ESSE s= MBL/WHC 1 D If II D 03D1 DDb; GIBA FOUNDATION COLLOQUIA ON AGEING Vol. I. General Aspects Two other series of books from The Ciba Foundation are issued as ""Colloquia on Endocrinology''' and ''General Symposia''. A leaflet giving details of all these volumes is available from the Publishers. /^9 v^ •^ , .* i\ ^ r M- M^l ^^^ I'^-'-T 0 '^ 1 1 > 1 1 i m =^^^^r- -^ ''^^^^ ■ '&!)•' 'tf *" ' ■m-- M'^ ^l ■^^ Thk Bkii)(;e of Life liepwduced by pcDuission from "The C'liances of Death, and other studies in evohition," by Karl Pearson. 1807. London: Kdward Arnold.) Frdtitlspifcf] ^ 7 /' ^ CIBA FOUNDATION COLLOQUIA ON AGEING VOLUME I General Aspects Editors for the Ciba Foundution G. E. W. WOLSTENHOLME, o.B.K, M.A., M.B., B.Ch. and MARGARET P. CAMERON, M.A., A.B.L.S. Assisted by JOAN ETHERINGTON With 38 lHustrations LITTLE, BROWN AND COMPANY 1955 BOSTON THE CIBA FOUNDATION for the Promotion of International Co-operation in Medical and Chemical Research 41 Portland Place, London, W.l. Trustees: The Right Hon. Lord Adrian, O.M., P.R.S. The Right Hon. Lord Beveridge, K.C.B., F.B.A. The Right Hon. Lord Horder, G.C.V.O. Mr. Raymond Needham, Q.C. Director, and Secretary to the Executive Council: Dr. G. E. W. \Volstenholme, O.B.E. Assistant Secretary: Miss N. Bland ' Scientific Assistant: Miss M. P. Cameron, M.A. Editorial . 1 ssi slants : Miss C. M. O'Connor, B.Sc. Miss E. C. P. Millar, A.H.VV.C. Librarian: Miss Joan Etherington All Rights Reserved This book may not be reproduced by any means, in xvhole or in part, with- out the permission of the Publishers Printed in Great Britain and published in London by J. db A. Churchill Ltd. PREFACE The Ciba Foundation's international conferences are arranged, as often as possible, at times when the members can on the same journey also participate in other appropriate conferences in Western Europe. As early as 1952 Professor R. E. Tunbridge asked whether the Ciba Foundation would be willing to hold a suitable conference near the time of the third Congress of the International Association of Geron- tology, to be held in London in July 1954, and to this the Director of the Foundation readily agreed. Early in 1954 the Trustees of the Ciba Foundation decided to embark on special measures in support, internationally, of basic research relevant to the problems of ageing, so that the conference already arranged became the first in what it is hoped will be a series of conferences on subjects in this field. In conformity with the Foundation's earlier series of "Collo- quia on Endocrinology" the new series is being called "Colloquia on Ageing," whilst the conferences on isolated subjects will continue to be described as "Symposia". In the arrangements for this colloquium on the General Aspects of Ageing, the Director was greatly helped by the advice he received from Professor Tunbridge, who also kindly undertook the duties of Chairman. The colloquium was intended to be a general exploration and appreciation of the present position in regard to opinion and experiment on the processes associated with or directly involved in the changes occurring in tissues with age, at whatever period of life of the organism from conception to death. To those to whom this book serves as an introduction to the activities of the Ciba Foundation it should be explained that it is an international centre, which is established as an educational and scientific charity under the laws of England. vi Preface It owes its inception and support to its founder, CIBA Ltd. of Switzerland, but is administered independently and exclu- sively by its distinguished British Trustees. The Foundation provides accommodation for scientific workers who visit London from abroad, organizes and holds international conferences, conducts (in conjunction with the Institut National d'Hygiene) a postgraduate medical exchange scheme between England and France, arranges informal meetings for discussions, awards an annual lectureship, has initiated a scheme to encourage basic research relevant to the problems of ageing (as mentioned above), assists international congresses and scientific societies, is building up a library service in special fields, and generally endeavours to give aid in all matters that may promote international co- operation in scientific research. Leading research w^orkers from different countries and in different disciplines are invited to attend the symposia or colloquia. The size of the group is, however, very strictly limited in order to obtain a free conversational manner of discussion — although the basic timetable of the programme is strictly observed. The smallness of the groups means the exclusion of many workers active and interested in the subjects discussed, and therefore the proceedings of these conferences are published and made available throughout the world. It is hoped that the papers and discussions in this book will prove not only informative and stimulating, but will also give to readers a sense of participation in an informal and friendly occasion. CONTENTS pa(;e Chairman's opening remarks R. E. TUNBRIDGE ....... 1 The definition and measurement of senescence by P. B. Med A WAR ..... Some remarks on the pathological basis of ageing by G. R. Cameron ....... 16 Discussion: Brull, Cameron, Comfort, Cowdry, Franklin, Krohn, Lansing, McCay, Medawar, Olbrich, Schulze, TUNBRIDGE, VeRZAR, ViSCHER ..... 23 Mental aspects of ageing by Aubrey Lewis ....... 32 Discussion: Comfort, Cowdry, Freeman, Lansing, Lewis, Olbrich, Shock, Tunbridge, Vischer .... 48 General Discussion: Albertini, Aub, Brull, Cameron, Christie, Cowdry, Franklin, Lansing, McCance, Medawar, Miescher, Rubin, Schulze, Shock, Tunbridge, Verzar ......... .52 Effects of ageing on respiratory function in man by D. V. Bates and R. V. Christie .... 58 Discussion: Brull, Christie, Franklin, McCance, NicoLAYSEN, Schulze, Shock, TtJNBRiDGE, Verzar . . 65 Changes with age in diffusion coefficients of solutes for human tissue membranes by J. E. Kirk and T. J. S. Laursen . . . .09 Discussion: Aub, Bean, Franklin, Kirk, Lansing, McCance, Medawar ......... 76 73499 viii Contents PAGE The changing incidence of certain vascular lesions of the skin with ageing by W. B. Bean . 80 J Ageing of elastic tissue and the systemic effects of elastase by A. I. Lansing ....... 88 Discussion: Albertini, Aub, Comfort, Cowdry, Lansing, McCance, Medawar, Tunbridge, Verzar . . . 108 Calcium metabolism in old age as related to ageing of the skeleton by O. J. Malm, R. Nicolaysen and L. Skjelkvai.e . 109 Discussion: Aub, Brull, McCay, Nicolaysen, Tunbridge . 123 17-Ketosteroid excretion in ageing subjects by Betty L. Rubin, R. I. Dorfman, and G. Pincus . 126 Discussion: Comf'Ort, Freeman, Krohn, Lewis, Medawar, MiEscHER, Olbrich, Parkes, Rubin, Shock, Tunbridge 137 Tissue transplantation technique applied to the problem of the ageing of the organs of reproduction by P. L. Krohn ........ 141 Preservation of tissue in vitro for the study of ageing by A. S. Parkes 162 Discussion: Bean, Comfort, Cowdry, Krohn, Medawar, Parkes, Verzar . . . . . . . .169 Research areas in gerontology nutrition that are nov^ neglected by C. M. McCay 173 A fantasy of ageing and the bearing of nutrition upon it by R. a. McCance and E. M. Widdowson . . .186 Too rapid maturation in children as a cause of ageing by H. M. Sinclair 194 Discussion: Aub, Bartlett, Brltll, Comfort, Cowdry, Franklin, Krohn, Lansing, McCance, McCay, Medawar, MiESCHER, Nicolaysen, Olbrich, Parkes, Shock, Sinclair, Tunbridge, Verzar ....... 201 Contents ix Psychological aspects of ageing page by Sir F. C. Bartlett 209 Discussion: Bartlett, Franklin, Lewis, Parkes, Sciiulze, Shock, J. Vkrzar ....... 213 Adrenocortical reactivity in aged schizophrenic patients by H. Freeman, G. Pincus, F. Elmadjian and Louise P. Romanoff ........ 219 Discussion: Freeman, Lewis, Nicolaysen, Olbrich, Rubin, Shock 236 General Discussion: Brull, Comfort, Cowdry, Franklin, Krohn, Lansing, Nicolaysen, Shock, Tunbridge . 238 List of those participating in or attending the Colloquium on Ageing: "General Aspects," 13th-15th July, 1954. A. VON Albertini, J. C. AuB Sir F. C. Bartlett W. B. Bean L. Brull G. R. Cameron R. V. Christie A. Comfort . E. V. COWDRY K. J. Franklin H. Freeman F. Gross V. R. Khanolkar J. E. Kirk . P. L. Krohn A. I. Lansing A. J. Lewis . R. A. McCance C. M. McCay P. B. Medawar Histopathological Inst., University of Zurich. Massachusetts General Hospital, Boston. Applied Psychol. Research Unit, Cambridge. Dept. of Internal Medicine, University Hospitals, Iowa City. Inst, de Clin. Medicale, Hopital de Baviere, Liege. Dept. of Morbid Anat., Univ. Coll. Hosp. Medical School, London. Dunn Laboratories, St. Bartliolomew's Hospital, London. Dept. of Zoology, University College, London. Dept. of Anatomy, Washington Univ. School of Med., St. Louis, Missouri. Dept. of Physiology, St. Bartholomew's Hospital Medical College, London. Worcester Foundation for Exper. Biology, Shrewsbury, Mass. Biology Division, CIBA Ltd., Basle. Indian Cancer Research Centre, Bombay. Div. of Gerontology, Washington University, St. Louis. Dept. of Anatomy, Birmingham University Medical School. Dept. of Anatomy, W^ashington Univ. School of Med., St. Louis; now at Dept. of Anatomy, Emory Univ., Georgia. Inst, of Psychiatry, Maudsley Hospital, London. Dept. of Exper. Medicine, University of Cam- bridge. Cornell University, Ithaca, X.Y. Dept. of Zoology, University College, London. Xll K. MiESCHER R. NiCOLAYSEN O. Olbrich . A. S. Parkes Betty L. Rubin w. schulze . J. H. Sheldon N. W. Shock H. M. Sinclair A. R. SOUTHWOOD R. E. TUNBRIDGE F. Verzar Jean Verzar A. L. Vischer List of Participants . CIBA Ltd., Basle. Inst, for Ernaeringsforsking, Oslo. Geriatric Unit, General Hospital, Sunderland. . National Institute for Medical Research, London. Worcester Foundation for Exper. Biology, Shrewsbury, Mass. University Clinic, Leipzig. Royal Hospital, Wolverhampton. Section on Gerontology, Baltimore City Hospitals. Lab. of Human Nutrition, University of Oxford. Dept. of Public Health, South Australia. Dept. of Medicine, University of Leeds. Physiological Inst., University of Basle. Physiological Inst., University of Basle. Basle, Switzerland. CHAIRMAN'S OPENING REMARKS R. E. TUNBRIDGE, O.B.E., M.D., F.R.C.P. Department of Medicine, University of Leeds. First of all, I know that it is your wish that I should thank the Ciba Foundation, on your behalf, for giving us this opportunity of meeting together to discuss the problem of ageing. The Director wisely chose as the title of the Sym- posium "Ageing — General Aspects". We have thus been spared any etymological discussion on what is the meaning of gerontology. During the course of our discussions we shall wander frc the most general aspects of the problem to the particular out I trust that even when we are discussing minutiae we shall keep in mind the general problem of ageing. Perhaps it is too much to hope that at the end of this Con- ference we shall be able to define ageing precisely and clearly — but possibly many of us would feel that a lot of fun had gone out of life if the riddle were to be solved so easily. Is ageing a chronological term, merely reflecting the passage of years, and if so, what years, or are the public right in assum- ing, as they generally do, that ageing is synonymous with senescence and/or decay? The concept of the elixir of life, like the philosophers' stone and the alchemists' dream of the transmutation of metals, has long served as a tremendous stimulus to mankind, to higher flights of imagination or some- times to derision. We shall not dwell upon these fantasies, nor shall we deal with that other very important aspect of the problem, what one might call the political, economic and social aspects of ageing, of which we as citizens cannot be unaware. We shall try, I hope, to confine ourselves to the more biological aspects in an attempt to define, if it is possible, the process of ageing. Is there an allotted span of life? Is there species specificity with regard to the life-span? Is the length of life determined 2 R. E. TUNBRIDGE by the structure of the chromosome? We can breed animals with longevity as a dominant, and it is well known that there are human families in which longevity is a striking characteris- tic. This is a fascinating aspect of th^ problem; Professor Medawar has very strong views on this approach to ageing, and we look forward to hearing his lucid exposition. Then again, is senescence itself a primary biological func- tion? Is it inevitable or is it merely the predominance of catabolism over anabolism? Is it a feature of cellular life, of the life of an organ or of an organism? Alexis Carrol in his experimental work suggested that environmental factors were highly important, if not the most important factor in sene- scence or the changes which we associate with senescence. Then we have the problem of the cyclic changes that occur in many organs, for example the thyroid or the ovary. Is the hypertrophy and involution a fundamental biological process, and is senescence allied to it, an exaggeration of it, or a similar process in another phase? I understand that the senescent or atrophied ovary of an adult animal can be trans- planted into a young ovariectomized animal of the same species, and that it then proceeds to take on cyclic function again. Therefore, is it merely the milieu in which the organ lives, the environmental factors, which determine age changes, or are they separate and distinct processes? We have with us experts like Dr. Betty Rubin, Dr. Parkes, Professor Krohn, who may not deal with this particular problem, but who as a result of their pre-eminence in the field of endocrinology are bound to touch on similar topics. Then, of course, there is the extraordinary power of organs or tissues to regenerate. We are familiar with the great powers of regeneration of, say, the liver, and not merely the growth but the readiness with which new cells take on apparently normal function, whereas in other tissues, like the nervous system, once the nerve cell is seriously damaged it is no longer capable of repair. We have with us many experts on the changes in tissues with ageing. Are the changes in the skin, the arteries, the bone, the lungs, just senescence? Are they Chairman's Opening Remarks 3 specific? Is senile osteoporosis, so common in elderly women, merely a manifestation of osteomalacia in a particular age group, or is it a specific degenerative change? It is interesting that elastic tissue so a})undantly laid down in youth seems to cease to be laid down, or only sparingly so, in old age, its place being taken by collagen. We have to decide on the relationship of these problems to the whole problem of ageing; the riddle is interminable, and possibly our own approach is coloured. I hope in our discussions we shall keep this in mind, and that we shall leave this colloquium invigorated and refreshed. The programme also refers to aspects of nutrition, and you will notice that because of the question of bias we have not neglected to include the psychiatric and psychological aspects of the subject. Here the field is less well-defined, and Professor Aubrey Lewds and Professor Sir Frederic Bartlett might say we are all highly biased and therefore truth is very difficult to come by. THE DEFINITION AND MEASUREMENT OF SENESCENCE P. B. MeDAWAK, D.Sc, f.r.s. Deportment of Zoology, University College, London. Introduction Nothing is clearer evidence of the immaturity of geronto- logical science than the tentative and probationary character of its system of definitions and measurements. "What is ageing?" is a question to which there is no agreed answer. It is therefore very natural that the first contribution to this colloquium should be an attempt to suggest the ways by which an answer might be arrived at. "Ageing", in the literal sense of merely growing older, might well be a beneficial process; i.e. there are certain reasons why an animal's expectation of further life should increase, instead of becoming shorter, with advancing years. Animals grow wiser as they grow older: so far as memory can influence their actions, the survivors of earlier hazards are less likely to be their victims when exposed to them for a second time. For example, Lack (1951) has shown that beyond the first few months of life the mortality of wild birds is independent of their age; but game birds (whose census is partly carried out by gunfire) may suffer a disproportionately high mortality during the first two years of life, after which the survivors presumably know better. Then again, animals become wiser in an immunological sense, for antibody formation is a process with a long memory, and a second response to an antigenic stimulus is much more effective than the first. (Immuno- logists were speaking of the anamnestic response, a not- forgetting of past behaviour, long before biologists came to regard the storage of information as a conceptual novelty.) Yet ageing in man is accompanied by that general deteriora- tion which the word itself colloquially implies, and the same Definition and Measurement of Senescence 5 is true of all mammals and birds which have been artificially kept alive long- enough to suffer the inroads of senile decay. The deterioration that goes with ageing could be described as a decline of "vitalitv" or a lowerino- of biolooical efficiencv — of an organism's power to maintain itself as a going concern; but this is more of an innuendo than a formal definition, and the question remains, how is senescence to be measured and defined with acceptable precision? (It will be convenient to use the seventeenth century word senescence to stand for the deterioration that accompanies ageing, and to leave ageing itself to stand for merely growing old.) The process of senescence and the state of senile deteriora- tion can be measured or assessed in two entirely different ways (Medawar, 1952). First, there is the kind of measure that may be called personal, because it purports to measure a process that takes place in the life history of an individual animal; second, there is a statistical or actuarial measure, which is founded upon the mortality of a population of individuals and which bears only indirectly upon the changes that occur within the lifetime of anyone. The former is used by the pathologist, physiologist, and (by an extension of his terms of reference) the embryologist. The latter is used by biologists generally, and in particular by ecologists and by students of the genetical theory of evolution. The Personal Measure of Senescence If we knew of some master factor or prime mover in the process of senescence — if, for example, every ingredient of decay was causally dependent upon changes occurring in the ])ineal gland — then a personal measure of senescence would be in principle entirely adequate. The one master process could be measured, and the rest should follow. But we know of no such master factor; the processes of senescence are in gear, but we do not know which is the driving wheel. All that can be done, therefore, is to record separately the several manifestations of decay — the wrinkling of the skin and grey- ing of hair, the anatomical involution of the organs, the 6 P. B. Medawar 4,yrowing insufticiency of endocrine output, the loss of acuity of the senses, the weakening of muscular power and coordina- tion, the decline of the specific growth rate and rate of cellular turnover, and so on. All these can be described and measured, in fact or in principle; but as mere description cannot allot causal priorities among their competing claims, no one of them can be held to be the measure of senile decay. Considered one by one, each of these measurements gives different estimates of the time of onset and rate of progress of senescence. Take, for example, the specific growth rate of the body, i.e. its rate of growth considered as a system increas- ing by compound interest, in which that which is formed by growth is itself capable of growing. Many years ago, Charles Minot pointed out that the specific or percentage rate of growth in man fell from birth onwards, rapidly at first, and then progressively more slowly. Man grows like money invested at a rate of compound interest which falls progressively at a rate which itself progressively falls. Minot argued that the specific growth rate was a particularly searching and intimate measure of vitality; it follows that senescence begins at birth and goes on faster in children than in their elders. Estimates of the rate of healing of wounds have been held to reveal a similar trend, but some of the data will not stand up to ill-disposed criticism. Measurements by other standards give different answers. Acuity of hearing is said to be sharpest at about the age of ten and resistance to infectious disease to be at its maximum at about fifteen. Muscular power and coordination are presumably at their peak at about age twenty-five. If therefore we think of the body as a multiplicative growth system, there is a sense in which senescence could be said to begin at birth; as an antibody forming system, at fifteen; and as a muscular machine at twenty-five. These several measures of vitality or senescence are therefore incongruent, and it is an important empirical fact that they are so; but for the time being we must concede that no one measure of senescence based upon the properties of individual organisms Definition and Measurement of Senescence 7 can yet claim causal precedence over any other, and we must therefore try our luck elsewhere. The Actuarial Measurement of Senescence The actuarial measurement of senescence is based upon the following presupposition: that although senescence is indeed a diverse and complex process, with manifestations in all the structures and activities of the body, its net total lOO.OOO age in years Fig. 1. effect is of one measurable kind — to increase vulnerability, the likelihood of dying, as life goes on. A measurement of vulnerability is implicit within the information set out in an actuary's life table. A life table may be thought of as a series of figures representing the sur- viving residue, year by year, of a population marked at birth and followed through life until all have died. The table there- fore begins with some arbitrary but large number of indivi- duals (Z^,, say 100,000) and ends at zero when none survive. In the form of a graph, a life table for human beings has the shape illustrated by Fig. 1. 8 P. B. Medawar The curve in Fig. 1 represents the faUing away of the num- ber of survivors (/^) in each year x of age. Its slope is the death rate, i.e. the rate of dechne of the number of survivors; where it is steep, the death rate is high; were it horizontal, the death rate would be zero. The slope is downward and therefore has a negative sign; to make the death rate positive, it may be written 1 1 ^^h death rate= — -r^ ax The death rate itself is almost useless as a measure of the likelihood of dying, for its numerical value clearly depends upon the number of individuals still left to die. (The death rate at 100 is lower than at any previous age, because so few are left in the running.) What is required is the age-specific death rate or force of mortality, the death rate at any age divided by the number still exposed to hazard, i.e. force of mortality = — - — ^, •^ l^ dx the negative sign being used, as before, to give the force of mortality a positive value. In a population of which the mem- bers do not deteriorate with increasing age — e.g. a population of plates or test tubes — the likelihood of being broken is no greater at any one age than at any other; at the end of each age-interval the number of survivors falls to a constant fraction of its value at the beginning of the interval; the force of mortality is therefore constant. The curve of the force of mortality in human beings has the shape illustrated by Fig. 2. It is high at birth, falls steeply to a very low value between the ages of eight and fourteen, and then rises without singu- larity or inflexion for the remainder of life. The question now is: can the force of mortality be used as a measure of the degree of senescence over that period of life in which the force of mortality is increasing? (There is no question of its being used for such a purpose during that- earlier epoch of life in which its value is falling.) Definition and Measurement of Senescence The difficulties and covert implications of such a scheme of measurement may now be commented upon one by one. (1) In the first place, senescence is a process. At their face value, the life table and its derivatives record, not a sequence of events which takes place in the lifetime of a single indivi- dual, but the age-frequency distribution in the population as a whole of a single event in life, viz: death, its end. 'a o E This is not a weighty criticism. It is a well understood convention that, with certain safeguards, the frequency of the occurrence of an event in a population can be taken to represent the probability of its happening to an individual. If these safeguards are observed (not all can be: see below) we can legitimately claim to be measuring that change in an individual's lifetime which makes him progressively more vulnerable to mortal hazards. (2) The use of the force of mortality assumes that all death is accidental in origin (using the word "accidental" in its 10 p. B. Medawar widest sense) and that senescence is the process of becoming ever more accident-prone. In other words, everyone is assumed to die of something, no-one to die what is most inaptly called a "natural" death. Most pathologists beheve this to be true (Aschoff among them), and all are obliged to do so by the laws of certification. The approach to natural death is therefore asymptotic. The lethal threshold falls until, in theory, the most trivial accident imaginable is a cause of death. (3) The measurement is corrupted by the existence of causes of death which are not random in their incidence, i.e. to which the entire population is not equally at risk. The barnacles of Hatton's (1938) life table* may well be equally exposed to hazard, but human beings are obviously not; special ages and occupations have special risks. Evidently the measurement must be corrected for causes of death having any kind of specific frequency of incidence. (4) A life table set up by tracing the life histories of a cohort of newborns from birth to death will certainly be corrupted by secular changes in the causes of mortality. Men aged fifty-four today were born in 1900, when the nature of the causes of death and their pattern of incidence were very different from what they are today. Animals reared in laboratories may not be so affected; nor need it be assumed that secular errors of this kind affect wild populations of animals with comparatively short lives. (5) If the force of mortahty is to be a reasonably efficient measurement of the degree of senescence, there should be no gross inherited differences of susceptibility between members of the population at risk. For let it be supposed that the population with which the life table begins is genetically subdivided into stalwarts and weaklings. Selection will then occur in the course of life, and the population which the later years of the life table deals with will be a quite unrepresenta- tive sample of that with which it began. In the laboratory, *For the literature of life tables for wild animals, see Deevey (1947), AlleC Emerson, Park, Park and Schmidt (1949), and Haldane (1953). Definition and Measuremi:nt of Senescence 11 the use of inbred strains or their F^ liybrids can eliminate this sx)urce of error. None of the five objections so far raised challenges the pririciple of using vulnerability as a measure of senescence, though they make it very evident that grave difficulties stand in the way of its practical use. They are difficulties which apply with particular force to human beings — in this respect, as in so many others, a tiresome and anomalous animal, rightly excluded from the syllabus of studies in most depart- ments of zoology. We may now consider two genuine short- comings of principle, one of which is reparable, the other not. The first is that senescence may reveal itself by a true and radical deterioration that is not accompanied by any increase in the likelihood of dying, viz: a deterioration of reproductive power. Should not any measurement of senescence take reproductive capacity into account? The question is pertinent, because in human females reproduction comes to an end over an epoch of life that is not distinguished by any dependent change in the likelihood of dying. (On the contrary, one cause of death, in childbirth, has been outlived.) It may have less bearing on other animals, in which reproduction, instead of stopping rather suddenly, merely slows down in step w^ith other senescent changes. A more fundamental measure of biological "efficiency" — of an animal's power not merely to maintain itself as a going con- cern, but to perpetuate its kind — should have regard to both mortality and fertility. Several such measures suggest them- selves, of which the most useful might be the total future expectation of offspring per head from a chosen age x onwards, viz: the reproductive value 1 r* — p^b^.dx, where b^ is the age-specific birth rate, and p^ (=lxlto) repre- sents the probability of living to age x (or, in effect, the value of /^ when /<, is taken to be unity). 12 P. B. Med A WAR There is therefore no difficulty in coming to terms with this first objection of principle, if need be. The second is of an entirely diff'erent kind. By the word "senescence" we clearly wish to describe a change which is of innate origin, which is built into the developmental structure of an animal, and which could take place even under the most "favourable" conditions it is possible to envisage. The force of mortality does indeed measure this inborn decline, but the difficulty is that it must also measure another sort of decline, of purely contingent origin, as well. It cannot be strictly true to say, as I said earlier, that plates and test-tubes are subject to a constant force of mortality, even when given equality of use and of exposure to the risk of being broken. They suffer indeed no innate decline (if we disregard such refinements as the slow cry- stallization of glass) but they may deteriorate in a diff'erent way. Plates and test-tubes get cracked and chipped, but may yet be usable, at least in the more frugal homes and laboratories. A cracked test-tube or plate is more vulnerable than one which is still intact; it is more likely to get broken in the course of ordinary use. There is therefore, or may be, an increase of vulnerability which has nothing at all to do with innate decline, but which is due to the persistent, there- fore cumulative, effect of recurrent injury or stress. There are innumerable examples in living organisms of an increase of vulnerability caused by the persistence and, with time, the summation of the effects of recurrent damage, stress, and biological error (cf. Medawar, 1952). Any injury which leaves any physical trace is an actuarial debt, and will add to the likelihood of dying. A single example will iUus- trate the distinction which may be drawn between deteriora- tion of innate and of merely contingent origin: the formation of ectopic flexure lines or creases in the skin, which (if the efforts made to conceal them are anything to go by) are among the most characteristic and reliable signs of the deterioration associated with increasing age. Definition and Measurement of Senescence 13 Ectopic flexure lines form on those ])arts of the body where the skin is habitually creased or folded, particularly on the face. A single creasing of the skin is insuttieient to form an overt flexure line, but presumably even one creasing leaves some trace in the dermal fibrous skeleton, because if the skin is creased sufficiently often, a flexure line will eventually appear. Old people have acquired lined faces for two clearly distinguishable reasons: {a) because an innate deterioration of their skin has made it more susceptible to the engravery of habitual usage, and {b) because they have smiled or frowned or raised their eyebrows more often than their juniors. There can be little doubt that the first factor, the true "innate" deterioration of the skin, by far outweighs the latter, and that it represents the process which most people understand by senile change, but both agencies are at work: the per- sistent and cumulative effects of recurrent stress, and the progressive decline of power of the skin to resist its action. This is a trivial example, but it would be idle to deny that purely contingent factors influence the vulnerability of animals, particularly when the organ or part of the body that is affected is constitutionally incapable of repair. For a higher animal to lose a limb may not in itself be fatal, but if it does so it will certainly be more vulnerable to the hazards which limbs are normally used to escape from. It may be that the actual life-span of wild carnivores is as closely bounded by the life of their teeth as by any other single factor. Teeth may become more fragile with increasing age; but whether they do or not, it is quite certain that one good reason why an old animal should have fewer sound teeth than a young one is because time alone, unconnected with senile decay, has given the older animal more opportunity to break them. In short, risks are not all of an all-or-nothing character; there is a gradient of efi'ect between being killed (which alone qualifies one for admission to the life table) and being left unscathed. An injury which falls short of being mortal does not necessarily leave an animal as well equipped as before to escape a hazard of the same or of another kind. The residue of injuries that 14 P. B. Medawar are not in themselves mortal may build up a debt which is due to age, because it depends upon the passage of time, but which is not due, in the colloquial sense, to "ageing". The two classes of agencies which contribute to increase of vulnerability need not be difficult to distinguish by experi- mental means, but in the actual records of mortality the two are inextricably combined. If w^e wish to use an actuarial measurement of senescence, this is a real shortcoming of principle. Yet this confounding of two causes of increase of vulnerability does not in any way detract from the biological value of the use of the force of mortality as a measure of senescence, for both are real agencies that govern the popu- lation dynamics of real communities of animals. It is up to the physiologist and pathologist to distinguish and keep apart the contribution to vulnerability that is made b}^ a true innate decline. In summary, the actuarial definition of senescence amounts simply to this: that senescence is a change which accompanies ageing and which makes an individual progressively more liable to die. (Such a definition may be challenged on many grounds, but its simplicity should not be one of them.) Senes- cence may be measured by the force of mortality estimated from the age frequency distribution of the incidence of death, but the measurement is valid only of a genetically uniform population subjected to a constant pattern of hazards impingeing randomly upon it, and is imperfect in proportion as these conditions fall short of being fulfilled. Increase of vulnerability is caused partly by an innate deterioration of the body and partly by the accumulation of an actuarial debt as a result of recurrent exposure to injuries that leave a persistent mark. The separation of the latter contribution from the former should depend upon the context in which the phenomenon of senescence is being studied. REFERENCES Allee, W. C, Emerson, A. E., Park, O., Park, T., and Schmidt, K. P. (1949). Principles of Animal Ecology. Philadelphia: W. B. Saun- ders Co. Definition and Measurement of Senescence 15 Deevey, E. S. (1947). Quart. Rev. Biol., 22, 283. Haldane, J. B. S. (1953). .7. hist. Actuaries, 79, 83. Hatton, H. (1938). Ann. Inst. Oceanogr., 17, 241. Lack, D. (1951). Population ecology in birds. Proc. Xth Int. Congr. Omit hoi., p. 409. Medawar, p. B. (1952). An unsolved problem of biology. London: ir. K Lewis & Co., Ltd. SOME REMARKS ON THE PATHOLOGICAL BASIS OF AGEING G. R. Cameron, d.Sc, f.r.c.p., f.r.s. Depar1?nn)f of Morbid .inatomy, University College Hospital Medical School, London. Two years ago, two Swiss doctors, A. L. Vischer and F. C. Roulet (1952) reported the death of two of their countrymen at or near the age of one-hundred-and-two. Both of these old people had led healthy active lives, escaping serious illness and remaining free from any disability up to the end. Death came to the man as the result of gangrene of a limb after influenza; the woman faded out after the death of a daughter, bronchopneumonia being the immediate cause of death. Careful investigation of bodily functions in both towards the end of their lives failed to disclose any serious disturbance. Basal metabolism, recovery from fractures, plasma proteins, haemoglobin index, blood cell counts, blood pressure and pulse were within normal limits. Autopsy on the one-hundred-and-two-year-old male dis- closed generalized arteriosclerosis, especially severe in the abdominal aorta and lower extremities, with thrombosis of both femoral, })opliteal and tibial arteries and of the accom- panying veins, gangrene of the left foot, terminal broncho- pneumonia and chronic pulmonary emphysema. Other findings included a carcinoid tumour in the lower ileum, regeneration nodules in the thyroid, and glandular and muscular increase in the prostate. The old lady, almost one-hundred-and-two years of age when she died, showed at post mortem examination cancer of the breast associated with cystic mastitis but no metastases, severe arteriosclerosis of the aorta, coronary disease with obstruction, calcification of the annulus fibrosis of the heart 16 Pathot.ogical Basts of Ageing 17 and brown atrophy of the heart and h\'er, eholehthiasis and a nodular goitre. To all intents and purposes, therefore, these two old people were healthy right up to the terminal illness, yet their bodies were a collection of morbid processes, any one of which might have flared up at any time during the last thirty years and ended life. No doubt an extended search of the pathological literature would bring to light further reports on extreme old age, but I am not aware of such information, so I set to work to collect data from the autopsy records of University College Hospital which bear on the topic of ageing, and these I now present in the hope that they will interest you. Our records over the years 1904-1953 comprise 13,630 autopsies among which are 1,872 on old people. Examinations w^ere conducted by five pathologists who were, on occasions, assisted by juniors; however, it is safe to claim that autopsy technique has altered very little during the fifty years, for each of us, with one exception, has been trained by his prede- cessor. In many instances microscopy on the most important organs was performed and such reports have been available for study. I am greatly indebted to two of my research colleagues, V. Udall and K. C. Basu Mallik, who sorted out the cases for me. Reports on males from sixty-five years onwards and females from seventy years onwards have been scrutinized. Since there were very many more males than females (1,328 against 544) most of my remarks wdll be confined to the former. Let me say at once that all I have endeavoured to do is to find out what was the pathological condition responsible for the decline and death of the individual. In many instances, of course, the immediate cause of death was some terminal accessory event, such as an acute infection, a severe haemorr- hage, septicaemia or uraemia. It is therefore more correct to speak of morbidity rather than mortality. I must first of all explain that our hospital is a teaching institution, and I gather from conversations with our senior clinicians that the 18 G. R. Cameron policy for many years has been to admit as varied a type of patient as is possible from the population that closely sur- rounds us. Of course, there will have been some selection and no-one would dare claim that our material represents a random sample of the London or English population. We have had surgeons, for instance, who for a time were especially interested in some kinds of cancer for which they were engaged m perfect- ing techniques for treatment. Likewise we have attracted, at one time or another, heart cases or patients referred to a lung or liver specialist. Caution must also be maintained in drawing conclusions from our ageing cases for they have been admitted to our hospital because they were suffering from some fairly obvious complaint. Against this can be placed figures such as those given by Drs. Trevor Howell and A. P. Piggot of St. John's Hospital, Battersea, whose material was still less likely to be selected than ours. These workers also give information collected by Dr. G. Stewart Smith in various municipal hospitals. Suffice it to say that there is reasonable agreement, in general, between the results obtained from the three independent series, although the classification and emphasis often differ. In Tables I and II, I summarize the principal morbid con- ditions encountered during five-yearly intervals. In every instance I have carefully scrutinized the autopsy records and endeavoured to decide what condition dominated the com- plex pathological pattern at death. In practically every case I have agreed with the considered opinion of the pathologist responsible for autopsy. You will see from the tables that 70-80 per cent of deaths are included under the diagnosis of cancer, arteriosclerosis, urinary disease (especially senile enlargement of the prostate and its complications), primary lobar or bronchopneumonia and chronic peptic ulceration of the stomach or duodenum. In both sexes cancer and arterio- sclerosis make up the majority of morbid conditions. Table III groups together the dominant conditions in the total groups. It turns out from these latter figures that 40-3 per cent of males die from cancer and 20-8 per cent die from Pathological Basis of Ageing 19 arteriosclerosis; 31 per cent of females die from cancer and 27- 1 })er cent from arteriosclerosis. Let me repeat once ai^ain that these pathological states, in my ()i)inion, were the cause Table I The Commoner Morbid Conditions Found at Autopsy in 1,328 Mai,es FROM Sixty-Five Years onwards. Percentage and total cases. Morbid Condition .ii>e Groii]>.s 65-70 71 75 76-80 81-85 8«i- Cancer Arteriosclerosis . 43-8% 17-9 41-0% 19 0 36-6% 26-4 31-6% 27 0 6-8% 48-3 Urinary disease Gastric and duodenal 0-7 90 8-4 00 6-8 ulceration 4-7 41 30 1-3 3-4 Pneumonia 3-7 3-3 7-4 15-8 13-6 Total cases in groups . 655 3()fi 202 76 29 Table II The Commoner Morbid Conditions Found at Autopsy in 544 Females FROM Seventy Years onwards. Percentages in total cases. Morbid Condition A^e Croups 70-75 76-80 81-85 86 Cancer Arteriosclerosis . Pneumonia 39% 19 5 26% 31 9 18% 41 15 21% 47 12 Total cases in groups . 275 153 73 43 of decline and often of death of the subjects but that fre- quently enough an acute complication was the immediate cause of death. But some of these figures are misleading, for all who have examined the organs of aged persons must have been struck 20 G. R. Cameron by the frequent association of a number of morbid conditions. In an attempt to assess this association 1 liave re-examined, with care, reports on 100 ageing male subjects from sixty-five years upwards suffering from cancer and compared them with autopsy reports on 100 subjects dying from causes other than cancer. All of these autopsies were performed by myself at University College Hospital during the years 1935-39 and Table III The Commoner Morbid Conditions Found in Ageing Males (Sixty-Five Years Onwards) and Females (Seventy Years Onwards). Total numbers. Morbid Conditions Mcdes Fenialt'.s Totid Tumours, mostly cancers Arteriosclerosis Dif^estive disease Respiratory ,, Urinary ,, Nervous ,, 555 277 162 116 108 13 178 149 74 ()2 14 4 788 426 286 178 122 17 Total cases 1281 481 1712 Total number of subjccls iiivostig.ited: 1872. 1946 and 1947. Table IV gives the results. Without attempt- ing to draw any quantitative comparisons I feel sure you will agree that the cancer cases show a very high incidence of coronary, aortic and renal arteriosclerosis, in other words, of generalized arterial disease. Similarly, the non-cancerous ageing subjects are very frequently affected with generalized arteriosclerosis. I believe that it is safe to conclude that cancer and arteriosclerosis are common morbid conditions in ageing persons and that they very frequently go together. I have likewise looked into cases of chronic urinary disease, peptic ulceration and bronchopneumonia and I have found with these, too, that severe arteriosclerosis is very common. Since it is much more difficult to collect adequate numbers for analysis than with cancer, I have not attempted to present Pathological Basis of Ageing 21 figures, but I can assure you that my conclusion seems definite enougli. Finally, I give you in Tables V and VI some details about the common types of cancer on the one hand and the distri- Table IV Aktkuiosclerosis in Cancer and Non-Cancer Male Cases. 100 Cancer Cases 100 Cases free from Cancer Age range 05-84 05-85 Heart weight (g.) range 210-570 210-935 Coronary disease Normal .... 8 3 Slight 42 32 Moderate .... 25 27 Severe ..... 25 38 Complete obstruetion 17 20 Aortic Normal .... 3 0 Slight 28 28 Moderate .... 35 38 Severe ..... 34 34 Renal Normal .... 24 11 Slight Moderate .... 32 32 45 32 Severe ..... 8 5 Pyelonephritis 4 7 Cerebral Normal . . ' . 35 33 Slight Moderate .... 10 5 14 16 Severe ..... 