Underpainting for ‘Calvados Bottle, Driftwood, and Sandstone’

Today I began the underpainting. As usual, I mixed up 9 shades of lead white and burnt sienna and numbered them right on the palette for easy reference.

For the underpainting, I work very loosely, while trying to preserve the drawing. I keep everything much lighter in value than it will be in the finished painting because the colors on top look more vibrant with a lighter ground beneath them. Also, if I need to change anything later on, it’s easier to correct on top of a lighter layer. If the under-layer were dark, it might show through the correction. I keep the paint layer very thin because it dries more quickly. Brush strokes are kept to a minimum so that they won’t interfere with the top layer. Any brush-strokes I make this early in the game are bound to be ill-considered, and I don’t want them showing through in the top layers of paint. An exception to this rule is the handling of areas that will be rough-textured, such as the sandstone. These actually do benefit from some roughness showing through, so I don’t bother to smooth them too much.

Another, less- obvious benefit of doing an under-painting is that it breaks the ice. It can be daunting applying finished paint to a white canvas with all of the pressure of having it be right. An under-painting is so simple and un-detailed that it’s easy to approach without fear. Small steps are best!

The drawing that I transferred was a bit light and hard to see in spots, so I sketched it in again right on the canvas. Even though I’d been so very careful to be accurate while I was tracing the drawing onto the tracing paper, and then onto the canvas, I found that many of the lines were incorrect! It reminds me of that childhood game in which everyone sits in a circle and whispers a short story into the ear of the person sitting next to them, on around the circle. By the time the story goes full-circle, it’s unrecognizable! The very act of copying introduces error. Every line drawn must be the result of checking with reality, not the second-hand tracing.

I haven’t painted the shadow cast onto the wall. Later, I will paint the entire wall as though it had no shadow. When that dries, I’ll glaze the shadow over. This nicely mimics the look of an actual shadow with the color and texture of the wall showing through the transparent glaze. The only problem with this method is that I lose the outlines of the shadows that I observed and drew so carefully. For the complicated shadows that I feel I’ll need some guidance with, I indicate them lightly. I will still glaze these, and they will still look transparent. This works especially well with the darker shadows, like the ones on the table-top.

This is as far as I’ll take the under-painting. I’ll let it dry for at least a week before I start painting.

A New Set-up

While waiting for the canvas to arrive for my sandstone painting, I decided to set up another still life. I thought I’d do a smaller one that wouldn’t take as long to finish as my big works. I love painting books and glass, so looked through my cupboards and shelves and picked out some promising specimens.

When I got everything back to the studio, this rough wood crate caught my eye. I loved the way the texture contrasted to the glass vases and glasses. It also echoed the roughness of the worn edges of the old books. Above is my first attempt. Looking at it through my view-finder, I remembered why this crate was always hard to place in compositions. The strong right vertical of the crate always seems to divide the picture in half. This is a very hard barrier for the eye to cross, so the picture reads as two separate sections, not one unified whole. I tried to think of a way to soften that edge and guide the eye across.

After trying placing a glass in front of the edge, I remembered this ribbon I had used in another painting. I draped it an angle trying to lead the eye across the troublesome barrier. It was an improvement! I still didn’t like the strong vertical so close to the middle of the composition, though.

I changed to a more square view-finder, which placed the vertical much further over to the right and eliminated the blue book and vase. I immediately loved this. The ribbon leads the eye through the composition. The crate is an important part of the whole, not just one half. The angles of the books and ribbon add drama and dynamism. I was worried that the ribbon would lead the eye straight out of the bottom right of the picture, so I added the blue stone to turn the eye back up. I replaced the shot glass on top of the crate with a stone. The glass was too big and shiny and was drawing attention away from the more important objects. I also didn’t like the strongly lit left side of the crate. Fortunately, the cupboard where I set this up has a door on the left. When I opened this door a bit, it cast a shadow on the vertical, darkening it. Also, by eliminating the right side of the composition, I’ll now have a small painting- just what I wanted. I might prefer the yellow crystal turned the way it was in the previous photo. I’ll experiment with it a bit more before I start the drawing.

