American Journal of Primatology 28:251-261 (1992) Primate Longevity: Its Place in the Mammalian Scheme STEVEN N. AUSTAD’ AND KATHLEEN E. FISCHER’ Harvard University,’ Department of Organismic and Evolutionary Biology, and ‘Department of Organismic and Evolutionary Biology, The Biological Laboratories, Harvard University, and Department of Anthropology, The Peabody Museum, Cambridge, Massachusetts Data on captive longevity in 587 mammalian species were analyzed in order to evaluate primate longevity in the context of general mammalian life history patterns. Contrary to some recurrent claims in the literature, we found that 1) primates are not the longest-lived mammalian order, either by absolute longevity, longevity corrected for body size, or metabolic expenditure per lifetime; 2) although relative brain size is highly correlated with longevity in primates, this is a n aberrant trend for mammals in general, and other body organs account for a n even greater amount of variation in longevity; and 3) there has been no progressive evolution of increased longevity among the primate superfamilies. The exceptional magnitude of primate longevity may, in keeping with evolutionary senescence theory, be due to a n evolutionary history of low vulnerability to environmentally imposed death due to their body size, arboreal habit, and propensity to live in social groups. 0 1992 Wiley-Liss, Inc. Key words: longevity, b o d y size, brain size, evolution, metabolism INTRODUCTION The exceptional longevity of primates, especially humans, relative to other mammals has had a major impact on scientific thought concerning the determinant(s) of animal life span [Sacher, 1959, 1975; Cutler, 1975, 19761. For instance, primate longevity has contributed to the notion that brain size (or some specific component of brain size such as cortex size or cerebro-cortical surface) and longevity are causally related [Sacher, 1959, 1975; Cutler, 1975; Mallouk, 1975; Weiss, 1981; Hofman, 19831. Second, it has been stated that primates exhibit a greater energetic expenditure per unit body mass per lifetime than other mammals, and this energetic longevity reflects greater metabolic efficiency [Cutler, 19781. Finally, it is commonly stated life span has been increasing through evolutionary time [Schultz, 1969; Cutler, 1975; Bowden & Jones, 1979; Lovejoy, 1981; Hofman, 1984; Napier & Napier, 19851. These ideas arose primarily from the analysis of the allometric relationship between a size variable (e.g., body mass, brain mass) and maximum captive longevity for a species. . Maximum captive longevity is likely to reveal a n important aspect of the rate Received for publication May 11, 1992; revision accepted July 4, 1992. Steven N. Austads current address is: Department of Biological Science, University of Idaho, Moscow, ID 83843.Send reprint requests to him there. 0 1992 Wiley-Liss, Inc. 252 I Austad and Fischer of aging, because it presumably reflects how long individuals can survive under optimal conditions before dying due to senescence. The study of captive longevity is therefore an indirect manner of studying aging. We can see the importance of the first two aspects of primate longevity in relation to general theories of aging, if we consider several of the historically important theories. An enormously influential general theory of aging has been the rate-of-living theory [Rubner, 1908; Pearl, 19281. Briefly, this idea posits that an organism’s life span depends upon a genetically determined metabolic potential and the rate at which that potential is expended. In its strong form, this notion holds that related species, such as birds or mammals, have similar metabolic potentials and consequently the diversity of life spans could be largely explained by differences in metabolic rate [Sacher, 1959; Lindstedt & Calder, 19761. In its more modern, weaker form, the theory holds that aging is an ineluctable consequence of imperfect physiochemical processes which ultimately lead to failure of cellular constituents due to the accumulation of biochemical errors or to the gradual buildup of toxic metabolic byproducts [Sohal, 19761. However a major prediction of the theory remains that other things being equal, organisms with low metabolic rates should outlive organisms with high metabolic rates. It is well known that there is a general trend