Nobel Lecture, December 11, 1929

The Earlier History of Vitamin Research

When the present century began, animal nutrition was being viewed too exclusively from the standpoint of energy requirements. The fundamental pioneer work of Rubner and its later extension to human subjects in the remarkable enterprise of Atwater, Benedict, Rosa, and others in the United States could not fail to produce a deep impression upon the thought of the time. The quantitative character of the data obtained and the attractive circumstance that such data appeared to supply a real measure of nutritional needs, independent of, and apparently superior to, considerations based upon chemical details, induced a feeling that knowledge concerning these needs had become highly adequate and was approximating even to finality. As a matter of fact, however, these calorimetric studies, invaluable in themselves, were then leading to doctrinal teaching which contained inherent errors. So fundamental an aspect of the then dominant doctrine as, for instance, the law of isodynamic equivalence among foodstuffs, is at the most approximately true, and fails entirely when the equivalence is tested by physiological results rather than by purely physical data. The assumption indeed that carbohydrates and fats can replace each other indefinitely in a diet, so long as the total energy supplied remains the same, has led to serious errors in practical dietetics. More serious in leading to error was the assumption that all proteins were of equal nutritive value, and most serious of all in this respect was, as we now know, the belief that proteins, carbohydrate, fat, and suitable in organic materials supplied in themselves all the needs of the organism. Inhibitory, moreover, was the odd assumption often to be detected in the writings of leading authorities that to view nutritional needs from the standpoint of energetics was not alone more convenient, but more scientific, and even more philosophical, than to discuss them in terms of the material supply.

It must, of course, be most fully recognized that the calorimetric studies so dominant twenty years ago and, no less, the quantitative studies of respiratory exchange which aided and extended them, provided knowledge which is of the utmost importance and a permanent asset of science. The prime demand of the active organism is for energy, and when all the more specific demands are met in a well-balanced food supply, the available energy becomes, of course, the limiting factor which determines the adequacy of that supply. The studies in question were, moreover, usually made upon subjects consuming natural foods, and the diet was therefore, in general, sufficiently well balanced. When it was not, the inevitable brevity of calorimetric observations failed to reveal effects of dietetic deficiencies which require time for their display. Such studies have therefore retained their value and will yet doubtless be extended: but had they continued to dominate the whole field of activity, had they not been supplemented by a more discriminative enquiry into the various factors which determine real adequacy in nutrition, our knowledge would have remained highly imperfect, and our views erroneous. For now that the organic material factors present in dietaries actually capable of maintaining normal nutrition for long periods have been more carefully explored, it has become evident that, so far from comprising protein, fat, and carbohydrate alone, they are many and diverse, and each is indispensable. We have even come to know that a diet which, apparently at least, will support the individual throughout life may yet lack a factor which is necessary to maintain an adequate capacity for parentage, so the diversity of materials indispensable for normality is seen to be yet greater. The complexity of these nutritional needs as we now view them is indeed astonishing.

We find them displayed in the details of the protein supply, in the call for a right balance among the inorganic constituents of a diet, and, particularly, in the urgent call of the body for a number of organic substances specific in nature and function, in respect of which, however, the quantitative supply is, in accordance with the demand, so small as to contribute little or nothing to the energy factor in nutrition. These substances, following the suggestion of Casimir Funk, we have agreed to call vitamins.

Who was the “discoverer” of vitamins? This question has no clear answer. So often in the development of science, a fundamental idea is foreshadowed in many quarters but has long to wait before it emerges as a basis of accepted knowledge. As in other cases, so with recognition of vitamins as physiological necessities. Their existence was foreshadowed long ago, but a certain right moment in the history of the science of nutrition had to arrive before it could attain to universal recognition. Some workers had discovered suggestive facts, but failed to realize their full significance. On the other hand, the work and words of true pioneers lay forgotten because published when average minds were not ready to appraise them at their right value.

Some fifteen or sixteen years ago, the importance of vitamins became somewhat suddenly recognized. So enormous now is the existing literature concerning them, so complex and, sometimes, so uncertain, are the issues raised, that it is impossible to survey the subject adequately in a single lecture. The circumstances of my most enviable position here today will justify me in dealing rather with the earlier history of the subject, and I will venture in virtue of that position to put before you certain personal experiences which have no place in the proper history of the subject. They have not been, and will not be, published elsewhere.

No one can deny that the recorded experience of voyagers and explorers in the eighteenth century, and particularly perhaps the records of the British Navy which deal with the incidence and cure of scurvy, would have directed thought towards our modern conception of vitamins, had the times been ripe. The knowledge concerning nutrition was then, however, entirely vague, and the days of experiment in such matters had not yet come. The earliest experiments indeed supplied evidence which was of an independent kind reached from a different angle.

