Bostick and Sullivan platinum and palladium kits

Five boxes out of Santa Fe, one instruction sheet that covers all five, and a supplier who is franker than most about chemistry and almost silent about hazard.

The ordinary bought product in this formulary hides its composition behind a marketing description and discloses fragments of it on a safety data sheet, because the law obliges a maker who sells a classified mixture to classify it. Bostick & Sullivan do neither thing. Their instruction sheet tells you that the iron bottles are 27 per cent, that one of them carries potassium chlorate as a contrast agent, that palladium needs twice as much of it as platinum, that ferric oxalate is the only light-sensitive substance in the box, and that the noble metals will plate out onto a metal container — real chemistry, plainly written, in a document meant for beginners. And then the company publishes eighty-nine safety data sheets — the count its own site returned on 6 September 2026 — and not one of them is for a mixture. Every sheet in the library is a single-substance sheet for something sold by the jar. There is no sheet for a kit, no sheet for a premixed solution, and no sheet for anything in these boxes except the two powders.

It gets stranger where the platinum is. The library’s only sheet filed under a chloroplatinate heading turns out to be a GFS Chemicals sheet for COLOR STANDARD SOLUTION, APHA 250 — the platinum-cobalt standard of water analysis, 97.75 per cent water, 2.15 per cent hydrogen chloride, and 0.05 per cent of a platinum(IV) salt that is not the platinum(II) salt in the bottle. The sheet filed under Sodium Platinum Chloride (Na2) is a 2012 ScienceLab MSDS for “Platinum Chloride, 10%”, which its own composition section gives as chloroplatinic acid and water, and whose chemical-name field reads “Not applicable”. Neither document describes anything the company sells under those names.

So the concentrations in the table below were not read off a hazard document. Two of them were read off the maker’s instruction sheet, one off a store listing, and two off safety data sheets for the powders in the box. Five of the nine lines have no concentration at all, because the maker publishes none — and an empty cell on this page means the maker was silent, never that the page was.

☠Level D: this page describes a product and gives no procedure, and two separate course rulings put it here

Platinum. The course ruled on 4 September 2026 that students perform palladium and never handle platinum. Potassium tetrachloroplatinate(II) carries H334 — may cause allergy or asthma symptoms or breathing difficulties if inhaled — and HSE’s EH40 sets halogeno-platinum compounds at 0.002 mg/m³ as platinum with the Sen notation, the tightest workplace exposure limit anywhere in this course. Ware’s own health warning is the practical form of it: never touch the surface of sensitised paper, keep ungloved fingers out of the baths, and treat dust as the worst hazard. Three of the five kits on this page contain platinum and the other two contain a platinum salt used as an oxidant.

Potassium chlorate. Its encyclopaedia entry is at Level D on the rubric’s second criterion rather than on acute hazard — a safer modern route to the same result exists and this course teaches it. Bostick & Sullivan supply the chlorate already dissolved, which removes the weighing operation from the darkroom; it does not remove the reason.

So this page gives no procedure. It has no coating routine, no drying regime, no exposure time offered as an instruction, no processing sequence and no protective-equipment list, and the dilutions and development fields of its product record are deliberately empty. What it does instead is describe a document and a box: what the maker publishes, where the publishing stops, and what a reader holding one of these bottles is actually holding. The drop system itself is published, once, on the open entry that owns it — the three-solution drop system — where it is set out so that a formula met in a manual or on a shelf can be identified, and for no other purpose.

Two neighbouring entries do open. The palladiotype sits at Level B once the contrast has moved from the bottle into the negative, and the kallitype sensitiser runs the identical iron photochemistry with silver as the image metal and nothing oxidising in the box at all.

Bostick & Sullivan, Inc. — sold as kit

Components the maker discloses — from the safety data sheet, which classifies hazards and does not state a formula

Component	Concentration as the sheet gives it	Hazard codes
Ferric oxalatenamed on the sheet as Ferric Oxalate Solution #125 ml, and the light-sensitive half of the sensitiser. The maker calls it "the classic 27% solution", says it "is the same for platinum and palladium", and states that it is "the only light sensitive compound in your kit". No unit is attached to the percentage, no hydrate is named and no assay is given	27%
Ferric oxalatenamed on the sheet as Ferric Oxalate Solution #225 ml, the same iron at the same stated strength with a contrast agent dissolved in it, and the bottle whose proportion in the shot glass is the contrast control. The sheet says the platinum and palladium versions differ; the store offers one product, in five bottle sizes and no variants	27%
Potassium chlorateThe oxidant dissolved in Ferric Oxalate Solution #2, and the whole of the traditional kits' contrast control. Published as a percentage on one store page and as "a small amount" in the instruction sheet, which also says the platinum and palladium bottles carry different amounts — so the single figure and the two-strength statement cannot both describe what is sold	1.2%
Palladium Solution #310 or 25 ml according to kit, and the image metal of every kit except the platinum one. The maker gives it no strength, no formula and no CAS number anywhere; one kit listing calls it "Sodium Palladium Solution #3" and the stock code reads SODPAL, and that is the entire published evidence of what metal salt is in the bottle	—
Potassium tetrachloroplatinate(II)named on the sheet as Platinum Solution #310 or 25 ml, the image metal of the platinum kit and a minority partner in the traditional combo kit. Named once, in a parenthesis, and never quantified: the sheet's only identification of it is the phrase "the traditional Platinum Solution #3 (Potassium Chloroplatinite)", written while explaining what the Na2 solution replaces	—
Sodium Platinum Na2 Solution #310 ml, and the contrast agent of the current kits rather than an image metal in the ordinary sense: it is "a high contrast Platinum solution which is used in place of both the Ferric Oxalate Solution #2 and the traditional Platinum Solution #3", and the sheet forbids in bold using either of those with it. The two strengths are the only numbers published about it; the compound is never named	5% or 20%, according to kit
Potassium oxalate monohydratenamed on the sheet as Potassium Oxalate DeveloperA quart, or 32 fluid ounces, of made-up developer, at a strength the maker never states. It contains no developing agent: it dissolves the photoproduct rather than reducing anything, which is why the sheet can say the print develops in a few seconds and why the same bath is used again and again	—
EDTA Clearing AgentCAS 13235-36-4250 g of powder, used with the bisulfite at two tablespoons of each to a quart of water. It is the chelating half of the clearing bath: it holds the unreduced iron in solution so that it washes out of the paper instead of hydrolysing in it. This is one of only two lines in this table for which the company publishes a sheet describing the thing in the box	95-100 %	H302, H315, H319, H331
Sodium metabisulfitenamed on the sheet as Sodium BisulfiteCAS 7681-57-4250 g of powder, the reducing half of the clearing bath, working with the EDTA rather than after it. The kit labels it sodium bisulfite; the CAS number the maker prints beside that name, and the sheet it files for it, are both sodium metabisulfite's	≥ 97 %

