Radars have been used successfully for many years to measure atmospher
ic motions over a wide range of altitudes, from ground level up to hei
ghts of several hundred kilometres into the ionosphere. In this paper
we particularly wish to concentrate on the accuracy of these measureme
nts for winds in the middle atmosphere (i.e. 10-100-km altitude). We b
egin by briefly reviewing the literature relating: to comparisons betw
een radar methods and other techniques. We demonstrate where the radar
data are most and least reliable and then, in parallel with a discuss
ion about the basic principles of the method, discuss why these differ
ent regimes have the different accuracies and precisions they do. This
discussion is used to highlight the strengths and weaknesses of radar
methods. Issues like radar volume, aspect sensitivity, gravity wave e
ffects and scatterer intermittency in producing wind biases, and the d
egree by which the intermittent generation of scatterers at quasi-rand
om points in space could skew the radar measurements, are all consider
ed. We also investigate the possibility that MF radar techniques can b
e contaminated by E-region scatter to heights as low as 92-95-km altit
ude (i.e. up to 8-10 km below the ionospheric peak echo). Within all t
hese comments, however, we also recognize that radar methods still rep
resent powerful techniques which have an important future at all level
s of the atmosphere.