Some workers have reported that their rats do not eat a ration of enriched white flour bread as well as they do a bread made from this flour supplemented with milk solids. It is possible that our basal ration and the rations containing the white flour, enriched white flour, and whole-wheat flour were "unpalatable," and that the addition of the dried skim milk made these rations more "palatable." Unquestionably this was, in a sense, a fact; why then were these rations without the skim-milk powder in them unpalatable to our rats? Was this unpalatability due to some quality of these rations irrespective of the nutrients in them; or was it a response, which is familiar, of animals depleted in respect to some particular nutrient, to a ration deficient in that nutrient? If the latter is the explanation, might it not be possible to change the "palatability" of these rations to our rats, not by altering the rations themselves, but by correcting the nutritional deficiency in the test animals? That we tried to do.

In related experiments we fed 333 pairs of litter-mate male rats and 112 pairs of litter-mate female rats exactly the same basal ration as the one used with the rats previously mentioned. In these experiments one rat in each pair received in addition a few milligrams daily of a liver extract. Here, as before, the rats on the unsupplemented basal ration grew at a rate that was not much more than half of normal. By "normal growth" is meant the growth of our rats that received the basal ration plus an amount of the liver extract adequate for optimum growth. But the feeding of the liver extract, which was given separately from the basal ration, improved the palatability of this ration, increased consumption, and nearly doubled the growth. In fact, it was not even necessary to feed the liver extract to obtain the results; we obtained exactly the same results when the liver extract was injected under the skin or into the muscles of our rats. And the same results were also obtained when, instead of a few milligrams of liver extract, a few micrograms a few millionths of a gram of concentrated preparations of the growth-promoting material in these liver extracts were fed daily, separate from the basal ration.

The results show also that the administration of the liver extract separate from our basal ration brought about the same growth as the feeding of dried skim milk incorporated in this ration. It is obvious that the unpalatability of our basal ration can be just as effectively overcome in either of these ways, and that it is due to a nutrient deficiency in the basal ration and in our test animals.

But what nutrient was involved, and why were our test animals deficient?

We stated that the basal ration used in the experiments contained adequate amounts of all known nutrients for normal growth. The statement covers quite a bit of territory. We have found that our basal diet and the wheat-flour diets were deficient in some nutrient that can be supplied in a liver extract or in skim-milk powder. Before concluding that this nutrient actually is some still unidentified factor, let us consider what we mean by "all known- nutrients," and how we have arrived at the conclusion that our basal rations is "adequate for normal growth" in respect to these nutrients.

There is nothing unique about the composition of what we have called "our basal ration." It is a ration often used in the biological assay of vitamin A, except that we added vitamin A to it. The minerals in the ration were supplied by a common mineral mixture that unquestionably is adequate; the B vitamins were supplied by 10 percent of yeast. No increase in the quantity or change in the kind of yeast improved the growth of our rats on this ration, and, as a result of many efforts to supplement the ration with various amounts of crystalline or pure vitamins, or to replace the yeast by mixtures of these vitamins, we concluded that our basal ration also supplied adequate amounts of all of the available fat-soluble and water-soluble vitamins-vitamins A, D, E, K, and C ( ascorbic acid) ; choline, thiamine, riboflavin, pyridoxin, pantothenic acid, nicotinic acid, para-aminobenzoic acid, inositol, biotin, folic acid (or pteroylglutamic acid), also xanthopterin.

The protein in our basal ration (25 percent) was comprised of about 5 Percent of yeast protein and 20 percent of casein, the principal protein in Milk. Both proteins are ordinarily considered complete and the total amount adequate, but the casein we used had been heated for long Periods with successive lots of alcohol as had been our custom in removing vitamin A from it. This method of removing the vitamin A was one that was formerly used in vitamin A assay work. Could our handling of this casein have injured its nutritive properties?

This question led to a study of the effect of using, in our basal ration, casein preparations made in various ways. One preparation that was being used" in vitamin-assay work in another laboratory, and that had been quite extensively washed with acidulated water and then extracted with alcohol for 24 hours, gave results similar to our casein; one commercial "vitamin free" casein gave hardly significantly better growth than our casein; another commercial "vitamin free" casein, and other caseins prepared by milder treatments with alcohol, or with alcohol and ether, all gave decidedly better growth than our casein.

None of these caseins gave normal growth. But samples of "commercial" caseins that were tested gave almost normal growth, and a casein precipitated from skim milk by an 11 day dialysis against water and another centrifuged out of milk at 50,000 revolutions a minute and repeatedly washed and re-centrifuged, gave fully normal growth.

These results certainly mean either one of two things : That many researches are now being conducted with caseins in which the nutritive value has been impaired by various methods of treatment in which case our casein would be one that had been the most impaired of any of those tested, or that many researches are being conducted today with caseins containing as an impurity more or less of some unidentified nutrient in which case our casein would be one of those containing the least amount of this nutrient.