Albumen solution with its chloride

Every other salting solution in this formulary puts the chloride into the paper. This one puts it into the binder, and that single change of address is the whole difference between a salt print and the standard photographic print of the nineteenth century. It is also the only preparation in the course that begins by damaging its principal ingredient on purpose, three separate ways, and then leaves it in a refrigerator for a week to finish the job.

Ingredient	Quantity	Form the source specifies
Ammonium chloride	15 g	dissolved in the water with the acid
Glacial acetic acid	2 mL	glacial, i.e. the concentrate
Water	30 mL, added	the vehicle for the two above, and nothing else
Albumen	1000 mL	fresh egg white, cleanly separated

Reilly braces the first three together with the words “combine and add to” and prints the litre of albumen beneath them. That brace is part of the formula and is recorded as the mixing order.

Purpose

To convert a litre of egg white into a photographic coating solution — which means doing four things at once, only one of which is obviously photographic.

To carry the halide. The 15 g of ammonium chloride is the reagent that a later bath of silver nitrate will convert to silver chloride. Silver chloride is insoluble and cannot be dissolved and painted on, so in every process in this family it has to be assembled in place out of two solutions applied separately and dried separately. What is unusual here is where it is assembled: not in the paper fibres, but in a transparent protein layer sitting on top of them.

To make the egg white coatable at all. Native egg white is the wrong material. It is stringy, its several proteins have different viscosities, and it will not lie down flat on a sheet of paper. Reilly is blunt about it: albumen is never used in the strictly native condition. Beating, chlorides and acid are three separate denaturing treatments, and this formula applies all three.

To make it acid. Reilly gives the pH of native egg white as 7.8, and states that fresh alkaline albumen is used only for matte papers, where it is always mixed with starch or something else. Glossy papers are prepared from partially decomposed, acidic albumen, because in that condition it gives a glossier surface, a more even coating, and less tendency to yellow after sensitisation. The 2 mL of glacial acetic acid starts that process; the week in the refrigerator finishes it.

To keep the water out. The 30 mL is the smallest amount that will carry 15 g of salt and 2 mL of acid into a litre of protein. Every millilitre beyond it dilutes the binder, and the amount of water in the albumen is the single variable that decides whether the finished paper is glossy, half-matte or indistinguishable from a plain salted sheet. That ladder is set out under Variants, and its middle rungs have their own entry at the albumenised salting solution.

Recommended uses

Glossy albumen paper, single- or double-coated, printed out by contact under a negative in ultraviolet light and then toned and fixed. This is the material on which most of the surviving photographs of the second half of the nineteenth century were printed, and it is the reason the albumen print has a process page of its own.

Floating, for one to one and a half minutes, on one side only, from a tray filled to a depth of two thirds to three quarters of an inch. Albumen paper cannot be immersed: it is coated on one face and the back must stay dry, because a wetted back will take silver later and print through.

In white light, at leisure. Nothing in this solution is light-sensitive. Neither is the coated, dried sheet. Albumenising is therefore a batch operation done at the sink on an ordinary afternoon, and the sheets keep very well before they are sensitised. The clock that matters starts at the silver bath, not here.

With a negative of shorter density range than a salted paper wants. Reilly’s ranking, from his chapter on tone reproduction: plain salted paper needs the longest scale in the negative, matte arrowroot and matte albumen slightly less, glossy albumen least of the four — and all of them far more than any develop-out enlarging paper. Hübl’s measurement puts glossy albumen level with platinum paper for scale length.

As the binder half of a matched pair. Paper coated with this solution is sensitised by floating for two and a half to three minutes on a silver nitrate bath of 10 to 12 per cent, which is the strength Reilly recommends for a modern worker using a chloride content of 1.5 per cent or above. The two numbers are chosen together and neither is meaningful alone.

As the teaching case for protein chemistry in photography. Gelatin is the binder that won, and it is so thoroughly domesticated that nobody thinks about it. Albumen is the binder you have to argue with. It coagulates when you look at it, it is alkaline when you want it acid, it goes off on a schedule you have to learn to read, and every one of those inconveniences turns out to be doing something useful.

When another formula is preferable

• For a matte or half-matte surface, the same solution diluted — which is a different formula and has its own page as the albumenised salting solution. Hardwich’s three volumes of albumen to one of water is the classic middle rung; Reilly’s 1:6 is the bottom one.
• For a plain salted paper, the plain salting solution: 20 g of common salt and 2 g of gelatin in a litre. It is far simpler, keeps badly rather than usefully, needs no refrigerator and no eggs, and gives a longer scale and a lower maximum density.
• For a matte paper with more brilliance than a plain salted one, the arrowroot salting solution, or Hübl’s matte albumen, which is this solution cut half and half with a salted arrowroot and belongs to the matte albumen print.
• For a thin negative, a lower chloride content — 1 to 1.5 per cent with a silver bath of 8 to 9 per cent, which is what the trade actually did between 1880 and 1900 to cope with the shorter scale of gelatin dry-plate negatives. Reilly records the practice and then advises against it for a modern worker, on the ground that the richer, more permanent print made by stronger salting and silvering is worth losing a few unprintable negatives for.
• If you cannot keep a protein solution for a week, or cannot face the smell, a ready-made salted albumen solution is sold. Its composition is not published — no chloride content, no acid, no preservative named — so it is not a formula and this course cannot tell you what is in it. What the supplier does publish is a keeping time of three months unrefrigerated and 24 to 36 months refrigerated, which is what a preservative buys, and a six-minute float, which tells you it is not the same liquid as this one.
• If what you want is a hand-coated print and not specifically a silver one, a cyanotype is one coating rather than two, needs no fixer, keeps indefinitely and costs a fraction as much.

Mixing

Nothing here is heated and nothing is hurried. The whole operation is about ten minutes of work spread over eight days.

One: separate the eggs, and separate them properly. Between about 35 and 40 large eggs make a litre. Reilly and Hardwich both give the yield as roughly one fluid ounce of white per egg; Wall is more conservative at seven drachms. The white must be free of yolk, of blood and of the stringy tissue called the chalazae. Reilly’s procedure is the one to copy: separate each egg over a small bowl and pour it into the large one only when it is clean, so that a single bad egg costs one egg rather than the batch. Wall gives the same instruction in 1912 and adds “take out the germ”.

🧮How many eggs, and how much chloride: the arithmetic Reilly leaves to you

1000 mL ÷ 28.4 mL per egg = 35 eggs

Eggs for a litre, at Reilly's and Hardwich's yield

1000 mL ÷ 24.9 mL per egg = 40 eggs

Eggs for a litre, at Wall's 7 drachms

27 dozen eggs ÷ 9 L = 36 eggs per litre, i.e. 28 mL each

Eggs for a litre, from Reilly's Dresden figures

Three independent statements, a century apart, land between 35 and 40. That is a stronger agreement than most numbers on this page enjoy, and it is worth having because eggs are what this formula actually costs.

