This article was downloaded by: [Monash University Library]On: 06 January 2015, At: 20:33Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Biodemography and Social BiologyPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/hsbi20

The potential contribution oflongitudinal twin studies: AnappraisalFrank Falkner M.D. aa Assistant Professor of Child Health School of Medicine ,University of LouisvillePublished online: 23 Aug 2010.

To cite this article: Frank Falkner M.D. (1957) The potential contribution of longitudinaltwin studies: An appraisal, Biodemography and Social Biology, 4:2, 81-91, DOI:10.1080/19485565.1957.9987312

To link to this article: http://dx.doi.org/10.1080/19485565.1957.9987312

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information(the “Content”) contained in the publications on our platform. However, Taylor& Francis, our agents, and our licensors make no representations or warrantieswhatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions andviews of the authors, and are not the views of or endorsed by Taylor & Francis. Theaccuracy of the Content should not be relied upon and should be independentlyverified with primary sources of information. Taylor and Francis shall not be liablefor any losses, actions, claims, proceedings, demands, costs, expenses, damages,and other liabilities whatsoever or howsoever caused arising directly or indirectly inconnection with, in relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden.Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

THE POTENTIAL CONTRIBUTION OFLONGITUDINAL TWIN STUDIES

An Appraisal*

FRANK FALKNER, M.D.Assistant Professor of Child Health

School of MedicineUniversity of Louisville

Introduction

IN GENERAL DISCUSSIONS INVOLVING repre-sentatives of the varying disciplines of hu-man biology, it is apparent that most au-thorities agree in considerable measure su-perficially, but there are certain basicfundamentals they do not agree upon. Thisleads to the proposition that more infor-mation is needed in the basic sciences inthe disciplines connected with problems ofgrowth and development.

In studying some of the many problemsthere is a need for cooperation of manydifferent investigators and a strong plea forthe interdisciplinary approach must belodged at the outset. To give an example,in studying the genetic effect on growthand development, there is no sharp divid-ing line between, for instance, psychologi-cal and physical investigation. It is vitalto record this view.

Clearly in the study of this complex sub-ject we must accept the fundamental hy-pothesis that the overall gene "stream" ofan individual runs through and undermany influences which may be summarizedas environmental, and, of course, heredi-tary. These two basic influences on growth

*From the Child Development Unit, Departmentof Pediatrics, University of Louisville. Based on apaper Tead at the 33rd annual conference of TheMilbank Memorial Fund on "The Nature andTransmission of the Genetic and Cultural Charac-teristics of Human Populations."

The author is Markle Scholar in medical science.He is indebted to Mr. Frederick Osborn and Pro-fessor Alex Steigman for encouragement and to Dr.Wilton Krogman and Dr. J. M. Tanner for con-stant stimulation.

and development with ultimate formationof a "type" of individual are irrevocablyinseparable. What is wanted in the manybranches of the fields of investigation un-der discussion is to know the weighting ofthe influences, and then how the resultsmay become set or altered; and how thegenetic stream may be affected by cultural,educational, physiological, and physicalinfluences, which are quoted as clear-cutexamples.

The human biologist eagerly seeksknowledge on the structural and behavioralpatterns and their variations. It is a con-stant stimulation that in spite of the greatamount of work carried out in the field inthe past we still know surprisingly littleabout patterns of development, and espe-cially correlations between them.

There are a great many facets of an in-dividual, and his changing curve of devel-opment is indeed complex. D'Arcy Went-worth Thompson (1917) wrote: "A curveof growth to run all life long is only sim-ple in the simplest of organisms, and is usu-ally a very complex affair." What is con-fusing is that the differing characteristicsare inter-related and influence each other.Some of these facets appear to change littlein life—and it may be upon these that werely to determine an individual's character,or "make-up," in the widest sense. Knowl-edge on such problems is fundamental forinvestigation in the basic sciences.

It is apparent that the final characteris-tics just discussed may be laid down inchildhood and influenced sharply by envi-ronmental influences in this period. Mind-

Dow

nloa

ded

by [

Mon

ash

Uni

vers

ity L

ibra

ry]

at 2

0:33

06

Janu

ary

2015

82 Eugenics Quarterly

ful constantly of genetic influence, it istherefore important to study the childfrom these aspects on an inter-disciplinaryand longitudinal basis.

It is perhaps important to say what ismeant by longitudinal basis. Clearly it iseasy, or easier, to obtain cross-sectional fig-ures upon growth and development in anypopulation. This consists of measuring forsome factor a large number of individualsat the same time, at a certain age in child-hood, or in adulthood. According to theworthiness or representation of the samplewe gain information on the state of thatmeasurement at certain points on a devel-opmental curve.

With a changing organism, and certain-ly from the research point of view, it ismore profitable and often necessary tostudy the organism as it changes to findout, for example, its speed of growth. Ifsomething is moving, you need to know atwhat speed it is moving. This can only bedone by the longitudinal approach.

