Hazard analyses of food preparation practices were conducted in three households in a new settlement in the rocky, dusty hillsides at the outskirts of Lima, Peru. These analyses consisted of watching all steps of preparation, recording temperatures throughout these steps, and collecting samples of the food and testing them for common foodborne pathogens and indicator organisms. The residents had migrated from different regions of the country; consequently, they prepared different foods. These included soya cereal, milk formula, rice and carrots for feeding a baby who had diarrhea, soups, mashed potatoes with spinach, carrot and beet salad, cow's foot soup, beans, rice and a mixture of beans and rice. The temperatures attained were high enough to kill vegetative forms of foodborne pathogens, but not their spores. During the interval between cooking in the morning and serving at either lunch or supper time, foods were held either on unheated ranges or in unheated ovens. This interval was long enough to permit some bacterial multiplication, but apparently not to massive quantities. Just before the evening meal, foods were reheated to temperatures that usually exceeded 70°C. Rice, however, was either served cold or if reheated, the center temperature rose a few degrees only. Critical control points for preparation of family meals are cooking, holding between cooking and serving, and reheating. Critical control points for milk formula for babies are using recently-boiled water for the formula, cleaning and boiling bottles and nipples, and, of particular importance, time of holding at room temperature.