Blog posts on Tartan Dictionary

Weaving in software — making the cloth measure true

Sun, 19 Jul 2026 00:00:00 +0000

Every woven swatch in the Dictionary — the samples on the variant pages, the favicon, the background tiles, the patches on the print sheets — is woven in software, thread by thread. This is the story of how that weaving works, and of a calibration problem it turned out to have: the woven cloth measured darker than the dye it was woven from.

The loom, pulled apart

Tartan is a 2/2 twill: each warp (vertical) thread passes over two weft (horizontal) threads, then under two, and each row of weft advances the pattern by one thread. Woven at a few fat threads, with gaps left open so every crossing is visible, the software's loom looks like this:

Where the Rainbow Sits in OKLab

Tue, 14 Jul 2026 00:00:00 +0000

The dictionary does all of its colour reasoning in OKLab — classifying shades, measuring ΔTartan, drawing dye-range maps. Those maps always show colours inside a gamut. This story asks the opposite question: where is the outer edge of all physically possible colour — the locus of pure monochromatic light, the curve that draws the famous horseshoe of the CIE 1931 chromaticity diagram — when you plot it in OKLab?

Threads of tradition — the Maasai shuka

Mon, 13 Jul 2026 00:00:00 +0000

The dictionary holds a set of Masai Shuka artefact tartans — checks read from the woolly red blankets worn across Kenya and northern Tanzania. Our interest is the one this dictionary always has: the pattern and its lineage — how the tartan check's grammar met a much older Maasai colour tradition and combined into a genuinely new set of setts. Their story is not a Highland one, and it is worth telling properly. This retelling draws on Paukwa's KE Textiles series essay, Threads of Tradition: The Tale of the Maasai Shuka, and on the sources behind our own records, below.

Neighbourhood maps — a flat picture of a many-dimensional space

Sun, 12 Jul 2026 00:00:00 +0000

Under the nearest tartans table, many pages in the Dictionary carry a small scatter — the neighbourhood map. Every grey dot is a cloth in the corpus, the red dot is the cloth whose page you are on, and the blue dots are its nearest neighbours. This post explains what the map is, and — just as important — what it is not.

Where the dots come from

ΔTartan describes each cloth by a bundle of measured features — its colour ladder, its proportions, how its sequence jogs against others — and measures the distance between two cloths in that feature space. That space has many dimensions: two tartans can differ in ground colour, in the shade of one overstripe, in the count of a band, in the order the colours cycle, in scale, each an axis of its own.

The Baronage of Angus and Mearns — family tartans of 1856

Sun, 12 Jul 2026 00:00:00 +0000

In 1856 David MacGregor Peter published The Baronage of Angus and Mearns: comprising the genealogy of three hundred and sixty families … being a guide to the tourist and heraldic artist — a county-by-county walk through the landed families of Angus and the Mearns, each article carrying arms, crest, motto, seat and anecdote. And, for thirty-eight of the families, a tartan: the sett spelled out stripe by stripe, exactly as James Logan had done twenty-five years earlier.¹

The Vindolanda Check

Wed, 08 Jul 2026 00:00:00 +0000

A generation before the Falkirk cloth was stuffed into its coin-pot, there was already checked wool being worn on the northern frontier of Roman Britain. It comes from the fort of Vindolanda, just south of Hadrian's Wall, and it is one of the earliest checked cloths to survive from Britain at all.

The find

Vindolanda's waterlogged early levels are a textile archive: the anaerobic mud under the first timber forts preserved leather, wood, writing-tablets and cloth that would have rotted anywhere drier. Among the woollens excavated there is a simple two-colour check, deposited in the fort's earliest levels, around AD 100 — a full century or so before the Falkirk hoard was buried. The textiles were published by J. P. Wild in the Vindolanda research reports, and the fleeces themselves were analysed by M. L. Ryder.¹²

ΔTartan: how far apart are two tartans?

Mon, 06 Jul 2026 00:00:00 +0000

Every cloth page in the Dictionary carries a nearest tartans table: the most similar cloths in the whole corpus, each with a number in the ΔTartan column. This post explains what that number measures — why a faded, two-hundred-year-old kilt sits a whisker from its own younger self while two tartans that share most of their colours sit far apart.