Now the chloride, which is the number the whole page turns on:

15 g ÷ 1000 mL = 1.50 % w/v

On the albumen, which is how every source states it

15 g ÷ 1032 mL = 1.45 % w/v

On the whole 1,032 mL of stated liquid

15 g ÷ 53.49 g/mol = 0.280 mol

As moles of chloride available to the silver

0.280 mol × 169.87 g/mol = 47.6 g

Silver nitrate that would exactly consume it

Reilly’s recommended bath is 100 to 120 g of silver nitrate per litre. Assuming — and it is an assumption, not a measurement — that the sheet carries away comparable volumes of the two baths, the silver arrives at roughly two to two and a half times the amount needed to convert the chloride, and the surplus stays in the sheet as free silver nitrate. That surplus is not waste; it is the second reagent of the printing-out process, and The mechanism below is about what it does.

One inconsistency worth noticing rather than hiding. Reilly’s general rule is that the silver bath should be about four times the strength of the salting solution, and 1.5 per cent × 4 is 6 per cent. His own recommendation for albumen paper is 10 per cent or above, which is nearly seven times. He gives the reason himself: better-looking and more permanent prints result from the increased silver image deposited by stronger salting and silvering. The four-times rule is a floor, not a target.

Two: dissolve the chloride and the acid in the 30 mL of water. Not in the albumen. Reilly’s reason for the chloride is that it should go in dissolved, in a minimum of water, before the beating starts; he adds that a blender makes this unnecessary, because a blender will dissolve solid salt in egg white by brute force.

✕The blender exemption covers the salt. It does not cover the acid.

Reilly writes that if a blender is used it is not necessary to dissolve the chloride in water first. He says nothing about the acid, and the course reads the exemption as not extending to it, for a reason that follows from his own chemistry rather than from a sentence of his: albumen is coagulated by acid, and 2 mL of glacial acetic acid poured neat into a litre of egg white is 2 mL of a 99 per cent acid meeting protein at the point of contact before it can be diluted by anything. The predictable result is a few small solid coagulated specks in an otherwise clear solution — exactly the particulate matter that the muslin straining step exists to remove, made deliberately.

This is a reading, not an instruction from Reilly. It costs nothing to follow: dissolving the salt and the acid in 30 mL of water takes ten seconds either way.

Three: pour the 30 mL into the litre of albumen and beat for three minutes. An electric mixer or a blender, until the entire mixture has been converted to froth. Do not stop early. The beating is not aeration for its own sake; it is the mechanical half of the denaturation, and Hardwich’s warning from 1864 is the one to keep in mind — if the albumen is not thoroughly beaten, flakes of animal membrane are left in the liquid and will streak the paper.

Expect less froth than you would get from unsalted egg white. Reilly names that explicitly as one of the effects of adding chlorides, and it is the most direct evidence a home worker will ever see that the salt is doing something to the protein rather than merely dissolving in it.

Four: let it stand, covered, for 24 hours. The froth collapses back to a liquid. The Getty’s atlas describes the same step in its own words: left to stand overnight, the froth liquefies, resulting in a much more homogenised and uniform solution of salted albumen. Hardwich and Towler both substitute settling for filtering at this stage and give the reason — albuminous liquids are too glutinous to run through a paper filter and are better cleared by subsidence.

Five: strain through muslin. Reilly warns that the liquid may have to be squeezed through with some pressure. This is where the membrane, the chalazae you missed and any coagulated specks are left behind.

Six: cover and refrigerate for one week. This is not storage. It is the last of the three denaturing treatments, running slowly at a temperature that keeps it slow. What comes out is Reilly’s target state: a yellowish, homogeneous liquid with a slight “aged” odour that signals its readiness for use.

Seven, immediately before coating: bring it up to temperature, filter it again, and wet it. Condition both the paper and the solution to the working temperature of the room. Filter through muslin into the tray. Then add a surfactant — Reilly gives 4 mL of Kodak Photo-Flo per litre of albumen, stirred in gently so as not to raise bubbles — for two stated reasons: it controls bubbles, which he calls one of the most troublesome aspects of the whole process, and it improves the runoff during drying, giving a more even coat. The trade word for a surfactant in a coating plant is a spreader, which is exactly what it is doing.

⚠One step of this preparation is above the level of the rest

The page is classified Level A and everything in it is Level A except one operation: measuring 2 mL of glacial acetic acid. The concentrate carries the signal word Danger with H314, causes severe skin burns and eye damage, and H226, flammable liquid and vapour, and the ILO-WHO safety card records that a harmful contamination of the air can be reached rather quickly on evaporation at 20 °C.

For that step alone, work at Level B: splash goggles rather than spectacles, nitrile gloves, an apron, eyewash within reach, ventilation at the bench, and the acid added to the water and never the water to the acid. It takes about ten seconds and then it is over; the rest of the preparation is eggs and salt.

The classification rule the course works to is on the safety classification page, and it prefers avoiding a raised step to controlling it. Here the avoidance is real: Kodak’s 1928 primer states that 28 per cent acetic acid is made by diluting three parts of glacial with eight of water, so 7.3 mL of 28 per cent acid delivers the same 2 mL of glacial acid and adds about 5 mL of extra water to a 1,032 mL batch. That arithmetic is the course’s, not Reilly’s, and the extra water is a real if tiny change to a formula whose whole point is how little water it contains. Kodak’s own warning runs the other way and is worth carrying: a formula that calls for glacial acid and receives 28 per cent instead gets less than one third of the acid it asked for.

What is not in the procedure, and why. No heat, at any point: albumen coagulates above 65 °C, so a warmed solution is a ruined one. No pH measurement, because Reilly publishes no target pH for the finished solution and the course will not invent one — he gives 7.8 for native egg white and then describes the destination qualitatively, by smell and colour. And no make-up volume, because there is none: this formula is a litre of albumen with things added to it, not a litre of solution.

Behaviour

It is a protein solution that is going off on purpose, at a controlled rate. That single sentence explains most of what is odd about handling it. There is no oxidation to guard against, no precipitate waiting to form, no alkali losing carbon dioxide to the air. What there is instead is a slow, wanted, irreversible change whose end point is judged by nose and eye rather than by a clock.

It thickens the froth less than egg white should. The chlorides cut the froth volume, which is visible from the first thirty seconds of beating.

It clears by settling, not by filtering. Hardwich’s observation from 1864 still holds: the liquid is too glutinous to pass a paper filter. Muslin, cheesecloth, sponge or a tuft of washed cotton wool are what the sources use, and Towler’s version of the procedure — a day covered from dust, a filter through sponge, then two more days of settling and a decant of the supernatant — is simply a slower route to the same clear liquid.

