OBITUARY NOTICE Keith John Worsley (1951--2009) Professor Keith Worsley died on 27 February 2009 in Chicago: he was 57. In November 2008 he was diagnosed with cancer. At the time of his death, Professor Worsley was Professor of Statistics at the University of Chicago and James McGill Professor of Mathematics and Statistics at McGill University in Montreal. He was a world leader in research on the geometry of random images in astrophysics and brain mapping and was recognized with the Gold Medal of the Statistical Society of Canada (2004), Fellowship of the Royal Society of Canada (2003), and Honorary Fellowship of the Royal Society of New Zealand (2008).  George P. H. Styan & David B. Wolfson, McGill University A Photographic tribute to Keith, compiled by George Styan and friends. Professor Worsley is one of the world’s leading statisticians. Over the last 20 or so years, he has made very significant contributions both to statistical theory and its applications, most importantly in the field of human brain mapping. He is a Fellow of the Royal Society of Canada and received the Gold Medal of the Statistical Society of Canada in 2004.  Keith Worsley graduated from the University of Auckland with a BSc (1972), MSc (1973) and PhD (1978). Professor Worsley’s main research is concerned with smooth random fields which, for example, serve as models for the data collected by neuroscientists using Positron Emission Tomography or Functional Magnetic Resonance Imaging. His work has focused on inferential questions to determine which areas of the human brain are active while subjects perform a given task. Professor Worsley’s approach to these questions is a unique mix of probability theory, statistics and geometry. In his work on the maxima of smooth random fields, he uncovered elegant connections between smooth random fields and classical integral geometry. In addition to his work on random fields, Professor Worsley has made important contributions to a number of other areas of statistics, including multiple testing and change point problems. Royal Society of New Zealand