SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Captured free-living male mallard Anas platyrhynchos at Abberton in southern Britain showed peak mass gain immediately prior to simultaneous remex moult. Individuals of both sexes were heavier before shedding wing feathers than when flightless confirming literature accounts that show mallard accumulate fat stores in anticipation of moult to contribute to meeting energy needs during remex re-growth. Over the course of four seasons, males lost 13 17% of initial body mass on average during re-growth of flight feathers, females 13 23%. Based on energy expenditure of 1.3 times BMR, male mallard were estimated to be able to fulfil 42 60% and females 41 82% of their energy needs throughout moult from stores. Free-flying male mallard fed ad libitum in a predator-free environment did not differ in starting body mass or rate of mass loss during wing moult compared to free-living Abberton birds, suggesting depletion of fat stores, irrespective of available sources of exogenous energy. Based on this evidence, we reject that the hypotheses that mass loss in moulting mallard is due to 1) simple energy stress and 2) restrictions on feeding and consider that 3) attaining the ability to fly at an earlier stage on incompletely grown flight feathers is not the primary factor shaping this trait. Rather, we consider the accumulation and subsequent depletion of fat stores, together with reductions in energy expenditure, enable mallard to re-grow feathers as rapidly as possible by exploiting habitats that offer safety from predators, but do not necessarily enable them to balance energy budgets during the flightless period of remex feather re-growth.

Virtually all wildfowl undergo simultaneous remex moult, re-growing flight feathers in a way that renders birds temporarily flightless. Mallard Anas platyrhynchos are flightless for 22 7 d (Owen and King 1979, Panek and Majewski 1990), during which time the inability to fly reduces the potential to escape predation and to move between profitable foraging opportunities (Hohman et al. 1992). In addition to these constraints, increased energy demand has been demonstrated to meet the costs of feather growth (Payne 1972, Thompson and Boag 1976, Dolnik and Gavrilov 1979, Qian and Xu 1986, Portugal et al. 2007), estimated at 1.3 times basal metabolic rate in mallard (Prince 1979). Faced with reduced mobility, mallard could meet the enhanced energy needs of growing new flight feathers through the adoption of one or more of three potential strategies: 1) compensatory food consumption (Hartman 1985); 2) reduction of energy consuming activities (inability to fly already profoundly reduces energy expenditure, Guillemette et al. 2007, but see also Portugal et al. 2010); 3) exhibit phenotypic plasticity, reconstructing body parts to minimise energy consumption and utilising body stores of fat and protein (accumulated prior to moult) to offset temporary increased demand (Young and Boag 1982, Fox and Kahlert 2005, Fox et al. 2008).

Young and Boag (1982) showed that during the flightless period, there was a significant decrease in mallard flight muscle mass and increase in leg muscle mass consistent with the predictions arising from the hypothesis of phenotypic plasticity (Ankney 1979, 1984, Fox and Kahlert 2005). Despite major changes in muscle architecture, these tend not to contribute to major overall changes in body mass during flightless moult in most studied Anatidae (Ankney 1979, 1984, Thompson and Drobney 1996). However, many European dabbling ducks (Sjöberg 1988, Panek and Majewski 1990, King and Fox 2012) and diving ducks (Fox and King 2011) conspicuously lose mass during flightless moult (although others do not, Fox et al. 2008), most likely due to consumption of body fat stores. Just as in the case of Saltholm moulting greylag geese Anser anser (Fox and Kahlert 2005), Folk et al. (1966) reported mass loss in moulting mallard and Young and Boag (1982) documented a reduction in fat stores in mallard through moult, although they asserted there was no overall change in total carcass lipid, total protein or total body mass. However, Panek and Majewski (1990) showed 12% declines in body mass amongst males and females through moult. So do mallard consistently lose body mass during moult and if so what is the explanation?

There has been debate as to whether such mass loss in moulting ducks is: 1) an adaptive trait, whereby fat stores provide an endogenous source of energy to regrow feathers as rapidly as possible whilst reducing reliance on external energy sources, access to which involves predation risk (as suggested for some geese, Fox and Kahlert 2005); 2) a simple reflection of the elevated energetic costs of feather synthesis which birds meet by catabolism of body ‘reserves’ (the energetic stress of Hohman 1993) rather than body ‘stores’ acquired specifically to fuel the process (sensu van der Meer and Piersma 1994); 3) due to predation risk, that imposes cryptic feeding behaviour and exploitation of habitats where foraging is less effective, such foraging constraints necessitate exploitation of fat (which does not necessarily preclude pre-moult accumulation of fat stores, Panek and Majewski 1990) and 4) because lighter body mass enables Anatidae to regain the capability of flight earlier on incompletely re-grown flight feathers earlier than if heavier (Owen and Ogilvie 1979, Brown and Saunders 1998).

