Human Evolutionary Biology

HUMAN EVOLUTIONARY BIOLOGY Wide-ranging and inclusive, this text provides an invaluable review of an expansive selection of topics in human evolution, variation, and adaptability for professionals and students in biological anthropology, evolutionary biology, medical sciences, and psychology. The chapters are organized around four broad themes, with sections devoted to phenotypic and genetic variation within and between human populations, reproductive physiology and behavior, growth and development, and human health from evolutionary and ecological perspectives. An introductory section provides readers with the historical, theoretical, and methodological foundations needed to understand the more complex ideas presented later. Two hundred discussion ques- tions provide starting points for class debate and assignments to test student understanding. Michael P. Muehlenbein is an assistant professor of anthropology at Indiana University, Bloomington. He holds an MsPH in both tropical medicine and biosta- tistics from Tulane University, as well as an MPhil and PhD in biological anthropol- ogy from Yale University. His research interests are focused most recently on (1) evaluating hormone-mediated immune functions in reference to evolutionary and life history theories, and (2) investigating potential zoonotic and anthropozoo- notic pathogen transmission associated with primate-based ecotourism. He has received teaching awards for his graduate and undergraduate courses on human biological variation, behavioral endocrinology, evolutionary medicine, and global health. In addition to running an endocrinology and infectious disease laboratory in Indiana, he presently conducts fieldwork in the United States, Malaysia, Dominica, and the Dominican Republic.



HUMAN EVOLUTIONARY BIOLOGY Edited by MICHAEL P. MUEHLENBEIN Indiana University, Bloomington

CAMBRIDGE UNIVERSITY PRESS Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sa ˜o Paulo, Delhi, Dubai, Tokyo Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521879484 # Cambridge University Press 2010 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2010 Printed in the United Kingdom at the University Press, Cambridge A catalogue record for this publication is available from the British Library Library of Congress Cataloguing in Publication data ISBN 978-0-521-87948-4 Hardback ISBN 978-0-521-70510-3 Paperback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Contents List of contributors page vii 11 Human Adaptation to High Altitude 170 Preface xi Tom D. Brutsaert 12 Skin Coloration 192 Part I Theory and Methods 1 Nina G. Jablonski 1 Evolutionary Theory 3 13 Classic Markers of Human Variation 214 Douglas J. Futuyma Robert J. Meier 2 The Study of Human Adaptation 17 14 DNA Markers of Human Variation 238 A. Roberto Frisancho Michael E. Steiper 3 History of the Study of Human Biology 29 15 Ten Facts about Human Variation 265 Michael A. Little Jonathan Marks 4 Genetics in Human Biology 48 16 The Evolution and Endocrinology Robert J. Meier and Jennifer A. Raff of Human Behavior: a Focus on Sex Differences and Reproduction 277 5 Demography 74 Peter B. Gray James Holland Jones 6 History, Methods, and General Applications of Anthropometry Part III Reproduction 293 in Human Biology 92 17 Human Mate Choice 295 Noe ¨ l Cameron and Laura L. Jones David P. Schmitt 7 Energy Expenditure and Body 18 Mate Choice, the Major Composition: History, Methods, Histocompatibility Complex, and Inter-relationships 113 and Offspring Viability 309 Peter S. W. Davies and Alexia J. Murphy Claus Wedekind and Guillaume Evanno 8 Evolutionary Endocrinology 127 19 Why Women Differ in Ovarian Function: Richard G. Bribiescas and Genetic Polymorphism, Developmental Michael P. Muehlenbein Conditions, and Adult Lifestyle 322 9 Ethical Considerations for Human Grazyna Jasienska Biology Research 144 20 Pregnancy and Lactation 338 Trudy R. Turner Ivy L. Pike and Lauren A. Milligan With commentary by Michael P. Muehlenbein 21 Male Reproduction: Physiology, Commentary: a Primer on Human Subjects Behavior, and Ecology 351 Applications and Informed Consents 150 Michael P. Muehlenbein and Michael P. Muehlenbein Richard G. Bribiescas Part II Phenotypic and Genotypic Variation 155 Part IV Growth and Development 377 10 Body Size and Shape: Climatic and Nutritional Influences on Human Body Morphology 157 22 Evolution of Human Growth 379 William R. Leonard and Peter T. Katzmarzyk Barry Bogin v

vi Contents 23 Variation in Human Growth Patterns due to 29 Evolutionary Medicine and the Causes Environmental Factors 396 of Chronic Disease 502 Stanley J. Ulijaszek Paul W. Ewald 24 Evolutionary Biology of Hormonal Responses 30 Beyond Feast–Famine: Brain Evolution, to Social Challenges in the Human Child 405 Human Life History, and the Metabolic Mark V. Flinn Syndrome 518 Christopher W. Kuzawa 25 Human Biology, Energetics, and the Human Brain 425 31 Human Longevity and Senescence 528 Benjamin C. Campbell Douglas E. Crews and James A. Stewart 26 Embodied Capital and Extra-somatic Wealth 32 Evolutionary Psychiatry: Mental in Human Evolution and Human History 439 Disorders and Behavioral Evolution 551 Jane B. Lancaster and Hillard S. Kaplan Brant Wenegrat 33 Industrial Pollutants and Part V Health and Disease 457 Human Evolution 566 Lawrence M. Schell 27 Evolutionary Medicine, Immunity, and Infectious Disease 459 34 Acculturation and Health 581 Michael P. Muehlenbein Thomas W. McDade and Colleen H. Nyberg 28 Complex Chronic Diseases Index 603 in Evolutionary Perspective 491 S. Boyd Eaton

Contributors Barry Bogin Paul W. Ewald Department of Human Sciences Department of Biology Loughborough University University of Louisville Loughborough, UK Louisville, KY, USA Richard G. Bribiescas Mark V. Flinn Department of Anthropology Department of Anthropology Yale University, University of Missouri New Haven, CT, USA Columbia, MO, USA Tom D. Brutsaert A. Roberto Frisancho Department of Exercise Science Department of Anthropology Syracuse University University of Michigan Syracuse, NY, USA Ann Arbor, MI, USA Noe ¨l Cameron Douglas J. Futuyma Department of Human Sciences Department of Ecology and Evolution Centre for Human Development and Ageing State University of New York, Stony Brook Loughborough University Stony Brook, NY, USA Loughborough, UK Peter B. Gray Benjamin C. Campbell Department of Anthropology and Ethnic Department of Anthropology Studies University of Wisconsin University of Nevada, Las Vegas Milwaukee, WI, USA Las Vegas, NV, USA Douglas E. Crews Nina G. Jablonski Department of Anthropology Department of Anthropology Ohio State University Pennsylvania State University Columbus, OH, USA University Park, PA, USA Peter S. W. Davies Grazyna Jasienska Children’s Nutrition Research Centre Department of Epidemiology and Population Discipline of Paediatrics and Child Health Studies The University of Queensland Institute of Public Health Royal Children’s Hospital Jagiellonian University, Collegium Medicum Herston, Australia Krakow, Poland S. Boyd Eaton James Holland Jones Department of Anthropology Department of Anthropological Sciences Emory University Stanford University Atlanta, GA, USA Stanford, CA, USA Guillaume Evanno Laura L. Jones Department of Ecology and Evolution Division of Epidemiology and Public Health University of Lausanne Nottingham City Hospital Lausanne, Switzerland Nottingham, UK vii

viii List of contributors Hillard S. Kaplan Royal Children’s Hospital Department of Anthropology Herston, Australia University of New Mexico Colleen H. Nyberg Albuquerque, NM, USA Department of Anthropology Peter T. Katzmarzyk University of Massachusetts Boston Pennington Biomedical Research Center Boston, MA, USA Baton Rouge, LA, USA Ivy L. Pike Christopher W. Kuzawa Department of Anthropology Department of Anthropology University of Arizona Northwestern University Tucson, AZ, USA Evanston, IL, USA Jennifer A. Raff Jane B. Lancaster Department of Anthropology Department of Anthropology University of Utah University of New Mexico Salt Lake City, UT, USA Albuquerque, NM, USA Lawrence M. Schell William R. Leonard Department of Anthropology Department of Anthropology State University of New York, Albany Northwestern University Albany, NY, USA Evanston, IL, USA David P. Schmitt Michael A. Little Department of Psychology Department of Anthropology Bradley University State University of New York, Binghamton Peoria, IL, USA Binghamton, NY, USA Michael E. Steiper Jonathan Marks Department of Anthropology Department of Anthropology Hunter College University of North Carolina, Charlotte New York, NY, USA Charlotte, NC, USA James A. Stewart Thomas W. McDade Department of Social and Behavioral Sciences Department of Anthropology Columbus State Community College Northwestern University Columbus, OH, USA Evanston, IL, USA Trudy R. Turner Robert J. Meier Department of Anthropology Department of Anthropology University of Wisconsin-Milwaukee Indiana University Milwaukee, WI, USA Bloomington, IN, USA Stanley J. Ulijaszek Lauren A. Milligan Institute of Social and Cultural Anthropology Department of Anthropology University of Oxford University of California, Santa Cruz Oxford, UK Santa Cruz, CA, USA Claus Wedekind Michael P. Muehlenbein Department of Ecology and Evolution Department of Anthropology University of Lausanne Indiana University Lausanne, Switzerland Bloomington, IN, USA Brant Wenegrat Alexia J. Murphy Department of Psychiatry and Behavioral Children’s Nutrition Research Centre Sciences Discipline of Paediatrics and Child Health Stanford University The University of Queensland Palo Alto, CA, USA

Preface In review of a different text, Moses Hadas (1900–1966) reviews of basic evolutionary biology, molecular bio- remarked that “this book fills a much-needed gap.” logy, biological anthropology, behavioral ecology, and Unlike the text he was referring to, the topic of human statistics. We have, however, tried to produce a text evolutionary biology deserves no such gap in our readable to a wide audience and organized in an intui- understanding. To identify one’s place in nature and tive fashion. Part I of the book begins where it should, to appreciate how human evolution has been guided by with basic and detailed reviews of theory, history, and the same evolutionary principles that guide other methods in human evolutionary biology. This includes organisms is humbling and necessary. We are products introductions to evolutionary theory, human adap- of evolution, and this is reflected throughout our bio- tability, genetics, demography, evolutionary endocri- logy and behaviors. nology, anthropometry, and nutrition/energetics, as The purpose of our text is to provide thorough and well as the history of the study of human biology. We modern reviews of a wide range of pertinent aspects of even introduce readers to some of the ethical consider- human evolutionary biology and contemporary human ations for human biology research. Clearly the purpose biological variation. The history of research on human of Part I is to provide readers with a basic foundation biological variation is a long one, and includes studies in theory and methodology that can be used as a basis on general human adaptability, variations in growth for understanding some of the more complex problems patterns, body sizes and shapes, genetic diversity, and presented by authors throughout the remainder of the race concepts. More recently, the study of human bio- volume. logy has included analyses of reproductive physiology Part II of the book focuses on phenotypic and geno- and behavior within evolutionary and ecological typic variation. This has been the bread and butter of frameworks. Other advancements include the science research on contemporary human biological variation. of evolutionary medicine, in which evolutionary Body size and shape, skin color, and adaptations to research on health and disease is used to elucidate high altitude are all revisited. Chapters 13 and 14 pro- medical research and practices. vide detailed accounts of classic (e.g., serological, etc.) The text before you is different from most others. and DNA markers of human variation, and Chapter 15 Unlike traditional texts on evolutionary biology, our importantly addresses the “race concept,” a long-held book focuses specifically on humans and the applica- discussion in physical and cultural anthropology. tion of evolutionary theories on understanding modern A chapter on human behavioral endocrinology logic- human variation and adaptability. Unlike traditional ally bridges our section on phenotypic/genotypic vari- texts on human biology, our book does not include ation with the following Part III: reproduction. This detailed descriptions of all human physiological and section begins with more discussion on human beha- anatomical systems. You will also not find detailed viors, specifically mate choice. Female reproductive accounts of the history of human evolution, where we ecology/physiology is addressed in Chapters 19 and came from and how we are related to other species. 20, and male reproductive ecology/physiology in What you will find are historical perspectives on the Chapter 21. Unfortunately, we were unable to provide study of human evolutionary biology, detailed reviews at this time a chapter on female reproductive senes- of modern methods for studying human evolutionary cence (i.e., the menopause). biology, descriptions of fundamental research on geno- As Part IV focuses on growth and development, it typic and phenotypic variation within and between contains discussions on the evolution of, and variation contemporary human populations, comprehensive in, rates and patterns of somatic growth. This includes discussions on human reproductive physiology and classic and modern ideas on the sensitivity of human behavior as well as evolutionary medicine. development to environmental factors like nutrition, It is not possible to produce one single inclusive disease, and even social challenges. Chapter 26 shows text on such a diverse topic. Supplemental materials us how human life histories, cognition, and body/brain to our book would include, among others, detailed development are intimately intertwined. Its discussion ix

x Preface on human longevity also serves as a logical transition contents of different chapters to their classmates. It is to our final section of the book. Part V focuses on never too early to learn how to give an effective presen- various aspects of human health from evolutionary tation. Readers are also encouraged to utilize the and ecological perspectives. This includes basic dis- questions listed at the end of each chapter to facilitate cussions of immunity and infectious diseases, the discussion. Identifying and stimulating future direc- evolution of chronic diseases (including the meta- tions of research should be the primary goal. bolic syndrome and mental disorders), the infectious Lastly, credit must be given where credit is due. causes of some chronic diseases, and human senes- Acknowledgements belong to Dr. Dominic Lewis and cence. We conclude the section and book with discus- others at Cambridge University Press for being brave sions on some of the cultural determinants of health enough to tackle this project. The friends, families, and that have and will continue to be influential on human colleagues who assisted each contributing author evolution. are too numerous to list, but all deserve recognition. We have intended this text to be used as a general We do not do this by ourselves or for ourselves. reference for professional scholars as well as graduate/ postgraduate students and advanced undergraduates . . . And we must acknowledge, as it seems to me, that man in evolutionary biology, biological anthropology, and with all his noble qualities, with sympathy which feels for the most debased, with benevolence which extends not only other academic programs. The ultimate goal is to to other men but to the humblest living creature, with his forward research on human evolutionary biology by god-like intellect which has penetrated into the movements identifying gaps in our understandings of this inclusive and constitution of the solar system – with all these exalted discipline. Admittedly, it may be difficult to cover all powers – Man still bears in his bodily frame the indelible chapters of this book in a single semester in a class- stamp of his lowly origin. room setting. In this case, instructors may choose to focus only on certain sections. Alternatively, instruct- Charles Darwin (1809–1882), The Descent of Man and ors may choose to have different students present the Selection in Relation to Sex (1871), p. 405.

Part I Theory and Methods “The natural phenomena of the evolutionary history of man claim an entirely peculiar place in the wide range of the scientific study of nature. There is surely no subject of scientific investigation touching man more closely, or in the knowledge of which he is more deeply concerned, than the human organism itself; and of all the various branches of the science of man, or anthropology, the history of his natural evolution should excite his highest interest.” Ernst Haeckel (1834–1919), The Evolution of Man (1892), p. 2 1



1 Evolutionary Theory Douglas J. Futuyma Our contemporary understanding of evolutionary material, and even changes in ploidy. Many mutations processes builds on theory developed during the have no effect on phenotype or fitness (are selectively “Evolutionary Synthesis” of the 1930s and 1940s, when neutral), such as synonymous mutations in protein- Darwin’s ideas, especially on natural selection, were coding regions, which do not alter amino acid sequence, joined with Mendelian genetics. Since, then, of course, and mutations that occur in pseudogenes and other our understanding of evolution has been greatly apparently nonfunctional regions. There exists greater advanced by the discoveries in molecular genetics, as potential for fitness effects of nonsynonymous muta- well as by continuing elaboration of the “neo-Darwinian” tions in coding regions, or of mutations in regulatory theory that issued from the Evolutionary Synthesis sequences. The rate of mutation (usually on the order (Futuyma, 1998, 2005). of 10 –9 per base pair per gamete) is usually too low A capsule summary of contemporary theory, to be to be a significant factor in driving allele frequency followed by more detailed explication, is as follows. change within a population, but it can determine the Elementary evolutionary change consists of changes rate of DNA sequence divergence in the long term, and in the genetic constitution of a population of organ- can influence the equilibrium level of standing genetic isms, or in an ensemble of populations of a species. variation. Considerable contemporary research con- These genetic changes may be reflected in change of cerns whether or not rates and directions of pheno- the population mean or variance of phenotypic charac- typic evolution are often constrained by the supply of teristics. Any change requires that genetic variation suitable mutations (Houle, 1998; Blows and Hoffmann, originate by mutation of DNA sequences, and/or by 2005). Mutation is a random process, in the sense recombination. The minimal evolutionary process is that the probability of occurrence of a particular muta- an increase in the frequency of a mutation, or a set of tion is not affected by environmental circumstances mutations, within a population, and the corresponding which would make it advantageous. That is, there is decrease in frequency of previously common alleles. no known mechanism by which the mutational process Such frequency changes are the consequence of random can be directed by the environment in advantageous genetic drift (leading to occasional fixation of nearly directions. neutral genetic variants) or of diverse forms of natural selection. Successive such changes in one or more VARIATION WITHIN POPULATIONS characteristics cumulate over time, yielding potentially indefinite divergence of a lineage from the ancestral Based on studies of many species, it appears that most state. Different populations of a species retain similar- populations carry substantial sequence variation in ity due to gene flow and perhaps uniform selection, but many gene loci, and that there exists some heritable can diverge due to differences in mutation, drift, and/or variation in many or most “quantitative” phenotypic selection. Some of the consequent genetic differences traits (continuous traits such as size, as well as the can generate biological barriers to gene exchange number of highly repeated unit characters, such as between populations, resulting in the formation of hairs or scales). Presence of two or more fairly common different biological species. alleles or genotypes within a population is referred to as polymorphism. The level of variation is enhanced by THE ORIGIN OF VARIATION mutation, recombination (often but not always), gene flow from other, genetically differentiated, populations, Mutational changes in DNA sequences are of many and some forms of natural selection (e.g., frequency- kinds, ranging from single base-pair alterations to dependent selection, below). It is eroded by genetic insertions, deletions, and rearrangements of genetic drift and by most forms of natural selection (including Human Evolutionary Biology, ed. Michael P. Muehlenbein. Published by Cambridge University Press. # Cambridge University Press 2010. 3

4 Douglas J. Futuyma directional and stabilizing selection on quantitative termed the heritability (in the narrow sense), defined traits). The analysis of genetic variation is based on more narrowly than the “broad sense heritability” the frequencies of the alleles and genotypes at individ- V G /V P . Because V A is a function of allele frequencies, ual genetic loci (for an introduction to population and V P includes the environmental variance V E ,an genetics, see Hartl and Clark, 1997) For sexually estimate of the heritability of a trait is valid only for reproducing populations, the Hardy–Weinberg (H-W) the particular population and the particular environ- ment in which it was estimated, since other popula- theorem states that the frequency of each allele (p i for allele i) will remain constant from generation to tions might differ in both these respects. Although generation unless perturbed by mutation, gene flow, many or most characters are genetically variable, sampling error (genetic drift), or selection, and that we do not know what fraction of this variation can the frequencies of the several genotypes will likewise contribute to evolution by natural selection, since it is remain constant, at values given by the binomial the- possible that a considerable portion of the variation 2 orem (p i for homozygote A i A i , and 2p i p j for hetero- may be deleterious under most circumstances. zygote A i A j ), if mating occurs at random. A single The “mapping,” or relationship, between a pheno- generation of random mating establishes H-W geno- typic character state (e.g., body mass) and the environ- type frequencies at any autosomal locus. Furthermore, ment (e.g., caloric intake) is a genotype’ norm of alleles at two or more polymorphic loci will become reaction. Genotypes may differ in their norms of reac- randomized with respect to each other (a state of tion; for example, some people may gain more weight linkage equilibrium) due to recombination. These prin- on a given diet than others. Such differences give rise to ciples have important consequences; for example, at a genotype X environment interaction, expressed at the H-W equilibrium, a rare allele exists mostly in hetero- population level by the variance component V GXE . The zygous state, and so is concealed if it is recessive. In “mapping” between genotypes and phenotypes, even fact, rare, deleterious recessive alleles exist at a great within a constant environment, often depends on devel- many loci in populations of most outcrossing species, opmental processes. For example, a trait may simply including humans. The frequency of heterozygotes increase or decrease additively and gradually as þ or (“heterozygosity”) at a locus in H-W equilibrium (2p i p j ) alleles (those that increase or decrease the character) is often used as a measure of genetic variation at are substituted in the genotype; or there may be non- that locus, since variation is maximized when allele linear effects, so that the character suddenly and frequencies are equal. changes from one to another discretely different state Phenotypic variation in most quantitative traits is when the number of þ alleles crosses a threshold. continuous or almost so, because it is polygenic, based A gene commonly affects two or more characters on segregating alleles at several or many loci, and also (pleiotropy), and so can contribute to a genetic correl- includes environmental effects on the development or ation (r G ) between them. Another possible cause of expression of a character (Falconer and Mackay, 1996; genetic correlation is linkage disequilibrium, nonran- Barton and Keightley, 2002). At many of the segregat- dom association of certain alleles at two or more loci ing loci, the individual effects of alleles on the charac- within a population (e.g., an excess of AB and ab com- ter typically are small, relative to the range of variation, binations and a deficiency of Ab and aB). A genetic but substantially larger effects are commonly contrib- correlation caused by pleiotropy may be the net effect uted by segregating alleles at a few loci. The variance in of both positive and negative components, since alleles phenotype (V P ) includes a genetic component (genetic at some loci may affect both characters in the same variance, V G ) and an environmental component (V E ), direction, and at other loci, in opposite directions. The and often an interaction effect (V G.E ) as well. An value of r G depends on the frequencies and phenotypic important component of V G is the “additive genetic effects of all contributing loci. It is estimated by the variance” (V A ), which is described by the correlation covariance between characters over a set of families, between the phenotype of parents and their offspring; just as the genetic variance is estimated for a single it is this component of variation that is most important character. Genetic correlations are important because for evolution by natural selection. This component if the population mean of one character is altered, consists of the “additive” effects of alleles, that is, the perhaps by natural selection, the other character will phenotypic effect of each allelic substitution, averaged also be changed. over all the genetic backgrounds in which it occurs. V A depends on the number of loci contributing to the character, on the evenness of allele frequencies at each GENETIC DRIFT locus, and on the average magnitude of the phenotypic 2 effect of different alleles. (V A ¼ 2Sp i p j a in the simplest Random genetic drift is simply random change in the case, where a is the average phenotypic effect and frequency of alleles (and consequently, of genotypes). S indicates summation over loci.) The ratio V A /V P is The genes carried by a generation of newly formed

