The modern rain forest stretches from the Atlantic to eastern [Congo], which is closer to the Indian Ocean. During glacial maxima, it seems, the rain forest shrank to three small patches, one near each end of its present extent and the third in between, in southern Nigeria and Cameroon.  These three oases sometimes added up to no more than about 20 percent of the present extent of the African rain forest.

- Steven M. Stanley, Children of the Ice Age, 1996

High above the equator gives a view of the right bank and the left bank of the Congo River, the dividing line between our cousins, the chimp and the bonobo, starting about 2.5 million years ago – about the same time that our earliest Homo ancestors split off from the bipedal apes, probably somewhere east or south of here.  Lots of swamp forest down below, which is where the last bonobos live.  (They’re called the “Left Bank Chimps” for other reasons, as well.)  They were trapped there by climate change, in one of the refugia from the cooling and drying episodes.  The bonobos in the other refugia probably died out.

     Certainly in the time since the orangutans split off the great ape tree at about 12 million years ago, climate has changed.  Globally it cooled – and, in East Africa, the effects were even more pronounced because the highlands of Kenya and Ethiopia were pushed up like a blister, so that now Nairobi is slightly more elevated than mile-high Denver.  The largest cities of a number of eastern and southern African countries are equally elevated (which is fortunate, as it gets them up above the lower-lying malaria zone).

     Temperatures decrease about 6°^C for every thousand meter rise, everything else being equal, so Nairobi ought to be about 10°^C cooler than Kenya’s Indian Ocean coastline – and up here in the stratosphere at 10 km, the air ought to be about 60°^C cooler than down on the ground.  The two East African blisters thinned out the forest and created more open woodland and even savanna (grasslands with the occasional tree).  Something similar probably happened in South Africa, where the first australopithecine skull was found.

     At the same time that things cooled, climate became much more variable – so the averages don’t tell you the whole story.  (Averages are just devices to keep us from thinking more deeply.)  Just as variability in populations is the key to thinking about how species can change, so variability in climate (especially the fire-prone droughts) is the key to thinking about what speeds up species change.

We don’t know much about how the brain reorganized during the last five million years, though we do know one tantalizing feature of average size.  Unlike upright posture which was pretty well established before four million years ago, brain size didn’t increase much until the Homo lineage was spun off from the australopithecine lineage about 2.4 million years ago.  Since then, the brain has increased about three-fold in volume (and more like four-fold in neocortical surface area, from more infolding).  Furthermore, sexual dimorphism decreases in the Homo lineage; instead of australopith males being twice as large as females (usually a sign of males fighting one another over access to females), the Homo erectus males revert to being only 20 percent larger than females, a pattern much like our chimpanzee cousins.

     And as I earlier mentioned, the archaeologists say that split-cobble toolmaking also started up about 2.5 million years ago.  Why was there all this prehuman action starting about 2.5 million years ago?

     The first clue is that it isn’t just the hominid lineage that spun off new species about then.  That’s also when the ancestral Pan lineage split into what is today the bonobo and the chimp.  The bonobos are confined to the left bank of the Congo River, and now limited to the forest immediately below me, here at Latitude Zero.  The various subspecies of the common chimpanzee extend across equatorial Africa from the Rift Valley into westernmost Africa.  About 2.5 million years ago is also about when, among the lesser apes in southeast Asia, the gibbons spun off the siamangs.  And numerous other major mammalian groups such as pigs and antelopes also show a lot of speciation happening about then, best documented in Africa.  This speciation fest, by itself, suggested big climate change between 3 and 2 million years ago.

     The second clue is that Northern Hemisphere glaciation intensified between 3.1 and 2.5 million years ago, thanks to all the moisture delivered to the far north via evaporation from a more vigorous Gulf Stream.  Ice sheets eventually built up to the height of mountain ranges over Canada and Scandinavia.  The ice mountains tended to melt off every 40,000 to 100,000 years, only to rebuild again.  Yet the site of the hominid speciation action was likely in the tropics, probably somewhere here in Africa.  And Africa wasn’t icy back then – a little cooler (3-5^°C), but not the sort of thing that would keep the tropics from being nice and warm in most places.  Most importantly, Africa was much drier (Lake Victoria, one of Africa’s largest lakes, dried up during the last ice age).

     So what do the Ice Ages have to do with stimulating all this evolutionary action among the mammals – and particularly among our African ancestors?  (Think drought, not cold.)  What might it have to do with the items on my little chunnel-train list of the big-time augmentations of chimpanzeelike behaviors?

     First let me explain some of the standard story about climate change and population fluctuations.  Then I’ll try to show what abrupt climate change adds to the story, which is a much less settled question, not yet in the textbooks.

