The purpose of this study was to evaluate the ability of a glass-ionomer cement-base material to prevent bacterial penetration along the dentin interface and to compare it with two conventional cement-base materials. A total of 107 Class 5 restorations was placed in Rhesus monkey teeth by means of three test materials [zinc oxide-eugenol (ZOE), copalite varnish + zinc phosphate cement base (V + ZP), and a glass-ionomer lining cement (GI)], with unetched and unbonded resin composite used alone as a control material and as a final restoration over the test base materials. Following disinfection, Class 5 cavities were prepared on the buccal surfaces of the teeth to the inner one-half of dentin. A sterile filter-paper disk was then placed on the axial wall and covered with a Teflon disk. Next, the cavities were based to the dento-enamel junction with one of the test base materials and finally restored with unetched and unbonded resin composite. After five and 16 weeks, the filter-paper disks were retrieved and cultivated for the presence and type of bacteria. The five-week results showed positive growth in two groups: the composite-only controls and the V + ZP group. The 16-week results showed growth in all of the test groups, but only one of nine teeth showed growth in the zinc oxide-eugenol group and one of 16 teeth in the glass-ionomer group. The results of this study indicate that under the conditions tested, a glass-ionomer base was capable of minimizing bacterial penetration along the material-tooth interface.