Loss of norepinephrine (NE) releasing neurons, in the locus coeruleus of the brainstem, is well documented to occur in Alzheimer's disease (AD). However, this process does not necessarily result in decreased release of NE, since compensatory mechanisms may produce increased release of this neurotransmitter. Independent of potential loss of locus coeruleus cells, brain NE levels may be elevated in some persons with AD, both before and during disease progression. Here I examine evidence that elevated, endogenous brain NE is an etiological factor in some cases of AD, and not merely an epiphenomenon of the disease. To explore this etiological hypothesis in AD, I examine the following eight lines of evidence: 1) direct evidence of elevated NE or its metabolites in AD; 2) studies of tricyclic antidepressants, which may principally boost NE; 3) studies of clonidine and other alpha2 adrenergic agonist drugs, which may principally lower the concentration of NE; 4) studies of beta adrenoceptor blocking drugs, including propranolol; 5) comorbidity of AD and bipolar disorder, where both disorders may involve elevated NE; 6) comorbidity of AD and hypertension; 7) comorbidity of AD and obesity; and 8) potential interaction between AD and psychological stress, where stressors are known to release NE. These lines of evidence tend to support the elevated NE etiological hypothesis.