In this paper, we first analyse changes in total body mass of free-living mallard captured during the flightless period to test whether or not they lose body mass during moult. Secondly, we look for support for the different explanations of how mallard would meet the enhanced energy demands during wing moult and assess whether body mass loss is an adaptive trait or should be explained by the alternative hypotheses. We use wing length as a measure of moult stage to examine how mass changes as birds move towards the point at which they regain the powers of flight. We also compare these results with those from free-flying male mallards provided with a non-limiting supply of food which select to moult in an artificially maintained predator-free area, to test whether observed patterns are the result of food limitation in the wild during the moult period or if the patterns can be considered adaptive. Males moult before females, so if mallard show adaptive mass (i.e. fat store) gain to meet the demands of wing feather replacement, we expect 1) body mass to peak just before the onset of moult and 2) that male mass would peak prior to their flightless period correspondingly earlier than females. The peak in body mass prior to wing moult would fail to support hypotheses 2) and 4) above, and if mallard fed ad libitum lost mass at the same rate as free flying birds, this would provide additional evidence for patterns of fat accumulation and depletion to meet the demands of moult, rather than simply result from energetic stress, i.e. consumption of reserves to meet energy demands contra stores (sensu van der Meer and Piersma 1994). Furthermore, because ad libitum fed males were present in a predator-free environment, we would expect there to be no feeding constraint on them and therefore if birds still lost mass under these circumstances this would fail to support hypothesis 3).

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Field determination of mass and moult stage

Mallard were caught annually in baited traps by RK at Abberton Reservoir, Essex in south-east England (50°40′N, 2°36′W) and morphometric data taken, along with details of plumage and ability to fly. Mallard were present and trapped before, after and during the wing moult period in 1978 (167 males and 93 females), while in 1983–1985 inclusive only moulting birds (i.e. those with new feathers that could not fly) were caught and processed, 86 males and 62 females in 1983, 88 males and 44 females in 1984 and 56 males and 29 females in 1985. Although it is known that wing feather re-growth does not show a linear relationship with time (Owen and King 1979), the length of primary feathers is the best index of moult stage of an individual at the point of capture. Measurements taken were maximum flattened and straightened wing length from the carpal joint to the tip of the longest primary feather, (to the nearest millimetre) and body mass (in grams, Owen and Montgomery 1978). Attempts were made to differentiate breeding females from failed or non-breeding females, but brood patches were not always evident in known breeding individuals later in moult and break up of broods late in the season precluded this method for confirming breeding status. We conclude it was therefore not possible to distinguish non-breeders from failed breeding females. Twenty-six flightless male mallard were rounded up in the captive bird enclosures within a fox-proof fence at the Wildfowl and Wetlands Trust (WWT) Centre at Slimbridge (51°44′N, 2°25′W) on 13 July 1979 and identical measurements were taken from birds in a situation where birds were attracted to virtually ad libitum provision of barley and wheat, both energy-rich sources of carbohydrate (Evans and Kear 1978). Since we show that the phenology of moult differs between the sexes and because body size and mass differs between the sexes (Cramp and Simmons 1977), in all the following analyses we treat the sexes separately. After confirmation that data were normally distributed, we subjected the data from both sexes to analysis of covariance to model body mass as the dependent variable based on wing length across years, using SAS PROC GLM (SAS 2011). In the case of females, we compared 1978, 1983, 1984 and 1985 data, for males, we included the Slimbridge ad libitum fed sample from 1979 as another ‘year’ class. Despite the fitting of such a composite model, we have here plotted individual regression lines to data from the different years for ease of interpretation and to compare year-specific body mass loss to estimate energy store depletion.

