Subrahmanyan Chandrasekhar, known to his friends and colleagues as Chandra, opened the door to our understanding of the death of stars. He was the first to calculate the possible final states of stars that have used up their supplies of energy. He did so in 1930, when he was a graduate student travelling by ship from his home in India to study at the University of Cambridge, UK.

Even before he got to Cambridge, Chandrasekhar knew more about relativity and quantum mechanics than most of his teachers. He knew how to take account of both when building mathematical models of cold stars that had stopped shining. On board the ship, once he he had finished his calculations, he came to a startling conclusion: he found that there exists a critical mass, now known as the Chandrasekhar limit, beyond which no cold star made of ordinary matter can exist. He calculated this critical mass, and found that it is a few times the mass of the Sun, the exact value depending on the chemical composition of the star.

When a star of less than the critical mass has used up all its fuel, it will slowly radiate away its energy and cool down to reach a state described by one of the models he had calculated. But once a star with greater than the critical mass has used up its fuel, it cannot cool down gradually and die quietly. It must either change into some totally different form of matter, or end its life in a violent collapse and explosion.

When Chandra discovered the critical mass, he had no idea what the ultimate fate of a massive star should be. He opened the door to understanding by raising the question: what happens to a massive star when it runs out of fuel and has no way to cool down?

Star performer: Subrahmanyan Chandrasekhar's insight helped to revolutionize astronomy. Credit: BETTMANN/CORBIS

The efforts of a whole generation of astronomers were needed to find the answer to Chandra's question, starting with Fritz Zwicky's observations of supernovae in the 1930s and ending with the identification of stellar-mass black holes using X-ray telescopes in space in the 1960s. We now know that stars with a mass greater than the Chandrasekhar limit mostly die in catastrophic explosions, which we call supernovae, leaving behind collapsed cores which may be either neutron stars or black holes. Chandra's question led the way to the modern view of the Universe as a dynamic arena dominated by violent events.

When Chandra arrived in Cambridge in 1930, his mentors had no inkling of the revolution that his question was to bring about. Chief among his mentors were Arthur Eddington and Edward Milne, two world-famous astronomers who thought they knew everything worth knowing about stars. Each of them had a private theory of the Universe that was incompatible with Chandra's calculation. They ignored his arguments and declared publicly that his conclusions were wrong.

But Chandra had a cool head. He published his work in reputable astronomical journals and waited for the next generation of astronomers to recognize its importance. He stayed in Cambridge for seven years and remained on friendly terms with Eddington and Milne. After their deaths many years later, he wrote warm and sympathetic memorial lectures for each of them.

Once I went for a long walk with Chandra in the woods around Princeton and listened to him talking about his friendships. His love and admiration for Eddington and Milne were genuine. He saw them clearly, on the one hand as misguided fools, and on the other hand as human beings of rare quality, worthy of honour and respect.

In 1937 Chandra moved to the University of Chicago, where he worked until his death in 1995. His output of research followed a regular pattern. At the beginning of each decade, he chose a fresh field of study. Then he wrote a series of papers solving the outstanding problems in that field. At the end of the decade he published a magisterial book, summarizing his results and presenting the whole field in a new and clearer light. He worked in each of six fields in turn: in his third decade, he worked on the structure of dying stars; in his fourth on the transport of radiation through stellar atmospheres; in his fifth on instabilities of fluid motions; in his sixth on Einstein's general theory of relativity; in his seventh on the theory of black holes; and in his eighth on a detailed historical study of Newton's Principia Mathematica.

Everything that Chandra did was done with elegance and style. He reached a deep understanding of the mathematical and physical properties of black holes, those objects of perfect symmetry that he saw as the crowning beauty of the Universe, a beauty to which Eddington and Milne and even Einstein had been blind. His book about black holes displays his unrivalled mathematical skill as well as his impressive command of the English language.

In his eighth decade, his first great discovery, the Chandrasekhar limit, was recognized with the award of a belated Nobel prize. His last book Truth and Beauty is a collection of meditations about the place of beauty in science, including a critical comparison of Newton with Shakespeare and Beethoven, and ending with eloquent tributes to his old enemies Eddington and Milne.