Judged the way a weaver would

A tartan is a small palette deployed in a strict order, and that is how ΔTartan reads it. Each sett becomes a colour ladder: its colours are grouped into hue families (red, green, blue…), ranked dark-to-light within each family, and split by prominence into the ground — the broad foundation colours that set the cloth's character — and the decoration — the thin overstripes and dark accents laid over it.

The Walkers Shortbread Tartan

Sun, 28 Jun 2026 00:00:00 +0000

Few tartans are seen by more people than the red one wrapped around a box of Walkers Shortbread — a Speyside biscuit, "established 1898", sold the world over in its tartan livery. In 2016 Walkers registered that packaging as a UK trade mark (UK00003180795), and the device shows the tartan plainly.

It is usually called Royal Stewart, but a glance at the cloth is enough to doubt that — the ground is right, but the green blocks are too large and there is no white overcheck. So the question for this post is the narrow, hands-on one: what is the sett?

Drummonds of Megginch Tartan Notes

Tue, 23 Jun 2026 00:00:00 +0000

Drummonds of Megginch Tartan

Originally written May 16, 2018

There has long been confusion about the tartan for the Drummonds of Megginch. This is due the tartan being very similar to the New Bruce / New Grant tartan that was recorded by Wilson in the 1819 weaving pattern book.

Drummonds of Megginch Tartan different from Wilson Grant pattern.

The earliest example I currently have of the tartans that the Drummonds of Megginch is a kilt dating from 1820 to 1849 which I wore from about 1975 to 2000. The picture is of the inside which has faded less the outer. I can be confident of the latter date as there is a picture of JMD (James Murray Drummond) wearing this with a sporran which I also have. The sporran was made in 1849 in Edinbugh and can be dated from the silver makers mark.

The fine grid — a colour space sampled evenly

Sat, 13 Jun 2026 00:00:00 +0000

A specifier book like Pantone or Munsell samples colour space unevenly — densely where its trade cares, thinly elsewhere — because it is a record of what people actually made. The Dictionary needs the opposite: a frame in which equal distance is equal perceived difference, filled evenly, so that "how far is this shade from that one" has a single honest answer. That frame is the fine grid.

What it is

Take OKLab — the perceptual colour space the colour list already uses — and simply fill it at one spacing in every direction:

The six-colour reference palette

Sat, 13 Jun 2026 00:00:00 +0000

The Dictionary names every pattern in a simplified six-colour vocabulary — white, black, red, green, yellow, blue, "the rest are shades of these six." These are not screen primaries (tartans are not woven in #FF0000); they are representative cloth shades, chosen to maximise discrimination — to keep the colour distinctions that genuinely separate one tartan from another and throw the rest away. This post re-anchors those six onto the fine grid and re-runs the experiment that picked them.

The Drummonds of Megginch Tartan Collection

Thu, 11 Jun 2026 00:00:00 +0000

For two centuries the Drummonds of Megginch have worn one tartan, and for two centuries no two weavings of it have been quite the same. The collection held at Megginch Castle — a plaid, three kilts, a child's kilt, even a carpet — is the same design at different scales, in different dyes, at different states of fading. This post gathers the whole collection in one place: every artefact, every thread count, every shade we have measured, and a printable poster of the lot.

Logan's Scottish Gaël — the first published tartans

Wed, 10 Jun 2026 00:00:00 +0000

In 1831 James Logan published The Scotish Gaël; or, Celtic Manners, as Preserved Among the Highlanders — two volumes of antiquarian sweep covering arms, dress, music and language. Tucked into the appendix of volume II is something genuinely new: a Table of Clan Tartans, the first time the setts of tartans had ever been published. Fifty-four patterns, each recorded stripe by stripe.¹

The Falkirk Tartan

Tue, 09 Jun 2026 00:00:00 +0000

Scotland's oldest surviving tartan is not a clan sett, is not dyed, and very nearly did not survive at all. It came out of the ground as a wad of cloth stuffed into the neck of a Roman money-pot.

The find

On 9 August 1933, workmen levelling ground near Falkirk — a short way from the Antonine Wall — broke into a buried Roman earthenware jar. Inside were close to two thousand silver denarii, a coin hoard whose latest coins date its burial to around AD 230, in the first half of the 3rd century. The cloth had been pushed into the mouth of the pot as a stopper, and that is the only reason a scrap of third-century wool survived to be found at all. Hoard and cloth are both now in the National Museum of Scotland.¹

The La Tène Tartan

Tue, 09 Jun 2026 00:00:00 +0000

Centuries before the Falkirk sett — and in full colour where Falkirk makes do with undyed fleece — Iron Age Celts were already weaving the combination at the heart of countless later tartans: a dark blue ground crossed by a fine red check. The cloth that proves it came out of a salt mine in the Austrian Alps.

