Billions of years later – as the book ends – milder winds, bearing perhaps an eighth of the oxygen levels we enjoy today, sweep across barren continents. In lakes and seas, animals in their millions swim, crawl and burrow around shores cloaked in algae and bacteria. This is the world of about 530 million years ago in the midst of the Cambrian explosion, a geological eye-blink about 5 million years long that saw the evolution of most of the major animal groups.

In Early Life on Earth more than 40 geologists, palaeontologists and evolutionary biologists attempt to trace the events that link the two scenes. This is not a book to read if you want a single, coherent story. Instead it offers a fascinating view of the researchers’ individual and sometimes conflicting opinions. Among other things, they discuss the evolution of life from its chemical precursors, the divergence of the fundamental groups of organisms, including eukaryotes (cells with nuclei) and prokaryotes (without nuclei), and the evolution of the main eukaryotic forms: plants, animals and fungi.

The tone of the book changes dramatically between the first group of papers, which discuss how, where and when life may have begun, and the rest, which talk about what happened next. The early papers leave you more intrigued than enlightened. Most biologists agree that life probably arose more than 4 billion years ago. But the oldest surviving rocks are only 3.8 billion years old, so we have no direct physical evidence about the environment that gave birth to life. How do you judge a model for the origin of life when its underlying assumptions are so uncertain? The later papers are less speculative because they have data from rocks and fossils to analyse. A helpful smattering of black-and-white photographs lets you see the traces of the ancient organisms for yourself.

Many of the papers in the book reflect the enormous impact of molecular biology on the study of early evolution. Biologists can build evolutionary trees by comparing DNA sequences taken from different organisms. They can also estimate how rapidly the DNA sequence of a particular gene may change, giving a “molecular clock” that puts rough dates next to the tree’s branches. For example, the Cambrian explosion saw the appearance of the first hard-bodied animals: so while the remains of bones and shells litter Cambrian rocks, those from Precambrian ages are nearly devoid of fossils. DNA-based techniques have now confirmed that this evolutionary explosion was not just about the invention of animal skeletons. An equal or greater leap in the variety of soft-bodied animals occurred at the same time.

Molecular techniques also suggest that the astonishingly rapid changes of the Cambrian explosion may not be unique. The data support the idea that evolution moves like a drunk: shuffling along slowly for a while then suddenly lurching sideways. If the molecular clock is right, then unicellular forms of animals, plants and fungi all diverged from a common ancestor in a rapid burst around a billion years ago. Several papers attempt to explain what caused such dramatic evolutionary events.

Andrew Knoll of Harvard University, Cambridge, discusses the long-standing theory that one trigger may be changes in the environment. The Cambrian explosion could not have happened until the oceans held sufficient oxygen to support the new forms of animal life. But Knoll admits that no one can say for certain whether the rise in the oxygen was the only cause of the explosion or was merely one of several preconditions. Other researchers offer alternative ways of lighting the blue touchpaper. James Valentine of the University of California at Berkeley suggests that the key to the Cambrian explosion may have been the evolution of a basic genetic programme to control animal development, which rapidly mutated to give all the different animal forms. In contrast, the editor of this volume – Stefan Bengtson of the Institute of Earth Sciences, Uppsala, Sweden – argues for the importance of ecological factors. He says that the appearance of hard-bodied animals could well have resulted from an evolutionary arms race between the teeth of predators and the skeletons of prey.

During the period covered by the book, however, some things changed much less than others. Malcolm Walter of Macquarie University in Australia discusses the evolution of microbial mats, which are preserved in fossil form as stromatolites. Today, microbial mats only occupy marginal habitats, but in the Precambrian era these photosynthetic communities of microorganisms dominated the beds of seas and lakes. For more than three billion years the mats captured and precipitated sediments, building reefs over a hundred metres high and hundreds of kilometres long, and changing in composition and complexity as new forms of life appeared. Their doom came finally with the evolution of organisms that they could neither ignore nor incorporate. Some time between 700 and 600 million years ago, early multicellular animals appear to have burrowed and grazed the mats to destruction – showing that humans are by no means the first to overexploit their environment.

Dan Nilsson of Lund University in Sweden and Richard Modlin at the University of Alabama found the peculiar eye in Dioptromysis paucispinosa, a crustacean which lives in shallow waters off Belize. Dioptromysis is, strictly speaking, not a true shrimp but a member of a related order called ‘opossum shrimps’ because females nurture larvae in a brood pouch.

Adult Dioptromysis are about 5 millimetres long and have roughly spherical eyes at the end of eyestalks. The surface of each eye is a honeycomb of hexagonal facets, except at the back, where a single giant facet peers towards the shrimp’s tail.

Apart from the giant facets in each eye, Dioptromysis’s eyes are similar to those of other opossum shrimps. Each has between 800 and 900 normal facets, and beneath every facet is a barrel-shaped light receptor called a rhabdom. However, Nilsson and Modlin have found that the two giant facets focus light onto bowl-shaped retinas made up of 120 long, narrow, closely packed rhabdoms. In effect, the giant facets form simple, camera-like eyes at the back of the shrimp’s compound eyes.

By studying Dioptromysis larvae, the researchers have shown that these simple eyes evolved through a relatively small change in the development of the compound eyes. In a normal compound eye, each facet and its accompanying rhabdom is formed by a group of cells called an ommatidium. The researchers found that each giant facet and its retina develops from 120 ommatidia. All 120 ommatidia produce rhabdoms but only one completes its development fully to form the overlying facet.

Nilsson and Modlin estimate that the giant facets offer more than six times the resolution of the rest of the eye. In fact, even though Dioptromysis’s eyes are small, its visual acuity rivals that of insects with large eyes, such as dragonflies. These have purely compound eyes and have evolved their sharp sight by increasing the size of hundreds of individual facets. Dioptromysis achieves the same clarity by possessing a simple eye behind a single giant facet.

Although the giant facets normally point backwards, videotapes of the shrimp show that Dioptromysis frequently rotates its eyestalks so that its simple eyes are pointing forwards. Nilsson and Modlin suggest this is the shrimp’s adaptation to a dimly lit environment. The rhabdoms behind the giant facets are narrower than those in the rest of the eye and each captures much less light. As a result, the giant facets are about 15 times less sensitive to light than normal facets. Dioptromysis may point its giant facets backwards so that the blind spots they produce in low light cover nothing more important than the shrimp’s tail.