It is fussy about temperature at the tray, and the fussiness is mechanical. The supplier of the ready-made solution states that albumen works best above 60 °F and that many papers will resist absorbing cold albumen, which is why it must come out of the refrigerator at least two hours before use. That is a statement about a different product, but the underlying property is the same one Towler described in 1864: in dry conditions the albumen does not attach itself easily to the paper, giving a coat that is thin at the top of the hung sheet and thick at the bottom.

✕60 °F is 15.6 °C, not 14 °C

The supplier’s own sheet prints “above 60F (14C)” and the conversion is wrong: 60 °F is 15.6 °C. The course quotes the Fahrenheit figure, which is the one the sheet is written around, and notes the slip rather than passing on either number silently. It makes no practical difference at a tray and it makes a great deal of difference to whether you can trust a document’s arithmetic elsewhere in it.

Bubbles are the principal enemy and they are worse here than on a salting solution. Towler says so directly: much more care is required in laying the paper on salted albumen than on a plain salting solution, because bubbles are more likely to form and are less easily removed. A bubble under the sheet is a circle that never receives albumen and therefore never receives chloride, and the silver nitrate that arrives later finds nothing to react with. The surfactant is the modern answer; the paper spatula drawn slowly across the tray to sweep bubbles to one side is the supplier’s.

It coagulates on contact with the silver bath, and that behaviour only appears at the next step. This is the property that makes albumen paper possible at all. Reilly: the reason albumen does not dissolve off the sheet during processing is that contact with silver nitrate coagulates it, forming an insoluble silver-albumen complex. A weak or freshly made silver bath coagulates it less well and can let the layer start to dissolve; an older bath, which has accumulated ammonium nitrate from the double replacement, coagulates it better. The bath gets better at its job as it is used, which is counter-intuitive and true.

A litre goes further in a factory than in a tray, for reasons that are geometry rather than chemistry. Reilly’s Dresden figures give 9 litres per ream of 480 sheets 46 × 58 cm. That is 53 sheets, about 14 m² of paper, to the litre. The supplier of the ready-made solution gives about 60 prints of 8 × 10 inch to the litre, which is about 3 m². The gap is not a contradiction: the supplier double-coats, and a hand worker must fill a tray to a stated depth before floating a single sheet.

🧮Why the tray is the real consumption

Reilly’s tray depth is two thirds to three quarters of an inch, which is 17 to 19 mm. Take an 8 × 10 inch tray with an internal floor of roughly 21 × 26 cm:

0.21 m × 0.26 m × 0.018 m = 0.00098 m³ ≈ 980 mL

Volume to fill it to 18 mm

So filling the tray takes about the whole litre, before a single sheet is floated. The albumen a session actually consumes is a small fraction of what has to be in the tray for the session to happen at all, and the rest goes back in the bottle. The tray dimensions are the course’s assumption and yours will differ; the conclusion does not, and it is the reason to use the smallest tray the sheet will fit and to plan a session’s sheets in advance.

It does not exhaust in any chemical sense. Nothing is consumed except by carry-off on the sheets. What ends a batch is decomposition, not depletion, and Reilly publishes no capacity figure at all for a hand worker.

Image characteristics

Glossy, and the gloss is the point. A transparent binder on the surface of the sheet minimises diffuse reflection and the scattering of light by the paper fibres. Whites look whiter and shadows look denser for the same quantity of silver, and fine detail survives because the image is not sitting inside a fibrous mat. The Getty records that the gloss was heavily criticised in the photographic literature of the 1850s by a public accustomed to matte salt prints, and that taste changed as people came to want the high definition and contrast that came with it.

Higher maximum density and a shorter required negative scale than any matte paper of the family. The two go together, and both follow from the same optical fact. Reilly’s ranking of required negative density range: plain salted greatest, matte arrowroot and matte albumen slightly less, glossy albumen lowest. A negative made for an enlarging paper will still print flat and empty; a negative made for a salt print will print harshly.

Long scale, distributed unevenly. Hübl measured glossy albumen as equal to platinum paper in scale length, with salted papers exceeding both by a considerable margin. He also found the shape of the curve: albumen shows a slow progression from the shadows to the middletones and then an abrupt jump from the middletones to white. The practical consequence is a rule for making negatives — put the detail in the highlights, because that is where this paper has the least room.

The slowest paper in the family. Reilly’s speed ranking is matte salted papers such as arrowroot fastest, plain salted next, and albumen the slowest of the lot — the exact reverse of the gloss ranking. He gives the order of magnitude too: an average exposure for albumen paper is 5 to 10 minutes in direct sunlight, and in shade anywhere from half an hour to several days.

Reddish brown untoned; brown, purple or aubergine with gold. Untoned printing-out silver is a warm colour, and the whole point of the alkaline gold toning introduced around 1855 was to widen the range and improve permanence. Two things you do at this tray change the colour before any toner is mixed. Less chloride gives a warmer colour — Hardwich states it directly and warns that it must not be reduced so far as to injure the contrast and depth of shadow. And citric acid put into the albumen makes the print reddish brown even under prolonged gold toning; put into the silver bath instead, it makes it browner still. Purple tones are difficult in the presence of citric acid, which is precisely why professionals went on sensitising their own paper after 1872 while amateurs bought it ready-made.

Double coating raises everything and costs something for each. Denser, glossier, better from thin negatives — and harder to tone, harder to fix, more brittle, more prone to curl. Reilly’s explanation of the toning and fixing difficulty is physical rather than chemical: albumen becomes increasingly less permeable as the coating thickens.

And it will yellow, over decades, partly because of what is in this bottle. The AIC’s conservation catalogue lists four proposed causes of the yellow highlight staining that is characteristic of surviving albumen prints. Two of them are properties of the material mixed here: the chemical bonding of image silver to the sulfur-containing side groups of the egg protein, forming silver sulfide; and the Maillard reaction between the glucose naturally present in egg white and the amino groups of the protein, forming a highly coloured insoluble compound whose formation is accelerated by high humidity and by alkaline conditions.

Both of those are in the bottle before any silver is. Nothing in this formula removes either, and the course has found no source saying whether the trade’s fermentation — which was pursued for gloss — consumed any of the sugar as a side effect. That would be a satisfying connection and it is not one the evidence supports, so it is left as an open question rather than an assertion.

The mechanism

There are two mechanisms on this page and they run in opposite directions. One takes a folded protein apart so that it will coat. The other assembles a light-sensitive salt inside the coating that results. They are joined by the fact that the same ammonium chloride does part of both jobs.

