[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Re: questions about the Odontochelys study

Mike Habib wrote:

> Ah, gotcha.  Very interesting species.  Don't know of any velocity
> measurements for them, but they are pursuit predators and pretty strong.
>  Interestingly enough, it may be the best living outgroup to the matamata -
> with makes ancestral state reconstruction for the punting stuff I was
> talking about before a real pain.  There are a few fossil Chelidae that
> might be able to help (can't recall how much material is there).

If both are fast, then it may be that velocity did not need the
punting. Alternatively, perhaps punting was a further refinement in an
already fast clade  (however, I do not know much about velocity in
other pleurodires and other aquatic turtles as emydids).

> It's the other way around - greater limb length will tend to improve
> punting, to an extent.  Punting is an odd form of locomotion, in that it
> incorporates aspects of both walking and running, from a biomechanical
> standpoint. To the extent that it acts like walking, there is gravitational
> potential utilization, and this improves with limb length.

Ok., punting beneficiates from limb lenght, granted. But what I tryed
to mean is that stride lenght is not correlated with use of
gravitational force for advance, but with excursion ranges at limb
articulations and limb lenght.

> It's the same
> reason that maximum walking speed increases in humans with longer lower limb
> lengths - individuals with shorter limbs transition to jogging gaits as
> lower velocities than those individuals with longer limbs.  Of course, part
> of what makes the whole thing odd is that there is also buoyancy involved,
> which is also a form of gravitational potential, so the calculations get a
> bit messy...

I think that the advantages of employing gravitational potential
energy for advance should be diminished when submerged in water,
because of the greater resistance to advance (and the involved
deceleration would counter aceleration produced by gravity); the
contribution of gravity seems to me to be too slow in that medium.

I think the pendular-like advance/fall of the center of gravity can be
left to gravity at low speeds. I think that at faster speeds, you
retract most actively the extremity to make advance of the center of
gravity quicker, because if not you would have to wait for gravity to
act, which I think would be a little slower. Thus, gravitational
energy should be of less use precisely when at top velocity.

> Different sort of problem, in part because you seem to be referencing top
> speed rather than walking efficiency.

Oops, I though we were talking about the more erected gait of the
matamata in relation to its velocity, when compared with that of
Hydromedusa. I think that except for really low locomotion, in the
water gravity-assisted advance would be of little contribution to
energy saving.

Now a question: ok., so gravity helps advance while the body is
descending, but is not then more energy wasted in elevating the body
again, in relation to animals that move less pendularly? I suppose the
energy employed in elevating the gravity center again should be
proportionately more the slower is the gait, as the inertia that would
help the body to advance would be lesser to counter gravity (you guess
it: I know little of the theme, and had no password to download the
turtle paper you mentioned).

> It's also a tough analogy because the
> lizard introduces the effects of anguilliform motion.

Well, but on the side of the mammal there is dorsoventral flexion
Carrier (1989) mentioned, which helps amplifying the stride too
(perhaps more exagerated in carnivores, but I think it is present at
greater or lower degree in all mammals, so that in marine mammals is
the prioritaty way of locomotion). In any case, I grant there are many
differences that make difficult concluding to which feature the faster
animal owes its greater velocity.

> Seems reasonable enough, but it doesn't actually speak to an adaptive versus
> drift-based paradigm - the aforementioned mechanism works equally well in
> either.

True, heterochronic changes can also have adaptive value.

> Whereas I see the possibility for a distinct advantage in mobility.

Yes, now I get your point on behalf of energy saving (however, I
expressed it above certain doubts of the usefulness of this
gravity-assisted walking on water). I based my skepticism in seeing
that you do not need to be parasagittal to walk in the bottom, not
seem to give greater velocity.

> Just as
> you are correct in asserting that even modest drift can prevail when
> selection is weak, even a small selective advantage can fix a trait - the
> performance benefit doesn't have to be huge, especially if costs are low.

But I think precisely that when selection is weak (that is because the
selective advantage is not great) is when it is less likely to fix
something because drift tends to distort its "achievements" in allele
frequency changes. Even if the slightly more adaptive variety
persists, it may be the result of the aleatory (for practical sakes at
least) drift. I did not know if anybody talk about "modest" or
"strong" drift, I suppose drift is just an stochastic and in aleatory

> In any case, both hypotheses are reasonable.  Testing for drift is
> fiendishly hard in this sort of circumstance, but a functional advantage
> might be tested for with some simple comparative studies of locomotor
> efficiency and/or velocity.

Inded I think one can test if there are differences in possible
movement in some sense in a turtle relative to the other. For example,
perhaps the notches for legs in turtles with less reduced shells can
achieve the same range of motility than in a turtle with a more
reduced carapace, and the range of motion of the limb be restricted at
the glenoid/acetabular articulations.

In the case of the snapping turtle, the plastron seems to be reduced
to a cross, and if that plastron reduction was considered as related
to increase ventral movility range , we would have to hypothesize the
femur can be adducted so as to point ventromedially, towards the other
side of the body! This may thus indicate that the plastron reduction
was at least not so necessary linked to limb mobility, or if it was,
was while the reduction can really permit the leg to be more adducted,
but once the femur cannot be adducted anymore, or is no necessary for
it to do so, the further reduction is possibly non-adaptive.

> I agree; but there are a few cases where plausible selective hypotheses are
> present and worth testing.  Pure drift, by contrast, is difficult to get a
> good handle on (but that doesn't make it untrue).

May be testing for drift is hard, because drift can fix either
adaptations and non-adaptations (this also means that noting that a
character provides an advantage does not demonstrate selection as it
could also have been produced by drift!!). This is a process that can
produce anything, but this does not mean you can not test non-adaptive
hypotheses of character origins.

You can test wether or not some feature represents an improvement for
a proposed function. However, non-adaptive reasons for the presence of
features can also be testable, for example, considering a character as
responding to a simple allometric transformation associated with size
change. For example, the supposed allometric correlation of size and
relative forelimb size in theropods can be refuted with small
theropods with short arms and large theropods with large arms.
Granted, this is not the current case. Hypotheses about independence
of a feature from habitat (which would result from consider the
character independent of adaptation to habitat) can be refuted noting
a biunivocal correlation of the character with only some kind of

At the end, we can test alternative hypotheses of adaptation, but we
cannot test if a character was adaptive or not, because there will
always be the possibility that the character is an adaptation to
something yet unknown or unimagined. Thus, you can not test if the
character was an adaptation, you can only test if it was an adaptation
to more efficiently use a resource in particular (and you can thus not
test whether or not there was selection, but if ther was a particular
kind of selection).