Moult phenology and patterns of mass accumulation/depletion

To determine the phenology of moult of mallard caught at Abberton, we constructed cumulative frequency distributions of moulting birds for each day for each of the years 1978 and 1983–1985. We summed the daily number of birds caught in each season that exhibited any stage of regrowth of remex feathers to the point where these could fly (determined by their behaviour on release after capture), and expressed these as a cumulative annual percentage to compare phenology between years and sexes. We tested for significant differences in the dates at which 50% of males and females were in remex moult using a Student t-test and we tested for divergence between years in these distributions within the sexes using Kolmogorov–Smirnov tests. Based on Abberton-caught mallard in 1978, we fitted regression models to determine mass from date of capture, with the expectation that a quadratic model would show a better fit than an ordinary linear model, quantified by a measurable improvement in the explanatory power. We reasoned that if mallard accumulated mass prior to moult, mass would peak prior to moult and subsequently decline through the flightless period. We differentiated the regression model equations to derive the date of peak mass for males and females.

Modelling energy consumption based on mass loss

Previous studies of greylag geese suggested mass loss during moult was almost entirely due to depletion of fat deposits accumulated prior to moult (Fox and Kahlert 2005). We here make the assumption that mass loss was directly equivalent to fat consumed to meet daily energy expenditure (Kahlert 2006) and that the flightless period equated to 26 d (Panek and Majewski 1990). We regressed the body mass of all flightless birds caught in the process of growing new feathers against wing length, which reflects the growth of the longest primary, and solved these equations to generate an average starting body mass (substituting wing lengths of 103 and 97 mm respectively, being the mean minimum values taken during the study) and that at close to the point of regaining flight (237 and 233 mm, mean maximum recorded flightless measurements) for males and females separately. The loss of mass between these points was converted to energy by multiplying by 38.9, the energy density of avian fat in kJ g−1 (Pennycuick 2008). Daily energy expenditure (DEE measured in kJ d−1) in Anatidae has been estimated on the basis of 2.6×basal metabolic rate (BMR) at other times of the year (Drent et al. 1981), but this incorporates periods spent flying, considered to consume energy at a rate equivalent to 10 12×BMR (Mooij 1992). Diving duck DEE has been experimentally determined to be between 1.7×and 2.5×(Bevan et al. 1995, de Leeuw et al. 1999) and moulting greylag geese moving daily between loafing and feeding areas were estimated to consume 1.7×BMR in DEE (Kahlert 2006). However, since wing moulting mallard neither fly nor dive and many duck species in wild studies and captivity show significant reductions in locomotion and feeding and increases in resting behaviour (Portugal et al. 2010 and references therein), it seems likely that their DEE would be less than 1.7× BMR and more likely to approach the 1.3×BMR of Prince (1979). We therefore estimate calculated DEE of male and female mallard during moult based on 1.7×, 1.3× and 1.0× BMR to assess the degree to which they could meet their DEE from exploitation of fat deposits alone. We used the mean final mass at the end of the flightless period (which we consider closer to metabolically active body mass than that at the start of moult, assuming the difference to be fat) to calculate BMR (kJ d−1) based on the equation BMR (kJ d−1) = 1.968 × mass0.767 (where mass is expressed in g, McNab 2009) and estimated DEE on the basis of different multiples of BMR, expressing the potential daily source of energy from fat stores as a percentage of this.

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Timing of wing moult

Males were recorded in wing moult at Abberton from 10 June until 12 September, females from 2 July to 29 September during 1978 and 1983 1985 inclusive and males moulted significantly earlier than females (Students t test corrected for unequal variances, t4= 7.0, p < 0.001, Fig. 1, Supplementary material Appendix 1). There were significant differences between the frequency distributions from all years amongst females (Kolomogorov Smirnov 0.159 <D < 0.482, p < 0.01) and males (0.274 < D < 0.699, p < 0.01), suggesting considerable variability in the annual timing of moult of both sexes at this one site. There was far greater variation in the start of moult amongst females (date by which 10% of birds were moulting was 5 July ± 2.0 d SE for males, 1 August ± 6.9 d for females). The date at which 50% of mallard were flightless was 20 July (± 2.1 d) for males, 22 August (± 4.2 d) for females.

image

Figure 1. Cumulative frequency of flightless moulting mallard per date amongst male and female mallard caught at Abberton in 1978 and 1983–1985 inclusive.

