Willis's potassium oxalate developer

Light does the reducing in this process, and then leaves the reducing agent where it cannot move. That is the whole problem this bath was invented to solve, and it is why the most famous developer in alternative photography contains no developing agent.

Ultraviolet turns the ferric oxalate in the coating into iron(II) oxalate, which is a reducing agent strong enough to put platinum metal into the paper — and which dissolves in water to the extent of 0.022 grams per hundred millilitres. It sits exactly where the photon left it. The platinum salt sits where it was coated. Neither can reach the other. William Willis’s third chemical problem, after finding a platinum salt he could buy and an iron salt that was sensitive enough, was to find something that would pick the iron up and carry it the few micrometres to the platinum without reducing anything on the way.

Ingredient	Quantity	Form the source specifies
Potassium oxalate monohydrate	7.78 g	120 grains, converted at Ware's 1 grain = 0.0648 g. Willis writes "potassic oxalate" and, in his first patent, "the neutral oxalate of potassium": the normal salt K2C2O4, not the acid oxalate. No hydrate is named in any nineteenth-century printing read for this page; Ware identifies the commercial article as the monohydrate, K2C2O4.H2O, formula weight 184.24
Water	28.4 mL, added	One imperial fluid ounce, at Ware's 28.413 cm3. Willis's line is "dissolving ... in one fluid ounce of water", an added volume rather than a make-up volume, so the finished bath is more than 28.4 mL. Distilled or deionised water, not because Willis says so but because calcium precipitates as calcium oxalate and pushes the iron(III) that has to leave the paper towards hydrolysis; the mechanism is set out under Incompatibilities.

Purpose

To dissolve the iron(II) that light has made, so that it can travel and hand two electrons to a platinum or palladium ion. Nothing else. Ware’s account of Willis’s three chemical obstacles names the developer’s job in one clause: potassium oxalate was needed “to dissolve the insoluble ferrous oxalate produced by light, providing a strongly reducing solution”, which could then reduce the platinum salt. The solution is strongly reducing because of what was dissolved into it, not because of what was in the bottle.

Two consequences follow immediately and they govern everything below.

The bath cannot exhaust in the ordinary sense. A developer for a silver gelatin paper is consumed: the developing agent is oxidised as it works and the bath dies. An oxalate ion that has complexed an iron(II) has not been chemically changed, and the concentrations involved are such that a working tray is in enormous excess over anything one sheet can ask of it. What ends a platinotype developer is contamination, dilution and sludge, not exhaustion.

Development stops when the iron runs out. This is why the modern kit sheet can print the sentence “You cannot overdevelop a Platinum Print” and why Wall could write in 1912 that “with correct exposure, over-development is impossible”. The quantity of image metal was fixed by the exposure; the developer only determines whether all of it is realised.

Recommended uses

Traditional development platinotype and palladiotype, coated with ferric oxalate and a noble-metal salt. This is the bath the platinotype was built around and the one both surviving kit suppliers ship. Photographers’ Formulary supplies 227 g of the solid in the platinum kit and 227 g in the palladium kit, differing only in the water it tells you to add; Bostick and Sullivan ship a 32-ounce bottle ready made and publish no strength for it at all.

Palladium printing in the American tradition. Willis specified an entirely different developer for his palladium papers — sodium citrate with citric acid — and Ware records that American workers largely ignored it, developing their palladiotypes in the used platinum oxalate bath already on the shelf. Paul Anderson published that practice in 1937 and 1938 with one correction: clear a palladium print in hydrochloric acid at 1:200, not the 1:60 used for platinum, which dissolves palladium. The palladiotype is the noble-metal print this course will let you make, and this is the developer most of its literature assumes.

Rescuing a pure platinum print made by the modern print-out route. Ware’s own print-out platino-palladiotype develops and clears in disodium EDTA, but he notes that with a 100 per cent platinum sensitiser the print-out is less vigorous, and that where highlight detail is deficient or the paper’s fibre structure shows as grain, the first bath may be replaced by “the more energetic traditional platinotype developer bath of ~30% potassium oxalate (poisonous!). It may even be used hot.” His parenthesis is quoted because it is his.

Reading a period manual. Between 1879 and 1937 essentially every commercial platinum paper sold anywhere was developed in this solution, and the manuals assume it without describing it. Knowing that “normal strength” meant a pound in fifty-four ounces, and that the working bath was that stock diluted one to two, is what makes a 1912 instruction legible.

When another formula is preferable

• When the metal is palladium and the print must be clean in the high values. Ware’s own tests find that the oxalate developer gives palladium slightly higher densities and a warmer colour but “does cause a perceptible chemical fog in the high values, which is worsened by humid conditions”. Willis’s sodium citrate developer leaves the highlights clear. The chemistry of that difference is set out under The mechanism, and it is not a matter of taste.
• When the print is a kallitype and the image is silver. Use the sodium citrate, Rochelle salt or borax developer. Those baths do a second job this one cannot: they hold iron(III) as well as iron(II), so the unexposed iron leaves the sheet in the developing tray. Oxalate holds iron(III) magnificently too, but a silver image in an acid oxalate bath is in exactly the wrong place, and the kallitype literature moved to alkaline developers for that reason.
• When you are making a print-out platinum or palladium image. The print-out process forms its image during the exposure and needs clearing rather than developing; a 30 per cent oxalate bath is available there as an option, not a requirement.
• When you want a colder, bluer black on a traditional platinotype. The Platinotype Company’s own answer was to sell a different developer, not a different paper: the phosphate-bearing “Special D Salts”, discussed under Variants. The course cannot publish it as a formula because the encyclopaedia has no page for potassium dihydrogen phosphate.
• When there is any silver in the system. Oxalate and silver salts are an incompatibility with a named product, silver oxalate, which the ILO-WHO chemical safety card describes as explosive. See Incompatibilities: this is not a preference, it is a rule.

Mixing

Seven and three-quarter grams of the salt into one fluid ounce of water. That is Willis’s own sentence from the 1880 patent, converted, and it is the entire published formula: “The developing solution of potassic oxalate is made by dissolving one hundred and twenty (120) grains or more of the salt in one fluid ounce of water.”

Nobody works in fluid ounces now, so here is the scaling, done in the open.

🧮From grains per fluid ounce to grams per litre, and what the per cent means

Ware’s Appendix VIII gives the two conversions this needs, and one shortcut:

1 grain = 0.0648 g · 1 imperial fluid ounce = 28.413 cm³ · 1 grain per fluid ounce = 0.228 per cent w/v

The units

So Willis’s bath, scaled to a litre of water, is:

120 gr × 0.0648 = 7.78 g in 28.413 mL → 7.78 × (1000 ÷ 28.413) = 274 g per litre of water

Willis 1880, scaled

and its nominal strength is 120 × 0.228 = 27.4 per cent w/v. Two cautions travel with that percentage, and both are the reason the course records the ingredient at Willis’s own scale rather than at the litre.

