Afarensis is a 3.5-2.8 million year old hominin from the Kada Hadar member of the Hadar formation in the Middle Awash, Ethiopia. He is approximately 41 inches tall, weighs approximately 60 pounds and has a cranial capacity of a whopping 410 cc (approximately). Afarensis is currently considered to be transitional between apes and humans and displays some traits of both. Since he spends a lot of time on the couch watching monster movies, some observers question whether he is an obligate biped (although no one has observed him climbing a tree). He also has a blog called 
Back in April of 2007 I wrote a brief post on a paper by Rak, Ginzberg, and Geffin. I had meant to write a more in depth post about it but kept procrastinating.
First, a picture. Below is a human jaw. The anatomy we are interested in is the coronoid process, the mandibular or sigmoid notch, and the condylar process.
The coronoid process is the site of the insertion of the temporalis muscle. The masseteric vessels and nerve cross over the sigmoid notch. The condyles articulate with the temporal to form the temporomandibular joint. Pre - and postnatally the mandible undergoes some size and shape changes before growth is complete. The picture below is that of a neonate mandible.
(Picture source: Baker, Dupras, and Tocheri 2005, The Osteology of Infants and Children)
As you can see, the ascending ramus is shorter, forms a greater angle with the corpus, has a much more prominent condyle, and has a smaller coronoid process. Additionally, the shape of the sigmoid notch is quite a bit different. In most modern humans the condyle is taller than the coronoid process and the sigmoid notch is relatively deep. In, most, apes the coronoid process is taller than the condyle, the sigmoid notch is shallow, and the ascending ramus is anteroposteriorly broad.
Which brings us to the Rak et al study. Rak and coworkers examined the morphology of the ramus in 146 primate specimens (orangutan, gorilla, pygmy and common chimpanzees, and modern humans from various regions around the globe). Here is how Rak et al describe their methodology:
To convey the anatomical differences in the upper ramal contour, we adopted a method based on Rak et al. (27), which consisted of capturing a digital image of the mandibular ramus with the camera centered at the vertical level of the mandibular notch and held perpendicular to the lateral surface of the ramus. Using FreeHand 9.0 for Macintosh (Adobe Systems, Seattle, WA), we traced the digital image of each ramus from the tip of the condylar process to the anterior margin of the ramus. (This step represents a slight modification of the original method, in which the contour extended only as far as the coronoid tip.)With the aid of the FreeHand software, we stretched the contour proportionally on the vertical and horizontal axes by dragging the contour's lower right corner until it occupied the entire width of the area of the fixed coordinates in the background template. This part of the procedure eliminated differences in size in the analysis. The posterior margin was aligned with the vertical line at 0, and the anterior margin was aligned at T.
This is the diagram accompanying the paper:
The intersection of each ramal contour with the lines A-T provided numerical values which were then analyzed using discriminant function analysis. The resulting function classified species correctly 82% of the time (overall, the function seemed to have problems distinguishing between the two species of chimps). According to Rak et al ramal morphology in apes and humans fell into two main groups. The first is the gorilla morphology with the coronoid process higher than than condyle and a narrow deep sigmoid notch. The second is found in chimps, orangutans, and humans and consists of a spacious mandibular notch, the deepest part of which is anteriorly oriented and the condyle is higher than the coronoid process (Rak et al consider this to be the primitive condition). Included in the analysis were two Australopithecus afarensis mandibles, two Australopithecus (Paranthropus) robustus mandibles, and one Ardipithecus ramidus mandible. Both species of Australopithecines grouped with gorillas, while Ardipithecus grouped with humans, chimps and orangutans.
So what is the problem? To understand that we need to look at an earlier paper written by the same group. In the earlier paper Rak et al performed the same type of analysis on nine Neanderthals, nine non-Neanderthal fossil hominins and 250 modern human mandibles. Results were similar to that of the australopithecine study. Neanderthals were separated from fossil hominins and modern humans by the same traits that separated australopithecines from modern humans. Here, for example is Regourdou:
Note that, like australopithecines, the coronoid process is higher the the condyle, the sigmoid notch is shallow, and the greatest depth is posterior. As with all things Neanderthal, this paper drew a response. In this case the response came three years later in a paper by Wolpoff and Frayer. They grant that some Neanderthal mandibles display the pattern Rak et al identified. On the other hand, they point out that La Ferrassie 1 looks a little different:
La Ferrassie 1 has a broad, deep, sigmoid notch. The greatest depth is about the center of the notch. The condyle displays evidence of arthritic flattening, without which, it would be as tall as the coronoid process. The situation is no different for early archaics. Ternifine 3, for example looks a lot like the australopithicines (robustus in particular):
Finally, the early modern (Aurignacian associated) Stetten 1 follows the australopithecine/Neanderthal pattern as well:
Wolpoff and Frayer go on to demonstrate that Neanderthal display a wide variety of mandibular morphologies that frequently overlap that of modern humans. Modern humans and non-Neanderthal hominins also display a wide variety of overlapping mandibular morphologies. Consequently, they argue that the three traits (and a fourth not used in the australopithecine paper) do not give a valid phylogenetic signal. The same critique can be made with the australopithecine paper. Does it sample the range of variation in the australopithecines (including Paranthropus)? Probably not...
