[Grut]

Plotted out as a graph, [fossil dimensions] form the classic bell-shaped curve found using data from modern humans.

I'm pretty sure that if you graphed, say, the body masses of a few thousand critters chosen at random you'd get a bell curve, too. Not too good an argument for their all being the same species.

[Blue_Ridge_Mtn_Geek]

Plotted out as a graph, [fossil dimensions] form the classic bell-shaped curve found using data from modern humans.

I'm pretty sure that if you graphed, say, the body masses of a few thousand critters chosen at random you'd get a bell curve, too. Not too good an argument for their all being the same species.

---

Actually, you would not. You would get a multi-modal distribution. For exaggerated example, take random sample of 1000 humans, 1000 dogs, 1000 elephants, 1000 possums, 1000 horses.

The point of the article is not that the distribution of these observations is normal, but that the observations fall well within the normal distribution of human population weights and heights, so it is not evident, at least from these measurements, that they are from different species.

[Doctor Stochastic]

I'm pretty sure that if you graphed, say, the body masses of a few thousand critters chosen at random you'd get a bell curve, too.

It's more likely that you'd get a superposition of bell curves. This would be either multi-modal or maybe just hyperkurtotic.