of a highlight and out of a shadow
3. 3In the highlight, that is most of what was thereThe little iron(II) the exposure made is gone before it can reduce any silver, and the highlight prints as paper
4. 4In the shadow, it is a small fractionPlenty of iron(II) survives, and the shadow prints nearly as dark as it would have
5. 5The result is a shorter exposure scaleThe light end has been cut off. That is a contrast increase, and it is bought with highlight detail rather than earned

The mechanism Ware states for potassium chlorate in the platinotype. It is a threshold effect, which is why a small quantity has a large consequence and why too much empties the highlights entirely.

That is a threshold mechanism, and it explains three things the practical sheets record without explaining. It explains why the quantities are tiny and why a small excess is disastrous — King’s warning that too much dichromate makes printing times increase considerably and the image “take on a granular look”. It explains why the additive costs exposure — Wynn White reports that one drop of 3 per cent dichromate in 12 drops of Van Dyke sensitiser means the exposure “must be nearly doubled”, because you are throwing away photoproduct and have to make more of it. And it explains what Ware calls the aesthetic objection: “the ‘false sparkle’ of lost gradation in the empty highlights”.

⚗How much of that mechanism is sourced for the dichromate, and how much is the course's

Sourced, for chlorate, by Ware, in the words quoted above: strong oxidising agent, re-oxidises some of the iron(II), truncates the exposure scale.

Sourced, for dichromate, as a classification rather than a mechanism. Ware groups it with chlorate and hexachloroplatinate(IV) as “contrast-enhancing agents” and describes Willis & Clements’ recommendation of “a very small amount of the strong oxidising agent potassium dichromate” added to the developer bath. He does not restate the re-oxidation mechanism for it.

Sourced, as a period description, by Wall in 1924: Thomson’s kallitype developer takes 0.02 to 0.8 per cent potassium bichromate “which keeps the whites pure and acts as a restrainer”. Keeps the whites pure is a description of exactly the threshold effect above, written by somebody who had no redox potentials to reason with.

The course’s own step, and it is marked as an inference rather than a finding: dichromate is a strong oxidising agent added at the same point in the same processes for the same stated purpose as chlorate, so the course reads it as acting by the same mechanism. What no source read for this course establishes is whether the dichromate also acts on the silver or on the paper, and a reader should treat the mechanism above as well-founded reasoning rather than as a measurement.

One thing Ware does establish about the dichromate specifically, and it is practical. Potassium dichromate “reacts readily with oxalic acid, being reduced to oxalato-chromium(III) complexes, so this additive will not be stable in an oxalate developer indefinitely, and will lose effectiveness, especially at low pH.” A dichromated oxalate developer is losing its contrast agent while it sits — which is another reason the effect is hard to reproduce and another reason the literature about it is inconsistent.

Potassium dichromate: taught in full, and not given

Every kit in this part except one ships a dichromate, and every practical source in this part gives a procedure for it. This course gives none, and the reasoning is worth setting out completely, because a refusal without an argument is just a prohibition.

What is used, and how much. The quantities are small and they are consistent across four independent sheets.

Source	Process	Where it goes	How much
Sandy King	Kallitype	The developer	1 to 16 mL of a 5 per cent potassium dichromate solution per litre, taking the printable negative range from about 1.2 to about 2.2
Photographers’ Formulary 07-0070	Kallitype	The developer	5 to 20 drops of a 10 per cent solution per 500 mL of mixed developer
Photographers’ Formulary 07-0080	Van Dyke	The first wash	9 to 10 drops of a 10 per cent solution in 500 mL, worth “about the loss of 1 step of a Kodak No 2 Step Table”
Bostick & Sullivan	Kallitype	The sensitiser	A 5 per cent ammonium dichromate “contrast booster”, by small additions, diluted further if one drop is too strong
Wynn White	Van Dyke	The sensitiser	One drop of a 1 to 5 per cent solution, with exposure nearly doubled at 3 per cent
Wall, 1924 (Thomson)	Kallitype	The developer	0.02 to 0.8 per cent potassium bichromate
Ware	New Cyanotype	The sensitiser	0.1 g of ammonium dichromate per 100 cc — optional, and omitting it diminishes both contrast and shelf life

Note the disagreement about where it goes, because it is instructive. Photographers’ Formulary print the instruction in their own emphasis on both kallitype and Van Dyke sheets: add it to the developer, not to the sensitiser. Bostick & Sullivan and Wynn White add it to the sensitiser. Ware’s New Cyanotype has it in the sensitiser, where it doubles as a preservative. Two of those cannot both be the best place, and no source read for this course adjudicates.

