Natural dyes - results of tests for lightfastness.

Apologies to Online Guild members, who should have seen this on our mailing list, but these results belong here also as I have been writing about my work with natural dyes.

I have just taken down my lightfastness test cards which have been taped to
the hall window for the past two months. The cards have yarn samples wrapped around them, and the top half of each card has been protected from sunlight with a double thickness of white card over the top of the yarns, so half of each yarn has been in the sunlight and half hidden.

I tried scanning the cards, but my scanner is old and just about dead. I tried photos, but they weren't particularly useful to show the more subtle changes, so here's a description of the results.

For most of these dyes I have samples mordanted (1) with alum, (2) with iron after-mordant. In most cases the samples treated with iron after-mordant are less faded.

The most light fast dyes, barely faded at all are:

Broad leaf dock root (Rumex obtusifolius) (1) and (2)
Heather (1) and (2)
Weld (1) and (2)
Marjoram (1) and (2) (this was unexpected!)
Eucalyptus coccifera (1) and (2)

Slightly faded, with dark greenish shades becoming brownish are:

Bracken (1) and (2)
Broad leaf dock leaves (2)
Feverfew (1) and (2)

Moderate fading:

Broad leaf dock leaves (1) (went from khaki green to light brown)
Henna (1) (strong orange to soft orange)
Stinging nettle (2) (deep green has gone a bit silverly - still a lovely colour)

Badly faded - in worst cases now off-white (x) are

Twilleys Freedom Spirit (yes, a commercial, chemical dyed yarn faded more than some natural dyes!)
cutch (1)
onion (1)
Stinging nettle (1)
Madder (1) (still very pretty)
Turmeric (1) (this has gone greyish!)
Buddleia (1) (x)
Dyers chamomile (1) (x)
Buttercup (1) (x) and (2)
Buckler leaved sorrel (1) (x) and (2)

Several surprises here, for example, Feverfew kept its colour far better than I anticipated, and the commercial yarn much worse! It was well worthwhile doing these tests because the results were in most cases not what I had expected. My next sample card will included more commercial yarns, it's a useful comparison as I think we would normally assume they don't fade much.

UPDATE...

Thanks to my Online Guild friends I have been able to work out why some of my test results were unexpected. Everything that faded more than expected was a Shetland sample. This may be due to the original dye take-up being less than the dyes used on the superwashed merino. Possible reasons why include the wool not being wet through properly before dyeing, the amount of lanolin in the wool, the superwash process may improve dye take-up.

Looking at the dyes on Shetland wool:

Weld dyed samples (yellow from just Alum and C of T, also two shades
green from iron after mordant) did not fade at all. Wonderful!

Madder went from pinky red to pink - significantly less faded than
other samples on the same card (except the weld).

Turmeric yellow went grey. Even where had been covered this yarn
was greatly faded.

Showing white in part or all of the faded sample were: Cutch, onion,
henna, Dyer's Chamomile, and Buddleia.