Albumenised salting solution

Three salting solutions in this formulary put a chloride into a sheet of paper and a binder around it. The plain one uses two grams of gelatin per litre — just enough to keep the image out of the fibres. The arrowroot one uses boiled starch, which does the physical job well and no chemical job at all, and has to carry an acid to make up for it. This one uses egg white, which does both jobs at once and is therefore not really a binder with a chloride dissolved in it: it is a second light-sensitive substance that happens also to be transparent, and the water you add to it is the control.

Ingredient	As Hardwich prints it	Metric
Chloride of ammonium	200 grains	12.96 g
Water	5 fluid ounces	142 mL, added
Albumen	15 fluid ounces	426 mL

Sum of the stated volumes, 568 mL. Chloride, 2.28 per cent w/v of that sum. Albumen, 75 per cent of the liquid by volume — three parts to one of water. Both percentages are the course’s arithmetic on Hardwich’s three lines; he states neither.

!Which of the albumen entries is this?

The register holds two solutions of salted egg white and they are not variants of one another.

This page is the diluted bath — the albumenised salting solution, in which water is a named ingredient and the amount of it decides whether the finished print is glossy, half-matte or indistinguishable from a salt print. The formula of record is Hardwich’s, from 1864, and the page is really about the axis rather than the single point on it.

The albumen solution with its chloride is the undiluted bath as a modern worker prepares it, aged on purpose, and it makes glossy albumen paper.

Neither is a modification of the other. They are two points on one line, and the line is how much water is in the egg white.

Purpose

To do in a single float what the plain and arrowroot salting solutions do with a binder that is inert or nearly so: put a chloride into the surface of the sheet, so that a later bath of silver nitrate can precipitate silver chloride in place; and put a transparent protein layer around it, so that the silver chloride forms in the albumen rather than in the paper.

The first purpose is common to the whole family and is set out in full on the plain salting solution page. Silver chloride is insoluble; it cannot be dissolved and coated; it has to be assembled inside the sheet from two solutions applied and dried in turn.

The second purpose is where the albumen earns its keep, and it does two different things at once that are easy to run together.

Physically, it is a transparent film. Reilly’s account of why that matters is the clearest in the literature: if the light-sensitive layer penetrates into the paper fibres, the maximum density obtainable falls and the print is matte, because light reflected from the paper is scattered and diffused by the fibres. If the layer sits compactly on top of the fibres, that scattering is minimised. And if the layer is itself smooth and transparent, the scattering in the white areas is minimised as well, so the paper looks more brilliant and carries more contrast. Detail comes with it.

Chemically, it is one of Reilly’s “active” organic substances — albumen, gelatin, and the organic acids citric, tartaric and oxalic. Active substances facilitate the more complete reduction of silver chloride, and they form light-sensitive silver salts of their own on contact with silver nitrate. Albumen’s is silver albumenate, and Reilly is explicit that this substance is itself light sensitive and makes an important contribution to image formation in albumen paper. Starch does none of this; gelatin does some of it; albumen does the most of it.

Hardwich reached the same division in 1864 by experiment rather than by classification, and his statement of it is worth having in his own terms:

the function of the Chloride is to impart sensitiveness in Photographic printing, and that of the Nitrate of Silver to give intensity, the Organic Matter acts by brightening the colour … organic compounds of Silver in the paper, produced by adding Albumen or similar substances to the salting bath, will afford him the means of obtaining these varied tones, and that without them the picture will lack richness of effect.

Which leaves the third purpose, the one that gives this entry its own page: to set the amount of albumen. Hardwich’s five fluid ounces of water are not a solvent of convenience. They are the difference between a glossy print and a matte one, and Reilly names the positions on the scale — pure albumen is the familiar glossy paper, a 1:1 dilution is half-matte, a 1:6 dilution gives a paper almost indistinguishable from other matte salted papers. The Getty’s atlas confirms from the other end that this was the main axis of variation in the historical literature: “the major difference between many of the different published recipes lay in whether the albumen salt solution was used as prepared, or if it was diluted with various amounts of water.”

Recommended uses

An albumenised salted paper of moderate gloss, from a negative of moderate range. At three parts albumen to one of water this bath gives a paper that has clearly left the plain salted class — the image sits up on the surface, the whites are cleaner and the detail is finer — without the mirror gloss of a double-coated albumen sheet. Hardwich says the difference between neat and diluted albumen in the plainest terms available: albumen alone, without any addition of water, “gives a more highly varnished appearance.”

Where the negative will not stand a matte paper. This is the practical headline, and it runs the other way from the way beginners expect. A glossy paper needs a shorter-range negative, not a longer one, because its transparent binder stops the paper fibres scattering light, so the whites read whiter and the shadows denser without any more silver being there. Reilly’s ranking of the required density range is plain salted paper greatest, arrowroot and matte albumen slightly less, glossy albumen lower again. An albumenised paper takes its place on that ladder according to how much water went into the bath, which means the bath is a contrast control you operate a week before you print.

Where you want the albumen’s colour. Hardwich’s ranking of the reddening action of the three binders is starch least, gelatine more, albumen most, with surface brilliancy following the same order. A print made on this bath and fixed without toning is a warm red-brown; the same negative on an arrowroot paper is yellower and flatter. Since the whole point of a warm red print is that it will go a full brown or black in a gold toner, the binder choice is made at the toning stage as much as at the coating stage.

In white light, at leisure, in batches. Nothing in this solution is light-sensitive. The salting and albumenising of a dozen sheets is an afternoon’s kitchen work, and the sheets then keep — “a long time in a dry place”, says Hardwich — until you have a negative worth silvering one for.

Where eggs are cheaper than gelatin sheets and starch is not what you want. Hardwich reckons one fluid ounce of white per egg, Wall about seven drachms, so this formula is fifteen to seventeen eggs. Wall’s smaller batch — six ounces of albumen, about seven eggs — he says is “sufficient for a quire”, twenty-four sheets.

When another formula is preferable

• For the simplest possible salt print, the plain salting solution. Twenty grams of salt and two of gelatin in a litre, no eggs, no beating, no settling, no smell. It gives a flatter, greyer, more sunken print with the longest tonal scale of the family, and for a great many pictures that is the better print.
• For a matte surface with more brilliance than plain salted paper, the arrowroot salting solution. It reaches part of the way towards this one without the protein, and pays for it with three grams of citric acid, because a starch binder contributes nothing chemically.
• For a full glossy albumen print, the albumen solution with its chloride — the undiluted, aged bath — and, if you want the surface the 1880s trade sold, the double-coated procedure on top of it. Reilly’s fresh-versus-aged distinction matters here and is not a detail: glossy papers are prepared from partially decomposed, acidic albumen because in that state it coats more evenly, gives more gloss and yellows less after sensitising.
• For a matte paper that is still an albumen paper, Hübl’s matte albumen — equal volumes of fresh albumen and a salted arrowroot solution. It reaches the matte end of the range by adding starch rather than water, which is a genuinely different answer and gives a genuinely different surface. It is set out under Variants, with the disagreement between its two printings.
• If what you want is a hand-coated print rather than specifically a silver one, the classic cyanotype sensitiser is one coating instead of two, needs no fixer, keeps no eggs and costs a fraction as much.