Evolutionary Theory 5 zygotes in a population are a sample of the genes Since DNA sequence data have become available, carried by the previous generation, to which the another theoretical approach to studying the dynamics parents belong. Because of random sampling error, of genetic variation, coalescent theory, has become the frequency (p) of an allele, say A i ,amongthe prominent (Hein et al., 2005). Looking back in time zygotes is unlikely to be exactly the same as in the from the present, the gene copies (at a particular gene previous generation, since there is likely to have locus) in the population today are descended from only been random mortality and random variation in some of the genes carried by the previous generation’s female reproduction (fecundity) and male reproduc- zygotes, due to sampling error; those zygotes in turn tion (number of mates) among individuals in the carried genes descended from only some of those in previous generation (here we are considering their parents’ generation; and so on. Pursuing this random, not selective, variation in survival and logic, it is inevitable that all the gene copies in the reproduction). So although the allele frequency in a population today are descended from one single ances- new generation of N zygotes (carrying 2N genes if tral gene copy (one DNA molecule) at some time in the species is diploid) is p on average (the same as in the past. The descendants of that gene form lineages the previous generation), the frequency distribution of genes, replicating down through the generations to of possible allele frequencies has a variance,givenby the present time, the set of lineages forming a gene tree the binomial expression Var (p) ¼ p(1–p)/(2N). This which, like a phylogenetic tree of species, portrays may be conceptualized as the variation among a their ancestry back (“coalesces to”) the common ances- large number of possible samples of 2N genes. The tral (CA) gene, which existed t CA generations ago. That greater Var (p) is, the greater the random change in ancestor was one of some number (say, 2N) of genes in allele frequency is likely to be, from generation to the population at that time, but the descendants of generation, and thus the faster the process of evolu- those other genes have not persisted to the present tionary change by genetic drift. The expression for time, due to random genetic drift. (When this history Var (p) tells us that this happens faster, the smaller was first described for human mitochondrial DNA, the population size N. N in this theory refers to the catchy phrase “mitochondrial Eve” was applied the effective size of the population, which is smaller to the female that carried the ancestor of all human than the “census size” if individuals vary in repro- mitochondrial genomes. Some people wrongly sup- ductive rate, if the sex ratio among breeding individ- posed that this meant the ancestral human population uals departs from 1:1, or if the population fluctuates consisted of only one woman [and presumably one in size. man].) The speed of genetic drift depends on popula- Since this variance holds in each generation, p tion size, so it will not be surprising to learn that for fluctuates at random from generation to generation a population of constant effective population size N with no corrective tendency to return to its starting (2N genes at a diploid locus), the average time back to point, in a “random walk” to a boundary from which the common ancestor of all contemporary genes, t CA , no return is possible: either loss of the allele A i from the is 4N generations (e.g., four million if the effective population or fixation of the allele A i , i.e., attainment of population size is one million individuals). p ¼ 1. (Movement away from this boundary is possible, A gene tree, representing the history of common however, if new variation enters the population by ancestry of a sample of gene copies from one or more mutation or by gene flow from other populations.) populations of a species, can be estimated by phylo- Hence, genetic drift results in loss of genetic variation genetic methods, using as characters the mutational within a population. differences (e.g., nucleotide substitutions) that have If a number of separate populations of the species accrued among the lineages during their descent from all began with the same initial p, different populations their common ancestor. would have different random paths, and in principle A i would become fixed in some and lost in others; thus, genetic drift results in variation (divergence) THE NEUTRAL THEORY OF MOLECULAR among populations. An allele is more likely to be lost EVOLUTION than to be fixed if its frequency is near zero, and conversely if its frequency is near 1.0; in fact, the Building on these principles, Motoo Kimura pioneered probability, at any time, t, that an allele will eventually the development of a neutral theory of molecular evo- become fixed is p t , its frequency at that time. A new lution that is the basis for analyzing DNA sequence mutation often exists, at first, as a single gene copy variation within and among species, and is often con- among the 2N genes carried by the N individual sidered the “null hypothesis” against which alternative organisms in a population, so its initial frequency is hypotheses, such as natural selection, must be com- 1/(2N), and this is its probability of fixation (if it is pared (Kimura, 1983; Nei and Kumar, 2000). Muta- selectively neutral). tional changes occur at many sites in a DNA sequence,

6 Douglas J. Futuyma at a total rate of, say, u T per gene per generation. suggest that some supposedly nonfunctional, “junk,” Of these, suppose some fraction f is selectively neutral, DNA sequences may have unknown functions, perhaps so the neutral mutation rate is u ¼ fu T . (The fraction in gene regulation.) The ratio o ¼ k A /k S , where k A and f may depend on the functional role of a DNA sequence k S are the rates of nonsynonymous and synonymous or the effect of a nucleotide change; for instance, a nucleotide substitution, respectively, is often used as synonymous mutation in a functional gene or any muta- an index of the degree to which a protein-coding DNA tion in a nonfunctional sequence such as a pseudogene sequence has been evolving neutrally, relatively free is more likely to be selectively neutral than a nonsynon- of functional constraint. If all mutations have been ymous mutation in a gene with a critical function.) selectively neutral, o should equal 1. Since 2N genes are carried by (diploid) zygotes in each Genetic variation is lost from a population by gen- generation, the total number of new mutations in etic drift, as we have seen. However, it is regenerated the population each generation is 2Nu,onaverage.We by mutations at many sites in a DNA sequence, and know from genetic drift theory that the probability of at equilibrium there exists variation in nucleotide fixation of a new neutral mutation is 1/(2N) in a diploid sequence within a population, when the rate of input population of constant size N,so2Nu  1/(2N) ¼ u new by neutral mutation balances the rate of loss by genetic mutations occur each generation that will eventually drift. A measure of polymorphism is the expected pro- be fixed. The time to fixation, we have just learned, portion of base pairs that differ between two gene is 4N generations, on average. Since this is the case copies taken at random (p) from a population. At equi- each generation, u mutations should be fixed in a popu- librium this equals 4Nu, i.e., it is proportional to the lation every generation on average. In other words, population size and the mutation rate. Consequently, population-wide substitutions of nucleotides in a DNA effective population size can be estimated from p/4u. sequence occur at a roughly constant rate, so DNA Because of polymorphism, the history of popula- sequence evolution theoretically acts as a molecular tion separation may not be the same as the history of clock, accumulating ut substitutions over the course of any one gene locus. Suppose an ancestral population t generations. If two populations (or species) are derived divides into two populations (or species) A and B at from a common ancestor and do not exchange genes time t 1 , and B later separates into populations B 1 and for t generations, and if mutations at different sites in B 2 at time t 2 . Populations B 1 and B 2 are more closely the DNA sequence are fixed in each population, the related to each other, by definition, than they are difference D between sequences taken from the two to population A. If population B became fixed for a populations will be D ¼ 2ut.Ifu (the neutral mutation new mutation, and thus for a different sequence than rate, which can vary among genes because of functional population A, the mutation would be inherited by differences or DNA repair processes) can be calibrated, populations B 1 and B 2 and provide evidence of their then the time since the two populations separated sister-group relationship. Suppose, however, that popu- can be estimated from the observed difference D,as lations A and B, and their common ancestor, have t ¼ D/2u. (Calibration is usually based on geologically effective size N, and that the time between successive dated events, such as fossils of the studied lineage or splits (between t 1 and t 2 ) is less than the 4N generations related lineages, or separation of two land masses on required for one or another sequence variant to be fixed which related taxa reside.) in each population by genetic drift. If the common Eventually, D increases at a lower rate and levels ancestor is polymorphic for sequences x and y (per- off, because mutational substitutions occur repeatedly haps differing by a new mutation in sequence x), fix- at the same sites within the sequence, erasing evidence ation may not occur until after the three populations of previous substitutions. This happens sooner for have become separate. Then one sequence (say, x) may rapidly than slowly evolving sequences. According to be fixed in both A and one of the derived B-populations the neutral theory, evolutionary change is more rapid (say B 1 ), and the other sequence (y) may become fixed if mutations do not affect organismal function, since in B 2 . The phylogeny of genes may be accurate (the mutations that affect protein function are more likely gene copies in B 1 are most closely related to those in A), to be deleterious and eliminated by natural selection. but it would differ from the phylogeny of the popula- Consequently, evolution is predicted, and found, to be tions. Therefore, it is important to use information more rapid at third-base than second-base positions in from several or many independently inherited genes codons, because third-base mutations are more often when analyzing the historical relationships among synonymous. Sequence evolution is also more rapid in populations or species that have become separated nonfunctional sequences, such as pseudogenes, than during a short time span. in functional sequences. (Indeed, the rate of sequence Summarizing this section, note that for selectively evolution between species is now used by molecular neutral mutations, whose fate is unaffected by natural biologists to target functionally important versus less selection, the theory of genetic drift and the related important sequences. This and other lines of evidence neutral theory of molecular evolution provide a basis

Evolutionary Theory 7 for many important inferences: e.g., inferring effective Let us consider selection among individual organ- population size, time since separation of populations isms in a sexually reproducing population that differ in (or since speciation), historical relationships among genotype at a single locus with two alleles, A and a. populations, and whether or not natural selection has In the simplest case, the fittest of the three genotypes affected DNA sequence divergence and polymorphism. AA, Aa, and aa is a homozygote. If aa is rare, because the environment previously favored AA and has only recently changed so that aa is now the fittest genotype, NATURAL SELECTION we speak of directional selection for aa. Once aa becomes the prevalent genotype, allele A, as well as There are so many nuances to the concept of natural any other disadvantageous alleles that may arise by selection that a simple, comprehensive definition is mutation, are reduced in frequency, and selection is difficult to devise, but it may suffice, for present pur- often termed purifying. These are two faces of the same poses, to define it as consistent (nonrandom) differences coin, selection that fixes the allele that, in homozygous in the rate of survival or reproduction among classes of state, maximizes fitness. The frequency q of the advan- entities that differ in inheritable characteristics. The tageous allele a attains the deterministic equilibrium term “reproductive success” is often used for “survival q ¼ 1 if only selection is operating, but if other alleles and reproduction,” since survival to reproductive age is repeatedly arise by mutation, the equilibrium fre- prerequisite for reproduction. “Entities” is deliberately quency will be set by the mutation rate relative to the vague, because selection can (in principle) act among strength of purifying selection (“mutation/selection various kinds of biological “individuals,” such as genes balance”). Similarly, if a locally disadvantageous allele or larger sections of genetic material, individual organ- (perhaps A) that is advantageous in a different geo- isms, groups of conspecific organisms, species, or graphic population enters the population by gene flow, clades (Williams, 1992). We speak of “classes” of genes, the genetic equilibrium is determined by the relative individuals, etc., because we cannot tell if a difference strength of gene flow and purifying selection. Gene in reproductive success is nonrandom from informa- flow from other populations can sometimes severely tion about a single individual of each kind; we require diminish the degree of adaptation of populations to samples of similar genes or individuals in order to see their local environment. if there is a consistent difference between different Suppose the advantageous allele a is very rare, types of alleles or phenotypically different organisms. either because it has recently originated by mutation Natural selection, in distinction from genetic drift, is or because it has formerly been disadvantageous marked by a consistent difference in mean reproduct- but nevertheless persisted in the population due to ive success within a given environment, not a random, mutation/selection balance. If the frequencies of A unpredictable difference; thus natural selection is the and a are p and q respectively, the Hardy–Weinberg antithesis of chance. frequencies of the two genotypes that contain the a 2 allele, Aa and aa, are 2pq and q , and the vast majority of the a genes are carried by heterozygotes. (For 2 MODES OF SELECTION example, if q ¼ 0.01, 2pq ¼ 0.0198, q ¼ 0.0001, and the ratio of heterozygotes to homozygotes is 198:1.) Most analyses of evolution by natural selection are Whether or not the a allele can increase (or “invade” concerned with individual selection: differences in fit- the population) depends almost entirely on the fitness ness, owing to a genetically variable phenotypic charac- of Aa relative to the prevalent homozygous genotype ter, among individual organisms within a population. (AA); at this stage the fitness of aa is almost irrelevant In the simplest models, the character is affected by because it is so rare. This means that even if aa is the variation at a single locus, and we suppose that the fittest genotype, the a allele will not increase if it fitness of each genotype can be estimated. In practice, reduces the fitness of the heterozygote. This illustrates this can be difficult, because fitness, defined as a geno- that natural selection acts only in the present, and type’s relative rate of increase, i.e., growth in numbers cannot look forward toward the best possible outcome. from one generation to the next, depends on several life- It also shows the value of the Hardy–Weinberg principle. history parameters. The rate of increase is a complex Directional (or purifying) selection eliminates gen- function of the probability of survival at each age from etic variation, but several other modes of selection birth to the oldest reproductive age class, and on the (balancing selection) may maintain genetic polymorph- age-specific values of female reproduction (fecundity) ism. The simplest model is heterozygous advantage, and male reproduction (affected by mating success in which the fitness of Aa is greater than that of either and sometimes by sperm competition). (In some cases, AA or aa, and all three genotypes segregate each gener- it may be affected also by other complicating factors, ation due to random mating. Several hemoglobin poly- such as genetic compatibility among uniting gametes.) morphisms in human populations, including sickle

8 Douglas J. Futuyma cell hemoglobin, are the best-known of the few well- genetic make-up, so þ alleles rise in frequency. If the documented examples of this mode of selection. fitness/phenotype relationship is “open-ended” (e.g., Unquestionably more important is frequency-dependent the unlikely circumstance that bigger is always better), selection, in which the fitness of each genotype is a selection will ultimately favor the genotype with þ decreasing function of its own frequency in the popu- alleles only (and subsequently, any mutations with still lation, relative to other genotypes; that is, each geno- greater effects), so þ alleles become fixed at all loci, type is more and more advantageous, the rarer it is. genetic variation is eliminated, and evolution ceases Many biological phenomena, including competition except insofar as it continues to arise by mutation. for resources, social interactions, and resistance to Thus the magnitude of the “mutational variance,” the different genotypes of parasites, can give rise to such per-generation increment in the variance of the char- frequency-dependent effects. Mathematically, this is a acter due to new mutations, will then limit the rate of powerful way of maintaining multiple alleles in a popu- subsequent response to selection. lation, and cases are known in which 100 or more What if the relationship between fitness and pheno- alleles appear to be maintained this way. Variable type is not monotonic, but instead has an intermediate selection, in which different homozygotes are advanta- maximum (“optimum”) that lies above the current geous at different times or in different microhabitats mean phenotype? Directional selection will increase within the area occupied by a breeding population, the frequency of þ alleles and bring the mean to the can also maintain polymorphism, although this is by new optimum. The character then becomes subject to no means guaranteed: mathematical models show that stabilizing selection: deviations in either direction even if both homozygotes (AA and aa) are advanta- from the mean are disadvantageous. Many different geous in different environmental states, only a rather combinations of þ and  alleles can add up to give narrow range of combinations of selection intensities the same optimal intermediate phenotypic value; some and environmental frequencies will maintain all the of them are highly heterozygous, and others are homo- genotypes indefinitely. (Note that persistence of both zygous for þ alleles at some loci and for  alleles at homozygotes because of their variable fitnesses also other loci. Mathematical theory has shown that one or implies persistence of heterozygotes, due to random another of the homozygous genotypes will eventually mating.) replace all the other genotypes, so that genetic vari- The phenotypic implications of these genetic ation will be eliminated by stabilizing selection. models depend on the relation between genotype and Studies of natural populations have shown that phenotype. In simple cases, in which there is either the most common forms of selection on quantitative complete dominance of one allele or additive inherit- phenotypic characteristics are stabilizing selection and ance (in which the heterozygote’s phenotype is inter- disruptive (also called diversifying) selection, in which mediate), persistent genetic polymorphism implies two or more phenotypes have higher fitness than do persistence of two or three phenotypic classes, respect- the intermediates between them (Endler, 1986). Dis- ively. Most of the consequences of the single-locus ruptive selection at a single locus generally implies that models carry over into thinking about the effects of the heterozygote for two alleles A and a has lower selection on a polygenic phenotypic trait, in which fitness than both homozygotes. Such a polymorphism each variable locus contributes a small amount to over- is unstable, however, and the population will become all variation. We consider the simplest case, an additive fixed for the initially more common allele. In models of character, measured in, say, millimeters, for which “þ” disruptive selection on an additive polygenic character, and “–” alleles at each of k loci add or subtract the same variation is not maintained indefinitely; instead, the amount. The mean and variance of the character are population mean evolves to one or the other of the determined by the frequency of the alleles at all of the superior phenotypes, and stabilizing selection then loci; the mean will clearly be higher (and the variance takes over and reduces variation. In both the single- lower) if most of the þ alleles have high frequency. locus and polygenic models, variation is maintained However, an intermediate mean could result from only if disruptive selection is frequency-dependent, many possible allele frequency arrays, ranging from a such that the fitness of the superior genotypes declines highly variable population with p ¼ 0.5 (i.e., þ and  as they become more abundant. The simplest example equally frequent) at each locus, to fixation of a single would be if the genotypes are each adapted for a differ- genotype that is homozygous for þ at half of the loci ent food or other limiting resource, so that competition and for  at the other half. becomes more intense, and fitness declines, as a par- Directional selection on the character occurs when ticular genotype becomes more abundant and depletes there is a monotonic relationship (at least over part of its resource. the range of possible phenotypes) between phenotype I have introduced frequency-dependent selection and fitness. For example, selection may favor larger as a negative feedback loop that can maintain stable phenotypes, namely those with more þ alleles in their coexistence of different genotypes in a single breeding

