An epidemiological study has been carried out an the scholastic population of L'Aquila (Abruzzo). This study had the aim of verifying possible correlations between feeding procedures and the outbreak of illnesses whose causes are unknown. The information was registered with the collaboration of the Medical Scholastic Service, in the period of May-June in 1988-1989, interviewing four hundred and seventy-seven mothers of students, the information regards the temporary period of 1973-1985. The results of their elaboration are as follows: 75.9% of babies took breast milk at birth and in following months breast feeding was the most frequent (34.8%). The procedure breast feeding-bottle feeding was carried out in 30.4% of cases and exclusively bottle feeding in 22.4%. The study has permitted to discover thirteen cases of illnesses whose cause are unknown among which a case of SIDS (sudden infant death syndrome) in a newborn baby fed till eighteen weeks with breast milk and through twenty days before dying with undiluted cows milk. These illnesses happened in 13.4% of babies fed by bottle milk and diluted with running water; in 7.7% by cow's milk diluted with running water and in 4.3% by undiluted cow milk. Some Authors have suspected a relationship between bacterial endotoxin introduced by diet and baby illnesses whose causes are unknown. For this a quantitative determination of such substances is necessary in baby diets.