Egg Tempera Forums

Egg Tempera Forums (
-   The Forum for Tempera Painting Issues (
-   -   Question Underpainting layer question (

Kathy 01-08-17 02:36 AM

Underpainting layer question
Hi all,

I'm experimenting with a new way to do an underpainting and wanted to see what you thought about it. I've been using my pigment paste (mars brown) with just water to do a tonal underpainting, as I like the way I can move the pigment around, and lift it off as needed till I'm happy with the drawing. When it is dry I put a coat of diluted egg yolk over the drawing to "fix" it. Then I proceed to paint with egg tempera as usual. Can anyone see a problem with that method? Will my underpainting layer take away from the adhesion since it has no egg in it? Or will the egg layer combine with the past layer to make a good bond. Thank you for your expertise!

arbrador 02-08-17 05:33 AM

Hi Kathy,

Very interesting idea! I'll be interested to hear what Koo Schadler has to say about your technique but in a way it doesn't sound too different than using homemade pigmented transfer paper which also has no binder.

In terms of locking it in with an egg wash my only thought is you may want to follow Koo's recommended ratio of 1 part water to 8 parts egg yolk so it's not too concentrated nor too dilute.

I like your idea so much that I may try it soon and will report back to ET headquarters!

I'm someone who, although I make numerous preparatory drawings and watercolor skethches, I end up composing quite a bit on the gesso panel. I use vine charcoal and erase with a kneaded eraser. What I do after that is strengthen the lines with regular ET paint. So you could just paint over your underdrawing and skip the nourishing layer.

I'll be interested in hearing what others have to say.

Thanks for a unique idea!

Lora Arbrador

Salamander 02-08-17 12:11 PM

Hi, I do an under painting with water colors then "fix"with a tempera wash.

Koo Schadler 02-08-17 02:48 PM

Hmmmm.....I have to think about it, but at the moment I don't see anything wrong with this approach. I agree with Lora,it's not much different from drawing on a panel with charcoal or applying lines via transfer paper (the Renaissance working method may include a charcoal underdrawing fixed by redrawing with india ink on top). I also like Salamander's idea of developing an image with watercolor than moving onto ET. But your idea should work too. Here are a few caveats:

1. A very heavy, covering layer of pigment could create a distinct, unbound layer between the support and tempera layers, and this could cause upper layers to lift or flake - but this sounds unlikely, as it'd have to be a substantial layer of pigment, which isn't what you are doing.

2. I agree with Lora about being attentive to the water to egg ratio in the fixative layer. There is a phenomena (discussed previously on this site) called Fatty Acid Migration (FAM) or Efflorescence. Excess lipids (or fats, which are a part of the yolk binder), that are not needed by the paint film are "expelled" and eventually travel up through paint layers to settle on the surface. (This can also happen in oil paintings, but less commonly). FAM is still not fully understood, continues to be studied. It can be exacerbated by the size of a painting (seemingly more common in large works) and changes in climate (moving a painting between varying percentages of humidity). The expelled lipids appear as a sort of crystalline, whitish fuzz on the painting's surface. FAM can occur anytime, from a painting's completion to decades later. It's not necessarily a problem in that the fuzz can be gently wiped off the surface without harming the image. Still, it's not desirable because it changes the appearance of a painting, and you don't necessarily know where a work is going to be in a few years or decades, and how the the fuzz will be treated.

(Aside: Many of Andrew Wyeth's paintings exhibit ongoing efflorescence, perhaps more so in paintings he moved between Chadd's Ford and Maine. There is a fun story about Andrew Wyeth directing conservators to remove FAM from all his paintings except for snow scenes, where he felt the efflorescence contributes to a wintery effect :-)).

Enough of a digression - suffice to say you don't want to inject excess lipids into a painting.

3. Even though the "fat over lean" principle isn't as critical in egg tempera as oil (because ET layers are so thin), an extreme of lean over fat (as in a very fatty underpainting followed by subsequent lean layers) can cause cracking; in fact, such cracks can appear almost immediately, as well as in the future. So keep the fixative layer of yolk on the lean side. Yolks vary slightly in richness depending on the hen's diet and health; with experience, you can feel the richness of a yolk. So, depending on the egg, I'd shoot for a 1:6, 1:8 or even higher ratio of yolk to water. The less yolk needed to fix the image, the better.

