Albumen sensitising solution

!Which of the albumen entries is this?

Albumen printing has five entries in this formulary and they are five different operations. The albumen solution and the albumenised salting solution are what you coat the paper with — undiluted egg white in the first, three parts egg white to one of water in the second, both carrying the chloride. The double-coated procedure is what you do if you want two coats. This page is the bath that comes next, and it is where the paper stops being a piece of stationery and becomes a photographic material. The generic silver bath for plain salted papers is the salted paper sensitiser; read that one for arrowroot, plain salted and matte papers, and this one for anything with albumen in it.

Ingredient	Quantity	Form the source specifies
Silver nitrate	5.83 g	90 grains. Reilly states that a technical or ACS grade is sufficiently pure for albumen and salted paper printing and that the highly purified grades needed for emulsion making are unnecessary; Hardwich adds that it is not worth examining the crystals for free nitric acid, because unless the sample is very impure the retarding effect is inappreciable and especially so on albumenised paper, which is usually slightly alkaline
Water	28.41 mL, added	One imperial fluid ounce, distilled. Hardwich specifies distilled water and no temperature. Reilly gives the reason a modern reader needs - a bath made with tap water throws a cloudy precipitate of silver chloride and silver carbonate as the silver meets the chloride and the hard-water ions already in the supply, which robs the bath of strength and makes sensitising difficult. The same reaction, run deliberately, is what clouds the first wash water after exposure

Hardwich prints it as two lines, under a heading that is also an instruction: “To render the paper sensitive. — This operation must be conducted by the light of a candle, or by yellow light. Take of — Nitrate of Silver, 90 grains. Distilled Water, 1 ounce.” That is 20.5 per cent w/v, or 205 g in a litre, and it is roughly twice the strength a modern worker would use. Both halves of that sentence matter, and the page below is largely about why.

Purpose

To do three jobs in one three-minute float. The salted paper sensitiser does two of them. The third is albumen’s, and it is the reason this entry exists.

One: make the light-sensitive substance. Silver chloride is insoluble in water and in most solvents, so it cannot be dissolved and coated. It has to be built inside the sheet from two solutions applied separately — a soluble chloride first, carried in the albumen itself, then this bath of soluble silver. Where they meet, silver chloride precipitates on the spot.

Two: leave far more silver behind than job one consumes. Reilly states the requirement plainly: the bath must contain enough silver nitrate not only to react with all the chloride present, but to leave a considerable excess of silver nitrate in the light-sensitive layer. That surplus is the second reagent of the printing-out process. Without it the sheet darkens to a grey that stops early and fixes out to almost nothing.

Three: coagulate the binder. This is the albumen-specific job. Albumen is water-soluble; a water-soluble binder on a sheet about to spend an hour in wet processing is no binder at all. Silver nitrate is one of the metal salts that coagulate egg protein, and the coagulation happens in this tray, at this concentration, in these three minutes. Hardwich makes the point by rejecting a rival method for it: some recommend pressing albumenised paper with a heated iron to set the layer, “but this precaution is unnecessary, since the coagulation is perfectly effected by the Nitrate of Silver used in sensitizing.”

Job three is coupled to jobs one and two through a single variable — the strength of the bath — and it does not want the same thing they do. Everything difficult about sensitising albumen paper comes out of that conflict.

Recommended uses

Albumenised paper of any dilution, floated three minutes for thin stock and four or five for thick. This is Hardwich’s own pairing: his albumenised salting solution is 200 grains of ammonium chloride in 20 fluid ounces of albumen and water, and this bath is what he sensitises it on.

Glossy albumen paper, single or double coated. The albumen solution and the double-coated procedure describe the sheet. Reilly’s float of two and a half to three minutes on a 10 to 12 per cent bath is stated to be adequate even for doubly albumenised paper, which is the harder case.

Plain salted paper, on the same bath. Hardwich’s Formula II, the plain paper, is “sensitized on the same solution of Nitrate of Silver as the albuminized paper”. If you are working through his chapter you make one bath and float both. A modern worker following Reilly would use the salted paper sensitiser at 12 per cent for the plain paper and the same bath again for the albumen.

Floating, not brushing. Reilly recommends brush sensitising as an alternative for matte salted papers but not for glossy albumen. A glossy albumen surface shows every stroke, and the coagulation begins the moment the brush touches it, which locks the unevenness in.

Under yellow light or a red safelight, and nowhere near the salting bench. Hardwich’s heading carries the instruction — candle or yellow light. Bostick and Sullivan’s version of the same rule is more useful to a reader with electric light: a red compact fluorescent or red LED works, the light can be very bright without harm, and every operation from this tray onwards is a safelight operation. Note the exact boundary they draw, because it is easy to get wrong in both directions: the silver nitrate solution itself is not light sensitive, but it becomes so the moment it meets an organic substance, which on this bench means the moment it touches the paper.

When another formula is preferable

• If you are printing today and buying silver today, Reilly’s 10 to 12 per cent. It is half the strength, it costs half as much per litre, and it is what the course recommends for modern work. See the salted paper sensitiser, which is that formula.
• If the sensitised sheet has to keep more than a day, the citric acid version, under Variants. Five per cent citric acid in this bath is Reilly’s maximum preservative effect; one per cent extends the life noticeably. It changes the print colour and makes purple tones difficult, so it is a trade and not an improvement.
• If the binder is starch — arrowroot, resin-arrowroot, matte albumen — the citric acid version again, and for a different reason. Starch is not an “active” organic substance in Reilly’s sense: it keeps the image on the surface but does not help reduce the silver chloride, and a starch paper needs an organic acid supplying what albumen would have supplied. See the arrowroot salting solution.
• If what you want is a developed image with a fixed speed and an enlarger, the silver chloride contact emulsion. Every rule on this page inverts: a develop-out material wants a halide excess and no free silver at all.
• If you want a hand-coated print without silver, the classic cyanotype sensitiser is one coating instead of two, needs no fixer, costs a fraction as much and carries none of the hazards below.
• Never ammonio-nitrate, on an albumenised sheet. It is more sensitive and gives a better colour, and it dissolves the albumen off the paper. Hardwich’s Formula III is prefaced with the reason: ammonio-nitrate paper “is always prepared without Albumen, which is dissolved by Ammonio-Nitrate of Silver.”

