Flight muscle atrophy and predation risk in breeding birds

Functional Ecology

2000

14

, 115–121

© 2000 British Ecological Society

115

Blackwell Science, Ltd

Flight muscle atrophy and predation risk in breeding birds

J. S. VEASEY, D. C. HOUSTON and N. B. METCALFE

Ornithology Group, Division of Environmental & Evolutionary Biology, Graham Kerr Building, IBLS, University of Glasgow, Glasgow G12 8QQ, UK

Summary

1.

The speed with which small birds can get airborne is critical to the effectiveness oftheir escape response when attacked by a predator. However, take-off ability is likelyto be affected by physiological changes occurring as a result of egg formation.

2.

To investigate whether reduced take-off velocity is a cost of reproduction, thephysiological costs of egg production in the Zebra Finch (

Taeniopygia guttata

) wereexperimentally manipulated by varying both the number of eggs a female laid andthe quality of her prelaying diet. The effect of changes in postlaying flight musclecondition and body mass upon alarmed flight take-off velocity (a measure of escape-ability in birds) was subsequently measured.

3.

Changes in muscle condition were found to correlate positively with changes invarious measures of flight velocity: treatments that caused the greatest declinesin muscle condition during egg-laying were associated with the greatest declines inflight performance over this period. In contrast, breeding attempts that caused thesmallest declines in muscle condition were associated with improvements in flightperformance (i.e. birds flew faster at the end of laying than at the start).

4.

These effects were independent of changes in body mass, and occurred post

-

laying, suggesting that the cost of egg production lies primarily in the formation ofthe eggs, rather than in carrying them as other studies had suggested. The observedtrade-off between muscle loss resulting from egg production and escape ability couldhave important implications for the evolution of optimal clutch size in birds.

Key-words

: Clutch size, pectoral muscle,

Taeniopygia guttata

, trade-off

Functional Ecology

(2000)

14

, 115–121

Introduction

One of the main assertions of life-history theory isthat reproduction is costly, and current reproductiveeffort can diminish the chances or amount of futurereproduction (Stearns 1992). However, the underlyingmechanisms of how reproduction affects survival arepoorly understood (Tatar & Carey 1995). A recentreview showed that the majority of studies (96·9%)investigating the costs of reproduction in birds haveemphasized the postlaying phase and so may havemissed costs occurring earlier in the breeding cycle(Monaghan & Nager 1997). This emphasis on post-laying costs may have arisen for two reasons. Firstit is easier to manipulate the number of chicksreared than it is to manipulate the number of eggslaid. Secondly, in the past, egg production has beenconsidered as inexpensive owing to the apparentease with which birds replace lost eggs or even entireclutches (Monaghan & Nager 1997). However, evid-ence is accumulating that egg production is costlyin terms of energetic expenditure (Ward 1996), female

survival (McCleery

et al

. 1996), subsequent breedingsuccess (Heaney & Monaghan 1995; Monaghan,Nager & Houston 1998) and muscle condition(Kendall, Ward & Bacchus 1973; Jones 1991; Houston

et al

. 1995a; Houston, Donan & Jones 1995b,c;Monaghan

et al

. 1998).Recent work has shown that a wide range of bird

species (21 out of the 29 for which muscle volumechanges have been studied) exhibit declines in musclecondition during the laying period (Houston

et al

.1995a). It has been suggested that these declines rep-resent a direct contribution of muscle proteins to eggproduction (Kendall

et al

. 1973; Jones & Ward 1976;Houston

et al

. 1995b,c). Birds can be classified ascapital breeders relying upon stored nutrients suchas those that might exist within the flight muscles, orincome breeders relying upon daily dietary intakefor egg formation (Drent & Daan 1980). It has typic-ally been assumed that small passerines must beincome breeders because clutch masses often repres-ent a high proportion of the female’s body mass.However, it has been shown that in female Zebra

FEC391.fm Page 115 Wednesday, March 1, 2000 5:58 PM

116

J. S. Veasey

et al.