Why it works so well is the threshold mechanism above, plus one property no substitute matches: dichromate is a strong enough oxidant to act at concentrations of a few hundredths of a per cent, so a drop moves the curve and the sensitiser is otherwise unchanged.

☠Why this course gives no dichromate procedure, at any level

The ruling. The course’s chromium policy was settled on 4 September 2026 and is stricter than a level assignment: chromium(VI) compounds are never used, at any level, anywhere in this course. The chemistry, the history and the reason it works are taught in full — this page is that teaching — and no page instructs a reader to weigh, dissolve, apply or dispose of a dichromate. Part XXVI carries the historical processes that depend on it, under the Level D policy.

The criteria that produce that decision, from the classification rubric. The rubric takes a procedure’s level from the highest hazard criterion any of its substances meets, read from a named, dated source. Potassium dichromate’s notified classification, aggregated from 491 reports across 19 ECHA notifications, is signal word Danger with six pictograms and twelve hazard statements. Four of them decide this on their own:

• H350, may cause cancer — 99.8 per cent of reports.
• H340, may cause genetic defects — 99.8 per cent.
• H334, may cause allergy or asthma symptoms or breathing difficulties if inhaled — 99.8 per cent.
• H317, may cause an allergic skin reaction — 98.6 per cent.

with H301, H312 and H330 for acute toxicity by all three routes, H314, H360, H372 and H410 alongside them. Ammonium dichromate’s record is the same shape, with H301, H312, H314, H317, H330, H334, H340, H350 and H372 each at 100 per cent of 194 reports.

And the exposure limit says what control actually looks like. HSE’s EH40 sets chromium(VI) compounds at 0.01 mg/m³ as chromium over eight hours, annotated Carc, Sen and BMGV — that last meaning there is a biological monitoring guidance value, which is to say that the recognised way of knowing whether the control worked is to test the worker’s urine after the shift. A domestic darkroom does not do that. The limit was tightened in the 2020 revision, which is a useful reminder that a figure in an older book is not the figure in force.

Two smaller reasons, which would not decide it alone. The sensitisation statements matter particularly for a substance a hobbyist meets repeatedly in small amounts: sensitisation is a cumulative, individual response, and once acquired it does not go away. And the disposal instructions printed on the kit sheets — wash the solid and its solutions down a drain with copious water — are ones this course would have to contradict on the same page it published the procedure.

What to do if your kit contains one. Leave it sealed. Do not follow the sheet’s disposal instruction. Take it, in its original labelled container, to the route your local hazardous waste guidance specifies, under the general chemical waste SOP and the unlabelled or unknown container SOP if the label has gone. Check your local regulations; they govern.

The alternatives, each assessed rather than recommended

Four substitutes circulate. Here is what the evidence actually supports for each, and what it costs.

Chlorate: documented, but for the platinum processes

What is established. Ware’s account of potassium chlorate is the fullest mechanism statement for any contrast agent in the siderotype family, and it is where this page’s restrainer principle comes from. Pizzighelli and Hübl built an entire drop-counting system on it in 1882, three sensitiser solutions deep, and it “has become standard practice today with many users”.

What is not. No source read for this course documents a chlorate restrainer in a kallitype, a Van Dyke or an argyrotype. It belongs to the platinum literature. Whether it transfers is a reasonable hypothesis and not a published fact.

What it costs. Two things, and the first is the one people forget. Ware records that chlorate “can cause a deterioration in image quality and an increase in image graininess”, and calls the empty highlights aesthetically unacceptable for the platinotype ethos. And potassium chlorate itself is classified Danger across 262 ECHA reports with H271 — may cause fire or explosion, strong oxidiser — at 99.2 per cent, plus H332 at 93.9, H302 at 84 and H411 at 93.9. Its redox potential is +1.45 V, which is what makes it work and also what makes it a fire hazard in contact with organic material. That is a substitution of a serious physical hazard for a serious health one, and this course does not present it as the safe option.

