Age, Longevity and Life-History Trade-Offs in the Collared Flycatcher (Ficedula Albicollis)

Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology 288

Age, Longevity and Life-History Trade-Offs in the Collared Flycatcher (Ficedula albicollis)

JOANNA SENDECKA

ACTA UNIVERSITATIS UPSALIENSIS ISSN 1651-6214 UPPSALA ISBN 978-91-554-6852-1 2007 urn:nbn:se:uu:diva-7787 ! ""! #"$"" % & % % '& &( )& * + &(

, - .( ""!( / /%01 ) 02%% & 3 & 4 5( ( 66( 77 ( ( 8,9: !60#0;;70<6; 0#(

% % ( 1 * 0 & % % ( )& = % % & % % 0 & & * & & % - * ( 3 % & 4 5 = & * & % 0 & > & % & * % %( )& & % 0 % & & & ( ? & * & % %0& 0 %% & * & ( )& % % & % % % *&& % * @ & ( 2 & & A % %% ( 1 * && @ & && & % 0( )& & %% & * @ & 0 ( %% @ & & ( * & * 0 && @ ( B % % && % ( )& & && @ *& 0 ( B %0 *& & % %0 4 5 & * ( )& & % & *&& & @ ( @ & %% 0 % %0& ( )&% - & % % %0&

(

% & 0 @ %% A %0& 0 %%

! " # $ $ # % $ # & '( # # $)*+,-. #

C . , - ""!

8,,: #<;#0< #7 8,9: !60#0;;70<6; 0# $ $$$ 0!!6! 4& $DD (-(D E F $ $$$ 0!!6!5 Lo! In the orient when the gracious light Lifts up his burning head, each under eye Doth homage to his new-appearing sight, Serving with looks his sacred majesty; And having climbed the steep-up heavenly hill, Resembling strong youth in his middle age, Yet mortal looks adore his beauty still, Attending on his golden pilgrimage: But when from highmost pitch, with weary car, Like feeble age, he reeleth from the day, The eyes, 'fore duteous, now converted are From his low tract, and look another way: So thou, thyself outgoing in thy noon Unlooked on diest unless thou get a son.

William Shakespeare The picture on the cover: ‘Åldrandets träd’ was painted by Hillevi Torell. List of papers

This thesis is based on the following five papers, which will be referred to in the text by their Roman numerals.

I Sendecka, J. & Gustafsson, L. Age-related changes in reproductive performance in Collared flycatchers early in life are shaped by individual quality and optimization of reproductive effort. (Manuscript).

II Sendecka, J., CichoĔ, M. & Gustafsson, L. (2007) Age- dependent reproductive costs and the role of breeding skills in the Collared flycatcher. Acta Zoologica 88, 95-100.

III Sendecka, J. & Bolund, E. Age-related benefits to female collared flycatchers from breeding on high quality territories. (Sub- mited).

IV CichoĔ, M., Sendecka, J. & Gustafsson, L. (2003) Age-related decline in humoral immune function in Collared Flycatchers. Journal of Evolutionary Biology 16, 1205-1210.

V Sendecka, J. & Gustafsson, L. Longevity and senescence in a wild bird population. (Manuscript).

Papers II and IV were reproduced with kind permission of the publisher, The Blackwell Synergy.

The order of authors reflects their relative contributions to the papers. I have personally written papers I, II, III and V, performed most of the statistical analyses for papers I, III and V and was deeply involved in the planning of all the experiments and analyses, collection of the field data for all the papers, and lab analyses for paper IV. In all papers coauthors contributed intel- lectually in planning the experiments, final stages of manuscript preparation and data preparation (III).

Contents

Introduction...... 9 Collared flycatchers on Gotland ...... 11 Age-dependent survival and reproduction ...... 13 Why do some individuals live longer than others? ...... 15 What causes age-related changes in reproductive success? ...... 16 Early increase in reproductive success ...... 16 Changes in reproductive success late in life ...... 21 Age-related cost of reproduction...... 24 Final conclusions and perspectives...... 26 Summary in Swedish (Sammanfattning) ...... 28 Summary in Polish (Streszczenie) ...... 34 Acknowledgements...... 39 References...... 42

Introduction

Longevity and age-related changes in organisms are fascinating topics and active areas of research. This is not only because they apply directly to humans, but also, and maybe even more so, because of the great variation in ageing patterns observed between taxa. For centuries people were consumed by their own ageing without considering the fact that animals also undergo similar ageing processes. Indeed, age-related changes in organisms’ functions, including the senescence process are widespread in nature (e.g. Rick- lefs & Scheuerlein, 2001). The basis for many of the major theories on ageing was provided by August Weismann (1834-1914). He proposed the Germplasm theory which distinguished two different lines within a multicel- lular organism: the germ line and the soma line. This separation provided an explanation for the fact that changes occur within the organism over time without being passed on to the next generation. Thus he provided the starting point for modern theories on ageing. As the field of biology developed, an increasing number of researchers started to address questions about age- related changes in the organism, most of them related either to organismal development or senescence. However, a greater understanding of how organisms can change with age and the possible mechanisms responsible for those changes was not possible until the development of modern genetics and evolution theories. In 1930 Ronald Aylmer Fisher (1890-1962) provided the basis for life-history theory in his book ‘The Genetical Theory of Natural Selection’. He proposed that organisms face decisions on the distribution of available resources between reproduction and self-maintenance. This as- sumption became a starting point for the theory of senescence, which identi- fies senescence as a result of underinvestment into self-maintenance. How- ever, most of the currently acknowledged hypotheses such as those address- ing age-related changes in reproductive success or senescence process were formulated later, in the second half of the 20th century (Medawar, 1952, Williams, 1957; Hamilton, 1966; Kirkwood, 1977; Charlesworth, 1980). Age-related changes in organismal functions are extremely common. Age-related patterns and signs of senescence in new species are described every day, and almost all the species have now being covered. Even single- celled, asexually reproducing metazoans which were traditionally believed to be immortal show signs of senescence, albeit at a very slow rate (Martinez & Leviton, 1992). Today age differences and ageing receive a lot of interest. The Web of Science displays over a hundred thousand scientific publications

9 containing the term ‘age’ published within only the last five years. Most of these results are related to animals. Why than study age and ageing? With all this research on the topic don’t we already know all there is to know about it? The answer is: absolutely not. Although there are many hypotheses re- garding the underlying causes of age-related patterns, most of these remain to be empirically tested. Studies on ageing in humans and captive mammals are leading the way shifting from external signs of senescence to mechanisms acting at the molecular and biochemical level (Arking, 1998). How- ever in field studies, purely age-related changes are often confounded by effects of the environment and interactions with other individuals, effects which are difficult to control for. Moreover, there is huge variation in age- related changes between individuals. Thus, there are large differences in age- related patterns both within and between species and patterns found in labo- ratory studies are not necessarily found in the wild. The aim of this study is explore the patterns of changes in survival probability, reproductive performance, reproductive costs and life-history decisions during the initial and final stages of a wild passerine bird’s lifetime. Although an organism’s functions change throughout its lifetime, the most pronounced changes occur at the initial and final stages of lifetime. The possible mechanisms underlying observed changes in survival rates, life-history trade-offs and reproduction will also be tested.

10 Collared flycatchers on Gotland

The collared flycatcher (Ficedula albicollis, Temminck, 1795) is a small (~13g), migratory passerine bird belonging to the Old World flycatcher family (Muscicapidae). Outside the breeding season both sexes possess a cryptic plumage of dull brown, with white patches on the wings and stripes on the tail. For the breeding season the males’ plumage changes to black with white patches on the wings. During this period males also possess the species- characteristic white collar and a white patch on the forehead, which is a sexually selected trait. Collared flycatchers breed in central and Eastern Europe but also on the Swedish Baltic islands of Gotland and Öland, where they form well established populations. In Eastern Europe and on Baltic islands the collared flycatcher cohabits with its sister species, the pied flycatcher (Ficedula hypoleuca, Pallas, 1764)(Fig 1). Sexual isolation between these two species is not complete and at the area of coexistence the two species form a hybrid zone (Haavie et al, 2004; Wiley, 2006; Svedin, 2006).

Figure 1. Distributions of the breeding areas of collared (black) and pied (light grey) flycatchers (dark grey area represents the area of sympatry). Baltic islands of Gotland and Öland are marked with a circle. Adapted from Haavie et al, 2004.

The collared flycatcher populations on Gotland and Öland are isolated from the main species range (Fig 1) and for this reason recapture and return rates of individuals are quite high (Gustafsson, 1986; Pärt & Gustafsson, 1989). Collared flycatchers arrive from their wintering areas in southern Africa to the breeding areas in late April to mid May. Females tend to arrive shortly after males and older individuals arrive earlier than young ones (Pärt

11 & Gustafsson, 1989; Mitrus, 1996). Males start occupying the territories just after arrival. Females choose mates on the basis of the territory he holds (Alatalo et al, 1986) and his secondary sexual characters (e.g. forehead patch size; Qvarnström, 1998). Pair formation is seasonal, with no mate fidelity between years. Most of the males are monogamic, but approximately nine per cent of males attract a secondary female (Gustafsson & Qvarnström, 2006). Moreover, extra-pair copulations in this species are quite frequent, resulting in 14.5-15.5% of all nestlings being extra pair young (Sheldon & Ellegren, 1999; Veen et al, 2001). Collared flycatchers on Gotland lay one clutch per year, consisting on average of six eggs (range of four to eight eggs). At the beginning of May, the first eggs are laid and incubation begins, a task which is undertaken solely by females. Nestlings hatch after approximately 14 days of incubation, but females can prolong the incubation time for up to 20 days if the weather is cold (personal observation). Nestlings remain in the nest for an additional 14-16 days and are fed by both parents. They reach the maximum nestling body mass at the age of 10-11 days and lose mass prior to fledging. After fledging, young flycatchers stay close to the nest for another two weeks and are still fed by the parents. Just after breeding or by the end of breeding season, the complete autumn moulting of adults begins and is followed by the autumn migration (Hemborg & Merilä, 1998). Prior to spring migration all adults, except for the yearlings, exchange their wing feathers again. This difference in moulting pattern makes is possible for yearlings and older individuals to be visually distinguished during the breeding season (Svensson, 1992). Research on collared flycatchers on Gotland was initiated in 1980 (Gustafsson, 1989) and is still continued. This population is a perfect study system for age-related patterns because of the high site-fidelity displayed by individuals (Pärt, 1991) and because the birds prefer nest boxes over natural tree holes (Gustafsson, 1986), which makes them easy to handle. Each year the majority of adults are caught, ringed and reproductive parameters are recorded, including; laying date, clutch size, and the number of hatchlings (eggs that hatched), fledglings (nestlings that fledged from the nest) and recruits (fledglings that returned to breeding areas in at least one of the next two years). Nowadays, our Gotlandic study population of collared flycatchers consists of 15 distinct nest-box areas in southern Gotland, where my study was performed, and over 15 small study areas in central northern Gotland. Dis- persal between the southern and the northern areas is negligible (personal observation). In collared flycatchers lifetime reproductive success- one of the most ac- curate measures of fitness- depends on the number of breeding seasons (paper I). This is a pattern observable in many other seasonally breeding bird species (e.g. Clutton-Brock, 1988). Thus, longevity is an important characteristic when assessing individual fitness and quality.