Mixing

Distilled or de-ionised water, and this is the one requirement every source states as a requirement. Hardwich writes “Distilled Water” into the formula. Reilly gives the mechanism: made with tap water, the silver nitrate reacts with the chloride and the hard-water ions already in the supply and throws a cloudy white precipitate of silver chloride and silver carbonate, which robs the bath of strength and makes sensitising difficult. You do not get a weak bath of known strength; you get an unknown bath and a tray of sediment.

Water first, then the salt. With one solute there is no order to argue about, and no source states one. Weigh 5.83 g of silver nitrate per fluid ounce of the finished quantity — in practice, work in litres and use the strengths under Dilutions — and stir until the last crystal has gone. Silver nitrate is freely soluble and needs no warming; no source in the corpus gives a mixing temperature for this bath, and none is needed.

Make enough to float on. Hardwich’s line is per ounce; his instruction is “Prepare a sufficient quantity of this solution, and pour it out into a porcelain dish.” A tray deep enough to float a sheet without the corners touching bottom takes several hundred millilitres, and at 205 g/L that is a serious quantity of silver sitting in an open dish. This is the single strongest practical argument for Reilly’s weaker bath, and it is a better argument than the chemistry.

Glass, porcelain or new plastic. Never stainless steel. Hardwich uses a porcelain dish; Reilly says the tray and every container should ideally be glass, that stainless steel is not recommended, and that a plastic tray must be new because plastic holds on to chemicals and a tray contaminated with developer or fixer will stain. The same rule reaches the small tools: Hardwich lifts sheets with bone forceps or tweezers tipped with sealing-wax, and Reilly warns that a brush with a metal ferrule will react with the silver nitrate and stain the print.

Grade. Reilly states that a technical or ACS grade is sufficiently pure and that the highly purified grades required for emulsion making are unnecessary. Hardwich makes the same point from the other end: it is not worth testing the crystals for free nitric acid, because unless the sample is very impure the retarding effect will be inappreciable, “and especially so in the case of albuminized paper, which possesses usually a slightly alkaline reaction”. The sheet neutralises the trace acid for you.

Kaolin goes in the storage bottle, not in the tray. Both Reilly and Bostick and Sullivan store the bath over kaolin — about 15 g per litre, or half a teaspoon per bottle — and shake it up after each session so that it can settle overnight and carry the dissolved organic matter down with it. Hardwich’s proportion is a quarter of an ounce of finely pulverised kaolin to twenty ounces of bath, shaken and then filtered through paper. The course has no chemical encyclopaedia page for kaolin; Hardwich calls it “pure white china clay”, it takes no part in the photographic chemistry, and the one caution about it is Hardwich’s and is a real one, given under Incompatibilities.

Behaviour

The float time and the bath strength are coupled, and the coupling runs the wrong way. This is the fact that separates an albumen sensitiser from every other silver bath in the formulary. Reilly states it directly: if the silver bath is very strong, the coagulating effect of silver nitrate on the albumen layer is correspondingly strong, and the solution therefore permeates the albumen more slowly, so a strong bath needs a longer float. And if the bath is relatively weak, prolonged floating is harmful for the opposite reason — the albumen, which is water-soluble until it is coagulated, may begin to dissolve off the sheet before it has been set.

So there is a floor and a ceiling on strength and they are set by different failures. Too weak, and the binder washes into the tray. Too strong, and the silver takes minutes to get through a layer it has already sealed. The published float times track this: Hardwich’s three minutes at 20.5 per cent, Wall’s three or four at 13.7, Reilly’s two and a half to three at 10 to 12, and Bostick and Sullivan’s six at 15 per cent on a doubly coated and heat-hardened sheet — that last one an outlier that says something about the sheet rather than about the bath.

A fresh bath behaves worse than an old one, and the reason is a by-product. The double replacement leaves ammonium or sodium nitrate in the solution, and those salts coagulate albumen too. A bath that has silvered fifty sheets has accumulated them; a bath mixed this morning has none. Reilly records that the dissolving-off problem is “especially noticeable with freshly made-up sensitizing baths” for exactly this reason, and that some old manuals advised adding ammonium or sodium nitrate deliberately. He then bounds the advice: it applied to silver-poor baths of 5 to 8 per cent and is not applicable at 10 per cent. On Hardwich’s 20.5 per cent bath it is irrelevant.

The bath goes brown, then black, and this is not spoilage. Some albumen and some size dissolve off every sheet. The organic matter reacts with a little of the silver, which is spontaneously reduced to the metallic state, and finely divided metallic silver is what colours the solution. Hardwich names the intermediate — a brown “Sub-Albuminate of Silver”, partially soluble in silver nitrate solution — and says the bath “may be used for sensitizing until it is nearly black”. Reilly is more conservative: slightly discoloured is fine, nearly black is not. Both clear it with kaolin. Bostick and Sullivan report the colour as red rather than brown and attribute it to salt carried out of the albumen; the observation is theirs and the difference is not resolved in the corpus.

A greasy scum forms on the surface and it will mark your prints. Hardwich again, and it is the kind of detail that only appears in a book written by somebody who did it: after the bath has been in use a short time, the albumen dissolved out of the papers forms a greasy scum on the liquid, “which if allowed to remain produces marbled stains upon the sensitive paper”. His remedy is a folded strip of blotting-paper cut to the exact breadth of the dish and drawn lightly along the surface before each sheet. Reilly’s equivalent is to filter the bath whenever a surface scum appears, and he notes the same defect from the other end — a metallic marbled sheen on exposed prints.

The bath weakens fast and in two ways at once. Silver leaves it by reacting with the chloride, and it also leaves mechanically, soaked into every sheet that comes out. Hardwich: “The strength of the Bath decreases rapidly by use, and hence an addition of Nitrate of Silver must occasionally be made.” Towler: the strength “has to be maintained at its original point by the addition of fresh silver every time it is used, for the bath soon becomes impoverished by the floating of paper”. Reilly quantifies the practical consequence — a weak bath needs more constant attention than a strong one because it reaches a dysfunctional point sooner, which is why he tells you to start at 12 per cent rather than 10 even though 10 is cheaper.

