Microinjection of muscarinic receptor subtype antagonists into the region of the cat retrotrapezoid nucleus (RTN) decreases blood pressure (greatest efficacy; M2 subtype) and both baseline phrenic activity and CO2 sensitivity (greatest efficacy; M3/M1 subtype). Here we examine, in cat medullary sections at the level of the RTN, the effects of the same antagonists on binding of the high affinity muscarinic agonist quinuclidinyl benzilate (QNB). 3H-QNB binding was saturated and highly specific at 1 nM concentration and stable over 30 to 120 min (Kd, 0.49 nM; Bmax, 136 fm/mg protein). Studied biochemically, we found IC50 values for whole sections of 4.9 x 10(-6) M (M1 antagonist pirenzepine); 1.0 x 10(-6) M (M2 antagonist AFDX); and 0.64 x 10(-7) M (M3 antagonist DAMP; P < 0.03 vs PZ). Densitometric analysis of whole medullary cross section autoradiograms resulted in similar IC50 values as in the biochemical approach. Specific analysis of the RTN region demonstrated the presence of 3H-QNB binding and similar competition by the antagonists. Average IC50 values determined by densitometry were 14 x 10(-6) M (pirenzepine); 1.3 x 10(-6) M (AFDX; P < 0.01 vs PZ); and 0.53 x 10(-7) M (DAMP; P < 0.01 vs PZ). All three subtypes of muscarinic receptors identifiable via pharmacological antagonists appear to be present in the RTN region but we could not distinguish a subtype-specific pattern of receptor distribution.