An exercise in the combined use of time series on consumers' expenditure in different countries, and in the use of first differences. A comparison of estimates from "within countries" and "between countries" regressions suggests that the former capture mostly short-run effects of income and price changes, the latter mostly long-run effects, and that the differences between these effects are highly significant both statistically and economically. In demand analysis it is therefore essential to specify the period of adjustment; there is no such thing as "the" elasticity of demand. Elasticities derived from time series for separate countries differ significantly from each other and from those derived from combined time series, but this does not invalidate the combined analysis because the estimates for separate countries are often unacceptable.