The aim of this investigation was to test the hypothesis that there is no difference in microleakage between composite-enamel and composite-wire interfaces, when different composites are used. Forty-five freshly extracted human mandibular incisors separated into three groups were used in the study. Multi-stranded 0.0215 inch diameter wire was bonded to enamel using two conventional (Transbond XT and Transbond LR) and a flowable (Venus Flow) orthodontic composite. The specimens were sealed with nail varnish, stained with 0.5 per cent basic fuchsine for 24 hours, sectioned and examined under a stereomicroscope, and scored for microleakage at the composite-enamel and composite-wire interfaces from the mesial and distal margins. Statistical analysis was performed using Kruskal-Wallis and Mann-Whitney U-tests with a Bonferroni correction. Little or no microleakage was observed between the composite-enamel interfaces for the three investigated composites, and any difference was not statistically significant. However, statistically significant differences were found between microleakage at the composite-wire interface for both the conventional and flowable composite groups (P < 0.001). Flowable composite showed the highest leakage (mean: 4.8 +/- 0.8 mm), while Transbond XT (mean: 0.5 +/- 0.3 mm) and Transbond LR (mean: 1.1 +/- 1.2 mm) showed significantly lower and comparable results. The amount of microleakage at the wire-composite interface was significantly greater than that at the enamel-composite interface of flexible spiral wire retainers (FSWRs). The null hypothesis is therefore rejected. Flowable composites may not be appropriate for bonding FSWRs.