Annual oak (Quercus robur L.) latewood d13C values are presented for the period 1895–1994 from two sites with different hydrological characteristics. Both ring width and d13C indices record high-frequency common forcing better than lower-frequency forcing. At both sites the high-frequency variance in the d13C indices of latewood cellulose is highly correlated with combined July and August environmental variables. The association between the high-frequency variance in annual ring width indices and climate is not as strong. Higher correlations with environmental variables were found for the high-frequency d13C indices at the dry site (Sandringham) than at the wet site (Babingley), but the differences are not statistically significant. These results illustrate the need for routine signal quantification in isotope records and hence a requirement for between-tree replication of isotope series in future studies. High-frequency (year-to-year) interseries correlation is shown to be relatively strong, indicating that only small numbers of replicate series are needed to represent interannual isotope variability accurately. However, common signal variance is diminished at lower (decadal and longer period) frequencies. This implies a need for increased sample replication in order to achieve chronology confidence equivalent to that routinely produced for simple ring width data. This work demonstrates that significant high-frequency climate signals are contained in isotopic measurements of trees whose ring widths contain little or no such information.