Download figure to PowerPoint

Overall changes in male body mass

Mean body mass of captured male mallard at Abberton peaked during late June 1978, confirmed by the first derivative by differentiation of the quadratic function identifying a local maximum on 21 June 1978 (Fig. 2a), i.e. during the period immediately prior to the moult period for the majority of male birds (Fig. 1). During July when most males were in moult, flightless birds were significantly lighter (by ca 6%, mean 1069.4 g ± 11.07 SE, n = 82) than those that had not yet commenced moult when captured (mean 1135.8 g ± 18.13 SE, n = 18, ANOVA F1,99= 6.99, p < 0.01, Fig. 2a, Supplementary material Appendix 2).

image

Figure 2. (a) Mass of male mallard captured at Abberton Reservoir on different dates between 19 May and 9 October 1978. Open symbols indicate birds capable of flight on unmoulted flight feathers, solid symbols indicate birds in moult or capable of flight on newly grown feathers. Fitted least squares second order polynomial regression model (fitted to all data) has the equation y =−0.0449x2+ 15.56x =− 257.21, r = 0.44, p < 0.001, where x = ordinal date). (b) Mass of female mallard caught during the period 19 May and 9 October 1978, conventions as for males. Best fit least squares second order polynomial regression model (fitted to all data) has the equation y =−0.0065x2+ 3.0027x + 571.78, r = 0.06, p > 0.05).

Download figure to PowerPoint

Analysis of covariance confirmed that male mallard lost mass in all years (ANCOVA, r = 0.61, F9,422= 27.5, p < 0.0001, Fig. 3a), but since the interaction term (wing ×year) failed to reach statistical significance (type III F4= 0.61, p = 0.65) the slopes of the regression were not significantly different between years. The final model incorporated wing length and year (ANCOVA, r = 0.61, F5,422= 48.4, p < 0.0001), but despite body mass being lighter in 1978 there was no significant difference in intercepts between this and other years (Fig. 3a). Note that the best fit single year regression models for Slimbridge in 1979 and Abberton in 1985 were almost identical (Fig. 3a).

image

Figure 3. Relationships between body mass and wing length from male (a) and female (b) mallard captured at Abberton Reservoir, Essex, south east England (‘Abb’, free flying birds in a wild setting) in 1978 and 1983–1985 and at Slimbridge, Gloucestershire, south west England (‘Slim’, free flying birds fed ad libitum) in 1979. Fitted lines indicate highly significant best least squared fitted regression models fitted to each set of data, although an ana lysis of covariance showed no significant differences in slopes between years (see text for details).

Download figure to PowerPoint

Overall changes in female body mass

Changes in mean body mass of captured females were more complex, probably because of the differential demands of brood rearing that added to the variance in individual timing of moult. Nevertheless, the differentiation of the best fit quadratic regression showed a local maximum body mass on 17 August 1978, although the model did not provide a significant fit. By mid August, 50% of females were moulting in 1978 (Fig. 2b). Females moulting flight feathers during the entire month of August (when most were in moult) where significantly lighter (by ca 17%, mean 865.3 g ± 12.19 SE, n = 69, ANOVA F1,86= 47.5, p < 0.001) than those that were not re-growing flight feathers when captured during the same period (mean 1041.8 g ± 20.51 SE, n = 19, Fig. 2b).

Analysis of covariance showed that female mallard lost mass in all years (ANCOVA, r = 0.63, F7,228= 20.9, p < 0.0001, Fig. 3b), but since the interaction term (wing ×year) failed to reach statistical significance (type III F3= 1.13, p = 0.34) the slopes of the regression were not significantly different between years. The final model incorporated wing length and year (ANCOVA, r = 0.62, F4,228= 35.6, p < 0.0001), with no differences in slopes between years (Fig. 3b).

Modelling energy consumption based on mass loss

On average, males lost 13–17% of average initial mass during the flightless period in any one year and females 13–23% (Table 1). Assuming 1) the body mass determinations relate to reasonably lean body mass in each case, 2) that all mass loss was due to fat depletion and 3) that DEE equated to 1.3× BMR during moult, the amount of energy available to males in the form of fat stores would satisfy 42–60% of total energy needs during moult and 41–82% of those of female DEE (Table 1). This rises to 55–78% and 53–107% respectively if DEE were brought to levels equivalent to resting metabolism. Energy expenditure based on 1.0 × and 1.7 × BMR are also shown in Table 1 for comparison.

