RTI (Reflectance Transformation Imaging) Workshop 2 – hidden features brought into the light

Friday 24 June 2022

At a workshop back in April I applied RTI (Reflectance Transformation Imaging) to the processing of images of a twelfth-century parchment leaf and found some details that were difficult to observe with the naked eye.

From the University Collections of St Andrews, this is a Latin leaf containing the homilies 31-32 by St Augustine on the Gospel of John (ms39089). The text, in double columns, was written in the first half of the twelfth century in Italy using a slightly rounded Carolingian script. The leaf was excised from its original manuscript codex; other leaves from the same original have been dispersed through a variety of auction sales since the 1960s, and may now be found in various libraries.

Figure 1 – pressure lines underneath the text

Firstly, with normal lighting, one can barely see any lines on the page to guide the scribe in aligning the text, but by handling it with RTI, the pressure lines underneath the text appear remarkably clear (Figure 1). There is no pigment on these lines, hence they were not drawn with a quill, but would have been created by pressing with a pointed tool – a process known as ‘dry-point’ ruling. Neatly spaced and parallel, they should be aligned with the prickings at the outer edge of the leaf (Figure 2). Meanwhile, one notices the intersections of horizontal and vertical lines in the central part of the leaf. Extending down from the punctures at the top, the four vertical lines frame the width of the body text (Figure 3). By adjusting the angle of light incidence and colour saturation in the RTI viewer, we obtain a more three-dimensional visual effect, so that the subtle changes in the depth of the parchment can be detected.

Figure 2 – prickings at the outer edge of the leaf
Figure 3 – vertical lines frame the width of the body text
Figure 4 – Incipit omelia XXXII in red

It is also with the assistance of a photograph with a low angle of incidence and low saturation (even black and white) that the variation in the scribe’s writing is revealed. If we zoom in around the large Incipit omelia XXXII in red, we can witness the difference in the copyist’s treatment of the heading in majuscule script and the text (Figure 4). In addition to using different colours, the scribe distinguishes the two with varying intensity – the dark brown text floats like fallen leaves on the surface of the parchment, while the red headings are embedded as deeply as an intaglio. In particular, the initials of Incipit and the preceding D – used as a visual guide for the convenience of reading to show the start of a section – show significant shading under RTI’s manipulation. The illuminator, similarly, rendered the decoration with various levels of intensity. In the decorated initial “M,” the yellow letter is clearly set deeper into the parchment, while the rest of the decoration lies on top of it, creating a somewhat three-dimensional effect.

Figure 5 – “normals visualization”
Figure 6 – black and white mode
Figure 7 – follicles

Last but not least, RTI delivers a more straightforward presentation of the texture of the leaf beneath the text. Switching the “rendering mode” in RTI viewer to “normals visualization”, one gets the normal mapping of the object (Figure 5). In this way, the text is completely obscured and the materiality of the leaf as an object is strongly emphasized. Layers of ridges can be observed centered somewhere in the middle of the leaf and spreading out to all sides; there are more pronounced wrinkles in the upper and lower right corners, which might be the result of long-term usage or exposure to damp. If we adjust to black-and-white mode, the ridges become more three-dimensional with their shadows (Figure 6). Especially in the lower right corner, the texture of the original animal skin appears to be sharper as well. Zooming in, we can even see traces of the original follicles (Figure 7).

Figure 8

Another interesting point about this area is that there seems to be a small sign at the edge of the page (Figure 8). We cannot confirm whether this is an inherent defect in the animal’s skin or a note made by the book’s compilers—it could be, for example, a quire mark to indicate the location of the leaf in its quire, and further in the whole manuscript. Either way, it confirms RTI’s amazing capacity to reveal minutiae that the human eye alone cannot detect.

Weiliang Xu

Leave a reply

By using this form you agree with the storage and handling of your data by this website.


Enter your email address here to receive notifications of new posts by email.

Join 6,499 other subscribers.