Abstract: The last few years have seen important developments in our understanding of the dementias. These include advances in genetics, pathology, imaging and biomarkers. In 2009 two large genome-wide association studies of Alzheimer's disease (AD) reported three new risk-factor genes. Unravelling the effects of these genes may offer new insights into pathogenesis. The debate about the relative pathogenic importance of τ and A β continues, however amyloid oligomers are increasingly recognised as potentially critical in a toxic cascade. Recent population-based autopsy studies have revealed the extent of the associations between neocortical amyloid plaques and dementia (weaker with increasing age). There is a growing emphasis on early detection of AD. A remarkable advance has been the ability to image cerebral amyloid in vivo using PET ligands. This is proving very sensitive for AD pathology: virtually 100% of AD patients have positive scans; however so do almost 30% of elderly individuals. CSF markers are now well-validated: reduced CSF-Abeta and raised CSF-tau carry strong positive predictive value for AD even at an early stage. These developments have contributed to the publication of new criteria for AD which include imaging (MRI and PET) and CSF markers as supportive features for a clinical diagnosis earlier than previous criteria permitted. These fit with evidence for a long preclinical period to AD characterised by increased rates of brain atrophy (particularly medial temporal lobe); cerebral hypometabolism; amyloid deposition; CSF changes and subtle cognitive decline. Just as significant have been the advances in non-AD dementias. The Progranulin (GRN) gene has joined the MAPT (microtubule-associated protein τ) gene as an (equally) important cause of dominantly inherited frontotemporal dementia (FTD). The pathological classification of FTD has also been changing rapidly with discoveries of new histopathological markers. Clinical criteria for dementia with Lewy bodies (DLB) continue to be refined and now incorporate supportive imaging features. Disappointing, despite an increasing number of drug-development strategies (at least for AD), a number of recent trials of disease-modifying therapies have reported negative results. In this context there is increasing interest in trialling treatments earlier and using the biomarkers mentioned above to identify for inclusion individuals at an early (even preclinical) stage of AD and then to use clinical, psychometric and imaging markers to assess efficacy. Recent advances have improved our ability to understand the dementias and to make early and accurate clinical diagnosis. Treatments that significantly ameliorate the clinical course of these diseases are urgently needed.