Developing the Print Consistently and Running a Printing Session

Everything before this page happened in light: the negative was read, the exposure was anchored, the contrast was chosen, the hands moved over the paper. Then the sheet goes into a tray and the printer stops being an optical instrument and becomes a chemist with a stopwatch — and this is where most of the print-to-print variation in an amateur darkroom is actually made.

The reason is structural. Every setting on the enlarger is discrete and recoverable: an aperture clicks, a filter has a number, a head height can be measured again next month. Every property of a tray is continuous and drifting: the temperature is falling towards the room, the developer is being consumed by every sheet that goes through it, and the solution’s own oxidation runs whether you print or not. The enlarger is a set of positions. The tray is a moving state.

So the discipline of this page is the opposite of the discipline of the last one. Local control is about deliberate difference between one sheet and the next. Development is about holding everything constant so that the deliberate differences are the only ones there are. Put a number on the scale of the problem. ILFORD publish an ISO range of R95 for MULTIGRADE FB CLASSIC at filter 2, which is 0.95 log exposure units, or about 3.2 stops from base white to maximum black; a burn of a third of a stop is therefore about a tenth of everything the paper has. A tray that has drifted two degrees, or done forty sheets of work, can move the print by more than that without anyone noticing.

Development to completion, at the bench

Part XVIII established the chemistry: a paper developer is designed so that the maximum black reaches a limit inside the recommended time and then stops rising, while base fog rises without a limit — so there is a broad plateau between them where the print is finished and not yet spoiled. ILFORD publish that plateau directly, recommending 1 min 30 s to 3 min for MULTIGRADE FB CLASSIC in MULTIGRADE developer at 1+9 — one part of concentrate to nine of water, which is how the course writes every dilution — and stating that development can be extended to six minutes without any noticeable change in contrast or fog. Kodak’s tray table for POLYMAX T gives 45 s to 4 min for a fibre-base paper. Those are windows inside a plateau, not a contrast control, and this page does not re-derive the point. It asks the practical question that follows from it: given a plateau, what exactly does a printer do with a stopwatch?

Four things, and each of them has to be the same on every sheet of the session.

A fixed time. Choose one time inside the published range and print at it for the whole session. The range is there because different papers, dilutions and preferences sit at different points in it, not so that you can move around inside it while you work. The moment the development time becomes a variable, the base exposure you spent a whole lab establishing no longer means anything, because the exposure and the development were calibrated together.

A fixed temperature. ILFORD’s figure for their paper developers is 20 °C ± 1 °C. That tolerance is tighter than most darkrooms achieve by accident, and the next section is about why it matters and how to hold it.

An agitation scheme that does not change. ILFORD’s beginner sheet rocks the dish continuously, backwards and forwards, for the whole time. Kodak’s J-5 asks for the emulsion side to be covered by solution as quickly as possible and for the tray to be rocked or the print kept in motion throughout. For several sheets at once, both makers describe the same manoeuvre: slide them in one at a time, emulsion side down — ILFORD call it interleaving — and then continuously take the bottom sheet and move it to the top. What matters is not which scheme you choose but that you choose one and repeat it, because agitation sets how fast fresh solution reaches the emulsion and that is a rate, which means it is part of the development.

A defined start and stop. This is the one printers most often leave loose, and ILFORD’s own worked instruction shows how tight it should be: slide the paper in quickly and smoothly with no air bubbles on the emulsion, start the timer, rock continuously, lift the sheet with tongs at 50 seconds and drain it, and slide it into the stop bath as one minute shows on the timer. The drain is inside the development time. Kodak’s AJ-3 sheet does the same arithmetic differently, draining for five seconds after the developer and two after the stop. Either convention is defensible. What is not defensible is stopping the clock when the sheet leaves the tray on one print and when it enters the stop on the next, because a ten-second drain on a ninety-second development is eleven per cent of the time.

✕Snatching is a fault, and the control you actually want is the enlarger

The print looks too dark in the tray at fifty seconds, so it comes out. Everybody does it once.

Kodak state the mechanism and the symptom in one paragraph of J-5, and it is worth reading as a diagnosis rather than as advice. Overexposed and underdeveloped prints, they write, often have a muddy-looking appearance, with low contrast and weak blacks, because they are pulled from the developer in much less than the recommended time — and those images are often mottled from uneven development. ILFORD’s RC sheet says the same thing more briefly: prints developed for shorter times may be underdeveloped and lacking in contrast and density.

Read the three symptoms as three separate mechanisms, because they are.

Weak blacks are the plateau not yet reached: the deepest tone is still climbing when the sheet leaves, so the print has no true black to hang its scale on.

Low contrast follows, because the shadows lose more than the highlights do. The heavily exposed areas have the most silver still to deposit and therefore the most left to lose.

Mottle is the interesting one, and it is why snatching cannot be rescued by giving a bit less exposure to the next sheet. Agitation is never perfectly even in the first thirty seconds; the developer reaches some parts of the sheet before others. If the print goes to completion, that head start does not matter, because everywhere eventually finishes. If the print comes out early, the head start is frozen into the picture as a pattern of the way the solution first flowed. The atlas entries are muddy print from snatched development and mottled development.

The correct control for a too-dark print is less exposure. Kodak put it as an instruction rather than a prohibition, which is the more useful way round: adjust the printing exposure so that the print develops to the correct density within the recommended time. The tray is not a dimmer. The enlarger is.

🕰The road not taken: factorial development for bromide paper

The modern rule — fix the development, vary the exposure — is a choice, and the alternative was worked out seriously and published.

Wall’s 1924 compendium reports Dr B. J. Glover’s application of the Watkins factorial system to bromide papers. Using the Kodak amidol formula at 17 °C, Glover found that exposure multiplied by development time in seconds was very nearly a constant for prints of the same depth — his figure was 1800, from exposures of 30, 15 and 10 seconds against total development times of 60, 120 and 180 seconds, and Wall records the three results as practically indistinguishable. Glover’s first rule is the one worth keeping: development must not be for a shorter time than that required to produce the maximum black of the paper. His third sets the other bound: the maximum development allowable is that which just stops short of fog or stain.

