KNM-ER 1470 is not a microcephalic
I keep seeing this story about Tim Bromage's "computer-simulated" reconstruction of KNM-ER 1470.
Bromage said his team's reconstruction includes biological principles not known at the time of the skull's discovery, which state that a mammal's eyes, ears and mouth must be in precise relationships relative to one another.
"It doesn't matter if you're a rat, a kangaroo, an elephant, a human or a dog -- their [facial features] are all organized to a very specific architectural plan," Bromage said.
Well, let's see. Here's the lateral view of the old and new reconstructions in the article:
The photo accompanying the article
Wow, that new reconstruction sure has a sloping face doesn't it? Oh wait! It's rotated at a different angle from the old reconstruction! Let's use Photoshop to fix that right up:
Same picture, with the reconstruction rotated to the Frankfort horizontal, like the old reconstruction.
Well, now, that's better. Now the vaults are at the same orientation. And the reconstruction does have a bit more sloping face. By about 5 degrees.
Now, a good cast of ER 1470 comes with the face and vault in separate pieces. There is only one join between them, at the nose, and it's not a very good one. Every graduate student in the world has probably taken these two pieces and rotated them back and forth to decide on the best angle. No doubt, there is five degrees of variation between them all. I'm perfectly willing to believe that the skull should have five more degrees of inclination to its face.
I wish that new reconstruction had the nasal bones pictured, though -- a greater slope makes the join there worse, which is probably why the original reconstruction was made with the more vertical orientation. Oh wait! It looks like the nasal bones have been crammed back under the frontal bone! That seems odd...considering the fronto-nasal suture is there underneath the small browridge. Hmm...
I think this particular issue is one for which more detail will be necessary. That seems like an unreasonable placement of the nasal bones, but we only have the one view to work with.
There is a lot of talk about the brain size of the specimen. I don't have any details of the presentation, and it is possible that Bromage was incorrectly quoted. Here is what the article says:
The new reconstruction suggests H. rudolfensis' jaw jutted out much farther than previously thought. The researchers say the cranial capacity of a hominid can be estimated based on the angle of the jaw's slope and they have downsized KNM-ER 1470's cranial capacity from 752 cubic centimeters to about 526 cc. (Humans have an average cranial capacity of about 1,300 cc.)
That, of course, is utter nonsense. Ralph Holloway produced an endocast, the joins between the fragments are good, and the volume of 752 cc was measured by water displacement. Why in the world would you estimate brain size from the face when you have a perfectly good vault? It has to be a misquote.
The article quotes Bob Martin as a skeptic:
"What they're claiming is you stick the face out, and because the face sticks out more the brain capacity has to be less. I don't follow that at all," said Martin, who is an expert on hominid skulls and who was not involved in the study.
"They haven't changed the skull at all; they've simply rotated the face outwards," Martin added.
He mentions that the 752 cc estimate is not a problem in comparison to other contemporary hominids. We can also mention the Dmanisi crania, the largest of which (D2280) has a brain size essentially the same as KNM-ER 1470, at 1.75 million years. KNM-ER 1470 is one of the most solid endocranial volume estimates in the fossil record. It's the face that's crummy!
Big arms, small sacrum
In case you're following the debate about Homo habilis limb proportions, there's a new contribution by Martin Haeusler and Henry McHenry in the JHE holding pen. They examined the partial KNM-ER 3735 skeleton.
KNM-ER 3735 is often assigned to Homo habilis, but it's not exactly an easy diagnosis. There are a few pieces of the skull preserving anatomy, including the cheek, frontal and temporal. Here's what Bernard Wood (1991) had to say about the skeleton:
The form of the mandibular fossa and malar region virtually preclude this specimen from being attributed to A. boisei. Its general affinities are with Homo. Some features (e.g. vault thickness) ally it with a Homo erectus-like hominid, but in other areas (e.g. the frontal) it is more like crania such as KNM-ER 1813, a conclusion endorsed by Walker (1987) and by Leakey et al. (1989). Tobias (1989) includes KNM-ER 3735 within H. habilis.