No Tartan for 2,000-Year-Old Galician Kings

Sun, 07 Jun 2026 00:00:00 +0000

The website tartan.galician.org claims Iron Age "Galician kings" wore tartan two millennia before Scotland. The figures are real — the Galaecian warrior statues — but the archaeology says otherwise: the carved garment ornament is a diagonal lozenge lattice read as symbolic, not woven check; no published pigment study supports the colours (which are modern artistic reconstructions); and the "tartan" reading appears only in cultural-revival material, never in the scholarly literature.

Summary

The page tartan.galician.org/kilts.htm presents colour images of "Galician kings wearing kilts/tartan." The figures are genuine and well studied — they are the Iron Age Galaecian warrior statues (guerreiros galaicos / estatuas de guerreiros lusitano-galaicos) of the Castro culture of north-west Iberia. But almost everything the website builds on top of them is unsupported by mainstream archaeology:

The Origins of Tartan

Sun, 07 Jun 2026 00:00:00 +0000

"For the purposes of this Act, a tartan is a design which is capable of being woven consisting of two or more alternating coloured stripes which combine vertically and horizontally to form a repeated chequered pattern."

— Scottish Register of Tartans Act 2008, section 2¹

The Falkirk tartan — Scotland's oldest surviving check, 3rd century AD.

This is how the Scottish Parliament defined tartan when it set up the Scottish Register of Tartans. It is the end of a journey that begins in ancient Eurasia. This post follows that whole line: from the birth of weaving to its eventual flowering, much later, in Scotland into a joyful way of celebrating cultural and family identity.

The Tarim Tartan

Fri, 05 Jun 2026 00:00:00 +0000

There is a piece of checked woollen cloth, woven on the rim of the Taklamakan desert about three thousand years ago, that looks unmistakably like tartan — the oldest surviving example of the 2/2-twill check tradition. Why that matters, and how the same weave reaches from Xinjiang to the Highlands, is the subject of the companion post, The Origins of Tartan. This post does the narrower, hands-on job: it takes the surviving Hami fragment, reads a sett from it, reconstructs the colours the burial bleached away — and sets my reading beside the other reconstructions people have made of the same cloth, because this is a guess, and guesses should be compared.

A tartan's colour order is nearly its name

Mon, 01 Jun 2026 00:00:00 +0000

Reduce a tartan to almost nothing — six colours, and the order they run in, with every thread count thrown away — and you would expect to be left with something far too coarse to tell one tartan from another. Hundreds of designs ought to collapse together. They don't. In the Scottish Register of Tartans, that stripped-down colour sequence is very nearly a fingerprint.

What we kept, and what we threw away

We took the register-of-tartans dataset on its own — 7,393 setts — so that every match is two genuinely distinct tartans, not the same tartan re-listed across sources. Each stripe was quantised to the nearest of six base colours (blue, black, red, green, yellow, white), and the thread counts were discarded entirely. What survives is the Pattern: the bare order of coloured stripes across the half-sett, folded so a reflective sett and its mirror count as one. So B24 K8 G24 R4 and B6 K2 G6 R1 both reduce to the same four-colour Pattern, BKGR.

Does a castle make a clan? The Drummonds' seats over time

Mon, 01 Jun 2026 00:00:00 +0000

The Dictionary already draws a line between a Clan and a Family. A Clan is a surname-and-territory grouping — Drummond; a Family is a set of tartans carried by one branch over more than a couple of generations — Drummond of Megginch, Drummond of Perth. That second name is the interesting part. "of Megginch", "of Perth" is a territorial designation: it names land. And land, in the Scottish tradition, usually means a seat — a castle.

Selecting Colours

Fri, 14 Apr 2023 00:00:00 +0000

This is a broader description about selecting colours. See the selection of colours for the choices of base colours using the Scottish Tartan Registry base colours.

This is partly my journey on discovering how to select and specify colours. When you start looking at colour spaces you can start with Wikipedia. I am keen to know how much a colour varies from a standard colour and what a standard colour is.

