The development of rodent models for Alzheimer's disease is a critical step for both understanding the disease and developing therapeutic drugs. Transgenic and knockout mouse models will elucidate some important aspects of the etiology of the disease and the development of pharmaceutical treatments. Here, we will focus on the advantages of nontransgenic models. In nontransgenic rat models, intraventricular infusion of A beta 1-40 (alone) generally results in diffuse deposition of A beta with very few focal plaque-like amyloid deposits after a 30-day intraventricular infusion. However, we have recently found that large numbers of scattered A beta immunoreactive plaque-like deposits can be produced in retired female Sprague-Dawley rat breeders using intraventricular infusion of A beta combined with neuropil injection of transforming growth factor beta 1(TGF beta). A beta that was not associated with the large deposits was often immunolocalized with neurons and cell processes. Immunogold electron microscopy demonstrated the presence of A beta in endosome/lysosomes of neuronal processes and glia and basal lamina. In some cases this labeling was clearly in lysosomes of degenerating neurites. This model allows one to introduce A beta and other plaque-associated factors without overexpression of potentially confounding APP domains. We conclude that A beta infusion models will be a useful complement to transgenic approaches to Alzheimer's pathology.