Socio-cultural similarity with host population rather than ecological similarity predicts success and failure of human migrations.

Demographers argue that human migration patterns are shaped by people moving to better environments. More recently, however, evolutionary theorists have argued that people move to similar environments to which they are culturally adapted. While previous studies analysing which factors affect migration patterns have focused almost exclusively on successful migrations, here we take advantage of a natural experiment during World War II in which an entire population was forcibly displaced but were then allowed to return home to compare successful with unsuccessful migrations. We test two competing hypotheses: (1) individuals who relocate to environments that are superior to their place of origin will be more likely to remain-The Better Environment Hypothesis or (2) individuals who relocate to environments that are similar to their place of origin will be more likely to remain-The Similar Environment Hypothesis. Using detailed records recording the social, cultural, linguistic and ecological conditions of the origin and destination locations, we find that cultural similarity (e.g. linguistic similarity and marrying within one's own minority ethnic group)-rather than ecological differences-are the best predictors of successful migrations. These results suggest that social relationships, empowered by cultural similarity with the host population, play a critical role in successful migrations and provide limited support for the similar environment hypothesis. Overall, these results demonstrate the importance of comparing unsuccessful with successful migrations in efforts understand the engines of human dispersal and suggest that the primary obstacles to human migrations and successful range expansion are sociocultural rather than ecological.

Is bigger better? The relationship between size and reproduction in female Asian elephants.

The limited availability of resources is predicted to impose trade-offs between growth, reproduction and self-maintenance in animals. However, although some studies have shown that early reproduction suppresses growth, reproduction positively correlates with size in others. We use detailed records from a large population of semi-captive elephants in Myanmar to assess the relationships between size (height and weight), reproduction and survival in female Asian elephants, a species characterized by slow, costly life history. Although female height gain during the growth period overlapped little with reproductive onset in the population, there was large variation in age at first reproduction and only 81% of final weight had been reached by peak age of reproduction at the population level (19 years). Those females beginning reproduction early tended to be taller and lighter later in life, although these trends were not significant. We found that taller females were more likely to have reproduced by a given age, but such effects diminished with age, suggesting there may be a size threshold to reproduction which is especially important in young females. Because size was not linked with female survival during reproductive ages, the diminishing effect of height on reproduction with age is unlikely to be due to biased survival of larger females. We conclude that although reproduction may not always impose significant costs on growth, height may be a limiting factor to reproduction in young female Asian elephants, which could have important implications considering their birth rates are low and peak reproduction is young - 19 years in this population.

The effects of resource availability and the demographic transition on the genetic correlation between number of children and grandchildren in humans.

Studies of evolutionary change require an estimate of fitness, and lifetime reproductive success is widely used for this purpose. However, many species face a trade-off between the number and quality of offspring and in such cases number of grandoffspring may better represent the genetic contribution to future generations. Here, we apply quantitative genetic methods to a genealogical data set on humans from Finland to address how the genetic correlation between number of children and grandchildren is influenced by the severity of the trade-off between offspring quality and quantity, as estimated by different levels of resource access among individuals in the population. Further, we compare the genetic correlation before and after the demographic transition to low mortality and fertility rates. The genetic correlation was consistently high (0.79-0.92) with the strongest correlations occurring in individuals with higher access to resources and before the demographic transition, and a tendency for lower correlations in resource poor individuals and after the transition. These results indicate that number of grandoffspring is a slightly better predictor of long-term genetic fitness than number of offspring in a human population across a range of environmental conditions, and more generally, that patterns of resource availability need to be taken into account when estimating genetic covariances with fitness.

Sibship effects on dispersal behaviour in a pre-industrial human population.

Understanding dispersal behaviour and its determinants is critical for studies on life-history maximizing strategies. Although many studies have investigated the causes of dispersal, few have focused on the importance of sibship, despite that sibling interactions are predicted to lead to intrafamilial differences in dispersal patterns. Using a large demographic data set from pre-industrial Finland (n = 9000), we tested whether the sex-specific probability of dispersal depended on the presence of same-sex or opposite-sex elder siblings who can both compete and cooperate in the family. Overall, following our predictions, the presence of same-sex elder siblings increased the probability of dispersal from natal population for both sexes, whereas the number of opposite-sex siblings had less influence. Among males, dispersal was strongly linked to access to land resources. Female dispersal was mainly associated with competition over availability of mates but likely mediated by competition over access to wealthy mates rather mate availability per se. Besides ecological constraints, sibling interactions are strongly linked with dispersal decisions and need to be better considered in the studies on the evolution of family dynamics and fitness maximizing strategies in humans and other species.

