1. The types of K+ channel which determine the membrane potential of arcuate artery smooth muscle cells were investigated by patch-clamp recording from isolated cells and lumenal diameter measurements from intact pressurized renal arcuate arteries. 2. Single cells had a mean resting potential of -38 mV and were depolarized by 130 mM K+ but not by the Cl- channel blocker 4,4'-diisothiocyanatostilbene-2, 2'-disulphonic acid (DIDS). 3. Iberiotoxin did not affect the resting potential but inhibited spontaneous transient hyperpolarizations. Iberiotoxin or 1 mM tetraethylammonium (TEA+) constricted intact arteries. 3,4-Diaminopyridine (3,4-DAP)-sensitive delayed rectifier K+ (KV) channel current was elicited by depolarization but 3,4-DAP did not affect the resting potential or induce constriction in the intact artery. 4. A voltage-independent K+ current was inhibited by >= 0.1 mM barium (Ba2+) and unaffected by iberiotoxin, glibenclamide, apamin, 3,4-DAP and ouabain. In six out of ten cells, 1 mM Ba2+ depolarized the resting potential, while in the other cells the potential was resistant to all of the K+ channel blockers and ouabain. Ba2+ (0.1-1 mM) constricted the intact artery, but 10 microM Ba2+, 1 microM glibenclamide or 100 nM apamin had no effect. 5. The data suggest that resting potential is determined by background K+ channels, one type being Ba2+ sensitive and voltage independent, and another type being poorly defined due to its resistance to any inhibitor. Large conductance Ca2+-activated K+ (BKCa) and KV channels do not determine the resting potential but have separate functions to underlie transient Ca2+-induced hyperpolarizations and to protect against depolarization past about -30 mV.