Not disclosed. Bostick & Sullivan publish two concentrations and withhold every other number that would let a reader know what is in the bottle. The iron is given as 27 per cent in both ferric oxalate bottles and the hydrate, the assay and the basis of the percentage are never stated — which matters here rather than in the abstract, because the formula weights in circulation for ferric oxalate run from 375.76 to 483.84, so two bottles both honestly labelled 27 per cent can differ by a quarter in iron, and the only iron sheet in the company's own library is a Sigma-Aldrich sheet for the hexahydrate at the top of that range which the company nowhere says describes the "exclusive yellow powder" its instruction sheet claims the solutions are mixed from. The chlorate is worse than unstated: the instruction sheet says the #2 bottles "have different amounts of potassium chlorate for platinum and palladium, the palladium requires twice as much", and the store listing for the same bottle prints a single figure, 1.2 per cent, with no platinum or palladium variant offered, so the two documents cannot both be describing what is sold. The noble metals are not characterised at all. Neither the instruction sheet nor the store gives a strength, a formula, a CAS number or a hydrate for Palladium Solution #3 or for Platinum Solution #3; the sheet identifies the platinum in a single parenthesis as potassium chloroplatinite and identifies the palladium nowhere, the word "sodium" reaching the metal only through a stock code and one student kit's contents list, and the four product listings for the metal bottles consist of a stock number and a list of sizes and no other text at all. The Na2 solution is sold at 5 and 20 per cent as a "sodium platinum" with no compound named. The developer is a quart of potassium oxalate whose strength is published nowhere, the clearing bath is specified in tablespoons and described by its own maker as "not a critical measurement", and no pH is given for anything. There is no sensitometry of any kind: no exposure scale, no maximum density, no speed, no curve, and an exposure time given only as "3 to 8 minutes" under a sun lamp for a 4 × 5 negative. No coating volume is published in millilitres, no drop volume is stated, and the drop charts in the instruction sheet do not agree with the drop charts on the company's own Na2 sheet for the same kit. And the boundary that matters most is documentary. The company maintains a library of eighty-nine safety data sheets, every one of them a single-substance sheet for a chemical it sells loose, and there is not one sheet in it for any kit, any premixed solution or any mixture the company sells — so of the eight things in these boxes exactly two, the EDTA and the bisulfite, are covered by a sheet in the form in which they are shipped. Of the rest, the sheet filed under Potassium Chloroplatinate is a GFS Chemicals sheet for a water-testing colour standard that is 97.75 per cent water and 0.05 per cent of a different platinum salt, and the sheet filed under Sodium Platinum Chloride (Na2) is a 2012 ScienceLab MSDS for 10 per cent chloroplatinic acid, a compound in a different oxidation state that the company does not claim to sell. A reader who wants to know what has been classified about the liquids they will actually handle will find that the answer is nothing at all.

Nearest open formula. Three-solution drop system for platinum and palladium — It is the same system with three of its four numbers moved. The open entry publishes the three-solution drop method — plain iron, iron with an oxidant, noble metal, mixed drop by drop in a shot glass — with the composition, the mechanism and the arithmetic in full; these kits are that method sold as objects, with a stronger iron at 27 per cent instead of 20, with the chlorate dissolved in the bottle instead of weighed into it as a sachet, with the drop charts indexed by print size instead of by contrast grade, and with EDTA and bisulfite in place of the citric acid clearing bath. What the box adds is that the four difficult purchases have already been made and the two solutions already mixed. What the open entry adds is everything the box withholds: the identity of the material behind the words "ferric oxalate", the disagreement between three suppliers who call the same bottle 20, 25 and 27 per cent, what an oxidant actually does to a tonal scale as against the maker's phrase "contrast agent", the redox chemistry that explains why palladium needs twice as much of it, and the two rulings that put both entries at Level D. Neither is a route this course opens: the open entry publishes a composition and no procedure, and this page publishes what the maker publishes and no procedure.

Purpose

To let somebody make a platinum or palladium print without owning a balance, without buying four difficult chemicals in small quantities, and without dissolving anything.

The sheet states that offer in its second paragraph and states it accurately: “This kit provides the essential chemicals in standardized form. No mixing is necessary, you can begin printing almost immediately without the time consuming weighing and dissolving steps.” Of the eight things in the largest box, six arrive as liquids already made up and two as powders that go into water by the tablespoon. There is nothing to weigh in any kit, which is not true of the competing supplier’s boxes, and nothing to dissolve except the two clearing powders.

The five kits are one system with the metal and the contrast agent changed, and the sheet covers all five at once because that is all the difference amounts to.

	Platinum 25 ml	Classic Palladium 25 ml	Traditional 35 ml	Na2 Combination 35 ml	Na2 Digital Negatives
Ferric Oxalate #1	25 ml, 27 %	25 ml, 27 %	25 ml, 27 %	25 ml, 27 %	25 ml, 27 %
Ferric Oxalate #2	25 ml (pt)	25 ml (pd)	25 ml (pd)	—	—
Palladium Solution #3	—	25 ml	25 ml	25 ml	25 ml
Platinum Solution #3	25 ml	—	10 ml	—	—
Na2 sodium platinum	—	—	—	10 ml at 20 %	10 ml at 5 %
Potassium oxalate developer	1 quart	1 quart	32 oz	32 oz	32 oz
EDTA / sodium bisulfite	250 g each	250 g each	250 g each	250 g each	250 g each
Yield claimed	25–30 8 × 10	none	40 8 × 10	none	35–40 8 × 10
Strength of the metal	not published	not published	not published	20 %	5 %
Safety data sheet for the kit	none exists	none exists	none exists	none exists	none exists

Two rows are the whole design argument. The metal strength row is empty for every traditional kit, which is the single most consequential silence in the maker’s literature: a printer who wants to compare these bottles with anybody else’s, or to make up a replacement, has nothing to compare. And the Na2 column replaces the chlorate bottle rather than filling it — the sheet is explicit that the Na2 solution “is used in place of both the Ferric Oxalate Solution #2 and the traditional Platinum Solution #3”, and forbids in bold using either with it. The contrast agent has stopped being a chlorate and become a platinum salt.

The sheet’s own account of the difference is a sales claim rather than a measurement, and is recorded here as one: the Na2 method “produces a Platinum print with superior fine line detail and separation, and an unmatched resistance to solarization and grain texture”, and “allows a much wider range of negatives to be printed, while at the same time using much less of the costly Platinum solution”. No supporting figure of any kind is offered for any part of that.

⁇The sheet's contents list does not agree with the store, or with its own drop charts

The instruction sheet says the ferric oxalate #2 bottle is in the “traditional kit or classic palladium kit only”. The store listing for the Platinum Printing Kit – 25 ml includes “25 ml – Ferric Oxalate Solution #2 Contrast Booster”, and the sheet’s own fifth drop chart, headed Platinum Printing Kit - 25ml, has a column for “Ferric Oxalate #2 (pt)” and uses it at every print size. Three documents from one company, two of which are the same document.

The same sheet lists “(1) 10ml Platinum solution” as though there were one, then gives it three identities in the same line — Platinum #3 in the traditional kit, 5 % Na2 in the digital negatives kit, 20 % Na2 in the combination kit — while the store sells the platinum kit with 25 ml of Platinum Solution #3. Neither reading is wrong for every kit and neither is right for all of them.

This page records the disagreements and does not resolve them. Resolving them would mean choosing which of the maker’s documents to believe, and nothing in the evidence supports the choice.

Recommended uses

What the sources record, and what this course does with them. The course recommends none of the printing uses below: it recommends the reading. The section exists because a reader who meets one of these bottles — on a shelf, in a bequest, in a workshop store, in a kit bought second-hand — has to know what the numbers on it were for.

Identifying a bottle. This is the use the course has for the page. A 25 ml bottle labelled Ferric Oxalate #1 is, on this supplier’s published composition, a 27 per cent solution — a strength no other supplier uses, which makes the label diagnostic on its own. Ferric Oxalate #2 is the same solution with potassium chlorate in it, published once as 1.2 per cent. A bottle marked Palladium #3 or Platinum #3 carries no published strength at all and must be treated as an unknown; a bottle marked Na2 is a platinum solution at 5 or 20 per cent that behaves as an oxidant rather than as an image metal, and must not be mixed with either of the two bottles it replaces. A 250 g packet marked Sodium Bisulfite from this supplier is, on the supplier’s own CAS number and its own safety data sheet, sodium metabisulfite.