© 2000 British Ecological Society,

Functional Ecology

,

14

, 115–121

Finches, muscle proteins are transferred directly to eggsduring their formation (Houston

et al

. 1995b,c), sothat females may subsequently undergo a reductionof around 14% in protein from their pectoral muscles(Houston

et al

. 1995a) despite the availability of

adlibitum

food throughout the laying cycle.Similar results have been reported for Red-Billed

Quelea (

Quelea quelea

), which have been shown to lose16% of their muscle mass during breeding despite anabundance of food (Kendall

et al

. 1973). This loss ofmuscle condition is roughly equivalent to being halfway between the peak condition ever recorded forthis species, and the condition of birds that have diedfrom starvation (Kendall

et al

. 1973; Jones & Ward 1976).Costs associated with such a substantial depletion ofprotein reserves are likely to be particularly high if theproteins are derived from contractile muscle fibresinvolved in flight. The flight muscle ratio (flight musclemass/body mass) positively correlates with residuallift capacity (Marden 1987), and so one might expecta reduction in foraging efficiency (see Heaney &Monaghan 1995) or speed of alarmed flight to occur asa result of reproduction in capital breeders. Since gettingoff the ground and gaining height rapidly is critical insurviving attacks from mammalian and avian predators(Rudebeck 1950; Howland 1974; Page & Whitacre 1975;Cresswell 1993; Bednekoff 1996), predation risk in smallbirds can be assumed to increase if alarmed flight speedsare reduced. Changes in flight muscle condition mighttherefore be expected to affect maternal survival rates.

The aim of this investigation, therefore, is to discoverthe extent to which the maximal flying ability of femalebirds, and hence their ability to escape from predators,is affected by losses in muscle condition resulting fromlaying. To do this trials were run in which the same groupof female Zebra Finches were allowed to lay a succes-sion of clutches. The extent of the change in musclecondition during each breeding attempt was varied bymanipulating the number of eggs laid and the qualityof the birds’ prebreeding diet. The relationship betweenthe changes in postlaying flight performance and flightmuscle condition was then examined.

Materials and methods

Twenty-four previously established pairs of Zebra

Finches (

Taeniopygia guttata

, Vieillot) were housed inseparate cages (0·6

×

0·5

×

0·4 m

3

) and maintained onan artificial 14L:10D light regime throughout theexperiment. Four treatments were utilized in order tomanipulate changes in female flight muscle conditionresulting from egg-laying. These treatments consistedof manipulations of both prebreeding diets and clutchsizes. The two possible prelaying diets were a low-proteindiet of

ad libitum

mixed seed, and a high-protein dietof

ad libitum

mixed seed with a high-quality proteinsupplement containing soya protein and homogenizedboiled hen’s eggs. This high-quality protein diet isknown to reduce the extent to which female ZebraFinches draw upon their pectoral muscles when pro-ducing a clutch (Selman & Houston 1996a). The twoclutch manipulation treatments encouraged the lay-ing of relatively large or small clutches: removal ofthe first four eggs on the day each was laid inducedthe laying of a larger clutch, while the addition of afalse egg daily to the nestbox for 4 days, starting 4 daysafter pairing, induced birds to lay smaller clutches(see Haywood 1993a,b). These two manipulationswere combined to make four treatments which mani-pulated both the clutch size produced and the nutri-tional cost of producing a given clutch (see Table 1).In all trials birds were given free access to water, gritand cuttlebone. The sexes were kept apart until thebreeding portion of each trial was started, at whichpoint pairs were reunited, cages provided with nestbuilding material and nestboxes, and all diets revertedto the low-protein type for the duration of the layingattempt. All eggs were removed from the nestboxon the day of laying, weighed and replaced with anartificial egg where appropriate.

Once all the birds had completed their clutches(when 2 days had passed without an egg being laid),the nestboxes were blocked and the sexes separatedagain. A period of 4–6 weeks was allowed for recov-ery before the onset of the next breeding attempt,during which time the birds were maintained on theappropriate prebreeding diet for the following trial.

Flight performance was assessed by recording thetime taken to complete vertical flights when alarmed,since such escape flights are a better indicator ofchanges in vulnerability to predation than unalarmed

Table 1. Description of the four treatments used in the study

Low-protein prebreeding diet High-protein prebreeding diet

Four eggs removed from clutch Treatment 1 Treatment 2Female induced to lay large clutchon poor-quality diet

Female induced to lay large clutchon high-quality diet.

