Milky Way's last major merger: new clues about galaxy formation thicken the plot.

Orbiting above and below the bulged center of the Milky Way is an elongated, wispy contingent of stars. It could be considered a mere footnote amidst the galaxy's stellar hordes, but astronomers are reading it as if it were an entire chapter early in the galaxy's autobiography. That's a surprising view, given that these stars didn't even start out in the Milky Way. Discovered over the past year, the orbiting contingent appears to be the tattered remnants of a galaxy that collided with ours billions of years ago. If this were confirmed, the intruder galaxy would have had about 15 percent of the stellar mass of our galaxy during that early epoch. So, it packed enough of a wallop to change the overall shape of the Milky Way. This collision about 10 billion years ago was probably,the most consequential event in our galaxy's 13-billion-year history. Since then, the galaxy has enjoyed a relatively peaceful existence.

Now, astronomers are reconstructing what happened during the Milky Way's big collision and wondering if many other galaxies across the universe have a similar history. New evidence suggests that this may be the case.

SPIRALING COMPLEXITY Astronomers have long recognized that the Milky Way contains a central bulge surrounded by a flat disk with spiral arms ablaze with newborn stars. An ancient spherical halo envelopes it all.

The cosmic collision jumbled all the stars that were then in the disk, says Gerard Gilmore of the University of Cambridge in England. Our galaxy's orderly spiral shape was destroyed. "It would have looked very irregular and distorted, like the sort of galaxies one sees in the Hubble Space Telescope images of the early universe," Gilmore says.

Over the next billon years, he calculates, the gas already in the Milky Way, along with the material added by the intruder galaxy, cooled down and formed new stars. This mix of gas and young stars reassumed the familiar spiral geometry.

But the collision had a permanent effect on the Milky Way, according to a new study by Gilmore, Rosemary F.G. Wyse of the Johns Hopkins University in Baltimore, and John E. Norris of the Australian National University in Canberra. When the collision heated and puffed up our galaxy's flat disk, it created a distended, tenuous fog of gas and stars--the so-called thick disk--that persists to the present day.

The collision also seems to have marked the end of the era of frequent impacts. According to an analysis by Gilmore and his collaborators, that ancient smackdown was the last major collision the Milky Way has endured, and the creation of the thick disk was the last significant change in the galaxy's structure.

If Gilmore and his colleagues are right, the Milky Way's evolution could provide a case study for cosmologists exploring galaxy formation throughout the universe. According to a well-accepted scenario, known as the hierarchical model of galaxy formation, galaxies start out small and grow bigger by capturing material from galaxies that collide with them.

Evidence of such collisions in faraway galaxies has been building for years, but to get a close look at the aftermath of such an event, scientists needed to find remnants of one in our own galaxy. That's why astronomers are excited about the newfound string of stars orbiting above and below the Milky Way's plane.

The finding that the galaxy suffered its last major impact about 10 billion years ago places new constraints on any model of galaxy formation, says theorist Martin D. Weinberg of the University of Massachusetts in Amherst. Adds Gilmore: "If we can understand the formation of one apparently typical galaxy--the Milky Way--that is a big step towards understanding them all."

CONDENSED MILK It was nearly 2 decades ago that Gilmore and a colleague, Neil Reid, both based at the University of Edinburgh in Scotland, found evidence that something was missing in the standard picture of the Milky Way. When they counted the galaxy's stars, the pair found that there were far more than the known components of the Milky Way could account for. To explain the discrepancy, the two scientists suggested that the galaxy's disk has a second part--a thicker, diffuse one extending about 3,000 light-years above and below the familiar flat disk. They calculated that this thick structure contains 5 to 10 percent of our galaxy's stars.

Over the next few years, as astronomers measured the age, composition, and distribution of these stars, they confirmed the existence of the so-called thick disk. In 1985, Wyse and Gilmore proposed that a massive intergalaxy collision created this structure.

