The ground based measurement techniques for the determination of the radar back-scatter of vegetation and soils as used in The Netherlands will be described. Two techniques are employed: one covering a large sample area (> 1000 m2) but working at low grazing angels only and one (short range) covering a small sample area of about 1 m2 only, but working at higher grazing angles. They will be compared with the requirements: 1. that all samples investigated must contain sufficient scatterers to give a Rayleigh distribution at the output of the sensor and 2. that the decorrelation time is sufficiently short to obtain an adequate number of uncorrelated samples in one measurement. With both equipments the depression angles between lo and 75o are covered. Results of measurements will be reported. They include measurements on coniferous trees , selected agricultural crops, grass and bare soils. The radar return parameter y as a function of wavelength and polarization is a useful classifier, Within the full dynamic range of y as met in nature its total variation for vegetation is a 20 dB. The radar back-scatter coefficient as a function of frequency and polarization seems to be the onìy possible classifier for vegetation species.