Category: learn to paint

Posted on

Cathedral of St. John the Divine, New York

Since Monday was a day of heavy rain, painting outdoors without shelter was out of the question (eventually the paint starts sliding off the canvas). The Cathedral kindly gave me permission to paint inside. I know the Cathedral fairly well, and have always been enthralled by Dutch church interiors (for example, see here, here, and here.)

My husband asked me, “How does it compare to Canterbury Cathedral?” It doesn’t have the patina of a thousand years of continuous use, but in fact it compares pretty well. For example, the stone carvings near the high altar are sensitive and traditional, yet very fresh and lively.

A serious fire swept through the north transept of the Cathedral in 2001, damaging tapestries, organ pipes and stone work. Today, the chancel and choir have been thoroughly cleaned and the great nave (601 feet long) should be finished by this fall. Perhaps this is my only complaint—I wish that in a thousand years, a docent could point high above to the vaulting and say, “These marks are traces of the Great Fire of 2001,” as a memorial to New York’s annus horribilis. For in addition to the World Trade Center cataclysm that year, the second-most-deadly plane crash in American aviation occurred in Queens in November. Although the Cathedral fire was far less important than the other catastrophes, it was etched into stone laid for perpetuity.

The Cathedral is a continual ongoing project. Although the cornerstone was laid in 1892, work has progressed in fits and starts (dictated by finances and two world wars) and is at this time moribund. I am totally charmed by the cinderblock-and-corrugated iron sheds along the south wall, which were built as temporary structures. It takes no effort to see them as wattle-and-daub huts pressed against medieval cathedrals-in-progress.

The light in the Cathedral was very dim, with the nave closed off and the high chancel windows dark on such a dreary day. That made paint mixing difficult, since I was literally guessing at color. Many tourists stopped to visit while I was painting, and I never let on that I was not from Gotham—it would have spoiled their fun.

Posted on

How did I end up with more than fifty tubes of paint in my studio?

OK, it’s not that I’m a packrat exactly, but how long do you suppose this tube of paint has been kicking around? (For the record, there are also 58 tubes of watercolors…)

I have the terrible habit of buying paints without checking my inventory first. There are paints from my teen years, squatters left by former students, and orphan colors I bought but don’t like. There are also specialty paints, including a few metallics and zinc white.

But most of the tubes of paint in my studio are there because of my carelessness. That’s how I ended up with seven started tubes of titanium white, five different dark reds ranging from alizarin crimson to mars violet, several phthalo green mixes, and many other overlapping pigments.

There is a four-character colour index international (CII) code listed on every tube of good paint. Recognizing pigments from these codes is an important skill for the painter.

Just as Benjamin Moore uses names like “Yukon Sky” to peddle grey paint, art paints are often marketed with evocative names. Generally these names appeal to our sense of tradition, even when the old paint has no relationship to its namesake. For example, Indian Yellow was once made from the urine of cattle which had been fed mango leaves. Today it is made from lightfast diarylide yellow (PY83).

Other obsolete paints are approximated by blends. Naples Yellow started as lead antimoniate, but today is approximated by a blend of four pigments.

Then there are the modern synthetic organic pigments, which I enjoy tremendously. These were developed for industrial purposes and have no historical antecedents. They are great for their high chroma and clarity when tinted with white. The problem comes when they are used to mimic more expensive pigments. For example, I once bought a paint called “viridian” which was not genuine but a blend including a phthalo green. It looked like viridian coming out of the tube but stained like crazy.

When I was sorting today, I found three tubes of cerulean blue. One is Gamblin’s cerulean (PB35), which is “true” cerulean, made of oxides of cobalt & tin. This is a pricey paint but invaluable in the plein air paint-box. The second is cerulean blue hue, which is a much less expensive paint designed to mimic the color and opacity of PB35. It is a mix of zinc white and phthalo blue (PW4, PB15). The third was an off brand which I chucked before noting the contents.

There are places I can substitute the hue for the real thing, but why buy a mix to do it when I already own both the phthalo and white? A good general rule is to stick to single pigment paints whenever possible and mix your own colors. This gives you the greatest latitude.

There are great resources on the web to learn more about pigments. For oils, see Gambin Paint, here. For watercolors, see Bruce MacEvoy’s Handprint.com, here. (I am personally grateful to my friend Kristin Zimmermann for teaching me about CII pigment identification.)

Posted on

Rip-rap on the Lake Ontario Shore

I learned two important things today.

  1. When electronics (like your work camera) go missing, it’s wisest to start by looking in your teen’s bedroom;
  2. It costs $.25 per picture to upload photos from your cell phone.

This is my way of apologizing for not having “in progress” shots of this little sketch of rip-rap on the Lake Ontario shore. This was an extremely quick study, done in a few hours. The most memorable part was the surf rising and spraying my easel, my palette, and my feet.

