In dog submandibular gland, the activity of myoepithelial cells was assessed by simultaneous measurement of intraductal pressure (P(du)) and subcapsular pressure (P(ca)) using catheter-tip pressure transducers; their resting values were 2.5 +/- 0.21 and 3.0 +/- 0.19 mmHg, respectively (n = 40). Retrograde infusion of saliva (collected from preceding parasympathetic nerve stimulation) increased P(du) (coefficient of 50 mmHg ml(-1) for rates < 1 ml min(-1) and 85 mmHg ml(-1) for higher rates) and P(ca) (coefficient of 0.47 mmHg ml(-1) for all rates). Blood flow changes did not affect P(du) but increased P(ca) (coefficient of 0.04 mmHg ml(-1)). Parasympathetic nerve stimulation increased P(du) but decreased P(ca) abruptly; the response threshold was 0.1 Hz, with maximal responses at 16 Hz. The coefficients for P(du) and P(ca) on salivary secretion to parasympathetic nerve stimulation in glands with spontaneous blood flow (5.3 x 10(-3) and 4.87 x 10(-2) ml min(-1) g(-1) mmHg(-1)) were close to their values in glands with constant-flow vascular perfusion (4.9 x 10(-3) and 3.68 x 10(-2) ml min(-1) g(-1) mmHg(-1)). The finding that P(ca) fell despite concomitant increased blood flow suggests contraction of myoepithelial cells. Additional ductal occlusion further increased P(du) and enhanced the fall in P(ca), suggesting that the myoepithelial cells can contract when distended. Atropine blocked salivary secretion and responses of P(du) and P(ca) to parasympathetic nerve stimulation. ACh elicited responses similar to that of parasympathetic nerve stimulation. VIP caused very scanty salivary secretion and gradual slight increases in P(du) and P(ca); the change in P(ca) was abolished in glands with constant-flow vascular perfusion. Hence, contraction of myoepithelial cells to parasympathetic nerve stimulation is via muscarinic receptors. Sympathetic nerve stimulation increased P(du) and decreased P(ca) abruptly; the response threshold was 0.1 Hz, with maximal responses at 16 Hz. The coefficients for P(du) and P(ca) on salivary secretion to sympathetic nerve stimulation in glands with spontaneous blood flow (3.0 x 10(-3) and 3.2 x 10(-3) ml min(-1) g(-1) mmHg(-1)) were similar to their values in glands with constant-flow vascular perfusion (3.2 x 10(-3) and 3.1 x 10(-3) ml min(-1) g(-1) mmHg(-1)). The finding that P(ca) fell even in glands with constant-flow vascular perfusion suggests contraction of myoepithelial cells. Superimposed sympathetic nerve stimulation immediately enhanced the pressure changes and secretory response to parasympathetic nerve stimulation, indicating that the two autonomic nerves act synergistically to evoke myoepithelial cell contraction. Phentolamine and prazosin but not propranolol and yohimbine blocked the sympathetic enhancement. The finding that phenylephrine, but not clonidine and isoproterenol, abruptly decreased P(ca) in glands with constant-flow vascular perfusion suggests that the sympathetic activation of myoepithelial cells is via the alpha(1)-adrenoceptors.