Measuring the strength: two methods, a century apart. Hardwich’s is a “Silver Meter”, a hydrometer in a small cylinder, graduated directly in grains per ounce. He is honest about it — “It is not quite accurate, but near enough for practical purposes” — and gives two cautions worth carrying: the cheap instruments need their scales verified against a freshly made 90-grain solution, and the readings stop being trustworthy if the bath contains alcohol or ether, because those change the density without changing the silver. (A printing bath will not normally contain either. A negative bath will, which is why he raises it.) Reilly’s method for serious work is a Volhard titration — sodium thiocyanate run into a measured, nitric-acidified sample against a ferric ammonium sulfate indicator until the colour change will not stir out. He records that the same method was in use for albumen sensitising baths as early as 1875.

Replenishing. Reilly’s rule of thumb is arithmetic you can do at the sink: replace the lost volume with a solution twice as strong as the bath. A litre of 12 per cent that has fallen to 850 mL takes 150 mL of 24 per cent. He calls it unsuitable for continuous use because the errors accumulate. Hardwich’s remedy is for a different and more specific failure, and it is worth keeping separate: when an old printing bath has gone alkaline from the reaction of the albumen, the coagulation becomes imperfect and a white precipitate falls away into the bath — add to each ounce ten grains of silver nitrate and half a drop of glacial acetic acid. That is a strength correction and a pH correction at once, and it treats the symptom he has just diagnosed.

🧮How much of the silver actually becomes silver chloride?

Hardwich’s own stoichiometry: “three parts by weight of Nitrate of Silver will precipitate nearly one part by weight of Salt, and a slight excess of the Nitrate of Silver will remain.” The modern figures confirm him — silver nitrate is 169.87 g/mol and sodium chloride 58.44, so the ratio is 2.91. For the ammonium chloride his own formula actually uses, 53.49 g/mol, it is 3.18. He states the chlorine equivalence too, and it also checks: 100 grains of ammonium chloride carry as much chlorine as 109 grains of sodium chloride.

Now put the two of his solutions side by side. His salting bath is 2.28 per cent w/v ammonium chloride; converting it needs 2.28 × 3.18 = 7.2 per cent silver nitrate. His sensitising bath is 20.5 per cent. If a sheet takes up similar volumes of the two baths, then about a third of the silver it carries is spent making silver chloride and about two thirds stays in the layer as free silver nitrate.

Two warnings about that number. The equal-uptake assumption is the course’s, not Hardwich’s: he measured the uptake of the salting bath — a quarter-sheet of 11 × 9 inches removes one and a half fluid drachms — and measured nothing for the silver bath, and Reilly’s coagulation argument says the two cannot be equal. And the ratio is not the point. The point is that the free silver is the larger share, which is what Talbot found empirically in the 1830s and what Vogel later explained.

Image characteristics

Vigour is the word the period used, and it is what a strong bath buys. Hardwich’s justification for a formula he admits is stronger than necessary is one sentence: “it has been found that paper floated on weak solutions is always more or less deficient in vigour.” What he means is set out in his own graded experiment, which is the most useful passage on this page and is worth reproducing as a table.

Bath	Per cent w/v	What Hardwich reports
20 grains to the ounce	4.6	Darkens, but never reaches the bronzed stage; very pale in the hypo; cold and slaty when toned; pale patches from uneven capillarity
40 grains	9.1	Deeper shadows, warmer colour; scarcely enough for foreign paper salted at 10 grains to the ounce, sufficient for English paper of the same salting
60 grains	13.7	Good
80 grains	18.2	Good, and differing very little from 60
100 grains	22.8	Good; the best of all if the float is cut to a single minute

Three things follow from that table. There is a threshold rather than a gradient — somewhere between 40 and 60 grains the print stops being deficient and starts being good, and above it the differences are small. The threshold moves with the paper: the same salting on a harder-sized English stock needs less silver than on a softer Continental one, because a hard size keeps the chemicals on the surface. And time trades against strength: the weak bath improves markedly if the sheet floats ten or fifteen minutes instead of two or three, “for the Albumen is allowed time to draw to itself more of the Nitrate of Silver”, and the strongest bath is only the best when the float is cut to a minute.

Bronzing is the visible end of the scale. Highly salted and sensitised papers darken rapidly and pass completely into the bronze stage — the greenish metallic sheen of a heavily printed-out shadow. Papers with less chloride darken more slowly and never bronze with the same intensity. Hardwich uses this as a printing control: double the salt for a feeble negative in dull weather, keep the chloride low for an intense negative in direct sun so that the deep shadows are not over-bronzed and the feeble rays have time to come forward.

Colour tracks the sensitising strength as well as the salting. A print on weakly sensitised paper comes out “unusually red after fixing, and of a brown or mulberry colour when toned”; strongly salted and sensitised paper approaches black. Reilly’s mechanism explains why a print changes colour twice after it leaves this tray: print-out silver is colloidal, its colour depends on the refractive index of the medium around it, and both fixing (which dissolves the remaining silver chloride) and drying (which changes the spacing of the particles) change that medium.

Grain size, and why a strong bath keeps the image on top. Reilly states that the strength of the sensitising solution regulates the size of the precipitated silver chloride grains, and that stronger solutions produce larger grains which are less likely to be absorbed down into the paper fibres. On a glossy albumen sheet this matters less than elsewhere, because the coagulated albumen is already holding everything on the surface; on a matte or heavily diluted albumen paper it matters a great deal, and it is the reason Reilly pairs starch binders with strong baths and short floats.

Speed and exposure. Reilly’s ranking is unambiguous: the speed of albumen and salted papers is exceedingly slow, matte salted papers such as arrowroot are the fastest, plain salted papers next, “and the slowest of the lot is albumen paper”. His average exposure for albumen in direct sunlight is 5 to 10 minutes, and in shade anywhere from half an hour to several days. There is no box speed and no development time; the print is watched under the negative and taken when it is right, which on this material means taken when it is too dark, because density is lost in the toning and fixing baths. Reilly’s starting point for a beginner is to overprint by about one and a half stops for albumen and two for salted papers.

And it self-masks, which is why the scale is so long. As silver builds up in the shadows of a printing-out print it behaves like extra density in the negative, delaying the arrival of maximum density and holding the shadows open until the highlights have had time to print in. Reilly reports Hübl’s finding that glossy albumen paper equalled platinum paper in scale length while salted papers exceeded platinum by a considerable margin, and that albumen’s contrast is distinctively shaped — a slow progression from shadows to middletones and a relatively abrupt jump from the middletones to white. A negative for albumen should therefore carry highlight detail worth printing.

The mechanism

The double replacement

The reaction is the same one that makes every salted paper, and it happens in the wet layer within seconds.

