One clear character of the record of investigations into human evolution is that, rather than becoming clearer as data increase, our origins become more of a puzzle. With every major fossil find the hominin clade or bush of descent acquires what appears to be another branch. With the recent publication of the genome of our closest living relative, the chimpanzee – and its earliest fossil remains – (Nature, v. 437, p. 47-108), it will hardly be surprising if the assumptions about a gene-based time of separation of the two clades (5-7 Ma) comes into question. Studies of the Y-chromosomes of living human males have suggested ‘bottlenecks’ in our recent evolutionary past, interpreted to indicate near-catastrophic declines in numbers to perhaps that of a few scattered bands. One such ‘near-extinction’ seems to have occurred about 70 thousand years ago, which has been linked to the huge explosion of the Toba ‘supervolcano’ in Indonesia in whose ash are poignantly preserved biface axes. Toba would have had a global climatic effect at a time when fully modern humans were migrating rapidly from Africa across Eurasia; thinly spread and easily isolated by disaster. What followed was an explosive development of both material and aesthetic culture, perhaps enabled by some serious selection amongst those who endured Toba’s global blast.

It is always tempting to restrict hypothesizing with the ‘Just gimme the facts’ outlook – as people of my generation will remember from the main detective in the Dragnet TV series. That is, ideas based on hominin remains alone. Yet all evolution takes place within a wider environmental context; for much of our history that of East Africa. Scanty knowledge of tropical climates there and a reliance on distant deep-sea records had led to the widespread belief that this centre of most hominin evolution gradually became drier since the late Miocene. Lake beds in the East African Rift system have held the key to a useful record, and now some of the detail is emerging (Trauth, M.H. et al. 2005. Late Cenozoic moisture history of East Africa. Science, v. 309, p. 2051-2053). Lakes in the Rift are handy for climate study because they span 8 degrees of latitude north and south of the equator, the spread helping to isolate more local effects of volcanism and tectonics on their sedimentary record from those of regional climate change. Many have little outflow and a local supply of water, so their levels depend mainly on the amount of local precipitation compared with evaporation. The actively subsiding basins in which they form have the opportunity to preserve unbroken, thick records of both lake and river sediments.

Trauth et al. compile environmental and chronological information from sediments in seven Rift basins, going back to about 3 Ma. Volcanic events provide plenty of dating opportunities to calibrate and correlate the sedimentary evidence. They show three rift-long episodes of deep lakes spanning broad periods from 2.7-2.5, 1.9-1.7 and 1.1-0.9 Ma. A few sections reveal lake-level fluctuations on Milankovich timescales. The longer episodes link in time to the intensification of Northern Hemisphere glaciation, to a shift in east-west air circulation over Africa and to the switch from the dominant glacial cyclicity of 41 ka to one of 100 ka, respectively. Wisely, they consider the climatic information to be crucial to studies of human evolution, but still too coarse to be used with confidence in relation to details of the fossil record. Long humid periods would have been ‘easy’, whereas the separating drier periods may have experienced ups and downs in humidity on Milankovich timescales. Fluctuating conditions would have been more stressful and likely to witness speciation. One very odd feature is that the 1.9-1.7 Ma period of deep rift lakes is the time when H. erectus became the first tooled-up being to migrate far beyond Africa. Many have regarded migration as a response to environmental stress, but just as likely is an expansion of opportunity.