Evolutionary Theory 9 population. It is possible, however, to imagine that the that together more than make up for the reproduction fitness of individuals of a particular genotype increases foregone at an earlier age. On the other hand, if abun- as the genotype’s frequency increases. This would dant inescapable predators make death at an early be a form of positive feedback that hastens fixation of age virtually inevitable, selection will favor early the genotype, eliminating variation. Such selection is reproduction, and mutations that defer senescence or easily envisioned for many social behavioral traits, enhance fecundity at advanced ages may well be dis- in which conformity to a predominant behavior pat- advantageous (if these effects reduce early fecundity). tern might be advantageous. The evolution of intrinsic senescence and mortality may therefore be affected by the age distribution of extrinsic mortality factors. Many potential adaptations COMPONENTS OF FITNESS have both benefits and costs, which may be environment- dependent. Genotypes may differ in fitness due to one or more A fitness component of particular interest is repro- components, most of which are generally considered ductive success achieved through success in mating, life history features (Stearns, 1992). These components which Darwin termed sexual selection (Andersson, contribute to the rate of increase (numbers/time) of a 1994). In many species of animals, the variation in genotype, relative to others. One may think of a popu- reproductive success, and therefore the potential inten- lation of organisms as consisting of subpopulations of sity of sexual selection, is greater in males than in different genotypes (or of alleles) that are all growing females. This difference is generally ascribed to the like a bank account, with compound interest. All else smaller and far more abundant gametes of males than being equal, a difference (in, say, survival probability females, but sexual selection acts more strongly on or fecundity) expressed at an earlier age generally has females of some species (e.g., phalaropes and some a bigger impact on growth in numbers (fitness) than a pipefish and seahorses), in which investment in pater- similar difference expressed at a later age. Suppose nal care of offspring limits the number of a male’s individuals reproduce from age three until ten, and potential mates. (Thus the “choosier” sex, that exerts then die. A mutation that increases the chance of sur- stronger sexual selection on the opposite sex, usually vival from age eight to nine has a smaller selective expends greater parental effort.) The two most com- advantage than one that provides a similar survival monly discussed modes of sexual selection are conflict advantage from age two to three, because survival between males, with winners gaining access to more enhancement in the older age classes will have much females, and female “choice” of some males over less effect on the number of offspring they might yet others, based on one or more characteristics that usu- produce (and the number of genes passed on). Simi- ally are actively displayed to females. (In many cases, larly, a mutation that increases reproductive output at the same trait seems to play a role in both male–male age three has a greater impact on the increase of the and male–female interactions.) There is considerable mutation’s frequency than if it affects reproduction at evidence that conflict between males selects for larger age ten, because (a) fewer individuals survive to age size, greater weaponry, and many other kinds of traits ten, so they don’t get the benefit of the mutation; that are used to establish dominance. The equilibrium and (b) the mutation expressed at the younger age mean value of such a trait will be set by balance effectively shortens the generation time, so more des- between the reproductive advantage it provides and cendants (grandchildren, great-grandchildren . . .) are disadvantages such as its energy costs or effects on produced per time unit than are produced by the geno- susceptibility to predation. Indeed, male investment type whose reproductive capacity is enhanced only at in features that enhance mating success, such as an older age. mating activity, weaponry, or display features, may Consequently, mutations that enhance survival reduce investment in maintenance (e.g., immune or the number of offspring (e.g., number of eggs or system) and survival. young) are expected to increase fitness, but the magni- There is considerable evidence from birds, insects, tude of increase depends on the age at which these and other animals that female “choice” imposes sexual effects come into play. Moreover, there may exist selection, but there is considerable uncertainty about trade-offs between different fitness components, or why females choose particular male phenotypes, such between a given component expressed at different as males with more vigorous displays or more highly ages, partly because an organism must partition elaborated ornaments or vocalizations. According to energy or nutrients (e.g., protein) among different one hypothesis, exaggerated male features indicate functions (the principle of allocation). For example, if high physiological vigor that may stem from superior reproduction reduces growth, it may be advantageous genetic constitution (the “good genes” hypothesis), and to delay reproduction until the individual is larger, females that choose such males will have fitter off- which may ensure a longer life and higher fecundity spring. There is some support both for this hypothesis

10 Douglas J. Futuyma and for several contenders. In models of runaway attempts to specify what the optimal character state sexual selection, a nonrandom association (linkage dis- ought to be, given some assumptions about benefits, equilibrium) develops between genes that affect a male costs, and constraints. This approach assumes that ornament and genes that affect the degree of female there has been enough time and enough genetic vari- preference for this character. Females that prefer more ation for natural selection to bring the mean character highly ornamented males have daughters that inherit state in a population nearly to its optimum value, and this preference (as well as unexpressed genes for large that the genetic details do not matter very much. male ornamentation) and sons that inherit larger orna- Whether or not these are reasonable assumptions ments (as well as unexpressed genes for heightened may depend on empirical information about such female preference). (Note that most features expressed things as genetic variation and evolutionary history by a single sex are encoded in the genome of both (e.g., inferences about how long a species has probably sexes.) Therefore, any increase in the average male been subject to consistent environmental selection). ornament in the population will cause a correlated Optimization is a common approach in the fields increase in the average female preference, and vice of functional morphology and physiology, in which it is versa, ratcheting both toward more extreme values assumed that fitness is enhanced by maximizing some until the process is halted either by counteracting function, subject to constraints such as costs in energy selection or by running out of genetic variation. or materials, or compromises with other functions. In a twist on sexual selection theory, females and For example, aerodynamic models have been used to males are engaged in sexual conflict: males reduce model flight and optimal wing morphology in birds, females’ fitness in various ways (e.g., incessantly in which compromises among speed, maneuverability, attempting to mate), females are selected to resist, and energy expenditure are taken into account. Among and selection favors males with ever more stimulating nonsocial aspects of behavior, models of optimal for- characteristics that can overcome female resistance aging describe when a foraging animal should give up (Arnqvist and Rowe, 2005). The scope for such inter- searching in one patch and move to another. actions appears greater than was formerly supposed, Social interactions entail complexities that make because it is clear that females of many species mate genetic modeling difficult, and have been analyzed with multiple males, even in species that form a sup- almost entirely by optimal models. The complexity posedly monogamous pair bond. Thus males have the arises from the frequency-dependent fitness of differ- potential of siring offspring by mating not only with ent trait values: the optimal behavior of an individual unmated, but also with previously mated, females. often depends on the behavior of other members of the The consequences include competition between sperm population. Among the most widely used approaches from different males. Probably because of the strong, is game theory (Maynard Smith, 1982). Suppose, for long-continued selection exerted by sperm competition example, that the problem is whether or not parental and sexual selection, reproductive characteristics, care, by either or both mated partners, will evolve by including male display features, genitalic morphology, individual selection. One might postulate two “strat- proteins from accessory reproductive glands, sperm egies,” “Stay and provide care to offspring” and “Defect morphology, and cell-surface proteins of gametes, are and attempt to reproduce again.” For each possible rapidly evolving characteristics that often are the pair (S♀/S♂,S♀/D♂,D♀/S♂,D♀/D♂), one postulates major differences among closely related species. for each partner the expected reproductive “payoff,” which depends on both the benefit to each partner (in terms of surviving offspring from this mating) and the MODELING ADAPTATION costs to each (in terms of the likely reproductive success sacrificed). The average fitness of each strategy, for In considering components of fitness, we have moved each sex, is then its payoff averaged over the possible from the very general theories of population and quan- pairings, and weighted by their frequency in a random- titative genetics, which apply to unspecified genes and pairing population. The best strategy, within the set of characters, to models of the evolution of specific strategies considered (here, S and D), is the one that, if classes of characters, such as life history features. fixed in the population, will remain fixed even if indi- Evolutionary analyses of adaptive evolution of specific viduals with alternative strategies attempt to invade. classes of characters employ several approaches to This is the evolutionarily stable strategy,orESS. modeling (Bulmer, 1994). The evolution of some fea- tures is best analyzed by genetic models. This is true of models of sexual selection by female choice, for LEVELS OF SELECTION example, because linkage disequilibrium is an essential component and it requires an explicit genetic approach. Natural selection was defined above as “consistent The major alternative is optimization, an approach that (nonrandom) differences in the rate of survival or

Evolutionary Theory 11 reproduction among classes of entities that differ in bearers on the reproductive success of individuals inheritable characteristics.” These “entities” may be at (kin) who carry the same allele due to common des- different, nested levels, and the effects of selection cent. (In this case, the “bearers” are parents, and the at different levels may be opposite (Okasha, 2006). “kin” are their offspring.) In the same way, genes that Consider, for example, the level “individual organism” enhance their bearers’ propensity to help more distant and the level “somatic cell lineage” within a multicellu- relatives may increase in frequency – but the conse- lar organism. If a cell lineage experiences a mutation quent increase in the relatives’ fitness must be greater, that causes rapid, unrestricted cell division, that lin- since their probability of sharing the “helping allele” is eage has a “selective advantage” relative to other cells, lower. William Hamilton formalized this relationship and will constitute an increasing proportion of cells in what has become known as Hamilton’s rule, which within the domain of the single organism (Nowak, states that “altruism” spreads if rb > c: an altruistic 2006). This proliferation – cancer – is clearly disadvan- trait can increase in frequency if the benefit (b) tageous to the higher-level entity (the organism), if it received by the donor’s relatives, weighted by their occurs before or during the organism’s reproductive relationship (r) to the donor, exceeds the cost (c)of ages. Selection among genetically variable individual the trait to the donor’s fitness. The relationship, r, organisms will favor genotypes that have the ability to between donor and recipient is the fraction of the suppress cancerous tumors. donor’s genes that, on average, are identical by descent We may likewise distinguish selection among indi- with any of the recipient’s genes. For example, r ¼ ½ vidual organisms with different genotypes (the level of between parent and offspring, so an allele for parental selection assumed so far in this chapter) from selection care should spread, even if it costs the parent her life at the level of the individual gene (locus). In asexual and subsequent reproduction, as long as her care organisms, there is little conflict between selection results in survival of more than two extra offspring at these levels, because the fate of a gene (survival, (compared to a parent that does not provide care). passage to subsequent generations) depends on that (Kin selection is only one of several explanations of the rest of the genotype to which it is bound. But of the evolution of co-operation among genes, cells, in sexually reproducing organisms, conflicts can arise. or conspecific organisms. For example, reciprocity A famous example is the “t locus” in house mice. More [“reciprocal altruism”] may evolve if individuals recog- than 90% of the sperm of males heterozygous for a nize one another and can benefit others or not, normal allele (T) and one of several recessive alleles depending on their history of behavior.) (t) carry the t allele (an example of meiotic drive). Some Because of kin selection, the family (mated pair of the recessive alleles cause embryonic death, and and associated offspring) is an obvious context in others male sterility, in homozygous condition. The which co-operation may evolve. Nevertheless, intrafa- differential transmission of T and t alleles constitutes milial interactions are riddled with conflict. Sexual differential “reproduction” at the gametic level (genic conflict inevitably arises from the sex difference in selection), opposing differential survival of individual gamete size (and some other features in some species): mice (individual selection). Genic selection accounts male fitness can be increased by mating with many for many phenomena, such as the proliferation of females, whereas all of a female’s eggs can usually be transposable elements (“selfish genetic elements”): fertilized by a single male. Female fitness is more likely DNA sequences that replicate more frequently than to be enhanced by her offspring’s survival, which may most of the genome. be increased by parental care or by “genetic quality.” Genic selection provides one way of viewing the Parental care increases the fitness of both parents, but evolution of co-operation, which stands in contrast to it entails costs, including lost reproductive opportun- the selfish individualism that generally characterizes ities. If offspring were as likely to survive with unipar- individual selection (Dawkins, 1982; Sober and Wilson, ental care as with biparental care, selection would 1998). Cells in multicellular organisms co-operate favor defection by one sex – the one for which parental because they are (generally) genetically identical: a care is more costly (Clutton-Brock, 1991). A further gene in a liver cell is replicated by virtue of the replica- complication is that if a caregiver were not actually tion of identical copies in the germ cell line – and the the parent of some or all of the offspring, he (or she) fate of the germ cell line depends on the gene copies would have less of a genetic interest in their survival. functioning in the liver. Likewise, the rate of increase “Extrapair copulation,” common in seemingly monog- of a parent’s gene over generations depends on the amous species of birds, therefore alters the costs and survival and replication of copies of that gene in benefits of parental care. In some species of primates the parent’s offspring – and so alleles that program and other mammals, a male that replaces another male parental care may increase in frequency. This is an kills his new mate’s offspring, since he has no genetic example of kin selection: selection in which alleles interest in them, and killing them enables him to father differ in fitness by influencing the effect of their his own offspring faster. (Killing some offspring can

12 Douglas J. Futuyma also be advantageous to parents if, by reducing compe- of geographic range), species may differ in their tition among the remaining offspring, it maximizes the probability of extinction or of speciation per unit time. number of healthy survivors.) Stephen Jay Gould and others, pointing out that these Trivers (1974) first pointed out that because a probabilities are the species-level analogues of differ- parent maximizes her (his) fitness by allocating care ential mortality and reproductive rates of individual among a number of offspring, present and future, the organisms, have proposed that species selection has optimal investment of care in any one offspring is shaped the frequency distribution of traits in large lower from the parent’s perspective than from the off- clades and biotas (Gould, 2002). The fraction of plant spring’s. The consequent parent–offspring conflict has species with flowers, for example, greatly increased many consequences for behavior, and even for preg- during the Mesozoic, and the world’s mammal fauna nancy. Haig (1993, 1997) has described how genes became increasingly dominated by placental euther- expressed in human and mouse fetuses that enhance ians in the late Mesozoic and Tertiary. Identifying the uptake of sugar and other nutrients from the mother trait(s) that have been the “target” of species selection are counteracted by maternally expressed genes that can be difficult, but is sometimes possible by compar- prevent the fetus from extracting too much. For ing the diversification rate of multiple clades that example, insulin levels are increased in pregnant have independently evolved a postulated diversity- women, decreasing blood glucose (just when one enhancing feature, compared to the diversification rate might expect it to be higher), in order to counteract a of sister clades that lack the feature. Herbivory in fetal hormone that has the opposite effect. insects, sexual dichromatism in passerine birds, and If an individual with an “altruism allele” dispenses low body mass in several orders of mammals seem benefits indiscriminately to both related and unrelated to be associated with increased diversification (Coyne individuals, the survival of both that allele and its and Orr, 2004). nonaltruistic (“selfish”) alternative allele are equally enhanced, but the donor suffers a cost (c), so the selfish allele increases. As a rule, individual selfishness SPECIATION increases within populations, even if the population as a whole suffers. In principle, extinction of whole There are several contending concepts of “species” populations of selfish individuals, and survival of because this word serves several purposes. As a term populations of co-operative individuals, could cause in classification, it may simply label phenotypically evolution of co-operation in the species as a whole. distinguishable populations. For instance, morpho- This would be group selection (also called population logically different sections of a single lineage in the selection and interdemic selection), in opposition to fossil record are sometimes given different names, individual selection. Some authors hold that group and may be termed “chronospecies.” Many systematists selection can play a role in evolution, especially if the use a phylogenetic species concept (PSC), according to groups are very small and temporary (Wilson, 1983). which a species is a diagnosably different cluster of (For example, if some groups of nestling birds include organisms, within which there is a parental pattern of co-operators and others do not, the overall productivity ancestry and descent. This would include asexual of those nests with co-operators may be higher, under forms. Most evolutionary biologists concerned with some circumstances.) However, because populations evolutionary processes use one or another version of of co-operators are likely to be invaded by immigrant the biological species concept (BSC), articulated by selfish genotypes (“cheaters”), which rapidly increase Ernst Mayr in 1942. A biological species is a group of within these populations, co-operation is unstable. actually or potentially interbreeding populations that Most evolutionary theorists therefore conclude that are reproductively isolated (by biological differences) group selection, evolution by differential extinction, from other such groups. The reproductive isolating or reproductive productivity of whole populations, is barriers (RIBs), which are usually genetically based, unlikely to play a major role in evolution. The import- include prezygotic barriers that reduce gene exchange ant consequence is that we generally do not expect before zygote formation (e.g., differences in habitat the evolution of characteristics that benefit the popula- association, timing of reproduction, behavior, pollin- tion or species, but which are disadvantageous to the ation [for plants], and failure of gametes to unite) and individual or its kin. postzygotic barriers that act after zygote formation, and Selection among groups, as long as it does not are expressed as diminished survival or reproduction oppose individual selection within groups, may how- of hybrid genotypes. The BSC applies only to sexual ever have evolutionary consequences. These may be organisms and may be more difficult to apply in prac- most evident when the groups are species. As a conse- tice than the PSC, since the potential ability of spatially quence of characteristics of their constituent individ- separated (allopatric) populations to interbreed may be uals, or of properties of the species as a whole (e.g., size difficult to evaluate. However, reproductive isolation

Evolutionary Theory 13 (RI) plays a critical role in long-term evolution, since it 2004). Only a few well-documented cases of sympatric enables populations, even if they eventually meet and speciation (completed or in process) satisfy skeptics overlap, to retain their distinct characteristics and to (Coyne and Orr, 2004). generate clades of species that subsequently elaborate What causes the evolution of RI between spatially those differences. It is even possible that speciation separated populations? As long as the populations are facilitates sustained evolution of morphological and allopatric, there cannot be selection to avoid hybridiza- other characteristics, by preventing the “slippage” that tion, so RI must evolve as a by-product of genetic interbreeding with other populations would cause divergence that transpires for other reasons. Genetic (Futuyma, 1987). drift, ecological selection, and sexual selection are the There is abundant evidence that many species postulated processes, but it is only recently that this of animals and plants form by genetic divergence of problem has been explicitly and rigorously studied. spatially disjunct (allopatric) or neighboring (parapatric) Divergent sexual selection, resulting in different populations of an ancestral species, since this reduces female preferences and male display traits, can result gene flow enough to allow divergent changes in allele in behavioral isolation. There is considerable indirect frequencies by natural selection or genetic drift. evidence for this hypothesis, in that the diversification Whether or not species are often formed sympatrically, rate of clades of birds and insects with features indica- i.e., by evolution of an intrinsic separation of a single tive of strong sexual selection has been higher than randomly mating population into two reproductively that of sister clades lacking those features. The high isolated forms, is controversial. Except for speciation rate of evolutionary divergence of genitalia, sperm, and by chromosome doubling (polyploidy), which is other features associated with reproduction is consist- common in plants but rare in animals, the evolution ent with this hypothesis. Diverse lines of evidence have of RI between populations, like the evolution of most also recently pointed to divergent ecological selection phenotypic differences, is gradual: arrays of popula- as a cause of RI. In some cases the effect of selection is tions can be found that display all degrees of pre- or fairly direct: beak size differences among Darwin’s postzygotic isolation from none to complete. Genetic finches, for example, are adaptations for feeding, but analyses, correspondingly, show that RI is polygenic, also are used by females to discriminate among males. based on contributions from several or many genes. In other cases, genes underlying ecological divergence Thus it is common to find partially reproductively may contribute to RI pleiotropically; examples include isolated populations (semispecies) that are not readily a correlation between copper tolerance and partial classified as the same or as different species. Moreover hybrid sterility among monkeyflower populations, and very importantly, hybridization between such and between adaptation to different host plants and forms in nature can result in some parts of the genome behavioral isolation in some herbivorous insects. readily “introgressing,” or penetrating from one semi- Genetic drift plays a role in Ernst Mayr’s (1954) species into the other, while other parts of the genome influential hypothesis of founder-effect speciation (also remain much more differentiated, because stronger called peripatric speciation). Mayr believed that select- divergent natural selection on these regions counter- ive advantage of one allele over another depends acts gene flow. strongly on which alleles it interacts with at other loci Because RIBs are usually polygenic, full RI requires (epistasis for fitness). He postulated that a population that two distinct clusters of different alleles be formed founded by few individuals will undergo genetic drift at loci that affect a RIB (AABBCC... and aabbcc. . .), at some loci, so that some previously rare alleles whereas recombination generates allelic mixtures become common by chance. This change in the (AaBBCc. . ., etc.) that are only partially reproductively “genetic environment” alters selection at interacting isolated and so form a “bridge” for gene exchange loci, so that previously disadvantageous alleles become between diverging populations. An extrinsic barrier advantageous. Consequently, natural selection, initi- (e.g., topography or unsuitable habitat) between diver- ated by genetic drift, alters the genetic constitution of ging populations, if it is seldom surmounted by disper- the newly founded population so that it may become sing individuals, reduces or eliminates this problem, reproductively isolated from the parent population. which is why allopatric speciation is easy. After allo- There is little evidence for Mayr’s hypothesis, which patric populations have diverged sufficiently, they may is considered implausible by some theoreticians, expand their ranges and overlap without interbreeding. but new theoretical developments suggest that it still Conversely, sympatric speciation is generally con- warrants consideration, albeit in modified form sidered difficult because divergent selection, unaided (Gavrilets, 2004). by an extrinsic barrier, must overcome recombination. If some genetic incompatibility has evolved Several models have been proposed by which this may between allopatric populations, and if they subse- occur, but the conditions and assumptions required quently expand their range, mate, and form hybrids for sympatric speciation are fairly stringent (Gavrilets, with low fitness (i.e., low survival or reproduction),

14 Douglas J. Futuyma individuals that mate conspecifically will have more repeatedly challenged, and it is certainly not possible to successful offspring that those that hybridize, and affirm that mutations of large effect, that radically alleles that enhance (reinforce) mating discrimination alter developmental pathways, have never contributed may be transmitted and increase in frequency. Such to evolution. Some evolutionary changes have been reinforcement of prezygotic isolation is the major con- discontinuous, such as neoteny in salamanders that text in which there is direct selection for RI. It appears retain larval morphology into reproductive adulthood, to be a fairly common component of speciation in and which is based on one or a few gene substitutions; some taxa. but this is an abbreviation of a developmental pathway The divergence of populations into distinct species that requires the action of many genes for its continu- often includes ecological differentiation, often initiated ation. Mutations in key regulatory genes, such as in allopatry due to environmental differences among the Hox genes that are important in establishing the regions, and sometimes enhanced via character dis- fundamental body plans of animals, can have drastic placement, the evolution of accentuated ecological effects, sometimes switching development of one body difference between sympatric populations of two region into another; but whether or not comparable species, to reduce competition. Bursts of speciation, mutations in these genes were the origin of major accompanied by ecological divergence, are referred to evolutionary changes in body plan is not known. as adaptive radiations, macroevolutionary episodes The complex developmental pathways that these genes that account for much of the extraordinary adaptive now control may have evolved incrementally. variety of organisms (Schluter, 2000). Classical evolutionary and embryological studies of morphological differences among taxa identified several common patterns, such as (a) allometric differ- FROM MICROEVOLUTION ences: evolved differences in the rate at which one TO MACROEVOLUTION structure or dimension grows compared to other struc- tures (compare limb proportions in humans and apes); Macroevolution, or evolution above the species level (b) heterochrony: differences in the timing of develop- (Levinton, 2001), includes the evolution of “higher mental events, including initiation or cessation of taxa,” which, most systematists today insist, should a structure’s growth (as in neotenic salamanders); be monophyletic groups of species that are recognized (c) heterotopy: differences in the location on the body by shared derived characters (synapomorphies), and where a feature develop (e.g., bone in the skin of arma- which usually are phenotypically quite different from dillos); (d) individualization of repeated elements (e.g, related higher taxa. At least among extant animals, for heterodont dentition of mammals compared to homo- example, mammals differ substantially in skeletal and dont dentition of most “reptiles”); (e) “standardiza- other features from other amniotes. During the 1930s tion,” or restriction of variation (e.g., digit number and 1940s, the major authors of the Evolutionary was higher and more variable in the earliest tetrapods Synthesis provided both theory and evidence that than in their pentadactylous descendants). A major the evolution of the distinctive features of higher taxa challenge is to understand how differences in genes consisted simply of incremental changes in each of are translated into such phenotypic changes, via their the differentiating characters, attributable to the pro- effects on the processes by which such characters cesses, outlined above, that operate within and during develop. Increasingly, this challenge is being met in the formation of species (Simpson, 1953). The many the field of evolutionary developmental biology (EDB, distinctive characters of mammals, for example, are or “evo-devo”), which is based on increased under- a product of mosaic evolution: largely independent standing of how certain genes regulate the time and evolution, often at different rates, of many distinct place of expression of other genes, often in hierarchical features. Each such feature, it was postulated, evolves sequences of control (Carroll, 2007). For example, by successive small steps rather than by large discrete changes in the expression pattern of certain Hox genes jumps. Such gradual evolution has been paleonto- along the embryonic anterior-posterior axis in turn logically documented for many skeletal features of control activation of other genes that govern the mammals, and for many other lineages. Often, evolu- form of vertebrae. In snakes, most precaudal vertebrae tion of a character is accelerated, and takes surprising have characteristic thoracic form, rather than sacral directions, because a feature may serve a new and and other vertebral forms, due to changes in the different function. domains of expression of certain Hox genes. Moreover, This neo-Darwinian view affirmed Darwin’s hypo- new regulatory connections between genes have often thesis that evolution is a gradual process – although, to evolved. For instance, although the canonical role be sure, it can sometimes be quite rapid, which makes of Hox genes is specification of anterior-posterior finding intermediate stages in an already very incom- domains of differentiation, a Hox locus also controls plete fossil record even less likely. Gradualism has been the development of bristles on the legs of Drosophila.