Reading a kit sheet critically, which is a skill this formulary exists to teach. This is unusually good practice material precisely because the chemistry in it is sound and the documentation around it is not. The sheet explains what the chlorate does and never says how much there is. It gives a shelf life that the store contradicts. It prints drop charts that the company’s own second sheet contradicts. It lists contents that its own tables contradict. A reader who can find those four without being told where they are has learned most of what Rule 7 asks for.

Understanding what the Na2 method actually changed. Richard Sullivan’s substitution is the most interesting thing in this entry: a contrast control that is not an oxidising salt of chlorine but an inert platinum(IV) complex, added in ones and twos of drops rather than as a proportion of the iron. Why an inert platinum(IV) salt can oxidise iron(II) at all without becoming image metal is set out on the open entry, together with the reason Willis’s own platinum(IV) experiments of 1872 failed.

Not for a negative made on purpose for the process. This is what the sources converge on and it is the position the course takes. Ware’s conclusion is that with modern negative-making, especially by digital means, contrast agents “become unnecessary when a correctly calibrated negative is made”. A correction curve on a digital negative buys the same contrast for nothing, without the graininess and without an oxidiser on a domestic bench — and the irony is that the maker’s own best-selling kit is sold for digital negatives and still ships a contrast agent.

Not as a first alternative process. The sheet says so itself, in the only piece of advice in it that is unambiguously about the reader’s safety rather than the print’s quality: “We recommend a beginning platinum/palladium printer be under the supervision of an experienced platinum/palladium printer.” A reader who cannot yet coat an even sheet should be practising on the classic cyanotype sensitiser or the kallitype, where a spoiled sheet costs pennies.

Not as a way of learning what a platinum print is. Read the sheet and you will know the drop counts and not the chemistry. It never says that the developer contains no developing agent, never explains why development is instantaneous, never says that the image is metal precipitated among the paper fibres, and never once uses the word siderotype.

When another formula is preferable

• The palladiotype, with the contrast in the negative. The course’s route to a noble-metal print, and the one that removes both Level D grounds at a stroke: no platinum and no oxidant. Willis’s potassium oxalate developer and the sodium citrate developer are its two published developers, and the clearing sequence is the step that decides whether the print survives.
• The three-solution drop system, if the question is what is actually in this box. The open entry this page compares against. It carries the composition, the drop tables, the mechanism and the arithmetic in full, and it is where the two suppliers’ numbers are set beside each other rather than averaged.
• Ware’s print-out platino-palladiotype, if the objection is to the traditional method itself. It replaces the ill-characterised ferric oxalate with ammonium iron(III) oxalate, takes its contrast from controlled humidity and from the metal ratio rather than from an oxidant, and prints out, so the exposure can be judged by eye. Every complaint this page makes about the specification of the iron falls out with the change of salt.
• The kallitype sensitiser, to learn the craft on. Identical iron photochemistry, identical coating, drying, exposing and clearing skills, silver instead of the noble metal, and a fraction of the cost per sheet. Its clearing sequence is the same problem in a cheaper material.
• The Pizzighelli and Hübl formulations of 1886, if the question is where the drop system came from. The four-bottle ancestor, in which the chlorate is Solution C.
• The Photographers’ Formulary kits, if the comparison wanted is between the two surviving suppliers. They sell the same three-bottle system at 20 per cent iron against this supplier’s 27, weigh the chlorate into a sachet — 0.18 g for platinum, 0.26 g for palladium — rather than dissolving it, organise the drop tables by contrast grade rather than by print size, clear in citric acid rather than EDTA, and publish a full contents table with a quantity against every line. On composition they publish more; on hazard, neither publishes anything for the kit. The numbers are not transferable in either direction, because the iron differs by more than a third.
• The Bostick & Sullivan cyanotype kits, for the same company doing better. That sheet asks for gloves, calls its own solution poisonous, and gives a keeping time. This one does none of those things, for a considerably more hazardous box. The Formulary’s cyanotype kit is the third corner of the same comparison.

What the maker publishes

A moderate amount, in two documents that do not entirely agree, and the useful way to read them is by what kind of statement each thing is.

The contents of each box, in a list. The instruction sheet gives the generic list — “(1) 25ml ferric oxalate #1 solution; (1) 25ml ferric oxalate #2 (pt or pd) solution; (1) 25ml Palladium #3 solution; (1) 10ml Platinum solution; (1) 32oz B&S potassium oxalate developer; (1) 250g EDTA clearing agent; (1) 250g Sodium Bisulfite (for clearing)” — and the six store listings give the specific ones, kit by kit, which is where the differences in the table above come from.

Two concentrations, and only two. The iron is 27 per cent, stated twice in the instruction sheet and twice again on the store listings. The chlorate is 1.2 per cent, stated once, on the listing for Ferric Oxalate Solution # 2 and nowhere else. That is the entire quantitative disclosure of what is in these bottles.

What the chlorate is for, and why palladium needs more of it. “This solution is also 27% and has a small amount of potassium chlorate added as a contrast agent. The #2 solutions have different amounts of potassium chlorate for platinum and palladium, the palladium requires twice as much chlorate to achieve contrast.” The competing supplier states the same fact as two weights, 0.18 g and 0.26 g, which is nearer half again than twice — the two houses agree on the direction and not on the factor, and neither publishes a measurement behind it.

Which bottle is light-sensitive, and only which bottle. “Ferric Oxalate is the only light sensitive compound in your kit. Limit its exposure to sunlight and intense UV.” That is correct, it is the single most useful sentence in the document for somebody setting up a working space, and it is stated once without elaboration: no safelight colour, no wavelength, no illumination level, no drying-room rule.

A real chemical warning about the metal bottles. “Be very careful not to contaminate them with ferric oxalate or any other chemicals and always use glass containers, never metal as the platinum or palladium will plate out on the container.” A supplier explaining electroless deposition in one clause, to a beginner, without using the word.

A shelf life, and a handling condition attached to it. “It has a shelf life of approximately 1 year, but will last longer if refrigerated, however it must be at room temperature when used.” The metal solutions, by contrast, “will last a very long time; they do not oxidize or age”, and refrigeration “is not necessary” for them.

A developer described as effectively immortal. Reused indefinitely, replenished for evaporation rather than exhaustion, filtered through a coffee filter when sludge accumulates, and expected to darken as dissolved metal builds up in it: “Many printers swear the older the developer, the better. Other printers prefer to start with fresh developer every once in a while.” No exhaustion test is offered and none is needed in the sheet’s account, because in this process the developer is a solvent rather than a reductant.

A clearing bath given in tablespoons. “Use 2 tablespoons of each to 1 quart of water. This is not a critical measurement; less will just take longer to clear, more will clear faster.” Followed by the one diagnostic in the document that a reader can actually apply: “If the print shows any yellowing in the highlights, it is not being properly cleared.”

A negative specification. “The negative should have a density range of 1.35 to 1.50.” Plus film and developer recommendations — Ilford HP5 and FP4, developed in D-19, Rollo Pyro or Pyrocat-HD — which is the sheet reaching outside its own catalogue to name products it does not sell.