Four eggs added to clutch Treatment 3 Treatment 4Female induced to lay small clutchon poor-quality diet

Female induced to lay small clutchon high-quality diet

FEC391.fm Page 116 Wednesday, March 1, 2000 5:58 PM

117

Flight muscle atrophy and predation risk

© 2000 British Ecological Society,

Functional Ecology

,

14

, 115–121

flights (Veasey, Metcalfe & Houston 1998). Femaleswere tested for flight performance prior to reunitingthe pairs, so as to obtain a measurement of theirprelaying flight performance (see Veasey

et al

. 1998for details). Once pairs had been re-established, femaleswere flown only on the days on which they laid anegg. Flights were conducted in a separate verticalflight aviary (1·9

×

1

×

1 m

3

), with 0·4 m high verticalbaffles positioned 0·2 m apart either side of a holdingchamber at the base of the aviary, in order to makethe birds fly in as vertical a trajectory as possible.Although body mass has been shown to have only aminor effect if at all on alarmed flight performance inthe Zebra Finch (Veasey

et al

. 1998), females wereflown at a time when their body mass was known tobe most stable, that is to say between 3 and 5 h afterdawn (Dall & Witter 1998), in order to minimize anyconfounding effects of the natural diurnal variationin body mass that is known to occur in the ZebraFinch (Metcalfe & Ure 1995; Dall & Witter 1998).

On each occasion that a bird was flown, it was firsttaken from its home cage, placed in a holding cham-ber and weighed to the nearest 0·1 g. The holdingchamber containing the bird was then placed in asheath at the base of the flight aviary from where itwas then startled into making a vertical escape flighttoward a perch at the top of the aviary. The bird wasthen re-caught, returned to the holding chamber andallowed to recover for a minimum of 30 s before theprocedure was repeated.

Flights were filmed using a camcorder. Each teston a given day consisted of recording three verticalflights for each bird. The measures obtained fromthese replicate flights have been shown to be highlyrepeatable (Veasey

et al

. 1998). The films were playedback on a VCR using a frame by frame facility tomeasure the time taken for each bird to fly a verticaldistance of 0·15, 0·3 and 1·15 m from the top of theholding cup. Mean values from the three flights werethen calculated for each bird, for each of the definedheights on each day on which the bird was flown.

Since the changes in maternal body mass during thecourse of the laying period will reflect changes inthe reproductive tissues within the bird, as well asthe presence or absence of eggs, ‘body mass’ will notgive a satisfactory measure of female condition. How-ever, mass changes are still recorded since there is alarge body of evidence that suggests that mass willaffect flight performance (for examples see Hartman1961; Pennycuick 1969; Metcalfe & Ure 1995). As analternative measure of condition, female flight musclecondition is considered. Although leg muscle condi-tion has been shown to decline in laying female ZebraFinches (Houston

et al

. 1995b), and leg musculatureis an important component in take-off performance(Marden 1987), this is not considered here for two

reasons. First, it would not be possible to monitorchanges adequately in the leg musculature of layingfemale Zebra Finches, and secondly, an analysis ofthe data from Houston

et al

. (1995b), on changes inflight and leg muscle mass during laying, showed thereto be a strong correlation between the two (Pearson’scorrelation:

r

= 0·97,

n

= 7,

P

= < 0·001).The size of the pectoral muscles of females was

measured using a technique modified from Selman& Houston (1996b). Females were measured bothimmediately prior to being reunited with their mate(to give a prelaying measurement), and on the com-pletion of the clutch (two days after the last egg waslaid). Individual birds were placed breast downwardsinto a 20-mm deep tray of dental alginate (Cavex CA37Superior Pink, Cavex, Haarlem, Holland), which givesa faithful body mould without adhering to feathers(Selman & Houston 1996b). Two people were requiredfor this procedure, one to hold the head, tail and the legsclear from the gel, while the other kept the wingsfrom flapping. The alginate sets approximately oneminute after mixing with warm water. Birds wereplaced in the gel

≈

45 s after mixing, so that theywere held in position for as short a time as possible(i.e.

≈

15–20 s). The birds were then eased from themould which was subsequently cut with a blade dorso-ventrally at the mid-point between the fulcra and theposterior portion of the sternum (Fig. 1a). The cutsurfaces of both halves of the mould were thenplaced in a plastic tray to maintain their rigidity, andpushed downwards onto an ink pad. Ten prints (fivefrom each half of the mould) were then taken of thecross-sectional profile of the pectoral region (Fig. 1b).This was carried out immediately upon the setting ofthe alginate to avoid errors resulting from shrinkage ofthe moulds. A horizontal line was then drawn 5 mmperpendicular from the base of the print (approxi-mately the average keel depth size in Zebra Finches)corresponding to where the keel would have beenat its deepest (Fig. 1b). The enclosed area was thenmeasured using a computer plotter (BBC Mastercomputer with cherry digitizer and puck), and a meanof these muscle cross-sectional areas taken of meas-ures from five separate prints, at least two being fromeach half of the mould. An analysis of variance forthe calculation of repeatability (Lessells & Boag 1987)performed across the five measurements confirmed thatthe repeatability was high (

F

19

,

99

= 17·825,

r

= 0·771).The mean cross-sectional pectoral muscle area ishereafter referred to as the muscle condition index.