12 Age-dependent survival and reproduction

Age-dependent changes in mortality and reproductive success of collared flycatchers were studied for the first time in 1990, ten years after the study of collared flycatchers on Gotland was initiated. No differences in survival probability between age classes were detected (Gustafsson & Pärt, 1990). However, other signs of senescence in collared flycatcher females were detected as early as their fourth year: a decline in clutch size and number of fledglings, trends for a reduced number of recruits, and later laying dates. I conducted a follow-up study investigating age-dependent changes in mortality and reproductive success of collared flycatchers 15 years later with approximately nine times the sample size (7276 versus 871 breeding re- cords). In addition to the parameters measured in Gustafsson & Pärt’s (1990) study, I also analyzed age dependent changes in survival and reproduction using individuals from the oldest age class (six to seven years of age). My study also augmented the previous study with the inclusion of male flycatchers. Unfortunately, the data on males is incomplete because males who failed to reproduce were not included and extra-pair paternity was not controlled for (for details see paper I).

Figure 2. Age-related survival of collared flycatcher. Females are represented by white columns and males by grey. Mean ± SE, sample size is placed over particular bars (number of survivors above number of observations)(see Paper I).

The patterns of survival and reproduction differed between my study and the earlier study by Gustafsson & Pärt (1990). Survival probability changed with an individual’s age for both male and female flycatchers (Fig 2). For females, survival probability increased significantly from the first to the

13 second year of life and then proceeded to decrease with each following year. For males, survival probability showed a trend to increase from the first to the third year of life, followed by a decrease with each year to the end of life. No individuals survived for longer than eight years (Fig 2). I found that reproductive success changed with age. Both males and females showed an increase in overall reproductive success and its underlying components during the initial stages of their life (one to two years) (Paper I; see Fig 3A for an example). Middle age classes did not differ significantly between each other. The later age classes (five to eight years) showed an overall decrease in reproductive success in both sexes, although in females there was a non-significant trend for an increase between sixth and seventh year. When long-lived females were analyzed separately; an initial increase in their reproductive success was not detectable (beside changes in laying date; see paper V) and they maintained a high reproductive success for most of their lifetime. However, in this group a trend for decline in reproductive success was already apparent after the age of three. This pattern resembled the pattern found in Gustafsson & Pärt’s (1990) study, although in my study the changes became significant no earlier than the age of five (Fig 3B).

Figure 3. Age-related changes in reproductive performance of female collared flycatcher. A – pattern for the whole population (see paper I for details), B – pattern for females who survived for at least five years (see paper V for details). Mean ± SE, sample sizes placed over error bars.

14 Why do some individuals live longer than others?

The secret of longevity has always intrigued people. For a long time we have tried to answer questions as to why some individuals live longer than others. Some of the first observations made in this area concerned the genetic com- ponent of longevity, by raising questions such as why some families live longer than the others. However, many additional factors to genes influence an individual’s lifespan to an even higher extent. Individuals differ from each other in both: current condition, defined as an internal property of the individual (after Rowe & Houle, 1996); and general quality defined as phenotypic status of individual shaped by genetic quality (as defined by Hunt et al, 2004) and developmental conditions. Both condition and quality influence the survival probability as well as reproductive success, creating a positive correlation between these factors. In paper I, I show that some poor quality individuals die after their first breeding season. However, this does not provide an explanation for lifespan differences later in life, since this pattern disappears after the second breeding season. This finding is also not useful in elucidating why some individuals are of higher quality than the others. I found this explanation by comparing nestling history of long-lived (at least five years) and short-lived (up to four years) individuals (paper V). Two characters best explained differences in longevity in females which survived at least five years; longer tarsi, and a lower body mass as fledglings than females which survived for a shorter time (Fig 4).

Figure 4. Body mass and tarsus length at fledgling in relation of female lifespan. The solid line and left axis depicts tarsus length, whereas body mass is de- picted by the dashed line and right axis. Numbers above error bars indicate sample size (same for both factors) (see paper V for details).

These opposing effects of fledgling tarsus length and body mass on longevity can be due to rearing conditions in the nest that a particular female originates from, and are discussed in depth in paper V. There are also affects on lifespan occurring later in a female’s life: females mated to more attractive males had shorter life spans (paper V). Attractive males were shown to

15 decrease their feeding effort, and thus females mated to them compensate by increasing their feeding rates (Qvarnström, 1997). This may generate a high reproductive cost to the female which plays a role in decreasing her life span.

What causes age-related changes in reproductive success?

The most pronounced changes in an individual’s reproductive success occur early in life (for small passerines typically between first and second year) and, later towards the end of their life. The patterns of age-related changes in reproductive success, especially those occurring early in life, have been widely studied in birds (e.g. Curio, 1983; Saether, 1990; Forslund & Pärt, 1995). The initial increase in reproductive success has received the most attention since it typically is more striking than the change in reproductive success later in life. In the following paragraphs I will attempt to review the major hypotheses explaining changes in reproductive success early and late in life (see also papers I and V). Generally they can be divided into two groups: constraints and reproductive effort adjustment. Hypotheses address- ing the constraints which may cause differences in reproductive success between age classes include those which predict that some individuals cannot perform better than they do because: i) they are of poor quality (quality hypothesis), ii) they are not experienced during first reproductive event (experience hypothesis), or iii) that they are in poor condition (e.g. declining as a result of senescence). In contrast, hypotheses concerning the adjustment of reproductive effort explain differences in reproductive success as being due to individual investment decisions (optimization hypothesis and ‘terminal investment’).

Early increase in reproductive success

Because the probability of mortality increases with age, the most common age class in the population is always the youngest, which may lead to those young individuals having the highest genetic input into the population. How- ever, very young individuals typically have a lower reproductive output than older ones (e.g. Newton, 1988). Thus, the genetic input of the youngest individuals into the population may in fact be smaller than predicted just from the survival patterns.

16 As outlined above, three major hypotheses have been raised to explain this early increase in reproductive success: i) the progressive appearance/ disappearance of phenotypes (differential quality hypothesis), ii) the experience hypothesis, and iii) the optimization hypothesis. These hypotheses are tested individually in paper I. However it must be noted that hypotheses explaining age-related differences in reproductive success are not mutually exclusive and it is very likely that more than one mechanism is at play.

Differential quality hypothesis

The basis for the quality hypothesis was proposed by Coulson (1968), who described the differences in individual quality and its relation to survival of Kittiwake males breeding in different parts of the colony. He considered age-related changes in reproductive success at the population level, but not on an individual level. The quality hypothesis assumes indirectly that an individual’s quality is constant throughout its lifetime, and that both its reproductive success and survival probability are dependent on its quality. The initial increase of average reproductive performance within the cohort results from two possible causes; i) differential survival, or ii) differential age at first reproduction for individuals of diverse quality (Curio, 1983; Forslund & Pärt, 1995; Espie et al, 2000). Thus, a progressive, quality-dependent appearance or disappearance of certain phenotypes in the cohort is observed. In the first scenario a progressive disappearance of certain phenotypes is predicted due to differential survival: poor quality individuals which are also poor breeders die after the first breeding event(s). This results in an increase of average reproductive success in this cohort. Such a correlation between reproductive success and survival has been shown in Merlins (Espie et al, 2000) and Great cormorants (Bregnballe, 2006). My study provided further evidence for progressive disappearance of poor breeders in both female and, to a lesser extent, male collared flycatchers (paper I). Females which did not survive also had a low overall reproductive success in the first year of life, with males showing a similar pattern only in the number of recruits (Fig 5). In the second scenario, a progressive appearance of certain phenotypes is predicted without consideration of quality-dependent survival differences. Instead, individuals differ in age at first reproduction, with high quality individuals being predicted to delay their reproduction. Thus, the initial increase in reproductive performance results from their appearance in the breeding population (Forslund & Pärt, 1995). This hypothesis is not very well supported by experimental evidence (Forslund & Pärt, 1995). In birds, it is un- clear why high quality individuals would delay reproduction, since their lifetime reproductive success depends often on the number of breeding at- tempts (e.g. Espie et al, 2000). In my study, individuals which delayed reproduction did not differ in reproductive success from individuals of the

17 same age which started reproduction already in their first year. Moreover, individuals which delayed reproduction had a low lifetime reproductive success, which suggests that they are rather of poor quality (paper I). In the light of these findings I conclude that this case of a progressive appearance of high quality phenotypes does not apply to collared flycatchers on Gotland.

Figure 5. Examples of the differences in reproductive success in first year of life between individuals surviving for over one year and for one year only. (ANOVA, Mean ± SE, sample sizes are placed next to the rele- vant error bars). Solid line represents females, dashed line – males. PC1 is a principal com- ponent of reproductive success calculated from laying date, clutch size, number of fledglings and number of recruits. See paper I for details.

Breeding experience hypothesis

Within-individual changes in reproductive performance cannot be explained by the mean differences in quality between age classes. Those differences are often explained by gaining breeding and/or foraging experience. The experience hypothesis predicts increasing breeding competence to be essen- tial for an increase in reproductive success (Curio, 1983; Forslund & Pärt, 1995). According to this hypothesis, young birds breeding for the first time have lower reproductive performance because they are yet to gain the skills which are crucial for reproduction. These skills include: finding a mate, se- curing a breeding site, foraging, and feeding the offspring (Forslund & Pärt, 1995).