It is now generally agreed that the first clear evidence, based upon experiment, for the existence of dietary factors of the nature of vitamins came from the school of Bunge at Basel. In 1881 Lunin, one of the workers in that school, fed mice upon an artificial mixture of the separate constituents of milk; of all the constituents, that is, which were then known, namely the proteins, fats, carbohydrates, and salts. He found that upon such a mixture the animals failed to survive and was led to conclude that “a natural food such as milk must therefore contain besides these known principal ingredients small quantities of unknown substances essential to life”. Such a statement, already half a century old, when allowed to stand out clear and apart from a context which tended to bury it, seems to contain the essentials of what is believed today.

It will not be uninstructive to seek for an explanation as to how such a significant remark could remain for years almost without notice, instead of being an immediate challenge to further investigation. It may be noted first of all that the title of Lunin’s paper, “Ueber die Bedeutung der anorganischen Salze für die Ernährung des Thieres” was one prone to conceal the important but incidental suggestion it contained. The thought of the Bunge school was, as a matter of fact, both then and long afterwards concerned largely with the inorganic factors of nutrition. After an interval of ten years, during which, so far as one can discover, no effort had been made to follow up the pregnant suggestion of Lunin’s work, another member of the school came again to Lunin’s point of view. He, however, was once more concerned primarily with the study of an inorganic element in nutrition. Socin, the worker in questin, though, as the title of his paper, “In welcher Form ist das isen resorbiert?”, indicates, he had not set out to seek further for those wholly unknown substances essential to life, the existence of which Lunin had predicted, was nevertheless owing to the nature of his experiments inevitably driven to a belief in them. Socin, it is true, was led to think that the ineffectiveness of the synthetic diets employed by him might be due to an inadequacy in the quality of the protein contained in them, a view not wholly unjustified and foreshadowing knowledge which only long afterwards became definite and proven. He became clear, however, that other and unknown substances must be sought, substances which, he remarks, are certainly present in whole milk and egg-yolk. He remarks, moreover, with conviction that to discover them was the first task to be faced, before new feeding experiments are undertaken. Here was a strong enough challenge to investigate, yet we have again, I think, no knowledge of any attempt in Bunge’s laboratory to follow up the challenge. Perhaps an explanation of this is to be found in the circumstances that Bunge, though in his well-known book he remarks that it would be worth-while to continue the experiments which had suggested the existence of unknown nutritional factors, was, as I happen to know, himself inclined to disbelieve in them. He thought that the real error in the synthetic diets used by his pupils (which was, so to speak, “dissected milk”) was that the method of its preparation had involved the separation of inorganic constituents from certain organic combinations in which latter form alone could they adequately subserve the purposes of metabolism. Other causes may have contributed to the long neglect, first of Lunin’s and then of Socin’s suggestions, and it must be admitted indeed that no experimentum crucis was carried out in Bunge’s laboratory.

In Lunin’s experiments the fate of six mice only (those placed by him upon a normal salt mixture) really suggested the existence of unknown factors; and no data are given as to their consumption of food. Neither Lunin nor Socin made any attempt to complete the evidence by making discriminative additions to the diets which had proved inadequate. Lastly, as I have already suggested, since the main intention of their work, and the titles of their publications, were remote from the special issue, their significant remarks might well appear as mere obiter dicta when read without the light of modern developments.

Yet the pregnant suggestions arising from the observations just discussed did ultimately, though not for fourteen years after the latest of them were published, awaken (as we may suppose) the interest of an investigator distinguished in many fields who was led to repeat and extend them. I allude to the late Professor Pekelharing, whose own observations (published in 1905) unhappily again remained unknown to the majority till very recently. It is indeed astonishing that the results of such significant work as his, though published in the Dutch language alone, should not have become rapidly broadcast. I cannot refrain from referring to the circumstance that the paper was not abstracted or mentioned in Maly’s Jahresbericht für Thierchemie, so adequate, and in general so complete, in its dealings with current literature. Many of us were accustomed to rely upon it for references to work published in journals that we could not consult, or in a language that we could not read. Though other work by Pekelharing was duly recorded at this time, no mention was made of the extraordinarily interesting paper in question. My own experiments began soon after the paper was published, and as a proportion of my own work was very similar to that of Pekelharing, I shall never cease to regret that, in common with so many others, I was then completely ignorant of the latter. After speaking of experiments carried out on lines similar to some of those done in Bunge’s laboratory, and indicating that they pointed to the existence of some unknown essential, Pekelharing goes on to say: “Till now my efforts, constantly repeated during the last few years, to separate this substance and get to know more about it, have not led to a satisfactory result, so I shall not say any more about them. My intention is only to point out that there is a still unknown substance in milk which even in very small quantities is of paramount importance to nutrition. If this substance is absent, the organism loses the power properly to assimilate the well-known principal parts of food, the appetite is lost and with apparent abundance the animals die of want. Undoubtedly this substance occurs not only in milk but in all sorts of foodstuffs both of vegetable and animal origin.” Here we have a clear statement of the vitamin doctrine already a quarter of a century old. It is noteworthy that Pekelharing records prolonged endeavours towards the isolation of a vitamin.

