[mrjesse]

Just a short point (not a lot of time tonight):

The statistical software that the paleontologists often use to study fossils is multiple discriminant function analysis. I used it a lot in grad school to study skeletons, so I am familiar with it.

That software is able to work with shapes, and ignore such variables as size. It also does a great deal toward taking the subjectivity out of the analyses.

The size issue that you keep bringing up is not nearly as important as you believe.

Glad to hear. Sounds like some paleontologists don't use the software, though. By the way, what's you're take on the lying issue - are little lies okay, and what's the defining criteria between an okay lie and a not okay lie?

The software sounds interesting. But the integrity of the operator is still an important question - data can be selected carefully before being fed into the software, for example.

I bring up the size issue to demonstrate that even vastly different species, due to in-species variation and grotesque deformations can cause the two species to appear closely related on at least some axis or criteria types.

Regarding the 17 or so blobs of fixall with chimp and human skull fragments stuck on, obviously a lot more human and chimp skulls were found then just those few, I'd say hundreds maybe thousands had to have been found. But I have no idea whether the ones shown are average or whether there really was quite a gap between humans and chips by and large that were bridged just barely by a few grotesque deformations in each group, allowing the line to be drawn.

For example, the below diagram shows just two axis ( for example cranial capacity and jaw length, or whatever) for two different species:(I don't have any two in mind; this is just to demonstrate the principle. I just randomly placed the dots by hand.)

Now if the two species were not related at all, and yet due to genetic drift and grotesque deformation, there were some far-out specimens which brought them close to eachother, then a person could, by picking and choosing just the right ones, form an apparent line of progression, as the red line depicts. And if all you were allowed to see of the above diagram was the dots through which the red line passes, you'd have no way of knowing about all the others - but it sure isn't a fair representation of them, is it!

So I'm not saying here and now specifically that this is the case of the chimp-human fixall lineup, but that I have no way of knowing, and the integrity of the folks who did the said fixall lineup does matter.

By the way, did you work any with the chip to human lineup specimens? I would be delighted to hear some first-hand report about those. Were some of the intermediate pieces the only one? Obviously the modern chimp and modern humans are plentiful, but just how sparse were the intermediates? What was the distribution like?

Thanks,

-Jesse

[Coyoteman]

Regarding your groups of dots in the above post--that is very much what the output from discriminant analysis looks like, but there is no line attempting to connect the dots. Rather what you have are the many dots (cases) separated by their measurements (variables) into one or more groups. If the many cases you input are indeed distinct, the program will group them separately; the distance between the groups shows just how different they are. The outliers can be interesting too. If you have evolution between one group and the next, the outliers should tend to be between the group centroids, not spread about at random. If the cases you input are all members of the same group you will not get two different groups, as your dots illustrate.

By the way, did you work any with the chip to human lineup specimens? I would be delighted to hear some first-hand report about those. Were some of the intermediate pieces the only one? Obviously the modern chimp and modern humans are plentiful, but just how sparse were the intermediates? What was the distribution like?

I did not do any statistics with the fossil man specimens. I did that only on human skeletons.

But the bone lab had hundreds and hundreds of casts. It had all of the famous ones that you see in the photos, such as the one photo you mention. But it had a lot of the bits and pieces as well. It was easy to see why the various fossils were grouped the way they were. You could take an early Australopithecus skull and then see where the various small parts belonging to other individuals of that group fit.

For comparison there were gorilla and various primate bones. What was particularly fascinating was the initial human osteology class. Once you had been through the bones of the skull, during the first few weeks of the class, you could pick up any primate skull and you knew all of the bones! They were of different shapes, and generally smaller, but once you knew the bones from a human skull you know all of the primate cranial bones as well. It was quite a surprise to see how similar they really were.