Fourth semester students of Spanish as a foreign language at the university level in the US participated in two kinds of comprehensible-input based instruction, an extensive reading class that combined assigned and self-selected reading, and a "Reading-Discussion" class that consisted of assigned reading, debates and discussions. Students in both classes outperformed those in a traditionally taught class on a check-list vocabulary test and on a grammar test. The Reading-Discussion group outperformed the Traditional students on a cloze test (P=0.105), but there was no difference between Traditional and Reading students on the cloze. The results confirm the efficacy of comprehensible-input based pedagogy at the intermediate level. (Contains 4 tables.)