I wouldn't call the American Astronomical Society conference the world's greatest party scene, but it's a lot more fun than many people think. As if getting to talk all day long about the fate of the universe or the search for habitable planets weren't enough, there are the actual parties, both semi-official and off-the-grid. The conference naturally generates lots of news, such as the possible binary supermassive black hole I wrote about a couple of weeks ago. But for me the real action is in the broad trends that don't make the front pages, but underlie everything that does.

If I had to pick a single theme that defined this past meeting, it would be galaxies. In one of the keynote talks, Sandy Faber of the University of California at Santa Cruz captured this spirit. She compared astronomers' understanding of galaxies today to their understanding of stars half a century ago. Back then, astronomers had an aha moment when they connected the observed patterns of star properties to the new theoretical understanding of nuclear fusion. After thousands of years of creative explanations involving sky chariots and battling gods, humans finally understood why the sun shines and where the chemical elements in our bodies ultimately come from.

Today, astronomers are doing the same for galaxies: connecting observed patterns of galaxy properties to new theoretical understanding. The theory involves the workings of gravity in systems with billions of moving parts. In the emerging synthesis, galaxies start off as clumps of dark matter of a range of sizes. Nobody knows what the dark matter is, but for these purposes, it makes no difference. All that's important is that dark matter is dead-simple: it moves around, exerts the force of gravity, and doesn't do much of anything else. The clumps grow by colliding and merging. Gas pools within them. The gas is what creates the visible galaxy -- the luminescent ball or pinwheel of stars.

For reasons best known to themselves, astronomers call the clumps "halos," like a shining, dainty ring above Simon Templar's head. In fact, they're not rings, they don't shine, and they've not above anyone's head. They're dark balls that encase the visible galaxy and outweigh it by 10 to 1. If anything, it's the visible galaxy that is the shining, dainty part.

In her talk and a paper released the same day, Faber proposed a simple law that does a tantalizingly good job of explaining galaxies' huge variety: When a growing halo reaches a critical mass, stars begin to form in it, and when the halo exceeds an upper mass threshold, stars stop forming. In other words, a galaxy is either "on" or "off" -- either forming stars or not. Like a house party, it doesn't get going until enough people show up, and all the cool kids bail out when it gets too crowded.

The value of the critical mass isn't well established, but might be about 10 million times the mass of the sun. Anything smaller than that lacks the gravity to hold onto its gas against the forces that disperse it, such as supernova explosions. The upper threshold is thought to be about a trillion times the mass of the sun. The reason for this threshold is still obscure, but might have to do with the sheer strength of a trillion sun's worth of gravity, which stirs up gas clouds too much for them to form stars. Black holes may also muck things up: a giant galaxy has a proportionately giant black hole at its core, which may blow the whistle on star formation -- almost literally: the hole blows gas out of the galaxy and chokes off star formation.

Faber's simple on-off law explains some seemingly contradictory observations. First, as far back in time as astronomers can see, they see giant galaxies. How can that be, if halos start small and build up? According to Faber's law, the answer is the critical mass. Some galaxies just got lucky to start off big, and we naturally see them first, before the small ones have grown enough to light up.

Second, if small halos are the basic building block of the galactic universe, why do astronomers see so few little galaxies? The theory predicts that hundreds of small galaxies should buzz around our Milky Way like angry bees, yet astronomers have found only a couple of dozen. According to Faber's law, that's because the small fry haven't switched on yet. Give 'em time. On this week's Sci Am podcast, Josh Simon of Caltech elaborates on the puzzling absence of these small galaxies. So much of astronomy focuses on the biggest, brightest, and heaviest, but some of the biggest remaining questions about galaxies concern the little ones.

A number of other talks at the conference fleshed out the synthesis. Once star formation turns on, it burns with a nearly constant brightness until it turns off. Shardha Jogee of the University of Texas described how even starbursts -- periods of frenetic star formation, often triggered by galaxy collisions -- represent only a modest step up from a galaxy's normal star formation rate. Robert Kennicutt of the University of Arizona said that there seems to be a simple law, applying to all galaxies under all conditions, relating the rate of star formation to the density of gas. No one knows why. Usually scientists say, "Well, it's more complicated than that…." But here's a case where it's simpler than people expected.

In my next blog entry, I'll discuss one of the puzzling aspects of galaxies that emerged at the meeting.