Fifty years ago the processes responsible for the formation of the ionospheric layers were well established, Appleton, Bartels and Chapman having been in the vanguard of the exploration phase of ionospheric research. Routine observations were being carried out in some locations, but in other areas, such as Antarctica, the ÿrst measurements were only being undertaken. Thus, a global perspective of the ionosphere was still emerging. In the intervening years, there have been many very signiÿcant ÿndings. Energy from the atmosphere both above and below the ionosphere has been shown to have a major in uence its structure and dynamics. Winds up to 1 km s -1 blow at F-region altitudes and the solar wind imposes a cross-polar cap potential ¿ 100 kV on the ionosphere when the interplanetary magnetic ÿeld has a strong southward component. The ionosphere itself is not simply a passive window on atmospheric and plasma processes. It is the major source of plasma for the magnetosphere, and its lower boundary is critical in the re ection of planetary waves back into the stratosphere and troposphere. Evolving technology has made a vast improvement in our ability to observe and model the ionosphere; some key developments are summarised. The paper also provides a few illustrations of the topics studied over the last 50 yr, namely F-region storms, plasma irregularities and long-term changes. Although the climatology of the ionosphere is well understood and can be modelled successfully, predicting the ionospheric weather accurately is not yet possible. The paper concludes by identifying some key topics where further understanding will be essential if reliable modelling and forecasting are to be achieved.