Literature Cited
Yoel Rak, Avishag Ginzburg, Eli Geffen 2002 Does Homo neanderthalensis play a role in modern human ancestry? The mandibular evidence AJPA 119: 199-204
Yoel Rak, Avishag Ginzburg, Eli Geffen 2007 Gorilla-like anatomy on Australopithecus afarensis mandibles suggests Au. afarensis link to robust australopiths PNAS 104(16): 6568-6572
Milford H. Wolpoff and David W. Frayer 2005 Unique ramus anatomy for Neandertals? AJPA 128(2):245-251
Humanities & Social Science
Comments
I was not terribly fond of the Rak et al. 2007 paper, either. For a class project I examined chimpanzee mandibles, looking for qualitative versions of their quantified "afarensis-gorilla" traits (i.e. coronoid process taller than condyle). While my results agreed with theirs, suggesting the afarensis rami were significantly different from the chimpanzee 'ancestral' morphotype, I still take issue with the paper (and I wish I'd designed a better study myself). Here are some thoughts:
For one thing, their method seems strikingly similar to that of geometric morphometrics, but falls short of GM's strengths. In this study, the points on the mandible that intersect lines A-T would be analogous to equally-spaced landmarks. Gunz et al. 2005 describe the problems with equally-spaced landmarks, and present an alternative method that 'minimizes bending energy' between specimens. A main issue with equal-spacing that the Gunz &al method attempts to circumvent is homology. The diagram from Rak &al's paper, posted above, clearly demonstrates that homologous points are not being compared.
Moreover, size may well be important in influence the morphology of the ramus. By removing size and never considering it, Rak &al overlook the possible consequences of allometry. Thus, I think if I were to address this study with (another) one of my own, I would treat the ramal contour with the latest GMM methods.
The study also potentially exaggerates the phylogenetic signal of the ramus. Using extant taxa as outgroups and ignoring fossil specimens further obfuscates phylogenetic inferences. Without looking at other Miocene hominoid mandibles, how sure can one be that extant orangutans haven't changed since their divergence from African apes? And you make a good point about the ill-sampled ramus. The Rak study looked at 3 afarensis rami--2 of which were antimeres from a single individual, and the 3rd was a fragment. There are 2 well-enough preserved Au. robustus included in the study. Other australopithecine (and early Homo) rami are either absent or too fragmentary to include in such an analysis. So like you said, we're probably missingthe true range of variation in ramal morphology of fossil taxa.
Posted by: zacharoo | January 12, 2009 10:05 PM
Interesting analysis but doesn't address a major problem: how are the original specimens oriented?
"With the aid of the FreeHand software, we stretched the contour proportionally on the vertical and horizontal axes by dragging the contour's lower right corner until it occupied the entire width of the area of the fixed coordinates in the background template. This part of the procedure eliminated differences in size in the analysis. The posterior margin was aligned with the vertical line at 0, and the anterior margin was aligned at T. The posterior ramal margin in the entire sample exhibits a slight concavity between the posterior end of the condyle and the insertion site of the posterior fibers of the masseter and medial pterygoid muscles; using these two posteriorly protruding structures, we were able to orient the posterior margin on a vertical line throughout the sample. The intersection of the ramal contour with each of the vertical lines, A through T, yielded 20 numeric variables for each ramus (SI Table 1). "
The difficulty is the posterior ramal margin is not a straight line, it features two protruding *curved* structures. Orienting the anterior point "T" to this line between two curves is misleading, there is an unanalyzed and/or unreported degree of variability in the placement of the line. This in turn means the degree of projection of the coronoid above the condyle across specimens is much less absolute than the article suggests. The other more important critique is that the coronoid as mentioned, is the anchor for mm temporalis. As such individual variation can be expected to be high, possibly higher than rigorous phylogenetic hypothesis testing requires.
John Y. Anderson, University of New Mexico
Posted by: JY Anderson | January 12, 2009 10:59 PM
You both make some excellent points. The lack of any discussion of allometry was a definite drawback. Other than a brief discussion of sexual dimorphism in gorillas size goes unmentioned. They dismiss any impact due to muscle size, etc, in about three sentences and, implicitly, any role for different diets. Their discussion of the possible convergent evolution between A. afarensis and the robust group amounted to hand waving.
Posted by: afarensis, FCD | January 12, 2009 11:31 PM