These big rocks appear to be white marble and something else hard—gneiss? The prevailing stone here is Medina sandstone, which is soft and tints the soil pinkish. These big, hard white boulders look alien here. Although they are weathering beautifully, I hesitate to paint them in detail because they aren’t part of my essential Lake Ontario.

Posted on

From 99 Âş F to 65 Âş in two days

At Rochester, the shore of Lake Ontario is flat and covered with fine cobblestones. The shoreline is very even. It is hard to break up the strong horizontal and diagonal lines, except by putting in the dark overhanging trees.

What the lake lacks in architecture it makes up for with incredible light and color. On a windy day, the water shifts from violet to emerald as cloud shadows fly across its surface.

It was 65 º F when we got there with steadily strengthening winds. By noon, the wind was so strong, there were whitecaps on the lake and my hat had blown away. Of course, this fellow was happy…


I start with a crude sketch in Transparent Earth Orange (Gamblin’s transparent version of Burnt Sienna). My first pass is a very static composition, since I’ve divided the canvas into three equal spaces with two well-balanced lumps in the foreground.


I move the horizon line up gradually. I’d wanted clouds racing across the sky but realize you can’t have it all. I hope to break the bottom diagonal with little hummocks of plants along the shore.


The horizon moves even higher and only the clouds distant over Ontario remain. In truth, the only darks are in the foliage on the shore but I don’t want to weight the bottom of my painting so much. The shadow color on the water ranges from tints of ultramarine to quinacridone violet, depending on when you look. The green ranges from yellowish to emerald green.

My first pass is mainly to establish darks. Often the highest chroma in Lake Ontario is at the horizon line and the color of the water becomes less saturated the nearer you are to it. (This is the opposite of most long views, where the color becomes lower key the farther away you look.)


Next I establish an overall color scheme. I like this little sketch at this point, but I am concentrating so much on the water lighting that I don’t notice I’m “regularizing” the shapes in the foreground. The mind wants so much to balance things, but that same symmetry will dissatisfy me farther on. (The lump on the left is a young box elder and the wind at this point was bending it nearly double.)


I need to set the diagonals of the breakers, which appear to change angle as you scan the shore because they are rolling in from the west (over my left shoulder). Although the angle changes, the waves break at about the same distance from the shore no matter what direction you are looking.


I begin to consider the interstices between the breakers and develop the foliage in the front. Unfortunately, my painting pal has to leave, so we call it a day.


My biggest issue with this painting is to make the foreground shapes more interesting. I also want to refine the waves. But again, I want to do this on location, rather than in studio.

Posted on

Painting on top of a ridge

This study is about half finished. It was done on a ridge above Naples, NY (in the Finger Lakes region). Since I don’t want to work from photos, I will go back next week.

The thermometer read 99Âş F; later we found out that was as high as it would go. The photo was taken at the end of our painting session and the light was then much bluer. The mountains were a more complex color earlier in the day.

My goals for next week are to refine the drawing/painting of the tree line and the fields in the foreground. There is an unfortunate confluence of line where the tree meets the valley—I need to resolve that too. My painting is a little cramped horizontally, but I won’t try to fix that on this draft; instead I’ll paint it again.

Here is the view from the top of the ridge.

Posted on

One more time with feeling–Tilt-a-Whirl


My last attempt at blogging was so difficult it scared me away. But here I am, back for more.

This was done last Saturday at Seabreeze Amusement Park in Rochester. It was painted from about 30 feet away. The figures are imaginary because the cars were whirling too fast to see people. I loved the odd pulsating thing on the right which rises up and down like a sea monster. The canvas and lighted bulb actually conceal the mechanical arms of the Tilt-a-Whirl.

The best I could hope for in the chaos of an amusement park was a straight rendering of what was happening. I finished at dusk so was able to suggest the lights coming on.

I realize after the fact that most people hold themselves very rigid as the Tilt-a-Whirl spins. My figures are too relaxed.

Posted on

Painting a small study using an indirect method

Note: I first posted this in April but managed to lose all the illustrations. Embarrassing for someone who prides herself on her computer literacy. I recently realized I was overworking the thing, and ought to just follow the software instead of trying to bend it to my will.

During Wednesday’s class I demonstrated a method of indirect painting where one starts with a monochromatic grisaille and then paints into it. I had so much fun with the demo that I decided to finish it, working with the monochromatic underpainting while still wet. (In true indirect painting, the grisaille is allowed to dry completely and then color is glazed into it in transparent layers. What I am doing is more properly called scumbling.)

I am working on a 5X7 RayMar linen art board (so these images are actually quite small). I start with a mix of two earth pigments, Burnt Sienna and Burnt Umber, thinned with Turpenoid.

The advantages of cartooning in paint instead of drawing are numerous: you can wipe out what you don’t like, you’re inclined to work in masses instead of lines, your sketch is temperamentally closer to the final painting, and you concentrate more on your painting than your drawing skills. Rubens drew in paint, as did Rembrandt.