Table 1.  Estimates of energy consumption of moulting mallard at Abberton Reservoir and Slimbridge, southern Britain. See text for full explanation of the methods; final mean body mass is calculated using the annual regression model of mass on wing length solved for the maximum wing length recorded for flightless birds (237 and 233 mm for males and females respectively).
     Predicted daily energy consumption (kJ d–1) based on initial body mass estimated by different levels of energy consumption relative to BMR (values in brackets indicate percentage of total energy needs supplied from fat stores during moult)
Sex, place and year of captureTotal imputed fat mass consumption (g) (values in brackets indicate percentage mass loss of starting mass)Imputed daily fat mass consumption based on flightless period of 26 d (g d−1)Energy supplied by this amount of fat (kJ d−1)Final mean body mass (g)1.7× BMR1.3× BMR1.0× BMR
Males Abberton 1978183.6 (16.5%)7.1274.7927.3631.4 (43.5%)482.9 (56.9%)371.4 (74.0%)
Males Slimbridge 1979205.8 (17.0%)7.9307.91006.8672.5 (45.8%)514.3 (59.9%)395.6 (77.8%)
Males Abberton 1983159.1 (13.8%)6.1238.0996.7667.4 (35.7%)510.3 (46.6%)392.6 (60.6%)
Males Abberton 1984144.5 (12.6%)5.6216.21006.2672.3 (32.2%)514.1 (42.1%)395.4 (54.7%)
Males Abberton 1985199.5 (16.4%)7.7298.51016.9677.7 (44.1%)518.3 (57.6%)398.7 (74.9%)
Females Abberton 1978229.7 (22.9%)8.8343.6771.4548.3 (62.7%)419.3 (82.0%)322.5 (106.5%)
Females Abberton 1983179.6 (17.0%)6.9268.8880.3606.7 (44.3%)464.0 (57.9%)356.9 (75.3%)
Females Abberton 1984160.3 (15.2%)6.2239.9890.5612.2 (39.2%)468.1 (51.2%)360.1 (66.6%)
Females Abberton 1985130.8 (12.5%)5.0195.6912.1623.5 (31.4%)476.8 (41.0%)366.7 (53.4%)

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Male mallard caught during the moult in 1978 at Abberton showed peak mass levels immediately prior to the most intensive period of wing moult for each sex. Females showed a similar pattern, although failing to reach statistical significance, but literature sources confirm that females attain peak body mass later than males in relation to moult (Cramp and Simmon 1977, Pehrsson 1987). During the peak period of moult, individuals of both sexes were significantly heavier pre-moult than those which had already shed old feathers and were re-growing new ones. Several studies have shown elevated body mass in both sexes amongst mallard immediately preceding remex moult (Folk et al. 1966, Young and Boag 1982 and Pehrsson 1987, the latter amongst captive birds). We therefore contend that Abberton mallard accumulated mass (in all probability fat stores) in advance of wing moult, a feature of the annual mass dynamics of mallard, gadwall Anas strepera, teal A. crecca and pintail A. acuta according to Oring (1969) and Cramp and Simmons (1977). All of these species then subsequently lose mass during flight feather moult (Cramp and Simmons 1977, Sjöberg 1988, King and Fox 2012). The fact that peak mass was attained immediately prior to the most intense period of moult earlier in males (21 June) than in females (17 August, given the delay and prolonged moult in females compared to males) provides some additional support for theory that this mass gain is specifically associated with meeting the future demands of wing moult, confirmed by Folk et al. (1966). This suggests that mallard acquire post-breeding fat stores to meet the energetic challenges of wing moult just as they do at other stages in the annual cycle. For instance, it is demonstrated that migratory mallard accumulate anticipatory fat stores for migration, to meet energy deficits during periods of harsh winter (when DEE exceeds energy supply from food) and potentially for pre-basic moult (Owen and Cook 1977, Heitmeyer 1988).

Furthermore, the comparisons of mass loss during different years at Abberton and amongst free-flying birds with access to ad libitum food at Slimbridge show that the degree of mass accumulation at the start of moult and rate of depletion through moult did not differ between years or treatments based on measurements from many different individuals (except in 1978, when both sexes apparently started moult with low body mass compared to other years, although differences were not statistically significant). Of course, we should be prudent about comparing between different seasons when the timing of the moult differed significantly between years and sexes, since this suggests that seasonal factors such as timing of breeding and food supply could affect phenology and restrict comparisons between years and sites. For this reason, we should also be careful when comparing data from males at Abberton and Slimbridge in different years. Nevertheless, the similarity in the rate of mass loss in birds from different seasons and sites was remarkable and suggests a common underlying mechanism. We also cannot fully rule out the possibility that the degree of fat accumulation under some circumstances is environmentally determined, with mallard acquiring less body mass in anticipation of moult in some seasons, perhaps because of food limitation, that could affect ultimate body mass on completion of moult.

