Talbot's calotype iodised paper

Every other formula in this book is trying to make a sheet of paper respond to light. This one is trying to stop it. Talbot’s iodised paper is a photographic material whose published design requirement is that light must have no effect on it — “scarcely sensitive to light”, he says, and then, as though it were the ordinary thing to boast of, “it may be kept for any length of time without spoiling or undergoing any change”.

The sensitivity is added later, by somebody else’s page. What this page makes is the storable blank.

The silver wash — "100 grains of crystallized nitrate of silver in six ounces of distilled water" — the silver, brushed on one side only and dried before the sheet meets any iodide

Ingredient	Quantity	Form the source specifies
Silver nitrate	38 g	3.8 per cent w/v, 0.224 mol/L — Ware's conversion of Talbot's 100 grains in six imperial fluid ounces, scaled by the course to a litre. Talbot specifies the crystallized salt and distilled water
Water	to make 1000 mL	Talbot's own volume is six fluid ounces, about 170 mL; the litre is the course's scaling of a strength, not a batch size he specified. He names distilled water, which matters, because tap water carries chloride.

The iodising bath — "500 grains of that salt dissolved in one pint of water" — the halide, in deliberate excess, which converts the silver and then kills the sheet's sensitivity

Ingredient	Quantity	Form the source specifies
Potassium iodide	57 g	5.7 per cent w/v, 0.343 mol/L — Ware's conversion of Talbot's 500 grains in one imperial pint, scaled by the course to a litre. Talbot names plain water here, not distilled
Water	to make 1000 mL	Talbot's own volume is one imperial pint, about 568 mL. He specifies distilled water for the silver solution and does not for this one.

Used in this order — Talbot's iodising, as published in 1841 — the whole of it

1. The silver wash — "100 grains of crystallized nitrate of silver in six ounces of distilled water" — Washed on with a soft brush, on one side only, and that side marked so it can be found again. Then dried cautiously at a distant fire, or left to dry spontaneously in a dark room. The whole operation is best done in the evening by candlelight.
2. The iodising bath — "500 grains of that salt dissolved in one pint of water" — two or three minutes — The sheet is dipped when dry or nearly so. Ware states what the time is doing: it is long enough for the iodide to be in chemical excess and to precipitate silver iodide within the fibres, and short enough that the iodide does not begin re-dissolving what it has just made.
3. Water — "Then dip it into a vessel of water" — a rinse, which carries away the potassium nitrate of the reaction and the unreacted iodide. Ware records that prolonging it makes the silver iodide fall off the sheet. The paper is then dried lightly with blotting-paper and finished at a fire, "which will not injure it even if held pretty near".

"Take a sheet of the best writing paper, having a smooth surface, and a close and even texture. The watermark, if any, should be cut off, lest it should injure the appearance of the picture. Dissolve 100 grains of crystallized nitrate of silver in six ounces of distilled water. Wash the paper with this solution, with a soft brush, on one side, and put a mark on that side whereby to know it again. Dry the paper cautiously at a distant fire, or else let it dry spontaneously in a dark room. When dry, or nearly so, dip it into a solution of iodide of potassium containing 500 grains of that salt dissolved in one pint of water, and let it stay two or three minutes in this solution. Then dip it into a vessel of water, dry it lightly with blotting-paper, and finish drying it at a fire, which will not injure it even if held pretty near: or else it may be left to dry spontaneously. All this is best done in the evening by candlelight."

The drying steps are not baths and the schema cannot hold them; they are in the notes above and in the Mixing section, where they belong, because a sheet dried at a fire while loaded with an oxidiser is a safety statement as well as a procedural one. Nothing is exposed at any point in this sequence, which is why `development` is empty.

☠Why this page carries no procedure, and why the reason is not the one you would expect

This is a historical-study entry. Notice first what the reason is not. It is not hazard: at Level B, a 3.8 per cent silver nitrate solution and a 5.7 per cent potassium iodide bath are milder than the chemistry the course teaches for real under salted paper. And it is not, as it was on the photogenic drawing paper page, that the quantities are a modern reconstruction. Talbot published both quantities exactly, and Hunt reprinted them verbatim three years later. The numbers on this page are as well attested as any in the formulary.

There are three reasons, and the third is the interesting one.

The course does not make calotypes. The process atlas says so plainly: the calotype is taught here as chemistry and history, and no written part of the curriculum makes a calotype negative. A formula page cannot unilaterally open a practice the curriculum has not decided to teach.

This sheet is one stage of three and is useless alone. An iodised sheet takes no picture. It becomes a photographic material only when it is excited with gallo-nitrate of silver, and it yields nothing until it is developed with more of the same and then fixed. Publishing one stage as a procedure would be publishing a third of a process.

And Talbot’s contemporaries testified, in print, that his directions could not be followed. This is the reason worth carrying away. George Cundell opened his 1844 paper on calotype practice by observing that had Talbot “thought fit to publish directions for the details and refinements of his process, as minute and explicit as those given by Daguerre, his invention, it is probable would now have stood in a very different position”; Lady Elizabeth Eastlake wrote in 1857 that the directions, “though sufficient for his own pre-instructed hand”, were too vague to help anyone else. Ware’s technical reading agrees from the chemistry side: the timing of both stages was critical, and either misjudgement gave poor negatives. A complete set of quantities is not the same thing as a procedure, and this formula is the clearest case in the course of the difference. What is missing is not the numbers. It is everything that goes with them.

If you want to put silver halide into a sheet of paper today, the course’s actionable routes are the plain salting solution with the salted paper sensitiser at Level B, or, for a developed image with a fixed speed, the silver chloride contact emulsion.

Purpose

To precipitate silver iodide inside the surface fibres of a sheet of writing paper, with the iodide in excess, so that the sheet is chemically finished, photographically dead, and will keep in a drawer until it is wanted.

Three things are being achieved at once, and only the first is obvious.

Silver iodide has to be made in place. It is the most insoluble of the three photographic halides, so it cannot be dissolved and brushed on; the only way to get it into paper is to put a soluble silver salt in first and a soluble iodide in afterwards, and let the two meet inside the sheet. That much this formula shares with every salted paper and every albumen paper of the century, and with the precipitation that begins a modern emulsion.

