The Hennessy Grading Probe (GP) and the SFK Fat-O-Meater (FOM) are instruments capable of automatically measuring fat and muscle thicknesses from the differential light reflectance of tissues. They may offer advantages in the precision and cost of measurement over the simpler optical probe (OP). The GP and FOM were used on the slaughterline at nine abattoirs to measure fat thickness over the m. longissimus dorsi (MLD) and MLD depth at the last rib (LR) and third to fourth from last rib (3/4 LR) positions. The probes were used alternatively on runs of 100 carcasses which were also measured for LR fat thickness by OP in the normal course of carcass classification. In all 4642 (GP) and 5760 (FOM) carcasses were measured. A further 130 carcasses sampled from six of the abattoirs were measured by all three probes and their left sides were dissected into component tissues. The performance of the automatic probes relative to the OP was found to differ between abattoirs. The FOM worked more effectively on the slaughterline, requiring fewer repeat readings to obtain a satisfactory measurement than the GP. Both automatic probes gave more precise results when used at the 3/4 LR position. In the overall sample, residual s.d. for the prediction of OP-measured fat thickness were 2·33 mm (GP) and 2·07 (FOM). Residual s.d.'s for the prediction of carcass lean content from 3/4 LR fat thickness were 30·8 g/kg (OP), 36·6 (GP) and 31·4 (FOM). Addition of MLD depth provided a significant improvement in precision; residual s.d. were 35·9 g/kg (GP) (P < 0·05) and 29 1 g/kg (FOM) (P < 0·001). These results rate the FOM as superior but the differences are not large enough to discount the GP.