Atkins and Dyl (1997) document a positive relationship between transactions costs and holding periods for stocks in the NYSE and the NASDAQ. We provide additional empirical evidence for the London Stock Exchange using data for all the stocks traded on the FTSE All Share Index between 1990 and 1999. We test for the appropriateness of the chosen OLS estimator against an Instrumental Variables alternative. Diagnostic tests do not reveal any misspecification. We find overwhelming evidence to suggest that there is a positive relationship between transactions costs as approximated by the bid-ask spread and holding periods.