for maximum life span to allometrically increase, and mass-specific metabolic rate to allometrically decrease, with body size among mammals [e.g., Sacher, 1959; Lindstedt & Calder, 1981; Calder, 19841. In combination these trends tend to support the rate-of-living theory. The idea that primates reputedly expend greater metabolic energy than other mammals has led to the speculation that they are a special case among mammals with some generalized physiological superiority, and that some unique primate characteristic besides metabolic rate was causally related to this physiological superiority [Sacher, 1959, 1975; Cutler, 1975; Mallouk, 1975; Weiss, 1981; Hofman, 19831. One glaringly obvious difference between primates and the rest of mammals is their relative large brain size. Hence brain size was a prime candidate for the source of primate longevity. A relatively large brain has been hypothesized to confer longevity either because of the improved precision of physiological homeostasis it putatively mediates [Sacher, 19591, some substance vital to tissue repair it secretes [Mallouk, 19751, or its specialized metabolic demands [Hofman, 19831. This theory would predict that if there has been a progressive evolutionary increase in brain size, there would also be a progressive increase in life span. A competing general theory is evolutionary aging, or senescence, theory [Medawar, 1952; Williams, 19571. Evolutionary senescence theory attributes aging to the declining strength of natural selection a t successive ages after sexual maturity. According to this idea, the power of natural selection to favor advantageous alleles, or eliminate disadvantageous ones, wanes with the age a t which these alleles are expressed. The waning of selection is due to environmental hazards which dictate the probability that an animal will be alive at the time the trait is expressed. Aging, then, is the genetic consequence of a) the accumulation of lateacting deleterious alleles which are nearly selectively neutral in their effects, and b) positive selection for alleles with beneficial early-acting effects and deleterious late-acting effects. This second mechanism has been called negative or antagonistic pleiotropy and could be manifested in traits, for instance, that favor a rapid rate of early reproduction at the expense of later survival [Williams, 19571. It is important to note that the rate at which natural selection declines with age will be related to the degree of environmental hazard, i.e., the probability of death which is unrelated to aging. Note also that this theory is one of physiological design optimization, such that organisms may be engineered t o reproduce rapidly Primate Longevity I 253 but not endure very long or the reverse. In any case, organisms’ aging rate should be related t o their ecological history and the trade-offs between early and late fitness. With respect to primates, evolutionary senescence theory suggests that their exceptional life spans would be due t o an evolutionary history of low environmental hazard and slow early reproduction, and that other mammal groups with similar histories should also exhibit exceptional longevity. Also according to this theory, there is no reason to expect a progressive evolutionary increase in longevity in any group. In this study we will assess primate longevity, lifetime metabolic expenditure, and relative brain size in light of the overall mammalian patterns. Previous studies on general patterns of mammalian life span have been performed using relatively few of the more than 4,300 species of extant mammals and approximately 180 living primate species [Corbet & Hill, 19911. Currently, data from many more species are available. In addition, improved husbandry and record keeping in zoos have substantially increased maximum longevity for less commonly kept species. Analyses of more extensive data sets have overturned some of the conventional wisdom from the past already [Prothero & Jurgens, 1987; Austad & Fischer, 19911 and we feel such an analysis is timely MATERIALS AND METHODS Our data base consists of body mass and maximum recorded longevity for 587 mammalian species, including 77 primates, gathered from standard secondary sources [Anonymous, 1960; Flower, 1931; Crandall, 1964; Haltenorth & Diller, 1977;Jones, 1982; Nowak & Paradiso, 1983; Harvey