In order to investigate the metabolic poise of the elasmobranch rectal gland, we conducted two lines of experimentation. First, we examined the effects of feeding on plasma metabolites and enzyme activities from several metabolic pathways in several tissues of the dogfish shark, Squalus acanthias, after starvation and at 6, 20, 30 and 48 h post-feeding. We found a rapid and sustained ten-fold decrease in plasma beta-hydroxybutyrate at 6 h and beyond compared with starved dogfish, suggesting an upregulation in the use of this substrate, a decrease in production, or both. Plasma acetoacetate levels remain unchanged, whereas there was a slight and transient decrease in plasma glucose levels at 6 h. Several enzymes showed a large increase in activity post-feeding, including beta-hydroxybutyrate dehydrogenase in rectal gland and liver, and in rectal gland, isocitrate dehydrogenase, citrate synthase, lactate dehydrogenase, aspartate amino transferase, alanine amino transferase, glutamine synthetase and Na(+)/K(+) ATPase. Also notable in these enzyme measurements was the overall high level of activity in the rectal gland in general. For example, activity of the Krebs' TCA cycle enzyme citrate synthase (over 30 U g(-1)) was similar to activities in muscle from other species of highly active fish. Surprisingly, lactate dehydrogenase activity in the gland was also high (over 150 U g(-1)), suggesting either an ability to produce lactate anaerobically or use lactate as an aerobic fuel. Given these interesting observations, in the second aspect of the study we examined the ability of several metabolic substrates (alone and in combination) to support chloride secretion by the rectal gland. Among the substrates tested at physiological concentrations (glucose, beta-hydroxybutyrate, lactate, alanine, acetoacetate, and glutamate), only glucose could consistently maintain a viable preparation. Whereas beta-hydroxybutyrate could enhance gland activity when presented in combination with glucose, surprisingly it could not sustain chloride secretion when used as a lone substrate. Our results are discussed in the context of the in vivo role of the gland and mechanisms of possible upregulation of enzyme activities.