Short answer
Diocletian's price edict of AD 301 capped purple-dyed silk at 150,000 denarii a pound, about twice the 72,000 it set for gold. Natural ultramarine is recorded at up to 11,000 French francs a kilogram before a synthetic version reached the market at 850 in 1828. When synthetic indigo arrived in 1897, India's exports of the natural dye fell by 74% in nine years. Each figure comes from one document with its own unit and purpose, so they show scale, not a single price history.
The Edict on Maximum Prices, issued under Diocletian late in AD 301, lists a ceiling price for more than 1,200 goods and services. Its chapter on purple sets raw silk dyed purple at 150,000 denarii for a Roman pound — the highest price in the whole document, shared with a male lion. Undyed white silk is capped at 12,000 denarii a pound, so the dye multiplied the value of the fibre about twelve times. Gold in bars or coin is set at 72,000. Wool dyed with the best purple is 50,000, and the grades step down through lighter and 'Tyrian' purples to wool dyed with archil, a lichen purple, at 300 to 600 denarii. The same colour name therefore covered a price range of several hundred to one, depending on whether the dye came from sea snails or from a lichen.
These are maximum prices decreed during an inflation, in a unit of account, and nobody knows how far they were observed. They show what the administration thought the top of each market was.
Natural ultramarine, extracted from lapis lazuli, is recorded as costing up to 11,000 French francs a kilogram. In 1824 the Société d'Encouragement pour l'Industrie Nationale in Paris offered 6,000 francs to anyone who could make an artificial ultramarine by an industrial process for less than 300 francs a kilogram — a target of well under a thirtieth of the natural pigment's top price. The prize went to Jean-Baptiste Guimet in 1828, after a rival claim by the German chemist Gmelin. The account used here gives the price of the French synthetic product as 850 francs a kilogram: far above the prize's ceiling, yet roughly a thirteenth of the upper price of the natural pigment. The fall was large enough to move ultramarine from a colour rationed in contracts to a staple blue.
Synthetic colour did not always arrive cheap. BASF bought the rights to the first synthesis of alizarin, the red of madder root, in 1869, and a workable sulfonation process was patented in England on 25 June that year, one day ahead of William Perkin's application for the same idea. According to a regional industrial-heritage account, about 230 kilograms were made within a year at 270 marks a kilogram, more than four times the 60 marks of natural madder red. After five years output had reached 100 tons and later more than 1,000 tons, and the price fell to 9 marks a kilogram — a thirtieth of the launch price and less than a sixth of the natural dye's. The same account says madder growing in France and the Netherlands then stopped being worthwhile. It gives no dates for the later prices.
Indigo has the best figures because the Government of India published its exports. In the 1895–96 trade year India exported 187,337 hundredweight of natural indigo, about 9,500 tonnes, at an average of 285.82 rupees a hundredweight. BASF put synthetic indigo on sale on 10 July 1897, after research said to have cost 18 million gold marks. By 1904–05 Indian exports were 49,252 hundredweight, about 2,500 tonnes, at 169.46 rupees. Quantity had fallen by 74%, the average price by 41% and the value of the trade by 84%. The quantity fell faster than the price: planters left the crop, so the market shrank instead of clearing at a much lower price. The decline had begun in 1896–97, before the synthetic dye was on sale, so not all of it can be laid at BASF's door.
Cochineal shows the limit of what can be cited. A museum account records that it became Mexico's second most valuable export after silver, grown almost entirely by Indigenous producers in Oaxaca, with Spain guarding the insect source until the eighteenth century. Annual registration figures for Oaxaca exist in the specialist literature, but Colourwise could not reach a source to confirm them, so no quantities appear here. The figures on this page also cannot be chained together. A denarius of 301, a franc of 1824, a mark of 1870 and a rupee of 1900 are different currencies in different price levels, and a legal ceiling, a prize condition, a firm's selling price and a customs average are different kinds of number. What they share is the pattern: scarcity priced by the pound, then a substitute that cut the price by an order of magnitude.