Hydrogen peroxide: not documented at all

The design for this page asked whether hydrogen peroxide additions to the sensitiser or the developer are documented and what evidence supports them. The answer is that no source read for this course documents any such addition, in any iron-silver process.

Where hydrogen peroxide does appear in this corpus, it is doing two completely different jobs. In cyanotype it is a post-processing bath — Bostick & Sullivan’s optional 100 mL of 3 per cent in 900 mL for 60 to 90 seconds — which re-oxidises Prussian white to Prussian blue and brings back density the wash reversed. And in one route to making ferric oxalate it is the oxidant that turns precipitated iron(II) oxalate into the iron(III) salt.

The chemistry that a proposal would rest on is real: peroxide is an oxidising agent, so it could in principle re-oxidise iron(II) exactly as chlorate does. But there is a difference between a mechanism that would work and a practice somebody has tested, and this page will not blur it. If you find a published account, it belongs in the record; until then, an untested oxidant added to a sensitiser is an experiment, and it should be run as one with a step tablet rather than adopted as a control.

Changing the iron-to-silver ratio: no figure exists

Covered above. One source raises it, names no proportion, cites no study, and describes the effect as density rather than contrast. Treat it as an experiment.

Changing the developer: real, and its size is unmeasured

Covered above. The colour effect is well documented and the contrast effect is not published for the three classical baths.

✕Why some published contrast controls should be distrusted, and how to tell which

Several claims in this area circulate as community practice with nothing behind them. Three tests will separate the ones worth trying from the ones worth ignoring.

Does the source state a mechanism, or only an effect? Ware states one for chlorate and for the argyrotype’s acid; he states none for the New Cyanotype’s citric acid, and says so by omission rather than by hedging. A claim with a mechanism can be predicted from and tested; a claim without one can only be repeated.

Does the source give a quantity, and does that quantity agree with anybody else’s? The dichromate quantities in the table above are consistent across four sheets, which is why the course is confident about what was done even while refusing to do it. Compare that with the iron-to-silver ratio, where one sheet says “some studies show” and names neither a number nor a study.

Is the claimed effect on contrast, or on something that looks like contrast? This is the one that catches people. A deeper maximum density is not more contrast. A cleaner highlight from better clearing is not more contrast. A print that looks snappier because the developer changed its colour is not more contrast. Contrast is a slope, and only a step tablet measures a slope.

And one specific warning that belongs here rather than in a footnote: metal additives are not contrast controls and they carry a cost the sources state plainly. King documents gold, platinum, palladium and mercury(II) chloride added to a kallitype sensitiser to modify colour and tonal range, and adds: “Unfortunately the employment of metal has one important negative effect. The image is more likely to stain and the print will be much more difficult to clear.” In a family whose characteristic failure is inadequate clearing, that is a large cost for an aesthetic change — and the mercury procedure is one this course does not reproduce at all.

Contrast control at the negative rather than in the tray

If the ranking at the top of this page is right, the largest lever is upstream of the darkroom, and there are two moves worth knowing that Part XXI’s negative lesson introduces and this part uses.

Make the negative for the process, and make it twice if the processes differ. A negative at a density range of 2.2 serves the New Cyanotype, the argyrotype, the kallitype and the platinum-palladium prints; the classic cyanotype needs a second, much shorter one. Ware’s point about composite printing is the practical form of this: one long negative serves nearly the whole cluster, and compromising it to suit the one short-scale process costs you the comparison that the cluster exists for.

Use the difference between ultraviolet and visual density deliberately. A staining developer builds an image stain alongside the silver, and that stain absorbs in the near ultraviolet far more than it absorbs visually. To a densitometer reading visual density the negative looks moderate; to the sensitiser under a UV lamp it is contrasty. Part VIII’s staining and tanning lesson owns the chemistry, and the two practical consequences here are that a stained negative can be a long-scale negative that does not look like one, and that you cannot calibrate it with an ordinary densitometer — which brings this page to its last section.

The measurement that settles any of this

Everything above is a claim about the shape of a curve. There is exactly one honest way to test such a claim, and it costs one sheet.

