Seven normal asymptomatic Hong Kong-Chinese (HKChinese) were recruited and their noninvasive tear break-up time (NITBUT) was measured. Measurements were performed eight times over a period of 8 weeks to test the reliability of the NITBUT test over time. The distribution of NITBUT data was positively skewed. The mean (±SD) NITBUT was 16.0 (±9.4) s and the mode was 17.5 s. Friedman two-way analysis of variance revealed no significant difference in the NITBUT taken on different days. For the clinical estimate of the NITBUT, we recommend taking five measures of NITBUT and average the three closest measures and use this as a representative value of tear stability for the subject.