et al., 19861,supplemented by material from the primary literature and by personal correspondence with zoo personnel, field biologists, and veterinarians. When several records of maximum longevity were found, we selected the longest. Nearly all the longevity data derive from captive populations with the exception of the bats, for which 82% (41150) of our species information comes from the maximum recapture interval of individuals from banded wild populations. The accuracy with which mark and recapture data reflects actual maximum longevity of bats in the wild will depend upon banding intensity and the frequency of capture attempts as well as on the length of the study. Consequently bat longevity is probably substantially underestimated in all species studied only in the wild. The data base does not represent all species for which information was available. Maximum captive longevity is likely to be misleadingly short if husbandry techniques are inadequately developed. Therefore we did not use available data for a species if we knew that a) breeding attempts in captivity had not been successful, b) greater than 50%of individuals died during their first year in captivity, or c) the author indicated that life span was probably shortened by captivity. Furthermore, because maximum longevity will be strongly influenced by sample size a t small numbers, we deleted species for which the sample of prospective life spans was known to be less than ten. Body mass was determined in several ways. Field measurements of body mass always took priority over captive data if both were available. If only body mass ranges were given, we used the arithmetic mean of those values. In highly sexually dimorphic species, we used the arithmetic mean of male and female body mass. If only maximum body mass was reported, we used 55% of that value, a convention used by Prothero and Jurgens 119871 after finding it to be the ratio of mean to maximum body mass reported for the Asian elephant. McNab  has compiled data on basal metabolic rate for 321 species, and we added to this information on 18 additional species from the primary literature. 254 I Austad and Fischer Both maximum life span and basal metabolic rate were available for 164 species. Primate brain sizes are from Harvey, Martin, and Clutton-Brock [19871, and other brain sizes are from Eisenberg 119811, Crile and Quiring 119401, and Pirlot and Stephan 119701. Data on the mass of other body organs (spleen, liver, kidney, and heart) are from Crile and Quiring 119401. Using species as independent data points can confound analyses [Pagel and Harvey 19881. Well-studied taxa, or those with many species, can dominate a sample and bias the analyses. To prevent our sample from being dominated by particular taxa for which many data points were available (e.g., Rodentia), we used nested mean values for all higher level comparisons. That is, order means were calculated from family means which were calculated from genus means which were, in turn, calculated from species values. Taxonomy is from Corbet and Hill , and statistical analyses from Wilkinson [19901. RESULTS It is well known that mammalian longevity increases with body size [e.g., Rubner, 1908; Sacher, 19591. Our previous work, using LOWESS regression (robust locally-weighted sum of the squares, a statistical technique that reveals the shape of the relationship between two variables by using the weighted average of nearby values of the dependent variable to calculate expected values [Cleveland, 1979, 198111, has indicated that if all our species are included, the relationship between the logarithms of body mass and maximum longevity is not even approximately linear. However, if bats and marsupials are deleted, the relationship becomes surprisingly linear [total 463 species, Austad & Fischer, 19911. The resulting regression line can be considered to yield the “expected longevity” for a mammal of a particular body size. We call the ratio of actual to expected longevity and longevity quotient (LQ),which gives the intuitively satisfying proportion of an “average” mammal’s longevity which the species in question exhibits. LQ in our data ranges from 0.14 (Australian water rat) to 5.39 (little brown bat). Life Span of Primates Relative to Other Mammals Primates are clearly among the longest-lived mammals. They exhibit the fourth