Finishing the Drawing

I’m happy with my study, and don’t need to move anything around on the drawing. Now I can go back and finalize the ellipses and correct the perspective.

Above, you can see all of the ellipses I drew and then transferred onto the drawing. You can see the construction lines and my notes on the right. 3@15, for example, means that the length of the ellipse (the major axis) is 3″ and it is 15 degrees below my eye level. To see more on how I construct an ellipse, see Portfolio and Jewelry: Drawing.

Above, I’ve transferred the ellipses to the drawing, using tracing paper rubbed on the back with graphite. Sometimes, even after all of the measuring, the ellipse looks wrong to me! I always believe my eyes, not the measurements. I adjusted the ellipse at the bottom of the bottle. It looked too shallow, so I deepened it. You can see both lines. I marked the shallow one with a tic-mark to remind me. It was easier than trying to erase.

I checked that the bottle was symmetrical by tracing the side I was most confident in. I included the vertical center line in my tracing. I then flipped the tracing over, lined up the center line, and traced the original line onto the other side. Amazingly, both lines matched up.

I took a little time indicating where the text would go on the label. This was probably wasted effort, as I like to paint in the whole label, getting the value transitions and colors right before painting the text. Because of this, all of my drawing would be covered up with opaque paint! Oh, well!

My canvas should arrive this week from my canvas guy, and I can finally begin to paint!

Finalizing the Value Study

My value study looked a bit bland to me today, so I decided to compare it to the photo I took of the set-up. I immediately noticed that there were more dramatic darks in the photo. I think that I composed the set-up and photographed it on a cloudy day, but painted the study on a sunny day. Being me, when painting the study, I had to paint all of the light that I saw in the set-up. At the time, I suspected that the darks weren’t dark enough. However, it’s much easier for me to begin with what I see. It’s not too hard to adjust later when all of the values are in.

Above, I’ve darkened most of the shadows. I’m much happier with the composition now. I’ll have to remember to work on the shadows on cloudy days or closer to evening. One of these days, I’ll have to replace the window coverings in my studio, so that I can control the light better!

Value Study

Before I finalized the ellipses and perspective, I wanted to paint my value study to make sure that I was happy with the composition. Sometimes it’s easier to judge a composition without the distraction that color can bring. The large areas of darks and lights must be harmonious and interesting, and serve to guide the viewer’s eye to the focal point. Details are unimportant at this point. I aim to work quickly and just roughly indicate the big shapes.

Above is the result of my first session. I’ll let this dry for a few days. If I tried to continue now, the different values of paint would smear together, and it’d be hard to achieve crisp lights and darks. Also, my first guess at the correct values is usually wrong. I need to see some paint covering the whole study, so that I can compare adjacent areas. Painting is all about comparing!

Above is the study after the second session. I adjusted values, corrected the drawing, and added some details. Ill live with this for a few days, and see if I’m satisfied with the composition.


I wanted the painting to be slightly smaller than life size. To achieve this, I measured the horizontal distance of my set-up and subtracted a bit. Since the view-finder I used to compose the painting was in a ratio of (2-to-3), I could then calculate the height. I drew a rectangle of that size on my drawing paper. I sub-divided the rectangle into halves, quarters, and thirds to correspond to the tic-marks on my view-finder. Using a thin knitting needle and the view-finder, I quickly sketched in the bottle, so that I could judge if it was the correct size. I found it to be a bit larger than life. I reduced the size of my rectangle, and tried again. Now, when I sketched the bottle, it measured a bit smaller than life-perfect!

