What Is a Constellation? Boundaries, Not Just Dot-to-Dot

Ask most people what a constellation is and they will draw a shape in the air, a dipper, a hunter, a scorpion. That's the folklore version, and it's not wrong exactly, it's just not what a constellation legally is anymore. Since 1930, a constellation has been a fenced area of sky with a hard border, and any object inside that fence belongs to it regardless of whether it's part of any recognizable shape at all. The stick figures on the Real Sky map are there for you to find your way around. The actual constellation is the territory around them.

The sky was carved into 88 fixed regions, and the process took eight years

In 1922, at its very first General Assembly in Rome, the newly formed International Astronomical Union settled on a list of 88 constellations covering the entire sky, each with a standard three-letter Latin abbreviation. That gave every constellation a name and a code, but not yet a shape with an edge. In 1925, at the second General Assembly in Cambridge, Belgian astronomer Eugène Delporte, working with Louis Casteels, proposed a rigorous system of boundaries drawn using lines of right ascension and declination. The IAU handed him the job of actually doing it.

Delporte worked alone between October 1925 and September 1927, circulating four progress reports as he went. His boundaries were approved at the Leiden General Assembly in 1928 and published in full in 1930, in a book called Délimitation Scientifique des Constellations with an accompanying star atlas, both in French. That's the real timeline: 1922 for the list, 1925 for the proposal, 1928 for the approval, 1930 for the finished, published borders. It took eight years to go from naming the sky to actually fencing it.

The lines zig-zag on purpose

If you look closely at a real constellation boundary map, the edges are not clean rectangles. They jog and step in odd little jags, and that's not sloppiness, it's a design constraint. Delporte was required to keep every star that already had a variable-star designation inside the constellation it had already been assigned to, even where that meant drawing an irregular notch rather than a straight line. The boundaries had to respect a century of prior labeling, not just carve up the sky from a blank slate.

That constraint also settled a genuinely messy problem. Before 1930, the same star could belong to different constellations depending on which atlas you consulted. Ptolemy's own Almagest described the star we now call Alpha Andromedae as shared between Andromeda's head and Pegasus's navel, and Beta Tauri as shared between Taurus's horn and Auriga's foot. Delporte's boundaries ended the ambiguity for good: Andromeda kept the head star, Taurus kept the horn.

A constellation is now a coordinate lookup, not a pattern match

The IAU states this about as plainly as an institution ever states anything: constellations, which started out as imaginary patterns of stars, became scientifically defined areas of sky, and when astronomers say something is in a given constellation, they mean it is simply within those defined boundaries, not that it is part of any star pattern. No stick figures or artistic outlines are formally defined by the IAU at all. Whether a star, a galaxy, or a stray asteroid is "in" Orion is a pure question of whether its right ascension and declination fall inside Orion's drawn polygon.

This is worth sitting with because it changes what the constellation figures on the Real Sky map actually are. The map draws 88 clickable stick-figure lines, one per constellation, but it does not draw the IAU's boundary polygons anywhere. The figures are there to help you recognize a shape and click into its lore card; the constellation itself, in the technical sense, is the invisible fenced region around that figure, not the figure itself.

One constellation is drawn in two disconnected pieces

Serpens, the Serpent, is the one genuine exception to "each constellation is one contiguous region." Delporte solved the problem of a snake that Ophiuchus's figure runs straight through by splitting it in two: Serpens Caput, the head, sits west of Ophiuchus, and Serpens Cauda, the tail, sits east of it. The IAU still counts this as a single constellation, not two, and on the sky map it's keyed once and its two drawn halves both resolve to the same lore entry. It's the only one of the 88 built this way, and its neighbor Ophiuchus has its own, separate reputation problem worth reading on its own: the constellation that supposedly ruined everyone's horoscope.

Why the borders look crooked on a modern chart

Delporte drew every boundary line as an arc of right ascension and a parallel of declination for the epoch 1875.0, chosen specifically to match the epoch Benjamin Apthorp Gould had already used for his own southern-hemisphere boundary work a few decades earlier, so the two hemispheres would form one consistent system. A boundary drawn as a clean vertical or horizontal line against the sky as it looked in 1875 is no longer vertical or horizontal against the sky as it looks today, because the whole coordinate grid slowly drifts under precession. The skew is real, it increases with time, and it's a direct, honest consequence of freezing a coordinate system to a fixed moment nearly 150 years in the past.

Roughly half the sky's figures have no ancient myth at all

Of the 88, exactly 50 carry a genuine Greco-Roman myth. The other 38 were invented outright in the 16th through 18th centuries, mostly by four people: Nicolas-Louis de Lacaille, who filled the southern sky with instruments of Enlightenment science; Petrus Plancius, a Dutch cartographer; Pieter Dirkszoon Keyser and Frederick de Houtman, who charted the southern stars from a ship's crow's nest; and Johannes Hevelius, a Danzig brewer-turned-astronomer who insisted on naked-eye measurements even though he owned telescopes. Those aren't lesser constellations. They're a completely different, more recent, and in some ways more traceable kind of story, one about real people naming real skies they had just seen for the first time.

What this means for you tonight

Open the sky map, turn on the constellation and names toggles, and pick one figure you don't already know. Click it, and look specifically at whether its lore card gives you a Greco-Roman myth or a modern origin note, because that split is not evenly distributed and it changes what kind of story you're actually reading. If it's a modern one, you're not missing an ancient legend, there simply isn't one to find, and the real story is a specific 16th to 18th century observer looking at a sky nobody in Europe had charted before. Then check its best month against tonight's date. If they're close, you're looking at that constellation near its best moment of the whole year, riding high around local midnight rather than skimming the horizon at dusk.