The standard deviation and standard error; how to deal with outlier; normal value; the box-and-whisker plot; the null hypothesis; why is the significant level 5per cent?; which should be used - less than 5per cent or 5per cent or less for p value?; the t-test; which should be used - a two-sided (2a) or one-side (a) test? the type I error (error of the first kind); Dunnett's test; the multiple comparison (range) test; Scheffe's multiple comparison test can be used to compare two groups; the differences in power between the t-test and Dunnett's test; the multiple comparison (range) test; Scheffe's multiple comparison test has a very low power; a sifnificant difference is observed by Dunnett's multiple comparison test even when a significant difference is not observed by analysis of variance; the rate of detection of a significance different in a low-dose group by Dunnett's multiple comparison test. (Part contents).