Wall's soft-working gaslight paper

An eleven-kilogram batch of paper emulsion from a book that says plainly where its formulas came from: they are given as used commercially. This is the only complete published unwashed chlorobromide paper the course has found, which makes it the formula Part V’s second project is built on, and its silver charge divides by a hundred onto the first project’s exactly.

Solution A — the salted gelatine, and the vessel the emulsion is made in

Ingredient	Quantity	Form the source specifies
Potassium bromide	25 g
Sodium chloride	200 g
Citric acid (anhydrous)	50 g
Gelatin	625 g
Ethanol	250 mL	printed as alcohol
Water	5000 mL, added	at 50 °C; Wall prints 50 °C (122 °F) as the temperature at which A stands when the silver goes in. He states the water as an amount rather than as a make-up volume, and the only made-up figure on the page is the 11,000 g of finished emulsion, which is a weight rather than a volume and is recorded in the combination below.
Added to the working bath, not to the stock
Chrome alum	3 g	in the finals, after the digestion, when the emulsion is made up to Wall's 11,000 g
The finals line is printed by Wall as "chrome alum 3 g, water to 11,000 g", and the chrome alum therefore appears above as an addition to the made-up emulsion rather than as part of the salted gelatine. It is attached to Solution A because A is the vessel that becomes the emulsion: B goes into it, the digestion happens in it, and the finals go into what it has become.

Solution B — the silver

Ingredient	Quantity	Form the source specifies
Silver nitrate	500 g
Water	1660 mL, added

Mixed in this order — the emulsification and the digestion

1. Start with the whole of Solution A — held at 50 °C
2. Then add the whole of Solution B

Solution B is added to Solution A at 50 °C (122 °F), and the emulsion is then digested for seventy-five minutes at 50 °C. It is not washed. The finals - chrome alum 3 g - go in with water to make the whole up to 11,000 g.

The made-up figure is a weight and not a volume: Wall works in a tared crock and finishes at 11,000 g of emulsion inclusive of the finals; the schema's make-up field holds millilitres, so the number lives here rather than in a water line where it would be printed as though it were 11 litres. Wall adds on the following page that this paper and its contrast companion both carry a fairly high ratio of gelatine in mixing and are then diluted down to the required bulk, and that this is to prevent the formation of a coarse grain - so the dilution to 11,000 g is part of the grain control and not merely a make-up.

Purpose

To make a gaslight paper — a slow developing-out printing paper of mixed halide, exposed by contact and processed in a tray. Wall’s own classification puts it with the papers that are as a rule unwashed and that contain varying ratios of bromide and chloride of silver, against the bromide papers, which are always washed.

Soft-working is his name for this particular one, and it distinguishes it from the contrast paper printed immediately beside it on the same page. The two differ in more than one variable, which matters when they are used as a pair.

What “chlorobromide” means here is worth checking against the arithmetic, because the word suggests a balance the formula does not have: this is 94 mole per cent chloride and 6 per cent bromide. It is a chloride paper with a bromide addition rather than a halved mixture.

Recommended uses

The second emulsion in a controlled series. Its 500 g of silver nitrate is exactly one hundred times the 5 g of the chloride contact paper, so a one-hundredth batch changes the halide ratio and holds the silver charge. That is a genuine one-variable experiment and it is rare in this literature.

Learning what an unwashed emulsion is. No wash means every by-product of the precipitation stays in the coating — the potassium and sodium nitrate the double decomposition made, and the whole halide excess. The formula is stabilised by what a washed emulsion throws away.

Reading a period formula as a formula. This entry gives salts, acid, gelatine, alcohol, water, temperature, silver charge, digestion time, a hardener dose and a made-up weight. Very few emulsion formulas anywhere are that complete, and comparing it with a modern tested recipe is the fastest way to see what a modern recipe leaves out.

Warm tones, with the caveat under Image characteristics. A chloride-rich paper reaches them readily, and reaching them is a matter of the developer as much as of the emulsion.

When another formula is preferable

• When you want the simplest possible first make, the chloride contact paper: two cups, no acid, no hardener, and a maker who has published the results.
• When you want an enlarging paper, the bromide paper. A gaslight paper is deliberately too slow for an enlarger.
• When you want higher contrast, Wall’s own contrast paper on the same page — with the reservation that it reaches contrast through cupric chloride and hydrochloric acid, and this course does not carry a copper(II) salt into a Level B project. Its quantities are under Variants.
• When you need a formula somebody has demonstrably made in the modern era, none of Wall’s. Nobody in this corpus has made any of them with modern inert gelatin, and that is the standing cost of choosing a Tier 1 emulsion formula over a Tier 2 one.

