Monotypic Taxa

A monotypic taxon is one that contains only a single subordinate taxon. It is most commonly used to describe genera that contain only one species. It can also be used for subfamilies that contain only a single genus (even if that genus contains more than one species, and is therefore not monotypic itself). In theory a species would be called monotypic if it does not contain subspecies, but in practice this is rarely, if ever, done (at least in ants).

From an evolutionary perspective, a monotypic taxon has relatively low diversity (contains only a single child taxon) for its degree of divergence from close relatives (it is highly distinctive from sister taxa). In practice, the "degree of divergence" involved is somewhat arbitrary and is largely based on creating a useful classification system. This classification system is restrained by phylogenetic relationships, but judgement does play a role.

An Example
As a real-world example, these are the relationships among the genera within the subfamily Pseudomyrmecinae:

Myrcidris contains a single species and is thus monotypic.

And these are the relationships among selected species within this subfamily:

If we hide the taxon names and only look at the relationships, it's not immediately obvious that there are three genera included in this diagram. To appreciate a classification system consisting of three genera we need to add "branch lengths", or a measure of the differences among these taxa. In simple terms, all of the Tetraponera species "look similar" in that they share a large number of morphological (and genetic) characters, as do the species of Pseudomyrmex. The single Myrcidris species differs from both and it is reasonable to place it in a separate genus from Pseudomyrmex. However, "reasonable" is undefined and is based on the judgement and knowledge of the taxonomist making these decsions. One could have argued just as well that Myrcidris is just a "strange" Pseudomyrmex and should not be treated as a separate genus. The available phylogenetic information does not impact on this choice as the relationships remain the same in both cases.

To complicate things further, if a strictly phylogenetic classification system were to be adopted, all three genera would be placed together in Pseudomyrmex (the oldest name) as currently Tetraponera is paraphyletic relative to the other two genera. This would mean that Myrcidris is not actually a monotypic genus, but rather a species of "Pseudomyrmex" that evolved after an African species (of Tetraponera) dispersed to the Americas, leading to the diversification of a New World clade of what is currently called "Pseudomyrmex".

However, within the subfamily Pseudomyrmicinae, it is very useful to recognise an Old World clade (Tetraponera) separate from two New World clades, one monotypic (Myrcidris) and one species-rich (Pseudomyrmex). This is in some ways similar to recognising "dinosaurs" as a group separate from "birds", even though birds are phylogenetically dinosaurs. Both of these classifications are useful even if inaccurate.