Alzheimer's disease (AD) is considered to be the number one health problem and seems to be reaching epidemic proportion in the USA. The cause of AD is not known, a reliable animal model of the disease has not been found and appropriate treatment of this dementia is wanting. The present review focuses on the possibility that a virus or exogenous toxic materials may gain access to the CNS using the olfactory mucosa as a portal of entry. Anterograde and retrograde transport of the virus/zeolites to olfactory forebrain regions, which receive primary and secondary projections from the main olfactory bulb (MOB) and which, in turn, project centrifugal axons to the MOB, may initiate cell degeneration at such loci. Pathological changes may, thus, be initially confined to projecting and intrinsic neurons localized in cortical and subcortical olfactory structures; arguments are advanced which favor the view that excitotoxic phenomena could be mainly responsible for the overall degenerative picture. Neurotoxic activity may follow infection by the virus itself, be facilitated by loss of GABAergic terminals in olfactory cortex, develop following repeated episodes of physiological long term potentiation (which unmasks NMDA receptors) or be due to excessive release, faculty re-uptake or altered glutamate receptor sensitivity. Furthermore, a reduction in central inhibitory inputs to the MOB might then result in disinhibition of mitral/tufted neurons and enhance the excitotoxic phenomena in the MOB projecting field. Within this context, and in line with recent studies, it is believed that pathology begins at cortical (mainly olfactory) regions, basal forebrain neurons being secondarily affected due to retrograde degeneration. In addition, failure to produce a critical level of neurotrophic factors by a damaged MOB and olfactory cortex, could adversely affect survival of basal cholinergic neurons which innervate both regions. Support for these hypothesis is provided, first, by recent reports on pathological findings in AD brains which seem to involve preferentially the olfactory and entorhinal cortices, the olfactory amygdala and the hippocampus, all of which receive primary or secondary projections from the MOB; secondly, by the presence of severe olfactory deficits in the early stages of the disease, mainly of a discriminatory nature, which points to a malfunction of central olfactory structures.