AgNO₃ + NH₄Cl → AgCl + NH₄NO₃

Hardwich's albumenised paper, which is salted with ammonium chloride

AgNO₃ + NaCl → AgCl + NaNO₃

The same reaction where the salt is sodium chloride, as Reilly writes it

Silver chloride is nearly insoluble, so it precipitates immediately as a fine white solid, dispersed through the albumen where the chloride ions were. The ammonium or sodium nitrate formed alongside it is soluble, takes no part in image formation, and either dissolves back into the bath or washes out during processing. Reilly notes the one thing it does do on its way past: it accumulates in the bath and helps coagulate albumen.

Why the excess silver is not slack in the formula

A sheet carrying only silver chloride prints grey, flat and slowly. Talbot found this empirically — papers with chloride and silver in equal amounts were hardly light-sensitive at all, and about six times more nitrate was needed. Vogel gave the explanation half a century later, and it is a cycle rather than a single step:

1. Light dissociates a unit of silver chloride into metallic silver and free chlorine.
2. The free chlorine would ordinarily recombine with the silver and undo the exposure.
3. Instead it meets free silver nitrate, and forms new silver chloride.
4. Light breaks that down too, and the cycle repeats.

Each turn deposits more metallic silver. The free silver nitrate is therefore not a reservoir of sensitivity but a chlorine scavenger, and the maximum density a print can reach is limited by how much of it is in the sheet. This is why a weak bath gives a print that “does not reach the bronzed stage, the action appearing to stop at a certain point”: the scavenger runs out.

The organic binder does the same job by a second route. Reilly classes albumen, gelatin and the organic acids — citric, tartaric and oxalic — as active organic substances, which both facilitate the more complete reduction of silver chloride and themselves form light-sensitive silver salts. Starch is inactive in that sense: it holds the image on the surface and contributes nothing to its formation. This is the chemical reason an albumen print is richer than an arrowroot one from the same negative, and the reason a starch paper wants citric acid in its silver bath and an albumen paper does not.

Coagulation, which is the part that is albumen’s alone

Albumen is a protein solution. Proteins are held in their folded, soluble state by weak forces that several things can break — heat above about 65 °C, alcohol, a shift in pH, and contact with the salts of metals. Silver nitrate is one of those salts, and it acts fast. The moment the sheet touches this bath, the albumen at the interface begins to denature and become insoluble, and in doing so it takes silver with it: Reilly names the product silver albumenate, and records the two facts that matter about it. It is insoluble, so it survives the wash, the toning bath and the fixer. And it is itself light-sensitive, so it is part of the image-forming system rather than merely the box the image sits in.

That is the mechanism behind Hardwich’s dismissal of the hot iron, and it is also the mechanism behind every failure mode on this page:

• A weak bath coagulates too slowly. “The weaker it becomes, the greater the tendency to dissolve away the Albumen without coagulating it. In such a case the greater part of the Nitrate of Silver is converted into Chloride, and not being properly retained upon the surface of the paper by the coagulated Albumen, falls away into the solution.” You lose the binder, the image layer and the silver together.
• An alkaline bath does not coagulate at all. “Alkaline Nitrate of Silver does not properly coagulate Albumen, and in consequence a Bath of this kind soon exhibits a white turbidity when the papers are floated upon it.” The white turbidity is your image layer, in suspension.
• A strong bath coagulates so fast that it seals itself out. Hence Reilly’s longer floats on stronger baths, and hence the whole difficulty of choosing a strength.
• Ammoniacal silver dissolves albumen instead of setting it. Hardwich: albumenised paper floated on ammonio-nitrate “loses its surface-varnish, and appears dead, like the plain paper”.

Where the brown colour of an old bath comes from

The same protein chemistry, running in the tray instead of on the sheet. Dissolved albumen and dissolved size react with a little of the silver to give what Hardwich calls a sub-albuminate of silver, partly soluble in silver nitrate solution and brown; over time some of that silver is spontaneously reduced to the metallic state, and finely divided metallic silver is black. Kaolin removes it by adsorption — the clay’s plate surfaces take up the organic matter and settle out carrying it. Curdy freshly precipitated silver chloride does the same thing less well, because, as Hardwich puts it, it “has an affinity for the brown Sub-Albuminate of Silver which produces the colour, and gradually carries it down”.

The identical reaction, on the sheet rather than in the tray, is why sensitised paper yellows in storage. Reilly: the yellow is spontaneously reduced metallic silver in a very finely divided state, and if left it goes on to reddish brown and finally to a bronzed black over some months of dark storage.

Function of every ingredient

Silver nitrate — 5.83 g per fluid ounce, 90 grains, 20.5 per cent w/v

What it is. Silver nitrate, AgNO₃, colourless crystals, freely soluble in water, made by the action of nitric acid on silver metal. It is the only invariable ingredient of any sensitising solution in this family, and here it is the only ingredient at all.

Why it is present. Because it is the only convenient source of silver ions that can be brought to a sheet of paper in solution. Silver chloride cannot be coated; silver nitrate can be, and then converted in place.

What it does — three things, which is one more than any other silver bath in this formulary.

1. It supplies the halide-forming ion. Ag⁺ meets Cl⁻ in the albumen layer and silver chloride precipitates immediately. About a third of the silver the sheet takes up is consumed doing this, on Hardwich’s own stoichiometry applied to his own two solutions.
2. It stays behind in large excess. The remaining two thirds dries into the sheet as unreacted silver nitrate and acts as the chlorine scavenger that lets the print-out reaction run to a real density.
3. It coagulates the albumen into silver albumenate, an insoluble, itself light-sensitive protein salt that anchors the image layer to the surface of the paper and survives every subsequent bath.

What photographic consequence follows. Everything visible about an albumen print except the toning. The maximum density, the bronzing of the shadows, the warmth of the untoned colour, the gloss that survives washing, and the fact that the image sits on the paper rather than in it are all downstream of this one salt at this one concentration.

With more of it. Up to a point, better prints: deeper shadows, warmer colour, a scale that runs further before it stops. Past about 60 grains to the ounce on Hardwich’s paper the improvement flattens out and you are buying very little for a great deal of silver — his own 80- and 100-grain baths differ “very little” from the 60. Past that again, three specific costs. The albumen coagulates faster, so the bath permeates the layer more slowly and needs a longer float. The tray holds an unreasonable amount of money. And every sheet carries more free silver into the wash, where it has to be removed before it can stain the print later.

