This paper describes the construction of a 0.5° latitude/longitude gridded dataset of monthly terrestrial surface climate for the period 1901-1996. The dataset includes seven climate variables: precipitation, mean temperature, diurnal temperature range, wet-day frequency, vapour pressure, cloud cover, and ground-frost frequency. The dataset is created by interpolating monthly climate anomalies relative to the 1961-1990 mean from surface climate data, and then combining these anomalies with a 1961-1990 mean monthly climatology to produce monthly climate grids. The primary variables (precipitation, mean temperature, and diurnal temperature range) are interpolated directly from station observations, while the secondary variables (wet-day frequency, vapour pressure, cloud cover, and ground-frost frequency) are interpolated from merged datasets, including station observations and synthetic data derived from primary variables. The dataset is available from the Climatic Research Unit. The new dataset represents an advance over previous products due to its higher spatial resolution, longer temporal coverage, more extensive suite of climate variables, and strict temporal fidelity. The dataset is useful for climate change detection, climate model evaluation, and environmental modeling. The primary variables are derived entirely from observed station data, while the secondary variables are constructed using a combination of observed data and empirical relationships with the primary variables. The dataset is available from the Climatic Research Unit.This paper describes the construction of a 0.5° latitude/longitude gridded dataset of monthly terrestrial surface climate for the period 1901-1996. The dataset includes seven climate variables: precipitation, mean temperature, diurnal temperature range, wet-day frequency, vapour pressure, cloud cover, and ground-frost frequency. The dataset is created by interpolating monthly climate anomalies relative to the 1961-1990 mean from surface climate data, and then combining these anomalies with a 1961-1990 mean monthly climatology to produce monthly climate grids. The primary variables (precipitation, mean temperature, and diurnal temperature range) are interpolated directly from station observations, while the secondary variables (wet-day frequency, vapour pressure, cloud cover, and ground-frost frequency) are interpolated from merged datasets, including station observations and synthetic data derived from primary variables. The dataset is available from the Climatic Research Unit. The new dataset represents an advance over previous products due to its higher spatial resolution, longer temporal coverage, more extensive suite of climate variables, and strict temporal fidelity. The dataset is useful for climate change detection, climate model evaluation, and environmental modeling. The primary variables are derived entirely from observed station data, while the secondary variables are constructed using a combination of observed data and empirical relationships with the primary variables. The dataset is available from the Climatic Research Unit.