Studies by van de Wetering and Cooke (2000) and Portugal et al. (2007) showed relationships between speed of moult and body mass loss and between body size and mass loss during moult in other Anatidae species during wing moult. These studies suggest that the degree of body resources available to an individual for feather growth at the start of flightless wing moult could affect the rate of re-growth of remex feather and/or the reliance of that individual on exogenous sources of energy. Such individual and seasonal patterns were beyond the scope of the present study, but clearly future research should address the extent to which starting moult in better condition (i.e. greater body stores) allows an individual to increase feather growth rate and/or spend more time resting, potentially reducing predation risk (Portugal et al. 2011).

The regular rate of subsequent depletion of stores during wing moult confirmed the findings of Pehrsson (1987) who also showed mallard fed ad libitum rolled oats also showed declines in body mass, which suggests this may be a heritable trait, since the loss of mass was not accentuated in wild birds (despite their exposure to food limitation and predation risk) compared to those fed ad libitum. The fact that male mallard with ad libitum access to food in a totally predator-free situation lost mass at a similar rate to free-living birds fails to support the hypothesis that loss of body mass relates to constraint on foraging (Panek and Majewski 1990). Rather we believe that the mallard may show restraint on foraging during moult, and that they choose (or are able) to feed less, especially avoiding risky habitats where predation may be more likely, and so reduce energy expenditure as well as predation risk by supplementing exogenous sources of energy by depleting fat stores (Portugal et al. 2010, 2011).

There remains some discussion about the true costs of wing feather replacement. Metabolic rate apparently increases during wing moult in Anatidae species, but not during body moult (as shown by elevated energy consumption, Qian and Xu 1986, Portugal et al. 2007). This suggests that increases in metabolic rate are associated with simultaneous replacement of the largest feathers or that this is a mechanism to facilitate rapid feather growth to foreshorten the flightless period. Either way, the mallard studied here appear to exploit fat stores laid down prior to moult, rather than burning mass of a form not stored as lipid in advance of this period of extra demand. For these reasons, despite apparent increased energetic demands, we reject the hypothesis that mass loss in mallard during wing moult is simply the result of energetic stress, i.e. individuals simply mobilising energy from body reserves (rather than fat stores, sensu van der Meer and Piersma 1994), i.e. burning the ‘engine’ rather than the ‘fuel’. Because both sexes accumulate body mass at slightly different periods in advance of moult, we also reject the suggestion that the mass loss is an effective shortening of the flightless period, by allowing the birds to fly on incompletely grown flight feathers – why accumulate stores immediately before remix moult if the end objective were to minimise mass in the latter part of the flightless period? Hence, the ability to fly on shorter wings is a consequence of the mass loss (Brown and Saunders 1998) but is unlikely to contribute to the explanation for this trait.

There is no doubt that testing of the theories behind mass dynamics and feather growth during remex growth in Anatidae can only be achieved by simultaneous measurement of feather growth, fat store depletion, food consumption and energy expenditure. Moulting waterbirds do not respond well to handling and stress under controlled captive conditions and manipulative experiments render themselves open to criticism about application of results to ‘wild’ situations. Nevertheless, there are limits to how much more study of free-living Anatidae can contribute to our understanding of how physiological state interacts with timing and rate of wing moult and with the quality of feather tissue produced under stress. There is no doubt that much more could be achieved by experimental studies to better enlighten moulting tactics of waterbirds replacing flight feathers in open habitats in relation to their energy needs.

Many species of dabbling and diving ducks deplete stores of body mass during moult (reviewed by Hohman et al. 1992), including some in the Southern Hemisphere

