Alzheimer’s disease is characterized neuropathologically by senile plaques and cytoskeletal changes. It has been proposed that amorphic plaques would locally induce anterograde propagation of cytoskeletal changes in consecutive neurons followed by amorphic plaque deposition at their axonal terminals. The Alzheimer changes would spread in this way along neural pathways. To test the ‘primary amyloid anatomical cascade hypothesis’, Congo red staining, β-protein/A4 (Aβ) antiserum and Alz-50, which recognizes cytoskeletal changes, were applied to the hypothalamus and adjoining brain areas of five Alzheimer’s disease patients of 40–90 years of age and five age- and sex-matched controls. The results showed that (1) virtually all Aβ plaques in the hypothalamus were of the Congo red-negative amorphic type; (2) amorphic plaques and Alz-50-stained cytokeletal changes were observed not only in all Alzheimer’s disease patients but also in a non-demented, 90-year-old control subject; (3) the density of amorphic plaques in the hypothalamus was unrelated to the duration of the dementia; (4) the density of amorphic plaques was unrelated to that of Alz-50-stained cytoskeletal changes; (5) double-labeling with anti-Aβ and Alz-50 did not show an evident topical relationship between amorphic plaque deposition and the occurrence of cytoskeletal changes; and (6) the distribution of Aβ and Alz-50 staining in five brain areas, for which essential anatomical information is available, did not support the primary amyloid anatomical cascade hypothesis. Amorphic plaques and cytoskeletal changes rather occur independently.