Five drop charts, indexed by print size. Not by contrast grade, which is the competing supplier’s organising principle and the historical one. Bostick & Sullivan’s tables answer “how much do I need for an 8 × 10” and leave contrast to be adjusted from the starting point: “These charts are starting points for making a print with each different kit. Adjust as needed to fit your personal printing style, negatives, and paper selection.”

Three safety sentences. They are set out in full under Safety below, because their content matters less than what is not among them.

What is not disclosed

The identity of the iron. Twenty-seven per cent of what? The instruction sheet says the solution is mixed “from our exclusive yellow powder”; the store listing says “fresh from powdered Ferric Oxalate”; neither says which powder, what its formula weight is, what hydrate it is, or how the percentage was computed. This is not a pedantic objection. Ware records formula weights in circulation for material sold as ferric oxalate running from 375.76 to 483.84, so two solutions both honestly labelled 27 per cent can differ by nearly a quarter in the iron they actually deliver. The only iron sheet in the company’s own library is a Sigma-Aldrich sheet for iron(III) oxalate hexahydrate, CAS 166897-40-1, molecular weight 483.84 — the heaviest of the candidates — and the company nowhere says that this is what its yellow powder is. It is filed under Ferric Oxalate Powder, which is a product the company also sells; it is not filed against the kits, and it is not a sheet for a 27 per cent solution.

Whether the acidity is controlled, and how. The competing supplier states that its iron solution “contains a slight excess of oxalic acid” and explains why. Bostick & Sullivan say nothing whatever about oxalic acid, about pH, or about the acidic and potassium double-salt forms of ferric oxalate. The excess acid sets the pH, keeps the iron soluble and decides how much of the image prints out during the exposure; a reader of this sheet alone would not know the question existed.

How much chlorate, in which bottle. The instruction sheet says the platinum and palladium #2 bottles carry different amounts. The store sells one product called Ferric Oxalate Solution # 2, offers it in five bottle sizes and no other variant, and prints one number for it: 1.2 per cent. Both statements are the maker’s, they are three sentences apart on the site, and they cannot both describe what is shipped. No weight is published anywhere, for either version.

The strength of the platinum. The strength of the palladium. The compound in either bottle. This is the largest silence and it is complete. The instruction sheet identifies the platinum once, in a parenthesis written to explain what Na2 replaces — “the traditional Platinum Solution #3 (Potassium Chloroplatinite)” — and identifies the palladium nowhere at all. The store listings for Palladium Solution #3, Platinum Solution #3, Sodium Platinum “Na2” 5% Solution and Sodium Platinum “Na2” 20% Solution consist, in every case, of a stock number and a list of bottle sizes, with no descriptive text of any kind. The word sodium attaches to the palladium only through the stock code SODPAL and one student-kit listing that calls it “Sodium Palladium Solution #3”.

⁇An inference, marked as one

A “sodium palladium” solution sold for this process is, on everything the course knows about the chemistry, sodium tetrachloropalladate(II), Na₂PdCl₄ — the salt Ware sets out as the standard palladium sensitiser stock, and the salt the competing supplier names in words. The company sells cesium and lithium variants of the same #3 bottle alongside it, which is the pattern of a family of alkali tetrachloropalladates. The identification is a course inference from the product name and the chemistry, not a statement by the maker, and it is why the palladium line in the table above carries no chemical link. The same applies to Na2, which the drop-system entry reads as sodium hexachloroplatinate(IV) on the same footing and marks the same way.

The strength of the developer. A quart of made-up potassium oxalate, with no percentage, no saturation statement, no specific gravity and no pH. The competing supplier at least publishes a weight and a volume from which a strength could in principle be worked out; here there is nothing.

Any pH, anywhere. Not for the sensitiser, the developer or the clearing bath.

Any sensitometry, anywhere. No exposure scale, no maximum density, no speed, no characteristic curve, no reciprocity note. The exposure guidance is “3 to 8minutes” — set as printed — under “a suntan type mercury vapor lamp” for a 4 × 5 negative, with dense negatives at “a half hour or longer”, and the lamp “3 to 4 inches above the paper”.

Any coating volume in millilitres. Only drops, and the drops themselves are never measured: no drop volume is stated in either document, so nothing in the maker’s literature converts a drop chart into a quantity. Ware’s independent figures — 24 to 36 cm³/m², about 1.4 cm³ for an 8 × 10 on cotton paper — are the only way to check a drop count at all, and they belong to a different account of the process.

Which drop chart is current. The instruction sheet and the company’s own Na2 Digital Negatives Kit Drop Count Sheet both give counts for the same kit and do not agree. For an 8 × 10 the sheet in the box says 20 drops of palladium #3, 20 of ferric oxalate #1 and “4 or 5” of the 5 per cent Na2; the drop count sheet, which describes itself as “Calibrated to B&S Photoshop Curves”, says 20 of ferric oxalate #1, 15 of palladium #3 and 5 of Na2. That is a third less palladium in one document than in the other, for the same print on the same kit. At 5 × 7 the gap is wider still.

And the whole of the hazard documentation. The company’s safety data sheet library contains eighty-nine sheets, every one for a single substance sold loose, and not one for any kit, any premixed solution or any mixture it sells. Of the eight items in the largest box, exactly two — the EDTA and the bisulfite — are covered by a sheet describing the thing in the form in which it ships. Of the remaining six: the iron sheet is for a catalogue powder, not for the 27 per cent solution; the chlorate sheet is for crystalline solid, not for a 1.2 per cent solution; the potassium oxalate sheet is for the crystalline monohydrate, not for a quart of made-up developer; there is no sheet of any kind for the palladium solution; the sheet filed under Potassium Chloroplatinate is a colour standard for water testing; and the sheet filed under Sodium Platinum Chloride (Na2) is a thirteen-year-old MSDS for a different platinum compound. A reader who does the responsible thing — look up the safety data sheet before opening the box — will find that for six of the eight bottles and packets there is nothing to find.

Disclosed components

Nine lines, of which two come from a hazard document and seven do not. The distinction is the page’s, not the maker’s, and it is worth stating before the details: a concentration read off a safety data sheet has been through a regulatory classification, and a concentration read off a shop page has not.

Ferric Oxalate Solution #1, 25 ml at 27 per cent. The light-sensitive half of the sensitiser, identical in both traditional and Na2 kits, and the one bottle the sheet warns about by name: “Ferric Oxalate is the only light sensitive compound in your kit.” Iron(III) in an oxalate complex is what actually responds to ultraviolet; the noble metal in the other bottle is inert to light and waits in the paper for the iron(II) the exposure makes. Twenty-seven per cent is the highest figure any of the three suppliers in this formulary publishes — the Formulary’s bottle is 20 per cent and Ware’s own standard solution is 25 — and because the hydrate behind the percentage is not stated, the three numbers are not directly comparable. Nothing in the maker’s literature says whether the solution carries free oxalic acid, which every independent account of ferric oxalate treats as essential.

Ferric Oxalate Solution #2, 25 ml at 27 per cent. The same iron at the same stated strength, with the contrast agent dissolved in it. In the traditional kits this bottle is the whole of the contrast control: the printer changes the ratio of #1 to #2 in the shot glass and changes nothing else. Because both bottles are 27 per cent, the iron delivered to the paper stays constant across every ratio and only the oxidant varies — which is the elegant part of the design, and is the same idea the Formulary implements with a weighed sachet. The bottle is supplied in a platinum version and a palladium version, and the sheet’s instruction for a mixed print is to use the palladium one: “If you are printing with a mixture of both platinum and palladium, you should use the ferric oxalate #2 solution for palladium. Just use less #2 for the proportion of platinum solution.”