What this is evidence for, and what it is not. It is evidence that exposure and development on paper trade against each other over some range, which is a real property of the material and is exactly why a snatched print is not simply a lighter one. It is not a licence to develop by eye: Wall himself notes that the factorial method is not applicable to the fast chloride development papers because the time of appearance is too short to measure, and the whole scheme depends on a particular developer, a particular temperature and a stopwatch on the moment of appearance. Modern manufacturers took the other branch — fix the time, control the exposure — because it is the branch that keeps agitation, colour and evenness constant as well as density.

The exception, and why it is not the same thing as snatching

Two techniques genuinely do interrupt development on purpose, and calling them “taking the print out early” is the error this section exists to prevent.

In water-bath development the sheet is moved from the developer into a tray of still water and left without agitation, then returned. In two-bath development the second tray is a weaker or differently balanced developer rather than plain water. In both, the sheet arrives in the second bath carrying developer inside the emulsion, and that developer keeps working. Where the image is dense it is consumed quickly and development stops locally; where the image is thin it lasts and development continues. The result is compression of the highlights with the shadow separation left alone. Part XVIII gives the mechanism and connects it to the adjacency effects Part VIII established for film.

The difference from snatching is the whole point, and it is a difference in whether development is still happening. A snatched print has been removed from the reaction: it stops where it stopped, unevenly, with the flow pattern of the first thirty seconds frozen in. A water-bath print is still developing, in a still bath, precisely because the reaction has not stopped — and the stillness is the mechanism, not carelessness, because agitation would refresh the surface and destroy the local exhaustion the technique depends on.

⁇What the course will and will not publish about these

The mechanism is not in dispute. Local exhaustion in an unagitated bath is the same physics as adjacency effects, applied to a tonal region instead of an edge.

No Tier 1 or Tier 2 source in this course’s corpus documents either technique for prints. Part XVIII searched for one and found none: no manufacturer sheet in the corpus mentions them, and no conservation publication describes them. What exists is practitioner convention, consistently reported but not traceable to a published procedure the course can cite.

So the course teaches them as mechanism and refuses to publish times for them. No number on this page tells you how long to leave a print in the water bath, because the course does not have one from a source it accepts and will not invent one. If you want to use the technique, treat it as an experiment of your own under Part IX’s framework: one variable, a control print made the ordinary way, and your own numbers recorded as yours.

And keep it out of a session whose purpose is consistency. A water-bath print is a different process from the print beside it in the drying rack, and a matched set cannot contain both.

Temperature, and what a tray does over an evening

A tray of developer is a thin layer of liquid with a very large free surface, sitting in a room. It does not stay where you mixed it. It moves towards room temperature, and it keeps moving, because the room is much larger than the tray and wins.

That gives the drift its direction, which is the first useful thing to know. In an unheated garage or a cellar in winter the tray falls, and it falls fastest in the first few minutes; in a small room with an enlarger lamp, a safelight and a person in it, the tray rises over the evening. ILFORD’s beginner sheet already assumes the first case and compensates for it in advance: mix the solutions at about 23 °C so that they settle at the working temperature of 20 °C ± 1 °C once the heat lost in mixing and cooling has gone. That is a manufacturer instructing the printer to aim off, which only makes sense if the drift is expected.

How much a couple of degrees is worth

Here the course has to be careful, because the evidence is thinner than the question deserves.

There is no published time-and-temperature compensation chart for paper in this corpus. ILFORD publish one, and it is titled a film development time and temperature compensation chart; its own text describes it as a guide for film and development combinations, and it carries a warning in capitals that development times below five minutes are not recommended because of the risk of uneven development. Every paper development time on this page is below five minutes. The chart is not merely un-transferable to paper; it explicitly excludes the region paper lives in.

What ILFORD do publish, in the machine-processing section of the MULTIGRADE RC sheet, is a table of development times against temperature for their 2000RT machine developer. It is the only paper developer time-against-temperature series in the corpus, and the course uses it for direction and order of magnitude only, with its limitations stated.

🧮The course's own derivation from ILFORD's machine table

ILFORD’s suggested development times for 2000RT developer, including transfer time to the next tank, are 46 s at 20 °C, 32 s at 25 °C, 22 s at 30 °C, 15 s at 35 °C and 12 s at 40 °C.

Take the ratio across each five-degree step and reduce it to a per-degree factor:

f = (t₂ ÷ t₁)^(1/ΔT)

Per-degree time factor

where t₁ and t₂ are the times at the two temperatures and ΔT is the difference between them in degrees Celsius. The four steps give 0.930, 0.928, 0.926 and 0.956. The first three agree closely: about a 7 per cent change in the required time per degree Celsius. The fourth does not, and the reason is visible in the table’s own caption — the times include transfer time to the next tank, which is a fixed mechanical interval that does not shrink with temperature, so it dominates once the development itself is down to a few seconds.

Consequence for a tray. Over the range a darkroom actually drifts through, a 2 °C fall requires about 15 per cent more time (1 ÷ 0.930² = 1.16) to reach the same development. A ninety-second development becomes about a hundred and four seconds.

What this is and is not. It is the course’s own arithmetic on ILFORD’s published figures for a machine developer in a replenished transport processor. It is not a compensation table for a tray of MULTIGRADE developer, and the course does not offer one. For comparison, ILFORD’s film chart implies about 9 per cent per degree from its own worked example — 8 minutes at 20 °C becoming 5 min 30 s at 24 °C — so the two independent figures agree on the order of magnitude and not on the coefficient. Use the direction, hold the temperature, and do not compute your way out of a drifting tray.

What the drift does to the print

Two effects, and only one of them is fixed by adding time.

Density and contrast. A tray two degrees cold, run at the nominal time, delivers roughly the development that 15 per cent less time would have given — which is to say the print is under-developed, and it shows the same three symptoms as a snatched print in milder form: blacks a little short, contrast a little low, and, if the shortfall is large enough, the beginnings of the uneven pattern.

