Albumen crazing and yellowing

Owned by The Albumen Print as an Object, and How It Decays, which treats both halves of this entry’s title as the two failures that dominate surviving examples and is careful about what follows for a maker: the yellowing is inherent to the material — silver bound at sulfur-bearing side chains that no fixer reaches, and the albumen’s own sugar attacking its own protein by the Maillard route — while the cracking is a brittle film glued to a support that moves with the humidity, and is the process’s clearest identification signature except on burnished prints. This entry was assembled from the albumen encyclopaedia entry and the conservation literature before that page was written.

What you see

A fine network of cracks across the whole surface, visible under magnification and, when severe, to the naked eye in raking light. And, separately or together, a yellow or yellow-brown cast in the highlights, often with the image itself faded to a flatter sepia.

✕One of these is not a defect, and the atlas has to say so

A network of surface microcracks is the identification signature of an albumen print, not evidence that something went wrong. The Getty’s atlas states it as the most typical microscopic characteristic of albumen photographs — a fine network rather uniform across the surface layer — and records that it is found in most albumen prints produced before 1870. Prints made after 1870, which were sometimes burnished and heat treated, may not fully show the pattern or may show no surface cracks at all.

So a period albumen print with microcracks is a period albumen print. What is a defect is cracking that is coarse, directional, or newly arrived — the kind that follows a bend, a roll or a flattening, and that the AIC attribute to humidity.

Likely causes

1. Humidity variation. The AIC’s finding is the mechanism: overall cracking of the albumen binder, likely caused by variations in humidity that caused differential dimensional changes between the image layer and the support.
2. Mechanical stress on a dry sheet. The Library of Congress state the general rule — low RH causes the binder to shrink and crack and the secondary support to curl — and advise against rolling, because a photograph may crack when it is unrolled for use.
3. Residual fixer or atmospheric sulfur, for the yellowing half — the mechanism that the residual-silver entry sets out in full.
4. A heavy or double coating. Reilly notes that two coatings can give a paper so heavily coated that it is brittle and hard to tone, and that thicker coatings are increasingly less permeable, so both faults are worse on them.

What is happening: the chemistry and physics

Two materials, two responses to water, one bond between them. Albumen swells and shrinks with humidity by a different amount and in a different way from the paper it sits on, so a change in humidity puts the layer into tension or compression. Do that repeatedly and the layer fails in the pattern that relieves the stress: a network of cracks. The AIC name variations rather than any single value, and that is the operative word.

The support makes it worse. Most albumen prints are on a lightweight paper, which the AIC record as particularly prone to tears, bends, creases, abrasion and losses — so the layer is bonded to something that moves easily and tears easily.

The yellowing is a different chemistry entirely and happens to share a page. Reilly’s account, given in full on the residual-silver entry, is that silver bonds to sulfur-containing side groups on the albumen protein, is not removed by hypo, and is converted to silver sulfide by labile sulfur from residual fixer or from the atmosphere. Albumen is unusually vulnerable because it supplies the binding sites, which is why Reilly says the phenomenon is peculiar to albumen paper and does not occur in quite the same way in otherwise similar gelatin and collodion papers — so much so that it often serves as an identification clue in its own right.

And the two faults keep company. Reilly records that severe highlight yellowing is accompanied in most cases by an apparent colour shift and density loss in the image, loss of highlight detail, and sometimes a greenish tinge; and that the time to onset is affected primarily by the moisture level and temperature of storage and the amount of residual thiosulfate. The same warm, damp cupboard drives both mechanisms.

!The confirming test: magnify the cracks, and compare the white

For the cracking, look at the pattern under magnification.

• A fine, uniform network across the whole surface: the process’s own signature, on the Getty’s account.
• Coarse cracks following a bend, a roll, a mat edge or a corner: mechanical, and recent. That is the defect.
• No cracks at all on a period print: consistent with a burnished, heat-treated post-1870 paper.

For the yellowing, use Reilly’s reference-white test. Lay a fresh, unsensitised sheet of similar paper beside the print. The human visual system pegs the lightest area present as white, which is precisely why the problem goes unnoticed; a true white beside it removes that adaptation.

And be careful about which fault you are treating, because the actions point in opposite directions: cracking asks for stable humidity and no flexing, and yellowing asks for clean chemistry and clean air. Neither is fixed by the other’s remedy.

Diagnostic questions

• Is the cracking uniform or directional? Uniform is the process; directional is damage.
• Has the print been rolled, flattened or flexed? Both conservation sources name that as the mechanical route.
• What has the storage humidity done? Variation is the cause, not any particular value.
• Is the print single or double coated? Double coating is more brittle and less permeable.
• How was it fixed and washed? For the yellowing half, this is the controllable variable.
• Is there density loss and a colour shift as well? Reilly’s association makes that corroboration rather than a second fault.

Corrective action

Neither fault is reversible in this course. For the yellowing, Reilly is explicit that no presently known treatment can remove it without completely altering the character of the print, because the silver sulfide is much more chemically stable than the colloidal silver of the image. For the cracking, a broken binder is a physical loss.

Stabilise the environment and stop handling it. That is the whole of the useful action for an existing print.

Support it rather than flattening it, as cracked emulsion from flattening argues at length.

Prevention

Stable humidity, not merely correct humidity. The AIC attribute the cracking to variation.

Do not roll, and if a sheet must be rolled, Reilly’s instruction for unsensitised albumenised paper is to roll it albumen side out, which minimises cracking of the layer.

Fix briefly, wash thoroughly and tone, which are the three controls on the yellowing half — and note Reilly’s own finding that a print that was inadequately fixed or washed will yellow in addition to fading.

Consider a single coating. Reilly’s own assessment of double coating is a list of costs: brittleness, reduced permeability, harder toning and fixing, and a greater tendency to curl.