The discovery in the distant past of massive galaxies containing more than 10 billion stars has astrophysicists scratching their heads over how such large objects could have formed so early.

Two teams writing today in Nature provide evidence for large, fully developed galaxies at the farthest reaches of the cosmos. The light from these objects takes time to reach the earth, thus astronomers are seeing them as they looked some 10 billion years ago. Because the big bang is believed to have occurred nearly 14 billion years ago, these giant galaxies--some as big as the largest present-day galaxies--must have completed most of their growth before the universe was a quarter of its age.

This early formation challenges the popular hierarchical model of galaxy evolution, which assumes that the first galaxies to appear were relatively tiny. According to this hypothesis, only through the merging of these smaller entities did larger galaxies slowly develop. The model does not predict the four massive galaxies that Andrea Cimatti of the Italian National Institute of Astrophysics and his colleagues observed with the Very Large Telescope in northern Chile. The assumed age of these mature galaxies (one of which is located at the center of the image above) implies that they formed when the universe was just two billion years old.

In a separate survey using the Gemini telescope in Mauna Kea, Hawaii, Karl Glazebrook of Johns Hopkins University and his collaborators discovered even more evolved galaxies over a broad time period. Although the researchers did see a decrease in the number of these galaxies as they looked farther back in time, the observed drop-off was not as rapid as the predicted one.

Because the hierarchical model has been so successful at explaining other observations, such as how galaxies cluster, scientists are unlikely to abandon it. Instead, they may rethink their understanding of how gas turns into stars. "Some new ingredient is required to make more stars form earlier in the big galaxies," Glazebrook comments. "But what that ingredient is, we don't yet know." --Michael Schirber