The excess halide is the design. In Talbot’s earlier paper the silver was in gross excess and the halide was starved, and the result printed out in the sun. Here the arrangement is reversed, and the reversal is not an accident of the order of coating: Ware states that the two-to-three-minute immersion “ensured that the iodide was in chemical excess”. A silver halide crystal sitting in a surplus of its own halide is what section 23.6 of Argyronomicon calls “fixed” silver halide, and fixed silver iodide “is completely insensitive even to direct sunlight”. Talbot did not have that vocabulary. He had the observation, and he built on it.

And the third purpose is the one that outlived the process. Because the sheet is inert, the work of preparing it can be separated in time from the work of making a photograph. Talbot says the iodising “may be performed at any time” and that the remaining part “is best deferred until shortly before the paper is wanted for use”. That division — a stable material made in advance, sensitised at the moment of use — is the shape every manufactured photographic material has taken since, and reading it as an ancestry rather than as a coincidence is this course’s interpretation rather than anything Talbot claimed. It is the reason this page exists in a book otherwise uninterested in a sheet that does nothing.

Recommended uses

Historical, all of them, and each is a use of the finished iodised sheet rather than of the two solutions.

As the substrate of every calotype negative Talbot made from 1841, and of most of those made by everyone else for the next fifteen years. Ware’s Table 1 enumerates the whole family of paper-negative processes and finds three distinguishable methods of iodising underneath all of them: Talbot’s, which is this one; Furlong’s; and Guillot-Saguez’s. Everything else that varies — waxed or plain paper, strong or dilute exciting solution, physical or chemical development — varies on top of a sheet made by one of those three.

As the material of the first paper photographs that could record a shaded subject at all. The gain the calotype brought was about a hundredfold over printing out, and it came from the latent image. But a latent image needs a halide that will hold one without spending itself, and it needs a sheet that is the same on Tuesday as it was on Monday. The iodised paper supplies both.

As Talbot’s own direct-positive experiment, which is the strangest use and is worth knowing because it shows how the same bath does two opposite jobs. In the specification Hunt quotes, a sheet of excited calotype paper is exposed to daylight until it visibly browns, then dipped in the same 500-grains-to-the-pint iodide solution, which apparently removes the discoloration; put in the camera and developed, it gives a positive. The iodide bath that makes a sheet insensitive at the start of the process is used here to switch a sheet off in the middle of it.

As the parent of a stored, prepared, dry material. By the 1843 patent Talbot was already pushing the idea further: Hunt’s appendix lists an iodo-gallic paper, made by washing iodised paper with gallic acid, which “will keep in a portfolio” and is made sensitive later by a wash of silver nitrate. The direction of travel from this formula to a boxed dry plate to a roll of film is one straight line.

And, in this course, as the counterexample that makes printing out intelligible. Read this page next to the photogenic drawing paper: the same two reagents, the same sheet of writing paper, opposite proportions, opposite order, opposite behaviour. Nothing else in the formulary isolates the variable so cleanly.

When another formula is preferable

The honest answer is “always, for anything practical”, so the list below is arranged by what you actually want to do.

• To make a print you can hang up, the plain salting solution and the salted paper sensitiser, fixed with a thiosulphate fixer. That is a printing-out process and this is not; an iodised sheet will not print out at all, which is the entire point of it.
• To understand printing out, the photogenic drawing paper page, which is this formula’s mirror image and has to be read with it.
• To make a developed image on paper with published, repeatable behaviour, the silver chloride contact emulsion — a halide in a binder, on a support, with a speed you can measure and a fixer that works cold.
• To see what an iodide addition does in a modern material, the iodobromide double-jet emulsion. Silver iodide never left photography; it stopped being the whole halide and became a few mole per cent of it, and the reasons are on that page.
• To iodise a paper with fewer ways to go wrong, and staying inside the calotype family: Furlong’s 1843 single-solution method or Guillot-Saguez’s 1847 silver-free one. Both are described under Variants, neither is published here as a formula of its own, and Ware’s verdict on Furlong’s is that it “at a stroke, overcame the two iodizing problems inherent in Talbot’s method”.
• To fix a silver iodide image, anything but this bath. Excess iodide leaves an image that fades, and Talbot’s own retrospective verdict on iodide as a preserving agent was that it “must be considered as not sufficiently certain to be retained in use”. A plain hypo fixing bath is the answer, warm.

Mixing

Two solutions, one sheet, and they must never meet in a vessel. That is the structural fact of this formula and the reason the entry has two solutions rather than one. Silver nitrate poured into potassium iodide gives a curd of silver iodide in the beaker and nothing else; the same two liquids, one dried into a sheet of paper and the other met a few minutes later, give a photographic material. Everything in the procedure exists to keep them apart until the paper brings them together.

Talbot’s own order, and every word of it is load-bearing. Take the best writing paper, smooth and close-textured. Cut off the watermark, “lest it should injure the appearance of the picture” — the watermark is a thinning of the sheet and would print. Brush the silver solution on one side only, and mark that side so you can find it again, because from this moment the sheet has a front and a back and the marked side is the one every later solution must go on. Dry it cautiously at a distant fire, or spontaneously in a dark room. When dry or nearly so, dip it in the iodide for two or three minutes. Dip it in water. Blot it. Dry it at a fire, which at this stage “will not injure it even if held pretty near”. Do all of it in the evening by candlelight.

Why candlelight, when the finished sheet is insensitive. Talbot does not explain, and the course will not put words in his mouth, but the chemistry gives a reading and it is offered as a reading. The sheet is only insensitive after the iodide bath. Between the silver wash and the dip it is paper loaded with silver nitrate — Ware’s account of why such a paper darkens at all points at the cellulose and the mill’s sizing as the reducing partner — and that is exactly the condition Wedgwood and Davy worked in forty years earlier. The candlelight protects the intermediate, not the product.

🧮From grains and pints to grams per litre, with the arithmetic shown

Talbot writes in the British imperial measures of his day, and the schema has no unit for either. Every figure in the table above is a conversion, so here is the whole of it.

