The effect of a polyalkenoic acid conditioner pretreatment on the bond strength of a glass-ionomer adhesive to tooth substrates with or without smear layer was evaluated. Smear-layer covered and smear-layer free dentin and enamel surfaces were prepared from 24 extracted human molars. Resin composite was bonded to the surfaces using FujiBond LC with or without a polyalkenoic acid conditioner and subjected to microtensile bond strength (microTBS) testing. Failure modes were determined using scanning electron microscopy. For dentin, smear-layer coverage and conditioner treatment did not reveal significant differences in microTBS, which ranged from 20 to 29 MPa. For enamel, smear-layer coverage did not significantly affect microTBS, whereas, the use of conditioner significantly improved microTBS, reaching the same microTBS-values as when bonded to dentin. Regarding failure mode, most dentin specimens failed mixed adhesive cohesively. For enamel, adhesive failures mostly occurred when no conditioner was used, though mixed failures were predominant when the specimens were conditioned beforehand. Bonding of the glass ionomer adhesive to dentin can be achieved without the separate use of a polyalkenoic acid conditioner, even with the interposition of a smear layer. However, instrumented and non-instrumented enamel requires separate conditioning to provide sufficient micro-mechanical retention.