One other option would be to spray a very dilute mix of shellac over a pigment underdrawing. Many a Renaissance tempera painting was done atop an india ink underdrawing.


Georgeoh 02-08-17 05:18 PM

Good points, Koo.

The cracking you mentioned in your reply (Item 3), I believe should be attributed to the binder-to-pigment ratio rather than to a fat-over-lean principle in egg tempera. Although these two concepts are related, because they depict the pigment volume concentration (PVC) in paint, they present themselves differently in egg tempera and oil painting. The fat-over-lean issue is about the reducing the PVC in successive layers of oil painting, where each pigment particle is enveloped by the binder (oil). Egg tempera is a high PVC paint, similar in that nature to watercolor, distemper, etc., hence each pigment particle is not enveloped by the binder (egg yolk), but mostly by air voids and protein glue bridges of the egg yolk.

We know that when the glue to pigment ratio of a traditional chalk or gesso ground exceeds a certain amount, the ground cracks as dries. This is largely due to the stress caused by the protein molecules shrinking as water evaporates. The same may be true for egg tempera as it dries when the binder to pigment ratio exceeds a certain amount.

Just another idea for you to mull over.

MBergt 03-08-17 03:53 PM

Great responses all, and I had the same initial impression as Lora—the raw pigment is not that different than a transfer paper using raw pigment. However, that transfer paper would probably be a very thin layer. I remember reading years ago, that egg temper binds very well to itself, but it's critical that the first layer binds well with the ground. That's my only question, the reason we need to paint on an absorbent ground is we need that first layer to adhere well, then the successive layers of ET will too. I've also had questions about laying in an initial layer of a glue wash before starting to paint. Is that advisable, or is it better to just begin painting on the gesso?


Georgeoh 03-08-17 04:42 PM

Paint is an adhesive and good adhesion is not based on mechanical adhesion of porous substrates. In fact adhesion of paint layers is based on dispersive adhesion, which is due to the forces of attraction between charges or dipole moments; different charges attract one another. Absorbent grounds can increase adhesive strength, but it is ultimately more dependent on dispersive forces.

In my opinion, an absorbent ground is needed for egg tempera primarily to allow some lipids to sink in otherwise this may lead to bloom and other issues related to lipids. (Koo may have a better idea on this, too.)

As long as there is good wetting of the ground by the egg tempera, it should not be necessary to apply a layer of glue or egg wash on the ground first.

For a more detailed discussion on adhesion, please visit this link:

Koo Schadler 03-08-17 05:31 PM

It's probably unwise to disagree with a genuine paint chemist (George) when I didn't even finish high school chemistry, but I'll give it a try. George has explained dispersive adhesion to me before (in his excellent "Painting Best Practices" workshop, which I've taken twice - it's that informative) and I agree with its importance. Still, as someone who's worked long term with egg tempera on traditional gesso, I feel the mechanical bond between the two is more than secondary, it is very important (and a lack of mechanical adhesion is why, in part, egg tempera is more apt to lift and flake on less absorbent surfaces).

In addition to the adherence question, there are two other important reasons to use an absorbent ground. One, as George says, is to take up excess lipids (which otherwise might migrate to the surface and lead to FAM, previously discussed). Two is to control the working properties of the paint. An absorbent surface absorbs tempera's solvent (water) more readily; this creates more controlled mark making and decreases lifting, which allows for the quick and endless layering that is part of egg tempera's magic.

Michael makes a good point - adhesion of the initial layer is important. When I've seen chipping or flaking in tempera paintings, it's always been all the way down to the gesso - that's where the delamination apparently takes place. So I agree, keep the initial layer of unbound pigment thin.

