In　this　paper,　an　adaptive　output　feedback　tracking　for　a　class　of　uncertain　nonlinear　systems　with　input　and　output　quantization　is　investigated.　A　new　controller　structure　is　introduced　in　which　a　modified　output　quantizer　is　proposed　so　that　its　first-order　derivative　exists,　and　an　observer　is　designed　with　quantized　input　and　output　as　its　input　signals.　Then,　with　dynamic　surface　control　as　a　tool,　an　adaptive　recursive　design　is　developed　which　can　avoid　higher-order　derivatives　of　the　quantized　output　signal.　It　is　proved　that,　with　the　proposed　scheme,　coarser　input　and　output　quantization　can　be　achieved　and　the　tracking　error　can　converge　to　a　small　residual　set　by　appropriately　choosing　some　design　parameters.