I've deluded myself into thinking there might be a couple of people (besides myself) interested in seeing a larger work develop from start to finish, so here goes. Click the images for larger views.

Here is the photo reference for the painting, a view in the canyon along Catherine Creek, near my home. I selected a portion of the image for the design.

A 40 x 30" canvas was stretched and primed with two coats of Utrecht heavy body acrylic gesso, applied with a wide plastic spatula. I don't have a set process for beginning a painting. Sometimes I begin on the white surface directly with color. In this case I decided to begin with an allover tone. Using a 3" housepainters brush I scrubbed a thinned layer of transparent yellow iron oxide artist's acrylic over the entire canvas. When thoroughly dry, the design was loosely sketched and basic shapes blocked in very broadly, using burnt umber oil paint thinned with mineral spirits.

With a #12 flat natural bristle I started laying in color, working as broadly and directly as possible. I usually mix my paints with a small amount of Winsor and Newton Liquin, but with this painting I tried Gamblin's Galkyd Lite for the first time. It's too early to tell for sure whether I like it, but it did provide an interesting resistance, keeping the paint from feeling too "slippery" while still allowing brushstrokes to stand.

Further development, trying to strike a balance between refining areas and maintaining a painterly feel. I tend to use natural bristles most of the time, resisting at this stage going smaller than a #8 flat to keep from nit-picking too much.

After laying the groundwork of values and colors, I work with somewhat smaller brushes in selective areas, trying to suggest some of the complexity of limbs without detailed rendering.

The right side of the painting has been bothering me. I was uncertain from the start how to create a feeling of equilibrium, to balance the weight of the mass of trees in the upper left without it looking forced or obvious. In the photo the forested hills fall away to the right, so I felt the need to somehow "lift" that side back up visually. I tried to use more cottonwood trees to do this. Here I use a paper towel dampened with mineral spirits to lift color off, preparing the surface for a tree form.

Here I begin to drag lighter color, a mixture of white, ultramarine and manganese violet, over darker tones, starting to suggest reflections in the moving water. I try to keep a painterly feel to the strokes, not being overly reliant on the photo. Photos of moving water show it in a frozen instant of time. That is accurate, of course, but not really true to how our eyes percieve the motion in real time. I search with the brush to find an equivalent for this kind of perception.

The painting feels close to completion, but the right side still bothers me. It is maddening sometimes how blind I can be to the obvious. Suddenly the awkwardness of the two trees on the right is apparent, echoing too much the ones on the left.

The trees breaking the horizon line on the right edge seem to call too much attention to themselves. Once again paint is wiped away with the towel, back down to the ground color. I decide to depart from the photo and add a distant hill, gently sloping up. This seems to me to better provide that equilibrium I've been after.

Here's the painting in its present state. The foreground felt too dark, so I decided to lighten a portion of the large rock, suggesting a subtle sunlight dapple. This seemed to help. In the later stages of a painting the photo reference should become less important, as one responds to what the painting "needs." This painting probably needs a little more work, which I'll try to do without overdoing it (maybe it's already too late.) I think I'll study it awhile from my special chair.