Oxford University Press, 2011 (Paperback)

Conclusion From turbulent rivers to auto-catalytic chemical reactions. With the "Nature's Patterns" trilogy, Philip Ball gives us an eclectic, yet surprisingly coherent overview of all kinds of patterns that are found in nature. It's an interesting and challenging read, but perhaps a single book would have sufficed.

Why are honeycomb cells hexagonal? Why do spotted animals tend to have striped tails? And, for that matter, why are animal pelts so often spotted or striped, rather than endowed with, say, a rectangular grid? Why does Jupiter have a giant red spot?

The diversity of the issues that Philip Ball takes on in his trilogy on nature's patterns is overwhelming. Most of them cannot even be said to have much in common: Jupiter's red spot cannot be explained in the same way as the shape of a honeycomb cell. Yet, despite his eclectic subject matter, Philip Ball manages to tell a coherent story. One that goes far beyond stamp collecting of interesting factoids.

A recurring theme in the three books (Shapes, Flow, and Branches) that together form Nature's Patterns: A Tapestry in Three Parts is that many patterns are 'emergent properties'. Spotted and striped pelts do not necessarily provide the best possible camouflage, so they cannot be fully explained in Darwinian terms. Nor is it practical (even if perhaps theoretically possible) to explain such patterns in terms of the laws of physics. Instead, Ball argues, to explain why spots and stripes are so common, you need to consider how animals grow and develop, which chemicals play a role in pelt colouration, etc. If you do this, you will find that spots and stripes are 'easy', in the sense that they readily emerge from the underlying processes. No excessive evolutionary tweaking is necessary to get a striped or spotty pelt: The potential for stripiness and spottiness was already there; Stripiness and spottiness are emergent properties.

None of this is new. And perhaps you will not even find it very surprising. After all, we all know that structures and patterns emerge naturally. Take, for example, the inverse tree-like structure of a river, which emerges when many small streams converge in fewer, larger streams. But even though you may be familiar with the general idea of emergence, I bet you will marvel at the wonderful examples and explanations that Ball has to offer. Sure, a honeycomb cell's shape must have something to do with the fact that they are stacked together in a (presumably) optimal way. But how exactly does this happen? And why does this lead to a hexagonal shape? These are the type of questions that Ball tries to answer. And he manages to do so in a clear and intelligible way.

Philip Ball is an exquisitely fluent writer. The illustrations are also wonderful, particularly the coloured plates in the centre of each book. Another major pro is that the books are different. Unlike most recent popular science, they don't deal with the brain or with theoretical physics. In fact, I cannot think of any other recent book that covers the same ground. For this reason alone, I would recommend picking up one of them.

But I also have a few small gripes. Occasionally, Ball delves a bit too much into the details. Specifically, he provides a fair amount of detail when describing some chemical reactions, which makes these sections hard to follow. And it also seems a bit off balance when compared with most other sections, which are much more casual. Finally, three books is a bit much, or at least it was for me. I positively loved the first book (on its own, I would give it a comfortable four stars), but I felt my interest slipping halfway the second.

In sum, Nature's Patterns is a unique and mostly successful attempt to illustrate the wonderful diversity of the patterns and structures that are found in nature. It's a bit much, though. So, initially at least, you might want to buy only the first book, Shapes.

Get 'Shapes' on amazon.com, amazon.co.uk, or bol.com