Now I proceeded with the drawing. Using the view-finder and knitting needle again, I located major points (where objects begin and end) on my paper. I find this much more effective than just guessing where all of the lines should be. I tend to draw things sight-sized- that is, the actual size they appear to my eyes. If I did this, the objects in the drawing would be way too small. Having some guidance helps me to draw larger, and get everything placed properly within my picture frame. For more on measuring techniques, see Getting Ready to Draw and Drawing Again.

Once I have determined where objects begin and end using my measuring method, I begin to draw simply by observation. I’ve discovered that if I stay in measuring mode for too long, errors are likely to creep in, because of the natural shaking of my hand while holding the view-finder. My measuring will tell me that one thing is correct, but when I stand back and observe, I can see that it’s not right. It seems like a conflict, but I know by now that my eye doesn’t lie! No measuring can replace the quick observing and comparing I do when sketching.

Above, I’ve put in some more detail on the driftwood, geode, and crystal.

After leaving the drawing for a day, I can return to it and see it with fresh eyes. I like to take advantage of these first few moments of fresh observation to spot errors. They really jump out! I’ve learned that time needs to pass for this to work. The longer I look at my drawing and work on it, the more I accept whatever is there as correct. It’s amazing what errors you accept by the end of the drawing session! Now I noticed that some of the shapes in the driftwood were off, and that the geode was placed too high. Also, the far right edge of the sandstone was drawn incorrectly.

I’ll postpone perfecting the ellipses and perspective until I’ve done my value study. I don’t want to spend time drawing these details now, in case I decide to change things around as a result of my study.

An Eye Level Problem

The first photo below is the set-up I liked the most. The second photo is the same set-up as I began to draw. Can you see the difference (aside from a slight color variance, due to the exposure)?

The second view is from a much higher vantage point- from my actual eye level, actually. It turns out that when I took the first photo, I held my camera at chest height to eliminate keystoning (which happens when the sides of a photo angle in due to parallax). This, of course, lowered the vantage point. When I returned to start my drawing, and looked at it from my usual eye level, everything looked wrong! I was viewing the set-up from too high. I could, of course, paint it this way, but I much prefer the look of a lower horizon line. I don’t like the viewer to feel as if they’re above the table looking down. I remember now that I had a similar problem with my bricks painting. I had to lower the horizon line and redraw everything from a lower level.

I tried lowering my stool, but it wouldn’t go low enough. So, after marking everything’s position with a pencil on the tabletop, I removed the set-up and set about raising the table.

Fortunately, bricks are useful not only as the subject of a painting, but also as structural material. I have this handy pile behind my studio, which I find useful for both purposes.

They worked nicely. I also had to raise the spotlight an equivalent amount. All looks in order now. Next session, I’ll figure out how big the painting should be so I can order my canvas.

Second Attempt

Above is the composition I worked on last. I lived with it for a while, and though I liked it, I thought that I’d try another.

I left the bottle where it was, and moved everything else around, just as an experiment. I also wondered how the bottle would look with its label showing. The result is above. The label seems distracting. Also, the wood bowl so close to the middle draws too much attention. Lastly, the pinecone points out of the picture- never a good thing!

I put the pinecone and the bowl back to their original positions, and experimented with moving the driftwood back into the shadow of the bottle. I removed the small brown bowl and added the orange geode and white shell. I turned the label away from the light. I like the label more, but this arrangement seemed fragmented- as if it were pulling the eye in two different directions. I thought I’d try removing the pinecone to shift the focus more to the center of the composition.

This seemed better, but something didn’t seem right. I tried looking at the image upside down. This is a great method for seeing the composition more clearly, since the eye doesn’t focus on the separate objects so much as on their abstract shapes and relationships.

When viewed upside-down, I noticed that I didn’t like the shape of the orange bowl and orange geode. They make a pronounced horizontal and vertical that didn’t seem harmonious with the rest of the composition. Also their brightness both in chroma and lightness in value drew the eye out of the frame.

