The microclimate of Great Northern (GN) dry bean lines with diverse plant architecture was investigated in terms of white mold (WM) incidence and yield. A split-plot design was used with protected (3 weekly sprays of benomyl 0.9 KG HA-1 after flowering) and unprotected treatments as main-plots and GN lines as sub-plots in a WM nursery (1990, 1991). Canopy density, erectness, leaf area index, and plant characteristics were measured. `Starlight' (upright) and `Tara' (prostrate) were selected for detailed microclimate studies. An infrared thermometer, humidity sensor, and a thermistor were placed within the canopy at the advent of flowering. Leaf wetness and its duration were estimated by the leaf temperature in combination with air temperature and dewpoint temperature. `Starlight' showed later and shorter duration of leaf wetness, lower humidity, and WM and higher yield than `Tara'. Severe WM and reduced yields occurred also on all other susceptible entries with dense prostrate plant habits in the unprotected plots. Fractal analysis was done on the images of the canopy to quantify the light interception within the canopy.