All treatments outlined in Table 1 were repeatedtwice (with the exception of treatment 4, i.e. thehigh-protein prebreeding diet with the addition ofeggs, which was only run once), in a sequence altern-ating from low to high protein prelaying diets to avoidany longer term effects of continued high proteinsupplementation (see Williams 1996). The percentagechange in muscle condition index from pre- to post-breeding was calculated for each breeding female,

FEC391.fm Page 117 Wednesday, March 1, 2000 5:58 PM

118

J. S. Veasey

et al.

© 2000 British Ecological Society,

Functional Ecology

,

14

, 115–121

along with the corresponding mean percentage changein flight times (for each of the flight parameters)between the day of the first egg and that of the lastegg. All statistical analyses are based on using onemean value of all females per trial (i.e. the samplesize is the number of trials, 7) to avoid problems ofpseudoreplication, while the significance of statisticaltests was evaluated using sequential Bonferroni cor-rection (Rice 1989). Over all seven trials, flight andmuscle condition data were collected for a mean of17 (

±

2·02 SE) females out of the 24 used in each trial.

Results

There were no significant differences between treatmentsin prebreeding muscle condition index (

F

3,115

= 1·645,

P

= 0·183, mean muscle condition index = 79·32 mm

2

±

0·75 SE), prebreeding body mass (

F

3,115

= 0·408,

P

= 0·748, mean body mass = 15·09

±

0·11 SE), andprebreeding flight velocity over all of the threeflight distances (

F

3,115

= 0·735,

P

= 0·533, mean flightvelocity to 0·15 m from take-off = 1·25 ms

±

0·01 SE,

F

3,115

= 0·705,

P

= 0·551, mean flight velocity to0·3 mm from take-off = 1·50 ms

±

0·01 SE,

F

3,115

= 0·455,

P

= 0·715, mean flight velocity to 1·15 mm from take-off = 1·99 ms

±

0·02 SE).For each of the three flight parameters (time to

reach a vertical height of 0·15, 0·30 and 1·15 m), linearregression analysis was used to examine whether thechanges in muscle condition index during breedinginfluenced changes in flight times between the firstand last egg across the seven breeding trials. Changesin muscle condition index produced significant changesin flight times to 0·15 m (

F

1,5

= 12·614, adjusted

R

2

= 0·659,

P

= 0·016,

B

= –1·450

±

0·408; Fig. 2a), to0·30 m (

F

1,5

= 25·383, adjusted

R

2

= 0·803,

P

= 0·004,B = –0·962

±

0·191; Fig. 2b), and to 1·15 m (

F

1,5

= 7·520,adjusted

R

2

= 0·521,

P

= 0·041, B = –1·198

±

0·437;Fig. 2c), all three tests still being significant aftersequential Bonferroni adjustment.

When breeding caused only a small change inpectoral muscle condition, birds tended to fly fasterat the end than at the beginning of egg-laying (i.e.the change in flight time was negative). However, ifthe experimental treatment caused the birds to losea larger amount of muscle condition, then their flightperformance deteriorated during the laying period

Fig. 2. The relationship between changes in female musclecondition index during the course of laying a clutch andchanges in time to fly to a height of (a) 0·15, (b) 0·3 and(c) 1·15 m above the point of take-off. Each data point is themean (+SE) value from a single breeding trial (n = 7 trials),involving an average of 17 ± 2·02 x–y females. Numbersbeside points correspond to treatment type outlined in

Fig. 1. (a) Schematic representation of an alginate mould taken from the pectoralmuscle area of a Zebra Finch showing the position where the mould is cut, andprints consequently taken. (b) Illustration of prints taken from alginate moulds withthe area measured (‘muscle condition index’) shown shaded. The left-hand print istypical of a prebreeding female, whereas the right-hand print is typical of apostbreeding female showing the pronounced keel.