Mixing

One crock, two additions, seventy-five minutes of heat and no wash.

1. Solution A. Wall prints the ingredients as a block and does not give a dissolving order, so none is printed here. What the block contains is 25 g of potassium bromide, 200 g of sodium chloride, 50 g of citric acid, 625 g of gelatine and 250 cm³ of alcohol in 5000 cm³ of water, brought to 50 °C.
2. Solution B. 500 g of silver nitrate in 1660 cm³ of water.
3. The emulsification. B into A at 50 °C.
4. The digestion. Seventy-five minutes at 50 °C.
5. No wash. Nothing is set, shredded or noodled; the emulsion goes straight to the finals.
6. The finals. 3 g of chrome alum, and water to bring the whole to 11,000 g.

!Wall's make-up is a weight, and the schema holds volumes

“Water to 11,000 g” is what the page says, and an emulsion is made up in a tared crock rather than in a measuring cylinder. That figure therefore appears on this page in the combination statement and in this callout rather than in a water line, because printing it as 11,000 mL would state something Wall does not say and would make every strength computed from it a little wrong.

It is the number the whole formula is calibrated against. Every concentration below — silver per kilogram, hardener as a fraction of gelatine, halide excess — is computed against the made-up weight, and the dilution to it is itself part of the design: Wall says these papers carry a high ratio of gelatine during the mixing and are then diluted to bulk, to prevent a coarse grain.

🧮What eleven kilograms of this emulsion actually contains

500 g AgNO₃ ÷ 169.87 = 2.943 mol

Silver

200 g NaCl ÷ 58.44 = 3.422 mol

Chloride

25 g KBr ÷ 119.00 = 0.210 mol

Bromide

(3.422 + 0.210) ÷ 2.943 − 1 = 23 mol per cent

Halide excess

chloride 94 mol per cent, bromide 6 mol per cent

Halide ratio

625 g ÷ 2.943 = 212 g per mole

Gelatine per mole of silver

3 g ÷ 625 g = 0.48 per cent

Hardener against gelatine

500 g in 11,000 g = 45 g of silver nitrate per kilogram

Silver concentration

Two of those figures deserve an argument rather than a nod.

The hardener. Wall’s own rule, from his hardening pages, is that the quantity used depends on the total quantity of the gelatine and not on the bulk of the emulsion — which is what makes the dose survive a change of scale. Against 625 g of gelatine, 3 g is 0.48 per cent. Duffin’s general band is 0.5 to 2 per cent, so this formula sits just below it; Wall’s own general range, one part of chrome alum to 250 or 300 of dry gelatine, is 0.33 to 0.40 per cent, so by his own standard it is generous. The two authorities do not quite agree and this formula lands between them, which is worth knowing before anybody “corrects” it upward. Expressed the other way — 3 g in 11,000 g, or 0.027 per cent of the emulsion — the number means nothing and does not survive scaling.

The silver concentration. 45 g of silver nitrate per kilogram is a negative emulsion’s concentration on Baker’s figures: he gives fifteen to twenty-five grams per litre for paper emulsions and forty to fifty for negative emulsions. Wall’s paper carries twice what Baker says a paper carries. The course does not resolve that. Two possibilities are open — that the two makers’ practices simply differ, or that Wall’s 11,000 g is not the bulk the paper was finally coated at — and nothing in the corpus decides between them. What follows practically is that a sheet coated with this emulsion at the volume the chloride paper uses receives about twice the coating weight, and the volume has to be halved to hold the weight constant.

A surfactant Wall names and does not dose. He writes that with all paper emulsions saponin or the quillaia tincture should be added and refers to a later page the course has not read. There is therefore no quantity to print, and neither substance has an entry in this course’s encyclopaedia. It is recorded here as a gap in the published formula rather than filled from elsewhere.

Behaviour

It is never washed, and everything follows from that. The by-product nitrates and the 23 per cent halide excess stay in the emulsion and end up in the coating. That is what stabilises a gaslight paper, and it is also why nobody can quote a keeping life for one: no source in this corpus gives a figure for how long an unwashed chlorobromide paper emulsion lasts.

It is digested rather than ripened by ammonia. Seventy-five minutes at 50 °C, with the halide excess as the only ripening agent. That is a long thermal digestion by the standards of the domestic recipes in this part, and it is the price of leaving ammonia out.