Image colour. This one is real and it is manufacturer-sourced, though not through temperature directly. ILFORD instruct that MULTIGRADE RC COOLTONE be developed for approximately double the usual time to obtain the coolest image colour, and note that the developer’s capacity is then approximately halved. Foma make the mirror statement for their warm-tone material, where further dilution with proportionally lengthened development times gives progressively warmer tones. So the extent of development moves the colour of the silver as well as its amount, and a tray that has drifted cold and is being run at the nominal time is developing to a different point on that scale as well as to a lower density. Kodak’s toning manual arrives at the same place from the other side, instructing printers to use the development temperature and time recommended in the paper instructions specifically in order to obtain uniform image colour.

What the course does not claim is that a cold development which has been fully compensated in time gives a different colour from a warm one that has not. That is a plausible consequence of the deposition-rate mechanism in Part XVIII, but no source in the corpus tests it, and it is left as an open question rather than stated.

Holding it, cheaply

In descending order of how much they are worth for what they cost.

1. Measure the tray, not the room. A thermometer standing in the developer, read before the first sheet and again halfway through the session. ILFORD ask for every bath to be within 1 °C of the working temperature before starting; there is no way to know that without a thermometer in each one.
2. Aim off when mixing, as ILFORD’s beginner sheet does, so the bath arrives at the working temperature rather than passing through it.
3. Use more solution. Thermal mass scales with volume and heat loss with area, so a deeper tray of solution drifts more slowly than a shallow puddle in the same dish. This costs money and buys capacity as well as stability, which is the argument the next section makes for its own reasons.
4. Stand the tray in a water bath. A larger dish or a shallow sink of water at the working temperature, with the developer tray sitting in it, is the standard laboratory answer. It works not because water has some special property the developer lacks — both are essentially water — but because there is far more of it, so it takes far longer to move. The honest caveat is that the jacket drifts too: it buys hours rather than permanence, and a second tray of liquid in a dark room is one more thing to knock over.
5. Heat the room rather than the chemistry. A room held at 20 °C holds every bath at 20 °C with no further apparatus at all, and it fixes the wash water, the fixer and your hands at the same time. It is the most expensive item on this list to install and the cheapest to operate, and where it is possible it is the best answer here.

The atlas entry for what this looks like when it is not controlled is control chart temperature error, and the session-level version is session exposure drift.

Capacity, and the shape of exhaustion across a session

The tray also changes because of the work it does. Every sheet that goes through it takes developing agent out and puts bromide and oxidation products in, and the published capacity figures are the makers’ statements of how far that can go.

Part XVIII collected the numbers and this page will not repeat the table. What it adds is the conversion that makes three manufacturers comparable, because they do not publish in the same unit — and the arithmetic that turns a capacity into a decision about tonight.

An 8 × 10 inch sheet is 0.2032 × 0.254 m, which is 0.0516 m². That one constant reconciles everything.

Developer, working strength	What the maker publishes	Fibre, per litre	RC, per litre
ILFORD MULTIGRADE 1+9	8 × 10 prints per litre	50 sheets (2.58 m²)	100 sheets (5.16 m²)
ILFORD MULTIGRADE 1+14	8 × 10 prints per litre	40 sheets (2.06 m²)	70 sheets (3.61 m²)
Foma FOMATOL LQN 1+7	square metres per litre	29 sheets (1.5 m²)	58 sheets (3.0 m²)
Kodak DEKTOL 1+2	8 × 10 sheets per US gallon	32 sheets, not split by base	32 sheets, not split by base
Kodak POLYMAX T 1+9	8 × 10 sheets per US gallon	32 sheets, not split by base	32 sheets, not split by base

The conversions are the course’s own arithmetic on the makers’ published figures: Foma’s 1.5 and 3.0 m² divided by 0.0516 m², and Kodak’s 120 sheets per US gallon divided by 3.785 L. Kodak’s single column is not an oversight to be corrected — E-103CP and J-5 both give one capacity figure for a tray developer without separating the supports, and the course reports that rather than splitting it on their behalf.

Two patterns survive the conversion. A fibre print costs about twice a resin-coated one in both of the manufacturers who separate them, which is the base soaking up solution and carrying it out of the tray. And the three makers disagree by nearly a factor of two on fibre, from Foma’s 29 sheets to ILFORD’s 50, which is a reminder that a capacity is a judgement about acceptable quality expressed as a number, not a measured physical constant.

🧮What your tray actually holds, and where sheet forty falls

Work from the dish rather than from the litre, because the dish is what you own.

A 10 × 12 inch dish is 254 × 305 mm. Filling it to a depth of 12 mm — enough to cover an 8 × 10 sheet and keep it covered while the dish rocks — takes

V = 254 mm × 305 mm × 12 mm = 930 000 mm³ ≈ 0.93 L

Tray volume

Call it 900 mL. At ILFORD’s published 50 fibre 8 × 10 prints per litre for MULTIGRADE at 1+9, that tray has a capacity of about 45 fibre sheets. Sheet forty is at just under ninety per cent of it, which is exactly the region where the symptoms below start to appear — so the covers question this page was set, what does the fortieth sheet get, has a numerical answer and it is: less than the first, in a tray that size.

Count test strips too, by area. A strip 25 mm wide off the long edge of a 10-inch sheet is 0.025 × 0.254 = 0.0064 m², which is 0.12 of an 8 × 10. Eight strips are one sheet. A session that makes thirty strips and fifteen whole sheets has put about nineteen 8 × 10 equivalents through the tray, not fifteen.

And if you want forty fibre sheets, mix at least 800 mL by the capacity figure alone, and rather more than that if the session will also carry strips — which is the same conclusion the temperature section reached from thermal mass, by a different route.

The two clocks, and which one stops first

A tray of paper developer is running out in two independent ways at once, and both are published.