Mixing

One vessel, and the order is not optional. Unlike its two siblings, this formula is not two solutions that meet at the end. It is a salt dissolved in a little water and then stirred into a protein, and everything after that is mechanical.

Dissolve the chloride in the water first. Twelve point nine six grams of ammonium chloride into the 142 mL, and wait until the solution is clear. Hardwich lists the ingredients in this order and does not say why; Reilly says why, in his own procedure — the chloride “should be dissolved in a minimum of water and added to the egg white before the beating process” — and Wall prints the instruction outright: “Dissolve the salt in the spirit and water, add to the albumen.” Solid chloride tipped straight into egg white dissolves at a useless rate and coagulates the protein locally around each crystal, giving you specks that will survive the beating and end up as marks on the paper.

Separate the eggs cleanly, one at a time. Fifteen eggs at Hardwich’s rate. Wall’s instruction is the sound one and is about damage limitation: “Crack each egg into a separate cup or measure before mixing with the bulk, so that in case of the yolk breaking the whole of the albumen may not be spoilt.” Take out the germ — Towler removes it with a glass rod, Wall says take it out, Reilly says leave no traces of yolk or blood behind. Yolk is fat; fat is exactly what a coating solution must not carry.

Add the salted water to the albumen, then beat the whole to a froth. Hardwich: “take a bundle of quills or a fork, and beat the whole into a perfect froth.” Wall: an egg-whisk, fifteen minutes. Reilly, with a modern kitchen available: an electric mixer or blender, three minutes, or until the entire mixture has been converted to froth.

The beating is not aeration for its own sake and it is not optional. Egg white is a mixture of proteins of different viscosities, and Reilly’s account is that beating denatures them and produces a homogeneous liquid that will form an even layer on the sheet — where native egg white, being stringy and unevenly viscous, will not. Hardwich states the failure mode from the other side: “if the Albumen be not thoroughly beaten, flakes of animal membrane will be left in the liquid, and will cause streaks upon the paper.” The chlorides help. Reilly notes that one of the effects of adding them is to reduce the volume of froth produced, because they are themselves denaturing the protein while you beat.

⚗Why the froth has to collapse again

A froth is not a coating solution. What you are after is what the froth turns back into when it is left alone — a liquid whose proteins have all been through the same mechanical shear and have come out at the same viscosity. Hardwich’s procedure separates the two stages deliberately: skim the froth off as it forms into a flat dish to subside, and only when it has partially subsided transfer it to a tall narrow jar and leave it several hours so that the membranous shreds settle to the bottom. Then pour off the clear upper portion. He adds the reason you do not simply filter it: “Albuminous liquids are too glutinous to run well through a paper filter, and are better cleared by subsidence.”

He also offers a shortcut that avoids the froth altogether — fill a bottle about three parts full with the salted mixture of albumen and water and shake it for ten to fifteen minutes “until it loses its glutinosity and can be poured out smoothly from the neck of the bottle” — while noting that makers of albumenised paper did not much use it.

Let it stand, then clear it. Hardwich: several hours in a tall jar, decant. Wall: allow to settle, then filter through a tuft of cotton-wool that has itself been well washed with distilled water. Towler is the most patient of the three — a day covered from dust, then a filtering through sponge, then a further two days’ settling before the supernatant is poured off. Reilly’s modern version is 24 hours covered, strained through muslin under pressure.

Do not warm anything. There is no heating step anywhere in this formula and there cannot be: Reilly gives 65 °C as the temperature above which albumen coagulates, and a coagulated salting bath is scrap. This is the one obvious respect in which the operation is easier than either of its siblings, both of which want hot water and one of which wants a boil.

Bring the bath and the paper to the same temperature before coating. Reilly’s instruction for albumen, and the general one for floating: a solution warmer than the room makes the sheets curl ferociously the moment they touch it.

Behaviour

It floats a sheet in eighty to ninety seconds, and the upper limit matters more than the number. Hardwich gives one minute and a half, Wall eighty seconds, Reilly one to one and a half minutes for undiluted albumen. Three independent sources across 116 years agree within ten seconds, which is more than can be said for most figures in this part of the formulary. But both period sources phrase it as a maximum and give the same reason. Hardwich:

The paper must not be allowed to rest upon the Salting Bath much longer than the time specified, because the solution of Albumen being alkaline … tends to remove the size from the paper and to sink in too deeply: thus losing its surface gloss.

Wall, in seven words: “not longer, or the albumen will sink into the body of paper.” Over-floating does not give you more albumen on the sheet. It gives you less, in the wrong place.

It is fussier to float than a plain salting solution, and everyone who wrote about it said so. Towler: “Much more care is required in the successful management of laying the paper on the salted albumen than upon the plain salting solution, for bubbles are more likely to be formed, and are less easily removed.” Wall gives the useful diagnostic — bubbles show as “numerous puckers” in the sheet within the first seconds, and are wetted down with a camel’s-hair brush. Hardwich’s method is to bow the sheet convexity downwards and lower the centre first so that the air is pushed outwards, and to replace any sheet that shows circular albumen-free spots for the same length of time again.

It runs downhill while it dries, and the bath’s own viscosity is why. Towler describes the consequence precisely: in dry weather the albumen does not attach itself easily to a dry sheet, and the coating flows off the upper part so that “the film on the upper part is much thinner than on the lower part, and a number of irregular marks and curves are apt to be formed on the lower part.” His remedy is to hang the sheet by its broadside, so that the distance the liquid can run is as short as possible. Hardwich’s runoff bead is blotted off repeatedly; Reilly’s is too, and he warns that a dried bead becomes “a thick rind of albumen” that makes the sheet unmanageable in later operations.

It takes about half as much again out of the tray as a plain salting solution does. Hardwich measured this, which almost nobody did, and the measurement is like for like because his plain paper formula carries the same 200 grains of chloride against the same 20 ounces of stated liquid. A quarter-sheet of 11 × 9 inches removes one and a half fluid drachms from the albumenised bath and only one drachm from the plain one, taking about a grain and three quarters of chloride with it — “the glutinous nature of the Albumen causes a third part more of Salt to be retained by the paper.”

It is not light-sensitive and it does not need a darkroom. The salting, the settling, the floating and the drying are all white-light operations, because there is no silver in the room yet. Everything changes at the sensitiser.

The coated sheet keeps and the coated sheet is fragile. Hardwich: “Albuminized paper will keep a long time in a dry place.” He adds a small practical instruction that reveals what the dried layer is like — cut it with a bone spatula, do not touch the albumenised surface with the fingers, press the sheets flat between boards, and avoid rolling them. Reilly, if they must be rolled, says roll them albumen side out, so the layer is stretched rather than compressed and cracks less.