With less of it. A sequence of failures that arrive in a definite order, and this is the most useful thing on the page for diagnosing a bad print. First the shadows stop short of bronze and the print goes pale in the fixer, cold and slaty. Then pale patches appear, worst at the edge that was uppermost while drying and least at the lower edge where the runoff drains and concentrates. Then the “measles” — Reilly’s word, taken from the old manuals — blotches and light spots in what should be dense areas, caused by insufficient sensitisation. And finally, below the coagulation threshold, the albumen itself starts dissolving off the sheet into the tray and you lose the coating.

What it interacts with. The chloride in the paper, which sets how much of it is consumed and therefore how much is left; the binder, which it coagulates and which dissolves part of it back; the water it is dissolved in, which must be distilled or it precipitates before it ever reaches the sheet; alkali of any kind, which stops it coagulating albumen; ammonia, which converts it into something that dissolves albumen instead; and every metal it touches.

Distilled water — 28.41 mL per ounce of bath

Not an inert diluent, and the only ingredient Hardwich qualifies. Ordinary tap water carries chloride and carbonate, and both react with silver nitrate the instant it is added. The bath clouds, the precipitate is silver chloride and silver carbonate, and the strength you weighed out is not the strength you have. Reilly makes the point vivid by pointing at the same reaction happening later on purpose: the cloudiness of the first wash water after exposure is the excess silver nitrate in the sheet meeting the ions in the tap supply. Bostick and Sullivan observe the practical corollary — a municipal supply high in chlorine clouds the first wash markedly, well water much less.

No source specifies a temperature. Silver nitrate dissolves readily at room temperature and there is nothing in the bath that needs warming.

Interactions

With the chloride: three published rules that do not agree, and one of them is this bath’s own.

Source	Rule	Applied to a 2.28 per cent salting bath
Talbot, via Reilly	About six times more nitrate than chloride	About 13.7 per cent
Hardwich, for albumenised paper floated	Nitrate bath about six times the salting bath, floated twice as long	60 grains to the ounce, 13.7 per cent
Reilly	Sensitiser about four times the salting solution	About 9.1 per cent
Hardwich, as actually printed	—	90 grains to the ounce, 20.5 per cent

Hardwich’s own rule and his own formula do not agree with each other, and he says so: the printed bath is “stronger than is actually necessary”. His rule and Talbot’s do agree, exactly, which is worth noticing given that the two men are working seventy years and one ocean apart from different starting points. Reilly’s rule is the outlier, and it is the recent one.

Hardwich also gives the two variants of his rule that the coating method demands, and they are a good demonstration that a “rule” here is really a statement about how much liquid stays on the paper. Salt by floating and sensitise by brushing, and the nitrate may be twelve times the salting strength. Salt by total immersion and sensitise by brushing, and the salt may be cut to a fifth, so that the proportions run about thirty of nitrate to one of salt.

With the binder. Albumen and gelatin are coagulated by the bath and starch is not, and that single difference reorganises the whole procedure. Reilly’s consequence: for starch binders — arrowroot, matte albumen, resin-arrowroot — use strong baths and short floats, because a starch layer stays permeable and a long float lets fine silver chloride sink into the fibres and dull the image. For glossy albumen, the layer seals itself and the float is longer.

With the paper size. Hardwich: a hard-sized English paper keeps the chemicals comparatively on the surface and does not need such concentrated solutions as a softer, spongier stock. This is not a small effect — it is the difference between his 40-grain bath being adequate and being inadequate on two papers salted identically. It also cuts the other way: an over-sized paper gives considerable gloss but tones and fixes with difficulty.

With citric acid. Reilly records both places it can go and distinguishes them. In the silver bath, 5 per cent gives the maximum preservative effect and 1 per cent a noticeable extension; in either the albumen or the silver bath it improves pre-processing stability and gives greater sensitivity and maximum density; but the print colour comes out more brownish when the acid is in the silver bath than when it is in the albumen, and purple tones are difficult in its presence either way. On a starch-bound paper it does more than preserve — it supplies the active organic substance the binder lacks and produces a redder, more brilliant print.

With the nitrates it makes. Ammonium and sodium nitrate accumulate with use and coagulate albumen. This is the one respect in which an old bath outperforms a new one, and it is why a fresh bath is the worst case for albumen dissolving off. Reilly’s boundary on the old advice to add them deliberately is firm: that was for baths of 5 to 8 per cent, and it does not apply at 10.

With alcohol and ether. Not in the bath, but in the measurement of it. Hardwich’s hydrometer reads density, and alcohol or ether present in a bath will make it read low by three or four grains an ounce. A printing bath should contain neither, but a sheet coated with an albumen carrying rectified spirit — Wall’s formula does, and Hardwich offers it as an option — will carry a trace in.

With the toner and the fixer downstream, and Reilly states the consequences rather than leaving them to be inferred. The free silver nitrate this bath leaves in the sheet is what makes the print, and it has to be taken out again before anything else happens. Reilly’s initial wash — “usually about 10 minutes in running water” — exists for that one purpose, and he gives both failures that follow from skipping it: silver nitrate still in the sheet “would retard or completely prevent any toning from taking place”, and if it is still there when the print is fixed, “black stains would be the result”. Bostick and Sullivan wash longer, 25 to 30 minutes with five complete changes, and warn that residual silver nitrate produces brown spots and foxing that may not appear for years. Reilly also gives the reason toning precedes fixing on a printing-out paper: toning forms silver chloride as a by-product, which would re-sensitise a print that had already been fixed. So the order is wash, tone, short wash, fix — and the first of those four steps is the one that belongs to this page.

Variants

Reilly’s modern bath, 10 to 12 per cent — the course’s recommendation. Published at salted-paper-sensitiser as 120 g of silver nitrate to a made-up litre. For albumen paper Reilly gives 10 to 12 per cent with a float of two and a half to three minutes and states that no additives are needed. Everything on this page about coagulation, exhaustion, kaolin and keeping applies to it unchanged; what changes is the amount of silver in the tray.

The citric acid bath, for keeping. Reilly’s “serviceable technique that requires the least effort” is to add 5 per cent citric acid to the sensitising bath — 50 g per litre alongside the silver. One per cent gives a noticeable extension of usable life. Two alternatives brush a 2 per cent citric acid solution on the paper instead, either before the salting or after the sensitising. Replenishment of the acid is on a volume basis rather than a strength basis: if a litre containing 50 g has fallen to 850 mL, the 150 mL of replenisher carries 7.5 g of citric acid as well as its silver. This is not published as a separate formulary entry because the register holds no id for it; that is bookkeeping, not a judgement about the formula.