It is diluted down to bulk on purpose. The high gelatine ratio during mixing and the dilution afterwards are Wall’s stated method of preventing a coarse grain, and the made-up weight is where that happens.

It is hardened in the finals, at the bottom of the band. Chrome alum goes in at the end, which is where Duffin also puts it — into the molten emulsion immediately before coating — with the failure mode that too rapid an addition coagulates the emulsion through a temporary local excess. Add it slowly, to a moving emulsion.

And the hardening will only work if the emulsion is not alkaline. Duffin records that chrome alum is effective at the usual coating pH of about 6 and far less effective above it. This emulsion carries 50 g of citric acid, so it is not a formula at risk of that — but the rule is why the citric acid matters to more than the fog.

Image characteristics

Soft-working, which is all Wall says, and the authorities disagree about why. Wall’s own general statement is that the more bromide an emulsion carries against the chloride, the faster it is, the longer the scale of gradation, and the less easy it is to obtain warm tones. Baker says an excess of chloride tends towards higher contrast and a shorter scale. Duffin says chlorobromide emulsions are of greater speed and softer in contrast as the bromide content increases. Baker and Duffin point in opposite directions on contrast, and this course does not pick one. What all three agree on is speed and scale: more bromide is faster and longer-scaled.

⁇Halide ratio does not set image colour, and the best source here is the one that tested it

The obvious thing to expect from a warm-tone paper is that its warmth comes from its halide ratio. Wall tested exactly that and reported the result against himself.

He made a series from pure chloride through 5, 10 and 20 per cent bromide to pure bromide, exposed them all under a test plate of densities 1 to 3.01 to three different light sources, and developed them all in one metol-hydroquinone developer. In every case the same coloured image was obtained, from the lightest deposit to the deepest. With a weaker developer and increased bromide he got warmer tones — but the same warmer tones on all the plates. His conclusion is that the rate of development, which is to say the rate of deposition of the silver, is what decides the colour of the image, and not the composition of the emulsion.

What the ratio does set is speed and the shape of the scale, and hence how easily a warm tone can be reached. A chloride-rich paper develops slowly and gives fine grain, and those are the conditions a warm tone is reached under. That is a real and useful relationship, and it is not the one the phrase “warm-tone emulsion” suggests.

Extremely fine grain, which Wall attributes to the chloride and chlorobromide class generally and describes as an image more like a stain than a normal silver deposit.

No published curve, no grade, no speed number. Wall gives none for this formula and neither does anybody else. Every number in the Image characteristics of this entry is a word.

The mechanism

Two precipitations at once, in the same crock, competing for the same silver.

AgNO₃ + NaCl → AgCl + NaNO₃

The chloride, which is 94 per cent of the halide

AgNO₃ + KBr → AgBr + KNO₃

The bromide, which is 6 per cent

Silver bromide is much the less soluble of the two, which means the bromide is taken up preferentially as the silver arrives, and a crystal grown in a mixed-halide melt is not a mechanical mixture of two salts. What forms is a mixed crystal, bromide-rich at the core where the silver first met the halide and chloride-rich outside it, and that structure is a large part of why a small bromide addition changes the emulsion out of proportion to its molar share.

The 23 per cent halide excess is what drives both reactions to completion and what keeps the crystals in a halide-rich condition afterwards. In a washed emulsion the excess is removed; here it stays and does its stabilising in the coating.

The seventy-five minutes at 50 °C is Ostwald ripening — small crystals dissolving and redepositing on larger ones — with the excess halide as the only solvent. Without ammonia, that is the whole ripening mechanism available, and it is why the time is long.

The citric acid holds the pH down through both the precipitation and the digestion, which restrains the reduction of silver that produces fog and, incidentally, leaves the emulsion in the condition chrome alum needs to work in.

The hardening is the same chemistry as every alum treatment in this formulary and is dealt with on the chrome alum page, including the point at which that page stops: how the chromium(III) ion binds gelatin at the molecular level is not established by any source meeting this course’s standard, and this entry does not settle it in passing.

Function of every ingredient

Silver nitrate, 500 g in 1660 cm³. The silver: 2.943 moles of it, and almost all of the cost of the batch. It arrives as a solution rather than dry, and it arrives into a halide excess, which is what stops any of it surviving as free silver nitrate in a coating that will never be washed. More silver at fixed halide eats into the 23 per cent excess and, past it, leaves unprecipitated silver nitrate in the paper, which fogs and stains; less thins the coating weight. Because this emulsion is unwashed, the silver is also the source of the by-product nitrates that stay in the layer.

