Previous studies have shown that shear stress, and molecules such as VEGF differentially increase the permeability to large and small molecules. However, it has not been possible to measure the solute permeability to both large and intermediate sized molecules within a single vessel at the same time. We have therefore designed and built a system to measure solute permeability to two different fluorescently labelled solutes simultaneously in capillaries and postcapillary venules of living animals. In this system, frogs (Rana temporaria, anaesthetised by immersion in MS222 (0.25g/l) solution) or rats (anaesthetised by 1:1:2 Hypnovel/Hypnorm/water; i.m. 0.1ml/100 g body wt) were laid supine and the limbs lightly secured to a supporting tray. An incision was made in the abdominal cavity and the exteriorised mesentery was gently stretched over a glass coverslip and pinned in place. The tray was placed on an extended microscope stage with the coverslip positioned over a 20x Fluotar objective on an inverted epifluorescence microscope (Leica DMIRB) and the mesentery was continuously superfused with frog Ringer solution, enabling visualisation of the mesenteric microvessels. A capillary or post-capillary venule that was straight, free-flowing, had at least 300μm between side-branches, 15-35μm in diameter and free of white cells adhering to, or rolling along the vessel wall was cannulated using a bevelled glass theta micropipette with a septum dividing it in two. A hole had been drilled in each side of the pipette and a cannula glued in place to allow refilling of either side of the micropipette independently. Blunt metal needles were glued in place at the back of the pipette allowing two pressure lines to be connected and the pressure on each side of the micropipette to be controlled independently. The vessel was perfused initially with 1% bovine serum albumin (BSA) in frog Ringer solution at pH 7.4 at a pressure of approximately 30cmH2O in one side of the pipette, and 1% BSA with 1mg/ml TRITC BSA and 100μg/ml sodium fluorescein on the other side. The pressure in the fluorescent side was then adjusted to balance the driving pressure of the BSA solution. The vessel was illuminated by a xenon light source (Cairn High Intensity Arc lamp, Cairn Research Ltd, UK). A rotating (50Hz) disc containing 480 and 530nm excitation filters was used to control the excitation wavelength. Fluorescence intensity (I_f) was measured within a defined window around the perfused vessel at 480 and 530nm wavelengths (F480 and F530). Fluorescence was measured during perfusion of 1% BSA, and when the perfusate was switched to the 1% BSA containing TRITC BSA and sodium fluorescein in the other side of the micropipette (pressures in either side of the micropipette were swapped). An infrared camera was used to set the window for the photomultiplier tube (PMT), incident white light was passed through an infra red filter (750nm longpass) to visualise the preparation with an IR camera (Watec WAT-902B). A dichroic filter was placed in front of the photometer to reflect light of 700nm passed through the dichroic filter to the IR camera. The PMT was controlled by a Cairn spectrophotometer, which was in turn connected to a PowerLab/4SP (AdInstruments) system. Thus, I_f was measured using Chart software, P_a was calculated from the initial increase in fluorescence following switching (△I_f), the radius of the vessel, and the rate of increase of fluorescence (dI_f/dt) for each fluorescent solute, P_a = (dI_f/dt)(1/△I_f)(r/2).