🕰What the citric acid bath actually created

The preservative effect of organic acids was known by 1860. What nobody had realised, Reilly says, was that increasing the quantity of citric acid in the silver solution would improve stability so greatly. That step was published in 1869 by a Viennese photographer named Adolf Ost, and in 1872 the Sensitized Paper Co. of Portsmouth, Ohio, put the first ready-sensitised albumen paper on the market. The Getty Conservation Institute’s timeline records the same 1872 date from the conservation side, and notes that commercial albumen paper had existed since 1854 — but unsensitised, still needing this bath.

Professionals kept sensitising their own for two reasons: purple tones are difficult in the presence of citric acid, and doing it at home was considerably cheaper. Amateurs did not, and after the gelatin dry plate arrived in the 1880s very few of them ever touched a silver bath at all. The manufacturers’ exact methods were guarded secrets, and remain unpublished.

Towler’s plain silver solution, 1864 — the acidified version. Nitrate of silver 2 ounces, rain-water 12 ounces, nitric acid 2 to 3 drops; filtered before every use, held at about 70 grains to the ounce with an argentometer, five-minute float. Two things distinguish it from Hardwich’s. The nitric acid is in from the start rather than added as a correction, on the principle that “the sensitizing solution must always be slightly acid, in order that the whites may be thoroughly preserved” — which is the same claim Hardwich makes about spontaneous reduction in the dark, stated as a rule instead of an observation. And the water is rain-water rather than distilled: soft, low in carbonate, and what a nineteenth-century worker could get a barrel of. A modern reader should use distilled water and should not read rain-water as an endorsement of anything collected off a roof today.

Wall’s 1912 bath, about 60 grains to the ounce. One sentence in the Dictionary: the albumenised paper is sensitised by floating it face downwards on a solution of silver nitrate containing about 60 grains to the ounce, three or four minutes being sufficient. It is Hardwich’s own six-times rule arrived at independently forty-eight years later, and it is the lowest of the three period figures — the band was falling.

Bostick and Sullivan’s 15 per cent, current. Shipped ready-made as a litre of 15 per cent solution with 100 g of kaolin to keep it clean, and floated six minutes. The sheet it is designed for is double-coated and heat-hardened in a dry-mount press, which is a much less permeable object than a single-coated sheet, and that is the likeliest reason for a float twice Reilly’s on a bath half as strong again.

Ammonia fuming — a treatment of the sheet, not a variant of the bath. After sensitising and drying, the sheets were hung in a closed box over a dish of strong ammonia for five to ten minutes to increase sensitivity and contrast. Towler describes the box in detail and floats his sheets four or five minutes on a plain bath of sixty to seventy grains beforehand. Reilly’s assessment is worth quoting for its method as much as its conclusion: fuming was common in the United States and hardly practised in Europe, which is why the manuals appear to contradict each other; much of its value can be had instead by simply increasing the strength of the silver bath or aiming at a different negative density range; its value was greatest where a paper of low chloride content was sensitised on a weak bath; and for modern practice “the troublesome and unpleasant process of ammonia fuming seems unnecessary.” This course does not teach it.

Ammonio-nitrate of silver — excluded, and the exclusion is the point. It is more sensitive, gives greater intensity, and improves the colour. It also dissolves albumen. Hardwich’s Formula III is plain paper for that reason and Towler’s account of it is about the different chemistry it needs. Reilly adds the history: photographers who tried to fix the problem by neutralising the ammonia with nitric acid were only making ammonium nitrate, which the bath accumulates anyway.

Safety

Level B — advanced home lab. The hazard is not the procedure, which is a sheet of paper floating on a tray. It is the reagent, its concentration and the fact that the whole operation happens in dim light with wet hands.

Silver nitrate is corrosive and it is the eyes that matter. PubChem’s aggregated GHS classification for silver nitrate carries the corrosion, health-hazard and environment pictograms. Reilly’s warning is specific about the mechanism of injury and should be read literally: silver nitrate can cause permanent, irreversible eye damage, and sheets dripping silver solution pose a special hazard to eyes — the splash comes from above, off a sheet you are lifting, at the moment your attention is on the sheet. Bostick and Sullivan repeat it as a rule of the kit: always wear glasses or goggles. Wear eye protection for the whole session, not just the mixing.

Argyria, and why the gloves are not optional. Infrequent contact and the resulting skin staining do not appear to be dangerous in themselves, but prolonged exposure leads to permanent staining, and silver absorbed through the skin is ultimately deposited around the body — the conjunctivae, blood vessel walls, gums and mucous membranes. Reilly notes that consequences more serious than staining have also been reported. Tightly fitting gloves, and wash your hands immediately after any contact.

The stains are trivial as an injury and severe as a nuisance. Black on skin, clothing and almost any surface; a week or more to fade off skin; very difficult to remove from hard surfaces. Cover the bench.

Safelight discipline. Every operation from this tray onwards is a safelight operation, and the point where it begins is exact: the albumenised sheet is not light sensitive, and neither is the silver solution, until the two touch. Reilly’s rule is that white light must be excluded from all of sensitising and subsequent handling — though he adds that a brief look under a low-wattage incandescent lamp is helpful when judging the progress of a print, because yellow light hides the separation of the middletones and shadows.

What is not a hazard here, and why. There is no vapour and no fume: silver nitrate solution is non-volatile, nothing is heated and nothing evolves a gas, so a fume cupboard is not among the controls and ordinary room ventilation is adequate. There is no fire risk in the bath itself. There is no inhalation exposure except from the dry crystals during weighing, which is a minutes-long operation done with the container closed between scoops. And the sensitised paper is not a hazard to handle once dry. The controls that matter are eye protection, gloves, dedicated vessels and containment of the liquid — and they matter continuously, not at one step.

Do not improvise the dish. No food vessels, no shared utensils, nothing that will go back into a kitchen. This is standard for the course and it applies with particular force to a bath that stains everything it touches.

