SA4503, a potent sigma(1) receptor agonist, is reported as having 103-fold higher affinity for sigma(1) (IC(50) = 17.4 nM) than sigma(2) (IC(50) = 1,784 nM) sites in guinea pig brain membranes. Modest structural changes appear to have major effects on sigma(1)/sigma(2) selectivity. The fluoroethyl analog, FE-SA4503, is described as having high primary affinity for sigma(2) sites (IC(50) = 2.11 nM) and a weaker interaction with sigma(1) sites (IC(50) = 6.48 nM). SA4503 and FE-SA4503 have been radiolabeled for PET studies, and both bind selectively to sigma(1) receptors in animal and human brain in vivo. We prepared SA4503 and FE-SA4503 as reference compounds for radioligand development efforts. In our hands, SA4503 is 14-fold selective for sigma(1) (K(i) = 4.6 nM) over sigma(2) (K(i) = 63.1 nM) sites in guinea pig brain homogenates. Further, FE-SA4503 exhibits the same 14-fold selectivity for sigma(1) (K(i) = 8.0 nM) over sigma(2) (K(i) = 113.2 nM) receptors. The main differences from previously reported values stem from sigma(2) affinity determinations. This protocol, displacement of [(3)H]DTG binding to sigma(2) sites using (+)-pentazocine (200 nM) to mask sigma(1) sites, was validated by the proper rank order of sigma(2) inhibitory potencies shown by a panel of additional ligands: ifenprodil > haloperidol > DTG >> (+)-pentazocine. Robust Pearson correlation (r = 1.0, P = 0.002; slope = 0.97) was observed for our pK(i) values against those from a prior study by others. The findings have bearing on structure-activity relationships for this active series, and on conclusions that might be drawn from experiments relying upon defined sigma(1)/sigma(2) binding selectivity.