The use of external representations has a potential to facilitate inquiry learning, especially in hypothesis generation and scientific reasoning, which are typical difficulties encountered by students. This study proposes and investigates the effects of a three-dimensional thinking graph (3DTG) that allows learners to combine in a single image, problem information, subject knowledge (key concepts and their relationships), and the hypothesizing and reasoning process involved in exploring a problem, to support inquiry learning. Two classes of eleventh grade students (97 in total) were randomly assigned to use either the 3DTG (i.e., experimental condition) or concept map (i.e., control condition) to complete a group-based inquiry task in an online environment. Data were collected from multiple data sources, including measures of group inquiry task performance, postknowledge scores, postquestionnaires of student perceptions of consensus building, and open-ended survey. The analysis of group task performance revealed that participants in the experimental condition performed better in the inquiry task than their counterparts specifically in generating hypotheses and reasoning with data, but not in drawing conclusions. The findings show the 3DTG's benefits in facilitating exploring and justifying hypotheses, and also suggest that the 3DTG can share equal value with concept mapping in terms of drawing conclusions and consensus building. In addition, students using the 3DTG achieved higher scores in the postknowledge test, although no difference was found in their perceptions of consensus building. These findings were validated by students' responses to the survey. Implications of this study and future work are also discussed.