Storage

The bath: a stoppered dark glass bottle, over kaolin, labelled and dated. Reilly’s routine is the one to copy — return the bath from the tray to its bottle after every session, shake it up with the kaolin that lives there permanently (about 15 g per litre), and let it settle overnight. Bostick and Sullivan give the same procedure with a half teaspoon per bottle and 12 to 16 hours of settling, and add the detail that the settled matter is a fine black precipitate and that the cloudy bottom layer should be left behind when decanting. Filter before each session — a paper coffee filter in the funnel is sufficient, and will catch a little black silver powder.

It keeps essentially indefinitely. Bostick and Sullivan state 10 or more years for their unused 15 per cent solution. A working bath does not have a shelf life at all in the ordinary sense: it has a strength, which falls, and a colour, which darkens, and both are managed rather than waited out.

The sensitised sheet: this is the short-lived thing, and the sources disagree about how short. See the callout below. Whatever figure you work to, the storage that extends it is the same: dark, dry, cool, and with air excluded.

⁇How long does sensitised albumen paper keep?

The disagreement. Reilly (1980) says albumen paper sensitised as he describes “will remain in good condition for 24 to 48 hours, depending on environmental conditions”, that in hot and humid conditions it may yellow in 8 to 12 hours or less, and that for best results you should sensitise, print and process on the same day. Hardwich (1864) says “Sensitive albuminized paper will usually remain good for several days, if protected from the light, but it afterwards turns yellow from partial decomposition.”

What each is actually claiming. They agree completely on the mechanism — spontaneous reduction of silver, accelerated by air and moisture, visible first as yellowing. They differ on how long it takes, by a factor of several. Neither states a temperature or a humidity.

What may explain it. Three things, none of them established. Hardwich’s sheets carry a much stronger bath, and a sheet with more free silver may simply take longer to look degraded relative to the density it can still give. “Good” is a period judgement against a period standard, and Hardwich immediately qualifies it — papers kept even by his best method “can scarcely be considered equal to those recently sensitized”. And a cool, dry, unheated London workroom is not the same environment as a modern centrally heated one, which is the variable Reilly explicitly makes his figure depend on.

What the course uses. Reilly’s 24 to 48 hours, and the practice of sensitising, printing and processing on one day. It is the conservative figure, it is the one stated with its dependence on conditions, and nothing is lost by following it. Hardwich’s preservation trick is worth knowing anyway and costs nothing to try: dry the sheets in a warm place and pile them up under pressure in a printing frame, which excludes the air. He prefers it to keeping them between the leaves of a book, on the ground that printers’ ink may act on the surface.

The unsensitised albumenised sheet keeps for months and is stored dry, flat, and rolled albumen side out if it must be rolled at all. That belongs to the albumenised salting solution page; the one thing to carry across is that a sheet which has become too dry will not take the silver evenly, and Reilly’s remedy is a night in a damp basement or a few hours in a closed box with a dish of water before it is floated.

Incompatibilities

Ammonia and anything ammoniacal. The clearest incompatibility in the family, stated by two sources 116 years apart. Ammonio-nitrate of silver dissolves albumen off the paper. Keep ammonia out of the room, not merely out of the tray, if fumed and unfumed work are going on at the same bench.

Alkalis, carbonates and anything that raises the pH. An alkaline silver bath does not properly coagulate albumen. There are three routes to one, and all three are documented. Continued use with albumenised paper, because native egg white is pH 7.8 and every sheet carries some in. Hard tap water, by way of carbonate. And — the specific trap — kaolin containing carbonate of lime. Hardwich: kaolin “often contains Carbonate of Lime, and effervesces with acids: if so, it must be purified by washing in dilute vinegar, or the Bath will become alkaline, and dissolve off the Albumen.” He gives the same warning about animal charcoal, which he lists as the inferior decolouriser: the common grade contains carbonate and phosphate of lime, and the carbonate renders silver nitrate alkaline in the presence of ammonium salts. Test a new bag of clay with a drop of vinegar before it goes anywhere near a litre of silver.

Chloride, everywhere except in the paper. Tap water, table salt, sweat, an unrinsed salting tray, the fixer. Any of them will precipitate silver chloride in the bath and take strength out of it. Dedicate the tray, the funnel and the bottle; Bostick and Sullivan supply two funnels with their kit for precisely this reason and say that cross-contamination is the biggest single source of problems in albumen printing.

Metals. Stainless steel trays are not recommended; a brush with a metal ferrule reacts with the silver and stains the print; Hardwich lifts sheets with bone or wax-tipped tweezers. Silver nitrate is reduced by most base metals, which puts metallic silver where you do not want it and takes silver out of solution.

Organic contamination of any kind. A plastic tray that has held developer or fixer will give unwanted staining. Skin oils transfer to the paper and print as marks. Reilly’s and Bostick and Sullivan’s advice is identical and is worth taking literally: clean gloves at all times, and dedicated vessels for everything.

Thiosulfate, upstream. Fixer in a sensitising tray is not a contamination problem, it is a destroyed bath — thiosulfate dissolves silver chloride, which is the whole point of it four steps later. Keep the fixing bath at the other end of the bench.

Waste

Nothing silver-bearing goes down the drain, and on this page that is a longer list than usual. The exhausted bath, the tray rinse, the kaolin sludge in the bottom of the storage bottle, the used filter papers and coffee filters, the blotting-paper strips used to lift the scum, the drip tray under the drying line, and — the big one — the first wash water off every print. That wash is carrying the free silver nitrate that was two thirds of what the sheet took out of this tray. Bostick and Sullivan’s instruction for the solution itself is direct: dispose of 15 per cent silver nitrate with your other hazardous waste.

Saving it is also the economical thing. Reilly’s method for a print workshop is to precipitate the silver from saved wash water with sodium chloride or sodium carbonate, decant the supernatant, and send the solid to a refiner. The silver nitrate page carries the regulatory detail, including the concentrations at which a liquid waste becomes a listed hazardous waste in the United States and the typical municipal limits. Local regulation governs, and the course does not give a jurisdiction-specific instruction as though it were universal — find your own authority’s rule before you dispose of anything.

The paper waste matters too. Off-cuts of sensitised paper, blotters and the drip catcher all carry silver. Collect them dry, in one labelled container, and take them to the same place as the liquid.

