Cerebral amyloid angiopathy (CAA) is a significant risk factor for hemorrhagic stroke in the elderly, and occurs as a sporadic disorder, as a frequent component of Alzheimer's disease, and in several rare, hereditary conditions. The most common type of amyloid found in the vasculature of the brain is beta-amyloid (A beta), the same peptide that occurs in senile plaques. A paucity of animal models has hindered the experimental analysis of CAA. Several transgenic mouse models of cerebral beta-amyloidosis have now been reported, but only one appears to develop significant cerebrovascular amyloid. However, well-characterized models of naturally occurring CAA, particularly aged dogs and non-human primates, have contributed unique insights into the biology of vascular amyloid in recent years. Some non-human primate species have a predilection for developing CAA; the squirrel monkey (Saimiri sciureus), for example, is particularly likely to manifest beta-amyloid deposition in the cerebral blood vessels with age, whereas the rhesus monkey (Macaca mulatta) develops more abundant parenchymal amyloid. These animals have been used to test in vivo beta-amyloid labeling strategies with monoclonal antibodies and radiolabeled A beta. Species-differences in the predominant site of A beta deposition also can be exploited to evaluate factors that direct amyloid selectively to a particular tissue compartment of the brain. For example, the cysteine protease inhibitor, cystatin C, in squirrel monkeys has an amino acid substitution that is similar to the mutant substitution found in some humans with a hereditary form of cystatin C amyloid angiopathy, possibly explaining the predisposition of squirrel monkeys to CAA. The existing animal models have shown considerable utility in deciphering the pathobiology of CAA, and in testing strategies that could be used to diagnose and treat this disorder in humans.