In addition to the classic allopatric model of speciation in which two lineages diverge through time, a population may “bud off” from an ancestral population and become adapted to a different habitat. The ancestral population (the progenitor or p species) remains largely unchanged in its original habitat while the population that budded off acquires novel characters and becomes the derivative (d) species. P‐d species pairs are appropriate systems for studying plant speciation because they represent recent divergence, making it more feasible to identify the few differences between the taxa and to infer possible early‐evolving barriers to gene flow. The evidence that has been used to designate species as p‐d pairs is reviewed and evaluated. P‐d species pairs are compared for several characters, with ecogeographic factors the only consistent difference between species pairs. Three genera in which p‐d species pairs have been studied in some depth are used to illustrate both the advantages and challenges of studying the process of speciation. It is suggested that there are some advantages in focusing future studies on even earlier stages of divergence, in which speciation is not yet complete, and p and d populations can be identified. Efforts should focus on ever more refined genetic analyses of differences between populations, especially factors reducing effective gene flow between them.