Troubleshooting

Paper-white round spots with hard, definite edges. Air bubbles trapped under the sheet, which prevented the bath from touching the paper at all, so those areas were never sensitised. Reilly’s distinction is diagnostic: the same defect from the salting stage gives spots that are light brown rather than paper-white, because they were sensitised but carry no chloride. Prevention is Hardwich’s technique — bow the sheet convexity downwards, lower the centre first and let the corners down gradually, so the bubbles are pushed out ahead of the contact line — plus Reilly’s habit of blowing gently on any bubbles before the sheet goes down, and starting the clock only when the sheet lies flat.

Marbled or streaky stains over the whole sheet. The greasy scum of dissolved albumen on the surface of the bath. Draw it off with a folded strip of blotting-paper the width of the dish before each sheet; if the bath itself has a scum that keeps returning, filter it.

Blotches and light spots in what should be the densest areas — the “measles”. Insufficient sensitisation. Two causes, and Reilly separates them: a too-weak silver solution, or insufficient residual sensitiser left in the paper. Nothing can be done for prints already exposed. Unexposed sheets suspected of the same fault can be re-floated or re-brushed on a stronger bath. If it is appearing regularly, the bath needs testing rather than topping up.

Pale patches, worst along one edge. Uneven capillarity in the paper, made visible by a bath that is too weak to leave enough silver anywhere. Hardwich’s observation is a useful confirmation of the diagnosis: the patches are most abundant at the edge that was uppermost while drying and nearly absent at the lower edge, “where the excess of liquid drains down and becomes concentrated by evaporation”. A stronger bath, or a longer float on the same bath, both fix it.

A cold, slaty print that goes very pale in the fixer and never bronzed. The classic weak-bath print. Check the strength before assuming anything else, with a hydrometer or a titration.

A single small white mark, sharp-edged, near a corner. Hardwich’s, and it is a handling fault rather than a chemical one: lift a sheet by a pin, then re-insert the pin without letting the corner drain, and decomposition of the silver nitrate at the puncture leaves a white mark. Let the corner drain in the hand first.

White precipitate falling through the bath while sheets are floating, and albumen visibly coming off. The coagulation is failing. Two causes with one remedy each: the bath has gone alkaline (test it — it should be faintly acid, never alkaline), or it has simply gone too weak. Hardwich’s correction addresses both at once — ten grains of silver nitrate and half a drop of glacial acetic acid to each ounce. If the bath is fresh rather than old, suspect the opposite problem: a new bath has none of the accumulated ammonium or sodium nitrate that helps coagulate albumen, so shorten the float rather than lengthening it.

The bath is nearly black. Kaolin, overnight, then filter. If it has also acquired a surface scum that shows as a metallic marbled sheen on prints, filtering is not optional.

Sheets yellowing before you can print them. Spontaneous reduction, accelerated by air, warmth and moisture. Sensitise, print and process on the same day; if that is impossible, use the citric acid bath and accept the colour shift.

Brown spots and foxing appearing on finished prints, weeks or months later. Residual silver nitrate that was never washed out. Bostick and Sullivan are explicit that this damage is not immediately visible and can appear at any point in the future, and prescribe 25 to 30 minutes of first wash with five complete changes of water. Albumen prints need long washing, like fibre-based silver gelatin paper.

Experiments

1. Repeat Hardwich’s series. This is the best single experiment in the alternative-process part of the course, because a nineteenth-century author has already published the expected result and you are checking it rather than guessing. Salt one batch of paper, cut it into six, and sensitise on baths of 20, 40, 60, 80 and 100 grains to the ounce — 4.6, 9.1, 13.7, 18.2 and 22.8 per cent, or 46, 91, 137, 182 and 228 g/L — plus one at Reilly’s 12 per cent, all floated three minutes. Print all six under the same negative to the same visual endpoint, tone and fix them together, and record exposure times. Hardwich’s prediction is that 20 is useless, 40 is marginal and depends on your paper’s sizing, and 60, 80 and 100 differ very little. The question worth answering for yourself is where your paper’s threshold sits, because he says plainly that it moves with the stock.

2. The time-against-strength trade. Take the weakest bath from experiment 1 and float sheets for 3, 5, 10 and 15 minutes. Hardwich predicts the result improves markedly with time, “for the Albumen is allowed time to draw to itself more of the Nitrate of Silver”. Then take the strongest and float for 1 minute and for 3. He predicts the 100-grain bath is the best of the lot at one minute. If both predictions hold you have demonstrated that strength and time are substitutes over a wide range, which is not obvious and is not true of most photographic operations.

3. The coagulation floor. Float identically albumenised sheets on baths of 4, 6, 8 and 10 per cent for five minutes each, and look at the bath rather than the print: at some strength the water goes cloudy and the sheet loses gloss, because the albumen is dissolving faster than it is being coagulated. Repeat with a bath that has been used for twenty sheets. Reilly predicts the used bath performs better at the same strength, because of the nitrates it has accumulated. This experiment produces no prints and teaches more than most that do.

4. Fresh against seasoned, controlled properly. The obvious version of experiment 3 confounds two variables — a used bath is both weaker and richer in nitrate. Control it by making up a fresh bath at the measured strength of the used one, and compare. Any remaining difference is the nitrate.

5. Calibrate a hydrometer, or learn a titration. Hardwich’s instruction is still the right one: make up a solution of known strength and check the instrument against it. If you have a balance good to 0.01 g you can make a 90-grain standard and a 12 per cent standard in ten minutes. Reilly’s Volhard titration is the accurate method and is the one to learn if you print regularly; the reagents are sodium thiocyanate, a ferric ammonium sulfate indicator and a little nitric acid, and the endpoint is a colour change that will not stir out.

6. Kaolin, measured. Split a discoloured bath, clear half with kaolin overnight, and compare the two halves on identically salted sheets. Does clearing the bath change anything about the print, or only about how the bath looks? Hardwich says a bath highly coloured by albumen “acts nearly the same as at first”, which is a testable claim and a slightly surprising one.

7. Citric acid, both ways. Reilly states that citric acid in the silver bath gives a browner print than the same acid in the albumen. Prepare two albumenised batches, one plain and one with the acid, and two baths, one plain and one at 5 per cent, and print the four combinations from one negative. Record the colour before toning, after toning and after drying — three different colours, for the reasons given under The mechanism — and record how long each unexposed sheet takes to yellow.

Record everything in the lab notebook format: paper stock and sizing, salting formula and strength, bath strength as measured rather than as mixed, float time from the moment the sheet lay flat, temperature, humidity if you can measure it, the interval between sensitising and printing, and the exposure. On this material the interval matters as much as the exposure.