Provisional taxonomic assessment: Homo sp. indet.
Well, that's not exactly a rousing endorsment. You can see the problem --- and it's a common tale for hominid fossils. It has a smaller brain than early H. erectus (that would be the "frontal looks like KNM-ER 1813 bit). But its cranial bones are thick. The most complete of the bones in the skeleton is a radius, but it's not complete. The best bone for estimating joint surface area is the sacrum; a femur shaft is there, but it falls short of the midshaft length.
And there's a problem: the radius seems pretty big, but the sacrum is little. If it were a human, the radius looks like it came from a body twice the size of the sacrum. There's something going on here. Previous work has assumed that the sacrum is more likely reflective of the size of the body, and the radius is therefore big compared to a small body mass. Maybe that means more climbing, leading to a greater role in weight support for the arms. Or maybe it means a retention of more apelike proportions.
This is a frustrating literature to follow, because pretty much every other early specimen except Lucy (AL 288-1) and the Nariokotome skeleton (KNM-WT 15000) present exactly the same problem. You can't estimate limb sizes very accurately from small pieces of bone. And you can't estimate proportions accurately at all without estimates of size. Plus, it's not clear that you can interpret limb proportions without a decent estimate of body mass. Two years ago, there was a huge go-around about the limb proportions of OH 62. Like KNM-ER 3735, it looks to have a relatively large arm compared to its body. Or maybe the legs are short. Or maybe the estimates are bad. You get the picture. So everybody has a different clever statistical transformation to try to make these fossils comparable to each other. I have no argument with any of the work; but it seems like the error involved in these assessments of proportions is pretty large relative to the information content of the bones.
Here's some of the conclusion from Haeusler and McHenry:
Our analyses suggest that the idea that KNM-ER 3735 had more primitive body proportions than A.L. 288-1 (e.g., Leakey et al., 1989) needs to be refined. We found a unique but distinct mosaic of modern and ape-like limb proportions in the two early hominid species. H. habilis shares a gracile humerus and radius and a small base of the hand phalanges with the earlier A.L. 288-1 and modern humans. In addition, other characteristics, including the relatively small size of the sacrum and a robust midshaft of the phalanges, are common to both early hominids and extant great apes. Surprisingly, however, those upper limb proportions that differ between the two fossil species, such as a robust scapula, a long radial neck, and a long forearm, are all more ape-like in H. habilis.
In KNM-ER 3735, the shoulder muscles that originate on the scapula (trapezius, deltoid, supraspinatus, and infraspinatus) as well as the biceps brachii were, therefore, probably not only more powerful than in modern humans, but also stronger than in A.L. 288-1. On the other hand, the extraordinarily short lever arm of A.L. 288-1's biceps muscle, the minute elbow size, and the small radial head may indicate a weaker arboreal component in its behavioral repertoire than in H. habilis. However, in the absence of modern correlates, caution is needed with respect to possible behavioral implications of the different forearm proportions in the two species.
They also note the Homo-like anatomy of the femur shaft, including a marked pilaster.
Seth Dobson (2005) claimed that that the sacrum of STW 431 (A. africanus) is also small -- it certainly yields a small mass estimate compared to other elements of the skeleton. Heck, all of the early hominid sacra yield small mass estimates. Well, you can see it's confusing.
References:
Haeusler M, McHenry HM. 2007. Evolutionary reversals of limb proportions in early hominids? Evidence from KNM-ER 3735. J Hum Evol (in press) doi:10.1016/j.jhevol.2007.06.001
Dobson SD. 2005. Are the differences between Stw 431 (Australopithecus africanus) and A.L. 288-1 (A. afarensis) significant? J Hum Evol 49:143-154. doi:10.1016/j.jhevol.2005.04.001
John Hawks Department of Anthropology
University of Wisconsin—Madison
Copyright © 2007 John Hawks