18 Inexperienced individuals have been shown to have lower breeding success than experienced ones in several bird species (Croxall et al, 1992; Fors- lund & Larsson,1992; Bregnballe, 2006). In addition, one study from the Gotlandic population of collared flycatchers has suggested that previous breeding experience explain at least part of the age differences in breeding performance early in life (CichoĔ, 2003; but see Pärt, 1995). According to the experience hypothesis we should observe an improvement in reproductive performance with age, within an individual. Addition- ally, as reproductive success is predicted to be the result of experience; birds which have bred before should express higher reproductive output than birds of the same age that have not bred before. As mentioned above, birds in our study population which delayed their reproduction do not differ in reproductive success from already experienced birds of the same age (paper I). More- over, when laying date is controlled for, young and older females do not differ in the quality of the territories they breed on (paper III). Thus, earlier experience does not seem to play a role in shaping age-related patterns of reproductive success in collared flycatchers on Gotland. However, when laying date was controlled for, we found that young females pay higher costs of reproduction (paper II) and do not profit as much from breeding at high quality territories as middle-age birds (paper III). These patterns are best explained by their lack of earlier breeding/foraging experience, and provide some evidence for the experience hypothesis.

Optimization of reproductive effort hypothesis

Another hypothesis explaining within-individual improvement in reproductive performance is the optimization hypothesis, also called the restraint hypothesis for early age classes (Williams, 1966; Sterns, 1992; Forslund & Pärt, 1995). Central to the theory are individual decisions on reproductive effort, with reproductive success depending on the reproductive effort an individual puts in to reproduction. Decisions as to the degree of reproductive effort are made on the basis of individual quality, residual reproductive value (Fisher, 1930), or future survival and breeding probability (Williams, 1966). The lower the reproductive value, the greater the effort an individual puts into reproduction and, consequently, the higher its reproductive success. Therefore, old individuals are expected to express higher reproductive success than young ones, because old individuals have lower residual reproductive value and/or survival probability than young individuals. According to optimization hypothesis, the early increase in reproductive success observed in many bird species is a result of a restrained reproductive effort by very young individuals. If the optimization of reproductive success plays a role in the early increase of reproductive success, then a negative correlation between repro-

19 ductive success and survival should be observed. This is because individuals with a low survival probability should invest more into reproduction, and thus have higher reproductive success (e.g. Forslund & Pärt, 1995). As mentioned before, collared flycatchers on Gotland show a positive correlation between reproductive success and survival (paper I). It is, however, likely that the optimization of reproductive effort is still an important mechanism, at least for some birds. This effect may just be masked by dominating patterns of individual quality differences. Birds with a long life span also have a relatively high reproductive success throughout their lifespan, whereas short- lived individuals start with a low reproductive output in the first year and improve during the following years (paper I). Moreover, young females show a higher immune response against novel antigen than old females (Fig 6), which suggests a relatively higher investment into self-maintenance in this age class (paper III and IV). These results point towards the mechanism of restrained reproductive effort early in life. However, the first finding suggests that it may only apply to short-lived lower quality birds (see paper I for details).

Figure 6. Immune responses to sheep red blood cells of females from different age classes (mean ± SE). ‘Young’ are 1-year old, ‘mid- age’ 3-year old females and ‘old’ is 5–6-year olds. Numbers below error bars denote sample size. Horizontal lines with numbers denote statistical significance after Tukey post hoc test. Graph A comes from paper IV, graph B from data in paper III.

20 Changes in reproductive success late in life

As mentioned above, survival probability decreases with age and old individuals are not as numerous as young ones in the population. Thus, old individuals’ input into the population is much smaller than that of young individuals and, for this reason, natural selection acts much stronger at early than at late ages (Medawar, 1952; Williams, 1957; Futuyma, 1997). Reproductive success is often observed to change towards the end of an individual’s life. The typical reproductive pattern for birds is a decrease late in life (e.g. New- ton, 1988). However, an increase in reproductive output towards the end of life may also occur (Sanz & Moreno, 2000) since old individuals often increase their reproductive effort (Pärt et al, 1992). The decrease in reproductive success observed frequently at old ages is most often explained by physiological senescence (Kirkwood, 1977). This results in an ageing individual not being able to improve its performance, regardless of its investment into reproduction. In contrast, an increase in reproductive success later in life is observed in some cases and is typically explained by an increase in reproductive effort by the end of life. In the following paragraphs I am going to briefly present both these ideas and relate the results from this thesis to their predictions.

Physiological senescence

Hypotheses describing particular mechanisms causing senescence are described in detail in paper V; however, they are not tested in this thesis. For the purpose of this summary it is important to note that senescence is defined as a slow decline in physiological functions which results in increasing mortality and decreasing reproductive value with age. Generally it is believed to occur as a result of unrepaired somatic damage (Williams, 1957; Hamilton, 1966; Charlesworth, 1980; Rose, 1991). As mentioned above, the strength of natural selection decreases with an individual’s age and the input into the population of old individuals is weak (Fisher, 1930; Hamilton, 1966). Thus, it is much more profitable to maintain and repair somatic tissues early in life (Rose, 1991). Because of this fact, it is possible that genes which are profitable early in life but deleterious late in life are not eliminated from the population (antagonistic pleiotropy - Williams 1957). Additionally, as the self- maintenance level decreases with age, the accumulation of unrepaired deleterious mutations is possible later in life (Medawar 1952). Nonetheless, it is a theoretical plausibility that senescence occurs due to purely somatic changes, without genetic influences. Scenario considering purely somatic changes is described by the soma hypothesis. It is based on a differential

21 allocation of resources into organismal processes. If self-maintenance is underinvested, the somatic damages accumulate and a physiological break- down occurs (Kirkwood 1977). The disposable soma hypothesis is based on the same mechanisms of differential investment in organismal processes as the optimization of reproductive effort hypothesis described above. However, they give different predictions for the shape of reproductive success at the end of life (take a different perspective). As in other senescence hypotheses, the disposable soma hypothesis predicts a decrease in reproductive success towards the end of life, as a consequence of physiological deterioration. In contrast, the mechanism of optimization of reproductive effort predicts an increase in reproductive success towards the end of life, as a result of increased effort (see also ‘terminal investment’ below). However, increased effort may reinforce the mechanism of disposable soma, as limited available resources are invested in an increased reproductive effort, resulting in low investment in self- maintenance. As described above, collared flycatchers on Gotland show an increased probability of mortality much earlier than expected as an effect of senescence. Females’ survival probability starts to decrease already after their third year of life (significantly after their fourth), whereas males’ survival probability starts to decline after their second year of life (significantly after their third; Paper I; Fig 2). Patterns of a decrease in reproductive success are not clear (see above) and different components of reproductive success change differently. Therefore, it is difficult to state when senescence starts. Such conclusions are always difficult, since every organism begins ageing at different moment of life, does so at a different rate, and in a different way (i.e. different functions are affected first; e.g. Arking, 1998). In the study described in paper IV, we found that females of at least five years of age show simultaneous decreases in both reproductive success components (fledglings’ body mass) and immune response. These results of simultaneous decrease in several functions follow a typical pattern for senescence.

‘Terminal investment’

The idea of ‘terminal investment’ is based on both senescence theory and on the optimization hypothesis, in particular the fact that residual reproductive value of an individual decreases with age and therefore its reproductive effort should be increased. Hence, an increased reproductive success may be observed towards the final stages of an individual’s life. It is, however, important to stress that this terminal increase in reproductive effort might also be relative i.e. higher in relation to individual condition and investment into self-maintenance, yet not necessarily higher in absolute value of reproductive effort (Clutton-Brock 1984; see also paper V). Thus, if individuals mak-

22 ing their terminal investment are in poor condition, the observed increase in reproductive output will be negligable. In the Gotlandic population of collared flycatchers, females have been shown to make a terminal investment. Pärt and colleagues (1992) found that old flycatcher females (at least five years old) were feeding their nestlings at a higher rate than middle-aged females. This resulted in higher loss of body mass during the reproductive period. In my study I was able to separate old females into particular age groups. I found that at the age of five some females show very high reproductive success but do not survive to the next year (Fig 7; paper V). This result indicates a terminal investment mechanism in this age class of females and partly explains why a decrease in reproductive success is not distin- guishable at later ages in this study population, whereas a decrease in survival is (see above). At the age of six the pattern is reversed; however, the increased investment can not be excluded because of detected quality differences between females (see paper V for details).

Figure 7. The differences in reproductive success (number of fledglings) of females in their fifth year of life, between survivors and non-survivors. (Nomi- nal Logistic Regression; Mean ± SE, sample size placed over particular error bars). (see paper V for details).

In paper IV, we tested the relative investment into self-maintenance by young, middle-aged and old females by comparing their immune response to a novel antigen. Old females (at least five years old) had a much lower immune response than the other age classes (Fig 5). This low immune response can be explained by either their low investment in self-maintenance or by their inability to elicit a high response because of senescence. In the study described in paper III, different females of these three age classes were exposed to the antigen again, but this time both old and middle-aged females had low immune responses when compared to young females. Since middle- aged females are not expected to senesce we can assume that a decrease in immune response with age is caused by a generally reduced investment in self-maintenance at older ages.

23 Age-related cost of reproduction

It is trivial to say that reproduction is costly. However, it is a unavoidable fact that reproduction is one of the processes competing for resources (e.g. Stearns, 1992). The amount of available resources is limited and dependent on current condition. The decision as to how these resources are allocated depends on many factors and has been widely studied. For example, infec- tion may cause redirection of resources towards immunity (Ilmonen et al, 2000), or towards reproduction in such a mechanism as described above by ‘terminal investment’ (Bonneaud, 2004). The optimization hypothesis (Wil- liams, 1966; see above) predicts that allocation of resources into reproduction or self-maintenance (e.g. immune response) should change with pro- gressing age and decreasing residual reproductive value (Fisher, 1930; Wil- liams, 1966). Following this optimization pattern is especially important for very young birds which are not experienced and/or are of lower quality and, thus, may pay higher penalties for investment mistakes (Sasvári et al, 2000). In the study described in paper II the higher reproductive costs experienced by young individuals are found in collared flycatchers on Gotland. Young females which were forced to raise enlarged broods laid smaller clutches in the next year (Fig 8).

Figure 8. Clutch size in the year following the experimental ma- nipulation of brood size in different female age classes. Dashed line – one year old females, solid line – middle-aged females. Bars represent standard errors (see paper II for details).

