Technical Notes on Nattier's "Portrait of a Woman as a Vestal Virgin"

Abstract

The busy and at times confusing x-radiograph of Jean-Marc Nattier's Portrait of a Woman as a Vestal Virgin, upon closer inspection, reveals a wealth of information about the con struction techniques and history of the painting (Fig. i). In most cases, the plainly visible, white brush markings are indicative of the initial paint application, when the artist lays in the contours and highlights of the composition with loaded, broad strokes prior to the application of the opaque body color or transparent glazes. This technique is similar to draw ing with paint and has its origins in Venetian painting of the sixteenth century. This application of paint is most effective for laying out the composition when the ground is a dark tone, in this case a gray color, and the use of white paint provides a workable contrast. Many preliminary brush markings became the foundation color for the highlights or accents as delin eated in the initial conception. The x-radiograph also reveals changes the artist made to the size of the painting, as well as possible changes in the way the painting was later displayed. The two vertical canvas strips added to the left and right sides of the painting indicate that the artist was working out the size of the composition as he was laying in the figure and background. The strip on the right (about eight cm wide) was most likely added when the canvas was initially prepared in the studio. Comparison of the density and consistency of the ground application, which contains white lead pigment, with the right addition and the main canvas reveals that the strip was already sewn on when the ground was applied. The width of the main canvas (124 cm) was apparently not enough for the initial conception of the composition. It is also apparent that the ground was applied after the painting was stretched onto the first strainer. Linear markings of thinner ground application at the inside edge of the strainer bars are evident along all sides, about two to three inches in from the edge, likely indicating the width of the strainer bar members. On the left side, the linear markings occur about two inches in from the sewn edge of the addition, which reveals that the canvas was first stretched onto a strainer with only the addi tion at the right, with a second stretching occurring later. The added strip on the left side, which is between ten and eleven centimeters wide, also has the thinner ground marking, visible as a vertical line in the x-radiograph because of the disruption in application from the inside edge of the second, expanded stretcher member. This strip addition appears to have been cut from a canvas that already had a ground layer and possibly even some paint application on it. The strip, sewn onto the trimmed, previously stretched left edge, reveals a much thicker white lead application than the main canvas. The canvas was then restretched onto a second, larger strainer. A new ground application was applied over the new strip and partially onto the main canvas to create a