We came to appreciate some years ago, that we were able to demonstrate in short experiments only faults of a pronounced nature. This is what is to be expected. If the tissues are subjected to only slight deviations from the optimum in their food supply, they will not deteriorate rapidly as they do when the faults in the diet are more pronounced. It was to be expected, however, that longer periods of observation would reveal defects of a lower order. This we have found to be the case. There was a time a few years ago when any investigator of a nutrition problem would have said that a diet which was adequate for the support of growth through a considerable portion of the period during which growth normally takes place, would certainly be adequate for the maintenance of health and vigor in the adult. The experiments with domestic animals conducted at the Wisconsin Experiment Station, clearly forecasted the necessity of including reproduction records as a means of refining the technic of nutrition studies so as to make possible more accurate estimations of quality in one or more factors in the diet. Miss Sim-monds and I have developed this idea and have applied this principle in our work since 1915. We realized the importance of attempting to answer the question: How sensitive is an animal to faulty diets of different types? Will not faults in the diet which are too slight to affect the animal promptly during the growing period to an extent which will be observable, become easily apparent if they exert their influence upon it at a time when it is meeting the unusual demand occasioned by pregnancy or the nursing of young? Will not the growth and well-being of the young and the capacity of a family to remain in a state of vigor during succeeding generations when confined to a certain type of diet, reveal shades of quality not to be discriminated in a short experiment? We have now tested these propositions sufficiently to enable us to give very definite answers to them. The results have amply demonstrated that any fault in the diet, however slight, which causes it to deviate from the optimum, will, if it exerts its influence over a sufficient period, surely be reflected somewhere in the life history of the individual or its descendants. The sensitiveness of the rat to faulty diets, which fall but little short of the optimum, is remarkably great, if such diets are persisted in for considerable periods. In certain critical conditions such as pregnancy, such effects become observable in a short time. The nursing period is almost equally critical in that small departures from the optimum composition of the diet leave their mark on the vitality of the young and frequently also on the vitality of the mother.