Early reproductive investment, senescence and lifetime reproductive success in female Asian elephants.

The evolutionary theory of senescence posits that as the probability of extrinsic mortality increases with age, selection should favour early-life over late-life reproduction. Studies on natural vertebrate populations show early reproduction may impair later-life performance, but the consequences for lifetime fitness have rarely been determined, and little is known of whether similar patterns apply to mammals which typically live for several decades. We used a longitudinal dataset on Asian elephants (Elephas maximus) to investigate associations between early-life reproduction and female age-specific survival, fecundity and offspring survival to independence, as well as lifetime breeding success (lifetime number of calves produced). Females showed low fecundity following sexual maturity, followed by a rapid increase to a peak at age 19 and a subsequent decline. High early life reproductive output (before the peak of performance) was positively associated with subsequent age-specific fecundity and offspring survival, but significantly impaired a female's own later-life survival. Despite the negative effects of early reproduction on late-life survival, early reproduction is under positive selection through a positive association with lifetime breeding success. Our results suggest a trade-off between early reproduction and later survival which is maintained by strong selection for high early fecundity, and thus support the prediction from life history theory that high investment in reproductive success in early life is favoured by selection through lifetime fitness despite costs to later-life survival. That maternal survival in elephants depends on previous reproductive investment also has implications for the success of (semi-)captive breeding programmes of this endangered species.

Selection on male longevity in a monogamous human population: late-life survival brings no additional grandchildren.

Humans are exceptionally long-lived for mammals of their size. In men, lifespan is hypothesized to evolve from benefits of reproduction throughout adult life. We use multi-generational data from pre-industrial Finland, where remarriage was possible only after spousal death, to test selection pressures on male longevity in four monogamous populations. Men showed several behaviours consistent with attempting to accrue direct fitness throughout adult life and sired more children in their lifetimes if they lost their first wife and remarried. However, remarriage did not increase grandchild production because it compromised the success of motherless first-marriage offspring. Overall, grandchild production was not improved by living beyond 51 years and was reduced by living beyond 65. Our results highlight the importance of using grandchild production to understand selection on human life-history traits. We conclude that selection for (or enforcement of) lifetime monogamy will select for earlier reproductive investment and against increased lifespan in men.

Influence of handler relationships and experience on health parameters, glucocorticoid responses and behaviour of semi-captive Asian elephants.

Declining wild populations combined with accumulating captive populations of e.g. livestock, pets, draught and zoo animals have resulted in some threatened species with substantial proportions of their populations in captivity. The interactions animals have with humans in captivity depend on handler familiarity and relationship quality and can affect animal health, growth and reproduction with consequences for the success of conservation programmes. However, assessments of how specific human-animal relationships affect a range of physiological and behavioural outcomes are rare. Here, we studied semi-captive Asian elephants with detailed records of elephant-handler (mahout) relationships and veterinary management, allowing assessment of multiple welfare indicators in relation to specific mahout-elephant relationship lengths and mahout experience. These included measures of physiological stress (faecal glucocorticoid metabolite [FGM], heterophil:lymphocyte ratio [H:L]), muscle damage (creatine kinase [CK]), immunological health (total white blood cell count [TWBC]) and behaviour (response to mahout verbal commands). We found no evidence that FGM or H:L related to aspects of the mahout-elephant relationship. Longer overall mahout experience (i.e. years of being a mahout) was linked to increased muscle damage and inflammation, but the lengths of specific mahout-elephant relationships were inversely associated with muscle damage in working-age elephants. Elephants responded more to familiar mahouts in behavioural tasks and faster to mahouts they had known for longer. In summary, our results found little evidence that the mahout-elephant relationship affects physiological stress in this population based on FGM and H:L, but mahout experience and relationships were linked to other physiological responses (CK, TWBC), and elephants require behavioural adjustment periods following mahout changes.

Selection for long lifespan in men: benefits of grandfathering?