It is a nominal figure, not a measured one. The 0.228 factor treats grains per fluid ounce as though the ounce were the finished volume. Willis dissolved the salt in an ounce of water, so the finished bath is more than an ounce and the true per cent w/v is lower. Every “27 to 30 per cent” in the platinum literature, Ware’s included, carries that same convention.

“Or more” is Willis’s, and the trade took him at his word. The published strengths for this one bath run from about 25 per cent to saturation:

Source	As printed	Nominal per cent w/v
Willis, patent No 1117, 1880	120 grains or more per fluid ounce	27.4
Willis, patent No 20,022, 1913 (Satista)	1 lb in 64 fluid ounces	24.9
Wall, 1912, “normal strength” stock	1 lb in 54 fluid ounces	29.5
Wall, 1912, cold-bath working bath	that stock, 1 part to 2 of water	about 9.8
Wall, 1924, cold-bath developer	250 g per 1000 ccm	25.0
Ware’s own standard, 2017	28 per cent w/v, room temperature	28.0
Photographers’ Formulary, platinum kit	227 g in 500 mL, saturated	about 32 at saturation
Photographers’ Formulary, palladium kit	227 g in 700 mL, saturated	about 32

The pound is read as avoirdupois, 7000 grains, throughout. That reading is not a guess: 7000 grains in 54 fluid ounces is 129.6 grains per fluid ounce, and Ware independently describes Willis’s own recommendation as “120 to 130 grains per fluid ounce”. The apothecaries’ pound of 5760 grains would give 106.7 and miss it.

Dissolve it in warm water, and expect it to take time. CAMEO’s datasheet for potassium oxalate monohydrate records that the solid “sinks in and mixes slowly with water”, which is the practical reason a 28 per cent bath is not made in five minutes. Wall’s 1912 dictionary gives the solubility as 1 part in 3 of water, so saturation is around 32 per cent w/v and a 25 to 28 per cent bath is close to it: the last of the solid goes reluctantly.

The modern kit deliberately overshoots saturation, and says so. Photographers’ Formulary’s platinum sheet is 227 g of solid into 500 mL of water, with the note “Not all of the solid will dissolve. This saturated potassium oxalate solution is the developer.” Wall’s 1-in-3 figure says how much will not: 500 g of water will hold about 167 g, leaving roughly 60 g of solid on the bottom of the bottle. That undissolved reservoir is the point. It keeps the decanted bath at saturation as water is drawn off and replaced, which is the same trick a saturated stock solution performs anywhere in chemistry — and it is a tidy answer to Willis’s “or more”.

The same supplier’s palladium sheet says 700 mL, and the difference is worth noticing. The identical 227 g goes into 700 mL there, with the identical sentence about not all of it dissolving. Against Wall’s solubility that is a different bath: 700 g of water will hold about 233 g, so the palladium version sits a hair below saturation with little or nothing left over, at a nominal 32 per cent, while the platinum version sits well above it with a reservoir. Neither sheet explains the change and no source read here reconciles them. The palladium sheet is also the only one that publishes a keeping statement — “It has an indefinite life and may be replenished with fresh solution to maintain volume” — which is recorded in this entry’s shelfLife and is the closest thing the literature offers to a capacity.

✕Which potassium oxalate did you weigh?

Nothing written before about 1950 names the hydrate, and the two articles on a supplier’s shelf are not interchangeable by weight. Potassium oxalate monohydrate has a formula weight of 184.23; the anhydrous salt is 166.22.

monohydrate: 280 ÷ 184.23 = 1.52 mol/L · anhydrous: 280 ÷ 166.22 = 1.68 mol/L

Oxalate delivered by 280 g in a litre

That is eleven per cent more oxalate per gram from the anhydrous grade. For the bath’s main job the difference is invisible, because the ligand is in five-hundred-fold excess either way — the arithmetic is under The mechanism. It matters in exactly one place: if you acidify the bath, the ratio of oxalate to added oxalic acid sets the pH, and the grade moves the ratio. Ware identifies the traditional salt as the monohydrate and works at 28 per cent w/v in it. Record what you weighed in the lab notebook.

Use distilled or deionised water. No nineteenth-century source read for this page says so, and the reason is chemical rather than fastidious. Calcium in hard water precipitates as calcium oxalate and, in doing so, strips oxalate from the iron complexes that have to leave the paper:

Ca²⁺ + Fe(C₂O₄)₃³⁻ + 2 H₂O → CaC₂O₄ + Fe(C₂O₄)₂(H₂O)₂⁻

Ware: calcium takes the oxalate and pushes the iron towards hydrolysis

Ware follows that arrow through aquation to iron(III) hydroxide and eventually to goethite, which no clearing bath will remove. A developer made in hard water starts that chain in the tray.

Nothing here needs a safelight, and nothing here needs heat to be safe. The developer contains no light-sensitive substance while it is fresh and may be mixed and bottled in room light — which stops being true the moment it has been used, for reasons under Storage. For the bench work, follow the SOP for weighing a solid and, if you are diluting a stock, the SOP for mixing from a stock.

Behaviour

The image is there before the sheet is flat. The Getty Conservation Institute’s atlas puts it in four words — “The platinotype image appears almost immediately” — and the Photographic Materials Group uses three: it “comes up instantly”. That instantaneous arrival is the first thing to plan around, because it means the bath cannot be approached casually.

Which is why every source that gives a handling instruction gives the same one. Wall’s 1924 direction is to take both ends of the sheet, immerse one end face down, draw the paper right through the solution and then turn it face up. Photographers’ Formulary, eighty years later and with a heated bath in mind, warns that “because development occurs almost immediately, it is important to submerge the print quickly and evenly into the developer solution to avoid streaking and uneven development”, and that an air bubble must be brushed away at once. Wall’s 1912 alternative for cold-bath papers is to pass the print quickly under the solution, or to float it face down. One pass, one motion, no hesitation.

Then you wait, and the waiting is not for the image. The published times run from thirty seconds to five minutes and they are not measuring the same thing. Wall 1912 gives thirty seconds at 60 °F for a cold-bath paper in the diluted trade developer. Wall 1924 gives one to two minutes with the dish rocked, calling development “comparatively slow”. Willis’s own 1913 patent says about one minute for Satista. The modern kit says at least two minutes at room temperature or at 90 to 100 °F. The spread is real and it reflects the paper, the coating weight and the temperature; what none of the sources suggests is that a longer time makes a heavier print once the reaction is done.

