BP), a later arrival (12,000-10,000 BP) of the Amerinds, and a third wave (10,000-7000 BP) of the Eskimos and the Na-Dene speakers. But the awkward fact remains that there is no direct archaeological evidence to support these earlier dates. The remains of only thirty-seven individuals had been found in America by AD 2000 which dated to earlier than 11,000 BP.23 The archaeological evidence for early man in the Americas suffers further because there are no securely dated sites in Alaska earlier than the Bluefish caves in the eastern Yukon Territory, which date to between 15,000 and 12,000 years ago.24 Nevertheless, there is little doubt that there are many features common to both sides of the Beringia area. One element is the 'Northwest Microblade' tradition, a particular type of microblade, which was wedge-shaped and made from a distinctive core, found all over Beringia.25 These cores have been associated with one site in particular, Denali, which, according to F. Hadleigh West, is the eastern outpost of Dyukhtai culture, with at least twenty locations in Alaska. (Denali is situated in and around Tangle Lakes in Alaska.) Dyukhtai culture is no older than 18,000 years ago and Denali was gone by 8000 BP.26 That early man crossed the Bering land bridge between 18,000 and 12,000 years ago is also supported by details from the Meadowcroft rock shelter in western Pennsylvania, where remains have been calculated, on eight separate occasions, to between 17,000 and 11,000 BC. And by the fact that the presence of early man at Tierra del Fuego, 'the end of the road' at the southern tip of South America, has been dated to about 9000 BC. However, there are still doubts about the dating of Meadowcroft, where the remains are corrupted by the presence of coal, which may make it seem older than it is. Early man's discovery of the New World may not seem, on the face of it, to fall into the category of 'ideas'. But there are three reasons for including it. One is because the conquest of cold was a major advance in early humans' capabilities. Second, in being cut off for so long, and from such an early date (say 15,000 BP to AD 1492, 14,500 years, and ignoring the possibility of Norse contacts, which were abortive) the parallel development of the Old World and the New provides a neat natural experiment, to compare how and in what order different ideas developed. Third, as we shall now see, this separation throws crucial light on the development of language. George Schaller, as mentioned before, has pointed out that lions hunt game in groups-fairly successfully- without the benefit of language. We cannot say, therefore, that as man turned to the hunting of big game he necessarily had more than the rudiments of language. On the other hand, it would seem highly unlikely that he could manufacture standardised tools, or cave paintings, or beads, without language. But these are all inferential forms of evidence. Is there anything more direct? We have to remember that many of the skulls of ancient men and women, on which these studies are based, have been in the ground for as much as 2 million years, with rock and earth bearing down on them. Their present-day configuration, therefore, may owe as much to those millennia of pressure as to their original form. Nevertheless, with this (all-important) proviso in mind, we may say as follows. Modern studies, of people living today, show that two areas of the brain are chiefly responsible for language-what are called Broca's area, and Wernicke's area. Broca's area is located in the left hemisphere, towards the front of the brain, and about half-way up. Individuals with damage to that area generally lose some of their facility with words. Wernicke's area, slightly larger than Broca's area, is also in the left hemisphere, but behind it, also about half-way up. Damage to Wernicke's area affects comprehension.27 There is much more to the brain than this, of course, in relation to language. However, studies of the skulls of H. habilis show that Broca's area was present with the earliest of the hominids but not with the australopithecines. Pongids (apes), who lack Broca's area, cannot produce any human speech sounds and they further appear to lack intentional voluntary control of vocal signals: for example, they cannot suppress food-barks even when it is in their best interest to do so.28 On the other hand, several experiments in the late twentieth century show that chimps possess a nascent language ability in that, although they couldn't speak, they could learn American Sign Language. This suggests (to some) that language ability is very old.29 In line with such reasoning, each of the skulls unearthed at Skhul and Qafzeh in Israel and dated to