You had better not look down

What determines the shape of life? Why are there no end of ants and parasitoid wasps but only two coelacanths? At the root of at least some of these questions is a need to estimate variation in rates of speciation and extinction. However, these rates are notoriously hard to estimate. Here Stilianos Louca and Matthew W. Pennell cast a harsh light on such uncertainties. They show that, for any given speciation–extinction model, there exists an infinite number of alternative birth–death models that can explain any given extant time-tree equally well as the candidate model. It is not just that the data we might accumulate are not up to the job—what is startling is that even an infinite amount of data is insufficient to distinguish between any given scenarios that generate a given tree.