Evolutionary Theory 15 Major challenges lie ahead in understanding how advances in developmental biology to understand the changes in gene regulatory pathways evolve, and the evolution of phenotypes, and many other concerns. extent to which understanding the genetic basis of New approaches are being developed to answer old development can predict possible constraints on the but difficult questions, such as the causes of speci- kind of phenotypic variation that can arise and be ation, the evolution of functionally integrated charac- available to natural selection. Moreover, development teristics, and the conditions under which populations is often responsive to environmental signals, so a geno- adapt to environmental change or become extinct – one type often expresses phenotypic plasticity, the ability to of the most important questions now, when humans develop a variety of adaptive phenotypic states (its are changing the Earth at a frightening pace. norm of reaction) under different environmental con- ditions (Pigliucci, 2001). Conversely, canalization, or developmental buffering, can evolve: the capacity of a DISCUSSION POINTS genotype to produce a consistent phenotype despite potentially destabilization by environmental variation 1. Although evolution consists of genetic change, we or gene mutations. Adaptive phenotypic plasticity and often do not have direct evidence that differences canalization are under genetic control, but how they among individuals, populations, or related species evolve is little understood. in a feature of interest have a genetic foundation. Are there ways in which we can have more or less confidence that interesting variation represents EVOLUTIONARY THEORY TODAY evolved differences rather than direct environmen- tal effects on the phenotype? When Darwin referred to “my theory,” he was speaking 2. What kinds of evidence might we use to judge of a little-tested hypothesis. Since then, all the major whether a character of interest, or a difference elements of his theory – the common ancestry of all between populations or species, is an adaptation organisms, the bifurcation of lineages in the origin of rather than the consequence of genetic drift? species, the evolution of adaptations by the action of 3. In what ways is evolution a matter of chance natural selection on hereditary variation, the diversifi- (random) versus a nonrandom process? cation of species by adaptation to different ecological 4. How do functionally complex characters, such as niches, or places in the “economy of nature” – have the vertebrate eye, evolve? Cast your answer both been abundantly supported, elaborated, and extended. in historical, phenotypic terms and in terms of When biologists speak today of “evolutionary theory,” genetics and selection. they refer not to a speculation or hypothesis, but to 5. In what ways may DNA sequence data inform us a mature scientific theory, an accepted complex of about the history of differentiation among human general principles that explain a wide variety of natural populations and the causes of the differences phenomena, as quantum theory and atomic theory do among them? in the physical sciences. 6. Taking into account principles of evolutionary No biologist today would think of publishing “new genetics and possible scenarios for environmental evidence for evolution” – it hasn’t been a scientific change, do you expect adaptive evolutionary issue for a century. However, although the major prin- changes to occur in human populations in the next ciples of evolutionary theory have been increasingly few hundred years? If so, discuss what changes may supported despite, and indeed because of, vast changes occur, and how. If not, why? in biological knowledge, many questions remain unanswered and substantial controversy persists, as it does in all active sciences, about some important ques- REFERENCES tions such as gradual versus discontinuous major changes in phenotype. Today, research in evolutionary Andersson, M. B. (1994). Sexual Selection. Princeton: Prince- ton University Press. biology is flourishing as never before. Because of Arnqvist, G. and Rowe, L. (2005). Sexual Conflict. Princeton: increasingly powerful and affordable molecular and Princeton University Press. computing technologies, documentation of newfound Barton, N. H. and Keightley, P. D. (2002). Understanding phenomena that call for evolutionary explanation (e.g., quantitative genetic variation. Nature Reviews Genetics, in genomics), and the development of new theory 3, 11–21. addressing both new and old questions, evolutionary Blows, M. W. and Hoffmann, A. A. (2005). A reassessment biology is now occupied with the phylogeny of all of genetic limits to evolutionary change. Ecology, 86, major groups of organisms, discerning and explaining 1371–1384. patterns of evolution of genes and genomes, explaining Bulmer, M. G. (1994). Theoretical Evolutionary Ecology. puzzling behaviors and other phenotypic traits, using Sunderland, MA: Sinauer.

16 Douglas J. Futuyma Carroll, S. B. (2007). The Making of the Fittest: DNA and Kimura, M. (1983). The Neutral Theory of Molecular Evolu- the Ultimate Forensic Record of Evolution. New York: tion. Cambridge: Cambridge University Press. W. W. Norton. Levinton, J. S. (2001). Genetics, Paleontology, and Macroevo- Clutton-Brock, T. H. (1991). The Evolution of Parental Care. lution, 2nd edn. Cambridge: Cambridge University Press. Princeton: Princeton University Press. Maynard Smith, J. (1982). Evolution and the Theory of Coyne, J. A. and Orr, H. A. (2004). Speciation. Sunderland, Games. Cambridge: Cambridge University Press. MA: Sinauer. Mayr, E. (1954). Change of genetic environment and evolu- Dawkins, R. (1982). The Extended Phenotype. Oxford: Oxford tion. In Evolution as a Process, J. Huxley, A. C. Hardy, and University Press. E. B. Ford (eds). London: Allen and Unwin, pp. 157–180. Endler, J. A. (1986). Natural Selection in the Wild. Princeton: Nei, M. and Kumar, S. (2000). Molecular Evolution and Princeton University Press. Phylogenetics. New York: Oxford University Press. Falconer, D. S. and Mackay, T. F. C. (1996). Introduction to Nowak, M. A. (2006). Evolutionary Dynamics: Exploring the Quantitative Genetics. Harlow, Essex: Longmans. Equations of Life. Cambridge, MA: Harvard University Futuyma, D. J. (1987). On the role of species in anagenesis. Press. American Naturalist, 130, 465–473. Okasha, S. (2006). Evolution and the Levels of Selection. Futuyma, D. J. (1998). Evolutionary Biology, 3rd edn. Oxford: Oxford University Press. Sunderland, MA: Sinauer. Pigliucci, M. (2001). Phenotypic Plasticity: Beyond Nature Futuyma, D. J. (2005). Evolution. Sunderland, MA: Sinauer. and Nurture. Baltimore: Johns Hopkins University Press. Gavrilets, S. (2004). Fitness Landscapes and the Origin of Schluter, D. (2000). The Ecology of Adaptive Radiation. Species. Princeton: Princeton University Press. Oxford: Oxford University Press. Gould, S. J. (2002). The Structure of Evolutionary Theory. Simpson, G. G. (1953). The Major Features of Evolution. Cambridge, MA: Harvard University Press. New York: Columbia University Press. Hartl, D. L. and Clark, A. G. (1997). Principles of Population Sober, E. and Wilson, D. S. (1998). Unto Others: the Evolu- Genetics. Sunderland, MA: Sinauer. tion and Psychology of Unselfish Behavior. Cambridge, MA: Haig, D. (1993). Genetic conflicts in human pregnancy. Harvard University Press. Quarterly Review of Biology, 68, 495–532. Stearns, S. C. (1992). The Evolution of Life Histories. Oxford: Haig, D. (1997). Parental antagonism, relatedness asymmet- Oxford University Press. ries, and genomic imprinting. Proceedings of the Royal Trivers, R. (1974). Parent–offspring conflict. American Society of London. Series B, 264, 1657–1662. Zoologist, 11, 249–264. Hein, J., Schierup, M. H. and Wulff, C. (2005). Gene Geneal- Williams, G. C. (1992). Natural Selection: Domains, Levels, ogies, Variation and Evolution. A Primer in Coalescent and Challenges. New York: Oxford University Press. Theory. Oxford: Oxford University Press. Wilson, D. S. (1983). The group selection controversy: Houle, D. (1998). How should we explain variation in the history and current status. Annual Review of Ecology and genetic variance of traits? Genetica, 102/103, 241–253. Systematics, 14, 159–187.

2 The Study of Human Adaptation A. Roberto Frisancho INTRODUCTION systems of organs, to entire organisms. For example, the lungs provide oxygen to the extracellular fluid to conti- Ever since hominids left Africa, they have expanded nually replenish the oxygen that is being used by the cells, throughout the world and have adapted to diverse the kidneys maintain constant ion concentrations, and environments, and acquired specific biological and the gastrointestinal system provides nutrients. cultural traits that have enabled them to survive in a Humans living in hot or cold climates must undergo given area. The conceptual framework of research in additional functional adjustments to maintain thermal biological anthropology is that evolutionary selection balance; these may comprise adjustments to the rate processes have produced the human species and that of metabolism, avenues of heat loss, heat conservation, these processes have produced a set of genetic charac- respiration, blood circulation, fluid and electrolyte teristics, which adapted our evolving species to their transport, and exchange. In the same manner, persons environment. Current investigations have demonstrated exposed to high altitudes must adjust through physio- that the phenotype measured morphologically, physio- logical, chemical, and morphological mechanisms, such logically, or biochemically is the product of genetic as increase in ventilation, increase in the oxygen-carrying plasticity operating during development. Within this capacity of the blood resulting from an increased concen- framework, it is assumed that some of the biological tration of red blood cells, and increased ability of adjustments or adaptations people made to their natural tissues to utilize oxygen at low pressures. Failure to and social environments have also modified how they activate the functional adaptive processes may result in adjusted to subsequent environments. The adjustments failure to restore homeostasis; which in turn results in we have made to improve our adaptations to a given the maladaptation of the organism and eventual incap- environment have produced a new environment to acitation of the individual. Therefore homeostasis is a which we, in turn, adapt in an ongoing process of new part and function of survival. The continued existence stress and new adaptation. of a biological system implies that the system possesses mechanisms that enable it to maintain its identity, despite the endless pressures of environmental stresses HOMEOSTASIS AND ENVIRONMENTAL (Proser, 1964). The complementary concepts of homeo- STRESS stasis and adaptation are valid at all levels of biological organization; they apply to social groups as well as to Central to the study of adaptation is the concept of unicellular or multicellular organisms (Proser, 1964). homeostasis and environmental stress. Environmental Homeostasis is a function of a dynamic interaction stress is defined as any condition that disturbs the normal of feedback mechanisms whereby a given stimulus functioning of the organism. Such interference eventu- elicits a response aimed at restoring the original equi- ally causes a disturbance of internal homeostasis. librium. Several mathematical models of homeostasis Homeostasis means the ability of the organism to main- have been proposed. In general, they show (as schema- tain a stable internal environment despite diverse, disrup- tized in Figure 2.1) that when a primary stress disturbs tive, external environmental influence (Proser, 1964). the homeostasis that exists between the organism and On a functional level, all adaptive responses of the the environment the organism must resort either organism or the individual are made to restore internal to biological or cultural-technological responses in homeostasis. These controls operate in a hierarchy at order to function normally. For example, when faced all levels of biological organization, from a single with heat stress, the organism may simply reduce its biochemical pathway, to the mitochondria of a cell, to metabolic activity so all heat-producing processes cells organized into tissues, tissues into organs and are slowed down, or may increase the activity of the Human Evolutionary Biology, ed. Michael P. Muehlenbein. Published by Cambridge University Press. # Cambridge University Press 2010. 17

18 A. Roberto Frisancho Lifetime: habituation Primary acclimation stress acclimatization Biological responses Restores Disturbs homeostasis homeostasis Growth: between between developmental organism organism acclimatization and and environment environment Cultural– Decrease technological environmental responses stress 2.1. Schematization of adaptation process and mechanisms that enable individual or population to maintain homeostasis in the face of primary environmental disturbing stress. From Frisancho (1993). heat-loss mechanisms. In either case, the organism processes that enable them to function and to be adapted may maintain homeostasis, but the physiological to their environment. processes will occur at a different set point. The attain- Not all responses made by the organism can be ment of full homeostasis or full functional adaptation, considered adaptive. Although a given response might depending on the nature of the stress, may require not be adaptive per se, through its effect on another short-term responses, such as those acquired during structure or function it may prove beneficial to the acclimation or acclimatization, or may require expos- organism’s function. Conversely, a given adaptive ure during the period of growth and development as in response may aid the organism in one function, but developmental acclimatization. actually have negative effects on other functions or In theory, the respective contributions of genetic and structures. Thus, within all areas of human endeavor environmental factors vary with the developmental stage a given trait is considered adaptive when its beneficial of the organism–the earlier the stage, the greater the effects outweigh the negative ones. In theory, this is a influence of the environment and the greater the valid assumption, but in practice, due to the relative plasticity of the organism (Proser, 1964; Timiras, 1972; nature of adaptation, it is quite difficult to determine Frisancho, 1975, 1993). However, as will be shown, the the true adaptive value of a given response. Every principle does not apply to all biological parameters; response must be considered in the context of the it depends on the nature of the stress, the developmental environmental conditions in which the response was stage of the organism, the type of organism, and the measured and within the perspective of the length of particular functional process that is affected. For time of the study and the subject population. example, an adult individual exposed to high-altitude hypoxia through prolonged residence may attain a level of adaptation that permits normal functioning in all ADAPTIVE PROCESSES daily activities and as such we may consider him adapted. However, when exposed to stress that requires The term adaptation is used in the broad generic sense increased energy, such as strenuous exercise, this of functional adaptation, and it is applied to all levels of individual may not prove to be fully adapted. On the other biological organization from individuals to populations. hand, through cultural and technological adaptation, A basic premise of this approach is that adaptation is a humans may actually modify and thus decrease the process whereby the organism has attained a beneficial nature of the environmental stresses so that a new adjustment to the environment (Mayr, 1963; Lewontin, microenvironment is created to which the organism 1957; Proser, 1964; Dubos, 1965; Baker, 1966; Lasker, does not need to make any physiological responses. 1969; Mazess, 1973; Frisancho, 1975, 1993). This adjust- For example, cultural and technological responses permit ment can be either temporary or permanent, acquired humans to live under extreme conditions of cold stress either through short-term or lifetime processes, and may with the result that some of the physiological processes involve physiological, structural, behavioral, or cultural are not altered. However, on rare occasions, humans changes aimed at improving the organism’s functional have been able to completely avoid an environmental performance in the face of environmental stresses. stress. Witness the fact that the Eskimos, despite their If environmental stresses are conducive to differential advanced technological adaptation to cold in their mortality and fertility, then adaptive changes may everyday hunting activities, are exposed to periods of cold become established in the population through changes stress and in response have developed biological in genetic composition and thus attain a level of

The Study of Human Adaptation 19 genetic adaptation. In this context, functional adapta- ACCLIMATIZATION tion, along with cultural and genetic adaptation, is viewed as part of a continuum in an adaptive process Acclimatization refers to changes occurring within that enables individuals and populations to maintain the lifetime of an organism that reduce the strain both internal and external environmental homeostasis. caused by stressful changes in the natural climate or Therefore the concept of adaptation is applicable to all by complex environmental stresses (Eagan, 1963; levels of biological organization from unicellular organ- Bligh and Johnson, 1973; Folk, 1974). If the adaptive isms to the largest mammals and from individuals traits are acquired during the growth period of to populations. This broad use of the concept of adapta- the organism, the process is referred to as either tion is justified not only in theory but also because it developmental adaptation or developmental acclima- is currently applied to all areas of human endeavor tization (Timiras, 1972; Frisancho, 1975, 1993). Stud- so that no discipline can claim priority or exclusivity in ies on acclimatization are done with reference to both the use of the term (Dubos, 1965). Functional adaptation major environmental stresses and several related involves changes in organ system function, histology, secondary stresses. For example, any difference in morphology, biochemical composition, anatomical the physiological and structural characteristics of relationships, and body composition; either independ- subjects prior to and after residence in a tropical ently or integrated in the organism as a whole. These environment is interpreted as a result of acclimatiza- changes can occur through acclimation, habituation, tion to heat stress. In addition, because tropical acclimatization or genetic adaptation. climates are also associated with nutritional and disease stresses, individual or population differences in function and structure may also be related to these ACCLIMATION factors. On the other hand, in studies of acclimation any possible differences are easily attributed to the Acclimation refers to the adaptive biological changes major stress to which the experimental subject has that occur in response to a single experimentally been exposed in the laboratory. For understanding induced stress (Eagan, 1963; Folk, 1974) rather than to the basic physiological processes of adaptation, stud- multiple stresses as occurring in acclimatization. As with ies on acclimation are certainly better than those of acclimatization, changes occurring during the process acclimatization. However, since all organisms are of growth may also be referred to as developmental never exposed to a single stress, but instead to acclimation (Timiras, 1972; Frisancho, 1975, 1993). multiple stresses, a more realistic approach is that of studying acclimatization responses. Thus, studies on both acclimation and acclimatization are essential for HABITUATION understanding the processes whereby the organism adapts to a given environmental condition. This Habituation implies a gradual reduction of responses rationale becomes even more important when the to, or perception of, repeated stimulation (Eagan, aim is to understand the mechanisms whereby 1963; Folk, 1974). By extension, habituation refers to humans adapt to a given climatic area, since humans the diminution of normal neural responses, for in a given area are not only exposed to diverse stresses example, the decrease of sensations such as pain. Such but have also modified the nature and intensity of changes can be generalized for the whole organism these stresses, as well as created new stresses for (general habituation) or can be specific for a given part themselves and for generations to come. of the organism (specific habituation). Habituation necessarily depends on learning and conditioning; which enable the organism to transfer an existing DEVELOPMENTAL ACCLIMATIZATION response to a new stimulus. A common confusion is that habituation can lead to adaptation. However, the The concept of developmental acclimatization (also extent to which these physiological responses are referred to as developmental adaptation) is based important in maintaining homeostasis depends on upon the fact that the organism’s plasticity and suscep- the severity of environmental stress. For example, with tibility to environmental influence is inversely related severe cold stress or low oxygen availability, failure to to developmental states of the organism, so that the respond physiologically may endanger the well-being younger the individual the greater is the influence and survival of the organism. Likewise, getting used to of the environment and the greater the organism’s tolerating high levels of noise implies ignoring the plasticity (Frisancho, 1975, 1993; Frisancho and stress, which eventually can lead to deafness. In other Schechter, 1997). Hence, variability in physiological words, habituation is a process that in the long run traits can be traced to the developmental history of produces negative side effects. the individual.