Print a step tablet through the process, before and after the change. A transmission step tablet has a known density difference between steps — 0.15 for a 21-step Stouffer, 0.10 for a 31-step — so the print of it is a direct read-out of the process’s exposure scale, in the same units as a negative’s density range.

🧮What to read, and what counts as a difference

Two numbers, off each print. The first step distinguishable from paper white, and the first step that has reached maximum density. The exposure scale is the difference between them, times the tablet’s step interval.

ES = (step at D-max − step at first tone) × step interval

Exposure scale from a step tablet

Then the discipline, which matters more than the arithmetic. One step of difference between two prints is at the edge of what a single pair can support: papers vary, coatings vary, and a reflection densitometer on a matte, textured, binder-free surface is not a precision instrument. Two steps is a finding. One step is “no difference I can distinguish.” That is Part XIII’s rule applied to a process where nobody has published the numbers you are producing.

And hold everything else still. One coating, cut in two after exposure; one exposure; one negative; both halves processed in the same session. Ware’s own test target for exactly this purpose is eight identical step tablets on one sheet, so that the sheet can be divided and processed differently and the comparison is within a single coating.

One caution about the tablet itself. Ware notes that a standard photographic step tablet has a density range of about 3, which always exceeds the exposure scale of these processes — no more than about 2.4 for the platinum ones — so the full tonal range should always be obtainable from it. That is what makes it a good instrument here. It is also a silver step tablet, so its ultraviolet density and its visual density are close; an ink-jet substitute is not, and would be measuring itself as much as the process.

How this connects, in both directions

Back to Part XXI. The same question arose there and was settled the same way. Ware’s New Cyanotype takes 0.1 g of ammonium dichromate per 100 cc of sensitiser, and he marks it optional: omitting it “diminishes the contrast and shelf-life of the sensitizer” and the process still works. The course therefore publishes the New Cyanotype without it and classifies the formula on its oxalate and its acid instead. His improved Classic sensitiser is a different case — there the dichromate is not optional, Ware’s own comment is that anyone prepared to go to that trouble would do better with the New Cyanotype, and the course gives no procedure for it at all.

Forward to Part XXV. The same problem, a different answer, and the difference is worth anticipating. Platinum printers have a contrast agent that is not a chromium compound: sodium hexachloroplatinate(IV), which Ware describes as a milder oxidising agent recommended in the nineteenth century, still sold, and “said to cause less image deterioration than potassium chlorate”. It is also, in his parenthesis, sold “at a very high price, of course” — it is a platinum salt — and there are conflicting opinions about whether it works with palladium, with one authority maintaining it works with palladium and not at all with platinum. That is the shape of the trade in this family: the contrast agents that are chemically comfortable are expensive, and the cheap one is a carcinogen.

And Ware’s own conclusion, which is where this page began. “With the availability today of better controls for modern negative-making, especially by digital means, contrast-enhancing agents, such as chlorate, hexachloroplatinate(IV) or dichromate, become unnecessary when a correctly calibrated negative is made. In the present work, to avoid introducing yet another variable, the use of these undesirable image-degrading agents will be avoided.”

Alternative route

This page needs no facility, and a reader with no wet space can do all of it as reading and arithmetic. What such a reader should do instead of the measurement is the exercise the last section describes on paper: take the published exposure scales in the first table, take the density range of a negative you already have, and work out which of the processes in this cluster it could print on. That calculation is the whole argument of the page and it needs nothing but a densitometer reading you may already own.

Without a densitometer but with a wet bench, the step-tablet test still works and gives you step numbers rather than densities. Count the steps; the interval is printed on the tablet; the arithmetic is the same. What you lose is the maximum density, which is a separate measurement.

Without a step tablet, the honest substitute is a contact-printed one: expose a strip of sensitised paper in timed increments a stop apart, process it, and use the resulting print as a relative scale. It will not give you an exposure scale in log-exposure units, and it will tell you whether a change moved the highlights, the shadows or both — which is most of what you want to know.

Recap

The negative is the control and everything else is an adjustment. These processes have a fixed exposure scale of roughly 1.8 to 2.4, and a negative made for grade 2 enlarging paper is not soft, it is unprintable here.