Over-development is not a failure mode here. Wall’s sentence is unqualified — “With correct exposure, over-development is impossible” — and Photographers’ Formulary repeats it. The reason is structural: the iron(II) is a fixed stock laid down by the exposure, so the reaction runs to completion and stops. Compare a silver gelatin print, where the developer will go on reducing unexposed halide until fog arrives.

Under-development is a failure mode, and a subtle one. Wall 1912 notes that development can be arrested early by plunging the print into the acid bath, but that the result “is not so good as in the case of normal exposure and development, and there is a tendency to granularity”. That granularity is the same defect Ware explains chemically: if the wet processing washes the soluble platinum salts out of the paper before they have reacted fully with the iron(II), “the image will appear weak and fibrous or grainy”. Development is a race between two dissolutions and a reduction, and stopping it early hands the race to the wrong runner.

The bath is reused, and it changes as it is. Bostick and Sullivan’s sheet is the plainest modern account: the developer “can be saved and used over and over again as long as you replenish what evaporates with fresh developer, and filter out the sludge that accumulates periodically using a coffee filter and plastic funnel. The color will darken due to dissolved metal accumulating in it. Many printers swear the older the developer, the better. Other printers prefer to start with fresh developer every once in a while.” That disagreement is trade opinion and is recorded as such; the chemistry that gives it some grounding is under Interactions.

No source read for this page publishes a capacity. Not in prints, not in sheets, not in square metres. That is unusual for a formulary entry and it is stated rather than filled in: this bath is worked until it is abandoned, not until it is spent.

Image characteristics

Colour is what this developer controls, and it controls it two ways. The Getty atlas lists them together: the tonality of a platinotype “could be controlled by processing (hot/cold developer, concentration of the developing solution) or by adding mercury salt to the developing solution”. Hotter is warmer; stronger is warmer; and the mercury route belongs to the paragraph at the foot of Variants that this course will not turn into a procedure.

But the warmth a hot bath buys is small, and the literature has overstated it for a century. Ware is direct about it: “It is sometimes mistakenly stated that ‘sepia platinotypes’ were simply made historically by employing a hot developer bath, at ca. 160 °F. However the degree of ‘warming’ of the neutral grey-black colour of normal ‘cold bath’ Platinotype paper achieved by this means is found to be only ‘very slight’.” Sepia platinotype was a mercuric-chloride sensitiser, sold as a different paper. The hot bath warms the black a little; it does not make a brown print.

Contrast is not this bath’s business. The Photographic Materials Group states it flatly: image contrast “is achieved mostly by exposure, not the developer, although contrasting agents can be applied to the process”. The developer’s contribution is real but second-order, and the two published statements about it point in the same direction. Wall 1924, on the black papers: “Dilution of the developer gives more brilliant prints.” Wall 1924 again, on the sepia papers: “The stronger the developer, the more rapid its action and the softer the print.” A weaker bath develops more slowly, loses more of the race to the washing-out of the salts, and the loss falls hardest where there was least iron to begin with — in the high values. That is a contrast increase bought by throwing away the weakest part of the image, which is exactly what Ware says every oxidising contrast agent does: the effect “is not to contract the tonal scale uniformly but simply to truncate the high values”.

Density comes from the coating, not from the tray. The developer cannot put down more platinum than the exposure made iron(II) for. What it can do is fail to put down all of it, and a thin, chalky maximum black on a correctly exposed sheet points at the developer — too weak, too cold, too brief, or too old — before it points anywhere else.

On palladium, the same bath is a different animal. Ware’s comparison of oxalate against citrate development on palladium sensitisers finds that the oxalate bath gives “slightly higher densities and warmer image colour”, higher speed, and “a perceptible chemical fog in the high values, which is worsened by humid conditions” — a brownish-grey fogging by palladium metal, quite distinct from the yellow stain of iron. Under The mechanism there is a reaction that explains all four observations at once.

The mechanism

Three things happen, in order, and only the second one is the developer’s.

One: light makes an insoluble salt and leaves it there

Ultraviolet decomposes the iron(III) oxalate in the coating. Ware gives the reaction Döbereiner first observed in 1831:

UV + Fe₂(C₂O₄)₃ → 2 FeC₂O₄ + 2 CO₂

Photolysis of iron(III) oxalate: a solid photoproduct and a gas

Two things are now true of the exposed sheet. It contains iron(II), which is a reducing agent. And that iron(II) is iron(II) oxalate, whose solubility Ware gives as 0.022 g per 100 cc of water — about 1.5 millimoles per litre at a formula weight of 143.86. It is a solid, sitting in the fibres where the photon found it, and Ware states the consequence: iron(II) oxalate “cannot reduce platinum(II) or palladium(II) salts in aqueous solution to the metal unless it is solubilised by complexation”.

Two: the developer dissolves it

FeC₂O₄ + C₂O₄²⁻ → [Fe(C₂O₄)₂]²⁻

Oxalate ions dissolve the photoproduct into a mobile complex

That is the only reaction the potassium oxalate takes part in. It is a ligand substitution, not a redox step: the iron is iron(II) before and iron(II) after, and the potassium ions are spectators from beginning to end, which is why Ware omits them from every balanced equation on the subject.

⚗Why a concentrated bath, and how much excess there really is

A 27 per cent solution of the monohydrate is about 1.5 mol/L in oxalate, and it needs to be. Two separate reasons pile up.

The complex has to be forced. Ware gives the stepwise formation constants for the iron oxalato complexes as roughly K₁ ≈ 10⁹, K₂ ≈ 10⁷, K₃ ≈ 10⁴. Those are for iron(III); the iron(II) analogues are far weaker, Ware giving an overall formation constant near 10⁵ against 10²⁰ for the iron(III) tris complex. Weak binding is dissolved by mass action, and mass action means concentration.

The bath has to hold the unexposed iron too. Most of the sheet was never exposed, and every part of it carries iron(III) that has to leave. The developer takes a share of that job before the clearing baths finish it.

How much excess is that? Take Willis’s own coating figure from the same patent: not less than 1.7 and preferably about 4 grains of potassium chloroplatinite per square foot.

80 in² = 0.556 ft² → 2.22 gr = 0.144 g K₂PtCl₄ ÷ 415.1 = 0.35 mmol Pt

One 8 by 10 sheet, at Willis's 4 grains per square foot

0.35 mmol Pt × 2 Fe(II) × 2 oxalate = 1.4 mmol · 500 mL at 1.49 mol/L = 745 mmol

What that asks of the developer, against what a small tray holds

More than five hundred times what the platinum on the sheet can possibly consume. This is the arithmetic behind “the bath does not exhaust”, and it is the course’s, from Willis’s coating weight and Ware’s formula weights.