The work clock is the capacity above. The standing clock is oxidation, which happens whether or not you print: ILFORD state that working-strength MULTIGRADE, PQ UNIVERSAL and BROMOPHEN left in an open dish should not be kept for more than one working day, and Kodak give DEKTOL and POLYMAX T a working-solution life in a tray of one working day regardless of how many sheets have gone through. Kodak’s 1949 handbook makes the mechanism visible by publishing dish lives for the older formulas that differ by nearly a factor of fifty — 24 hours for D-72, 8 hours for the extra-warm-tone D-166 and thirty minutes for the amidol developer D-170. Part XVIII owns the mechanism as aerial oxidation and attributes most of that spread to the sulfite each formula carries, while warning that the formulas differ in their agents and their alkali too, so no single-variable law can be read off three rows. The session-level consequence is simply that whichever clock finishes first ends the bath, and for most home printers it is the standing clock, because a session rarely reaches forty sheets but always reaches the end of the evening.

📓The symptoms arrive in an order, and the first one is free to measure

Part XVIII lists them: slower emergence, then weaker blacks, then a warmer tone, then staining. The session-level point is which of the four you can catch before it costs you a print.

Emergence time is the one. It is published — approximately 10 seconds on a correctly exposed RC print and about 20 seconds on fibre with MULTIGRADE developer at 1+9 — it needs no equipment beyond the timer already running, and it moves first. So the cheapest process control in the whole of fine printing is one line in the log: the image appeared at N seconds, written for the first sheet of the session and for every fifth sheet after it.

When N has grown by half again on a sheet you know is correctly exposed, the bath is telling you the same thing the sheet count would have told you later, and it is telling you now. The atlas entries are exhausted developer and grey highlights from developer exhaustion.

Replenishment or replacement

The question has a clean sourced answer, and it is worth stating precisely because the literature does contain replenishment schemes for paper developers — they are simply not for trays.

Every replenishment figure in this corpus is for machine processing. ILFORD publish a replenishment rate of 150 to 250 mL per square metre of paper for their 2000RT developer, in the machine-processing section of the MULTIGRADE RC sheet. Foma publish 200 mL per square metre and label it, in as many words, as the recommended rate for the automatic processing. Kodak’s replenishable paper developer, POLYMAX RT, is described as a liquid concentrate for continuous or roller-transport processors, while the two tray developers in the same table, DEKTOL and POLYMAX T, carry no replenishment column at all and a one-working-day tray life instead.

No manufacturer in this corpus publishes a replenishment scheme for a tray of paper developer processed by hand. The course therefore says: replace it.

The mechanism behind the makers’ distinction is worth having, because it explains why this is not timidity. Replenishment holds a bath at a working point by adding a solution formulated to put back what a known area of paper removed — which requires that you know the area, that the throughput is steady, and that the accumulating bromide is being diluted out at the same rate. A transport processor satisfies all three by construction. A tray satisfies none of them: the throughput is irregular, the carry-out varies with the paper and the drain, and topping up a used tray with fresh concentrate gives a bath at an unknown bromide concentration and an unknown pH — which is to say a bath whose image colour and speed you no longer know, in a session whose entire purpose is that you do.

The tray line

Five positions, and the reason for each. The line is the same for fibre and resin-coated paper; only the times and the risks change.

The five positions, and what each one is for

1. 11. DeveloperFixed time, fixed temperature, unchanged agitation, drained on a fixed count. 20 °C ± 1 °C for the ILFORD paper developers; 1 min 30 s to 3 min for MULTIGRADE FB CLASSIC at 1+9, 1 min for MULTIGRADE RC.
2. 22. Stop bathILFOSTOP at 1+19 for 10 s, 18–24 °C, on both ILFORD sheets; Kodak Indicator Stop Bath at 1+63 for 10 s on RC and 15 s on fibre. ILFORD: strongly recommended, because it stops development immediately, reduces the risk of staining and extends the life of the fixer.
3. 33. First fixerHalf the single-bath time. This bath meets the sheet loaded with halide and does nearly all of the dissolving, so it is the one that wears out.
4. 44. Second fixerThe other half. It sees a sheet with almost nothing left to dissolve, so it stays close to fresh — and it is the last fixer the print meets, which is the whole argument.
5. 55. Holding waterA dish of clean water for prints not going straight to the wash, on ILFORD’s own instruction. Sheets separated and moving; for RC, total wet time under 15 minutes.

Times and dilutions as ILFORD and Kodak publish them for their own products; the sequence is common to all three manufacturers in the corpus. The wash that follows position five belongs to Part XII and to the finishing lab.

Why the stop bath is not optional even though development has finished. Development has finished in the sense that the plateau has been reached, but the reaction has not stopped — the sheet leaves the tray carrying alkaline developer inside the gelatin, and it goes on working until something removes it. Part X owns the chemistry. The session-level reasons are the two ILFORD give on both paper sheets: the risk of staining, and the life of the fixer, which is shortened by every millilitre of carbonate that arrives in it.

Why the second fixer is not optional for a print you will tone or keep. Kodak’s G-23 makes the case as a chain of consequences rather than a rule, and it is the clearest statement in the corpus: improper fixing is probably the major cause of stains in toned prints; an exhausted fixing bath contains insoluble silver compounds that cannot be removed completely by washing; and when those compounds later meet a toner they form a dark yellow stain, especially noticeable in the borders and the highlights. The instruction that follows is one sentence — use two-bath fixing for best results.

J-5 supplies the arithmetic underneath it. A single tray of POLYMAX T Fixer is rated at 100 8 × 10 prints per US gallon, 26 per litre. In the two-bath rotation the first bath is run to two hundred prints per gallon — twice as far — and then discarded, the second is promoted to first, and a fresh second is mixed; the cycle repeats four times before both are replaced, and both are replaced after a week of use whatever the count. The capacity of the set roughly doubles, but that is the smaller benefit. The larger one, in Kodak’s own words, is that the first bath does most of the work while the second remains relatively fresh — so the last fixer the print ever meets has always been nearly new. Part XI works the chemistry and the two-bath rotation SOP is the bench procedure.