Image characteristics

Surface: whatever the water says. This is the characteristic the formula exists to control. Hardwich’s own dilution is glossy but not varnished. Reilly’s ladder gives the rest of the range, and the Getty’s analytical categories give the same range read backwards off a finished print: an albumen-rich or double-coated print, a single-coated albumen print, and — for prints containing much less albumen, “probably produced using diluted albumen” — the category the atlas explicitly calls albumenized photographs. This entry is the bath that makes the third of those on purpose.

Contrast: intermediate, and the negative must match. Wall puts matte albumen paper as giving “results intermediate between albumenised and plain salted paper”, which is a period statement of the same ladder. Reilly gives the mechanism: a glossy paper’s transparent binder minimises diffuse reflection and scattering by the fibres, so the same quantity of reduced silver reads as a deeper shadow, and the negative may accordingly be shorter in range. Dilute the albumen and you push the paper back up the ladder towards the long-scale negative a plain salted paper wants.

Tonality: long, self-masking, and weighted towards the shadows. Every printing-out paper has the self-masking property Reilly describes — the silver building in the shadows behaves as extra density in the negative and delays the arrival of maximum density, so highlight detail has time to print in and prints avoid the soot-and-chalk failure of develop-out papers. Hübl’s own measurements, reported by Reilly, put the scale length of glossy albumen paper equal to that of platinum paper, with salted papers exceeding both by a considerable margin. Hübl also found that albumen has a slow progression from shadows to middletones and then an abrupt jump from middletones to white, from which Reilly draws the practical conclusion that a negative intended for albumen printing should emphasise highlight detail at the expense of shadow detail.

Colour: warm red or reddish brown untoned; whatever the toner says afterwards. Reilly’s explanation of why printing-out prints are this colour at all is worth having, because it is not obvious: the image is colloidal silver, not filamentary, and colloidal particles absorb some wavelengths and not others depending in part on the refractive index of the medium they sit in. Change the binder and you change the colour. Fix the print and you change it again, because dissolving the unreduced silver chloride changes the refractive index of the whole system and packs the particles closer. This is precisely why an albumenised print is a different colour from an arrowroot print made from the same negative on the same day: the medium is different.

Speed: slow, and slower the more albumen there is. Reilly’s ranking is matte salted papers fastest, plain salted papers next, albumen slowest. This looks like a contradiction of the contrast ranking and is not. A matte print needs more reduced silver to look black, so it needs a long exposure; a glossy print needs less silver to look black, but the albumen layer is also slower to build what silver it does build. The two effects are separate and both are real.

The mechanism

The chloride, and why it is not enough on its own

The light-sensitive substance is silver chloride, formed inside the coated layer when the sheet is floated on the silver bath. The reaction is a double replacement, and with an ammonium chloride salting bath it runs:

NH₄Cl + AgNO₃ → AgCl + NH₄NO₃

Silver chloride is formed in place; the ammonium nitrate stays in solution

The ammonium nitrate has no part in the image. Some of it washes out and some of it accumulates in the silver bath, where — as Reilly notes — it does one useful thing, which is to help coagulate the albumen of later sheets.

Silver chloride by itself makes a poor print. Reilly: pure silver chloride paper “produces gray and flat images”, and successful printing depends on two further things, an excess of silver nitrate and an active organic substance. The excess is Talbot’s discovery, and Reilly quantifies it: Talbot found that when chloride and silver nitrate were present in equal amounts his papers were hardly sensitive at all, and that about six times more nitrate was necessary, establishing a salting solution of 2 to 4 per cent against a silver bath of 12 per cent.

Vogel supplied the reason, and it is the most useful single idea in printing-out chemistry. Light splits a unit of silver chloride; the liberated chlorine, instead of escaping, meets the excess silver nitrate sitting beside it and forms fresh silver chloride; light splits that; and the cycle repeats. Where there is a large excess of silver nitrate, more image silver is formed and a greater maximum density is reached. Where there is not, the reaction stalls and the print is grey.

Hardwich, without the vocabulary, had assembled the same picture experimentally by 1864, and his strip test is still the clearest demonstration of it: prepare paper strips with washes of silver nitrate at 5, 10, 20, 40, 80 and 100 grains to the ounce, print them all under the same negative, and watch the pictures grow more vigorous with the strength of the bath — up to a point beyond which further increase adds nothing.

The albumen, and what it does that starch cannot

Now put the protein in. Three things happen that would not happen with starch.

It is coagulated by the silver bath and becomes insoluble. Reilly: albumen is coagulated by alcohol, by temperatures above 65 °C and by contact with salts of metals, and “the reason why albumen does not dissolve off the sheet during processing is because contact with silver nitrate in the sensitizing bath coagulates it”. A starch layer is never coagulated by anything and stays permeable throughout, which is why an arrowroot paper must be silvered briefly on a strong bath or the silver chloride migrates down into the fibres. An albumenised sheet sets itself.

It forms silver albumenate, which prints out. The coagulation is not merely physical. It produces a new insoluble silver–protein complex which Reilly names silver albumenate and describes as “itself light sensitive”, making “an important contribution to image formation in albumen paper”. So an albumenised paper is printing on two substances at once, and part of the image is not silver chloride at all. This is what “active organic substance” means in Reilly’s classification, and it is the difference between this bath and the arrowroot one that no amount of coating skill can make up.

It is alkaline, and that has consequences the moment the chloride goes in. Reilly gives the pH of native egg white as 7.8. Hardwich, without a pH meter, had identified both the fact and the cause:

Albumen is not a neutral fluid, but possesses an alkaline reaction, due to the presence of a small quantity of Soda; hence, on adding Chloride of Ammonium to Albumen a development of free Ammonia takes place, easily perceptible to the smell, and Ammonia is a solvent of the materials used in sizing paper.

That is a correct piece of chemistry from 1864 and it explains a practical rule. Ammonium chloride is the salt of a weak base and a strong acid; put it into an alkaline solution and it gives up ammonia:

NH₄⁺ + OH⁻ → NH₃ + H₂O

Why a freshly salted albumen bath smells of ammonia

The free ammonia is why Hardwich’s float is a maximum and not a target: leave the sheet on the bath and the alkali attacks the paper’s own size, the albumen sinks into the body of the sheet, and the gloss you were coating for is lost. Sodium chloride does not do this — it is the salt of a strong acid and a strong base, which is elementary chemistry rather than a statement of Hardwich’s — and that is one real difference between the two chlorides that the sources otherwise treat as interchangeable.

The dilution, and why it is a contrast control

The chain runs: water in the bath → albumen on the sheet → where the image sits → how much light the paper scatters → what negative you need.

