how to do a lightfast test for natural dyes

There’s nothing complicated about a lightfast test, so either people don’t know about it, or they choose to ignore it. All it requires is some cardboard or heavy paper, a couple of clips, and your yarn/thread/fabric, a window, and some patience.

A couple winds of thread or yarn, or a 4×4″ piece of fabric is all that’s needed to do this. Cover half with the cardboard or paper so it’s enclosed and clip it so it’s tight. Put in a very bright window. Wait. Wait some more. Leave it for at LEAST a month. Go do whatever else it is you do, but DON’T commit that larger stretch of yarn/thread, or cloth to any project, or sell it.

After a month, unclip it. What’s happened? Here’s hollyhock bloom dyed yarn after a month :

There’s a VERY distinct line where the cardboard covered it. All of my blooms went into the compost. I overdyed this in indigo and while the shade may eventually change *because* of the underlying hollyhock hue, i’m okay with that in work i keep for myself, or to use as stitch samples.

I have a big thick natural dye journal that i started in 2010 at the beginning of this journey. Yesterday i was looking through it and found the photo on the left taken at the time of the tests of alum mordanted fibres and hollyhock, hibiscus and marigold, and compared it to a 9 year stint in the journal, completely covered, and look at the difference. Even the vaunted marigold flower has faded.

Though i am happy to use marigold, dyer’s chamomile and coreopsis, they will stay in my studio, as this proves to *me* that flowers just don’t have enough strong dye compounds to be useful for long term work. I will never again use flowers like hibiscus, hollyhock or any others that have anthocyanins in them, as the beautiful reds, blues and purples they share, are shared only briefly. EDIT: Since people are getting argumentative about the hollyhock (!!!), see the expert’s opinion here, last paragraph.

Another hibiscus sample below, stored in a drawer for 3 years:

Below, annatto, the folded over top section the area that was exposed to light for a month:

Not as bad a change, but still obvious. Again, i’ll use as a base, or keep in the studio for personal use.

Some natural dyes last better than others, notably madder, indigo and weld, the “grand teint” dyes. Lesser (still very good results in most cases) dyes are most of the others including brazilwood, logwood, fustic, cochineal, sandalwood, cutch, lac etc, ie dyes that are sold by reputable dealers. Tannin rich dyes will eventually oxidize–and that includes avocado–to browns, beiges and mud. Iron will *improve* lightfastness, but does not make anything truly permanent.

Below, a FB friend’s results of lightfastness of avocado:

Left, the first result of avocado dyeing, right is after a lightfastness test.  You can see the darkening of the tannin as it “ages”, even after covering, and to the right in the right photo 🙂 , the lightening of the colour as it is exposed to sun. Ximena lives in South America, and has access to local lore and plant materials that are indigenous to the continent. We’re shared much information about avocados, tannin use, and “true” dyes, with both of us pointing the way to more scholarly papers, something that either people don’t know about, or don’t care enough about to go so far with research, which is sad because we live in such an information rich world now!

Note: Carol Lee, the one who got deep reds from avocado, has said that even after storage for 10 years, her reds turned to brown, proving avocado colours are tannin only, not a true dye for reds and pinks. The tannin oxidized even away from light.

NOTHING FIXES FUGITIVE, so forget the flowers, and food waste unless you’re happy with “play silks” for the kids, or things for personal use that you can re-dye. Please, don’t sell these. All of this is not to say don’t play, BUT please be aware of longevity and the final use of the finished “product”. Again, i say don’t sell fugitive dyed fabrics and yarns.

I still intend to continue experimenting (playing? 🙂 ) with locally foraged dyes, but now armed with a section from the Boutrup/Ellis book, i know what to look for in these forays. Admittedly, early peoples, and settlers to this continent discovered certain plants dyed well enough for their end use, so we’ll go with that. Accurate scouring, mordanting, possible post mordanting and/or modifying, light fast and wash fast tests DO make things last better.

What it all boils down to is: whenever you are trying a plant material for a natural dye, and can find no information on it as to whether or not it works, do a lightfast test.

2 responses to “how to do a lightfast test for natural dyes

Your thoughts? (Spammers, good luck on that.)

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.