highest absolute longevity among the eighteen mammalian orders examined, and the third highest LQ (Table I). They are also clearly not the longest-lived mammals for their body size. They are surpassed in this respect by the monotremes (echnidas and duck-billed platypus) and the bats. Contrary to past assumptions [e.g., Sacher, 19591,the tendency to hibernate or enter daily torpor is not correlated with exceptionally long life among mammals, as shown by the fact that nonhibernating bats and marsupials are equally long-lived as related species that do not hibernate or enter torpor [Jiirgens & Prothero, 1987; Austad & Fischer, 19911. Multiplying mass-specific basal metabolic rate times maximum longevity yields lifetime basal energy expenditure-a quantity which has been hypothesized to be approximately constant throughout the mammals [Rubner, 1908; Sacher, 1959; Stahl, 1962; Cutler, 19761. Primates have the second highest lifetime energetic expenditure by this measure, but it is still significantly less than that of bats = 0.02). (Table I) (Tll = 2.297, Pone-tail Organ Size and Longevity in Primates and Other Mammals Primates have exceptionally large brains for their body size [Jerison, 19731, and brain size has been implicated in their longevity [Sacher, 1959, 1975; Mallouk 1975; Cutler, 1976; Hofman, 19841. A convenient measure of brain size, corrected for body size, is encephalization quotient (EQ), or the ratio of actual brain size to Primate Longevity I 255 TABLE I. Mean Maximum Longevity, and Longevity Quotient (LQ), Encephalization Quotient (EQ),and Lifetime Energy Expenditure (LEE) for 18 Mammalian Orders [Taxonomy based on Corbet & Hill,19911 Order Chiroptera (bats) Monotremata (monotremes) Primata (monkeys & apes) Dermoptera (flying ‘‘lemurs”) Edentata (edentates) Scandentia (tree shrews) Proboscidea (elephants) Carnivora (carnivores) Artiodactyla (even-toed ungulates) Tubulidentata (aardvark) Perissodactyla (odd-toed ungulates) Rodentia (rodents) Hyracoidea (hyraxes) Marsupialia (marsupials) Lagomorpha (rabbits, hares) Insectivora (insectivores) Pholidota (pangolins) Macroscelidea (elephant-shrews) Number of species Maximum longevity (yr) LQ EQ LEE (kcal/g/life) 50 14.9 2.75 0.94 602 3 32.6 2.41 0.83 228 77 27.9 1.92 2.54 420 1 17.5 1.56 - - 9 21.4 1.44 0.95 171 2 12.0 1.44 1.34 397 2 64.5 1.36 1.59 - 109 21.4 1.07 1.22 283 92 22.0 0.84 0.84 209 1 24.2 1.06 - 130 12 36.7 0.96 0.92 - 131 10.0 0.95 0.99 250 3 12.1 0.92 0.90 200 67 10.1 0.88 0.61 149 7 8.1 0.85 0.62 241 14 6.2 0.79 0.55 228 1 13.1 0.72 - - 6 4.1 0.62 - 87 that expected from the general relation between mammalian body size and brain size [Jerison, 19731. The most complete analysis of the mammalian brain size-body size relation was by Eisenberg and Redford, who assembled data on 547 mammalian species in similar relative proportion to that actually represented among taxonomic orders [Eisenberg, 19811. Using their formulation for EQ, we find that primates have by far the largest brains among mammals (Table I). Furthermore, in primates EQ is indeed significantly related to both maximum longevity (r = 0.656, N = 73, P < < 0.001) and LQ (r = 0.621, P < < O.OOl), although this relationship explains less than half of the variation and is highly leveraged by the outlying values for the Hominoidea. The nonhominoid relationship remains statistically significant, but 256 I Austad and Fischer TABLE 11. Correlations Between Encephalization Quotient (EQ) and Two Measures of Longevity-Maximum Longevity and Longevity Quotient (LQ).P Values are Corrected by a Sequential Bonferroni Technique for Simultaneous Tests [Rice, 19891. Order Artiodactla Carnivora Chiroptera Insectivora Marsupialia Primata Rodentia EQ-LQ Correlation (Pearson’s r) EQ-maximum longevity N -0.512 -0.052 -0.125 -0.085 -0.097 0.621*** 0.371 -0.672* -0.438 0.066 -0.174 -0.329 0.656*** 0.049 21 33 23 12 24 73 39 *P < 0.05. ***P << 0.0001. is even looser (r = 0.506, N = 69, P < 0.001 and r = 0.478, P < 0.001, respectively). On the other hand, primates are rather unusual in this relationship. Among the seven other mammalian orders for which we have a reasonably sizable sample (N 2 12), no other order has a similarly significant positive relationship between EQ and LQ or EQ and maximum longevity, in fact most of the correlations are negative, and the only other significant