The protein: three denaturing treatments, in the order the formula applies them

What albumen is. Not a compound. The chemical encyclopaedia’s albumen entry records that the FDA’s substance registry classes egg white as structurally diverse rather than as a molecule: there is no formula and no molar mass, only CAS 9006-50-2. Reilly gives its specific gravity as 1.040 and notes that many specific proteins can be identified in it, referred to collectively as albumen. Their viscosities differ, and that difference is the problem the beating solves.

Treatment one, mechanical. Beating denatures the various proteins and produces a homogeneous liquid that will form an even layer on paper. Reilly’s phrasing is worth reading carefully: it is the differences between the proteins that are being levelled, not the proteins that are being removed.

Treatment two, ionic. The chlorides help to denature the albumen. Reilly states this as a fact of practice and gives the observable consequence — the froth volume falls.

Treatment three, acid. Reilly’s sentence is the mechanistic one: the chemical forces which bind together the enormous molecules of protein grow weaker as the pH is lowered, and the physical properties of the substance change as a result. Adding acid and beating therefore change egg white completely and irreversibly.

⚗How far the course will go on protein denaturation, and where it stops

Reilly’s account is the one this page rests on and it is stated above in his terms. The general picture behind it is standard protein chemistry and is stated here only at the level his sentence supports: a native protein is a chain folded into a particular shape and held there by interactions between its side chains, some of which depend on those side chains being charged; lowering the pH protonates acidic side chains and takes their charge away, so the interactions that depended on them fail, and the chain opens out. A solution of opened-out chains is more homogeneous and less stringy than a solution of folded ones.

What the course does not state, because it has not read a source that would support it: the isoelectric points of the individual egg-white proteins, the pH this formula actually reaches, the proportion of the change that is due to the acid rather than to the chloride or the beating, and any rate for the week of refrigerated ageing. No source on this page publishes any of them. A reader with a pH meter can supply the second of those for their own batch in a minute, and the Experiments section asks them to.

Why the acid condition is wanted rather than merely tolerated. Reilly gives three consequences and they are all practical: partially decomposed acidic albumen creates a glossier surface, coats more evenly, and has less tendency to yellow after sensitisation. Hardwich, writing sixty-six years earlier without the vocabulary, describes the same transformation from observation — albumen deliberately evaporated in an open vessel becomes more limpid, acquires a rather offensive odour and an acid reaction to litmus, and in that state runs on the paper easily and does not dissolve the size.

That last clause is a mechanism in itself, and it is the reason a photographer might reach for acid even without Reilly’s gloss argument. Hardwich also records that albumen is not neutral but alkaline, from a small quantity of soda, and that on adding chloride of ammonium to albumen a development of free ammonia takes place, easily perceptible to the smell — and that ammonia is a solvent of the materials used to size paper.

NH₄⁺ + OH⁻ ⇌ NH₃ + H₂O

Why an ammonium salt smells of ammonia in an alkaline liquid, and why acid stops it

The ammonium ion is the acid form of a weak base. In an alkaline solution the equilibrium above lies to the right and free ammonia is released; acidify, and it lies to the left and the ammonia stays put as the ammonium ion. So the acetic acid does two jobs at once: it denatures the protein, which is Reilly’s stated reason, and it suppresses the very ammonia that Hardwich says attacks the sizing of the paper. The second of those is the course’s reading joining two sources, not a claim either author makes. It is offered as an interpretation and should be treated as one.

The photograph: what happens after this solution has dried on paper

Step one, at the silver bath: a double replacement, inside the binder.

NH₄Cl + AgNO₃ → AgCl + NH₄NO₃

Silver chloride formed in place, in the albumen rather than in the paper fibres

Reilly writes the sodium version of this equation out in his sensitising chapter. Two products, and both matter. The silver chloride is the light-sensitive substance and it is now a fine precipitate held in a transparent layer. The ammonium nitrate is soluble and washes out later — but while it is there it accumulates in the silver bath, and Reilly records that these nitrates have a coagulating effect on albumen, so much so that some old manuals advised adding them deliberately to weak baths.

Step one and a half, simultaneously: the binder becomes a second sensitive substance. Silver nitrate coagulates albumen and forms an insoluble complex Reilly calls silver albumenate. He states plainly that this substance is itself light-sensitive and makes an important contribution to image formation in albumen paper. This is the sense in which albumen is an “active” binder in his classification, as against starch, which is inert and needs an organic acid brought in beside it.

Step two, under the negative: photolysis, and a reaction that wants to run backwards.

2 AgCl + light ⇌ 2 Ag + Cl₂

Print-out, and the back-reaction that would stop it

Silver chloride is reduced to metallic silver and chloride is oxidised to chlorine. Left alone, the chlorine re-forms silver chloride and the print-out comes to a halt at a density too low to see.

Step three: the free silver nitrate takes the chlorine away.

Cl₂ + H₂O + Ag⁺ → AgCl + HOCl + H⁺

What the surplus silver nitrate and the paper's own water do with the liberated chlorine

This is what the excess computed in the maths callout is for, and it is why Reilly insists that the silver bath must be strong enough not merely to convert the chloride but to leave a considerable excess in the layer. The fresh silver chloride made in that reaction is broken down by light again, and the cycle repeats. More free silver nitrate therefore means more image silver and a greater maximum density — which is the practical statement of the whole mechanism, and the reason Reilly prefers a 10 per cent bath over an 8 per cent one for a modern worker.

Step four: where the silver ends up, and why it looks the way it does. The image of a printing-out process is colloidal silver in very fine particles rather than the filamentary silver of a developed image; the AIC states the consequence, which is a large exposed surface area and extreme vulnerability to oxidation and chemical attack. (The particle size printed in that entry is not quoted here: the figure as it stands is in micrometres, which is three orders of magnitude larger than anything that could be called colloidal, and the course reads it as a slip rather than a measurement.) Because the chloride was dissolved in the binder rather than soaked into the paper, those particles form in a transparent layer on the surface. That is the entire difference between this formula and a salting solution, and it is worth stating once without hedging: the chloride’s address decides the photograph.

Function of every ingredient

Ammonium chloride, 15 g — a 1.5 per cent chloride. The halide, and the only ingredient that becomes part of the light-sensitive substance. It waits in the dried albumen layer for the silver bath and is converted to silver chloride exactly where it lies.

Its quantity is the most consequential number in the formula and Reilly gives a full account of what moves when it moves. Below about 1.5 per cent — his 1 to 1.5 per cent band — the paper is less sensitive and tends to produce slightly more contrasty prints from thin negatives, but it does so at the expense of a rich, dense image. Between 1.5 and 2.5 per cent is his normal band. Above what is necessary, more chloride buys nothing at all: it only raises silver consumption. His summary recommendation for a modern worker is to keep it at 1.5 per cent or above, with a silver bath of 10 per cent or above.

