Stepp prepared a bread made from milk and wheat flour, and another from milk and "protamol," a protein-rich preparation from rice. These he extracted thoroughly with alcohol and with ether. Mice were fed either the extracted or the unex-tracted breads. Those fed the extracted food died within thirty days, whereas those fed the untreated bread remained in good health during the six or eight weeks they were under observation. Stepp interpreted these results to mean that certain lipins (fatlike substances) are essential in the diet. He next sought to discover which of the individual lipins was indispensable for the maintenance of health.

The method employed was to restrict mice to the lipin-free diet (extracted) plus one or more of the individual lipins, the chemistry of which is more or less well understood, and which can be prepared in a state of purity. He tested out in this way lecithin, cholesterol, cephalin and cerebrosids, but in no instance was he successful in making the extracted diet complete by the addition of one or more of these purified substances. He also made an extract of egg yolk by shaking with cold alcohol, and divided the solution into two portions. One of these he heated with 95 per cent alcohol for two days in a water bath, while the other was heated long enough to evaporate the alcohol. Each of these preparations was added to a portion of the extracted bread and the two mixtures served as rations for two groups of mice. The group fed the heated preparation all died within thirty days, while the other group receiving the unheated preparation remained in good health. Stepp thus discovered that the substances which dissolved from his breads when they were treated with alcohol and ether were changed by heating during several days, so that they were no longer capable of rendering the extracted breads satisfactory for the nutrition of mice.