FEC391.fm Page 118 Wednesday, March 1, 2000 5:58 PM

119

Flight muscle atrophy and predation risk

© 2000 British Ecological Society,

Functional Ecology

,

14

, 115–121

(Fig. 2). The incorporation of changes in body massinto these analyses (using multiple regression) did notimprove the models for the flights to distances of 15and 30 cm, but did for the 115-cm flight times (mul-tiple linear regression:

F

2,4

= 16·822, adjusted

R

2

= 0·841,

P

= 0·011, B = –0·970

±

0·261) though changes inbody mass alone were not significant over this height(linear regression:

F

1,5

= 5·570, adjusted

R

2

= 0·432,

P

= 0·065, see Fig. 3). The results demonstrate thatthe mean change in muscle condition index duringbreeding tended to be the most significant factoraffecting the change in flight times during the courseof a breeding attempt, particularly for the immediatetake-off measures, while changes in body mass weregenerally unimportant. Moreover, the potential effectof body mass appeared to be in the opposite direc-tion to that of muscle volume, as the loss of bodymass over the laying period tended to cause birds tofly faster (compare Figs 2c and 3).

Discussion

It has previously been stated that ‘in a class of verte-brates [birds] in which the young depend on healthyparents for food and protection, protein loss frommuscles cannot normally be so great as to impairlocomotion’ (Jones 1991). In this series of experi-ments, we have demonstrated that this is not the case,with muscle wastage resulting from egg productionbeing sufficient to depress alarmed flight speeds tothe extent that predation risk may be increased. In allthree measures of flight performance, the range inchanges in mean muscle condition during egg-layingresulted in a variation of

≈

20% in the time taken tofly the defined trial distances. Previous studies invest-igating the effects of reproduction upon locomotionin birds (Lee

et al

. 1996), reptiles (Schwarzkopf &Shine 1992) and crustacea (Berglund & Rosenqvist1986) have primarily considered the costs of carryingeggs. That the changes in flight times here appear to

be independent of changes in body mass, particularlyat the take-off stage, and are seen to persist into thepostlaying phase, indicates that the real cost in thesebirds may lie not in the carrying of developing eggs,but rather in the production of those eggs and themuscle wastage that this incurs. These results demon-strate not only the existence of a cost of egg produc-tion, but also the underlying mechanism of that cost.The implications of this are wide reaching, and sug-gest that postlaying maternal predation risk mayform part of a physiological trade-off between repro-duction and survival, where investment in develop-ing eggs competes directly for resources vital to themaintenance of flight muscle tissue. Such costs ofreproduction may be sufficient to depress clutchsizes below the maximum which parents are capableof rearing, especially since it has been shown thatchanges in muscle condition are proportional to thenumber of eggs laid (Monaghan

et al

. 1998). In anenvironment where predators are present, femalesmust balance gains in fecundity with losses in musclecondition and the subsequent increase in predationrisk that this may entail, since the ratio of flightmusculature to body mass positively correlates withtake-off ability (Marden 1987). Such reductions intake-off velocity would greatly reduce the chancesof surviving an attack (Rudebeck 1950; Page &Whitacre 1975; Cresswell 1993; Bednekoff 1996).

Studies attempting to demonstrate a link betweenreproductive output and subsequent maternal preda-tion risk have typically focused upon the increasedlikelihood of a predatory attack rather than theincreased risk once an attack has been initiated(Magnhagen 1991; Székely, Karsai & Williams 1994)or have highlighted the decline or cessation in repro-ductive activity of animals under increased risk frompredation (Magnhagen 1991; Korpimaki, Nordahll& Valkama 1994). The importance of predation onparents as a cost of reproduction may have beenunappreciated in the past since most detailed studieshave taken part in the northern temperate zones wherethe effect of predators upon adult birds has beengreatly reduced by the activities of humans in remov-ing predators from ecosystems (Lima 1987). It shouldalso be noted that clutch size manipulations them-selves tend to decrease predation risk because ofthe continued presence of humans and the use ofpredator-proof nestboxes (Stearns 1992). However,since predators have begun to return to near historicallevels, the importance of predation upon adultsurvival, and subsequently evolution, has begun to beappreciated more (see Gosler, Greenwood & Perrins1995; Slagsvold & Dale 1996).

A number of studies have looked at the effects ofreproduction upon predator vulnerability once anattack has been launched, but these have typicallyfocused upon increases in mass associated with repro-duction in flying animals (Jones 1987; Hughes &Rayner 1993). However, recent work has shown that

Fig. 3. The relationship between changes in female bodymass during the course of laying a clutch and correspondingchanges in time to fly to a height of 1·15 m above the take-off point. Numbers beside points correspond to treatmenttype outlined in Table 1.

