A review is given of the aircraft instrumentation used for cloud physics investigations. Below the cumuliform cloud the nature of the convection is examined by means of rapidly acting thermometers, hygrometers, hot wire anemometers and aircraft accelerometers. In the lowest 1000 feet where the lapse-rate is slightly super-adiabatic, continuously turbulent conditions are found. Above this bubbles or pulses 300–600 feet across rise sporadically in an environment whose lapse-rate is approximately dryadiabatic. Just below the cloud base the lapse-rate is usually stable and long wave fluctuations of temperature in the horizontal are found. Within the cloud temperatures are difficult to measure accurately. Attempts to use vortex thermometers for this purpose have been unsuccessful. Vertical currents can be estimated roughly from aircraft displacements. Water content in cumuliform cloud probably attains local values of about 5 g/m3 for short periods, but up to the present none of the instruments developed have been capable of measuring amounts of this magnitude. The progress made so far in developing hot wire water content meters and refrigerated icing disc meters is described. Cloud particles can readily be sampled from aircraft when they are small but satisfactory methods have not yet been developed for the larger particles (e.g. hail, snow and rain). Droplet sampling over England has shown the presence of large droplets (100 microns diameter or more) in moderate sized cumulus clouds. Examples are shown of rain being produced by stratiform warm sector clouds with tops well below the freezing level and precipitation in large cumulus clouds apparently preceeding the ice crystal stage. A comparison between average cloud top temperatures for clouds which produce rain and for those whose tops are composed of ice crystals rather than supercooled droplets suggests that rainfall in England may often be produced in clouds in which theBergeron mechanism does not operate.