Substantive and Adjective: Which Dyes Need a Mordant and Which Do Not

An adjective dye has no useful affinity for the fibre by itself. The mordant bonds to the fibre and the dye bonds to the mordant, and the result is fast. This family includes nearly all the great historical colours: madder and its alizarin reds, cochineal and kermes, weld, dyer's broom, fustic, logwood, brazilwood, osage, coreopsis and marigold. Two consequences follow from the structure and both are useful: **The mordant changes the colour, not just the fastness.** The same weld gives lemon on alum and olive on iron, because the colour comes from a dye-metal complex and the metal is part of it. One dyestuff and three mordants is three colours. **Unmordanted fibre in an adjective bath takes a stain, not a dye.** It will look like a pale version of the colour coming out of the pot, and it will rinse and wash away. That is worth seeing once, deliberately, on a scrap.

A substantive dye bonds to the fibre directly, usually because it is rich in tannin and tannin has an affinity for both protein and cellulose. The usual ones are the browns and the tan-to-black end of the range: walnut hulls, oak bark and galls, sumac, myrobalan, cutch, tea, pomegranate rind, persimmon tannin. They will dye an unmordanted skein and the colour will stay. That makes them the right place to start for anyone dyeing for the first time — no mordant step, no chemistry, no disappointment. Two caveats: **A mordant still changes them.** Walnut on alum is a warmer brown; walnut on iron is nearly black. Not needing a mordant is not the same as not responding to one. **They mordant the next dye.** A skein dyed in walnut or oak gall is a tannin-mordanted skein, whether that was intended or not, which is exactly the trick the cellulose sequence uses on purpose.

Indigo and woad are neither adjective nor substantive, because indigo is not soluble in water. It cannot be dissolved and therefore cannot be absorbed. The vat works by chemically **reducing** the indigo — removing oxygen from it — in an alkaline liquor, which turns it into a soluble yellow-green form called leuco-indigo. The fibre goes into that, takes it up, and is then lifted into the air. The oxygen of the air turns it back into insoluble blue indigo, now trapped inside the fibre. That is why an indigo vat is greenish-yellow with a coppery skin, why the fibre comes out green and turns blue in front of you, and why depth is built by repeated short dips rather than one long soak. It also explains the care the vat needs: air is the enemy while the fibre is in it, so nothing is stirred, nothing is dripped back in, and the vat is kept covered. Every bubble undoes some of the reduction. No mordant is used or needed. The indigo is held mechanically, not chemically, which is also why indigo crocks — see the fastness rung.

For an unfamiliar plant, three small tests answer it in an afternoon. **Test one: unmordanted scrap.** Dye a scoured but unmordanted scrap, rinse it hard, and wash it with detergent. Colour that survives means substantive. Colour that mostly leaves means adjective. **Test two: three mordants side by side.** Alum, iron, and none, all in the same bath at the same time. This tells you the range the dyestuff covers as well as which family it is in, and it is the single most informative dye test there is. **Test three: does it dissolve?** A dyestuff that gives a clear coloured solution in hot water is being extracted. One that gives a suspension of particles that settle out is a pigment, not a dye, and will sit on the surface rather than bond — which is how most "failed" flower dyes actually fail. Write all three on one card with the scraps stapled to it. That card is worth more than any amount of reading about the plant, and it is the beginning of the dye record.