Corrosion of the stained glass in a 15th-16th-century window in the title church involved the red glass almost exclusively and covered 3.7% of the total window surface. The chemical composition of corroded and uncorroded samples was determined by electron microprobe analysis and is tabulated. The corroded and uncorroded glasses differed mainly with respect to the type of alkaline flux used, which was potassium in the former case and sodium in the latter. In the Middle Ages, flux was composed of plant ash. Sodium-containing flux was of Venetian origin, whereas the potassium flux was from inland Europe, probably Germany. The corrosions of the exterior and interior faces of the glass had the same composition and consisted almost exclusively of gesso (CaSO₄ . 2H₂O), with traces of singenite (K₂Ca(SO₄)₂ . H₂O). Th e sulphur came from atmospheric sulphur dioxide. The initial phase of corrosion is apparently due to atmospheric water, which, in contact with the glass, extracts the alkali and alkaline earth elements by an ion-exchange process. Most of the potassium-containing glass was ruby red in color, a tint obtained with copper and melting under reducing conditions. The copper en-richment corresponded to an enrichment in lead and, to a lesser extent, in manganese, iron, and aluminum. The ruby glass consisted of a central colored layer and two surrounding trans-parent layers, but copper had diffused from the center and was present in relatively high con-centrations even in areas that were not colored. -- AATA