5 12 Not examined 45 25 bution of arteriosclerosis on the other. You will notice how certain sites are often picked out by cancer, especially stomach, oesophagus, rectum and lung. I might say that the list of possible sites is a long one, and the rarer examples are 22 G. R. Cameron not included. So, too, with arteriosclerosis there is a pre- dominance of certain types, especially coronary and the generalized type with its associated congestive cardiac failure. Table V Types of Cancer at Different Ages in Males. Tupc of Cniicer Age Groups 65-70 71-75 76-80 81-85 86- Stomach Lunjr Rectum ffisopliagus Colon Bladder Pharynx Larynx Prostate 49 38 25 35 22 14 13 10 10 24 27 17 17 8 4 3 5 11 15 4 5 7 12 3 0 () (i 3 4 3 2 0 2 1 1 2 1 0 0 0 1 0 0 0 0 Total cases 287 150 74 22 2 Table VI Types of Arteriosclerosis at Different Ages in Males. Total numbers Types 65-70 71-75 76-80 81-85 86- Coronary disease Generalized AS with CCF Cerebral thrombosis Cerebral haimorrha('t may not be generally valid. However, in Lchistcs, at any rate in tlie laboratory, it seems very much as if the form of the life-table is going to be very nearly the same as in a mammal. As regards gro\\i:h-pattern, the male of Lebistes ceases to grow at a relatively early age, whereas the female goes on growing like the female plaice, we think probably throughout life; the measurements are not yet complete. Yet the sex differential in longevity is not a very great one. I agree very much w ith Dr. Lansing, that the work of Dr. McCay is the hub of the whole thing and the point from which we should open our investigation. Is it the delay of maturation in the retarded rat which makes it live longer or is it the prolongation of gro\\i:h? In other words, are we to consider that gTO\Hh per se has some medicinal virtue in pro- longing the life of the animal and a\ arding off senescence, or is it that the animal senesces when a certain sequence of operations has been fed into it? I have interpreted Dr. McCay's work — Fd very much like to hear what he has to say about this — as simply slowing down the perforated tape you're feeding to the calculating machine, and senescence as taking place when the tape is exhausted; if you play the tape slowly then the whole cycle is stretched out in time. If it's reall}^ a matter of gro^^i:h rather than of sequence of operations, it should be possible, in the light of some work Moon* and his colleagues have recently published on giving gro\\i:h hormone to mice, to maintain some measure of growth in adults — adult rodents, at any rate — and see whether that does have any detectable effect on the life-table. I would like very much to know, tirst whether anybody has any data whatsoever regarding the senescence or non-senescence of animals other than birds and mammals, and in the second place I'd like to hear w^hether Dr. McCay has anj^thing to tell us now about his more recent work which would throw light on how far gro\\'th and how far differentiation and development are the essential components in determining the age of senescence in mammals like the rat. McCay: I can't answer the basic problem at all, though we have worked with other species, with insects and with trout. We did about fifteen years' work with trout of various species and it seems to hold there. But we failed with dogs; experimental methods are so extremely difficult with dogs. The dog seems to seal its epiphyses at about ten or twelve months, unlike rodents, and the retarded dog is much more subject to parasites than the normal dog. Comfort: I wonder if I might ask you further whether you have any evidence on another point. You delayed your rats and kept them at a low rate of growth; you then allowed them to grow up and go through their life-cycle. Could that be done in steps? Could they be allowed to grow part of the way and then, when they had become sexually mature, *Moon, H. D., Simpson, M. K., Li, C. H. and Evans, H. i\I., 19.52, Cancer lies., 12, 448. 30 General Discussion retarded again and kept in a state of suspended development, or was it all or none? McCay: I've never let them first beeome sexually mature and then stopped them. You'd ereate another variable there, the variable of body-weight and adipose tissue. We have taken rats half through life and then foreed them to become very thin, and again, "the thin rat goes to the funeral of the fat rat". Comfort: That was a rather separate effect, wasn't it? McCay: Yes, that's a separate effect that you run into as soon as you try that type of experiment beyond sexual maturity. Comfort: I have a medical bias here, because while it would be nice to keep patients in an arrested state as children, as you kept your rats children, the real medical problem is to prolong the period of adult vigour. And although, I suppose, restriction of diet is beneficial, you have no evidence so far that one can bring about the same slowing down of the record in adult rats that one can in young ones. Olbrich: Dr. McCay, did you measure any type of function, say liver function, in your retarded rats for a prolonged period and correlate it with the time? McCay: No, I never did. In our hands rats are difficult for functional measurements and we'd always hoped we could get dogs where M^e could make functional measurements. Olbrich: Have you any impression that range of function is correlated to time changes in a mathematical fashion in your retarded rats? McCay: No, I couldn't answer that. Olbrich: They don't get any infection? McCay: No. We have no trouble with infection — that's a virtue of the rat, and we hope Prof. Verzar's functional tests continue to develop and we'll apply them to the retarded rats. Krohn: Have you any notion about what is happening that prevents actual maturation of the reproductive tract? Is there any particular reason why the gonadotrophins should fail to be produced in your own animals? McCay: No. We've been trying to find out how late in life the rat can reproduce. Asdell made initial studies on the ccstrus cycle a long while ago, but now we are actually trying to breed them very late in life after they've long been retarded. We are using those same rats for certain types of psychological testing at present. We're also hoping to use parabiosis, and suture a retarded rat to a rat that's growing old. Lansing: There have been a number of studies in lower organisms which show that retardation through dietary control is effective only prior to maturation during the period of gTOwth, in the Cladocera and so on; that if the dietary restriction is controlled after maturation there is no beneficial effect, in fact in some studies there was actually a reduction in total life-span. The extension of life-span is not directly proportional to the period of extension of the growth period, in fact there is a very great disproportion. One or two days' extension of the growth period will result in a 10- or 20-fold increase in life-span. General Discussion 31 Franklin: I should like to suggest for consideration later that a definition of senescence is a decrease with time in tlie pliysiological reserves of tissue and in the physiological capacity to increase the number of cells. MENTAL ASPECTS OF AGEING Aubrey Lewis, m.d., f.r.c.p. Institute of Psychiatry, Maudsley Hospital, London. Ageing is a stage in the total process of growth and decay, and it might be expected that its mental aspects, like those of childhood, would cover most of psychology, since age is a variable that can hardly ever be left out of account in any psychological inquiry. But there is no such extensive literature about the latter end of life as there is about the earlier. For this there are obvious sentimental, social and technical reasons. In spite of the relatively meagre data it is not easy to survey the present stage of knowledge, because it reflects changing theories and methods in regard to such fundamental matters as learning and personality. Any general question as to the essential nature of ageing does not admit of a psychological answer: the question is a biological one and concerns the whole organism, rather than only its psychological aspect. I shall therefore limit myself to considering some questions about the manner in which we age mentally. Chiefly these are: Is there a differential rate and degree of decline in psychological abilities, which may permit reorganisation and even gain just as it may entail disorganisation and decay? What are the characteristics which distinguish healthy ageing from pathological ageing? Are they differences only of degree? Are some mental changes of age compensatory and adaptive? What is the physical substrate of mental ageing? How far do social and cultural influences affect the ageing individual, either by hastening and aggravating his handicaps or by allowing readjustments and opportunities that suit him? Before seeing what can be said in reply to these questions I should like to look briefly at a concept which crops uj) almost as 32 Mental Aspects of AcxEIng 33 often as adaptation or stress. This is the concept of regression. It would be more correct to say the concepts, for behind this word there are evidently several meaninf^s when it is applied to ageing. The simplest is that originally given it by Ribot when he introduced the term to denote the retention in memory of the earliest perception and experiences, whereas other memories are lost in the reverse order of their acquisi- tion: this is based on crude, and on the whole erroneous, observations, but is at any rate a defined and appropriate use of the word. Another, commoner meaning refers to the way in which, when a man is confronted with "harsh experi- ences, difficult or impossible to adjust to, his personality reverts or regresses to earlier levels of integration": it has the disadvantage of itself resting upon some vague conceptions and not indicating how the regressive process affects different features of the personality nor how far back it pushes the personality. A third meaning of the term, by no means open to these objections, is perhaps open to others of a more theoretical kind: I mean the psychoanalytical use of the term, to denote the redirection of libidinal energy towards infantile fantasies. This is illustrated by Karl Abraham's explanation of why so many senescent people pay excessive attention to their bowels: "the more the genital zone retires into the background as a source of pleasure, the more many individuals turn back to oral and anal erotism" as in infancy. Yet another meaning of regression is that which Kurt Lewin gave it; he regarded various sorts of disorganisation and reduction of the field of activity as regressive, pointing out that the patterns of behaviour observable in senility or imprisonment "strongly indicate that the different aspects of regression are to a certain degree independent of each other". The term has yet other meanings in this context: Spielmeyer used it, for instance, to describe the arterial changes in senile brains; and of course in German psychiatry late middle life is called the period of regression (Riickbildung). A word so confusingly various in its meaning and application is suspect in psychology: there are too many ambiguous words and concepts as it is. 34 Aubrey Lewis But it may be said that beneath this variety there is one com- mon notion, and an important one — that the ageing man reverts to the psychological ways of his earlier life. This is a very ancient and familiar idea: Aristophanes wrote that old men are boys again, and no doubt it goes much further back than that. But the "second childhood" view of old age derives too obviously from some superficial features of senile dementia. Old men, it is true, may become self-centred, greedy, boastful, prone to tantrums, sulky and in other ways "childish", as we say: but I see no more reason for equating this behaviour with the corresponding behaviour of a young child than for equating an old man's tottering gait with a youngster's toddle, or an old man's urinary incontinence with a baby's bedwetting. It is a purely descriptive similarity, whereas the concept of "regression" implies essential identity. Unlike Hughlings Jackson's principle of dissolution of function, to which it has a deceptive affinity, it neither explains nor describes. The psychoanalytical use of the term, however, cannot be criticised on these grounds. Granted the psychoanalytical theory, regression does in part describe the psychopathology of many of the emotional changes and morbid beliefs of the senescent or senile patient. It is almost entirely, however, a matter of inference and analogy: very few elderly people have had an actual psychoanalysis. And it fails us, as do so many concepts based on study of the abnormal, if we wish to ac- count for the constructive readjustments which often appear in later life, in spite of reduced capacity in some respects. The first general question, and the most difficult, is to dis- tinguish the intrinsic, or normal psychological accompani- ments of chronological ageing from those that are morbid. The problems in this are not different from those we meet when we try to distinguish between physiological and patho- logical ageing: but I think they are incomparably more difficult to solve. Apart from the common troub le in defining any essential distinction between normal and morbid, we have very little ground for correlating mental with structural changes; we cannot isolate mental functions readily, and Mentat. Aspects of AcxEINO .35 many of the most important wc cannot measure; study of animals takes us only a little way; and we cannot control some variables wliieli are probably powerful influences upon the rate and signs of mental ageing in human beings. Conse- quently we are deprived of many of the methods which are available for separating pathological from normal in the somatic features of old age. It might be said that these difficulties can evidently be overcome, since they have been overcome at the other end of life. To some extent that is true. We know a great deal about mental changes in children as they develop, but in them also the difficulty of telling what is pathological may arise in matters of emotion and behaviour, as the experience of every child guidance clinic attests. In any case ageing is spread over a longer period than childhood is, its timing and form are probably less ordained in advance by heredity, and the strong environmental influences that may play upon it are harder to control for purposes of investigation. Hence the greater obstacles in studying its normal character. If we could relate the normal mental characteristics of old age to structural changes in the ageing brain, we might be better off. This we cannot do. Such changes occur, of course, but have been very little studied in the normal. The only thorough examination of cerebral changes in normal old age, uncomplicated by dementia, is Gellerstedt's, which was carried out on 50 brains. His careful findings lead to a dead end, so far as the relating of cellular to psychological phen- omena goes. Thus, speaking of senile plaques, he says, "Manchen andern Autoren beistimmend, konnen wir als unsere Auffassung aussprechen dass diese Gebilde, ob nicht oder ob in grossen Mengen vorhanden, durchaus keine siehere Aussage iiber den vorhergehenden psychischen Zustand des Patienten erlauben". Similarly Oskar and Cecile Vogt in 1941 compared the minute structure of the thalamus in the brains of four distinguished old scholars who had preserved their intellectual faculties till their death with those of men who had died in senile dementia: they concluded that the 36 Aubrey Lewis involutional process in the thalamus (consisting chiefly of nerve-cell degeneration and death in tlie centro-median nucleus) is the same in normal old men witli unimpaired intellect as in senile dements. This does not preclude a, relation between the normal mental deterioration of old age and the plaques, neurofibrillary thickening and other changes seen in ageing brains, but it makes it dubious, and certainly does not help in discriminating normal from pathological mental changes. It has similarly been impossible so far to link the kind and degree of involutional mental changes to either cerebral arterio-venous oxygen difference or to the electro-encephelo- gram or to endocrine findings. Pincus and his colleagues at Worcester have given us data about limited stress responses which show no falling-ofP with age in healthy men, and I hope Dr. Freeman will further illuminate this matter in his paper. Up to now, however, purely psychological methods of assessing what is normal have been unchecked by reference to the material basis or accompaniments of the functions in question. Most of the detailed information about normal psychological ageing has regard to cognition. Intelligence, perception, memory, mental efficiency and speed have been measured, and the results indicate a decline, it is true, in intellectual powers but one that is not uniform, nor certain, nor predict- able. On the analogy of physical functions this is what one would expect: allometric growth, and allometric decline: in some ripe old age, and in others premature senility: no tokens of rapid decay to come, no guarantee against it. There is, however, very little exact information about how any individual ages mentally: the data we have are derived from analysing the scores obtained by sample populations contain- ing more or less adequate numbers of people of different age groups. This method, which has been so fruitful in determin- ing the mental growth of children, has led to rather doubtful conclusions when apphed to ageing adults. Samples have been small or biased: and it has not been safe to assume that Mental Aspects of Ageing 37 the measuring' instrument was measuring the same abilities at different ages. When does dechne begin? Earher work seemed to show that the average scores obtained on intelhgence tests dimin- ished at an increasing rate from the third decade onwards. But this is now questionable. Patrick Slater, on a carefully selected and stratified sample of 2,500 men aged 18 to 42, found no evidence of an increased rate of decline from thirty onwards. Vincent, more recently, examined the scores of 7,297 subjects on a verbal intelligence test and found that between the ages of twenty and sixty the mean score declined steadily, showing a linear regression. The annual decline of the mean score was found to be likewise constant when other tests had been used: it was between -026 and -030 of the standard deviation of the test. Such a finding, though important, is less valuable than one which would establish whether the annual decline in an individual is related to the maximal level attained, say, at sixteen or twenty, or to the rate of increment during the earlier years of life. As far as I know there is no evidence at all on these points. A score on an intelligence test is a crude summation. If, as seems to be the case, the decline of some abilities is more rapid than that of others, the decrement will be small on some subtests and large on others. The abilities least affected have been those concerned with vocabulary and information, while most rapid decline was evident in the ability to deduce relations and adapt to new situations. A mean total score would conceal these differences, as well as the differences between individuals. The long dispute about a general factor and special factors is relevant to this. Balinsky, using Thurstone's method of factor analysis and intercorrelating scores on subtests of the Wechsler Bellevue Scale at different ages, concluded that mental organisation changes over a span of years. From childhood onwards there is increasing special- isation, and then later the various special abilities are inte- grated into a flexible whole, but with subsequent changes in the relative extent to which special abilities are utilized. No 38 Aubrey Lewis matter how the factors are named or interpreted, it is evident from Bahnsky's data that the same tests measure different abihties in the adult at different ages. Birren hkewise by factor analysis of data obtained from the Wechsler Bellevue Scale administered to subjects between sixty and seventy years of age, found that four factors could be extracted from the intercorrelations: they were — verbal comprehension, closure (non-verbal organization of visual perceptual material), memory, and perhaps induction. The greater scores usually attained on tests requiring verbal comprehension depends mainly on the store of verbal in- formation retained by ageing persons, while ability to acquire fresh information and face fresh situations diminishes. "It seems reasonable that the age of optimum learning and the age of maximum stored information are not coincident, and do not correspond to the age of maximum scores on our intelligence tests". It follows that a scale like the Wechsler Bellevue needs to be docked of some items and amplified with some others if it is to cover the mental abilities of ageing people and permit comparison of one elderly person with another or with a standard population of the same age. It seems clear from the many studies of decline in intelli- gence in ageing people that a wide range of tests and a series of longitudinal studies of the same individuals are essential if we are to make progress. A wide range of tests is called for to cover the primary mental abilities as fully as possible: and longitudinal studies are needed to allow for the very wide individual differences in optimal performance level and in performance decrement as age advances — individual differ- ences which are submerged and undetectable when the mean scores of large groups of people of different ages are compared. This last difficulty is a most serious one. The great success of the familiar methods of assessing level of intelligence in children of different ages has caused us to concentrate more on performance relative to others of the same age than on rate and kind of change in the individual. So long as we are ignorant of the performance level of an individual in his Mental Aspects of Ageing 39 })rinie, we cannot tell whether a low score on one or many tests betokens a neoligible decrement on an originally low achievement or, on the other hand, a gross falling off from a previously high attainment. It is true that we attempt to distinguish between these by devices which assume that some abilities (especially vocabulary) are almost wholly retained while others decline; we therefore detect and measure the dechne in terms of this disparity. But such devices are, in practice, weak: they hardly do justice to the individual differ- ences in constancy or otherwise of vocabulary and information as people grow older, and they cannot tell us at what rate decline has occurred, or will occur in the future. Moreover, the decline must be fairly gross to be detected with confidence. It has been urged that the degree of decline in any indivi- dual is inversely related to the optimal level of intelligence: the more stupid he is, the faster he will become stupider still as he grows old: if he is bright he will remain bright. These findings bespeak an academic bias. It is true that in 45 University professors, aged sixty to eighty, exacting tests showed no unambiguous evidence of decline: but perhaps 45 manual labourers aged sixty to eighty who had lived satisfy- ing and healthy lives would also show no unambiguous evi- dence of decline from their more modest level of intelligence. Moreover in old people so many allowances must be made for sensory and motor defects and unfamiliarity with test procedures that assertions about the more rapid decline of intelligence in the less intelligent must be scrutinized very critically. On the question of deterioration in the highly intelligent, however, we can profit by the painstaking labours of Harvey Lehman. Limiting himself to the relationship between chronological age and outstanding achievement in science, medicine, philosophy, music, art and literature, he found that in all of these the maximum production of first-rate work had occurred before the age of forty, but that there are many instances of exceptional powers retained into the eighth decade of life. Verdi, for example, composed Otello at seventy-three 40 Aubrey Lewis and Falstaff at seventy-nine: Humboldt wrote the five volumes of his Kosmos between the ages of seventy-six and eighty-nine: Goethe composed the second part of Faust between seventy and eighty: Galileo at seventy-four, Euler at seventy-two, Laplace at seventy-five were still making valuable original contributions; and there are many others. It was, however, in the continuation of work which had occupied their youth or their prime that these old men were engaged, and as Lehman points out, they chiefly assembled knowledge or completed an edifice they had been building all their lives. But although qualities of the highest order, such as are essential for great scientific discoveries and the creation of literary and musical works of genius, decline with advancing years, many of the requisites for distinguished work have clearly been retained. It is not yet possible to define or to measure these qualities, to which we give loose names such as mental energy, or determination. But, as Terman has suggested, it is desirable that as full medical and psychological observations as possible should be made of "contrasting groups of creative persons, including (a) individuals whose productivity declined early and rapidly and (b) others who have remained productive to a much later age. Such con- trasting groups can be found in the faculty of any large University. An important feature of the clinical approach would be the study of motivation in the individual and how these are affected both by specific influences of the immediate environment and by the large social and cultural trends characteristic of the given time and place". The effort of the Guilford group at the University of Southern California to develop tests of "creative thinking" may permit psychometric estimates more relevant to the contrast than existing intelli- gence tests. Other measurements have been made, notably of memory, but these have not added much to the common observation that memory declines with advancing years, that individuals differ widely in this respect, and that in the same individual the amount of decline varies with the sort of material learned Mental Aspects of Ageing 41 and the time when it was learnt. We are still some way from being able to measure effective memory; it cannot be taken for granted that "digit span", "sentence repetition", "retention of Turkish-Enghsh vocabulary" and other such tests, or the recognition of pictures, give a fair estimate of anyone's capacity to retain and reproduce what he has perceived, or tried to learn. Moreover performance on most memory tests is influenced by intelligence and habit patterns: a comparison of an elderly group with a youthful group can therefore be fallacious. Such groups have usually been matched on vocabulary score or socio-economic status, neither of which is a wholly safe criterion. Even so the more intelligent of the elderly subjects do much better than the rest, and those who have least need to organize their habits in order to learn an intellectual or a motor task show least deficit. The learning process, from initial perception to final performance, is now being examined with such energy, in relation to a complex theory, that retention in the human subject at different ages will perhaps soon be elucidated, but for the present we cannot dissect the disability that the old have in learning unfamiliar tasks. Possibly this is to make a necessity out of a virtue. Sir Frederic Bartlett wrote not so long ago that it is insufficient to collect, analyse and study measures of simple bodily or mental functions carried out in artificial isolation from all others; what is needed is to collect, analyse and study measures of skill. This of course is what has been done by Welford and his colleagues at Cambridge. As Sir Frederic will be speaking later on about this, and about remembering, I would only say now that the Cambridge experiments have demonstrated the trouble which the elderly have in organising new incoming data and perhaps also out- going action. Whether this accounts in part for the greater slowness of the aged in performing many intellectual and motor operations, and for their relative or total failure in speeded tasks, is still doubtful. Compensatory tendencies may enable a good level of achievement to be maintained in spite of impaired capabilities, so long as the method and 42 Aubrey Lewis timing of the task are under the ageing person's control. Many studies have shown, however, that the steady process of slowing in simple reactions and choice-reactions probably begins in the late teens or early twenties. Personality and awareness of defect may be as important as degree of defect. When an elderly man uses a rather different method from that of a younger man to carry out a particular task, because he can no longer command the capacities requisite for the earlier method, he may not be aware that he is doing so, or, much less often, his new method may be the outcome of reflection and experiment. Personality factors will affect his attitude to disabilities and his success in coping with them. But on personality changes in the elderly we have little systematic information. Most of it comes from cross- sectional studies which cannot tell us whether a trait or con- stellation of traits is prominent in the elderly only when the same traits have been evident in them when they were young. It is, for example, commonly held that elderly people become more obstinate, more self-centred, more rigid and conservative but the opposite qualities are also observed in many — undue pliancy, gossiping curiosity, vacillation and uncritical accept- ance: these seem less to be the characteristics of age than underlined features of individual personality. At present many of the characteristics of old age, as ordinarily under- stood, are attributed to the interplay between the individual and his social environment which moulds his personality as it influences his activities and interests, but Kallmann's reports of senescent one-egg twins who have lived their lives far apart suggest that too much weight may be given to social influences in determining personality changes in old age. Rorschach test results have been interpreted as showing in the aged reduced responsiveness to emotional stimuli, inability to use inner resources, difficulty in forming satis- factory social relationships and lessened instinctual control; but these findings are b}^ no means regularly encountered. Daniel Schuster's old gentleman of one hundred and six vividly illustrates this. He provided ''a rich and productive Mental Aspects of Ageing 43 (Rorschach) record whose varied content suggests an indivi- dual with a multiphcity of interests. He is aggressive with a strong drive for achievement, a drive that is accompanied by good creative abihties . . . Emotionally he presents a picture of a basically introverted individual who in emotionally stimulating situations is capable of responding in a warm mature manner ... he depends upon his ow^n resources . . . Though he may almost appear extra verted in his social con- tacts, (he) actually maintains these on a superficial level for the most part, with the exception of a few^ carefully selected w^arm associations". The interpretation of a Rorschach record is ahvays open to question, but in this instance a man who was very old indeed failed to show the changes supposed to be characteristic of the elderly. Other means of inquiry into the personality of the ageing have been interviews, attitude questionnaires, and inven- tories. For the most part the populations studied were small and diversely composed. The Chicago inquiry, however, showed in a large sample of elderly people a lessening of social contacts, satisfactions, plans and zest. The limitations of large-scale inquiries about personality are great. They are necessarily, as a rule, concerned with self-reported behaviour and with subjective accounts of happiness, worries and pre- occupations; they tell more about what a particular society imposes upon elderly people of a particular socio-economic class, and what effect infirmities can have upon conduct, than about lasting changes of personality. There are those who hold with William James that by the age of thirty the character has "set like plaster and will never soften again". This arbitrary opinion implies that adult personality has a recognisable structure with unchanging components. What these components are, and how they can be detected and their mode of organisation classified, is still controversial. It is, however, generally agreed that when certain attitudes and interests are steadily manifested in a consistent way and occur often enough with certain other attitudes, we can infer — and by appropriate methods derive — 44 Aubrey Lewis a siniile more ruiulanieiital factor which manilcsts itsclt' in them; similarly we measure traits such as perseveration or persistence, which represent an abstraction from several forms of conduct or test response that evidently have a common factor. To determine whether personality changes with advancing- a^e, and how it does so, the surface manifestations of attitude and interest have to be distinouished from the underlyino common factors which make up the basic per- sonality. As far as I know systematic studies of age changes have been almost entirely concerned with the former: no one has applied to the findings the methods of statistical analysis which would reveal whether there are changes in the common underlying characteristics or in the way in which they are or- ganized. It is perhaps a vice in this predominantly statistical approach to the problems of personality that it treats the cross-section of attitudes and test results at a given moment as a constitutional attribute almost immune from time and change: yet there is no inherent reason for this except the great practical dilliculty of large-scale retesting of the same I)opuIation, or of another population strictly comparable in all relevant resj)ccts except age. Many of our assumptions about normal personality changes in older people are influenced by the features which mental disorder wears in them. Psychiatrists in particular often allow their experience of paranoid and hypochondriacal trends in senile psychoses to bias their opinion about the normal changes of personality in old age. This is understand- able, since most psychiatrists accept the postulate that mental ilhiesses not accompanied by constant demonstrable tissue changes — the "functional" psychoses — express characteristics of })ersonahty writ large. But do the mental disorders of the elderly in fact exhibit features peculiar to this stage of life? It is in "involutional melanchoha" that this (jucstion has been most obviously disputed, backwards and forwards, for the last lifty years. The characteristics said to be typical of these dei)ressi()ns oi' later life are: hysterical symptoms; hypo- chondriacal and paranoid ideas; extreme fear and agitation; Mkntai. Aspects of Ageing 45 lialliicinations; and catatonic features. ThoujL^h common, tlicse sym})toms cannot be regarded as specific. None of them, and no combination of them, is restricted to the elderly: there is no clinical ])ictiire supposedly typical of involutional psychosis that I have not seen in the same form in men and women in their twenties — though it will then be given a different diagnostic label. Why agitation, hypochondriacal delusions (especially about the bowels) and paranoid distortion should be so much oftener seen in later life is unclear. Some would relate them to the painful experiences, generating fear and insecurity, wdiich arise out of waning powers, lessened status, and narrowing field of activity as age advances. Others have put forward psychoanalytical concepts of enhanced narcissism, introjection of ambivalently loved and hated objects which are then identified with over-libidinized viscera, and frustration of sexual desires leading to the revival and entrance into consciousness of infantile fantasies; but as these explanations are equally applicable to similar phenomena in younger people, they do not account for the frequency of the constellation of symptoms in the elderly. Briefly, then, mental illness in middle-aged and elderly people can occur in every form that occurs at other phases of adult life, but is more often characterized by somatic delu- sions, terror and distrust. If the illness is less severe, these characteristics may still be there, but may seem no more than harmless neurotic symptoms. They are then taken to be accentuated traits expected in the elderly — "senes morosi, philauti, deliri, suspiciosi". That these traits — suspicion, anxious self-concern, agitation — are in fact typical of old age is no more true than that their opposites are — credulity, self- confidence, apathy. The mental abnormalities of late middle life, apart from dementia, are the late realisation of consti- tutional tendencies, though the genetical factors seem com- plicated, and not specific to senile or involutional illness. The healthier the personality in early life, the less chance of failure to adjust to the changed circumstances and capacitise of advancing age. 40 AiTEREY Lewis When prooTcssivc dementia and focal symptoms, especially apraxia, apliasia, agnosia, point to a cerebral degenerative jmx'ess of the presenile Pick-x41zheimer type, and this is conlirmed post-mortem, it is tempting to attribute all the essential symptoms to cellular death in specified parts of the brain; yet the cellular changes found in Alzheimer's Disease have been found also in the brains of undemented old people. The same can be said of senile dementia. The most obvious explanation is that the cellular damage essentially respon- sible for the dementing process is not disclosed by the histo- logical methods of examination usually employed on the brain. Alternatively, the brain — or the whole organism — may be credited with such great powers of compensating for defect due to local damage that mental abilities may seem intact in one person while in another with a similarly damaged brain dementia would be evident. Each of these views is open to objection; they are not, of course, mutually exclusive and may well be complementary. The neuropathologists will no doubt sooner or later tackle the histological problem inherent in the first: as for the second, psychologists have already leant heavily on the concept of compensatory adj ustment in explaining individual differences in degree and rate of ageing. This concept — compensatory adjustment — is still imprecise, but indispensable. The falling off in certain abilities; the anxieties about one's material needs, health and survival; the necessity for finding new social roles and preserving social contacts, companionship and affection — these are stresses which can only be met by adaptation calling for fuller use of unimpaired abilities and opportunities to make up for those that are lost. No doubt the ability to compensate needs much analysis. It is hardly likely that the same ability is responsible for the reorganisation of effector processes in a skilled task and for the activation of an unsuspected potenti- ality whereby social misfortunes or bodily decay can be coped with. If the driving force is the need to maintain, in spite of declining or defective ability, a standard of attainment, it is Mentat- Aspects oe Ageing 47 evident that this has no single source, but de})ends on })sycho- logical constitution, and social and cultural influences. Moreover it is not indifferently eni^aned in conipensatin R, I and R constants. In a whole population an exponential decline of survivals but no real ageing A\'ill occur: X = Ae -kt (or kt = In A — In x) e = basis of natural logarithms t = time x = number of survivals at time t A = number of survivals at time O k = rate of destruction There is ageing when: (c) I > R, I constant, R declining (e.g. following an exponential course) X = Ae -kte"* (equation of Puetter, 1921) a = ageing factor (rate of ageing). k = destructive factor (rate of destruction) General Discussion 57 The equation of Puetter corresponds to the known sigmoid survival rate in man. Tliere are two different ageing processes: one shoAvs a decline from birth to death, the other is characterized by an up and down of develop- ment (youth, maturity and senescence). There is no strict parallelism between the changes of the different organs of the organism, nor their parts. This is true for physical and mental faculties. Life shows different levels of vitality (tonus) which are characterized by the ease with which an individual holds its ground physically and mentally. The higher the level of vitality, the better the organism is capable of withstanding injuries and even over-strain. Not mere longevity is the aim, but the maintaining of a level of vitality as high as possible even in the declining phase of life. Verzdr: I wonder whether Dr. Miescher would be willing to change his first sentence so that the last word is not "vitality" but "adaptab lity". I could agree to the definition then. Miescher: I think that the concept of vitality is the broader one, embracing adaptability. Besides, vitality has two aspects, a physical and a psychological one. Verzdr: I think adaptability is the elasticity of tissues or the com- pensatory hypertrophy of organs, or might be the adaptability to different physical changes, atmospheric and other conditions, and even to new psychological situations. Miescher: In any case different levels of vitality or adaptability have to be considered. Tunbridge: But you are not wanting the man of sixty to be the man of twenty. You are accepting the fact that there might be standards of decline? There is not very much difference between you: Dr. Miescher is saying he wants 100 per cent vitality of the standard for the age, which I think Prof. Verzar would accept as adaptability to the highest function for that age. Miescher: In fact it is simply a question of definition. ^EFFECTS OF AGEING ON RESPIRATORY FUNCTION IN MAN D. V. Bates, m.d., m.r.c.p., AND R.V. Christie, m.d., d.Sc, f.r.c.p., Dunn Laboratories, St. Bartholomew's Hospital, London. I HAVE been asked to discuss the effects of ageing on respiratory function. What I propose to do is to describe briefly some of the tests which are now used to assess res- piratory function and to contrast the performance of young individuals w ith the performance of those who have reached middle age. I also propose to bring the disease emphysema into the picture because as far as respiratory function is concerned emphysema is a gross exaggeration of w hat happens to the lungs with advancing years. Fig. 1 shows the age distribution of the various groups investigated. The sole reason for choosing thirty-five years as the dividing line between young and old was that the primary purpose of this investigation was to study the changes which occur in emphysema, a disease which is rarely found before the age of thirty-five. The vital capacity is clearly reduced with age, a fact which has been known for at least half a century. In Fig. 2 the "young normals" have a mean vital capacity of 4-8 litres while in the "old normals" the mean is 3-5 litres, a difference which is statistically significant. This change is not due to any decrease in the distensibihty of the lung itself; it is almost certainly due to fixation of the thoracic cage, arising from a variety of anatomical changes such as decreased elasticity and immobility of the costal cartilages. In emphysema the vital capacity is still further reduced. The volume of the lung increases with age (Fig. 3). In the younger group this averages 2-9 litres while in the middle- ♦Presented by Prof. R. V. Christie. 58 Respiratory Function and Age 59 YOUNG NORMAL EMPHYSEMA fI 80 YtABS Fig. 1. The age (listribution of the sul)jects iuvestijiated. this and subsetiuent similar tigures, the ordinate is the centage of the total number of subjects studied. In per- YOUNG NORMAL NORMAL EMPHYSEMA 40 30 %20 lO 40 30 OLD o; /b20 lO 40 30 /^20 MeAN«4>8 80^ 1-2 NO. 35 MEAN" 3-5 %oi 1-2 26 40 ■ o 2 4 1 a "5 LITRES MEAN* I -9 SO - 0-6 NO. 88 "3 UTRES Pig. 2. The distribution of the Vital Capacity. 60 D. V. Bates and R. V. Christie aged it averages 3-5 litres and in emphysema the average rises to 4-7 htres. One reason for this increase, and possibly the sole reason, is that the elastic recoil of the lung diminishes with age. During life the lungs remain distended because the elastic recoil is counterbalanced by traction of the chest wall and diaphragm, and expansion of the lung can be brought 40 MeAN.2-^ YOUNG 30 NORMAL />20 $0 2- 0-7 HO. 35 lO 12 3 4 1 C 5 6 7 8 LITRES 40 OLD o> ^ NORMAL /o20 1 MEAN. 3-5 soi 09 .o. 2 6 lO iIl P C 1 i 3 4 Ti 7 ^ LITRES 40 30 EMPHYSEMA ^20 MEAN. 4-7 soi 09 HO. 87 ID rid in - i i 3 Fig. 3. The distribution of the Functional Residual Capacity (i.e. the volume of air in the lungs at the end of an ordinary quiet expiration). about either by increased traction, such as occurs on inspira- tion, or by diminution in elastic recoil. That the elasticity of the lung does in fact diminish with age can be shown by direct measurement, and I will just say a word about the technique we have used. On inspiration the intrapleural or intrathoracic pressure becomes more negative, and this fall in pressure is a direct measure of the force exerted on the lungs by the muscles of inspiration. The tidal air is a measure of the volume change which occurs. If these two, the intra- pleural pressure and the tidal air, are measured simultan- eously it is quite easy to calculate the amount of work the Respiratory Function and Age 61 respiratory muscles liave to do in order to distend the lungs. It is also possible to indicate how much of this work is ex- pended on the elastic resistance of the lungs and how much on the viscous or non -elastic resistance (Mcllroy, Marshall and Christie, 1954). In Fig. 4 is shown a summary of our results. In eight healthy individuals below the age of thirty-five, between 70 90- PI 80- n " -| ^« °' ,c. TOTAL 60- WORK 50- r I n EXPENDED 40- AGAINST 30- -j -, ELASTIC 20- - RESISTANCE lO- 0- YOU sJG OLD EMPHYSEMA NOR MAL NOP MAL Fio. 4. The proportion of respiratory work expended against the elastic resistance of the lung. and 80 per cent of the work of breathing was expended against elastic resistance. In seven healthy individuals over the age of fifty, only between 50 and 60 per cent was expended against elastic resistance, and in emphysema the percentage is even lower. It is fair to conclude that with advancing years the elastic resistance or elastic recoil of the lungs is reduced (Mcllroy and Christie, 1954). So much for the mechanical aspects of lung function. The sole function of the lung is to ventilate the blood, and if it is to do this with maximal efficiency it must have at least two characteristics. Firstly, the lung should be venti- lated evenly throughout; in other words each alveolus should 62 D. V. Bates and R. V. Christie get just its fair share of eacli breath that is taken. Seeondly, the air in eaeh alveolus should I)e fully exposed to the lung capillaries; in other words, the alveolar air should eonie into proper contact with the pulmonary blood. Both of these functions can be measured, and I will first say a word about the efficiency of ventilation which, surprisingly enough, can be measured quite easily and accurately. HEALTH EMPHYSEMA 0 12 3 MINUTES < 5 2 ' 1 ■ ■ i. ; 8 5 > ip n — ' 1 — — — \ — ri — 1 — I — 1 ,iii.|ij'i||,— III II U| — . mn— — «- - Nil ill iiii m fe p W^ mn PP Hp^-i fT^-Brf^ 1 1 S^H / / ' — / — -/- 1 — -^^ — = 1 — 1 — — ( ««» M=i 1 1 - T I J < ^ ) b 7 ( i < ) 1 0 II MINUTES Fig. 5. Tlie efficiency of gas mixing in the lung measured by the hehum dilution method. In a healthy young individual (on left) equilibrium between spirometer and the lungs is reached after 80 breaths; in a patient with emphysema (on right) efjjuilibrium is only reached after 150 breaths. If an inert and relatively insoluble gas such as helium is breathed it will mix with the other gases in the alveoli but practically none will be absorbed (Bates and Christie, 1950). In Fig. 5 is shown the record of a young normal individual who starts to breathe from a spirometer containing 13-5 per cent helium. The helium percentage, which is measured electrically by a katherometer, falls, and after two minutes, Respiratory Function and Age 63 or 30 breaths, equilibrium is esta])lislu'(l ])ctweon luno and spirometer, with a hehum percentage of 1) 12. 