That leaves Reilly’s own sample preparation sitting exactly on the boundary of his own advice, and it is worth being explicit about which basis makes it so. On the litre of albumen, 15 g is 1.50 per cent and the formula just satisfies the rule. On the 1,032 mL of total stated liquid it is 1.45 per cent and the formula just fails it. Reilly does not say which he means. The course reads it on the albumen, because every other source on this page states a chloride as grains of salt per ounce of albumen, and because a formula whose author immediately afterwards recommends “1.5 per cent or above” is unlikely to have intended its own example to fall below the line. It is a reading, and the practical answer for anyone worried by it is to weigh 16 g instead, which is 1.55 per cent on either basis and is within the band whichever way you read him.

There is a second thing the chloride does, and it is not photographic. It is one of the three denaturing agents, and the visible sign of it is the reduced froth. As the salt of a weak base in an alkaline liquid it also liberates ammonia, which Hardwich both smelled and warned about; the acid in the formula is what keeps that in check.

Ammonium or sodium? Reilly says prints of similar colour and contrast may be expected from either, or from both in combination in the correct amount, and that ammonium chloride was the most commonly used in the nineteenth century. The Getty says the same in its process description. There is one substitution the sources rule out rather than in: Hardwich records that chloride of barium was sometimes used in salting paper and is contra-indicated when alkaline gold toning is adopted, because the sodium carbonate of the toner throws down carbonate of baryta in the paper.

Glacial acetic acid, 2 mL. The acidifier, and the third of the three denaturing treatments. It is 2 mL of a roughly 99 per cent acid in 1,032 mL, and its job is to shift an alkaline protein solution into the acid condition in which it coats glossier and more evenly and yellows less after sensitising.

Less acid, or none, leaves the albumen nearer its native pH of 7.8: stringier, harder to coat evenly, more inclined to yellow at the sensitising bath, and — on Hardwich’s account — more inclined to release ammonia that attacks the paper’s sizing and lets the coating sink in instead of sitting on top. Towler’s and Hardwich’s own pure-albumen formulas contain no acid at all, so a workable paper can certainly be made without it; what they relied on instead was time, and in the trade’s case fermentation.

More acid is the direction in which the course has no published guidance and will not guess. What can be said is where the ceiling comes from: acid coagulates albumen, so somewhere above this dose the solution stops being a solution. Reilly gives 2 mL and no range. The Experiments section proposes a series rather than a number.

Why acetic acid rather than another one? Reilly names acids generally as the third denaturing treatment and gives acetic as his example. The course’s observation, marked as such, is that acetic acid is the one organic acid in this family that has no photographic side effect worth naming: it is weak, volatile, and — unlike citric acid — does not form a light-sensitive silver salt of its own, does not preserve the sensitised sheet, and does not push the print colour towards red. Citric acid in the albumen does all three, which makes it a different decision rather than a better acid, and it is set out under Variants.

Albumen, 1000 mL. Both the solvent and the second sensitive substance, which is why it is listed last and measured first.

As a binder it does what gelatin does in a plain salting solution, only far more of it: it holds the chloride, and later the silver chloride and the image silver, in a coherent transparent layer on top of the sheet instead of letting them soak into the fibres. That is where the gloss, the maximum density, the fine detail and the shorter required negative scale all come from.

As a reagent it forms silver albumenate on contact with the silver bath — insoluble, which is why the coating survives processing, and light-sensitive, which is why it contributes to the image.

More or less of it is not a matter of weighing differently; it is the dilution ladder, because the albumen is the bulk of the solution. Undiluted is glossy albumen paper. One part to one part of water is half-matte. One to six is almost indistinguishable from a matte salted paper. Reilly adds that even a 2 per cent solution of albumen improves depth and contrast over a paper with no organic binder at all, so there is no floor at which it stops earning its place, only a point below which you cannot see it. Going the other way, a second coat raises the gloss and the density and makes the sheet harder to tone and fix and more brittle.

Two of its properties are constraints rather than functions and both are worth knowing before you handle it: it coagulates above 65 °C, and it coagulates on contact with alcohol and with salts of metals. Both facts are used deliberately elsewhere in the process — alcohol to harden a first coat, heat in the modern kit’s dry-mount press — and both will ruin a batch by accident.

Water, 30 mL, added. Not a filler and not a make-up volume. It is the vehicle that lets 15 g of a solid and 2 mL of a corrosive concentrate enter a litre of protein already dissolved and already diluted, and its smallness is a statement of what the formula is for. Thirty millilitres is 2.9 per cent of the batch. Hardwich’s albumenised salting bath, by comparison, is 25 per cent water by volume, and it is a different material with a different surface. Reilly gives no temperature for it and none is needed; nothing here is warmed.

Interactions

With the silver bath, which is the interaction the formula is designed around. The two solutions never meet in a tray — they meet inside the coated sheet — and their ratio is the design of the process. Reilly’s rule of thumb is that the silver bath should be about four times the strength of the salting solution; his specific advice for albumen paper is 1.5 per cent chloride or above with a bath of 10 per cent or above, which is nearly seven times, and his reason is that the extra silver buys a better-looking and more permanent print. The historical alternative is on the record too: 1 to 1.5 per cent chloride with an 8 to 9 per cent bath, used between 1880 and 1900 to print the thinner negatives of the gelatin dry plate.

With the silver bath again, physically. The strength of the bath decides how fast it coagulates the albumen, and therefore how long the sheet must float. A strong bath coagulates the surface quickly and penetrates slowly, so it needs a longer float; a weak or freshly made bath coagulates slowly enough that a long float may start dissolving the albumen off the sheet. An older bath is better at this than a new one because it carries ammonium nitrate from the reaction itself.

With alcohol, deliberately. Between two coats the first has to be made insoluble or the second float simply dissolves it. Reilly’s practical route is a brief immersion in 70 per cent isopropyl alcohol, and he explains the choice of strength from both ends: pure alcohol is too strong and coagulates the layer unevenly, while too dilute a solution lets the albumen partially dissolve before it coagulates. The interaction that catches people out is the next sentence — whatever chloride the albumen contains must also be put into the alcohol, or the chloride leaches out of the coating into the bath. For this formula that is 1.5 per cent, not the 2 per cent his example happens to use. Wall’s 1912 version of the same trick is two volumes of alcohol to one of water, with ethanol as the rectified spirit of the period.

With heat, deliberately and accidentally. Drying at 30 to 50 °C was standard in the nineteenth-century factories and buys gloss and depth. The modern kit hardens each coat in a dry-mount press at 250 °F or under a hot iron. Both work by coagulating the surface — and both are the same property that ruins the liquid above 65 °C.

With the rawstock’s own sizing. The papers the trade used were machine-made all-rag stock sized internally with starch and resin soaps, chosen for wet strength, light weight and freedom from metallic impurities that would show as black specks. An alkaline albumen dissolves size; an acid one, on Hardwich’s account, does not. That is a direct interaction between the acid in this formula and the paper you choose.

