[Justification of repeated animal experiments and determination of the required number of animals according to the German Animal Protection Act].

The repetition of animal experiments requires justification according to the German Animal Protection Act (Tierschutzgesetz). On the other hand, repetition of experiments is considered necessary for the improvement of the precision and reliability of the results. In this paper it is shown that such improvement is not achieved by a repetition of the experiment but by repeated inclusion of animals into the experiment. The more animals are included, the more precisely and reliably the results can be generalized to further situations. The number of animals necessary for a proposed precision can be determined with statistical methods. The basic ideas are explained in the paper and demonstrated with two examples. In long term experiments the precision can be improved, if all experimental conditions are applied within temporally homogenous blocks to few animals and the blocks are repeated till the necessary number of animals is reached (block experiments). By sequential interim analyses, where at each analysis on the basis of all available data a decision on continuation or conclusion of the experiment is drawn, a further reduction of the number of animals can be achieved. These methods are discussed in the paper. Repetitions of total animal experiments are justified, if there exist legitimate doubts on the soundness of the published results or if in so called ring-experiments the variability of the results between different laboratories should be investigated.