If you find the idea of a rich cocktail of famous scientists writing clearly and elegantly about topics close to their hearts appealing, then this is the book for you. For instance, the book offers Paul Davies, Richard Dawkins, Stephen Jay Gould and Ian Stewart, and altogether it has contributions from 34 "names", many of whom have written best-sellers of their own. Unfortunately, the proportion of women writers is rather low, much lower than I would have wished.

Each essay explaining "how things are" as the writer sees the world, is short and would be accessible to anyone with a modicum of interest in science. In fact, one of the most striking features of the book is its frequent references to the experience of being a child or to the way children view the world. This might make the book most appealing to young and budding scientists.

Anthropologist Mary Catherine Bateson considers the impact of humans on nature and examines what it means to be "natural". She gives the impression that we interfere far too much with nature and expresses the desire that we "invent new forms and learn some new things: limits; moderation; fewer progeny; the acceptance of our own dying."

Evolutionary biologist Richard Dawkins, who writes his essay in the form of a letter to a child, likens scientists to detectives, the evidence they gather being good reasons for believing what they say. Bad reasons, he says, are tradition, authority and revelation. But Dawkins admits that scientists can have hunches that turn out to be right, though he says these are "not worth anything until they are supported by evidence".

ÁñÁ«ÊÓƵ

Theoretical physicist Paul Davies, known for his books on the origin of the universe, opens the section of the book devoted to origins. He asks the age-old question of why time should suddenly "switch on". He may provide few answers but the journey is fascinating. Chemist Robert Shapiro considers how life becomes organised and wonders how we can find out more, suggesting that gifted amateurs in schools and colleges might help with some simple experiments. Also in this section are Jack Cohen's essay on how a fly becomes a fly; Lewis Wolpert on how cells in the embryo know what to do; and Lynn Margulis reflecting on the facts of death. Margulis uses the example of a curd dairy drink called Kefir, a combination of 30 microbes, to explore life without sex and death; here she rides her hobby-horse of symbiosis.

Evolutionary biologist Stephen Jay Gould opens the section on evolution by claiming that our present view of evolution concentrates far too much on the concept of progress, "the idea that evolution possesses a driving force or manifests an overarching trend towards complexity". Wilford Wolpoff examines the evidence for the relationships between humans and apes, and concludes that "in terms of genealogical relations we are apes, albeit of a very special kind". Geneticist Steve Jones ponders the reasons why some people are black, and offers some suggestions about the causes of albinism. But the theories he mentions seem to me beyond scientific resolution, on a par with the ridiculous notion that homosexuality is genetically inherited.

ÁñÁ«ÊÓƵ

Cognitive scientist Daniel Dennett, opening the section on mind, offers a diffuse, if elegantly written, ramble around the advantages of making mistakes - in evolution, in academe, and on the stage. William Cohen also considers trial and error in evolutionary terms, trying to find some sort of link between Darwinian ideas and what he calls "the rest of psychology and neurobiology". Cognitive psychologist Roger Schank berates the education system for teaching children facts instead of how to do things. Computers and children alike should try to learn from their mistakes, in other words learn how to generalise.

In one of the longer essays, neuroscientist Steven Rose uses the analogy of the printed page to help him explain how he sees the functioning of the mind and the brain. He then introduces the subject of depression to examine the claims of biological and social psychiatrists concerning what happens to neurotransmitters in the brain. Dismissing mind/brain dualism, he views the mind/brain as a single entity and he concludes that we need to talk about it in two languages, which he calls "neurologese" and "psychologese". These mind and brain languages can then be translated into each other. The problem of the mind/brain scientists is to decipher the rules governing this translation. "Knowing the biology of how we learn and remember," says Rose, "adds to our human appreciation of the richness of our own internal living processes."

Opening the fifth section, on the cosmos, theoretical physicist Lee Smolin examines the debate "between time as absolute and preexisting and time as an aspect of the relations of things". At present, he says, we have two theories of nature, general relativity and quantum mechanics, which are based on two different notions of time. The key problem of theoretical physics is how to combine these two theories. Inevitably, his discussion delves into the mysteries of black holes. Mathematician Ian Stewart returns to a favourite topic, symmetry, describing it as "invariance under transformation". Invariance means that the end result looks the same as the starting point. Stewart then refers briefly to group theory, which deals with whole collections of transformations. He says that mathematicians and physicists have now "begun to use group theory to understand how the symmetries of the universe affect how nature behaves".

The last part of the book considers the future. Here, physicist Freeman Dyson takes issue with an essay by Richard Gott in Nature in which he attempts to revive the "Copernican Principle": that human beings do not occupy a special position in the universe. Dyson maintains that many of Copernicus's assumptions about the longevity of species are implausible and no longer valid. Paleontologist Niles Eldridge also uses the device of writing a letter to children, in this case to his sons. He begins by saying that "life is terrifically resilient", but continues by painting a rather dismal picture of how human beings have tampered with nature, primarily through agriculture. The final essay is by cosmologist Martin Rees. who asks "Can science answer every question?" He concludes: "To understand ourselves, we must understand the stars. We are stardust - the ashes from long-dead stars."

ÁñÁ«ÊÓƵ

Each essay ends with a brief biography and this is accompanied by a short guide to further reading.

Overall, the book highlights both the joys and limitations of the scientific method, of gathering evidence in support of novel hypotheses. Each contributor provides fascinating insights into the scientific psyche, allowing us glimpses into the many ways that hunches become established science. What the essays demonstrate most of all, however, is the vast potential for further discovery.

Robbie Vickers was science books editor of The THES from 1980-94.