This posting was placed on the s.a.p. forum on 3rd June 2004.
Click this link for the full thread for the context...

No I'm not, except perhaps only in why so few still take the most
obvious solution to the question of bipedalism (i.e. wading)
seriously.

> While indeed there's reason to question
> whether or not measurements in a controlled laboratory environment (in
> this case efficiency on a treadmill) mirror measurements in a real
> environment, what he's doing is asserting that a conclusion is invalid
> because there *may* be a discrepency between the lab and nature.  He

> He asserts that if a biped has to pick up its foot slightly more than is
> necessary over completely level ground then the "cost is increased to such
> an extent that Rodman & McHenry's argument...is lost."

> Does he know this?  

Yes.

> Has he done measurements quantifying the reduced
> efficiency of walking over to know that the difference disappears?  

Yes.

> I  sincerely doubt it.  I suspct that he's making an assertion in absence
> of data.  If he's really got some data on the decay in efficiency of a human
> over  uneven ground versus a chimp over similarly uneven ground , why the
> hell hasn't he published it?

Because it's from a pilot study and I need to get a lot more data
before I can publish - you know scientific rigour and all that. But
it's no big surprise. Anyone can try it. All you need is a douglas bag
and equipment to measure VO2. Measure energy usage of walking at slow
speed (say 1 m/s) on a very flat surface, then repeat the experiment
somewhere where the substrate requires the feet to actually be picked
up a few cm from the ground.

It's kind of obvious if you thought about it for one whole second. The
reason human bipedalism is so efficient is *because* of the inverted
pendulum. That's why Crompton et al were so easily able to pour doubt
upon the BHBK gait as a putative gait in a'piths. But, the point
everyone seems to have missed is that for an inverted pendulum to
actually work - that is for the foot in the swing phase (note that
term *swing*) to actually swing through - there cannot be any
obstacles in the way of that swing, right? Or at least if the knee is
flexed to any degree more than it is in humans during the swing phase
any energy gains could easily be squandered away.

My preliminary data (O2/kg/min) are:

Walking normally (avg +/-std)       9.57 +/- 1.46  N=6
Walking in light bush              16.04 +/- 2.97  N=4

This represents a 68% increase in energy expenditure in walking
through an imperfect substrate. As you will know this more than
negates the 45% benefit cited by Rodman & McHenry for bipedal
advantage over chimp quadrupedalism.

In other words human bipedal efficiency can *only* be a factor in it's
evolution in substrates which were flat, firm and vegetation free -
and only then, assuming that the anatomical changes had already been
geared towards taking advantage of this mechanism.

The interesting question is: Where on earth do you get such
substrates? ... Any guesses?

> Yes, treadmills don't exactly mirror ground. But this is true for the
> quadruped as well.  What Algis would like you to believe is that the
> difference in efficiency disappears when you take chimps and humans off
> the treadmill.  This is possible.  He has no evidence of this though and
> there is a possibility that the difference may actually be greater.  Do
> not chimps have to lift up their legs (and arms) to get over branches as
> well?

This is a fair point. I suspect, however, that quadrupedalism is
favoured in such substrates, or at least it does not incur the same
disadvantage as bipedalism. Does chimp quadrupedalism include two
doubled up inverted pendulums? I think not. In fact there is a growing
body of evidence to clearly show that it is not (e.g. D'Aout et al
2004.) There's also quite a bit of anecdotal evidence that moving
quadrupedally through bush is easier than bipedalism. For example...
"The chimps knuckle-walk at about the same pace that I walk behind
them. The difference is they don't change their pace on achingly steep
hills, whereas I slow to a crawl.... But my disadvantage as an upright
walker is that I am several feet too tall to slip easily through the
thorny thickets." (Stanford 2003:16)

> The point isn't even so much that he's not coming to a conclusion based on
> the available evidence.  Perhaps (though he's not adequately voiced it)
> there IS reason to believe that the measurements taken aren't sufficiently
> valid and the difference may not exist.  But he's not saying this.  He's
> saying that there's sufficient reason to believe that that the
> measurements taken aren't suffiently valid and then asserting that the
> difference does not exist.  I suspect that he doesn't actually understand
> the difference though.  That's the sad part.

No, Jason. The point is this: You and others like you have jumped on
the bandwaggon of energetic efficiency as a driver for human evolution
without even asking the most basic question: Is the treadmill a
substrate that mirrors the real world or is it *over-extending*
(remember that phrase?) my argument to just assume that it is?

What I'm actually saying (please don't try to put words in my mouth),
based on primary data, is that there is a big enough difference
between human walking efficiency on a treadmill and on substrates that
negate the benefits of the inverted pendulum effect to overcome the
undoubted benefit that human bipedalism has over chimp quadrupedalism
in perfect substrates (R & McH's whole argument.) My argument is that
this either dismisses the energy efficiency argument for human
bipedalism completely or, at least, confines it into a specific
habitat/location where such benefit could reasonably be expected.

Algis Kuliukas

Refs:

D'Aout, Kristiaan; Vereecke, K; De Clarq, D; Van Elsacker, Linda;
Aerts, Peter (2004). Locomotion in bonobos (Pan paniscus): differences
and similarities between bipedal and quadrupedal terrestrial walking,
and a comparison with other locomotor modes. Journal of Anatomy
Vol:204(5) Pages:353-363

Stanford, Craig (2003). Upright - The Evolutionary Key to Becoming
Human. Houghton Mifflin Company (Boston)