Three: the iron(II), now mobile, reduces the metal

2 [Fe(C₂O₄)₂]²⁻ + [PtCl₄]²⁻ → 2 [Fe(C₂O₄)₂]⁻ + Pt + 4 Cl⁻

Ware's development reaction: two iron(II) per atom of platinum

Two iron(II) ions per platinum atom, because platinum(II) needs two electrons. The same equation with tetrachloropalladate gives the palladium print. The metal comes down as nanoparticles trapped among the cellulose fibres, and that is the photograph.

The thermodynamics, in one table. Ware measures the reducing power of the complexed iron and lists the noble metals against it. A couple with a more negative potential reduces one with a more positive potential.

Couple	Potential	What it means here
Fe(III)/Fe(II), uncomplexed	+0.771 V	A moderate oxidising agent. Useless for this
Fe(III)/Fe(II) as oxalato complexes	+0.02 V	A moderate reducing agent. This is what the developer creates
Fe(III)/Fe(II) as citrato complexes	+0.372 V	Weaker. Ware states it does not reduce platinum(II) or palladium(II) under the printing conditions
PtCl42− / Pt	+0.73 V	Reduced by the oxalato couple, with 0.7 V to spare
PdCl42− / Pd	+0.62 V	The same, more easily

The whole of the developer’s power is in that second row. Complexing iron(III) by oxalate drags the couple down from +0.771 V to +0.02 V — three-quarters of a volt — and Ware gives the reason: oxalate binds iron(III) enormously more strongly than iron(II), with formation constants of about 10²⁰ against 10⁵, so the oxidised form is stabilised and the reduced form becomes eager to give its electron away. Choosing the ligand is choosing the reducing power. That is the single most transferable idea on this page, and it is why swapping oxalate for citrate changes what the bath can reduce rather than merely how fast it does it.

Deeper: why the bath must be neutral, or barely acid

Every text Ware cites says the developer should be neutral or just slightly acid, and both halves of that instruction have a reason.

Why not alkaline. Above about pH 4, iron(III) begins to hydrolyse. Ware sets out the four stages — rapid reversible dimers, then a “red cationic polymer” sol of 2 to 4 nm particles over minutes, then larger polymers over days and weeks, then, over years, crystallisation to goethite, α-FeO(OH), which is “quite insoluble in dilute acids”. A print that leaves the developer with hydrolysed iron in it is a print with a yellow stain scheduled for some time in the next decade. Ware’s warning against using tetrasodium EDTA at pH about 10 as a first bath is the same warning in modern dress: at high pH the iron(III) “is not effectively complexed, but rather hydrolysed”.

Why not more than barely acid. Ware, citing Crawford: if much more than a trace of acid is present, “it is said to inhibit the formation of platinum”. No mechanism is given for that in any source read here, and the course does not supply one.

Why any acid at all. Because the bath is reused, and it accumulates alkali from the water and from the paper. A trace of oxalic acid holds the line. The doses in the literature vary by a factor of twenty: some recipes carry 0.1 per cent or less; Irving Penn’s bath was 30 per cent oxalate with 0.5 per cent oxalic acid; Ware records that some modern practitioners of the traditional process use about 2 per cent excess oxalic acid, and offers as a presumption rather than a finding that this lowers the pH below 6 and inhibits the hydrolysis of iron(III), avoiding staining.

⁇A conflict between two sourced statements about acid

The tension is real and neither side is wrong within its own frame. Ware relays a traditional rule that more than a trace of acid inhibits platinum formation, and separately relays that contemporary printers add twenty times a trace, with a plausible reason. The Photographic Materials Group adds a third statement that cuts across both: “the higher the pH, the lighter the print and conversely, the lower the pH, the darker the print”, together with the observation that the process is inherently acidic and that more acidic processing produces browner tones.

What the course does. It reports all three, attributes each, and notes that no source read for this page publishes a pH figure for the working bath, a titration, or a controlled comparison. That is why targetPH on this entry carries a sentence rather than a number. The experiment at the foot of the page is the smallest test that would settle it for one paper and one coating.

Deeper: the oxalate can reduce platinum by itself, and that fact wrecked one of Willis’s processes

Everything above treats oxalate as an innocent solvent. Thermodynamically it is not. Ware gives the direct reaction and the potentials that permit it:

C₂O₄²⁻ + [PtCl₄]²⁻ → 2 CO₂ + Pt + 4 Cl⁻

Oxalate reducing platinum(II) without any iron involved

with E(PtCl₄²⁻/Pt) = +0.73 V against E(2 CO₂, 2 H⁺ / H₂C₂O₄) = −0.49 V. It is allowed. What stops it, for platinum, is kinetics: Ware describes the reaction as “probably very slow with platinum” — and adds, in the same breath, that “it has been observed with palladium”.

Three separate observations fall out of that one sentence.

Willis’s 1888 process failed on it. From 1888 to 1892 the Platinotype Company sold a “cold bath” paper carrying no platinum at all, the whole of the metal being dissolved in the oxalate developer. Ware records what Willis admitted to the London Camera Club: both oxalate of platinum and platinum metal itself soon precipitated out of the stored developer, “which quickly became black and unusable”. A concentrated oxalate solution is a slow reducing bath for platinum(II), and a bottle of it is a slow plating reaction. Willis withdrew the process in 1892.

Palladium fogs in this developer. Palladium(II) is a second-row transition metal and its complexes are far more labile than platinum’s; Ware attributes the whole behavioural gap between the two metals to that difference in ligand-field activation energy. The reaction that is too slow to matter with platinum is fast enough to matter with palladium, and the result is precisely the “brownish-grey fogging of the highlights by palladium metal” that Ware’s own comparisons found — chemical fog laid down where there was no image, by the developer itself. It also explains the higher density and the higher speed he measured, and it explains why Willis, “a shrewd and observant chemist, painstaking in his perfection of process”, devised a citrate developer for palladium rather than reusing the one he already had.

A used bath is still a plating solution. Ware notes that a re-used platinotype developer always contains sensitiser reagents in solution and remains photosensitive; left in the light it “will steadily precipitate particulate platinum metal, which if left in suspension could find its way onto the surface of subsequent prints”. That is one of the hypotheses for the “plague of black spots”.

Deeper: what the temperature was actually doing, and what 1892 changed

Before 1892, the trade developed hot: 140 to 170 °F, which is 60 to 77 °C. Ware’s reason is kinetic — to accelerate the chemistry of platinum precipitation and so provide an adequate image quality. The underlying fact is in his comparison of the two metals: platinum is a third-row (5d) transition metal, its complexes have larger ligand-field activation energies, and its reactions are generally slower than those of a second-row metal like palladium. The hot bath existed because platinum sits one row further down the periodic table than palladium does.