Three constraints on the fixing positions that a printing session gets wrong more often than it gets the times wrong:

• Do not extend the fixing time. ILFORD state there is no benefit beyond the recommended time and that long times can cause image etching and, on RC, affect the paper’s image colour. Kodak set an upper bound of about ten minutes for fibre and two for RC, because prolonged fixing expands the paper and drives solution into the base where washing cannot reach it, and J-5 adds that over-fixing can reduce the silver image itself, most visibly in the highlights. The answer to a doubtful print is a fresher bath, not a longer time.
• Do not use a hardening fixer. ILFORD’s reason is washing efficiency; Kodak’s is that a hardened emulsion is less receptive to toner. Both point the same way, and both ILFORD RAPID FIXER and HYPAM are non-hardening. Part XI compares the types.
• Keep the sheets apart and moving. G-23 records what happens when they are not: stains where prints stick together or float, and round purple stains from air bubbles trapped between sheets, which show up after toning. The atlas entry is purple spots from fixer air bubbles.

The holding bath is water, and it has a time limit. ILFORD’s beginner sheet is explicit: a print that is not being washed straight away is stored in a dish of clean water. It is not a fixer, and it is not a wash — its job is to keep sheets wet, separated and out of the fixer while the session goes on. For resin-coated paper the constraint that governs it is ILFORD’s warning that prolonged immersion causes edge penetration and curl, so total wet time stays under fifteen minutes; a holding bath that has become a two-hour soak is a defect being manufactured. The storing unfixed prints SOP covers the other case, where the sheet has not yet reached the fixer at all.

Carry-over and contamination

Every position in that line is connected to the next by a wet sheet, and the sheet is a pump.

How much does it carry? No manufacturer in this corpus publishes a carry-over volume for a print. Part X derived one by running published stop-bath capacities backwards against the acid demand of the developer formulas, and reached 13 to 28 mL for an 8 × 10 fibre-base sheet, with the width of that range coming from the drain time and the endpoint criterion rather than from noise. The course claims that as its own inference and not as a manufacturer figure, and it has no figure at all for a resin-coated sheet — the polyethylene keeps the base dry, so almost all of what an RC print carries is the surface film and the swollen emulsion, which is much less, but much less is not a number.

🧪Measure your own carry-over in five minutes

It is one of the few numbers in fine printing you can get exactly, for your paper, your drain count and your hands.

Weigh a dry 8 × 10 sheet on the balance from Part II. Put it through the developer for its normal time, lift it and drain it for the count you actually use, and weigh it again. The difference in grams is the carry-over in millilitres to within a few per cent, because a working developer is mostly water. Repeat for a resin-coated sheet and for a wetter and drier drain.

Two results usually surprise people. The fibre figure is larger than they expected, and the drain count changes it more than the paper does — which is why a fixed drain is one of the four things that has to be the same on every sheet.

What it does downstream, in order of how much it matters.

Developer into the stop is the designed case, and Part X quantifies it: the alkali the sheet brings is what consumes the acid, and it is why a stop bath has a capacity at all.

Developer into the fixer is what the stop bath exists to prevent. Alkaline developer arriving in an acid fixer neutralises it, shortens its life, and — with an exhausted bath already carrying dissolved silver — is one of the classic conditions for dichroic fog.

Fixer into the developer is the direction that ruins prints, and it is the one the discipline is built around. Kodak state the symptom in G-23 without naming the contaminant: avoid contaminating the developer with other solutions, because developer contamination or exhaustion can cause image-colour variations. The mechanism the course offers for the specific case of thiosulfate is an inference from Part XI’s chemistry rather than a manufacturer’s statement — thiosulfate is a silver-halide solvent, so its presence in a developer dissolves halide that was about to be developed, and Wall’s own account of dichroic fog names a silver-bromide solvent in the developer as a condition that favours it. Either way the practical rule is not in doubt: a few millilitres of fixer in a tray of developer costs you the tray.

The discipline that prevents it is four habits, and ILFORD’s beginner sheet specifies three of them directly.

1. Dishes that cannot be confused. Different colours, or labelled DEV, STOP and FIX, and kept in the same left-to-right order every session so that the hand goes to the right place in dim light.
2. Two pairs of tongs with fixed jobs. ILFORD’s rule is directional and worth copying exactly: the developer tongs lift the print out of the developer and must not be contaminated by the stop bath, and the second pair, which lives with the fixer, is what moves the print from stop to fix. A pair of tongs that has been in the fixer never touches the developer.
3. A drained sheet, on a fixed count. The drain is the carry-over control and it is free.
4. Hands, and the geometry of the bench. The wet side stays away from the dry side and the paper box, splashes go forwards rather than sideways, and a hand that has been in the fixer is dried before it opens the paper. Handling paper in dim light and darkroom opening and closing are the standing procedures.

⚠The consistency controls are also the exposure controls

Print tongs and a dry-hand rule are on this page because they keep the baths apart, and the same two habits are the reason a fine printing session puts very little skin in contact with the chemistry — an alkaline developer, for which ILFORD publish pH 10.45 to 10.55 at MULTIGRADE 1+9, or a fixer. Where hands do go into a tray — and for interleaved sheets they sometimes must — nitrile gloves are the control, and the specific hazards, classifications and glove data for each substance are on the chemical pages and in Part II.

Waste. Every bath on this page ends the session as a waste stream: spent developer, spent stop, and fixer that carries dissolved silver and is the one that matters most. Part XII covers collection and recovery. What goes where is governed by your local regulations, which differ between jurisdictions and change; check them rather than following a procedure written elsewhere.

Sequencing the session

A printing session is a queue, and the order it runs in changes the results even when nothing else does.

Batch by paper and by filter. Every change of paper costs a re-anchored base exposure, and above the speed-matched range every change of filter costs one too — the filtration page has the arithmetic. Grouping the work so that those changes happen a few times instead of a few dozen buys back both time and paper, and it also keeps the comparison honest, because two prints made ten minutes apart on the same paper through the same filter differ by less than two made an hour apart across a change of both.

Print the difficult negative first. Two reasons and they point the same way. The tray is at its freshest, so the print you will judge most harshly is the one made in the most predictable chemistry. And you are at your freshest, which matters more than printers like to admit: the fourth hour of a session is where the burns get vague and the drain counts get sloppy.

