A new paradigm for the pathogenesis of ovarian cancer has recently been proposed which helps to explain persistent problems in describing the development and diverse morphology of these neoplasms. The paradigm incorporates recent advances in our understanding of the molecular pathogenesis of epithelial 'ovarian' cancer with new insights into the origin of these tumors. Correlated clinicopathologic and molecular genetic studies led to the development of a dualistic model that divides all the various histologic types of epithelial ovarian carcinomas into two broad categories designated 'type I' and 'type II'. The prototypic type I tumor is low-grade serous carcinoma and the prototypic type II tumor is high-grade serous carcinomas (HGSCs). As the serous tumors comprise ∼70% of all epithelial ovarian tumors and account for the majority of deaths, the serous tumors will be the subject of this review. There are marked differences between the low-grade and high-grade serous tumors. Briefly, the former are indolent, present in stage I (tumor confined to the ovary) and develop from well-established precursors, so-called 'atypical proliferative (borderline) tumors,' which are characterized by specific mutations, including KRAS, BRAF and ERBB2; they are relatively genetically stable. In contrast, HGSCs are aggressive, present in the advanced stage, and develop from intraepithelial carcinomas in the fallopian tube. They harbor TP53 mutations in over 95% of cases, but rarely harbor the mutations detected in the low-grade serous tumors. At the time of diagnosis they demonstrate marked chromosomal aberrations but over the course of the disease these changes remain relatively stable. Along with the recent advances in understanding the molecular pathogenesis of these tumors, studies have demonstrated that the long sought for precursor of ovarian HGSC appears to develop from an occult intraepithelial carcinoma in the fimbrial region of the fallopian tube designated 'serous tubal intraepithelial carcinoma (STIC)' and involves the ovary secondarily. Another possible mechanism for the development of ''ovarian'' HGSC is implantation of normal fimbrial epithelium on the denuded ovarian surface at the site of rupture when ovulation occurs. We speculate that this tubal epithelium can result in the formation of a cortical inclusion cyst (CICs) that can then undergo malignant transformation. Thus, serous tumors may develop from inclusion cysts, as has been previously proposed, but by a process of implantation of tubal (müllerian-type) tissue rather than by a process of metaplasia from ovarian surface epithelium (OSE, mesothelial). The dualistic model serves as a framework for studying ovarian cancer and can assist investigators in organizing this complex group of neoplasms. In conjunction with the recognition that the majority of 'ovarian' carcinomas originate outside the ovary, this model also facilitates the development of new and novel approaches to prevention, screening and treatment of this devastating disease.