Purpose- To recreate iris green pigment for use by illuminators as detailed in multiple historic sources¹.

Background- Production of iris green pigment is detailed in De Arte Illuminandi (14th c. Italian) and the “Paduan manuscript” c.1650² among others³. In both texts, the flowers of the iris are collected, but two significant themes are outlined. Either the extraction is done by crushing, fermenting, treating with alum, then filtering the flowers and treating with quicklime or by forming a filtered solution and absorbing onto cloth. While superficially, these appear to be different methodologies, the results on a chemical level would be significantly different by definition, as explained in results.

Process- Flowers of the Iris germanica were gathered from around a dozen plants and placed into a clay vessel, then were crushed roughly with a clay pestle and mixed with alum (around 10g to 200g of flowers). This mixture was left to ferment for a week, and notably including a few ounces of rainwater in the process. This mixture was then strained through cloth into a glass tray (Fig 1). At this stage calcium hydroxide⁴ was added in the place of quicklime. Of note when this touched the extraction bag, an immediate precipitate of green was formed (fig 2). This was then left to dry out of sunlight with the assistance of a fan for around five days until completely dry.

Results- The dry glass tray appeared blue with a green and yellow underlayer. Upon initial scraping three distinct layers were found (Fig 3). The bottom layer was pale yellow and typical for degradation products of organic material expose to a strong basic solution with noted yellowish tint. The next layer appeared to be green powder consistent with solids dropped from solution on reaction. The top layer was a slightly gummy blue layer that appeared to remain in solution until the last of the water had evaporated. Upon washing the tray afterwards, this blue was found to still be water soluble. The blue and green, where physically separate were collected in small amounts, then the rest of the powder collected. These were ground and mulled separately and made into paint by adding liquid gum arabic for the resultant colors in the main photo.

Discussion- the aforementioned different processes seem to make sense upon reflection. The clothet method in De Arte Illuminandi would concentrate the blue/green shades, but would experience significant decrease in the overall amount of pigment produced by 10-20 times less. It could also be used in different ways due to it still being soluble, such as a wash or glaze. The rather large amount⁵ obtained from the mixed layers ground easily and both produced great color on pergamenata. Further experimentation could be used to first test the viability of the clothet method if a large enough quantity of raw materials were obtained versus the color shift if the solution obtained from squeezing the flowers could be finely filtered to remove organic detritus. The current samples are going to be left in a sunny window to test colorfastness and correlate with the idea that this is why they are not included in Cennini.

Please send feedback with question, further suggestions, or if you would like a sample.

1. I would like to thank the writings of Lady Marion Forester, modernly known as Lisa Adams, who has taken the time and care to document her processes and results at https://dragonflyscribeblog.wordpress.com/ which was both foundational and inspirational to this process.

2. As detailed in Medieval and Renaissance Treatises on the Arts of Painting: Original Texts with English Translations by Mary P. Merrifield

3. Although noticeably absent from Cennini’s Il Libro dell' Arte, who has a dim view of fugitive or light sensitive pigments. See discussion.

4. Calcium hydroxide aka pickling lime is a soluble base that results in a similar pH change but is notably safer than quicklime or lye to have around the house.

5. Enough from ~20 blooms (a fist size mass after squeezing) to fill a 3 ounce container