Between the publications of Lunin, Socin and that of Pekelharing appeared the extremely important papers of Professor Eijkman. There is no need for me to discuss those remarkable publications. A most adequate review of them is fresh in your minds. The fundamental importance of Professor Eijkman’s. work to the whole subject of vitamins and deficiency diseases is now universally recognized. It would be difficult to rejoice wholeheartedly in one’s own good fortune had I not the happiness of sharing with him the Nobel Prize for Medicine.

In my own earlier thought about the subject, though in so far as I was familiar with Eijkman’s observations I could not fail to recognize their importance, yet I thought of them in a wrong and too narrow category. Eijkman’s own earlier teaching as based on his results was that the function of the substance in the cortex was to neutralize a nutritional error due to excess of carbohydrate in a diet of rice. A substance which functions in the neutralization of an error is not the same thing as a substance universally necessary, and it was to the existence of substances of the latter type that my own thoughts had turned. Eijkman did not at first visualize beriberi clearly as a deficiency disease; but the view that the cortical substance in rice supplied a need rather than neutralized a poison was soon after put forward by Grijns and ultimately accepted by Professor Eijkman himself. I was myself, however, quite ignorant of these later views when my experiments were begun twenty-four years ago.

It is abundantly clear that before the last century closed, there was already ample evidence available to show that the needs of nutrition could not be adequately defined in terms of calories, proteins, and salts alone. How came it then that this limited definition was then in vogue and so remained for the first ten years of the present century, while no effective attention was given to the facts which have been discussed? The main reason, I feel sure, is that the minds of the leaders of thought in nutritional science were obsessed by a sense of the overwhelming importance of calorimetric studies with their impressive technique. A few experiments done mostly on small animals, and not very quantitative or crucial in their nature, were at best unimpressive. The former studies were clearly scientific and sound; the results of the latter appeared to be on a less sound basis, while deficiency diseases, in so far as their existence was proven, seemed to stand in so special a category that their indications might be neglected when the needs of normal nutrition were being estimated.

No general or widespread belief in the view that an adequate diet must contain indispensable constituents other than adequate calories, a minimum of protein, and a proper mineral supply, could be said to exist till the years 1911-1912. Those years saw the appearance of my own publications. Thereafter started a period of great activity in the study of the facts: immediately in the United States of America, a little later, and still more after the war, in many centres. I will now refer to my own experiments, and will also intrude upon your patience with a reference to those personal experiences of which I spoke at the beginning of this address.

Early in my career I became convinced that current teaching concerning nutrition was inadequate, and while still a student in hospital in the earlier eighteen nineties I made up my mind that the part played by nutritional errors in the causation of disease was underrated. The current treatment of scurvy and rickets seemed to me to ignore the significance of the old recorded observations. I had then a great ambition to study those diseases from a nutritional standpoint; but fate decreed that I was to lose contact with clinical material. I had to employ myself in the laboratory on more academic lines. I realized, however, as did many others at the last century’s close, that for a full understanding of nutrition, no less than for an understanding of so many other aspects of biochemistry, further knowledge of proteins was then a prerequisite; and when I was first called to the University of Cambridge I did my best to contribute to that knowledge.

