This article focuses on some of the long-standing problems that exist in the classification of the vascular dementias. To clarify the definition of vascular dementia, mental status and other cognitive processes were investigated in 12 patients with a single cerebral infarct (mean age, 74.8 years), 17 patients with multiple cerebral infarcts (mean age, 71.4 years), 21 vascular patients with no cerebral infarcts (mean age, 76.9 years), and 16 demographically equivalent normal elderly persons (mean age, 70.4 years). The following null hypotheses/hypotheses were tested: (a) The cognitive impairment of single-infarct, multiple-infarct, and noninfarct vascular patients is not/is significantly greater than and outside the range of that in normal aging, and (b) there are/are not significant differences between single-infarct, multiple-infarct, and noninfarct vascular patients on mental status and other cognitive measures. Measures used included the Mini-Mental State Examination, Dementia Rating Scale, Western Aphasia Battery, and Boston Naming Test. Results indicate that vascular disorders involve a decrement in mental status and other cognitive functions that is significantly greater than the age-associated cognitive impairment of normal aging. Also, results from the study indicate that there were no robust, reliable significant differences between single-infarct, multiple-infarct, and noninfarct patients. The validity of the distinction of focal versus generalized or diffuse cerebral lesions is questioned. The nosologic entity of noninfarct vascular dementia is introduced. The future nosology of vascular dementia should provide for noninfarct vascular dementia, defined as vascular dementia caused by underlying vascular mechanisms other than cerebral infarction.