And the race it was winning is the one Willis himself described. In 1892 he told the Camera Club of London that he had found a way of preparing ordinary platinotype paper “so that during development the rapidity of solution of the salts shall not overtake the rapidity of reduction — or, to express it differently — so that the image shall be developed before the salts which cause its formation have been removed from the paper”. That is a photographer’s statement of a kinetic competition, made forty years before anyone could have written the rate laws, and it names both competitors correctly.

🕰The one thing the platinotype never told anybody

Willis followed that sentence with: “For reasons which will be understood I cannot make this method public.” He never patented the cold-development paper and never published it. Ware records that no technical records of the Platinotype Company are known to have survived the London Blitz, so the manufacture of the most successful platinum papers ever sold is a lost secret.

What can be inferred is narrow and Ware states it as inference: Willis said “ordinary Platinotype paper”, which points at the sizing or the coating rather than at the sensitiser, and he is known to have avoided gelatin sizing from about 1880 because it inhibited the platinum chemistry. No source read for this page explains, chemically, what made cold development work. The course records the date, the effect and the gap.

The sepia papers never made the transition. Willis admitted in 1892 that he doubted sepia would ever be obtained below 150 °F, and in 1893 that “to make good sepia paper is a heart-breaking problem”. His one partial success was to add glycerine to the developer, which allowed a lower temperature; the Platinotype Company’s lists still called sepia a “hot bath paper” in 1906, and by 1911 the commercial sepia papers were still preferably developed at 160 to 170 °F.

Function of every ingredient

Potassium oxalate monohydrate, 7.78 g per fluid ounce of water — 120 grains, in Willis’s words, “or more”. It is the only substance in the formula, and it is worth being precise about which of its two ions does the work and which does none.

What it is. The neutral, or normal, dipotassium salt of oxalic acid, K₂C₂O₄·H₂O, formula weight 184.23. Willis specifies “the neutral oxalate of potassium”, which distinguishes it from the acid oxalate — potassium hydrogen oxalate, the old salt of sorrel — and the distinction is not pedantry: the acid salt would bring the bath’s pH down far past the “barely acid” the process wants.

Why it is there. To supply oxalate anions at a concentration high enough to dissolve a nearly insoluble solid. Its 1.5 mol/L of free oxalate converts the iron(II) oxalate that light made into bisoxalatoferrate(II), which can diffuse. The potassium is a spectator; the water of crystallisation is freight.

What it does chemically. Three things at once, in order of importance. It complexes iron(II), mobilising the reducing agent. It lowers the iron redox couple from +0.771 V to +0.02 V, which is what makes the mobilised iron capable of reducing platinum(II) at all — a solvent that did not also depress the potential would dissolve the photoproduct and achieve nothing. And it complexes iron(III), both the iron(III) produced by the development reaction and the great excess that was never exposed, starting the removal that the clearing baths finish.

What follows on the print. A full maximum black in the shadows, provided the reaction finishes before the salts wash out; and, in the highlights, whatever the exposure left, faithfully.

More. A stronger bath develops faster and further; Wall’s rule is that the stronger the developer, the more rapid its action and the softer the print, and that dilution gives more brilliant prints. Warmer tone, too, on the historical evidence. Past about 32 per cent w/v nothing more dissolves, and the extra sits on the bottom as a saturation reservoir.

Less. Slower development, a longer race against the dissolving salts, and — where the exposure was marginal — a weak, fibrous, granular image with a poor maximum black. Wall’s cold-bath developer C is nothing but developer A at double strength, prescribed for exactly this: “in cases where prints appear granular, or where intense blacks are wanted”. Very much less, and Ware’s substitution of the oxalate bath for EDTA in pure platinum printing tells you what the bottom of the range looks like: a 5 per cent chelating bath develops a print-out image adequately, and a 30 per cent oxalate bath is the “more energetic” option.

What it interacts with. Silver, dangerously (see Incompatibilities). Calcium, which takes its oxalate. Acid, which it neutralises — CAMEO classes it under Salts, Basic and records that it neutralises acids with the generation of heat. Oxidising agents, with which CAMEO records that this weak reducing agent can generate carbon dioxide, and which is the reason a dichromate addition to this bath does not keep.

Water, one fluid ounce, added. Not a make-up volume: Willis dissolves the salt in an ounce of water, and the finished bath is more than an ounce. It is the reaction medium and the transport medium, and its purity is a chemical variable rather than a housekeeping one — see Mixing for the calcium oxalate reaction that hard water starts. Its temperature is the second variable in the formula and was the trade’s principal control for a decade and a half: 60 to 77 °C for the hot-bath papers, room temperature after 1892, 32 to 38 °C in the modern kit sheet as an optional accelerant.

Interactions

With the sensitiser it develops. This bath is written for a coating of ferric oxalate and a noble-metal chloride — the three-solution drop system in its modern form. The pairing is not arbitrary: the sensitiser’s ligand is already oxalate, so the developer adds more of what is there rather than exchanging one ligand for another. Ware notes the consequence for the sensitiser side of the same equilibrium: adding excess oxalic acid to a ferric oxalate sensitiser displaces water and hydroxide from the iron’s coordination sphere in favour of the tris-oxalato complex, which increases print-out at the expense of development and improves the ease of clearing.

With an ammonium iron(III) oxalate sensitiser, it is largely redundant. In the Malde-Ware print-out chemistry the photoproduct is already a soluble complex, so the ions can move without help and the image forms during the exposure. That is the clearest possible demonstration that this developer’s whole function is solubilisation: change the iron salt so the photoproduct dissolves by itself, and the developer becomes optional.

With an oxidising agent, wherever it is put. Ware’s account of contrast agents applies identically to potassium chlorate in the sensitiser and potassium dichromate in the developer: the oxidant re-oxidises part of the iron(II) the light made, so less of it is available to reduce the metal, and the exposure scale is truncated. In the developer there is an extra wrinkle Ware supplies: 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 contrast control that decays in the bottle is a variable pretending to be a constant.

With the clearing baths that follow, which are one operation with it. Development pulls iron into solution; the clearing sequence removes what is left. The handover matters because the developer is close to neutral and the first clearing bath must be acid: Ware’s first bath is disodium EDTA at pH 3 to 4, and his instruction not to use the tetrasodium salt at pH 10 is the sharpest warning in the platinum literature about what an alkaline bath does to iron(III). The traditional route was three baths of hydrochloric acid at 1:60 for platinum, and Willis’s own preference in the 1880 patent was a weak acid, “preferably citric acid”.

