and not in others. One of these is vaccenic acid which it has recently been claimed is a component of butter but not of vegetable oils. When the first report of a mysterious fatty acid in butter was made by Schantz and the group at the University of Wisconsin 6 or 7 years ago, we made a thorough study on rats to determine whether we could obtain variations in growth when a diet which was optimal in all respects contained butterfat or various vegetable oils. In this series of tests, rats were fed on such complete diets where corn, cottonseed, peanut, olive or soybean oils or a margarine fat or butterfat was present in the diet. The growth which we observed during the active period of growth on rats was the same in the butter group as on the vegetable oil series which included the margarine fat. That this was true growth was shown by the fact that the bone growth as determined by X-ray pictures was similar on all diets. Another proof that the activity of the fats was identical was that the composition of the body was quite similar after different diets were fed. We reached the conclusion that butterfat does not possess any fatty acids necessary for growth of the rat which also are not present in the vegetable fats studied.

We have followed a number of different leads in order to see whether differences could be demonstrated between vegetable oils and butterfat. In addition to the growth tests cited earlier, another series of experiments involved the determination of the rate of growth of rats which were weaned prematurely. It had been claimed by the Wisconsin group that requirements for the butter-fatty acids were especially acute in the prematurely weaned animals. No such results were demonstrated with our own animals which were weaned at 14 days instead of the usual 21-day period. In fact, Zialciti and Mitchell at the University of Illinois were able to demonstrate that an identical rate of growth occurred in rats weaned at 7 days of age, irrespective of whether they were fed on a diet containing butterfat or a diet containing corn oil. These are some of the most ingenious experiments which have been carried out and it takes a very beautiful technique to be able to maintain life processes in a 7-day-old rat by artifical feeding.

Another index of nutritive value of fat is its ability to allow a normal pregnancy. The requirements of food for pregnancy are even greater than those for growth. In studies in our laboratory which were made on a series of vegetable fats, margarine fat, and butterfat no differences were obtained in the number of litters which could be produced or the size of the litters or of the well-being of the newly born animals irrespective of whether the animals had continuously fed from weaning on diets containing vegetable fats or butter.

Nutritionists feel that an even more critical method of evaluation of the efficacy of a diet is by its ability to sustain a normal lactation or a normal milk production. Although, in the case of rats, it is impossble to do quantitative measurements on this function, one can obtain some information about it from the well-being of the baby rats as determined by their ability to survive and their weight at the time of weaning. In a series of tests designed to investigate this point, it was found that the results on vegetable fats and margarine were certainly no less satisfactory than on butter. One can definitely state that margarine fat, corn, cottonseed, peanut, and soybean oils are as satisfactory as butter in such diets in allowing a successful lactation.

Another point to consider is that while cow's butterfat is not necessarily essential for children, it is a fat par excellence for the calf. This is because the composition of such butterfat is entirely different from the fat obtained from human milk or from the milk of other species. In fact, the differences are so marked between cow's milk and human milk that Hilditch and Meara concluded the following in the British Biochemical Journal-(I might state that Hilditch is the dean of all people working in fats-"Human milk fat in regard to its component acids has more resemblance to a typical margarine fat blend than to butterfat." One may conclude therefore that on the basis of composition that whereas cow's milk fat may be the ideal fat for growing calves, there is no reason to suppose that a similar superior nutritive value might hold for the human baby or for the young of other species of animals. However, I do not wish to leave the impression that I do not feel that cow's milk is a fine food for the child as well as for the adult. This superiority of milk, however, is rather to be traced to its excellent proteins and inorganic salt mixture-calcium and phosphorus-rather than being due to any especial superiority of its fat make-up.

Finally, in determining the nutritive value of any substance, we still have a final more critical measure for estimating such nutritive effects than any previously mentioned. So far as we have found that growth offers an important criterion, the requirements for pregnancy are somewhat more critical and that lactation puts the greatest strain on the animal as far as diets of high quality are concerned. It is conceivable that a diet might be satisfactory for all these conditions and might still fail after a period of generations. We have therefore carried out an experiment to test this point. A series of tests on a single strain of rats was started in April 1940. These animals have been fed on a diet which contains the whole wheat flour, skimmed milk powder, and margarine fat in place of the butterfat normally present in whole dried milk powder. Such animals have been given this diet continuously and their descendants continued on it for a period of over 20 generations. I may state that at the twenty-first generation the animals have maintained their vigor. Their growth rate is similar to that of the original group. Moreover, no failures have occured in pregnancy or lactation and it would appear that the animals could continue on such a diet indefinitely.

