rationalskepticism.org

CharlieM wrote:

Rumraket wrote: Common ancestry is evidence for evolution. It might not be direct evidence for darwinian evolution, but evolution nonetheless. Evolution in this context simply meaning change over time through succesive generations.

Additionally, homologous proteins IS evidence for mutations. Though it in itself doesn't specify the cause of these mutations or whether they are random or somehow planned.

CharlieM:
I have no problem with this meaning of evolution and I agree that homology is not evidence for either random or planned change.

This is not what I meant. I meant homology IS evidence of change, but not WHICH of the two(random or planned).
Now, it just so happens that specific homology is a testable prediction of the theory of evolution. Specific homology in this context of the bacterial flagellum meaning that a protein found in the filament/propeller, should have homology to proteins secreted by whatever relevant secretory transport system secreets them in related organisms. It would obviously be highly confusing if the filament proteins were homologous to proteins found involved in something like a specific carbohydrate metabolism vastly unrelated to flagella or secretory systems.
But this is not the case. All the homologies we find, are homologies where we SHOULD find them. This is the testable prediction, this is what makes the testable prediction make sense and this is why the testable prediction so massively supports the darwinian evolutionary hypothesis for the origin of the flagellum.

CharlieM wrote:

Rumraket wrote:
For example the hook, or a pre-hook(before it's function was a hook that transferred torque into the propeller), still attached to the drive and a secretory system, can provide the bacteria with the ability to quickly spray some compound into it's environment... a trick often used by bacteria as a defense mechanism. Many bacteria under physical and chemical stress will eject some kind of ooze that protects it from drying out or acids etc. etc. A rotating spraying mechanism will function superiorly to a stationary one.

CharlieM:
From this it sounds to me as though you think of the hook as rotating like a garden sprinkler. It rotates around its central axis while pointing in one direction.

Ahh yes I can see how it comes out like that. That was badly worded and the garden sprinkler analogy sounds almost exactly like what I described, lol.
What I mean was simply the ability of a pre-hook, a pilus of some sort(before it was under selection for rigidity for the transfer of torque) to be flexible, to be able to move from side to side. Vibrate? I have a hard time pinning down the word.

CharlieM wrote:A universal joint is not something that would aid the ejection of matter a spinning rigid tube would do that much better.

Well yes, a garden sprinkler would be superior to a vibrating/wiggling device, for the delivery of a compound over a larger area. But natural selection has no foresight, remember? We can come up with a billion intermediates with superior function than what they propably had. We have foresight though

CharlieM wrote:The only thing I've ever seen universal joints used for is to transmit torque through a changing angle of axis. Why would evolution select for a complex mechanical device when a simple fairly rigid tube would suffice.

I'm not sure I understand what you mean here.
In any case, I really think you should read that link I gave through carefully. There are several submitted hypotheses for various selective advantages in the individual steps leading to the modern flagellum, and they don't nessecerily involve motility, even at later stages.
Remember the part about the protein homologies being homologues to what we would expect them to evolve from. We find these proteins in the correct places. The proteins of the hook and filaments are homologous to proteins ejected from secretory systems. The proteins of the motor unit are homologous to proteins found in the Tol/Pal machine. Their location in the flagellum is identical to the location wherein we find the Tol/Pal machine on it's own and so on and so forth. The evolutionary model simply makes sense. I honestly don't see these supposedly unreasonable hurdles of extrapolation we have to overcome. The mechanism for the combinations of these molecular machines are duplications and shufflings with positive selection, their later and slower refinement resulting in the modern flagellum is mutation and negative/positive selection. It fits.