More water means less albumen per unit area when the sheet dries. Less albumen means the silver chloride forms partly in the fibres rather than wholly in a film above them. An image in the fibres is lit by light that has been scattered and diffused by cellulose, so its maximum density is lower and its whites are duller — a matte print. To make that matte print look as black as the glossy one, you need more reduced silver, which means a longer exposure, which means a negative with enough highlight density to hold the light tones back while you make it. Hence Reilly’s ranking of required negative density range, and hence the counter-intuitive rule that the glossier paper is the one for the flatter negative.

That whole chain is operated by one number in this formula, and the number is 142 mL.

Function of every ingredient

Albumen — 426 mL, fifteen fluid ounces, about fifteen eggs

What it is. The clear white of a hen’s egg: not one substance but a collection of proteins, referred to collectively as albumen. Reilly gives its specific gravity as 1.040 and notes that it dries at room temperature to a brittle, transparent mass. It is also sold dried, as a powder, of which a 15 per cent solution in water approximates native egg white — more costly and less convenient, but available to anyone who cannot face separating fifteen eggs. Its chemical identity, such as it is, is on the albumen page.

Why it is here. For the two jobs described under The mechanism: it is a transparent film that keeps the image out of the paper, and it is an active organic substance that forms light-sensitive silver albumenate and assists the reduction of silver chloride. No other binder in this formulary does both.

What it does chemically. On meeting the silver bath it coagulates and forms silver albumenate, an insoluble complex which is itself light-sensitive. Before that, in the bath, it is being deliberately damaged: Reilly names three denaturing treatments — beating, the chlorides themselves, and acidification — and says flatly that albumen is never used in the strictly native condition. Its alkalinity (pH 7.8) liberates ammonia from the ammonium chloride, which is a nuisance during floating and a positive danger if the sheet is left too long.

What happens if you use more or less. This is the whole formula, so the answer is long, and it is the dilution ladder set out in this entry’s own data. Briefly: more albumen gives more gloss, more detail, more brilliance, more contrast, a shorter-range negative, a redder untoned print and a slower paper. Less gives the reverse. Reilly’s ladder — pure, 1:1 half-matte, 1:6 nearly matte — is the map, and even a 2 per cent albumen solution measurably improves depth and contrast over a paper with no organic binder at all.

What it interacts with. Everything. It is coagulated by alcohol, by heat above 65 °C and by metal salts — which is why the double-coating procedure works, why nothing is warmed, and why the silver bath sets it. It is dissolved by ammoniacal silver, which rules out one whole class of historical sensitiser. It carries its own sulfur, which over decades becomes silver sulfide in the highlights and is one of the proposed causes of the yellowing of surviving albumen prints. And it takes about a third more chloride onto the sheet than a gelatin bath of the same strength would, because it is more viscous.

Ammonium chloride — 12.96 g, 200 grains, 2.28 per cent of the stated liquid

What it is. Sal ammoniac, NH₄Cl: the acidic salt of a weak base and a strong acid, and the chloride the nineteenth century reached for by default. Reilly: “ammonium chloride was the most commonly used in the 19th century”. Its properties and hazards are on the ammonium chloride page.

Why it is here. It is the halide. Without a soluble chloride in the sheet there is no silver chloride to form, and the silver bath would give at best the feeble print-out of silver nitrate on cellulose.

What it does chemically. Two things. It supplies chloride ion for the double replacement with silver nitrate. And, being an ammonium salt in an alkaline protein solution, it acts as one of Reilly’s three denaturing agents: it helps break the protein down towards the homogeneous liquid the coating wants, and Reilly notes the visible sign of it, which is that the chlorides reduce the volume of froth the beating produces.

What happens if you use more or less. The sources agree closely and one of them is very old. Reilly’s modern statement is the fullest: papers with a low chloride content of 1 to 1.5 per cent are less sensitive and tend to produce slightly more contrasty prints from thin negatives, but the gain comes at the expense of a rich, dense image, so ordinarily the chloride should be kept at 1.5 per cent or above; and using more chloride than is necessary buys nothing but higher silver consumption. Hardwich, a century earlier, states the colour half of the same trade-off in one sentence: “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.”

Hardwich’s 2.28 per cent sits at the top of Reilly’s normal band of 1.5 to 2.5 per cent, and Wall’s 1.72 per cent sits in the middle of it. Towler’s independent 1864 figure for pure albumen — ten grains of chloride of ammonium to each ounce — works out at 2.28 per cent as well, by the same conversions. Four sources, three centuries of practice between them, one band.

What it interacts with. The albumen, alkaline, which it partly denatures and from which it liberates ammonia. The silver bath, whose strength has to be matched to it: Reilly’s rule of thumb is that the sensitiser should be about four times the strength of the salting solution, which for this bath would put the silver at about 9 per cent. Both period sources use far more than that, and the discrepancy is discussed under Interactions. And the paper’s own sizing, which the liberated ammonia will attack if the sheet floats too long.

Water — 142 mL, five fluid ounces

What it is. The vehicle for the chloride and the whole of the dilution. Wall specifies distilled; Hardwich does not, though he requires distilled or at least rain-water for the silver bath and notes that spring water “will often answer” there.

Why it is here. Two jobs, and it is worth separating them because they pull in opposite directions. It dissolves the chloride so the chloride can be distributed evenly through the albumen — that job could be done by a much smaller volume, and Reilly’s undiluted formula uses only 30 mL per litre of albumen for exactly that purpose. And it dilutes the albumen, which is a decision about the finished print and not a matter of convenience at all.

What it does chemically. Nothing directly. Everything indirectly, by setting the concentration of both other ingredients at once — which is the trap in this formula and is dealt with under Interactions.

What happens if you use more or less. More water: a thinner albumen layer, a matte surface, a longer-scale negative required, a lower maximum density, a lower chloride percentage unless you compensate. Less: gloss, brilliance, detail, a shorter-scale negative, and — at zero — Hardwich’s separate chloride figure of 5 to 8 grains per ounce for neat albumen, which is 1.14 to 1.82 per cent, lower than this formula’s, because the percentage is reckoned on a smaller total.

What it interacts with. The chloride, arithmetically and inescapably. Every millilitre of water you add to increase the matteness also dilutes the halide.

Interactions

The dilution changes two things at once, and only one of them is what you wanted. This is the most important practical point on the page. Take Hardwich’s bath and add an equal volume of water to move towards Reilly’s half-matte paper: the albumen falls from 75 per cent to 37.5 per cent, which is the intention, and the chloride falls from 2.28 per cent to 1.14 per cent, which is not. At 1.14 per cent you are below Reilly’s floor of 1.5 per cent and into the territory he describes as producing prints that “lack brilliance and density”. The fix is to keep the chloride at the same percentage of the new total — which for a 1:1 dilution means adding another 12.96 g. That instruction is the course’s reading of Reilly’s statements about chloride content, not a sentence in any of the three sources; it is stated here as a reading, and it is the first thing to test if a diluted bath disappoints.

The chloride sets the silver bath, and the period sources do not obey the modern rule. Reilly’s rule is that the sensitiser must be approximately four times the strength of the salting solution, with the worked example of a 3 per cent salting solution calling for approximately 12 per cent silver. Applied to this formula’s 2.28 per cent, that gives about 9 per cent — and Reilly’s own advice for albumen papers, 10 per cent or above with a chloride of 1.5 per cent or above, points the same way.