1 grain = 64.79891 mg exactly

The grain, defined against the pound

1 fl oz = 28.4131 mL; 1 pint = 20 fl oz = 568.261 mL

The imperial fluid ounce and pint

100 gr = 6.4799 g in 6 fl oz = 170.478 mL, which is 3.80 % w/v

The silver wash

500 gr = 32.399 g in 1 pint = 568.261 mL, which is 5.70 % w/v

The iodising bath

Those are Ware’s two figures, 3.8 and 5.7 per cent w/v, recovered to three significant figures. The agreement settles a question the conversion could otherwise have got wrong by five per cent: Talbot’s ounce is the imperial one of 28.41 mL, not the American one of 29.57 mL, which is what you would expect of a British gentleman writing seventeen years after the imperial system was defined, and is now checked rather than assumed. A per cent w/v is grams of solute in 100 mL of finished solution, so scaling to a litre is multiplication by ten: 38 g/L and 57 g/L. Molarity follows from the relative molecular masses on the silver nitrate and potassium iodide pages:

38 g/L ÷ 169.873 g/mol = 0.224 mol/L

Silver nitrate

57 g/L ÷ 166.003 g/mol = 0.343 mol/L

Potassium iodide

Both match Ware’s parenthetical molarities exactly, which is the second check.

Now the number Talbot never states and the whole formula turns on. Converting the silver in the wash needs one iodide ion per silver ion, so a bath at chemical equivalence with a 0.224 mol/L silver solution would be 0.224 mol/L of potassium iodide:

0.224 mol/L × 166.003 g/mol = 37.2 g/L, a 3.7 % bath

Iodide that would exactly consume that silver

Talbot’s bath is 5.7 per cent — about one and a half times equivalence, litre for litre. And that understates it badly, because the comparison is not litre for litre. The sheet carries away a thin film of the silver solution and is then immersed in a reservoir of the iodide, so the real excess is set by the volume of the dish and is larger by whatever factor you care to imagine. Ware’s summary needs no arithmetic at all: the immersion “ensured that the iodide was in chemical excess”.

Set that beside the earlier formula and the reversal is stark. Photogenic drawing paper met a salting bath of about 1.1 per cent w/v sodium chloride, 0.19 mol/L, with a silver bath of about 20 per cent w/v, 1.18 mol/L. Equivalence there would have been a 3.2 per cent silver bath, so Talbot ran roughly six times the silver the halide could use. Two years later, with the same two elements in his hands, he ran the halide in excess instead. Same chemist, same paper, opposite sign — and the two papers do opposite things.

The paper is an ingredient and the schema cannot say so. Ware identifies Talbot’s stock as John Whatman’s “Turkey Mill”, a gelatin-sized rag paper. The sizing is not inert packaging: section 23.3 of Argyronomicon assigns it the job of retaining the sensitiser in the surface fibres, which is what keeps the silver iodide where the image will be instead of letting it wander through the sheet, and it influences the colour and stability of the silver eventually formed. English mills sized with gelatin and continental mills with starch, which is why the same two solutions gave different papers in London and in Paris — and why Guillot-Saguez’s free-iodine indicator, described under Variants, turned violet on English paper and blue on French.

What the water dip is for. The reaction leaves potassium nitrate behind, soluble and useless, and the sheet is also wet with unreacted potassium iodide. The Image Permanence Institute names both: the coats “created the desired precipitate of silver iodide as well as a byproduct of potassium nitrate that was then rinsed off the paper”. The course has no encyclopaedia entry for potassium nitrate, which is why it is not linked; it is a spectator here and leaves in the rinse.

And the drying is where the safety statement lives. A sheet of paper carrying an oxidising salt, held near a fire, is a fire risk in the plainest possible terms, and Talbot’s instruction to dry the silver-loaded sheet only “at a distant fire” while allowing the iodised sheet to be “held pretty near” is, whether or not he knew why, exactly the right distinction: by the second drying the nitrate is gone.

Behaviour

It does nothing, and that is the specification being met. Talbot’s phrase is “scarcely sensitive to light”, and Ware’s is that the sheet “was quite stable to light and could be stored for long periods because of its insensitivity”. The development array on this page is empty because there is nothing to develop: no exposure belongs to this stage, no time, no temperature, no printing out.

Its two failure modes are both timing, and they pull in opposite directions. This is the behaviour a practitioner actually had to manage, and Ware states it as a pair. Too long in the iodide bath and the silver iodide begins to re-dissolve. Too long in the water rinse and the silver iodide falls off the sheet. Talbot’s “two or three minutes” and his single dip are the window between the two, and he gives no tolerance on either.

The first of those was understood, chemically and correctly, within three years. Hunt in 1844 credits Dr Ryan with the warning and states the mechanism outright: if the paper is left too long, “the iodide of silver will be dissolved, that salt being soluble in an excess of iodide of potassium”. Hunt’s own practice was more cautious than Talbot’s — “simply passing the paper through the solution appears to answer every purpose effectually”.

Sheet-to-sheet variation was the dominant complaint of the period. Ware’s judgement is that in 1843 “the difficulties of most calotypists stemmed from an almost complete absence of clear, detailed directions”, and Furlong’s single-solution iodising is presented as the thing that let calotypists “iodize their papers with a degree of certainty that previously came only through trial and error”. That is a statement about this stage in particular, and it is the strongest evidence there is that the formula above is harder to execute than it looks.

It keeps. Not for weeks — indefinitely, on the maker’s own published claim, and the mechanism under “The mechanism” makes the claim credible rather than merely optimistic. Set against the excited sheet’s few hours, that is a difference of at least three orders of magnitude in keeping, achieved by withholding the silver.

Image characteristics

There is no image, and the page would be dishonest to imply otherwise. What the sheet has is a colour and a uniformity, and both matter later.

The colour is “a uniform pale yellow coating of iodide of silver”. That is Talbot’s own identification of the compound, made from its colour, and it is correct: silver iodide is the yellow one of the three photographic halides, where silver chloride is white and silver bromide pale cream.

The uniformity is the only property that can be inspected before use, and it is not cosmetic. Hunt makes the consequence explicit in 1844, writing about printing from the finished negative: the light “has to permeate a piece of paper, the yellow tint of which, offers considerable interruption to those rays which are active in producing chemical change; we must therefore, be exceedingly careful to preserve the prepared sheet of as pale and uniform a tint as possible”. Every unevenness brushed in at this stage becomes an unevenness in every positive ever printed from the negative — and because the absorbing species is yellow, what it absorbs is precisely the blue and ultraviolet the printing paper needs.

The yellow survives the whole process unless it is deliberately removed, and Talbot spent his second patent on the removal. Hunt’s appendix lists the method: plunge the finished picture into a bath of hyposulphite of soda in ten times its weight of water, heated nearly to boiling, for about ten minutes, then wash in warm water and dry. A hot bath, because silver iodide is only dissolved by thiosulphate with difficulty. Ware’s own reading of surviving objects is that the step remained optional, and that plenty of calotypes carry their yellow to this day.

