Comparison of a variance with a given value, estimation of a variance, comparison of two variances, estimation of the ratio of two variances, and the same procedures for a mean with known or unknown variance are dealt with. Techniques are valid for sample elements drawn at random and being independent, when the distribution of the observed variable is normal, for a sample size not too small (5 to 10 at least) approximately also when the distribution does not deviate very much from the normal. Techniques of verification of the hypothesis of normality are briefly dealt with in the examples.