It's well known that the Milky Way is a spiral galaxy, a swirl of stars in an extended, many-armed disk. But the structure of the galaxy is far from two-dimensional. Above and below those familiar spiral arms is a lesser-known feature, a spherical swarm of stars that makes up a halo around the disk.

For decades the presence of the halo has prodded astronomers to ask big questions about its nature: How is it structured? How do stars in the halo compare with disk stars such as our sun, or to stars elsewhere in the halo? And just how did the halo get there? In recent years a group of astronomers has suggested an answer to some of those big questions by drawing on a large telescopic survey of the sky.

The halo, they have concluded, is composed of at least two distinct populations of stars, with different chemical makeups and different orbits. One group of stars, dubbed the inner halo, generally orbits closer to the galactic center, and its members tend to contain more heavy elements such as iron than do stars farther out. (Halo stars as a whole are depleted in these heavy elements, relative to stars in the galactic disk.) Stars of the outer halo occupy somewhat wider orbits around the galactic center, contain lower levels of heavy elements, and—unlike the inner halo—tend to follow retrograde orbits, circling the Milky Way in a direction counter to the rotation of the galactic disk.

"We don't think it's just one halo," says Timothy Beers, an astronomer at the National Optical Astronomy Observatory and Michigan State University, who was lead author on a recent study in The Astrophysical Journal. Beers, Daniela Carollo of Macquarie University in Australia and their colleagues based their analysis on data from the Sloan Digital Sky Survey, a long-running telescopic campaign based at Apache Point Observatory in New Mexico. "We advocate the position that we are looking at a minimum of a dual halo," he says.

As the Milky Way built up by accretion of smaller galaxies, the inner and outer halo would represent two different epochs of galactic assembly. "We actually think that the formation scenario was something you could describe as a multiphase assembly," Beers says. The inner halo would represent the remnants of relatively massive dwarf galaxies, which coalesced early on. Lighter-weight galaxies would have attached themselves later on in a very gradual agglomeration to form the outer halo.

The inner and outer halo are not cleanly divided, but the differences in how the two populations move could aid astronomers in finding extremely primitive stars, which contain primarily hydrogen and helium. Those were the raw materials for the first generation of stars, early in the history of the universe; subsequent generations contained heavier elements that were fused in stellar cores and supernovae and then released into interstellar space. "Knowing that you have this dichotomy helps direct us to finding these interesting low-metallicity stars," Beers says. Outer-halo stars could be identified for detailed study by their distinctive motions on the sky. "Those are the ones that tell the story of how the universe built its elements," Beers says.

But not everyone agrees that the facts support the dual-halo interpretation. "I have a very relaxed opinion about single halos, dual halos, multiple halos," says astrophysicist Ralph Schönrich, a NASA Hubble Fellow at The Ohio State University. "I don't mind any idea of a dual halo. It's just that I don't see any evidence for it."

Schönrich and his colleagues published a 2011 paper, "On the Alleged Duality of the Galactic Halo." In typical fashion for astrophysics, where drafts of research papers often appear online months before going to press in peer-reviewed journals, the back-and-forth between Beers's group and Schönrich's has taken a dizzying, seemingly time-bending course, in which one paper refuted another that had not yet been published.

First Beers, Carollo and their colleagues published two studies, in 2007 and 2010, arguing for the dual nature of the halo. Then, later in 2010, Schönrich and his co-authors posted their rebuttal on, a preprint server in wide use by astronomers and physicists. Beers and his group responded on the arXiv in 2011 with the analysis that eventually appeared in The Astrophysical Journal. But by the time that that study went to press, Schönrich’s team had already published its updated rebuttal in Monthly Notices of the Royal Astronomical Society, including a response to the Beers group's as-yet unpublished rejoinder to the initial draft of the rebuttal. The final, published version of the Beers paper, in turn, takes aim at the final, published version of the Schönrich paper in an appendix.

The details of the Schönrich critique, and of the Beers critique of that critique (and so on), quickly veer into the technical and arcane. But one issue at the heart of the disagreement is whether the distances to stars in the proposed outer halo have been significantly overestimated. In particular, "turnoff stars," which have expended their nuclear fuel and departed the so-called main sequence of stellar evolution, are difficult to diagnose from afar. "If you say it's a subgiant and in truth it's a dwarf star, then you're making a huge mistake," Schönrich says, adding that distance estimates for turnoff stars can be off by as much as 50 percent.

With flawed distances from the sun, which itself is zooming through the galaxy, stars can be incorrectly identified as moving in a retrograde fashion. Schönrich likens this to watching trees from a car: trees just beside the road zoom past in a blur, whereas trees set back by a few hundred meters appear almost stationary. So if a passenger in a moving vehicle sees trees flying rapidly past the window, and incorrectly believes those trees to be far away, he or she would have to conclude that the trees themselves are moving backward. "The point was really that we could see some of their signatures when we used only their stars," he says. "But they went away when we took out those turnoff stars that are contaminated by distance estimates."

For their part, Beers and his colleagues contended that Schönrich's group used a flawed distance scale in their critique. (Schönrich counters that he and his collaborators used multiple distance measures, so their analysis does not rest on any one method of estimating distances.) And Beers notes that recent work in theoretical astrophysics has shown that the two-halo interpretation fits naturally with the current picture of how galaxies take shape. "If we didn't find this inner-outer halo structure," he says, "then the question before us would be, 'Why not?'"

The debate remains unsettled as to what the data from the Sloan Digital Sky Survey show, but the larger question of how the Milky Way's halo is structured may soon be resolved. The European Space Agency is planning to launch a spacecraft called Gaia in 2013 to track the motions of roughly one billion stars with exquisite precision. Once Gaia returns its data, Beers says, the structure of the halo "will be absolutely apparent—it will be clear." And forthcoming ground-based telescopic campaigns, such as the one that will be carried out using the Large Synoptic Survey Telescope in Chile, will produce floods of data on astronomical objects near and far. “With the Sloan data we're just at the ragged edge of being able to see it," he says. "With the next big surveys it will be a slam dunk."