The negative the sheet eventually carries is grey rather than warm, and Ware gives the reason on the physics rather than the chemistry: development builds silver particles “relatively much larger than those of the print-out process”, so a calotype has a neutral grey appearance where a photogenic drawing has the reds and purples of colloidal silver. The Photographic Materials Group, cataloguing the family from the other end, records the range across photogenic drawings, salt prints and calotypes together as warm brown to cool grey, with the image layer being silver “deposited directly in the paper support” and no binder at all.

The mechanism

The precipitation, and why it has to happen inside the paper

Silver iodide is insoluble in water — the most insoluble of the three halides, which is the property the whole formula is built on and also the property that makes it impossible to coat. So it is assembled in place, from two soluble salts that arrive separately:

Ag⁺ + I⁻ → AgI

Ware's equation: silver iodide precipitated wherever the two ions meet

Written with the salts as the sheet actually meets them, it is a double decomposition with a soluble spectator left over:

AgNO₃ + KI → AgI + KNO₃

The whole of the iodising reaction, potassium nitrate included

The potassium nitrate has no job. It is dissolved out in the water dip, and its removal is the only purpose that dip serves that is not simply washing away surplus iodide.

The geometry matters as much as the stoichiometry. The silver went on with a brush, on one side only, so the silver iodide forms in the fibres near that surface rather than throughout the sheet — Ware notes that this surface coating is what confines the image to one side, and that photographers who instead immersed the paper in every bath produced sheets on which it is hard to tell which side the image was made on.

Why iodide in excess makes the sheet dead

This is the part Talbot could observe and not explain, and it is the reason the page is worth reading.

Section 23.4 of Argyronomicon starts from a fact of analytical chemistry: a precipitated silver halide is non-stoicheiometric at its surface, and which ion it adsorbs there depends on which ion is in excess in the liquid it was precipitated from. Ware names the two cases. Where silver ions are in excess, the crystal carries adsorbed silver ions and a net positive surface charge: “sensitized” silver halide. Where the halide ion is in excess, it carries adsorbed halide and a net negative surface charge: “fixed” silver halide.

Photolysis itself is the same in both cases:

2 AgX + light → 2 Ag + X₂

The net photolysis, for any halide X

What differs is what happens next, and everything depends on it. In a sensitized crystal, the positive surface charge attracts the photoelectrons outward, silver specks grow at the surface free of the lattice, and — crucially — the liberated halogen meets water and free silver ions and is disposed of:

X₂ + H₂O + Ag⁺ → AgX + HOX + H⁺

Ware's halogen acceptor: water and silver(I) acting together

That is a halogen acceptor at work. It takes the halogen out of circulation so it cannot re-oxidise the silver just made, and it recycles some of it into fresh silver halide. Without one, Ware’s first listed outcome applies: “the halogen will react with the surface photolytic silver, reversing the reaction above and re-forming the halide.”

In a fixed crystal there is no acceptor, and worse than that, the surface charge is now negative and repels the photoelectrons back into the interior. Ware’s own three-line summary of the visible result is the most useful piece of chemistry on this page:

• Fixed silver chloride changes from colourless to dull violet, from colloidal silver formed inside the crystal.
• Fixed silver bromide responds less, becoming pale grey, “because the bromine disproportionation is very slight in the absence of excess silver ions”.
• Fixed silver iodide “is completely insensitive even to direct sunlight, remaining a pale yellow colour”, because iodine released by photolysis is not disproportionated at all, so the back-reaction is the most efficient of the three.

⚗Which is the answer to a question this page has not asked yet: why iodide?

Talbot had three halides available and knew about all of them. He had used chloride for photogenic drawing, he fixed his calotypes with bromide, and in 1834 at Geneva he had found silver iodide “not sensitive at all” — a result he reports in The Pencil of Nature as a disappointment.

Read through section 23.6, that disappointment is the specification. Of the three halides, only iodide gives a sheet that is genuinely dead. A chloride paper made the same way would go violet in sunlight; a bromide paper would go grey. Either would fail the one requirement this formula exists to meet, which is that the prepared sheet must survive being taken out of a drawer in daylight and must still be the same material a year later.

So the property that made silver iodide useless for printing out is exactly the property that made it the right halide for a stored, developed process. That is the inversion at the heart of the calotype, and it is why this page and the photogenic drawing paper are not two variations on a theme but opposites.

One honest boundary. The paragraph above is the course reading Ware’s data back onto Talbot’s choice; Ware does not claim that Talbot reasoned this way, and neither does this page. Talbot found by trial that iodised paper kept and excited paper did not. The explanation waited a century.

Where the sensitivity comes from afterwards

Nothing on this page makes the sheet sensitive. The next stage does, and the mechanism is the mirror of the one above. Ware: “the silver iodide already existing in the paper was made sensitive to light at this stage by the excess silver ions introduced in the ‘exciting’ solution”. Brushing on gallo-nitrate of silver floods the sheet with silver ions, which convert the crystal surfaces from fixed to sensitized — adsorbed iodide displaced by adsorbed silver, negative surface charge to positive, no halogen acceptor to a good one. The same crystals, in a different environment, doing the opposite thing.

That is why the sheet can be switched on at the moment of use and why it cannot be switched on for long: the acetic acid in the exciting solution slows the spontaneous decomposition, and even so Ware says the excited paper “remained critically unstable”.

The limit on how much iodide you can use

The excess that makes the sheet dead will, past a point, dissolve the sheet’s silver iodide altogether. Ware gives the equation:

AgI + 2 I⁻ → [AgI₃]²⁻

Complexation: silver iodide dissolving in its own excess iodide

The product is the tri-iodoargentate(I) anion, and Ware names it as such. It is the same class of reaction as the dissolution of silver halide in thiosulphate that makes a complex ion out of an insoluble solid, and the reason a bath cannot simply be made stronger to be safer. Hunt states the practical consequence in 1844, from Dr Ryan, in language a working photographer could use: a sheet left too long in the bath loses its silver iodide.

The course’s reading of the second failure, the silver iodide falling off after a prolonged rinse, is offered as a reading and marked as one. Ware reports the observation without a mechanism. Two things are true of this material and either would do it: there is no binder, so the silver iodide is held only by the fibres and by whatever sizing the mill supplied; and any silver iodide that had gone into solution as the complex will re-precipitate when the rinse dilutes the iodide — Furlong built a whole method on that reverse reaction — but it will re-precipitate wherever it happens to be, not necessarily anchored where it was. A loose deposit is a deposit that washes away.