Nonetheless, under favorable conditions reproductive cost can be reduced. In paper III I show that females occupying territories rich in caterpillars (the highest quality food) make fewer feeding visits yet still produce fledglings of

24 at least the same condition as females breeding in poorer territories. More- over, females breeding in caterpillar-rich territories reduce their reproductive costs i.e. lose less in body mass during the breeding season and have higher survival rates (Fig 9).

Figure 9. Influence of relative caterpillar availability on (A) female body mass loss during breeding and (B) female survival. Solid lines represent young females, dashed lines - middle-aged and dotted – old females. Interactions between female age and prey type are significant (see paper III for details).

This study also detected differences in reproductive costs between females of different ages. Young females do not show any marked benefits from breeding in high quality territories in body mass changes, and old females do not show any differences in survival (Fig 9). The reduced benefit to young females may be explained by the fact that they generally pay higher reproductive costs, or by their lack of breeding experience. Consequently, the group of old females consists of birds older than five years. Thus, some of them may senesce or undertake a ‘terminal investment’, whereas some may be unaffected by age. Therefore, there will be differences in resource allocation patterns within this group of birds. This may explain the observed, inconsistent pattern of territory quality-related reduction in reproductive costs.

25 Final conclusions and perspectives

I did not attempt to find the ultimate secret of longevity or everlasting youth in this study. Nonetheless, this thesis provides broad data on age-related changes in survival and reproduction in a wild population of passerine birds. First, it clearly shows that senescence also occurs in the wild and has similar overall effects on the organism as it does in humans. I found that an individual’s lifespan depends on many different factors: individual quality, developmental condition, reproductive effort, mate choice and the quality of its breeding territory. Even more importantly, my study showed that many mechanisms shaping investment patterns, immunity, reproductive costs, and the benefits of breeding in a high quality territory act differently in different age classes. Age is often a neglected factor in all kinds of ecological studies, yet a whole spectrum of characters change with individual age alone. My findings show that pooling individuals of various ages together may result in confu- sion, since individuals belonging to different age classes may follow different patterns.

Nonetheless, this study reveals only the tip of the iceberg. Each time I answered one of my research questions I found many additional ones to be addressed. However, isn’t this typical in science!? The problems which remain to be solved definitively are: x What is the relative importance of genetic and environmental factors in- fluencing longevity? My findings suggest that parental quality may be quite important for shaping individual lifespan either directly through good genes, or by creating favorable developmental conditions. x What is the mechanism, triggering investment decisions at different ages? Can we measure changes in body condition which lead to the reallocation of resources from self-maintenance to reproduction? Or, is it the other way around, with the ‘wrong’ investment decisions triggering the a cas- cade of events with no way back to an increased investment in self- maintenance? x What directly makes young females pay higher costs of reproduction? Is it individual quality, lack of breeding/foraging experience, physiological immaturity, or other factors?

26 x How is it possible that some individuals can afford maintaining very high levels of reproductive output through their lifetime? How can they afford paying elevated reproductive costs? Are those costs really higher for them? x How can collared flycatchers recognize the quality of a breeding territory? Is that ability determined genetically (instinctual) or learned? x Do birds follow optimal investment patterns at a certain age? For in- stance, is the observed high immune response of young females an accu- rate measure of the potential risks they are exposed to or is it higher than is necessary to give protection from disease?

Most of these questions require further addition of data to the extensive collared flycatcher database or additional experiments, and therefore, cannot be answered yet. For some of these questions, we do not have sufficient experimental methods at hand yet. This thesis, however, gives a good basis for further investigations into the topic of age and its effects in the wild.

27 Summary in Swedish (Sammanfattning)

Åldrandet och åldersrelaterade förändringar är ett biologiskt fenomen som är generellt för alla levande organismer. Det har fascinerat mänskligheten i alla tider. Inte bara för att det drabbar oss själva, utan också för att naturen uppvisar så många olika variationer i livslängd och åldrande både mellan olika arter och mellan individer inom en art. I århundraden, ja rent av årtusenden har människan sökt efter ’ungdomens källa’ för att insupa dess begärliga dryck i en strävan efter evigt liv. Detta utan att reflektera över det uppenbara faktum att alla levande organismer, precis som människan, genomgår ett stadium av åldrande innan den oundvikliga döden. Grunden för den moderna biologins syn på åldrandet har sin utgångspunkt i August Weismann (1834-1914) som föreslog (1886), att det centrala för flercelliga organismer är att de har två separerade celllinjer, de reproduktiva könscellerna (spermier och ägg) respektive soma-linjen (kroppscellerna). De reproduktiva cellerna förs vidare oförändrade till nästa generation, opåver- kade av förändringar i resten av kroppens celler. Detta i sig omöjliggör att förvärvade egenskaper eller förändringar i kroppens celler kan nedärvas. Denna insikt blev fundamental för neo-darwinismen och utgör startpunkten för dagens evolutionära idéer om åldrandet. En annan framstående forskare Sir Ronald Fisher (1890-1962) formulera- de ett ytterligare ett viktig steg i förståelsen för skillnader mellan och inom arter i åldrandeprocessen. Han skrev (1930) att: ’Det vore av största intresse att förstå inte bara den fysiologiska mekanismen med vilken resurser förde- las mellan gonader (till reproduktion) och resten av kroppen, utan också hur detta förändras under livet och i olika miljöer. Särskilt vilka betingelser som gör det lönsamt att fördela mer eller mindre av tillgängliga resurser till reproduktionen’. Denna insikt var före sin tid och formulerar den grund- läggande idéen om betydelsen av resursallokeringar (trade-offs) och hur dessa kan förändras under en individs liv. Det tog ända tills slutet på 50-talet innan detta kopplades ihop med åldrandeprocessen. Då framväxte en solid evolutionsteori om hur åldrandet kan vara generellt för alla organismer. Sir Peter Medawar (1952), George Williams (1957) och William Hamil- ton (1966) utvecklade en genetisk förklaring om varför åldrandet kan vara en optimal livshistoria, trots att det i sig reducerar en individs fitness. Det centrala i deras idéer är att det varje år dör ett visst antal individer. På grund av detta kommer det alltid att finnas relativt få gamla individer i relation till antalet unga. Därför kan inte den naturliga selektionen lika effektivt elimine-

28 ra genotyper som uppvisar dåliga drag för reproduktion och överlevnad sent i livet. Istället premieras de positiva dragen som uttrycks tidigt i livet. De yngre individerna utgör större delen av en population och därför blir effekten av den naturliga selektionen effektivare på de drag (positiva eller negativa) som utrycks tidigt i livet. Varför har jag valt att studera åldrandet idag på 2000-talet? Vet vi inte redan allt värt att veta? Svaret är nej, långt ifrån allt. De senaste åren tycks ett förnyat intresse ha växt fram för studier av åldrandet och processer relaterade till åldrandet. En möjlig förklaring kan vara, dels de nya möjligheter som framsteg inom molekylärbiologin givit oss beträffande orsakssamband på gennivå och dels tillgång till allt fler långtidsstudier av vilt levande djurpo- pulationer. Slutligen, förstås, den eviga olösta frågan om variationer i vårt eget åldrande. Allt detta har lett till att vi nu har möjligheten att verkligen testa de underliggande orsakerna till åldrandet och processer relaterade till åldrandet. Vid studier av åldrandet hos människor och djur i fångenskap har forskarna redan gått från att studera de ytligt synliga och mätbara tecknen på åldrande till de underliggande mekanismerna på molekylär och biokemisk nivå. Vid studiet av vilda populationer är rent åldersrelaterade förändringar ofta påverkade av förändringar i miljön och interaktioner med andra individer. Dessa interaktioner kan vara av både inomarts- och mellanartsnatur. Dessutom har varje individ sin egen specifika åldrande historia. Detta sammantaget gör att de åldrandeprocesser hos djur i fångenskap och på laboratoriet, nödvändigtvis inte är de samma som hos djur ute i naturen. Därför har jag valt att fokusera studierna i min avhandling på frilevande fåglar. De mest spektakulära ålders- relaterade förändringar hos de flesta djur sker både i början av livet och i slutet av livet. Det är anledningen till att jag i den här avhandlingen har valt att försöka förklara och testa några av de mekanismer som kan antas ligga bakom åldersrelaterade förändringar i överlevnad, reproduktion och livshi- storieavvägningar. Jag har valt en, för detta ändamål, idealiskt population av halsbandsflugsnappare på södra Gotland. Här har fåglarna studerats sedan 1980 och populationen består av mellan 400 till 500 par som årligen häckar i uppsatta fågelholkar. Alla individer, både föräldrar och deras ungar, har ringmärkts med individuella ringar och följts från födseln till livets slut. En mängd data om deras reproduktion, morfologi, immunologi och beteende har insamlats genom åren.

Åldersberoende överlevnad och reproduktion

I den här studien har jag använt mig av data från 7276 häckningar insamlade under 25 år. De äldsta individerna blev åtta år. Det fanns dock ett par individer som hade blivit minst nio år. De användes inte i studien för att de var så få och för att deras exakta födelse år inte är känt.

29 Överlevnaden hos både honor och hanar förändrades med ålder och ökade från drygt 40% som ettåriga till 45% för två- till treåriga. Därefter sjönk överlevnaden gradvis till 20% hos sex- och sjuåringar. Ingen åttaårig fågel överlevde. Detta gjorde att det fanns mycket få individer som blev äldre än fem år. Reproduktionsframgången förändrades också med åren. Både honor och hanar ökade antalet överlevande ungar från år ett till år två. Så gjorde också alla de andra karaktärer som har med häckningen att göra, som t.ex. häck- ningsdatum, kullstorlek (antalet ägg), antalet kläckta ägg och antal ungar som flög ut ur boet. Mellan år två till år fyra skedde inga påtagliga föränd- ringar utan först från år fem till år åtta sjönk den generella häckningsfram- gången. Sex och sju åriga honor betedde sig lite oväntat och hade högre vär- den. Mer om detta längre fram.

Varför lever vissa individer längre än andra?

