Ecological studies are often hampered by the lack of long-term data sets against which trends or meaningful variability can be measured. This is particularly true when year to year change may be slight and overall variance is high.  Historical records have distinct limitations since technological changes may affect the precision of data acquisition, and differences in observers create additional “noise” in results. Nonetheless, historical data sets, especially those collected using a common methodology and/or by a single individual, can provide at least order-of-magnitude indicators of change. Gilbert White (1720-1793) has been identified as one of the most important founders of modern Natural History and indeed of ecological studies in general.  Besides making extensive observations of plants, animals, soils and geology in southern England, White organized his work around a system of phenology, in which he recorded possible relationships between weather conditions and biological processes.

Results/Conclusions

Much of White’s phenological data are qualitative: occurrence of first and last “hard frosts”, budding of plants, return of migrant birds, etc. but he also includes  daily temperatures and monthly totals of rainfall for the period 1782-1793. White’s journals provide a window into climatic conditions in the last quarter of the 17th century as the Industrial Revolution was beginning –he notes even at this early stage and at a range of over 70 km the "London Smoke" when the wind is from the north-east. Although there are obvious technological limitations to White’s quantitative records, comparison of his phenology with contemporary conditions and records supplied by the U.K. Meteorological Office is suggestive of shifts in meteorological and ecological patterns across more than two centuries. Besides potential insights into climate change and the impact of variation in land-use, the examination and comparison of White’s journals in light of more recent data illustrates the value of systematic long-term Natural History observation in identifying both pattern and variance. I suggest that ecologists in general and climate scientists in particular could benefit from the systematic identification of reliable historical data sets and their pairing with contemporary records, adjusted where possible for changes in methodological accuracy.