To all other telescopes these objects have remained deeply obscured by dust, which means that with these observations ISO is revealing the hidden side of the early universe - dust and cool objects can only be seen in the infrared.

The gravitational arcs are the images of distant galaxies magnified by factors between 2 and 10. The magnification phenomenon is done by giant clusters of galaxies whose tremendous gravitational field bends the light from the objects beyond and amplifies it, just like a lens; the effect often distorts the image of the lensed object, stretching it into an arc - hence the name 'gravitational arcs'.

The present results are a victory for those who, against all pessimistic predictions, bet on ISO's sensitivity to detect gravitational arcs by infrared light. Before ISO no gravitational arc had been detected in the mid-infrared, where the emission of the young galaxies becomes so significant. At first, even trying to do so with ISO was almost considered a waste of time by many astronomers, for they believed that the sensitivity and resolution required by such a search exceeded ISO's capabilities.

Nevertheless, the ISO astronomers started their search looking through selected galaxy clusters and succeeded. The detection of a giant gravitational arc in the galaxy cluster Abell 370, plus the thirty lensed and possibly young-colliding galaxies in the galaxy cluster Abell 2390, settled the issue.

In the case of Abell 2390 it took ISO an unusually long exposure-time of 16 hours to collect the light, the sensitivity of the infrared camera on board ISO, ISOCAM, having been pushed to the limit.

"It's like getting a good hand in poker. It feels very good when you look at the last card and it's the fifth diamond", says group leader Leo Metcalfe, at the ISO Science Operations Centre in Villafranca (Madrid, Spain). "We knew what we wanted to uncover, but we also knew we could find nothing after ten years of waiting. But it worked! It is aesthetically a really nice experiment. We use this exceptional instrument (ISO) and boost its power by using galaxy clusters as magnifying glasses. We see extremely deep".

The group has already identified more lensed sources in other galaxy clusters and are hoping to use future infrared space telescopes such as ESA's FIRST (Far Infrared and Submillimetre Telescope).

A vigourous early universe

These data allow astronomers to catch a glimpse of a side of the early universe that had remained hidden up to now. During the last years suspicion has been raised that important processes taking place in the early universe have been disregarded or underestimated because the dust enshrouding them made them invisible to optical telescopes, and now there's growing evidence that indeed it has been so.

Thanks to ISO's unique ability to see through the dust, it is now clear that our view of the early universe has been so far hampered by the dust.

On the whole, ISO is unveiling a young universe much more vigourous than it was thought to be. The conjecture, based on results like those obtained in Abell 2390, is that the newly-found population of young galaxies are merging to form larger ones, a process during which strong starbursts are generated.

As Metcalfe explains, "In Abell 2390 we've found a population of objects which brighten dramatically as one moves into the infrared. Viewed in the visible they merge inconspicuously with other field objects, but at 15 microns [the so-called 'mid infrared'] they emerge as a distinctive population. This is strong evidence for hierarchical galaxy formation with a large number of dust obscured starbursts. A fairly vigourous history for the universe, and a big role played by dust in concealing it - so it was just waiting for ISO to come along".

The result is exciting as it will help to discriminate among different versions of the history of the universe, and to say something about how its structures developed. For instance, it seems to give support to the model suggesting that the current galaxy types, spiral and elliptical, are the result of gradual merging of smaller galaxies into larger and larger structures, starting from dwarf irregulars which merge to form spirals and later giant ellipticals. Knowledge on these issues must ultimately constrain models of how the universe itself formed.

Commenting on the discovery, Martin Harwit, from Cornell University (US), said: "This is an audaciously beautiful piece of work. Standing on the shoulders of giant clusters of galaxies, Leo Metcalfe and his colleagues have seen farther out into the Universe and further back in time than any infrared astronomer before. In the years ahead, the novel techniques they have developed will undoubtedly be widely used to unravel the early history of the cosmos when stars and galaxies were just beginning to form."