11" the luii^ had been perfectly eftieient from tlie point <»!' view of a ventilation enoineer — i.e., if all the alveoli had been equally ventilated — it can be calculated that equilibrium should have been reached in 23 or 24 breaths, whereas in fact 30 breaths were required, and it is true that most normal individuals YOUNG NORMAL 204060 ?o i5o% Fig. G. The efliciency of gas niixino in the hino, expressed as a percentage of perfect elliciency, is shown on the abscissa. operate at 70 or 80 per cent efficiency as far as ventilation is concerned. The record of a patient with emphysema is also shown in Fig. 5 and here the story is a very different one. Mixing was only complete after ten minutes or 150 breaths. In Fig. 6 is shown a summary of our results. Young normals operate at an average of 75 per cent efficiency while the figure for the older group is only 54 per cent, a difference which is statistically significant. This can only mean that with ad- vancing years ventilation of the lungs becomes more uneven, and this is presumably a sequel of loss of elasticity. 64 D. V. Bates and R. V. Christie Lastly, is tlierc any change in the abihty of the inspired air to make proper contact with the pulmonary blood? It nii^ht seem that the obvious way to settle this point would be to analyse the arterial blood and find out just what has happened to the blood as it passed through the lungs. To do this with any accuracy it is not enough to measure the Og saturation of the arterial blood; both the alveolar and arterial YOUNG 80 60 emphysema) 40 20 20 40 60 ao /i"*^*" MEAN " 31 sot 6-4 NO 72 % OJ -G o CO p^ t- c.t: '^ 'S ^^ > a; O . 3^ 2 a> w a; ^ — ^ CO — .ti •^ «^ ^ ^ . ?: fc^ a; So; u ^ S • !|l "£"-/:! o "a) o ^ I: !S e; cs +^ 1^5 w ra 25 Age and Membrane Diffusion Coefficients 71 tops ground plane. The membrane is interposed between tlic ends of the syringes and is held in place by two metal dia- phragms and two hard rubber rings. A needle is inserted through each rubber ring; the outer ends of the needles are closed by insertion of a one-way metal stopcock. The gas or non-gaseous compound to be investigated is dissolved in buffer medium heated to 37° C. and the solution introduced into one of the compartments of the apparatus. The other compartment contains plain buffer medium. After both compartments have been filled with solution the diffu- sion apparatus is placed horizontally in a thermostat at 87° C. and rotation of the apparatus started. The rotation is pro- vided by an electric motor. Twenty to thirty minutes are allowed to elapse for establishment of temperature equilibrium and for initial penetration of the gas or compound through the membrane before withdrawal of samples for analysis. The technique provides for the withdrawal of samples from the diffusion apparatus at any time during an experiment. In studies on gas diffusion samples of the solution are trans- ferred without contact with the air to the extraction chamber of a Van Slyke apparatus for gas analysis. If conditions of analysis permit, samples should be withdrawn in immediate succession from the two compartments Of the diffusion apparatus at the beginning and end of a diffusion period. This will be possible in the case of diffusion studies on non- gaseous solutes, and in investigations on gas diffusion, provided two Van Slyke apparatuses are available. The diffusion coefficient is defined as the units of the sub- stance diffusing through 1 cm.^ of the membrane in one minute at a concentration gradient of 1 unit per ml. per cm. (Hill, 1928-29). For experiments in which samples for analysis are withdrawn in immediate succession from the two compartments at the beginning and end of a diffusion period the coefficient may be calculated from the equation: -'■■n(^^v^> (Cg— C4) = (Ci— c'a)^ 72 J. E. Kirk and T. J. S. Laursen where, A: = diffusion coefficient, Ci= concentration of solute on donor side at beginning of period, C2=concentration of solute on recipient side at beginning of period, ^3^ concentration of solute on donor side at end of period, ^4= concentration of solute on recipient side at end of period, A = area of membrane in cm. 2, L=thickness of membrane in cm., Vi and ¥2= volumes of solution in the two compartments expressed in ml., and ^=time in minutes. Results Aortic tissue Determinations have been made of the diffusion coeffi- cients of nitrogen, oxygen, carbon dioxide, lactate, iodide and glucose in experiments on 51 samples of intima preparations and 50 samples of media preparations. The age of the indivi- duals from whom the samples were obtained ranged between ten and eighty years. The mean diffusion coefficients observed for these prepara- tions are shown in Table I. The table further contains the calculated values for the products: diffusion coefficient X VMWXIO^ It will be seen from the table that a fair agreement was found between the product values for nitro- gen, oxygen and carbon dioxide, and that the product values for these gases were over two times greater than the product values for lactate and glucose. These findings might indicate the presence in the aortic membrane of a set of smaller pores (permitting the passage of compounds of MW 28 to 44) and a set of larger pores (permitting the passage also of compounds of MW up to 180). Age and Membrane Diffusion Coefficients 73 Table I Average Diffusion Coefficients Anti Mflan Valqes for Products : k x I4w x 10^ Intima (with attached ssbiatimal tissue) Media Solute Mtf m Diffusion coefficient Diffuaion eoefficieiut X ^^ X 105 Diffusion coefficient Diffusion coefficient xi/Sw X 105 Nitrogen 28 5.29 0.000^69 248 0.000551 291 Oxygen 32 5.65 0.000502 283 0.000579 327 Carbon dioxide U 6.63 0.00040^ 267 0.000375 24« Uctate 90 9.A9 0.000123 116 0.000064 80 Iodide 131" 11.^5 0.000318 365 0.000258 296 Glucose 180 13.-^ 0.000104 139 0.000076 102 • Radioactive isotope Table II Mean Diffusion Coefficienta ( x 10^) Observed for Various Age Groups Age group Number of Nitrogen Oxygen Carbon Lactate Iodide Glucose ( Years ) samples dioxide Intima (with attached subintlmal tissue) 10 - 39 13 39.3 ^3.9 35.9 9.3 25.3 7.4 AO - 59 20 50.0 55.0 AO.O 9.3 31.3 10.1 60-80 18 A9.2 50.1 4^.2 16.6 36.3 12.8 Media 10 - 39 13 49.9 50.5 32.9 6.4 23.0 6.1 AO - 59 20 54.3 60.7 36.7 7.6 24.4 7.5 60 - 80 17 59.7 6^.5 41.9 10.7 28.9 9.1 74 J. E. Kirk and T. J. S. Laursen Tlie average diffusion coefficient values observed for the age groups ten to thirty-nine, forty to fifty-nine, and sixty to eighty years are shown in Table II. It will be seen from the table that a definite tendency was found for the diffusion coefficients to increase with tlie age of the individuals. The significance of this tendency is also evident from the calculated coefficients of correlation between age and the diffusion coefficients (Table III). Table III PreDarations of the Human Aorta Intlma Media r t r t Age / Nitrogen + 0.28 2.04 + 0.35 2.60 Age / O^cygen + 0.18 1.27 + 0.29 2.08 Age / Carbon dioxide + 0.23 1.66 + 0.36 2.67 Age / Uctato + 0.^6 3.60 + o.a 3.12 Age / Iodide + 0.38 2.89 + 0.39 2.93 Age / Glucose + 0.58 4.99 + o.u 3.35 In connection with the data reported in Table III it should be noted that the total lipid and cholesterol contents of the aortic membranes were likewise found to increase witli advancing age. The coefficient of correlation age/total lipid was -|-0-48 for the intima samples and -f-0'32 for the media samples. The corresponding r values for the correlation, age/cholesterol were +0-40 and +0-31. Tentorium cerebelli Diffusion studies with this connective tissue membrane of comparatively simple structure were carried out on 35 samples derived from individuals ranging in age between one and Age and Membrane Diffusion Coefficients 75 seventy-eight years. Only the diffusion coefficients of carbon dioxide and glucose were estimated. The mean diffusion coefficient values for the age group one to thirty-nine, forty to fifty-nine, and sixty to seventy-eight years are shown in Table IV. In contrast to the findings in the experiments on the aortic samples the data fail to show any consistent change in permeability of the cerebellar tentorium with age. Table IV Mean Diffusion Coefflclenta ( x 10^) Observed for Various Age Groups for Human Tentorium CerebelH, Age Group Number of Carbon dioxide Glucose ( Years } samples 1-39 8 ^3.6 12.9 ^0 - 59 17 U.2 11.4 60-78 10 U.9 12.8 Summary Diffusion experiments were carried out with nitrogen, oxygen, carbon dioxide, lactate, iodide and glucose on prepar- ations of the intima and the media of the human thoracic aorta, using the diffusion procedure of Kirk and Johnsen (1951). The diffusion coefficient values showed a definite tendency to increase with the age of the individuals, indicating an increase in tissue permeability with age. In contrast to the findings irr the experiments on aortic tissue, diffusion studies on preparations of the human ten- torium cerebelli failed to show any increase in permeability of that tissue with age. The investigation was supported by a research grant (PHS-891) from the National Heart Institute of the National Institutes of Health, Public Health Service. 76 General Discussion REFERENCES Hill, A. V. (1928-29). Proc. roy. Soc. Ser. B., 104, 39. Kirk, J. E., and Johnsen, S. G. (1951). Circulation, 4, 478. DISCUSSION McCance: To what extent do you consider these tissues are ahve? I could not see any part of that apparatus designed to keep those cells aerated, and we all know now from the work which is going on at a cellular level, that if you do not keep even a single cell oxygenated its metabolism is completely disorganized; it swells up, changes its size, changes its osmotic pressure, and it can scarcely be regarded as a cell at all until you put it back into oxygen and allow it to recover. That seems to me to be a major criticism of this work, if it is valid. Kirk: I did not go into all the details of the experiments because of the limitation of time. In the diffusion studies which I reported at least one of the two compartments of the apparatus contained oxygen. It has been demonstrated in a previous study (J. Gerontol., 9, 10, 1954) that human aortic tissue after removal from the body will continue to show respiration for periods of up to several weeks when kept under sterile conditions. The Q02 values of such tissue preparations can be measured by means of a technique which we developed recently (J. hiol. Chem., 199, 675, 1952; 208, 17, 1954). This technique is considerably more sensitive than the ordinary Warburg procedure, since it permits deter- mination of Q02 values as low as 0 01. We have shown that under ordi- nary circumstances the aorta can be supplied in a thickness of 1 • 2 mm. with oxygen by diffusion, leaving a reserve margin of about 30 per cent. McCance: It's extremely interesting how little oxygen tension is needed to provide, say, the cell of the intima with the amount of oxygen it requires because, physiologically, it is bathed, as a rule, in a medium containing so much. x\m I right in thinking that you separate the muscular tissue from the intima before incubating the latter? Kirk: No. In the media we have some smooth muscle cells, but the values I gave for the Qo^ refer to the aortic wall. And there is not much difference in the Qo.^ of the human intima and media. The maximum value is about 0-30. Franklin: Have you done any experiments on animals? Because I think it might answer Dr. McCance's question if you took some of these pieces and used them to repair blood vessels in other animals. Kirk: Well, I have carried out determinations of the rate of respiration and glycolysis of the dog aorta (J. Gerontol., 9, 10, 1954). The studies showed that the Qoo of the dog aorta is approximately twice that of the hmuan aorta; about 0 • 65-0 • 70 is the maximum Q02 found in dogs. Our observations on the respiration of aortic tissue stored at refrigerator temjicrature between tests are in essential agreement with the data by Pierce and his associates {Ann. Surg., 129, 333, 1949), who showed that General Discussion 77 it was possible in many instances to obtain growth of fibroblasts in tissue cultures from segments of aortas stored for periods of thirty to fifty days at 4>° C in a salt solution-serum medium. Franklin: You haven't used a piece of the old vessel for repair? Kirk: No, I have done no transplantation studies. McCance: I think there is more in it than whether the tissue is alive sufficiently to use for "repair" purposes. If, for instance, you take a piece of kidney and cut thin slices of it and immerse it in the solution containing no oxygen, the cells swell, but they can be restored to nor- mality by replacing them in oxygen. The cells can also be made to swell and become very abnormal by putting them for a time in cyanide solution or some other poison: yet they may recover after they have been removed from the cyanide and placed in a good oxygenated saline. It is not so much a question of what the aorta will do later, as its exact condition at the moment at which Dr. Kirk is studying it. And I would feel that the value of this work would be enormously enhanced if he could establish that his tissue was, when he was studying it, absolutely normal, or as normal as it can be when it is old. Kirk: Of course, it is very difficult to establish that a tissue is absolutely normal. We have measured the thickness of the aortic mem- branes before and after the experiments, and the variations we have found have only been of the magnitude of + 5 per cent. We believe that one reason for our ability to keep the membrane of the same thickness for some period of time has been the use of the buffer medium indicated by Aebi {Helv. physiol. acta, 10, 184, 1952). This buffer has a rather high potassium and calcium concentration. Aebi's studies have shown that with a buffer of such composition no appreciable swelling of the tissue occurs. Furthermore, the loss of nitrogenous material from the tissue is very small. Our observations on aortic tissue (J. Gerontol., 9, 10, 1954) have confirmed Aebi's findings. Aub: Your CO2 diffusion is lower than the oxygen and nitrogen; isn't CO2 much more diffusible? Kirk: This has been discussed in the literature for a long time. As far as I can see, the conflicting statements with regard to the diffusion rate of carbon dioxide gas in solution compared with that of oxygen are due to the fact that two different definitions of the diffusion have been employed. One of these definitions is based on the difference in partial pressure of the gas on the two sides of the membrane (Krogh, 1918-19, J. Physiol., 52, 391), whereas the other is based on the difference in the quantity of gas per volume of fluid (i.e. the concentration) (Hifl, Proc. roy. Soc, B, 1928-29, 104, 39). If one employs the definition of the diffusion coefficient introduced by Krogh the value for carbon dioxide wall be much greater than that for oxygen. This is due to the fact that the absorption coelficient of carbon dioxide in water at 38° C is 0-550, whereas that of oxygen is only 0 • 023. For the same partial pressure of the gases the carbon dioxide content of the w ater will therefore be 23 • 8 times greater than the content of oxygen. Since the diffusion of gases in water is inversely proportional to the square root of their molecular weights, the diffusion rate of carbon 78 General Discussion dioxide will be 23 • 8 x \/S2/\/4>4>, or 20 • 3 times greater than that of oxygen (at the same partial pressnre difference). In the field of chemistry, however, especially when dealing with diffusion kinetics, Hill's definition of the diffusion coefficient is generally employed, and — as mentioned in the presentation of my data — it was this coefficient which was used in the present study. According to Hill's definition of the diffusion coefficient, the ratio: Diffusion coefficient for carbon dioxide/Diffusion coefficient for oxygen, will be equal to \/ 32/ V 44, or 0 • 853. In the experiments on the 51 human aortic intima samples presented today the average diffusion coefficients observed for carbon dioxide and oxygen were 0- 000404 and 0- 000602 respectively. This gives a ratio between the coefficients of 0-805, a value which is in fair agreement with the ratio of 0 • 853 obtained for the diffusion of the gases in water. Lansing: I wonder if one can properly compare the intima of the young with the intima of the old, since there are so manj^ anatomical differences. In stripping the intima and the associated sub-intimal material of the young by blunt dissection, one winds up with the endothelium and a very thick elastica interna. In this stripping the elastica interna goes with the endothelium. The elastica interna has periodic fenestrations which are quite gross and rather widely separated so that between the fenestrations there are very definite barriers to diffusion. Now, if one examines the separated intima of the old, it's a very different story. One finds the endothelium, a very rich bed of fibroblasts, collagenous fibres, ground substances and so on, which may measure as much as a millimetre or two, depending upon the pathological state of the tissue, and then, far beneath all this material, one finds the fragmented and disrupted elastica interna which is so badly damaged that often it does not go along with the sub-endothelial material in stripping. Thus in the young there may be a mechanical barrier to diffusion in the elastica interna which is lacking in the old. This alone may contribute to your data. The implication would be that the sub-endothelial material be- tween the elastica interna and endothelium is not a barrier to diffusion. Medaivar: Dr. Kirk, if the membranes you are working with are deliberately killed by non-precipitants, or narcotized, do you still get the age differences? And you mentioned using oxygen uptake as a criterion of the viability of the cells — are there age differences in the oxygen uptakes of these various membranes? Kirk: I have not made any studies on the effect of enzyme inhibitors or various poisons on the membrane permeability. And as far as the oxygen consumption is concerned, our data (J. Gerontol., 9, 10, 1954) failed to show any change with age in the oxygen consumption of the aortic tissue membranes. Bean: Do you believe that the increase in cholesterol is a cause, an effect, or an independent parallel of the change in permeability you found in relation to arteriosclerosis? Kirk: I have no real opinion on that point. We merely carried out determinations of the total lipid and cholesterol content of the aortic membranes to have a quantitative index of the pathological changes in General Discussion 79 the tissue, but I have no opinion on the direct relation between the cholesterol content and the diffusion cocllicients. Bean: Was there an indi\idual correlation as well as an average correlation? Kirk: We have only carried out experiments on 50 membranes, and the number of samples for each five-year group is too small in my opinion to make a computation. Franklin: What is the total amount of fluid diffusing through the wall of the aorta as compared with that in the capillaries? Kirk: I don't know \ The data that I reported refer only to the diffusion of solutes through the membrane. One of the valuable features of the procedure is that the free-moving syringe plungers in the diffusion apparatus ensure a constant total pressure on both sides of the mem- brane, so there should be no mechanical movement of fluid from one side to the other. We usually measure the volumes of fluid in both compart- ments at the end of the experiments. They generally correspond quite closely to what should be expected, which indicates that there has been no net transfer of fluid from one side to the other. THE CHANGING INCIDENCE OF CERTAIN VASCULAR LESIONS OF THE SKIN WITH AGEING W. B. Bean, m.d., Department of Medicine and the University Hospitals of the College of Medicine, State University of Iowa, Iowa City, U.S.A. The subdued tones of discussion of the sombre aspects of ageing have echoed down the immemorial lanes of history; and the subject has been treated variously by writers and philosophers from Cicero to Ponce de Leon, from Seneca to Browning. Our recent preoccupation with gerontology in an ageing society casts oblique Darwinian light on those who, growing old, pontificate upon growing old. Gray hair, obesity, wrinkles, presbyopia, loss of muscle tone, the menopause, the burden of disease in later decades — all these have had increasing attention. Yesterday's introductory probings, aimilig for a definition of ageing without a prior definition of life or of time — Newtonian or Einsteinian — illustrate the large difficulties of our problem. Is ageing merely the running down of a wound-up clock, a Calvinistic view, or is it such a process in a clock which is buffeted about and thus does not run as long as its unbuffeted control? Yesterday's suggestion that we get more data by observing the process and progress of ageing as it goes, rather than look at the artificially stopped frame in a moving picture of panoramic dimensions, gives a mere clinical observer some excuse for participating in this colloquium which I take to mean talking to each other. Thus as junior in wisdom and in years but still a member of an ageing society, I join in our increasing and self-conscious focus upon gerontology. I hope I may avoid a geriatric hazard not always escaped in the 80 (a) (b) (c) (tl) Fk;. 1. (a) Arterial spiders of the skin. Black and wiiite i)hoto<>raph. (b) Same area oi' the skin showino the failure of arterial spiders to show in an infra red photograph, and the location of the lesions away from veins, (c) Venous star. Black and white photograph, (d) Venous star. Infra red photoorai)h showing tlie connection of the lesion with tlie larger veins to which it is tributary. [ To face jxige S 0 Fic;. 4. The caviar lesion under the tongue. Vascut.ar Lesions of Skin 81 academic arena, that of becoming' mellow before I have grown ripe. My brief discussion today concerns vascular lesions in the skin, and their increasing incidence with advancing years. For the most part it emphasizes observation rather than interpretation. I have no evidence to demonstrate preponderant environmental or genetic influences. The study represents merely the first step in clinical research, which is systematic observation and recording of data. I have only vague notions of why the several types of lesions occur at all and can give you only speculation about the effect of increasing age upon them. The observations began about fifteen years ago when I failed to get any help about the nature of arterial spiders from the Nestors of American medicine at Johns Hopkins or Harvard. I was sufficiently curious to attack the problem from the simple approach of the natural historian. Having become interested in one kind of vascular alteration of the skin I noticed others closely or distantly related, or sometimes merely confused in the medical writings on the skin. The data reported here are based on a study of about 700 persons with arterial spiders and palmar erythema, 75 persons with Osier's disease, and more than 1,000 patients surveyed in detail for cherry angiomas, venous stars and caviar lesions. The venous lake on the ear has been studied systematically only recently. Spider Naevi To try to throw light on the significance of the eruptive spider naevus (Fig. 1) of liver disea;Se and pregnancy I made a series of studies which led to the hypothesis that such vascular lesions were overgrown and hypertrophic end arteries whose increase in size was probably caused by oestrogenic hormones. In chronic disease of the liver these hormones are in high concentration in the blood because the liver cannot ade- quately inactivate, destroy or excrete them. In pregnancy the liver functions well but the level is high because the hormones 82 W. B. Bean are produced in great quantity. Since approximately 10 per cent of normal persons may have one or more vascular spiders their clinical significance is important only when there are many, when new ones appear or when they enlarge. Age: The eruptive spider naevi occur in the childbearing period in women and in the cirrhotic patients mostly in the thirties and forties. There is no apparent relation to age per se. They occur in very young or very old persons. Palmar erythema as an acquired vascular change in the skin has the same backgound, course and age incidence as spider nsevi. The relationship to hypercestrinaemia is not so close though there is the same clinical background. Venous Stars As the name implies the venous star (Fig. 1) is a stellate system of collecting veins visible in the skin and larger than those in normal skin. They are small varicose veins, vari- cules, appearing usually upstream in venous collecting systems of the skin where pressure is regularly or intermit- tently high. They are common on the lower legs and thighs in association with larger varices. Blood flows from the periphery to the centre, thence to an underlying collecting vein. Such lesions may appear in large numbers over the chest in persons with obstruction of the superior vena cava during the period of development of collaterals. Age: Fig. 2 shows that as age increases many more persons are found to have venous stars. There is a consistent tendency for more women than men to be affected in any age group. This is associated with the well known tendency for women to have more and worse varicose veins than men. Cherry Angiomas There are many names for the cluster of small dilated capillary vessels which form the characteristic bright red ruby spot, De Morgan's spot, senile angioma or capillary angioma. The size and shape vary but the typical spot is Vascuiar Lesions of Skin 83 round, the small ones flat, the lar<>e ones elevated and dome- shaped. In most persons when the large ones oceur, magni- fication of twenty times reveals many smaller ones with but one 10 - 19 ;" 20-29 7 30-39^ S 40-49 7^ 50-59;" 60-69 7 70-79 7 m ^^ 20 40 60 PERCENT 80 100 Fig. 2. The increasing incidence of venous stars with increasing age. In every decade women outnumber men. The bars indicate the percentage of all subjects found to have one or more lesions. 10- 19 70-79 0 20 Fig. 3. Incidence of cherrv angioma by sex and decade. 40 60 80 100 PERCENT or a few capillary tufts. Pressure partly blanches the lesion and the colour and the natural shape are restored after about a minute. Age: Fig. 3 shows the rapidly rising incidence 84 W. B. Bean with increasing years so that in the thirties about three- (juarters of the ])ersons observed have the lesions and they are almost always found in the aged. Not only do more people become affected but the spots are more numerous and larger. Caviar Lesions The caviar lesion is a small, roughly spherical or dome- shaped varicose enlargement of the poorly supported veins 10- 19 20-29 80 100 Fig. .'5. Incidence of caviar tongue lesions l)y sex and decade. in the collecting system under the tongue. The small ones develop as an outpouching of a branch of a communicating vein near the ranine veins. Pressure with a glass slide causes them to empty. Knots and clusters of such vessels with a dark blue or black colour resemble caviar or buck shot (see Fig. 4). Anatomically the caviar lesion is a dilated vein with hypoplastic endothelium but a fairly thick wall. There are no inflammatory changes. Age: There is a sharp and fairly steady increase in incident with increasing age; and the lesions grow larger and more numerous (Fig. 5). Vascular Lesions of Skin 85 Osier's Disease {Hereditary Hcemorrhagic Telangiectasia) The cutaneous lesion in Osier's disease is usually flat, punctiform, with sharp margins. Solid bluish lesions may occur. The mucous surfaces of the body, especially of the mouth and lips, are favourite sites. Bleeding is common as would be expected from the vascular fault, a deficiency of smooth muscle and elastic tissue in small veins, arteries and capillaries. The trait is transmitted without sex linkage as a Mendelian dominant. Recently a number of persons with this disorder have been observed to develop pulmonary arteriovenous aneurysms. Age: I have observed some children with no skin lesions who bled from the nose and after the age of puberty atypical telangiectases appeared in the skin. Bleeding is rare before puberty. It most often becomes a clinical problem during the third decade and so remains throughout life. There may be slow increase in size of some lesions, others may vanish but the change is usually gradual over a period of years. Venous Lakes In recent years I have made observations on a dilated venous structure found on the ears and lips of old persons. They are rare in women, but not so rare in old men. In colour they are dark blue. Pressure on them causes fading or com- plete blanching. They refill slowly. Histological section reveals a thin vascular structure with walls resembling veins. Discussion The life of man witnesses many changes in blood vessels — the great proliferations during embryonic life, the sharp shifts of the neonatal period, the tendency for birthmarks and haemangiomas generally to disappear during the early years of life. But there are vascular alterations of another class readily observed in vessels of the skin, which increase in size, 86 W. B. Bean number and in frequency of occurrence with ageino-. In Fig. 6 the effect of ageing is noticed in the somewhat reciprocal relation of persons exhibiting all three of the venous lesions and those who have none. While the cherry angioma is proliferative and grows larger as time goes on the vessel walls are thin and atrophic. Caviar lesions are dilated and varicose veins of small size. Their enlargement may be favoured by increases in pressure especially because they are not protected 100 80 60 40 20 10- 19 20-29 30-39 O 40-49 7 50-59 60-69 70-79 ALL ABSENT / Cherry Angioma Venous Stars Caviar Tongue Lesion / ALL PRESENT 10-19 20-29 30-39 40-49 50-59 60-69 70-79 20 40 60 80 100 PERCENT Fig. 6. Age and sex distrilmtion of persons with all three venous lesions and those with none. by supporting tissue without or valves within. Venous stars increase with ageing chiefly in relation to chronic stasis in large veins, but may occur whenever the venous pressure remains high. These observations provide a starting point for detailed work on the increase in frequency, in size, and in number with ageing. Nothing emerges to tell whether they are the result of the intrinsic process of ageing, the mere passage of time, or whether they result from the stress and strain inherent in life and increase with ageing because environmental stimuli accrue as time passes. Should further work help solve the Vascular Lesions of Skin 87 puzzle it might be said with more justification than is now possible that "small inferior veins From me receive that natural competency Whereby they live." Coriolanus, Act /, Scene I. REFERENCES Medicine, Baliimore, 24, 243. Trans. Ass. Anier. Phycns., 64, 100. Quart. Bull. Northw. Univ. med. Sch., 27, 89. Trans. Anier. clin. climat. Ass., 64, 40. Bean, W. B. (1953). Circulation, 8, 117. Bean, W. B. (1953). Trans. Ass. Amer. Phycns., 64, 240. Bean, W. B. (1945). Bean, W. B. (1951). Bean, W. B. (1952). Bean, W. B. (1952). AGEING OF ELASTIC TISSUE AND THE SYSTEMIC EFFECTS OF ELASTASE Albert I. Lansing, ph.d. Department of Anatomy, Emory University, Georgia. It is interesting to note that although collagenic and elastic fibres have closely related origins, they are remarkably different functionally, chemically, and anatomically. The collagenic fibre is a tough, stretch-resistant element that is quite rich in hydroxyproline and is characterized by regular transverse periodicity. The elastic fibre, as the name indicates, is very elastic, contains only a trace of hydroxyproline and is annoy ingly free of structural characteristics. The latter is especially true of elastic fibre preparations in the electron microscope. A propos of the present discussion it may also be added that these two fibre types differ in their suscepti- bility to pathological alterations; there are a great number of diseases affecting collagen and the generalized fibrosis of ageing is almost a diagnostic feature of that process. On the other hand there are but a limited number of diseases of elastic tissue, and the ageing of this material is highly localized. Hass (1939) has reviewed much of the data concerning elastic tissue diseases. In my present discussion it is intended to review some of the data on age changes in arterial elastic tissue and its relation to human arteriosclerosis, and to note that the age changes in dermal elastic tissue of the human are strikingly different from those found in arteries. Arteriosclerosis, edited by Cowdry in 1933, reveals a sur- prising unanimity of opinion in regard to the nature of this disease. Fox, in outlining the phylogenetic aspects of arterio- sclerosis, emphasized the significance of changes in the 88 Ageing of Elastic Tissue 89 elastic tissue of the media and tlie relations of such elastic tissue changes to atheromata. "The only feature that is clear is that the less definite a lamina (elastic) protecting the media, the more definite is the atherosclerosis". Ophuls observed that "the most important change caused by ageing of the arterial wall is a gradual diffuse distension due to the pro- gressive deterioration of the elastic tissue". Wells argued strongly for the importance of elastic tissue failure in arteriosclerosis. The fact is that a large volume of data had been accumulated describing a characteristic fraying and fragmentation of elastic tissue in arteriosclerosis, frequent reduplication of the elastica interna, and development of a profound affinity by elastic tissue for calcium salts. Curiously enough, Anitschkow, who fathered experimental cholesterol atherosclerosis, pointed out that lesions of the arterial wall coupled with intimal fibrosis predispose to atheromata in the presence of hypercholesterolaemia. It is quite understandable that Anitschkow's production of atheromata by the administration of cholesterol resulted in an intensive effort over the last twenty years to link arterio- sclerosis to faulty lipid metabolism. These studies have been strongly reinforced by the lipoprotein studies of Gofman and of Barr, by the dietary studies of Keys, and by many others too numerous to enumerate. As these studies developed it became increasingly difficult to relate elastic tissue changes in the arterial wall to the aetiology of arteriosclerosis. Our laboratory has been led to the conclusion that arterio- sclerosis may be considered to be not one, but two diseases. One disease involves a defect in cholesterol metabolism or circulation, while the second disease is manifested by a breakdown in the structure of tlie elastic elements in the media of arteries accompanied by a calcification of this elastic material. This point of view was supported by our observations on human material that the elastic tissue break- down occurs prior to the formation of atheromata, that the changes in elastic tissue are associated with age and may occur without atheromata (the converse is not true), and that 90 Albert I. Lansing when these two lesions co-exist the accumulation of choles- terol in the intima occurs after the elastic tissue of the underlying media has broken down. Whether or not this formal hypothesis is sound is not too important. The cardinal point is that in human arterio- sclerosis there is an almost invariable association of medial elastic tissue degeneration with the conspicuous atheromata. The questions are: what relation is there between cholesterol accumulation in the intima and elastic tissue breakdown in the media; and, is there a common factor responsible for the production of both of these lesions? It would be excessively repetitious to present once again all of the histological and analytical data which have contri- buted to the development of the point of view of my labora- tory. For the purposes of this discussion it might be more appropriate to make a few points on the chemistry of elastic tissue in relation to age and arteriosclerosis and to attempt to evaluate factors that condition elastic tissue in these two states. Since all of our data depend upon the method of preparation of elastin this procedure will be described before making three points; first, that although elasticity of arteries decreases with age, there is no apparent loss of elastin; second, that elastin extractable from old human aortas has an amino acid composition distinct from that of young elastic tissue; and third, that elastic tissue degeneration as measured by calcification thereof occurs as a function of age. Preparation of elastic tissue Elastin may be prepared in a reproducible form and with several objective measures of purity. A slight modification of the method of Lowry, Gilligan and Katersky (1941) was applied to the tunica media of fresh aortas. The tissue was refluxed in methanol or ethanol for one hour and in acetone for a second hour. This defatted material was then digested at 98° C in 0 • 1 N NaOH and small samples taken off at five minute time intervals for chromatographic, histological, and chemical analyses. Ageing of Elastic Tissue 91 Two dimensional paper chromatography appears to be an effeetive means of characterizing elastin. After complete hydrolysis, tryptophane exists as a trace in elastin (Lansing et al., 1951) but is a significant component of other tissue proteins. By tracing the disappearance of the ninhydrin colorized tryptophane spot in the chromatogram one can readily determine when contaminants of elastin are removed. This w^as done in the NaOH digestion series and it was determined that the tryptophane spot disappeared after forty to fifty minutes. At this time, glycine, proline, leucine, isoleucine and valine were readily identified in the chromato- gram and minor spots for aspartic and glutamic acid were present. The finding of these amino acids as the principal components of elastin is consistent with published analytical data for the amino acid composition of elastin (Neuman and Logan, 1950). There w^as a progressive reduction in the amount of residual material through forty to forty-five minutes of digestion, after which time the residual dry weight was constant through sixty minutes. These data seem to indicate that digestion of elastic tissue in hot, dilute alkali for less than forty-five minutes is not adequate to remove extraneous materials from the tissue (Fig. 1). Microscopic examination of samples digested in NaOH for time periods less than forty-five minutes revealed variable amounts of collagen and muscle after staining with Mallory's procedure. At and after forty-five minutes no collagen or muscle could be demonstrated and the elastin was optically homogenous, refractile, and apparently free from contami- nating structures. Likewise, elastin prepared from ligamen- tum nuchae was free of collagen in so far as could be determined by light microscopy and electron microscopy. It w^as bire- fringent when stretched, or dried, had a refractive index of 1 • 534, was resistant to digestion by crystalline trypsin (Armour) and stained effectively with orcein, resorcin- fuchsin, or the Verhoeff procedure. 92 Albert I. Lansing Constancy with age of human arterial elastin A total of 110 human aortas and 106 human puhiionary arteries from which both intima and adventitia had been removed by stripping were analysed for elastin content by the method just outlined. The results obtained by these procedures on aortas ranging in age from stillborn to one hundred and three years are 90 80 70 >- 50 I 40 Z 5 ^ o cc u «^ 20 10 10 15 ■35 — -to — tS — ^3 — 55" 20 25 30 MINUTES Fk;. 1 . Hydrolysis of human aortas in 0 • 1 n XaOH showing that incubation at 98° C. for at least 45 minutes is needed to procure a constant residuum of insoluble elastin. 60 summarized in Fig. 2. The average elastin content of the media during the first two decades of life is slightly over 48 per cent while the average elastin content in the third decade and thereafter drops to values ranging between 41-1 and 44 • 1 per cent. Because of the limited number of specimens available for analj^sis in the first two decades it is difficult to decide whether or not this apparent drop in medial elastin AgeinCx of Elastic Tissue 93 content is .significant. Certainly there is no significant gain or loss of elastin after the third decade of life. The data for the ])ulnionary arteries followed a somewhat different pattern from those of the aortas. During the first three decades of life the media of the pulmonary artery q:60 <50 Q UJ ?40 < d30 20 10 D- AORTA (means) •- PULMONARY • •*• • 20 AGE 80 Too Fig. 2. Analyses of the elastin content of human aortas and pulmonary arteries at various ages. The point for aortas at 10 years is of doubtful significance because of the small number of analyses involved. contained slightly less than 31 per cent elastin. This increased slowly but steadily throughout life to 34 per cent in the seventh decade and to almost 37 per cent in the eighth decade. The increase with age of elastin, although small, is statistically significant. It would appear then that at least after the second decade of life the elastin content of elastic vessels such as the aorta and pulmonary artery either remains constant or actually 94 Albert I. Lansing increases. The loss of arterial elasticity with age cannot be attributed to a loss of elastic tissue. It is more likely due to a change in the properties of elastic tissue with age. Amino acid composition of young and old arterial elastin Microbiological assay of aortic and pulmonary elastin prepared in the usual manner was effected by the method of Roberts, Ramasarma and Lewis (1950). Nitrogen analyses indicated a content between 15 and 16 per cent suggesting that elastin is largely protein. Highly significant increases in the contents of aspartic and glutamic acids were noted in the samples of old aortic elastin as compared to those of the young elastin, in confirmation of the chromatographic findings. There were also increases in the contents of amide nitrogen. However, there was a four-fold increase in the excess of dicarboxylic acids over amide nitrogen, showing that the quantity of free carboxyl groups was increased significantly in the older specimens. The mean values for valine, proline, and glycine contents were somewhat lower in the older samples, while the leucine and isoleucine contents in the two groups were closely similar. In contrast to the findings for aortic elastin, the pulmonary elastin did not show increases in aspartic and glutamic acid contents with age. The other amino acids also did not change significantly. The contents of the proline, leucine, isoleucine, and valine were virtually identical with those found in young aortic elastin. However, the contents of glycine and the dicarboxylic amino acids were higher in the pulmonary elastin. The quantity of free carboxyl groups was at least twice as great as that found in young aortic elastin, while the range showed some overlap with that of the old aortic elastin. It was found that a separa- tion of fractions having differing specific gravities could be achieved by differential centrifugation of finely ground elastin (Wiley mill) in sucrose solution (sp.gr. 1 -30). Most of the material in samples of aortic elastin from very young individuals floated on the solution after centrifugation, while Ageing of Elastic Tissue 95 most of that from old individuals settled to the l^ottom. A sample of the suitably washed light fraetiou (young light) was prepared from several young aortas and some of the heavy material (old heavy) was prepared from old aortas. Both of these materials were analysed for calcium and the contents of 18 amino acids. Table I Percentage of Principal Amino Acids in Purified Elastin from Aorta Young Senile Aspartic Glutamic . Glycine Valine Proline Leucine Isoleucine . 014 1-6 26-5 16-4 13-6 6-8 3-5 1-5* 4-4* 230 14-3 13-9 7-4 3-7 ♦Significant increase. In Table II are shown the contents of 18 amino acids in the light fraction of young aortic elastin and the heavy fraction of old aortic elastin prepared by suspension in sucrose in the manner described previously. In the case of the 7 amino acids which were determined on whole elastin (Table I), the differences observed between the young light and old heavy fractions were all similar to those found for the young and old whole elastin. The old heavy elastin show^ed increases in aspartic and glutamic acids and decreases in the contents of glycine, proline, and valine. There ^was also an increase in the number of free carboxyl groups in the old heavy sample. In addition, there was a decrease in the content of alanine. All of the other amino acids showed increases of varying degree in the old heavy elastin. Approximately 90 per cent of the nitrogen of both samples was accounted for by the analyses. The light fraction contained 1-14 per cent calcium while the heavy fraction contained 6-39 per cent. 96 Albert I. Lansing Table II Amino Acid Composition of Young Light and Old Heavy Elastin Amino Acid Young "■light"* g N per 100 gN Old "heavin g N per 100 gN Aspartic Glutamic 0-38 1-83 111 301 Glycine Proline 26 10 1010 21-30 9-20 Leucine 4-52 4-76 Isoleucine . 210 2-33 Valine 13 00 11-50 Alanine 2318 21-58 Lysine Arginine Histidine . 0-49 1-78 015 117 4-35 0-75 Cystine Methionine 006 006 010 0-35 Phenylalanine Tyrosine Tryptophan Serine 1-73 1-45 006 0-29 1-97 1-76 0-24 0-70 Threonine . 