20 A. Roberto Frisancho ACCOMMODATION AND ADAPTATION is an important mechanism that facilitates human biological adaptation (Thomas, 1975; Rappaport, 1976; The term accommodation is used to describe responses Moran, 1979). It may be said that cultural adaptation to environmental stresses that are not wholly success- during both contemporary times and in an evolutionary ful because, even though they favor survival of the perspective, represents humanity’s most important tool. individual, they also result in significant losses in some It is through cultural adaptation that humans have important functions (Waterlow, 1990). For example, been able to survive and colonize far into the zones subjects when exposed to a low intake of leucine for of extreme environmental conditions. Humans have three weeks can achieve body leucine balance at the adapted to cold environments by inventing fire and expense of reducing protein synthesis and protein clothing, building houses, and harnessing new sources turnover (Young and Marchini, 1990). Since low of energy. The construction of houses, use of clothing in protein synthesis and protein turnover diminishes the diverse climates, certain behavioral patterns, and work individual’s capacity to successfully withstand major habits, represent biological and cultural adaptations to stresses, such as infectious diseases (Frenk, 1986), climatic stress. The development of medicine, from its under conditions of low-dietary protein intake achiev- primitive manifestations to its high levels in the present ing body leucine balance represents only a temporary era, and the increase of energy production associated accommodation, which in the long run is not adaptive. with agricultural and industrial revolutions are represen- In other words, accommodation is a stopgap that tative of human cultural adaptation to the physical ultimately produces negative side effects. environment. Culture and technology have facilitated biological adaptation, yet they have also created, and continue to INDIVIDUALS VERSUS POPULATIONS create, new stressful conditions that require new adap- tive responses. A modification of one environmental Whatever the method employed, geographical or experi- condition may result in the change of another, and mental research in human adaptation is concerned with such a change may eventually result in the creation of populations, not with individuals; although the research a new stressful condition. Advances in the medical itself is based on individuals. There are two related sciences have successfully reduced infant and adult reasons for this. mortality to the extent that the world population is The first is a practical consideration. Studying all growing at an explosive rate, and unless world food members of a given population, unless its size is small resources are increased, the twenty-first century will enough, is too difficult to be attempted by any research witness a world famine. Western technology, although team. Therefore, according to the objectives of the upgrading living standards, has also created a polluted investigation, the research centers on a sample that is environment that may become unfit for good health considered representative of the entire population. and life. If this process continues unchecked, Based on these studies, the researchers present a environmental pollution will eventually become picture of the population as a whole, with respect to another selective force to which humans must adapt the problem being investigated. through biological or cultural processes, or else face The second reason is a theoretical one. In the study extinction. Likewise, cultural and technological adap- of adaptation, we usually focus on populations rather tation has resulted in the rapid increase of energy than on individuals because it is the population that availability and has decreased energy expenditure; survives and perpetuates itself. In the investigation causing a disproportionate increase in the develop- of biological evolution, the relevant population is the ment of degenerative diseases associated with meta- breeding population because it is a vehicle for the bolic syndrome. This mismatch between biology and gene pool, which is the means for change and hence lifestyle threatens our survival as human species evolution. The study of an individual phenomenon (Eaton et al., 2002). Therefore, adaptation to the world is only a means to understand the process. The adapta- of today may be incompatible with survival in the tion of any individual or individuals merely reflects the world of tomorrow unless humans learn to adjust their adaptation that has been achieved by the population of cultural and biological capacities. which he is a member. CULTURAL AND TECHNOLOGICAL GENETIC ADAPTATION AND ADAPTABILITY ADAPTATION Genetic adaptation refers to specific heritable charac- Cultural adaptation refers to the nonbiological responses teristics that favor tolerance and survival of an of the individual or population to modify or ameliorate individual or a population in a particular total envir- an environmental stress. As such, cultural adaptation onment. A given biological trait is considered genetic

The Study of Human Adaptation 21 when it is unique to the individual or population and Environmental influences when it can be shown that it is acquired through during growth and development biological inheritance. A genetic adaptation becomes have a high influence and established through the action of natural selection elicit a high morphological (Neel, 1962; Livingstone, 1958; Neel et al., 1998; Mayr, response 1997). Natural selection refers to the mechanisms whereby the genotypes of those individuals showing the greatest adaptation or “fitness” (leaving the most Morphological Genetic Genetic descendents through reduced mortality and increased phenotypic trait mediation fertility) will be perpetuated, and those less adapted to response the environment will contribute fewer genes to the population gene pool. Natural selection favors the features of an organism that bring it into a more Environmental influences efficient relationship with its environment. Those during adulthood gene combinations fostering the best-adapted pheno- have a low influence and elicit a low morphological types will be “selected for,” and inferior genotypes will response be eliminated. The selective forces for humans, as for other 2.2. Schematization of interaction of genetic and environment mammals, include the sum total of factors in the nat- and the phenotypic morphological outcome. Morphological and ural environment. All the natural conditions, such as physiological diversity reflects the responses and adaptations that the organism makes to particular environment during and hot and cold climates and oxygen-poor environments, development. From Frisancho (1993). are potential selective forces. Food is a selective force by its own abundance, eliminating those susceptible to obesity and cardiac failures, or by its very scarcity, and individual flexibility. Extinct populations are those favoring smaller size and slower growth. In the same which were unable to meet the challenges of new manner, disease is a powerful selective agent, favoring conditions. Thus, contemporary fitness requires both in each generation those with better immunity. The genetic uniformity and genetic variability. natural world is full of forces that make some individ- Contemporary adaptation of human beings is both uals, and by inference some populations, better the result of their past and their present adaptability adapted than others because no two individuals or (Lasker, 1969; Frisancho, 1993). It is this capacity to populations have the same capacity of adaptation. adapt that enables them to be in a dynamic equilibrium The maladapted population will tend to have lower in their biological niche. It is the nature of the living fertility and/or higher mortality than that of the organism to be part of an ecosystem whereby it modifies adapted population. the environment and, in turn, is also affected by The capacity for adaptation (adaptability) to envir- such modification. The maintenance of this dynamic onmental stress varies between populations and even equilibrium represents homeostasis; which, in essence, between individuals. The fitness of an individual or reflects the ability to survive in varying environments population is determined by its total adaptation to the (Dubos, 1965; Proser, 1964). The ecosystem is the environment – genetic, physiological, and behavioral fundamental biological entity – the living individual (or cultural). Fitness, in genetic terms, includes more satisfying its needs in a dynamic relation to its habitat. than just the ability to survive and reproduce in a given In Darwinian terms, the ecosystem is the setting for environment; it must include the capacity for future the struggle for existence, efficiency and survival are survival in future environments. The long-range fitness the measures of fitness, and natural selection is the of a population depends on its genetic stability and process underlying all products (Proser, 1964). variability. The greater the adaptation, the longer the In general, the morphological and functional features individual or population will survive, and the greater reflect the adaptability or capacity of the organism to the advantage in leaving progeny resembling the respond and adapt to a particular environment parents. In a fixed environment, all characteristics (Figure 2.2). The effect and responses to a given environ- could be under rigid genetic control with maximum mentalconditionaredirectlyrelatedtothedevelopmental adaptation to the environment. On the other hand, in stage of organism; so that the younger the age, the a changing environment a certain amount of variabil- greater the effect and the greater the flexibility to respond ity is necessary to ensure that the population will and adapt. Conversely, the later the age and, especially survive environmental change. This requirement for during adulthood, the effect of the environment is less variability can be fulfilled either genetically or pheno- likely to be permanent and the capacity to respond and typically or both. In most populations a compromise adapt is also diminished when compared to a developing exists between the production of a variety of genotypes organism.

22 A. Roberto Frisancho MULTILEVEL SELECTION AND EVOLUTION single defector (Traulsen and Nowawk, 2006). Hence, OF CO-OPERATION this simple condition ensures that selection favors co-operators and opposes defectors. Ever since Darwin (1871) indicated that the competi- The concept of group selection has been a major tion between groups can lead to selection of co-opera- tenet of behavioral ecology. The major premise of tive behavior, the concept of multilevel selection has behavioral evolutionary ecology is that genetic and been developed. He stated that, “there can be no doubt behavioral traits are two distinct expressions of a that a tribe including many members who were single evolutionary process. (Trivers, 1971; Cronk, always ready to give aid to each other and to sacrifice 1991; Strier, 2000; Silk, 2001). In behavioral ecology, themselves for the common good, would be victorious behaviors are treated like any other biological trait over other tribes; and this would be natural selection” and are potentially subject to natural selection. The (Darwin, 1871, p. 166). Over many years, Wilson and processes involved in behavioral evolution are colleagues have been the main proponent of the idea of equivalent to those in genetic evolution: natural selec- group selection (Wilson, 1975; Sober and Wilson, 1998; tion influences the frequency of a trait transmitted Traulsen and Nowak, 2006). It is assumed that group from parent to offspring through differential fertility selection is an important organizing principle that and mortality. In the evolutionary perspective, bio- permeates evolutionary processes from the emergence logical structures have been custom tailored to motiv- of the first cells to development of nations. According ate behaviors that are likely to enhance individual to multilevel selection, groups consist of genetically fitness. Therefore, behavioral variants with a high unrelated individuals, and successful groups attract fitness have been favored and these perpetuate the new individuals, which learn the strategies of others evolutionary origin of fitness-enhancing biological in the same group. A population can be subdivided traits. It follows then, that the behavioral traits that into groups, and the individuals interact with other enhance fitness also accentuate biological fitness. members of the group, and depending on their repro- In other words, a change occurring in one system ductive fitness, individuals can lead to larger groups entails a change in the fitness governing evolution in that split more often. In other words, higher-level or the other system. Therefore, both genetic and behav- group selection emerges as a by-product of individual ioral selection tend to favor those existing variants reproduction. whose net effect is to increase the average fitness A fundamental condition for the success of the of the individual and population to the prevailing group, therefore, must be co-operation among indi- conditions. Studies of primates indicate that they viduals, and thus group selection favors co-operative use a diversity of behaviors that increase the likeli- altruistic behavior and opposes defectors. The fitness hood of gaining access to mates and guarantee the of an individual, and the group at large, also depends survival of their offspring; which, in turn, insures the on the altruistic behavior of nonrelatives. When an passing of their traits to the next generation. In this altruist gives an alarm call, it benefits not only his or context, behavioral actions that lead to a higher her relatives, but also other unrelated members of reproductive success will become adaptive and the thegroupbecauseaprimatetroopdoesnotonly genes associated with such behavior will be trans- include relatives. Thus, altruism can be selected ferred to the next generation faster than those that if these nonrelated individuals can be counted on to are less adaptive. Therefore, the differences in fitness reciprocate the favor when the need arises. between individuals and populations will determine A recipient of an altruistic behavior who fails to recip- the behavioral pattern of a given primate group. rocate is a cheater. Studies of nonhuman primates In other words, a specific behavioral strategy that indicate that a cheater may gain in the short run by contributes to the survival and reproductive fitness receiving aid without any costs to their own fitness of the individual – and eventually the population – (Strier, 2000). However, reciprocity is necessary becomes part of the genetic milieu of the species. for future support in the long run because the In summary, co-operation and altruism evolve by cheater’s fitness is lower when compared to the group selection or multilevel selection. Human behav- individual who reciprocated. In view of the fact that ioral ecology rests upon a foundation of evolutionary primates constantly need to protect themselves from theory, which include sexual selection, whereby indi- neighboring communities and predators, one can viduals within one sex secure mates and produce assume that reciprocal altruism must have been offspring at the expense of other individuals within selected for because it enhanced their fitness, not the same sex, which can cause changes in gene because the animals are conscious of their motives frequency across generations that are driven at least or the reproductive consequences of their behavior. in part by interactions between related individuals Mathematical models indicate that a single co- referred to as kin selection, and be expressed as the operator has a greater fixation probability than a sum of an individual’s own reproductive success.

The Study of Human Adaptation 23 CURRENT DIRECTIONS IN ADAPTATION mellitus was a quick insulin trigger. Insulin’s main func- RESEARCH tion is to assist in the homeostasis of glucose in the blood. Specifically, when blood glucose levels are too In the 1970s I postulated the hypothesis of develop- high, the pancreas releases insulin to increase tissue mental adaptation to explain the enlarged lung volume uptake of glucose, thus reducing blood glucose levels. and enhanced aerobic capacity that characterize the Conversely, when blood glucose levels are low, the Andean high altitude natives. According to the organism secretes glucagon and growth hormone, which developmental adaptation hypothesis, “adult biological in turn, induce the release of stored glucose and fatty traits are the result of the effects of the environment acids into the blood stream raising serum glucose levels. and the physiological responses that the organism The insulin response is to activate an uptake of glucose makes during the developmental state” (Frisancho, into the muscle cells for storage, and in liver cells it 1970, 1975, 1977). This concept is based upon the fact influences the conversion of glucose to fatty acids for that the organism’s plasticity and susceptibility to envir- storage in fat (adipose) tissue. This response was an onmental influence is inversely related to developmen- asset during times of abundance because it would allow tal states of the organism, so that the younger the an individual to build-up energy reserves more quickly individual the greater is the influence of the environ- and thus better survive times of food scarcity. Under ment and the greater the organism’s plasticity these conditions, the thrifty gene was selected to regulate (Frisancho, 1975, 1977, 1993). Hence, variability in efficient intake and utilization of fuel stores. In other physiological traits can be traced to the developmental words, during periods of food shortage and famine, history of the individual (Figure 2.2). Currently this those with the thrifty genotype would have a selective concept has been applied to explain the variability in advantage because they relied on larger, previously adult behavioral traits such as in learning and crime and stored energy to maintain homeostasis; whereas those delinquency (Yueh-Au Chien, 1994; Sroufe et al., 2005; without “thrifty” genotypes would be at a disadvantage Kruger et al., 2008), in sensory inputs and auditory and less likely to survive and reproduce. However, under spatial processing (Martin and Martin, 2002), in toler- modern conditions of abundant food and sedentary ance to surgical intervention (Faury et al., 2003), in lifestyle, this genotype becomes perversely disadvanta- variability in oxygen consumption and mitochondrial geous. With a constant abundance of food, insulin levels membrane potential in energy metabolism of rat remain high, resulting in tissues becoming less sensitive cortical neurons (Schuchmann et al., 2005), and in to the effects of insulin. This reduced sensitivity to the variability in increased risk of adult obesity and cardio- effects of insulin results in chronically elevated blood vascular problems associated with the metabolic glucose levels type II diabetes and related chronic health syndrome (Barker, 1994). A common denominator of problems (e.g., obesity). all these studies is that humans and many other A test of the genetic predisposition to type II organisms are conditioned by experiences during devel- diabetes involved a comparative study of the Pima opment and as developmental experiences is an import- Indians of southern Arizona and the Pima Indians of ant contributor to variability in adult phenotypic the Sierra Madre mountains of northern Mexico behavioral and biological traits. (Knowler et al., 1990; Price et al., 1992). These two In this section I will summarize the evidence groups, which were separated 700 1000 years ago, supporting the applicability of the concept of develop- differ in their life style. The Arizona Pima live under mental adaptation to account for the origins of the high conditions of access to a high fat, highly refined diet risk of the adult metabolic syndrome incorporating and low energy expenditure. In contrast, the Mexican information derived from thrifty gene, thrifty pheno- Pima still pursue a much more traditional lifestyle type, and epigenetics. The evolution of the metabolic and have a diet based on the occasional intake of syndrome is also discussed at length in Chapter 30 of lamb and poultry, but mainly on beans, corn, and this volume. potatoes, grown by traditional, and physically very energy-demanding, techniques. These two groups differ significantly in the frequency of obesity and DEVELOPMENTAL ADAPTATION AND THE diabetes. The Arizona Pima adults have a body mass 2 THRIFTY GENOTYPE index (BMI) of 33.4 kg/m ; compared to a BMI of 24.9 2 kg/m in the Mexican Pima (Ravussin et al., 1994). Neel and colleagues (Neel, 1962; Neel et al., 1998) Likewise, in the Arizona Pima 37% of men and 54% attempted to explain the epidemic proportions of of women were diabetic, while in the Mexican Pima diabetes in Native American populations, such as the only 2 of 19 women and 1 of 16 men were diabetic Pima Native Americans, by postulating the existence of (Knowler et al., 1990; Price et al., 1992). In other a “thrifty gene” that increased the risk of type II diabetes. words, although the Mexican Pima share the “thrifty According to this hypothesis, the basic defect in diabetes gene” with Arizona Pima, their increased frequency of

24 A. Roberto Frisancho obesity and diabetes is more evidence that an abun- Poor maternal nutrition dance of fatty foods and modern sedentary lifestyles associated with fetal are the real culprits. Thus, it is not the presence of a undernutrition “thrifty gene” alone that results in increased rates of diabetes, but rather the interaction with modern diet- Fetal programing: Cellular, ary and lifestyle conditions the results in increased physiological, and metabolic rates of the chronic health problems. compensatory responses In summary, the thrifty genotype hypothesis has resulting in energy conservation been used to explain the epidemic levels of obesity and diabetes among non-Western populations, such Adult poor Adult good as South Pacific Islanders, sub-Saharan Africans, postnatal postnatal nutrition and high nutrition and low Native Americans in the southwestern United States, energy energy Inuit, Australian aborigines, etc. (Eaton et al., 1988; expenditure expenditure O’Dea, 1991); all of whom were newly introduced to industrialized diets and environments. The fact that the frequency of type II diabetes has recently increased Nondiabetic Type II diabetic and good health and polor health among Europeans that were not subjected to periodic famines cannot be attributed to the action of a 2.3. The thrifty phenotype. The risk of type II diabetes and so-called “thrifty” gene. metabolic syndrome in adulthood is associated with prenatal undernutrition resulting in efficient physiological adaptation that becomes detrimental when food is abundant and energy expenditure is low. DEVELOPMENTAL ADAPTATION AND THE THRIFTY PHENOTYPE the Dutch famine of World War II were found to have Recently, Barker and colleagues (Barker, 2007; Hales impaired glucose tolerance and increased adiposity and Barker, 1992) have reported an inverse relationship in adulthood (Stein et al., 1975, 2007). between birthweight and the risk of hypertension, cardiovascular disease, and type II diabetes in adulthood when the individual is well nourished postnatally. DEVELOPMENTAL ADAPTATION AND To account for these observations, Barker and colleagues EPIGENETICS proposed that adverse effects in utero induce cellular, physiological, and metabolic compensatory responses, Epigenetics refers to the transmission of phenotypic such as insulin resistance, high blood plasma levels of traits from one generation to the next that do not fatty acids, which result in energy conservation and depend on differences in DNA sequence (Waddington, reduced somatic growth that enable the fetus to survive 1952; Jablonka, 2004; Holliday, 2006). During the last undernutrition. This response is referred to as the thrifty two decades, there has been an accumulation of obser- phenotype hypothesis (Armitage et al., 2005). These vations indicating that the expression of DNA traits can responses that were adaptive under poor prenatal condi- be affected by environmental factors acting during tions become a problem if food becomes abundant. development. Specifically, experimental studies In this view, thrifty physiological mechanisms are showed that identical twin mice differ in the color of adaptive in nutritionally poor environments, but in rich fur; one has brown fur and will grow up to be lean and environments are maladaptive. That is, what was positive healthy, while the other has yellow fur and becomes under reduced availability of nutrients, particularly obese and prone to cardiovascular disease. The differ- during periods of rapid development, becomes negative ent phenotypes are due to the addition of a methyl in rich environments because it facilitates nutrient group (-CH 3 ); which is referred to as methylation. absorption and hence increases the risk of adult obesity and the suite of risk factors for cardiovascular disease Methylation known as the metabolic syndrome (Figure 2.3). In summary, it appears that nutrition and other Methylation refers to the altering of the genetic envir- environmental factors during prenatal and early post- onment through the addition of a methyl group (-CH 3 ) natal development influence cellular plasticity; thereby to the fifth position of cytosine, which is largely altering susceptibility to adult cardiovascular disease, confined to CpG dinucleotides. This addition, by modi- type II diabetes, obesity, and other chronic diseases fying the CpG islands, prevents signaling molecules referred as the adult metabolic syndrome. This hypoth- from reaching the promoter site to turn the gene on esis is supported by the finding that the offspring of and prevent the expression of the dark coat color. women who were starved and became pregnant during In other words, the additional methyl group attaches

