So the ancestor of all Costaceae was a generalist; specialists (bird- and bee-pollinated plants) evolved later. Does specializing on a particular pollinator have any advantages or disadvantages? Do “jacks-of-all-trades” do better in some way if they give up their free-wheeling lifestyle and specialize on one particular thing? Chelsea hypothesized that specializing on a single pollinator-type would increase pollination efficiency which could lead to increased diversification.

Specifically, she thought that a specialist pollination syndrome (bird or bee as opposed to generalist) might act as a “key innovation.” A key innovation is a new adaptation that allows an organism to use a resource or niche that would have been unavailable without the adaptation. For example, as birds evolved the ability to fly (the key innovation), they found themselves with a new set of habitats to occupy and food sources of which to take advantage (the resources). Perhaps evolving red, tubular flowers opened up a whole new range of pollination possibilities or a new level of pollination efficiency for these plants.

The evolution of a key innovation may trigger an adaptive radiation, or a burst of speciation events. Biologists think that key innovations allow a species to access many new and different niches, which may make it easier for a lineage to speciate. This is the idea that Chelsea wanted to test in her plants — that lineages that evolved specialized pollination syndromes would experience a surge of speciation events. If that hypothesis were true, then we’d expect to find many cases in which lineages with specialized pollination syndromes had higher speciation rates than generalist lineages. As shown at right, in order to calculate a speciation rate, you need to know the number of speciation events a lineage has experienced in a particular length of time. So in order to find out if the evidence supported her hypothesis, Chelsea needed to know which clades had evolved key innovations (in this case specialized pollination syndromes), how many speciation events had occurred in each of those clades, and when those clades first evolved. The first two pieces of evidence came straight from the phylogeny — but getting the third piece of evidence (the dates of origination of different clades) required more data.