correlation ( A ~ t i o d ~ c t yisZ a~negative ) one (Table 11). Moreover, the mammalian groups with the highest LQs, monotremes and bats, have EQs below the general mammalian average (mean = 0.83, N = 2 and mean = 0.85, N = 22, respectively). In fact, there is no correlation between EQ and maximum longevity (r = 0.452, N = 14, P = 0.105) or LQ (r = 0.335, P = 0.335) for the mammalian orders generally, or for all the nonprimate species when considered individually (EQ-LQ, r = -0.002, N = 158, P = .998; EQmaximum longevity, r = 0.047, P = 562). A second point is that other organ weights are also correlated with longevity (Table 111). In fact in our data for nonprimate species, heart, liver, kidney, and spleen sizes all show a stronger correlation with maximum longevity (or its logarithm) than does brain size. Progressive Lengthening of Life in Primates Do primates exhibit increased longevity throughout their history, as exemplified by extant forms that are considered to represent earlier stages in primate evolutionary development as some have hypothesized [e.g., Schultz, 1969; Lovejoy, 1981]? This is an orthogenetic notion which implies that the longer ago a group diverged from the direct ancestry of the hominids, the more reduced will be its longevity. The most straightforward manner in which to assess this notion is to look at the relationship between body size and maximum longevity in each of the six extant primate superfamilies (Lemuroidea, Lorisoidea, Tarsioidea, Ceboidea, Cercopithecoidea, and Hominoidea) [Koop et al., 19891. In absolute terms, mean longevity is clearly the greatest in the Hominoidea, second greatest in the Cercopithecoidea; the other four superfamilies are similar to one another, with no discernible pattern relating divergence and longevity (Table IV). However, cercopithecoids and hominoids are also substantially larger than the other primate groups. This size bias can be corrected by comparing LQ rather than absolute values for longevity. While hominoids are still clearly the longest-lived group, the Primate Longevity I 257 TABLE 111. Relationships Between Maximum Longevity and Organ Mass for Nonprimates Organ Maximum longevity Correlation (Pearson’s r) Log (maximum longevity) Brain Heart Liver Kidney Spleen .579 .604 .639 ,613 ,677 ,417 ,428 .443 .446 .535 N 60 54 50 52 26 TABLE IV. Mean Body Mass, Maximum Longevity, Longevity Quotient, and Mean Residuals From the Primate Log Maximum Longevity-Body Mass Regression in the Six Monouhvletic Grouus of Primates Grour, Lemuroidea Lorisoidea Tarsioidea Ceboidea Cercopithecoidea Hominoidea Body mass (ka) Longevity (yr) LQ Mean residuals 1.97 1.65 0.16 1.46 10.18 42.02 17.66 16.02 12.65 15.60 26.13 57.66 1.57 1.65 1.67 1.47 1.68 2.82 -0.340 0.006 0.042 -0.002 -0.013 0.102 other primate superfamilies share remarkably similar LQs, and there is certainly no scala naturae, no trend toward decreasing longevity as relationships become more distant from the hominoids. An alternative approach to the same issue is to construct a primate-specific regression line between log body mass and maximum longevity (Fig. 11, and to analyze the residuals from this line. A one-way analysis of variance of these residuals by superfamily shows no significant differences between groups (F5,71= 1.115, P = 0.3601, and absolute values of the residuals show no trend for decreasing longevity as one diverges farther and farther from the hominid lineage (Table IV). DISCUSSION Primates are indeed relatively long-lived mammals, but they are clearly not the longest-lived either in absolute longevity or in longevity corrected for body size. Relative brain size is correlated with maximum longevity within the primates, but this is apparently a primate idiosyncrasy, because a similar relationship is missing in the other mammalian orders or across the mammals generally. Even within primates, the correlation between relative brain size and longevity is not particularly tight and consequently brain size cannot be used to predict maximum longevity with any degree of confidence. Finally, although hominoids are substantially longer-lived than the rest of the primates, both in relative and absolute terms, the other primate superfamilies have very similar LQs to one another. The original implication of brain size in the aging process derived from the fact that brain size was more closely correlated with maximum longevity than was body size [Sacher, 19591. It is a well known, if often ignored, caveat in statistical analysis that correlation does not imply causation regardless of the strength of the correlation [Radinsky, 1982; Calder, 19841. The relationship between brain size 258 I Austad and Fischer 2.0 I I I H I r ;k v ~ 1.7 . U d 5M a 3 ti 1.4 B 1 Ei .d s X s 2 1B B 4M 1.1 / C B M M 0.8 -2 -1 0 1 2 3 Log Body Mass (kg) Fig. 1. Maximum longevity as a function of body mass in the primates. Regression line: Log (maximum longevity) = 1.238 + ,227dog (body mass). 