Life-history theory suggests that individuals should live until their reproductive potential declines, and the lifespan of human men is consistent with this idea. However, because women can live long after menopause and this prolonged post-reproductive life can be explained, in part, by the fitness enhancing effects of grandmothering, an alternative hypothesis is that male lifespan is influenced by the potential to gain fitness through grandfathering. Here we investigate whether men, who could not gain fitness through reproduction after their wife's menopause (i.e. married only once), enhanced their fitness through grandfathering in historical Finns. Father presence was associated with reductions in offspring age at first reproduction and birth intervals, but generally not increases in reproductive tenure lengths. Father presence had little influence on offspring lifetime fecundity and no influence on offspring lifetime reproductive success. Overall, in contrast to our results for women in the same population, men do not gain extra fitness (i.e. more grandchildren) through grandfathering. Our results suggest that if evidence for a 'grandfather' hypothesis is lacking in a monogamous society, then its general importance in shaping male lifespan during our more promiscuous evolutionary past is likely to be negligible.

Reproductive investment in pre-industrial humans: the consequences of offspring number, gender and survival.

The number and gender of offspring produced in a current reproductive event can affect a mother's future reproductive investment and success. I studied the subsequent reproductive outcome of pre-industrial (1752-1850) Finnish mothers producing twins versus singletons of differing gender. I predicted that giving birth to and raising twins instead of singletons, and males instead of females, would incur a greater reproductive effort and, hence, lead to larger future reproductive costs for mothers. I compared the mothers' likelihood of reproducing again in the future, their time to next reproduction and the gender and survival of their next offspring. I found that mothers who produced twins were more likely to stop breeding or breed unsuccessfully in the future as compared with women of a similar age and reproductive history who produced a same-gender singleton child. As predicted, the survival and gender of the offspring produced modified the costs of reproduction for the mothers. Giving birth to and raising males generally appeared to be the most expensive strategy, but this effect was only detected in mothers who produced twins and, thus, suffering from higher overall costs of reproduction.

Adaptive sex ratio variation in pre-industrial human (Homo sapiens) populations?

Sex allocation theory predicts that in a population with a biased operational sex ratio (OSR), parents will increase their fitness by adjusting the sex ratio of their progeny towards the rarer sex, until OSR has reached a level where the overproduction of either sex no longer increases a parent's probability of having grandchildren. Furthermore, in a monogamous mating system, a biased OSR is expected to lead to lowered mean fecundity among individuals of the more abundant sex. We studied the influence of OSR on the sex ratio of newborns and on the population birth rate using an extensive data set (n = 14,420 births) from pre-industrial (1775-1850) Finland. The overall effect of current OSR on sex ratio at birth was significant, and in the majority of the 21 parishes included in this study, more sons were produced when males were rarer than females. This suggests that humans adjusted the sex ratio of their offspring in response to the local OSR to maximize the reproductive success of their progeny. Birth rate and, presumably, also population growth rate increased when the sex ratio (males:females) among reproductive age classes approached equality. However, the strength of these patterns varied across the parishes, suggesting that factors other than OSR (e.g. socioeconomic or environmental factors may also have influenced the sex ratio at birth and the birth rate.

Induced abortion ratio in modern Sweden falls with age, but rises again before menopause.

A woman's reproductive value decreases over her reproductive life span and it is therefore predicted that the likelihood of termination of investment in a child decreases with increasing age. An eventual increase in termination ratio in the oldest age groups, as is often found in abortion statistics, could depend on older women on average having larger families rather than on age per se. We used data on abortions and births in Sweden during 1994 to investigate how abortion ratio is related to age and parity of women. We found that age-specific abortion ratio is U-shaped (i.e. that it is highest for the youngest and for the oldest age groups) in each parity class from zero to four children but that age-dependence breaks down in higher parity classes (5, >/=6). Thus, for each of the parity classes 0-4, the incidence of abortion decreases with age up to a point, but increases again as women approach menopause. This late increase in induced abortion ratio seems to depend on age per se. The data indicate that abortion ratio is an inverse function of fertility, and that investment in new reproduction gradually decreases as a woman approaches menopause. Assuming grandmothering as an important driving force in human life history evolution, such a pattern might indicate that the transition from behavioural investment in one's own children to one's grandchildren is a gradual process similar to the decline in ovarian function.

Menopause: why does fertility end before life?

Menopause is associated with an ultimate cessation of child-bearing potential. Medical research on menopause focuses mostly on the underlying physiological changes associated with menopause. By contrast, evolutionary biologists are interested in understanding why women lose their potential to reproduce before the end of their lives. Evolution by natural selection predicts that the behaviors that we observe today are products of generations of selection on the genes that govern those behaviors. Since one would expect an individual reproducing throughout its life to produce more offspring than an individual stopping early, one would seldom expect genes for menopause to be selected for during our evolutionary past. This article discusses how menopause and prolonged lifespan might be explained by evolutionary theory, and highlights some angles for future research.