A child investigated more than onceprovides two measurements; the first meas-urement is subtracted from the second andthe result is data vital to the researcher inthe field—the incremental figure. It is notpermissible, as one often sees in the litera-ture, to deduce incremental figures fromcross-sectional data. It is fully realized thatthe longitudinal approach to developmen-tal problems is expensive and fraught withevery kind of difficulty, but in the investi-gation of a changing organism it is difficultto see an alternative approach.

"Is maximum growth optimum growth,"and such other stimulating questions can-not be answered unless you follow thegrowing organisms over a long period oftime, and, indeed, there are some problemswhich cannot be investigated unless theorganism is followed from birth to ma-turity. In fact, it is pertinent to note thatenvironmental influences do not start tooperate at birth but theoretically at themoment of conception. The fact that onecan come from a good home or a bad home

in utero is important and all too oftenneglected. Means of studying this pre-natal environment are difficult but by nomeans impossible, as will be suggested.

A possible suitable method of study,based upon investigations of twins ideallyfrom before birth, will be proposed.

It may be asked why it is necessary tocenter such a study on childhood. Thecrux is that as differentiations appear theypresumably become fixed at some time inchildhood. The varied influences can beseen acting, and their effect assessed. Clear-ly many problems, as has been mentioned,can only be solved by following childrenuntil they actually become adults, for insome children differentiations in childhoodmay be temporary and may not always per-sist to adulthood. The main aim, then,would be to study the inter-relation andidentification of the many factors whichlead to a differentiation of the characterand "type" of the child or adult. "Type" isused here in the very widest sense. Thediverse influences acting on the develop-ment of the child may be summarizedfrom the investigation point of view underthe headings psychological, physical, andphysiological.

Mindful of these introductory remarks,it is obvious that investigators in thegrowth field and geneticists might turn tostudies of identical twins. Such studies havebeen done before, and this is an idea, ofcourse, which is by no means new.

HistoricalFrancis Galton founded the science that

might be termed "Gemellology" in 1875and 1883. First, his paper on "The His-tory of Twins as a Criterion of the Rela-tive Powers of Nature and Nurture," andthen his book "Inquiries into Human Fac-ulty and Its Development," showed that theproposed investigations were to throw lighton the weight on "just scales of the effectsof Nature and Nurture, and to ascertaintheir respective shares in framing the dis-position and intellectual ability of men."

Dow

nloa

ded

by [

Mon

ash

Uni

vers

ity L

ibra

ry]

at 2

0:33

06

Janu

ary

2015

83

Thorndike in 1905, and Poll in 1914, pub-lished results of the measuration of twins,and in 1923 Apert and Siemens separatelyprovided more comprehensive publicationsdealing with biological and pathologicalconsiderations. These scientists may besaid to have been the pioneers. Dahlberg(1926) gave the first detailed work on theinheritance of physical characteristicswhich is a milestone in this history. Similarwork was carried out by Stocks in 1930. Itdoes not seem to be widely known thatimportant publications were being pro-duced from the Maxim Gorky Medical-Genetical Research Institute in Moscow upto 1936.

Professor S. G. Levit, its director, disap-peared some short time after this, andlittle more has been heard from this im-portant center. A large number of care-fully documented twins were followed andinvestigations covered a wide range. Sobo-lera and Ignatier (1936) on the role ofheredity and environment in variability ofheight and weight; Lichtenstein and Bas-sina (1936) on the mouth cavity in twins;and Lunia and Mirennova (1936) on thestability of the effect of training, serve asexamples of the important work usingtwins as subjects which was carried out bymembers of this Institute. The productionof appreciable differences in monozygoustwins after the action of "exogenous" post-natal influences was shown by Komai andFukuoka (1934). The classical work ofNewman, Freeman and Holzinger waspublished in 1931 and described the inves-tigations into identical twins who had beenseparated at birth or early in life and hadbeen brought up in different environments.A general account of twins under study hasbeen described by Gedda (1951). Penrosein 1937 reported a pair of monovular twinsin which only one had congenital syphilis.This stimulated the very necessary interestin pre-natal influences upon identicaltwins, and these recognized pre-natal fac-tors have been given in review form byBass (1952). It was Price (1950) who

pointed out our need to consider thesebiases of pre-natal onset which he main-tains influence the development of twins.He divides them into natal factors, lateralinversion, and the effects of mutual circu-lation. These biases may affect both typesof twins, but to a varying degree. Neel(1956) has suggested that "evidence on themagnitude of the biases must be obtainedif we are to determine whether the twinmethod can be properly implemented as atool for the appraisal of a nature-nurtureinteraction."

Despite the past excellent work doneupon twins and briefly reviewed here, thereis a need for a longtitudinal interdis-ciplinary study, in the widest sense, carriedout over a number of years.

In summary, unusual opportunities areafforded in the study of monozygous anddizygous twins and their siblings on a lon-gitudinal basis to elucidate the weightingsof genetic and environmental influences.For a long time the dramatic concordanceof the monozygous twin in psychological,physical, physiological, and pathologicalconsiderations has been an attractive andnoticeable feature, but it should be broughtto mind that it will be discordant featuresthat will lead to greater knowledge con-cerning our problems.