Keep every strip with its print. A test strip is evidence, and evidence separated from what it was evidence for is litter. Number the sheets in the order they were exposed, write the number on the back in pencil in the margin before the sheet goes in the developer, and keep the strips for a negative clipped together with the prints made from them. The local control lab depends on being able to lay the versions out in order afterwards.

Interleave carefully or not at all. ILFORD describe the interleaving method for developing several sheets at once, and Kodak’s J-5 gives the same manoeuvre with the bottom-to-top rotation. It works, and it is how a production darkroom gets through a set. But it puts the sheets into the developer at different times, needs both hands in the tray, and makes the fixed-drain discipline harder. For a session whose purpose is a matched set rather than throughput, one sheet at a time is the more conservative choice, and it is what the times on the datasheets are written for.

Stop before the session degrades, and decide in advance what “before” means. Three things end a session and only one of them is about you: the developer’s working-day life, the fixer’s running total, and the point at which your own judgements stop converging. The signature of the third is a printer making the same change twice in opposite directions. The session exposure drift entry gives the confirming test: keep the first negative, and make an identical print at the end of the evening. Two prints, one variable — everything that happened in between.

Matching a set

An edition, a portfolio sequence or a triptych is a different problem from a single fine print, because the thing being judged is the relationship between the sheets rather than any one of them. A set of six prints each individually excellent and each half a grade different from its neighbour is a failure; six slightly ordinary prints that sit together is not.

Matching is easier within one tray life than across two, and the reason is now arithmetic rather than folklore. Everything else on the printing map is discrete and recoverable — the filter, the aperture, the head height, the times. The bath is the one variable that cannot be re-created exactly: it has a temperature that is drifting, a sheet count that only goes up, and a standing time that only goes up. Two sessions a week apart differ by a whole freshly-mixed tray whose behaviour you can describe but not reproduce. So the practical rule is simply print the set in one session, and if it will not fit into one, mix a full fresh tray for each part rather than letting one bath straddle the boundary.

ILFORD publish the technique that makes a batch converge, and it is the most useful sentence in their paper developer sheet for anyone printing a set: to maintain print-to-print consistency when batch processing a large number of either RC or FB prints, it may be advantageous to reduce exposure slightly and extend development.

⚗Why reducing exposure and extending development stabilises a batch

It works by moving the print’s working point further along the plateau, and the plateau is the whole reason there is a technique here at all.

Recall the shape: maximum black rises steeply and then flattens, and base fog stays low and then climbs. In the steep region a small change in the developer’s activity — two degrees, or fifteen sheets of work — produces a proportionally large change in the density that results, because the curve is steep there. Further along, where the curve has flattened, the same change in activity produces almost no change in density, because there is nothing left to develop.

A print exposed a little less and developed a little longer sits further into the flat region. It reaches the same final density by a route that is less sensitive to the thing that is drifting. The cost is real and should be recorded: you are spending part of the plateau’s headroom, so you have less of it left before base fog begins to lift, and the extra time reduces the number of sheets the tray will deliver.

This is a manufacturer’s recommendation, and the explanation above is the course’s reading of it against the mechanism Part XVIII established. ILFORD state the technique; they do not state this reason.

Three more things a matched set needs, in descending order of how often they are forgotten.

One box of paper. Paper is coated in batches, and the packaging carries codes that identify which. Where a set has to match, take it from one box, copy whatever code the box carries onto every map, and treat a change of box as a reason to re-anchor rather than as an accident. Part XVIII’s printing map already asks for the field; this is the situation that makes it earn its place.

A reference print in the room. Print the reference sheet first and judge every subsequent sheet against the object rather than against a memory — comparison against a physical print is a measurement, comparison against a recollection of one from twenty minutes ago is not. A wet reference will do for tracking within the session, provided the final judgement is made on dry sheets, because the wet-to-dry offset applies to the reference too.

Judgement on dry sheets, in one light. Dry-down applies to every sheet equally, which means a set judged wet may be internally consistent and uniformly too dark. The set is finished when it matches dry, under the light it will be seen in.

Reading the datasheet’s processing summary

Every claim on this page came out of a table like the ones below, and reading those tables properly is a skill in itself. Here are the three manufacturers’ own summaries for a fibre-base and a resin-coated paper, at the makers’ own recommended dilutions.

Step	ILFORD MULTIGRADE FB CLASSIC	ILFORD MULTIGRADE RC	Foma FOMABROM (FB)	Foma FOMASPEED (RC)	Kodak, POLYMAX T
Developer	MULTIGRADE 1+9, 1:30–3:00, 20 °C	MULTIGRADE 1+9, 1:00, 20 °C	FOMATOL LQN 1+7, 90–120 s, 20 °C	FOMATOL LQN 1+7, 60–90 s, 20 °C	1+9, 0:45–4:00 FB and 0:45–3:00 RC, 20 °C
Stop	ILFOSTOP 1+19, 10 s, 18–24 °C	ILFOSTOP 1+19, 10 s, 18–24 °C	Acetic acid at 2 per cent as published, or Fomacitro 1+19, 20–30 s	Acetic acid at 2 per cent as published, 10 s, or Fomacitro 1+19, 10–20 s	Indicator Stop Bath 1+63, 15 s FB and 10 s RC
Fixer	RAPID FIXER or HYPAM 1+4, 1:00	RAPID FIXER or HYPAM 1+4, 0:30	Fomafix 1+5, 3 min, or acid fixer 5 min	Fomafix 1+5, 90 s, or acid fixer 3 min	5:00–10:00 FB single bath, 3:00–5:00 in two; 2:00 RC
Wash	30–45 min, running water above 5 °C	2 min	30 min above 12 °C, 45 min below	2 min above 12 °C, 4 min below	1 h FB with a change every 5 min; 4 min RC

Two notes on the table before the reading. The Kodak column is POLYMAX T in its own J-5 sheet, whose developer times cover both supports where the paper sheets quote one; and Foma publish their acetic acid stop bath as 2 per cent without stating whether that is w/v or v/v, so the course reproduces the figure as they give it and does not resolve the ambiguity on their behalf. Their own Fomacitro at 1+19 is unambiguous and is the easier of the two to reproduce.

