Trimethaphan, a ganglionic blocking agent which is administered by intravenous drip to produce controlled hypotension during surgery, produces a complete neuromuscular blockade at the isolated phrenic nerve-hemidiaphragm preparation of the rat at a concentration of 0.3 mmol X l-1. This blockade is not reversed by neostigmine, a cholinesterase inhibitor, nor by calcium chloride, and this action is attributed to the local anaesthetic activity of the drug. Trimethaphan (1.5 X 10(-2) mmol X l-1) interacts with the following aminoglycoside antibiotics: gentamicin (0.04), streptomycin (0.05), netilmicin (0.06), amikacin (0.11), sisomicin (0.14), kanamycin (0.17), tobramycin (0.18) and dibekacin (0.21 mmol X l-1) to produce a complete neuromuscular blockade. These pharmacodynamic interactions of trimethaphan and aminoglycoside antibiotics occur at significantly reduced concentrations of the interacting drugs which are very close to the ones obtained after administration of therapeutic doses. When trimethaphan or aminoglycoside antibiotics are used alone at the above reduced concentrations they do not exert any neuromuscular blocking activity. The neuromuscular blockade which is obtained after the interaction of trimethaphan with aminoglycoside antibiotics is not reversed by either neostigmine or calcium chloride, although the neuromuscular blockade which is produced by aminoglycoside antibiotics alone is reversed by calcium chloride. It is concluded that the local anaesthetic effect of trimethaphan is the predominant factor of the mechanism of the above interactions. These interactions may produce severe respiratory disturbances (respiratory depression or apnoea) to the patients, during the perioperative period, which can be reversed only with artificial ventilation.