Linear transformations

On my computer I have an sRGB monitor with a gamma of 2.2 which seems quite accurate (Microsoft Surface checked with a DataColour spyder). However the gamut is limited compared to the Adobe gamut.

Tartan Threadcounts

Thu, 02 Feb 2023 00:00:00 +0000

Aim is to define the way thread counts are specified.

This uses an ebnf specifcation which is elaborated here which includes further references.

Tartan = WarpThreadCountList { "!" WeftThreadCountList } {Palette}. 
Palette = ColourSpec { ColourSpec}
ColourSpec = ColourThreadCode HexChar ColourHex | ColourThreadCode "~" OKLabColour
HexChar = "#"|"$"
ColourHex = Red Green Blue
Red = Hex
Green = Hex
Blue = Hex
Hex = HexChar HexChar
OKLabColour = LightNess "%" A-Axis "~" B-Axis
LightNess = "1" | "0" {"." Digit {Digit {Digit {Digit}}}}
A-Axis = AxisNumber
B-Axis = AxisNumber
AxisNumber = "0" | {"-"} "1" | {"-"} "0" "." Digit {Digit {Digit {Digit}}}
Digit = "0" | "1" | "2" | "3" | "4" | "5" | "6" | "7" | "8" | "9" ;
HexChar = Digit | "A" | "B" | "C" | "D" | "E" | "F" 
WarpThreadCountList = ThreadCountList .
WeftThreadCountList = ThreadCountList .
ThreadCountList = FirstThreadCount {ThreadCount} {LastThreadCount}.
ThreadCount = ColourCode Integer .
FirstThreadCount = ColourCode ReflectiveSettSpecifier Integer | RepeatingSettSpecifier ThreadCount
LastThreadCount = ColourCode ReflectiveSettSpecifier Integer | ThreadCount RepeatingSettSpecifier 
ReflectiveSettSpecifier = "/" 
RepeatingSettSpecifier = "..."
ColourThreadCode - ColourCode | ColourCode LowercaseChar
CoreColourCode = "W" |"R" | "G" | "B" | "Y" | "K" 
ColourCode = CoreColourCode | "LR" | "DR" | "O" | "DO" | "LY" | "DY" |
 "LG" | "DG" | "LB" | "DB" | "LP" | "P" | "DP" | 
 "LN" | "N" | "DN" | "LT" | "T" | "DT".
LowercaseChar = "a" | "b"
 | "c" | "d" | "e" | "f" | "g" | "h" | "i"
 | "j" | "k" | "l" | "m" | "n" | "o" | "p"
 | "q" | "r" | "s" | "t" | "u" | "v" | "w"
 | "x" | "y" | "z"

Notes:

Tartan Colours

Sun, 22 Jan 2023 00:00:00 +0000

A dictionary needs a colour list because the same design is woven in many shades: you must allow variation and still call it the same thread count. So the real question is not "what are the colours" but "when are two shades the same colour" — a distance question, best asked in a perceptually uniform space (OKLab), not in sRGB.

Is the Tartan Registry list good enough?

The Register's list is the obvious starting point. Run through OKLab it turns out to be a list of 131 example shades under 24 codes (not 19), and as a partition of colour space it does not hold up:

Tartan Symmetry and Normalisation

Sun, 15 Jan 2023 00:00:00 +0000

Symmetry.

The simplest tartans have reflective symmetry in both the warp and weft (x,y) directions and are identical. They are defined by pivot points in the symmetry.

So.

R/2 B2 G/2

creates a non repeating pattern:

R2 B2 G2 B2 R2 B2 G2 B2 R2 B2 G2 B2 R2

This is a repeating pattern of:

...R2 B2 G2 B2...

but whereas the reflective pattern has defined end points the repeating pattern can choose any starting point so the following are all the same pattern:

1997 Kilt

Mon, 02 Jan 2023 00:00:00 +0000

This tartan was woven for my wedding. It was done by Dalgeish in Selkirk and they have muted the designs.

Dark Blue: RGB 40 44 57 Nearest Pantone colours Black C, 2695 C and 275 C.

Thread counts

R/14 DB2 R4 DB4 R70 LB4 R4 DB20 R4 G4 R4 G74 R6 DB4 R/12

What is Tartan?