Function of every ingredient

Silver nitrate — 38 g/L, 0.224 mol/L, brushed on one side and dried.

What it is. AgNO₃, relative molecular mass 169.873, the one silver salt soluble enough in water to be useful and the starting point of every silver process in this book. Talbot specifies the crystallized salt, which is how it was sold and is a purity statement rather than a hydrate one: silver nitrate has no hydrate.

Why it is here. It is the sole source of silver in the formula, and the silver is the future image. Every atom of metallic silver in a finished calotype negative was either laid down in this step or deposited on top of it later by physical development.

What it does chemically. It is dried into the fibres as the intact salt, sits there through the candlelit interval, and is then converted wholesale by the iodide bath. It should end the formula having entirely ceased to exist, its silver as silver iodide and its nitrate rinsed away as potassium nitrate.

What follows photographically. The amount of silver the sheet carries away from the brush sets a hard ceiling on how much silver iodide can ever be in it, and therefore on the maximum density the finished negative can reach. Guillot-Saguez’s later method proves the point by removing this step entirely and finding it had to compensate with a much stronger silver solution at the next stage.

More or less of it. Hunt is the source here and is unusually blunt: “A much weaker solution of the nitrate may be used, and this, on the score of economy, is important.” Channing of Boston went the other way, to 60 grains to the ounce — about 13.7 per cent w/v, three and a half times Talbot’s — and Hunt reports that little improvement can be made on those proportions either. The band of workable strength is evidently wide. What is not wide is the relationship to the iodide: put on more silver than the bath can convert in two or three minutes and the surplus stays in the sheet as free silver nitrate, at which point the paper is light-sensitive, will not keep, and has silently become a bad photogenic drawing paper instead of a good iodised one.

What it interacts with. Everything, and mostly badly. It must not meet the iodide bath in a vessel. It must not meet chloride, which is why Talbot specifies distilled water for this solution and does not for the other. It is an oxidiser, and the sheet carrying it is dried at a fire. And it is reduced by every organic reducing agent in a darkroom, gallic acid included — which is not a flaw here but the next stage’s whole mechanism.

Potassium iodide — 57 g/L, 0.343 mol/L, a bath the dried sheet is immersed in for two or three minutes.

What it is. KI, relative molecular mass 166.003, very soluble in water, cheap, stable as a solid and the standard laboratory source of iodide ion. Talbot names plain water for it and no purity grade.

Why it is here. To supply iodide, and to supply too much of it. Both halves are deliberate and the second half is the invention.

What it does chemically. Two jobs in one dip, and they happen in that order. First it converts: every silver ion in the sheet is precipitated as silver iodide, and the potassium and nitrate that are left over pair off as a soluble salt. Then, having run out of silver to convert, the surplus iodide adsorbs onto the surfaces of the crystals it has just made and stays there — which is what turns the freshly made silver iodide from a photographic material into an inert yellow stain. The transition from the first job to the second is the moment the sheet stops being sensitive, and it happens inside the two or three minutes.

What follows photographically. The sheet becomes storable, and it becomes yellow. The first is the process’s greatest practical advantage over everything before it; the second is the tax it pays, levied on every print made from the negative afterwards.

More or less of it. This is the ingredient with the sharp edges. Less, or a shorter dip, and conversion is incomplete: free silver nitrate remains, the sheet fogs and will not keep. More, or a longer dip, and the complexation begins — the silver iodide dissolves back into the bath as tri-iodoargentate, the sheet ends up thin and weak, and Ware records that either misjudgement gave poor negatives. There is also a slower penalty, downstream and easy to miss: Argyronomicon 7.5.1 records that an image sitting in excess iodide fades even in the dark, its nanoparticle silver oxidised back to pale silver iodide by atmospheric oxygen, “especially if the solution of potassium iodide was too concentrated”. Iodide left in a sheet is not neutral. It is patient.

What it interacts with. Silver nitrate, absolutely and only inside the paper. Its own solution, which yellows on keeping as free iodine appears; a yellowed stock is no longer potassium iodide but iodine dissolved in iodide, and that mixture attacks silver rather than precipitating it. And image silver, as above.

Water, which is not an ingredient in the object but is a step in the sequence. The dip between the iodide bath and the blotting-paper is doing real chemical work: it removes the potassium nitrate the reaction made and the potassium iodide it did not use. That second removal is what stops the excess iodide going on quietly damaging the sheet in storage. It is also the step with a stated limit in the other direction — prolonged, it takes the silver iodide with it.

The paper, which the schema has no field for and Talbot treats as an ingredient. Best writing paper, smooth surface, close and even texture, watermark removed; Ware identifies Whatman’s Turkey Mill from Talbot’s practice. Its gelatin sizing holds the sensitiser in the surface fibres, and its evenness decides whether the yellow is even. A sheet is not a neutral support in a process that has no binder of its own: in a calotype the paper is the emulsion’s matrix.

Interactions

Silver nitrate and potassium iodide — order first, proportion second. Reverse the order and you do not get a worse version of this formula, you get a different one: iodide into the paper first and silver afterwards is Guillot-Saguez’s process of 1847, in which no silver is present at the iodising stage at all. Ware’s judgement is that deferring the silver to the exciting stage put it “where it was more controllable”. Both work. They are not interchangeable, because the second cannot leave the sheet in a state of halide excess in the way the first does.

Iodide excess and the sheet’s insensitivity — the interaction that is the formula. Adsorbed iodide is what makes the paper storable, and the same adsorbed iodide is what must be displaced by silver ions before the paper will do anything. Every property of this material, good and bad, comes out of that one equilibrium.

Iodide excess and silver iodide itself. Past about one and a half times equivalence in the bath, and past two or three minutes in it, the excess starts dissolving the product. The formula’s tolerance is the gap between “enough excess to poison the crystals” and “enough excess to dissolve them”, and nobody in 1841 published where the edges of that gap were.

Iodide and image silver, on a timescale of years. The same excess oxidises finished image silver back to silver iodide in the presence of air and humidity, and light accelerates it. Talbot exploited that deliberately in his abandoned “Leucotype” direct-positive experiments, and suffered from it in every photograph he fixed with iodide.

Silver iodide and thiosulphate. Not an incompatibility but a difficulty, and one that shaped the history: silver iodide is dissolved by thiosulphate only with difficulty and needs heat, which is why Talbot’s 1843 patent specifies a bath heated nearly to boiling and why removing the yellow was a separate operation rather than part of fixing.

