Microphone windscreens are used to attenuate wind noise. They add impedance between the source and microphone, not accounted for when the system is checked by acoustical calibrator or when used in the field. The procedure for the characterization of the attenuation has recently been addressed in ANSI S1.17/2000 part 1. But to date, no windscreens have been tested in accordance with that standard. Partially, the reason is that the precision of the procedure has not been determined. As of January 2002 a round robin is addressing just that issue. By June 2002 the results of the uncertainty determination will be available. In the meantime, we have tested some of the samples used and others in a small chamber approaching free-field conditions. We present the results of the tests of insertion loss (not in accordance with S1.17 because we do not have a reverberation room). We will show that, in some cases, the use of a windscreen can easily change a measurement using type 1 instruments to type 2 or worse. Without knowing information about a particular windscreen, the use of a windscreen in still air can drastically change uncertainty of measurement. In moving air conditions we expect even more severe problems.