With citric acid, if you add it. Citric acid in the albumen preserves the sensitised sheet, raises sensitivity and maximum density, and holds the print colour reddish brown even under prolonged gold toning. In the silver bath instead it does the same job and makes the colour browner. Either way, purple tones become difficult. This is a genuine three-way trade between keeping, colour and toning latitude, and it is decided at this tray as much as at the silver bath.

With the toner and the fixer, through thickness. Albumen becomes less permeable as it thickens, so a double-coated sheet tones and fixes more slowly than a single-coated one. Decisions made here set the times two operations later.

With its own sugar and its own sulfur, over decades. The glucose in egg white and the sulfur-containing side chains of the protein are both in the bottle and both are implicated in the yellowing of surviving prints, on the AIC’s account. Nothing in this formula removes either, and no source the course has read says whether fermentation did.

Variants

Blanquart-Evrard’s original, 1850. Reilly gives it in one sentence: white of egg beaten to a froth with 25 per cent by weight of a saturated salt solution, allowed to settle overnight, floated for one minute.

🧮What Blanquart-Evrard's 25 per cent actually means, and why the course does not resolve it

The phrase cannot be turned into a chloride percentage without deciding two things the sentence does not settle.

What is saturated brine? The sodium chloride entry gives 36.0 g per 100 g of water at 25 °C, so a saturated solution is 36/136 = 26.5 per cent salt by weight.

What is the 25 per cent reckoned on? If it is 25 g of brine per 100 g of egg white:

25 × 0.265 = 6.6 g salt in 125 g of mixture = 5.3 % w/w

Reading A: 25 parts brine to 100 parts egg white

If it is 25 g of brine per 100 g of the finished mixture:

25 × 0.265 = 6.6 g salt in 100 g of mixture = 6.6 % w/w

Reading B: brine is 25 per cent of the mixture

Both readings put the first albumen paper at three to four times the chloride of Reilly’s modern preparation, and the course cannot choose between them from the text it has. What survives the ambiguity is the direction, which is the useful part: the process began salt-rich and was worked down over the following decades as photographers learned what the excess was costing them in silver.

Towler, 1864 — the same undiluted case, with no acid. “Albumen can be used either pure or diluted. With pure albumen the prints are very brilliant, but the paper is not so easily prepared.” He then gives ten grains of chloride of ammonium to each fluid ounce of albumen, dissolved in the least quantity of distilled water, beaten to a thick white froth, stood ten minutes, skimmed onto a hair sieve, stood a day covered from dust, filtered through sponge, settled two more days and decanted.

10 grains ÷ 28.4131 mL = 0.648 g ÷ 28.4131 mL = 2.28 % w/v of the albumen

Towler's chloride, in modern units

Half as much again as Reilly’s, no acid at all, and a five-day preparation instead of an eight-day one. The conversions are the course’s arithmetic on Towler’s apothecaries’ measure, at 1 grain = 64.79891 mg and 1 imperial fluid ounce = 28.4131 mL.

Hardwich, 1864 — the same case again, at the other end of the range. In the middle of his diluted Formula I he adds the undiluted case in a sentence: “When pure Albumen is used without water, from 5 to 8 grains of Salt to each ounce will be sufficient.”

5 to 8 grains ÷ 28.4131 mL = 1.14 to 1.82 % w/v of the albumen

Hardwich's chloride for undiluted albumen

Reilly’s 1.5 per cent sits inside that band, 116 years later. Hardwich also gives the trade-off in the same breath, and it is the clearest statement of it anywhere on this page: the less the quantity of Salt the warmer the colour, but it must not be so far reduced as to injure the contrast and depth of shadow in the print.

Wall, 1912 — the alcohol version, and a diluted one. Albumen 6 ounces, chloride of ammonium or sodium 60 grains, rectified spirit 96 minims, distilled water 14 drachms; dissolve the salt in the spirit and water, add to the albumen, beat with an egg-whisk for fifteen minutes, settle, and filter through a tuft of well-washed cotton wool. That is about 1.7 per cent chloride on the whole liquid and it is a diluted albumen, so it belongs to the albumenised salting solution rather than here. It is worth noting for one detail: the alcohol is in the coating solution itself, doing at the outset a little of what Reilly’s 70 per cent isopropyl does between coats.

Sodium chloride in place of ammonium. Reilly’s own permitted substitution, at the same percentage, with similar colour and contrast expected. The one thing that changes is the by-product at the silver bath — sodium nitrate rather than ammonium nitrate — and both are recorded as having a coagulating effect on albumen.

Powdered albumen. Reilly records that dried albumen is available as a powder and that a 15 per cent solution in water approximates native egg white, while noting that it is more costly and less convenient than eggs. This is not a trivial variant: it is how the only commercially sold salted albumen solution the course has found is made, and it means the water content of a powder-based solution is a decision somebody has already taken for you.

🕰Fermented albumen, and why Dresden could be identified by the smell

The one-week refrigerated ageing in this formula is the tame descendant of something the trade did much harder. Photographers found independently, through the 1850s, that partially decomposed albumen gave a glossier and more even coating; decomposed albumen passes into an acid condition and loses the uneven viscosity and stringiness of native egg white. Some albumenisers went further and let the albumen ferment at elevated temperatures for several days, using the bacteria already in it.

Fermented albumen was in use as a collodion substratum on glass by the mid-1850s and became standard procedure in the Dresden factories that supplied most of the world’s albumen paper from the early 1870s. Reilly: Dresden paper could readily be identified by the smell. The fermented material gave glossier paper that toned more easily. Other European and American producers made good paper from aged but not actually fermented albumen — which is the tradition Reilly’s week in the refrigerator belongs to.

The scale is worth recording because it explains why this was an industry rather than a craft. One Dresden company produced 18,674 reams in 1888. At 480 sheets to the ream and 27 dozen eggs to every nine litres of albumen, that is about nine million sheets and over six million eggs, in one year, in one factory.

Citric acid in the albumen. Not a different acid but an additional one, and a different decision: it buys keeping time and sensitivity and maximum density, costs purple tones, and pushes the colour reddish brown. Reilly’s fullest figures are for citric acid in the silver bath — 5 per cent for maximum preservative effect, 1 per cent for a noticeable extension — and he states the albumen route qualitatively rather than numerically, so the course has no quantity to publish for it here.

The ready-made commercial solution. Sold as a salted albumen made from food-grade powdered egg whites with a food-grade preservative. The supplier publishes a working procedure — six-minute float, double-coated, each coat hardened in a dry-mount press at 250 °F, sensitised on a 15 per cent silver bath — and publishes no composition at all: no chloride content, no acid, no preservative named. Under this course’s rules that makes it a product rather than a formula, and the honest statement is that a six-minute float on an unknown liquid is not evidence about a one-and-a-half-minute float on this one.