9 = 0.562, +.x = 0.140. L, Lorimidea; M, Lemuroidea; T, Tarsioidea; B, Ceboidea; C, Cercopithecoidea;H, Hominoidea. and longevity should provide yet another cautionary example of the reason for this caveat, because larger data sets and more extensive analyses demonstrated that other organs, such as the liver, were correlated at least as closely with maximum longevity as brain size [Economos, 1980; Prothero & Jurgens, 19871. These results suggest that organ mass may be a better indicator of overall body size than body mass, and the relative magnitude of these correlations does not contain much insight into factors governing longevity. The reason for the general mammalian correlation between body size and longevity isn’t clear, but it is not due to the relation between body size and basal metabolic rate [Austad & Fischer, 19911. One current idea is that body size is a surrogate variable which reflects vulnerability to environmental hazards or mortality risk [Read & Harvey, 1989; Austad & Fischer, 19911, a factor which life history theory suggests is central to the evolution of life history characteristics [e.g., Stearns, 1976; Charlesworth, 19801. According to this idea, larger animals would, on average, be less susceptible to predators, and perhaps because of their lower metabolic rate more resistant to food or water shortages. Resistance to environmentally imposed mortality of this nature should lead to the evolution of retarded aging [Medawar, 1952; Williams, 1957; Hamilton, 19663. By this reasoning, small organisms that for some special reason exhibit reduced vulnerability to environmental hazard should also exhibit long-for-size life spans. Owing to their aerial habits, small, volant and gliding mammals (bats, dermopterans, gliding marsupials, and “flying” squirrels) and birds are likely to be at reduced risk from many forms of predation [Pomeroy, 19901 and should therefore show increased longevity. Gliding mammals, including marsupials, average 1.7 time the expected life span of nonflying eutherians [Austad & Fischer, 19911, Primate Longevity / 259 while birds live about 2.4 times the mass-specific life span of eutherian mammals [Lindstedt and Calder, 19761. Historically, considerations of broad interspecific patterns of longevity have led to insight into potential mechanisms through which aging might occur. The rate-of-living theory, though now virtually discarded by biogerontologists [e.g., McCarter et al., 1985; Finch, 1990; Rose, 19911, has stimulated very productive research into damaging by-products of normal metabolic processes. One such byproduct is the generation of highly reactive free oxygen radicals [Harman, 1962; Halliwell & Gutteridge, 19851, which damage an array of cellular components from membranes to nucleic acids. A second type of damaging by-product is the nonenzymatic attachment of glucose and other reducing sugars to proteins and nucleic acids to form nonreversible “advanced glycosylation end-products” (AGE’S) [Cerami, 1985; Monnier, 19891. What the diversity of mammalian life spans in the face of a variety of metabolic demands tells us is that a key to understanding mechanisms of aging is not simply understanding these how these damaging by-products are generated, but more importantly in understanding evolutionarily important mechanisms of defense against these by-products. The study of antioxidant defenses is well developed and currently a very active field of research [Halliwell & Gutteridge, 19851 and the study of anti-glycosylation defense systems is just beginning [Cerami et al., 1987; Monnier et al., 19911. One thing that the preceding analysis indicates is that primates such as humans are not the only animals with highly developed defenses against normal metabolic