Sodium chloride, 200 g. The bulk of the halide — 3.422 moles, 94 mole per cent of the total — and what makes this a gaslight paper rather than an enlarging paper. Chloride gives the class its slowness, its short scale, its very fine grain and its readiness to yield a warm image under a weak developer. More chloride raises the excess and slows the paper further; less shifts the balance towards bromide, which on all three authorities means more speed and a longer scale, and on two of them means an argument about contrast.

Potassium bromide, 25 g. Six mole per cent of the halide and the ingredient the word chlorobromide is doing all its work for. Because silver bromide is much less soluble than silver chloride, this small addition is taken up first and ends up concentrated in the interior of the mixed crystals, so its effect on speed and scale is out of proportion to its share by weight. More bromide moves the paper towards the enlarging speeds of the bromide papers and makes warm tones harder to reach; less takes it back towards the pure chloride paper. It is also, in a different role entirely, the restrainer that a paper developer carries — which is a good reason not to think of bromide as a single-purpose ingredient.

Citric acid, 50 g. The acid, and it does two jobs that are easy to conflate. Through the make it holds the emulsion acid, which restrains the reduction of silver that shows as fog — the more valuable of the two here, because an unwashed emulsion has no wash to take away what provokes it. And it leaves the finished emulsion near the pH at which the chrome alum in the finals will actually harden, which Duffin puts at about 6 and falling away sharply above it. More acid than this would begin to attack the gelatin’s setting behaviour; less removes the fog restraint from a formula that has no other. Note that Wall’s contrast paper reaches its contrast partly through more acid, so the quantity is not photographically neutral.

Gelatin, 625 g — 212 g per mole of silver. Protective colloid during the precipitation, binder in the coating, and here a third thing as well: Wall states that this paper carries a high ratio of gelatine during mixing and is then diluted to bulk, and that this is done to prevent a coarse grain. So the gelatine concentration at the moment of precipitation is a grain control, and the dilution afterwards is part of the same decision. More gelatine makes a finer grain and a thicker, slower-drying coat; less coarsens the grain. Its type matters as much as its mass: Wall’s formulas assume an active gelatin whose own sulfur contributes sensitivity, and a modern inert photographic gelatin carries one to two parts per million of active sulfur against about a hundred. An inert gelatin in this formula will give a slower and cleaner emulsion than Wall got, and that is the single largest reason nobody can promise you his result.

Ethanol, 250 cm³, printed as alcohol. A coating aid rather than a reagent. It lowers the surface tension of the melt so it wets the paper and flows out, and it speeds the setting and the drying. More would begin to affect the setting of the gelatine; less costs coating quality on a sized paper. It is one of the standing differences between a paper emulsion and a plate emulsion, and Wall’s commercial chloride finals carry a great deal more of it than this.

Chrome alum, 3 g, in the finals rather than in the mix. The hardener, and its position in the sequence is the formula: it goes into the made-up emulsion, which is where Duffin also puts it — introduced carefully into the molten emulsion immediately before coating, because the reaction is fast and because too rapid an addition coagulates the batch through a local excess. Against the gelatine it is 0.48 per cent, which is just below Duffin’s 0.5 to 2 per cent band and above Wall’s own 0.33 to 0.40 per cent rule. More hardens further and makes the layer harder for the fixer to penetrate, which shows up weeks later as a thick patch that has darkened; less leaves a tender layer. It is chromium(III), not the chromium(VI) of dichromate, and the difference is the whole of why this course permits it — see chromium.

Water, 5000 cm³ in A and 1660 cm³ in B, and then enough to make 11,000 g. Not a filler at any of the three points. The 5000 cm³ sets the concentration at which the crystals form; the 1660 cm³ sets how fast the silver arrives per unit of stirring; and the make-up is the grain control described above. Wall states A’s water as an amount at a stated temperature rather than as a make-up volume, and this page keeps that distinction because it is his.

Interactions

With the developer, which decides the image colour. This is the interaction Wall’s own experiment establishes, and it is the reason a warm-tone paper is not warm on its own.

With light, from the moment the silver goes in. A chlorobromide reaches further into the blue than a pure chloride, so the safelight margin that served the chloride paper should be re-tested rather than assumed. The transmitted colour of the wet emulsion is itself a rough guide — white for a chloride, yellow for a bromide — and the practical answer is a safelight test on this material.

