Fluid secretion by ant Malpighian tubules is driven by active K+-transport, Chloride follows passively, Using voltage- and ion-sensitive (K+,Cl-) microelectrodes, electrochemical driving forces were estimated in control conditions (51 K+, 143 Cl- in mmol/l and in slow (low bath-K+ or Cl-) or fast (cAMP present) fluid secretion conditions, Low bath-Cl- (14 mmol/l sulphate-substitution) reduced cell-Cl- to 49% and luminal Cl- to 19% of the control values, Membrane potentials and K+ concentrations were not affected, if bath K+-activity was kept constant, Passive Cl--driving forces turned cell-outward basally but were still lumen-directed across the luminal and transepithelial barriers, In Cl--free solution secretion dropped to 30% of control, When compared to control conditions (''100%'') low bath-K+ (5 mmol/l; Na+-substitution) caused a drop in the cellular and luminal concentration of K+ (to 50 and 72%, respectively) and of Cl- (to 57 and 91%, respectively), Basal membrane potential(V-bl) hyperpolarized and the Cl--driving force turned cell-outward but was still lumen-directed luminally and transepithelially. Fluid secretion dropped to 70%, (3) In control solution a membrane-permeant cAMP analogue depolarized the luminal and transepithelial potential and collapsed the Cl--driving forces, although secretion was increased, V-bl and the basal inward driving force were not affected, The anion-channel blocker DIDS (1 mmol/l) reduced fluid secretion rate by 40%, both in the absence and presence of cAMP, The results may be understood by assuming parallel and ''shunt'' pathways for Cl-: In high-K+/Cl-, the cellular pathway may be conductive, poorly selective, regulated by cAMP and DIDS-sensitive. In low-k(+)/Cl-, a basal cation/Cl- symport may mediate basal Cl- uptake.