Everett suggested that there’s no such thing as wavefunction collapse. He hypothesized that for an idealized spin measurement the apparatus evolves into a superposition on the pointer basis of two apparatuses, each displaying one of the two outcomes which are standardly thought of as alternatives. As a result, the observer ‘splits’ into two observers, each perceiving a different outcome. There have been problems. Why the pointer basis? Decoherence is generally accepted by Everettian theorists to be the key to the right answer there. Also, in what sense is probability involved, when all possible outcomes occur? Everett’s response to that problem was inadequate. A first attempt to find a different route to probability was introduce by Neil Graham in 1973 and the path from there has led to two distinct models of branching. I describe how the ideas have evolved and their relation to the concepts of uncertainty and objective probability. Then I describe the further problem of wavefunction monism, emphasized by Maudlin, and make a suggestion as to how it might be resolved.