Safety

Level A, with one declared Level B step, which is set out in full in the callout under Mixing: measuring 2 mL of glacial acetic acid.

Taking the ingredients in turn. Albumen has no GHS classification anywhere the course could look — no PubChem compound record, no ILO-WHO safety card, no entry in HSE’s EH40 list — and its encyclopaedia page is explicit that this is an absence of classification and not a finding about hazard. What controls it does need are microbiological and are dealt with under Storage. Ammonium chloride carries the signal word Warning with H302, harmful if swallowed, and H319, causes serious eye irritation; it is a solid handled with ordinary care and a technique that does not raise dust. Glacial acetic acid is the one reagent here that can injure you: Danger, H314, H226, and a concentrate that reaches a harmful air concentration quickly on evaporation at 20 °C. Handle it as the callout says and then put it away.

What is not a hazard here, and why. There is no silver in this solution, so nothing stains, nothing has to be collected as silver-bearing waste, and the whole silver nitrate regime belongs to the next bath rather than this one. Nothing here is light-sensitive, so there is no safelight requirement and no dim-light handling; this is a white-light operation from the eggs to the dried sheet. Nothing is heated, so there is no hot-work hazard and no thermal decomposition — the only temperature that matters is the 65 °C above which the albumen coagulates, which is a spoilage limit and not a safety one. There is no alkali, no oxidiser and no reducing agent, and therefore no route to a gas from anything in the tray. The one vapour in the room comes out of the acetic acid bottle before it is diluted, and it is gone as soon as the bottle is closed.

The hazard that is real and is not chemical. This formula involves raw eggs, a deliberately decomposing protein and a week of storage. Treat the whole operation as food handling done in a laboratory: wash your hands, keep the batch covered, use vessels that have never held food and never will, label them per the labelling SOP, and do not taste, smell deeply or store it where food is stored. The albumen encyclopaedia page reaches the same conclusion from the same absence of toxicological data, and its wording is the one to follow — a covered vessel, refrigeration, dated labels, and a low threshold for discarding a batch that has gone beyond the state the process wants.

What is downstream and is not Level A. Everything that makes this paper into a photograph: silver nitrate at the sensitising bench, an ultraviolet exposure, and a gold toner. None of them is in this bottle, and the fact that this step is safe should not be read as a statement about the process.

Storage

The prepared solution. Covered, refrigerated, dated. Reilly’s schedule is 24 hours of settling, then a week in the refrigerator before use, then several weeks of remaining useful — and no number for the end of it, because there cannot be one. The material is wanted partly decomposed, so the boundary between a batch at its best and a batch that has gone is a judgement rather than a rule. Reilly gives the three senses to judge it by: the smell, the colour and the sedimentation. His target state is a yellowish homogeneous liquid with a slight aged odour. What you are looking for is a change of degree, and the honest advice is that you will not recognise it the first time and will never miss it afterwards.

Use the refrigerated labelling SOP: the date it was mixed, the date it came out of the settling, the chloride percentage and the acid, because a bottle of yellowish liquid in a refrigerator is indistinguishable from any other bottle of yellowish liquid.

Make what you will use, but understand what “use” means. The tray arithmetic under Behaviour is the governing consideration: a session needs about a litre in the tray whether it coats one sheet or twenty, and what does not go onto paper goes back into the bottle. Filter it on the way back in.

The albumenised sheet. Keeps very well, cool and dry, and nothing on it is light-sensitive yet. Press the sheets flat under weights, which also makes them supple enough to handle for the next float. If they must be rolled, roll them albumen side out, so that the layer is stretched rather than compressed and cracks less. Do not let them get bone dry before sensitising: an over-dry sheet will not properly absorb the silver solution, and Reilly’s remedy is a night in a damp place or a few hours in a closed box with a dish of water.

The sensitised sheet. Sensitise, print and process on the same day. That is a different page’s problem and it starts at a different tray, but it is the reason albumenising is worth doing in batches: the operation with no deadline should be separated from the operation with a twelve-hour one.

The dry chemicals. Ammonium chloride takes up water from the air and cakes; a closed container. The glacial acetic acid needs its own attention for a reason that is not a hazard: it freezes solid at moderately low temperatures, which is why it is called glacial and why a bottle in an unheated darkroom in winter can be a solid block when you want 2 mL out of it. Warm it gently in a water bath rather than attempting to chip it.

Incompatibilities

Silver nitrate, absolutely and by design. The two halves of this process must meet only inside the paper. A drop of silver solution in the albumen tray coagulates the protein where it lands and precipitates silver chloride at the same time, and the resulting stringy mess is not recoverable. Separate trays, separate funnels, separate measuring vessels — the supplier’s kit ships two funnels specifically so that neither is ever the wrong one — and the albumen bench kept away from the silver bench. See incompatibilities.

Alcohol, except where you mean it. Albumen is insoluble in alcohol and is coagulated by it. That is the basis of the hardening step between two coats and it is also a way to destroy a batch with a carelessly rinsed measuring cylinder.

Salts of metals, generally. Reilly’s list of what coagulates albumen is alcohol, temperatures above 65 °C, and contact with salts of metals. Silver nitrate is the case the process uses; the rest are cases to keep out of the tray. A metal tray, a metal spatula or a metal whisk is not a good idea, and the nineteenth-century preoccupation with metal flecks in the rawstock is the same worry one step earlier.

Heat above 65 °C. Not a hazard, an irreversibility. There is no way back from a coagulated batch.

Alkali. An alkaline albumen dissolves the size out of the paper and sinks into the fibres instead of sitting on them, which is the fault the acid in this formula exists to prevent. Anything that raises the pH of the tray — a carbonate carried in on a wet hand, an ammoniacal solution nearby — works directly against the formula.

Yolk, blood and the chalazae, which is a contamination incompatibility rather than a chemical one but destroys a batch just as thoroughly. So does fixer carried backwards on a wet hand, which does nothing visible to the albumen and shows up as a dead spot on a print two operations later.

Waste

This is the one tray in the albumen workflow that is not a silver stream, and knowing which streams are which is most of what responsible practice means here. Spent albumen is dilute protein, a little ammonium chloride, a trace of acetic acid and some paper fibre. It carries no silver, no heavy metal, no thiosulfate and no chromium.

What it does carry is organic load: protein in water is oxygen-demanding, and a decomposing batch more so. The supplier of the ready-made solution instructs disposal by pouring down the drain with plenty of cold water, which is a manufacturer’s instruction for their own product in their own jurisdiction and is recorded here as that rather than as a rule. Local regulation decides what may go to drain where you are; see disposal and the general chemical waste SOP.

