Teaching students to use and interpret representations in science is critically important if they are to become scientifically literate and learn how to communicate their understandings and learning in science. This study involved 248 students (119 boys and 129 girls) from 26 grade 6 teachers' classes in nine primary schools in Brisbane, Australia. Teachers were randomly allocated by school to one of three conditions: the contemporary science + representations condition (Experimental[superscript a]), the contemporary condition (Experimental[superscript b]), or the comparison condition as they participated in an eight-week inquiry-science unit on Natural Disasters. Data on students' discourse were collected at two time points during the implementation of the unit and data on the concept maps were collected pre- and post-intervention while data on the reasoning and problem-solving (RP-S) task were collected following the intervention. The results show that when students participate in an inquiry-based science unit that is augmented with a variety of multimedia resources presenting a range of current contemporary events (Experimental[superscript a] and Experimental[superscript b] conditions), they demonstrate significantly more social language and basic scientific language and marked increases in moderate scientific language than their peers in the comparison condition. Interestingly, although there were no significant differences on the Personal Concept Map scores between the conditions at Times 1 and 2, the students' scores in all conditions improved decidedly across time. It appears that as the children had more time to engage with the material, participate in cooperative peer discussions, and receive encouragement from their teachers to provide elaborated feedback to each other, their conceptual understandings of earthquakes were enhanced. However, although the children in the experimental conditions demonstrated significantly more social and scientific language than their peers in the comparison condition, these oral language skills did not transfer to the RP-S task, possibly because they may not have had enough time to consolidate their application in a novel context where they had to work independently.