0-65 1-13 Amide N 2-90 2-86 Total 90-83 90 17 ♦This preparation contained 1 • 14 per cent calcium and 14-9 per cent N, fThis preparation contained 6-39 per cent Ca and 12-5 per cent N. These data indicate that the amino acid composition of young and old arterial elastin is significantly different. It is doubtful that the shift in amino acid distribution represents a change with time in the amino acid composition of a pure protein. The elastin studied may be a mixture of two or more proteins, and the changes observed may be related to differences in the ratios of the separate constituents. How such a change can come about is not yet known. It is possible that the proteins laid down later in life may be difPerent from those formed earlier because they are formed under the influence of fibroblasts which themselves have undergone age changes. Ageing of Elastic Tissue 97 Age calcification of medial elastic tissue The question has repeatedly arisen whether the breakdown and calcification of medial elastic tissue is a sequel to athero- matosis or whether this is an age dependent phenomenon. In an attempt to resolve this question atheroma-free segments of the upper abdominal portions of human aortas were collected and both adventitia and intima were separated by stripping. To guard against chemical contamination of the 7- MEDIAL ELASTIN FROM PLAQUE-FREE AREAS OF HUMAN AORTAS AGE Fig. 3. Analyses of the calcium content of elastin from human aortas using grossly normal specimens. In plaque-free areas there is a progressive increase with age in tlie calcium content of the elastin. selected areas by adjacent and overlying atheromata or necrotic and calcified plaques we selected for analysis 14 aortas which possessed an over-all minimum of atheroma- tosis. Little difficulty was experienced in collecting atheroma- free tissue in the young group while considerable selection was required in the older specimens. The separated tissues collected in this age series were analysed for calcium content by the method of Salomon, Gabrio and Smith, and for cholesterol content by the method of Kingsley and Schaffert. The data as summarized in Fig. 3 show that there is no significant age change in the amount of 98 Albert I. Lansing medial elastin or in the calcium and cholesterol contents of the intima. There is a small increase in the cholesterol content of the media and a thirty-fold increase in the calcium content of the medial elastin. It would appear from these data that the development of an affinity for calcium by arterial elastic tissue is an age-conditioned process. It seems clear that change in composition of human arterial elastic tissue, which is almost invariably associated with overlying atheromatosis, is age-conditioned. That age alone is not effective is attested to by the fact that the pulmonary artery is normally resistant to elastic tissue change at all ages. The pulmonary artery does develop typical elastic tissue changes as well as atheromatosis when it is subjected to stress as in pulmonary hypertension. In presenting these several forms of data, I have not been attempting to propose that the primary lesion of arterio- sclerosis is elastic tissue breakdown rather than accumulation of lipid in the intima. My point is that arteriosclerosis is a complex disease; at least in the human, arteriosclerosis is a product of both accumulation of lipid in the intima and degeneration of elastic tissue in the media of arteries. The question is: Is there a common basis to both of these lesions or is their occurrence coincidental? I suspect that the former possibility is valid and that the material called elastase may be implicated in both elastic tissue and lipid metabolism. The data are far from complete and elastase is as yet poorly defined but sufficient experimental data are available to warrant serious consideration of the role of this material in arteriosclerosis. Elastase, exclusively extractable from the pancreas, was originally prepared and characterized as an enzyme by Balo and Banga (1950). It is apparently specific for elastic tissue and acts upon the substrate to convert a fibrous protein to the globular form; in solubilizing elastic tissue there is no apparent release of peptides or amino acids. In this labora- tory elastase has been used to work out the fine structure of elastic fibres. The enzyme solubilizes both elastic fibrils and Ageing of Elastic Tissue 09 a matrix material which together make up the elastie fibre. In a recent study of the i)hyl()i^enetic distribution of elastase, Lansing, Rosenthal and Alex (11)53) not only found this enzyme in a teleost fish but also determined that it was present in the islet rather than acinar tissue. Lophius piscatorkis (goose, monk, angler fish) like several other teleosts has islet tissue that is anatomically separate from the acinar portion of the pancreas. Each, then, can be collected separately and in relatively pure form. We found the elastase activity to be confined exclusively to the islet tissue and were thus led to suspect that elastase has a systemic rather than digestive function. This suspicion is supported by the w^ell- known observation that elastic fibres resist digestion in the alimentary tract; adult human pancreatic juice has failed to show elastase activity; the yield of elastase from whole pancreas is extremely low; and beef pancreas is an excellent source of elastase. Since this animal is herbivorous there would appear to be no need for a digestive enzyme for elastic tissue. Balo and Banga have recently indicated (1953) that the elastase content of the pancreas of human arteriosclerotics is substantially less than normal. Close inspection of their tabular data shows that age is as much a variable in their material as is arteriosclerosis. For example, the group of arteriosclerotics w^hich yield a mean of 9 elastase units per gram had an average age of sixty-one years; the group of non-arteriosclerotics who came to autopsy because of other diseases had an average age of thirty-four years and a mean of 155 elastase units per gram; lastly, the group of non- arteriosclerotics w^ho came to autopsy through violent death had an average age of thirty-one years and 208 elastase units per gram. The possibility appears to exist that loss of elastase may be related to arteriosclerosis and/or ageing. Although elastase has been crystallized by a rather labor- ious procedure with a very low yield, this material is generally prepared as little more than a pancreatic extract by acid 100 Albert I. Lansing extraction and salting. In such extracts the elastase activity in vitro or in vivo is destroyed by boihng. The elastase activity of pancreatic extracts persists after fat extraction and dialysis. Curiously enough, although these conditions would suggest that elastase is a protein, the systemic effects to be outlined were obtained by oral administration of the material. A unit dose per rabbit was that amount of elastase extractable from 5 grams of pancreatin (Viokase, Viobin Corp., or Wilson's concentrated pancreatin). In two groups of tracer studies using ^*C labelled acetate (COOH) and glycine (CH3) with and without elastase, it appears that the oral administration of this material has significant effects on the metabolism of aortic elastic tissue. Rabbits, in groups of five, were given 0 • 1 mc of either of the two isotopes by intraperitoneal injection and followed for two to twenty-eight days. Animals were sacrificed at various time intervals after injection of the glycine or acetate, the aorta from the arch to the diaphragm was removed, fat- extracted, reduced to elastin, hydrolyzed and transferred to planchets for counting. The only difference between the controls and experimental rabbits was that the latter received daily doses of elastase for the duration of the experiment. The data are summarized in tabular and graphic form (Table III & Fig. 4). The results with the controls indicate that the metabolism of elastic tissue is considerably like that of collagen as established by Neuberger using comparable doses of the isotopes. Every indication is that the turnover of elastic tissue is very low, and indeed after administration of elastase is even lower than in the control. It is difficult to propose a formal explanation for this result; one possibility is that elastase removes a particular component from elastic tissue, one that is substantially labelled in the control. Whatever the final explanation for this phenomenon may be, it is clear that oral administration of elastase has an in vivo effect on the isotopic labelling of elastic tissue. The information we have on ageing in elastic tissue in human skin is much less complete than that available for AcEiNG OF Elastic Tissue 101 arterial elastic tissue; the data are largely histocliemical. As a matter of fact there is some question in my mind as to whether or not the elastic tissue changes in senile elastosis should be referred to as an age change. In so far as can be determined senile elastosis occurs only in the dermis of skin that is chronically exposed to the elements, such as face, Table III Uptake of i* C Labelled Acetate in Normal Rabbit Aortic Elastin AND IN ElASTIN FROM THE AORTAS OF ElASTASE TREATED Rabbits Animal Exp. Dry wgt. Dry wgt. % Activity S.A. Number AORTA (mg) ELASTIN (mg) ELASTIN ct»/min ct »/min/ 100 mg 121 2 day 102 43 42.2 6.8 15.8 122 cont. 119 48 40.2 6.1 12.7 123 94 34 36.7 8.4 24. 3 124 116 45 38.8 6.4 14.2 125 135 52 38.6 39.3 13.7 26.7 18. 7 126 146 60 41.1 6.5 10.8 127 2 day 116 50 43.1 5.4 10.8 128 Elastase 131 56 42.8 5^7 10.2 129 109 41 37.7 8.% 20. S UO UO 41 40.0 40.9 7.3 16.6 13.8 21 94 32 34.0 361.6 1064 22 7 day 105 39 37.1 94.0 253 23 cont. 151 67 44.3 59.2 134 24 112 43 38.4 26.6 71.3 25 102 36 35.3 -37.a 380.5 26 169 63 37.3 30.5 81.7 27 7 day 126 48 38.2 18.0 47. 1 28 Elastase 126 54 42.9 14.8 34.5 29 142 51 35.9 14.5 40.3 30 135 55 40.8 12.2 29.8 TZ =— back of the neck, and hands; it does not occur in skin that is normally covered by clothing. It would seem that this elastic tissue lesion is a product of external stresses operating for long periods of time. In senile elastosis the normal bed of delicate elastic fibres in the pars papillaris of the dermis is replaced by a coarse mat of fibres 3-5 times thicker in diameter than normal. In severe lesions this mat may extend proximally to include the pars reticularis. These elastic fibres have several properties 102 Albert I. Lansing which distinguish them from the elastic elements found in senile arteries. The senile elastic tissue of arteries manifests a weak acidophilia whereas the opposite is true in senile elastosis. When the elastic fibres of the latter lesion are stained with a dye such as methylene blue or toluidine blue the fibres exhibit ? 100 ' Fig. 4. Graphs sliowing the uptake of ^*C labelled glycine in normal rabbit aortic elastin and in aortic elastin of rabbits treated with elastase. a strong basophilia. No evidence of metachromasia has been observed so far by us. Another striking difference between ageing of arterial and dermal elastic tissue exists in the calcium content of the two. As has already been indicated the ageing of arterial elastic tissue may be measured in terms of the amount of calcium it contains. Juvenile arterial elastic tissue is free of calcium; with age it progressively takes up calcium, as much as 18 per cent of the dry weight of the Ageing of Elastic Tissue 103 purified elastin. In micro-incinerated preparations the calcium in elastic fibres may be recognized as very heavy deposits of a brilliant white ash when the specimen is viewed with dark- field. We have made micro-incinerated sections of a number of specimens of skin exhibiting senile elastosis; to date we have found no evidence of a tendency toward mineralization of the elastic fibres in this lesion. There has been some question as to the nature of the fibres found in senile elastosis; all of the evidence found in our laboratory would establish that the fibres are elastic. They stain typically with the usual stains for elastic tissue, resorcin- fuchsin, orcein, and the Verhoeff reagent. Thin sections of osmic acid-fixed fibres of normal skin and of senile elastosis are indistinguishable; both are quite free of distinctive architecture. Lastly, incubation of small cubes of fresh skin exhibiting senile elastosis with elastase shows typical solubil- ization of the elastic elements therein. The digested areas are negative after Mallory or elastic tissue staining. It would appear that we are dealing with a bizarre type of elastic tissue. REFERENCES Balo, J., and Banga, I. (1950). Biochem. J., 46, 384. Balo, J., and Banga, I. (1953). Acta physiol. hung., 4, 187. CowDRY, E. V. (1933). Arteriosclerosis. New York: Macmillan Co. Hass, G. M. (1939). Arch. Path., 27, 334, 583. Lansing, A. I., Roberts, E., Ramasarma, G. B., Rosenthal, T. B., and Alex, M. (1951). Proc. Soc. exp. Biol. N.Y., 76, 714. Lansing, A. I., Rosenthal, T. B., Alex, M., and Dempsey, E. W. (1952). Anat. Rec, 114, 555. Lansing, A. I., Rosenthal, T. B., and Alex, M. (1953). Proc. Soc. exp. Biol., N.Y., 84, 689. LowRY, O. H., GiLLiGAN, D. R., and Katersky, E. M. (1941). J. biol. Chem., 139, 795. Neuberger, a., Perrone, J. C, and Slack, H. G. B. (1951). Biochem. J., 49, 199. Neuman, R. E., and Logan, M. A. (1950). J. biol. Chem., 186, 549. Roberts, E., Ramasarma, G. B., and Lewis, H. B. (1950). Proc. Soc. exp. Biol. N.Y., 74, 237. DISCUSSION McCance: I was fascinated by Dr. Lansing's communication because the possible effect of age on the chemical structure of proteins has 104 General Discussion always interested me. Foetal haemoglobin is quite a different proposition from adult human haemoglobin, and we know there are a number of proteins in the foetus which change biochemically but do not apparently change functionally with development. In the study of ageing might 1 call your attention to the possibihties of the placenta, which ages very rapidly and calcifies very rapidly also. You might get some interesting leads and connections betM^een foetal life and adult life by a study of the ageing placenta. Tunhridge: I am sorry Dr. Hall is not able to be here, because he has found a number of very interesting facts dealing with less drastic methods of getting elastin. I think Dr. Lansing will agree that even one-tenth normal sodium hydroxide is very drastic chemically. It will remove much of the non-protein material present, since it destroys mucopoly- saccharides. It will also, however, cause hydrolysis of peptide linkages in the main protein chains themselves and therefore it is surely not the best reagent for the preparation of a pure undegraded sample of "elastin". Using methods which, though still chemically drastic, are less likely to produce the effects of alkalis, we do find marked differences. F'irst of all the polysaccharide is retained in quit^ liigh concentration. We do not agree with Balo and Banga that the amount of carbohydrate is of the order of 10 per cent, because if you look at their methods, they have not allowed for the amino acid content of their hydrolysates, which would probably be included in any determinations of reducing sugars. In fact, after correspondence with them, they now agree that it is more of the order of 5 per cent. I suggest most tentatively that although the amino acid composition of so-called elastin and of collagen differ from one another, the main reason for their differing physical pro- perties lies elsewhere, probably in the fact that their polysaccharide content differs, both qualitatively and quantitatively. The difference in physical properties would suggest that elastin has a rubber-like con- stitution with enormous stretch right up to the point of breaking, as opposed to collagen which only stretches slightly before breaking. It is difficult to get pure specimens, and one hesitates to draw conclusions, but I do think Hall's work shows that there is a definite difference between the two proteins. It is not clearcut; the chemical differences when a more gentle method of separation is used are not as great as those usually suggested. It is fascinating, and it is in line with what McCance has said. Auh: I liked Dr. Lansing's presentation, but there is one question that puzzles me, and that is the extraordinary increase in calcium in this elastin. If I'm correct in my chemistry, calcium phosphate would still be intact in 1/10 n sodium hydroxide. You only spoke of calcium and not phosphate, but in your last slide I saw that phosphate also went up in your preparation. I wonder whether the calcium is really attached to the elastin or whether it is a residue of calcium phosphate or of an apatite which remains in your solution after it is digested. Then it would be just calcification remaining behind and not attached to the elastin. Its significance would be very different. Lansing: That is a question which, of course, came to our attention General Discussion 105 very early in the work, and I can say that the calcium exists largely as an apatite-like crystal. At least the ratios are proper for an apatite. I indicated, in referring to the dicarboxylic acids, that we may have only a primary linkage, meaning that perhaps the initial calcium may have gone to one of the carboxyl groups and then formed a nidus for the gro\\i;h of a crystal-like chain. Now it is not adventitious calcium phosphate derived from non-elastic tissue, because one can expose this calcified old elastin to dilute acids, and remove all of the calcium and then, in rendering the solution slightly alkaline, add calcium ions under controlled conditions and effect recombination. Not all of it goes back, we are not reconstituting the long chain calcium apatite but we do recombine a significant amount of calcium to the elastin. We can't do that with young elastin. So apparently there are groups in old elastin which have an affinity for calcium. Aub: But you get such a large percentage of calcium — a tremendous change. Is that all attached to your elastin? Could some of it be phosphate or apatite? Lansing: I think it is calcium phosphate. And, as I say, some of it at least is directly linked to the elastin molecule. How far the change goes in crystallization I can't say. Anatomically in micro-incineration all the mineral is directly in or on the elastic fibres. Visual observation, of course, doesn't establish chemical continuity, but anatomically all the calcium salts are in or on the elastic fibres when viewed in dark field micro-incinerations . Medawar: I'm interested in the idea raised by Lansing that the elastic fibre systems of the skin and of the blood vessels may have very different properties. Dr. Bean mentioned that in Osier's disease there was im- perfect development of elastic fibres in vessels; I presume they are all right in the skin. Is it not also the case that there are congenital affec- tions in which elastic fibres are very poorly developed in the skin but are normal in the arteries? If so, that would tend to reinforce the idea that we are really dealing with two somewhat independent fibre systems which might well be expected to undergo different types of senescent change. Tunbridge: If I may butt in here, I must cross swords with Lansing rather strongly on certain statements he has made. First of all, the amount of elastic tissue in normal skin is very small. The criteria for that are three-fold. One is qualitative and histological, which is always unreliable. The second is qualitative but makes use of more advanced techniques, particularly electron microscopy in which, as Lansing said, the pattern of collagen is very distinct, and the pattern of elastin, although amorphous, by that very fact is distinctive. In qualitative studies from all sites of the body we have found very little elastin— a finding substantiated by Wyckoff using the glass knife, w hich cuts a very thin section and minimizes the criticism that with the mechanical tear- ing techniques the elastic fibres might be lost. This was a serious technical criticism, in view of the widely held opinion that there were a lot of elastic fibres in ordinary skin. One of Prof. Astbury's workers has taken large masses of skin from different sites and dealt with it on a 106 General Discussion gross scale — ^a square foot of skin at a time — ^aiid obtained slightly higher results than those we have quoted earlier. But whichever way you look at it, be it by weight, volume or crude analysis, it would seem that at the most there is not more than 5 per cent, and probably less, elastin in dermal tissue. The other point is one which I thought we had proved, but Dr. Lansing disagrees with us: this question of senile elastosis. The elastic staining material is not elastic tissue, it is altered collagen — ^considerably altered, but still collagen. This answers Bean's point, I think, because the blood vessels are less well supported and they are very vulnerable to trauma. The extraordinary thing is that it increases with age, it is present in about 2-3 per cent of the population at the age of sixty, and in 25 per cent of all we have been able to examine at ninety; there is a linear relationship. With it goes scarring, tendency to haemorrhage and so on, but it is definitely altered collagen. Dr. von Albertini would agree that it is a totally different collagen degradation product from that wliich you get in the rheumatic nodules, both chemically and in its resistance to enzyme action. The lesion is usually said to be due to exposure, but we had one old lady of ninety who for various reasons had lived in hospital since the age of seven, and she had this change in a very striking degree — -where she got the sunlight in a hospital in Leeds, I do not know! Lansing: Well, apropos of your first point, there certainly is no issue on which to cross swords. I don't believe that there is a rich bed of elastic tissue in normal skin, and indeed the illustration I showed of the abdominal surface showed a remarkable lack of elastic elements, at least in the pars papillaris; there was rather a coarse net deep in the reticular portion of the skin, but I think we would all agree that there is probably less than 1 per cent extractable elastin in skin on a dry weight basis. We differ very sharply as to what this material is that we see in senile elastosis. This may be a matter of semantics or one of simple definition, but the fact is that this peculiar bed of material appears in the pars papillaris of human skin, usually but not necessarily in exposed portions of the body. This material stains like elastic tissue. We have very few criteria for defining elastic tissue; the best ones we have depend on tinctorial reactions. Affinity for orcein is rather specific, and this material stains rather well with orcein. The resorcin-fuchsin reagent stains elastic tissue quite specifically, it does not stain collagen, and this material in senile elastosis takes the resorcin-fuchsin dye. Similarly the Verhoeff's dye which I illustrated here, is a little less specific but stains effectively. So, tinctorially, this stuff behaves like elastic tissue. I pointed out that the elastase which is ratiier generally specific for elastic tissue, with one rather recent exception (the heat denatured collagen), responds to it, but within an hour after exposure to elastase at a pH of 9 • 0 the material that normally stains like elastic tissue in senile elastosis disappears entirely, it does not take up any of the elastic tissue dyes. And lastly, that material when exposed to the aniline blue or any of the Mallory reagents for collagen is entirely negative. General Discussion 107 Tunbridge: I tiike you up on two points there. Collagen aeted on chemically or enzymatically, for example with pepsin, will then take up stains which are said to be characteristic for elastin, so I think that classifying tissue as elastin on staining data alone is a generalization we should use with care. Then there is the question of enzyme digestion. The elastase which you mentioned, which Balo and Banga first isolated, definitely seems to have a dual function, at least one component appear- ing to be more in the nature of a mucase than anything else. The point about pH I want to raise particularly. Owing to the retention of sulphated polysaccharide in our elastin preparations and its subsequent release after elastase action there is a marked lowering in pH, the opti- mum effect being at two pH values, namely 8-7 and 7-8. The aorta acts very differently from ligamentum nuchae, which as you are aware has a much simpler form of elastic tissue, a smaller percentage of collagen interwoven with the elastic fibres, and what is more important a lower polysaccharide content. Therefore in the absence of adequate buffering you get far more acid liberated when you act on aortic media than you have from ligamentum nuchae, this may carry the pH from the optimum of one enzyme component (8 • 7) to that of the other (7-8) with a consequent alteration in the nature of the liberated products. In adequately buffered systems you find your elastase first liberates poly- saccharide and acid in a combination which is very closely allied to that of chondroitin sulphuric acid. You then get an action which is proteo- lytic in nature, and so far no elastase has been purified to such a degree as to be without this; its first action therefore is on the carbohydrate and secondly there is a delayed action on the protein itself. Lansing: You mean in England it hasn't been purified; it has in the States. Tunbridge: Well, we have been very careful, and Banga agrees with us on this entirely, that once you get rid of the polysaccharide, trypsin then has a most dramatic action. Lansing: I would like to reaffirm that this is material which morpho- logically looks like elastic tissue, tinctorially shows all the reactions of elastic tissue, and is entirely dissolved by elastase. These are the chief criteria we have for defining elastic tissue. It appears that there is presumptive evidence that this is elastic tissue. Comfort: I wonder if I might ask the combatants whether this material in the skin possesses structure under the electron microscope? Lansing: One slide I showed [not published] was of a section serially adjacent to a somewhat thicker specimen which was viewed in the phase contrast microscope to make sure we were dealing with the right area of senile elastosis. Comfort: And that is the material which you describe as elastin and Prof. Tunbridge as modified collagen? Lansing: Yes. Tunbridge: I think under the electron microscope this so-called elastin-like substance has a high percentage of broken fibres, an enormous amoimt of debris and amorphous material, and only when you have cleared it by enzymatic action with trypsin do you find a few normal 108 General Discussion collagen fibres remaining. So, under the electron microscope the struc- ture is very difterent from ordinary collagen with its perfect fibres — in bundles and cross striations. Albertini: May I show an electron microscopic pattern of a skin biopsy with senile keratosis and very typical so-called elastosis which we also call "elastoid degeneration". When I examined the slide in the light microscope I saw the same tremendous masses of "elastoid material" as Dr. Lansing has already shown. Using the low power of the electron microscope (Fig. 1) we again see the same masses. But with higher power (Fig. 2) it becomes obvious that this material must be degenerated collagen and not degenerated elastic fibres. Don't you think so? Lansing: But in between the bundles of collagen, do you think the tissue is actually empty there or do you think perhaps you have a rather fragmented section? Tunbridge: Of course, when one employs the electron microscope, obviously some of the spaces are artifacts due to the technique. Some may be intracellular substances and matrix of which we know very little and cannot get rid of. But I think that you can compare section with section. I do not know if Prof. Albertini will agree with me on that. Albertini: Yes, I do. Lansing: The very low level of electron density in the area represented by what I call space is awfully suggestive of naked collodion membrane and not ground substance which has a significant electron density. I think that we are not dealing here with a continuous section and I think the only way one can answer that would be to have two sections side by side — a thick section in the phase microscope and an adjacent serial section cut much thinner and introduced into the electron microscope to enable one to check area against area. That's what we've been doing and we don't get empty spaces. Cowdry: Dr. Lansing, could you say a word about purifying the enzyme? Lansing: Using organic solvents under standardized conditions of precipitation, we have a material that seems to be free of trypsin, and is well along towards purification, but it still has a long way to go. I think in another few months we may be able to talk more fuU}^ about it. Verzdr: I am glad that the discussion has now reached basic problems of the ageing of cells and even of proteins. I think that the admirable work of Dr. Lansing is really of a more general importance than when he speaks only of elastin and collagen. The loss of elasticity is so general in ageing tissue that one can find it in skin, in muscle, in blood vessels, and even in nerve. I think that this leads to the basically important question of the changes of tissue adaptation which might be caused by changes in the proteins. I would also like to say that a lot of time was spent in our laboratory with elastase estimations in young and old animals, and large differences in the pancreases of young and old animals were found but unfortunately they were not related to age. .• -%. Fig. 1 (.Vlbertini). Klastoid-deoenerated masses in the dermis (El. micr. < 2475). Fig. 2 (Albertini), Transforniatiou iVom coUatien into elastoid masses (Kl. micr. ■ 1 5()()()). [To face page 108 ^CALCIUM METABOLISM IN OLD AGE AS RELATED TO AGEING OF THE SKELETON O. J. Malm, m.d., R. Nicolaysen, m.d., and L. Skjelkvale Inst, for Ernceringsforsking, Oslo. Ageing of the skeleton implies dissolution of structural units (osteons and trabeculae) with loss of calcium salts from the body and with a reduction of the breaking strength. It is not known if the breaking strength is always proportionate to density of mineralisation. The systematics of bone diseases have been discussed in a number of text books and monographs, for example Albright and Reifenstein (1948). Senile osteoporosis is said to be frequent in old age, but no good statistics have been produced. The problem is how, if at all, this disease is related to dietary intake of calcium and vitamin D. Further, the course of calcium metabolism is not known during the development of the disease; but it must be negative over a period of years. Recently, the problem has in part been ably discussed by Hegsted, Moscoso and Collazos (1952), who consider that the calcium requirement of adult man is low and that the frequent occurrence of osteoporosis is not related to calcium intake. On the other hand numerous discussions of the Ca require- ment of man have ended with a recommendation for a defined intake. Lately the U.S. Food and Nutrition Board revised their recommendation of 1 g. to a rninimum of 0-8 g. daily for adult men. The implication must be that adult people consuming considerably less Ca than is recommended run the risk of getting into continuous negative Ca balance which in the course of time must lead to recognizable osteoporosis. Such differences of opinion can only be settled by adequate ♦This is to be considered as a preliminary report, presented by Prof. Nicolaysen. 109 110 O. J. Malm, R. Nicolaysen and L. Skjelkvale •and representative experimental work. Tlie problem of Ca balance and Ca requirement of elderly organisms has been subjected to a number of studies in recent years. According to McCay (1952) various species tend to get into negative Ca balance when about two thirds through the span of life. Liu and McCay (1953) found that they had trouble in maintaining metabolic equilibrium in old dogs unless the diet was relatively rich in calcium. Owen (1939) and Owen et at. (1940), in studies in old osteo- porotic men, observed a marked tendency for Ca retention. Bogonoff et al. (1953) studied the Ca balance in seven elderly men (sixty-six to eighty-three years of age) of whom three had osteoporosis. High, intermediate, and very low levels of Ca intake were used in periods of three to four weeks. The tendency to retention on high levels was small; but the loss on the low level (100-130 mg. daily) was also small (100-200 mg. daily). In contrast Ackermann and Toro (1953a), in a study of the Ca balance over fifty to seventy days in eight old men given only a high Ca diet, concluded that seven of the men needed 18-5 mg. Ca/kg. body weight to maintain equilibrium. The eighth subject lost 0- 5-1-0 g. Ca daily in spite of a very high level of Ca in the diet. In these experiments, as well as in earlier animal experi- ments, a high faecal Ca loss was observed; in other words absorption seemed to be deficient, whereas the urinary Ca in these old men was in the lower range of "normal". Schilling and Laszlo (1951) studied various bone diseases, among them seven senile osteoporotic cases. The urinary Ca was low, but they retained somewhat less of an intravenously injected amount of Ca salt than the normal controls (personal communication). In view of the fact (see below) that vitamin D is a dominant factor in Ca absorption in both adult and in old age, it is of interest to note that Ackermann and Toro (19536) observed a substantially increased Ca absorption in their subjects following vitamin D administration. Balance was achieved, but no retention, in the person who lost 0-5-1 -0 g. Ca daily. Calcium Metaboijsm in Old Age 111 The physiology of calcium metabolism was recently reviewed by Nicolaysen, Eeg-Larsen aiul Malm (1953), and the biochemistry and physiology of vitamin D by Nicolaysen and Eeg-Larsen (1953). Reference is made in these publications to a number of aspects of the field. In Oslo we have for several years studied the calcium metabolism of rats, and we have also conducted a long term study in man. The studies are in part finished, in part they are continuing. The main purpose here is to report briefly on these experiments, which will be published in full at a later date. Experiments in Rats The object has been to study the adaptation of the aVjsorp- tion of calcium in adult and old rats. Some of the characteristic results are reported below. Experimental. Adult rats of known age taken from our stock colony and given a vitamin D free diet (842 g. whole wheat, 100 g. casein [Ca "free" and fat free], 30 g. dried brewers yeast, 10 g. NaCl, 50 g. B12 "mixture", 8 g. KH2PO4, 50 g. CaCOg), with a Ca content of 0 • 25 per cent. At intervals the CaCOg is removed and the diet then contains 0-03-0 -04 per cent Ca. With the aid of 0 • 5 per cent Na oxalate added to this Ca poor diet the animals are at given intervals deprived of body calcium. The Ca absorption or the Ca balance is studied continu- ously. In Figs. 1 and 2 two experiments are reported, one on a D free diet to which vitamin D was added when the animals had suffered a measured loss of Ca, the second in very old rats who received 0*5 i.u. vitamin D daily throughout. Results The contrast between the group of rats given a vitamin D free diet (Fig. 1) in the periods following Ca deprivation (but before vitamin D was given), and the rats given vitamin D throughout (Fig. 2) is striking. Thus rats without vitamin D can lose 15-20 per cent of their Ca and the net absorption still remains very low. When vitamin D is given to such rats 112 O. J. Mai.m, R; Nicola ysen and L. Skjelkvale the absorption quickly reaches very high values. This result has been confirmed in a number of other experiments. The results achieved in the very old (thirty-one months) rats have so far been confirmed in fifteen to twenty months old rats. Vitamin D free diet ma. Ca/day 20 jVit. D 20 m./day| absorbed total Ca loss 610 m^. 640 ing. H -^640 — -4 0.25 % 0.04% leveb c£ Ca in the diet 30(^ A^e 51 weeks 300^. body 85 we«ks wt. Fig. 1. The absorption of Ca in six female rats. (Ca intake on the diet containing 0-25 per cent Ca, about 40 nig. daily.) The urinary Ca was measured throughout in the very old rats. The daily output was about 0-5 mg. in the period of deprivation. In the following two weeks it increased to about 2 mg. daily. Thus nearly all Ca absorbed was retained. The loss of bone minerals suffered in the period of deprivation could quickly be made good. Calcium Metabolism in Old Age 113 It is worth while reporting that the technique used in the above experiments matured in consequence of a great number of experiments in which adaptation (significantly increased absorption) could not be observed. In these earlier experi- ments rats given vitamin D throughout were used. Short Age I 30 months m^. Ca/day 22 T absorbed 16 6- O ■4 ■8H ■12 -293 jng. 0.25 % 0.04% levels of Ca in the diet Fig. 2. The Ca absorption in six old rats given 0-5 I.U. vitamin D daily. (Ca intake on the 0 • 25 per cent Ca diet, about 40 mg. daily.) periods of ten to fourteen days on the 0 • 25 per cent Ca diet were used to test the absorption. In between the rats were given the 0-04 per cent Ca diet for weeks and months. Some suffered losses of 50-100 mg. (2-4 per cent of the total), but this did not result in a clearcut increase of the absorption from the 0 • 25 per cent diet. Other groups of rats very quickly reduced their faecal output to very low values, so that they hardly lost any body Ca at all. 114 O. J. Malm, R. Nicola ysen and L. Skjelkvale As a result, we introduced oxalate as a Ca "trapper". It needs emphasis, that the lack of adaptation of tlie absorption in rats is in reality nothing but a confirmation of the principle which can be extracted from all earlier observations in vitamin D deficiency in man, not least from results in late rickets and in osteomalacia (see Nicolaysen and Eeg-Larsen, 1953, for the metabolic pattern). Experiments in Men These were long term balance experiments conducted in our main prison. Only several consecutive two- week periods were used for calculations, and all observations were based on analyses of food as served and eaten. The procedure was that any person taken for experiment was first observed for a period ranging from some months to a year on the level of Ca intake of about 900 mg. daily. When the Ca "charac- teristic" of a given person had thus been established at this level, the Ca intake was reduced. The vitamin D intake was at least 200 i.u. daily. Results In all thirty-eight men were studied. However, some were released before an adaptation experiment could be conducted over a sufficient period of time. Therefore only twenty-five out of the thirty-eight were subjected to such a procedure. In Fig. 3 the balances of all thirty-eight men have been given, and in Table I are condensed and summarized the results of the adaptation study in twenty-five men. In Fig. 3 the figures have been arranged according to age, whereas in Table I they are according to increased, calculated requirement. Most of the men were overweight according to the Metro- politan Insurance Company's standard table which was used for correction. The figures for urine and faecal Ca are given with their standard errors. In some the variations were irregular, but in other persons long waves in the faecal Ca output occurred at one level of intake. A definite trend of the curves for Calcium Metaboijsm in Old Age 115 urinary and faecal Ca occurred when the intake was reduced. In such instances the standard deviation is an exaggeration of the error of the single observations. The method of calculation used for estimated requirement rested on a number of assumptions and approximations. age 20 50 60 80 years wet wt. 20 10 30o-| mgCii/day 100 0 - -100-" age 20 "true" absorption -balance 30 40 50 60 70 80 years. 42-68 44-G6 54-67 43-74 51-69 % "True" absorption calculated accorcjing to formula (Nicola vsen et ah, 1953, Physiol. Rev., 33, 424): (■-r Ca (feces) a intake -\- Ca secreted *500 nig. used here. ») 100 Fig. 3. Top: Asb content of wbole corp. vertebr.T in men killed by accidents. BcUnv: Calcium balances in 38 men. Continued observations over 100-500 days; daily intake about 900 mg. Ca. 116 O. J. Malm, R. Nicolaysen and L. Skjelkvale C 10 C 05 "f 01 GO >0 ^ Oi ^ CO CO 00 01 © T}( CD 01 05 OO ^ CD X if i^ CO 00 CO »0 O »0 r-H -t o 01 2 2 01 UO © 01 rH O 01 X ■CO 05 O CO' 01 CO rH CO z -fl-H 4 ■H-H H-l-l-H (1 f H 1 -II -II-HH4HI4I4^-H -W \\-H-\\ 1 ^?5 CO CO O 05 CO ® i 4j O CO CO rH Tfl QO rH rj( 01 CO +1^ »o X t-XOIOICOtO-"*-* ^0 05 >0 ^ -H -H ~H -H •2 r-* CO »0 JO O i- CD I- o >0 >0 i ■<* 05 CO I- CD^J>i>0105©X rH >f5 CO rH 05 © i> CO rH O 05 CO »ra C5 fl5 J>. ~J u " 01 '"' ^ 01 CO 01 in c5 1^ "-" ^SSIh ill 01 CO is^L^^ 01 l^ rH X iO >0 CO CO CD t- 05 © 05 J> CO IC o :-^- 00 r-^ 01 CO t^ i- 10 »0 X 05 © rH ^r -^ '0 to •o •0 >0 05 i> »0 >0 CO CO ^ LO J> r^ 05 >0 l^ Ci O © 1> o o © © © © © © lO IC rj* 01 r> %i 1 1 01 X O O i> 1 1 1 1 o X o 01 i^ CO 01 ^ S 2 g ® 2 ^^ 1 1 ' CO 01 CO rH 1 1 1 r^ ■^ ss OO O rH ^ J> -^ ^ in '^ 01 Oi 05 © 05 Oi »0 rH CO 05 CD X 01 X 1 (M » 1 1 CO 1 CO ^ o Tt^ tP I- X I- i> ^ r-{ CO 01 CO rH ^ 1 1 ' ' 1 ^ 1 1 1 1 1 1 It ::t 2 C O 01 QO X I- Ci X CO © CO CO © «D O X CO in 01 01 l> © 'i? :t r-' rH ig c^ CO CO CO -f i^ »n »0 >0 »0 CO X © rH »o 01 © CO "O ■^s 'f -* ^ CO CO ^ -+ o -^ 05 -^ 05 05 »0 CO 'f oo ^ 't 05 Tji ^g, S5 3; >£ (M O O CO CD -t X CO X X 01©CDXCO©CD© X ■^ X ■^ X CO 01 -f CO C0»0C0©O^CD'f' C5 X O X S"2 't »0 CO CO 0-1 01 rH (M ^ "2 ■.: + + + + + + + + + + ^*^ ■^ .S£ Tf lO CO © © © to © >o P° :** 05 01 »^ »o o C^-o t> t> t- CO o ^^^ |.i o SI CI CO 'f ^ CO »g X © -^ ^^ CO CO 00 ■^ ^ o o 01 CO »o co§ c c c a r; 0^ ESC • © o ''t* ^ 01 rH »0 05 << Ki^ Ct) o CO 01 05 ■^ '^ rH^ W CO CO 01 Calcium Metabolism in Old Age 117 c CO o:r-t 01 I- O I- I- >o I- .0 05 C 01 I- rH -^ CO rH »0 CO >0 >0 01 O 01 01 c: c '-' o 05 '^ O 1> LT r- r^ »0 CO -f l^ rH O X 05 Ol 0 0 05 Tp rH I— 1 I— < r-i Ol 01 T-H T^ ,-H Ol Ol Ol Ol r- CO Ol 01 •^ 01 CO -t -H-4H -H-H-H ^4^-IH-H-H 41 -H -i\^ 41 -H-^ 41 -H 4141^ -W41-H4I O GO O i-O (X> >0 01 ;0 05 i:- o c 05 05 01 01 c o o 05 ^ ^ b- Tj< b- 01 t- •^ O X Tf O Ol 05 Tf* X rH C! rH O X CO X 05 lO I- 01 rH 10 10 0 »o -* 01 01 O CO Ol 01 r- CO 01 I- CO CO CO 01 I- CO I- ^ CO -H CO 01 ^ Tt C I- i> Ol 01 C X I- o Ol X l> CO rH 05 »0 X CO CO 05 CO ■>* 01 if: ^ 00 Ci rt TT t^ Tf CO CO 01 X O CO l> »o »o X i> »o 0 05 1> 10 rH rH ^^ -H-H+l -H^-H4H-H ^^^^ ^4H ^4H41 4141 414141 41-^4141 '* o o o c X »0 i-O 01 tJi 01 »o C Tf CO CO O »o o X t- 05 01 Tj. 0 X as i> CO 01 c CO C c o « 10 05 I- X -f -f< X CO o I- CO Tt i> »0 CO >o 01 01 Ol 01 01 01 Ol 01 0^ r-* 01 01 01 01 10 01 I- 01 o t^ ».o 05 -f> C t- r-i rH X 't 01 rH CO CO CO »o t- ^ 05 X C -* o ;0 i^ i^ lO C' ;o I- X »o 1> »0 CO >0 CO I- rf CO ^ to CO LO I- I- CO Ol 01 -f 10 CO CO „ „ CO 05 o o o » CO CO I- I- I- l^ O 01 o ^ o o o CO rH CO o c c o coo o c coo 000 01 Ol ^ ^ o rH »0 lO rH rH Ol lO 05 Tjl >0 CO rH 01 CO CO CO Ol 0 05 05 CO 1 CO 01 rH 1 1 ^ 1 1 00 1 1 17 1 1 1 CO CO C C Ol 05 O »0 X T^ CO I- I- Ol t- X 05 O CO 0^0 0 01 I- rH CO 01 10 rH ;0 O O ^ rH i> »o X CO CO o 01 Tj< 01 ^ r- X 05 X ' 1 1 1 1 1 - 1 1 171 t 1 »0 X 01 ^ CO 05 I- »0 X CO rH »0 O 05 rH 'f 05 CO 0 rH O Oi Oi 01 rH •^ Ol O »0 CO CO 05 t- CO CO lO CO X »o CO »o »0 01 01 05 05 05 CO X O lO 05 ^ Tf T}< Tj« 05 CO 05 ^ 05 Tf* Tjt 05^ Oi ^ ^ cc O -f 01 o o O O CO O X 01 't X CO Ol -f X X o 01 X X -t 01 -t 0 t- 00 o <* -* rH ^ Ol Tfl 05 X X 05 Ol »0 X 05 05 I- rH 05 CO 01 rH X I- Ol r^ rH 01 r-i r-t 01 01 r^ •i + ^.2 + CO 1 o ^ ® ^ + + o + + oti Ci o ^ o rH CO CO I- X - ^ CO »o i- I- a « fc Cfi l^ ^f -^ I- CO i> CO 0 J-' c ^' Oh > o rr. ^ 'k t^ o QO 43 Cu o CO 01 Ol >o »0 t- 01 «l Tf^ 01 CO '^ Ol lO 00 ^ d >J 5 £ a CO c a 05 01 d 0 o fi 12^8 ^ Ol 01 o 0« »o ^ 2 »o CO CO 01 o o c^ s X CLh z > p ij II II II + -H I II S a s 118 O. J. Malm, R. Nicolaysen and L. Skjelkvale 5; 1 CO C5 o CO lO b. »o o rH X CO '^ CO X f^ 10 00 F^ 01 b- lO O rl< lO CO rH rH rH rH t> S- (M rH rH l-H 01 ^ * ^^^^4^+^ -H-4H-H 4^^-H ^^7^5 -^^-H P GO CO ■* CO Ci I- Tj* o o o »o CO CO X rH 01 »0 O X 05 1 '^ (N 05 •^ ^ ^ l> CO i> I- ^ X » »o ^ ^ »0 CO CO O CO CO CO 01 rH Si X C^ 05 •<* ^ 01 05 Tf 05 05 I- 05 O CO X o CO 10 X o '^ 05 05 ^ CO 01 -^ 00 r^ O CO o 1 T—i Ol r-t 4^-+^4^-+^ 5-H4H -H4^4^ -H+l-^ ^^-n7^ s O O J> 00 X 05 Ci t- C 01 l> 05 X O ^ X "£ O 00 i> »o O CO X ^ 05 ^ ^^^ 01 »0 lO ^ I— 1 I— 1 I— 1 I— ( (M (N rH Ol rH 01 CO CO CO ,•1^ ^sj-g o »o Ci CO 01 X rH O 01 J> Tf 05 '^ 05 ■* 01 lO lO »o o o -o o »o o » I- l> X »0 CO i> a,a c^ 01 X ? t , X X 05 ;J ^'^ 00 00 00 00 00 05 01 ^ 05 ft 1-^ 1-i ^ti lO '^ »0 M C r^ O :| to i> CO CO Oi O I- »0 01 05 r-^ » CO O lO 'f^ to (M O OJ C X t- 05 rH I- 01 -^ Ol ^ c 1 1 ^ 1 1 7 1 -" 1 '7 It CO to X -^ CO 05 -i* »0 rH ,H 05 »0 O O t' CO Q -ii rH X O X Ol I- o »0 01 lO O 05 X 05 lO '* 3i 05 I- »0 "O o; '^ >o 05 ^ '^ 05 10 lO X i> »o II 01 X X '^ Ol o o « O 't O Tf X -f X o t^ O 05 X X -* 01 05 -^ X rH 01 05 05 CO X h.-o 0,^ Ol 01 01 r-i 01 1^. + + + -H + r-^ X o r-^ ,00 "* ^■g-* b- -^ 00 01 o 05 O-o ® i> o "U CO o- o — .22 ^.e 05 CO ^ «^^ X o -^1 01 »0 CO d ^1 c a a ^g 8 1 X X y-l §55 8 rH c/3 s ° § so Calcium Metabolism in Old Age 119 One assumption obviously not completely valid was that adaptation had been fully developed and that it had reached its maximal power. The three approximations used were the following: 1. The "true" percentage absorption was calculated according to formulae (see Fig. 1 and Nicolaysen, Eeg-Larsen and Malm, 1953). 2. It was assumed that this calculated percentage would be the same when the level of intake varied somewhat up or down. Two examples will illustrate the procedure. 