Caroline Ross And Kate E Jones

Introduction

A thought experiment: take two animals, one male and one female. Provide their offspring with unlimited food and allow them to breed. If animals die only of old age, how many animals will you have after 2 years, 5 years, 10 years, 20 years or 50 years? Figure 4.1 shows the results of this experiment using the intrinsic rate of population increase (rm, see Table 4.1, p. 80, for a definition) for six primate species. It is clear from this that some animals can increase their population size more rapidly than others. The bushbaby (Galago moholi) population contains over 2000 females after only 12 years and 3600 million (3.6 x 1014) individuals after 50 years, whereas the gorilla population contains only 29 females after 50 years.

The intrinsic rate of population increase depends on three variables: age at first reproduction (AR), birth rate (b) and age at last reproduction (L). Birth rate has an important influence on the rate of population increase, which is illustrated by the examples given of Macaca silenus and Macaca sylvanus the females of which start breeding at about five years of age. However, in Macaca sylvanus, birth rate is higher, leading to a higher intrinsic rate of population increase. Animals that start breeding at an early age also show a higher intrinsic rate of population increase than do those that start breeding later. This has a knock-on effect over the generations, as a female that breeds early produces female infants that also breed early, thus increasing the total number of females that are producing infants. For example, both Macaca sylvanus and Erythrocebus patas breed annually, but E. patas reaches maturity nearly two years before M. sylvanus. Over time, this difference in AR has a large effect on the population size, with the E. patas population taking 33 years to grow to 2000 females and the M. sylvanus population taking 41 years (Fig. 4.1). In contrast, variation in the age at last reproduction makes little difference to the intrinsic rate of population increase. Titus (Callicebus moloch) and patas monkeys (E. patas) both start breeding at three years of age and produce an infant annually, but the patas monkeys live far longer than the titis (21 as

Fig. 4.1 Population growth for six primate species, assuming the population is started with a single adult female and the sex ratio is 1: 1. The rate of population growth is calculated by assuming that there is no mortality until the age at last reproduction is reached. Curves were calculated using the following life history variables: Callicebus moloch (a = 3.0 years, b = 1.0 offspring/year, L = 12.0 years rm = 0.27), Erythrocebus patas (3.0,1.0, 21.6, 0.26), Galago moholi (1.0, 3.2, 16.5, 0.96), Gorilla gorilla (10.0, 0.26, 50.0, 0.07), Macaca silenus (4.9, 0.72, 38.0, 0.17), Macaca sylvanus (4.8, 1.0, 22.0, 0.21).

Fig. 4.1 Population growth for six primate species, assuming the population is started with a single adult female and the sex ratio is 1: 1. The rate of population growth is calculated by assuming that there is no mortality until the age at last reproduction is reached. Curves were calculated using the following life history variables: Callicebus moloch (a = 3.0 years, b = 1.0 offspring/year, L = 12.0 years rm = 0.27), Erythrocebus patas (3.0,1.0, 21.6, 0.26), Galago moholi (1.0, 3.2, 16.5, 0.96), Gorilla gorilla (10.0, 0.26, 50.0, 0.07), Macaca silenus (4.9, 0.72, 38.0, 0.17), Macaca sylvanus (4.8, 1.0, 22.0, 0.21).

compared to 12 years maximum recorded longevity). Despite this, Figure 4.1 shows that the population growth of the patas is only slightly more rapid than that of the titis, with the latter taking only two years more to reach a female population of 2000 (35 years compared to 33 years).

Life history theory seeks to explain this variation, asking not only, 'Why do gorillas breed slowly and bushbabies breed rapidly?' but also, 'Why do the reproductive rates of primates differ from those of other animals?'. This chapter examines these questions, both by reviewing previous studies and by presenting the results of some new analyses. It starts with a brief overview of life history theory and then explores the primate data in the light of these ideas.

Reproductive rates and life history theory

Design constraints, body size and phylogeny

Variation in reproductive rates is correlated with body size. Generally speaking, large animals take longer to reach maturity and breed more slowly than do small animals. There are usually allometric relationships between life history variables and body weight (e.g. Peters, 1983; Calder, 1984), many of which have been described in primates (e.g. Kirkwood, 1985; Harvey, Martin and Clutton-Brock, 1987; Ross, 1988; Lee, Majluf and Gordon, 1991). For many parameters, the association between body weight and a life history parameter may be very strong; e.g. body weight variation in primates explains about 80% of the variation in female age at first reproduction. Explanations for these high correlations between life history parameters and body size fall into two categories: those that link body size with life history parameters because of reasons of 'design constraint', and those that suggest an 'adaptive link'.

The 'design constraint' model suggests that body size acts as a constraint on life history evolution, so that selection for body size 'carries along' life history characters with it. Life-history parameters are therefore constrained within certain limits by the size of the organism (e.g. Western, 1979; Western and Ssemakula, 1982). Although design constraints will be imposed by an animal's size, other aspects of its physiology and anatomy will also constrain the evolution of reproductive rates within certain limits. For example, eutherian mammalian reproduction does not allow the evolution of egg laying and birds are unlikely to evolve asexual reproduction. Constraints such as these are likely to be shared by closely related organisms that have a similar anatomy and physiology and these related species are therefore likely to have similar life histories.