damage. Indeed instead of focusing virtually all research effort into anti-aging mechanisms on animal models such as laboratory rodents which have demonstrably poor defenses, scientists might fruitfully turn to other organisms with as effective, or even more effective, defense systems than humans. Unexploited animal models that would seemingly be ideal for this sort of research are bats and birds. In conclusion, our results show that primates’ exceptional longevity is not anomalous, but is a pattern that fits into the larger context of the evolution of mammalian life spans. The theories that 1) brain size is a determinant of maximum life span, 2) that life span has been increasing through evolutionary time among primates, and 3) that metabolic efficiency and life-time energy expenditure are determinants of maximum life span are not supported by our analyses. We suggest that the exceptional primate longevity may be better explained as a result of a generalized low mortality risk due to an arboreal habit and the propensity for social grouping. Because primates have been the subjects of intensive, long-term studies, they offer an ideal opportunity t o address some of the problems concerning mammalian life history evolution. ACKNOWLEDGMENTS This work was supported in part by grants to S.N.A. from the Henry Rosovsky Fund and Milton Fund of Harvard University and by U.S.N.I.H. grant AG0970001. We are grateful to C. Finch, J. Robinson, P. Waser, and R. Wrangham for helpful comments on the text. REFERENCES Anonymous. Longevity survey: Length of life of mammals in captivity at the London Zoo and Whipsnade Park. INTERNATIONAL ZOO YEARBOOK 11~288-299, 1960. Austad, S.N.; Fischer, K.E. Mammalian aging, metabolism, and ecology: Evidence from the bats and marsupials. JOURNAL OF GERONTOLOGY B47-B53,1991. 260 I Austad and Fischer Bowden, D.M.; Jones, M.L. Aging research in nonhuman primates. In AGING IN NONHUMAN PRIMATES. D.M. Bowden, ed. New York, Van Nostrand Reinhold, 1979. Calder, W.A. 111. SIZE, FUNCTION, AND LIFE HISTORY. Cambridge, MA: Harvard University Press, 1984. Cerami, A. Hypothesis: Glucose as a mediator of aging. JOURNAL OF THE AMERICAN GERIATRIC SOCIETY 33:626-634, 1985. Cerami, A,; Vlassara, H.; Brownlee, M. Glucose and aging. SCIENTIFIC AMERICAN 256(5):90-96, 1987. Charlesworth, B. EVOLUTION IN AGESTRUCTURED POPULATIONS. Cambridge, Cambridge University Press, 1980. Cleveland. W.S. Robust locally weighted regression ' and smoothing" scatterplots. JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION 74~8294336,1979. Cleveland, W.S. LOWESS: A program for smoothing scatterplots by robust locally weighted regression. AMERICAN STATISTICS 3354, 1981. Corbet, G.B.; Hill, J.E. A WORLD LIST OF MAMMALIAN SPECIES. 3rd Ed. New York, Facts on File Publications, 1991. Crandall, L.S. THE MANAGEMENT OF WILD MAMMALS IN CAPTIVITY. Chicago: University of Chicago Press, 1964. Crile, G.; Quiring, D.P. A record of the body weight and certain organ and gland weights of 3690 animals. OHIO JOURNAL OF SCIENCE 40~219-259,1940. Cutler, R.G. Evolution of human longevity and the genetic complexity governing aging rate. PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCE USA 72:4664-4668,197 5. Cutler, R.G. Evolution of longevity in primates. JOURNAL OF HUMAN EVOLUTION 5:169-202, 1976. Cutler, R.G. Evolutionary biology of senescence. Pp.311-360 in THE BIOLOGY OF AGING, 2nd Ed. J.A. Behnke; C.E. Finch; G.B. Moment, eds. New York, Plenum Press, 1978. Economas, A.C. Brain-lifespan conjecture: a reevaluation of the evidence. GERONTOLOGY 26~82-89,1980. Eisenberg, J.F. THE MAMMALIAN RADIATIONS. Chicago: University of Chicago Press, 1981. Finch, C.E. LONGEVITY, SENESCENCE, AND THE GENOME. Chicago, University of Chicago Press, 1990. Flower, S.S. Contributions to our knowledge of the duration of life in vertebrate animals. V. mammals. PROCEEDINGS OF THE ZOOLOGICAL SOCIETY OF LONDON, 1931. Hamilton, W.D. The moulding of senescence by natural selection. JOURNAL OF THEORETICAL BIOLOGY 12:12-45,1966. Halliwell, B.; Gutteridge, J.M.C. FREE RADICALS IN BIOLOGY AND MEDICINE. Oxford, Clarendon Press, 1985. Haltenmorth, T.; Diller, H. MAMMALS OF AFRICA. London, Collins, 1977. Harman, D. Role of free radicals in mutation, cancer, aging and maintenance of life. RADIATION RESEARCH 16:752-763, 1962. Harvey, P.H.; Martin, R.D.; Clutton-Brock, T.H. Life histories in comparative perspective. In PRIMATE SOCIETIES. B.B. Smuts, D.L. Cheney, R.M. Seyfarth, R.W. Wrangham, T.T. Struhsaker, eds., Chicago, University of Chicago Press, 1986. Hofman, M.A. Energy metabolism, brain size and longevity in mammals. QUARTERLY REVIEW OF BIOLOGY 58:495512,1983. Jerison, H.J. EVOLUTION OF THE BRAIN AND INTELLIGENCE. New York, Academic Press, 1973. Jones, M.L. Longevity of mammals in captivity. ZOOLOGISCHE GARTEN 52:113128,1982. Jurgens, K.D.; Prothero, J. Scaling of maximal lifespan in bats. COMPARATIVE BIOCHEMISTRY AND PHYSIOLOGY. 