Can a star change its colour in the blink of an eye? The theory of stellar evolution says no, but Sirius has yet to be explained.
Do stars always stay the same colour? Astronomers know this is not the case, but most people do not even know that stars have colours. In fact, it takes only a little perseverance to see. Take Orion, for instance, a great constellation that rises early at night in September with its distinctive belt of three stars and two bright stars on each side. The first, Betelgeuse, which sits to the north of the belt, is obviously red; the second, Rigel, south of the belt, is blue.
Stars have different colours, but can a star change colour before your eyes? According to the modern theory of stellar evolution, a star can change colour only very slowly, except for supernovae and novae, when it may disappear in a gigantic explosion.
Now for our story. Sirius, the brightest star in the sky, is located roughly southeast of Orion's belt and has been the focus of much interest since the dawn of humanity. Claudius Ptolemy, a Greek astronomer living in Alexandria, mentioned in the star catalogue for which he was famous which stars were red in colour, such as Betelgeuse, Aldebaran, Antares, Pollux, Arcturus andI Sirius. All the others are definitively red, but if one observes the September night sky, Sirius appears brilliant white. What has happened? I came across this enigma more than 20 years ago in an article by Kenneth Brecher of the Massachusetts Institute of Technology in the book Astronomy of the Ancients , and I thought it worthy of more attention.
Could a change in the colour of Sirius reveal a severe flaw in the theory of stellar evolution? Like many others before me, I was dubious about the case at first. Could it not be simply a wrong transcription or a misinterpretation of Ptolemy? Some independent records might exist, but where? I immediately thought of China, the greatest ancient astronomical civilisation, with records spanning thousands of years.
A contract between my institute, the French Atomic Energy Institute and the Chinese Academy of Science took me to Beijing for one year in 1988. With the help of Chinese historians, we discovered a clear historical record that says Sirius changed colour about the time of Ptolemy.
As a specialist in binary stars, I thought of a new hypothesis. What if an as-yet-undetected small star was orbiting Sirius in a very elliptical orbit, similar to those of the comets around the Sun, and it caused severe disturbances each time it passed close to the bright star? Sirius is already known as a double star, but its detected companion, Sirius-B, is too distant to cause such perturbation. I realised that nothing else was known because Sirius is so bright it dazzles the big telescopes needed to see faint stars.
The chase was on. Funded by the European Southern Observatory, we took our first image in 1986 in Chile using a special mask called a coronograph to block the light from the bright star and reveal for the first time nearby fainter stars. Our second image was taken after a 15-year interval, in 1999, at an observatory in the French Pyrennees. Our idea was to follow the movement of Sirius through the sky to search for a small star going along with it. The analysis of the two images has yielded a precise view of Sirius's motion but so far has failed to detect a good candidate.
This year we will use a new technique called "adaptive optics" at the ESO to search for our star in the immediate vicinity of Sirius. If we do not find it there, astronomers will have to worry about stellar evolution, and the Sirius mystery will deepen.
Jean-Marc Bonnet-Bidaud is an astrophysicist, Service d'Astrophysique, French Atomic Energy Institute.