1. No. 273: 20 mg. — — =28 mg. which added to the 349 mg. results in the figure used. 2. No. 121: 49 mg. ——=69 mg. which subtracted from 450 mg. results in the figure used. 3. In a few persons a slight linear correction up or down of the urinary Ca was used in consequence of a more substantial correction needed for intake. (See e.g. 92 and 408c.) Thus in 408c., 25 mg. were added to the 186 mg. The balance was measured in fourteen men above fifty years of age and adaptation was studied in nine men above fifty years of age. Discussion Absorption and its adaptation It appears that the absorption is more efficient than usually thought. The introduction of e.g. 300 and 900 mg. respectively for the daily Ca content of the digestive juices will not materially alter this contention. It also appears that the power of absorption is increased in the course of time in practically all persons studied. This contention rests on the assumption that the digestive juice Ca is not systematically altered in a period of negative balance 120 O. J. Malm, R. Nicolaysen and L. Skjelkvale turning positive. Anyhow, it seems reasonable in view of the undoubtedly established adaptation of absorption in rats, to assume that an increase of the power of absorption is responsible for most of the adaptation of the absorption. Persons above fifty years of age show on an average no less ability to adapt their absorption than younger persons. The results of Bogonoff et al. (1953) in old men on very low Ca diets strengthened this view. Urinary Ca and its adaptation The quantitative aspects have recently been discussed (Nicolaysen et al., 1953), and the linearity within individuals as well as the variability between individuals has been stressed. In a few persons no reduction of the urinary output follows halving of the intake, but in most persons some reduction follows, and in some a substantial reduction is responsible for quite a large part of the quick achievement of balance following the reduction of Ca intake. However, when two consecutive periods on a low level of intake are compared, the general impression is that absorption improves in nearly all, whereas the urinary Ca level mostly remains horizontal. Ca requirement The validity of the figures rests entirely upon the method of study. The material here presented reveals the number of weaknesses inherent in such a study of arbitrarily chosen subjects of mostly unknown previous dietary habits and status of nutrition as regards Ca and vitamin D. Judged from the number of persons able to retain substantial amounts of Ca it may be said that they did not represent a population well nourished as regards nutrients essential to maximal skeletal calcification. On the other hand such a contention would lend support to the hypothesis that the limit of adapta- tion may have been reached in some. Calcium Metabolism in Old Age 121 The approximations used cannot lead to serious errors affecting the main results. In the main most of the figures for Ca requirement support the view of Hegsted et al. (1952) that Ca requirement for maintenance is low in men. On the other hand the higher figures reached are at variance with such a view and some comments on these figures are needed. The experiment in No. 551 over twenty-six months leads to a comparatively well established figure of 12-8 mg. Is he an exception in the material presented? In view of the results in No. 242, the figure reached in No. 288 can be quite off the mark, and the figure reached for No. 408c is not well founded due to the "abnormal" figure for faecal Ca in the third period. In the course of time this high faecal output would have been followed by an inverse wave, with a much lower figure for Ca require- ment as the result. On the other hand the figure of 9 mg. for No. 200 would seem to be not too poorly founded. Elementary statistics on the requirement figures for the first 24 persons results in a mean of 6 -3 mg./kg. with s.d. of 2 • 1. This would imply that No. 551 is one in a thousand but it does not solve the problem. The men above fifty years of age seem not to behave differently as regards adaptation, and it appears from Fig. 3 that a number of them retained Ca remarkably well. The development of osteoporosis implies loss of calcium from the skeleton, either by a real increase of urinary Ca, or by a non- reduction of urinary Ca combined with faulty absorption. The results in No. 790 on the highest intake level are instructive as to how the absorption may seem to be low when only high levels of Ca in the diets are used, and this may throw some additional light on the apparent faulty absorption in the old persons of Ackermann and Toro (1953). The results of Bogonoff et al. (1953) on very low levels of Ca in the diet are in line with the results here presented. As regards urinary Ca the variability between individuals makes it impossible to judge if the urinary Ca in a given person has been increased. Observations over many years 122 O. J. Malm, R. Nicola ysen and L. Skjelkvale would be required. On the other hand the considerable material of Ohlson et al. (1952) in women can be included in the discussion, and this has recently been done (Nicolaysen et al., 1953). Nothing indicates that urinary Ca is increased in elderly men and women. However, such a contention is far from conclusive, because comparison between indivi- duals suffers from great variability. No. 200 as well as No. 551 were not only high urinary excretors, but they seemed to keep the high urinary Ca level with tenacity. It may be that they represent the type that in the course of time may develop senile osteoporosis with hospitalisation as a consequence. The two different types of observations presented in Fig. 3 may indicate that the men above fifty, who retained Ca so well, did so because they had been somewhat demineralized in the years prior to the actual balance study. Nothing in the material here presented indicates that the Ca metabolism of men above fifty differs from that in younger men. On the other hand the limitations of the material are obvious, so that it does not allow of any final conclusion as regards the origin of senile osteoporosis as related to dietary intake of calcium and vitamin D. It does not seem profitable to continue to w^ork with an arbitrarily chosen material; however, adequate balance studies in a selected material of senile osteoporotics combined with the load method of Schilling and Lazslo (1951) might in the course of some years contribute definitely to the solution. REFERENCES ACKERMANN, P.G., and ToRO, G. (1953a). J. Gerotit., 8, 289. AcKERMANN, P.O., and ToRO, G. (19536). J. Geront., 8, 451. Albright, F., and Reifenstein, E. C. (1948). The Parathyroid Glands and Metabolic Bone Disease. Baltimore: Williams & Wilkins. BoGONOFF, M. D., Shock, N. W., and Nichols, M. P. (1953). J. Geront., 8, 272. Hegsted, D. M., Moscoso, I., and Collazos, Ch. J. I. (1952). J. Nutrit., 46, 181. Liu, C. H., and McCay, C. M. (1953). J. Geront., 8, 264. General Discussion 123 McCay, C. M. (1952). Problems of ageiiijj, 3rd edition, edited by A. I. Lansing, p. 1G9. Baltimore: Williams & Wilkins. KicoLAYSEN, R., and Eeg-Larsen, N. (1953). Vitamins and Hormones, 11, 29. New York: Academic Press. Nicola YSEN, R., Eeg-Larsen, N., and Malm, O. J. (1953). Physiol. Rev., 33, 424. Ohlson, M. a., Brewer, W. D., Jackson, L., Swanson, P. P., Roberts, P. H., Mangel, M., Leverton, R. M., Chaloupka, M., Gram, M. R., and Lutz, R. (1952). Fed. Proc, 11, 775. Owen, E. C. (1939). Biochem. J., 33, 22. Owen, E. C, Irving, J. T., and Luall, A. (1940). Act. med. scand., 103, 235. Schilling, A., and Laszlo, D. (1951). Proc. Soc. exp. Biol. N.Y., 78, 286. DISCUSSION Brull: Urinary calcium, as far as I can see, is rather a leakage, and is more or less independent of the rest of the calcimn balance, and much more influenced by diuresis and organic acid metabolism than by any- thing else, except in disease of the parathyroid gland. What do you think about senile osteoporosis, when it has gone so far that the vertebrae break down? Do you believe that it is due to lack of intake or do you think there is another factor? Nicolaysen: My opinion changes from day to day! I think you just have to do long-term studies. There is material indicating that old osteoporotic men retain calcium remarkably well, but that may again be over shorter periods of time. We may have a certain capacity for repair but not a complete one; but it may be that the periods of observa- tion are not long enough. Brull: I have done many calcium balances, and I have now some doubt about the value of them at all. If you take a case of osteoporosis, and do the calcium balance, you very often find normal requirements. If you investigate their background, you find that some people who become osteoporotic have had a normal calcium intake, and others, who have no disease, had a low calcium intake. Then there is the striking fact that when you take a patient who has just a normal balance but who has osteoporosis, and you give him very small doses of follicular hormone, there are enormous changes in calcium balance. This suggests another origin of osteoporosis. Nicolaysen: Well, in ordinary persons you can introduce new matrix with the aid of sex hormones, you can force calcium deposition. Brull: You mentioned the problem of the requirement of vitamin D in adults and old people. As far as we could see the requirement of vitamin D in adults and old people is about 3-4 /xg./day. Nicolaysen: I agree with you that they need vitamin D, but I am not sure how much they need; I think that remains to be discovered. Brull: There is a very sensitive method to find it out, and that is by the determination of the proportion of phosphorus in the faeces and urine. 124 General Discussion Tlie amount of phosphates in urine is above that in the faices when the intake of vitamin D is sufficient, and the faeces content goes above that of the urine when the intake of vitamin D is below normal requirements. As soon as you give the necessary amount of vitamin D (about 4 ;ug./day), then the proportion of phosphates in the urine is above the phosphate in the faeces. That is a very sensitive method in man. In vitamin D deficiency, when practically no phosphates are absorbed, vitamin D can produce a tremendous increase in phosphate absorption with no effect on calcium. Nicolaysen: I am a little sceptical of that method. I have done a great many studies on the ratio of calcium to phosphate in the faeces, and it only tells me that if there is no phytic acid in the faeces, the phosphate varies with the calcium. Auh: In the faeces, yes, but not in the urine. Phosphate in the urine is a very complicated problem — ^calcium excretion appears simpler, or at least is better understood. There may be a very low phosphate in the blood stream but a very large excretion in the urine, and the reverse situation may also occur. The differences appear to be dependent largely on the amount of reabsorption in the renal tubules. Brull: May I remind you that I demonstrated a threshold for phos- phate about twenty-five years ago, and the influence of hormones and other factors upon it? Nicolaysen: If you were to accumulate the phosphates in the faeces in a soluble form you would soon get diarrhoea. Auh: I agree that phosphate excretion appears to depend on the amount of calcium in the faeces. It appears in the faeces largely as dibasic calcium phosphate, but it doesn't always appear in the urine in that way, the excretion there appears more complicated. McCay: It is interesting that you have found vitamin D of importance. We have tried for many years, off and on, and we have never been able to show a vitamin D requirement in old rats. You have different diets and conditions, and it shows that results vary enormously under different experimental conditions. Nicolaysen: We have conducted a life-time study in rats which is instructive on this point. The diet was free of vitamin D but when vitamin D was added optimal growth was seen (in the D-free rats the final body weight was 10 per cent lower). The calcium balance was studied continuously over eighteen months. The difference in the absorption of calcium between the +D group and the D-free group disappeared after twelve weeks of experiment. The total calcium retention per 100 g. body weight was about 10 per cent lower in the D-free rats. Vitamin D given to such D-free rats resulted in a con- siderably increased absorption. This was true for rats at the age of thirty-five weeks as well as at seventy-five weeks. The amount of Ca absorbed was many times higher than that seen in rats of the same age which had been given vitamin D throughout. McCay: Absorption from the intestine is confirmed very well by radio- active calcium work. There seems to be no difference in absorption of radioactive calcium from the tracts of young and old dogs or young and General Discussion 125 old rats, but there is a tremendous differenee in the rate at whieh that calcium mf)ves out of the blood, ft is out of the blood within twenty- four hours in a youno- rat, but you can still measure it very easily at the end of a week in an old rat or dog. So the absorption seems to lie be- tween the blood vessel and the bone rather than between the intestinal wall and the blood. NicoUujsen: Yes, absorption is different in young and old adults. The removal from the blood is very complicated, as you know; it is worked out in an equation with four exponential terms. The first two are identical to those for the removal for sodium, then follows the slower one, which is a summary of ionic exchanges and recrystallization, and the last one represents the term excretion. McCay: There is a problem of adaptation in the rat: if one maintains rats from early youth on a low calcium diet, one finds that they go into negative calcium balance later in old age, and it is easier to maintain their calcium in old age than if they are from early youth on high calcium diets. And also, if rats are fed diets very rich in milk, one finds that their bones are denser and have more calcium when they die in old age and they excrete more calcium in their urine than if the diet had some other source of calcium like bone meal. So there must be other factors in milk that modify this utilization of calcium in old age, don't you think? Nicolaysen: In three years of preliminar}^ experiments we studied the adaptation of Ca absorption in adult rats. The absorption was tested on a 0 -25 per cent Ca diet following intervals of up to four months on a 0 • 04 per cent Ca diet, in the course of w hich the rats lost 2-4 per cent body Ca. However, we could in no instance find an improved absorption. Therefore, we introduced a more severe Ca deprivation, either by giving a diet free of Ca or by giving the 0 • 04 per cent Ca diet with added oxalate. Now the losses increased above 5 per cent, and in the foUow^ing weeks on the 0-25 per cent diet a greatly increased speed of absorption was seen, falling off successively in a staircase-like way. Thus adaptation had been clearly developed. Vitamin D was given to all these rats throughout. Auh: If you are looking for osteoporosis, why do you use men instead of women? Isn't it somewhat rarer in men? Nicolaysen: Well, an eminent clinician told me he had seen it in many men, and again on the way over here I asked the Head of a hospital for old people, and he said that it was very frequent in old men. Tunbridge: I think Prof. Nicolaysen htis fully demonstrated the need for longitudinal studies as well as longevity studies. 17-KETOSTEROID EXCRETION IN AGEING SUBJECTS* Betty L. Rubin, ph.d., R. I. Dorfman, Ph.D., AND G. PiNCUS, D.Sc. Worcester Foundation for Experimental Biology, Shreivsbury, Mass. Many studies of the changes in adrenal function with increasing age or with disease have been based on the indices of total urinary corticoid excretion and total urinary 17- ketosteroid excretion. They have demonstrated a decreased excretion of both classes of steroid with increasing age. How- ever, these decreases are not parallel. To cite only one study, that of Pincus, RomanofP and Carlo (1954), carried out at the Worcester Foundation on men and women aged twenty to ninety, it was found that the 17-ketosteroids decreased more markedly with age than did the corticoids. Two recent developments have made it possible to derive more information about the function of the adrenal gland in aged individuals from a more detailed study of urinary 17- ketosteroids. First, metabolism studies with the steroid hormones have established the relationships between endo- genous steroids and the individual 17-ketosteroids of the urine, and secondly, methods have been developed which make possible the quantitative determination of six indivi- dual a- 17-ketosteroids in a urine extract containing as little as 2-5 mg. of Zimmermann-reacting material. The metabolic relationships for which there is experimental evidence are presented in Figs. 1 and 2. Fig. 1 shows what we consider to be the principal pathways in the biosynthesis of the biologically active adrenal steroids. Cholesterol, a Cgg steroid with the 3/3-hydroxy-zl^ configuration, and other ♦Investigations described in this paper have been supported by grant number H-1659(C6) from the U.S. Pubhc Health Service. 126 17-Ketosteroid Excretion in Ageing Subjects 127 128 Betty L. Rubin, R. I. Dorfman and G. Pincus precursors, sucli as acetate, first are converted to androst-5- en-3j8-ol-17-one (dehydroepiandrosterone), a Cjg steroid, and pregn-5-en-3j8-ol-20-one, a C21 steroid, both still having the 3jS-hydroxy-z]^ structure. These are then oxidized to the corresponding Zl*-3-ketones, such as androst-4-ene-3,17-dione and pregn-4-ene-3,20-dione (progesterone). Oxygenation at carbons 11, 17, and 21 takes place after the oxidation of ring A, giving rise to androst-4-en-llj3-ol-3,17-dione in the C^g series, and in the C21 series through six compounds containing, in various combinations, one or two more hydroxyl groups than progesterone, to the tri-hydroxy steroid pregn-4-ene-llj8, 17a,21-triol-3,20-dione (Cortisol). Among the compounds intermediate between progesterone and Cortisol, those which can contribute to the urinary 17-ketosteroids are pregn-4-en- 17a-ol-3,20-dione (17-hydroxyprogesterone), pregn-4-ene-llj8, 17a-diol-3,20-dione, and pregn-4-ene-17a,21-diol-3,20-dione ( 1 1 -desoxy Cortisol) . Fig. 2 shows the relationships between these steroids and the urinary 17-ketosteroids. Androsterone and aetiocholan- 3a-ol-17-one excretion reflect the adrenal production of the C19O2 steroids, androst-4-ene-3,17-dione and dehydroepi- androsterone (and in the male, testosterone produced by the gonads), with a minor contribution from C21O4 steroids, such as 11-desoxycortisol. An indication of the production by the adrenal of C19O3 steroids, such as androst-4-en-llj8- ol-3,17-dione, can be obtained from the urinary excretion of 11 -oxygenated steroids having the 5a (or androstane) con- figuration, while excretion of 11 -oxygenated 5p steroids (aetiocholane configuration) may be regarded as an index of production of C21O5 steroids, the major one being Cortisol. The quantitative method used at the Worcester Foundation for analysing urine extracts may be briefly described. Urines were collected as twenty-four hour specimens and hydrolyzed by boiling for eight minutes with 15 volumes per- cent of hydrochloric acid. They were then extracted with ether and fractionated according to the method of Pincus (1945). After separation with the Girard reagent (Girard 17-Ketosteroid Excretion in Ageing Subjects 129 Androst-4-gne-3,17-dione 0^' (lldesoxy-corfisolj y Anc[rost-1cn-![/3-o(5j7dinr]e CWzOH Prggn-4-m-ll5J7c(,Z!-tnol-5,20-d(one (Corttsolj •^ HO ^ H Andro5ten-3c(-ol-l7-one (Androsterone) .0 /ttiocho[an-5o?.-3UV}sojipuY b -fl b b 4^ b b -H ? b b b ? b -H b b -fl b I b -H b o b 4^ CO CO b 9U0-11 05 b -H OS M b 4^ CO o CO CO b 4^ OS b b -fl 00 b -H «o b 4^ CI o 4^ OS b 3UOX9)SOXpUy CO b ■+^ CO b -H CO CO 00 +1 b -H b W5 b CO M5 b o CO b 4^ b 4^ CO b SN-ii-"" imox C5 b b o o b OS b CO b 4^ 'J CO o I- 00 b 4^ CO 00 01 CO b 4^ CO s}D3Cqng fo xdqiunj^ (N oo CO >o 01 ^ CO 00 9Suv}i aSy OS ! «5 7 g W5 CO 1 01 CO X X9S :? s S s fe fe fe fe 17-Ketostekoii) Excretion in Aceing Subjects 133 androstane-3a,ll/S-diol-17-oiie, 0-2-0 -3 mo-, as aiulrostan-3a- ol-ll,17-dione, 0-24-0 -34 mg. as 8etiocholane-3a-ll/3-diol-17- one, and 0-41-0-52 mg. as 3etiocholan-3a-ol-ll,17-dione. These figures are in agreement with Dobriner (1953) except for 8etiocholane-3a-ll^-diol-17-one excretion. In older people, the total excretion has fallen to 3-2-3-4 mg. Androsterone and 8etiocholan-3a-ol-17-one have de- creased to 0-6-0-8 and 0-9 respectively. The decreases in 11 -oxygenated steroids were less severe. Androstane-3a,ll/3- diol-17-one excretion was 0-30-0-37 mg., androstan-3a-ol- 11,17-dione excretion was 0-1 mg., 8etiocholane-3a,ll/3-diol- 17-one w^as 0-2 mg., and aetiocholan-3a-ol-ll,17-dione was 0-21-0-23 mg. per day. Only three men and three women in an intermediate age group (fifty to fifty-seven) have been studied. Though it would appear that the decrease in 17-ketosteroid excretion in the men appears only with more advanced age, the impression is erroneous. The high standard error for the male group indicates the wide variation, and, in fact, one of the male subjects had a 17-ketosteroid excretion which was double that of the mean for the younger men. The other two men showed the same degree of decline which may be seen for the women in the middle-aged group. Discussion Since men and women of the same age group differ only slightly from each other in the excretion of any of the urinary a-17-ketosteroid components studied, it appears that the contribution of testosterone to the urinary 17-ketosteroids is small in quantity, and therefore its production is relatively small quantitatively. Even this small quantity, however, is sufficiently greater in androgenicity than the androgenic steroids of adrenal origin to determine the structural and functional diff'erences between male and female. As shown in Fig. 1, it is likely that the principal pathway of adrenocortical steroid biosynthesis is from the 3^-OH-J^ structure to the J*-3-ketones, either C^g or Cgi, and that 11, 134 Betty L. Rubin, R. I. Dorfman and G. Pincus 17, and 21 -oxygenation take plaee after the conversion to the Zl*-3-ketone. Analyses of urinary 17-ketosteroid excretion in subjects with adrenal hyperactivity have led us to the conclusion that those syndromes are due in part to imbalances in the concentrations of the various biosynthetic enzymes (Rubin et al., 1954; Dorfman, 1954). It is also possible to discuss the variations from the young normal pattern which are seen in old age in terms of the same enzyme systems. In old age, androsterone and 8etiocholan-3a-ol-17-one excretion are decreased to about 20-30 per cent of the young adult level, while ll-oxygenated-5a compounds are being excreted at about 50 per cent of the younger value, and 11- oxygenated-5j8 steroids at about 60 per cent. Therefore, we must assume that the enzyme systems converting cholesterol, acetate, or other precursors to dehydroepiandrosterone are the enzyme systems functioning at the lowest level in com- parison with those of young adults. This would account for the low excretion of 11-desoxy C^g steroids. The conversion from A^-S^-OH to J*-3-ketone and the 11 -oxygenating function are less impaired and such dehydroepiandrosterone as is formed is converted readily to androst-4-en-ll^-ol-3,17- dione. Cgi steroids are also being formed and oxygenated at C-11 relatively efficiently, as shown by excretion of two-thirds of the normal young adult amount of ll-oxygenated-5^- steroids. The connection between the indicated changes in production of individual adrenal hormones and the signs and symptoms of old age has not yet been established. To this end, a study has been started at the Worcester Foundation, in collabora- tion with Dr. Eric Bloch and with Dr. Harry Freeman and the research staff of the Worcester State Hospital, in which twelve elderly men are to be given, for many months, steroids calculated to restore the urinary a-17-ketosteroid pattern to that seen in young men. Concurrently, ])eriodic tests are made of recent memory, as an index of mental function, and of weight-lifting ability, as an index of muscle jicrformancc. The daily sublingual dose of steroids consists of 30 mg. of 17-Ketosteroid Excretion in Ageing Subjects 185 testosterone, 5 nig. of androst-4-ene-3,ll,17-trione (adren- osterone), 7 mg. of Cortisol acetate, and 2 mg. of pregn-4- ene-ll^,21-diol-3,20-dione (corticosterone). On this regime, the mean total 17-ketosteroid excretion increased from 3-5^ 0 • 4 mg. per twenty-four hours to 15 • 8 ±1.2 mg. The urinary excretion pattern has so far been determined for only one subject. It is shown in Table II. Total 17-ketosteroid excre- tion was 21 -8 mg., of which 9 mg. were androsterone, 6-6 mg Table II Urinary a-17-KETOSTEROiD Excretion of a Seventy-Seven-Year-Old Man Treated with Steroids.* mg./24 hrs. Pre-treatment total 17-KS .... 4-2 Treatment total 17-KS 21-8 Androsterone ....... 90 ^tiocholan-3a-oi-17-one ..... 2-6 Androstane-3a, ll/3-diol-17-one .... 0-4 Androstan-3a-ol-ll, 17-dione .... 0-22 ^tiocholane-3a, ll/3-diol-17-one .... 014 ^tiocholan-3a-ol-ll, 17-dione 0.58 ♦Daily Sublingual Dose =30 nig. Testosterone. = 5 mg. Adrenosterone. = 7 nig. Cortisol. = 2 mg. Corticosterone. were 3etiocholan-3a-ol-17-one, 0-4 mg. was androstane-3a,ll/3- diol-17-one, 0-22 mg. was androstan-3a-ol-ll, 17-dione, 0-14 mg. was 8etiocholane-3a,ll^-diol-17-one, and 0-58 mg. was 3etiocholan-3a-ol-ll, 17-dione. The total 17-ketosteroids and the 11-desoxy C^g steroids are higher than the mean for young men, but within the range for young men. The daily dose of testosterone is apparently too high. The 11-oxygen- ated steroids approximate young normal values, although the 11 -oxygenated androstane derivatives are rather low. 136 Betty L. Rubin, R. I. Dorfman and G. Pincus Apparently the daily dose of adrenosterone should be twice as great, or the steroid administered should be androst-4-en- llj8-ol-3,17-dione, which was demonstrated by Romanoff, Hudson and Pincus (1953), to be the steroid produced by the adrenal gland. This fact was demonstrated after the estab- lishment of the experimental regime for this study. It is as yet too early in the study to make a statement as to its probable outcome. Summary Quantitative studies have been made of six a-17-ketosteroids in the urines of normal young men and women (aged nineteen to forty) and of normal old men and women (aged seventy to eighty-eight). There is no significant difference in the excretion of any of these compounds between men and women of the same age group. When the excretion of the steroids by old men and women is compared with that by young men and women, it is found that androsterone and 8etiocholan- 3a-ol-17-one excretion are decreased to 20-30 per cent of the values for the younger group, while excretion of 5a-ll- oxygenated steroids is 50 per cent of that of the younger group, and excretion of 5jS- 11 -oxygenated steroids is 60 per cent of that of the younger group. If this difference be dis- cussed in terms of those adrenal enzyme systems involved in converting cholesterol, acetate, or other precursors, to the steroid hormones, it would appear that the enzyme systems functioning in old age at the lowest level in comparison with those of young adults are the ones which convert the pre- cursors to dehydroepiandrosterone. The conversion from Zl^-3^-hydroxy to Zl*-3-ketone and the 11 -oxygenating func- tion are less impaired. Cgi steroids are also being formed and oxygenated at C-11 relatively efficiently. This is in agreement with those studies which have demonstrated a greater decrease with age in total 17-ketosteroid excretion than in corticosteroid excretion. The significance of this finding with respect to the signs and symptoms of old age is a subject only for speculation 17-Ketosteroid Excretion in Ageing Subjects 137 at the present time, but a study is in progress in whicli recent memory and muscle strength tests are being taken at intervals by aged men on a regime of steroid administration designed to restore the urinary pattern to that of normal young men. REFERENCES Butt, W. R., Henly, A. A., and Morris, C. J. O. R. (1948). Biochem. J. 42, 447. DoBRiNER, K. (1953). J. clin. Invest., 32, 940. DoRFMAN, R. I. (1955). Ciba Foundation CoUoquia on Endocrinology, 8, 112, London: J. & A. Churchill, Ltd. GiRARD, A., and Sandulesco, G. (1936). Helv. chim. acta, 19, 1095. PiNCUS, G. (1945). J. clin. Endocrin., 5, 291. PiNCUS, G., Romanoff, L. P., and Carlo, J. (1954). J. Geront., 9, 11 *. Romanoff, E, B., Hudson, P., and Pincus, G. (1953). J. clin. Endocrin. 13, 1546. Rubin, B. L., Dorfman, R. I., and Pincus, G. (1954). Rec. Progr. Hormone Res., 9, 213. DISCUSSION Tunbridge: Dr. Freeman, would you like to elaborate Dr. Rubin's last remark? Freeman: We gave 15 subjects a placebo for one month, and then the potent medication for a period of five months. The strength test consisted of lifting 5 lb. dumb-bells, one on each arm, from the side to the level of the shoulder, 20 times a minute until the individual was tired out. This is a crude test; it is somewhat subjective, somewhat objective. On the other hand you couldn't afford to exhaust these people completely and have them pass out. Then there was a steadiness test, in which they held a rod in a hole, and if there was a tremor an electrical contact would be made. There were a series of holes, and at the smallest hole in which they could get a certain percentage of not touching the edges we ran off a series on that particular hole. There were psychological tests — arithmetic, memory, mazes, and we also had brain waves but these haven't been reported on yet. The psychological tests showed absolutely no change from beginning to end; if anything there was a slight decrease in memory quotient. This may be due to the natural ageing of the individual, and we will have to check this by running a placebo group for this whole time, preferably two groups, one receiving the placebo and the other the potent medication, with the investigator not knowing which was which. However, there is the possibility that in a larger number, if this memory defect persists, it may be an alarming phenomenon. Steadiness was not affected at all. Apparently this is a factor dependent on integration of the central nervous system rather than muscular activity. The general muscular strength however was definitely improved, starting off at a control level of sixty seconds, and going up in one month to eighty-five seconds, and then to about ninety 138 General Discussion in two to three months, ninety-five in four months, and staying at that level for the fifth month — an increase of about 04 per cent. It may be that this group was a select group of very well-preserved men, and with a less well-preserved group you might get more on the psychological side. I don't know. Tunbridge: There is some work, I believe, to suggest that when old people are kept immobile there is a difference in excretory values from when they are mobile and active. You've not done any studies on people in bed, have you? Rubin: Most of our subjects do live in an old people's home, but they are not bedridden. Tunbridge: There is evidence from fracture cases, for what it's worth, that the level when they are outside living in the country, for example farmers, is not so low. When we bring them in after a fracture the level drops lower than in your cases. Rubin: But with fracture cases you are running into the stress factor. Olbrich: How do you exclude the increase in muscle strength by daily exercise? Freeman: The strength test is only run once a month, so there is no practice effect at all. There was one session at the beginning of the placebo period, and one at the end, and the mean of each of the two tests was within a second of the other. Olbrich: And what was the interval for the memory test? Freeman: That was done in a control period and three and five months after the initial level. Comfort: Did they have subjective effects, either during the placebo period or during the medication? Freeman: I believe the placebo period was too short. During the medication period some said they felt better and there were varying reports of increased or decreased urination. The general report by the majority was of not feeling so tired. I wouldn't trust this unless I ran a group on placebos where the investigator didn't know which was which. There have been an increasing number of reports on the effect of placebos on pain and headache — ^I believe that in arthritis it has been shown that 25 per cent of cases show some improvement on placebos. Miescher: We spoke yesterday about the vitality factor. Do you think that by giving testosterone there is improvement of the physical vitality, but not of the psychological one — ^that the two don't run parallel? Freeman: Apparently not for this short period of time. I don't know whether a longer period might show anything, but certainly over a five-month period in which the hormone output was increased four to five times on an average, nothing happened psychologically. It was just an effect on their physical well-being, but that's all. Lewis: I think it is rash to assume that this is a test of muscular strength. It could ecjually be regarded as a test of endurance. Gordon, O'Connor and Tizard found, for example, in defectives, that if you give a test of this tyi)e, people may change in their endurance, and conse- ((uently in their achievement. They carry out the performance more and more effectively as their standard rises, or as they learn what they General Discussion 139 achieved the last tunc and try to improve on it. It may just as likely be a psychological test as a physical one. Freeman: That is true. As a matter of fact these people didn't know what their particular endurance was. Parkes: Dr. Rubin, if you had a sample of urine given you, could you tell with certainty wiiether it came from an old or a young man, or an old or young woman? Rubin: Not with certainty, because as I pointed out there was this one man of fifty-two who had a level which w as twice that of the mean of the young men. But I was very much surprised at his result. The young ones range from a total of 5-6 mg. to 18 or 20, whereas the older people will range from 1 • 9 to 5, if that. There is occasionally an overlap. But the values in our charts show standard error, not standard deviation, and there is a significant difference between old and young people on that basis. Parkes: But not between male and female? Rubin: No. Krohn: Miss Rubin, about the precursor of all these hormones: you say it can be either cholesterol or acetate. These two substances seem to be so different chemically that I do not see how you imagine any particular enzyme system coping with both of them. Rubin: I was very careful to hedge about that. Dr. Hechter and his group are at present engaged on perfusing adrenals with various precursor substances, but the whole thing is very much in a state of flux. I can only say that the major block seems to be in the production of dehydroepiandrosterone. Krohn: Have you any information about the response of these old people to ACTH? Are they perhaps leading a less exhausting and stressful life, and would they respond to a standard dose of ACTH in exactly the same way as the younger people? Rubin: We did some work several years ago when we gave 25 mg. ACTH a day to various age groups, and we couldn't show any difference either in total excretion or in individual patterns. We couldn't demon- strate any great response to the dose of ACTH at any age. It may be that it should have been given by continuous intravenous infusion. Shock: We carried out studies on this subject in our laboratory some years ago, both in terms of response to a single dose of ACTH* and also the metabolic responses to continued administration over a period of six weeks t. We observed our subjects on a metabolic balance regime throughout the entire period. Balances for nitrogen, Ca, P, Na, K etc., were all followed. Although we found individual differences in responsive- ness to the continued administration of ACTH, they were not related to the age of the subject. We w'cre very disappointed. Medawar: Why do you call that a disappointing result? It seems to me an intensely interesting one, of perhaps fundamental character. Shock: Well, it was disappointing because at that time, which was * Solomon, D. H. and Shock, N. W. (1950) ./. Geront., 5, 203. t Duncan, L. E., Solomon, D. H., Rosenberg, E. F., Nichols, M. P. and Shock, N. W. (1952) J. Germit., 7, 351. 140 General Discussion when ACTH lirst reared its ugly head, there was considerable clinical opinion that older individuals responded to stresses of various kinds less effectively than the young, and presuming that ACTH represented a physiological stimulus to the adrenal cortex, it was thought that one ought to see differences in response to a physiological stimulus. Rubin: I'd like to point out once more that in these old people the 11-oxygenated 17-ketosteroids which come from the C21 steroids were not nearly so markedly decreased as those reflecting C19 steroid pro- duction: they were still excreting 50-60 per cent of the young adult level of 17-ketosteroids coming from C21 steroids, though the total 17-ketosteroid excretion was down to 30 per cent, reflecting the greater decrease of the non- 11 -oxygenated steroids. I think the physiological tests of metabolic response of which Dr. Shock speaks would be due to the 11-oxygenated C21 steroids, so one would not expect a marked difference in young and old subjects. TISSUE TRANSPLANTATION TECHNIQUES APPLIED TO THE PROBLEM OF THE AGEING OF THE ORGANS OF REPRODUCTION P. L. KrOHN, B.A., B.M., B.Ch. Department of Anatomy, University of Birmingham. The interpretation that is attached to the word "ageing" depends very much on the particular tissue whose cells are ageing. An individual neurone in the central nervous system, for example, may divide by mitosis until the end of the first year of life; it may increase in size later in life, and it is also capable of regenerating a new piece of fibre if its axon is cut. An unknown proportion of such cells dies as time goes on, but seemingly none of them are ever replaced. The compon- ents of other tissues, such as skin or the epithelium of the gut, are continually being replaced and the cells that one sees in an old animal are, therefore, not old in the way that neurones are old, for they may just have been formed by the most recent of a series of cell divisions which stretch back into childhood and earlier. 'Both tissues have aged but the individual cells that make up the tissues have not both aged in the same way. Other types of specialised tissue, e.g. the mammary gland or the uterus, undergo cycles of activity during the life of an individual. The cells grow, mature, function, become atrophic or senile and then after an interval for recuperation may repeat the series of changes. The functional capacity of the tissue alters as its experience of stimulation by hormones increases with each cycle of activity. For the mammary gland this is demonstrated by the increasing amount of milk which is produced in successive lactations. The previous experience of the uterus to hormone also conditions its^ subsequent sensitivity to further stimuli. 141 142 P. L. Kroiin The ovary is a composite organ, some of whose cells undergo cyclic changes similar to and in fact responsible for the changes in the mammary glands and uterus. The other main group of cells, the oocytes, may, as will be seen later, behave rather like some neurones. Their lifespan may indeed rival that of neurones, although the chance that any particular oocyte will survive as long is much less. It is about the age changes in the ovary and the ways in which the methods of tissue trans- plantation can be used to study the changes that I wish to talk today. The Life of the Ovary Temporal changes in the sensitivity of the ovary to its hormonal environment (a) During intra-uterine life and before puberty A discussion of the process of ageing in the ovary cannot be restricted to the adult or old organ but must also consider the changes that take place during the prepubertal period. The ovary can, indeed, be permanently affected by its hor- monal environment very early in intra-uterine life. It is generally believed, for instance, that the ovary of the free- martin female cattle twin develops abnormally because it is affected by the transplacental passage of hormones from its brother. The gonads of a foetal foal may, at one stage, be twice the size of the ovaries of the parent (Amoroso and Rowlands, 1951). This enlargement, due mainly to an increase in the amount of interstitial tissue, is attributed to the effects of oestrogen rather than gonadotrophin because it occurs during the latter part of pregnancy, when the serum gonad- otrophin levels characteristic of the earlier stages have fallen, and because serum gonadotrophin does not cross the placental membrane. Bacsich (1949, 1950) has demonstrated regressive changes in the human ovary just after birth which he relates to tlie withdrawal, at the time of delivery, of a hormonal stimuhis. Like the adrenal cortex, therefore, to which it is so Age Changes tn the Ovary 14.'3 closely related embryologically, the ovary may undergo considerable hyperplasia in utero, which is followed later by degeneration and regression. ^lore numerous papers emphasize the converse finding that the young postnatal ovary of various experimental animals is insensitive to hormones. Absolute insensitivity of the ovaries of four-week-old rabbits has been reported by Hertz and Hisaw (1934), and Parkes (1943) has shown that female rabbits do not respond to a gonadotrophin preparation from horse pituitary glands until they weigh more than 1 • 3 kg. Adams (1953) reports that ten-week-old rabbits are insensitive to a similar type of pituitary extract whatever their weight. Their capacity to respond increases rapidly during the next two weeks and super-ovulation can be achieved when a rabbit is three months old. The ova which are released can be fertilised but fail to develop in the imma- ture animals, not, however, because the ova themselves are inadequate but because the necessary uterine conditions have not yet been achieved. The ova will develop fully if they are transferred to the uterus of a mature host. The duration of this early phase of absolute insensitivity of the ovaries varies considerably from one species to another. It appears to be very short in the rat (Price and Ortiz, 1944) and the young heifer. Harden (1951) has reported that a three- week-old calf will respond to injections of gonadotro- phin at least as w^ell as an eight to eleven-month-old calf. On the other hand, young female monkeys are insensitive for relatively much longer. In general, the interstitial tissue of the ovary responds at an earlier age than does the follicular apparatus, which seems to grow slowly but autonomously until the stage of antrum formation is reached. {b) Changes at puberty In some species the pituitary gland contains gonadotrophin even before the animal is born. The hormone gradually accumulates in the gland as the age of puberty is approached (Lauson, (iolden and Severinghaus, 1939) but none is released. 