(Douthwaite 1976, Halse and Skead 1983, Halse and obbs 1985, Ndlovu et al. 2010), so the pre-wing moult fat accumulation trait to assist meet the energetic needs of flight feather re-growth is likely widespread in the Anatidae, although certainly not ubiquitous (Hohman et al. 1992, Fox et al. 2008). For this reason, it would be instructive to better understand the circumstances under which such adaptive traits have evolved and how these enable different populations to manage energetic balance during this period of enhanced energy cost, lost mobility and elevated predation risk. We conclude that the accumulation of body mass in mallard represents the build up of fat stores which are depleted during the flightless period to enable rapid regrowth of feathers whilst exploitating habitats (typically open water) that offers safety from predation, but do not necessarily enable the ducks to balance their energy budgets during the vulnerable flightless period of wing feather re-growth. We encourage the examination of other species to determine how widespread such a trait may be.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We gratefully acknowledge the inspiration of Hugh Boyd who fostered our interest in duck wing moult, and thank Sievert Rohwer and Steve Portugal for many stimulating discussions on this subject. Thanks to the ground staff at Slimbridge for allowing us to catch up birds there in 1979 and to the Nature Conservancy and Nature Conservancy Council for funding ringing activities at Abberton. We also thank Rich Hearn for reading the script and Johnny Kahlert and Barbara Helm for generous comments and improvements on an earlier version, and finally the then Essex Water for allowing WWT to catch and ring ducks at Abberton Reservoir.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • Ankney C. D. 1979. Does the wing molt cause nutritional stress in lesser snow geese? Auk 96: 6872.
  • Ankney C. D. 1984. Nutrient reserve dynamics of breeding and molting brant. Auk 101: 361370.
  • Bevan R. M., Speakman J. R. and Butler P. J. 1995. Daily expenditure of tufted ducks: a comparison between indirect colorimetry, doubly labelled water and heart rate. Funct. Ecol. 9: 4047.
  • Brown R. E. and Saunders D. K. 1998. Regulated changes in body mass and muscle mass in molting blue-winged teal for an early return to moult. Can. J. Zool. 76: 2632.
  • Cramp S. and Simmons K. E. L. 1977. Handbook of the birds of Europe the Middle East and North Africa. The birds of the Western Palearctic. Volume I. Oxford Univ. Press.
  • de Leeuw J. J., van Eerden M. R. and Visser G. H. 1999. Wintering tufted ducks Aythya fuligula diving for zebra mussels Dreissena polymorpha balance feeding costs within narrow margins of their energy budget. J. Avian Biol. 30: 182192.
  • Drent R., Ebbinge B. and Weijand B. 1981. Balancing the energy budgets of arctic-breeding geese throughout the annual cycle: a progress report. Verh. Ornithol. Ges. Bayern 23: 239264.
  • Dolnik V. R. and Gavrilov V. M. 1979. Bioenergetics of molt in the chaffinch (Fringilla coelebs). Auk 96: 253264.
  • Douthwaite R. J. 1976. Weight changes and wing moult in the red-billed teal. Wildfowl 27: 123127.
  • Evans M. E. and Kear J. 1978. Weights and measurements of Bewick's swans during winter. Wildfowl 29: 118122.
  • Folk C., Hudec K. and Toufar J. 1966. The weight of the mallard, Anas platyrhynchos, and its changes in the course of the year. Zool. Listy 15: 249250.
  • Fox A. D. and Kahlert J. 2005. Changes in body mass and organ size during wing moult in non-breeding greylag geese Anser anser. J. Avian Biol. 36: 538548.
  • Fox A. D. and King R. 2011. Body mass loss amongst moulting pochard Aythya ferina and tufted duck A. fuligula at Abberton Reservoir, south east England. J. Ornithol. 152: 727732.
  • Fox A. D., Hartmann P. and Petersen I. K. 2008. Body mass and organ size change during wing moult in common scoter. J. Avian Biol. 39: 3540.
  • Guillemette M., Pelletier D., Grandbois J.-M. and Butler P. J. 2007. Flightlessness and the energetic cost of wing molt in a large sea duck. Ecology 88: 29362945.
  • Halse S. A. and Skead D. M. 1983. Wing moult, body measurements and condition indices of spur-winged geese. Wildfowl 34: 108114.
  • Halse S. A. and Dobbs J. C. 1985. Carcase composition of spur-winged geese with comments on reproduction and moulting. Afr. J. Ecol. 23: 171178.
  • Hartman G. 1985. Foods of male mallard before and during moult, as determined by faecal analysis. Wildfowl 36: 6571.
