A long term (1998-2006) study of annual precipitation and cloud-to-ground (CG) lightning has been made at 31 stations over Taiwan. The CG-lightning data were collected by the ground-based Lightning Location System (LLS) built by Tai-Power Company of Taiwan while the precipitation data were collected from the Central Weather Bureau (CWB) of Taiwan. For the present study, a spatial scale of 0.1° latitude × 0.1° longitude (≈ 102 km2) is selected to determine the flash density. CG-lightning and precipitation data are used to compute the values of "rain yield", defined as the mass of rain produced per CG-lightning flash in units of kg fl- 1 over a given surface area. The rain yield is found to vary considerably with seasonal and climatic conditions, and geographical location. A positive linear correlation is observed between precipitation and lightning flash density with a highest correlation coefficient of 0.70 over inland stations. Out of the 31 stations, 13 stations are inland stations and these stations show higher rain yields clustering close to a mean of 0.7 × 1010 kg fl- 1, compared to the coastal stations which show a mean value 1.4 × 1010 kg fl- 1. When the stations are classified according to seasonal climate zones, the winter and winter-dominant rainfall stations show comparatively higher value of rain yield with an average of 2.8 × 1010 kg fl- 1 than the summer and summer-dominant rainfall stations which exhibit a significantly lower value of rain yield of 2.1 × 1010 kg fl- 1. Inland stations exhibit a lower value of rain yield with a mean of 1.6 × 109 kg fl- 1 and 1.4 × 1010 kg fl- 1 respectively during warm and cold seasons compared to the coastal stations. For each station, the average cold season rain yields are significantly higher than that of warm season values. These differences in rain yield values are attributed to local surface heating which indirectly controls such parameters as cloud base height and convectively available potential energy (CAPE) in the atmosphere. The variation of rain yield with geographical, seasonal, and climatic conditions, found in our observations, are in good agreement with studies found in the literature from other parts of the world.