The Study of Human Adaptation 25 to and shuts off the gene that controls dark fur color the organism; so that the younger the individual, the and allows the yellow color to be expressed. Thus, the greater the epigenetic marks, including CpG methyla- process of methylation works as a kind epigenome that tion. Despite the great interest in molecular genetics, dictates which genes in the genome are turned on and there is scant incontrovertible evidence indicating which are not. This process can differ even between epigenetic effects in humans. Considering society’s identical twins. increased concern about environmental pollutants, this Recently, experimental studies indicate that area of research should a good direction for human bisphenol A (BPA) can alter gene expression and biologists. affect adult phenotype by modifying CpG methylation at critical epigenetically labile genomic regions (Water- land and Jirtle, 2004). Bisphenol A is used in the produc- OVERVIEW tion of polycarbonate and plastic containers and in the organism acts like the body’s own hormones. Thus, The term adaptation encompasses the physiological, there is concern that long-term exposure to BPA may cultural, and genetic adaptations that permit individ- induce chronic toxicity in humans (vom Saal and uals and populations to adjust to the environment in Hughes, 2005). Fortunately, the effects of methylation which they live. These adjustments are complex, and are not permanent but reversible, as shown by the fact the concept of adaptation cannot be reduced to a simple that the yellow agouti (Avy), whose diet was supple- rigid definition without oversimplification. The func- mented with folic acid, vitamin B 12 , choline, betaine, tional approach in using the adaptation concept permits and zinc, counteracted the DNA methylation and its application to all levels of biological organization changed coat color from yellow to dark brown coat from unicellular to multicellular organisms, from early (Dolinoy and Jirtle, 2008), which is associated with a embryonictoadultstages,andfromindividualstopopu- low risk of cardiovascular disease. lations. In this context, human biological responses to environmental stress can be considered as part of a continuous process whereby past adaptations are modi- Transgenerational epigenetic effects fied and developed to permit the organism to function It has been suggested that the epigenetic modifications and maintain equilibrium within the environment to brought about by parental conditions may be which it is daily exposed. expressed even in grandchildren. Extensive records of The mechanisms for attaining full functional adap- a population in Overkalix cohorts, northern Sweden, tation include acclimation, acclimatization, habitu- found that an association between grandparental ation, and accommodation. The role played by each prepubertal slow growth periods (SGP) or rapid of these processes depends on the nature of the stress growth periods (RGP), and parental periods of low or or stresses, the organ system involved, and the devel- high food availability, with grandchildren’s mortality opmental stage of the organism. It is emphasized that and disease risk (Kaati et al., 2007). If the SGP of the the goal of the organism’s responses to a given stress is grandparent was a period of high food availability, to maintain homeostasis within an acceptable normal then the male grandchild had reduced longevity but range with itself and with respect to other organisms an increased mortality. The extent to which these and the environment (as schematized in Figure 2.1). associations represent multigenerational epigenetic Such adaptations are usually reversible, but the revers- effects is unwarranted, in part because ruling out ibility depends on the developmental stage of the genetic and societal confounders, and in the absence organism at which the adaptive response occurs and of molecular evidence, is extremely difficult. Hence, the nature of the environmental stress. This character- future research must be focused on long-term transge- istic allows organisms to adapt to a wide range of nerational studies whereby many birth cohorts are environmental conditions. Furthermore, an adaptation studied using intensive prenatal and perinatal genotyp- is always a compromise between positive and negative ing across generations. Only then can variability in the effects. Every adaptation involves a cost. The process expression of phenotypic traits can be attributed to of adaptation is always positively beneficial; without epigenetic changes. which the organism would be worse off, however the In summary, epigenetic effects exist that are not organism has to pay a price for the benefit. The benefit necessarily adaptive, and in many of these cases, the derived from a given response depends on the circum- inherited phenotype is actually detrimental to the organ- stances and the conditions where it occurs. As recently ism. Environmental exposure to nutritional, chemical, pointed out (Young and Marchini, 1990), every adap- and physical factors can alter gene expression and affect tation involves a choice. For example, a man has adult phenotype: a process known as epigenetics. In all 6 hours in which to walk 11 km. If he walks slowly, of these studies, the extent of DNA methylation depends he saves energy expenditure, and therefore it may be on and is inversely related to the developmental state of adaptive if the energy resources are limited; however

26 A. Roberto Frisancho he has no time left to do anything else. On the other 7. Compare developmental and adult adaptation. hand, if he walks fast, he saves time at the cost of using Which is more likely to be reversible and why? more energy. Thus, the adaptive importance of given 8. Discuss the applicability of the concept of develop- type of response depends on the conditions. mental adaptation to the hypothesis thrifty geno- The concept of developmental adaptation has type and thrifty phenotype that account for the become a major focus for studying the origins of human increased frequency of the adult metabolic syn- diversity (Figure 2.2). The applicability of this research drome among native and nonnative populations. strategy is based upon the premise that human biological 9. Discuss the relationship of the concept of develop- responses to environmental stress represent a continu- mental adaptation to the field of epigenetics. ous process whereby past adaptations are modified and developed to permit the organism to function and main- REFERENCES tain equilibrium within the environment to which it is daily exposed. From the studies of the thrifty genotype Armitage, J. A., Taylor, P. D. and Poston, L. (2005). Experi- and thrifty phenotype, and their relationship to the mental models of developmental programming: conse- etiology of metabolic syndrome, it is evident that what quences of exposure to an energy rich diet during was positive under reduced availability of nutrients, development. Journal of Physiology, 565, 3–8. particularly during periods of rapid prenatal develop- Baker, P.T. (1966). Human biological variation as an ment, becomes negative in rich environments. Research adaptive response to the environment. Eugenics Quarterly, 13,81–91. in epigenetics may provide the bridge between the thrifty Barker, D. J. P. (1994). Mothers, Babies, and Disease in genotype and thrifty phenotype to unravel the interrela- Later Life. London: BMJ Publishing. tions of how the impact of early diet helps how the Barker, D. J. P, (2007). The Origins of the Developmental organism adapts to a given environmental condition that Origins Theory. Journal of Internal Medicine, 261, differs in nutritional resources; resulting in the diverse 412–417. phenotypic expression of physiology, body size, and Bligh, J. and Johnson, K. G. (1973). Glossary of terms for health risk of contemporary and past populations. thermal physiology. Journal of Applied Physiology, 35, The study of individuals exposed to stressful conditions 941 961. in natural and laboratory environments is one of Cronk, L. (1991). Human behavioral ecology. Annual Review the most important approaches for understanding the of Anthropology, 20, 25–53. mechanisms whereby human populations adapt to a Darwin, C. (1871). The Descent of Man and Selection in given environment. Knowledge of human adaptation is Relation to Sex, 2nd edn. London: Murray. basic in our endeavors to understand past and present Dolinoy, D. C. and Jirtle, R. L. (2008). Environmental epige- human variation in morphology and physiological nomics in human health and disease. Environmental and performance. The insights we have gained during the Molecular Mutagenesis, 49, 4–8. Dubos, R. (1965). Man Adapting. New Haven, CT: Yale last decade have stimulated new approaches to study University Press. human adaptation, not only for understanding behavior, Eagan, C. J. (1963). Introduction and terminology. Federation but of how ecology shapes function both in the present Proceedings, 22, 930–932. and in the past, and for understanding variability in the Eaton, S. B., Konner, M. and Shostak, M. (1988). Stone agers immune system, thermo-regulation, coping with limited in the fast lane: chronic degenerative diseases in evolu- and excessive amounts of foods, low oxygen pressure, tionary perspective. American Journal of Medicine, 84, and high and low solar radiation. 739–749. Eaton, S. B., Strassman, B. I., Nesse, R. M., et al. (2002). Evolutionary health promotion. Preventative Medicine, DISCUSSION POINTS 34, 109–118. Faury, G., Pezet, M., Knutsen, R.H., et al. (2003). Develop- 1. What are the four processesoffunctional adaptation? mental adaptation of the mouse cardiovascular system to 2. Why does the organism have to respond to environ- elastin haplo insufficiency. Journal of Clinical Investigation, mental stimuli? Give examples. 112, 1419–1428. Folk, G. E. Jr (1974). Textbook of Environmental Physiology. 3. When does a given functional adaptation become a Philadelphia: Lea and Febiger. genetic adaptation? Frenk, S. (1986). Metabolic adaptation in protein-energy 4. Discuss the role of multiple-level selection in the malnutrition. Journal of the American College of Nutrition, expression of biological and behavioral traits. 5, 371–381. 5. Compare accommodation and adaptation. Give Frisancho, A. R. (1970). Developmental responses to high specific examples. altitude hypoxia. American Journal of Physical Anthropology, 6. Discuss how technological and cultural adapta- 32: 401–407. tions have created new environmental stresses. Frisancho, A.R. (1975). Functional adaptation to high altitude Give examples. hypoxia. Science, 187, 313–319.

The Study of Human Adaptation 27 Frisancho, A. R. (1977). Developmental adaptation to high Transactions of the Royal Society of London. Series B, altitude hypoxia, International Journal of Biometeorology, 334, 233–240. 21, 135–146. Price, R. A., Lunetta, K., Ness, R., et al. (1992). Obesity in Frisancho, A. R. (1993). Human Adaptation and Accommo- Pima Indians. Distribution characteristics and possible dation. Ann Arbor, MI: University of Michigan Press. thresholds for genetic studies. International Journal of Frisancho, A. R. and Schechter, D. E. (1997). Adaptation. Obesity and Related Metabolic Disorders, 16, 851–857. In History of Physical Anthropology: an Encyclopedia, Proser, C. L. (1964). Perspectives of adaptation: theoretical vol. 1. Garland, TX: Garland Press, pp. 6–12. aspects. In Handbook of Physiology, vol. 4, D. B. Dill, E. F. Hales, C. N. and Barker, D. J. (1992). Type 2 (non-insulin- Adolph and C. G. Wilber (eds). Washington, DC: American dependent) diabetes mellitus: the thrifty phenotype Physiological Society, pp. 11–26. hypothesis. Diabetologia, 35, 595–601. Rappaport, R. A. (1976). Maladaptation in social systems. Holliday, R. (2006). Epigenetics: a historical overview. In Evolution in Social Systems, J. Friedman and Epigenetics, 1, 76–80. M. Rowlands (eds). London: Gerald Duckworth, pp. 49–71. Jablonka, E. (2004). Epigenetic epidemiology. International Ravussin, E., Valencia, M. E., Esparza, J., et al. (1994). Journal of Epidemiology, 33, 929–935. Effects of a traditional lifestyle on obesity in Pima Indians. Kaati, G., Bygren, L. O., Pembrey, M., et al. (2007). Transge- Diabetes Care, 17, 1067–1074. nerational response to nutrition, early life circumstances Schuchmann, S., Buchheim, K., Heinemann, U., et al. (2005). and longevity. European Journal of Human Genetics, Oxygen consumption and mitochondrial membrane poten- 15, 784–790. tial indicate developmental adaptation in energy meta- Knowler, W. C., Pettitt, D. J., Saad, M. F., et al. (1990). Dia- bolism of rat cortical neurons. European Journal of betes mellitus in the Pima Indians: Incident, risk factors Neuroscience, 21, 2721–2732. and pathogenesis. Diabetes/Metabolism Research and Silk, J. (2001). Ties that bind: the role of kinship in primate Reviews, 6, 1–27. societies. In New Directions in Anthropological Kinship, Kruger, D. J., Reisch, T., and Zimmerman, M. A. (2008).Time L. Stone (ed.). Lanham, MD: Rowman and Littlefield perspective as a mechanism for functional developmental Publishers, pp. 71–92. adaptation. Journal of Social, Evolutionary, and Cultural Sober, E. and Wilson, D. S. (1998). Unto Others: the Evolu- Psychology, 2, 1–22. tion and Psychology of Unselfish Behavior. Cambridge, Lasker, G.W. (1969). Human biological adaptability. Science, MA: Harvard University Press. 166, 1480–1486. Sroufe, L. A., Egeland, B., Carlson, E., et al. (2005). The Lewontin, R. C. (1957). The adaptations of populations to Development of the Person: the Minnesota Study of Risk varying environments. In Cold Spring Harbor Symposia and Adaptation from Birth to Adulthood. New York: on Quantitative Biology, vol.22. Boston, MA: Cold Spring Guilford Press. Harbor Laboratory of Quantitative Biology. Stein, Z., Susser, M., Saenger, G., et al. (1975). Famine and Livingstone, F. B. (1958). Anthropological implications of Human Development. The Dutch Hunger Winter of 1944 45. sickle-cell gene distribution in West Africa. American New York: Oxford University Press. Anthropologist, 60, 533–562. Stein, A. D., Kahn, H. S., Rundle, A., et al. (2007). Anthropo- Martin, P. and Martin, M. (2002). Proximal and distal metric measures in middle age after exposure to famine influences on development: The model of developmental during gestation: Evidence from the Dutch famine. Ameri- adaptation. Developmental Review, 22, 78–96. can Journal of Clinical Nutrition, 85, 869–876. Mayr, E. (1963). Animal Species and Evolution. Cambridge, Strier, K. B. (2000). Primate Behavioral Ecology. Boston: MA: The Belknap Press of Harvard University Press. Allyn and Bacon. Mayr, E. (1997). The objects of selection. Proceedings of the Thomas, R.B. (1975). The ecology of work. In Physiological National Academy of Sciences of the United States of America, Anthropology, A. Damon (ed.). New York: Oxford University 94, 2091 2094. Press, pp. 59–79. Mazess, R. B. (1973). Biological adaptation: aptitudes and Timiras, P. S. (1972). Developmental Physiology and Aging. acclimatization. In Biosocial Interrelations in Population New York: Macmillan. Adaptation, E. S. Watts, F. E. Johnson and G. W. Lasker Traulsen, A. and Nowak, M. A. (2006). Evolution of cooper- (eds). The Hague: Mouton, pp. 9–18. ation by multilevel selection. Proceedings of the National Moran, E.F. (1979). Human Adaptability: an Introduction to Academy of Sciences of the United States of America, Ecological Anthropology. North Scituate, MA: Duxbury Press. 103, 10952–10955. Neel, J. V. (1962). Diabetes mellitus: A “thrifty” genotype Trivers, R. L. (1971). The evolution of reciprocal altruism. rendered detrimental by “progress”? American Journal of Quarterly Review of Biology, 46, 35–57. Human Genetics, 14, 353–362. Vom Saal, F. S. and Hughes, C. (2005). An extensive new Neel, J. V., Weder, A. B. and Julius, S. (1998). Type II dia- literature concerning low-dose effects of bisphenol betes, essential hypertension, and obesity as “syndromes A shows the need for a new risk assessment. Environmen- of impaired genetic homeostasis”: the “thrifty genotype” tal Health Perspectives, 113, 926–933. hypothesis enters the twenty-first century. Perspectives in Waddington, C. H. (1952). Epigenetics of Birds. Cambridge: Biology and Medicine, 42, 44–74. Cambridge University Press. O’Dea, K. (1991). Traditional diet and food preferences of Waterland, R. and Jirtle, R. (2004). Early nutrition, epigen- Australian aboriginal hunter-gatherers. Philosophical etic changes at transposons and imprinted genes, and

28 A. Roberto Frisancho enhanced susceptibility to adult chronic diseases. Nutrition, Young, V. R. and Marchini, J. S. (1990). Mechanisms and 20, 63–68. nutritional significance of metabolic responses to altered Waterlow, J. C. (1990). Nutritional adaptation in man: intakes of protein and amino acids, with reference to General introduction and concepts. American Journal of nutritional adaptation in humans. American Journal of Clinical Nutrition, 51, 259–263. Clinical Nutrition, 51, 270–289. Wilson, D. S. (1975). A theory of group selection. Proceedings Yueh-Au Chien, C. (1994). Developmental Adaptation of of the National Academy of Sciences of the United States of Social Learning Theory: the Etiology of Crime and Delin- America, 72, 143–146. quency. Denver, CO: University of Colorado Press.

3 History of the Study of Human Biology Michael A. Little INTRODUCTION biology will be traced through the contributions of key authors, their ideas, and the role of key institutions The field of human biology is broad based but with its in its history. principal origins in studies of variation in living popu- lations within physical or biological anthropology. Today, human biology incorporates a majority of sci- RACE AND TYPOLOGY entists trained in anthropology, but also counts among its members scientists trained in other specialties of The tangled roots of interests in human biology and the human sciences. During the late nineteenth and human variation lie deep in antiquity. In the early early twentieth centuries, our stem science – physical modern era, Europeans became acutely aware of human anthropology – was oriented toward skeletal studies, population variation during the Age of Exploration and gross anatomy, and human variation as represented by the opening of new regions of the world in the fifteenth, race. Skeletal studies included those of both living and sixteenth, and seventeenth centuries (Eiseley, 1958). prehistoric populations. Anatomically oriented studies Newly discovered populations from the Americas, of the skeleton and of the living were focused on struc- Africa, the Pacific, and Asia, were classified, along with ture and origins, but with less interest in function and Europeans, into broad racial groups according to phys- evolutionary causality. In the his Manual of Physical ical characteristics. Such racially based physical charac- Anthropology, Juan Comas (1960) presented an excel- teristics were erroneously thought to be tightly bound to lent overview of the history of physical anthropology mental, emotional, intellectual, and cultural attributes, from its earliest origins, including the origins of and as well. Some races were identified as clearly inferior to connections with natural history, racial classification, others on a typological scale – an extension of The Scale craniometry, prehistory and paleoanthopology, and of Being – from primitive to advanced. Ideas of fixed, evolution. In later chapters, he reviewed the histories unchanging racial categories were tightly bound to of growth studies, somatology, constitutional typology, Judeo-Christian religious beliefs associated with, among craniology, osteology, and racial classification. other things, the creation or origin of humans. Some The beginning of an integrated human biology in believed that humans originated from a single creation, the United States dates back to the late 1920s when monogenism, where the more primitive races had then Darwin’s ideas of population variation, adaptation, degenerated from the original people living in the selection, and evolution began to be reconceptualized Garden of Eden. Another belief was in polygenism,where by a number of leading scientists. At that time, evolu- God had created several independent peoples, some who tionary theory, and concepts from behavioral sciences, were elevated to civilization and others doomed at demography, genetics, and child growth began to be the outset to perpetual savagery. In both of these beliefs consolidated into a science of human biology. In later of human creation, Western, civilized populations decades, body composition, physiology, nutrition, and were held, with a few exceptions, to be superior to non- ecology were added to the mix leading to an even Western races from Asia, Africa, the New World, and the greater understanding of the variations in, the evolu- Pacific. tion of, and the characteristics of humans in all of their Within Western societies, the two sexes were con- complex biobehavioral states. sidered unequal – men were believed to be superior to A history of anything is a reflection of traditions, women, by virtue of their very nature, and women were connections, and innovations. These can be found as often viewed as incapable of rational thought (Stocking, well in a history of human biology. In the narrative that 1987). Women were more emotional than men and tied follows, the development of the science of human more closely to nature through their role in reproduction Human Evolutionary Biology, ed. Michael P. Muehlenbein. Published by Cambridge University Press. # Cambridge University Press 2010. 29

30 Michael A. Little and childbirth. The upper socioeconomic classes were and the first chapter by Mielke et al. (2006) provide believed to be innately superior to the lower classes, histories of these controversies about race. Fundamen- because class differences were thought to have a heredi- tally, all classifications of this nature are essentialist in tary basis. Hence, aristocratic males (“good breeding”) character, that is, they focus on fixed types that obscure were alleged to be at the pinnacle of God’s creations. the variation that contributes to human population diver- There were some who questioned the dominance of sity around the globe. It was the rejection of typological nature over nurture, however. For example, in the last thinking (fixed species and races) that enabled Darwin to chapter of the popularly known Voyage of the Beagle conceptualize ideas of variation, adaptation, and natural (Darwin, 1839), a thoughtful Charles Darwin noted, in selection (Mayr, 1972), and it was this same transition in the context of slavery: “.. . if the misery of the poor be thinking about human populations that moved human caused not by the laws of nature, but by our institutions, biologists toward a truly scientific approach in this field. great is our sin . . .” (Darwin, 1839, p. 433). Questions about the contributions of nature (heredity) and nurture (environment) to human diversity persist and are FRANZ BOAS AND HIS CONTRIBUTIONS debated up to the present. Eighteenth-century racial classifications were Although the late nineteenth and early twentieth established by Carolus Linnaeus (1707–1778), Georges centuries were characterized by beliefs in fixed racial Louis Leclerc, the Comte de Buffon (1707–1788), types, accompanied by beliefs in the superiority of Georges Cuvier (1769–1832), and, of course, Johann some racial groups over others (racism or racialism), Friedrich Blumenbach (1752–1840) (Gould, 1996; there was a contrary opinion in the form of Franz Molnar, 2002). Classifications ranged from Linnaeus’s Boas (1858–1942). Boas, often referred to as the Asian, European, African, and American groups to founder of American anthropology and the four-field Blumenbach’s Caucasian, Mongolian, Ethiopian, approach, made remarkable contributions to dispel- American, and Malayan. ling the myth of fixed or pure races and the importance Twentieth-century classifications of human popula- of the environment in structuring the character of tions have been devised in numerous combinations human populations (Figure 3.1). These interests in and have been based on a variety of criteria. Earnest human plasticity in the context of race were almost A. Hooton (1931), the Harvard professor who trained certainly based on his research on child and adolescent most of the modern generation of physical anthropolo- growth and development (Tanner, 1959, 1981). Hence, gists before World War II, identified “composite races,” he can be identified also as one of the founders in addition to the “primary races and subraces.” of human biology in the United States because of The composite races were formed by admixture of his contributions to: (1) debunking the idea of fixed primary races. Carleton Coon was one of Hooton’s early races; (2) establishing a migration research design that students who had written a book classifying The Races of Europe (Coon, 1939). Several years later, Coon et al. (1950), in a mid-twentieth century approach, presented a six-fold geographical classification and then divided these major races into thirty subpopulations, whose char- acteristics were based on external physical and skull attri- butes. In that same year, Boyd (1950) identified six races according to their blood group genetics. About a decade later, Garn (1961), who was a coauthor of the Coon et al. (1950) book, refined these classifications and identified geographical, local, and micro-races, in a hierarchy of populations and subpopulations. These three works were different from previous efforts at classification in that they attempted to apply contemporary evolutionary, genetic, and ecological principles to the identification of racial (population)variation around the world.They were transitional in the sense that they applied modern theory to an outdated typological system in which bound- aries between populations were fixed. There has been considerable controversy in scientific writings over the intervening years about the utility and social impact of racial classification and the validity of the concept itself. 3.1. Franz Boas (1858–1942) in 1906. Photograph with permis- Works by Shipman (1994), Marks (1995), Brace (2005) sion from the American Philosophical Society.

