The measurement principle of forward scatter visibilimeter is presented in this paper according to the theory of estimating atmospheric visibility with scatter energy.The curve fitting is worked out between visibilities man-observed and real field signals based on from the design field experimental data of a visibilimeter’s prototype.It shows that the basic principle of the prototype is correct although it has the limitations such as bigger errors and the smaller range of dynamic state.The main factor leading to the limitationis is random background noises especially in clear day.They severely affect the measurement results.The problems in the prototype are discussed and improving measures are put forward.