Twenty years ago there were large disagreements between instruments measuring stratospheric H2O and NO2, and there were no reliable long-term records. Now, there is greatly improved agreement between techniques, there are 20-year records of profiles of stratospheric H2O and of total NO2 at single sites, there is a qualified H2O record extending back 40 years and 12-year records of total NO2 from many sites, and there are reliable global measurements from satellites to form the basis of climatologies. This excellent progress is marred by a discrepancy between the observed trend in lower stratospheric H2O and temperatures at the tropical cold point, and by a possible discrepancy between the observed trend in total NO2 and the trend in the source of NO2. The second discrepancy would be resolved by a trend in the residual circulation in the stratosphere, in the same way as variability in the residual circulation was responsible for variability in the trend in H2O in the upper stratosphere in the 1990s.