Potassium chlorate, 1.2 per cent in Solution #2. The oxidant. It re-oxidises the iron(II) that light has made, so that only the most strongly exposed areas retain enough iron(II) to reduce the noble metal — which raises contrast by suppressing the low values rather than by extending the high ones. Ware’s account of what this does to a tonal scale is not the maker’s phrase “contrast agent”: it truncates the high values rather than compressing the scale uniformly, and it costs graininess and a false sparkle. The Getty atlas adds that the chlorate leaves no analysable residue in a finished print but is responsible for a visually detectable patchiness in the platinum image. The 1.2 per cent is published on one store page; the instruction sheet says the two versions of the bottle differ and gives no number; the company’s own sheet for the substance is for POTASSIUM CHLORATE, CRYSTAL, REAGENT, ACS, signal word Danger, Oxidizing solids Category 1, “May cause fire or explosion; strong oxidizer” — a sheet for the jar, not for the bottle.

Palladium Solution #3, 10 or 25 ml, strength not published. The image metal of four of the five kits. Nothing about it is disclosed: not the compound, not the strength, not the hydrate, not a CAS number. The maker’s only substantive statement about this bottle is shared with the platinum one and is about handling — it does not oxidise or age, it must not be contaminated, and it must be kept in glass because the metal plates out on metal. The nearest sheet in the company’s library is for palladium(II) chloride, Fisher P6-5, classified Skin Sensitization Category 1 with “May cause an allergic skin reaction” — a solid the company sells loose, and not what is in this bottle.

Platinum Solution #3, 10 or 25 ml, strength not published. Named once, in the parenthesis “(Potassium Chloroplatinite)”, which is the older name for potassium tetrachloroplatinate(II), K₂PtCl₄. Ware records that this complex aquates slowly in solution — a half-time of about 2.4 hours at 20 °C against 0.17 for the palladium analogue — which is the chemical reason platinum prints and palladium prints behave differently in the developer even when everything else is equal. The maker publishes none of that and no strength. The library’s only chloroplatinate sheet is for a water-testing colour standard containing 0.05 per cent of the platinum(IV) salt, which is neither this compound nor this concentration nor this product.

Sodium Platinum Na2 Solution #3, 10 ml at 5 or 20 per cent. A platinum solution used as a contrast agent, not as an image metal: “This is a high contrast Platinum solution which is used in place of both the Ferric Oxalate Solution #2 and the traditional Platinum Solution #3 (Potassium Chloroplatinite). DO NOT use traditional platinum solution #3 or Ferric Oxalate solution #2 with this solution.” The two strengths are published; the compound is not. The 20 per cent version comes with the only dilution instruction in the document — “20% Na2 can be diluted with distilled water if it seems to strong”, set as printed — and its drop counts are given as “1 or less” at every print size, which is a control with one usable position.

Potassium Oxalate Developer, one quart, strength not published. It contains no developing agent. Oxalate ion complexes the iron(II) the exposure has made and carries it to the noble-metal salt sitting beside it in the paper, and the metal comes down where the iron(II) is: that is why “the print will develop immediately” and why “development is complete within a few seconds”. It is also why the bath does not exhaust in the ordinary sense and can be replenished for evaporation and filtered rather than replaced. The maker’s own sheet for the substance covers POTASSIUM OXALATE, MONOHYDRATE, CRYSTAL, REAGENT, ACS, CAS 6487-48-5, classified Acute toxicity Oral 4 and Dermal 4, Skin corrosion/irritation 2 and Serious eye damage/eye irritation 2A — the crystal in the jar, not the quart in the box.

EDTA Clearing Agent, 250 g, at 95–100 per cent on its own sheet. The chelating half of the clearing bath. Unreduced iron(III) is chemisorbed onto the cellulose after development and hydrolyses in place above pH 4; left there it goes on to goethite, which no ordinary bath will remove and which is the yellow stain in the highlights of a badly cleared print. A strong chelator holds the iron in solution long enough for it to wash out. This is one of only two components in the box for which the company publishes a sheet describing what is actually shipped, and the sheet is for EDTA Tetrasodium Salt, CAS 13235-36-4, signal word Danger, with H302, H315, H319 and H331 — toxic if inhaled — against a store listing that calls the same powder a “Safe, easy to use clearing agent”. Tetrasodium EDTA is strongly alkaline in solution, which is worth knowing next to a process whose stain chemistry turns on pH, and the sheet’s inhalation classification is a dust classification: 250 g of powder is a dust operation.

Sodium Bisulfite, 250 g, at 97 per cent or more on its own sheet. The reducing half of the clearing bath, working alongside the EDTA rather than after it: it keeps the iron in the more soluble iron(II) state while the chelator carries it away. The name on the packet and the substance in it are not the same thing. The maker’s store listing prints “CAS No. 7681-57-4” — which is sodium metabisulfite, not sodium bisulfite, CAS 7631-90-5 — beside the synonym “Sulfurous Acid Monosodium Salt”, which is bisulfite’s. The sheet the company files for it is titled Sodium (Meta)bisulfite and is an Esseco USA sheet for SODIUM METABISULFITE, CAS 7681-57-4, at 97 per cent or more, stamped with a Univar despatch header naming Bostick & Sullivan as the customer. Two of the three documents say metabisulfite and the packet says bisulfite. The course records the CAS number as decisive and the label as loose, and notes that the two are not interchangeable by weight: they differ in how much sulfur dioxide equivalent a spoonful delivers, which is exactly the property the clearing bath is using.

📓Why five of the nine lines have no concentration

The schema for this page kind allows a concentration only where a document prints one. Five lines are blank because the maker publishes no figure: the two metal solutions, the developer, and — in the sense that matters — the chlorate’s two supposed strengths, of which only one number exists. Nothing here has been computed, interpolated, or taken from a third party’s analysis of a competing product. An empty cell is a fact about the maker’s literature.

Behaviour

Everything happens in a shot glass. The maker’s own instruction is to “count out the number of drops for all printing solutions into the same small plastic or glass container (about the size of a whiskey shot glass)”, which is also how the conservation literature describes the process being done. The sensitiser has no keeping time because it is made one print at a time; nothing is mixed that is not used within minutes.

Coating is fast, and the sheet says so bluntly. “Quickly pour the coating solution across the center of the paper and rapidly spread it as evenly as possible. Soon the wetness of the paper will dull and you won’t be pushing puddles around the paper. Stop brushing at this point.” Then ten minutes for the solutions to soak in, then drying to completion — “An exposed image will have splotchiness in the areas that were not completely dry before exposure”, which is the maker naming its own commonest defect.

Exposure is by contact, under ultraviolet, and is not published as a number worth quoting. Sun, a sunlamp, or a UV unit; 3 to 8 minutes for a 4 × 5 with a sunlamp; dense negatives half an hour or more. The sheet offers no way to calibrate any of that and no test strip procedure.

Development is over before it has begun. “The print will develop immediately. Development is complete within a few seconds. Most printers leave the print in the developer for 1 to 2 minutes.” This is the single behavioural fact that most distinguishes a siderotype from a silver print, and the sheet reports it without explaining it. The reason is in The mechanism below.

The developer’s temperature is a picture control, and the only one the maker names. “The developer can be used from room temperature up to the boiling point; the print color and contrast will vary with the temperature.” Which direction is not stated. The Getty atlas supplies it: hotter development gives smaller metal particles and warmer tones.

Clearing is the step that decides whether the print survives, and it is the step the sheet is vaguest about: one or two baths, five minutes each, two tablespoons of each powder to a quart, “not a critical measurement”, a dozen 8 × 10s per fresh batch, water hardness a variable, and a single visual test — any yellowing in the highlights means it has not worked. Then half an hour of gently flowing water.

