This study aimed to evaluate the marginal microleakage of Class V cavities of All-bond 2 (Wet - bonding system), Gluma (Adhesion of resin to exposed collagen fibers), and Scotchbond Multi-purpose(Mild Etching System). Hundred extracted human teeth divided into a control and three experimental groups consisting of eight teeth. The experimental group was further subdivided into All- bond2, Gluma and Scotchbond MP groups, Vitrebond served as the control. The positive control group consisted of specimens filled with resin and with no etching, primer and bonding procesure. Polished specimens underwent temperature changed from $5^{circ}C$ to $55^{circ}C$ a thousand times. After thermocycling, speciemens were placed in 2 % methylene blue dye solution for 24 hours in an incubator set at $37^{circ}C$. The teeth were sectioned buccolingually and the degree of dye penetration was observed with a stereomicroscoped(*20). The following results were obtained. 1. Both the control and the experimental group showed a lower degree of dye penetration on enamel than on dentin margins(p<0.05) 2. Gluma exhibited a statistically significant lower degree of dye penetration than All-bond 2 on enamel margins(p<0.05) Scotchbond MP also exhibited a lower degree of dye penetration than All-bone 2 but was no statistically significant. Gluma and Scotchbond MP exhibited a similar degree of dye penetration. 3. The degree of dye penetration of All-bond 2, Gluma and Scotch bond MP showed no statistically singnificance on enamel margins but was significantly lower than in the control using Vitrebond. 4. All-bond 2 exhibited a statistically significant lower degree of dye penetration than Gluma on dentin margins. All- bond 2 and Scotchbond MP showed a similar degree of dye penetration. 5. The degree of dye penetration of All-bond 2, Gluma and Scotchbond MP showed no statistically significance on dentin margins. There was neither a statistical significance with the control.