The procedure for the development of a measure of humour appreciation among a convenience sample of 53 primary school children attending a one-week summer program held in Hong Kong is described. Two sets of visual stimuli depicting pictures of local celebrities portrayed either veridically (congruously) or incongruously were established from a pool of 30 pictorial items. The pictures were ranked for their humour properties yielding a 16-item test battery. Analyses of children's responses clearly indicated that incongruous pictures were rated as significantly funnier than congruous pictures. (Contains 4 tables.)