The energy absorbed by the impact would have heated the Milky Ways thin, primordial disk, causing it to swell to its present proportions. The available evidence puts that impact in the distant past. May, the thick disk is composed only of stars that are at least 10 billion years old. In contrast, the thin disk contains stars of all ages, from newborns to ancients that are 13 billion years old. "If the thick disk were produced by later merging events, we would expect to see younger stars from the thin disk puffed up into the thick disk," explains Kenneth Freeman, also of the Australian National University in Weston.

An early collision between our galaxy's flat disk and an intruder galaxy plausibly accounts for the extra stars that Gilmore and Reid first detected in the early 1980s, but the researchers lacked direct evidence for their model.

Two years ago, Gilmore, Wyse, and Norris went hunting for the remnants of that ancient collision. Using a spectrograph on the Anglo-Australian Telescope in Coonabarabran, which can examine 400 objects simultaneously, the researchers measured the velocities of a group of 2,000 stars in the thick disk.

Orbital velocity is an important diagnostic tool, notes Gilmore, because "it's one of the very few conserved quantifies" in the aftermath of a collision.

The team found that the stars' velocities--on average about 100 kilometers per second--didn't match those of any known population of stars in the Milky Way's disk or surrounding halo. Finding a batch of stars with a unique velocity is a sign that another galaxy had infiltrated our own, Gilmore says. Moreover, the researchers report in the July 20A strophysical Journal, this velocity is just what would be expected of stars from an intruder galaxy.

The researchers add that the narrow spread of stellar velocities indicates that the thick disk was generated by a single collision.

The team's results are "interesting but not very secure yet," Freeman says. He notes that it's difficult to distinguish by velocity alone a grouping of stars in the thick disk from the plethora of stars found in the Milky Ways enveloping halo. Showing that the chemical composition of the stars studied by Gilmore and his colleagues differs from that of those in the halo would make a more convincing case that a separate galaxy indeed had intruded, he says.

DODGING PUNCHES The findings by Gilmore's team raise a new question: Why hasn't the Milky Way been struck by another massive galaxy, during the past 10 billion years? Weinberg says that it could be mere happenstance. He notes, for example, that there are no large galaxies nearby. If the Large Magellanic Cloud, a massive satellite galaxy of the Milky Way, resided closer, the Milky Way would probably have been pummeled in the more recent past.

Further, says Weinberg, collisions are but one element of a galaxy's life story. Theorists now see galaxy evolution as a two-phase process. Early in the universe, when galaxies were closer together, collisions were commonplace. Then, after about 7 billion years of cosmic expansion, collisions became less frequent. In this second phase of evolution, galaxies have accumulated matter more slowly and less dramatically by gravitationally coaxing material from its near neighbors.

Many observers argue that the first, collisional era was the most critical for generating the structure of galaxies. And for the Milky Way, at least, there's no evidence that its most recent acquisitions, such as its capture of the Sagittarius dwarf galaxy, has significantly altered our galaxy's structure (SN: 4/9/94, p. 228).

Astronomers have recently found evidence that thick disks may be routine features of spiral galaxies. In a study of 47 spiral galaxies that span a wide range of masses, Julianne J. Dalcanton of the University of Washington in Seattle and Rebecca A. Bernstein of the University of Michigan in Ann Arbor found that all of them have thick disks. The astronomers note that they can't study these thick disks with anywhere near the detail that Gilmore has uncovered in the Milky Way. But the red hue of these disks suggests they are composed of old, red stars--just as the thick disks of our galaxy--the researchers reported in the September Astronomical Journal.

Dalcanton cautions that the color data are insufficient for accurately pinning down the age of thick disks. She also cautions that the galaxies she and Bernstein examined may undergo fewer collisions than the average spiral galaxy.

Nonetheless, if these thick disks were created during a collision--the way Gilmore's team proposes our own galaxy's thick disk arose--then their home galaxies have also been free of major collision for billions of years.

"When people have constructed galaxy-formation models, they've accounted for the bulge of galaxies and the halo and the thin disk," Dalcanton says. The detection of so many thick disks could prove to be an important step forward in deciphering the puzzle of how galaxies form, she says.