I've often heard about "reticulation" of gelatin without ever encountering it...

Alain Oguse - July 2025

Does it exist? Can it be treated?

While I was learning the professions of color lab technician and then photographer at the end of the 60s, I often heard about "reticulation" of gelatin. Yet I had never seen it firsthand. This danger seemed to me a bit like the threat made to a child that if he lies, his nose will grow.

Let's look today at this portrait of a friend from 1972, printed at the time in 18 x 24 cm on an enlarger. Nothing special to note. It's just a photo among friends; technically, we paid little more attention to it. The print is probably stored somewhere... Fifty years later, we love to find these old negatives. And if we digitize it? What would a digitization with my process in green point light give?

IMPORTANT➔ : The images in this article try to reproduce the rendering for specific paper formats. However, web browsers primarily respect your system's display settings rather than the actual size of pixel images. Thus, for a Windows system with display settings set to "125% recommended," instead of the rendering of an 18x24 cm, you will get 25% more, almost a 24x30 cm. The issue is different under Linux, though somewhat similar. On Mac, the problem seems less present.

I strongly encourage you to make the necessary adjustments so that the images display at their actual pixel size. Otherwise, it is hardly possible to evaluate the nuances presented here.

To do this, two solutions are possible:
- The most precise is to temporarily set the display parameter to 100%,
- The simplest is to reduce the browser zoom (Ctrl + mouse wheel). In the case of a display at 125%, reduce to 80% for an approximate result.

Digitization in Point Light (Green)

Well yes, this is the first reticulation I have encountered. This deterioration affects the entire surface of the image uniformly. The white cracks are very present, invasive. Moreover, the grain seems disorganized, confused, uglified. Finally, the negative is scratched and very dirty. It will need to be cleaned delicately with isopropyl alcohol and undergo a good "retouching" session.

What happened?

Searching online, I discovered that the images presented regarding photographic film reticulation are very different from what we observe here. The only indicated cause is a temperature deviation during processing, so significant that it could only be intentional https://cmsw.mit.edu/wp/wp-content/uploads/form-maker/Photographic%20Film%20Reticulation%20Due%20to%20Temperature%20Changes%20during%20Film%20Development.pdf. In our case, this is excluded. Everything was done at exactly 20°C, I am certain. So then?

This TriX 135 negative was carefully stored in a binder with 30 sheets (i.e., 30 films of 36 exposures). It turns out that ALL are in the same condition. After having tortured the owner for a long time, I finally learned that this black plastic binder had been forgotten for a few days on a desk near a south-facing window in the middle of summer. In such a situation, the surface of the binder could have reached, and perhaps even exceeded, the pain threshold temperature of 50°C. This cause seems plausible, doesn't it?

Can this image be saved?

There's no question of missing such a great opportunity to confront my recent experiences with such a rebellious reality. The first idea is to try to soften the reticulation cracks using the Diffusion or Sharpness module of Darktable/Ansel, somewhat like I've gotten used to doing to reduce grain in point light (when the subject allows it). This would be a good opportunity to further explore the multiple (but delicate and demanding) possibilities of this module. The first attempts were disappointing. Then I discovered a passage in the documentation that states:

"Natural diffusion usually occurs from points of high potential (high energy or high particle concentration) to points of low potential (low energy or low particle concentration). In an image, this means that diffusion always occurs from the brightest pixels to the darkest pixels."

If I interpret this correctly, in our case, the diffusion should extend from the dark parts (the non-reticulated parts) to the bright parts (the reticulation cracks). Therefore, the module should be moved in the "pipe" to place it before inversion by the Doctor nega module. Didn't I think of this sooner? Immediately, the results improve. But it's still not magical. I had to experiment a lot to achieve a tangible reduction in reticulation, without causing a blur even more detestable than the reticulation. The balance is delicate to find, and I am far from mastering all the intimidating possibilities of this module.

Digitization in "Attenuated" Point Light

Here I find my limits in trying to mimic diffuse light with this technique. Certainly, in small doses, to reduce grain for example, I get good results, but on the other hand, to hide the large cracks of reticulation, I would have to push the sliders to the point where the details of the image are too affected.

Digitization in Diffuse Light (Frosted Glass)

So, another solution must be found. That's when I remembered a discussion with Aurélien Pierre about his Diffusion or Sharpness module. If I remember correctly, he was very clear: even if the module provides useful results for reducing grain, the mathematical and physical principles of this module cannot produce diffusion exactly similar to that of diffuse light. So, perhaps we should try a digitization directly in diffuse light? For convenience, I just added a diffuser 1 cm above the condensers. I even tested three different diffusers: tracing paper, a plastic diffuser from a broken LED TV, and the frosted glass from an old 4x5 camera. And it was the latter that gave, by far, the best results.

And miraculously, there is no longer any trace of reticulation while the added blur remains relatively limited. Diffuse light like keeping secrets indeed  ;-). As for point light, it seems to be just a complication for maniacs who dream of complicating people's lives by revealing the finest details that could have remained discreetly hidden without bothering anyone... Nyah!

At this stage, for lack of a better option, this is a possible temporary solution.

Shall we take the plunge?

However, I couldn't forget a little idea I had during these trials, which I had dismissed as too "old school." Since reticulation seems to be the result of the contraction of gelatin that was too dehydrated by heat, would it be possible to reduce its effects by rehydrating it? Soaking the negative for 2 to 3 hours in a tray of water at room temperature with a few drops of wetting agent (PhotoFlo or even simple dish soap)? Then letting it dry in the open air?

This idea might seem contrary to the preservation requirements of originals to a photographer from the digital age who has never dipped a film or photographic paper into a developer bath. It seemed, on the contrary, a relevant and harmless attempt to the old dog  Of those who are said to be unable to learn new tricks ;-)I am, who has developed so many films of all types: B&W, color, negative, positive, Ektachrome E2-E3, and all formats from 135 to 20x25 cm sheet films.

The reticulation has completely disappeared! It's hardly believable.
In the larger formats, it is clearly evident that the result is better than anything that could have been reasonably hoped for.
The best solution is therefore in the good old lab techniques...

In Summary

To quickly compare the results of the different trials, here are 4 portions of a 40 x 60 cm (A2)

And finally, since I am making comparisons, I might as well go all the way and see if my efforts really make sense. Digitization with an Epson V600 scanner. Chosen resolution: 6400 dpi. As with the others: portion of a 40 x 60 cm (A2)

I am perplexed by this last frankly poor result. So I searched online to verify if I might have misused it. And I found this, which confirms my result but above all revolts me on the part of a major brand: Source: an excellent site available in French and English, but which unfortunately no longer seems to be updated. Very useful for information on old film scanner models. https://www.filmscanner.info/en/EpsonPerfectionV600Photo.html « According to our resolution table, the effective resolution is 1,560 dpi. This is less than a quarter of the 6,400 dpi resolution advertised by the manufacturer. » And this seems to have barely improved for the V850 with 2300 dpi. Therefore, it should only be used for negatives in medium formats and above. Or at most for 135 but not exceeding 13 x 18 cm.

⊕