The developer changes a kallitype’s colour, certainly; its scale, unmeasured. More Rochelle salt is sepia, more borax is black, two of the three are specified warm — and nobody has published their exposure scales, which is why the lab measures them.

Humidity is a genuine contrast control for the print-out platinum and palladium processes and is not documented for the kallitype. Record it; do not transfer it.

Every chemical contrast agent in this family works by re-oxidising iron(II) before it can reduce the metal. It is a threshold effect, which is why the doses are tiny, why they cost exposure, and why a small excess empties the highlights.

The traditional agent is a chromium(VI) compound and this course will not give it as a procedure. Potassium dichromate carries H350, H340, H334 and H317 in essentially every ECHA report, and a workplace limit of 0.01 mg/m³ whose recognised verification is testing the worker. The chemistry is taught here in full and no page gives a quantity as a working instruction.

The substitutes are weaker than their reputations. Chlorate is documented for platinum and not for these processes, and brings a fire hazard and graininess of its own. Hydrogen peroxide is not documented as a contrast agent anywhere in this corpus. The iron-to-silver ratio has no published figure. Multiple coating is documented for density, not for scale.

And one step tablet settles more than the whole literature. Two numbers, one coating cut in half, and two steps of difference before you call it a finding.

Check your understanding

Question 1. By what mechanism does a small amount of a strong oxidising agent raise contrast in an iron process, and which end of the scale does it act on?

• It oxidises the image metal in the shadows, lowering maximum density and steepening the curve
• It re-oxidises some of the iron(II) photoproduct before it can reduce the metal salt; because it removes the same amount everywhere, it takes nearly all of a highlight's iron(II) and only a fraction of a shadow's, so it truncates the scale at the light end
• It hardens the sizing, so the sensitiser sits nearer the surface and prints more contrastily
• It lowers the pH, which dissolves silver in the highlights

Show the answer and why

Answer: It re-oxidises some of the iron(II) photoproduct before it can reduce the metal salt; because it removes the same amount everywhere, it takes nearly all of a highlight's iron(II) and only a fraction of a shadow's, so it truncates the scale at the light end

This is Ware's statement for potassium chlorate in the platinotype, and it is a threshold effect: a fixed quantity of oxidant removed from a highlight that had very little iron(II) is most of what was there, while the same quantity removed from a shadow is a small fraction. That explains the tiny doses, the large exposure increase - Wynn White found one drop of 3 per cent dichromate nearly doubled his Van Dyke exposure - and the "false sparkle" of lost gradation Ware objects to. The last option is a real mechanism but a different one: it is what extra sulfamic acid does in an argyrotype, acting on the silver rather than on the iron.

Question 2. Your negative has a visual density range of 1.1 and your process needs about 1.8. What are the three changes, in order?

• Add a restrainer to the sensitiser, then to the developer, then change paper
• Remake the negative for the process - longer development, a staining developer, or a digital negative with a correction curve - and only then consider the developer and the sensitiser; the restrainer is last and, in this course, not available
• Double-coat the paper, warm the developer, then add dichromate
• Change to a shorter-scale process, since a negative of 1.1 cannot be printed at all

Show the answer and why

Answer: Remake the negative for the process - longer development, a staining developer, or a digital negative with a correction curve - and only then consider the developer and the sensitiser; the restrainer is last and, in this course, not available

A gap of 0.7 in density range is far larger than anything below the negative in the ranking can close: King's full dichromate series takes the printable range from about 1.2 to 2.2, which is the biggest chemical effect published for this family, and it is a chromium(VI) procedure. Photographers' Formulary give the same three negative routes in the same order, including a software contrast boost of 10 to 30 per cent. The last option is not quite wrong - a classic cyanotype at an exposure scale of about 0.9 really would print that negative - but it changes the picture rather than solving the problem.

Question 3. Which of these does the literature read for this course actually document? (Select all that apply.)