International CooperationAt this point it is necessary to digress

and to describe some international growthstudies which are at present in progress inEurope and Africa. Very briefly, the his-torical beginning of these growth studiesstarted with an interdisciplinary longi-tudinal growth study in the widest sense inLondon approximately six years ago. Thiswas not concerned with twins in any way.It was a study on so-called "healthy," ran-domly-selected children from an urbanpopulation, and the principle was to studythem on a birth-to-maturity basis.

It was the desire of French colleaguesthat they start in Paris a similar growthstudy, the plan being to run on exactlyparallel lines to the London study. It then

Dow

nloa

ded

by [

Mon

ash

Uni

vers

ity L

ibra

ry]

at 2

0:33

06

Janu

ary

2015

84 Eugenics Quarterly

became apparent that other centers wereinterested in cooperating in this way andthe writer foresaw the possibility of spend-ing a life going around the capitals of Eu-rope starting growth studies. This wasalarming. The bringing of order, into whatpromised to be a possibly chaotic house,was effected by sponsorship of the Interna-tional Children's Center in Paris. In short,there are now running five European longi-tudinal growth studies. They are all inter-disciplinary. The principle behind themis that all have agreed to follow a verysimple—and simple is emphasized—"baseline" of investigations in all the disciplines(Falkner, 1957i). It is essential to note

that no central "authority" is directingthese growth studies. They all have veryspecial interests and in some branches aredoing very detailed work. That, in a way,is not the concern of the internationalgroup. What all have undertaken to do isto follow this agreed international baseline.

Clearly, the base line must be improved,criticized and altered continually, especiallyon the psychological side, as the childrenare growing older. It is, therefore, neces-sary to meet regularly to do battle, criticizeeach other, compare data, plan work forthe following year, and to standardizemethods and techniques.

This international cooperative work ispresented here, for its potential advantagesare obvious—the production of compara-tive international data. There are twostudies in Africa which are clearly ongroups of very differently constituted sam-ples of children, malnutrition being one ofthe big factors of difference. Their par-ticipation will lead to comparisons of greatinterest. These centers are in West Africaand East Africa, and the base line is usedas far as possible. Already, although someof the growth studies are comparativelynew, it is apparent that the African child,certainly in the first year of life, has a dif-ferent pattern of mental and physical devel-opment compared to the European child.

Allbrook (1956) has recently shown thatthere is a reduction in the adrenal corticalmass in the East African compared to theEuropean. Such work stimulates one tospeculate on the basis for such differingpatterns of development and enhances theplea for more of such basic comparativedata. The importance of studying the Ne-gro comparatively in the United States isalso clear.

Quite apart from the advantages of us-ing truly comparative international data,the solution of such problems as the influ-ence of racial factors upon human varia-bility may be further investigated follow-ing such original observations as those ofMuller (1954).

The Longitudinal ApproachTo gain approval for such a proposed

type of project is not easy. The reason isclear. Experience has shown that this typeof study has not always been satisfactoryfrom every point of view.

It is fully recognized that some problemscan only be studied by following childrenfrom birth to maturity. The objections tothis method are that it has been thoughtnecessary to wait a great time before ana-lyzing the data. Such growth studies havebeen attempted in the past and, with oth-ers, have collected large quantities of datawhich, excellent as they are, have been in-sufficiently analyzed because of difficultiesto which all are fully sympathetic. Thereis a further danger in interdisciplinarylong-term studies of this kind in that thevarious disciplines become bogged downwith data which is sequestered and not co-ordinated.

Mindful of these disastrous pitfalls, andwith these uppermost in thought, a solu-tion is to follow children routinely on a"mixed longitudinal basis." The basicprinciple behind this is to recruit childrenin groups at all ages. This method enablesinvestigation to be carried out on a wholeage range of development by means of sev-eral overlapping longitudinal runs. Con-

Dow

nloa

ded

by [

Mon

ash

Uni

vers

ity L

ibra

ry]

at 2

0:33

06

Janu

ary

2015

85

veniently, these would be in five-year pe-riods, and the important adolescent groupwould be included in the last two. Theadult will be mentioned later. This methodproduces results much more quickly, andthe investigators will clearly at the sametime gain a much larger overall experiencewith the growing child in his entirety andin all the disciplines.

Although the work would be plannedbasically in runs of five years, it will be toeveryone's advantage to follow some of thechildren for as long as possible, and tohope to follow the majority to maturity. Itwould be very desirable to recruit as manyidentical twins as possible before birth,and the expressed interest, for example, ofobstetrical and radiological colleagueswould ensure this. After recruitment ofgroups of monozygous and dizygous twins,the siblings ideally should be studied. Anenergetic attempt should also be made tofind the greatest possible number of mono-zygous twins who had been separated inchildhood. It is recognized that these arevery rare.