Five things to notice, because each of them is a different kind of claim.

A time is a range, and the range is not an invitation to wander. It brackets the plateau for that paper in that developer. Pick one number inside it and keep it.

A temperature is a condition on the time, not a preference. Every time in that table is quoted at a stated temperature and means nothing away from it.

The fixing times disagree wildly, and the disagreement is real. ILFORD ask for one minute on fibre in a non-hardening rapid fixer; Kodak’s range for a fibre print in a single bath is five to ten minutes. They are not describing the same chemistry — a rapid ammonium-thiosulfate fixer works far faster than a conventional sodium bath — and ILFORD’s short time is bound up with their optimum-permanence sequence, where a brief fix in fresh fixer is deliberately chosen so that less is driven into the base. Use the fixer’s own sheet, not another maker’s.

A capacity is a claim about acceptable quality, not a physical limit. ILFORD say as much on the RAPID FIXER sheet, of their own paper capacities: the figures for paper may be exceeded whenever print stability is not critically important. So the number tells you where the maker thinks quality falls off, and how far you follow it is a decision about what the print is for.

A wash time assumes a flow. Kodak’s one hour for fibre specifies a complete change of water every five minutes, and their four minutes for RC specifies a flow sufficient to change the water completely within it. A figure without its flow rate is not a wash specification. Part XII owns the physics and the archival sequence SOP is the version the course uses for prints it intends to keep.

What changes if you deviate is the last question, and the honest answer is that the makers tell you for some deviations and not for others. Published: shorter development gives weak, flat, mottled prints (ILFORD, Kodak); longer fixing etches the image and traps solution in the base (ILFORD, Kodak); an over-concentrated stop bath causes mottle in the base of a toned print (Kodak); heat drying shifts the tone cooler with some paper and toner combinations (Kodak); Hypo Clearing Agent is not recommended with resin-coated papers (Kodak). Unpublished, in this corpus: what a two-degree tray does to a tray-processed paper, what carry-over actually is in millilitres, and how a tray developer behaves when replenished. Where the sheet is silent, the course’s position is to say so and to measure rather than to guess.

What to record, and what a session log is for a year later

Part XVIII defines the two records and this page does not restate them: the printing map is one sheet per print, and the session log is one line per batch — date, bath temperatures, paper and box code, developer and dilution, emergence times, the sheet count against the published capacity, the fixer’s running total, the wash route, the drying method, the light the prints were judged under, and where the waste went.

What this page adds is the reason the second one exists, which only becomes obvious when you need it.

A printing map tells you what you did to a print. A session log tells you what the print was made in. A year later the map still works — the aperture, the filter and the increments are all recoverable — but every one of those settings was chosen against a particular bath, and the bath is gone. The session log is what lets you tell the difference between the map is wrong and this tray is not that tray, which are the two hypotheses you will have and cannot otherwise separate.

Concretely, when a repeat print comes out darker than the original, the log answers the questions the map cannot: was the original made at sheet three of a fresh tray or sheet forty of a tired one; was the developer bottle opened that week or six months earlier; was the room 16 °C in January and 23 °C in July; is this the same box of paper. Each of those is a plausible cause of a half-stop discrepancy, and without the log all four stay open.

!What a map and a log can and cannot promise

They can promise that every decision you made is recoverable, that a discrepancy has a short list of candidate causes instead of an unbounded one, and that a stranger could follow the instructions.

They cannot promise an identical print, and the course will not pretend otherwise. Paper is made in batches; makers reformulate and discontinue products; a developer concentrate ages in its bottle. The honest claim is that a map and a log get a repeat print close enough that the remaining difference is small, one-directional and diagnosable — and that without them the difference is large, arbitrary and mute.

Record the box code precisely because of this. It is the field that lets a future disagreement be attributed to the paper rather than to you.

Alternative route

Exposing a sheet of enlarging paper needs a room that can be made dark, and there is no substitute for that. But this page is mostly about what happens after the exposure, and a good deal of it survives without a permanent darkroom.

Processing needs darkness only until the print is in the fixer. ILFORD’s beginner sheet is encouraging on the point: print-making needs only a room that can be blacked out, and no running water, because the washing stage happens in daylight. Part XVI’s reversible blackout is enough, and the whole tray line above can be run at night on a temporary bench in a bathroom or a utility room, with the wash and the drying done afterwards in the light. Not on a food-preparation surface: the course’s laboratory discipline separates the two, and a kitchen worktop that carries trays of developer and fixer is the one temporary darkroom Part II rules out.

Without a permanent bench, the session discipline is what changes, not the chemistry. A temporary line has to be laid out identically every time, because the muscle memory that keeps fixer out of the developer in a fixed darkroom does not exist yet. Photograph your own layout once and set it up from the photograph; keep the tongs with their trays rather than in a drawer; and mix fresh working solution each session, which a temporary darkroom is going to do anyway and which is what ILFORD instruct directly and Kodak’s one-working-day tray life implies.

Where the physical session is impossible, three parts of this page still work on their own and are worth doing before you ever book a darkroom hour. The capacity arithmetic is a spreadsheet exercise on your own tray dimensions and your paper’s published figures. Reading the processing summaries is a reading task, and it is the one that makes an hour of paid darkroom time productive rather than exploratory. And the session log can be designed, printed and rehearsed in advance, which matters more in a hired darkroom than in your own, because you cannot go back tomorrow to check what the temperature was.

What has no alternative is the tray itself. There is no way to learn what a developer feels like at sheet forty except by taking a tray to sheet forty, and no simulation of a print’s emergence time substitutes for having watched a hundred of them. A community darkroom, a college facility or a shared session with another printer is the route; the honest statement is that this page describes an activity that requires trays of liquid and cannot be done any other way.

