Within the context of the Berlin Aging Study, we examined the distribution of anxiety symptoms and disorders in a representative community sample. The participants were beyond the age of 70 years, thereby extending results from other studies not covering this age range. Additionally, we analyzed the distribution of anxiety symptoms and syndromes not fulfilling specified diagnostic criteria. A sample of 258 old (70 to 84 years) and 258 very old (85 to 103 years) subjects were examined. Anxiety disorders as defined in DSM-III-R and according to clinical judgment (diagnoses termed NOS) were assessed. In addition, items from the Geriatric Mental State-A (GMS-A) covering a wide range of symptoms of anxiety were subjected to factor analysis. The raw score distributions of anxiety subscales obtained by this procedure are examined by age, gender, education, personal living situation, and psychiatric comorbidity. The weighted overall prevalence of anxiety in the elderly community is 4.5% (n = 17), including specified (n = 8) anxiety disorders according to the DSM-III-R and unspecified (n = 9) disorders. Prevalence rates in the younger old were 4.3% and in the older old 2.3%. Weighted prevalence rates for males were 2.9% and for females 4.7%. The most common comorbid disorders were affective disorders in both age groups as well as both genders. Independently of the nosological level, 52.3% reported one or more symptoms of anxiety. Factor analysis of anxiety-related symptoms yielded 5 independent subscales, reflecting hypochondriasis, panic, phobia, worries, and vegetative anxiety. There were more phobic symptoms in the younger age group (P < .001). Except for worries and hypochondriac symptoms, females showed significantly higher anxiety in all other anxiety dimensions. There was no relation between anxiety and cognitive status or socioeconomic status (SES). Only for subjects living alone was more phobic-type anxiety found. Anxiety disorders in old and very old persons are less frequent than other psychiatric disorders of old age, and do not increase with age. Gender differences can still be observed. The symptomatic structure of anxiety seems similar to that found in younger cohorts. Thus, anxiety disorders in old age do not seem substantially different from those in younger age. Their relative contribution to the spectrum of mental disorders seems to decrease, rather than increase, with age, while at the same time anxiety symptoms are an almost daily experience.