Hemligheten med ett långt liv har alltid konfunderat mänskligheten. Tidigt gjordes observationer att individer från vissa familjer oftare blev mycket gamla. Det skulle tyda på att finns en stark ärftlig komponent i livslängd hos människor. Andra, nyligen gjorda, studier av halsbandsflugsnappare tyder på en mycket låg genetisk bakgrund till skillnader i livslängd. Det finns dessutom en mängd av faktorer under en individs livstid som kan påverka livs- längden. Individer varierar både i sin nuvarande kondition (definierad som påverk- bar fysiologiskt status) och generell kvalité (definierad som konstant status bestämd under uppväxten av både gener och miljö). Både kondition och kvalité påverkar överlevnad och reproduktionsframgång vilket gör att individer med hög status kan ha både hög överlevnad och reproduktionsfram- gång. Varför är då vissa individer av högre kvalité? De honor som blev långli- vade hade från födseln större kroppsstorlek och var mindre feta än de som var kortlivade, vilket tyder på att uppväxtbetingelserna har en avgörande effekt på livslängden. Det kan tyckas märkligt, att de som var mindre feta som ungar blev långlivade. Normalt förväntar man sig i naturen, att de som får mycket mat blir de mest levnadskraftiga individerna. Så är det också för flugsnapparna, i alla fall tidigt i livet. Däremot kan det vara så att lagom är bäst, vad det gäller pålagrade fettreserver, för ett långt liv. Det finns teorier om livslängd och restriktioner av kalorier som är påvisade hos flugor och möss och kan tänkas gälla också för människor och flugsnappare. En restrik- tion av intaget av kalorier under det tidiga livet gav längre livslängd. En annan effekt tycktes vara att honor som hade en far med stor vit pann- fläck (en fjäderteckning som signalerar hög kvalité) också blev långlivade. En annan avgörande effekt för honornas livslängd var huruvida deras part- ners senare i livet hade en stor pannfläck. Detta påverkade deras överlevnad

30 negativt. Det har nämligen visat sig att honor parade med storfläckade hanar får arbeta mer med matningen av ungarna medan dessa hanar är förhållande- vis lata och matar i mindre utsträckning. Denna ökade arbetsbörda leder då till ökad dödlighet för dessa honor.

Varför ökar reproduktionsframgången tidigt i livet?

Det finns tre olika hypoteser som framlagts för att förklara varför reproduk- tionsframgången ökar de första åren av livet. Selektionshypotesen förklarar ökningen med, att de individer som är av högre kvalité (de som också har högre reproduktionsframgång) är de som överlever från år ett till år två. Er- farenhetshypotesen förklarar ökningen med, att de individer som överlevt har skaffat sig större erfarenhet av att häcka och föda upp ungar. Optime- ringshypotesen förklarar ökningen med, att de individer som begränsar sin ansträngning tidigt, för att öka den senare i livet, blir de som får högst repro- duktionsframgång totalt under livet. Dessa tre hypoteser utesluter inte var- andra utan kan alla bidra till ökningen. Hos både hon och han halsbandsflugsnappare visade det sig att selektionshypotesen förklarar största delen av ökningen, då de överlevande ettåringarna hade högre reproduktionsframgång än de som dog. Dessutom visade det sig att om man analyserade de långlivade (5-8 år) individerna för sig hade de en hög och konstant reproduktionsframgång under hela livet. Individer med hög kvalité har högre initial reproduktionsframgång och upp- gången beror nästan enbart på borttagandet (döendet) av individer med låg kvalité efter första året. Erfarenhetshypotesen kan inte förklara varför reproduktionsframgången ökar de första åren av livet, eftersom det inte fanns någon påtaglig ökning när varje livslängdsgrupp analyserades var för sig. Dessutom skiljde sig inte de individer, som överlevt till år två men inte häckat år ett, från de som häckat båda åren. Därmed är inte sagt att erfarenhet inte har någon betydelse. Mina resultat tyder på att erfarenheten spela en viss roll när det gäller att optimera de reproduktiva ansträngning det första levnadsåretåret. När ettåri- ga honor fick kullen förstorad matade de ungarna så mycket mer, att deras överlevnad eller reproduktion nästa år påverkades negativt. Detta blev inte fallet när treåriga honor fick sin kull förstorad. Ettåriga honor kunde heller inte fullt utnyttja resurserna i form av mat till ungarna som fanns i reviret, om de råkade ha ett högkvalitativt revir. Optimeringshypotesen fick lite motstridiga resultat som förklaring till ök- ningen de första åren av livet. Kullförstoringsexperimentet visade, att ettåri- ga honor faktiskt var benägna att öka sin reproduktionsansträngning tidigt i livet tvärt emot de teoretiska förutsägelserna. När vi, däremot, studerade hur olika åldrar fördelar resurser till immunförsvaret respektive reproduktionsan- strängningen, visade det sig att ettåringar var mer benägna att satsa på im- munförsvaret än äldre honor. Detta för att öka sina överlevnadsmöjligheter.

31 Varför minskar reproduktionsframgången sent i livet?

Förutom den genetiskt baserade evolutionsteorin om hur åldrandet kan vara generellt för alla organismer har en rent fysiologisk mekanism föreslagits. Kirkwood beskrev sin fysiologiska ”disposable soma hypotes” 1977. Den ser åldrandet som ett resultat av minskade resurser till underhållandet av för- störda kroppsceller och kroppsfunktioner. Denna idé står inte i motsatsför- hållande till den genetiskt baserade evolutionsteorin utan fungerar som en fysiologisk förklaring till de underliggande genetiska mekanismerna. Det visade sig hos halsbandsflugsnappare, att man kan förklara den tidigt i livet ökande överlevnaden och reproduktionsframgången framför allt med en skillnad i kvalité´ hos de individer som överlevde och de som dog. När det gäller mönstren senare i livet är bilden mer komplicerad. Överlevnaden sjunker accelererande från treårsåldern medan reproduktionsframgången inte har ett lika klart mönster. Detta kan troligen i huvudsak förklaras med optimeringshypotesen och av att individer börjar åldras vid olika ålder beroende på deras kvalité. Tydligt var att immunfunktionerna blev sämre från fem års ålder. En orsak till att reproduktionsframgången hålls på en relativt hög nivå under slutet av livet, trots att överlevnaden minskar kraftigt, är att individer väljer att göra en så kallad terminal investering det sista levnadsåret. Idén om terminal investering baseras på optimeringshypotesen och den förväntade framtida reproduktionsframgången. När åldrandet startat minskar överlevna- den drastiskt och chanserna att kunna genomföra en lyckad framtida häck- ning minskar därför också drastiskt. Därför är det bättre att vid hög ålder fördela mer resurser till den pågående reproduktionen och mindre till sin egen överlevnad som i immunfunktionsexemplet. Alltså satsa allt på en sista reproduktion (terminal investering) och sedan dö. En sådan terminal investering stämmer speciellt för femåriga honor. Samma mönster fanns inte hos ännu äldre honor utan kunde istället förklaras med deras ovanligt höga kvali- té. Dessa verkligt gamla individer tycks utgöra extremfall och är kanske exempel på så kallade Darwinska demoner som försöker trotsa grundläggan- de teorier om livshistoria.

Betydelse av ålder i naturliga populationer och framtida forskning

Ålder är ofta en förbisedd faktor i ekologiska och evolutionära studier. Den här avhandlingen har visat betydelsen av att studera individer med känd ål- der och att separera grupper med olika livslängd, för att kunna dra rätt slut- satser. Att negligera detta leder ofta till förvirring och felaktiga tolkningar eftersom individer som tillhör olika åldersklasser och som skiljer sig i livs- längd följer olika mönster av optimering under sin livshistoria.

32 Min avhandling visar bara toppen av isberget. Varje gång jag kom fram till ett resultat ledde det till nya frågor som kräver svar. Några specifika problem att lösa i framtiden är: x Vilken är den relativa betydelsen av gener och miljö för livslängden? Jag har visat att vad jag kallat kvalité har stor betydelse, men den är både beroende av gener och miljöfaktorer under den tidiga uppväxten. x Vilken fysiologisk mekanism är det som startar ett beslut att allokera mera resurser till reproduktion och mindre till kroppsfunktioner för över- levnad? Eller är det så att ett felaktigt beslut sätter igång åldrandet som en irreversibel process utan återvändo? x Varför gör unga honor oftare felaktiga investeringsbeslut? Är det honor med låg kvalité eller beror det på att ettåriga honor saknar tidigare erfarenhet av att häcka? x Hur kan vissa individer ha råd att ha en hög reproduktionsnivå hela livet och samtidigt ha lång livslängd? Eller är deras kostnad i själva verket inte högre än andras? x Hur kan individer så precis följa de optimala investeringsavvägningar de hela tiden måste göra? Eller är vissa investeringar helt enkelt inte optimala som t.ex. den höga immunrespons som ettåriga honor hade?

Dessa frågor kräver ytterligare många års studier inom existerande lång- tidsstudier och en mängd smart uttänkta experiment. Det kommer att vara fruktbart att kombinera nya tekniker från olika vetenskapliga discipliner som molekylärgenetik, kvantitativ genetik, immunologi, fysiologi, statistiska metoder med ekologiska metoder och ett evolutionärt perspektiv. Slutligen kan en luttrad doktorand i zooekologi konstatera: ’Att åldras tycks vara det enda tillgängliga sättet att få ett långt liv’ och ’ingenting kan förstås inom biologin utom i ljuset av evolutionen’.

