1. Long-term clinical and biochemical vitamin C (ascorbic acid and dehydroascorbic acid) status and dietary intake of vitamin C were monitored for 18 months in twenty-three relatively-healthy elderly subjects living at home in the north of England.
2. Plasma vitamin C showed a strong positive correlation with buffy-coat vitamin C both cross-sectionally between subjects and longitudinally within subjects; plasma levels, therefore, were almost as good an index of long-term status as buffy-coat levels.
3. Vitamin C intake was strongly correlated with plasma and with buffy-coat levels both between subjects and within subjects. This contrasts with the poor correlation observed between riboflavin intake and biochemical riboflavin status in the same subjects.
4. Subjects with relatively high average intakes showed considerable seasonal variation, and several widely-spaced measurements would be needed to characterize their long-term status accurately. Those with low average intakes and blood levels showed little variation during the study.
5. The strongest within-subject correlation was obtained by relating the biochemical values to the 7 d dietary intake directly preceding the biochemical analysis. Blood and tissue levels therefore appear to be strongly related to the current dietary intake.
6. Although some subjects had blood vitamin C levels consistently within the region associated with biochemical deficiency, none showed clinical evidence of specific deficiency symptoms.
7. After the main study, fifteen of the subjects received supplementary vitamin C for 2 months. Plasma and buffy-coat levels rose sharphy, but fell to presupplementation levels within 1 month of withdrawal, emphasizing the transitory nature of increased tissue levels. No significant changes were detected in the following collagen-related urinary ratios: hydroxyproline: creatinine, proline: creatinine, proline: total amino-nitrogen and proline: hydroxyproline in hydrolysates either of whole urine or of various fractions. These variables thus appear to be insensitive to short-term changes in vitamin C status over the ranges encountered in this study.