Hominid Brain Evolution: Looks Can Be Deceiving

Every now and then a newly discovered fossil comes along that changes the big picture of hominid evolution. In 1985, the Nariokotome skeleton from Kenya (KNM-WT 15000) created a stir because it was surprisingly tall for early Homo (1). A year later, a robust australopithecine dubbed the Black Skull (KNM-WT 17000) caused a sea change in how paleoanthropologists drew their family trees because its date was surprisingly early (2). Thanks to a report by Conroy et al. on page 1730 of this issue (3), another new specimen (Stw 505), tentatively assigned to Australopithecus africanus, is about to wreak havoc on our view of hominid evolution. This time it is the cranial capacity that is a surprise. The word from the paleoanthropological grapevine (a hardy variety) was that the capacity of this word-of-mouth famous fossil was about to break the australopithecine piggy bank by measuring over 600 cm³ (compare this with the average modern human skull at 1350 cm³). It did not. Although the capacity of the Stw 505 skull of 515 cm³ is the largest for any known gracile australopithecine, it is a far cry from the blockbuster volume we were all expecting.

Stw 505 does not just represent another pretty cranium. Its significance lies in the fact that a larger capacity was expected because of rumors that the interior of its braincase looked enormous compared with those of other australopithecines. Conroy et al. now suggest that these other australopithecines may be associated with published cranial capacities that are too large, and that this explains the great, if frustrated, expectations for Stw 505. They note, for example, that Sts 71 has a published capacity of 428 cm³ but that their water-displacement measurement of a detailed cast renders it closer to 370 cm³. This sent me running to my own endocast and skull collection. They are right--the published capacity of Sts 71 is clearly too big. If the capacity of Sts 71 is the 370 to 375 cm³ that my water volume measurements confirm, then my eyes tell me that the approximately 3.2 million-year-old hominid endocast from Hadar, Ethiopia (AL 162-28) must be considerably smaller than its published estimate of 375 to 400 cm³ (4). And so it goes for a number of early hominid casts in my collection, including both gracile and robust australopithecines. But then all I have are my eyes, calipers, and some molding compound. Conroy's team, on the other hand, uses sophisticated three-dimensional computed tomography technology with Seidler and his colleagues in Austria who honed their analytical teeth on the Tyrolean Iceman and then moved on to apply their medical imaging technology to fossil hominids (5). Importantly (and not to be lost in all the hoopla), Conroy et al. have now validated the use of "virtual endocasts" for estimating cranial capacity by testing the reliability of this procedure on 10 Homo sapiens skulls (3).

The implication of Stw 505's surprising cranial capacity is that something is very wrong with the published record of early hominid cranial capacities. If so, the ramifications for hominid brain evolution may be profound. For example, what now of the received wisdom that cranial capacity took off in early Homo relative to australopithecines somewhere around 2 million years ago (see figure) (6, 7)? Is this true, or did cranial capacity begin its sharp increase much earlier in (some) australopithecines and simply continue in their Homo descendants? To answer this, ideally we need to see verified or revised cranial capacities of all early hominids plotted against time. And what about the capacities of robust compared with gracile australopithecines? Did robust australopithecines have somewhat larger brains as previously believed, or is brain size equivalent in the two types of hominids, or even bigger in gracile australopithecines? The answer to this will determine the encephalization quotients of the two types of early hominids (8), which is an important indicator of relative "braininess," an issue that is now up for grabs. Are the published capacities for the earliest Homo specimens accurate, or are some of them also too big, as Conroy et al. hint? Hopefully, Conroy and his colleagues will now revisit the cranial capacities of as many early fossil hominids from Africa as possible. Come to think of it, they might want to examine the trends in cranial capacity at the front end of the fossil record too, because controversial new dates for some Java Homo erectus specimens (9) indicate that they (and their cranial capacities) may have occurred much earlier than previously believed. If so, whether different patterns of cranial capacity evolution distinguished H. erectus from early H. sapiens (10) is also ripe for reconsideration. Fortunately, Seidler's team is already collecting computed tomography data from skulls at both ends of the hominid fossil record (3, 5), and I, for one, cannot wait to see what if anything the results will do to the big picture regarding evolution of hominid brain size.

1. R. E. F. Leakey and A. C. Walker, Natl. Geogr. Mag. 168, 624 (1985) [GEOREF].
2. A. Walker, R. E. Leakey, J. M. Harris, F. H. Brown, Nature 322, 517 (1986) [GEOREF].
3. G. C. Conroy et al., Science 280, 1730 (1998).
4. D. Falk, Nature 313, 45 (1985) [GEOREF]; R. L. Holloway, in Origins of the Human Brain, J.-P. Changeux and J. Chavaillon, Eds. (Clarendon, Oxford, 1995), pp. 42-54.
5. H. Seidler et al., J. Hum. Evol. 33, 691 (1997) [Medline].
6. D. Falk, Evolution of the Brain and Cognition in Hominids (James Arthur Lecture, American Museum of Natural History, New York, 1992).
7. D. Falk, Annu. Rev. Anthropol. 16, 13 (1987).
8. J. Kappelman, J. Hum. Evol. 30, 243 (1996).
9. C. C. Swisher III et al., Science 263, 1118 (1994) [Medline].
10. S. R. Leigh, Am. J. Phys. Anthropol. 87, 1 (1992) [Medline].
11. This work supported by NSF grant SBR-9729796.