33 Summary in Polish (Streszczenie)

Ludzie od zawsze interesowali siĊ zmianami zachodzącymi z wiekiem w ich organizmach, czego dowody znajdujemy zarówno w sztuce i literaturze jak i w pierwszych opracowaniach medycznych. Przez dáugi okres czasu nie zdawano sobie jednak sprawy z tego, jak bardzo zjawisko to jest powszechne w caáej naturze. Tymczasem starzenie siĊ i innego typu zmiany zachodzące z wiekiem dotyczą nie tylko ludzi i zwierząt, ale równieĪ roĞlin, a nawet mikroorganizmów. Mimo tak powszechnie obserwowanego wpáywu wieku na organizmy Īywe oraz duĪego zainteresowania tym zjawiskiem, teorie naukowe próbujące wyjaĞniü podstawy tych zmian zostaáy uksztaátowane dopiero w XX-stym stuleciu. Podstawa tych teorii leĪy w rozróĪnieniu dwóch linii w obrĊbie organizmu: linii somatycznej i páciowej, co pozwala wyjaĞniü dlaczego zachodzące z wiekiem zmiany nie są przekazywane nastĊpnym pokoleniom. Istotnym zaáoĪeniem, szczególnie dla hipotez wyjaĞniających zjawisko starzenia siĊ, jest takĪe fakt dysponowania przez organizm ograniczoną iloĞcią zasobów, które muszą zostaü podzielone pomiĊdzy poszczególne funkcje, jak na przykáad rozród, odpornoĞü czy teĪ naprawy uszkodzonych tkanek. Kompromis w rozdziale puli zasobów pomiĊdzy róĪnymi funkcjami organizmu jest uwaĪany za mechanizm wyjaĞniający istnienie kosztów reprodukcji, kiedy nadmierne inwestowanie w rozród powoduje spadek kondycji, w skrajnych przypadkach prowadzący nawet do Ğmierci. Te dwa zaáoĪenia stanowią podstawĊ licznych hipotez próbujących wyjaĞniü mechanizmy zmian zachodzących z wiekiem w róĪno- rodnych funkcjach organizmu. Pomimo tego, Īe zjawiska dojrzewania i starzenia siĊ cieszą siĊ obecnie ogromną popularnoĞcią wĞród naukowców, wiĊkszoĞü prowadzonych badaĔ dotyczy ludzi i trzymanych w niewoli zwierząt. W związku z tym nadal niewiele jest wiadomo na temat przyczyn zmian związanych z wiekiem u gatunków Īyjących dziko. Celem niniejszej pracy byáo badanie zmian w funkcjach organizmu zachodzących z wiekiem, a takĪe testowanie ich prawdopodobnych przyczyn w oparciu o najwaĪniejsze hipotezy. Studium to obejmuje zarówno zmiany w prawdopodobieĔstwie przeĪycia i w sukcesie rozrodczym, jak i zmiany w kosztach reprodukcyjnych i decyzjach inwestycyjnych opisanych w teorii historii Īyciowych (ang. life-history theory). Zmiany te są najsilniej akcentowane w początkowym i koĔcowym etapie Īycia, dlatego teĪ na te wáaĞnie etapy poáoĪony jest w tej pracy najwiĊkszy nacisk.

34 Badania przeprowadzono w populacji muchoáówki biaáoszyjej (Ficedula albicollis) zasiedlającej poáudniową czĊĞü baátyckiej wyspy Gotlandii. Populacja ta stanowi idealny model do badaĔ zmian zachodzących w organiĨmie z wiekiem, poniewaĪ jest populacją izolowaną. Oznacza to, Īe wiĊkszoĞü osobników, które urodziáy siĊ na Gotlandii powraca tam kaĪdego roku po przezimowaniu w Afryce. Badania nad muchoáówką biaáoszyją zostaáy zapoczątkowane na Gotlandii juĪ 1980 roku. KaĪdego roku znacząca wiĊkszoĞü osobników jest odáapywana, oraz gromadzone są dane dotyczące ich sukcesu rozrodczego i przeĪywalnoĞci. Zaowocowaáo to ogromną bazą danych, pozwalającą na dáugoterminowe analizy. WiĊkszoĞü osobników zostaje zaobrączkowana juĪ jako pisklĊta w gnieĨdzie, dlatego teĪ w póĨniejszym okresie ich dokáadny wiek jest áatwy do okreĞlenia na podstawie unikalnego numeru obrączki. Muchoáówka biaáoszyja jest stosunkowo krótko Īyjącym ptakiem (max. 9 lat), który wykazuje typowe takĪe dla innch gatunków ptaków, związane z wiekiem zmiany w przeĪywalnoĞci i sukcesie rozrodczym. Zarówno prawdopodobieĔstwo przeĪycia do nastĊpnego roku, jak i sukces rozrodczy wzrasta u muchoáówki w początkowych latach Īycia, a nastĊpnie maleje pod koniec Īycia (zobacz równieĪ rozdziaá I).

Dlaczego niektóre osobniki Īyją dáuĪej niĪ inne?

Wyniki badaĔ opisanych w rozdziale I jednoznacznie wykazują, Īe najwaĪniejszym powodem róĪnic w przeĪywalnoĞci miĊdzy poszczególnymi osobnikami są u muchoáówki biaáoszyjej róĪnice w fenotypowej jakoĞci osobników. RóĪnice te zaznaczają siĊ najwyraĨniej w pierwszych latach Īycia, kiedy osobniki niskiej jakoĞci mają zarówno niski sukces rozrodczy jak i niską przeĪywalnoĞü. Ponadto, dáugo Īyjące samice wykazują wysoki sukces rozrodczy przez wiĊkszoĞü Īycia, z jedynie niewielkim spadkiem pod koniec Īycia, co dodatkowo potwierdza ich wysoką jakoĞü. Niemniej jednak, wyniki te nie wyjaĞniają dlaczego osobniki róĪnią siĊ miĊdzy sobą jakoĞcią. W badaniach opisanych w rozdziale V testowane byáy moĪliwe przyczyny tych róĪnic poprzez porównanie samic muchoáówki Īyjących relatywnie dáugo lub krótko. DáugoĞü Īycia, a zatem takĪe fenotypowa jakoĞü samicy, wydaje siĊ zaleĪeü od warunków jakich doĞwiadczyáa ona podczas wzrostu. Dáugo Īyjące samice byáy, jako pisklĊta tuĪ przed opuszczeniem gniazda, stosunkowo duĪe, ale równoczeĞnie lekkie w porównaniu z samicami Īyją- cymi krótko. BezpoĞrednie przyczyny takiego wzorca są dyskutowane w rozdziale V. Ponadto, dáugo Īyjące samice pochodziáy od ojców ze stosunkowo duĪym ornamentem páciowym, którym u muchoáówki biaáoszyjej jest biaáa plama na czole. JednakĪe utworzenie pary z bardziej atrakcyjnym samcem wydaje siĊ zmniejszaü prawdopodobieĔstwo przeĪycia samicy. Jest tak prawdopodobnie dlatego, Īe atrakcyjne samce czĊsto redukują swój

35 wysiáek rozrodczy, co z kolei jest kompensowane przez samice, zwiĊkszając równoczeĞnie jej wáasny wysiáek, a zatem i koszt reprodukcyjny.

Dlaczego sukces rozrodczy wzrasta w máodym wieku?

W badaniach opisanych w rozdziale I testowane są trzy gáówne hipotezy wyjaĞniające wzorzec początkowego wzrostu sukcesu rozrodczego. NajwiĊksze znaczenie dla wzrostu sukcesu rozrodczego we wczesnych latach Īycia muchoáówki okazaáy siĊ mieü róĪnice w przeĪywalnoĞci miĊdzy osobnikami róĪnej jakoĞci. JednakĪe ten mechanizm, jakkolwiek znakomicie wyjaĞniający wzrost sukcesu rozrodczego z wiekiem na poziomie populacji, nie wyjaĞnia takiego wzrostu w sukcesie rozrodczym danego osobnika. Mimo, Īe wiele innych badaĔ dowodzi, iĪ indywidualny wzrost sukcesu rozrodczego spowodowany jest zdobywaniem kompetencji koniecznych do efektywnego rozrodu, moje analizy takiej zaleĪnoĞci nie wykazaáy. Osobniki w tym samym wieku, ale róĪniące siĊ wczeĞniejszym doĞwiadczeniem rozrodczym, nie róĪniáy siĊ pod wzglĊdem sukcesu rozrodczego. W takim przypadku, indywidualny wzrost sukcesu rozrodczego w pierwszych latach Īycia jest najlepiej táumaczony przez mechanizm opisany hipotezą optymalizacji wysiáku rozrodczego. Zgodnie z tą hipotezą osobniki máode, o wy- sokim prawdopodobieĔstwie przeĪycia i przyszáego rozrodu, powinny in- westowaü wiĊcej w utrzymanie dobrej kondycji, a mniej w rozród. Z wiekiem przewidywane jest stopniowe przesuniĊcie inwestycji w kierunku rozrodu. Hipoteza ta znajduje czĊĞciowe potwierdzenie w badaniach opisanych w rozdziale III i IV, gdzie máode samice wykazują relatywnie wysoką odpowiedĨ immunologiczną, sugerującą zwiĊkszone inwestycje we wáasną kondycjĊ w tej grupie ptaków w porównaniu z osobnikami starszymi. Jed- nakĪe, jak wspomniano wyĪej, w badanej populacji dáugo Īyjące osobniki wysokiej jakoĞci nie wykazują wzrostu w sukcesie rozrodczym na początku Īycia, gdyĪ jest on wysoki juĪ w pierwszym roku Īycia. Za to osobniki Īyjące krócej początkowo osiągają niski sukces rozrodczy, który nastĊpnie wzrasta z wiekiem. Sugeruje to, Īe wzorzec optymalizacji wysiáku rozrodczego jest modyfikowany w odniesieniu do indywidualnej jakoĞci, kiedy to osobniki wysokiej jakoĞci prawdopodobnie nie muszą ograniczaü wysiáku rozrodczego w pierwszych latach Īycia.