Those who were in Wednesday’s class know that my first sketch had two figures. But when I returned to it on Thursday, I could no longer remember where I was going with it, so I wiped it out. I decided (for no particular reason) to paint “Ruth amid the corn” for my subject.

While the first drawing was a sensitive rendering, it is a static composition, completely in profile. So I wiped it out and redrew it. The second draft is a rather brutish woman, but I liked the pose.

At this point I began to consider what I wished to say about Ruth. In the Bible story, she is starving, but remains beautiful enough to attract Boaz. She is an alien, but her story culminates in her being the great-grandmother of King David. She is both a poor woman and the ancestor of a King. The other women working in the field are separated from Ruth but aware of her. (The Book of Ruth can be read here.) I developed this stage by adding layers of paint and wiping it away where necessary. These layers were not completely dry as I was working over them, in contrast to a true grisaille, but I was able to add additional pigment without disturbing the lower layers. The goal at this point was to clearly develop the values (darks and lights) and begin to add texture. As I was not working with any reference (either a model or photos), the figure and face are unfortunately somewhat stylized.

I chose a transparent palette with mainly 20th century pigments: chromatic black, phthalo emerald, hansa yellow, Indian yellow, napthol red, quinacridone violet, ultramarine and phthalo blue. I chose this palette based on personal preference, because it certainly isn’t historically correct. Since the 20th century pigments were developed for direct painting, the end result is kind of quirky.

I glazed Ruth’s dress in blue as befits the ancester of a King (see representations of the Virgin). When the color was added, I noted a number of drafting errors—in the arm, in the basket, and in the placement of the eye. I could have corrected these in the transparency but didn’t feel like scrubbing them out on such a small canvas. Instead, I will correct them with opaque paint.

At this point I added two whites to my palette: zinc (for glazing) and titanium (for opaques). Zinc white tends to be warmer and more brittle than titanium white, which is extremely cool. It is important to begin using medium at this stage.


I began to experience the limitations of my brushes. I have been direct painting for so long that I no longer have sable or synthetic brushes. The hog bristle brushes I have are adequate but leave brushmarks unless you overwork the surface. I also have no tiny brushes, and I am working on a very small surface. Oh, well.


As I began adding white, I was also modeling chromatically. By underpainting in a warm tone, I have set up the painting so the light areas must be cool and the shadows warm. In these blues, I used a combination of violet and phthalo in the highlights, phthalo blue and emerald in the shadows, and ultramarine in the midtones.

I became aware of an annoying composition problem in the lower right corner, where the leg slices off a triangle. I could solve this by reducing the contrast and chroma in that corner, or by changing my drawing. I don’t like the bare leg that much anyway; it might contribute to a sense of motion, but is nonsensical for Biblical-era women’s dress.

I began to add the white to the flesh tones, which only emphasized the difficulties in the lower right corner. Note that scumbling the transparent (zinc) white over the wet transparent earth tones in the blouse results in warm and interesting shadows.

I realized I need to add some opaque colors in order to develop the skin tones and background. I choose yellow ochre, chromium oxide green, and raw sienna (modern earth pigments are more opaque than the historic ones). It’s a little picture, so I put out small amounts of paint.


The flesh is rather flat (I’m blaming my brushes), and my drafting errors have become more obvious. Note that I don’t strive for transparency in the skin tones at all. Rembrandt had a delightful way of painting solid faces into figures which were essentially transparencies (here, for example). I love it, even if I am not particularly good at it.

I decided to reduce the size of the figures in the background, and integrated a pale sky into the painting. Note that I was constantly refining the figure, the draperies, and the face with every iteration. In addition, I was reducing the contrast in the blue drape.

It was time to rid myself of that pesky leg. (I could have fixed the drafting, but that wasn’t the leg’s primary problem; its position was the issue.) Note that I used chromatic modeling on the white skirt even though the russet tones weren’t pulled up from the underpainting. Also, I was steadily reducing the contrast in the blue drapes while increasing the chromatic range.

One issue with using 20th century pigments was becoming apparent. Many of them (the phthalos in particular) are high-stain colors. With little provocation they will bleed or track into neighboring sections of the painting. The painter must take extra care to manage this.


But—gack!—that eye! I’ve put up with it in the wrong place long enough. I had great fun introducing a stone wall and ruffles on the white skirt. It’s become somewhat more opaque than I intended but there are still passages which are transparent. Have developed the background figures as far as I want to… I think.

In that I have to teach tomorrow, I will let it dry thoroughly and then look at it again. There are shadows which need resetting, but perhaps I will take my friend Toby’s advice and work this up as a larger painting.

Just for fun, here are some other renderings of Ruth.

Marc Chagall, The Meeting of Ruth and Boaz

William Blake, Naomi entreating Ruth and Orpah

Julius Schnorr von Carolsfeld, Ruth im Feld des Boaz