Silver iodide and gelatin sizing. Section 23.3 is careful here and this page will be too. Gelatin is not an effective halogen acceptor at print-out exposure levels, so it is not what makes any of this work. What it does is mechanical and colloidal: it retains the sensitiser in the surface fibres, and it can protect the silver particles eventually formed, influencing their colour and stability. On a French sheet sized with starch, the same solutions meet a different substrate.

The finished sheet and light. Nothing, which is the point, and is the only line in this section where the interaction is an absence.

Variants

None of the variants below is published as a formulary entry of its own; each is described here because Ware’s Table 1 treats the method of iodising as one of the four axes on which the whole calotype family varies, and a page about Talbot’s method that did not name the other two would be misleading about how the process was actually practised.

Talbot’s own second patent, No. 9,753 of 1 June 1843. Hunt’s appendix lists seven improvements and two of them bear on the storable sheet. The third is iodo-gallic paper: iodised paper washed with gallic acid, which “will keep in a portfolio” and is made sensitive later by a wash of nitrate of silver — the storable intermediate pushed one stage further down the process. The fourth goes further still: iodised paper washed with a mixture of twenty-six parts saturated gallic acid to one part of the usual silver solution “can then be dried without fear of spoiling, may be kept a little time, and used without further preparation”. The others — the hot hyposulphite bath for the yellow, the waxing, the warm iron behind the paper in the camera — belong to later stages and appear elsewhere on this page.

Channing of Boston, 1842, via Hunt. The first published simplification: wash with 60 grains of crystallized nitrate of silver to one ounce of water, and when dry with 10 grains of iodide of potassium to the ounce; wash with water, dry between blotting paper. A stronger silver and a much weaker iodide than Talbot’s. Hunt reports a second, more sensitive paper from the same source, made with five grains of the iodide and five of common salt to the ounce — a mixed-halide sheet, twelve years before mixed halides became ordinary — and adds that “little, if any, improvement can be made upon these proportions”. Two cautions. Hunt gives no immersion time for the iodide and describes it as a wash rather than a dip, so whether Channing’s sheet ends in halide excess at all is not established by the text; and the course has not read Channing’s own paper in the American Journal of Arts and Sciences of July 1842, only Hunt’s report of it.

Furlong, St Andrews, 1843 — a single solution. W. H. Furlong replaced Talbot’s two treatments with one. An eight-fold excess by weight of potassium iodide added to a silver nitrate solution throws silver iodide and then redissolves it to a clear solution of the double iodide, which is imbibed into the sheet; diluting with an excess of water drives the reaction backwards and re-precipitates the silver iodide in the surface fibres.

AgI + 2 I⁻ → [AgI₃]²⁻

Furlong's method run forwards, in the bottle

Ware’s verdict: “Despite the complexity of its preparation, this was a chemically elegant method that, at a stroke, overcame the two iodizing problems inherent in Talbot’s method. With this method, calotypists could now iodize their papers with a degree of certainty that previously came only through trial and error.” It removes both failure modes at once, because the complexation that was Talbot’s enemy is now the mechanism, and the re-precipitation happens on dilution rather than in competition with it.

⁇Furlong, Adamson, and how much of Hill and Adamson is a chemistry footnote

Ware raises, as a question rather than a claim, whether Furlong’s method is part of why Robert Adamson’s calotypes were good enough to impress David Octavius Hill in 1843. The announcement preceded the Disruption by weeks; Hill is on record thinking the process before that “was chemically and artistically speaking a very miserable affair”; and Ware writes “Perhaps Furlong’s method of iodizing was one of the things that Adamson ‘knew and others did not’”, noting how easily such a formulation would have stayed inside the St Andrews and Edinburgh circle.

The course records this as Ware’s speculation, phrased by him as a question, and does not adopt it. There is a documented fact underneath it — Furlong’s letter was read to the St Andrews Literary and Philosophical Society on 3 April 1843 by John Adamson, Robert’s brother — and an undocumented inference on top. A second, separate item in the same footnote is on firmer ground and is also attributed rather than asserted: Furlong, writing in Photographic Notes in January 1856, credits John Adamson with the “wonderful improvement” of sunning the iodised paper to improve its sensitivity. The course has not read that paper and reports it as Ware reports it.

Guillot-Saguez, Paris, 1847 — no silver at all. The third distinguishable method, and the boldest: the paper is iodised with a single 4.1 per cent solution of potassium iodide and nothing else, all the silver being deferred to a strong 7.1 per cent aceto-nitrate at the exciting stage. Ware calls the modification “bold and significant” and says the stability was greatly improved by also withholding gallic acid, at the cost of losing the option of diluting the exciting solution, since a dilute one would have left the negative short of silver. It has a diagnostic side-effect worth knowing: on standing in moist air the potassium iodide coating oxidises slightly to free iodine, whose violet colour — blue on starch-sized French papers — is discharged when the exciting solution has fully penetrated the sheet, giving the worker a visible signal that Talbot’s method never offered.

Later French practice, from a translation. Ware prints Arsène Pelegry’s iodising bath as an example of how elaborate the recipes eventually became: rice water with potassium iodide, potassium bromide, free iodine and milk sugar, or the same salts in clarified milk serum. The course records that such formulations existed and does not reproduce their quantities, because Ware himself prints them from a machine translation supplied by a correspondent and flags it as such, and because a bath containing free iodine is a different hazard from the two salts above.

What did not vary. Ware’s summary is the useful closing fact: no matter how elaborate the recipe, “nothing overall was chemically different. At its heart, the calotype process functioned simply by precipitating silver iodide in paper; by sensitizing it with excess silver ions; by impressing a latent image upon it with light; and by developing that image with gallic acid.”

Safety

Level B, and one ingredient carries almost all of it.

Silver nitrate is the governing hazard, even at 3.8 per cent w/v. Its harmonised classification under Regulation (EC) No 1272/2008 is signal word Danger, with H272 (may intensify fire; oxidiser), H314 (causes severe skin burns and eye damage), H400 and H410 (very toxic to aquatic life, with long-lasting effects). A dilute solution is less corrosive than a concentrated one and is not thereby safe: it stains skin, cloth and every surface it touches, the stain appears hours after the splash, and it has to wear off with the skin. Sealed splash goggles rather than safety glasses, nitrile gloves, an apron, and eyewash within reach before the bottle is opened. The full regime is in the silver nitrate handling SOP and the whole classification, with workplace exposure limits and the argyria warning, is on the chemical’s own page.

