A 1987 survey of the atmospheric and oceanic science community on university instruction in instrumentation and observations revealed that an apparent imbalance existed between observational and theoretical/numerical components of the atmospheric sciences. The committee analyzing the survey results identified several factors contributing to the perceived imbalance. From a follow—up survey conducted of the same community in 1997 to assess the level of change over 10 years, it is found that some progress has been achieved over this period, but that more can be done. Rather than try to implement all of the recommendations from analysis of the initial survey, it is suggested that attention be focused on seven items that build on partial successes over the last 10 years and that offer likelihood of success in the near future.