Press Release

Discovery of a New Gravitational Lens System

22 October 1987

The discovery of a new gravitational lens system in the southern constellation Cetus comes as a first exciting and fundamental result obtained by a group of European astronomers in the frame of the systematic search program they are carrying out at the ESO La Silla observatory. Not only did they find that the image of the highly luminous quasar (ESO8712 and ESO8713) UM673 is double, but they were also able to observe the distant galaxy that is responsible for this effect. Continued monitoring of this rare object may actually lead to cosmologically significant results about the size and the age of the Universe.

Quasars are known to be the most luminous as well as the most distant objects in the Universe. If a galaxy (or a cluster of galaxies) lies near the line of sight to a quasar, the result of gravitational bending of the light [1] may be such that more than one image of this single quasar will be seen by an observer on Earth. Such a phenomenon is referred to as gravitational lensing.

The relative positions and intensities of the multiple quasar images in the sky depend on the amount and d istribution of mass in the intervening object(s), as well as on the geometric configuration between quasar, deflector and observer. Therefore, accurate observations of such images may lead to a determination of the mass of the deflector as well as that of the relative size of the Universe.

Since the well known discovery in 1979 of the first gravitational lens system (Q0956+561 A and B), very few additional lensed quasars have been identified. Furthermore, not all reported candidates have been confirmed by subsequent, detailed observations. The present discovery at ESO is remarkable because, for the first time, a gravitational lens system has been identified by using a purely optical observational strategy, according to which the images of selected quasars are systematically obtained under the best possible seeing conditions.

The observations were carried out with the ESO/MPI 2.2m and ESO 3.6m telescopes. The data show that the quasar UM673 (the 673rd quasar catalogued during the University of Michigan survey) appears in the sky as a double object having a separation of 2.2 arcsec, and that a faint galaxy is superimposed over the two images (see accompanying photo). It is this galaxy that causes the gravitational lensing and it could only be seen after the two stellar-like images of the QSO (quasi-stellar object) had been removed by computer processing.

The conclusive evidence that the two images are of one single quasar comes from a detailed comparison of the spectra. A careful study shows that the two QSO images have exactly the same spectrum. The measured redshifts [2] are found to be identical, z(A) = z(B) = 2.72, corresponding to an apparent recession velocity of the order of 86 % the speed of light. Because of the enormous distance that separates us from the quasar, we see the latter today as it was 13 billion years ago. It was also possible to determine the redshift of the intervening galaxy as z = 0.49, indicating that it lies between the quasar and us. With this information, the mass of the lensing galaxy has been estimated as 240 billion times the mass of our Sun.

The discovery of gravitational lenses is important for at least three different reasons:

The study of gravitational lens systems should lead to an independent estimate of the amount of hidden matter in the Universe. Indeed, several lines of arguments appear to indicate that the Universe may contain as much as 10 times more matter than can presently be directly observed (the “missing mass problem"). Since the quasar light is deflected by all the mass in an intervening object, a comparison with the visible mass deduced from other observations may lead to an estimate of the invisible mass.

Another important cosmological measurement that is possible with gravitational lens systems is an independent verification of the distance scale in the Universe. This is feasible because the lengths of the light paths of the two quasar images are different. If, as expected in this type of objects, the brightness of the quasar changes with time, then the variation will first be seen in the image that corresponds to the shortest path, and later in the other. A measurement of the time delay, taken together with the known parameters of the lens, will determine the absolute size of the system, and therefore the value of the Hubble parameter H0 which indicates the expansion rate of our Universe. For the quasar UM673, the time delay for a variation in the brightness of the two images is expected to be as small as a few months. Continued monitoring of the brightness of the two images is actually being carried out at La Silla. It is hoped that variations in the two images will soon be detected which will make it possible to accurately determine the time delay and thereby, for the first time, to make a reliable, independent determination of H0 by this method.

Finally, it should be mentioned that UM673 appears to be one of the most luminous quasars because its apparent luminosity has been amplified by gravitational lensing (approximately by a factor of 10). In the systematic search for gravitational lens systems among a selected sample of the most highly luminous quasars, the present team of astronomers has now identified several additional promising candidates. Spectroscopic observations of these are under way at ESO. These findings naturally raise the following, very fundamental question: "To what extent is the observed, high luminosity of quasars really intrinsic?". Since gravitational lensing may give rise to the formation of cosmic mirages, it may be that lensing effects also deform our own view of the entire Universe.


[1] Since the total solar eclipse of 1919, when astronomers observed for the first time an apparent displacement in the position of stars near the limb of the Sun, it is recognized that light beams can be bent, not only in optical systems, but also in gravitational fields. This effect was predicted by Einstein within his general theory of relativity.

[2] In astronomy, the redshift z denotes the fraction by which lines are shifted towards longer wavelengths in the spectrum of a distant galaxy or quasar receding from us with the expansion of the Universe. The observed redshift gives a direct estimate of the apparent recession velocity, which is itself a function (Hubble relation) of the distance of the object under study.

[3] Since the total solar eclipse of 1919, when astronomers observed for the first time an apparent displacement in the position of stars near the limb of the Sun, it is recognized that light beams can be bent, not only in optical systems, but also in gravitational fields. This effect was predicted by Einstein within his general theory of relativity.

More information

Detailed accounts of the observations and interpretation of the gravitational lens system UM673 are contained in two scientific papers to appear in the British journal Nature (on 22 October 1987) and in the European journal Astronomy & Astrophysics.

The group consists of J. Surdej and J.P. Swings (Institut d'Astrophysique, Université de Liège), P. Magain (formerly in Liège, presently at the European Southern Observatory, La Silla, Chile), T.J.-L. Courvoisier (Space Telescope European Coordinating Facility, Garching near Munich, F.R.Germany), H. Kuhr (Max Planck Institut für Astronomie, Heidelberg, F.R.Germany), S. Refsdal, U. Borgeest and R. Kayser (Hamburger Sternwarte, F.R.Germany) and K. Kellermann (National Radio Astronomy Observatory, Virginia).


Richard West
Garching, Germany
Tel: +49 89 3200 6276

Connect with ESO on social media

About the Release

Release No.:eso8715
Legacy ID:PR 14A/87
Name:Abell 370, ESO8712, ESO8713, UM 673
Type:Local Universe : Galaxy : Activity : AGN : Quasar
Facility:ESO 3.6-metre telescope, MPG/ESO 2.2-metre telescope


The gravitational lens system UM673
The gravitational lens system UM673
Spectrum of the giant arc in Abell 370
Spectrum of the giant arc in Abell 370