With the paper, which is not inert. Ware’s clearing chapter notes that some iron(III) chemisorbs onto the hydroxyl groups of cellulose, and that a calcium carbonate buffer in the sheet will precipitate calcium oxalate and drive hydrolysis. A buffered watercolour paper fights this developer.

With the last print’s chemistry, through the tray. This is the interaction that has no analogue in silver work and it is the most interesting thing about a reused bath. Ware’s investigation of Alfred Stieglitz’s palladium prints found platinum and mercury in prints that were nothing but palladium on the label. His explanation is the developer: a used platinotype bath from Stieglitz’s own shelf would have accumulated salts of platinum and mercury in the ordinary course of its previous life, and “would chemically precipitate these metals into the palladium image during the development process”. A tray of this developer is a slow, accumulating record of everything printed in it — which is one reading of the trade opinion that an old bath prints better, and a decisive argument for keeping separate bottles for platinum and palladium.

With air and light, slowly. Air re-oxidises iron(II) to iron(III); Ware gives the reaction and notes that it is slow in acid and rapid in base, which is one more reason not to let this bath drift alkaline. Light, acting on the sensitiser residues a used bath carries, precipitates metal.

Variants

The Platinotype Company’s “Special D Salts”, sold as the cool alternative. Ware reports a modern analysis of the Company’s proprietary developer as approximately:

Component	Proportion
Potassium oxalate monohydrate	85 %
Potassium dihydrogen phosphate	12 %
Oxalic acid dihydrate	3 %

made up to a working strength of only 8.6 per cent w/v for the black papers at room temperature, and 10.3 per cent for the sepia Japine papers at 70 to 80 °C. Ware prints those proportions once as per cent w/w and once as per cent w/v; the course reads them as proportions of the dry mixture, which is the only reading in which they can be a recipe for a tin of salts.

Two observations belong with that table. The working strength is a third of the patent bath, and it agrees closely with the independent figure that falls out of Wall’s 1912 instruction — dissolve a tube of D salts in 48 ounces of water and dilute one part with one of water. And the added ingredient is a second ligand. Willis’s 1887 patent is explicit that phosphates are chosen as salts “in which the ferrous oxalate image produced by exposure to light is insoluble or nearly so”, used with or without potassium oxalate “to increase the rapidity and vigour of the development”. A phosphate developer holds iron differently from an oxalate one, and the Company sold the difference as a colder, bluer black.

This course cannot publish it as a formula, and the reason is worth stating plainly: the chemical encyclopaedia has no entry for potassium dihydrogen phosphate or for the di-potassium salt, so a reader could not look up what they were being asked to weigh. Inventing the entry is forbidden, and so is a stub standing in for one. It is recorded here, with its source, as history.

The phosphate developers generally. They are a family, not a curiosity. Wall’s 1912 dictionary gives a cold-bath developer of potassium oxalate 2 oz and potassium phosphate ½ oz in 28 oz of water, and a hot-bath sepia developer of oxalate, phosphate, citric acid and potassium chloride. Wall 1924 gives 100 g oxalate with 50 g potassium phosphate per litre as the alternative cold-bath developer. Ware records that the Autotype Company’s three platinum papers of 1895 were all developed in potassium oxalate plus potassium hydrogen phosphate. The same encyclopaedia gap applies to all of them.

The acidified versions. Adding oxalic acid to the oxalate bath is the commonest variation in the whole literature and the doses span a factor of twenty: Wall’s 1 part of saturated oxalic acid to 20 of the mixed cold-bath developer; the sepia bath at 10 parts of normal-strength oxalate to 1 of saturated oxalic acid; Penn’s 0.5 per cent; the “0.1 per cent or less” of some recipes Ware cites; and the “about 2 per cent” he attributes to some modern practitioners of the traditional process. What the acid is for is under The mechanism, and so is the countervailing rule that more than a trace inhibits platinum formation. The course publishes none of these as its own formula because no two of them agree and none is offered with a measured result.

Willis’s own successors to this bath, which are separate formulas. Three of them, all his: the 1878 patent’s developer of potassium oxalate carrying 7 grains of potassium chloroplatinite to the fluid ounce; the 1887 patent’s phosphate developers carrying all of the platinum, of which the best known is 100 grains of di-potassic orthophosphate, 40 grains of potassium oxalate and 10 grains of potassium chloroplatinite per fluid ounce; and the 1880 patent’s own alternative, a sodium citrate developer at 120 grains to the ounce, which he preferred among the tartrates, citrates and acetates and which became the palladiotype developer thirty-seven years later. The first two are, chemically, the ancestors of the process that failed in 1892 for the reason given under The mechanism; the third is the one that survived.

The contrast additions. Wall 1912: “the addition of one grain (in no case to exceed two grains) of bichromate of potash to every 20 ozs. of developer gives noticeably increased contrasts. This property gradually wears off in the developer” — the decay Ware later explained. Wall 1924: 2 to 5 per cent of a 1 per cent solution of potassium bichromate, or 0.5 to 1 per cent of ammonium persulphate, which “shortens the scale of gradation”. Willis and Clements recommended a very small amount of dichromate in the oxalate bath for “brilliant” palladium prints, and Paul Strand and Ned Scott used it on flat negatives.

☠The dichromate and mercury variants, and why no quantity of either is offered as a procedure

Chromium(VI). Potassium dichromate is notified as Danger with H350 and H340 — may cause cancer, may cause genetic defects — along with H317 and H334 sensitisation statements and an aquatic hazard. The chromium policy governs, and the quantities above are printed as what the period sources said rather than as a dose to use. The mechanism does not need chromium: any oxidant that reaches the iron(II) first will truncate the scale, and Ware’s own position is that with modern negative-making “contrast-enhancing agents, such as chlorate, hexachloroplatinate(IV) or dichromate, become unnecessary when a correctly calibrated negative is made”. He calls them “undesirable image-degrading agents”.

Mercury. Mercuric chloride in the developer is how the first sepia platinotypes were made, and Wall’s 1912 dictionary prints a full glycerine-retarded mercury developer with six working ratios. The course publishes none of it. Ware’s chemistry says what happens: mercury(II) in a concentrated oxalate bath becomes the bisoxalatomercurate(II) complex, whose potential of +0.39 V is still well within reach of the iron(II), so mercury metal co-precipitates with the platinum — and the system is photosensitive, which is why the period manuals said the sepia developer “must be kept in the dark”. The commercial verdict was delivered in 1911 by the Platinotype Company’s own manager, W. H. Smith: the Company “had never advocated the use of mercury in their developers simply because it was not stable, and they would not include a formula in their instructions which would injure the reputation of platinum printing for permanency”. The course’s separate entry for mercuric sepia platinotype covers the sensitiser route at Level D. The modern substitute for a warm-toned noble-metal print is palladium, which prints brown without any mercury at all.