Reuse is asymmetric. The sensitiser is one-shot by construction. The clearing bath is thrown away after a session. The developer is kept indefinitely, replenished for evaporation and filtered when sludge appears, and is expected to darken as dissolved metal accumulates. The product record’s reuse field records the developer’s behaviour because it is the only reuse the maker instructs.

Image characteristics

Colour is not a control the maker offers. Nothing in either document says what colour any kit gives. The only lever named is developer temperature, and its direction is left to the reader. What the conservation literature records is that platinum reads neutral to cool grey, palladium warm-black to sepia, and mixtures fall between according to the ratio, with paper, humidity and development temperature all modulating the result.

Contrast is a ratio, and the maker publishes the ratios without saying what they do. In the traditional kits the printer changes the proportion of ferric oxalate #2; in the Na2 kits, the number of Na2 drops. The published charts run from one part in eleven to three in twenty for the chlorate bottle, and from one drop to five for the Na2, and the sheet says only “Adjust as needed to fit your personal printing style”. There is no grade scale, no measured density range for any setting, and no worked example.

The negative carries the specification instead. “The negative should have a density range of 1.35 to 1.50. This will give a print with a full rich tonal range.” That is the closest thing to sensitometry in the maker’s literature, and it is a statement about the input rather than the output. Ware puts the unmodified platinotype sensitiser’s exposure scale at about 2.0, which is the number the whole drop-counting apparatus exists to shorten.

Nothing visible during the exposure is the image. The faint brownish figure that appears under the negative is photochemically generated ferrous oxalate, not metal; the metal arrives in the developer, within seconds. A printer judging exposure by the visible print-out is judging the wrong thing, and the sheet never mentions the distinction.

The sheet’s one aesthetic paragraph is about softness, and it is the most characterful writing in the document: “Platinum and palladium have the unique ability to make soft prints with a strength and character. Printers at the turn of the century used this quality to their advantage and produced beautiful luminous soft images. Prints like this are not often being made today. Like avocados, the soft print is an acquired taste.”

The mechanism

The chemistry is a siderotype: iron does the photography and the noble metal does the picture.

Light reduces iron(III) to iron(II) in an oxalate complex. Ultraviolet drives an internal electron transfer in the iron(III) oxalato complex; one oxalate is destroyed as carbon dioxide and the iron is left in the +2 state. Nothing that looks like an image has been made — the faint brown that appears is ferrous oxalate, not metal.

The iron(II) then reduces the noble metal. The iron oxalato couple sits at about +0.02 V, the tetrachloroplatinate couple at +0.73 V and the tetrachloropalladate at +0.62 V, so the reduction is strongly favoured in both cases. It does not happen in the dry paper at any useful rate; it happens when the developer arrives and mobilises the iron(II). That is the whole explanation of a development that is “complete within a few seconds”: the developer is not reducing anything, it is delivering a reductant already made and already in place. Ware’s electron micrographs put the resulting platinum particles at 15 to 25 nm, ellipsoidal, within the surface cellulose fibres — which is why the image sits in the paper rather than on it, and why the surface texture of the sheet is part of the picture.

The chlorate reverses the first step selectively. Potassium chlorate in Solution #2 oxidises iron(II) back to iron(III). In lightly exposed areas, where little iron(II) was made, the oxidant wins and nothing is left to reduce metal; in heavily exposed areas enough survives. So contrast rises by removing the bottom of the scale rather than by extending the top, which is the technical content of Ware’s objection that an oxidant truncates rather than compresses. Palladium needs more of it than platinum, and the maker says so without saying why.

⚗An inference the maker does not make

The reason palladium needs more oxidant follows from the two numbers above: the palladium couple sits about 0.11 V below the platinum one, so palladium is the easier metal to reduce, and more of the photolytically generated iron(II) has to be destroyed before the low values stop printing. The maker states the practical rule — twice as much chlorate — and offers no reason at all; the competing supplier states it as “palladium is less sensitive to contrast control with potassium chlorate than is platinum”, which is the observation rather than the cause. The redox account here is this course’s reading of Ware’s data, not a claim made by either supplier, and it is set out at length on the open entry.

Na2 substitutes an inert platinum(IV) salt for the chlorate, and the same logic applies with one extra step. A platinum(IV) complex can oxidise iron(II) thermodynamically, but it is kinetically far too inert to be reduced to metal in the few minutes a development takes — so it acts as an oxidant and never becomes image silver’s equivalent, image platinum. That is why it can be a contrast agent rather than a second image metal, and why the sheet insists it replaces both the chlorate bottle and the traditional platinum bottle rather than joining them. That identification of Na2 as a hexachloroplatinate(IV) is an inference this course marks as an inference; the maker names no compound.

The developer is a solvent for the photoproduct, not a reductant. Oxalate complexes the iron(II), mobilises it, and lets it reach the metal salt beside it. Potassium oxalate is the classical choice; the sodium citrate and ammonium citrate developers of the same family do the same job with different tone and speed, and are on their own entries.

Clearing is a race against hydrolysis. After development the paper still holds iron(III), chemisorbed to the cellulose. Above about pH 4 it hydrolyses in place, and the hydrolysis product goes on eventually to goethite, which no bath removes. EDTA holds the iron in solution as a chelate; the bisulfite keeps it reduced and therefore more soluble; between them they get it out of the sheet before it sets. The maker’s “if the print shows any yellowing in the highlights, it is not being properly cleared” is that chemistry seen from the tray. Calcium carbonate in a buffered paper or in hard water makes the same problem worse by stripping oxalate from the iron complex and raising local pH.

The nearest open formula

The three-solution drop system is the same method, published as a formula rather than as a shopping list, and the comparison is the useful part of this page.

They agree on the architecture completely: plain iron, iron with an oxidant, noble metal, combined drop by drop at the moment of coating, exposed by contact under ultraviolet, developed in an oxalate, cleared, and washed. They differ in four places, and every difference is a number this supplier moved.

	These kits	The open entry
Ferric oxalate	27 %, hydrate unstated	20 % as the Formulary publishes it, 25 % as Ware’s standard, with the polymorphism and the 375.76–483.84 range of formula weights set out
Contrast agent	chlorate dissolved at 1.2 % in bottle #2, or Na2 at 5 or 20 %	chlorate weighed — 0.18 g for platinum, 0.26 g for palladium — with the arithmetic and the historical origin
Drop charts	indexed by print size, “starting points”	indexed by contrast grade, five grades, with the metal held constant across all of them
Clearing	EDTA and bisulfite, two tablespoons each per quart	citric acid, and the historical hydrochloric sequence, with the pH chemistry of why it matters
Noble metal	strength not published	20 % platinum and 15 % palladium as published, with the salts named and their aquation rates given

What the box adds is that the four difficult purchases have already been made, the two solutions have already been mixed, and nothing needs weighing. That is a real service and it is the whole commercial reason these kits exist. What the open entry adds is everything the box withholds: what the words “ferric oxalate” denote, why three suppliers’ bottles at 20, 25 and 27 per cent are not comparable, what an oxidant does to a tonal scale, why palladium needs more of it, and the two rulings that put both entries at Level D.

Neither is a route this course opens. The formula this course does teach in this family is the palladiotype, with the contrast put into the negative instead of into the bottle.

