The present study was conducted to investigate the relationship between proactive and reactive Focus on Form (FonF) and gestures. Two English classes in a public high school in Japan taught by a native speaker of English were observed and videotaped for 96 min and analyzed. Overall, 64 proactive FonF and 43 reactive FonF were provided. Among them, 36% of the proactive FonF and 14% of the reactive FonF were accompanied by gestures. The target feature of proactive FonF was exclusively lexicon whereas those for reactive FonF varied. For reactive FonF, gestures were used the most for grammatical errors and the least for pronunciation errors. In addition, explicit correction was the most likely to be provided with gestures whereas elicitation was the least. Overall, deictic was the most common gesture for both proactive and reactive FonF. Furthermore, it was found that a deictic gesture was often essential in the sense that it was used to supplement the meaning of the demonstrative pronoun, "this," in proactive FonF. In contrast, when reactive FonF or proactive FonF without the demonstrative pronoun was provided, gestures tended not to carry any crucial meaning related to FonF by which they were accompanied, rendering the gestures subsidiary. (Contains 7 tables.)