Above, I eliminated the bowl and re-positioned the orange geode. This is better, but the bottom of the composition seemed rather dull. I thought that I’d try adding the small brown bowl back, but further forward, overlapping the edge of the tabletop.

This seems promising. I like the way that the upward angle made by the top left side of the sandstone is echoed in the line of the fool’s gold, green geode and the lit top of the driftwood. This angle is also echoed in the line of the brown bowl and orange geode. Repeating angles can help to unify a composition. I feel like maybe it could use some more light values and brighter colors.

Here’s the original composition. I don’t know which I prefer. I’d better give it a few days!

A New Set-up

When it’s time to begin a new painting, often I wander around my studio and house, looking for something to paint, feeling like I’ll never have another idea again. Other times, I see some object and I think “That’ll be the start of a new painting!” Last week I was sitting in my herb garden drinking my coffee and looked down and saw this piece of sandstone lying on top of the bricks of the patio. I’m not sure why it was there, or why I hadn’t noticed it before! Immediately, I thought that I wanted to paint it. I liked the shape, and the textures were unusual and beautiful. I set it it the area that I had used for my last Japanese basket painting, which was still set up in my studio. I could use it as a tabletop or as a vertical backdrop. I thought that it would be more visible as a vertical.

I liked the way it looked with the rice paper behind it, so I left that up. Also, the colors were harmonious with the brown board I had used as the tabletop, so that stayed too. Glass would be a nice contrast with the rough stone, so I selected a green bottle, whose color also complemented the stone. Two objects do not a still life make, though!

I gathered some other objects in subdued colors. This set-up didn’t call for any bright colors. The bottle was more visible on the right, so I moved it. It cast an interesting shadow. I added the glass vase. Though it looked nice, I thought that it was competing with the bottle.

I moved the dish with the pinecone to the left and added the driftwood for some more textural interest. The whole set-up was looking very dark. It’s usually a good idea to have a full range of values from white to black in a composition, so I put in the white stone and shell.

To add even more light tones, I took out the yellow crystal and substituted the geode with the white center. I arranged all of the objects in the front in a shallow curve bending downwards.

The curve drew too much attention away from the top of the composition, so I re-arranged the objects in front in more of an upward curve. I substituted some fool’s gold for the white stone. I didn’t like that the geode was almost dead center. That can be very distracting in a composition. I switched it with the brown dish. I prefer this.

I wasn’t fully satisfied with the driftwood. I rotated it to find a more pleasing angle. I also raised the geode so that it intersected the sandstone.

Above, I tried yet another angle with the driftwood.

And another! I think that maybe in this position, the driftwood attracts the eye too much. I prefer the version two photos up.

I’ll let it sit for a while before I decide.

Fine-Tuning the Composition

Since I was almost finished with the painting, I decided to stand back and see if the composition was working as I had intended. I wanted the eye to start at the basket on the left, then follow a curve through the blue paperweight, the turquoise, up to the top of the black bottle, and then over to the small blue crystal on the left and out. I was mostly satisfied, but thought that the trip up from the turquoise to the top of the bottle could be stronger.

I made many subtle adjustments to achieve this. First, I strengthened the vertical highlight on the right side of the vase to lead the eye up. For the same reason, I then brightened the reflected light on the left side of the vase (though this is hard to see in the photo). Also hard to see, I made the highlights at the top rim of the vase stronger because bright lights against a dark background draw the eye. I added some more details to the front surface and edges of the brick behind the vase (smudges and tiny nicks). Detail also draws the eye. I lightened the bit of brick showing to the right of the neck of the vase to contrast more with the vase. I lightened the third brick from the left so that it contrasted more with the brick behind the vase. Contrast in value draws the eye. I thought that all of these details would attract attention and lead upward. The brightness of the tabletop was directing the eye down, so I painted in the woodgrain. This both darkened it and added some horizontal movement that keeps the eye from exiting through the bottom. Lastly, I lightened the brick on the far right, adding more highlights on its top.

I’m pretty happy with it now!