History of the Study of Human Biology 31 continues to be used up to the present; (3) incorpor- acknowledged the plasticity in the immigration study ating the social and material environment as influen- when he stated: “These changes do not obliterate the cing human biology (plasticity); and (4) making differences between genetic types [characters] but they numerous discoveries about the patterns of growth in show that the type as we see it contains elements that are children and adolescents. not genetic but an expression of the influence of the The migrant study, designed to test the idea of races environment” (Boas, 1936, p. 523). as static types, was initiated in 1908 with modest funding Boas growth studies date back to 1888 when from the US Immigration Commission. The pilot study he took an academic position at Clark University in began in June 1908 with studies of height and cephalic Worcester, Massachusetts. His first important discov- index of Eastern European Jewish boys in New York ery was based on data gathered by the growth studies City Schools (Stocking, 1974). The results of the pilot pioneer and Harvard professor, Henry Bowditch study and the more extensive study of Bohemians, (Tanner, 1959). Boas discovered that the asymmetrical Sicilians, Neapolitans, Polish, Hungarians, and Scots distributions in Bowditch’s data on height during established differences between those born to parents adolescence could be explained by the individual before and those born to parents after they migrated to variations in growth rates, which he called “tempo of the United States (Tanner, 1959). Head form or cephalic growth.” This is the first indication of Boas’s sensitivity index differed between the two groups, refuting the idea to the importance of longitudinal growth data in of fixed races and demonstrating the influence of the uncovering subtle changes in growth during childhood environment on human variation in physical characteris- and adolescence. In 1891, he initiated a longitudinal tics. The completed study, published by Boas (1912), was survey of Worcester school children that confirmed his accompanied by the publication several years later of the sense of the value of longitudinal growth data. Between raw data of more than 18 000 subjects who were meas- 1892 and 1941, Boas published numerous papers on ured in the migration study (Boas, 1928). The reanalysis growth in Science, Human Biology, and other journals of these data stimulated a new controversy (Sparks and in which he made other significant discoveries, each a Jantz, 2002; Gravlee et al., 2003) over Boas’s analyses and reflection of his remarkable knowledge of statistics and interpretations and whether he really demonstrated his genius. Over the years, he: (1) produced the first plasticity in these migrant populations. National Growth Standards or norms; (2) introduced Relethford (2004), in exploring this controversy, the concept of physiological or developmental age and identified three ways that craniometric variation can observations that males were behind females as early change over time: (1) developmental plasticity through as five years of age; (3) observed that working class and environmental change; (2) long-term changes through poor children from large sibships tend to be smaller on natural selection; and (3) within-group and among- average than those from small sibships; (4) established groups variation by gene flow. Each of these has been relationships between “age of peak velocity” during ado- shown to operate, but what Relethford (2004) noted was lescence and other measures of size during adolescence that the debate about Boas’s study centered on the rela- and adulthood; and (5) observed that children from the tive importance of these three causes of craniometric Horace Mann School of Columbia University had change. Based on the two major studies, the question become larger between 1909 and 1935 (now known as that Relethford raised, “.. . is whether developmental the secular trend in growth) (Tanner, 1959, 1981). plasticity has a significant effect on craniometric vari- ation” (Relethford (2004, p. 380). Relethford approached this question by inspecting the changes in craniometric PRE-WORLD WAR II AND HUMAN variation among the groups Boas studied. Three of the POPULATION BIOLOGY seven European ethnic groups showed no statistical dif- ference between US-born and European-born migrants In addition to Boas’s contributions to a developing field (Hungarians, Polish, and Scots). The remaining four of human biology through his growth studies, there are groups did show statistically significant differences, but several other lines of continuity from pre-World War II these were relatively slight differences. Based on this, to the present. In 1929, Raymond Pearl (1879–1940) Relethford (2004) suggested that these data do demon- founded the journal Human Biology and served as editor strate developmental plasticity, but this plasticity does until his death in 1940 (Crawford, 2004). Pearl was an not obscure the underlying genetic differences that accomplished population biologist who was Professor separate the ethnic groups. In other words, both genetic of Biometry and Vital Statistics in the School of Hygiene characters and developmental plasticity contribute to and Public at Health at Johns Hopkins University and the variation, but the genetic contribution to the total who had broad interests in genetics, fertility, evolution, variation, in this case, is the stronger of the two. Releth- nutrition, disease, duration of life, senescence, and ford’s (2004) conclusion was anticipated by Boas in a physical anthropology (Figure 3.2). According to Kings- paper published years after the original, where Boas land (1984, p. 8), “Pearl considered himself to be first

32 Michael A. Little was editor in the 1930s. It is probably the case that Ales ˇ Hrdlic ˇka, one of the founders and first president of the AAPA, and Pearl did not share the same scientific philosophy because of Hrdlic ˇka’s strong dislike and avoidance of statistics and Pearl’s commitment to their use (Lasker, 1989). This added an additional dimension to the differences between the AJPA and HB,since Hrdlic ˇka was editor of the former and Pearl was editor of the latter. Perhaps the most extensive pre-war research in human biology was that of human growth conducted at several centers in the United States. Some of these longitudinal growth studies (sequential measurements over time of the same individuals) were the Fels Longi- tudinal Study (Yellow Springs Ohio), the Bolton-Brush Study at Western Reserve University (Cleveland, Ohio), the Berkeley Growth Study at the Institute of Human Development, University of California (Berkeley), the Child Research Council Study at the University of Colorado (Denver), and the Harvard School of Public 3.2. Raymond Pearl (1879–1940) sometime in the late 1920s or Health Growth Study (Boston) (Roche, 1992, pp. 1–2). early 1930s. Photograph with permission from the American Each of these was founded from the 1920s to the early Philosophical Society. 1930s principally to determine the effects of the Great Depression and to assess the means to help impover- and foremost a human biologist”; that is, he was totally ished children. As Tanner noted: these longitudinal committed to a human population biology that was studies were part of “. . . a powerful child welfare move- quite akin to some of the holistic and integrated science ment [that] arose in the 1920s and provided the soil for practiced today. a crop of longitudinal studies, whose harvesting The journal, Human Biology,publishedavarietyof shaped the whole pattern of human auxology in the papers in, among other topics, genetics, growth, health, years 1935–1955” (Tanner (1981, p. 299). Here he uses human population studies, mathematical and statistical the term auxology, which refers to the study of human modeling, and human evolution. Goldstein (1940), in a physical growth, which he (James Tanner) pioneered survey of the first decade of Human Biology (HB)andtwo in the 1950s and 1960s in the United Kingdom decades (1920s and 1930s) of the American Journal of (Figure 3.3). It is not clear to what extent Boas’s work Physical Anthropology (AJPA), suggested that the journal HB published articles more in the realm of “biological anthropology” whereas the AJPA was more prone to publish in “anatomical anthropology.” These patterns of topical publication imply that the earliest development of a professional human biology “identity” began about this time. As Goldstein (1940) reported, there were strong connections between Pearl and his journal and physical anthropology – connections with the journal that have continued to the present. Many early papers in Human Biology were published on topics related to physical anthropology (Lasker, 1989). Raymond Pearl and Franz Boas were acquainted at that time because Pearl was active in a number of professional societies in which Boas was a member, including the American Association of Physical Anthropologists (AAPA). In fact Pearl was well known among physical anthropologists of this time as evidenced by his election as the third president of the AAPA from 1934 to 1936. Moreover, both Pearl and Boas were members of the National Academy of Sciences, both were sophisticated biostatisticians, and Boas published 3.3. James M. Tanner (1920–) at a conference in 1982. Photo- several papers on growth in the journal HB while Pearl graph courtesy of Barbara Garn.

History of the Study of Human Biology 33 stimulated these longitudinal growth studies, but himself in studies of body composition. The study began principle investigators must have known the value of in 1944 with 36 conscientious objectors who volunteered longitudinal series from Boas work before the turn for the full-year study that consisted of an equilibration of the twentieth century and later papers in the 1930s. period on a normal diet, 3 months of semi-starvation For example, when Frank Shuttleworth (1889–1958) (1600 kcal/day), and a period of refeeding. In addition, analyzed the Harvard Growth Study data, he employed the volunteers were expected to walk 22 miles per week some of the same procedures that Boas employed in to increase their energy expenditure. The study was analyzing data of adolescent girls (Tanner, 1981, p. 310). intensive, with almost daily biochemical, physiological, Boas’s influence was also seen in students from psychological, and body composition measurements Harvard and elsewhere who used his migration model taken. The result of the study were published in two in their own research during the 1930s and 1940s massive volumes that stand as state-of-the-art research (Little and Leslie, 1993). For example, Harry Shapiro even today, particularly since it probably would not be and Frederick Hulse (1939) studied Japanese migrants permitted to conduct such a high-risk project during to Hawai’i, Marcus Goldstein (1943) measured Mexican- present times (Keys et al., 1950). Americans, and Gabriel Lasker (1946) studied immigrant The Rochester Desert unit was charged with the and American-born Chinese. The migration-research research task of determining water and food require- design was also used in a great deal of human biology ments, sweat rates, energy balance, heat tolerance, research up to the end of the twentieth century. work capacity, and the probability of survival while living under the hot-dry conditions of desert environ- ments. Much of the desert research was conducted on THE WORLD WAR II YEARS military personnel on maneuvers in southern Califor- nia. Other studies were conducted of men on life rafts The enormity of World War II brought a halt to a great without water, and some limited tests were conducted deal of academic research in the United States and under hot-wet conditions in Florida to simulate troops elsewhere, but stimulated research that could be directly in tropical forests. In the preface to the volume that applied to the war effort. Some of this research carried reported this research, the authors identify this work over to the post-war years and not only led to additional as a part of the “. . . recently developing field of environ- academic research, but even contributed to changes in mental physiology” (Adolph and Associates, 1947, thinking about the causes of human variation. Much of p. vii). What was not known then was that this work the wartime research centered on the maintenance and was to stand as the first major work in climatic stress health of US troops in stressful environments and ameli- physiology, and one that would serve as the basis for orating the stress of warfare and hunger in the civilian heat-stress studies both in environmental physiology populations in Europe and Asia. During the war, US and human biology. military personnel were stationed and fighting in In addition to the pioneering research in nutritional Europe, East Asia, Southeast Asia, North Africa, the and environmental physiology represented by these Pacific, and Alaska and the Aleutian Islands. Hence, they two wartime studies, a relatively new area of human were expected to be able to work strenuously at high variation was being developed that would be explored levels while at the same time being exposed to climatic by anthropologists and human biologists interested in variation that ranged from Arctic and temperate zone growth, environmental stress, exercise physiology, and winter cold to tropical and temperate summer heat. nutrition – body composition. These early studies intro- In addition to stresses from dietary restriction and star- duced hard-tissue anthropologists (bone and teeth) to vation and from climatic extremes, there was interest in the importance of soft tissue (muscle and adipose tissue) the interaction between diet and climate, particularly in human function, human growth, and in studies of in the context of providing enough food calories for human adaptation to the environment. troops fighting in cold or hot geographic zones (Mitchell and Edman, 1951). Two of the most prominent wartime studies will be surveyed briefly: the semi-starvation POST-WAR UNITED STATES AND study conducted at the University of Minnesota and the UNITED KINGDOM desert physiology study conducted by the Rochester Desert Unit whose research was conducted at several There were parallel transformations of physical anthro- places in the United States. pology in the years directly following World War II in The Minnesota semi-starvation study was initiated by both the United States and in the United Kingdom. Ancel Keys (1904–2004), a stress physiologist best known In the United States, Sherwood L. Washburn (1911– for having developed K-rations during the early part of 2000) successfully promoted a scientific agenda of prob- the war, and several colleagues including Josef Broz ˇek lem solving, application of evolutionary theory, and (1913–2004), a psychologist who later distinguished understanding of human variation rather than racial

34 Michael A. Little the 15th Cold Spring Harbor Symposium on Quantitative Biology (Warren, 1951) and was initiated by the Cold Spring Harbor Institute Director, Milislav Demerec (1895–1966). Washburn’s co-organizer of the Sympo- sium, called “Origin and Evolution of Man,” was the distinguished geneticist, Theodosius Dobzhansky (1900–1975), whom he had met at Columbia University. Although the conference was successful in bringing together geneticists, evolutionists, and anthropologists, there were some differences among the more traditional physical anthropologists, particularly the typological “constitutionalists” such as William Sheldon and Earnest Hooton on the one hand, and the more for- ward-looking physical anthropologists and geneticists on the other concerning topics of race and population variation. “Constitutional somatology” is an outmoded and typological area of study of body form and associated behavior that has a long history (see Comas, 1960, 319 ff.) and was promoted by Sheldon (Sheldon et al., 1940) 3.4. Sherwood L. Washburn (1911–2000) at the 1952 Wenner- and Hooton (1939) at the Cold Spring Harbor meeting. Gren Foundation International Symposium of Anthropology. Many years later, the daughter of the Institute Director, Photograph courtesy of the Wenner-Gren Foundation for Rada Dyson-Hudson, who had attended the Symposium Anthropological Research. as a young Swarthmore College student, remarked that after the afternoon sessions, the geneticists would typology and simply descriptive anthropometric survey gather to discuss the talks given that day, while the (Washburn, 1951). Washburn, who had been trained at anthropologists would head for the nearest local tavern! Harvard, among many others, under Earnest A. Hooton Despite these differences, new ground had been broken, (1887–1954), rejected many of the traditional ideas of his and Washburn’s “scientific and evolutionary” physical mentor and began an active program to bring his new anthropology was gaining momentum. ideas to other anthropologists, especially younger ones In the United Kingdom, the transformation also (Figure 3.4). With support from a private anthropological involved experimental scientific approaches to investi- foundation in 1946, he organized the Viking Fund gating human variation and the application of evolu- Summer Seminars in Physical Anthropology. These tionary theory as a basis for explaining human Summer Seminars were dynamic “state-of-the-art” meet- variation. The United Kingdom’s transformation ings in which many new ideas in physical anthropology moved in two directions: first, toward an adaptive were explored for the first time. Washburn, despite and ecological view of human biological function, and having received his PhD degree only six years earlier second toward an evolutionary view of human genetic (1940), already was a dynamic force in the field of population structure. The leaders in this movement physical anthropology. He was Secretary-Treasurer were Joseph S. Weiner (1915–1982) and his colleague of the AAPA (1943–1946), had been teaching anatomy at Nigel A. Barnicot (1914–1975) (Harrison, 1982). Weiner, Columbia University since 1939, and persuaded Paul who was trained in physiology, anatomy, and anthropol- Fejos (1897–1963), Director of the Viking Fund (later the ogy in South Africa, came to England in 1937. With post- Wenner-Gren Foundation), to sponsor these Summer war academic posts at Oxford with the distinguished Seminars held in New York City. In addition, Washburn anatomist Le Gros Clark, and later at the London enlisted a new Harvard PhD, Gabriel W. Lasker School of Hygiene and Tropical Medicine, he contrib- (1912–2002) to edit a new Yearbook of Physical Anthro- uted to the training of a whole generation of human pology to report on the Summer Seminars and other news biologists, promoted the transformation of physical of the profession as well as to reprint important papers anthropology, and was instrumental in organizing not easily found in American anthropology journals. and managing human adaptability research during the The Yearbook of Physical Anthropology has continued to International Biological Programme (IBP) from 1962 to the present from the year of its first publication in 1946. 1974 (Harrison and Collins, 1982). Barnicot was trained In 1950, the fifth year of the Summer Seminars, in zoology and physiology, and was in the Anthropology Washburn organized a major conference that served Department at University College, London for most to bring together physical anthropologists and human of his professional life. He worked on blood genetics, geneticists with hopes to produce a fresh synthesis and skin and hair color in West African populations, on in human variation and evolution. The conference was studies of nonhuman primates and, in later years, on the

History of the Study of Human Biology 35 human biology of Tanzanian Hadza hunter-gatherers (Sunderland, 1975). Joseph Weiner was also a strong supporter of biocultural approaches in human biology, in addition to his commitment to solid experimental studies in human physiology. In a 1958 paper on “. . . training in physical anthropology and human biology,” he noted, “.. . it should be emphasized that in their [students’] research interests an important field of overlap exists between biological and cultural anthropologists” (Weiner, 1958, p. 47). Shortly before his death and nearly 25 years after the earlier statement, Weiner wrote: The simple fact that his unit of study is a defined community has made it imperative for the human biologist to take full account of the sociocultural properties of his community. Population structural analysis . . . cannot be pursued without close attention to social factors whatever parameter is under examination. Genetic analysis is inseparable from demographic and mating patterns; nutrition and energetics 3.5. Gabriel W. Lasker (1912–2002) at his desk. Photograph are inseparable from food production and food distribution, courtesy of Bernice A. Kaplan. and land holding; climatic adaptation must encompass the technology of housing and clothing; biomedical fitness is related to population size, sanitary systems, health services (1945) on Chinese migrants; James N. Spuhler (1946) and life-style (Weiner, 1982, p. 19). on human genetics; Stanley M. Garn (1948) on human Several other leading human biologists in the United hair composition and distribution; William S. Laughlin Kingdom who were junior to Weiner and influenced (1949) on Aleut populations; Edward E. Hunt, Jr (1951) by his ideas maintained this biocultural perspective. They on Micronesian populations; and Paul T. Baker (1956) on are Geoffrey A. Harrison, Derek F. Roberts, and James M. desert-heat stress. With the exception of Alice Brues and Tanner. Roberts conducted pioneering demographic and Paul Baker, all of these young anthropologists attended genetic research on Southern Sudan peoples (Roberts, some or most of the six Summer Seminars organized by 1956). Harrison and Tanner, along with Barnicot and Washburn that were held in New York City, and many Weiner, wrote the first modern textbook in Human contributed directly to these sessions, as well. Gabriel Biology in 1964, that continued the British tradition of Lasker’s (1999) memoir describes some of his associ- anthropological and biocultural perspectives in human ations with Hooton’s students at this time. biology (Harrison et al., 1964). In each section of the Again, in parallel with the United Kingdom, Ameri- book, culture figured prominently in examples that were cans Gabriel W. Lasker and Paul T. Baker provided presented. Overseas field experience and a commitment inspiration and guidance to junior colleagues and to comparative studies were attributes that exemplified students by promoting a biocultural framework. As most of this generation of biomedically trained human noted, Lasker (Figure 3.5) had built on Boas’s migration biologists. Each of these attributes led to an appreciation model in studies of Chinese (Lasker, 1946) and Mexican of the need to study human biology in the context of (Lasker, 1952) migrants. In addition to his research, human behavior and culture. which he pursued actively for more than 50 years, a In the United States, with the stimulus of Washburn’s major and long-term contribution to students and ideas on science and evolution, Earnest Hooton’s former junior colleagues was by way of support of their papers students at Harvard were moving the human biology submitted to the journal Human Biology, which he of living populations forward. Washburn contributed edited for 35 years. Baker (Figure 3.6), published widely very little to the growth of human biology because his with his own strong biocultural commitment, and also interests were largely in skeletal biology, functional socialized his students about the value of biocultural morphology, and primatology. Hooton’s students, who research. In a paper published four and a half decades completed their PhDs between 1940 and 1956, were to ago on “Climate, culture, and evolution,” he stated: become the leaders in human biology during the next “To completely reject the concept of climatic selection generation. These included, with PhD year in paren- without cultural involvement would be premature, but thesis: Alice Brues (1940) on the genetics of eye, skin, from anthropology’s present theoretical framework it is and hair; Marshall T. Newman (1941) on the peopling of more accurate always to consider the role of culture the American Southeast; Joseph B. Birdsell (1942) on when trying to formulate an evolutionary process Australian Aborigine populations; Gabriel W. Lasker related to climate” (Baker, 1960, p. 5). Twenty years later

