The qbits and pieces that make up the big picture

Alcohol leads inexorably to volcanoes, or it does in Nigel Calder's fat square book about the big ideas in science from A to Z - well, at least to V. In true apophatic style, the author tells us that this is not an encyclopedia, nor a dictionary, textbook or directory; rather it is a spider's web of connected scientific themes.

By any standards, this is a superb creation. Calder has spun a vast web of key ideas that spans the sciences. These ideas are then illuminated in elegant essays that combine factual description, historical anecdote and personal reflection. The result is a massive collection of high-quality pieces of journalism that draw on the author's lifetime of experience as a science populariser in many media.

Calder's emphasis is on current knowledge and the unsolved problems that characterise new fields. Unlike the reader of an encyclopedia of science, you will not be anaesthetised by lists of facts, dates and names; nor will you be taken through the history of almost everything. This is like a gigantic magazine. While it is possible to read this book's 720 pages from cover to cover, few will do that. Most will use it as an occasional reference to learn in a coherent way about topical science ideas - El Niño, qbits, buckyballs, climate change, dark matter, immortality, sparticles, prions, space weather - that pop up in the news and gradually become part of the mysterious body of things that every reader of The Times Higher is supposed to know about.

Calder's style is very much that of the science journalist, seamlessly mixing direct quotes from the principal architects of new developments with telling historical details and metaphors to explain the science to outsiders. But unlike some recent ventures by bestselling writers into the expository science business, one does not get the impression that Calder has swotted up on his material only the night before regurgitating it for the consumption of uncritical readers. He is at ease with science across a range of disciplines and the book has the luxury of being broad without the necessity to be complete in the way an encyclopedia would need to be. There was no team of editors drawing up a list of key words to include.

Cross-referencing is good, layout is inviting and there are two full indexes - even a bibliography of sources for quotes from original sources.

The achievement of writing more than 100 engaging and accurate scientific essays is considerable and this is a great book to recommend to anyone who wants a wide introduction to science. The most impressive testimony to its eloquence and clarity is that it does all this so successfully without using a single picture.

John D. Barrow is professor of mathematical sciences, University of Cambridge.