As an ultimate outcome of my experiments dealing with the relative metabolic importance of individual amino acids from protein, my attention was inevitably turned, without, I think, knowledge, or at any rate without memory, of the earlier work, to the necessity for supplying other factors than the then recognized basal elements of diet if the growth and health of an animal were to be maintained. This indeed must at any time come home to every observer who employs in feeding experiments a synthetic dietary composed of adequately purified materials. It was the experience of the workers in Bunge’s laboratory long ago; it was, as we have lately learned, the experience of Pekelharing. A good many investigators using synthetic dietaries have, it is true, from time to time expressed doubts upon the point, but we now know that it was because the constituents they used were not pure and not free from adherent vitamins. In 1906-1907 I convinced myself by experiments, carried out, as were those of Lunin and Socin, upon mice, that those small animals at any rate could not survive upon a mixture of the basal foodstuffs alone. I was especially struck at this time, I remember, by striking differences in the apparent nutritive value of different supplies of casein in my possession. One sample used as a protein supply in a synthetic dietary might support moderate growth, while another failed even to maintain the animals. I found that a sample of the former sort, if thoroughly washed with water and alcohol, lost its power of support, while addition of the very small amount of extract restored this power and also, if added to the samples originally inadequate, made them to some degree efficient in maintaining growth. I found further at that time (1906-1907) that small amounts of a yeast extract were more efficient than the casein extracts. Similar experiences were encountered when otherwise adequate mixtures of amino acids were used to replace intact proteins. By sheer good fortune, as it afterwards turned out, I used butter as a fat supply in these early experiments. Upon the evidence of these earlier results I made a public statement in 1907 which has been often quoted. I cannot, however, justly base any claims for any sort of priority upon it, as my experimental evidence was not given on that occasion. It was indeed not till four years later that I published any experimental data. In explanation of this delay I would ask you to consider the circumstances of the time. The early experiments of Lunin and others had been forgotten by most; the calorimetric studies held the field and tentative suggestions concerning their inadequacy were, I found, received with hesitation among my physiological acquaintances. It seemed that a somewhat rigid proof of the facts would be necessary before publication was desirable. Thus came the great temptation to endeavour to isolate the active substance or substances before publication, and I can claim that throughout the year 1909 I was engaged upon such attempts, though without success. At this time I was using what is now the classic subject for vitamin studies, namely the rat. As I was concerned with the maintenance of growth in the animal, the tests applied to successive products of a fractionation took much longer than those which could be used in studying the cure of polyneuritis in birds by what we have learned to call vitamin B, so the work occupied much time. I may perhaps be allowed to mention what was for me a somewhat unfortunate happening in the beginning of 1910, as it is instructive. A commercial firm had prepared for me a special extract of a very large quantity of yeast made on lines that I had found effective on a small scale. With this I intended to repeat some fractionations which had appeared promising. I thought, however, upon trial that the whole product was inactive and it was thrown away. The real explanation, however, was that instead of using butter, as in earlier experiments, I at this moment determined to use lard, and my supply of this, as I learned to understand much later, was doubtless deficient in vitamins A and D; I was now giving my animals in the main the B-group alone. If I had then had the acumen to suspect that any of the substances I was seeking might be associated with fat I should have progressed faster. Later in 1910, if I may intrude so personal a matter, I suffered a severe breakdown in health, and could do nothing further during the year. On my return to work I felt that the evidence I had by then accumulated would be greatly strengthened by a study of the energy consumption of rats, on the one hand when failing on diets free from the accessory factors (as I had then come to call them), and, on the other hand, when, as the result of the addition of minute quantities of milk, they were growing vigorously. These experiments took a long time, but they showed conclusively, as at that time it seemed necessary to show, that the failure in the former group of animals was not due primarily, or at the outset of the feeding, to any deficiency in the total uptake of food.

My 1912 paper is sometimes unfairly quoted as though its bearing applied only to the influence of minimal quantities of milk upon nutrition. It will be found, however, that it emphasizes on general lines the indispensable nature of food constituents which were then receiving no serious consideration as physiological necessities.

In a personal endeavour to estimate the influence of my publications in 1912 upon the opinion of the time, and their relative importance in the initiation of that great activity in kindred studies which shortly followed their appearance, I have found it necessary to consider at the same time, and particularly, the work and writings of Casimir Funk. It is sure that, until the period 1911-1912, the earlier suggestions in the literature pointing to the existence of vitamins lay buried. There is no evidence, I think, that they were affecting the orientation of any authoritative teaching concerning the phenomena of normal nutrition either at the time in question or indeed, in any effective sense, before.

A few years ago in the American journal Science, Funk published a short article in which, after giving me some credit for prophetic vision, he protests against my being called the “discoverer of vitamins”. In this protest he was justified; I have certainly never made any personal claim to be their “discoverer”, and all the past circumstances of which I have reminded you, have deprived perhaps every individual worker of that clear title. Funk, however, further remarks in the article mentioned that my chief paper appeared too late to affect the situation to any appreciable degree; a remark which I believe to be entirely unjust.