• Potassium chlorate as a contrast control in platinum printing, with a stated mechanism
• Potassium chlorate as a contrast control in kallitype or Van Dyke printing
• Extra sulfamic acid raising argyrotype contrast by dissolving silver in the highlights
• Hydrogen peroxide added to an iron-silver sensitiser or developer to raise contrast

Show the answer and why

Answer: Potassium chlorate as a contrast control in platinum printing, with a stated mechanism, Extra sulfamic acid raising argyrotype contrast by dissolving silver in the highlights

Chlorate is thoroughly documented for the platinum processes - Pizzighelli and Hübl built a three-solution drop-counting system on it in 1882 - and not documented at all for the iron-silver ones in the sources read here. The argyrotype acid is Ware's own, with the mechanism in one clause. Hydrogen peroxide appears in this corpus twice, as a re-oxidation bath for a finished cyanotype and as an oxidant in one route to making ferric oxalate, and never as a contrast agent. That it would plausibly work is a hypothesis; running it as an experiment with a step tablet is the right response, and adopting it as a control is not.

Question 4. Design a measurement that would establish whether humidity changes contrast in your kallitype process. What is the control and what is the one variable?

• Print two negatives on two days with different weather, and compare the prints
• Coat one sheet, cut it in two after drying, equilibrate one half over a saturated salt solution and one in the room, expose both to the same step tablet at the same dose in the same session, and develop both together in the same bath - the coating, the exposure and the developer are held; the humidity at exposure is the variable
• Coat two sheets on the same day, expose one in the morning and one in the evening, and process them together
• Develop two prints in the same bath at two humidities and compare their maximum densities

Show the answer and why

Answer: Coat one sheet, cut it in two after drying, equilibrate one half over a saturated salt solution and one in the room, expose both to the same step tablet at the same dose in the same session, and develop both together in the same bath - the coating, the exposure and the developer are held; the humidity at exposure is the variable

One coating cut in two is what removes the largest nuisance variable in hand-coated printing, and Ware's own test target of eight identical step tablets on one sheet exists for the same reason. The second option is the only one that holds the coating, the exposure and the developer while moving humidity alone. Note also what the design should expect to find: humidity is a documented contrast control for the print-out platinum sensitisers and is not documented for kallitype at all, so a null result would confirm the reading in this page and a positive result would be a new finding worth reporting.

Question 5. Why does this course teach the dichromate contrast procedure in detail and refuse to give it?

• Because the quantities in the sources disagree too much to publish safely
• Because the chromium policy makes chromium(VI) never usable at any level: the classification carries H350, H340, H334 and H317 in essentially every ECHA report, and the workplace limit of 0.01 mg/m³ is annotated with a biological monitoring value, meaning the recognised check on the control is testing the worker
• Because dichromate is unstable in an oxalate developer and therefore does not work reliably
• Because the effect can be obtained more cheaply with hydrogen peroxide

Show the answer and why

Answer: Because the chromium policy makes chromium(VI) never usable at any level: the classification carries H350, H340, H334 and H317 in essentially every ECHA report, and the workplace limit of 0.01 mg/m³ is annotated with a biological monitoring value, meaning the recognised check on the control is testing the worker

The classification and the exposure limit are the whole basis of the ruling, and the rubric takes a procedure's level from the highest criterion any substance in it meets. The quantities are in fact consistent across four independent sheets, which is why the course can describe what was done with confidence. Ware's observation that dichromate is reduced by oxalic acid to oxalato-chromium(III) complexes is a real practical objection and would not on its own justify a refusal. And no source read for this course documents peroxide as a contrast agent at all.

Question 6. You double-coat a Van Dyke and the shadows are visibly deeper. Have you raised the contrast?

• Yes - deeper shadows against the same paper white is a longer scale and therefore more contrast
• Not necessarily: contrast is a slope, and whether the scale lengthened depends on whether the second coat also built density in the light values, which no source read for this course has measured
• No - multiple coating only affects maximum density and never contrast
• Only if the second coat is applied at right angles to the first

Show the answer and why

Answer: Not necessarily: contrast is a slope, and whether the scale lengthened depends on whether the second coat also built density in the light values, which no source read for this course has measured

A deeper maximum density is one of the three things this page names as looking like contrast without being it - the others are a cleaner highlight from better clearing, and a colour change from a different developer. Whether the exposure scale changed depends on what happened at the light end, and that is a step-tablet question. Wynn White's claim is about weak dark areas, which is a density claim, and neither manufacturer mentions double coating at all; so the honest position is that the effect on scale is unmeasured and a reader with a step tablet can settle it for their own paper in one evening.

Check my answersClear