Setting the scene The development of aircraft attack on locust swarms in Africa since 1945 and the start of operational research on control systems

The work described here was essentially a search for innovation, to replace existing, totally inadequate, methods of locust control. It was, from the start, quantitative, not only because the experiments would not otherwise be fully meaningful but also because the huge scale of locust control required forward planning, efficiency and economy. From these premises arose emphasis on devising methods of assessment of locust numbers, alive or dead, on measuring everything relevant, and on analysing causes of mistakes and other failures. W hat was wanted was not merely good techniques of killing locusts but complete systems that could be used in a variety of circumstances. Developments after our first moves in 1945 and 1947 are also outlined. They were so successful that less than 30 years later, in spite of wars and other international difficulties, as well of those of the control systems themselves, an exceptionally widespread upsurge of Desert Locusts was checked successfully (Rainey, Betts & Lumley, this symposium). A period of unusual freedom from Desert Locust plagues resulted.