The stranger additives, recorded for identification only. Ware notes Professor Borlinetto’s 1885 claim that 10 per cent of a saturated copper chloride solution added to a hot potassium oxalate developer at 176 °F produced a “warm sepia-brown tone”, and a sepia developer attributed to Dr Jacoby containing zinc oxalate, of which “the larger the quantity, the warmer the tone”. Neither substance has an encyclopaedia entry here and neither claim has been tested in any source read for this page.

No safer course variant of the quantities is offered, because there is nothing to make safer in the formula itself. The bath is one salt in water. Its variants are other people’s formulas, and the two that carry real hazard are the two above.

Safety

Level B, on the substance rather than on the procedure. Potassium oxalate monohydrate is notified to the ECHA inventory as Warning with GHS07 and the statements H302 harmful if swallowed, H312 harmful in contact with skin, H315 causes skin irritation and H319 causes serious eye irritation. That classification alone would sit near the A/B boundary. Three things push it over.

The suppliers’ own warnings go further than the notified classification. Photographers’ Formulary singles the substance out: “The platinum printing kit contains a chemical that needs special attention: Potassium Oxalate. This compound is an anticoagulant (prevents blood clotting) and a poison. Since this chemical is used as a developer, it can easily come into contact with your skin. It is strongly advised that you use tongs to develop platinum prints or wear rubber gloves.” Ware’s own instruction, in the middle of a technical paragraph, is the single word “(poisonous!)”. CAMEO’s health hazard entry describes ingestion as capable of causing burning pain, vomiting, severe purging, weak pulse and cardiovascular collapse, and states that inhalation of the dust can cause systemic poisoning.

It is a tray bath, worked with wet hands nearby, and used again tomorrow. Unlike a one-shot developer poured away after a film, this solution lives in an open tray during a session and in a bottle between them, so the exposure is repeated and the labelling has to survive years.

The hot-bath versions are a scald hazard before they are a chemical one. A tray of solution at 60 to 77 °C is handled with the fingers, and Ware’s summary of what Willis’s 1892 announcement offered the trade is development at room temperature “rather than at the elevated temperatures which had caused so many workers previously to scald their fingers” — Ware’s words about the practice, not a quotation from Willis. Nothing in modern practice requires the heat; the kit sheets’ optional 32 to 38 °C does not.

Personal protection. HSE’s COSHH essentials sheet for manual film development takes single-use nitrile gloves as splash protection where the safety data sheet gives no more specific advice, and eye protection where splashing is possible. Print tongs are the primary control for a developing tray and the suppliers say so. Dust control matters when weighing the solid: EH40 sets no limit for potassium oxalate, and states in terms that absence from the list does not mean a substance is without risk; the nearest published figure is oxalic acid at 1 mg/m³ over eight hours and 2 mg/m³ short-term.

⚠What is not a hazard in this bath, and why

There is no platinum in this developer. In the process as Willis perfected it in 1880 and 1892, and in every kit sold today, the whole of the noble metal is in the coating and none is in the tray. That matters because the course’s tightest exposure limit anywhere is EH40’s halogeno-platinum compounds at 0.002 mg/m³ as platinum with the Sen notation, and the reason the platinotype is a Level D process. It is the sensitiser that carries that hazard, not this bath — which is why the course can teach the developer at Level B and the palladiotype as a process a student may perform. Willis’s 1878 and 1887 developers, which did carry a platinum salt, are history rather than practice, and are described above without a procedure.

There is no vapour and no flammability. Nothing here is volatile, nothing is a solvent, and CAMEO records that the salt’s thermal decomposition to the carbonate is not a hazardous reaction. A hot bath raises humidity and nothing else.

There is no acid or alkali strong enough to burn. The neutral salt is a weak base whose solution neutralises acids with some warming; the optional oxalic acid additions are the strongest thing that ever enters this tray, and even at 2 per cent that is a weak acid. The corrosive baths in this process are elsewhere, in the hydrochloric acid clearing sequence.

What that leaves is one thing, and it is a real one: the solution is toxic by ingestion and readily absorbed through wet skin, so the entire control is do not get it on you and do not eat at the bench.

Storage

Bottled, labelled, dated, and kept dark once it has been used. Store it away from food and food containers, as any toxic solution is stored; the darkness, though, is specific to this bath and matters only after its first use, because a used developer carries dissolved sensitiser and is photosensitive. Ware’s observation is that such a bath, if not protected from actinic light, will steadily precipitate particulate platinum metal into suspension.

The solid effloresces. Potassium oxalate monohydrate loses its water of crystallisation in warm dry air, which makes it stronger per gram than the label implies. A tightly closed container is not tidiness; it is the difference between weighing the monohydrate and weighing something between the two hydrates.

Keep separate bottles for platinum and for palladium. The Stieglitz finding under Interactions is the argument: a used bath accumulates the metals printed in it, and will precipitate them into the next print. A bottle that has developed platinum will put platinum into a palladium print.

Replenish evaporation and filter the sludge. That is Bostick and Sullivan’s instruction and it is the only maintenance regime published: top up what evaporates with fresh developer, and pass the bath through a coffee filter periodically. Ian Leake’s observation, relayed by Ware, gives a second reason to filter: filtering the sensitiser did not reduce black spots in his prints, but filtering the developer did.

Do not store it near, above, or in the same cabinet as any silver salt. See Incompatibilities.

Incompatibilities

Silver compounds, without exception. The NIOSH pocket guide lists silver compounds among oxalic acid’s incompatibilities, and the ILO-WHO International Chemical Safety Card names the reason: silver oxalate is explosive. A darkroom that runs salted paper or kallitype alongside platinum work has silver nitrate on the shelf and silver-bearing waste in the sink, and this developer, its stock salt and its spent bath are kept away from all of it — different shelf, different waste container, different funnel. This is the one hard rule on the page.

Calcium, and therefore hard water and chalk-buffered paper. Calcium oxalate is essentially insoluble. It precipitates from the bath, it clouds the solution, and by taking oxalate out of the iron complexes it starts the hydrolysis chain that ends in an unremovable stain.

Acids, beyond a trace. CAMEO classes the salt under Salts, Basic: it neutralises acids with the generation of heat. In this bath the relevant incompatibility is photographic rather than thermal — Ware’s texts say that more than a trace of acid inhibits the formation of platinum.

Alkalis, and anything that raises the pH. Above about pH 4 the iron(III) hydrolyses rather than complexing, and Ware’s warning about a high-pH first bath is unambiguous. Tap water carrying carbonate alkalinity does this slowly, over many sessions, which is the reason for the trace of oxalic acid in the first place.

