Expanding on the studies of Hernandez (2008) and de la Fuente (2009), this investigation examines the combined effect of explicit instruction (EI) and input flood (IF) vs. IF alone on students' use of discourse markers to narrate a past event. Participants included 91 English-speaking adults enrolled in fourth-semester college Spanish courses, assigned to the EI + IF group (n = 36), the IF group (n = 30), or the control group (n = 25). The EI + IF group was provided with EI on the function and use of discourse markers. This group also received a flood of input containing the target forms. Students were provided with communicative practice and feedback. In contrast, the IF group did not receive EI + IF on the use of discourse markers. This group received the same flood of input as the EI group. The IF group was presented with communicative practice although there was no feedback. Results of a speaking task administered as a pre-test, immediate post-test, and delayed post-test indicated that both treatments had a positive effect on students' use of discourse markers. Discourse transcripts show that the EI + IF group used more discourse markers to shape their output in student--student communicative exchanges than the IF group. Taken together, results suggest that the combined effect of EI and IF was not superior to IF alone in promoting student use of discourse markers as measured on the post-test speaking tasks. (Contains 2 figures, 5 tables, and 4 notes.)