A non-destructive technique is presented for verifying torsional spring constants used in lateral force microscopy. Various calibrations of the microscope are required and these are detailed. The technique produces reasonable values which tend to be larger than those predicted >from considerations of the cantilever dimensions. The differences are discussed in terms of length corrections and particularly the uncertainty in the thickness of the cantilevers, which has an enormous effect on the values obtained through a priori calculations. Methods for inferring the thickness are discussed. Further, artefacts in conventional force measurements related to the experiments performed here are discussed.