Alternatively, the correlation between body size and life history parameters may be adaptive as the body size of an organism may influence the way in which it experiences its environment (Pianka, 1970). For example, a large-bodied animal will not be as threatened by environmental fluctuations as will a smaller animal, all other things being equal. In this way, body size itself can influence life history characters.

Despite the strong correlations between body size and reproductive rates, neither the design constraint model nor an adaptive link between body size and reproductive rates can fully explain life history variation. As noted by Pagel and Harvey (1993), this is for two basic reasons. Firstly, it does not explain why body size should vary: if being big must lead to slow reproduction, why should animals waste time and resources growing large? Secondly, it does not explain the variation in life histories that is not correlated with body weight. This second point is important as several studies have shown that life history traits co-vary predictably, so that animals fall along a continuum from those with high mortality, short lifespans, high birth rates and fast development to those with low mortality, long lifespans, low birth rates and slow development, even if the influence of body size is removed (e.g. Stearns, 1983; Promislow and Harvey, 1990). Numerous models have been proposed that try to explain these observations. These models assume that natural selection will act to optimise life histories, so as to maximise the lifetime reproductive success of individuals (i.e. they assume that life histories are not only varying due to 'design constraints') and are similar in that they recognise the importance of trade-offs in life history evolution.

Trade-offs and primate reproduction

If breeding had no costs to the animal, it would be selected to breed as early as possible, as rapidly as possible and for as long as possible so as to maximise its contribution to future generations. Both common sense and a large body of evidence (Boyce, 1988; Stearns, 1992) tell us that producing offspring is not free of cost, and resources devoted to an infant are resources that cannot be used elsewhere. Selection for large infant size will result in fewer infants; the limited resources animals have for reproduction means that they must trade numbers of infants against infant size. Similarly, trade-offs may be made between allocating resources to the production of young or into parental survival. If investing less than the maximum possible in one reproductive attempt increases the survival of the parent, a lowering of reproductive effort per reproductive attempt may result in greater numbers of offspring being produced in an individual's lifetime.

As primates produce only one or two young per litter, there is little scope for trading infant numbers against infant quality. The evolutionary 'decision' to trade the disadvantages of producing few young against the advantages of producing large, 'high-quality' young was probably made early in the line of primate ancestry. Primates do, however, produce twins, some species typically do so and even those that typically produce singletons may produce twins occasionally. Within species, there is a considerable amount of evidence to suggest that an increase in litter size results in smaller young being produced (e.g. humans: Wilson, 1979). There is also evidence of trade-offs when the comparative evidence is considered, with a multiple regression showing that both body weight (W) and litter size contribute to the variation in primate neonatal weight (N) (n = 82 species, 80 contrasts, r2 = 0.58; W: coefficient = 0.613, p < 0.0001; N: coefficient = — 0.266, p = 0.0005, see below for methods).

Several intraspecific studies of primates have tried to find evidence for trade-offs between other reproductive parameters, particularly maternal survival versus fecundity and early reproduction versus fecundity. The results of these studies are variable, with only limited evidence that such trade-offs do occur. Some evidence that fecundity may trade-off against maternal survival comes from Altmann, Hausfater and Altmann's (1988) study of savannah baboons (Papio cynocephalus) in which females caring for infants have higher mortality rates than those without infants. A cost of early reproduction was shown in rhesus macaques (Macaca mulatta) on Cayo Santiago, where young primaparous females ( < 4 years) suffer greater mortality of their infants than older primaparous females ( > 4 years) (Sade, 1990). Younger females do tend to have a longer interbirth interval to their second infant than do older females, suggesting that there is a trade-off between early reproduction and fecundity (although the result is not statistically significant). However, Sade (1990) found no difference in maternal survival between young and old primaparous rhesus macaques.

One reason for the differences found between studies is that simple trade-offs between two variables are often likely to be complicated by variation in a third variable (e.g. animals' rank or age), so that trade-offs may not always be observed easily. Despite this, the available evidence does seem to show that primates have the flexibility to trade-off reproductive variables against one another. This is important as it indicates that intraspecific variation in life histories is likely when individuals of a species are experiencing different environments that select for different balances of possible trade-offs. What then, are the selective pressures whose variation brings about diversity in life history strategies?

r and K selection theory and age-specific mortality

The theory of r and K selection was originally proposed by MacArthur and Wilson (1967) to explain the processes of island colonisation and was extended to include other situations in which mortality is primarily density independent (Pianka, 1970). Where a population is expanding into unoccupied habitats, or into an empty niche caused by a 'population crash', an individual with a high intrinsic rate of population increase will rapidly fill the 'space' with its descendants, faster than slower breeding survivors can. Hence, such conditions will select for a high intrinsic rate of population increase and 'r selection' occurs. Alternatively, when a population is at the carrying capacity (K) of its habitat, MacArthur and Wilson reasoned that there would be a high incidence of density dependent mortality, i.e. mortality due to competition for limiting resources rather than stochastic events. Hence, selection would favour individuals that use the available resources most efficiently, thus maximising the carrying capacity. A K-selected population will be expected to have a higher competitive ability and, thus, a lower reproductive effort than will an r-selected species.

