The purpose of this quantitative correlational study was to assess the relationship between teachers' emotional intelligence (EI) and their students' achievement in a private high school located in Florida. Participants were 46 teachers and their students in grades 9--12. The secondary data analyzed in this study included teacher EI overall and end-of-the-year class average scores for English, mathematics, history, science, art, world languages, and music. Teachers' EI scores ranged between 284 in art and 302 in world languages, with high scores of 299 in English and science and lower scores in art (287) and history (284). Given the very small subsamples by subject (ranging between 9 English teachers and 3 Music teachers), most results of the nonparametric Kendall's tau-b correlations by subject between teacher EI and the corresponding students' performance did not reach statistical significance. The only exception was in science (with a subsample of 8 teachers), where the results showed significant and consistently inverse correlations between students' performance and teachers' total emotional intelligence (taub = --.951, p = 0.03), emotional expression (taub = --.919, p = 0.006), emotional awareness of others (taub = --.821, p = 0.013), and emotional self-control (taub = --.651, p = 0.046). The results suggest that, at the school where the study was conducted, science teachers' EI may be counterproductive in terms of their students' academic achievement in that subject. Other nonsignificant correlations suggested direct relationships in English and inverse relationships in math. These findings differ from the consistently strong and direct relationships between leader/manager EI and subordinates' performance in business.