Joined: Jan. 2006
It appears that Dr Hunter is confused in his argumentation. I’ll try to help him with my best English.
He seems to argue that convergence is a problem for the theory of evolution while, actually, he shows us that convergence is merely a problem for the inference of reliable phylogenies, which is a completely different issue.
Dr Hunter, if you want to use convergence as an argument against the ToE, you'll have to prove that convergent evolution is impossible. You haven't done anything of the sort, yet. We await.
Regarding the phylogenetic inference: yes, homoplasy (be it by convergence, reversion or parallelism) can be problematic. In fact, the whole field of cladistics aims at resolving issues due to homoplasy. If hompoplasy didn’t exist, there wouldn’t be any researcher in phylogenetics, we would just use the good old principle of parsimony to infer phylogenies that would always be 100% accurate.
But how do we solves the problems?
First, we use our own experience and logic. We know, from experience, that some poorly defined traits like gross morphology can be extremely labile and cannot be used to support common ancestry (otherwise, dolphins would belongs to the fishes, wouldn't they?).
However, there are been confusions, especially in taxonomy, for example when defining the superclass of "pisces", separate from tetrapods, while in fact, we are more closely related to the trout than the trout is related to a shark. But the anatomy of the skeleton indicated us the correct topology. Sharks (chondrichthyes) don't have any bones, while humans and trouts (osteichtythes) do. The fish-like shape can be easily explained by the fact that fish have to swim, but why would humans and trouts have bones, if not by common ancestry?
Another way to get around homoplasy is using data from paleontology. We know, by the fossil record, that tetrapods evolved from particular osteichthyes, sometime in the Devonian, while chondrichthyes (sharks) already existed. It confirms us that tetrapods and other osteichthyes share a more recent ancestor than osteichtyes and chondrichthyes.
Lastly, we now mainly resort to DNA sequences to infer phylogenies. In this case, all characters (nucleotides) are considered equal. Homoplasy is still common, but researchers have defined models of evolution which are used to detect the most probable tree. And those trees, with a very few exceptions, confirm the phylogenetic relationships that were established thanks to anatomy and paleontology (chondrichthyes and osteichthyes for instance).
I hope that helps.