FEC391.fm Page 119 Wednesday, March 1, 2000 5:58 PM

120

J. S. Veasey

et al.

© 2000 British Ecological Society,

Functional Ecology

,

14

, 115–121

within the natural mass range of a bird, mass maynot be significant in affecting alarmed flight velocityover short distances during take-off (Witter, Cuthill& Bonser 1994; Lee

et al

. 1996; Kullberg, Jakobsson& Fransson 1998; Kullberg 1998; Veasey

et al

. 1998;Lind

et al

. 1999). Although a number of studieshave investigated the effect of egg carrying upon loco-motion in a variety of animals such as prawns(Berglund & Rosenqvist 1986), lizards (Schwarzkopf& Shine 1992) and starlings (

Sturnus vulgaris

) (Lee

et al

. 1996), we are not aware of any previous demon-stration that the consequences of egg production(rather than of carrying eggs) may increase predationrisk in iteroparous animals.

Lack (1947) was the first to propose that clutchsizes may have evolved to an optimal level, wherebybirds will lay a clutch that gives rise to the mostyoung fledged. He hypothesized that as brood sizesincreased, each chick would receive less resources andsubsequently survive less well (Lack 1947), and so fora long time clutch size was considered to have beendetermined by the trade-off between the number andfitness of offspring rather than by any costs to theparent. This was apparently corroborated by thework of Pettifor, Perrins & McCleery (1988), whofound the addition or subtraction of hatchlings to anest reduced recruitment in Great Tits (

Parus major

),but did not affect parental survival. However, theexistence of widespread reproductive costs in birds(such as increased predation risk) is suggested by thefact that mean clutch sizes in many birds appears tobe consistently lower than the most productive clutchsize (Stearns 1992). The trade-off between repro-duction and predation risk is such that a marginalincrease in current reproductive output such as lay-ing an extra egg, or laying a clutch when in a poorernutritional state, may jeopardize a female’s residualreproductive value (Clark 1994). This asymmetrictrade-off would tend to encourage birds to be con-servative in the number of eggs they lay. Although wehave shown that egg production may affect flight per-formance in the short term, the long-term effects areunknown, and the consequences for optimal clutchsizes may be difficult to determine. However, shorter-lived species are likely to be less conservative sincethe potential fitness gains of laying an extra egg inthe current clutch are a greater proportion of totallifetime reproductive output. Since the life expect-ancy of the Zebra Finch in the wild is low (51 days athatching), and few breeding attempts are made perseason (1·7 + 0·9) (Zann 1996), it is apparent thatfuture reproduction may be a relatively small propor-tion of the lifetime reproductive success of the aver-age wild Zebra Finch. Consequently, in comparisonto longer-lived birds, Zebra Finches may be morewilling to invest in the production of the currentclutch to the extent of impaired flight capacity.

It is likely that females would attempt to minim-ize any increases in predation risk by behavioural

adjustments at the time of laying. It has been shownfor example that activity in female birds declines atthe time of laying (Houston

et al

. 1995c). This phe-nomenon has previously been interpreted as attemptsto protect developing eggs or to minimize energeticexpenditure (Fogden & Fogden 1979; Houston

et al

.1995c). However, Schifferli (1976) found that develop-ing House Sparrow (

Passer domesticus

) eggs werenot vulnerable to mechanical damage during normalactivity, and in this series of experiments in which weflew 37 females on a total of 2505 occasions, not asingle individual showed symptoms of having damagedan egg. It seems unlikely that the observed reductionin female activity at this time is a result of an attemptto minimize energetic costs during laying, as egg pro-duction is energetically demanding and so one mightexpect an elevation in activity to allow for theincreased intake of resources needed for the repro-ductive attempt. It is suggested that the decline inactivity is more likely to be an adaptive responseto increased predation risk resulting from musclewastage, whereby females deliberately reduce theirexposure to predators at the time of laying.

Acknowledgements

We are grateful to the help of Reudi Nager and GraemeRuxton in the preparation of this paper, and toDorothy Armstrong for looking after the birds. Thisresearch was supported by NERC grant GR3/9995.

References

Bednekoff, P.A. (1996) Translating mass dependent flightperformance into predation risk: an extension of Metcalfe& Ure.

Proceedings of the Royal Society of London SeriesB

263

, 887–889.Berglund, A. & Rosenqvist, G. (1986) Reproductive costs in

the prawn

Palaemon adspersus

: effects on growth andpredator vulnerability.