Sukces rozrodczy pod koniec Īycia

U wiĊkszoĞci organizmów obserwowany jest spadek sukcesu rozrodczego w póĨnym wieku, co jest zwykle táumaczone ogólnym spadkiem wydajnoĞci organizmu na skutek starzenia siĊ. Niemniej jednak w niektórych przypadkach obserwowano wzrost sukcesu rozrodczego pod koniec Īycia osobnika. Taki scenariusz jest przewidywany przez wspomnianą wyĪej hipotezĊ optymalizacji wysiáku rozrodczego, gdy osobniki stare, o niskiej

36 przewidywanej dáugoĞci Īycia i niskim prawdopodobieĔstwie nastĊpnego rozrodu inwestują wszystkie zasoby w ostatni rozród. Mechamizm ten nazywany jest inwestycją terminalną (ang. terminal investment), poniewaĪ zwykle wystĊpuje tuĪ przed Ğmiercią osobnika. Jak opisano w rozdziale V, w badanej populacji samice, które nie przeĪyáy do nastĊpnego roku po rozrodzie w piątym roku Īycia, wykazują znacznie wyĪszy sukces rozrodczy niĪ samice, które przeĪyáy, co sugeruje wystĊpowanie wspomnianego wyĪej mechanizmu. JednakĪe w szóstym roku Īycia samice wykazują odwrotny wzorzec, gdzie samice, które przeĪyáy do nastĊpnego roku mają równieĪ wysoki sukces rozrodczy. ZaleĪnoĞü ta nie wyklucza zwiĊkszonej inwestycji w rozród w póĨnym wieku i moĪe wynikaü z dalszych róĪnic w indywidualnej jakoĞci. JeĞli róĪnice w indywidualnej jakoĞci pozostają istotne równieĪ w póĨnych latach Īycia, nieliczne osobniki które doĪywają szeĞciu lub siedmiu lat mogą byü w stanie utrzymaü wysoki poziom wysiáku rozrodczego przez caáe Īycie, bez znaczących kosztów dla ich kondycji lub prawdopodobieĔstwa przeĪycia. JednakĪe nawet osobniki bardzo wysokiej jakoĞci podlegają prĊdzej czy póĨniej procesowi starzenia siĊ. Na poziomie populacji starzenie siĊ u badanej muchoáówki biaáoszyjej wyraĪa siĊ poprzez spadek prawdopodobieĔstwa przeĪycia i sukcesu rozrodczego obserwowany pod koniec Īycia. W eksperymencie opisanym w rozdziale IV samice powyĪej czwartego roku Īycia wykazują, charakterystyczny dla procesu starzenia siĊ równoczesny spadek sukcesu rozrodczego i odpowiedzi immunologicznej. JednakĪe przeĪywalnoĞü, jak i róĪne parametry sukcesu rozrodczego zaczynają spadaü w róĪnym wieku u róĪnych osobników, co jest równieĪ widoczne w ogólnych trendach zmian zachodzących z wiekiem (zobacz rozdziaá I). Tak wiĊc, opisana równieĪ w rozdziale V, pozytywna zaleĪnoĞü miĊdzy prawdopodobieĔstwem przeĪycia a sukcesem rozrodczym moĪe wynikaü z róĪnego tempa starzenia siĊ osobników, gdzie niektóre osobniki w szóstym roku Īycia mają nadal wysoki sukces rozrodczy i wysokie szanse na przeĪycie podczas gdy u innych osobników te parametry są niskie na skutek zmian nastĊpujących w wyniku procesu starzenia siĊ.

Koszty reprodukcji

Jak opisano powyĪej, organizm dysponuje ograniczonymi zasobami, w związku z czym zwiĊkszenie inwestycji w rozród powoduje niedoinwestowanie innych funkcji organizmu. Zjawisko to okreĞlane jest jako koszt reprodukcji. W rozdziale II pokazano, Īe koszt ten moĪe byü róĪny dla osobników w róĪnym wieku. Máode, jednoroczne samice, które zostaáy zmuszone do wychowywania eksperymentalnie zwiĊkszonej liczby piskląt, skáadaáy w nastĊpnym roku mniej jaj niĪ samice wychowujące naturalną liczbĊ piskląt. Takiej zaleĪnoĞci nie wykazaáy samice starsze, co sugeruje, Īe ich koszty reprodukcyjne są niĪsze niĪ dla samic máodych, co

37 moĪe wynikaü z róĪnic we wczeĞniejszym doĞwiadczeniu rozrodczym lub teĪ z róĪnic w indywidualnej jakoĞci. Koszty reprodukcji mogą w okreĞlonych warunkach zostaü zredukowane, poprzez na przykáad gniazdowanie na wysokiej jakoĞci terytorium o wysokiej dostĊpnoĞci pokarmu. W badaniu opisanym w rozdziale III samice gnieĪdĪące siĊ na terytoriach bogatych w gąsienice, bĊdące najlepszym Ĩródáem pokarmu dla muchoáówki, wykazywaáy wyĪsze prawdopodobieĔstwo przeĪycia i mniejszy spadek masy ciaáa podczas sezonu rozrodczego. Wyniki te wskazują jednoznacznie, Īe jakoĞü terytorium i dostĊpnoĞü pokarmu moĪe zredukowaü koszty rozrodu. Máode samice wykazaáy jednakĪe ograniczone korzyĞci z gniazdowania na terytoriach bogatych w gąsienice. Ich spadek masy ciaáa podczas rozrodu nastĊpowaá niezaleĪnie od rodzaju terytorium, co moĪe wynikaü z opisanych powyĪej ogólnie wyĪszych kosztów rozrodu máodych osobników. TakĪe bardzo stare samice wykazaáy zmniejszone korzyĞci z gniazdowania na terytoriach bogatych w gąsienice, gdyĪ ich prawdopodobieĔstwo przeĪycia nie róĪniáo siĊ pomiĊdzy poszczególnymi typami terytoriów. MoĪe to wynikaü zarówno z przewidywanego zwiĊkszenia przez te osobniki wysiáku rozrodczego, jak i bezpoĞrednio z faktu, Īe te osobniki starzeją siĊ i dlatego mają niską przewidywaną dáugoĞü Īycia, niezaleĪnie od ponoszonych kosztów rozrodu.

Podsumowanie

Niniejsza praca jest jednym z najszerszych opracowaĔ dotyczacych zmian w przeĪywalnoĞci, sukcesie rozrodczym i parametrach historii Īyciowych u dziko Īyjącego gatunku ptaka. Wykazuje ona, Īe dáugoĞü Īycia osobnika uzaleĪniona jest od róĪnorodnych czynników, od indywidualnej jakoĞci, wynikającej z warunków rozwojowych poczynając, a na wyborze partnera i decyzjach inwestycyjnych koĔcząc. Co wiĊcej, praca ta pokazuje, Īe róĪnorodne mechanizmy ksztaátujące zmiany w sukcesie rozrodczym, decyzje inwestycyjne, koszty rozrodu i redukcjĊ tych kosztów, mogą dziaáaü róĪnorodnie w zaleĪnoĞci od wieku, a czasem równieĪ od indywidualnej jakoĞci osobnika. Praca ta pokazuje jak waĪne są czynniki wieku i indywidualnej jakoĞci w ksztaátowaniu tak waĪnych cech organizmu jak przeĪywalnoĞü i sukces rozrodczy. Dlatego teĪ powinny byü one zawsze brane pod uwagĊ w badaniach dotyczących historii Īyciowych.

38 Acknowledgements

First of all, I would like to thank my supervisors Lars Gustafsson and Anna Qvarnström for giving me the opportunity to undertake my PhD at Uppsala University and work on the famous Gotlandic population of collared flycatchers. Thank you both for endless discussions, advice and for the proof- reading my papers over and over. Lars, thank you for giving me the space to develop my own ideas and do things my own way. I hope you do not regret these decisions now, because in my opinion it is exactly what made you the best supervisor I could have. Anna, many thanks for always having time to answer my numerous questions, not only the scientific ones. Ingela and Marianne thank you for all the help with administrative tasks and all the other things. I would have been lost without you. Mariusz, thank you for showing me that there is a whole world outside Jagiellonian University and for teaching me not only how to work with birds, but also many other aspects of field-work. I owe very special thanks to all the people who agreed (despite my warn- ings) to work for and with me on Gotland during these years: Adam, Elisa- beth, Hajnalka, Ingrid, Joasia, Kinga, Mirek, Mårten, Natalia, Sara, Silvia & Staszek. You have done a great job! I would have never achieved so much without you. Anna D., Anna Q., Blandine, Chris, Katherine, Jukka, Mariusz, Måns, Mårten, Natalia & Thor, it was great to have you around during all these years of my fieldwork. You have made the hard time of fieldwork much easier with your advice and great company. Two girls made the office I was sitting in the nicest place in the Depart- ment. Nina and Johanna, you have been great companions for all these years, always ready to talk science and non-science, always ready to help, and always ready to laugh. This place would not have been the same without you. I wish I could take you both with me to wherever it is I am going now. The time I spent in Jyväskylä was one the best during my PhD studies thanks to Rauno and Heli, Elina, Hanna, Mario, Inez, Judith, Andres, Chris- topher, Joasia and many other people from the Department of Biological and Environmental Science. Thank you all for the great time and all I learnt. Marlena, Artur and Wojciech, thank you for inviting me to spend some time at Adam Mickiewicz University in PoznaĔ and for all your help with my PCR analyses. These results are not in my thesis, because of the other

39 complications, but I learned a lot from you and had a fantastic time in PoznaĔ. The Department of Animal Ecology in Uppsala is a very special one since there is always someone here ready to talk science, help with statistics and discuss other problems. Göran Arnqvist, a special thanks to you for all the hours you spent explaining statistical problems to me. Mats, thank you for your optimism, you raised my spirits many times. Reija, Eevi and Marta, it is impossible to overrate the importance of all the help I got from you while working in the lab. Måns, thank you so much for all your enthusiasm and for endless discussions of science and life when I just started my studies and was a bit lost. Very special thanks to Chris, Damian, Ted and especially Sandra for correcting my hopeless English and to Anna D. and Magdalena for comments to summary in Polish. My time at Uppsala University would not have been the same also without all the other PhD students, post-docs and visitors who made the Depart- ment of Animal Ecology a fun place: Alex, Anders Ö., Björn, Cath, Claudia, Emma, Göran, Johanna A., John, Jonas, Jukka, Kalev, Katherine, Mari, Maria, Marnie, Martin, Mårten, Niclas, Olivia, Olle, Sara, Sarah & Theresa. Thank you all for the time I spent with you here. There is also the bunch of people who made my life so much better, especially during the first years when they made Uppsala feel like home. Mi- chael, thanks for introducing me to The Bunch and for being a true friend for all the years before and during my PhD studies. Silke, thank you for taking me in to your apartment when I started my studies and had no place to stay. You were a great “flat” mate. Marta and Debora, it was a pleasure to share an apartment with you. Marta, if you ever want to wash a cat again, just give me call. Martina, Ariane & Claudia thank you for endless talks on important topics, serious and not serious. They have been priceless to me. To the First Bunch members, Amilcar, Daniela, David, Johiris, Marta, Michael, Silke, Sonia, Thomas & Vittorio: thanks for all the hours of good laughs with you, watching movies and photos, and for all the nights of dancing. You made me a very happy person by being both here with me and here for me, thank you. Magdalena, Dominik, Betty & Artur, thank you for all the great times we spent together. You all became very important to me- especially you, Magda. Thank you for all these years of true friendship, endless talks and support. I also owe a very special thanks to a number of people a bit older than me, who were always ready to support me with kind words or advice, and to restore my self confidence. Hanna and Wáadysáaw Kuhl, thanks for giving me the feeling that there is someone in Uppsala I can always count on and who is always truly interested in my work and its progress. Maágorzata Packalen and Hillevi Thorell, thank you for all your invaluable advice, friendliness and all the hours laughing.