F. Röhmann, an experienced worker on nutritional problems, and much concerned with the chemical side of them, but one who never fully believed in the claims made for vitamins, wrote in 1916 after discussing the earlier literature “Als der geistige Vater der Vitaminlehre ist wohl Gowland Hopkins zu betrachten, während die Bezeichnung Vitamine von Casimir Funk herstammt.” Such a statement without extension does of course far less than justice to Funk’s influence, which in many ways was important. Funk himself, however, made no experiments which bore upon the physiological functions of vitamins until long after my paper had appeared.

Funk’s first entry into kindred fields was in a paper published in December 1911, describing his earliest efforts to isolate the curative substance from rice polishings. He continued this effort, and further papers appeared in 1912, when Suzuki and others were also describing their endeavours to isolate the substance. Funk’s attempt was extremely praiseworthy, and his publications doubtless awakened new interest in Eijkman’s original discovery. He did not, however, succeed in isolating any substance which has since been accepted as being in fact any actual vitamin of the B-group, and the papers in question contain no suggestion bearing on the general’ physiological importance of vitamins.

In June 1912, however, Funk published a paper dealing with the Etiology of Deficiency Diseases, which was a valuable summary of the existing knowledge concerning such conditions, and here he showed more clearly than had been shown before, how definite a group is constituted by these diseases. He also emphasized in an interesting way and in advance of general opinion, the view that pellagra would prove to be one of them. This review contains, however, no account of personal work other than the attempts already mentioned to isolate the anti-beriberi factor. A short final section purports to relate the available knowledge concerning deficiency diseases to the facts of normal metabolism and nutrition. I feel justified in saying that this section is written in a manner which is essentially disingenuous. The author says, for instance, “I suppose that the substance facilitating growth found in milk is similar to, if not identical with, the vitamins described by me.” But Funk had then “described” no vitamins. He had in this review – the merits of which I have already emphasized – on theoretical grounds and with reliance mainly on the work of others – only made the doubtless very significant suggestion that each deficiency disease might depend upon the absence of its own specific factor. He admits knowledge of the existence of my experimental results; knowledge obtained – according to a footnote – through a “private communication”. In the article in Science to which I have earlier referred, Funk states on the other hand that he had in 1912 no knowledge of my work. I am entitled to say that early in October 1911, I gave a very full account of my ‘results at a meeting of the English Biochemical Club, and I certainly obtained the impression during the succeeding months that they had become well-known to English physiologists and biochemists generally.*

In closing these references to Casimir Funk’s writings and to his earlier work I would like to make clear my belief that he has not received too much, but too little, credit for his vitamin work as a whole. I venture to think, however, that he was in no sense my predecessor in the physiological field. I may say that till now I have had no intention of commenting on his remarks in the Science article of 1925. Only the peculiar circumstances of my present position, which seem incompatible with his view, have led me to discuss it.

Very soon after my chief paper appeared the study of vitamins was, as you know, developed with great energy and success in the United States. We owe especially to Osborne and Mendel, and to McCollum and his co-workers, the all important work which continued during the earlier years of the war. The proof on the part of the American authors of the distinction between what were then known as the “water soluble” and “fat soluble” vitamins was the fast clear evidence of diversity among the vitamin factors required for growth. This discovery made all later studies more discriminating, and the pioneer work of the authors mentioned was of the greatest importance. So prominent indeed was the American work at this time, and so large a proportion did it form of the total output from 1912 to near the end of the war that, if I wished to claim that my own publications exerted any real and effective influence in starting a new movement in the study of dietetics, I should have to convince myself that they helped to direct the thoughts of the Harvard and Baltimore investigators. Anyone reading with care the succession of papers describing their experimental studies before and after the appearance of my own publications in 1912, will, I think, become convinced that such directive influence was indeed exerted. This circumstance and much correspondence received at the time from European colleagues made me feel then that my papers had served the purpose I had wished for it, namely, to direct thought concerning normal nutrition into a channel which, if not new, had been long and strangely neglected. For a time indeed I thought that channel to be even new. I was at least a pioneer whose efforts were not wasted, and I am always now content to recall an opinion expressed by the late Franz Hofmeister, the most just, if also the most generous of critics. Hofmeister in 1918, after an exhaustive study of the whole literature, speaks of me as the first to realize the full significance of the facts. If that be true, and if, as well may be, that has been the view of the Nobel Commissioners who have thought me worthy of so great a reward, I can happily enjoy my good fortune.

* On February 28th and again on March 8th, 1911, the London Daily Mail gave great publicity to certain statements of my own which, though distorted for journalistic purposes, made essentially clear the conception of vitamins as based upon my personal experiments. These articles were quoted in the continental and, with especial freedom, in the American press. At the time I much regretted this unsought publicity. My main paper, the publication of which was much delayed in the press, appeared in the month following Funk’s review.