144 P. L. Krohn At puberty the material which has been locked up in the pituitary is set free to act on ovaries which have also become progressively more responsive. The process which triggers the initial release is, however, altogether unknown. If Kallas (1929) is correct, the immature gonads already hold the pituitary in check before puberty. He found that a 15-20 g. rat in parabiosis with a castrated immature rat of the same weight rapidly came into oestrus. Kallas believed that the absence of the gonads in one parabiont allowed its pituitary to release gonadotrophin which stimulated growth in the gonads of the other. We now know, however, that the non-specific stress of surgical operations may be sufficient to induce precocious puberty (Mandl and Zuckerman, 1951). (c) During maturity The reproductive processes continue to mature after the first pregnancy and maximum efficiency, as determined by the number, size and development of the individual foetuses, is not reached until several pregnancies have occurred. This is true both of laboratory and agricultural animals. Not only are the number of ovulations increased (either because more gonadotrophin is being secreted or because the ovaries are more sensitive) but the intra-uterine conditions evidently become more satisfactory since the individual foetuses are larger as well as more numerous. {d) The end of reproductive life The maximum reproductive capacity is maintained for several pregnancies before it begins to decline. The most dramatic and rapid change is seen in women when the onset of the menopause, with which is associated a decreased production of sex hormones and the disappearance of oocytes from the ovaries, indicates the end of the child-bearing epoch. Whether the increased amounts of gonadotrophin circulating in the blood and urine at this time represent extra production by the pituitary or decreased utilisation by the gonads is not Age Changes in the Ovary 145 clear, nor is it certain for how long into old age the increased titres of gonadotrophin continue. It is certainly reasonable, at first sight, to believe that the menopause develops in women because the ovary is no longer capable of responding to gonadotrophin, but the evidence on this point is not very strong and even contradictory. Kurzrok and Smith (1938) found that the response to treat- ment with a pregnant mare serum extract gradually decreased as the age of the patient increased. The ovaries of two women aged sixty-three and sixty-six did not respond at all. Westman (1934), on the other hand, believed that he restored ovarian function in two women aged fifty-one and forty-eight, who had stopped menstruating, by giving them transfusions of blood from pregnant women which contained an anterior pituitary-like gonadotrophin. Waldeyer (1934), in a criticism of this w^ork, described seven menopausal women in whom ovarian function apparently restarted after an interval of four or five years without any treatment and referred to other reports that regular menstrual periods may recur seven or even nine years after the menopause. Waldeyer also quoted the work of Tschertok and Penkow who treated seven meno- pausal women with extracts of pregnancy urine unsuccessfully. The disappearance of ova from the ovary does not necess- arily mean that the post-menopausal ovary is an entirely inert organ (see Wallart, 1942). Woll, Hertig, Smith and Johnson (1948) also believe that the stromal elements may become hyperplastic as a result of stimulation by pituitary gonadotrophin and that the occurrence of carcinoma of the uterus can be correlated with such changes. As far as is known no such abrupt menopause occurs in other species, but the information available is extremely scanty, either because the animals die too soon in their natural habitat, or are killed too soon for economic reasons. Information about the reproductive life in apes and monkeys is especially lacking. Zuckerman (1947) describes a baboon which was fully mature and menstruating in 1929 and which was still menstruating in 1947. (The animal is still alive and 146 P. L. Krohn menstruating at rather irregular intervals in 1954. Whether it is ovulating or not is unknown.) Another baboon also menstruated over a period of at least fourteen years. A mon- key in the Birmingham colony accidentally died during preg- nancy at an age of fourteen to fifteen, and other spayed animals which are being used to compare the sensitivity to hormones of the uterus in youth and old age are now approxi- mately twenty years old. The only other information is provided by Hartman (1938) who mentions that his colony of rhesus monkeys contained two animals which reached the ages of eighteen and seventeen before they died. Both were still menstruating but they had not ovulated for the preceding three and one-and-a-half years respectively. Since the menarche occurs at about the age of four, these fragmentary observations suggest that reproductive life in monkeys may last for as long as twenty years, a period which probably represents a greater proportion of the total lifespan than does the phase of reproductive life in women. Strain differences in the rate of changes w^ith age The rate at which reproductive function comes to an end seems to vary considerably even within a single species. Fekete (1953) has described the histological appearance of the ovaries of different aged mice from eight inbred strains. Her observations are only qualitative but they show, for instance, that the number of follicles is noticeably diminishing in mice of the RIII and C57 strains when they have reached the age of eight months, while similar changes have not taken place in the C3H strain at thirteen months of age. Loeb (1948) also provides qualitative information about inter-strain differences in the number of follicles and in the histological appearance of other components of the ovaries of mice. Some similar observations have been reported by Wolfe and Wright (1943) for two strains of rats, which, at equal ages, differed not only in the relative proportions of the cell types in the pituitary but also in the numbers of normal and atretic follicles and of corpora lutea. However, Wolfe and Age Changes in the Ovary 147 Writrht only counted follicles over 0-3 mm. in diameter and therefore omitted all the smaller follicles which make up 90-95 per cent of the total population. The Age of Ova at the Time of Ovulation and the Problem of Oogenesis It is becoming increasingly clear that the age of the mother at the time of ovulation may affect the future characteristics of the offspring in all sorts of ways. Since most of the evidence bearing on this point has been brought together in a sj^mpos- ium organized by the New York Academy of Sciences (1954) there is no need to do more than mention a few examples. The increased risk of giving birth to a Mongol child when the mother is more than forty years old has long been recognised. The defect, whatever its cause, seems to be related to the age of the mother only and not to the number of pregnancies that she has had. The chances of finding a wide range of embryological defects and the susceptibility of animals to neoplastic disease are also related to the age of the mothers (Strong, 1954; Law, 1954). The relative significance of, on the one hand, the ageing ovum in the ovary, and the ageing uterine environment on the other, cannot yet be decided. It remains important, there- fore, to determine whether the ovum that is fertilised should be regarded as a young or an old cell. We are thus led immedi- ately to consider the problem of oogenesis, which has formed one of the major points of controversy among reproductive physiologists for a very long time. Two entirely contrary views have taken it in turn to become the accepted dogma. The older view is that the ovary is endowed during embryonic development with a fixed number of primordial germ cells from which the oocytes subsequently develop. These cells are not derived from the germinal epithelium which covers the ovary, but migrate to the ovary before birth. Multiplication by ordinary mitosis may occur during intra-uterine life but has come to an end soon after birth; from then on the ovary merely uses up the material 148 P. L. Kroiin that it already contains. If this view is correct then the hfe- span of an ovum may a})proach that of a neurone and one might expect that an ovum which was fertihsed when a woman was eighteen years old would behave differently from one that was fertilised when she was forty years old. Lansing's (1954) demonstration in rotifers that the lifespan and repro- ductive capacity of the offspring are greatly influenced by the age of the mother may well turn out to be relevant in this connection. It is believed that a fixed number of future oocytes are incorporated into the ovaries of these animals during embryological development and that no further formation of oocytes takes place. Hence, the lifespan of the offspring depends directly on the length of life of the individual ovum before fertilization. The alternative view% which has become more popular recently, is that there is a continuous process of destruction and neoformation going on in the ovary throughout repro- ductive life. It derives to some extent from the feeling that the signs of active atresia are so evident in histological sections of the ovary that neoformation must also occur to make up, even partially, for the great loss of cells. Though a possible contribution of other elements has never been ruled out, the cells in the germinal epithelium covering the ovary are usually thought to be the equivalent of the layer of spermatogonia in the testis and to })e responsible for the replacement of the oocytes that are destroyed by atresia. The rates of the two processes of oogenesis and spermato- genesis thus become much more similar. It has been suggested, indeed, that an ovum lives no longer than a red blood cell and that its lifespan is to be measured in days rather than years. If this second view is correct, an ovum that was released when a woman was forty would possibly have suffered no greater age changes than one released when she was eighteen, and the effects of maternal age on the foetus might depend more on changes in tlie uterine evironment than in tlie ovary. It would also mean that attempts to prolong reproductive life in women could be based not only on Age Changes in the Ovary 119 controlling the destructive processes of atresia, but also on stimulating continued division in the cells of the germinal epithelium after they have ordinarily become inactive. There is no need to discuss fully the evidence for and against these two views since it has been summarised and criticised in detail by Zuckerman (1951) who supports the older view in his summing up. Two things stand out. The first is the extraordinary paucity of reliable counts of the total number of follicles in the ovary of any species at different ages, and the difficulty of forming a satisfactory picture of oogenesis by piecing together a series of discontinuous histological observations. Second, neoformation, even if it occurs at all, is totally insufficient in women and partly insufficient in other animals to compensate for the atresia that takes place. Age changes in the ovum However long or short the life of an oocyte before its final maturation may ultimately turn out to be, the ovum certainly does not go on living for very long after ovulation and the present trend of opinion is to reduce even more the interval during which the ovum is thought capable of being fertilised. Probably its entire lifespan is less than twenty-four hours and even within this interval the ovum's capacity to develop into a normal embryo falls off progressively. The work of B landau and of Young (1953) has shown that, in both guinea pigs and rats, the proportion of embryos, with defects of development increases as the time interval between ovulation and fertili- sation is increased. The ability of the ovum to become fertilised is not, however, altered in the same way. The proportion of fertilised ova in the tubes is not changed and it is only in the later stages of implantation and development that the influence of the ovum's age becomes apparent. This finding applies to cattle as well, according to Barrett (1948), who has studied the conception rates in dairy cows artificially inseminated at varying intervals after ovulation. These observations may be of importance in any consideration of the problem of abortion in women since the natural 150 P. L. Krohn mechanisms of ccstrous behaviour which ensure mating only near the optimum time in animals do not operate in human beings. Transplantation Techniques Applied to the Study of the Ageing of the Ovum and Ovary The method of tissue transplantation has been used for the study of a wide range of biological problems. The next part of this paper will discuss how the technique has been applied to the particular problems presented by the ageing ovum and ovary. Finally, some of the general hazards that beset the tissue transplanter, the ways of overcoming them, and some experimental results will be described. The transfer of ova The technique for the collection of fertilized ova from the Fallopian tubes and their transfer to either the tubes or the uterus of another animal is now well established and has been used especially in the study of reproductive physiology in rabbits and cows; such experiments have not usually, however, been directed towards the study of age changes in the repro- ductive tract itself. The method has also been used by Fekete (1947) to investigate the effect of transferring ova from one strain of mice into an environment provided by another strain of mice. Her work shows, for example, that the eggs of the two strains, dba and C57, have about equal powers of survival but that the environment provided by one strain is more conducive to survival than that provided by the other. Once again, however, the work was not directed towards the study of age change but it indicates the potentialities of the method. The grafting of whole ovaries The onset of puberty Some of the earliest work using ovarian transplantation was concerned with the factors which influence the onset of puberty. It was, indeed, carried out long before an interpre- tation of the results in terms of pituitary gonadotrophin Age Changes in the Ovary 151 could be attempted. Foa (1900, 1901) transplanted newborn rabbit ovaries into rabbits of different ages. One of his experimental groups, in which twelve to eighteen-month-old rabbits were used as hosts, demonstrated for the first time that transplantation into a mature environment accelerates the maturation of a young ovary. This group, however, only contained three experiments of which two gave positive results. In another group the hosts were five to six years old and the newborn ovary never survived the transplantation. Conversely, Lipschutz (1925) showed that the transplantation of mature ovaries failed to induce rapid feminization if the host guineapigs were prepubertal. Other experiments, which indicate that ovaries of newborn animals develop precociously when grafted into older animals, have also been described by Lipschutz (1925), Del Castillo (1929), Engle (1929) and Pfeififer (1934). Goodman (1934), May (1940) and Dunham, Watts and Adair (1941) showed that immature ovaries which are implanted into the anterior chamber of the eye of adult male and female rats of the same inbred strain also develop more rapidly than normal. CEstrous cycles are restored after only a very slightly longer delay (thirteen to seventeen days) than follows the grafting of adult ovaries (seven to fourteen days). In many of these experiments there is some doubt about the final result and Del Castillo, for instance, believes that the oestrous cycles which appear in adult spayed animals grafted with young ovaries, soon disappear again. He could find no trace of the grafts at autopsy. Hertz and Hisaw (1934), whose work on the sensitivity of young ovaries to gonadotrophin has already been mentioned, grafted four-week-old ovaries under the renal capsule of twelve to fourteen-week-old rabbits. Ten days later they began treatment with gonadotrophin. In four experiments in which the grafts took successfully the hosts' ovaries responded normally, but the grafts were not affected. They interpreted this observation to indicate that the failure of the very young gonad to respond resided specifically in the ovary and was not influenced by conditions in the rest of the intact animal. 152 P. L. Krohn It is now known, however, tliat grafts of any age often fail to respond to a dose of gonadotrophin when the host's ovaries remain in situ, a finding which is attributed by Lane and Markee (1941) to differential take-up of the hormone by the host glands which retain an intact normal blood supply. The anterior pituitary gland of a young animal also develops more rapidly after it has been transferred into an adult host. Harris and Jacobsohn (1951) found that pituitary tissue which was transplanted from two-day-old female rats to the sella turcica of their hypophysectomized mother, very soon functioned like a normal mature gland and restored normal cestrous cycles. Harris and Jacobsohn also found that the pituitaries of newborn male rats were just as effective in producing cestrous cycles in the female host. When ovaries are transplanted into adult males, however, only follicular development takes place and no corpora lutea are formed. The mature male pituitary apparently fails to secrete the luteinizing hormone which is required for ovulation and the formation of corpora lutea. There is, therefore, a functional difference between mature male and female pituitaries. By gonad transplantation studies Pfeiffer (1936) was able to show that, at birth, the character of the hormone production of the pituitary is still undetermined and can be directed in one direction or the other by bringing the gland under the influence of male or female gonads. Final determi- nation has normally occurred before the age of puberty, after which attempts to change the function of the gland are unsuccessful. Pituitary transplants and the end of activity of the ageing ovary Wiesner's preliminary observations (1932) indicate that the ovaries of old rats, unlike those of women, are still capable of responding to gonadotrophic stimuli, which, in ordinary circumstances, the pituitary no longer provides. Compensa- tory hypertrophy after unilateral gonadectomy did not occur in either male or female rats which were over twenty-four Age Changes in the Ovary 153 months old and which showed evidence of gonadal insuffici- ency. Ovarian function could, however, be restored by treatment either with extracts of pregnancy urine or with grafts of anterior pituitary tissue. Hoffman (1931) also restored oestrous cycles in old anoestrous mice by transplants of calf pituitary tissue. It is evident from the illustrations provided by Hoffman that the ovaries of these mice, al- though atrophic, still contained many primordial follicles which could respond to gonadotrophin. If these experiments are valid, it would seem that, while the function of the ovary fails before that of the pituitary in women, such is not the case in rodents. This inter-species difference is further shown by work on the development of ovarian tumours in the ageing ovaries of rodents transplanted into the spleen. Ageing of the ovary and the development of ovarian tumours It is known that tumours develop in ovaries that are autotransplanted to the spleen of spayed rats or mice (Bis- kind and Biskind, 1944). The usual explanation for the development of these tumours is that oestrogen from the grafted ovary passes directly into the portal circulation and is destroyed by the liver before the hormone can inhibit the release of gonadotrophin from the pituitary (Golden and Severinghaus, 1938). The imbalance of hormone production by the pituitary persists and stimulates abnormal growth of the graft. While this is true for the rat it may not be so for the rhesus monkey (Van Wagenen and Gardner, 1950) or guinea pig (Wenner and Hoffman, 1950). It has already been mentioned that the level of circulating gonadotrophin in women rises as ovarian function declines at the menopause. If there is any reactive tissue remaining in the ovary, one would expect ovarian tumours to become more common in older women. This expectation is not, however, borne out by the facts and one must therefore assume either that the level of gonadotrophin is not high 154 P. L. Krohn enough or tliat the ovaries of women have become refractory to stimulation of any sort. In tests on rodents, however, Li and Gardner (1950) found that the ovaries of 340-491-day-old mice were just as sus- ceptible to tumour formation as were the ovaries of younger animals. Klein (1952) who extended these observations to a group of old mice whose average age was seven hundred and ninety-nine days, also found that old ovaries may develop tumours, but there was some evidence that they were not as sensitive to the stimulus as are young ovaries. The positive findings in these experiments can be correlated with the continued presence of active ovarian tissue in such old ovaries and serve to distinguish once again the ageing ovary in women from the ovaries of rodents. General methodological problems of ovarian transplantation Much of the earlier work on ovarian transplantation must face the objection that the experiments were not designed in a way that would avoid the difficulties introduced by the immunity reaction which develops when any tissue is homo- grafted. One way of overcoming this difficulty is to use what may be called delayed autografting. Tissues can be removed from an animal while it is still young, preserved for any required length of time by storage at low temperatures and then regrafted into the original donor. The results of this form of experimental approach will shortly be discussed by Parkes. The method suffers from the inevitable dis- advantage that tomorrow's older tissues can never be trans- ferred to yesterday's younger animals. It is only possible to graft tissues of their more youthful days into animals that have got older. There is also the added disadvantage that the tissue may deteriorate during prolonged storage. An alternative approach depends on the fact that some strains of mice are so inbred that tissues can be transferred from one member of the strain to another without stimulating an immunity reaction. Tissues from both parent strains of Age Changes in the Ovary 155 mice are -Aso acceptable to the liybrid produced from crossing the two strains. In this way it becomes possible to study what happens not only when an old ovary is placed in a young environment and vice versa, but also when ovaries of two strams, which may difPer in the rate in which they become old (see above), are put into a common environment. (Cf. Huseby and Bittner [1950] who were interested in the influence on the incidence of mammary cancer of ovaries from strains A and Z when grafted into AxZ hybrids.) Many difPerent sites for transplants have been chosen and the functional activity of the grafts has usually been assessed by the reappearance of oestrous cycles and by histological methods. Most sites, such as the subcutaneous tissues, the spleen or the anterior chamber of the eye, necessarily preclude the most valuable and, for our purpose, the essential criterion of function. The ovary must be transplanted so that the ova can reach the fallopian tubes, become fertilized and implant in the uterus. Such orthotopic grafting of ovaries was first carried out in rabbits and guinea pigs, which do not have a complete ovarian capsule, some fifty years ago. Foa (1901) and Castle and Phillips (1913) believed that pregnancies which took place after grafting were derived from the grafted ovary and not from regenerating host ovary. Much more recently Whitney (1946) believed that he had successfully exchanged orthotopically the ovaries from a bloodhound to a foxhound, although the animals failed to become pregnant afterwards. The fact that its ovary is surrounded by a capsule makes it easier to achieve this purpose in the mouse than in the other two species that have been mentioned. It is relatively simple to open the capsule, remove the^ animal's own ovary and replace it with another. This method of intracapsular grafting of the ovary has been employed mainly by Robertson and by Russell and Hurst (1945). Robertson (1942) has been able to differentiate, for example, between the genetic and environ- mental factors which are responsible for the intra-uterine death of homozygous yellow mice. The Bar Harbor workers, 156 P. L. Krohn who have developed an ingenious marker gene technique for distinguishing between the offspring from the grafted ovary and from any ovarian tissue that regenerates, have shown that htters can be obtained even from embryonic ovaries that have been transplanted into young adult animals. Russell and Douglass (1945) have also used the method to study what happens to the ovaries when they are submitted to abnormal hormonal influences during a temporary stay in the spleen of castrated hosts before being returned to a normal endocrine environment. They find that the oocyte producing apparatus is unable to withstand such treatment, although the endocrine function is unaffected. My own experiments using transplantation methods have been directed along the following lines. Problem of technique The only serious difficulties raised by the technique of intra-capsular grafting are incomplete removal of the ovary and bleeding from the cut ovarian vessels. Even the most meticulous surgery cannot always obviate the first difficulty, but the effects can be mitigated by using the gene marker methods introduced by the Bar Harbor workers. Patience seems to be the only way of overcoming the second. The use of a diathermy-cautery was unsatisfactory because it damaged the delicate tissues of the fallopian tubes. The mere fact of transplantation is sufficient, according to Hummel, Fekete and Little (1953) to increase the proportion of ovarian tumours that develop in later life. The results of any work which uses the method may, therefore, have to be interpreted with some caution. The survival of ovarian homografts In view of the widespread differences of opinion about the chances of survival of ovarian homografts, discussed elsewhere (Krohn, 1955), I have carried out a further series of experi- ments in which ovaries from one strain of mouse have been transplanted orthotopically into the ovarian capsule of mice Age Changes in the Ovary 157 of another strain. The host animals were necessarily spayed before grafting and conditions for the snr\ i^'al of liomografts should have been optimal. In '25 experiments the liomografts that were recovered at autopsy, between nineteen and forty- six days after grafting, had all been destroyed by a homograft reaction in which round cell infiltration and replacement with fibrous tissue were prominent features. In some of these experiments oestrous cycles had continued after the operation. The explanation for this fact, however, lay not in survival of the homograft but in the presence of small masses of ovarian tissue of host origin left behind after the ovariectomy. Differentiation between homograft and autogenous ovarian tissue was quite obvious when the material was examined histologically. In further experiments the liomografts were removed either six or nine days after grafting. In all a reaction against the grafts was well under way; in most of them the tissue organisation of the grafts had already been destroyed, but a few small follicles and some luteal tissue still looked reasonably healthy. The follicular apparatus does not withstand the rigours of homotransplantation as well as the luteal cells, scattered clumps of which may still be seen some time after the rest of the ovary has been destroyed. The "Potential Immortality'* of the ovary The fact that tissue cultures of fibrol>lasts can be main- tained for very long periods of time is well know^n. The same sort of cellular immortality can be demonstrated for tumour cells by repeated passages of the malignant tissue from one host to another in such a way that the descendant cells of the timiour have far outlived the animal from which the tumour was originally derived. Very few attempts have been made to carry out similar experiments with normal organs. Loeb (1945) reports the serial transplantation of thyroid glands in Strong A mice. The number of serial transplantations ranged from three to seven and the transplants remained in each host for three-and-a-half to six months. In some of these experiments the thyroid tissue survived the process of serial 158 P. L. Krohn trans])lantation and the oldest tissue recovered was forty-one months old, Avhich is ])robably outside the ordinary lifespan of the strain of mice used. Loeb (1945) mentions that similar experiments with grafts of ovary failed entirely. Some comments by Dunham, Watts and Adair (1941) that are incidental to their main work, also suggest that the ovary does not tolerate being transplanted more than twice. In our own experiments ovaries have been autotrans- planted from one side of the body to the other at intervals of two or six weeks. Control autografts readily survived a period of up to twelve weeks and gave rise to normal vaginal oestrous cycles. In the first group of animals the period between successive transplantations was two weeks. After the second transplantation two out of eight animals continued to show definite oestrous cycles and four of the eight showed occasional and irregular periods of oestrus. After the third transplantation four out of five animals showed no cestrous cycles at all and only one gave a doubtful positive result. The same results were obtained in another group which were continued to a fourth transplantation. In the final experiment six weeks were allowed to elapse before the second trans- plantation. (Estrous cycles did not recur during the succeed- ing six weeks in any of the four animals used. The experiments have, therefore, failed in their purpose. They are being repeated using orthotopic grafts and allowing a longer interval to elapse between successive graftings. The behaviour of old and young ovaries in environ- ments of different ages Combinations of orthotopic grafts have ' been employed, either young ovaries into old hosts, old into young, young and old into young, and young and old into old. Only Strong A strain mice have been used. It is too early to say what the effects on the numbers of oocytes and on reproduction will be, but it would appear that old ovaries grafted into young animals behave more satisfactorily, as far as the data on the oestrous cycles go, than young ovaries grafted into old animals. Age Changes in the Ovary 159 This would confirm the general impression that ovarian function in rodents is related to pituitary activity and not to changes inherent in the ovaries themselves. Using CBA and A strain mice for the work, we are also satisfied that a hybrid host can tolerate grafts of ovaries from both parent strains. Of the six animals used all have shown cestrous cycles and after mating have had vaginal plugs and become pseudo-pregnant. So far two of them have become pregnant. Before proceeding any further with the experiments it will be necessary to establish more exactly the rate at which the total number of follicles in the ovaries (both normal and after transplantation) falls off as the mice get older. Only when this information is available will one be able to say anything with certainty about the interaction of the ageing environment and the ageing ovary. Conclusion It is a far cry from the days of Voronoff and the expectation that transplants of gonadal tissue would serve to rejuvenate the senescent. It is now plain that such hopes, which rested on the possible benefits of heterotransplantation, were bound to be illusory. Today, even homotransplants of gonadal or of endocrine tissue have no practical clinical value. In the future, however, the method of transplantation may contri- bute more as a research tool to the problem raised by the ageing of tissues than it ever could ha^^e done in the past as a form of treatment. Acknowledgement I am very grateful to Miss E. Holt for her assistance in carrying out the experiments. REFERENCES Adams, C. E. (19;")8). In Ciba Foundation Symposium, Mammalian Germ Cells, p. 198. London: J. & A. Churchill, Ltd. Amoroso, E. C, and Rowlands, I. W. (1951). J. Endocrin., 7, 1. Bacsich, p. (1949). J. Endocrin., 6, i. Bacsich, p. (1950). J. Endocrin., 7, xiv. 160 References Barrett, G. R. (1948). Ph.D. Thesis, University of Wisconsin. Quoted by Young, 1953. BisKiND, M. S., and Biskind, G. S. (1944). Proc. Soc. exp. Biol. N.Y., 55, 176. Castle, W. K., and Phillips, J. C. (1913). Science, 38, 783. Del Castillo, E. B. (1929). C. R. Soc. Biol, Paris, 102, 457. Dunham, L. J., Watts, R. M,, and Adair, F. L. (1941). Arch. Path., 32, 910. Engle, E. T. (1929). Amer. J. Anat., 44, 121. Fekete, E. (1947). Anat. Rec, 98, 409. Fekete, E. (1953). Anat. Rec, 117, 93. FoA, C. (1900). Arch. ital. Biol. 34, 43. FoA, C. (1901). Arch. ital. Biol., 35, 364. Golden, J. B., and Severinghaus, E. L. (1938). Proc. Soc. exp. Biol. N.Y.,39, 361. Goodman, L. (1934). Anat. Rec, 59, 223. Harris, G. W., and Jacobsohn, D. (1951). Proc. Roy. Soc. B., 139, 263. Hartman, C. G. (1938). Les Hormones Sexuelles. p. 103. Paris: Hermann & Cie. Hertz, R., and Hisaw, F. L. (1934). Amer. J. Physiol. 108, 1. Hoffman, J. (1931). Amer. J. Obstet. Gynec, 22, 231. Hummel, K. P., Fekete, E., and Little, C. C. (1953). Anat. Rec, 117, 628. HusEBY, R. A., and Bittner, J. J. (1950). Steroid Hormones: A Sym- posium, p. 361. University of Wisconsin Press. Kallas, H. (1929). C. R. Soc Biol, Paris, 100, 979. Klein, M. (1952). J. Nat. Cancer Inst., 12, 877. Krohn, p. L. (1955). Ann. N.Y. Acad. Sci., 59, 443. KuRZROK, R., and Smith, P. E. (1938). Ass. Res. Nerv. Ment. Dis., 17, 340. Lansing, A. I. (1954). Ann. N.Y. Acad. Sci., 57, 455. Lane, C. E., and Markee, J. E. (1941). Growth, 5, 61. Lauson, H. D., Golden, J. B., and Severinghaus, E. L. (1939). Amer. J. Physiol., 125, 396. Law, L. W. (19.54). Ann. N.Y. Acad. Sci., 57, 575. Li, M. H., and Gardner, W. U. (1950). Cancer Res., 10, 162. LiPSCHUTZ, A. (1925). C. R. Soc. Biol., Paris, 93, 1066. Little, C. C, Hummel, K. P., Eddy, M., and Rupple, B. (1951). Proc. nat. Acad. Sci., Wash., 37, 666. LoEB, L. (1945). Biological Basis of IndividuaHty. p. 104. Springfield: C. C. Thomas. LoEB, L. (1948). Arch. Path., 46, 401. Mandl, a., and Zuckerman, S. (1951). J. Endocrin., 7, 335. Marden, W. G. R. (1951). J. Physiol., 115, 22.P. May, R. M. (1940). Bull. Histol. appliq. Physiol., 17, 51. Parkes, a. S. (1943). J. Endocrin., 3, 268. Pfeiffer, C. a. (1934). Proc Soc exp. Biol., N.Y., 31, 479. Pfeiffer, C. a. (1936). Amer. J. Anat. 58, 195. Price, D., and Ortiz, E. (1944). Endocrinology, 34, 215. References 161 Robertson, J. J. (1942). J. exp. ZooL, 89, 197. Russell, W. I^., and Douglass, P. M. (1945). Proc. nat. Acad. Sci.y Wash., 31, 402. Russell, W. L., and Hurst, J. G. (194.5). Proc. nat. Acad. Sci., IVash., 31, 267. Strong, L. C. (1954). Ann. N.Y. Acad. Sci., 57, 507. Van Wagenen, G., and Gardner, W. U. (1950). Endocrinology, 46, 265. Waldeyer, L. (1934). Zbl. ges. Gynak., 58, iii, 2882. Wallart, J. (1942). Bull. Iiistol. appliq. Physiol., 19, 195. Wenner, R., and Hoffman, K. (1950). Gyniecologia, 130, 382. Westman, a, (1934). Zbl. ges. Gynak., 58, ii, 1090. Whitney, L. F. (1946). Science, 103, 654. WiESNER, B. P. (1932). Brit. med. J., 2, 585. Wolfe, J. M., and Wright, A. W. (1943). Cancer Res., 3, 497. WoLL, E., Hertig, a. T., Smith, G. van S., and Johnson, L. C. (1948). Amer. J. Obstet. Gynec, 56, 617. Young, W. C. (1953). Pregnancy Wastage. Edited E. T. Engle. p. 38. Springfield: C. C. Thomas. ZucKERMAN, S. (1947). J. Eudocrin., 5, 220. ZucKERMAN, S. (1951). Rcccnt Progr, Hormone Res., 6, 63. PRESERVATION OF TISSUE IN VITRO FOR THE STUDY OF AGEING A. S. Parkes, Sc.d., f.r.s. National Institute for Medical Research, London. The general question of ageing of parts of the body in relation to ageing of the whole can be investigated in various ways. I want to deal with a specialized development of one of them. In the case of organs or tissues which can be grafted it is possible to transfer the organ or the tissue from one animal to another in such a way as to have an old tissue in a young animal or a young tissue in an old animal. Theoreti- cally, as Dr. Krohn has indicated, it is thus possible to distinguish between functional breakdown of a tissue due to internal causes and due to failure of its environment. Unfor- tunately, some of the organs most suitable for homografting and for estimation of functional activity — the endocrine organs other than the pituitary body — are of limited signi- ficance in this connection, because they are known to be dependent on environmental stimulation, and, in the case of the gametogenic function of the ovary, to be of limited potentiality. Nevertheless, instructive results could be expected, especially where successive passages of the same tissue from old animals to young animals as each host becomes senile makes it possible to secure a big difference between tissue age and body age. All such experiments, however, require that the part shall be subjected to the hazards of homografting. I want now to discuss a recently developed technique which permits of experiments where the part is physiologically younger than the whole without the hazards of homografting from a young animal to a contemporaneously older one. I refer, of course, to the technique of low temperature preservation by which tissue can be held in a state of sus- pended but potential viability, free from the complications 102 Preservation of Tissue in vitro 1G3 of ageing, while the donor animal wends its way to senility. The idea of achieving a state of suspended animation at low temperatures is, of course, very old, and depends in its modern form on the reas()nal)le assumption that all vital j)rocesscs would be suspended at temperatures such as those obtained by the use of dry ice (-79° C.) or liquid air ( — 190° C). Until recently, however, the act of freezing and thawing was fatal to almost all normal cells of vertebrates, so that little pro- gress was possible in the study of experimental biostasis. This situation was altered radically a few years ago by the observation that glycerol has remarkable properties in pro- tecting living cells against the otherwise fatal effects of freezing to very low temperatures (Polge, Smith and Parkes, 1949). This discovery has made it possible to keep a variety of living cells at —79° C. or —190° C. in an apparently stable state for periods of months or years, and the work has added greatly to our knowledge of the dynamics of the living cell and in particular of the effects of low temperatures. Resistance to Freezing and Thawing It appears that freezing damage is primarily of two kinds, one due to the hypertonicity of the residual fluid as water freezes out as ice, and the other to thermal shock arising from rapid change of temperature either above or below freezing point. The former can be avoided to a large extent by ultra-rapid freezing, the latter by slow cooling. In other words, it is almost impossible to freeze a normal cell by ordinary methods without subjecting it to one or other type of damage. Glycerol apparently exerts its effect by decreasing the hypertonicity hazard, and permitting cooling to be slow enough to avoid thermal shock (Lovelock, 1953). Stability at Low Temperatures The loss of cells on freezing and thawing without storage at the low temperature may be negligible, appreciable, sub- stantial or very large according to the conditions and the 104 A. S. Parkes type of cell. On storage at the low temperature the cell may be apparently stable, as with bull spermatozoa (Rowson and Polge, 1953) and rat ovarian tissue (Parkes and Smith, 1953), or the loss caused by the initial freezing may be increased, as with fowl spermatozoa or human red cells, according to the type of cell, the medium and the temperature. In any case, cells that survive storage have apparently not aged. Frozen bull spermatozoa, for instance, after thawing, survive at normal temperatures as well as do fresh sperm, too 0 20 40 60 80 100 120 DAYS AFTER TRANSFUSION Fig. 1 . Survival after transfusion of red cells stored at — 79° C. for six months and at +4° C. for eleven days. Survival is normal. (Mollison, Sloviter and Chaplin, Lancet, 1952, 2, 501. By permission of the Lancet.) and human red cells, after long periods of storage, have a normal expectation of life in the circulation after transfusion (Mollison, Sloviter and Chaplin, 1952). A significant deduction is to be drawn from these observa- tions. It may be assumed that biochemical changes are arrested at the very low temperatures employed. The changes taking place in cells which suffer damage on storage must therefore be physical ones occurring in the absence of the usual biochemical processes of repair. The various factors involved in the case of the red blood cell have been investi- gated in detail by Lovelock (1954), according to whom the damage caused by storage under unfavourable conditions I PUESEIIVATION OF TlSSUE iu vltt'o 1G5 arises from tlie failure of bioeliemieal processes to rei)air biophysical dissolution. With the red cell it appears that damage of sub-lethal degree occurring during storage may be repaired when the cell is returned to its normal environ- ment. It is possible that the general thesis set out by Lovelock is of more general applicability to the exhaustion and repair of cells; it might perhaps be possible to regard ageing as the result of increasing failure to repair by active biosynthesis the passive dissolution due to biophysical causes. Isolated Tissues I want to turn now to the application of this work to the study of the ageing of parts in relation to the whole. I have already indicated that ovarian tissue of rats can be preserved for long periods in liquid air at —190° C, that there is some loss of cells on freezing and thawing, but that the loss is not increased on storage and that the cells which survive have not apparently aged. Details of technique and data for tissue frozen for up to one year were given at a previous symposium (Smith and Parkes, 1954). By the time similar material had been frozen for two years a number of complica- tions had appeared. For one thing, several of the donor rats died during the second year. For another, several known failures in the maintenance of the flasks of liquid air had occurred, and tissue known to have warmed up had been discarded. As a result of these circumstances there was avail- able after approximately two years a number of old ovariecto- mized females without corresponding stored ovaries as well as stored ovaries without corresponding old animals. An experi- ment was therefore designed in which in addition to the auto- grafting of long-stored ovaries, fresh ovarian tissue from young females was homografted into old ovariectomized females and long-stored ovaries into newly ovariectomized young females (Table I). The results of the first experiment sug- gested that the tissue autografted after storage for approxi- mately two years (line one of the table) did not take well. 166 A. S. Parkes The second experiment (line two of the table) seemed con- clusive on the point. The homografting of long-stored ovaries into newly ovariectomized young females (line three of the table) and the reciprocal experiment of homografting fresh tissue from young females into the old ovariectomized Table I Grafting of Long-Frozen Ovarian Tissue into Original Donors or Young Ovariectomized Females and of Fresh Ovarian Tissue from Young Females into Long-Ovariectomized Old Females (Donors all young females approx. 100 g. B. Wt.) Xalare of orafl Time uvariaN 1 issue frozen '{(lai/s) Hecipients {all ovariectomized) Proportion of animals (ievelopinfi active iiraft Average interval between firaftinii and cornijicd smear (dans) Period of observation offer iirafting (days) Total No of cycles observed 1 Autograft 674-747 Donors, approx. two years after ovariectomy 5/9 18 50 15 2 Autograft 719—776 Donors, approx. two years after ovariectomy 1/13 12 33 2 3 4 Homograft 668-777 Young females 4/23 23 32-50 11 Homograft Fresh tissue Old females approx. two years after ovariectomy 9/9 9 52 37 5 Homograft (control) Fresh tissue Different young females 9/10 12 28 29 females (line four of the table) showed clearly that the long- stored tissue, not the old females, was at fault. It must be assumed, therefore, that there had been an undetected breakdown in maintenance of the liquid air flasks, or that the tissue had deteriorated spontaneously during its second year of storage. It has not yet been possible to distinguish between these alternatives. Isolated Organs In some ways the possibility of long-term preservation of isolated organs is of even greater interest than that of cells or tissues. An approach has been made to this problem by All Kai Onaklvx Graft; Fig. 2. Autooraft of long-stored ovaries to old ovariectomized females, showing mature follicle and normal egg. Ovaries removed from young rat of approximately 50 g. body weight, stored at —190° C. in 15 per cent glycerol- saline for 747 days and then autografted back to donor. Vaginal cornification occurred 14 days later and four cycles were observed before the graft was re- moved 50 days after transplantation. (I^FO 80.) m^' Fig. 3. Homograft of long-stored ovaries to young ovariectomized female, showing corpus luteum and follicles. Ovaries removed from yoimg rat, stored at — 190° C. in 15 per cent glycerol-saline for 720 days and then homografted to newly ovariectomized young female. Vaginal cornification reappeared 20 days later and two cycles were observed before the graft was removed 32 days after transplantation. (LFO 220.) [To face page 166 M ^ % ■■ 'x^^. '^ -zU Fig. 4. Ilomouialt of ficsli ovaries to old ovariectoniized rat sliowino- mature follicle and normal cuus. Ovaries removed from youno female and lionio- grafted to old lat ()(;4 days after ovariectomy. Va' year. Antlers ean be eollected and weighed and analysed or even studied when they are growing. It is a beautiful animal for studying the ability to regenerate. Here is a reproducible growth impetus every year, it goes down gTadually as the animal grows older. But you can change the development of the antlers, you can control it by changing the endocrine environment. Lansing: There is a very large literature on the role of nutrition in experimental animals. It doesn't need another fifteen years to be resolved, it has been resolved for more than fifteen years — some of the work was done in about 1900, on silk worm, Cladocera (Daphnia in particular), on rotifers, on Planaria and other fiat worms, even on rats and mice. Hoelzel and Carlson several years ago published extensively on this subject, and I'm surprised at the amount of general agTcement there is. The consensus is, with amazingly little difference of opinion, that restriction of the diet prior to maturation does seem to effect a prolongation of life-span — restriction of diet, or super-alimentation, after maturation, having no effect at all. Shock: But isn't the problem that we are raising here a little deeper than that? We are asking "what is the mechanism whereby this pro- cess is carried out"? It seems to me that although we do have a fairly large body of information on various animal species, we have not yet any fruitful ideas on the mechanism of the process. McCay: In the other direction (and in regard to Dr. Olbrich's question), we have left a substantial group of rats to become middle-aged and fat, and then forced some of them to become tliin by exercise or restriction of food, and that lengthens the life of the rat — not as significantly as retarded growth, but one does get a significant result. Tunbridge: You have some work on your mental patients, haven't you, relating to the calorie requirements in elderly people? Is it correct that their calorie intake is diminished? McCay: There is work on that, but I didn't do it. That gets involved with basal metabolism. It looks as if evidence is very good that the BMR drops in older people, but I've never found any sound evidence for any experimental animal that the BMR drops with age. Assuming those data are sound, man may be unique in that respect. Shock: I think Davis (Amer. J. Physiol., 1947, 119, 28) studied the basal oxygen uptake of rats, and as I recall found a reduction in the basal metabolism of the rat as well as in man. McCay: We have data on both sides — you'll find one can jump either way. Shock: We have collected evidence on changes in basal metabolism with age, and it is true that the oxygen uptake of the total animal diminishes as the individual gets older (Shock and Yiengst, 19.5.5, J. Geront., in press). But if one uses some other estimate of the amount of functional tissue present, instead of surface area which is our usual standard, there is no change in basal oxygen uptake with increasing age. We have, for example, measured the total body water content in individuals in parallel with our determinations of basal oxygen uptake — and when one computes the oxygen uptake per unit of body water, all General Discitssion 207 evidence for an age change in basal metabolism disappears*. So I am inclined to question whether there is an actual slo\\ing of the cellular metabolism as you get older. I think tlie droj) in oxygen consumption which we all see is simply a reflection of the loss of fimctioning cells and their replacement with other non-mctal)olising or at least slowly metabolising tissues. Cowdry: However, it used to be thought that there was a significant dehydration, reduction in amount of water per unit of weight in older tissues. And then Hastings and others came along and seemed to show that this was not the case. What is the situation now? Shock: 1 think there is no question but that the total water content of the animal decreases with age. Using anti-p>Tene as an index we have found an average change from 34-6 1. to 29-7 1. between the ages of forty and ninety years. Again, if you want to indulge in some fancy assumptions, you can show that this is a loss of functioning protoplasm, because the extracellular water phase does not change significantly, at least in our group of subjects. Tiiis is probably a physiological reflection of the histologically observed loss of numbers of cells with increasing age. McCance: I don't know anything about the water, but if you starve a person over eighty he derives exactly the same percentage of his basal calorie requirements from breaking down of his own tissues as a young adult does. So right on into old age that balance seems to be preserved. Coivdry: Can you say that the basal metabolic rate in females is in general somewhat lower than males, for the same age, and that that is in some way related to increased life span? McCance: Our work has all been done on old men, and I'm afraid I can't tell you anything about old women. McCay: There is no proof of any sex difference in any animal other than man. Cowdry: Tliere is a difference in man, isn't there? McCay: Apparently, yes. Shock: But the female has more fat on her from the start, so that this may again be a question of relative amounts of fat. Unfortunately we have never done any studies on women, and I'd like to know if one were to relate oxygen consumption with body water, whether there would be any sex difference left. Sinclair: I think Ancel Kej^s has concluded that tliere is no sex difference in terms of active metabolising tissue — it is that there is more fat. Cowdry: But there would be a difference in terms of total weight? Sinclair: Yes, of course. Or surface area. Briill: May I remind you that at the first International Conference on Gerontology I reported that we had started a study of the influence of nutrition and feeding on the ageing of inbred mice. We started this * Shock, N. W., Yiengst, M. J. and Watkin, D. M. (1953). J. Geront., 8, 338. See also : Shock, N. W. in Symposium on Problems of Gerontology, National Vitamin Foundation, Nutrition Symposium Serv,, No. 9, 1954. 