40 Last, but not least, I would like to thank my family for all the support I received from them. DziĊkujĊ Wam, Rodzice za ogromne wsparcie, które zawsze od Was otrzymywaáam. Mamo, dziĊkuje Ci, Īe zawsze zachĊcaáaĞ mnie do stawiania nowych pytaĔ i Īe juĪ dawno pokazaáaĞ mi jak ciekawa moĪe byü biologia. Tato, nawet nie wiesz jak waĪne są dla mnie Twoje slowa Īebym száa swoją drogą, bo to jest moje Īycie. DziĊkujĊ Ci za nie. Moje kochane Ciocie: Janina i Maágorzata Witek, dziĊkujĊ Wam, Īe zawsze wspieraáyĞcie mnie w moich decyzjach i radziáy w problemach. Po rozmowie z Wami zawsze miaáam lepszy humor i radoĞniej patrzyáam na Ğwiat. Staszek, Sáoneczko, wiesz? DziĊkujĊ Ci, Īe przyjechaáeĞ ze mną do Uppsali, byáeĞ przy mnie i wspieraáeĞ mnie przez te wszystkie lata. Nie umiem wyraziü jakie to bylo i jest dla mnie waĪne. Jakub, maáy-wielki cudzie, i Tobie teĪ dziĊkujĊ, ze wniosáeĞ tyle Ğmiechu i radoĞci w moje Īycie i Īe zmuszasz mnie do oderwania siĊ od mojej pracy i do bycia przez chwilĊ dzieckiem, razem z Tobą.

The part of collared flycatcher study included in this thesis was financially supported by the Swedish National Research Council, Swedish Research Council for Environmental, Agricultural Sciences and Spatial Planning, Zoo- logical Foundation of Uppsala University, Tullbergs Stiftelse (II) & State Committee for Scientific Research of Poland (II & IV).

41 References

Alatalo, R. V., Lundberg, A. & Glynn, C. (1986). Female pied flycatchers choose territory quality and not male characteristics. Nature 323(6084): 152-153. Arking, R. (1998). Biology of Ageing: observations and principles. 2nd Edition. Sinauer Associates Inc. Sunderland. Bonneaud, C., Mazuc, J., Chastel, O., Westerdahl, H. & Sorci, G. (2004). Terminal investment induced by immune challenge and fitness traits associated with major histocompatibility complex in the House sparrow. Evolution 58(12): 2823- 2830. Bregnballe, T. (2006). Age-related fledgling production in great cormorants Pha- lacrocorax carbo: influence of individual competence and disappearance of phenotypes. Journal of Avian Biology 37(2): 149-157. CichoĔ, M. (2003). Does prior breeding experience improve reproductive success in collared flycatcher females? Oecologia 134(1): 78-81. Clutton-Brock, T. H. (1984). Reproductive effort and terminal investment in iteroparous animals. The American Naturalist 123(2): 212-229. Clutton-Brock, T. H. (1988). Reproductive success. Studies of individual variation in contrasting breeding systems. The University of Chicago Press, Chicago and London. Coulson, J. C. (1968). Differences in the quality of birds nesting in the centre and on the edges of a colony. Nature 217: 478-479. Croxall, J. P., Rothery, P. & Crisp, A. (1992). The effect of maternal age and experience on egg-size and hatching success in Wandering albatrosses Diomedea ex- culans. Ibis 134(3): 219-228. Curio, E. (1983). Why do young birds reproduce less well?. Ibis 125:400-404. Espie, R. H. M., Oliphant, L. W., James, P. C., Warkentin, I. G. & Lieske, D. J. (2000). Age-dependent breeding performance in Merlins (Falco columbarius). Ecology 81(12): 3404-3415. Fisher, R. A. (1930). The genetical theory of natural selection. Clarendon Press, Oxford. Forslund, P. & Larsson, K. (1992). Age-related reproductive success in the barnacle goose., Journal of Animal Ecology 61: 195-204. Forslund, P. & Pärt, T. (1995). Age and reproduction in birds – hypotheses and tests, Tends in Ecology and Evolution 10(9), pp. 374-378. Futuyma, D. J. (1997). Evolutionary Biology 3rd Edition. Sinauer Associates, Inc. Sunderland, Massachusetts. Gustafsson, L. (1986). Lifetime reproductive success and heritability: empirical support for Fisher’s fundamental theorem. The American Naturalist 128: 761- 764. Gustafsson, L. (1989). Lifetime reproductive success in the Collared Flycatcher. In: Lifetime reproductive success in birds. (ed. I. Newton), pp. 75-89. New York: Academic Press.

42 Gustafsson, L. & Pärt, T. (1990). Acceleration of senescence in the collared flycatcher Ficedula albicollis by reproductive costs. Nature 347(6290): 279-281. Gustafsson, L. & Qvarnström, A. (2006). A test of the “Sexy Son” hypothesis: Sons of polygynous collared flycatchers do not inherit their fathers’ mating status. The American Naturalist 167 (2):297-302. Haavie, J., Borge, T., Bureš, S., Garamszegi, L. Z., Lampe, H. M., Moreno, J., Qvarnström, A., Török, J. & Saetre, G.-P. (2004). Flycatcher song in allopatry and sympatry – convergence, divergence and reinforcement. Journal of Evolu- tionary Biology 17:227-237. Hemborg, C. & Merilä, J. (1998). A sexual conflict in collared flycatchers, Ficedula albicollis: Early male moult reduces female fitness. Proceedings of the Royal Society of London B 26: 2003-2007. Hunt, J., Buissiere, L. F., Jennions, M. D. & Brooks, R. (2004). What is genetic quality? Trends in Ecology and Evolution 19(6): 329-333. Ilmonen, P., Taarna, T. & Hasselquist, D. (2000). Experimentally activated immune defense in female pied flycatchers results in reduced breeding success. Proceed- ings of the Royal Society of London B 267(1444): 665-670. Kirkwood, T. B. L. (1977). Evolution of ageing. Nature 270: 301- 304. Martinez, D. E. & Leviton, J. S. (1992). Asexual metazoans undergo senescence. Proceedings of the National Academy of Sciences 89: 9921-9923. Medawar, P. B. (1952). An unsolved problem of biology. H.K. Lewis, London. Mitrus, C., Walankiewicz, W., Czeszczewik, D. & Jablonski, P. M. (1996). Age and arrival date of Collared Flycatcher Ficedula albicollis males do not influence quality of natural cavities used. Acta ornithologica 31(2): 101-106. Newton, I. (1988). Age and reproduction in the Sparrowhawk. In: Reproductive success (Clutton-Brock T.H. ed), pp. 201-219. University of Chicago Press. Pärt, T. & Gustafsson, L. (1989). Breeding Dispersal in the Collared Flycatcher (Ficedula albicollis): Possible Causes and Reproductive Consequences. Journal of Animal Ecology 58(1): 305-320. Pärt, T. (1991). Philopatry pays: A comparison between collared flycatcher sisters. The American Naturalist. 138(3): 790-796 Pärt, T., Gustafsson, L. & Moreno, J. (1992). "Terminal investment" and a sexual conflict in the collared flycatcher (Ficedula albicollis). The American Naturalist 140(5): 868-882. Pärt, T. (1995). Does breeding experience explain increased reproductive success with age? An experiment. Proceedings of the Royal Society of London B 360(1357): 113-117. Qvarnström A. (1997). Experimentally increased badge size increases male competi- tion and reduces male parental care in the collared flycatcher. Proceedings of the Royal Society of London B 264:1225-1231. Qvarnström A. (1998). Sexual selection in the collared flycatcher (Ficedula albicollis). Acta Universitatis Upsaliensis, PhD Thesis, Uppsala University, Sweden. Ricklefs, R. E. & Scheuerlein, A. (2001). Comparison of aging-related mortality among birds and mammals. Experimental Gerontology 36(4-6): 845-857. Rose, M. R. (1991). The Evolutionary Biology of Ageing. Oxford University Press, Oxford. Rowe, L. & Houle, D. (1996). The lek paradox and the capture of genetic variance by condition dependent traits. Proceedings of the Royal Society of London B 263: 1415-1421. Saether, B.-E. (1990). Age specific variation in reproductive performance of birds. Current Ornithology 7: 251-283.

43 Sanz, J. J. & Moreno, J. (2000). Delayed senescence in a southern population of the pied flycatcher (Ficedula hypoleuca). Ecoscience 7(1): 25-31. Sasvári, L., Hegyi, Z., Csörgõ, T. & Hahn, I. (2000). Age-dependent diet change, parental care and reproductive cost in tawny owls Strix aluco. Acta Oecologica, 21: 267-275. Sheldon, B. C. and H. Ellegren (1999). Sexual selection resulting from extrapair paternity in collared flycatchers. Animal Behaviour 57(2): 285-298. Stearns, S.C., (1992). The evolution of life histories, Oxford University Press, New York. Svedin, N. (2006). Natural and sexual selection in natural hybrid zone of Ficedula flycatchers. Acta Universitatis Upsaliensis, PhD Thesis, Uppsala University, Sweden. Svensson L. (1992). Identification guide to European passerines. Märtsatryk, Stockholm. Veen, T., Borge, T., Griffith, S. C., Saetre, G-P., Bureš, S., Gustafsson, L. & Shel- don, B. C. (2001). Hybridization and adaptive mate choice in flycatchers. Na- ture 411(6833): 45-50. Wiley, C. (2006). Speciation – What can be learned from Flycatcher hybrid zone? Acta Universitatis Upsaliensis, PhD Thesis, Uppsala University, Sweden. Williams, G. C. (1957). Pleiotropy, natural selection and the evolution of senescence. Evolution 11: 398-411. Williams, G. C., (1966). Natural selection, the costs of reproduction, and a refine- ment of Lack’s principle. The American Naturalist 100:687-690.

Acta Universitatis Upsaliensis Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology 288

Editor: The Dean of the Faculty of Science and Technology

A doctoral dissertation from the Faculty of Science and Technology, Uppsala University, is usually a summary of a number of papers. A few copies of the complete dissertation are kept at major Swedish research libraries, while the summary alone is distributed internationally through the series Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology. (Prior to January, 2005, the series was published under the title “Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology”.)

ACTA UNIVERSITATIS UPSALIENSIS Distribution: publications.uu.se UPPSALA urn:nbn:se:uu:diva-7787 2007