Oxidising agents. CAMEO classes the salt as a weak reducing agent that can generate carbon dioxide with oxidisers. In practice this shows up as the decay of a dichromate contrast addition into oxalato-chromium(III) complexes, and it means that any oxidant put into this bath has a shelf life.

Metal trays and metal tongs. Bostick and Sullivan’s warning is written about the metal solutions — always use glass, never metal, because the platinum or palladium will plate out on the container — and the same argument applies with more force to a bath that Ware shows is thermodynamically capable of reducing platinum(II) on its own. Use plastic, glass or wood.

Waste

Collect it; do not tip it. Spent platinum or palladium developer is largely potassium oxalate carrying dissolved iron and traces of a platinum-group metal. Three reasons to bottle it: the oxalate is toxic; the noble metal is worth recovering and Ware directs that a spent first bath “should be saved for recovery of precious metals”; and the incompatibility with silver waste means it needs its own container anyway.

Never combine it with silver-bearing waste. The same rule as above, applied to the drain and the waste carboy. Follow the SOP for silver-bearing waste for the silver side and the general chemical waste SOP for this one.

On pH, which is the parameter a sewer code is most likely to name. Kodak’s J-52 publication gives 5.6 to 9.4 as the window effluent regulations most frequently set. A neutral oxalate developer is inside that window, which decides nothing: the reasons to collect this bath are the oxalate and the metal, not its pH. ILFORD’s advice to domestic users is to bottle wastes separately, label them, and take them to a household waste and recycling centre. Local regulation governs, and no jurisdiction-specific instruction is given here.

Troubleshooting

A weak, fibrous or granular image, with a poor maximum black. The classic platinotype failure, and Ware gives the mechanism: the soluble platinum salts washed out of the paper before they had reacted fully with the iron(II). The developer’s contribution is to lose the race — too dilute, too cold, or the print pulled early. Wall’s 1912 answer is his developer C, the same bath at double strength, “in cases where prints appear granular, or where intense blacks are wanted”; the historical answer was heat.

Streaks and uneven density across the sheet. Development begins on contact, so any part of the paper that meets the solution late develops late. One smooth pass, immersing one end first and drawing the sheet through, is the instruction every source gives. Brush away air bubbles the moment they form.

Brownish-grey fog in the highlights of a palladium print. Ware’s finding: the oxalate developer itself reduces some palladium(II), and humidity makes it worse. It is distinguishable from the yellow stain of residual iron by colour. The remedies in the literature are to use Willis’s citrate developer instead, or to clear in stronger hydrochloric acid, which etches the fog away — Ware’s dry note being that in American practice “the two procedural faults in US processing tend to be self-cancelling”.

A grey fog in the highlights of a print made on old paper. Not the developer’s fault. Ware attributes it to slow thermal decomposition of the ferric oxalate in stored paper, which is invisible in the dry sheet and becomes visible the moment the oxalate solubilises the iron(II) it has produced.

Random black spots, appearing only after development. The “plague of black spots”. Ware records that the cause is uncertain, that the spots are paper-dependent, that “they appear to increase in number and density with the vigour of the developer (i.e. oxalate produces a worse effect than citrate)”, and that they are more common with palladium. Two hypotheses are in play: microcrystals of iron(II) oxalate precipitated in the coating by reducing impurities in the paper, invisible until the developer dissolves them; or particles carried in from the developer itself, since filtering the developer helped and filtering the sensitiser did not. Filter the bath; try another paper.

A yellow stain in the highlights, appearing weeks or years later. Residual iron(III), hydrolysed and eventually crystallised as goethite. The developer’s share of the blame is small but real: a bath that has drifted alkaline hydrolyses iron rather than complexing it. The cure is on the clearing side and it has to happen before the print dries.

The developer has gone dark and prints are getting worse. Expected, up to a point: the colour darkens as dissolved metal accumulates, and trade opinion is divided on whether that is a virtue. Filter it, replace what has evaporated, and if it has been standing in the light, suspect precipitated metal in suspension.

Solid in the bottom of the storage bottle. If the bath was made at saturation, that is the reservoir working as intended. If it appeared later in a bath that was clear, suspect calcium oxalate from hard water or from a chalk-buffered paper.

Experiments

One: the strength series, which is the page’s central claim under test. Coat and expose six identical sheets from one negative and one sensitiser batch, then develop one each at 10, 15, 20, 25, 30 per cent w/v and at saturation, all at the same temperature and for the same time. Read maximum density and the highlight densities. Wall’s two published rules predict opposite-looking results — dilution gives more brilliant prints; a stronger developer is more rapid and softer — and they are the same rule seen from two ends. Plot density range against developer strength and find out where, for your paper, brilliance turns into a lost highlight. Record everything on the formula version record.

Two: the temperature series, and how much warmth a hot bath really buys. The same sheets, one developer strength, developed at 20, 30, 40 and 50 °C. Ware’s claim is that the colour shift is “only very slight” and that the real gain is completeness of development. Measure both: the maximum density tests the kinetic claim, and a visual comparison against a neutral grey tests the colour claim. This is the experiment that decides whether a hundred years of “hot for warm tones” is a real effect or a misattribution, and the course takes Ware’s side without having tested it. Keep the bath below 50 °C: the historical 60 to 77 °C is a scald risk with no compensating advantage in a modern palladium print.

Three: the pH question the sources leave open. Develop matched sheets in the same bath with 0, 0.1, 0.5 and 2 per cent added oxalic acid, measure the pH of each with a calibrated meter, and compare maximum density, highlight cleanliness and, after clearing and a fortnight of daylight, the yellow stain in the borders. The competing claims are set out under The mechanism: acid inhibits platinum formation, acid suppresses iron hydrolysis, and lower pH darkens the print. This is the smallest test that would settle which dominates in one darkroom.

Four: does an old bath really print better? Keep two bottles from the same batch, one used continuously for a season and one kept fresh, and print the same negative in each at intervals. Ware’s Stieglitz finding gives a mechanism by which the answer could be yes for the wrong reasons — the old bath is quietly plating another metal into your print. Test for it by printing pure palladium in a bath that has only ever seen palladium, and in one that has seen platinum, and comparing image colour.

Five: prove that the developer reduces nothing. Put a few millilitres of the palladium solution into a test tube of the developer, in the dark, with no iron and no paper anywhere near it, and leave it a week. Ware’s chemistry says a slow direct reduction is thermodynamically allowed and has been observed with palladium; Willis found out the hard way when his 1888 platinum-bearing developer turned black in the bottle. Whatever you see, this is the experiment that shows why the bath on this page has nothing in it but a salt.