Hardwich sensitises on 90 grains to the ounce, which is 20.5 per cent. Towler keeps his bath “at about 70 grains to the ounce”, 16 per cent. Wall, in 1912, gives about 60 grains to the ounce, 13.7 per cent. Every one of them is well above the modern rule, and the trend across the three is downward. Hardwich is aware of the discrepancy and answers it directly: “A Bath prepared by the above formula is stronger than is actually necessary. But it has been found that paper floated on weak solutions is always more or less deficient in vigour.”

⁇How strong should the silver bath for an albumenised paper be?

The disagreement. Reilly (1980) derives the sensitiser strength from the salting strength by a factor of about four and recommends 10 to 12 per cent for albumen papers. Hardwich (1864) uses 20.5 per cent, Towler (1864) about 16 per cent and Wall (1912) about 13.7 per cent — all of them above it, with the period figures roughly double the modern derivation.

What each side is actually claiming. Reilly’s four-times rule is about stoichiometry plus excess: enough silver to convert all the chloride and leave the large excess that Talbot and Vogel showed the print-out mechanism needs. Hardwich’s claim is about result: weak baths give prints deficient in vigour, whatever the arithmetic says. They are not answering the same question, and neither is obviously wrong.

What may explain it. Three things, none of them established. Period baths were used until nearly exhausted and topped up by guesswork, so a nominal 90 grains may often have been working at much less; Hardwich says outright that “the strength of the Bath decreases rapidly by use”. Period floating times were short and the albumen coagulates faster on a stronger bath, so the strong bath may be buying speed of coagulation rather than silver. And a stronger bath produces larger silver chloride grains, which Reilly notes are less likely to be absorbed into the fibres — an argument that applies with most force to the diluted, albumen-poor coatings this page is about.

What the course uses. Reilly’s figures, because they are the ones offered to a modern worker with a modern purpose, and because they use less silver. The period figures are recorded because they are what these formulas were actually printed with, and a reader reproducing Hardwich exactly should know that he is not reproducing Reilly.

Albumen and ammoniacal silver are incompatible, and two sources 116 years apart say so. Hardwich introduces his third formula with the reason built into the first line: “Formula III. Ammonio-Nitrate Paper. — This is always prepared without Albumen, which is dissolved by Ammonio-Nitrate of Silver.” Reilly’s account is the same and adds the history: when albumen paper came into use in the early 1850s the disadvantages of the ammonia-nitrate bath multiplied, since it tended to dissolve the albumen off the paper, and photographers’ attempts to fix this by neutralising with nitric acid amounted only to making ammonium nitrate, which accumulates in a bath anyway.

The binder ranks the colour, and the ranking is a hundred and sixteen years stable. Hardwich: “The reddening action of Gelatine, although greater than that of Starch, is less than that produced by Albumen, and the surface brilliancy is also less.” Reilly’s classification of active and inactive organic substances says the same thing in modern terms, and his note that colloidal silver takes its colour partly from the refractive index of its medium says why.

The rawstock and the binder choose each other. Hardwich records the trade’s opinion: starch-sized paper “offers more mechanical advantages when Albumen is to be used in the salting solution”, whereas gelatine sizing “gives a better surface layer of Chloride of Silver in the case of plain salted paper prepared without Albumen”. He names Papier Rive as taking a high gloss on the albumen because it is hard, smooth and not easily permeable, and the more porous Saxe as taking less. And he gives the limit in the other direction: too heavily sized paper gives considerable gloss but prints that tone and fix with difficulty.

Citric acid put in here is a different decision from citric acid put in the silver bath. Reilly records both and distinguishes them: citric acid in either place improves the pre-processing stability of albumen paper and gives greater sensitivity and maximum density, but the print colour comes out more brownish when the acid is in the silver solution than when it is in the albumen, and in both cases the print turns reddish brown even with prolonged gold toning. If purple is what you want, keep the acid out of both.

Variants

Wall’s 1912 albumenised paper — the same formula by a different route

The ninth edition of Wall’s Dictionary of Photography prints, under “Albumenised Paper”:

Ingredient	As printed	Metric
Albumen	6 ounces	170 mL
Chloride of ammonium or sodium	60 grains	3.89 g
Rectified spirit (ethanol)	96 minims	5.7 mL
Distilled water	14 drachms	49.7 mL

Total stated liquid 226 mL; chloride 1.72 per cent; albumen 75.5 per cent by volume. The conversions are the course’s, using 480 minims and 8 fluid drachms to the imperial fluid ounce.

The agreement with Hardwich is the striking thing. Forty-eight years apart, two authors with no common source of quantities arrive at three parts albumen to one part aqueous diluent — 75.0 per cent against 75.5 — with chlorides of 2.28 and 1.72 per cent, both inside the 1.5 to 2.5 per cent band Reilly would name as normal in 1980. That is not a corroboration of a formula, since the numbers differ; it is something arguably more useful, which is independent evidence that this ratio is where the albumenised paper of the period lived.

Wall differs in three respects worth noting. He allows sodium chloride as an equal alternative, which Reilly also does — “prints with similar color and contrast may be expected with the use of either of these chlorides, or both used in combination in the correct amount” — and which removes the ammonia problem described under The mechanism. He beats for fifteen minutes with a whisk rather than to a froth with quills. And he includes alcohol as a standing ingredient, which brings us to the next variant.

Hardwich’s spirits of wine — the period’s wetting agent

Hardwich’s own improvement on his own formula, given in the paragraph following it:

The Ox-Gall has been thought to make the paper tone more slowly in the Gold Bath, and consequently the Writer now omits it, and adds two drachms of Spirits of Wine (previously diluted with water to prevent coagulation) to each four ounces of Albumen. This removes the greasiness in a great measure, and lessens the chance of streaks.

Two fluid drachms to four fluid ounces is 6.25 per cent of the albumen by volume, or about 27 mL for this formula’s 426 mL. Wall’s is a little over half that rate — 96 minims to 6 ounces, which is 3.3 per cent of the albumen.

Note the parenthesis, because it is the whole chemistry of the addition: the spirit must be diluted with water before it goes in, or it will coagulate the albumen where it lands. Reilly states the general property — albumen is insoluble in alcohol and alcohol coagulates it — and the same property is what makes the double-coating procedure possible, where a 70 per cent isopropanol or a two-to-one alcohol-and-water bath is used deliberately to set the first layer before the second is floated.

The modern equivalent is a surfactant, and Reilly gives a rate: 4 mL of a wetting agent per litre of albumen, stirred in gently immediately before a batch is coated, avoiding bubbles. His two stated purposes are exactly Hardwich’s two — it controls bubbles, and it improves the runoff characteristics so the coating dries more evenly. The coating-plant jargon for such additives, he notes, is “spreaders”.

