Hi thereal and Bgood,
Thanks for your replies, but you still have not answered the question about the testability of natural selection as a mechanism for evolution.
Natural selection requires variation in a trait, fitness differences in consistent relation to the trait, and the trait be heritable. So while you showed the standard evolutionary story of how environmetal pressures act on organisms which then somehow are seen as better suited to that environment, it still does not show testability. The variations have to occur in the loci that regulates the control network of the organistic ontogeny. So what is needed to test natural selection, is to demonstrate the presence of heritable variation throughout early ontogeny, and to show thus how mutagenesis can be the basis for adaptive variation.
thereal:
Your idea of no offspring surviving in the long run is incorrect in terms of gene flow. Obviously, no individual is immortal, but if if an individual reproduces successfully it has passed on some of its genetic material, so in this way the aspects of successful individual do outlive the actual individual. The most successful individuals are not those who live the longest, but those who produce the most offspring that then reproduce themselves.
While that is a nice theory that I have heard several times, it requires some nifty footwork to accomplish it over a complete biocosm. And it repeats back to me what I said, that the successful individuals are the ones that procreates best, while the ones that procreates best are the successful ones. But it does not explain how it tests natural selection.
You have two species of deer whose ranges meet in the the middle of a country, call them species 1 and 2. Natural selection would predict that each species would be best suited to their local habitat, so let's say species 1 is best suited to survive the rugged mountainous terrain around the Rocky mountains while species 2 is best suited to survive the warmer climate of the Southeast U.S. By looking at their reproductive strategies across their respective ranges we can make inference to how natural selection has shaped the two species. We would predict that species 1 should have thicker fur, bigger body size, etc. to handle the colder climate, while the converse should be true for species 2; where the ranges meet, we may predict that an intermediate condition is favored. The next step would be to actually look at these species to see if the predictions of morphology are correct, as well as look at their reproductive strategies to see if, for example, the furriest individuals are the most reproductively successful in the Rockies and the smaller individuals are the most successful in the Southeast. Also, if we see a gradual transition whereby the two species are very similar where there ranges meet, that would provide further support for our assumptions of natural selection shaping these species.
I have several issues here, but will stivck with the main one for now. How does natural selection predict anything? It has no transcendent properties, it is merely an arandom blind process acting on current impulses. The same question applies to how it can shape anything?
Darwin himself said:"It may be said that natural selection is daily and hourly scrutinising, throughout the world, the slightest variations; rejecting those that are bad, preserving and adding up all that are good; silently and insensibly working, whenever and wherever opportunity offers, at the
improvement of each organic being in relation to its organic and inorganic conditions of life.
Some have even imagined that natural selection induces variability, whereas it implies only the preservation of such variations as arise and
are beneficial to the being under its conditions of life. . . . In the literal sense of the word, no doubt, natural selection is a false term."
From the quote it is clear that even Darwin knew that natural selection only preserved changes, but does not itself have the ability to create variability.
From an evolutionary perspective, one way of testing whether these two species evolved from one another or a common ancestor would be to look at genetic data. If, for example, you look at a certain sequence of DNA and find that that sequence becomes more and more different as you move towards each coast from the center of the country, it could support a theory that at one point an ancestral species existed where the ranges meet in the middle. This theory could also be tested by determining whether individuals that live where the ranges meet can interbreed while those on the coasts cannot; if this is observed, it would support a theory that the species may have been one and gradually split due to expanding ranges and different environmental conditions. You can also use genetic data to test whether these species came from a common ancestor that may still be present, for if the genetic information between species 1 and 2 is similar where there ranges meet AND similar to a third species that inhabits this area, one theory may be that these two species evolved from the third species.
A fine description of micro-evolution, but still does not say anything about how homology of DNA accounts for the testability of natural selection. In fact, I have in another thread argued that genetic homology is not testable or falsifiable since there are no defined margins of variation that would falsify it.
Bgood:
The fossils themselves are evidence. The lineage and relations between them is an application of the theory and is therefore not evidence, and as you said cannot be tested to a certain extant. The only way to test it is for future fossil findings to fall within reasonable lines. Seeing that the evidence is not complete there is always leeway for changing divergence dates etc. Making testing very questionable. However if lets say homo sapien bones are found among dinosaur bones then it presents problems. But I see what your saying about testability.
The fossils themselves are evidence only of the presence of an organism at a specific point in time. Fossil lineage is considered the evidence for gradual changes over time, and I am yet to see how that can be tested.
We know this because offspring die before reaching sexual maturity. If organisms do not produce enough offspring to replace the current generation logically there will be no future generations.
I'm not sure how that answers my question, but ok. If all life produced more than could be supported, why do we see extinction? Logically, all organisms should survive if all life produced more than could be supported. I was just curious, as I always am when I see the word "all".
This only requires survival long enough to procreate.
Individuals are unique. We know some mutations are harmful, this may cause an individual to perish before reaching sexual maturity, thus removing the harmful mutation from the gene pool. Mutations which are not harmful will pass on to the next generation.
So are you saying that harmful mutations do not pass on, and precludes the organism from procreating? To suggest natural selection happens purely by eliminating harmful mutations through lack of procreation does not explain it when tested against empirical observation.
The collection of nonharmful mutations increases and changes the gene pool throughout time. This allows a population to have more variety(traits).
But that still does not answer my question, how does natural selection account for biochemical pathways to new species? No-one is debating that within species differentiation happens, but it does not account for new phyla, for example. And while nonharmful mutations increase, since the overwhelming majority of mutations are neutral or harmful, those will increase at the same rate.
It is easy to test. Select a trait you wish to give selective advantage. Lets say big feet. Now raise some mice and actively select for large feet. After several generation you should expect to see the average foot size increase. Of course this will only work on traits which actually exist.
I guess all of your preceding info can be captured here. The problem I have with this is that if I select a trait, that is adding an outside influence or purpose into the process, and then I agree, it can easily be done. But natural selection does not have purpose, or any other transcendental characteristics, so while your experiment may work, it does not test natural selection, it tests Bgoods selection. And why should it work for traits only actually exist? Surely genetic manipulation will allow you to grow a gills on a mouse, since it is genetically homologous with sharks, through a common ancestor?
In all, the examples which you spent so much time writing, which I appreciate, did not show the testability of natural selection as a mechanism for evolution. It demonstrated microevolution nicely, described genetic variances on which natural selection can or cannot act, but not how that can be tested, merely how it is supposed to work. It does not account for why or how procreation happened in the first place, and how mutations of the simplest lifeforms lead to the complexity of life seen today, how does it account for eyes, ears, gender etc? After all, natural selection only preserves, without purpose.