The meeting gave a good picture of the main problems being presently pursued in convective reactor heat transfer. Models for the prediction of the flow structures and temperature distributions in fuel subassemblies have made significant progress and are reaching a high level of sophistication. These models cover steady one-phase and two-phase flow as well as transient and accident conditions. Experiments on rod clusters investigating mixing in the cluster and the effect of spacers on it, as well as voidage and velocity distributions in subchannels are providing improved input data. Heat transfer on rough surfaces is still being studied, but no general reliable method for predicting the effect of such surfaces seems to be available yet. New data have been obtained on heat transfer in less common gases. Research on burnout seems to be concentrated mainly on transient situations. Increased activity is noticeable in the research of flow and heat transfer in typical heat exchangers of nuclear plant. In the field of reactor safety the main effort is concentrated on studies of the effect of blockages and accident conditions in fast reactor fuel assemblies and of blow-down and emergency flooding in water reactors. It is probably in this region that further experimentation is most urgently required. Some important problems not covered at the meeting have been pointed out on the previous pages.