Random Constellation

How this random constellation generator works

Press Generate and you get a constellation. Turn up the Number box for several at once, all different. Copy takes them to your clipboard.

Behind the button sit the 88 constellations recognised by the International Astronomical Union, in their Latin names. Not 87. Not 90. Eighty-eight, exactly, and the exactness is the interesting part.

The three filters trim before the draw. Starts with and Ends with match the first and last letter, ignoring case. Contains matches anywhere in the name.

A constellation is not a pattern of stars

This is the thing almost everybody has wrong, including people who own telescopes.

A constellation is an area of the sky. It is a region with a border drawn around it, exactly the way a country is a region with a border drawn around it. The stars inside it are incidental. They are not connected to each other and mostly they are nowhere near each other.

Since the borders were fixed, every single point in the sky belongs to exactly one constellation. Not most points. Every point. The 88 tile the entire celestial sphere with no gaps and no overlaps, the way floor tiles cover a floor.

Which means when an astronomer says a supernova went off "in Cassiopeia", they are not saying it happened among those particular stars. They are giving you a map reference.

It also means the Big Dipper is not a constellation. It never was. It is seven stars inside Ursa Major, which is a much larger region containing a great many other stars. A recognisable shape made of stars is called an asterism.

Why there are exactly 88 and not 87

Because in 1922, a committee decided.

The International Astronomical Union held its first General Assembly in Rome that year, and its Commission on Notations and Units agreed on a list of 88 constellations covering the whole sky, each with a three-letter abbreviation.

They had to. Nineteenth-century star charts had drifted into chaos. Astronomers had been inventing new constellations for a century and a half, sometimes to honour a patron, sometimes to fill an awkward gap, and by the end nobody could be sure that two charts meant the same thing by the same name.

The 88 that survived are a strange, honest mixture. Forty-eight of them come almost unchanged from Ptolemy's catalogue in the second century. The rest were added by European astronomers as they mapped the southern sky, which is why the far south is full of scientific instruments and the north is full of Greek heroes.

Thirty-six lie mostly in the northern sky. The other fifty-two lie mostly in the south. Apus, Cassiopeia, Cetus, Coma Berenices, Dorado, Lacerta, Ophiuchus, Pyxis, Vela, Vulpecula.

The man who drew the lines, and the rule he had to follow

Agreeing on 88 names was the easy part. The 88 still had no borders. Star charts showed boundary lines wandering vaguely between the figures, and they differed from map to map, so a faint star could sit in one constellation on one chart and another on the next.

At the IAU's second General Assembly in Cambridge in 1925, Eugène Delporte of the Royal Observatory in Brussels proposed drawing the borders along lines of right ascension and declination, which are the sky's version of longitude and latitude. Everything would be a rectangle, or a staircase of rectangles. No curves, no argument.

And they gave him one specific constraint, which is the detail that makes this a human story rather than an administrative one. Every known variable star had to stay inside the constellation it had already been assigned to.

Think about what that means. Variable stars are named after their constellation. Redraw a border carelessly and a star that has been called by one name in every paper ever published about it suddenly lives somewhere else. So Delporte's staircases zigzag, because they had to walk around stars that could not be moved.

The boundaries were approved in 1928 and published in 1930, in a volume called Délimitation scientifique des constellations. That book is why every point in the sky has an address.

The borders are already crooked

Delporte drew his lines for the epoch B1875.0. That is, he used the coordinate grid as it stood in the year 1875, half a century before he started work.

But the Earth wobbles. Its axis traces a slow circle over roughly 26,000 years, and as it goes, the celestial coordinate grid slides across the fixed stars. Astronomers call it the precession of the equinoxes.

So the borders Delporte drew as perfect verticals and horizontals are no longer vertical or horizontal. Plot them on a modern star chart, which uses the coordinates of the year 2000, and they lean. Every one of them.

Nobody is going to fix this, because fixing it would move stars between constellations and rename things. So the sky is divided by a grid that stopped being aligned with anything in about 1876, and it will keep leaning further every year, forever.

The constellations that were deleted

Eighty-eight survived. Plenty did not.

In 1922, when the IAU handed out three-letter abbreviations, it issued 89 of them. The extra one was Argo Navis, the ship of the Argonauts, an enormous southern constellation that was eventually broken up into three: Carina the keel, Puppis the stern, Vela the sails.

Quadrans Muralis, the mural quadrant, was invented in 1795 and later split between Draco and Boötes. Its only surviving trace is the Quadrantid meteor shower, which still bears the name of a constellation that no longer exists.

This is the part people find unsettling. The sky did not change. The names did. Somebody decided Argo Navis was too big, and it stopped existing.

Ways people actually use this

• Learning the sky one piece at a time. Generate one, look up where it is, then go outside and find it. Eighty-eight of them is about three months at one a night.
• Naming things. Spacecraft, ships, projects, servers, racehorses. Latin, evocative, and unlikely to be taken.
• Astronomy quizzes. Generate five and ask which hemisphere each one belongs to. Most people cannot do it for more than two.
• Writing. If a character looks up, they should look up at something specific.
• Losing an argument gracefully. Check whether what you generated is a constellation or an asterism. Orion is. The Big Dipper is not.

Getting more out of the filters

• Type ium into Contains and you get Horologium, Microscopium, Telescopium. A clock, a microscope, a telescope. That suffix is the fingerprint of the eighteenth-century astronomers who filled in the southern sky.
• Type austral and you get Corona Australis and Triangulum Australe. Type boreal and you get Corona Borealis. Southern crown, southern triangle, northern crown. The Latin is telling you where to look.
• Starts with C is the crowded letter. Ends with S catches most of the masculine and plural Latin endings in one go.
• Look for the two-word names. Ursa Major and Ursa Minor. Canis Major and Canis Minor. The second word is usually doing the work of telling two similar things apart.

Questions people ask

How many constellations are there?

Eighty-eight, and the number is exact rather than approximate. It was fixed by the International Astronomical Union in 1922 and has not changed since.

How many constellations are in the zodiac?

Twelve, by tradition. But the ecliptic, the path the Sun appears to take through the sky, passes through thirteen of the 88. The thirteenth is Ophiuchus, and it was left out of the zodiac long before anybody drew official borders.

Is the Big Dipper a constellation?

No. It is an asterism, a recognisable shape made of stars, and it sits inside the constellation Ursa Major. The Plough is the same seven stars under a different name.

Which constellation is the biggest?

Hydra, the water snake, which is a question about area rather than about stars. The 88 between them cover the whole celestial sphere, about 41,253 square degrees.

Are the stars in a constellation near each other?

Almost never. They are at wildly different distances and only look adjacent because we see them projected onto a flat sky. A constellation is a line of sight, not a neighbourhood.

References

1. The constellations, International Astronomical Union
2. Constellation boundaries, Ian Ridpath
3. IAU designated constellations by solid angle