The fire hazard is specific to this formula and Talbot walked straight into it. Silver nitrate is not combustible itself but accelerates the burning of combustible material, and a sheet of paper carrying an oxidising nitrate and held to a fire to dry is the textbook case. His own instruction distinguishes a distant fire for the silver-loaded sheet from one it may be “held pretty near” after iodising, which is the correct distinction for the wrong-or-unstated reason. Nobody should copy the fire-drying, and nothing in this course asks anyone to.

Potassium iodide is milder but is not nothing, and the classification is genuinely disputed. PubChem’s aggregation of the ECHA C&L Inventory returns signal word Danger with H302, H315, H317, H319, H334, H372, H373 and H411 — but 13 per cent of the reports state that the substance does not meet GHS criteria at all, and no single statement is given by even half the notifiers. That is a spread of opinion rather than a ruling, and the honest summary is: treat it as a skin and respiratory sensitiser, avoid dust when weighing, and read the supplier’s own safety data sheet, which is the authority for the grade in your hand. Its own encyclopaedia page carries the detail.

Silver iodide, the product, is classified for the environment and for nothing else. GHS09 only, H400 and H410 — no signal word Danger, no health-hazard pictogram and no skin, eye or respiratory statement above PubChem’s display threshold. That is a narrower classification than either of the other two photographic halides carries, and it is the reason the finished sheet is handled as an ordinary piece of paper and the waste is not.

Free iodine, if the stock has yellowed. A potassium iodide solution that has gone deep yellow has liberated iodine, which is volatile, stains, irritates the eyes and airway, and — photographically — attacks silver rather than precipitating it. A yellow stock is a different reagent, not a tired one.

And the reason this page still gives no procedure is set out in the callout at the head of it, and it is not any of the above. The hazards here are ones the course teaches people to manage.

Storage

The finished iodised sheet — the claim this whole formula exists to support. Talbot: it “ought to be kept in a portfolio or a drawer, until wanted for use. It may be kept for any length of time without spoiling or undergoing any change, if protected from the light.” Ware confirms it as a property rather than an optimism: the sheet “was quite stable to light and could be stored for long periods because of its insensitivity”, and section 23.6 supplies the reason. Note the small inconsistency Talbot leaves in his own text and does not resolve — the paper is “scarcely sensitive” and yet “ought” to be kept in a drawer. Insensitive is not inert, and he evidently knew it.

The excited sheet, which is not this page’s material but is the comparison that gives it meaning. Talbot: three months in a press, often, “but this is not uniformly the case”, and use it within a few hours. Hunt in 1844: two or three months has been found, but it is “often rendered useless by spontaneous change, in the dark”. Ware, reading the practice: an hour or two before it fogged. Three figures, from three vantage points, and the course records all three rather than picking one. What they agree on is the direction, and the direction is the argument for iodising as a separate, storable stage.

The two solutions. A silver nitrate solution is decomposed by light faster than the solid is: amber glass, capped, cool, dark, in a vessel that has never held food, and labelled with strength and date per the labelling SOP. A potassium iodide solution keeps less well than it looks, and its own page carries Kodak’s 1928 finding that such solutions often turn deep yellow from liberated free iodine — dark brown bottles for both the solid and the solution, and a yellowed stock discarded rather than used.

The finished negative, as an object. Two things, and they are different. The chemical one is Talbot’s own proviso in The Pencil of Nature: the number of copies a negative will yield is almost unlimited “provided that every portion of iodine has been removed from the picture before the copies are made”, because sunlight and a trace of iodine acting together, though neither alone, will decompose the silver. The conservation one is the Photographic Materials Group’s, which treats photogenic drawings, salt prints and calotypes as one storage problem: a stable temperature between 18 and 30 °C to avoid embrittlement, at 30 to 50 per cent relative humidity. The paper support, not the image, is what that figure protects.

Incompatibilities

Silver nitrate and any soluble halide, in a vessel. The two solutions of this formula are the example. They meet in the paper or they are both ruined; a splash of the iodide bath into the silver bottle throws a curd of silver iodide and ends both. Separate vessels, separate dishes, separate brushes, separate everything. See incompatibilities.

Silver nitrate and tap water. Chloride in the supply precipitates silver chloride and quietly converts part of the silver into the wrong halide. Talbot specifies distilled water for the silver solution and, tellingly, not for the iodide bath, where the same impurity does no harm.

Silver nitrate and combustible material. Paper, cloth and wood loaded with an oxidiser and then warmed. The historical instance is on this page.

Silver nitrate and any reducing agent. In a darkroom that means every developing agent, and it also means gallic acid — which the next stage introduces on purpose, and which is why Talbot warns that the gallo-nitrate mixture “does not keep long without spoiling” and why it must be mixed in small quantities as wanted. Hunt saw the same reaction from the other side in 1844: gallic acid and silver nitrate mixed together precipitate almost immediately in weak diffused light, while the same mixture “will often remain clear for many hours in the dark”.

Excess iodide and image silver. An incompatibility inside the process rather than between two bottles. Iodide left in a finished photograph oxidises its silver back to silver iodide over months and years, faster in humid air and faster still in light.

Potassium iodide stock, air and light. Free iodine develops, and iodine bleaches a silver image. The reagent turns into its own enemy on the shelf.

Thiosulphate carried backwards. A trace of hypo on a wet hand, in a sensitising dish or on a brush, does nothing visible at the time and produces a dead patch later. In any silver process contamination travels upstream at least as often as downstream.

Waste

Both baths are silver-bearing streams by the time the work is done, and the second one is the surprising one.

The silver solution obviously carries silver. The iodide bath carries it too, in two ways: as dissolved tri-iodoargentate complex where the immersion ran long, and as silver iodide particles rinsed or shed off the sheets. So does the water rinse. None of it goes to a drain. All of it is collected into one labelled container under the silver-bearing waste SOP.

The reason is not administrative. Silver and its compounds carry H400 and H410 — very toxic to aquatic life with long-lasting effects — and silver iodide’s classification consists of essentially nothing else. A bath that looks like clear water and is legally an aquatic toxicant is exactly the kind that gets poured away.

Spoiled and off-cut iodised paper is a solid silver-bearing waste, not paper recycling.

Local regulation decides what may be discharged, and this course cannot tell you what it says where you are. See disposal and the general chemical waste SOP.