36 Michael A. Little AFTER THE MIDDLE OF THE TWENTIETH CENTURY After 1950, new directions were taken in climatic morph- ology and physiology, genetics, disease, and adaptation to the environment. The book by Coon et al. (1950) outlined applications to climatic rules in humans includ- ing Bergmann’s rule (body size) and Allen’s rule (size and shape of extremities), and may have stimulated several biogeographic studies of human form. For example, Derek F. Roberts (1952, 1953) from the United Kingdom and Eugene Schreider (1950, 1951) from France explored relationships between body size and basal metabolism and climate in human populations around the world. They found that humans, as with other warm-blooded animals, conformed to Allen’s and 3.6. Paul T. Baker (1927–2007) (left) and Stanley M. Garn Bergmann’s rules. Other anthropologists from both (1922–2007) (right) some time during the late 1950s or early sides of the Atlantic conducted similar studies of the 1960s. Photograph courtesy of Barbara Garn. climatic environment and adaptation of humans to both heat and cold stress (Newman, 1953, 1956; Weiner, Baker (1982) continued to emphasize this theme 1954; Newman and Munro, 1955; Baker and Daniels, about the importance of the linkage between human 1956; Baker, 1958a, 1958b). Some of these studies were biology and culture in a Huxley Memorial Lecture. experimental (either laboratory or field) and were influ- In his 1996 Pearl Memorial Lecture, he summarized enced by the physiological studies conducted of heat and his ideas and underlined the “. . . need for all human cold stress of South Africans (Wyndham et al., 1952), biologists to have some training in relevant aspects Australian Aborigines (Scholander et al., 1958), Arctic of cultural anthropology” (Baker, 1997). Such a pers- Indians (Irving et al., 1960), Inuit/Eskimos (Brown and pective seemed not only intuitively correct and Page, 1952; Meehan, 1955), Alakaluf Indians (Hammel, logical to Baker, but it also led to more productive 1960), and Laplanders (Scholander et al., 1957). These lines of research and testable hypotheses. Weiss and and other studies demonstrated quite clearly that Chakraborty (1982, p. 383) observed that the complex human populations distributed around the world had influence of culture on evolution “. . . is a point still adapted morphologically, physiologically,andbehavior- largely misunderstood or ignored by many researchers ally to the climatic conditions of their environments. without anthropological training.” Harrison (1982, The difference between the physiologists’ studies and p. 471) suggested that it is much easier to give “equal the anthropologists’ studies was in the anthropologists’ attention . . . to cultural as to biological . . .” processes in understanding of the role of culture in these patterns of the United States than in Europe, because the subfields adaptation to temperature extremes (Baker, 1960). The of anthropology can often be found in the same depart- anthropologists were more concerned with the whole ment in the United States. culture/population and variations in responses by age Human biology research and student training and sex, whereas the physiologists’ contributions were during the immediate post-war years was supported more focused on physiological mechanisms in small enormously by the private Wenner-Gren Foundation samples of men tested under controlled laboratory for Anthropological Research (founded as The Viking conditions. Both kinds of studies enriched what we Fund in 1941). Partly because of the close personal learned and were complementary. relationship between its Director, Paul Fejos, and One of the post-war military institutions that stimu- Sherwood Washburn, the Foundation supported the lated work in anthropology and human biology was the Summer Seminars, the Yearbook of Physical Anthropol- Natick, Massachusetts Quartermaster Corps Climatic ogy, and other activities associated with the develop- Research Laboratory (Francesconi et al., 1986). Being ment of physical anthropology during that time close to Harvard University it could draw on talented (Szathma ´ry, 1991). During these early years, and up individuals from this university and also influence to the present, it is hard to imagine how human biol- research directions there and elsewhere, as well. Many ogy, within the larger field of physical anthropology, of the young scientists at the Climate Research Labora- could have developed into a mature science without tory were environmental physiologists who became the vision of Paul Fejos and later directors of the well-known scientists in later years, while a few of the Wenner-Gren and the financial support of the Founda- scientists were trained in physical anthropology tion (Baker and Eveleth, 1982). (e.g., Paul Baker and Russell Newman). Since the

History of the Study of Human Biology 37 Natick laboratories were also engaged in clothing and research in the anthropology department. Another nutrition studies, there were also strong interests in major contribution in human genetics in the 1950s was body size (anthropometry) and body composition vari- Livingstone’s (1958) paper on the anthropology of sickle ations in military personnel. These government studies cell in West Africa. Based on fieldwork in Liberia and contributed to the mix of interests in nutrition, disease, extensive literature research on West Africa, he demon- body composition, and climate in the context of human strated the relationships among malaria, sickle cell adaptation to variable environments. distribution, mosquito ecology, agriculture (forest trans- By the early 1960s, studies of body composition in formation), and language and human migration. Only human biology were on the rise (Broz ˇek and Henschel, four years after Allison’s (1954) publication, Livingstone 1961; Broz ˇek, 1963; Garn, 1963). Interests among had shown the importance of culture in structuring anthropologists were derived, in part, from knowledge evolutionary change through its influence on malaria of skeletal biology and were linked closely to studies of prevalence and gene-frequency distributions. Culture nutritional adaptation (Broz ˇek, 1956; Newman, 1960, change had produced evolutionary change! 1962), and child growth (Garn and Shamir, 1958). Other important genetics work in the 1950s and 1960s Anthropometric (Garn, 1962) and physique (Broz ˇek, by anthropologists outside the United States and the 1956) surveys were carried out to evaluate nutritional United Kingdom included research by Jean Hiernaux status, and associations between growth and nutrition in (1966), from Paris, who studied blood-group genetics in stressful environments were studied (Roberts, 1960; African populations. Another major figure is Francisco Schraer and Newman, 1958). Josef Broz ˇek (1999) Salzano (Salzano and Callegari-Jacques, 1988), from provided a personal history of body composition studies Brazil, who first did post-doctoral research with James in human biology. Neel at the University of Michigan in the late 1950s, and During this period, human population genetics who went on to become the major human geneticist to focused almost exclusively on blood polymorphisms. study tropical forest Native Americans in South America. A good example was Alice Brues’s (1954) important paper demonstrating that selection must have operated on the ABO blood group system because of the nonran- HUMAN ADAPTABILITY, ECOLOGY, AND dom distribution of ABO allele frequencies around INTERNATIONAL PROGRAMS the world. Electrophoresis was first used to separate proteins, and it was at this time that “.. . the staggering Among human biologists, the 1960s ushered in a magnitude of [human] genetic variation . . .” became mature sense of scientific problem solving, an increasing apparent (Cavalli-Sforza et al., 1994, p. 3). One of the commitment to understanding evolutionary process great discoveries of the 1950s was the relationship and adaptation to the environment, and a growing inter- between the sickle cell gene and malaria and the protec- est in integrated, collaborative studies. In anthropology, tion afforded against malaria by the sickle cell heterozy- broadly, there was a movement toward empiricism gote. James Neel (1915–2000) had worked out the with interests in ecological approaches and cultural genetics of sickle cell disease (Neel, 1949), and then materialism. At the same time, human biologists were described the anemias of sickle cell disease and thalas- exploring ecological models in the context of adaptation semia in his 15th Cold Spring Harbor Symposium paper to the environment (Baker, 1962; Little, 1982). In 1964, (Neel, 1951). Neel proposed mutation or selection as four distinguished British human biologists published hypotheses to explain the high prevalence of these the first edition of an important introductory textbook dangerous anemias in the Mediterranean and in Africa, (Harrison et al., 1964). In this book, Geoffrey A. Harrison but it remained for Anthony Allison to show that and Nigel A. Barnicot dealt with genetics and phenotypic the major force was natural selection. Allison (1954) variation, James M. Tanner covered human growth, and demonstrated that the sickle cell gene was a balanced Joseph S. Weiner wrote the sections on human adapta- polymorphism maintained by the selective pressure of tion and human ecology. This was an important work malaria. He described in some detail the history of his because it defined the field of human biology, was early work and the conditions surrounding the discovery synthetic, and it was state of the science for the 1960s. in an autobiographical paper (Allison, 2002). Weiner’s final section on “Human Ecology” was the first James Neel established the first human genetics definition of this important perspective in human department at the University of Michigan in 1956. biology: topics included nutrition, disease, climate, and At that time, this program was one of the strongest in population (demography), and a central emphasis of the the country, partly because of its connection with work was on “ecological adaptive processes.” anthropology. William J. Schull worked with Neel in Two other important events took place in 1964. human genetics and had a joint appointment in anthro- Firstly, the International Council of Scientific Unions pology, while James N. Spuhler (1917–1992) and Frank (ICSU) (now the International Council for Science) in B. Livingstone (1928–2005) were conducting genetic Paris established the International Biological Program

38 Michael A. Little (IBP) with a planning phase from 1964–1967, a DeVore, 1976) and the Andean Biocultural Studies research phase from 1967–1972, and a synthesis initiated in 1962 (IBP affiliated, Baker and Little, 1976). phase to follow from1972–1974. The orientation of this Two other important IBP projects included the worldwide program was ecological, and its theme was International Studies of Circumpolar Peoples, a four- “The Biological Basis of Productivity and Human nation investigation of the health of Inuit/Eskimos in Welfare.” A number of sections of the IBP were designed Alaska, Canada, and Greenland (Milan, 1980), Popula- to cover various components of ecology and ecosystems tion Genetics of the American Indian, a project that studies. A separate section called Human Adaptability focused on the Yanomama horticulturalists from Brazil (HA) was to cover “the ecology of mankind” from a and Venezuela (Neel et al., 1977), and the Solomon variety of perspectives including health and welfare, Islands Project, organized by Albert Damon and others environmental physiology, population genetics, child at Harvard (Damon, 1974; Howells 1987). These and growth, anthropology, and demography (Weiner, 1965). other HA projects were discussed in detail by Hanna Secondly, although some preliminary preparation had et al. (1972) and from theoretical perspectives by begun in 1962, an important HA planning conference Lasker (1969). was held in Austria at the Wenner-Gren Foundation Burg Other major contributions from the IBP were the Wartenstein Conference Center in the July of that year. surveys of human growth (Eveleth and Tanner, 1976) It was at that meeting that Weiner (1966) outlined the and human gene frequencies (Collins and Weiner, major categories of planned research, and the other, 1977) that were conducted on populations around the more than 20 internationally represented, participants globe that contributed primary data. In many of these reported on current knowledge over a wide variety of smaller Third World projects, research design and topics in human biology for North and South American, “problem orientation” were subordinated to data European, Asian, African, and Australian populations collection, but these important data sets, then, became (Baker and Weiner, 1966). Weiner, who was Inter- available for comparative, historical, and in-depth national Convener (director) of the Human Adaptability analyses for future investigators (Weiner, 1977). Section of the IBP prepared a handbook (Weiner and Following the close of the IBP in the 1970s, a new Lourie, 1969) of standardized methods, and at the end international program arose from UNESCO (United of the IBP published a compendium of the completed Nations Educational, Scientific, and Cultural Organiza- research (Collins and Weiner, 1977). The planning and tion) called the Man and the Biosphere Program (MaB). research that followed resulted in the participation of Additional multidisciplinary projects were affiliated with 40 nations, the completion of more than 230 projects, the MaB program including the Multinational Andean and several thousand publications under the HA banner. Genetic and Health Project (Schull and Rothhammer, In a retrospective review of the HA research, there has 1990) and the Samoan Migrant Project (Baker et al., been some criticism based on topical omissions, but the 1986). Other projects, such as the South Turkana contributions far outweigh the shortcomings (Ulijaszek Ecosystem Project (Little and Leslie, 1999), the Ituri and Huss-Ashmore, 1997). Harrison (1997, p. 25) noted, Forest Project (Bailey, 1991), and the Siberian Evenki in a contribution to the same review: “Notwithstanding Project (Crawford et al., 1992) followed somewhat later. its limitations, it played a major part in converting the old These multidisciplinary projects promoted international defunct physical anthropology into the vibrant and excit- collaboration, gathered fundamental data on popula- ing component of biological anthropology as it is today.” tions that were on the brink of extinction, were instru- In addition, it is quite clear that the associations made mental in developing new field and experimental between the British and American human biologists methods, and provided research training for a whole during the IBP were of remarkable value by serving to new generation of human biologists. cross-fertilize ideas and to reinforce the biocultural and environmental perspectives shared by most of the partici- pants (Baker, 1988). DNA ANALYSIS AND MOLECULAR GENETICS One of the major conceptual contributions of the IBP was in the organization of multidisciplinary research Weiss and Chakraborty (1982) provided a detailed (Little et al., 1997). Because complex ecosystems required review of the history of human genetics up to the late broad expertise, many individuals from different fields of 1970s. The late 1970s and 1980s, however, marked a science were required to collaborate. Ecologists from the dramatic transition in human population genetics IBP were organized according to major ecosystems (e.g., from a reliance on phenotypic proteins and inference tropical forests, arctic tundra, deserts, and grasslands), about genes (phenotype-based approach) to the direct while some of the HA projects focused on single human measures of genes via DNA (the new molecular genet- populations in special environments. Two early projects ics). This transition was based on the new laboratory were the Kalahari Research Group to study the !Kung methods that enabled DNA to be extracted and purified Bushmen initiated in 1963 (not IBP affiliated; Lee and from Blood and other tissues (Crawford, 2000). These

History of the Study of Human Biology 39 methods included: use of restriction enzymes to clip concern of human population biologists about the HGP nucleotide segments of DNA (restriction fragment was that it would provide a generic genome; that is, little length polymorphisms), DNA hybridization, and the or no information would be provided on human genetic polymerase chain reaction for amplification of DNA. variation in individuals and populations around the Some of the new DNA methods were used in human world. Such variation was of value to studies of human biology for phylogenic reconstruction of our ancestors, evolution and population history, as well as applied to reconstruct human population movements over research in the genetic bases for disease (resistance, space and time, and for forensic and ancient DNA susceptibility, environmental interactions, etc.) and analysis. New methods of DNA extraction and analysis genetic epidemiology. also led to the Human Genome Project and the contro- In response to these concerns, the Human Genome versial Human Genome Diversity Project. Diversity Project (HGDP) was founded in 1991 under Mitochondrial DNA (mtDNA), which is only trans- the leadership of L. L. Cavalli-Sforza, the distinguished mitted through the maternal line and does not undergo Stanford University geneticist. Also, the project was recombination, has been very useful in a variety of identified as urgent because of the disappearance of approaches to human evolution (Cann, 1986). One many small Mendelian populations. The proposed of the most remarkable studies of mtDNA was the HGDP sparked a controversy for a variety of factors: so-called “Mitochondrial Eve” observation that startled first, there was distrust by anthropologists of the gen- scientists interested in modern human origins. Cann eticists in their conceptualization of race (linked to et al. (1987) found in a sample of 147 people from typological race and eugenics); second, there were con- around the world that “All these mitochondrial DNAs cerns by Native American groups about their being stem from one woman who is postulated to have exploited again, about race stereotyping, and about lived about 200 000 years ago, probably in Africa” the potential use of DNA for commercial gain (Cann et al., 1987, p. 31). This and other DNA work (patenting genes); third, there was an element of polit- led to the “Out of Africa” hypothesis on modern ical naivete ´ of the organizing geneticists about conflict- human origins (Stoneking and Cann, 1989; Vigilant ing beliefs about race and, initially, there were few et al., 1991). More recent research incorporated the anthropologists included in the program and its design Y chromosome, which is inherited through paternal (Reardon, 2005). Because of conflicts and protests the lineages (Hammer, 1995). original plan to establish cell lines of DNA samples A variety of studies have been conducted to trace from 25 individuals from each of 400 populations was population distributions and migrations in the historic never satisfied, and funding was not forthcoming. and prehistoric past. Cavalli-Sforza et al. (1988) and Later the HGDP was revived under the title of HGDP- Sokal (1988) combined genetic, archaeological, and CEPH (Centre d’Etude du Polymophisme Humain), linguistic information to reconstruct population and now DNA is available for 1064 individuals from expansion in Europe. This geographic genetic research 52 populations. These DNA samples have begun to be continues up to the present (Barbujani, 2000). Another used for research (Ramachandran, 2005). area of interest that had been dominated by archaeo- During the past half century, there has been logical evidence soon saw the application of genetic increasing scientific activity in what Derek Roberts data to the question the peopling of the New World (1965) first referred to as “anthropological genetics.” (O’Rourke, 2000). Based on mtDNA and Y-chromosome Crawford (2007) outlined the differences between markers, Merriwether (2002) suggested a single-wave “anthropological genetics” and “human genetics,” migration, possibly originating from ancient Mongolia where the former incorporates a biocultural perspec- through Siberia and into the New World between 20 000 tive, focuses on the population and, often, non-Western and 30 000 BP. All of these dates fall within the late populations, and centers on questions of interest to Pleistocene when the Wisconsian glaciation covered anthropologists, particularly evolutionary questions. parts of North America and the cold Beringia land Research in molecular anthropology, genetic epidemi- bridge between Siberia and Alaska was exposed. ology, forensic anthropology via DNA analysis, human By the late 1980s, it was feasible to sequence the origins, and the history of human migration and whole human genome of nucleotides and DNA on the dispersal are all areas of exploration that are being human chromosomes. This culminated in the Human pursued by anthropological geneticists. Genome Project (HGP), a massive DNA-sequencing effort that was estimated to cost several billion dollars. In the United States, the Department of Energy (DOE) and REPRODUCTION AND CHILD GROWTH the National Institutes of Health (NIH) supported the HGP (Crawford, 2000). The work conducted by an inter- Interests in reproduction in anthropology date back to national consortium began in 1990 and was completed the early part of the twentieth century and are linked in 2003, several years ahead of schedule. The principal closely to two primary emphases in human biology:

40 Michael A. Little evolution/natural selection and human health. In although energy balance does play an important role. evolution, differential fertility is one of the main However, Frisch’s research was truly pioneering in that driving forces of selection, and fertility (measured as it contributed to the development of a new ecology of number of live-born offspring) is also a prime indicator reproduction that began to explore the evolution and of adult health and well being. Studies of growth are an ecology of reproductive function in a variety of West- extension of reproduction and, of course, date back in ern and non-Western peoples (Howell, 1979; Ellison, human biology to Franz Boas. 1990, 2001; Leslie et al., 1994). These and other investi- Pioneering studies by Sophie Aberle, who was trained gations explored relationships among nutritional status in genetics at Stanford and medicine at Yale, were and availability of food resources, disease, physical conducted of sex and reproduction in San Juan Pueblo activity, endocrine function, body composition, behav- Indians in New Mexico (Aberle, 1928, 1931). She was a ior, growth, and reproduction – truly an integrated, member of the National Research Council Committee for behavioral and environmental science of human Research in Problems of Sex (Aberle, 1953) and worked reproduction. for the Bureau of Indian Affairs. From her research she At the same time as this new research direction found that the poor reproductive pattern of Pueblo in fertility and reproduction was being taken, so were Indians was not based on reduced fertility, which in fact new discoveries being made in infant, child, and adoles- was very high, but rather was based on extraordinarily cent growth studies. Surveys of the worldwide variation high infant mortality rates (about 25%). Ashley Montagu in human growth, which were originally compiled from (1905–1999) conducted early work of female sexual the IBP HA studies, synthesized the available knowledge maturation by discovering post-menarcheal sterility in on population variations in human growth up to the late adolescent girls (Ashley-Montagu, 1939a, 1939b, 1946). 1980s (Eveleth and Tanner, 1976, 1990). Somewhat later, He found that, for a year or more following menarche in Michelle Lampl conducted new longitudinal research of girls, there was a markedly reduced fecundity or even individual growth patterns (Lampl et al., 1992) in an sterility, and “.. . that puberty and the power to procreate ingenious study of infant length, where some infants are not synchronous events . ..” (Ashley-Montagu, 1939b, were measured every day for more than a year. She and p. 213). The sociologists, Kingsley Davis and Judith Blake her colleagues found that rather than being a continuous (Davis and Blake 1956) formulated a groundbreaking process as believed, growth proceeded in “incremental model of factors affecting fertility in the context of bursts” (saltatory growth) followed by periods of stasis. social behavior. This provided the basis for systematic Somemeasurements of infants showed daily increases in hypothesis testing and refinement of the model, which length of up to 1 cm in length! This work was extended was done by Bongaarts (1978). Campbell and Wood to adolescents, who also showed saltatory growth (1988) further refined the model for “proximate deter- (Lampl and Johnson, 1993), and has led to new lines minants” of fertility in non-Western (noncontraceptive) of research in bone and soft tissue growth and in the as well as Western populations. One of the first studies to endocrine control of growth. document an important variable, not considered by In the 1980s and 1990s, increasing research in Davis and Blake (1956), was the central importance growth has been directed toward health and the total of breast-feeding in fertility control by Konner and life span from conception to senescence and death. Worthman (1980) in studies of !Kung Bushmen. This This research has had both basic and applied compon- and other research work stimulated a number of anthro- ents related to health across what has been called the pological studies of nursing, energetics, and fecundity in “life span approach” to the study of human biology traditional populations (Bentley, 1985; Gray, 1994; (Leidy, 1996). A general principle of this approach is that Vitzthum, 1994). “. . . no single stage of a person’s life (childhood, middle An important series of studies was conducted in the age, old age) can be understood apart from its antece- 1960s and 1970s that explored relationships among dents and consequences” (Riley, 1979, p. 4). Activity energy balance (diet and activity), body composition levels, diet, and exposure to adverse substances during (fat and muscle), and reproduction (age of menarche) childhood all influence health during adulthood and sen- in adolescent girls. The work by Rose Frisch (Frisch escence. It is also the case that longitudinal study of the and Revelle, 1970; Frisch and McArthur, 1974) led to a same individuals through time, in the tradition of Franz hypothesis the there was a critical weight, and, then Boas, is one of the best methods to demonstrate these life later, a critical level of body fat that both triggered span correlations. David J. P. Barker (Barker et al., 1989; menarche and maintained normal menstrual cycles. Barker, 1992) developed an extension of this approach to Frisch’s ideas were severely criticized, partly because understanding the life span. He discovered from early of her unyielding adherence to this basic hypothesis. twentieth-century archival birthweight data that the Subsequent research has demonstrated that reproduct- lower the birthweights of infants, the greater the risk of ive controls on the ovarian cycle are more complex coronary heart disease in adulthood. He also found that than just being a function of the energy of body stores, lower birthweight increases the risk of hypertension,