With bromide in the developer. A soft-working paper developer of the period already carries several grams of potassium bromide per litre as a restrainer, which is the same ion the emulsion carries. The two are doing different jobs at different concentrations and it is worth keeping them apart in your head when you are diagnosing fog.

With the coating temperature. A hardened emulsion sets fast, and chrome alum’s action is described as rapid; this is not a melt to leave standing.

With alkali of any kind, which takes the pH away from the point where the chrome alum works.

With copper, which appears in the neighbouring contrast formula and not in this one. Wall uses cupric chloride as a contrast agent; the course leaves it out and says so under Variants.

Variants

The course’s own one-hundredth scaling, which is the version Part V performs. 0.25 g of potassium bromide, 2.0 g of sodium chloride, 0.5 g of citric acid, 6.25 g of gelatin, 2.5 mL of ethanol, 50 mL of water in A; 5 g of silver nitrate in 16.6 mL in B; the finals 0.03 g of chrome alum, delivered as 0.6 mL of a 5 per cent stock; made up to 110 g. That scaling is the course’s arithmetic and not Wall’s, and the lesson labels it as a variant the course owns. Two things about it are worth stating here. Scaling preserves every ratio — halide to silver, gelatine to silver, salt to salt, hardener to gelatine — and those are the numbers that define the emulsion. Scaling preserves none of the reactor: addition rate per unit volume, heat transfer and stirring are all different in a 60 mL cup from a 45-litre crock, and addition rate is the variable that decides how many crystals form. The full working is in the chlorobromide project.

Wall’s contrast paper, printed beside this one on the same page: potassium bromide 13 g, crystalline strontium chloride 433 g, gelatine 625 g, sodium acetate 100 g, alcohol 160 cm³, hydrochloric acid 25 cm³ and water 5000 cm³; silver nitrate 500 g in 1650 cm³; one hour at 60 °C; finals of chrome alum 3 g, hydrochloric acid 25 cm³, cupric chloride 5 g and water to 11,000 g. Wall states that the contrast is controlled by the quantity of the acid and the copper solution and that reducing them reduces the contrast. The course does not carry it: a copper(II) salt adds a hazard and a waste stream to a Level B project for an effect reachable by other means, and strontium chloride has no encyclopaedia entry here. The two papers also differ in five variables at once, so they are not a controlled pair.

No safety variant of this formula is offered, because there is nothing in it to substitute: the whole list is a halide, a common salt, a food acid, gelatine, alcohol and a chromium(III) hardener at half a per cent of the gelatine.

Safety

Level B, and silver nitrate sets it — Danger, with the oxidiser, corrosive, health-hazard and environmental pictograms. At Wall’s own scale this formula would put half a kilogram of it on a bench, which is one of the reasons the course performs the scaled version.

Chrome alum is the second reagent to plan for: Warning, GHS07, skin and eye irritation with possible respiratory irritation, no sensitisation and no carcinogenicity notified, and a UK limit of 0.5 mg/m³ as chromium. Weigh it once, over a tray, in still air, and dose from a 5 per cent stock thereafter — 0.03 g is not a weighing anybody should attempt.

This is chromium(III). The chromium policy holds the distinction once for the whole course and is not restated here.

Citric acid is Level A, and the remaining ingredients are table salt, a photographic halide, gelatin and drinking-grade alcohol.

The ethanol is the fire hazard nobody expects on an emulsion bench, because it is being handled beside a 50 °C water bath.

Nothing here needs ammonia, and that absence is deliberate: Part V excludes ammoniacal makes at Level B, and this formula is one of the reasons it can.

Storage

No source publishes a keeping life for an unwashed chlorobromide paper emulsion, and none is invented here. That is a real gap and it is the practical reason to make an emulsion at the size you will coat.

What the literature does give is a direction: Wall notes that bromide emulsions are sometimes poured into ice-cold pans immediately after digestion to set quickly and stop the ripening. An emulsion that is still warm is still ripening.

Refrigerate, dark, lightproof, and date it. That is the practice the modern tested recipes in this part use, and it is transferable even though their keeping figures are not.

The coated paper is dried in the dark and stored lightproof. The only keeping criterion anywhere in the corpus for a coated material is Baker’s industrial oven test — ten days at 105 °F with no more than 0.02 extra fog density — which is a factory specification and not a shelf life.