We have carried out such tests on rats because of the limitations that such an experiment would have when performed on human subjects. It would, of course, be impossible to so limit the diet of a man and woman over a considerable period and certainly not over a generation. Figuring that the generation in man is about 30 to 33 years, such experiments on rats if translated into human tests would represent a maintenance of normal function over a period exceeding 600 years where the fat in the diet was almost completely a vegetable-margarine fat.

There are still other ways also in which we have attempted to compare butterfat with various vegetable fats. In one case, we have attempted to compare the ability of fats during periods of undernutrition, where the animals did not get their full coloric requirements. Here again no differences obtained. Another series of tests involved a determination whether the vegetable oils were equally effective with butter in sustaining exaggerated growth which could be produced by

the use of the growth hormone. You know that in the pituitary gland, which is a very small gland located in almost the center of the skull, a series of hormones are produced which regulate many vital body processes. One of these hormones which has been sepa-. rated and investigated is the so-called growth hormone. When this is injected in man or in animals, it will cause a very much greater growth than normal. When such pituitary preparations were injected into animals fed diets containing the various vegetable fats, margarine, or butter, all animals were equally able to sustain the exaggerated growth.

In the literature during the past few years there has been considerable discussion about vaccenic acid which is presumably present in butter and which is not present in vegetable fats. It has been claimed by some Dutch workers in experiments carried on during the war that animals raised on butterfat grew at a greater rate than those which were raised on rapeseed oil. These Dutch workers ascribe the variations in growth to the presence of vaccenic acid in the butter. They believed that they had proved that the addition of what they thought was pure vaccenic acid to the diet prevented the differences in growth. In experiments which we completed a few months ago in my laboratory, we attempted insofar as we could to repeat the experiments of the Dutch workers. However, in these tests we again demonstrated that no difference in growth rate occurred when butterfat and cottonseed oil were compared in two different series of tests. Moreover, although we obtained a somewhat lower growth rate when rapeseed oil was substituted as the fat in such diet, we could show no beneficial effect whatsoever when an extremely pure vaccenic acid was added to the diet, nor could we demonstrate any improvement when a natural source of vaccenic acid, namely hydrogenated China wood oil or hydrogenated tung oil, was also added to the diet. Vaccenic acid did not accelerate the growth of the animals receiving cottonseed oil. The results of these tests have only become available by publication in the March number of the Journal of Nutrition.

On the basis of all these facts, you can see that there is not any mysterious X factor present in butter which is necessary to growth. Certainly one cannot demonstrate any difference in the rate of growth between butter and vegetable oils when the diets fed to the animals are complete in respects other than the fat and when care is taken to see that comparisons are not made between animals that are on one hand eating a large amount of food and animals in another case which for some reason or other have refused to eat a comparable amount of the diet.

Based on such data, I should certainly be definite in my statement that butter is in no way superior to margarine. I share my opinion on the comparative nutritive value of butter and margarine with many other authorities in the field. The committee on public health relations of the New York Academy of Medicine has formally recommended:

That wide publicity, both lay and professional, be given the fact that oleomargarine, fortified by vitamin A, is nutritionally equal to butter.

The food and nutrition board of the National Research Council in the reprint and circular series No. 118, August 1943, reviews this subject in full and I quote one relevant sentence:

The present available scientific evidence indicates that when fortified margarine is used in place of butter as a source of fat in a mixed diet, no nutritional differences can be observed.

The American Medical Association, in a formal published report of its Council on Food and Nutrition, takes a similar position.

The United States Federal Security Agency, in its attempt to popularize nutrition during the war, emphasized the needs of seven basic food classes. It included butter, and fortified margarine in one group, attributing equal value to each.

Finally, other evidence on the satisfactory nutritive status of margarine has only within the last 2 months been published in the Journal of the American Medical Association. This article, of which Carlson and others were coauthors, is a report of experiments carried on over a 2-year period on a number of children in an orphanage. Ine group received butter and the other received margarine during the entire period. That is in addition, of course, to other foods, but margarine and butter were the main fats. There was no evidence that any difference in growth or in well-being obtained between the two groups. In fact, there was some indication that the results on the margarine tests were somewhat superior to those to the butter tests although there were no differences which were statistically significant. It should be emphasized that these were experiments carried on by a large group of individuals at a time of growth when their requirements for any specific nutrients in the fat are the greatest. Certainly if there was a requirement for specific fat components at this period, it should have become evident by some variation in weight or physical well being of the children who were receiving a deficient diet. This did not occur.