In view of past experiences with growthstudies in general the following is statedwith considerable trepidation, but honestyrequires that it be stated: It must be recog-nized that many basic problems of growthand behavior are unsolved. In any longi-tudinal multi-disciplinary project data iscollected in quantity. In addition to clear-cut hypotheses to be tested—and these areessential—it is always probable that profit-able lines of research and possible correla-tions, and so forth, will develop as theresearch does. It is best to clearly state this,and this concept is strongly claimed tojustify some part of such a study's useful-ness. The great pitfalls apparent whenusing this perhaps comforting hypothesismust be, and are, recognized.

A further aim is to provide informationfor other workers in the field. It is obviousthat possible correlations would be verylarge. If the data is in a workable form, itis invaluable from time to time for special-

ized researchers who may require informa-tion from one part of the work. This hasoccurred successfully five times within thelast three years in the cooperative Europeanstudies.

There is always the problem of keepingcontact with a "longitudinal sample" andin large countries the geographical consid-erations of the moving members of thesamples can be a source of worrying loss.It is, therefore, important to prophesy sucha loss in order to minimize a final resultof finishing with a sample too depleted tosustain reliable results.

One aspect of such growth studies ingeneral is neglected: Their value as ateaching tool is not always appreciated.They can contribute much to post-graduateand graduate education during their actualroutine and progressive existence while theeducational value to the investigators them-selves is large. In making a plea for thiskind of work it is necessary to produceevery possible advantage.

The great pitfalls, then, apparent whenfollowing the methodology of this type ofresearch, are: (1) Over-collection and stor-age of large quantities of data. This maybe largely overcome by the use of mechan-ically recorded data, and using modern sta-tistical techniques and computation (Heu-mann, 1956). It is fully realized that byplacing data on IBM cards, for instance,the problems of actual analysis of data areby no means solved. It does, however, placethe data in a workable condition. (2) Lackof constant review of data. Such reviewshould be stimulated by: (a) An AdvisoryBoard, and (b) the availability of data foruse of other research workers.

A word on "an Advisory Board": Enter-ing the field of investigation into problemsof growth and development is not unliketaking off across the high wire. Falling offis easy but staying on is not, and to under-take investigation into this sphere withoutthe constant supervision and skilled helpof a geneticist and biometrist is asking tofall very early in the precarious journey.

Dow

nloa

ded

by [

Mon

ash

Uni

vers

ity L

ibra

ry]

at 2

0:33

06

Janu

ary

2015

86 Eugenics Quarterly

The advice and planning of a social scienceexpert is also essential in order to be sureto investigate skillfully the all-important"socio-environmental" background of thesubjects. Thus, the interdisciplinary ap-proach must be carefully planned and theinvestigations themselves supervised bysuch a regularly meeting advisory body.

The Twin StudyBefore carrying out any investigations, it

is clearly essential to classify the twinsstudied and designate accurately themonozygous from the dizygous. Study ofthe placenta is important, and micro-scopic examination may be necessary buthot conclusive. It is common that mono-zygous twins may exist in utero in separatefetal membrances for example (Von Versh-ner, 1939), and indeed may have separateplacentas (Lassen, 1931). Other such diag-nostic factors as concordance of sex, physi-cal features, skin, hair and eye color, tasteperception for phenylthiocarbamide, tonguecurling, and fingerprints and palm printsare all of help in determining mono-zygosity. But at best their conclusiveness isnever of complete certainty. It is of inter-est that authorities now make clear thatfingerprints of identical twins are notnecessarily identical (Federal Bureau ofInvestigation, 1953).

Happily the basic work of Race andSanger (1950); Fisher (1951); Weiner(1935, 1940, 1952); Bryce, et al. (1951);and Walsh (1955) on the concordance ofblood factors has made it possible to deter-mine the truly monozygous pair with greataccuracy. For theoretical considerations,use of the previously mentioned concord-ance factors together with determination ofthe ABO, MNS, and RH blood groups willensure correct diagnosis.

Basic HypothesesThere are three hypotheses to be tested

which are vital and basic to interpretationof twin study data. In the past these in-fluences may have been recognized as pos-

sibilities, but their effect has not been as-sessed. Hence, all work in this field isbased upon assumptions which have notbeen and must be tested. These are: (1)that differing pre-natal and natal influencesupon each twin of a monozygous pair causediscordant patterns of development. (Forexample, abnormal positioning in utero ofone twin.) (2) That concordance in pat-terns of development of dizygous twins maybe due to, or influenced by, a common cir-culation while in utero. Concordance heldto be due to similar environment may be,in effect, due to this factor. (3) That cer-tain definite maternal influences pre-natallyproduce discordance in monozygous twins.For example, one pair of monozygous twinshad congenital syphilis and not the other.(Penrose, 1937)

To test these hypotheses, a group of pre-natally recruited twins are needed and thecooperation of skilled obstetric colleaguesrequired. After the thirty-second week ofpregnancy positioning of each twin is rarelyaltered and careful observation at the birthshould label each twin. Routine radio-graphic examinations are normally carriedout after the thirty-second week and thisfurther gives information as to the posi-tioning. (Extra radiography for this studyis not justified with the current conscious-ness of the effect of radiation on the fetus.)Regular plotting of the position of eachfetal heart will be an added indicator ofposition, since the differing fetal heart ratescommonly remain so. Labeling and accu-rate microscopic and macroscopic examina-tions of the placentas are necessary andpractical, while an accurate maternal his-tory during pregnancy is mandatory.