208 General Discussion seven years ago, and it is still in progress. Of course we lived with the illusion that we can avoid disease. This illusion covers all studies in gerontology, and will remain an illusion, because we shall never be able to study pure ageing without the influence of exogenous factors. To go back to Dr. Lansing's contribution about changes in the constitution of tlie tissue according to age, we are trying to analyse the whole animal. We feed groups of animals of different ages on a standard diet, and tlien we dry-powder the whole animal. The first result of these analyses is that the main effect of ageing is the putting on of fat and lipid substances. But this is already known. One of my previous masters, Terroine in Strasbourg, demonstrated thirty years ago that when you analyse tissues of animals, if you are to have comparative figures you must starve your animals to death before you analyse. When you do that, and you take kidneys or liver or muscle of different animals of the same species, the results are comparable. So we are going to do that. But as regards comparing results or exchanging of possibilities, we could just as well exchange powdered animals. I shall do what I can to analyse the powdered animals from many points of view, I shall do gross analyses and some of the amino acids etc., but many other laboratories may do what I am unable to do. So if we are thinking of working along those lines, we might help each other. I have not pub- lished anything yet about these mice, perhaps I never shall, but I am ready to compare and to exchange powdered mice of that inbred colony with anyone who wants help in that way. PSYCHOLOGICAL ASPECTS OF AGEING Sir Frederic Bartlett, c.b.e., f.r.s. Applied Psychology Research Unit, Cambridge. About seven years ago the Nuffield Trustees, with tlie approval of the University, established a group at Cambridge for the experimental psychological study of the effects of ageing in the case of normally fit persons. It is a few of the chief results achieved by this group which I am now going to describe and consider. The principal merit for the achieve- ments is due to Mr. A. T. Welford, the Director of the Group, and to his fellow- workers. If there are mistakes of interpreta- tion they are mine. It was from the beginning agreed that we should not be much concerned with the extremely old, but rather with the critical age ranges from round about forty-five to sixty-five. It was also agreed that we should not confine our attention to relatively isolated items of behaviour, such as sensory threshold measures, reaction times, immediate or remote memory span disconnected from any other activity. We should, rather, look at these in that intimate association with continued work which they have in practically all forms of day by day actual behaviour. This meant that our main pre-occupation must be to try to find out something about the nature and conditions of skilled behaviour, whether of body or mind; and that our initial problems would concern demonstrable changes in skilled activities which normally occur with advancing age. We had, of course, some guide, both from the results of a great amount of earlier experimentation, and from commonly held opinions. Nearly everybody, for instance, agrees that most actions are slowed up with increasing age, and many experimenters had already reported this. But when does this AGEING 209 15 210 Sir Frederic Bartlett slowing-up begin? Once begun does it continue and perhaps increase without recovery? Particularly are there certain phases of complex skilled activity which, more than the others, exhibit this slowing-up? It appears that there is some slowing-up of skilled actions normally in the late twenties. This is followed by considerable or complete recovery, but it reappears m the late thirties or middle forties. There is again recovery, until the early or middle fifties, and then the slowing up seems generally to continue and perhaps to increase. But there is nothing very remarkable or dramatic about it in the case of the fit person, and if it is adequately countered probably it need have no important practical significance. I am speaking of course of skilled performance which must be done rapidly, but does not require any great expenditure of muscular effort. If the latter is required, and still more if work has to be done at relatively rapid rates, and with accompanying overall bodily mobility, the final slowing-up appears earlier, usually in the early forties, and with little or no subsequent recovery. Very much more interesting is the question of whether there are certain stages in the performance of skilled move- ments of all kinds which are especially susceptible to change. It rapidly became clear that overall time measures of per- formance may be extremely misleading (Welford, 1951). For example, in one of the early experiments a small object was set into apparently random movement over a surface marked off in small squares. When it came to rest its position had to be accurately identified by manipulation and if this was done successfully the object was set into further random movement; and so the experiment proceeded. If the overall time was recorded for a series of correct locations, it could, as indeed anyone would expect, be exactly the same with great and significant differences in the internal pattern of manipulative movement. It was from this experiment that we got the first suggestion that the most unstable element in any series of accurate adjustments to changing stimulation is not the speed of movement, and not the reaction time (if Psychological Aspects of Ageing 211 this is defined in tlie nsiial way as the interval which elapses lietween the appearance of a stimulus or signal for movement and the lieginnino- of the movement), but the resting; time, or recovery time, between an adjustment which is in one direction and an immediately following adjustment in a different direction (Leonard, 1953). This has since been confirmed in many other experiments, and, recently, in an analysis of time-motion study films. It is the stationary items in apparently continuous movement series that are most liable to change. They are the constituents that are most readily accelerated or decelerated. They tend to be drawn out under the influence of certain drugs, of fatigue, and, almost certainly, of increasing age. The interesting thing is that these stationary constituents in a flow of movements, or of ideas, occur almost always, perhaps always, where something specific about signals for action has to be perceived, such as the shape, size, colour, disposition and number of objects; or where, in what is usually called "mental" skill, there is a shift from one topic to another, or some new consideration has to be brought into a process of thinking. This is to say that the most variable elements in skill, and the ones that are most likely to be affected adversely with increasing age, unless precautions are taken, are just those that most indubitably demand central, neurological activity and control. To have demonstrated an increasing importance of halts in a flow of movements, or ideas, with increasing age is a step in advance. But by itself it is not much of a step, and when we try to go further we are at once brought up against diffi- culties of interpretation which can be resolved only by more experimentation. There is considerable evidence now that older people — I mean broadly from the middle forties upwards — tend to look both at what they are doing and at the signals for what they are to do, in most kinds of bodily skill, when accurate and rapid work is required, far more than younger people (Szafran, 1951). It may even be the case in general that increasing 212 Sir Frederic Bartlett age prefers to have cues for performance and of the course of performance from as many soin'ces as possil)le. In particular it may be that sensitivity for proprioceptive feed-back diminishes. For this however there is no direct evidence as yet, or, in fact, of significantly lowered threshold response in any direction. It is safer, though of course less satisfying, to treat the demand for a greater amount and diversity of confirmiup" evidence as one of the signs of the increased caution which is alleged to accompany advances in age. It is more intriguing to consider the possibility that the range and function of anticipation may change with age. We have decisive experimental evidence that far and away the most important character in skill performance is what has to be called its temporal structure. The way in which any item of skill is dealt with depends upon its relation to preceding and succeeding items in the sequence of which it forms a part. If a signal for action is given at an appropriate interval before the moment of action it is the resting time between move- ments in one direction and another, whether of muscle or of mind, that is significantly reduced. It is a reasonable sugges- tion that the general effect of accumulation of experience is to enhance the influence of preceding items in any skilled series and to depress that of succeeding items. There may therefore be a retraction of anticipation range with age and if this is so it would undoubtedly produce just that preference for shorter advances and more and longer halts which older performers are known to show. But only further experiments can decide. There is yet another possibility. Every human action exhibits a character which may be most forcibly described as a "point of no return". That is, it reaches a phase beyond which putting in new signals produces no effect whatsoever. Before this stage is reached there is a range within which incoming signals produce actions, but the action is mal- adjusted: a mistake is made. It has become a matter of great theoretic and practical importance to determine both the absolute and the relative "points of no return" for all kinds Psychological Aspects of Ageing 213 of skilled response, within fairly wide age ranges. One charae- teristic of behaviour in the middle age ranges and upwards is an unusual persistence of mistakes. If a mistake has been made once or twice there seems to be a strong tendency for the performer to move to the required response by way of the imwanted one, even when it is perfectly clear that the latter is unwanted (Kay, 1951). This is just what would be expected if the course of an action is determined at a relatively early stage of the action. Supposing that there were both some retraction of antici- pation range, and some advance of the "point of no return'' all the slowing-up of skilled work which has been experiment- ally demonstrated would be fully accounted for. Whether they do occur and if so by what human mechanisms they are effected are questions perfectly amenable to an experimental approach; but the experiments have not yet been done. One final question about these results of which I have spoken and a good many others, both in practice and in training, which I might have discussed had there been time: are they inevitable? That depends upon the meaning which is attached to "inevitability". If "inevitable" means results that will occur unless special precautions are taken, they are. If "inevitable" means results that are bound to occur what- ever is done about them, simply because everybody advances in age, the strong probability is that they are not. But that opens up another, and perhaps a still wider, field for experimental study. REFERENCES Kay, H. a. (1951). Quart. J. exp. Psychol., 3, 1G(3. Leonard, J. A. (1953). Quart. J. exp. Psychol., 5, 141. SzAFRAN, J. (1951). Quart. J. exp. PsythoL, 3, 111. Welford, a. T. (1951). Skill and Age. Oxford University Press for the Nuffield Foundation, p. 26. DISCUSSION Shock: In investigating problems of this kind, can one deal with a wide variety of skills, or is it necessary to confine the tc.its to fairly definite acts? In other words, how broad an experience can one dea' \Wth experimentally in this kind of approach? 214 General Discussion Bartlett: For the most part, our initial work was carried out with skills that involved the notion of tracking. You have an object which is moving in a given direction, and its movement either has to be pro- gressively followed or controlled in some way. Then there was an ex:- tension from this, and the things that I have spoken of have been tried out with a considerable variety of bodily skills, with a good many mental skills involving memory and anticipation processes proper, and they have also been taken into the factory and applied to repetitive activities such as are involved in assembly work and conveyor belt work. The same features have occurred over and over again. The thing which is variable is the interval between a movement that is required to locate something in one position, let us say, and then take it from that position to another. The same thing occurs in connection with a lot of skills in games, which we have studied. It also occurs in a good many operative procedures in the laboratory, where what you have got to do is to locate a particular tiling that you want to do something about and then move it into a given position in relation to something else. A skilled process always involves a sequence of movements, mental or physical, inter- spersed with changes of direction, and it is the halts between changes of direction that are the most unstable feature; it is these which are most affected when what we call tiredness or fatigue sets in, which are most influenced by drugs, and which become more marked with increasing age. Franklin: Is there any difference between the sexes? Bartlett: None at all, as far as we know, but most of these experiments have been done with men in the factory and in the laboratory. But wherever we have had older women they have shown the same character- istics, and there doesn't seem to be any important difference at all. Lewis: Have any of these studies been carried out on people whose mental powers were obviously failing — perhaps in the early stage of dementia? Bartlett: No, we try to keep away from that as far as possible and to have a range of people who would satisfy the requirements of normal mentality. We have kept off people with special disabilities because we think that complicates the picture at this stage. I think it possible that you would have these things happening, but a good many other things superimposed on them as well. Lewis: I was wondering whether, since these are central happenings, they would be more evident in people with dementia. Bartlett: Yes, it could be, but I don't know about that. Franklin: How far does the ageing of the eye come into this? Bartlett: Since we have tried to keep to people with normal eyesight, we have no evidence that anytliing very much is dependent upon visual responses as such, except that the optimum distance of the signal for action from the viewing point will tend to go a bit further away. Shock: Do you think that the way to look for age changes is in an activity that is well practised in the individual, one that he has done repeatedly? Or are you more likely to see age differences when you pre- sent him with an entirely new activity? General Discussion 215 Bartlett: It is quite true that from the fifties upwards, the attempt to introduce new conditions of performance, new types of stimulation is Hkely to be met by a relatively long learning stage. But I think that is about the only thing that we can say about that. Most of the work that we have been doing involves repetitive activities which have been completely learnt. J. Verzdr: In that connection, and relative to Dr. Welford's work, we have been doing some maze experiments with rats. We have tried to study two problems: learning at different ages in a multiple T maze, and relearning or memory of it at different stages during life. We have found slightly lower learning in extreme old age, but the most interesting thing was that certain of the old animals, about 40 per cent of the groups, were not able to learn at all when presented with the maze for the first time in old age, whereas the other 60 per cent learnt it within the normal range, although rather more slowly than the young rats. The second point was that in memory tests with groups of rats which had learnt the maze at different ages, again in extreme old age about 40 per cent forgot the task which they had learnt in earlier life, and were quite unable to remember it, whereas the other 60 per cent of that same group remembered as well as the younger rats. I don't know how far that correlates with the work of Dr. Welford. Bartlett: I didn't say anything about the learning side of all this, because if one doesn't take special precautions what you discovered about the rats is equally true about human beings. That is to say the learning processes which are characteristic of middle and older age are quite different from those which are characteristic of the relatively young. If the older people are presented with the same conditions of learning as the young, they either learn less quickly or perhaps fail to learn at all. But so far the evidence is that this is more a question of the conditions under which the tests are given than it is of any absolute capacity to learn. It seems extremely likely that as far as human beings are concerned, there is no particular reason why people shouldn't go on learning things as long as they live, or at any rate up to extreme old age, provided they are given conditions of learning which are suitable for their particular range. I think, too, there is a great deal of agreement between the results about relearning which you have indicated with the rats and similar cases with human beings. The only thing about it is that it seems as if you can get rid of all these differences if you adjust your learning conditions to the age range that is conqerned. It is very difficult to do that with rat experiments, I think, but a little easier to do it with human experiments. The whole progranune of our unit was to be concerned also with learning, and we have got a good deal of evidence about the learning processes which are characteristic of age ranges, and they certainly are different in human beings for different age ranges, but I haven't time to say anything about that. Parkes: I suppose in the case of the animal experiments the stimulus applied to learning is the same for young as for old, is it? Loss of food, or falling into a tank of water, or whatever? 216 General Discussion J. Verzdr: Yes. It was hunger motivation, not very extreme, daily fasting. Parkes: And tliat again makes the penalty for failing to learn quite different from what might be applied to human experiments. J. Verzdr: Yes. But we were interested in these individual differences, that some old beasts could still remember and others had total loss of memory. Parkes: Does that kind of sanction mean the same to an old animal as to a young one? J. Verzdr: There is admittedly the possibility of an age difference there. Parkes: If I remember rightly, Sir Frederic, one of the original ideas behind this project was to find out what particular jobs old people might be suitable for. I take it that the answer is still rather in the future, is it? Bartlett: That is quite true, but I think the very general answer is that provided one avoids the difficulties inherent in a great many current industrial processes of rapid forced pacing, there are probably practically no jobs for which the upper age ranges are not suitable. Parkes: Provided they are taught appropriately or allowed to take things at their own pace? Bartlett: Provided conditions of operation are readjusted at the right age range. If one looks at the structure of British industry — if we are talking about the practical side of this — at the moment, one of the most astonishing things is (it is reported over and over again, and is I think quite definitely established) that there is a very great proportion of older people doing heavier work and doing it successfully. I didn't say anything about heavy work, but in point of fact if a man remains fit, he seems to be perfectly capable of going on doing heavy work, provided he can do it in his own time. A great amount of the heavy work in industry at the moment is done at the operator's own pace. So that is one of the many reasons which tend to make the heavy work be passed over to the older people. Parkes: Does that mean that old people now are more capable of doing heavy work than old people were previously? Bartlett: No. I don't think there is any reason to think they are more capable of doing heavy w ork. They are more capable of doing slow work, and most heavy work is slow work. There are other reasons — ^one is that all the heavy jobs in a modern industrial community are relatively unmechanized, and all the forces of current education are in favour of people getting into higlily mechanized industries early, leaving the people who haven't had that kind of training to do the unmechanized jobs. That is going to become much less possible every year because there are fewer and fewer of such operations. Therefore if one is looking at this thing in a practical sense, the really important thing is not to bother very much about the heavier work at the moment but to find out what could be done, given proper conditions, with the type of activity which is obviously going to predominate in the community in a few years' time. General Discussion 217 Parkes: In general, are assembly-line types oi" jobs unsuitable for old people? Would it be too awkward to ask whether there is any evidence that the inereasing expectation of life at birth is bringing an}- increased expectation of potential working life? Bartlett: I don't think it would be possible to give an absolutely categorical answer to that. The indications are that the increasing expectation of life has already lengthened the potential working life, when you mean by the potential working life the period during which an individual may be of economic service in a community. I think this is true, but it is an opinion rather than anything that can be definitely proved. But it seems to me certain that the two things must go to- gether, because the increasing expectation of life is simply an expression of increasing fitness of human beings. At least I think that is true. I don't believe — ^this is again a matter of opinion — that the increasing expectation of life is very much tied up with the fact that the older people are being better looked after, I think it is that they are potentially much more active than they used to be. Parkes: The increasing expectation of life is presumably an expression of the increasing control of diseases which killed children, or adults in their prime, not of the control of senescence. Bartlett: The question of whether in fact, because people could go on working for longer, with satisfaction to themselves and with economic gain to the community, they will be expected to do that, is quite another matter, of course. I think they will have to be. Schulze: I msh to mention a problem of the psychology of ageing, which seems to be remarkable to me, that is the changing of the psychical conception of the speed of time as a function of ageing. All of us will remember that, when children, we considered a holiday lasting six weeks an almost infinite space of time. As adults we. experience now an apparent shortening of our idea of this period. In old age we shall feel the same space of time to be a trifle. This fact may correspond to the change of our mental attitude to the course of time in general. In childhood we are living merely in the present and actual events seize upon our fancies. When we gTow adult, the future achieves more and more predominance in our minds, whereas the elderly are looking back and lose their power of recollection of recent events. This quick-motion change of our spiritual idea of time flow is a very interesting phenomenon, and I wonder if any experimental work has been done in this field to realize the acceleration of our "internal chronometer". Bartlett: I think that there are a lot of investigations about this, but in so far as these have been experimental they are not terribly con- vincing. It is very difficult to find any way by which you could get experiments giving clearcut measurable results. Shock: This is rather an expression of the Weber-Fechner law, isn't it? The boy's total lifespan is ten years, so that proportionately a week represents a large part of it, whereas the fifty-five-year-old individual has a lot of years to base his judgment on, and a week is only a smaU portion of his total life experience. Accordingly, one would expect the 218 General Discussion boy to regard a week as much longer than would the fifty-five-year-old man. Bartleti: I think the truth is that anticipated time and remembered time — if we mean by time a time span as measured by a clock or some- thing of that kind — appears longer to the young and shorter to the old; but experienced time, that is the span within which we live, appears shorter to the young and longer to the old. To a young child, what to an old man is a small period of anticipated time seems long, or a small period of remembered time may seem long; but to the old man what is a relatively short period of lived time may seem to be long relative to what it is in the case of the child. However, usually this is held to be dependent upon the number of items that fill up an interval. You can do this experimentally, if you like; if you have an interval which is started by a sound and terminated by another sound, and if you fill this with a series of sounds, then the apparent length of the interval is dependent to a large extent on the number of sounds that are used to fill up the interval. If you don't fill it up at all, then it always seems a much longer interval. I think that is true for all ages, but in general the intervals in a young life are less fully filled up than they are in the older fife, until you get to retiring age, when time then may appear to go more slowly. Leivis: I think it has been found that people's subjective impression of how time is passing is not related in any recognizable way to their actual judgment of how much time has passed between two specified intervals. Bartlett: Yes, that is the case. ADRENOCORTICAL REACTIVITY IN AGED SCHIZOPHRENIC PATIENTS* H. Freeman, m.d., G. Pincus, Ph.o., F. Elmadjian, Ph.D., AND Louise P. Romanoff, a.b. Worcester State Hospital, Worcester, Mass., and Worcester Foundation for Experimental Biology, Shrewsbury, Massachusetts Previous investigations by Pincus et al. (1949) and Hoagland et al. (1953) have described abnormal findings in adrenocortical responsivity in young and middle-aged schizo- phrenic patients. The present paper deals with a study of adrenocortical reactivity in elderly normal and schizophrenic men to determine whether senescence modifies the results obtained in the younger groups. The normal subjects included 34 men, ranging in age from sixty to ninety-one years with a mean of seventy-six years, who were obtained partly from the community-at-large and partly from a home for the aged. All were active and were free from any obvious disease except the usual stigmata of the ageing process. The patients, 33 in number, were all residents of the Worcester State Hospital. They ranged in age from sixty to eighty-one years with a mean of sixty-eight years. They had been confined to the hospital an average of 31 -3 years, the range being 10-1 to 49-6 years. The majority of these patients were of the paranoid type, with a scattering of the other usual subtypes. All the patients likewise, for their age, were in good physical condition. They were a selected group in so far as co-operation was an essential element for the completion of the studies. Data on young subjects are included in this study to exempUfy specific points as to the ageing process. ♦Investigations described in this paper have been aided by grants from the Scottisli Rite Committee for Research in Dementia Praecox, the United States Pubhc Health Service (H-1659 and M-G57) and by an institutional grant from the American Cancer Society. 219 220 Freeman, Pincus, Elmadjian and Romanoff Two test procedures were used: (1) the injection of 25 mg. of ACTH intramuscularly as a direct means of studying the effects of the stimulated adrenal cortex, (2) the ingestion of glucose, using the Exton-Rose technique, which acts on the adrenal cortex presumably via the anterior pituitary. In the ACTH test, under fasting conditions, measurements of lymphocytes and eosinophils were made immediately prior to the injection and one-half, two and four hours after the injection. Urine collections were made from the period of awakening (after discarding the first morning urine) to the time of the injection (average time two hours), for two hours after the injection, and for another two hours. In the case of the glucose tolerance test, the lymphocyte and eosinophil counts were made before the ingestion of the first 50 grams of glucose, a half hour later just before the ingestion of the second 50 grams of glucose, a half hour and two hours after this. Urine collections were made for two hours before the onset of the test and collected during the first hour and the succeeding two hours after the ingestion of the first dose of glucose. In the urine collections analyses were made for creatinine, sodium, potassium, 17-ketosteroids, corticosteroids as neutral reducing lipid and inorganic phosphates. The two tests were done three days apart, the glucose tolerance always being performed first so that no residual effects would be obtained in the second test. Before discussing results which deal largely wdth urinary metabolites, a first consideration in the aged should be an evaluation of the efficiency of the renal excretory apparatus. Previous investigations by Shock (1946), Da vies and Shock (1950) and Miller, McDonald and Shock (1951) have indicated a decrease in the functional capacity of the kidney in the aged. It is with this point in view that we have analysed some of the data concerning the excretion of creatinine. The mean values for urinary creatinine during the ACTH and glucose tolerance tests are shown in Table I. The subjects are divided into two groups on the basis of age, and further subdivided into normal and schizophrenic men in each of the Adrenal Activity in Aged Schizophrenics 221 age categories. The l)asal values for the young groups are ahiiost twice as great as the corresponding values for the old groui)s. This is to be expected on the basis of the greater muscular mass of the young subjects, although a renal factor may be implicated. The normal subjects, both young and Table I Mean Creatinine Values for Young and Old Normal and Schizophrenic Men Durinc; ACTH and Glucose Tolerance Tests ACTH Age years Subjects Urinary Creatinine {^./24 1 r.) Basal 2hr. -i hr. 19-39 Normal Schiz. 1 • 8222 1 • 5428 1 • 8521 1 • 572" 1-8321 1 • 4228 (50-90 Normal Schiz. 1-0819 0 • 83^5 1-4019 1-2P5 1-2019 1 - 3730 GLUCOSE Age years Subjects Urinary Creatinine (^./24 hr.) 1 Basal 1 hr. • 3 hr. 19-39 Normal Schiz. 1-802 5 l-(>929 1-952* 1 - 6529 l-952« 1-7129 00-90 Normal Schiz. 1 - 0423 0-91^1 1 - 3723 1-4131 1-4521 1 - 4412 N.B. Raised figures indicate nuinber of subjects. old, excrete slightly more creatinine in the basal state than the corresponding patient groups, but the difference is statisti- cally significant only between the two groups of young subjects. The administration of ACTH results in no essential change in the values for the younger groups but does show a definite increase in the amounts of creatinine excreted in the two 222 Freeman, Pincus, Elmadjian and Romanoff older groups. An analysis of the change in values from the basal level to the two and four hour readings shows a statisti- cally significant difference between tlie young and old schizo- phrenic subjects but not between the yoinig and old normal men. The greater excretion of creatinine following ACTH in the aged may be influenced by the differences in the initial levels between young and old. There is a negative relationship between the basal values and the change in four hours so that the lower initial values show greater changes than the higher basal values. Correlation coefficients for the four groups are as follows: young normal, —0-23; young schizophrenic, —0-66; old normal, —0-67; old schizophrenic, —0-66. Only in the young normal group is this relationship not significant. The data on creatinine excretion following the ingestion of glucose show the same trend as with ACTH (Table I). Again the young subjects, both normal and schizophrenic, exhibit no significant change in values over the experimental period while the old subjects show a distinct increase. The increase in excretion at the three-hour point is significantly higher in both older groups than for the corresponding- younger subjects. Further, the relationships between the basal values and the changes at three hours are of the same order of magnitude as in the case of ACTH. It may be that the initially higher values of the young subjects represent not merely the degradation products of greater muscular activity but also a more efficient renal mechanism which, under stress, is able to maintain a given level of excretion. The impairment of renal function in the old may be manifested in an inability to resorb the increased amount of creatinine presented by the stimulating agent. Not merely does creatinine vary but the other urinary metabolites vary with it. In Table II are shown the relation- ships between basal values for creatinine and the other functions studied, in the old normal and schizophrenic subjects. The correlations are of a good high order of magnitude, the highest relationships being obtained with phosphates, sodium Adrenal Activity in Aged Schizophrenics 223 Table II Relationship Between Basal Values for Creatinine And Other Urinary Metabolites in Elderly Men Urinary metabolites Cnrr. coefficients* Creatinine vs. 17-ketosteroids ..... ,, vs. cortin ...... ,, vs. uric acid ...... ,, vs. sodium ...... ,, vs. potassium ...... ,, vs. inorgan. phosphates .... + 0-G8 + 0G5 + 0-92 + 0-84 +0-07 +0-73 *AI1 correlation coefficients are statistically significant. and uric acid. This degree of co-variance furnishes additional evidence for the renal origin of creatinine variation in the aged. As a result of the excellence of the relationships, we have analysed all the urinary data on the basis of creatinine ratios and they will be so presented. The mean values for the two groups of subjects during the ACTH test are shown in Table III. The normal subjects have higher initial levels of eosinophils, a finding noted in younger groups (Pincus et al., 1949). The difference is not statistically significant, however, in this group. The eosinopenia follows similar trends in both groups, again with no statistical significance. This, too, has also been noted in younger groups of subjects by Pincus et al. (1949) and Hoagland et al. (1953). The response of the eosinophils to the administration of ACTH cannot be used, therefore, as a point of differentiation between normal and schizophrenic subjects at any age. Ageing itself, apparentl3% has no effect upon the basal eosinophil level according to Olbrich (1947) or upon the eosinopenic response to ACTH (Solomon and Shock, 1949). In connection with the magni- tude of this response (a 48 per cent fall) note should be taken of the diurnal variation in eosinophils observed by Stevenson, Metcalf and Hobbs (1953) that from 8.00 a.m. 224 Freeman, Pincus, Elmadjian and Romanoff to noon there is a decrease of 21 per cent in the eosinophil count. The lymphocyte values (Table III) likewise show no sig- nificant differences either in initial or in the post-injection Table III Values for Bi^ood and Urinary Metabolites in Elderly Normal and Schizophrenic Men During ACTH Test J'ariahle Subject Basal i hr. 2 fir. 4 Ar. o s Eosinophils Norm. Schiz. 126" 18()^» lOS*" 159*« 8734 123^^ 01" I^ymphocytes Norm. Schiz. 2527" 293.'i»* 249129 2394^* 221429 2148" 19242* 2079" a c .2 t a >, i 17-KS (nig.) Creatinine (g.) Norm. Schiz. 3 • 64" 5 17'^ 3-83" 4-67" 4-55" 4-73" Cortin (mg.) Creatinine (g.) Norm. Schiz. *2 • 3-4»' *3 13»2 2 14" 3.1732 2-54" 315" Uric Acid (mg.) Creatinine (g.) Norm. Schiz. 455" 440='* 512" 490" 574" 51929 Sodium (mg.) Norm. Schiz. 3324>"' 3312" **4777" **3524" **5499" **298G28 Creatinine (g.) Potassium (mg.) Norm. Schiz. 2555>9 2352" 3301" 2G02" **4987" **302728 Creatinine (g.) Phosphates (mg.) Norm. Scliiz. *577" *287»'' ** 542*9 ** 413" **535" **4422s 1 Creatinine (g. ) ♦Statistically significant difTerences between control values in the two groups. ♦♦Statistically significant differences between the changes in the 2 or 4 hour values from the basal levels between the two groups. N.B. Raised figures represent number of cases. values between the normal and the schizophrenic men. This ne^yative finding has also been noted in younger groups of subjects (Hoagland, et al., 1953). Ageing apparently does not affect the lymphopenic response to ACTH, as both Solomon and Shock (1949) and Pincus (1950) have found. In regard to the 17-ketosteroid/crcatinine ratios (Table III) the basal levels of the patients are higher than those of the Adrenal Activity in Aged Schizophrenics 225 normal subjects as is true in the younger groups studied by Fincus et al. (1949) and Hoagland et al. (1953). The response to ACTH is an increase in the normal subjects and a decrease in the elderly schizophrenic patients. There are, however, no statistically significant differences between the normal subjects and the patients either in the control values or in the trends after the injection of ACTH. A similar lack of differences in reactivity has been noted between young normal and schizophrenic men (Pincus et al., 1949). The difference in trends between the normal and schizo- phrenic men may be due, as has been attributed in younger subjects (Pincus et al., 1949), to a factor of adrenocortical dysfunction. It may also be related in this older group to the difference in the initial levels between the two groups of subjects. While in the younger subjects there is no relation- ship between the basal values and the absolute change at four hours, in this group there is a negative correlation of high value {r= —0-70). The normal subjects with their initial low values show increases at four hours, while the patients with their initial high values show decreases. This relationship holds good as well for the 17-ketosteroid values alone as for the creatinine ratios. The basal values for the neutral reducing lipids or cortins (Table III) show a significantly higher level in the elderly schizophrenic patients. While the normal subjects show a greater rise after ACTH than do the patients, the trends are not significantly different. This is true for younger subjects also (Pincus et al., 1949). Again, the slightly greater reactivity of the normal group may be due, at least in part, to the difference in the initial levels. Ther.e is a negative relationship (/■= —0-47) between the initial values and the absolute change at four hours, so the normal subjects, initially lower, exhibit greater increases than the patients. In the values for the uric acid/creatinine ratios (Table III) there is no difference between the initial levels of elderly normal and psychotic men. After the administration of ACTH there is an increase in the excretion in both groups 226 Freeman, Pincus, Elmadjian and Romanoff which is greater in the normal group than in the patients, but not to a statistically significant degree. In younger individuals the normal subjects show a significantly greater response (Hoagland et al., 1953). The greater reactivity noted in the response of the normal subjects is not due to any influence of the initial level. No relationship can be demonstrated between the control levels and the magnitude of the change at the four-hour reading. The sodium/creatinine excretion data in Table III show identical basal values for the two groups of subjects. In younger subjects, patients usually excrete more sodium (Pincus et al, 1949). Their response to the injection of ACTH is distinctly different. The normal subjects show a steady rise, more marked in the first two hours, the final level reached being an increase of 65 per cent. In the case of the patients, there is an insignificant rise at the two-hour reading and then a decrease in excretion to a point slightly below the initial leveJ. At both the two- and four-hour points, the response of the normal subjects is statistically significantly greater than that of the patients. A similar lesser responsivity is characteristic of young schizophrenic patients also. Again, the responsivity of sodium to ACTH is independent of the initial level and, since both groups have the same mean basal values, the difference in trends is therefore more striking. Turning next to the data on potassium-creatinine ratios (Table III), we see that in the basal levels the patients excrete a lesser amount than the normal subjects. The reverse is true in younger subjects (Hoagland et al., 1953). After the administration of ACTH, the excretion is increased in both groups, but significantly more in the normal subjects. In this respect again, the greater response of the elderly normal subjects as compared with the patients is a reflection of what has been noted in younger subjects (Pincus et al., 1949). The initial level plays no role in the greater responsivity of the normal subjects, there being no relationship between the control values and the amount of change at the four-hour point. Adrenal Activity in Aged Sctiizopitrenics 227 The phosphate/creatiniiie Aahies (Table III) show that the normal subjeets exerete tw iee as much inor<;»anic phosj^hates on a creatinine l^isis as do ])atients, a statistieally sionifi- eant difference. After ACTII the normal subjects show no increase but rather a mild decrease; the patients show a striking increase. This difference in trends is statistically significant. The lesser excretion of phosphates, in the basal state, in schizophrenic patients is a consistent phenomenon through- out the various age decades (Hoagland et al., 1953). The excessive phosphate excretion of the patients after the injection of iVCTH has also been noted in the younger groups (Hoagland et aL, 1953). To what extent this is a feature inherent in the schizophrenic psychosis is, however, a question. Like the 17-ketosteroids and the cortins, there is a negative relationship between the initial levels and the amount of change after ACTH, the correlation coefficient being —0-64. For phosphate values alone, r= —0-55. Since the normal subjects have the high initial levels they show little effect from ACTH, while the schizophrenic patients with low control values show increases. The greater reactivity of the patients to ACTH may thus be just a reflection of their initial levels. In normal men, there has been noted in all the urinary metabolites a decrease in excretion values with advancing age (Pincus, 1950; Kirk, 1949; Hamilton and Hamilton, 1948; Kowalewski, 1949). A point of interest is whether ageing affects schizophrenic patients in this regard as much as normal individuals. In Table IV we have shown the mean values for the various constituents in young and old normal and schizo- phrenic groups. In order to minimize the intra-individual variation the control values for several tests have been grouped for each subject and an individual mean taken. The results of the change from the young to the old group are expressed on a percentage basis in the last line. Several points are of interest. The creatinine and phosphate values are lower in the patients consistently. The sodium and 228 Freeman, Pincus, Ei.madjian and Romanoff Table Means of Basai, Valuks i'ok Urinary Constituents Kon . toe oroiiiis Years Cr (g.l2 'lit. 4hr.) 17.KS (ina.l hr.) (Drtin (m^./24 hr.) 19-39 N S N S N S 1-80158 1.63108 0-54153 0.70110 3 02148 2.29i«« 60-90 1 • 06^2 0 • 87«» 0 1942 0.1768 2-1139 2- 45" Per cent change Iroin voung to old ' . -41 -47 -65 -76 -30 + 7 N.B. Raised figures represent luiniber of tests. potassium values are higher in the young patients than in the young normals, but not in the old patients. The cortin values arc lower in the young patients than in the opposing group, but there is no essential difference between the two groups in the old population. The uric acid values are essenti- ally the same for the two groups for corresponding ages. The percentage changes show that with increasing age, decreasing excretion appears in approximately the same proportion in the normal and schizophrenic groups in the creatinine, 17- ketosteroids, uric acid and phosphates. The cortins show no decrease in the patients, an indication possibly that this portion of the adrenal cortex is functioning at a low level consistently throughout life. The sodium and potassium decrease more in the patients than in the normal subjects, primarily because of their higher values in youth. Thus, it is only in the corticosteroids and minerals that advancing age affects the schizophrenic subject differently from the normal subject. It is evident from the percentage changes that ageing could not be expressed with the creatinine ratios, since, with the exception of the 17-ketosteroids, the creatinine decrease is sufhcient in magnitude to nullify or even reverse the tendency to a (iecreasino excretion with aoe. Adrenal Activity in Aged Sc'irizopiiRENus IV Normal (N) and Schizopuhknic (S) Mkx at Two Aok Lkvkf,s 2-2!) i'ric Acid (mg.lmin.) Sodium {rng./hr.) N Potasaium (mg.lhr.) Pliosplidtr (mg.lmin.) {)-5V^ {)3V 39 OoO^ 1901 306" IG"; lin 0-28«« 138^ 134«3 112^2 68" 44 ■27 00 33 (i.j 0-541 0 • 20»" 0-30" 0 17" 33 A further point of interest is the question as to whether schizophrenic patients show a change in their reactivity to ACTH as they grow older. Previous studies by Pincus (11)50) and Solomon and Shock (1949) on ageing in normal men show that the adrenocortical responsivity is essentially unaffected, although the latter authors observed some deficiency in the renal handling of the uric acid. Analysis of the data for young and old normal and schizophrenic men bears out this con- clusion for patients also. In Table V are shown the percentage changes from the basal values four hours after the injection of ACTH in young and old men of the normal and psychotic groups. It is evident that for the normal subjects the only marked changes in ageing are a lesser reactivity in the excretion of uric acid and in potassium. In the patients, as a result of senescence, there is a moderate increase in tlie potassium excretion and a decrease in the phosphate excretion. We may conclude, therefore, that on the whole, schizo- phrenic patients show the same degree of reactivity to ACTH in old age that they do in youth and that whatever aspects of lesser responsivity they exhibit are preserved throughout life. Inter-ivhitioiiships between the effects on tlie various functions studied have been notably lacking. It seems that 230 Freeman, Pincus, Elmadjian and Romanoff each function operates independently of the others. How much of this is dependent on technical variation and how much is due to inherent differential effects of adrenocortical reactivity upon the various target organs or physiological systems is impossible to assess. It does, however, make it difficult to determine causal relationships. Table V Mean Values for Per Cent Changes from Basal Levels in Urinary Metabolites/Creatinine Ratios Four Hours After the Injection of 25 MG. ACTH IN Elderly Normal (N) and Schizophrenic (S) Men Age Groups Years \7-KS Cortin llru: Acid Sodium Potassium Phosphates Great. Great . Great. Great. Great. Great. N S N S N S N 5' N s N .S' 19-3!) + 25 0 + 21 -13 + 64 + 11 + 56 - 9 + 124 + 7 -16 + 73 60-90 + 25 -8 + 9 + 1 + 26 + 18 + 65 -10 + 95 + 29 — 7 + 54 The data on the response to glucose is shown in Table VI. So far as the blood sugar is concerned, the trends in the two groups are quite similar except that the patients show a significantly greater elevation at the one-hour reading than do the normal subjects. This decrease in tolerance is charac- teristic of schizophrenic patients also (Pincus et al., 1949; Freeman and Elmadjian, 1950). One of the usual concomitants of ageing is a tendency to show a diabetic-like blood glucose trend, whether sugar is given by mouth (Smith, 1948) or intravenously (Smith and Shock, 1949). In young normal subjects the thirty-minute value is usually higher than the sixty-minute reading. In old normal subjects the thirty-minute reading is higher than that of the younger individuals and the sixty-minute value is further elevated over the thirty-minute one. In young schizophrenic patients, the sixty-minute reading is usually higher than the thirty-minute level, and in old schizophrenic subjects the trend is exaggerated. The factors causing the reduction in glucose tolerance of the schizophrenic patients are unknown. Adrenal Activity in Aged Schizophrenics 231 One possible cause has been thought to be emotional tension which has acted via the hypothalamus on the anterior pituitary, stimulating the production of corticotrophins and adrenocortical steroids resulting in giuconeogenesis from Table VI Values for Blood and Urinary Constituents in Elderly Normal and Schizophrenic Men During Exton- Rose Glucose Tolerance Test Variable Subject Basal Mr. Ihr. 3hr Sugar Norm. Schiz. 