Hübl’s matte albumen — the other way to reach a matte surface

The albumenised salting solution reaches matte by taking albumen out. Baron Arthur von Hübl reached it by putting starch in, and the resulting paper — matte albumen — was a commercial article from 1898 until the late 1920s under names like Albumat, Albumon, Alboidin and Albumatpapier. The Getty’s atlas credits the variant to Hübl in 1895 and describes the method as mixing albumen stock with a salted solution of starch.

Two printings of the formula exist in the course’s sources and they do not agree:

	Wall, 1912	Reilly, 1980
Albumen	1 ounce (28.4 mL)	100 mL
Arrowroot solution	1 ounce (28.4 mL)	100 mL
Salt	30 grains (1.94 g)	4 g in the 100 mL of arrowroot solution
Chloride of the whole	3.4 per cent	2.0 per cent
Starch strength	“as described”	not restated in the procedure; Hübl’s original given elsewhere as 2 per cent

Both agree on the ratio — equal volumes of albumen and salted arrowroot solution — and on the fresh, unaged albumen, whipped to a froth and left twenty-four hours to clear. Wall adds that the two are mixed when the arrowroot has cooled to 90 °F and then brushed on; Reilly says the albumen is used at most 24 hours after settling, kept refrigerated, and filtered through muslin immediately before use.

They disagree on the chloride by a factor of about 1.7, and there is no third printing in the corpus to break the tie. The course records the disagreement rather than averaging it, which is the rule this formulary works by; corroboratedBy has no opposite and there is no field for a disagreement between printings, which is open item 5 of docs/FORMULARY-SCHEMA.md. Anyone making this paper should note that Reilly’s figure sits inside his own recommended band and Wall’s sits well above it, and should probably start at Reilly’s.

A note attaches to this variant that does not attach to the main formula: the binder it substitutes had no page in this course’s chemical encyclopaedia when this was written, and the register held neither arrowroot nor starch. Arrowroot now has one, written because the arrowroot salting solution needed it, and it carries the material’s granule chemistry and its one quantified control — HSE’s workplace exposure limit for starch dust, 10 mg/m³ inhalable over eight hours — alongside what Reilly’s chapter on starches and Towler’s on the amylaceous substances say about it. Starch deliberately remains unregistered as an alias, because tapioca, sago, rice and wheat starches are different substances that the sources name separately: a formula asking for “starch” should fail until somebody says which one.

The chlorideless albumen papers, for the record

The oldest variant of all is the one with no chloride in it. Reilly’s history begins with the letter signed “H.L.” in The Athenaeum of 11 May 1839 — “Wash the paper with a mixture of equal parts of the WHITE OF EGG and water, afterwards with the solution of nitrate of silver” — which is an albumenised salting solution with the salt left out, depending entirely on the light sensitivity of silver albumenate. Reilly notes that the idea was rediscovered in 1865 by Schultner and proposed as a practical method in 1866 by Schnauss, that such papers required a weaker than usual silver solution and a fairly contrasty negative, and that the results have “surprising vigor and tone well in gold toners” — but that chlorides give a far more versatile and satisfactory paper, and that in nineteenth-century practice they were always used.

The course does not publish this as a formula, because “equal parts of white of egg and water” with no quantity of anything else is a proportion rather than a formulation, and because the one thing it demonstrates — that silver albumenate prints out on its own — is better learned as the experiment suggested at the end of this page.

🕰Blanquart-Evrard's original, and what it tells you about the chloride

Reilly gives Blanquart-Evrard’s own 1850 recipe as white of egg beaten to a froth with 25 per cent by weight of a saturated salt solution, settled overnight, floated for one minute and hung to dry. It is not entered as a formula here because “a saturated salt solution” is a quantity only if you also specify a temperature, and he does not. But the shape is worth seeing: the inventor’s version is undiluted albumen carrying a strong brine, which puts a great deal of chloride into a small volume, and the whole subsequent history of the process is a retreat from that towards less chloride and, where a matte surface was wanted, more water.

Safety

Classified Level A, the standard home darkroom footing, and it is one of the few formulas in this part of the course that genuinely deserves it. There is no silver in this bath, no acid stronger than a food acid, nothing volatile unless you add the spirit, and nothing that is heated.

Ammonium chloride carries GHS07 with the signal word Warning and the statements H302, harmful if swallowed and H319, causes serious eye irritation, aggregated on PubChem from a very nearly unanimous set of notifications to the ECHA inventory. In practice that means gloves and eye protection while you weigh and dissolve it, and no dust. It is the one substance in the formula with a workplace exposure limit, and the limit is for the fume rather than for a solution; nothing here generates fume. The ammonium chloride page carries the detail.

Albumen carries no GHS classification, and that is an absence of information rather than a finding of safety. The albumen page states the position carefully: there is no PubChem compound record, no ILO-WHO safety card, and no entry in the HSE’s exposure limit list, which itself says that absence from the list does not indicate that a substance is without risk. What is certain is that this is raw hen egg white, so two ordinary food risks apply and neither is photographic. It is a major food allergen; if anyone in the household has an egg allergy, this is not a formula to make in the kitchen. And raw egg is a microbiological material: work clean, refrigerate, keep it covered, label it, and keep it away from food and from food utensils.

The dedicated-utensil rule applies in both directions here, which is unusual. Ordinarily a darkroom keeps its utensils away from the kitchen because of the chemicals. Here you also want to keep the kitchen’s utensils away from this, because a bowl that has held fat will spoil a coating and a bowl that has held raw egg should not go back into food service without a proper wash.

Do not heat it. Not a health hazard, a material one: above 65 °C the batch is scrap.

The hazard arrives at the next bath, not this one. Everything about the risk profile of this process changes when the sheet meets the sensitiser. Silver nitrate causes permanent, irreversible eye damage, and Reilly is emphatic about eye protection because sheets dripping silver solution are a particular hazard to the eyes; prolonged skin absorption leads to argyria. That is a Level B operation and it is described on its own page. Salting is not.

If you add the spirit, ethanol is flammable, and the working rule is the ordinary one — no flames, no hotplates, a stoppered bottle, and dilute it with water before it goes into the albumen, which Hardwich requires for a chemical reason anyway.

Storage

The made-up bath: cold, covered, dated, and used within a few weeks — but read the next paragraph before you throw a batch away. No source in the corpus publishes a keeping time for an albumenised salting solution. Reilly’s nearest statement, for his undiluted albumen, is that the solution remains useful for several weeks after its one-week ageing period, and that “the smell of the albumen and its color, sedimentation, etc., will reveal when it has decomposed too badly to use.” That is a judgement rather than a rule, and it is the only kind of answer this material admits of.

The complication is that decomposition is partly the point. This is the one place where albumen behaves unlike any other reagent in the formulary. Hardwich describes deliberate ageing as a technique:

makers of albuminized paper have found it advantageous to expose the Albumen in an open vessel to the air, until a considerable amount of evaporation has taken place … the fluid often becomes gradually more limpid, acquiring a rather offensive odour, and an acid reaction to Litmus paper. Albumen so prepared runs upon the paper very easily, and does not dissolve the size.

