1. We previously reported that parotid gland secretion is decreased in rats deprived of vitamin D (Glijer, Peterfy & Tenenhouse, 1985). In the present study we examine whether this effect is a direct result of the absence of vitamin D or due to the secondary systemic effects of vitamin D deficiency. 2. Offspring of rats maintained on a calcium-supplemented (1.2%), vitamin-D-deficient diet were weaned onto the same diet and examined after 8 weeks. Using this method it was possible to maintain serum calcium and parathyroid hormone concentrations within normal limits. Serum 25-hydroxyvitamin D (25(OH)D3) was not detectable, but 1,25-dihydroxyvitamin D (1,25(OH)2D3) concentrations were normal. 3. Pilocarpine-stimulated flow of parotid saliva was reduced 57% in vitamin-D-deprived animals, but amylase secretion was unchanged. Treatment with vitamin D3 returned flow rates to normal. 4. The concentration of calcium in parotid saliva was normal in vitamin-D-deprived rats, although total parotid calcium output was reduced 57%. 5. Pilocarpine-stimulated salivary flow from submandibular gland, a tissue which does not possess 1,25(OH)2D3 receptors, was normal in vitamin-D-deprived rats. 6. Heart rate and arterial blood pressure changes in response to I.V. pilocarpine administration were identical in normal and vitamin-D-deficient rats. 7. Auriculotemporal nerve-stimulated flow of parotid saliva was also reduced by 50% and administration of vitamin D3 to these rats corrected this abnormality. 8. It is concluded that fluid and electrolyte secretion from parotid gland is directly dependent on vitamin D; abnormal parotid gland function seen in vitamin-D-deficient rats is not due to secondary hypocalcaemia or hyperparathyroidism, nor can it be explained by haemodynamic changes evoked during systemic administration of pilocarpine. We further conclude that the metabolite of vitamin D responsible for this effect is not 1,25(OH)2D3.