The subsequent groups of monozygousand dizygous twins form control groups.

Research on human twins has often hadto be retrospective and obviously has in-volved twins who have both lived a certaintime and usually a comparatively longtime. The method of ante-natal study al-lows of the possibility of studying twinswith the prospects of a long-term follow-

Dow

nloa

ded

by [

Mon

ash

Uni

vers

ity L

ibra

ry]

at 2

0:33

06

Janu

ary

2015

87

up; and perhaps more important of study-ing twins where one or both die beforebirth; and aborted twin pregnancies. In-vestigation of such aborted or still-borntwins, coupled with histo-chemical andbiochemical studies on healthy and dis-eased placentas in monozygous and di-zygous pregnancies should reveal valuableinformation.

Further InvestigationsIn general, the lines of investigation of

the proposed interdisciplinary study wouldproceed along the main lines already dis-cussed and hypotheses other than the abovebasic three would be tested in the differentdisciplines of the research. It would be un-desirable to carry out a great mass ofhighly detailed investigations on all thetwins, for example. Sub-groups would beformed where suitable.

It may be profitable to discuss at thispoint some examples of problems from thevarious disciplines which should be inves-tigated in such a study. Along the physi-cal line of approach, Palmer (1934) has sug-gested that genetic determination of therate of growth is likely to be independentof genes influencing size achieved. Furthergeneral work on this subject would clearlyemerge. A most profitable feature wouldbe to break down the growth in the twinsto the unit involved: as an example, the"whole concept of skeletal maturity is underdiscussion, for its accurate measurement isadvantageous in many spheres. (Falkner,1957ii). More information is needed on thegenetic influence on the pattern of the indi-vidual variations in skeletal maturation be-fore such improved assessment is possible.(Reynolds, 1943; Garn, 1956; Acheson,1956). Patterns of development are alwaysof interest to the growth investigator andthese would, clearly, merge. A particularcurrent interest is centered on the pattern-ing of fat growth in the human, and the in-vestigation of true genetic influence onobesity and the "normal" large variation

of fat distribution in the child and adultwould be profitable.

Physiologically certain problems presentthemselves for disentanglement. Two im-portant examples are given: Are enzymelevels in the child genetically determined?And possibly more important, is the patternof antibody formation in the human ge-netically determined? It is more correct tosay in place of "genetically determined," isthe weighting very heavy on the side of thegenetic influence? Basic information onantibody formation is vital to us in the fieldof preventive medicine in the future—oursynthesization of antibiotics may not forlong outstrip the rate of sensitization oforganisms.

Dobzhansky (1957) has pointed out tous that growth does not stop at the age of20 years in the human but is representedby a trajectory which falls off at senescence.This is true, but it is continually fascinat-ing in examining the incremental curvesof size of many parts of the body to see howdramatically these incremental figures dropto zero at approximately the age of 20years. Why does growth of size stop at thispoint and stay at the same point for often40 years? Growth, to make a generaliza-tion, occurs by cell division rather thanincrease in cell size. (Neural tissue is aclear exception.) In some organs or tis-sues there are, then, two processes occur-ring during the active growing phase—celldivision, and repair or replacement for re-generative processes. Clearly calories areneeded for both; but a highly speculativesuggestion is made that a constant supplyand amount of a substance (say M, signi-fying mythical ) is available which isquantitatively genetically determined, andwhich activates the division and regenera-tion process. When a certain size of thattissue mass is obtained, the regenerationnow necessary for its constancy will use allthe substance Af available, and hence fur-ther calories above a basic amount will notresult in any cell division and thus no in-

Dow

nloa

ded

by [

Mon

ash

Uni

vers

ity L

ibra

ry]

at 2

0:33

06

Janu

ary

2015

88 Eugenics Quarterly

crease in size. Investigation at a basic levelalong this avenue might have a wide appli-cation.

In the area of pathological processes, itwill be particularly helpful to establish theweighting of conditioning of disease by in-heritance. Twin studies are a profitablefield for such investigation. Warkany, et al.(1955) have described discordance inmonozygous twins resulting in diabetesmellitus in one twin and obesity in theother. They point out the importance ofthis and similar observations for the studyof the etiology of diabetes mellitus and ofobesity.

Work on the inheritance of susceptibilityto infection by tuberculosis was stimulatedby Diehl (1933, 1936) and recently byworkers in France. These are importantindications of the part twin investigationscan play in elucidating such important in-formation of this type.

It must have been evident by now thatthis discussion has been centered verylargely on fields of investigation other thanthe psychological one. It must be empha-sized very strongly that this is not be-cause of a reluctance to admit of its greatimportance, but it is simply due to an un-willingness to discuss something in whichthe writer has no special knowledge. Withthis reservation, however, some of his col-leagues' observations and general ideas areput forward. Thompson (1957 has shownus that with modern mathematical, statis-tical and experimental psychological meth-ods we are able to be far more objectivetoday than we ever were before. On a verysimple level, the psychological investiga-tion methods as used in the internationalbase line are summarized. This will showthe basis of proposed work in this field foruse in a twin study.