Recap

• The tray is the only part of printing that drifts, so consistency is four fixed things: the time, the temperature, the agitation scheme, and a start and stop that includes the drain.
• Snatching fails in three separate ways — weak blacks, low contrast, and mottle frozen in from the first thirty seconds of uneven flow. Kodak’s constructive form: set the exposure so the print reaches the right density within the recommended time.
• Water-bath and two-bath development are not the same thing, because in them the reaction is still running. No Tier 1 or Tier 2 source in this corpus documents either for prints, so the course teaches the mechanism and publishes no times.
• No paper time-and-temperature chart exists in this corpus; ILFORD’s is for film and excludes times below five minutes. From their machine-developer table the course derives about 7 per cent per degree Celsius, so a 2 °C drift is roughly 15 per cent of the time — a direction, not a table.
• Capacity converts. An 8 × 10 sheet is 0.0516 m². ILFORD give 50 fibre and 100 RC sheets per litre at 1+9; Foma give 29 and 58; Kodak 32 without splitting the supports. A 900 mL tray holds about 45 fibre sheets, eight test strips count as one, and two clocks run at once — both makers cap a tray at one working day whatever the count. Replace, do not replenish: every replenishment rate in the corpus is for machine processing.
• The line is developer, stop, two fixers, holding water. The second fixer is the argument, because it is the last one the print meets: G-23 traces the yellow stain in a toned print to insoluble silver compounds left by an exhausted single bath, which washing cannot remove.
• Carry-over is the pump between the positions. No maker publishes a figure; Part X’s inference is 13 to 28 mL for an 8 × 10 fibre sheet, and a balance plus a fixed drain count gives you your own. Two pairs of tongs with fixed jobs is the control that matters most.
• A matched set is made in one tray life, from one box, against a reference print, judged dry — and ILFORD’s batching advice, reduce exposure slightly and extend development, works by moving the print further onto the plateau where the drift matters less.
• The map records what you did; the log records what you did it in. A year later the log is what separates a wrong instruction from a different tray.

Check your understanding

Question 1. A fibre print at 90 seconds looks too dark in the tray. What is the correct response?

• Lift it out now and fix it; the density is what matters
• Leave it for the full development time and give the next sheet less exposure
• Move it to a tray of water so development slows down
• Extend development to six minutes so the highlights catch up

Show the answer and why

Answer: Leave it for the full development time and give the next sheet less exposure

A print pulled early is not lighter, it is incompletely developed, and it fails in three separate ways: the maximum black has not reached its limit, contrast falls because the heavily exposed areas lose the most, and the uneven flow of the first thirty seconds is frozen into the picture as mottle. Kodak state the constructive form of the rule in J-5 — adjust the printing exposure so that the print develops to the correct density within the recommended time. Moving it to water is a different technique with a different purpose, and extending development would make it darker still.

Question 2. Your darkroom is 18 °C and the developer tray has drifted to 18 °C with it. Using the course derivation from ILFORD published machine-developer table, roughly how much longer would the same development take?

• About 3 per cent
• About 15 per cent
• About 40 per cent
• It makes no measurable difference below 5 °C of drift

Show the answer and why

Answer: About 15 per cent

The table gives 46 s at 20 °C and 32 s at 25 °C, which is a per-degree factor of about 0.93 — roughly 7 per cent per degree Celsius. Two degrees compounds to 1 divided by 0.93 squared, which is 1.16, so about 15 per cent more time. The figure is the course own arithmetic on a machine developer in a transport processor and transfers as a direction rather than as a compensation table: the sound response is to warm the bath, not to compute a new time.

Question 3. You have a 10 by 12 inch dish holding 900 mL of ILFORD MULTIGRADE developer at 1+9, and a session of twenty-four whole 8 by 10 fibre sheets plus forty test strips 25 mm wide. Is the tray within its published capacity?

• Yes, comfortably: twenty-four sheets against a capacity of forty-five
• Yes, but only just: the strips add about five sheet-equivalents, giving twenty-nine of forty-five
• No: the strips count as whole sheets, giving sixty-four
• The question cannot be answered because capacity is published per print, not per area

Show the answer and why

Answer: Yes, but only just: the strips add about five sheet-equivalents, giving twenty-nine of forty-five

Count by area. An 8 by 10 sheet is 0.0516 square metres and each strip is 0.025 by 0.254 metres, or 0.0064 square metres, which is 0.12 of a sheet. Forty strips are therefore about five sheet-equivalents, and the session total is about twenty-nine against the roughly forty-five that 900 mL supports at ILFORD published fifty fibre sheets per litre. The tray is inside its capacity — but the one-working-day standing life still applies whatever the count, and for most home sessions that clock stops first.

Question 4. Why does Kodak G-23 insist on two-bath fixing for a print that will be toned?

• Because a single bath cannot dissolve all the silver halide in a fibre-base emulsion
• Because an exhausted bath leaves insoluble silver compounds that washing cannot remove, and they form a yellow stain when they meet a toner
• Because toners require a hardened emulsion and only the second bath hardens it
• Because the second bath removes the developer the stop bath missed

Show the answer and why

Answer: Because an exhausted bath leaves insoluble silver compounds that washing cannot remove, and they form a yellow stain when they meet a toner

G-23 traces the chain: an exhausted fixing bath contains insoluble silver compounds that cannot be removed completely by washing, and on contact with a toner they form a dark yellow stain, especially noticeable in borders and highlights. Two-bath fixing does not make any single bath better; it guarantees that the last bath the print meets is nearly fresh, which is why Kodak J-5 rotation promotes the second bath and mixes a new one rather than simply replacing both. Note also that G-23 asks for a non-hardening fixer for prints intended for toning, which rules out the third option.

Question 5. A print you made in March comes out half a stop darker when you repeat it in November from the printing map. Which record tells you whether the map is wrong?

• The printing map itself, since it carries the aperture and the exposure
• The session log, which records the bath the March print was made in
• Neither: paper batch variation makes the comparison meaningless
• The exposure log for the negative

Show the answer and why

Answer: The session log, which records the bath the March print was made in

The map records every decision, and all of those decisions are recoverable. What it cannot record is the state of a bath that no longer exists. The session log carries the March tray temperature, the sheet count against capacity, the developer bottle and its age, and the paper box code — which is exactly the list of candidate causes for a half-stop discrepancy. Paper batch variation is real and is one of those candidates, which is why the box code is logged rather than a reason to give up on the comparison.

Check my answersClear