Darwin saw that climate had repeatedly changed but, unlike others before him, he successfully figured out a mechanism whereby animal species could change with it, to adapt body and behavior to the new climate regime.  Just spawn a lot of variations in each generation and, given the high mortality among the young, only those variants better adapted to the current environment will survive long enough to reach reproductive age.  And those lucky variants will spawn additional variations around their body-and-behavior traits, to further explore “fits” to the environment’s opportunities and perils.  Those variants better suited to some other climate simply tend not to grow up and reproduce.

     But note that this need not be sustainable change.  When the climate changes back to the original, the adaptations can track it back again.  (Remind me later to explain how speciation can ratchet the adaptation, so it doesn’t drift back so easily).  Furthermore, adaptations may mostly happen when there is no other choice.  At least on some time scales, climate’s influence needs to be viewed with some skepticism, as most species react to cooling and drying episodes by moving elsewhere, places where their suite of adaptations still works.

     Well, moving is something of an euphuism; if there are regional subpopulations, some of them may die out while others continue.  With serious climate change, this may leave only a few subpopulations in refugia, places where the species still has all of the essentials for making a living and reproducing.  Let climate improve, and they will “expand their range” to live in more places, with refugia pioneers rediscovering those old places where the species once thrived.

     Population size is always fluctuating like this.  A shrink-and-expand cycle produces more evolutionary change than adaptations-in-place, as I mentioned earlier, but other factors that truly fragment the central population may prove even more important in transforming the species, particularly (as I’ll mention in a minute) because of the chance aggregations that occur when things fragment.

     Population fragmentations are what happens when a lake almost dries up.  As the water level drops, you get a series of small ponds and puddles, in which life continues – but there’s now a lot of inbreeding because they are trapped and cannot circulate.  There may be some selection for living in the increasingly salty ponds.  Most little ponds dry up completely, and the life in them doesn’t contribute to what happens later (there are some exceptions, animals whose lay-them-and-leave-them young can survive desiccation).  If only the population in one pond survives and then re-expands, we see a classical “population bottleneck“ where the re-enlarged population is comprised of only closely-related individuals.  Note that much of the pre-existing variety may vanish, even though little natural selection affected the survivors directly (it just eliminated much of their competition by chance).  Refugia are common on land, too, and land animals can be similarly restricted to inbreeding for awhile, with a great reduction in genetic variety because of sheer chance.  Cheetahs, all very similar genetically, likely re-expanded from one such small surviving population.  This means that natural selection no longer has much variation to operate on, preventing evolution until mutations and cross-over breaks eventually generate some new variation on which recombination can act.

     But more often, multiple ponds survive the downsizing and fragmentation.  When the old lake refills with the rains, multiple small groups form the basis of the re-enlarged animal population.  Each group may have survived for a different reason, some developing adaptations but most not.  It is presumably only when the land refugia are also under stress, when they too become excessively cool or dry or dusty, that selection can efficiently operate to improve thermal regulation or kidneys or noses.  Or to select for rare abilities that, for once, make a difference.  This seems fundamentally different from “who survives” during ordinary population contractions into unstressed refugia.

     And there is really nothing to suggest that it was all that cold in tropical refugia for our ancestors.  Drought, however, is another matter, as is an ecosystem that fire has severely disrupted.  Cooling is just the easiest thing to measure in reconstructing paleoclimate, and not necessarily the most relevant thing to survive and thrive.

     Lake Victoria, right on the equator over in East Africa, dried up during the last ice age and abruptly refilled about 15,000 years ago.  The cichlid fish in the East African lakes split into many new species about then.

     Selection during downsizing isn’t the only way that evolution operates.  There are also opportunities to be exploited when conditions improve.  And for an ape-like creature already adapted to making a living on the savanna, the opportunities were of boom-time proportions.

A changing climate drives different populations apart and brings them together again.  This could have facilitated speciation.

- Richard Potts, 2001

The later part of the Pleistocene had been a period of extreme fluctuation in climate.  Vegetation zones had moved north and south, up-mountain and down.  Extensive woodlands had been fragmented by invading steppe or savannas, and had been rejoined as forests returned. Glaciers and harsh periglacial climates made vast areas of northern Eurasia periodically uninhabitable by hominids, presumably spurring major migrations and causing local extinctions.  Sea levels had risen and fallen, alternately creating islands and land bridges.  Perhaps no period in the history of the globe had been more conducive to the emergence of new species and to competition between related species newly in contact – in other words, to evolutionary change.  And the variety seen among later Pleistocene hominid fossils was, in fact, exactly the kind of thing one might expect to find under those conditions.

- Ian Tattersall, 1995