Mon, 02 Jan 2023 00:00:00 +0000

Tartan or checked weaving pattern has ancient origins. Perhaps it came with the spread of the ancient Bronze age Celts from Austria and Turkey. To the East there may be a link to the Tarim Mummies as was popularised by Elizabeth Barber¹. This currently has got caught up in origin stories of the Han. Notable mummies are the tall, red-haired "Chärchän man" or the "Ur-David" (1000 BCE); who wore a red twill tunic and tartan leggings. Elizabeth Barber highlights the similarities between this and fragments recovered from salt mines associated with the Hallstatt culture.

Old Kilt

Wed, 28 Dec 2022 00:00:00 +0000

Old Kilt

The earliest example I currently have of the tartans that the Drummonds of Megginch is a kilt dating from 1849 which I wore from about 1975 to 2000. The picture is of the inside which has faded less the outer. I can be confident of the date as there is a picture of JMD (James Murray Drummond) wearing this with a sporran which I also have. The sporran was made in 1849 in Edinbugh and can be dated from the silver makers mark.

Old Plaid

Wed, 28 Dec 2022 00:00:00 +0000

Old Plaid

My mother wore this for our wedding and the picture is on [Tatlers site][]. [Tatlers site]:https://media.tatler.com/photos/6141d8016b716d36b22fccff/master/w_2580%2Cc_limit/Lady-Strange-Sir-Ian-Rankin-tatler-06may15-Zygmunt-Sikorski-Mazur.jpg

This also featured in Vanity Fair with Liam Neeson when he was sitting on it.

Detail of Tartan

This was one of the best unfaded squares.

Then there is the fine detail:

This for part of the tartan gives a thread count of:

Drummonds New Bruce 1819

Sat, 18 Jun 2022 00:00:00 +0000

"XR#D40000; A#5C8CA8; HG#285800; DP#440044;","XR/12 DP2 XR4 DP4 XR64 A2 XR4 DP16 XR4 HG4 XR4 HG48 XR4 DP4 XR/12"

Drummond of Megginch Tartan thread counts

Fri, 10 Jun 2022 00:00:00 +0000

Given thread counts produce table

Wait for this to come out

test = "R14 DB2 R4 DB4 R70 LB4 R4 DB20 R4 G4 R4 G74 R6 DB4 R12"

import re

PATTERN = re.compile("([a-zA-Z]*)(\d*)")

def split_thread(threadcount):
 """Given a thread count such as R14, will split to R and integer 14"""
 match = re.search(PATTERN, threadcount)
 if match:
 return match.group(1),match.group(2)
 return threadcount,""
 

class Stripe:
 def __init__(self, short_colour, count):
 self.short_colour = short_colour
 self.count = int(count)

 def __str__(self):
 return f"{self.short_colour}{self.count}"

class Tartan:
 def __init__(self):
 self.stripes = []
 
 @classmethod
 def from_space_threadcount(cls, threadcount):
 tartan = cls()
 threads = threadcount.split(" ")
 for thread in threads:
 colour,count = split_thread(thread)
 tartan.stripes.append(Stripe(colour, count))
 return tartan

 def __str__(self):
 s= ""
 for stripe in self.stripes:
 s += f"{stripe} "
 return s

 @property
 def threadcount(self):
 tc = 0
 for stripe in self.stripes:
 tc += stripe.count
 return tc

a = Tartan.from_space_threadcount(test)
print(f"Pattern = {a}")
print(f"Num threads = {a.threadcount}")

Pattern = R14 DB2 R4 DB4 R70 LB4 R4 DB20 R4 G4 R4 G74 R6 DB4 R12 
Num threads = 230

Creating images

My first attempt at creating images was using a simple turtle graphics package:# Free after the example from the Python 3.5 manual

Details of New proposed Tartan

Wed, 01 Jun 2022 00:00:00 +0000

Colours are the for the moment in sRGB perceptual space.

The proposal is this one: 2023 Kilt (There may be some tweaking but will be updated here.)

This is a list of the kilts and proposals:

1820 Plaid See also the post around this 1820 post
1849 Kilt See also the post around this 1849 post Note the colours on this need a little correction.
1890 Child's kilt
1967 Carpet
1997 Kilt Note the colours have been done on this although the light blue may need work.
2022 Proposal Initial proposal with Lexa and Bertie's colouring back to the 1849 colouring
2023 Kilt As the 2022 proposal This reverts to the thread count of the 1849 kilt rather than 1997 with the colours of the 1997.

The following is a picture of the proposed new tartan sample.