Troubleshooting

The faults below come from three places and are labelled: what Talbot published, what Hunt and his contemporaries reported in the 1840s, and what Ware establishes from the chemistry. Where a reading is the course’s own, it says so.

A thin, weak, barely yellow sheet after a long dip. Reported by Hunt in 1844, crediting Dr Ryan, and by Ware. The silver iodide has been re-dissolved by the excess iodide as tri-iodoargentate. The period remedy was Hunt’s: shorten the immersion drastically — “simply passing the paper through the solution appears to answer every purpose effectually”.

The yellow coating comes off in the rinse. Reported by Ware. Prolonged rinsing causes the silver iodide to fall off. Course reading, offered as one: there is no binder holding it, and any material that had gone into solution as the complex re-precipitates on dilution wherever it happens to be rather than anchored in the fibres.

A sheet that fogs, darkens by itself or will not keep. Course reading from Ware’s mechanism, not a fault Talbot names. Conversion was incomplete and free silver nitrate remains in the paper. A sheet in that state is not an iodised paper at all; it is a weak photogenic drawing paper, with all the instability that implies, and it will have to be judged by whether it darkens in daylight rather than by its colour.

An uneven or too-deep yellow. Reported by Hunt in 1844. Brushwork, or a bath that has been allowed to work too long. It cannot be corrected and it is not cosmetic: the tint absorbs the blue and ultraviolet that every subsequent print needs, so an uneven sheet gives an unevenly printing negative for the rest of its life.

The watermark in the picture. Named by Talbot in the first sentence of the procedure. Cut it off before starting.

Poor negatives with no identifiable cause. Reported by Ware, Cundell and Lady Eastlake. This is the characteristic failure of the whole method rather than of one step: “the timing of these two stages was critical … Either misjudgement could result in poor negatives”, and Talbot’s published directions did not say what the tolerances were. Furlong’s variant exists because of this fault.

A finished negative that prints very slowly. Reported by Hunt in 1844 and by Talbot’s 1843 patent. Residual silver iodide, doing what yellow does to blue light. The period cure is the hot hyposulphite bath, and Ware records that many photographers did not bother.

A negative that fades after yielding several prints. Reported by Talbot in the Pencil of Nature. Iodine and sunlight acting together decompose the silver. His own remedy was to wash the sheet again with gallo-nitrate and warm it, which redeposits silver on the image and revives it — a repair no modern developed material admits of, because a modern developer has only the silver already in the grain.

A negative that fades in the dark. Established by Ware, section 7.5.1. Excess iodide, atmospheric oxygen and humidity, without any light at all, “especially if the solution of potassium iodide was too concentrated”. The fault is in this formula even though the symptom appears years later and on somebody else’s page.

Experiments

None of these makes a calotype, because this page is not a procedure and the course does not teach the process. Each instead tests a claim this page makes, using chemistry the course already teaches at Level B: the salting solution and salted paper sensitiser, with potassium iodide and potassium bromide added from the shelf. Read the silver nitrate handling SOP first, and note that every one of these produces silver-bearing waste.

The fixed-halide experiment, which tests the mechanism of this entire page. Hypothesis: a silver halide sitting in excess of its own halide is insensitive, and how insensitive depends on which halide it is. Control: a sheet sensitised and left with silver in excess, as an ordinary salted paper. Variable: which halide is put in excess afterwards.

Salt and sensitise four identical strips of the same stock in the usual way. Leave the first alone. Bathe the second in a strong sodium chloride solution, the third in potassium bromide and the fourth in potassium iodide, rinse briefly, dry all four in the dark, then expose all four side by side to direct sunlight for an hour and photograph them hourly against a grey card. Ware’s section 23.6 predicts the outcome precisely and it is unusually specific for a prediction that costs so little to test: the control prints out to a full brown-black; the chloride-treated strip goes dull violet and stops; the bromide strip goes pale grey; and the iodide strip stays pale yellow and unchanged, even in direct sunlight. If the fourth strip changes, the account on this page of why Talbot’s iodised paper keeps is wrong.

The complexation limit, in test tubes, in ten minutes. Hypothesis: silver iodide dissolves in excess iodide as a complex, and the reaction reverses on dilution. Precipitate silver iodide by adding a few drops of silver nitrate solution to potassium iodide solution, then keep adding potassium iodide to the yellow suspension until it clears. Record the iodide concentration at which it does. Then add water and watch the silver iodide come back. You have just performed both halves of Furlong’s 1843 method and, in the first half, the fault that ruined Talbot’s sheets. Record the clearing concentration against the 5.7 per cent of the bath above and see how much margin the formula actually had.

The rinse experiment. Hypothesis: prolonged rinsing removes silver iodide from an unbound sheet. Control: a strip blotted straight from the halide bath. Variable: rinse duration — 15 seconds, 1 minute, 5 minutes, 20 minutes. Use the iodide-treated salted-paper strips from the first experiment, dry them, and compare the depth of yellow, by densitometer through a blue filter if one is available and by eye against a white tile if not. Ware reports the effect qualitatively; nobody appears to have published the curve.

What the yellow costs, measured. Hypothesis: residual silver iodide meaningfully slows printing by absorbing actinic light, which is Hunt’s 1844 claim and Talbot’s reason for a whole patent clause. Print a step wedge onto salted paper twice, identically, once through a clear sheet of the paper stock and once through an iodide-treated strip of the same stock. The difference in exposure needed to match the two is the tax, in stops. Expect it to be substantial in ultraviolet-rich light and to depend strongly on the light source, which is itself worth recording.

The keeping experiment, which takes a year and is worth starting anyway. Hypothesis: a halide-excess sheet keeps and a silver-excess sheet does not. Control and variable in one: two strips from the same batch, one left silver-rich as an ordinary sensitised salted paper, the other bathed in potassium iodide and rinsed. Both go into the same drawer, in the same envelope, on the same day. Photograph both against a grey card monthly. Talbot’s claim is “any length of time without spoiling or undergoing any change”, and it is a claim nobody in the modern literature seems to have re-tested with a control alongside.

Read the sizing, not the halide. For anyone with access to a papermaking supplier or an old stock of drawing papers: repeat the first experiment on a gelatin-sized sheet and a starch-sized sheet, and compare the evenness of the coating and the colour of the silver. Ware and the Getty atlas both record that English and continental mills sized differently and that photographers found their papers would not behave alike; the effect has a documented history and, so far as this course can find, no published measurement.