88A:361-367,1987. Koop, B.F.; Tagle, D.A.; Goodman, M.; Slightom, J.L. A molecular view of primate phylogeny and important systematic and evolutionary questions. MOLECULAR BIOLOGY AND EVOLUTION 6580-612, 1989. Lindstedt, S.L.; Calder, W.A. Body size and longevity in birds. CONDOR 78:91-94, 1976. Lindstedt, S.L.; Calder, W.A. Body size, physiological time, and longevity of homeothermic vertebrates. QUARTERLY REVIEW OF BIOLOGY 56:l-16,1981. Lovejoy, C.O. The origin of man. SCIENCE 211:341-350,1981. McCarter, R.; Masoro, E.J.; Yu, B.P. Does food restriction retard aging by reducing the metabolic rate? AMERICAN JOURNAL OF PHYSIOLOGY 248:E488-E490, 1985. McNab, B. Complications inherent in scaling the bzsal rate of metabolism in mammals. QUARTERLY REVIEW OF BIOLOGY 63~25-63,1988. Mallouk, R.S. Longevity in vertebrates is proportional to relative brain weight. FEDERAL PROCEEDINGS 34:21022103, 1975. Medawar, P.B. AN UNSOLVED PROBLEM IN BIOLOGY. H.K. Lewis, London, 1952. Monnier, V.M. Toward a Maillard Reaction Primate Longevity I 261 theory of aging. Pp. 1-22 in THE MAILLARD REACTION IN AGING, DIABETES, AND NUTRITION. J.W. Baynes; V.M. Monnier, eds. New York, Alan R. Liss, 1989. Monnier, V.M.; Sell, D.R.; Nagaraj, R.H.; Miyata, S. Mechanisms of protection against damage mediated by the Maillard reaction in aging. GERONTOLOGY 37: 152-165, 1991. Napier, J.R.; Napier, P.$. THE NATURAL HISTORY OF PRIMATES. Cambridge, MA, MIT Press, 1985. Nowak, R.M.; Paradiso, J.L. WALKERS MAMMALS OF THE WORLD, 4th Ed. Baltimore, Johns Hopkins University Press, 1983. Pagel, M.D.; Harvey, P.H. Recent developments in the analysis of comparative data. QUARTERLY REVIEW OF BIOLOGY 63: 413-440,1988. Pearl, R. THE RATE OF LIVING. New York and London, Knopf, 1928. Pirlot, P.; Stephan, H. Encephalization in chiroptera. CANADIAN JOURNAL OF ZOOLOGY 48:433-444,1970. Pomeroy, D. Why fly? The possible benefits for lower mortality. BIOLOGICAL JOURNAL OF THE LINNEAN SOCIETY 40: 53-65, 1990. Prothero, J.; Jurgens, K.D. Scaling of maximal lifespan in mammals; a review. Pp. 49-74 in EVOLUTION OF LONGEVITY IN ANIMALS. A.D. Woodhead; K.H. Thompson, eds. New York, Plenum Press, 1987. Radinskv. L. Some cautionary notes on making inf6rences about relative brain size. Pp. 29-37 in PRIMATE BRAIN EVOLUTION. E. Armstrong; D. Falk, eds. Plenum Press, New York, 1982. Read, A.F.; Harvey, P.H. Life history differ- ences among the eutherian radiations. JOURNAL OF ZOOLOGY 219~329-353, 1989. Rice, W.R. Analyzing tables of statistical tests. EVOLUTION 43:223-225, 1989. Rose, M.R. EVOLUTIONARY BIOLOGY OF AGING. New York and Oxford, Oxford University Press, 1991. Rubner, M. DAS PROBLEM DER LEBENSDAUER. Berlin, Germany, 1908. Sacher, G.A. Relation of lifespan to brain weight and body weight in mammals. CJBA FOUNDATION COLLOQUIA 5: 115-141,1959. Sacher, G.A. Maturation and longevity in relation to cranial capacity in hominid evolution. Pp. 417-441 in PRIMATE FUNCTIONAL MORPHOLOGY AND EVOLUTION. R.H. Tuttle, ed. The Hague, Mouton, 1975. Schultz, S. THE LIFE OF PRIMATES. New York, Universe Books, 1969. Sohal, R.S. The rate of living theory: A contemporary interpretation. in INSECT AGING. K.-G. Collatz, R.S. Sohal, eds. Berlin, Springer-Verlag, 1986. Stahl, W.R. Similarity and dimensional methods in biology. SCIENCE 137:205212,1962. Steams, S.C. Life-history tactics: A review of the ideas. QUARTERLY REVIEW OF BIOLOGY 51:3-47, 1976. Weiss, K.M. Evolutionary perspectives on human aging, Pp. 25-58 in OTHER WAYS OF GROWING OLD. P.T. Amoss; S. Harrell, eds. Stanford, CA, Stanford University Press, 1981. Wilkinson, L. SYSTAT: THE SYSTEM FOR STATISTICS, Evanston, IL, SYSTAT, Inc., 1990. Williams, G.C. Pleiotropy, natural selection and the evolution of senescence. EVOLUTION 11:398-411,1957.