Safety

⚠Level D, on two independent grounds, and neither of them is the palladium

Platinum, for irreversibility. Potassium tetrachloroplatinate(II) carries H334 — may cause allergy or asthma symptoms or breathing difficulties if inhaled — and HSE’s EH40 sets halogeno-platinum compounds at 0.002 mg/m³ as platinum with the Sen notation, against 5 mg/m³ for platinum metal: a limit 2,500 times tighter for the compound than for the element. Ware’s health warning is the practical form of the same fact — never touch the surface of platinum-sensitised paper, never put ungloved fingers in the processing solutions, treat dust inhalation as the worst hazard.

Potassium chlorate, for the availability of a better route. Signal word Danger, Oxidizing solids Category 1, “May cause fire or explosion; strong oxidizer” on the supplier’s own sheet. Dissolving it at 1.2 per cent removes the dry-handling operation that drives most of that classification, which is a real improvement over a weighed sachet; it does not remove the reason the course prefers a calibrated negative.

None of the above is a control, and this page offers none. Under the Level D policy, a classification says nothing about how hard the controls would be to arrange; it says that a page has no business assuming they are in place.

The instruction sheet’s safety section, in full. It is three sentences long and this is all of it: “ALL CHEMICALS SHOULD BE USED WITH CAUTION AND KEPT OUT OF REACH OF CHILDREN.” — “When using a hair dryer to dry your paper, tiny particles of emulsion may be blown into the air. If inhaled, the dust could be harmful. It is advised to wear a dust mask when drying prints with a hair dryer.” — “We recommend a beginning platinum/palladium printer be under the supervision of an experienced platinum/palladium printer.” A fourth sentence appears later, under the ferric oxalate: “DO NOT STORE WHERE A CHILD CAN REACH IT.”

What is not in those seven pages. The words glove, eye, ventilation, disposal, first aid, poison, toxic, hazard and safety data sheet do not occur once. There is no eye-protection instruction, no glove instruction, no first-aid measure, no emergency contact, no ventilation requirement, and no disposal instruction of any kind — the word drain appears exactly once, in “Drain off excess developer off the print”. By way of comparison, the same company’s cyanotype instruction sheet asks for gloves for handling, coating and processing, and says in terms that “the solution should be considered poisonous”. This is the more hazardous box and the thinner document.

The dust warning that is there is the right warning in the wrong place. A hair dryer aimed at a sensitised sheet aerosolises platinum and palladium salts, and the sheet is correct to ask for a dust mask. But dust is also what happens when 250 g of EDTA and 250 g of bisulfite are tipped into a quart jug, and the EDTA’s own sheet classifies it Acute toxicity Inhalation (Dusts/Mists) Category 3, H331 — toxic if inhaled, with “May cause respiratory irritation” and “May cause drowsiness or dizziness” alongside. The store listing for the same powder calls it “Safe, easy to use”. The two statements are from the same company about the same jar.

The bisulfite carries the hazard most likely to catch somebody out. Its sheet — for sodium metabisulfite, whatever the packet says — records that it “may cause severe and possibly fatal allergic reactions if inhaled or swallowed by some asthmatics and other ‘sulfite-sensitive’ individuals”, that it “reacts with acids to form toxic and irritating sulfur dioxide gas”, and that it releases sulfur dioxide above 150 °C. Sulfite sensitivity is common enough that a darkroom shared with other people should treat it as foreseeable, and the acid reaction matters here specifically: a printer who has read about citric acid clearing and decides to combine the two approaches in one tray is generating sulfur dioxide on purpose without knowing it. Signal word DANGER, “Causes serious eye damage”.

The developer is the bath a printer’s hands go near, and it has a systemic classification. Potassium oxalate on the supplier’s own sheet: Acute toxicity Oral 4 and Dermal 4, Skin corrosion/irritation 2, Serious eye damage/eye irritation 2A — harmful if swallowed, harmful in contact with skin, causes skin irritation, causes serious eye irritation. Oxalate is also an anticoagulant and a systemic poison, which is a property of the ion rather than of any particular salt. The sheet in the box says nothing about it at all, and the sheet in the library is for the crystal rather than the quart. Hands do not go in the developer. See gloves and first aid.

The iron is classified and barely documented. The only iron sheet the company publishes gives Acute toxicity Oral 4 and Dermal 4 with H302 + H312, records that it contains no substances with occupational exposure limit values, and then answers every question in its toxicological and ecological sections with “no data available”, closing with the admission that the properties “have not been thoroughly investigated”. Absence of data is not evidence of safety.

The eye hazard nobody labels. Four of the classified substances in this box carry a serious eye classification — the EDTA at Category 2, the bisulfite at Category 1, the potassium oxalate at 2A, the chlorate at 2B — and the instruction sheet never mentions eyes. Splash protection is not optional around any of these trays, whatever the document omits.

And the sentence that is not there. The competing supplier ends its safety section by transferring all risk to the buyer and offering a refund, which this course records without approval. Bostick & Sullivan do not do that, because they do not have a safety section to end.

Storage

What the sources record.

The iron is the perishable one and the two documents disagree about how perishable. The instruction sheet gives “approximately 1 year, but will last longer if refrigerated, however it must be at room temperature when used”; the store listings give “approximately 1-2 years before it should be tested or discarded”. Both say the solution is mixed shortly before it ships, so the clock starts at despatch on either account, and neither says what “tested” means. The sheet is candid about the underlying difficulty: “As with all photosensitive materials, it is very difficult to know precisely how long the material will last.”

Refrigeration, with the condition attached. Cold extends the iron’s life and the solution must be back at room temperature before use — a two-part instruction of which the second half is the one people forget, and a cold sensitiser mixed with a room-temperature metal is a coating that behaves differently for reasons nobody looks for.

The metal solutions keep and their enemy is contamination. “These solutions will last a very long time; they do not oxidize or age.” Glass only, never metal, and a separate dropper for each bottle: the noble metals plate out onto a more reactive metal surface, which rules out metal vessels, metal spatulas and metal droppers throughout. A dropper used in both an iron bottle and a metal bottle reduces metal in the bottle rather than in the paper, and ruins the most expensive liquid in the room.

Both iron bottles look identical and one of them is a different solution. Ferric Oxalate #1 and Ferric Oxalate #2 are the same colour at the same strength, and the difference between them is a dissolved oxidant that nothing about the bottle reveals. Anything decanted from either is a container nobody can identify: write the substance name, the percentage and the date on it — see labelling a container — and if you have inherited one that is unmarked, the unlabelled container procedure exists for exactly that.

Light, for one bottle out of six. Ferric oxalate is the only light-sensitive substance in the kit and the instruction is to “limit its exposure to sunlight and intense UV”. No safelight is specified, no illumination level, and no rule for storing coated paper — which is the gap that matters most, because a dried sensitised sheet left out is both a fogging problem and, for platinum, a dust problem.

Two 250 g jars of powder in a room full of open trays. Neither powder has a published keeping time. Tetrasodium EDTA and sodium metabisulfite are both hygroscopic in practice; a jar that has taken up water is a jar whose two tablespoons are no longer two tablespoons, which matters more for the bisulfite, whose reducing capacity decays as it takes up moisture and oxidises.

Incompatibilities

Iron against the metal bottles, in the bottle. The maker’s own warning and the most expensive mistake available: “Be very careful not to contaminate them with ferric oxalate or any other chemicals.” Iron(II) reaching a platinum or palladium solution reduces metal in the bottle. One dropper per bottle, and a brush washed thoroughly — the sheet makes that point too, for a different reason: “any solution left on the brush will be exposed and contaminate your next print.”