  • Heitmeyer M. E. 1988. Body composition of female mallard in winter in relation to annual cycle events. Condor 90: 669680.
  • Hohman W. L. 1993. Body composition dynamics of ruddy ducks during wing molt. Can. J. Zool. 71: 22242228.
  • Hohman W. L., Ankney C. D. and Gordon D. H. 1992. Ecology and management of postbreeding waterfowl. – In: Batt B. D. J., Afton A. D., Anderson M. G., Ankney C. D., Johnson D. H., Kadlec J. A. and Krapu G. L. (eds), Ecology and management of breeding waterfowl. Univ. of Minnesota Press, Minneapolis, pp. 128189.
  • Kahlert J. 2006. Effects of feeding patterns on body mass loss in moulting greylag geese Anser anser.– Bird Study 53: 2031.
  • King R. and Fox A. D. 2012. Moulting mass dynamics of female gadwall Anas strepera and male wigeon A. penelope at Abberton Reservoir, south east England. Bird Study 59: 252254.
  • McNab B. K. 2009. Ecological factors affect the level and scaling of avian BMR. Comp. Biochem. Physiol. A 152: 2245.
  • Mooij J. H. 1992. Behaviour and energy budgets of wintering geese in the Lower Rhine area of North Rhine-Westphalia, Germany. Wildfowl 43: 121138.
  • Ndlovu M., Cumming G. S., Hockey P. A. R. and Bruinzeel L. W. 2010. Phenotypic flexibility of a southern African duck Alopochen aegyptiaca during moult: do northern hemisphere paradigms apply? J. Avian Biol. 41: 558564.
  • Oring L. W. 1969. Summer biology of the gadwall at Delta Marsh, Manitoba. Wilson Bull. 81: 4454.
  • Owen M. and Cook W. A. 1977. Variations in body weight, wing length and condition of mallard Anas platyrhynchos and their relationship to environmental change. J. Zool. 183: 377395.
  • Owen M. and Montgomery S. 1978. Body measurements of the mallard caught in Britain. Wildfowl 29: 123134.
  • Owen M. and King R. 1979. The duration of the flightless period in free-living mallard. Bird Study 26: 267269.
  • Owen M. and Ogilvie M. A. 1979. Wing molt and weights of barnacle geese in Spitzbergen. Condor 81: 4252.
  • Panek M. and Majewski D. P. 1990. Remex growth and body mass of mallards during wing molt. Auk 107: 255259.
  • Payne R. B. 1972. Mechanisms and control of molt. – In: Farner D. S. and King J. R. (eds), Avian biology. Vol. 2. Academic Press, pp. 103155.
  • Pehrsson O. 1987. Effects of body condition on moulting in mallards. Condor 89: 329339.
  • Pennycuick C. J. 2008. Modelling the flying bird. – Elsevier.
  • Portugal S. J., Green J. A. and Butler P. J. 2007. Annual changes in body mass and resting metabolism in captive barnacle geese (Branta leucopsis): the importance of wing moult. J. Exp. Biol. 210: 13911397.
  • Portugal S. J., Isaac R., Quinton K. L. and Reynolds S. J. 2010. Do captive waterfowl alter their behaviour patterns during their flightless moult? J. Ornithol. 151: 443448.
  • Portugal S. J., Green J. A., Piersma T., Eichhorn G. and Butler P. J. 2011. Greater energy stores enable flightless moulting geese to increase resting behaviour. Ibis 153: 868874.
  • Prince H. H. 1979. Bioenergetics of post-breeding dabbling ducks. – In: Bookhout T. A. (ed.), Waterfowl and wetlands: an integrated review. Proceedings of the 1977 Symposium of the North Central Section of The Wildlife Society, Madison, WI, pp. 103117.
  • Qian G. H. and Xu H. F. 1986. Molt and resting metabolic rate in the common teal Anas crecca and the shoveler A. clypeata. Acta Zool. Sinica 32: 6873.
  • SAS 2011. SAS proprietary software release 9.3. – SAS Inst., Cary, NC.
  • Sjöberg K. 1988. The flightless period of free-living male teal Anas crecca in northern Sweden. Ibis 130: 164171.
  • Thompson D. C. and Boag D. A. 1976. Effect of molting on the energy requirements of Japanese quail. Condor 78: 249252.
  • Thompson J. E. and Drobney R. D. 1996. Nutritional impli cations of molt in male Canvasbacks: variation in nutrient reserves and digestive tract morphology. Condor 98: 512526.
  • van de Wetering D. and Cooke F. 2000. Body weight and feather growth of male Barrow's goldeneye during wing molt. Condor 102: 228231.
  • van der Meer J. and Piersma T. 1994. Physiologically inspired regression models for estimating and predicting nutrient stores and their composition in birds. Physiol. Zool. 67: 305329.
  • Young D. A. and Boag D. A. 1982. Changes in the physical condition of male mallards (Anas platyrhynchos) during moult. Can. J. Zool. 60: 32203226.

Supplementary material (Appendix JAB5785 at < www.oikosoffice.lu.se/appendix >). Appendix 1 2.