| Item | Maximum price | Multiple of the gold price |
|---|---|---|
| Raw silk dyed purple | 150,000 | 2.083 |
| Refined gold, in bars or coin | 72,000 | 1 |
| Wool dyed purple (top grade) | 50,000 | 0.694 |
| Wool dyed lighter purple | 32,000 | 0.444 |
| Wool dyed bright Tyrian purple | 16,000 | 0.222 |
| White raw silk, undyed | 12,000 | 0.167 |
| Wool dyed twice in best Milesian purple | 12,000 | 0.167 |
| Wool dyed scarlet with kermes | 1,500 | 0.021 |
| Wool dyed with best archil (lichen) purple | 600 | 0.008 |
| Wool dyed with fourth-grade archil purple | 300 | 0.004 |
| Colourant | What was priced | When | Price |
|---|---|---|---|
| Ultramarine | Natural ultramarine from lapis lazuli, upper price | before 1828 | 11,000 French francs per kg |
| Ultramarine | Synthetic ultramarine by Guimet's process | c. 1828 | 850 French francs per kg |
| Ultramarine | Price ceiling set by the Société d'Encouragement's 6,000-franc prize | 1824 | 300 French francs per kg |
| Alizarin red | Natural madder red | c. 1870 | 60 marks per kg |
| Alizarin red | BASF synthetic alizarin in its first year (230 kg made) | c. 1870 | 270 marks per kg |
| Alizarin red | BASF synthetic alizarin at scale | after output passed 1,000 tons | 9 marks per kg |
| Trade year | Exports (cwt) | Exports (tonnes) | Average price (rupees per cwt) | Quantity index | Value index |
|---|---|---|---|---|---|
| 1895–96 | 187,337 | 9,517 | 285.82 | 100 | 100 |
| 1896–97 | 169,523 | 8,612 | 257.83 | 90 | 82 |
| 1897–98 | 133,849 | 6,800 | 228.42 | 71 | 57 |
| 1898–99 | 135,187 | 6,868 | 219.73 | 72 | 55 |
| 1899–1900 | 111,420 | 5,660 | 241.65 | 59 | 50 |
| 1900–01 | 102,491 | 5,207 | 208.37 | 55 | 40 |
| 1901–02 | 89,750 | 4,560 | 206.38 | 48 | 35 |
| 1902–03 | 65,377 | 3,321 | 184.42 | 35 | 23 |
| 1903–04 | 60,410 | 3,069 | 178.15 | 32 | 20 |
| 1904–05 | 49,252 | 2,502 | 169.46 | 26 | 16 |
Why: The comparison comes from one decree, for purple-dyed silk specifically; cheaper purples in the same decree cost under a hundredth of the gold price.
Fix: Give the item, the year and the document: purple-dyed raw silk, AD 301, a legal ceiling about twice that of gold.
Why: No conversion holds across a debased Roman unit of account, pre-industrial France and colonial India.
Fix: Compare within one document — dye against gold, synthetic against natural — and leave the currency as recorded.
Why: Early alizarin was priced several times above madder until production scaled up.
Fix: Separate the date of invention from the date the price crossed below the natural product.
Each statement is labelled by kind — established fact, a standard’s requirement, observed market data, a convention, or Colourwise’s own interpretation or analysis — with the strength of the evidence behind it.
FactModerate evidence
Diocletian's Edict on Maximum Prices of AD 301 set a ceiling of 150,000 denarii for a pound of raw silk dyed purple, against 72,000 for a pound of gold and 12,000 for a pound of undyed white silk.
Caveat: Maximum prices by decree, reconstructed from fragmentary inscriptions and read here in translation; not evidence of prices actually paid.
FactModerate evidence
Natural ultramarine cost up to 11,000 French francs a kilogram; an 1824 prize of 6,000 francs asked for a synthetic process below 300 francs a kilogram, was awarded to Guimet in 1828, and the synthetic pigment was produced at 850 francs a kilogram.
Caveat: Figures are from the historical introduction of a chemistry paper citing earlier literature; other accounts quote Guimet's price per pound, and the date of the 11,000-franc figure is not given.
Observed market dataStrong evidence· India· 1895–96 to 1904–05
India's exports of natural indigo fell from 187,337 hundredweight in 1895–96 to 49,252 in 1904–05, and the average export price from 285.82 to 169.46 rupees a hundredweight.
Measured: Natural indigo exported from Indian ports, by trade year. Method: Official trade returns as reported in the Indian Planters' Gazette, 8 September 1906, and tabulated by Kumar (2004)
Caveat: Exports only, read at second hand through a dissertation; cultivation, domestic use and stocks are not measured, and prices are averages in nominal rupees.
Source: Facing Competition: The History of Indigo Experiments in Colonial India, 1897–1920
Colourwise analysisModerate evidence
Over those nine years the quantity of indigo exported fell by 74%, its average price by 41% and the value of the trade by 84%.
Based on: Calculated by Colourwise from the first and last rows of the export table reproduced on this page.
Caveat: The fall began a year before synthetic indigo was sold, so it cannot be attributed to the synthetic dye alone.
Source: Facing Competition: The History of Indigo Experiments in Colonial India, 1897–1920
FactLimited evidence
BASF's first synthetic alizarin sold at 270 marks a kilogram, more than four times the 60 marks of natural madder red, and fell to 9 marks a kilogram once output passed 1,000 tons; its synthetic indigo went on sale on 10 July 1897.
Caveat: From a regional heritage project's unfootnoted account; the later prices are undated and the unit for madder red is implied, not stated.
Source: Alizarin 1869, Indigo 1890, Indanthrene 1901, Ammonia synthesis 1913
FactModerate evidence
Cochineal became Mexico's second most valuable export after silver and was produced almost exclusively in Oaxaca by Indigenous growers.
Caveat: A museum exhibit summary with no quantities, prices or dates for the ranking.
Source: Cochineal: The Color of Power
Reviewed 1 October 2026. Colourwise summarises its sources in its own words and does not reproduce standards text or proprietary colour data. Spotted an error? Tell us.