Abstract
29 laboratory animal diets from 5 manufacturers for rats, mice, rabbits, guineapigs, primates and dogs were analysed for calcium, total phosphorus, phytate phosphorus, magnesium, sodium, potassium, iron, zinc and copper. Variations between diets for the same species were found to be much larger than differences between means for different species. The analyses were compared with data in the manufacturers' catalogues. The best controlled element was potassium. Most deviations of other elements were distributed in a skewed fashion and quite serious differences between analytical and catalogue data were found. In general, deviations from the catalogue were less than variations between diets. It is recommended that catalogues should give ranges of composition rather than means.