1. The interrelation of psychologicalinfluences with others affecting the be-havior and development of the personalityof the child is studied. In avoiding detaileddescription, the principles of investigationmay be divided into (a) interview tech-

niques. Here data obtained from inter-viewing the mother and close relatives ofthe child, and older child himself, is col-lected, (b) Standardized testing techniquesfor the assessment of the child's mentalability. These vary naturally according tothe age of the child. It is pertinent to notethat work has already been carried out bythe international teams on attempting toassess present tests and devising new ones,particularly for the very young child andthe infant. In respect to both of the abovebasic methods it should be pointed outthat both are constantly subject to reviewand attempt at improvement, (c) Assess-ment of the temperament of the child bystandard and experimental methods.

2. Assessment of the changing socio-en-vironmental background of the child. Thispart of the investigation can be said to bevitally important to all sides of the re-search, but for the sake of convenience it isincluded under the psychological heading.

This sketchy account is merely the pres-entation of a non-expert in the field whorealizes its paramount importance.

ConstitutionFinally it is necessary to traverse the

most slippery part of the high wire. In dis-cussing the child and adults in growth anddevelopment studies, we are constantlycoming across the concept of "types,""somatotypes," "body-types," "personality-types," and so on. It is clear that each ofus has an individual subjective method oftyping our fellow humans. This is not anew idea.

"Let me have men about me that arefat, sleek-headed men and such as sleepof nights! Yon Cassius has a lean andhungry look, he thinks too much; suchmen are dangerous."

. . . Julius CaesarAct 1, Scene 2.

Hippocrates (450 B.C.) described twotypes of people, apoplectic and phthisic,and the great clinicians in the past com-monly gave us excellent constitutional de-scriptions of individuals with certain dis-

Dow

nloa

ded

by [

Mon

ash

Uni

vers

ity L

ibra

ry]

at 2

0:33

06

Janu

ary

2015

89

ease entities. Beneke (1878), the Germanpathologist, perhaps may be said to havefounded the interest in constitutional dis-ease. Two Frenchmen, Rostan in the mid-dle of the nineteenth century and Sigaudat the beginning of the twentieth, startedand continued the French "school" of con-stitutional classification of the human, andthey divided man into "digestive," "mus-cular," "respiratory," and "cerebral" types.(Some individuals were not able to befitted into these divisions and presumablywere just unfortunate misfits.) There arestill some French workers who adhere tothis classification, and the school certainlystimulated work in the field. It was Kretsch-mer in 1921 who achieved considerable no-tice through his furtherance of this princi-ple and he divided individuals into threegroups: pyknic, athletic, and asthenic. Al-though Kretschmer was in fact workingbasically with mentally abnormal persons,and his notations have since been widelyand very incorrectly used, his was the firstreal attempt to correlate character withbody-type. His method was highly subjec-tive, but he did at one time try to be moreobjective by physical measurements.

Since 1919, Viola, an Italian physician,has worked with the problem of constitu-tion and its correlation with disease, andby physical measurements he devised an in-dex of "morphological type." Here was atruly objective attempt, but he, too, placedthe individual in three separate categories.It might repay us to study Viola's princi-ples anew especially in relation to disease.

Up to this point the "constitutionalists"divided us into clear-cut categories. Anewer development was that of factor anal-ysis (examples: Burt, 1938; McCloy, C. H.,1940) which, reducing description to avery basic level, is a method of describingany physique conveniently as factors afterstatistical analysis of a series of correlatedmeasurements. These measurements areanthropometric ones and what is actuallymeasured is clearly important. (As far as isknown no follower of this method takes

into consideration the distribution of fatin the body.) Certainly this index movedus a step forward and did not place us in apigeonhole.

Sheldon since 1940 (1940, 1942, 1949,1954), has steadily put forward his impor-tant concept of constitutional factors. He,too, used a number of divisions of physicaltypes, three—but with this vital difference—they became components and each indi-vidual type was assessed by means of thequantity present of each component. Shel-don's original work was highly objectiveand based on measurements carried out onstandardized photographs of over 4000young adult males. Since this time it hasbeen realized that this method was too"static," and the present method now is tolabel the individual by division of subjectsinto age groups, the use of a height-to-weight ratio, and anthroscopy (or inspec-tion of standardized photographs). Thismethod, perhaps surprisingly to some, issatisfactorily objective (Tanner, I954i.)I have carefully refrained from describingin words any of the physical types usedover the centuries. Sheldon's nomenclaturefor the three basic components has causedsome criticism, but I submit the actualnames have really little importance. Asa basis for research into the difficultproblems of biological constitution it isclear a great amount of work has to bedone, and it is timely to state that Sheldonhimself has never stopped pointing out thathis work and the work of others in the field(examples: Dupertuis and Michael, 1953;Tanner, 1954ii) is provocative of new re-search. The relationship of physique totemperament, that most fascinating ofproblems in this field, is the most contro-versial part of Sheldon's work, and it ishere lies the stimulation for the beginningof a great amount of needed research. Dowe not, all of us, have our individual meth-ods of "rating" temperament as well asphysique? Is there a true relationship be-tween the two? Can temperament be al-tered? Is physique largely a static phe-

