Martin Kemp is Professor of the History of Art at Oxford University, and writes extensively on imagery in art and science. This book is a compilation of his columns from Nature, for which he's been writing since 1997. The articles range over a wide variety of topics, from Galileo's sketches of light and shade on the moon, through the modelling of scenes in Vermeer's paintings, to Hooke's description of the eye of the housefly and the naturalistic sculptures of Andy Goldsworthy. Working to a 500-600 word limit in each piece, the author briefly sets the scene before describing a specific image and drawing out some general points from his analysis and, in some cases, making connections with images discussed in other articles.
I greatly enjoyed this book, coming to it from a strong - even professional - interest in visualization, and having a vague idea about how it can act as a conduit between ideas and explanation in art and science. The author examines this in some depth, drawing on a large set of examples from different points in history and which use scales from the cosmic to the microscopic. I finished this book with a renewed appreciation of the power of visualization (and even learnt a new word - bucephadonically apparently means using S-shaped motion). The author also draws together some general points nicely in some specially-written material to bookend the articles, with some illuminating and erudite musings on the aesthetic impulse. The only slight criticism I had of the book was that I found the articles' alliterative titles too tiresome (e.g. "Lucid looking", "Maculate moons", "Abbott's absolutes", etc). But I soon developed the knack of not reading them.