In an ongoing project an electronic nose is being studied and developed for detection of volatile organic compounds (VOC) emitted from wood colonised and decayed by fungi. The electronic nose consists of an array of gas sensors with different selectivity patterns for different groups of volatile organic compounds (VOC). The use of pattern recognition routines implemented by artificial neural networks (ANN) is used to evaluate data from the sensor array. The responses from the sensor array have been correlated to weight loss of and contents of chitin in the decayed wood samples. The results obtained so far indicate that the electronic nose qualitatively can detect significant differences between sound and decayed sapwood of Scots pine (Pinus sylvestris). Preliminary results also indicate that the electronic nose can detect differences between decay types like brown rot (Lentinus lepideus) and soft rot (Phialophora A). Results from a study aiming at investigating the abilities of the electronic nose to quantitatively detect different stages of decay is now being analysed statistically. The influence on the responses from the sensor array due to variation in relative humidity, moisture content of the decayed wood samples and temperature have also been studied.