What is a species tree?

Because I maybe expected a lot from the last journal club (Brunes et al. 2010) for us to understand the concept of Species tree, I was a bit disappointed. So, I googled this morning to search for some answers. I found some, though not many. So, you can find below the two sources where I found definitions for "Species tree". Then, I propose a tentative definition and try to cover all the implications I could find when attempting to resolve a species tree. Hope it will help!

Sources:

www.encyclopedia.com

species tree: A phylogenetic tree which depicts the evolutionary relationships of a set of species. Inferred trees, based on gene trees or other character trees, are often presented as estimates of the true species tree.

www.nature.com/nrmicro/journal/v1/n2/glossary/nrmicro751_glossary.html

A phylogenetic tree that represents evolutionary relationships between species as a whole, as opposed to phylogenetic trees for individual genes

A tentative definition of Species tree:

“Species tree” is the conceptual view of the real evolutionary relationships between species. It is opposed to the phylogenetic trees drew from individual genes that give potentially different evolutionary relationships between species.

What does it mean:

I think that the basis of this concept reside in the fact that the evolutionary relationships constructed from individual genes are false in general, maybe very false compared to the species tree. To resolve a species tree, the alternative is to use several or numerous genes (depending on the model), expecting to find a great improvement in the resolution of the tree. I think one can find an ultimate resolution of the species tree if and only if one can find concordant evolutionary relationships between species using different sets of genes. Probably, sometimes it is possible, sometimes not. So, as we saw in this paper of Brunes et al. (2010), finding supports for the different nodes using several genes allowed them to reach the species tree of these tree frogs. I guess the worse cases would be those involving radiations where no support could be found because the relationships between the species represent a polytomy (all lineages diverged at the same time). In such cases, I guess people would keep trying to add more and more genes without getting any concordant resolution.