I’ve been hiding out under a rock for the past two months. Part of the quietude here is because I’ve been working hard to write up my dissertation, which compares patterns of jaw growth in humans and the our extinct relative Australopithecus robustus. Even though I presented some extremely preliminary results last year, I’ve generally been hesitant to talk about my work here on Lawnchair.
But in an effort to break my dissertation silence, and to begin thinking about starting to consider crawling out from under my rock, here’s a pretty pretty picture I made:
The green box and whiskers are humans, and the blue boxes A. robustus. Each box and whiskers represent all the individual mandible “sizes” that can be calculated for each species in each dental stage (stage 1 has only baby teeth erupted, stage 5 has nearly all its teeth erupted). It’s sort of like a mandible growth curve for each species, but not exactly.
The problem is I want to see whether I can distinguish patterns size change, from infancy to just before adulthood, in humans and A. robustus. But fossils don’t preserve well, and not all specimens share all the same measurable parts. So I devised a special test that measures a mandible’s “size” based on the traits it compares with other individuals.
Now, clearly from the figure A. robustus and human mandibles differ in mandible size throughout childhood. But questions arise: given the range of size variation within each species (especially stage 4), what are the chances of seeing the same amount of size change between dental stages in each sample? What traits or measurements on the mandible are contributing to these differences? Do these ‘sizes’ reflect the development of each species’ unique mandible shape? Well you’ll just have to stay tuned to find out…
…Or you could check this poster I presented at this year’s annual meeting of the American Association of Physical Anthropologists.
Lawn Chair Anthropology 
Can you tell us more about this? I’d like to find out
more details.
Hi Markus, I’m happy to elaborate on anything you have questions about. You can download my dissertation here (a lot of formatting errors arose when I converted it from Mac to PC and then into a PDF, sorry!). Basically, the picture there shows the size of each human and A. robustus mandible, plotted by dental eruption stage (a rough proxy for age). “Size” of a complex shape, for instance, has often been measured as the geometric mean of a set of variables. The difficulty with fossils is that not all specimens preserve the same parts, so the same (i.e., homologous) size measure might not be able to be obtained for all specimens. To deal with this, I calculated all sizes that could be homologously measured between all pairs of a fossil and a human jaw – that’s what shown in the figure. So for the first age group, eruption stage 1, I sampled 1 fossil and 1 human jaw, measured size by the variables they shared in common, then sampled a different fossil-human pair and measured their size based on shared variables, etc. etc. This creates a distribution (the box and whiskers in the plot) of sizes for each species in each age group. There are some statistical issues that arise with such a strategy, and that’s what a major portion of my dissertation is devoted to. In the end, it’s a useful way to compare patterns of variation – in this case growth-related variation – between groups when one of them (the fossils) has a small sample size and poor preservation. I hope this helps!