Reilly gives the modern explanation — the chemical forces binding the protein molecules weaken as the pH falls, so an acidified albumen is a differently behaved albumen — and records how far the trade took it: the Dresden factories fermented their albumen at elevated temperatures for several days, and their paper could be identified by the smell.

But note what Reilly also says, and it is the reason this page’s bath is not an ageing one: partially decomposed acidic albumen is used for glossy papers, and fresh alkaline albumen is used for matte papers, where it is always mixed with starch or another substance. An albumenised bath that is being diluted towards matte is therefore not obviously a candidate for long ageing, and Hübl’s matte albumen explicitly uses albumen no more than 24 hours old. The course has found no source that states the optimum age of a diluted albumen bath, and does not invent one.

Hardwich also states the limit of the technique, and it is a real one. Decomposition eventually produces sulphuretted hydrogen from the albumen’s own sulfur, and stale albumenised paper is destructive downstream: “The Nitrate Bath is soon rendered turbid by the use of stale albuminized paper, and the sensitiveness to light is injured, so that the half-tones of the picture do not appear until after a prolonged exposure.” A bath you are unsure about will cost you a silver bath as well as a sheet of paper.

The coated paper: dry, flat, and not rolled. Hardwich says albumenised paper keeps a long time in a dry place, and gives the handling that goes with it — cut with a bone spatula, do not touch the coated face, press flat between boards or in a printing frame, avoid rolling. Reilly adds that if sheets must be rolled they should be rolled albumen side out to minimise cracking of the layer, and that they should not be excessively dry when they are sensitised, so a night in a damp place before silvering is worth the trouble.

The sensitised sheet: same day. Reilly’s instruction for albumen and salted papers generally is to sensitize, print and process all on the same day. The whole nineteenth-century ready-sensitised paper trade after 1872 existed to get round that, using citric acid, and the details belong on the sensitiser page.

Incompatibilities

Ammoniacal silver. The clearest incompatibility in the whole family, stated by two sources 116 years apart. Hardwich: ammonio-nitrate paper “is always prepared without Albumen, which is dissolved by Ammonio-Nitrate of Silver.” Reilly: the ammonia-nitrate bath “tended to dissolve the albumen off the paper”. If you are working with an ammoniacal sensitiser, you are working with a plain salted paper, not this one.

Heat above 65 °C. Coagulates the albumen irreversibly, at any stage. There is no hot-water step in this formula, and if you find yourself wanting one you have the wrong formula.

Undiluted alcohol. Coagulates it on contact. Hardwich’s spirits of wine must be diluted with water first, and he says so in the same breath as he recommends it. The property is exploited deliberately in the double-coated procedure, where a 70 per cent isopropanol bath — or Wall’s two volumes of alcohol to one of water — is used to set the first coating so that the second does not dissolve it.

Metal salts generally. Reilly lists contact with salts of metals among the things that coagulate albumen. In practice this means clean glass, porcelain or new plastic vessels; it also means that a tray with a history is a liability, and Reilly warns separately that plastic tends to hold on to chemicals, so a tray that has held developer or fixer will stain.

Barium chloride, if you are toning with alkaline gold. Hardwich’s own warning, and a good example of an incompatibility that is invisible until three steps later: “Chloride of Barium, sometimes used in salting paper instead of Chloride of Ammonium, is contra-indicated when the Alkaline Gold-toning process is adopted, since the Carbonate of Soda would throw down Carbonate of Baryta in the paper.” Barium chloride is also acutely toxic, has no page in this encyclopaedia, and has nothing to recommend it; it is recorded because Hardwich records it.

Fat, in any quantity. Yolk, greasy fingers, an unwashed bowl. It is the reason every source insists on separating the eggs one at a time and taking out the germ, and the reason Hardwich reaches for ox-gall and then alcohol to cut the “greasiness”.

Alkali in the paper’s own sizing, in the other direction. The free ammonia this bath generates attacks the size, which is why the float is a maximum. A lightly sized or badly sized rawstock will suffer faster than a well-sized one.

Waste

The bath itself is food waste with a little ammonium chloride in it. Diluted albumen in rinse water is oxygen-demanding organic load and nothing more exotic; the chloride at 2.28 per cent is a dilute solution of a salt used by the tonne in food and metalwork. Nothing in this formula is a heavy metal, a complexing agent or a persistent organic. A spent salting bath is, in ordinary domestic terms, the least troublesome liquid in this part of the course.

It is still not nothing, and the sensible order is: don’t make more than you need, then dispose of what is left as the organic waste it is. Check your local rules, which govern; some jurisdictions treat any bulk protein liquid as trade effluent, and the smell of a fermented batch is a nuisance issue in its own right.

The moment a sheet has been silvered, none of the above applies. The rinse from a sensitised sheet carries silver nitrate and belongs in the silver stream, and the first wash of a printed sheet carries a great deal of it. Bottle it, label it with what it is and the date, and keep it separate. The alkaline fixing bath page and the washing and permanence material carry the silver-recovery detail.

Eggshells and yolks. Fifteen yolks are a real by-product and the nineteenth century took them seriously: yolks were preserved with salt and sold to bakeries and to tanners, who used them to finish kid leathers, and Reilly records that the money realised from them was a large factor in the profit-and-loss statements of the giant Dresden albumen-paper producers. For those working below industrial scale he reproduces the British Journal of Photography’s cheesecake recipe of 2 September 1861. The modern answer is the same one — use them — and it is not a joke: it is the only waste stream in this formulary that improves with a lemon.

Troubleshooting

Streaky lines that bronze under light after sensitising. Hardwich names this as the principal difficulty of albumenising and gives both causes. Either the albumen was not thoroughly beaten, so flakes of membrane remain in the liquid — the fix is at the mixing stage and there is no rescue for the sheet — or the sheet was lowered onto the bath with a pause in the movement: “lower the paper on the liquid by one steady movement, since if a pause be made, a line will be formed.” His third remedy, for a paper the albumen does not wet readily, is the diluted spirits of wine under Variants.

Circular white spots on the print. Air bubbles under the sheet during floating, which left that patch uncoated. Wall’s diagnostic catches them in the first seconds — bubbles show as puckers in the floating sheet — and are wetted down with a camel’s-hair brush; Hardwich’s remedy for a sheet already lifted is to replace it on the bath for the same length of time again. The same fault at the sensitising stage gives the same spots for a different reason, and Reilly is explicit that bubbles prevent sensitisation where they occur.

A thin, dull coating at the top of the sheet and a heavy marked one at the bottom. Runoff, and gravity. Towler describes it exactly and gives the fix: hang the sheet by its long edge so the liquid has the shortest distance to run. Blot the accumulating bead repeatedly while it dries — Hardwich and Reilly both do — because a dried bead forms a rind that makes the sheet unmanageable. If the effect is severe, dry weather and a too-dry sheet are usually contributing; Towler notes that albumen does not attach itself easily to dry paper in dry conditions.

