Butler's Methodology

“The Eclipse occurred within a second of the time astronomers had calculated for it, and I was told I would have just 112 seconds in which to make the picture. I kept hoping that their calculations might be wrong and that it would take longer, because under ordinary circumstances, the kind of picture I wanted to paint, would have taken from ten to twelve sittings. I knew I should have to devise some swift system, therefore, and the first one I hit upon was the muffin-pan method. By this, I planned to have half-a-dozen panels the color the sky was to be, with dishes filled with the corona colors besides me, and to grab the colors and paint the corona as best I could. But the more I practiced with this plan the less I liked it, and finally I gave it up entirely for an old method I had used when painting clouds in landscapes before.

This was a system based on the color values of pigments. The value of a color may be defined as the light and shade in a color. Nature has a full scale of 100 values ranging from the deepest dark to the brightest light and every pigment has its own value, but with this difference that the rose color, for instance, like that in the corona which you get in pigment, will be much farther down in the scale of values. To get the rose color in pigment which would represent that of the corona, I experimented with the rose in the in the spectrum and finally arrived as near to that as possible.

My plan for the actual drawing was to sketch in the outline of the eclipse rapidly, jot down the color value present in the sky, moon, clouds, corona, hydrogen flames; each according to its numeric value in the scale, and after the eclipse was over fill them in at leisure. I did not bother with an accurate outline of sun and moon, preferring to depend on photographs for that. I used an easel protected by wind guards, white cardboard, and pencil, and with these I had ten drills preparatory to painting the eclipse.

When I came actually to do the sketch with the model posed before me for 112 seconds, I found I had ample time. I began with the sky, allotting ten seconds to that, then ten seconds for the moon and twenty for the corona, twenty for observation through the binoculars to survey the red prominences, and so on. When I got through I found I had jotted down these numbers at various points: sky, 30; clouds, 36; corona, 30–60; red prominences, 63; hydrogen flames, 70.

Two days later I was shown the negatives and the positives from photographs taken from the time, and with them and my own color observations, I painted the first drafts of my pictures.”