Adaptive cluster sampling is often of more practical interest than conventional sampling methods, especially in the fields of ecological, environmental and social sciences. When the target population is a rare, clustered or patched one, it can select more meaningful sample for the field investigators and provide more efficient estimates of the population quantity of interest than the conventional sampling methods. In many survey situations, one would like to utilize the auxiliary information collected together with the population variable of interest in the estimation stage. Ratio estimator is a commonly used estimation method utilizing the auxiliary information. Though it is design-biased, it is often more efficient than the usual unbiased estimator which solely makes use of the information of the variable of interest. The ratio estimator under adaptive cluster sampling is studied in this article. Empirical study shows that it produces better estimation results than the original estimator of adaptive cluster sampling, and the ratio estimator under a comparable conventional sampling design. The property of the associated 100(1 — α) confidence interval is investigated as well. In practice, often there are more than one auxiliary variable available, this article is meant to be an initial research of the utilization of the auxiliary information in adaptive cluster sampling.