Last week, I talked about results from the American Astronomical Society conference early this month. For me, the biggest news from the meeting wasn't something that gets headlines. It was a quiet revolution: the emerging synthesis for explaining the incredible diversity of galaxies. Ironically, though, one of the first shots of this revolution has been called into question.


Several years ago, astronomers found intriguing patterns in the properties of galaxies and the supermassive black holes at their cores. Nearly every galaxy seemed to have such a black hole. The only exceptions were completely flattened galaxies. To host a giant black hole, a galaxy had to have a ball-shaped region in its middle (a so-called bulge). The biggest holes lay in elliptical galaxies, which are purely bulges, without any outer flattened area at all.

Quantitatively, the mass of each black hole was about 0.2 percent of the mass of the bulge that hosted it -- a rule that practically acquired the status of a new law of nature. What caused the sizes of hole and bulge to become so closely intertwined? Some astronomers thought the black hole came first and catalyzed the formation of the bulge; others argued that the stars in the bulge came first and then agglomerated into a giant hole; and still others thought the hole and bulge arose in unison.

Since then, it has emerged that the pattern isn't quite what astronomers thought. To begin with, the mass of the hole turned out to be correlated not with the mass of the bulge so much as with the velocities of the stars in the bulge. Now astronomers have found plenty of bulgeless galaxies with giant black holes. At the astronomy conference, Shobita Satyapal of George Mason University listed eight of them. The black holes had eluded detection because the galaxies were choked with dust, but the Spitzer infrared space telescope saw right through the muck.

Maybe the black holes formed first after all, in which case these counterexamples are bulges in the act of forming. Satyapal also suggested an interesting alternative. Within the Spitzer sample, the eight galaxies had above-average amounts of dark matter, suggesting that black holes aren't related to the bulge per se, but to the dark matter. If so, the earlier bulge-hole patterns were merely a proxy for this dark-matter effect.

Even leaving aside the black holes, the fact that some galaxies -- a full fifth, by some studies -- don't have bulges is weird. Bulges are an inevitable byproduct of large-scale galactic collisions. Somehow those bulgeless galaxies have been peculiarly successful at galactic dodgeball.

For instance, our own Milky Way galaxy has a substandard bulge. In separate talks at the conference, Jason Kalirai of UC Santa Cruz and Thomas Brown of the Space Telescope Science Institute discussed how our galaxy doesn't seem to have undergone a big collision for the past 10 billion years -- a much less violent history than the neighboring Andromeda galaxy.

Even odder, Elizabeth McGrath of UC Santa Cruz showed Hubble Space Telescope pictures of the most massive galaxies -- about 10 times the mass of the Milky Way -- in the ancient universe. Most were highly flattened, so they must not have undergone any major collisions. How did they bulk up, then? McGrath suggested they didn't assemble from smaller pieces. Instead, a titanic gas cloud went Whoosh! and collapsed down to a galaxy all at once. Astronomers' understanding of galaxies remains an exhilarating mix of order and enduring mystery.