Best Before DNA, taxonomists use detailed comparative anatomy. As Louis Dollo, one of the greatest anatomists of all time, points out, evolution is irreversible.
Dollo's principle states:
"An organism is unable to return, even partially, to a previous stage already realized in the ranks of its ancestors.". According to this hypothesis a structure or organ that has been lost or discarded through the process of evolution will not reappear in exactly the same form in that line of organisms."
It means, that, no matter how closely two organisms may resemble one another, the products of convergent evolution can never resemble one another in every single detail. By examining them closely, evidence of separate ancestry will always reveal themselves. Dollo's principle has served taxonomy very well, although some taxonomists have unwisely abandoned comparative anatomy and rely instead on superficial similarity. They claim, for example, that bird hands must be the same as theropod hands because both of them have 3 fingers. Comparative anatomists, OTOH, would point out that bird hands have fingers 2-3-4 because of evidence from developmental biology and the 3 fingers of theropod hands are 1-2-3 based on fossil evidence, since primitive theropods such as Eoraptor and Herrerasaurus have vestigial fingers 4 and 5. Therefore, using Dollo's principle, we have determined that birds and theropods have evolved their 3 fingered hands independently of each other from a common ancestor with a 5 fingered hand.
Finally, to answer the question of whether snakes and worms share a recent common ancestor, we apply Dollo's principle by examining both animals in detail. Snakes have a vertebral column. They have jaws. Snakes have a lung and they have scales made of beta keratin. These structures are absent from worms. OTOH, these same structures are found only in vertebrates, but not in all vertebrates. The particular subgroup of vertebrates that have these uniquely shared characters are in fact the reptiles. Snakes therefore are reptiles, not worms. Further, since snakes and lizards share the unique character of paired hemipenes, snakes are in fact close relatives of lizards. Using Dollo's principle to study worms, we can also determine that not all worms share a recent common ancestor either. Some worms for example, have a coelom, but some have a pseudocoelom while others, such as the flat worms, have no coelom at all. Using comparative anatomy, it has been determined that annelid worms are in fact more closely related to arthropods and molluscs than they are to the round or nematode worms.
Dollo's principle has in fact resulted in a highly stable and useful classification. However, many contemporary taxonomists are either ignorant of it or disdains its use. The result is taxonomic chaos generated by those who fail to realize how powerful and useful Dollo's principle really is.
