Now that reduction in beta-cell mass has been clearly established in humans with type 2 diabetes mellitus (T2D), the debate focuses on the possible mechanisms responsible for decreased beta-cell number. Appropriate inbred rodent models are essential tools for this purpose. The information available from the Goto-Kakizaki (GK) rat, one of the best characterized animal models of spontaneous T2D, is reviewed in such a perspective. We propose that the defective beta-cell mass in the GK model reflects mostly a persistently decreased beta-cell neogenesis. The data discussed in this review are consistent with the notion that poor proliferation and/or survival of the endocrine precursor cells during GK foetal life will result in a decreased pool of endocrine precursors in the pancreas, and hence an impaired capacity of beta-cell neogenesis (either primary in the foetus or compensatory in the newborn and the adult). As we also demonstrated that beta-cell neogenesis can be pharmacologically reactivated in the GK model, our work supports, on a more prospective basis, the concept that facilitation of T2D treatment may be obtained through beta-cell mass expansion after stimulation of beta-cell regeneration/neogenesis in diabetic patients.