K-selected populations will also be expected to produce a smaller number of young per litter so that their parental investment per offspring can be maximised, conferring increased competitive ability on the offspring. A K-selected species will therefore be characterised by having a later age at first reproduction, longer development time, lower birth rate and fewer, larger young than a more r-selected species.

r and K selection theory was widely used in early studies of life history variation to explain the observed fast-slow continuum, with slow-growing, slow-reproducing and long-lived animals being assumed to be K selected and fast-growing, fast-reproducing and short-lived animals r selected. However, the theory has now been largely replaced by more recent models that place emphasis on the effects that mortality has on individuals of different ages, rather than on the causes of mortality (i.e. density dependent or density independent). This change has mainly arisen because a large number of studies has shown that the r/K model does not explain the pattern of life histories seen in many taxa and that artificial selection experiments do not produce the results predicted by r/K theory (Stearns, 1977, 1992). This lack of correspondence between prediction and observation suggest that there are deficiencies in the r/K model.

Another major problem is that the terms 'density-dependent' and 'density-independent' mortality try to describe the reasons why mortality occurs but they do not describe who mortality affects or when it acts. In many natural populations, mortality is not evenly distributed amongst animals of all ages. In mammals, the usual pattern is that the very young and very old are more likely to die than are adults in their prime. Models that include information on the way in which mortality patterns are distributed have been shown to explain life history variation far more comprehensively than has the r/K model. Despite the replacement of r/K theory by other models, there is still general recognition that selection may be expected to act differently when mortality (particularly juvenile mortality) is primarily density dependent rather than density independent (Boyce, 1988; Stearns, 1992). In particular, the prediction that density-dependent selection is likely to lead to a decreased reproductive effort and an increased need for competitive infants may still be incorporated into other models (Purvis and Harvey, 1995). One model that does not incorporate density-independent mortality has, nevertheless, received some support from empirical comparative evidence.

Charnov's model

Chamov (1993) presents a model of mammalian life history evolution. The model explains life history allometry by assuming that the age at maturity is determined by adult mortality rates, which are in turn determined by the environment. When mortality rates are high, animals are expected to mature rapidly, so as to minimise their chances of dying, and thus maximise their lifetime reproductive success. Charnov assumes that growth continues up to the age of maturity and then stops, with resources then being diverted to reproduction. Thus, the model predicts that animals that mature late will have a larger body size than those that mature early, and that this will influence fecundity. In a stable population, fecundity is balanced by mortality and hence fecundity will be expected to be positively correlated with juvenile mortality rates.

Charnov's model makes several predictions about the type of relationships one would expect to find between body weight, mortality rates and reproductive rates. In some cases, studies of mammals appear to support this model (Charnov, 1993; Purvis and Harvey, 1995), although Kozlowski and Weiner (1997) suggest that the model may only be appropriate in certain limited situations. One problem with Charnov's model is that it predicts that relative size at independence (the ratio of weaning weight to adult weight, S) is a constant, something that is not always supported by the data (Purvis and Harvey, 1995). However, Charnov's predicted value of S = 0.33 is found for primates (Lee et al., 1991; Charnov, 1993) and his model may therefore be appropriate for use in this group.

Charnov and Berrigan (1993) and Charnov (1993) pointed out that primates are unusual amongst mammals in having an allometric exponent for a (a might be female juvenile period length or age at first reproduction in Charnov (1993); it is unclear which is used in the analysis) against body weight that is greater than 0.25 (calculated from species values). Charnov (1993) has linked this difference in the scaling of a to primates having an unusual form of the basic growth function that relates body weight (W) to growth rate (DW/dT): DW/dT = AW°'75. The value of the constant A is approximately 1 for most mammals, but for primates Charnov calculates the value as 0.42. This suggests that primates differ from other mammals in having very low production rates, so that they take longer to grow both themselves and their infants than do other mammals, something that may also account for both the late maturation and the slow breeding rates of primates. Some of Charnov's predictions are tested below for a larger primate data set than originally used, and some possible causes and consequences of the low growth rate of primates are discussed.

Table 4.1. Description of variables used in analyses

Parameter

Symbol Definition

Adult female body weight

Age at first reproduction

Juvenile period

Pregnancy And Childbirth

Pregnancy And Childbirth

If Pregnancy Is Something That Frightens You, It's Time To Convert Your Fear Into Joy. Ready To Give Birth To A Child? Is The New Status Hitting Your State Of Mind? Are You Still Scared To Undergo All The Pain That Your Best Friend Underwent Just A Few Days Back? Not Convinced With The Answers Given By The Experts?

Get My Free Ebook


Post a comment