This study evaluated in vitro the influence of five solutions used to treat cavity preparations on microleakage of amalgam restorations. Seventy-two standard Class V cavities were prepared in buccal and lingual surfaces of 36 recently extracted human third molars. The cervical wall was located in cementum and the occlusal wall was located in enamel. Specimens were randomly divided into six groups (n=12) according to the solution employed to treat the cavities: Group I-2.5% sodium hypochlorite; Group II-Calcium hydroxide solution; Group III-1.23% Acidulated phosphate fluoride; Group IV-2% digluconate of chlorhexidine; Group V-anionic detergent solution (1.25% sodium lauryl sulfate) and Group VI-control group. The solutions were applied for one minute, followed by washing and drying. Two coats of copal varnish were applied in each cavity. The cavities were filled with capsuled amalgam GS-80 (SDI). After finishing and polishing, the specimens were submitted for thermal cycling followed by immersion in methylene blue. Then, the specimens were sectioned and microleakage was evaluated based on a standard ranking under magnification (40x). Data were subjected to statistical analysis using non-parametric tests. Results of the study concluded that leakage was higher in cementum than enamel (p<0.01). The substances employed have no influence on microleakage except for sodium hypochlorite, which increased leakage values in enamel (p<0.05).