Where the silver actually goes is the sensitising bath, the first wash after exposure — the clouding of that water is excess silver nitrate meeting the ions in tap water — the toner and the fixer. The silver-bearing waste SOP governs all four. None of it applies to this tray, and pretending otherwise blunts the rule where it matters.

The eggs are the largest waste stream by mass, and the nineteenth century had a better answer than we do. Reilly records that the yolks were salted and sold to bakeries and to tanners, who used them to finish kid leathers, and that the money from yolks was a significant line in the accounts of the Dresden factories. Thirty-five yolks is a serious quantity of food. The shells and any spoiled batch are ordinary domestic organic waste.

Troubleshooting

Streaky lines on the dried sheet, which bronze strongly under light when the paper is sensitised. Two causes and they are distinguishable. If the streaks are fine and irregular, the albumen was not thoroughly beaten and flakes of membrane were left in it — Hardwich’s diagnosis, and the reason the straining step exists. If they are single hard lines, the sheet was paused on the way down; Hardwich’s remedy is to lower the paper in one steady movement, because if a pause is made a line will be formed.

Paper-white circles, sometimes with a light brown stain around them. Bubbles under the sheet during the albumen float, so that area got no binder and no chloride, and the silver nitrate that arrived later had nothing to react with and stained the paper on its own. Sweep the tray surface before every sheet, add the surfactant, and start timing only when the sheet lies flat with no bubbles. With thin paper the bubbles show through as light circles on a dark ground; with thicker paper you must lift each sheet and look at its underside.

A coating that is thin at the top of the hung sheet and thick at the bottom. Gravity, and it is unavoidable rather than a fault — Reilly says there will always be some difference. Reduce it: hang the sheet by its broadside so the run is as short as possible, which is Towler’s instruction from 1864; blot the accumulating bead several times with a cloth as it dries, or the last droplet forms a thick rind that makes the sheet very difficult to handle; and if you are double coating, mark the lowest edge and hang it uppermost the second time so the two gradients cancel.

The albumen will not wet the paper, and the coat comes out patchy. Towler’s own observation, and he blames dry paper and dry weather. The solution may also simply be cold: the ready-made solution’s supplier states that many papers resist absorbing cold albumen and asks for it to be out of the refrigerator two hours before use. The surfactant addresses the same problem chemically; Hardwich’s period answer was a few drops of a spirituous solution of bile, or a fragment of prepared ox-gall.

The second coat dissolves the first, and the sheet is no thicker than it was. The hardening step was omitted or was ineffective. Reilly is explicit that some hardening or coagulating step is necessary between coats — 70 per cent isopropyl alcohol carrying the same chloride content as the albumen, a current of steam, or in the trade six months in a warm loft.

The albumen dissolves off the sheet in the silver bath. The bath is too weak, or too freshly made, or the float was too long for its strength. An older bath containing accumulated ammonium nitrate coagulates better than a new one. Reilly’s floor for a modern worker is a 10 per cent bath.

The highlights go yellow immediately after sensitising, before any printing. Reilly’s account of why glossy papers are made from acidified, partly decomposed albumen names exactly this: fresh alkaline albumen has a greater tendency to yellow after sensitisation. A batch used too soon, or under-acidified, is the first suspect. Do not confuse it with the slow yellowing of a finished print over decades, which has different causes and is discussed under Image characteristics.

Flat prints without depth, on a well-coated sheet. Work backwards along the chain: the negative’s density range first, because this paper needs a long one; then the silver bath strength, because free silver nitrate is what builds density; then the chloride, and only then the albumen. Reilly’s warning is that lowering the chloride costs a rich dense image, and that raising it above what is necessary buys nothing but silver consumption.

The batch smells wrong, has separated, or has grown a haze. It has gone past the state the process wants. Reilly names smell, colour and sedimentation as the three indicators and gives no rule for reading them; discard it and make another. Eggs are cheap and a session is not.

Experiments

The chloride series, which is the experiment this page exists for. Four batches of 250 mL of albumen, salted at 1.0, 1.5, 2.0 and 2.5 per cent, with the acid held at 2 mL per litre and everything downstream identical, sensitised on the same 10 per cent silver bath and printed from the same negative to the same visual endpoint. You are testing Reilly’s own account against your own paper: that below 1.5 per cent the paper is less sensitive and slightly more contrasty from a thin negative but gives a thinner image, and that above the optimum the extra chloride only consumes silver. Record the exposures, because they will differ, and keep the strips: this series is also the fastest way to find the chloride your rawstock actually wants.

The acid series, which nobody has published. The same albumen, split four ways, at 0, 1, 2 and 4 mL of glacial acetic acid per litre. Measure four things that are all cheap to measure: the froth volume immediately after beating, the time the froth takes to liquefy, the pH of each batch after a week — the number no source on this page publishes — and, at the end, whether the sensitised sheet yellows before it is printed. The 0 mL batch is the control and is also Towler’s and Hardwich’s formula, so a null result is a historically interesting one rather than a failure. This design is the course’s, not Reilly’s; he gives one figure and no range.

The ageing series. One batch, coated at day 1, day 8, day 21 and day 45. Reilly says a week and then several weeks; this measures what “several” is on your bench, at your refrigerator’s temperature, which is the variable he cannot know. Judge on gloss, on coating evenness, on yellowing after sensitisation, and on maximum density in the finished print.

The dilution ladder, which turns this page into its neighbour. Undiluted, 1:1 and 1:6 with water, with the chloride corrected each time so that all three carry the same percentage of the whole solution. Reilly predicts glossy, half-matte and near-matte. Print all three from one negative and measure the maximum density and the scale length of each with a step tablet. The chloride correction is the part that matters methodologically: without it you would be changing two variables and learning nothing about either.

Single coat against double coat. One negative, four sheets: single-coated, double-coated with the alcohol hardening, and — if you can spare the time — a pair coated in the two possible hanging orientations, to see how much of the gravity gradient a reversed second coat actually cancels. Time the toning and the fixing of each, because Reilly’s claim that thicker albumen is less permeable is directly testable with a clock.

Ammonium against sodium chloride, at the same percentage, everything else held. Reilly expects similar colour and contrast. If you can smell free ammonia over the ammonium batch during beating and not over the sodium one, you have reproduced Hardwich’s 1864 observation with no equipment at all.

Measure the scale length properly, with a 21-step tablet. Sensitise a test sheet, print it under the tablet until the step-1 patch matches the density of the margin outside it, process normally and count the steps you can distinguish. That count is the paper’s scale length and it tells you what density range the negatives for this paper should have. It is the only way to stop guessing, and it is the measurement that lets you test Hübl’s finding that glossy albumen matches platinum paper and that the paper’s tonal progression is slow in the shadows and abrupt at the top.