The concentration of lead in the blood stream was measured for a sample of 50 children form a large school near a busy main road. The sample mean was 10.12 ng/mL, and the standard deviationwas 0.64 ng/mL. Approximately how large should the sample have been to reduce the range of the confidence interval to 0.2 ng/mL (i.e. ±0.1 ng/mL)?