Distant galaxy formed stars only 250 million years after the Big Bang

Little is known about the star-birth activity of the earliest galaxies. Observations of a particularly distant galaxy provide evidence for such activity when the Universe was just 2% of its current age.
Rychard Bouwens is at the Leiden Observatory, Leiden University, Leiden NL-2333, the Netherlands.

Search for this author in:

Astronomers have only a limited understanding of star and galaxy formation in the first 300 million years of the Universe’s history1,2. Two important processes would have been accretion, whereby objects increase in mass by gravitationally attracting nearby matter, and the cooling of gas. But the overall picture lacks the sensitive observational data required to inform or guide its construction. In a paper in Nature, Hashimoto et al.3 present observations of an extremely distant galaxy, and report that star-birth activity began there just 250 million years after the Big Bang. The authors’ results suggest that future telescopes could detect such early episodes of star formation in similar galaxies.

The first stars in the Universe are thought to have formed in regions with high densities of matter1,2. These regions grew over cosmic time by accretion, and eventually developed into galaxies. From a theoretical standpoint, there is great interest in establishing observationally when the Universe had its first major star-birth activity, but current constraints are poor. One way in which to investigate the onset of star formation is to detect starlight from extremely distant galaxies. Because light travels at a finite speed, observations of the distant Universe act as a time machine, allowing us to look back into the past.

The galaxy probed by Hashimoto and colleagues is one of the farthest known objects from Earth for which light can be detected (Fig. 1a). It was discovered4 in 2012, and was dubbed MACS1149-JD1. What the authors have added to this discovery is a precise measurement of the galaxy’s redshift. The redshift of a light source tells us the factor by which the Universe has expanded since the source emitted its light, as well as the distance to the source and the time at which the light was released.

Figure 1 | Observations of MACS1149-JD1. a, Hashimoto et al.3 report that star formation in the distant galaxy MACS1149-JD1 (indicated by the box) began when the Universe was only 2% of its current age. Other sources of light in the image are nearby galaxies. Scale bar, 2 arcseconds (one arcsec is 1/3,600th of a degree). b, The authors’ result relied on a detailed study of an emission line in the galaxy’s spectrum that is produced by doubly ionized oxygen. The contours join points that have equal emission-line intensity. Scale bar, 0.5 arcsec. (Adapted from Extended Data Fig. 2a of ref. 3.)

Hashimoto et al. determined the redshift of MACS1149-JD1 by studying the properties of an emission line in the galaxy’s spectrum. They report a redshift of about 9.11, which implies that the galaxy is being viewed as it was when the Universe was roughly one-tenth of its current size and about 550 million years old. This is the highest redshift ever inferred from a spectral line5, and is only slightly lower than a redshift reported by using a broader, possibly less robust, spectral feature6.

The authors’ measurement was made possible thanks to the continuously improving capabilities of the US$1.4-billion Atacama Large Millimeter/submillimeter Array (ALMA) observatory in Chile. ALMA studies of galaxies with redshifts greater than 4 frequently use an emission line that is produced by singly ionized carbon79. By contrast, Hashimoto and colleagues relied on a line that is associated with doubly ionized oxygen (Fig. 1b). This line was shown to be readily detectable in previous theoretical and observational work using ALMA10,11. The authors’ result showcases ALMA’s capabilities as a tool for precisely measuring the redshifts of distant galaxies, as was also illustrated earlier this year7.

Next, Hashimoto et al. considered the optical colours observed in MACS1149-JD1 by NASA’s Hubble and Spitzer space telescopes. Such colours provide clues about the number of stars that formed early in a galaxy’s lifetime4,12,13. The authors show that the colours represent a substantial episode of star formation in the galaxy when the Universe was only 250 million years old. The authors’ precise redshift measurement was essential in arriving at this conclusion because it allowed them to rule out the possibility that the colours arise instead from strong recombination lines — spectral features that are commonly associated with the intense ionizing radiation produced by hot stars in the earliest galaxies1416.

However, MACS1149-JD1 is just one galaxy, and it remains unclear whether such early star-birth activity occurred in other galaxies. Observations of the cosmic microwave background — the relic radiation from the Big Bang — by the Planck Collaboration17 indicate that there was less star formation in the early Universe than the present authors’ results suggest. In addition, measurements of the prevalence of galaxies at similar epochs to that of MACS1149-JD1 suggest a lack of star-birth activity at these early times18. Nevertheless, a high rate of star formation in the early Universe could explain the discovery earlier this year of an unexpectedly large absorption signal in the spectrum of the cosmic microwave background19.

Hashimoto and colleagues have not only set a record for the highest redshift inferred from a spectral line, but they also did it using ALMA, which is a first for the facility. The possibility that the galaxy they observed had substantial star formation at early times is intriguing. Their discoveries seem certain to inspire similar studies of other galaxies in the distant Universe, and provide fuel for observations using the future James Webb Space Telescope.

Nature 557, 312-313 (2018)

doi: 10.1038/d41586-018-05114-z
Nature Briefing

Sign up for the daily Nature Briefing email newsletter

Stay up to date with what matters in science and why, handpicked from Nature and other publications worldwide.

Sign Up


  1. 1.

    Loeb, A. & Barkana, R. Annu. Rev. Astron. Astrophys. 39, 19–66 (2001).

  2. 2.

    Abel, T., Bryan, G. L. & Norman, M. L. Science 295, 93–98 (2002).

  3. 3.

    Hashimoto, T. et al. Nature 557, 392–395 (2018).

  4. 4.

    Zheng, W. et al. Nature 489, 406–408 (2012).

  5. 5.

    Zitrin, A. et al. Astrophys. J. Lett. 810, 12 (2015).

  6. 6.

    Oesch, P. A. et al. Astrophys. J. 819, 129 (2016).

  7. 7.

    Smit, R. et al. Nature 553, 178–181 (2018).

  8. 8.

    Capak, P. L. et al. Nature 522, 455–458 (2015).

  9. 9.

    Willott, C. J., Carilli, C. L., Wagg, J. & Wang, R. Astrophys. J. 807, 180 (2015).

  10. 10.

    Inoue, A. K. et al. Astrophys. J. Lett. 780, 18 (2014).

  11. 11.

    Inoue, A. K. et al. Science 352, 1559–1562 (2016).

  12. 12.

    Bradač, M. et al. Astrophys. J. 785, 108 (2014).

  13. 13.

    Hoag, A. et al. Astrophys. J. 854, 39 (2018).

  14. 14.

    Roberts-Borsani, G. W. et al. Astrophys. J. 823, 143 (2016).

  15. 15.

    Smit, R. et al. Astrophys. J. 784, 58 (2014).

  16. 16.

    Laporte, N. et al. Astron. Astrophys. 562, 8 (2014).

  17. 17.

    Planck Collaboration. Astron. Astrophys. 596, 107 (2016).

  18. 18.

    Oesch, P. A., Bouwens, R. J., Illingworth, G. D., Labbé, I. & Stefanon, M. Astrophys. J. 855, 105 (2018).

  19. 19.

    Bowman, J. D., Rogers, A. E. E., Monsalve, R. A., Mozdzen, T. J. & Mahesh, N. Nature 555, 67–70 (2018).

Download references