Oikos

46

, 349–354.Clark, C.W. (1994) Antipredator behaviour and the asset-

protection principle.

Behavioural Ecology

5

, 159–170.Cresswell, W. (1993) Escape responses by redshanks,

Tringatotanus

on attack by avian predators.

Animal Behaviour

46

, 609–611.Dall, S.R.X. & Witter, M.S. (1998) Feeding interruptions,

diurnal mass changes and daily routines in the zebrafinch.

Animal Behaviour

55

, 715–725.Drent, R. & Daan, S. (1980) The prudent parent: energetic

adjustments in avian breeding.

Ardea

68

, 225–252.Fogden, M.P.L. & Fogden, P.M. (1979) The role of fat

and protein reserves in the annual cycle of grey-backedcamaraptera in Uganda (Aves: Sylvidae).

Journal ofZoology

189

, 233–258.Gosler, A.G., Greenwood, J.J.D. & Perrins, C.M. (1995)

Predation risk and the cost of being fat.

Nature

377

,621–623.

Hartman, F.A. (1961) Locomotor mechanisms of birds.

Smithsonian Miscellaneous Collections

.

143

, 1–91.Haywood, S. (1993a) Role of extrinsic factors in the control

of clutch size in the blue tit

Parus caeruleus

.

Ibis

135

, 79–84.Haywood, S. (1993b) Sensory control of clutch size in the

zebra finch (Taeniopygia guttata). Auk 110, 778–786.Heaney, V. & Monaghan, P. (1995) A within-clutch trade-off

FEC391.fm Page 120 Wednesday, March 1, 2000 5:58 PM

121Flight muscle atrophy and predation risk

© 2000 British Ecological Society, Functional Ecology, 14, 115–121

between egg production and rearing in birds. Proceedingsof the Royal Society of London Series B 261, 361–365.

Houston, D.C., Donnan, D., Jones, P., Hamilton, I. &Osborne, D. (1995a) Changes in the muscle condition ofthe female zebra finch Poephila guttata during egg layingand the role of protein storage in bird skeletal muscle. Ibis137, 322–328.

Houston, D.C., Donnan, D. & Jones, P. (1995b) The sourcesof nutrients required for egg production in zebra finchesPoephila guttata. Journal of Zoology 235, 469–483.

Houston, D.C., Donnan, D. & Jones, P. (1995c) Use of labelledmethionine to investigate the contribution of muscleproteins to egg production in zebra finches. Journal ofComparative Physiology B 164, 161–164.

Howland, H.C. (1974) Optimal strategies of predatoravoidance: the relative importance of speed andmanoeuvrability. Journal of Theoretical Biology 47, 333–350.

Hughes, P.M. & Rayner, J.M.V. (1993) The flight of pipist-relle bats Pipistrellus pipistrellus during pregnancy andlactation. Journal of Zoology 230, 541–555.

Jones, G. (1987) Sexual chases in sand martins (Ripariariparia): cues for males to increase their reproductivesuccess. Behavioural Ecology and Sociobiology 19, 179–185.

Jones, M.M. (1991) Muscle protein loss in laying housesparrows (Passer domesticus). Ibis 133, 193–198.

Jones, P.J. & Ward, P. (1976) The level of reserve protein asthe proximate factor controlling the timing of breedingand clutch size in the red-billed quelea. Ibis 118, 547–574.

Kendall, M.D., Ward, P. & Bacchus, S. (1973) A proteinreserve in the pectoralis major muscle of Quelea quelea.Ibis 115, 600–601.

Korpimaki, E., Norrdahll, K. & Valkama, J. (1994) Re-productive investment under fluctuating predationrisk: microtine rodents and small mustelids. EvolutionaryEcology 8, 357–368.

Kullberg, C. (1998) Does diurnal variation in body massaffect take-off ability in wintering willow tits? AnimalBehaviour 56, 227–233.

Kullberg. C., Jakobsson, S. & Fransson, T. (1998) Predator-induced take-off strategy in great tits (Parus major). Pro-ceedings of the Royal Society of London Series B 265,1659–1664.

Lack, D. (1947) The significance of clutch size. Ibis 89, 302–352.

Lee, S.J., Witter, M.S., Cuthill, I.C. & Goldsmith, A.R. (1996)Reduction in escape performance as a cost of reproductionin gravid starlings, Sturnus vulgaris. Proceedings of the RoyalSociety of London Series B 263, 619–624.

