Afarensis

As I mentioned in the previous post in this series, Semicircular canals are fully formed by about age two. This makes them an interesting and useful object to study, but in depth studies did not become common until the creation of CT scanners. Previous techniques were highly invasive and destructive. Starting in the mid 1990's and continuing to the present, Fred Spoor embarked on an ambitious program studying mammalian semicircular canals. Primates and hominins were a special focus of this research. As we shall see below, several other researchers performed studies of the semicircular canals as well.

Before going further, let's talk about the size and shape of the semicircular canals seen in modern great apes and modern humans. The anterior and posterior canals are smaller in the great apes than in humans (with the exception of the posterior canal in gorillas, which is similar in size to humans). The lateral canal in humans is smaller than in the great apes (with the exception of the chimpanzee, interestingly enough the lateral canal in humans is also the same size as in some species of Cercopithecus). Once upon a time, the relationship between bipedal locomotion and semicircular canals was fairly straightforward. By 1998 Spoor et al were arguing:

It should be emphasized that the functional interpretation of hominin semicircular canal morphology does not relate to issues of body posture or bipedalism as general locomotor mode. What canal arc sizes appear to relate to in extant primates, and in mammals and birds in general, is the overall agility of locomotion, and the requirement for fast and accurate coordination of locomotor movements...

I will get back to this later. Two questions need to be looked at. First, what was the ancestral condition for humans? The semicircular canals of a number of extinct apes have been examined. Dryopithecus had similar morphology to modern extant apes. Oreopithecus has been examined as well and displays a morphology similar to the modern great apes. Consequently, it is a safe to assume that the great apes display the ancestral condition for humans.
The second question, then, is when did the human condition evolve? To answer that we need to look at the morphology of the semicircular canals in other fossils.

The semicircular canal of a number of australopithecines have been examined including Taung, STS 5, STS 19, MLD 37/38, SK 46, SK 47 and SK 879. Note that this includes species referable to A. africanus and A. (Paranthropous) robustus. Interestingly enough in these specimens the pattern is similar to that seen in great apes. Other specimens examined include: Homo erectus (OH 9 and Sangiran 2 and 4), Neanderthals (Dederiyeh 93002, Gibralter 1 and 2, La-Chapelle-Aux-Saints, La Ferrassie 1-3, La Quina 5, Le Moustier 1, Pech de l'Aze 1, Petit - Puymoyen 5, Spy 1 and 2, and Tabun C1), European Upper Paleolithic (Abri-Pataud 1 and 3, Cro-Magnon 1, Laugerie Base 1), Early anatomically modern (Qafzeh 6, Skhul 5) and European Middle Pleistocene (Abri - Svard, Reilingen and Steinheim). Several groups are missing, among them being H. habilis, early H. ergaster and Mladec. Consequently, interpretation is somewhat contigent. What does seem clear is that the first recognizably human patterns of semicircular canal morphology appear in Homo erectus. Neanderthal semicircular canal morphology has it's antecedents in the European Middle Pleistocene specimens. The Neanderthals had smaller anterior and posterior canals and larger lateral canals than modern humans. What is striking, though, is that early modern and European Upper Paleolithic populations were more different from the Neanderthals than are Holocene humans (compare with this). Modern human morphology, on the other hand, can be traced back through the Upper Paleolithic and early modern Skhul/Qafzeh populations to African and Asian Homo erectus.

But what does this have to do with whales? Thewissen (mentioned in the first post of this series) and Spoor have teamed up to do a paper on the semicircular canals in whales and their ancestors:

We also measured the semicircular canal size of four Eocene cetaceans. They represent respective nodes on the cetacean phylogenetic tree, and are: Ichthyolestes, a pakicetid from Pakistan; Remingtonocetus, a remingtonocetid from India; Indocetus, a protocetid from India; and Dorudon, a dorudontid from Egypt...

One of their conclusions is highly interesting:

The first appearance of small semicircular canals in the middle Eocene genera Remingtonocetus and Indocetus shows that the reduction in canal size took place rapidly and early in cetacean evolution, within five million years of the origin of the order. The palaeoenvironmental context of the fossils.. demonstrates that this event accompanied the invasion of marine environments...The newly evolved and highly derived vestibular sensory regime was almost certainly incompatible with any terrestrial locomotion beyond cautious beach-bound crawling, which indicates that dedicated agile swimming of cetaceans originated as a rapid and fundamental shift in behaviour. As such, the modification of the semicircular canal system represented a crucial 'point of no return' event in early cetacean evolution, which excluded a prolonged semiaquatic phase. These findings strongly contrast with the pattern of gradual change shown by the postcranial skeleton, with the nearmodern body plan and fluke-based propulsion of dorudontids emerging eight million years after the canal system modified... In that period the well-developed hindlimbs of remingtonocetids and protocetids... had a newly acquired function in powerful swimming, with some evidence pointing at a persisting degree of weight-bearing...

The interesting thing about this is the difference in timing. The postcranial anatomy of whales changed gradually while the semicircular canal morphology changed relatively rapidly. We see the same gradual change in the postcranial anatomy during the course of human evolution. As of this writing the semicircular canal morphology of early Homo is unknown (but currently being researched) but I would not be surprised if it was more ape-like than human-like...

Literature Cited

Philip D. Gingerich, et al. 2001 Origin of Whales from Early Artiodactyls: Hands and Feet of Eocene Protocetidae from Pakistan, Science 293, 2239-2242

Hublin et al 1996 A late Neanderthal from Arcy-sur-Cure associated with Upper Paleolithic artefacts, Nature Vol 381:224-226

Rook et al 2004 The bony labyrinth of Oreopithecus bambolii, Journal of Human Evolution 46:347-354

George Gaylord Simpson 1942 The principles of classification and a classification of mammals

Spoor et al 1994 Implications of early hominid labyrinth morphology for evolution of human bipedal locomotion, Nature Vol 369:645-648

Spoor et al 2002 Vestibular evidence for the evolution of aquatic behaviour in early cetaceans, Nature Vol 417:163-166

Spoor et al 2003 The bony labyrinth of Neanderthals, Journal of Human Evolution 44:141-165

Spoor and Zonneveld 1998 Comparative Review of the Human Bony Labyrinth, YRBK of Physical Anthropology 41:211-251

Thewissen et al 2001 Skeletons of terrestrial cetaceans and the relationship ofwhales to artiodactyls, Nature Vol 413:227-281