The New Zealand divaricates are a collection of shrubs, short trees and tree juveniles whose crowns are made of tough interlaced twigs branching at wide angles and bearing small leaves. These species represent c. 13% of the native woody flora, a proportion not seen in any other region of the world. Since the late 19ᵗʰ century, ecologists and botanists have sought to understand the of drivers this unique case of convergent evolution. Debate has been dominated by two main competing hypotheses invoking (1) the effect of browsing by now-extinct avian herbivores (moa) and (2) a response to frosty and droughty Plio-Pleistocene climates. Observational and experimental evidence, as well as theoretical discussions, have not clearly favoured one over the other. More recently, a synthetic hypothesis involving both climate and browsing has been proposed, but has not been specifically tested yet: the divaricate habit did not become advantageous as an anti-browsing defence until Plio-Pleistocene climatic constraints prevented young trees and shrubs from growing quickly out of reach of ground-dwelling herbivores. These hypotheses imply different expected divergence periods between divaricate species and their non-divaricate relatives. The focus of this PhD project is to produce a dated phylogeny including as many divaricates as possible, with the aim of bringing new evidence to help tease apart the various hypotheses about their evolution. This PhD thesis is built from a previously published paper and draft manuscripts intended for submission to scientific journals. The first two papers report phylogenies of small genera (Pennantia J.R.Forst. & G.Forst. and Corokia A.Cunn.) that include divaricate species. These studies helped develop lab methods and offered the opportunity to test diverse phylogenetic methods on a smaller scale in preparation for the third paper, which is the core dating work of this PhD. The phylogenies of Pennantia and Corokia provided in this thesis (Chapter 2 and 3, respectively) are the first published dated phylogenies of all the species of each genus. Building the phylogeny of Corokia was also the occasion to discuss two theories trying to explain the distribution of extant species on landmasses formerly part of Gondwana: vicariance and long-distance dispersal—Corokia appears to be one of an increasing number of cases where long-distance dispersal is indicated. The dated phylogeny of most New Zealand divaricate species presented in Chapter 4 reveals that, in the great majority of genera with divaricate representatives, the divaricate habit appeared in New Zealand within the last 5 My, i.e. since the beginning of the Pliocene. On one hand, this research constitutes a valuable methodological addition to the field of molecular phylogenetics by (1) disseminating a method for retrieving extra genetic markers, at no marginal cost, from Next Generation Sequencing shotgun sequencing data of DNA samples enriched for a specific set of markers, and (2) developing a guide to using a piece of phylogenetic reconstruction software (treePL), which was missing from the literature and needed by users. On the other hand, dating the emergence of the divaricate habit brought new and crucial evidence to the debate over what promoted the evolution of the divaricate habit in New Zealand: the findings are clearly consistent with a major effect of Plio-Pleistocene climates—and given evidence and discussion from past studies of these plants and similar plants around the world, the effect of browsing by moa was also probably involved.