The evidence which I have recited to you is to a very considerable extent first-hand evidence from my own experiment, but it has been supported by a very considerable amount of investigation. I am thoroughly convinced that the fat requirements can just as effectively be served if margarine is present in the diet as is butter. As a biochemist and a nutritionist, I feel that restrictive taxes should be repealed from such an excellent food as margarine.

In considering all aspects of the butter-margarine controversy, one should consider the need for fat in the diet. There seems to be definite experimental evidence that growth, physical endurance and various other indices of nutritive value are improved when the animals consume a diet containing relatively high proportions of fats as contrasted with those on a low-fat or fat-free regime. In some recent experiments, that we have carried on, it was found that the best performance was obtained when fat was present at a 20 to 40 percent level in the diet. Somewhat poorer results were obtained on a 10 percent level, still poorer results on a 5 percent level and much inferior results on a regime which contained practically no fat but did contain the unsaturated fatty acids. Of course, all the necessary vitamins were present.

If lower priced fats are more generally available for human consumption, there undoubtedly will be a tendency to use them much more freely than when an extremely expensive fat as butter is being employed. Therefore, the removal of the restrictions on margarine would encourage a greater use of margarine in cooking and on the

table, which in turn would seem to me to probably result in a further improvement in nutritional status of the average individual.

I should like to thank your committee for the attention you have given my testimony and will be glad to elaborate on any question which you may have.

Mr. ANDRESEN. Mr. Chairman.

The CHAIRMAN. Mr. Andresen.

Mr. ANDRESEN. I would like to ask you a few questions, Doctor. Your experiments have been going on since 1940.

Dr. DEUEL. They have been going on since 1917.

Mr. ANDRESEN. That is, your personal experiments in the field of margarine?

Dr. DEUEL. My personal experiments in the field of margarine have largely been since 1940, but I have worked on many vegetable fats long before that, on rates of digestibility, on rates of absorption, on the effect of different fats in producing what happens to them when they get in the body. I have been working on that since 1917. Mr. ANDRESEN. Do you have detailed descriptions of each experiment that you have conducted?

Dr. DEUEL. A large proportion of the experiments are published in some of these 175 papers, which include some of the scientific series from the Department of Agriculture as early as 1918.

Mr. ANDRESEN. Would you submit the details on your various experiments to this committee?

Dr. DEUEL. I will be happy to, except a number of the earlier experiments; a number of the earlier reprints are no longer available. I can submit a list of all the experiments and the available reprints will be submitted to you.

Mr. ANDRESEN. Would you submit the details of those experiments on which your statement is based?

Dr. DEUEL. Yes, sir.

Mr. ANDRESEN. So that we can make those a part of the record. Dr. DEUEL. Yes, sir.

(The information is as follows:)

[Reprinted from the Journal of Nutrition, Vol. 27, No. 1, January 1944]

STUDIES OF THE COMPARATIVE NUTRITIVE VALUE OF FATS I. GROWTH RATE AND EFFICIENCY OF CONVERSION OF VARIOUS DIETS TO TISSUE 1

(Harry J. Deuel, Jr., Eli Movitt, Lois F. Hallman, and Fred Mattson, with the technical assistance of Evelyn Brown, Department of Biochemistry and Nutrition, University of Southern California School of Medicine, Los Angeles)

[Received for publication August 4, 1943)

There are conflicting reports in the literature as to whether various animal and vegetable fats have similar nutritive properties aside from their vitamin content. It is known that with the exception of a few fats having abnormally high melting points, all are equally well absorbed in human subjects, the coefficients of digestibility approaching 95 percent (Holmes and Deuel, 1921; Deuel and Holmes, 1922). The requirement for certain unsaturated fatty acids present in

1 This work was carried out under a research grant from The Best Foods, Inc. The authors wish to acknowledge the helpful advice of Prof. Anton J. Carlson of University of Chicago, of Prof. Arthur W. Thomas of Columbia University and of Dr. H. W. Vahlteich of The Best Foods, Inc. during the course of the experiments.