Lessells, C.M. & Boag, P.T. (1987) Unrepeatable repeatabil-ities: a common mistake. Auk 104, 116–121.

Lima, S.L. (1987) Clutch size in birds: a predation perspect-ive. Ecology 64, 1062–1070.

Lind, J., Fransson, T., Jakobsson, S. & Kullberg, C. (1999)Reduced take-off ability in robins due to migratory fuelload. Behavioural Ecology and Sociobiology 46, 65–70.

Magnhagen, C. (1991) Predation risk as a cost of reproduc-tion. Trends in Ecology and Evolution 6, 183–186.

Marden, J.H. (1987) Maximum lift production during take-off in flying animals. Journal of Experimental Biology 130,235–258.

McCleery, R.H., Clobert, J., Julliard, R. & Perrins, C.M.(1996) Nest predation and delayed cost of reproductionin the great tit. Journal of Animal Ecology 65, 96–104.

Metcalfe, N.B. & Ure, S.E. (1995) Diurnal variation in pre-

dation risk in birds. Proceedings of the Royal Society ofLondon Series B 261, 395–400.

Monaghan, P. & Nager, R.G. (1997) Why don’t birds laymore eggs? Trends in Ecology and Evolution 12, 270–274.

Monaghan, P., Nager, R.G. & Houston, D.C. (1998) Theprice of eggs: increased investment in egg productionreduces the offspring rearing capacity of parents. Pro-ceedings of the Royal Society of London Series B 265,1731–1735.

Page, G. & Whitacre, D.F. (1975) Raptor predation onwintering shorebirds. Condor 77, 73–83.

Pennycuick, C.J. (1969) The mechanics of bird migration.Ibis 111, 525–556.

Pettifor, R.A., Perrins, C.M. & McCleery, R.H. (1988)Individual optimisation of clutch size in great tits. Nature336, 160–162.

Rice, W.R. (1989) Analyzing tables of statistical tests. Evolu-tion 43, 223–225.

Rudebeck, G. (1950) The choice of prey and modes ofhunting of predatory birds with special reference to theirselective effort. Oikos 2, 65–68.

Schifferli, L. (1976) Factors affecting weight and condition inthe house sparrow Passer domesticus particularly whenbreeding. DPhil Thesis, Oxford University.

Schwarzkopf, L. & Shine, R. (1992) Costs of reproductionin lizards, escape tactics and susceptibility to predation.Behavioural Ecology and Sociobiology 31, 17–25.

Selman, R.G. & Houston, D.C. (1996a) The effect of pre-breeding diet on reproductive output in zebra finches.Proceedings of the Royal Society of London Series B263, 1585–1588.

Selman, R.G. & Houston, D.C. (1996b) A technique formeasuring the lean pectoral muscle mass in live smallbirds. Ibis 138, 348–350.

Slagsvold, T. & Dale, S. (1996) Disappearance of femalepied flycatchers in relation to breeding stage and experi-mentally induced moult. Ecology 77, 461–471.

Stearns, S.C. (1992) The Evolution of Life Histories. OxfordUniversity Press, Oxford.

Székely, T., Karsai, I. & Williams, D. (1994) Determina-tion of clutch size in the Kentish plover Charadriusalexandrius. Ibis 136, 341–348.

Tatar, M. & Carey, J.R. (1995) Nutrition mediates repro-ductive trade-offs with age-specific mortality in the beetleCallosobruchus maculatus. Ecology 76, 2066–2073.

Veasey, J.S., Metcalfe, N.B. & Houston, D.C. (1998) A reas-sessment of the effect of body mass upon flight speed andpredation risk in birds. Animal Behaviour 56, 883–889.

Ward, S. (1996) Energy expenditures of female barn swallowsHirundo rustica during egg formation. PhysiologicalZoology 69, 930–951.

Williams, T.D. (1996) Variation in reproductive effort infemale zebra finches (Taeniopygia guttata) in relation tonutrient-specific dietary supplements during egg laying.Physiological Zoology 69, 1255–1275.

Witter, M.S., Cuthill, I.C. & Bonser, R.H.C. (1994) Exper-imental investigations of mass dependent predation risk inthe European starling, Sturnus vulgaris. Animal Behaviour48, 201–222.

Zann, R.A. (1996) The Zebra Finch: A Synthesis of Fieldand Laboratory Studies. Oxford University Press, Oxford.

Received 9 June 1999; revised 11 August 1999; accepted 18August 1999

FEC391.fm Page 121 Wednesday, March 1, 2000 5:58 PM