Noble metals against metal containers. Platinum and palladium plate out onto more reactive metals, so glass throughout and no metal trays. The sheet extends the rule to development — “Do not use metal trays for developing the print as this will adversely affect the print” — which is the same electrochemistry acting on a tray instead of a bottle.

Bisulfite against acid. The clearing powder’s own sheet: it “reacts with acids to form toxic and irritating sulfur dioxide gas”. The kit’s own clearing bath is not acidic, but the competing supplier’s is citric acid, the historical sequence uses hydrochloric, and a printer combining accounts from two suppliers in one tray is the specific way this goes wrong. See incompatibilities.

Chlorate against combustible material. A 1.2 per cent solution is a very different proposition from a jar of crystals, and dissolving the oxidant is the safest thing about this supplier’s design. The classification on the company’s own sheet — Oxidizing solids Category 1, “May cause fire or explosion” — belongs to the solid, and is the reason a spill that is allowed to dry on paper or cloth should not simply be left there.

Oxalate against heavy metals and against calcium. Oxalate developers are conventionally kept away from heavy-metal salts, and in this process the developer meets the noble metal on purpose, in the paper, which is the point; what should not happen is a metal solution finding its way into the developer jar. Calcium is the other half of the same chemistry and matters in the wash: calcium carbonate in a buffered paper or hard water strips oxalate from the iron complex, raises local pH, and starts the hydrolysis that ends in an unremovable stain. This is why the clearing instruction mentions water hardness, and it is why an alkaline-buffered paper is a poor choice even though the sheet never says so.

EDTA against the assumption that it is inert. Tetrasodium EDTA is a strong chelator and strongly alkaline in solution. It will strip metal ions out of things that are not the print — including, over time, from equipment — and its alkalinity is the wrong direction for a process whose stain chemistry begins above pH 4. That the bath works anyway is because the chelate is strong enough to win; it is not because the pH is convenient.

Cross-contamination between processes. Ferricyanide is not in this kit, but it is in the cyanotype kit from the same supplier and it is the standard spot test for ferrous contamination in ferric oxalate. Any acid near potassium ferricyanide liberates hydrogen cyanide, and Prussian blue appearing in an iron solution is the sign of exactly the contamination the test looks for. See cross-contamination between alt processes.

Waste

The bulk of it is iron, and it is the clearing baths. One or two trays of EDTA and bisulfite carrying the unreduced iron out of the paper, thrown away after a session on the maker’s own instruction, plus half an hour of wash water behind them. The standing arrangement is on the disposal page and in general chemical waste, and the first thing to establish is what your local authority accepts.

The instruction sheet gives no disposal advice at all, for any of the eight items, in seven pages. That is recorded here as a fact about the document rather than as a criticism of any particular omission: there is nothing to disagree with.

The developer is the stream to keep out of a drain on its own merits. A quart or more of oxalate solution, reused for years, carrying dissolved noble metal — with an acute oral and dermal classification on the supplier’s own sheet and an ecological section that has nothing to say. Absence of data is not evidence of safety.

Spent metal is worth recovering rather than discarding, and this is the one waste stream in the formulary with a positive value. Platinum and palladium in an old developer, in a spent clearing bath, or on a spoiled print are the most expensive thing in the room, and a refiner will take them. The maker’s observation that the developer’s “color will darken due to dissolved metal accumulating in it” is a description of a metal-bearing waste stream that nothing in the sheet identifies as one.

The bisulfite is the one that must not be acidified in a sink. Sulfur dioxide is generated by the reaction the sheet warns about, and a shared sink is where a bisulfite bath and somebody else’s acid stop bath meet.

Troubleshooting

These are the maker’s own faults and remedies where it gives them, and the course’s reading of the chemistry where it does not. None of it is a procedure and none of it authorises the process.

“The highlights are yellow.” The print has not been cleared. The maker’s remedy is direct: “Increase the concentration of the clearing agent or increase the time in the bath.” Behind it is the hydrolysis above pH 4 that ends in goethite — so the fault is time-critical, and a print left overnight before clearing may not be recoverable. Hard water lengthens the clearing time; the sheet says so. See platinum stain from incomplete clearing and iron-silver highlights that will not clear.

“The image is splotchy in patches.” The maker’s own diagnosis, and the only defect it names: “An exposed image will have splotchiness in the areas that were not completely dry before exposure.” See mottled coating from humidity and coating mottle and drying marks.

“The print is grainy and the highlights sparkle.” Ware’s account of what an oxidising contrast agent does at high concentration. Less ferric oxalate #2, and more contrast in the negative.

“The whites are dirty and the print is flat.” Iron(II) already present in the sensitiser before the exposure. The maker gives no test; the competing supplier publishes one, using two crystals of potassium ferricyanide in a little water — a blue colour means iron(II), and the deeper the blue the worse the bottle. The relevant fact here is the shelf life: a year on the instruction sheet, one to two years on the store listing, and no way offered to tell which.

“The developer has gone dark and there is sludge in it.” Expected, on the maker’s own account, and not a fault: filter it through a coffee filter and a plastic funnel, top up for evaporation, and carry on. The sludge is metal-bearing and should not go down a drain.

“The palladium bottle has a deposit on the glass.” Contamination, and the sheet’s own warning covers it: iron in a metal bottle, or a metal container. There is no recovery instruction and there should not be an improvised one.

“The two sheets tell me different drop counts.” They do. The instruction sheet and the Na2 drop count sheet disagree by about a third on the palladium for an 8 × 10, and the course has no basis for choosing. A reader who owns one of these kits owns the more specific document — the one calibrated to the supplier’s own correction curves — and should note that the other exists.

“I have a bottle and no label.” The unlabelled container procedure. Note that on this supplier’s system an unlabelled 25 ml bottle of yellow solution is one of two things that look identical and behave differently, and there is no test the course offers that distinguishes them.

Experiments

Reading experiments. None of them involves the kit, and none of them is a printing procedure.

Enumerate the safety data sheet library and check it against the catalogue. The company publishes eighty-nine sheets; list them, then list what the company sells. The exercise is a useful one for any supplier, and the general result is more interesting than this particular instance: a substance library is not a product library, and the gap between them is where a bought mixture lives. Then find the two sheets in this library that are filed under names that do not match their contents.

Compare three suppliers’ ferric oxalate on the page. Twenty per cent, twenty-five, twenty-seven — three numbers for the same bottle from three sources, none of which states a hydrate, against a range of formula weights from 375.76 to 483.84. Work out what could be concluded about the relative iron content of the three, and then write down honestly what cannot. This is the single best exercise in the formulary for understanding why Rule 1 forbids computing what a source did not print.

Find the four internal contradictions in seven pages. Contents against drop chart; shelf life against store; drop chart against drop chart; single chlorate figure against two-strength statement. All four are findable without any chemistry at all, and finding them is the skill this formulary is trying to teach.

Track the word that is missing. Read the sheet and note every point at which a safety instruction would belong in a document of this kind, then check whether one is there. Then read the same company’s cyanotype sheet, which asks for gloves. The comparison is the finding.

Trace one CAS number. The packet says sodium bisulfite; the store page prints CAS 7681-57-4; the filed sheet says sodium metabisulfite. Look up both substances, work out what the difference means for the strength of a clearing bath made by the tablespoon, and decide which of the three documents you would act on. Then read sodium bisulfite and sodium metabisulfite and see whether you were right.

Read the open entry beside this one. The three-solution drop system publishes the composition, the tables and the mechanism. Read it and then re-read the instruction sheet, and notice how much of what you now understand was in the box all along and unexplained.