104="' 9133 149'" 151" ♦ ♦16330 ♦♦178" lliso 95" Eosinophils Norm. Schiz. ISS^" 128" 17030 108" 165^» 102'" 142" 106" Lymphocytes Norm. Schiz. 2932" 2949''» 2071" 2083" 2185" 1990" 2539" 2813" (1 3 O •t <>* O a 17-KS (mg.) Norm. Schiz. 4-23" 5 IS" 4 06" 4-51" 3-84" 3-87" Creatinine (g.) Cortin (mg.) Norm. Schiz. *2-1022 *3 15" 1-60" 2-56" Creatinine (g.) Uric Acid (mg.), Creatinine (g.) Norm. Schiz. 385" 497="^ 441" 507" ♦♦50120 **430" Sodium (mg.) Norm. Schiz. 3007^3 3792" **3213" **2700" **3354" ♦♦2782" Creatinine (g.) Potassium (mg.) Creatinine (g.) Norm. Schiz. 2500" 1844" 2465" 1592" **2329" ♦ ♦115121 Phosphates (mg.) Creatinine (g.) Norm. Schiz. ♦47622 ♦ 28132 **493" **503" ♦ ♦403«» ♦ ♦32022 ♦Statistically significant differences between basal values in the two groups. ♦♦Statistically significant differences between the changes in the 2 or 4 hour values from the basal levels between the two groups. N.B. Raised figures represent number of cases. protein. In older people, it may be that the elimination of sugar through the kidney and not the production is at fault (Shock, 1946). At any rate, both the factors involved in the psychosis and in ageing result in the maintenance of a reduced sugar tolerance in the elderly psychotic group. In the eosinophilic response (Table VI), the normal subjects again have a higher (but not significantly so) initial value. 232 Freeman, Pincus, Elmadjian and Romanoff The trend is steadily downward over the four hour period. There are no significant differences between the two groups of subjects in this regard. A similar lack of differentiation has been noted in younger groups (Hoagland et al., 1953). Ageing plays no role in the eosinophil response, the decrease being of the same magnitude roughly at all ages (Solomon and Shock, 1949). In view of the fact that the diurnal vari- ation of eosinophils results in a decrease of approximately 21 per cent in the morning hours (Stevenson, Metcalfe and Hobbs, 1953) and that the decrease seen above is about this magnitude, the effect of the glucose as a cause of the eosino- penia is somewhat doubtful. The values for the lymphocytes are shown in Table VI. After the ingestion of glucose the lymphocytes show a drop in values at the half-hour reading, which continues for the next half hour and then returns almost to the original values by three hours. These trends are similar in both groups of subjects, indicating no difference in response. This is true also for younger groups (Pincus et al., 1949). The lymphocyte response is, in the main, a mirror image of the blood glucose tolerance trend. In the time relationship to the glycaemic curve, the lymphocytes react more quickly than the eosinophils. In the case of the 17-ketosteroid/creatinine ratios (Table VI) the patients' initial values are higher (but not significantly so) than those of the normal subjects, a finding also noted in younger subjects (Pincus et al., 1949). In both normal and schizophrenic subjects the values show a downward trend, more steeply for the patients, so that the levels at the three- hour reading are the same. In neither group is there any indication of adrenocortical stimulation. The greater decrease in values in the patients may be due to their initially higher values, since there is a negative relationship (r= —0-60) between the basal value and the absolute change at the three-hour point. So far as ageing is concerned, the response to glucose diminishes with increasing years. The younger normal Adrenal Activity in Aged Schizophrenics 233 subjects, but not the patients, show an increase at the oiie- and three-hour points while, as we have seen, both of the older groups show a decrease (Pincus, et al., 1949). The cortin/creatinine ratios, again as in the case of the ACTH data, show significantly higher values in the case of the patients. In the younger groups, the reverse is usually true (Pincus et al., 1949). There is no response to the glucose, there being a decrease at the one-hour reading in each case, the trend being the same for both normal and psychotic groups. Ageing seems to play no role in the response to glucose in this function, the younger groups being equally unreactive to glucose (Pincus, 1950). The uric acid values in Table VI show initially higher levels of uric acid excretion in the patients, a trend noted in younger subjects (Hoagland et at., 1953). After the ingestion of glucose the normal subjects show an increase in excretion at the one-hour and three-hour readings. The schizophrenic patients at the one-hour show a slight upward trend and at three hours a downward trend. At the three-hour point, the difference between the increase in the normal subjects and the decrease in the schizophrenic patients is statistically significant. This lesser responsivity of elderly schizophrenic patients has been noted also in younger schizophrenic patients (Pincus et al., 1949). The initial levels of uric acid/creatmine excretion bear no relationship to the response following glucose, so that any difference between the trends in the normal and psychotic groups cannot be attributed to the differences in their initial levels. In the values for sodium/creatinine ratios (Table VI) the patients have initially higher levels than the normals, but not to a statistically significant degree. During the three-hour test period the normal subjects show, on the whole, a steady mild upward trend. The patients show a marked fall at one hour, with a subsequent levelling off of the values for the next two hours. The difference in trends l)etween the normal and schizophrenic groups is statistically significant at the one-hour and three-hour readings. This difference in trends between 234 Freeman, Pincus, Elmadjian and Romanoff the two groups is similar to that found in younger subjects (Pincus et al., 1949), but it may be, in part, due to the difference in the initial levels. There is a negative relationship (r= —0-46) between the basal values and the absolute changes at three hours, so that the higher values of the patients tend to sho\v the greatest decreases. The potassium/creatinine values (Table VI) again show the normal subjects to have somewhat higher values than the patients, although the difference is not statistically significant. During the three-hour period the normal subjects show a slight decline in excretion values, while in the patients the decrease is quite marked to a value approximately 40 per cent below the inital level. This difference in trends is significantly different between the two groups. The initial levels have some influence in the degree of response at three hours. There is a negative relationship (r= — 0-40) between the control levels and the absolute change at the three-hour reading. On this basis the normal subjects with initial higher values would be expected to show the greater decrease. That they do not, emphasizes the real nature of the difference in trends between the two groups. The values for the phosphate/creatinine ratios (Table VI) show the characteristically low initial levels of the schizo- phrenic pateints. After the ingestion of glucose, in the normal subjects there is a mild rise at the one-hour reading, with a subsequent fall below the initial level at the three-hour reading. In the patients there is a marked increase at the one-hour reading to the level of the normal subjects and then a marked fall almost back to the original level. The trends at both one and three hours as compared with the initial levels are statistically significantly different between the two groups of subjects. It would seem, then, that in the normal subjects the ingestion of glucose produces a conservation of phosphate in the body, so that less is excreted. In the patients, on the other hand, more phosphate is excreted. Here the difference in initial levels between the two groups of subjects may play a Adrenal Activity in Aged SciiizoniRENics 285 role in this varying" type of reaction since there is a hi,<»iily negative relationship {r= —0-91) between the initial level and the absolute change at the three-hour point. Thus, the patients with their low values would show a rise and the normal subjects with their higher values would show a fall. Again, as in the case following ACTH injection, we may say that the initially low level of urinary phosphates is a charac- teristic of schizophrenia but the reaction to the stress on stimulus is largely dependent on that level and may have little bearing on the dysfunction in the psychosis. Analysis of the data for inter-relationships between various functions again, as in the case of ACTH, proved disappointing. No significant co-variance was noted between any of the variables studied. Discussion It may be said in general that elderly schizophrenic patients exhibit the same differences from elderly normal subjects that young patients do from young normal individuals. There is less tendency in the old to show the increases in ketosteroid and corticoid excretion than in the young, but the differences in responses in the mineral elements are essentially identical throughout the life-span. It is probably true that the renal mechanisms in the aged play a greater role in the responsivity of the organism to the two stress situations employed, but this fact does not obviate the persistence of differential responses between the normal and schizophrenic subjects. The maintenance of these abnormalities in the ageing schizophrenic group poses an interesting problem as to whether the adrenal cortex really suffers from exhaustion over the many years in which these patients have experienced a profound psychotic reaction. It may be, however, that the sheltering of such patients from the stresses of existence out- side may minimize the stimulating effect upon the adrenal cortex of the constant emotional reaction. In any case, it is evident that the physiological defects in these elderly schizo- phrenic patients, whatever their cause, have no deleterious effects upon the life expectancy of such subjects. 236 Freeman, Pjncus, Elmadjian and Romanoff Summary A study was made of the adrenoeortieal responsivity of 34 elderly normal men and 33 aged schizophrenic patients hospitalized for an average period of thirty years. Following the injection of 25 mg. of ACTH and the ingestion of glucose the blood and urinary functions under investigation showed the same degree of dysfunction as had been noted in young schizophrenic subjects. REFERENCES Davies, D. F., and Shock, N. W. (1950). J. din. Invest., 29, 4<)(>. Freeman, H., and Elmadjian, F. (1950). Anier. J. Psychiat., 106, 660. Hamilton, H. B., and Hamilton, J. B. (1948). J. clin. Endocrin.^ 8, 483. HoAGLAND, H., Pincus, G., Elmadjian, F., Romanoff, L. P., Free- man, H., Hope, J. M., Ballon, J., Berkeley, A., and Carlo, J. (1958). Arch. Neurol. Psychiat. Chicago, 69, 470. Kirk, E. (1949). J. Geront., 4, 84. KowALEWSKi, K. (1949). J. Geront., 4, 222. Miller, J. H., McDonald, R. K., and Shock, N. W. (1951). J. Geront., 6, 218. Olbrich, O. (1947). Edinb. med. J., 54, 306. PiNCUs, G. (1950). Psychosom. Med., 12, 225. PiNCus, G., HoAGLAND, H., Freeman, H., Elmadjian, F., and Romanoff, L. P. (1949). Psychosom. Med., 11, 74. Shock, N. W. (1946). Geriatrics, 1, 282. Smith, L. E. (1948). J. Geront., 3, 66. Smith, L. E., and Shock, N. W. (1949). J. Geront., 4, 27. Solomon, D. H., and Shock, N. W. (1949). J. Geront., 4, 802. Stevenson, J. A. F., Metcalf, E. V., and Hobbs, G. E. (1958). Arch. Neurol. Psychiat. Chicago, 70, 802. DISCUSSION Nicolaysen: If you observe the urinary composition from tiie moment you wake up and throughout the day, you observe cychc changes in many constituents. The concentration may be nearly doubled from say seven o'clock up to eleven, and then it falls back again — ^that's in experiments with no food or water intake. Freeman: Well, diurnal variation is something we have to take into considerati(jn always. As a matter of fact there isn't too much good literature on this particular point, because individuals have been studied at various times, but most of the experimenters have not got the data for the exact hours in which our tests were performed. One can say that although diurnal variation played a r61e in modifying the response General Discussion 237 to ACTH or glucose, at least the same factors were present in both normals and schizophrenics, and from that point of view it doesn't alter our facts. Xicolaysen: A long time ago 1 studied these cyclic changes in the urinary composition. Some years later parallel studies were reported in schizophrenics. The results were the same as in normal subjects, never- theless they were interpreted as being due to the schizophrenic state. Freeman: As a matter of fact I think that one should have a control day in which one studies this diurnal variation to evaluate the effect of medication. I think that sometimes the ACTH produces much less response than is assimied, Nicolaysen: I am returning to my point about controls, because in those days it was maintained that pituitrin influenced tubular re- absorption of fixed bases. However, placing my curves over the curves from the experiments in which pituitrin had been injected showed that the two sets of observations were identical. Obviously your conclusions would not necessarily be invalidated, but they might be somewhat modified. Freeman: Yes, I think that has probably not been taken into account sufficiently in all the studies on ACTH. Practically nobody in the American literature has done anything on this. Olbrich: I would like to ask if you have any comparative studies as to insulin sensitivity, say with the blood sugar? Freeman: None in old schizophrenics. The young schizophrenic patients have on the whole shown some insulin resistance. This was also noted by Meduna, that they drop down to a lesser degree in a given time, if you take rapid intervals, and on the whole they don't show as quick or as great a response to a given dose of insulin as normal individuals do. Olbrich: You see the same thing in senile dementias or in any type of neurological lesions with focalizing sites. Frec7nan: It is undoubtedly true that reduced glucose tolerance is found in other types of psychosis, because we have studied manic depressives and other kinds, and it may be just a reaction to a psychosis rather than specific to schizophrenia. Olbrich: What method did you use for the creatinine tests? Rubin: Incubation with sodium hydroxide and determination by colorimeter. It's just a simple determination of creatinine excretion in the urine, as a check on accuracy of collection; we don't give them creatinine. Freeman: This ratio is probably not so important in young individuals because the creatinine doesn't vary with the ACTH or glucose, but it certainly does vary in old people. Shock: A study is just being completed in our laboratory by Drs. Silverstone and Brandfonbrener, who have been investigating the response of blood sugar to insulin in normal old people. We have found that older individuals uniformly show a reduced response to insulin. In these experiments, a standard amount of insulin was administered intravenously along with 25 g. of glucose also given intravenously. 238 General Discussion Blood sugar levels were determined at ten minute intervals for a period of two hours. In all instanees the rate of (lisapi)earanee of glueose from the blood of the older grouj) was substantially less than it was for the young and middle-aged groups. Sinee we also ran standard intravenous glucose tolerance eiuA^es on each subject we were able to show that insulin had a greater effect in the young than in the old. Lewis: Dr. Freeman, had any of these schizophrenics received electroconvulsive therapy? Freeman: No. GENERAL DISCUSSION Tunbridge: I think that Prof. Brull and Prof. Verzar suggested almost complementarily that any assessment of ageing should be related to function, and that the functional reserve or the capacity of the indivi- dual under strain was much more likely to give an indication of ageing. In other words, failure of function could be used as a measure of ageing. Prof. Medawar suggested that there might be a heredity factor, and in addition what he referred to as traumatic or environmental factors. Dr. Comfort, would you say something on the role of heredity? We can discuss other aspects afterwards. Comfort: I have been waiting in the hope that somebody who knows some genetics would say something about this, but I don't feel that we can just let it pass, and I only hope that you will correct me where I'm wrong. I think we agreed that longevity in laboratory stocks such as we use is hereditary, and longevity in man is "hereditary", but it is rather difficult to specify exactly what we mean by that. We know that some strains we work with live longer than others. In Gruneberg's book on the mouse there are two very nice life tables for inbred lines of mice, showing a very distinct characteristic difference between the two strains. And we know that it is to some extent possible to breed for longevity. Strong conducted some experiments on those lines, but in most cases I think it is fair to say he was selecting against "short-evity", or against specific causes of death. And we know that longevity in man runs in families to some extent also. Now the point I want to make is this. Beeton and Pearson (Biometrika, 1901, 1, 50) a long time ago studied the sib-sib and parent-sib correlations of ages in a long-lived family, and they showed very clearly that there was far greater inter-sib correlation than there was parent-child correlation. In the same series, moreover, the parent-child correlation was, I think, only about one- quarter of that which one would expect in the case of stature. That is, the correlation between tall father and tall son was about four times that between long-lived father and long-lived son. Haldane (Ann. Genet., 1949, 14, 288) pointed out that this type of pattern is what you w^ould expect in any case where the heterozygote was inherently fitter than the homozygote, and that does raise some considerable difficulties. I don't want to comment on this at great length, but I wonder if I might show you two figures to demonstrate rather briefly the sort of genetical difficulty that arises. Fig. 1 shows the survival curves of General Discussion 239 eight generations of Drosophila subobscura. They were bred in each generation from the eggs hiid by flies after tlie thirtieth day of iniaginal hfe. The objeet of the experiment was to try to find tlie same thing as Dr. Lansing found in rotifers, and to see whether tlie life-span of the off- spring of old parents decreased cumulatively. It didn't do so in this experiment; there is no significant trend in those figures over eight genera- tions occupying about a year. But even more interesting is the stability of the life-span. In each case we were breeding from the longest-lived Fig. 1. (Comfort). Survival curves oi Drosophila subobscura strain K, bred in each generation from eggs laid by parents after the 30th day of imaginal life. members of the cohort. You'll see that in each case the thirty-day mark comes down to about the 75 per cent mortality point, we had only about a quarter of the flies left on each occasion, and only quite a small pro- portion of those were fertile. But by inbreeding long-lived flies, in a strain which had been in laboratory culture for some three years, and which was genetically fairly stable, we were not able to effect any appreciable increase at all in the imaginal life-span. I've got permission from my colleagues Miss J. Clark and Dr. Maynard-Smith (J. Genet., 53, 172) to show an experiment which they did on the same strain of Hies. Fig. 2 shows the survival curves for (dotted line on the left) the same Kiissnacht strain which was used in my experiments, and (solid line on the left) a strain B which was an inbred line from another kind of 240 General Discussion Drosophila. On the right you have the survival curves under the same K R conditions for the two reciprocal hybrids, - and — I thinj^ that is B K rather an instructive example of the effect of heterosis, of induced hybrid vigour, on the life-span. The degree of increase — I did superim- pose the pictures — is almost exactly comparable quantitatively to the degree of decrease in life-span which was obtained in the rat experiments 100 .^^ ".r — — — --^ 90 \ ■ ■X---V B/K 80 - 70 - \ K/B^N V 60 > - V\ /B \.. ■s \Y ^:y t 50 _ V\ i3 E K-^\ 140 - 30 \ \ \ \ 20 ■ \ V "n;,. 10 - 1 1 1 1 1 ^ -^^^--'-|-» 1 1 ">-■•:.. 1- 10 20 30 40 50 60 70 80 90 100 Age in days Fig. 2. (Comfort) Drosophila subobscura. Survival curves of inbred lines B and K, and of the reciprocal hybrids between them (sexes combined.) Clarke, J. M. and Maynard Smith, J. (1955) .7. Genet., 53, 172. which Dr. McCay published. There was approximate doubling of life- span and no change in the slope of eventual mortality; what has been done is to push the whole thing over, with a much longer plateau. I wanted to show that as an indication of the extremely large effect which can be produced by a genetic factor, in this case heterosis. Now had we started with that particular hybrid cross, and inbred from it, I have no doubt that over five or six more generations we should slowly push the life table back again to the position of curves B and K, the equili- brium position which we have when we have a line that has been partially inbred for some time. That is rather an awful warning in so far as these laboratory stocks are concerned. We can't, I know, go General Discussion 241 here into the nature of heterosis, but I think it is fair to point out that while Dr. Maynard-Smith has shown (,/. Genet., 1950, 52, 152) that in this particuhir case an orthodox Mendehan explanation, a chromosomal explanation, is necessary, vigour is a v ery heterogenous thing, and there are quite a number of instances in which an apparent hybrid vigour can be induced by non-chromosomal means, even, I believe, by cross- suckling or by transplantation of foetuses from one animal to another. There is only one remaining point that I should like to make, and that is that we are working with stocks that have been in the laboratory for some time, and the wild Drosophila, which are not "wild-type" but are actually caught in the wild, have often been noticed to have a great deal longer life-span than our domestic Drosophila. The longest-lived one which Dr. Maynard-Smith has observed had a life-span after catching — it may have lived already for some days before that — which suggested that it would live for about as long as those vigorous hybrids. Of course, we have no evidence as yet that heterosis, hybridization, can produce a greater life-span than is found in the wild animal. I just want to stress the dangers into which we run if we do ageing experiments on inbred lines which are showing the degree of stability and the degree of inbreeding depression which we found in our Drosophila. Brull: I am glad that you have stressed the genetical influence, which is so important in ageing. If you can say that in small animals genetical influence is so gTcat, you could go further in mankind and say that genetics in mankind not only influence duration of life but also the form of ageing. In this field much can be done and should be done. If I may make a suggestion for future research and for future meetings, it would be to have meetings of clinicians who have studied the special field of heredity in man, and cross clinicians with geneticists. Comfort: And thereby obtain some hybrid vigour? Lansing: I think Dr. Comfort has raised a very interesting point in connection with hybrid vigour, and I'm very much impressed with the studies on Drosophila that attempted to repeat the w^ork that we did with rotifers and did not succeed. I wonder if there is a fundamental difference in our material? We of course have flirted with the possibility that there are differences. The rotifer ceases all cell division in the embryo, the newly-hatched rotifer has all the cells it is going to have, they merely get larger. If I'm not mistaken, in Drosophila division in the ovarj^ continues, and so we may have there a very fundamental difference between these stocks of material. The rotifer eggs have time to age in the ovary, they age along with tjie mother, while in Drosophila young eggs are available in the ovary of the older mother. Comfort: And of course here we are only dealing with the part of the life of Drosophila that follows eclosion, when it has completed its develop- ment and has ceased somatic growth, apart from in the ovary. I don't know^ to what extent somatic mitosis continues in Drosophila. An entomologist might be able to tell us; but we are only dealing here with imaginal life. Timbridge: Would you now like to say anything about your definition of ageing, Prof. Franklin? AGEING 17 242 General Discussion Franklin: No, I think it is straightforward as it stands [see p. 31, Ed.^. But I should hke to say that I am glad you have tackled this problem because we should otherwise be like the negro who was chased by a bear and, asked by someone, when he was going at full speed, Mhere he was going, said "Ah ain't goin' nowhere, Ah'm comin' away from some place". Shock: Does a definition of ageing have to be limited to decreasing functions? Franklin: Change of function, surely. Cowdry: Yes, it can be increasing. Krohn: Well, that is the meaning of ageing, isn't it? You use the word "ageing" to mean any change as the organism gets older. You have to use perhaps "senescing" for deteriorative changes. Cowdry: I have a definition. Ageing is change with time in the life cycle. Lansing: Would you care to qualify that and make it change with time in the adult organism? Cowdry: No. Krohn: When would you define somebody as having reached adult- hood? Lansing: When he has reached maximum size. Gross: There should be a distinction between development and ageing. Comfort: In Prof. Medawar's temporary absence I would like to put in a plea for his definition of senescence, as the increase in liability to die with advancing age. It may be proper to distinguish ageing from senes- cence, but in that case I think we can scrap ageing altogether and call it development, because gerontology is an entity which only comes into existence to describe a process human beings don't like, a deteriorative process, and I take it that it is senescence with which we are concerned here. Earlier in the meeting Dr. Lansing made a declaration of faith on the subject of the overall unity of the senescent process. He said that we ought to look for underlying processes which explain all senescence. Up to a point I think that that is so, in that I have no doubt that the exhaustion of enzyme systems in cells which don't divide might provide us with something of the kind. But if we do accept, as I suggest we should, the idea of senescence simply as the increasing liability to die with increasing age, then the most striking thing in comparative studies is its diversity. We have insects in which the mechanical wearing out of the cuticle appears to be an important feature, we have others which don't feed at all and appear to deplete their reserves, and we have some animals, as has already been mentioned, which die automatically after breeding, possibly from depletion; we have the elephant, which, I'm told, dies in the wild state when its teeth wear out. I have no doubt that if you gave the elephant false teeth, he would die of some other senescent process eventually. I don't want to speak out of turn, but I'm somewhat sceptical of this underlying unity of any ageing process; I think we should be empirical about it, and treat senescence simply as a name for that whole group of causes which make animals have a determinate life-span instead of an indeterminate one. General Discussion 243 Lansing: But take the male rotifer: it is born, it has no ahmentary traet and dies of starvation within twenty-four hours after fertihzinj^^ Does he die of seneseenee? I'd rather put him in a speeial eateojory, as a very degenerate eharaeter wlio starves to death in tlie twenty-four-hour period that he is busy fertihzing. Comfort: But you ean't prove, altliough I think ifs highly improbable, that even the mammal does not die from depletion of something. We know^ so little about the processes which actually operate. I really don't see how we can discriminate. Lansing: Well, I rather hope that we can. It seems to me that there are (in spite of my declaration of faith yesterday) in biology a number of bizarre phenomena and organisms which can be pointed out as the exceptions to the rule. When I think of senescence I think of something that happens not to children or to infant rotifers, but to the organism that has become an adult and then undergone some type of change, to wind up dead sooner or later. That's what I mean by senescence. The maturation of the embryo, the new born child, the adolescent, the changes with time prior to maturation to me are not senescence. Cowdry: Yours is the downswing of life, then. Lansing: Yes, after adulthood has been reached. I can't define adulthood too well, and in some cases the changes that occur in adult- hood are said to be improvements rather than losses. Cowdry: You don't have to define it if you just call it the downswing, that implies that after a height you start to go down. Comfort: Do you agree then that for various organisms the factors that contribute to that downswing tend to differ very radically from phylum to phylum? Lansing: I'm not prepared to agree to that. I think we have special cases which bring about death, but not all death is due to senescence. Nicolaysen: If you had the ideal solution, you would know about every cell in the body, what cells are ageing, and then what enzymes are declining, and then you would come to a critical stage where life accord- ing to our orthodox meaning will stop. If we had all that knowledge, we could give a chemical definition of ageing. So we could then work backwards and see how much we know about the vital organs in old age. This might be a useful approach. Lansing: The declaration of faith I made yesterday stems in part from the various types of survival curves that Dr. Comfort showed us. The survival curve for Drosophila was virtually identical with the sur- vival curve that one gets for man, except that the time scale is a little off, one is measured in years and the other in days. If one projects the survival curve for rotifers, it's the same sigmoid curve, except that we telescope it a bit further; and we have an identical curve for mice and rats and every other organism that has been measured, which makes me doubt that there are different mechanisms operative. It would be quite a coincidence if all these processes all expressed themselves in the same way. Comfort: Raymond Pearl plotted a survival curve for automobiles which was again the same shape! 244 General Discussion Shock: X tliink the nrjiunicnt that because two different ijhenomena can be made to fit the sainc jnatheniaticjd fornuilation they liave common j)rocesscs behind them is an extremely hazardous one. Lansinii: I said only that it's a possibility, I'm not prepared to say that we have as many kinds of proto})lasm as we have species. I think there is a common proto})lasm with basic properties of multiplication and o'rowth, decline, irritability and so on, varying in detail, not in principle. Comfort: Does that imply that j^ou think that senescence is probably inherent in all metazoa? Lansing: In so far as we can observe it, it is a natural phenomenon. I would Hke to avoid the implication that because it may be inherent it must be inevitable. Comfort: I was wondering about the case of the sea anemone. At Edinburgh there Mere sea anemones which were in captivity for about ninety years without any change in shape; they apparently grew new cells at one end and knocked off cells at the other end, and they continued to form like a cloud over a mountain. They appear to be an example of indeterminate groM-th combined with definite size. And I see Brien in a recent article (Biol. Rev., 1958, 28, 808) has been claiming the same thing for Hydra. Lansing: But for instance the protozoa were held to be contradictory to this hypothesis until the work was done, and then we found that there is senescence in protozoa just as there is in any other species, imless a particular biological process intervenes, autogamy or conjuga- tion. In the absence of the latter, Paramecia go on to die, following survival curves very much like those for man. Shock: I would agree that protoplasm is probably fundamentally much the same stuff, although we know that various tissues develop different functions, so that their enzyme systeins must vary quite widely between different cells in the same animal. To that extent, I would agree that perhaps if you knew what it was that caused a cell to lose its ability to maintain concentration gradients, maintain its metabolic processes, you would be a long way toward understanding the ageing process. But it seems to me that the techniques that we have for inv^estigating single cells are very meagre. Dr. Cowdry feels that if you take a cell out of its tissue it is no longer a cell. If we accept this position we are limited to unicellular organisms for study, but unfortunately most of these species simply divide and form two new cells so that "ageing" fails to occur. Thus, we are faced with the problem of studying more complex animals or tissue, using both biochemical and physio- logical techniques. Since changes in the environment of the cell, pro- duced by changing the diet of the animal, will often result in alterations in cellular enzymes, it seems to me that perhaps we are going to have to look at the problem of ageing from a number of different levels simul- taneously and not try at the moment to conceptualize the entire f)roblem in one framework. Prof. Medawar has approached the problem from a statistical evaluation of life tables; I am not prepared to accept this a{)proach as the only way out of the dilficulties. I think the examina- General Discussion 2i5 tiou ol" life tcil)les iiiij>ht be an index as to what you were doing to a proeess, but if you are goin5, 00 antispasmodics in, 67 in rats, 67 sex distribution, (58 in Swiss centenarians, 16 Endogenous ageing, 27, oS, 54, 65, 243, 244 in fish, 28, 29 in tortoise, 28 Ephedrine in emphysema, 67 Epiphyses, seahng of, and matura- tion, 27, 29 Erythema, })ahiiar, 82 Exogenous ageing, 12, 18, 27, 58, 54, 65, 242 Expectation of life {sec also Life- span), 201-202, 217 of birds, 4 and retirement, 49, 50 Familial aspects of longevity, 288- 241 Fish, elastase in, 99 endogenous ageing in, 28, 29 Ufe-span, 29 usefuhiess in ageing research, 176 Flexure lines, 12, 13, 23, 27 Food intake see Nutrition Force of mortality, 8, 9, 10, 12, 14 Freezing, skin stored by, 170 tissue preservation, 162-169, 171, 172 of whole body for preservation, 167-169 Functional reserve, 0,') Functional residual capacity, age changes, 60 Ganglion cells, disappearance, 49 Gastric ulcer, incidence in aged, 18, 19, 20 Genetics, 27, 204, 238-241 Geographical distribution ol arteriosclerosis, 24 of cancer, 24 Glucose tolerance tests in aged schizophrenics, 220, 230-236, 237 Gonadotrophins in retarded rats, 30 Grafting of long-stored ovaries, 165- 166 of tissues, problems, 150, 154-157 Grow^th hormone on life-span, 29 Grow^th rate in Lebistes, 29 and life-span, 29, 30, 195 specific, 6 Guinea pigs, usefulness in ageing research, 181 Hcenialoeocciis pluvialis 187 Hsematopoiesis as measure of seiescence, 24 Hamsters, freezing of whole bodv, 167-169 usefuhiess in ageing research. IM Hearing acuity, as measure of senescence, 6 Heart disease in Swiss centenarians, 16, 17 Height as criterion of maturation, 201, 202 nutrition and, 195-199 racial differences in children, 201 Helium dilution method, (»2, 63, 66 Hereditary haemorrhagic telan- giectasia, 85 Heredity and life-span, 27, 204, 238-241 Heterosis, 204, 240, 241 Histological tests of ageing, 24, 25, 53 Homeostasis, 27 Homografts, 155, 166, 171 Hormone sensitivity of ovary, age changes, 142-146 Hybrid vigour, 204, 240, 241 Industry, older people in, 216 "Inevitability" of psychological age changes, 213 Infection, resistance to, as measure of senescence, 6 Injury, recurrent, 12, 13 Insects, usefulness in ageing research, 175 Insemination, artificial, of long- stored sperm, 171 Insulin tolerance in aged schizo- phrenics, 237 "Intellect and pathological changes, 49 Intelligence, decline with age, 36, 37, 38, 39, 52 tests in aged, 37, 38, 42 Involutional melancholia, 44 Isotopes in elastase studies, 100-101 Keratosis, senile, 108 17-Ketosteroid excretion in aged schizophrenics, 223-236 252 Subject Index 17-Ketosteroid excretion, chauycs with age, 12(5-137, 188, 140 ^ sex difference, 139 17-Ketosteroids, relation to adreno- cortical steroids, 128, 129 Learning in old people, 41, 215 experiments in rats, 215-21(5 Lebistes, «Towth rate, 29 life tables, 29 Life expectancy, 201-202, 217 and retirement, 49, 50 Life-span of birds, 4 and delayed maturation, 27, 28. 29, 30," 201, 204 and eclampsia, 205 of liamsters, 181 and heredity, 27, 238-241 of naturalists, 52 and nutrition, 18(5-193, 194-199. 202, 203, 204, 205, 20(j psychological fa cto rs i n H ue nc i n <> 52 of rats, 179 retention of teeth and, 13, 242 of sea anemones, 244 oi Tokophn/fi, 187, 191 Life tables, 7, 8, 9, 10 in Drosophila, 239, 240 in Lebistes, 29 Lipid content of aortic membrane, 74,78 Literary achievement, 39, 40. 49, 50 Longevity, see Life-span Longitudinal studies 38, 48, 49 204 Lung function, changes with age, 58-65 sex difference, (J8 volume, age changes, 58, 59, 00, 05. GO, 08 Lungs, ageing effects of lowered O2 partial pressure, 08 blood flow, 07 efficiency in man and animals, 08 vital capacity, 58, 59 (50 Maternal age, importance, 147, 238- 241 Maturation, definition and criteria, 201, 202 delay and life-si)an, 27, 29, 30, 201 early, as cause of ageing, 194-199 nutrition on, 30, 189-190, 194-199, 203, 206 Maturity, sexual, and life-span, 29, 30, 203 Maze experiments in rats, 215 Measurement of cerebral function, 51 of senescence, 4-14, 24 Melancholia, involutional, 44 Membrane permeability, age changes, 09-75, 70, 77, 78, 79 Memory in age, 40, 41 ex})eriments in rats, 215 tests after steroid administration, 137, 138 Menarche, onset as criterion of maturity, 202, 203 Menstruation, restoration after cessation, 144-145 Mental aspects of ageing, 32-48, 50, 51, 52, 215 disorders in aged, 44-40 patients, adrenocortical activity, 219-230, 237 use in ageing research, 183, 184 Metabolic balance studies after ACTH administration, 139 Mice, starved, analysis of tissues, 208 usefulness in ageing research, 177- 178 Mitral stenosis, CO uptake in, 00 Monkeys, duration of reproductive life, 145-140 usefulness in ageing research, 182 Morbidity in elderly, 10-23 Mortality, l)irds, 4 force of, 8, 9, 10, 12, 14 Muscle function after steroid administration in aged, 137, 138 power as measure of senescence, 0 Musical achievement of aged-, 40 Natural death, 10, 20 Naturalists, life-span, 52 Nervous disease, incidence in aged, 20 Nutrition and delay of sexual maturity, 203 on life-span, 180-193, 194-199, 202, 203, 204, 205, 200 on life of cell, 191, 192 neglected areas of research, 173-185 Obesity on life-span, 194, 195, 203 Oocytes, life-span, 142 neogenesis 172 Oogenesis, 147-149, 172 Subject Index 253 Osier's disease, S"), 105 Osteoporosis, calciimi nietal)oIisin in, 110, IKi-ns, 121, 122 senile, incidence, 125 Ovary, aiieing of, 141-159 age changes in sensitivity to hormones, 142-14() isolated, long storage of, 1 05-100 life-span of cells, 142 ])otential immortality, 157-159 transplantation, 150-152, 154-157, 158-159 tumours in grafted, 153, 154 Overnutrition (see also Nutrition) of cliildren as cause of ageing, 194- 199 Ovum, age changes, 149-150 life-span, 147-149 maternal age and, 172 transplantation, 150 Oxygen partial pressure, loAvered, effects, 08 quotient of aortic tissue. 70, 78 uptake in age, 200, 207 Palmar erythema, 82 Pancreas, elastase in, 99, 108 Parabiosis in ageing research, 170, 180 in retarded rats, 30 Pdramecium, 55 Parental age, importance, 147, 238-241 Pathogenesis of arteriosclerosis, 28 Pathological basis of ageing, 10-23 changes in brain, 34, 35, 30, 49 in cells, 54 Pathology at autopsy, lack of findings, 20 specific, of age, 54 in Swiss centenarians, 10, 17 Peptic ulcer, incidence in aged, 18, 19, 20 Permeability of membranes, 09-75, 70, 77, 78 Personality changes, 42-44 Phosphate absorption, 124 Pigmentation of cell, 53, 54 Pigs, life-span changed bv nutrition, 188-190, 192 Pneumonia, incidence in aged, 18, 19, 20 Populations, homogeneity of, 204 Poultry, usefulness in ageing research, 177 Pre-senile dementia, 44, 45, Ki Preparation of elastin, 90, 91 Preservation ol" skin by free/ing, 170 of tissues hy freezing, 1(>2-1()9, 171 of whole body by freezing, 107- 1 09 Pressure, atmospheric, adaptation to, 24 Prisoners, usefulness for ageing research, 184 Proteins of bodv, age on structure, 103, 104 Protozoa, r^o, 180, 187, 188, 191 Psychological aspects of ageing, 30-39, 209-213, 214, 215-218 changes, sex difference, 214 tests after steroid administration in aged, 137, 138 Psychoses, adrenocortical function in, 219-230, 237 senile, 44 Puberty, onset as criterion of maturity, 202-203 Pulmonary artery, elastin content, 92, 93, 94 function, changes with age, 58-05, 00 Rabbits, usefulness in ageing research, 181 Rats, calcium metabolism, 111-114, 124, 125 emphysematous broncliitis in, 07 ieai'ning experiments, 215 life-span, 179, 180 usefulness in ageing research, 178 vitamin D requirement, 124, 125 Red cell regeneration, 24 Regression, mental, in senescence 33, 34 Reproduction, effect of transplanta- tion of ovary, 158, 159 Reproductive capacity, 11 as criterion of maturity, 202, 203 cycle in retarded rats, 30 life, duration in monkeys, 145, 14() Reserve, fvmctional, 55 tissue, 55 Resistance to infection as measure of senescence, 0 Respiratory disease, incidence in aged, 18, 19, 20 function, changes with age, 58-()5 Retardation of m.aturation, 27, 28, 29, 30, 201 254 Subject Index Retirement, effect on lire span. 41) 50 Rotifers, life-span, 241. 24:5 Schizophrenics, aged, adrenocorti- cal activity, 219-236, 287 Scientific achievement of aged, 40 Sea anemones, life-span, 244 Second childhood, 34 Senescence, definition, 31, 242, 243, 244, 245 and measurement, 4-14, 24 Senile dementia, 44, 45, 40 age changes in skill, 214 elastosis, 101-103, 106, 108 keratosis, 108 osteoporosis, 110, 116-118, 121-122, 125 calcium balance, 123 Senility, mental, in centenarians, 25 mental changes, 34 Sex difference in animal life-spans, 29, 180, 181 in 17-ketosteroid excretion, 139 in psychological age changes, 214 distribution of lung changes with age, 68 hormones on calcium balance, 123 Sexual maturity and life-span, 29, 30, 203 Sheep, usefulness in ageing research, 182 Skilled action, age changes, 210-212, 214 Skin, ageing of, 53, 54 cancer, 53 collagen in, 106, 107, 108 creases, 12, 13, 23, 27 elastic tissue in, 105, 106, 107 elastoid degeneration, 53 long-stored, autografts of, 170 vascular lesions, 80-87 Social influences, 47, 49, 50 Specific growth rate, 6 pathology of age, 54 Spermatozoa, long-storage of, 164- 171 Spider naevi, 81-82 Starvation, late effects, 205 Steroid administration, effect on aged, 134-137, 138 Steroids, adrenocortical, bio- synthetic pathways, 126-128, 133, 134, 139 Stress, recurrent, 12, 13, 27, 65, 242 Survival curves, 239, 240, 243 Teeth, retention of, on life-span, 13, 242 Tentorium cerebelli, permeability, ()9, 74, 75 Ti'trahijnicNd in nutrition researcli, 186' Thalamus, pathological studies in aged, 35, 36, 49 Time, conception of, age changes, 217, 218 Tissue reserve, 55 transplantation problems, 150, 154- 157 Tokophrya in ageing research, 187, 191 Tortoise, endogenous ageing in, 28 Transplantation of long-stored skin, 170 of long-stored ovaries, 165-166 of ova, 150 of ovaries, 150-154 of tissues, prol)lems, 150, 154-157 Trauma, recurrent, 12, 13, 27, 65, 242 Trout, usefulness in ageing research, 29, 176 Trypsin in preparation of elastase, 107, 108 Tumours in grafted ovaries, 153, 154 Turtles, usefulness in ageing research, 176 Underfeeding, chronic, late effects, 205 Undernutrition {see also Nutrition) of children, 194-199 Unicellular organisms, 55, 186 187, 188, 191 Urinary disease, incidence in aged 18, 19, 20 metabolites in aged sehizof>hrenics, 223-236 17-ketosteroids, changes with age, 126-127, 128, 140 sex difference, 139 Uterus, isolated, long-storage of, 166-167 Vascular disease changes, and ageing changes, 54, 55 lesions of skin, 80-87 Venous lakes, 85 stars, 82, 83, 86 Ventilation of ]>lood, 62, 63, 64, 66, 68 Subject Index 255 Vital capacity of lungs, 58, 59, 0(5 Vitality factor in aj[»eing, 5, (5, 50, 57 nieasiued by adaptability, 24 Vitamin B^.^, effect on growtii, 11)7-199 Vitamin D and calcium absorption, 110-122 requirement in man, 128, 124 in rats, 124, 125 Volume of lirngs, 5S, 59, 00. 05, 00, ()8 Vulnerability as measure of senescence, 7-14, 28 Water metabolism in age, 200, 207 Work, loss of, 49, 50 Wound healing as measure of senescence, 0 Wrinkles, 12, 18, 28, 27 PiUNTED BY Richmond Hill Printing Works Ltd.. 23-25, Abbott Road, Bournemouth ''HIT' i ^>% .^'1,' '1'^ , . ,' nisi' r'"^'''.