The recent technical comments on Ardipithecus have left some of us scratching our heads and thinking about how to define a meaningful phylogenetic trait. Drew Rendall and Tony DiFiore wrote one of my favorite papers on the subject, which deals specifically with the perceived “special” status of behavior in human and primate evolution. I think those of us who deal with cold, hard bones can all stand to be reminded every once in awhile that there are other things that are meaningful, too, even if you can’t dig them up!
But if we are stuck with just looking at the morphology, we need to know how to interpret those bones. Different mechanisms act to shape our bones in different ways, and not all traits are equal.
We know that morphogenetic fields shape a lot of variation between species, and I would argue that traits which we can trace back to these fields are the ones that we should be using. This presents a few challenges, though. A single morphogenetic field can wreak havoc simultaneously on many of the atomized traits that we’re used to dealing with. Without fiddling with these fields in the lab, we’re left trying to figure out which traits are correlated with each other. And then, if those correlations are meaningful because of a shift in one morphogenetic field which affects all of our traits, or if they represent two different instances of selection.
A prime example of the difficulty of this approach is the skull. Are brow ridges an adaptation, or a simple by-product of selection for something else? What about the extreme canine reduction in the robust australopithecines- was that selected for, or the did it just happen because their giant molars needed the space?
The pelvis is a good example, too, and one that is relevant to our discussion. It seems silly to make a list of landmarks and check them off, rather than looking at the Total Morphological Pattern (to borrow a phrase from Le Gros Clark). We have a short, broad ilium and a greater sciatic notch, but is the notch really a distinct trait, or just the result of the restructuring of the ilium? Yes, both Ardi and Oreopithecus have an anterior inferior iliac spine, but is it really the same structure, derived from the same shift in morphogenetic field? To answer that, you have to look at the rest of the pelvis. White et al argue that it is not the same structure, in part because humans, australopithecines, and Ardi all share a unique growth plate in the anterior inferior iliac spine, which Oreopithecus apparently lacks. Oreo also lacks the shortened iliac “isthmus” that we have in our lineage as well, which means that its sacroiliac joint is further away from the acetabulum than in humans. I doubt that White et al. have had the final say in this discussion, and I look forward to a more rigorous comparison.
Pelvis of (A) Pan, (B) Proconsul, (C) Ardi, and (D) Lucy. The "Iliac isthmus" is highlighted in blue. It was elongated in Pan (a derived trait!) and shortened in the human lineage. Image from Lovejoy et al. 2009, Supplementary Online Material.
Here is the Oreopithecus pelvis for comparison. I'm not comfortable drawing in the AIIS because I don't think this image is in the same plane as the others, but it may offer a rough comparison. From Rook et al. 1999.
And then there are other traits that can tell us about how an animal moved around, but don’t tell us anything about who it’s related to. Certain activities can leave “scars” on your bones as you develop. Squatting facets on the tibia, for example, develop when a child continuously squats throughout his or her development. That kind of trait can tell you what the kid was doing- squatting- but it isn’t a useful trait to use in phylogenetic analyses. Sarmiento suggested that the fused os centrale of the wrist was a phylogenetically informative trait, but it seems to me that this trait might better reflect the loading pattern during life than phylogenetic affinity. He also mentions a non-articular styloid process in the wrist- ulnar withdrawal- which I think is a good, phylogenetically meaningful trait. It was most likely selected for to allow a plamigrade ape to adduct and supinate the wrist, and was later co-opted by the brachiators. But Pierolapithecus possesses ulnar withdrawal at 12 million years ago, so it’s most likely primitive for humans, chimps, and gorillas.
These things are important to think about, and I’m definitely considering these issues in my own research. For those of you who work in different fields, how do you pick out the seeds from the chaff?
References:
Rook, L. (1999). Oreopithecus was a bipedal ape after all: Evidence from the iliac cancellous architecture Proceedings of the National Academy of Sciences, 96 (15), 8795-8799 DOI: 10.1073/pnas.96.15.8795
Lovejoy, C., Suwa, G., Spurlock, L., Asfaw, B., & White, T. (2009). The Pelvis and Femur of Ardipithecus ramidus: The Emergence of Upright Walking Science, 326 (5949), 71-71 DOI: 10.1126/science.1175831
Rendall D, & Di Fiore A (2007). Homoplasy, homology, and the perceived special status of behavior in evolution. Journal of human evolution, 52 (5), 504-21 PMID: 17383711
White, T., Suwa, G., & Lovejoy, C. (2010). Response to Comment on the Paleobiology and Classification of Ardipithecus ramidus Science, 328 (5982), 1105-1105 DOI: 10.1126/science.1185462
Sarmiento, E. (2010). Comment on the Paleobiology and Classification of Ardipithecus ramidus Science, 328 (5982), 1105-1105 DOI: 10.1126/science.1184148
This isn’t something I have a huge amount of experience in, but isn’t one of the points of doing a cladistic analysis to find out which characters are homoplastic (noisy) and which are phylogenetic informative (correlated)?
Pingback: Weekly Link Conglomerate | Ninjameys
The question of what is meaningful is an important one. I think that one assumption that we make all the time — we have to — is that various linear measurements, landmarks and indices have any biological meaning whatsoever. At pretty much every step, the things we quantify for systematic study are an abstraction, a step or two removed from their biological reality. If I want to study something complicated like iliac flare in hominids, I need to come up with some metric that accurately quantifies this. Perhaps a metric is good at distinguishing taxa, but does it convey any other information (say about development)? Or does it make a biologically meaningful difference if you measure a trait one way as opposed to another? Of course, in all likelihood the type of measurement taken will depend on the question (and the preservation of the sample).
Aside from methodological questions, another is what traits are actually meaningful evolutionarily? If Neandertals and Paleolithic humans differ in the rate of tooth enamel deposition, is this taxonomically significant, even though similar differences within modern humans are not?
Sometimes, when the questions and doubts start piling up like this, it’s good to take a minute, sit down, and have a beer.
Well, there are assumptions, and then there are assumptions. If we’re studying biomechanics, useful measurements abound. They won’t necessarily tell you anything taxonomically important, but if you just want to study how things move there are plenty of ways of figuring that out. You just have to know your lever arms and fulcrums.
I think that, if we understand developmental biology well enough, we can make some pretty well-grounded assumptions about traits based on that, too. Things like finger:thumb ratio based on Hoxd11 and Hoxd13, or lumbar and somite counts. The trick is, we have to use our knowledge of developmental biology to guide our measurements and not the other way around. Once we’ve correlated genetic structures with morphological ones, then we can perhaps make inferences about the development of fossil taxa, but it’s still going to be tricky.
If we’re looking at something complicated like the iliac flare, that’s when it’s even more important to understand pattern formation. How do we get a short or flared ilium? Are those two different questions?
Again, we may not necessarily get anything evolutionarily useful out of a developmental genetics approach, but we stand a much greater chance of finding something that could potentially be the target of selection if we understand the genetic structure of a trait across different taxa.