Dow

nloa

ded

by [

Mon

ash

Uni

vers

ity L

ibra

ry]

at 2

0:33

06

Janu

ary

2015

90 Eugenics Quarterly

nomenon? Clearly not—but to what de-gree not? An ex-chief and friend, who wasvery much not of muscular athletic build(heavily endowed with "mesomorphy")and was very round, once asked me: "Weshould all like to be mesomorphs — howcan I change; could I have done so whilst ababy?"

It is clear to me—and I am impressed bythe number of colleagues who agree withme—that babies and children can be indi-vidually "typed" using the word verybroadly. With the babies this is highlysubjective. What is wanted is informationon the type, not necessarily a static typereached at "maturity," but its continualtrend or pattern of development. Thework of animal physiologists (Hammond,1954, 1955), stimulates us to investigatethe changes of body build in the humanbasically genetically determined, for theyhave shown that differing nutritional statusmay produce great changes in configura-tion and actual body composition.

In the realm of environment, we areback into the concepts of the founders ofconstitutional medicine. I quote an exam-ple: Colleagues (Dowling et al., 1956) areinvestigating the disease of infantile eczema(Prurigo Besnier) in a group of children.It is considered probable that this condi-tion is a good example of "psychosomaticdisease." It is becoming apparent to us thatthese children of all ages have a very largecomponent of muscularity in their build.If this is proved and accepted at a scien-tific level, it is only one small step forward.The next and vital question is, what was,and is, the weighting of genetic and envi-ronmental factors which produced or al-tered the behavior, physique, and physio-logical state of a child so that he was moresusceptible to this condition than if hewere the child next door with a differentmake-up?

Newer objective provocative work onthese fundamentals is being produced withthe adult, examples of which are Gertleret al. (1951) on coronary occlusion, and

Sutter and Tabah (1953) in France on therelation of consanguinity to the generalstructure of mortality.

In assessing the weighting of the geneticinfluence, it appears that a valuable contri-bution to this whole difficult field might bemade by a long-term study of the humantwin. It cannot be emphasized enoughthat great harm has been done in the pastby some unscientific approaches to theseproblems and any future work from what-ever angle must be made from an interdis-ciplinary standpoint and with rigid scien-tific standards. A collective approach tothese problems should yield valuable re-sults with simultaneous assessment ofshape and constitution using psychological,anthropometric, biochemical, photogram-metric, radiological measurement, andvolumetric estimations. But is an unravel-ing of the place of the genetic, physiologi-cal, and psychological biases as influenceson human constitution likely to contributeto biological knowledge of the human?Surely, for ". . . morbidity and mortalityconstitute the long range criteria of theimportance of differences in human phy-sique" (Brozek, 1956).

Some varying components of tempera-ment, behavior and body composition areinfluenced greatly by the human variableswhich we have discussed. All these variableaspects present a problem for further re-thinking and research. Valid informationon this subject gained by impeccable sci-entific attempt is needed and may be eval-uated in degrees of future human healthand happiness. It is a challenge of highendeavor to human biologists—a challengefull, I am sure, of heartbreaks but it mustbe taken up for the rewards will be verygreat.

REFERENCES

ACHESON, R. M., 1956. Personal Communicationsand unpublished data.

ALLBROOK, D., 1956. Lancet. 2 : 607.APERT, E., 1923. Bibliotheque des Conaissances

MMicales, Paris.BASS, M. H., 1952. Adv. Int. Med., 5 s 15.BENEKE, F., 1878. De Anatomischen Grundlagen

dor Constitutionsan O Malien, Marburg.

Dow

nloa

ded

by [

Mon

ash

Uni

vers

ity L

ibra

ry]

at 2

0:33

06

Janu

ary

2015

91

BROZEK, J., 1956. ed. Body Measurements and Hu-man Nutrition. Wayne University Press.

BRYCE, L. M., et al., 1951. Ann. Eugenics 15 : 271.BURT, C., 1938. Brit. J. Med. Psychol. M. 518.DAHLBERG, G., 1926. Twin Births and Twins from

a Hereditary Point of View. T. Trycken:Stockholm.

Diehl, K., 1933; 1936. Zwillingstuberkulose I & II.Fischer. Jena.

DOBZHANSKY, T., 1957. Proceedings of the 33rdAnnual Conference of the Milbank GeneralFund. The Nature and Transmission of theGenetic and Cultural Characteristics of HumanPopulations. In press.

DOWLING, G. B., CALNAN, C., FALKNER, F., 1956.Awaiting publication.

DUPERTUIS, C. W., & MICHAEL, W. B., 1953. ChildDevel. 24: 203.

