torsdag 10 maj 2012

The problem with clades

The problem with clades is that each of their internal branches (ie, between nodes) consistently represents two entities: one descendant and one ancestor, because it means that these branches consistently contain changes between character states and thus can possess mutually exclusive properties. This fact is difficult to interpret, but among other things, it means that a "true" clade, like the idea of a "true tree of life", is shorter than the "most parsimonious" such tree. However, since this state is impossible, the practical implication of the fact is instead that there are several just as parsimonious trees. The fact thus means that the idea of a "true" clade in practice is a matter of gray scale rather than of black and white. Every bifurcating process that has to be represented with at least one internal branch does in fact offer several just as parsimonious tree illustrations, all of which are contradictory. There is thus no consistent tree illustration of any bifurcating process that has to be represented with at least one internal branch to be found.

This impossibility may be difficult to understand, but the alternative is to search for something that can't be found.  

Inga kommentarer:

Skicka en kommentar