No gloss, and a print that looks sunken even though you used plenty of albumen. You floated too long. Both period sources say so and give the same mechanism: the alkaline bath dissolves the paper’s size, the albumen sinks into the body of the sheet instead of resting on it, and the surface goes with it. Time from the moment the sheet lies flat and every bubble is broken, and treat ninety seconds as a ceiling.

Grey, flat prints with poor maximum density. Work through it in this order. Is the chloride below about 1.5 per cent — which happens automatically the moment you dilute the bath without adding more chloride? Is the silver bath strong enough for that chloride, at four times it or better? Is the rawstock too porous, so that the image is in the fibres whatever the binder is doing? And is there enough excess silver nitrate present for Vogel’s recycling to run at all? Reilly’s short answer for this whole class of failure is that lowering the chloride content “tends to produce prints that lack brilliance and density”.

The bath smells strongly of ammonia when you make it. That is expected and it is Hardwich’s observation, not a fault: an ammonium salt meeting an alkaline protein liberates ammonia. It is, however, a reminder that the sheet must not sit on the bath, since the same ammonia attacks the size.

A sour or sulfurous smell, a change of colour, heavy sedimentation. The batch has gone past the useful degree of decomposition. Reilly’s test is exactly this — smell, colour and sedimentation — and his answer is to discard. Hardwich’s warning about the consequences of persisting is the one to weigh: a stale batch will turbidify your silver bath and reduce the paper’s sensitivity, so you lose more than the eggs.

The silver bath goes brown and then nearly black. Expected, and not immediately a fault. Hardwich says the solution “becomes after a time discoloured by the Albumen, but may be used for sensitizing until it is nearly black”, and gives two clarifications: shake twenty ounces of bath with a quarter of an ounce of finely pulverised kaolin and filter through paper, or shake it with freshly precipitated silver chloride in the curdy state, which has an affinity for the brown sub-albuminate of silver causing the colour. His caution on the first is worth repeating: kaolin containing carbonate of lime must be washed in dilute vinegar first, “or the Bath will become alkaline, and dissolve off the Albumen.”

Marbled stains on the sensitised sheet. A greasy scum of dissolved albumen floating on the silver bath. Hardwich’s remedy is mechanical and effective — draw a folded strip of blotting-paper, cut to the exact breadth of the dish, lightly along the surface before each sheet.

A print that will not tone purple however long it sits in the gold. Citric acid, most likely, either in the albumen or in the silver bath. Reilly records that it drives albumen prints reddish brown even under prolonged gold toning, and that this was one of the two reasons professionals kept sensitising their own paper after ready-sensitised albumen appeared in 1872.

Experiments

The dilution series, which is the experiment this page exists for. Make one batch of Hardwich’s formula and split it four ways: neat, 1:1 with water, 1:3, and 1:6, adding chloride to each dilution to hold the concentration at 2.28 per cent of the new volume. Coat the same rawstock the same way, silver all four on the same bath, print all four from the same negative to the same visual endpoint, tone and fix them together. You are testing Reilly’s ladder directly — glossy, half-matte, matte, almost-a-salt-print — and you will see the exposure times lengthen as you go down it. Record them, because the speed change is the half of the prediction people forget.

The same series with the chloride left alone. The control that makes the first experiment mean something. Dilute the same four ways and do not top up the chloride, so the halide falls with the albumen. If the course’s reading under Interactions is right, the 1:6 sheet should be visibly poorer than its topped-up twin — thinner, less dense, lacking what Reilly calls brilliance — and the difference between the two sets is the whole of the argument.

Silver albumenate on its own, after “H.L.”. Coat a sheet with albumen and water, equal parts, with no chloride at all, dry it, and sensitise it on a weak silver bath — Reilly says a chlorideless albumen paper wants a weaker solution than usual and a fairly contrasty negative. Print it beside a chlorided sheet from the same negative. What you are looking at is the contribution of silver albumenate alone, which is the claim in Reilly’s classification that this page rests on and which almost nobody ever tests. Expect it to be slow and to have, in his words, “surprising vigor”.

Ammonium chloride against sodium chloride, at the same molar concentration. Wall and Reilly both say the two are interchangeable for colour and contrast; Hardwich’s ammonia mechanism says they cannot be quite interchangeable in handling, because only one of them liberates ammonia into an alkaline albumen. Make the two baths at equal chloride-ion concentration — note that 12.96 g of ammonium chloride and 14.16 g of sodium chloride carry the same amount of chloride, since the molar masses are 53.49 and 58.44 — and compare the smell, the behaviour of the size at 90 and at 180 seconds of floating, and the finished prints. This is a test of whether a real chemical difference has a visible photographic consequence, which is the most interesting kind of question this course asks.

Fresh albumen against aged albumen, at one dilution. Split a batch, use half within 24 hours and refrigerate the other half for a week, then coat and print both. Reilly’s prediction is specific and falsifiable: the aged half should coat more evenly, give more gloss, and yellow less after sensitising. Wall’s period claim is that “many professional albumenisers prefer stale eggs as giving a more even and lustrous coating”, so the two sources agree about the direction. Hübl’s use of fresh albumen for a matte paper suggests the advantage may vanish or reverse at high dilution, which is the version of the experiment worth doing twice.

Hardwich’s silver-strength strips, unchanged from 1864. Prepare strips of albumenised paper and sensitise them on silver baths of 5, 10, 20, 40, 80 and 100 grains to the ounce — 1.1, 2.3, 4.6, 9.1, 18.2 and 22.8 per cent — and print them all under one negative. His prediction is that vigour rises with strength up to a point and then stops rising. Reilly’s four-times rule predicts where that point should be for a 2.28 per cent salting solution. This is the cheapest available test of the disagreement set out under Interactions, and it settles it for your paper, which is the only place it can be settled.

Measure the scale length with a 21-step tablet, and place your paper on Reilly’s ladder. His method: sensitise a test sheet, print it under the tablet long enough that the step-1 patch matches the density of the margin outside the tablet, process normally, and count the distinguishable steps. Do it for this bath at two dilutions, for a plain salted sheet and for an arrowroot sheet. If Reilly and Hübl are right, the order of required negative density range should come out plain salted, arrowroot, albumenised, glossy albumen — and you will have the four numbers that tell you which negative to print on which paper for the rest of your working life.

Weigh what a sheet actually takes, and check Hardwich’s capacity figure. Weigh the tray and bath before and after coating a quarter-sheet of 11 × 9 inches. He says 1.5 fluid drachms, about 5.3 mL; he also says a plain salting solution gives up only 1 drachm to the same sheet. Repeat it for the plain bath and for a dilution of this one, and you will have measured the viscosity effect he describes and found out whether it scales with the albumen content. Almost no nineteenth-century number in this formulary is this easy to check.