The dry chemicals by their own encyclopaedia entries: silver nitrate dark and dry; potassium bromide airtight, because it cakes; chrome alum dry, closed and away from anything carrying sulfite.

Incompatibilities

Silver nitrate with organic matter, dust, skin and bare metal. See incompatibilities.

Chrome alum with sulfite, which destroys its hardening power — a fact about the finals bottle rather than about this emulsion, since there is no sulfite in the make.

Chrome alum with an alkaline emulsion, which is a real risk in a formula without the citric acid and not in this one.

Thiosulfate anywhere near the make, which dissolves silver halide. One set of vessels for the emulsion.

Copper and iron, both of which produce their own defects in an emulsion; the copper here is a matter of not adopting the neighbouring formula rather than of contamination.

Bacteria and mould. This formula has no bacteriocide, and eleven kilograms of warm gelatine is a nutrient broth. Small batches, cold storage and use.

Waste

Silver-bearing throughout, and there is no wash to dilute it. An unwashed emulsion means the silver stays in the emulsion until it reaches the coating and then the processing baths, so the waste is concentrated in the fixer rather than in a wash tank.

Chromium(III) is present in the finals and therefore in everything downstream — the coating vessel rinse, the processing baths, the wash water. It is a small quantity, 3 g in 11,000 g of emulsion, and it is still a chromium stream: follow the chromium(III) rinse route and the disposal ruling.

Bottle and label the two streams separately where you can tell them apart, and label the mixture honestly where you cannot.

Local regulation decides, and this course states the chemistry without computing a threshold.

The option that puts no chromium anywhere is the course’s default of no hardener at all — which for this formula would be a variant, because the hardener is in the published finals.

Troubleshooting

Coarse grain. Wall’s own diagnosis and remedy are in the formula: a high gelatine ratio during mixing, diluted to bulk afterwards. If you have scaled the formula and mixed it at a lower gelatine concentration than the ratio implies, that is the first thing to check.

Fog on an unwashed paper. The citric acid is the formula’s only fog restraint, and it is easy to under-measure at the scaled size. Check the acid, then the digestion time, then the safelight — in that order, because the first two are in the formula and the third is not.

The coat is twice as heavy as the chloride paper’s at the same volume. It is: this emulsion carries about twice the silver halide per unit weight. Halve the volume per sheet to hold the coating weight constant, and record the volume rather than the number of pourings.

The emulsion coagulated when the finals went in. Duffin’s failure mode exactly: the chrome alum was added too fast, or into a still emulsion, and a local excess did the rest. Add it slowly to a moving emulsion, and add it as a dilute stock rather than as a solid.

The layer frills in the developer. At 0.48 per cent of the gelatine the hardener is at the bottom of Duffin’s band, and a hand coat is thick. Fix in an acid hardening fixer rather than raising the chrome alum, and if you do raise it, record it as a variant.

The prints are colder than expected. That is the developer, not the emulsion. Wall’s series is the evidence: a weaker developer and a slower deposition give warmer tones on any of these papers.

The paper is slower than the chloride paper rather than faster. Possible, and worth recording rather than explaining away: an inert modern gelatin removes a sensitivity Wall’s active gelatin supplied, and nobody in this corpus has measured how much.

Experiments

The halide-ratio series, done properly. This formula at one hundredth, the chloride paper, and a bromide make, all at 5 g of silver nitrate, all coated at the same computed coating weight, all printed on the same negative and read on the same step wedge. Speed and scale should move as all three authorities say; colour, on Wall’s evidence, should not.

Test the colour claim directly. One emulsion, four developers — a normal paper developer, the same one diluted, a warm-tone developer and one carrying extra bromide — and the same negative on each. Wall says the developer moves the colour and the emulsion does not. This is the experiment he actually ran, at a size you can afford.

Vary the hardener against the gelatine, not against the emulsion. Batches at 0.33, 0.48 and 1.0 per cent of the dry gelatine, and a fixing test on each. You will be sampling from Wall’s rule, through this formula, into Duffin’s band, and the frilling behaviour is the measurement.

Coat at the trade’s weight and at the hand coater’s. Baker’s one litre to sixty or eighty square feet against what your rod actually lays down. Then look at maximum black on both, remembering his warning that too thin a coating is mistaken for poor maximum black.

Put an active gelatin against an inert one. The single largest uncontrolled variable in reproducing any formula from this book. Two batches, everything else held, both read on a wedge. Expect the active one to be faster and foggier, and record how much of each.