FALKNER, F., 1955. Editor: Une base-commune derecherches pour les étudds longitudinales surla croissance de l'Enfant (A Base Line of In-vestigation for Longitudinal Growth Studiesof the Child). Published in French and Eng-lish, Centre International de l'Enfance, Paris.

—, 1957i. Editor and Contributor: Inter-nationally Cooperative Research into ChildDevelopment. (Published in French and Eng-lish) . International Children's Centre, Paris,1957. In Press.

, 1957ii. Skeletal Maturity. Proceedingsof the 26th Annual meeting. Am. Assoc. ofPhys. Anthropologists.

FEDERAL BUREAU OF INVESTIGATION. 1953. Law En-forcement Bulletin. February.

FISHER, R. A., 1951. Heredity. 5: 95.GALTON, F., 1875. Erasers Magazine. 12: 566.GARN, S. M., Personal Communications and un-

published data.GEDDA, L., 1951. Studio dei gemelli, Edizioni Oriz-

zonte Medico, Rome.GERTLER, M., GARN, S. M., & WHITE, P. D., 1951.

J. Amer. Med, Assn. 147: 621.HEUMANN, K. F., 1956. Science. 124: 3226. 773.HAMMOND, J., 1954-55. Progress in the Physiology

of Farm Animals. Vol. I & II, Butterworth,London.

KRETSCHMER, E., 1921. Korperbau Und Karakta.Springer, Berlin.

KOMAI, T., & FUKUOKA, G., 1934. J. Hered. 25 :423.

LASSEN, M. T., 1931. Arch. Gynak. 147: 48.LicHTENSTEiN, E. A., & BASSINA, I. L., 1936. Proc.

Maxim-Gorky Medico-Genetical Res. Inst. IV.LUNIA, A. R., & MIRENNOVA, A. N., 1936. Proc.

Maxim-Gorky Medico-Genetical Res. Inst. IV.MCCLOY, C. H., 1940. Child Devel. 11: 249.

MULLER, H. J., 1954. Proc. Maxim-Gorky Medico-Genetical Res. Inst., IV.

NEEL, J. V., and SCHULL, W. J., 1956. HumanHeredity. Univ. of Chicago Press, Chicago.

NEWMAN, H. H., FREEMAN, F. N., & HOLZINGER, K.,1937. Twins. A Study of Heredity and Envir-onment. Univ. of Chicago Press, Chicago.

PALMER, C. E., 1934. Child Devel. 5 : 351.PENROSE, L. S., 1937. Lancet 1: 322.POLL, H., 1914. Z. Etho. S. 87.PRICE, B., 1950. Am. J. Hum. Genet. 2 : 293.RACE, R. R., & S(&758, R., 1950. Blood Groups in

Man. Blackwell, Oxford.REYNOLDS, E. L., 1943. Amer. J. Phys. Anthropol.1: 405.ROSTAN, L., 1936. Coude de Midecine Clinique.

Dumont. Brussels.SIEMENS, H. W., 1924. Die Zwillungs Pathologue.

Springer, Berlin.SIGAUD, C., 1908. Forme Humaine. Flammarion,

Paris.SHELDON, W. H., 1940. Varieties of Human Phys-

ique. Harper, New York., 1942. Varieties of Temperament. Har-

per, New York., 1949. Varieties of Delinquent Youth.

Harper, New York., 1954. Atlas of Men. Harper, New

York.SOBOLERA, G. V., & IGNATIER, M. V., 1936. Proc.

Maxim-Gorky Medico-Genetical Res. Inst. IV.STOCKS, P., 1930. Ann. Eugenics, IV. 1 & 2.SUTTER, J., & TABAH, L., 1953. Population 8: 511.TANNER, J. M., 1954 i. Amer. J. Phys. Anthropol.12: 257.

, 1954 ii. Eugen Rev. 46: 149.THOMPSON, D. W., 1917. On Growth and Form.

Cambridge University Press, Cambridge.THOMPSON, W. R., 1957. Proceedings of the 33rd

Annual Conference of the Milbank MemorialFund. The Nature and Transmission of theGenetic and Cultural Characteristics of HumanPopulations. In Press.

THORNDIKE, E. L., 1905. Arch. Philos. Psychol.No. 1.

VIOLA, G., 1919. La Constitutione Individuals, Bo-logna.

VON VERSHNER, O., 1939. Proc. Roy. Soc. LondonB. 128: 62.

WALSH, R. J., & KOOPTZOFF, O., 1955. Austral J.Exp. Biol. 3 3 : 189.

WARKANY, J., GUEST, G. M., & COCHRANE, W. A.,1955. Am. J. Dis. Child. 89 : 689.

WEINER, A. S., 1935. Hum. Biol. 7 : 229.WEINER, A. S., 1952. Am. J. Hum. Genet. :4 37.WEINER, A. S., & LEFF, I. L., 1940. Genetics 2 5 :

187.

Dow

nloa

ded

by [

Mon

ash

Uni

vers

ity L

ibra

ry]

at 2

0:33

06

Janu

ary

2015