Just to clarify (so no one is confused), the issue of excess lipids showing up on a painting's surface is called Fatty Acid Migration or Efflorescence. It may also be referred to as bloom - and it is a sort of bloom. However bloom is also used for a different phenomenon: a cloudy appearance under a varnish caused by moisture or mold. This sort of "bloom" is distinct from FAM. So to keep the two phenomena distinct, I use the term FAM for excess lipids, bloom for what can happen under a varnish. Am I missing anything with this distinction, George?

Thanks for the clarification, George, regarding cracking due to excess binder - I agree, I've seen both gesso and egg tempera crack due to excess binder. Both things tend to happen as soon as the paint dries. Tempera's PVC isn't as changeable as oil's, and tempera paint layers are very thin - still, would you say it's possible, if someone is not getting the correct ratio of egg to pigment and is over tempering the initial layers, then under tempering upper layers, to have a lean over fat problem akin to what happens in oil? Or is it just not possible due to how yolk cures?


Georgeoh 03-08-17 06:09 PM

The reason I believe the fat-over-lean principle does not apply to egg tempera is because egg yolk does not envelop pigment particles as does oil in oil paint films, and this is critical to the properties of the paint film. In other words, the change of film properties due to variations of the pigment ratio around the critical pigment volume concentration (CPVC) does not apply to egg tempera due to its being a high pigment volume concentration paint.

It is also related to the point you inferred which is how egg tempera dries, because this is somewhat true of acrylic paints and other thermoplastic film forming paints.

Not to beat this point into the ground, but the reason I say that mechanical adhesion plays a secondary role in the adhesion (not in other factors) of paint is that it possible to degrade what might otherwise be good mechanical adhesion when the paint exhibits poor wetting and there is poor dispersive adhesion. In the Painting Best Practices workshop, I use the example of water beading up on textile after a water-repellant substance is applied to the surface of the textile. The textile remains absorbent but the water does not "adhere" to the surface of the fibers due to poor wetting.

Koo Schadler 03-08-17 06:28 PM

Got it, George - thanks for all of those clarifications. Very helpful to have a scientist chime in.


Kathy 03-08-17 06:28 PM

Thanks all for your impressive expertise, this forum is so great! I'm attaching a jpg of my underpainting (hope it's not too big). Hopefully it's thin enough. I like all the other ideas of underpainting methods too, I'll have to try them, I'm not as good on the tracing method.

Koo, I'm very interested in the info on efflourescence, since I'm having a problem with that in my oils. I store them in my garage studio which is far from climate controlled, can be anywhere from under 10 degrees to over 80. but it's my only storage option. I get white crystals, especially in the darks, and sometimes they will keep coming back. I'm getting the impression it might be a phenomenon of modern materials, at least in oils. If you know of any of your contacts who I could ask about it as regards to oils, I'd be interested.

What's interesting about this piece I've started is that I'm using a gessoed panel prepared in 1939 by magic realist painter Louise Mariennetti (1916-2009) who prepared several panels but never used them. I was lucky enough to be given them, so this painting better do justice to the panel! It's a bit rough but I'm loving the feel of the surface, I assume it's "old school" gesso, rabbit skin glue and white lead...

Kathy 03-08-17 06:31 PM

Sorry, I guess that did display too big! Will remember in the future!

MBergt 03-08-17 08:05 PM

I'm thankful for this exchange, because I've learned a lot. It seems to me that laying in an initial egg wash might in some way contribute to the migration of lipids, since you're essentially adding egg before you begin painting??? Also, if one of the roles of the gesso is to absorb lipids, then anything one does to decrease that ability (like adding a layer of shellac or fixative) might unwittingly contribute to the formation of bloom. Good to know that fat over lean doesn't really apply to ET as it does Oil paint, and that the paint bonding in ET is related to more than just a mechanical bond. Even though we all know a slick surface isn't the best for adhesion, and egg rich paint layers result in a "waxy" paint that's not very durable.

Georgeoh 03-08-17 08:42 PM

Whether an initial egg wash contributes to fatty acid migration and resulting issues (i.e., haze or efflorescence) would depend on factors such as the absorbency of the ground and total amount of egg yolk in the layers of paint.

All times are GMT. The time now is 01:44 PM.

Powered by vBulletin® Version 3.8.6
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.